UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of relaxin in prostate cancer Thompson, Vanessa Camille 2007

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2007-319390.pdf [ 10.5MB ]
Metadata
JSON: 831-1.0100760.json
JSON-LD: 831-1.0100760-ld.json
RDF/XML (Pretty): 831-1.0100760-rdf.xml
RDF/JSON: 831-1.0100760-rdf.json
Turtle: 831-1.0100760-turtle.txt
N-Triples: 831-1.0100760-rdf-ntriples.txt
Original Record: 831-1.0100760-source.json
Full Text
831-1.0100760-fulltext.txt
Citation
831-1.0100760.ris

Full Text

T H E R O L E OF R E L A X I N I N P R O S T A T E C A N C E R by V A N E S S A C A M I L L E T H O M P S O N B . S c , The University of Victoria, 2001 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Genetics) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A July 2007 © Vanessa Camil le Thompson, 2007 Abstract Prostate cancer is the leading cause of cancer in men, and there is currently a lack of novel treatment options for this disease. Relaxin, part of the insulin superfamily, is a potent peptide hormone normally produced and secreted by the human prostate. c D N A and tissue microarray analyses, indicated that relaxin is highly overexpressed in prostate cancer progression to androgen independence (Al), and is negatively regulated by androgens. Characterization of the relaxin receptor, L G R 7 , in xenografts and human patient tissue microarrays (TMAs) indicated that L G R 7 is expressed in the stroma and epithelia, suggesting that relaxin may act in an autocrine and/or paracrine fashion. Relaxin is known to increase angiogenesis through upregulation of vascular endothelial growth factor, and matrix metalloproteases. To elucidate novel pathways that may be upregulated by relaxin in prostate cancer, the effects of relaxin overexpression in the LNCaP prostate cancer xenograft model were analysed by gene expression microarrays. A novel discovery is that the protocadherinY (PCDHY)/Wnt pathway is upregulated by relaxin, resulting in P-catenin translocation from the cell membrane to the cytoplasm and upregulation of W n t l l , which can stimulate proliferation, transformation, and migration. Relaxin is well characterized in the uterus and ovary to increase insulin-like growth factor-I (IGF-I); relaxin may also increase IGF-I in prostate cancer. Since IGF-I and various other growth factor signaling is mediated by insulin receptor substrate-2 (IRS-2), and IGF-I signaling is important in prostate cancer progression to A l , we investigated the role of IRS-2 in LNCaP cells. IRS-2 mRNA is dramatically upregulated by androgens, and decreased by IGF-I in androgen treated cells. IRS-2 was knocked down using IRS-2 s iRNA to explore the interaction between the IGF axis and androgen signalling. Combined IGF-I and androgen treatment reduced transcription of androgen regulated genes, n prostate specific antigen, insulin degrading enzyme, vascular endothelial growth factor, and clusterin. This indicates that in the presence of IGF-I treatment, IRS-2 m R N A is required for normal androgen mediated transcription. Due to the potential for relaxin to induce proliferation and metastases, and the necessity of IRS-2 to mediate androgen action, relaxin and IRS-2 are intriguing targets for novel, targeted therapeutics for prostate cancer. in Table of Contents Abstract i i Table of Contents iv List of Tables vi List of Figures vii List of Abbreviations ix Acknowledgements xi Dedication xii Co-Authorship Statement xiii Chapter 1. Introduction 1 1.1 The Prostate 1 1.2 Relaxin 9 1.3 Hypothesis 24 1.4 Objectives 24 1.5 References 26 Chapter 2. Relaxin Becomes Upregulated During Prostate Cancer Progression to Androgen Independence and is Negatively Regulated by Androgens 40 2.1 Introduction 40 2.2 Materials and Methods 43 2.3 Results 50 2.4 Discussion 59 2.5 References 66 Chapter 3. Relaxin Drives Wnt Signalling Through Upregulation of P C D H Y in Prostate Cancer 71 3.1 Introduction 71 iv 3.2 Materials and Methods 74 3.3 Results 80 3.4 Discussion 92 3.5 References 98 Chapter 4. Suppression of Androgen Signaling by IGF-I due to s iRNA Depletion of IRS-2 102 4.1 Introduction 102 4.2 Materials and methods 104 4.3 Results 107 4.4 Discussion 117 4.5 References 121 Chapter 5. Discussion 125 5.1 Relaxin regulation in prostate cancer progression 125 5.2 Relaxin regulated Wntl 1 pathway 127 5.3 IRS-2 mediates androgen regulated gene expression 130 5.4 Future Directions 132 5.5 References 136 Appendix I. Genes upregulated at least 5-fold in R l 881-treated cells 139 Appendix II. Genes downregulated at least 5-fold in R l 881-treated cells 140 Appendix III. University of British Columbia Animal Care Certificate 141 Appendix IV. Primers for quantitative RT-PCR 142 v List of Tables Table 3.1 List of Figures Figure 1.1 Histologic grades of prostate adenocarcinoma 4 Figure 1.2 L H R H axis for anti-androgen therapy 6 Figure 1.3 Relaxin structure 12 Figure 1.4 Schematic of LGR7 tertiary structure 13 Figure 1.5 Relaxin signaling pathways used by the relaxin receptor, LGR7 in multiple cell types 15 Figure 2.1 Relaxin levels in LNCaP cells decrease with increased androgen concentration 52 Figure 2.2 Relaxin levels in LNCaP cells decrease with increased duration of androgen treatment 53 Figure 2.3 Relaxin mRNA levels increase with progression to androgen independence in LNCaP subcutaneous xenograft model in nude mice 56 Figure 2.4 Relaxin levels increase in human prostate cancer tissue with increased duration of neoadjuvant hormone therapy (NHT) 58 Figure 3.1 Relaxin is overexpressed in L N C a P - R L X stable cell lines 81 Figure 3.2 Relaxin and LGR7 protein levels are higher in L N C a P - R L X xenografts grown in intact mice compared to controls, using IHC 82 Figure 3.3 LGR7 is expressed in malignant epithelium and stromal cells of prostate. 84 Figure 3.4 L N C a P - R L X xenograft tumours grow more rapidly than LNCaP-GFP xenografts in intact mice 86 Figure 3.5 W n t l l staining is increased after six months NHT, and sustained in A l metastases in human patient samples 90 vi i Figure 3.6 Relaxin overexpression causes p-catenin to translocate to the cytoplasm in intact mice 91 Figure 3.7 Model of relaxin driving upregulation of P C D H Y , with P-catenin translocating to the cytoplasm, and downstream upregulation of Wntl 1 96 Figure 4.1 IGF-I treatment reduces IRS-2 mRNA levels 109 Figure 4.2 A R siRNA knocks down A R mRNA and abrogates androgen mediated transcription of PSA and IRS-2 110 Figure 4.3 IRS-2 siRNA dose curve 112 Figure 4.4 IRS-2 is required for androgen mediated transcription in the presence of R1881 and IGF-I 113 Figure 4.5 Effect of IRS-2 siRNA on A R and IGF signaling molecules 116 viii List of Abbreviations A l androgen independence A R androgen receptor A T B F 1 at-binding transcription factor 1 Bcl-2 B-cel l CLL/ lymphoma 2 b H L H class B basic helix-loop-helix c A M P cyclic A M P C 0 2 carbon dioxide C R E B c A M P response element binding protein CSS charcoal stripped serum D H T dihydrotestosterone D R E digital rectal exam E G F epidermal growth factor E R K extracellular signal-regulated kinase F C S fetal calf serum F O X O forkhead box 0 F S H follicle stimulating hormone G F P green fluorescent protein G n R H gonadotrope releasing hormone G P R G protein coupled receptor G S T glutathione-S-transferase IDE insulin degrading enzyme IGF insulin-like growth factor I G F B P IGF binding protein IGF-IR insulin-like growth factor receptor type I IHC immunohistochemistry I N S L insulin-like peptides IRES internal ribosome entry site IRS insulin receptor substrate K G F keratinocyte growth factor K L K kallikrein L G R 7 relaxin receptor L G R 8 INSL3 receptor L H luteinizing hormone L H R H luteinizing hormone releasing hormone L N C a P -G F P L N C a P xenografts stably transfected with empty vecot control L N C a P -R L X L N C a P xenografts stably overexpressing relaxin M metastasis M A P K mitogen activated protein kinase M E K M A P K / E R K kinase M M P matrix metalloprotease N lymph node N H T neadjuvant hormone therapy N O nitric oxide N O S nitric oxide synthase O D C ornithine decarboxylase P C D H protocadherin P C R polymerase chain reaction PI3K PI3 kinase P I N prostatic intraepithelial neoplasia P K A protein kinase A P K C protein kinase C P S A prostate specific antigen P T E N phosphatase and tensin homolog Q - P C R semi-quantitative real time - P C R relaxin relaxin H2 R L X relaxin R T K receptor tyrosine kinase s.c. sub cutaneous SF serum free s i R N A short interfering R N A S R C steroid receptor coactivator S R E sterol response element S R E B P sterol response element-binding protein S T A T signal transducer and activator of transcription T tumor growth T-cell factor and lymphoid enhancer factor family of T C F / L E F 1 transcription factors T F E transcription factor for immunoglobulin heavy-chain enhancer T I M P tissue inhibitors of metalloprotease T M A tissue microarray T N M tumor, node, metastasis T R A M P transgenic adenocarcinoma of mouse prostate U V ultraviolet V E G F vascular endothelial growth factor Wnt wingless-type M M T V integration site family x Acknowledgements M y experience as a PhD student has been made more enjoyable and fulfilling by many people. I would like to acknowledge the friends and colleagues who have helped along the way. John Cavanagh and Robert Shukin are two great friends whose expertise and insight have been invaluable along the way. M y fellow graduate students have inspired me and kept me sane. I would like to thank Dr. Dawn Cochrane, Dr. Lindsay Brown, Leah Prentice, and Melanie Lehman for all their feedback and support. Dr. Susan Moore and Dr. Tanya Day have been extremely generous with their assistance. I would like to thank my supervisor, Dr. Colleen Nelson, for her guidance, encouragement, and motivation. x i Dedication To my husband, Cory Wood. Your love and support throughout this process have been immeasurable. Co-Authorship Statement The work presented in this thesis was conducted under the guidance of my supervisor, Dr. Colleen Nelson, and is due to collaboration with several individual researchers. The specifics of each researcher's involvement are detailed below. Chapter 2. Tanis Morris performed the xenograft experiment, and contributed the northern blot, Ladan Fazli stained the tissue microarrays and scored them for staining intensity, John Cavanagh developed the antibody against relaxin, Tatyana Hamilton designed the relaxin Q - P C R primers and probes. Chapter 3 . Antonio Hurtado-Coll photomicrographed tissue microarrays and managed automated analysis software, Dmitry Turbin scored for subcellular P-catenin localization in xenograft tissue microarrays, Ladan Fazl i manually scored all remaining tissue microarrays. Chapter 4. Steve Hendy treated cells for microarray analysis; Manuel Altamirano hybridized c D N A to microarrays, and analyzed microarray data with Mauricio Neira. Melanie Lehman performed statistical analyses; A m y Lubik performed western blots. Xl l l Chapter 1. Introduction Prostate cancer is the leading cause of cancer in men, and there is currently a lack of targeted treatment options for this disease. Novel research performed in our laboratory demonstrated that an increase in relaxin, a normal prostate secretory protein, correlated with progression to androgen independence in the L N C a P prostate cancer cell line model. Relaxin is of interest due to its pleiotrophic properties, including its involvement in the upregulation of several factors known to be angiogenic or metastatic. In this dissertation, I describe androgen regulation of relaxin in prostate cancer, and novel roles for relaxin in prostate cancer. Novel relaxin-regulated pathways were first elucidated using c D N A microarray analysis. From these studies it was shown that W n t l 1 and P C D H Y are upregulated in relaxin overexpressing prostate cancer cells, which may be a mechanism by which relaxin increases cellular migration. In addition, I show that IRS-2, a signaling molecule for insulin and insulin-like growth factor pathways, is also downregulated by relaxin, and is strongly upregulated by androgens. A s relaxin belongs to the insulin superfamily, the interaction between relaxin and IRS-2 is of interest. Based on the results presented in this dissertation, relaxin and its downstream targets may be suitable therapeutic targets for prostate cancer. 1.1 The Prostate Prostate Cancer Prostate cancer is the most common form of cancer in men, with an expected 22,300 new cases being diagnosed this year in Canada, and the third leading cause of cancer related deaths (Canada, 2007). Prostate cancer is thought to progress from 1 prostatic intraepithelial neoplasia (PIN) to adenocarcinoma, and then spread to secondary sites including bone, lymph node, liver, and lung, with bone being the most predominant site of metastasis (Bagi, 2005). One other major form of prostate disease is benign prostatic hyperplasia, which is marked by increased prostate size, but is not considered a precursor to the malignant disease (Chokkalingam et al., 2003). Prostate cancer is a multifocal and multiclonal disease; one individual may have multiple tumors, which arise independently through different mechanisms such as loss of P T E N (Jin-Tang Dong, 2006; L i et a l , 1997; Steck et a l , 1997). P T E N loss occurs in 23% of high grade PIN, and 68% of prostate adenocarcinomas, indicating a role for P T E N in prostate cancer development (Yoshimoto et al., 2006). Other tumor suppressor genes deleted or mutated in prostate cancer include A T B F 1 (Sun et al., 2005), and p53 (Grignon et al., 1997). However, there appears to be a lack of consistencies of mutations in particular genes and it is accepted that multiple pathways are likely involved in the development and progression of prostate cancer. There are several established risk factors associated with prostate cancer; most important is age, as the risk of developing prostate cancer increases with age. In addition to age, a family history of prostate cancer is also associated with increased risk of prostate cancer, as one first degree relative (father, brother, or son) with the disease increases risk two-fold, and two first degree relatives further increases the relative risk to 8.8 (Klein, 1995). Ethnic background is also a risk factor; men of African descent have a relative risk of 1.8 over men of European descent (Morton, 1994), and men from As ia have the lowest risk. However, second generation Asians-Americans have increased risk, indicating that lifestyle is important. One lifestyle difference is diet; a higher fat 2 diet has a relative risk of 1.31 (Key, 1995), while various dietary components have been indicated to have apparent protective effects, including lycopenes, in some studies, components of green tea, vitamin E and selenium (Reviewed in Gupta, 2006). These are all antioxidants, and may serve to protect the cells from D N A damage. This is particularly interesting, considering that expression of glutathione-S-transferase (GST), which prevents oxidative stress-induced D N A damage, is commonly lost early in prostate cancer due to promoter methylation (Brooks et al., 1998; Lee et al., 1994; Mi l la r et al., 1999). In addition to established risk factors, several recent reports have shown that several abnormalities, including small nucleotide polymorphisms, along 3 regions of chromosome 8q24 are independently associated with an increased risk for the development of prostate cancer (Gudmundsson et al., 2007; Haiman et al., 2007; Platz, 2007; Schumacher et al., 2007; Suuriniemi et al., 2007; Wang et al., 2007; Witte, 2007; Yeager et al., 2007). This sheds some light on the poorly understood genetic risk of prostate cancer. Prostate cancer is detected through the combined use of a digital rectal exam (DRE), prostate specific antigen (PSA) tests, and a transrectal biopsy. The P S A test measures the concentration of serum P S A , an enzyme produced in the normal prostate, which, when present in the serum, is an indication of tumor size in untreated prostate cancer. A n elevated serum P S A of greater than 4-10 ng/ml, and a positive D R E suggest a cancerous lesion, which would then result in a transrectal biopsy for characterization of the potential tumor. Using the Gleason grading system (Gleason and Mellinger, 1974), prostate cancer is given a value of 1-5 based on glandular characteristics, with a lower grade given to more differentiated and normal prostate glands, and a higher grade is 3 given to less differentiated glands (Fig. 1.1). The Gleason score is the additive value of the Gleason grade from the most common pattern with the grade of the second most common; the Gleason score ranges from 2-10. A well defined cancer could have a Gleason score of 1+2=3 and an aggressive cancer could have a Gleason score of 4+5 = 9 (Humphrey, 2004). A well differentiated cancer is often smaller than a dedifferentiated cancer, although these generalizations do not always hold true. In part for this reason, P R O S T A T I C A D E N O C A R C I N O M A ( H i s t o l o g i c G r a d e s ) Figure 1.1 Histologic grades of prostate adenocarcinoma, as characterized by Dr. Gleason. Adapted from Humphrey et al. (Humphrey, 2004). 4 prostate cancer can also be characterized clinically using tumor, node, metastasis ( T N M ) staging, which indicates the extent of tumor growth (T), the number of local lymph nodes affected (N), and the degree of metastatic spread (M) (Hoedemaeker et al., 2000). Based on Gleason score, T N M staging, and P S A value, the urologist can predict pathological stage, and determine appropriate treatment (Ayyathurai et al., 2006; Partin et al., 1997; Partin et al., 2001). Androgen dependence and treatment Prostate cancer is readily curable i f the cancer is still contained within the prostatic capsule. In this condition, common primary treatment strategies include radiation and surgery. In addition, patients with low volume cancer or those with other co-morbidities may receive "watchful waiting" - observation of the disease without any treatment. I f the cancer has progressed beyond the capsule, the major form of treatment is androgen withdrawal therapy, as prostate cancer cells are androgen dependent (Huggins and Hodges, 1941). This treatment targets the prostate and metastatic cells, yet is not without side effects, including osteoporosis, fatigue, diminished cognitive function, and loss of libido. Androgen deprivation therapy functions through two mechanisms, firstly through luteinizing hormone releasing hormone ( L H R H ) agonists or orchiectomy to reduce total circulating androgen levels via reduction of testosterone production, and secondly through A R antagonists by directly inhibiting androgen receptor (AR) signaling (Tammela, 2004). The original and most effective means of reducing circulating androgen levels is orchiectomy, as the removal of the testes eliminates the major source of androgens with 5 the adrenal glands being the remaining source of androgens (Huggins and Hodges, 1941). Disrupting the L H R H axis is also another option. L H R H acts on the gonadotropic cells of the anterior pituitary, resulting in the synthesis and release of luteinizing hormone (LH) and follicle stimulating hormone (FSH). These two gonadotropes stimulate the production of testosterone, which in turn provides the negative feedback signal to reduce testosterone production (Fig. 1.2). Treatment with L H R H agonists, such as leuprolide, goserelin, and buserelin, blocks L H secretion. hypothalamus j J gland prostate Figure 1.2 L H R H axis for anti-androgen therapy Initially testosterone production increases for 4-5 days; however, castrate level testosterone is present within 2-3 weeks. L H R H antagonists block L H R H action directly, resulting in immediate blockage of the L H R H axis, and testosterone production, presenting a benefit over L H R H agonists (reviewed in Brawer, 2006; Debruyne, 2004; Goldenberg et a l , 1995; Tammela, 2004). Another treatment option is the use of antiandrogens, which inhibit androgen signaling through direct binding of the A R . Nonsteroidal antiandrogens, such as flutamide, nilutamide, and bicalutamide, bind to the A R ligand binding domain, and inhibit activation of the receptor. Antiandrogen-bound A R often binds D N A at target elements but prevents transcription of androgen regulated genes. However, nonsteroidal antiandrogens block the negative feedback effect of testosterone in the hypothalamus, resulting in higher testosterone levels and permitting normal sexual potency (Farla et al., 2005; Furr and Tucker, 1996). Steroidal antiandrogens, such as cyproterone acetate, compete with ligand for receptor binding similar to nonsteroidal antiandrogens, but impart a negative feedback on the L H R H axis, thereby reducing testosterone levels (reviewed in Tammela, 2004; Varenhorst et al., 1982). Intermittent androgen suppression is an alternate form of treatment which seeks to improve quality of life and time to androgen independence (A l ) , through cycling on and off androgen ablation. Recent clinical trials have suggested that intermittent androgen suppression is equal to or better than the current standard treatment. A clear benefit of this regime is the significant decrease in time a patient is subjected to hormonal treatment, which positively impacts a patient's quality of life (Brawer, 2006; Bruchovsky et al., 2006; reviewed in Tammela, 2004). 7 Androgen independence While androgen ablation is initially effective, inevitably the tumour wi l l progress to A l prostate cancer, in which A R is reactivated (Zhang et al., 2003). Feldman and Feldman (Feldman and Feldman, 2001) described 5 potential mechanisms for the development of A l , which are as follows; 1) A R hypersensitivity, permitting the receptor to utilize what little levels of androgen are present in functionally castrated men. One mechanism, present in 30% of patients studied, involves gene amplification of the A R , resulting in excess A R , and increasing the probability that low levels of ligand wi l l be bound by receptor (Koivisto et al., 1997; Visakorpi et al., 1995). Another mechanism leading to A R hypersensitivity involves increased 5-a-reductase activity, resulting in more intracellular dihydrotestosterone, the most active androgen. 2) The promiscuous receptor, which can be transactivated by ligands other than androgen, permits the transcription of androgen regulated genes in the absence of androgens. For instance, the L N C a P cell line which was derived from a prostate metastasis to lymph node and harbors a T877A mutation, permits A R transactivation by estrogen, progestens, and some anti-androgens (Veldscholte et a l , 1992). 3) A R activation in the absence of ligand may occur through phosphorylation of the A R by Her2/neu (Berger et al., 2006; Craft et a l , 1999), insulin-like growth factor-I (IGF-I), epidermal growth factor (EGF), or keratinocyte growth factor ( K G F ) signaling (Culig et al., 1994). 4) Growth and survival may occur through proliferative and anti-apoptotic mechanisms, such as an increase in Bcl -2 expression (Chi et al., 2001; Gleave et al., 1999), with no requirement of A R or ligand. 5) Cells are present in the prostate that are A l prior to treatment, and are clonally selected for by the absence of androgens (Isaacs, 1999). 8 1.2 Relaxin History Relaxin was initially discovered in 1926 by Frederick Hisaw, who discovered that serum from pregnant guinea pigs could lengthen the interpubic ligament in female virgin guinea pigs (Hisaw, 1926). The relaxin protein was isolated from the pregnant porcine corpora lutea of the ovaries, and was termed so for its lengthening or relaxing effects (Fevold et al., 1930). Relaxin was originally considered a hormone facilitating the ease of birth in pregnant mammals, through the lengthening of pubic ligaments and sacroiliac joints, cervical ripening, and inhibiting uterine contractions (Guico-Lamm and Sherwood, 1988; Hughes and Hollingsworth, 1997; Hwang et al., 1989; Hwang and Sherwood, 1988; Krantz et al., 1950). The effects on the ligaments, cervix and joints are due in part to collagen remodeling, characterized by decreased collagen production, increased matrix metalloprotease ( M M P ) production, and decreased tissue inhibitors of metalloproteases (TIMPs). The degree to which relaxin affects collagen remodeling is species specific, with the effect in humans and ruminants not as great as in rodents (guinea pig, and rat) and pig. In humans, elongation of the sacroiliac joints by relaxin is greater than the pubic symphysis; whereas in cows, relaxin causes cervical ripening (Fevold et al., 1930). In rats and pigs, relaxin expression is increased in the third trimester of pregnancy, and there is a relaxin surge near parturition (Ivell and Einspanier, 2002). In humans there is no relaxin surge at term; in fact, serum relaxin levels peak in the first trimester, but remain elevated throughout pregnancy (Bell et al., 1987; Eddie et al., 1986; Ivell and Einspanier, 2002; MacLennan et al., 1986a). Elevated serum relaxin 9 levels in the last trimester of pregnancy can result in joint pain due to increased elongation of the pubic symphisis and sacroiliac joints (MacLennan et al., 1986b). Relaxin production was initially found in the ovary, but further studies found relaxin production in the endometrium (Goldsmith and Weiss, 2005; Pardo et al., 1980; Pardo and Larkin, 1982; Pardo et al., 1984; Schmidt et al., 1984), the decidua, and the placenta (Koay et al., 1985; Schmidt et al., 1984) of humans, guinea pigs, pigs, and rhesus monkeys. Additionally, the human prostate (Ivell et al., 1989; Sokol et al., 1989; Yki-Jarvinen et al., 1983), and dogfish and boar testes (Dubois and Dacheux, 1978) were also found to express relaxin. Relaxin is essential to mammary growth, nipple development, and lactation in rats and mice (Hwang et al., 1991; Krajnc-Franken et al., 2004; Kuenzi and Sherwood, 1992). Other studies found that relaxin is expressed in and can bind to the heart (Osheroff and Ho, 1993), and brain (Burazin, 2000; Osheroff and Ho, 1993), suggesting non-reproductive functions for the hormone. Relaxin has also been shown to affect water intake, and is postulated to be responsible for water retention during pregnancy, indicating a role for relaxin binding in the brain (Danielson et al., 1999; Sunn et al., 2002; Zhao et al., 1995). The peptide structure A large increase in relaxin research occurred in the late 80's early 90's with the identification of the human relaxin genes, H I and H2, and the discovery that H2 was the predominant transcript of the ovary and prostate (Hudson et al., 1983; Ivell et al., 1989). Relaxin H I is transcribed, but protein has not been definitively detected in the prostate. Relaxin H2 is the primary relaxin protein of the prostate, and thus was the focus of study 10 within this thesis. Relaxin H2 wi l l be referred to as "relaxin" for the remainder of this thesis. The relaxin protein shares surprisingly little homology with rat or porcine relaxin, sharing less than 50% identity. In addition, relaxin homology differs by 6% between the two baleen whales, Balaeonoptera acutorostrata and B. edeni, while by less than 10% between the two whales and pig (Schwabe and Bullesbach, 1990). This lack of homology does not indicate a lack of cross reactivity between species. In fact, porcine relaxin is bioactive in many different species, including rat and human (Schwabe and Bullesbach, 1990). Due to similar chemical characteristics, relaxin was postulated to share structural homology with insulin; this was confirmed upon discovery of the porcine relaxin sequence in 1977 (Schwabe and McDonald, 1977). Great quantities of relaxin were obtainable from porcine ovaries; however, the paucity of human ovaries of pregnant women precluded discovery of the human relaxin gene until 1984 (Hudson et al., 1984). Relaxin is similar to insulin and the insulin-like growth factors in that they are translated as follows: signal peptide, B-chain, C-chain, A-chain (Gast, 1982; Stults et al., 1990). For both insulin and relaxin, the C-chain is cleaved from the A and B chains to create the mature peptide (Gast, 1982). Human relaxin tertiary structure is similar to both insulin and porcine relaxin (Eigenbrot et al., 1991). The A - and B-chains are composed of interacting alpha helices connected by two disulfide bonds; as well , the A-chain contains an intrachain disulfide bond (Fig. 1.3) (Canova-Davis et al., 1991; Stults et al., 1990). The insulin superfamily is made up of insulin, IGF-I, IGF-II, relaxin H I , relaxin H2, relaxin H3, and insulin-like peptides (INSL) 3-6. INSL3 is essential for the descent of the testes, through interaction with its receptor, L G R 8 (Kumagai et al., 2002; 11 Yamazawa et a l , 2007; Zimmermann et al., 1999). Insulin and IGF-I signal through the insulin receptor and the IGF-I receptor (IGF-IR), respectively, which are both receptor tyrosine kinases (RTKs) . Activation of the receptor results in activation of the intracellular signaling intermediates, insulin receptor substrate (IRS)-l and IRS-2, which then signal to downstream pathways such as Akt (Dearth et al., 2007). Relaxin and INSL3 bind to the G protein coupled receptors (GPR), L G R 7 and L G R 8 , respectively (Hsu et al., 2002). C oomain Figure 1.3 Relaxin structure. A ) Tertiary sturcutre of relaxin, demonstrating interacting alpha helices and cystine bonds. B) Schematic diagram of preprorelaxin, indicating signal peptide, cleavage sites, A - , B - , and C-chains, and cystine bonds. Adapted from (Ivell and Einspanier, 2002). 12 The receptor L G R 7 was identified as the relaxin receptor nearly 80 years after relaxin was first discovered (Hsu et al., 2002). L G R 7 is a leucine rich repeat G P R (Fig. 1.4) that shares Figure 1.4 Schematic of LGR7 tertiary structure. The extracellular domain is composed of leucine-rich repeats and an N-terminal ectodomain. The receptor spans the membrane seven times, characteristic of a GPR, and has an intracellular C-terminus. Adapted from (Ivell and Einspanier, 2002). structural homology with receptors for the gonadotropins, L H and F S H , and thyrotropin (Hsu et a l , 2000). Relaxin also binds to L G R 8 in vitro (Hsu et al., 2002), while INSL3 specifically binds to L G R 8 (Kumagai et al., 2002). Relaxin H3 specifically activates L G R 7 but not L G R 8 (Sudo et al., 2003). Additional GPRs, GPCR135 and GPCR142 , have been found to bind relaxin H3 in vitro; however, GPCR135 is likely to be the relaxin H3 receptor, whereas GPCR142 is activated by INSL5 (Sutton et al., 2005; Sutton et al., 2006). INSL4 and INSL6 are the remaining members of the insulin superfamily and currently no receptors have been identified for these two proteins. 13 L G R 7 m R N A is expressed in many tissues previously characterized to bind relaxin, such as the brain, kidney, testis, placenta, prostate, uterus, ovary, skin heart, adrenal (Hsu et al., 2002). Several L G R 7 splice variants that encode truncated proteins have been discovered and they can inhibit full length L G R 7 activation without interfering with relaxin binding to the membrane bound receptor (Scott et al., 2006). Relaxin action was originally characterized by increased cyclic A M P ( c A M P ) in the pubic symphysis (Braddon, 1978), and rat uterus (Cheah and Sherwood, 1980). c A M P is considered such a definitive response of relaxin bioactivity that it was the characteristic endpoint of several bioassays. Initially uterine strips were assayed for c A M P production, until cultured uterus cells were available that were sensitive and specific to relaxin adminstration (Fei et al., 1990; Kramer et al., 1990). Furthermore, the THP-1 monocyte cell line responds to relaxin administration with increased levels of c A M P (Parsell et al., 1996). The mechanism by which relaxin activation causes increased c A M P levels has not been fully elucidated. There is evidence indicating that G-protein signaling is involved in activating adenylate cyclase, as would be expected from a G P R (Hsu et al., 2002). However, there is evidence that mitogen activated protein kinase ( M A P K ) and tyrosine kinases are involved in relaxin signaling. Using R T K inhibitors, Bartsch et al (Bartsch et al., 2001) determined that relaxin-induced increases in c A M P were due to activated adenylate cyclase through inhibition of phosphodiesterase (Fig. 1.5). Relaxin causes an increase in activated M A P K , dependent on M E K activation (Zhang et al., 2002). One downstream effector activated by M A P K may be c A M P response element 14 Figure 1.5 Relaxin signaling pathways used by the relaxin receptor, L G R 7 in multiple cell types. Gas couples L G R 7 signalling to adnylate cyclase, resulting in increased c A M P . L G R 7 coupling to GPy subunits activates PI3K, which then activates P K C which in turn activates adenylate cyclase. PI3K can also signal through c-Raf to activate the M A P K signaling pathway, and can activate Akt . Adapted from (Bathgate et al., 2006). binding protein ( C R E B ) , although C R E B may also be activated by c A M P (Zhang et al., 2002). However, this M A P K activation can be blocked by protein kinase A ( P K A ) inhibition, indicating that relaxin-induced M A P K signaling is downstream of c A M P signaling (Bathgate et al., 2006). There is a biphasic c A M P response to relaxin, through two signaling mechanisms (Halls et al., 2006; Nguyen and Dessauer, 2005). The first, acute response signals through the G-protein Gas, resulting in increased c A M P through activation of adenylate cyclase. A second mechanism responsible for the sustained elevation of c A M P acts through GPy to activate PI3 kinase (PI3K) (Dessauer and Nguyen, 2005; Halls et al., 2005; Nguyen and Dessauer, 2005). The mediator between PI3K activation and adenylate cyclase has been shown to be protein kinase C (PKC) (, (Nguyen and Dessauer, 2005). Not all signaling pathways are activated by relaxin in 15 every cell, in support of different responses to relaxin in different cells (Dessauer and Nguyen, 2005; Dschietzig et al., 2003; Nguyen and Dessauer, 2005). Relaxin functions in different tissues The major site of relaxin production is the corpus luteum, which expresses far more relaxin in pregnant females than cycling females (Bogie et al., 1995; Soloff et al., 1992). Relaxin expression in the ovary is secreted into the blood stream, and acts in an endocrine fashion on several different tissues throughout pregnancy. Relaxin is mitogenic in porcine follicular granulosa and thecal cells (Bagnell et al., 1993). Relaxin is also synthesized in the placenta, decidua, brain, prostate, pancreas, and kidney (Bogie et a l , 1995; Gunnersen et a l , 1995). Relaxin binds to the uterus, likely to L G R 7 (Osheroff et al., 1990; Zhao et al., 1999), resulting in multiple downstream effects. Relaxin increases uterine growth and elevates E-cadherin protein levels (Hall et al., 1990; Hal l et al., 1992; M i n et al., 1997; Ryan et al., 2001). Relaxin increases vascular endothelial growth factor ( V E G F ) in endometrial cells in vitro, and women treated with systemic relaxin experienced menometrorrhagia, likely due to increased angiogenesis in the endometrium (Palejwala et al., 2002; Unemori et al., 1999). Increased endometrial angiogenesis occurs in part by V E G F , and facilitates growth of the endometrium throughout the menstrual cycle and pregnancy (Girling and Rogers, 2005). Relaxin has been shown to inhibit murine myometrial contractility, likely as a means to inhibit premature parturition. This is at least in part mediated by increased nitric oxide production, through upregulation of endothelial nitric oxide synthase (Bani et al., 1999a). 16 Relaxin is expressed at the same time or prior to prolactin expression, during progression of the menstrual cycle, followed by IGF binding protein-1 (IGFBP-1) expression late in the cycle. A l l three proteins are expressed in early pregnancy, while they are expressed strongly at late pregnancy in the stroma (Bryant-Greenwood et al., 1993). A relaxin-induced increase of IGFBP-I and prolactin in human decidual and endometrial stromal cells (Gao et al., 1995; Gao et al., 1994) is mediated by P K A signaling for both genes (Tang et a l , 2005). Additionally, the relaxin induction of prolactin is mediated by M A P K activation (Tang et al., 2005). Relaxin induces uterine growth in the pig, which may be mediated in part by increased in IGF-I, IGF-II, I G F B P -2, and IGFBP-3 . IGF-I and IGF-II were elevated only at the protein level, suggesting either increased translation, or increased half life of the protein through interaction with IGFBP-2 and IGFBP-3 . Interestingly, IGFBP-1 was not affected by relaxin in this model (Ohleth et al., 1997). The proliferative effects associated with relaxin in the uterus may be because relaxin upregulates the IGF-I axis. Relaxin binding in the cervix is enhanced by estrogen priming (Huang et al., 1993; Osheroff et al., 1990), and results in cervical softening, growth (Hwang et al., 1989), and increased extensibility in rats (Sherwood et al., 2000). The mechanisms by which relaxin alters the cervix are varied among species. In the rat, relaxin increases cervical size through inhibition of apoptosis (Zhao et al., 1999). Nitric oxide mediates the relaxin induced growth, but not the extensibility or softening of the cervix in rats (Sherwood et al., 2000). However, in human cervical stromal cell culture, relaxin does not change growth of cervical cells, but causes collagen remodeling to soften cervix (Hwang et al., 1996). In the rat, relaxin affects cervical softening through less compact 17 collagen fiber bundles and shorter and less dense elastin fibers (Lee et al., 1992), implying collagen remodeling is not limited to pubic ligaments and joints. In the porcine uterine cervix, relaxin-mediated collagen remodeling is mediated by elevated TIMP-1 and -2 expression (Lenhart et al., 2002). Relaxin plays a minor role in mammary gland growth and development in the rat and mouse (Kass et al., 2001), but is essential for nipple development and lactation (Hwang et al., 1989). Relaxin depletion in pregnant rats and mice results in impaired nipple development, with decreased size and altered histology, resulting in inability of pups to attach to the nipple, and stimulate prolactin (Kuenzi and Sherwood, 1992; Zhao et al., 1999). In addition, in mice there is no lengthening of the pubic symphisis, resulting in impaired delivery (Zhao et al., 1999). L G R 7 knockout mice recapitulate the same phenotype, implying that relaxin signals through L G R 7 to cause nipple development, and pubic symphisis elongation. Several binding sites in the rat brain have been characterized, with binding to the circumventricular organs, (subfornical organ and organum vasculosum of the lamina terminalis) and the paraventricular and supraoptic nuclei, being potential ways by which relaxin controls vascular volume and blood pressure (Osheroff et al., 1990; Osheroff and Phillips, 1991). Relaxin acts on the subfornical organ, an interface between the blood, brain, and cerebrospinal fluid, to increase drinking in rats (Hornsby et al., 2001; Sunn et al., 2002). Relaxin acts through increasing angiotensin II receptors in the subfornical organ (Hornsby et al., 2001). Depleting relaxin in the subfornical organ results in a cessation of drinking urges associated with pregnancy in rats (Hornsby et al., 2001). 18 Relaxin binds to the heart through L G R 7 regardless of estrogen treatment, unlike the uterus, which requires estrogen priming for relaxin receptor production (Osheroff et al., 1992; Zhao et al., 1999). Relaxin protects against myocardial injury through protection from ischemia, in part through involvement of nitric oxide (Bani et al., 1998; Masini et al., 1997). Nitric oxide production in mast cells is elevated by relaxin (Masini et al., 1994; Masini et al., 1995), which inhibits histamine release from mast cells in the heart (Masini et al., 1997), and results in an inhibition of cardiac anaphylaxis (Masini et al., 2002; Ndisang et al., 2001). In addition, relaxin can inhibit apoptosis in cardiomyocytes through an Akt pathway (Moore et al., 2007). This has led to preclinical research into relaxin as a treatment for acute myocardial infarction, as relaxin improves ventricular performance and protects cardiomyocytes through reduced damage and apoptosis (Perna et al., 2005). Relaxin causes collagen remodeling in pubic symphisis ligaments; further studies indicated the molecular mechanisms by which relaxin caused collagen remodeling (Wahl et al., 1977). In human dermal fibroblasts, relaxin caused a decrease in collagen secretion, a decrease in tissue inhibitor of metalloproteases, and an increase in procollagenase, M M P - 1 (Unemori and Amento, 1990). The same characteristics are present in a pulmonary fibrosis model, resulting in decreased fibrosis (Unemori et al., 1996). In a renal fibrosis model, relaxin again decreases collagen production, and increases procollagenase (Masterson et al., 2004). Relaxin decreases T I M P - 1 , and increases procollagenase and M M P - 3 , two enzymes which degrade collagen I, the primary collagen component of the human uterine cervix (Palejwala et al., 2001). The large body of evidence that relaxin can remodel the extracellular matrix and reduce 19 fibrosis, led to clinical studies using relaxin to treat systemic scleroderma, a disease of systemic fibrosis (Seibold et al., 2000). While Phase I and II trials were successful in reducing fibrosis, characterized by improving skin elasticity in scleroderma patients, Phase III trials did not support this data, and the treatment development was discontinued (Gavino and Furst, 2001). Regulation of relaxin expression While much is known about the function of relaxin, very little is known about the peptide's regulation. Signal transducer and activator of transcription 3 (STAT3) is bound to the rat relaxin promoter when relaxin levels are low, whereas STAT5a is bound to the relaxin promoter when relaxin levels are elevated during pregnancy, suggesting a role for these transcription factors in regulating relaxin transcription (Soloff et al., 2003). In the rabbit, relaxin transcription is negatively regulated by retinoids (Bernacki et al., 1998), while in a human choricarcinoma cell line, relaxin is positively regulated by glucocorticoid and progesterone (Garibay-Tupas et a l , 2004). In pigs, L H and IGF-I can increase relaxin expression and secretion individually or in combination (Huang et al., 1992; Ohleth and Bagnell, 1999); relaxin is regulated by prolactin and L H in the rat (Peters et al., 2000). However, gonadotropin treatment does not affect relaxin secretion in human seminal fluid (Colon et al., 1994). There is significant species and tissue variation in the expression of this hormone, and the characterization of relaxin transcriptional regulation in the prostate has been limited to reporter assays (Brookes et al., 1998). 20 In prostate In males, relaxin was originally detected in human seminal plasma, and further analysis indicated that relaxin was produced in the prostate but not the testis (Essig et al., 1982b; Ivell et al., 1989; Loumaye et a l , 1980). Relaxin m R N A expression further indicated that this gland produces, and does not just sequester, relaxin (Bogie et al., 1995). The relaxin produced by the prostate is H2 relaxin, the same as is produced in the ovary (Ivell et al., 1989; Winslow et al., 1992). The presence of relaxin in this gland led to investigations aimed at determining the function of relaxin in the prostate. Relaxin can affect the motility of sperm, and has been found to maintain sperm motility over time in aged semen samples (Essig et al., 1982a; Lessing et al., 1986; Lessing et al., 1984; Sarosi et al., 1983). Relaxin can increase sperm motility, and this effect is larger in sperm with compromised motility (Han et al., 2006; Lessing et al., 1986). However, only optimal doses of relaxin are capable of increasing motility as too much or too little relaxin wi l l decrease motility (reviewed in Weiss, 1989). It is possible that relaxin enhances the ability of sperm to penetrate the cervical mucosa (Brenner et al., 1984; Colon et al., 1989). Relaxin can increase the fertilization of oocytes by sperm of suboptimal quantity or motility (Han et al., 2006; Park et al., 1988). Relaxin knockout studies have also suggested a role for relaxin in prostate development in mice; prostate weight is lower in relaxin knockout mice more than three months old than in wi ld type controls (Samuel et al., 2003). Also, these mice had increased collagen levels in the prostate, testes, and epididymis. In all tissues examined, the relaxin deficient mice had a denser extracellular matrix, suggesting a decrease in M M P production (Samuel et a l , 2003). 21 Various functions in cancer Relaxin is present in normal and neoplastic breast tissue (Tashima et al., 1994). In breast cancer cell lines, relaxin increased invasive potential of M C F - 7 and S K - B R 3 , through increased production of M M P - 2 , -7, and -9 (Binder et al., 2002). Previous studies found relaxin to have a biphasic effect on proliferation and differentiation of M C F - 7 cells in vitro and in xenografts in vivo (Bani et al., 1999b; Bigazzi et al., 1992; Sacchi et al., 1994). Relaxin overexpression in a canine mammary cell line, CF33.Mt, increased cell motility, but did not affect mitogenesis (Silvertown et al., 2003). These results indicate that relaxin increases invasive potential without increasing proliferation. Furthermore, relaxin serum levels are elevated in breast cancer patients with metastases, and patients with elevated serum levels of relaxin had decreased overall survival compared to patients with no detectable serum relaxin (Binder et al., 2004). Since serum relaxin is elevated in breast cancer patients with metastases, this hormone may be involved in promoting an invasive phenotype through increased M M P s (Binder et al., 2004). Relaxin has clearly been characterized to be expressed and function in the reproductive tissues (Bani, 1997; Gunnersen et al., 1995; Hal l et al., 1990; Lenhart et al., 2002; Schmidt et al., 1984). A survey of relaxin expression in normal and cancerous gynecological tissues indicated relaxin was expressed in the normal endometrium, hydatiform mole, and choriocarcinoma, but was not detectable in cervical adencarcinoma, ovarian cystadenocarcinomas, or endometrial adenocarcinoma ( Y k i -Jarvinen et al., 1983). The absence of relaxin expression in endometrial cancer is 22 surprising given that relaxin is mitogenic to normal endometrial tissue (Yki-Jarvinen et al., 1983). However, Kamat et al (2006) found relaxin is expressed in 67% of high grade tumours, and 37% of low grade tumours, indicating that the small sample size (n=10) in the original study may have precluded detection of relaxin-expressing endometrial carcinomas (Yki-Jarvinen et al., 1983). Similar to the function of relaxin expression in mammary carcinoma, relaxin appears to increase invasion of endometrial cell lines through increased levels of M M P - 2 and M M P - 9 , and is associated with invasive phenotype of endometrial carcinoma (Kamat et al., 2006). Recently, it was shown that relaxin is produced in thyroid carcinoma cell lines, and overexpression of relaxin in these cell lines causes increased migration, without increasing proliferation, in keeping with results seen in other tissues (Hombach-Klonisch et al., 2006). Interestingly, the increased migratory potential was attributed to a novel target of relaxin, cathepsin-L and cathepsin-D (Hombach-Klonisch et al., 2006), which increases invasiveness in several cancer cell lines (Akiharu Dohchin, 2000; Kirschke et a l , 2000). Turning the field of relaxin in cancer on its head, K i m et al ( K i m et al., 2006) have cloned relaxin into a lytic adenovirus to enhance the lytic properties of the virus. The results of this are twofold: 1) relaxin expression appears to increase apoptosis in 3 cell lines, using this system, and 2) relaxin expression results in decreased collagen content, facilitating increased transmission of the virus throughout tumours. This novel combination treatment abrogates lung metastases of mice with C 3 3 A or A375-mlnl . luc (Ganesh et al., 2007) tumours, and completely inhibits tumour growth by the H e p l hepatocellular carcinoma ( K i m et a l , 2006). This does not indicate that relaxin itself 23 increases apoptosis; rather, relaxin upregulates collagenase activity to increase the lytic properties of this particular virus. 1.3 Hypothesis Relaxin is involved in the activation of key pathways involved in invasion and angiogenesis during the progression of androgen independent prostate cancer, and therefore may be a suitable target for therapy. 1.4 Objectives Objective 1. To characterize androgen regulation of relaxin Preliminary research indicated that relaxin transcription was affected by androgens. Surprisingly, very little is known about the transcriptional regulation of relaxin, especially in the human prostate. The effect of androgens on relaxin levels in prostate cancer is important for those patients who receive androgen withdrawal therapy. The work of Chapter 2 characterizes androgen effects on relaxin transcription, and the relevance of this work to prostate cancer. Objective 2. To identify novel pathways regulated by relaxin in prostate cancer Much work has been done to characterize relaxin, and it is established that relaxin causes an increase in migration and invasiveness in various cancers, although exactly how relaxin affects each cancer differs based on tissue type. c D N A microarray technology is an exciting opportunity to explore novel pathways regulated in prostate cancer by relaxin. One novel pathway regulated by relaxin involves upregulation of 24 W n t l 1 and P C D H 1 1 Y , two genes negatively regulated by androgens, similar to relaxin. Chapter 3 describes the upregulation of the W n t l 1/PCDH11Y pathway by relaxin in prostate cancer. Additional analysis of the c D N A microarray data indicated that relaxin altered IRS-2 levels. IRS-2 is a signaling molecule for the insulin receptor and IGF-IR. IRS-2 was an attractive candidate gene to study for many reasons. 1) Relaxin and IGF-I are both part of the insulin superfamily. 2) Relaxin can cause an increase in IGF-I protein or m R N A , depending on the tissue. 3) The signaling pathway of relaxin itself is not fully characterized. Chapter 4 describes a role for IRS-2 in androgen signaling in IGF-I stimulated prostate cancer cells. These studies identify the dysregualtion of relaxin expression in prostate cancer progression and characterize some of the downstream pathways regulated by relaxin in prostate cancer. 25 1.5 References Akiharu Dohchin, J.-i.S., Hideyuki Seki, Manabu Masutani, Hiroshi Shiroto, Yoshikazu Kawakami,. 2000. Immunostained cathepsins B and L correlate with depth of invasion and different metastatic pathways in early stage gastric carcinoma. Cancer. 89:482-487. Ayyathurai, R., K . Ananfhakrishnan, R. Rajasundaram, R.J . Knight, H . Toussi, and V . Srinivasan. 2006. Predictive ability of partin tables 2001 in a Welsh population. Urol Int. 76:217-22. Bagi , C M . 2005. Targeting of therapeutic agents to bone to treat metastatic cancer. Adv Drug Deliv Rev. 57:995-1010. Bagnell, C . A . , Q. Zhang, B . Downey, and L . Ainsworth. 1993. Sources and biological actions of relaxin in pigs. J Reprod Fertil Suppl. 48:127-38. Bani, D . 1997. Relaxin: a pleiotropic hormone. Gen Pharmacol. 28:13-22. Bani, D. , M . C . Baccari, S. Nistri , F. Calamai, M . Bigazzi , and T .B . Sacchi. 1999a. Relaxin up-regulates the nitric oxide biosynthetic pathway in the mouse uterus: involvement in the inhibition of myometrial contractility. Endocrinology. 140:4434-41. Bani, D . , D . Flagiello, M . F . Poupon, S. Nistri , F. Poirson-Bichat, M . Bigazzi , and T. Bani Sacchi. 1999b. Relaxin promotes differentiation of human breast cancer cells M C F - 7 transplanted into nude mice. Virchows Arch. 435:509-19. Bani, D . , E . Masini , M . G . Bello, M . Bigazzi , and T .B . Sacchi. 1998. Relaxin protects against myocardial injury caused by ischemia and reperfusion in rat heart. Am J ' Pathol. 152:1367-76. Bartsch, O., B . Bartlick, and R. Ivell. 2001. Relaxin signalling links tyrosine phosphorylation to phosphodiesterase and adenylyl cyclase activity. Mol Hum Reprod. 7:799-809. Bathgate, R . A . , R. Ivell, B . M . Sanborn, O.D. Sherwood, and R.J . Summers. 2006. International Union of Pharmacology L V I I : Recommendations for the Nomenclature of Receptors for Relaxin Family Peptides. Pharmacol Rev. 58:7-31. Be l l , R.J . , L . W . Eddie, A . R . Lester, E . C . Wood, P .D. Johnston, and H . D . N ia l l . 1987. Relaxin in human pregnancy serum measured with an homologous radioimmunoassay. Obstet Gynecol. 69:585-9. Berger, R., D.I. L i n , M . Nieto, E . Sicinska, L . A . Garraway, H . Adams, S. Signoretti, W . C . Hahn, and M . Loda. 2006. Androgen-dependent regulation of Her-2/neu in prostate cancer cells. Cancer Res. 66:5723-8. Bernacki, S.H., A . Medvedev, G . Holloway, M . Dawson, R. Lotan, and A . M . Jetten. 1998. Suppression of relaxin gene expression by retinoids in squamous differentiated rabbit tracheal epithelial cells. Mol Cell Endocrinol. 138:115-25. Bigazzi , M . , M . L . Brandi, G . Bani, and T .B. Sacchi. 1992. Relaxin influences the growth of M C F - 7 breast cancer cells. Mitogenic and antimitogenic action depends on peptide concentration. Cancer. 70:639-43. 26 Binder, C , T. Hagemann, B . Husen, M . Schulz, and A . Einspanier. 2002. Relaxin enhances in-vitro invasiveness of breast cancer cell lines by up-regulation of matrix metalloproteases. Mol Hum Reprod. 8:789-96. Binder, C , A . Simon, L . Binder, T. Hagemann, M . Schulz, G . Emons, L . Trumper, and A . Einspanier. 2004. Elevated concentrations of serum relaxin are associated with metastatic disease in breast cancer patients. Breast Cancer Res Treat. 87:157-66. Bogie, L . V . , M . Mandel, and G .D . Bryant-Greenwood. 1995. Relaxin gene expression in human reproductive tissues by in situ hybridization. J Clin Endocrinol Metab. 80:130-7. Braddon, S.A. 1978. Relaxin-dependent adenosine 6',5'-monophosphate concentration ' changes in the mouse pubic symphysis. Endocrinology. 102:1292-9. Brawer, M . K . 2006. Hormonal therapy for prostate cancer. Rev Urol. 8 Suppl 2:S35-47. Brenner, S.H., J .B. Lessing, C. Schoenfeld, R .D . Amelar, L . Dubin, and G . Weiss. 1984. Stimulation of human sperm cervical mucus penetration in vitro by relaxin. Fertil Steril. 42:92-6. Brookes, D .E . , D . Zandvliet, F. Watt, P.J. Russell, and P .L . Mol loy . 1998. Relative activity and specificity of promoters from prostate-expressed genes. Prostate. 35:18-26. Brooks, J.D., M . Weinstein, X . L i n , Y . Sun, S.S. Pin, G.S. Bova, J.I. Epstein, W . B . Isaacs, and W . G . Nelson. 1998. C G island methylation changes near the GSTP1 gene in prostatic intraepithelial neoplasia. Cancer Epidemiol Biomarkers Prev. 7:531-6. Bruchovsky, N . , L . Klotz , J. Crook, S. Malone, C. Ludgate, W.J . Morris , M . E . Gleave, and S.L. Goldenberg. 2006. Final results of the Canadian prospective phase II trial o f intermittent androgen suppression for men in biochemical recurrence after radiotherapy for locally advanced prostate cancer: clinical parameters. Cancer. 107:389-95. Bryant-Greenwood, G.D. , E . M . Rutanen, S. Partanen, T . K . Coelho, and S.Y. Yamamoto. 1993. Sequential appearance of relaxin, prolactin and IGFBP-1 during growth and differentiation of the human endometrium. Mol Cell Endocrinol. 95:23-9. Burazin, T., Davern, PJ, McKin ley , M J , Tregear, G W . 2000. Identification of relaxin and relaxin responsive cells in the rat brain. In Third international conference on relaxin and related peptides. V o l . 1. G . Tregear, Ivel R, Bathgate, R A , Wade, JD, editor, Broome Australia. 209-214. Canada, N.C.I .o. 2007. National Cancer Institute of Canada: Canadian Cancer Statistics 2007, Toronto, Canada. Canova-Davis, E . , T.J. Kessler, P.J. Lee, D.T. Fei, P. Griffin, J.T. Stults, J.D. Wade, and E. Rinderknecht. 1991. Use of recombinant D N A derived human relaxin to probe the structure of the native protein. Biochemistry. 30:6006-13. Cheah, S.H., and O.D. Sherwood. 1980. Target tissues for relaxin in the rat: tissue distribution of injected 1251-labeled relaxin and tissue changes in adenosine 3',5'-monophosphate levels after in vitro relaxin incubation. Endocrinology. 106:1203-9. Ch i , K . N . , M . E . Gleave, R. Klasa, N . Murray, C. Bryce, D . E . Lopes de Menezes, S. D 'Alois io , and A . W . Tolcher. 2001. A phase I dose-finding study of combined 27 treatment with an antisense Bcl-2 oligonucleotide (Genasense) and mitoxantrone in patients with metastatic hormone-refractory prostate cancer. Clin Cancer Res. 7:3920-7. Chokkalingam, A . P . , O. Nyren, J.E. Johansson, G . Gridley, J .K. McLaughl in , H .O . Adami, and A . W . Hsing. 2003. Prostate carcinoma risk subsequent to diagnosis of benign prostatic hyperplasia: a population-based cohort study in Sweden. Cancer. 98:1727-34. Colon, J., N . Santoro, L . Goldsmith, R. Whitcomb, D . Musket, and G . Weiss. 1994. Relaxin secretion into human semen independent of gonadotropin stimulation. Biol Reprod. 50:187-192. Colon, J . M . , C . Gagliardi, C . Schoenfeld, R .D . Amelar, L . Dubin, and G . Weiss. 1989. Human relaxin stimulates human sperm penetration of bovine cervical mucus. Fertil Steril. 52:340-2. Craft, N . , Y . Shostak, M . Carey, and C . L . Sawyers. 1999. A mechanism for hormone-independent prostate cancer through modulation of androgen receptor signaling by the HER-2/neu tyrosine kinase. Nat Med. 5:280-5. Cul ig , Z . , A . Hobisch, M . V . Cronauer, C. Radmayr, J. Trapman, A . Hittmair, G . Bartsch, and H . Klocker. 1994. Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. Cancer Res. 54:5474-8. Danielson, L . A . , O.D. Sherwood, and K . P . Conrad. 1999. Relaxin is a potent renal vasodilator in conscious rats. J Clin Invest. 103:525-33. Dearth, R . K . , X . Cu i , H.J . K i m , D . L . Hadsell, and A . V . Lee. 2007. Oncogenic transformation by the signaling adaptor proteins insulin receptor substrate (IRS)-1 and IRS-2. Cell Cycle. 6:705-13. Debruyne, F . M . 2004. Gonadotropin-releasing hormone antagonist in the management of prostate cancer. Rev Urol. 6 Suppl 7:S25-32. Dessauer, C .W. , and B .T . Nguyen. 2005. Relaxin stimulates multiple signaling pathways: activation of c A M P , PI3K, and PKCzeta in THP-1 cells. Ann N Y Acad Sci. 1041:272-9. Dschietzig, T., C. Bartsch, C. Richter, M . Laule, G . Baumann, and K . Stangl. 2003. Relaxin, a pregnancy hormone, is a functional endothelin-1 antagonist: attenuation of endothelin-1-mediated vasoconstriction by stimulation of endothelin type-B receptor expression via E R K - 1 / 2 and nuclear factor-kappaB. CircRes. 92:32-40. Dubois, M . P . , and J .L. Dacheux. 1978. Relaxin, a male hormone? Immunocytological localization of a related antigen in the boar testis. Cell Tissue Res. 187:201-14. Eddie, L . W . , R.J . Be l l , A . Lester, M . Geier, G . Bennett, P .D. Johnston, and H . D . Nia l l . 1986. Radioimmunoassay of relaxin in pregnancy with an analogue of human relaxin. Lancet. 1:1344-6. Eigenbrot, C , M . Randal, C. Quan, J. Burnier, L . O'Connell, E . Rinderknecht, and A . A . Kossiakoff. 1991. X-ray structure of human relaxin at 1.5 A . Comparison to insulin and implications for receptor binding determinants. J Mol Biol. 221:15-21. Essig, M . , C . Schoenfeld, R . D . Amelar, L . Dubin, and G . Weiss. 1982a. Stimulation of human sperm motility by relaxin. Fertil Steril. 38:339-43. 28 Essig, M . , C . Schoenfeld, R.T. D'Eletto, R. Amelar, B . G . Steinetz, E . M . O'Byrne, and G . Weiss. 1982b. Relaxin in human seminal plasma. Ann N Y Acad Sci. 380:224-30. Farla, P., R. Hersmus, J. Trapman, and A . B . Houtsmuller. 2005. Antiandrogens prevent stable DNA-binding of the androgen receptor. J Cell Sci. 118:4187-98. Fei , D.T. , M . C . Gross, J .L. Lofgren, M . Mora-Worms, and A . B . Chen. 1990. Cycl ic A M P response to recombinant human relaxin by cultured human endometrial cells—a specific and high throughput in vitro bioassay. Biochem Biophys Res Commun. 170:214-22. Feldman, B . J . , and D . Feldman. 2001. The development of androgen-independent prostate cancer. Nat Rev Cancer. 1:34-45. Fevold, H . L . , F . L . Hisaw, and R . K . Meyer. 1930. The relaxative hormone of the corpus luteum. Its purification and concentration. J. Am. Chem. Soc. 52:3340-3348. Furr, B . J . , and H . Tucker. 1996. The preclinical development of bicalutamide: pharmacodynamics and mechanism of action. Urology. 47:13-25; discussion 29-32. Ganesh, S., M . Gonzalez Edick, N . Idamakanti, M . Abramova, M . Vanroey, M . Robinson, C O . Y u n , and K . Jooss. 2007. Relaxin-expressing, fiber chimeric oncolytic adenovirus prolongs survival of tumor-bearing mice. Cancer Res. 67:4399-407. Gao, J., J. Mazella, and L . Tseng. 1995. Activation of human insulin-like growth factor binding protein-1 gene promoter by a distal regulatory sequence in a human endometrial adenocarcinoma cell line. Mol Endocrinol. 9:1405-12. Gao, J .G., J. Mazella, and L . Tseng. 1994. Activation of the human IGFBP-1 gene promoter by progestin and relaxin in primary culture of human endometrial stromal cells. Mol Cell Endocrinol. 104:39-46. Garibay-Tupas, J .L. , K . J . Okazaki, L . S . Tashima, S. Yamamoto, and G . D . Bryant-Greenwood. 2004. Regulation of the human relaxin genes H I and H2 by steroid hormones. Mol Cell Endocrinol. 219:115-25. Gast, M . J . 1982. Studies of luteal generation and processing of the high molecular weight relaxin precursor. Ann N YAcad Sci. 380:111-25. Gavino, E.S. , and D . E . Furst. 2001. Recombinant relaxin: a review of pharmacology and potential therapeutic use. BioDrugs. 15:609-14. Girl ing, J.E., and P .A . Rogers. 2005. Recent advances in endometrial angiogenesis research. Angiogenesis. 8:89-99. Gleason, D.F. , and G.T. Mellinger. 1974. Prediction of prognosis for prostatic adenocarcinoma by combined histological grading and clinical staging. J Urol. 111:58-64. Gleave, M . , A . Tolcher, H . Miyake, C. Nelson, B . Brown, E . Beraldi, and J. Goldie. 1999. Progression to androgen independence is delayed by adjuvant treatment with antisense Bcl-2 oligodeoxynucleotides after castration in the L N C a P prostate tumor model. Clin Cancer Res. 5:2891-8. Goldenberg, S.L., N . Bruchovsky, M . E . Gleave, L . D . Sullivan, and K . Akakura. 1995. Intermittent androgen suppression in the treatment of prostate cancer: a preliminary report. Urology. 45:839-44; discussion 844-5. Goldsmith, L . T . , and G . Weiss. 2005. Relaxin regulates endometrial structure and function in the rhesus monkey. Ann N YAcad Sci. 1041:110-7. 29 Grignon, D.J . , R. Caplan, F . H . Sarkar, C . A . Lawton, E . H . Hammond, M . V . Pilepich, J.D. Forman, J. Mesic, K . K . Fu, R . A . Abrams, T.F. Pajak, W . U . Shipley, and J.D. Cox. 1997. p53 status and prognosis of locally advanced prostatic adenocarcinoma: a study based on R T O G 8610. J Natl Cancer Inst. 89:158-65. Gudmundsson, J., P. Sulem, A . Manolescu, L . T . Amundadottir, D . Gudbjartsson, A . Helgason, T. Rafnar, J.T. Bergthorsson, B . A . Agnarsson, A . Baker, A . Sigurdsson, K . R . Benediktsdottir, M . Jakobsdottir, J. X u , T. Blondal, J. Kostic, J. Sun, S. Ghosh, S.N. Stacey, M . Mouy, J. Saemundsdottir, V . M . Backman, K . Kristjansson, A . Tres, A . W . Partin, M . T . Albers-Akkers, J. Godino-Ivan Marcos, P .C. Walsh, D . W . Swinkels, S. Navarrete, S.D. Isaacs, K . K . Aben, T. Graif, J. Cashy, M . Ruiz-Echarri, K . E . Wiley, B . K . Suarez, J .A. Witjes, M . Frigge, C. Ober, E . Jonsson, G . V . Einarsson, J.I. Mayordomo, L . A . Kiemeney, W . B . Isaacs, W.J . Catalona, R . B . Barkardottir, J.R. Gulcher, U . Thorsteinsdottir, A . Kong, and K . Stefansson. 2007. Genome-wide association study identifies a second prostate cancer susceptibility variant at 8q24. Nat Genet. 39:631-7. Guico-Lamm, M . L . , and O.D. Sherwood. 1988. Monoclonal antibodies specific for rat relaxin. II. Passive immunization with monoclonal antibodies throughout the second half of pregnancy disrupts birth in intact rats. Endocrinology. 123:2479-85. Gunnersen, J . M . , R.J . Crawford, and G .W. Tregear. 1995. Expression of the relaxin gene in rat tissues. Mol Cell Endocrinol. 110:55-64. Gupta, S. 2006. Prostate cancer chemoprevention: Current status and future prospects. Toxicol Appl Pharmacol. Haiman, C . A . , N . Patterson, M . L . Freedman, S.R. Myers, M . C . Pike, A . Waliszewska, J. Neubauer, A . Tandon, C. Schirmer, G.J. McDonald, S.C. Greenway, D .O. Stram, L . Le Marchand, L . N . Kolonel , M . Frasco, D . Wong, L . C . Pooler, K . Ardlie, I. Oakley-Girvan, A . S . Whittemore, K . A . Cooney, E . M . John, S.A. Ingles, D . Altshuler, B . E . Henderson, and D . Reich. 2007. Multiple regions within 8q24 independently affect risk for prostate cancer. Nat Genet. 39:638-44. Hal l , J .A. , T .C. Cantley, B . N . Day, and R . V . Anthony. 1990. Uterotropic actions of relaxin in prepubertal gilts. Biol Reprod. 42:769-74. Hal l , J .A. , T .C. Cantley, J . M . Galvin, B . N . Day, and R . V . Anthony. 1992. Influence of ovarian steroids on relaxin-induced uterine growth in ovariectomized gilts. Endocrinology. 130:3159-66. Halls, M . L . , R . A . Bathgate, and R.J . Summers. 2005. Signal switching after stimulation of L G R 7 receptors by human relaxin 2. Ann N YAcad Sci. 1041:288-91. Halls, M . L . , R . A . D . Bathgate, and R.J . Summers. 2006. Relaxin Family Peptide Receptors R X F P 1 and R X F P 2 Modulate c A M P Signaling by Distinct Mechanisms. Mol Pharmacol. 70:214-226. Han, Y . J . , A . G . Miah , M . Yoshida, H . Sasada, K . Hamano, T. Kohsaka, and H . Tsujii. 2006. Effect of relaxin on in vitro fertilization of porcine oocytes. J Reprod Dev. 52:657-62. Hisaw, F . L . 1926. Experimental relaxation of the pubic ligament of the guinea pig. Proc Soc Exp Biol Med. 23:661 -663. Hoedemaeker, R.F. , A . N . V i s , and T . H . Van Der Kwast. 2000. Staging prostate cancer. Microsc Res Tech. 51:423-9. 30 Hombach-Klonisch, S., J. Bialek, B . Trojanowicz, E . Weber, H. -J . Holzhausen, J.D. Silvertown, A . J . Summerlee, H . Dralle, C. Hoang-Vu, and T. Klonisch. 2006. Relaxin Enhances the Oncogenic Potential of Human Thyroid Carcinoma Cells. Am J Pathol. 169:617-632. Hornsby, D.J . , B . C . Wilson, and A . J . Summerlee. 2001. Relaxin and drinking in pregnant rats. Prog Brain Res. 133:229-40. Hsu, S.Y., M . Kudo, T. Chen, K . Nakabayashi, A . Bhalla, P.J. van der Spek, M . van Duin, and A . J . Hsueh. 2000. The three subfamilies of leucine-rich repeat-containing G protein-coupled receptors (LGR) : identification of L G R 6 and L G R 7 and the signaling mechanism for L G R 7 . Mol Endocrinol. 14:1257-71. Hsu, S.Y., K . Nakabayashi, S. Nishi , J. Kumagai, M . Kudo, O.D. Sherwood, and A . J . Hsueh. 2002. Activation of orphan receptors by the hormone relaxin. Science. 295:671-4. Huang, C , Y . L i , and L . L . Anderson. 1993. Stimulation of collagen secretion by relaxin and effect of oestrogen on relaxin binding in uterine cervical cells of pigs. J Reprod Fertil. 98:153-8. Huang, C.J. , Y . L i , M . H . Stromer, and L . L . Anderson. 1992. Synergistic effects of insulin-like growth factor I and gonadotrophins on relaxin and progesterone secretion by ageing corpora lutea of pigs. J Reprod Fertil. 96:415-25. Hudson, P., J. Haley, M . John, M . Cronk, R. Crawford, J. Haralambidis, G . Tregear, J. Shine, and H . Nia l l . 1983. Structure of a genomic clone encoding biologically active human relaxin. Nature. 301:628-31. Hudson, P., M . John, R. Crawford, J. Haralambidis, D . Scanlon, J. Gorman, G . Tregear, J. Shine, and H . Nia l l . 1984. Relaxin gene expression in human ovaries and the predicted structure of a human preprorelaxin by analysis of c D N A clones. Embo J. 3:2333-9. Huggins, C , and C V . Hodges. 1941. Studies on prostatic cancer. I. The effect of castration, estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Research. 1:293-297. Hughes, S.J., and M . Hollingsworth. 1997. Relaxin as a relaxant of the isolated rat uterus: comparison with its mechanism of action in vivo. Gen Pharmacol. 29:829-33. Humphrey, P . A . 2004. Gleason grading and prognostic factors in carcinoma of the prostate. Mod Pathol. 17:292-306. Hwang, J.J., A . B . Lee, P . A . Fields, L . M . Haab, L . E . Mojonnier, and O.D. Sherwood. 1991. Monoclonal antibodies specific for rat relaxin. V . Passive immunization with monoclonal antibodies throughout the second half of pregnancy disrupts development of the mammary apparatus and, hence, lactational performance in rats. Endocrinology. 129:3034-42. Hwang, J.J., D . Macinga, and E . A . Rorke. 1996. Relaxin modulates human cervical stromal cell activity. J Clin Endocrinol Metab. 81:3379-84. Hwang, J.J., R . D . Shanks, and O.D. Sherwood. 1989. Monoclonal antibodies specific for rat relaxin. IV . Passive immunization with monoclonal antibodies during the antepartum period reduces cervical growth and extensibility, disrupts birth, and reduces pup survival in intact rats. Endocrinology. 125:260-6. 31 Hwang, J.J., and O.D. Sherwood. 1988. Monoclonal antibodies specific for rat relaxin. III. Passive immunization with monoclonal antibodies throughout the second half of pregnancy reduces cervical growth and extensibility in intact rats. Endocrinology. 123:2486-90. Isaacs, J.T. 1999. The biology of hormone refractory prostate cancer. Why does it develop? Urol Clin North Am. 26:263-73. Ivell, R., and A . Einspanier. 2002. Relaxin peptides are new global players. Trends Endocrinol Metab. 13:343-8. Ivell, R., N . Hunt, F. Khan-Dawood, and M . Y . Dawood. 1989. Expression of the human relaxin gene in the corpus luteum of the menstrual cycle and in the prostate. Mol Cell Endocrinol. 66:251-5. Jin-Tang Dong. 2006. Prevalent mutations in prostate cancer. Journal of Cellular Biochemistry. 97:433-447. Kamat, A . A . , S. Feng, I .U. Agoulnik, F. Kheradmand, N . V . Bogatcheva, D . Coffey, A . K . Sood, and A . I . Agoulnik. 2006. The role of relaxin in endometrial cancer. Cancer Biol Ther. 5:71-7. Kass, L . , J .G. Ramos, H . H . Ortega, G.S. Montes, L . E . Bussmann, E . H . Luque, and M . Munoz de Toro. 2001. Relaxin has a minor role in rat mammary gland growth and differentiation during pregnancy. Endocrine. 15:263-9. Key, T. 1995. Risk factors for prostate cancer. Cancer Surv. 23:63-77. K i m , J .H. , Y . S . Lee, H . K i m , J .H. Huang, A . R . Yoon, and C O . Y u n . 2006. Relaxin expression from tumor-targeting adenoviruses and its intratumoral spread, apoptosis induction, and efficacy. JNatl Cancer Inst. 98:1482-93. Kirschke, H . , R. Eerola, V . K . Hopsu-Havu, D . Bromme, and E . Vuorio. 2000. Antisense R N A inhibition of cathepsin L expression reduces tumorigenicity of malignant cells. Eur J Cancer. 36:787-95. Kle in , E . A . 1995. A n update on prostate cancer. Cleve Clin J Med. 62:325-38. Koay, E.S., C . A . Bagnell, G .D . Bryant-Greenwood, S.B. Lord, A . C Cruz, and L . H . Larkin. 1985. Immunocytochemical localization of relaxin in human decidua and placenta. J Clin Endocrinol Metab. 60:859-63. Koivisto, P., J. Kononen, C. Palmberg, T. Tammela, E . Hyytinen, J. Isola, J. Trapman, K . Cleutjens, A . Noordzij, T. Visakorpi, and O.P. Kall ioniemi. 1997. Androgen receptor gene amplification: a possible molecular mechanism for androgen deprivation therapy failure in prostate cancer. Cancer Res. 57:314-9. Krajnc-Franken, M . A . M . , A . J . M . van Disseldorp, J.E. Koenders, S. Mosselman, M . van Duin, and J .A. Gossen. 2004. Impaired Nipple Development and Parturition in L G R 7 Knockout Mice . Mol. Cell. Biol. 24:687-696. Kramer, S . M . , U . E . Gibson, B . M . Fendly, M . A . Mohler, D . W . Drolet, and P .D. Johnston. 1990. Increase in cyclic A M P levels by relaxin in newborn rhesus monkey uterus cell culture. In Vitro Cell Dev Biol. 26:647-56. Krantz, J . C , Jr., H . H . Bryant, and C J . Carr. 1950. The action of aqueous corpus luteum extract upon uterine activity. Surg Gynecol Obstet. 90:372-5. Kuenzi , M . J . , and O.D. Sherwood. 1992. Monoclonal antibodies specific for rat relaxin. VII . Passive immunization with monoclonal antibodies throughout the second half of pregnancy prevents development of normal mammary nipple morphology and function in rats. Endocrinology. 131:1841-7. 32 Kumagai, J., S.Y. Hsu, H . Matsumi, J.-S. Roh, P. Fu, J.D. Wade, R . A . D . Bathgate, and A . J . W . Hsueh. 2002. INSL3/Leydig Insulin-like Peptide Activates the L G R 8 Receptor Important in Testis Descent. J. Biol. Chem. 277:31283-31286. Lee, A . B . , J.J. Hwang, L . M . Haab, P . A . Fields, and O.D. Sherwood. 1992. Monoclonal antibodies specific for rat relaxin. V I . Passive immunization with monoclonal antibodies throughout the second half of pregnancy disrupts histological changes associated with cervical softening at parturition in rats. Endocrinology. 130:2386-91. Lee, W . H . , R . A . Morton, J.I. Epstein, J.D. Brooks, P . A . Campbell, G.S. Bova, W.S. Hsieh, W . B . Isaacs, and W . G . Nelson. 1994. Cytidine methylation of regulatory sequences near the pi-class glutathione S-transferase gene accompanies human prostatic carcinogenesis. Proc Natl Acad Sci U S A. 91:11733-7. Lenhart, J .A., P .L . Ryan, K . M . Ohleth, S.S. Palmer, and C A . Bagnell. 2002. Relaxin increases secretion of tissue inhibitor of matrix metalloproteinase-1 and -2 during uterine and cervical growth and remodeling in the pig. Endocrinology. 143:91-8. Lessing, J.B., S.H. Brenner, J . M . Colon, F .W. Ginsburg, C. Schoenfeld, L . T . Goldsmith, P. Sarosi, R . D . Amelar, L . Dubin, and G . Weiss. 1986. Effect of relaxin on human spermatozoa. J Reprod Med. 31:304-9. Lessing, J.B., S.H. Brenner, C. Schoenfeld, P. Sarosi, R. Amelar, L . Dubin, and G . Weiss. 1984. The effect of an anti-insulin antiserum on human sperm motility. Fertil Steril. 42:309-11. L i , J., C . Yen , D . Liaw, K . Podsypanina, S. Bose, S.I. Wang, J. Puc, C. Miliaresis, L . Rodgers, R. McCombie , S.H. Bigner, B . C . Giovanella, M . Ittmann, B . Tycko, H . Hibshoosh, M . H . Wigler, and R. Parsons. 1997. P T E N , a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science. 275:1943-7. Loumaye, E . , S. De Cooman, and K . Thomas. 1980. Immunoreactive relaxin-like substance in human seminal plasma. J Clin Endocrinol Metab. 50:1142-3. MacLennan, A . H . , R. Nicolson, and R . C . Green. 1986a. Serum relaxin in pregnancy. Lancet. 2:241-3. MacLennan, A . H . , R. Nicolson, R . C Green, and M . Bath. 1986b. Serum relaxin and pelvic pain of pregnancy. Lancet. 2:243-5. Masini , E . , D . Bani, M . G . Bello, M . Bigazzi , P.F. Mannaioni, and T .B . Sacchi. 1997. Relaxin counteracts myocardial damage induced by ischemia-reperfusion in isolated guinea pig hearts: evidence for an involvement of nitric oxide. Endocrinology. 138:4713-20. Masini , E . , D . Bani , M . Bigazzi , P.F. Mannaioni, and T. Bani-Sacchi. 1994. Effects of relaxin on mast cells. In vitro and in vivo studies in rats and guinea pigs. J Clin Invest. 94:1974-80. Masini , E . , M . G . D i Bello, D . Bani, M . Bigazzi , T. Bani Sacchi, and P.F. Mannaioni. 1995. Relaxin inhibits histamine release from mast cells: involvement of nitric oxide production. Inflamm Res. 44 Suppl l:S12-3. Masini , E . , G . Zagli , J.F. Ndisang, M . Solazzo, P.F. Mannaioni, and D . Bani . 2002. Protective effect of relaxin in cardiac anaphylaxis: involvement of the nitric oxide pathway. Br J Pharmacol. 137:337-44. 33 Masterson, R., T .D. Hewitson, K . Kelynack, M . Martic, L . Parry, R. Bathgate, I. Darby, and G . Becker. 2004. Relaxin down-regulates renal fibroblast function and promotes matrix remodelling in vitro. Nephrol. Dial. Transplant. 19:544-552. Mi l la r , D.S. , K . K . Ow, C . L . Paul, P.J. Russell, P .L . Mol loy , and S.J. Clark. 1999. Detailed methylation analysis of the glutathione S-transferase pi (GSTP1) gene in prostate cancer. Oncogene. 18:1313-24. M i n , G . , M . G . Hartzog, R . L . Jennings, R.J . Winn, and O.D. Sherwood. 1997. Evidence that endogenous relaxin promotes growth of the vagina and uterus during pregnancy in gilts. Endocrinology. 138:560-5. Moore, X . - l . , S.-l. Tan, C.-y. L o , L . Fang, Y . - D . Su, X . - M . Gao, E . A . Woodcock, R.J . Summers, G .W. Tregear, R . A . D . Bathgate, and X . - J . Du. 2007. Relaxin Antagonizes Hypertrophy and Apoptosis in Neonatal Rat Cardiomyocytes. Endocrinology. 148:1582-1589. Morton, R . A . , Jr. 1994. Racial differences in adenocarcinoma of the prostate in North American men. Urology. 44:637-45. Ndisang, J.F., R. Baronti, G . Cecere, E . Masini , D . Bani, and P.F. Mannaioni. 2001. Relaxin generates nitric oxide and provides protection against cardiac anaphylaxis. Inflamm Res. 50 Suppl 2:S122-3. Nguyen, B.T. , and C . W . Dessauer. 2005. Relaxin stimulates protein kinase C zeta translocation: requirement for cyclic adenosine 3',5'-monophosphate production. Mol Endocrinol. 19:1012-23. Ohleth, K . M . , and C A . Bagnell. 1999. Relaxin secretion and gene expression in porcine granulosa and theca cells are stimulated during in vitro luteinization. Biol Reprod. 60:499-507. Ohleth, K . M . , J .A. Lenhart, P .L . Ryan, S.V. Radecki, and C A . Bagnell. 1997. Relaxin increases insulin-like growth factors (IGFs) and IGF-binding proteins of the pig uterus in vivo. Endocrinology. 138:3652-8. Osheroff, P .L . , M . J . Cronin, and J.A. Lofgren. 1992. Relaxin binding in the rat heart atrium. Proc Natl Acad Sci USA. 89:2384-8. Osheroff, P .L . , and W . H . Ho. 1993. Expression of relaxin m R N A and relaxin receptors in postnatal and adult rat brains and hearts. Localization and developmental patterns. J Biol Chem. 268:15193-9. Osheroff, P .L . , V . T . Ling , R . L . Vandlen, M . J . Cronin, and J .A. Lofgren. 1990. Preparation of biologically active 32P-labeled human relaxin. Displaceable binding to rat uterus, cervix, and brain. J Biol Chem. 265:9396-401. Osheroff, P .L . , and H.S. Phillips. 1991. Autoradiographic localization of relaxin binding sites in rat brain. Proc Natl Acad Sci USA. 88:6413-7. Palejwala, S., D . E . Stein, G . Weiss, B .P . Monia , D . Tortoriello, and L . T . Goldsmith. 2001. Relaxin positively regulates matrix metalloproteinase expression in human lower uterine segment fibroblasts using a tyrosine kinase signaling pathway. Endocrinology. 142:3405-13. Palejwala, S., L . Tseng, A . Wojtczuk, G . Weiss, and L . T . Goldsmith. 2002. Relaxin gene and protein expression and its regulation of procollagenase and vascular endothelial growth factor in human endometrial cells. Biol Reprod. 66:1743-8. 34 Pardo, R., L . H . Larkin, and P . A . Fields. 1980. Immunocytochemical localization of relaxin in endometrial glands of the pregnant guinea pig. Endocrinology. 107:2110-2. Pardo, R.J . , and L . H . Larkin. 1982. Localization of relaxin in endometrial gland cells of pregnant, lactating, and ovariectomized hormone-treated guinea pigs. Am J Anat. 164:79-90. Pardo, R.J . , L . H . Larkin, and R . H . Renegar. 1984. Immunoelectron microscopic localization of relaxin in endometrial gland cells of the pregnant guinea pig. Anat Rec. 209:373-9. Park, J . M . , K . Ewing, F. Mi l le r , C.I. Friedman, and M . H . K i m . 1988. Effects of relaxin on the fertilization capacity of human spermatozoa. Am J Obstet Gynecol. 158:974-9. Parsell, D . A . , J .Y. Mak, E.P. Amento, and E . N . Unemori. 1996. Relaxin binds to and elicits a response from cells of the human monocytic cell line, T H P - 1 . J Biol Chem. 271:27936-41. Partin, A . W . , M . W . Kattan, E . N . Subong, P.C. Walsh, K . J . Wojno, J.E. Oesterling, P.T. Scardino, and J.D. Pearson. 1997. Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update. Jama. 277:1445-51. Partin, A . W . , L . A . Mangold, D . M . Lamm, P.C. Walsh, J.I. Epstein, and J.D. Pearson. 2001. Contemporary update of prostate cancer staging nomograms (Partin Tables) for the new millennium. Urology. 58:843-8. Perna, A . - M . , E . Masini , S. Nistri , V . Briganti, L . Chiappini, P. Stefano, M . Bigazzi , C . Pieroni, T. Bani Sacchi, and D . Bani. 2005. Novel drug development opportunity for relaxin in acute myocardial infarction: evidences from a swine model. FASEB /.:04-3664fje. Peters, C . A . , E .T. Maizels, M . C . Robertson, R.P. Shiu, M . S . Soloff, and M . Hunzicker-Dunn. 2000. Induction of relaxin messenger R N A expression in response to prolactin receptor activation requires protein kinase C delta signaling. Mol Endocrinol. 14:576-90. Platz, E . A . 2007. Genetic variation at 8q24 as a susceptibility factor for prostate cancer: definitive results from epidemiologic studies? Cancer Res. 67:2905-7. Ryan, P .L . , D . L . Baum, J .A. Lenhart, K . M . Ohleth, and C . A . Bagnell. 2001. Expression of uterine and cervical epithelial cadherin during relaxin-induced growth in pigs. Reproduction. 122:929-37. Sacchi, T .B . , D . Bani, M . L . Brandi, A . Falchetti, and M . Bigazzi . 1994. Relaxin influences growth, differentiation and cell-cell adhesion of human breast-cancer cells in culture. Int J Cancer. 57:129-34. Samuel, C.S. , H . Tian, L . Zhao, and E.P. Amento. 2003. Relaxin is a key mediator of prostate growth and male reproductive tract development. Lab Invest. 83:1055-67. Sarosi, P., C. Schoenfeld, J. Berman, R. Basch, G . Randolph, R. Amelar, L . Dubin, B . G . Steinetz, and G . Weiss. 1983. Effect of anti-relaxin antiserum on sperm motility in vitro. Endocrinology. 112:1860-1. 35 Schmidt, C . L . , P. Sarosi, B . G . Steinetz, E . M . O'Byrne, J.E. Tyson, K . Horvath, M . Sas, and G . Weiss. 1984. Relaxin in human decidua and term placenta. Eur J Obstet Gynecol Reprod Biol. 17:171-82. Schumacher, F.R. , H.S. Feigelson, D . G . Cox, C A . Haiman, D . Albanes, J. Buring, E . E . Calle, S.J. Chanock, G . A . Colditz, W.R. Diver, A . M . Dunning, M . L . Freedman, J . M . Gaziano, E . Giovannucci, S.E. Hankinson, R . B . Hayes, B . E . Henderson, R . N . Hoover, R. Kaaks, T. Key, L . N . Kolonel , P. Kraft, L . Le Marchand, J. M a , M . C Pike, E . Ribol i , M . J . Stampfer, D .O. Stram, G . Thomas, M . J . Thun, R. Travis, J. Virtamo, G . Andriole, E . Gelmann, W . C Willett, and D.J . Hunter. 2007. A common 8q24 variant in prostate and breast cancer from a large nested case-control study. Cancer Res. 67:2951-6. Schwabe, C , and E .E . Bullesbach. 1990. Relaxin. Comp Biochem Physiol B. 96:15-21. Schwabe, C , and J .K. McDonald. 1977. Relaxin: a disulfide homolog of insulin. Science. 197:914-5. Scott, D.J . , S. Layfield, Y . Yan, S. Sudo, A . J . Hsueh, G .W. Tregear, and R . A . Bathgate. 2006. Characterization of novel splice variants of L G R 7 and L G R 8 reveals that receptor signaling is mediated by their unique low density lipoprotein class A modules. J Biol Chem. 281:34942-54. Seibold, J.R., J .H. Korn, R. Simms, P.J. Clements, L . W . Moreland, M . D . Mayes, D .E . Furst, N . Rothfield, V . Steen, M . Weisman, D . Collier, F . M . Wigley, P . A . Merkel , M . E . Csuka, V . Hsu, S. Rocco, M . Erikson, J. Hannigan, W.S. Harkonen, and M . E . Sanders. 2000. Recombinant human relaxin in the treatment of scleroderma. A randomized, double-blind, placebo-controlled trial. Ann Intern Med. 132:871-9. Sherwood, O.D., L . M . Olson, S. Zhao, and H.R. Little. 2000. Inhibition of nitric oxide synthase activity diminishes the acute effects of relaxin on growth, but not softening, of the cervix in the rat. Endocrinology. 141:2458-64. Silvertown, J.D., B . J . Geddes, and A . J . Summerlee. 2003. Adenovirus-mediated expression of human prorelaxin promotes the invasive potential of canine mammary cancer cells. Endocrinology. 144:3683-91. Sokol, R .Z . , X . S . Wang, J. Lechago, P .D. Johnston, and R.S. Swerdloff. 1989. Immunohistochemical localization of relaxin in human prostate. J Histochem Cytochem. 37:1253-5. Soloff, M . S . , S. Gal , S. Hoare, C A . Peters, M . Hunzicker-Dunn, G . D . Anderson, and T .G . Wood. 2003. Cloning, characterization, and expression of the rat relaxin gene. Gene. 323:149-55. Soloff, M . S . , A . R . Shaw, L . E . Gentry, H . Marquardt, and P. Vasilenko. 1992. Demonstration of relaxin precursors in pregnant rat ovaries with antisera against bacterially expressed rat prorelaxin. Endocrinology. 130:1844-51. Steck, P .A . , M . A . Pershouse, S.A. Jasser, W . K . Yung, H . L i n , A . H . Ligon, L . A . Langford, M . L . Baumgard, T. Hattier, T. Davis, C. Frye, R. Hu , B . Swedlund, D . H . Teng, and S.V. Tavtigian. 1997. Identification of a candidate tumour suppressor gene, M M A C 1 , at chromosome 10q23.3 that is mutated in multiple advanced cancers. Nat Genet. 15:356-62. Stults, J.T., J .H. Bourell , E . Canova-Davis, V . T . L ing , G.R. Laramee, J.W. Winslow, P.R. Griffin, E . Rinderknecht, and R . L . Vandlen. 1990. Structural 36 characterization by mass spectrometry o f native and recombinant human re laxin . Biomed Environ Mass Spectrom. 19:655-64. Sudo, S., J . K u m a g a i , S. N i s h i , S. L a y f i e l d , T . Ferraro, R . A . Bathgate, and A . J . Hsueh . 2003. H 3 re lax in is a specific l igand for L G R 7 and activates the receptor by interacting wi th both the ectodomain and the exoloop 2. J Biol Chem. 278:7855-62. Sun, X . , H . F . Fr ie rson , C . Chen , C . L i , Q. R a n , K . B . Otto, B . L . Cantarel , R . L . Vesse l l a , A . C . G a o , J . Petros, Y . M i u r a , J . W . S imons , and J .T. D o n g . 2005. Frequent somatic mutations o f the transcription factor A T B F 1 i n human prostate cancer. Nat Genet. 37:407-12. Sunn, N . , M . E g l i , T . C . B u r a z i n , P . Burns , L . C o l v i l l , P . Dave rn , D . A . Den ton , B . J . O l d f i e l d , R . S . Weis inger , M . Rauch , H . A . S c h m i d , and M . J . M c K i n l e y . 2002. C i r c u l a t i ng re lax in acts on subfornical organ neurons to stimulate water d r ink ing in the rat. Proc Natl Acad Sci U S A. 99:1701-6. Sutton, S . W . , P . Bonaventure, C . K u e i , D . Nepomuceno , J . W u , J . Z h u , T . W . Lovenbe rg , and C . L i u . 2005. G-prote in-coupled receptor ( G P C R ) - 1 4 2 does not contribute to relaxin-3 b ind ing i n the mouse brain: further support that relaxin-3 is the phys io log i ca l l igand for G P C R 1 3 5 . Neuroendocrinology. 82:139-50. Sutton, S . W . , P . Bonaventure , C . K u e i , D . Nepomuceno , J . W u , J . Z h u , T . W . Lovenberg , and C . L i u . 2006. G-Pro te in -Coup led Receptor ( G P C R ) - 1 4 2 Does N o t Contr ibute to Re lax in -3 B i n d i n g i n the M o u s e B r a i n : Further Support that Re lax in -3 Is the Phys io log i ca l L i g a n d for G P C R 1 3 5 . Neuroendocrinology. 82:139-150. Suur in iemi , M . , I. A g a l l i u , D . J . Schaid , B . Johanneson, S . K . M c D o n n e l l , L . Iwasaki , J / L . Stanford, and E . A . Ostrander. 2007. Conf i rmat ion o f a Pos i t ive Assoc i a t i on between Prostate Cancer R i s k and a L o c u s at Ch r om os om e 8q24. Cancer Epidemiol Biomarkers Prev. 16:809-814. T a m m e l a , T . 2004. Endocr ine treatment o f prostate cancer. J Steroid Biochem Mol Biol. 92:287-95. Tang , M . , J . M a z e l l a , H . H . Z h u , and L . Tseng. 2005. L i g a n d activated re lax in receptor increases the transcription o f I G F B P - 1 and prolact in i n human decidual and endometr ial s tromal cel ls . Mol Hum Reprod. 11:237-43. Tash ima , L . S . , G . M a z o u j i a n , and G . D . Bryan t -Greenwood . 1994. H u m a n relaxins i n normal , ben ign and neoplastic breast tissue. J Mol Endocrinol. 12:351-64. U n e m o r i , E . N . , and E . P . A m e n t o . 1990. R e l a x i n modulates synthesis and secretion o f procollagenase and col lagen by human dermal fibroblasts. J Biol Chem. 265:10681-5. U n e m o r i , E . N . , M . E . E r i k s o n , S .E . R o c c o , K . M . Sutherland, D . A . Parse l l , J . M a k , and B . H . G r o v e . 1999. R e l a x i n stimulates expression o f vascular endothelial growth factor i n normal human endometrial cells i n vi t ro and is associated w i t h menometrorrhagia i n women . Hum Reprod. 14:800-6. U n e m o r i , E . N . , L . B . P i ck fo rd , A . L . Salles, C . E . P ie rcy , B . H . G r o v e , M . E . E r i k s o n , and E . P . A m e n t o . 1996. R e l a x i n induces an extracellular matr ix-degrading phenotype i n human lung fibroblasts i n vi t ro and inhibits lung fibrosis i n a murine mode l i n v i v o . J Clin Invest. 98:2739-45. 37 Varenhorst, E . , L . Wallentin, and K . Carlstrom. 1982. The effects of orchidectomy, estrogens, and cyproterone acetate on plasma testosterone, L H , and F S H concentrations in patients with carcinoma of the prostate. Scand J Urol Nephrol. 16:31-6. Veldscholte, J., C . A . Berrevoets, C. Ris-Stalpers, G . G . Kuiper, G . Jenster, J. Trapman, A . O . Brinkmann, and E . Mulder. 1992. The androgen receptor in L N C a P cells contains a mutation in the ligand binding domain which affects steroid binding characteristics and response to antiandrogens. J Steroid Biochem Mol Biol. 41:665-9. Visakorpi, T., E . Hyytinen, P. Koivisto, M . Tanner, R. Keinanen, C. Palmberg, A . Palotie, T. Tammela, J. Isola, and O.P. Kall ioniemi. 1995. In vivo amplification of the androgen receptor gene and progression of human prostate cancer. Nat Genet. 9:401-6. Wahl , L . M . , R.J . Blandau, and R . C . Page. 1977. Effect of hormones on collagen metabolism and collagenase activity in the pubic symphysis ligament of the guinea pig. Endocrinology. 100:571-9. Wang, L . , S.K. McDonnel l , J.P. Slusser, S.J. Hebbring, J . M . Cunningham, S.J. Jacobsen, J.R. Cerhan, M . L . Blute, D.J . Schaid, and S.N. Thibodeau. 2007. Two common chromosome 8q24 variants are associated with increased risk for prostate cancer. Cancer Res. 67:2944-50. Weiss, G . 1989. Relaxin in the male. Biol Reprod. 40:197-200. Winslow, J.W., A . Shih, J .H. Bourell, G . Weiss, B . Reed, J.T. Stults, and L . T . Goldsmith. 1992. Human seminal relaxin is a product of the same gene as human luteal relaxin. Endocrinology. 130:2660-8. Witte, J.S. 2007. Multiple prostate cancer risk variants on 8q24. Nat Genet. 39:579-80. Yamazawa, K . , Y . Wada, I. Sasagawa, K . A o k i , K . Ueoka, and T. Ogata. 2007. Mutation and polymorphism analyses of INSL3 and L G R 8 / G R E A T in 62 Japanese patients with cryptorchidism. Horm Res. 67:73-6. Yeager, M . , N . Orr, R . B . Hayes, K . B . Jacobs, P. Kraft, S. Wacholder, M . J . Minichiel lo, P. Fearnhead, K . Y u , N . Chatterjee, Z . Wang, R. Welch, B . J . Staats, E . E . Calle, H.S. Feigelson, M . J . Thun, C. Rodriguez, D . Albanes, J. Virtamo, S. Weinstein, F.R. Schumacher, E . Giovannucci, W . C . Willett, G . Cancel-Tassin, O. Cussenot, A . Valeri , G . L . Andriole, E.P. Gelmann, M . Tucker, D.S. Gerhard, J.F. Fraumeni, Jr., R. Hoover, D.J . Hunter, S.J. Chanock, and G . Thomas. 2007. Genome-wide association study of prostate cancer identifies a second risk locus at 8q24. Nat Genet. 39:645-649. Yki-Jarvinen, H . , T. Wahlstrom, and M . Seppala. 1983. Immunohistochemical demonstration of relaxin in the genital tract of men. J Reprod Fertil. 69:693-5. Yoshimoto, M . , J.C. Cutz, P .A . Nuin, A . M . Joshua, J. Bayani, A . J . Evans, M . Zielenska, and J .A. Squire. 2006. Interphase F ISH analysis of P T E N in histologic sections shows genomic deletions in 68% of primary prostate cancer and 23% of high-grade prostatic intra-epithelial neoplasias. Cancer Genet Cytogenet. 169:128-37. Zhang, L . , M . Johnson, K . H . Le, M . Sato, R. Ilagan, M . Iyer, S.S. Gambhir, L . Wu , and M . Carey. 2003. Interrogating androgen receptor function in recurrent prostate cancer. Cancer Res. 63:4552-60. 38 Zhang, Q., S.H. L i u , M . Erikson, M . Lewis, and E . Unemori. 2002. Relaxin activates the M A P kinase pathway in human endometrial stromal cells. J Cell Biochem. 85:536-44. Zhao, L . , P.J. Roche, J . M . Gunnersen, V . E . Hammond, G . W . Tregear, E . M . Wintour, and F. Beck. 1999. Mice without a functional relaxin gene are unable to deliver milk to their pups. Endocrinology. 140:445-53. Zhao, S., C . H . Malmgren, R . D . Shanks, and O.D. Sherwood. 1995. Monoclonal antibodies specific for rat relaxin. VIII. Passive immunization with monoclonal antibodies throughout the second half of pregnancy reduces water consumption in rats. Endocrinology. 136:1892-7. Zimmermann, S., G . Steding, J . M . Emmen, A . O . Brinkmann, K . Nayernia, A . F . Holstein, W . Engel, and I . M . Adham. 1999. Targeted disruption of the Insl3 gene causes bilateral cryptorchidism. Mol Endocrinol. 13:681-91. 39 Chapter 2. Relaxin Becomes Uprequlated During Prostate Cancer Progression to Androgen Independence and is Negatively Regulated by Androgens1 2.1 Introduction Prostate cancer is the most common cancer in men and the second leading cause of cancer related death. Organ confined early stage disease is curable by surgery; however, advanced prostate cancer is typically treated by androgen ablation which causes an initial apoptotic regression of the cancer. Unfortunately, the remission is usually temporary and A l cells arise that are able to survive and proliferate in the absence of androgens. Therapeutic options at this stage are limited (Craft et al., 1999; Rennie and Nelson, 1998). This process of progression is considered to be multifactorial, likely driven by genetic aberrations, changes in gene expression and activation of various growth factor pathways. Investigation of these dysregulated pathways wi l l help to identify new therapeutic targets and biomarkers for prognosis. In the human male relaxin is produced in the secretory epithelial cells of the prostate (Gunnersen et al., 1995a; Sokol et al., 1989; Yki-Jarvinen et al., 1983b), and is 1 A version of this chapter has been published. Thompson, V . C . , Morris, T .G. , Cochrane, D.R, Cavanagh, J., Wafa, L . A . , Hamilton, T., Wang, S., Fazl i , L . , Gleave, M . E . , and Nelson, C C . 2006. Relaxin becomes upregulated during prostate cancer progression to androgen independence and is negatively regulated by androgens. Prostate. 66:1698-709. 40 released into the seminal fluid (De Cooman et al., 1983; Schieferstein et al , 1989) where it is thought to play an important role in fertility. In a study of serum biomarkers in various cancerous and non-cancerous tissue types, relaxin HI was up-regulated in tumors compared to normal prostate tissue (Welsh et al., 2003). The role of relaxin in prostate cancer progression has not been explored. Relaxin has been primarily studied in the context of its role in pregnancy and pelvic girdle relaxation, where it derives it name. In various tissue types, relaxin causes collagen remodelling through increased collagenases and a concomitant decrease in collagen and tissue inhibitors of metalloproteinases (TIMPs). Relaxin also causes increased angiogenesis through upregulation of vascular endothelial growth factor (VEGF) and matrix metalloproteases (MMPs) and vasodilation through increased nitric oxide (NO) levels, caused by upregulation of nitric oxide synthase (NOS) (reviewed in Bani, 1997b). The transcriptional regulation of relaxin in the prostate has not yet been thoroughly elucidated. Promoter-reporter based assays using the relaxin promoter linked to luciferase have been performed (Brookes et al., 1998; Fu, 2000; Gunnersen et al., 1995b), in prostate cancer cell lines in which Brookes et al (Brookes et al., 1998) found the relaxin promoter to be positively regulated by androgens in the PC-3 prostate cell line transfected with the androgen receptor (AR). However, androgens failed to increase reporter expression under the control of the relaxin promoter in LNCaP cells and non-prostate cells examined: liver, kidney, bladder, lung, breast, and ovarian. These studies suggested that the relaxin 5' upstream regulatory region could be activated by androgens, but this regulation of relaxin was cell type dependent. 41 In rat luteal cells, relaxin m R N A is upregulated by prolactin or rat placental lactogen, likely signalling through the prolactin receptor. This prolactin induced expression is independent of E R K , and requires P K C delta, however P K C delta is not sufficient for relaxin expression. STAT-3 complexes with P K C delta (Peters et al., 2000), and has been shown to bind upstream of the relaxin gene. However, STAT-3 appears to repress relaxin expression, while STAT-5 upregulates relaxin expression in the rat (Soloff et al., 2003). AP-1 has also been implicated in relaxin expression (Bernacki et al., 1998; Garibay-Tupas et al., 1999; Garibay-Tupas et al., 2004; Soloff et al., 2003), possibly mediated through a functional AP-1 binding site in the relaxin H2 upstream region (Garibay-Tupas et al., 1999). In contrast to previous observations, we have found that the endogenous relaxin H2 gene in L N C a P cells in vitro is repressed by the administration of androgens. In L N C a P tumors grown as xenografts in male mice, relaxin is expressed at a low level and upon castration relaxin expression dramatically increases during A l progression. Furthermore, using tissue arrays of prostate cancer patient samples that have undergone androgen ablation prior to surgery, we have found that relaxin increases with the duration of androgen ablation and is highly expressed compared to untreated or hormone refractory samples. Our observations suggest that relaxin is negatively regulated by androgens in the prostate and hormone ablation in prostate cancer patients leads to an in increase in relaxin production. 42 2.2 Materials and Methods Plasmid construction A R expression vector was constructed as previously described by Snoek et. al (Rennie et al., 1993; Snoek et al., 1996). SRC1 and C B P expression constructs were obtained from Tsai (Onate et al., 1995) and Curran (D'Arcangelo and Curran, 1995), respectively. Cel l Culture and Media Cell Culture - L N C a P cells, passage 35-50, were maintained in R P M I 1640 with 5% fetal calf serum (FCS) and penicillin/streptomycin at 37 °C and 5% CO2. Hormone Treatments - Fetal bovine serum (FBS) was incubated with charcoal and dextran (0.1% final), then filtered before use as charcoal stripped serum (CSS). A t approximately 60% confluency, L N C a P cells received R P M I + 2% CSS and were allowed to grow for two days. The stock solution of synthetic androgen, R1881, was dissolved in absolute ethanol, and applied to cells in 20%) ethanol. For microarray R N A extraction, L N C a P cells were seeded in 10 cm plates (3 X 10 6 cells/plate) and incubated until 60-70%) confluency at 37 °C. Cells were transfected with 1.5 jig A R expression vector, 15 ug of empty p R C C M V vector or coregulator (SRC1 or C B P ) , and 0.5 \ag of pRL-tk renilla luciferase reporter. The A R expression plasmid was transfected into L N C a P cells because the effect of co-regulators is more marked when the receptor and the co-regulator are overexpressed at the same time; any co-regulator effect is less noticeable when only endogenous receptor is present. Cells were treated with or without 43 I n M R1881 ligand in 5% dextran-coated charcoal stripped F B S for 24 hours before R N A extraction. Preparation of total R N A The L N C a P cells were lysed in Trizol (Invitrogen) and total R N A was isolated according to the manufacturer's protocol. Total R N A samples were quantitated using the Pharmacia Biotech Ultrospec 3000. Microarray procedures 15-20 ug total R N A was used to label c D N A using Cy3 dUTP (GE Healthcare) during reverse transcription by Superscript II (Invitrogen). The labelled c D N A was cleaned using a Qiagen P C R purification kit, spiked with Cy5 labelled reference oligonucleotide and 20 ixg glycogen. Precipitated D N A was resuspended in 50 ul hybridization solution (55 ul formamide, 5 X SSC, 0.1% SDS, 25 ug poly dA, 25 ug yeast t R N A , 20 ug salmon testes D N A ) and denatured, then applied to denatured slide. Human Named (Research Genetics) library spotted c D N A slides representing 7,500 genes were prepared by washing in 0.1 % SDS, blocking in 0.1 M boric acid, p H 8.0 and 0.42 g succinic anhydride in l-methyl-2-pyrrolidinone, prior to denaturation in 95 °C water. After overnight hybridization at 42 °C, slides were then washed, dried, and scanned using a Virtek ChipReader. 44 L N C a P xenograft model A l l animal procedures were performed with approval of the University of British Columbia Committee on Animal Care. 2 x l O 6 L N C a P cells were inoculated s.c. with equal volumes (120 uL) of Matrigel (Becton Dickinson Labware) in 6 sites (right and left shoulder, right and left flank and right and left hip) in eighteen 6- to 8-week-old athymic male nude mice ( B A L B / c , Charles River Laboratory) under mefhoxyfluorane anaesthesia (Pitman-Moore) as previously described (Gleave et al., 1992). Determination of Serum PSA Levels - Blood serum samples were obtained from tail vein incisions every 3 - 5 days to follow the progression to A l gene expression. Serum P S A levels were determined by an enzymatic immunoassay kit (Abbott I M X ) , according to the manufacturer's protocol (Gleave et al., 1992). Surgical Procedures and Harvesting of Tumors - Tumors were grown for 6-8 weeks at which time the tumors were approximately 1.0 cm in diameter. 2 animals were sacrificed using CO2 asphyxiation and the tumors removed surgically, ensuring removal of any stromal layers. Tumors were placed into 4 m L of Trizol (Invitrogen) and immediately frozen at -80°C. The remaining animals were surgically castrated via the scrotal route under mefhoxyfluorane anaesthesia (Pitman-Moore) as previously described (Gleave et al., 1992). The animals were sacrificed, and the tumors harvested at the indicated times post-castration. R N A was isolated from each of the tumors homogenized using a Polytron power homogenizer (Kinematica) in Trizol , using Invitrogen instructions. 45 Generation of Northern Blot Probes The relaxin H2 gene was cloned using the primer pairs h R L X H 2 r e v (5' atactcgagatcactattatcagcaaaatctagcaagag 3'), h R L X H 2 f w d (5' ataggatccatgcctcgcctgttttttttccac 3') from Operon, and L N C a P c D N A as template. The L N C a P c D N A was generated using random hexamer [p(dN)6, (Roche)] primers and Superscript II (Life Technologies, Inc.). Relaxin H2 was cloned into p IRES-GFP (Stratagene), using B a m H l and X h o l double digest. This was used as template to generate the northern blot probe. P S A was digested from pBluescr ip tPSA construct, using E c o R l , and used as template to generate northern blot probes, as described previously (Gleave et al., 1991). Northern Blot Analysis 15-20 jig of total R N A was denatured in de-ionized formaldehyde/ formamide/ M O P S sample buffer and electrophoresed on a M O P S formaldehyde agarose gel. The R N A was transferred to a nylon membrane in 20X SSC (3 M sodium chloride, 0.3 M sodium citrate, p H 7.0) for 20 h, then crosslinked to the membrane using a U V Stratalinker (Stratagene) according to manufacturer's instructions. The membrane was prehybridized in Expresshyb (Clontech) containing denatured salmon testes D N A for 3 hrs at 65 °C on a rotating rack in a hybridization oven. Radioactivity was incorporated into the full length relaxin c D N A probe using Ready-To-Go D N A labelling beads (Amersham Pharmacia Biotech) to a specific activity of l - 2 x l 0 8 dpm/ug. The probes were allowed to hybridize to the membrane overnight at 65 °C. High stringency washes were performed on the membranes at 65 °C. The membranes were exposed to Kodak M S 46 film or a phosphoimager screen, and band intensities were analysed using Imagequant (Amersham). Q- P C R 2 ug total R N A isolated from L N C a P cells was treated with 1 U DNase I (amplification grade, Invitrogen) in 1 X DNase reaction buffer. The reaction was stopped with 2.27 m M E D T A and 10 min incubation at 65 °C. DNase treated R N A was added to a mix of I X M M L V first strand synthesis buffer (Invitrogen), 0.01 M D T T (Invitrogen), 1 m M dNTPs (Invitrogen), 1.33 U / u l R N A s i n (Promega), 6.67 U/u l M -M L V reverse transcriptase (Invitrogen), and 8.33 U/u l random hexamers for reverse transcription. The reverse transcription mix was incubated at room temperature for 10 min, 37 °C for 1 hour, 95 °C for 5 min, then cooled to 4 °C prior to use in quantitative real time P C R (Q - P C R ) . The Q- P C R mix contained c D N A from the reverse transcription reaction above, mixed with I X T A Q M A N mix (Applied Biosystems), 0.9 u M forward relaxin primer with sequence 5' - T G A A G C C G C A G A C A G C A G T - 3' (Operon), 0.9 u M reverse relaxin primer with sequence 5' - A A C A T G G C A A C A T T T A T T A G C C A A - 3 ' (Operon), and 0.2 u M relaxin Taqman probe with sequence 5' - V I C -A A A A T A C T T A G G C T T G G A T A C T C A T T C T C G A A A A A A G A G A - T A M R A - 3' ( U B C N A P S ) to detect relaxin m R N A . 18S r R N A was used as the endogenous control, using I X primer and probe set (Applied Biosystems). The transcript levels were detected using the w A B I 7900 H T Sequence Detection System, with cycling conditions of 50 °C for 2 min, 95 °C for 10 min, and then 40 repeats of 95 °C for 15 sec and 60 °C for 1 min. 47 Antibody to relaxin H2 A n antibody to relaxin H2 was made by immunizing mice to synthesized relaxin A chain conjugated to K L H at the C-terminus using the Imject Immunogen E D C conjugation kit with m c K L H (Pierce). The relaxin A chain was synthesized without a pyroglutamate at the N-terminus to alleviate possible problems during synthesis. A l l peptide synthesis steps were performed by Dalton Chemical Laboratories. The anti-relaxin polyclonal antibody was generated by immunizing 3 rabbits with the KLH-conjugated relaxin A chain in Freund's adjuvant, then boosted after 14 and 28 days in Freund's incomplete adjuvant. Rabbits were again boosted at 4 to 6 week intervals in Freund's incomplete adjuvant prior to bleeding. The antiserum was prepared, and the antibody isolated with Protein A / G P L U S agarose beads (Santa Cruz Biotechnology). The in-house antibody was later tested against the commercial antibody to validate the antibodies. Tissue Microarrays and Immunohistochemistry Formalin-fixed, paraffin-embedded human prostate tissue cores were selected from donor tissue blocks, and arranged into two arrays in recipient paraffin blocks using a tissue arrayer (Beecher Instrument, Silver Spring, M D ) , and 5 um sections cut from the recipient blocks and mounted to slides (Kiyama et al., 2003). On the first array, T M A -N H T - V 2 , 111 patient samples were arrayed in triplicate, representing patients treated with neoadjuvant hormone therapy (NHT) for 0 (n=21), <3 (n=21), 3-6 (n=28), and >6 months (n=28), and A l patients (n=13). A l patient samples were collected from distal 48 sites, including bone (n=6), liver (n=2), and lymph nodes (n=3), adrenal (n=2). Tumours from untreated patients had Gleason grades in the range of 3 to 4, with 95% of the tumours having Gleason grade of 3. Tumours in hormone ablated patients can no longer be characterized by Gleason grade, due to morphometric changes induced by androgen withdrawal. However, all patient samples were roughly matched for grade and stage, for both T M A - N H T - V 2 and T M A - N H T V I . The second array, T M A - N H T - V 1 , was described previously (Kiyama et al., 2003). 185 specimens were arrayed in duplicate, representing patients treated with neoadjuvant hormone therapy (NHT) for 0 (n=25), 1 month (n = 56), 3 months (n = 24), and 6-8 months (n = 56), and A l patients (n=24). A l patient samples were collected from transurethral resections. Neoadjuvant hormone therapy involved medical castration via L H R H analogue, with or without non-steroidal anti-androgen, for patients on both T M A - N H T - V I and T M A - N H T - V 2 . Informed consent was obtained from each patient to use their tissues for research. Sections were rehydrated, then steamed in citrate buffer (8.2 m M sodium citrate, 1.8 m M citric acid) for antigen retrieval. The sections were blocked with 10%> B S A , and endogenous peroxidase activity was blocked using 3%> H2C»2 in P B S , prior to incubating with a 1:100 dilution of rabbit anti-relaxin antibody (ImmunDiagnostik for T M A - N H T -V2or in-house for T M A - N H T - V 1 ) overnight. To detect relaxin staining, D A K O Envis ion + System, H R P (Dako) was applied, Nova-red (Vector, Burlingame, C A ) was used as chromogen after P B S washes, and then slides were counter-stained with Hematoxylin (Vector, Burlingame,CA). Slides were dehydrated, and cover slip applied. Slides were scored for relaxin staining by a pathologist (L.F.) with a system of four grades of staining: 0 for no staining, 1 for weak staining, 2 for moderate staining, and 3 49 for high staining intensity. Slides were also photomicrographed using the B L I S S system (Bacus Laboratories, Inc. Lombard, IL), and scored with automated quantitative image analysis software, Image Pro-Plus (Media Cybernetics, Carlsbad, C A ) . Statistical Analysis For microarray analysis, ImaGene 4.2 (BioDiscovery) software was used for quantification of spot intensity; GeneSpring 4.2 (Silicon Genetics) was used for subsequent analysis and normalization. To analyse the dose response and timecourse data, a repeated measures A N O V A was performed, followed by the Tukey-Kramer multiple comparisons test. A l l statistical analyses for timecourse data was conducted using GraphPad Prism (GraphPad Software Inc, San Diego, C A ) . To analyse the difference between precastrate relaxin levels and levels 35 days post-castration, the best fit line for the curve was generated using Microsoft Excel , providing a slope and an R 2 value. Standard error of the mean was used to generate error bars for graphs of both Q-P C R and northern blot results. Immunohistochemistry scoring results of tissue microarrays stained for relaxin were analysed using Wilcoxon Rank sum test. 2.3 Results Relaxin is down-regulated by androgens in L N C a P cells in tissue culture In order to examine the changes in gene expression that occur due to androgen ablation and subsequent A l , we have used microarray-based profiling with the L N C a P human prostate cancer cell line. To determine changes in gene expression due to androgens, total R N A isolated from L N C a P cells transfected with the A R and treated 50 with or without the synthetic androgen, R1881, in vitro was analyzed using c D N A microarrays. Normalized intensity of relaxin in the presence of R1881 was 0.17, but in the absence of R1881 was 2.12. From this data, relaxin was downregulated more than 12-fold by androgens. In subsequent experiments, the same trend was seen irrespective of co-transfected steroid receptor coactivator-1 (SRC-1) or CREB-binding protein (CBP) , two co-regulators of the A R . In these microarrays representing 7,500 genes, 26 genes were upregulated 5-fold or greater by androgens (Appendix I), and 21 genes were downregulated 5-fold or greater by androgens (Appendix II). To confirm i f the relaxin gene was negatively regulated by androgens in prostate cancer cells, Northern blots of total R N A isolated from L N C a P cells cultured in vitro in a dose response curve to R1881 were analyzed (Fig. 2.1). These data showed that in L N C a P cells treated with increasing concentrations of R1881 for 72 hours, relaxin m R N A levels were significantly decreased up to 2-fold in a dose dependent manner. Relaxin m R N A levels were significantly lower when LNCaPs were treated with 1.0 n M or 10 n M R1881 (p<0.01, A N O V A , and Tukey comparison test) compared to untreated L N C a P cells. To ensure that the L N C a P cells were responding to androgen treatment as expected, P S A m R N A levels were also examined by northern blot, and found to increase in response to increased androgen levels. To further determine the regulation of relaxin by androgens, L N C a P cells were treated in vitro with 1.0 n M R1881 in a time course for up to 72 hours. The maximal decrease in relaxin m R N A levels occurred by 72 hours of R1881 treatment (Fig. 2.2, p<0.001, A N O V A , Tukey comparison test), as determined using Q- P C R to analyse total R N A isolated from the treated L N C a P cells. These results showed that the level of 51 e P - P 0 LU o o v- t - n M R 1 8 8 1 A B c , • r ] 2 8 s 0.0016 D 0.0014 0.0012 i§ 0.0010 co 0.0003 c 0.0006 | 0.0004 0.0002 0 0 0.01 0.1 1 10 R1881 concentration (ntvl) Figure 2 . 1 Relaxin levels i n L N C a P cells decrease with increased androgen concentration. Northern blot analyses of androgen titration in L N C a P cells with 72 hours androgen exposure in vitro, probed for A ) relaxin, B ) P S A . C) Ethidium bromide stain of total R N A run on 1% formaldehyde agarose gel at 90V for 1 hour prior to blotting to nylon membrane. D) Graph shows normalized relaxin levels decrease in a dose dependent manner. Values represent average of at least three experiments, with standard error of the mean represented by error bars. * - significantly lower than control (p-<0.01, A N O V A and Tukey comparison test). 52 w <D > 0) X 0) mm CD > re 24 48 72 R1881 treatment duration (hours) Figure 2.2 Relaxin levels in LNCaP cells decrease with increased duration of androgen treatment. Q- P C R analysis of relaxin levels in L N C a P cells treated with 1 n M R1881 for 0 to 72 hours. Graph shows ratio of relative relaxin m R N A levels in R l 881-treated over control samples, and is representative of results of performed in triplicate, with error bars representing standard error of the mean. * Significantly different (p<0.001, A N O V A with Tukey comparison test) from 0 hr timepoint. 53 relaxin m R N A decreased dramatically by 24 hours of androgen exposure. Further treatment with androgens resulted in a continued suppression of relaxin m R N A levels compared to untreated control (p<0.001, A N O V A , Tukey comparison test), with a non-significant decrease in relaxin levels from 24 hours to 72 hours. Relaxin levels are increased by androgen ablation in prostate tumors In order to examine the changes in relaxin gene expression that occur due to androgen ablation and subsequent A l we have used the L N C a P human xenograft prostate tumor model. This tumor model was generated from a lymph node metastasis of advanced prostate cancer and is representative of the human disease in that it is of secretory epithelial origin, expresses an androgen receptor (albeit with a mutated A R that alters ligand specificity) and produces and secretes P S A in response to androgens. However, unlike the clinical disease the L N C a P tumor model fails to undergo an apoptotic regression upon castration when grown in nude mice. These features of the L N C a P model make it very useful for studying gene expression changes regulated by androgen ablation, without the confounding affects of induction of apoptosis. In this tumor model we have studied the effects of castration on prostate tumor gene expression and the subsequent progression to A l , as defined by the re-expression of prostate specific antigen (PSA) in the absence of androgens (Bladou et al., 1997; Gleave et al., 1992; Sato et al., 1996). P S A is a serine protease under the control of androgens, normally expressed by the secretory epithelial cells of the prostate (Gleave, 1995), and levels of P S A detectable in the serum are proportional to tumor burden in an intact animal (Gleave et al., 1992). After castration in the L N C a P model, P S A levels fall and the 54 tumor volume plateaus for about 21 days, at which time the cells regain the ability to express P S A and grow in the absence of androgens (Bladou et al., 1997; Gleave et al., 1992; Sato et al., 1996). Since relaxin levels are down-regulated by androgens in L N C a P cells in vitro, we sought to determine the effect of removal of androgens in vivo, through castration, in the L N C a P xenograft model. A s expected, serum P S A levels dropped after castration, reached nadir level at day 8, and then increased continually through to 35 days post castration, indicating the androgen independent expression of P S A by L N C a P tumor cells. Northern blots were used to assess changes in relaxin and P S A m R N A levels in the L N C a P xenograft model. P S A m R N A levels followed those of serum P S A levels, with nadir levels at 8 days post castration (Fig. 2.3A). Relaxin transcript levels began to increase 9 days after castration, and increased continuously with progression to A l (Fig. 2.3B). Relaxin levels were highest at A l , at approximately 28 times that of pre-castrate levels, as determined by densitometry (Fig. 2.3C). This data indicates that relaxin levels increase dramatically with increased duration of androgen withdrawal. To analyze protein expression levels of relaxin we created a polyclonal antibody by immunizing rabbits to a synthetic peptide fragment of relaxin H2 A chain. This antibody was validated using recombinant relaxin obtained from ImmunDiagnostik. To determine i f relaxin levels also increased in human patient samples following androgen ablation, a tissue microarray ( T M A ) of prostate cancer samples from 111 patients treated with androgen ablation prior to prostatectomy [neo-adjuvant hormone therapy (NHT)] and control patients ( T M A - N H T - V 2 ) , was probed by a commercial antibody (ImmunDiagnostik) to detect relaxin by immunohistochemistry (IHC). This T M A -55 >< x x x o O O O n O O • ' JL ' . i T -o "o TJ £ "o -a ts ^ ^_ LO Q- CD CO 03 i - CNJ CO A B c Figure 2.3 Relaxin m R N A levels increase with progression to androgen independence in L N C a P subcutaneous xenograft model in nude mice. Northern analysis of 20 ug total R N A from the L N C a P xenograft probed with (A) P S A and (B) Relaxin H2 c D N A . C) Ethidium bromide stain of total R N A run on 1% formaldehyde agarose gel at 90V for 1 hour prior to blotting to the nylon membrane. 56 N H T - V 2 contains samples from 5 groups of patients: those who, prior to surgery, underwent hormonal treatment for 0 months, less than 3 months, 3-6 months, and greater than 6 months, as well as those who had A l prostate cancer. The T M A sections were given a score from 0 (no staining) to 3 (intense staining). T M A - N H T - V 2 was also scored by automated quantitative software, and scoring results using the software were statistically similar to the results obtained by the pathologist. The results of the IHC demonstrated that relaxin levels were significantly higher in patients who had been treated with N H T for an increased period of time, with most intense relaxin staining in the >6 month N H T group (Fig. 2.4A). The staining in the >6 month N H T group was significantly different (p<0.05, Wilcoxon rank sum) from the untreated and <3 month N H T groups. Representative staining is shown in figures 2.4B-E . For T M A - N H T - V 2 , A l prostate cancer samples were harvested from several metastatic sites: bone, liver, and lymph nodes. Bone metastasis samples had the highest average relaxin staining score (1.44) compared with the liver and lymph node metastases, with scores of 0.5 and 0.83, respectively (data not shown). Differential staining intensity is shown in the biopsy cores from bone and lymph nodes in figures 2.4F and G , respectively. This trend is similar in the T M A - N H T - V 1 array (data not shown), which was probed for relaxin using the anti-relaxin antibody generated in our laboratory. Additiaonally, T M A - N H T - V 1 contained A l cores from transurethral resections, and these had similar staining scores (1.0) to the samples from patients who had not undergone N H T (1.025) (Data not shown). Relaxin IHC staining was limited to the prostatic secretory epithelial cells and present in most patient tumor samples, regardless of duration of N H T treatment. 57 A NHT Groups (months) , . V KIT J " v . . • p . „ . . - " - p >«».1 T1 Figure 2.4 Relaxin levels increase in human prostate cancer tissue with increased duration of neoadjuvant hormone therapy (NHT). Paraffin embedded tissue samples from patients receiving radical prostatectomy after varying durations of androgen ablation were arrayed in a tissue microarray and assayed immunohistochemically. A) Immunohistochemistry scoring results of tissue microarrays stained for relaxin expression are presented as median score for each timepoint, with standard error indicated. * >6 month treatment group is statistically significant from <3 month and 0 treatment groups (p-value <0.05, Wilcoxon rank sum). ** 3-6 month treatment group is statistically significant from 0 treatment group (p-value <0.05, Wilcoxon rank sum). B-G) Representative images of paraffin embedded sections from tissue array, T M A - N H T - V 2 , stained for relaxin, x400 magnification. B) 0 month NHT, C) <3 month NHT, D) 3-6 month NHT, E) >6 month NHT, F) A l bone, and G) A l lymph node samples are shown. 58 Staining was more diffuse throughout the cell with increased N H T duration. In addition to staining being more diffuse, staining was more intense in sections with higher scoring (Fig. 2.4B-G).These results indicate that relaxin levels are elevated in the absence of androgens in human prostate cancer patients, as they are in vitro. 2.4 Discussion Advanced prostate cancer is commonly treated by androgen ablation therapy, which induces an apoptotic regression of prostate tumors and normal secretory prostate epithelial cells. Unfortunately, this remission is typically temporary with the eventual emergence of A l prostate cancer in the majority of patients. The L N C a P xenograft model has been used extensively to characterize changes in gene expression during progression to A l (Ettinger et al., 2004; Gimenez-Bonafe et al., 2004; Miyake et al., 2000; Sato et al., 1997). In this study, we found that relaxin levels increase with progression to A l in the L N C a P xenograft model, and in human patient samples following androgen ablation. We also provide evidence that relaxin is negatively regulated by androgens in prostate cancer through the A R . Since in L N C a P cells, relaxin levels increased in a manner corresponding to the removal of androgens in microarray experiments, we hypothesized that relaxin may be negatively regulated by androgens in the prostate. Our data demonstrate that relaxin levels in vitro decrease in the presence of synthetic androgen, R1881 (Figs. 2.1 and 2.2). The exact mechanism underlying this regulation is under current investigation. The relaxin promoter has been described as being regulated positively by androgens when recombinantly linked to a C A T reporter assay in PC-3 cells transfected with the 59 A R (Brookes et al., 1998). The difference in results may simply be due to a difference in cell lines, as PC-3 cells express a different repertoire of proteins than L N C a P . Alternatively, the difference in results could be because a recombinant reporter assay does not take into consideration all the possible mechanisms for control of relaxin expression, as only 300 and 3000 bp 5' flanking regions were analysed for androgen regulation. It is possible that the negative regulation is not present in the 3000 bp 5 ' flanking region , but instead within an intron (Martinez et al., 1999; Nonaka et al., 2001; Watt et al., 2001), or that relaxin regulation is dependent on features of chromatin structure. Dyscoordinate regulation between endogenous promoter and reporter constructs have been noted in other studies. For example, in L N C a P cells and in mice, ornithine decarboxylase (ODC) is upregulated by androgens (Betts et al., 1997; Levi l la in et al., 2005; Levil lain et al., 2003). However, when a portion of the O D C promoter is transfected into L N C a P , O D C promoter activity is negatively regulated by androgens (Bai et al., 1998), suggesting that the reporter assay is not representative of the biological conditions within the cell. Fu et al (Fu, 2000) studied consensus regions in the human and rat relaxin promoters by the electrophoretic mobility shift assay using rat nuclear extracts derived from rat tissue samples. Some consensus regions were bound by factors present only in pregnant rats, and other consensus regions were bound by factors present in pregnant and non-pregnant rats. N o studies parallel to the study by Fu et al (Fu, 2000) have been reported in human models, although there have been studies involved in elucidating some of the regulatory regions of the 5' flanking regions of relaxin H I and H2 genes 60 (Brookes et al., 1998; Fu, 2000; Garibay-Tupas et al., 1999; Garibay-Tupas et al., 2004; Gunnersen et al., 1995b). In pregnant rats, prolactin stimulates relaxin expression (Peters et a l , 2000; Sortino et al., 1989), while in humans, prolactin levels increase due to relaxin expression ( K i m et al., 1998; Telgmann and Gellersen, 1998; Tseng and Mazella, 1999). There is no direct evidence that relaxin expression is stimulated by prolactin in humans. A s relaxin was negatively regulated by androgens in L N C a P cells in vitro, the effect of androgens on relaxin levels in vivo was studied. Relaxin levels increase during progression to A l in the L N C a P xenograft model, with peak relaxin levels at 35 days post-castration, at which time P S A is re-expressed in the absence of androgens, heralding A l in this model (Fig. 2.3). The increase in relaxin levels is also seen in human samples treated with androgen withdrawal therapy (Fig. 2.4A); levels peak following greater than 6 months N H T group, while comparatively decreased in the A l samples. While the L N C a P tumor model mimics many aspects of the clinical disease, there are distinctive differences. Following castration in the L N C a P xenograft, P S A levels decrease as seen in patients; however, L N C a P tumors do not regress, unlike the N H T treated patients. Furthermore, while the L N C a P cells are considered moderately well differentiated, they would be considered more aggressive than the patient tumors represented in the N H T T M A , because they are derived from a human patient lymph node metastasis. Therefore, the difference in peak relaxin levels may be due to these inherent attributes of the L N C a P model, compared to the clinical disease. A s well , the A l specimens on N H T - T M A - V 2 are from different secondary sites, reducing the sample size on the T M A for any one site. As each metastatic lesion would be exposed to a 61 different microenvironment, more samples are necessary to fully explore the protein expression profiles at these different sites. Despite the differences in the peak relaxin levels, relaxin dramatically increases following androgen ablation in both the L N C a P xenograft and patient samples. A n intriguing finding of this study is that relaxin levels increase following androgen ablation, while P S A levels decrease and reach nadir, prior to increasing with progression to A l , which characterized by re-expression of P S A . Some genes may be regulated in the same manner during A l as they were prior to androgen loss. P S A is an excellent example of a gene which is upregulated by androgens and is decreased in hormone responsive tumours upon initiation of androgen ablation therapy, but then is re-expressed in A l tumours. Relaxin is upregulated during androgen ablation, and is subsequently expressed at A l similarly to expression in untreated tumours. The relaxin protein expression profile in the N H T - T M A s is consistent with our hypothesis that relaxin may be involved in activating key pathways in invasion and angiogenesis during the progression of androgen independent prostate cancer. There are a number of possible mechanisms which may lead to A l , as described by Feldman and Feldman (Feldman and Feldman, 2001). Further testing is required to determine the exact mechanisms by which androgens and other factors control and modulate relaxin expression in the prostate. Relaxin levels are increased above normal levels in other neoplastic tissues including gastric cells (Stemmermann et al., 1994), breast cells (Tashima et al., 1994), chioriocarcinoma, and hydatiform mole, (Yki-Jarvinen et al., 1983a), as detected by IHC. This indicates that relaxin may .play a role in tumor progression in these cancer 62 types. Relaxin H2 has been implicated in aggressive cancer, for example, serum relaxin levels are elevated in patients with metastatic breast cancer, compared to localised disease (Binder et al., 2004). Relaxin H I has been reported to be overexpressed 4.6 times in tumor tissue over normal in the prostate (Welsh et al., 2003). Most research into relaxin in cancer has been done in breast cancer (Bani, 1997a; Bani et al., 1999b; Bani et a l , 1994; Bigazzi et al., 1992; Binder et al., 2002; Sacchi et a l , 1994; Tashima et al., 1994), with a few studies in ovarian cancer (Yki-Jarvinen et al., 1983a) and prostate cancer cell lines (Brookes et al., 1998; Gunnersen et al., 1995b). To our knowledge, this is the first evidence that relaxin levels increased during progression to A l prostate cancer. In the present study, there is an apparent differential level of relaxin expression based on tumor site. Bone metastases exhibit the highest relaxin staining of all A l cells studied (Fig. 2.4F-G) as compared to liver and lymph node metastases, and primary site samples. Prostate cancer bone metastases are the most common metastatic lesion of prostate cancer and modulate bone turnover through the degradation of the extracellular matrix (Bogdanos et al., 2003). Relaxin may play a role in general in the invasion phenotype of metastases by increasing collagen remodeling, through increased M M P s and decreased TIMPs. The bone environment may have increased levels of growth factors that further stimulate the production of relaxin, such as has been noted for IGF-1 signalling increasing relaxin in the placenta (Huang et al., 1992). Like relaxin, other genes are similarly downregulated by androgens, and become upregulated during A l progression. ET-1 is upregulated with progression to A l , and downregulated by androgens in PC-3 cells with transfected androgen receptor (Granchi 63 et al., 2001). Other groups have done large-scale analyses of genes and proteins regulated by androgens in the prostate and found a series of genes to be repressed by androgens; however, fewer genes or proteins have been reported to be down-regulated by androgens than up-regulated (Clegg et al., 2002; Jiang and Wang, 2003; Martin et al., 2004; Wang et al., 1997). For instance, in the rat ventral prostate, 1.8% of genes studied were up-regulated 2.3 fold or more, while 1.6 % of genes were down regulated by androgens (Jiang and Wang, 2003). Those upregulated genes were classified into several groups, including metabolism, chaperones and trafficking, and protein sysnthesis. The down regulated genes were of diverse functional ontologies including insulin like growth factor binding protein-3 (IGFBP-3), ATP-binding cassette protein (Abca8), M H C class I antigen gene, Fos-related antigen, and TIMP-3 . In accordance with those studies, TIMP-3 and M H C class I antigen are also down-regulated in the microarrays used in this study. O f note is that of 52 genes downregulated by androgens 2.3 fold or more, 6(11%>) are M H C antigens or M H C receptor chains (Jiang and Wang, 2003). Using mass spectrometry to determine androgen regulation of secreted proteins, Martin et al (2004) found that 9% of proteins were upregulated, while 2.5 % were downregulated, including IGFBP-2 . Like relaxin, IGFBP-2 has been shown to increase during progression to A l (Kiyama et al., 2003). Relaxin is a provocative potential therapeutic target, as levels increase with progression to A l . It has been previously described that relaxin upregulates matrix metalloproteases ( M M P s ) in many tissues, as well as in human breast cell lines, M C F - 7 and S K B R 3 (rev. in Bani, 1997b; Binder et al., 2002). Relaxin also increases V E G F levels in several tissues, (Palejwala et al., 2002; Unemori et al., 1999; Unemori et al., 64 2000). A n increase in angiogenesis would be most pronounced i f both V E G F and M M P s were being upregulated by relaxin, as both are angiogenic factors (Chiarugi et al., 2000; Galligioni and Ferro, 2001), and could be a key factor in metastases (Weidner et al., 1993). Relaxin increases nitric oxide through the upregulation of nitric oxide synthase (NOS), and this could lead to increased blood flow to the tumor region, lending a proliferative advantage to prostate cancer cells (Bani et al., 1999a; Bani et al., 1998; Bani et al., 1995; Danielson et al., 1999). Nitric oxide has likewise been linked to angiogenesis (Cooke, 2003). Studies are underway to determine i f these linkages exist in prostate cancer. Clearly relaxin is tightly regulated in a tissue specific manner. We have shown that in the human prostate secretory epithelial cells relaxin is negatively regulated by androgens in that in the absence of androgens, as in androgen ablation, relaxin levels increase in a dose and time-dependent manner. Future studies w i l l investigate the interaction between the androgen receptor and the relaxin promoter, and the role of relaxin in prostate cancer progression, with respect to downstream effectors including V E G F , N O S , M M P s , and TIMPs. 65 2.5 References Bai , G . , S. Kasper, R.J . Matusik, P.S. Rennie, J .A. Moshier, and A . Krongrad. 1998. Androgen regulation of the human ornithine decarboxylase promoter in prostate cancer cells. J Androl. 19:127-35. Bani, D . 1997a. Relaxin and breast cancer. Bull Cancer. 84:179-82. Bani, D . 1997b. Relaxin: a pleiotropic hormone. Gen Pharmacol. 28:13-22. Bani, D . , M . C . Baccari, S. Nistri , F. Calamai, M . Bigazzi , and T . B . Sacchi. 1999a. Relaxin up-regulates the nitric oxide biosynthetic pathway in the mouse uterus: involvement in the inhibition of myometrial contractility. Endocrinology. 140:4434-41. Bani, D . , P. Fa i l l i , M . G . Bello, C. Thiemermann, T. Bani Sacchi, M . Bigazzi , and E . Masini . 1998. Relaxin activates the L-arginine-nitric oxide pathway in vascular smooth muscle cells in culture. Hypertension. 31:1240-7. Bani, D . , D . Flagiello, M . F . Poupon, S. Nistri , F. Poirson-Bichat, M . Bigazzi , and T. Bani Sacchi. 1999b. Relaxin promotes differentiation of human breast cancer cells M C F - 7 transplanted into nude mice. Virchows Arch. 435:509-19. Bani, D . , E . Masini , M . G . Bello, M . Bigazzi , and T .B . Sacchi. 1995. Relaxin activates the L-arginine-nitric oxide pathway in human breast cancer cells. Cancer Res. 55:5272-5. Bani, D . , A . Riva, M . Bigazzi , and T. Bani Sacchi. 1994. Differentiation of breast cancer cells in vitro is promoted by the concurrent influence of myoepithelial cells and relaxin. Br J Cancer. 70:900-4. Bernacki, S.H., A . Medvedev, G . Holloway, M . Dawson, R. Lotan, and A . M . Jetten. 1998. Suppression of relaxin gene expression by retinoids in squamous differentiated rabbit tracheal epithelial cells. Mol Cell Endocrinol. 138:115-25. Betts, A . M . , I. Waite, D . E . Neal, and C . N . Robson. 1997. Androgen regulation of ornithine decarboxylase in human prostatic cells identified using differential display. FEBSLett. 405:328-32. Bigazzi , M . , M . L . Brandi, G . Bani, and T .B . Sacchi. 1992. Relaxin influences the growth of M C F - 7 breast cancer cells. Mitogenic and antimitogenic action depends on peptide concentration. Cancer. 70:639-43. Binder, C , T. Hagemann, B . Husen, M . Schulz, and A . Einspanier. 2002. Relaxin enhances in-vitro invasiveness of breast cancer cell lines by up-regulation of matrix metalloproteases. Mol Hum Reprod. 8:789-96. Binder, C , A . Simon, L . Binder, T. Hagemann, M . Schulz, G . Emons, L . Trumper, and A . Einspanier. 2004. Elevated concentrations of serum relaxin are associated with metastatic disease in breast cancer patients. Breast Cancer Res Treat. 87:157-66. Bladou, F., M . E . Gleave, F. Penault-Llorca, G . Serment, P . H . Lange, and R . L . Vessella. 1997. [In vitro and in vivo models developed from human prostatic cancer]. Prog Urol. 7:384-96. Bogdanos, J., D . Karamanolakis, R. Tenta, A . Tsintavis, C. Milathianakis, C. Mitsiades, and M . Koutsilieris. 2003. Endocrine/paracrine/autocrine survival factor activity of bone microenvironment participates in the development of androgen ablation 66 and chemotherapy refractoriness of prostate cancer metastasis in skeleton. Endocr Relat Cancer. 10:279-89. Brookes, D .E . , D . Zandvliet, F. Watt, P.J. Russell, and P .L . Mol loy . 1998. Relative activity and specificity of promoters from prostate-expressed genes. Prostate. 35:18-26. Chiarugi, V . , M . Ruggiero, and L . Magnelli . 2000. Molecular polarity in endothelial cells and tumor-induced angiogenesis. Oncol Res. 12:1-4. Clegg, N . , B . Eroglu, C. Ferguson, H . Arnold, A . Moorman, and P.S. Nelson. 2002. Digital expression profiles of the prostate androgen-response program. J Steroid Biochem Mol Biol. 80:13-23. Cooke, J.P. 2003. N O and angiogenesis. Atheroscler Suppl. 4:53-60. Craft, N . , Y . Shostak, M . Carey, and C . L . Sawyers. 1999. A mechanism for hormone-independent prostate cancer through modulation of androgen receptor signaling by the HER-2/neu tyrosine kinase. Nat Med. 5:280-5. Danielson, L . A . , O.D. Sherwood, and K . P . Conrad. 1999. Relaxin is a potent renal vasodilator in conscious rats. J Clin Invest. 103:525-33. DArcangelo, G . , and T. Curran. 1995. Human genetics. Smart transcription factors. Nature. 376:292-3. De Cooman, S., P. Gil l iaux, and K . Thomas. 1983. Immunoreactive relaxin-like substance in human split ejaculates. Fertil Steril. 39:111-3. Ettinger, S.L., R. Sobel, T .G . Whitmore, M . Akbari , D.R. Bradley, M . E . Gleave, and C . C . Nelson. 2004. Dysregulation of sterol response element-binding proteins and downstream effectors in prostate cancer during progression to androgen independence. Cancer Res. 64:2212-21. Feldman, B.J . , and D . Feldman. 2001. The development of androgen-independent prostate cancer. Nat Rev Cancer. 1:34-45. Fu, P., Gunnersen, J, Tregear, G , Roche, P. 2000. Identification of consensus sequences in the rat relaxin promoter which bind nuclear proteins. In Third international conference on relaxin and related peptides. V o l . 1. G . Tregear, Ivel R, Bathgate, R A , Wade, JD, editor. Kluwer Academic Publishers, Broome Australia. 149-150. Galligioni, E . , and A . Ferro. 2001. Angiogenesis and antiangiogenic agents in non-small cell lung cancer. Lung Cancer. 34:3-7. Garibay-Tupas, J .L. , K . Csiszar, M . Fox, S. Povey, and G .D . Bryant-Greenwood. 1999. Analysis of the 5'-upstream regions of the human relaxin H I and H2 genes and their chromosomal localization on chromosome 9p24.1 by radiation hybrid and breakpoint mapping. J Mol Endocrinol. 23:355-65. Garibay-Tupas, J .L. , K . J . Okazaki, L .S . Tashima, S. Yamamoto, and G .D . Bryant-Greenwood. 2004. Regulation of the human relaxin genes H I and H2 by steroid hormones. Mol Cell Endocrinol. 219:115-25. Gimenez-Bonafe, P., M . N . Fedoruk, T .G . Whitmore, M . Akbari , J .L. Ralph, S. Ettinger, M . E . Gleave, and C .C . Nelson. 2004. Y B - 1 is upregulated during prostate cancer tumor progression and increases P-glycoprotein activity. Prostate. 59:337-49. Gleave, M . , J. Hsieh, C. Gao, A . von Eschenbach, and L . Chung. 1991. Acceleration of human prostate cancer growth in vivo by factors produced by prostate and bone fibroblasts. Cancer Res. 51:3753-3761. 67 Gleave, M . E . , J.T. Hsieh, H . C . Wu , A . C . von Eschenbach, and L . W . Chung. 1992. Serum prostate specific antigen levels in mice bearing human prostate L N C a P tumors are determined by tumor volume and endocrine and growth factors. Cancer Res. 52:1598-605. Gleave, M . E . , S.L. Goldenberg, andN. Bruchovsky. 1995. Prostate specific antigen as a prognostic predictor in prostate cancer. In Hormone-Dependent Cancer. Granchi, S., S. Brocchi, L . Bonaccorsi, E . Baldi , M . C . V i n c i , G . Forti, M . Serio, and M . Maggi . 2001. Endothelin-1 production by prostate cancer cell lines is up-regulated by factors involved in cancer progression and down-regulated by androgens. Prostate. 49:267-77'. Gunnersen, J . M . , R.J . Crawford, and G .W. Tregear. 1995a. Expression of the relaxin gene in rat tissues. Mol Cell Endocrinol. 110:55-64. Gunnersen, J . M . , P.J. Roche, G .W. Tregear, and R.J . Crawford. 1995b. Characterization of human relaxin gene regulation in the relaxin-expressing human prostate adenocarcinoma cell line L N C a P . F G C . J Mol Endocrinol. 15:153-66. Huang, C.J . , Y . L i , M . H . Stromer, and L . L . Anderson. 1992. Synergistic effects of insulin-like growth factor I and gonadotrophins on relaxin and progesterone secretion by ageing corpora lutea of pigs. J Reprod Fertil. 96:415-25. Jiang, F. , and Z . Wang. 2003. Identification of androgen-responsive genes in the rat ventral prostate by complementary deoxyribonucleic acid subtraction and microarray. Endocrinology. 144:1257-65. K i m , J.J., R . C . Jaffe, and A . T . Fazleabas. 1998. Comparative studies on the in vitro decidualization process in the baboon (Papio anubis) and human. Biol Reprod. 59:160-8. Kiyama, S., K . Morrison, T. Zellweger, M . Akbari , M . Cox, D . Y u , H . Miyake, and M . E . Gleave. 2003. Castration-Induced Increases in Insulin-Like Growth Factor-Binding Protein 2 Promotes Proliferation of Androgen-independent Human Prostate L N C a P Tumors. Cancer Res. 63:3575-3584. Levil lain, O., J.J. Diaz, O. Blanchard, and H . Dechaud. 2005. Testosterone down-regulates ornithine aminotransferase gene and up-regulates arginase II and ornithine decarboxylase genes for polyamines synthesis in the murine kidney. Endocrinology. 146:950-9. Levil lain, O., A . Greco, J.-J. Diaz, R. Augier, A . Didier, K . Kindbeiter, F. Catez, and M . Cayre. 2003. Influence of testosterone on regulation of O D C , antizyme, and N l -S S A T gene expression in mouse kidney. Am J Physiol Renal Physiol. 285:F498-506. Martin, D . B . , D .R. Gifford, M . E . Wright, A . Keller, E . Y i , D .R . Goodlett, R. Aebersold, and P.S. Nelson. 2004. Quantitative proteomic analysis of proteins released by neoplastic prostate epithelium. Cancer Res. 64:347-55. Martinez, A . , A . M . Lefrancois-Martinez, M . Manin, S. Guyot, C. Jean-Faucher, G . Veyssiere, A . Kahn, and C. Jean. 1999. 5'-flanking and intragenic sequences confer androgenic and developmental regulation of mouse aldose reductase-like gene in vas deferens and adrenal in transgenic mice. Endocrinology. 140:1338-48. Miyake, H . , M . Pollak, and M . E . Gleave. 2000. Castration-induced up-regulation of insulin-like growth factor binding protein-5 potentiates insulin-like growth 68 factor-I activity and accelerates progression to androgen independence in prostate cancer models. Cancer Res. 60:3058-64. Nonaka, M . L , G . Wang, T. M o r i , H . Okada, and M . Nonaka. 2001. Novel androgen-dependent promoters direct expression of the C4b-binding protein alpha-chain gene in epididymis. J Immunol. 166:4570-7. Onate, S.A., S .Y. Tsai, M . J . Tsai, and B . W . O'Malley. 1995. Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science. 270:1354-7. Palejwala, S., L . Tseng, A . Wojtczuk, G . Weiss, and L . T . Goldsmith. 2002. Relaxin gene and protein expression and its regulation of procollagenase and vascular endothelial growth factor in human endometrial cells. Biol Reprod. 66:1743-8. Peters, C . A . , E.T. Maizels, M . C . Robertson, R.P. Shiu, M . S . Soloff, and M . Hunzicker-Dunn. 2000. Induction of relaxin messenger R N A expression in response to prolactin receptor activation requires protein kinase C delta signaling. Mol Endocrinol. 14:576-90. Rennie, P.S., N . Bruchovsky, K . J . Leco, P .C. Sheppard, S.A. McQueen, H . Cheng, R. Snoek, A . Hamel, M . E . Bock, and B.S . MacDonald. 1993. Characterization of two cis-acting D N A elements involved in the androgen regulation of the probasin gene. Mol Endocrinol. 7:23-36. Rennie, P.S., and C .C . Nelson. 1998. Epigenetic mechanisms for progression of prostate cancer. Cancer Metastasis Rev. 17:401-9. Sacchi, T .B. , D . Bani, M . L . Brandi, A . Falchetti, and M . Bigazzi . 1994. Relaxin influences growth, differentiation and cell-cell adhesion of human breast-cancer cells in culture. Int J Cancer. 57:129-34. Sato, N . , M . E . Gleave, N . Bruchovsky, P.S. Rennie, S.L. Goldenberg, P . H . Lange, and L . D . Sullivan. 1996. Intermittent androgen suppression delays progression to androgen-independent regulation of prostate-specific antigen gene in the L N C a P prostate tumour model. J Steroid Biochem Mol Biol. 58:139-46. Sato, N . , M . D . Sadar, N . Bruchovsky, F. Saatcioglu, P.S. Rennie, S. Sato, P . H . Lange, and M . E . Gleave. 1997. Androgenic induction of prostate-specific antigen gene is repressed by protein-protein interaction between the androgen receptor and AP- l / c - Jun in the human prostate cancer cell line L N C a P . J Biol Chem. 272:17485-94. Schieferstein, G . , W. Voelter, H . Seeger, and T . H . Lippert. 1989. Immunoreactive relaxin in seminal plasma of man. IntJFertil. 34:215-8. Snoek, R., P.S. Rennie, S. Kasper, R.J . Matusik, and N . Bruchovsky. 1996. Induction of cell-free, in vitro transcription by recombinant androgen receptor peptides. J Steroid Biochem Mol Biol. 59:243-50. Sokol, R .Z . , X . S . Wang, J. Lechago, P .D. Johnston, and R.S. Swerdloff. 1989. Immunohistochemical localization of relaxin in human prostate. J Histochem Cytochem. 37:1253-5. Soloff, M . S . , S. Gal , S. Hoare, C A . Peters, M . Hunzicker-Dunn, G . D . Anderson, and T .G . Wood. 2003. Cloning, characterization, and expression of the rat relaxin gene. Gene. 323:149-55. Sortino, M . A . , M . J . Cronin, and P . M . Wise. 1989. Relaxin stimulates prolactin secretion from anterior pituitary cells. Endocrinology. 124:2013-5. 69 Stemmermann, G . N . , W . Mesiona, F .C . Greenwood, and G .D . Bryant-Greenwood. 1994. Immunocytochemical identification of a relaxin-like protein in gastrointestinal epithelium and carcinoma: a preliminary report. J Endocrinol. 140:321-5. Tashima, L .S . , G . Mazoujian, and G .D . Bryant-Greenwood. 1994. Human relaxins in normal, benign and neoplastic breast tissue. J Mol Endocrinol. 12:351-64. Telgmann, R., and B . Gellersen. 1998. Marker genes of decidualization: activation of the decidual prolactin gene. Hum Reprod Update. 4:472-9. Tseng, L . , and J. Mazella. 1999. Prolactin and its receptor in human endometrium. Semin Reprod Endocrinol. 17:23-7. Unemori, E . N . , M . E . Erikson, S.E. Rocco, K . M . Sutherland, D . A . Parsell, J. Mak, and B . H . Grove. 1999. Relaxin stimulates expression of vascular endothelial growth factor in normal human endometrial cells in vitro and is associated with menometrorrhagia in women. Hum Reprod. 14:800-6. Unemori, E . N . , M . Lewis, J. Constant, G . Arnold, B . H . Grove, J. Normand, U . Deshpande, A . Salles, L . B . Pickford, M . E . Erikson, T . K . Hunt, and X . Huang. 2000. Relaxin induces vascular endothelial growth factor expression and angiogenesis selectively at wound sites. Wound Repair Regen. 8:361-70. Wang, Z . , R. Tufts, R. Haleem, and X . Cai . 1997. Genes regulated by androgen in the rat ventral prostate. Proc Natl Acad Sci USA. 94:12999-3004. Watt, F. , A . Martorana, D . E . Brookes, T. Ho, E . Kingsley, D.S. O'Keefe, P.J. Russell, W . D . Heston, and P .L . Mol loy . 2001. A tissue-specific enhancer of the prostate-specific membrane antigen gene, F O L H 1 . Genomics. 73:243-54. Weidner, N . , P.R. Carroll, J. Flax, W . Blumenfeld, and J. Folkman. 1993. Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. Am J Pathol. 143:401-9. Welsh, J.B., L . M . Sapinoso, S.G. Kern, D . A . Brown, T. L i u , A . R . Bauskin, R . L . Ward, N . J . Hawkins, D.I. Quinn, P.J. Russell, R . L . Sutherland, S.N. Breit, C A . Moskaluk, H .F . Frierson, Jr., and G . M . Hampton. 2003. Large-scale delineation of secreted protein biomarkers overexpressed in cancer tissue and serum. Proc Natl Acad Sci US A. 100:3410-5. Yki-Jarvinen, H . , T. Wahlstrom, and M . Seppala. 1983a. Immunohistochemical demonstration of relaxin in gynecologic tumors. Cancer. 52:2077-80. Yki-Jarvinen, H . , T. Wahlstrom, and M . Seppala. 1983b. Immunohistochemical demonstration of relaxin in the genital tract of men. J Reprod Fertil. 69:693-5. 70 Chapter 3. Relaxin Drives Wnt Signalling Through Upregulation of PCDHY in Prostate Cancer 1 3.1 Introduction In men, prostate cancer is the most common cancer and the third leading cause of cancer related death. Early stage, organ confined disease is curable by surgery; however, advanced stage disease is commonly treated with androgen ablation, causing an initial apoptotic response, and decrease in tumour size. Inevitably, androgen independent (Al ) cells arise which can survive and proliferate in the absence of androgens, for which there are limited therapeutic options (Clarke, 2006; Craft et al., 1999; Rennie and Nelson, 1998). Progression to A l is considered to be multifactorial, likely driven by genetic aberrations, changes in gene expression and activation of various growth factor pathways. Investigation of these dysregulated pathways w i l l help to identify new therapeutic targets and biomarkers for prognosis. In the human prostate, the peptide hormone, relaxin, is produced by the secretory epithelial cells (Gunnersen et al., 1995; Sokol et al., 1989; Yki-Jarvinen et al., 1983), and secreted into the seminal fluid (De Cooman et al., 1983; Schieferstein et al., 1989) where 1 A version of this chapter w i l l be submitted for publication. Thompson, V . C . , Hurtado-Coll, A , Turbin, D , Fazli , L , Huntsman, D , Gleave, M E , Nelson, C C . Relaxin drives Wnt signalling through upregulation of P C D H Y in prostate cancer. 71 it is thought to play an important role in fertility. Historically, relaxin has been considered a pregnancy hormone, and studies into pelvic girdle relaxation have elucidated the pleiotrophic functions of this hormone. Relaxin-induced collagen remodelling is achieved through increased collagenases within the M M P family and decreased tissue inhibitors of matrix metalloproteinases (TIMPs) and collagen. Relaxin also upregulates nitric oxide through upregulation of nitric oxide synthase, and vascular endothelial growth factor ( V E G F ) , which, in conjunction with a decrease in TIMPs and increase in matrix metalloproteases, increases angiogenesis (reviewed in Bani, 1997). Relaxin has been shown to increase angiogenesis and tumour volume of PC-3 xenografts (Silvertown et al., 2006), to facilitate A l growth of L N C a P xenografts overexpressing a p53 mutation (R273H) (Silvertown et al., 2007; Vina l l et al., 2006), and decrease survival in the transgenic adenocarcinoma of mouse prostate ( T R A M P ) mice systemically overexpressing relaxin. Knocking down relaxin function through relaxin antagonists or relaxin receptor (LGR7) s i R N A decreases growth of prostate tumour models (Feng et al., 2007; Silvertown et al., 2007). Relaxin expression is negatively regulated by androgens in L N C a P cells in vitro and in L N C a P tumors, and these observations are also seen in clinical prostate cancer specimens following androgen ablation (Thompson et al., 2006). L N C a P cells overexpressing protocadherinY ( P C D H Y ) are resistant to stress conditions, such as androgen ablation (Chen et al., 2002). P C D H Y m R N A in L N C a P xenografts is very low in intact nude mice and 2 days post castration, but elevated 28 days post castration. However, P C D H Y protein appears to increase by 2 days post castration. P C D H Y is localized to the cytoplasm, and binds /?-catenin through a serine 72 rich catenin binding domain (Chen et al., 2002). /5-catenin is exclusively localized to the membrane in untreated LNCaPs in tissue culture (de la Taille et al., 2003), but is driven to the nucleus in P C D H Y overexpressing cells, increasing Wnt signalling via T C F elements in the colon cancer cell line, HCT116, and in several prostate cell lines regardless of A R status (Yang et al., 2005). P C D H Y m R N A expression is higher in human prostate tumour cells than normal cells, is increased further with androgen ablation therapy, and is highest at A l (Terry et al., 2006a). P C D H Y overexpression confers invasive potential, and A l growth of xenografts implanted into castrated nude mice (Terry et al., 2006a). Several neuroendocrine differentiation markers in prostate cancer cells are increased by P C D H Y , including neuron-specific enolase, c-myc, cyclin D I , c-ret, and cox-2. Additionally, several Wnts are upregulated by P C D H Y , including Wnt3, Wnt lOA, Wnt7B, and W n t l 1 (Yang et a l , 2005). W n t l 1 functions in gastrulation, causing E-cadherin to dissociate from the plasma membrane (Ulrich et al., 2005). In IEC6 intestinal epithelial cells, W n t l 1 causes cell mobility, transformation, and proliferation (Ouko et al., 2004b). Human prostate cancer samples express elevated W n t l 1, proportional to Gleason Grade (Zhu et al., 2004b). W n t l 1 is inhibited by androgens in L N C a P cells, although this is relieved in A l xenografts (Zhu et al., 2004b). To study the role of relaxin in prostate cancer, we generated L N C a P xenografts stably overexpressing relaxin ( L N C a P - R L X ) . L N C a P - R L X tumours displayed an increased take rate and showed an accelerated growth rate compared to controls. Microarray gene expression analysis of R N A isolated from xenografts indicated W n t l 1 is strongly upregulated by relaxin. As Wnt l 1 is upregulated by P C D H Y , we examined 73 whether P C D H Y was upregulated by relaxin. Relaxin overexpression in L N C a P cells upregulates P C D H Y in the presence and absence of androgens; we postulate relaxin acts in an autocrine fashion upstream of P C D H Y in regulating increased Wnt signaling and W n t l l levels. 3.2 Materials and Methods Plasmid construction The relaxin H2 gene was cloned from L N C a P c D N A [generated using random hexamer (Roche, Palo Alto , C A , U S A ) primers and Superscript II (Invitrogen, Carlsbad, C A , U S A ) ] , and the primer pairs hRLXH2rev (5' A T A C T C G A G A T C A C T A T T A T C A G C A A A A T C T A G C A A G A G 3'), h R L X H 2 f w d (5' A T A G G A T C C A T G C C T C G C C T G T T T T T T T T C C A C 3') (Operon, Huntsville, A L , U S A ) , into p IRES-GFP (Stratagene, L a Jolla, C A , U S A ) . pExchange nodule neoR (Stratagene) was inserted into p I R E S - G F P - R L X and pIRES-GFP. Cel l Culture and Media Cell Culture - L N C a P cells, passage 35-50, were maintained in R P M I 1640 with 5% fetal bovine serum (FBS) and penicillin/streptomycin at 37 °C and 5% CO2. Hormone Treatments - A t approximately 60% confluency, L N C a P cells received R P M I + 2% charcoal stripped serum (CSS) (Hyclone, Logan, U T , U S A ) for 48 h prior to R1881 treatment. 74 Stable cells - Wells of 6-well plates were transfected with 2jag p I R E S - R L X vector, or empty vector control (pIRES-GFP) using lipofectin (Invitrogen, Carlsbad, C A , U S A ) selected using 300 ug/ml G418 (Invitrogen), and screened for G F P fluorescence. Preparation of total R N A L N C a P cells were lysed in Trizol (Invitrogen) and total R N A isolated according to manufacturer's protocol; R N A was quantified using the ND-1000 Spectrophotometer (Nanodrop Technologies, Wilmington, D E , U S A ) . For microarray analysis, total R N A concentration and quality were evaluated using the 2100 Bioanalyser (Agilent, Santa Clara, C A , U S A ) . L N C a P xenograft model 2 x l O 6 L N C a P - R L X , or L N C a P - G F P cells were inoculated s.c. with equal volumes Matrigel (Becton Dickinson Labware, Bedford, M A , U S A ) in four sites (right and left shoulder and hip) in eight 6- to 8-week-old athymic male nude mice ( B A L B / c , Charles River Laboratory, Wilmington, M A , U S A ) per cell line (intact mouse experiment), or in three sites (right and left shoulder, right hip) in twelve 6- to 8-week-old athymic male nude mice ( B A L B / c , Charles River Laboratory) per cell line (castrated mouse experiment), under mefhoxyfluorane anaesthesia (Pitman-Moore, Mississauga, O N , Canada) as previously described (Gleave et al., 1992). Tumour length, width and height were measured thrice weekly, and volume calculated using the formula V = t t L W H / 6 . Tumors were grown for 42 days to approximately 100 mm 3 . 8 animals per group were surgically castrated via the scrotal route under mefhoxyfluorane anaesthesia 75 (Pitman-Moore) as previously described (Gleave et al., 1992). The animals were sacrificed 18 days post-castration, or 50 days post-inoculation for intact mice, using CO2 asphyxiation, and the tumors removed surgically, ensuring removal of any stromal layers. Tumours were dissected and frozen in liquid nitrogen for protein or R N A isolation, or formalin fixed for paraffin embedding. R N A was isolated from homogenized tumours in Trizol as previously described (Thompson et al., 2006). A l l animal procedures were performed with approval of the University of British Columbia Committee on Animal Care (Appendix III). Northern Blot Analysis 15-20 ug total R N A was analysed using northern blot analysis as previously described (Thompson et al., 2006). Briefly, R N A was electrophoresed, transferred to a nylon membrane, and prehybridized prior to hybridizing the membrane with 3 2 P-labelled probe overnight. The membranes were exposed to Kodak M S film or a phosphoimager screen. Microarray Analysis Total R N A was treated for microarray hybridization using the 3 D N A Array 350 kit (Genisphere, Hatfield, P A , U S A ) , as described previously (Kojima et al., 2006). Briefly, c D N A synthesized from 15 ug total R N A , was concentrated using a Microcon Y M - 3 0 Centrifugal Filter (Millipore, Billerica, M A , U S A ) . Samples were run in duplicate in each channel, with Universal Human Reference R N A (Strategene) as normalization control in the alternate channel. 76 c D N A in formamide-based hybridization buffer was hybridized to microarrays (Array Facility of The Prostate Centre at Vancouver General Hospital, Vancouver, Canada) spotted in duplicate with 21 329 70mer probe Human Operon Version 2.0 library (Operon, Huntsville, A L , U S A ) . 3 D N A Array 350 Capture Reagent was hybridized in formamide buffer to labelled c D N A previously hybridized and microarrays scanned on a Scan Array Express (Perkin Elmer, Wellesley, M A , U S A ) . Microarray spot intensity was quantified using ImaGene 4.2 software (BioDiscovery, EI Segundo, C A , U S A ) , normalized using GeneSpring 4.2 (Agilent Technologies, Palo Alto , C A , U S A ) , and further analyzed using Ingenuity Pathways Analysis (Ingenuity Systems, Redwood City, C A , U S A ) . Semi-Quantitative Real Time - P C R 2 ug total L N C a P R N A was DNase I (Invitrogen) treated, then reverse transcribed in I X M - M L V first strand synthesis buffer, 0.01 M D T T , 1 m M dNTPs, 6.67 U / u l M - M L V reverse transcriptase (all from Invitrogen), 1.33 U / u l R N A s i n (Promega, Madison, WI, U S A ) , and 8.33 U/u l random hexamers. The semi-quantitative real time - P C R (Q-PCR) mix contained c D N A from the reverse transcription reaction, I X TaqMan mix (Applied Biosystems, Foster City, C A , U S A ) , and primers and probes for relaxin, (see Appendix I V for sequences), W n t l 1, IRS-2 (Applied Biosystems), and 18S r R N A as the endogenous control (Applied Biosystems). P C D H Y was detected using S Y B R Green (Invitrogen) and P-actin as the endogenous control. Transcript levels were detected using the A B I 7900 H T Sequence Detection System (Applied Biosystems). 77 Tissue Microarrays Formalin fixed xenograft tumours were paraffin embedded, arrayed into two tissue microarrays (TMAs) , sliced, and affixed to slides. The intact mouse-TMA contained 10 L N C a P - G F P tumours, and 12 L N C a P - R L X tumours, in quadruplicate. The castrated mouse-TMA contained 24 L N C a P - G F P tumours, and 28 L N C a P - R L X tumours, in triplicate. The T M A of 111 human patient samples was generated as previously described as T M A - N H T - V 2 in Chapter 2 (Kiyama et al., 2003a; Thompson et al., 2006). Briefly, formalin-fixed, paraffin-embedded human prostate tissue cores were selected, arrayed into recipient paraffin blocks, and 5 | im sections cut and mounted to slides (Kiyama et al., 2003b). This T M A consisted of 111 patient samples arrayed in triplicate, representing patients roughly matched for stage and grade, treated with neoadjuvant hormone therapy (NHT) for 0 (n=21), <3 (n=21), 3-6 (n=28), and >6 months (n=28), and A l patients (n=13). A l patient samples were collected from distal sites, including bone (n=6), liver (n=2), lymph nodes (n=3), and adrenal (n=2). Neoadjuvant hormone therapy involved medical castration via L H R H analogue, with or without non-steroidal anti-androgen. Informed consent was obtained from each patient to use their tissues for research. Immunohistochemistry Immunohistochemical (IHC) staining was performed with heat-activated antigen retrieval in citrate buffer (Thompson et al., 2006) for L G R 7 , relaxin, W n t l 1, and |3-78 catenin. L G R 7 was detected using rabbit polyclonal ant i-LGR7 antibody (GeneTex, San Antonio, T X , U S A ) diluted 1:100; relaxin was detected using rabbit polyclonal anti-relaxin H2 antibody (Thompson et al., 2006) diluted 1:100, and W n t l 1 was detected using goat anti-mouse W n t l 1 antibody ( R & D Systems, Minneapolis, M N , U S A ) diluted 1:100, then treated with biotinylated secondary antibody ( D A K O L S A B + System, L I N K , Mississauga, Canada). IHC staining for P-catenin was performed with Vector M . O . M Immunodetection K i t (Vector, Burlingame, C A , U S A ) and mouse anti-P-catenin antibody (BD Transduction, San Jose, C A U S A ) diluted 1:100. IHC utilized the Discovery X T (Ventana Medical Systems, Tucson, A Z ) . Slides were scored for relaxin, L G R 7 , and W n t l 1 staining by a pathologist (L.F.) with a system of four grades of staining: 0 - absent, to 3 - high staining intensity. Slides were scored for subcellular localization of P-catenin by a pathologist (D.T.), using a system of 3 grades of staining at the cytoplasmic membrane and the cytoplasm. Slides were photomicrographed using the B L I S S system (Bacus Laboratories, Inc. Lombard, IL, U S A ) , and scored with automated quantitative image analysis software, Image Pro-Plus (Media Cybernetics, Carlsbad, C A , U S A ) . Statistics To analyze the relaxin expression by Q-PCR, and IHC, a two-tailed student's t-test was performed using GraphPad Prism (GraphPad Software, Inc., San Diego, C A ) . Standard error of the mean was used to generate error bars for graphs of both Q - P C R and tumour volume results. 79 3.3 Results Relaxin overexpression results in larger tumours through paracrine or autocrine fashion To study the role of relaxin in prostate cancer, we created L N C a P cells stably overexpressing relaxin ( L N C a P - R L X ) , using a p IRES-GFP vector (Fig. 3.1 A ) . Northern blot analysis indicated that in vitro L N C a P - R L X expressed higher levels of relaxin than empty vector (LNCaP-GFP) or parental controls (Fig. 3.IB). The relaxin message in L N C a P - R L X samples is larger than the endogenous relaxin m R N A species due to the IRES and G F P m R N A sequences present on the bicistronic message. This was validated by probing for G F P , which co-migrated with the overexpressed relaxin band, was detectable in L N C a P - G F P , and was not present in parental LNCaPs . L N C a P - R L X and L N C a P - G F P were inoculated s.c. into intact male nude mice and resulting xenografts analysed for relaxin expression. L N C a P - R L X xenografts had substantially higher levels of relaxin m R N A than control, which was barely detectable using northern blot analysis (Fig. 3.1C). Tissue microarrays ( T M A s ) of paraffin embedded xenografts were generated to characterize protein expression differences in L N C a P - R L X and L N C a P - G F P . Protein levels of relaxin were 1.4 fold (p<0.05, t-test) higher than control tumour as measured by immunohistochemistry (IHC) in L N C a P -R L X - T M A - I (Fig. 3.2A). Relaxin staining is largely cytoplasmic, and restricted to the epithelial cells (Fig. 3.2B-C). To help determine i f relaxin overexpression could have a biological effect through an autocrine or paracrine mechanism, L N C a P - R L X - T M A - I was analyzed for the relaxin receptor, L G R 7 . A l l tumors analyzed had detectable levels 80 P C M V Re lax in I R E S G F P R p "D C TO o •1 O Z X _J ct CL CD O r P C M V ^ > Re lax ing IRES ^ > GFP ^ > P C M V ^ > I R E S ^ > GFP ^ > LNCaP-GFP LNCaPftLX RLX- IRES-GFP relaxin RLX- IRES-GFP RLX- IRES-GFP relaxin Figure 3.1 Relaxin is overexpressed in L N C a P - R L X stable cell lines, a) Schematic drawing of p IRES-RLX-neo construct, indicating pIRES sequence between relaxin and G F P genes. L N C a P - R L X expresses higher levels of relaxin-containing transcript than L N C a P - G F P using Northern blot b) in tissue culture and c) in xenografts, detected using 15 ug total R N A on Northern blots probed for relaxin. 81 s o -X 0) OJ > < 3 2.5 2 1.5 1 0.5 0 m H rr O — 3 2.5 2 1.5 1 0.5 0 L N C a P - G F L N C a P - R L X L N C ; i P - G F P L N C a P R L X Figure 3.2 Relaxin and L G R 7 protein levels are higher in L N C a P - R L X xenografts grown in intact mice compared to controls, using I H C . a) Average intensity of relaxin in L N C a P - R L X and L N C a P - G F P xenografts in L N C a P - R L X - T M A - I . Representative images of relaxin staining in b) L N C a P - G F P and c) L N C a P - R L X xenografts, d) L G R 7 protein levels are slightly higher in relaxin overexpressing L N C a P xenografts in intact mice compared to empty vector controls. Representative images of L G R 7 staining in e) L N C a P - G F P and f) L N C a P - R L X xenografts. Error bars are standard error of the mean. 82 of L G R 7 staining, with relaxin overexpressing tumors having marginally 1.17 fold (p<0.03, t-test) higher levels of L G R 7 than control tumors (Fig. 3.2D). Both L N C a P -R L X and L N C a P - G F P xenografts stained positively for L G R 7 in prostate cancer epithelial cells (Fig. 3.2E-F), indicating that relaxin is capable of acting in an autocrine fashion in xenografts in vivo. L G R 7 levels were examined in a T M A of human patient samples treated for various durations of androgen ablation prior to radical prostatectomy ( T M A - N H T - V 2 ) (Fig. 3.3). L G R 7 levels were detectable in stromal and malignant epithelial cells at all conditions. L G R 7 levels in stromal cells of patients were slightly suppressed in the less than three month N H T group compared to the untreated group, but increased with duration of androgen ablation to the greater than 6 month androgen ablation group. L G R 7 levels were low in the stroma of androgen independent (Al ) tissue, especially bone metastases. In the cancerous epithelial cells of patients, L G R 7 was present in the untreated group, and subsequently declined slightly with increasing duration of androgen ablation, with the lowest point at 3-6 months treatment. The epithelial L G R 7 levels were higher at the greater than 6 month treatment group, and highest in the A l samples. The presence of L G R 7 in stroma and epithelia indicates potential autocrine and paracrine functions for relaxin in the prostate. We have previously shown that relaxin is upregulated in A l progression of prostate cancer (Thompson et al., 2006); therefore, we examined the effect of relaxin on prostate cancer growth using the L N C a P xenograft model in intact mice. The data demonstrated that the tumour take rate was higher in L N C a P - R L X cell line with 41% tumour take rate, compared to a 28% tumour take rate in the control group; however, the 83 2.5 P?1.5 O'J 1 H 0.5 H 3 <3M 3-6M >6M NHT Groups (months) A] 1 -E -Figure 3 .3 L G R 7 is expressed in malignant epithelium and stromal cells of prostate. The average scoring results of N H T array (human patients) stained for L G R 7 in stroma (light shading), and cancer cells (dark shading) a). Representative L G R 7 staining of b) 0, c) 3 months, d) 3-6 months, and e) 9 months N H T , and f) A l bone metastases. Error bars are standard error of the mean. X4 size of tumours in intact mice was not statistically larger in L N C a P - R L X tumours over control (Fig. 3.4A). To determine the effect of relaxin overexpression in an androgen depleted environment, L N C a P - R L X and L N C a P - G F P were inoculated s.c. into male nude mice that were subsequently castrated after six weeks. L N C a P - R L X xenografts had 45 fold (p<0.009, t-test) higher levels of relaxin m R N A than control, as determined by Q-PCR. T M A s were generated from paraffin embedded xenografts for protein expression and localization analyses, and labeled L N C a P - R L X - T M A - C . L N C a P - R L X xenografts from castrate mice did not exhibit higher relaxin protein levels than control, determined by IHC of L N C a P - R L X - T M A - C ; however, the staining intensity was higher than in the L N C a P - G F P samples from intact mice (Data not shown). This suggests that endogenous relaxin protein levels are elevated in castration, supporting R N A data in Chapter 2, and that the relaxin protein levels in the control xenografts are similar to those in L N C a P - R L X in castrated mice. A s expected, there was no change in L G R 7 between the 2 groups using IHC in L N C a P - R L X - T M A - C . Despite this, L N C a P - R L X tumours displayed an increased take rate of 83% compared to 67% in the L N C a P - G F P group. The tumours in the L N C a P - R L X group grew more rapidly than those in the L N C a P - G F P group by as early as 39 days post inoculation (Fig. 3.4B), at which point the mice had not been castrated. After castration at day 42, growth rates of L N C a P - R L X and control xenografts were approximately the same. At termination of the experiment, 21 days post castration, the average L N C a P - R L X tumour size was 172 m m 3 compared to control at 89 mm 3 . We anticipate that since endogenous relaxin is elevated by castration (Chapter 2: Thompson et al., 2006), and there is no difference in protein levels of relaxin 85 B time post inoculation (days) Figure 3.4 L N C a P - R L X xenograft tumours grow more rapidly than LNCaP-GFP xenografts in intact mice. Nude mice were inoculated with 2 x 10 6 L N C a P - R L X or L N C a P - G F P cells, and examined for tumor volume, a) L N C a P - R L X tumors tended to be larger than L N C a P - G F P xenografts in intact mice, though the difference was not significant, b) In a subsequent experiment, L N C a P - R L X xenografts grew faster than L N C a P - G F P prior to castration 42 days after inoculation, resulting in L N C a P - R L X xenografts being larger than L N C a P - G F P xenografts; however, tumour growth rates were similar after castration for both L N C a P - R L X and L N C a P - G F P . Error bars are standard error of the mean. 86 between L N C a P - G F P and L N C a P - R L X in the xenograft T M A s from castrated mice, the effects of overexpressed recombinant relaxin are less evident. Relaxin overexpressing tumours upregulate W n t l 1 by upregulating P C D H Y To investigate novel mechanisms by which relaxin overexpression might affect prostate tumours, R N A from L N C a P - R L X and L N C a P - G F P xenografts grown in intact mice was expression profiled using high density oligomer microarrays representing 21 329 genes. Assembling this data set into Ingenuity Pathway Analysis helped highlight biological pathways that might be regulated by relaxin in prostate tumours. One gene highly upregulated by relaxin was W n t l 1, which was upregulated an average 3.56 fold in 8 tumours. The results were validated using Q-PCR, and found to be upregulated 5.6 fold in the same set of tumours (Table 3.1). Two androgen regulated genes, P S A and IRS-2 (Chapter 4), were downregulated in L N C a P - R L X compared to L N C a P - G F P in microarray analysis, and validated using Q - P C R (Table 3.1). W n t l 1 levels are not highly upregulated in relaxin overexpressing tumours compared to L N C a P - G F P controls in castrated mice (Table 3.1), but this is likely due to W n t l 1 being upregulated by androgen withdrawal, a condition which also upregulates relaxin as we have previously shown (Thompson et al., 2006). W n t l 1 was also slightly upregulated in L N C a P - R L X cells grown in tissue culture, compared to L N C a P - G F P controls (Table 3.1). A s we found W n t l 1 to be upregulated by relaxin overexpression, and W n t l 1 is upregulated in the absence of androgens, we examined i f androgen withdrawal treatment in human patients resulted in increased W n t l 1 levels. Indeed W n t l 1 levels in A l metastases and 87 Table 3.1 Relaxin overexpression upregulates P C D H Y and W n t l l . Summary of results of relaxin overexression in vivo and in vitro. Results are fold change above L N C a P - G F P . cell line treatment P C D H Y Wnt11 IRS 2 L N C a P - G F P intact mouse 1.00 1.00 1.00 L N C a P - R L X intact mouse 1.83 5.60 0.58 L N C a P - G F P Cx Mouse 1.00 1.00 1.00 L N C a P - R L X Cx Mouse 2.63 1.23 0.84 L N C a P - G F P in vitro 1.00 1.00 1.00 L N C a P - R L X in vitro 3.64 1.32 0.50 C x : castrated samples from patients treated with N H T for greater than six months were significantly higher than in untreated patient samples (Fig. 3.5). Since W n t l 1 has been shown to be upregulated by P C D H Y (Yang et al., 2005), we investigated whether relaxin upregulated P C D H Y as well . B y Q - P C R we found that P C D H Y is upregulated in L N C a P - R L X cells in tissue culture over L N C a P - G F P , as well as in xenografts, in both castrated and intact mice. In intact mice, relaxin upregulates P C D H Y 1.83 fold (p<0.15, t-test) over control tumours, whereas in castrated mice, relaxin upregulates P C D H Y 2.63 fold (p<0.05, t- test) over control (Table 3.1). In tissue culture, L N C a P - R L X has 3.6 fold higher P C D H Y levels compared to L N C a P - G F P (Table 3.1). This suggests that relaxin upregulates P C D H Y in an autocrine fashion, independently of androgen withdrawal, and elevated P C D H Y in turn upregulates W n t l 1. Relaxin overexpression causes P-catenin to migrate to the cytoplasm P C D H Y expression causes /?-catenin to leave the cytoplasmic membrane and move to cytoplasm or nucleus (Yang et al., 2005). To determine i f the increase in P C D H Y expression in L N C a P - R L X cells resulted in altered /?-catenin localization, L N C a P - R L X - T M A - I and L N C a P - R L X - T M A - C were stained for y9-catenin and scored for subcellular localization (Fig. 3.6). There was no detectable change in /?-catenin localization in castrated mice, utilizing IHC. /J-catenin is located primarily at the cytoplasmic membrane in L N C a P - R L X and control xenografts from intact nude mice (Fig. 3.6A); however, L N C a P - R L X samples have 1.4 fold higher cytoplasmic /5-catenin levels than the controls, though the change is statistically significantly different (p<0.2, t-test) (Fig. 3.6B). These results indicate that relaxin upregulation causes increased 89 D 2.5 CO S 1.5 ? 1.0 o w 0.5 0.0 m l <3 3-6 >6M NHT Groups (months) Figure 3.5 W n t l l staining is increased after six months N H T , and sustained in A l metastases in human patient samples. Representative W n t l 1 I H C staining of the epithelia in patients receiving N H T for a) O mo, b) <3 mo, c) 3-6 mo, d) >6 mo, and in A l metastases in e) bone or f) liver, g) The average Wnt l 1 scoring results. Error bars are standard error of the mean. 90 q g i/i s 2 L N C i i P -R L X B 5 o •ca If) iS a. o >. O 1 6 1.2 0 8 0 4 L N C a P -R L X Figure 3.6 Relaxin overexpression causes p-catenin to translocate to the cytoplasm in intact mice, a) Membrane P-catenin staining is no different in L N C a P - R L X than L N C a P - G F P xenografts in intact mice (p-value >0.5, t-test). b) L N C a P - R L X xenografts have a higher proportion of cytoplasmic P-catenin than control xenografts in intact mice, though this is not statistically significant (p-value <0.2, t-test). 91 cytoplasmic P-catenin localization, likely due to an upregulation of P C D H Y in intact mice. 3.4 Discussion The aim of this work was to identify biological pathways regulated by relaxin in prostate cancer cells. Relaxin is a peptide hormone of diverse functions in many tissues. We show that relaxin overexpression enhances tumour growth in xenografts in intact mice, and illuminate a novel relaxin-regulated pathway involving P C D H Y and W n t l 1. Additionally the distribution of the relaxin receptor, L G R 7 , is characterized for the first time here in human prostate cancer patient samples. L N C a P - R L X xenografts in intact male nude mice exhibited no significant difference in tumour volume from control, although in castrated nude mice L N C a P - R L X were larger. This difference in tumour volume was apparent before castration of mice, indicating that relaxin enhances prostate cancer cell growth in vivo in the presence of androgens. Endogenous relaxin is upregulated in prostate cancer xenografts from castrated mice (Data not shown), and in patients following androgen ablation (Chapter 2: Thompson et al., 2006). This may explain why there is no difference in tumour growth rate after castration between L N C a P - R L X and L N C a P - G F P . In support of this, relaxin protein levels were similar in L N C a P - G F P and L N C a P - R L X groups when the tumours were harvested post-castration. Other groups have seen relaxin increase tumour volume through various mechanisms in prostate cancer, including increased angiogenesis (Silvertown et al., 92 2006), increased proliferation (Feng et al., 2007), and p53 mutation (Vinal l et al., 2006), and in other cancers, including mammary (Binder et al., 2002; Radestock et al., 2005), thyroid (Hombach-Klonisch et al., 2006), and endometrial cancers (Kamat et al., 2006). We sought, in an unbiased approach, novel mechanisms of relaxin action using microarray analysis. To see i f prostate cancer cells could themselves be responsive to relaxin, we examined L G R 7 levels and cell type distribution. L G R 7 is present in L N C a P xenografts in intact male nude mice, and has been detected in L N C a P cells in vitro (Vinall et al., 2006) and in PC3 cells in vivo (Silvertown et al., 2006). Xenograft results were validated in human prostate cancer samples. We characterize for the first time L G R 7 protein in malignant epithelial and stromal prostate cells, indicating the potential of paracrine and autocrine roles of relaxin in prostate cancer. Compared to epithelial relaxin expression, the increase of stromal L G R 7 is delayed in androgen ablated prostate cancer patients (Thompson et al., 2006). The epithelial cells maintain L G R 7 throughout androgen ablation with minor fluctuations. L G R 7 is almost exclusively localized to the malignant epithelial cells of A l bone metastases, which also express elevated levels of relaxin (Thompson et al., 2006). Taken together, these results indicate an autocrine role for relaxin in prostate cancer. Microarray analysis of L N C a P - R L X xenografts compared to control indicated upregulation of W n t l 1, a novel function for relaxin. Like relaxin, W n t l 1 has been reported to be upregulated by androgen ablation (Zhu et al., 2004a). W n t l 1 is negatively regulated by androgens in prostate cancer, and negatively regulates A R transcriptional activity in androgen dependent cells. This complex regulation is relieved, however, in 93 A l prostate cancer cells (Zhu et a l , 2004a). W n t l 1 is upregulated in prostate cancers with increased Gleason score (Zhu et al., 2004a), has been shown to transform and enhance migration of mammary (Christiansen et al., 1996) and intestinal epithelial cells (Ouko et al., 2004a). In some tissues, relaxin is capable of enhancing cell migration (Kamat et al., 2006; Silvertown et al., 2003), and W n t l 1 expression may be a novel mechanism by which relaxin mediates this effect. A s W n t l 1 is upregulated by P C D H Y (Yang et al., 2005), another gene upregulated in A l (Terry et al., 2006a), we sought to determine the hierarchy of this pathway. P C D H Y and W n t l 1 were both upregulated in L N C a P - R L X xenografts in the presence or absence of androgens, indicating that relaxin is likely upstream of both P C D H Y and W n t l 1. P C D H Y enables L N C a P growth in castrated hosts (Terry et a l , 2006a), suggesting a novel method by which relaxin may aid L N C a P grow faster than controls in castrated mice. In addition, L N C a P - R L X cells grown in vitro in full serum exhibit upregulated levels of P C D H Y and W n t l 1, further suggesting an autocrine mechanism of relaxin effect on L N C a P cells. P C D H Y is a protocadherin that lacks a signal sequence; therefore, it is localized to the cytoplasm (Chen et a l , 2002). P C D H Y shares homology with the protocadherins, and shares homology at its C-terminal with the P-catenin binding domain of E-cadherin. p-catenin and P C D H Y colocalize to the cytoplasm, permitting P-catenin mediated signalling (Chen et al., 2002). To confirm that P C D H Y mediated upregulation of W n t l 1, we assayed T M A s of L N C a P - R L X and L N C a P - G F P xenografts for p-catenin subcellular localization, and determined that P-catenin is driven to the cytoplasm in L N C a P - R L X , but not in L N C a P - G F P . The cytoplasmic localization of p-catenin by IHC 3 94 is indicative of P-catenin signalling (Chesire et al., 2000; Terry et al., 2006a). Our work supports previous results, indicating that the canonical p-catenin pathway is upregulated through upregulation of P C D H Y (Chen et a l , 2002; Terry et al., 2006a; Yang et al., 2005) . Wnt signalling has been postulated to be relevant in prostate cancer through a number of lines of evidene, with about 5% of cancers having P-catenin or adenomatous polyposis coli mutations (Chesire et al., 2000; Voeller et al., 1998). P-catenin has been studied in malignant mesothelioma (Fox and Dharmarajan, 2006), lung (Daniel et al., 2006) , multiple myeloma (Pearse, 2006), and colon cancer (Radtke and Clevers, 2005). In colon cancer, there are inactivating mutations in the proteins involved in P-catenin degradation, as well as activating mutations in p-catenin (Terry et al., 2006b). Very few activating mutations of P-catenin have been found in prostate cancer, although in some cases there is P-catenin accumulation without activating mutations in P-catenin itself (Chesire et a l , 2000). P-catenin is involved in cell-cell interactions through E-cadherin and a-catenin, is localized to the cytoplasmic membrane, and in the absence of Wnt signalling, cytosolic P-catenin is degraded. Upon Wnt signalling, P-catenin dissociates from the cytoplasmic membrane, localizes in the cytosol and ultimately enters the nucleus, to activate T C F / L E F 1 transcription factors (reviewed in Terry et al., 2006b). P-catenin interacts with the A R in a ligand - dependent fashion (Mulholland et al., 2002) which is competitive with T C F / L E F 1 interaction (Song et al., 2003). This interaction is ligand and A R dose dependent, and can overcome activating mutants of P-catenin (Chesire and Isaacs, 2002; Mulholland et al., 2003). However, in the absence of androgen, relaxin is 95 CRJ3T) - — — upregulates B e t a - c a t e n i n Birds to anc ca j ses dissociation from the -nt-v'Tibran©, resulting in upregulation of beta-caienm-responsive genes WNT11 Figure 3.7 Model of relaxin driving upregulation of P C D H Y , w i th p-catenin translocating to the cytoplasm, and downstream upregulat ion o f W n t l 1. 96 upregulated (Thompson et a l , 2006), upregulating P C D H Y , and this stabilizes P-catenin in the cytoplasm. In our model (Fig. 3.7), there is limited ligand for the A R , thus the competition between A R and T C F binding for P-catenin is low, with P-catenin binding T C F . This is evidenced by enhanced P-catenin activation seen in P C D H Y overexpressing cells (Terry et al., 2006a). To summarize, relaxin is negatively regulated by androgens, and is upregulated with progression to A l prostate cancer. Overexpression of relaxin results in upregulation of P C D H Y and W n t l 1. Relaxin overexpression results in p-catenin translocating from the cytoplasmic membrane to the cytoplasm. P C D H Y upregulates W n t l 1, likely through P-catenin signalling, in L N C a P cells (Yang et al., 2005). We add to the previous work to postulate that relaxin is driving this process by upregulating P C D H Y . Interestingly, relaxin, W n t l 1, and P C D H Y are negatively regulated by androgens in L N C a P cells. Relaxin overexpression is able to upregulate P C D H Y and W n t l 1 even in the presence of androgens, in tissue culture conditions and in xenografts, therefore relaxin is likely upstream of W n t l 1 and P C D H Y . 97 3.5 References Bani, D . 1997. Relaxin: a pleiotropic hormone. Gen Pharmacol. 28:13-22. Binder, C , T. Hagemann, B . Husen, M . Schulz, and A . Einspanier. 2002. Relaxin enhances in-vitro invasiveness of breast cancer cell lines by up-regulation of matrix metalloproteases. Mol Hum Reprod. 8:789-96. Chen, M . W . , F . Vacherot, A . De L a Taille, S. Gil-Diez-De-Medina, R. Shen, R . A . Friedman, M . Burchardt, D . K . Chopin, and R. Buttyan. 2002. The emergence of protocadherin-PC expression during the acquisition of apoptosis-resistance by prostate cancer cells. Oncogene. 21:7861-71. Chesire, D.R. , C M . Ewing, J. Sauvageot, G.S. Bova, and W . B . Isaacs. 2000. Detection and analysis of beta-catenin mutations in prostate cancer. Prostate. 45:323-34. Chesire, D.R. , and W . B . Isaacs. 2002. Ligand-dependent inhibition of beta-catenin/TCF signaling by androgen receptor. Oncogene. 21:8453-69. Christiansen, J .H. , S.J. Monkley, and B.J . Wainwright. 1996. Murine WNT11 is a secreted glycoprotein that morphologically transforms mammary epithelial cells. Oncogene. 12:2705-11. Clarke, N . W . 2006. Management of the spectrum of hormone refractory prostate cancer. Eur Urol. 50:428-38; discussion 438-9. Craft, N . , Y . Shostak, M . Carey, and C . L . Sawyers. 1999. A mechanism for hormone-independent prostate cancer through modulation of androgen receptor signaling by the HER-2/neu tyrosine kinase. Nat Med. 5:280-5. Daniel, V . C , C D . Peacock, and D . N . Watkins. 2006. Developmental signalling pathways in lung cancer. Respirology. 11:234-40. De Cooman, S., P. Gil l iaux, and K . Thomas. 1983. Immunoreactive relaxin-like substance in human split ejaculates. Fertil Steril. 39:111-3. de la Taille, A . , M . A . Rubin, M . - W . Chen, F. Vacherot, S.G.-D. de Medina, M . Burchardt, R. Buttyan, and D . Chopin. 2003. {betaj-Catenin-related Anomalies in Apoptosis-resistant and Hormone-refractory Prostate Cancer Cells. Clin Cancer Res. 9:1801-1807. Feng, S., I .U. Agoulnik, N . V . Bogatcheva, A . A . Kamat, B . Kwabi-Addo, R. L i , G . Ayala , M . M . Ittmann, and A . I . Agoulnik. 2007. Relaxin promotes prostate cancer progression. Clin Cancer Res. 13:1695-702. Fox, S., and A . Dharmarajan. 2006. W N T signaling in malignant mesothelioma. Front Biosci. 11:2106-12. Gleave, M . E . , J.T. Hsieh, H . C Wu, A . C von Eschenbach, and L . W . Chung. 1992. Serum prostate specific antigen levels in mice bearing human prostate L N C a P tumors are determined by tumor volume and endocrine and growth factors. Cancer Res. 52:1598-605. Gunnersen, J . M . , R.J . Crawford, and G .W. Tregear. 1995. Expression of the relaxin gene in rat tissues. Mol Cell Endocrinol. 110:55-64. Hombach-Klonisch, S., J. Bialek, B . Trojanowicz, E . Weber, H.J . Holzhausen, J.D. Silvertown, A . J . Summerlee, H . Dralle, C. Hoang-Vu, and T. Klonisch. 2006. 98 Relaxin enhances the oncogenic potential of human thyroid carcinoma cells. Am J Pathol. 169:617-32. Kamat, A . A . , S. Feng, I .U. Agoulnik, F. Kheradmand, N . V . Bogatcheva, D . Coffey, A . K . Sood, and A . I . Agoulnik. 2006. The role of relaxin in endometrial cancer. Cancer Biol Ther. 5:71-7. Kiyama, S., K . Morrison, T. Zellweger, M . Akbari , M . Cox, D . Y u , H . Miyake, and M . E . Gleave. 2003a. Castration-induced increases in insulin-like growth factor-binding protein 2 promotes proliferation of androgen-independent human prostate L N C a P tumors. Cancer Res. 63:3575-84. Kiyama, S., K . Morrison, T. Zellweger, M . Akbari , M . Cox, D . Y u , H . Miyake, and M . E . Gleave. 2003b. Castration-Induced Increases in Insulin-Like Growth Factor-Binding Protein 2 Promotes Proliferation of Androgen-independent Human Prostate L N C a P Tumors. Cancer Res. 63:3575-3584. Kojima, S., D.J . Mulholland, S. Ettinger, L . Fazli , C .C . Nelson, and M . E . Gleave. 2006. Differential regulation of IGFBP-3 by the androgen receptor in the lineage-related androgen-dependent L N C a P and androgen-independent C4-2 prostate cancer models. Prostate. 66:971-86. Mulholland, D.J . , H . Cheng, K . Reid, P.S. Rennie, and C .C . Nelson. 2002. The androgen receptor can promote beta-catenin nuclear translocation independently of adenomatous polyposis coli . J Biol Chem. 277:17933-43. Mulholland, D.J . , J.T. Read, P.S. Rennie, M . E . Cox, and C . C . Nelson. 2003. Functional localization and competition between the androgen receptor and T-cell factor for nuclear beta-catenin: a means for inhibition of the T c f signaling axis. Oncogene. 22:5602-13. Ouko, L . , T.R. Ziegler, L . H . Gu , L . M . Eisenberg, and V . W . Yang. 2004a. W n t l 1 signaling promotes proliferation, transformation, and migration of IEC6 intestinal epithelial cells. J Biol Chem. 279:26707-15. Ouko, L . , T.R. Ziegler, L . H . Gu , L . M . Eisenberg, and V . W . Yang. 2004b. W n t l 1 Signaling Promotes Proliferation, Transformation, and Migration of IEC6 Intestinal Epithelial Cells. J. Biol. Chem. 279:26707-26715. Pearse, R . N . 2006. Wnt antagonism in multiple myeloma: a potential cause of uncoupled bone remodeling. Clin Cancer Res. 12:6274s-6278s. Radestock, Y . , C. Hoang-Vu, and S. Hombach-Klonisch. 2005. Relaxin downregulates the calcium binding protein S100A4 in M D A - M B - 2 3 1 human breast cancer cells. Ann N YAcad Sci. 1041:462-9. Radtke, F. , and H . Clevers. 2005. Self-renewal and cancer of the gut: two sides of a coin. Science. 307:1904-9. Rennie, P.S., and C .C . Nelson. 1998. Epigenetic mechanisms for progression of prostate cancer. Cancer Metastasis Rev. 17:401-9. Schieferstein, G . , W . Voelter, H . Seeger, and T . H . Lippert. 1989. Immunoreactive relaxin in seminal plasma of man. IntJFertil. 34:215-8. Silvertown, J.D., B . J . Geddes, and A . J . Summerlee. 2003. Adenovirus-mediated expression of human prorelaxin promotes the invasive potential of canine mammary cancer cells. Endocrinology. 144:3683-91. 99 Silvertown, J.D., J. N g , T. Sato, A . J . Summerlee, and J .A. Medin. 2006. H2 relaxin overexpression increases in vivo prostate xenograft tumor growth and angiogenesis. Int J Cancer. 118:62-73. Silvertown, J.D., J.C. Symes, A . Neschadim, T. Nonaka, J .C .H . Kao, A.J .S . Summerlee, and J .A. Medin. 2007. Analog of H2 relaxin exhibits antagonistic properties and impairs prostate tumor growth. FASEB J. 21:754-765. Sokol, R .Z . , X . S . Wang, J. Lechago, P .D. Johnston, and R.S. Swerdloff. 1989. Immunohistochemical localization of relaxin in human prostate. J Histochem Cytochem. 37:1253-5. Song, L . N . , R. Herrell, S. Byers, S. Shah, E . M . Wilson, and E.P. Gelmann. 2003. Beta-catenin binds to the activation function 2 region of the androgen receptor and modulates the effects of the N-terminal domain and TIF2 on ligand-dependent transcription. Mol Cell Biol. 23:1674-87. Terry, S., L . Queires, S. Gil-Diez-de-Medina, M . W . Chen, A . de la Taille, Y . Al lory , P .L . Tran, C C . Abbou, R. Buttyan, and F. Vacherot. 2006a. Protocadherin-PC promotes androgen-independent prostate cancer cell growth. Prostate. 66:1100-13. Terry, S., X . Yang, M . W . Chen, F. Vacherot, and R. Buttyan. 2006b. Multifaceted interaction between the androgen and Wnt signaling pathways and the implication for prostate cancer. J Cell Biochem. 99:402-10. Thompson, V . C . , T .G . Morris, D.R. Cochrane, J. Cavanagh, L . A . Wafa, T. Hamilton, S. Wang, L . Fazl i , M . E . Gleave, and C C . Nelson. 2006. Relaxin becomes upregulated during prostate cancer progression to androgen independence and is negatively regulated by androgens. Prostate. 66:1698-709. Ulr ich , F., M . Krieg, E . M . Schotz, V . Link, I. Castanon, V . Schnabel, A . Taubenberger, D . Mueller, P . H . Puech, and C P . Heisenberg. 2005. W n t l 1 functions in gastrulation by controlling cell cohesion through Rab5c and E-cadherin. Dev Cell. 9:555-64. Vina l l , R . L . , C . G . Tepper, X . B . Shi, L . A . Xue, R. Gandour-Edwards, and R . W . de Vere White. 2006. The R273H p53 mutation can facilitate the androgen-independent growth of L N C a P by a mechanism that involves H2 relaxin and its cognate receptor L G R 7 . Oncogene. 25:2082-93. Voeller, H.J . , C.I. Truica, and E.P. Gelmann. 1998. Beta-catenin mutations in human prostate cancer. Cancer Res. 58:2520-3. Yang, X . , M . W . Chen, S. Terry, F. Vacherot, D . K . Chopin, D . L . Bemis, J. Kitajewski, M . C Benson, Y . Guo, and R. Buttyan. 2005. A human- and male-specific protocadherin that acts through the wnt signaling pathway to induce neuroendocrine transdifferentiation of prostate cancer cells. Cancer Res. 65:5263-71. Yki-Jarvinen, H . , T. Wahlstrom, a n d M . Seppala. 1983. Immunohistochemical demonstration of relaxin in the genital tract of men. J Reprod Fertil. 69:693-5. Zhu, H . , M . Mazor, Y . Kawano, M . M . Walker, H . Y . Leung, K . Armstrong, J. Waxman, and R . M . Kypta. 2004a. Analysis of Wnt gene expression in prostate cancer: mutual inhibition by WNT11 and the androgen receptor. Cancer Res. 64:7918-26. 100 Zhu, H . , M . Mazor, Y . Kawano, M . M . Walker, H Y . Leung, K . Armstrong, J. Waxman, and R . M . Kypta. 2004b. Analysis of Wnt Gene Expression in Prostate Cancer: Mutual Inhibition by WNT11 and the Androgen Receptor. Cancer Res. 64:7918-7926. 101 Chapter 4. Suppression of Androgen Signaling by IGF-I due to s iRNA Depletion of IRS-21 4.1 Introduction In men, prostate cancer is the most commonly diagnosed cancer and the third leading cause of cancer related deaths (Canada, 2007). Prostate cancer is dependent upon androgens for survival, and is commonly treated by androgen ablation, resulting in tumour regression. In approximately 80% of men this treatment eventually fails, resulting in androgen independent (Al ) prostate cancer, for which treatment options are limited (Rennie and Nelson, 1998). This is in part due to a lack of understanding of the mechanisms driving A l in prostate cancer cells. Therefore, new insights into these mechanisms are invaluable for developing new therapies for A l prostate cancer. One possible means by which prostate cancer cells may be able to proliferate in the absence of testicular androgens is by reactivation of the androgen receptor (AR) . There is evidence to suggest that A R signaling can be modulated by various growth factors that have been shown to activate the A R in the absence of androgens (Culig et al., 1994; reviewed in Feldman and Feldman, 2001), while other lines of evidence show that A version of this chapter w i l l be submitted for publication. Thompson, V . C . , Lehman, M . , Lubik, A . , Moore, S., Neira, M . , Altamirano, M . , Hendy,S., Nelson, C . C . Suppression of androgen signaling by IGF-I due to s i R N A depletion of IRS-2. 102 prostatic androgens may be responsible for reactivation of A R (Mostaghel et al., 2007; Titus et al., 2005). IGF-I binds to and activates the IGF-I receptor (IGF-IR), which is a receptor tyrosine kinase. Insulin receptor substrates, IRS-1 and IRS-2, dock with the ligand-bound IGF-IR, are phosphorylated, and activate PI-3 kinase, resulting in Ak t activation. The downstream effects of Akt activation are increased proliferation, and inhibition of apoptosis and differentiation (reviewed in Meinbach and Lokeshwar, 2006). A n additional pathway activated by IGF-IR is the M A P K pathway, with an end result of cell proliferation (reviewed in Krueckl et al., 2004; Meinbach and Lokeshwar, 2006). In mice, IRS-1 and IRS-2 are transforming in breast (Dearth et al., 2006), while overexpression of IGF-I in basal cells results in prostate cancer (DiGiovanni et al., 2000) . IGF-I and IGF-IR are elevated in A l prostate cancer models (Nickerson et al., 2001) . Increased plasma levels of IGF-I in humans result in an increased risk of prostate cancer (Chan et al., 1998). This indicates that IGF-I is capable of transforming the prostate, and may enable prostate cells to survive and proliferate in the absence of androgens. IGF-I has been shown to alter A R signaling, although the results are somewhat controversial; in some cases IGF-I increases androgen receptor activation (Culig et al., 1994), and in other cases represses androgen receptor activity (Lin et al., 2003; L i n et al., 2001; Plymate et a l , 2004). We found that IRS-2 m R N A is strongly upregulated by androgens in L N C a P cells, and that the treatment of these cells with IGF-I can lower IRS-2 m R N A levels. We knocked down IRS-2 using s i R N A as a means of exploring the interaction between the IGF axis and androgen signaling. IRS-2 knockdown led to 103 altered regulation of androgen responsive genes, prostate specific antigen (PSA), insulin degrading enzyme (IDE), vascular endothelial growth factor ( V E G F ) , and clusterin, indicating that in the presence of IGF-I treatment IRS-2 is required for androgen regulation of these gene targets. 4.2 Materials and methods Cel l Culture and Media Cell Culture - L N C a P cells, passage 35-50, were maintained in R P M I 1640 with 5% fetal bovine serum (FBS) and penicillin/streptomycin at 37 °C and 5% CO2. Hormone Treatments - A t approximately 60% confluency, L N C a P cells received R P M I + 2% charcoal stripped serum (CSS) (Hyclone, Logan, U T , U S A ) for 48 h prior to 1 n M R1881 treatment or ethanol control. For microarray analysis, cells were treated with R1881 for 72 h. For IGF-I treatment, cells were treated +/- R1881 for 72 h prior to serum free (SF) media for 2 h, followed by 16 h 50 ng/ml IGF-I. IRS2 siRNA experiments - Initial s i R N A treatment dose curve included 5, 10, 25 and 50 n M s i R N A in F B S ; 5 n M s i R N A was used in further experiments. For studies on effect of IGF-I and R1881 in IRS-2 s i R N A treated cells, cells were treated with R1881 for 48 h prior to treatment of negative control s i R N A or IRS-2 s i R N A (Ambion, Austin, T X ) . using a 1:100 ratio of Lipofectamine 2000 (Invitrogen, Carlsbad, C A , U S A ) to SF media for 4 hours, followed by replacement CSS and R1881. 24 h later, 50 ng/ml IGF-I was added for 16 h incubation without SF media pretreatment. 104 AR siRNA experiments - experiments were conducted in the presence and absence of 1 n M R1881as described for IRS2 s i R N A , except no IGF-I treatment was given, and cells were harvested 48 hours post s i R N A addition. A s well , 25 n M A R s i R N A was used. Preparation of total R N A L N C a P cells were lysed in Trizol (Invitrogen) and total R N A isolated according to manufacturer's protocol; R N A was quantified using the ND-1000 Spectrophotometer (Nanodrop Technologies, Wilmington, D E , U S A ) . For microarray analysis, total R N A concentration and quality were evaluated using the 2100 Bioanalyser (Agilent, Santa Clara, C A , U S A ) . Microarray Analysis Total R N A was treated for hybridization to microarrays using the 3 D N A Array 350 kit (Genisphere, Hatfield, P A , U S A ) , following manufacturer's instructions, and as described previously (Kojima et al., 2006). Briefly, c D N A was synthesized from 10 ug total R N A with Superscript II, then concentrated using a Microcon Y M - 3 0 Centrifugal Filter (Millipore, Billerica, M A , U S A ) . Samples, in triplicate, were labelled with Cyanine 5 dUTP, with Universal Human Reference R N A (Strategene, L a Jolla, C A , U S A ) as normalization control in the Cyanine 3 dUTP channel. c D N A in formamide-based hybridization buffer was hybridized to microarrays (Array Facility of The Prostate Centre at Vancouver General Hospital, Vancouver, Canada) spotted in duplicate with 21 329 70mer probe Human Operon Version 2.0 library (Operon, Huntsville, A L , U S A ) . 3 D N A Array 350 Capture Reagent was 105 hybridized in formamide buffer to labelled c D N A previously hybridized and microarrays scanned on a Scan Array Express (Perkin Elmer, Wellesley, M A , U S A ) . ImaGene 4.2 (BioDiscovery, EI Segundo, C A , U S A ) software was used for microarray spot intensity quantification; GeneSpring 4.2 (Agilent Technologies, Palo Alto , C A , U S A ) was used for normalization, and further analysis utilized Ingenuity Pathways Analysis (Ingenuity Systems, Redwood City, C A , U S A ) . Additional analysis was performed, including preliminary quality analysis of the microarray hybridizations using in-house quality scripts written in R, based on Bioconductor packages 'arrayQuality' and 'Limma'. Normalization and statistical analysis of differential expression across different conditions was performed using linear models for microarrays implemented in 'Limma' (Smyth, 2005). Semi-Quantitative Real Time - P C R 2 ug total L N C a P R N A was DNase I (Invitrogen) treated, then reverse transcribed in I X M - M L V first strand synthesis buffer, 0.01 M D T T , 1 m M dNTPs, 6.67 U / u l M - M L V reverse transcriptase (all from Invitrogen), 1.33 U/u l R N A s i n (Promega, Madison, WI , U S A ) , and 8.33 U / u l random hexamers. The semi-quantitative real time - P C R (Q-PCR) mix contained c D N A from the reverse transcription reaction, I X TaqMan mix (Applied Biosystems, Foster City, C A , U S A ) , and 0.9 u M forward P S A primer, 5' - T C T G C G G C G G T G T T C T G - 3', 0.9 u M reverse P S A primer, 5' - G C C G A C C C A G C A A G - 3', and 0.2 u M P S A Taqman probe, 5' - F A M - C T A C C C A C T G C A T C A G G A A C A A - T A M R A - 3', or 0.9 u M forward clusterin primer, 5' - G A G C A G C T G A A C G A G C A G T T T - 3' , 0.9 u M reverse clusterin 106 primer, 5' - C T T C G C C C T T G C G T G A G G T - 3' , and 0.2 u M clusterin Taqman probe, 5' - ( 6 - F A M ) A C T G G G T G T C C C G G C T G G C A - ( T A M R A - Q ) - 3 ' (all from Operon), IRS-2, V E G F , or IDE primer and probe mixes, (Applied Biosystems), and 18S r R N A primer and probe mix as the endogenous control (Applied Biosystems). Transcript levels were detected using the A B I 7900 H T Sequence Detection System (Applied Biosystems). Statistics Error bars represent standard error of the mean. Data from the A R s i R N A experiment were analysed with a one-way A N O V A , followed by Tukey's multiple comparison test. Data from the IRS-2 s i R N A dose curve experiment were log transformed prior to analysis by one way A N O V A , followed by Dunnett's multiple comparison test. Data from the IRS-2 s i R N A plus IGF-I experiment were log transformed prior to analysis by Student's t-test with Benjamin and Hochberg multiple test correction. 4.3 Results IRS-2 is upregulated by androgens, and decreased by IGF-I To identify genes strongly upregulated by androgens, we compared expression profiles of L N C a P cells treated in vitro with I n M R1881 for 72 hours with ethanol control treated LNCaPs , using high density oligo microarrays representing 21 329 genes. We found that IRS-2 is consistently upregulated 12-fold by androgens, regardless of any additional treatment of LNCaPs , using microarray analysis (data not shown). R1881 107 consistently induced IRS-2 levels as much as 3.6 fold in L N C a P s using Q - P C R in several experiments (Figs. 4.1 and 4.4A). When L N C a P cells are grown in serum free (SF) conditions for 2h prior to 16 hour IGF-I treatment in SF media, there is an approximately 50% reduction in IRS-2 levels regardless of whether cells are co-treated with androgen or not; however, IRS-2 levels are almost 100% higher when cells are treated with R1881 and IGF-I than when treated with IGF-I alone (Fig. 4.1). The decrease in IRS-2 in response to IGF-I treatment has been characterized at the protein level, as IGF-I treatment increases proteasomal degradation of IRS-2 (Briaud et al., 2005; K i m et al., 2005; More l l i et al., 2003; Rui et a l , 2001); however, the effect of IGF-I on IRS-2 levels in the prostate has not been explored. To further characterize the androgen mediated change in IRS-2, we used s i R N A to knock down the A R (Fig. 4.2). In these R1881 treated cells, IRS-2 levels were reduced by approximately 30% compared to negative control treated cells. The A R levels were reduced by more than 50%, as was the hallmark androgen regulated gene, P S A . This further indicates that IRS-2 is regulated by androgens, and this regulation is mediated through the A R . 108 3 2.5 2 1 5 o •4—' UJ an + GO 00 Si-Figure 4.1 IGF-I treatment reduces IRS-2 mRNA levels. L N C a P cells were treated with or without 1 n M R1881 in 2% CSS for 80h, serum starved for 2 h, then treated with 50 ng/ml IGF-I. Q - P C R results are fold change in m R N A compared to ethanol control, and representative of at least 3 experiments. Error bars are standard error of the mean. 109 Lipofectamine control Negative siRNA ARs iRNA Figure 4.2 A R s i R N A knocks down A R m R N A and abrogates androgen mediated transcription of P S A and IRS-2 . L N C a P cells were treated with 1 n M R1881 in 2% CSS for 48h, prior to treatment with 25 n M A R s i R N A or negative s i R N A . A R expression: dark shading; P S A expression: hatch marks; IRS-2 expression: light shading. Q - P C R results are fold change in m R N A compared to R1881 treated lipofectamine control, and representative of at least 3 experiments. * pO.001 compared to negative s i R N A or liopfectamine control, ** p<0.05 compared to negative s i R N A or liopfectamine control, f p<0.05 compared to lipofectamine control ( A N O V A and Tukey's Multiple Comparison Test). Error bars are standard error of the mean. 110 Decreasing IRS-2 levels alters IGF-I and androgen regulation of androgen responsive genes Since IGF-I, which signals through IRS-2, has been reported to activate the A R , and we demonstrated that IRS-2 is upregulated by androgens in L N C a P cells, we wanted to examine the effect of IRS-2 s i R N A treatment on androgen regulated gene expression in L N C a P cells. We first determined that 5 n M s i R N A was the minimal concentration suitable for knocking down IRS-2 m R N A ; IRS-2 m R N A was reduced to 39% of negative control levels, using Q - P C R (Fig. 4.3). Compared to negative s i R N A treated cells, IRS-2 s i R N A reduces IRS-2 levels to 28% with R1881 treatment alone, and to 19%o in R1881 and IGF-I co-treated cells. IRS-2 is decreased by IRS-2 s i R N A in all treatment conditions other than control, compared to respective treatment conditions in negative s i R N A . However, IRS-2 levels do not drop below ethanol control treated levels seen in negative s i R N A cells except when IGF-I is added to IRS-2 s i R N A treated cells (Fig. 4.4A). Treating L N C a P cells with negative control s i R N A decreased the magnitude of androgen-mediated upregulation of IRS-2, likely due to an off-target effect of the s i R N A which in part decrease A R m R N A levels (Fig. 4.2). IGF-I alone decreases IRS-2 levels in cells treated with IRS-2 s i R N A (Fig 4.4A). In R1881 treated cells transfected with s i R N A , IRS-2 levels are decreased slightly by IGF-I . These results likely differ from those seen in Figure 4.1 since cells treated with s i R N A were not serum starved prior to treatment with IGF-I. I l l 2 CN CO CC CD cn a as sz o .0 5 10 nM siRNA 25 50 F i g u r e 4.3 I R S - 2 s i R N A dose c u r v e . Effect of IRS-2 s i R N A to knock down IRS-2 m R N A was compared to negative s i R N A at 5, 10, 25, and 50 n M s i R N A , in 5% F B S -R P M I . Q - P C R results are fold change in m R N A compared to lipofectamine control (0 n M s iRNA) , when treated with negative s i R N A (light shading), or IRS-2 s i R N A (dark shading). Results are representative of experiment performed in triplicate. 112 siRNA Neg IRS-2 s iRNA Neg IRS-2 R1B81 IGF-I + + + + + + - + + + + - + 6 5 o 4 ce OT £ 2 I 0 40 35 30 2 25 < 20 Q _ 15 10 5 0 14 a 1 2 ^ 1 0 CD 8 t o = 6 o H 4 S 2 0 f r rf * ** * i — ; —C n trr. * A n , 1*1 Pi (1 n t r n w n L _ r—i I—I ft T * * i-n ' **' f •** * i t l 1 -ft n n PI r*i R1881 IGF-I 5 D 4.5 4 3.5 9 2 1.5 1 0.5 0 3.5 3 0 2 . 5 2! 2 ° 1 5 a; 0.5 0 + + + - + + + + - + + + + - + ] : fr-it :p:*=ti=op l*L k~ j r -j Figure 4.4 IRS-2 is required for androgen mediated transcription in the presence of R1881 and I G F - I . L N C a P cells were treated with or without 1 n M R1881 in 2% CSS for 48h, then treated with 5 n M IRS-2, negative s i R N A , or lipofectamine control for 24 h, then treated with 50 ng/ml IGF-I for 16 h. Q - P C R results are fold change in m R N A compared to ethanol control treated lipofectamine control, and representative of at least 3 experiments. IRS-2 s i R N A decreases A ) IRS-2 m R N A , and several A R -regulated genes, including B) P S A , C) clusterin, D) IDE, and E) V E G F m R N A in the presence of R1881 and IGF-I. 113 To explore the effects of IRS-2 on androgen signaling in L N C a P cells, we examined P S A and total clusterin, two androgen-regulated genes, in IRS-2 s i R N A treated cells using Q - P C R (Fig. 4.4 B and C). P S A and total clusterin levels were increased by 1 n M R1881 as expected (Cochrane et al., 2007). P S A and clusterin leVels were not significantly affected by IGF-I treatment except when cells were treated with IRS-2 s i R N A and R1881, which caused a very large decrease in P S A and clusterin m R N A . The effect of IGF-I on P S A expression agrees with others who have shown that IGF-I does not affect the androgen mediated increase of P S A in L N C a P cells (Culig et al., 1994). In the negative control s i R N A treatment group, P S A m R N A levels are not as strongly upregulated by androgens. This is similar to the effect of negative s i R N A treatment on IRS-2 levels which we have shown result in a slight decrease of A R levels (Fig. 4.2). However, treatment with negative s i R N A causes an increase in total clusterin m R N A levels compared to treatment with lipofectamine only. The decrease of P S A and total clusterin in IRS-2 s i R N A treated cells due to IGF-I and R1881 indicates that IRS-2 is required for androgen signaling in the presence of IGF-I. We have also identified in our microarray and expression studies herein that IDE and V E G F are both upregulated by R1881 or IGF-I alone, with R1881 causing a much larger increase than IGF-I in untreated cells. V E G F is decreased by IGF-I alone in IRS-2 s i R N A treated cells. In control treated cells, V E G F is upregulated by R1881 regardless of IGF-I treatment. In summary, all of these androgen responsive genes are downregulated by IGF-I and R l 881 with knockdown of IRS-2 with s i R N A . IGF-I signaling is not simply abrogated by decreased IRS-2, but is actually negatively affecting AR-regulated gene expression, as decreased IRS-2 results in decreased androgen activity 114 in the presence of IGF-I. This implies that IRS-2 is required for optimal androgen-mediated transcription in the presence of IGF-I. Effect of IRS-2 s i R N A on downstream signaling molecules To determine that androgen signaling was affected by IRS-2, and not merely by loss of A R or IGF-IR, we looked at protein levels of IRS-2, A R and IGF-IR. IRS-2 is undetectable in any samples, (Fig. 4.5 A ) likely due to low levels of IRS-2 protein. To ensure that the s i R N A treatment did not affect the IGF-IR, we examined IGF-IR protein levels. IGF-IR is present in all conditions, indicating that the lack of androgen signaling in the IRS-2 s i R N A group treated with R1881 and IGF-I is not due to a lack of IGF-IR (Fig. 4.5A). Likewise, we examined A R levels, and found that A R is much lower in IRS-2 s i R N A treated cells treated with R1881in the presence and absence of IGF-I (Fig. 4.5 A ) . However, the A R levels appear to be similar in R1881 or R1881 and IGF-I treated cells. The androgen regulated genes had very different expression levels in these two groups, with co- treatment of IGF-I and R1881 greatly decreasing levels of androgen responsive genes, P S A , V E G F , clusterin, and IDE, compared to R1881 treatment alone. This indicates that while A R levels are lower in this condition, this is not the source of decreased androgen signaling in the presence of R1881 and IGF-I . To determine i f signaling was occurring through IRS-2 in response to androgen treatment, we performed an immunoassay to detect downstream signaling molecules. Akt , p38, and E R K are present in all samples (Fig. 4.5B), indicating that these proteins would still be capable of transmitting IGF-I signal. This indicates that IRS-2 downregulation may have a role in the IGF-I mediated inhibition of androgen signaling. 115 siRNA Neg IRS-2 A R1881 - - + + - - + + . . + + IGF-I - + - + - + - + - + - + AR » T — ff, — 1 — - — * — * * " ' vinculin — —.«•»< FJ IGF-IR beta _ SiRNA Neg _j_RS-2 R1881 + + - - + + - - + + IGF-I - + - + - + - + - + . + Akt p38 • vinculin • E R K mm vinculin _; Figure 4.5 Effect of IRS-2 s i R N A on A R and I G F signaling molecules. Western blots of 50 u.g total protein indicate IRS-2, A R , and IGF-IR are expressed in all conditions A ) , as are phospho-IRS-2 and downstream signaling molecules, Akt , p38 and E R K B) . Under each blot vinculin loading control is shown. Blots were stripped and reprobed for vinculin. 116 4.4 Discussion In this study, we show for the first time that androgens upregulate IRS-2 m R N A in prostate cancer cells. IRS-1 has been shown to be upregulated by testosterone in preadipocyte and myoblast cell lines (Chen et al., 2006); however, this is the first indication that IRS-2 is upregulated by androgens. Previously it has been shown that progesterone upregulates IRS-2 (Cui et al., 2003; Vassen et al., 1999), and estrogen upregulates IRS-1 (Lee et al., 1999; Oesterreich et al., 2001) in breast cancer models. IGF-IR has also been shown to be upregulated by androgens (Krueckl et al., 2004) suggesting androgens likely accentuate IGF-I signalling through multiple mechanisms. IRS-2 regulation, though, is complex; IRS-2 m R N A is downregulated by IGF-I in SF media (Fig. 4.1), but not in CSS media (Fig 4.3A). In other studies it has been shown that E G F blocks the IGF-I mediated proteasomal degradation of IRS-1 in prostate (Zhang et al., 2000), but m R N A levels were not examined. IRS-2 is downregulated by IGF-I treatment in serum starved cells by proteasomal degradation, but transcriptional regulation was not found (Briaud et a l , 2005; K i m et al., 2005; Rui et al., 2001). It is possible that using Q - P C R to detect m R N A changes is more sensitive than previous methods used or that the cell culture context results in differential observations. While there is no evidence of IRS-2 m R N A regulation by IGF-I , there is evidence that IRS-2 m R N A is downregulated by insulin. Interestingly, desensitization by insulin, through prolonged insulin treatment, results in IGF-I desensitization, and vice versa. There is no desensitization to IR by platelet derived growth factor or F B S . This may be due in part to heterodimerization between the insulin receptor (IR) and IGF-IR. However, elevated insulin levels decrease IRS-2 m R N A and protein (Ide et al., 2004; 117 Pirola et al., 2003; Shimomura et al., 2000; Zhang et a l , 2001). This decrease in IRS-2 transcription is mediated through the Akt /PI3K pathway in liver cells (Hirashima et al., 2003), and L6 muscle cells (Pirola et a l , 2003). P T E N overexpression or PI3K inhibition in the P T E N null breast cell line, M D A - M B - 4 6 8 , resulted in increased IRS-2 R N A and protein (Simpson et al., 2001). Therefore, it is possible that the observed increased IRS-2 bound to the p85 PI3K subunit in the presence of PI3K inhibitor in L N C a P cells is due to increased total IRS-2 levels (Krueckl et al., 2004). The mechanism of IRS-2 transcriptional regulation has been further defined through knockout studies in mice. In IRS-2-/- mice, insulin effects are attenuated due to loss of IRS-2, but sterol response element-binding protein-lc ( S R E B P - l c ) is upregulated, as in wi ld type mice (Shimomura et al., 2000). S R E B P s bind to the sterol response element (SRE) and inhibit IRS-2 transcription by inhibiting co-activator binding to the promoter (Zhang et al., 2001). F O X O l and F O X 0 3 a increase IRS-2 (Ide et al., 2004) synergistically with TFE3 , a b H L H protein and transcription factor, through an E-box in the insulin response element (Nakagawa et al., 2006). SRC-3 , an A R coactivator (Wu et al., 2004; Zhou et al., 2005), and AP-1 also positively regulate IRS-2 transcription in the PC3 prostate cancer cell line (Yan et al., 2006). The effect of IGF-I treatment on prostate cancer cells has been widely studied. Increased IGF-I levels have been correlated with increased risk of prostate cancer (Chan et al., 1998), and IGF-I is elevated in prostate cancer patients over controls (Scorilas et al., 2003). IGF-I has been seen to augment the activity of the A R in the presence or absence of androgens (Culig et al., 1994). The increase in P S A m R N A can be used to measure A R transactivation in LNCaPs , and in the present study we found IGF-I in the 118 presence of CSS to have little to no effect on P S A levels. While the lack of effect of IGF-I alone on P S A levels agrees with previous results in L N C a P s (Culig et al., 1994), the R1881 and IGF-I result differs from the same study (Culig et a l , 1994), which found that IGF-I and 5 n M R1881 increased A R transactivation. There are several reasons why the results of this study differ from the results in L N C a P in the previous study, including duration and concentration of R1881 and IGF-I treatments. It is possible that increased duration of IGF-I treatment is required for AR-stimulating effects. A n additional study in L N C a P s has shown no difference in androgen response by IGF-I treatment in an androgen responsive luciferase reporter assay (Krueckl et al., 2004). The endogenous androgen regulated genes, P S A , V E G F , and clusterin, are upregulated by R1881 when IRS-2 levels are depressed by IRS-2 s i R N A , but expression is severely depressed when co-treated with R1881 and IGF-I. This is not due to a decrease in A R or IGF-IR, since there is no difference in these proteins between R1881 alone or R1881 and IGF-I treatment when IRS-2 s i R N A is knocked down. These results demonstrate that IRS-2 is required for androgen mediated gene expression in the presence of IGF-I in L N C a P cells. Knocking down IRS-2 in L N C a P cells should diminish IGF-I activation of the Ak t signaling pathway, because there is no IRS-1 in L N C a P s (Krueckl et al., 2004; Nickerson et al., 2001). The M A P K response to IGF-I in L N C a P cells has been described as minimally present (Krueckl et al., 2004; Rochester et al., 2005). M A P K is activated by IRS-2 signalling in L 6 cells (Zhang et al., 2001), and could play some role in LNCaPs . Akt , p38, and E R K are present in all conditions, indicating that signaling could occur through any of these molecules, so any changes due to IRS-2 s i R N A are 119 likely through depletion of IRS-2 and not downstream signaling molecules. Phosphorylation studies at short durations, up to 30 minutes, post IGF-I addition wi l l have to be conducted to determine signaling pathways involved. Perhaps the Erk signaling pathway is negatively affecting A R transactivation when IRS-2 is downregulated at the m R N A level. The expected result of knocking down IRS-2 would be to abrogate the signaling cascade of IGF-I. However, the results herein show a marked inhibitory effect of IGF-I on androgen regulated gene expression when IRS-2 is knocked down. It is difficult to directly compare the effect of IRS-2 knockdown to other studies that disrupt the IGF axis, such as IGF-IR inhibition or knock down (Rochester et al., 2005), as the end points in these two studies were different. The IGF-IR knockdown study suggests that IGF-IR signaling is necessary for phosphorylated Akt . Perhaps in our study, the treatment with IRS-2 s i R N A inhibits the phosphorylation of Akt , resulting in altered androgen signalling. Regardless of the mechanism involved, we have shown that IRS-2, an androgen regulated gene in prostate cancer cells, is required for androgen signaling in the presence of IGF-I. These studies indicate that targeting IRS-2 in prostate cancer patients may result in decrease androgen receptor function and IGF signaling. 120 4.5 References Briaud, I., L . M . Dickson, M . K . Lingohr, J.F. McCuaig , J.C. Lawrence, and C.J . Rhodes. 2005. Insulin receptor substrate-2 proteasomal degradation mediated by a mammalian target of rapamycin (mTOR)-induced negative feedback down-regulates protein kinase B-mediated signaling pathway in beta-cells. J Biol Chem. 280:2282-93. Canada, N.C.I .o. 2007. National Cancer Institute of Canada: Canadian Cancer Statistics 2007, Toronto, Canada. Chan, J . M . , M . J . Stampfer, E . Giovannucci, P . H . Gann, J. M a , P. Wilkinson, C . H . Hennekens, and M . Pollak. 1998. Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science. 279:563-6. Chen, X . , X . L i , H . Y . Huang, and J.F. L i n . 2006. [Effects of testosterone on insulin receptor substrate-1 and glucose transporter 4 expression in cells sensitive to insulin]. Zhonghua YiXue Za Zhi. 86:1474-7. Cochrane, D.R. , Z . Wang, M . Muramaki, M . E . Gleave, and C . C . Nelson. 2007. Differential regulation of clusterin and its isoforms by androgens in prostate cells. JBiol Chem. 282:2278-87. Cu i , X . , Z . Lazard, P. Zhang, T .A . Hopp, and A . V . Lee. 2003. Progesterone crosstalks with insulin-like growth factor signaling in breast cancer cells via induction of insulin receptor substrate-2. Oncogene. 22:6937-41. Cul ig , Z . , A . Hobisch, M . V . Cronauer, C. Radmayr, J. Trapman, A . Hittmair, G . Bartsch, and H . Klocker. 1994. Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. Cancer Res. 54:5474-8. Dearth, R . K . , X . Cu i , H.J . K i m , I. Kuiatse, N . A . Lawrence, X . Zhang, J. Divisova, O .L . Britton, S. Mohsin, D . C . Allred, D . L . Hadsell, and A . V . Lee. 2006. Mammary tumorigenesis and metastasis caused by overexpression of insulin receptor substrate 1 (IRS-1) or IRS-2. Mol Cell Biol. 26:9302-14. DiGiovanni , J., K . Kiguchi , A . Frijhoff, E . Wilker, D . K . B o l , L . Beltran, S. Moats, A . Ramirez, J. Jorcano, and C. Conti. 2000. Deregulated expression of insulin-like growth factor 1 in prostate epithelium leads to neoplasia in transgenic mice. Proc Natl Acad Sci USA. 97:3455-60. Feldman, B . J . , and D . Feldman. 2001. The development of androgen-independent prostate cancer. Nat Rev Cancer. 1:34-45. Hirashima, Y . , K . Tsuruzoe, S. Kodama, M . Igata, T. Toyonaga, K . Uek i , C . Kahn, and E. Araki . 2003. Insulin down-regulates insulin receptor substrate-2 expression through the phosphatidylinositol 3-kinase/Akt pathway. J Endocrinol. 179:253-266. Ide, T., H . Shimano, N . Yahagi, T. Matsuzaka, M . Nakakuki, T. Yamamoto, Y . Nakagawa, A . Takahashi, H . Suzuki, H . Sone, H . Toyoshima, A . Fukamizu, and N . Yamada. 2004. SREBPs suppress IRS-2-mediated insulin signalling in the liver. Nat Cell Biol. 6:351-7. K i m , B . , C M . van Golen, and E X . Feldman. 2005. Insulin-like growth factor I induces preferential degradation of insulin receptor substrate-2 through the 121 phosphatidylinositol 3-kinase pathway in human neuroblastoma cells. Endocrinology. 146:5350-7. ' Kojima, S., D.J . Mulholland, S. Ettinger, L . Fazli , C C . Nelson, and M . E . Gleave. 2006. Differential regulation of IGFBP-3 by the androgen receptor in the lineage-related androgen-dependent L N C a P and androgen-independent C4-2 prostate cancer models. Prostate. 66:971-86. Krueckl , S.L., R . A . Sikes, N . M . Edlund, R . H . Be l l , A . Hurtado-Coll, L . Fazl i , M . E . Gleave, and M . E . Cox. 2004. Increased insulin-like growth factor I receptor expression and signaling are components of androgen-independent progression in a lineage-derived prostate cancer progression model. Cancer Res. 64:8620-9. Lee, A . V . , J .G. Jackson, J .L. Gooch, S.G. Hilsenbeck, E . Coronado-Heinsohn, C . K . Osborne, and D . Yee. 1999. Enhancement of insulin-like growth factor signaling in human breast cancer: estrogen regulation of insulin receptor substrate-1 expression in vitro and in vivo. Mol Endocrinol. 13:787-96. L i n , H . K . , Y . C Hu, L . Yang, S. Altuwaijri, Y . T . Chen, H . Y . Kang, and C. Chang. 2003. Suppression versus induction of androgen receptor functions by the phosphatidylinositol 3-kinase/Akt pathway in prostate cancer L N C a P cells with different passage numbers. J Biol Chem. 278:50902-7. L i n , H . K . , S. Yeh , H . Y . Kang, and C. Chang. 2001. Akt suppresses androgen-induced apoptosis by phosphorylating and inhibiting androgen receptor. Proc Natl Acad Sci USA. 98:7200-5. Meinbach, D.S. , and B . L . Lokeshwar. 2006. Insulin-like growth factors and their binding proteins in prostate cancer: cause or consequence? Urol Oncol. 24:294-306. More l l i , C , C. Garofalo, M . Bartucci, and E . Surmacz. 2003. Estrogen receptor-alpha regulates the degradation of insulin receptor substrates 1 and 2 in breast cancer cells. Oncogene. 22:4007-16. Mostaghel, E . A . , S.T. Page, D . W . L i n , L . Fazli , I . M . Coleman, L . D . True, B . Knudsen, D . L . Hess, C C Nelson, A . M . Matsumoto, W.J . Bremner, M . E . Gleave, and P.S. Nelson. 2007. Intraprostatic androgens and androgen-regulated gene expression persist after testosterone suppression: therapeutic implications for castration-resistant prostate cancer. Cancer Res. 67:5033-41. Nakagawa, Y . , H . Shimano, T. Yoshikawa, T. Ide, M . Tamura, M . Furusawa, T. Yamamoto, N . Inoue, T. Matsuzaka, A . Takahashi, A . H . Hasty, H . Suzuki, H . Sone, H . Toyoshima, N . Yahagi, and N . Yamada. 2006. TFE3 transcriptionally activates hepatic IRS-2, participates in insulin signaling and ameliorates diabetes. Nat Med. 12:107-13. Nickerson, T., F. Chang, D. Lorimer, S.P. Smeekens, C . L . Sawyers, and M . Pollak. 2001. In vivo progression of L A P C - 9 and L N C a P prostate cancer models to androgen independence is associated with increased expression of insulin-like growth factor I (IGF-I) and IGF-I receptor (IGF-IR). Cancer Res. 61:6276-80. Oesterreich, S., P. Zhang, R . L . Guler, X . Sun, E . M . Curran, W . V . Welshons, C . K . Osborne, and A . V . Lee. 2001. Re-expression of estrogen receptor alpha in estrogen receptor alpha-negative M C F - 7 cells restores both estrogen and insulin-like growth factor-mediated signaling and growth. Cancer Res. 61:5771-7. Pirola, L . , S. Bonnafous, A . M . Johnston, C. Chaussade, F. Portis, and E . Van Obberghen. 2003. Phosphoinositide 3-Kinase-mediated Reduction of Insulin 122 Receptor Substrate-1/2 Protein Expression via Different Mechanisms Contributes to the Insulin-induced Desensitization of Its Signaling Pathways in L 6 Muscle Cells. J. Biol. Chem. 278:15641-15651. Plymate, S.R., M . K . Tennant, S.H. Culp, L . Woodke, M . Marcel l i , I. Colman, P.S. Nelson, J . M . Carroll, C.T. Roberts, Jr., and J.L. Ware. 2004. Androgen receptor (AR) expression in AR-negative prostate cancer cells results in differential effects of D H T and IGF-I on proliferation and A R activity between localized and metastatic tumors. Prostate. 61:276-90. Rennie, P.S., and C C . Nelson. 1998. Epigenetic mechanisms for progression of prostate cancer. Cancer Metastasis Rev. 17:401-9. Rochester, M . A . , J. Riedemann, G.O. Hellawell, S.F. Brewster, and V . M . Macaulay. 2005. Silencing of the IGF1R gene enhances sensitivity to DNA-damaging agents in both P T E N wild-type and mutant human prostate cancer. Cancer Gene Ther. 12:90-100. Rui , L . , T . L . Fisher, J. Thomas, and M . F . White. 2001. Regulation of insulin/insulin-like growth factor-1 signaling by proteasome-mediated degradation of insulin receptor substrate-2. J Biol Chem. 276:40362-7. Scorilas, A . , M . Plebani, S. Mazza, D . Basso, A . R . Soosaipillai, N . Katsaros, F. Pagano, and E.P. Diamandis. 2003. Serum human glandular kallikrein (hK2) and insulin-like growth factor 1 (IGF-1) improve the discrimination between prostate cancer and benign prostatic hyperplasia in combination with total and %free P S A . Prostate. 54:220-9. Shimomura, I., M . Matsuda, R . E . Hammer, Y . Bashmakov, M . S . Brown, and J .L. Goldstein. 2000. Decreased IRS-2 and increased S R E B P - l c lead to mixed insulin resistance and sensitivity in livers of lipodystrophic and ob/ob mice. Mol Cell. 6:77-86. Simpson, L . , J. L i , D . Liaw, I. Hennessy, J. Oliner, F. Christians, and R. Parsons. 2001. P T E N Expression Causes Feedback Upregulation of Insulin Receptor Substrate 2. Mol. Cell. Biol. 21:3947-3958. Smyth, G . K . 2005. Limma: linear models for microarray data. In Bioinformatics and Computational Biology Solutions using R and Bioconductor. V . C . R . Gentleman, Dudoit, S., Irizarry, R., Huber, W. , editor. Springer, New York. 397-420. Titus, M . A . , M . J . Schell, F . B . L i h , K . B . Tomer, and J.L. Mohler. 2005. Testosterone and Dihydrotestosterone Tissue Levels in Recurrent Prostate Cancer. Clin Cancer Res. 11:4653-4657. Vassen, L . , W . Wegrzyn, and L . Klein-Hitpass. 1999. Human Insulin Receptor Substrate-2 (IRS-2) Is a Primary Progesterone Response Gene 10.1210/me. 13.3.485. Mol Endocrinol. 13:485-494. W u , R .C . , J. Qin, P. Y i , J. Wong, S.Y. Tsai, M . J . Tsai, and B . W . O'Malley. 2004. Selective phosphorylations of the SRC-3/AIB1 coactivator integrate genomic reponses to multiple cellular signaling pathways. Mol Cell. 15:937-49. Yan, J., C.T. Y u , M . Ozen, M . Ittmann, S.Y. Tsai, and M . J . Tsai. 2006. Steroid receptor coactivator-3 and activator protein-1 coordinately regulate the transcription of components of the insulin-like growth factor/AKT signaling pathway. Cancer Res. 66:11039-46. 123 Zhang, H . , H . Hoff, and C. Sell. 2000. Insulin-like growth factor I-mediated degradation of insulin receptor substrate-1 is inhibited by epidermal growth factor in prostate epithelial cells. J Biol Chem. 275:22558-62. Zhang, J., J. Ou, Y . Bashmakov, J.D. Horton, M . S . Brown, and J .L. Goldstein. 2001. Insulin inhibits transcription of IRS-2 gene in rat liver through an insulin response element (IRE) that resembles IREs of other insulin-repressed genes. Proc Natl Acad Sci USA. 98:3756-61. Zhou, H.J. , J. Yan , W . Luo, G . Ayala, S.H. L i n , H . Erdem, M . Ittmann, S.Y. Tsai, and M . J . Tsai. 2005. SRC-3 is required for prostate cancer cell proliferation and survival. Cancer Res. 65:7976-83. 124 Chapter 5. Discussion 5.1 Relaxin regulation in prostate cancer progression This thesis describes the effect of androgens on relaxin and the effect of relaxin on androgen regulated genes in prostate cancer. The negative effect of androgens on relaxin expression and the clinical relevance of this finding are described in Chapter 2. This study was the first to characterize the negative regulation of relaxin by androgens in the prostate. Vina l l et al (2006) noticed an increase relaxin expression in androgen free conditions in a concurrent study, further supporting our observations. These results illustrate a possible discoordinate regulation between the endogenous relaxin promoter and exogenous relaxin promoter regions used in a chloramphenicol acetyltransferase reporter assay (Brookes et al., 1998). This discoordinate regulation is similar to that seen between endogenous and exogenous ornithine decarboxylase promoters in L N C a P cells (Bai et al., 1998; Betts et al., 1997). The human patient tissue microarray results for relaxin, L G R 7 , and W n t l 1 expression are supportive of the results obtained in the model system to human prostate cancer. Also , the large patient sample size used in T M A s permits greater confidence in findings. These studies would further benefit from correlation of staining intensity to survival data; no survival data is available at this time due to lack of sufficient followup time of this contemporary cohort. Survival results could be examined with the use of a different prostate cancer tissue collection sufficiently mature for survival data. Regardless, the elevated relaxin levels seen in patients undergoing androgen ablation therapy indicate there may be a role for relaxin in promoting prostate cancer progression to androgen independence. 125 The function of relaxin overexpression in prostate cancer is addressed in Chapter 3. The relaxin receptor, L G R 7 , was characterized to be expressed in human patient prostate cancer and prostate stroma, indicating that increased relaxin levels can act in these cells through autocrine or paracrine mechanisms. Our work adds to recent results that demonstrated L G R 7 m R N A to be expressed in cancerous and noncancerous tissues (Feng et a l , 2007), through identifying protein expression, and its localization to the stromal and cancerous tissue. It is possible that the effects described for relaxin overexpression in prostate cancer may result from relaxin signaling through L G R 7 in both paracrine and autocrine mechanisms (Feng et al., 2007; Silvertown et al., 2006; Silvertown et al., 2007; Thompson et al., 2006; Vina l l et a l , 2006). We used in vivo studies to examine gene expression affected by relaxin overexpression because this considers the endocrine effects of the host and paracrine effects of the surrounding tissue. A n example of the difference between in vitro and in vivo expression is that L G R 7 is not expressed at detectable levels in L N C a P cells in vitro in our lab (Personal observations), but is detectable in vivo. Pretreatment with estrogen enhances relaxin effects in some tissues of the female reproductive tract (Huang et al., 1993; Kapi la and Xie , 1998; Mushayandebvu and Rajabi, 1995; Sherwood et a l , 2000), indicating that L G R 7 levels are affected by cell type interactions and factors that may not be present in growth media. It is not unlikely that the in vivo environment, such as that described in this thesis, contains factors that alter expression profiles of L G R 7 and other molecules that are relevant to prostate cancer compared to what is seen in cell culture. 126 5.2 Relaxin regulated Wnt11 pathway Through use of a gene expression microarray screen, I identified that the W n t l 1 /PCDH11Y pathway is upregulated in relaxin overexpressing L N C a P xenografts (Chapter 3). P C D H 1 1 Y , a newly discovered protein, upregulates Wnt signaling and permits A l growth of prostate cancer cells (Terry et al., 2006; Yang et al., 2005). P C D H 1 1 Y is negatively regulated by androgens (Chen et al., 2002), and its upregulation by relaxin, which is itself upregulated in the absence of androgens (Chapter 2), is interesting. Upregulation of P C D H 1 1 Y by relaxin resulted in similar results as androgen withdrawal-mediated upregulation of P C D H 1 1 Y , in that increased P C D H 1 1 Y expression resulted in P-catenin moving to the nucleus, and increased W n t l 1 levels (Chapter 3). P C D H 1 1 Y upregulates other Wnt molecules (Yang et al., 2005); therefore, it may be possible that relaxin is associated with the increase in additional Wnt molecules. W n t l 1 is characterized in embryonic development and in intestinal epithelial cells to stimulate cellular motility (De Calisto et a l , 2005; Ouko et al., 2004; Zhu et al., 2006); this may be a novel mechanism by which relaxin increases motility of prostate cancer cells (Feng et al., 2007). The presence of L G R 7 and elevation of relaxin levels prior to W n t l 1 levels in T M A s of samples from patients undergoing androgen ablation further suggests that relaxin may act through its receptor to increase W n t l 1 in prostate cancer specimens (Chapters 2 & 3). Relaxin has been implicated separately in prostate and cancer growth (Bigazzi et al., 1992; Binder et al., 2002; Samuel et al., 2003); therefore the effects of relaxin on L N C a P xenograft growth were of interest. The finding that relaxin enhances tumour 127 growth agrees with other reports of relaxin overexpression in prostate cancer xenografts (Feng et al., 2007; Silvertown et al., 2006; Silvertown et al., 2007), and is the first time this study has been performed on L N C a P cells in a suitable murine host, nude mice. Previous studies into the effect of relaxin overexpression on L N C a P growth in SCID mice has resulted in minimal tumour take and growth, and little information of relaxin effect on L N C a P growth is reported (Silvertown et a l , 2007). Our data led us to suggest a model of relaxin upregulation causing increased P C D H 1 1 Y levels, which in turn increase W n t l 1, potentially resulting in increased migration. The s.c. L N C a P xenograft model has some inherent limitations for determining the effects of relaxin overexpression on prostate cancer. This model system is limited to monitoring tumour volume. The mechanism by which tumour volume changes can not be readily addressed in this system. L N C a P xenografts are well vascularized, and the effect of relaxin on angiogenesis may not be detectable in this system. Also , it may be difficult for further increases in angiogenesis to occur in the s.c. model, and another model, such as the kidney capsular model may be more appropriate. However, the increased tumour volume in the relaxin overexpressing xenografts suggests the possibility that relaxin may increase angiogenesis. Additionally, relaxin overexpressing PC-3 xenografts were shown to be more well vascularized that controls (Silvertown et al., 2006). The other major relaxin-mediated effect hypothesised, was increased metastatic capability; however, the s.c. L N C a P xenograft model in nude mice is not appropriate for this observation, as L N C a P cells rarely metastasize in this model. A more appropriate model for examining this phenotype would be the orthotopic L N C a P xenograft in SCID mice. The L N C a P xenograft model has inherent heterogeneity, with 128 variations in tumour size and vascularity between tumours, evidenced even between different tumours on the same mouse. Additionally, the exact dimensions of a tumour can be difficult to measure. A s a result, the average tumour volume between two groups can be expected to have quite large variance. These limitations of the s.c. L N C a P xenograft model can also explain the difference in results for the two in vivo experiments (Fig. 3.4 A & B) . A s well , the two experiments differed slightly, with the initial experiment utilizing intact mice having four inoculation sites, while the second experiment involving mice castrated after tumours were established had only three inoculation sites per mouse. The L N C a P model is well characterized to be androgen sensitive, proliferating in the presence of androgens, and surviving in the absence of androgens (Gleave et al., 1992). It secretes P S A , which can be used to monitor the androgen dependent status of the tumor (Gleave et al., 1992). These characteristics mimic the human disease, and make this cell line a useful model system. However, L N C a P tumors do not regress in the absence of androgens, and this is different from the human disease (Gleave et al., 1992). L N C a P cells grown as xenografts can vary in vascularization, and tumour take, which may be a source of variation among biological replicates. Regardless, this xenograft is a suitable model of the human disease in its progression from androgen dependence to androgen independence. 129 5.3 IRS-2 mediates androgen regulated gene expression In examining microarray results, I noticed that relaxin affects the regulation of genes that are characterized to be regulated by androgens, in the presence or absence of androgens. I have discussed the regulation of P C D H 1 1 Y and W n t l 1 above. Relaxin overexpression also results in downregulation of IRS-2 (Chapter 3), which is highly upregulated by androgens (Chapter 4). This is the first indication that IRS-2 is regulated by androgens; however, similar results have been found for IRS-1 in preadipocytes and moblasts (Chen et a l , 2006), and IRS-1 and IRS-2 are also known to be regulated by other steroids in other tissue types (Cui et al., 2003; Lee et al., 1999; Oesterreich et al., 2001; Vassen et al., 1999). IRS-1 is present in the prostate, prostatic intraepithelial neoplasia, and some prostate cancers (Liao et al., 2005). IRS-1 is expressed at low levels in DU145, and in prostate cell lines derived from normal prostate cells (Reiss et al., 2000); however, this protein is not expressed in LNCaPs , likely due to promoter methylation (Nickerson et al., 2001; Reiss et al., 2000), and is not expressed in L A P C 4 or L A P C 9 (Nickerson et al., 2001). Effects of IGF-I on IRS-2 m R N A levels have not been described previously. IRS-2 has a role in androgen-mediated signaling in the presence of IGF-I , as may be inferred from other studies into effects of IGF axis on A R activity (Culig et al., 1994); however, this is the first report of the requirement of IRS-2 for normal androgen mediated signaling in the presence of IGF-I stimulation. The decrease in IRS-2 by relaxin (Chapter 3) may explain the altered androgen signaling seen by relaxin overexpression. It seems that relaxin, like IGF-I, may alter androgen mediated signaling. However, unlike IGF-I, relaxin seems to inversely affect androgen signaling. 130 To attempt to characterize the effect of IRS-2 in androgen and IGF-I signaling in L N C a P cells, I used IRS-2 s i R N A . s i R N A is still a novel technology, and there are "off-target" effects to consider when conducting these experiments. The use of a negative s i R N A aims to control for this. Using antisense oligonucleotides, neutralizing antibodies, or small molecule inhibitors are additional ways to examine the effect of knocking down individual targets. A n y results observed consistently among multiple mechanisms would increase the likelihood that the results observed are due to knocking down the target, and not an artifact of the treatment itself. A n additional novel technology used extensively for this thesis is gene expression microarrays, and this technology is continually evolving. Gene lists are being re-annotated on a regular basis as new information on gene sequences becomes available. New methods for hybridizing, printing, scanning, and analyzing arrays have emerged throughout the course of my candidacy. In fact, there are several generations of microarrays used in this thesis. The microarrays used in the identification of relaxin as a gene negatively regulated by androgens are quite different from those used in the relaxin overexpression study. The methods used to analyse the results of the relaxin overexpression study and the androgen regulation of IRS-2 differed. Regardless, the existing results have been confirmed using independent techniques such as Q - P C R or northern blots. The use of large numbers of biological and technical replicates increased the confidence of the microarray results. New array developments permit for more reliable data mining, with fewer falsely identified changes in gene expression. Furthermore, advances in pathway analysis such as Ingenuity Pathway Analysis has helped tremendously in understanding the large data sets generated in this thesis. 131 5.4 Future Directions We characterized quite fully the effect of androgens on relaxin expression, the involvement of the A R in this regulation, and the relevance in xenografts and human patient samples. However, we have not described the mechanism by which the A R mediates repression of relaxin, or what factors are driving relaxin expression in the prostate. Future studies to address this include treating prostate cells with prolactin, a hormone known to increase relaxin expression in some tissues, which is also implicated in prostate cancer; and studies to see i f the A R is directly interacting with the relaxin promoter, such as D N A footprinting, chromatin immunoprecipitation analysis, and gel electrophoretic mobility shift assays. Current evidence indicates that relaxin is involved in increasing tumor invasiveness through increasing tumour volume, motility and angiogenesis (Chapter 3, Feng et al., 2007; Silvertown et al., 2006), but an increase in metastases has not been shown (Feng et al., 2007). Further research into metastatic spread using the orthotopic murine model of prostate cancer may provide additional information. Patients with metastatic breast cancer had elevated serum relaxin levels (Binder et al., 2004), and this may be an area worth studying in prostate cancer. Relaxin has been difficult to study in the sera due to poor antibodies and low circulating levels of the hormone, but newer antibodies may permit for more sensitive studies on serum relaxin levels. The results of relaxin overexpression indicate an association between relaxin overexpression and altered P C D H 1 1 Y , W n t l 1, and IRS-2 expression. Relaxin can be knocked down in vivo using anstisense oligonucleotides to relaxin. The results of knocking down relaxin on P C D H 1 1 Y , Wnt l 1, and IRS-2 w i l l further characterize the 132 function of relaxin in the upregulation of these genes. Another way of identifying the role of relaxin on P C D H 1 1 Y or IRS-2 expression is using reporter assays to see i f relaxin dosing causes increased P C D H 1 1 Y or decreased IRS-2 promoter activity in L N C a P cells. These cells may need to be transfected with L G R 7 in order for in vitro assays to be successful. Another option to see the effects of relaxin in vitro may be to co-culture L N C a P s with prostatic stromal cells expressing L G R 7 . W n t l 1 has been shown to be transforming and to increase motility in other cell lines (Christiansen et al., 1996; Ouko et al., 2004). Treating L N C a P s with W n t l 1 would aid in determining i f this increases cellular motility. Increased motility has been characterized for relaxin previously (Feng et al., 2007), and W n t l 1 would be a novel means of increasing relaxin-induced motility. The PCDH11 Y / W n t l 1 and IRS-2 pathways were not the only ones identified in microarray analysis of relaxin overexpressing LNCaPs . Several interesting pathways were identified and validated using microarray analysis, and warrant further study. Clathrin-D and clathrin-L were recently identified to be regulated by relaxin in thyroid carcinoma (Hombach-Klonisch et al., 2006), and would also be of interest to study. Current research arising from this thesis involves examining the effects of blocking 5-a-reductase activity on androgen regulated gene expression at the protein level, using prostate cancer samples from patients treated with the 5-a-reductase blocker, dutasteride. The effects of dutasteride on P C D H 1 1 Y , IRS-2, relaxin, and other proteins w i l l be analyzed. The work done to identify IRS-2 as an androgen and IGF-I regulated gene is in early stages. It would be of great interest to fully characterize the effects of IGF-I and 133 androgens on IRS-2 levels and androgen signaling. Further information would be obtained by determining the effect that serum and knocking down IRS-2 have on these treatments in vitro. This characterization would require survival and proliferation assays, and an examination of the downstream signaling pathways involved in IGF-I mediated androgen signaling. Additional analysis of the effect of IGF-I on androgen responsive genes could be achieved through ARE-luciferase assays. One way to screen for novel targets regulated by androgen and IGF-I treatment of cells is through microarray analysis of m R N A , microRNA, and protein, and these experiments are currently being performed in our lab in IRS-2 s i R N A and A R s i R N A treated cells. The comparison of genes differentially regulated between these two experiments w i l l help to identify genes regulated by both androgens and IRS-2. The finding that IRS-2 s i R N A is highly effective at decreasing A R mediated signaling in IGF-I stimulated LNCaPs indicates that this could make a good therapeutic target in patients with demonstrated increased levels of IGF-I. It would be interesting to compare all of the IRS-2 knockdown studies done in L N C a P with similar studies in another cell line that expresses IRS-1. There is significant, though not complete, redundancy between these molecules. In a patient expressing both IRSs in normal tissue, but who has lost one of IRS-1 or IRS-2 in a tumour for which IGF-I signaling is deemed important, knocking down the remaining IRS molecule could be a fairly low-toxic therapy. There may be sufficient cross talk to enable the patient to survive with relatively minimal side effects. Another potential therapeutic approach may be to inhibit multiple targets at once, such as knockdown of both IRS-2 and relaxin simultaneously or in sequence in the presence and absence of androgens respectively. 134 The hypothesis of this thesis is that relaxin is involved in the activation of key pathways involved in invasion and angiogenesis during the progression of androgen independent prostate cancer, and therefore may be a suitable target for therapy. The technologies and models used in this thesis were unable to determine i f relaxin increases angiogenesis in prostate cancer; however, relaxin is involved in increased invasive qualities, can increase tumour size, and upregulates genes that promote A l growth, and indicate that this may be a suitable target for novel therapeutics. 135 5.5 References Bai , G . , S. Kasper, R.J . Matusik, P.S. Rennie, J .A. Moshier, and A . Krongrad. 1998. Androgen regulation of the human ornithine decarboxylase promoter in prostate cancer cells. JAndrol. 19:127-35. Betts, A . M . , I. Waite, D . E . Neal, and C . N . Robson. 1997. Androgen regulation of ornithine decarboxylase in human prostatic cells identified using differential display. FEBSLett. 405:328-32. Bigazzi , M . , M . L . Brandi, G . Bani, and T .B . Sacchi. 1992. Relaxin influences the growth of M C F - 7 breast cancer cells. Mitogenic and antimitogenic action depends on peptide concentration. Cancer. 70:639-43. Binder, C , T. Hagemann, B . Husen, M . Schulz, and A . Einspanier. 2002. Relaxin enhances in-vitro invasiveness of breast cancer cell lines by up-regulation of matrix metalloproteases. Mol Hum Reprod. 8:789-96. Binder, C , A . Simon, L . Binder, T. Hagemann, M . Schulz, G . Emons, L . Trumper, and A . Einspanier. 2004. Elevated concentrations of serum relaxin are associated with metastatic disease in breast cancer patients. Breast Cancer Res Treat. 87:157-66. Brookes, D . E . , D . Zandvliet, F. Watt, P.J. Russell, and P .L . Mol loy . 1998. Relative activity and specificity of promoters from prostate-expressed genes. Prostate. 35:18-26. Chen, M . W . , F. Vacherot, A . De L a Taille, S. Gil-Diez-De-Medina, R. Shen, R . A . Friedman, M . Burchardt, D . K . Chopin, and R. Buttyan. 2002. The emergence of protocadherin-PC expression during the acquisition of apoptosis-resistance by prostate cancer cells. Oncogene. 21:7861-71. Chen, X . , X . L i , H . Y . Huang, and J.F. L i n . 2006. [Effects of testosterone on insulin receptor substrate-1 and glucose transporter 4 expression in cells sensitive to insulin]. Zhonghua YiXueZaZhi. 86:1474-7. Christiansen, J .H. , S.J. Monkley, and B.J . Wainwright. 1996. Murine WNT11 is a secreted glycoprotein that morphologically transforms mammary epithelial cells. Oncogene. 12:2705-11. Cu i , X . , Z . Lazard, P. Zhang, T .A . Hopp, and A . V . Lee. 2003. Progesterone crosstalks with insulin-like growth factor signaling in breast cancer cells via induction of insulin receptor substrate-2. Oncogene. 22:6937-41. Cul ig , Z . , A . Hobisch, M . V . Cronauer, C. Radmayr, J. Trapman, A . Hittmair, G . Bartsch, and H . Klocker. 1994. Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. Cancer Res. 54:5474-8. De Calisto, J., C . Araya, L . Marchant, C.F. Riaz, and R. Mayor. 2005. Essential role of non-canonical Wnt signalling in neural crest migration. Development. 132:2587-2597. Feng, S., I .U. Agoulnik, N . V . Bogatcheva, A . A . Kamat, B . Kwabi-Addo, R. L i , G . Ayala , M . M . Ittmann, and A . I . Agoulnik. 2007. Relaxin promotes prostate cancer progression. Clin Cancer Res. 13:1695-702. 136 Gleave, M . E . , J.T. Hsieh, H . C . Wu , A . C . von Eschenbach, and L . W . Chung. 1992. Serum prostate specific antigen levels in mice bearing human prostate L N C a P tumors are determined by tumor volume and endocrine and growth factors. Cancer Res. 52:1598-605. Hombach-Klonisch, S., J. Bialek, B . Trojanowicz, E . Weber, H J . Holzhausen, J.D. Silvertown, A . J . Summerlee, H . Dralle, C. Hoang-Vu, and T. Klonisch. 2006. Relaxin enhances the oncogenic potential of human thyroid carcinoma cells. Am J Pathol. 169:617-32. Huang, C , Y . L i , and L . L . Anderson. 1993. Stimulation of collagen secretion by relaxin and effect o f oestrogen on relaxin binding in uterine cervical cells of pigs. J Reprod Fertil. 98:153-8. Kapila , S., and Y . X i e . 1998. Targeted induction of collagenase and stromelysin by relaxin in unprimed and beta-estradiol-primed diarthrodial joint fibrocartilaginous cells but not in synoviocytes. Lab Invest. 78:925-38. Lee, A . V . , J .G. Jackson, J .L. Gooch, S.G. Hilsenbeck, E . Coronado-Heinsohn, C . K . Osborne, and D . Yee. 1999. Enhancement of insulin-like growth factor signaling in human breast cancer: estrogen regulation of insulin receptor substrate-1 expression in vitro and in vivo. Mol Endocrinol. 13:787-96. Liao, Y . , U . Abe l , R. Grobholz, A . Hermani, L . Trojan, P. Angel , and D . Mayer. 2005. Up-regulation of insulin-like growth factor axis components in human primary prostate cancer correlates with tumor grade. Human Pathology. 36:1186-1196. Mushayandebvu, T., and M . Rajabi. 1995. Relaxin stimulates interstitial collagenase activity in cultured uterine cervical cells from nonpregnant and pregnant but not immature guinea pigs; estradiol-17 beta restores relaxin's effect in immature cervical cells. Biol Reprod. 53:1030-1037. Nickerson, T., F. Chang, D . Lorimer, S.P. Smeekens, C . L . Sawyers, and M . Pollak. 2001. In vivo progression of L A P C - 9 and L N C a P prostate cancer models to androgen independence is associated with increased expression of insulin-like growth factor I (IGF-I) and IGF-I receptor (IGF-IR). Cancer Res. 61:6276-80. Oesterreich, S., P. Zhang, R . L . Guler, X . Sun, E . M . Curran, W . V . Welshons, C . K . Osborne, and A . V . Lee. 2001. Re-expression of estrogen receptor alpha in estrogen receptor alpha-negative M C F - 7 cells restores both estrogen and insulin-like growth factor-mediated signaling and growth. Cancer Res. 61:5771-7. Ouko, L . , T.R. Ziegler, L . H . Gu, L . M . Eisenberg, and V . W . Yang. 2004. W n t l l signaling promotes proliferation, transformation, and migration of IEC6 intestinal epithelial cells. J Biol Chem. 279:26707-15. Reiss, K . , J .Y . Wang, G . Romano, F . B . Furnari, W . K . Cavenee, A . Morrione, X . Tu, and R. Baserga. 2000. IGF-I receptor signaling in a prostatic cancer cell line with a P T E N mutation. Oncogene. 19:2687-94. Samuel, C.S. , H . Tian, L . Zhao, and E.P. Amento. 2003. Relaxin is a key mediator of prostate growth and male reproductive tract development. Lab Invest. 83:1055-67. Sherwood, O.D. , L . M . Olson, S. Zhao, and H.R. Little. 2000. Inhibition of nitric oxide synthase activity diminishes the acute effects of relaxin on growth, but not softening, of the cervix in the rat. Endocrinology. 141:2458-64. 137 Silvertown, J.D., J. N g , T. Sato, A J . Summerlee, and J .A. Medin. 2006. H2 relaxin overexpression increases in vivo prostate xenograft tumor growth and angiogenesis. Int J Cancer. 118:62-73. Silvertown, J.D., J.C. Symes, A . Neschadim, T. Nonaka, J .C .H . Kao, A.J .S . Summerlee, and J .A. Medin. 2007. Analog of H2 relaxin exhibits antagonistic properties and impairs prostate tumor growth. FASEB J. 21:754-765. Terry, S., L . Queires, S. Gil-Diez-de-Medina, M . W . Chen, A . de la Taille, Y . Al lory , P .L . Tran, C C . Abbou, R. Buttyan, and F. Vacherot. 2006. Protocadherin-PC promotes androgen-independent prostate cancer cell growth. Prostate. 66:1100-13. Thompson, V . C , T .G . Morris, D.R. Cochrane, J. Cavanagh, L . A . Wafa, T. Hamilton, S. Wang, L . Fazl i , M . E . Gleave, and C C Nelson. 2006. Relaxin becomes upregulated during prostate cancer progression to androgen independence and is negatively regulated by androgens. Prostate. 66:1698-709. Vassen, L . , W . Wegrzyn, and L . Klein-Hitpass. 1999. Human Insulin Receptor Substrate-2 (IRS-2) Is a Primary Progesterone Response Gene. Mol Endocrinol. 13:485-494. Vina l l , R . L . , C . G . Tepper, X . B . Shi, L . A . Xue, R. Gandour-Edwards, and R . W . de Vere White. 2006. The R273H p53 mutation can facilitate the androgen-independent growth of L N C a P by a mechanism that involves H2 relaxin and its cognate receptor L G R 7 . Oncogene. 25:2082-93. Yang, X . , M . W . Chen, S. Terry, F. Vacherot, D . K . Chopin, D . L . Bemis, J. Kitajewski, M . C Benson, Y . Guo, and R. Buttyan. 2005. A human- and male-specific protocadherin that acts through the wnt signaling pathway to induce neuroendocrine transdifferentiation of prostate cancer cells. Cancer Res. 65:5263-71. Zhu, S., L . L i u , V . Korzh, Z . Gong, and B . C . Low. 2006. R h o A acts downstream of Wnt5 and W n t l 1 to regulate convergence and extension movements by involving effectors Rho kinase and Diaphanous: use of zebrafish as an in vivo model for GTPase signaling. Cell Signal. 18:359-72. 138 Appendix I. Genes upregulated at least 5-fold in R1881-treated cells UP W I T H R1881 (5 fold or higher) Gene Accession Number "kall ikrein 2, prostatic" AI927872 chymotrypsinogen B1 A A 8 4 5 1 6 8 " C D C 1 4 (cell division cycle 14, S. cerevisiae) homolog A " N68854 alkylation repair; a lkB homolog A A 6 0 9 6 0 9 vi l l in- l ike AI887514 " C D C 1 4 (cell division cycle 14, S. cerevisiae) homolog B " A A 4 1 7 3 1 9 leukocyte-associated Ig-like receptor 1 A A 9 9 1 1 9 6 "inositol 1,4,5-triphosphate receptor, type 2" A A 4 7 9 0 9 3 " C D 3 Z antigen, zeta polypeptide (TiT3 complex)" AI289821 brain-specific angiogenesis inhibitor 3 H17398 " T A T A box binding protein (TBP)-associated factor, R N A AI493402 polymerase I, A , 48kD" "protein kinase, AMP-act iva ted , gamma 3 non-catalytic subunit" A A 2 5 6 3 8 3 prostate epithelium-specific Ets transcription factor AI668916 hypothetical protein A A 7 0 8 8 8 6 " P O U domain, class 3, transcription factor 1" AI807330 v-myb avian myeloblastosis viral oncogene homolog N49284 cytochrome c oxidase subunit V i l a polypeptide 1 (muscle) A A 8 7 2 1 2 5 leukocyte membrane antigen H84077 tyrosinase-related protein 1 A A 4 2 4 9 9 6 "butyrophilin, subfamily 3, member A 3 " A A 4 7 8 5 8 5 cathepsin K (pycnodysostosis) R00859 C18B11 homolog (44.9kD) R32439 "butyrobetaine (gamma), 2-oxoglutarate dioxygenase (gamma-butyrobetaine hydroxylase) 1" A A 4 5 5 9 8 8 a disintegrin and metalloproteinase domain 3a (cyritestin 1) A A 4 3 5 9 6 3 glycoprotein A repetitions predominant A A 1 2 2 2 8 7 "collagen, type X V , alpha 1" A A 4 5 5 1 5 7 139 Appendix II. Genes down regulated at least 5-fold in R1881-treated cells D O W N W I T H R1881 (5 fold or higher) Gene Accession Number " C D C 2 0 (cell division cycle 20, S. cerevisiae, homolog)" A A 5 9 8 7 7 6 hypothetical protein M G C 2 4 8 7 AI279844 regulator o f G-protein signalling 11 A A 8 8 7 5 3 0 Human normal keratinocyte m R N A AI024655 diptheria toxin resistance protein required for diphthamide biosynthesis (Saccharomyces)-like 2 AI018643 Kaiso AI151208 C D 14 antigen A A 7 0 1 4 7 6 L I M domain only 4 H27986 "sirtuin (silent mating type information regulation 2, S.cerevisiae, A A 9 3 5 5 6 4 homolog) 7" " E S T , H igh ly similar to C A 3 4 H U M A N C O L L A G E N A L P H A 3(IV) AI952285 C H A I N P R E C U R S O R [H.sapiens]" toll- l ike receptor 7 N30597 GPI-anchored metastasis-associated protein homolog A A 4 7 9 6 0 9 "branched chain keto acid dehydrogenase E l , beta polypeptide (maple syrup urine disease)" A A 4 2 7 7 3 9 intercellular adhesion molecule 3 A A 4 7 9 1 8 8 hypothetical protein D K F Z p 7 6 1 D 1 8 2 3 A A 7 7 5 4 0 5 toll- l ike receptor 3 R76099 tight junction protein 3 (zona occludens 3) A A 4 0 2 0 4 0 nuclear body protein S p l 4 0 H66484 formin (limb deformity) AI040235 N-Acetylglucosamine kinase AI669862 " M A D (mothers against decapentaplegic, Drosophila) homolog 3" W72201 140 Appendix IV. Primers for quantitative RT-PCR Primer/probe Sequence relaxin forward 5' - T G A A G C C G C A G A C A G C A G T - 3' relaxin reverse 5' - A A C A T G G C A A C A T T T A T T A G C C A A - 3 ' relaxin probe 5' - V I C - A A A A T A C T T A G G C T T G G A T A C T C A T T C T C G A A A A A A G A G A - T A M R A - 3' P C D H 1 1 Y forward 5' - A A A C A T T A T C T C C A G G A G T T T G G A - 3 ' P C D H 1 1 Y reverse 5' - T T G C C C T G G A T A C C T T T C A G A - 3 ' p-actin forward 5' - G C T C T T T T C C A G C C T T C C T T - 3 ' P-actin reverse 5' - C G G A T G T C A A C G T C A C A C T T - 3 ' 142 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0100760/manifest

Comment

Related Items