Prostate Cancer Resistance to Cabazitaxel Chemotherapy

University of Kentucky UKnowledge

Theses and Dissertations--Toxicology and Cancer Biology Toxicology and Cancer Biology

2020

Diane Begemann University of Kentucky, [email protected] Author ORCID Identifier: https://orcid.org/0000-0002-1742-4825 Digital Object Identifier: https://doi.org/10.13023/etd.2020.187

Right click to open a feedback form in a new tab to let us know how this document benefits ou.y

Recommended Citation Begemann, Diane, "Prostate Cancer Resistance to Cabazitaxel Chemotherapy" (2020). Theses and Dissertations--Toxicology and Cancer Biology. 31. https://uknowledge.uky.edu/toxicology_etds/31

This Doctoral Dissertation is brought to you for free and open access by the Toxicology and Cancer Biology at UKnowledge. It has been accepted for inclusion in Theses and Dissertations--Toxicology and Cancer Biology by an authorized administrator of UKnowledge. For more information, please contact [email protected]. STUDENT AGREEMENT:

I represent that my thesis or dissertation and abstract are my original work. Proper attribution has been given to all outside sources. I understand that I am solely responsible for obtaining any needed copyright permissions. I have obtained needed written permission statement(s) from the owner(s) of each third-party copyrighted matter to be included in my work, allowing electronic distribution (if such use is not permitted by the fair use doctrine) which will be submitted to UKnowledge as Additional File.

I hereby grant to The University of Kentucky and its agents the irrevocable, non-exclusive, and royalty-free license to archive and make accessible my work in whole or in part in all forms of media, now or hereafter known. I agree that the document mentioned above may be made available immediately for worldwide access unless an embargo applies.

I retain all other ownership rights to the copyright of my work. I also retain the right to use in future works (such as articles or books) all or part of my work. I understand that I am free to register the copyright to my work.

REVIEW, APPROVAL AND ACCEPTANCE

The document mentioned above has been reviewed and accepted by the student’s advisor, on behalf of the advisory committee, and by the Director of Graduate Studies (DGS), on behalf of the program; we verify that this is the final, approved version of the student’s thesis including all changes required by the advisory committee. The undersigned agree to abide by the statements above.

Diane Begemann, Student

Dr. Natasha Kyprianou, Major Professor

Dr. Isabel Mellon, Director of Graduate Studies

PROSTATE CANCER RESISTANCE TO CABAZITAXEL CHEMOTHERAPY

______

DISSERTATION ______

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the College of Medicine at the University of Kentucky

By Diane Begemann Lexington, Kentucky

Co-Directors: Dr. Natasha Kyprianou, Professor of Urology, Pathology and Oncological Sciences, Vice Chair of Basic Science Research, Department of Urology, Icahn School of Medicine at Mount Sinai New York, New York And Dr. John D’Orazio, Professor of Toxicology & Cancer Biology, University of Kentucky Lexington, Kentucky 2020

ABSTRACT OF DISSERTATION

PROSTATE CANCER RESISTANCE TO CABAZITAXEL CHEMOTHERAPY

The plasticity of prostate tumors contributes to the heterogeneity in response and acquisition of therapeutic resistance in advanced prostate cancer. Disruption of the phenotypic landscape via epithelial-mesenchymal transition (EMT) enables prostate tumors to invade and metastasize. Our previous studies demonstrated that cabazitaxel (a 2nd generation FDA-approved taxane chemotherapy) that is used for treatment of castration-resistant prostate cancer (CRPC), causes reversal of EMT to mesenchymal- epithelial transition (MET). The present study examined the effect of sequencing cabazitaxel chemotherapy mediated MET on prostate tumor re-differentiation and its impact on overcoming resistance in models of advanced prostate cancer. The presence of DHT (1nM) decreases the efficacy of cabazitaxel treatment in vitro. Cabazitaxel treatment in vivo induced MET in LNCaP xenograft tumors as shown by increased E-cadherin and decreased N-cadherin expression. Sequencing cabazitaxel after ADT improves tumor response in androgen sensitive LNCaP models but not in CRPC 22Rv1 models. In addition, a novel AR-HSET kinesin interaction is identified as a potential androgen receptor axis therapeutic target. Our findings provide new insights into re-programming prostate cells into an epithelial phenotype, towards re-sensitizing the cell to therapeutic targeting. There is high translational impact in therapeutic sequencing (cabazitaxel chemotherapy and ADT), an avenue towards improved therapeutic response in patients with advanced lethal CRPC.

KEYWORDS: Prostate Cancer, Androgen Receptor, Microtubules, Taxanes, Anti- Androgen, Castration Resistant Prostate Cancer

Diane Begemann (Diane Begemann)

May 13, 2020 Date

PROSTATE CANCER RESISTANCE TO CABAZITAXEL CHEMOTHERAPY

By Diane Begemann

Dr. Natasha Kyprianou Co-Director of Dissertation

Dr. John D’Orazio Co-Director of Dissertation

Dr. Isabel Mellon Director of Graduate Studies

May 11, 2020 Date

Acknowledgements

The following dissertation, while an individual work, benefited from the insights and direction of many people. First, my mentor and dissertation advisor Dr. Natasha Kyprianou, exemplifies the high quality scholarship to which I aspire. Dr. Isabel Mellon provided incredible support, as Director of Graduate Studies, and a sounding board in times of need. I wish to thank the complete dissertation committee, and outside examiner, respectively: Dr. John D’Orazio (co-chair), Dr. Qiou Wei, Dr. Binhua Zhou, Dr. Isabel Mellon, and Dr. Brett Spear who all provided timely and instructive comments and evaluation at every stage of the dissertation process, allowing me to complete this project on schedule. I would like to thank Dr. Hong Pu (Kyprianou lab, University of Kentucky), who taught me so many valuable skills and lent her expertise in countless in vitro and in vivo experiments. I would like to thank Dr. Xiaoqi Liu, Department of Toxicology & Cancer Biology Chair, for guiding the department and for kindly hosting me in the Liu Lab for my final experiments. Each individual provided insights that guided and challenged my thinking, substantially improving the finished product. In addition, I received equally important assistance from the research team in the Department of Urology at Mount Sinai. Special thanks to Dr. Ash Tewari, Dr. Sujit Nair, Kamala Bhatt, Rachel Weil, and Jennifer Stockert (Tewari lab) for helping me navigate a major change of scenery. To our collaborators, Dr. Wei Yang with Cedars Sinai Medical Center and Dr. Nora Navone with MD Anderson to supporting my research with their expertise. My family and friends provided on-going support throughout the dissertation process. Thanks to my parents, Ann and John Begemann, for always supporting and encouraging me. To Ross Guffey, who was always patient, supportive, and caring, even when the pursuit of this degree led me to New York City. My friends and steady supporters, who helped keep me focused when I needed to focus, and kept me distracted when I needed a break – Veronica Feth, Carli McGinley, and Nannette and Harold Eichell in particular. While no one in my friends and family knew anything about prostate cancer 5 years ago, I think they all do now!

iii

Table of Contents Acknowledgements ...... iii

List of Figures ...... vii List of Tables ...... ix

Chapter One: Introduction ...... 1 The Normal Prostate Gland ...... 1 Prostate Cancer ...... 3 Cell Death Failures in Cancer Development ...... 4 Epithelial-Mesenchymal Transition (EMT) in Cancer Cell Dynamics ...... 7 Threading the Metastatic Journey of Cancer Cells ...... 10 EMT in the Tumor Microenvironment ...... 12 a. Transforming Growth Factor-β ...... 12 b. The Androgen Receptor (AR) Axis ...... 13 AR Targeting Therapies ...... 15 a. Androgen Deprivation Therapy (ADT): ...... 15 b. Enzalutamide (MDV3100): ...... 16 c. Abiraterone Acetate: ...... 17 d. N-terminal targeting, an approach for AR-variant inhibition: ...... 18 Contributions of Androgens to EMT ...... 19 Taxane Chemotherapy in Advanced Prostate Cancer: Stabilizing Microtubules ...... 20 a. Paclitaxel and Docetaxel: ...... 20 b. Cabazitaxel:...... 22 Microtubule Transport Proteins ...... 23 a. KIFC1/HSET: ...... 24 b. KIF2C/MCAK ...... 25 c. Dynein motor proteins: ...... 26 Taxanes Mechanisms of Action Beyond Microtubule Stabilization ...... 26 Targeted Therapeutics in Prostate Cancer ...... 29 Conclusions ...... 37

Chapter Two: Impact of Androgens on Therapeutic Response to Chemotherapy in Models of Advanced Prostate Cancer ...... 49 Background and Significance ...... 49 Experimental Approach ...... 51 Cell Lines: ...... 51 Cell Viability Assays:...... 52 In vivo xenograft experiments: ...... 52 Immunohistochemical Analysis: ...... 53 Apoptosis Detection: ...... 53 Co-immunoprecipitation: ...... 54 Western Blotting: ...... 54 Immunofluorescent Confocal Microscopy: ...... 54 Targeted mass spectrometry quantification of cabazitaxel: ...... 55 Statistical Analysis: ...... 56 Results ...... 58 Effect of androgens (DHT) on cell viability response to cabazitaxel chemotherapy in vitro: ...... 58 Treatment sequencing of ADT and cabazitaxel modifies the tumor dynamics in LNCaP and 22Rv1 prostate tumor xenografts: ...... 59 Novel targetable interaction between HSET and AR by cabazitaxel: ...... 60 HSET expression in in vivo xenograft tumors: ...... 61 HSET and HSET/AR interaction are molecular targets of cabazitaxel: ...... 61 Conclusions ...... 63

Chapter Three: Androgen Independent Contributor to Prostate Cancer Therapeutic Resistance ...... 89 Significance...... 89 Approach ...... 91 Cell Lines: ...... 91 Cell Viability Assays:...... 91 In vitro Cell Migration and Invasion ...... 91 In vivo Xenograft Models of Advanced Prostate Cancer: ...... 92 Immunohistochemical Analysis: ...... 92

Apoptosis Detection: ...... 93 Western Blotting: ...... 93 Immunofluorescent Confocal Microscopy: ...... 94 Quantitative real-time reverse transcription polymerase chain reaction analysis ...... 94 RNA-sequencing analysis ...... 95 Targeted Mass spectrometry Quantification of Cabazitaxel: ...... 95 Statistical Analysis: ...... 96 Table 3: Primary antibodies used ...... 97 Results ...... 98 Conclusions ...... 105

Chapter Four: Discussion ...... 136

References ...... 144

Vita ...... 160

List of Figures

Figure 1.1: Anatomy and histology of a normal prostate gland and BPH...... 40 Figure 1.2: AR axis targeting therapeutics ...... 41 Figure 1.3: Androgen Receptor and Variants Share the N-Terminal Domain...... 43 Figure 1.4: Taxane Mechanism of Antitumor Action...... 44 Figure 1.5: Effect of EMT to MET interconversion on anoikis sensitivity...... 45 Figure 1.6: EMT and anoikis are navigated by AR and TGF-β cross-talk signaling...... 46 Figure 1.7: Chemical Structures of FDA-approved Therapeutics for Prostate Cancer Treatment...... 48 Figure 2.1: DHT increases prostate cancer cell viability in vitro: Dose response studies...... 67 Figure 2.2: DHT affects cabazitaxel efficacy in AR-negative cells...... 68 Figure 2.3: DHT monotherapy does not affect cell viability in PC3 cells...... 69 Figure 2.4: Prostate tumor xenograft response to cabazitaxel and ADT in vivo...... 70 Figure 2.5: Effect of cabazitaxel and ADT on prostate tumor growth...... 72 Figure 2.6: Reversal of EMT to MET in advanced prostate tumor xenografts, androgen sensitive LNCaP and CRPC 22Rv1...... 75 Figure 2.7: Effect of ADT and cabazitaxel on AR expression in prostate tumor xenografts...... 77 Figure 2.8: Significance of HSET Kinesin in Prostate Cancer...... 80 Figure 2.9: Effect of cabazitaxel treatment on kinesin and AR expression and localization...... 82 Figure 2.10: HSET and HSET/AR interaction are molecular targets of cabazitaxel...... 85 Figure 2.11: DHT challenges cabazitaxel effect in vitro in prostate cancer cells...... 86 Figure 2.12: Interface of cabazitaxel and ADT on AR and HSET cellular distribution .... 87 Figure 3.1: Cabazitaxel inhibits androgen independent prostate tumor xenograft growth and increases tumor apoptosis...... 109 Figure 3.2: Cabazitaxel treatment alters HSET localization in androgen-independent prostate cancer cells...... 111 Figure 3.3: EMT phenotype in cabazitaxel resistant cells (PC3-CR) ...... 113 Figure 3.4: Cabazitaxel resistance is associated with migratory and invasive properties in prostate cancer cells...... 115 Figure 3.5: RNA profile of cabazitaxel resistance...... 116

vii

Figure 3.6: Loss of SLCO1B3 expression in cabazitaxel resistant cells...... 118 Figure 3.7: SLCO1B3 alterations correlate with survival: TCGA analysis...... 119 Figure 3.8: SLCO1B3 loss increases both cabazitaxel uptake and resistance to cabazitaxel...... 121 Figure 3.9: Treatment with cabazitaxel and ADT decreases membrane localization of SLCO1B3...... 123 Figure 3.10: Bcl-2 overexpression sensitizes PC3 prostate cancer cells to cabazitaxel. . 125 Figure 3.11: Patient derived xenografts (PDX) models from cabazitaxel treated, metastatic CRPC...... 127 Figure 3.12: Epithelial and mesenchymal phenotype in PDX tumors...... 128 Figure 3.13: Cellular differentiation corresponds with epithelial character in PDX tumors...... 130 Figure 3.14: HSET Kinesin expression is consistent across PDX tumors...... 132 Figure 3.15: SLCO1B3 expression correlates with epithelial phenotype in cabazitaxel resistant prostate tumors...... 134

viii

List of Tables

Table 1. Summary of PCa therapeutics and the cellular processes they target...... 39 Table 2. List of primary antibodies used, Chapter Two...... 57 Table 3. List of primary antibodies used, Chapter Three...... 97

Chapter One: Introduction The Normal Prostate Gland

The human prostate is a pyramidal-shaped endocrine organ located in the male pelvis beneath the bladder. The prostate surrounds the urethra, and normally weighs 15-

20g (1). The prostate is essential for the process of fertilization – it is responsible for secreting a fluid component of semen in ejaculation that protects and nourishes sperm (2).

The organ is surrounded by a fibrous layer deemed the prostatic capsule or prostatic fascia, and fibromuscular tissue that contracts during ejaculation to expel prostatic fluid with sperm (1, 2). Though it is one gland, there are three distinct histological zones – peripheral, transition, and central zones. The peripheral zone comprises the posterior aspect of the gland and surrounds the distal urethra, and it is estimated that 70% of all prostate cancers develop in the peripheral zone (1). In aging men, benign prostatic hyperplasia (BPH) is a frequent cause of lower urinary tract symptoms, and is defined by enlarged prostatic volume (>30mL) (3). Approximately 50% of men over 50 years of age have pathological evidence of BPH (3). A summary of prostate anatomy, histology, and BPH is shown

(Figure 1.1).

During embryonic development the prostate develops from the endodermal urogenital sinus (UGS), which develops outgrowths that differentiate into prostatic glandular epithelium in response to fetal androgens (4). The prostate remains a glandular organ, consisting of mostly pseudostratified epithelium but also containing areas of cuboidal, squamous, and transitional epithelium (5). Prostatic acini and ductal epithelium consists of three types of cells: luminal, basal, and neuroendocrine (1). Columnar luminal

cells secrete specialized products into the gland lumen, which contribute to seminal fluid

– one of these products is prostate specific antigen (PSA), a protein that is measured in the blood for prostate cancer detection (6). The numerous ducts within the gland drain into canals, and then into 15-20 main ducts, which drain into the urethra (5). Although the prostate is a glandular organ, there is a significant amount of prostatic stroma, a fibromuscular component containing smooth muscle, fibroblasts, blood vessels, and nerves that supports the epithelial cells with both physical infrastructure and through cell-cell signaling (7).

In prostate development and normal function, androgens promote cell growth, survival, and differentiation (8, 9). The principle male steroidal androgen, testosterone, is secreted from the testes and converted within the prostate epithelial cell by 5α-Reductase into its metabolite 5α-dihydrotestosterone (DHT) (9). DHT has a high binding affinity for the androgen receptor (AR), which causes AR dimerization and binding to co-regulators, translocation to the nucleus where it acts as a transcription factor for target genes via binding to the androgen response element (ARE) (8, 9). Multiple signaling cascades can result from this AR activity; increased free intracellular calcium, activation of protein kinase A (PKA), protein kinase C (PKC), and transcription of roughly 500 genes that promote growth and proliferation both in a normal prostate and as a critical pathway in the development of prostate cancer (9, 10).

Prostate Cancer

Prostate cancer affects hundreds of thousands of men in the United States, with over

174,000 new cases and over 31,000 estimated deaths for 2019 alone (11). Prostate cancer has the highest number of new diagnoses in males, and is the second most lethal male cancer (11). Over 75 years ago, in a ground breaking innovation, male steroid hormones were discovered to be involved in prostate cancer growth and proliferation (12). Hormone withdrawal, or androgen deprivation therapy (ADT), was discovered to cause growth regression in prostate cancer by pioneer urologic surgeon-scientist Dr. Charles Huggins.

However, for the majority of prostate cancer patients with androgen-dependent tumors,

ADT is not curative, and disease regression/relapse after ADT is common (13-15). After

ADT, emergence of tumors that are resistant are deemed castration resistant prostate cancer

(CRPC), which can then develop into the worst clinical outcome for patients – metastatic castration resistant prostate cancer (mCRPC) (13-15). Unfortunately, the length of time before disease recurs after hormone therapy is variable and unpredictable, but can be as low as only several months (16).

A patient at risk for prostate cancer risk is primarily monitored through prostate specific antigen (PSA) screening, a simple blood test, and a physical/digital rectal exam (DRE). If a patient has a rising PSA after radiation treatment or surgery, the disease is termed biochemically recurrent. An estimated 50-65% of patients have biochemically recurrent disease (17) returning within a median of 18-24 months after ADT (18). While local disease is highly curable with radical prostatectomy surgery, robotic prostatectomy or

radiation therapy, those who progress to mCRPC have a median survival of less than two years and 90% will have metastasis to the bone (18).

Prostate epithelial cells undergo apoptosis in response to ADT treatment roughly 72 hours after hormone withdrawal (19-21). However, apoptosis occurs in only roughly 2-

3% of the total cell population alongside decreased proliferation in the majority of the cell population (21). Apoptosis is the most studied cell death mechanism in response to hormone withdrawal, but ADT can also induce cellular senescence, and is linked with non- programmed and non-specific cell death (necrosis) in a minority of cells (22).

Cell Death Failures in Cancer Development

Eukaryotic cells die via three modes: (1) apoptosis, programmed cell removal triggered by either internal cellular damage or an external signal; (2) anoikis, death induced by detachment from the extracellular matrix (ECM) and loss of contact from the microenvironment; and (3) necrosis, a non-specific process resulting in cell lysis and inflammation (23, 24). Cancer cells have developed mechanisms to avoid programmed pathways anoikis and apoptosis, engaging in abnormal survival mechanisms to sustain their aberrant growth (23). Evasion of apoptosis and emergence of anoikis resistance are critical mechanisms contributing to cancer initiation, metastasis and therapeutic resistance, thus rendering restoration of cell death mechanisms an attractive therapeutic platform for the treatment of advanced recurrent tumors (25-28). Overcoming apoptosis is essential for

initial tumor formation, acquisition of metastatic properties, and secondary site colonization (25, 29, 30)

Mechanistically, there are two primary pathways to initiate apoptosisthe extrinsic/death-receptor (CD95 or FAS) pathway, and the intrinsic/mitochondrial pathway

(31). Extrinsic apoptosis is activated by death-receptor ligand binding, such as FASL or

TRAIL (Fas ligand and TNF Related Apoptosis Inducing Ligand, respectively) (23).

Binding of these ligands induces receptor polymerization and subsequent activation of caspase-8 (26, 30). Caspase-8 triggers the activation of executioner caspases, such as caspase-3, 6, and 7, which carry out the apoptosis signaling cascade (32). Intrinsic apoptosis begins through intracellular signals, such as mitochondrial or endoplasmic reticulum stress and DNA damage leading to apoptosome formation (cytochrome c, Apaf1, and caspase-9) and subsequent activation of executioner caspases 3, 6, and 7 (23).

Extracellular signals can trigger the intrinsic pathway; for example lack of survival signals

(e.g. growth factor withdrawal) can result to intrinsic pathway activation (33, 34). In response to pro-apoptotic proteins like Bak and Bax (bcl-2 antagonist killer and bcl-2 like protein 4, respectively) are activated while anti-apoptotic proteins Bcl-2 and Bcl-xL are subdued (35, 36). An increase in pro-apoptotic proteins bak/bax and decrease in anti- apoptotic proteins Bcl-2/Bcl-xL cause cytochrome c release, apoptosome formation, caspase cascade activation and apoptosis (32, 37).

Anoikis is the mode of apoptotic cell death induced upon cell detachment from the

ECM (25, 29). Normal epithelial and endothelial cells adhere to ECM, and die via anoikis, if ECM contact is lost. Mechanisms conferring resistance to anoikis facilitate the cancer cell dissemination, functionally contributing to metastasis and therapeutic resistance (30,

38). Under conditions of normal homeostasis, detachment of epithelial or endothelial cells from the ECM leads to an increase of FAS and FASL, FAS polymerization and caspase-8 activation, and ultimately anoikis induction (30). ECM detachment directly causes increased Bak activity and decreased Bcl-xL activity, thus triggering apoptosis via the intrinsic pathway (36). Anoikis resistant cells fail to activate caspase-8 after losing ECM anchoring, and engage Bcl-xL activity and consequently survive (30, 39). Normal cell-

ECM interactions are mediated through integrins, transmembrane proteins that are critical for cell attachment, survival, differentiation, migration and invasion (32). Signal transduction from the extracellular environment to the intracellular network through the integrin network dictates anoikis resistance, cell invasion, and metastasis outcomes (34).

Through aberrant integrin function, epithelial or endothelial cells can avoid anoikis by signal modifications including the integrins directly, downstream signaling effectors (37,

38) or activation of survival pathways dependent on integrins (23, 33, 40, 41).

AR signaling can induce anoikis through downregulation of p38, and ultimately suppress metastasis (42). Intracellular reactive oxygen species in prostate cancer have been shown to activate pro-survival signals and anoikis resistance, with sensitivity to anoikis restored upon antioxidant treatment in prostate epithelial cells (43). Chronic redox damage leads to Src activation regardless of cellular attachment, conferring anoikis resistance to highly damaged cells (43). If detachment of cells from the prostate epithelium occurs, basal progenitor cells are stimulated to divide and differentiate into an epithelial phenotype to replace them (44).

Epithelial-Mesenchymal Transition (EMT) in Cancer Cell Dynamics

Epithelial-mesenchymal transition (EMT) is a process through which an epithelial cell acquires migratory and invasive characteristics, and is a pivotal mechanism utilized by cells to escape their native microenvironment (35). Cells at the leading edge/periphery of tumors have undergone EMT, implicating the process in tumor development and expansion

(45). Epithelial cells are normally very cohesive, held together in a single layer via cell- cell interactions and navigated by cadherins, specifically E-Cadherin (46). Adherence junctions are formed to connect stable epithelial cells, containing E-cadherin and other proteins such as β-catenin and actin to form stable junctions (46). Through linking the cellular actin network to that of neighboring cells, adhering junctions effectively form a single, unified sheet from individual epithelial cells (46). Loss of E-Cadherin expression and adhering junctions directly causes a loss in normal epithelial cell polarity and de- differentiation into an amorphous cell without its anchor to other epithelial cells (Figure

1.5). Upon EMT induction, adherence junctions disintegrate, releasing β-catenin from the junction into the cell cytosol, where it can translocate to the nucleus and act as a transcription factor (46); moreover other transcriptional regulators like E-Cadherin repressors snail and twist (46). Epithelial cells have a distinct polarity, with proteins localized to either the apical or basolateral membrane surface. Membrane proteins claudin and occludin maintain these junctions, and are associated with cytosolic proteins such as

Ankyrin to maintain polarity (47). In order for epithelial cells to lose their polarity, these tight junctions must be dissolved.

The phenotypic landscape of EMT is characterized by loss of epithelial markers (e.g.

E-Cadherin, β-catenin, occludin) and a gain of mesenchymal markers (e.g. N-Cadherin,

Zeb-1/2, vimentin, snail, twist) in addition to a loss of cell polarity (35). Signaling pathways through extracellular receptors can trigger EMT reprogramming, such as platelet- derived growth factor receptor (PDGFR), fibroblast growth factor receptor (FGFR), transforming growth factor-β (TGF-β), and insulin-like growth factor-1 receptor (IGF-1R)

(48). Intracellular signaling pathways such as PI3K, Akt, and mTOR (48) can also navigate

EMT outcomes. A key event in EMT induction is transcriptional regulation by primary transcription factors Snail, Slug, and Twist, which suppress expression of epithelial markers (E-cadherin) and upregulate corresponding mesenchymal proteins (N-cadherin)

(48).

Upon acquisition of the mesenchymal phenotype cells no longer need attachment for survival – therefore, cells that have undergone EMT do not die through anoikis but acquire freedom and survival, liberated from the “home” epithelial layer (47). The key transcriptional regulators of EMT such as Slug and Snail confer anoikis resistance by targeting cell-cell adhesion molecules, as well as intracellular signaling effectors (25).

Recent evidence supports the ability of E-cadherin to sensitize cells to anoikis, an intuitive role if the structural integrity of adhering junctions is disturbed (47). Transcriptional repressors such as these silence downstream genes through epigenetic control (6). Post- translationally, E-cadherin can be degraded/cleaved or endocytosed under control of the

GTPase Arf6 (6). Consequential to EMT induction, E-cadherin transcription and expression are decreased. Loss of E-cadherin has been shown to make cells resistant to anoikis, as demonstrated in E-cadherin knockout models, indicating that loss of an

epithelial phenotype is directly linked to loss of anoikis-mediated cell killing (47, 49). E- cadherin loss leads to a decrease in other proteins, such as β-catenin and Ankyrin-G. When not sequestered in adhering junctions with E-cadherin, β-catenin is released from the cytosol into the nucleus, acting as a transcription factor in control of genes regulating proliferation and migration. While knockout of E-cadherin decreases anoikis, simultaneous loss of both E-cadherin and β-catenin restores sensitivity to anoikis, indicating that β-catenin is also a common control denominator between EMT and anoikis

(47, 50). Ankyrin-G, a protein that connects the cytoskeleton to the cell membrane, is also lost when cells undergo EMT and downregulate E-cadherin (47). Loss of Ankyrin-G leads to reduced expression of p14ARF, a tumor suppressor that promotes anoikis – therefore, anoikis is also regulated through interactions between E-cadherin and Ankyrin-G (47), further implicating traditional EMT players as regulators of anoikis. The impact of interfacing EMT and anoikis in the tumor microenvironment landscape includes loss of cell-cell contact, acquisition of chemoresistance, immune escape (51), and metastatic spread (35). In advanced disease, loss of E-cadherin expression correlates with higher

Gleason score, the pathological scoring of prostate cancer on a scale of 1 (no cancer) to 5

(widespread cancer) (36), extended local tumor progression, higher clinical stage, increased likelihood of metastasis and higher chance of biochemical recurrence after radical prostatectomy (52). Thus an integrated EMT signature of the tumor landscape provides a unique biomarker value in prostate cancer progression.

Threading the Metastatic Journey of Cancer Cells

In tumor progression, transitioning between a small, initial solid tumor into a metastatic cancer involves deregulation of several critical cellular processes.

Neovascularization to the tumor has to occur before the tumor can exceed 2mm (24).

Anoikis suppression by cancer cells promotes tumor neovascularization, making anoikis resistance not only mandatory for cell dissociation from its microenvironment, but also necessary for tumor expansion (24). After volume expansion and the accumulation of mutations in the cancer cells, some cells manage to disassociate from and degrade the associated basement membrane/ECM and connective tissue, travel through ECM, and gain access to lymphatic or blood vessels (46). Cancer cells must evade anoikis in order to continue surviving in circulation. Anoikis evasion has been correlated with increasing numbers of prostate tumor circulating cells (CTC) (53), with the number of CTC correlated with poor prognosis, late disease recurrence and patient mortality, indicating the importance of anoikis evasion in disease progression (54, 55). The combined effect of undergoing EMT and avoiding anoikis increases both cellular invasiveness and metastasis

(56, 57)

Anoikis resistance confers invasive properties in cancer cells, leading to metastasis, contributing to therapeutic resistance and ultimately impacting cancer patient mortality

(58). The significance of anoikis evasion and activation of survival pathways as leading contributors to the emergence of CRPC has been established (27, 38, 59). The Src- navigated survival pathway whose activation is functionally linked the emergence of CRPC therapeutic resistance, leads to anoikis resistance via engaging the PI3K/Akt

(phosphoinositide-3 kinase and protein kinase B, respectively) pathways (27). Moreover,

loss of the PTEN tumor suppressor gene encoding for a protein that directly suppresses the

PI3K/Akt pathway; loss of PTEN therefore increases PI3K/Akt activity, increases cell survival, and is common in poorly differentiated CRPC (28). In addition, overexpression of Bcl-2 and Bcl-xL results in apoptosis resistance, and is related to tumor survival after chemotherapy and radiotherapy (60). A transcriptional repressor of E-cadherin, Snail, activates cell survival pathways such as PI3/Akt, inhibits executor caspase-3 antagonizes p-53 mediated apoptosis, in addition to downregulating E-cadherin (61). Snail has been directly associated with chemoresistance (61). Snail is activated by Notch1, a protein associated with EMT and commonly found in advanced tumors and recurring prostate cancer (62). Twist, another transcriptional repressor, is associated with poor prognosis and the emergence of the mesenchymal phenotype (63) and can upregulate Bcl-2, enhancing survival and resistance to radiotherapy and chemotherapy (64, 65).

Considering that extravasation has likely already occurred at the time of a patient’s diagnosis (66) renders preventative measures to impair tumor metastatic spread a challenge. Cells are either invading or dividing, but not both simultaneously (67); thus invasive cells are arrested in G1, capable of re-entering the cell cycle upon settling in a secondary location (67). Circulating cells that have evaded anoikis are therefore not sensitive to classic chemotherapeutics, which target dividing cells, allowing the arrested cells to persist and seed secondary tumors (67, 68). Phenotypic reversal of EMT to MET is temporally required in the colonizing tumor cells to retain their growth kinetics at the distant site of metastasis and confer a proliferative capacity with the cells return to a dividing phenotype and undergo mitosis (69).

EMT in the Tumor Microenvironment

a. Transforming Growth Factor-β

TGF-β (Transforming growth factor β) signals from the stroma and tumor microenvironment have a great effect on cancer cells (46, 63, 70). TGF-β signaling in cancer is a complicated process, and can be simplified by understanding TGF-β as having a dual function. In normal cells and initial stages of cancer, TGF-β acts as a tumor suppressor, but in later stages of cancer switches to a more classical growth factor and is associated with increased metastatic potential and worse patient prognosis (70). A mechanism for this switch in TGF-β responses is not yet fully understood. Excess TGF-β is correlated with rapid disease metastasis, and cancer metastases often contain higher

TGF-β levels than the original seed/solid tumor (70).

Metastatic cells from primary prostate tumors that have formed “quiescent micro- metastatic deposits” in the secondary location of bone are stimulated to maintain their invasive characteristics through endogenous TGF-β signaling (46). These deposits of a single or a few cells can further develop into large metastases through continued EMT promotion (46). Efforts to understand/define the contribution of TGF-β to EMT in the tumor microenvironment led to a link between high metastases occurrence and EMT of metastatic cells in the pre-metastatic niche (46, 63) (as illustrated on Figure 1.6).

Mechanistically, TGF-β signaling begins with the TGF-β ligand binding a type II receptor

(TβRII), which activates type I receptors (TβRI) and further initiating signal transduction through intracellular Smad2 and Smad3 phosphorylation (71). Active/phosphorylated

Smad2 and 3 translocate to the nucleus and form a complex with Smad4 (71). The complex

of Smad 2, 3, and 4 acts as a transcription factor, recruits other cofactors, and increase the production of EMT transcription factors Snail, Twist, and Zeb1/2 (71). Through this canonical Smad signaling, TGF-β is linked to EMT. In a non-canonical/Smad independent manner, TGF-β can determine EMT outcomes through downstream effectors MAP kinase and PI3K/Akt pathways (72). In addition to promoting EMT, TGF-β also regulates production of matrix metalloproteinases, enhancing the ability for cells undergoing EMT to migrate through the extracellular matrix as it is being digested by matrix metalloproteinases (63). TGF-β signaling also dictates the actin cytoskeleton remodeling by regulating cofilin expression and increasing filopodia formation and giving cells the ability to migrate and metastasize (62, 73). b. The Androgen Receptor (AR) Axis

Activation of AR signaling in prostate epithelial cells occurs upon ligand binding and AR translocation to the nucleus where it functions as a transcription factor (Figure 1.2).

However, in many cancers there is aberrant AR protein and signaling, with over 20 splicing mutations known (74). AR can gain functional activity by diverse mechanisms including constitutive activation, ability to bind androgenic metabolites, and resistance to antiandrogen therapy, via mutations in the ligand binding domain (74). Microtubule-targeting taxane chemotherapy can also block AR translocation to the nucleus and therefore inhibit

AR transcriptional activity regardless of the AR mutation load (74, 75).

Dihydrotestosterone (DHT), the circulating form of androgen, suppresses of E- cadherin expression through an increase in Snail (48). However, this effect requires a threshold low level of AR, suggesting an inverse relationship between level of AR expression and degree of EMT induction through DHT/AR signaling (48). A common

therapy could be promoting an aggressive phenotype through induction of EMT. Work from this lab has shown that use of an AR antagonist such as abiraterone can negate the

EMT changes induced by canonical androgen signaling and ADT (48). By preventing

EMT, AR antagonists function on more than one level as a treatment therapy; stopping androgen-dependent growth while also promoting an epithelial phenotype (48). Androgen signaling linked to EMT contributes to high number of metastatic CRPC cases and there is a growing body of evidence supporting the involvement of EMT in the development of

CRPC (48, 76, 77). Low AR expression induced by ADT contributes to the aggressive nature of CRPC through inducing EMT, facilitating metastasis, and decreasing therapeutic efficacy (48, 77). This evidence indicates that low AR signaling promotes prostate cancer, but it is also known that gain of function AR mutations and AR overexpression promote prostate cancer growth as well (48, 78). Intermittent ADT, where androgen therapy is administered with scheduled gaps, is a treatment scheme designed to avoid AR downregulation and has provided some promise of improved clinical response to ADT

(48). Unfortunately, similar to what is seen with first generation ADT, many patients with

CRPC become resistant to second generation antiandrogens enzalutamide and abiraterone.

Resistance can develop through various mechanisms, including AR mutations, splice variants, constitutive activation, increased AR expression, or increased androgen synthesis from adrenal or intra-tumoral sources (78). Because several AR splice variants and mutants have altered C-terminal ligand binding domain, the un-mutated N-terminal domain has become a target of AR inhibition in recent research in order to treat the variants and mutants as effectively as the wild type AR (74).

AR Targeting Therapies

In a subpopulation of prostate cancer patients, ADT leads to the emergence of castration resistant prostate cancer (CRPC) and the cancer no longer responds to ADT.

Targeting androgen signaling and increased cell proliferation through a combination of anti-androgens and microtubule targeted chemotherapy has survival benefits for patients with metastatic CRPC (mCRPC), but therapeutic resistance can develop, resulting in lethal disease (79-82). Therapy resistance can result from the addiction of CRPC cells to androgen receptor (AR) signaling and constitutive activation of ligand independent AR splice variants (83).

a. Androgen Deprivation Therapy (ADT): Androgens have been defined as a growth factor for prostate gland growth and in prostate cancer since a seminal discovery by Huggins and Hodges (12). Since that discovery, ADT in prostate cancer treatment has been targeting the AR signaling pathway (Figure 1.2). More specifically, all clinically approved, existing endocrine targeting therapies in prostate cancer target the ligand binding domain (LBD) of the AR. The original method of ADT was the removal of testicular androgens by orchiectomy surgery, introduced in late 1940s by the pioneering work of Dr Charles Huggins (12, 84, 85). Patients were also treated with estrogens to block testicular androgens, but these patients experienced severe cardiovascular side effects (84).

Shortly after the discovery of the androgen receptor, the first generation antiandrogen flutamide was discovered in 1967 and was later found to compete for AR binding and prevent its nuclear translocation (86). Second generation antiandrogen, bicalutamide, was discovered in the 1980’s and has a four-fold greater binding affinity for AR binding than

its predecessor flutamide (86). Luteinizing-hormone releasing hormone (LHRH) analogs were then developed, such as leuprolide (Lupron) and goserelin (Zoladex). Systemic treatment with a LHRH analog/agonist results in GnRH receptor downregulation in the pituitary, which then results in a decrease of luteinizing-hormone (LH) and follicle stimulating hormone (FSH) production, followed by inhibition of testicular testosterone production (85). Finasteride, the first type II 5α-Reductase inhibitor discovered, was initially used in patients with BPH to reduce testosterone conversion to DHT, and was found in the Prostate Cancer Prevention Trial to decrease risk of men developing prostate cancer (87, 88). Finasteride is also used in patients to treat androgenic alopecia, working through reducing scalp DHT level and preventing hair loss (87). Side effects of long term finasteride treatment include increased risk of urinary problems, sexual side effects, and a paradoxical increased risk of high-grade prostate cancer despite overall decreased risk of prostate cancer (88).

b. Enzalutamide (MDV3100): The second generation antiandrogen flutamide was FDA approved in 1989, bicalutamide in 1995; then there was a lack of progress in AR inhibitors until third generation antiandrogen enzalutamide was discovered in 2008 and FDA approved in 2012 (86). Enzalutamide is an AR antagonist that binds to the AR with a higher affinity than bicalutamide, preventing AR translocation to the nucleus, the recruitment of AR cofactors, and blocking AR binding to DNA (86, 89). The

AFFIRM Phase III, double blinded, placebo controlled clinical trial showed that enzalutamide had therapeutic benefit in metastatic castration-resistant prostate cancer patients that had previously been treated with taxanes (90, 91). Enzalutamide-treated patients had PSA reduction, increased time to first skeletal related event, and an overall

survival of 18.4 months versus 13.6 months for placebo treated patients (91). In a direct comparison trial, patients that had been pre-treated with docetaxel had a worse response to enzalutamide than taxane-naïve patients, indicating that there may be a cross resistance mechanism occurring when docetaxel is given before enzalutamide, or that patients who progressed through docetaxel have disease that is less responsive to enzalutamide (92).

Enzalutamide action puts a selective pressure on prostate cancer cells to restore androgen signaling and continue proliferating. AR-variants, such as AR-v7, and mutations, commonly the F876L point mutation, are common mechanisms of resistance to enzalutamide (93). In addition, in an AR-independent mechanism of resistance, glucocorticoid receptor expression is increased in cells that are resistant to enzalutamide, found in both pre-clinical and clinical studies (93, 94). Stimulation with glucocorticoid agonist dexamethasone is sufficient to confer enzalutamide resistance, while a glucocorticoid antagonist restores sensitivity (94). Alarmingly, corticosteroids are routinely administered to patients undergoing docetaxel and abiraterone treatment to ameliorate side effects (94). The chemical structure of enzalutamide is shown in Figure

1.7.

c. Abiraterone Acetate: Abiraterone, a novel anti-androgen therapy, was

FDA approved in 2018 and targets adrenal androgen synthesis (95). CYP17 is an enzyme critical in testosterone synthesis; through irreversible inhibition of CYP17, abiraterone blocks precursors for testosterone and DHT from being generated (95). In an initial clinical trial on abiraterone, the COU-AA-301 randomized, double blinded, placebo controlled

Phase II clinical trial in metastatic castration resistant prostate cancer patients, found that abiraterone provided a survival benefit of 4.6 months when compared to control (96, 97).

In patients who had already received enzalutamide alone or in combination with docetaxel, abiraterone was substantially less effective (98, 99). Similar to enzalutamide, treatment with abiraterone can pressure cells to develop mutant or variant androgen receptors to maintain AR signaling for growth and proliferation. The chemical structure of abiraterone is shown in Figure 1.7.

d. N-terminal targeting, an approach for AR-variant inhibition:

Structural changes in the AR, such as splice variants, are functional contributors to therapeutic resistance. Many of the AR-variants have the ligand binding domain either mutated or deleted, located at the C-terminus of the AR (Figure 1.3) (83, 100). Since most traditional anti-androgens bind the C-terminus, ADT does not work on AR-variants; it is also probable that the loss of the C-terminal domain is from selective pressure by C- terminal targeting ADT (74, 100, 101). Targeting the AR N-terminal domain (NTD) offers a broader AR population that can be effected by therapy, and this is the approach of the

EPI small molecule inhibitors of the AR NTD (74, 101).

A functional N-terminal domain is necessary for AR translocation to the nucleus, emphasizing the importance of a functional AR retaining a full length NTD (102). EPI-

001 is a small molecule inhibitor of the AR N terminal domain, and prevents AR translocation to the nucleus through inhibiting protein-protein interactions necessary in the

AR transcriptional complex such as CREB-binding protein and RAP74 (102-104). EPI-

001 effectively inhibits constitutively active AR splice variants that lack the ligand binding domain and are resistant to other ADT therapies (102). EPI-001/002 in combination with docetaxel improves outcome in pre-clinical models of prostate cancer, decreases tumor growth, and increases apoptosis induction (74). In the same drug family, EPI-506 works in

the same manner of AR NTD inhibition and is being investigated in phase I/II clinical trials for safety, dose, and efficacy in patients with prostate cancer (101). Clinical trials are ongoing with the EPI drug family, with clinical potential for patients that express constitutively active AR-variants that are resistant to other AR targeting therapies.

Contributions of Androgens to EMT

The plasticity of fully differentiated prostate tumor glandular epithelium allows individual cells to de-differentiate into mesenchymal-like derivatives via the process of epithelial-mesenchymal transition (EMT) and re-differentiate via the reverse process, mesenchymal-epithelial transition (MET). While EMT is a normal process in embryonic development (105), it can contribute to developing metastasis and advanced disease (106).

EMT is characterized by a loss of epithelial cell markers (e.g. E-cadherin) and gain of mesenchymal cell markers (e.g. N-cadherin, vimentin) particularly at the invasive edge of a tumor (106). Tumor cell EMT results from a set of transcriptional reprogramming, induced by signals from transforming growth factor-β (TGF-β), insulin-like growth factor-

1 receptor (IGF-1R), fibroblast growth factor receptor (FGFR), and platelet derived growth factor receptor (PDGFR) (107, 108). The androgen axis and TGF-β signaling intertwine in androgen-sensitive prostate cancer cells (109); moreover, prostate tumor cells undergo

EMT more readily with a low AR content (59), indicating that there is multilayered complexity in the interaction between androgens and TGF-β towards EMT. ADT has been shown to induce EMT in a sub-population of prostate cancer cells (110, 111), while our recent studies supported the impact of cabazitaxel to reverse EMT to MET (112).

Taxane Chemotherapy in Advanced Prostate Cancer: Stabilizing Microtubules

Taxane chemotherapy is the standard of care for patients with metastatic CRPC who develop resistance to second generation anti-androgens such as abiraterone or enzalutamide (113, 114). For over a decade, taxanes were the only class of Food and Drug

Administration (FDA)-approved chemotherapeutic agents to confer additional survival and palliative benefit to CRPC patients (113, 114). Second line taxane, cabazitaxel (Jevtana), was optimized from first line taxane chemotherapy (docetaxel, paclitaxel) for increased cellular retention due to its lower affinity for the p-glycoprotein efflux pump (114).

Taxanes exert their primary antitumor action by stabilizing β-tubulin microtubule subunits, preventing microtubule depolymerization which induces G2-M arrest and apoptosis (Figure

1.4) (113). Taxanes pro-apoptotic effects are able to counteract Bcl-2 overexpression that is frequently observed in prostate cancer (115-118).

Optimizing treatment with taxanes and ADT is a continual therapeutic challenge.

Patients who have been treated with abiraterone have decreased response to docetaxel, indicating a cross-resistance between the two therapies in clinical settings (119). Preclinical evidence also shows evidence for cross resistance between AR targeting agents abiraterone, enzalutamide and taxane chemotherapy (120).

a. Paclitaxel and Docetaxel: Paclitaxel was the first clinically available taxane, but was rapidly surpassed by docetaxel, a semisynthetic taxane that inhibits taxane de-polymerization approximately twice as effectively as paclitaxel in pre-clinical models

(121). Clinical benefit for the use of docetaxel in patients with CRPC originated with a

series of clinical trials, TAX327 and SWOG-9916, which found a survival benefit of docetaxel (122). Despite the fact that docetaxel is the first-line treatment for CRPC, response in some patients is limited and resistance develops (122, 123). Surprisingly, PSA decrease is observed in only 45% of patients with CRPC when on docetaxel treatment

(122). In an effort to combat resistance to docetaxel, preclinical studies have tested the efficacy of giving small molecules that disrupt the anti-apoptotic reaction of Bcl-2/Bcl-xL with BAX/BAK (ABT-263 and ABT-737) and have found increased antitumor effects with combination treatment (123-125).

Docetaxel was tested after ADT in the STAMPEDE trial in patients with non- metastatic disease (31 men), and no benefit was determined by adding docetaxel after ADT

(126). A different STAMPEDE arm for patients with metastatic disease (550 men), in which docetaxel therapy was given a median of 9 weeks after beginning ADT, showed a

22% reduction in risk of all-cause death and a median of 10 months increased survival

(126). These STAMPEDE results indicate that docetaxel is effective in metastatic disease when given after ADT, but no benefit was found in non-metastatic disease.

Mutations in β-tubulin can occur, preventing taxane binding and altering microtubule monomer interactions (127). Docetaxel has a high binding affinity for P-

Glycoprotein-1, an ATP-dependent drug efflux pump, thus increasing its movement out of the cell (128, 129). Additionally, increased P-Glycoprotein-1 is observed throughout prostate cancer progression, an event that is either unrelated to docetaxel treatment but still aids in docetaxel resistance inadvertently, or an increase that develops in response to therapies (128, 129). The chemical structure of docetaxel is shown in Figure 1.7.

b. Cabazitaxel: The second line taxane chemotherapy cabazitaxel (Jevtana) has been shown to be effective in docetaxel-resistant CRPC patients (130, 131).

Cabazitaxel is one of a very limited therapeutic options for patients who have failed docetaxel therapy and hormonal treatment (131). Pre-clinical studies of cabazitaxel have shown that cabazitaxel stabilizes microtubules as effectively as docetaxel, but was 10-fold more potent than docetaxel in chemotherapy-resistant tumor cells (IC50 for docetaxel,

0.17-4.01 µmol/L; cabazitaxel, 0.013-0.414 µmol/L) (114). In cell lines resistant to docetaxel, cabazitaxel was still effective, indicating that it might be effective in clinical settings for patients with docetaxel resistant disease. Cabazitaxel is taken up by the cells

10 times faster than docetaxel; in addition, docetaxel has a half-life of 12 hours, while cabazitaxel is retained much longer and has a half-life in humans of 95 hours (132). Pre- clinical results indicate that although cabazitaxel likely binds to microtubules at the same binding site as docetaxel, it stabilizes microtubules more strongly than docetaxel, is taken up more quickly, and is retained longer in the cell (132). The chemical structure of cabazitaxel, with structural differences from docetaxel, is shown in Figure 1.7.

In TROPIC, a randomized phase III clinical trial in men with mCRPC who had received previous hormone therapy and had progressed through docetaxel, cabazitaxel was found to increase median survival from 12.7 months to 15.1 months (133). Cabazitaxel treated patients had a 30% relative reduction in risk of death, doubled rate of progression- free survival, and showed PSA and tumor response (133). Following this study, cabazitaxel was approved by the FDA in 2017 for use in docetaxel resistant CRPC patients.

Published seven years after TROPIC, a different phase III trial, PROSELICA, investigated a lower dose of cabazitaxel (20mg/m2 compared to 25 mg/m2) in an effort to avoid severe

side effect symptoms for patients, and it was effective in decreasing treatment-emergent adverse effects, but secondary efficacy endpoints favored the higher dose of cabazitaxel

(134).

Microtubule Transport Proteins

Microtubules are cytoskeletal polymeric structures present in all eukaryotic cells.

Composed of repeating α/β tubulin heterodimers that alternate in a line to form a protofilament, 13 protofilaments align in parallel to form a hollow tube (135).

Microtubules are extremely versatile, being about one hundred times stiffer than actin and intermediate filaments, but also flexible enough to bend laterally and grow in varied directions without breaking (135).

Motor proteins are categorized into three super-families that exhibit directed movement on either microtubule or actin filaments. The first are myosin motor proteins that move on actin filaments inside muscle cells, the second are dynein motor proteins that move on microtubules, and the third are kinesin motor proteins that also move on microtubules (136). All three families utilize the energy from ATP to power the structural changes necessary for movement (136). Kinesin and dynein motor proteins have been implicated in many signaling pathways involved in cancer progression (136).

Kinesin-1 was the first molecular motor protein discovered to move cargo along microtubules (137). There is a highly conserved, ~340 amino acid sequence shared by the majority of kinesin proteins that comprises the motor domain (138). Though the motor domain is conserved, there is high variability between protein association and cargo- binding domains that accounts for the high variability seen between individual kinesins

and their unique cellular function (138, 139). Some kinesins have been found to regulate microtubule dynamics with no movement function at all, even though they possess the conserved motor protein domain (kinesin-13), while other kinesins both move along microtubules and modulate microtubule dynamics (kinesin-8) (136). Kinesins move uni- directionally, the majority towards the (+) end of microtubules, that is, away from the microtubule organizing center (MTOC) that is anchored in the centrosome (136, 140).

However, the kinesin-C-terminal motor/kinesin-14 subfamily notably moves towards the

(-) end of microtubules unlike the rest of the kinesin superfamily (141).

a. KIFC1/HSET: KIFC1, also known as HSET, is one of the rare kinesin proteins that is a minus-end transporter, that is, moves towards the (-) end of microtubules

(141, 142). In mammalian cells, HSET is the only (-) directional, kinesin-14 family member (143). HSET plays many critical roles in mitosis and meiosis (144). Mitotic spindles change length rapidly when a cell is dividing, and HSET activity contributes to spindle length during cell division events (144). HSET overexpression results in very long mitotic spindles, while HSET RNAi reduced the average length of microtubules (144). In addition to manage spindle length in mitosis and meiosis, HSET is known to cross-link and slide microtubules as needed to increase microtubule length (144). The normal function for HSET is in the nucleus; experiments that mutate the nuclear localization signal (NLS) of HSET result in cytoplasmic localization due to failure of nuclear import protein importin

α/β to associate with the HSET NLS (144). Work done with HSET-GFP fusion proteins reveals nuclear localization under baseline conditions (144). The function of HSET in spindle morphogenesis is dictated by a Ran gradient within the nucleus to aid in chromosomal alignment, microtubule cross linking, and successful cell division (144, 145).

HSET has been found to be overexpressed in a variety of cancer cells, and is positively correlated with both the stage and malignancy of a tumor (145). In non-small cell lung cancer, HSET overexpression is associated with poor survival, and had dramatically increased expression when compared to paired normal tissue from the same patient (146). In gastric cancer, HSET protein expression is up-regulated in 37% of cases, and in esophageal squamous cell carcinoma HSET is upregulated in cancer tissue when compared to neighboring non-plastic mucosa (147). In breast cancer, HSET overexpression gives cells an inherent resistance to docetaxel chemotherapy (148). In addition, HSET overexpression was found in docetaxel-resistant tumor samples, suggesting that HSET expression may play a role in response to taxane chemotherapy (148,

149). Previous work has shown that cabazitaxel treatment lowers HSET mRNA and protein expression in vitro (112), indicating that cabazitaxel may be an optimal therapeutic for patients that have a high HSET expression.

b. KIF2C/MCAK: KIF2C, also known as mitotic centromere-associated kinesin (MCAK) is a member of the kinesin-13 subfamily, a non-motile kinesin that has a unique role within the cell of removing tubulin subunits from the microtubule end during mitotic progression (150-152). In normal cells, MCAK is involved in kinetochore attachments to the centromere during mitosis (152, 153). MCAK is regulated spatially by the Rac1-Aurora-A kinase pathway, which inhibits the microtubule depolymerizing activity of MCAK (154). Through Aurora-A-mediated phosphorylation, MCAK is inhibited at the leading edge of motile cells, thereby allowing microtubules to grow within the cell towards the direction of movement (154). MCAK promotes fast microtubule growth on a compliant 2D ECM surface, and slows microtubule growth on a compliant 3D

ECM surface (154). In prostate tumors, MCAK expression increases with each disease stage from prostate adenocarcinoma, more so in hormone-sensitive prostate carcinoma, and even higher in CRPC in a dose-dependent pattern (153). Because MCAK expression increases with increasing disease stage, there may be therapeutic potential for MCAK inhibition pending further investigative research (153).

c. Dynein motor proteins: Dyneins, like kinesins, motor in a unidirectional manner on microtubules to transport various cargo and organelles through the cell.

However, dynein motor proteins move towards the (-) end of microtubules, that is moving towards the MTOC (136). Unlike kinesins that move in a uniform 8nm step along one microtubule protofilament, dynein motors frequently step backwards and sideways around the surface of the microtubule (155-157). Dynein motor proteins transport cargo, and are important in AR transport from the cytoplasm to the nucleus (158). Co- immunoprecipitation analysis has shown AR-dynein intermediate chain association is increased upon R1881 (synthetic DHT) stimulation (158), indicating that the dynein motor protein is important in efficient AR transport from the cytoplasm into the nucleus under androgen stimulation.

Taxanes Mechanisms of Action beyond Microtubule Stabilization

Taxanes have many other effects on the cell aside from inducing mitotic arrest and apoptosis as a result of microtubule stabilization. Forkhead box proteins comprise a family of transcription factors that are a direct downstream target of the PI3K/Akt pathway and

increase the transcription of a wide range of genes including those affectingcell cycle inhibition, apoptosis, DNA repair and defense from oxidative stress (159). One forkhead box protein, FOXO1, is increased upon taxane treatment – this is critical in prostate cancer because FOXO1 is also a transcriptional repressor of AR (160). Through increasing a transcriptional repressor of AR, taxanes can decrease AR activity in a mechanism independent of androgen or endocrine signaling events (160). Paclitaxel has been shown to induce apoptosis in breast and ovarian cancer cells via inducing FOXO1 nuclear localization, causing upregulation of the pro-apoptotic factor Bim (159-161). FOXO3a, another forkhead box protein, is relocated into the nucleus upon paclitaxel treatment, a signaling event dependent on JNK and Akt signaling, and is likely a mediator in paclitaxel- induced apoptosis (161). FOXM1 is overexpressed in and correlated with docetaxel resistance in breast cancer, and gefitinib resistance in non-small cell lung cancer (162).

FOXM1 is a critical protein in resistance to docetaxel and paclitaxel through regulating the microtubule destabilizing protein Stathmin and changing the microtubule dynamic, rendering the cell less responsive to the chemotherapeutic treatment (162). Treatment with a FOXM1 inhibitor, thiostrepton, reverses taxane resistance, indicating a complicated, dynamic relationship between Forkhead box protein family and the efficacy of taxanes, and the potential for a combination treatment to optimize response to taxanes (162).

Work from out lab provided the first evidence that taxane chemotherapy prevents

AR translocation to the nucleus, resulting in AR cytoplasmic sequestration and inhibition of AR activity (Figure 1.4) (75). In a tissue microarray of clinical specimens from docetaxel and untreated patients, AR immunohistochemistry revealed a dramatically different AR localization in specimens that were treated with docetaxel (75). AR intensity

was not changed, but docetaxel treated patients had a 38% decrease in AR nuclear localization, bringing to light not only the importance of cellular localization for proper function but also novel effects of docetaxel treatment that is independent of mitotic arrest and apoptosis induction (75). This effect has been confirmed by other groups in circulating tumor cells from docetaxel-treated patients that showed decreased nuclear localization of

AR (158). Further analysis into this effect revealed that the AR’s critical NTD is necessary for docetaxel to inhibit AR translocation, which provides therapeutic potential for preventing AR-variant translocation in addition to full length AR (75).

Although taxanes are known for inducing apoptosis through microtubule stabilization and mitotic arrest, paclitaxel has also been shown to induce apoptosis through phosphorylation of the pro-survival protein Bcl-2 (163). Mechanistically, the transcriptional repressor of E-cadherin, TWIST, has been shown to mediate paclitaxel resistance through the Akt and Bcl-2 pathways (164). Bcl-2, as a functional mediator between taxane chemotherapy and apoptosis, with taxanes inducing a direct, pro-apoptotic effect on Bcl-2 but also Bcl-2 protecting the cell from taxane-induced-apoptosis (163, 164).

This is confirmed when the combination of paclitaxel and ABT-263 (venetoclax), a Bcl-2 inhibitor, increased therapeutic response to paclitaxel in vitro (165). In sum, taxanes have many other effects within the cell that may be “off-target” for a microtubule stabilizing drug, but these additional effects contribute to apoptosis in a multi-faceted manner.

Targeted Therapeutics in Prostate Cancer

Current therapeutic options for mCRPC are limited in efficacy and have high rates of therapeutic resistance. Options include androgen axis-targeting agents abiraterone and enzalutamide, microtubule stabilizing taxane chemotherapeutics cabazitaxel, radiotherapy

Radium-223, and limited immunotherapy sipuleucel-T (78, 166). A summary of the current therapeutics targeting cell death processes in prostate tumors is shown on Table 1.

a. Kinesin Targeted Therapeutics: In vitro evidence indicated that HSET knockout cell lines have growth inhibition, reduced cell cycle kinetics, altered cell membrane, aneuploidy, and a disordered chromatin density (145). HSET knockout cells have dramatically lower wound closure times, decreased colony formation, decreased spheroid formation, decreased cell proliferation, and disordered microtubule transport within the nucleus (145-147). All of these factors come together and paint a favorable picture for HSET inhibition as an attractive therapeutic target for a variety of cancers.

CW069, a novel HSET inhibitor, showed high selectivity for HSET in in vitro drug development, and decreases cell growth, centrosome clustering, centrosome amplification in cancer cells (167). In pre-clinical models of prostate cancer, where HSET was discovered to induce docetaxel resistance (149), CW069 is effective in both docetaxel- sensitive and resistant cell lines (168). Promising work shows that CW069 sensitizes docetaxel-resistant prostate cancer cell lines to docetaxel treatment, indicating that there is potential clinical value in such a drug to be used in monotherapy or in combination with taxane chemotherapy (168).

Early attempts to target MCAK via inhibitors have been done through kinesin spindle protein (KSP) inhibition, an approach that inhibits a wide range of kinesins

including MCAK and ten others (169). MCAK was identified as a target of synthetic sulfoquinovosylacylglycerols (SQAGs), compounds with an 18 carbon long fatty acid chain on the glycerol moiety, after SQAGs were found to induce S and M phase arrest in cultured mammalian cells (170). The other identified molecular target of SQAGs was

DNA polymerase, and narrowing the cytotoxic effect to MCAK inhibition versus the effect of DNA polymerase inhibition complicates any beneficial results from SQAG treatment

(170).

b. Anoikis Inducing Agents: Quinazoline-based a1-adrenoreceptor antagonists such as Doxazosin and Terazosin have been previously shown to induce anoikis in urologic cancers (171). Doxazosin has been heavily studied in prostate cancer and induces apoptosis by activating Caspase-3, which cleaves focal adhesion kinase (FAK)

(172). Active FAK suppresses anoikis, but cleaved FAK is non-functional and the cell undergoes anoikis (172). Studies from our group on structural optimization of Doxazosin lead to the generation of a lead derivative, DZ-50, which exerts a potent anti-tumor effect via anoikis induction in both prostate and renal cancers (171, 173). Induction of anoikis by DZ-50 was demonstrated by the loss of integrin-linked kinase and talin expression on prostate cancer lines PC3 and DU145 (49) leading to prostate tumor growth suppression and impairing metastasis in tumor models (49, 171). In renal cell carcinoma, DZ-50 and doxazosin treatment resulted in significant suppression of migration and invasion while quinazoline derivatives exert antiangiogenic properties via anoikis, impairing tumor vascularity and growth (171). Activation of p38 by Platycodin D activates p38 MAPK signaling and increases anoikis in other cancers (174). Another therapeutic approach for anoikis resistant metastatic renal cell carcinoma centers (mRCC) is based on the fact that

tyrosine receptor kinase B (TrkB) is commonly overexpressed in mRCC (175). Treatment of mRCC tumors using the first line drug sorafenib leads to the emergence of resistance in cells overexpressing TrkB; inhibiting TrkB sensitized cells to anoikis and sorafenib, with a therapeutic promise for anoikis-resistant tumors (176).

c. EMT-MET Interconversions Rule the Microenvironment: First line treatment for patients with CRPC is to administer taxane chemotherapy, a microtubule targeting chemotherapeutic drug that functions by stabilizing microtubules and inducing apoptotic cell death (177), as well as impairing AR activity and interfering with EMT (59).

The second generation taxane cabazitaxel, is clinically used only after patients have failed to respond to first generation taxanes treatment and hormonal therapy/second generation antiandrogens. Recent work from our lab identified the ability of cabazitaxel to reverse

EMT to mesenchymal-endothelial transition (MET) in both androgen-responsive and

CRPC prostate cancer cells (177) indicating that the taxane mechanism of action extends beyond mitotic arrest and apoptosis. Interconversion of EMT to MET and subsequently inducing tumor re-differentiation through cabazitaxel treatment could be utilized to overcome therapeutic resistance to both AR-targeted and microtubule-targeting taxane therapies (177). Combination of ADT with taxane chemotherapy into a fashioned chemohormonal therapy provides a therapeutic advantage to patients with advanced disease, compared to ADT alone (178) and calls for optimizing sequencing of the treatment regimens to increase survival. Recent optimization strategies include combination approaches of docetaxel with a Notch inhibitor to induce MET [88]. Inhibiting Notch causes deactivation of Snail, the transcriptional repressor of E-cadherin, thereby restoring transcription of E-cadherin, assisting the cancer cells in maintaining an endothelial

phenotype, metastatic cell reattachment, and promote MET (69). Selective targeting of the stromal cells in the tumor microenvironment has thus emerged as an attractive option for eradicating prostate cancer (179). Endothelial cells surrounding prostate tumors have been found to secrete growth factors such as IL-6 (180). Phenotypic changes in the landscape of the prostate microenvironment after castration engage signaling through IL-6, AR downregulation, and associated upregulation of TGF-β, to promote tumor cell survival and invasiveness (181). Blocking IL-6 signaling in co-cultured cancer epithelial cells and endothelial cells abrogated the effect of endothelial cells, enabling a new therapeutic platform (181).

Further exploitation of EMT-MET interconversion within the tumor microenvironment enables development of new therapeutic platforms for targeted therapies against advanced tumors (63). Local TGF-β signaling can induce distinct morphological changes in the tumor through the EMT regulators snail and twist (63).

Inhibiting TGF-β signaling presented with clinical benefit in patients involved in ongoing clinical trials (182) and in our preclinical models of advanced prostate cancer (205). In the prostate microenvironment, androgens secreted in an autocrine/paracrine manner, rather than endocrine, enable androgen dependent tumor growth after ADT (182). To combat this, the second generation antiandrogen, abiraterone is given because it inhibits androgens from both auto/paracrine and endocrine sources, including the adrenals (182).

d. Integrin-targeting agents: Integrins are not only involved in cell-cell adhesion, but also navigate several signal transduction networks from the tumor microenvironment to intracellular signals (34), thus becoming attractive targets for cancer therapeutics. Specifically, the αvβ3 and αvβ5 integrins have a pharmacologic inhibitor

cilengitide, designed to antagonize integrin signaling and disrupt cell motility through the microenvironment (34). However, in a mCRPC phase II clinical trial reported in 2011, cilengitide treatment did not meet standards for patient improvement (183). In this trial, the majority of patients had disease progression during drug treatment, hypothesized to be from cancer cells increasing integrin production and signaling in response to the drug treatment (183). Alternative integrin signaling-blocking therapies include monoclonal antibodies CNTO 95 and MEDI-522, in development for blocking integrin signaling in prostate cancer (184). Despite disappointing clinical trials of cilengitide, the αvβ3 integrin has been found necessary for prostate cancer metastasis to the bone, supporting integrin signaling as an attractive drug target if more effective therapies can be designed (182).

e. Anti-angiogenic agents: Increased expression of vascular endothelial growth factor A (VEGFA) and platelet-derived growth factor β (PDGFβ) in tumors regulate local blood flow to dividing cells and target organs (48). Sunitinib and sorafenib are two small-molecule tyrosine kinase inhibitors with potent anti-angiogenic properties via inhibiting VEGF receptor 2 and PDGF receptor β, depriving the tumor of signals that contribute to its survival (185). Bevacizumab is a monoclonal antibody which binds and neutralizes VEGF (186), and has been in clinical use through phase II and III clinical trials for the treatment of advanced renal cell carcinoma (52). Axitinib and pazopanib are newer anti-angiogenic drugs that work by inhibiting VEGFRs, causing decreased vascular growth, increased vascular permeability and tumor cell apoptosis (63, 70). Still in clinical trials, anti-angiogenic drugs aflibercept and cediranib are being tested in combination therapy with prednisone and/or docetaxel for treatment of mCRPC (182). The anoikis inducing agent DZ-50 was first demonstrated from this laboratory to impair angiogenesis

through downregulation of the angiogenesis modulator thrombospondin 1, further increasing the translational potential of this novel attractive therapeutic (6, 176).

f. Inhibition of PI3K/Akt/mTOR Survival Pathway: mTOR is an evolutionarily conserved serine/threonine kinase downstream of the PI3K/Akt survival pathway that has been implicated in human cancers, including prostate cancer.

Phosphorylation and aberrant activation of this pathway results in cell growth and proliferation without specific growth or survival signals, giving selective advantage to those cells and facilitating tumorigenesis (7). The imidazo-quinoline derivative NVP-

BEZ235, a dual PI3K/mTOR inhibitor, shows antitumor activity against prostate tumors through limiting vasculature leakage in the tumor tissue microenvironment and reducing tumor interstitial fluid pressure effects (187-189). NVP-BEZ235 is effective against prostate tumors regardless of PTEN status (188). Combination treatment with Akt inhibitor

MK-2206 and mTOR inhibitor MK-8669 (ridaforolimus) inhibits CRPC in a pre-clinical model (189-191). The Akt inhibitor MK-2206 is currently undergoing clinical trials in non-urologic cancers as a combination therapy with initial promising results (192).

g. Inhibition of TGF-β Signaling: TGF-β is a tumor suppressor in the early stages of prostate tumorigenesis, but a tumor promoter in advanced disease through multiple mechanisms (193, 194). TGF-β research increased in 1990 after a seminal study showed that pre-treatment of adenocarcinoma cells with TGF-β1 in vitro increased tumorigenic and metastatic characteristics when injected into rats, characteristics attributed to increased matrix metalloproteases at the time (195). TGF-β promotes tumorigenesis via induction of induction of EMT (196, 197). If TGF-β signaling is blocked, such as through a dominant-negative TGF-β Receptor II (DN-TGFβRII) mutation, TGF-β signaling

becomes a potent tumor promoter and tumors are much more aggressive in pre-clinical models of prostate cancer, supporting the role of TGF-β as an early-stage tumor suppressor

(198). Dysfunctional TGF-β signaling in in vivo models accelerates malignant changes in the prostate through EMT induction, increased inflammation, increased proliferation, and significantly increased AR expression (198). Clinically, TGFβRII mutations are more frequent than in the other TGF-β receptor, TGFβRI (199). Interestingly, DN-TGFβRII mutations increase TGF-β ligand mRNA expression, possibly through autocrine and paracrine feedback effects, where the increase in TGF-β ligand stimulate stromal cells and encourage angiogenesis, extracellular matrix degradation, and EMT, thus increasing tumor invasion and metastasis (198, 200).

For patients with advanced prostate cancer, inhibiting TGF-β signaling is an attractive therapeutic target. Galunisertib (LY2157299 monohydrate), a novel TGF-β receptor 1 (TGFBRI) inhibitor, is being investigated as monotherapy or in combination with standard therapies (201). Galunisertib has been shown to inhibit TGF-β effector protein (SMAD) phosphorylation and activation alone and in the presence of free TGF-β ligand (202). Galunisertib safety and efficacy has demonstrated in patients with hepatocellular carcinoma and glioblastoma (203, 204), and has been tested in pre-clinical models of prostate cancer (205).

The use of galunisertib for patients with prostate cancer shows promise after pre- clinical studies showed efficacy in combination therapy with enzalutamide (205).

Treatment with galunisertib increased epithelial phenotype, indicating EMT to MET reversion, causing re-differentiation of prostate tumors, marked by increased cytokeratin-

18 expression (205). Cofilin, a small protein regulator of actin cytoskeleton, decreased

with galunisertib treatment – the dramatic cofilin depletion compromises the integrity of the skeleton and may contribute to MET (73, 205). Galunisertib had no effect on AR nuclear localization or intensity in a TRAMP mouse model, and did not decrease prostate tumor weight in monotherapy, but increased therapeutic response to enzalutamide in pre- clinical models of advanced prostate cancer (205). TGF-β signaling in prostate cancer is a multi-faceted pathway with effects on the actin-cytoskeleton, cell phenotype, growth, invasion and metastatic ability. Galunisertib provides a promising potential therapeutic for patients with advanced prostate cancer, bypassing AR signaling. The re-differentiation via

MET also provides a therapeutic approach where galunisertib triggers a reversion to an epithelial phenotype, optimizing therapeutic response.

h. Natural Therapeutic Agents: Research of natural compounds as an alternative or an addition to traditional therapeutics is increasing. Extracts from the nettle leaf (Urtica Dioca) has been shown to increase caspase 3 and 9 mRNA levels, de-regulate apoptosis inhibitor Bcl-2, and inhibit proliferation in breast and prostate cancer cells in vitro (206). Two compounds, punicalagin and ellagic acid, found in pomegranate peel extract (Punica Granatum), can inhibit growth by inducing apoptosis in prostate cancer cells in vitro through loss of mitochondrial transmembrane potential (207). In addition, pomegranate peel extract inhibits proteins involved in metastasis through downregulation of matrix metalloproteinases MMP2 and MMP9, and upregulation of tissue inhibitor of metalloproteinase 2 (TIMP2) (207). Pomegranate fruit extract also contains antitumor effects in mouse models through increase of pro-apoptotic Bax and Bak expression, corresponding downregulation of pro-survival Bcl-X, Bcl-2, cyclins D1, D2, E, cyclin dependent kinases cdk2, 4, and 6 in PC3 cells (208). Outside of direct apoptosis inducers,

pomegranate fruit extract also decreased AR expression in a dose dependent manner in prostate cancer cell lines (208). Pomegranate fruit extract administered in drinking water significantly decreased CWR22 prostate cancer tumor volumes, decreased serum PSA expression, and delayed tumor onset in a xenograft mouse model of prostate cancer (208).

Terpenoids, a large class of natural compounds found in a wide range of fruit and vegetables, exhibit anti-oxidative and anti-inflammatory properties (209). In breast and prostate cancer cells, terpenoids have been found to activate apoptosis through Bcl-2, NF-

κB and ubiquitin-proteasome inhibition (209). It is notable that many of the terpenoids are not water soluble, which limits possible clinical application, and their pharmacokinetic characteristics have not yet been determined.

Conclusions

The therapeutic challenge in patients with advanced metastatic CRPC is to effectively increase survival more than a few months. At the molecular level, the sequencing of therapeutic targeting to reverse EMT to MET and induce anoikis in metastatic cancer cells might be an emerging paradigm to be exploited. While EMT and resistance to anoikis are currently a therapeutic challenge, it is a characteristic that separates cancer cells from normally functioning cells, and any process that is unique to cancer cells can be exploited for therapeutic use. Therapeutics that target EMT are in their infancy and need further development before a reliable, widely used therapeutic platform is constructed. EMT programming can bypass the cell’s ability to die upon detachment, causing a temporal interconversion of its phenotype extending survival, enter the bloodstream, metastasize, invade secondary tissues, and successfully colonize in the distant

site. The clinical benefits of exploiting the anoikis-EMT connection is prevention of metastasis before it can begin while overcoming therapy resistance. Recent advances in molecular profiling of prostate tumor landscape have defined several molecular signatures based on dissection of significant processes operational in prostate cancer cells. How can a therapeutic window in these temporal events then be capturing a subset of tumor epithelial cells ready to be targeted after the anoikis resistance offense is down regulated by the EMT effectors? Prostate cancer is one of the most lethal male cancers in the US today, but restoration of anoikis sensitivity and a reprogramming of EMT towards cell execution will aid in overcoming therapeutic resistance and increasing survival.

Inhibiting anoikis and related cellular events including EMT, increased motility and chemoresistance would restore a sort of neutrality to the tumor through prevention of disease spread. To block extravasation would mean blocking metastatic events before they can begin and keep disease states in a realm where therapies are more effective – treating the primary tumor rather than advanced disease. When metastatic events occur, almost all therapeutic regimes universally fail. The “holy grail” of effectively treating metastatic disease might be to prevent the metastic journey and cell colonization from occurring in the first place by overcoming resistance to anoikis via sensitizing the cells to EMT-MET reversal and acquisition of the epithelial phenotype.

Docetaxel has many effects on prostate tumor cells, including microtubule stabilization but also inhibits AR translocation and increases FOXO1 expression (160).

Cabazitaxel has the same primary mechanism of action through microtubule stabilization.

However, cabazitaxel shows efficacy in docetaxel resistant tumors, potentially through cabazitaxel’s unique effect on microtubule destabilizing kinesins MCAK and HSET. The

goal is to avoid the seemingly inevitable therapeutic resistance in CRPC patients that is one of the biggest challenges in prostate cancer research today. The scope of these studies towards this thesis is to determine new mechanisms driving therapeutic resistance in cabazitaxel chemotherapy in advanced prostate cancer. Cabazitaxel’s efficacy in the postdocetaxel therapeutic landscape stresses the clinical relevance of overcoming cabazitaxel resistance to avoid lethal disease. In the subsequent chapters of this thesis, I describe the results from studies investigating the molecular mechanisms of cabazitaxel resistance in in vitro and in vivo pre-clinical models of prostate cancer progression. Based on enhanced understanding of these novel mechanisms of resistance, I proposed new treatment schemes to optimize therapeutic response, and identified molecular targets that will enable new treatment therapeutic platforms for combination therapy to increase response and decrease resistance in patients with late stage prostate cancer. Table 1. Summary of PCa therapeutics and the cellular processes they target.

Therapeutics Targeted Process Quinazolines Anoikis Taxane Chemotherapy Apoptosis, EMT-MET ADT/2nd Generation Anti-Androgens Apoptosis, EMT Integrin Targeting Agents Anoikis Anti-Angiogenic Agents Anoikis, Autophagy PI3K/Akt/mTOR Inhibitors Apoptosis, Anoikis

Figure 1.1: Anatomy and histology of a normal prostate gland and BPH.

The prostate is located underneath the male bladder, and encloses the urethra. There are three distinct histological zones. Benign prostatic hyperplasia (BPH) classically originates in the transitional zone, and is defined by increased proliferation of both epithelium and stroma on histological staining. Figure from Chughtai et al., 2016. Nature Reviews. (3).

Figure 1.2: AR axis targeting therapeutics

The hypothalamus-pituitary axis controls testicular androgen synthesis by regulating the release of luteinizing hormone (LH) from the pituitary. LHRH analogs Leuprolide and

Goserelin disrupt this through suppression of LH release. Orchiectomy (surgical castration) removes testosterone production from the testes. Finasteride and dutasteride inhibit testosterone conversion to DHT in the prostate. Abiraterone and ketoconazole inhibit steroid biosynthesis. Enzalutamide (MDV3100) works within the cancer cell to inhibit AR activity. AR-variants, through losing the ligand binding domain, are inherently resistant to these AR therapies that target ligand production and blocking ligand binding.

Docetaxel chemotherapy in addition to target microtubule stabilization also prevents AR translocation to the nucleus and ultimately transcriptional activity; cabazitaxel 2nd line chemotherapy impacts the microtubule dynamics, and decreased microtubule destabilizing kinesin proteins with novel effects on AR binding (See Chapter 2).

Figure 1.3: Androgen Receptor and Variants Share the N- Terminal Domain.

The full length AR protein is translated from exons 1-8, containing the NTD, DNA binding domain (DBD), hinge domain, and ligand binding domain (LBD) that binds DHT. AR splice variants have modified or deleted domains. The most common deletion in AR- variants is the LBD domain, which renders the AR ligand independent and constitutively active due to no longer requiring DHT binding for activation. Modified from Van der Steen et al., 2013. International Journal of Molecular Sciences. (210).

Figure 1.4: Taxane Mechanism of Antitumor Action.

Docetaxel and cabazitaxel prevent microtubule depolymerizing into tubulin monomers.

Docetaxel prevents AR translocation to the nucleus, transported by a yet to be identified microtubule transport protein. Cabazitaxel blocks the activity and lowers protein level of microtubule-depolymerizing kinesins.

Figure 1.5: Effect of EMT to MET interconversion on anoikis sensitivity.

Normal epithelial cells are polarized with strong cell-cell and cell-ECM junctions, securing homeostatic control. When cells undergo EMT, they lose polarity and gain a mesenchymal phenotype via breakdown of E-cadherin mediated cell-cell interactions. EMT is consequential to increased TGF-β signaling, a change of AR expression, and integrin detachment. Reversal of EMT to MET calls for a remarkably dynamic and a highly transient phenotype; cells undergoing EMT and becoming mesenchymal upon detachment, can potentially resist anoikis, evade death and continue surviving and able to seed secondary metastasis.

Figure 1.6: EMT and anoikis are navigated by AR and TGF- β cross-talk signaling.

Signaling exchanges by TGF-β (Smad protein effectors) and AR induce EMT through activation of transcription factors (Snail and Twist) that transcriptionally repress E-

Cadherin expression. Degradation of cell-cell junctions consequential to the loss of E-

Cadherin, claudin and occludin lead to anoikis and apoptotic execution via FAS signaling and caspase-8 activation. FAK/PI3K signaling confers resistance to anoikis via Akt activation, caspase blocking and NFκB activation.

Figure 1.7: Chemical Structures of FDA-approved Therapeutics for Prostate Cancer Treatment

Chemical structures of select therapeutic agents FDA-approved (antiandrogens and taxane chemotherapeutics) used for the treatment of patients with advanced prostate cancer. (A) 2nd generation antiandrogens Enzalutamide, (B) Abiraterone acetate, (C) 1st line chemotherapy Docetaxel, (D) 2nd line chemotherapy Cabazitaxel. Red Circle emphasizes the structural difference between docetaxel and cabazitaxel.

Chapter Two: Impact of Androgens on Therapeutic Response to Chemotherapy in Models of Advanced Prostate Cancer

Background and Significance

In early stage, localized disease, patients are treated with radiation therapy and/or radical prostatectomy surgery. If disease spreads, patients with metastatic prostate cancer are treated with ADT to block the growth stimulating effect that circulating androgens have on prostate cancer (15). In some patients, ADT leads to the emergence of castration resistant prostate cancer (CRPC) and the cancer no longer responds to ADT. Targeting androgen signaling through combination of anti-androgens and microtubule targeted chemotherapy has survival benefits for patients with metastatic CRPC (mCRPC), but therapeutic resistance ultimately develops in a majority of patients, resulting in lethal disease (79-82). Therapy resistance can result from the addiction of CRPC cells to AR signaling and constitutive activation of ligand independent AR splice variants (83). Many of the AR variants have lost the ligand binding domain, conferring constitutive AR activity independent of ligand/androgen binding (100). The most clinically relevant AR variant,

AR-v7, has been linked to increased ADT resistance, decreased survival, and poor clinical outcomes; however, AR-v7 detection is not associated with primary resistance to taxane chemotherapy, indicating that AR-v7 positive patients may respond better to taxanes than to AR-targeting therapies (83, 100).

Taxane chemotherapy is the standard of care for patients with metastatic CRPC who develop resistance to second generation anti-androgens such as abiraterone or enzalutamide. For over a decade, taxanes were the only class of Food and Drug

Taxanes exert antitumor action by stabilizing β-tubulin microtubule subunits, preventing microtubule depolymerization which induces G2-M arrest and apoptosis (113). More recent evidence has shown that taxane chemotherapy exerts its antitumor effects in prostate cancer by impairing AR transport along the microtubules, resulting in AR cytoplasmic sequestration and inhibition of AR activity (59, 75).

Cabazitaxel has been shown to target and lower cellular level of microtubule- binding kinesin protein KIFC1 (HSET) (112). Increased HSET expression has been associated with docetaxel resistance (149), and recently a HSET inhibitor CW069 has been shown to induce apoptosis and re-sensitize resistant cells to docetaxel treatment (168).

Currently, the full role of HSET in both ADT and cabazitaxel treatment is poorly understood.

EMT more readily with a low AR content (59), indicating that there is multilayered complexity in the interaction between androgens and TGF-β towards EMT.

ADT induces EMT in a sub-population of prostate cancer cells (110, 111), and cabazitaxel chemotherapy reverses EMT to MET in prostate cancer cells in vitro and in vivo (112). In the present study we investigated the impact of cabazitaxel and anti-androgen treatment in combination and sequential administration, on prostate tumor cell dynamics.

The efficacy of sequencing therapies in both androgen responsive and CRPC pre-clinical models of advanced prostate cancer was geared towards improving and optimizing the therapeutic response of patients with lethal disease.

Experimental Approach

Cell Lines: The human prostate cancer cell lines, the castration-resistant cell line 22Rv1 and androgen-sensitive LNCaP cells were obtained from American Type Culture

Collection (ATCC, Manassas, VA). Cells were maintained in RPMI 1640 (Invitrogen,

Grand Island, NY) and 10% fetal bovine serum (FBS) (Invitrogen, Grand Island, NY),

100units/ml penicillin and 100µg/ml streptomycin in a 5% CO2 incubator at 37C. For experiments examining responses to androgen, cells were seeded in 10% charcoal-stripped serum and phenol-red free RPMI 1640 (Invitrogen, Grand Island, NY) and were stimulated every 24hrs with dihydrotestosterone (1nM) (DHT) (Sigma-Aldrich, St. Louis, MO). PC3-

CR (cabazitaxel resistant) cell lines were generated through incremental increase of cabazitaxel, starting with 0.5nM, over a period of 2 months.

Cell Viability Assays: The effect of the various treatments on prostate cancer cell viability was evaluated using the Thiazolyl Blue Tetrazolium bromide (MTT) assay. Cells were seeded into 24-well plates, grown to 50-65% confluence and treated. Following treatment, media was aspirated and cells rinsed with PBS then treated with 250μl/well MTT

(1mg/ml) for 30mins at 37°C. After incubation, MTT was aspirated and formazan crystal was solubilized with DMSO. Absorbance was measured at 570nm using μQuant

Spectrophotometer (Biotech Instruments Inc., Winooski, VT).

In vivo xenograft experiments: All animal experiments were performed in accordance with the guidelines approved by the Animal Care and Use Committee of the University of

Kentucky and according to the NIH (Bethesda, MD) recommendations and reporting standards. 6-8 week old male Hsd:Athymic nude Foxn1nu mice (Envigo-Harlan) were injected with 22g needle into the right flank of the mice with LNCaP (4x106 cells in 100uL media) or 22Rv1 (2x106 cells in 100uL media) mixed with 100uL matrigel (Corning

356230, Corning, NY). After tumors were palpable, mice were divided into four groups 5 mice/group: (i) control group receiving sham castration at start, and vehicle (1:1:18 v/v ethanol: polysorbate 80: 5% w/v glucose in sterile water) every 3 days via intraperitoneal

injection thereafter, (ii) surgical castration and cabazitaxel (3mg/kg every 3 days), (iii) sham castration and cabazitaxel (3mg/kg every 3 days), (iv) castration then cabazitaxel

(3mg/kg every 3 days) three days after castration. Prostate tumors were measured three times a week, and the volume was calculated using the formula length x width x 0.5236.

Mice were sacrificed with CO2 gas according to institutionally approved protocols and NIH guidelines, and tumors were excised.

Immunohistochemical Analysis: Tissue specimens from mouse xenograft tumors were formalin fixed and paraffin-embedded; sections (5µ) were subjected to immunohistochemical analysis using rabbit monoclonal antibodies. E-cadherin antibody was obtained from Cell Signaling Technology Inc. (Danvers, MA), rabbit polyclonal antibodies against nuclear antigen, Ki67 (cell proliferation marker) and N-cadherin were from Abcam

Inc. (Cambridge, MA); the rabbit polyclonal antibody against AR was obtained from Santa

Cruz Biotechnology (Santa Cruz, CA). Sections are exposed to specific primary antibody and immunostaining was detected by biotinylated goat anti-rabbit IgG and horseradish peroxidase–streptavidin conjugate (Millipore, Billerica, MA). Color development was performed using a FAST 3,3′-diaminobenzidine-based kit (Sigma–Aldrich) and counterstained with hematoxylin. Images are captured with an Olympus BX51 microscope

(Olympus America, PA). The number of Ki-67 immunoreactive positive cells was counted from three different fields per section and the average value was determined.

Apoptosis Detection: The incidence of apoptosis was evaluated in situ using the terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) assay (Millipore, Billerica, MA). Prostate tissue sections were counterstained with methyl green and TUNEL-positive cells were counted per high power field (40x10). Quantitative

analysis of apoptosis was done by counting the average number of positive cells from three different fields of view per section.

Co-immunoprecipitation: Cells were lysed using Pierce IP Lysis buffer (Thermo

Scientific, Rockford IL). Co-IP was performed using a Dynabeads Protein G kit, protocol, and reagents (Thermo Scientific, Rockford IL). Anti-KIFC1 (ab172620) (abcam,

Cambridge, MA) was used (at a concentration of 5µg/sample) for HSET pull downs.

Western Blotting: Protein samples were prepared using the NE-PER nuclear- cytoplasmic fraction kit (Thermo Scientific, Rockford, IL). Protein content was quantified using the Pierce BCA Protein Assay Kits (Thermo Scientific) and protein samples were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (4-15% SDS- polyacrylamide gels; Bio-Rad, Hercules, CA), and transferred to Hybond-C membranes

(Amersham Pharmacia Biotech, Piscataway, NJ). Following incubation with primary antibody (overnight at 4°C), membranes were exposed to species-specific horseradish peroxidase-labeled secondary antibody (Jackson ImmunoResearch Laboratories, West

Grove, PA). Signal detection was achieved with HyGLO Quick Spray Chemiluminescent

HRP Antibody Detection Reagent (Denville scientific, Metuchen, NJ).

Immunofluorescent Confocal Microscopy: Cells were plated (1x105) on chamber slides coated with fibronectin (Invitrogen). After 24-48hrs, or at about 40% confluence, cells were exposed to medium (RPMI 1640 with 10% CSS) in the presence of DHT (1nM, ethanol), cabazitaxel (50nM, ethanol) or in combination of the two agents. Following treatment, cells were fixed in 4% (v/v) paraformaldehyde and permeabilized with 0.1%

Triton X-100 in sterile phosphate buffer saline (PBS). Fixed cells were incubated overnight at 4°C with primary antibody. Fluorescent secondary antibody was applied with gentle

rocking and the appropriate Alexa-Fluor (Life Technologies, Grand Island, NY) fluorescent secondary (1.5 hrs, room temperature). Slides were mounted using Vectashield mounting medium with DAPI and were visualized using a FV1000 Confocal Microscope

(Markey Cancer Center Core, University of Kentucky) or Zeiss LSM780 (Microscopy

Core, Icahn School of Medicine at Mount Sinai).

Targeted mass spectrometry quantification of cabazitaxel: Mass spec analysis was done at Cedars Sinai Institute, Los Angeles, CA. Cell pellets were lysed by adding

100 μL dionized water to each sample, followed by sonication in a water-bath sonicator.

Two microliters of 50μg/mL cabazitaxel-d9 (Cayman Chemical, #28488) were spiked into each sample as an internal standard. After vortexing, 10 μL solution was used for protein concentration measurement using the Pierce 660 nm assay kit (Thermo Scientific) according to the manufacturer’s manual. To extract cabazitaxel, 1mL tert-butyl methyl ether was added to the remaining solution. The sample tubes were vortexed, incubated in a ThermoMixer at 1,200 rpm for 10 min at room temperature, and centrifuged at 16,000× g for 10 min at 4°C. The upper layer that contains cabazitaxel and cabazitaxel-d9 was dried down in a SpeedVac and reconstituted by 75 μL 50% methanol. To quantify cabazitaxel, a parallel reaction monitoring (PRM)-based targeted mass spectrometry assay (Dr. Wei

Yang’s laboratory; Cedars-Sinai Medical Center) was applied. Briefly, 20 μL solution was loaded onto a 15-cm Accucore Vanquish C18 column (Thermo Scientific, #27101-152130) and separated by ultraperformance liquid chromatography (UHPLC) at a flow rate of 0.3 mL/min using an Ultimate 3000 XRS (Thermo Scientific). The mobile phase A and B consisted of 2 mM ammonium formate and 100% acetonitrile, respectively. A liquid chromatography gradient of 50% B in 1 min, 50-85% B in 2 min, 85% B in 2 min, 85-50%

B in 0.1 min, and 50%B in 2.9 min. PRM analysis was conducted using an Orbitrap Fusion

Lumos Tribrid mass spectrometer (Thermo Scientific) operated in the negative ion mode.

MS1 scans were acquired with an Orbitrap resolution of 120,000 and a scan range of m/z

870-900. Target masses include m/z 880.36 and 889.414, which respectively correspond to cabazitaxel and cabazitaxel-d9 with a formate adduct. Target-triggered MS2 fragmentation was performed using higher-energy collisional dissociation (HCD) with collision energy of 30%. MS2 scans were acquired with an Orbitrap resolution of 60,000.

The PRM files were analyzed by Skyline (v4.2) to quantify the m/z 880.376/206.042 and

889.433/206.042 transitions, which are specific signatures for cabazitaxel and cabazitaxel- d9, respectively.

Statistical Analysis: The numerical data are analyzed for statistical significance using the un-paired student’s t-test by GraphPad Prism 8. Values are expressed as the mean ± standard error of the mean (SEM). Statistical differences were considered significant at p value of <0.05.

Table 2. List of primary antibodies used, Chapter Two. (IF=immunofluorescence, IP=immunoprecipitation, IHC=immunohistochemistry, WB=western blot). Target Host Uses Source, Catalog #

KIFC1/HSET Rabbit IF, IP, WB, IHC Abcam, ab172620

Androgen Receptor Mouse IF Thermo, MA5-13426

Androgen Receptor Rabbit WB, IHC Santa Cruz, sc-816

AR-v7 Rabbit IHC Revmab, 31-1109-00

Ki-67 Rabbit IHC Cell Signaling, D2H10

N-Cadherin Rabbit IHC Santa Cruz, sc-59987

E-Cadherin Rabbit IHC Cell Signaling, 24E10

E-cadherin Mouse IF Cell Signaling, 4A2

Results

Effect of androgens (DHT) on cell viability response to cabazitaxel chemotherapy in vitro:

We first investigated the role of androgens on the in vitro response to cabazitaxel, by performing time-course studies in three prostate cancer cell lines with different AR status (Figure 2.1, panels A, B and C). The addition of DHT (1nM) every 24 hours had a modest protective effect on CRPC 22Rv1 cells, (harboring full length AR but also variants

2, 3, 4, 5, 5/6, 7, and 120 (100) (Figure 2.1B). In the androgen-sensitive LNCaP cells, containing full length AR and AR variant 3, DHT supplementation had a great protective value from 48-96 hours of cabazitaxel treatment (100) (Figure 2.1A). In the AR-negative

PC3 cells, DHT also had a great protective effect from 24-96 hours (Figure 2.1C). A dose response of PC3 cells to cabazitaxel with and without DHT (1nM) revealed a delay in cytotoxicity and a protective effect provided by DHT supplementation. However, in PC3-

CR, a cabazitaxel resistant cell line developed in our lab, there is no protective effect of

DHT (Figure 2.2A). Interestingly, DHT does not have the same protective effect on PC3 cells when treated with 1st generation taxane, docetaxel (Figure 2.2B). There is no statistically significant increase in viability in PC3 or PC3-CR cells between charcoal stripped serum (CSS) and 1nM DHT supplementation in neither a dose response (Figure

2.3A) nor the time course analysis (Figure 2.3B). These data indicate that DHT is interfering with cabazitaxel efficacy in a mechanism independent of AR binding and growth stimulation through downstream transcription products of the androgen response element.

Treatment sequencing of ADT and cabazitaxel modifies the tumor dynamics in LNCaP and 22Rv1 prostate tumor xenografts:

In order to determine pre-clinical efficacy of combining ADT and sequencing therapy by giving ADT first to prime the tumor for cabazitaxel induced MET, we investigated the effect of androgens on cabazitaxel chemotherapy in in vivo xenograft models of prostate cancer. 20 male, 6-8 week old nude mice were randomized into 4 treatment groups, as indicated in the treatment scheme (Figure 2.4A). There was no significant change in tumor volume (cm3) between treatment groups in either LNCaP

(Figure 2.4B) or 22Rv1 tumors (Figure 2.4C).

Treatment with cabazitaxel three days post-ADT led to an increase in apoptosis in androgen sensitive LNCaP cells (Figure 2.5A, B left); but not in castration resistant CRPC

22Rv1 cells (Figure 2.5C, D left). As shown on Figure 2.5 (panel B), each treatment reduced proliferation as measured by Ki67 staining in LNCaP cells, although sequencing cabazitaxel after ADT had the biggest effect (Figure 2.5B, right). In 22Rv1 xenograft models, treatment administration led to reduced cell proliferation when compared to control, but sequential application of taxanes and ADT had no effect when comparing the treatment groups to each other (Figure 2.5D, right). Representative images of TUNEL

(apoptosis detection) and Ki67 staining are shown for the LNCaP and 22Rv1 xenograft tumor sections (Figure 2.5, panels A, C respectively). Previous work from our group established that cabazitaxel induces reversal of EMT to MET (112), consistent with our current observations in LNCaP tumors treated with cabazitaxel, as detected by increased

E-cadherin expression (Figure 2.6 A). Moreover there was a parallel decrease in

mesenchymal marker N-cadherin expression in cabazitaxel treated tumors when compared to control (Figure 2.6A). In CRPC 22Rv1 tumors, cabazitaxel does not induce E-cadherin expression, but instead causes decrease in both E and N cadherins (Figure 2.6B). ADT causes an increase in mesenchymal marker N-cadherin in CRPC 22Rv1 tumors (Figure

2.6B), indicating the differential response of CRPC to treatment compared to androgen- sensitive LNCaP xenografts.

Novel targetable interaction between HSET and AR by cabazitaxel:

Cabazitaxel treatment given alone or simultaneously with ADT reduced the number of cells with nuclear AR positivity (Figure 2.7A, left). Nuclear intensity was measured via H-Score [as previously described, (211)] and treatment did not induce a marked change in nuclear AR intensity (Figure 2.7A, right). In 22Rv1 xenografts, only cabazitaxel and

ADT given together resulted in a significant decrease in AR immunoreactivity (Figure

2.7B, left). Nuclear AR intensity was widely distributed in 22v1 tumors, with nearly 50% of positive cells at intensity of 1. Cabazitaxel treatment decreased the proportion of cells that had a low intensity score, while increasing the proportion of high intensity scores

(Figure 2.7B, right). LNCaP tumors had a heterogeneous response to ADT when imaging for AR intensity, with some tumors showing little to no AR expression and others with unchanged AR expression (Figure 2.7C). 22Rv1 CRPC xenograft tumors showed consistent AR intensity without any effect by ADT (Figure 2.7D). There was increased detection of AR V7 however in 22Rv1 CRPC in all three treatment groups compared to control (Figure 2.7E).

HSET expression in in vivo xenograft tumors:

We previously identified HSET kinesin (KIFC1) as a molecular target of cabazitaxel (112); in subsequent studies I subsequently examined the effect of cabazitaxel treatment on HSET expression in prostate cancer xenografts. Analysis of the TCGA database (Figure 2.8A) revealed a correlation between HSET mutations in prostate cancer with poor survival in Kaplan-Meier survival estimates. This implicates a potentially important role of HSET in lethal prostate cancer. Cabazitaxel monotherapy increased

HSET positive cells (Figure 2.8B, D left), in contrast with the cabazitaxel-induced decrease in HSET levels observed consistently in vitro (112). Combination of cabazitaxel and ADT decreased HSET positive cells in vivo (Figure 2.7D, left). In 22Rv1 xenografts, there is no change in HSET positivity in any treatment groups (Figure 2.7D, right).

Immunohistochemical images of HSET nuclear localization and intensity in LNCaP xenografts (Figure 2.7B) and 22Rv1 xenografts (Figure 2.7C) are shown at 20x (top) and

100x (bottom) magnification.

HSET and HSET/AR interaction are molecular targets of cabazitaxel:

In vitro confocal microscopy of LNCaP cells show a substantial decrease of HSET in cabazitaxel treated (under castrate conditions), with a DHT induced protective effect in

CBZ treated cells in the presence of DHT (Figure 2.9A). E-cadherin shows a slight increase in intensity and a more uniform, epithelial cell shape after cabazitaxel treatment, indicating MET has occurred (Figure 2.9A, top row). The results from confocal microscope analysis shown on Figure 2.9 (Panel B), indicate that LNCaP cells in CSS there

increased AR expression in response to DHT (as expected), and reduced AR expression when cells were treated with cabazitaxel. Cells exhibited high HSET or AR, and in the presence of androgens (DHT) there was an increased association of AR-HSET.

Significantly, cells positive for AR were not affected by cabazitaxel induced multinucleation nor were undergoing apoptosis. Co-localized nuclei with AR/HSET are highlighted with a yellow arrow (Figure 2.9B, bottom right).

Previous work has shown cabazitaxel treatment decreases HSET expression in

LNCaP cells (112). In the present studies our findings in the PC3 androgen-independent cells resonate with the original evidence (Figure 2.10A). This association was further supported by co-immunoprecipitation of LNCaP lysate; As shown on Figure 2.10, panel

B, the interaction between AR and HSET, is partially blocked partly by cabazitaxel treatment, after 24hrs (50nM) and blocked completely at 48 and 72 hours of treatment in

FBS cultured cells (Figure 2.10B). In castrate (CSS) conditions, there is no detectable AR-

HSET binding. After 48hours of exposure to DHT (1nM), the AR-HSET association was detected, but it was antagonized by cabazitaxel (50nM) (Figure 2.10C).

At a lower dose of cabazitaxel, there is still a cytotoxic delay when DHT is present.

LNCaP cells treated with cabazitaxel (35nM) with and without DHT, exhibit a delayed loss of cell viability under androgenic conditions (DHT is present) (Figure 2.11A). Mass spectrometry analysis shows that there is no statistical change in intracellular cabazitaxel level when LNCaP cells are treated with cabazitaxel (50nM) with or without (1nM) (Figure

2.11B). A proposed schema of the results is shown (Figure 2.12A).

Conclusions

The emergence of cabazitaxel as a 2nd line taxane chemotherapy effective postdocetaxel was an important step forward to combatting CRPC and improvement in patient survival (114). While cabazitaxel exerts the same “classic” stabilization of microtubules as docetaxel, additional mechanisms have been revealed as drivers of therapeutic response

(114). Recently, growing molecular evidence has revealed targetable mechanisms contributing to therapeutic resistance to docetaxel in CRPC including ERG (212), and

Notch/Hedgehog signaling pathways (213, 214). The pre-clinical efficacy of cabazitaxel for prostate cancer was first determined in AR negative cell line DU-145 (114), and has later been established in a panel of prostate cancer cell lines with different AR profiles

(112). One of the major clinical challenges in the treatment of advanced metastatic prostate cancer, is the understanding that despite the monumental involvement of AR and androgen signaling as drivers of prostate cancer progression, in the clinical setting the AR status does not single handedly determine sensitivity to cabazitaxel (112). Significantly the presence of AR-v7, a splice variant commonly associated with advanced disease, is not associated with primary resistance to taxane chemotherapy (83, 215). Our findings are in accord with evidence reported by other investigators that the therapeutic response to cabazitaxel is significantly modified by the presence of testosterone in AR-containing, hormone responsive, pre-clinical models of prostate cancer (216).

Sequencing cabazitaxel after ADT in androgen responsive LNCaP models had a more advantageous effect than in the 22Rv1 CRPC model. While tumor volume showed no significant changes between treatment groups (Figure 2.3B-C), there is a shift in the biology within the tumor towards MET in the androgen sensitive LNCaP model. These

findings provide proof of-principle that that sequential treatment with taxane chemotherapy is more effective earlier on in prostate cancer progression to optimize therapeutic response.

The 22Rv1 cell model is more representative of CRPC and our data indicate that ADT and cabazitaxel combination fails to provide enhance efficacy. In contrast, in the LNCaP

(androgen sensitive) model we found a significant increase in apoptosis and decrease in proliferation when cabazitaxel is given immediately following ADT.

We speculate that ADT is “priming” the cell for cabazitaxel to induce apoptosis, before resistance to ADT and EMT occurs. In addition, cabazitaxel-induced MET (177) can work in tandem with ADT to decrease cancer progression in androgen sensitive models. We thus propose that cabazitaxel and ADT sequencing would be more effective if given earlier in the treatment scheme for patients, prior to the emergence of CRPC. In the STAMPEDE trials, patients with metastatic disease (550 men) who started docetaxel therapy a median of 9 weeks after beginning ADT, showed a 22% reduction in risk of all- cause death and a median of 10 months increased survival(126). In patients with non- metastatic disease (31 men), no benefit was determined by adding docetaxel after ADT

(126). These STAMPEDE results indicate that 1st generation taxane docetaxel is effective in metastatic disease. Our data indicates that there is potential therapeutic value in attempting a similar trial with ADT and cabazitaxel in patients to test for therapeutic value.

Our data indicate that DHT is interfering with cabazitaxel efficacy, potentially via a mechanism independent of AR, gain support from studies by an independent investigative team (216). DHT is interfering with cabazitaxel efficacy in the AR-null PC3 prostate cancer cell line, indicating that the mechanism is more complicated. The lack of response in cabazitaxel resistant PC3-CR cells to DHT indicates that developing cabazitaxel

resistance includes breaking away from the effect that DHT has on the cell. Aside from

AR binding, DHT has been shown to protect against apoptosis induction in glial cells (217), and induce Akt phosphorylation (217, 218). Considering the recent evidence (216, 219,

220), one may argue that DHT is interfering with cabazitaxel entry into the cell.

My work led to the first evidence on the interaction of AR and the kinesin

KIFC1/HSET in prostate cancer cells. AR has been previously shown to be transported by

KIF5B (221), and that KIFC1/HSET is downregulated by cabazitaxel (177), but cabazitaxel has never been shown to influence the AR-Kinesin interaction. This is of translational significance as it provides a new targeting platform in the AR targeting axis, and a druggable protein interaction independent of the action of the current therapeutics available. Patients with mutated HSET have lower progression free survival in the TCGA prostate tumor database (Figure 2.7A), indicating that altered HSET contributes to more advanced disease progression. Preventing AR and AR-variant cellular trafficking by inhibiting kinesins, or by blocking interaction, could be a way of blocking AR activity in an ADT-independent manner. Ongoing studies are examining whether the AR variants interact with kinesins, and the druggable value of such an interaction to impact tumors harboring AR variants (222). The schematic diagram on Figure 2.12 summarizes the consequences of cabazitaxel treatment on protein effectors and the potential targetable interactions between protein effectors that beg for therapeutic exploitation to overcome resistance in lethal disease. Impairing AR and AR-variant nuclear transport and activity by blocking the kinesin interaction, may lead to a new avenue of blocking AR CRPC driver action. Ongoing studies are examining whether the AR variants interact with kinesins, and the therapeutic value of compromising such an interaction to impact tumors harboring AR

variants. My studies provide new insights into the impact of treatment sequencing strategies to overcome therapeutic resistance in advanced prostate cancer. Our findings identified a) the antagonistic effect of androgens in the antitumor action of cabazitaxel that is overcome by ADT; and b) a novel association, between AR and kinesin HSET, which is targeted by cabazitaxel. This evidence is of translational significance in targeting AR-

HSET interactions to improve therapeutic response and survival in patients with lethal disease.

66 67

Figure 2.1: DHT increases prostate cancer cell viability in vitro: Dose response studies. (A) Time course of LNCaP cell viability in response to cabazitaxel (50nM) in the presence or absence of DHT (1nM). (B) Time course of 22Rv1 cell death response after cabazitaxel

(50nM) treatment in the presence or absence of DHT (1nM). (C) Time course response of

PC3 cells to cabazitaxel (50nM), in the presence or absence of DHT (1nM). Values represent the average of three independent experiments performed in duplicate. (*=p<0.05,

**=p<0.005, ***=p<0.0005).

67 68

Figure 2.2: DHT affects cabazitaxel efficacy in AR-negative cells. (A) Dose response of PC3 and PC3-CR (cabazitaxel resistant) cells to increasing doses of cabazitaxel, in the presence or absence of DHT (96 hours). (B) Dose response of PC3 cells to docetaxel with and without DHT (1nM) over 96 hours. All experiments done in phenol- red free media supplemented with charcoal serum stripped FBS. Values represent the average of three independent experiments performed in duplicate. (*=p<0.05,

**=p<0.005, ***=p<0.0005).

68 69

Figure 2.3: DHT monotherapy does not affect cell viability in PC3 cells. (A) PC3 and PC3-CR dose response to increasing doses of DHT over 96 hours. Results show 2 trials, experiment was completed in phenol-red free RPMI with charcoal stripped

FBS. (B) Time course of PC3 cells with no treatment and with DHT (1nM) over 96 hours.

All experiments done in phenol-red free media supplemented with charcoal serum stripped

FBS. Values represent the average of three independent experiments performed in duplicate. (*=p<0.05, **=p<0.005, ***=p<0.0005).

69 70

Figure 2.4: Prostate tumor xenograft response to cabazitaxel and ADT in vivo. (A) Treatment scheme for mice (n=5/group). Mice were treated with surgical castration

(ADT) and cabazitaxel (CBZ, 3mg/kg I.P.) for 14 days. (B, C) Tumor volumes (cm3) of

LNCaP (A) and 22Rv1 (B) xenografts, averaged within treatment groups (n=4).

70 71

71 72

Figure 2.5: Effect of cabazitaxel and ADT on prostate tumor growth.

(A). Representative images (20X magnification) of LNCaP prostate xenograft tumors after

TUNEL (top) and Ki-67 staining (bottom). Scale bar indicates 100 µm. (B) Left – LNCaP

TUNEL positive cells were quantified. Right – LNCaP Ki67 positive cells were quantified

TUNEL (top) and Ki-67 staining (bottom). Scale bar indicates 100 µm. (D) Left – 22Rv1

TUNEL positive cells were quantified. Right – 22Rv1 Ki67 positive cells were quantified.

(All counting done at 40X, 3 fields/tumor, 2 independent observers. *=p<0.05,

**=p<0.005).

72 73

73 74

74 75

Figure 2.6: Reversal of EMT to MET in advanced prostate tumor xenografts, androgen sensitive LNCaP and CRPC 22Rv1.

(A) Profiling protein markers of EMT Representative images (20X magnification) of

LNCaP xenograft tumors after epithelial marker E-cadherin staining (top) and mesenchymal marker N-cadherin (bottom). Cabazitaxel treatment increasing E-cadherin expression and decreasing N-cadherin expression is indicative of EMT reversal into MET.

(B) Representative images (20X magnification) of CRPC 22Rv1 xenograft tumors after E- cadherin staining (top) and N-cadherin (bottom). Representative images from 4-5 tumors per treatment group. Cabazitaxel induced MET is not present in 22Rv1 xenograft tumors as it is in LNCaP tumors (A). Scale bar indicates 100 µm.

75 76

76 77

Figure 2.7: Effect of ADT and cabazitaxel on AR expression in prostate tumor xenografts. (A) Left – LNCaP AR positive cells were quantified. Right – LNCaP nuclear AR strength was quantified by H-Score. (B) Left – 22Rv1 AR positive cells were quantified. Right –

22Rv1 nuclear AR strength was quantified by H-Score as previously described (211). (C)

Characteristic images of AR immunoreactivity in LNCaP tumor xenografts, magnification at 20x. Heterogeneity within CBZ + ADT and CBZ post ADT treatment groups is shown with 2 images vertically aligned. Scale bar indicates 100 µm. (D) AR immunohistochemical detection in prostate tumor sections shown at 20x. Scale bar indicates 100 µm. (E) 22Rv1 AR-v7 immunohistochemistry, shown at 20x and 100x magnification. Treatment with cabazitaxel increases AR-v7 expression in monotherapy and in combination with ADT, while sequencing cabazitaxel post ADT has a diminished increase in AR-v7 expression. Shown, representative images from 4-5 tumors per treatment group. Scale bar indicates 100 µm (20x images) and 20µm (100x images).

*=p<0.05, **=p<0.005.

77 78

78 79

79 80

Figure 2.8: Significance of HSET Kinesin in Prostate Cancer.

(A) HSET mutations in the prostate ate TCGA database decreases progression free Kaplan-

Meier estimate. Accessed via cBioPortal on November 25, 2019, querying 1329 samples.

Mutations were found in 4 of the queried samples. (B) HSET immunohistochemistry for

LNCaP treated xenografts, 20x and 100x magnification. (C) HSET immunohistochemistry for 22Rv1 treated xenografts, 20x and 100x magnification. (D) Left – HSET positive cells were quantified in LNCaP tumors. Right – HSET positive cells were quantified in 22Rv1 tumors. (*=p<0.05, error bars represent SEM). Representative images from 4-5 tumors per treatment group. Scale bar indicates 100 µm (20x images) and 20µm (100x images).

80 81

81 82

Figure 2.9: Effect of cabazitaxel treatment on kinesin and AR expression and localization.

(A) Confocal microscopy analysis of HSET and E-Cadherin (100x magnification) of

LNCaP cells treated with cabazitaxel (50nM), DHT (1nM), or combination for 48 hours and imaged for E-cadherin as marker of EMT (Alexafluor 594), HSET (Alexafluor 488), and DAPI. Cabazitaxel treatment caused a decrease in HSET expression, which was restored with DHT. (B) Confocal microscopy images (100x) of LNCaP cells in charcoal stripped FBS, phenol-red free media treated with 50nM cabazitaxel, DHT (1nM), or combination for 48hours and imaged for AR (Alexafluor 594), HSET (Alexafluor 488), and DAPI. Arrows highlight cells with AR/HSET co-localization. Nuclei are positive for either HSET or AR, but not both, except in DHT monotherapy treated cells.

82 83

83 84

84 85

Figure 2.10: HSET and HSET/AR interaction are molecular targets of cabazitaxel. (A) Cabazitaxel (50nM) decreases HSET expression over 72 hours of treatment in PC3, in

FBS supplemented media. (B) HSET co-immunoprecipitation of LNCaP cells treated with

50nM cabazitaxel, in FBS supplemented media. (C) HSET co-immunoprecipitation of

LNCaP cells in CSS supplemented, phenol-red free media treated with cabazitaxel (50nM) and DHT (1nM); the protein profile shows the effect of cabazitaxel on reducing the association between the two proteins.

85 86

Figure 2.11: DHT challenges cabazitaxel effect in vitro in prostate cancer cells. (A) 1nM DHT increases cell viability in LNCaP cells treated with cabazitaxel (35nM). (B)

LNCaP cells treated with cabazitaxel (50nM) with and without DHT (1nM) for 48hours were analyzed for intracellular cabazitaxel level.

86 87

Figure 2.12: Interface of cabazitaxel and ADT on AR and HSET cellular distribution

(A) Scheme of combining ADT and cabazitaxel and the effects on select intracellular elements. Cytosolic AR binds DHT and translocates to the nucleus, where it aids transcription of AR related genes, including downstream effects on growth, proliferation, and EMT. AR-HSET associate upon DHT stimulation, blocked by cabazitaxel, with unknown effects on translocation. Cabazitaxel lowers intracellular level of HSET independently of AR-HSET interaction. ADT blocks both extracellular testosterone (T) and intracellular DHT. Testosterone enters the cell through SLCO1B3 transporter, and cabazitaxel is proposed to enter the cell through the same transporter.

87 88

88 89

Chapter Three: Androgen Independent Contributor to Prostate Cancer Therapeutic Resistance Significance

Therapy resistant prostate cancer is a complex, multi-faceted disease. In the previous chapter, I discussed the validity of therapeutic sequencing to optimize response to cabazitaxel chemotherapy, and uncovered the effect of circulating androgens on cabazitaxel efficacy. In this chapter, I explore a potential mechanism if DHT interferes with cabazitaxel chemotherapy, uncovered by the cabazitaxel resistant cell line PC3-CR.

The membrane transporter protein SLCO1B3, also referred to as OATP1B3, was identified to be involved in cabazitaxel resistance through analysis of cabazitaxel resistant cell line

PC3-CR. SLCO1B3 has been identified within recent years to be involved in docetaxel as well as testosterone cellular uptake (216, 219, 223, 224).

Bcl-2 (B-cell lymphoma 2) is a pro-survival protein that has been identified in prostate cancer therapeutic resistance (36, 225). Bcl-2 is found in the basal epithelium of the prostate, and is closely tied to androgen signaling (226, 227). ADT has been shown to increase Bcl-2 expression in the prostate, directly contributing towards therapeutic resistance and androgen independent disease (227-229). Overexpression of bcl-2 as a critical apoptosis suppressor and driver of prostate cancer apoptosis evasion has also been shown to be associated with progression from androgen-dependent disease to the more aggressive androgen-independent disease stage (230). Targeting Bcl-2, through both pharmacologic inhibition and altering protein levels, is of great interest in prostate cancer.

Research has shown that decreasing Bcl-2 expression through antisense transcript increases

89 90 apoptosis and sensitizes in vitro prostate cancer cells to chemotherapy (231). For radiation therapy, the Bcl-2 ratio with its pro-apoptotic family member Bax was predictive of a patients’ response to radiation (226) and Bcl-2 overexpression decreases apopotosis when prostate cancer cells were exposed to radiation (225). Pharmacologic inhibition of Bcl-2 increased efficacy of docetaxel therapy in prostate cancer cells through increasing apoptosis induction, indicating the therapeutic value of combination therapy for resistant disease (232).

To interrogate mechanisms of clinical significance contributing to therapeutic resistance of prostate cancer patients with advanced disease we utilized pre-clinical, PDX models from patients with metastatic CRPC resistant to cabazitaxel in collaboration with by Dr. Nora Navone and Chris Logothetis at MD Anderson Cancer Center, Houston TX

(233, 234). These PDX samples are from patients who have metastatic CRPC, and have been treated with radiation, ADT, and chemotherapy including 2nd line taxane chemotherapy, cabazitaxel. In the following study the tumors are profiled for EMT phenotype, and expression and cellular localization of HSET kinesin, and SLCO1B3 transporter.

Our hypothesis is that the membrane transporter protein SLCO1B3 and kinesin

HSET expression and cellular distribution are affected by androgens and can impact cabazitaxel uptake, ultimately affecting drug resistance and therapeutic response among selected prostate tumor cell populations. In the following chapter I described the results of the investigation of alternative mechanisms of resistance in cabazitaxel chemotherapy, potential therapeutic value of sequential targeting of AR and microtubules as a mechanisms of overcoming treatment resistance in advanced prostate tumors.

90 91

Approach

Cell Lines: The human prostate cancer cell lines, PC3 (androgen independent, AR- negative), androgen-sensitive LNCaP cells were obtained from American Type Culture

Collection (ATCC, Manassas, VA). Cells were maintained in RPMI 1640 (Invitrogen,

Grand Island, NY) and 10% fetal bovine serum (FBS) (Invitrogen, Grand Island, NY),

CR (cabazitaxel resistant) cell lines were generated through incremental increase of cabazitaxel, starting with 0.5nM, over a period of 2 months.

(1mg/ml) for 30mins at 37°C. After incubation, MTT was aspirated and formazan crystal was solubilized with DMSO. Absorbance was measured at 570nm using μQuant

Spectrophotometer (Biotech Instruments Inc., Winooski, VT).

In vitro Cell Migration and Invasion: Prostate cancer cells was analyzed via wound assay. Cells (104 cells/well) were seeded in 6 well plates, allowed to grow to 60-70% density of cell monolayer and a wound was induced using a pipet tip. After wounding, cells were exposed to cabazitaxel. The number of migrating cells was counted in three different fields,

91 92 under microscopic examination. Invasion potential was evaluated using a Biocat Matrigel

Transwell Chamber (Beckon Dickinson, Franklin Lakes, NJ). Cells were seeded into the upper chamber of a transwell insert pre-coated with matrigel in serum-free medium (50,000 cells/well). After 24hrs non-invading cells were removed from the upper chamber and invading cells were stained with Diff-Quick Solution (IMEB Inc., San Marcos, CA).

In vivo Xenograft Models of Advanced Prostate Cancer: All animal experiments were performed in accordance with the guidelines approved by the Animal Care and Use

Committee of the University of Kentucky and according to the NIH (Bethesda, MD) recommendations and reporting standards. 6-8 week old male Hsd:Athymic nude Foxn1nu mice (Envigo-Harlan) were injected with 22g needle into the right flank of the mice with

PC3 cells (4x106 cells in 100uL media). After tumors were palpable, mice were divided into two groups 5 mice/group: (i) control group receiving vehicle (1:1:18 v/v ethanol: polysorbate 80: 5% w/v glucose in sterile water) every 3 days via intraperitoneal injection thereafter, (ii) cabazitaxel (3mg/kg every 3 days). Prostate tumors were measured three times a week, and the volume was calculated using the formula length x width x 0.5236.

Mice were sacrificed with CO2 gas according to institutionally approved protocols and NIH guidelines, and tumors were excised.

Immunohistochemical Analysis: Tissue specimens from mouse xenograft tumors were formalin fixed and paraffin-embedded; sections (5µ) are subjected to immunohistochemical analysis using rabbit monoclonal antibodies. E-cadherin antibody was obtained from Cell Signaling Technology Inc. (Danvers, MA), rabbit polyclonal antibodies against nuclear antigen, Ki67 (cell proliferation marker) and N-cadherin were from Abcam

92 93

Inc. (Cambridge, MA); the rabbit polyclonal antibody against AR was obtained from Santa

(Olympus America, PA). The number of Ki-67 immunoreactive positive cells was counted from three different fields per section and the average value was determined.

Grove, PA). Signal detection was achieved with HyGLO Quick Spray Chemiluminescent

93 94

HRP Antibody Detection Reagent (Denville scientific, Metuchen, NJ).

Immunofluorescent Confocal Microscopy: Cells were plated (1x105) on chamber slides coated with fibronectin (Invitrogen). After 24-48 hrs, or at about 40% confluence, cells were exposed to medium (RPMI 1640 with 10% CSS) in the presence of DHT (1nM, ethanol), cabazitaxel (50nM, ethanol) or in combination of the two agents. Following treatment, cells were fixed in 4% (v/v) paraformaldehyde and permeabilized with 0.1%

Triton X-100 in sterile phosphate buffer saline (PBS). Fixed cells were incubated overnight at 4°C with primary antibody. Fluorescent secondary antibody was applied with gentle rocking and the appropriate Alexa-Fluor (Life Technologies, Grand Island, NY) fluorescent secondary (1.5 hrs, room temperature). Slides were mounted using Vectashield mounting medium with DAPI and were visualized using a FV1000 Confocal Microscope

(Markey Cancer Center Core, University of Kentucky) or Zeiss LSM780 (Microscopy

Core, Icahn School of Medicine at Mount Sinai).

Quantitative real-time reverse transcription polymerase chain reaction analysis: Molecular profiling was conducted using real-time polymerase chain reaction

(PCR) analysis of prostate cancer cells. Total RNA was extracted from prostate cancer cells using TRIzol reagent (Invitrogen/Life Technologies) and was subjected (1 mg) to reverse transcription into complementary DNA using a reverse transcription kit from Promega

(Madison, WI) under the following conditions: 25°C for 10 minutes, 42°C for 60minutes, and 95°C for 5minutes. Primer pairs and TaqMan probes (Applied Biosystems, Foster City,

CA) were used to determine Snail (SNAI1, Hs00195591_m1) Slug (SNAI2,

Hs00950344_m1), and E-cadherin (CDH1, Mm01247357_m1) gene expression. Primer pairs to assess Twist-1 messenger RNA (mRNA) expression were designed using Primer

94 95

Express software (Applied Biosystems, Branchburg, NJ). The following primers were used: Twist-1, forward, 50CCGGAGACCTAGATGTCATTGTT-30, reverse, 50-

AGTTATCCAGCTCCAGAGTCTCTAGAC-30. Real-time PCR was conducted in an

ABI Prism 7300 system (Applied Biosystems, Branchburg, NJ) in TaqMan Fast universal

PCR master mix (Life Technologies, Austin, TX) and Power SYBR Green PCR master mix (Life Technologies, Woolston Warrington, UK). Numerical data (normalized to 18S mRNA and b-actin levels) indicate mean values 6 standard error of the mean (SEM; n =

4–6).

RNA-sequencing analysis: RNA-seq analysis was done using the Human Clariom-S array analysis with the assistance of the University of Kentucky Microarray Core Facility.

Total RNA was extracted from prostate cancer cells using TRIzol reagent (Invitrogen/Life

Technologies) and recommended protocol for RNA-isolation. Pathway association studies were completed using NIH DAVID software KEGG pathway association analysis.

Targeted Mass spectrometry Quantification of Cabazitaxel: Mass spec analysis was done at Cedars Sinai Institute, Los Angeles, CA. Cell pellets were lysed by adding

100 μL dionized water to each sample, followed by sonication in a water-bath sonicator.

Two microliters of 50 μg/mL cabazitaxel-d9 (Cayman Chemical, #28488) were spiked into each sample as an internal standard. After vortexing, 10 μL solution was used for protein concentration measurement using the Pierce 660 nm assay kit (Thermo Scientific) according to the manufacturer’s manual. To extract cabazitaxel, 1 mL tert-butyl methyl ether was added to the remaining solution. The sample tubes were vortexed, incubated in a ThermoMixer at 1,200 rpm for 10 min at room temperature, and centrifuged at 16,000× g for 10 min at 4°C. The upper layer that contains cabazitaxel and cabazitaxel-d9 was dried

95 96 down in a SpeedVac and reconstituted by 75 μL 50% methanol. To quantify cabazitaxel, a parallel reaction monitoring (PRM)-based targeted mass spectrometry assay established by

Dr. Wei Yang’s laboratory at Cedars-Sinai Medical Center was applied. Briefly, 20 μL solution was loaded onto a 15-cm Accucore Vanquish C18 column (Thermo Scientific,

#27101-152130) and separated by ultraperformance liquid chromatography (UHPLC) at a flow rate of 0.3 mL/min using an Ultimate 3000 XRS (Thermo Scientific). The mobile phase A and B consisted of 2 mM ammonium formate and 100% acetonitrile, respectively.

A liquid chromatography gradient of 50% B in 1 min, 50-85% B in 2 min, 85% B in 2 min,

85-50% B in 0.1 min, and 50%B in 2.9 min. PRM analysis was conducted using an Orbitrap

Fusion Lumos Tribrid mass spectrometer (Thermo Scientific) operated in the negative ion mode. MS1 scans were acquired with an Orbitrap resolution of 120,000 and a scan range of m/z 870-900. Target masses include m/z 880.36 and 889.414, which respectively correspond to cabazitaxel and cabazitaxel-d9 with a formate adduct. Target-triggered MS2 fragmentation was performed using higher-energy collisional dissociation (HCD) with collision energy of 30%. MS2 scans were acquired with an Orbitrap resolution of 60,000.

The PRM files were analyzed by Skyline (v4.2) to quantify the m/z 880.376/206.042 and

889.433/206.042 transitions, which are specific signatures for cabazitaxel and cabazitaxel- d9, respectively.

96 97

Table 3. List of primary antibodies used, Chapter Three. (IF=immunofluorescence, IP= immunoprecipitation, IHC=immunohistochemistry, WB=western blot). Target Host Uses Source, Catalog #

KIFC1/HSET Rabbit IF, IP, WB, IHC Abcam, ab172620

E-Cadherin Rabbit IHC Cell Signaling, 24E10

E-cadherin Mouse IF Cell Signaling, 4A2

SLCO1B3 Rabbit IHC, WB Abcam, ab224064

Vimentin Rabbit IHC Cell Signaling, D21H3

Cytokeratin-18 Rabbit IHC Abcam, ab189444

Β-Tubulin Mouse IF Santa Cruz, sc-5274

97 98

Results

PC3 prostate cancer xenografts (androgen-independent), treated with either vehicle or cabazitaxel for 14 days showed a statistically significant decrease in tumor growth in cabazitaxel treated mice (Figure 3.1A). Cabazitaxel treated tumors had significantly less vimentin expression, indicating that treatment induced MET and decreased mesenchymal characteristics (Figure 3.1B). In addition to decreasing mesenchymal markers, cabazitaxel treatment induced apoptosis induction in PC3 xenograft tumors significantly (Figure 3.1C).

HSET kinesin protein, in the control tumors, was located primarily in the nucleus, with occasional cells having positive staining in the cytosol; however, in cabazitaxel treated tumors, HSET became localized primarily in the cytosol (Figure 3.2A). This is in contrast to in vitro data in LNCaP and 22Rv1 cells previously discussed in Chapter 2. However, in

PC3 and PC3-CR cells, cabazitaxel treatment for 48 hours (50nM) causes HSET to become localized to the cytosol in PC3 cells. In cabazitaxel resistant PC3-CR cells, HSET retains a cytosolic localization in both untreated and treated cells (Figure 3.2B).

Cabazitaxel resistant cells, PC3-CR, were developed from parental PC3 cells through continual exposure to increasing doses of cabazitaxel. PC3-CR cells have increased cell viability when treated with cabazitaxel for 96 hours when compared to PC3 cells (Figure 3.3A). In addition, PC3-CR cells have an inherent cross resistance to docetaxel treatment as shown in Figure 3 (Panel B). Interestingly, PC3-CR cells have a more mesenchymal phenotype when compared to the epithelial PC3 parental cells, indicating that the cells have undergone EMT as part of the emerging cabazitaxel resistance

(Figure 3.3C). Upon undergoing EMT, PC3-CR cells exhibit acquisition of migratory

98 99 characteristics. PC3-CR cells have increased migration at 24 and 48 hours when compared to parental PC3 cells, and maintain migratory qualities when treated with cabazitaxel

(50nM and 100nM) (Figure 3.4A). While PC3 cells are not invasive, PC3-CR cells have an average of over 250 invasive cells per matrigel membrane after 48 hours (Figure 3.4B).

RNA-seq analysis (Human Clariom S) comparing PC3-CR to PC3 gene expressions show changes in untreated gene expression between the two cell lines (Figure

3.5A). Transcriptional repressors of E-Cadherin (emphasized with blue boxes) had variable responses; ZEB1, Slug (SNAI2), and Snail (SNAI1) decreased in expression, while Twist (TWIST1) and ZEB2 increased in expression. E-cadherin (CDH1) expression was unchanged in 2 replicates and decreased in one replicate, and N-Cadherin (CDH2) expression decreased, contradicting the phenotypic EMT that the cell shape indicated.

Pathway analysis, completed using NIH DAVID KEGG pathway association showed PC3-

CR cells had increased DNA repair pathways (mismatch repair, homologous recombination), cell cycle, oxidative phosphorylation, and pathways in cancer when compared to PC3 cells (Figure 3.5B).

The RNA seq analysis revealed downregulation of the membrane transporter

SLCO1B3 in the cabazitaxel-resistant prostate cancer cells compared to the PC3 parental control cells (Figure 3.5A). Previous evidence has demonstrated that SLCO1B3 is responsible for the transport of docetaxel and testosterone into cells (120, 216, 235).

SLCO1B3 RNA was confirmed to be decreased in PC3-CR cells via qRT-PCR (Figure

3.6A), and protein level is non-detectable via western blot (Figure 3.6B). I hypothesize that the effect that DHT has on PC3 cells in vitro, through increasing cell viability despite

PC3 being an AR-negative cell line, is due to PC3 cells possessing the SLCO1B3

99 100 transporter while PC3-CR cells do not (Figure 2.2A). Clinically, SLCO1B3 alterations lead to decreased progression free Kaplan-Meier estimate, and decreased overall survival

(Figure 3.7A) in the TCGA prostate cancer tumor database. Alterations in SLCO1B3 are most commonly a shallow deletion (light blue), and a deep deletion (dark blue), with rare amplification (red) within the TCGA database (Figure 3.7B) contributing to the decreased months progression free and decreased survival.

For functional analysis, I performed a knockdown of SLCO1B3 in PC3 cells in order to determine the functional consequences of the transporter loss on cabazitaxel resistance. Stable shRNA knockdowns were generated, with a shCONTROL vector and 3 shSLCO1B3 constructs selected. All three shSLCO1B3 cell lines generated had decreased

SLCO1B3 protein level when analyzed via western blot (Figure 3.8A). Dose response treatment for a period of 96 hours, demonstrated that shSLCO1B3-C statistically significant increase in cell viability to increasing doses of cabazitaxel (Figure 3.8B), compared to vector control and shSLCO1B3-A cells. Unexpectedly, DHT presence in

PC3 cells significantly increased cabazitaxel uptake into the cell, and knockdown of

SLCO1B3 (shSLCO1B3-C) also significantly increased cabazitaxel uptake (Figure 3.8C).

In the untreated controls LNCaP xenograft tumors (Figure 2.3), there was a defined membrane localization of SLCO1B3 in vivo. In response to treatment (either with cabazitaxel as a monotherapy or and cabazitaxel and ADT expression of SLCO1B3 was detected as diffused distribution, with modest changes in the cellular localization of the protein with a faint cytosolic localization and no clear membrane boundary (Figure 3.9A).

In vitro confocal microscopy of LNCaP cells shows a similar effect. Untreated (in castrate conditions) LNCaP cells have a membrane and cytosolic localization of SLCO1B3, with

100 101 cabazitaxel treatment slightly lessening signal intensity but not entirely abrogating the signal (Figure 3.9B). Cabazitaxel treatment (50 nM) in the presence of DHT resulted in

SLCO1B3 protein cytosolic localization without traces of membrane distribution. E-

Cadherin, shown in green, is increased in cabazitaxel treated cells and shows that LNCaP cells have undergone MET in response to cabazitaxel treatment (Figure 3.9B).

B-cell lymphoma 2 (Bcl-2) overexpression in PC3 cells changes response to taxane chemotherapy in vitro. First, a panel of prostate cancer cell lines was profiled for

Bcl-2 expression (Figure 3.10A), and only PC3-bcl2 cells had detectable Bcl2 expression.

Bcl-2 overexpression in PC3 cells conferred statistically significant increase in viability/decrease in cell death when treated with docetaxel or paclitaxel chemotherapy dose response studies (as expected) (Figure 3.10B). In contrast, Bcl-2 overexpression sensitized prostate cancer cells PC3 cells to cabazitaxel in a dose response study on cell viability (Figure 3.10C). Interestingly enough, Bcl-2 expression increased E-cadherin expression in these cells as detected by western blot (Figure 3.10D), suggesting a phenotypic reversal of EMT to MET conferred by Bcl-2; moreover there was increased expression of pro-mitotic kinesin proteins HSET and MCAK in prostate cancer cells overexpressing bcl-2.

To interrogate mechanisms of clinical significance contributing to therapeutic resistance of prostate cancer patients with advanced disease we obtained PDX samples established by Drs. Nora Navone and Christopher Logothetis at MD Anderson Cancer

Center, Houston TX (233, 234). The clinical origin and the pathological characteristics of prostate cancer specimens from which the PDX models were established and used in our analysis, are summarized on Figure 3.11. Eight samples were studied, from five, n=5)

101 102 patients, four of which received cabazitaxel chemotherapy. Patient 5 has four samples obtained longitudinally over time (Patient 5a, 5b, 5c, and 5d), before and after cabazitaxel treatment. The clinicopathological features including tumor site, diagnosis, treatment status, and therapies received are summarized on Figure 3.10.

Two patients, patient 1 and 2, had adenocarcinoma of the prostate. Patient 1, with

PDX established from a bone metastasis, was treatment naïve, and is being used as a control for this sample set. Patient 2, with PDX established from peritoneal fluid, in contrast, received multiple therapies including prostatectomy surgery, radiation, continuous use of antiandrogen, and docetaxel with carboplatin followed by cabazitaxel. The tumor from

Patient 3, classified as small cell carcinoma, was isolated from a lymph node, and was heavily treated with radical prostatectomy, radiation, and antiandrogens. Patient 4’s PDX was isolated from circulating tumor cells (CTC), and is pathologically adenocarcinoma.

Patient 4 was treated with antiandrogens, radical prostatectomy surgery, then cabazitaxel with carboplatin. Patient 5 had four PDX tumors established from his disease, the first one

(5a) isolated from pelvic soft tissue and was classified as malignant epitheliod and spindle cell neoplasm with osteoid differentiation. At this point in time, patient 5 had received antiandrogens and a VEGF inhibitor. After sample 5a was isolated, sample 5b was isolated from the same patient. Patient sample 5b was isolated from the prostate, after the patient was treated with second generation antiandrogen abiraterone and cabazitaxel with carboplatin, and then with DNA targeting agents, and microtubule targeting agent vincristine. Samples 5c and 5d were isolated in the same surgery, 5c was extracted from the bladder and 5d was removed from a lymph node. Both tumors are classified as adenocarcinoma with sarcomatoid and small cell carcinoma. A summary table of all

102 103 known information about patient samples from PDX tumors can be found in Figure 3.11.

First, a profile for EMT markers was completed on the PDX tumors. Patients 1 and 4 had substantial E-cadherin staining (Figure 3.12A), with patient 2 having consistent but weaker positivity. Patients 3 and 5 (5a-d) had detectable but extremely weak and diffuse E- cadherin staining, suggesting these patients had more mesenchymal tumors.

Complementary staining for mesenchymal marker vimentin showed a similar effect

(Figure 3.12B). Patients 1 and 4, with strong E-cadherin, had very weak vimentin staining.

Patients 2 and 5 (5a-d) had strong positive staining for vimentin, further indicating mesenchymal tumors. Interestingly, patient 5 had increasing vimentin intensity in later samples (5c and 5d) indicating that this patient’s tumor became more mesenchymal in phenotype as the patient underwent further therapies. Patient 3, with small cell carcinoma, was positive in neither epithelial marker E-Cadherin nor mesenchymal marker vimentin.

Patients 1 and 4 had the most epithelial character, perhaps due to patient 1 being treatment naïve. Patient 4 could possess strong epithelial markers due to MET from extensive cabazitaxel with carboplatin chemotherapy (4 cycles) before sample collection or because the PDX was generated from circulating tumor cells and not a solid tumor site.

Cytokeratin-18, a marker of cellular differentiation, was strongly present in the epithelial tumors from patients 1 and 4, and weakly in patient 2 (Figure 3.13). The co-positivity of cytokeratin 18 and E-cadherin indicates that EMT reduces cellular differentiation as cells become more mesenchymal and mobile in nature.

We found that the prostate tumor specimens from all the PDX models exhibited strong immunoreactivity for nuclear HSET kinesin (Figure 3.14). Patient 3 had the weakest staining, but it was widespread, and there is no visible change in the prevalence of HSET

103 104 positive cells between the different tumors. HSET is not altered appreciably between epithelial and mesenchymal tumors, and that different disease stages does not alter HSET expression visibly. Strong staining for the novel transporter protein, SLCO1B3, was detected (in two) of the cabazitaxel resistant tumors with extremely weak, diffuse staining in the other tumors (Figure 3.15). The strongest SLCO1B3 expression was found in patient

4, an epithelial phenotype tumor isolated from circulating tumor cells.

104 105

Conclusions

Cabazitaxel treatment significantly decreased growth in PC3 xenograft tumors.

Tumors, along with a statistically significant decrease in growth, showed EMT reversal into MET indicated by decreased levels of the mesenchymal marker vimentin. The EMT reversal is indicative of changes in tumor biology independent of tumor volume, but related to restoration of an epithelial phenotype that could increase sensitivity to treatment.

Corresponding with the decreased tumor size, cabazitaxel-treated tumors showed an increase in apoptosis, fitting with cabazitaxel primary mechanism of action, apoptosis induction through microtubule stabilization.

Cabazitaxel treatment of PC3 xenografts resulted in nearly complete cytosolic localization of HSET. Since HSET is the only known kinesin in human cells that moves cargo towards the (-) end of microtubules, that is, towards the microtubule organizing center near the nucleus, locking HSET in the cytosol instead of its normal nuclear location would have dramatic effects on HSET functionality (141, 142). It is possible that the reason HSET is responding to cabazitaxel differently in PC3 xenografts than in the LNCaP and 22Rv1 xenografts (where HSET was exclusively nuclear in localization) is because

PC3 cells do not express any androgen receptor. The AR-HSET interaction discovered in

Chapter 2 could have more mechanistic consequences in the cell, with AR transport providing a critical role for HSET in the nucleus – without AR, HSET could lose a nuclear role. Further research into this hypothesis is necessary.

The cabazitaxel resistant PC3-CR cell line provided a valuable model for investigating cellular differences acquired with cabazitaxel resistance. Importantly, PC3-

105 106

CR cells were found to be locked in a mesenchymal phenotype, with different cell shape, increased migration, and dramatically increased invasion. The observation that PC3-CR cells exhibit cross-resistance to docetaxel (1st line taxane chemotherapy) treatment (Figure

3.3B) supports the functional overlap in terms of taxane mechanism of action with cabazitaxel, the 2nd line taxane chemotherapy. PC3-CR cells also exhibited a cytosolic

HSET localization, rather than nuclear, as in the cabazitaxel sensitive PC3 cells. RNA sequencing comparing the sensitive and resistant PC3 cells identified membrane transporter SLCO1B3, a protein implicated in androgen and docetaxel uptake, as heavily downregulated in PC3-CR cells (216, 224). SLCO1B3 has not been investigated in the context of cabazitaxel resistance, until our novel findings on SLCO1B3 knockdown conferred statistically significant increases in viability when treated with cabazitaxel in vitro (Figure 3.8B). Alterations in SLCO1B3 decrease progression free survival and overall survival in the TCGA database, a relevant clinical trial despite the limited sample size on the database (Figure 3.7A). In the future, with larger database availability and accessibility increasing, a more extensive analysis of SLCO1B3 protein expression in prostate cancer is necessary.

Finding that SLCO1B3 knockdown increased cabazitaxel cellular uptake in LNCaP cells was a paradox. It is possible that SLCO1B3 knockdown caused the cells to upregulate other membrane transporter proteins that therefore caused an increase in cabazitaxel uptake, and further research is needed to determine the role of SLCO1B3 in cabazitaxel uptake. Additionally, DHT presence triggered a paradoxical increase in cabazitaxel uptake, in contrast to predictions based on previous research. Once again, determining the complicated relationship between DHT, cabazitaxel, and cellular uptake is needed. LC-

106 107

MS/MS was an incredibly invaluable tool to collect these data, and the data was normalized to microgram of lysed intracellular protein. It would be interesting in the future to normalize the amount of cabazitaxel to the number of cells present in the sample to determine if there is discrepancies in the data due to cell lysis.

Despite the information gained from PC3-CR cells, it is not a model without weakness. Cell authentication revealed that while PC3-CR cells are identical in cell markers to the parental PC3 line, the cell line became contaminated with mouse origin materials, indicative of contamination with a mouse origin cell line. The data collected from the PC3-CR cell line will need to be validated in the future with different cell lines.

An interesting and paradoxical result was found with Bcl-2 overexpression in prostate cancer cell lines. PC3-bcl2, a cell line overexpressing Bcl-2, was found to be resistant to first line taxane chemotherapies docetaxel and paclitaxel. Being a pro-survival, anti-apoptotic protein (36, 225, 230), it was surprising to find that PC3-bcl2 cells were more sensitive to second line taxane cabazitaxel (Figure 3.10C). This effect contradicts the main body of literature, and needs to be investigated further in more diverse models. It is possible that, because Bcl-2 inhibitors sensitize human prostate cancer cells to docetaxel

(124, 125), that Bcl-2 inhibition would be even more effective for cabazitaxel chemotherapy and is an attractive target for combination therapy to overcome cabazitaxel resistance.

The patient-derived xenograft tumors allowed a momentary snapshot into the biology of advanced prostate cancer that has progressed through cabazitaxel chemotherapy in clinical settings. One of the most interesting conclusions from the PDX results is how the tumors, based on staining, fit into either an epithelial or a mesenchymal phenotype.

107 108

Patient 1, who was treatment naïve, had an epithelial marker positive tumor, with well- defined E-cadherin expression and no detectable vimentin. Similarly, patient 4 had an epithelial phenotype tumor. The other tumors had weak or no E-cadherin expression, and a corresponding positivity in mesenchymal marker vimentin. Cytokeratin-18, a marker for cellular differentiation, corresponded with E-cadherin expression, further supporting that the prostate tumors from patients 1 and 4 are of glandular epithelial phenotype.

Furthermore the SLCO1B3 membrane localization correlated with epithelial character as well. Patients 1 and 4 had membrane localization, with patient 4 having a cytosolic presence as well. Patient 5 had a transient (5b only) cytosolic presence, but as with the other mesenchymal tumors, did not have membrane localization of SLCO1B3. These data support the hypothesis that SLCO1B3 functionally contributes to maintaining an epithelial phenotype, perhaps through inducing MET, or if an epithelial cell has more SLCO1B3 expression (and membrane localization) as a consequence of changes in cell polarity. This work also supports a new avenue for investigation of a potential therapeutic value of

SLCO1B3, towards overcoming therapeutic resistance and a druggable protein to target in order to improve clinical outcome in prostate cancer patients with lethal disease. Ongoing studies focus on establishing a functional link between SLCO1B3 transporter and the phenotypic configuration of prostate cancer cells. This will validate my findings in profiling the distribution of the transporter in PDX prostate tumor specimens from treatment-resistant human tumors as a novel mechanism to induce MET in mesenchymal, invasive prostate tumors towards acquisition of re-differentiation properties.

108 109

Figure 3.1: Cabazitaxel inhibits androgen independent prostate tumor xenograft growth and increases tumor apoptosis. (A) Male, nude, SCID mice were injected with PC3 human prostate cancer cells and treated with either cabazitaxel (3mg/kg) or vehicle (n=5) for 14 days. A statistically significant decrease was found in tumor growth in cabazitaxel treated mice. (B)

Immunohistochemical staining of mesenchymal marker vimentin in control/vehicle and cabazitaxel treated xenograft tumors. Scale bar indicates 100 µm (20X images) and 20 µm

(100X images). (C) Cabazitaxel treatment induced statistically significant increase in apoptosis in PC3 xenograft tumors. Each point is representative of one tumor, with an average of 6 fields of view quantified at 20X magnification. *=p<0.05, **=p<0.005 students t-test.

109 110

110 111

Figure 3.2: Cabazitaxel treatment alters HSET localization in androgen-independent prostate cancer cells. (A) Immunohistochemical staining of kinesin protein HSET in control/vehicle and cabazitaxel treated xenograft tumors. Scale bar indicates 100 µm (20X images) and 20 µm

(100X images). (B) Immunofluorescence imaged with confocal microscopy (100X) of β- tubulin, HSET, and DAPI on PC3 and cabazitaxel resistant PC3-CR cells treated with control or cabazitaxel (50nM) for 48 hours, FBS. HSET is nuclear in untreated PC3 cells, and cytosolic in untreated PC3-CR cells. Cabazitaxel treatment alters HSET localization in PC3 cells, but not in PC3-CR cells. Scale bars indicates 10 µm.

111 112

112 113

Figure 3.3: EMT phenotype in cabazitaxel resistant cells (PC3-CR)

(A) Dose response of parental PC3 and cabazitaxel resistant PC3-CR cells to cabazitaxel,

96 hours. (B) Dose response of PC3 and PC3-CR cells to docetaxel, 72 hours. (C)

Brightfield images of live, cultured PC3 and PC3-CR cells (10X magnification).

***=p<0.005, students t-test.

113 114

114 115

Figure 3.4: Cabazitaxel resistance is associated with migratory and invasive properties in prostate cancer cells.

(A) The migratory potential of PC3 and PC3-CR cells was assessed via wound healing assay after exposure to cabazitaxel treatment for 24 and 48 hours, 0nM, 50nM, and 100nM.

Cells migrating into the wound were counted on three different fields of view. (B) Invasive potential of untreated PC3 and PC3-CR cells were assessed via matrigel invasion assay for

24 hours. Membranes were counted on three different fields of view.

115 116

Figure 3.5: RNA profile of cabazitaxel resistance. (A) Human Clariom S RNA sequencing analysis of PC3 and PC3-CR cells. Cadherin proteins (Red box) E-cadherin (CDH1) and N-cadherin (CDH2). Transcriptional repressors of E-cadherin are shown with blue boxes. (B) NIH DAVID KEGG pathway association analysis, fold enrichment compares PC3-CR expression to parental PC3.

116 117

117 118

Figure 3.6: Loss of SLCO1B3 expression in cabazitaxel resistant cells. (A) mRNA expression of SLCO1B3 was evaluated by RT-PCR analysis. qRT-PCR results revealed that untreated PC3-CR cells exhibited total loss of mRNA expression for the transporter compared to PC untreated parental cells (normalized to 18s). (B) Western blot analysis of SLCO1B3 protein expression in untreated PC3 and PC3-CR cells.

***=p<0.0005, student’s t-test.

118 119

Figure 3.7: SLCO1B3 alterations correlate with survival: TCGA analysis.

(A) Progression free Kaplan-Meier estimate (left) and overall survival Kaplan-Meier estimate (right) of patients with alterations in the SLCO1B3 gene (red) compared to no alterations (blue). (B) SLCO1B3 expression, done by RNA Seq, in 2015 TCGA prostate database (left), TCGA PanCan prostate database (middle) and TCGA prostate database

(right). Gray circles indicate no SLCO1B3 mutation, light blue indicates a shallow copy deletion, royal blue indicates a deep copy deletion, red indicates copy number amplification.

119 120

120 121

Figure 3.8: SLCO1B3 loss increases both cabazitaxel uptake and resistance to cabazitaxel. (A) Stable knockdown of SLCO1B3 using shRNA in PC3 cells decreases SLCO1B3 protein expression, (compared to shCONTROL). (B) Cell viability dose response analysis of stable knockdown PC3-shSLCO1B3 when treated with increasing doses of cabazitaxel for 96 hours. (C) Intracellular concentration of cabazitaxel measured via mass spec analysis and normalized to protein content in three cell lines, PC3 (black), PC3- shCONTROL (blue), and PC3-shSLCO1B3-C (green) treated with cabazitaxel (50nM) and

DHT (1nM). *=p<0.05, **=p<0.005, student’s t-test.

121 122

122 123

Figure 3.9: Treatment with cabazitaxel and ADT decreases membrane localization of SLCO1B3.

(A) Immunohistochemistry of SLCO1B3 in LNCaP xenograft tumors. Representative images shown at 20X and 100X magnification. Scale bar indicates 100 µm (20X images) and 20 µm (100X images). (B) Immunofluorescence of E-cadherin (green) and SLCO1B3

(red) in LNCaP cells treated with cabazitaxel (50nM), DHT (1nM) or combination for 48 hours shows SLCO1B3 cytosolic localization after treatment. Scale bar indicates 20 µm.

123 124

124 125

Figure 3.10: Bcl-2 overexpression sensitizes PC3 prostate cancer cells to cabazitaxel. (A) Western blot analysis shows Bcl-2 protein expression in a panel of human prostate cancer cell lines. (B) Cell viability dose response assay of PC3 and PC3-bcl2 cells to docetaxel and paclitaxel for 72 hours, demonstrates resistance to treatment, and (C) cabazitaxel for 96 hours, showing increased sensitivity to cabazitaxel. (D) Western blot analysis of PC3 and PC3-bcl2 cells, bcl2 overexpression shows increased bcl2, E-cadherin, and kinesins HSET and MCAK. **=p<0.005, ***=p<0.0005, student’s t-test.

125 126

126 127

Figure 3.11: Patient derived xenografts (PDX) models from cabazitaxel treated, metastatic CRPC.

(A) Table of de-identified information of prostate cancer tumors from patients treated with cabazitaxel. Sample date indicates the date at which the tumors were frozen, not collected. Tumor site indicates the location from which the sample was collected.

Patient 5 has four samples collected from three different surgeries.

127 128

Figure 3.12: Epithelial and mesenchymal phenotype in PDX tumors.

(A) Immunohistochemical staining of PDX tumors for E-cadherin, shown at 20X and inserts at 100X magnification. (B) Corresponding immunohistochemical staining of PDX tumors for mesenchymal cytoskeletal element vimentin, shown at 20X and inserts at 100X magnification.

128 129

129 130

Figure 3.13: Cellular differentiation corresponds with epithelial character in PDX tumors.

(A) Immunohistochemical staining of PDX tumors for marker of cellular differentiation, cytokeratin-18, shown at 20X and 100X magnification.

130 131

131 132

Figure 3.14: HSET Kinesin expression is consistent across PDX tumors.

(A) Immunohistochemical staining of PDX tumors for HSET kinesin, shown at 20X and

100X magnification. HSET is localized in cell nuclei, with rare diffuse cytosolic staining.

132 133

133 134

Figure 3.15: SLCO1B3 expression correlates with epithelial phenotype in cabazitaxel resistant prostate tumors. (A) Expression and cellular localization of the transporter protein SLCO1B3 was examined by immunohistochemistry of paraffin embedded sections of PDX prostate tumors for membrane localization. The images reveal the immunoreactivity for the transporter at 20X and 100X magnification. Protein expression is weak and diffuse, with the strongest stain in patient 4, an epithelial phenotype tumor isolated from circulating tumor cells.

134 135

135 136

Chapter Four: Discussion

The current therapeutic landscape for treatment of prostate cancer is lacking options, for CRPC in particular, and in resistant disease. Understanding the biology that underlies therapeutically resistant disease is the primary way to improve treatment options and thus improve patient outcome.

Our findings identified a novel interaction, AR-HSET, which could provide a new targetable mechanism to impair AR signaling. Moreover we show that cabazitaxel and

ADT in combination and in optimized sequencing is therapeutically effective in androgen- sensitive prostate cancer. The observation that DHT has a strong effect on the AR-null

PC3 (androgen-independent) cell line, implicates a mechanism of DHT action independent of AR binding. Interestingly, DHT presence increased cabazitaxel cellular entry in PC3 cells, but not in LNCaP cells or PC3-shSLCO1B3 cells. The complicated relationship between androgens and taxane therapy is far from being fully understood, but assisting in cabazitaxel cell entry could be advantageous if cabazitaxel were given without ADT being present.

HSET, throughout my studies, was consistently and reliably located in the nucleus.

However, in PC3 xenograft tumor staining, HSET was present in the cytosol in a minority of untreated tumors, and was present in the cytosol in the majority of cabazitaxel-treated tumors. Viewed with confocal microscopy, cabazitaxel treated PC3 cells in vitro had cytosolic localization of HSET. In addition, PC3-CR cells had cytosolic HSET both before and after treatment. The cellular localization of HSET could be a contributing factor in cabazitaxel response or resistance, and future mechanistic research is necessary to

136 137 determine the importance of localization and physiological response. In contrast to PC3 cells, cabazitaxel treated LNCaP and 22Rv1 cells, both in vitro and in vivo xenograft tumors, showed exclusively a nuclear localization of HSET after treatment. These data require further investigation before the cabazitaxel-induced nuclear localization is considered to be reliable and repeatable. It is possible that the reason HSET is responding to cabazitaxel differently in PC3 xenografts than in the LNCaP and 22Rv1 xenografts

(where HSET was exclusively nuclear in localization) is because PC3 cells do not contain any androgen receptor. The AR-HSET interaction discovered in Chapter 2 could have more mechanistic consequences in the cell, with AR transport providing a critical role for

HSET in the nucleus – without AR, HSET could lose a nuclear role. Further research into this hypothesis is necessary.

Early evidence indicated that DHT or testosterone transport/blocked SLCO1B3 transporter but more thorough evidence showed the opposite. Stable SLCO1B3 knockdown cells showed an increased intracellular concentration of cabazitaxel, indicative of knockdown increasing cabazitaxel cellular uptake. These data indicate that perhaps a decreased SLCO1B3 triggers the upregulation of other membrane transporters, resulting in increased cabazitaxel uptake. The complex, novel relationship between DHT and

SLCO1B3 discovered here is definitively involved because DHT increased cabazitaxel uptake only when SLCO1B3 is present at full biological level – shRNA knockdown of

SLCO1B3 rendered cabazitaxel uptake independent from DHT’s effect. This integral role of SLCO1B3 in cabazitaxel uptake has not been explored before, and it warrants future research as to whether DHT is activating the transporter, or causing upregulation of

SLCO1B3 or other, undiscovered cabazitaxel transporters. The current state of

137 138 pharmacologic inhibitors of SLCO family proteins is limiting – there is no available, specific inhibitor for SLCO1B3. If a specific inhibitor for SLCO1B3 is developed, there can be a more thorough investigation into cabazitaxel uptake through specific

SLCO/OATP family proteins. Targeting this transporter may have tremendous side effects in normal surrounding tissue.

Therapeutic sequencing in in vivo models done in Chapter 2 of my work indicated that in androgen dependent models, sequencing ADT and cabazitaxel will potentially lead to a better treatment outcome. Interestingly, this theory was published in a clinical trial when my pre-clinical experiments were underway (236). The FIRSTANA trial, a randomized phase III trial, tested the efficacy of giving cabazitaxel instead of docetaxel as a first chemotherapy treatment in patients with chemotherapy-naïve mCRPC (236). The majority of these patients had received first generation AR-targeted therapies, with a small number having received abiraterone or enzalutamide treatment. There was no change in median overall survival, with cabazitaxel (20mg/m2) having a median survival of 24.5 months, cabazitaxel (25mg/m2) of 25.2 months, and docetaxel (75 mg/m2) of 24.3 months

(236). There was no significant change in overall survival or progression free survival comparing either of the two cabazitaxel doses to docetaxel (1st line chemotherapy).

However, tumor responses (complete or partial response) were higher in patients receiving cabazitaxel (25mg/m2) compared to docetaxel (236).

One major limitation of the FIRSTANA trial was the very few patients that had received second generation antiandrogens abiraterone or enzalutamide. This was a matter of timing, as these therapies were just being introduced into the clinic when the FIRSTANA trial was beginning (80, 236). The other very interesting element of this trial is that it was

138 139 done exclusively on mCRPC patients. My pre-clinical data aligns with the outcome of

FIRSTANA, because no change in sequencing therapies was found in advanced, CRPC models of prostate cancer. The therapeutic potential in my pre-clinical data is in androgen sensitive models, before CRPC develops. It would be interesting to perform a clinical trial testing cabazitaxel chemotherapy in patients before CRPC or mCRPC develops, in order to prevent mCRPC from developing. This change may be transformative in the treatment regime for prostate cancer patients. The therapeutic potential of preventing mCRPC is worthy of testing in clinical trials, and continued pre-clinical studies. Taken together, this data is of high translational significance in identifying novel, potentially targetable interactions in advanced lethal prostate tumors.

In the subsequent segment, I wish to discuss the limitations of my dissertation work.

Firstly, the PC3-CR cell line had detectable DNA of mouse origin, as discussed in Chapter

3. The cell line was majority human, with all detectable markers specific for PC3 cells, indicating that the human cells were of PC3 prostate cancer origin. The contamination with murine cells is a limitation, experiments with the cell line were halted once the test results were received. The increase in mesenchymal character correlated with increased migratory and invasive characteristics. This effect is well established both within prostate cancer and in other cancers (237-239). The cross-resistance with docetaxel highlights the similar mechanisms of action between microtubule targeting agents docetaxel and cabazitaxel, emphasizing the similar mechanisms of resistance that are still unknown. Not to be overlooked, the decrease in SLCO1B3 expression in cabazitaxel resistant cells spurred further research into SLCO1B3’s role in cabazitaxel treatment, a line of inquiry that requires continued research to understand. As discussed above, the lack of current

139 140 specific pharmacological inhibition of SLCO1B3 is limiting for the breadth of studies than can be done at this time, and in the current therapeutic landscape. If a specific SLCO1B3 inhibitor is developed, the efficacy can be tested in pre-clinical models as a combination therapy with cabazitaxel and/or anti-androgens to re-sensitize resistant models to available therapeutics.

It is readily apparent that a significant amount of my data is immunohistochemical staining, and conclusions drawn from stained slides can be done through qualitative or quantitative measures. The majority of staining, such as E-Cadherin, N-Cadherin,

Vimentin, and Cytokeratin-18 are analyzed through qualitative measures, visually determining the strength of the protein expression. Some markers, such as AR and HSET, were quantified by a board certified pathologist and either myself or the principal investigator. An approach that could have been used is quantification through computer program, but was decided against due to access to pathologists and possible errors in interpretation by computer analysis.

A limitation both in my work and persistent throughout the prostate cancer research field as a whole is limiting cell models. The standard research in prostate cancer is done on a variety of cell lines, each modeling different stages of disease. How I utilized this was LNCaP, the hormone naïve, androgen sensitive cell line represented androgen sensitive disease, 22Rv1 cell line modeled castration resistant prostate cancer, and PC3 represented androgen independent disease. There are, of course, differences between the cell lines aside from just androgen receptor presence or absence. Each cell line was isolated from a different patient with prostate cancer, from different organs of metastasis, have different doubling times, and even different karyotypes. To validate the findings in my

140 141 dissertation, using more biologically precise models would be prudent. For example, to study AR-v7, instead of using LNCaP (possessing AR) and 22Rv1 (possessing AR, AR- v7 and other AR variants), the PC3 cell line could be transfected to express AR or AR-v7 to guarantee that adding AR or AR-v7 is the only difference in the models studied. As noted earlier, this is a weakness in the prostate cancer field as a whole; the disease progression is so extreme from benign to metastatic disease that it is only natural that there is a wide spread of cell lines to represent disease stages, but the field can strive to be as precise as possible in experimentation.

A limitation in the pre-clinical xenograft study examining the effect of tumor sequencing is the lack of ADT/castration alone treatment group. Experimental design for this study accounted for the plentiful, established literature regarding the effect of castrating on LNCaP and 22Rv1 cells (112, 240-242). The goal of my xenograft experiments was to examine the therapeutic efficacy of sequencing ADT and cabazitaxel chemotherapy to optimize therapeutic response, a goal that was achieved in androgen sensitive LNCaP cells but resulted in no effect in CRPC 22Rv1 cells. A limitation, which can be focused into future studies, is the lack of survival studies done with preclinical mouse models to determine if treatment had an effect on overall mouse survival – a study that would be particularly interesting for treatment sequencing.

Future directions derived from my studies are multifaceted at the mechanistic and translational level in the context of developing effective strategies to overcome therapeutic resistance in prostate cancer patients with lethal disease. First, the interaction between AR and HSET, and potential therapeutic consequences of targeting this interaction needs to be explored further. Studies will need to be completed using the HSET inhibitor CW069

141 142

(167), examining the efficacy of the inhibitor in prostate cancer cells both in vitro and using pre-clinical in vivo models. Studies have been done with CW069 and docetaxel in combination therapy, an avenue that has promise towards re-sensitizing docetaxel resistant cells to therapy (149, 168). Cabazitaxel and CW069 in combination therapy has not been studied, and future research in this area should take into account the interaction of AR and

HSET that is inhibited by cabazitaxel chemotherapy. A secondary avenue to further confirm the interaction between AR and HSET that should be pursued is a crystal or computer generated model of the protein-protein interaction. While the co- immunoprecipitation studies showed a clear binding, the technique should be confirmed with more extensive investigation. Along the same thread, potential binding of HSET to

AR variants, particularly AR-v7, should be investigated as a potential therapeutic target in advanced disease that is resistant to standard agents.

The paradoxical finding of SLCO1B3 knockdown causing increased intracellular cabazitaxel concentration warrants further research. SLCO1B3 is a strictly influx transporter (243, 244) so it is not possible that decreased SLCO1B3 level resulted in less drug efflux unless there was a concurrent loss of efflux proteins such as ATP-binding cassette transporters or multi-drug resistance-associated proteins (ABC, MDR protein families, respectively) (245). One may consider the possibility that stable SLCO1B3 functional loss knockdown triggered a feedback loop that resulted in upregulation of other

SLCO family membrane transporters, and these other transporters allowed increased efflux of cabazitaxel. Other SLCO family protein level will be profiled after SLCO1B3 knockdown for both protein and RNA level, to determine if there is an upregulation of other protein family members resulting in increased cabazitaxel entry. SLCO1B3 plays a

142 143 role in cabazitaxel resistance, as shown by SLCO1B3 knockdown inducing a modest resistance to cabazitaxel, but it is not the only protein involved. If SLCO1B3 were the only cellular uptake protein for cabazitaxel, knockdown would have had a more complete effect, and would have conferred a more substantial resistance with SLCO1B3 knockdown.

The most epithelial-like PDX tumors also had the highest expression of SLCO1B3.

The goal of my dissertation work has been to investigate the underlying mechanisms that contribute to therapeutic resistance in prostate cancer. Cabazitaxel remains the only antiproliferative chemotherapeutic option as a 2nd line taxane chemotherapy for prostate cancer patients with metastatic disease, alongside AR targeting therapies, radiation, and limited immunotherapy. There are currently more than 650 Phase I and II clinical trials, and over

140 Phase III trials for prostate cancer treatment in the United States alone (246). There is clinical innovation happening through the ongoing clinical trials, but continued advancements in basic science research are much needed to provide mechanism-based evidence for supporting pre-clinical guidance based studies towards innovative clinical applications with potential impact in patients. The findings described in this thesis provide novel insights into the molecular and phenotypic characterization of the prostate tumor landscape that may potentially drive new therapeutic platforms through identifying novel, actionable targets in advanced prostate tumors in patients with lethal disease. Pending and future clinical trials in prostate cancer patients with lethal disease will take several years to be interpreted, with the overarching goal to have an impact on patients’ survival so that it can be increased from a few months to a few years with thoughtful research enhancing our understanding of both tumor biology and therapeutic resistance.

143 144

References

1. Ittmann M. Anatomy and Histology of the Human and Murine Prostate. Cold Spring Harbor Perspectives in Medicine. 2018;8(5):a030346. 2. Huggins C, Scott WW, Heinen JH. Chemical Composition of Human Semen and of the Secretions of the Prostate and Semenal Vesicles. American Journal of Physiology-Legacy Content. 1942;136(3):467-73. 3. Chughtai B, Forde JC, Thomas DDM, Laor L, Hossack T, Woo HH, et al. Benign prostatic hyperplasia. Nature Reviews Disease Primers. 2016;2(1):16031. 4. Sadley T. Langman's medical embryology (14th Ed). Philadelphia: Wolters Kluwer; 2019. 5. Satandring S. Gray's anatomy: the anatomical basis of clinical practice. 41st ed. Philadelphia. 2016. 6. Yilmaz M, Christofori G. EMT, the cytoskeleton, and cancer cell invasion. Cancer and Metastasis Reviews. 2009;28(1-2):15-33. 7. Lee H, Park S, Kim JB, Kim J, Kim H. Entrapped doxorubicin nanoparticles for the treatment of metastatic anoikis-resistant cancer cells. Cancer Letters. 2013;332(1):110-9. 8. Cunha GR, Ricke W, Thomson A, Marker PC, Risbridger G, Hayward SW, et al. Hormonal, cellular, and molecular regulation of normal and neoplastic prostatic development. 2004;92(4):221-36. 9. Heinlein CA, Chang C. The Roles of Androgen Receptors and Androgen-Binding Proteins in Nongenomic Androgen Actions. Molecular Endocrinology. 2002;16(10):2181-7. 10. Bolton EC, So AY, Chaivorapol C, Haqq CM, Li H, Yamamoto KR. Cell- and gene-specific regulation of primary target genes by the androgen receptor. Genes & Development. 2007;21(16):2005-17. 11. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69(1):7-34. 12. Huggins C, Hodges CV. Studies on Prostatic Cancer: I. The Effect of Castration, of Estrogen and of Androgen Injection on Serum Phosphatases in Metastatic Carcinoma of the Prostate. Journal of Urology. 1941;168(1):9-12. 13. Huggins C, Hodges CV. Studies on Prostatic Cancer: I. The Effect of Castration, Of Estrogen and of Androgen Injection on Serum Phosphatases in Metastatic Carcinoma of the Prostate. 1972;22(4):232-40. 14. Mohler JL. Castration-Recurrent Prostate Cancer Is Not Androgen-Independent. Hormonal Carcinogenesis V: Springer New York; 2008. p. 223-34. 15. Kahn B, Collazo J, Kyprianou N. Androgen Receptor as a Driver of Therapeutic Resistance in Advanced Prostate Cancer. International Journal of Biological Sciences. 2014;10(6):588-95. 16. Hu R, Denmeade SR, Luo J. Molecular processes leading to aberrant androgen receptor signaling and castration resistance in prostate cancer. Expert Review of Endocrinology & Metabolism. 2010;5(5):753-64.

144 145

17. Rosenbaum E, Partin A, Eisenberger MA. Biochemical Relapse After Primary Treatment for Prostate Cancer: Studies on Natural History and Therapeutic Considerations. Journal of the National Comprehensive Cancer Network. 2004;2(3):249-56. 18. Hotte SJ, Saad F. Current management of castrate-resistant prostate cancer. Current Oncology. 2010;17(0). 19. Kyprianou N, Isaacs JT. Activation of Programmed Cell Death in the Rat Ventral Prostate after Castration*. Endocrinology. 1988;122(2):552-62. 20. Smitherman AB, Gregory CW, Mohler JL. Apoptosis levels increase after castration in the CWR22 human prostate cancer xenograft. The Prostate. 2003;57(1):24-31. 21. Westin P, Stattin P, Damber JE, Bergh A. Castration therapy rapidly induces apoptosis in a minority and decreases cell proliferation in a majority of human prostatic tumors. Am J Pathol. 1995;146(6):1368-75. 22. Etheridge T, Damodaran S, Schultz A, Richards KA, Gawdzik J, Yang B, et al. Combination therapy with androgen deprivation for hormone sensitive prostate cancer: A new frontier. Asian Journal of Urology. 2019;6(1):57-64. 23. Danial NN, Korsmeyer SJ. Cell Death. Cell. 2004;116(2):205-19. 24. Simpson CD, Anyiwe K, Schimmer AD. Anoikis resistance and tumor metastasis. Cancer Letters. 2008;272(2):177-85. 25. Frisch SM, Screaton RA. Anoikis mechanisms. Current Opinion in Cell Biology. 2001;13(5):555-62. 26. McKenzie S, Kyprianou N. Apoptosis evasion: The role of survival pathways in prostate cancer progression and therapeutic resistance. Journal of Cellular Biochemistry. 2005;97(1):18-32. 27. Rennebeck G. Anoikis and Survival Connections in the Tumor Microenvironment: Is There a Role in Prostate Cancer Metastasis? Cancer Research. 2005;65(24):11230-5. 28. Gilmore AP. Anoikis. Cell Death and Differentiation. 2005;12:1473-7. 29. Frisch SM, Ruoslahti E. Integrins and anoikis. Current Opinion in Cell Biology. 1997;9(5):701-6. 30. Sakamoto S, McCann RO, Dhir R, Kyprianou N. Talin1 Promotes Tumor Invasion and Metastasis via Focal Adhesion Signaling and Anoikis Resistance. Cancer Research. 2010;70(5):1885-95. 31. Tan K, Goldstein D, Crowe P, Yang J-L. Uncovering a key to the process of metastasis in human cancers: a review of critical regulators of anoikis. Journal of Cancer Research and Clinical Oncology. 2013;139(11):1795-805. 32. Winograd-Katz SE, Fässler R, Geiger B, Legate KR. The integrin adhesome: from genes and proteins to human disease. Nature Reviews Molecular Cell Biology. 2014;15(4):273-88. 33. Igney FH, Krammer PH. DEATH AND ANTI-DEATH: TUMOUR RESISTANCE TO APOPTOSIS. Nature Reviews Cancer. 2002;2(4):277-88. 34. Desgrosellier JS, Cheresh DA. Integrins in cancer: biological implications and therapeutic opportunities. Nature Reviews Cancer. 2010;10(1):9-22. 35. Riedl SJ, Salvesen GS. The apoptosome: signalling platform of cell death. Nature Reviews Molecular Cell Biology. 2007;8(5):405-13.

145 146

36. Cory S, Adams JM. The bcl2 family: regulators of the cellular life-or-death switch. Nature Reviews Cancer. 2002;2(9):647-56. 37. Alanko J, Mai A, Jacquemet G, Schauer K, Kaukonen R, Saari M, et al. Integrin endosomal signalling suppresses anoikis. Nature Cell Biology. 2015;17(11):1412- 21. 38. Díaz-Montero CM, Wygant JN, McIntyre BW. PI3-K/Akt-mediated anoikis resistance of human osteosarcoma cells requires Src activation. European Journal of Cancer. 2006;42(10):1491-500. 39. Attwell S, Roskelley C, Dedhar S. The integrin-linked kinase (ILK) suppresses anoikis. Oncogene. 2000;19(33):3811-5. 40. Taddei ML, Giannoni E, Fiaschi T, Chiarugi P. Anoikis: an emerging hallmark in health and diseases. The Journal of Pathology. 2011;226(2):380-93. 41. Nagata S. Fas Ligand-Induced Apoptosis. Annual Review of Genetics. 1999;33(1):29-55. 42. Wen S, Niu Y, Lee S, Chang C. Androgen Receptor (AR) Positive vs Negative Roles in Prostate Cancer CellDeaths including Apoptosis, Anoikis, Entosis, Necrosis and Autophagic Cell Death. Cancer Treat Rev. 2014;40(1). 43. Giannoni E, Fiaschi T, Ramponi G, Chiarugi P. Redox regulation of anoikis resistance of metastatic prostate cancer cells: key role for Src and EGFR-mediated pro survival signals. Oncogene. 2009;28(20):2074-86. 44. Toivanen R, mohan A, Shen M. Basal progenitors contribute to repair of prostate epithelium following induced luminal anoikis. Stem Cell Reports. 2016;6(5):660- 7. 45. Thiery JP. Epithelial–mesenchymal transitions in tumour progression. Nature Reviews Cancer. 2002;2(6):442-54. 46. van der Pluijm G. Epithelial plasticity, cancer stem cells and bone metastasis formation. Bone. 2011;48(1):37-43. 47. Frisch SM, Schaller M, Cieply B. Mechanisms that link the oncogenic epithelial– mesenchymal transition to suppression of anoikis. Journal of Cell Science. 2013;126(1):21-9. 48. Zhu ML, Kyprianou N. Role of androgens and the androgen receptor in epithelial- mesenchymal transition and invasion of prostate cancer cells. The FASEB Journal. 2009;24(3):769-77. 49. Derksen PWB, Liu X, Saridin F, van der Gulden H, Zevenhoven J, Evers B, et al. Somatic inactivation of E-cadherin and p53 in mice leads to metastatic lobular mammary carcinoma through induction of anoikis resistance and angiogenesis. Cancer Cell. 2006;10(5):437-49. 50. Onder TT, Gupta PB, Mani SA, Yang J, Lander ES, Weinberg RA. Loss of E- Cadherin Promotes Metastasis via Multiple Downstream Transcriptional Pathways. Cancer Research. 2008;68(10):3645-54. 51. Xu Y-R, Dong H-S, Yang W-X. Regulators in the apoptotic pathway during spermatogenesis: Killers or guards? Gene. 2016;582(2):97-111. 52. Umbas R, Isaacs WB, Bringuier PP, Xue Y, Debruyne FMJ, Schalken JA. Decreased E-Cadherin expression is associated with poor prognosis in patients with prostate cancer. Cancer Research. 1994;54(14):3929-22.

146 147

53. Folkman J. What Is the Evidence That Tumors Are Angiogenesis Dependent? JNCI Journal of the National Cancer Institute. 1990;82(1):4-7. 54. Radisky D, Muschler J, Bissell MJ. Order and Disorder: The Role of Extracellular Matrix in Epithelial Cancer. Cancer Investigation. 2002;20(1):139-53. 55. Overall CM, Kleifeld O. Tumour microenvironment — Opinion: Validating matrix metalloproteinases as drug targets and anti-targets for cancer therapy. Nature Reviews Cancer. 2006;6(3):227-39. 56. Tiwari N, Gheldof A, Tatari M, Christofori G. EMT as the ultimate survival mechanism of cancer cells. Seminars in Cancer Biology. 2012;22(3):194-207. 57. Kang Y, Massagué J. Epithelial-Mesenchymal Transitions. Cell. 2004;118(3):277- 9. 58. Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. Journal of Clinical Investigation. 2009;119(6):1420-8. 59. Zhu ML, Kyprianou N. Role of androgens and the androgen receptor in epithelial- mesenchymal transition and invasion of prostate cancer cells. FASEB J. 2010;24(3):769-77. 60. Frisch SM. Disruption of epithelial cell-matrix interactions induces apoptosis. The Journal of Cell Biology. 1994;124(4):619-26. 61. Kroemer G, Galluzzi L, Brenner C. Mitochondrial Membrane Permeabilization in Cell Death. Physiological Reviews. 2007;87(1):99-163. 62. SK Martin MK, Natasha Kyprianou. Cytoskeleton Targeting Value in prostate Cancer. Am J Clin Exp Urol. 2015;2(1):15-26. 63. Smith B, Bhowmick N. Role of EMT in Metastasis and Therapy Resistance. Journal of Clinical Medicine. 2016;5(2):17. 64. Nguyen DX, Bos PD, Massagué J. Metastasis: from dissemination to organ-specific colonization. Nature Reviews Cancer. 2009;9(4):274-84. 65. Howard EW, Leung SCL, Yuen HF, Chua CW, Lee DT, Chan KW, et al. Decreased adhesiveness, resistance to anoikis and suppression of GRP94 are integral to the survival of circulating tumor cells in prostate cancer. Clinical & Experimental Metastasis. 2008;25(5):497-508. 66. Liotta LA. Adhere, Degrade, and Move: The Three-Step Model of Invasion. Cancer Research. 2016;76(11):3115-7. 67. Matus David Q, Lohmer Lauren L, Kelley Laura C, Schindler Adam J, Kohrman Abraham Q, Barkoulas M, et al. Invasive Cell Fate Requires G1 Cell-Cycle Arrest and Histone Deacetylase-Mediated Changes in Gene Expression. Developmental Cell. 2015;35(2):162-74. 68. Yano S, Miwa S, Mii S, Hiroshima Y, Uehara F, Yamamoto M, et al. Invading cancer cells are predominantly in G0/G1resulting in chemoresistance demonstrated by real-time FUCCI imaging. Cell Cycle. 2014;13(6):953-60. 69. Cui D, Dai J, Keller JM, Mizokami A, Xia S, Keller ET. Notch Pathway Inhibition Using PF-03084014, a -Secretase Inhibitor (GSI), Enhances the Antitumor Effect of Docetaxel in Prostate Cancer. Clinical Cancer Research. 2015;21(20):4619-29. 70. Glick AB. TGF? Back to the Future: Revisiting its Role as a Transforming Growth Factor. Cancer Biology & Therapy. 2004;3(3):276-83. 71. Massagué J. TGFβ in Cancer. Cell. 2008;134(2):215-30.

147 148

72. Zhang YE. Non-Smad pathways in TGF-β signaling. Cell Research. 2009;19(1):128-39. 73. Collazo J, Zhu B, Larkin S, Martin SK, Pu H, Horbinski C, et al. Cofilin Drives Cell-Invasive and Metastatic Responses to TGF- in Prostate Cancer. 2014;74(8):2362-73. 74. Martin SK, Banuelos CA, Sadar MD, Kyprianou N. N-terminal targeting of androgen receptor variant enhances response of castration resistant prostate cancer to taxane chemotherapy. Molecular Oncology. 2014;9(3):628-39. 75. Zhu ML, Horbinski CM, Garzotto M, Qian DZ, Beer TM, Kyprianou N. Tubulin- Targeting Chemotherapy Impairs Androgen Receptor Activity in Prostate Cancer. Cancer Research. 2010;70(20):7992-8002. 76. Clyne M. Prostate cancer: Androgen deprivation causes EMT in the prostate. Nature Reviews Urology. 2011;9(1):4-. 77. Sun Y, Wang BE, Leong KG, Yue P, Li L, Jhunjhunwala S, et al. Androgen Deprivation Causes Epithelial-Mesenchymal Transition in the Prostate: Implications for Androgen-Deprivation Therapy. Cancer Research. 2011;72(2):527-36. 78. Chi K, Hotte S, Joshua A, North S, Wyatt A, Collins L, et al. Treatment of mCRPC in the AR-axis-targeted therapy-resistant state. Annals of Oncology. 2015;26:2044- 56. 79. de Bono JS, Logothetis CJ, Molina A, Fizazi K, North S, Chu L, et al. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med. 2011;364(21):1995-2005. 80. Scher HI, Fizazi K, Saad F, Taplin ME, Sternberg CN, Miller K, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;367(13):1187-97. 81. Kahn B, Collazo J, Kyprianou N. Androgen receptor as a driver of therapeutic resistance in advanced prostate cancer. Int J Biol Sci. 2014;10(6):588-95. 82. Mohler JL. Castration-recurrent prostate cancer is not androgen-independent. Adv Exp Med Biol. 2008;617:223-34. 83. Antonarakis ES, Lu C, Luber B, Wang H, Chen Y, Nakazawa M, et al. Androgen Receptor Splice Variant 7 and Efficacy of Taxane Chemotherapy in Patients With Metastatic Castration-Resistant Prostate Cancer. JAMA Oncol. 2015;1(5):582-91. 84. Labrie F, Dupont A, Belanger A, Lachance R, Giguere M. Long term treatment with luteinising hormone releasing hormone agonists and maintenance of serum testosterone to castration concentrations. BMJ. 1985;291(6492):369-70. 85. Hoda MR, Kramer MW, Merseburger AS, Cronauer MV. Androgen deprivation therapy with Leuprolide acetate for treatment of advanced prostate cancer. Expert Opinion on Pharmacotherapy. 2017;18(1):105-13. 86. Crawford ED, Schellhammer PF, McLeod DG, Moul JW, Higano CS, Shore N, et al. Androgen Receptor Targeted Treatments of Prostate Cancer: 35 Years of Progress with Antiandrogens. Journal of Urology. 2018;200(5):956-66. 87. Chaudhary UB, Turner JS. Finasteride. 2010;6(7):873-81. 88. Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, et al. The Influence of Finasteride on the Development of Prostate Cancer. New England Journal of Medicine. 2003;349(3):215-24.

148 149

89. Tran C, Ouk S, Clegg NJ, Chen Y, Watson PA, Arora V, et al. Development of a Second-Generation Antiandrogen for Treatment of Advanced Prostate Cancer. Science. 2009;324(5928):787-90. 90. Hoffman-Censits J, Kelly WK. Enzalutamide: A Novel Antiandrogen for Patients with Castrate-Resistant Prostate Cancer. Clinical Cancer Research. 2013;19(6):1335-9. 91. Scher HI, Fizazi K, Saad F, Taplin M-E, Sternberg CN, Miller K, et al. Increased Survival with Enzalutamide in Prostate Cancer after Chemotherapy. New England Journal of Medicine. 2012;367(13):1187-97. 92. Nadal R, Zhang Z, Rahman H, Schweizer MT, Denmeade SR, Paller CJ, et al. Clinical activity of enzalutamide in Docetaxel-naïve and Docetaxel-pretreated patients with metastatic castration-resistant prostate cancer. 2014;74(15):1560-8. 93. Culig Z. Molecular Mechanisms of Enzalutamide Resistance in Prostate Cancer. Current Molecular Biology Reports. 2017;3(4):230-5. 94. Vivek, Schenkein E, Murali R, Sumit, Wongvipat J, Minna, et al. Glucocorticoid Receptor Confers Resistance to Antiandrogens by Bypassing Androgen Receptor Blockade. Cell. 2013;155(6):1309-22. 95. Rehman Y, Rehman Y. Abiraterone acetate: oral androgen biosynthesis inhibitor for treatment of castration-resistant prostate cancer. 2012:13. 96. De Bono JS, Logothetis CJ, Molina A, Fizazi K, North S, Chu L, et al. Abiraterone and Increased Survival in Metastatic Prostate Cancer. New England Journal of Medicine. 2011;364(21):1995-2005. 97. Fizazi K, Scher HI, Molina A, Logothetis CJ, Chi KN, Jones RJ, et al. Abiraterone acetate for treatment of metastatic castration-resistant prostate cancer: final overall survival analysis of the COU-AA-301 randomised, double-blind, placebo- controlled phase 3 study. The Lancet Oncology. 2012;13(10):983-92. 98. Noonan KL, North S, Bitting RL, Armstrong AJ, Ellard SL, Chi KN. Clinical activity of abiraterone acetate in patients with metastatic castration-resistant prostate cancer progressing after enzalutamide. 2013;24(7):1802-7. 99. Loriot Y, Bianchini D, Ileana E, Sandhu S, Patrikidou A, Pezaro C, et al. Antitumour activity of abiraterone acetate against metastatic castration-resistant prostate cancer progressing after docetaxel and enzalutamide (MDV3100). 2013;24(7):1807-12. 100. Wadosky KM, Koochekpour S. Androgen receptor splice variants and prostate cancer: From bench to bedside. Oncotarget. 2017;8(11). 101. Antonarakis ES, Chandhasin C, Osbourne E, Luo J, Sadar MD, Perabo F. Targeting the N-Terminal Domain of the Androgen Receptor: A New Approach for the Treatment of Advanced Prostate Cancer. The Oncologist. 2016;21(12):1427-35. 102. Myung J-K, Banuelos CA, Fernandez JG, Mawji NR, Wang J, Tien AH, et al. An androgen receptor N-terminal domain antagonist for treating prostate cancer. 2013;123(7):2948-60. 103. Aarnisalo P, Palvimo JJ, Janne OA. CREB-binding protein in androgen receptor- mediated signaling. Proceedings of the National Academy of Sciences. 1998;95(5):2122-7.

149 150

104. McEwan IJ, Gustafsson JA. Interaction of the human androgen receptor transactivation function with the general transcription factor TFIIF. 1997;94(16):8485-90. 105. Thiery JP, Chua K, Sim WJ, Huang R. [Epithelial mesenchymal transition during development in fibrosis and in the progression of carcinoma]. Bull Cancer. 2010;97(11):1285-95. 106. Huber MA, Kraut N, Beug H. Molecular requirements for epithelial-mesenchymal transition during tumor progression. Curr Opin Cell Biol. 2005;17(5):548-58. 107. Xie L, Law BK, Chytil AM, Brown KA, Aakre ME, Moses HL. Activation of the Erk pathway is required for TGF-beta1-induced EMT in vitro. Neoplasia. 2004;6(5):603-10. 108. Thiery JP, Sleeman JP. Complex Networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol. 2006;7(2):131-42. 109. Bruckheimer EM, Kyprianou N. Dihydrotestosterone enhances transforming growth factor-beta-induced apoptosis in hormone-sensitive prostate cancer cells. Endocrinology. 2001;142(6):2419-26. 110. Tsai YC, Chen WY, Abou-Kheir W, Zeng T, Yin JJ, Bahmad H, et al. Androgen deprivation therapy-induced epithelial-mesenchymal transition of prostate cancer through downregulating SPDEF and activating CCL2. Biochim Biophys Acta Mol Basis Dis. 2018;1864(5 Pt A):1717-27. 111. Sun Y, Wang BE, Leong KG, Yue P, Li L, Jhunjhunwala S, et al. Androgen deprivation causes epithelial-mesenchymal transition in the prostate: implications for androgen-deprivation therapy. Cancer Res. 2012;72(2):527-36. 112. Martin SK, Pu H, Penticuff JC, Cao Z, Horbinski C, Kyprianou N. Multinucleation and Mesenchymal-to-Epithelial Transition Alleviate Resistance to Combined Cabazitaxel and Antiandrogen Therapy in Advanced Prostate Cancer. Cancer Res. 2016;76(4):912-26. 113. Huizing MT, Misser VH, Pieters RC, ten Bokkel Huinink WW, Veenhof CH, Vermorken JB, et al. Taxanes: a new class of antitumor agents. Cancer Invest. 1995;13(4):381-404. 114. Vrignaud P, Semiond D, Lejeune P, Bouchard H, Calvet L, Combeau C, et al. Preclinical Antitumor Activity of Cabazitaxel, a Semisynthetic Taxane Active in Taxane-Resistant Tumors. Clinical Cancer Research. 2013;19(11):2973-83. 115. Bruckheimer EM, Kyprianou N. Dihydrotestosterone Enhances Transforming Growth Factor-β-Induced Apoptosis in Hormone-Sensitive Prostate Cancer Cells 1. 2001;142(6):2419-26. 116. Bruckheimer EM, Kyprianou N. bcl-2 antagonizes the combined apoptotic effect of transforming growth factor-? and dihydrotestosterone in prostate cancer cells. The Prostate. 2002;53(2):133-42. 117. Debes JD, Tindall DJ. Mechanisms of Androgen-Refractory Prostate Cancer. New England Journal of Medicine. 2004;351(15):1488-90. 118. Oliver CL, Miranda MB, Shangary S, Land S, Wang S, Johnson DE. (−)-Gossypol acts directly on the mitochondria to overcome Bcl-2- and Bcl-X_L- mediated apoptosis resistance. Molecular Cancer Therapeutics. 2005;4(1):23-31. 119. Mezynski J, Pezaro C, Bianchini D, Zivi A, Sandhu S, Thompson E, et al. Antitumour activity of docetaxel following treatment with the CYP17A1 inhibitor

150 151

abiraterone: clinical evidence for cross-resistance? Annals of Oncology. 2012;23(11):2943-7. 120. van Soest RJ, van Royen ME, de Morrée ES, Moll JM, Teubel W, Wiemer EAC, et al. Cross-resistance between taxanes and new hormonal agents abiraterone and enzalutamide may affect drug sequence choices in metastatic castration-resistant prostate cancer. European Journal of Cancer. 2013;49(18):3821-30. 121. Schutz FA, Buzaid AC, Sartor O. Taxanes in the management of metastatic castration-resistant prostate cancer: Efficacy and management of toxicity. Critical Reviews in Oncology/Hematology. 2014;91(3):248-56. 122. Tannock IF, De Wit R, Berry WR, Horti J, Pluzanska A, Chi KN, et al. Docetaxel plus Prednisone or Mitoxantrone plus Prednisone for Advanced Prostate Cancer. New England Journal of Medicine. 2004;351(15):1502-12. 123. Kroon J, Kooijman S, Cho N-J, Storm G, Van Der Pluijm G. Improving Taxane- Based Chemotherapy in Castration-Resistant Prostate Cancer. Trends in Pharmacological Sciences. 2016;37(6):451-62. 124. Tamaki H, Harashima N, Hiraki M, Arichi N, Nishimura N, Shiina H, et al. Bcl-2 family inhibition sensitizes human prostate cancer cells to docetaxel and promotes unexpected apoptosis under caspase-9 inhibition. 2014;5(22):11399-412. 125. Parrondo R, De Las Pozas A, Reiner T, Perez-Stable C. ABT-737, a small molecule Bcl-2/Bcl-xL antagonist, increases antimitotic-mediated apoptosis in human prostate cancer cells. PeerJ. 2013;1:e144. 126. Carthon B, Antonarakis ES. The STAMPEDE trial: paradigm-changing data through innovative trial design. Transl Cancer Res. 2016;5(3):S485-90. 127. Huzil JT, Chen K, Kurgan L, Tuszynski JA. The roles of beta-tubulin mutations and isotype expression in acquired drug resistance. Cancer Inform. 2007;3:159-81. 128. Abdulla A, Kapoor A. Emerging novel therapies in the treatment of castrate- resistant prostate cancer. Canadian Urological Association Journal. 2011;5(2):120- 33. 129. Attard G, Greystoke A, Kaye S, De Bono J. Update on tubulin-binding agents. Pathologie Biologie. 2006;54(2):72-84. 130. Galsky M, Dritselis A, Kirkpatrick P, Oh W. Cabazitaxel. Nature Reviews 2010;9:677-8. 131. Sartor AO. Progression of metastatic castrate-resistant prostate cancer: impact of therapeutic intervention in the post-docetaxel space. Journal of Hematology & Oncology. 2011;4(1):18. 132. Azarenko O, Smiyun G, Mah J, Wilson L, Jordan MA. Antiproliferative Mechanism of Action of the Novel Taxane Cabazitaxel as Compared with the Parent Compound Docetaxel in MCF7 Breast Cancer Cells. Molecular Cancer Therapeutics. 2014;13(8):2092-103. 133. De Bono JS, Oudard S, Ozguroglu M, Hansen S, Machiels J-P, Kocak I, et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. The Lancet. 2010;376(9747):1147-54. 134. Eisenberger M, Hardy-Bessard A-C, Kim CS, Géczi L, Ford D, Mourey L, et al. Phase III Study Comparing a Reduced Dose of Cabazitaxel (20 mg/m2) and the Currently Approved Dose (25 mg/m2) in Postdocetaxel Patients With Metastatic

151 152

Castration-Resistant Prostate Cancer—PROSELICA. Journal of Clinical Oncology. 2017;35(28):3198-206. 135. Pampaloni F, Florin E-L. Microtubule architecture: inspiration for novel carbon nanotube-based biomimetic materials. Trends in Biotechnology. 2008;26(6):302- 10. 136. Sweeney HL, Holzbaur ELF. Motor Proteins. Cold Spring Harbor Perspectives in Biology. 2018;10(5):a021931. 137. Vale R, Reese T, Sheetz M. Identification of a novel force-generating protein, kinesin, involved in microtubule-based motility. 1985;42(1):39-50. 138. Lawrence CJ, Dawe RK, Christie KR, Cleveland DW, Dawson SC, Endow SA, et al. A standardized kinesin nomenclature: Table I. 2004;167(1):19-22. 139. Miki H, Setou M, Kaneshiro K, Hirokawa N. All kinesin superfamily protein, KIF, genes in mouse and human. Proceedings of the National Academy of Sciences. 2001;98(13):7004-11. 140. Wu J, Akhmanova A. Microtubule-Organizing Centers. Annual Review of Cell and Developmental Biology. 2017;33(1):51-75. 141. Dagenbach EM. A new kinesin tree. Journal of Cell Science. 2004;117(1):3-7. 142. Lawrence CJ, Dawe RK, Christie KR, Cleveland DW, Dawson SC, Endow SA, et al. A standardized kinesin nomenclature. The Journal of Cell Biology. 2004;167(1):19-22. 143. Kim N, Song K. KIFC1 Is Essential for Bipolar Spindle Formation and Genomic Stability in the Primary Human Fibroblast IMR-90 Cell. Cell Structure and Function. 2013;38(1):21-30. 144. Cai S, Weaver LN, Ems-Mcclung SC, Walczak CE. Kinesin-14 Family Proteins HSET/XCTK2 Control Spindle Length by Cross-Linking and Sliding Microtubules. 2009;20(5):1348-59. 145. Wei Y-L, Yang W-X. Kinesin-14 motor protein KIFC1 participates in DNA synthesis and chromatin maintenance. Cell Death & Disease. 2019;10(6). 146. Liu Y, Zhan P, Zhou Z, Xing Z, Zhu S, Ma C, et al. The overexpression of KIFC1 was associated with the proliferation and prognosis of non-small cell lung cancer. Journal of Thoracic Disease. 2016;8(10):2911-23. 147. Imai T, Oue N, Yamamoto Y, Asai R, Uraoka N, Scentani K, et al. Overexpression of KIFC1 and its association with spheroid formation in esophageal squamous cell carcinoma. Pathology 2017:1388-93. 148. De S, Cipriano R, Jackson MW, Stark GR. Overexpression of Kinesins Mediates Docetaxel Resistance in Breast Cancer Cells. Cancer Research. 2009;69(20):8035- 42. 149. Sekino Y, Oue N, Shigematsu Y, Ishikawa A, Sakamoto N, Sentani K, et al. KIFC1 induces resistance to docetaxel and is associated with survival of patients with prostate cancer. Urologic Oncology: Seminars and Original Investigations. 2017;35(1):31.e13-31.e20. 150. Desai A, Verma S, Mitchison TJ, Walczak CE. Kin I Kinesins Are Microtubule- Destabilizing Enzymes. 1999;96(1):69-78. 151. Ogawa T, Nitta R, Okada Y, Hirokawa N. A Common Mechanism for Microtubule Destabilizers—M Type Kinesins Stabilize Curling of the Protofilament Using the Class-Specific Neck and Loops. 2004;116(4):591-602.

152 153

152. Rath O, Kozielski F. Kinesins and cancer. Nature Reviews. 2012;12:527-39. 153. Sircar K, Huang H, Hu L, Liu Y, Dhillon J, Cogdell D, et al. Mitosis Phase Enrichment with Identification of Mitotic Centromere-Associated Kinesin As a Therapeutic Target in Castration-Resistant Prostate Cancer. PLoS ONE. 2012;7(2):e31259. 154. D’Angelo L, Myer NM, Myers KA. MCAK-mediated regulation of endothelial cell microtubule dynamics is mechanosensitive to myosin-II contractility. Molecular Biology of the Cell. 2017;28(9):1223-37. 155. Wang Z, Khan S, Sheetz MP. Single cytoplasmic dynein molecule movements: characterization and comparison with kinesin. 1995;69(5):2011-23. 156. Mallik R, Petrov D, Lex SA, King SJ, Gross SP. Building Complexity: An In Vitro Study of Cytoplasmic Dynein with In Vivo Implications. 2005;15(23):2075-85. 157. Ross JL, Wallace K, Shuman H, Goldman YE, Holzbaur ELF. Processive bidirectional motion of dynein–dynactin complexes in vitro. 2006;8(6):562-70. 158. Darshan MS, Loftus MS, Thadani-Mulero M, Levy BP, Escuin D, Zhou XK, et al. Taxane-Induced Blockade to Nuclear Accumulation of the Androgen Receptor Predicts Clinical Responses in Metastatic Prostate Cancer. Cancer Research. 2011;71(18):6019-29. 159. Goto T, Takano M, Hirata J, Tsuda H. The involvement of FOXO1 in cytotoxic stress and drug-resistance induced by paclitaxel in ovarian cancers. British Journal of Cancer. 2008;98(6):1068-75. 160. Gan L, Chen S, Wang Y, Watahiki A, Bohrer L, Sun Z, et al. Inhibition of the Androgen Receptor as a Novel Mechanism of Taxol Chemotherapy in Prostate Cancer. 2009;69(21):8386-94. 161. Sunters A, Madureira PA, Pomeranz KM, Aubert M, Brosens JJ, Cook SJ, et al. Paclitaxel-Induced Nuclear Translocation of FOXO3a in Breast Cancer Cells Is Mediated by c-Jun NH2-Terminal Kinase and Akt. Cancer Research. 2006;66(1):212-20. 162. Li X, Qiu W, Liu B, Yao R, Liu S, Yao Y, et al. Forkhead box transcription factor 1 expression in gastric cancer: FOXM1 is a poor prognostic factor and mediates resistance to docetaxel. Journal of Translational Medicine. 2013;11(1):204. 163. Haldar S, Chintapalli J, Croce CM. Taxol Induces bcl-2 Phosphorylation and Death of Prostate Cancer Cells. Cancer Research. 1996;56(6):1253-5. 164. Kwon CH, Park HJ, Choi Y, Won YJ, Lee SJ, Park DY. TWIST mediates resistance to paclitaxel by regulating Akt and Bcl-2 expression in gastric cancer cells. Tumor Biology. 2017;39(10):101042831772207. 165. Wang C, Huang S-B, Yang M-C, Lin Y-T, Chu IH, Shen Y-N, et al. Combining Paclitaxel with ABT-263 Has a Synergistic Effect on Paclitaxel Resistant Prostate Cancer Cells. PLOS ONE. 2015;10(3):e0120913. 166. D'Amico A. US Food and Drug Administration Approval of Drugs for the Treatmen tof Prostate Cancer: A New Era has Begun. J of Clinical Oncology. 2013:362-264. 167. Ciorsdaidh, Frances, Bender A, Peter, Korb O, Kern O, et al. Design, Synthesis, and Biological Evaluation of an Allosteric Inhibitor of HSET that Targets Cancer Cells with Supernumerary Centrosomes. Chemistry & Biology. 2013;20(11):1399- 410.

153 154

168. Sekino Y, Oue N, Koike Y, Shigematsu Y, Sakamoto N, Sentani K, et al. KIFC1 Inhibitor CW069 Induces Apoptosis and Reverses Resistance to Docetaxel in Prostate Cancer. Journal of Clinical Medicine. 2019;8(2):225. 169. Rickert KW, Schaber M, Torrent M, Neilson LA, Tasber ES, Garbaccio R, et al. Discovery and biochemical characterization of selective ATP competitive inhibitors of the human mitotic kinesin KSP. Archives of Biochemistry and Biophysics. 2008;469(2):220-31. 170. Aoki S, Ohta K, Yamazaki T, Sugawara F, Sakaguchi K. Mammalian mitotic centromere-associated kinesin (MCAK). FEBS Journal. 2005;272(9):2132-40. 171. Moreno JG, Miller MC, Gross S, Allard WJ, Gomella LG, Terstappen LWMM. Circulating tumor cells predict survival in patients with metastatic prostate cancer. Urology. 2005;65(4):713-8. 172. Walden P, Globina Y, Nieder A. Induction of anoikis by doxazosin in prostate cancer cells is associated with activation of caspase-3 and a reduction of focal adhesion kinase. Urol Res. 2014;32:261-5. 173. Barrallo-Gimeno A. The Snail genes as inducers of cell movement and survival: implications in development and cancer. Development. 2005;132(14):3151-61. 174. Chun J, Joo E, Kang M, Kim Y. Platycodin D induces anoikis and caspase-mediated apoptosis via p38 MAPK in AGS human gastric cancer cells. J of Cell Biochem. 2013;114:456-70. 175. Zhang P, Xing Z, Li X, Song Y, Zhao J, Xiao Y, et al. Tyrosine receptor kinase B silencing inhibits anoikis‑resistance and improves anticancer efficiency of sorafenib in human renal cancer cells. International Journal of Oncology. 2016. 176. Paoli P, Giannoni E, Chiarugi P. Anoikis molecular pathways and its role in cancer progression. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research. 2013;1833(12):3481-98. 177. Martin SK. Mechanisms of therapeutic resistance in castration resistant prostate cancer. UKnowledge. 2015:1-187. 178. Sweeney CJ, Chen Y-H, Carducci M, Liu G, Jarrard DF, Eisenberger M, et al. Chemohormonal Therapy in Metastatic Hormone-Sensitive Prostate Cancer. New England Journal of Medicine. 2015;373(8):737-46. 179. Thiery JP, Sleeman JP. Complex networks orchestrate epithelial–mesenchymal transitions. Nature Reviews Molecular Cell Biology. 2006;7(2):131-42. 180. Peinado H, Olmeda D, Cano A. Snail, Zeb and bHLH factors in tumour progression: an alliance against the epithelial phenotype? Nature Reviews Cancer. 2007;7(6):415-28. 181. Wang X, Lee SO, Xia S, Jiang Q, Luo J, Li L, et al. Endothelial Cells Enhance Prostate Cancer Metastasis via IL-6->Androgen Receptor->TGF- ->MMP-9 Signals. Molecular Cancer Therapeutics. 2013;12(6):1026-37. 182. McCabe NP, De S, Vasanji A, Brainard J, Byzova TV. Prostate cancer specific integrin αvβ3 modulates bone metastatic growth and tissue remodeling. Oncogene. 2007;26(42):6238-43. 183. Bradley DA, Daignault S, Ryan CJ, DiPaola RS, Smith DC, Small E, et al. Cilengitide (EMD 121974, NSC 707544) in asymptomatic metastatic castration resistant prostate cancer patients: a randomized phase II trial by the prostate cancer clinical trials consortium. Investigational New Drugs. 2010;29(6):1432-40.

154 155

184. Corn PG. The tumor microenvironment in prostate cancer: elucidating molecular pathways for therapy development. Cancer Management and Research. 2012:183. 185. Kudo-Saito C, Shirako H, Takeuchi T, Kawakami Y. Cancer Metastasis Is Accelerated through Immunosuppression during Snail-Induced EMT of Cancer Cells. Cancer Cell. 2009;15(3):195-206. 186. McConkey DJ, Choi W, Marquis L, Martin F, Williams MB, Shah J, et al. Role of epithelial-to-mesenchymal transition (EMT) in drug sensitivity and metastasis in bladder cancer. Cancer and Metastasis Reviews. 2009;28(3-4):335-44. 187. Park SI, Zhang J, Phillips KA, Araujo JC, Najjar AM, Volgin AY, et al. Targeting Src Family Kinases Inhibits Growth and Lymph Node Metastases of Prostate Cancer in an Orthotopic Nude Mouse Model. Cancer Research. 2008;68(9):3323- 33. 188. Davies M, Koul D, Dhesi H, Berman R, McDonnell T, McConkey D, et al. Regulation of Akt/PKB activity, cellular growth, and apoptosis in prosate carcinoma cells by MMAC/PTEN. Cancer Research. 1999;59(11):2551-6. 189. Hong S-W, Shin J-S, Moon J-H, Kim Y-S, Lee J, Choi EK, et al. NVP-BEZ235, a dual PI3K/mTOR inhibitor, induces cell death through alternate routes in prostate cancer cells depending on the PTEN genotype. Apoptosis. 2014;19(5):895-904. 190. McCarty M. Targeting multiple signaling pathways as a strategy for managing prostate cancer: multifocal signal modulation therapy. Integr Cancer Ther. 2004;3(4):349-80. 191. Floc'h N, Kinkade CW, Kobayashi T, Aytes A, Lefebvre C, Mitrofanova A, et al. Dual Targeting of the Akt/mTOR Signaling Pathway Inhibits Castration-Resistant Prostate Cancer in a Genetically Engineered Mouse Model. Cancer Research. 2012;72(17):4483-93. 192. Wisinski KB, Tevaarwerk AJ, Burkard ME, Rampurwala M, Eickhoff J, Bell MC, et al. Phase I Study of an AKT Inhibitor (MK-2206) Combined with Lapatinib in Adult Solid Tumors Followed by Dose Expansion in Advanced HER2+ Breast Cancer. Clinical Cancer Research. 2016;22(11):2659-67. 193. Massagué J, Blain SW, Lo RS. TGFβ Signaling in Growth Control, Cancer, and Heritable Disorders. Cell. 2000;103(2):295-309. 194. Kim SJ, Im YH, Markowitz SD, Bang YJ. Molecular mechanisms of inactivation of TGF-β receptors during carcinogenesis. Cytokine & Growth Factor Reviews. 2000;11(1-2):159-68. 195. Welch DR, Fabra A, Nakajima M. Transforming growth factor beta stimulates mammary adenocarcinoma cell invasion and metastatic potential. 1990;87(19):7678-82. 196. Li P, Yang R, Gao W-Q. Contributions of epithelial-mesenchymal transition and cancer stem cells to the development of castration resistance of prostate cancer. Molecular Cancer. 2014;13(1):55. 197. Liu G-L, Yang H-J, Liu T, Lin Y-Z. Expression and significance of E-cadherin, N- cadherin, transforming growth factor-β1 and Twist in prostate cancer. 2014;7(1):76-82. 198. Pu H, Collazo J, Jones E, Gayheart D, Sakamoto S, Vogt A, et al. Dysfunctional Transforming Growth Factor- Receptor II Accelerates Prostate Tumorigenesis in the TRAMP Mouse Model. Cancer Research. 2009;69(18):7366-74.

155 156

199. Brattain MG, Ko Y, Banerji SS, Wu G, Willson JKV. Defects of TGF-β receptor signaling in mammary cell tumorigenesis. Journal of Mammary Gland Biology and Neoplasia. 1996;1(4):365-72. 200. Akhurst RJ, Derynck R. TGF-β signaling in cancer – a double-edged sword. 2001;11(11):S44-S51. 201. Lahn M, Herbertz S, Sawyer JS, Stauber AJ, Gueorguieva I, Driscoll KE, et al. Clinical development of galunisertib (LY2157299 monohydrate), a small molecule inhibitor of transforming growth factor-beta signaling pathway. Drug Design, Development and Therapy. 2015:4479. 202. Yingling JM, McMillen WT, Yan L, Huang H, Sawyer JS, Graff J, et al. Preclinical assessment of galunisertib (LY2157299 monohydrate), a first-in-class transforming growth factor-β receptor type I inhibitor. Oncotarget. 2018;9(6). 203. Faivre SJ, Santoro A, Kelley RK, Merle P, Gane E, Douillard J-Y, et al. A phase 2 study of a novel transforming growth factor-beta (TGF-β1) receptor I kinase inhibitor, LY2157299 monohydrate (LY), in patients with advanced hepatocellular carcinoma (HCC). Journal of Clinical Oncology. 2014;32(3_suppl):LBA173-LBA. 204. Carpentier AF, Brandes AA, Kesari S, Sepúlveda JM, Wheeler H, Chinot OL, et al. Safety interim data from a three-arm phase II study evaluating safety and pharmacokinetics of the oral transforming growth factor-beta (TGF-ß) receptor I kinase inhibitor LY2157299 monohydrate in patients with glioblastoma at first progression. Journal of Clinical Oncology. 2013;31(15_suppl):2061-. 205. Paller C, Pu H, Begemann D, Wade C, Hensley PJ, Kyprianou N. TGF-β receptor I inhibitor enhances response to enzalutamide in a pre-clinical model of advanced prostate cancer. Prostate 2019;79(1):31-43. 206. Safarzadeh E, Delazar A, Kazemi T, Orangi M, Shanehbandi D, Esnaashari S, et al. The Cytotoxic and Apoptotic Effects of Scrophularia Atropatana Extracts on Human Breast Cancer Cells. Advanced Pharmaceutical Bulletin. 2017;7(3):381-9. 207. Deng Y, Li Y, Yang F, Zeng A, Yang S, Luo Y, et al. The extract from Punica granatum (pomegranate) peel induces apoptosis and impairs metastasis in prostate cancer cells. Biomedicine & Pharmacotherapy. 2017;93:976-84. 208. Malik A, Afaq F, Sarfaraz S, Adhami VM, Syed DN, Mukhtar H. Pomegranate fruit juice for chemoprevention and chemotherapy of prostate cancer. 2005;102(41):14813-8. 209. Yang H, Dou P. Targeting Apoptosis Pathway with Natural Terpenoids: Implications for Treatment of Breast and Prostate Cancer. Curr Drug Targets. 2010;11(6):733-44. 210. Van Der Steen T, Tindall D, Huang H. Posttranslational Modification of the Androgen Receptor in Prostate Cancer. International Journal of Molecular Sciences. 2013;14(7):14833-59. 211. Hensley PJ, Cao Z, Pu H, Dicken H, He D, Zhou Z, et al. Predictive and targeting value of IGFBP-3 in therapeutically resistant prostate cancer. Am J Clin Exp Urol. 2019;7(3):188-202. 212. Galletti G, Matov A, Beltran H, Fontugne J, Miguel Mosquera J, Cheung C, et al. ERG induces taxane resistance in castration-resistant prostate cancer. Nature Communications. 2014;5(1):5548.

156 157

213. Qiu S, Deng L, Bao Y, Jin K, Tu X, Li J, et al. Reversal of docetaxel resistance in prostate cancer by Notch signaling inhibition. Anticancer Drugs. 2018;29(9):871- 9. 214. Mimeault M, Rachagani S, Muniyan S, Seshacharyulu P, Johansson S, Datta K, et al. Inhibition of hedgehog signaling improves the anti-carcinogenic effects of docetaxel in prostate cancer. oncotarget. 2015;28(6):3887-903. 215. Shimizu Y, Tamada S, Kato M, Hirayama Y, Takeyama Y, Iguchi T, et al. Androgen Receptor Splice Variant 7 Drives the Growth of Castration Resistant Prostate Cancer without Being Involved in the Efficacy of Taxane Chemotherapy. Journal of Clinical Medicine. 2018;7(11):444. 216. Mout L, Wit Rd, Sturrman D, Verhoef E, Mathijssen R, Ridder Cd, et al. Testosterone Diminishes Cabazitaxel Efficacy and Intratumoral Accumulation in a Prostate Cancer Xenograft Model. EBioMedicine. 2018;27:182-6. 217. Yao K, Wu J, Zhang J, Bo J, Hong Z, Zu H. Protective Effect of DHT on Apoptosis Induced by U18666A via PI3K/Akt Signaling Pathway in C6 Glial Cell Lines. Cellular and Molecular Neurobiology. 2016;36(5):801-9. 218. Liu R, Ding L, Yu M-H, Wang H-Q, Li W-C, Cao Z, et al. Effects of dihydrotestosterone on adhesion and proliferation via PI3-K/Akt signaling in endothelial progenitor cells. Endocrine. 2014;46(3):634-43. 219. Hamada A, Sissung T, Price DK, Danesi R, Chau CH, Sharifi N, et al. Effect of SLCO1B3 Haplotype on Testosterone Transport and Clinical Outcome in Caucasian Patients with Androgen-Independent Prostatic Cancer. Clinical Cancer Research. 2008;14(11):3312-8. 220. Sissung TM, Ley AM, Strope JD, McCrea EM, Beedie S, Peer CJ, et al. Differential Expression of OATP1B3 Mediates Unconjugated Testosterone Influx. Molecular Cancer Research. 2017;15(8):1096-105. 221. Li J, Fu X, Cao S, Li J, Xing S, Li D, et al. Membrane-associated androgen receptor (AR) potentiates its transcriptional activities by activating heat shock protein 27 (HSP27). Journal of Biological Chemistry. 2018;293(33):12719-29. 222. Antonarakis ES, Lu C, Wang H, Luber B, Nakazawa M, Roeser JC, et al. AR-V7 and Resistance to Enzalutamide and Abiraterone in Prostate Cancer. New England Journal of Medicine. 2014;371(11):1028-38. 223. Iusuf D, Hendrikx JJMA, van Esch A, van de Steeg E, Wagenaar E, Rosing H, et al. Human OATP1B1, OATP1B3 and OATP1A2 can mediate the in vivo uptake and clearance of docetaxel. International Journal of Cancer. 2015;136(1):225-33. 224. De Morrée ES, Böttcher R, Van Soest RJ, Aghai A, De Ridder CM, Gibson AA, et al. Loss of SLCO1B3 drives taxane resistance in prostate cancer. British Journal of Cancer. 2016;115(6):674-81. 225. Natasha Kyprianou EK, David Bradbury, Juong Rhee. . Bcl2 overexpression delays radiation induced apoptosis without affecting the clonogenic survival of human prostate cancer cells. Int J Cancer. 1997;70:341-8. 226. Li Yan Khor JM, William Shipley, Alan Pollack. Bcl2 and Bax expression predict prostate cancer outcome in men treated with androgen deprivation and radiotherapy on radiation therapy oncology group protocol 92-02. Clin Cancer Res. 2007;13(12):3585-90.

157 158

227. McDonnell TJ, Troncoso P, Brisbay SM, Logothetis C, Chung LWK, Hsieh J-T, et al. Expression of the Protooncogene Bcl-2 in the Prostate and Its Association with Emergence of Androgen-independent Prostate Cancer. Cancer Research. 1992;52(24):6940. 228. Tsuji M, Murakami Y, Kanayama H, Sano T, Kagawa S. Immunohistochemical analysis of Ki-67 antigen and Bcl-2 protein expression in prostate cancer: effect of neoadjuvant hormonal therapy. Br J Urol. 1998;81(1):116-21. 229. Zellweger T, Ninck C, Mirlacher M, Annefeld M, Glass A, Gasser T, et al. Tissue microarray analysis reveals prognostic significance of syndecan-1 expression in prostate cancer. . The Prostate. 2003;55(1):20-9. 230. Yuting Lin JF, Richard Hiipakka, John Kokontis, Jialing Xiang. Up-Regulation of Bcl2 is required for the progression of prostate cancer cells from an androgen dependent to an androgen independent growth stage. Cell Research. 2007;17:531- 6. 231. Xu Shi PG, Jared Muenzer, Ralph deVere White Bcl2 antisense transcripts decrease intracellular bcl2 expression and sensitze LNCaP Prostate Cancer Cells to Apoptosis Inducing Agents. Cancer Biotherapy and Radiopharmaceuticals. 2001;16(5). 232. Tamaki H, Harashima N, Hiraki M, Arichi N, Nishimura N, Shiina H, et al. Bcl-2 family inhibition sensitizes human prostate cancer cells to docetaxel and promotes unexpected apoptosis under caspase-9 inhibition. Oncotarget. 2014;5(22):11399- 412. 233. Navone NM, Van Weerden WM, Vessella RL, Williams ED, Wang Y, Isaacs JT, et al. Movember GAP1 PDX project: An international collection of serially transplantable prostate cancer patient-derived xenograft (PDX) models. The Prostate. 2018;78(16):1262-82. 234. Zhang W, Liu B, Wu W, Li L, Broom BM, Basourakos SP, et al. Targeting the MYCN–PARP–DNA Damage Response Pathway in Neuroendocrine Prostate Cancer. Clinical Cancer Research. 2018;24(3):696-707. 235. Ellen Morree RvS, Ashraf Aghai, Corrina de Ridder, Peter de Bruijn, Inge Ghobadi, Herman Burger, Ron Mathijssen, Erik Weimer, Ronald de Wit, Wytske van Weerden. Understanding Taxanes in prostate Cancer; Importance of Intratumoral Drug Accumulation. The Prostate. 2016;76:927-36. 236. Oudard S, Fizazi K, Sengeløv L, Daugaard G, Saad F, Hansen S, et al. Cabazitaxel Versus Docetaxel As First-Line Therapy for Patients With Metastatic Castration- Resistant Prostate Cancer: A Randomized Phase III Trial—FIRSTANA. Journal of Clinical Oncology. 2017;35(28):3189-97. 237. Saitoh M. Involvement of partial EMT in cancer progression. The Journal of Biochemistry. 2018;164(4):257-64. 238. Singh S, Chakrabarti R. Consequences of EMT-Driven Changes in the Immune Microenvironment of Breast Cancer and Therapeutic Response of Cancer Cells. Journal of Clinical Medicine. 2019;8(5):642. 239. Wang S, Huang S, Sun YL. Epithelial-Mesenchymal Transition in Pancreatic Cancer: A Review. BioMed Research International. 2017;2017:1-10. 240. Wu C, Chen W-C, Chen M-F. The Response of Prostate Cancer to androgen Deprivation and Irradiation Due to Immune Modulation. Cancers. 2019;11(20).

158 159

241. Byrne NM, Nesbitt H, Ming L, McKeown SR, Worthington J, McKenna DJ. Androgen deprivation in LNCaP prostate tumour xenografts induces vascular changes and hypoxic stress, resulting in promotion of epithelial-to-mesenchymal transition. 2016;114(6):659-68. 242. Olson BM, Gamat M, Seliski J, Sawicki T, Jeffery J, Ellis L, et al. Prostate Cancer Cells Express More Androgen Receptor (AR) Following Androgen Deprivation, Improving Recognition by AR-Specific T Cells. Cancer Immunology Research. 2017;5(12):1074-85. 243. Ieiri I, Higuchi S, Sugiyama Y. Genetic polymorphisms of uptake (OATP1B1, 1B3) and efflux (MRP2, BCRP) transporters: implications for inter-individual differences in the pharmacokinetics and pharmacodynamics of statins and other clinically relevant drugs. Expert Opinion on Drug Metabolism & Toxicology. 2009;5(7):703-29. 244. Bronger H, König J, Kopplow K, Steiner H-H, Ahmadi R, Herold-Mende C, et al. ABCC Drug Efflux Pumps and Organic Anion Uptake Transporters in Human Gliomas and the Blood-Tumor Barrier. Cancer Research. 2005;65(24):11419. 245. Fletcher JI, Haber M, Henderson MJ, Norris MD. ABC transporters in cancer: more than just drug efflux pumps. Nature Reviews Cancer. 2010;10(2):147-56. 246. Prostate Cancer Clinical Trials. Prostate Cancer Foundation. 2020.

159 160

Vita

Education B.S. Biochemistry & Molecular Biology, Centre College, Danville Kentucky, 2014 Positions and Honors

Positions and Employment 2015-2020 Ph.D. Candidate, Graduate Research Assistant at the University of Kentucky- Department of Toxicology & Cancer Biology

2014-2015 Emergency Medical Technician (EMT-B), Rural Metro Ambulance Services 2012-2014 Technical Assistant for Biology & Molecular Biology Lab 2012-2013 Technical Assistant for Organic Chemistry II Lab

Other Experience and Professional Memberships 2018- Student mentor for incoming graduate students in the departments 2019 of Toxicology & Cancer Biology and Integrated Biomedical Sciences 2017- Graduate Student Congress: Member, Departmental 2019 Representative 2016- Biomedical Graduate Student Organization: Member, Treasurer 2019 2016- Women in STEM: Member, student mentor 2019

Peer-Reviewed Publications 1. Begemann D, Wang Y, Yang W, Kyprianou N. Androgens Modify Therapeutic Response to Cabazitaxel in Models of Advanced Prostate Cancer. 2020. The Prostate. In publication. 2. Pu H, Begemann D, Kyprianou N. Aberrant TGF-β Signaling Drives Castration Resistant Prostate Cancer in a Male Mouse Model of Prostate Tumorigenesis. 2017. Endocrinology. DOI: 10.1210/en.2017-00086. 3. Begemann D, Anastos H, Kyprianou N. Cell death under epithelial- mesenchymal transition control in prostate cancer therapeutic response. Review. International Journal of Urology. 2017. DOI: 10.1111/iju.13505. 4. Paller, C., Pu, H., Begemann, D., Wade, C., Hensley, P., Kyprianou, N. TGF-β Receptor 1 inhibitor enhances response to enzalutamide in a pre-clinical model of advanced prostate cancer. The Prostate. 2018. DOI: 10.1002/pros.23708.

160 161

Oral and Poster Presentations Department of Toxicology & Cancer Biology Seminar 2/2019 Oral research presentation “Overcoming Taxane Resistance in Advanced Prostate Cancer”

Society for Basic Urology Research (SBUR) National Meeting 11/2018 Poster Presentation, Rancho Mirage CA “Significance of Prostate Tumor Phenotypic Reprogramming in Resistance to Cabazitaxel Chemotherapy”

Markey Cancer Center Research Day 5/2018 Poster Presentation, University of Kentucky “TGF-B Signaling Cross Talk with AR Drives Castration Resistant Prostate Cancer”

Sanofi at University of Kentucky 6/2017 Research update on Sanofi pharmaceutical product Cabazitaxel “Journal Club” style article presentation of relevant literature

Centre College 8/2012 RICE Symposium (Research Internships and Creative Endeavors) Collaborative undergraduate research symposium Poster and oral presentation to interdisciplinary audience

161