Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
The list of abbreviations Abbreviations Full name ALS Acid-labile subunit AKT Protein kinase B ADT Androgen deprivation therapy AI Androgen independent APES 3-aminopropyl triethoxy-saline AR Androgen receptor BAD BCL-2 associated death promoter BCL-2 B-cell CLL/lymphoma 2 bFGF basic fibroblast growth factor BLAST Basic local alignment search tool BPH Benign prostatic hyperplasia BSA Bovine serum albumin CAM Chorioallantoic membrane cDNA Complementary DNA C-FABP Cutanious fatty acid binding protein CK Cytokeratin DAB 3, 3’-Diaminobenzidine DBD DNA binding domain DEPC Diethylpyrocarbonate DHT 5-α-dihydrotestosterone Page | 1 Abbreviations Full name DMSO Dimethyl sulphoxide DNA Deoxyribonucleic acid EBM EndoGRO basal medium ECL Enhanced Chemiluminescence ECM Extra cellular matrix EDTA Ethylenediaminetetraacetic acid EGF Epidermal growth factor EGFP Enhanced green fluorescent protein EHS Engelbreth-Holm-Swarm ET-1 Endothelin-1 EZH2 Enhancer of zeste homologue 2 FABP Fatty acid binding protein FGF Fibroblast growth factor FISH Fluorescence in-situ hybridization Flt-1 Fms-like tyrosyl kinase-1 FCS Foetal calf serum G418 Geneticin GSTP1 Glutathione S Tranferase π Grb2 Growth factor receptor-binding protein 2 HPC Hereditary prostate cancer gene HIF Hypoxia-induced factor HRP Horse radish peroxidise Page | 2 Abbreviations Full name HUVEC Human umbilical vein cells IFGBP Insulin-like growth factor binding protein IGF Insulin-like growth factor IRS1 Insulin receptor substrate-1 KDR/Flk-1 Kinase insert domain receptor LB Lysogeny broth LBD Ligand binding domain LCFA Long chain fatty acid LOH Loss of heterozygosity LUTS Lower urinary tract symptoms MAPK Mitogen-activated protein kinase MVD Microvessel density NBF Neutral buffered formalin NCRI National cancer research institute NES Nuclear export signal NLS Nuclear localization signal OD Optical density P13K Phosphorylates phosphatidylinositol 3’-kinase PAP prostatic acid phosphatase PBS Phosphate buffered saline PCR Polymerase chain reaction PDB Protein data bank Page | 3 Abbreviations Full name PDK Phosphatidylinositol-dependent kinase PDK1 3-phosphoinositide dependent protein kinase 1 PEN-STREP Penicillin/streptomycin PIN Prostatic Intraepithelial Neoplasia PIP3 Phosphatidylinositol (3,4,5)-trisphosphate PlGF Placental growth factor PPARs Peroxisome proliferator-activated receptors PPRE Peroxisome proliferator response element PTEN Phosphatase and tensin homologue PTHrP Parathyroid hormone related protein PSA Prostate specific antigen RANK Receptor activator of nuclear factor - RB Retinoblastoma susceptibility RIN RNA integrity number RNA Ribonucleic acid rpm round per minute RT Room temperature RT-PCR Reverse transcription polymerase chain reaction RXR Retinoic acid receptor SC Subcutaneous SD Standard deviation SHC Src- and collagen-homology Page | 4 Abbreviations Full name siRNA Short interfering RNA SOB Super optimal broth SOC SOB medium plus glucose SV40 Simian virus 40 TBS Tris-buffered salin T/E Trypsin/EDTA TMB 3, 3’, 5, 5’-Tetramethylbenzidine TSG Tumor suppressor gene uPA urokinase-type plasminogen activator VEGF Vascular endothelial growth factor min (s) Minute (s) g Gram mg Milligram µg Microgram ng Nanogram pg Picogram ml Millilitre µl Microlitre Page | 5 Chapter 1 Introduction 1.1 The epidemiology of prostate cancer 1.1.1 Incidence and mortality of prostate cancer Prostate cancer is the most common form of cancers in men in the UK. Nearly 35,000 men are diagnosed with prostate cancer in the UK each year and the figure is still going up due to longer life expectancy of men. It accounts for approximately a quarter (24%) of all new male cancer diagnoses. Although there has been a pronounced rise in prostate cancer incidence over the last 20 years, the increase in mortality rate has been much more slowly. Much of the increase in incidence was due to the increased detection of prostate cancer by a wider use of prostate specific antigen (PSA) testing. Prostate cancer accounts for around 13% of male deaths from cancer in the UK and is the second most common cause of cancer mortality in men, after lung cancer. It has been reported 10,000 deaths in the UK from prostate cancer in 2005 as shown in Figure 1.1. Figure 1.1 Age standardised incidence and motility rates for prostate cancer 1975–2005 [1, 2] Page | 6 1.1.2 Worldwide Epidemiology Geographic variations exist in the impact of prostate cancer. Clinical incidence and mortality of prostate cancer are low in Asian men, higher in population of developed western countries and highest in African-Americans [3-5]. Interestingly, once individuals from a low incidence/mortality region move to a high incidence/mortality region, the risk of prostate cancer becomes higher. The length of time living in the new environment seems also to be correlated with the increase [6, 7]. 1.1.3 Factors affecting the prostate cancer 1.1.3.1 Age The prostate cancer has been known as a disease of elderly men. The diagnosis is very rare before the age of 50 years. However, beyond this age both incidence and mortality rate increased exponentially. The risk of prostate cancer increases faster with age than any other major cancer and, with a growing age population grow due to longer life expectancy, the burden of illness from prostate cancer will probably continue to increase in the future. The studies have shown that the age specific prevalence of histologic prostate cancer is similar in Japan and United States but there is a remarkable difference in the age specific prevalence of clinical prostate cancer between Japanese and American men [8]. These data indicated that the initiation rate of prostate cancer may be the same in both groups however there are differences in the rate of promotion or progression to clinically evident prostate cancer. Therefore, the presence of histologic prostate cancer appears to be related to age, other risk factors that increase the development of prostate cancer probably affect the promotion or progression steps of the malignant transformation pathway. Page | 7 1.1.3.2 Genetic factors There is an increase in prostate cancer incidence in male relatives of patients with prostate cancer – Cater et al. showed that prostate cancer was diagnosed about two times (15% of men who had a father or brother affected by the prostate cancer) higher than in control group (8% of control population with no affected relatives) [9]. It has also been reported that a significant number of early onset prostate cancers may be inherited in an autosomal dominant fashion and suggest that hereditary factors are most commonly affected men with early onset of prostate cancer, however, they are responsible for only a small proportion of prostate cancer [9]. Genetic linkage analysis on potential genetic susceptibility regions which include the hereditary prostate cancer gene 1 (HPC1) on 1q23-25, hereditary prostate cancer gene 2 (HPC2) on 17p, hereditary prostate cancer gene X (HPCX) on Xq27-28 and Predisposing for Cancer Prostate (PCAP) on 1q42-43 have identified several strong candidate genes such as RNASEL, ELAC2, and MSR1 [10-15]. A number of studies provide support, both functional and epidemiological, that these genes play a role in hereditary prostate cancer. Yet other studies have suggested that their role may be small. Further work will be required in order to understand how these genes involved in prostate cancer and to sort out the roles of the various variants of these genes [16]. 1.1.3.3 Dietary factors Dietary patterns have been recognised as one of the key factor in the development of different type of cancer malignancies. Consumed food component may be metabolized to oncogene, may alter hormonal balance, or may protect against the development of Page | 8 certain cancer. As mentioned previously (section 1.1.3), the rising rate of prostate cancer in both black South African and Japanese men who migrated to the United States is thought to be due to the change in their dietary patterns, and risk was higher when they migrated at a younger age. As the “Westernization” diet pattern is closely linked with high dietary intake of fat, the increased risk for the development of prostate cancer has a relationship with dietary fat, and it has been proved by several studies that dietary factors in black, white, and Asian-Americans with prostate cancer found a positive association between prostate cancer risk and total fat intake in all ethnic groups [7]. Similarly, for mortality, a multination survey highly correlated prostate cancer deaths with total fat consumption [17, 18] (see Figure 1.2). Some studies propose that obese men have increased prostate cancer mortality by 2.5-fold [7]. Figure 1.2 The positive correlation between age-adjusted mortality for prostate cancer and animal fat consumption by country [19]. Page | 9 Further evidence suggested that not only the amount of fat consumption but also the type of fat had an effect on prostate cancer growth. Dietary fats are classified as either saturated fatty acids or unsaturated fat acids as can be seen in Table 1.1. It has been suggested that saturate fat consumption has been constantly associated with prostate cancer risk [20] whereas some studies have been published on the role of monounsaturated fats in association with protecting prostate cancer [21]. An epidemiological research also noted that in countries where people consume larger amount of olive oil which is rich in all kinds of mono-unsaturated fats, a lower incidence of several cancers including prostate cancer was found [19]. In addition, a number of researches have been performed on effect of various kinds of poly-unsaturated fatty acids. A negative effect of fish oil, a highly unsaturated omega-3 fatty acid, on tumour development was suggested when nude mice fed menhaden (omega-3 rich fish oil concentrate) had decreased growth of injected Du145 prostate cancer cells [22]. Saturated Fatty Acids CH3(CH2)16COOH (No double bond) Unsaturated Fatty Acids Mono-unsaturated CH3(CH2)7CH=CH(CH2)7COOH (One double bond) Poly-unsaturated CH3(CH2)4CH=CHCH2CH=CH(CH2)7COOH (Two or more double CH3CH2CH=CHCH2CH=CHCH2CH=CH bond) (CH2)7COOH Table 1.1 Classification of fatty acids Page | 10 Retinoic acid, also known as vitamin A, is a lipophilic molecule that plays an important role in regulating cell growth, differentiation, cell-cycle arrest and apoptosis [23-25]. Retinoic acid shows the capacity of reducing tumourigenicity and is currently used as a therapeutic agent in several human cancers [26]. It has been also reported that intake of retinoic acid is inversely proportional to prostate cancer incidence [27, 28]. However, results of some studies were conflicting – in some tissues retinoic acid appears to promote rather than inhibit cell survival and retinoic acid treatment can enhance skin tumour formation [29-31]. Recent studies from Schug et al. demonstrated that retinoic acid mediated cell biological process by two contradictory signalling pathways: retinoic acid inhibits cell proliferation by signalling through its receptors; the ‘non-traditional’ pro-proliferative activities of retinoic acid may be mediated by nuclear receptor PPARβ/δ. The response through witch pathway might depend on the ratio of FABP5 / CRABP-II [32]. The relationship between other diet factors and prostate cancer is also complex. A cohort study found that increasing consumption of beans, lentils, tomatoes, dates and dried fruit significantly reduced risk for prostate cancer [33]. Studies also showed that diets riching in soya protein had a protective effect on prostate cancer [34, 35]. In general, the results from dietary intake studies support the concept that a high fibre and low fat diet may protect men against the development of prostate cancer. 1.1.3.4 Other factors The association of prostate cancer with other risk factors have been evaluated in a number of studies. However, the statistic data for risk factors such as socioeconomic Page | 11 status, occupation, cigarette smoking, alcohol consumption and sexual activity were not convincing that these risk factors resulted in an increasing incidence of prostate cancer. 1.2 The pathology of prostate cancer 1.2.1 Anatomy and physiology of the prostate gland The prostate is an accessory gland and present only in males. The prostate is located in front of the rectum, just below the urinary bladder and it surrounds the urethra, which carries urine from the bladder out through the penis. Between the rectum and the bladder, attached to the prostate are a pair of glands called the seminal vesicles which provide nutrients for the semen (see Figure 1.3). Figure 1.3 Interior of male urinary system and prostate (Picture is taken from http://www.pioa.org/c_anatomy.html) Page | 12 Prostate is one of the exocrine gland which secrete through ducts to the outside of the body and partly made up of muscular fibres and partly glandular, with ducts opening into the prostatic portion of the urethra. The prostate gland secretes a slightly alkaline fluid that contains a number of enzymes, citric acid, prostate specific antigen and zinc [36-38]. The fluid neutralizes and protects sperm from the acidic environment of the female vagina. The prostate is a small acorn shaped gland and composes of four morphological regions known as the anterior fibromuscular stroma, peripheral zone, central zone and transition zone [39] (see Figure 1.4). The anterior fibromuscular stroma forms entire surface of the prostate as the prostatic capsule and consists of approximately 5% of the prostatic weight. This zone is usually devoid of glandular components because this zone is composed only of muscular and fibrous tissue. The anterior fibromuscular stroma is mainly responsible for the secretion of prostatic fluid during ejaculation by initiating muscular contractions. The peripheral zone constitutes approximately 70% of the normal prostate gland. It is the posterior portion of the prostate gland which surrounds the distal urethra. More than 64% prostate cancer originated from this zone. The central zone makes up 25% of the total tissue mass of the prostate gland. This zone surrounds the ejaculatory ducts and accounts for only up to 2.5% of prostate cancers although these cancers tend to be more aggressive [40]. The remaining 5% of the prostate gland is formed by the transitional zone which is the innermost part of the prostate gland and surrounds the urethra where it passes through the prostate. The transitional zone is the primary site of Benign Prostatic Hyperplasia (BPH). Page | 13 Figure 1.4 Schematic of prostate gland describes zonal anatomy. PZ = peripheral zone, TZ = transition zone, CZ = central zone [41] 1.2.2 Normal prostatic cells The prostatic epithelium mainly consists of three anatomically distinct cell populations: Luminal cell, basal cell and neuroendocrine cells as shown in Figure 1.5. The secretory luminal cells are the most predominant cells in the prostate and responsible for secretion of prostatic proteins including prostate-specific antigen (PSA) and prostatic acid phosphatase (PAP), two important clinical biomarkers for prostate cancer. These cells express androgen receptor (AR) and require androgens for maintenance of their function and viability [42]. Luminal cells can be characterised by their expression of the filament high molecular weight proteins, cytokeratins (CK) 8 and CK18. Most of the prostate cancers are adenocarcinomas (95%) which are considered to differentiate towards Page | 14 secretory cells as the majority of cancer cells express luminal cell specific markers such as CK8, CK18 and PSA [43]. Figure 1.5 Schematic view of the cell types within human prostate epithelium. Picture is taken from [44] Basal cells are the second major epithelial cell type in prostate. These cells are located between the luminal cells and the underlying basement membrane and are able to express high molecular weight protein markers such as CK5 and CK14. Unlike luminal cells, the basal cell population neither produce prostatic secretory proteins nor express AR and they do not undergo apoptosis in response to androgen ablation [45, 46]. However, it has been reported that basal cells express factors that have a protective function against DNA damage, such as anti-apoptotic gene Bcl-2 and the free-radical scavenger Gst-π [39]. In addition, it is speculated that the basal cell population contains a subset of cells acts as a stem cell reservoir in order to generate all three types of cells [47]. Page | 15 Finally, the third prostatic epithelial cell type is the neuroendocrine cells which can be found in all area of prostate. The marker used to distinguish this cell population is Chromogranin A. Neuroendocrine cells provide specialized neuropeptides or neuroendocrine factors that support the vitality and proliferation of luminal epithelial cells [39]. Some studies indicated a positive correlation between the level of neuroendocrine factors and the more advanced stage of prostate adenocarcinoma [48, 49]. 1.2.3 Prostate cancer initiation 1.2.3.1 Benign Prostatic Hyperplasia (BPH) BPH has been characterized as non-malignant overgrowth of prostate tissue. BPH shares some similarities with prostate cancer: (1) Both of them are age related disease; (2) Both of them require androgen stimulation. However, BPH is not a precursor of prostate cancer. BPH develops in the transition zone, a ring of tissue around the urethra and its growth is inward toward the prostate's core, constantly tightening around the urethra and associated with lower urinary tract symptoms (LUTS). Most of prostate carcinoma, on the other hand, begin in the outer peripheral zone of the prostate and invade the surrounding tissues. Although the pathogenesis of BPH is still poorly understood, some studies indicated that BPH is related to hormonal changes that occur as men age. The testis produces the hormone testosterone, which is converted to 5-alpha-dihydrotestosterone (DHT) and estrogen in certain tissues. High level of DHT, a testosterone derivative involved in prostate growth, may trigger for initiation of BPH [50]. Page | 16 1.2.3.2 Prostatic Intraepithelial Neoplasia (PIN) PIN is the earliest morphologically identifiable stage in prostatic carcinogenesis and is defined as abnormal proliferation of secretory cells of prostatic glands without invasion into surrounding stroma [51]. PIN has been classified into two groups: low grade and high grade PIN. In low grade PIN, cell nuclei are enlarged, remarkable vary in size, have a normal or slightly increased chromatin content and possess small or inconspicuous nucleoli [52]. In high grad PIN, cell nuclei are also enlarged but at the similar size and shape variation, have dramatically increased density and clumping of chromatin content and possess prominent nucleoli that are similar to those of invasive carcinoma cells. In addition, the basal cell layer shows frequently interruptions in high grade PIN but intact or rarely interrupted in low grade PIN. Currently, conventional use of term PIN without qualification refers to high grade PIN which must be included as a part of pathologic report. Four main patterns of high grade PIN have been distinguished: flat, tufting, micropapillary and cribriform [53]. Tufting is the most common pattern found in PIN lesion (97%) although multiple patterns can be found in most of cases at the same time. There are some common properties between high grade PIN and prostate cancer: epidemiologically, the incidence and extend of both high grade PIN and prostate cancer is associated with age; histopathologically, high grade PIN and prostate cancer are both multifocal and share a same location in peripheral zone; Cytogenetic studies showed that they also share similar genetic alterations and proliferation activity. Many studies exhibited correlation between PIN and prostate cancer and the appearance of high grade PIN is an important marker for detection of potential prostate cancer [54]. Page | 17 1.2.4 Gleason grading system of prostate cancer Gleason score, a grading system for prostate cancer devised by Dr. Donald Gleason in 1977 as a method for evaluating the prognosis of men with prostate cancer, is the most widespread method of prostate cancer tissue grading used today [55]. The classic Gleason scoring diagram identified five distinct prostate cancer patterns that are technically referred to as tumour grades which have been numbered ranging from 1 to 5, with Grade 1 being the least aggressive and Grade 5 being the most aggressive cancer (see Figure 1.6). Figure 1.6 Gleason grading system diagram (Picture is taken from http://en.wikipedia.org/wiki/Gleason_staging_system) Gleason score is a sum of the primary grade representing the dominant of tumour pattern and a secondary grade which is assigned to second most prevalent tumour pattern and is a number ranging from 2 to 10. The Gleason grading system allows patients with tumours of similar biologic behaviour to be compared in studies and as a strong prognosis factor, it also provides useful information for patient treatment options. Page | 18 1.2.4 Prostate cancer cell lines To understand the pathogenesis and malignant progression of prostate cancer, prostate cancer cell lines are wildly used in the laboratory studies. However, the most common used prostate cancer cell lines are: PNT2, LNCaP, Du145 and PC-3. 1.2.4.1 PNT2 The cell line is derived from a normal human prostate of a 33 year old male at post mortem and has been immortalized by transfection with a plasmid containing simian virus 40 (SV40) with a defective replication origin [56]. The PNT cells possess a well differentiated morphology with the expression of cytokeratin 8, 18 and 19 which are the markers of differentiated luminal cells of the glandular prostate. Cytokeratin 14, a marker of epithelial basal cells, is not expressed. In addition, the cells are not tumorigenic when injected in Nude mice [57]. 1.2.4.2 LNCaP The cell line was established in 1977 by J.S. Horoszewicz et al., from a needle aspiration biopsy of the left supraclavicular lymph node of a 50 year old male patient with confirmed diagnosis of metastatic prostate carcinoma. Prostatic specific markers, such as androgen receptor (AR), prostate specific antigen (PSA) and prostatic acid phosphatise (PAP), are presented in cultured cells [58, 59]. LNCaP cells are androgen sensitive, do not produce a uniform monolayer, but grow in clusters and they are responsive to 5-alpha-dihydrotestosterone [60]. LNCaP cells rarely form tumours in mice by subcutaneous inoculation and no metastatic spread was noted in any of the animals given injections [61]. Page | 19 1.2.4.3 DU-145 The cell line is derived from prostate carcinoma that was metastatic to central nervous system lesion of a 69 year old patient. Not only cytokeratin markers of luminal cells of the glandular prostate but also MUC1 are found in DU-145 cells [62]. However, expression of prostatic associated markers (AR, PAP and PSA) are not present in this cell line [62]. Tumours were formed in 29 out of 32 of nude mice given subcutaneous injection of DU-145 cells but no evidence of metastasis has been found [63]. 1.2.4.4 PC-3 and PC-3M The cell line was initiated from a bone marrow metastasis of a grade IV prostatic adenocarcinoma from a 62 year old male patient [64]. The cultured cells show androgen independent growth and produce both high incidence of tumorigenicity and metastasis in nude mice [65]. The PC-3M human prostate cancer cell line was isolated from liver metastasis produced in nude mice subsequent to intrasplenic injection of the parental PC-3 cells. As a derivative of PC-3 cell line, PC-3M shares some similar properties with PC-3 but more aggressive than the rest of prostate cancer cell lines including PC-3 [66]. 1.3 Oncogenes and tumor suppressor genes in prostate cancer The molecular biological events involved in initiation, promotion and progression of prostate cancer are still unclear. Apparently, it is generally believed that prostate cancer formation and metastasis are due to mutations of specific genes which play important roles in regulating cell cycle control, proliferation, differentiation and apoptosis. The genes have been divided into two main groups: oncogenes and tumour suppressor genes. Page | 20 1.3.1 Oncogenes in prostate cancer Oncogenes can be defined as genes are partly or solely responsible for tumourigenesis. Oncogenes are generally mutated forms of certain normal cellular genes known as proto-oncogenes which provide signals to regulate cell biological behaviour such as cell growth, cell differentiation and apoptosis. There are a large number of oncogenes are now recognized, but only a relatively small number have been associated with prostate cancer. 1.3.1.1 C-MYC oncogene The C-MYC oncogene is located at 8q24 and involves in regulation of cell proliferation, cell differentiation and apoptosis [67]. This well known oncogene, characterized as a transcription factor, increases its copy numbers by amplification or/and chromosomal gain [68]. Both over expression and amplification of C-MYC have been detected in prostate carcinomas. Fluorescence in-situ hybridization (FISH) with a specific probe for C-MYC was performed in PIN, localized prostate cancer and its metastasis [69]. A number of studies also showed that the level of C-MYC mRNA is enhanced in prostate cancer compared to BPH or normal prostate tissue. These results indicated that the amplification of C-MYC may be potential marker for prostate cancer progression. However, the contribution of C-MYC to prostate cancer development and progression is infrequent: 8% of primary prostate cancers and about 30% of metastatic lesions. Page | 21 1.3.1.2 C-ERB-B2 oncogene The C-ERB-B2 oncogene also known as HER-2/NEU is located at chromosome 17q1121. It encodes the p185neu transmembrane glycoprotein which belongs to the epidermal growth factor receptor family of tyrosine kinase [70]. It has been reported that over expression of C-ERB-B2 may contribute to androgen independence in prostate cancer [71]. It has been reported that C-ERB-B2 can activate the androgen receptor pathway in absence of ligands such as steroid hormones. In addition, casodex, an androgen antagonist, blocks the effects of growth factors but not C-ERB-B2 on androgen receptor function [71]. This indicates that either C-ERB-B2 signalling cascade produces a molecule that binds at a competitively higher affinity than casodex, or C-ERB-B2 activates androgen receptor function on fresh non-optimal androgen receptor ligand binding domains rather than to enhance androgen receptor function on high-affinity binding sites. Sadasivan et al. found that C-ERB-B2 expression correlated with advanced stages of prostate cancer and higher Gleason scores [72]. However, reports on frequencies of C-ERB-B2 over-expression in prostate cancer vary wildly. 1.3.1.3 BCL-2 oncogene The B-cell CLL/lymphoma 2 (BCL-2) oncogene is located at chromosome 18q21.3 and is an inner mitochondrial membrane protein that plays an important role in regulating cell apoptosis. It has been reported that BCL-2 promotes tumorigenesis by inhibiting programmed cell death instead of increasing cell division [73]. In the prostate gland, BCL-2 normally expressed in basal epithelial cells which are not affected by androgen deprivation but not in luminal cells. In prostate cancer cells, BCL-2 oncogene contribute to accelerating progression of androgen dependent LNCaP cells to androgen Page | 22 independence via inhibiting castration-induced apoptosis initiated by androgen deprivation [74]. Similarly, McDonnell et al. (1992) also demonstrated that transfection of BCL-2 into LNCaP cells enabled the survival of those cells during androgen withdrawal. These results indicated that over expression of BCL-2 prevented apoptosis of prostate cancer cells subjected to androgen withdrawal and related to progression to androgen independence in prostate cancer cells. 1.3.1.4 EZH2 polycomb oncogene The enhancer of zeste homologue 2 (EZH2) oncogene is located at chromosome 7q35 and is known as a transcription repressor. It has been reported that the expression of both EZH2 mRNA and protein progressively increase from BPH to primary prostate tumor, to metastatic tumors, indicating that elevation of EZH2 protein level may accrue in advance of the development of metastasis [75]. Furthermore, over expression of EZH2 protein in prostate cancer cells leads to transcriptional repression as a result, increases cell proliferation [76]. These results suggested that EZH2 protein level, as a marker of prostate cancer progression and metastasis, might be helpful to predict patient outcome after prostatectomy. 1.3.1.5 RAS oncogenes The RAS oncogene family, consisting of Harvey (H) RAS (located at chromosome 11p15.5), Kirsten (K) RAS (located at chromosome 12p12.1) and neuroblastoma (N) (located at chromosome 1p13.2), encodes cytoplasmic GTPase signal transduction proteins with molecular weights of approximately 21kDa. The RAS proteins are localized at the inner surface of the plasma membrane and are involved in regulating Page | 23 cell proliferation, adhesion, apoptosis and cell migration activated by point mutations at codon 12, 13 or 61 [77, 78]. Some studies indicated that over expression of RAS protein seems to link to prostate cancer progression and could be correlated with development of bone metastasis [79]. However, activation of RAS oncogenes is uncommon in prostate cancer with a frequency of less than 5% mutant RAS oncogenes in North American. 1.3.2 Tumor suppressor genes in prostate cancer Tumor suppressor genes (TSGs) are normal genes involved in reducing cell proliferation, promoting cell differentiation, repairing DNA mistakes and inducing cell apoptosis. The importance of TSGs is that loss of function mutations of TSGs leads to breakdown of cellular order and deregulation, which results in turning a normal cell into a tumor cell. Mutations in TGSs are frequently seen in many types of cancers including prostate cancer. There are three major ways to identify which TSGs are involved in prostate cancer: 1. Identifying prostate cancer TSGs mutated in prostate cancer pedigrees and positional cloning 2. Studying TSGs that had already been found to be important in other types of cancers 3. Examining prostate cancer for regions of constantly chromosomal loss of heterozygosity (LOH) Similar to the study of oncogenes, the investigation of TSGs in prostate cancer will provide insights into initiation and progression of prostate cancer. It will also bring new prognostic markers and future treatment strategies. Page | 24 1.3.2.1 P53 TSG The p53 gene is located at chromosome 17q13 and encodes a 53kDa nuclear phosphoprotein. The normal functions of p53 appear to be cellular growth suppression, regulation of cell cycle which contributes to maintenance of DNA stability and cell apoptosis [80, 81]. When the cell DNA damaged, p53 increasing the expression of mitotic inhibitors such as p21/WAF1, which binds to and inhibits cyclin kinase complexes, preventing phosphorylation of specific substrates and arresting the cell cycle in G1 phases to allow for repairing damaged DNA. If the damaged DNA is unrepairable, p53 induces apoptosis [82]. However, mutation of p53 TSG results in a loss of p21 gene expression and loss of regulation of the cell cycle which enables mutated cells to divide in an unchecked manner and increases tumourigenic potential. The p53 TSG is inactivated by point mutations which are the most commonly observed genetic alterations (50%) in human cancers analysed. In addition, the wild type p53 protein has a very short half-life within cells, however, the mutant p53 protein has a prolonged half-life and it is detectable by immunohistochemistry. In prostate cancer, the p53 protein has been widely examined using molecular and immunohistochemical assay. No evidence of p53 activation has been found in BPH and a low frequency (10-30%) of p53 TSG mutations has been observed in primary prostate tumours. However, p53 activation rate ranges from 50 to 80% in patient with metastatic prostate cancer [83, 84]. These results showed that mutations in the p53 TSG rarely occur in early stage of prostate carcinoma and are more commonly activated in advanced and metastatic prostate cancers. These evidences indicated that p53 TSG activation is associated with malignant progression of prostate cancer and worse prognosis. Page | 25 1.3.2.2 RB TSG The retinoblastoma susceptibility (RB) gene is located at chromosome 13q14 within one of the most commonly deleted regions in prostate cancer. RB TSG encodes a nuclear phosphoprotein which suppresses cell proliferation by arresting cell cycle in G1 phase through its interactions with transcription factors, notably the E2F family. Loss of RB occurred at relatively early stage of the tumour progression as the studies showed a similar frequency in a low grade localised prostate cancer as well as in more advanced prostate carcinoma [51]. Bookstein et al. (1990) showed that transfection of wild type RB gene into a prostate cancer cell line Du145 resulted in a dramatic reduction (~15fold) in the tumor formation of the transfected cells in nude mice. The results above suggested that RB TSG could play an important role in tumorigenicity in prostate cancer. Furthermore, approximately one third LOH of the RB TSG were found in prostate cancer cases. However, some data demonstrated the lack of correlation between LOH at RB and absent RB protein expression, which indicated the existence of another TSG in this region [85]. 1.3.2.3 PTEN TSG The phosphatase and tensin homologue (PTEN) TSG is located at chromosome 10q23.3 and acts as a (3,4,5)-trisphosphate lipid phosphatase (PIP3) by that inactives dephosphorylation. the phosphatidylinositol PIP3 initiates the phosphatidylinositol-dependent kinase (PDK) / AKT signaling pathway to increase cell proliferation and cell survival [86]. In normal cells, PTEN expression blocks the PDK/AKT signal transduction pathway that results in cell apoptosis. On the contrary, in tumor cells, LOH and mutation of PTEN results in promoting AKT activity, thereby blocking apoptotic signal. Page | 26 It has been reported that PTEN LOH or/and mutation are rare in early stage of prostate cancer whereas these changes are most commonly found in metastatic prostate cancer [87]. McMenamin et al. also found that complete loss of PTEN is associated with highstage and grade prostate cancer. These results suggest that PTEN may play an important part in the acquisition of metastatic potential in prostate cancer. 1.3.2.4 GSTP1 TSG Glutathione S Tranferase π (GSTP1) TSG is located at chromosome 11q13 and encodes the detoxifying isoenzyme GSTP1 which inactivates electrophilic carcinogens by conjugation with glutathione. In addition, GSTP1 also protects DNA from oxidative damage [88]. The studies demonstrated that the activity of GSPT1 in prostate tumors was inhibited by hypermethylation which is the most common (>90%) reported epigenetic alteration in prostate cancer [89]. It is also been reported that the transcription of GSTP1 is completely blocked in the LNCaP cell line because of hypermethylation of the gene promoter. Harden et al. demonstrated that the quantitation of GSTP1 hypermethylation using real-time PCR is highly sensitive and specific to prostate cancer (not present in BPH specimens or normal tissue), which results in improving accuracy of the detection. As GSTP1 gene hypermethylation occurs in early stage of prostate cancer, it may serve as a useful prognostic marker. Page | 27 1.4 Molecular mechanisms of metastasis in prostate cancer The main cause of prostate cancer treatment failure is metastasis which is defined as a process that cancer cells escape from the site of primary tumour and form secondary tumours at distant sites [90]. The molecular mechanism of metastasis in prostate cancer is still not clear, however, there are a few hypotheses to explain the site specific pattern of metastatic displacement. One of the most well known hypothesis is the “Seed and Soil” theory [91] which consists of a number of steps: Firstly, the malignant cancer cells lose their cell-cell and cell-matrix adhesive characteristics, become motile and gain the capability to degrade proteins that make up the surrounding extracellular matrix using degradative enzymes [92]. Secondly, once the malignant cancer cells escaped from the primary tumor and enter the circulation (blood and/or lymphatic vessels), they interact with the host immune system resulting cell apoptosis. However, very few cells survived in the circulation known as the “seed”. Finally, the surviving cancer cells arrest at the specific distant site known as the “soil” because the environments of different organs are biologically unique. In other words, their endothelial cells express different cell-surface receptors and growth factors that may influence the cell-endothelial interactions. The cells then extravasate from the circulation, invade to the interstitium and proliferate to form a metastatic tumor. These processes require various changes in cell properties, such as: cell proliferation, cell invasion, and angiogenesis [90]. In addition, the metastasis site also capable of producing further metastasis [93]. Page | 28 1.4.1 Mechanisms of bone metastasis in prostate cancer The metastasis of prostate cancer shows a significantly strong predilection for bone. The studies demonstrated that approximately 90% of advanced prostate cancer metastasizes to bone [94]. The osseous tissue consists of two different structures: compact bone which is mineralized cortical bone and trabecular bone which is metabolically active cancellous bone. The outer layer of the trabecular bone contains bone marrow, where the osteoblasts and osteoclasts are generated. The most frequent sites of skeletal metastasis are the trabecular bones with high content of bone marrow [95]. 1.4.1.1 Osteoclastic bone resorption The first important steps of prostate cancer cells metastasis to bone is the activation of osteoclasts resorption which results in not only allowing the seeding of cancer cells but also releasing survival and growth factors that promote prostate cancer metastasis [96]. Among these factors, the most prominent is the Receptor Activator of Nuclear Factor ligand (RANK ligand) produced by osteoblastic cells. RANK ligand is mainly located on the surface of osteoblasts and bind to its specific receptor (RANKR) which is expressed on the surface of the osteoclastic precursors. The studies showed that osteotropic factors such as parathyroid hormone, 1, 25dihydroxyvitamin D3, and parathyroid hormone can induce osteoclast formation and activation by up-regulation of RANKR [97]. Osteoprotegerin, a member of tumor necrosis factor receptors, acts as a competitive inhibitor against RANK ligand binding RANKR and thus blocks the RANKR - RANK ligand interaction [96]. On the other hand, some contraventional results demonstrated that osteoprotegerin also bind to tumor Page | 29 necrosis factor related apoptosis inducing ligand results in inhibition of cancer cell apoptosis [98]. Nevertheless, osteoprotegerin remarkably decreases prostate cancer metastasis to bones because it inhibits bone resorption [99]. Therefore, the development of competitive antibodies, such as recombinant osteoprotegerin, against RANK ligand can be potential treatments for bone metastasis (Figure 1.7). Figure 1.7 The ratio of RANK ligand to osteoprotegerin determines the level of osteoclastogenesis. 1.4.1.2 Domination of the osteoblastic lesions Osteoblasts are bone forming cells which come from mesenchymal stem cells [100]. Many factors can regulate the growth, survival and differentiation of osteoblasts, such as Endothelin-1 (ET-1), basic fibroblast growth factor (bFGF), Insulin-like growth factor (IGF) and Parathyroid hormone related protein (PTHrP). Some of them are also produced by metastatic prostate cancer cells. Page | 30 ET-1, a peptide of 21 amino acid residues, is synthesized in vascular endothelial cells and is involved in angiogenesis, the formation of the new bone and osteoblastic bone metastasis [101]. The previous studies demonstrated that ET-1 is one of the most overexpressed factors in osteoblast and can increase prostate cancer cell proliferation probably by enhancing the mitogenic effects of IGF and epidermal growth factor. Recent studies showed that overexpress ET-1 can up-regulate the vascular endothelial growth factor (VEGF) expression through the endothelin receptor A in osteoblast and thus could act as an angiogenic factor in the formation of bone metastasis [101]. IGF-1 plays an important role in cell proliferation and inhibiting cell apoptosis in prostate cancer cells. Insulin-like growth factor binding protein-3 (IFGBP3) together with acid-labile subunit (ALS) binds IGF-1(75% - 90% of circulating IGF-1) and the ligands binding trigger the downstream pathway of IGF-1 receptor (IGF-1R). The ligands binding activate the intrinsic tyrosine kinase of IGF-1R which results in phosphorylation of IGF-1R substrates - insulin receptor substrate-1 (IRS1), Src- and collagen-homology (SHC) and growth factor receptor-binding protein 2 (Grb2). Then, these phosphorylated factors active with Ras – Raf – mitogen-activated protein kinase (MAPK) cascade, which result in increasing cell proliferation [102]. The IRS-1 also phosphorylates phosphatidylinositol 3’-kinase (P13K) and AKT, which promotes antiapoptotic survival of the cancer cells as shown in Figure 1.8. In addition, it has been reported that elevated concentration of urokinase-type plasminogen activator (uPA) which is produced by prostate cancer cells also stimulate the local activity of IGF-1, therefore, increases prostate cancer cells proliferation and induces the osteoblastic reaction. Page | 31 Figure 1.8 IGF-I–IFGBP3–ALS binds to IGF-IR to trigger a series of ligandmediated activations including RAS–RAF–MAPK and PI3K–AKT signalling pathway, controlling cell proliferation and cell apoptosis, respectively. Picture is taken from [102]. 1.5 Androgen 1.5.1 Hormone therapy for advanced prostate caner Androgen, also known as androgenic hormones or testoids, has key influence on regulating the development and maintenance male characteristics through binding to its receptors. The predominant and most active androgen is testosterone which is mainly (approximately 90%) synthesized by cells in testes and an important factor in the development and function of the prostate gland. Page | 32 In prostate cancer, radical prostatectomy is the most common treatment when the disease is at localized stage. For patients diagnosed with advanced or metastatic disease, androgen deprivation therapy (ADT) is the standard treatment which includes medical or surgical castration, anti-androgens, and combined androgen blockade, thereby resulting in a suppression of androgen signalling. ADT is an effective initial treatment with 80%-90% primary response rate, however, the duration of response to ADT is limited to 14 – 20 months (reviewed in [103]). Then, the prostate cancer become refractory to additional treatment due to the fact that prostate cancer cells eventually developed to an androgen independent (AI) state. In most cases with ADT, patients developed castrate resistant prostate cancer, which is no longer in response to androgen deprivation. Nevertheless, the majority of AI tumor cells remain expressing androgen receptor (AR), which appears to act as normal in spite of the castrate level of androgen. 1.5.2 Multiple pathways to androgen independence 1.5.2.1 Androgen receptor The testicular synthesized testosterone transports to prostate tissue and is converted into 5α-dihydrotestosterone (DHT), which has higher affinity for AR than testosterone, by a 5α-reductase enzyme. Altogether, testosterone and DHT bind to AR and exert their biological function roles including regulation of specific gene transcription. The AR is a member of the steroid hormone receptor family of ligand-dependent transcription factor. Like other member of nuclear receptors, AR protein is composed of several functional domains: (1) N-terminal regulatory domain, which mainly involved in most ligandactivated transcriptional activity; (2) DNA binding domain, which is the best conserved region and important for recognizing androgen response elements; (3) hinge domain, Page | 33 which connects the DNA binding domain and ligand binding domain and it also contains a ligand dependent nuclear translocation signal; (4) ligand binding domain, which is involved in androgens binding induced activity as well as in nuclear localization, receptor dimerization and interaction with other proteins [104]. The molecular mechanisms of prostate cancer progression from an androgen-dependent phase to an androgen-independent stage are still not fully understood. However, there are several signalling pathways may account for the development of androgenindependent prostate cancer such as: AR amplification, AR mutations and AR coregulators overexpression. 1.5.2.2 Androgen receptor amplification Amplification of the AR gene is a potent mechanism that enables prostate cancer cells to become more sensitive to the reduced level of circulating androgens. It has been reported that amplification of AR was rarely found in untreated primary prostate cancer. However, approximately 30% of androgen-independent prostate tumors had increased levels of AR gene expression after ADT (reviewed in [105]). Furthermore, studies have found that increased level of AR gene expression in progression from androgen sensitive to androgen resistance using xenograft models [106]. However, AR amplification only occurred in a proportion of androgen independence prostate cancers suggested that AR amplification is not the only mechanism responsible for development of androgen independence. Page | 34 1.5.2.3 Androgen receptor mutations Studies have found that the frequency of mutations in the AR is significantly increased in tumors collected after ADT compared with few AR mutations in primary tumor samples without the therapy [107, 108]. These results suggested that gaining of AR mutation may be one of the mechanisms leading to the development of prostate cancer’s androgen resistance. AR may acquire transcriptional activity not only by testosterone and DHT specifically but also by other circulating steroid hormones such as adrenal androgens, estrogen and cortisol et.al due to mutations in the ligand binding domain of the AR. This relaxation of ligand specificity allows AR to recruit other substitute androgens when reduced the level of androgen and thus promote the proliferation and avoid the apoptosis of prostate cancer cells. LNCaP cells were the first cell line to be identified with this type of mutation in the AR ligand-binding domain [109]. AR mutations in LNCaP cells extended its ligand binding with other steroid hormones leading to increasing AR expression and cell proliferation. A study by Hara et.al demonstrated that long-term culture of LNCaP-FGC cells in androgen depleted medium with 1µM anti-androgen bicalutamide can produce point mutants W741C, W741L, and T877A located in AR ligand binding domain [110]. Furthermore, researches showed that AR T877A mutation allowed the hydroxyflutamide (anti-androgen) to trigger AR transcriptional activity and promote LNCaP cells’ proliferation [111]. Page | 35 Constant with AR mutation described above, many AR mutations found in clinical cancer cases and associated with progression of prostate cancer and the failure of ADT but not all androgen independent prostate tumors carried AR mutation. Therefore, there must be mechanisms other than AR mutations that promote androgen independence. 1.5.2.3 Other pathways Further studies revealed additional mechanisms including, outlaw AR pathways: activation of receptor tyrosine kinases resulted in phosphorylation of AR by either the P13K/AKT pathway or the mitogen activated protein kinase (MAPK) pathway, producing a ligand independent AR [112]; bypass AR pathway: inhibition of apoptotic response by activation of anti-apoptotic candidate genes such as BCL-2 which can promote progression to androgen independence [113]. As prostate cancer cells employ multiple pathways to survive in reduced level of androgens it is important to identify the pathways in individual case to design the most efficient therapeutics. 1.5.3 Androgen receptor and growth factors interactions It has reported that growth factors, such as insulin-like growth factor, fibroblast growth factor and epidermal growth factor (EGF), can increase AR transactivation under absence or low levels of androgen [114, 115]. A study by Gregory et.al suggested that EGF increased level of transcriptional intermediary factor 2/glucocorticoid receptor interacting protein 1 through mitogen activated protein kinase (MAPK) signalling pathway; therefore, promote AR transcriptional response in prostate cancer cells [116]. Another angiogenesis growth factor, vascular endothelial growth factor (VEGF), has Page | 36 been frequently studied recently. The elevated expression of VEGF, which resulted in promoting vascular growth and the endothelial cell proliferation, was found in Androgen-stimulated prostate growth [117]. Other reports also showed that functions of VEGF were activated through hypoxia-induced factor (HIF) regulated by androgens and/or androgen receptor [118]. Further studies demonstrated the significant correlation between HIF expression with AR and VEGF expression, and provided firm support for this control system [119]. Understanding of such cross-talk between AR cascade and growth factors signalling may provide new opportunities for therapy of aggressive prostate cancer. 1.6 Angiogenesis Angiogenesis is a multiple and physiological progress involving the growth of new capillaries from pre-existing blood vessels. Angiogenesis happens to repair damaged tissue when wounds are healing; therefore, normal cells can switch on the growth of blood vessel by releasing angiogenic factors. On the other hand, they also can produce antiangiogenic factors which switch blood vessel growth off. Tumor angiogenesis is that cancer cells stimulate the growth of hundreds of capillaries from the nearby blood vessels, which grow around and/or into the tumor, to establish an independent supply of nutrients and oxygen. Angiogenesis is not only important to tumor development but also a critical step in tumor metastasis. The immature, highly permeable blood vessels, which have little basement membrane and few intercellular junctional complexes, Page | 37 provide efficient route of exit for tumor cells to leave the primary site, enter the blood stream, travel to another part of the body and begin to grow there. 1.6.1 Fibroblast growth factors Fibroblast growth factor (FGF) family is composed of 23 members which are heparinbinding proteins and the size of these single chain peptides is range from 16 to 18 kDa [120]. FGFs regulate cell functions by binding FGF receptors (FGFR1, FGFR2, FGFR3 and FGFR4) located on the cell surface in the presence of heparin proteoglycans. FGFs were found to involve various biological functions including cell proliferation, cell differentiation, angiogenesis and tissue repair. It has been reported that FGF prototypes, such as FGF-1 and FGF-2, stimulate tumor cell growth rate and organize endothelial cells to form tube-like structures which results in promoting angiogenesis. In vitro studies demonstrated that FGF-1 significant increase vascularisation when transfected to the chicken chorioallantoic membranes (CAMs) and the level of protein kinase B (AKT) mRNA also enhanced in the experiment indicating the involvement of the AKT kinase signaling pathway. Moreover, adding the AKT inhibitor significantly reduces FGF-1 induced angiogenesis in CAMs [121]. Another important heparin binding prototype of FGF family, FGF-2 has also been correlated with tumor progression by activation of second of secondary angiogenesis factors, such as vascular endothelial growth factor (VEGF). Hori et al. provided the evidence that monoclonal antibody against FGF-2 inhibit tumour growth and vascularization [122]. Other studies also show that FGF-2 is associated with low differentiation, metastasis and poor survival in cancer patients [123-125]. Page | 38 1.6.2 Vascular endothelial growth factor (VEGF) There are at least five members of the vascular endothelial growth factor (VEGF) family existed in human: VEGF-A (usually referred to as VEGF), VEGF-B, VEGF-C and VEGF-D and a structurally related molecule, placental growth factor (PlGF). All members of VEGF family are crucial important angiogenic factors which primarily target vascular endothelial cells. They promote new blood vessel formation by binding tyrosine kinase receptors (also known as VEGF receptors) located on cell surface. Three VEGF receptors have been recognized: VEGFR-1 or fms-like tyrosyl kinase-1(Flt-1), VEGFR-2 or Kinase insert domain receptor (KDR/Flk-1) and VEGFR-3 or Flt-4. Both VEGF and VEGF-B bind to VEGFR-1 which is involved in the organization of development of blood vessel, haematopoiesis and enhance VEGF-induced VEGF-2 signaling during abnormal angiogenesis. VEGF but not VEGF-2 binds to VEGFR-2 results in promoting cell proliferation, mitosis, vascular permeability and angiogenesis, whereas VEGF-C and VEGF-D are ligands for VEGFR-3, which stimulates lymphangiogenesis [126]. 1.6.2.1 The isoforms of VEGF VEGF is a ~45kDa homodimeric heparin-binding glycoprotein which was firstly identified by Senger et al. in 1983. VEGF gene consists 8 exons separated by 7 introns and is located at chromosome 6p21.3 [127]. So far, there are 12 VEGF isoforms which have been divided into two families according to their terminal exon (exon 8) splice site: the proximal splice site, designated as VEGFxxx or distal splice site, designated as VEGFxxxb [128]. The identified multiple proteins of VEGF expressed in human tissues and cells are named as VEGF121, VEGF121b, VEGF145, VEGF145b, VEGF165, VEGF165b, Page | 39 VEGF189, VEGF189b and VEGF206 according to different number of amino acids. VEGF121 is thought to be the most diffusible isoform due to the absence of a heparinbinding domain. VEGF165, the predominant VEGF isoform secreted by various of cells, not only exists in diffusible location but also remains bound to cell surface and extra cellular matrix (ECM). However, the larger isoforms such asVEGF189 and VEGF206 with high affinity to heparin remain localized within ECM (reviewed in [126]). 1.6.2.2 VEGF and prostate cancer Most of tumor cell lines secrete VEGF in vitro indicating that VEGF may play a crucial role in tumor angiogenesis. This has been confirmed by in situ hybridization studies which demonstrated the VEGF mRNA expressed in majority human carcinomas such as breast cancer, lung cancer, bladder cancer, prostate cancer and ovary caner (reviewed in [126]). In prostate cancer, VEGF is expressed differently in normal, benign and prostate cell line. It has been reported that the expression of VEGF mRNA was detected in PIN, and poorly differentiated tissues, but not in normal prostate tissue [129]. Further studies also showed that the microvessel density (MVD) was increased significantly in metastatic prostate cancer sample when compared with non-metastatic prostate cancer samples[130]. In addition, Du et al. detected the higher expression of hypoxia-inducible factor (HIF), a key regulator of VEGF expression, in malignant prostate cancer compared with the normal and benign prostate tissue [131]. Clinical studies revealed that the level of VEGF expression in serum, plasma or urine was correlated with higher Gleason grade, metastasis and disease-specific survival [132, 133]. On the other hand, inhibition of VEGFR-1 and VEGFR-2 using AZD-2171 (Cediranib, AstraZeneca) induce tumor shrinkage in 56.5% patients (13 out of 23 patients with measurable Page | 40 disease) with 4 meeting the criteria for partial response [134]. The results above indicated that VEGF plays crucial role in prostate cancer at both early initiating stage and later stage for tumor progression and metastasis. VEGF interacts with VEGF-2 to stimulate endothelial cell proliferation through the mitogen activated protein kinase (MAPK) pathway and promote vascular permeability, and subsequently with VEGFR-1 to assist the organization of new capillary tubes. 1.6.3 Peroxisome proliferator-activated receptors Peroxisome proliferator-activated receptors (PPARs) are part of the nuclear receptors family of transcription factors and originally identified in Xenopus frogs as receptors that induce the proliferation of peroxisome in cells in the early 1990s [135]. The functional role of these ligand activated transcription factors is involved in regulating fatty acid catabolism, lipid storage and glucose metabolism in the body. There are three isoforms of PPARs: PPARα, PPARβ/δ and PPARγ. They are expressed in various tissues and play different roles. Page | 41 1.6.2.1 Structure of PPAR Despite the location and functional differences, PPAR isoforms share similar structure and functional domains. Like other nuclear receptors, PPARs are composed of four major domains: A/B, C, D and E/F domains (Figure 1.9). Figure 1.9 The schematic view of the domain structure of PPARs. The N-terminal region contains PPARs A/B domains which are different in both length and amino acid sequence. The A/B domain contains a transactive domain independent of the presence of the ligand, which is termed activation function 1 (AF1). The activity of this domain can be regulated by protein kinase phosphorylation [136]. The C domain, which is the most conserved region, consists of a DNA binding domain (DBD). This domain contains two typical zinc finger motifs which bind to specific sequences of DNA and responsible for binding PPAR to peroxisome proliferator response element (PPRE) within the promoter region of the target gene. The D domain is a highly flexible hinge region. The C-terminal region of PPAR is E/F domain which contains ligand binding domain (LBD) and ligand dependent activation domain (AF2). The LBD of PPARs consists 13 α helices and a β sheet forming a hydrophobic pocket which is about twice larger than other nuclear receptors [137]. This region is important for activation of PPAR binding to the PPRE and also plays an important role in nuclear localization. In addition, the researches demonstrated that PPARs could be activated by a variety of Page | 42 fatty acid at a certain concentration [138]. PPARs form heterodimers with the 9-cis retinoic acid receptor (RXR), which result in a conformational change of both receptors, allowing the heterodimers to bind to the PPRE DNA promoter region containing repeat motifs (AGGTCANAGGTCA) and regulate the gene transcription (Figure 1.10). Figure 1.10 Model for PPARs signalling pathway. 1.6.2.2 PPAR β/δ PPAR β/δ is expressed ubiquitously throughout the body with different levels in certain cell types. PPAR β/δ can be activated by natural ligands, such as fatty acids, retinoid acid and prostaglandins as well as synthetic ligands including GW501516, GW0742 and L165041. Several studies demonstrate that PPAR β/δ plays important roles in cell proliferation, differentiation and apoptosis. Activation of PPAR β/δ can stimulates the cell proliferation in colon cancer, liver cancer, lung cancer and breast cancer cell lines (reviewed in [139]). It has also been reported that PPAR β/δ increases cell proliferation in PNT and LNCaP but not in DU145 and PC-3 prostate cancer cell lines [140]. Page | 43 However, the result data on the role of PPAR β/δ in cell proliferation are inconsistent and controversial. In addition, a number of recent studies suggest that activation of PPAR β/δ with GW501516 also can stimulate angiogenesis by enhancing the expression of VEGF [141, 142]. Strong evidence suggests that ligand activation of PPAR β/δ can enhance the expression of 3-phosphoinositide dependent protein kinase 1 (PDK1), subsequently activation of the downstream protein kinase: protein kinase B (AKT) by phosphorylation. Activation of PDK1/AKT signalling pathway has a direct effect on cell apoptosis through phosphorylation of the BCL-2 family member BCL-2 associated death promoter (BAD) thereby suppressing cell apoptosis and promoting cell survival. Recent work by Schug et al. also supports this PDK1/AKT anti-apoptosis signalling pathway regulated by PPAR β/δ [32]. 1.6.2.3 PPAR γ PPARγ has been the most extensively studied isoform of PPAR family. It is mainly expressed in adipose tissues and plays a crucial role in adipocyte differentiation and lipid metabolism. Previous observation showed PPARγ ligands promote the VEGF production in bladder and prostate cancer cells [143, 144]. Analogous findings were also reported by Cho et al. who detected a significant increase in VEGF and its receptors in epithelial cells treated with PPARγ synthetic ligand, thiazolidinediones. In addition, activation of PPARγ also stimulates the tumor formation in colon cancer [145]. On the other hand, there are several lines of evidence to support that PPARγ ligand activation are capable of inhibiting angiogenesis progression under certain conditions. It is suspected that the various time and dose of PPARγ ligand treatment or the variety of organisms and cells examined may lead to such discrepancies. Page | 44 1.7 Fatty acid binding proteins 1.7.1 Fatty acid binding protein family As mentioned in previously in 1.1.3.3, several studies highlighted a possible association between high incidence of clinical prostate cancer in western developed nations and high dietary fat intake. Fatty acids, which are produced by hydrolysis of ester linkage in oil or fat, commonly have a chain of 4 to 28 carbons (short or medium chain: aliphatic tails are less than 12 carbon atoms; long chain: aliphatic tails are between 12 and 24 carbon atoms). Fatty acids not only function as an energy source but also act as cell signalling molecules which regulate gene expression, proliferation and apoptosis pathways [146]. Fatty acids are very soluble once they secreted from fat cell to the aqueous cytoplasm environment. Then, free fatty acids cross the plasma membrane either by passive diffusion or though transporters. Once fatty acids reach the cytoplasm, they either enter a metabolic pathway or bind to intracellular fatty acid binding proteins which carry them to different organelles. Fatty acid binding proteins (FABPs) are a family of cytoplasmic proteins that reversibly bind to long chain fatty acids and other lipophilic substances such as eicosanoids and retinoids, with high affinity [147]. FABP super family can be classified into two groups: (1) FABPpm, located in the plasma membrane; (2) FABPc (usually referred to as FABPs), intracellular or cytoplasmic proteins. Since the initial discovery of FABPs in 1972, so far nine members of FABP have been identified as shown in Talbe 1.2. Page | 45 Table 1.2 Family of fatty acid binding proteins [146] Page | 46 FABPs are small intracellular proteins with similar molecular masses around 15kDa. Members of FABP family display various pattern of tissues expression. Although FABPs were named according to the tissue in which they were first identified, none of members is specific for one tissue or cell type and some tissues express more than one FABP, which indicates that FABPs may have unique functions. Some of FABPs expressed in few specific tissues but some of them exhibit broad tissue distribution. For instance, LiverFABP (FABP1) is highly expressed in the liver and is also found in the intestine, pancreas, kidney and lung. Whereas E-FABP (FABP5), which is one of the most ubiquitously expressed FABP is widely expressed in the skin, tongue, adipocyte, macrophage, dendritic cell, mammary gland, brain, intestine, kidney, liver, lung, heart, skeletal muscle, testis, retina, lens and spleen [146]. Expressions of FABPs are significantly higher in hepatocytes, adipocytes and myocytes, where fatty acids are prominent substrates for lipid biosynthesis, storage, or breakdown compared to amounts found in other tissues [148]. It has been reported that increasing fatty acid also elevated the level of FABP expression in most cell types [149]. These observations indicate that FABPs response to increased fatty acid concentration and trigger the downstream signalling pathways regulating cells behaviours. 1.7.2 Structures of fatty acid binding proteins The structures of FABPs have been investigated by X-ray crystallography, nuclear magnetic resonance and other techniques. The members of FABPs have an abroad range of sequence diversity varying from 17-67% [150]. Despite the differences between FABPs sequences, these proteins show a great similarity in their three-dimensional structures as shown in Figure 1.11. Page | 47 Figure 1.11 Three-dimensional structures of fatty acid binding proteins [146] Ligand-binding FABPs are shown above (The figures were created using PyMOL). (a) FABP1 (Protein Data Bank (PDB) code: 1lfo); (b) FABP2 (PDB code: 2ifb); (c) FABP3 (PDB code: 2hmb); (d) FABP4 (PDB code: 2ifb); (e) FABP5 (PDB code: 1b56); (f) FABP8: (PDB code: 1pmp); (g) FABP6: (PDB code: 1o1v); (h&i) FABP7: (PDB code: 1fe3 & 1fdq). Page | 48 All members of FABPs form a β barrel structure, which consists of two orthogonal β strands with five antiparallel β sheets in each strand. The fatty acid binding site is located inside the β barrel structure capped by helix-loop-helix motif which also connects the two β strands [151, 152]. Within the β barrel structure is a large cavity lined with hydrophobic and polar amino acids [147]. Most FABPs bind only a single fatty acid which is oriented inwards and anchored in the cavity through the interaction with tyrosine and 2 arginine residues [153]. The conserved fingerprint, providing a signature for all FABPs, has recently been revealed as shown in Figure 1.12 (Prints pattern FATTYACIDBP; PR00178). Page | 49 Figure 1.12 Fingerprint for fatty acid binding proteins [146] Three highly conserved motifs are presented with the ribbon diagram. FATTYACIDBP1 (blue ribbon) forms part of first β strands (βA). It also includes a nuclear localization signal (NLS) and a hormone-sensitive lipase (HSL) binding site. The nuclear export signal (NES) domain is located in FATTYACIDBP2 (green ribbon) which includes β sheet 4 and 5 (βE). FATTYACIDBP3 (red ribbon) encodes β sheet 9 (βI) and 10 (βJ). The key amino acids are also marked with blue, green and red respectively. Page | 50 1.7.3 Functional role of fatty acid binding proteins Various functions have been proposed for intracellular FABPs which increase the solubility of fatty acids and facilitate the transport of them to specific organelles, such as to the lipid droplet for storage; to the endoplasmic reticulum for signalling and membrane synthesis; to mitochondria or peroxisome for oxidation; to cytosolic to regulate enzymes activity (reviewed in [146]). Numerous studies also found that FABPs appear to be involved in regulating nuclear transcription factors (PPARs) activities through ligand-dependent translocation to the nucleus. The functional roles of FABPs in intracellular fatty acids are shown in Figure 1.13. Figure 1.13 The functions of FABPs in the cell [146] Page | 51 Growing evidence suggests that FABPs bind to fatty acids and interact directly with PPARs. Further studies also detected the presence of FABPs in the nucleus of different cell types, such as hepatocytes, heart myocytes, cancer cells and cell lines transfected with FABPs using immunofluorescence imaging or imaging of living cells expressing GFP-tagged FABPs [154, 155]. In addition, Tan et.al revealed that C-FABP located into the nucleus in presence of PPARβ, whereas A-FABP undergoes a nuclear localization in response of ligand for PPARγ but not PPARβ [156]. These results indicated that individual FABPs may have unique signalling pathway in nucleus to control the specific gene expression. 1.7.4 Cutanious fatty acid binding protein in prostate cancer Cutanious fatty acid binding protein (C-FABP), also known as FABP5, E-FABP and K-FABP, is a typical member of FABP family with a molecular weight of 15kDa. CFABP was originally isolated from psoriatic skin and located at chromosome 8q21.13. Similar to other FABP members, it binds with high affinity to long chain fatty acid. Our previous studies identified and characterized a metastasis-inducing gene, which is coding for the C-FABP using systematic differential display [157]. Jing et.al also showed that C-FABP is differentially expressed in normal, benign and malignant breast and prostate cell lines and capable to develop metastasis in vivo when transfected CFABP into Rama37 cells which are non-metastatic rat mammary cells [157]. Further investigation on the possible down-steam mediator demonstrated that overexpression of C-FABP significantly increases the expression of VEGF, a crucial factor for tumor growth and expension and also elevated the microvessel density of the primary Page | 52 tumor [158]. The immunocytochemical analysis detected that the level of C-FABP expression is associated with prostate cancer malignant progression. However, no relation between expression level of C-FABP and degree of malignancy has been found as C-FABP expression is not greatly different in between low and high Gleason grades [159]. Suppression of C-FABP expression by antisense transfection in PC-3M cells decreased the invasive capacity in vitro and reduced the tumorigenicity in vivo. Moreover, the suppression of VEGF expression has also been detected in these transfected cell lines [159]. The results above indicate that overexpression of C-FABP may stimulate the expression of VEGF and subsequently promote the angiogenesis to facilitate tumor formation and metastasis. Recent studies by Morgan et.al also confirmed that suppressing the expression of C-FABP in PC-3M cells using siRNA silencing technique resulted the significant reduction of tumorigenicity both in vitro and in vivo [160]. However, how exactly C-FABP elevates the expression of VEGF is still unclear. Page | 53 1.8 Aim of this thesis To investigate whether the fatty acid binding activity is essential for CFABP to promote the tumorigenicity of prostate cancer cells transfected with C-FABP To examine biological behaviours and the fatty acid transporting activity in prostate cancer cells before and after transfected with C-FABP and its mutants To study whether the elevation of C-FABP expression in transfectants increases the expression level of VEGF Page | 54 Chapter 2 Materials and Methods 2.1 Materials 2.1.1 Materials for cell culture Reagents Supplier Cell culture plates and filter cap flasks Nunc Labofuge 400R (Centrifuge) Heraeus Cryogenic vials Nunc DMSO Sigma EndoGRO basal medium Millipore EndoGRO LS supplement kit Millipore Eppendorf tubes (1.5ml) Sarstedt Fetal calf serum Biosera Fluorescence microscope GeneJammer transfection reagent Stratagene Geneticin (G418) PAA Hemocytometer slide Weber scientific international Hydrocotisone Sigma Incubator (Cell culture) Thermo Scientific L-Glutamine BioWhittaker Penicillin/Streptomycin BioWhittaker Phosphate buffered saline (Tablet) Gibco RPMI 1640 PAA Page | 55 Reagents Supplier Sodium pyruvate Sigma Testosterone Sigma Tissue culture pipettes (5ml – 25ml) Greiner bio-one Trypsin Gibco Universal tube (25ml) Greiner bio-one Versene Gibco Water bath Grant Instruments 2.1.2 Materials for fatty acid uptake assay Reagents Supplier BSA Sigma BODIPY 558/568C12 Invitrogen EPICS XL Flow Cytometer Beckman Coulter 2.1.3 Materials for cell proliferation assay Reagents Supplier 96-well plate Nunc MTT Sigma Multiscan plate reader Labsystem Page | 56 2.1.4 Materials for cell invasion assay Reagents Supplier 24-well Boyden chamber plate Boyden chamber Crystal violet Matrigel BD 2.1.5 Materials for soft agar assay Reagents Supplier GelCount Oxford Optronix Low melting point agarose Sigma 2.1.6 Materials for Tumorigenicity in vivo Reagents Supplier Balb/C immuno-incompetent nude mice Matragel Needles and Syringes BD Page | 57 2.1.7 Materials for Histology Reagents Supplier Formalin Tissue-Tek VIP Sakura Embedding cassette Microtome HM355 Mirrom Stainless steel disposable blade Superfrost microscope slides Menzel-Glaster Sodium citrate Sequenza Slide Rack Tween-20 Sodium chloride Tris-base Bovine serum albumin Coverslip (20x40mm) 2.1.8 Materials for measurement of VEGF Reagents Supplier Sodium carbonate 3, 3’, 5, 5’-Tetramethylbenzidine Recombinant human VEGF Phenol red-free RPMI 1640 medium Page | 58 Charcoal stripped FCS VEGF microplate Sulfuric acid HRP-Streptavidin solution Wash solution Biotinylated secondary antibody 2.1.9 Materials for in vitro angiogenesis assay Reagents Supplier In vitro angiogenesis assay kit Human umbilical vein cells EndoGRO basal medium EndoGRO LS supplement kit Crystal violet 2.1.10 Materials for general molecular biology Reagents Supplier Agarose DEPC Ethanol Microcentrifuge Sodium acetate Page | 59 Bromophenol blue Xylene cyanol FF Glycerol Ethidium bromide NanoDrop ND-1000 NanoDrop Technologies Restriction enzymes T4 DNA ligase 2x quick Ligation buffer 2.1.11 Materials for transformation of competent bacteria Reagents Supplier DH5 E.coli bacteria Bactotryptone Yeast extract NaCl KCl LB agar MOPS CaCl2 Glycerol MgCl2 MgSO4 Glucose Page | 60 Ampicillin Kanamycin LB broth 2.1.12 Materials for RT-PCR Reagents Supplier 10x PCR buffer Taq DNA polymerase dNTPs Thermocycler DTT Oligo (dT)20 RNaseOUT 5x first strand buffer SuperScriptIII reverse transcriptase 2.1.13 Materials for Western blot Reagents Supplier CelLytic-M Sigma Roller mixer Dye reagent Bio-Rad Bradford reagent Page | 61 Nitrocellulose membrane Coomassie blue Protoblock reagent A &B ACTG (National diagnostics) Page | 62 2.2 Methodology 2.2.1 Cell culture 2.2.1.1 Routine cell culture All routine cell culture manipulations took place in a tissue culture hood located in the Tissue Culture Laboratory. Cells were cultured in 75cm2 cell culture flasks and maintained at 37oC in a humidified incubator which contains 5% (v/v) CO2. LNCaP cells were cultured in RPMI1640 nutrient medium supplemented with 10% (v/v) Foetal calf serum (FCS), 2mM L-glutamine, 1mM Sodium pyruvate, 100IU penicillin/streptomycin (PEN-STREP), 5µg/ml 5-dihydroxytestosterone and 5µg/ml hydrocortisone. HUVEC cells were cultured in EndoGRO basal medium (EBM) with EndoGRO LS supplement kit. For transfected cell lines, geneticin (G418) was added to the previous components gave the final working concentration of 1mg/ml. Both routine and experimental culture media were replaced every three days. 2.2.1.2 Thawing of cells Cryogenic vials of cells were defrosted in a water bath at 37oC and immediately transferred into a universal tube with 20ml of complete cell culture medium. The cells were then pelleted by centrifugation at 900rpm lasting for 3 minutes. Then, the supernatant was discarded and the pellet was resuspended in a suitable volume of complete medium contained with 10% FCS. Finally, cells were plated out in a 75cm2 cell culture flask and maintained within a controlled atmosphere of 5% CO2 at 37oC. Page | 63 2.2.1.3 Sub-culture of cell lines Cells that had grown to approximately 60-80% confluence were passaged to allow continuous growth. Medium was completely removed from the flasks. Then, growing cells were washed by phosphate buffered saline (PBS) and incubated at 37 oC with 3.5ml of 2.5% (v/v) trypsin/versene solution for a period 6 minutes or until the cells had rounded up and were starting to detach from the culture dish. The trypsin/versene solution was inactivated by adding 10 ml of complete medium. The cells were then transferred to a universal tube and centrifuged at 900rpm for 3mins. The supernatant was removed and cells were resuspended in cell culture medium. The cells were aliquotted at the required concentration and maintained within a controlled atmosphere of 5% CO2 at 37oC. 2.2.1.4 Counting cells The number of cells was calculated by using an improved Neubauer double counting chambers hemocytometer. Cells were detached as described previously. Mix cell suspension thoroughly and load 10µl of cell suspension to a counting chamber of hemocytometer with 9 (3x3) squares. Then, count cells in four corner squares under the microscope. The total number of cells is calculated using the equation listed below: Total number of cells = Average cell count per square x Dilution x 104 x Total volume (ml) Page | 64 2.2.1.5 Freezing cells Cells were selected for freezing when they were approximately 60-80% confluence. Before freezing, the cell culture medium was replaced with fresh medium for 24 hours. The cells were detached as described in 2.2.1.3 and resuspended in 10ml of completed medium. Then, the total number of cells was counted as mentioned in 2.2.1.4. The cells were re-centrifuged at 900 rpm for 3 minutes. The supernatant was removed and the cell pellet was broke up by repeated pipetting with freezing medium (Complete medium with 7.5% (v/v) DMSO) to give a concentration of 1x106 cells/ml. Aliquots (1ml) of the cell suspension were pipetted into cryogenic vials and placed into a nalgene cryo-preserver box containing 250ml of isopropyl alcohol. The box stored in a freezer at –80oC overnight before cryogenic vials transferred for long term storage in liquid nitrogen. 2.2.2 Cell transfection 2.2.2.1 Stable transfection of the LNCaP cell line using GeneJammer A polyamine based method of transfection (GeneJammer) was used to insert pIRES2-EGFP into LNCaP cells. Prior to transfection, LNCaP cells (2x105) were seeded in a 6-well tissue culture plate and grew the cells overnight in complete LNCaP culture medium to reach 50-70% confluence. In the mean time, 97µl of sterile, room temperature (RT), serum-free, antibiotic-free RPMI medium was transferred to a 1.5ml eppendorf tube and added with 3µl GeneJammer transfection reagent. After incubation for 5mins at RT, 2µg of DNA was added to the diluted GeneJammer transfection reagent and mixed gently. The mixture was incubated at RT for 30 minutes and added dropwise to the wells of the 6-well tissue culture plate. Then, the transfection mixture Page | 65 was distributed evenly by rock the plate back and forth. The control transfections were also performed as shown in Figure 2.1. After incubation in standard growth conditions for 48 hours, cells were split into 5 separate 9cm cell culture plate and cultured in selective medium containing 0.5mg/ml Geneticin for 10 days. The Geneticin was removed and the surviving transfected clones were allowed to recover in routine medium for one week. Figure 2.1 Three mock-transfection controls were performed by putting cells through the transfection procedure without adding GeneJammer transfection reagent or DNA or both of them. Page | 66 2.2.2.2 Transfected clone selection by cell image Cells were then re-cultured in selective medium for a further 3-4 weeks until the cells in control transfection died out and healthy cell clones had formed in the cell culture plate. As pIRES2-EGFP vector (as presented in Figure 2.2) contained the enhanced green fluorescent protein (EGFP) coding region, successfully transfected cell colonies with the expression of fluorescent marker were detected under a fluorescence microscope. Because the proportion of cells transfected within a population is relatively low (less than 10%), the ring cloning methodology was used to isolate colonies. For each batch transfectants, 5 colonies with best expression of EGFP were marked on the down-side of 9cm plate for ring cloning. Figure 2.2 Circular maps and lists of features for the pIRES2-EGFP vector Page | 67 2.2.2.3 Ring cloning of transfected cells The tops of 1ml pipette tips were cut to larger the cloning rings which were sterilised with silicon grease and forceps by autoclaving. The selective medium from the transfectant plates was removed and the plates were washed with PBS. Using forceps, the cloning rings were dipped in silicon grease and placed over the colonies. The small amount of silicon grease allowed the rings to stick to the plate and formed a watertight seal around the colony. Then, 100µl of 2.5% (v/v) trypsin/versene solution was added to isolated colonies and incubated for 6 minutes or until the cells were rounded-up. The detached cells were transferred to a 1.5ml eppendorf tube and the trypsin inactivated by adding an equal volume of routine cell culture medium. The cells were centrifuged for 3 minutes at 900rpm. After remove of supernatant, the cell pellet was resuspended in 1ml selective culture medium containing 0.5mg/ml geneticin and plated into a 24-well plate and maintained within a controlled atmosphere of 5% CO2 at 37oC. 2.2.3 Fatty acid uptake assay Fatty acid uptake assay is designed in order to test long chain fatty acid (LCFA) uptake kinetics in different transfectant cells containing fatty acid transporters. Conventional protocols utilizing radioactivity often require cell lysis and processing at very low temperature. Whereas, uptake of fluorescently-labeled fatty acids requires neither radioisotope usage nor cell lysis and also the fluorescence intensity in each individual cell can be measured by flow cytometry precisely. The red fluorescence-labeled LCFA, BODIPY 558/568C12 (4, 4-difluoro-5-(2-thienyl)-4-bora-3a, 4a-diaza-s-indacene-3dodecanoic acid), was used as the analogue of natural LCFA in the fatty acid uptake assay. BODIPY analogue behaves much like natural fatty acids: it becomes activated by Page | 68 acyl-CoA attachment; is incorporated into triglycerides; and accumulates in intracellular lipid droplets. In addition, the BODIPY analogue is also known as the substrate for fatty acid transporters since its uptake by adipocytes and can be competed by natural LCFA. Cultured cells (1x105 cells) in 2ml routine culture medium were seeded into 6-well plates and incubated at 37oC, 5% CO2 overnight. After replacing the medium with 2ml of solution containing 25μg BODIPY 558/568C12 in 200μm BSA/PBS, the cells were incubated for 30mins at 37oC, 5% CO2. The fatty acid uptake was stopped by removal of BODIPY fatty acid solution followed by addition of 3ml of an ice-cold stop solution (PBS containing 0.5% BSA). The stop solution was discharged after 2 min and the culture plates were washed another two times by fresh ice-cold stop solution. Then, cells were detached using 2.5% (V/V) trypsin/versene at 4oC and the fluorescence intensity of each cell line was measured with an EPICS XL Cytometer at the wave length of 570 nm to assess the fatty acid uptakes. Page | 69 2.2.4 Cell proliferation assay 2.2.4.1 Preparation of growth curve Cells were grown to 60-80% confluence in 175cm2 flasks and harvested as previously outlined and resuspended in 10ml of complete culture medium. Each cell line was counted using a hemocytometer as previously mentioned in section 2.2.1.4 and was made up in 5x105/ml in 4mls of complete culture medium. A standard growth curves was prepared in serial dilution at: 6.25x103/ml, 1.25x104/ml, 2.5x104/ml, 5x104/ml, 1x105/ml, 2.5x105/ml and 5x105/ml. Then, 200µl of cell suspension from each dilution was plated into a 96-well plate in triplicate. Each cell line requires its own standard curve. In the mean time, the cells for proliferation assay were also prepared at the concentration of 5x104/ml. Similar as preparation of standard curve, 200µl cell suspension containing 1.25x104 cells was plated into five separate 96 well plates in triplicate. After overnight incubation with 5% CO2 at 37oC, the cell growth of the standard curve and first experimental day analysed by the MTT assay as described in section 2.2.4.2. The cell proliferation assay was set to run for five days at each plate per day. 2.2.4.2 Assessing cell numbers using MTT assay MTT is a yellow coloured chemical and it entered into the mitochondria of the cells. In mitochondria, MTT is oxidised into an insoluble blue dye known as formazan which forms blue crystals in the cytoplasm. The crystals are dissolved by adding DMSO and leave a coloured liquid, which is directly proportional to the number of cells present. Page | 70 The optical density (OD) of the cell population colour is measured accurately with a Multiscan plate reader at 570nm. For each cell line, a standard curve was constructed by plotting OD (570nm) against the number of cells. The cell number of the test samples was determined by extrapolating from the standard curve. MTT stock solution is prepared at a concentration of 5mg/ml (100mg MMT in 20ml PBS) and stored at 4oC. At each time point (every 24 hours), 50l of MTT stock solution was added to each well in a 96-well plate and the cells were incubated at 37oC, 5% CO2 for 4 hours. The cell culture medium and MTT is removed gently to avoid disturbing cells followed by adding 200l of DMSO to each well including blank control wells mixed by pipetting up and down. The plate was incubated for a further 10 minutes at 37oC, 5% CO2 and read at 570nm in the optical density plate reader. The growth rate for each cell line is quantified against its own standard curve to obtain the number of cells/well in each day for five days. Graphs of cell number/time were plotted. 2.2.5 Cell migration assay in Boyden chamber system The Boyden chamber system was used to examine the difference in invasive ability in cell lines expressing wild-type and mutant C-FABPs. The results are often variable, as many factors are involved in the assay, including the number of cells added to the upper chamber, the duration of the assay and the way to measure the number of the cells [161]. Therefore, it is important for the system gives reproducible results by setting up an optimized procedure considered the variable mentioned above. All materials used in the section can be found in section 2.1. Page | 71 Cells to be used in invasion assay were cultured in flasks until 60-80% confluent. Cell invasion assays were carried out using a modified Boyden chamber system. Wells of the 24-wells plate were divided into upper and lower compartment by 6.5mm diameter polycarbonate membrane containing 8µm pores. The polycarbonate filters were wetted with 200μl RPMI medium for 30mins and the medium was removed followed by adding further 100μl RPMI medium containing 30μg of matrigel to the upper side of the filters. The plates were incubated at 37oC, 5% CO2 for 2 hours to allow the matrigel solidification. The assays were set up with 1x105 cells in the upper compartment in 200μl of 2% (v/v) FCS growth medium and 500μl of 10% (v/v) FCS growth medium was placed to the lower chamber. After 24 hours incubation at 37oC, 5% CO2, the cells remaining in the upper side of the filters were removed gently by cotton swabs and the cells attached to the lower side of the filter were fixed and stained by crystal violet reagent. None of cell was found in lower compartment indicated that all cells invaded through the matrigel coated filter were attached to the lower side of the filter. The numbers of the invaded cells were counted under microscope. The results were finally normalized by the cell number in the 24-well plate after 20 hours incubation because the cells after typsinisation may vary in viability. In this way, the cell motilities measured from different cell lines were more comparable. The results are the mean ± SD of three separate experiments. Page | 72 2.2.6 Soft agar assay Soft agar assay is designed in order to examine the tumorigenicity of each cell lines in anchorage-independent environment. The assay was carried out in 6-well plates which were pre-coated with 2ml of 2% (w/v) low melting point agarose in routine culture medium with 10% (v/v) FCS and the mixture solidified in refrigerator at 4oC for 20mins. The cells were routinely grown to 60-80% confluent in flasks, detached and counted as previously described in section 2.2.1.3 and 2.2.1.4. Then, the cells were washed once by centrifugation and resuspended in routine culture medium with 10% (v/v) FCS at the concentration of 1x104/ml. The cell suspension was mixed with 2% low melting point agarose in a ratio of 1:1 and 2ml of mixtures were seeded to the precoaded wells to make the final cell concentration at 1x104 per well. The plates were placed at 4oC until solidified. Once the top layer was set, the 6-well plates were placed in the incubator at 37oC, 5% CO2 for 4 weeks. During this period, 2 drops of routine culture medium were added to each well twice a week to keep the soft agar moist. At the end of the soft agar assay, colonies were dyed by adding 2ml of MTT (5mg/ml) followed by incubation at 37oC, 5% CO2 for 4 hours. Colonies larger than 150μm in diameters were counted by the GelCount. Page | 73 2.2.7 Tumorigenicity in vivo The in vivo tumorigenicity of the transfectant lines were explored using Balb/C immuno-incompetent nude mice (4-6 weeks old). The experiments were carried out using four groups of mice (8 animals in each group) to assess the level of tumorigenecity in LNCaP cells transfected with wild-type C-FABP, two mutant C-FABPs and pIRES2-EGFP vector only. All Balb/C immuno-incompetent nude mice were housed in the University of Liverpool animal unit and experiments were conducted in accordance to UKCCCR guidelines under Home Office Project Licence PPL 40/2963 to Professor Y. Ke. The cells for the in vivo tumorigenecity assay were grown up to ~80% confluence in 175cm2 flasks and harvested as described earlier in section 2.2.1.3. Then, the cells were centrifuged at 900rpm for 3mins and resuspended in PBS at the concentration of 1x107/ml. The cell suspension was mixed with matrigel in a ratio of 1:1 and kept on ice during transport to the animal unit. The cells were inoculated subcutaneously into right flank of the nude mice at the density of 2x106 cells in 200l PBS/matrigel (1:1) mixture using a 1ml syringe. Tumor volume was measured every three days and calculated by the following formula: Length x Width x Height x 0.5236 [162] and the primary tumour measurement was taken by weighing the tumour at autopsy. Page | 74 2.2.8 Histology 2.2.8.1 Dissection of nude mouse tissues The nude mice were examined every three days and autopsy after 60 days or when primary tumor volume exceed one third of the body size of nude mice. Samples of the primary tumor, liver, pancreas, lungs, kidneys, lymph nodes and any other tissues of abnormal appearance were removed from the animals and fixed in 10% neutral buffered formalin (NBF) for 24 hours before processed routinely for histological evaluation. 2.2.8.2 Embedding in paraffin wax and sectioning The fixed tissue samples were trimmed and placed within an embedding cassette and processed on a Tissue-Tek VIP5 processor. The processed tissue was then embedded in paraffin wax at 60οC and cooled on ice. After solidification, the blocks were sectioned at RT on a Microm HM355 microtome using a microtome knife containing a stainless steel disposable blade. Sections were cut to 4μm thickness, placed on 3-aminopropyl triethoxy-saline (APES) coated Superfrost microscope slides and incubated at 37οC overnight prior to processing with Haematoxylin and Eosin staining using a Varistain 24-4 automated stainer or immunohistochemistry. 2.2.8.3 Immunohistochemistry Immunohistochemistry is a technique that widely used in research to understand distribution and localization of biomarkers and differentially expressed proteins in different parts of a tissue and it enables the localization and visualization of tissue antigens in-situ by the use of specific antibodies. Firstly, specific antibodies are applied to bind to the tissue antigen. Secondly, antibody-antigen complex is bound by a horse Page | 75 radish peroxidise (HRP)-conjugated secondary antibody. Finally, the enzyme-labeled complex can be visualized by the use of the 3, 3’-Diaminobenzidine (DAB) which produces a brown-colored precipitate in the presence of HRP. The step-by-step immunohistochemical staining procedures are listed in Table 2.1. Table 2.1 Step 1 Step 2 The immunohistochemical staining procedures The mounted sections were de-waxed in two 5mins changes of xylene. Shake off excess liquid and dehydrate slides in two changes of fresh absolute ethanol for 1min each. Step 3 Slides were incubated in 3% (v/v) hydrogen peroxide/methanol for 12mins to quench endogenous peroxidise activity. Slides were rinsed in running water for 3mins and antigen retrieval was Step 4 carried out by microwaving tissue sections for 15mins at full power in 10mM sodium citrate buffer (pH 7.6). Step 5 Slides were allowed to cool down for further 15mins and placed in running water before being racked into sequenza slide rack. 1. Slides were washed twice by Tris-Buffered Saline Tween-20 (TBST) and Step 6 incubated with primary antybody in 200µl of 1% Bovine serum albumin (BSA)/TBST for 1 hour. Negative controls were incubated with BSA/TBST alone for the same period of time. Step 7 Slides were washed three times using TBST and incubated in 200 µl of Envision FLEX/HRP for 20mins. Slides were washed three times using TBST and incubated in 200 µl of Step 8 EnvisionTm FLEX DAB+ chromogen mixed with EnvisionTm FLEX substrate (1dop/ml) for 20mins. Page | 76 Step 9 Slides were removed from sequenza slide rack and washed in running tap water for 1 min. Step 10 Slides were counterstained in Haematoxylin solution for 50 seconds and washed briefly in running tap water. Step 11 Slides were dipped in 1% acid alcohol for five times and washed briefly in running tap water. Step 12 Slides were incubated in Scott’s tap water blueing agent for 30 seconds and washed thoroughly in running tap water. Step 13 Slides were dehydrated in five changes of absolute methanol (30 seconds each change). Step 14 Slides were cleared of any remaining water by incubating in two changes of Xylene (30 seconds each change). Step 15 Slides were mounted with cover slips (20x40mm) using DPX synthetic resin. Step 16 The stained sections were observed by a light microscope and graded as negative, weakly positive, moderately positive and strongly positive. Page | 77 2.2.8.4 The scoring system used for immunohistochemical staining The immunohistochemical scoring system used in this study was a method devised by Remmele et.al [163]. The intensity and distribution of positive staining was evaluated two score system by three observers. The first evaluation employed a four-point scale for intensity with slides being scored as 0, 1, 2 and 3 as listed below: 0 = no positive staining observed 1 = predominantly weak staining observed 2 = predominantly moderate staining observed 3 = predominantly strong staining observed Eight random fields of each slide were counted at x40 magnification according to a positive control sample and a negative control sample. The second evaluation employed a five-point scale for distribution with slides being scored as 0, 1, 2, 3 and 4 as listed below: 0 = Less than 10% of the cells positively stained 1 = 10-29% of the cells positively stained 2 = 30-49% of the cells positively stained 3 = 50-79% of the cells positively stained 4 = 80-100% of the cells positively stained The scores from both intensity and distribution were multiplied to give a final score for grading which is listed below: (-) = 0 (negative expression of the protein) (+) = 1-4 (weak expression of the protein) (++) = 5-8 (moderate expression of the protein) (+++) = 9-12 (strong expression of the protein) Page | 78 2.2.9 Immunosorbent assay for the measurement of VEGF The quantitative measurements of human VEGF expression in LNCaP transfectants were carried out using the RayBio human VEGF ELISA (enzyme-linked immunosorbent assay) kit. This ELISA assay employs an antibody specific for human VEGF coated on a 96-well plate. Standards and samples are pipette into the wells and VEGF is bound to the wells by the immobilized antibody. The well are washed and incubated with biotinylated anti-human VEGF antibody. After washing away unbound biotinylated antibody, the HRP-conjugated streptavidin is added to bind to the secondary antibody. Finally, 3, 3’, 5, 5’-Tetramethylbenzidine (TMB) is pipetted to the wells and color develops in proportion to the amount of VEGF bound. 2.2.9.1Preparation of standard curve The recombinant human VEGF was diluted to a concentration of 18,000pg/ml standard solution in a diluent, 50mM sodium carbonate (pH 7.6). Then, 400µl standard solution was pippeted into a tube to produce a dilution series as shown in Figure 2.3. The diluent serves as the zero standard (0pg/ml). Figure 2.3 Serial dilutions for standard curve Page | 79 2.2.9.2 Human VEGF ELISA The cells were grown up to ~80% confluence in flasks, detached and counted as described in section 2.2.1.3. Then, 1x106 cells were seeded to 75cm2 culture flasks and cultured in growth factor deprived medium (phenol red-free RPMI 1640 medium containing 10% charcoal stripped FCS) for 48 hours. The conditioned medium (cell culture supernatant) of each transfected cell lines was collected and 100µl of each standard (see Figure 2.2) and sample was added into VEGF microplate wells. The plate was covered with the lid and incubated over night at 4oC with gentle shaking. On the following day, the conditioned medium was removed completely and the plate was washed four times with 300µl of wash solution provided by the VEGF Elisa kit. Then, the biotinylated secondary antibody (100µl) was added to each well and incubated for 1 hour at RT. After removing the unbound second antibody, the plate was washed as mentioned above and 100µl of HRP-Streptavidin solution was added to the wells and incubated for 45mins at RT. Discard the solution and repeated the wash step as mentioned above. Finally, 100µl of TMB substrate reagent was pipetted to each well to visualize the level of VEGF and incubated for 30mins at RT with gentle shaking followed by adding 50µl of 2M sulfuric acid to stop the reaction. The plate was then read at 450nm in the optical density plate reader immediately. The expression level of VEGF secreted in the conditioned medium for each cell line is quantified against the standard curve to obtain the concentration of VEGF (pg/ml). The results are the mean ± SD of three separate experiments. Page | 80 2.2.10 In vitro angiogenesis assay In vitro angiogenesis for conditioned medium from each tansfectant was assessed using the Millipore in vitro angiogenesis assay kit which provided a timesaving system for evaluation of tube formation by endothelial cells. When cultured on ECMatrix, a solid gel of basement proteins prepared from the Engelbreth-Holm-Swarm (EHS) mouse tumor cells, these endothelial cells could achieve fully developed tube structures between 4-8 hours under the effect of pre-angiogenic factors. 2.2.10.1 Preparation of the ECMatrix coated plate ECMatrix was thawed on ice overnight prior to the in vitro angiogenesis assay and then mixed with 10 x dilution buffer in a sterile microfuge tube. The tip of pipette tip was cut off by a sterile knife to ease pipetting and avoid pipetting air into the solution as ECMatrix is highly viscous. The diluted ECMatrix (50µl) was transferred to each well of a 96-well tissue culture plate. All equipments such as pipette tips, plates and tubes were pre-cooled and all procedures were carried out in a cold room. The ECMatrix coated plate was incubated at 37oC for 2 hours to allow the matix solution to solidify. 2.2.10.2 Angiogenesis assay Human umbilical vein cells (HUVEC) were grown to 60-80% confluence in 75cm2 culture flasks in EndoGRO basal medium added with EndoGRO LS supplement kit. The HUVEC were harvested as described in section 2.2.1.3 and resuspended in EndoGRO basal medium containing 5% charcoal stripped FCS. Then, 100µl of 1 x 104 HUVEC were seeded onto the surface of the polymerized ECMatrix of each well followed by adding 100 µl of conditioned medium from each transfected cell line. The Page | 81 assay plate was incubated at 37oC, 5% CO2 for 6 hours. At the end point of the assay, cell-tubes were visualized by adding 50µl of crystal violet (0.5% crystal violet in a solution of 50% ETOH/PBS containing 5% formaldehyde) at RT for 10mins and quantified using light microscope in 40X magnification. 2.2.10.3 Quantitation of tube formation Tube formation is a multi-step process, stating with cell migration and alignment, followed by the development of capillary tubes, sprouting of new capillaries and finally the formation of the cellular networks. There are several measurements for quantitation of tube formation such as branch point counting and total capillary tube length measurement. However, the pattern recognition quantitation method works best in in vitro angiogenesis assay involving activators of angiogenesis. A numerical value to each pattern was allocated as presented in Table 2.2. Five random view-fields per well were examined and the values averaged. Table 2.2 The degree of angiogenesis progression and its numerical value Page | 82 2.2.11 General molecular biology methods 2.2.11.1 Calculation of DNA and RNA concentration and purification The DNA or RNA samples were examined for concentration and quality using the NanoDrop ND-1000 spectrophotometer. Nucleic acid sample (1µl) was loaded onto the lower measurement pedestal. After closing the sampling arm, the spectral measurement was initiated using the operating software on the computer. The DNA or RNA sample concentration was presented on the screen in ng/µl based on absorbance at 260nm using the modified Beer-Lambert equation. The purity of the sample was also assessed using ratio of absorbance at 260nm and 280nm: a ratio of ~1.8 was generally accepted as pure for DNA; a ratio of ~2.0 was generally accepted as pure for RNA. 2.2.11.2 Ethanol precipitation of DNA and RNA The DNA or RNA to be precipitated was mixed with 2.5 volumes of chilled state temperature absolute ethanol and 10% volumes of 3M sodium acetate (pH5.2). The mixture was mixed gently by inversion and incubated at –80oC freezer overnight for DNA or RNA to precipitate. The precipitated DNA or RNA was pelleted by centrifuge at 16,000 rpm at 4oC in a microcentrifuge for 20 mins. The supernatant was removed and the pellet was washed by 100μl of pre-chilled 70% (v/v) ethanol. The DNA or RNA was recentrifuged at 16,000rpm at 4oC for 15 mins and the supernatant was decanted. The DNA or RNA was air dried and resuspended in sterile DEPC-treated water. Page | 83 2.2.11.3 Agarose gel electrophoresis of DNA Agarose gel electrophoresis was carried out in 1xTBE buffer (0.089M Tris-base, 0.089M Boric acid, 0.002M EDTA pH8.0). The agarose concentration of the gel was 0.8% to 2.0% (w/v) according to DNA size. The agarose was dissolved in 200ml of 1xTBE buffer by microwave heating for 2mins and cooled to 40-50oC. Then, 10µl of 500ng/ml ethidium bromide was added for visualisation. The gel mixture was poured and placed at 4oC to solidify. DNA samples were prepared by adding 5x agarose gel loading buffer [0.25% (w/v) bromophenol blue, 0.25% (w/v) xylene cyanol FF, 30% (v/v) glycerol in water] to a 1x concentration and loaded into the gel alongside DNA size marker. The gel was run for 50mins at 90V and visualised on a UV transilluminator. 2.2.11.4 Purification of DNA from agarose gels DNA fragments were separated using agarose gel electrophoresis as described previously in section 2.2.11.3. The required fragments were visualised by ethidium bromide and excised using a sterile blade. Each excised gel slice was weighed and three times the volume of buffer QG was added. Then, the mixture was dissolved in a 50oC water bath for 10 mins with occasional vortex mixing. The melted agarose and DNA solution was applied to a QIAquick spin column and centrifuged for 1 minute at 16,000 rpm in a microcentrifuge and the flow through was discarded. The membrane, bound with DNA, was washed by passing through 750l of buffer PE and re-centrifuged for a further minute to remove any residue. The DNA was eluted from the column by adding 30μl of distilled water and quantified as described above. The DNA was stored at –20oC until needed. Page | 84 2.2.11.5 Restriction Enzyme digestion of plasmid DNA Plasmid DNA was digested with two different restriction enzymes (XhoI and PstI) to check for the presence of insert or to produce the DNA fragment for ligation. The restriction enzyme digest mixture was prepared as shown in Table 2.3: Table 2.3 Water 35µl 10x NEBuffer 3 5µl XhoI 20units/µl 1µl PstI 20units/µl 1µl BSA 10µg/µl 5µl DNA 1µg/µl 3µl The restriction enzyme digestion mixture (total volume 50µl) The restriction enzyme digest mixture was incubated at 37oC for 2 hours to allow complete digestion of the template DNA. Then, the enzyme reaction was inactivated by heating to 68oC for 10 minutes and the products were analysed by gel electrophoresis. Page | 85 2.2.11.6 DNA ligation Fragments of DNA were removed from pBluescript vector using restriction enzyme digestion as described in section 2.2.11.5 and inserted into transfection vector pIRES2-EGFP by ligation reaction using T4 DNA ligase. The ligation mixture was prepared as shown in Table 2.4. Insert DNA (180ng/µl) 2µl Vector DNA ( 110ng/µl) 1µl 2x quick Ligation buffer 10µl T4 DNA Ligase 1µl ( 400 cohesive end units /µl) Table 2.4 H2 O 6µl Total volume 20µl The ligation reaction mixture After overnight incubation at 4oC, 5µl of ligation mixture was transformed into competent bacterial cells as described in section 2.2.12. The recombinant DNA was recovered from bacterial cells using a Qiagen miniprep or midiprep kit as described in section 2.2.13. Page | 86 2.2.11.7 cDNA sequencing The recombinant plasmid DNAs or cDNA sequences generated by reverse transcription polymerase chain reaction (RT-PCR) were sent for sequencing at the University of Liverpool, School of Tropical Medicine. Sequencing was carried out by using a big dye terminator cycle sequencing reaction kit with an ABI prism 377 automated DNA sequencer. Sequence alignment was determined by using BioEdit for searching Genbank (BLAST). 2.2.12 Transformation of competent bacteria with plasmid DNA 2.2.12.1 Preparation of competent bacterial cells E.coli (DH5) glycerol stock was streaked onto LB-agar plate and incubated at 37C overnight. On the following day, a single colony was inoculated into a conical flask containing 10mls LB broth and incubated at 37oC overnight with shaking at 225rpm. Then, 1ml of overnight culture was transferred into 100ml of SOB medium (2% w/v bactotryptone, 0.5% w/v yeast extract, 10mM NaCl, and 2.5mM KCl) and incubated at 37C with shaking at 225rpm until the OD550 reached 0.4. The culture solution was split into 8x 12ml aliquots and placed on ice to cool for 10mins. The DH5 bacteria were pelletted by centrifuging at 2500rpm for 10mins at 4oC and the supernatant was discarded. The bacterial cell pellets were resuspended in 8.25ml of pre-cooled RF1 buffer (100mM KCl, 50mM MgCl2.4H20, 30mM KCl, 10mM CaCl2 and 15% v/v glycerol, pH6.8) and incubated on ice for 10mins. After a further centrifuging at 2500rpm for 10mins at 4oC, the cell pellet was resuspended in 2ml of RF2 buffer Page | 87 (10mM MOPS, 10mM KCl, 75mM CaCl2.2H2O and 15% v/v glycerol, pH6.8). The DH5 bacteria solution was dispensed into 1ml aliquots in cryovials, frozen in liquid nitrogen and transferred immediately to –80oC freezer for storage. 2.2.12.2 Transformation A vial of DH5 competent cells was removed from the -80 C freezers and thawed on ice. Then, 50ng of plasmid DNA or 10μl of the ligation mixture was added to 50µl of competent DH5 cells, mixed by flicking gently and incubated on ice for 30mins. The cells were heat shocked at 42C for exactly 90 seconds and placed on ice for a further 2mins followed by adding 800l of SOC (2% w/v bactotryptone, 0.5% w/v yeast extract, 10mM NaCl, 10mM MgCl 2, 10mM MgSO4, 2.5mM KCl, 20mM glucose) and incubated 37C for 1 hour in a shaking incubator at 225rpm. Transformed bacteria solution (200l) was plated onto LB plates containing 50µg/ml ampicillin or 30µg/ml kanamycin for antibiotic selection and incubated at 37C overnight. Finally, colonies were picked from the plate and grown in LB broth containing antibiotic at 37C, 225rpm overnight and plasmid minipreps were conducted. Page | 88 2.2.13 Isolation of plasmid DNA The isolation of plasmid DNA was conducted using extraction kits provided by Qiagen. 2.2.13.1 Miniprep extraction of plasmid DNA The overnight cultured bacteria (5ml) in LB medium containing antibiotic was harvested by centrifuging at 6,000rpm for 1min and the supernatant was removed. The cell pellet was resuspended in 250l of cell suspension buffer P1 (50mM Tris-HCl pH8.0, 10mM EDTA and 100μg/ml RNaseA) by vortexing, followed by adding 250l of cell lysis buffer P2 (200mM NaOH and 1% w/v SDS) and mixed by gentle inversion of the microcentrifuge tube six times. Then, 350l of neutralisation buffer N3 (4.2M Gu-HCl, 0.9M potassium acetate pH4.8) was added and mixed by gentle inversion of the microcentrifuge tube six times. The mixture was centrifuged at 13,000rpm for 10mins and the supernatant was loaded to a miniprep spin column. The column was centrifuged at 13,000rpm for 1min and the flow through was discarded. The column was then washed by adding 750l of wash buffer PE (10mM Tris-HCl pH7.5 and 80% ethanol) and centrifuged at 13,000rpm for 1min. After removing the flow through, the spin column was centrifuged for an additional minute to remove residual wash buffer. The column was placed in a clean 1.5ml microcentrifuge tube and added with 30l of distilled water to the centre. After 1min incubation at RT, the plasmid DNA was eluted by centrifuging at 13,000rpm for 1min. The concentration and quality of the DNA were examined as described in section 2.2.11.1 and isolated plasmid DNA was stored in a 20C freezer. Page | 89 2.2.13.2 Midiprep extraction of plasmid DNA A single colony of transformed DH5 bacteria was transferred into flasks containing 10ml of LB medium with antibiotic and left to grow overnight at 37C, 225rpm. The overnight cultured bacterial cells (5ml) were then transferred to 200ml LB medium containing antibiotic and incubated for 5 additional hours at 37C, 225rpm. The transformed bacterial cells were harvested by centrifuging at 6,000rpm for 10min at 4 oC and the supernatant was removed. The cell pellet was resuspended in 4ml of buffer P1 (50mM Tris-HCl pH8.0, 10mM EDTA, 100μg/ml RNase A) by vortexing until no clump was remained. The bacterial cells were lysed by adding 4ml of lysis buffer P2 (200mM NaOH, 1% w/v SDS), gently mixed by six times inversion. After incubation at RT for 5mins, 4ml of pre-chilled P3 neutralising buffer (3.0M potassium acetate pH5.5) was added and the mixture was mixed by inverting six times, incubated on ice for 15mins and then centrifuged at 16,000rpm, 11C for 30mins. The supernatant containing plasmid DNA was transferred to a clean 30ml centrifuge tube and recentrifuged at 16,000rpm, 11C for further 15mins. In the mean while, a Qiagen midiprep column was equilibrated by applying 4ml of equilibration buffer QBT (750mM NaCl, 50mM MOPS pH 7.0 and 15% v/v isopropanol) and allowed to empty by gravity flow. Once the Qiagen midiprep column was equilibrated, the cleared supernatant containing plasmid DNA was loaded into the column and allowed to flow under gravity. The column was washed twice with 10ml of wash buffer QC (1.0M NaCl, 50mM MOPS pH 7.0, 15% v/v isopropanol) and the plasmid DNA was eluted from the column using 5ml of elution buffer QF (1.25 M NaCl, 50mM Tris-HCl pH 8.5, 15% v/v isopropanol). The eluted DNA was precipitated by adding 3.5ml of isopropanol, mixed and centrifuged at 4C, 11,000rpm for 30mins. The supernatant was Page | 90 carefully removed and the pellet was washed with 2ml of 70% (v/v) ethanol and centrifuged at RT, 11,000rpm for 10mins. After removing the supernatant, the pellet was air-dried for 10mins and the plasmid DNA was then dissolved in 200l of nucleasefree water. The concentration and quality of the DNA were examined as described in section 2.2.11.1 and the isolated plasmid DNA was stored at -20C 2.2.14 Mutagenesis of C-FABP gene and construction of CFABPs transfection vectors In previous works, a 436 base pair fragment of the protein coding region of C-FABP had been amplified by RT-PCR from total mRNA isolated from PC-3M cells. The amplified C-FABP product was T-A cloned into a PGEM-Teasy plasmid and then transferred into pBluescript II SK vector as shown in Figure 2.4. Transformertm SiteDirected Mutagenesis Kit was used to generate one or two point mutations to the CFABP gene. The transformer mutagenesis strategy, as shown in Figure 2.5, can not only introduce single or multiple specific base mutations with high efficiencies (70-90%) but also benefit by avoiding subcloning and using single-stranded vectors or specialized double-stranded plasmids. Furthermore, the use of T4 DNA polymerase instead of PCR reduces the risk of generating spurious mutations. Firstly, the mutagenic primer and the selection primer containing a mutation in a restriction enzyme site are annealed to one strand of the denatured doublestranded plasmid. Page | 91 Secondly, after DNA extension, ligation and a primary selection by restriction digest the mixture of mutated and unmutated plasmids is transformed into a mutS E.coli strain defective in mismatch repair. Thirdly, the DNA was isolated from pooled transformants and digested with the selective restriction enzyme. Finally, the selective digested DNA was transformed into DH5α bacterial cells and the mutated plasmid was recovered by midiprep extraction. Page | 92 Figure 2.4 Circular maps and lists of features for the pBluescript II SK vector The pBluescript II SK vector contains an ampicillin selection marker which is essential for Transformer site-directed mutagenesis kit because the bacterial strain BMH 71-18 mutS has a Tn10 transposon and therefor is tetracycline resistant. The efficiency of the chosen restriction enzyme digestion has been examined at a low DNA concentration (0.1µg/30µl) before the mutagenesis assay procedure. Page | 93 Figure 2.5 Transformer site-directed mutagenesis strategy (The picture is taken from Clontech transformer site-directed mutagenesis kit user manual) Page | 94 2.2.14.1 Selection and mutagenic primers Selection primer: 5’- CAAGATCTCGATGGCGGTGGCG-3’ Mutagenic primer (R109A): 5’- CGTGTTATGTCGTTTTAACTTTC -3’ Mutagenic primer (R129A): 5’- GTGGACATGACGCTAGATAC -3’ The changed base pairs of the selection and mutagenic primers are underlined. All primers were synthesized with an incorporated 5’ phosphate. 2.2.14.2 Generation of the mutant plasmid DNA The primer/plasmid annealing reaction mixture was prepared with 2µl of 10x annealing buffer (200mM Tris-HCl pH7.5, 100mM MgCl2, 500mM NaCl), 0.1µg plasmid DNA, 2µl of 0.1µg/µl selection primer, 2µl of 0.1µg/µl mutagenic primer (the selection and mutagenic primers are listed in section 2.2.14.1) and nuclease-free water (adjust to a total volume of 20µl). After denaturation at 100oC for 3mins, the reaction mixture was immediately placed on ice for 5mins. Then, the reaction mixture was added with 3µl of 10x synthesis buffer, 1µl of T4 DNA polymerase (3units/µl), 1µl of T4 DNA ligase (2.84 WEISS units/µl), 5µl of nuclease-free water and incubated at 37oC for 2 hours to synthesize the mutant DNA strand. The reaction was stopped by heating at 70oC for 8mins and cool down to RT. 2.2.14.3 Selection by restriction digestion The DNA in synthesis/ligation mixture was ethanol precipitated as mentioned in section 2.2.11.2 and resuspended in restriction enzyme digestion mixture with 0.5µl of NotI (20units/µl), .3µl of 10x NEBuffer3 and 26.5µl of nuclease-free water for primary selection. The reaction mixture was incubated at 37oC for 2 hours and heated at 70oC for Page | 95 5mins to stop the digestion. Then, 11µl of the digested plasmid DNA was transformed to BMH 71-18 mutS competent cells as described in section 2.2.12. The mixed plasmid pool was isolated using miniprep as described in section 2.2.13.1 and digested with NotI restriction enzyme. The digested plasmid DNA was transformed to DH5α competent cells followed the same procedure as for the first transformation. The mutant DNA was recovered by miniprep or midiprep and the mutation was verified by directly sequencing the mutagenized region as previously described in section 2.2.11.7. The mutagenic primer (R109A) was applied to introduce single mutation and the double mutations were generated using both mutagenic primer (R109A) and mutagenic primer (R129A). 2.2.15 Total RNA isolation Total RNA was isolated from cultured cells using the RNAeasy Mini Kit. Two 75cm2 flasks cells were cultured to 60-80% confluence and harvested as described in section 2.2.1.3. The cells were washed with PBS which was removed by centrifuging at RT, 900rpm for 3mins and 350µl buffer RLT containing 3.5µl -Mercaptoethanol was added to lyse the cells. The cell lysate was then homogenised by centrifuging through a PrepEase filter unit at 16,000rpm for 1min and precipitated in one volume of 70% ethanol. The resulting solution was transferred to an RNAeasy mini column sitting in a 2ml collection tube and centrifuge for 30sec at 11,000rpm. The column was then washed by 700μl buffers RW1 and twice with 500μl RPE buffer by centrifuging at 11,000rpm for 30sec. The RNeasy spin column was centrifuged again at 16,000rpm for 1min to eliminate any carryover buffer. After 30μl of RNase free water was added, the RNeasy spin column was incubated at RT for 1min. Finally, the total RNA was eluted Page | 96 by centrifuging the column for 1min at 16,000rpm. The total RNA yield and purity was determined by NanoDrop ND-1000 spectrophotometer as mentioned in section 2.2.11.1. 2.2.16 Reverse transcription polymerase chain reaction 2.2.16.1 First strand cDNA synthesis First strand cDNA was synthesized using total RNA isolated from LNCaP cells as described in section 2.2.15. Total RNA (3μg) was mixed with 1μl of 50μM Oligo (dT)20 primer and 2μl of dNTP mixture (10mM each dATP, dCTP, dGTP, dTTP at neutral pH). After the volume was adjusted to 13μl with nuclease-free water, the mixed solution was incubated at 65°C for 5mins and chilled on ice for 1min. Then, the solution was mixed with 4μl of 5x first strand buffer, 1μl of 0.1M DTT, 1μl of RNaseOUT (40units/μl) and 1μl of SuperScriptIII reverse transcriptase (200units/μl). The reaction mixture was incubated at 50oC for one hour, followed by incubation at 70oC for 15mins to inactivate the reaction. The First strand cDNA was used as a template for amplification. Page | 97 2.2.16.2 Polymerase chain reaction Polymerase chain reaction (PCR) was performed to amplify a specific region of the DNA template. The forward and reverse primers are listed as below: Forward primers sequence: 5’- ACCATGGCCACAGTTCAGCA -3’ Reverse primers sequence: 5’- CCTGTCCAAAGTGATGATGGAA -3’ The DNA template was amplified in a total volume of 50μl reaction mixture as shown in table 2.5. DNA template 50-200ng 10x PCR buffer 5µl MgCl2 (25mM) 3µl Forward primer (10µM) 1µl Reverse primer (10µM) 1µl dNTPs mixture (10mM each) Taq polymerase ( 5U /µl) Nuclease-free water Table 2.5 1µl 1µl Up to 50µl PCR mixture The reaction was incubated in a thermal cycler at 95oC for 5mins, followed by 35 cycles of PCR amplification, each cycle consisting of denaturing: 95oC for 30 seconds; annealing: 55oC for 45 seconds; extension: 72oC for 1 min, followed by one cycle incubation at 72oC for 5mins and then maintained the reaction at 4oC. PCR amplification products were analysed by agarose gel electrophoresis and purified as previously described in section 2.2.11.3 and 2.2.11.4. Page | 98 2.2.17 Real-time PCR The real-time PCR, also known as quantitative real time polymerase chain reaction (qPCR), was applied in order to quantify the level of C-FABP mRNA in the transfected LNCaP cells. Traditional PCR or RT-PCR uses agarose gel electrophoresis for detection of amplified DNA fragments at the end of the reaction. However, real-time PCR is designed to collect data in the exponential growth phase when the reaction is in progress, which significantly increased the accuracy for DNA and RNA quantitation compare to the traditional PCR or RT-PCR. Several methods exist for real-time quantitation of amplification products, including fluorescence resonance energy transfer techniques using fluorescently labeled molecular beacons, SYBR Green I. During the PCR amplification, the fluorescence of SYBR Green I increases 100- to 200-fold when bound to double stranded DNA as presented in Figure 2.6 The number of amplicons generated is directly proportional to the increase in reporter fluorescent signal which was detected at the end of the elongation step of the PCR reaction by Chromo4 fluorescence detector. Figure 2.6 SYBR Green Dye binds to the double stranded DNA and fluoresces Page | 99 2.2.17.1 Real-time PCR primer design The real-time PCR primers were designed spanning exon-exon junctions of C-FABP gene in order to avoid amplifying any genomic DNA contamination. The size of PCR amplicon was limited to between 50bp to 200bp and the primers self complimentarity and hair pins was also checked as SYBR Green Dye bound to all double stranded PCR products, including non-specific PCR products. Similarly, the house-keeping gene, β actin primers were designed as listed below: Real-time PCR primers for C-FABP Forward primers sequence: 5’- CATTGGTTCAGCATCAGGAG -3’ Reverse primers sequence: 5’- TTCATGACACACTCCACCACT -3’ Real-time PCR primers for β actin Forward primers sequence: 5’- ACCATGGCCACAGTTCAGCA -3’ Reverse primers sequence: 5’- CCTGTCCAAAGTGATGATGGAA – 3’ 2.2.17.2 Relative real-time PCR The total RNAs were extracted from each cell line using RNAeasy Mini Kit as described in section 2.2.15 and reverse transcribe 1µg of total RNA to cDNA using SuperScriptIII reverse transcriptase as mentioned in section 2.2.16.1. The real-time PCR mixture for both C-FABP and β actin was prepared with 5µl of 2× Brilliant SYBR Green qPCR master mix (containing SureStart Taq DNA polymerase, dNTPs mixture, MgCl2 and optimized buffer), 1µl of forward primer, 1µl of reverse primer, 1µl of cDNA generated by reverse transcription and 2µl of nuclease-free water. After gentle Page | 100 mix, the reactions were centrifuged briefly and placed to a real-time PCR thermocycler. The real-time PCR program is listed in table 2.6. Temperature Time Step1 Denaturation 95oC 15mins Step2 Denaturation 94oC 15 seconds Step3 Annealing 60oC 30 seconds Step4 Extension 72oC 30 seconds Step5 Plate reading 57oC 15 seconds Step6 Go back to step 2 and repeat 38 cycles Step7 Final extension 72oC 10mins Step8 Melting curve 65-95oC 1oC increment for 10mins Table 2.6 Real-time PCR program for amplification of short target DNAs (50-400bp) 2.2.17.3 Relative quantitation analysis Melting curve analysis was carried out to detect the presence of nonspecific products and the primer dimmers. The relative fold differences of C-FABP mRNA between transfected cell lines and parental cell line LNCaP were obtained using the formula listed below: Relative fold difference=2 -ΔΔCt ΔCt was calculated as the average Ct for the gene of interest minus the average Ct for the house keeping gene, β actin. ΔΔCt was calculated as the ΔCt of the test sample minus the ΔCt of the calibrator sample Page | 101 2.2.18 Analysis of protein expression using Western blot 2.2.18.1 Isolation of protein extracts from cultured cells Cells were selected for protein isolation when they were grown to approximately 80% confluence. The cells were detached (As described in section 2.2.1.3) and suspended in 10ml of completed medium to inactive the trypsin. Cells suspension was transferred to a 25 ml sterile universal tube, centrifuged at 900 rpm for 5 mins, decanted the supernatant and washed with PBS which was removed by centrifuging. The cells was lysed by adding CelLytic-M reagent and incubated on a roller mixer for 15 mins at RT (5x106 cells / 100µl). The mixture was transferred into a 1.5ml microcentrifuge tube and then centrifuged for 20 mins at 10,000rpm to pellet the cellular debris. The supernatant was harvested into a fresh microcentrifuge tube and the pellet was discarded. 2.2.18.2 Determination of protein concentration The concentration of protein extract was quantified using the Bradford dye-binding assay. The dye reagent was diluted 5-fold by adding distilled water (Bradford reagent: H2O = 1:4) and filtered through Whatman 540 paper. A concentration standard curve was created by measuring the absorbance (595nm) of a serial dilution of BSA standards (from 50µg/µl to 500µg/µl) after 15 mins incubation with Bradford reagent. Protein samples were also diluted by PBS and incubation with Bradford reagent for 15 mins before measuring the absorbance at 595nm. The sample concentration was calculated using the equation from established standard curve. Page | 102 2.2.18.3 Sodium dodecyl sulphate polyacrylamide protein gel electrophoresis (SDS-PAGE) Proteins isolated from cells were analysed using the Bio-Rad miniprotean system with 1 mm spacers. Proteins were separated by SDS-PAGE with 4% (w/v) acrylamide stacking gel and 10% or 15% (w/v) acrylamide resolving gel according to the molecular size of the target protein. The mixture (20μl) of 2x loading buffer containing equal amount of protein (v/v) was heated at 100oC in a water bath for 10 mins prior to loading, which linearized the protein by breaking hydrogen bonds of the tertiary structure of the polypeptide and reveal the antigen to be recognised by the antibody. The mixture was then chilled on ice for 2 mins and spun down before loading into the SDS gel. Electrophoresis was performed in 500ml of 1x running buffer (50 mM Tris-HCl, 200 mM glycine and 0.1% SDS) at 80V through the stacking gel and 160V through the resolving gel until the dye front had reached the end of the resolving gel. 2.2.18.4 Transfer of proteins from SDS gel to nitrocellulose membrane Separated Proteins were transferred from SDS gel to a nitrocellulose membrane using the BioRad mini trans-blot system. Six sheets of Whatman 3mm filter paper and the nitrocellulose membrane were cut according to the size of the SDS gel and soaked in 1x transfer buffer for 5 mins. The cassette, with black side down, was placed on a tray containing transfer buffer and assembles the cassette in following order (from black side): a pre-wet fiber pad, three sheets of Whatman filter paper, equilibrated SDS gel, nitrocellulose membrane, three sheets of Whatman filter paper and another pre-wet fiber pad. The air bubbles were removed from each step using a glass roller. The transfer was Page | 103 performed at 100V, 4oC for 1.5 hours in pre-chilled 1x transfer buffer. At the end of the process, the transfer efficiency was assessed through the Coomassie blue staining: the gel was stained with Coomassie blue for an hour at room temperature before being destained in destaining buffer overnight. 2.2.18.5 Immunoblotting for detection of protein expression The nitrocellulose membrane containing the proteins was blocked with 10ml of 5% Protoblock for an hour at room temperature on a gentle agitation to prevent non-specific binding of the primary antibody. The membrane was then incubated on a shaker with a primary antibody in an appropriate concentration (Table 2.7) at 4 oC overnight. The following day, the membrane was washed with 1x T-TBS 4 times for 10 mins each time to remove unbound primary antibody and incubated with a secondary antibody in an appropriate concentration (Table 2.7) for 2 hours at room temperature. Then, the membrane was washed with 1x T-TBS 4 times for 10 mins each time again. The probed antigens on the membrane were visualized by incubation in ECL reagents for 2 minutes at room temperature. Chemiluminescent images were recorded on Kodak films with 0.5-10 mins exposure. The films were developed and fixed in the dark room. Page | 104 Target protein C-FABP Primary antibody Monoclonal Rabbit Anti-human C-FABP (1:200) VEGF Polyclonal Rabbit Anti-human VEGF (1:500) β-Actin Monoclonal Mouse Anti-β-Actin (1:5000) Table 2.7 Secondary antibody Polyclonal Swine Anti-rabbit Immunoglobulins/HRP (1:1000) Polyclonal Swine Anti-rabbit Immunoglobulins/HRP (1:1000) Polyclonal Rabbit Anti-mouse Immunoglobulins/HRP (1:2000) Primary and secondary antibodies used in Western blot To standardize the loading difference, the level of β-actin expression was examined. The expression level (EL) of each target protein was calibrated using the formula listed below: Normalised EL of target protein = EL of target protein / EL of β-Actin Page | 105 2.2.19 Statistic method The Student’s t-test, introduced in 1908 by William Sealy Gossett, is the most commonly used method to evaluate the differences in averages between two groups. In this thesis, the Student’s t-test was used to compare any differences observed between each experimental group and the control group. The results from most of assays were statistically assessed using the Student’s t-test. For all Students’s t-test calculated by Microsoft Office Excel 2003, the P-value less than 0.05 was regarded as statistical significance. Page | 106 Chapter 3 Transfection of wild type and mutant C-FABP into LNCaP prostate cancer cells 3.1 Introduction As previously mentioned in Chapter 1, cutaneous fatty acid binding protein (C-FABP) is a FABP family member binding to long chain fatty acids with high affinity. C-FABP has been identified as a gene which was not only involved in malignant progression of prostate cancer but also able to promote the growth of primary tumors and induce metastasis when it transfected into benign rat Rama 37 model cells [157]. Recently, it was demonstrated that C-FABP was a prognostic marker for patient outcome and a target of tumor-suppression for prostate cancer [159]. Previous studies suggested that C-FABP might promote the malignant progression of prostate cancer by up-regulating the expression of the gene for vascular endothelial growth factor (VEGF), a potent factor for angiogenesis which is essential for growth and expansion of solid tumours [158, 159, 164]. However, how C-FABP up-regulates the expression of VEGF is still not clear. Since the common biological function of FABPs, including C-FABP, is to transport the intracellular fatty acids into cells, therefore it is interesting to know whether the tumourigenicity-promoting function of C-FABP is related to its activity of transporting fatty acids. It has been confirmed that there are three key amino acids (Arg109, Arg129, and Tyr131) of C-FABP, which are highly conserved amongst the FABP protein family, and which are responsible for Page | 107 binding to the carboxylate group in the fatty acids [165]. Mutation of one or two of the three key amino acids would either partially or completely deprive of C-FABP fatty acid-binding ability. In order to investigate whether binding to fatty acids is essential for C-FABP to promote cancer malignant progression, one and two site-directed point mutations were introduced to the C-FABP cDNA region which contains the fatty acid-binding motif. The mammalian expression vector pIRES2-EGFP containing wild type and mutated C-FABP cDNAs were tansfected respectively into the LNCaP prostate cancer cells, which did not express C-FABP prior to the transfection, to establish cell lines expressing wild type C-FABP and mutated C-FABPs. Page | 108 3.2 Results 3.2.1 Generation of mutations in C-FABP cDNA Point mutations were introduced into fatty acid binding domain of the C-FABP cDNA using Transformertm Site-Directed Mutagenesis Kit as described in Chapter 2 section 2.2.14. The pBluescript II SK vectors containing wild type, single-mutation and double-mutated C-FABP cDNA were transformed into DH5α E.coli competent cells respectively for amplification. After overnight incubation at 37oC, 225rpm, the plasmid vectors were isolated from DH5α cells using midiprep and the presence of mutations were confirmed by nucleotide sequence analysis using specific C-FABP cDNA forward and reverse primers as described in Chapter 2 section 2.2.16.2. The constructed harboring wild type of C-FABP cDNA was designated as C-FABP-WT. Those containing the single and double-mutated C-FABP cDNAs were denoted as C-FABP109A and C-FABP-109/129A, respectively. As showing in Figure 3.1, sequencing analysis results demonstrated that the sequence of the C-FABP cDNA region containing the first mutation site was successfully altered by converting Arginine 109 to Alanine 109. Thus the triplet AGA was changed into GCA. In the second mutation site the triplet CGG was changed into GCG by replacing Arginine 129 with Alanine 129. In the construct of C-FABP-109A, only one amino acid was changed in first mutation site and no change was made in second mutation site. In the construct of C-FABP-109/129A, amino acids in both mutation sites were changed. Page | 109 Figure 3.1 Detection of mutations in C-FABP cDNA by sequencing analysis (A) Point mutantion in first site is shown in panel A. The Alanine (GCA), mutated from Arginine (AGA), is presented both in C-FABP-109A and C-FABP-109/129A constructs. (B) Point mutation in second site is shown in panel B. The Alanine (GCG), converted from Arginine (CGG), is presented in C-FABP-109/129A but not in single-mutated C-FABP, C-FABP-109A and C-FABP-WT constructs. Page | 110 3.2.2 Insertion of wild type C-FABP and mutant C-FABPs into a mammalian expression vector - pIRES2-EGFP The C-FABP-WT, C-FABP-109A and C-FABP-109/129A were removed from pBluescript II SK vector by restriction enzyme digestion using XhoI/PstI restriction enzymes. The products of restriction enzyme digestion were separated by agarose gel electrophoresis (1%) as presented in Figure 3.2A. The DNA fragments of C-FABP-WT, C-FABP-109A and C-FABP-109/129A were recovered from agarose gel respectively as described in Chapter 2 section 2.2.11.4. The mammalian expression vector pIRES2-EGFP was linearized using XhoI/PstI restriction enzymes as shown in Figure 3.3A and purified from agarose gel. The purified DNA fragment of wild type and mutant C-FABPs were inserted into pIRES2-EGFP vector respectively using T4 DNA ligase. The recombinant DNAs were transformed into DH5α bacterial cells and plated onto LB plates in the presence of antibiotics (50µg/ml ampicillin). Single colonies of transformed DH5 bacterial was transferred into flasks containing 10ml of LB medium with 50µg/ml ampicillin and left to grow overnight at 37C, 225rpm. The overnight cultured bacterial cells (5ml) were harvested and the pIRES2-EGFP vectors containing C-FABP-WT, C-FABP-109A and C-FABP-109/129A were isolated from bacterial cells respectively using miniprep as described in Chapter 2 section 2.2.13.1. The ligation was examined using restriction enzyme digestion as shown in Figure 3.3B. The sequencing analysis also confirmed the presence of the insertions in pIRES2-EGFP vector as shown in Appendix. Page | 111 A B 1 Figure 3.2 2 3 4 M The products of restriction enzyme digestion for the ligation reaction (A) After DNA sequencing analysis, the pBluescript II SK vector containing C-FABP-WT (Lane 2), C-FABP-109A (Lane 3) and CFABP-109/129A (Lane 4) was digested by XhoI/PstI for the ligation reaction. The control vector digested by XhoI/PstI is presented in lane1. For lanes 2, 3 and 4, the bottom band represents DNA fragment of CFABP-WT, C-FABP-109A and C-FABP-109/129A respectively. The top band in lanes 1, 2, 3, and 4 represents digested pBluescript II SK vector (3.0 kb). (B) The DNA standard marker is denoted in number of base pairs. Page | 112 A B 1 Figure 3.3 2 M 1 2 3 4 5 M Identification of the recombinant DNAs by restriction enzyme digestion (A) Two restriction enzyme digestions (Lanes 1 and 2) of mammalian expression vector pIRES2-EGFP (5.3 kb) using XhoI/PstI restriction enzymes to linearize the plasmid for the ligation reaction. (B) After ligation reaction, the recombinant DNAs were digested using XhoI /BamHI to assure that the DNA fragment of C-FABP-WT, C-FABP-109A and C-FABP-109/129A was inserted into the vector. The empty pIRES2-EGFP vector was used as control in lane 1. The vector containing C-FABP-WT digested with BamHI was shown in lane 5 for comparison with the recombinant DNAs digested with XhoI/PstI restriction enzymes. For lanes 2, 3 and 4, the bottom band represents DNA fragment of C-FABP-WT, C-FABP-109A and C-FABP-109/129A respectively. The top band in lanes 1, 2, 3, 4, and 5 represents digested pIRES2-EGFP vector. In panel A and B, the DNA standard marker in lane M is denoted in number of base pairs. Page | 113 3.2.3 Detection of wild type and mutated C-FABP mRNAs in transfected LNCaP cells by RT-PCR LNCaP cells were stably transfected with the empty pIRES2-EGFP vector and the vector containing C-FABP-WT or one of the two mutated C-FABP: C-FABP-109A and C-FABP-109/129A using a polyamine based transfection reagent. The cell line transfected with C-FABP-WT, C-FABP-109A, C-FABP-109/129A and the empty pIRES2-EGFP vector were designated LNCaP-WT, LNCaP-R109A, LNCaPR109/129A and LNCaP-V. It was difficult to establish transfected clones as LNCaP cells had an extremely low survival rate when plated at very low concentrations in tissue culture flask [166]. Furthermore, the Geneticin was also significantly slow the growth rate of the transfected cells in selection step which makes the formation of transfected clones more difficult. In order to overcome the difficulties in generating LNCaP clones, the modified transfection method was developed using GeneJammer transfection reagent which is a proprietary formulation of polyamine offering high efficiency with low cytotoxity. After selection step, the survived clones were analyzed by fluorescence microscopy as described in Chapter 2 section 2.2.2.2. The successful transfected clones for LNCaP-WT, LNCaP-R109A, LNCaP-R109/129A and LNCaP-V were isolated using ring clone and cultured in selective medium separately. The pool of wild type C-FABP and mutant C-FABP transfected clones was Page | 114 also established by random selection of five single clones. The abundant green fluorescence images for pooled transfected cell lines were presented in Figure 3.4 confirmed the expression of recombinant pIRES2-EGFP vectors. For LNCaP-WT, LNCaP-R109/129A and LNCaP-V transfected cell lines, the fluorescence-positive cells represented a subset of approximately 90% of the whole population. For LNCaP-R109A, approximately 80% of cells displayed strong green fluorescent emission and about 10% of LNCaP-R109A transfectant expressed less green fluorescence. The rest 10% populations of all four pooled transfected cell lines were fluorescence-negative. The fluorescence-positive cell population stayed at same level when increased the concentration of Geneticin or removed the antibiotic for a period of time (4 weeks). Similar observations were found in single cloned transfectants. Page | 115 Figure 3.4 The green fluorescent emission images of successful transfected cells The green fluorescence images of pooled LNCaP-WT, LNCaP-R109A, LNCaP-R109/129A and LNCaP-V were shown in panel A1, A2, A3 and A4 respectively. All four cell lines appeared strong green fluorescence confirmed the expression of recombinant pIRES2-EGFP vectors. Page | 116 To confirm the presence of the wild type and mutated C-FABP mRNAs existence in the transfected cell lines, total RNA was isolated from each pooled clones of transfectants. The total RNAs were amplified by PCR using a specific C-FABP forward primer which was listed in Chapter 2 section 2.2.16.2 and a specific pIRES2-EGFP plasmid reverse primer to ensure the amplification products were from the transgenes instead of the endogenous gene for C-FABP. The products of PCR amplification were analysed by agarose gel electrophoresis as shown in Figure 3.5. No band was detected in PCR amplification without reverse transcriptase indicating that the amplified bands were not arising from total RNA isolation. The RT-PCR products on the agarose gel were excised and purified as previously mentioned in Charpter 2 section 2.2.11.4 and send for sequencing analysis (Liverpool University, School of tropical Medicine, Sequencing Unit). The RT-PCR products of LNCaP-R109A and LNCaP-R109/129A were the same size as LNCaP-WT (623kb). In contrast, no band was detected in LNCaP-V and negative control amplifications. Subsequent nucleotide sequencing also confirmed the presence of C-FABP-WT, C-FABP-109A, and C-FABP-109/129A. Page | 117 Figure 3.5 Detection of mRNAs from transfected cell lines by RT-PCR Reverse transcriptase PCR was carried out using the Taq polymerase with specific primers to amplify the wild type and mutated C-FABP gene. Total RNAs were isolated from LNCaP cell line (Lane 1) and transfected cell lines LNCaP-V (Lane 2), LNCaP-WT (Lanes 3 and 4), LNCaP-R109A (Lanes 5 and 6) and LNCaP-R109/129A (Lane 7 and 8). Negative controls are presented in lanes 4, 6 and 8 in which the reverse transcriptase was omitted. The PCR products were analyzed on a 2% agarose/EtBr gel in 1xTBE buffer alongside 100 base pair ladder DNA marker (Lane M). Page | 118 3.2.4 Expression of wild type and mutant C-FABP proteins in LNCaP cell line Levels of wild type and mutant C-FABP proteins expressed in transfected LNCaP cells were examined by Western Blot as described in Chapter 2 section 2.2.18. The LNCaP-V cells and highly malignant prostate cancer cells, PC-3M were used as negative and positive control respectively. The C-FABP specific primary antibody used in Western Blot analysis was provided by our Japanese colleagues. The level of β-actin expression in each cell line was also examined and used as loading control. The C-FABP expression in transfectant cells was quantified by relating to that in positive control cells using scanning densitometry. The expression of wild type C-FABP (15 KDa) in LNCaP-WT and PC-3M cells are shown in Fegure 3.6A and B. When the level of C-FABP expression in positive control PC-3M cells was set as 1, the level of C-FABP in LNCaP-WT was significantly increased (Students t-test P<0.05) to approximately 70% of that expressed in the highly malignant PC-3M cells. No band was observed in LNCaP-R109A, LNCaP-R109/129A and LNCaP-V. In addition, no significant difference was detacted in wild type C-FABP levels between single cloned and pooled LNCaP-WT cells (Students t-test P>0.05). The expression of β-actin, served as a loading control, is also examined as shown in Fegure 3.6A at the size of 42 KDa. Page | 119 A B Figure 3.6 Detection of protein expression levels of wild type C-FABP in single cloned and pooled transfectant LNCaP cell lines (A) Cellular lysates from three individual cloned (Lanes 3, 5 and 7) and three pooled (Lane 4, 6 and 8) of transfectant lines: LNCaP-WT (Lanes 3 and 4), LNCaP-R109A (Lanes 5 and 6) and LNCaP-109/129A (Lanes 7 and 8), were subjected to Western blot analysis. C-FABP protein band was recognized by a monoclonal anti-human C-FABP antibody and visualized as described in Chapter 2 section 2.2.18. The β-actin antibody was used to correct the possible loading artifacts. (B) The relative levels of C-FABP expression in different transfectants. The results (mean±S.D. of three experiments) were obtained by densitometry analysis of the band intensities. Page | 120 Both wild type and mutant C-FABPs were picked up using a rabbit polyclonal antihuman C-FABP antibody. It was observed that intensities of the bands in LNCaP-WT, LNCaP-109A and LNCaP-109/129A were very similar, whereas the expression of C-FABP in control cell line LNCaP-V and parental cell line LNCaP is barely detectable as shown in Figure 3.7A. The β-actin protein expression level in each cell lines, served as a loading control, was also presented. Further relative quantifications by scanning densitometry were shown in Figure 3.7B. When wild type C-FABP expression in LNCaP-WT was set as 1, the level of mutant C-FABPs were 1.09 and 0.946 in LNCaP-R109A and LNCaP-109/129A cells respectively. No significant expression differences have been found between the level of wild type C-FABP and level of mutant C-FABPs (Students t-test P>0.05) in the transfectants. LNCaP-V with empty pIRES2-EGFP vector and parental LNCaP cells exhibited undetectable level of C-FABP. Page | 121 A B Figure 3.7 Detection of protein expression levels of wild type and mutant C-FABPs in pooled transfected LNCaP cell lines (A) The mutant C-FABP proteins in LNCaP-R109A cells (Lane 2) and LNCaP-109/129A cells (Lane 3) were recognized using a polyclonal anti-human C-FABP antibody. The wild type C-FABP in LNCaP-WT cells (Lane 1) was also detected by the antibody. LNCaP-V and LNCaP cells were used as negative control in Lane 4 and 5 respectively. (B) The mean and S.D. of the relative band intensities of each cloned and pooled cell lines from three individual experiments. Page | 122 3.2.5 Measurement of levels of wild type and mutant C-FABP mRNAs in different transfectants by Real-time PCR To confirm that the transfected wild type and different mutant cDNAs were expressed, the relative C-FABP mRNA levels in each transfectant cell line was measured by real-time PCR. The parental cells, LNCaP were used as the calibrator. Total RNAs from each cell lines were isolated routinely and purified by RNAeasy Mini column. As poor quality RNA samples can lead to spurious Real-time PCR results, the integrity and quality of the total RNAs were determined by Agilent 2100 bioanalyzer using RNA 6000 Nano Kit. The results (as shown in Figure 3.8A) were analyzed by the 2100 Expert Software and visual output were confirmed that the isolated total RNAs were in high quality with both rRNA bands (18S/28S) distinguishable as shown in Figure 3.8B. The RNA integrity numbers (RIN) for each isolated RNA were between 9.0 to 9.6 out of 10 and no degradation was observed. The RNA concentration of each sample was also calculated by the 2100 Expert Software. After RNA purification, relative quantification real-time PCR was preformed to measure the mRNA levels of wild type and mutant C-FABPs using exon-exon junction primers as described in Chapter 2 section 2.2.17. Figure 3.9 summarizes the relative level of C-FABP mRNA in total RNA isolated from each pooled cell line. The LNCaP-WT, LNCaP-R109A and LNCaP-R109/129A cell line showed a significant increase in the relative C-FABP mRNA level when compared to parental cell line Page | 123 LNCaP. When the C-FABP mRNA level in the parental cell line LNCaP was set at 1, the levels of expression was 137.5±18.1, 114.1±15.6, and 115±25.2 in LNCaP-WT, LNCaP-R109A and LNCaP-R109/129A, respectively. On the other hand, the C-FABP mRNA level in LNCaP-V cells was only slightly increased. The C-FABP mRNA level between LNCaP-R109A and LNCaP-R109/129A did not show significant difference (Students t-test P>0.05). Similarly, no difference was found when they compared to LNCaP-WT (Students t-test P>0.05). However, the relative levels of C-FABP mRNA expressed in all three transfectant lines (LNCaP-WT, LNCaP-R109A and LNCaP-R109/129A) were significantly increased when compared to the control LNCaP-V cells (Students t-test P<0.05). Page | 124 A B Figure 3.8 Quality checks of total RNA samples by Agilent 2100 bioanalyzer The quality and quantification of total RNA were assessed using the RNA 6000 Nano LabChip on Agilent 2100 bioanalyzer. The gel electropherogram images for LNCaP-WT (Lane 1), LNCaP-R109A (Lane 2), LNCaP-R109/129A (Lane 3), LNCaP-V (Lane 4) and LNCaP (Lane 5) were shown in panel A. The fluorescence plots with two peaks of 18S and 28S ribosomal RNAs of each sample were shown in panel B. The qualities of RNAs were presented by RINs which were calculated from the plots. Page | 125 Figure 3.9 Relative level of C-FABP mRNA in transfectant and LNCaP cell lines The Real-time PCR products of LNCaP-WT (A), LNCaP-R109A (B), LNCaP-R109/129A (C), LNCaP-V (D) and LNCaP (LNCaP) were analyzed by Chromo4 fluorescence detector. The mean and S.D. of relative C-FABP mRNA level for each cell line from three individual experiment were shown in the bar chart. LNCaP cells were used as calibrator. Page | 126 3.3 Discussion C-FABP, which was originally isolated from psoriatic skin, is a typical FABP family member with high affinity to long chain fatty acid. It has been revealed that overexpression of C-FABP may promote VEGF to facilitate angiogenesis which contributed to tumor progression such as tumor formation and metastasis [158, 159]. However, the exact mechanism is still not clear. In order to establish transfectants expressing mutant forms of C-FABP which are not able to bind to fatty acids, point mutations were introduced to two key amino acids (Arg109 and Arg129) which are responsible for fatty acid-binding in C-FABP gene. As a result, the relationship between the fatty acid binding capacity and the tumor promoting function can be assessed by investigating whether the transfectants expressing mutant C-FABP can still promote tumor growth. Two mutated gene have been successfully generated from the wild type C-FABP gene using Transformertm Site-Directed Mutagenesis Kit: (1) C-FABP-109A where Arginine 109 was converted to Alanine 109; (2) C-FABP-109/129A where Arginine 109 and Arginine 129 were converted to Alanine 109 and Alanine 129. The point mutations in both mutated genes were confirmed by sequencing analysis. Then, the wild type and mutant C-FABPs were transferred into a mammalian expression vector pIRES2-EGFP with enhanced green fluorescent protein coding region using XhoI and PstI restriction digestion enzymes. The pIRES2-EGFP empty vector and vectors containing C-FABP-WT, C-FABP-109A, and C-FABP-109/129A were transfected into LNCaP cells to create transfected cell lines LNCaP-V, LNCaP-WT, LNCaP-R109A and LNCaP-R109/129A respectively. Page | 127 LNCaP cell line, established from a needle aspiration biopsy of the lymph node of male patient with diagnosed metastatic prostate carcinoma, is a wildly used androgen sensitive human prostate cancer model cell line [60]. Two properties of LNCaP cells make them a particularly difficult cell line to generate transfectants compared to other cell lines --- (1) LNCaP cells have a low growth rate and a tendency to form aggregates; (2) LNCaP cells have a poor survival rate when seeded at low concentration in tissue culture plate [166], which can significantly affect the efficiency of transfection as the cell colonies were developed from a single transfected cell. To overcome the difficulties in cloning LNCaP cells, the modified transfection method was carried out using a polyamine based transfection reagent, GeneJammer which provided a higher transfection efficiency and lower cytotoxity to LNCaP cells compared to calcium phosphate method [167]. The transfected cell lines were selected by 0.5mg/ml antibiotic G418 for 10 days in a 75cm2 tissue culture flask. Then, the G418 was removed and 500µl of untransfected LNCaP cells at a concentration of 2x104cells/ml were seeded into the flask as a feeder layer. The cells were allowed to grow in routine culture medium until cellular confluence reached at 80%. Then, the cells were re-select followed the same procedure as for the first selection. After 5 round selections, the cells were maintained in selective culture medium and successfully transfected colonies were determined by green fluorescence imaging and recovered by ring cloning. Page | 128 Transfectant lines: LNCaP-WT transfected with wild type C-FABP, LNCaP-R109A transfected with single site mutated C-FABP, LNCaP-R109/129A transfected with double site mutated C-FABP and LNCaP-V (control) transfected with pIRES2-EGFP vector only were successfully generated. The wild type or mutant C-FABP cDNAs in each transfected cell line were examined by RT-PCR. The Western Blot assays were applied to determine the protein level of C-FABP-WT, C-FABP-R109A and C-FABP-R109/129A in each transfected cell line. The results demonstrated that expression level of C-FABP-WT in pooled LNCaP-WT cells and mutant C-FABPs in pooled LNCaP-R109A and LNCaP-109/129A cells were at similar level, which indicated that these transfectants were suitable for comparison. However, there is no C-FABP cDNA or protein has been detected in LNCaP cells transfected with empty vector and paraental cells indicating that transfected with pIRES2-EGFP vector did not change the expression level of C-FABP. The same pattern of relative C-FABP mRNA level obtained by Real-time PCR analysis also confirmed the Western Blot results. The results in this chapter demonstrated the successful establishment of tansfected cell lines with expression of wild type or mutant C-FABPs. However, the effect of wild type or mutant C-FABP expression in these transfectants is unknown. These aspects such as fatty acids uptake ability, cell growth rate, invasion and tumor formation will be examined and presented in the Chapter 4. Page | 129 Chapter 4 Effect of overexpression of wild type and mutant C-FABPs on fatty acids uptake ability and tumorigenicity in prostate cancer cells 4.1 Introduction C-FABP is associated with malignant tumor progression as mentioned previously [160]. However its mechanism of action is still not clear. In Chapter 3, four transfected LNCaP cell lines with expression of wild type and mutant C-FABPs were successfully established to generate transfectants with different fatty acids uptake capacities. To reveal whether the fatty acid binding ability is essential for C-FABP to promote cancer development progression, fatty acids uptake ability of transfected cell lines were monitored. In this Chapter, experiments were performed to study the effect of reduced fatty acid binding ability on the tumor-promoting activity of C-FABP. Long chain fatty acids are not only served as metabolic fuel but also act as endogenous ligands for nuclear receptors such as PPAR [146]. Within the cells, it is possible that excessive intracellular fatty acids may contribute to both breast and prostate cancer risk and disease progression [168-170]. Although the three key amino acids conserved in FABPs were suggested to be the fatty acid binding site, direct evidence on the effect of changing this site on fatty acid binding ability of C-FABP is lacking. Thus, it is important to examine the long chain fatty acids uptake capacities of transfected cell lines expressing wild type and mutant C-FABPs. The fatty acid uptake assay was performed to assess relationship between the key amino acids mutation and the fatty acid binding capability of C-FABP. Page | 130 Cell proliferation and invasion ability are two prominent features contributory to tumorigenicity and metastasis. The recent studies showed that cell proliferation rate was significantly reduced (by 2.5-fold) in siRNA transfected prostate cancer cell line PC-3M, in which the expression level of C-FABP was suppressed, compared to the parental cell line [160]. Furthermore, it has been demonstrated that suppression of C-FABP expression resulted in great reduction of the cell invasiveness and inhibition of metastasis [159, 160]. On the other hand, Fang et.al reported that overexpression of C-FABP in transfected human oral squamous carcinoma cells was associated with promoting the cancer cells growth rate and invasion ability [171]. These findings suggest that increase expression of C-FABP is related to malignant progression of these cells. Tumorigenesis is complex progresses, which involves multiple mutations of tumor suppress genes or proto-oncogenes resulted in losing control of its cell proliferation. The studies showed that androgen-dependent human prostate cancer LNCaP cells rarely formed tumors with 0%-50% tumorigenicity rate when inoculated subcutaneous (SC) in immunodeficient nude mice [60, 172] and even if the tumor was formed, the tumor growth was very slow. Morgan et.al showed that suppression of C-FABP expression level in PC-3M cells significantly reduced tumor formation ability of the cancer cells both in vitro and in vivo [160]. Similar results had been achieved by using antisense RNA to suppress C-FABP expression [159]. However, the tumorigenicity of LNCaP cells with overexpression of C-FABP has not been reported. In contrast to PC-3M cells which express very high level of C-FABP, the weakly malignant LNCaP cells do not Page | 131 express detectable level of C-FABP. It is interesting to know whether the tumorigenicity of LNCaP can be increased by elevated C-FABP expression. Since the biological activity of C-FABP in normal condition is to bind to fatty acids and to transport them into cells, it is important to understand whether the ability of binding to fatty acids is related to the tumorigenicity promoting function of C-FABP in experiments of Chapter 4. Firstly, the fatty acid uptake capacity of cloned transfectants LNCaP-WT, LNCaP-R109A, LNCaP-R109/129A and the control transfectant LNCaPV was measured and compared. Then transfectants were subjected to different in vitro and in vivo assessments to analyze the effect of overexpression of wild type and mutant C-FABP on the cell properties associated with tumor development and malignancy. Page | 132 4.2 Results 4.2.1 The effect of increased expression of wild type and mutant of C-FABPs on cellular fatty acid uptake The orange-red florescence-labeled fatty acid BODIPY 558/568 C12 {4, 4-difluoro-5(2-thienyl)-4-bora-3a, 4a-diaza-s-indacene-3-dodecanoic acid}, which act as the analogue of natural LCFA, was applied to examine the level of fatty acid uptake for each transfectant. The fluorescence intensity of each single cell before and 30mins after adding labeled LCFA was measured using an EPICS XL Cytometer at the wave length of 570 nm to assess the transfectants fatty acid uptake capacities as described in Chapter 2 section 2.2.3. The fatty acid uptakes in different transfected cell lines were shown in Figure 4.1. The peaks in the each diagram represented the red fluorescence intensities which were quantified by manufacture’s software as demonstrated in X-Mean. The number and proportion of cells in each cell line were also indicated in the diagram column number and %Gated respectively. Relative levels of LCFA uptake were shown in Figure 4.2. When the level of fatty acid uptake in LNCaP-V cells was set at 1, levels in LNCaP-WT, LNCaP-R109A, and LNCaP-R109/129A were 3.1±0.7, 1.7±0.5 and 1.4±0.5 respectively. In comparison with the control transfectant, the fatty acid uptake by LNCaP-WT cells was significantly higher (Student t-test, P=0.0064). However, the fatty acid uptake between LNCaP-R109A and the control LNCaP-V, or that between LNCaP-R109/129A and the control transfectant, was at similar level (Student t-test, P =0.095, P = 0.216). In addition, no significant difference was observed in fatty acid uptakes between LNCaP-R109A and LNCaP-R109/129A (Student t-test, P=0.584). Page | 133 Figure 4.1 Fatty acid uptakes by mutant and wild-type C-FABP cDNA transfectants at absence and presence of red florescence labeled LCFA. Flow cytometry analysis of LCFA uptake by different cell lines. Florescence labeled BODIPY 558/568C12 was added to the cultured cells and plated at 1x106cells/well in a 6-well plate. The fluorescence intensity of each transfected cell lines was assessed (X-Mean) and plotted as histogram before (left panel) and after 30min (right panel) incubation. A1: LNCaP-WT; A2: LNCaP-R109A; A3: LNCaP-R109/129A and A4: LNCaP-V. Page | 134 Figure 4.2 Relative levels of LCFA uptake by mutant and wild-type C-FABP cDNA transfectants Relative levels of LCFA uptake by LNCaP-WT (Column 1), LNCaP-R109A (Column 2) and LNCaP-R109/129A (Column 3) and the control cells LNCaP-V (Column 4) were measured by flow cytometer. LNCaP-WT had the highest fatty acid uptake capacity and significantly higher than the rest transfectants (Student t-test P<0.05). The relative level of LCFA uptake by LNCaP-R109A was higher than that of LNCaP-R109/129A and the control tansfectant LNCaP-V but the differences are not significant (Student t-test P>0.05). Data expressed as means±SD of three independent experiments. Page | 135 4.2.2 Effect of wild type and mutant C-FABPs on cell proliferation To find out whether overexpression of wild type or mutant C-FABPs were correlated with the cell growth rate, proliferation assays were conducted to assess cell proliferation. Standard curve of each tested cell line was constructed as described in section 2.2.4.1. Diagrams in Figure 4.3 presented standard curves for the four transfectants. The MTT forms crystals in cytoplasm and the crystals were dissolved by DMSO. The colorimetric measurement was performed at 570nm wavelength using a spectrophotometer. The values obtained from the plate reader were converted to cell numbers which were extrapolated from the standard curve. The detailed proliferation assay results were demonstrated in Figure 4.4. During the first three days, growth rates of all four transfectant lines were very similar. However, from the 3rd day on, the proliferation rate of LNCaP-WT started to be higher than those of other lines. On the 4th and the 5th day, the numbers of LNCaP-WT cells were 44070±1314 and 50508±1193 respectively. Whereas in the same time points, numbers of cells from the control line were 34445±1044 and 40095±2947 respectively. The proliferation rate of LNCaP-WT was increased 28% (Day 4) and 26% (Day 5) compared with LNCaP-V and significantly higher than the control (Student t-test, P=0.001). On the other hand, cell numbers of LNCaP-R109A and LNCaP-R109/129A were not significantly different when compared to control transfectant LNCaP-V cells during the 5 day culture (Student t-test, P=0.482, P=0.055). The final cell counts on day 5 and statistical comparison were summarised in Table 4.1. Page | 136 Figure 4.3 Standard curves of different transfected cell lines For each cell line, a standard curve was established by plotting absorbance (OD at 570nm) (Axis Y) against the number of cells (Axis X). The cell number of samples was determined by comparing serially diluted standard from the standard curve. The curve equation and regression value of standard curve are presented in the diagrams. Page | 137 Figure 4.4 Time course curve of proliferation rate for wild type and mutant C-FABPs transfected LNCaP cells The number of cells for each transfectants, shown as mean of three individual experiments, were obtained every day during the 5 day period. Cell lines Mean number of cells ± SD P Value LNCaP-WT 50508 ± 1193 0.001 LNCaP-R109A 41271 ± 1088 0.482 LNCaP-R109/129A 36115 ± 1587 0.055 LNCaP-V 40095 ± 2947 --- Table 4.1 Cell counts of transfectants at the end point of proliferation assay The final cell counts on day 5 and the statistic analyses were show above. P values were obtained by comparing sample groups to control LNCaP-V. Page | 138 4.2.3 Invasiveness of different transfected cell lines A aodified Boyden Chamber assay was applied to evaluate the invasiveness of different transfected cell lines. This chemotactic directional migration assay was used wildly to reveal the cell invasion ability which correlates well with the metastasis potential [173]. The cells were examined in a Boyden Chamber with 6.5mm diameter polycarbonate membrane coated with 30μg of matrigel to the upper compartment before the cells were seeded. After 24 hours incubation, cells remaining in the upper compartment of the filters were removed and the cells attached to the lower side of the filter were fixed and stained by crystal violet reagent. The numbers of migrated and invaded cells were counted under a light microscope at 10/0.25 magnification. On each filter, nine random fields were counted and the representatives of each transected cell line were shown in Figure 4.5. The results were normalized by the cell number after 20 hours routine culture and were shown in Figure 4.6. All migrated cells remained attaching to the low surface of the membrane and no cells were found in the culture medium of the low chamber. When the number of the invaded cells from LNCaP-V was set at 1, the numbers of the invaded cells from LNCaP-WT, LNCaP-R109A and LNCaP-R109/129A were 2.2±0.3, 1.6±0.2, and 1.2±0.2 respectively. The number of invaded cells from LNCaP-WT cells was the highest in all transfected cell lines and more than 2-fold higher than the number of invaded cells from LNCaP-vector cells (Student t-test, P=0.002). The invasiveness of LNCaP-R109A cells was higher than LNCaP-R109/129A and LNCaP-vector cells, but Page | 139 was significantly lower than that of LNCaP-WT cells (Student t-test, P=0.003). The invasiveness of LNCaP-R109/129A cells was at similar level to that of the control LNCaP-Vector cells (Student t-test, P=0.247). Difference of invaded cells between LNCaP-R109A and LNCaP-R109/129A was not significant (Student t-test, P=0.06). Page | 140 1 2 3 4 Figure 4.5 The images of invaded cells of different transfectants on the lower filter surface of the Boyden Chamber The appearances of invaded cells from LNCaP-WT (Panel 1), LNCaP-R109A (Panel 2), LNCaP-R109/129A (Panel 3) and LNCaP-V (Panel 4). Page | 141 Figure 4.6 The invasiveness of cell lines transfected with wild type and mutant C-FABP genes The columns represent the invasiveness of LNCaP-WT (Column 1), LNCaP-R109A (Column 2), LNCaP-R109/129A (Column 3) and LNCaP-V (Column 4). The invasiveness of LNCaP-WT which expresses of wild type C-FABP was significantly higher than that of LNCaP-R109A, LNCaP-R109/129A and LNCaP-V. The invasiveness was at a downward trand from column 1 to column 4 but no statistical difference was found between either LNCaP-R109A and LNCaP-R109/129A or LNCaP-R109/129A and LNCaP-V. The results are the mean ± SD of three individual experiments. Page | 142 4.2.4 Soft agar colony formation assay of transfected cell lines The tumourigenicity of the control and the transfectant cell lines was tested by examining their anchorage-independent growth in a soft agar assay. The images of soft agar assay for each transfectant were shown in Figure 4.7 and the results of the soft agar assays were shown in Figure 4.8. The accurate quantitative assessment and P values of Student t-test for each transfected cell line compared to control LNCaP-V were presented in Table 4.2. The number of colonies produced in the soft agar by LNCaP-V, LNCaP-WT, LNCaPR109A and LNCaP-R109/129A were 130±48, 798±181, 220±66, and 157±38, respectively. In comparison with the control, the number of colonies produced by LNCaP-R109A and LNCaP-R109/129A cells was increased slightly but was not statistically different (Student t-test, P=0.1267, P=0.4855). The most significant change was observed in LNCaP-WT cells which produced more than 6-fold increase in the number of colonies formed in soft agar assay comparing to the control. Further analysis showed that no statistical difference was found between the number of colonies produced by control LNCaP-R109A cells and the number of colonies produced by LNCaP-R109/129A cells (Student t-test, P=0.2240). However, the differences between the colony numbers produced by LNCaP-R109A or LNCaP-109/129A were significantly reduced when compared to LNCaP-WT (Student t-test, P<0.005). Page | 143 A B C D Figure 4.7 The colonies produce by transfectants in soft agar assay After incubation for four weeks, 3 hours staining by MTT was performed on colonies developed from LNCaP-WT (Panel A), LNCaP-R109A (Panel B), LNCaP-R109/129A (Panel C) and LNCaP-V (Panel D) were displayed above. Cells from each transfected cell lines were tested in three plates. Page | 144 Figure 4.8 Colonies formation ability of transfected cell lines in soft agar assay The number of colonies was counted using GelCount with optimised setting: Edge detaction: 30; Centre detaction: 50; Colony diameter: 150µm-800µm; Circularity: 60; Density: 0.75. Cell lines Mean number of cells ± SD P Value LNCaP-WT 798 ± 180.6 0.003 LNCaP-R109A 220 ± 65.7 0.1267 LNCaP-R109/129A 157 ± 38.1 0.4855 LNCaP-V 130 ± 47.5 --- Table 4.2 The numbers of colonies produced by transfectants in soft agar assay Numbers of colonies (mean ± SD) produced by different transfectants in three separate soft agar assays were shown above and P values were obtained by comparing sample groups to control group LNCaP-V. Page | 145 4.2.5 Expression of wild type C-FABP increased tumourigenicity in vivo To examine the tumourigenicity of wild type and mutant C-FABPs transfected cell lines, cells were harvested and injected subcutaneously into the right flank of Balb/C immuno-incompetent nude mouse as described in Chapter 2 section 2.2.7. A representative tumor developed inside a mouse from each transfectant group was displayed in Figure 4.9A and tumor masses resected from the corresponding the tumorbearing animals were also displayed in the lower panel of Figure 4.9A. The appearance of two representative tumors from mouse inoculated with LNCaP-WT or LNCaP-V cells were displayed in Figure 4.9B. In addition, the tumor growth curves of all transfectants were shown in Figure 4.10 according to the average tumor volume measured and calculated. Finally, the incidence of tumors produced by each transfected cell lines in nude mice and tumor weight at autopsy were summarized in Table 4.3. Both Figure 4.9 and Figure 4.10 clearly showed that the size of tumors produced by LNCaP-WT cells were much bigger than those produced by the rest groups of mice. It can be noticed in Figure 4.9B that the tumor produced by cells overexpressing C-FABP appeared dark-reddish, as opposed to a pale appearance of tumors generated by control transfectant LNCaP-V. The latent period shown in Table 4.3 was different amongst different mice and ranged from 13 to 25 days, no significant difference was found on the average length of latent time amongst the different groups. However, the shortest length of latent period for LNCaP-WT cells was 13 days while for rest transfectants the shortest length was 18 days. The Table 4.3 also showed that 3 out of 8 (37.5%) mice Page | 146 inoculated with LNCaP-R109/129A and LNCaP-V cells and 4 out of 8 (50%) mice inoculated with LNCaP-R109A cells produced tumors. Whereas in the group inoculated with LNCaP-WT cells, 7 out of 8 (88.9%) mice developed tumors. The volumes of tumors produced by the four groups of mice at different time points in nude mice were shown in Figure 4.10. The average volume of tumours produced in the mice inoculated with LNCaP-WT cells was significant greater than those produced by the mice inoculated with LNCaP-R109A, LNCaP-R109/129A and LNCaP-V cells respectively. At autopsy (nine weeks after inoculation), the average weight of the tumours produced by LNCaP-WT, LNCaP-R109A, LNCaP-R109/129A and LNCaP-V was 614.3±192mg, 217.5±69.5mg, 59.3±19mg and 46.7±15.3mg respectively. While the average weight of tumours produced by LNCaP-R109A or LNCaP-R109/129A was not significantly increased (Mann-Witney U-test, P=0.059, P=0.879), the average weight of tumours produced by the LNCaP-WT cells was significantly heavier than that produced by LNCaP-V cells by 13.2-times (Mann-Witney U-test, P=0.003). Page | 147 A B Figure 4.9 1 2 The tumor produced by nude mice inoculated with transfected cell lines (A) Each picture is a representative animal from each group: LNCaP-WT (Lane 1), LNCaP-R109A (Lane 2), LNCaP-R109/129A (Lane 3) and LNCaP-V (Lane 4). Tumor masses resected from the corresponding the animals were also displayed below each Lane respectively. The tumor produced by LNCaP-WT was pointed out by an arrow but the tumors produced by the rest transfectants were imperceptible. (B) The representative tumor produced by LNCaP-WT (Panel 1) and LNCaP-V (Panel 2). Page | 148 Figure 4.10 Average tumor volume of each nude mice group inoculated with different transfected cell lines Wild type, two mutant and control transfectants were inoculated subcutaneously into right flank of four groups of Balb/C immuno-incompetent nude mice (8 animals in each group). The tumors became visualized and measurable after average three weeks latent phase. Tumor volume was measured every three days and calculated by the following formula: Length x Width x Height x 0.5236. Page | 149 Table 4.3 Incidence of tumors produced by transfectants in nude mice and tumor weight at autopsy a Tumor incidence is the percentage of the number of mice with tumors/ total number of inoculated animals. b Latent period is the number of days from the time of inoculation to the time of first appearance of tumor. c Tumor weight was measured at autopsy performed 60 days after inoculation of the transfectant cells. Page | 150 4.3 Discussion In the first part of this chapter, the relationship between the tumourigenicity-promoting function of C-FABP and its fatty acid-binding ability was investigated. As mentioned in Chapter 2, C-FABP binds to fatty acids through a binding motif which consists of three key amino acids (Arg109, Arg129, and Tyr131) and replacing any one of these key amino acids with another will almost completely deprive C-FABP of its fatty acid binding ability [165, 174]. Fatty acid uptake is a strictly regulated process and it is related to the levels of certain FABPs in LNCaP cells [175]. The fatty acid binding assay revealed that the fatty acid uptake capacity of LNCaP cells was significantly increased when C-FABP was forced to be overexpressed. On the other hand, the fatty acid uptake capacities stayed at same level between either LNCaP cells with expression of single point mutated C-FABP (C-FABPR109A) and control cells (LNCaP) or LNCaP cells with expression of double points mutated C-FABP (C-FABP-R109/129A) and LNCaP cells. These results indicated that elevate the expression of C-FABP in LNCaP cells can result in enhancing its fatty acid uptake capability. Furthermore, it has been also demonstrated that expression of mutated CFABPs generated by site-directed mutagenesis can not increase the fatty acid uptake capability compared to the control cells, which suggested that the ability of C-FABP to bind and to transport fatty acids into these transfectant cells depends on the structural integrity of its fatty acid-binding motif and changing amino acids in this part can result losing most fatty acid-binding and transporting capability. Page | 151 In order to investigate whether up-regulation of intracellular fatty acid signaling activities is associated with tumor malignancy progression, different bioassays were applied to examine the transfected cells’ biological properties. Bioassays showed that the proliferation rate, invasiveness and in vitro tumourigenicity of the LNCaP-WT cells, which expressed high level of wild type C-FABP, capable of up-taking high level of fatty acids, was significantly higher than those of the LNCaP-V cells. These results indicated that increasing C-FABP expression was able to promote the malignant progression in LNCaP cells which did not express C-FABP prior to the transfection. Both LNCaP-R109A and LNCaP-R109/129A cells, which were generated by C-FABP cDNA with point mutations and have a similar ability of up-taking fatty acids, exhibited similar levels of invasiveness and tumourigenicity compared to the control LNCaP-V cells. These results indicated that the tumourigenicitypromoting activity exerted by C-FABP is related to its ability of binding and transporting fatty acids. When transfected cells’ ability of uptake fatty acids was reduced to the similar levels to the control cells by double mutations, C-FABP lost its biological function of promoting tumourigenicity. As a result, LNCaP-R109/129A produced similar number of colonies as LNCaP-V. Furthermore, when the ability of C-FABP to bind and transport fatty acids was reduced, but not completely eliminated by a single mutation as observed in LNCaP-R109A cells, cell growth rate, invasiveness and tumourigenicity is significantly lower than that of the LNCaP-WT cells, but higher than that of the control cells. It appeared that the biological function of C-FABP to promote tumor malignant progression was reduced as its reducing capability of uptake fatty acids which indicated that the increased uptake of fatty acids played a key role for tumourigenicity promotion. Page | 152 When inoculated in nude mice, 7 out of 8 (88.9%) mice inoculated with LNCaP-WT cells developed tumours, but only 3 out of 8 (37.5%) mice inoculated with LNCaP-R109/129A and LNCaP-V cells respectively produced tumours. At the end of the test, the average weight of the tumours produced by LNCaP-WT was 5.8- and 7.3- times of those produced respectively by LNCaP-r109/129a and LNCaP-V cells. This result showed that not only that the wild type C-FABP-expressing cells produced more but also larger tumours than those cells either not expressing C-FABP or expressing mutated C-FABP which was not able to bind and transport fatty acids. In contrast to this finding, our previous work demonstrated that suppressing expression of C-FABP by either antisense mRNA or RNA interference in the highly malignant prostate cancer cell line PC-3M has greatly inhibited its ability of forming tumours in nude mouse [159, 160]. The results of the current and previous work in nude mouse further confirmed that C-FABP promotes malignant progression of prostate cancer by its binding and transporting intracellular fatty acids into the cancer cells. Page | 153 Chapter 5 Effect of expression wild type and mutant C-FABPs on VEGF and angiogenesis 5.1 Introduction Vascular endothelial growth factor (VEGF) plays an important role in angiogenesis which is essential for growth and expansion of solid tumor [176]. Previous studies demonstrated that overexpression of C-FABP in rat Rama 37 cell model significantly increased the level of VEGF mRNA and protein. In addition, the transfected Rama 37 cells with higher expression of C-FABP and VEGF produced larger tumors and metastases (58.8%) when compared to the non-metastasis control cells [158]. Recent studies also showed that suppressed expression of C-FABP in PC-3M cells had greatly reduced their tumorigenecity, probably through down-regulating the expression of VEGF [159]. These results suggested a strong positive correlation between C-FABP and VEGF. Angiogenesis is the process of generating new blood vessel and involves multiple steps: (1) Endothelial cells escape from the stable location by breaking through the basement membrane that envelops the existing blood vessels (2) Endothelial cells migrate toward angiogenic stimulates such as wound associated macrophages or released from cancer cells (3) Endothelial cells proliferate to provide certain amount of cells for new vessels (4) Endothelial cells reorganize into a tubular structure Page | 154 Angiogenesis is a vital process in progression of cancer from small, localized neoplasms to larger, growing tumors. It has been wildly reported that VEGF is a key mediator of angiogenesis in prostate cancer and the expression level of VEGF in advanced prostate cancer is significantly increased compared with non-malignant prostate tissue [177, 178]. The significance of angiogenesis in prostate cancer has been well established and researches showed that angiogenesis has correlated with Gleason score, tumor stage, metastasis and survival rate in prostate cancer [179-181]. Using the chorioallantoic membrane (CAM) assay, Jing et.al found that overexpression of C-FABP in Rama37 cells resulted in an enhanced angiogenic activity in the conditioned medium. On the other hand, the angiogenesis was inhibited in CAM assay when anti-VEGF antibody was added [158], indicating C-FABP may promote angiogenesis through up-regulated VEGF expression. To investigate the possible correlation between the expression level of wild type, mutant C-FABPs and the expression of VEGF, immunohistochemistry, western blot and enzyme-linked immunosorbent assays were carried out to analyze the level of VEGF expression in primary tumors from nude mice, in transfected cell lines and in their conditioned medium respectively. In this chapter, the angiogenesis abilities of each transfected cell lines were also assessed. Page | 155 5.2 Results 5.2.1 Detection of VEGF protein expression in transfectants and their conditioned medium Western blot was applied to measure the level of VEGF expression in transfected cell lines as shown in Figure 5.1A. VEGF expression level in transfected cell lines were quantified using scanning densitometry as shown in Figure 5.1B. The β-actin expression in each cell lines was also analyzed as loading control. The results showed that all four transfectant cell lines expressed VEGF protein (23KDa). Noticeably, VEGF expression in LNCaP-WT was significantly higher compared to the rest of transfectant cell lines (Students t-test P<0.05). When the level of VEGF expression in control LNCaP-V cells was set as 1, the level of VEGF in LNCaP-WT was significantly increased by more than 3-fold (3.48±0.350). In addition, LNCaP-R109A showed higher VEGF expression (1.71±0.294) compared to LNCaP-R109/129A (0.97±0.177) but no significant difference has been found when compared to control cell line LNCaP-V (Students t-test P>0.05). Page | 156 A B Figure 5.1 Expression level of VEGF in LNCaP transfected cell lines (A) The VEGF expression in LNCaP-WT (Lane 1), LNCaP-R109A (Lane 2), Rama37 (Lane 3), LNCaP-109/129A (Lane 4) and LNCaP-V (Lane 5) was analyzed by Western blot. (B) The relative levels of VEGF expression in different transfectants: LNCaP-WT, LNCaP-R109A, LNCaP-109/129A and LNCaP-V were shown in column 1, 2, 3 and 4 respectively. The results (mean±S.D. of three individual experiments) were obtained by densitometry analysis of the band intensities. Page | 157 The level of VEGF protein, which was secreted by transfectant cells, in cell culture supernatant was measured using ELISA analysis and quantified by comparing to a standard curve, which is obtained from a serial dilution of a known-concentration solution of human VEGF, as presented in Figure 5.2A. The results in Figure 5.2B demonstrated that the amount of VEGF protein produced by control transfectant cells LNCaP-V in 100µl conditioned medium (96.0±23.7pg) was almost the same as that produced by LNCaP-R109/129A (93.3±5.2pg). The amount of VEGF protein produced in the same period of time by LNCaP-R109A (109.3±31.4pg) was slightly higher than LNCaP-R109/129A or LNCaP-V but no statistical difference has been observed either between LNCaP-R109A and LNCaP-R109/129A or between LNCaP-R109A and LNCaP-V (Students t-test P>0.05). However, the amount of VEGF protein produced by LNCaP-WT (452.7±44.1pg) was 4.1, 4.9 and 4.7 times greater than those secreted by LNCaP-R109A, LNCaP-R109/129A and LNCaP-V cells respectively. The Students t-test also showed the significant difference between LNCaP-WT and the rest of transfectants (Students t-test P<0.05). Moreover, the relative difference of VEGF expression in conditioned medium was greater than the cell lysate but exhibited a similar pattern. Page | 158 A B Figure 5.2 Analysis VEGF protein level in conditioned medium (A) A standard curve generated for VEGF ELISA analysis. The quantity of VEGF was determined by comparing serially-diluted concentrations from the standard curve. (B) The levels of VEGF expression in conditioned medium of each transfectants - LNCaP-WT, LNCaP-R109A, LNCaP109/129A and LNCaP-V were shown in column 1, 2, 3 and 4 respectively. Page | 159 5.2.2 Detection of C-FABPs, VEGF and angiogenesis in primary tumor tissue from nude mice The expression levels of C-FABPs and VEGF in tumors produced by four groups of nude mice were examined using immunohistochemistry as shown in Table 5.1 and 5.2 respectively. The microvessels were also stained by anti-CD34 in each tumor section as shown in Figure 5.3. Representative picture of immunohistochemical stained sections for each protein was shown in Figture 5.4. The C-FABPs and VEGF expression were quantified according to the scoring system explained in Charpter 2, Section 2.2.8.4. Among the cases examined using C-FABP antibody (Table 5.1), all seven (100%) stumor from LNCaP-WT nude nice group were positively stained. In this group, six out of seven samples (86%) were moderately or strongly stained and only one tumor (14%) was weakly stained. Very similar intensities were observed in tumors produced by LNCaP-R109A and LNCaP-R109/129A cells. All their samples were positively stained by C-FABP antibody. Most of them were stained moderately or strongly positive (LNCaP-R109A: 75%; LNCaP-R109/129A: 100%). For tumors removed from LNCaP-V group, on the other hand, one tumor was unstained and two were weakly stained. Page | 160 Table 5.1 Immunohistochemical detection of C-FABP in tumors produced by different transfectants Strained with C-FABP antibody Tumor Origin No. of Tumors - + ++ +++ Total Positive No. (%) No. (%) No. (%) No. (%) No. (%) LNCaP-WT 7 0 (0) 1 (14) 3 (43) 3 (43) 7 (100) LNCaP-R109A 4 0 (0) 1 (25) 2 (50) 1 (25) 4 (100) LNCaP-R109/129A 3 0 (0) 0 (0) 2 (66) 1 (33) 3 (100) LNCaP-V 3 1 (33) 2 (67) 0 (0) 0 (0) 2 (67) Immunohistochemical staining in each sample is graded negative (-), weakly positive (+), moderately positive (++) and strongly positive (+++) Page | 161 When stained with anti-VEGF antibody (Table 5.2), all seven (100%) tumors from LNCaP-WT group were positive. In this group, more than half of the samples (57%) were stained strongly positive, two (29%) were stained moderately positive and only one (14%) was stained weakly positive. When tumors produced in other groups were stained, most of the samples were weakly positive. For tumors produced by LNCaP-R109A cells, three out four were stained (75%) weakly positive and one was moderately positive. For tumors removed from LNCaP-R109/129A group and LNCaP-V group, all samples (100%) were stained weakly positive. Strong positive staining was only observed in tumors from LNCaP-WT group (four out of seven) and no sample had been stained strongly positive in other groups. Page | 162 Table 5.2 Immunohistochemical detection of VEGF in tumors produced by different transfectants Strained with anti-VEGF antibody Tumors Origin No. of Tumors - + ++ +++ Total Positive No. (%) No. (%) No. (%) No. (%) No. (%) LNCaP-WT 7 0 1 (14) 2 (29) 4 (57) 7 (100) LNCaP-R109A 4 0 3 (75) 1 (25) 0 (0) 4 (100) LNCaP-R109/129A 3 0 3 (100) 0 (0) 0 (0) 3 (100) LNCaP-V 3 0 3 (100) 0 (0) 0 (0) 3 (100) Immunohistochemical staining in each sample is graded negative (-), weakly positive (+), moderately positive (++) and strongly positive (+++) Page | 163 The CD34-related antigen-stained cells were used to assess microvessel density as shown in Figure 5.3. For each section stained, 8 randomly selected fields were counted. The average microvessel density per field in tumors produced by LNCaP-WT cells was 4.67±1.15, compared to 1.67±1.15, 1.33±0.58 and 0.67±0.58 per field in tumors produced by LNCaPR109A, LNCaP-R109/129A and LNCaP-V cells respectively. When the microvessel density per field in tumors produced by LNCaP-V cells was set as 1, the relative microvessel density in tumors produced by LNCaP-WT was significantly (Students t-test P<0.05) increased by 7-fold (7.0±1.73). The relative vessel density in tumors produced by LNCaP-R109A and LNCaP-R109/129A was also slight increased to 2.5±1.73 and 2.0±0.87 respectively but no statistical difference has been found when compare to the density in tumors produced by LNCaP-V (Students t-test P>0.05). Page | 164 Figure 5.3 The relative microvessel densities of tumors produced by transfectants in nude mice The average number of microvessel per field (mean±SD) was calculated from all tumors in each group. The microvessel density in tumors produced by LNCaP-V (column 4) was set as 1 and average numbers of microvessel per field in rest groups were calculated relative to that of tumors produced by LNCaP-V group. Column 1: LNCaP-WT; Column 2: LNCaP-R109A and Column 3: LNCaP-109/129A. Page | 165 1 2 3 A B C D (Original magnification, ×200) Figure 5.4 Immunohistochemical detection of C-FABP, anti-VEGF and CD34 expressed in tumors developed in nude mice Representative samples were immunohistochemically stained with specific antibodies against: C-FABP (1), VEGF (2) and CD34 (3). Samples in each of the four rows were removed from tumors produced by LNCaP-WT (A), LNCaP-R109A (B), LNCaP-109/129A (C) and LNCaP-V (D). Page | 166 5.2.3 Measurement of angiogenic activity of VEGF produced by wild type and mutant C-FABPs transfectants To analysis biological functions of the VEGF produced and secreted by each transfected cell lines, the endothelial tube formation assay was carried out using the Millipore In Vitro Angiogenesis Assay Kit which provided a simple model of angiogenesis in which the induction or inhibition of tube formation by exogenous signals can be easily monitored. The human umbilical vein endothelial cells (HUVEC) were seeded onto the surface of the polymerized ECMatrix with the conditioned medium from each transfectant cells. After 6 hours incubation at 37oC, cells or tubes were visualized by provided dye as shown in Figure 5.5 and the cell networking structures were quantified according to the scoring system explained in Charpter 2, Section 2.2.10.3. The average numerical values of three individual experiments for each transfected cell line were shown in Figure 5.6. The HUVEC tube formation ability was strongly promoted by the conditioned medium from positive control (human VEGF, 10ng/ml) and LNCaP-WT. The HUVEC cells were able to form well-assembled and complex mesh like structures in the presents of conditioned medium from positive control and LNCaP-WT (Figure 5.5). For positive control and LNCaP-WT group, the average numerical value associated with tube formation pattern was 5 and 4.25±0.83 respectively. In contrast, the HUVEC cells remained randomly separated without any signs of organization and unable to complete tube formation when applied with the culture medium only (negative control). For negative control, the average numerical value associated with tube formation pattern Page | 167 was 1.25±0.83. The HUVEC cells were begin to migrate and align themselves and some of them were able to formed capillary tubes in presence of the conditioned medium from LNCaP-R109A, LNCaP-R109/129A and LNCaP-V but no closed polygons pattern were observed. For LNCaP-R109A, LNCaP-R109/129A and LNCaP-V group, the average numerical value associated with tube formation pattern was 2.75±0.44, 2.25±0.83 and 2.5±0.50 respectively. With the treatment of conditioned medium from LNCaP-WT, HUVEC cells tube formation ability were significantly increased by 1.5-, 1.9-, 1.7- and 3.4-fold compared to the effect of LNCaP-R109A, LNCaP-R109/129A, LNCaP-V and negative control respectively (Students t-test P<0.05). Differences of HUVEC cells tube formation ability either between LNCaP-R109A and LNCaP-V or between LNCaP-R109/129A and LNCaP-V were not significant (Students t-test P>0.05). It has also been observed that the number of branch points formed in LNCaP-WT group was higher than that formed in LNCaP-R109A, LNCaP-R109/129A, LNCaP-V and negative control group. Page | 168 A B C D E F Figure 5.5 HUVEC cell network formations on ECMatrix with conditioned medium from different transfected cell lines Images of HUVEC cell network structure were taken after 6 hours incubation in endothelia cell growth media with conditioned medium from different cell lines. Positive control (A), negative control (B), LNCaP-WT (C), LNCaP-R109A (D), LNCaP-R109/129A (E), and LNCaP-V (F). Page | 169 Figure 5.6 Cell network formation values of HUVEC cells with conditioned medium from different transfected cell lines The effect of conditioned medium from different transfected cell lines was assessed and the cell network formation values were quantified according to the cell network pattern. The values from five random fields per well were averaged and three individual assays were performed (mean±SD). Positive control (P), LNCaP-WT (1), LNCaP-R109A (2), LNCaP-R109/129A (3), LNCaP-V (4) and negative control (N). Page | 170 In order to determine whether the biological activities of VEGF played an important role in tube formation, HUVEC cells in conditioned medium from different cell lines were treated with anti-VEGF in the angiogenesis assay. After 6 hours incubation at 37oC, cells were stained with reagent from Millipore In Vitro Angiogenesis Assay Kit as shown in Figure 5.8 and the cell network patterns were analyzed as described previously. With the treatment of anti-VEGF, HUVEC cells in LNCaP-WT conditioned medium were unable to form polygons pattern and most of them were randomly separated without any signs of organization. For LNCaP-WT group, the average numerical value associated with tube formation pattern was 1.75±1.26 which is significantly reduced by 2.4-fold compared to samples without anti-VEGF (Students t-test P<0.05). The average numerical values for the rest of the groups including the negative control stayed as same level as samples without anti-VEGF (Students t-test P>0.05). No statistical difference was found between the average numerical value of LNCaP-WT group and that of LNCaP-V group in presence of anti-VEGF (Students t-test P>0.05). Page | 171 A 1 2 B Figure 5.8 Inhibition of cell network formation by anti-VEGF antibody (A) Images of HUVEC cell network structure were taken after 6 hours incubation with treatment of anti-VEGF antibody in conditioned medium from LNCaP-WT (1) and LNCaP-V (2). (B) The cell network formation values were quantified according to the cell network pattern: LNCaP-WT (1) LNCaP-R109A (2), LNCaP-R109/129A (3) and LNCaP-V (4). The values from five random fields per well were averaged and three individual assays were performed (mean±SD). Page | 172 5.3 Discussion The relationship between the tumourigenicity-promoting function of C-FABP and its ability of binding to and transporting fatty acids were examined and showed that fatty acid-binding and transporting ability is essential for C-FABP to promote tumor growth and expansion. However, whether the increased expression of wild type and mutant C-FABPs are associated with the level of putative target VEGF is still not clear. In the first part of this chapter, the levels of VEGF expression both in transfected cell lines and in their conditioned medium were measured by Western blot and ELISA. These studies demonstrated that VEGF expression in LNCaP cells transfected with C-FABP was significantly increased compared to VEGF level in cells transfected with mutant C-FABPs or the vector. These results suggested that increase expression of C-FABP can elevate the expression level of VEGF in LNCaP cell modle. Our previous studies results on Rama 37 cell model and prostate cancer cell PC-3M model as mentioned previously also supported these findings [158, 159]. VEGF, as the most proangiogenic factor, plays a crucial role on solid tumor development and expansion. It was reported that the order of VEGF expression in different prostate cancer cell lines as judged by intensity of immunohistochemical staining was PC-3 (a parental cells of PC-3M) > Du145 > LNCaP [182]. The malignancy ranking of these prostate cancer cell lines is in similar sequence as VEGF expression levels. In clinical research, the higher levels of VEGF expression in plasma were observed in prostate cancer patients compared with healthy controls. Based on these findings, it has been indicated that VEGF may facilitate to prostate cancer malignant progression possibly by stimulate angiogenesis. Page | 173 In the second part of Chapter 5, the angiogenic function of VEGF produced by different transfected cell lines was examined. Immunohistochemistry was performed to investigate the relationship between C-FABP, VEGF and angiogenesis in tumors produced by nude mice inoculated with different transfected cell lines. The tumors from LNCaP-WT group, strong expression of VEGF has been detected and microvessel densities in these tumors were increased 2.8-, 3.5- and 7.0-fold respectively in comparison to those developed from LNCaP-R109A, LNCaP-R109/129A and LNCaP-V. These results suggested that higher expression of C-FABP can stimulate angiogenesis through upregulation of VEGF. The tube formation assay was also carried out using the Millipore In Vitro Angiogenesis Assay Kit in which the angiogenic function of VEGF in conditioned medium from each transfected cell lines can be qualitatively and quantitatively measured. The results showed that tube formation ability of HUVEC cells with the treatment of conditioned medium from LNCaP-WT cells were significantly increased by 1.5-, 1.9-, 1.7- and 3.4-fold compared to the effect of LNCaP-R109A, LNCaP-R109/129A, LNCaP-V and negative control respectively. On the other hand, the angiogenesis ability of HUVEC cells in LNCaP-WT conditioned medium was suppressed when co-cultured with anti-VEGF antibody. These results combined with those reported previously [158, 159] confirmed that the angiogenesis was indeed produced by VEGF, rather than other possible angiogenesis factors. Page | 174 Chapter 6 General discussion and future work Prostate cancer is the most common form of cancer affecting men in the western countries and the second leading cause of death after lung cancer. In the UK, more than 35,000 men were diagnosed with prostate cancer and around 10,000 men die from prostate cancer each year. Current treatment of prostate cancer mainly relies on a combination of androgen ablation, radiotherapy and chemotherapy. However, tumors often return with more aggressive hormone independent form. So it is important to improve the understanding of molecular mechanisms involved in prostate cancer malignant progression, which may lead to new therapies. The risk factors for prostate cancer include age, ethnicity, family history and diet. It has been reported that prostate cancer onset has been attributed to a variety of diet factors ranging from lack of selenium supplementation [183] to vitamin D [184] and high dairy fat intake was also associated with increased risk of prostate cancer [185-187]. A longterm study demonstrated that high blood levels of trans fatty acids were associated with an increased prostate cancer risk [188]. This thesis investigated the relationship between the tumourigenicity-promoting function of C-FABP and its fatty acid-binding ability. Firstly, transfectant cells with different abilities of binding to fatty acids have been established and the tumourigenicity of these transfectants were compared to assess whether the reduced fatty acid-binding capability is associated with the decreased ability of producing Page | 175 tumours. Like other FABP family proteins, C-FABP binds to fatty acids through a binding motif which consists of 3 key amino acids (Arg109, Arg129, and Tyr131). Thus we used site-directed mutagenesis to convert Arg109or Arg109 and Arg129 into Ala109 or Ala109 and Ala129 to generate two mutant cDNAs and transfected them and wild type cDNA into the LNCaP cells to generate transfectant cell lines. Realtime PCR measurement on C-FABP mRNA showed that the C-FABP mRNA level in LNCaP-WT cells was increased by 137.5-fold comparing with the control cells. Western blot also confirmed the overexpression of C-FABP in LNCaP-WT cells. From the results obtained by our previous studies, it was demonstrated that the biological function of C-FABP is promote malignant progression of cancer cells. In this study, the effect of high level increment of C-FABP expression on cell malignant properties such as proliferation rate, invasiveness and anchorage-independent growth as an indication of tumourigenecicy, was investigated. It is also showed that the mutant C-FABP mRNA levels expressed respectively in LNCaP-R109A and LNCaP-R109/129A cells were also increased to the levels similar to that in LNCaP-WT cells. Thus these transfectants are ideal cell lines used for comparing biological functions exerted by wild type and mutant C-FABPs. As mentioned previously, fatty acid uptake is regulated by certain FABPs in LNCaP cells [175]. Quantitative assessment showed that the relative levels of fatty acid uptakes in LNCaP-WT, LNCaP-R109A and LNCaP-R109/129A were increased 3.1-, 1.6- and 1.4- fold of that detected in control cells. In another word, in LNCaP-WT, LNCaP-R109A and LNCaP-R109/129A cell lines, 210%, 60% and 40% more fatty acid uptakes occurred comparing to that in the control cells. This result showed that Page | 176 increased wild type C-FABP expression by 137.5 times in mRNA level can result in an increment in fatty acid uptake to 3.1- time of that in control; whereas one mutation to the 3 key amino acids in fatty acid binding motif of C-FABP deprived its fatty acid up-taking ability by 72% (1-60%÷210%). Similarly, two mutations to C-FABP caused 81% reduction in its fatty acid up-taking ability. These results suggested that the structural integrity of its fatty acid-binding motif is important for the ability of C-FABP to bind and to transport fatty acids into these transfectant cells and changing one key amino acid in this part can result losing most fatty acid-binding and transporting ability. The biological properties of transfected cells with high up-taking level of fatty acids seem to be much more aggressive when compared to those of control cells, LNCaP-V. When tested in in vitro assays, the proliferation rate, invasiveness and tumorigenicity of LNCaP-WT cells with high level of wild type C-FABP expression were significantly enhanced. However these properties in LNCaP-R109A and LNCaP-R109/129A cells which express high level of mutated C-FABP, stayed at similar level as those in LNCaP-V cells. The results obtained from in vivo assay for tumorigenesis have same the pattern as in vitro assays. When inoculated with LNCaP-WT cells, 88.9% nude mice developed tumors. On the other hand, only 50%, 37.5% and 37.5% of the mice inoculated with LNCaP-R109A, LNCaP-R109/129A and LNCaP-V cells respectively yielded tumuors. The average tumor weight produced in LNCaP-WT group was also increased more than 13-times to 614.3±192mg compared that in LNCaP-V group (46.7±15.3mg). These data showed that increase the ability of up-taking fatty acids in LNCaP cells by elevate C-FABP expression was able to promote the proliferation, invasiveness and tumorigenicity both in vitro and in vivo. It has also been demonstrated Page | 177 that the function of C-FABP to stimulate tumourigenicity was reduced when its up-taking fatty acids capability was decreased. Taken together, these findings indicated that the increased up-taking of fatty acids played an important role in tumourigenicitypromotion. Our previous work demonstrated that suppressing expression of C-FABP in the highly malignant prostate cancer cell line PC-3M has inhibited its ability of forming tumours in nude mouse [189, 190]. The results of the current and previous work further confirmed that C-FABP promotes malignant progression of prostate cancer depending on its binding and transporting intracellular fatty acids capacity. Angiogenesis plays an important role in growth, malignant progression and metastasis by promoting endothelial cell proliferation, invasion and capillary differentiation [191193]. In addition, Weidner et.al has reported that the pathological surrogate for angiogenesis (microvessel density) is correlated with malignancy of prostate cancer [130]. The current studies showed that angiogenesis-associated genes such as VEGF significantly increased the prostate cancer risk. The results in chapter 5 showed that increased expression of biologically active VEGF was detected in wild type C-FABP transfectants, but not in the mutant types of the transfectant cells. These results suggested that increased VEGF expression may be caused by wild type C-FABP and that the ability of C-FABP to increase VEGF depends on its fatty acid-binding ability. Fatty acids appeared to play important roles. Fatty acids were identified as signalling molecules [194] which may be recognised by their nuclear peroxisome proliferatoractivated receptors (PPARs), particularly PPARβ/δ [32]. Page | 178 According to the results in this study and those reported by some other groups, it is reasonable to hypothesis that there may be fatty acid-initiated signalling pathways leading to malignant progression in prostate cancer cells. The route of these pathways may be as following: the elevated expression of C-FABP may give rise to an increased total uptake of fatty acids and enhanced the possible fatty acid signalling activity. Thus, it may be possible that excessive levels of free fatty acids are transported by C-FABP into the nucleus of the cancer cells to activate some as yet unknown mechanisms which might initiate a chain of molecular reactions resulted in promoting cancer growth and expansion. One of such actions regulated through this unknown mechanism is to up-regulate some important down-stream “cancer-promoting” genes, such as VEGF [164, 189], and hence contribute to carcinogenesis. Thus C-FABP which acted as an intracellular fatty acids transporter may play a key role during this process. A schematic hypothesis for the involvement of C-FABP in cancer malignant progression is shown in Figure 8.1. Page | 179 Figure 8.1 Model for possible C-FABP signaling pathways in cancer malignant progression Page | 180 As a first step of testing this hypothesis, we investigated the relationship between the tumourigenicity-promoting function of C-FABP and its ability of binding to and transporting fatty acids in this study and demonstrated that fatty acid-binding and transporting ability is essential for C-FABP to promote tumor growth and expansion. To fully establish this proposed route of fatty acids signaling pathways, the future work should be performed to study the detailed molecular mechanisms on how VEGF is up-regulated? Effort is also needed to study whether there are any other cancerpromoting molecules up-regulated by this mechanism? Page | 181 1. Cancer Incidence and Mortality. 2008, Northern Ireland Cancer Registry. 2. Office for National Statistics, Cancer Statistics registrations: Registrations of cancer diagnosed in 2005, England. 2008, National Statistics 3. Parker, S.L., et al., Cancer statistics, 1997. CA Cancer J Clin, 1997. 47(1): p. 5-27. 4. Parkin, D.M. and C.S. Muir, Cancer Incidence in Five Continents. Comparability and quality of data. IARC Sci Publ, 1992(120): p. 45-173. 5. Hsing, A.W., L. Tsao, and S.S. Devesa, International trends and patterns of prostate cancer incidence and mortality. Int J Cancer, 2000. 85(1): p. 60-7. 6. Tsugane, S., et al., Cancer mortality among Japanese residents of the city of Sao Paulo, Brazil. Int J Cancer, 1990. 45(3): p. 436-9. 7. Whittemore, A.S., et al., Prostate cancer in relation to diet, physical activity, and body size in blacks, whites, and Asians in the United States and Canada. J Natl Cancer Inst, 1995. 87(9): p. 652-61. 8. Carter, H.B., S. Piantadosi, and J.T. Isaacs, Clinical evidence for and implications of the multistep development of prostate cancer. J Urol, 1990. 143(4): p. 742-6. 9. Carter, B.S., et al., Hereditary prostate cancer: epidemiologic and clinical features. J Urol, 1993. 150(3): p. 797-802. 10. Smith, J.R., et al., Major susceptibility locus for prostate cancer on chromosome 1 suggested by a genome-wide search. Science, 1996. 274(5291): p. 1371-4. 11. Xu, J., et al., Evaluation of linkage and association of HPC2/ELAC2 in patients with familial or sporadic prostate cancer. Am J Hum Genet, 2001. 68(4): p. 901-11. 12. Tavtigian, S.V., et al., A candidate prostate cancer susceptibility gene at chromosome 17p. Nat Genet, 2001. 27(2): p. 172-80. 13. Kibel, A.S., et al., Xq27-28 deletions in prostate carcinoma. Genes Chromosomes Cancer, 2003. 37(4): p. 381-8. 14. Cancel-Tassin, G., et al., PCAP is the major known prostate cancer predisposing locus in families from south and west Europe. Eur J Hum Genet, 2001. 9(2): p. 135-42. 15. Carpten, J., et al., Germline mutations in the ribonuclease L gene in families showing linkage with HPC1. Nat Genet, 2002. 30(2): p. 181-4. Page | 182 16. Schaid, D.J., The complex genetic epidemiology of prostate cancer. Hum Mol Genet, 2004. 13 Spec No 1: p. R103-21. 17. Pienta, K.J. and P.S. Esper, Risk factors for prostate cancer. Ann Intern Med, 1993. 118(10): p. 793-803. 18. Armstrong, B. and R. Doll, Environmental factors and cancer incidence and mortality in different countries, with special reference to dietary practices. Int J Cancer, 1975. 15(4): p. 617-31. 19. Rose, D.P., A.P. Boyar, and E.L. Wynder, International comparisons of mortality rates for cancer of the breast, ovary, prostate, and colon, and per capita food consumption. Cancer, 1986. 58(11): p. 2363-71. 20. Strom, S.S., et al., Saturated fat intake predicts biochemical failure after prostatectomy. Int J Cancer, 2008. 122(11): p. 2581-5. 21. Norrish, A.E., et al., Men who consume vegetable oils rich in monounsaturated fat: their dietary patterns and risk of prostate cancer (New Zealand). Cancer Causes Control, 2000. 11(7): p. 609-15. 22. Rose, D.P. and L.A. Cohen, Effects of dietary menhaden oil and retinyl acetate on the growth of DU 145 human prostatic adenocarcinoma cells transplanted into athymic nude mice. Carcinogenesis, 1988. 9(4): p. 603-5. 23. Park, D.J., et al., CCAAT/enhancer binding protein epsilon is a potential retinoid target gene in acute promyelocytic leukemia treatment. J Clin Invest, 1999. 103(10): p. 1399-408. 24. Altucci, L., et al., Retinoic acid-induced apoptosis in leukemia cells is mediated by paracrine action of tumor-selective death ligand TRAIL. Nat Med, 2001. 7(6): p. 680-6. 25. Donato, L.J. and N. Noy, Suppression of mammary carcinoma growth by retinoic acid: proapoptotic genes are targets for retinoic acid receptor and cellular retinoic acid-binding protein II signaling. Cancer Res, 2005. 65(18): p. 8193-9. 26. Soprano, D.R., P. Qin, and K.J. Soprano, Retinoic acid receptors and cancers. Annu Rev Nutr, 2004. 24: p. 201-21. Page | 183 27. Reichman, M.E., et al., Serum vitamin A and subsequent development of prostate cancer in the first National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. Cancer Res, 1990. 50(8): p. 2311-5. 28. Hsing, A.W., et al., Serologic precursors of cancer. Retinol, carotenoids, and tocopherol and risk of prostate cancer. J Natl Cancer Inst, 1990. 82(11): p. 941-6. 29. Jacobs, S., et al., Retinoic acid is required early during adult neurogenesis in the dentate gyrus. Proc Natl Acad Sci U S A, 2006. 103(10): p. 3902-7. 30. Plum, L.A., et al., Retinoic acid combined with neurotrophin-3 enhances the survival and neurite outgrowth of embryonic sympathetic neurons. Exp Biol Med (Maywood), 2001. 226(8): p. 766-75. 31. Verma, A.K., E.A. Conrad, and R.K. Boutwell, Differential effects of retinoic acid and 7,8-benzoflavone on the induction of mouse skin tumors by the complete carcinogenesis process and by the initiation-promotion regimen. Cancer Res, 1982. 42(9): p. 3519-25. 32. Schug, T.T., et al., Opposing effects of retinoic acid on cell growth result from alternate activation of two different nuclear receptors. Cell, 2007. 129(4): p. 723-33. 33. Mills, P.K., et al., Cohort study of diet, lifestyle, and prostate cancer in Adventist men. Cancer, 1989. 64(3): p. 598-604. 34. Yan, L. and E.L. Spitznagel, Soy consumption and prostate cancer risk in men: a revisit of a meta-analysis. Am J Clin Nutr, 2009. 89(4): p. 1155-63. 35. Badger, T.M., et al., Soy protein isolate and protection against cancer. J Am Coll Nutr, 2005. 24(2): p. 146S-149S. 36. Clemente, C.D., Gray's Anatomy of the Human Body 30th ed. 1985: Lea and Febiger: Philadelphia. 37. Costello, L.C. and R.B. Franklin, Citrate metabolism of normal and malignant prostate epithelial cells. Urology, 1997. 50(1): p. 3-12. 38. Wang, M.C., et al., Purification of a human prostate specific antigen. Invest Urol, 1979. 17(2): p. 159-63. 39. Abate-Shen, C. and M.M. Shen, Molecular genetics of prostate cancer. Genes Dev, 2000. 14(19): p. 2410-34. Page | 184 40. Cohen, R.J., et al., Central zone carcinoma of the prostate gland: a distinct tumor type with poor prognostic features. J Urol, 2008. 179(5): p. 1762-7; discussion 1767. 41. Curran, S., et al., Endorectal MRI of prostatic and periprostatic cystic lesions and their mimics. AJR Am J Roentgenol, 2007. 188(5): p. 1373-9. 42. Denmeade, S.R., X.S. Lin, and J.T. Isaacs, Role of programmed (apoptotic) cell death during the progression and therapy for prostate cancer. Prostate, 1996. 28(4): p. 251-65. 43. Okada, H., et al., Keratin profiles in normal/hyperplastic prostates and prostate carcinoma. Virchows Arch A Pathol Anat Histopathol, 1992. 421(2): p. 157-61. 44. Bok, R.A. and E.J. Small, Bloodborne biomolecular markers in prostate cancer development and progression. Nat Rev Cancer, 2002. 2(12): p. 918-26. 45. Bonkhoff, H. and K. Remberger, Widespread distribution of nuclear androgen receptors in the basal cell layer of the normal and hyperplastic human prostate. Virchows Arch A Pathol Anat Histopathol, 1993. 422(1): p. 35-8. 46. Harper, M.E., et al., Expression of androgen receptor and growth factors in premalignant lesions of the prostate. J Pathol, 1998. 186(2): p. 169-77. 47. Foster, C.S. and Y. Ke, Stem cells in prostatic epithelia. Int J Exp Pathol, 1997. 78(5): p. 311-29. 48. Ather, M.H. and F. Abbas, Prognostic significance of neuroendocrine differentiation in prostate cancer. Eur Urol, 2000. 38(5): p. 535-42. 49. Abrahamsson, P.A., Neuroendocrine differentiation in prostatic carcinoma. Prostate, 1999. 39(2): p. 135-48. 50. Roehrborn, C.G., Clinical management of lower urinary tract symptoms with combined medical therapy. BJU Int, 2008. 102 Suppl 2: p. 13-7. 51. Foster, C.S., et al., Cellular and molecular pathology of prostate cancer precursors. Scand J Urol Nephrol Suppl, 2000(205): p. 19-43. 52. Joniau, S., et al., Prostatic intraepithelial neoplasia (PIN): importance and clinical management. Eur Urol, 2005. 48(3): p. 379-85. 53. Bostwick, D.G., Prostatic intraepithelial neoplasia (PIN). Urology, 1989. 34(6 Suppl): p. 16-22. Page | 185 54. Ayala, A.G. and J.Y. Ro, Prostatic intraepithelial neoplasia: recent advances. Arch Pathol Lab Med, 2007. 131(8): p. 1257-66. 55. Gleason, D.F., Histological grading and clinical staging of prostatic carcinoma, in Urologic pathology: the prostate. Lea & Febiger, Philadelphia, T. M, Editor. 1977. p. 171-198. 56. Cussenot, O., et al., Immortalization of human adult normal prostatic epithelial cells by liposomes containing large T-SV40 gene. J Urol, 1991. 146(3): p. 881-6. 57. Berthon, P., Cussenot, O., Hopwood, L., Le Duc, A., Maitland, N.J, Functional expression of SV40 in normal human prostatic epithelial and fibroblastic cells: Differentiation pattern of non-tumorigenic cell lines. International Journal of Oncology, 1995. 6(2): p. 333–343. 58. Papsidero, L.D., et al., Prostate antigen: a marker for human prostate epithelial cells. J Natl Cancer Inst, 1981. 66(1): p. 37-42. 59. Horoszewicz, J.S., et al., The LNCaP cell line--a new model for studies on human prostatic carcinoma. Prog Clin Biol Res, 1980. 37: p. 115-32. 60. Horoszewicz, J.S., et al., LNCaP model of human prostatic carcinoma. Cancer Res, 1983. 43(4): p. 1809-18. 61. Tuxhorn, J.A., et al., Stromal cells promote angiogenesis and growth of human prostate tumors in a differential reactive stroma (DRS) xenograft model. Cancer Res, 2002. 62(11): p. 3298-307. 62. Mitchell, S., et al., Phenotypic and genotypic characterization of commonly used human prostatic cell lines. BJU Int, 2000. 85(7): p. 932-44. 63. Mickey, D.D., et al., Heterotransplantation of a human prostatic adenocarcinoma cell line in nude mice. Cancer Res, 1977. 37(11): p. 4049-58. 64. Kaighn, M.E., et al., Establishment and characterization of a human prostatic carcinoma cell line (PC-3). Invest Urol, 1979. 17(1): p. 16-23. 65. Shevrin, D.H., K.I. Gorny, and S.C. Kukreja, Patterns of metastasis by the human prostate cancer cell line PC-3 in athymic nude mice. Prostate, 1989. 15(2): p. 187-94. 66. Kozlowski, J.M., et al., Metastatic behavior of human tumor cell lines grown in the nude mouse. Cancer Res, 1984. 44(8): p. 3522-9. Page | 186 67. Grandori, C., et al., The Myc/Max/Mad network and the transcriptional control of cell behavior. Annu Rev Cell Dev Biol, 2000. 16: p. 653-99. 68. Nupponen, N.N., et al., Genetic alterations in hormone-refractory recurrent prostate carcinomas. Am J Pathol, 1998. 153(1): p. 141-8. 69. Jenkins, R.B., et al., Detection of c-myc oncogene amplification and chromosomal anomalies in metastatic prostatic carcinoma by fluorescence in situ hybridization. Cancer Res, 1997. 57(3): p. 524-31. 70. Coussens, L., et al., Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with neu oncogene. Science, 1985. 230(4730): p. 1132-9. 71. Craft, N., et al., A mechanism for hormone-independent prostate cancer through modulation of androgen receptor signaling by the HER-2/neu tyrosine kinase. Nat Med, 1999. 5(3): p. 280-5. 72. Sadasivan, R., et al., Overexpression of Her-2/neu may be an indicator of poor prognosis in prostate cancer. J Urol, 1993. 150(1): p. 126-31. 73. Katsumata, M., et al., Differential effects of Bcl-2 on T and B cells in transgenic mice. Proc Natl Acad Sci U S A, 1992. 89(23): p. 11376-80. 74. Gleave, M., et al., Progression to androgen independence is delayed by adjuvant treatment with antisense Bcl-2 oligodeoxynucleotides after castration in the LNCaP prostate tumor model. Clin Cancer Res, 1999. 5(10): p. 2891-8. 75. Varambally, S., et al., The polycomb group protein EZH2 is involved in progression of prostate cancer. Nature, 2002. 419(6907): p. 624-9. 76. Visser, H.P., et al., The Polycomb group protein EZH2 is upregulated in proliferating, cultured human mantle cell lymphoma. Br J Haematol, 2001. 112(4): p. 950-8. 77. Bos, J.L., The ras gene family and human carcinogenesis. Mutat Res, 1988. 195(3): p. 255-71. 78. Santos, E. and A.R. Nebreda, Structural and functional properties of ras proteins. FASEB J, 1989. 3(10): p. 2151-63. 79. Fan, K., Heterogeneous subpopulations of human prostatic adenocarcinoma cells: potential usefulness of P21 protein as a predictor for bone metastasis. J Urol, 1988. 139(2): p. 318-22. Page | 187 80. Cadwell, C. and G.P. Zambetti, The effects of wild-type p53 tumor suppressor activity and mutant p53 gain-of-function on cell growth. Gene, 2001. 277(1-2): p. 15-30. 81. Yonish-Rouach, E., et al., Wild-type p53 induces apoptosis of myeloid leukaemic cells that is inhibited by interleukin-6. Nature, 1991. 352(6333): p. 345-7. 82. Balint, E.E. and K.H. Vousden, Activation and activities of the p53 tumour suppressor protein. Br J Cancer, 2001. 85(12): p. 1813-23. 83. Navone, N.M., et al., p53 mutations in prostate cancer bone metastases suggest that selected p53 mutants in the primary site define foci with metastatic potential. J Urol, 1999. 161(1): p. 304-8. 84. Meyers, F.J., et al., Very frequent p53 mutations in metastatic prostate carcinoma and in matched primary tumors. Cancer, 1998. 83(12): p. 2534-9. 85. Cooney, K.A., et al., Distinct regions of allelic loss on 13q in prostate cancer. Cancer Res, 1996. 56(5): p. 1142-5. 86. McCall, P., et al., Is PTEN loss associated with clinical outcome measures in human prostate cancer? Br J Cancer, 2008. 99(8): p. 1296-301. 87. Cairns, P., et al., Frequent inactivation of PTEN/MMAC1 in primary prostate cancer. Cancer Res, 1997. 57(22): p. 4997-5000. 88. Meiers, I., J.H. Shanks, and D.G. Bostwick, Glutathione S-transferase pi (GSTP1) hypermethylation in prostate cancer: review 2007. Pathology, 2007. 39(3): p. 299-304. 89. Harden, S.V., et al., Quantitative GSTP1 methylation clearly distinguishes benign prostatic tissue and limited prostate adenocarcinoma. J Urol, 2003. 169(3): p. 1138-42. 90. Fidler, I.J., The pathogenesis of cancer metastasis: the 'seed and soil' hypothesis revisited. Nat Rev Cancer, 2003. 3(6): p. 453-8. 91. Paget, S., The distribution of secondary growths in cancer of the breast. Lancet 1, 1889: p. 571-573. 92. Clarke, N.W., C.A. Hart, and M.D. Brown, Molecular mechanisms of metastasis in prostate cancer. Asian J Androl, 2009. 11(1): p. 57-67. Page | 188 93. Welch, D.R., P.S. Steeg, and C.W. Rinker-Schaeffer, Molecular biology of breast cancer metastasis. Genetic regulation of human breast carcinoma metastasis. Breast Cancer Res, 2000. 2(6): p. 408-16. 94. Bubendorf, L., et al., Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Hum Pathol, 2000. 31(5): p. 578-83. 95. Bogdanos, J., et al., Endocrine/paracrine/autocrine survival factor activity of bone microenvironment participates in the development of androgen ablation and chemotherapy refractoriness of prostate cancer metastasis in skeleton. Endocr Relat Cancer, 2003. 10(2): p. 279-89. 96. Msaouel, P., et al., Mechanisms of bone metastasis in prostate cancer: clinical implications. Best Pract Res Clin Endocrinol Metab, 2008. 22(2): p. 341-55. 97. Hsu, H., et al., Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation and activation induced by osteoprotegerin ligand. Proc Natl Acad Sci U S A, 1999. 96(7): p. 3540-5. 98. Holen, I., et al., Osteoprotegerin (OPG) is a survival factor for human prostate cancer cells. Cancer Res, 2002. 62(6): p. 1619-23. 99. Mori, K., et al., DU145 human prostate cancer cells express functional receptor activator of NFkappaB: new insights in the prostate cancer bone metastasis process. Bone, 2007. 40(4): p. 981-90. 100. Aubin, J.E., Bone stem cells. J Cell Biochem Suppl, 1998. 30-31: p. 73-82. 101. Medinger, M., et al., Angiogenesis and the ET-1/ETA receptor system: immunohistochemical expression analysis in bone metastases from patients with different primary tumors. Angiogenesis, 2003. 6(3): p. 225-31. 102. Kojima, S., et al., Implications of insulin-like growth factor-I for prostate cancer therapies. Int J Urol, 2009. 16(2): p. 161-7. 103. Sharifi, N., J.L. Gulley, and W.L. Dahut, Androgen deprivation therapy for prostate cancer. JAMA, 2005. 294(2): p. 238-44. 104. Kuiper, G.G., et al., Structural organization of the human androgen receptor gene. J Mol Endocrinol, 1989. 2(3): p. R1-4. 105. Feldman, B.J. and D. Feldman, The development of androgen-independent prostate cancer. Nat Rev Cancer, 2001. 1(1): p. 34-45. Page | 189 106. Chen, C.D., et al., Molecular determinants of resistance to antiandrogen therapy. Nat Med, 2004. 10(1): p. 33-9. 107. Marcelli, M., et al., Androgen receptor mutations in prostate cancer. Cancer Res, 2000. 60(4): p. 944-9. 108. Tilley, W.D., et al., Mutations in the androgen receptor gene are associated with progression of human prostate cancer to androgen independence. Clin Cancer Res, 1996. 2(2): p. 277-85. 109. Veldscholte, J., et al., The androgen receptor in LNCaP cells contains a mutation in the ligand binding domain which affects steroid binding characteristics and response to antiandrogens. J Steroid Biochem Mol Biol, 1992. 41(3-8): p. 665-9. 110. Hara, T., et al., Novel mutations of androgen receptor: a possible mechanism of bicalutamide withdrawal syndrome. Cancer Res, 2003. 63(1): p. 149-53. 111. Veldscholte, J., et al., A mutation in the ligand binding domain of the androgen receptor of human LNCaP cells affects steroid binding characteristics and response to anti-androgens. Biochem Biophys Res Commun, 1990. 173(2): p. 534-40. 112. Cohen, M.B. and O.W. Rokhlin, Mechanisms of prostate cancer cell survival after inhibition of AR expression. J Cell Biochem, 2009. 106(3): p. 363-71. 113. McDonnell, T.J., et al., Expression of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res, 1992. 52(24): p. 6940-4. 114. Culig, Z., et al., Regulation of prostatic growth and function by peptide growth factors. Prostate, 1996. 28(6): p. 392-405. 115. Culig, Z., et al., Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. Cancer Res, 1994. 54(20): p. 5474-8. 116. Gregory, C.W., et al., Epidermal growth factor increases coactivation of the androgen receptor in recurrent prostate cancer. J Biol Chem, 2004. 279(8): p. 7119-30. Page | 190 117. Joseph, I.B., et al., Androgens regulate vascular endothelial growth factor content in normal and malignant prostatic tissue. Clin Cancer Res, 1997. 3(12 Pt 1): p. 2507-11. 118. Boddy, J.L., et al., The androgen receptor is significantly associated with vascular endothelial growth factor and hypoxia sensing via hypoxia-inducible factors HIF-1a, HIF-2a, and the prolyl hydroxylases in human prostate cancer. Clin Cancer Res, 2005. 11(21): p. 7658-63. 119. Banham, A.H., et al., Expression of the forkhead transcription factor FOXP1 is associated both with hypoxia inducible factors (HIFs) and the androgen receptor in prostate cancer but is not directly regulated by androgens or hypoxia. Prostate, 2007. 67(10): p. 1091-8. 120. Auguste, P., S. Javerzat, and A. Bikfalvi, Regulation of vascular development by fibroblast growth factors. Cell Tissue Res, 2003. 314(1): p. 157-66. 121. Forough, R., et al., Role of AKT/PKB signaling in fibroblast growth factor-1 (FGF-1)-induced angiogenesis in the chicken chorioallantoic membrane (CAM). J Cell Biochem, 2005. 94(1): p. 109-16. 122. Hori, A., et al., Suppression of solid tumor growth by immunoneutralizing monoclonal antibody against human basic fibroblast growth factor. Cancer Res, 1991. 51(22): p. 6180-4. 123. Shou, Y., et al., Influence of angiogenetic factors and matrix metalloproteinases upon tumour progression in non-small-cell lung cancer. Br J Cancer, 2001. 85(11): p. 1706-12. 124. Volm, M., R. Koomagi, and J. Mattern, PD-ECGF, bFGF, and VEGF expression in non-small cell lung carcinomas and their association with lymph node metastasis. Anticancer Res, 1999. 19(1B): p. 651-5. 125. Ito, H., et al., Expression of vascular endothelial growth factor and basic fibroblast growth factor in small adenocarcinomas. Oncol Rep, 2002. 9(1): p. 119-23. 126. Ferrara, N., Vascular endothelial growth factor: basic science and clinical progress. Endocr Rev, 2004. 25(4): p. 581-611. 127. Vincenti, V., et al., Assignment of the vascular endothelial growth factor gene to human chromosome 6p21.3. Circulation, 1996. 93(8): p. 1493-5. Page | 191 128. Nowak, D.G., et al., Expression of pro- and anti-angiogenic isoforms of VEGF is differentially regulated by splicing and growth factors. J Cell Sci, 2008. 121(Pt 20): p. 3487-95. 129. Huss, W.J., et al., Angiogenesis and prostate cancer: identification of a molecular progression switch. Cancer Res, 2001. 61(6): p. 2736-43. 130. Weidner, N., et al., Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. Am J Pathol, 1993. 143(2): p. 401-9. 131. Du, Z., et al., Expression of hypoxia-inducible factor 1alpha in human normal, benign, and malignant prostate tissue. Chin Med J (Engl), 2003. 116(12): p. 1936-9. 132. Bok, R.A., et al., Vascular endothelial growth factor and basic fibroblast growth factor urine levels as predictors of outcome in hormone-refractory prostate cancer patients: a cancer and leukemia group B study. Cancer Res, 2001. 61(6): p. 2533-6. 133. Duque, J.L., et al., Plasma levels of vascular endothelial growth factor are increased in patients with metastatic prostate cancer. Urology, 1999. 54(3): p. 523-7. 134. J. J. Karakunnel, J.L.G., P. Arlen, M. Mulquin, J. Wright, I. B. Turkbey, P. Choyke, W. D. Figg and W. Dahut Cediranib (AZD2171) in docetaxel-resistant, castration-resistant prostate cancer (CRPC). in 2009 ASCO Annual Meeting 2009: Journal of Clinical Oncology. 135. Dreyer, C., et al., Control of the peroxisomal beta-oxidation pathway by a novel family of nuclear hormone receptors. Cell, 1992. 68(5): p. 879-87. 136. Juge-Aubry, C.E., et al., Regulation of the transcriptional activity of the peroxisome proliferator-activated receptor alpha by phosphorylation of a ligand-independent trans-activating domain. J Biol Chem, 1999. 274(15): p. 10505-10. 137. Hihi, A.K., L. Michalik, and W. Wahli, PPARs: transcriptional effectors of fatty acids and their derivatives. Cell Mol Life Sci, 2002. 59(5): p. 790-8. 138. Keller, H., et al., Fatty acids and retinoids control lipid metabolism through activation of peroxisome proliferator-activated receptor-retinoid X receptor heterodimers. Proc Natl Acad Sci U S A, 1993. 90(6): p. 2160-4. Page | 192 139. Peters, J.M. and F.J. Gonzalez, Sorting out the functional role(s) of peroxisome proliferator-activated receptor-beta/delta (PPARbeta/delta) in cell proliferation and cancer. Biochim Biophys Acta, 2009. 1796(2): p. 230-41. 140. Stephen, R.L., et al., Activation of peroxisome proliferator-activated receptor delta stimulates the proliferation of human breast and prostate cancer cell lines. Cancer Res, 2004. 64(9): p. 3162-70. 141. Piqueras, L., et al., Activation of PPARbeta/delta induces endothelial cell proliferation and angiogenesis. Arterioscler Thromb Vasc Biol, 2007. 27(1): p. 63-9. 142. Han, J.K., et al., Peroxisome proliferator-activated receptor-delta agonist enhances vasculogenesis by regulating endothelial progenitor cells through genomic and nongenomic activations of the phosphatidylinositol 3-kinase/Akt pathway. Circulation, 2008. 118(10): p. 1021-33. 143. Fauconnet, S., et al., Differential regulation of vascular endothelial growth factor expression by peroxisome proliferator-activated receptors in bladder cancer cells. J Biol Chem, 2002. 277(26): p. 23534-43. 144. Haslmayer, P., et al., The peroxisome proliferator-activated receptor gamma ligand 15-deoxy-Delta12,14-prostaglandin J2 induces vascular endothelial growth factor in the hormone-independent prostate cancer cell line PC 3 and the urinary bladder carcinoma cell line 5637. Int J Oncol, 2002. 21(4): p. 915-20. 145. Pino, M.V., M.F. Kelley, and Z. Jayyosi, Promotion of colon tumors in C57BL/6JAPC(min)/+ mice by thiazolidinedione PPARgamma agonists and a structurally unrelated PPARgamma agonist. Toxicol Pathol, 2004. 32(1): p. 58-63. 146. Furuhashi, M. and G.S. Hotamisligil, Fatty acid-binding proteins: role in metabolic diseases and potential as drug targets. Nat Rev Drug Discov, 2008. 7(6): p. 489-503. 147. Zimmerman, A.W. and J.H. Veerkamp, New insights into the structure and function of fatty acid-binding proteins. Cell Mol Life Sci, 2002. 59(7): p. 1096-116. 148. Haunerland, N.H. and F. Spener, Fatty acid-binding proteins--insights from genetic manipulations. Prog Lipid Res, 2004. 43(4): p. 328-49. 149. Veerkamp, J.H. and H.T. van Moerkerk, Fatty acid-binding protein and its relation to fatty acid oxidation. Mol Cell Biochem, 1993. 123(1-2): p. 101-6. Page | 193 150. Chmurzynska, A., The multigene family of fatty acid-binding proteins (FABPs): function, structure and polymorphism. J Appl Genet, 2006. 47(1): p. 39-48. 151. Sacchettini, J.C., J.I. Gordon, and L.J. Banaszak, The structure of crystalline Escherichia coli-derived rat intestinal fatty acid-binding protein at 2.5-A resolution. J Biol Chem, 1988. 263(12): p. 5815-9. 152. Xu, Z., D.A. Bernlohr, and L.J. Banaszak, The adipocyte lipid-binding protein at 1.6-A resolution. Crystal structures of the apoprotein and with bound saturated and unsaturated fatty acids. J Biol Chem, 1993. 268(11): p. 7874-84. 153. LaLonde, J.M., et al., Adipocyte lipid-binding protein complexed with arachidonic acid. Titration calorimetry and X-ray crystallographic studies. J Biol Chem, 1994. 269(41): p. 25339-47. 154. Wolfrum, C., et al., Fatty acids and hypolipidemic drugs regulate peroxisome proliferator-activated receptors alpha - and gamma-mediated gene expression via liver fatty acid binding protein: a signaling path to the nucleus. Proc Natl Acad Sci U S A, 2001. 98(5): p. 2323-8. 155. Schroeder, F., et al., Role of fatty acid binding proteins and long chain fatty acids in modulating nuclear receptors and gene transcription. Lipids, 2008. 43(1): p. 1-17. 156. Tan, N.S., et al., Selective cooperation between fatty acid binding proteins and peroxisome proliferator-activated receptors in regulating transcription. Mol Cell Biol, 2002. 22(14): p. 5114-27. 157. Jing, C., et al., Identification of the messenger RNA for human cutaneous fatty acid-binding protein as a metastasis inducer. Cancer Res, 2000. 60(9): p. 2390-8. 158. Jing, C., et al., Human cutaneous fatty acid-binding protein induces metastasis by up-regulating the expression of vascular endothelial growth factor gene in rat Rama 37 model cells. Cancer Res, 2001. 61(11): p. 4357-64. 159. Adamson, J., et al., High-level expression of cutaneous fatty acid-binding protein in prostatic carcinomas and its effect on tumorigenicity. Oncogene, 2003. 22(18): p. 2739-49. 160. Morgan, E.A., et al., Expression of cutaneous fatty acid-binding protein (CFABP) in prostate cancer: potential prognostic marker and target for tumourigenicity-suppression. Int J Oncol, 2008. 32(4): p. 767-75. Page | 194 161. Kassis, J., et al., A role for phospholipase C-gamma-mediated signaling in tumor cell invasion. Clin Cancer Res, 1999. 5(8): p. 2251-60. 162. Janik, P., P. Briand, and N.R. Hartmann, The effect of estrone-progesterone treatment on cell proliferation kinetics of hormone-dependent GR mouse mammary tumors. Cancer Res, 1975. 35(12): p. 3698-704. 163. Remmele, W. and H.E. Stegner, [Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue]. Pathologe, 1987. 8(3): p. 138-40. 164. Chen, H.J., et al., Angiogenically active vascular endothelial growth factor is over-expressed in malignant human and rat prostate carcinoma cells. Br J Cancer, 2000. 82(10): p. 1694-701. 165. Vorum, H., et al., Expression of recombinant psoriasis-associated fatty acid binding protein in Escherichia coli: gel electrophoretic characterization, analysis of binding properties and comparison with human serum albumin. Electrophoresis, 1998. 19(10): p. 1793-802. 166. Bennett, S., A. Joshua, and P.J. Russell, Reliable method of isolating transfected clones from the LNCaP human prostatic cell line. Biotechniques, 1997. 23(1): p. 66, 68, 70. 167. Orth, P., et al., Analysis of novel nonviral gene transfer systems for gene delivery to cells of the musculoskeletal system. Mol Biotechnol, 2008. 38(2): p. 137-44. 168. Hughes-Fulford, M., Y. Chen, and R.R. Tjandrawinata, Fatty acid regulates gene expression and growth of human prostate cancer PC-3 cells. Carcinogenesis, 2001. 22(5): p. 701-7. 169. Collett, G.P., et al., Peroxisome proliferator-activated receptor alpha is an androgen-responsive gene in human prostate and is highly expressed in prostatic adenocarcinoma. Clin Cancer Res, 2000. 6(8): p. 3241-8. 170. Rose, D.P., Dietary fatty acids and cancer. Am J Clin Nutr, 1997. 66(4 Suppl): p. 998S-1003S. 171. Fang, L.Y., et al., Fatty-acid-binding protein 5 promotes cell proliferation and invasion in oral squamous cell carcinoma. J Oral Pathol Med, 2009. Page | 195 172. Gleave, M., et al., Acceleration of human prostate cancer growth in vivo by factors produced by prostate and bone fibroblasts. Cancer Res, 1991. 51(14): p. 3753-61. 173. Tuszynski, G.P., et al., Thrombospondin, a potentiator of tumor cell metastasis. Cancer Res, 1987. 47(15): p. 4130-3. 174. Ke, Y., et al., A rapid procedure for production of human basic fibroblast growth factor in Escherichia coli cells. Biochim Biophys Acta, 1992. 1131(3): p. 307-10. 175. Pinthus, J.H., et al., Androgen-dependent regulation of medium and long chain fatty acids uptake in prostate cancer. Prostate, 2007. 67(12): p. 1330-8. 176. Ferrara, N., Vascular endothelial growth factor. Eur J Cancer, 1996. 32A(14): p. 2413-22. 177. Ryan, C.J., A.M. Lin, and E.J. Small, Angiogenesis inhibition plus chemotherapy for metastatic hormone refractory prostate cancer: history and rationale. Urol Oncol, 2006. 24(3): p. 250-3. 178. Bigler, S.A., R.E. Deering, and M.K. Brawer, Comparison of microscopic vascularity in benign and malignant prostate tissue. Hum Pathol, 1993. 24(2): p. 220-6. 179. Ferrer, F.A., et al., Angiogenesis and prostate cancer: in vivo and in vitro expression of angiogenesis factors by prostate cancer cells. Urology, 1998. 51(1): p. 161-7. 180. Borre, M., et al., Microvessel density predicts survival in prostate cancer patients subjected to watchful waiting. Br J Cancer, 1998. 78(7): p. 940-4. 181. Bono, A.V., et al., Microvessel density in prostate carcinoma. Prostate Cancer Prostatic Dis, 2002. 5(2): p. 123-7. 182. Harper, M.E., et al., Vascular endothelial growth factor (VEGF) expression in prostatic tumours and its relationship to neuroendocrine cells. Br J Cancer, 1996. 74(6): p. 910-6. 183. Etminan, M., et al., Intake of selenium in the prevention of prostate cancer: a systematic review and meta-analysis. Cancer Causes Control, 2005. 16(9): p. 1125-31. Page | 196 184. Stewart, L.V., et al., Vitamin D receptor agonists induce prostatic acid phosphatase to reduce cell growth and HER-2 signaling in LNCaP-derived human prostate cancer cells. J Steroid Biochem Mol Biol, 2005. 97(1-2): p. 37-46. 185. Kurahashi, N., et al., Dairy product, saturated fatty acid, and calcium intake and prostate cancer in a prospective cohort of Japanese men. Cancer Epidemiol Biomarkers Prev, 2008. 17(4): p. 930-7. 186. Crowe, F.L., et al., Fatty acid composition of plasma phospholipids and risk of prostate cancer in a case-control analysis nested within the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr, 2008. 88(5): p. 1353-63. 187. Hughes-Fulford, M., et al., Arachidonic acid, an omega-6 fatty acid, induces cytoplasmic phospholipase A2 in prostate carcinoma cells. Carcinogenesis, 2005. 26(9): p. 1520-6. 188. Chavarro, J.E., et al., A prospective study of trans-fatty acid levels in blood and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev, 2008. 17(1): p. 95-101. 189. Adamson, J., et al., High-level expression of cutaneous fatty acid-binding protein in prostatic carcinomas and its effect on tumorigenicity. Oncogene, 2003. 22: p. 2739-2749. 190. Morgan, E.A., et al., Expression of cutaneous fatty acid-binding protein (CFABP) in prostate cancer: Potential prognostic marker and target for tumourigenicity-supression. Int. J. Oncol., 2008. 32: p. 767-775. 191. VanCleave, T.T., et al., Interaction among variant vascular endothelial growth factor (VEGF) and its receptor in relation to prostate cancer risk. Prostate. 70(4): p. 341-52. 192. Kaya, A., et al., The prognostic significance of vascular endothelial growth factor levels in sera of non-small cell lung cancer patients. Respir Med, 2004. 98(7): p. 632-6. 193. Sfar, S., et al., Association of VEGF genetic polymorphisms with prostate carcinoma risk and clinical outcome. Cytokine, 2006. 35(1-2): p. 21-8. 194. Glatz, J.F., et al., Role of membrane-associated and cytoplasmic fatty acidbinding proteins in cellular fatty acid metabolism. Prostaglandins Leukot Essent Fatty Acids, 1997. 57: p. 373-8. Page | 197