Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Identification and characterization of genes associated with the hormonal progression of prostate cancer Quayle, Steven Norman 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2006-200711.pdf [ 9.47MB ]
Metadata
JSON: 831-1.0092958.json
JSON-LD: 831-1.0092958-ld.json
RDF/XML (Pretty): 831-1.0092958-rdf.xml
RDF/JSON: 831-1.0092958-rdf.json
Turtle: 831-1.0092958-turtle.txt
N-Triples: 831-1.0092958-rdf-ntriples.txt
Original Record: 831-1.0092958-source.json
Full Text
831-1.0092958-fulltext.txt
Citation
831-1.0092958.ris

Full Text

IDENTIFICATION AND CHARACTERIZATION OF GENES ASSOCIATED WITH THE HORMONAL PROGRESSION OF PROSTATE CANCER by STEVEN N O R M A N Q U A Y L E B.Sc, The University of British Columbia, 2000 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (PATHOLOGY AND L A B O R A T O R Y MEDICINE) THE UNIVERSITY OF BRITISH COLUMBIA August 2006 © Steven Norman Quayle, 2006 ABSTRACT Prostate cancer is the most frequently diagnosed cancer as well as the second leading cause of cancer deaths among American men. Androgen deprivation therapy for patients with advanced prostate cancer inevitably fails as the disease progresses to the terminal, hormone-refractory stage. At the present time no effective therapies can be offered to patients in this stage of the disease. Currently, the molecular alterations that contribute to the development of hormone-refractory disease are poorly understood. Here, suppression subtractive hybridization was performed to identify genes of interest in a model of progression of prostate cancer to androgen independence. Subtractive hybridization identified a number of novel transcripts that had previously not been described. From this work it was hypothesized that (1) the novel transcripts identified may prove informative in the clinical management of prostate cancer, and (2) that the known genes identified may function in the progression of prostate cancer. One of these novel transcripts was shown to encode a truncated splice variant of the transmembrane protein T M E F F 2 . T M E F F 2 is specifically expressed in the prostate and the brain, and other groups have developed antibodies targeting T M E F F 2 for the treatment of metastatic prostate cancer. The isoform described here, T M E F F 2 - S , encodes a protein secreted by prostate cancer cells. Being a prostate-specific, secreted protein, T M E F F 2 - S may prove informative in predicting patient outcome. Additionally, the secretion of T M E F F 2 - S may influence the effectiveness of immunotherapy approaches targeting T M E F F 2 . Subtractive hybridization, together with other high-throughput screening techniques, also led to the examination of the potential functions of the 14-3-3 family of ii proteins in the progression of prostate cancer. These experiments demonstrated that 14-3-3 sigma specifically increased the activity of the androgen receptor in the absence of androgens. Thus, this finding may have implications for patients undergoing androgen deprivation therapy. Further studies will be required to determine if 14-3-3 sigma is active in a subset of patients with hormone refractory prostate cancer. If so, its role in activating the androgen receptor may potentially be targeted for the development of novel therapeutics for the treatment of hormone-refractory disease. in TABLE OF CONTENTS Page No. Abstract ii Table of Contents iv List of Tables vii List of Figures viii List of Abbreviations x Acknowledgements xii Co-Authorship Statement xiv 1 Introduction 1 1.1 Natural history of prostate cancer 1 1.1.1 Diagnosis of prostate cancer 1 1.1.2 Treatment of prostate cancer 4 1.1.3 Recurrence of prostate cancer 6 1.2 Androgen receptor 7 1.3 Mechanisms of androgen independent growth 11 1.4 Models for studying androgen-independent disease 16 1.5 Differential gene expression 19 1.5.1 Gene expression changes in prostate cancer 19 1.5.2 Techniques for transcriptional profiling 21 1.6 Suppression subtractive hybridization 23 1.7 Summary and research objectives 26 1.7.1 Hypothesis 27 1.7.2 Specific aims 27 1.7.3 Thesis overview 28 1.8 References 30 2 Novel expressed sequences identified in a model of androgen-independent prostate cancer 50 2.1 Introduction 50 2.2 Materials and methods 52 2.2.1 Cell culture 52 2.2.2 LNCaP Hollow fibre model 52 2.2.3 Suppression subtractive hybridization 54 2.2.4 RT-PCR 55 IV 2.3 Results 57 2.3.1 Subtractive hybridization 57 2.3.2 Novel transcripts were represented in the subtracted cDNA libraries 57 2.3.3 Characterization of novel clone sequences 58 2.3.4 Analysis of SAGE tags in the novel clones 65 2.3.5 The novel subtracted clones are expressed by a variety of cell types 67 2.3.6 Two of the novel clones are part of larger untranslated regions 69 2.4 Discussion 70 2.5 References 77 3 A truncated isoform of TMEFF2 encodes a secreted protein in prostate cancer cells 85 3.1 Introduction 85 3.2 Materials and methods 85 3.2.1 Cell culture and transfection 85 3.2.2 RT-PCR 86 3.2.3 Plasmids 87 3.2.4 Western blot analysis 87 3.2.4 Fluorescent microscopy 88 3.3 Results 89 3.3.1 A novel splice variant is transcribed from the TMEFF2 locus 89 3.3.2 TMEFF2-S was expressed in normal human tissues and produced a truncated protein 91 3.3.3 The truncated isoform of TMEFF2 produced a secreted protein 94 3.4 Discussion 97 3.5 References 100 4 14-3-3 sigma increases the transcriptional activity of the androgen receptor in the absence of androgens 103 4.1 Introduction 103 4.2 Materials and methods 105 4.2.1 Cell culture '. 105 4.2.2 Plasmids 105 4.2.3 Transfection and luciferase assay 105 4.2.4 AR siRNA 106 4.2.5 Western blot analysis 107 4.2.6 Co-immunoprecipitation 108 4.3 Results 108 4.3.1 14-3-3 sigma specifically enhanced AR transcriptional activity 108 4.3.2 14-3-3 sigma enhanced AR activity in a dose-dependent manner in the absence of androgens 112 4.3.3 14-3-3 sigma increased the sensitivity of AR to low levels of androgens 112 V 4.3.4 Induction of PSA reporters by 14-3-3 sigma was dependent on A R 112 4.3.5 14-3-3 sigma did not alter AR protein levels 117 4.4 Discussion 119 4.5 References 124 5 Concluding Chapter 130 5.1 Expression profiling of novel clones 130 5.1.1 Custom cDNA microarray 130 5.1.2 Low-density TaqMan® array 132 5.2 Summary and future directions 133 5.2.1 Novel transcripts 134 5.2.2 TMEFF2-S 137 5.2.3 14-3-3 sigma 139 5.3 Conclusions 142 5.4 References 144 6 Appendix 150 6.1 Animal Care Certificate 150 6.2 Biohazard Approval Certificate 151 vi LIST OF TABLES Page No. Table 2.1 Primers used for RT-PCR 56 Table 2.2 Summary of sequenced clones 59 Table 2.3 Sequence characteristics of unannotated clones 62 Table 2.4 Sequence characteristics of clones matching previously identified ESTs 63 Table 2.5 Sequence characteristics of clones matching previously annotated cDNAs... 64 V l l LIST OF FIGURES Page No. Figure 1.1 Mechanism of ligand-dependent activation of the androgen receptor 9 Figure 1.2 Proposed mechanisms of androgen-independent growth and survival of prostate cancer 12 Figure 1.3 Schematic diagram of suppression subtractive hybridization protocol 25 Figure 2.1 Distribution of sequence matches in the suppression subtractive hybridization clone set 60 Figure 2.2 The novel subtracted clones were expressed in a variety of human tissues 68 Figure 2.3 Two of the novel clones were part of transcripts containing a neighbouring SAGE tag 71 Figure 3.1 A novel EST is part of a larger transcript derived from the TMEFF2 locus 90 Figure 3.2 The novel splice variant of TMEFF2 is expressed in normal tissues 92 Figure 3.3 The novel splice variant of TMEFF2 produced a truncated protein 93 Figure 3.4 The truncated variant of TMEFF2 is not localized in the plasma membrane but is secreted by transfected cells 95 Figure 4.1 The transcriptional activity of the AR was activated specifically by 14-3-3 sigma 110 Figure 4.2 Increasing amounts of 14-3-3 sigma increased the amount of A R activity in the absence of androgens 113 Figure 4.3 14-3-3 sigma sensitized the AR to low levels of androgen 114 V l l l f Figure 4.4 Activation of the ARPv3-tk-luciferase reporter construct by 14-3-3 sigma was dependent on the AR 115 Figure 4.5 14-3-3 sigma did not interact with or alter the amount of A R protein in the cells 118 IX LIST OF ABBREVIATIONS A D T Androgen deprivation therapy A R Androgen receptor A R E Androgen response element bp base pairs DHT Dihydrotestosterone DRE Digital rectal exam E 2 Estradiol EST Expressed sequence tag FBS Fetal bovine serum GAPDH Glyceraldehyde-3-phosphate dehydrogenase HRPC Hormone-refractory prostate cancer Hsp Heat shock protein IL-6 Interleukin-6 L B D Ligand binding domain LNCaP Lymph node-derived carcinoma of the prostate M A P K Mitogen-activated protein kinase NCoR Nuclear receptor co-repressor ORF Open reading frame PAP Prostatic acid phosphatase PSA Prostate-specific antigen R A C E Rapid amplification of cDNA ends RT-PCR Reverse-transcriptase polymerase chain reaction SAGE Serial analysis of gene expression SRC-1 Steroid receptor co-activator 1 SSH Suppression subtractive hybridization tk Thymidine kinase TMEFF2 Transmembrane protein with EGF-like and two follistatin-like domains 2 UTR Untranslated region XI ACKNOWLEDGEMENTS A number of people at the BC Cancer Research Centre have helped me with the experiments described in this thesis. Nasrin Mawji and Dr Rob Snoek provided me with seemingly endless advice and assistance in the design and execution of a number of experiments. Heidi Hare provided me with excellent technical assistance. Dorothy Hwang and Jean Wang maintained the LNCaP hollow fibre model and monitored the experiments. Dr Allen Delaney and Scott Zuyderduyn provided bioinformatics support and performed comparisons to various SAGE libraries. And Dr Steven Jones provided invaluable assistance and knowledge of the world of bioinformatics. I was also fortunate to have the help of three undergraduate students - Stephanie Warner, Sherman Tung, and Michelle Joe. A number of people have also provided helpful discussions and advice on everything from experimental design to converting computer files. In no particular order, they are: Tammy Romanuik, Dr Katie Meehan, Dr Brad Hoffman, Elizabeth Hajen, Amara Garcia, Brad Coe, Angela Beckett, Dr James Goldie, and Dr Nicholas Bruchovsky. Other members of the Sadar laboratory have also provided helpful suggestions and comments: Dr Viola Akopian, Dr Tom Kim, Natasha Mendelev, Hans Adomat, Natalie Blaszczyk, Dr Richard Sobel, Gang Wang, Helen Chiu, Teresa Tarn, and Dr Barbara Comuzzi. Dr Wan Lam allowed me to use his microarray scanner and provided much encouragement. A number of administrative assistants have also helped me over the years, including Susan Chou, Penny Woo, Louise Clarke, and Mary Flores. I was financially supported by a number of agencies. Scholarships were received from the Canadian Institutes of Health Research, the Michael Smith Foundation for Health Research, the University of British Columbia, and the Department of Pathology and Laboratory Medicine. Travel awards were provided by the Society for Basic Urologic Research, Keystone Symposia, and Albert B. and Mary Steiner. I am of course indebted to my supervisor, Dr Marianne Sadar. She provided me with a rich research environment and financial support for both my stipend and my research (through a Centre of Excellence for Prostate Cancer Research grant from Health Canada). And she was always generous with her time while we discussed experiments, conferences, and holidays. On a personal note, I have to acknowledge my girlfriend, Lynn Mar, for all that she has given me during my training: endless advice on cloning, critical analysis of my experiments, and constant support when we were exhausted and it seemed like we would never be finished. And finally, this thesis would not have been possible without the unlimited support and encouragement of my family. Thank you all. X l l l CO-AUTHORSHIP STATEMENT The experiments described in this thesis were conceived, designed, conducted, and analyzed by myself, Steven Quayle, and Dr Marianne Sadar. However, a number of people contributed to the work described in Chapter 2. Dr Steven Jones and Dr Marco Marra helped conceive and design the experiments. The experiments were performed by Steven Quayle, Heidi Hare, Dr Allen Delaney, Dorothy Hwang, and Jacqueline Schein. Steven Quayle analyzed the data. Finally, Steven Quayle prepared the manuscripts and this thesis. xiv 1 INTRODUCTION 1.1 Natural history of prostate cancer The prostate is a small secretory organ surrounding the urethra at the base of the bladder. Its normal function is to secrete factors into the semen to increase survival of the sperm. The prostate is divided into three histologic regions: the transition zone, the central zone, and the peripheral zone. The transition zone is the smallest (5% of prostate volume), followed by the central zone (25% of prostate volume), with the peripheral zone being the largest (70% of prostate volume) [1,2]. The entire prostate is surrounded by a "capsule" of fibromuscular stromal cells [3]. Androgens regulate the differentiation, proliferation, secretion of proteins, as well as various biosynthetic pathways in the normal prostate. In fact, the prostate is an androgen dependent tissue and the removal of androgens results in apoptotic cell death [4,5]. It is this dependence on androgens that forms the basis for the treatment of advanced prostate cancer with androgen deprivation therapy (ADT), which will be discussed in more detail below. Androgens exert their effects on the prostate through direct binding to the androgen receptor (AR), a transcription factor that is a member of the nuclear receptor superfamily [6]. 1.1.1 Diagnosis of prostate cancer Carcinoma of the prostate is the most frequently diagnosed cancer as well as the second leading cause of cancer death among American men [7]. The majority of prostate cancers are located in the peripheral zone [2,8]. Since the peripheral zone lies adjacent to the rectum, it is possible to detect abnormal prostate morphology, indicative of 1 hyperplastic growth, through a digital rectal exam (DRE). The DRE is most useful in diagnosing peripheral zone cancers as cancerous foci located in the transition and central zones are less likely to be detected by this technique. The DRE is one of three tests used to diagnose prostate cancer. The second means of diagnosis relies on the measurement of serum levels of the prostate-specific antigen (PSA) protein. PSA is a 34 kDa serine protease normally expressed and secreted into the semen by the luminal epithelial cells of the prostate [9,10]. It was first purified in 1979 and was later shown to be important for the liquefaction of ejaculated semen [9-11]. However, development of a neoplastic growth in the prostate, or invasion of a prostatic carcinoma through the capsule, can result in elevated circulating levels of PSA in the bloodstream. Serum PSA levels can be accurately measured in these patients through the use of various techniques, including radioimmunoassay. Numerous studies indicate PSA expression is generally restricted to the prostate and that the level of serum PSA correlates well with the palpable stages of prostate cancer, leading to its use as a diagnostic marker in screening for prostate cancer [12-14]. The final, and most definitive, method of diagnosing prostate cancer remains the pathologic identification of invasive carcinoma after biopsy. Despite the value of PSA as a marker of prostate cancer there is currently a great deal of controversy in the field regarding its usefulness for screening programs. There is a growing body of literature suggesting that prostate cancer is present in men with no clinical symptoms of the disease and normal diagnostic parameters, including PSA levels. A classic study by Sakr et al. [15] performed pathological analyses on the prostates of a 2 large number of men who died of causes other than prostate cancer and who had no clinical symptoms of prostate cancer. This study found that up to 80% of men from 70 -80 years of age had precancerous lesions in their prostates [15]. Additionally, the Prostate Cancer Prevention Trial examined a large population of men and found that amongst men with "normal" PSA and DRE tests, approximately 15% could be diagnosed with prostate cancer based on a biopsy [16]. Results such as these have led some to conclude that prostate cancer is currently being underdiagnosed. Over the last decade there has been a steady decrease in the number of patients diagnosed with "high-risk" disease and a concomitant increase in diagnosis of "low-risk" disease, suggesting that prostate cancer is now being detected at earlier stages [17]. It has been recommended that lowering the PSA threshold may aid in identifying more cases of prostate cancer in younger men [18]. However, as cutoff values for "abnormal" PSA are lowered the sensitivity and specificity of the PSA test will also decrease. This means that more men will be referred for prostate biopsies, but a lower percentage of those men will be diagnosed with prostate cancer. In fact, a recent study has called into question the diagnostic value of serum PSA measurements below 22 ng/mL [19]. Further complicating the issue is the fluctuations in PSA level that are known to occur with time, and particularly with disturbances of the prostate [13,20]. New markers are thus needed to classify patients into high- and low-risk categories. Regardless of these controversies, prostate cancer is diagnosed in a large number of men each year [7]. The question then is how they should be treated. 3 1.1.2 Treatment of prostate cancer Upon diagnosis of prostate cancer by biopsy, these patients are typically offered a number of treatment options (for recent, in depth reviews, see [21-24]). The first, and most common, is radical prostatectomy where the entire prostate and the seminal vesicles are removed by surgery. The second is external beam radiation therapy targeting the prostate. More recent advances have led to the development of brachytherapy for prostate cancer where radioactive pellets are implanted directly into the prostate where they steadily emit a low level of radiation. Each of these interventions is undertaken with curative intent. However, given the generally slow progression of prostate cancer [25,26], it is not always advantageous to attempt a full cure. For example, active surveillance, or watchful waiting, may be offered to either those at lowest risk of disease recurrence or those who would suffer excessive morbidity from the other treatment options. The intent of this approach is obviously not curative, but simply to maximize quality of life while still being prepared to actively treat the patient should their situation worsen. Finally, androgen deprivation may be offered prior to or in conjunction with all of the above treatments. Two studies have recently looked at long term outcomes in patients whose cancers were not treated initially but were instead monitored by watchful waiting [25,26]. Both studies suggest that aggressive treatment is appropriate for patients whose cancers are poorly differentiated at diagnosis, but not for patients with low-grade disease. Encouragingly, a phase III study comparing radical prostatectomy to watchful waiting in patients with intermediate-risk disease found that surgery did result in decreased disease recurrence and decreased likelihood of death due to prostate cancer [27]. So while 4 patients diagnosed with more aggressive disease have a poorer outcome on watchful waiting, active treatment is able to extend their lives. Unfortunately, it is not always easy to distinguish those patients who are likely to die from their disease, and thus require treatment, from those whose disease is likely to remain clinically insignificant. The result is that many men with intermediate-grade disease may be being over-treated [21,23]. Stamey et al. [28] found that the percentage of high grade cancer and the total cancer volume at diagnosis were the best predictors of biochemical relapse after radical prostatectomy. Other studies have tried to obtain better prognostic information through the examination of PSA kinetics in these patients. For example, a rate of rise of PSA, or PSA velocity, at diagnosis of greater than 2.0 ng/mL/year correlated with an increased risk of dying from prostate cancer after radical prostatectomy [29]. However, the classical indicators of progression - PSA level, grade, and clinical tumour stage at diagnosis - were also predictive of increased risk of death in this study [29]. New biomarkers capable of distinguishing clinically significant from clinically indolent disease need to be identified to aid in making these treatment decisions. While the value of PSA as a diagnostic tool has been questioned, monitoring PSA levels post-diagnosis is an informative method of following the patient's response to therapy. Initial curative therapy typically results in an undetectable level of PSA, but the appearance of increased serum PSA after therapy, referred to as biochemical failure, is associated with the development of distant metastases and an overall poor prognosis [13,30,31]. In fact, in men treated by radical prostatectomy alone, Pound et al. [32] found that disease recurrence was only observed in men demonstrating biochemical 5 relapse. In this study group of 1,997 men, 15% exhibited PSA recurrence in a median time of 2.3 years post-surgery [32]. After biochemical failure, 34% of patients developed clinical metastases within a median time of 5 years [32]. Thus, biochemical relapse is a signal to begin more active screening for metastases. From the appearance of clinical metastases, the median time to death due to prostate cancer was 5 years. However, 82% of the study group were metastasis free 15 years after surgery [32]. 1.1.3 Recurrence of prostate cancer Recurrence of the disease at distant metastatic sites necessitates a systemic therapy; A D T is the only therapeutic approach shown to result in regression of the disease. Chemical, via luteinizing hormone releasing hormone agonists such as goserelin, or surgical castration, via orchiectomy, results in decreased serum testosterone levels, decreased serum PSA levels, and regression of the tumour [4,31,33,34]. Additionally, non-steroidal antiandrogens, such as bicalutamide, can be used to block the activity of the A R directly, but these therapies are costly and it is not yet clear if bicalutamide alone increases overall survival [35,36]. Finally, these therapies have been tested together to achieve greater inhibition of AR activity, but again this treatment did not lead to a significant increase in survival [34,35,37,38]. Unfortunately, A D T is not curative, and even with therapy the median survival of these patients is only about 2.5 years [37]. The optimal timing to begin ADT is still under debate (for recent reviews, see [24,35]). Often, ADT would be initiated at the point of biochemical recurrence after initial curative therapy or in cases where metastatic disease is already present at the time 6 of diagnosis. Increasingly, though, ADT is being offered soon after primary curative therapy (i.e. surgery), and before biochemical relapse. Finally, phase III trials are now underway to assess the effectiveness of intermittent androgen deprivation in delaying the eventual development of hormone refractory disease (reviewed by [39]). Regardless of when A D T is started during disease progression, the development of hormone-refractory prostate cancer (HRPC) leaves patients with few effective treatment options. Hormone-refractory, or androgen-independent, prostate cancer is that stage of disease after the failure of ADT, usually characterized by a rising serum PSA while still on therapy. The prognosis for patients at this stage of the disease is poor; the median survival is approximately 16 months [40-42]. In 2004, two studies showed an increase in overall survival among patients with HRPC who were treated with the chemotherapeutic drug docetaxel [41,42]. These were the first trials to demonstrate a therapy could increase survival in this group of patients, even though it was only a two month increase. There is currently a significant lack of understanding of the molecular mechanisms leading to the development of hormone-refractory prostate cancer. Greater comprehension of these mechanisms may aid in the creation of novel therapies for these patients. Additionally, new biomarkers are needed to identify those patients who may benefit from aggressive therapy and those who should be offered palliative approaches to increase their quality of life. 1.2 Androgen receptor As discussed above, the prostate is an androgen dependent organ and thus requires androgens for normal growth and differentiation. The effects of androgens are 7 mediated through binding of the hormone to its receptor, the AR. Unliganded A R is predominantly found in the cytoplasm, but upon binding ligand, such as dihydrotestosterone (DHT) or testosterone, the AR forms homodimers and translocates to the nucleus where it acts as a transcription factor for androgen regulated genes (Fig. 1.1) [43,44]. Nuclear translocation begins within 5-10 minutes after treatment of the cells with androgens, and is nearly complete within 60 minutes of treatment [44-46]. Once in the nucleus the A R binds androgen response elements (ARE) in the promoter and enhancer regions of its target genes and regulates transcriptional activity in association with co-activator proteins, such as steroid receptor co-activator 1 (SRC-1) [47,48], and co-repressor proteins, such as nuclear receptor co-repressor (NCoR) [49,50]. Examples of well-characterized androgen-regulated genes include PSA and ornithine decarboxylase [51,52]. The primary structure of the AR consists of four functional domains: an N-tefminal transactivating domain, a DNA-binding domain, a hinge region, and a C-terminal ligand binding domain (LBD; reviewed in [53]). The amino (N)-terminal region contains the activating function (AF)-l region (amino acid residues 110-379) that is primarily responsible for the transcription activating activity of the AR. This region also shows the least amount of homology across the nuclear receptor family. The DNA-binding domain contains two zinc-finger motifs (residues 559-614) that determine the specificity of interaction of the receptor with AREs and are also involved in the protein-protein interaction with the second A R molecule. Translocation of the A R to the nucleus is predominantly mediated by a ligand-dependent classical nuclear localization signal located within the DNA binding domain and hinge region of the AR (residues 617-633) 8 Figure 1.1- Mechanism of ligand-dependent activation of the androgen receptor. In the absence of ligand A R is bound to chaperone proteins in the cytoplasm. Binding of ligands, such as DHT, leads to dissociation from its chaperone complex, phosphorylation, dimerization, and translocation to the nucleus. There it regulates the expression of target genes in conjunction with co-activators such as SRC-1. See text for details. AR, androgen receptor; DHT, dihydrotestosterone; hsp, heat shock protein; SRC-1, steroid receptor coactivator-1; RNApol II, RNA polymerase complex. Adapted from [54]. 9 [55,56]. The carboxyl (C)-terminal domain, extending from residues 676-919, is the region responsible for binding ligand. While ligand binding is predominantly mediated by a large hydrophobic cavity known as the ligand binding pocket (approximately residues 735-787 and 855-865), deletion of small regions of the ligand binding domain tends to result in greatly decreased ligand binding efficiency [57]. Steroid receptors that are not bound to ligand do not exist as free cytosolic monomers. Instead, they are found in large heteromeric complexes with heat shock proteins (Hsp) and other members of the cellular chaperone machinery (reviewed in [58]). Each A R molecule is bound to Hsp70 and Hsp40, which results in the recruitment of Hsp90 to the complex. Various other cochaperones are required for the formation of this complex, but these three factors are obligately associated with unliganded steroid receptors. Complexing with heat shock proteins may increase the stability of the unliganded receptor's conformation so it can bind ligand. Alternatively, these chaperones may act to inhibit indiscriminate activation of the unliganded receptor by decreasing the efficiency of DNA binding and dimerization. Binding of agonists, such as DHT, by the receptor results in a conformational change in the AR, followed by dissociation of the multiprotein complex and trafficking of the activated A R through the cytoplasm and into the nucleus [59-61]. On the other hand, binding of antagonists, such as bicalutamide, prevents dissociation of the chaperone proteins and may inhibit the initial conformational change, but still induces nuclear import of the receptor [60,62,63]. Once A R has been activated and enters the nucleus it binds to AREs present in the enhancer regions of androgen-regulated genes. The AR binds as a homodimer to the consensus inverted repeat ARE, GGTACAnnnTGTTCT [64]. While numerous 10 androgen-regulated genes have been found, AR binding to AREs in the PSA promoter has been the most extensively studied [48,50,65]. Expression of the PSA gene is known to be androgen-regulated, and thus, monitoring PSA transcript levels is a simple method of assaying for transcriptional activation of the AR, in addition to its role in monitoring therapeutic response in prostate cancer patients. 1.3 Mechanisms of androgen independent growth A growing body of evidence indicates the AR is transcriptionally active in androgen-independent prostate cancer cells. Androgen-independent disease is marked by increased expression of the AR and re-expression of AR-regulated genes, such as PSA, after recurrence of an androgen-deprived tumour [66,67]. In fact, the work of Chen et al. [68] suggested that the AR was the only gene consistently upregulated amongst several models of androgen independent disease, and increased expression of AR was also seen in clinical samples of HRPC [67,69]. Several models have been proposed to explain how normally androgen-dependent prostate cells are able to proliferate in the absence of androgens (Fig. 1.2) [70,71]. One such model postulates that the A R is activated in a ligand-independent manner through cross-talk with other signal transduction pathways (Fig. 1.2A). Evidence has been presented that non-androgenic compounds and growth factors, such as forskolin, insulin-like growth factor-I and interleukin (IL)-6, are able to induce nuclear import and the transcription activation function of AR even in the absence of androgens [72-76]. This effect may be dependent on phosphorylation of additional sites within the A R or in receptor-associated proteins such as the co-activator SRC-1 [74,76,77]. Phosphorylation 11 Figure 1.2 - Proposed mechanisms of androgen-independent growth and survival of prostate cancer. (A) Signalling events downstream of a GF receptor result in transcriptional activation of the AR in the absence of ligand. (B) Mutation or amplification of the A R enable the receptor to become activated after binding a wider range of ligands. (C) Increased expression of Bcl2 and/or loss of PTEN lead to increased survival signals. See text for details. AR, androgen receptor; GF, growth factor; coA, co-activator; hsp, heat shock protein; SRC-1, steroid receptor co-activator-1; RNA pol II, RNA polymerase complex; E 2 , estradiol. Adapted from [70]. 13 of serine-650 in A R was also observed in response to both androgens and forskolin [78,79]. A number of in vitro experiments have shown that forskolin is able to induce ligand-independent activation of the AR [73,74]. This effect occurs through an increase in intracellular cyclic AMP levels, leading to activation of protein kinase A and phosphorylation of its downstream targets. The result is the production of endogenous PSA mRNA dependent on the activity of the AR [74]. Forskolin was also shown to increase the transcriptional activation function of the N-terminal domain of A R as well as increase the affinity of AR for binding to AREs [74]. There is also evidence suggesting ligand-independent activation of the A R may occur in vivo. An increase in serum levels of IL-6 has been associated with the progression of prostate cancer [80-82]. Nakashima et al. [80] demonstrated that serum IL-6 levels were significantly correlated with clinical stage in 74 patients at various stages of disease progression. Another group [82] found that IL-6 was elevated in androgen-independent disease relative to normal controls, benign prostatic hyperplasia, and localized and recurrent disease. Binding of IL-6 to its receptor results in downstream signalling through the Janus kinase family of proteins as well as through the mitogen-activated protein kinase (MAPK) cascade. MAPK, a kinase that regulates SRC-1 activity through phosphorylation, has been suggested to play a role in androgen-independent prostate cancer [83]. It has also been shown that treatment with IL-6 alone is sufficient to induce transcription of several androgen responsive promoters as well as the endogenous PSA gene in vitro [75,76,84,85]. Additionally, IL-6 treatment has been shown to increase the proportion of A R in the nucleus as well as increase the binding of A R to 14 androgen response elements [85]. Finally, the co-activator p300 was recently suggested to activate the expression of PSA independent of the AR in LNCaP cells after long term treatment with IL-6 [86]. A second mechanism by which the AR may become active in the absence of androgens is through increased promiscuity of the receptor (Fig. 1.2B). Increases in the copy number of AR and the amount of AR protein have been correlated with development of HRPC in 20-30% of patients [87,88]. Increased AR expression was also shown to sensitize prostate cancer cells to lower levels of androgens as well as the antiandrogen bicalutamide [68]. A number of mutations have also been identified in the A R of prostate cancer cell lines and patients that allow the receptor to be activated after binding a wider range of ligands, as well as antiandrogens [62,89-93]. However, approximately 10% of patients with HRPC had AR mutations, though this may be dependent on the length of therapy, and there was no correlation between the presence of A R mutations and overall survival [92,94]. It is therefore likely that the mechanism of androgen independence in approximately 60-70% of HRPC patients is not dependent on amplification or mutation of the AR. Additionally, these studies demonstrate that the A R is expressed in a large proportion of HRPC patients, suggesting that the loss-of-function of A R is not the predominant mechanism of androgen-independent growth. The survival of prostate cancer cells after ADT indicates these cells are proliferating in the absence of androgens, perhaps due to the mechanisms discussed above. However, it is also possible that these surviving cancer cells may not undergo the typical apoptotic response to androgen withdrawal (Fig. 1.2C). These alternative theories are not mutually exclusive, and in fact are likely complementary. Consistent with this, 15 expression of the anti-apoptotic gene BCL2 has been found to increase with progression to HRPC [95]. Additionally, loss of the tumour suppressor PTEN has been correlated with prostate cancer progression and has been shown to increase cell survival through activation of Akt [96-98]. Mechanisms such as these can be utilized by the tumour to overcome the pro-apoptotic signals resulting from the removal of androgens, allowing the cancer cells to survive and grow in an androgen-independent manner. The A R is expressed in the nucleus of a high proportion of HRPC tumours. This finding suggests the growth of the tumour may be promoted by A R activated in the absence of androgens. However, these cells may have additional abnormalities that also help them to survive ADT therapy. It is likely that there are multiple mechanisms leading to the androgen-independent growth of prostate cancer cells, and no single mechanism is sufficient to explain the growth of all cases of HRPC. 1.4 Models for studying androgen-independent disease Given the difficulty of obtaining androgen-independent tissue from HRPC patients, a number of groups have attempted to develop in vitro and in vivo models that recapitulate the hormonal progression of prostate cancer. The most frequently used prostate cancer cell line is the lymph node-derived carcinoma of the prostate (LNCaP) line originally derived from a lymph node metastasis of a human prostate cancer. This cell line expresses a functional AR, is responsive to androgens such as DHT, and expresses the prostatic marker prostatic acid phosphatase (PAP) [99,100]. LNCaP cells also express PSA mRNA and protein, which can be used as a natural androgen-responsive reporter [51,101]. The AR expressed by these cells 16 contains a point mutation in the LBD that broadens ligand specificity of the receptor, allowing it to also be activated by progesterone, estradiol, and anti-androgens [89]. Inoculation of LNCaP cells subcutaneously or intraprostatically in male SCID or athymic nude mice results in the formation of a tumour; serum levels of PSA secreted by these tumours correlates with the overall tumour volume [102,103]. Removal of androgens through surgical castration reduces serum levels of PSA, as in the clinical situation. Interestingly, 2-4 weeks post-castration the serum PSA levels of these mice exceeds the pre-castration level, indicating that PSA is expressed in an androgen-independent manner in these mice [102,103]. Unlike the clinical situation, though, LNCaP cells are not androgen dependent and thus do not regress after castration [102]. Regardless, LNCaP cells are useful for studying AR-dependent gene expression, both in the presence and the absence of androgens. The LNCaP xenograft model is thus a valuable system for studying changes in gene expression with the hormonal progression of prostate cancer. However, a disadvantage of this model is that these tumours are highly vascularized and contain a large proportion of murine stromal tissues. Given the high degree of sequence similarity between humans and mice it is difficult to monitor gene expression changes while being confident that only human transcripts are being measured. To address this issue, Sadar et al. developed a novel in vivo prostate cancer tumour model that progresses to androgen-independence upon castration of the host [104]. This model utilizes LNCaP cells grown inside of hollow fibres that are implanted subcutaneously into male nude mice. The fibres that these cells are grown in are porous enough to allow nutrients and wastes to be exchanged, but not so porous as to allow LNCaP cells to escape or mouse cells to enter. 17 This provides a source of uncontaminated DNA, RNA, and protein extracts for experiments investigating the molecular changes involved in the progression of prostate cancer to androgen-independence. In addition to LNCaP there are numerous other human prostate cancer cell lines and xenograft models. The two most common in vitro models, besides LNCaP, are the PC3 and DU145 cell lines, both of which are androgen-independent. The PC3 line was derived from a bone metastasis of a human prostate cancer and is more poorly differentiated than LNCaP [105]. These cells do not express AR and are unresponsive to androgens. They also show greater tumourigenicity and a higher growth rate than LNCaP when implanted in male nude mice. The DU145 line was derived from a brain metastasis of a human prostate cancer [106]. These cells are more similar to PC3 in that they also lack a functional AR, are unresponsive to androgens, and show a more malignant phenotype than LNCaP cells when grown as a xenograft. Since neither of these lines express the AR they are not ideal systems in which to study AR-dependent gene expression, nor do they represent the majority of clinical cases of the disease. An additional model of prostate cancer that has received attention of late is the CWR22 xenograft model. This serially transplantable xenograft was originally derived from a primary prostate cancer [107]. Like LNCaP xenografts, CWR22 xenografts show decreased expression of serum PSA after castration with eventual progression to androgen-independent growth and PSA expression, but on a much slower time scale [108]. Importantly, though, the CWR22 model also regresses after castration, in contrast to the LNCaP model and similar to the clinical situation [109]. However, the androgen-dependent cells are difficult to maintain in vitro, thus limiting the experimental 18 manipulations which may be performed. And, like the LNCaP subcutaneous xenograft, CWR22 xenografts also contain a significant amount of murine tissues. While each of the available models of prostate cancer has different advantages, we chose to perform our experiments with the LNCaP hollow fibre xenograft. This decision was made based on availability of the models at the time, our desired experimental designs and need of samples free of mouse tissues, and to give us the flexibility to perform follow-up experiments in either the LNCaP subcutaneous xenograft or LNCaP in vitro models. 1.5 Differential gene expression A number of genes important in the development of cancer have been identified by first demonstrating altered expression between the normal and cancerous state. Perhaps most famously, the p53 protein was first identified as being upregulated in cells transformed with SV40 T antigen [110]. Of course, p53 is now known to play critical roles in regulating apoptosis, proliferation, and senescence. It has thus become desirable to perform high-throughput gene expression analyses to quickly assay the expression status of large numbers of genes in a given model or treatment condition. 1.5.1 Gene expression changes in prostate cancer Previous studies have specifically examined gene expression changes in clinical samples of prostate cancer to identify potential markers whose expression levels correlate with patient outcome. For example, patients with decreased expression of KAI1 or E -cadherin were more likely to develop metastatic disease and have poorer overall survival 19 [111,112]. Likewise, samples of prostate cancer were much more likely to display decreased expression of the glutathione S-transferase gene GSTP1 than were samples of normal tissues [113]. Expression of the cell cycle inhibitor p27 was also shown to be lower in hyperplastic and cancerous prostate tissues relative to normal cells [114]. Samples of early stage prostate cancer were also shown to express increased levels of the kinase PIM1 and the serine protease hepsin [115]. Additionally, tumours expressing increased levels of the MUC1 proteoglycan were more likely to display higher levels of intratumoural angiogenesis [116]. Finally, expression of both EZH2 and BCL2 was shown to be higher in samples of hormone-refractory disease, indicating expression of these proteins may be a marker of hormonal progression [95,117]. However, while increased expression of one of these genes may correlate with one stage of the disease, other stages of the disease may be associated with decreased expression of the same gene. Such results can be difficult to interpret, and consequently, none of these genes have been successfully translated into clinical use. Given this diversity of results, it is interesting to note again the study of Chen et al. [68]. This study examined gene expression changes associated with hormonal progression of prostate cancer in multiple model systems. The final interpretation of their results suggested that increased expression of the A R was the only change consistently shared amongst these models. Importantly, though, these are strictly disease models, whereas the expression changes noted above were observed directly in clinical specimens. Since focusing on individual genes may be inefficient, other groups have attempted to build prognostic models based on the expression levels of multiple genes in a single sample [118]. Unfortunately, these studies have so far not resulted in altered 20 clinical management of the disease. Thus, new markers and expression signatures correlating with patient outcome still need to be identified. Interestingly, the identification of novel genes may provide new candidate genes that have not been considered previously. 1.5.2 Techniques for transcriptional profiling A variety of techniques are now available for monitoring gene expression profiles (see [119] for a review). Microarrays are perhaps the most common method of assaying gene expression of a large number of transcripts. There are two types of microarrays: those spotted with full-length cDNAs of specific genes [120] and those generated with oligonucleotide spots representing regions of known transcripts [121]. Each of these platforms is able to measure the expression of tens of thousands of transcripts. However, microarray experiments are only able to monitor the expression of genes for which prior knowledge of the transcript sequence is available. Microarrays also lack the sensitivity to detect transcripts expressed at very low levels and only provide a relative, not absolute, expression level. As sequencing capacity increased with the advent of the human genome project some groups began to sequence expressed transcripts as a method of identifying genes. These single-pass sequencing reads, called expressed sequence tags (EST), provided information on what genes were expressed in a sample and was able to identify novel genes [122,123]. However, this method is inefficient both for detecting rare transcripts and for monitoring gene expression changes. 21 Serial Analysis of Gene Expression (SAGE) is an alternative method that was developed to provide an unbiased and quantitative overview of the entire transcriptome of a sample [124]. It is based on a similar idea as EST sequencing, but SAGE provides for more cost-effective and extensive analyses. SAGE overcomes some of the difficulties of microarrays as this technique does not require prior knowledge of transcript sequence and is thus able to identify and measure the expression level of novel transcripts [124,125]. Additionally, SAGE provides a quantitative measure of gene expression [124]. However, SAGE is an extremely expensive technique requiring significant sequencing capacity. And while SAGE allows for the detection of novel transcripts, the small amount of sequence information in a SAGE tag increases the difficulty of directly studying these novel transcripts. Differential display was developed in order to specifically amplify and isolate transcripts differentially expressed between two samples [126]. This method also does not require prior sequence knowledge and allows for the isolation of novel differentially expressed transcripts. However, differential display usually identifies only a few candidates, and given the overall complexity of a cDNA sample it would be difficult to isolate low abundance transcripts. Suppression subtractive hybridization (SSH) is similar to the approach of EST sequencing, but SSH includes a normalization step that enriches for rare transcripts in a population of RNAs, while at the same time selecting against highly abundant transcripts [127,128]. Additionally, subtractive hybridization is able to detect entirely novel transcripts for which no previous annotation exists [129,130]. Thus, subtractive hybridization is a powerful tool to detect less abundant transcripts and the novel 22 transcripts that tend to be expressed at low levels. In support of this concept, a significant proportion of the transcripts identified by subtractive hybridization were shown to be expressed at levels below the detection limit of Affymetrix GeneChip® microarrays [131]. Subtractive hybridization also identified a number of novel transcripts which were not even represented on these arrays [131], confirming the value of not being dependent on prior knowledge of transcript sequence. However, subtractive hybridization is not as efficient as microarrays or SAGE at generating expression profiles for large numbers of genes. But, in contrast to SAGE, SSH recovers longer sequence fragments, thus simplifying the process of identifying and characterizing novel differentially expressed transcripts. The wealth of data from the human genome project has greatly changed the means by which high-throughput gene expression analyses are performed. More recent modifications of SAGE and microarrays have also addressed some of their original disadvantages by providing longer sequence tags and higher density arrays to represent a larger proportion of the transcriptome [125]. However, at the time this project began, SSH was the most efficient and cost-effective technique for identifying the novel and low abundance transcripts we were interested in examining. 1.6 Suppression subtractive hybridization Suppression subtractive hybridization was first described in 1996 by Diatchenko, et al. [127]. The relative advantages and disadvantages of this technique were described in Chapter 1.5. The original subtractive hybridization technique was developed to identify genes expressed under one treatment condition that were not expressed under a 23 second treatment condition [132]. SSH differs from subtractive hybridization by an additional normalization mechanism that equalizes the relative abundance of cDNAs within a sample, enabling more efficient isolation of transcripts expressed at low levels [127]. The SSH experiments described in this thesis were performed with the PCR-Select™ cDNA Subtraction Kit from Clontech Laboratories, Inc. (Palo Alto, CA). The manufacturer's protocol and the theory behind the technique are briefly described below. SSH allows the relative comparison of gene expression levels in two samples of interest. The idea is that one cDNA population (the driver) is "subtracted" from the other cDNA sample (the tester), leaving just those transcripts that are upregulated in the tester sample relative to the driver sample. Reversing this procedure identifies transcripts downregulated in the tester relative to the driver. First, double-stranded cDNA is generated from the total RNA populations of interest. These cDNAs are then digested with Rsal restriction endonuclease to produce cDNA fragments of approximately equal sizes, thus resulting in similar hybridization efficiencies for all cDNA fragments. Next, the tester cDNA population is split in two groups and different adaptors are ligated to each group (see Fig. 1.3). An excess of driver cDNA is then added to each reaction, the DNA denatured, and hybridization performed. Any cDNA fragments in common between the tester and driver samples will hybridize with each other. High abundance transcripts are more likely to re-anneal with their counterpart cDNAs, thus "normalizing" the concentrations of high and low abundance transcripts. The two different reactions are then combined and fresh denatured driver is added. Again, the driver cDNA will hybridize with any cDNA fragments also found in 24 Tester cDNA Tester cDNA Adaptor 1 on Driver cDNA (in excess) Denature cDNA First hybridization Fresh denatured driver Tester cDNA Adaptor 2 Second hybridization I T PCR amplification Figure 1.3 - Schematic diagram of suppression subtractive hybridization protocol. Two rounds of subtractive hybridization normalize the relative amount of each cDNA fragment while subtracting cDNA fragments shared between the tester and driver populations. Of the resulting cDNA molecules, only those with unique adaptors at each end are exponentially amplified by PCR. See text for details. Adapted from [127]. 25 the tester sample. As the tester cDNAs are not denatured again, only single-stranded fragments with Adaptor 1 will be free to anneal with their counterpart fragments with Adaptor 2. Thus, these newly annealed molecules have unique adaptors at opposite ends of the double-stranded cDNA. These are also the cDNA molecules that have increased expression in the tester sample relative to the driver sample. Unique PCR primers can then be used to selectively amplify only those cDNA fragments containing different adaptors on either end (see Fig. 1.3). After two rounds of PCR, the differentially expressed genes have been selectively amplified and they can then be cloned and characterized further. 1.7 Summary and research objectives Prostate cancer is the most frequently diagnosed cancer as well as the second leading cause of cancer deaths among American men. Androgen ablation therapy for patients with advanced prostate cancer inevitably fails as the disease progresses to an androgen-independent stage, leaving patients with no further effective treatment options. It is not understood how prostate cancer cells, normally dependent on androgens for survival, are able to adapt and thrive in this androgen-depleted environment. Current research, though, indicates the androgen receptor is transcriptionally active in many cases of hormone-refractory disease. However, the mechanism(s) by which the A R is activated in the absence of androgen are not fully understood. Additionally, new markers of disease progression are needed to differentiate those patients who will benefit from aggressive treatment from those who will not. 26 1.7.1 Hypothesis My overarching hypothesis was that the application of subtractive hybridization at different stages of hormonal progression in the LNCaP hollow fibre model would identify genes of interest in prostate cancer progression and allow for the development of new diagnostic, prognostic, and/or therapeutic targets for the treatment of HRPC. The subtractive hybridization experiments resulted in the isolation of a number of known and novel transcripts associated with the progression of prostate cancer. From these results it was then hypothesized that (1) the characterization of these novel transcripts would identify novel proteins with clinical relevance in the treatment of prostate cancer, and (2) that the known genes isolated here may function in the development of hormone refractory disease. 1.7.2 Specific aims My specific aims were: (1) Perform subtractive hybridization to identify the genes expressed at distinct stages of hormonal progression. This work will identify transcripts that may be associated with the development of androgen-independent prostate cancer. (2) To isolate and identify novel genes expressed during hormonal progression. A significant effort has been expended on examining known genes associated with prostate cancer. This aim will attempt to identify previously uncharacterized genes that are potential markers or therapeutic targets in prostate cancer. (3) To characterize, where possible, the potential mechanisms by which the identified genes contribute to the hormonal progression of prostate cancer. Some of the 27 genes identified in Aim 1 may be potential markers of disease progression. However, we also wanted to examine genes that may be responsible for activation of the A R in the absence of androgens. 1.7.3 Thesis overview The remainder of this thesis describes in detail the experiments I performed to address these specific aims. In Chapter 2,1 describe my work with SSH to characterize changes in gene expression during hormonal progression. I examined a novel in vivo tumour model, the LNCaP hollow fibre model, that mimics the hormonal progression of the prostate cancer. This model enabled the isolation and molecular analysis of prostate cancer cells (free from contamination by host cells) at various stages of hormonal progression. Aside from identifying hundreds of known genes, these experiments identified 25 expressed sequences representing novel human transcripts that had not been previously identified. Reverse transcriptase PCR (RT-PCR) for six randomly chosen sequences demonstrated they were expressed in a variety of normal human tissues as well as clinical samples of prostate cancer. Further analysis of two of these novel sequences indicated they were derived from large untranslated regions not previously identified. This study underlined the value of using complementary techniques in the annotation of the human genome. The work in Chapter 3 was performed to characterize another of the novel expressed sequences described in Chapter 2. This sequence was found to derive from a novel splice variant of the transmembrane protein TMEFF2, which has recently been proposed as a potential immunotherapeutic target for the treatment of prostate cancer. 28 This variant transcript, expressed by both normal and cancerous prostate cells, was shown to encode a secreted form of the TMEFF2 protein. The potential functions of this isoform and its potential impact on current immune-based treatment strategies are discussed. As described above, the AR is a ligand-dependent transcription factor that regulates numerous target genes, including PSA. My SSH experiments, as well as experiments performed by other members of the Sadar laboratory, indicated members of the 14-3-3 family of proteins may be differentially expressed with hormonal progression. In Chapter 4,1 describe my experiments to examine the ability of each member of this family to modulate transcription of PSA through the AR. Despite significant homology within the 14-3-3 family I observed differences in the ability of each isoform to alter the transcriptional activity of the AR. Significantly, 14-3-3 sigma activated PSA reporters not only at castrate levels of androgens, but also in the complete absence of androgens. Knockdown of the AR by siRNA oligonucleotides abolished activation of these reporters by 14-3-3 sigma, suggesting the A R is required for this activation to occur. I also discuss the potential mechanisms for these findings as well as their potential clinical significance. Finally, Chapter 5 summarizes all these experiments and discusses the relevance of these findings to prostate cancer research. This chapter also discusses current work and proposes future experiments to build on the results described in this thesis. 29 1.8 References 1. McNeal JE. Regional morphology and pathology of the prostate. Am J Clin Pathol 1968;49(3):347-357. 2. McNeal JE. The zonal anatomy of the prostate. Prostate 1981;2(l):35-49. 3. Ayala A G , Ro JY, Babaian R, Troncoso P, Grignon DJ. The prostatic capsule: does it exist? Its importance in the staging and treatment of prostatic carcinoma. Am J Surg Pathol 1989;13(l):21-27. 4. Huggins C, Hodges CV. Studies on prostatic cancer: I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1941;1:293-297. 5. Kerr JF, Searle J. Deletion of cells by apoptosis during castration-induced involution of the rat prostate. Virchows Arch B Cell Pathol 1973;13(2):87-102. 6. Evans RM. The steroid and thyroid hormone receptor superfamily. Science 1988;240(4854):889-895. 7. Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ, Thun MJ. Cancer statistics, 2005. CA Cancer J Clin 2005;55(1): 10-30. 8. McNeal JE. Origin and development of carcinoma in the prostate. Cancer 1969;23(l):24-34. 9. Lilja H. A kallikrein-like serine protease in prostatic fluid cleaves the predominant seminal vesicle protein. J Clin Invest 1985;76(5): 1899-1903. 10. Borgono CA, Michael IP, Diamandis EP. Human tissue kallikreins: physiologic roles and applications in cancer. Mol Cancer Res 2004;2(5):257-280. 30 11. Wang M C , Valenzuela L A , Murphy GP, Chu T M . Purification of a human prostate specific antigen. Invest Urol 1979; 17(2): 159-163. 12. Wang M C , Papsidero LD, Kuriyama M , Valenzuela L A , Murphy GP, Chu T M . Prostate antigen: a new potential marker for prostatic cancer. Prostate 1981;2(l):89-96. 13. Stamey TA, Yang N, Hay AR, McNeal JE, Freiha FS, Redwine E. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med 1987;317(15):909-916. 14. Stamey TA, Kabalin JN. Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. I. Untreated patients. J Urol 1989;141(5):1070-1075. 15. Sakr WA, Grignon DJ, Haas GP, Heilbrun L K , Pontes JE, Crissman JD. Age and racial distribution of prostatic intraepithelial neoplasia. Eur Urol 1996;30(2): 138-144. 16. Thompson IM, Goodman PJ, Tangen C M , Lucia MS, Miller GJ, Ford L G , Lieber M M , Cespedes RD, Atkins JN, Lippman SM, Carlin SM, Ryan A, Szczepanek C M , Crowley JJ, Coltman CA, Jr. The influence of finasteride on the development of prostate cancer. N Engl J Med 2003;349(3):215-224. 17. Cooperberg MR, Lubeck DP, Mehta SS, Carroll PR. Time trends in clinical risk stratification for prostate cancer: implications for outcomes (data from CaPSURE). J Urol 2003;170(6 Pt 2):S21-25; discussion S26-27. 31 18. Punglia RS, D'Amico A V , Catalona WJ, Roehl K A , Kuntz K M . Effect of verification bias on screening for prostate cancer by measurement of prostate-specific antigen. N Engl J Med 2003;349(4):335-342. 19. Stamey TA, Caldwell M , McNeal JE, Nolley R, Hemenez M , Downs J. The prostate specific antigen era in the United States is over for prostate cancer: what happened in the last 20 years? J Urol 2004;172(4 Pt 1): 1297-1301. 20. Kadmon D, Weinberg AD, Williams RH, Pavlik V N , Cooper P, Migliore PJ. Pitfalls in interpreting prostate specific antigen velocity. J Urol 1996; 155(5): 1655-1657. 21. Cooperberg MR, Moul JW, Carroll PR. The changing face of prostate cancer. J Clin Oncol 2005;23(32):8146-8151. 22. Speight JL, Roach M , 3rd. Radiotherapy in the management of clinically localized prostate cancer: evolving standards, consensus, controversies and new directions. J Clin Oncol 2005;23(32):8176-8185. 23. Klotz L. Active surveillance for prostate cancer: for whom? J Clin Oncol 2005;23(32):8165-8169. 24. Ryan CJ, Small EJ. Early versus delayed androgen deprivation for prostate cancer: new fuel for an old debate. J Clin Oncol 2005;23(32):8225-8231. 25. Johansson JE, Andren O, Andersson SO, Dickman PW, Holmberg L, Magnuson A, Adami HO. Natural history of early, localized prostate cancer. Jama 2004;291(22):2713-2719. 32 26. Albertsen PC, Hanley JA, Fine J. 20-year outcomes following conservative management of clinically localized prostate cancer. Jama 2005;293(17):2095-2101. 27. Bill-Axelson A, Holmberg L, Ruutu M , Haggman M , Andersson SO, Bratell S, Spangberg A, Busch C, Nordling S, Garmo H, Palmgren J, Adami HO, Norlen BJ, Johansson JE. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med 2005;352( 19): 1977-1984. 28. Stamey TA, McNeal JE, Yemoto C M , Sigal B M , Johnstone IM. Biological determinants of cancer progression in men with prostate cancer. Jama 1999;281(15):1395-1400. 29. DAmico A V , Chen M H , Roehl K A , Catalona WJ. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med 2004;351(2):125-135. 30. Stamey TA, Kabalin JN, Ferrari M . Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. III. Radiation treated patients. J Urol 1989;141(5):1084-1087. 31. Stamey TA, Kabalin JN, Ferrari M , Yang N. Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. IV. Anti-androgen treated patients. J Urol 1989;141(5):1088-1090. 32. Pound CR, Partin AW, Eisenberger M A , Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. Jama 1999;281(17): 1591-1597. 33 33. Geller J, Albert J, Yen SS, Geller S, Loza D. Medical castration of males with megestrol acetate and small doses of diethylstilbestrol. J Clin Endocrinol Metab 1981;52(3):576-580. 34. Crawford ED, Eisenberger M A , McLeod DG, Spaulding JT, Benson R, Dorr FA, Blumenstein BA, Davis M A , Goodman PJ. A controlled trial of leuprolide with and without flutamide in prostatic carcinoma. N Engl J Med 1989;321(7):419-424. 35. Loblaw DA, Mendelson DS, Talcott JA, Virgo KS, Somerfield MR, Ben-Josef E , Middleton R, Porterfield H, Sharp SA, Smith TJ, Taplin M E , Vogelzang NJ, Wade JL, Jr., Bennett CL, Scher HI. American Society of Clinical Oncology recommendations for the initial hormonal management of androgen-sensitive metastatic, recurrent, or progressive prostate cancer. J Clin Oncol 2004;22(14):2927-2941. 36. Wirth MP, See WA, McLeod DG, Iversen P, Morris T, Carroll K. Bicalutamide 150 mg in addition to standard care in patients with localized or locally advanced prostate cancer: results from the second analysis of the early prostate cancer program at median followup of 5.4 years. J Urol 2004;172(5 Pt 1): 1865-1870. 37. Eisenberger M A , Blumenstein BA, Crawford ED, Miller G, McLeod DG, Loehrer PJ, Wilding G, Sears K, Culkin DJ, Thompson IM, Jr., Bueschen AJ, Lowe BA. Bilateral orchiectomy with or without flutamide for metastatic prostate cancer. N Engl J Med 1998;339(15): 1036-1042. 38. Labrie F, Dupont A, Belanger A, Lacoursiere Y, Raynaud JP, Husson JM, Gareau J, Fazekas AT, Sandow J, Monfette G, et al. New approach in the treatment of 34 prostate cancer: complete instead of partial withdrawal of androgens. Prostate 1983;4(6):579-594. 39. Bhandari MS, Crook J, Hussain M . Should intermittent androgen deprivation be used in routine clinical practice? J Clin Oncol 2005;23(32):8212-8218. 40. Halabi S, Small EJ, Kantoff PW, Kattan MW, Kaplan EB, Dawson NA, Levine E G , Blumenstein BA, Vogelzang NJ. Prognostic model for predicting survival in men with hormone-refractory metastatic prostate cancer. J Clin Oncol 2003;21(7): 1232-1237. 41. Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi K N , Oudard S, Theodore C, James ND, Turesson I, Rosenthal M A , Eisenberger M A . Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004;351(15): 1502-1512. 42. Petrylak DP, Tangen C M , Hussain M H , Lara PN, Jr., Jones JA, Taplin M E , Burch PA, Berry D, Moinpour C, Kohli M , Benson M C , Small EJ, Raghavan D, Crawford ED. Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004;351(15):1513-1520. 43. McKenna NJ, Lanz RB, O'Malley BW. Nuclear receptor coregulators: cellular and molecular biology. EndocrRev 1999;20(3):321-344. 44. Tyagi RK, Lavrovsky Y, Ahn SC, Song CS, Chatterjee B, Roy A K . Dynamics of intracellular movement and nucleocytoplasmic recycling of the ligand-activated androgen receptor in living cells. Mol Endocrinol 2000; 14(8): 1162-1174. 35 45. Roy A K , Tyagi RK, Song CS, Lavrovsky Y, Ahn SC, Oh TS, Chatterjee B. Androgen receptor: structural domains and functional dynamics after ligand-receptor interaction. Ann N Y Acad Sci 2001;949:44-57. 46. James AJ, Agoulnik IU, Harris JM, Buchanan G, Tilley WD, Marcelli M , Lamb DJ, Weigel NL. A Novel Androgen Receptor Mutant, A748T, Exhibits Hormone Concentration- Dependent Defects in Nuclear Accumulation and Activity Despite Normal Hormone-Binding Affinity. Mol Endocrinol 2002;16(12):2692-2705. 47. Ma H, Hong H, Huang SM, Irvine RA, Webb P, Kushner PJ, Coetzee GA, Stallcup MR. Multiple signal input and output domains of the 160-kilodalton nuclear receptor coactivator proteins. Mol Cell Biol 1999; 19(9):6164-6173. 48. Wang Q, Carroll JS, Brown M . Spatial and temporal recruitment of androgen receptor and its coactivators involves chromosomal looping and polymerase tracking. Mol Cell 2005; 19(5):631-642. 49. Cheng S, Brzostek S, Lee SR, Hollenberg AN, Balk SP. Inhibition of the dihydrotestosterone-activated androgen receptor by nuclear receptor corepressor. Mol Endocrinol 2002; 16(7): 1492-1501. 50. Shang Y, Myers M , Brown M . Formation of the androgen receptor transcription complex. Mol Cell 2002;9(3):601-610. 51. Henttu P, Liao SS, Vihko P. Androgens up-regulate the human prostate-specific antigen messenger ribonucleic acid (mRNA), but down-regulate the prostatic acid phosphatase mRNA in the LNCaP cell line. Endocrinology 1992;130(2):766-772. 36 52. Betts A M , Waite I, Neal DE, Robson CN. Androgen regulation of ornithine decarboxylase in human prostatic cells identified using differential display. FEBS Lett 1997;405(3):328-332. 53. MacLean HE, Warne GL, Zajac JD. Localization of functional domains in the androgen receptor. J Steroid Biochem Mol Biol 1997;62(4):233-242. 54. Meehan K L , Sadar MD. Androgens and androgen receptor in prostate and ovarian malignancies. Front Biosci 2003;8:D780-800. 55. Simental JA, Sar M , Lane M V , French FS, Wilson E M . Transcriptional activation and nuclear targeting signals of the human androgen receptor. J Biol Chem 1991;266(1):510-518. 56. Zhou ZX, Sar M , Simental JA, Lane M V , Wilson E M . 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. J Biol Chem 1994;269(18): 13115-13123. 57. Jenster G, van der Korput HA, van Vroonhoven C, van der Kwast T H , Trapman J, Brinkmann AO. Domains of the human androgen receptor involved in steroid binding, transcriptional activation, and subcellular localization. Mol Endocrinol 1991;5(10):1396-1404. 58. Cheung J, Smith DF. Molecular chaperone interactions with steroid receptors: an update. Mol Endocrinol 2000;14(7):939-946. 59. Veldscholte J, Berrevoets CA, Zegers ND, van der Kwast T H , Grootegoed JA, Mulder E. Hormone-induced dissociation of the androgen receptor-heat-shock 37 protein complex: use of a new monoclonal antibody to distinguish transformed from nontransformed receptors. Biochemistry 1992;31(32):7422-7430. 60. Kuil CW, Berrevoets CA, Mulder E. Ligand-induced conformational alterations of the androgen receptor analyzed by limited trypsinization. Studies on the mechanism of antiandrogen action. J Biol Chem 1995;270(46):27569-27576. 61. Schaufele F, Carbonell X, Guerbadot M , Borngraeber S, Chapman MS, Ma A A , Miner JN, Diamond MI. The structural basis of androgen receptor activation: intramolecular and intermolecular amino-carboxy interactions. Proc Natl Acad Sci U S A 2005;102(28):9802-9807. 62. Veldscholte J, Berrevoets CA, Brinkmann AO, Grootegoed JA, Mulder E. Anti-androgens and the mutated androgen receptor of LNCaP cells: differential effects on binding affinity, heat-shock protein interaction, and transcription activation. Biochemistry 1992;31(8):2393-2399. 63. Georget V, Terouanne B, Nicolas JC, Sultan C. Mechanism of antiandrogen action: key role of hsp90 in conformational change and transcriptional activity of the androgen receptor. Biochemistry 2002;41(39): 11824-11831. 64. Roche PJ, Hoare SA, Parker M G . A consensus DNA-binding site for the androgen receptor. Mol Endocrinol 1992;6(12):2229-2235. 65. Kang Z, Janne OA, Palvimo JJ. Coregulator recruitment and histone modifications in transcriptional regulation by the androgen receptor. Mol Endocrinol 2004; 18(11):2633-2648. 66. Gregory CW, Hamil K G , Kim D, Hall SH, Pretlow TG, Mohler JL, French FS. Androgen receptor expression in androgen-independent prostate cancer is 38 associated with increased expression of androgen-regulated genes. Cancer Res 1998;58(24):5718-5724. 67. Holzbeierlein J, Lai P, LaTulippe E, Smith A, Satagopan J, Zhang L , Ryan C, Smith S, Scher H, Scardino P, Reuter V, Gerald WL. Gene expression analysis of human prostate carcinoma during hormonal therapy identifies androgen-responsive genes and mechanisms of therapy resistance. Am J Pathol 2004;164(l):217-227. 68. Chen CD, Welsbie DS, Tran C, Baek SH, Chen R, Vessella R, Rosenfeld M G , Sawyers CL. Molecular determinants of resistance to antiandrogen therapy. Nat Med 2004;10(l):33-39. 69. 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(9):3550-3555. 70. Feldman BJ, Feldman D. The development of androgen-independent prostate cancer. Nat Rev Cancer 200T;l(l):34-45. 71. Scher HI, Sawyers CL. Biology of progressive, castration-resistant prostate cancer: directed therapies targeting the androgen-receptor signaling axis. J Clin Oncol 2005;23(32):8253-8261. 72. Culig Z, Hobisch A, Cronauer M V , 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(20):5474-5478. 39 73. Nazareth L V , Weigel NL. Activation of the human androgen receptor through a protein kinase A signaling pathway. J Biol Chem 1996;271(33): 19900-19907. 74. Sadar MD. Androgen-independent induction of prostate-specific antigen gene expression via cross-talk between the androgen receptor and protein kinase A signal transduction pathways. J Biol Chem 1999;274(12):7777-7783. 75. Hobisch A, Eder IE, Putz T, Horninger W, Bartsch G, Klocker H, Culig Z. Interleukin-6 regulates prostate-specific protein expression in prostate carcinoma cells by activation of the androgen receptor. Cancer Res 1998;58(20):4640-4645. 76. Ueda T, Bruchovsky N, Sadar MD. Activation of the androgen receptor N-terminal domain by interleukin-6 via M A P K and STAT3 signal transduction pathways. J Biol Chem 2002;277(9):7076-7085. 77. Ueda T, Mawji NR, Bruchovsky N, Sadar MD. Ligand-independent activation of the androgen receptor by interleukin-6 and the role of steroid receptor coactivator-1 in prostate cancer cells. J Biol Chem 2002;277(41):38087-38094. 78. Zhou ZX, Kemppainen JA, Wilson E M . Identification of three proline-directed phosphorylation sites in the human androgen receptor. Mol Endocrinol 1995;9(5):605-615. 79. Gioeli D, Ficarro SB, Kwiek JJ, Aaronson D, Hancock M , Catling A D , White F M , Christian RE, Settlage RE, Shabanowitz J, Hunt DF, Weber MJ. Androgen receptor phosphorylation. Regulation and identification of the phosphorylation sites. J Biol Chem 2002;277(32):29304-29314. 40 80. Nakashima J, Tachibana M , Horiguchi Y, Oya M , Ohigashi T, Asakura H, Murai M . Serum interleukin 6 as a prognostic factor in patients with prostate cancer. Clin Cancer Res 2000;6(7):2702-2706. 81. Shariat SF, Andrews B, Kattan MW, Kim J, Wheeler T M , Slawin K M . Plasma levels of interleukin-6 and its soluble receptor are associated with prostate cancer progression and metastasis. Urology 2001 ;58(6): 1008-1015. 82. Drachenberg DE, Elgamal A A , Rowbotham R, Peterson M , Murphy GP. Circulating levels of interleukin-6 in patients with hormone refractory prostate cancer. Prostate 1999;41(2): 127-133. 83. Gioeli D, Mandell JW, Petroni GR, Frierson HF, Jr., Weber MJ. Activation of mitogen-activated protein kinase associated with prostate cancer progression. Cancer Res 1999;59(2):279-284. 84. Chen T, Wang L H , Farrar WL. Interleukin 6 activates androgen receptor-mediated gene expression through a signal transducer and activator of transcription 3-dependent pathway in LNCaP prostate cancer cells. Cancer Res 2000;60(8):2132-2135. 85. Lee SO, Lou W, Hou M , De Miguel F, Gerber L, Gao A C . Interleukin-6 Promotes Androgen-independent Growth in LNCaP Human Prostate Cancer Cells. Clin Cancer Res 2003;9(l):370-376. 86. Debes JD, Comuzzi B, Schmidt LJ, Dehm SM, Culig Z, Tindall DJ. p300 regulates androgen receptor-independent expression of prostate-specific antigen in prostate cancer cells treated chronically with interleukin-6. Cancer Res 2005;65(13):5965-5973. 41 87. Visakorpi T, Hyytinen E, Koivisto P, Tanner M , Keinanen R, Palmberg C, Palotie A, Tammela T, Isola J, Kallioniemi OP. In vivo amplification of the androgen receptor gene and progression of human prostate cancer. Nat Genet 1995;9(4):401-406. 88. Edwards J, Krishna NS, Grigor K M , Bartlett JM. Androgen receptor gene amplification and protein expression in hormone refractory prostate cancer. Br J Cancer 2003;89(3):552-556. 89. Veldscholte J, Ris-Stalpers C, Kuiper GG, Jenster G, Berrevoets C, Claassen E, van Rooij HC, Trapman J, Brinkmann AO, Mulder E. A mutation in the ligand binding domain of the androgen receptor of human LNCaP cells affects steroid binding characteristics and response to anti-androgens. Biochem Biophys Res Commun 1990;173(2):534-540. 90. Culig Z, Hobisch A, Cronauer M V , Cato A C , Hittmair A, Radmayr C, Eberle J, Bartsch G, Klocker H. Mutant androgen receptor detected in an advanced-stage prostatic carcinoma is activated by adrenal androgens and progesterone. Mol Endocrinol 1993;7(12):1541-1550. 91. Gaddipati JP, McLeod DG, Heidenberg HB, Sesterhenn IA, Finger MJ, Moul JW, Srivastava S. Frequent detection of codon 877 mutation in the androgen receptor gene in advanced prostate cancers. Cancer Res 1994;54(11):2861-2864. 92. Taplin M E , Bubley GJ, Shuster TD, Frantz M E , Spooner A E , Ogata GK, Keer HN, Balk SP. Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer. N Engl J Med 1995;332(21): 1393-1398. 93. Tan J, Sharief Y, Hamil K G , Gregory CW, Zang DY, Sar M , Gumerlock PH, deVere White RW, Pretlow TG, Harris SE, Wilson E M , Mohler JL, French FS. Dehydroepiandrosterone activates mutant androgen receptors expressed in the androgen-dependent human prostate cancer xenograft CWR22 and LNCaP cells. Mol Endocrinol 1997; 11(4):450-459. 94. Taplin M E , Rajeshkumar B, Halabi S, Werner CP, Woda BA, Picus J, Stadler W, Hayes DF, Kantoff PW, Vogelzang NJ, Small EJ. Androgen receptor mutations in androgen-independent prostate cancer: Cancer and Leukemia Group B Study 9663. J Clin Oncol 2003;21(14):2673-2678. 95. Colombel M , Symmans F, Gil S, O'Toole K M , Chopin D, Benson M , Olsson CA, Korsmeyer S, Buttyan R. Detection of the apoptosis-suppressing oncoprotein bcl-2 in hormone-refractory human prostate cancers. Am J Pathol 1993;143(2):390-400. 96. Cairns P, Okami K, Halachmi S, Halachmi N, Esteller M , Herman JG, Jen J, Isaacs WB, Bova GS, Sidransky D. Frequent inactivation of PTEN/MMAC1 in primary prostate cancer. Cancer Res 1997;57(22):4997-5000. 97. Suzuki H, Freije D, Nusskern DR, Okami K, Cairns P, Sidransky D, Isaacs WB, Bova GS. Interfocal heterogeneity of PTEN/MMAC1 gene alterations in multiple metastatic prostate cancer tissues. Cancer Res 1998;58(2):204-209. 98. Sun H, Lesche R, Li D M , Liliental J, Zhang H, Gao J, Gavrilova N, Mueller B, Liu X, Wu H. PTEN modulates cell cycle progression and cell survival by regulating phosphatidylinositol 3,4,5,-trisphosphate and Akt/protein kinase B signaling pathway. Proc Natl Acad Sci U S A 1999;96(11):6199-6204. 43 99. Horoszewicz JS, Leong SS, Chu T M , Wajsman ZL, Friedman M , Papsidero L , Kim U, Chai LS, Kakati S, Arya SK, Sandberg AA. The LNCaP cell line-a new model for studies on human prostatic carcinoma. Prog Clin Biol Res 1980;37:115-132. 100. Horoszewicz JS, Leong SS, Kawinski E , Karr JP, Rosenthal H, Chu T M , Mirand EA, Murphy GP. LNCaP model of human prostatic carcinoma. Cancer Res 1983;43(4):1809-1818. 101. Henttu P, Vihko P. Steroids inversely affect the biosynthesis and secretion of human prostatic acid phosphatase and prostate-specific antigen in the LNCaP cell line. J Steroid Biochem Mol Biol 1992;41(3-8):349-360. 102. Gleave M E , Hsieh JT, Wu HC, von Eschenbach A C , Chung LW. Serum prostate specific antigen levels in mice bearing human prostate LNCaP tumors are determined by tumor volume and endocrine and growth factors. Cancer Res 1992;52(6):1598-1605. 103. Sato N, Gleave M E , Bruchovsky N, Rennie PS, Beraldi E, Sullivan LD. A metastatic and androgen-sensitive human prostate cancer model using intraprostatic inoculation of LNCaP cells in SCID mice. Cancer Res 1997;57(8):1584-1589. 104. Sadar MD, Akopian V A , Beraldi E. Characterization of a New in Vivo Hollow Fiber Model for the Study of Progression of Prostate Cancer to Androgen Independence. Mol Cancer Ther 2002;l(8):629-637. 44 105. Kaighn M E , Narayan KS, Ohnuki Y, Lechner JF, Jones LW. Establishment and characterization of a human prostatic carcinoma cell line (PC-3). Invest Urol 1979;17(l):16-23. 106. Stone KR, Mickey DD, Wunderli H, Mickey GH, Paulson DF. Isolation of a human prostate carcinoma cell line (DU 145). Int J Cancer 1978;21(3):274-281. 107. Pretlow TG, Wolman SR, Micale M A , Pelley RJ, Kursh ED, Resnick MI, Bodner DR, Jacobberger JW, Delmoro C M , Giaconia JM, et al. Xenografts of primary human prostatic carcinoma. J Natl Cancer Inst 1993;85(5):394-398. 108. Nagabhushan M , Miller C M , Pretlow TP, Giaconia JM, Edgehouse NL, Schwartz S, Kung HJ, de Vere White RW, Gumerlock PH, Resnick MI, Amini SB, Pretlow TG. CWR22: the first human prostate cancer xenograft with strongly androgen-dependent and relapsed strains both in vivo and in soft agar. Cancer Res 1996;56(13):3042-3046. 109. Wainstein M A , He F, Robinson D, Kung HJ, Schwartz S, Giaconia JM, Edgehouse NL, Pretlow TP, Bodner DR, Kursh ED, et al. CWR22: androgen-dependent xenograft model derived from a primary human prostatic carcinoma. Cancer Res 1994;54(23):6049-6052. 110. Linzer DI, Levine AJ. Characterization of a 54K dalton cellular SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal carcinoma cells. Cell 1979;17(l):43-52. 111. Umbas R, Isaacs WB, Bringuier PP, Schaafsma HE, Karthaus HF, Oosterhof GO, Debruyne F M , Schalken JA. Decreased E-cadherin expression is associated with 45 poor prognosis in patients with prostate cancer. Cancer Res 1994;54(14):3929-3933. 112. Dong JT, Suzuki H, Pin SS, Bova GS, Schalken JA, Isaacs WB, Barrett JC, Isaacs JT. Down-regulation of the KAI1 metastasis suppressor gene during the progression of human prostatic cancer infrequently involves gene mutation or allelic loss. Cancer Res 1996;56(19):4387-4390. 113. Lee WH, Morton RA, Epstein JI, Brooks JD, Campbell PA, Bova GS, Hsieh WS, Isaacs WB, Nelson WG. Cytidine methylation of regulatory sequences near the pi-class glutathione S-transferase gene accompanies human prostatic carcinogenesis. Proc Natl Acad Sci U S A 1994;91(24):11733-11737. 114. Cordon-Cardo C, Koff A, Drobnjak M , Capodieci P, Osman I, Millard SS, Gaudin PB, Fazzari M , Zhang ZF, Massague J, Scher HI. Distinct altered patterns of p27KIPl gene expression in benign prostatic hyperplasia and prostatic carcinoma. J Natl Cancer Inst 1998;90(17): 1284-1291. 115. Dhanasekaran SM, Barrette TR, Ghosh D, Shah R, Varambally S, Kurachi K, Pienta KJ, Rubin M A , Chinnaiyan A M . Delineation of prognostic biomarkers in prostate cancer. Nature 2001;412(6849):822-826. 116. Papadopoulos I, Sivridis E, Giatromanolaki A, Koukourakis MI. Tumor angiogenesis is associated with mucl overexpression and loss of prostate-specific antigen expression in prostate cancer. Clin Cancer Res 2001;7(6): 1533-1538. 117. Varambally S, Dhanasekaran SM, Zhou M , Barrette TR, Kumar-Sinha C, Sanda M G , Ghosh D, Pienta KJ, Sewalt RG, Otte AP, Rubin M A , Chinnaiyan A M . The 46 polycomb group protein EZH2 is involved in progression of prostate cancer. Nature 2002;419(6907):624-629. 118. Singh D, Febbo PG, Ross K, Jackson DG, Manola J, Ladd C, Tamayo P, Renshaw A A , D'Amico A V , Richie JP, Lander ES, Loda M , Kantoff PW, Golub TR, Sellers WR. Gene expression correlates of clinical prostate cancer behavior. Cancer Cell 2002;l(2):203-209. 119. Liang P, Pardee AB. Analysing differential gene expression in cancer. Nat Rev Cancer 2003;3(ll):869-876. 120. Schena M , Shalon D, Davis RW, Brown PO. Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 1995;270(5235):467-470. 121. Lockhart DJ, Dong H, Byrne M C , Follettie MT, Gallo M V , Chee MS, Mittmann M , Wang C, Kobayashi M , Horton H, Brown EL. Expression monitoring by hybridization to high-density oligonucleotide arrays. Nat Biotechnol 1996;14(13):1675-1680. 122. Adams MD, Kelley JM, Gocayne JD, Dubnick M , Polymeropoulos M H , Xiao H, Merril CR, Wu A, Olde B, Moreno RF, et al. Complementary DNA sequencing: expressed sequence tags and human genome project. Science 1991 ;252(5013): 1651 -1656. 123. Adams MD, Dubnick M , Kerlavage AR, Moreno R, Kelley JM, Utterback TR, Nagle JW, Fields C, Venter JC. Sequence identification of 2,375 human brain genes. Nature 1992;355(6361):632-634. 47 124. Velculescu V E , Zhang L, Vogelstein B, Kinzler KW. Serial analysis of gene expression. Science 1995;270(5235):484-487. 125. Saha S, Sparks AB, Rago C, Akmaev V, Wang CJ, Vogelstein B, Kinzler KW, Velculescu V E . Using the transcriptome to annotate the genome. Nat Biotechnol 2002;20(5):508-512. 126. Liang P, Pardee AB. Differential display of eukaryotic messenger RNA by means of the polymerase Chain reaction. Science 1992;257(5072):967-971. 127. Diatchenko L, Lau YF, Campbell AP, Chenchik A, Moqadam F, Huang B, Lukyanov S, Lukyanov K, Gurskaya N, Sverdlov ED, Siebert PD. Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc Natl Acad Sci U S A 1996;93(12):6025-6030. 128. Gurskaya NG, Diatchenko L, Chenchik A, Siebert PD, Khaspekov GL, Lukyanov K A , Vagner L L , Ermolaeva OD, Lukyanov SA, Sverdlov ED. Equalizing cDNA subtraction based on selective suppression of polymerase chain reaction: cloning of Jurkat cell transcripts induced by phytohemaglutinin and phorbol 12-myristate 13-acetate. Anal Biochem 1996;240(l):90-97. 129. Xu J, Stolk JA, Zhang X, Silva SJ, Houghton RL, Matsumura M , Vedvick TS, Leslie KB, Badaro R, Reed SG. Identification of differentially expressed genes in human prostate cancer using subtraction and microarray. Cancer Res 2000;60(6): 1677-1682. 48 130. Xu J, Kalos M , Stolk JA, Zasloff EJ, Zhang X, Houghton RL, Filho A M , Nolasco M , Badaro R, Reed SG. Identification and characterization of prostein, a novel prostate- specific protein. Cancer Res 2001;61(4): 1563-1568. 131. Cao W, Epstein C, Liu H, DeLoughery C, Ge N, Lin J, Diao R, Cao H, Long F, Zhang X, Chen Y, Wright PS, Busch S, Wenck M , Wong K, Saltzman A G , Tang Z, Liu L, Zilberstein A. Comparing gene discovery from Affymetrix GeneChip microarrays and Clontech PCR-select cDNA subtraction: a case study. B M C Genomics 2004;5(1):26. 132. Zimmermann CR, Orr WC, Leclerc RF, Barnard EC, Timberlake WE. Molecular cloning and selection of genes regulated in Aspergillus development. Cell 1980;21(3):709-715. 49 2 NOVEL EXPRESSED SEQUENCES IDENTIFIED IN A MODEL OF ANDROGEN INDEPENDENT PROSTATE CANCER* 2.1 Introduction Prostate cancer is the most frequently diagnosed cancer as well as the second leading cause of cancer death among American men [1]. Androgen ablation therapy for patients with advanced prostate cancer inevitably fails as the disease progresses to an androgen-independent stage [2]. Few effective treatment options are available to these patients, and these increase survival by only a matter of months [3,4]. We examined an in vivo human prostate cancer tumour model to identify the underlying molecular events involved in hormonal progression. The LNCaP hollow fibre model differs from xenograft models by growing the LNCaP human prostate cancer cell line within fibres that are implanted subcutaneously in host mice [5]. These fibres prevent host cells from infiltrating, and contaminating, the tumour cell population. Upon castration of the host the LNCaP cells progress to an androgen-independent stage as determined by a rising titre of serum prostate-specific antigen (PSA), mimicking this aspect of clinical disease [5]. Many genes important in the development and progression of cancer have been identified by first detecting their altered expression at different stages of the disease. It has thus become desirable to perform high-throughput gene expression analyses to quickly assay the expression status of large numbers of genes in a given model or treatment condition. A variety of techniques are available for monitoring gene * A version of this chapter is a manuscript in preparation. Quayle, SN, Hare, H , Delaney, A D , Hwang, D, Schein, J, Jones, SJM, Marra, M A , Sadar M D . Novel expressed sequences identified in a model of androgen independent prostate cancer. In preparation. 50 expression profiles, with microarrays and Serial Analysis of Gene Expression (SAGE) being the most widely used. However, microarray experiments are only able to monitor the expression of genes for which prior knowledge of the transcript sequence is available. Microarrays also lack the sensitivity to detect transcripts expressed at very low levels. The SAGE technique is capable of detecting novel transcripts [6,7], but SAGE is also not optimal for detecting low abundance transcripts. In contrast, suppression subtractive hybridization includes a normalization step that enriches for rare transcripts in a population of RNAs [8,9]. Subtractive hybridization is also able to detect entirely novel transcripts for which no previous annotation exists [10,11]. Thus, subtractive hybridization is a powerful tool to detect less abundant transcripts and the novel transcripts that tend to be expressed at low levels. In support of this concept, a significant proportion of the transcripts identified by subtractive hybridization were shown to be expressed at levels below the detection limit of Affymetrix GeneChip® arrays [12]. Additionally, subtractive hybridization identified a number of novel transcripts which were not represented on these arrays [12]. Gene expression changes occurring with the hormonal progression of prostate cancer have been examined in various systems (see [13-15], for example). Our goal was to utilize the LNCaP hollow fibre model to identify genes that had not been previously associated with prostate cancer in order to gain new insights into the hormonal progression of this disease. The application of subtractive hybridization resulted in the identification of a number of novel expressed sequences in this model of androgen independence. These sequences exhibit low protein-coding potential and low 51 conservation across species, but RT-PCR experiments confirmed their expression in samples of prostate cancer and in a variety of human tissues. 2.2 Materials and methods 2.2.1 Cell culture LNCaP cells obtained from Dr. L.W.K. Chung (Emory University School of Medicine, Atlanta, GA) were maintained in RPMI 1640 supplemented with 5% (v/v) fetal bovine serum (FBS) (HyClone, Logan, UT), 100 units/mL penicillin, and 100 ug/uL streptomycin. All chemicals were purchased from Sigma, unless stated otherwise. Androgen stimulation was performed by treatment with the synthetic androgen R1881 (PerkinElmer, Woodbridge, Canada). 2.2.2 LNCaP Hollow Fibre Model The LNCaP hollow fibre model has been described in detail previously [5]. Briefly, LNCaP cells (2xl07) in RPMI 1640 with 20% (v/v) FBS were sealed inside polyvinylidine fluoride (PVDF) fibres (500 kDa molecular weight cutoff; 1 mm internal diameter; Spectrum Medical Co, Houston, TX) and incubated overnight at 37°C. The fibres were then cut into fragments of approximately 2 cm and inserted subcutaneously into anesthetized male 6-8 week old athymic nude mice (BALB/c strain) obtained from Charles River Laboratory (Montreal, Canada). Serum samples were obtained from the dorsal tail vein of mice every 7 days, and PSA levels measured by an immunoenzymatic assay (Abbott IMX, Montreal, Canada). Serum samples were always obtained prior to the performance of any procedure. After one week mice were castrated by ligation of the 52 vas deferens through a small incision in the scrotum. Control (intact) animals were not castrated, but all other procedures were performed on the same schedule. Hollow fibres were removed on the day of castration, 10 days post-castration, and 45-60 days post-castration when serum PSA levels had risen. All fibres were immediately placed on ice, washed three times in sterile phosphate buffered saline (PBS), and wiped with sterile, moistened lab wipes. Any fibre that appeared to be contaminated by mouse tissue was set aside. To harvest cells, lmL of ice-cold TRIZOL® Reagent (Invitrogen, Burlington, Canada) was flushed through the fibres and the cells homogenized with a 21-G needle prior to storage at -80°C. All animal procedures were performed according to protocols approved by the Animal Care Committee at the University of British Columbia (see Appendix). Two unique sets of hollow fibre samples were analyzed in these experiments. In the first set of samples, the total amount of time the cells were maintained in subcutaneous fibres was kept constant, with all samples being harvested from mice approximately 60 days after implantation of the fibres. Samples were obtained from (A) two mice that were not castrated (intact); (B) one mouse castrated 10 days prior; (C) one mouse castrated 60 days prior that had progressed to androgen independence; and (D) one mouse castrated 61 days prior that had not progressed to androgen independence. Samples B, C, and D were then compared to the intact samples (A) using subtractive hybridization. These samples were all obtained from fibres maintained in different animals. The second set of samples controlled for any differences that may be observed due to maintenance of fibres in different animal hosts. In this experiment, fibres were 53 implanted in a single mouse, the mouse was castrated 21 days later, and serum PSA levels in this mouse were deemed to be androgen-independent 63 days after implantation of the fibres. Fibres were removed from this mouse immediately prior to castration (E; 21 days post-implant), 10 days after castration (F; 31 days post-implant), and 42 days after castration (G; 63 days post-implant). Samples E, F, and G were then compared using subtractive hybridization. These samples were all isolated from the same animal, but the cells had been cultivated subcutaneously in hollow fibres for different amounts of time. 2.2.3 Suppression subtractive hybridization The SMART PCR cDNA Synthesis Kit (Clontech, Palo Alto, CA) was used to generate full-length cDNA from 1 pg of starting total RNA according to the manufacturer's protocol. Suppression subtractive hybridization was then performed with the PCR-Select™ cDNA Subtraction Kit (Clontech) according to the manufacturer's protocol. Briefly, the cDNA was digested with Rsal restriction endonuclease to generate fragments of approximately equal lengths. The digested cDNA was purified and split into two populations before ligation of Adaptor 1 or Adaptor 2R. An excess of driver cDNA was added to each reaction, the DNA denatured, and hybridization performed for 8 hours at 68°C. The two populations of cDNA were then combined and fresh denatured driver cDNA added before hybridizing for an additional 16 hours. The final cDNA population was subjected to two rounds of PCR to specifically amplify the differentially expressed cDNA transcripts. The efficiency of subtraction was determined by monitoring levels of the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase 54 (GAPDH) in these samples. The subtracted cDNA pools were then ligated into pCR 2.1-TOPO vector (Invitrogen) and transformed into competent bacteria. Positive colonies were selected and their inserts sequenced unidirectionally with T7 primer on an ABI 3700 automated sequencer. Clones containing novel inserts were sequenced bidirectionally to obtain the entire insert sequence. All sequences obtained in these experiments have been deposited in dbEST (Accession numbers EC093848 - EC094057) and GenBank (Accession numbers DQ668378 - DQ668403). 2.2.4 RT-PCR Total RNA was isolated from LNCaP cells maintained in vivo using TRIZOL® Reagent according to the manufacturer's protocol. Total RNA samples from several human tissues were purchased from Stratagene (La Jolla, CA). Samples of total RNA from cases of prostate cancer, and their matched normal samples, were purchased from Genomics Collaborative (Cambridge, MA). Reverse transcription (RT) was performed using M M L V - R T (Invitrogen) with 1 fig of template RNA. Subsequent PCR reactions were performed using 1 uL of the resulting cDNA as template. The primers used to amplify the clones are summarized in Table 2.1. PCR products of interest were cloned into pCR®2.1-TOPO vector and sequenced by the NAPS facility at the University of British Columbia. 55 Table 2.1- Primers used for RT-PCR Clone Sequence (5'-3') 1B09 F CACAGGAGACCCTGTCTTACCT R AAGCTCTTGCTAGGCATGTAGG 1E05 F AATAGATTGGCAGGCCTTTG R TGGGATGAGCAGGATATCAA 2A03 F AGAGATGCAAACGGACGAAC R TCACTTACTGGCTCCTGCAC 1AC02 F AAGGTTTCCATTGCATCAGG R CCTGAAAGGCTGGTCTTCAA 2DB01 F CCACAACTTGGAAGCAATCA R TTCCCTGTCCCCTAACTCCT 2DD06 F TGGCCATCATCAAGTCGATA R GAGGGATGGTGAAATCACTG GAPDH F CCGAGCCACATCGCTCAGA R CCCAGCCTTCTCCATGGTG 56 2.3 Results 2.3.1 Subtractive hybridization Suppression subtractive hybridization was used to isolate novel transcripts expressed at different stages of hormonal progression in the LNCaP hollow fibre model. This model enables the isolation and molecular analysis of prostate cells (free from contamination by host cells) at multiple stages of hormonal progression. Samples from intact mice, from mice 10 days after castration (at PSA nadir), and mice 45-60 days post-castration were used as the tester and driver samples in independent subtractive hybridization experiments. Each experiment compared different time points, and both forward and reverse subtractions were performed for each comparison. Of the known genes identified in these independent experiments, neural cell adhesion molecule 2 (NCAM2) and a disintegrase and metalloproteinase domain 2 (ADAM2) were determined to be upregulated in androgen independence relative to the intact state in both subtractive hybridization experiments. Additionally, peroxiredoxin 3 was found to be increased at 10 days post-castration in both experiments. The identification of these genes in independent experiments supports the conclusion that subtractive hybridization has detected consistent gene expression changes in these samples. Furthermore, the expression of Nkx 3.1, a gene known to be downregulated with prostate cancer progression, was decreased in androgen independence. 2.3.2 Novel transcripts were represented in the subtracted cDNA libraries To identify novel genes associated with prostate cancer we examined the sequences of all 428 of the subtracted clones isolated from our subtractive hybridization experiments. 57 These sequences were filtered to remove poor quality sequences and any clones containing less than 25 bp of sequence, more than one insert, or regions of repetitive sequence (Table 2.2). Of the remaining 340 clones, 103 clones contained inserts that were represented by other clones in the library. BLAST analysis of the Ensembl database (v. 35 - Nov 2005) [16] was then used to identify the transcript represented in each of the remaining 237 nonredundant clones. First, the clones were searched against the population of annotated Ensembl cDNAs, identifying 150 clones (63.3%) derived from previously annotated human cDNAs (Fig. 2.1). The remaining 87 clones were then mapped to the human genome using the Ensembl database, with all mappings verified using the UCSC Human Genome Browser (hgl7 - May 2004) [17] and GenBank databases. This search identified 57 clones (24.1%) matching expressed sequence tags (EST) and mRNAs which had not yet been classified as Ensembl cDNAs. A clone exhibiting any overlap with a transcript was considered to be part of that transcript. Of the 30 remaining clones, 5 (2.1%) did not align uniquely to the genome, but the other 25 (10.5%o) did map uniquely to unannotated regions of the human genome and were considered to represent novel transcripts (Fig. 2.1). These 25 clones mapped to a variety of locations, including within introns of annotated transcripts, nearby annotated transcripts, as well as intergenic regions. This analysis suggested that novel transcripts were represented in our subtractive hybridization libraries. 2.3.3 Characterization of novel clone sequences We next assessed the sequence characteristics of these novel clones to determine if there were differences between those mapping to unannotated regions of the genome 58 Table 2.2 - Summary of sequenced clones. Experiment 1 (%) Experiment 2 (%) Total (%) Total sequenced clones 211 217 428 Mouse 23(10.9) 1 (0.5) 24 (5.6) Multiple inserts 2(0.1) 8 (3.7) 10 (2.3) Poor quality 30 (14.2) 24 (11.1) 54(12.6) Redundant 60 (28.4) 43(19.8) 103 (24.1) Remaining Nonredundant 96 (45.5) 141 (65.0) 237 (55.4) No unique match 1 (1.0) 4 (2.8) 5(2.1) Annotated gene 74 (77.1) 76 (53.9) 150 (63.3) EST 12 (12.5) 45 (31.9) 57 (24.1) Unannotated 9 (9.4) 16(11.3) 25(10.5) 59 Figure 2.1 - Distribution of sequence matches in the suppression subtractive hybridiza-tion clone set. The 237 non-redundant clones identified through sequence analysis of the subtractive hybridization libraries were mapped to the human genome (v. 35) and classi-fied as representing an annotated Ensembl cDNA, a previously identified expressed sequence tag (EST), or an unannotated region of the genome. Clones not mapping uniquely to the genome were classified as ambiguous matches. 60 relative to those clones mapping to known transcripts. Five characteristics were examined for each clone: (1) their protein coding potential, (2) their evolutionary conservation, (3) whether they showed evidence of splicing, (4) whether they demonstrated homology to known non-coding RNAs, and (5) whether they contained a poly-adenylation site. For this analysis, the 25 novel clones (Table 2.3) were compared to 25 clones randomly chosen from the 57 that mapped to previously identified ESTs (Table 2.4), and another 25 clones randomly chosen from the 150 mapping to annotated transcripts (Table 2.5). First, to assess the likely coding potential of each clone, the length of the longest open reading frame (ORF) was determined using the ORF Finder tool at NCBI. The clones derived from known transcripts were much more likely to contain an ORF spanning most of the sequence. It should also be noted that the sequences used for this analysis were derived from single-pass sequencing reads, and thus potential sequencing errors may interrupt longer ORFs. Next, we qualitatively assessed the degree of conservation of each of these clones using the Vertebrate and Multiz Alignment and Conservation track at UCSC [18]. This track provides a measure of evolutionary conservation of a genomic region amongst 17 vertebrate species. As predicted, clones originating from known transcripts tended to show high conservation of their respective genomic loci, while the clones representing novel transcripts mapped to regions of the genome that were poorly conserved. This was consistent with the prediction that highly conserved transcripts would be expressed by a larger number of species, thus increasing the likelihood of the transcript having been detected in previous studies. 61 Table 2.3 - Sequence characteristics of unannotated clones. Clone Accession Length (bp) Open Reading Frame (%)T Degree of Conservation* Splicing polyA++ 11_B09 DQ668378 306 25.2 + no no 11_E05 DQ668379 1338 8.1 + no no 12_A03 DQ668380 543 46.8 + no no 12_A06 DQ668381 324 39.2 + no no 12_E10 DQ668383 576 46.7 + no no 13_F09 DQ668384 454 50.7 + no yes 13_G01 DQ668385 261 41.0 + no yes 13_G06 DQ668386 389 32.1 + no yes 14_D01 DQ668387 134 77.6 +++ no no 1a_A04 EC093934 245 25.3 + no no 1a_C02 DQ668388 1853 12.4 ++ no no 1a_E09 DQ668389 796 29.5 + no no 1b_C12 DQ668390 555 16.8 + no no 1c_D03 DQ668391 292 18.5 + no no 1c_E10 DQ668392 679 14.9 + no no 1c_F02 DQ668393 475 26.7 + no no 1d_E01 DQ668394 771 26.7 + no no 1d_F04 DQ668395 465 21.9 +++ no yes 2c_A06 EC094025 658 10.8 ++ no no 2c_C10 DQ668396 371 28.0 ++ no yes 2c_G10 DQ668397 969 19.8 + no no 2d_B01 DQ668398 1100 15.2 + no no 2d_D06 DQ668399 598 37.0 + no yes 2d_D09 EC094055 627 26.6 + no no 2d_E01 DQ668400 722 34.8 + no no * Length of longest open reading frame as a percentage of the total length of each clone * Degree of conservation of each clone, +, little to no conservation; +++++, high degree of conservation. 62 Table 2.4 - Sequence characteristics of clones matching previously identified ESTs. Clone Accession Length (bp) Open Reading Frame (%)T Degree of Conservation* Splicing polyA++ 11_C08 EC093853 470 19.6 + no no 13_F07 EC093910 281 44.5 + no yes 13_G08 EC093913 126 75.4 + no no 14_D02 EC093923 671 27.6 + no no 14_G04 EC093933 521 42.4 + no no 1a_B02 EC093936 579 25.7 + no no 1a_D01 EC093941 451 20.4 ++ no no 1a_F05 EC093946 478 19.9 +++++ no yes 1a_H09 EC093956 683 33.7 + no no 1b_A09 EC093959 619 30.4 +++ no no 1b_A11 EC093961 284 63.0 + no yes 1b_C01 EC093963 584 27.9 + no no 1c_A02 EC093968 514 25.5 ++ no yes 1c_D12 EC093979 215 59.5 +++++ no no 1c_E05 EC093980 429 49.4 + yes yes 1c_G06 EC093989 525 26.1 + no no 1d_C12 EC094007 596 22.0 ++ no yes 1d_G02 EC094018 351 36.5 +++++ no no 2c_B04 EC094028 484 21.5 + no yes 2c_B10 EC094032 427 44.0 + no no 2c_D10 EC094037 487 37.0 + no no 2c_F04 EC094043 384 45.8 + no yes 2c_H03 EC094044 642 48.0 + no no 2d_B10 EC094049 299 54.8 +++ no no 2d_D07 EC094054 299 60.5 + no no T Length of longest open reading frame as a percentage of the total length of each clone. * Degree of conservation of each clone, +, little to no conservation; +++++, high degree of conservation. 63 Table 2.5 - Sequence characteristics of clones matching previously annotated cDNAs. Clone Accession Length (bp) Open Reading Frame (%)* Degree of Conservation* Splicing polyA++ 11_F09 EC093861 149 100.0 + no yes 11_G07 EC093866 169 86.4 ++++ yes no 11J309 EC093868 323 43.3 ++ no yes 12_F01 EC093889 426 91.3 ++ yes no 12_F10 EC093893 276 74.6 ++++ yes no 12_G09 EC093898 279 49.1 ++ no no 12_G11 EC093899 433 73.2 +++++ yes yes 14_C11 EC093921 269 94.1 ++++ yes no 14_D07 EC093926 77 96.1 +++++ no no 14_E12 EC093929 318 43.1 ++++ no yes 14_F09 EC093932 180 100.0 +++ yes no 1a_B07 EC093937 477 69.0 +++++ yes yes 1a_E11 EC093945 655 33.7 ++++ yes no 1c_B12 EC093972 555 32.3 +++ yes yes 1c_D05 EC093975 325 27.4 +++ no yes 1c_D07 EC093977 425 37.9 ++ no no 1d_B08 EC094000 408 34.3 ++++ no no 1d_D04 EC094008 245 100.0 +++++ no no 1d_E07 EC094012 555 89.2 +++++ yes no 1d_E11 EC094014 670 60.6 ++ no yes 1d_F08 EC094016 297 91.9 +++++ yes no 1d_H01 EC094020 701 44.8 +++++ yes no 2c_B03 EC094027 92 100.0 +++++ yes no 2c_D08 EC094036 475 37.7 +++++ no no 2d B03 EC094046 388 29.9 + no no 1 Length of longest open reading frame as a percentage of the total length of each clone * Degree of conservation of each clone, +, little to no conservation; +++++, high degree of conservation. 64 We were also interested in whether these sequenced clones derived from spliced regions of transcripts. While over half of the annotated clones appeared to be spliced, only one clone matching an EST was spliced, and none of the novel clones showed evidence of splicing. It may be that the clones matching ESTs and unannotated regions were more frequently derived from untranslated regions of transcripts. This would be consistent with the decreased size and number of ORFs in these clones. A number of families of non-coding RNAs have also now been identified, such as ribosomal RNAs, transfer RNAs, and more recently, micro RNAs. To determine if the unannotated sequences identified here may be non-coding RNAs, these sequences were queried against the Rfam database of non-coding RNA families [19]. None of the novel sequences displayed homology to any of the known RNA families in this database. Next we examined all these clones for the presence of poly-adenylation sites. Approximately 30% of all clones, regardless of their mapping, were poly-adenylated, suggesting the novel clones were derived from mRNA transcripts. This result also showed that the known and novel transcript fragments were equally likely to contain poly-adenylation sites, suggesting that novel clones were not specifically biased towards the 3' untranslated regions of transcripts. 2.3.4 Analysis of SAGE tags in the novel clones SAGE is a second technique that has been used to identify large numbers of novel transcripts [6,7]. We sought to determine if the 25 novel clones had been previously detected in SAGE experiments performed with human tissues. Subtracted clones matching unannotated regions of the human genome were searched for potential Nlalll 65 restriction sites and the 17 bp of downstream sequence was considered to represent a LongSAGE tag. Twenty of our unannotated sequences contained uniquely mapping regions which could be potential LongSAGE tags, and many clones contained multiple possible tags. We then compared our predicted LongSAGE tags to a metalibrary containing approximately 11 million uniquely mapping human SAGE tags previously identified at the B.C. Cancer Agency Genome Sciences Centre or in the Cancer Genome Anatomy Project (CGAP) at NCBI. These tags were identified in 79 SAGE libraries representing a variety of human tissues, including embryonic and haematopoietic stem cells, samples of normal and cancerous lung, brain, pancreas, breast and colon, as well as some human cancer cell lines. Only three of the 20 unannotated clones contained LongSAGE tags that were previously detected in a SAGE experiment; none of these tags mapped to annotated transcripts. These tags, from clones IcElO, 2dB01, and 2dD06, occurred once, thrice, and once, respectively, in the entire human LongSAGE metalibrary, providing in silico evidence that these clones represent novel transcripts that are expressed at very low levels. A LongSAGE tag and a novel clone may also derive from the same transcript without the two sequences directly overlapping. We therefore looked for LongSAGE tags mapping to the 50 kb of genomic DNA sequence flanking our novel sequences to identify any neighbouring regions predicted to be expressed based on the previous SAGE experiments. The genomic neighbourhoods of all 20 of the analyzed subtracted clones contained LongSAGE tags that were not from annotated transcripts. As stated above, 3 of these clones mapped directly to a LongSAGE tag. Of the remaining 17 clones, 7 mapped within 2.5 kb of a previously detected LongSAGE tag, while the remaining 10 66 clones were up to 23 kb from their nearest tag. These tags also occurred at low frequency in the human metalibrary, with each tag being detected from one to three times. These regions of the genome do not appear to encode abundantly expressed transcripts, providing a possible explanation why these transcripts were not detected previously. This analysis supported the concept that our novel subtracted clones were likely to represent entirely novel transcripts. 2.3.5 The novel subtracted clones are expressed by a variety of cell types To confirm that the novel clones detected by subtractive hybridization are naturally expressed in cells we performed a series of RT-PCR experiments. We chose 6 novel clones and examined their expression in LNCaP cells from the hollow fibre model. Al l six of the chosen clones were clearly expressed in LNCaP cells (Fig. 2.2A). The absence of bands in the reverse transcriptase (RT)-negative samples confirmed that these products did not derive from contamination with genomic DNA. The tested clones included two clones containing a previously identified LongSAGE tag, three clones for which the nearest tag was between 2 to 8 kb away, and one clone that was approximately 14 kb from the nearest LongSAGE tag. Thus, the novel transcript fragments isolated here were derived from expressed transcripts that had not previously been documented in any database. To determine the tissue distribution of these novel transcripts we performed RT-PCR with a panel of RNAs derived from "normal" human tissues. Each of the 6 clones chosen displayed unique tissue expression profiles, with 4 clones being highly specific to the prostate and testes or the testes alone, while the remaining 2 clones were expressed in 67 A CM T - CO w o> It) CO o o o ~z 6 o e o & a m CD UJ < CO T> T5 T - T - CN r - CM CM CQ ' Z"~ - 5 0 0 b p • — 1 0 0 b p RT: + - + - + - + - + . + -B w c . *•> . 2 3 re S o o m < » * ; c: o c § "5 I § S I £ zJS m o _ i O o . H 3 _ i m - 1B09 1E05 • I <sm • mm %mmm 2A03 m 1aC02 - 2dB01 2dD06 **. w j * • « • « * » « GAPDH RT: + - + - + - + - + . + - + - + C re O c T - T - C N c N f O f O Z i S Z H Z I - Z I - J Z 1B09 — — — 1E05 ** 2A03 1aC02 2dB01 2dD06 — • GAPDH Figure 2.2 - The novel subtracted clones were expressed in a variety of human tissues. (A) RT-PCR was performed with primer pairs specific for each of six novel clones using cDNA from the LNCaP Hollow Fibre Model. RT indicates the presence or absence of Reverse Transcriptase. ( B ) RT-PCR was performed using cDNAs generated from normal human tissues with each of the six novel clones. All clones were expressed in at least one of the normal tissues. (C) RT-PCR was performed using cDNAs generated from three samples of prostate cancer (Tl - T3) and their respective matched normal samples (Nl - N3). All six of the clones were expressed in at least one of these samples. 68 all the tissues tested (Fig. 2.2B). Clone 2A03 was almost exclusively expressed in the prostate, with only a faint product observed in the testes. Even the two clones observed in all the tissues tested appeared to be expressed at variable levels in each of the tissues. Thus, while the expression of some of these clones appeared higher in LNCaP cells, the expression of these novel sequences was not unique to this cell line. Next we examined the expression of these 6 novel transcripts in three samples of prostate cancer with matched samples of "normal" prostate. While the cancer specimens were not microdissected, these samples were all scored by a pathologist to contain 65-80% tumour tissue. Five of the transcripts were expressed in both the normal and cancer specimens (Fig. 2.2C). Clone 2dB01 showed the most limited expression, being detectable in only one sample of cancer. The expression of this transcript is quite limited as LNCaP cells and the single cancer specimen were the only non-testicular samples in which expression could be demonstrated (Fig. 2.2B, C). While expression levels can not be precisely measured via RT-PCR, some of these clones appeared to show differential expression in the normal samples relative to the cancer samples. From these experiments we conclude that the novel clones derived from our subtractive hybridization experiments in fact represent previously unannotated transcripts that are expressed in a variety of human tissues. 2.3.6 Two of the novel clones are part of larger untranslated regions The novel clones isolated here were cDNA fragments presumably derived from larger transcripts. We wanted to isolate more sequence information from the full-length transcripts in order to understand the possible functions of these novel transcripts. The 69 above analysis with the human SAGE metalibrary identified a number of tags mapping near the novel clones isolated using subtractive hybridization, suggesting they may represent the same transcript. RT-PCR experiments were then performed utilizing primers spanning clones 1E05 and 2dB01 and their neighbouring LongSAGE tags. These experiments demonstrated that some of these tags were in fact derived from larger transcripts that also contained these novel subtracted clones (Fig. 2.3). These PCR products were not detected in reverse transcriptase negative samples, confirming they were not derived from genomic DNA. Upon sequencing of these PCR amplicons, clone 1E05 was demonstrated to be part of an additional 6 kb untranslated region (accession nos. DQ668402 and DQ668403) downstream of a previously identified, but uncharacterized, mRNA represented in the GenBank database (AL832227). Interestingly, mRNA AL832227 was originally identified in human testis tissue. This provides validation of the tissue expression profiles we observed as clone 1E05 was only detected in cDNA from testis (Fig. 2.2B). Finally, clone 2dB01 was part of a 1.6 kb transcript that did not contain any significant ORFs (accession no. DQ668401). This transcribed region is likely an untranslated region of a larger transcript that has yet to be identified. 2.4 Discussion While the precise number of genes in the human genome remains unknown, it is clear that an even greater number of transcripts are produced by a myriad of alternative splicing events. The recognition of non-coding RNAs has also resulted in a greater focus on transcripts not containing open reading frames. Thus, even with the completed 70 A co m if) i n o O o LU HI LU r-4 kb-200 bp— RT: + - + - + B CO o CO •o CM 1.65 kb-o CO •o CM i i CM CO 200 bp-RT: + - + - + Figure 2.3 - Two of the novel clones were part of transcripts containing a neighbouring SAGE tag. RT-PCR amplified transcript regions between clones 1E05 (A) and 2dB01 (B) and SAGE tags identified nearby each clone. RT indicates the presence or absence of Reverse Transcriptase. 71 sequence of the genome further experiments are required to fully annotate the functional units transcribed from the genome in the different cell types during normal growth and development, as well as in diseased tissues. High-resolution tiling microarrays have been used to generate predictions of likely transcripts, many of which mapped to intergenic and intronic regions of the genome that were not previously annotated [20-22]. Similarly, approaches focused on sequencing full-length cDNAs continue to identify a large number of novel transcripts, many of which appear to be non-protein coding [23,24]. Our results here provide further evidence that a significant number of genes and transcripts remain to be identified. Studies now suggest that the majority of the human genome is transcribed, but the function(s) of most of these transcripts has not yet been demonstrated (reviewed in [25]). The lack of defined function for these transcripts has led some to propose that they arise indirectly through spurious transcription of the genome. However, new functions of non-coding transcripts continue to be identified, indicating that these "spurious" transcripts likely have functions that just have not been characterized. For example, steroid receptor RNA activator (SRA) was demonstrated to act as an RNA transcript to regulate the transcriptional activity of steroid nuclear receptors, including the androgen receptor [26]. Another non-coding transcript, expressed at low levels in various tissues, was recently demonstrated to regulate the nuclear trafficking of nuclear factor of T cells (NFAT), and has been renamed non-coding RNA repressor of NFAT (NRON) [27]. Such studies confirm that low-abundance, non-coding, RNA transcripts perform diverse functions and regulate multiple biological processes. 72 Subtractive hybridization has been used to characterize gene expression changes associated with prostate cancer [10,13,28-31], and some of these studies have also characterized novel genes that were identified from their subtracted libraries [10,11,28]. However, these reports examined only a few novel genes that were differentially expressed in their experimental systems, and few studies have used subtractive hybridization to examine changes with hormonal progression in an in vivo model [13]. Our study is unique in that we sequenced all the clones arising from the subtractive hybridization experiments performed with in vivo samples from the LNCaP hollow fibre model. This approach identified a large number of transcripts that had not previously been detected in prostate cancer cells and may be of prognostic or therapeutic value. We identified 25 completely unannotated clones, and an additional 57 clones matching previously sequenced ESTs that were otherwise unannotated. The isolation of these ESTs specifically from prostate cancer cells may prove informative at a later date. Furthermore, we considered any overlap of an EST or annotated gene with one of our clones to signify that that clone derived from a previously identified transcript. However, several of the clones matching ESTs and annotated transcripts displayed only partial overlap with the known sequence, suggesting that the subtracted clone may still represent an unidentified splice variant of the known transcript. The functional relevance of these novel transcripts in the hormonal progression of prostate cancer remains to be elucidated. In our experiments, 4 of the 6 clones tested were expressed only in the normal prostate and testes; this limited tissue expression profile suggests these novel transcripts may function specifically in these organs. This also suggests the expression of these transcripts is tightly regulated, as would be expected 73 for a functional transcript. Furthermore, a related publication from our group identified a novel variant of TMEFF2 which encodes a secreted form of the protein [32]. This alternate form of the protein was identified after a novel clone from our subtractive hybridization library (clone 2A06) was mapped to the fourth intron of the TMEFF2 gene. The expression of TMEFF2 has been shown to increase with progression to androgen independence [13,33], consistent with our subtractive hybridization experiments. TMEFF2 is also currently being investigated as a target for antibody-based therapy in the treatment of prostate cancer [34,35], confirming that our approach identified novel transcripts which may be of interest in the study of prostate cancer. Changes in expression of some of these transcripts may also be valuable as a marker for disease progression. To characterize the function(s) of these novel transcripts it will first be necessary to identify the full-length cDNA sequence. Multiple techniques are available to recover full-length cDNA molecules starting from only cDNA fragments. For example, Rapid Amplification of cDNA Ends (RACE) is widely used to obtain full-length cDNA sequences [36]. However, this technique is most effective when enough sequence information is available to enable the design of nested PCR primers and when the orientation of that sequence is known. We performed preliminary R A C E experiments with some of our novel subtracted clones, but the efficiency of these experiments was low as evidenced by the large number of non-specific amplicons that were produced (data not shown). This may be partially due to the inability to determine the genomic orientation of subtractive hybridization clones, necessitating the consideration of all possible transcripts when designing R A C E experiments. Another technique to identify 74 full-length sequence for these novel transcripts would be to screen existing cDNA libraries [6,11]. However, given the relatively low expression of these transcripts this approach would likely require extensive screening of such libraries. Margulies et al. [37] recently described a highly parallel sequencing by synthesis approach that demonstrated an increased throughput for sequencing of genomic DNA. Our group has combined sequencing by synthesis and random shotgun analysis to generate ESTs and characterize the transcriptome of LNCaP cells grown in tissue culture (MN Bainbridge et al, submitted). This study isolated approximately 180,000 ESTs, and of these, 1,900 (1.0%) mapped to the human genome in regions not previously annotated in the Ensembl database. One of these ESTs mapped directly to clone 2A03 described here, while 12 more mapped within 1.5 kb of clones lcD03 and 2dB01. However, the remaining 22 novel clones identified here were still not detected by this alternative approach. SAGE has also been used to identify potentially novel transcripts [6,7,38]. Unfortunately, while SAGE provides sufficient sequence information to accurately map the tags to the human genome, there is often little other information available to aid in the design of experiments to derive more sequence data from these novel transcripts. In contrast, subtractive hybridization provides longer sequence fragments, but it is not possible to determine the orientation of these fragments. Our data also suggest that subtractive hybridization was able to detect transcripts that had not previously been found using SAGE or high throughput sequencing by synthesis. It is possible that subtractive hybridization may be more sensitive to detecting transcripts expressed at low levels. Alternatively, subtractive hybridization may isolate those transcripts that can not be 75 efficiently detected by SAGE, for instance transcripts lacking an Nlalll restriction site. Common results from multiple techniques gives greater confidence in the identification of novel transcripts and underlines the value of using complementary techniques to achieve a more thorough analysis of the human transcriptome. In conclusion, our subtractive hybridization experiments have identified novel transcripts that are specific for the prostate and/or the urogenital tissues. It may be of clinical value to further develop these novel transcripts as prognostic or therapeutic markers for prostate cancer and hormonal progression. 76 2.5 References 1. Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ, Thun MJ. Cancer statistics, 2005. C A Cancer J Clin 2005;55(1): 10-30. 2. Pirtskhalaishvili G, Hrebinko RL, Nelson JB. The treatment of prostate cancer: an overview of current options. Cancer Pract 2001;9(6):295-306. 3. Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi K N , Oudard S, Theodore C, James ND, Turesson I, Rosenthal M A , Eisenberger M A . Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004;351(15): 1502-1512. 4. Petrylak DP, Tangen C M , Hussain M H , Lara PN, Jr., Jones JA, Taplin M E , Burch PA, Berry D, Moinpour C, Kohli M , Benson M C , Small EJ, Raghavan D, Crawford ED. Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004;351(15):1513-1520. 5. Sadar MD, Akopian V A , Beraldi E. Characterization of a New in Vivo Hollow Fiber Model for the Study of Progression of Prostate Cancer to Androgen Independence. Mol Cancer Ther 2002;l(8):629-637. 6. Velculescu V E , Zhang L, Vogelstein B, Kinzler KW. Serial analysis of gene expression. Science 1995;270(5235):484-487. 7. Saha S, Sparks AB, Rago C, Akmaev V, Wang CJ, Vogelstein B, Kinzler KW, Velculescu V E . Using the transcriptome to annotate the genome. Nat Biotechnol 2002;20(5):508-512. 77 8. Diatchenko L, Lau YF, Campbell AP, Chenchik A, Moqadam F, Huang B, Lukyanov S, Lukyanov K, Gurskaya N, Sverdlov ED, Siebert PD. Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc Natl Acad Sci U S A 1996;93(12):6025-6030. 9. Gurskaya NG, Diatchenko L, Chenchik A, Siebert PD, Khaspekov GL, Lukyanov K A , Vagner L L , Ermolaeva OD, Lukyanov SA, Sverdlov ED. Equalizing cDNA subtraction based on selective suppression of polymerase chain reaction: cloning of Jurkat cell transcripts induced by phytohemaglutinin and phorbol 12-myristate 13-acetate. Anal Biochem 1996;240(l):90-97. 10. Xu J, Stolk JA, Zhang X, Silva SJ, Houghton RL, Matsumura M , Vedvick TS, Leslie K B , Badaro R, Reed SG. Identification of differentially expressed genes in human prostate cancer using subtraction and microarray. Cancer Res 2000;60(6): 1677-1682. 11. Xu J, Kalos M , Stolk JA, Zasloff EJ, Zhang X, Houghton RL, Filho A M , Nolasco M , Badaro R, Reed SG. Identification and characterization of prostein, a novel prostate- specific protein. Cancer Res 2001;61(4): 1563-1568. 12. Cao W, Epstein C, Liu H, DeLoughery C, Ge N, Lin J, Diao R, Cao H, Long F, Zhang X, Chen Y, Wright PS, Busch S, Wenck M , Wong K, Saltzman A G , Tang Z, Liu L, Zilberstein A. Comparing gene discovery from Affymetrix GeneChip microarrays and Clontech PCR-select cDNA subtraction: a case study. B M C Genomics 2004;5(1):26. 78 13. Mohler JL, Morris TL, Ford OH, 3rd, Alvey RF, Sakamoto C, Gregory CW. Identification of differentially expressed genes associated with androgen-independent growth of prostate cancer. Prostate 2002;51(4):247-255. 14. Chen CD, Welsbie DS, Tran C, Baek SH, Chen R, Vessella R, Rosenfeld M G , Sawyers CL. Molecular determinants of resistance to antiandrogen therapy. Nat Med 2004;10(l):33-39. 15. Holzbeierlein J, Lai P, LaTulippe E, Smith A, Satagopan J, Zhang L, Ryan C, Smith S, Scher H, Scardino P, Reuter V, Gerald WL. Gene expression analysis of human prostate carcinoma during hormonal therapy identifies androgen-responsive genes and mechanisms of therapy resistance. Am J Pathol 2004;164(l):217-227. 16. Hubbard T, Andrews D, Caccamo M , Cameron G, Chen Y, Clamp M , Clarke L, Coates G, Cox T, Cunningham F, Curwen V, Cutts T, Down T, Durbin R, Fernandez-Suarez X M , Gilbert J, Hammond M , Herrero J, Hotz H, Howe K, Iyer V, Jekosch K, Kahari A, Kasprzyk A, Keefe D, Keenan S, Kokocinsci F, London D, Longden I, McVicker G, Melsopp C, Meidl P, Potter S, Proctor G, Rae M , Rios D, Schuster M , Searle S, Severin J, Slater G, Smedley D, Smith J, Spooner W, Stabenau A, Stalker J, Storey R, Trevanion S, Ureta-Vidal A, Vogel J, White S, Woodwark C, Birney E. Ensembl 2005. Nucleic Acids Res 2005;33(Database issue):D447-453. 17. Kent WJ, Sugnet CW, Furey TS, Roskin K M , Pringle TH, Zahler A M , Haussler D. The human genome browser at UCSC. Genome Res 2002;12(6):996-1006. 79 18. Siepel A, Bejerano G, Pedersen JS, Hinrichs AS, Hou M , Rosenbloom K, Clawson H, Spieth J, Hillier LW, Richards S, Weinstock G M , Wilson RK, Gibbs RA, Kent WJ, Miller W, Haussler D. Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res 2005;15(8):1034-1050. 19. Griffiths-Jones S, Moxon S, Marshall M , Khanna A, Eddy SR, Bateman A. Rfam: annotating non-coding RNAs in complete genomes. Nucleic Acids Res 2005;33(Database issue):D121-124. 20. Kampa D, Cheng J, Kapranov P, Yamanaka M , Brubaker S, Cawley S, Drenkow J, Piccolboni A, Bekiranov S, Helt G, Tammana H, Gingeras TR. Novel RNAs identified from an in-depth analysis of the transcriptome of human chromosomes 21 and 22. Genome Res 2004;14(3):331-342. 21. Bertone P, Stole V, Royce TE, Rozowsky JS, Urban A E , Zhu X, Rinn JL, Tongprasit W, Samanta M , Weissman S, Gerstein M , Snyder M . Global identification of human transcribed sequences with genome tiling arrays. Science 2004;306(5705):2242-2246. 22. Cheng J, Kapranov P, Drenkow J, Dike S, Brubaker S, Patel S, Long J, Stern D, Tammana H, Helt G, Sementchenko V, Piccolboni A, Bekiranov S, Bailey DK, Ganesh M , Ghosh S, Bell I, Gerhard DS, Gingeras TR. Transcriptional maps of 10 human chromosomes at 5-nucleotide resolution. Science 2005;308(5725):1149-1154. 23. Imanishi T, Itoh T, Suzuki Y, O'Donovan C, Fukuchi S, Koyanagi K O , Barrero RA, Tamura T, Yamaguchi-Kabata Y, Tanino M , Yura K, Miyazaki S, Ikeo K, Homma K, Kasprzyk A, Nishikawa T, Hirakawa M , Thierry-Mieg J, Thierry-80 Mieg D, Ashurst J, Jia L, Nakao M , Thomas M A , Mulder N, Karavidopoulou Y , Jin L, Kim S, Yasuda T, Lenhard B, Eveno E, Suzuki Y, Yamasaki C, Takeda J, Gough C, Hilton P, Fujii Y, Sakai H, Tanaka S, Amid C, Bellgard M , Bonaldo Mde F, Bono H, Bromberg SK, Brookes AJ, Bruford E, Carninci P, Chelala C, Couillault C, de Souza SJ, Debily MA, Devignes MD, Dubchak I, Endo T, Estreicher A, Eyras E, Fukami-Kobayashi K, Gopinath GR, Graudens E, Hahn Y, Han M , Han ZG, Hanada K, Hanaoka H, Harada E, Hashimoto K, Hinz U, Hirai M , Hishiki T, Hopkinson I, Imbeaud S, Inoko H, Kanapin A, Kaneko Y, Kasukawa T, Kelso J, Kersey P, Kikuno R, Kimura K, Korn B, Kuryshev V , Makalowska I, Makino T, Mano S, Mariage-Samson R, Mashima J, Matsuda H, Mewes HW, Minoshima S, Nagai K, Nagasaki H, Nagata N, Nigam R, Ogasawara O, Ohara O, Ohtsubo M , Okada N, Okido T, Oota S, Ota M , Ota T, Otsuki T, Piatier-Tonneau D, Poustka A, Ren SX, Saitou N, Sakai K, Sakamoto S, Sakate R, Schupp I, Servant F, Sherry S, Shiba R, Shimizu N, Shimoyama M , Simpson AJ, Soares B, Steward C, Suwa M , Suzuki M , Takahashi A, Tamiya G, Tanaka H, Taylor T, Terwilliger JD, Unneberg P, Veeramachaneni V, Watanabe S, Wilming L, Yasuda N, Yoo HS, Stodolsky M , Makalowski W, Go M , Nakai K, Takagi T, Kanehisa M , Sakaki Y, Quackenbush J, Okazaki Y, Hayashizaki Y, Hide W, Chakraborty R, Nishikawa K, Sugawara H, Tateno Y, Chen Z, Oishi M , Tonellato P, Apweiler R, Okubo K, Wagner L, Wiemann S, Strausberg RL, Isogai T, Auffray C, Nomura N, Gojobori T, Sugano S. Integrative annotation of 21,037 human genes validated by full-length cDNA clones. PLoS Biol 2004;2(6):el62. 24. Dalla E, Mignone F, Verardo R, Marchionni L, Marzinotto S, Lazarevic D, Reid JF, Marzio R, Klaric E, Licastro D, Marcuzzi G, Gambetta R, Pierotti M A , Pesole G, Schneider C. Discovery of 342 putative new genes from the analysis of 5'-end-sequenced full-length-enriched cDNA human transcripts. Genomics 2005;85(6):739-751. 25. Willingham AT, Gingeras TR. TUF love for "junk" DNA. Cell 2006;125(7):1215-1220. 26. Lanz RB, McKenna NJ, Onate SA, Albrecht U, Wong J, Tsai SY, Tsai MJ, O'Malley BW. A steroid receptor coactivator, SRA, functions as an RNA and is present in an SRC-1 complex. Cell 1999;97(1): 17-27. 27. Willingham AT, Orth AP, Batalov S, Peters EC, Wen BG, Aza-Blanc P, Hogenesch JB, Schultz PG. A strategy for probing the function of noncoding RNAs finds a repressor of NFAT. Science 2005;309(5740):1570-1573. 28. Hubert RS, Vivanco I, Chen E, Rastegar S, Leong K, Mitchell SC, Madraswala R, Zhou Y, Kuo J, Raitano AB, Jakobovits A, Saffran DC, Afar DE. STEAP: a prostate-specific cell-surface antigen highly expressed in human prostate tumors. Proc Natl Acad Sci U S A 1999;96(25): 14523-14528. 29. Stubbs AP, Abel PD, Golding M , Bhangal G, Wang Q, Waxman J, Stamp GW, Lalani EN. Differentially expressed genes in hormone refractory prostate cancer: association with chromosomal regions involved with genetic aberrations. Am J Pathol 1999; 154(5): 1335-1343. 82 30. Porkka KP, Visakorpi T. Detection of differentially expressed genes in prostate cancer by combining suppression subtractive hybridization and cDNA library array. J Pathol 2001;193(l):73-79. 31. Chetcuti A, Margan SH, Russell P, Mann S, Millar DS, Clark SJ, Rogers J, Handelsman DJ, Dong Q. Loss of annexin II heavy and light chains in prostate cancer and its precursors. Cancer Res 2001;61(17):6331-6334. 32. Quayle SN, Sadar MD. A truncated isoform of TMEFF2 encodes a secreted protein in prostate cancer cells. Genomics 2006;87(5):633-637. 33. Glynne-Jones E, Harper M E , Seery LT, James R, Anglin I, Morgan HE, Taylor K M , Gee JM, Nicholson RI. TENB2, a proteoglycan identified in prostate cancer that is associated with disease progression and androgen independence. Int J Cancer 2001;94(2):178-184. 34. Afar DE, Bhaskar V, Ibsen E, Breinberg D, Henshall SM, Kench JG, Drobnjak M , Powers R, Wong M , Evangelista F, O'Hara C, Powers D, DuBridge RB, Caras I, Winter R, Anderson T, Solvason N, Strieker PD, Cordon-Cardo C, Scher HI, Grygiel JJ, Sutherland RL, Murray R, Ramakrishnan V, Law DA. Preclinical validation of anti-TMEFF2-auristatin E-conjugated antibodies in the treatment of prostate cancer. Mol Cancer Ther 2004;3(8):921-932. 35. Zhao X Y , Schneider D, Biroc SL, Parry R, Alicke B, Toy P, Xuan JA, Sakamoto C, Wada K, Schulze M , Muller-Tiemann B, Parry G, Dinter H. Targeting tomoregulin for radioimmunotherapy of prostate cancer. Cancer Res 2005;65(7):2846-2853. 83 36. Frohman M A , Dush M K , Martin GR. Rapid production of full-length cDNAs from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc Natl Acad Sci U S A 1988;85(23):8998-9002. 37. Margulies M , Egholm M , Altman WE, Attiya S, Bader JS, Bemben L A , Berka J, Braverman MS, Chen YJ, Chen Z, Dewell SB, Du L, Fierro JM, Gomes X V , Godwin BC, He W, Helgesen S, Ho CH, Irzyk GP, Jando SC, Alenquer M L , Jarvie TP, Jirage KB, Kim JB, Knight JR, Lanza JR, Leamon JH, Lefkowitz SM, Lei M , Li J, Lohman K L , Lu H, Makhijani VB, McDade K E , McKenna MP, Myers EW, Nickerson E, Nobile JR, Plant R, Puc BP, Ronan MT, Roth GT, Sarkis GJ, Simons JF, Simpson JW, Srinivasan M , Tartaro KR, Tomasz A, Vogt K A , Volkmer GA, Wang SH, Wang Y, Weiner MP, Yu P, Begley RF, Rothberg JM. Genome sequencing in microfabricated high-density picolitre reactors. Nature 2005;437(7057):376-380. 38. Wahl MB, Heinzmann U, Imai K. LongSAGE analysis significantly improves genome annotation: identifications of novel genes and alternative transcripts in the mouse. Bioinformatics 2005;21(8): 1393-1400. 84 3 A TRUNCATED ISOFORM OF TMEFF2 ENCODES A SECRETED PROTEIN IN PROSTATE CANCER CELLS* 3.1 Introduction The human TMEFF2 (transmembrane protein with EGF-like and two follistatin-like domains 2) gene was initially described by a number of independent groups [1-6]. Together, these reports described TMEFF2 as a proteoglycan predominately expressed in the brain and prostate. TMEFF2 was also shown to be hypermethylated in bladder and colorectal tumour cells. While its full name illustrates its predicted domains, a physiological role for TMEFF2 has not been determined. Overexpression of full-length TMEFF2 inhibited the proliferation of two aggressive prostate cancer cell lines [6] while the soluble extracellular domain of TMEFF2 increased survival of neuronal cells in primary cultures [3]. Additionally, the TMEFF2 extracellular domain elicited weak tyrosine phosphorylation of erbB-4 [1]. TMEFF2 has received recent attention as a possible immunotherapeutic target for the treatment of metastatic prostate cancer [7,8]. This interest is due to its restricted expression profile as well as its increased expression in androgen-independent prostate cancer [5,9]. We report here the identification of a novel splice variant of TMEFF2 that encodes a secreted protein (TMEFF2-S). 3.2 Materials and methods 3.2.1 Cell culture and transfection LNCaP cells obtained from Dr. L.W.K. Chung (Emory University School of Medicine, Atlanta, GA) were maintained in RPMI 1640 supplemented with 5% (v/v) fetal * A version of this chapter has been published. Quayle, SN, Sadar M D . A truncated isoform of TMEFF2 encodes a secreted protein in prostate cancer cells. Genomics 2006;87(5):633-637. 85 bovine serum (FBS) (FfyClone, Logan, UT), 100 units/mL penicillin, and 100 ug/uL streptomycin (Invitrogen, Burlington, Canada). All chemicals were purchased from Sigma, unless stated otherwise. Androgen stimulation was performed by treatment with the synthetic androgen R1881 (10 nM, PerkinElmer, Woodbridge, Canada) for 24hr. For transfections, LNCaP cells were plated at 3.0xl05 cells per well in 6-well plates. After 24 hr, transfections were performed using LIPOFECTIN® Reagent (Invitrogen) according to the manufacturer's protocol. Plasmids encoding full-length or truncated TMEFF2 were transfected at 1 ug and 3 ug/well in separate wells. The total amount of transfected DNA was normalized to 3 ug/well by the addition of pRc-CMV (Invitrogen) plasmid encoding the C M V promoter but lacking a downstream gene. After 24 hr RPMI 1640 was added to the cells to a final concentration of 5% FBS. 3.2.2 RT-PCR Total RNA was isolated from LNCaP cells maintained in vivo [10] or in vitro, PrEC cells (Clonetics, Guelph, Canada), or PC3 prostate cancer cells using TRIZOL® Reagent (Invitrogen) according to the manufacturer's protocol. Total RNA samples from several normal human tissues were purchased from Stratagene (La Jolla, CA). Total RNA from clinical specimens of prostate cancer, and their matched normal samples, was purchased from Genomics Collaborative (Cambridge, MA). Reverse transcription (RT) was performed using M M L V - R T (Invitrogen) with 1 ug of template RNA. Subsequent PCR reactions were performed using 1 uL of the resulting cDNA as template. Primers used to amplify TMEFF2 and EST 2A06 were as follows: Primer 1, 5'-C C C G T C A T G C T A C T C A T C G T - 3 ' ; Primer 2, 5'-86 C T G A A C A A A A G A G T G A A G G G A A C - 3 ' ; Primer 3, 5'-C C C T T G T G C A A A T G G G T T A - 3 ' ; and Primer 4, 5 ' - A A T G C T C A C A C T T C C C A T G C -3'. The primers used to detect GAPDH were: Forward 5'-C C G A G C C A C A T C G C T C A G A - 3 ' and Reverse 5 ' -CCCAGCCTTCTCCATGGTG-3' . 3.2.3 Plasmids A Mammalian Gene Collection expression clone (Clone ID 4180709) containing the entire transcript for TMEFF2 was purchased from Invitrogen. The open-reading frames of full-length and truncated TMEFF2 were cloned into pcDNA3.1™D/V5-His-TOPO® vector (Invitrogen) using the forward primer 5'-C A C C A T G G T G C T G T G G G A G T C C - 3 ' ; TMEFF2-reverse 5'-G A T T A A C C T C G T G G A C G C T C - 3 ' ; and TMEFF2-S-reverse 5'-A A T A T C T G A A A T G G A A C A T A C - 3 ' . The same forward primer was used to amplify both open reading frames. These inserts were then removed by digesting with HindJR and EcoKV restriction endonucleases, followed by ligation into the Hindlll and Smal sites of the pEGFP-N3 vector. All plasmids were sequence verified. 3.2.4 Western blot analysis LNCaP cells were harvested and washed twice in ice-cold PBS. Cell pellets were lysed in 2x SDS-PAGE loading buffer and boiled for five minutes prior to loading on 12% SDS-PAGE gels. Media from transfected cells was removed after incubation for 24 hours, centrifuged to remove unattached cells, and 25 uL samples were analyzed on 12% SDS-PAGE gels. The separated proteins were transferred to nitrocellulose membranes 87 (Amersham Biosciences, Piscataway, NJ) and blocked in Odyssey Blocking Buffer (LI-COR Biosciences, Lincoln, NE) for one hour at room temperature or at 4°C overnight, and then washed three times. The blots were probed with anti-GFP (B-2)(l :500)(Santa Cruz Biotechnology) or anti-P-actin (ab6276)(l:5000)(Abcam, Cambridge, MA) for 1 to 2 hours at room temperature, washed three times, and incubated with the appropriate IR-labelled secondary antibody (1:5000, Molecular Probes) for 30 minutes at room temperature. All washes were performed with PBS containing 0.1% Tween20. Primary antibodies were diluted in Odyssey Blocking Buffer with 0.1% Tween20, while secondary antibody dilutions contained an additional 0.01% SDS. Visualization was performed with the Odyssey Infrared Imaging System (LI-COR Biosciences). 3.2.5 Fluorescent microscopy LNCaP cells (3.0x105 cells per well) were plated on sterile coverslips in 6-well plates. Transfections were performed as described above. The cells were fixed for 30 minutes in 4% paraformaldehyde at room temperature, followed by two washes in sterile PBS. Nuclei were counterstained with 300 nM 4',6-Diamidino-2-phenylindole (DAPI) for 5 minutes at room temperature. The cells were again rinsed in sterile PBS prior to mounting on glass slides. Images were captured using an AxioCam high resolution colour camera on an AxioPlan 2 fluorescent microscope (Carl Zeiss, Toronto, Canada). Images of the GFP and DAPI channels were merged using Adobe Photoshop software. 88 3.3 Results 3.3.1 A novel splice variant is transcribed from the TMEFF2 locus Previous experiments in our laboratory have generated a collection of novel expressed sequence tags (EST; see Chapter 2) derived from a LNCaP xenograft model of prostate cancer [10]. Sequencing of one such EST (referred to as EST 2A06) revealed a 330bp sequence with 99.1% homology to a region of chromosome 2q32.3. Upon further analysis, EST 2A06 was mapped to the fourth intron of the gene TMEFF2 (also known as tomoregulin-2, HPP1, or TENB2; Fig. 3.1 A). RT-PCR with primers anchored in the known transcript and EST 2A06 amplified an unidentified product of approximately 3.3 kb, indicating EST 2A06 represented a novel splice variant of the TMEFF2 gene (Fig. 3. IB). However, the cDNA identified did not include the 3' region of the known TMEFF2 transcript. Sequencing of the 3.3 kb band revealed a cDNA encoding the first four exons of TMEFF2 followed by 2.9 kb of the fourth intron (GenBank accession number DQ133599). A predicted translation of this cDNA suggested it encoded a 175 amino acid open reading frame (ORF) consisting of the first 146 residues of TMEFF2 followed by 29, mostly hydrophobic, residues not found in the normal TMEFF2 protein sequence. These 29 novel amino acids (GRSCLFTYLKIYWWILLCIFTYVCSISDI) did not exhibit significant similarity to any proteins in GenBank. Significantly, this predicted isoform, designated TMEFF2-S, lacked one follistatin-like domain, the EGF-like domain, and the transmembrane domain (Fig. 3.1C). 89 TMEFF2 Locus EST 2A06 Primers 2 3 4 ^ CO CO "if ^ t • I t 1 T - CM r- rs CD 4.0 kb — ~~ 850 b p — S B ^ _ 200 b p — * • » c S S F S F S E G F T M jf—|l TMEFF2 S S F S N TMEFF2-S Figure 3.1 - A novel EST is part of a larger transcript derived from the TMEFF2 locus. (A) Schematic cartoon summarizing the TMEFF2 genomic locus (transcribed from left to right) and the relative location of the sequenced EST 2A06 in the fourth intron (not to scale). Horizontal lines represent introns, vertical lines represent exons, and filled boxes represent untranslated regions (UTRs). Numbered arrows represent the locations of the primers used in (B). (B) Results of RT-PCR experiment using the primer pairs from (A) indicated above each lane. These results suggest EST 2A06 and TMEFF2 share a single transcript. (C) Schematic cartoon showing the domain structure of the peptide predicted from the sequence of the novel splice variant. SS, signal sequence; FS, follistatin-like domain; EGF, EGF-like domain; TM, transmembrane domain; N, novel amino acid residues. 90 3.3.2 TMEFF2-S was expressed in normal human tissues and produced a truncated protein To determine the tissue distribution of TMEFF2-S, we performed RT-PCR with cDNA samples obtained from the prostatic cell lines LNCaP, PrEC, and PC3, from normal human tissues, and from samples of prostate cancer and their matched normal samples (Fig. 3.2). While the prostate cancer specimens were not microdissected, they did contain 65-80% tumour tissue. Additionally, tumour sample T3 expressed the TMEFF2-S transcript while the matched normal sample N3 showed little to no expression of this transcript, suggesting that in this case TMEFF2-S was only expressed by the cancer cells (Fig. 3.2C). No bands were detected in the reverse-transcriptase (RT) negative samples, indicating this product was not due to contamination with genomic DNA. Identical results were obtained from PCR with two other primer pairs located in different regions of the TMEFF2-S transcript. Expression of this isoform was limited to LNCaP cells, normal prostate, and testis. Importantly, TMEFF2-S expression was also observed in prostate cancer specimens. To demonstrate that this splice variant produced a truncated protein as predicted from its cDNA sequence, GFP was fused to the carboxyl terminus of both full-length TMEFF2 and TMEFF2-S. Each isoform was transfected into LNCaP cells and expression of the fusion proteins monitored by western blot analysis of whole cell protein lysates using an anti-GFP antibody. As predicted, the full-length TMEFF2-GFP fusion protein was approximately 75 kDa (Fig. 3.3); there were, however, other products of this fusion protein, most likely due to post-translational modifications. The extracellular domain of TMEFF2 was previously shown to be cleaved at multiple sites, leaving small 91 00 oo JS or ,c CO + Q- Q. 0. co ro re O O O Z z z _J _ l _ l EST 2A06 LU m 0. QL 500bp— -EST 2A06 100bp— GAPDH RT: + - + - + - + - + -B e o I > s s a « m o _i o o. K 5 cQ 500bp— 100bp— m - « - - » _ GAPDH RT: + - + - + - + - + - + - + -CL re O = Z H Z H Z H _ l 4 a 500bP— _ * EST 2A06 i o o b P - "* GAPDH Figure 3.2 - The novel splice variant of T M E F F 2 is expressed in normal tissues. R T - P C R was performed with primers located within EST 2A06 using c D N A from L N C a P , P r E C , and PC3 cell lines (A) , from normal human c D N A s (B), and from three samples of prostate cancer (TI , T2, T3) and their respective matched normal samples ( N l , N 2 , N3) (C). R T - P C R for G A P D H confirmed the presence and quality of c D N A in each sample. The novel T M E F F 2 transcript was expressed specifically in prostate-derived tissues. A l l images are representative of multiple experiments. 92 •a OJ CO c CO 9 CN LL LL 111 Q-LL o I to i OJ LL LL LU U a i M g 3MS i p g 3 p g 7 5 k D a -5 0 k D a -3 7 k D a -2 5 k D a -2 0 k D a -<«-TMEFF2-GFP ••-TMEFF2-S-GFP ^c leaved TMEFF2-GFP «-GFP ^-cleaved GFP •p Actin Figure 3.3 - The novel splice variant of TMEFF2 produces a truncated protein. Immu-noblot demonstrating expression of the GFP fusion protein constructs. Whole cell lysates from LNCaP cells not transfected or transfected with pRc-CMV, TMEFF2-GFP, or TMEFF2-S-GFP plasmids were immunoblotted using an anti-GFP antibody. Probing for P-actin confirmed the amount and quality of protein in each lane. Expression of TMEFF2-S produced a single 50kDa protein. Identical results were obtained in replicate experiments; a representative result is shown. 93 portions of the full-length protein in the plasma membrane [1,5]. This protein is also known to be glycosylated, thus altering its predicted migration pattern in acrylamide gels [1,5]. In contrast, the TMEFF2-S-GFP fusion protein produced a single molecular weight species of approximately 50 kDa, close to the predicted size of 48 kDa. A doublet band of about 27 kDa probably represents free GFP produced from internal ribosome entry into the fusion transcripts as this band was present regardless of the fusion protein being expressed. An unexpected band of 21 kDa was also seen. The identity of this band is unclear, but probably represents a cleavage product of the GFP protein as the antibody used was raised against the full-length GFP sequence. These results confirm that the TMEFF2-S transcript identified does encode a truncated protein. 3.3.3 The truncated isoform of TMEFF2 produced a secreted protein The predicted protein sequence of TMEFF2-S lacks a transmembrane domain (Fig. 3.1C); this may alter its subcellular localization as compared with full length TMEFF2. To determine this, LNCaP cells transfected with the GFP fusion proteins were examined by fluorescence microscopy (Fig. 3.4A). As predicted, cells transfected with TMEFF2-GFP showed diffuse cytoplasmic localization with strong signal in the plasma membrane (Fig. 3.4A, top panels). In contrast, cells expressing the TMEFF2-S-GFP splice variant showed no signal in the plasma membrane (Fig. 3.4A, bottom panels). All TMEFF2-S-GFP fluorescence was observed in bright foci in the cytoplasm, again in contrast with the diffuse cytoplasmic signal seen with full-length TMEFF2. Thus, the subcellular localization of TMEFF2-S was distinct from TMEFF2. 94 Figure 3.4 - The truncated variant of TMEFF2 is not localized in the plasma membrane but is secreted by transfected cells. (A) Fluorescent micrographs showing subcellular localization of GFP fusion proteins. LNCaP cells transfected with either TMEFF2-GFP (top row) or TMEFF2-S-GFP (bottom row) constructs were fixed and counterstained with DAPI prior to visualization by fluorescence microscopy. The left panel shows GFP signal, the middle panel shows DAPI-stained nuclei, and the right panel shows the overlay of these images. TMEFF2-S-GFP was not present in the plasma membrane, but instead all GFP signal was present in discreet foci in the cytoplasm. All images were captured at 63 OX magnification. (B) Immunoblot demonstrating the presence of the TMEFF2-S-GFP fusion protein in the media of transfected cells. LNCaP cells were not transfected or transfected with pRc-CMV, TMEFF2-GFP, or TMEFF2-S-GFP. All images are representative of multiple experiments. 96 Given that the truncated TMEFF2 protein still contained the putative signal sequence while lacking the transmembrane region, we proposed to determine whether the truncated TMEFF2 protein was secreted. Western blot analysis using anti-GFP antibody detected a distinct band of approximately 50 kDa only in the media of cells expressing the TMEFF2-S-GFP protein (Fig. 3.4B). The intensity of this band increased with increasing amounts of plasmid DNA transfected (Fig. 3.4B). The TMEFF2-S-GFP protein detected in the media migrated the same distance as TMEFF2-S-GFP in whole cell lysates. No GFP-reactive proteins were seen in the media of cells expressing full-length TMEFF2 or in culture media. The secretion of TMEFF2-S suggests this isoform may have a function distinct from that of TMEFF2. 3.4 Discussion The expression of splice variants from a single genomic locus allows for greater diversity of gene products without the need for enlarging the entire genome. Alternative splicing events have been associated with a number of genetic diseases, including cancer [11]. Often, these splice variants alter the function of the other gene products expressed from that locus [12,13]. There are multiple examples of transmembrane proteins with splice variants lacking the transmembrane domain, resulting in a secreted isoform of the protein [13,14]. This is the first report to describe a secreted isoform of TMEFF2. Full-length TMEFF2 is expressed predominately in the brain and in the prostate, with even stronger expression seen in prostate cancer [3,5,7]. In agreement with these data, expression of TMEFF2-S was detected in LNCaP prostate cancer cells, normal and cancerous human prostate tissue, and weakly in human testis tissue. Previous studies 97 have detected an unidentified minor band of approximately 3.5 kb when probing for TMEFF2 by Northern blot analysis [3,5,6]. The probes used in those experiments would overlap with the novel splice variant described here, and thus the 3.5 kb band observed may represent this truncated form of TMEFF2. Full-length TMEFF2 expression appears to be androgen-regulated, and also repressed by the c-Myc protein [5,6,15], but whether the truncated isoform is also regulated in this manner is not yet known. It will be of interest to assess whether the expression level of TMEFF2-S changes with the progression of prostate cancer, as this is seen for full-length TMEFF2 [6]. Intriguingly, the expression of the minor 3.5 kb band decreased after castration in the CWR22 model, but the kinetics of this decrease appeared different from that of full-length TMEFF2 [6]. The expression patterns of these isoforms in both normal and cancerous tissues needs to be assessed to help elucidate their potential roles in carcinogenesis. The precise function(s) of TMEFF2 need to be examined further, as does the relationship of these isoforms. It will be interesting to determine if TMEFF2-S functionally interacts with TMEFF2 through co-operative signalling events or competitive binding. A thorough examination of the different binding partners of the various forms of TMEFF2 may help in this regard. Expression of full-length TMEFF2 in the prostate is restricted to the luminal epithelial cells [7,8]. Given the similarities in their expression profiles it is likely both proteins are expressed in this cell type, suggesting these isoforms may have roles in the normal secretory functions of the prostate. The extracellular domain of TMEFF2 can be cleaved, resulting in at least two different soluble proteins [1,5]. This cleaved form of TMEFF2 may be responsible for its functional effects and the soluble TMEFF2-S protein may then negatively regulate, or 98 synergize with, cleaved-TMEFF2. One form of cleaved-TMEFF2 produces a soluble protein with both follistatin-like (FS) domains intact, but without the EGF-like domain, suggesting these FS domains act in concert. Follistatin regulates signalling downstream of several members of the transforming growth factor-P (TGF-P) family of growth factor ligands. In contrast to full-length TMEFF2, TMEFF2-S contains only one FS domain. The absence of one FS domain may then enhance, or inhibit, the activity of this secreted isoform relative to the cleaved-TMEFF2 peptide. Additionally, the loss of the EGF-like domain implies TMEFF2-S would not induce phosphorylation of erbB4, in contrast to TMEFF2 [1]. The delineation of the functions of these isoforms will be made simpler once a reliable screen for TMEFF2 activity has been established. Recently, two groups have published reports proposing the use of antibodies directed against TMEFF2 in the treatment of prostate cancer [7,8]. While both studies reported favourable results, the presence of a soluble form of TMEFF2 may not have been considered when these antibodies were designed. Clearly, further studies will be required to examine the circulating levels of TMEFF2-S in these patients and whether the secreted isoform may cross-react with the antibodies used. If so, the TMEFF2-S protein may decrease the efficacy of this approach by competitively binding the antibody away from the tumour cells. Alternatively, targeting of TMEFF2-S may provide added benefit to patients by altering the normal function of this novel isoform. 99 References Uchida T, Wada K, Akamatsu T, Yonezawa M , Noguchi H, Mizoguchi A, Kasuga M , Sakamoto C. A novel epidermal growth factor-like molecule containing two follistatin modules stimulates tyrosine phosphorylation of erbB-4 in MKN28 gastric cancer cells. Biochem Biophys Res Commun 1999;266(2):593-602. Liang G, Robertson KD, Talmadge C, Sumegi J, Jones PA. The gene for a novel transmembrane protein containing epidermal growth factor and follistatin domains is frequently hypermethylated in human tumor cells. Cancer Res 2000;60(17):4907-4912. Horie M , Mitsumoto Y, Kyushiki H, Kanemoto N, Watanabe A, Taniguchi Y, Nishino N, Okamoto T, Kondo M , Mori T, Noguchi K, Nakamura Y , Takahashi E, Tanigami A. Identification and characterization of TMEFF2, a novel survival factor for hippocampal and mesencephalic neurons. Genomics 2000;67(2):146-152. Young J, Biden K G , Simms L A , Huggard P, Karamatic R, Eyre HJ, Sutherland GR, Herath N, Barker M , Anderson GJ, Fitzpatrick DR, Ramm GA, Jass JR, Leggett BA. HPP1: a transmembrane protein-encoding gene commonly methylated in colorectal polyps and cancers. Proc Natl Acad Sci U S A 2001;98(l):265-270. Glynne-Jones E, Harper M E , Seery LT, James R, Anglin I, Morgan HE, Taylor K M , Gee JM, Nicholson RL TENB2, a proteoglycan identified in prostate cancer that is associated with disease progression and androgen independence. Int J Cancer 2001;94(2): 178-184. 6. Gery S, Sawyers CL, Agus DB, Said JW, Koeffler HP. TMEFF2 is an androgen-regulated gene exhibiting antiproliferative effects in prostate cancer cells. Oncogene 2002;21(31):4739-4746. 7. Afar DE, Bhaskar V, Ibsen E, Breinberg D, Henshall SM, Kench JG, Drobnjak M , Powers R, Wong M , Evangelista F, O'Hara C, Powers D, DuBridge RB, Caras I, Winter R, Anderson T, Solvason N, Strieker PD, Cordon-Cardo C, Scher HI, Grygiel JJ, Sutherland RL, Murray R, Ramakrishnan V, Law DA. Preclinical validation of anti-TMEFF2-auristatin E-conjugated antibodies in the treatment of prostate cancer. Mol Cancer Ther 2004;3(8):921-932. 8. Zhao X Y , Schneider D, Biroc SL, Parry R, Alicke B, Toy P, Xuan JA, Sakamoto C, Wada K, Schulze M , Muller-Tiemann B, Parry G, Dinter H. Targeting tomoregulin for radioimmunotherapy of prostate cancer. Cancer Res 2005;65(7):2846-2853. 9. Mohler JL, Morris TL, Ford OH, 3rd, Alvey RF, Sakamoto C, Gregory CW. Identification of differentially expressed genes associated with androgen-independent growth of prostate cancer. Prostate 2002;51(4):247-255. 10. Sadar MD, Akopian V A , Beraldi E. Characterization of a New in Vivo Hollow Fiber Model for the Study of Progression of Prostate Cancer to Androgen Independence. Mol Cancer Ther 2002;l(8):629-637. 11. Faustino NA, Cooper TA. Pre-mRNA splicing and human disease. Genes Dev 2003;17(4):419-437. 101 12. Chew LJ, Pan H, Yu J, Tian S, Huang WQ, Zhang JY, Pang S, Li L Y . A novel secreted splice variant of vascular endothelial cell growth inhibitor. Faseb J 2002;16(7):742-744. 13. Yonekura H, Yamamoto Y, Sakurai S, Petrova RG, Abedin MJ, Li H, Yasui K, Takeuchi M , Makita Z, Takasawa S, Okamoto H, Watanabe T, Yamamoto H. Novel splice variants of the receptor for advanced glycation end-products expressed in human vascular endothelial cells and pericytes, and their putative roles in diabetes-induced vascular injury. Biochem J 2003;370(Pt 3): 1097-1109. 14. Hotoda N, Koike H, Sasagawa N, Ishiura S. A secreted form of human A D A M 9 has an alpha-secretase activity for APP. Biochem Biophys Res Commun 2002;293(2):800-805. 15. Gery S, Koeffler HP. Repression of the TMEFF2 promoter by c-Myc. J Mol Biol 2003;328(5):977-983. 102 4 14-3-3 SIGMA INCREASES THE TRANSCRIPTIONAL ACTIVITY OF THE ANDROGEN RECEPTOR IN THE ABSENCE OF ANDROGENS* 4.1 Introduction Prostate cancer is the second leading cause of cancer death among American men [1]. Organ-confined disease is treatable by means of radical prostatectomy or radiotherapy, but while initially effective, androgen deprivation therapy for metastatic disease is strictly palliative. The eventual emergence of androgen-independent disease leaves patients with few effective treatment options, none of which extend survival by more than a few months. The molecular mechanism of disease progression remains uncertain (see [2,3] for recent reviews). The androgen receptor (AR) is a ligand-dependent transcription factor required for the regulation of androgen-dependent genes, such as prostate-specific antigen (PSA). Increased expression of P S A in the sera of men after androgen deprivation therapy is the hallmark for identifying androgen-independent disease. Numerous other androgen-regulated genes also become re-expressed with progression to androgen independence, indicating that the A R is active in these tissues [3,4]. Ligand-independent activation of the A R is suggested to be a significant contributor to this progression to androgen-independent disease [3,5]. One proposed mechanism by which the A R is activated in the absence of androgens is by the action of co-regulatory proteins stimulated by other signalling pathways [5,6]. For example, interleukin-6 (IL-6) caused ligand-independent activation of the A R through the activity of steroid receptor co-activator 1 (SRC1), the mitogen-activated protein kinase ( M A P K ) * A version of this chapter is a manuscript in preparation. Quayle, SN, Sadar M D . 14-3-3 sigma increases the transcriptional activity of the androgen receptor in the absence of androgens. In preparation. 103 pathway, and signal transducer and activator of transcription 3 (STAT3) [7,8]. Increased expression and/or activity of these proteins has also been correlated with prostate cancer and the development of androgen independence [9-12]. Additionally, the co-activator p300 was recently suggested to activate the expression of PSA independent of the A R in LNCaP cells after long term treatment with IL-6 [13]. The 14-3-3 family of proteins regulate a large number of cellular processes, including apoptosis, response to DNA damage, and mitogenic signalling (for a review see [14]). Members of this family form homo- and heterodimers, which can then bind to a diverse array of target proteins. Binding of 14-3-3s to target proteins modulates the function of those targets by, for example, altering their subcellular localization or protein interaction partners. This family consists of 7 different isoforms that share a high degree of sequence identity and conservation [15]. Despite this conservation, numerous isoform-specific observations have been described, including differences in binding partners, subcellular localization, and structure [16-18]. 14-3-3 eta was previously shown to enhance A R transcriptional activity in the presence of ligand, with no effect seen in the absence of ligand [19]. Unfortunately, the effects of other members of this protein family were not reported. Therefore, we investigated the effect(s) of all seven members of the 14-3-3 family of proteins on the transcriptional activity of the A R in prostate cancer cells. We demonstrated that 14-3-3 sigma specifically increased the transcriptional activity of the AR, particularly in the absence of androgens. To our knowledge, this is the first report to examine the effects of all seven 14-3-3 isoforms in the same system under identical conditions. 104 4.2 Materials and methods 4.2.1 Cell culture LNCaP cells obtained from Dr. L.W.K. Chung (Emory University School of Medicine, Atlanta, GA) were maintained in RPMI 1640 supplemented with 5% (v/v) fetal bovine serum (FBS) (HyClone, Logan, UT), 100 units/mL penicillin, and 100 ug/uL streptomycin. All chemicals were purchased from Sigma, unless stated otherwise. Androgen stimulation was performed by treatment of cells with the synthetic androgen R1881 (PerkinElmer, Woodbridge, Canada). 4.2.2 Plasmids The ARR3-tk-luciferase reporter plasmid has been described previously [20]. PSA (6.1kb)-luciferase was a kind gift of Dr. J.-T. Hsieh (University of Southwestern Medical Center, Dallas, TX). The expression vectors for the seven isoforms of 14-3-3 were generously provided by Dr. H. Fu (Emory University School of Medicine, Atlanta, GA) [21]. The control expression vector pDEST26 was generated by removing the BsrGI fragment from the 14-3-3 zeta-encoding plasmid, eliminating the entire coding sequence. All plasmids were sequence verified to ensure they contained the entire known coding sequence of each 14-3-3 isoform. 4.2.3 Transfection and luciferase assay LNCaP cells were plated at 3.0x105 cells per well in 6-well plates. After 24 hr, transfections were performed in serum-free media using LiPOFECTiN® Reagent (Invitrogen) according to the manufacturer's protocol. Each 14-3-3 expression vector 105 was transfected at 500 ng/well unless indicated otherwise, and each luciferase reporter construct was transfected at 1.0 ug/well. The total amount of transfected DNA was normalized to 3 ug/well by the addition of a promoterless plasmid [7,8]. After 24 hr serum-free RPMI 1640 containing R1881 (0.001 - 10 nM), or an equal amount of vehicle (ethanol), was added to the cells. Cell lysates were harvested after 24 or 48 hr incubation using Passive Lysis Buffer (Promega, Nepean, Canada). Luciferase activities were measured on a multiplate luminometer (EG & G Berthold, Germany) using the Dual Luciferase Assay System (Promega). Luciferase activities were normalized to the protein concentration of the lysates [7,22] as determined by Bradford assays [23]. All transfection experiments were performed in triplicate wells and repeated at least 3 times. The results are presented as fold induction relative to pDest26 alone. 4.2.4 ARsiRNA A 21-nucleotide double-stranded siRNA duplex targeting nucleotides 293-312 (5'-aagcccatcgtagaggcccca-3') of the AR was obtained from Dharmacon Research (Boulder, CO). A control double-stranded siRNA (AR-SCR) targeting the inverse sequence of nucleotides 293-315 (5'-accccggagatgctacccgaa-3') of the A R was also generated. The efficacy of these molecules was shown previously [24]. LNCaP cells were plated at 2.0x105 cells per well in 6-well plates and incubated in RPMI 1640 with 5% FBS for 24 hr. Cells were fed fresh serum-free RPMI 1640 and transfected with 100 nM of siRNA duplex using OLIGOFECTAMINE® Reagent (Invitrogen) according to the manufacturer's protocol. After 24 hr the cells were incubated for a further 24 hr in the presence of 5% FBS prior to transfection with the luciferase reporters (1.0 ug/well) and 106 14-3-3 isoforms (1.0 pg/well) in serum-free media as described above. Again, the total amount of DNA transfected was normalized to 3 pg/well by the addition of a promoterless plasmid. 4.2.5 Western blot analysis LNCaP cells were grown and treated in parallel with the siRNA-transfected cells analyzed for luciferase activity. These cells here harvested and washed twice in ice-cold PBS. The cell pellets were lysed in 2x SDS-PAGE loading buffer and boiled for five minutes prior to loading on 12% SDS-PAGE gels. Alternatively, cell lysates from the luciferase assays were also used for western blotting. The separated proteins were transferred to nitrocellulose membranes and blocked in Odyssey Blocking Buffer (LI-COR Biosciences, Lincoln, NE) for one hour at room temperature or at 4°C overnight, and then washed three times. The blots were probed with anti-AR (441)(1:500) or anti-His (H-15)(l:200) (Santa Cruz Biotechnology, Santa Cruz, CA) for 1 to 2 hours at room temperature, washed three times, and incubated with the appropriate IR-labelled secondary antibody (1:5000, Molecular Probes) for 30 minutes at room temperature. All washes were performed with PBS containing 0.1% Tween20. Primary antibodies were diluted in Odyssey Blocking Buffer with 0.1% Tween20, while secondary antibody dilutions contained an additional 0.01% SDS. Visualization was performed with the Odyssey Infrared Imaging System (LI-COR Biosciences). 107 4.2.6 Co-immunoprecipitation LNCaP cells were seeded at 2.0xl06 cells per 10 cm plate. After 24 hr, the cells were either mock transfected or transfected as above with the 14-3-3 sigma or pDest26 plasmids. The cells were incubated for a further 24 hr prior to treatment with 10 nM R1881 or an equal amount of vehicle (ethanol) as indicated. After 24 hr the cells were harvested and washed in Buffer 1 (40 mM Tris pH 7.4, lx protease-inhibitor cocktail, 1 mM sodium orthovanadate, 50 mM sodium fluoride, 1 mM phenylmethylsulphonylfluoride). The resulting cell pellet was freeze-thawed and re-suspended in modified RIPA buffer (20 mM Tris pH 7.4, 140 mM sodium chloride, 0.5% sodium deoxycholate, 0.5% Nonidet P40, 10% glycerol, 20 mM sodium fluoride, lx protease inhibitor cocktail) by rotation at 4°C for 15 min. Approximately 10% of each input lysate was removed and the lysates were then precleared with 4 ug/mL agarose-conjugated anti-mouse immunoglobulin prior to immunoprecipitating the A R with 2 ug/mL anti-AR (441) antibody and 20 uL of beads conjugated with Protein A / G (Santa Cruz Biotechnology). The precipitated proteins were washed successively with modified RIPA buffer. The final pellet was then re-suspended in 2x SDS-PAGE loading buffer and western blots performed as above. 4.3 Results 4.3.1 14-3-3 sigma specifically enhanced AR transcriptional activity To examine the effects of all the 14-3-3 isoforms on the transcriptional activity of the A R we utilized two different luciferase reporter constructs. The PSA (6.1 kb)-luciferase reporter construct contains both the enhancer and promoter regions of the 108 human PSA locus and is highly sensitive to androgens [25]. The second construct, APvR3-thymidine kinase (tk)-luciferase, is an artificial reporter containing three tandem repeats of the ARE1 and ARE2 regions of the rat probasin gene [20]. Both of these reporters were transiently transfected into LNCaP cells along with each isoform of 14-3-3, or the empty vector pDest26, and then treated with the synthetic androgen R1881 (10 nM) or its carrier, ethanol (0.00055 % v/v). Both reporter constructs displayed similar fold induction trends for each 14-3-3 isoform when compared to the empty vector (Fig. 4.1A,B). LNCaP cells treated with R1881 showed small inductions of reporter activity by the various isoforms of 14-3-3 relative to the empty vector (Fig. 4.1 A). One-way analysis of variance (ANOVA) followed by a multiple comparison test determined that both the sigma (2.2-fold) and tau (2.0-fold) isoforms caused a statistically significant increase (p<0.01) in PSA (6.1)-luciferase reporter activity relative to the empty vector. Surprisingly, though, 14-3-3 sigma was also found to greatly increase the activity of both reporters in the absence of androgens (Fig. 4. IB; /K0.001). This effect was specific to the sigma isoform as no other isoform significantly increased the activity of either reporter construct in the absence of androgens. Western blotting with an anti-His antibody demonstrated that all the isoforms were expressed in these cells (Fig. 4.1C). Together, these data indicate that the 14-3-3 sigma protein is uniquely able to increase the transcriptional activity of the AR, especially in the absence of androgens. 109 Figure 4.1 - The transcriptional activity of the AR was activated specifically by 14-3-3 sigma. LNCaP cells were transfected with each 14-3-3 isoform or the empty vector pDest26 and either the PSA (6.1)-luciferase or the ARR3-tk-luciferase reporter construct, followed by treatment with either 10 nM R1881 (A) or vehicle (B). The results are presented as fold induction relative to empty vector alone, and the error bars represent the standard deviation of the mean of three independent experiments. *,/><0.001 as determined by one way A N O V A and Tukey's multiple comparison test. (C) Immunoblot demonstrating expression of each His-tagged 14-3-3 isoform in transfected cells. Probing for p-actin confirmed the amount and quality of protein in each lane. All transfections and treatments were performed in the absence of serum. i l l 4.3.2 14-3-3 sigma enhanced AR activity in a dose-dependent manner in the absence of androgens We next examined the effect of increasing amounts of 14-3-3 sigma expression on reporter activity. All levels of 14-3-3 sigma caused a small, but consistent increase in reporter activity when cells were treated with R1881 (Fig. 4.2A). In contrast, reporter activity increased in a dose-dependent manner when the cells were grown in the absence of androgens (Fig. 4.2B). This suggests that only small amounts of 14-3-3 sigma protein were sufficient to activate AR-dependent reporters, but in the absence of androgens increasing amounts of this isoform were able to further increase reporter activity. 4.3.3 14-3-3 sigma increased the sensitivity of AR to low levels of androgens Since 14-3-3 sigma strongly activated these reporter constructs in the absence of androgens, we investigated the effect of the sigma isoform at the low levels of androgens observed in patients undergoing androgen deprivation therapy. To test this, a range of R1881 concentrations (0.001 - 10 nM) were used. While the absolute levels of luciferase activity increased with increasing R1881, the fold induction caused by 14-3-3 sigma was actually greatest at low concentrations of R1881 (Fig. 4.3). These results implicate a mechanism whereby 14-3-3 sigma is able to sensitize the A R to low levels of androgen. 4.3.4 Induction of PSA reporters by 14-3-3 sigma was dependent on AR Finally, to determine whether the activation of these reporter constructs by 14-3-3 sigma was mediated through the AR, a siRNA oligonucleotide strategy targeting the A R was used to reduce the amount of AR protein prior to the expression of 14-3-3 sigma in 112 R1881 5 c o 4 o 1 3 2 2 o B o s 4 3 1 3 2 2^ 14-3-3a 0 pDest26 500 14-3-30 0 pDest26 500 100 400 200 300 ng/well 300 200 400 500 100 0 Vehicle 100 400 200 300 ng/well 300 200 400 100 500 0 Figure 4.2 - Increasing amounts of 14-3-3 sigma increased the amount of A R activity in the absence of androgens. LNCaP cells were transfected with the indicated amounts of 14-3-3 sigma (o) and the total amount of DNA transfected was kept equal with the addition of pDest26 empty vector. The cells were then treated with either 10 nM R1881 (A) or vehicle (B). The results are presented as fold induction relative to empty vector alone, and the error bars represent the standard deviation of the mean of three independent experiments. All transfections and treatments were performed in the absence of serum. 113 CO CM CO 0> a a. CO > c o o 3 T3 C o LL 10 -i i l 0.001 0.01 0.1 R1881 (nM) 10 Figure 4.3 - 14-3-3 sigma sensitized the A R to low levels of androgen. LNCaP cells transfected with equal amounts of 14-3-3 sigma or pDest26 empty vector were treated for 48 hours with the indicated concentrations of R1881. The results are presented as fold induction relative to empty vector alone, and the error bars represent the standard deviation of the mean of three independent experiments. All transfections and treatments were performed in the absence of serum. 114 Figure 4.4 - Activation of the ARR3-tk-luciferase reporter construct by 14-3-3 sigma was dependent on the AR. LNCaP cells were transfected with an siRNA oligonucleotide targeting the A R (AR-siRNA) or a scrambled oligonucleotide (AR-SCR) prior to transfection with 14-3-3 sigma (a) or pDest26 empty vector. The cells were then treated with 10 nM R1881 (A) or vehicle (B). Activity of the luciferase reporter is expressed as fold induction relative to empty vector plus AR-SCR oligonucleotide, and the error bars represent the standard deviation of the mean of three independent experiments. * p<0.02 as determined by Student's t-test between cells treated with R1881 and vehicle in the presence of the AR-SCR control oligonucleotide. (C) Immunoblot demonstrating an efficient decrease in AR protein levels in the presence of the AR-siRNA oligonucleotide under both treatment conditions. Probing for P-actin confirmed the amount and quality of protein in each lane. All transfections and treatments were performed in the absence of serum. 116 these cells. Consistent with our previous results, cells transfected with a control scrambled oligonucleotide (AR-SCR) and 14-3-3 sigma showed a statistically significant increase in reporter activity in both the presence and absence of androgens (Fig. 4.4A,B; /?<0.02). Transfection with an AR-specific siRNA oligonucleotide (AR-siRNA) greatly reduced the expression of AR protein in these cells (Fig. 4.4C). As expected, knock-down of A R protein levels led to decreased reporter activity regardless of the presence of 14-3-3 sigma or treatment with androgens. Significantly, though, expression of 14-3-3 sigma in AR-siRNA-transfected cells did not result in increased reporter activity (Fig. 4.4A,B). These data suggest that the AR was required for 14-3-3 sigma to activate this reporter construct. 4.3.5 14-3-3 sigma did not alter AR protein levels We next investigated possible mechanism(s) by which 14-3-3 sigma could increase A R transcriptional activity. We first sought to determine if 14-3-3 sigma could bind directly to the A R to alter its activity, as was previously observed for the glucocorticoid receptor (GR), a related steroid nuclear receptor [26]. However, we did not detect a direct interaction between these proteins by co-immunoprecipitation under the conditions tested (Fig. 4.5A). Samples of input lysates confirmed the expression of A R and 14-3-3 sigma in these cells. Immunoprecipitation enriched the relative amount of A R protein, but no 14-3-3 sigma was detected in these samples. 14-3-3 sigma was also previously shown to inhibit the transcriptional activity of GR by sequestering the receptor in the cytosol [26]. Therefore, we monitored the subcellular localization of a green fluorescent protein (GFP)-AR fusion protein in the presence and absence of 14-3-3 117 14-3-3a CN h -to LU Q Q. ^ fl) E >< £ B CM CO 00 CM X o LU n ra > 0- > 0. Vehicle R1881 to <o ^ CM D W CM D w •jJ c o c o J 0 c o a) c o c o flj c o c o Q 4 4 Q 4 4 & T - T" Q . T - T -AR 14-3-3CT AR P-actin Figure 4.5 - 14-3-3 sigma did not interact with or alter the amount of A R protein in the cells. (A) LNCaP cells were mock transfected or transfected with 14-3-3 sigma (a) or pDest26 empty vector prior to treatment with 10 nM R1881 or vehicle for 24 hours and immunoprecipitation with an anti-AR antibody. The lysates and immunoprecipitated samples were immunoblotted with antibodies specific for the AR and 14-3-3 sigma. ( B ) LNCaP cells were transfected with either 14-3-3 sigma, 14-3-3 epsilon (e), or pDest26 and treated as indicated for 48 hours. All transfections and treatments were performed in the absence of serum. Whole cell lysates were harvested and immunoblotted with an antibody specific for the AR. Probing for P-actin confirmed the amount and quality of protein in each lane. 118 sigma. No significant difference in AR localization was observed in the presence or absence of androgens when compared to cells transfected with empty vector alone. Finally, it was recently shown that expression of 14-3-3 eta increased GR protein levels, by inhibiting degradation of the receptor, which lead to increased transcriptional activity of the GR [27]. However, western blot analysis demonstrated that the presence of 14-3-3 sigma did not alter the level of AR protein in these cells (Fig. 4.5B). Thus, while the A R must be present for 14-3-3 sigma to activate these reporter constructs, the expression of 14-3-3 sigma does not increase AR activity by simply increasing the amount of A R protein present in the cell or by significantly altering the subcellular localization of the AR. 4.4 Discussion Members of the 14-3-3 protein family are involved in regulating a wide range of cellular events. Various isoforms of this family have also been previously shown to regulate the transcriptional activity of their target proteins, including steroid nuclear receptors [14,19,28]. However, the studies investigating the A R and GR have focused on individual isoforms and the experiments have been performed in a variety of cell types. This is the first report to examine the effects of all seven isoforms on the transcriptional activity of the A R in a single system. We have shown that 14-3-3 sigma, in contrast to the other family members, greatly enhanced the transcriptional activity of the A R in the absence of androgens. In contrast to our findings with the AR, Kino, et al. [26] demonstrated that 14-3-3 sigma inhibited the transcriptional activity of the GR by sequestering the receptor in the 119 cytosol. Additionally, 14-3-3 eta was shown to enhance GR activity via two different mechanisms. This isoform was shown to alter the subcellular localization of the GR co-repressor RIP 140 and increase the amount of GR protein, with both mechanisms resulting in increased GR transcriptional activity [27,29]. Finally, 14-3-3 eta was also shown to increase the transcriptional activity of the AR, but only in the presence of androgens, and the mechanism by which this occurred was not examined [19]. Our results showed minimal induction of A R activity in the presence of 14-3-3 eta expression, but this difference in results is likely due to differences in the way these experiments were performed. Our experiments utilized the LNCaP cell line, which expresses an endogenous AR, and the PSA (6.1kb) and ARR3-thymidine kinase (tk)-luciferase reporters, which are well characterized androgen-regulated reporter constructs. In contrast, the experiments of Haendler et al. [19] were performed in PC3 prostate cancer cells stably transfected with the AR and employed a novel androgen-regulated reporter based on the CRISP-1 promoter. Previous studies have focused mainly on activity of the receptor in the presence of ligand, and little to no increase in transcriptional activity was seen in the absence of ligand [19]. In contrast, our experiments highlight the unique ability of 14-3-3 sigma to enhance A R transcriptional activity in the absence of androgens. Consequently, our work suggests that 14-3-3 sigma may play an important role in sensitizing the AR to low levels of androgens. These findings may be most relevant in the setting of androgen independent prostate cancer. Androgen withdrawal therapy results in decreased, but detectable, levels of dihydrotestosterone in the prostate [30]; the expression of 14-3-3 sigma may enable the AR to still respond strongly to these castrate levels of androgen. 120 We investigated a number of mechanisms by which 14-3-3 sigma could activate the AR. Direct protein-protein interactions were previously demonstrated between 14-3-3 sigma and GR [26] as well as between 14-3-3 eta and both the GR and the A R [28,29]. However, we did not detect a direct interaction between the AR and 14-3-3 sigma in our co-immunoprecipitation experiments. Our results also suggest that expression of 14-3-3 sigma did not alter the amount or subcellular localization of AR protein. LNCaP cells do not express 14-3-3 sigma; therefore, the lack of difference was not due to endogenous levels of 14-3-3 sigma [31]. One potential mechanism yet to be investigated is whether 14-3-3 sigma may alter the amount of, or localization of, an AR co-repressor, leading to increased AR transcriptional activity. A comparable situation was observed for the GR where 14-3-3 eta was able to sequester the GR co-repressor RIP 140 in the cytosol, thus allowing the GR to have greater activity in the nucleus [29]. While it would be interesting to determine what protein(s) 14-3-3 sigma is binding to mediate its effects on AR, 14-3-3 sigma has been demonstrated to interact with hundreds of proteins [32]. It is therefore difficult to speculate which of these interactions may be responsible for increasing AR activity. Alternatively, 14-3-3 sigma may influence AR activity indirectly, for instance by leading to altered phosphorylation of the AR or one of its co-regulators. It would be possible to assess the phosphorylation status of A R in the presence of 14-3-3 sigma using antibodies specific for each phosphorylation site, but unfortunately only a few such site-specific antibodies exist. Furthermore, much work remains to identify all A R phosphorylation sites activated in vivo as well as the kinases and conditions responsible for these events [33]. 121 14-3-3 sigma appears to have a number of characteristics unique from the other 14-3-3 isoforms. For example, 14-3-3 sigma tends to only form homodimers while other 14-3-3 isoforms readily form heterodimers with multiple other isoforms [18]. The determination of the crystal structure of 14-3-3 sigma revealed a structural basis for this finding [18,34]. Various other isoform-specific functions have also been described, for instance in the subcellular localization of the sigma and zeta isoforms [17]. Additionally, all other isoforms of this family were shown to bind the nuclear localization sequence of p27, but only 14-3-3 sigma was unable to bind this region [35]. Finally, Cdc25C was also shown to be bound by other 14-3-3 isoforms, but not sigma [16,36]. Structural analyses revealed a secondary surface on 14-3-3 sigma that influenced the specificity of its interacting partners [18]. This region contained three residues perfectly conserved in all isoforms except sigma. Mutation of these residues to the consensus sequence of the other isoforms enabled 14-3-3 sigma to interact with Cdc25C [18], revealing a potential mechanism by which 14-3-3 sigma is able to exert specific effects on target proteins. Thus, it is not unexpected that 14-3-3 sigma was uniquely able to alter the transcriptional activity of the A R in our experiments. Loss of 14-3-3 sigma has recently been correlated with the development of prostatic intraepithelial neoplasia, and the majority of patients with invasive prostate cancer displayed low or no expression of this isoform [37-39]. This decreased expression is due to promoter hypermethylation in both primary prostate cancer samples and in the LNCaP cell line [31,38]. However, these studies also showed that a subset of patients appeared to retain 14-3-3 sigma expression. Significantly, Cheng et al. [37] also observed an increased frequency of expression of 14-3-3 sigma in the more poorly 122 differentiated samples of prostate cancer. Huang et al. [40] correlated increased nuclear localization of 14-3-3 sigma with decreasing differentiation of tumours. Consistent with this, strong expression of 14-3-3 sigma was observed in the androgen-independent prostate cancer cell lines PC3, DU145, and 22Rvl, and expression of 14-3-3 sigma has been correlated with increased chemoresistance [31,38,41]. Increased expression of 14-3-3 sigma was also observed in response to treatment with androgens, and a reporter containing the 14-3-3 sigma promoter was induced within three hours of androgen treatment [40]. Previous studies have suggested that genes regulated by the A R are re-expressed in androgen independent disease [4]. It will be interesting to examine the expression of 14-3-3 sigma in androgen-independent tissues as the presence of 14-3-3 sigma protein may lead to increased activation of the AR in a subset of patients with hormone-refractory prostate cancer. 123 4.5 References 1. Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ, Thun MJ. Cancer statistics, 2005. C A Cancer J Clin 2005;55(1): 10-30. 2. Ryan CJ, Eisenberger M . Chemotherapy for Hormone-Refractory Prostate Cancer: Now It's a Question of "When?" J Clin Oncol 2005;23(32):8242-8246. 3. Scher HI, Sawyers CL. Biology of progressive, castration-resistant prostate cancer: directed therapies targeting the androgen-receptor signaling axis. J Clin Oncol 2005;23(32):8253-8261. 4. Gregory CW, Hamil K G , Kim D, Hall SH, Pretlow TG, Mohler JL, French FS. Androgen receptor expression in androgen-independent prostate cancer is associated with increased expression of androgen-regulated genes. Cancer Res 1998;58(24):5718-5724. 5. Culig Z, Steiner H, Bartsch G, Hobisch A. Mechanisms of endocrine therapy-responsive and -unresponsive prostate tumours. Endocr Relat Cancer 2005;12(2):229-244. 6. Heinlein CA, Chang C. Androgen receptor (AR) coregulators: an overview. Endocr Rev 2002;23(2): 175-200. 7. Ueda T, Bruchovsky N, Sadar MD. Activation of the androgen receptor N-terminal domain by interleukin-6 via M A P K and STAT3 signal transduction pathways. J Biol Chem 2002;277(9):7076-7085. 124 8. Ueda T, Mawji NR, Bruchovsky N, Sadar MD. Ligand-independent activation of the androgen receptor by interleukin-6 and the role of steroid receptor coactivator-1 in prostate cancer cells. J Biol Chem 2002;277(41):38087-38094. 9. Gregory CW, He B, Johnson RT, Ford OH, Mohler JL, French FS, Wilson E M . A mechanism for androgen receptor-mediated prostate cancer recurrence after androgen deprivation therapy. Cancer Res 2001;61(11):4315-4319. 10. Gioeli D, Mandell JW, Petroni GR, Frierson HF, Jr., Weber MJ. Activation of mitogen-activated protein kinase associated with prostate cancer progression. Cancer Res 1999;59(2):279-284. 11. Dhir R, Ni Z, Lou W, DeMiguel F, Grandis JR, Gao AC. Stat3 activation in prostatic carcinomas. Prostate 2002;51(4):241-246. 12. Mora LB, Buettner R, Seigne J, Diaz J, Ahmad N, Garcia R, Bowman T, Falcone R, Fairclough R, Cantor A, Muro-Cacho C, Livingston S, Karras J, Pow-Sang J, Jove R. Constitutive activation of Stat3 in human prostate tumors and cell lines: direct inhibition of Stat3 signaling induces apoptosis of prostate cancer cells. Cancer Res 2002;62(22):6659-6666. 13. Debes JD, Comuzzi B, Schmidt LJ, Dehm SM, Culig Z, Tindall DJ. p300 regulates androgen receptor-independent expression of prostate-specific antigen in prostate cancer cells treated chronically with interleukin-6. Cancer Res 2005;65(13):5965-5973. 14. Hermeking H. The 14-3-3 cancer connection. Nat Rev Cancer 2003;3(12):931-943. 15. Wang W, Shakes DC. Molecular evolution of the 14-3-3 protein family. J Mol Evol 1996;43(4):384-398. 16. Chan TA, Hermeking H, Lengauer C, Kinzler KW, Vogelstein B. 14-3-3Sigma is required to prevent mitotic catastrophe after DNA damage. Nature 1999;401(6753):616-620. 17. van Hemert MJ, Niemantsverdriet M , Schmidt T, Backendorf C, Spaink HP. Isoform-specific differences in rapid nucleocytoplasmic shuttling cause distinct subcellular distributions of 14-3-3 sigma and 14-3-3 zeta. J Cell Sci 2004;117(Pt 8):1411-1420. 18. Wilker EW, Grant RA, Artim SC, Yaffe MB. A structural basis for 14-3-3sigma functional specificity. J Biol Chem 2005;280(19):18891-18898. 19. Haendler B, Schuttke I, Schleuning WD. Androgen receptor signalling: comparative analysis of androgen response elements and implication of heat-shock protein 90 and 14-3-3eta. Mol Cell Endocrinol 2001;173(l-2):63-73. 20. Snoek R, Rennie PS, Kasper S, Matusik RJ, Bruchovsky N. Induction of cell-free, in vitro transcription by recombinant androgen receptor peptides. J Steroid Biochem Mol Biol 1996;59(3-4):243-250. 21. Subramanian RR, Zhang H, Wang H, Ichijo H, Miyashita T, Fu H. Interaction of apoptosis signal-regulating kinase 1 with isoforms of 14-3-3 proteins. Exp Cell Res 2004;294(2):581-591. 22. Rowan BG, Weigel NL, O'Malley BW. Phosphorylation of steroid receptor coactivator-1. Identification of the phosphorylation sites and phosphorylation through the mitogen-activated protein kinase pathway. J Biol Chem 2000;275(6):4475-4483. 23. Bradford M M . A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248-254. 24. Wright M E , Tsai MJ, Aebersold R. Androgen receptor represses the neuroendocrine transdifferentiation process in prostate cancer cells. Mol Endocrinol 2003; 17(9): 1726-1737. 25. Cleutjens KB, van der Korput HA, van Eekelen CC, van Rooij HC, Faber PW, Trapman J. An androgen response element in a far upstream enhancer region is essential for high, androgen-regulated activity of the prostate-specific antigen promoter. Mol Endocrinol 1997; 11(2): 148-161. 26. Kino T, Souvatzoglou E, De Martino M U , Tsopanomihalu M , Wan Y, Chrousos GP. Protein 14-3-3 sigma interacts with and favors cytoplasmic subcellular localization of the glucocorticoid receptor, acting as a negative regulator of the glucocorticoid signaling pathway. J Biol Chem 2003;278(28):25651-25656. 27. Kim YS, Jang SW, Sung HJ, Lee HJ, Kim IS, Na DS, Ko J. Role of 14-3-3 eta as a positive regulator of the glucocorticoid receptor transcriptional activation. Endocrinology 2005;146(7):3133-3140. 28. Wakui H, Wright AP, Gustafsson J, Zilliacus J. Interaction of the ligand-activated glucocorticoid receptor with the 14-3-3 eta protein. J Biol Chem 1997;272(13):8153-8156. 29. Zilliacus J, Holter E, Wakui H, Tazawa H, Treuter E, Gustafsson JA. Regulation of glucocorticoid receptor activity by 14—3-3-dependent intracellular relocalization of the corepressor RIP140. Mol Endocrinol 2001;15(4):501-511. 30. Geller J, Albert J, Nachtsheim D, Loza D, Lippman S. Steroid levels in cancer of the prostate—markers of tumor differentiation and adequacy of anti-androgen therapy. Prog Clin Biol Res 1979;33:103-111. 31. Urano T, Takahashi S, Suzuki T, Fujimura T, Fujita M , Kumagai J, Horie-Inoue K, Sasano H, Kitamura T, Ouchi Y, Inoue S. 14-3-3sigma is down-regulated in human prostate cancer. Biochem Biophys Res Commun 2004;319(3):795-800. 32. Benzinger A, Muster N, Koch HB, Yates JR, 3rd, Hermeking H. Targeted proteomic analysis of 14-3-3 sigma, a p53 effector commonly silenced in cancer. Mol Cell Proteomics 2005;4(6):785-795. 33. Edwards J, Bartlett JM. The androgen receptor and signal-transduction pathways in hormone-refractory prostate cancer. Part 1: Modifications to the androgen receptor. BJU Int 2005;95(9): 1320-1326. 34. Benzinger A, Popowicz G M , Joy JK, Majumdar S, Holak TA, Hermeking H. The crystal structure of the non-liganded 14-3-3 sigma protein: insights into determinants of isoform specific ligand binding and dimerization. Cell Res 2005;15(4):219-227. 35. Sekimoto T, Fukumoto M , Yoneda Y. 14-3-3 suppresses the nuclear localization of threonine 157-phosphorylated p27(Kipl). Embo J 2004;23(9): 1934-1942. 128 36. Peng CY, Graves PR, Thoma RS, Wu Z, Shaw AS, Piwnica-Worms H. Mitotic and G2 checkpoint control: regulation of 14-3-3 protein binding by phosphorylation of Cdc25C on serine-216. Science 1997;277(5331): 1501-1505. 37. Cheng L, Pan CX, Zhang JT, Zhang S, Kinch MS, Li L , Baldridge L A , Wade C, Hu Z, Koch MO, Ulbright T M , Eble JN. Loss of 14-3-3sigma in prostate cancer and its precursors. Clin Cancer Res 2004;10.(9):3064-3068. 38. Lodygin D, Diebold J, Hermeking H. Prostate cancer is characterized by epigenetic silencing of 14-3-3sigma expression. Oncogene 2004;23(56):9034-9041. 39. Mhawech P, Greloz V, Assaly M , Herrmann F. Immunohistochemical expression of 14-3-3 sigma protein in human urological and gynecological tumors using a multi-tumor microarray analysis. Pathol Int 2005;55(2):77-82. 40. Huang D, Liu X, Plymate SR, Idowu M , Grimes M , Best A M , McKinney JL, Ware JL. Proteomic identification of 14-3-3 sigma as a common component of the androgen receptor and the epidermal growth factor receptor signaling pathways of the human prostate epithelial cell line Ml2. Oncogene 2004;23(41):6881-6889. 41. Han B, Xie H, Chen Q, Zhang JT. Sensitizing hormone-refractory prostate cancer cells to drug treatment by targeting 14-3-3 {sigma}. Mol Cancer Ther 2006;5(4):903-912. 129 5 CONCLUDING CHAPTER 5.1 Expression profiling of novel clones 5.1.1 Custom cDNA microarray Preliminary work has been performed to examine the expression profiles of the genes identified in the SSH experiments described in Chapter 2. Expression levels of a handful of these clones were examined using other techniques, such as northern blot analysis, but given the large number of candidate genes this was not an efficient method of validating the results of these experiments. To overcome this problem, we chose 156 of these clones and spotted them on glass slides as a custom cDNA microarray, allowing us to examine the expression level of all these genes in a single experiment. We also selected 57 genes whose expression was previously shown to change with disease progression. A number of positive and negative control genes were also selected to allow us to monitor hybridization efficiency and control for sample-to-sample variations. Using the custom cDNA microarray I hoped to obtain data regarding a number of questions. First, I examined normal human tissue RNA samples to determine if any of the selected clones represented genes that were exclusively expressed in the prostate. Such genes could be important in the diagnosis and monitoring of prostate cancer. Second, I treated prostate cancer cells in tissue culture with synthetic androgens to identify novel androgen regulated genes. Third, I have screened a few samples from the LNCaP hollow fibre model to identify genes whose changes in expression consistently correlate with hormonal progression. These genes are the most likely to be involved in the process of hormonal progression and could thus represent diagnostic or therapeutic targets. 130 I also examined this data to try to identify a "signature" of gene expression levels representing different stages of hormonal progression. A number of studies have now been performed where large sets of clinical samples have been profiled using microarrays. These studies have found that distinct clinical subtypes of particular cancers, with different clinical outcomes, can be identified based strictly on the expression "signature" of the tumour [1-4]. Another study determined that the likelihood of metastasis could also be determined by examining the expression profile of the primary tumour [5]. While these findings are exciting, it should be noted that these gene expression signatures have not proven as reliable when used by other groups; a recent study has suggested that thousands of patient samples would be required to develop a truly robust expression signature [6]. While the preliminary analysis of the data from our custom cDNA array identified interesting trends, significant conclusions could not be drawn at this point as in each case more samples will be required in order to generate higher confidence in the data. There were other difficulties that also limited the utility of these custom microarrays. For instance, the microarray approach was not sensitive enough to consistently detect some of the rare transcripts expressed at low levels. Since a number of the transcripts seemed to be at the limit of detection the reproducibility of the technique was diminished. To compensate for this I attempted to use greater sample volumes. While this may have increased the reproducibility, the increased sample demands limited the number of samples for which sufficient material was available. Finally, because of initial concerns about the sensitivity of the technique, only a few of the novel cDNA fragments were spotted on this version of the array. Thus, the major 131 limitation of microarrays - the ability to only detect those transcripts on the array -prevented us from examining the expression levels and changes of all the novel genes identified by SSH. 5.1.2 Low-density Taqman® array As discussed in Chapter 1.5 there are a number of techniques available for monitoring expression profiles for multiple genes. There are also many techniques for examining the expression levels of individual transcripts. Northern blot analysis, for example, is a standard technique for characterizing expression levels and transcript size. However, the most sensitive technique is likely PCR. Unfortunately, standard PCR reactions are not quantitative and are thus not appropriate for determining expression level. On the other hand, real time quantitative PCR was developed specifically to allow for accurate quantitation of genomic DNA and transcript levels [7]. Briefly, in addition to a standard gene-specific primer pair, this TaqMan® approach also utilizes a gene-specific probe located between the two primers. This probe carries both a fluorescent molecule and a quencher, and the degradation of the probe via the 5' nuclease activity of the polymerase results in the release of fluorescence at levels proportional to the amount of PCR product. Amplification of a specific product can then be followed in real-time using a specially equipped thermal cycler. Applied Biosystems (Foster City, CA) has extended this technology with the development of the TaqMan® Low Density Array. This "array" consists of 384 unique real time PCR assays positioned at defined locations on a 384-well microfluidics card. Thus, this Low Density Array is analogous to a microarray in that pre-defined targets are 132 placed at discrete sites on the card. In contrast, though, this microfluidics card equally distributes a cDNA sample amongst the 384 wells, thus allowing for the expression monitoring of up to 384 genes in a single experiment. Additionally, real time PCR is extremely sensitive and highly reproducible [7]. Given the large number of potentially novel transcripts identified by SSH, we hoped to examine their expression levels in samples of prostate cancer. However, such experiments would be limited by the amount of sample available. A new study is currently being designed by Dr Sadar in which real time PCR assays would be designed for each novel transcript. These assays, as well as assays for genes previously associated with prostate cancer, such as PSA and AR, will be combined in a single TaqMan® Low Density Array. This array will then be screened with samples obtained via biopsy or radical prostatectomy from >50 patients with low grade prostate cancer who were treated with curative intent. At least 5 years of clinical follow-up data is also available for these patients. It is hoped that this method will identify which, if any, of the novel transcripts identified by SSH may be a useful marker for predicting clinical outcome after primary therapy. Again, there may not be enough samples to generate a highly reproducible gene expression signature, but it is hoped that the reproducibility of the TaqMan® technique will identify specific genes and gene signatures worthy of extensive follow-up. 5.2 Summary and future directions The molecular mechanisms resulting in the development of HRPC are not completely understood. The AR appears to play a significant role in this process; however, it is not apparent how the AR is being activated in the absence of androgens. 133 Additionally, new markers of disease progression need to be identified to distinguish disease subtypes that are associated with different prognoses and therapeutic responses. In this thesis I have described the experiments I performed to identify and characterize genes associated with the hormonal progression of prostate cancer. 5.2.1 Novel transcripts Suppression subtractive hybridization has been used in the past to study differential gene expression in the development and progression of prostate cancer [8,9]. However, few of these studies have examined changes in expression associated with the development of androgen independent disease, and fewer still have examined in vivo models of progression [9]. Additionally, these studies generally focused on transcripts derived from known genes. In contrast, our study sequenced all the cDNA fragments isolated from SSH in order to identify novel transcripts associated with hormonal progression in the LNCaP hollow fibre model. This model is unique in that it allows the in vivo study of hormonal progression with pure populations of prostate cancer cells. Aside from the known genes that were isolated, this SSH approach also identified sequences from 25 novel transcripts. These transcripts had not been previously detected and thus may represent novel genes of interest in prostate cancer. Additionally, a significant number of the isolated cDNA fragments corresponded to or overlapped with previously sequenced ESTs for which no other annotation was available. Such clones are informative for two reasons: (1) this is the first report to identify many of these ESTs in prostate cancer, and (2) some of the clones overlapping with known transcripts may 134 derive from novel splice variants of these genes. These studies have provided a rich source of novel transcripts in the hormonal progression of prostate cancer. As described in Chapter 5.1, ongoing experiments are examining the expression levels of the transcripts identified by SSH. These experiments are aimed at (1) identifying transcripts whose expression is regulated by the AR, (2) identifying genes specifically expressed in prostate tissues and/or prostate cancer, and (3) identifying transcripts whose expression correlates with the development and progression of prostate cancer. Aside from assessing expression changes in the LNCaP hollow fibre model, such experiments will also allow for the examination of other models of disease recurrence, such as the CWR22 xenograft model described in Chapter 1.4 [10,11]. Confirmation of the observed expression changes in multiple models of this process will increase confidence in their validity. The ultimate goal will be to validate these findings in clinical specimens of HRPC. However, such tissues are difficult to obtain, and thus, only the most promising candidate genes would be likely to progress to this stage. Additionally, other members of the Sadar Laboratory are examining expression changes of known genes in the LNCaP hollow fibre model using multiple high-throughput techniques such as Affymetrix GeneChip® arrays, SAGE, and Isotope-coded affinity tags (ICAT). It is hoped that these experiments will provide complementary data to aid in the selection of target genes for future experiments. Such an approach led to the selection of the 14-3-3 family for the functional studies described in Chapter 4. As discussed in Chapter 1.1, new markers are needed to distinguish those patients likely to have a poor prognosis. The expression of such markers would ideally 135 be restricted to, or greatly upregulated in, prostate cancer cells. For example, PIM1 and hepsin were identified as potential biomarkers in prostate cancer through a microarray analysis [12]. Given their restricted expression profile, such markers are also potential targets of novel therapeutics. For example, both TMEFF2 (discussed in Chapter 3) and STEAP (six-transmembrane epithelial antigen of the prostate) were proposed as therapeutic targets after being identified as differentially expressed in prostate cancer through the use of subtractive hybridization [9,13]. Therapeutic antibodies targeting TMEFF2 have now shown favourable results in independent pre-clinical trials [14,15]. The potential utility of these novel transcripts in defining prognostic and diagnostic gene signatures should also be considered. Previous studies have used microarrays to identify gene expression signatures associated with prostate cancer recurrence after surgery and with the development of metastatic cancer [4,5]. But, as previously discussed, one of the great limitations of microarrays is that one can only detect transcripts that are represented on the array. Future studies defining the expression profiles of the novel transcripts described in Chapter 2 may thus identify novel biomarkers or expression profiles for later use in the clinical management of prostate cancer. Finally, in order to begin functional studies on any of the identified transcripts it would first be necessary to isolate full-length cDNAs for these genes. Rapid amplification of cDNA ends (RACE) is a powerful technique to isolate full-length coding sequences, and it may prove to be more efficient when used on a gene-by-gene basis. Having this sequence information would allow for the prediction of protein coding 136 regions and possible functional domains, as well as comparison to previously identified genes in other species. 5.2.2 TMEFF2-S The transmembrane protein TMEFF2 is primarily expressed in the brain and the prostate, with higher expression levels observed in prostate cancer [14,16,17]. As discussed in Chapter 3, one of the novel cDNAs isolated via SSH was shown to derive from a splice variant of the TMEFF2 gene. This truncated splice variant encodes a secreted form of the TMEFF2 protein that lacks an EGF-like and follistatin-like domain. This is the first report to identify a secreted isoform of the transmembrane protein TMEFF2. Further studies will have to be performed to determine the expression profile of the TMEFF2-S variant during the development and hormonal progression of prostate cancer. Being a secreted protein, it may also be possible to detect TMEFF2-S in the serum of patients. This is one of the ideal characteristics of a biomarker as it allows for easy, relatively non-invasive monitoring in patients. These studies are timely and of high interest because Protein Design Labs, Inc. (Freemont, CA) and Berlex Biosciences (Richmond, CA) are developing antibodies to TMEFF2 for the potential treatment of metastatic prostate cancer based upon the limited tissue distribution of full-length TMEFF2 [14,15]. However, these groups may not have considered the presence of a secreted isoform of TMEFF2 when designing their antibodies. The potential mechanism(s) of action of these antibodies and the strategies used for epitope selection may have to be reconsidered in light of these results. 137 Additionally, this novel isoform of TMEFF2 may be a valuable reagent for examining the potential function and mechanism of the TMEFF2 protein in the etiology of prostate cancer and the development of androgen-independent disease. Having both isoforms would allow for the direct comparison and characterization of each form of the protein. For example, a thorough analysis of the binding partners of each isoform may delineate those protein-protein interactions which are isoform-specific. This knowledge may aid in the design of future experiments to examine the functional roles of each variant. Since TMEFF2-S is a secreted factor it is easily amenable to studies of its effects on other cells. For instance, the culture media from cells expressing the TMEFF2-S-GFP fusion protein could be removed and applied to cells not expressing this construct. Comparison with media isolated from control cells would indicate if the presence of the secreted protein alters the growth characteristics or morphology of the treated cells. Alternatively, TMEFF2-S protein could be generated from an in vitro translation system or purified from a bacterial expression system. Given the EGF-like and follistatin-like domains it is possible that TMEFF2 and TMEFF2-S modulate the signalling events of a growth factor protein. Therefore, it may also be necessary to test the effect of TMEFF2-S before and after treatment with a variety of growth factors. The results of such experiments could be highly relevant to the understanding of HRPC as a number of growth factors have been implicated in the ligand-independent activation of the AR [18-20]. Finally, it is unknown whether TMEFF2 and TMEFF2-S would act at the level of the cell surface or in the extracellular milieu. For example, the latent TGF-P binding 138 proteins (LTBP) regulate the availability and activity of secreted TGF-P ligand [21]. Alternatively, a homolog of TMEFF2, called TMEFF1, acted at the level of the membrane to inhibit signalling downstream of the TGF-P family member Nodal. Nodal normally binds the type I and II receptors, but also requires binding to the co-receptor Cripto. TMEFF1 was shown to specifically inhibit signalling through Nodal at the level of the membrane by directly binding the Cripto co-receptor [22,23]. Only the transmembrane form of TMEFF1 exhibited this activity; a secreted TMEFF1 construct had no effect on Nodal signalling [22]. Once the binding partners of TMEFF2 are known it will be interesting to determine if a similar scheme is in place for TMEFF2. However, TMEFF1 is not currently known to encode a secreted isoform, thus it is difficult to predict what role TMEFF2-S may be playing. 5.2.3 14-3-3 sigma When androgen-responsive prostate cells are maintained in the absence of androgens the majority of AR protein becomes localized in the cytoplasm, indicating it is transcriptionally inactive [24]. This is consistent with the low level of PSA expression observed in LNCaP cells maintained in androgen-free medium [25]. HRPC is characterized by the re-expression of PSA and other androgen-regulated genes, suggesting the AR is active in these tissues [26,27]. Much research effort is now focused on the mechanisms enabling the AR to be highly transcriptionally active in the presence of little to no ligand. The consistent identification of various members of the 14-3-3 family of proteins in high-throughput screens of differential expression in LNCaP cells led us to predict these proteins may be of interest in the development of HRPC. Chapter 139 4 described my experiments to show that 14-3-3 sigma may specifically sensitize the A R to low levels of androgens. This is the first report to demonstrate that a 14-3-3 family member could increase the transcriptional activity of a steroid nuclear receptor in the absence of its natural ligand. A previous study showed that three amino acid residues in 14-3-3 sigma were critical for altering its binding specificity relative to the other 14-3-3 isoforms [28]. Given that only 14-3-3 sigma significantly altered AR activity in my study, it would be interesting to determine if mutating these three residues back to the consensus sequence would prevent 14-3-3 sigma from activating the AR. Such an experiment would confirm that a specific interaction with 14-3-3 sigma is required for this effect. However, it is difficult to predict which protein, or proteins, 14-3-3 sigma is interacting with in this situation. Work by Benzinger, et al., demonstrated that 14-3-3 sigma interacted with hundreds of different proteins under the conditions tested [29]. It is also likely that different cell types and treatment conditions would express a unique repertoire of binding partners. Nonetheless, it would be of interest to determine the protein-protein interactions mediating the effect of 14-3-3 sigma on AR. This could be done with a candidate-by-candidate approach, where known A R co-regulators, such as SRC-1 [30] and NCoR [31], could be tested for their ability to interact with 14-3-3 sigma. One method to limit the possible list of candidate binding partners may be to assess the subcellular localization of 14-3-3 sigma in the presence and absence of androgens. This may establish in which cellular compartment the interaction is occurring. For instance, if the majority of cells show expression of 14-3-3 sigma only in the cytoplasm, this would suggest that most 14-3-3 sigma interactions would also be 140 occurring in the cytoplasm. In this case, it would be likely that 14-3-3 sigma was sequestering an AR co-repressor in the cytoplasm, thus allowing the A R to be more active in the nucleus. If, however, most 14-3-3 sigma was seen in the nucleus, it may suggest that 14-3-3 sigma is recruiting an AR co-activator to the nucleus, thus allowing the A R to be more active in the absence of androgens. The work of Huang, et al., showed that 14-3-3 sigma was mostly in the cytoplasm under androgen-free conditions in M12 prostate epithelial cells stably transfected with the A R [32]. Interestingly, treatment of these cells with either DHT or epidermal growth factor (EGF) resulted in the nuclear accumulation of 14-3-3 sigma [32]. Experiments need to be performed to determine if similar results are found in the LNCaP cell line, which expresses an endogenous AR. This same study demonstrated that transcription from the 14-3-3 sigma promoter was activated by treatment with androgens, suggesting that 14-3-3 sigma may be an androgen-regulated gene [32]. Future studies should determine if the AR directly regulates 14-3-3 sigma transcription by examining whether AR can bind its promoter. Sequence analysis of the 14-3-3 sigma promoter could be performed to see if consensus AREs are present. Additionally, chromatin immunoprecipitation (ChIP) could be utilized to determine if the AR is able to directly or indirectly interact with regions of the 14-3-3 sigma promoter. Several studies have shown that 14-3-3 sigma is downregulated with the development of prostate cancer [33-35]. However, both increased frequency of expression and increased nuclear localization of 14-3-3 sigma have been seen in more poorly differentiated tumours [32,33]. Studies examining the expression of 14-3-3 sigma need to be performed to determine its expression frequency in HRPC tissues. As 141 discussed in Chapter 1.4, there are multiple potential mechanisms leading to the survival and proliferation of prostate cancer cells in the absence of androgens. The ability of 14-3-3 sigma to increase the transcriptional activity of the AR in the absence of androgens indicates this may be a potential mechanism of ligand-independent activation of the A R in the subset of HRPC patients whose tumours express 14-3-3 sigma. Finally, exogenous factors are also able to influence the regulation of the transcriptional activity of the AR. For example, Zhu, et al. [36] recently showed that tumour-associated macrophages secreted IL-ip. Signalling events downstream of IL-ip altered the recruitment of the NCoR co-repressor to A R transcriptional complexes, with the result that the antiandrogen bicalutamide now acted as an AR agonist [36]. Experiments such as these demonstrate that the control of AR activity is a tightly regulated process, and that numerous events can lead to alterations in the activity of AR. It will be interesting to examine the upstream mechanisms leading to the expression of 14-3-3 sigma and whether external factors can influence the ability of 14-3-3 sigma to increase the transcriptional activity of the AR. 5.3 Conclusions In conclusion, the application of subtractive hybridization to the LNCaP hollow fibre model has identified numerous genes of interest in the hormonal progression of prostate cancer. The secreted isoform TMEFF2-S provides an excellent example of the capacity of subtractive hybridization to aid in the discovery of novel transcripts in such a model. Since it encodes a secreted protein, further examination of this splice variant may identify it as informative in predicting patient outcome. Furthermore, it will be 142 interesting to determine if the TMEFF2-S protein is bound by the therapeutic antibodies targeting TMEFF2. SSH, in conjunction with other high-throughput screening techniques, also led me to focus on the possible functions of the 14-3-3 family of proteins in advanced prostate cancer. The discovery that 14-3-3 sigma specifically increases the activity of the A R in the absence of androgens may be highly significant. Further studies will be required to determine if 14-3-3 sigma is active in a subset of HRPC patients. If so, its role in activating the AR may potentially be targeted for the development of novel therapeutics for the treatment of androgen-independent disease. 143 5.4 References 1. Alizadeh A A, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, Powell JI, Yang L, Marti GE, Moore T, Hudson J, Jr., Lu L, Lewis DB, Tibshirani R, Sherlock G, Chan WC, Greiner TC, Weisenburger DD, Armitage JO, Warnke R, Staudt L M , et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000;403(6769):503-511. 2. Perou C M , Sorlie T, Eisen MB, van de Rijn M , Jeffrey SS, Rees CA, Pollack JR, Ross DT, Johnsen H, Akslen L A , Fluge O, Pergamenschikov A, Williams C, Zhu SX, Lonning PE, Borresen-Dale A L , Brown PO, Botstein D. Molecular portraits of human breast tumours. Nature 2000;406(6797):747-752. 3. Sorlie T, Perou C M , Tibshirani R, Aas T, Geisler S, Johnsen H, Hastie T, Eisen MB, van de Rijn M , Jeffrey SS, Thorsen T, Quist H, Matese JC, Brown PO, Botstein D, Eystein Lonning P, Borresen-Dale A L . Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 2001 ;98(19): 10869-10874. 4. Singh D, Febbo PG, Ross K, Jackson DG, Manola J, Ladd C, Tamayo P, Renshaw A A , D'Amico A V , Richie JP, Lander ES, Loda M , Kantoff PW, Golub TR, Sellers WR. Gene expression correlates of clinical prostate cancer behavior. Cancer Cell 2002;l(2):203-209. 5. Ramaswamy S, Ross KN, Lander ES, Golub TR. A molecular signature of metastasis in primary solid tumors. Nat Genet 2003;33(l):49-54. 144 6. Ein-Dor L, Zuk O, Domany E. Thousands of samples are needed to generate a robust gene list for predicting outcome in cancer. Proc Natl Acad Sci U S A 2006;103(15):5923-5928. 7. Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative PCR. Genome Res 1996;6(10):986-994. 8. Stubbs AP, Abel PD, Golding M , Bhangal G, Wang Q, Waxman J, Stamp GW, Lalani EN. Differentially expressed genes in hormone refractory prostate cancer: association with chromosomal regions involved with genetic aberrations. Am J Pathol 1999; 154(5): 1335-1343. 9. Mohler JL, Morris TL, Ford OH, 3rd, Alvey RF, Sakamoto C, Gregory CW. Identification of differentially expressed genes associated with androgen-independent growth of prostate cancer. Prostate 2002;51(4):247-255. 10. Wainstein M A , He F, Robinson D, Kung HJ, Schwartz S, Giaconia JM, Edgehouse NL, Pretlow TP, Bodner DR, Kursh ED, et al. CWR22: androgen-dependent xenograft model derived from a primary human prostatic carcinoma. Cancer Res 1994;54(23):6049-6052. 11. Nagabhushan M , Miller C M , Pretlow TP, Giaconia JM, Edgehouse NL, Schwartz S, Kung HJ, de Vere White RW, Gumerlock PH, Resnick MI, Amini SB, Pretlow TG. CWR22: the first human prostate cancer xenograft with strongly androgen-dependent and relapsed strains both in vivo and in soft agar. Cancer Res 1996;56(13):3042-3046. 145 12. Dhanasekaran SM, Barrette TR, Ghosh D, Shah R, Varambally S, Kurachi K, Pienta KJ, Rubin M A , Chinnaiyan A M . Delineation of prognostic biomarkers in prostate cancer. Nature 2001;412(6849):822-826. 13. Hubert RS, Vivanco I, Chen E, Rastegar S, Leong K, Mitchell SC, Madraswala R, Zhou Y, Kuo J, Raitano AB, Jakobovits A, Saffran DC, Afar DE. STEAP: a prostate-specific cell-surface antigen highly expressed in human prostate tumors. Proc Natl Acad Sci U S A 1999;96(25): 14523-14528. 14. Afar DE, Bhaskar V, Ibsen E, Breinberg D, Henshall SM, Kench JG, Drobnjak M , Powers R, Wong M , Evangelista F, O'Hara C, Powers D, DuBridge RB, Caras I, Winter R, Anderson T, Solvason N, Strieker PD, Cordon-Cardo C, Scher HI, Grygiel JJ, Sutherland RL, Murray R, Ramakrishnan V, Law DA. Preclinical validation of anti-TMEFF2-auristatin E-conjugated antibodies in the treatment of prostate cancer. Mol Cancer Ther 2004;3(8):921-932. 15. Zhao X Y , Schneider D, Biroc SL, Parry R, Alicke B, Toy P, Xuan JA, Sakamoto C, Wada K, Schulze M , Muller-Tiemann B, Parry G, Dinter H. Targeting tomoregulin for radioimmunotherapy of prostate cancer. Cancer Res 2005;65(7):2846-2853. 16. Horie M , Mitsumoto Y, Kyushiki H, Kanemoto N, Watanabe A, Taniguchi Y, Nishino N, Okamoto T, Kondo M , Mori T, Noguchi K, Nakamura Y, Takahashi E , Tanigami A. Identification and characterization of TMEFF2, a novel survival factor for hippocampal and mesencephalic neurons. Genomics 2000;67(2):146-152. 146 17. Glynne-Jones E, Harper ME, Seery LT, James R, Anglin I, Morgan HE, Taylor KM, Gee JM, Nicholson RL TENB2, a proteoglycan identified in prostate cancer that is associated with disease progression and androgen independence. Int J Cancer 2001;94(2): 178-184. 18. 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(20):5474-5478. 19. Hobisch A, Eder IE, Putz T, Horninger W, Bartsch G, Klocker H, Culig Z. Interleukin-6 regulates prostate-specific protein expression in prostate carcinoma cells by activation of the androgen receptor. Cancer Res 1998;58(20):4640-4645. 20. Feldman BJ, Feldman D. The development of androgen-independent prostate cancer. Nat Rev Cancer 2001;l(l):34-45. 21. Rifkin DB. Latent transforming growth factor-beta (TGF-beta) binding proteins: orchestrators of TGF-beta availability. J Biol Chem 2005;280(9):7409-7412. 22. Chang C, Eggen BJ, Weinstein DC, Brivanlou AH. Regulation of nodal and BMP signaling by tomoregulin-1 (X7365) through novel mechanisms. Dev Biol 2003;255(1):1-11. 23. Harms PW, Chang C. Tomoregulin-1 (TMEFF1) inhibits nodal signaling through direct binding to the nodal coreceptor Cripto. Genes Dev 2003;17(21):2624-2629. 24. Tyagi RK, Lavrovsky Y, Ahn SC, Song CS, Chatterjee B, Roy AK. Dynamics of intracellular movement and nucleocytoplasmic recycling of the ligand-activated androgen receptor in living cells. Mol Endocrinol 2000; 14(8): 1162-1174. 147 25. Henttu P, Liao SS, Vihko P. Androgens up-regulate the human prostate-specific antigen messenger ribonucleic acid (mRNA), but down-regulate the prostatic acid phosphatase mRNA in the LNCaP cell line. Endocrinology 1992;130(2):766-772. 26. Gregory CW, Hamil K G , Kim D, Hall SH, Pretlow T G , Mohler JL, French FS. Androgen receptor expression in androgen-independent prostate cancer is associated with increased expression of androgen-regulated genes. Cancer Res 1998;58(24):5718-5724. 27. Holzbeierlein J, Lai P, LaTulippe E, Smith A, Satagopan J, Zhang L , Ryan C, Smith S, Scher H, Scardino P, Reuter V, Gerald WL. Gene expression analysis of human prostate carcinoma during hormonal therapy identifies androgen-responsive genes and mechanisms of therapy resistance. Am J Pathol 2004;164(l):217-227. 28. Wilker EW, Grant RA, Artim SC, Yaffe MB. A structural basis for 14-3-3sigma functional specificity. J Biol Chem 2005;280(19): 18891-18898. 29. Benzinger A, Muster N, Koch HB, Yates JR, 3rd, Hermeking H. Targeted proteomic analysis of 14-3-3 sigma, a p53 effector commonly silenced in cancer. Mol Cell Proteomics 2005;4(6):785-795. 30. Ma H, Hong H, Huang SM, Irvine RA, Webb P, Kushner PJ, Coetzee GA, Stallcup MR. Multiple signal input and output domains of the 160-kilodalton nuclear receptor coactivator proteins. Mol Cell Biol 1999; 19(9):6164-6173. 31. Cheng S, Brzostek S, Lee SR, Hollenberg A N , Balk SP. Inhibition of the dihydrotestosterone-activated androgen receptor by nuclear receptor corepressor. Mol Endocrinol 2002; 16(7): 1492-1501. 148 32. Huang D, Liu X, Plymate SR, Idowu M , Grimes M , Best A M , McKinney JL, Ware JL. Proteomic identification of 14-3-3 sigma as a common component of the androgen receptor and the epidermal growth factor receptor signaling pathways of the human prostate epithelial cell line Ml2. Oncogene 2004;23(41):6881-6889. 33. Cheng L, Pan CX, Zhang JT, Zhang S, Kinch MS, Li L , Baldridge L A , Wade C, Hu Z, Koch MO, Ulbright T M , Eble JN. Loss of 14-3-3sigma in prostate cancer and its precursors. Clin Cancer Res 2004;10(9):3064-3068. 34. Lodygin D, Diebold J, Hermeking H. Prostate cancer is characterized by epigenetic silencing of 14-3-3sigma expression. Oncogene 2004;23(56):9034-9041. 35. Mhawech P, Greloz V, Assaly M , Herrmann F. Immunohistochemical expression of 14-3-3 sigma protein in human urological and gynecological tumors using a multi-tumor microarray analysis. Pathol Int 2005;55(2):77-82. 36. Zhu P, Baek SH, Bourk E M , Ohgi K A , Garcia-Bassets I, Sanjo H, Akira S, Kotol PF, Glass CK, Rosenfeld M G , Rose DW. Macrophage/cancer cell interactions mediate hormone resistance by a nuclear receptor derepression pathway. Cell 2006;124(3):615-629. 149 6 APPENDIX 6.1 Animal care certificate The University of British Columbia Animal Care Certificate Appl ication N umber: A05-1794 Investigator or Course Director: Marianne Sadar Department: Medicine, Department of Animals Approved: Mice Male athymic Nude mice, B A L B / c Strain 180 Start Date: November 1,2005 Approval Date: January 6,2006 Funding Sources: Funding Agency: National Institutes of Health Funding Title: Genomic and proteomic analysis of androgen independent prostate cancer Unfunded title: N / A The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the C C A C and some granting agencies. Funding Agency: Funding Title: Health Canada Proteomics associated with the progression of prostate cancer to androgen-independence. A copy of this certificate must be displayed in your animal facility Office of Research Services and Administration 102, 6190 Agronomy Road, Vancouver, V6T 1Z3 Phone: 604-827-5111 Fax: 604-822-5093 150 6.2 Biohazard approval certificate T h e University of British C o l u m b i a Biohazard Approval Certificate P R O T O C O L N U M B E R : H03-0057 I N V E S T I G A T O R O R C O U R S E D I R E C T O R : Sadar, Marianne D E P A R T M E N T : Medicine P R O J E C T O R C O U R S E T I T L E : Genomic and proteomic analysis of androgen independent prostate cancer A P P R O V A L D A T E : 06-01-04 A P P R O V E D C O N T A I N M E N T L E V E L : 2 F U N D I N G A G E N C Y : National Institutes of Health The Principal investigator/Course Director is responsible for ensuring that all research or course work involving biological hazards is conducted in accordance with the Health Canada, Laboratory Biosafety Guidelines, (2nd Edition 1996). Copies of the Guidelines (1996) are available through the Biosafety Office, Department of Health, Safety and Environment, Room 50 - 2075 Wesbrook Mail, UBC, Vancouver, BC, V6T 1Z1, 822-7596, Fax: 822-6650. Approval of the UBC Biohazards Committee by one of: Chair, Biosafety Committee Manager, Biosafety Ethics Director, Office of Research Services This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required. A copy of this certificate must be displayed in your facility. Office of Research Services 102. 6190 Agronomy Road, Vancouver VST 1Z3 Phone: 604-827-5111 FAX: 604-822-5093 151 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092958/manifest

Comment

Related Items