The roles of osteoclasts and RUNX2 in the progression of prostate cancer bone metasta- ses Junchi Huang Department of Urology , Institute of Clinical Sciences at Sahlgrenska Academy University of Gothenburg Gothenburg, Sweden, 2021 The roles of osteoclasts and RUNX2 in the progression of prostate can- cer bone metastases © 2021 Junchi Huang Junchi.huang@gu.se ISBN: 978-91-8009-550-1 (PRINT) ISBN: 978-91-8009-551-8 (PDF) Printed in Borås, Sweden 2021 Printed by Stema Specialtryck AB Cover illustration is immunohistochemistry image of LNCaP-19 bone me- tastases To my family Abstract Metastasis to the skeleton is the major cause of death from prostate cancer (PC). Patients with metastatic PC are treated with androgen deprivation therapy (ADT) to decrease testosterone levels and thereby inhibit further growth of the tumor. However, a castration-resistant prostate cancer (CRPC) inevitably develops, often with activated androgen receptor (AR) signaling due to increased sensitivity of the AR or intratumoral steroido- genesis enabling AR activation despite castrate levels of testosterone in the circulation. In the skeleton, PC cells interact with the bone microenvironment. It is known that degradation of bone matrix, one effect of androgen deprivation, releases growth factors stimulating tumor growth. We have previously identified the role of osteoblasts, bone-building cells, in promoting tumor growth and intratumoral steroidogenesis. A direct effect of their balancing counterpart, the bone-degrading osteoclasts, has previously not been inves- tigated. In the present thesis, osteoclasts were found to stimulate proliferation and inhibit apoptosis of both osteolytic and osteoblastic CRPC cells in an in vitro co-culture model. Gene expression was affected by osteoclast co-cul- ture, more extensively so in the osteolytic model, where for example DNA repair genes were upregulated by osteoclasts. In both cell lines, genes re- lated to endoplasmic reticulum stress-induced apoptosis were downregu- lated by osteoclasts. Osteoclasts were also found to increase expression of the osteoblast transcription factor RUNX2 in the osteoblastic CRPC cell line, while the high levels in the osteolytic cell line was not affected. RUNX2 was found to promote expression of steroidogenic enzymes and the androgen regulated prostate-specific antigen (PSA) in CRPC cells co- cultured with osteoclasts. To evaluate the importance of RUNX2 for CRPC growth in bone, RUNX2 was knocked-out (KO) in osteoblastic CRPC cells which then was im- planted in the tibia of immune-deficient mice. Magnetic resonance imaging (MRI) was validated and used to give accurate size and location of in- tratibial CRPC xenografts. It was shown that RUNX2-KO CRPC cells grew slower and formed smaller intratibial tumors compared to control cells. Both expression of steroidogenic enzymes and PSA-expression was inhibited by depletion of RUNX2. In conclusion, this thesis show that both osteoclasts and RUNX2 affect CRPC growth and steroidogenesis. Treatment for metastatic PC include further targeting of the AR axis, a strategy that may be counteracted by both RUNX2 and interaction with osteoclasts. In addition, the effect of tar- geted therapy towards cells with defect DNA repair, such as PARP- inhibitors, may be affected by the action of osteoclasts. �us, the results of the present thesis suggest targeting osteoclasts or RUNX2 in combinatory therapeutic approaches to be investigated. Keywords Castration-resistant prostate cancer, Androgen, Bone metastases, RUNX2, Osteoclasts Sammanfattning på svenska Prostatacancer är en av de cancerformer som orsakar flest dödsfall bland män i Sverige. Det är oftast inte tumören i prostata som orsakar lidande och död, utan cancer som spridit sig till andra organ i kroppen och bildat dottertumörer, metastaser. Om cancern hittas tidigt, innan den hunnit sprida sig kan man i många fall behandla och bota sjukdomen. Metasta- serad cancer går idag inte att bota, även om det finns många läkemedel som förlänger livet betydligt. Skelettet är det vanligaste organet för metastaser från prostatacancer. Om man upptäcker metastaser från prostatacancer behandlas patienten oftast med så kallad hormonbehandling, vilken stoppar produktionen av det man- liga könshormonet steroiden testoster on och hindrar därmed cancercel- lerna från att växa ytterligare. Denna behandling har god effekt hos de flesta, men tyvärr brukar sjukdomen ta fart igen efter några år. Den har då blivit kastrationsresistent, d.v.s. den växer trots att halterna av testosteron i blodet fortfarande är låga. I skelettet växer tumören i benmärgen och metastaser från prostatacancer leder oftast till att ben-nybildningen ökar, i motsats till flera andra cancer- former, som bröstcancer och lungcancer, vilka oftast leder till en ökad ned- brytning av benet. Cancercellerna samverkar med olika typer av celler som finns i benet och benmärgen, och det är känt sedan tidigare att nedbrytning av benvävnad frigör flera substanser som kan driva på cancercellernas till- växt. I tidigare studier har vi studerat hur osteoblaster, dvs de celler som bygger benvävnad, interagerar med prostatacancerceller och funnit att de påverkar cancercellerna så att de förbättrar sin förmåga att själva bilda testosteron och stimulera sin tillväxt. Osteoklaster är de celler som bryter ner benvävnad och som i ett friskt ben samverkar med osteoblaster för att ständigt förnya och behålla benvävna- den i gott skick. Om och på vilket sätt osteoklaster direkt kan påverka prostatacancercellernas egenskaper har tidigare inte varit känt. I denna avhandlings första delarbete odlas prostatacancerceller tillsammans med osteoklaster och effekterna av detta studeras. Osteoklaster påverkar cancercellerna till att dela sig fortare och dö mer sällan. Uttrycket av ca viii 3500 gener förändrades i cancercellerna efter samodling med osteoklaster. Generellt påverkades prostatacancerceller som ger upphov till ben-nedbry- tande metastaser mer än de prostatacancerceller som ger upphov till ben- bildande metastaser. Gener som medverkar till att DNA kan repareras uppreglerades av osteoklaster främst i de ben-nedbrytande prostatacancer- cellerna. En grupp gener som reglerar hur cellerna påverkas av en viss storts stress påverkades i båda cancercelltyperna. RUNX2 är ett protein som i normala fall styr hur osteoblaster bildas. Det är sedan tidigare känt att RUNX2 också finns i prostatacancerceller och gör dessa mer elakartade. Vi har i tidigare studier sett att osteoblaster ökar mängden RUNX2 i benbildande prostatacancerceller. I avhandlingens andra delarbete studerar vi hur osteoklaster påverkar RUNX2 i prostatacan- cerceller och hur RUNX2 reglerar förekomst av vissa enzymer som med- verkar i steroidsyntesen och bildningen av testosteron. På samma sätt som osteoblaster, ökar osteoklaster mängden RUNX2 i de benbildande cancer- cellerna, men inte i de ben-nedbrytande, som redan har höga nivåer av RUNX2. Genom att blockera RUNX2 hämmades flera centrala enzymer i steroidsyntesen särskilt då de benbildande cancercellerna odlades tillsam- mans med osteoklaster. Vi kunde även visa att RUNX2 på detta sätt påver- kade mängden av prostata-specifikt antigen, PSA, som styrs av testosteronets bindning till sin receptor, androgenreceptorn. I delarbete två visades alltså att RUNX2 kan driva på tumörcellernas egen förmåga att bilda testosteron och därigenom aktivera androgenreceptorn, vilket är en mycket viktig överlevnadsfunktion hos prostatacancerceller. I det tredje delarbetet fortsatte studierna om hur RUNX2 påverkar cancercel- lernas tillväxt och egenskaper när de växer i skelettet, genom att cancercel- ler med blockerat RUNX2 injicerades i benmärgen på möss. Det är omöjligt att utifrån se hur dessa tumörer växer under försökets gång, och därför utvärderades en metod där magnetkamera användes för att följa och mäta tumörerna. Resultaten visar att magnetkamerametoden lämpar sig mycket väl för denna typ av studier och vi såg att både tumörernas storlek och benbildning mätt med magnetkamera överensstämde väl med andra analysmetoder. Med hjälp av magnetkamerametoden och andra analyser visades att tumörceller utan RUNX2 växte sämre i benmärgen och hade lägre nivåer av steroidenzymer och mindre aktivering av sin androgenre- ceptor jämfört med tumörceller med normal mängd RUNX2. Sammantaget visar denna avhandling att både osteoklaster och RUNX2 på- verkar centrala funktioner hos prostatacancerceller. Det finns idag många läkemedel som används för att behandla metastaserad prostatacancer och som förlänger livet på dessa patienter. Flera av dem syftar till att blockera androgenreceptorns funktion ytterligare, något som alltså tycks motverkas av både osteoklaster och RUNX2. Andra läkemedel som utprovas, som PARP-hämmare, riktar särskilt in sig på cancerceller med otillräcklig för- måga att reparera sitt DNA, en förmåga som till viss del verkar påverkas av osteoklaster. Läkemedel som hämmar osteoklasters funktion finns redan och används inom prostatacancervården, främst för att skydda skelettet och motverka så kallade skelettrelaterade händelser (SRE), men de påverkar ty- värr inte livslängden hos patienterna. Denna avhandling visar på en möjlig positiv effekt av kombinationsbehandling av läkemedel som hämmar os- teoklaster och RUNX2 med flera typer av de läkemedel som är aktuella för behandling för patienter med metastaserad prostatacancer. Detta behöver studeras vidare i djurmodeller för att sedan kunna utvärderas i kliniska stu- dier på patienter. List of papers �is thesis is based on the following studies, referred to in the text by their Roman numerals. I . Junchi Huang, Eva Freyhult, Jan-Erik Damber, Karin Welén Osteoclasts directly influence castration-resistant prostate cancer cells Manuscript. I I . Junchi Huang, Malin Hagberg Thulin, Jan-Erik Damber, Karin Welen The roles of RUNX2 and osteoclasts in regulating expression of steroidogenic enzymes in castration-resistant prostate cancer cells Mol Cell Endocrinol. 2021 Sep 15;535:111380. III . Junchi Huang, Mikael Montelius, Claes Ohlsson, Matti Poutanen, Jan-Erik Damber, Karin Welén RUNX2 as a mediator of castration resistant prostate cancer growth in the bone microenvironment Manuscript. 12 CONTENT Content Abbreviations _______________________________14 1. Introduction ______________________________16 1.1 The Prostate 16 1.1.1. Anatomy and Function 16 1.1.2. Morphology 17 1.1.3. Androgenic actions in prostate 18 1.2 Prostate Cancer 20 1.2.1 General background 20 1.2.2 Diagnosis and Prognosis 21 1.2.3 Prostate oncogenesis 23 1.2.4 Castration-resistant prostate cancer 24 1.2.5 Endoplasmic reticulum (ER) stress, DNA damage response (DDR) and PC 26 1.2.6 Treatment of prostate cancer 28 1.3 Bone Microenvironment and Prostate Cancer Bone Metastasis 29 1.3.1 Bone Microenvironment 29 1.3.2 Prostate cancer bone metastases 32 2. Aims ____________________________________37 3. Methods and methodological considerations ___38 3.1 In Vitro Experiments 38 3.1.1 Cell lines and cell culture 38 3.1.2 Co-culture model 39 3.1.3 RNA interference 39 3.1.4 Knock out experiment 39 3.1.5 Proliferation and Apoptosis Assay 40 3.1.6 Scratch-wound assay 40 3.1.7 RNA isolation and reverse transcription 40 3.1.8 Quantitative RT-PCR 41 3.1.9 Protein Assay 41 3.1.10 RNA sample handling and sequencing 41 3.1.11 Gene ontology and pathway analysis 42 CONTENT 13 3.2 In Vivo Experiments 43 3.2.1 Animal models 43 3.2.2 Animal handling and intratibial implantation 43 3.2.3 MRI experiments 44 3.2.4 Immunohistochemistry 44 3.2.5 Steroid measurements 45 3.2.6 Statistics 45 4. Results and comments _____________________46 Paper I. 46 “A broad investigation that demonstrates that OCs directly influences the growth and function of CRPC cells.” Paper 2. 49 “RUNX2 as an important mediator of steroidogenesis in CRPC cells with potential to influence castration resistant AR signalling. OCs promote RUNX2 regulated induction of key steroidogenic enzymes, influencing activation of AR in CRPC cells.” Paper 3. 52 “RUNX2 may be a crucial component in PC bone metastasis as well as a possible treatment target for mCRPC” 5. Discussion and future perspective ___________55 Bone metastases, the life wrecker for PC patients 55 The focus on osteoclasts and why 56 Bone targetting agents, beyond restoring bone homeostasis 59 RUNX2, an osteoblastic factor highly involved in PC progression and potential treatment target 61 The in vitro and in vivo models, advantages and limitations63 MRI, “eye opener” for in vivo studies 64 6. Conclusion _______________________________65 Acknowledgement ___________________________66 References _________________________________68 14 ABBREVIATIONS Abbreviations ADT Androgen deprivation therapy AKR1C3 Aldo-keto reductase family 1 member C3 Akt Protein kinase B ALP Alkaline phosphatase AR Androgen receptor ATF Activating transcription factor BPH Benign prostatic hyperplasia CDH Cadherin CRPC Castration-resistant prostate cancer CTC Circulating tumor cells CYP Cytochrome P450 family DDR DNA damage response DHEA Dehydroepiandrosterone DHT Dihydrotestosterone DNA Deoxyribonucleic acid EGF Epidermal growth factor EMT Epithelial–mesenchymal transition ER Endoplasmic Reticulum ERalpha Estrogen receptor alpha ERbeta Estrogen receptor beta ERs Estrogen receptors ET1 Endothelin 1 ETS E26 transformation-specific ETV 1 ETS Variant Transcription Factor 1 GAPDH Glyceraldehyde-3-Phosphate Dehydrogenase GnRH Gonadotropin-releasing hormone GR Glucocorticoid receptor HNPC Hormone-naïve prostate cancer HSD Hydroxysteroid dehydrogenase HSD17B Hydroxysteroid 17-Beta Dehydrogenase ABBREVIATIONS 15 IGF Insulin like growth factor IHC Immunohistochemistry IL Interleukin IRE1 Inositol-requiring enzyme 1 mCRPC Metastatic castration-resistant prostate cancer M-CSF(CSF-1) Macrophage colony-stimulating factor MET Mesenchymal–epithelial transition MMP Matrix Metalloproteinases MRI Magnetic resonance imaging mRNA Messenger RNA mTOR Mechanistic Target Of Rapamycin Kinase OB Osteoblast OC Osteoclast OPG Osteoprotegerin OPN Osteopontin PARP Poly (ADP-ribose) polymerase PC Prostate cancer PERK Protein kinase R like endoplasmic reticulum kinase PI3K Phosphoinositide 3-kinase PIN Prostatic intraepithelial neoplasia PSA Prostate-Specific Antigen PTHrP Parathyroid hormone-related protein qPCR Quantitative polymerase chain reaction RANK Receptor activator of NF-κB RANKL Receptor activator of NF-κB ligand RNA Ribonucleic acid ROI Region of interest RUNX2 Runt-Related Transcription Factor 2 siRNA Small interfering RNA SRD5A steroid 5 alpha-reductase TMPRSS2 Transmembrane Serine Protease 2 TRAP Tartrate-resistant acid phosphatase UPR Unfolded protein response VEGF Vascular endothelial growth factor XBP1 X-Box Binding Protein 1 16 1 . INTRODUCTION 1. Introduction 1.1 The Prostate 1.1.1. Anatomy and Function �e prostate is a walnut shaped gland in the male reproductive system as well as a controller for urination and ejaculation switching. �e prostate gland produces the fluid that forms part of semen. Anatomically, the prostate is located below the bladder with the urethra passing through the center of it. In macroscopy level, it is consisted of lobes, and in microanatomy level, it is defined by zones. It is enveloped by an elastic, fibromuscular capsule and contains glandular tissue as well as connective tissue [1]. Approximately 30 % of the seminal fluid origi- nates from the prostate; the prostate fluid contains high levels of zinc magne- sium, citric acid and also several proteins [2]. One important protein is Prostate- Specific Antigen (PSA), which have proteolytic properties and is of importance for the liquefaction of the seminal fluid [3]. �e exact function of prostate secre- tion is unknown but is generally believed to be of importance for sperm function and for the protection of ascending infections [4]. Figure 1. Schematic illustration of the location and anatomy of the prostate 1 . INTRODUCTION 17 �e internal structure of the prostate has been described as consisting of three glandular zones: Peripheral zone (PZ), Central zone (CZ), Transition zone (TZ) and an anterior fibromuscular stroma [5]. �e PZ constitutes the majority, about 65 %, of the normal glandular prostate, the CZ approximately 30 %, while the TZ only constitutes about 5 %. �e PZ ducts exit posterolaterally from the ure- thral wall on both sides of the prostate from the seminal colliculus to the prostate apex, with branches that curve anteriorly and posteriorly. �e CZ arises on the seminal colliculus to surround the ejaculatory duct orifices [6]. �e TZ is from the seminal colliculus to the bladder neck. �e anterior fibromuscular stroma is viewed as a wedge shaped stromal barrier, extending from the apex to the base, and hugging the prostatic urethra and glandular zones [7, 8]. 1.1.2. Morphology �e prostatic glandular structure is lined by a pseudo-stratified epithelium that mainly contains three cell types, basal cells, secretory luminal cells, and neuro- endocrine (NE) cells. Studies have shown that there are intermediate cells, which express both luminal and basal markers in the developing and adult prostate [9- 11]. Figure 2. Schematic illustration of the prostate epithelium cellular structure Secretory luminal cells are characterized by expression of cytokeratin (CK) 8, CK18, androgen receptor (AR) and secretory proteins like prostate specific anti- gen (PSA), also AR is essential for luminal cell survival. Basal cells are identi- fied by the expression of CK5, CK14 and p63 [12, 13]. NE cells, spread in the 18 1 . INTRODUCTION basal cell layer, are AR negative and thus androgen independent. �e secretory products from the interspersed NE cells are important for the growth and sur- vival of surrounding epithelial cells [14]. 1.1.3. Androgenic actions in prostate 1.1.3.1 Androgens and Androgen Receptor Androgens are the male sex hormones regulating the differentiation and matura- tion of the prostate and other male reproductive organs [15], and exert their ac- tion by binding to the ligand-regulated transcription factor, AR. �e structure of AR is similar to other members of the steroid receptor superfamily, consisting of three major functional domains, the N-terminal domain (NTD), a conserved DNA binding domain (DBD), a connecting hinge region, and the C-terminal lig- and-binding domain (LBD). �e DBD binds the AR with AR-regulated genes’ promoter and enhancer regions through direct DNA binding to activate function- ing of the NTD and LBD to stimulate transcription of these genes [16-18]. DHT could bind and activate the AR, which would go through a conformational change and release heat shock proteins and translocate to the nucleus, where it transcriptionally regulates the downstream target genes’ expression [19, 20]. Figure 3. Schematic illustration of androgens in androgen receptor (AR) signalling and AR structure 1 . INTRODUCTION 19 Testosterone, which is synthesized by Leydig cells in the testes, is the primary androgen in the circulation [21]. Adrenal glands also produce a small amount of androgens, such as dehydroepiandrosterone (DHEA) and androstenedione (4-di- one) [22]. Testosterone goes through passive diffusion into prostate cells, where it is transformed by steroid 5-α reductases to the more potent androgen dihydro- testosterone (DHT) [23]. However, there are more pathways for DHT synthesis besides 5α-reduction of T. Androstenedione can also be turned into DHT with the help of 17β-hydroxysteroid dehydrogenases (HSD17B) [24]. In addition, the backdoor pathway, enables for progesterone and androstanediol to be converted into DHT [25]. Figure 4. Androgen biosynthesis pathways and metabolism in prostate cancer 1.1.3.2 Androgens and prostate Androgens, especially DHT, is necessary for prenatal prostate development and functions [15]. 20 1 . INTRODUCTION �e functional AR is essential in the initiation of prostate development. AR me- diates the formation and growth of prostatic buds in the epithelia-stroma interac- tion from the urogenital mesenchyme [26, 27]. AR function is also involved in final morphogenesis of the prostate and the prostate secretory protein expression in late fetal or early neonatal development as AR expression is presented in the prostate epithelium [28, 29]. Besides the development of the prostate, androgens also function in promoting the survival of the secretory epithelium [30, 31]. In the normal prostate, the rate of cell death is 1–2% per day harmonized by the same rate of proliferation. In hu- mans, the proliferation occurs predominantly in the TZ of the prostate [30]. Apart from androgens acting directly on the epithelium via epithelial AR to bring out differentiated function, androgens could also operate indirectly via stromal AR and interactions with a range of mediators, such as epithelial growth factor (EGF) and fibroblast growth factor (FGF), nerve growth factor (NGF) and insulin-like growth factor 1 (IGF-1) [32, 33]. �ese factors can diffuse from the stromal compartment to epithelial cells, hence regulate epithelial growth and dif- ferentiation [34]. 1.2 Prostate Cancer 1.2.1 General background During the past twenty years, prostate cancer (PC) has become the most com- mon cancer among European men and the third most frequent for men in the world [35, 36]. Sweden, among other Nordic countries, is on the top of the list of incidence and mortality rate of PC [37]. It is a disease mostly occurring in older men with majority of the cases diagnosed in men aged 65 years old or older with an average age of about 66 at the time of cancer diagnosis [38]. PC is described as abnormal prostate gland growth from abnormally dividing cells in the gland [39, 40]. Death from PC is mainly resulted from metastasis when cancer cells spread to other parts of the body including liver, lung, urinary bladder lymph nodes and most commonly to the skeleton [41]. 1 . INTRODUCTION 21 Multiple signaling pathways have been discovered to be critical for prostate can- cer initiation and progression, for example the phosphoinositide 3-kinase (PI3K) pathway, ErbB pathway, angiogenesis and apoptosis pathways, of which the an- drogen signaling pathway has proved to be one of the most prominent [34, 42, 43]. 1.2.2 Diagnosis and Prognosis 1.2.2.1 Early detection �e prostate-specific antigen (PSA) blood test has been the most used for early detection, however its low specificity has been criticized [44]. Aside from the PSA test, other markers have been developed including the phi (Prostate Health Index) test that combines the results of total PSA, free PSA, and proPSA [45], and the Four–Kallikrein Panel which contains the results of total PSA, free PSA, intact PSA, and human kallikrein 2 [46]. Further, mRNA biomarkers like TMPRSS2:ERG fusion gene in the urine have been found to have potential in prostate cancer diagnosis [44, 47]. Today, blood tests, such as PSA, are com- bined with multiparametric magnetic resonance imaging (MRI) in order to yield specificity before the diagnostic procedures [48]. 1.2.2.2 Diagnosis MRI is used to improve the sensitivity and specificity of the diagnostic perfor- mance [49, 50].Since the PSA-tests are not optimal in specificity, the diagnosis of prostate cancer is based on microscopic evaluation of prostate tissue obtained via needle biopsy often targeting the suspicious areas observed on MRI. Conven- tionally, the prostate biopsy is performed using ultrasound to obtain tissue sam- ples in a grid-like pattern but also fusion technology is used combining ultrasound with MRI imaging, MRI guided biopsy has provided improved per- formance in cancer detection and diagnosis [51]. 22 1 . INTRODUCTION Figure 5. Schematic illustration of the ultrasound grid biopsy and MRI guided biopsy 1.2.2.3 Staging and grading �e TNM staging system has been used to describe the tumor stage. It describes the size and location of the tumor (T), spreading of the tumor to lymph nodes (N), and metastasis to other areas of the body (M). [52]. Gleason score expressed in Gleason grade groups (GGGs) is the main system for grading prostate cancer. �e Gleason score is based on growth pattern architec- ture comparing tumor tissue with normal tissue. A scale of 1 to 5 in Gleason score where 3-5 are considered pathological. In the grade groups, the most com- mon score is combined with the next most common determining the Gleason score. A Gleason score of 6 (3+3) is the lowest score referring to low-grade can- cer (ISUP GG1), a Gleason score of 7 points to an intermediate-grade cancer (Gleason 3+4 is ISUP GG 2, Gleason 4+3 is ISUP GG 3), and a score of 8–10 in- dicates to a high-grade cancer (Gleason 8 is ISUP GG 4, and Gleason 9-10 is ISUP GG 5) [53, 54]. 1 . INTRODUCTION 23 1.2.3 Prostate oncogenesis 1.2.3.1 Prostate cancer initiation PC is considered heterogeneous and multifocal since the primary prostate tumors often contain multiple genetically distinct and independent histologic foci of cancer [55]. On the other hand, metastatic PC is believed to develop from the se- lective advantage of individual clones during cancer progression [56]. High grade prostatic intraepithelial neoplasia (HGPIN) is commonly believed to be a precursor for prostate cancer. HGPIN is characterized by the appearance of luminal epithelial hyperplasia, reduction in basal cells, enlargement of nuclei and nucleoli, cytoplasmic hyperchromasia, and nuclear atypia as well as an exalta- tion of proliferation markers [57-59]. TZ is the primary zone for benign prostatic hypertrophy (BPH) but seldom the initiation site of PC. Studies on prostate gland zones showed that only 24% of cancers are initiated in TZ, the majority of the tumors (68%) emerged in the PZ related with high grade HGPIN or dysplasia [60, 61]. 1.2.3.2 Androgen signaling and prostate cancer Studies have failed to establish a link between increased incidences of PC in men administered testosterone, as well as reduced risks of PC in men with low serum androgen levels, suggesting that there is no correlation between PC risks and ex- cessive androgen alone [14, 62]. Despite that androgens do not seem to be important for PC initiation, it is well recognized that early PC relies on androgens, and its signaling by AR. In the de- velopment process of prostate cancer, AR plays roles in proliferation, apoptosis, migration and invasion [63]. One way for AR to influence the development of PC is via the fusion gene TMPRSS2-ERG. �e AR regulated TMPRSS2 gene and the E-twenty-six (ETS) transcription factor genes fusion is one of the most common genetic alterations in prostate tumors. �rough gene fusions between TMPRSS2 gene promoter and the coding region of the ETS family members like ETS-related gene (ERG) and ETS variant 1 (ETV1), granting androgen response to ETS transcription factors, resulting in cell-cycle progression [64-66]. 24 1 . INTRODUCTION Besides the direct stimulation of PC cells through AR, androgens also facilitate the proliferation of prostate epithelial cells with a wide range of growth factors secreted from AR regulated stromal cells [67, 68]. Also in the PC cell migration process, the role of androgens have been shown to be important, for example by upregulation and activation of matrix metalloproteinase (MMP-2) [69]. In addi- tion, PC angiogenesis, important for both local growth and metastatic spread, is driven by androgen-regulated vascular endothelial growth factor (VEGF) synthe- sis [70].Androgens are also able to control the IGF signaling and IGF1 could pro- mote the AR activities in the low androgen environment, an interaction especially important for tumor survival in the castrated situation [68, 71]. AR also interact with the transforming growth factor-β (TGFβ) pathways in PC. TGFβ family is tightly related to cell proliferation and apoptosis. Upregulated TGFβ1 has been discovered in patients with PSA increment and PC bone metas- tases [72]. AR expression is able to downregulate the Smad3 in TGFβ1-Smad pathway and reduce the Smad3 and Smad4 growth inhibition [73]. 1.2.3.3 Prostate cancer progression �e majority of PC conforms to AR expressing acinar adenocarcinomas, which develop in the glandular cells that line the prostate gland. Other forms of PCs like ductal adenocarcinoma, urothelial cancer also exist but are less common [74]. PC that grows exclusively within the prostate gland is defined as localized PC. When the cancer has emerged through the prostate capsule to neighboring tis- sues, it has developed into locally advanced PC. Generally, advanced PC or met- astatic PC implies that the cancer has spread from the prostate to other parts of the body, of which the bones and lymph nodes are the main metastatic sites [75, 76]. 1.2.4 Castration-resistant prostate cancer Patients diagnosed with metastatic hormone-sensitive PC usually get temporary control of the cancer with androgen blockade by chemical or surgical castration, the castrate level of testosterone leads to decreasing cancer cell proliferation and further inducing the apoptosis [77]. However, patients who show treatment re- sponse will eventually develop disease progression that is no longer sensitive or 1 . INTRODUCTION 25 resistant to the hormonal therapy. �is stage is defined as ‘castration-resistant prostate cancer’ (CRPC). CRPC is currently incurable despite several life-pro- longing therapeutic approaches exist [78, 79]. CRPC was for long believed to be androgen independent but it is now known that AR signaling is still important for the progression of the disease [80]. Sev- eral possible mechanisms have been raised to contribute to CRPC progression, for example, increased sensitivity of the AR to its agonists, ligand-independent AR activation, AR mutations that grant the receptor responsive to non-androgen ligands, and other mechanisms without AR involvement [80, 81]. Figure 6. Schematic illustration of mechanisms contribute to castration-re- sistant prostate cancer (CRPC) development Abnormal AR activities such as increased AR expression have been demon- strated in metastatic lesions and castration resistance prostate cancer (CRPC) [82, 83]. AR mutations have frequently been detected in PC cells, and some vari- ants are closely related to the progression of PC. �e deletion of the C-terminal makes it possible for the variants to be constitutively active, and mutated AR could stimulate and cause receptor nuclear localization without androgen [84, 85]. Higher levels of variants have been found in bone metastases comparing to hormone-naïve PC and the AR variant 7 (ARV7) is one of the variants which are upregulated during PC progression [86]. �e ARV7 receptor is regulated by anti- 26 1 . INTRODUCTION apoptotic protein kinase Akt, and it has been discovered to be associated with re- sistance to both enzalutamide and abiraterone in patients with CRPC in recent clinical studies [87, 88]. Another mechanism involved in the development of CRPC is intratumoral steroidogenesis. Evidence has shown that CRPC has a higher level of intra- tumoral testosterone than hormone-naïve PC, similar to the level in prostates of healthy men [89]. Alteration of the steroidogenic enzymes are believed to con- tribute to androgen synthesis of the CRPC metastases, since increased expres- sion of important enzymes like CYP17A1, CYP19A1 and AKR1C3 have been detected in CRPC metastases compared with primary tumors [90, 91]. In CRPC, AR could also be activated by tyrosine kinase receptor-activating lig- ands, for example, IGF-1 and EGF through PI3K/AKT/mTOR pathway [92]. In some cases, CRPC can bypass AR signaling completely by regulating cancer cells growth and survival via Stat3 signaling, anti-apoptotic protein Bcl-2 or up- regulated glucocorticoid receptor (GR) [93-95]. 1.2.5 Endoplasmic reticulum (ER) stress, DNA damage response (DDR) and PC ER in the eukaryotic cell is affiliated with protein synthesis and calcium signal- ling. One of the most important responsibilities of ER is the proper folding and transportation of proteins before secretion [96]. ER stress indicates the accumu- lation of misfolded proteins in the lumen of the ER due to disrupted ER function caused by environmental factors, for example altered glycosylation, oxidative stress, and nutrient deprivation. ER stress leads to unfolded protein response (UPR) which activates IRE1 (inositol‐requiring enzyme 1), the PERK (double stranded RNA dependent protein kinase PKR-like ER kinase) and ATF6 (acti- vating transcription factor 6) to restore the ER homoeostasis by pausing protein synthesis and increasing protein degradation or induce apoptosis when the stress is beyond adaptation. [97-99]. In the pro-survival pathway, IRE1 mediates splicing of α-x-box binding protein 1 (XBP1) to induce ER associated degradation (ERAD). PERK phosphorylates eIF2α to activate transcription factor 4 (ATF4) to re-establish ER homeostasis. ATF6α upregulates the ERAD [100]. For pro-apoptotic side, IRE1 partners with TNF receptor-associated factor2 (TRAF2) and apoptosis signal-regulating kinase 1 (ASK1) to activate downstream apoptosis pathways. PERK and ATF6 activate 1 . INTRODUCTION 27 apoptotic reaction through mediating ATF4 and DNA damage-inducible tran- script 3 (DDIT3) pathways and C/EBP Homologous Protein (CHOP) [101, 102]. Cancer cells are constantly exposed to high demand of protein production and external environment stress like hypoxia. Consequently, ER stress and UPR are involved in cancer progression. Cancer cells can adapt UPR to profit from the pro-survival effect as a survival strategy for progression [103]. Androgen signal- ling has proved to be able to activate the IRE1α arm of UPR and upregulate the IRE1α and XPB1 expression in PC cells, leading to the IRE1α/XPB1 induction of the oncogenic c-MYC expression, which further contributes to PC progression [98]. ATF3, which suppresses androgen signalling, is significantly lower of ex- pression in CRPC cell lines than androgen dependent cell line [104]. DNA damage response (DDR) pathways are essential for maintaining the ge- nomic stability. �e DDR, which is induced by internal or external events such as mutations, DNA strand breaks, or UV stimulus, will lead to DNA replication regulation, and cell cycle check and arrest, followed by DNA damage repair. Common DNA damage types like DNA double-strand breaks (DSBs), single- strand breaks (SSBs) and DNA adducts by base alkylation are detected by spe- cific sensors and trigger different kinds of responses [105, 106]. Altered or defect DDR pathways have shown to be associated to cancer develop- ment and progression. Cancer cells are known for genomic instability, tumors are able to harvest alterations in DDR pathways which induce genomic instabil- ity and oncogenesis [107]. Cancer cells often have lost one or more DDR path- way, which makes them to rely on certain pathways only [108]. �is gives a chance for therapy development, of which poly (ADP ribose) polymerase (PARP), an essential factor in the SSB repair, appears to become an important target in advanced PC [106, 109]. As to PC, inactivation of DDR related genes, alteration of DDR related kinases and further DDR defects have been found in clinical CRPC samples. Eight per cent of samples had germline DDR pathway aberrations and 23% contained DDR pathway alteration in a study comprising 150 mCRPC tumors [110, 111]. PC tumors show inactivation or mutation of DDR protein coding genes, of which the BRAC1, BRAC2 are common targets. BRAC1/2 are critical in homolo- gous recombination (HR), HR repair is largely included in DSB repairs and ho- mologous recombination deficiency (HRD) is a common factor in oncogenesis [112, 113]. Alteration in ataxia telangiectasia mutated (ATM) kinase, PARP1 and 28 1 . INTRODUCTION cyclin dependent kinase 12 (CDK12) have also been detected [114]. Crosslinks between AR functions and DDR pathways have also been discovered. AR acti- vation relates to DNA DSBs since PARP-1 is correlated positively with AR ac- tivities in PC and abnormal AR activation could induce downstream DNA damages and other genomic malfunction related to PC [115, 116]. In addition, AR is proved to regulate the expression of genes involved in DDR pathways like Fanconi anemia, and PARP mentioned above [117]. PARP inhibitor (PARPi) has shown clinical benefits in mCRPC patients, together ADT, researchers have dis- played a delay effect in developing castration resistance [118, 119]. 1.2.6 Treatment of prostate cancer PC treatments depend on the clinical presentation of the disease. For low-risk lo- calized PC, active surveillance (AS) is a common option. Radical prostatectomy as well as radiation therapy (RT) are used in intermediate to localized high risk PC [120]. Figure 7. Systemic treatment options for patients in different stages of prostate cancer 1 . INTRODUCTION 29 Androgen deprivation therapy (ADT) is typically applied in advanced PC and metastatic hormone naïve prostate cancer (HNPC). It slows the progression of PC by achieving castrate levels of testosterone. ADT is performed using gonado- tropin releasing hormone (GnRH) agonists or antagonists inhibiting the produc- tion of testosterone by interfering with the GnRH-receptor in the pituitary gland. [121, 122]. Additional inhibition of the androgen pathway can be achieved by blocking the steroid synthesis pathway, specifically CYP17A1, by abiraterone ac- etate, or inhibiting the AR by the second-generation anti-androgens such as en- zalutamide, apalutamide and darolutamide [123-126]. For metastatic HNPC chemotherapy, docetaxel is used in combination with ADT as also abiraterone and new antiandrogens. In different combinations and sequences, these drugs are also used in CRPC as well as another taxane, cabazitaxel [127, 128]. Furthermore, for treatment of bone metastases, bone-targeting therapy can be ap- plied. One example is Radium-223 dichloride (radium-223), which targets the high bone turnover areas incorporating the alpha-emitter in the bone metastases inducing local cytotoxicity in the bone stroma [129]. Denosumab, which is a hu- man monoclonal antibody against Receptor activator of NF-κB (RANK) ligand (RANKL) is another option. It blocks osteoclast (OC) differentiation thus inter- feres with the bone cells interacting with the tumor cells in the bone. Bisphos- phonates are also used, and reduce excessive bone turnover while preserving bone structure and mineralization [124, 130-132]. Denosumab and Bisphospho- nates do not prolong survival but have effect on skeletal related events (SREs) [133]. 1.3 Bone Microenvironment and Prostate Cancer Bone Metastasis 1.3.1 Bone Microenvironment 1.3.1.1 Bone cells �e skeletal lineage consists mainly of osteoblasts (OBs), osteocytes and chon- drocytes, which are all heavily involved in the bone development and formation. 30 1 . INTRODUCTION Osteoclasts (OCs) and OBs are located on the surface of bone, while chondro- cytes which produce cartilaginous matrix consisting of collagen and proteogly- cans, are the only cell type found in healthy cartilage[134]. Osteoclasts Multinucleated OCs are primarily responsible for the bone degradation and re- sorption. �e OC precursor arise from the hematopoietic lineage to the bone ma- trix surface, and the OC is a result of the fusion of bone marrow derived monocytes/macrophages into large multinucleated cells [135, 136]. In order for OC to mature and function properly, macrophage colony-stimulating factor (M- CSF) and RANKL, which are produced by OBs, osteocytes and immune cells in the bone environment [137], are necessary. M-CSF promotes proliferation of OC precursors and RANKL boosts differentiation of OC precursors to adhere to the bone and mature OCs and maintains normal functions of OCs [138, 139]. Addi- tion to RANKL and M-CSF, other factors like interleukin-6 (IL-6), interleukin-11 (IL-11), parathyroid hormone–related peptide (PTHrP) could activate OC for- mation [140, 141]. RANKL activates the OCs to establish the junction between basal membrane and bone surface for the bone resorption. OC then secretes osteolytic enzymes like Tartrate-resistant acid phosphatase (TRAP) and Cathepsin K (CatK) into the junction conducting resorbing and degrading actions. �e products from the bone resorption, for example collagens and other soluble factors, are then han- dled by OCs before releasing into the circulation [142, 143]. Osteoblasts OBs are the main characters in bone formation. OBs secrete essential matrix pro- teins like collagens, alkaline phosphatase (ALP), osteocalcin and osteopontin for building the bone [144]. �e expression of SRY-Box Transcription Factor 9 (SOX9) initiates the osteoprogenitor cells, which develop into Runt-related tran- scription factor 2 (RUNX2) expressing preosteoblasts. Finally, expression of RUNX2 and WNT–β-catenin signaling induced osterix (SP7) leads to mature OBs [135, 145, 146]. A part of the mature OBs will go into apoptosis, while some population of OBs secrete extracellular matrix components, interact with each other, and attach into the matrix of the bone, forming osteocytes. Osteocytes, which account for the major population of the cells in fully formed bones, are responsible for regulat- ing the bone maintenance by reacting to mechanical pressure and interacting be- tween OBs and OCs [146, 147]. 1 . INTRODUCTION 31 Figure 8. Schematic illustration of the differentiation process of osteoblasts and osteoclasts 1.3.1.2 Bone remodeling Bone is not a static tissue; in fact, about 10% of the bone mass is replaced every year, the remodeling mechanism helps bone development and repair [148]. In bone environment, the regulated destruction of bone matrix by OCs is closely connected to the formation of new bone by OBs with growth factors , such as TGFβ and IGF1, released from the degraded bone matrix as the mediator in be- tween [149]. In turn, the OBs produce RANKL, which stimulate OC precursors and lead to maturation of functional OCs. Furthermore, OBs produce the decoy receptor osteoprotegerin (OPG), which serves as a balance regulator by binding to RANKL, inhibiting the RANK signaling and stop the OC maturation [150]. �e dynamic balance of bone remodeling forms a suitable ground for tumor cells to take advantage to seed and grow, especially in trabecular bone, which have high turnover rate and are rich in vasculature [151]. 32 1 . INTRODUCTION 1.3.1.3 Hormonal regulation of bone remodeling Sex hormones such as androgens and estrogens are critical for bone growth and health [152]. Decreased level of estrogens and androgens is associated with ac- celerating bone turnover events [153]. Estrogen provides an anti-bone-resorption effect, it directly affect the bone re- modelling by regulating the OBs and OCs [154]. Estrogen stimulates the OBs to upregulate OPG, which block the RANKL signalling and reduce the OC matura- tion, thereby inhibiting bone degradation [155]. Androgens can be converted into estradiol through aromatization by CYP19A1 and may therefore be important for bone development and health [156]. How- ever, androgens can directly affect bone cells. AR is abundantly expressed in OBs, but is also present in OCs and osteocytes [157]. Androgens could stimulate the proliferation of OB precursors through AR signalling [158] and promote OB differentiation [159]. Studies have also shown the anti-apoptotic effects of DHT in OB by mediating the extracellular signal-regulated kinases (ERK) and Src ki- nase activities [160]. Furthermore, androgens could supress the RANKL induced OC formation through reducing membrane associated RANKL which is neces- sary for OC precursor maturation [161]. 1.3.2 Prostate cancer bone metastases 1.3.2.1 General background For progressive PC, metastases generally accompany or follow the inevitable transition into CRPC, and in most cases are established in skeleton [162]. Life quality of patients with bone metastases is largely decreased with the complica- tion of bone pain, fracture, and symptomatic hypercalcemia [163]. Although the detection, diagnosis and treatment for PC has advanced a lot, bone metastasis re- mains the major problem [164]. �e increment of bone density (osteoblastic phenotype) is the feature of most of PC bone metastases. However, prostate carcinomas are also able to form osteo- lytic (bone dissolving) metastases in dispersion through the osteoblastic metasta- ses [165, 166]. 1 . INTRODUCTION 33 1.3.2.2 The metastatic process to the bone Circulating tumor cells (CTC) in the blood stream and lymphatics, are tumor cells that go through an epithelial-mesenchymal transition (EMT) –like process and dissemble from the primary tumor [167]. However, most of them do not es- tablish metastases. To be metastasis-initiating cells, the cells need to equip them- selves with tumor initiating properties and undertake mesenchymal-epithelial transition (MET) to settle at the metastasis site [168, 169]. �e vessels to the bone marrow pave the way for metastasizing PC cells to arrive at the bone. In the bone, a vicious cycle of interactions between bone cells and cancer cells facilitate the growth and survival of cancer cells in the bone microenvironment [170]. PC cells employ the OCs degradation of mineralized bone matrix; the deg- radation results in releasing tropic factors for tumor growth [171]. Cancer cells benefitting from these growth factors secrete stimuli to activate OBs, which in turn (via RANKL) stimulate OCs to further bone degradation to close the circle [172]. PC cells interact with bone cells in several ways by secreted factors. PC cells produce endothelin-1 (ET-1), which is an osteoblast mitogen [173]. PSA, which is also produced by PC cells, is a PTHrP activator [174]. PC derived PTHrP acti- vates OCs for osteolysis, which facilitate the establishment and progression of PC bone lesions [151, 175]. Many types of tumor cells produce TGF-β, including PC cells. It can increase the production of PTHrP by tumor cells and boost the EMT of PC to enhance their metastatic potential. Further, TGF-β can increase production of IL-6, IL-11 and VEGF by tumor cells, which enhance bone metas- tasis and OC formation and survival [176, 177]. 34 1 . INTRODUCTION Figure 9. Schematic illustration of the bone homeostasis in normal condition and cancer bone metastasis vicious cycle 1.3.2.3 Hormonal regulation of PC bone metastases Intratumoral steroidogenesis is one of the major mechanisms for CRPC develop- ment and progression. mCRPC could maintain intratumoral androgens through the regulation of enzymes involved in intracrine steroidogenesis and androgen catabolism [178]. �e bone metastases are the primary site for development of CRPC. �e reason for that is not defined, but emerging evidence suggest that bone cells can affect the steroidogenic profile of CRPC cells. OBs empower CRPC cells to employ cholesterol as the source of steroidogenesis through upregulating Cytochrome P450 Family 11 Subfamily A Member 1 (CYP11A1), which turns cholesterol into pregnenolone [179]. Furthermore, increased levels of AKR1C3 (mediator of con- 1 . INTRODUCTION 35 version of adrenal androgen into testosterone) and HSD3B1 (converts andros- tenediol into testosterone) have been shown in in vivo studies of CRPC bone tu- mors. High expression levels of AKR1C3 was also displayed in a sub-group of patients with PC bone metastases [180-182]. 1.3.2.4 RUNX2 Runt-related transcription factor 2 (RUNX2) is a critical transcription factor in skeletal development. RUNX2 plays an important role in regulating the expres- sion of OB genes like osteopontin (OPN), which is important in bone turnover process [183]. During early skeletal development, RUNX2 activity is essential for OB differentiation [184]. RUNX2 is also expressed in normal prostate epithelium and PC cells [185, 186]. An overexpressed level of RUNX2 could frequently be found in cancers that fa- vours in bone metastasis and RUNX2 level have been positively correlated with increased Gleason score and metastasis of prostate tumors [187, 188]. �e function of RUNX2 in cancer cells has been associated with increased meta- static potential [185]. In cancer cells, RUNX2 activates expression of bone ma- trix and adhesion proteins that strongly link to PC metastasis process, as well as matrix metalloproteinases (MMPs) and angiogenic factors, which contribute to the metastatic potential [189]. Besides supporting tumor growth, RUNX2 is able to promote metastasis-related factors such as VEGF MMP-9, MMP-13, and se- creted bone-resorbing factors such as IL-8 and PTHrP in the PC osteolytic me- tastases model [190, 191]. RUNX2 is also an important mediator of TGFβ/SMAD and steroidogenesis pathways. RUNX2 binds Smad proteins that mediate downstream effects of TGFβ, which relates to the TGFβRII mutation caused resistance to TGFβ sup- pressing effect in prostate tumors [191, 192]. RUNX2 has been shown to functionally cooperate with androgen signalling, which could be indicated by RUNX2 co-expression enhance DHT-stimulated and AR-mediated transcriptional activation [193], and RUNX2 and AR also dis- played mutually suppressing effects possibly mediated by the AR DBD [194]. SNAI2, which is an important factor in PC metastases, is regulated by RUNX2 and androgens together [195]. In addition, in phosphatase and tensin homolog 36 1 . INTRODUCTION (PTEN)-deficient PC cells, RUNX2 interferes with steroidogenic enzymes activ- ities, for instance, RUNX2 signalling could induce CYP11A1 and CYP17A1 ex- pression. [191]. 2 . AIMS 37 2. Aims Overall aim The overall aim of this thesis is to increase the knowledge about the interaction between PC cells and the bone microenvironment. Specific aims 1. To identify any direct effect of OCs on PC cell phenotype 2. To analyse the role of OCs in regulation of intratumoral steroidogenesis via RUNX2 3. To determine the importance of RUNX2 in PC progression in the bone 38 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS 3. Methods and methodological considerations 3.1 In Vitro Experiments �e in vitro experiments in the present thesis have been designed to explore the crosstalk of OCs and CRPC cells. Furthermore, knocking out RUNX2 enables a clear and direct view of the impact of this mediator in CRPC cells. 3.1.1 Cell lines and cell culture Osteoclasts RAW 264.7 macrophage cells were used as the OC precursor. �ey originated from Abelson leukemia virus transformed cell line derived from BALB/c mice [196] and has been proven to be an useful tool to study OC due to its ability to differentiation to OCs [197]. �e OC maturation was induced by RANKL and mature OCs were defined by TRAP positivity, which is an important marker in OC bone resorption [198]. Prostate cancer cell lines Using CRPC cell lines serves to understand the different effects of RUNX2 and OCs to the different phenotypes of CRPC. LNCaP-19 is an in vitro derived castration resistant sub-clone of the widely used- LNCaP cell line originally established from a lymph node metastases from a pa- tient with CRPC [199]. Previous studies from our group have shown that LNCaP-19 cells grow without androgen but express AR and are still responsive to androgens, it has increased angiogenic potential and displays several molecu- lar signs of increased metastatic potential; in bone, it forms osteoblastic lesion in nude mice tibia model [200, 201]. 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS 39 �e PC-3 cell line is also a castration resistant cell line. It was isolated from a bone metastasis of a prostatic adenocarcinoma [202]. Different from LNCaP-19, the PC3 cell line exhibits low 5-alpha reductase activities and does not express AR or PSA [203, 204]. PC-3 forms mainly osteolytic lesions when injected in bone [205]. 3.1.2 Co-culture model CRPC cells were co-cultured with osteoclasts using the Transwell permeable support in 6-well plates. �e CRPC cells and OCs were seeded separately and assembled together when the CRPC cells reached 80% confluence. As control cells were used the original RAW 264.7 cells without RANKL stimulation. Both OCs and controls were washed with PBS, and LNCaP-19/PC-3 culture medium was added prior to the co-culture. �e co-culture system, although excluding the direct contact effects, is a good in vitro model mimicking the condition of CRPC cells growth in the presence of OC. �e 0.4 µm pore sized trans-permeable structure allows the communication between CRPC cells and OCs, as well as separate collection of the CRPC cells for various analysis. 3.1.3 RNA interference In order to examine the effects of knocking down endogenous RUNX2 on steroidogenic targets, small interfering RNA (siRNA) targeting RUNX2 or non- silencing control siRNA were used to transfect LNCaP-19 and PC-3 cells in Pa- per 2. siRNA has high specificity and potent silencing ability [206]. Cells were transfected using the Amaxa®Nucleofector® system using LONZA transfection reagents from LONZA according to manufacturer’s instructions. After 48 hours, the transfected cells were seeded for co-culture or further analysis. 3.1.4 Knock out experiment To follow up on the results from knock-down experiments, a knock-out of RUNX2 using CRISPR/Cas9 plasmid in LNCaP-19 cells were carried out for permanent inhibition of RUNX2 expression suitable for long-term in vivo experi- 40 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS ments. In CRISPR gene knockout, the Cas9 protein induces a site-specific dou- ble strand break (DSB) in the genomic DNA with the guide of gRNA sequences, which blocks expression of the gene. �is method is a powerful tool due to its high efficiency and specificity in stable knock out experiments [207]. Control-plasmid or RUNX2 double nickase plasmid were transfected using Mirus TransIT-X2® Dynamic Delivery System according to the manufacturer’s protocol. After incubation, GFP-positive cells were manually isolated under mi- croscope for successfully-transfected cells to proceed with puromycin screening growth for successfully transfected cells. 3.1.5 Proliferation and Apoptosis Assay �e proliferation and apoptosis assay were carried out with flow cytometry. PE Annexin V, with its ability of binding to phosphatidylserine, was employed as a marker for apoptosis. EdU (5-ethynyl-2’-deoxyuridine is incorporated into newly synthesized DNA by cells) and was used for assaying proliferation. A flu- orescent azide iFluor-488, small enough to diffuse freely through native tissues, is added and covalently cross-links to the EdU in a 'click' chemistry reaction. Flow cytometry results were statistically evaluated with unpaired t-test with Welch’s correction. 3.1.6 Scratch-wound assay In order to investigate the effect on PC cell migration, the scratch-wound assay was employed. LNCaP-19 Control and LNCaP-19 RUNX2 KO cells were plated in 6-well plates with normal growth medium 48 hours before the assay. When the cells were at 95% confluence, a wound was scratched with 200-μl pipette tips through the center of the culture wells. �e cells were maintained in normal cul- ture conditions and images of wounds were taken at the start and after 24 hours. By comparing the images taken at start and different time point, the migration rate was evaluated [208]. 3.1.7 RNA isolation and reverse transcription Total RNA was extracted from tumor cells using the RNeasy Mini Plus accord- ing to the manufacturer’s instructions. RNA concentration was measured on a 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS 41 NanoDrop. A total of 1 µg RNA per reaction was reverse transcribed into cDNA using the VILO Superscript cDNA synthesis kit according to the manufacturer’s instructions. 3.1.8 Quantitative RT-PCR For investigating genes affected by RUNX2 alteration, single Taqman assays were employed. Gene expression analysis was performed on an ABI Prism 7500 Fast Sequence Detector. �e expression levels of each sample were normalized against GAPDH and the ΔΔCt method was used for relative mRNA quantifica- tion. PCR reactions were performed in duplicates for all samples, originating in at least three independent biological replicates. 3.1.9 Protein Assay Proteins were isolated from whole cells in RIPA Lysis supplemented with phos- phatase inhibitor and protease inhibitor cocktail and collected by centrifugation. Protein concentrations were determined using the Bradford assay. In the Western blot analysis, 20 μg protein of cell lysates were resuspended in NuPage LDS sample buffer/NuPage reducing agent before loaded and separated on a 4–12% Bis–Tris gradient gel with MOPS-running buffer and transferred onto polyvinyldifluoride (PVDF) membranes using the i-Blot gel transfer sys- tem. �e membranes were blocked in Clear Milk Blocking Buffer for 1 hour in RT and incubated with primary antibodies over night at 4 °C. Membranes were washed with PBS-T and incubated with secondary antibodies for 1 hour at RT. �e immunoreactions were detected using the Amersham™ ECL Select™ west- ern blotting detection reagent (GE Healthcare; UK) following manufacturer’s in- struction. 3.1.10 RNA sample handling and sequencing �e consideration behind utilizing the RNA sequencing is the possibility of get- ting a broad picture of the effects on genes and pathways, from which the most related or changed genes could be selected for further validation and studies re- garding the interaction between osteoclasts and CRPC cells. 42 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS �e RNA samples were prepared as described in 3.1.7, after normalization for equal input, samples were spiked with ERCC RNA Spike-In Control Mix (�er- moFisher). Libraries were prepared using QuantSeq 3’ mRNA-Seq (FWD) Li- brary Prep Kit for Illumina (Lexogen) including individual indexing for multiplex sequencing. Finalized libraries were quality checked with capillary electrophoresis (Fragment Analyzer) and quantified with Picogreen (�er- moFisher). �e libraries were normalized and pooled before NextSeq500 se- quencing (Illumina). �e sequencing was performed by TATAA Biocenter (Göteborg, Sweden) 3.1.11 Gene ontology and pathway analysis �e pipeline Human (GRCh38) Lexogen QuantSeq 2.6.1 was used for the se- quence data analysis. After mapping the reads against the ENSEMBL reference genome, we merged the reads count files using a custom bash script. ENSEMBL identifiers were translated into HGNC gene names using a custom R script that retrieved the corresponding gene names from the BioMart database. Differential gene expression was performed using the R-package edgeR. Multiple testing correction is performed using Benjamini-Hochberg false discovery rate (FDR) method. A gene is considered differentially expressed if FDR < 5% and the log- transformed fold change (log2(FC)) > 1 (up-regulated) or log2(FC) < −1 (down- regulated). All non-protein coding RNAs were removed prior to analysis. Genes that were statistically significantly affected by OC co-culture compared to control cells were categorized in 10 groups based on their fold change in PC-3 and LNCaP-19. �e upregulated genes were grouped in those that were upregu- lated in LNCaP-19 and those upregulated in PC-3, those that were upregulated in both cell lines (UP_UP), and those that were upregulated uniquely in one of the cell lines. �e same grouping was applied for the down-regulated genes. �ese gene lists were assessed for GO ontology enrichment using PANTHER Overrepresentation Test (released 2020-07-28) annotated with GO Ontology da- tabase DOI: 10.5281/zonodo.4081749 (released 2020-10-09). �e annotation Data set used was GO biological process complete the analysis was performed using the Fisher’s Exact test with Bonferroni correction for multiple testing. Only GO terms with a Bonferroni adjusted P-value < 0.05 and a fold enrichment > 2 was taken into account. For each enriched ontology family tree the term with highest Bonferroni adjusted P-value is shown in the table. If that term was displayed as a parental term, also its child term with the highest fold enrichment is presented (if adjusted P < 0.05). Significantly changes genes subdivided in the same way as 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS 43 above were associated with KEGG pathway terms using the Database for Anno- tation, Visualization and Integrated Discovery (DAVID) Bioinformatics Re- sources 6.8 (https://david.ncifcrf.gov/). Differences in gene expression in the osteogenesis PCR array were statistically evaluated using Student’s t-test. �ese calculations were performed in IBM SPSS Statistics. 3.2 In Vivo Experiments 3.2.1 Animal models Immunodeficient mouse models have been widely used in cancer research in- cluding prostate cancer. �e BALB/c nude mice used in the present study lack a thymus and are unable to produce T cells, resulting in an inadequate immune system, enabling growth of human cells in vivo. A drawback is that effects of the immune system on tumor progression could not be fully addressed in this model. Another limitation is that the sex steroid metabolisms are not exactly the same between human and mice. Compared to human, mice lack circulating sex hor- mone binding globulin (SHBG), thus the effect of low circulating T could be more potent in mice [209]. Based on the metastatic patterns (trabecular bones, long bones like femur, tibia, ribs,) of PC we used the intratibial model with the purpose of optimally explor- ing the interaction of CRPC cells and bone microenvironment as well as the role of RUNX2 in the process. 3.2.2 Animal handling and intratibial implantation �e protocol of animal experiments was approved by the animal ethic committee in Gothenburg; the procedures have been performed tightly following the in- structions of the Laboratory for experimental medicine (EBM) at Sahlgrenska Academy, University of Gothenburg. �e male mice were 8 to 9 weeks old (sexually mature) when experiments started. To simulate the human CRPC condition, we castrated the male mice be- fore tumor implantation. 44 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS �e intratibial implantation was carried out by injecting the CRPC cells into the bone marrow of tibia bones through the knee joint. Matrigel was used as the basement membrane together with the cell mix to facilitate tumor cell establish- ment. 3.2.3 MRI experiments In order to better understand how RUNX2 is affecting CRPC-bone metastasis, we applied MRI to follow the tumor growth and progression in real time. MRI is a non-invasive technique widely used in diagnostic medicine and bio- medical research, which can provide comprehensive and multi-parametric infor- mation of the patients. It uses a strong magnetic field, gradient magnets, and radio waves to produce its images. CT is often used in in vivo studies for tumor, however, CT/X-ray imaging involves imaging contrast agents with high atomic numbers to better absorb X-rays and thus provide the signal. �ese radiation agents could affect the tumor activities in vivo. By using MRI, higher quality im- ages of tumors were acquired without the disturbance of radiation. �e mice were maintained anaesthetized during the imaging experiment. A re- gion of interest (ROI) was then created on each section of the MRI volume by manually tracing out the tumor border. Pixels intersecting the tumor border were excluded in order to avoid partial volume effects. �e tumor volume was then calculated by summing up the number of pixels within all ROIs and multiplying it by the voxel dimensions, including the slice gap using a in-house developed graphical user interface in Matlab. 3.2.4 Immunohistochemistry Immunohistochemistry (IHC) provides a forthright scope of the localization and expression for the markers of interest. �e markers were chosen based on the mechanisms we aimed to exam: proliferation and survival, migration, AR signal- ling, and steroidogenesis. �e intratibial tumor tissue and attached tibia were dissected and fixed in forma- lin, decalcified in EDTA and embedded in paraffin. Tissue sections were pre- 3 . METHODS AND METHODOLOGICAL CONSIDERATIONS 45 heated, deparaffinized and rehydrated before antigen retrieval, endogenous pe- roxidase blockage and antibody incubation. Section incubated without primary antibody used as negative control. Positive percentage of KI67 staining was calculated using IHC profiler in Image J in paper 3. Evaluation of the other staining was performed using a semi-quanti- tative scoring system multiplying staining intensity and area of positive tumor cells. Intensity and area was scored as 0 to 5 and multiplied to give a value for five different microscopic fields per section. 3.2.5 Steroid measurements To attempt to detect the difference in serum hormone levels due to RUNX2 knockout of tumor cells implanted intratibially in mice, serum concentration of progesterone, androstenedione, testosterone, estradiol, estrone and dihydrotestos- terone (DHT) were measured in a single run by GC-MS/MS. After addition of isotope-labeled standards, steroids were extracted to chlorobutane, purified on a silica column and derivatized using pentafluorobenzylhydroxylamine hydrochlo- ride followed by pentafluorobenzoyl chloride. Steroids were analyzed in multiple reaction monitoring mode with ammonia as reagent gas using an Agilent 7000 triple quadrupole mass spectrometer equipped with a chemical ionization source. �e limits of quantification (LOQ) for estradiol, estrone, testosterone, DHT, pro- gesterone and androstenedione were 0.5, 0.5, 8, 2.5, 74, and 12 pg/ml, respec- tively [210]. 3.2.6 Statistics In Paper 1, differences in gene expression in the osteogenesis PCR array were us- ing Student’s t-test. Flow cytometry results were analyzed with Welch’s t-test. In Paper 2, statistical differences between different treatments were measured with Student’s t-test. In Paper 3, Mann–Whitney test was employed in IHC anal- ysis. All data are presented as mean ± SEM. �e MRI method authentications were using Pearson correlation and Bland-Altman analysis. A P-value of <0.05 was considered significant. 46 4 . RESULTS AND COMMENTS 4. Results and comments Paper I. “A broad investigation that demonstrates that OCs directly influences the growth and function of CRPC cells.” Cancer cells benefit from growth factors released from the OC degraded bone matrix, and the cancer cells secrete stimuli to activate the OBs, and in turn (via RANKL) OCs and bone turnover, and thereby the access to more released growth factors [211]. Previous research in our group has identified the importance of OBs in regulating PC cells in the progress of osteoblastic CRPC metastases [179]. Few studies have been done on the specific roles of OCs in PC progres- sion, and based on our previous knowledge of the osteoblastic phenotype, this study was designed to investigate the effect of mature OCs on CRPC cell pheno- type and the difference between osteoblastic and osteolytic CRPC cells in this aspect. To evaluate the impact on proliferation or apoptosis, a flow cytometry assay was performed on CRPC cells after co-culturing with mature OCs or control RAW cells. �e osteoblastic LNCaP-19 cells co-cultured with OCs displayed a signifi- cant decrease of end stage apoptotic cells compared to the control condition, while the decrease in apoptosis in osteolytic PC-3 cells was not statistically sig- nificant. Furthermore, mature OCs promoted proliferation in both LNCaP-19 and PC-3 cells. �ese results proved that OCs have a direct effect on CRPC apoptosis and proliferation, especially potent on osteoblastic CRPC cells. Based on the significant effect from the flow cytometry assay, we further exam- ined the gene expression changes and affected pathways induced by the co-cul- ture using RNA-sequencing. Forty-six percent of mRNAs were significantly affected by OC stimulation. Of these, we focused on the 3566 protein coding genes. Gene expression in PC-3 was generally more responsive to OC stimula- tion compared to LNCaP-19. 4 . RESULTS AND COMMENTS 47 Gene ontology (GO) enrichment analyses revealed that LNCaP-19 and PC-3 re- spond to OC stimulation largely in the same manner. GO terms that were en- riched based on the lists of upregulated genes in PC-3 as well as in LNCaP-19 were DNA repair, DNA replication and DNA replication initiation; based on the same groups of affected genes, KEGG pathway analysis identified DNA replica- tion, Fanconi anemia pathway, and Metabolic Pathways. GO terms that appear as enriched based on significantly altered gene lists both for PC-3 and LNCaP-19 were cell cycle arrest and response to ER stress. However, differences between the cell lines were also observed. As suggested by the number of affected genes, more GO terms turned out as significantly enriched in PC-3 compared to LNCaP-19. GO terms that were only enriched among upregulated genes in PC-3 cells were protein transport along microtubule and glycoprotein biosynthetic pro- cess. �e only GO term significantly enriched uniquely among upregulated genes in LNCaP-19 was mitotic cell cycle, which mirrors the more significant re- action in proliferation with LNCaP-19 cells, and indicate that these are more sus- ceptible to bone cell stimulation. Genes that were downregulated only in PC-3 were associated with GO terms such as response to decreased oxygen and regu- lation of MAPK cascade. ER stress is a common cellular stress response which is caused by disturbance of cellular homeostasis, and activates cell apoptosis [212]. From the enriched GO terms and the specific gene analysis, we could show that OCs decreased gene ex- pression related to ER stress induced apoptosis in both cell lines. ATF3, ATF4, DDIT3, TRIB3, and CHAC1 were among the genes shared by both cell lines in the top 20 most downregulated genes list, and are heavily involved in ER stress induced apoptosis mechanism [213, 214]. Further protein analysis showed that ATF3 and DDIT3 were confirmed as downregulated, with DDIT3 protein level decreased in both cell lines, and ATF3 downregulated only in LNCaP-19, which could be related to the larger impact on apoptosis when co-cultured with OCs. �e overall lower expression of genes associated with ER stress induced apopto- sis may indicate that this mechanism for apoptosis induction was specifically de- creased by OCs. Fanconi anemia pathway is a common DNA repair pathway that resolves DNA interstrand cross-links, lethal DNA lesions to prevent DNA replication and tran- scription blockage [215]. Several of the Fanconi anemia genes upregulated by OCs overlaps with the homologous recombination repair pathway (BRCA1, BRCA2, PALB2, BRIP1, and RAD51C) implicated as important for sensitivity to treatment with PARP-inhibitors [216]. As to the protein level, BRCA1 and PALB2 were increased in PC-3, while the effect was not seen as clearly in 48 4 . RESULTS AND COMMENTS LNCaP-19. Defects in DNA repair of cancer cells have been used as a treatment strategy for chemotherapy that causes DNA damage [217]. �us, the upregula- tion of this certain pathway by OCs may suggest an increasing resistance of CRPC against certain therapies like radiation therapy and PARPi mentioned above [218]. Besides the survival and proliferation benefits from the OCs, to progress in the bone environment, the CRPC cells need to be equipped with bone like proper- ties. We used an osteogenesis focused PCR panel to study the effects of OCs on gene expression related to bone. In line with the previously described data, that osteoblastic LNCaP-19 cells, rather than PC3, respond more to the stimulation of OBs [219], osteolytic PC-3 was more sensitive to OC co-culture compared to LNCaP-19. �ese changes indicate a pattern where OCs induce PC-3 cells to ex- press genes for secreted factors that inhibit OC differentiation and activation [35]. In addition, we could show that OCs downregulated both osteoblast cad- herin (CDH11) and N-cadherin (CDH2) indicating a decreased capacity of attach- ment of the PC-3 cells to other cell types in the bone microenvironment. A limitation of the present study lies in macrophage precursor cells intermixed with the osteoclasts in the co-culture system. �e matured osteoclasts and pre- cursor cells could still have interactions and affect each other, which might im- pact the tumor cells in undetermined ways. Also, more specific studies on the effect of the precursor could be helpful to determine the function of pure OCs. In Paper 1, we demonstrated that OCs directly influences the growth and function of CRPC cells. With the help of RNA sequencing, we pointed out ER stress in- duced apoptosis and DNA repair pathways as targets for the OCs influence, to- gether with promoting survival and proliferation of CPRC. Furthermore, we displayed the difference responses of osteoblastic CRPC cells and osteolytic CRPC cells to the stimulation of OCs regarding bone biology, indicating an im- portant role for OCs on the phenotype of CRPC progression in bone. 4 . RESULTS AND COMMENTS 49 Paper 2. “RUNX2 as an important mediator of steroidogenesis in CRPC cells with potential to influence castration resistant AR signalling. OCs promote RUNX2 regulated induction of key steroidogenic enzymes, influencing activation of AR in CRPC cells.” As previously described, OBs are critical for bone formation and steroidogenesis in osteoblastic CRPC [179]. Our studies have also demonstrated that RUNX2 could mediate the de novo steroidogenesis of CRPC under the influence of OBs [220]. Although OCs are one of the crucial components in the “vicious cycle” of bone metastases, a possible role of OCs in intratumoral steroidogenesis has not been defined. Moreover, the regulatory role of RUNX2 in bone-cell induced ex- pression of steroidogenic enzymes in mCRPC cells is less known, especially in osteolytic CRPC. In Paper 2, we further investigated the function of OCs on steroidogenesis and the involvement of RUNX2 in this process for both osteo- blastic and osteolytic CRPC cells. We combined in vitro and in vivo experiments to examine the expression of RUNX2 in LNCaP-19 and PC-3 cells and we could show that OCs have the po- tential of altering RUNX2 levels in osteoblastic LNCaP-19 cells. �e in vitro mRNA levels of RUNX2 as well as the in vivo expression of RUNX2 in in- tratibial tumors of LNCaP-19 are generally lower than those of PC-3, although the difference in vivo was less clear. However, the osteoblastic LNCaP-19 cells showed a greater response to co-culture with OCs by significantly increased ex- pression of RUNX2 compared to the not significantly affected osteolytic PC-3 cells. Interestingly, in the intratibial model, RUNX2 staining of LNCaP-19 and PC-3 tumors appeared with different patterns, AR-negative PC-3 tumors having prominent nuclear staining compared to the LNCaP-19 tumors. How OCs induce differential expression patterns of steroidogenic enzymes in os- teogenic and osteolytic CRPC cells could be concluded from the androgen signa- ture array plate on osteoblastic LNCaP-19 cells and osteolytic PC-3 cells stimulated by OCs. Compared to insignificant effect from OCs in LNCaP-19 cells, steroidogenesis related genes were generally increased by OCs in the oste- olytic PC-3 cells. �ese observations complemented with previous data on osteo- blastic LNCaP-19, which greatly responded with increased expression of steroidogenesis related genes to OB stimulation [221]. �e results suggest that 50 4 . RESULTS AND COMMENTS different bone cells may play different roles in corresponding phenotype of CRPCs. Furthermore, we adopted the same array plates to demonstrate the role of RUNX2 in regulation of gene expression related to steroidogenesis and ster- oid signaling with or without OC stimulus by knocking down the expression of RUNX2 with siRNA in LNCaP-19 and PC-3 cells. In line with previous observa- tions, PC-3 and LNCaP-19 cells displayed opposite reactions to RUNX2 silenc- ing, with an overall promoting influence from RUNX2 in LNCaP- 19 and a suppressive in PC-3. OCs in the co-culture condition with RUNX2, severed a generally positive effects, enforcing RUNX2 promoting patterns in LNCaP-19 and compensating the suppressed effect from RUNX2 in PC-3 cells. Expanding from the array to single Taqman expression assay, we could display a clearer mechanism of RUNX2 together with OCs in mediating the effects on steroidogenic pathways in LNCaP-19 and PC-3 cells. In LNCaP-19 cells, the RUNX2 regulation of steroidogenesis signaling is pro- foundly dominant compared to that of OCs. OCs did not significantly promote expression of any of the steroidogenic genes in LNCaP-19. However, expression of important steroidogenic genes (CYP11A1, AKR1C3, HSD3B1, and CYP17A1) were significantly decreased due to silencing of RUNX2. �e effect on CYP11A1 and HSD3B1 was solely enforced by OC stimulation with RUNX2 knockdown, while AKR1C3 was regulated by RUNX2 in the control setting and CYP17A1 was purely regulated by RUNX2. AR signaling was affected by the combination of RUNX2 and OCs in LNCaP-19 cells. OC stimulation revoked the downregula- tion of AR by silencing RUNX2. KLK3 expression was increased by osteoclasts co-culture in LNCaP-19 cells, possibly due to the increased levels of RUNX2 in- duced by OCs, which could directly promote KLK3 expression [188, 222]. In contrast, RUNX2 silencing increased KLK3 in the control setting, which conceiv- ably could be attributed to the RUNX2-AR inhibition mechanism [223]. �e above discoveries could suggest that OCs together with RUNX2 are involved in the complex AR activation mechanism in CRPC and have complementary roles in promoting and inhibiting actions. Lack AR-expression, PC-3 cells displayed quite different reactions to RUNX2 and OCs, indicating alternative pathways of steroidogenic signaling and func- tions in CRPC progression. Unlike the co-dependent RUNX2 and OC mediation in LNCaP-19, the expression of CYP11A1 in PC3 was increased by OCs only, which may be attributed by the generally high level of RUNX2 in PC-3 cells. �e expression of the enzymes more downstream in the steroid synthesis path- way, AKR1C3, HSD17B3, and CYP19A1, were all downregulated by silencing of 4 . RESULTS AND COMMENTS 51 RUNX2 in the presence of OCs. However, CYP19A1, which converts testosterone to estrogen, was increased by RUNX2-silencing in the control setting. Estrogen receptor β (ERβ), encoded by ESR2, has been proved to inhibit proliferation and induce apoptosis of PC cells both in vitro and in vivo conditions [224]. We showed that the expression of ESR2 was repressed by RUNX2, which possibly lead to that RUNX2 promoting proliferation of PC-3 cells. �is suggestion was further supported by the decreased MKI67 proliferation marker in PC-3 cells with silenced RUNX2. In the PC-3 cells, RUNX2 and OCs together serve as an important part in promoting proliferation rather than the steroidogenesis. In Paper 2, we report that RUNX2 together with OCs, are regulators of steroido- genesis and proliferation in the CRPC cells. �e OC and RUNX2 co-operation may also play a part in AR signaling of LNCaP-19. �e difference in reaction to the osteoclast-RUNX2 stimulation between LNCaP-19 and PC-3 cells points out the need for further investigation about whether the cause of this observation is the osteo-phenotype or AR status of the CRPC cells. 52 4 . RESULTS AND COMMENTS Paper 3. “RUNX2 may be a crucial component in PC bone metastasis as well as a possible treatment target for mCRPC” From the results in Paper 2, we identified the importance of RUNX2 in steroido- genesis and AR activities in LNCaP-19 cells in vitro as well as a promoting role in survival and proliferation. However, how RUNX2 affects the CRPC cells in bone microenvironment in vivo still needs to be investigated. In Paper 3, we es- tablished a RUNX2 knock-out LNCaP-19 cell line and used an intratibial tumor model together with MRI to study how RUNX2 influenced tumor progression in the bone environment. MRI is a non-invasive technique widely used in diagnostic medicine and bio- medical research, which can provide comprehensive information of the patients’ diseases. Our group has previously used MRI to successfully follow orthotopic prostate cancer xenografts [225], and in the present study we further explored its possibility for the intratibial setting. �e validation of MRI in assessing the tu- mor positivity and volume was performed by comparing to histological sections. In defining the tumor positivity, we showed that MRI had a sensitivity and speci- ficity of 81.3 % and 88.0 %, respectively. Besides the capacity of MRI to illus- trate the location and structure of the tumors precisely, our result show that MRI also provides a good view of tumor features, as demonstrated by a comparison with histological staining (HS). We could successfully identify the same struc- ture from the MRI and HS, for example osteoblastic and osteolytic tumor areas, the bone leakage and tumor growth outside the tibia. Additionally, our MRI model also proved to provide high reliability in assessing the tumor volume, the calculations of tumor volumes using MRI and HS correlated well. �e evalua- tions demonstrated that MRI is an approved method for tibial tumor detection and growth assessment. With the MRI as a competent tool for analysing tibial tumors, we could show that RUNX2 affects the tumor growth and proliferation in the bone environment. Not only the tumor volume of the RUNX2 knock-out tumor was significantly smaller, but also the growth rate of tumor was lower compared to the control condition. Complementing the in vivo growth data, the proliferation marker Ki67 4 . RESULTS AND COMMENTS 53 was significantly lower in the lack of RUNX2, both in vitro and in vivo, support- ing the importance of RUNX2 for tumor growth Past studies have shown that RUNX2 boosts tumor progression by affecting the cadherin switching and stimulating the expression of MMPs [226]. RUNX2 me- diates the interaction of PC bone metastases and bone remodelling system by in- fluencing the cellular response to TGFβ/BMP signalling [192]. TGFBR1 is expressed at abundant levels in most of PC cell lines, and has been shown as an inhibitory pathway [227]. In line with previous findings, we identified a decrease in migration activity in vitro in the RUNX2 KO group, which is possibly accom- plished by mediating the EMT signalling. In line with this, the in vitro analysis showed a decreased expression of N-cadherin (CDH2) as well as of the expres- sion of TGFβ and TGFBR1, known regulators for cell migration and N-cadherin expression [228]. Interactions between RUNX2 and AR regulating their transcriptional activity have been described above, and the present study displayed that with knock-out of RUNX2, AR together with KLK3 protein levels were decreased in vitro. �e downregulation was also confirmed in the intratibial tumor samples, which fur- ther emphasized the strong connection of RUNX2 and AR in the LNCaP-19 CRPC cells. Intratumoral steroidogenesis has been recognized as an essential factor of CRPC [89]. Since we have shown tumor growth difference and AR alteration between RUNX2 knock-out and control/wt cells, steroidogenesis, was investigated. Cor- responding to findings in Paper 2, the RUNX2 knock-out LNCaP-19 cells showed a generally downregulated pattern of steroidogenesis. �e expression of CYP11A1, and ARK1C3 were significantly decreased both in vitro and in in- tratibial RUNX2-KO tumors. However, CYP17A1 and HSD3B1 expression in- creased on the mRNA level in the RUNX2 knockout cells. �e dynamic alterations in steroidogenic enzymes suggest a possible change in steroidogene- sis of LNCaP-19 cells. Instead of de novo synthesis from cholesterol, other up- takes might be dominant in lack of RUNX2. However, since the increase in HSD3B1 could not be determined in KO-RUNX2 cells in vivo, the bone micro- environment may influence this effect, as could also be seen in Paper 2. To deter- mine the effect of these changes on the resulting steroid levels, the serum steroid levels were measured in the mice. However, no detectable differences were found between the RUNX2-KO tumors group and the control/wt-tumors group, so whether the affected steroidogenic signalling resulted in changed steroidogen- esis level could not be determined in the current study. 54 4 . RESULTS AND COMMENTS �is study has some limitations that should be considered when interpreting the results. �e overall tumors establishing rate after the injections were low, with the most efficient establishment in the wild-type (wt) group. After a qualitative comparison of knockout-control tumors and wt-tumors for investigated proper- ties, they were judged to behave similarly although the numbers of tumors were small. �erefore, we combined the RUNX2 Knockout-control group (i.e. trans- fected with CRISPR-Cas9 control constructs) with the LNCaP-19 wild-type group (not subjected to any transfection) to enable group wise comparisons to the RUNX2 KO group. �is makes any conclusions regarding RUNX2 function in intratibial tumors preliminary, and complementing studies are needed. To conclude, Paper 3 established a functioning and reliable MRI detection method for CPRC bone metastases. We demonstrated that RUNX2 is involved in CRPC progression in bone microenvironment through mediating EMT signalling and altering intratumoral steroidogenesis pathways. Our results suggest that RUNX2 is a promising target for treatment and prevention of CRPC bone metas- tases. 5 . DISCUSSION AND FUTURE PERSPECTIVE 55 5. Discussion and future per- spective Bone metastases, the life wrecker for PC patients A contributing cause of death for the vast majority of patients with PC is bone metastases, which display a spectrum of osteoblastic, osteolytic or mixed bone responses [229, 230]. The skeletal complications mostly occur in the mCRPC phase, not only do they heavily disrupt the life quality of patients with pain, frac- ture, defective mobility and symptomatic hypercalcemia, but also have great im- pact on the prognosis of PC. Commonly, the bone metastases represent the mass of tumor burden for patients with advanced PC [231, 232]. Solid tumors like PC could elicit the bone marrow and further induce osteolytic reaction and abnormal formation of woven bones. It has been shown that the bone metastases of PC present mixed osteoblastic and osteolytic phenotypes with osteoblastic being the predominant property [233]. The distribution of bone metastases is extensively related to clinical outcomes of PC patients and continu- ing attention has been drawn on the bone health of PC patients, as well as mark- ers that predict and indicate SRE. In a Danish study of a 23,087 incident patients cohort, the adjusted 1-year mortality rate ratio of patients with bone metastases and SRE is 6.6 compared with bone metastasis free patients, suggesting a signifi- cantly poorer prognosis [234]. Iglesias-Gato et al. have demonstrated that the bone metastatic PC tumors were more heterogeneous, and that the expression of proteins involved in DDR pathways are increased compared with primary PC tu- mors, supporting the strong connection of bone metastases and PC progression [235]. ADT, the frontline treatment method for mPC, is associated with bone related side effects. PC patients receiving chronic ADT suffer from significant bone loss, especially during the initiation of the ADT treatment, which leads to bone fracture and bone pain [236]. The ADT initiating patients have substantially lower level of estradiol (47% lower than healthy males; 31% lower than ADT free PC patients), which is critical for bone mass in men [237]. Thus, the consid- eration of preventive or combination therapy towards bone loss condition is of great benefits for PC patients receiving ADT treatment. 56 5 . DISCUSSION AND FUTURE PERSPECTIVE For patients with mCRPC, monitoring of SREs are important. PSA is usually a biomarker for PC progression. However, for patients receiving bone metastases targeting treatments, bone remodelling markers like serum total osteocalcin, bone-specific alkaline phosphatase (BALP)and N- terminal telopeptide (NTx) are highly indicative in evaluating treatment outcomes [232, 238]. Bone Scan In- dex (BSI) is a method to quantify metastatic lesions from bone scans; on-treat- ment BSI shift is an indicator of patients’ survival [239, 240]. Researchers have also succeeded in developing automated BSI to predict the overall survival of mCRPC patients and suggesting potential of value of automated BSI for PSA monitoring of mCRPC patients receiving enzalutamide treatment [241]. Berruit et al. have showed the predictive usage of other bone related indicators like bone disease extent and bone pain intensity in risk of SREs [242]. Increasing knowledge of SREs from ADT or bone metastases could facilitate the development of more targeted and precise treatment plans for PC patients. For example, the timing for introducing bone targeting treatments like bisphospho- nates and whether to start with palliative treatments with radiation. �e focus on osteoclasts and why The OB is the bone cell most obviously coupled to the phenotype of PC bone metastases, since they are mainly described as osteoblastic, but the role of OCs is equally important. As described in the introduction, the normal remodelling of bone is shouldered by both OBs and OCs. During the PC bone metastatic process, the abnormal bone formation is coupled with excessive bone resorption, and the dysregulation both OBs and OCs con- tribute to the progression of PC bone lesions. Recent studies have successfully linked the high level of NTx, a bone resorption marker used for osteoporosis, with poor prognosis of PC patients [243, 244], showing that bone resorption is a critical component also in osteoblastic metastatic disease. Bone is a rich soil of growth factors and bone matrix factors; one major role for OCs in the bone metastases is the digger for the soil to fertilize the tumor cells. Growth factors like TGFβ, FGFs, and IGF1 are beneficial for tumour growth [245], and other factors released through bone resorption like BMPs,ET-1 and extracellular calcium are important for OBs’ bone formation [246]. The release 5 . DISCUSSION AND FUTURE PERSPECTIVE 57 of such factors from the degraded bone is the basis for the tumor promoting vi- cious cycle in which the OC function is critical. This importance of OCs in progression of PC bone metastases has been demon- strated by in vivo studies. For example, Guise et al. treated tumor-bearing mice with atrasentan (an ET-1 inhibitor), bisphosphonate (osteoclast inhibitor) and ve- hicle. Their study showed that each of the two inhibitors was able to reduce oste- blastic response with similar efficiency, and the combination treatment of the two provided the best outcome on tumor volume and PSA levels [247]. Hence, it is of crucial importance to investigate the regulation of OCs to identify targets for treatments of PC tumour bone metastases. OCs have for long time been a target for therapy of PC bone metastases, for in- stance, the commonly used treatments with bisphosphonates (zoledronic acid) and Denosumab. Bisphosphonates tackle OC by a direct effect on OCs’ activities and indirect through interfering with OB-OC signalling. The accumulated bisphosphonates in the bone released by bone resorption could interfere with the OC osteolytic enzyme activities and further induce OC apoptosis [248]. In addi- tion, bisphosphonates are able to react with OBs to increase the OPG production and reduce the RANKL level, which diminishes the OC maturation [249]. Deno- sumab, a monoclonal antibody, acts as a RANKL blocker. By binding to RANKL, it prevents the activation of RANK and differentiation of OCs [250]. Denosumab can decrease the functioning OC number to near zero and thereby moderate the bone resorption actions. Bisphosphonates and denosumab could reduce the risk of SREs and prolong the time to new events, reduce the bone pain and improve the quality of life of PC patients, and they have been recommended to use with the bone metastases diag- nosis [251, 252]. Zoledronic acid has been demonstrated to reduce the risk of SREs to 11% in CRPC patients in 2 years compared to placebo [253]. Pamidro- nate has proved to significantly reduce the bone resorption and prevent the bone loss in ADT patients [254]. As to denosumab, it has showed better performance than zoledronic acid in preventing SREs for CRPC patients with bone metastases in several phase 3 trials [255]. 58 5 . DISCUSSION AND FUTURE PERSPECTIVE Figure 10. Mechanism of action of denosumab and bisphosphonates In the present study, we demonstrate that OCs directly influence the CRPC cells, suggesting OCs to play a more specific role in mediating PC growth in both os- teolytic and osteoblastic metastases. Under the exposition of OCs, osteolytic PC cells would go through changes that are more aggressive and tend to be more in- vasive. We found that AR became overexpressed by the effect of OCs in an androgen sensitive CRPC cell line, which could indicate that OCs induce the PC cells to go through changes of proliferation and other behaviours regulated by andro- gens. Interestingly, in the AR negative and androgen insensitive CRPC cell line PC-3, OCs caused an increased expression of CYP11A1 and CYP19A1. The ef- fect on CYP19A1, converting testosterone to estradiol, indicates a possible switch to an estrogen dominated steroidogenic profile. Estrogens and estrogen receptors (ERs) are expressed in PC-3 cells, and it has been described that si- lencing ERα was able to suppress PC3 cell proliferation and estrogen enhanced invasion of PC cells [256, 257]. Moreover, studies have shown that aromatase (CYP19A1) is affected in PC tissue and CYP19A1 expression is highly elevated in PC metastatic tissue, as compared with primary tumors [258]. Combined with 5 . DISCUSSION AND FUTURE PERSPECTIVE 59 the results from our study, it may be suggested that OCs could affect androgen insensitive CRPC cells through the estrogen signalling pathway. Bone targetting agents, beyond restoring bone homeostasis Besides the osteoclastic focused agents, such as bisphosphonates and deno- sumab, several novel agents have been used in clinical trials or applied for use. �e focus for these agents was to restore the balance of disrupted OB and OC ac- tivities. Following that purpose, pathways including ET-1, ERs, AR and andro- gen synthesis, and certain protein kinases have been used as target sites [259]. One target of interest is ET-1, which is a proliferation and invasion factor for PC cells. �e interaction of the PC cell produced ET-1, and OB expressed ET-A re- ceptor is related to increased osteoblastic activities and decreased OC actions [260, 261]. However, neither Atrasentan (ET-A receptor antagonist ) nor Zi- botentan (ET-A receptor inhibitor) did show significant clinical improvement in phase III trials in patients with advanced PC [262]. As indicated above, estrogens and androgens are important for PC progression as well as bone remodelling. �e selective ERs modulator Toremifene is an antago- nist of ERα, but it also shows estrogenic effect in bone [263]. It has shown posi- tive effects in SREs in a two-year clinical trial for PC patients receiving ADT; compared with placebo. Toremifene not only reduced the new vertebral frac- tures, but also increased the bone mass density of hip and spine [264]. Consider- ing the involvement of estrogen signalling in PC regulation, the ERs targets do display potential. Protein kinases like tyrosine kinase receptor Met and Src kinases are influential in PC apoptosis, proliferation and invasion [265]. Met is generally upregulated in PC bone metastases, which could possibly be due to the bone mimicry property of mPC cells [266]. Src is also a mediator for OB and OC activities, especially in OC. Src deficiency could lead to disruption in bone resorption and OC migra- tion [267] [268, 269]. A c-Met inhibitor (Cabozantinib) has shown positive re- sults in reducing SREs for CRPC patients in phase II trials and been signed as a post-docetaxel option in phase III trials [270]. Dasatinib (Src inhibitor) dis- played reductions in PSA level and SREs in phase II trials [271], indicating that this pathway may be targetable in future PC treatment strategies 60 5 . DISCUSSION AND FUTURE PERSPECTIVE Radiation is an established treatment method for bone metastases. �e alpha emitter Ra-223 is a bone seeking calcium mimetic radium therapy. It causes non- reparable DSB in DNA, hence showing toxicity to tissues [272]. Ra-233 specifi- cally concentrates in active bone mineralization sites where OB is highly active. �is mechanism is attributed to Ra-233 reducing tumor induced bone formation [273]. �e ALSYMPCA clinical trial has proved the benefit of Ra-223 in im- proving overall survival for CRPC patients with chemotherapy resistance, as well as its direct effects on bone metastases [274]. So far, no drugs specifically targeting bone cells have shown benefit in terms of increased survival, but as demonstrated above, increasing PC bone metastases therapy developments have been focusing on the targets point to both PC tumors and bone cells. �e optimal “two birds one stone” agents require deeper under- standing for the interactions between PC cells and the bone microenvironment. In paper 1, we have demonstrated that OC is able to influence the ER stress and DDR related pathways in CRPC cells. Upregulation of IRE1α-XBP1 pathway in cancer cells has been shown to be a possible cause of resistance to docetaxel based chemotherapy [275]. In addition, the IRE1α-XBP1 is associated with bone pain regulation [276]. Prostaglandin binding with prostanoid receptor contributes to bone pain, and XBP1 is a cyclooxygenase-2 (COX-2) activator, an essential protein in prostaglandin production [277]. Besides COX-2 inhibitor reduces the bone pain, COX-2 inhibitors also shows inhibitory effects on bone resorption and tumor growth [278]. Interestingly in this context is the observation from the Stampede study group that the combined treatment with a Cox-2 inhibitor, zoledronic acid and ADT increased survival in a randomized trial [279]. �e ER pathway related skeletal effects and tumor progression together with OC mediated ER pathways might suggest OCs could be a possible target for ER re- lated therapy with benefits in SRE attenuation. In DDR pathways, BRCA1/2 and PALB2 have been identified to be essential in DSB repair by homologous re- combination [280, 281]. As a target for mCRPC therapy, Olaparib (PARP- inhibitor) has in a clinical trial showed promising outcome on bone metastases survival and PSA reduction [282]. Furthermore, DDR alterations are strongly re- lated with Ra-223 treatment outcome. mCRPC patients with deleterious DDR ab- errations have an enhanced reaction to Ra-233 therapy leading a significantly better bone metastatic response and overall survival [283]. Combination therapy of Ra-233 and PARP-inhibitors thus has been carried out in mCRPC patients with bone metastases. We revealed the possible effects of OCs on DDR path- ways especially BRAC1/2 and PALB2, and employing targeting of OCs could possibly aid the bone metastases treatment through DDR and Ra-233. 5 . DISCUSSION AND FUTURE PERSPECTIVE 61 RUNX2, an osteoblastic factor highly involved in PC progression and a potential treatment target Previous studies of RUNX2 are mainly focused on the OB regulation and bone remodelling, since it is recognized as the “master” transcription factor in OB dif- ferentiation and it promotes the expression of bone matrix components like oste- ocalcin [284]. RUNX2 expression is related to the development of tumors. Studies on human PC tissues have detected robust expression of RUNX2 in metastatic lesions, and expression levels of RUNX2 is correlated with the prognosis of PC [285]. �e RUNX2 staining in human PC tissue shows that the metastatic lesions have the highest intensities compared to PIN lesions and advanced primary tumors [286, 287]. Recent studies revealed the pro-metastatic role of RUNX2 in progression of ex- perimental PC to the bone. By overexpressing or silencing the expression of RUNX2 in PC cells lines, the direct effects of promoted metastatic activities or intercepted tumor growth and bone lesion formation have been observed [288]. �is could possibly be due to that RUNX2 could regulate the metastatic related MMPs such as MMP9 and MMP13 [286], VEGF, secreted bone-resorbing fac- tors like IL-8 and PTHrP [190]. Considering that RUNX2 is recognized as a pro-metastatic factor, we still lack the knowledge of the detailed signalling pathways of how RUNX2 mediate mCRPC cells in the bone. Ectopic expression of RUNX2 in the tumor cells is consequential for the osteoblastic growth of CRPC [289]. Previous work from our group has demonstrated that RUNX2 mediates genes involved in osteomim- icry and steroidogenesis in osteoblastic CRPC cells, and RUNX2 is induced in PC cells close to OBs in the mouse intratibial tumor model [220]. Moreover in the present study, RUNX2 showed different effects on tumorigenesis of osteo- lytic and osteoblastic PC cells. Combining the past work with the current study we found that neither OB or OC stimulate RUNX2 expression in the osteolytic PC cells PC-3, which is possibly due to the existing high level of RUNX2 in PC- 3 [290]. However, by silencing RUNX2, osteoblastic phenotype is developed, which points out that levels of ectopic RUNX2 relate to the different bone lesion phenotypes. [179, 291, 292]. Intratumoral steroidogenesis is critical in CRPC bone metastases development and drug resistance. In paper 2, we presented that RUNX2 cooperating with OCs 62 5 . DISCUSSION AND FUTURE PERSPECTIVE is an important mediator of steroidogenesis in CRPC cells with potential to influ- ence castration resistant AR signalling in LNCaP-19 cells. OCs come across to impact the function of RUNX2 since we observed exclusive effects from silenc- ing of RUNX2 in co-culture with OCs, for example a significant downregulation of CYP11A1 in LNCaP-19 cells and CYP19A1 in PC-3 cells. In PC cells, RUNX2 forms a protein complex with AR that largely affects the tumor progression [293], and our results suggest that this interaction influences the expression of KLK3 in response to AR in a context dependent manner. In addition, in Paper 3, where the LNCaP-19 cells were present in the bone mi- croenvironment, we further demonstrated the mechanism of RUNX2 mediating osteoblastic LNCaP-19 progression. RUNX2 is tightly related to androgen re- sponses in LNCaP-19 cells. It promoted the AR responsive expression of the prostate-specific marker PSA and affected steroidogenic enzymes like CYP11A1, CYP17A1, HSD3B1, and AKR1C3. RUNX2 is involved in mediating intratibial tumor growth and proliferation in vivo, as well as PC cell invasion and migration properties in vitro. In addition, the expression of the EMT protein N-cadherin was decreased due to RUNX2 knock out. We also showed that the expression of RUNX2 in LNCaP-19 cells is associated with cell growth and tumor development in the bone microenvironment, which both could be due to the regulatory effects on steroidogenic genes and AR activation in LNCaP-19 cells together with OCs. For PC-3 cells, we demon- strated a possible tumor suppressor estrogen receptor beta (ERβ) related pathway that RUNX2 is incorporated in mediating the PC progression, since expression of ESR2, encoding ERβ, was increased by silencing RUNX2 in vitro. Several studies have successfully impaired the progression of metastatic cancer and metastatic bone disease by targeting RUNX2 using microRNAs (miRs). For instance, the miR-196b inhibits lung cancer cell growth and metastasis, and the miR-135 and miR-203 could suppress breast cancer cell proliferation and migra- tion, as well as RUNX2 related osteolysis [294, 295]. miR replacement therapy has been in development and some have entered phase 1 trials [296], which is able to make RUNX2 targeted therapy possible in the future. To be conclusive, the present studies in paper 2 and 3 further revealed the possi- ble crucial role for RUNX2 in the growth of PC bone metastases. �e potential as a “two birds-one stone” treatment target may be specifically beneficial for mCRPC due to its connection with both bone cells and PC cells. Many signalling pathways in PC cells that promote tumour progression and bone metastases like 5 . DISCUSSION AND FUTURE PERSPECTIVE 63 BMP, TGFβ1 and Wnt is linked with RUNX2 [297], why targeting bone cells and RUNX2 expressing PC cells could be a great benefit for PC patients. �e in vitro and in vivo models, advantages and limitations PC cell lines have been employed in preclinical research since the 1980s. �e cell lines LuCaP, PC-3, LNCaP, DU145, and VCaP are most used in studies on tu- morigenesis, metastatic related events and potential therapeutic targets [298, 299]. Based on the original PC cell lines, new sublines for more specific purposes have been developed through various methods for example castrating xenografts, in- duced genetic alterations, chemical mutation, or genetic transformation [300]. �e LNCaP-19 cell line used in the present studies is derived from culturing the LNCaP cells in steroid-deficient media for numbers of passages in vitro [201]. �e advantages of PC cell line models are obvious, the low maintenance and un- limited growth ability make them suitable for gene modifications and drug screening, the data availability and further usage of in vivo are important to study the desired signalling pathways. However, the limitations of the cell line model have been criticized in many in- stances. Firstly, the PC cell lines lack heterogeneity compared to patient derived PC cells. �ey are selected from specific tumor subsets that can survive in vitro culture conditions. �e cancer cell lines selected have likely lost the relevant components of the tumor microenvironment [301]. Although, certain PC cell lines can represent individual subgroups of patients thus are clinically signifi- cant, the interpretation of results should be under curtailment. Secondly, the un- stable aspect between PC cell passages could cause unnecessary variations in the studies. �ese limitations drive the studies to focus on patient-derived xenografts (PDX). However, the PDX sample acquiring difficulty is an obvious obstacle. PDX is still restrained in the tumorigenesis studies, because PC cells have al- ready modified their AR and other steroidogenesis pathways when derived from CRPC patients [302, 303]. Immune deficient male mice with human PC cells have been the primary setting for in vivo used in PC research. It is efficient and by surgically castrating male mice, it is possible to mimic the castrated human environment. Besides the lack of immune T cells, several concerns still need to be raised for future studies in the present topics. �e mouse steroidogenesis is different from human, variations 64 5 . DISCUSSION AND FUTURE PERSPECTIVE of serum T levels between ages and individuals are much larger in mice than in human. Due to the lack of production and secretion of adrenal androgens, cas- trated adult mice keep a steadily low level of serum T (2 to 20 pg/ml), which are significantly lower than in human ADT patients, almost the same level as abi- raterone treated CRPC patients [304-306]. �is could cause different reactions from human PC cells. As to the considerations of intratibial model, the lack of metastatic steps may be the most obvious flaw. By injecting the PC cells directly into the tibiae, the impact of the existing bone marrow condition for the meta- static process and the effect of the injection itself usually are neglected. MRI, “eye opener” for in vivo studies One problem with the intratibial model has been the difficulty in monitoring tu- mor growth, which can easily be performed in the subcutaneous model. �e end- point gives only the final result, and tells nothing about the way the local devel- opment. �ere are some different conditions to be met to achieve the desired information. �e first is the possibility of continuous monitoring and data generating based on the different time points following the tumor progression. In addition, low level of harm and distress are desired, since the nude mice are sensitive to environ- ment change and stimulus, especially tumor bearing mice. Providing good visu- alization and quantitative ability within the not easily accessible intratibial location are critical. Last but not the least; the method should put minimum in- terference with the tumor progression and mice health conditions, since the ob- truding factors could impair the conclusion of any observed effects. Computed tomography (CT) was one of the options, however the required exog- enous contrast agents could influence the tumor activities [307]. In its clinical use, MRI has showed great potential in diagnose and manage prostate cancer with increasing accuracy, depleted over-diagnosis and treatment and reduced pain for patients [308]. Our group have previously used MRI for examination of orthotopic PC xenografts [225], where MRI displayed great advantages in visual- izing soft tissues like tumors. In paper 3, we successfully demonstrated the accu- racy and beneficial utility for analysing bone metastases in mice, which further prove MRI as a profitable tool for in vivo tumor researchers especially for inac- cessible locations. 6 . CONCLUSIONS 65 6. Conclusions • OCs influence CRPC cell proliferation and apoptosis in vitro. • OCs alter gene expression in CRPC cells in a way that may render them to be more resistant to external stress. • OCs can influence the role of RUNX2 in steroidogenesis in CRPC cells. • RUNX2 promotes expression of steroidogenic enzymes and AR signal- ling in the AR-positive CRPC model LNCaP-19. • Depletion of RUNX2 suppresses the AR signalling pathway and in- tratibial growth of LNCaP-19. • MRI is a useful tool for monitoring intratibial tumor growth in mice. 66 ACKNOWLEDGEMENT Acknowledgement During my PhD years, I am very lucky to receive countless help and support from my supervisors, colleagues, friends and family. Here, I would like to ex- press my gratitude in my humble words. My supervisor Karin Welen, thank you for the PhD opportunity in your super- vision, it is a truly valuable experience for me. I am grateful for the freedom to my research and numerous discussions we had regarding the obstacles I encoun- tered and directions leading the study. �anks for the patience and advices on my manuscripts, my thesis, I would not progress so much without your help. Jan-Erik Damber, my co-supervisor. �ank you for the trust and support in my PhD years, they are very important for me to be able to finish my studies. Your knowledge and feedbacks in clinical perspective broadened my horizon and that is something I will keep benefiting from in the future. �anks to all the previous and current members of Damber group. Karin Larsson, my “lab mom”. Your encouraging words and cheers always pull me out of low spirits, and thanks for always being here listening to me. �ank you for your instructions in the IHC, I really learned a lot. Anna Linder, and Malin Hagberg �ulin, thanks for all the scientific discus- sions and sparkles that assisted my PhD project. �ank you for being inspiring for me when I needed it. Malin Hagberg �ulin, Tajana Tesan Tomic, Helene Gustavsson and Karin Jennbacken for laying the foundation for my researches. Daniel Åhs for the company and help in and out of the lab. I would also like to acknowledge Anna Nordin, Åsa Jellvert, Andreas Josefsson and all the members in department of urology for the assistance. �anks to all the collaborators in my PhD project. �anks for Mikael Montelius in department of radiology for the assistance and scientific contribution in the MRI experiments and result analysis. Eva Freyhult from Uppsala University for the assistance and help in the bioinformatics. Prof. Claes Ohlsson, Prof. Matti Poutanen and Andreas Landin from Centre for Bone and Arthritis Research for the assistance in the serum steroid measurement. ACKNOWLEDGEMENT 67 I would also thank staff in Experimental Biomedicine (EBM) for the help in ani- mal maintenance. Special thanks to Prof. Eva Forssell-Aronsson and associate Prof. Khalil Helou for help me during my difficult time. �anks to Prof. Jonas Hugosson to give me the opportunity to finish my PhD in Department of Urology. �anks to the funding supports for the project, �e Swedish Cancer Society; �e Swedish Prostate Cancer Federation; �e Swedish state under the agreement be- tween the Swedish government and the county councils, the ALF-agreement; Assar Gabrielssons Foundation and Ulla och Karl-Erik Winbergs Foundation. My deep appreciation to all my former and present colleagues and friends in SCCR for the help in and out of science aspects. Delila, Elin, Sandra, Gustav, Andreas. I do not know how you people could put up with me, but thanks for all the listening, lunches, fika and all the wonderful time together, thanks for turning grumpy into happy. Amanda and Sara, thanks for the company and cheering me up. Manuel, my desk mate, thanks for the help with the bioinformatics and al- lowing me to occupy your screen. Toshima, Dorota, Agnieszka, Agota, Jana and Gautam, thanks for the company since the start of my time here. �erese and Shawn for the time I am on 4th floor. My friends Dusan, Elias, Philip and Anders for being there not only because of charming personality; p) but also for my up and downs in life. To my family, 谢谢我的爸爸妈妈和姐姐，家庭是我永远的港湾，谢谢你们作为我最强大的 后盾和支持者，没有你们就没有今天的我。 68 REFERENCES References 1. Fine, S.W. and V.E. Reuter, Anatomy of the prostate revisited: implications for prostate biopsy and zonal origins of prostate cancer. Histopathology, 2012. 60(1): p. 142-52. 2. Rosenthal, M., Human sexuality: From cells to society. 2012: Cengage Learning. 3. Lilja, H. and P.A. Abrahamsson, Three predominant proteins secreted by the human prostate gland. The Prostate, 1988. 12(1): p. 29-38. 4. Wang, T.J., et al., PSA concentrations in seminal plasma. Clinical chemistry, 1998. 44(4): p. 895a-896. 5. McNeal, J.E., Regional morphology and pathology of the prostate. Am J Clin Pathol, 1968. 49(3): p. 347-57. 6. Mcneal, J.E., The prostate and prostatic urethra: a morphologic synthesis. The Journal of urology, 1972. 107(6): p. 1008-1016. 7. McNeal, J.E., The zonal anatomy of the prostate. Prostate, 1981. 2(1): p. 35-49. 8. McNeal, J.E., Origin and evolution of benign prostatic enlargement. Investigative urology, 1978. 15(4): p. 340-345. 9. Wong, L. and W. Bushman, Comment on basal epithelial stem cells as efficient targets for prostate cancer initiation. Stem Cell Res Ther, 2010. 1(2): p. 16. 10. Lawson, D.A., et al., Basal epithelial stem cells are efficient targets for prostate cancer initiation. Proc Natl Acad Sci U S A, 2010. 107(6): p. 2610-5. 11. Hudson, D.L., et al., Epithelial cell differentiation pathways in the human prostate: identification of intermediate phenotypes by keratin expression. Journal of Histochemistry & Cytochemistry, 2001. 49(2): p. 271-278. 12. Toivanen, R. and M.M. Shen, Prostate organogenesis: tissue induction, hormonal regulation and cell type specification. Development, 2017. 144(8): p. 1382-1398. 13. Hudson, D.L., Epithelial stem cells in human prostate growth and disease. Prostate Cancer Prostatic Dis, 2004. 7(3): p. 188-94. 14. Shen, M.M. and C. Abate-Shen, Molecular genetics of prostate cancer: new prospects for old challenges. Genes Dev, 2010. 24(18): p. 1967- 2000. 15. Singh, M., et al., Stromal androgen receptor in prostate development and cancer. Am J Pathol, 2014. 184(10): p. 2598-607. REFERENCES 69 16. Tan, M.H., et al., Androgen receptor: structure, role in prostate cancer and drug discovery. Acta Pharmacol Sin, 2015. 36(1): p. 3-23. 17. Shang, Y., M. Myers, and M. Brown, Formation of the androgen receptor transcription complex. Molecular cell, 2002. 9(3): p. 601-610. 18. McEwan, I.J., Molecular mechanisms of androgen receptor-mediated gene regulation: structure-function analysis of the AF-1 domain. Endocrine-related cancer, 2004. 11(2): p. 281-293. 19. Claessens, F. and T. Visakorpi, Endocrinology of prostate cancer. Mol Cell Endocrinol, 2018. 462(Pt A): p. 1-2. 20. Griffiths, K., M.E. Harper, and W.B. Peeling, Some aspects of the biology and endocrinology of prostate cancer. Scand J Clin Lab Invest Suppl, 1995. 221: p. 23-31. 21. Kelly, D.M. and T.H. Jones, Testosterone: a metabolic hormone in health and disease. J Endocrinol, 2013. 217(3): p. R25-45. 22. Patt, M., et al., Profiling of anabolic androgenic steroids and selective androgen receptor modulators for interference with adrenal steroidogenesis. Biochem Pharmacol, 2020. 172: p. 113781. 23. Wilson, E. and F. French, Binding properties of androgen receptors. Evidence for identical receptors in rat testis, epididymis, and prostate. Journal of Biological Chemistry, 1976. 251(18): p. 5620-5629. 24. Lynn, W.S. and R.H. Brown, The conversion of progesterone to androgens by testes. Journal of Biological Chemistry, 1958. 232(2): p. 1015-1030. 25. Auchus, R.J., The backdoor pathway to dihydrotestosterone. Trends in Endocrinology & Metabolism, 2004. 15(9): p. 432-438. 26. Marker, P.C., et al., Hormonal, cellular, and molecular control of prostatic development. Developmental biology, 2003. 253(2): p. 165- 174. 27. Cunha, G.R., et al., Hormonal, cellular, and molecular regulation of normal and neoplastic prostatic development. The Journal of steroid biochemistry and molecular biology, 2004. 92(4): p. 221-236. 28. Donjacour, A.A. and G.R. Cunha, Induction of prostatic morphology and secretion in urothelium by seminal vesicle mesenchyme. Development, 1995. 121(7): p. 2199-2207. 29. Cunha, G.R., et al., The endocrinology and developmental biology of the prostate. Endocrine reviews, 1987. 8(3): p. 338-362. 30. Buttyan, R., et al., Regulation of apoptosis in the prostate gland by androgenic steroids. Trends in endocrinology & Metabolism, 1999. 10(2): p. 47-54. 31. Dehm, S.M. and D.J. Tindall, Regulation of androgen receptor signaling in prostate cancer. Expert review of anticancer therapy, 2005. 5(1): p. 63-74. 32. Zhu, M.L. and N. Kyprianou, Androgen receptor and growth factor signaling cross-talk in prostate cancer cells. Endocr Relat Cancer, 2008. 15(4): p. 841-9. 70 REFERENCES 33. Wu, Y., et al., Identification of androgen response elements in the insulin-like growth factor I upstream promoter. Endocrinology, 2007. 148(6): p. 2984-93. 34. Lonergan, P.E. and D.J. Tindall, Androgen receptor signaling in prostate cancer development and progression. J Carcinog, 2011. 10: p. 20. 35. Gronberg, H., Prostate cancer epidemiology. Lancet, 2003. 361(9360): p. 859-64. 36. Bray, F., et al., Prostate cancer incidence and mortality trends in 37 European countries: an overview. Eur J Cancer, 2010. 46(17): p. 3040- 52. 37. Cazzaniga, W., et al., Mini Review on the Use of Clinical Cancer Registers for Prostate Cancer: The National Prostate Cancer Register (NPCR) of Sweden. Front Med (Lausanne), 2019. 6: p. 51. 38. Zhou, C.K., et al., Prostate cancer incidence in 43 populations worldwide: an analysis of time trends overall and by age group. International journal of cancer, 2016. 138(6): p. 1388-1400. 39. Cackowski, F.C. and E.I. Heath, Prostate cancer dormancy and recurrence. Cancer Lett, 2021. 40. Puhr, M., et al., Inflammation, Microbiota, and Prostate Cancer. Eur Urol Focus, 2016. 2(4): p. 374-382. 41. Torre, L.A., et al., Global Cancer Incidence and Mortality Rates and Trends--An Update. Cancer Epidemiol Biomarkers Prev, 2016. 25(1): p. 16-27. 42. Di Lorenzo, G., et al., Expression of epidermal growth factor receptor correlates with disease relapse and progression to androgen- independence in human prostate cancer. Clinical Cancer Research, 2002. 8(11): p. 3438-3444. 43. Kaushal, V., et al., Stage-specific characterization of the vascular endothelial growth factor axis in prostate cancer: expression of lymphangiogenic markers is associated with advanced-stage disease. Clinical cancer research, 2005. 11(2): p. 584-593. 44. Filella, X. and L. Foj, Novel biomarkers for prostate cancer detection and prognosis. Cell & Molecular Biology of Prostate Cancer, 2018: p. 15-39. 45. Mikolajczyk, S.D., et al., Proenzyme forms of prostate-specific antigen in serum improve the detection of prostate cancer. Clin Chem, 2004. 50(6): p. 1017-25. 46. Vickers, A.J., et al., A four-kallikrein panel predicts prostate cancer in men with recent screening: data from the European Randomized Study of Screening for Prostate Cancer, Rotterdam. Clinical Cancer Research, 2010. 16(12): p. 3232-3239. 47. Schatten, H., Brief Overview of Prostate Cancer Statistics, Grading, Diagnosis and Treatment Strategies, in Cell & Molecular Biology of Prostate Cancer: Updates, Insights and New Frontiers, H. Schatten, Editor. 2018, Springer International Publishing: Cham. p. 1-14. REFERENCES 71 48. Hambrock, T., et al., Prospective assessment of prostate cancer aggressiveness using 3-T diffusion-weighted magnetic resonance imaging–guided biopsies versus a systematic 10-core transrectal ultrasound prostate biopsy cohort. European urology, 2012. 61(1): p. 177-184. 49. Catalá, V., et al., Questions and answers on prostate multiparameter magnetic resonance imaging: Everything a urologist should know. Actas Urológicas Españolas (English Edition), 2016. 40(6): p. 339-352. 50. Mohler, J., et al., Prostate cancer. Journal of the National Comprehensive Cancer Network, 2010. 8(2): p. 162-200. 51. Anastasiadis, A.G., et al., MRI-Guided Biopsy of the Prostate Increases Diagnostic Performance in Men with Elevated or Increasing PSA Levels after Previous Negative TRUS Biopsies. European Urology, 2006. 50(4): p. 738-749. 52. Schatten, H., Brief Overview of Prostate Cancer Statistics, Grading, Diagnosis and Treatment Strategies. Adv Exp Med Biol, 2018. 1095: p. 1-14. 53. Loeb, S., et al., Evaluation of the 2015 Gleason Grade Groups in a Nationwide Population-based Cohort. Eur Urol, 2016. 69(6): p. 1135- 41. 54. Giannico, G.A. and O. Hameed, Evaluation of prostate needle biopsies. Molecular & Diagnostic Imaging in Prostate Cancer, 2018: p. 69-86. 55. Cooper, C.S., et al., Analysis of the genetic phylogeny of multifocal prostate cancer identifies multiple independent clonal expansions in neoplastic and morphologically normal prostate tissue. Nature genetics, 2015. 47(4): p. 367-372. 56. Knudsen, B.S. and V. Vasioukhin, Mechanisms of prostate cancer initiation and progression. Adv Cancer Res, 2010. 109: p. 1-50. 57. Zhou, M., High-grade prostatic intraepithelial neoplasia, PIN-like carcinoma, ductal carcinoma, and intraductal carcinoma of the prostate. Mod Pathol, 2018. 31(S1): p. S71-79. 58. Nakai, Y. and N. Nonomura, Inflammation and prostate carcinogenesis. Int J Urol, 2013. 20(2): p. 150-60. 59. Omabe, M. and M. Ezeani, Infection, inflammation and prostate carcinogenesis. Infect Genet Evol, 2011. 11(6): p. 1195-8. 60. Lepor, H., Evaluating men with benign prostatic hyperplasia. Rev Urol, 2004. 6 Suppl 1: p. S8-S15. 61. McNeal, J.E., et al., Zonal distribution of prostatic adenocarcinoma. Correlation with histologic pattern and direction of spread. The American journal of surgical pathology, 1988. 12(12): p. 897-906. 62. Shiao, S.L., G.C. Chu, and L.W. Chung, Regulation of prostate cancer progression by the tumor microenvironment. Cancer Lett, 2016. 380(1): p. 340-8. 63. Culig, Z. and F.R. Santer, Androgen receptor signaling in prostate cancer. Cancer Metastasis Rev, 2014. 33(2-3): p. 413-27. 72 REFERENCES 64. Tomlins, S.A., et al., Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. science, 2005. 310(5748): p. 644-648. 65. Lin, B., et al., Prostate-localized and androgen-regulated expression of the membrane-bound serine protease TMPRSS2. Cancer research, 1999. 59(17): p. 4180-4184. 66. Rajput, A.B., et al., Frequency of the TMPRSS2: ERG gene fusion is increased in moderate to poorly differentiated prostate cancers. Journal of clinical pathology, 2007. 60(11): p. 1238-1243. 67. Byrne, R., H. Leung, and D. Neal, Peptide growth factors in the prostate as mediators of stromal epithelial interaction. British journal of urology, 1996. 77(5): p. 627-633. 68. Kaaks, R., A. Lukanova, and B. Sommersberg, Plasma androgens, IGF-1, body size, and prostate cancer risk: a synthetic review. Prostate cancer and prostatic diseases, 2000. 3(3): p. 157-172. 69. Liao, X., et al., Androgen stimulates matrix metalloproteinase-2 expression in human prostate cancer. Endocrinology, 2003. 144(5): p. 1656-1663. 70. Häggström, S., et al., Testosterone induces vascular endothelial growth factor synthesis in the ventral prostate in castrated rats. J Urol, 1999. 161(5): p. 1620-5. 71. Wu, Y., et al., Identification of androgen response elements in the insulin-like growth factor I upstream promoter. Endocrinology, 2007. 148(6): p. 2984-2993. 72. Adler, H.L., et al., ELEVATED LEVELS OF CIRCULATING INTERLEUKIN-6 AND TRANSFORMING GROWTH FACTOR-beta 1 IN PATIENTS WITH METASTATIC PROSTATIC CARCINOMA. The Journal of Urology, 1999. 161(1): p. 182-187. 73. van der Poel, H.G., Androgen receptor and TGFbeta1/Smad signaling are mutually inhibitory in prostate cancer. Eur Urol, 2005. 48(6): p. 1051-8. 74. Gandhi, J., et al., The molecular biology of prostate cancer: current understanding and clinical implications. Prostate Cancer Prostatic Dis, 2018. 21(1): p. 22-36. 75. Litwin, M.S. and H.J. Tan, The Diagnosis and Treatment of Prostate Cancer: A Review. JAMA, 2017. 317(24): p. 2532-2542. 76. Chang, A.J., et al., High-risk prostate cancer-classification and therapy. Nat Rev Clin Oncol, 2014. 11(6): p. 308-23. 77. Westin, P., et al., Castration therapy rapidly induces apoptosis in a minority and decreases cell proliferation in a majority of human prostatic tumors. The American journal of pathology, 1995. 146(6): p. 1368. 78. Mansinho, A., et al., Castration-Resistant Prostate Cancer: Mechanisms, Targets and Treatment. Adv Exp Med Biol, 2018. 1096: p. 117-133. REFERENCES 73 79. Smith, M.R., et al., Natural history of rising serum prostate-specific antigen in men with castrate nonmetastatic prostate cancer. Journal of Clinical Oncology, 2005. 23(13): p. 2918-2925. 80. Watson, P.A., V.K. Arora, and C.L. Sawyers, Emerging mechanisms of resistance to androgen receptor inhibitors in prostate cancer. Nat Rev Cancer, 2015. 15(12): p. 701-11. 81. Harris, W.P., et al., Androgen deprivation therapy: progress in understanding mechanisms of resistance and optimizing androgen depletion. Nature clinical practice Urology, 2009. 6(2): p. 76-85. 82. Taplin, M.-E., et al., Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer. New England Journal of Medicine, 1995. 332(21): p. 1393-1398. 83. Han, G., et al., Mutation of the androgen receptor causes oncogenic transformation of the prostate. Proceedings of the National Academy of Sciences of the United States of America, 2005. 102(4): p. 1151-1156. 84. Lapouge, G., et al., Unexpected paracrine action of prostate cancer cells harboring a new class of androgen receptor mutation—A new paradigm for cooperation among prostate tumor cells. International journal of cancer, 2007. 121(6): p. 1238-1244. 85. Hörnberg, E., et al., Expression of androgen receptor splice variants in prostate cancer bone metastases is associated with castration- resistance and short survival. PloS one, 2011. 6(4): p. e19059. 86. Dehm, S.M. and D.J. Tindall, Alternatively spliced androgen receptor variants. Endocrine-related cancer, 2011. 18(5): p. R183. 87. Antonarakis, E.S., et al., AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. New England Journal of Medicine, 2014. 371(11): p. 1028-1038. 88. Antonarakis, E.S., et al., Androgen receptor splice variant 7 and efficacy of taxane chemotherapy in patients with metastatic castration- resistant prostate cancer. JAMA oncology, 2015. 1(5): p. 582-591. 89. Montgomery, R.B., et al., Maintenance of intratumoral androgens in metastatic prostate cancer: a mechanism for castration-resistant tumor growth. Cancer research, 2008. 68(11): p. 4447-4454. 90. Chandrasekar, T., et al., Mechanisms of resistance in castration- resistant prostate cancer (CRPC). Translational andrology and urology, 2015. 4(3): p. 365. 91. Kumagai, J., et al., Intratumoral conversion of adrenal androgen precursors drives androgen receptor‐activated cell growth in prostate cancer more potently than de novo steroidogenesis. The Prostate, 2013. 73(15): p. 1636-1650. 92. 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 research, 1994. 54(20): p. 5474-5478. 74 REFERENCES 93. Isikbay, M., et al., Glucocorticoid receptor activity contributes to resistance to androgen-targeted therapy in prostate cancer. Hormones and Cancer, 2014. 5(2): p. 72-89. 94. Kumar, R., Emerging role of glucocorticoid receptor in castration resistant prostate cancer: A potential therapeutic target. Journal of Cancer, 2020. 11(3): p. 696. 95. Hoshi, S., et al., Upregulation of glucocorticoid receptor‐mediated glucose transporter 4 in enzalutamide‐resistant prostate cancer. Cancer science, 2021. 112(5): p. 1899. 96. Madden, E., et al., The role of the unfolded protein response in cancer progression: From oncogenesis to chemoresistance. Biol Cell, 2019. 111(1): p. 1-17. 97. Guo, J., et al., PERK controls bone homeostasis through the regulation of osteoclast differentiation and function. Cell Death Dis, 2020. 11(10): p. 847. 98. Sheng, X., et al., IRE1alpha-XBP1s pathway promotes prostate cancer by activating c-MYC signaling. Nat Commun, 2019. 10(1): p. 323. 99. Hiramatsu, N., et al., Multiple Mechanisms of Unfolded Protein Response-Induced Cell Death. Am J Pathol, 2015. 185(7): p. 1800-8. 100. Lin, J.H., et al., Divergent effects of PERK and IRE1 signaling on cell viability. PloS one, 2009. 4(1): p. e4170. 101. Sovolyova, N., et al., Stressed to death–mechanisms of ER stress- induced cell death. Biological chemistry, 2014. 395(1): p. 1-13. 102. Ariyasu, D., H. Yoshida, and Y. Hasegawa, Endoplasmic reticulum (ER) stress and endocrine disorders. International journal of molecular sciences, 2017. 18(2): p. 382. 103. Scriven, P., et al., Activation and clinical significance of the unfolded protein response in breast cancer. British journal of cancer, 2009. 101(10): p. 1692-1698. 104. Pelzer, A.E., et al., The expression of transcription factor activating transcription factor 3 in the human prostate and its regulation by androgen in prostate cancer. The Journal of urology, 2006. 175(4): p. 1517-1522. 105. Matt, S. and T.G. Hofmann, The DNA damage-induced cell death response: a roadmap to kill cancer cells. Cell Mol Life Sci, 2016. 73(15): p. 2829-50. 106. Basu, A.K., DNA Damage, Mutagenesis and Cancer. International Journal of Molecular Sciences, 2018. 19(4): p. 970. 107. Roos, W.P., A.D. Thomas, and B. Kaina, DNA damage and the balance between survival and death in cancer biology. Nat Rev Cancer, 2016. 16(1): p. 20-33. 108. Liang, Y., et al., DNA damage response pathways in tumor suppression and cancer treatment. World journal of surgery, 2009. 33(4): p. 661- 666. REFERENCES 75 109. O'Connor, M.J., Targeting the DNA Damage Response in Cancer. Mol Cell, 2015. 60(4): p. 547-60. 110. Burdak-Rothkamm, S., W.Y. Mansour, and K. Rothkamm, DNA Damage Repair Deficiency in Prostate Cancer. Trends Cancer, 2020. 6(11): p. 974-984. 111. Robinson, D., et al., Integrative clinical genomics of advanced prostate cancer. Cell, 2015. 161(5): p. 1215-1228. 112. Ira, G., et al., DNA end resection, homologous recombination and DNA damage checkpoint activation require CDK1. Nature, 2004. 431(7011): p. 1011-1017. 113. Sztupinszki, Z., et al., Detection of Molecular Signatures of Homologous Recombination Deficiency in Prostate Cancer with or without BRCA1/2 Mutations. Clin Cancer Res, 2020. 26(11): p. 2673- 2680. 114. Pritchard, C.C., et al., Inherited DNA-Repair Gene Mutations in Men with Metastatic Prostate Cancer. N Engl J Med, 2016. 375(5): p. 443- 53. 115. Ju, B.G., et al., A topoisomerase IIbeta-mediated dsDNA break required for regulated transcription. Science, 2006. 312(5781): p. 1798-802. 116. Tomlins, S.A., et al., Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science, 2005. 310(5748): p. 644-8. 117. Asim, M., et al., Synthetic lethality between androgen receptor signalling and the PARP pathway in prostate cancer. Nat Commun, 2017. 8(1): p. 374. 118. Brenner, J.C., et al., Mechanistic rationale for inhibition of poly(ADP- ribose) polymerase in ETS gene fusion-positive prostate cancer. Cancer Cell, 2011. 19(5): p. 664-78. 119. Schiewer, M.J., et al., Dual roles of PARP-1 promote cancer growth and progression. Cancer Discov, 2012. 2(12): p. 1134-49. 120. Evans, A.J., Treatment effects in prostate cancer. Mod Pathol, 2018. 31(S1): p. S110-121. 121. Francini, E. and M.E. Taplin, Prostate cancer: Developing novel approaches to castration‐sensitive disease. Cancer, 2017. 123(1): p. 29-42. 122. Bostwick, D.G. and I. Meiers, Diagnosis of prostatic carcinoma after therapy: Review 2007. Archives of pathology & laboratory medicine, 2007. 131(3): p. 360-371. 123. Antonarakis, E.S., et al., Randomized, noncomparative, phase II trial of early switch from docetaxel to cabazitaxel or vice versa, with integrated biomarker analysis, in men with chemotherapy-naive, metastatic, castration-resistant prostate cancer. Journal of Clinical Oncology, 2017. 35(28): p. 3181. 76 REFERENCES 124. Tran, C., et al., Development of a second-generation antiandrogen for treatment of advanced prostate cancer. Science, 2009. 324(5928): p. 787-790. 125. De Bono, J.S., et al., Abiraterone and increased survival in metastatic prostate cancer. New England Journal of Medicine, 2011. 364(21): p. 1995-2005. 126. Cortes, J.E. and R. Pazdur, Docetaxel. Journal of Clinical Oncology, 1995. 13(10): p. 2643-2655. 127. Gravis, G., et al., Androgen-deprivation therapy alone or with docetaxel in non-castrate metastatic prostate cancer (GETUG-AFU 15): a randomised, open-label, phase 3 trial. Lancet Oncol, 2013. 14(2): p. 149-58. 128. de Bono, J.S., et al., Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet, 2010. 376(9747): p. 1147-54. 129. Sartor, O., et al., Effect of radium-223 dichloride on symptomatic skeletal events in patients with castration-resistant prostate cancer and bone metastases: results from a phase 3, double-blind, randomised trial. The Lancet Oncology, 2014. 15(7): p. 738-746. 130. Nilsson, S., et al., A randomized, dose–response, multicenter phase II study of radium-223 chloride for the palliation of painful bone metastases in patients with castration-resistant prostate cancer. European journal of cancer, 2012. 48(5): p. 678-686. 131. Fleisch, H., et al., Bisphosphonates: mechanisms of action. Principles of bone biology, 2002: p. 1361-XLIII. 132. Baron, R., S. Ferrari, and R.G. Russell, Denosumab and bisphosphonates: different mechanisms of action and effects. Bone, 2011. 48(4): p. 677-92. 133. Leung, A.K., Optimizing skeletal-related events prevention in patients with advanced prostate cancer. Asia Pac J Clin Oncol, 2020. 16 Suppl 3: p. 4-6. 134. Datta, H.K., et al., The cell biology of bone metabolism. J Clin Pathol, 2008. 61(5): p. 577-87. 135. Salhotra, A., et al., Mechanisms of bone development and repair. Nat Rev Mol Cell Biol, 2020. 21(11): p. 696-711. 136. Burger, E.H., J.W. van der Meer, and P.J. Nijweide, Osteoclast formation from mononuclear phagocytes: role of bone-forming cells. J Cell Biol, 1984. 99(6): p. 1901-6. 137. Kearns, A.E., S. Khosla, and P.J. Kostenuik, Receptor activator of nuclear factor kappaB ligand and osteoprotegerin regulation of bone remodeling in health and disease. Endocr Rev, 2008. 29(2): p. 155-92. 138. Xiong, J., et al., Matrix-embedded cells control osteoclast formation. Nat Med, 2011. 17(10): p. 1235-41. REFERENCES 77 139. Kawai, T., et al., B and T lymphocytes are the primary sources of RANKL in the bone resorptive lesion of periodontal disease. Am J Pathol, 2006. 169(3): p. 987-98. 140. Weilbaecher, K.N., T.A. Guise, and L.K. McCauley, Cancer to bone: a fatal attraction. Nat Rev Cancer, 2011. 11(6): p. 411-25. 141. Martin, T.J., Parathyroid Hormone-Related Protein, Its Regulation of Cartilage and Bone Development, and Role in Treating Bone Diseases. Physiol Rev, 2016. 96(3): p. 831-71. 142. Burgess, T.L., et al., The ligand for osteoprotegerin (OPGL) directly activates mature osteoclasts. The Journal of cell biology, 1999. 145(3): p. 527-538. 143. Fuller, K., et al., TNFα potently activates osteoclasts, through a direct action independent of and strongly synergistic with RANKL. Endocrinology, 2002. 143(3): p. 1108-1118. 144. Zaidi, M., Skeletal remodeling in health and disease. Nat Med, 2007. 13(7): p. 791-801. 145. Soltanoff, C.S., et al., Signaling networks that control the lineage commitment and differentiation of bone cells. Crit Rev Eukaryot Gene Expr, 2009. 19(1): p. 1-46. 146. Stein, G.S., et al., Transcriptional control of osteoblast growth and differentiation. Physiol Rev, 1996. 76(2): p. 593-629. 147. Rubin, J., C. Rubin, and C.R. Jacobs, Molecular pathways mediating mechanical signaling in bone. Gene, 2006. 367: p. 1-16. 148. Berli, M., et al., Localized tissue mineralization regulated by bone remodelling: A computational approach. PLoS One, 2017. 12(3): p. e0173228. 149. Feng, X. and S.L. Teitelbaum, Osteoclasts: New Insights. Bone Res, 2013. 1(1): p. 11-26. 150. Raggatt, L.J. and N.C. Partridge, Cellular and molecular mechanisms of bone remodeling. Journal of Biological Chemistry, 2010. 285(33): p. 25103-25108. 151. Keller, E.T., et al., Prostate carcinoma skeletal metastases: cross-talk between tumor and bone. Cancer and Metastasis Reviews, 2001. 20(3): p. 333-349. 152. Horstman, A.M., et al., The role of androgens and estrogens on healthy aging and longevity. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 2012. 67(11): p. 1140-1152. 153. Martin, A., et al., Estrogens and androgens inhibit association of RANKL with the pre-osteoblast membrane through post-translational mechanisms. J Cell Physiol, 2017. 232(12): p. 3798-3807. 154. Bord, S., et al., The effects of estrogen on osteoprotegerin, RANKL, and estrogen receptor expression in human osteoblasts. Bone, 2003. 32(2): p. 136-41. 155. Eastell, R., Role of oestrogen in the regulation of bone turnover at the menarche. J Endocrinol, 2005. 185(2): p. 223-34. 78 REFERENCES 156. Wiren, K.M., A. Chapman Evans, and X.W. Zhang, Osteoblast differentiation influences androgen and estrogen receptor-alpha and - beta expression. J Endocrinol, 2002. 175(3): p. 683-94. 157. Abu, E., et al., The localization of androgen receptors in human bone. The Journal of Clinical Endocrinology & Metabolism, 1997. 82(10): p. 3493-3497. 158. Kasperk, C.H., et al., Gonadal and adrenal androgens are potent regulators of human bone cell metabolism in vitro. J Bone Miner Res, 1997. 12(3): p. 464-71. 159. Kasperk, C.H., et al., Androgens directly stimulate proliferation of bone cells in vitro. Endocrinology, 1989. 124(3): p. 1576-1578. 160. Kousteni, S., et al., Nongenotropic, Sex-Nonspecific Signaling through the Estrogen or Androgen Receptors: Dissociation from Transcriptional Activity. Cell, 2001. 104(5): p. 719-730. 161. Notelovitz, M., Androgen effects on bone and muscle. Fertility and sterility, 2002. 77: p. 34-41. 162. Carlin, B.I. and G.L. Andriole, The natural history, skeletal complications, and management of bone metastases in patients with prostate carcinoma. Cancer: Interdisciplinary International Journal of the American Cancer Society, 2000. 88(S12): p. 2989-2994. 163. Jimenez-Andrade, J.M., et al., Pathological Sprouting of Adult Nociceptors in Chronic Prostate Cancer-Induced Bone Pain. The Journal of Neuroscience, 2010. 30(44): p. 14649-14656. 164. Adami, S., Bisphosphonates in prostate carcinoma. Cancer: Interdisciplinary International Journal of the American Cancer Society, 1997. 80(S8): p. 1674-1679. 165. Orr, F.W., et al., Pathophysiologic interactions in skeletal metastasis. Cancer: Interdisciplinary International Journal of the American Cancer Society, 2000. 88(S12): p. 2912-2918. 166. Roudier, M.P., et al., Phenotypic heterogeneity of end-stage prostate carcinoma metastatic to bone. Human pathology, 2003. 34(7): p. 646- 653. 167. Pantel, K. and M.R. Speicher, The biology of circulating tumor cells. Oncogene, 2016. 35(10): p. 1216-24. 168. Reya, T., et al., Stem cells, cancer, and cancer stem cells. Nature, 2001. 414(6859): p. 105-11. 169. Kelly, K. and J.J. Yin, Prostate cancer and metastasis initiating stem cells. Cell Res, 2008. 18(5): p. 528-37. 170. DiNatale, A. and A. Fatatis, The Bone Microenvironment in Prostate Cancer Metastasis. Adv Exp Med Biol, 2019. 1210: p. 171-184. 171. Mundy, G.R., Mechanisms of bone metastasis. Cancer, 1997. 80(8 Suppl): p. 1546-56. 172. Peinado, H., et al., Melanoma exosomes educate bone marrow progenitor cells toward a pro-metastatic phenotype through MET. Nature medicine, 2012. 18(6): p. 883-891. REFERENCES 79 173. Nelson, J.B., et al., New bone formation in an osteoblastic tumor model is increased by endothelin-1 overexpression and decreased by endothelin A receptor blockade. Urology, 1999. 53(5): p. 1063-1069. 174. Cramer, S.D., Z. Chen, and D.M. Peehl, Prostate specific antigen cleaves parathyroid hormone-related protein in the PTH-like domain: inactivation of PTHrP-stimulated cAMP accumulation in mouse osteoblasts. The Journal of urology, 1996. 156(2): p. 526-531. 175. Morrissey, C. and R.L. Vessella, The role of tumor microenvironment in prostate cancer bone metastasis. J Cell Biochem, 2007. 101(4): p. 873-86. 176. Chen, G., C. Deng, and Y.-P. Li, TGF-β and BMP signaling in osteoblast differentiation and bone formation. International journal of biological sciences, 2012. 8(2): p. 272. 177. Yang, Q., et al., VEGF enhancement of osteoclast survival and bone resorption involves VEGF receptor-2 signaling and β3-integrin. Matrix Biology, 2008. 27(7): p. 589-599. 178. Montgomery, R.B., et al., Maintenance of intratumoral androgens in metastatic prostate cancer: a mechanism for castration-resistant tumor growth. Cancer Res, 2008. 68(11): p. 4447-54. 179. Hagberg Thulin, M., et al., Osteoblasts promote castration-resistant prostate cancer by altering intratumoral steroidogenesis. Mol Cell Endocrinol, 2016. 422: p. 182-191. 180. Jernberg, E., et al., Characterization of prostate cancer bone metastases according to expression levels of steroidogenic enzymes and androgen receptor splice variants. PLoS One, 2013. 8(11): p. e77407. 181. Hofland, J., et al., Evidence of limited contributions for intratumoral steroidogenesis in prostate cancer. Cancer research, 2010. 70(3): p. 1256-1264. 182. Hamid, A.R.A., et al., Aldo-keto reductase family 1 member C3 (AKR1C3) is a biomarker and therapeutic target for castration- resistant prostate cancer. Molecular Medicine, 2012. 18(11): p. 1449- 1455. 183. Carlinfante, G., et al., Differential expression of osteopontin and bone sialoprotein in bone metastasis of breast and prostate carcinoma. Clinical & experimental metastasis, 2003. 20(5): p. 437-444. 184. Komori, T., Runx2, a multifunctional transcription factor in skeletal development. J Cell Biochem, 2002. 87(1): p. 1-8. 185. Ge, C., et al., Role of Runx2 phosphorylation in prostate cancer and association with metastatic disease. Oncogene, 2016. 35(3): p. 366-376. 186. Li, Y., C. Ge, and R.T. Franceschi, Role of Runx2 in prostate development and stem cell function. Prostate, 2021. 81(4): p. 231-241. 187. Yun, S.J., et al., Transcriptional repression of RUNX2 is associated with aggressive clinicopathological outcomes, whereas nuclear location of the protein is related to metastasis in prostate cancer. Prostate Cancer Prostatic Dis, 2012. 15(4): p. 369-73. 80 REFERENCES 188. Fowler, M., et al., RUNX1 (AML-1) and RUNX2 (AML-3) cooperate with prostate-derived Ets factor to activate transcription from the PSA upstream regulatory region. J Cell Biochem, 2006. 97(1): p. 1-17. 189. Yeung, F., et al., Regulation of human osteocalcin promoter in hormone-independent human prostate cancer cells. Journal of Biological Chemistry, 2002. 277(4): p. 2468-2476. 190. Akech, J., et al., Runx2 association with progression of prostate cancer in patients: mechanisms mediating bone osteolysis and osteoblastic metastatic lesions. Oncogene, 2010. 29(6): p. 811-21. 191. van der Deen, M., et al., The cancer-related Runx2 protein enhances cell growth and responses to androgen and TGFbeta in prostate cancer cells. J Cell Biochem, 2010. 109(4): p. 828-37. 192. Arumugam, B., et al., Characterization of Runx2 phosphorylation sites required for TGF-beta1-mediated stimulation of matrix metalloproteinase-13 expression in osteoblastic cells. J Cell Physiol, 2018. 233(2): p. 1082-1094. 193. Baniwal, S.K., et al., Repression of Runx2 by androgen receptor (AR) in osteoblasts and prostate cancer cells: AR binds Runx2 and abrogates its recruitment to DNA. Mol Endocrinol, 2009. 23(8): p. 1203-14. 194. Kawate, H., et al., Mutual transactivational repression of Runx2 and the androgen receptor by an impairment of their normal compartmentalization. The Journal of Steroid Biochemistry and Molecular Biology, 2007. 105(1): p. 46-56. 195. Little, G.H., et al., Differential effects of RUNX2 on the androgen receptor in prostate cancer: synergistic stimulation of a gene set exemplified by SNAI2 and subsequent invasiveness. Cancer Res, 2014. 74(10): p. 2857-68. 196. Taciak, B., et al., Evaluation of phenotypic and functional stability of RAW 264.7 cell line through serial passages. PLoS One, 2018. 13(6): p. e0198943. 197. Biswas, S.K. and A. Mantovani, Macrophage plasticity and interaction with lymphocyte subsets: cancer as a paradigm. Nat Immunol, 2010. 11(10): p. 889-96. 198. Collin-Osdoby, P. and P. Osdoby, RANKL-mediated osteoclast formation from murine RAW 264.7 cells. Methods Mol Biol, 2012. 816: p. 187-202. 199. Castanares, M.A., et al., Characterization of a novel metastatic prostate cancer cell line of LNCaP origin. The Prostate, 2016. 76(2): p. 215- 225. 200. Jennbacken, K., et al., Prostate cancer progression into androgen independency is associated with alterations in cell adhesion and invasivity. Prostate, 2006. 66(15): p. 1631-40. 201. Gustavsson, H., K. Welen, and J.E. Damber, Transition of an androgen-dependent human prostate cancer cell line into an androgen- independent subline is associated with increased angiogenesis. Prostate, 2005. 62(4): p. 364-73. REFERENCES 81 202. 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. 203. Tai, S., et al., PC3 is a cell line characteristic of prostatic small cell carcinoma. Prostate, 2011. 71(15): p. 1668-79. 204. Schulze, J., et al., Osteolytic prostate cancer cells induce the expression of specific cytokines in bone-forming osteoblasts through a Stat3/5- dependent mechanism. Bone, 2010. 46(2): p. 524-33. 205. Shevrin, D.H., et al., Development of skeletal metastasis by human prostate cancer in athymic nude mice. Clin Exp Metastasis, 1988. 6(5): p. 401-9. 206. Elbashir, S.M., et al., Functional anatomy of siRNAs for mediating efficient RNAi in Drosophila melanogaster embryo lysate. EMBO J, 2001. 20(23): p. 6877-88. 207. Doudna, J.A. and E. Charpentier, Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science, 2014. 346(6213): p. 1258096. 208. Liang, C.C., A.Y. Park, and J.L. Guan, In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat Protoc, 2007. 2(2): p. 329-33. 209. Michiel Sedelaar, J.P., S.S. Dalrymple, and J.T. Isaacs, Of mice and men--warning: intact versus castrated adult male mice as xenograft hosts are equivalent to hypogonadal versus abiraterone treated aging human males, respectively. Prostate, 2013. 73(12): p. 1316-25. 210. Nilsson, M.E., et al., Measurement of a Comprehensive Sex Steroid Profile in Rodent Serum by High-Sensitive Gas Chromatography- Tandem Mass Spectrometry. Endocrinology, 2015. 156(7): p. 2492-502. 211. Schneider, A., et al., Bone turnover mediates preferential localization of prostate cancer in the skeleton. Endocrinology, 2005. 146(4): p. 1727-36. 212. Gorman, A.M., et al., Stress management at the ER: regulators of ER stress-induced apoptosis. Pharmacol Ther, 2012. 134(3): p. 306-16. 213. Edagawa, M., et al., Role of activating transcription factor 3 (ATF3) in endoplasmic reticulum (ER) stress-induced sensitization of p53- deficient human colon cancer cells to tumor necrosis factor (TNF)- related apoptosis-inducing ligand (TRAIL)-mediated apoptosis through up-regulation of death receptor 5 (DR5) by zerumbone and celecoxib. Journal of Biological Chemistry, 2014. 289(31): p. 21544-21561. 214. Li, T., et al., DDIT3 and KAT2A Proteins Regulate TNFRSF10A and TNFRSF10B Expression in Endoplasmic Reticulum Stress-mediated Apoptosis in Human Lung Cancer Cells. J Biol Chem, 2015. 290(17): p. 11108-18. 215. Clauson, C., O.D. Scharer, and L. Niedernhofer, Advances in understanding the complex mechanisms of DNA interstrand cross-link repair. Cold Spring Harb Perspect Biol, 2013. 5(10): p. a012732. 216. Murai, J., et al., Trapping of PARP1 and PARP2 by clinical PARP inhibitors. Cancer research, 2012. 72(21): p. 5588-5599. 82 REFERENCES 217. Jo, U. and H. Kim, Exploiting the Fanconi Anemia Pathway for Targeted Anti-Cancer Therapy. Mol Cells, 2015. 38(8): p. 669-76. 218. Stuparu, A.D., et al., Mechanisms of Resistance to Prostate-Specific Membrane Antigen–Targeted Radioligand Therapy in a Mouse Model of Prostate Cancer. Journal of Nuclear Medicine, 2021. 62(7): p. 989- 995. 219. Hagberg Thulin, M., et al., Osteoblasts stimulate the osteogenic and metastatic progression of castration-resistant prostate cancer in a novel model for in vitro and in vivo studies. Clin Exp Metastasis, 2014. 31(3): p. 269-83. 220. Hagberg Thulin, M., The regulatory role of osteoblasts in castration- resistant growth of prostate cancer, u. Göteborgs, Editor. 2015, Diss. Göteborg : Göteborgs universitet, 2015: Göteborg. 221. Hagberg Thulin, M., et al., Osteoblasts promote castration-resistant prostate cancer by altering intratumoral steroidogenesis. Mol Cell Endocrinol, 2016. 422: p. 182-191. 222. Nordstrand, A., et al., Bone Cell Activity in Clinical Prostate Cancer Bone Metastasis and Its Inverse Relation to Tumor Cell Androgen Receptor Activity. Int J Mol Sci, 2018. 19(4). 223. Baniwal, S.K., et al., Repression of Runx2 by androgen receptor (AR) in osteoblasts and prostate cancer cells: AR binds Runx2 and abrogates its recruitment to DNA. Molecular Endocrinology, 2009. 23(8): p. 1203-1214. 224. Horvath, L.G., et al., Frequent loss of estrogen receptor-β expression in prostate cancer. Cancer research, 2001. 61(14): p. 5331-5335. 225. Jennbacken, K., et al., The prostatic environment suppresses growth of androgen-independent prostate cancer xenografts: an effect influenced by testosterone. Prostate, 2009. 69(11): p. 1164-75. 226. Chimge, N.O., et al., Regulation of breast cancer metastasis by Runx2 and estrogen signaling: the role of SNAI2. Breast Cancer Res, 2011. 13(6): p. R127. 227. Ratz, L., et al., TMPRSS2:ERG gene fusion variants induce TGF-beta signaling and epithelial to mesenchymal transition in human prostate cancer cells. Oncotarget, 2017. 8(15): p. 25115-25130. 228. Zhao, H.-J., et al., Bone morphogenetic protein 2 promotes human trophoblast cell invasion by upregulating N-cadherin via non-canonical SMAD2/3 signaling. Cell death & disease, 2018. 9(2): p. 1-12. 229. Bubendorf, L., et al., Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Hum Pathol, 2000. 31(5): p. 578-83. 230. Roudier, M.P., et al., Phenotypic heterogeneity of end-stage prostate carcinoma metastatic to bone. Hum Pathol, 2003. 34(7): p. 646-53. 231. Abu-Hamar Ael, H. and T.A. Gameel, Prognostic Significance of PSA, Gleason Score, Bone Metastases in Patients with Metastatic Prostate Cancer Under Palliative Androgen Deprivation Treatment. J Egypt Natl Canc Inst, 2009. 21(3): p. 229-36. REFERENCES 83 232. Yamashita, K., et al., Prognostic significance of bone metastases in patients with metastatic prostate cancer. Cancer, 1993. 71(4): p. 1297- 302. 233. Harryman, W.L., et al., The cohesive metastasis phenotype in human prostate cancer. Biochimica et Biophysica Acta (BBA)-Reviews on Cancer, 2016. 1866(2): p. 221-231. 234. Nørgaard, M., et al., Skeletal Related Events, Bone Metastasis and Survival of Prostate Cancer: A Population Based Cohort Study in Denmark (1999 to 2007). The Journal of Urology, 2010. 184(1): p. 162- 167. 235. Iglesias-Gato, D., et al., The Proteome of Prostate Cancer Bone Metastasis Reveals Heterogeneity with Prognostic Implications. Clin Cancer Res, 2018. 24(21): p. 5433-5444. 236. Greenspan, S.L., Approach to the prostate cancer patient with bone disease. J Clin Endocrinol Metab, 2008. 93(1): p. 2-7. 237. Greenspan, S.L., et al., Bone loss after initiation of androgen deprivation therapy in patients with prostate cancer. J Clin Endocrinol Metab, 2005. 90(12): p. 6410-7. 238. Garnero, P., et al., Markers of bone turnover for the management of patients with bone metastases from prostate cancer. British Journal of Cancer, 2000. 82(4): p. 858-864. 239. Imbriaco, M., et al., A new parameter for measuring metastatic bone involvement by prostate cancer: the Bone Scan Index. Clin Cancer Res, 1998. 4(7): p. 1765-72. 240. Dennis, E.R., et al., Bone scan index: a quantitative treatment response biomarker for castration-resistant metastatic prostate cancer. J Clin Oncol, 2012. 30(5): p. 519-24. 241. Anand, A., et al., Automated Bone Scan Index as a quantitative imaging biomarker in metastatic castration-resistant prostate cancer patients being treated with enzalutamide. EJNMMI Res, 2016. 6(1): p. 23. 242. Berruti, A., et al., Predictive factors for skeletal complications in hormone-refractory prostate cancer patients with metastatic bone disease. British Journal of Cancer, 2005. 93(6): p. 633-638. 243. Ganesan, G.R. and P.V. Vijayaraghavan, Urinary N-telopeptide: The New Diagnostic Test for Osteoporosis. Surg J (N Y), 2019. 5(1): p. e1- e4. 244. Brown, J.E., et al., Bone turnover markers as predictors of skeletal complications in prostate cancer, lung cancer, and other solid tumors. J Natl Cancer Inst, 2005. 97(1): p. 59-69. 245. Morrissey, C., et al., The expression of osteoclastogenesis‐associated factors and osteoblast response to osteolytic prostate cancer cells. The Prostate, 2010. 70(4): p. 412-424. 246. Sanders, J.L., et al., Extracellular calcium-sensing receptor expression and its potential role in regulating parathyroid hormone-related 84 REFERENCES peptide secretion in human breast cancer cell lines. Endocrinology, 2000. 141(12): p. 4357-4364. 247. Guise, T.A., et al., Basic mechanisms responsible for osteolytic and osteoblastic bone metastases. Clin Cancer Res, 2006. 12(20 Pt 2): p. 6213s-6216s. 248. Kavanagh, K.L., et al., The molecular mechanism of nitrogen- containing bisphosphonates as antiosteoporosis drugs. Proceedings of the National Academy of Sciences, 2006. 103(20): p. 7829-7834. 249. Viereck, V., et al., Bisphosphonates pamidronate and zoledronic acid stimulate osteoprotegerin production by primary human osteoblasts. Biochemical and biophysical research communications, 2002. 291(3): p. 680-686. 250. Coleman, R.E., et al., Predictive value of bone resorption and formation markers in cancer patients with bone metastases receiving the bisphosphonate zoledronic acid. J Clin Oncol, 2005. 23(22): p. 4925-35. 251. Fizazi, K., et al., Randomized Phase II Trial of Denosumab in Patients With Bone Metastases From Prostate Cancer, Breast Cancer, or Other Neoplasms After Intravenous Bisphosphonates. Journal of Clinical Oncology, 2009. 27(10): p. 1564-1571. 252. Diel, I., et al., Real-world use of denosumab and bisphosphonates in patients with solid tumours and bone metastases in Germany. Supportive Care in Cancer, 2020. 28(11): p. 5223-5233. 253. Saad, F., et al., Long-term efficacy of zoledronic acid for the prevention of skeletal complications in patients with metastatic hormone-refractory prostate cancer. Journal of the National Cancer Institute, 2004. 96(11): p. 879-882. 254. Smith, M.R., et al., Pamidronate to prevent bone loss during androgen- deprivation therapy for prostate cancer. New England Journal of Medicine, 2001. 345(13): p. 948-955. 255. Lipton, A., et al., Superiority of denosumab to zoledronic acid for prevention of skeletal-related events: a combined analysis of 3 pivotal, randomised, phase 3 trials. Eur J Cancer, 2012. 48(16): p. 3082-92. 256. Kowalska, K. and A.W. Piastowska-Ciesielska, Oestrogens and oestrogen receptors in prostate cancer. SpringerPlus, 2016. 5(1): p. 1- 9. 257. Huang, J.K. and C.R. Jan, Mechanism of estrogens‐induced increases in intracellular Ca2+ in PC3 human prostate cancer cells. The Prostate, 2001. 47(3): p. 141-148. 258. Yu, L., et al., Estrogens promote invasion of prostate cancer cells in a paracrine manner through up-regulation of matrix metalloproteinase 2 in prostatic stromal cells. Endocrinology, 2011. 152(3): p. 773-781. 259. Mollica, V., et al., Bone Targeting Agents in Patients with Metastatic Prostate Cancer: State of the Art. Cancers (Basel), 2021. 13(3). REFERENCES 85 260. Basaria, S., et al., Hyperglycemia and insulin resistance in men with prostate carcinoma who receive androgen-deprivation therapy. Cancer, 2006. 106(3): p. 581-8. 261. Miller, K., et al., Phase III, randomized, placebo-controlled study of once-daily oral zibotentan (ZD4054) in patients with non-metastatic castration-resistant prostate cancer. Prostate Cancer Prostatic Dis, 2013. 16(2): p. 187-92. 262. Quinn, D.I., et al., SWOG S0421: Phase III study of docetaxel (D) and atrasentan (A) versus docetaxel and placebo (P) for men with advanced castrate resistant prostate cancer (CRPC). Journal of Clinical Oncology, 2012. 30(15_suppl): p. 4511-4511. 263. Wiseman, L.R. and K.L. Goa, Toremifene. Drugs, 1997. 54(1): p. 141- 160. 264. Smith, M.R., et al., Toremifene to reduce fracture risk in men receiving androgen deprivation therapy for prostate cancer. J Urol, 2010. 184(4): p. 1316-21. 265. Gururajan, M., et al., SRC family kinase FYN promotes the neuroendocrine phenotype and visceral metastasis in advanced prostate cancer. Oncotarget, 2015. 6(42): p. 44072. 266. Koeneman, K.S., F. Yeung, and L.W. Chung, Osteomimetic properties of prostate cancer cells: a hypothesis supporting the predilection of prostate cancer metastasis and growth in the bone environment. The Prostate, 1999. 39(4): p. 246-261. 267. Soriano, P., et al., Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell, 1991. 64(4): p. 693-702. 268. Knudsen, B.S., et al., High expression of the Met receptor in prostate cancer metastasis to bone. Urology, 2002. 60(6): p. 1113-1117. 269. Agoulnik, I.U., et al., Role of SRC-1 in the promotion of prostate cancer cell growth and tumor progression. Cancer research, 2005. 65(17): p. 7959-7967. 270. Smith, D.C., et al., Cabozantinib in patients with advanced prostate cancer: results of a phase II randomized discontinuation trial. Journal of clinical oncology, 2013. 31(4): p. 412. 271. Evan, Y.Y., et al., Phase II study of dasatinib in patients with metastatic castration-resistant prostate cancer. Clinical Cancer Research, 2009. 15(23): p. 7421-7428. 272. Bruland, Ø.S., et al., High-linear energy transfer irradiation targeted to skeletal metastases by the α-emitter 223Ra: adjuvant or alternative to conventional modalities? Clinical cancer research, 2006. 12(20): p. 6250s-6257s. 273. Suominen, M.I., et al., Radium-223 Inhibits Osseous Prostate Cancer Growth by Dual Targeting of Cancer Cells and Bone Microenvironment in Mouse Models. Clin Cancer Res, 2017. 23(15): p. 4335-4346. 274. Hoskin, P., et al., Efficacy and safety of radium-223 dichloride in patients with castration-resistant prostate cancer and symptomatic 86 REFERENCES bone metastases, with or without previous docetaxel use: a prespecified subgroup analysis from the randomised, double-blind, phase 3 ALSYMPCA trial. The Lancet Oncology, 2014. 15(12): p. 1397-1406. 275. Zhao, N., et al., Pharmacological targeting of MYC-regulated IRE1/XBP1 pathway suppresses MYC-driven breast cancer. The Journal of clinical investigation, 2018. 128(4): p. 1283-1299. 276. Xu, L., et al., Endoplasmic Reticulum Stress in Bone Metastases. Front Oncol, 2020. 10: p. 1100. 277. Chopra, S., et al., IRE1α–XBP1 signaling in leukocytes controls prostaglandin biosynthesis and pain. Science, 2019. 365(6450). 278. Sabino, M.A.C., et al., Simultaneous reduction in cancer pain, bone destruction, and tumor growth by selective inhibition of cyclooxygenase-2. Cancer research, 2002. 62(24): p. 7343-7349. 279. Mason, M.D., et al., Adding celecoxib with or without zoledronic acid for hormone-naïve prostate cancer: long-term survival results from an adaptive, multiarm, multistage, platform, randomized controlled trial. Journal of clinical oncology, 2017. 35(14): p. 1530. 280. Abeshouse, A., et al., The molecular taxonomy of primary prostate cancer. Cell, 2015. 163(4): p. 1011-1025. 281. Ren, S., et al., Whole-genome and Transcriptome Sequencing of Prostate Cancer Identify New Genetic Alterations Driving Disease Progression. Eur Urol, 2018. 73(3): p. 322-339. 282. Fong, P.C., et al., Inhibition of poly (ADP-ribose) polymerase in tumors from BRCA mutation carriers. New England Journal of Medicine, 2009. 361(2): p. 123-134. 283. Isaacsson Velho, P., et al., Efficacy of Radium-223 in Bone-metastatic Castration-resistant Prostate Cancer with and Without Homologous Repair Gene Defects. Eur Urol, 2019. 76(2): p. 170-176. 284. Franceschi, R.T., et al., Multiple signaling pathways converge on the Cbfa1/Runx2 transcription factor to regulate osteoblast differentiation. Connective tissue research, 2003. 44(1): p. 109-116. 285. Zhang, H., et al., FOXO1 inhibits Runx2 transcriptional activity and prostate cancer cell migration and invasion. Cancer research, 2011. 71(9): p. 3257-3267. 286. Lim, M., et al., Runx2 regulates survivin expression in prostate cancer cells. Laboratory Investigation, 2010. 90(2): p. 222-233. 287. Akech, J., et al., Runx2 association with progression of prostate cancer in patients: mechanisms mediating bone osteolysis and osteoblastic metastatic lesions. Oncogene, 2010. 29(6): p. 811-821. 288. Akech, J., et al., Runx2 association with progression of prostate cancer in patients: mechanisms mediating bone osteolysis and osteoblastic metastatic lesions. Oncogene, 2010. 29(6): p. 811-821. 289. Shiozawa, Y., et al., Human prostate cancer metastases target the hematopoietic stem cell niche to establish footholds in mouse bone marrow. J Clin Invest, 2011. 121(4): p. 1298-312. REFERENCES 87 290. Senbanjo, L.T., et al., Characterization of CD44 intracellular domain interaction with RUNX2 in PC3 human prostate cancer cells. Cell Communication and Signaling, 2019. 17(1): p. 1-13. 291. Hagberg Thulin, M., et al., Osteoblasts stimulate the osteogenic and metastatic progression of castration-resistant prostate cancer in a novel model for in vitro and in vivo studies. Clin Exp Metastasis, 2014. 31(3): p. 269-83. 292. Huang, J., et al., The roles of RUNX2 and osteoclasts in regulating expression of steroidogenic enzymes in castration-resistant prostate cancer cells. Mol Cell Endocrinol, 2021. 535: p. 111380. 293. Kawate, H., et al., Mutual transactivational repression of Runx2 and the androgen receptor by an impairment of their normal compartmentalization. The Journal of steroid biochemistry and molecular biology, 2007. 105(1-5): p. 46-56. 294. Bai, X., et al., MicroRNA-196b Inhibits Cell Growth and Metastasis of Lung Cancer Cells by Targeting Runx2. Cell Physiol Biochem, 2017. 43(2): p. 757-767. 295. Taipaleenmaki, H., et al., Targeting of Runx2 by miR-135 and miR-203 Impairs Progression of Breast Cancer and Metastatic Bone Disease. Cancer Res, 2015. 75(7): p. 1433-44. 296. Austin, T.b.M., First microRNA mimic enters clinic. Nature biotechnology, 2013. 31(7): p. 577. 297. Ge, C., et al., Role of Runx2 phosphorylation in prostate cancer and association with metastatic disease. Oncogene, 2016. 35(3): p. 366-76. 298. 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. 299. Sobel, R.E. and M.D. Sadar, Cell lines used in prostate cancer research: a compendium of old and new lines--part 1. J Urol, 2005. 173(2): p. 342-59. 300. Namekawa, T., et al., Application of Prostate Cancer Models for Preclinical Study: Advantages and Limitations of Cell Lines, Patient- Derived Xenografts, and Three-Dimensional Culture of Patient-Derived Cells. Cells, 2019. 8(1). 301. Daniel, V.C., et al., A primary xenograft model of small-cell lung cancer reveals irreversible changes in gene expression imposed by culture in vitro. Cancer research, 2009. 69(8): p. 3364-3373. 302. Lam, H.M., H.M. Nguyen, and E. Corey, Generation of Prostate Cancer Patient-Derived Xenografts to Investigate Mechanisms of Novel Treatments and Treatment Resistance. Methods Mol Biol, 2018. 1786: p. 1-27. 303. Sedelaar, J.M. and J.T. Isaacs, Tissue culture media supplemented with 10% fetal calf serum contains a castrate level of testosterone. The Prostate, 2009. 69(16): p. 1724-1729. 88 REFERENCES 304. Ho, C.K., et al., Calculated free testosterone in men: comparison of four equations and with free androgen index. Annals of clinical biochemistry, 2006. 43(5): p. 389-397. 305. Van Weerden, W., et al., Adrenal glands of mouse and rat do not synthesize androgens. Life sciences, 1992. 50(12): p. 857-861. 306. Luu-The, V., G. Pelletier, and F. Labrie, Quantitative appreciation of steroidogenic gene expression in mouse tissues: new roles for type 2 5α-reductase, 20α-hydroxysteroid dehydrogenase and estrogen sulfotransferase. The Journal of steroid biochemistry and molecular biology, 2005. 93(2-5): p. 269-276. 307. Ashton, J.R., J.L. West, and C.T. Badea, In vivo small animal micro-CT using nanoparticle contrast agents. Front Pharmacol, 2015. 6: p. 256. 308. Kasivisvanathan, V., et al., MRI-targeted or standard biopsy for prostate-cancer diagnosis. New England Journal of Medicine, 2018. 378(19): p. 1767-1777. Illustration figures are created with BioRender.com 89