University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content HOMOLOGOUS CELL SYSTEMS FOR THE STUDY OF PROGRESSION OF ANDROGEN-DEPENDENT PROSTATE CANCER TO CASTRATION- RESISTANT PROSTATE CANCER by Mengmeng Liang A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (PATHOBIOLOGY) December 2013 Copyright 2013 Mengmeng Liang ii “Nothing in life is to be feared, it is only to be understood. Now is the time to understand more, so that we may fear less.” Maria Sk łodowska-Curie (1867 – 1934) Polish physicist and chemist, First Female Nobel Laureate iii DEDICATION This dissertation is dedicated to my whole wonderful family, for their tremendous affection, endless patience and unconditional support. To my mother, Zhibo Yao, my parents-in-law Baohui Zhou and Chunrong Wang, and my husband, Xinyang Wang: I am so blessed to have you in my life. I also dedicate this work to my late father Weiming Liang. His words of inspiration and encouragement in pursuit of happiness and excellence still linger on and will be forever. iv ACKNOWLEDGMENTS I wish to express my greatest gratitude to my mentor Dr. Pradip Roy- Burman, first and foremost, for his scientific guidance, confidence placed in me and tremendous support throughout my Ph.D. studies. This work would not have been integrated into a complete body of research without his intelligence, experience and well-grounded knowledge in prostate cancer research. During the past six years, he has not only guided me extensively in science, but influenced me deeply with his own charisma to grow into a better person full of integrity and responsibility. What I have appreciated the most from Dr. Roy- Burman is the significant improvement I have gone through, transforming from initially being self-doubtful to become a self- confident individual. He has helped me to realize more potential that would be beneficial for my future life. It is definitely my greatest honor to become the 25th and also the last Ph.D. student Dr. Roy-Burman has trained during his entire well-accomplished mentoring life. I would like to acknowledge my committee members Dr. Cheng-Ming Chuong and Dr. Robert Maxson, along with my oral qualifying committee members Dr. Florence Hofman, Dr. Louis Dubeau and Dr. Randy Widelitz, for providing insightful advice as well as the enormous efforts to pass on a lot of interesting and useful knowledge in the first three-year course work. In addition, I also sincerely appreciate the help and encouragement from Dr. Baruch Frenkel, Dr. Jeremy Jones, Dr. Michael Stallcup, Dr. Wei Li and Dr. Libo Yao in the present and past. I wouldn’t have achieved what I have today without any of v them. I would like to especially thank Dr. Baruch Frenkel and Dr. Jeremy Jones for their tremendous help and moral support in the past several months. I thank Dr. Michael Stallcup and Dr. Wei Li for giving me the precious opportunity to start my scientific training at USC. It gives me great pleasure in acknowledging my aunt, Dr. Libo Yao, a successful scientist in the biomedical research, whose diligence, passion and dedication to science have initially aroused my interest in biology and pushed me further to pursue the Ph.D. study. I am very grateful for all my previous and current lab members: Kumkum Mittra, Dr. Chen Zhong, Dr. Shangxin Yang, Dr. Chun-Peng Liao, Dr. Minyoung Lim, Dr. Linda Pham, Dr. Helty Adisetiyo, Dr. Lauren Geary, Dr. Ari Aycock- Williams, Gohar Saribekyan, Erik Haw, Lily and Engracia. Thank you for having created a great working atmosphere in Roy-Burman’s lab, to make my every day in lab full of energy and never boring and to inspire me to learn new thingsand even face tough situations. I must acknowledge two people in particular with whom I have spent so much good times. I am so lucky to have our lab mom – Kumkum Mittra, who is a lab manager, a friend and sometimes a psychiatrist, amazingly all at once. With her help, we’d never need to worry about the material supplies and other lab management issues. She serves as one of my best oral English mentors by sharing so many intriguing stories as well as giving wise advice. Talking with her in my lunch break is my favorite relaxing time every day. As for Dr. Helty Adisetiyo, I don’t know exactly which category I should have put her to acknowledge, for the reason that she has been everywhere whenever I vi need her. She is the best technical support who has instructed me numerous experiments and shared all her expertise without concealing anything. She is one of the most helpful colleagues whom I have collaborated in at least five projects within my PhD studies. She is my best friend who has always laughed with me together, listened to my complaints and helped me through difficult times. I am so blessed to have this “acquired” sister. This thesis would not have been possible without the tremendous help from my former and current lab members, classmates and fellow collaborators. A special acknowledgement must be extended to Dr. Cheng-Ming Chuong, Dr. Randy Widelitz and Dr. Jeremy Jones who have first shown their interest on my androgen receptor splice variants. I would like to thank all these people for their contribution on this project: Dr. Gill Parkash for lending his brilliant scientists – Dr. Xiuqing Li and Dr. Ren Liu to assist essentially on cloning and immunoprecipitation assays, Dr. Robert Matusik for providing the plasmid, Dr. Jeremy Jones for performing a large amount of luciferase assays which have conveyed more interesting potentials of these variants, Dr. Helty Adisetiyo for conducting western blot analyses, Han Wang for his important guidance at the initial stage and Dr. Jiang Zhong for the single-cell analyses. I am also thankful to Cathleen Chiu and Ang Li of Dr. Cheng-Ming Chuong’s lab, Yang Li, Dr. Ni Zeng and Anketse Kassa of Dr. Bangyan Stiles’s lab, and Niyati Jhaveri of Dr. Hofman’s lab for generously lending reagents and lab equipment. I share the whole credit of the cell line project with the following collaborators: Dr. Chun- vii Peng Liao, Dr. Michael Cohen, Dr. Andrea Flesken-Nikitin, Dr. Joseph Jeong, and Dr. Alexander Nikitin. I would also like to thank Ryan Park, Lindsey Hughes and Grant Dagliyan from the USC Molecular Imaging Center, Ernesto Barron and Douglas Hauser from the USC Norris Comprehensive Cancer Center Cell and Tissue Imaging Core for providing technical and equipment support regarding this project. Without their help, this work wouldn’t have been completed in such a short time with solid data. I am indebted to Lisa Doumak, Raquel Rodriguez, Cynthia Carreon, Lisa Mendoza and Michele King for their help and support on the administrative matters. Their constant efforts have brought up so many fun and memorable experience for me as a Ph.D. student in the Department of Pathology. My life at USC would not have been this delightful without all my good friends I have met here: Cathleen Chiu, Dr. Xiuqing Li, Dr. Fang Chen, Niyati Jhaveri, Dr. Ni Zeng, Zhi Liu, Dr. Yujiao Sun, Dr. Biquan Luo, Dr. Shuting Sun, Dr. Vicky Yamamoto, Dr. Kaijie He, Dr. Hua Fang, Dr. Ying Liu, Xinyi Wang, Leng-Ying Chen, Yang Li, Zhengfei Lu, Ang Li, Dr. Wei Liang, and Likun Fei. I truly appreciate their technical discussions and moral support in every aspect. I am also grateful for Dr. Shanshan Xu, Dr. Jie Du, Zhongxiao Yang, Ying Liu, Dr. Ning Lin, Dr. Jianfu Chen, Dr. Haoyuan Liu, Fei Xu, Dr. Yu Wu, Kun Qian and Dr. Ke Yuan for their company and for listening whenever I needed them. I would like to thank my old best friends: Yanyan Zhang, Miao Li, Yan He, Dr. Bin Xu and Meng Meng for their care and attention from every corner of the world. A special viii gratitude would go to Dr. Shengxi Guan, Dr. Jingjing Wang, Dr. Jin Su and Dr. Jian Zhangfor their patient and meticulous training in the early stage of my lab life that has paved the way for my current progress. Last but definitely not least, I owe everything to my family: my grandmother Tao Jiang, my mother Zhibo Yao, my parents-in-law Baohui Zhou and Chunrong Wang, and my husband Xinyang Wang. I can’t find words to describe how much I love and cherish every one of them in my life. I want to thank my grandma - Tao Jiang, for her constant love and care. My mom, Zhibo Yao, is the most wonderful person in the world. She is kind, smart and independent. She is my role model that I would always strive to become like. I admire her unprecedented strength undergoing the toughest moment when my dad passed away. She has then made the selfless but difficult decision to push me back to finish my Ph.D. study. I have to also extend my appreciation to her thoughtfulness in the last year. She has come to lab with me many times on weekends, accompanied me when I stayed up late and relieved my pressure by cooking so much good food. I would have never arrived at this step without her. I express my sincere gratitude to my parents-in-law, Baohui Zhou and Chunrong Wang, for their extensive understanding and support throughout my Ph.D. study. I am extremely thankful to my husband, my best friend and my love for life, Xinyang Wang, for his unconditional love and patience. He indulges me, takes the best care of me and stands by me whenever I feel lost. I appreciate him ix wholeheartedly for waiting patiently in the past two years for me to finish. With him around, I have always felt more fearless and energetic to achieve my goals. x TABLE OF CONTENTS Epigraph ii Dedication iii Acknowledgments iv List of Tables xii List of Figures xiii List of Abbreviations xiv Abstract xvi Chapter One: Introduction 1.1 Prostate cancer 1 1.2 Androgen action in non-malignant and cancerous tissues of prostate 2 1.2.1 Androgen-androgen receptor (AR) signaling axis 2 1.2.2 Androgen action in normal prostate 5 1.2.3 Androgen action in prostate cancer 6 1.3 Current model systems used in prostate cancer research 10 1.3.1 Human immortalized cell lines and xenografts 11 1.3.2 Mouse models used for PCa 12 1.3.3 cPten -/- L mouse model of prostate cancer used in this study 13 1.4 Hypotheses and experimental strategies 15 Chapter Two: Mouse Prostate Cancer Cell Lines Established from Primary and Post-castration Recurrent Tumors 2.1 Abstract 20 2.2 Introduction 22 2.3 Materials and methods 24 2.4 Results 29 2.5 Discussion 48 Chapter Three: Novel Androgen Receptors Splice Variants in Homologous Systems of Mouse Prostate Cancer 3.1 Abstract 53 3.2 Introduction 55 3.3 Materials and methods 57 xi 3.4 Results 64 3.5 Discussion 90 Chapter Four: Discussions and Remarks about Future Studies 4.1 Discussions 96 4.2 Future studies 101 Bibliography 105 xii LIST OF TABLES Table 2-1: Primer sets used in real-time PCR and genotyping 26 Table 2-2: Representative karyotypes of the cell lines 38 Table 2-3: The comparison of the incidence and weights of the tumors formed 38 Table 3-1: 3’ RACE and nested PCR primers 60 Table 3-2: PCR primers for cloning of mAR-Vabc 60 Table 3-3: PCR primers for cloning of mAR-Va-myc and mAR-Vc-myc 60 Table 3-4: Primer sets for conventional PCR 62 Table 3-5: ARV specific primer sets for real-time PCR 62 Table 3-6: Summary of transcription and translation products of mAR-Vs 72 xiii LIST OF FIGURES Figure 2-1: Characteristics of the E2/E4 and cE1/cE2 cell lines 32 Figure 2-2: The comparison of proliferation rates of the cell lines in culture 35 Figure 2-3: Microanatomic analyses of tumors induced by the cell lines in male or female in NOD.SCID mice. 39 Figure 2-4: Analysis for the expression of EMT-related markers in the cell lines. 42 Figure 2-5: Characteristics of a new murine prostatic epithelial cell line E8. 45 Figure 3-1: Androgen receptor (AR) protein expression status in prostate cancer cell lines and prostatic tissues 66 Figure 3-2: Identification of new mouse AR splice variants 69 Figure 3-3: mRNA Expression of mARVs in mouse prostate cancer cell lines. 74 Figure 3-4: The relative expression of mAR-Vs were increased in response to androgen withdrawal. 78 Figure 3-5: Expression of mAR-Vs in mouse prostate 82 Figure 3-6: The transcriptional activity of AR-FL is inhibited by mAR-Vs in the presence of androgen. 87 Figure 3-7: Schematic illustrations of mAR-Vabc with previously published human and mouse ARVs. 92 Figure 3-8: Molecular and protein structures of mAR-Vd 95 xiv LIST OF ABBREVIATIONS AD – Androgen dependent ADCa – Androgen-dependent prostate cancer ADI-Ca – Androgen depletion independent cancer ADT – Androgen deprivation therapy AF – Activation function AIS – Androgen insensitivity syndromes AR – Androgen receptor AREs – Androgen response elements AR-FL – AR full-length AR-Vs – Androgen receptor splice variants BLI – Bioluminescence imaging CAFs – Cancer-associated fibroblasts CEs – Cryptic exons CgA – Chromgranin A CK – Cytokeratin CR – Castration-resistant CRPC – Castration-resistant prostate cancer CSCs – Cancer stem cells CSS – Charcoal stripped serum CTD – COOH-terminal domain DBD – DNA-binding domain xv DHT – Dihydrotestosterone EMT– Epithelial-mesenchymal transition EST – Expressed sequence tags GEMMs – Genetically engineered mouse models hAR – Human AR HAT – Histone acetyltransferase HSP – Heat shock protein LBD – Ligand binding domain mAR – Mouse AR mPIN – Murine prostatic intraepithelial neoplasm NLS – Nuclear localization signal NTD – NH 2 -terminal domain PAP – Prostatic acid phosphatase PB – Probasin PCa – Prostate cancer PSA – Prostate-specific antigen PSCA – Prostate stem cell antigen PTEN – Phosphatase and tensin homolog deleted on chromosome 10 RACE – Rapid amplification of cDNA ends Syph – Synaptophysin TAU – Transactivation units xvi ABSTRACT Despite immense research progress made in recent years, the recurrence of castration-resistant prostate cancer (CRPC) still remains poorly understood. Androgen receptor (AR), an essential player during the differentiation and maintenance of normal prostate as well as the initiation of prostate cancer, has been widely accepted to be the “Partner-in-Crime” even in the stage of CRPC, mediating and being regulated by numerous factors and signaling pathways. Therefore, to decipher the role of AR and its regulation in contributing to the occurrence of CRPC has become a priority topic in the prostate cancer (PCa) research. This dissertation describes establishment of a study system composed of murine PCa cell lines derived from tumors of the conditional Pten deletion mouse model of prostate adenocarcinoma obtained at two distinct phases of the namely androgen-dependent (AD) growth phase and the castration-resistant growth phase. We have characterized five mouse PCa cell lines - E2/E4 and E8 from two separate AD tumors and cE1/cE2 from one CR tumor. All the cell lines manifest biallelic deletion of the Pten gene, corresponding to epithelial origin since the Cre/LoxP system used in the modeling is targeted to prostate epithelium. Analyses of various molecular expressions as well as morphology of these cells suggest a degree of epithelial-mesenchymal transition (EMT) in E2/E4 but not much in E8, which possesses increased epithelial phenotype. The cell lines from the CRPC tumor generally display epithelial characteristics and xvii demonstrate significantly better growth capacity as compared to E-series in the absence of androgen, although sensitivity to supplemented androgen is retained. All the cell lines established are able to induce in vivo tumor growth in immunodeficient mice. However, while E8 and cE1/cE2 give rise to adenocarcinomas, E2 and E4 yields tumors that are akin to sarcomatoid carcinoma expressing protein makers of both epithelial and stromal cells. In this regard, there is a fairly good correlation between the in vitro and in vivo properties. Among the multiple mechanisms by which AR activity may be aberrantly regulated, we have focused our studies of the cell lines to determine the status of the androgen receptor splice variants (AR-Vs). We have identified a set of novel mouse AR-Vs (termed as mAR-Vabc) in these homologous mouse PCa cell systems. Their expression is present in all five cell lines and is generally up- regulated in response to androgen ablation. The relative proportion of these mAR-Vs varies with respect to the stage of the disease from which the cell lines originated. In cE-series, mAR-Vb and mAR-Vc are more abundant than that in E- series. However, mAR-Va displays an opposite profile of expression. The mRNA of these mAR-Vs is detected in the normal prostates; however, their abundance relative to the full-length AR (mAR-FL) is significantly increased in the tumor tissues. Interestingly, among the normal prostates, aged prostates harbor the highest levels. These findings implicate a potential role of these mAR-Vs in both etiological and normal biological processes. Structurally, mAR-Va is similar to xviii mAR-FL except lacking the ligand binding domain, and mAR-Vb and AR-Vc, differing in C-terminal sequences, basically harbor the N-terminal transcription activation domain (NTD) of mAR-FL. Functionally, as tested via reporter gene transcription assays in non-prostate heterologous COS-1 cells, all of these variants appear to lack the ability to activate AR response elements whether androgen is supplemented or not. However, each one, to a variable extent, can suppress the activity of AR-FL in co-transfection assays and in the presence of a low level of androgen. These variants appear to serve like transduced NTD “decoy” molecules that were reported before. Based on these findings, it is speculated that AR, that is necessary for both AD and CR prostatic tumors may be negatively regulated by production of the AR natural decoys to maintain a balance in overall AR functions in the activated and reactivated states. These decoys may also have a similar biological role in suppressing or down-regulating AR function in the ageing process. These are exciting clues and deserve to be further investigated. A critically important topic then would be to determine the mechanisms by which abnormal splice variants are formed. Once this mechanism is unveiled, it might be possible to design small molecular inducers for increased expression of these decoys for potential therapy or prevention of prostate cancer, in a unique but still natural manner. 1 CHAPTER ONE: INTRODUCTION 1.1 Prostate cancer Prostate cancer (PCa) is the most common malignancy and second leading cause of cancer-related mortality affecting men in the western world. Approximately 1 man in 6 will be diagnosed with prostate cancer, and about 1 out of 36 will die of this disease in their lifetime. According to the estimates by American Cancer Society in 2013, around 238, 590 new cases will be diagnosed and about 29,720 men will die of prostate cancer in the United States, that is, 28% of new cancer cases and 10% of cancer-related deaths in men 1 . If detected early, the localized disease can be cured by the conventional treatment regimens including surgery and/or irradiation therapy. Since advanced PCa initially develops and progresses in an androgen-dependent manner, androgen deprivation therapy (ADT) has been applied as a gold standard therapy for over 70 years to treat patients having androgen-dependent prostate cancer (ADCa), which is usually performed by medical (administration of gonadotropin-releasing hormone analogs and/or androgen receptor antagonist) or surgical castration 2, 3 .Most patients respond well to ADT, exhibiting a rapid tumor regression along with the amelioration of symptoms. Unfortunately, within a period of approximate 18-36 months, most tumors become refractory to androgen deprivation and manifest a relapse in a more aggressive and metastatic phenotype, best described as androgen depletion independent cancer (ADI-Ca) or castration- 2 resistant prostate cancer (CRPC) 4, 5 . The modified nomenclature castration resistant prostate cancer (CRPC) has replaced the previous use of term androgen independence, as it has been a central concept that the growth and survival of these advanced tumors remain dependent on AR function, as discussed below. At present, the lack of significantly effective therapies for the treatment of CRPC has made it one of the major clinical challenges in prostate cancer 6 . Therefore, the elucidation of mechanisms underlying the recurrence of CRPC has been a key topic in prostate cancer research, which could lead to the development of new therapeutic approaches. 1.2 Androgen action in non-malignant and cancerous tissues of prostate Androgens, the male sex steroids, are essential for male sexual differentiation, as well as the development and maintenance of male reproductive system including prostate. More importantly, its signaling via the androgen receptor (AR) axis is of vital importance during the initiation and progression of prostate cancer. 1.2.1 Androgen-androgen receptor (AR) signaling axis Testosterone, the primary circulating androgen, is produced mainly by the testis (90%-95%), with a minor amount (5%-10%) synthesized in the adrenal glands 7-9 . In prostate cells, systemic testosterone is rapidly converted into more potent metabolite dihydrotestosterone (DHT) by 5-alpha-reductase. Testosterone 3 and DHT exert their actions mainly by binding to androgen receptor (AR), a 110 kDa phosphoprotein and a member of steroid and nuclear receptor transcription factor superfamily 10 . In the absence of ligand, AR exists in the cytoplasm associated with heat shock protein (HSP) complexes 2, 11 . Once bound to androgens, AR dissociates from the HSP complex, undergoes conformational change, and translocates into nucleus where it dimerizes and engages with androgen response elements (AREs) in the cis-regions of a variety of target genes, such as prostate-specific antigen (PSA), prostatic acid phosphatase (PAP), TMPRSS2 etc 12 . The transcriptional activation of these target genes also relies on the recruitment of a plethora of co-activators, including those with intrinsic histone acetyltransferase (HAT) activity (P/CAF, CBP/p300) and SRC/p160 family, which together facilitate the chromatin remodeling and link AR with the basal transcription machinery to trigger the transcription of target genes 2, 11, 13, 14 . In addition to the canonical genomic activity, AR can also mediate the non-genomic responses, potentiating numerous signaling cascades independent of its nuclear translocation and DNA binding as a transcription factor, which usually originates at the plasma membrane or in the cytoplasm 11, 13, 15 . For example, in LNCaP cells, androgen binding can result in AR-mediated activation of Ras/ERK pathway through interacting with membrane c-Src tyrosine kinase, which can lead to increased cellular proliferation and protection from apoptosis 16 . Another study has also indicated that this Src-Ras-ERK-CREB pathway 4 induced by non-genomic action of AR has become constitutively active in LNCaP cells with an androgen-independent phenotype, which further consolidates the non-genomic role AR has played 17 . Accordingly, both genomic and non-genomic mechanisms must be considered all together for a thorough understanding of AR activity. Human and mouse AR gene are both located on their X chromosomes, composing of eight exons. As other steroid hormone receptors, AR protein modularly organizes into four distinct functional motifs: a NH 2 -terminal domain (NTD, encoded by exon1) harboring AR activation function -1 (AF-1), a DNA- binding domain(DBD, exon 2 and 3), a hinge region (exon 4) , and the COOH- terminal ligand binding domain (CTD, LBD, exon 4-8) containing AF-2 coactivator binding surface 2, 11, 13, 18-20 . The DBD and LBD of AR are highly conserved among different species and members of steroid nuclear receptor superfamily, whose 3-dimensional structures have already been determined by x-ray crystallography 21 ; conversely, AR NTD is greatly flexible and fails to be structured in solution 18, 22 . Mouse AR protein shares 97% and 85% of identical amino acids with respect to that of rat and human, mostly divergent on the sequences of NTD 23 . Starting from the conserved regions, LBD cavity at the CTD of AR facilitates binding of the ligand, resulting in the exposure of AF-2 coactivator region which, in turn, then serves as the binding site for AR coregulators with LxxLL motif (the SRC/p160 coactivators) 24 . More importantly, AF-2 regulates 5 the N/C interaction via FxxLF- and WxxLF motif of NTD, which may inhibit the transcriptional activity until AR engages with AREs of target genes in the nucleus 11, 18, 25 .The DBD recognizes the promoter and enhancer regions of androgen- responsive genes, composed of two zinc finger motifs. The hinge region, separating DBD and LBD, contains part of a bipartite nuclear localization signal (NLS) that is encoded together by DBD and hinge 18, 20, 26 . DBD/hinge region is essential for AR nuclear localization, receptor dimerization and DNA binding. NTD, accounting for ~57% of mouse AR protein 27 , plays multiple and dynamic roles in mediating the transcriptional activity of AR. AF-1, encompassing two transactivation units (TAU) termed as TAU1 and TAU5, is considered to be the major domain triggering AR transcription, independent of the CTD 28 . The FxxLF motif contained in NTD, in addition to regulating the N/C interaction with CTD, can also serve as the docking sites for various coactivators/corepressors, which in turns intricately mediates the relative architecture and activities of AR NTD 29- A better understanding of structural and functional elements of AR is required for the ultimate goal of developing novel therapeutic regimens for the treatment of advanced stage PCa. 1.2.2 Androgen action in normal prostate Prostates in human and mouse are functionally equivalent, which arrange into complex tubulo-alveolar glands composed of glandular acini embedded within the fibromuscular stroma 6, 33 .Three main cells types comprise the 6 glandular acini in the epithelial compartment of the prostate, including basal, secretory luminal and neuroendocrine cells that can be respectively differentiated by their cytokeratin (CK) expression patterns as well as their growth dependence/independence on the Androgen/AR 8, 33 . In normal prostate, androgens/AR signaling axis regulates a variety of key physiological possesses involved in the maintenance of prostate tissue, such as cellular proliferation, differentiation, as well as metabolic and secretory functions 2, 8 . The androgen- dependence of the epithelium also requires the paracrine factors provided by stromal AR positive cells 8, 34 . This complicated epithelial-stromal interaction contributes in maintaining the homeostasis in the normal prostate gland 2, 6, 8 . 1.2.3 Androgen action in prostate cancer The direct correlation between hormone and risk of prostate cancer was first identified by Huggins in 1941 35 , a discovery that led his Noble Prize in 1966, and that served as the basis for androgen ablation therapy, the current mainstay of treatment for advanced prostate cancer 9, 36 . In ADCa, the growth and survival of malignant prostate cells are still dependent on Androgen/AR function, thereby demonstrating a tumor regression concomitant with a reduced expression of PSA when treated with ADT 2, 6, 37 . Even in the stage of CRPC, despite the depleted androgen, AR continues to be critical for the growth and survival of the vast majority of tumors which is supported by the evidence that most CRPC patients retains high levels of AR 7 and PSA expression 38 . Moreover, unbiased gene expression studies demonstrate that AR induces altered gene expression patterns in CR tumor cells, which seems to be correlated with the growth of CRPC 39, 40 . Numerous molecular mechanisms have been described to restore the signaling activity of AR in CRPC 5, 20, 40-43 , including AR gene amplification 44 , increased expression of AR mRNA and protein 39, 45, 46 , selected mutations of AR that may confer a greater sensitivity to androgens and/or broader ligand specificity 47, 48 , altered ratio of expression levels of AR and its coregulators 42, 49 , up-regulation of constitutively active AR splice variants (ARV) 19, 50, 51 , endogenous expression of androgen synthetic enzymes that lead to de novo androgen synthesis 52 and up- regulation of cross-talk signaling pathways that activate AR 53-55 . Intensive efforts have been made to decipher the mechanisms how AR regains the aberrant activity in CRPC with the application of numerous in vitro and in vivo models of PCa, which seem to have woven an extremely complex network related to AR activation 56 . It is most likely that all of these mechanisms act in a non-mutually- exclusive fashion with certain pathways surpass over others, closely dependent on the specific phenotype of the disease 2, 5 . Staring from 2008, a series of alternatively spliced AR mRNA have been identified in various CR model systems, including human cell lines 22R ν1 and CWR-R1 derived from CWR xenografts, LuCaP xenografts, human VCaP and murine Myc-CaP cell lines 57-64 . With the application of RACE, EST (Expressed Sequence Tags) database search and yeast functional assays, many cryptic 8 exons (CEs) gathered within the intron 2/3a are shown to splice downstream of either exon 2 or exon 3 of prototype AR, which seem to encode the truncated AR proteins encompassing NTD and partial or intact DBD, but lack of hinge region and LBD. Seven ARVs containing varied compositions at 3’-end have been first discovered in 22Rv1 cells, including the best-characterized AR-V7/AR3 which is the mere one up to date proven to be translated into protein in human PCa cell lines and tissues 62, 63 . AR V567es , with the skip of exon 5-7, is found in the LuCaP 86.2 and LuCaP 136 xenografts derived from metastases of CRPC patients, through the regular RT-PCR using specific primer sets 59 . Although it remains unclear whether AR V567es would be translated into the endogenous protein in these xenografts, the predicted protein encoded by this ARV seems to contain NTD, DBD and the short hinge followed by 10 unique amino acids caused by a frame shift. Later on, more advanced technologies like next-generation sequencing analysis and genomic tilling arrays have also been employed in the search for ARVs in 22Rv1 and VCaP cells, leading to the discovery of more cryptic exons located in intron 3 and one novel “exon 9” downstream of exon 8 58, 61 . Compared to the diverse types of ARVs identified in human PCa systems, four ARVs (named as mAR-V1-4) are also detected in Myc-CaP cells derived from the transgenic mouse PCa model with prostate-specific oncogene myc expression 58 . Interestingly, mAR-V2 and mAR-V4 harbor the unique cryptic exons located in the distal regions on X chromosome outside the murine AR gene locus. Functionally, the majority of ARVs, despite their specific molecular 9 organizations with different sequences at 3’end, have displayed constitutive and ligand-independent transcriptional activity that in general could promote the CRPC growth in various model systems. More important, it has been implicated that detection of higher expression of AR-V7/AR3 and AR V567es in CRPC bone metastases than that of hormone-naïve patients, might be associated with the poor prognosis and clinical outcome 62, 63, 65, 66 . Meanwhile, there seems to be an exception in ARV family: AR8, the latest identified human ARV in CWR-R1 cells, is upregulated in CRPC cells but deduced to encode a protein composing of only NTD followed by the unique 33-amino acids due to a novel splice acceptor site at exon 3. This membrane-associated AR8 is likely to function through the intriguingly non-genomic mechanisms by potentiating the formation of signaling complex AR/EGFG/Src as well as enhancing AR phosphorylation, which contributes the CR growth and survival 57 . Despite the extensive advances in the quality and functional indications of ARVs, the mechanism how ARVs may contribute to convert ADCa to CRPC has yet to be elaborated. Should one or more ARVs be proven to be essential for the emergence of CRPC, it might initiate an innovative therapeutic mode to target the aberrant AR activity often found in CRPC stage. Although the resurgent AR activity in CRPC has been widely accepted nowadays thereby evoking extensive studies, it still remains largely elusive that which one of the mechanisms described earlier may function as the key determinants for the occurrence of CRPC. Of note, there is a great amount of 10 conflicting data regarding AR activity in the progression of PCa obtained from the studies using cell lines, xenografts, animal models and even clinical specimens, which are closely reflected by the vast heterogeneity of this disease 67-73 . Even more complicated, recent clinical data indicate that selected CRPC patients may benefit from androgen replacement therapies 74, 75 . Taken together, it has called attention to us that, besides the aberrantly active AR signaling in CRPC, we should not forget to consider the other side of the picture that many mechanisms function actively as well to counteract it. Therefore, rather than being as simple as that AR is essential for the initiation and growth of ADCa, AR signaling in the stage of CRPC might be under the intricate regulation through an extremely complex mechanism invoking both positive and negative control of AR in the cancer tissue. 1.3. Current model systems used in prostate cancer research Miscellaneous models have been applied in the field of PCa research to investigate the basic disease mechanisms and assess the preclinical efficacy 33, 76-81 . These models can be divided into three major categories: in vitro immortalized human cell lines established from metastatic lesions and xenografts, in vivo xenograft model using immunodeficient mouse as a recipient and in vivo genetically engineered mouse models (GEMMs) that mimic the human disease initiation and progression 76, 77 . Prostate cancer is extensively heterogeneous displaying both morphological and molecular genetic 11 complexities; therefore, it is difficult to elucidate the etiology of this disease through just one “perfect” system 33, 82, 83 . Instead, both in vitro and in vivo models should be employed in conjunction for the establishment of improved models that may mimic the disease onset and progression in a more accurate manner, which would advance our overall understandings in prostate cancer and develop more effective treatment. 1.3.1 Human immortalized cell lines and xenografts Immortalized human cell lines have been used extensively in the study of PCa research. The most commonly used human cell lines LNCaP, PC-3 and Du- 145, which are established from PCa lymph node, bond and brain metastases respectively, have been successful in the use of characterizing intracellular mechanisms underlying the disease progression 76 . Among these three cell lines, LNCaP is the only AR + cell line with the expression of a mutated AR (T887A), which is also used for the establishment of various CRPC subline variants 84-86 . 22Rv1 and CWR-R1 are derived from the relapsed CWR22 xenografts which are reported to express a novel AR mutation containing both mutation at H874Y and an in-frame tandem duplication of exon 3 87, 88 . As described in last section, a variety of ARVs have been initially identified in these two cell lines. However, the biggest weakness of the cell line system is the lack of the complex paracrine regulation derived from the tumor microenvironment. Xenografts models allow the foreign human tumor tissue/cell lines to propagate 12 and expand repetitively. However, an immune deficient system may not accurately recapitulate the normal processes of tumor development for the reason that the interaction between tumor cells and immune microenvironment is essential for the growth and metastasis of human PCa. 1.3.2 Mouse models used for PCa Although challenges still exist when extrapolating research data from mice to human, due to their high homology with human genome and short gestation time, a broad class of genetically engineered mouse models (GEMMs) has made the significant contribution to the study of prostate development and disease 33, 77, 81-83 . The first published transgenic mouse line inducing the expression of SV 40 viral large T-antigen to develop PCa in the mouse was the C3 (1) –Tag model, which was achieved with the use of a potentially prostate-restricted promoter, that is of the prostate steroid-binding protein C3(1) gene to drive expression of the SV40 89 . Several other models have been developed using other types of promoters to drive prostate-specific expression of viral oncogene, including the most well studied TRAMP model 90 . The major limitation of this model is that the most frequent malignancy in these mice has manifested the neuroendocrine phenotype, which is relatively rare in human PCa 33, 83 . 13 1.3.3 cPten -/- L mouse model of prostate cancer used in this study One of the major contributions our lab has made in the field of PCa research is the establishment of conditional Pten-deletion mouse model of prostate adenocarcinoma 91 . The tumor suppressor PTEN (phosphatase and tensin homolog deleted on chromosome 10) is one of the most frequently mutated/deleted genes in human prostate cancers which are reported in approximately 30% of primary and 63% of metastatic prostate cancers, respectively 91 . PI3K, leading to is the activation of downstream AKT. PTEN converts PIP3 to PIP2 through its phosphatase activity thereby serving as a negative regulator of the PI3K/AKT signaling pathways associated with unchecked growth and survival of the cells. Due to the essential functions of Pten in the early embryonic development, Pten +/- heterozygous mouse model has been used in the previous study which generally manifests a long latency for development of PIN. Therefore, ARR2PB-Cre (PB-Cre4) was combined with Pten flox/flox mice to generate a Pten homozygous prostate-specific deletion model 91 . With the application of Cre-loxP system, Pten deletion is targeted in the prostate epithelium achieved by the Cre expression driven by the enhanced rat prostate epithelium specific probasin (PB) promoter 92 . This conditional Pten knockout mouse model of prostate cancer serves as a potentially ideal system to study the disease mechanisms because it has achieved the following requirements: 1.) This model exhibits the hyperproliferation and hyperplasia by 4 weeks of age in prostate epithelial cells, leading to the formation of murine PIN 14 (mPIN) by 6 weeks. 2.) Invasive adenocarcinoma is subsequently developed in these mice by 9 weeks of age, which has the epithelial origin indicated by cytokeratin/AR positive expression. 3.) Metastases are detected in the lymph nodes and pulmonary alveolar spaces as early as 12 weeks of age, which represents an animal model that manifested a metastatic phenotype. 4.) Mice are able to benefit from androgen depletion therapy demonstrated by increased apoptosis and decreased tumor volume following castration. 5.) Castrated Pten- deletion mice seem to have acquired the CR growth indicated by the higher proliferation index and Ki67 positive folds than that from age- and genetic background- matched normal mice. The conditional Pten knockout model generally recapitulates the natural initiation and progression of human prostate cancer 91 . The conditional Pten-null mouse model is then improved by combining it with the conditional expression of luciferase reporter, resulting in the induced luciferase expression simultaneously with Pten-deletion in the prostate epithelial cells, designated as cPten -/- L 93 . Thus, we can now monitor the onset and growth of primary ADCa non-invasively over time by bioluminescence imaging. More importantly, this cPten -/- L model system allows us to monitor the tumor regression post-surgical castration and the subsequent recurrence of CRPC which is reflected by the initial decrease followed by the recovered bioluminescence signal, all in living animals. Taken together, this refined cPten -/- L mouse model may serve as an ideal system to trace the initiation and 15 progression of prostate cancer in a non-invasive fashion, which can be further employed as the powerful tools to study PCa with respect to the specific stage of the disease. 1.4. Hypothesis and experimental strategies Despite tremendous advances in understanding the etiology of prostate cancer (PCa) progression, castration-resistant prostate cancer (CRPC) remains to be one of the major challenges owing to the high mortality and no effectively clinical intervention for this aggressive disease 6 . How the vital molecular events work in concert resulting in the emergence of CRPC is yet to be elucidated. It has been well accepted that androgen receptor (AR) signaling axis, necessary for the male sexual differentiation and the growth of ADCa (androgen-dependent prostate cancer), persists to be functional even in the castration-resistance (CR) stage 5, 41-43 . However, significant amount of conflicting data acquired in the various study systems have made the picture of AR activity in CRPC extremely elusive. One of the major hurdles to exploring the mechanisms that how ADCa converts to CRPC is the dearth of good model systems, either in vitro or in vivo, which could accurately recapitulate the progression of human prostate cancer. As described above, the conditional Pten knockout model has been the most commonly used mouse PCa model system that mimics the disease initiation and progression to a considerable extent. Our further refined cPten -/- L model allows us to non-invasively monitor the growth of ADCa, regression post- 16 castration and recurrence of CRPC over time. More importantly, the potential application of this model has been indicated to serve as a powerful tool to study the disease progression in a disease stage-specific manner, which is supported by our experience to make the collections of the primary tumor tissues with respect to the specific phase. To date, the insights into the mechanisms on how ADCa advances to CRPC are mostly gained from the in vitro human cell lines and/or in vivo xenografts 56, 84 . Several essential limitations exist in these systems that may impede the extrapolation the data to the disease progression. First, most immortalized human PCa cell lines are established from late-stage metastatic lesions post ADT that are thus not the suitable systems to study early stage of ADCa. E006AA is one of the few human cell lines derived from a primary tumor 94 . However, it has been shown that E006AA cells express a dominant negative AR due to an S599G mutation in the DBD of AR, which suggests that this cell line does not serve as an ideal system for ADCa since the majority of primary tumors are still dependent on androgens. Second, a variety of LNCaP subline variants are obtained from the long-term culture in androgen deprived media to represent the CRPC phenotype. The problems with these models are that they are established via the “artificial” manipulation without any support from the homologous microenvironment. Third, even if some other LNCaP-derived CRPC variants are generated from the three-dimensional athymic nude mice xenograft models, as discussed above, the communication between tumor cells and immune system is required for the disease progression 17 and metastasis; therefore, they still may not serve as the putatively optimal CRPCs. Last but not the least, based on the extensive experience from previous studies, roles of regulators in the progression of PCa (e.g. AR) are highly dependent on the different cell systems used for the functional analyses, which indicates the necessity to establish the systems encompassing ADCa and CRPC side-by-side. Owing to the fact that there are not any available in vitro models to study ADCa and CRPC in parallel, we set the overall objective in this dissertation to characterize the homozygous systems established from the specific disease stage at either ADCa or CRPC in the cPten -/- L mouse PCa model, which can then be used to study various mediators involved in the progression of ADCa to CRPC, with our focus on the “Partner in Crime” – androgen receptor (AR). Our hypothesis is that, AR signaling axis is highly and dynamically regulated by numerous mechanisms which intricately promotes the occurrence of CRPC. Discovery of androgen receptor splice variants (ARVs) in human prostate cancer led us to propose that by using the homologous mouse systems we may be able to define the specific patterns of expression and function of AR splice variants that may contribute to the overall AR activity in the differential context of CRPC cancer tissues. For this objective, we first aim to isolate and establish the murine PCa cell lines from different stages of the disease, which could be more conveniently handled for future investigation of ADCa and CRPC. We followed the general 18 guideline to establish the stable cell lines and then to characterize them both phenotypically and biologically. The AD/CR features in these cell lines were examined – first, western blot analyses displayed the protein expression of AR; second, AR transcription levels were compared by real-time PCR in response to androgen depletion; third, their growth capacity in the absence of androgen were measured by in vitro growth curve. Real-time PCR was conducted to detect their cytokeratin expression pattern. Their tumorigenicity was also determined by s.c. xenografting followed by histological, immunohistochemical, and immunofluoresent analyses. The individual characteristics of each cell line established from different stages of the disease are described in chapter 2. Now that we have the well-characterized homologous systems composing of both AD and CR cell lines 95 , the next aim is to scrutinize the AR activity in these systems that how it may be regulated and/or contributing to the conversion of ADCa to CRPC. Our preliminary FL), a short AR species was often detected in both AD and CR murine cancer cell lines. These western blot analyses have shown that, in addition to the AR full-length (AR- intriguing results have drawn our interests to search for their compositions and potential roles during the disease progression, which are hinted by the recent identifications of various ARVs in human CRPC models. The ARVs, with the structures slightly different from each other but all lacking the LBD at COOH-terminus, are likely to function as the constitutively functional, ligand-independent transcription factors. Early evidence has shown that expression of two identified ARVs is associated with 19 poor prognosis, which implicates their roles in prostate cancer progression 62, 63, 65, 66 . To determine the role of ARVs in relation to the emergence of CRPC, we have chosen two mouse PCa cell lines, each one representing either ADCa or CRPC, and proceeded to differential expression and activity of ARVs. A special PCR method - 3’ RACE (rapid amplification of cDNA ends) was conducted to detect the putative ARVs that may end up with the novel sequences at the 3’- end. Subcloning and sequence analyses were applied to determine the structural basis of the identified ARVs. The mRNA expression profiles of these ARVs were quantified by the ARV-specific real-time PCR among different cell lines and culturing conditions. The functional analyses were achieved by dual- luciferase reporter assays and co-immunoprecipitation. The results obtained to date on the murine ARVs identified in our mouse prostate cancer model are reported in Chapter 3. 20 CHAPTER TWO: MOUSE PROSTATE CANCER CELL LINES ESTABLISHED FROM PRIMARY AND POST-CASTRATION RECURRENT TUMORS The following chapter has described the establishment and characterization of in total five different murine prostate cancer cell (PCa) lines derived from our conditional Pten-null mouse model of prostate 93 . We have first generated two pairs of prostate cancer cell lines from two tumors: E2 and E4 were from one ADCa tumor; cE1 and cE2 were from one CRPC tumor. This work was originally published in Hormones and Cancer 95 . Later, in order to expand the cell lines at the stage of ADCa, E8 has been established from another ADCa and further characterized using the similar manner. 2.1 Abstract The clinical course of prostate cancer is grouped into two broad phases. The first phase, which is the growth of the androgen-dependent cancer (ADCa) responds well to androgen depletion treatment while the second phase, that could be termed as castration-resistant prostate cancer (CRPC) does not. We first used two separate prostate tumors, one ADCa and one CRPC from the conditional Pten deletion mouse model to generate from each a pair of cell lines. The ADCa cell lines (E2 and E4) and the CRPC cell lines (cE1 and cE2) display bi-allelic deletion at the Pten gene locus, an event which is specific for the prostate epithelium for this mouse model, and a fairly similar level of expression 21 of the androgen receptor (AR). The CRPC cell lines (cE series) grow well in the absence of androgen, display increased AR transcription under androgen- deprived environment, and retain the sensitivity to increased proliferation when androgen is supplemented. The ADCa cell lines (E series) grow slowly in the absence of androgen, and, unlike cE cells, do not show increased AR expression when maintained in the absence of androgen. The detection of epithelial cell markers, such as CK8, CK14, CK18 and E-cadherin in the cE series is conforming with the polygonal epithelial morphology of these cells in culture. The E cells also present mostly polygonal-shaped morphology with a small percent of cells with fibroblastoid morphology, and produce little or very low levels of cytokeratins, but increased levels of vimentin, Twist and Slug, the markers known to be associated with epithelial-mesenchymal transition. Each of the cell lines, when inoculated subcutaneously into male or female NOD.SCID mice induced tumors within eight weeks with 100% incidence. Histopathological examinations of the tumor sections, however, led to noticeable biological differences. The cE series engenders adenocarcinomas, particularly in male hosts, and the E series induces sarcomatoid carcinomas (positively stained for CK8 and AR as well as vimentin expression) in either male or female hosts. A new murine prostate adenocarcinoma cell line, E8, was then isolated from a prostate tumor in the androgen dependent phase of the same mouse PCa model. Similar as E2 and E4 established from intact ADCa tumor, proliferation capacity of E8 is significantly inhibited in the absence of androgen. However, E8 22 displays a typical epithelial rather than an EMT phenotype. These cells present in vitro polygonal-shaped morphology with high levels of CK8 but low to none expression of EMT transcription factors. E8 cells are also able to induce the 100% incidence of in vivo tumorigenesis in NOD.SCID mice, resulting in the formation of adenocarcinomas as cE-series does. These new cell lines are promising models for the elucidation of the androgen metabolism and their role in prostate cancer. 2.2 Introduction Androgen depletion has been a gold standard therapy in the clinical management of prostate cancer with localized or metastasized disease 13, 96 . Because androgen is necessary to maintain the function of the prostate gland as well as for the initial growth of the prostate tumors, this therapy of the androgen- dependent cancer (ADCa) rapidly leads to the shrinkage of the tumors. Unfortunately, within a period of two years, most patients do, however, manifest a relapse resulting in recurrent tumor, which may be best described as androgen depletion-independent cancer (ADI-Ca) 4 , or more recently, castration-resistant prostate cancer (CRPC). How prostate cancer cells survive and proliferate under the low-androgen environment remains unclear, although multiple mechanisms, such as, increased level of expression of androgen receptor (AR) in the cancer cells, mutations or post-translational modifications of AR that broaden its ligand 23 specificity, bypassing of the androgen-AR signaling to an extent through utilization of alternative pathways, etc., have been implicated 46, 97-99 . There is currently a strong interest in the application of the knowledge gained from the analyses of human prostate cancer in modeling this disease in mice. Several mouse models of prostate cancer have been developed based on genetic alternations which are found frequently in human prostate cancer in order to recapitulate its natural course of initiation and progression 33, 100-104 . Among these models, the conditional Pten deletion mouse model (cPten -/- ) demonstrates a short latency for the development of adenocarcinoma 91 . A refined (cPten -/- L) model further increases the utility of the conditional Pten deletion model by combining it with a conditional luciferase expression, which is mediated by the same Cre/LoxP regulation system that is activated in the prostate epithelial cells 93 . In this model, the progression of tumor development and growth can be longitudinally monitored in living animals using bioluminescence imaging (BLI) technology. Importantly, the tumor regression after castration and then relapse can also be similarly monitored in these animals. For studies of the mechanisms of cancer, cancer cell lines are powerful tools. A limited number of human prostate cancer lines have been established from primary or metastatic cancers 105-108 , and there is indeed a dearth of parallel cell lines developed from both ADCa and CRPC phases. Mouse models of prostate cancer can readily provide tissue materials for establishing cell lines from different phases of the disease progression. However, only a few murine 24 prostate cancer cell lines have been derived from the primary tumors to date 109- 111 , and none from the CRPC phase. Here, we report establishment and characterization of four prostate cancer cell lines from the cPten -/- L mouse model, two derived from an ADCa and the other two from a separate CRPC. 2.3 Materials and Methods Animals. Mice with prostate epithelium-specific inactivation of Pten coincidental with luciferase reporter gene activation, as well as castration experiments after tumor growth were as described before 93 . All mice were maintained under identical conditions and animal experimentation was conducted using the standards for humane care in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Generation of cell lines from cPten -/- L prostate cancer model. Freshly collected prostate tumor tissues were minced with crossed scalpels (size 11 blades), transferred to a 50 ml tube and incubated in 5 ml of a digestion medium at 37 ℃ overnight on a rotator (37 ℃, 140 rpm). The digestion medium contained DMEM/F12 (Invitrogen), 10% fetal bovine serum (FBS; Gemini), collagenase (1 mg/ml), hyaluronidase (1 mg/ml) and DNase I (1 g/ml). Collagenase, hyluronidase and DNase I were purchased from Sigma-Aldrich, St. Louis, MO. After low speed centrifugation, single cells and cell clumps were collected and subjected to treatment with 0.05% Trypsin-EDTA for 10 min at 37 ℃ and 25 sequencially passed through 100 m and 40 m cell strainers (BD Bioscience). The preparation was then cultured in a maintenance medium that contained DMEM, 10% FBS, 25 µg/mL bovine pituitary extract (Invitrogen), 5 µg/mL insulin (Sigma-Aldrich) and 6 ng/mL recombinant human epidermal growth factor (rhEGF) (Invitrogen). When cell colonies with epithelial cell morphology were observed, cells were trypsinized, diluted with culture medium to obtain a concentration of approximately one cell/ 100µL, and seeded in 96-well plates (100µL/ well). After culturing overnight at 37 ℃, each well was carefully examined under the bright field in microscopy. Wells containing single cells with the epithelial morphology were marked, and the culture medium in such wells was changed every three days. Expanded cell populations derived from these colonies were subjected to further characterization. Quantitative RT- PCR. Total cellular RNA (1 µg), extracted by RNAqueous- 4PCR Kit (Ambion) was reverse-transcribed by random hexamers using qScript™ cDNA Synthesis Kit (Quanta), and the reverse transcription reaction (1µL) was then subjected to PCR amplification using FastStart Universal SYBR Green Master (Roche). PCR signals were recorded and analyzed in Stratagene MX3000P qPCR system with MxPro software (Stratagene; v4.01). Primer sets are listed in Table 2-1. PCR genotyping. DNA extracted from cell lines were used as templates for PCR reactions using specific primer sets for Pten and Cre. These primer sets are also 26 listed in Table 2-1. Table 2-1. Primer sets used in real-time PCR and genotyping* Gene Forward primer (5’ to 3’) Reverse primer (5’ to 3’) CK5 ACCTTCGAAACACCAACGAC TTGGCACACTGCTTCTTGAC p63 GAAGGCAGATGAAGACAGCA GGAAGTCATCTGGATTCCGT AR AACCAACCAGATTCCTTTGC ATTAGTGAAGGACCGCCAAC CK14 GACTTCCGGACCAAGTTTGA CCTTGAGGCTCTCAATCTGC CK8 ATCGAGATCACCACCTACCG TGAAGCCAGGGCTAGTGAGT CK18 ACTCCGCAAGGTGGTAGATG GCCTCGATTTCTGTCTCCAG PSCA GCTGCTACTCTGACCTGTGC TTCACAATCGGGCTATGGTA CgA GGGAGCTGGAACATAAGCAG TGTCCTCCCATTCTCTGGAC Syph CTTTGTGAAGGTGCTGCAAT GTCTTGTTGGCACAATCCAC E-cadherin TCCAGGAACCTCCGTGATG GGGTAACTCTCTCGGTCCAG N-cadherin AGCCTGGGACGTATGTGATG ATGTTGGGTGAAGGTGTGCT Vimentin CAAGTCCAAGTTTGCTGACCT TCTTCCATCTCACGCATCTG Twist CTCGGACAAGCTGAGCAAG ACGGAGAAGGCGTAGCTGAG Snai1 ACCCACACTGGTGAGAAGC GACCAAGGCTGGAAGGAGTC Slug CACAGTTATTATTTCCCCATATCT GCAGTCTCTCCTCTTCGTCA β-actin † AGTGTGACGTTGACATCCGT CTTGCTGATCCACATCTGCT Pten AAGCAAGCACTCTGCGAAACTG GATTGTCATCTTCACTTAGCCATTGGT Cre GATCCTGGCAATTTCGGCTAT TTGCCTGCATTACCGGTCGAT *Most of the primers were as described 110 except for Twist 112 and those we designed for E-cadherin, N-cadherin, vimentin, Snail, Slug, Pten, and Cre. † For use as a control gene in real-time real-time PCR. 27 Western blot analysis. The whole-cell lysates were prepared as described 92 . The antibodies used in this Western blots were anti-AR (1:100; Santa Cruz) and anti-actin (1:1000; Santa Cruz). Assays for growth in serum-free medium. For testing the efficiency of growth of these cell lines in an androgen-free environment, a modified serum-free medium (SFM) was used 113 . It was composed of DMEM/F-12 supplemented with 0.5% glucose (Sigma), and 2mM glutamine, 10g/ml insulin, 5.5 g/ml transferrin, 6.5 ng/ml selenium, and 6 ng/ml rhEGF (all from Invitrogen). Initially, cells (5x10 4 ) were plated in 6-well plates in the maintenance medium for overnight, washed two times with 1xPBS, and then cultured in SFM for 5 days. The culture medium was changed every 2 days, and the cell proliferation rate was determined at those time points by cell counting. SFM containing 1 or 5 nM of methyltrienolone (R1881) (Perkin Elmer), a synthetic androgen agonist, was utilized to determine the effect of androgen on cell proliferation. Tumorigenicity assays. Cells (1x10 6 ) were mixed with 100 µL Matrigel (BD bioscience) / maintenance medium at a ratio of 1:1 and then inoculated subcutaneously into NOD.SCID mice of 8-12 weeks of age. Grafts were collected surgically at eight weeks post-inoculation from the euthanized animals. Immunostainings. Paraffin sections (5 µm) of tissues fixed in 4% PFA were stained for either H&E or immunohistochemistry (IHC) or immunofluorescence (IF), following the published protocol. Antigen retrieval for IHC was conducted by 28 boiling the slides in 10 mM citrate buffer, pH 6.0 for 15 min, and then cooling down to room temperature. The sections were allowed to react with primary antibodies against androgen receptor (AR; 1:200; Santa Cruz), CK8 (1:100; TROMA-1 antibody; Developmental Studies Hybridoma Bank, University of Iowa), Vimentin (1: 50; Cell Signaling) or Ki67 (1:200; Vector Laboratories) for overnight at 4 o C, followed by incubation with biotinylated secondary antibodies (1:200) against rabbit or rat IgG (Vector Laboratories) for 30 min, and then stained using the ABC Elite Kit (Vector Laboratories) and 3,3’-diaminobenzidine (Dako) as substrate. IF analysis of the parallel sections was done by using primary antibodies against CK8, E-cadherin (1:100; Cell Signaling), N-cadherin (1:200; Abcam), and α-smooth muscle actin (SMA) (Cy3-conjugated; 1:200; Sigma-Aldrich). FITC conjugated secondary antibodies against rabbit or rat IgG (1:80; Sigma-Aldrich) were then applied. After staining, sections were mounted with medium containing DAPI for the labeling the nuclei. Karyotyping. Analysis of karyotypes was performed essentially as described in 114 . Briefly, actively growing cells were cultured in the presence of 0.05 µg / ml colcemid (Invitrogen) for 4 hours, harvested after trypsin dissociation and collected in 15-ml conical centrifuge tubes. Cell pellets were suspended in 1.5 ml of fresh growth medium, 10 ml of a 37°C potassium chloride solution (0.075 M) was slowly added followed by 17 minutes incubation at 37°C. Then, two to three drops of ice-cold fixative (methanol: glacial acetic acid = 3:1) were mixed into the cell suspension. After immediate centrifugation, the fixation step with 10 ml ice- 29 cold fresh fixative was repeated three times. Finally, the cell pellets were suspended in 1 to 0.5 ml fixative and chromosome spreads were made by dropping 70 µl onto glass slides. Slides stained with Giemsa dye were counted using a 100x or a 40x oil immersion objective. At least 25 spreads were counted for each cell line. A statistical analysis was performed with InStat 3.05 (GraphPad). Statistical analysis. The results of cell growth and marker analyses were evaluated as the mean±SE of at least two different experiments performed in triplicate. Statistical calculations were done with Microsoft Excel analysis tools. Differences between individual groups were analyzed by independent t test. P values of <0.05 were considered statistically significant. 2.4 Results Characterization of the new murine prostate tumor cell lines. We first isolated cells from an ADCa tissue from an intact mouse of 8 months of age. After culturing of single cells in 96-well plates, five colonies from different wells were obtained and cultured to enlarge the population, two from which, named E2 and E4, were selected for further characterization. By using the same method, cell lines were derived from a CRPC tumor collected from a 13 month-old male mouse, which was castrated at the age of 8 months. Initial decrease followed by increase in BLI signals were monitored in this mouse for 30 the growth of the recurrent tumor during the 5 months of observation after castration. From this recurrent tumor at the primary site, 3 single cell colonies were produced, two of which were named cE1and cE2, and subjected to further characterization. As illustrated in Fig 2-1A, E2 and E4 cells appeared to display generally a polygonal shaped morphology along with the presence of a small fraction of cells (<5%) that assumed a flattened and elongated shape (indicated by arrows). The cE1, cE2 cells mostly exhibited polygonal shapes but apparently with increased intracellular adhesion as these cells tend to grow in clumps in the culture. To determine the Pten gene status in these four cell lines, a genotyping analysis was performed. Genomic DNA extracted from these cell line cells were subjected to PCR reactions with a primer set specific for the exon 5 region of the Pten floxed allele. This primer set was able to amplify a 1 kb-long DNA fragment in the absence of DNA recombination. However, in the event of Cre-mediated recombination at the floxed site, there should not be any amplified product. Genomic DNA extracted from the tail tissue of cPten -/- L mice and their normal counterparts were used as controls for the PCR products. The sizes were expected to be 1.0 kb and 0.9 kb for the floxed allele and the wild-type allele, respectively. As illustrated in Fig 2-1B, such appropriate bands were detected in the tail DNA collected from the cPten -/- L and normal mice, with no evidence for such products from any of the DNAs isolated from the cell lines. The results implied homozygous deletion of the floxed Pten allele in each of the epithelial cell 31 lines. A similar analysis with the DNA isolated from the cancer-associated fibroblasts (CAFs) from the same mouse model showed the presence of only unrecombined floxed allele, reconfirming the prostate epithelium- specificity of the Cre-loxP system in the model 92 . The results supported an epithelial origin for each of the four cell lines established. The retention of the Cre gene in the cell lines was also confirmed by PCR analysis (Fig 2-1C). When the cell lines were subjected to karyotyping analysis, some individual differences became apparent. As shown in Table 2-2, E2, E4 and cE2 appeared near tetraploid, whereas cE1 was clearly aneuploid. For determining the characteristics of molecular expressions, we used western blot to analyze AR and real-time PCR technique for mRNAs for basal cell markers (CK5, p63, CK14), luminal cell markers (CK18, CK8), neuroendocrine phenotype markers 115-117 , namely chromgranin A (CgA) and synaptophysin (Syph), and the so-called prostate stem cell antigen, PSCA 118 , a marker for intermediate epithelial cell differentiation. The values obtained from each PCR reaction was normalized to that of β-actin. As shown in Fig 2-1D, the level of AR expression was fairly similar in all of these cell lines. Each of the cell lines expressed only very low levels of CK5, p63, CgA and PSCA, and a barely detectable level of Syph (Fig 2-1E). The cE1 and cE2 cells, but not E2, E4 cells, displayed significant levels of CK8, CK14 and CK18 (Fig 2-1F). 32 33 Figure 2-1. Characteristics of the E2, E4 and cE1, cE2 cell lines. A, comparison the morphology of cells under the bright field in microscopy. While all cell lines appear to display polygonal-shaped cells that are distinctive of epithelial cells, E2 or E4 cells also manifest a small amount of cells with elongated morphology, as indicated by arrows. Bar, 100 μm. B. PCR genotyping of the cell lines using Pten-specific primers. (a: normal mouse tail tissue; b: cPten -/- L mouse tail tissue). C. PCR detection of the Cre gene in the cell lines. D. Analysis of AR expression in the mouse cell lines by western blots; human prostate cancer cell lines used as controls were: PC3, negative; LNCaP, positive. E,F. Comparative real-time PCR analysis of marker gene expression including basal (CK5, p63, CK14), luminal (CK18, CK8), neuroendocrine cell markers (CgA, Syph), and PSCA. 34 We next examined these cell lines for the effect of androgen on the proliferation rate. In SFM, E2, E4 cells grew slowly. However, the proliferation rate for cE1, cE2 cells under identical conditions was higher displaying an approximately 5-fold increase in the cell number after 5-day culture (Fig 2-2A). The cell proliferation rate of E2, E4 cells in SFM could be increased by about 1.2 to 1.7 fold by the addition of 1 to 5 nM concentration of androgen (R1881), as illustrated for the E4 cells in Fig 2-2B. The response of cE1, cE2 cells to the same levels of androgen was higher displaying a 3.0- and 3.5- fold increase in the proliferation rate (Fig 2-2C-D). To determine the effect of androgen on the AR expression in these cell lines, we cultured the cells in a medium composed of DMEM and 10% charcoal stripped serum (CSS) up to 18 days. RNA extracted from the cells cultured either in the maintenance medium or CSS medium were examined using real-time PCR with primers specific to AR. As illustrated in Fig. 2- 2E, a significant (2.0- to 2.5-fold; P<0.01) increase of AR transcripts was detected in cE series cell lines cultured in CSS medium relative to cells in maintenance medium. However, In contrast to cE cells, there was no striking increase in the AR expression level between E cells cultured in the maintenance and the CSS medium; rather a slight decrease was observed. 35 Fig 2-2. The comparison of proliferation rates of the cell lines in cultures. A. Growth rates in serum-free-medium (SFM). The differences in the number of cells counted at day 5 were significant between the E and cE series (P< 0.01). B. Illustration of the effect of low concentrations of androgen (R1881) on the growth of E4 cells in SFM (similar results were also obtained with E2 cells). The growth rate was significantly increased (P<0.05) in the presence of 5 nM, but not 1 nM R1881. C. Effect of similar concentrations of androgen on the growth of cE1, and cE2 cells (D). At day 5, the stimulation by androgen was significant (P<0.01) with both cell lines. E. Analyses of the AR expression levels in E and cE series cultured in the maintenance or an androgen-free medium derived by using charcoal-stripped serum in the place of normal serum (CSS). The differences between cE1 and cE1-CSS and between cE2 and cE2-CSS were significant (P<0.01). 36 Tumorigenicity of the cell lines. For testing the tumorigenicity of the cell lines we used grafting on both male and female NOD.SCID recipients to examine the influence of gender on the intake of grafts and tumor histology. Mice were euthanized after 8 weeks later, and grafts were collected, weighed, and tissue sections prepared. Our data demonstrated that the tumor incidence for all cell lines was 100% (Table 2-3), and, in general, grafts formed in male were larger than those formed in female. When the tumors formed in the males from different cell lines were compared, the tumor weights appeared to be similar for the E2, cE1, and cE2 cells, while the grafts from E4 weighed 3-fold higher; in females, E series grafts weighed 3- to 18-fold more than the cE series grafts. When the grafts from the male mice were examined for pathology, a major difference between the cell lines from ADCa and CRPC series was noted. As illustrated in Fig 2-3A, H &E staining of E2, E4 grafts demonstrated spindle cell neoplasms. Since neoplastic cells expressed not only AR, but also vimentin and CK8, these neoplasms were diagnosed as sacrcomatoid carcinoma as per the “Pathological Classification of Prostate Lesions in Genetically Engineered Mice”. Distinct from the E2, E4 grafts (Fig 2-3A, upper part), cE1, cE2 grafts demonstrated the histology of adenocarcinoma with presentation of glandular structures composed of round-shaped neoplastic epithelial cells expressing AR and CK8 (Fig 2-3B), and with such epithelial cells that had penetrated into the stromal layers. Vimentin-positive cells could be detected in the stromal regions, 37 but not inside the epithelial layers. Sections from E2, E4 grafts from the female mice also demonstrated histopathology of sarcomatoid carcinoma, except that the percentage of CK8+ cells was higher than that in the E2, E4 grafts in male mice (Fig 2-3A). While cE1, cE2 cells were also able to form tumors in the female mice, the morphology of glandular structures in these grafts was not as clear as those observed in the male. The IHC staining demonstrated that cE1, cE2 grafts in male had positive nuclear staining for AR. In contrast, their counterparts in the female displayed mostly a diffuse cytoplasmic AR staining. Ki67, the cell proliferation marker, positive cells were detected in all grafts (Fig 2-3C). 38 Table 2-2. Representative karyotypes of the cell lines Table 2-3. The comparison of the incidence and weights of the tumors formed Cell Line Chromosome Number Sample Number E2 Near 4N (79.3±14.9) 40 E4 Near 4N (76.1±23) 31 cE1 Near 3N (63.8±15.1) 26 cE2 Near 4N (78.6±29.5) 25 Host Cell Line Incidence (%) Tumor Weight (g) Male E2 4/4 (100) 0.25 ±0.24 Male E4 4/4 (100) 0.80±0.54 Male cE1 4/4 (100) 0.25±0.08 Male cE2 4/4 (100) 0.26±0.07 Female E2 4/4 (100) 0.17±0.03 Female E4 8/8 (100) 0.56±0.15 Female cE1 8/8 (100) 0.05±0.01 Female cE2 4/4 (100) 0.03±0.02 39 40 Figure 2-3. Microanatomic analyses of tumors induced by the cell lines in male or female in NOD.SCID mice. A. H&E and IHC staining of sections of E2 and E4 grafts for AR, CK8 and vimentin. A higher magnification of an area marked by the arrow is shown in upper-left corner of each section. Bar, 100 μm. B. Similar staining experiments with sections from cE1 and cE2 grafts. Bar, 100 μm. C. IHC staining of the grafts for Ki67. Bar, 100 μm. 41 Analysis for EMT markers. Suspecting that E2, E4 cells might have undergone some degree of EMT conversion in cultures, we further tested for the expression levels of certain EMT markers in all cell lines by using real-time PCR. As illustrated in Fig 2-4A, while E-cadherin expression was present in cE1 and cE2 cells, it was practically absent in E2 and E4 cells; N-cadherin levels were very low in all cell lines. E2, E4 cells appeared to produce significantly more vimentin than the cE1, cE2 cells. Expression of Twist, Slug and Snail, the known transcriptional repressors for EMT 117-120 , was also increased in the E2, E4 cells relative to the cE1, cE2 cells (Fig 2-4B). A few markers were also examined at the protein level using the grafts formed in vivo. Tissue sections adjacent to those used for H&E and IHC staining, were analyzed by IF using antibodies against CK8, E-cadherin, N- cadherin and α-SMA. As illustrated in Fig 2-4C-E, cell stained for α-SMA did not appear to simultaneously express CK8, E-cadherin, or N-cadherin on sections from cE1 and cE2 grafts grown either in male or female mice. However, many such cells (arrows) were detected in the sections from E2 and E4 grafts. 42 Figure 2-4. Analysis for the expression of EMT-related markers in the cell lines. A. Relative expression of E-cadherin, N-cadherin and vimentin, and B, that of Twist, Snail and Slug as determined by real-time PCR. The difference in the level of expression of each of vimentin, Twist, Snail or Slug in the E series relative to the cE series was significant (P<0.01). C-E. Illustrations of co- immuofluorescence results when tumor grafts from the E and cE cell lines were stained red for α-SMA and green for either CK8(C), E-cadherin (D) or N-cadherin (E). White arrows indicate cells with co-expression (yellow). Bar, 50 μm 43 44 Establishment and characterization of the third ADCa cell line A new line of androgen dependent prostate cancer cells was established from the conditional Pten null mouse model. This cell line, dubbed E8, was isolated in the same manner as E2 and E4. E8 cells appeared polygonal with somewhat spindle shaped morphology and grew very rapidly; unlike cE1 cells they did not grow in clumps (Fig 2-5A). Using real-time PCR, we checked for the expression of epithelial markers, CK8, p63, and E-cadherin (Fig 2-5B). We saw that there was high expression of CK8 and no detectable expression of p63, likening the cells to luminal epithelial cells. E-cadherin expression was only moderate, consistent with the lack of cell to cell adhesion. Since other androgen dependent cell lines isolated from this mouse model have a tendency to undergo an epithelial-mesenchymal like transition (EMT), we also checked E8 cells for expression of fibroblast and EMT markers. We found that there was high expression of vimentin and moderate expression of N-cadherin, indicating some de-differentiation occurred, but little to no expression of EMT transcription factors, such as, Twist, Snail, and Slug (Fig 2-5C). E8 cells were isolated from an intact mouse and should be androgen dependent, so to determine how proliferation would be affected by the presence of androgens, growth assays were done in a modified serum-free media containing 0, 1, or 5 nM R1881. Fig 2- 5D shows that in the absence of androgen, E8 proliferation was inhibited. In fact, practically no proliferation was observed after three days in culture. The addition 45 of R1881, however, helped sustain cell proliferation and increased the calculated doubling time by two-fold at the concentration used (1 or 5 nM R1881). To determine if E8 cells possessed the ability to induce tumors in vivo, 1×10 6 E8 cells were subcutaneously injected into male NOD.SCID mice. Tumor incidence was observed to be 100% (6/6 mice). Histological analysis showed that the grafts indeed resembled adenocarcinoma containing glandular structures composed of multiple layers of epithelial cells expressing AR and CK8 (Fig 2-5E) and breaching into the surrounding stroma. Positivity for CK5, a basal epithelial cell maker, could be seen in cells lining the lumen. Grafts also stained positive for Ki67. 0.00E+00 5.00E-02 1.00E-01 1.50E-01 2.00E-01 2.50E-01 CK8 p63 E-Cadherin AR Mean Ratio to Actin A B 46 0.00E+00 2.00E-01 4.00E-01 6.00E-01 8.00E-01 1.00E+00 1.20E+00 1.40E+00 1.60E+00 1.80E+00 Vimentin N-Cadherin Tw ist Snail Slug Mean Ratio to Actin C D 47 Figure 2-5. Characteristics of a new murine prostatic epithelial cell line, E8. A: 100X brightfield micrograph of E8 cells in their normal growth media showing polygonal shaped morphology and lack of cell attachment. Bar, 100 μm. B: epithelial (CK8, p63, E-cadherin) and C, fibroblast (Vimentin, N-Cadherin) and EMT (Twist, Snail, Slug) expression profile for E8 cell lines as observed by real- time PCR analysis. Shown are mean +/- standard deviation. D: Comparison of E8 growth rate when cultured in SFM containing 0, 1, or 5 nM R1881. Cultures containing androgen had significantly higher cell numbers than those that were androgen free. E: 400X micrographs of H&E and IHC staining of sections of E8 grafts for AR, CK8, CK5, and Ki67 showing an adenocarcinoma-like morphology. Bar, 50 μm. **p<0.01. H&E CK8 CK5 AR Ki67 E 48 Discussion The importance of this report is that it describes isolation and properties of new tumorigenic cell lines derived from two different tumors developed in the cPten -/- L mouse model. These cell lines add to the limited repertoire of cell lines currently available for the study of prostate cancer. Of the four cell lines we describe here, E2, E4 were derived from an ADCa, and cE1, cE2 from CRPC of a separate mouse. While each of the cell lines is strong in AR production, the two series vary in their growth rates, levels of expression of certain cell markers, and sensitivity to androgen exposure. In general, E2, E4 cells, expressing low levels of basal cell markers (CK5 and p63) but practically no luminal cell markers, show a requirement for androgen for optimal growth in cell cultures. E2 and E4 cells appear to express a few EMT markers, namely vimentin, Twist, Slug and Snail, at levels significantly higher than the cE series. The cE1, cE2 cells, which express significantly higher levels of two luminal cell markers (CK8 and CK18) as well as a basal cell marker (CK14) than the E2 and E4 cells, can grow well in the absence of androgen, and at the same time retaining the capacity to produce increased AR transcripts when deprived of androgen, and to increase proliferation when androgen is supplemented. Since AR expression is similar in both series, the difference in the androgen sensitivity is likely to be regulated by mechanisms much more complex than simply by the levels of AR. It has been reported that AR remains critical for cell-cycle progression of human prostate cancer cell lines that are not dependent on androgen for growth 39, 121 , and the 49 loss of PTEN function in a set of mouse neoplastic epithelial cell lines enhances androgen independence 110 . In the Pten-deleted cell lines we have generated, a clear picture of the effect of androgen on cell growth is also not derived. We find that the two cell lines obtained from the androgen-dependent phase of the mouse prostate tumor can slowly proliferate for a limited time under androgen deprivation, and the cell lines from the CR phase of the tumor show the ability to grow significantly more efficiently under such an environment while retaining the potential of increased proliferation when exposed to androgen. Since the study was performed using the E and cE lines from two different mice, it is, however, unclear whether the cE cells represent an evolution from an androgen-dependent state correlating with a linear progression of the disease, or an entirely independent phenomenon. To address this issue, it would be necessary to analyze samples of the same tumor and derived cell lines from a single animal at successive stages of progression from ADCa to CRPC. Another interesting observation we made is that these two series of cell lines drastically differ in their ability to induce tumor types in the in vivo assays. E series cell lines formed sarcomatoid carcinoma in both male and female mice, whereas cE series cell lines formed adenocarcinoma in males but undifferentiated tumors in the female (Fig 2-3B). Consistent with the detection of reduced cell-cell adhesion properties as well the presence of a small fraction of cells with elongated and flattened shapes in the in vitro cultures of the E series, the real-time PCR results from these cells also indicate an increased levels of 50 some EMT markers in these cell lines, as stated above. The possibility that the E series cells might have acquired a degree of EMT is supported by the demonstration that spindle-shaped cells co-expressing both epithelial and mesenchymal markers could be readily detected in the tumors formed in vivo from these cells. No such evidence was obtained with the cE cell lines (Fig 2-4 C-E). In the tumors induced by the cE series, however, a major influence of the gender of the host animal was clearly evident. Tumors induced in the males by the cE cells present distinct glandular structures with the presence of CK8 positive cells both inside and outside the lumen, whereas, there is a dense distribution of CK8 positive cells in the undifferentiated tumors formed in the females, especially in cE2 grafts (Fig 2-3B). These results suggest that the cE cells retain the capacity to generate lesions akin to prostatic adenocarcinoma, and this property appears to be dependent on the levels of male hormones. In this connection, the AR staining in the tumor sections of both E and cE cell lines that are formed in the male mice is found to be located mostly in the nuclei, but are detected primarily in the cytoplasm of the tumor cells in the grafts formed in the female mice, where the circulating androgen levels are expected to be low. Thus, a regulation by androgen appears to be more prominent in the formation of adenocarcinomas from the cE1 and cE2 cells, and less for the histopathological phenotype of the tumors induced by E2 and E4 cells. Induction of sarcomatoid carcinoma by the E series, however, remains quite intriguing. First of all, sarcomatoid carcinomas of the prostate are rare both in humans and mice. 51 Second, contrary to our expectation, these cell lines from the relatively early phase of the prostate tumorigenesis appear to be less differentiated than those from the late CRPC phase. It is, thus, unclear whether the isolation of E2 and E4 cells is related to clonal variations that may not be representative of the majority of the cancer cells but are coincidentally selected in the cultures or their presence in the tumors may have a hitherto unknown biological implication in tumorigenesis induced by Pten gene activation in the prostate epithelium. Isolation and characterization of additional neoplastic cell clones from the ADCa phase will be necessary for further evaluation of these issues. In summary, we have generated two pairs of new mouse prostate cancer cell lines with homozygous deletion of the Pten tumor suppressor gene from two different prostate tumors: one pair of cell lines from an androgen-dependent tumor and the other pair from an androgen depletion-independent (recurrent) tumor. The particular cell lines, derived from spontaneous tumor models of a similar genetic background, display properties that are distinctive. Each of two cell lines derived from the recurrent tumor grows well in the absence of androgen, displays the ability to increase AR transcription when deprived of androgen, maintains sensitivity to androgen levels for increased proliferation, and can produce adenocarcinomas in male NOD.SCID mice. The pair of cell lines obtained from the androgen-dependent phase requires androgen for optimal growth, exhibits certain EMT-like properties in vitro, and can induce sarcomatoid carcinomas displaying a hybrid epithelial-mesenchymal cell phenotype in vivo 52 when transplanted in either male or female hosts. These cell lines would likely to be important resources to explore the mechanisms in prostate cancer progression and manifestation of the diverse histopathological phenotypes. Recognizing that E2 and E4 display a significant EMT-like morphology, we proceeded to generate another mouse prostate cancer cell line from an independent ADCa. The resulting cell line, E8 although having some mesenchymal-like characteristics perhaps due to its transformed state, mostly expresses epithelial markers and little epithelial-mesenchymal transition markers, such as Twist, Snail, and Slug, and thus is not actively undergoing EMT. This phenotype is stable in culture and when subcutaneously injected, forms adenocarcinoma in vivo. E8 is also androgen dependent as it has limited proliferative capacity without androgen. 53 CHAPTER THREE: NOVEL ANDROGEN RECEPTOR SPLICE VARIANTS IN HOMOLOGOUS SYSTEMS OF MOUSE PROSTATE CANCER 3.1 Abstract Androgen receptor (AR) activity that is necessary for the normal prostate development remains active throughout the progression of prostate cancer. Even at castrate levels of androgen AR remains functional, possibly through several non-mutually exclusive mechanisms including expression of constitutively active AR splice variants (ARVs). In relation to the various ARVs identified to occur in human prostate cancer cell lines or xenografts, not much is known on the presence and role of ARVs in disease progression in any defined mouse models of prostate cancer. In this study, by using the murine prostate cancer cell lines established from the conditional Pten-null mouse model of prostate adenocarcinoma, we report three distinct types of mouse AR-Vs which we named as mARV-a,-b, and -c. Structurally, mAR-Va only lacks ligand-binding domain (LBD) retaining exon 1-4 followed by the inclusion of a novel exon 4 (m4a). ARV isoforms (mAR-Vb and -Vc) are novel, merely containing the N- terminal domain (NTD) encoded by exon 1 before the addition of two different novel exons at their 3’-terimus, termed as m1b and m1c, respectively. Although these mAR-Vs occur in all murine cell lines established from either androgen- dependent (AD) or castration-resistant (CR) tumors, their proportion appears to vary with respect to the stage of the disease from which the cell lines originated. 54 In cE-series (CR cell lines), AR-Vb and AR-Vc are more abundant than that in E- series (AD cell lines). AR-Va, however, displays an opposite profile of expression. We found that mRNA expression levels of these AR-Vs are increased in response to androgen depletion in the mouse prostate cancer cell lines. We also detected the mRNA expression of these ARVs in the normal prostates; however, their relative abundance to the prototype AR is significantly increased in tumor phenotypes. Unexpectedly, mAR-Vabc variants are found to significantly inhibit the transcriptional activity of AR full-length in the presence of androgen, indicating a complex regulation of AR activity in the prostate epithelial cells. It appears that AR variants, particularly AR-Vb and AR-Vc identified here, may have an opposite role than what was anticipated; they may serve as brakes in high AR activation, thereby potentially benefitting the persistence or survival of cancer cells at the castrate levels of androgen. If this observation could be validated in other mouse models of recurrent prostate cancer and in human prostate cancer, it may be argued that the force of evolved ligand-independent activation of AR by various previously implicated means, such as, interaction with co-regulators or cofactors, gene mutations and amplifications, post-translational modifications, etc., needs to be restrained and, here AR variants come to play a crucial role. This potentially unique suppressive role of certain AR molecular structures, however, remains to be better defined at the biochemical level and in the progression of prostate cancer. 55 3.2 Introduction Prostate cancer is sensitive to androgen deprivation at the primary androgen-dependent growth stage (ADCa). However, it invariably acquires resistance to androgen withdrawal leading to the recurrence of castration- resistant prostate cancer (CRPC) which remains fatal and incurable 2, 6, 9 . Intensive efforts are being placed to understand the mechanism underlying the recurrence of CRPC, although the task is difficult owing to the phenotypic heterogeneity of tumor cells 33 . The androgen receptor (AR) plays a critical role in the development and progression of prostate cancer, in both ADCa and CRPC 5, 41, 43 . Multiple mechanisms have been implicated how cancer cells acquire the castration resistance under androgen-depleted condition, such as the de novo steroidogenesis, AR amplification, gene rearrangements or mutations, altered expression of co-regulators, etc 5, 6, 8, 37, 43, 122 . More recently, several studies with human prostate cancer cell lines and tissues identified the presence of AR splice variants (ARVs) which are up-regulated and seem to promote the survival and growth of tumor cells in the stage of CRPC 19, 50, 51 . The androgen receptor (AR) belongs to the steroid receptor transcription factor family composed of an N- terminal transcriptional activation domain (NTD, exon1), DNA-binding domain (DBD, exon 2 and 3), a hinge region (exon 4), and the C-terminal ligand binding domain (LBD, exon 4-8) 18, 20 . Mouse androgen receptor (mAR) is 97% and 85% homologous to that of rat and human at protein levels, respectively, consisting of the fully conserved DBD and LBD 23 . To date up to 15 different ARVs lacking 56 different portions of LBD have been reported in human prostate cancer cell lines CWR-R1, 22rv1, VCaP and LuCaP xenografts 57-64 . On the other hand, AR isoforms remain understudied in mouse prostate cancer tissues or cell lines; detection of only two ARVs was reported in Myc-CaP cells 58 . Therefore, we proceeded to obtain a better definition of ARVs in relation to distinct phases of this disease using an appropriate mouse model of prostate cancer. Mouse models, recapitulating the natural course of disease progression, have been applied extensively in the studies of human prostate cancer. Our laboratory had contributed to the development of the conditional Pten null mouse model which recapitulates many features of the initiation and progression of the human prostate cancer 91 . We then further refined this model by incorporating the conditional expression of luciferase reporter gene into the system, which allows longitudinal monitoring of initiation, growth, regression after castration and recurrence of tumors in a non-invasively manner by bioluminescence imaging (BLI) 93 . From this cPten −/ − L mouse model, we have established so far five murine prostate cancer cell lines, three E-series (E2, E4 and E8) generated from two primary ADCa and two cE-series (cE1 and cE2) from one CRPC tumor 95 . The CRPC cell lines (cE series) display a better growth capacity in the absence of androgen than that of the ADCa cell lines (E series), although all of them still retain the sensitivity to increased proliferation when androgen is supplemented. These cell lines can be used as the potential models to elucidate 57 mechanisms and roles of AR signaling in the epithelium at different stages of prostate cancer. In this study, we have identified three AR splice variants (named as mAR- Vabc) in the mouse PCa cell lines established from our conditional Pten-null mouse model of prostate cancer. Our data show that these mARVs display the distinct expression profile in respect to the specific stage of prostate cancer from which the cell lines were established; and they are up-regulated at mRNA levels when androgen is depleted. Of note, these mAR-Vs seem to be present in normal prostates as well, with an increased relative expression in cancerous prostates when compared to AR-FL. Intriguingly, expression patterns of AR- Vs/AR-FL seems to alter with ageing. More interestingly, despite their diverse molecular organizations, mAR-Vabc are all act to inhibit the transcription activity of AR-FL in the presence of androgen in the context of certain cell systems tested; no similar ARVs with such functions have been described before. This work provides a new platform to seek insights into the role of ARVs in disease progression, and then, potentially, to explore the clues to therapeutic approaches. 3.3 Materials and Methods Cell culture and mouse prostate samples. E4, E8, cE1 and cE2 were established from our conditional Pten-null mouse model of prostate adenocarcinoma as described previously. LNCaP and PC-3 cells were 58 purchased from ATCC. C4-2B and COS-1 cells were kindly provided by Dr. Jeremy Jones (City of Hope, LA, CA) and Dr. Allen Epstein (University of Southern California, LA, CA), respectively. The derivatives of parental cell lines were generated by culturing them in DMEM with 10%CSS (Charco-stripped Serum, Cellgro) up to 18 days. E8 and cE2 cells were also treated in normal maintaining media with the addition of 10 µM of anti-androgen, bicalutamide (Sigma-aldrich) for 6 days. Fresh media containing either CSS or bicalutamide were changed every other day. Mice with prostate epithelium-specific Pten knockout in combination with luciferase expression that developed spontaneous prostate cancer have been described 93 . Prostatic tissues were collected from mice bearing ADCa or CRPC tumors as well as wild-type ones at different age groups. Dissected prostates were processed by rapid freezing in liquid nitrogen and stored in -80 °C for further analysis. Cloning and constructs. Total cellular RNAs were isolated from E8 and cE1 by RNAqueous-4PCR Kit (Life technologies) used as the source. The detection of mouse AR splice variants from them was achieved by GeneRacer® Kit with SuperScript® III RT and TOPO TA Cloning® Kit for Sequencing (life as per manufacture’s protocols. Briefly, extracted RNAs were first reverse transcribed using the GeneRacer® oligo (dT) primer and SuperScript® III RT module from the kit. RACE-ready cDNAs were subjected to the 3’ rapid amplification of cDNA end PCR (3’ RACE-PCR) using a forward gene-specific primer (GSP) and reverse GeneRacer® 3’ primer from the GeneRacer ®Module. 59 Another round of nested PCR was conducted for further amplification using a forward GSP nested primer and reverse GeneRacer® 3’ Nested primer provided in the kit. SuperTaq™ Plus Polymerase (Life technologies) was used for both 3’RACE and nested PCR. Forward GSP and nested primers anchored within exon1 were listed in Table 3-1. 3’RACE and nested PCR products were then examined by standard TA subcloning and sequencing using the TOPO TA Cloning Module. Plasmid constructs used in this study included: Rat AR full- length (Dr. Robert Matusik, Vanderbilt University, Nashville, TN), human AR full- length (Dr. Jeremy Jones, City of Hope, Duarte, CA), and PSA-luciferase reporter and PCDNA3.1-myc (Dr. Parkash Gill, University of Southern California, LA, pCR®4-TOPO® was purchased in the TOPO TA Cloning® Kit for Sequencing (life technologies). CDS (coding sequences) for mARVa, Vb and Vc were amplified by PCR from cDNAs of E8 and cE1 using the primer sets enclosing the entire open reading frame (ORF), and subcloned into pcDNA3.1-myc-HisC. To generate the ARVa-myc and ARVc-myc constructs, reverse primers were re-designed to be anchored upstream of their own stop codons and in-framed with the ORF of expression vector. Hifi AccuStart™ Taq DNA Polymeras (Quanta Biosciences) and HotStar Taq DNA Polymerase (Qiagen) were applied in PCR cloning. Sequences of primer used to clone Va, Vb,Vc, Va-myc and Vc-myc were provided in Table 3-2 & Table 3-3. 60 ApE-A plasmid Editor was used in this study to analyze sequencing results, predict protein sequences and generate text maps for mARVs. Table 3-1. 3’ RACE and nested PCR primers Sequence Forward GSP primer TCTGTCTCTGTATAAATCTGGAGCA Forward GSP Nested primer ACCACCTCTTCTTCCTGGCATACTC Table 3-2. PCR primers for cloning of mAR-Vabc Table 3-3. PCR primers for cloning of mAR-Va-myc and mAR-Vc-myc Forward Reverse ARVa ACGTGAATTCGAAGCTACAGA CAAGCTCAA ACGTCTCGAGCTGAGTCTCAACTCACTCAA ARVb ACGTCTCGAGTTGGTCAAAAGGAGGCATTT ARVc ACGTCTCGAGATTTCCTCTATCATCAGGCA Forward Reverse mAR-Va-myc ACGTGAATTCGAAGCTA CAGACAAGCTCAA ACGTCTCGAGCTACGAAAATCACTTCTCCA mAR-Vc-myc ACGTCTCGAGCTTTCCTCTTGTAGTGCTTG AAATTTCGCATGTCCCCATAAGGT 61 PCR and Quantitative real-time PCR. Total cellular RNAs were extracted from all parental and other derivative cell lines and then reverse transcribed using oligo (dT) primer and SuperScript® III First-Strand Synthesis (Life technologies). RNA samples used for various assays were isolated from cells with different batches. Conventional PCR was performed on cDNAs using the same forward primer anchored within exon1 paired with reverse primer specific for mAR-Vabc and AR-FL, respectively (Table 3-4). Quantitative real-time PCR and data analysis have been described previously. The real-time primer sets designed specifically and exclusively to amplify each ARV and AR-FL were listed in Table 3-5. 62 Table 3-4. Primer sets for conventional PCR Table 3-5. ARV specific primer sets for real-time PCR Forward Reverse ARVa AGCCACCACCTCTTCTTCCT GCATCCCACATCCTCATTCT ARVb AAGATGACAGTCCCCACGAG ARVc AATGGAGAGTGACGCAAAGG AR-FL TACTGAATGACCGCCATCTG Forward Reverse ARVa TCCTTTGCTGCCTTGTTATC GCATCCCACATCCTCATTCT ARVb GGGACCTTGGATGGAGAACT GGCGGAACATTTCACAAGAT ARVc AGTGAAATGGGACCTTGGAT AAATGGAGAGTGACGCAAAG AR-FL TCCAGGATGCTCTACTTTGC TTTTGATTTTTCAGCCCATC 63 Transfection and luciferase reporter assay. 1-2 days prior to transfection, cells were placed in media containing charcoal-stripped serum (CSS). For all transfections, pools of cells were transfected using Lipofectamine Plus (Invitrogen) with empty vector control plasmid, full-length AR plasmid, or AR splice variant plasmids along with the androgen-responsive constructs MMTV- firefly luciferase or PSA-firefly luciferase 123 and the androgen-insensitive renilla luciferase control pRL-SV40 (Promega). Empty vector control plasmid was used to equilibrate total DNA in all transfections. The following day, the cells were plated in quadruplicate with drugs in 96 well plates. 24hrs later luciferase activities were quantified (Dual luciferase assay kit, Promega) on a plate reader (Tecan) and the firefly signal normalized to the renilla signal to control for cell number. Co-Immunoprecipitation assay.COS cells were transfected with mARVa-myc or mARVc-myc together with Rat AR-FL using Lipofectamine 2000 (Life technologies) as per the manufacture’s protocol. About 48 hrs post-transfection cells were lysed with 1% CHAPS lysis buffer (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.5 mM CaCl2, 0.5 mM MgCl2 supplemented with protease inhibitor cocktail (Pierce). Then 500 ul cleared cell lysate were immunoprecipitated by incubation with 5 ug anti-myc mAb (Roche) and Protein G PLUS-Agarose (EMD Millipore) overnight at 4°C. The isotype matched control mAbs were included in all immunoprecipitation experiments as negative control. Immunoprecipitates 64 were washed twice in 0.1% lysis buffer followed by elution separation on 4-20% Criterion Tris-HCl Gel (Bio-Rad) under reduced conditions. Western blot analysis.The whole-cell lysates from cell lines and primary tissues were prepared as described previously 124 . Protein samples were loaded on 10% polyacrylamide gels (Thermo Scientific) and subjected to immunoblot analysis with anti-AR (N20; Santa Cruz). After detection of signals membranes were stripped and re-probed against actin (Santa Cruz) to confirm equivalent loading and transfer of protein. The elutions from the co-immunoprecipitation were resolved by SDS- PAGE and analyzed by western blot. The antibodies used in this study included anti-AR (C19, Santa Cruz), anti-Myc and anti-Actin (Santa Cruz). The molecular weight and expression levels of protein samples were assessed by Gel Doc™ XR+ system (Biorad). 3.4 Results Short AR species are often detected in both PCa cell lines and prostatic tissues isolated from conditional Pten-null mouse model of prostate cancer. It is reported that smaller size of bands in the range of 75-80 kDa are detected in CWR-R1, 22RV1 and LNCaP human prostate cancer cell lines when antibody specific for the NH 2 -teriminus of AR is used 50, 125 . To examine the AR protein status in our conditional PTEN deletion mouse model of prostate 65 adenocarcinoma, our mouse prostate cancer (PCa) cell lines E8, cE1 and their charco-stripped treated derivatives (E8-CSS and cE1-CSS) were loaded together with human PCa cell lines LNCaP and PC3. Western blot analysis showed that similar to LNCaP, in addition to the AR full-length (AR-FL, ~110kDa), three major bands were detected in all the parental and CSS-treated mouse prostate cancer cell lines (Fig 3-1A). Among them, one with an estimated molecular weight of ~85 kDa was larger than that of ~75kDa in LNCaP which seemed to be a truncated AR as previously described 126, 127 . As for the lower categories composed of ~56 kDa and ~50 kDa, they were present in both human and mouse PCa cell lines that we tested. Previous studies have shown that the expression levels of several human ARVs were increased in CRPC bone metastases and associated with significant poor prognosis; therefore, the protein expression of AR was also examined in the prostatic tissues isolated from the age-matched wild-type, androgen-dependent prostate cancer (ADCa) and castration-resistant prostate cancer (CRPC) mice. Short AR species, especially those ~50kDa bands, were universally present in prostatic tissues, even in normal mice. Interestingly, in contrast to the expression pattern in cell lines, AR-FL was detected in the lysates isolated from prostates with a variable level. Weak to non-detectable AR-FL bands were found in 2 out of 3 wild-type, 1 out of 3 ADCa and 2 out of 3 CRPC prostates, respectively. These findings prompted us to begin to examine the structural and functional implications of the various ARVs in prostate cancer progression in our mouse model. 66 Figure 3-1. Androgen receptor (AR) protein expression status in prostate cancer cell lines and prostatic tissues A. The whole cell lysates harvested from LNCaP, PC3, E8, CE2 and and their derivatives treated with charcoal- stripped serum (CSS) were probed with anti-AR (top) and anti- βactin (bottom). Molecular weight of the short AR species was estimated by GelDocTM XR+ software, as 85 kDa, 56 kDa and 50 kDa. 30 µg of protein was loaded into each lane. B. AR expression in the cell lysates extracted from wild-type mice, androgen-dependent prostate cancer (ADCa) and castration-resistant prostate cancer (CRPC) developed from the conditional Pten-deletion mouse PCa model. Tumors were dissected from mice bearing AD and CR tumors at the age of 13- month-old. Prostates of age-matched wild-type mice in 129/BALB/c background were dissected and lysed into cells. AR expression was detected by western blot using antibody recognizing epitope at the NH2-terminus. 67 Characterization of structurally novel ARVs in murine PCa cell lines established from either ADCa or CRPC tumors. We performed 3’RACE on the cDNAs generated from E8 and cE1 with the forward primers anchored within the exon1 of mouse AR to fish out all kinds of AR mRNA with a Poly(A) tail, particularly, if there is any ARV with novel 3’- sequences. Through the initial 3’RACE PCR followed by a nested PCR, we found several products (Fig 3-2A.). Based on the rationale of 3’RACE, all AR mRNA kinds should be amplified using the forward gene-specific primer (GSP) anchored within exon1 paired with 3’RACE reverse primer composing of the Poly (A) stretch followed by the adaptor sequences. As we expected, the longest ~2 kb band, referred here as AR-FL, contained sequences upstream of the Poly(A) region that matched with the 3’UTR of mouse AR (mAR) mRNA. Cloning and sequence analyses of the ~ 850 bp PCR products indicated the existence of different mouse ARVs (mARVs) in E8 and cE1 cells. One short AR isoform was found in E8, retaining exon 1-4 from mAR mRNA, followed by a novel 175-bp sequence that was aligned to the AR intron region adjacent to exon 4, which we termed as mouse exon 4a (m4a) (Fig 3-2B). The molecular structure of this ARV was analogous to the previously described mouse mAR-V4 58 and human ARV 567es 59 on their 5’ – retaining exon 1-4 but differed in the compositions of 3’- sequences from m4a. Unexpectedly, two novel ARVs with potentially unique alternative splicing sites have been detected in cE1 cells. When mapping their sequences to the mouse genome, we found that one 495-bp sequence and 68 another 442-bp sequence from different positions at intron 1 were spliced to exon 1, which we referred to m1b and m1c. These two novel ones appeared to represent a type of ARVs that were not encountered before (Fig 3-2C& 2D). To avoid potential conflict with the nomenclature of those ARVs reported in human prostate cancer models, we named these three ARVs as mouse AR-Va,b,c (mAR-Vabc), of which mAR-Va was first identified in E8, while mAR-Vb and mAR-Vc in cE1 (Table 3-6). The molecular weight of the proteins encoded by mAR-Vabc seemed to be corresponding to that of the lower bands detected by the western blot. Although the length of the mRNAs is similar, the deduced protein products of mAR-Vabc are expected to be different. mAR-Va spliced after exon 4 followed by m1a would generate a short AR transcript consisting of NTD, DBD, the hinge and 42aa translated by m1a-derived sequences (Fig 3-2D). On the other hand, mAR-Vb and mAR-Vc containing only exon 1 before the addition of m1b and m1c would encode a smaller ARV composing of NTD only. The splicing of m1b to exon1 in mAR-Vb results in a stop codon right at the end of exon1; whereas eight extra amino acids can be extended to the NTD in mAR-Vc protein (Fig 3- 2D). To what extent the additional 42 aa and/or 8 aa translated by m4a and m1c might compensate for the loss of LBD and/or other functional domains remained to be evaluated. The schematic structures of mAR-Vabc transcripts and their encoding protein products are summarized in Fig 3-2D. 69 70 71 Figure 3-2. Identification of new mouse AR splice variants (mARVs). A. 3’RACE and a nested PCR were performed with the templates that were reversed transcribed from total RNA extracted from E8 or cE1. B.C.D. Text maps were generated by ApE- A plasmid Editor demonstrating the RACE-PCR sequencings of mARVabc. Pink highlights the primers used in PCR. Blue refers to sequences mapped with mAR mRNA. Green indicates the novel exons in mAR-Vs. Deduced stop codon is labeled with red star. E. Schematic structure of mAR-Vabc and AR full-length (AR-FL). Solid-colored boxes represent the common exons of AR-FL; hatched cassettes indicate the cryptic exons. Bolded straight lines refer to the transcribed sequences in their mRNAs. The predicated amino acid sequence translated by m4a and m1c is listed. 72 Table 3-6. Summary of transcription and translation products of mAR-Vs Distribution of mAR-Vs in the cell lines appears to vary with respect to the stage of the disease To examine the expression of mAR-Vabc identified by 3’RACE, we designed a forward primer anchored within exon 1 and individual reverse primer complementary to the specific sequences from m4a, m1b and m1c; a reverse primer anchored within exon 5 was used to detect AR-FL when paired with the same forward primer (Table 3-4). Although first identified separately from E8 or cE1, mAR-Vs could be amplified by conventional PCR from all the cell lines tested (Fig 3-3A). Relative abundance of AR-FL versus mAR-Vs was indicated by the ten times more input of templates of mAR-Vs as that of AR-FL in PCR. Real-time PCR was then performed to further quantitate the expression levels of these ARVs in our mouse prostate cancer cell lines. For measurement of AR-FL mRNA, we used the forward primer anchored within exon 5 paired with exon 7 reverse primer. mAR-Vabc-specific primers were designed on the unique junction mRNA (bp) Protein (amino acid) Molecular weight (KD) Cell Origin mAR-Va 2320 746 ~80 E8 mAR-Vb 2081 518 ~54 cE1 mAR-Vc 2029 526 ~56 cE1 73 sequences of each ARV to obtain the similar size of PCR products. The specificity of the quantitative real-time PCR was confirmed by gel electrophoresis (Fig 3-3B). In all E- and cE- cell lines analyzed, quantitative analysis of these ARVs by ARV-specific real-time PCR demonstrated that the mRNA expression levels of mAR-Vabc were substantially less than those of AR-FL, ranging from 1.9% to 23% of AR-FL mRNA (Fig 3-3C). In contrast to relatively equal expression of AR-FL among different cell lines, mRNA expression profile of mAR-Vabc seemed to vary between E- and cE- cell lines. mAR-Va expressed significantly more in E4 and E8 than that of cE1 and cE2 cells; conversely, cE- series had an increased enrichment of AR-Vb and AR-Vc compared to E-series (Fig 3-3D). Similar to the mRNA level, hypothetical mAR-V protein products corresponded to the levels detected, although it is based on a rough quantification of the Western blot data from Fig 1A (Fig 3-3E). Together these results showed that mAR-Vs are all expressed in the murine prostate cancer cell lines established from conditional Pten-null mouse model; however, their enrichment seemed to vary with respect to the stage of the disease from which the cell lines were established. 74 75 76 Figure 3-3. mRNA Expression of mARVs in mouse prostate cancer cell lines. A. Examination of the presence of mAR-Vs in all E- and cE- cell lines by conventional PCR. B. Real-time PCR products of mAR-Vabc and AR-FL amplified from E8 cDNA were examined by DNA gel electrophoresis. The loading amount of AR-FL was 10 times more than those of mARVs. C. Real-time PCR was performed to quantitate the mRNA expression levels of mAR-Vabc in comparison to that of AR-FL. The expression profiles of mARVs were further plotted in a higher resolution (shown in D). D. mARVs expressed as the minor species in the murine prostate cancer cells. However, the enrichment of individual kind of ARVs seemed to be specifically correlated with the disease phrase. The ARVs-specific primers used for the regular and quantitative real-time PCR were described in Table 3&4. Each column in the charts represented the mean ratio of independent triplicates with normalization to β-actin. *: p<0.05 E. Protein levels of mAR-Vabc and AR-FL in E8 and cE1 were estimated on the band intensities of ~85 kDa, 56 kDa and 50 kDa by Gel Doc™ XR+ system. Data were normalized by actin levels and then plotted. 77 mAR-Vs are up-regulated at mRNA levels in response to androgen depletion There are reports that the expression level of the human ARVs is generally increased in a variety of castration-resistant human PCa cell lines, xenograft models and tumor specimens as well 59, 62-66 . To quantitatively examine the extent of mAR-Vs variations in the androgen depleted environment, we performed the real-time PCR using ARV-specific primers to compare mRNA levels of mAR-Vabc between parental and their CSS- or casodex-treated derivatives. In E-series cells, relative abundance of all three mAR-Vs to AR-FL at mRNA level was significantly increased in E8-CSS but just modestly up- regulated in E4-CSS compared to the parental cells. In cE-series, CSS treatment seemed to be less affective in mAR-Vs expression. Only mAR-Va and mAR-Vc of cE1-CSS cells were increased at mRNA levels than that of cE1; whereas neither mARV showed any difference in their expression levels between cE2 and cE2-CSS (Fig 3-4A). On the other hand, all mAR-Vabc were significantly up- regulated in proportion to AR-FL when E8 and cE2 cells were treated with casodex, among which the expression level of mAR-Va was increased the most drastically (Fig 3-4B). Casodex is an AR antagonist competing with DHT for binding to the AR-LBD 128 . The mARVs we have identified were all devoid of LBD, which then should be left untargeted. Thus, the relative abundance of mARVs to AR-FL was more significantly up-regulated when AR-FL was directly inhibited via androgen depletion through the use of CSS. Together, our data 78 indicated that when murine cancer cells were treated with either CSS or casodex, the relative abundance of ARVs was generally up-regulated suggesting a negative correlation between mARVs and AR-FL mRNA levels in response to androgen ablation or antagonism (Fig 3- 4 A & B). 79 Figure 3-4. The relative expression levels of mAR-Vs were increased in response to androgen withdrawal. A. Quantitative expression of mARVs was compared using real-time PCR between primitive cancer cell lines and their CSS derivatives. B. Total RNA of E8 and cE2 treated with or without casodex were extracted and reverse transcribed into cDNA. Expression profiles of mAR-Vabc in these samples were then quantified. Columns, the mean ratio of independent triplicates adjusted by AR-FL mRNA levels, with normalization to β-actin; bars, SE. *: p<0.05 80 mAR-Vs expression is increased in mouse prostatic tumors and in aged normal prostates Considering the cellular heterogeneity in a tumor mass as well as the expression of AR in prostatic fibroblasts 8, 33, 72 , we attempted to analyze the relative abundance of AR and its truncated products in the whole tumors of the conditional Pten knockout mouse PCa model. In most of ADCa tumors that were collected from different age groups, we could detect expression of AR-FL. In contrast, 6 out of 7 CRPC tumors appeared to lack a level of AR-FL that could be detected by the N20 antibody used (Fig 3-5A). Whether this lack of detection is due to mutational or post-translational modifications of the amino terminus of the AR protein, however, remains to be examined before a real absence or reduction in AR-FL in CRPC could be assessed. Another interesting observation was that the level of AR-FL detection appeared to decrease in the aged prostates of normal mice. As illustrated in Fig. 3-5B, strong expression of AR-FL detected in young males gradually decreased to almost undetectable levels in the prostates of 20 month-old males. In contrast, short ARV species could be detected, although at variable levels, in all samples irrespective of origin from normal, ADCa or CRPC. To obtain a better determination of the distribution of mAR- Vs/AR-FL in these prostatic tissues, we performed the quantitative real-time PCR using the ARV specific- primers on mRNAs extracted from age-matched prostates of each wild-type, ADCa or CRPC. The relative abundance of mAR-Vs to AR-FL was significantly increased at mRNA levels in both murine ADCa and 81 CRPC prostatic tumors compared to the wild-type one (Fig 3-5C), indicating a potential etiological role of the variants in disease development. It is also noteworthy that while certain human ARVs, such as, AR-V7/AR3 and ARV567es that were previously correlated to poor prognosis 62, 63, 66 , it is of interest that expression of mAR-Vabc could be detected in normal mouse prostates. However, the relative distribution of mAR-Vs and AR-FL was drastically altered in the ageing mice. In 2 month-old mice, mAR-Vs were definitely the minor AR species with expression levels ranging from 0.3%~27% of AR-FL (Fig 3-5D top). One of the 5 month-old mouse showed a significant enrichment of mAR-Vs with its mAR-Vc expression reaching up to 73.8% of AR-FL (Fig 3-5D middle). Furthermore, in the prostate derived from a 23 month-old mouse, expression of mAR-Vs was determined higher than AR-FL, making them dominant species in this sample (Fig 3-5D bottom). This gradual decline in mRNA expression profile of AR-FL seemed to correlate with the observed pattern of AR protein expression detected in western blot analyses (Fig 3-5B & 5D). Taken together, our data suggested that, besides a potential etiological role of mAR-Vs in the development and progression of prostate cancer, mAR-Vs might also be involved in other aspects of biology, such as, ageing. 82 83 Figure 3-5. Expression of mAR-Vs in mouse prostates. A. AR protein status in the cell lysates extracted from androgen-dependent prostate cancer (ADCa) and castration-resistant prostate cancer (CRPC) developed from the conditional Pten-deletion mouse model of prostate adenocarcinoma. Tumors were dissected from mice bearing AD tumors at the age of 2-, 13- and 23- month-old, or those having CR tumors at the age of 13- and 23- month-old. B. Prostates of wild-type mice in 129/BALB/c background at the age of 2-, 5-, 13- and 20- month-old were dissected and lysed into cells. AR expression was detected by western blot using antibody recognizing epitope at the NH2-terminus. C. Total RNAs were extracted from age-matched wild-type, ADCa and CRPC prostatic tissues and subjected to ARV specific-real-time PCR. D. Total RNAs were extracted from the prostates dissected from wild-type mice at the age categories of 2-, 5- and 23-month-old, which were then subjected to ARV specific-real-time PCR. Expression profiles of mAR-Vabc obtained from C. and D. were then quantified. Columns, mean ratios of independent triplicates adjusted by AR-FL mRNA levels, with normalization to β-actin, bars, SE. 84 The transcriptional activity of AR-FL is suppressed by certain mAR-Vs in the presence of low levels of androgen To determine the biological activity of the mAR-Vs identified in the mouse prostate cancer cell lines we used transcription activation assays. An AR negative non-prostate cell line, namely simian fibroblast-like COS-1 was selected to avoid any possible interference from endogenous ARVs/AR-FL. Plasmid expression vectors containing the individual AR-Vs and AR-FL sequences were transfected into COS-1 cells along with one of two AR-responsive luciferase reporters (MMTV-Luc or PSA-Luc) and a control renilla luciferase reporter (pRL- SV40) that was not sensitive to androgens 123 . Expression of proteins of mAR-Va, b, c encoded by each expression plasmid was verified by detection of appropriate size protein bands in the transfected COS-1 cells by western blots (Fig 3-6A). Expression of human AR-FL (hAR-FL) resulted in the activation of both MMTV and PSA reporters following DHT treatment, although stronger induction of PSA-Luc indicated that it functioned as a more sensitive reporter in COS-1 cells (Fig 3-6B). However, none of mAR-Vs was determined to have the ability to induce the MMTV- or the PSA- Luc activity with/without DHT supplement. This observation was inconsistent with the previous reports that the majority of human ARVs and mouse AR-V4 tested displayed constitutive transcription activity in the absence of androgens. Although having a similar molecular organization to AR V567es 59 and mAR-V4 58 , mAR-Va failed to activate the reporters by itself in the COS-1 cells. It was, however, not surprising that 85 novel mAR-Vbc didn't display transcriptional activity as they didn’t contain the DBD region required for binding to the AR response elements 18, 23 . To investigate whether they could function cooperatively with AR-FL, we transfected COS-1 cells with mAR-Vs, PSA-Luc and the expression plasmids encoding either full-length human AR or rat AR. What we found was that, rather than the anticipated positive role, each one of AR-Vabc reduced the transcriptional activity of AR-FL to variable extent and in a dose dependent fashion (Fig 3-6C& D). Following the DHT treatment at the concentration of 0.03 nM (nanomolar) and 0.3 nM, mAR-Va and mAR-Vc significantly inhibited the transcription activity of hAR-FL by 31% with only 25 ng (nanogram) input in the transfection assays compared to control vector, which has increased to 92.8% suppression by mAR- Va and 89.7% by mAR-Vc with the addition of 1000 ng mAR-Vs. On the other hand, mAR-Vb didn’t display its significantly suppressive function until 250 ng was added in COS-1 cells and resulted in a lesser inhibition on hAR by 55.8% at most (Fig 3- 5C). Considering the lower degree of sequence homology between human and mouse AR as compared to rat and mouse AR (97% amino acid sequence identity) 23 , the expression plasmid encoding rat AR-FL (rAR) was also assessed together with mAR-Vs in the luciferase reporter assays. At an environment of low level of added DHT (0.03nM), mAR-Vs were determined to be effective also to inhibit the transcription activities of rAR-FL. The degree of inhibition ranged from 45% to 92.3%, and based on their effectiveness, could be ranked as mARVb<mARVc<mARVa (Fig 3- 6D). To further delineate the 86 mechanism by which mAR-Vs suppress the activity of AR-FL, we performed the co-immunoprecipitation to study if these mAR-Vs would have physically associated with AR-FL. Plasmids encoding Myc-tagged mAR-Va and mAR-Vc were established for this assay based on the fact that these two ARVs inhibited the AR-FL more significantly. We co-transfected COS-1 cells with rat AR and mAR-Va-myc (Va-myc) or mAR-Vc-myc (Vc-myc), and then immunoprecipitated the mAR-Va/Vc proteins from whole cell lysates using anti-myc antibody. Western blot analyses were then conducted using an antibody specific for the C- terminus of AR which only recognized AR-FL. We found that the 110 kDa band was detected in the Va-myc immunoprecipitates suggesting that mAR-Va could form a stable complex with AR-FL. However, since a significant AR-FL band was not detected in the Vc-myc immunoprecipitates, we inferred that such stable association was not formed between AR-FL and AR-Vc (Fig 3-6E). Taken together, our data suggested that mAR-Vabc identified from our conditional Pten- deficient mouse model of prostate cancer, irrespective of their ability to associate with the AR-FL, could function as negative regulators of AR activity, at least with respect to the promoters and the cell system used for the analyses. This functionality was found to require a low level of androgen and the magnitude of suppressor activity appeared to be dose-dependent. 87 88 89 Fig 3-6. The transcriptional activity of AR-FL is inhibited by mAR-Vs in the presence of androgen. A. COS-1 cells were transfected with MMTV- (top) or PSA- (bottom) luciferase reporter, along with each expression plasmid as indicated. At 24h post-transfection, cells were treated with or without 1 nM of DHT for 24 hrs and luciferase activities were measured. B. Overexpression of each expression plasmid used for the assay was examined by western blot analyses using anti-AR (N-terminus) and anti-Actin. C. COS-1 cells were transfected with PSA-luc reporter and expression plasmids encoding human AR- FL, in conjunction with increasing doses of mAR-Vs expression vectors. Empty vector was used to equilibrate total DNA in all transfections. Cells were treated with either 0.03 nM or 0.3 nM DHT for 24 hrs followed by luciferase measurement. D. Rat AR-FL was also used in the transient transfection and luciferase reporter assays as described in C. E. COS-1 cells were co-transfected with plasmids encoding rat AR and either empty vector, Va-myc or Vc-myc. 48 hr post-transfection, cell lysates were immunoprecipitated with anti-myc or IgG followed by western blot analysis using anti-AR (C-19). 90 3.5 Discussion Mechanisms of AR restoration that accompany the emergence of CRPC have yet to be elucidated. Extensive studies on this field have demonstrated that it is an extremely dynamic and complex process that involves multiple regulatory pathways including both genetic and epigenetic alterations 5, 42, 43 . AR overexpression has been found in up to 30% of CRPC tumors 44 ; while on the other hand, most clinical studies did not establish any significant correlation between AR expression and prognosis 67, 129, 130 . Although methodological limitation in detecting the expression of AR must be carefully taken into consideration, it has been implicated that other regulatory pathways might act simultaneously to achieve a balance of AR activity in certain cancer cells 69, 70 . However, the retention of AR sensitivity in our cE-series cells 95 as well as other evidence obtained from various CRPC model systems have all suggested that AR activity seems to be aberrantly maintained even in the stage of CRPC. Recent identification of several ARVs in human CRPC model systems has indicated a new mechanism by which AR activity may be restored in the castrated level of androgen. Functionally, the majority of ARVs identified so far are predicated to encode proteins lacking the LBD but retaining the transactivation NTD, acting as the constitutively active transcription factors independent of androgens which have conferred the castration resistance in prostate cancer 57-59, 62-64 . To scrutinize the process that how these ARVs may occur thereby contributing to the emergence of CRPC has become of great 91 importance to validate the hypothesis that these ARVs may serve as the potentially new targets in the clinical intervention. Therefore, our homologous systems of mouse ADCa and CRPC would provide a platform to assess the formation and functional implications in disease progression. In this study, we have defined a set of three mouse AR-Vs composed of novel molecular structures. Interestingly, the expression of these mAR-Vs is not restricted to the CRPC cell lines as they are detected in those of the ADCa tumor cells as well. mAR-Va, retaining the contiguous exon 1-4 followed by read- through into intron 4, is roughly analogous to the previously reported mAR-V4 58 and AR V567es 59 except for the unique amino acid sequence encoded by the 3’- end (Fig 3-7). mAR-Vb and mAR-Vc are composed of two different regions located within intron 1 to be spliced after exon 1, leading to the deduced ARVs protein containing only NTD. AR8 identified in 22rv1 cells was the closest human counterpart containing NTD followed by a longer 33-aa tail encoded by an Exon 3’ located within intron 3 57 . The schematic comparison of mAR-Vabc and their analogues is illustrated in (Fig 3-7). By co-immunoprecipitation we demonstrate that mAR-Va, like AR V567es 59 , could interact with AR-FL. mAR-Vb and mAR-Vc lacking the dimerization interface located within DBD 18 , as expected, could not interact similarly with AR-FL. 92 Fig 3-7. Schematic illustrations of mAR-Vabc with previously published human and mouse ARVs. The predicated amino acid sequence translated by m4a and m1c was listed. A. mAR-Va has retained the NTD, DBD and hinge, resembling the previously reported mouse AR-V4 and human ARV567es. B. mAR-Vb and mAR-Vc were composed of only NTD, similar as their human counterpart AR8 except the differences on their CTD encoded by novel cryptic exons. 93 In reporter assays, we couldn’t detect any significant transcriptional activity for mAR-Va, Vb or Vc in COS-1 cells. However, all three ARVs could significantly inhibit the transcriptional activity of AR-FL in the presence of low levels of androgens. The lack of transcription activity of mAR-Va alone is in contrast to its previously reported analogues but could be potentially attributed to the presence of 42 a.a. long tail at its C-terminus (Fig 3-7). However, the constitutive activity of the previously reported ARVs has been shown to be specifically dependent on the cellular context 51 . Thus, the inhibitory transcription activity of mAR-Vs to AR-FL, in particular the mAR-Va which still contains its DBD, should be further investigated and confirmed in other cell lines. Of note, the observation that mAR-Vb and Vc can function as inhibitor of AR-FL activity is in agreement with a previously reported work that expression of transduced NTD alone can serve as decoy molecules to interfere with AR activity in prostate cancer cell lines 131, 132 . What is even more interesting is that these novel NTD only ARVs seem to be the dominant ARV species in cE-series suggested by the fourth kind of mAR-V identified in cE1 cells, termed as mAR-Vd (Fig 3-8). AR-Vd was discovered indeed when we intended to clone mAR-Vc using the reverse primer complementary to the specific m1c of mAR-Vc. Surprisingly, 8 out 10 colonies have turned out to be mAR-Vd indicating the more abundance of mAR-Vd over mAR-Vc in cE1 cells. Compared to mAR-Vb and mAR-Vc, mAR-Vd seems to be generated through a more complicated splicing event, encoding a protein composing of the 20 a.a. tail followed by NTD (Fig. 3- 94 8). It would be very interesting to examine whether there would be any functional implication of the conceivable enrichment of the NTD only mAR-Vs in the cE- series cells. It was proposed that AR NTD would prevent activation of the AR-FL by interacting with some unidentified co-regulators that co-operate with the endogenous AR 131, 132 . Our study, for the first time describes the natural occurrence of these AR NTDs in the mouse prostate cancer cells and in tissues, thereby raising a considerable level of interest in these molecules. Furthermore, we present evidence that the proportion of these decoy NTDs appear to increase at CRPC tumors while there is a decline in AR-FL transcripts. Most interestingly, we also present clues that truncated molecules like the NTDs are enriched in the ageing prostates of the otherwise normal mouse. Thus, there seems to be a correlation between advanced prostate cancer and aged prostate, at least in the mouse, with respect to substitution of AR-FL by increasing levels of decoy like peptides. 95 Fig 3-8. Molecular and protein structures of mAR-Vd. The text map of mAR- Vd was generated by ApE- A plasmid Editor. The primer sets which were originally designed to clone AR-Vc were highlighted in Pink. Blue refers to the novel exon spliced specifically into mAR-Vd, termed as m1d. The common sequences shared between m1c and m1d were highlighted in green. Red highlight indicated the deduced stop codon of the protein encode by mAR-Vd. 96 CHAPTER FOUR: DISCUSSIONS AND REMARKS ABOUT FUTURE STUDIES 4.1 Discussions The primary significance of this dissertation is that we have established a homologous study system composed of ADCa and CRPC cell lines that were derived from our cPten -/- L mouse model of prostate cancer 95 . Of note, these cell systems are indeed first to be established from a genetically engineered mouse model representing spontaneous tumor formation and recurrence in animals with intact immune system. These cell lines are likely to serve as useful resources for understanding various parameters that may discriminate between ADCa and CRPC cancer cells. In fact, they are already being utilized by several other research groups to investigate many parameters such as, detection of circulating tumor cells, syngeneic tumor transplantation, miRNAs, etc 133-137 . The second significance of this dissertation is that we have first identified a set of mouse androgen receptor splice variants (termed as mAR-Vabc) that occur in these cells. The novelty of these mAR-Vs lies in their structural elements and in functional attributes. Understanding of their biology is likely to be important at two major fronts: a role in suppression of AR overactivity in prostate cancer and in ageing prostate, and in potential insight to new strategies to down-regulate AR for therapeutic or preventive measures. 97 Mechanisms of AR reactivation that accompany the emergence of CRPC largely remain to be elucidated. What is becoming increasingly evident is that the process is extremely complex involving multiple regulatory pathways that are influenced by both genetic and epigenetic modifications. Furthermore, it is also apparent that a balance in AR activity may be a crucial requirement for the cancer cells to survive and proliferate. Besides overexpression of AR in a large fraction of CRPC, reactivation has been implicated through AR mutations, gene rearrangements and post-translational AR protein modifications in various cases of CRPC 47, 48, 129, 138-140 . While such mechanisms coupled with in situ androgen synthesis 52 may account for stimulation of AR function in the cancer cells in the environment of castrate levels of androgens, there must be evolution of simultaneous processes that restrict overactivation of AR. For example, previous studies have reported independently inconsistent expression status of AR. One study has shown increased mRNA but reduced protein expression of AR in CRPC tumors derived from their in vivo CRPC xenograft models 68 . Another study has demonstrated that specific CpGs on AR promoter might be aberrantly methylated in the development of hormone independence leading to a subset of CRPC with no AR expression 141 . There is also evidence that AR may promote entosis, a non-apoptosis cell death mechanism to suppress prostate cancer cell growth 70 . Moreover, several other studies have proposed that AR inactivation might stimulate the transdifferentiation of neuroendocrine which tend to be enriched in some CRPC tumors 69, 73 . In this regard, it is not surprising why it is 98 difficult to establish any significant correlation between AR levels and clinical prognosis 129 ; although it is clear that AR activity is generally maintained throughout the prostate pathogenesis and cancer progression. Recent studies have identified another mechanism by which AR may escape the ligand dependence. The occurrence of multiple AR splice variants has been reported in human CRPC tissues and cell lines and a limited number of them have been functionally characterized. Particularly, the ARVs that lack ligand binding domain but retain DNA binding and amino terminal transactivation domains are of much interest since they are shown to have the ability to act in the absence of ligands. This type of constitutive activity could be related to CRPC status. However, tissues or cell lines derived from the same patient at both ADCa and CRPC are not available for undertaking a systematic study of ARVs in disease progression. Here lies the importance of our study. We utilized a homologous system of mouse ADCa and CRPC to begin to evaluate the genesis of mARVs. In Chapter 2, we described establishment and characterization of the two series of murine prostate cancer cell lines, one from the primary tumor (ADCa) of the cPten -/- L mice, and the other from the CRPC of the same mouse model. The biallelic loss of Pten is confirmed in all murine cell lines which supports the epithelial origin and purity for these cells. E2 and E4 were established from one ADCa tumor. These cell lines manifest some degree of EMT (epithelial- mesenchymal transition) that is supported by the high expression of EMT specific 99 markers. In contrast, E8, also established from an independent ADCa tumor, doesn’t seem to be actively undergoing EMT due to the expression epithelial markers and low levels of EMT markers. The differential patterns of morphology and behavior between E2/E4 and E8 imply existence of epithelial cell heterogeneity within the primary tumor of the model system. In contrast, none of the cE1 and cE2 cell lines derived from one CRPC tumor appears to have EMT phenotype and they maintain high mRNA expression of epithelial markers. Compared to the cells of the E-series, cE-series demonstrate better proliferative capability in the environment of low levels of androgen, consistent with their origin from the CR phenotype. E-series, no matter whether undergoing EMT or not, displays a limited proliferation rate under androgen starvation, a trait which is consistent with the traditional AD features. As expected, when supplemented with the additional ligands, the growth rates of all five cell lines are increased. The fact that cE-series retain the sensitivity to androgen but do not depend on the presence of androgen for their optimal growth, is reminiscent of the human C42B and CWR22rv1 PCa cell lines that harbor AR activity 84 , just as most tumors of the CRPC patients. The state of functionally active AR in cE-series has been further substantiated by the nuclear AR expression detected by immunohistochemistry (IHC) in the epithelial cells located with the glandular structures of the subcutaneously grown tumor grafts. In Chapter 3, we describe the identification and characterization of three novel mouse AR-Vs (termed as mAR-Vabc) in our homologous mouse PCa systems. They are expressed in all 100 five cell lines. The transcription levels of mAR-Vabc are up-regulated in response to androgen ablation. More interestingly, we have found that the relative abundance of mAR-Vabc to prototype AR has enriched more in the tumors and aged normal prostates, indicating the potential roles of these mAR-Vs during the etiological and other biological processes. Potentially, these mAR-Vs may participate in prostate cancer progression through a unique mechanism by serving as negative regulators of the transcription activity of the prototype AR. Once their putatively negative functions on the prototype AR have been proved in other PCa models, these mAR-Vs may lead to development of new strategies towards control of prostate tumorigenesis and progression. If their expression is to restrain AR function in the tumor cells, it may be argued that their overexpression may induce increased inhibition of AR to the point that cancer cell survival may be compromised. In this regard, it is noteworthy that in a recent study when a truncated version of AR carrying the NTD alone was genetically expressed in prostate cancer cells, a significant inhibition of AR function was achieved 131, 132 . These decoy products are basically similar to naturally expressed products of mAR-Vb and mAR-Vc of our study. Thus, it is logical to believe that if synthesis of these natural NTDs could be induced and elevated, a stronger suppression of AR activity may be accomplished. In fact, our results show a NTD dose related effect on AR activity in the COS-1 cells. 101 Furthermore, these natural decoys of AR may have a biological role analogous to cancer in suppressing or down-regulating AR function in the ageing process. These are exciting clues and deserve to be further investigated. A critically important topic would be to determine the mechanisms by which abnormal splice variants are indeed formed. Could this process be stimulated in prostate cancer or inhibited in aged animals by the design of small chemical inducers or suppressors, respectively? It may be suggested that once the factors that drive AR NTD genesis in the prostate cells, chemical inducers of the processes could be developed to augment their formation in prostate cancer to impede AR activity in a hitherto novel way to suppress prostate cancer progression. These peptides being naturally endogenous, their toxicity may be minimal and they are likely to be immunologically tolerant, thereby enhancing potential therapeutic efficacy. 4.2 Future studies The homologous system composed of mouse PCa cell lines derived from two different stages of the disease that we established should continue to serve as an important resource for both basic and pre-clinical investigations of prostate cancer. A future direction that I envision is that these cells could provide a test system for the evaluation of stem-like cell activity that may be central to both primary and recurrent prostate cancer. We have already successfully enriched a subpopulation of cancer stem/progenitor cells from cE1 cells on the 102 basis of our previously established methodology. This subpopulation is able to form spheroids mimicking prostate glandular structures, and the ability to form spheroid is enhanced when exposed to factors secreted from the cancer- associated fibroblasts. A protein factor has been identified that displays spheroid enhancing activity. Similarly, stem cell-like cells may be enriched from other cell lines of the cE and E series for the analysis of their response to this or other factors for the evaluation of parameters that drive CRPC induction and growth. The novel mAR-Vs we have identified raise several new questions. First, the significant enrichment of these mAR-Vs in proportion to the decreasing prototype AR in ageing mouse prostate has prompted us to consider that a physiological role may be in play. Several naturally occurring AR variants have been reported in normal tissues as well as the androgen insensitivity syndromes (AIS) patients 19, 59, 62 . For example, AR45, containing the entire AR DBD, CTD but a novel 7 a.a. sequence replacing the wild-type NTD, is expressed in several tissues including heart, muscle, uterus, and prostate, and described as a dominant negative regulator of AR signaling 142 . The molecular structures of those AR variants, however, differ from the ones identified in our prostate cancer model in that they reflect exon skipping, probably due to point mutations within the introns. Therefore, it would be of great interest to assess whether the mAR- Vs might also be associated with other diseases or enrich in non-prostate tissues of the body with impact on pathological and physiological processes. Second, previous studies have indicated the differential role of AR in different types of 103 prostate cancer cells, suggesting that AR may function as a tumor suppressor in the epithelial cells but an oncogene in the stromal cells of the prostate cancer 71, 72 . These findings may account for one of the reasons why so much controversial data have accumulated regarding AR signaling in the disease progression, suggesting that it would be necessary to dissect the role of AR in the individual cellular compartments of the prostate. Thus, it would be logical to ask if or how the expression pattern of these mAR-Vs may differ in the fibroblastic cells of the mouse prostate that we established from the cancer-associated fibroblasts (CAF) from the same model system. If there is a differential role of mAR-Vs in the regulation of cell proliferation and survival pathways with respect to their specific distributions in the microenvironment, it would be very important information for our understanding of the mechanisms underlying the disease progression. Lastly, although in general most of AR-Vs identified so far have expression in low levels in either tumor cell lines or tissues, it does not exclude the possibility that some of these variants may enrich preferentially in certain subpopulation of cells, such as cancer stem/progenitor cells. In fact, we have obtained some preliminary data, based on single cell real-time PCR analysis using a microfluidic system, to suggest that mRNA of these mAR-Vs is not expressed at similar level in every single cell. More interestingly, we have also obtained clues that the enriched cancer stem/progenitor cells are responsive to the increased levels of androgen, thereby implicating AR functionality in the cancer stem cells (CSC). Therefore, these preliminary data prompt us to further assess the expression and/or activity 104 of mAR-FL and mAR-Vs in the CSC population. The findings should be valuable for a better understanding of how these novel ARVs may be formed in response to certain cellular stress and environment. 105 BIBLIOGRAPHY [1] American Caner Society A, GA, USA, 2013.: http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/documents /document/acspc-037124.pdf. [2] Basu S, Tindall DJ: Androgen action in prostate cancer. Hormones & cancer 2010, 1:223-8. [3] Figg WD, Feuer JA, Bauer KS: Management of hormone-sensitive metastatic prostate cancer. Update on hormonal therapy. Cancer practice 1997, 5:258-63. [4] Roy-Burman P, Tindall DJ, Robins DM, Greenberg NM, Hendrix MJ, Mohla S, Getzenberg RH, Isaacs JT, Pienta KJ: Androgens and prostate cancer: are the descriptors valid? Cancer biology & therapy 2005, 4:4-5. [5] Lamont KR, Tindall DJ: Minireview: Alternative activation pathways for the androgen receptor in prostate cancer. Molecular endocrinology 2011, 25:897- 907. [6] Shen MM, Abate-Shen C: Molecular genetics of prostate cancer: new prospects for old challenges. Genes & Development 2010, 24:1967-2000. [7] Labrie F, Luu-The V, Labrie C, Simard J: DHEA and Its Transformation into Androgens and Estrogens in Peripheral Target Tissues: Intracrinology. Frontiers in Neuroendocrinology 2001, 22:185-212. [8] Dehm SM, Tindall DJ: Molecular regulation of androgen action in prostate cancer. Journal of cellular biochemistry 2006, 99:333-44. [9] Harris WP, Mostaghel EA, Nelson PS, Montgomery B: Androgen deprivation therapy: progress in understanding mechanisms of resistance and optimizing androgen depletion. Nature clinical practice Urology 2009, 6:76-85. [10] Lamb DJ, Weigel NL, Marcell M: Androgen receptors and their biology. Vitamins & Hormones: Academic Press, 2001. pp. 199-230. [11] Bennett NC, Gardiner RA, Hooper JD, Johnson DW, Gobe GC: Molecular cell biology of androgen receptor signalling. The international journal of biochemistry & cell biology 2010, 42:813-27. [12] Shang Y, Myers M, Brown M: Formation of the Androgen Receptor Transcription Complex. Molecular cell 2002, 9:601-10. 106 [13] Heinlein CA, Chang C: Androgen Receptor in Prostate Cancer. Endocrine Reviews 2004, 25:276-308. [14] Naar AM, Lemon BD, Tjian R: Transcriptional coactivator complexes. Annual review of biochemistry 2001, 70:475-501. [15] Foradori CD, Weiser MJ, Handa RJ: Non-genomic actions of androgens. Front Neuroendocrinol 2008, 29:169-81. [16] Migliaccio A, Castoria G, Di Domenico M, de Falco A, Bilancio A, Lombardi M, Barone MV, Ametrano D, Zannini MS, Abbondanza C, Auricchio F: Steroid- induced androgen receptor-oestradiol receptor beta-Src complex triggers prostate cancer cell proliferation. The EMBO journal 2000, 19:5406-17. [17] Unni E, Sun S, Nan B, McPhaul MJ, Cheskis B, Mancini MA, Marcelli M: Changes in Androgen Receptor Nongenotropic Signaling Correlate with Transition of LNCaP Cells to Androgen Independence. Cancer Research 2004, 64:7156-68. [18] Dehm SM, Tindall DJ: Androgen receptor structural and functional elements: role and regulation in prostate cancer. Molecular endocrinology 2007, 21:2855- 63. [19] Dehm SM, Tindall DJ: Alternatively spliced androgen receptor variants. Endocrine-related cancer 2011, 18:R183-96. [20] Bain DL, Heneghan AF, Connaghan-Jones KD, Miura MT: Nuclear receptor structure: implications for function. Annual review of physiology 2007, 69:201-20. [21] Sack JS, Kish KF, Wang C, Attar RM, Kiefer SE, An Y, Wu GY, Scheffler JE, Salvati ME, Krystek SR, Jr., Weinmann R, Einspahr HM: Crystallographic structures of the ligand-binding domains of the androgen receptor and its T877A mutant complexed with the natural agonist dihydrotestosterone. Proceedings of the National Academy of Sciences of the United States of America 2001, 98:4904-9. [22] Faus H, Haendler B: Post-translational modifications of steroid receptors. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie 2006, 60:520-8. [23] Charest NJ, Zhou ZX, Lubahn DB, Olsen KL, Wilson EM, French FS: A frameshift mutation destabilizes androgen receptor messenger RNA in the Tfm mouse. Molecular endocrinology 1991, 5:573-81. 107 [24] Hur E, Pfaff SJ, Payne ES, Gron H, Buehrer BM, Fletterick RJ: Recognition and accommodation at the androgen receptor coactivator binding interface. PLoS Biol 2004, 2:E274. [25] Klokk TI, Kurys P, Elbi C, Nagaich AK, Hendarwanto A, Slagsvold T, Chang CY, Hager GL, Saatcioglu F: Ligand-specific dynamics of the androgen receptor at its response element in living cells. Molecular and cellular biology 2007, 27:1823-43. [26] Zhou ZX, Sar M, Simental JA, Lane MV, Wilson EM: A ligand-dependent bipartite nuclear targeting signal in the human androgen receptor. Requirement for the DNA-binding domain and modulation by NH2-terminal and carboxyl- terminal sequences. The Journal of biological chemistry 1994, 269:13115-23. [27] Faber PW, King A, van Rooij HC, Brinkmann AO, de Both NJ, Trapman J: The mouse androgen receptor. Functional analysis of the protein and characterization of the gene. The Biochemical journal 1991, 278 ( Pt 1):269-78. [28] Jenster G, van der Korput HA, Trapman J, Brinkmann AO: Identification of two transcription activation units in the N-terminal domain of the human androgen receptor. The Journal of biological chemistry 1995, 270:7341-6. [29] Zhu P, Baek SH, Bourk EM, Ohgi KA, Garcia-Bassets I, Sanjo H, Akira S, Kotol PF, Glass CK, Rosenfeld MG, Rose DW: Macrophage/cancer cell interactions mediate hormone resistance by a nuclear receptor derepression pathway. Cell 2006, 124:615-29. [30] Baek SH, Ohgi KA, Nelson CA, Welsbie D, Chen C, Sawyers CL, Rose DW, Rosenfeld MG: Ligand-specific allosteric regulation of coactivator functions of androgen receptor in prostate cancer cells. Proceedings of the National Academy of Sciences of the United States of America 2006, 103:3100-5. [31] Dotzlaw H, Moehren U, Mink S, Cato AC, Iniguez Lluhi JA, Baniahmad A: The amino terminus of the human AR is target for corepressor action and antihormone agonism. Molecular endocrinology 2002, 16:661-73. [32] Poukka H, Karvonen U, Janne OA, Palvimo JJ: Covalent modification of the androgen receptor by small ubiquitin-like modifier 1 (SUMO-1). Proceedings of the National Academy of Sciences of the United States of America 2000, 97:14145-50. [33] Roy-Burman P, Wu H, Powell WC, Hagenkord J, Cohen MB: Genetically defined mouse models that mimic natural aspects of human prostate cancer development. Endocrine-related cancer 2004, 11:225-54. 108 [34] Litvinov IV, De Marzo AM, Isaacs JT: Is the Achilles' heel for prostate cancer therapy a gain of function in androgen receptor signaling? J Clin Endocrinol Metab 2003, 88:2972-82. [35] Huggins C, Hodges CV: Studies on prostatic cancer. I. The effect of castration, of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. CA: a cancer journal for clinicians 1972, 22:232-40. [36] Litvinov IV: Is the Achilles' Heel for Prostate Cancer Therapy a Gain of Function in Androgen Receptor Signaling? Journal of Clinical Endocrinology & Metabolism 2003, 88:2972-82. [37] Feldman BJ, Feldman D: The development of androgen-independent prostate cancer. Nat Rev Cancer 2001, 1:34-45. [38] Buchanan G, Irvine RA, Coetzee GA, Tilley WD: Contribution of the androgen receptor to prostate cancer predisposition and progression. Cancer metastasis reviews 2001, 20:207-23. [39] Chen CD, Welsbie DS, Tran C, Baek SH, Chen R, Vessella R, Rosenfeld MG, Sawyers CL: Molecular determinants of resistance to antiandrogen therapy. Nature medicine 2004, 10:33-9. [40] Wang Q, Li W, Zhang Y, Yuan X, Xu K, Yu J, Chen Z, Beroukhim R, Wang H, Lupien M, Wu T, Regan MM, Meyer CA, Carroll JS, Manrai AK, Janne OA, Balk SP, Mehra R, Han B, Chinnaiyan AM, Rubin MA, True L, Fiorentino M, Fiore C, Loda M, Kantoff PW, Liu XS, Brown M: Androgen receptor regulates a distinct transcription program in androgen-independent prostate cancer. Cell 2009, 138:245-56. [41] Pienta KJ, Bradley D: Mechanisms underlying the development of androgen- independent prostate cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 2006, 12:1665-71. [42] Nacusi LP, Tindall DJ: Androgen receptor abnormalities in castration- recurrent prostate cancer. Expert review of endocrinology & metabolism 2009, 4:417-22. [43] Knudsen KE, Kelly WK: Outsmarting androgen receptor: creative approaches for targeting aberrant androgen signaling in advanced prostate cancer. Expert review of endocrinology & metabolism 2011, 6:483-93. [44] Koivisto P, Visakorpi T, Kallioniemi OP: Androgen receptor gene amplification: a novel molecular mechanism for endocrine therapy resistance in 109 human prostate cancer. Scandinavian journal of clinical and laboratory investigation Supplementum 1996, 226:57-63. [45] Mohler JL: Castration-recurrent prostate cancer is not androgen- independent. Advances in experimental medicine and biology 2008, 617:223-34. [46] Linja MJ, Savinainen KJ, Saramaki OR, Tammela TL, Vessella RL, Visakorpi T: Amplification and overexpression of androgen receptor gene in hormone- refractory prostate cancer. Cancer Res 2001, 61:3550-5. [47] Taplin ME, Bubley GJ, Shuster TD, Frantz ME, Spooner AE, Ogata GK, Keer HN, Balk SP: Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer. The New England journal of medicine 1995, 332:1393-8. [48] Gottlieb B, Beitel LK, Wu JH, Trifiro M: The androgen receptor gene mutations database (ARDB): 2004 update. Human mutation 2004, 23:527-33. [49] Knudsen KE, Scher HI: Starving the addiction: new opportunities for durable suppression of AR signaling in prostate cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 2009, 15:4792- 8. [50] Haile S, Sadar MD: Androgen receptor and its splice variants in prostate cancer. Cellular and molecular life sciences : CMLS 2011, 68:3971-81. [51] Guo Z, Qiu Y: A new trick of an old molecule: androgen receptor splice variants taking the stage?! International journal of biological sciences 2011, 7:815-22. [52] Locke JA, Guns ES, Lubik AA, Adomat HH, Hendy SC, Wood CA, Ettinger SL, Gleave ME, Nelson CC: Androgen levels increase by intratumoral de novo steroidogenesis during progression of castration-resistant prostate cancer. Cancer Res 2008, 68:6407-15. [53] Culig Z, Hobisch A, Cronauer MV, Radmayr C, Trapman J, Hittmair A, Bartsch G, Klocker H: Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. Cancer Res 1994, 54:5474-8. [54] Craft N, Shostak Y, Carey M, Sawyers CL: A mechanism for hormone- independent prostate cancer through modulation of androgen receptor signaling by the HER-2/neu tyrosine kinase. Nature medicine 1999, 5:280-5. 110 [55] Murillo H, Huang H, Schmidt LJ, Smith DI, Tindall DJ: Role of PI3K signaling in survival and progression of LNCaP prostate cancer cells to the androgen refractory state. Endocrinology 2001, 142:4795-805. [56] Tamburrino L, Salvianti F, Marchiani S, Pinzani P, Nesi G, Serni S, Forti G, Baldi E: Androgen receptor (AR) expression in prostate cancer and progression of the tumor: Lessons from cell lines, animal models and human specimens. Steroids 2012, 77:996-1001. [57] Yang X, Guo Z, Sun F, Li W, Alfano A, Shimelis H, Chen M, Brodie AM, Chen H, Xiao Z, Veenstra TD, Qiu Y: Novel membrane-associated androgen receptor splice variant potentiates proliferative and survival responses in prostate cancer cells. The Journal of biological chemistry 2011, 286:36152-60. [58] Watson PA, Chen YF, Balbas MD, Wongvipat J, Socci ND, Viale A, Kim K, Sawyers CL: Constitutively active androgen receptor splice variants expressed in castration-resistant prostate cancer require full-length androgen receptor. Proceedings of the National Academy of Sciences of the United States of America 2010, 107:16759-65. [59] Sun S, Sprenger CC, Vessella RL, Haugk K, Soriano K, Mostaghel EA, Page ST, Coleman IM, Nguyen HM, Sun H, Nelson PS, Plymate SR: Castration resistance in human prostate cancer is conferred by a frequently occurring androgen receptor splice variant. J Clin Invest, 120:2715-30. [60] Marcias G, Erdmann E, Lapouge G, Siebert C, Barthelemy P, Duclos B, Bergerat JP, Ceraline J, Kurtz JE: Identification of novel truncated androgen receptor (AR) mutants including unreported pre-mRNA splicing variants in the 22Rv1 hormone-refractory prostate cancer (PCa) cell line. Human mutation 2010, 31:74-80. [61] Hu R, Isaacs WB, Luo J: A snapshot of the expression signature of androgen receptor splicing variants and their distinctive transcriptional activities. The Prostate 2011, 71:1656-67. [62] Hu R, Dunn TA, Wei S, Isharwal S, Veltri RW, Humphreys E, Han M, Partin AW, Vessella RL, Isaacs WB, Bova GS, Luo J: Ligand-independent androgen receptor variants derived from splicing of cryptic exons signify hormone- refractory prostate cancer. Cancer Res 2009, 69:16-22. [63] Guo Z, Yang X, Sun F, Jiang R, Linn DE, Chen H, Chen H, Kong X, Melamed J, Tepper CG, Kung HJ, Brodie AM, Edwards J, Qiu Y: A novel androgen receptor splice variant is up-regulated during prostate cancer progression and promotes androgen depletion-resistant growth. Cancer Res 2009, 69:2305-13. 111 [64] Dehm SM, Schmidt LJ, Heemers HV, Vessella RL, Tindall DJ: Splicing of a novel androgen receptor exon generates a constitutively active androgen receptor that mediates prostate cancer therapy resistance. Cancer Res 2008, 68:5469-77. [65] Zhang X, Morrissey C, Sun S, Ketchandji M, Nelson PS, True LD, Vakar- Lopez F, Vessella RL, Plymate SR: Androgen receptor variants occur frequently in castration resistant prostate cancer metastases. PloS one 2011, 6:e27970. [66] Hornberg E, Ylitalo EB, Crnalic S, Antti H, Stattin P, Widmark A, Bergh A, Wikstrom P: Expression of androgen receptor splice variants in prostate cancer bone metastases is associated with castration-resistance and short survival. PloS one 2011, 6:e19059. [67] Donovan MJ, Osman I, Khan FM, Vengrenyuk Y, Capodieci P, Koscuiszka M, Anand A, Cordon-Cardo C, Costa J, Scher HI: Androgen receptor expression is associated with prostate cancer-specific survival in castrate patients with metastatic disease. BJU international 2010, 105:462-7. [68] Zhou J-R, Yu L, Zerbini LF, Libermann TA, Blackburn GL: Progression to androgen-independent LNCaP human prostate tumors: Cellular and molecular alterations. International Journal of Cancer 2004, 110:800-6. [69] Wright ME, Tsai MJ, Aebersold R: Androgen receptor represses the neuroendocrine transdifferentiation process in prostate cancer cells. Molecular endocrinology 2003, 17:1726-37. [70] Wen S, Shang Z, Zhu S, Chang C, Niu Y: Androgen receptor enhances entosis, a non-apoptotic cell death, through modulation of Rho/ROCK pathway in prostate cancer cells. The Prostate 2013. [71] Niu Y, Chang TM, Yeh S, Ma WL, Wang YZ, Chang C: Differential androgen receptor signals in different cells explain why androgen-deprivation therapy of prostate cancer fails. Oncogene 2010, 29:3593-604. [72] Niu Y, Altuwaijri S, Lai KP, Wu CT, Ricke WA, Messing EM, Yao J, Yeh S, Chang C: Androgen receptor is a tumor suppressor and proliferator in prostate cancer. Proceedings of the National Academy of Sciences of the United States of America 2008, 105:12182-7. [73] Marchiani S, Tamburrino L, Nesi G, Paglierani M, Gelmini S, Orlando C, Maggi M, Forti G, Baldi E: Androgen-responsive and -unresponsive prostate cancer cell lines respond differently to stimuli inducing neuroendocrine differentiation. International journal of andrology 2010, 33:784-93. 112 [74] Morris MJ, Huang D, Kelly WK, Slovin SF, Stephenson RD, Eicher C, Delacruz A, Curley T, Schwartz LH, Scher HI: Phase 1 trial of high-dose exogenous testosterone in patients with castration-resistant metastatic prostate cancer. European urology 2009, 56:237-44. [75] Mathew P: Prolonged control of progressive castration-resistant metastatic prostate cancer with testosterone replacement therapy: the case for a prospective trial. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO 2008, 19:395-6. [76] Wu X, Gong S, Roy-Burman P, Lee P, Culig Z: Current mouse and cell models in prostate cancer research. Endocrine-related cancer 2013, 20:R155- R70. [77] Valkenburg KC, Williams BO: Mouse models of prostate cancer. Prostate cancer 2011, 2011:895238. [78] Sobel RE, Sadar MD: Cell lines used in prostate cancer research: a compendium of old and new lines--part 1. The Journal of urology 2005, 173:342- 59. [79] Sobel RE, Sadar MD: Cell lines used in prostate cancer research: a compendium of old and new lines--part 2. The Journal of urology 2005, 173:360- 72. [80] Powell WC, Cardiff RD, Cohen MB, Miller GJ, Roy-Burman P: Mouse strains for prostate tumorigenesis based on genes altered in human prostate cancer. Current drug targets 2003, 4:263-79. [81] Pienta KJ, Abate-Shen C, Agus DB, Attar RM, Chung LW, Greenberg NM, Hahn WC, Isaacs JT, Navone NM, Peehl DM, Simons JW, Solit DB, Soule HR, VanDyke TA, Weber MJ, Wu L, Vessella RL: The current state of preclinical prostate cancer animal models. The Prostate 2008, 68:629-39. [82] Irshad S, Abate-Shen C: Modeling prostate cancer in mice: something old, something new, something premalignant, something metastatic. Cancer metastasis reviews 2013, 32:109-22. [83] Hensley PJ, Kyprianou N: Modeling prostate cancer in mice: limitations and opportunities. Journal of andrology 2012, 33:133-44. [84] Sampson N, Neuwirt H, Puhr M, Klocker H, Eder IE: In vitro model systems to study androgen receptor signaling in prostate cancer. Endocrine-related cancer 2013, 20:R49-64. 113 [85] Ishikura N, Kawata H, Nishimoto A, Nakamura R, Ishii N, Aoki Y: Establishment and characterization of an androgen receptor-dependent, androgen-independent human prostate cancer cell line, LNCaP-CS10. The Prostate 2010, 70:457-66. [86] Culig Z, Hoffmann J, Erdel M, Eder IE, Hobisch A, Hittmair A, Bartsch G, Utermann G, Schneider MR, Parczyk K, Klocker H: Switch from antagonist to agonist of the androgen receptor bicalutamide is associated with prostate tumour progression in a new model system. British journal of cancer 1999, 81:242-51. [87] Sramkoski RM, Pretlow T, II, Giaconia J, Pretlow T, Schwartz S, Sy M-S, Marengo S, Rhim J, Zhang D, Jacobberger J: A new human prostate carcinoma cell line, 22Rv1. In Vitro CellDevBiol-Animal 1999, 35:403-9. [88] Tepper CG, Boucher DL, Ryan PE, Ma A-H, Xia L, Lee L-F, Pretlow TG, Kung H-J: Characterization of a Novel Androgen Receptor Mutation in a Relapsed CWR22 Prostate Cancer Xenograft and Cell Line. Cancer Research 2002, 62:6606-14. [89] Maroulakou IG, Anver M, Garrett L, Green JE: Prostate and mammary adenocarcinoma in transgenic mice carrying a rat C3(1) simian virus 40 large tumor antigen fusion gene. Proceedings of the National Academy of Sciences of the United States of America 1994, 91:11236-40. [90] Greenberg NM, DeMayo F, Finegold MJ, Medina D, Tilley WD, Aspinall JO, Cunha GR, Donjacour AA, Matusik RJ, Rosen JM: Prostate cancer in a transgenic mouse. Proceedings of the National Academy of Sciences of the United States of America 1995, 92:3439-43. [91] Wang S, Gao J, Lei Q, Rozengurt N, Pritchard C, Jiao J, Thomas GV, Li G, Roy-Burman P, Nelson PS, Liu X, Wu H: Prostate-specific deletion of the murine Pten tumor suppressor gene leads to metastatic prostate cancer. Cancer cell 2003, 4:209-21. [92] Wu X, Wu J, Huang J, Powell WC, Zhang J, Matusik RJ, Sangiorgi FO, Maxson RE, Sucov HM, Roy-Burman P: Generation of a prostate epithelial cell- specific Cre transgenic mouse model for tissue-specific gene ablation. Mechanisms of development 2001, 101:61-9. [93] Liao C-P, Zhong C, Saribekyan G, Bading J, Park R, Conti PS, Moats R, Berns A, Shi W, Zhou Z, Nikitin AY, Roy-Burman P: Mouse Models of Prostate Adenocarcinoma with the Capacity to Monitor Spontaneous Carcinogenesis by Bioluminescence or Fluorescence. Cancer Research 2007, 67:7525-33. 114 [94] D'Antonio JM, Vander Griend DJ, Antony L, Ndikuyeze G, Dalrymple SL, Koochekpour S, Isaacs JT: Loss of androgen receptor-dependent growth suppression by prostate cancer cells can occur independently from acquiring oncogenic addiction to androgen receptor signaling. PloS one 2010, 5:e11475. [95] Liao CP, Liang M, Cohen MB, Flesken-Nikitin A, Jeong JH, Nikitin AY, Roy- Burman P: Mouse prostate cancer cell lines established from primary and postcastration recurrent tumors. Hormones & cancer 2010, 1:44-54. [96] Bonkhoff H, Berges R: From pathogenesis to prevention of castration resistant prostate cancer. The Prostate 2009. [97] Brooke GN, Bevan CL: The role of androgen receptor mutations in prostate cancer progression. Curr Genomics 2009, 10:18-25. [98] Waltering KK, Helenius MA, Sahu B, Manni V, Linja MJ, Janne OA, Visakorpi T: Increased expression of androgen receptor sensitizes prostate cancer cells to low levels of androgens. Cancer Res 2009, 69:8141-9. [99] Chen Y, Clegg NJ, Scher HI: Anti-androgens and androgen-depleting therapies in prostate cancer: new agents for an established target. Lancet Oncol 2009, 10:981-91. [100] Abate-Shen C, Brown PH, Colburn NH, Gerner EW, Green JE, Lipkin M, Nelson WG, Threadgill D: The untapped potential of genetically engineered mouse models in chemoprevention research: opportunities and challenges. Cancer Prev Res (Phila Pa) 2008, 1:161-6. [101] Trotman LC, Niki M, Dotan ZA, Koutcher JA, Di Cristofano A, Xiao A, Khoo AS, Roy-Burman P, Greenberg NM, Van Dyke T, Cordon-Cardo C, Pandolfi PP: Pten dose dictates cancer progression in the prostate. PLoS Biol 2003, 1:E59. [102] Ma X, Ziel-van der Made AC, Autar B, van der Korput HA, Vermeij M, van Duijn P, Cleutjens KB, de Krijger R, Krimpenfort P, Berns A, van der Kwast TH, Trapman J: Targeted biallelic inactivation of Pten in the mouse prostate leads to prostate cancer accompanied by increased epithelial cell proliferation but not by reduced apoptosis. Cancer Res 2005, 65:5730-9. [103] Zhou Z, Flesken-Nikitin A, Corney DC, Wang W, Goodrich DW, Roy- Burman P, Nikitin AY: Synergy of p53 and Rb deficiency in a conditional mouse model for metastatic prostate cancer. Cancer Res 2006, 66:7889-98. [104] Bruxvoort KJ, Charbonneau HM, Giambernardi TA, Goolsby JC, Qian CN, Zylstra CR, Robinson DR, Roy-Burman P, Shaw AK, Buckner-Berghuis BD, 115 Sigler RE, Resau JH, Sullivan R, Bushman W, Williams BO: Inactivation of Apc in the mouse prostate causes prostate carcinoma. Cancer Res 2007, 67:2490-6. [105] van Bokhoven A, Varella-Garcia M, Korch C, Johannes WU, Smith EE, Miller HL, Nordeen SK, Miller GJ, Lucia MS: Molecular characterization of human prostate carcinoma cell lines. The Prostate 2003, 57:205-25. [106] Navone NM, Olive M, Ozen M, Davis R, Troncoso P, Tu SM, Johnston D, Pollack A, Pathak S, von Eschenbach AC, Logothetis CJ: Establishment of two human prostate cancer cell lines derived from a single bone metastasis. Clinical cancer research : an official journal of the American Association for Cancer Research 1997, 3:2493-500. [107] Chen ME, Lin SH, Chung LW, Sikes RA: Isolation and characterization of PAGE-1 and GAGE-7. New genes expressed in the LNCaP prostate cancer progression model that share homology with melanoma-associated antigens. The Journal of biological chemistry 1998, 273:17618-25. [108] Ishikura N, Kawata H, Nishimoto A, Nakamura R, Ishii N, Aoki Y: Establishment and characterization of an androgen receptor-dependent, androgen-independent human prostate cancer cell line, LNCaP-CS10. The Prostate 2009. [109] Foster BA, Gingrich JR, Kwon ED, Madias C, Greenberg NM: Characterization of prostatic epithelial cell lines derived from transgenic adenocarcinoma of the mouse prostate (TRAMP) model. Cancer Res 1997, 57:3325-30. [110] Jiao J, Wang S, Qiao R, Vivanco I, Watson PA, Sawyers CL, Wu H: Murine cell lines derived from Pten null prostate cancer show the critical role of PTEN in hormone refractory prostate cancer development. Cancer Res 2007, 67:6083-91. [111] Choi J, Burns AA, Williams RM, Zhou Z, Flesken-Nikitin A, Zipfel WR, Wiesner U, Nikitin AY: Core-shell silica nanoparticles as fluorescent labels for nanomedicine. J Biomed Opt 2007, 12:064007. [112] He W, Dai C, Li Y, Zeng G, Monga SP, Liu Y: Wnt/beta-catenin signaling promotes renal interstitial fibrosis. Journal of the American Society of Nephrology : JASN 2009, 20:765-76. [113] Leong KG, Wang BE, Johnson L, Gao WQ: Generation of a prostate from a single adult stem cell. Nature 2008, 456:804-8. 116 [114] Nagy A, Gertsenstein M, Vintersten K, Behringer R: Manipulating the Mouse Embryo. A Laboratory Manual, Third edn. . Cold Spring Harbor Laboratory Press 2003:389. [115] Leiblich A, Cross SS, Catto JW, Pesce G, Hamdy FC, Rehman I: Human prostate cancer cells express neuroendocrine cell markers PGP 9.5 and chromogranin A. The Prostate 2007, 67:1761-9. [116] Zhou Z, Flesken-Nikitin A, Nikitin AY: Prostate cancer associated with p53 and Rb deficiency arises from the stem/progenitor cell-enriched proximal region of prostatic ducts. Cancer Res 2007, 67:5683-90. [117] Isayeva T, Chanda D, Kallman L, Eltoum IE, Ponnazhagan S: Effects of sustained antiangiogenic therapy in multistage prostate cancer in TRAMP model. Cancer Res 2007, 67:5789-97. [118] Tran CP, Lin C, Yamashiro J, Reiter RE: Prostate stem cell antigen is a marker of late intermediate prostate epithelial cells. Mol Cancer Res 2002, 1:113- 21. [119] Hansel DE, Herawi M, Montgomery E, Epstein JI: Spindle cell lesions of the adult prostate. Mod Pathol 2007, 20:148-58. [120] Shappell SB, Thomas GV, Roberts RL, Herbert R, Ittmann MM, Rubin MA, Humphrey PA, Sundberg JP, Rozengurt N, Barrios R, Ward JM, Cardiff RD: Prostate pathology of genetically engineered mice: definitions and classification. The consensus report from the Bar Harbor meeting of the Mouse Models of Human Cancer Consortium Prostate Pathology Committee. Cancer Res 2004, 64:2270-305. [121] Yuan X, Li T, Wang H, Zhang T, Barua M, Borgesi RA, Bubley GJ, Lu ML, Balk SP: Androgen receptor remains critical for cell-cycle progression in androgen-independent CWR22 prostate cancer cells. Am J Pathol 2006, 169:682-96. [122] Agoulnik IU, Weigel NL: Androgen receptor action in hormone-dependent and recurrent prostate cancer. Journal of cellular biochemistry 2006, 99:362-72. [123] 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 Dev 2007, 21:2005-17. [124] Liao CP, Adisetiyo H, Liang M, Roy-Burman P: Cancer-associated fibroblasts enhance the gland-forming capability of prostate cancer stem cells. Cancer Res 2010, 70:7294-303. 117 [125] Tepper CG, Boucher DL, Ryan PE, Ma AH, Xia L, Lee LF, Pretlow TG, Kung HJ: Characterization of a novel androgen receptor mutation in a relapsed CWR22 prostate cancer xenograft and cell line. Cancer Res 2002, 62:6606-14. [126] Tsai HC, Boucher DL, Martinez A, Tepper CG, Kung HJ: Modeling truncated AR expression in a natural androgen responsive environment and identification of RHOB as a direct transcriptional target. PloS one 2012, 7:e49887. [127] Libertini SJ, Tepper CG, Rodriguez V, Asmuth DM, Kung HJ, Mudryj M: Evidence for calpain-mediated androgen receptor cleavage as a mechanism for androgen independence. Cancer Res 2007, 67:9001-5. [128] Shen HC, Balk SP: Development of androgen receptor antagonists with promising activity in castration-resistant prostate cancer. Cancer cell 2009, 15:461-3. [129] Ford OH, 3rd, Gregory CW, Kim D, Smitherman AB, Mohler JL: Androgen receptor gene amplification and protein expression in recurrent prostate cancer. The Journal of urology 2003, 170:1817-21. [130] Palmberg C, Koivisto P, Kakkola L, Tammela TL, Kallioniemi OP, Visakorpi T: Androgen receptor gene amplification at primary progression predicts response to combined androgen blockade as second line therapy for advanced prostate cancer. The Journal of urology 2000, 164:1992-5. [131] Quayle SN, Mawji NR, Wang J, Sadar MD: Androgen receptor decoy molecules block the growth of prostate cancer. Proceedings of the National Academy of Sciences of the United States of America 2007, 104:1331-6. [132] Myung JK, Banuelos CA, Fernandez JG, Mawji NR, Wang J, Tien AH, Yang YC, Tavakoli I, Haile S, Watt K, McEwan IJ, Plymate S, Andersen RJ, Sadar MD: An androgen receptor N-terminal domain antagonist for treating prostate cancer. J Clin Invest 2013, 123:2948-60. [133] Xu T, Xu Y, Liao C-P, Lau R, Goldkorn A: Reprogramming murine telomerase rapidly inhibits the growth of mouse cancer cells in vitro and in vivo. Molecular cancer therapeutics 2010, 9:438-49. [134] Jeong JH, Bhatia A, Toth Z, Oh S, Inn K-S, Liao C-P, Roy-Burman P, Melamed J, Coetzee GA, Jung JU: TPL2/COT/MAP3K8 (TPL2) activation promotes androgen depletion-independent (ADI) prostate cancer growth. PloS one 2011, 6:e16205. 118 [135] Ellis L, Lehet K, Ramakrishnan S, Adelaiye R, Pili R: Development of a castrate resistant transplant tumor model of prostate cancer. The Prostate 2012, 72:587-91. [136] Fedr R, Pernicová Z, Slabáková E, Straková N, Bouchal J, Grepl M, Kozubík A, Sou ček K: Automatic cell cloning assay for determining the clonogenic capacity of cancer and cancer stem ‐like cells. Cytometry Part A 2013. [137] Aycock-Williams AN, Pham LK, Liang M, Adisetiyo HA, Geary LA, Cohen MB, Casebolt DB, Roy-Burman P: Effects of fenbendazole and vitamin E succinate on the growth and survival of prostate cancer cells. [138] Kraus S, Gioeli D, Vomastek T, Gordon V, Weber MJ: Receptor for activated C kinase 1 (RACK1) and Src regulate the tyrosine phosphorylation and function of the androgen receptor. Cancer Res 2006, 66:11047-54. [139] Guo Z, Dai B, Jiang T, Xu K, Xie Y, Kim O, Nesheiwat I, Kong X, Melamed J, Handratta VD, Njar VC, Brodie AM, Yu LR, Veenstra TD, Chen H, Qiu Y: Regulation of androgen receptor activity by tyrosine phosphorylation. Cancer cell 2006, 10:309-19. [140] Gioeli D, Ficarro SB, Kwiek JJ, Aaronson D, Hancock M, Catling AD, White FM, Christian RE, Settlage RE, Shabanowitz J, Hunt DF, Weber MJ: Androgen receptor phosphorylation. Regulation and identification of the phosphorylation sites. The Journal of biological chemistry 2002, 277:29304-14. [141] Kinoshita H, Shi Y, Sandefur C, Meisner LF, Chang C, Choon A, Reznikoff CR, Bova GS, Friedl A, Jarrard DF: Methylation of the androgen receptor minimal promoter silences transcription in human prostate cancer. Cancer Res 2000, 60:3623-30. [142] Ahrens-Fath I, Politz O, Geserick C, Haendler B: Androgen receptor function is modulated by the tissue-specific AR45 variant. The FEBS journal 2005, 272:74-84.

Abstract (if available)

Abstract Despite immense research progress made in recent years, the recurrence of castration-resistant prostate cancer (CRPC) still remains poorly understood. Androgen receptor (AR), an essential player during the differentiation and maintenance of normal prostate as well as the initiation of prostate cancer, has been widely accepted to be the "Partner-in-Crime" even in the stage of CRPC, mediating and being regulated by numerous factors and signaling pathways. Therefore, to decipher the role of AR and its regulation in contributing to the occurrence of CRPC has become a priority topic in the prostate cancer (PCa) research. ❧ This dissertation describes establishment of a study system composed of murine PCa cell lines derived from tumors of the conditional Pten deletion mouse model of prostate adenocarcinoma obtained at two distinct phases of the disease, namely androgen-dependent (AD) growth phase and the castration-resistant (CR) growth phase. We have characterized five mouse PCa cell lines - E2/E4 and E8 from two separate AD tumors and cE1/cE2 from one CR tumor. All the cell lines manifest biallelic deletion of the Pten gene, corresponding to epithelial origin since the Cre/LoxP system used in the modeling is targeted to prostate epithelium. Analyses of various molecular expressions as well as morphology of these cells suggest a degree of epithelial-mesenchymal transition (EMT) in E2/E4 but not much in E8, which possesses increased epithelial phenotype. The cell lines from the CRPC tumor generally display epithelial characteristics and demonstrate significantly better growth capacity as compared to E-series in the absence of androgen, although sensitivity to supplemented androgen is retained. All the cell lines established are able to induce in vivo tumor growth in immunodeficient mice. However, while E8 and cE1/cE2 give rise to adenocarcinomas, E2 and E4 yields tumors that are akin to sarcomatoid carcinoma expressing protein makers of both epithelial and stromal cells. In this regard, there is a fairly good correlation between the in vitro and in vivo properties. ❧ Among the multiple mechanisms by which AR activity may be aberrantly regulated, we have focused our studies of the cell lines to determine the status of the androgen receptor splice variants (AR-Vs). We have identified a set of novel mouse AR-Vs (termed as mAR-Vabc) in these homologous mouse PCa cell systems. Their expression is present in all five cell lines and is generally up-regulated in response to androgen ablation. The relative proportion of these mAR-Vs varies with respect to the stage of the disease from which the cell lines originated. In cE-series, mAR-Vb and mAR-Vc are more abundant than that in E-series. However, mAR-Va displays an opposite profile of expression. The mRNA of these mAR-Vs is detected in the normal prostates

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

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

PDF

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

PDF

Studies of murine prostate cancer stem / progenitor cells

PDF

Bimodal effects of bone morphogenetic proteins in prostate cancer

PDF

Targeting molecular signals involved in the development of castration resistant prostate cancer

PDF

Identification of novel androgen receptor target genes in prostate cancer

PDF

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

PDF

PTEN deletion induced tumor initiating cells: Strategies to accelerate the disease progression of liver cancer

PDF

Functional analyses of androgen receptor structure pertaining to prostate cancer

PDF

Monoamine oxidase A inhibitors and androgen receptor antagonists regulate mitochondrial function in prostate cancer cells

PDF

The effect of tumor-mediated immune suppression on prostate cancer immunotherapy

PDF

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

PDF

Co-expression of monoamine oxidase A and prostate cancer stem cell markers in Pten knockout mice

PDF

Insights from mouse models on the origin of ovarian cancer and its predisposing mechanisms

PDF

Functional characterization of a prostate cancer risk region

PDF

MAO a deficient mice exhibit an altered immune system in the brain and prostate

PDF

RNA methylation in cancer plasticity and drug resistance

PDF

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

PDF

Role of a novel transmembrane protein, TMEM56, in tumorigenic growth of human PC3 prostate cancer cell line

Asset Metadata

Creator Liang, Mengmeng (author)

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

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Pathobiology

Publication Date 08/22/2013

Defense Date 07/24/2013

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

Tag androgen receptor,androgen receptor splice variants,androgen-dependent and castration-resistant prostate cancer,conditional Pten-deletion mouse model of prostate cancer,mouse prostate cancer cell lines,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Roy-Burman, Pradip (committee chair), Chuong, Cheng-Ming (committee member), Maxson, Robert E., Jr. (committee member)

Creator Email liangmm.usc@gmail.com,mengmengxy@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-323393

Unique identifier UC11295024

Identifier etd-LiangMengm-2018.pdf (filename),usctheses-c3-323393 (legacy record id)

Legacy Identifier etd-LiangMengm-2018.pdf

Dmrecord 323393

Document Type Dissertation

Format application/pdf (imt)

Rights Liang, Mengmeng

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

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