University of Calgary PRISM: University of Calgary's Digital Repository
Graduate Studies The Vault: Electronic Theses and Dissertations
2017 The Role of the ING3 Epigenetic Regulator in Prostate Cancer
Nabbi, Arash
Nabbi, A. (2017). The Role of the ING3 Epigenetic Regulator in Prostate Cancer (Unpublished doctoral thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/28360 http://hdl.handle.net/11023/3584 doctoral thesis
University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca UNIVERSITY OF CALGARY
The Role of the ING3 Epigenetic Regulator in Prostate Cancer
by
Arash Nabbi
A THESIS
SUBMITTED TO THE FACULTY OF GRADUATE STUDIES
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
GRADUATE PROGRAM IN MEDICAL SCIENCE
CALGARY, ALBERTA
JANUARY, 2017
© Arash Nabbi 2017 Abstract
INhibitor of growth (ING) proteins are epigenetic regulators and stoichiometric members of histone acetyltransferase (KAT) or histone deacetylase (KDAC) complexes. By reading the histone mark H3K4me3, they direct their complexes to chromatin to alter gene expression. This thesis focuses on the role of ING3 in prostate cancer biology.
Since rigorous characterization of antibodies is a prerequisite to acquire reliable results, we began by characterizing a new mouse monoclonal antibody against ING3. We profiled the expression of ING3 protein in normal human tissues and found that it is highly expressed in bone marrow, suggesting high expression in hematopoietic cell precursors. We also reported that
ING3 protein levels are highest in proliferating tissues of the small intestine and epidermis.
These data suggest a role for ING3 in promoting cell growth and renewal.
In the second part of this study, we investigated the effects of ING3 on the androgen receptor (AR) pathway in prostate cancer (PC). We hypothesized that ING3 by virtue of being an essential member of TIP60 KAT complex, plays a role in post-translational modifications of AR protein and thereby contributes to PC progression. We found that the levels of ING3 and AR are positively correlated in patient samples and cell lines. ING3 potentiates androgen effects, activating expression of androgen responsive genes and AR-regulated reporters. We showed that
ING3 interacts with the binding domain of AR and this interaction happens in the cytoplasm in the absence of androgens. ING3 increases AR-TIP60 interaction, promoting AR acetylation and nuclear translocation. The activating role of ING3 is independent of its ability to target the TIP60 complex to H3K4me3, identifying a previously unknown function for ING3. Knockdown of
ING3 inhibits PC cell proliferation and migration, establishing ING3 as a positive regulator of growth in PC.
ii Lastly, we asked whether ING3 could serve as a biomarker to distinguish latent versus aggressive PC. ING3 levels are higher in aggressive PC, with high levels of ING3 predicting shorter overall survival. Analysis with other predictive factors shows that including ING3 levels provides more accurate prognosis in PC.
iii Acknowledgements
I would like to express my gratitude to my doctoral supervisor Dr Karl Riabowol for the opportunity to pursue my graduate studies and to be part of his research group. His patience, support and supervision have been undoubtedly crucial to develop this project from start to completion and to help me think independently and test my own ideas and hypotheses.
I would like to thank my supervisory committee members, Dr Randal Johnston, Dr Frank
Jirik and Dr Tarek Bismar, who have been tremendously helpful with their advice and guidance during my PhD program. I would also like to thank Dr John Lewis and Dr Donald Morris for taking time out of their schedule to attend my PhD defense as external examiners.
I would like to thank all past and present members of Riabowol laboratory - Keiko, Uma,
Alex, Laura, Yang, Fangwu, Satbir and Tae-sun for their help, friendship and scientific suggestions throughout my work. My special thanks goes to Subhash, who has been always kind and patient to teach me the nuts and bolts of techniques and to provide me with his invaluable comments during my journey in science. It would not have been possible nor enjoyable without his friendliness and encouragement.
This work would not have been possible without the inter- and intra-departmental support. Thanks to Ms Donna Boland from SACRI antibody facility, who generated and screened numerous hybridoma antibodies and provided me with unending supply of antibodies to perform my experiments. I would like to express my appreciation to members of translational laboratories, especially Dr Emeka Enwere and Ms Michelle Dean for their contributions to my study on tissue samples. I would like to thank Dr Shirin Bonni and her laboratory for helping me optimize my luciferase system and accommodating me to use their equipment in a set of my experiments. I would like to appreciate Dr Olivier Binda and Dr Ula McClurg from Newcastle
iv University and Dr Dieter Fink from Vienna University for ongoing collaborations to take this project to next step.
And finally, special thanks to my parents for their constant care and encouragement during difficult times and my siblings, Christoph, Negi and Ali who have never left me without support, laugh and happiness in face of disappointment and frustration.
v Table of Contents
Abstract ...... ii! Acknowledgements ...... iv! Table of Contents ...... vi! List of Tables ...... viii! List of Figures and Illustrations ...... ix! List of Symbols, Abbreviations and Nomenclature ...... xi!
CHAPTER ONE: INTRODUCTION ...... 1! 1.1. Prostate cancer epidemiology ...... 2! 1.2. Diagnosis and grades of prostate cancer and current challenges ...... 5! 1.3. Management of prostate cancer ...... 9! 1.4. Molecular mechanisms of prostate cancer progression ...... 16! 1.4.1. Androgen receptor (AR) pathway ...... 16! 1.4.2. Mechanisms of Castrate Resistant Prostate Cancer (CRPC) progression ...... 21! 1.5. The INhibitor of growth (ING) family of epigenetic regulators ...... 28! 1.5.1. ING3 is a member of the TIP60 complex ...... 36! 1.6. Hypothesis and specific aims ...... 37!
CHAPTER TWO: MATERIALS AND METHODS ...... 39! 2.1.Cell culture and transfections ...... 40! 2.2. Generation of ING3 mouse monoclonal antibody ...... 40! 2.3. Cloning and viral preparations ...... 41! 2.4. SDS-PAGE and western blotting ...... 42! 2.5. Immunoprecipitation (IP) ...... 42! 2.6. In vitro acetylation assay ...... 42! 2.7. Luciferase assay ...... 43! 2.8. Chromatin immunoprecipitation (ChIP) and quantitative PCR (qPCR) ...... 44! 2.9. Immunofluorescence ...... 47! 2.10. Alamar Blue metabolic survival assay ...... 47! 2.11. Colony forming assay ...... 47! 2.12. Patient cohort ...... 47! 2.13. Immunohistochemistry and automated immunofluorescence ...... 50! 2.14.Transwell migration assay ...... 51! 2.15. Wound healing assay ...... 51! 2.16. Statistical analysis ...... 51!
CHAPTER THREE: RESULTS ...... 53! 3.1. Higher ING3 correlates with proliferation ...... 54! 3.1.1. Characterization of ING3 antibody ...... 54! 3.1.2. ING3 protein profiling in normal human tissues ...... 61! 3.1.3. ING3 promoter analysis ...... 69! 3.2. ING3 acts as a co-activator of AR ...... 73! 3.2.1.ING3 levels correlate with AR levels ...... 73! 3.2.2. ING3 activates the AR pathway ...... 78! 3.2.3. ING3 interacts with AR ...... 87!
vi 3.2.4. ING3 promotes TIP60 function and interaction with AR ...... 94! 3.2.5. ING3 promotes AR acetylation and nuclear translocation ...... 97! 3.2.6. ING3 affects prostate cancer proliferation and migration ...... 102! 3.3. ING3 is a novel prognostic biomarker for prostate cancer ...... 111! 3.3.1. ING3 levels correlate with Gleason score in low AR subgroup ...... 111! 3.3.2. ING3 levels predict prostate cancer patient overall survival and recurrence113!
CHAPTER FOUR: DISCUSSION ...... 121! 4.1. ING3 profiling in normal human tissues ...... 122! 4.2. ING3 is a co-activator of AR in prostate cancer ...... 126! 4.3. ING3 as a novel biomarker for prostate cancer ...... 135!
REFERENCES ...... 139!
APPENDIX A: PERMISSIONS TO REUSE ...... 156!
vii List of Tables
Table 1. List of primers for qPCR experiments ...... 46!
Table 2. Patient characteristics in prostate cancer cohort and the derived datasets ...... 49!
Table 3. Ten transcription factors predicted to bind the ING3 promoter ...... 70!
Table 3. Cox proportional hazard model for derivation dataset ...... 118!
Table 4. Cox proportional hazard model for validation dataset ...... 119!
viii List of Figures and Illustrations
Figure 1. New cases (A) and cancer related deaths (B) of different types of cancer in Canada. ... 3!
Figure 2. Gleason grading system...... 8!
Figure 3. Management of prostate cancer...... 10!
Figure 4. Clinical course of prostate cancer and possible treatment options in each stage...... 15!
Figure 5. The structure of the AR protein...... 18!
Figure 6. Summary of AR aberrations in CRPC...... 22!
Figure 7. Point mutations of the human AR gene...... 24!
Figure 8. ING family splicing isoforms and domains...... 31!
Figure 9. Characterization of a new ING3 monoclonal antibody...... 55!
Figure 10. ING3 protein levels in normal mouse tissues...... 57!
Figure 11. Indirect immunofluorescence using the 2A2 ING3 antibody...... 59!
Figure 12. ING3 staining by 2A2 in blood cells...... 63!
Figure 13. ING3 staining by 2A2 in human epithelium...... 64!
Figure 14. ING3 staining pattern in normal human tissues...... 65!
Figure 15. Quantification of ING3 staining in normal human tissues...... 66!
Figure 16. ING3 expression in growing and quiescent human epithelial cells...... 68!
Figure 17. ING3 promoter and location of RUNX1 binding sites...... 71!
Figure 18. BloodChIP database search for factors binding the ING3 locus...... 72!
Figure 19. ING3 levels correlate with androgen receptor (AR) levels in patient samples...... 74!
Figure 20. ING3 levels correlate with AR levels...... 76!
Figure 21. ING3 promotes androgen-induced gene expression...... 79!
Figure 22. ING3 regulates AR recruitment to the FKBP5 gene androgen response element...... 82!
Figure 23. Validation of an ARE-driven luciferase reporter system...... 84!
Figure 24. ING3 promotes AR transactivation...... 85!
ix Figure 25. ING3 interacts with AR...... 88!
Figure 26. ING3 and AR interact in the cytoplasm...... 90!
Figure 27. ING3 interacts with the DNA-binding domain of AR independent of the PHD...... 92!
Figure 28. ING3 promotes TIP60-AR interaction...... 95!
Figure 29. ING3 affects TIP60-mediated acetylation of AR...... 99!
Figure 30. ING3 overexpression alters AR localization...... 100!
Figure 31. ING3 knockdown reduces AR nuclear translocation...... 101!
Figure 32. ING3 affects prostate cancer cell proliferation...... 103!
Figure 33. ING3 affects proliferation in AR-positive prostate cancer cells...... 105!
Figure 34. ING3 knockdown reduces AR-mediated filopodia formation...... 108!
Figure 35. ING3 knockdown reduces prostate cancer cell migration...... 109!
Figure 36. ING3 levels correlate with Gleason score in AR low subset...... 112!
Figure 37. Validation of datasets based on Gleason score...... 114!
Figure 38. Kaplan Meier curves for patient overall survival based on ING3 levels...... 116!
Figure 39. Kaplan Meier curve for patient biochemical recurrence rate based on ING3 levels. 120!
Figure 40. Model of ING3 function in the AR pathway...... 133!
x List of Symbols, Abbreviations and Nomenclature
ADT Androgen Deprivation Therapy AF Activation Function ANOVA Analysis of Variance AR Androgen Receptor ATCC American Type Culture Collection ChIP Chromatin Immunoprecipitation cm centimeter CRPC Castrate Resistant Prostate Cancer CSS Charcoal-stripped serum d Day DAPI 4`,6-diamidino-2-phenylindole DBD DNA Binding Domain DMEM Dulbecco's modified Eagle's medium Dox Doxycycline EDTA Ethylenediamine-tetraacetic acid FBS Fetal Bovine Serum HAT Histone Acetyltransferase HDAC Histone Deacetylase hr Hour LBD Ligand Binding Domain LUC Luciferase MB Mibolerone mM Milimolar NLS Nuclear Localization Signal PBS Phosphate Buffered Saline PCA Prostate Cancer qPCR quantitative Polymerase Chain Reaction RNAi Ribonucleic Acid interference rpm revolutions per minute RPMI Rosewell Park Memorial Institute medium SDS-PAGE Sodium Dodecyl Sulfate-Polyacrylamide gel electrophoresis shRNA short hairpin RNA siRNA small interfering RNA TMA Tissue Microarray TSA Trichostatin A µg Microgram µl Microlitre µM Micromolar
xi
CHAPTER ONE: INTRODUCTION
1
1.1. Prostate cancer epidemiology
In 1853, the very first case report was published in Lancet describing prostate cancer (PC) as a
"rare disease" (1). Now in 2016, prostate cancer is recognized as the most common malignancy among men, especially in more developed countries. Its prevalence increases dramatically through later decades of life with a median age of 66 at the time of diagnosis according to 2016
American Cancer Society statistics (2). One could imagine, as our life expectancy increases more than ever thanks to healthcare research and our understanding of the aging process, the incidence rate of prostate cancer will also continue to rise.
Although there is discordance between the number of new prostate cancer cases and deaths in different countries, numbers point to a consistent trend. In the US, there were 180,890 new cases and 26,120 deaths from prostate cancer reported in 2016 (2). Canadian cancer statistics published in 2015 indicate there were 24,019 new cases and about 4,141 deaths
(Canadian Cancer Statistics 2015, www.cancer.ca) (Fig 1). In North America, prostate cancer is the most common cancer and the second cause of cancer-related deaths among men after lung cancer. According to GLOBOCAN 2012, prostate cancer is the second most commonly diagnosed cancer worldwide and the 5th cause of cancer-related deaths. These statistics depend on various factors including level of development in the countries and risk factors as mentioned below.
2
Figure 1. New cases (A) and cancer related deaths (B) of different types of cancer in Canada. A) Total number of new cancer cases in Canada in 2015 and the percentage of different types of cancer are shown. The most common malignancy among men is prostate cancer. B) Total number of cancer-related death in Canada in 2015. Prostate cancer is the third most common cancer death in men. (Canadian Cancer Statistics 2015)
3
There are a number of different risk factors involved in determining the incidence and mortality of prostate cancer. Age remains a strong risk factor for prostate cancer. As mentioned before, prostate cancer occurs only through later decades of life. The emergence of the PSA screening test has led to an increase in the reported incidence of the disease earlier in life, particularly in more developed countries where such screenings are commonly performed.
Different ethnic groups also have different incidence rates for prostate cancer with
African Americans being at highest risk of incidence and mortality followed by Whites and
Hispanics according to the Centre for Disease Control and Prevention (www.cdc.gov). A combination of genetic and environmental factors are believed to play a role in this disparity.
Exposure to androgens due to diet differences, polymorphisms in androgen receptor as well as
DHT metabolizing enzymes are among the contributing factors for prostate cancer disparity (3-
5).
Similar to other cancer types, family history is linked to prostate cancer occurrence;
Parental and/or fraternal incidence of prostate cancer significantly increases the probability of occurrence (6). There are now over 100 genetic loci that have been shown to be involved in about one-third of heritable prostate cancers. This list includes BRCA variants that are mostly studied in breast cancer (7). There have been reports across numerous patient cohorts with different ethnic backgrounds and they cumulatively (albeit partially) can predict the onset and the aggressiveness of the disease (8-11).
Other than the clearly defined risk factors noted above, there are a number of potential factors that appear to contribute to prostate cancer incidence and/or mortality including high
BMI, smoking and calcium/vitamin supplements. The association of obesity and physical exercise with prostate cancer is still a subject of debate. Due to many biases across such research,
4
a conclusion is difficult to draw particularly with regard to prostate cancer incidence and the response to Androgen Deprivation Therapy (ADT); however, the current understanding is that obesity is positively associated with more aggressive prostate cancer (12). The effects of weight loss have not been clearly defined in this context, but physical activity has been reported to have inverse correlation with prostate cancer mortality (13). These epidemiological studies point to possible associations, but the underlying biological mechanisms remain unknown and less vigorously investigated and it is, therefore, too early to make solid conclusions.
The role of smoking in prostate cancer incidence is also inconclusive. However, it has been associated with an increased rate of mortality (14,15). Biologically, smoking is linked in a number of reports to DNA methylation in prostate cancer (16).
The reported effects of calcium, vitamin D and dairy products on prostate cancer have also been mixed. The role of vitamin D variants in prostate cancer have been investigated but the overall effects on risk at the population level remains inconclusive (17,18). Since calcium levels are tightly controlled physiologically, we could speculate, unless at extreme levels, they may not be a definitive modifiable risk factor in prostate cancer.
1.2. Diagnosis and grades of prostate cancer and current challenges
Diagnosis
As mentioned in the last section, prostate cancer is one of the most common cancers worldwide and so screening tests have been developed for early and reliable detection. There are a number of diagnostic methods that are currently used in the clinic with the PSA screening test and Digital
Rectal Examination (DRE) being used most commonly.
Prostate specific antigen (PSA) was first reported in the New England Journal of
Medicine as a possible serum biomarker in prostate cancer. It was known that it is not specific to
5
the cancer and can increase in a number of benign conditions such as Benign prostatic hyperplasia (BPH) (19). Since then, it has become a widespread and accepted less invasive common "screening" test for prostate cancer. Despite its lack of specificity, PSA screening in the early detection of prostate cancer has had clear benefits. In fact, according to a recent study, the number of patients with metastatic prostate cancer at presentation declined three fold thanks to
PSA screening (20).
However, the risks of PSA screening risk sometimes outweigh its benefits. For example, in 2003 an investigation regarding the effects of Finasteride in prevention of prostate cancer was performed by recruiting 18,882 men. They were subject for annual PSA screening for a period of
7 years and at the end of that period biopsies were taken. The analysis of the placebo group was interesting in the sense that PSA screening failed to detect prostate cancer in several patients in the trial (with a common cutoff point of 4 ng/ml) (21). In addition, there was a dispute regarding over-diagnosis caused by the PSA screening; a fraction of the diagnosed prostate tumours involves a latent form of the disease and may not be clinically significant. In 2009, the results of the prostate, lung, colorectal and ovarian cancer (PLCO) screening trial was published, concluding that using 7 and 10 year follow up data, there is no clear benefit of PSA screening for prostate cancer mortality (22). In 2012, the same study was published with a 13 years follow up period with almost identical conclusions (23). Such studies, and the lack of optimal PSA cutoff point, made the usefulness of prostate cancer screening questionable. In fact, such doubts led to modified recommendations for PSA screening from the American Cancer Society including less frequent tests and adaptation of further caution only in high-risk men (24). Also in 2012, recommendations by the US Preventive Services Task Force (USPSTF) included one against using PSA screening diagnostically, especially in men >75 years old (25).
6
Another common method of prostate cancer diagnosis is DRE. This method was widely used by physicians well before PSA screening. It offers a cheap, easy and relatively accurate detection for prostate cancer based on obvious nodules. However, it is extremely subjective and cannot detect early disease. The false positive rate is relatively high and varies between different studies. In a recent study using the PLCO trial data, the authors indicated that only 2% of people with suspicious DRE were found to have significant disease upon biopsy and this number increased to 20% when the DRE and PSA tests were combined (26). This demonstrates the extremely subjective and misleading nature of DRE examinations, which results in unnecessary costs and invasive procedures for geriatric patients.
Grading system
The most widespread grading system in prostate cancer was generated by Dr. Donald
Gleason. This grading system is based on structural alterations of the prostate gland and is commonly used by pathologists to develop a prognosis and design appropriate medical intervention. Figure 2 shows Dr. Gleason's drawing and the corresponding scores. He showed that the first and second most prominent patterns are critical for determining the grade of the disease (27). Since then, common practice is to report the Gleason pattern 1 and 2. The sum of both patterns will be then reported as Gleason score. For example, if the first prominent pattern is
4 and the second is 3, the Gleason score will be 4+3 =7. This system has been modified in 2005 by participation of pathologists and urologists with addition of newly discovered variants of prostate cancer (28).
7
Figure 2. Gleason grading system. Gleason pattern 1 consists of well differentiated and regular glands. Pattern 2 is similar to the first pattern with slight irregularities in shape of the glands. Gleason pattern 3 involves moderate irregularities in size and shape of the glands. Pattern 4 consists of large fused glands and Gleason pattern 5 is poorly differentiated with no gland formation. (29) (Reproduced with permission)
8
Although it is a useful system to grade cancer and is widely accepted, Gleason score system has a number of limitations leading to confusion for both patients and clinicians. There has been inconsistency among pathologists, especially when it came to lower grades of prostate cancer (Gleason score of < 6) (30). This grading method is subjective and therefore dependent on experience of the doctors. General pathologists are more prone to under- or over-estimate
Gleason score than specialized pathologists (31). The Gleason score of 7 has also been misleading due the fact that outcome of patients with the 3+4 pattern has been reported to be very different from those with 4+3 (32). Ongoing modifications and new grading systems have been developed in an attempt to address these complications with the aim of reaching a consensus (28,32,33).
1.3. Management of prostate cancer
The fact that most patients diagnosed with prostate cancer are elderly, and have other comorbidities makes the management of the disease more challenging. Not all prostate tumours grow at a similar rate and it is clinically difficult to distinguish indolent versus aggressive prostate tumours. While some patients require immediate intervention either therapeutically or surgically, others may benefit from watchful waiting. When patients are selected properly, active surveillance can have a very good outcome, avoiding potential side effects of therapy. Apart from the advanced stages of the disease requiring chemotherapy, treatment options of prostate cancer can be divided into three major groups: Surgery, Radiation Therapy and Hormonal
Therapy. Figure 3 shows the most common management scheme for prostate cancer.
9
Figure 3. Management of prostate cancer. Active surveillance is a good approach for low-grade asymptomatic prostate cancer in those with life expectancy of less than 10 years. Treatment of high-grade localized disease mainly involves surgery and/or radiation therapy, while a combination of hormonal and radiation therapy is used for advanced PC. (adopted from Textbook of Therapeutics with permission)
10
Radical prostatectomy (RP) is a common surgical procedure for prostate cancer. The typical patient who may be eligible for RP is someone who has an organ-confined disease and reasonable life expectancy, without other major medical conditions. The patient survival based on this procedure has been studied with generally good outcomes, with disease-specific mortality rate as low as 7% in one study (34,35). The procedure outcome and complications have been greatly improved as a result of new surgical techniques and equipment. However, like all other invasive methods, this procedure retains a range of complications affecting the patient's quality of life.
Apart from common surgical complications such as infection, one of the concerning post-
RP adverse effects is urinary incontinence. The subjective nature of reporting incontinence and lack of quantitative measurements make it difficult to reach consistency across studies. Some studies report excellent continence rate with different surgical procedures (36), while others report higher number of patients requiring pads due to post-operative stress incontinence (37).
Although, it usually improves with time after surgery, incontinence remains a concern for patients undergoing RP.
Another major complication of such procedures is potency. In a recent study of 1,288 men who underwent RP, 28% of men reported maintenance of potency 60 months after surgery
(38). The post-operative erectile dysfunction rate is, however, highly dependent on the patient's age (38). Similar to urinary incontinence, technical improvements have led to a decline in such complications.
Radiation Therapy (RT) is another treatment option available as monotherapy or combinatorial therapy with hormones or surgery. Coupled with new imaging techniques, this treatment modality delivers therapeutic dosages of ionizing radiation (usually in the range of 70-
11
80 Gy) to tumours with less exposure of normal tissues. The efficacy and the rate of treatment failure depend on the total prescribed dose as well as the delivery techniques. The general findings from several trials suggest mere benefits at the upper limits of radiation dose (more than
78 Gy), albeit with higher risks of toxicity (39-41).
Although well tolerated due to using fractionated doses and technical improvements, short- or long-term toxicities can occur upon completion of RT. The lower GI tract and genitourinary system are subject to radiation exposure and therefore are major sites of complications. Impotence remains a major concern especially with young high-risk patients as they receive higher dose of radiation in combination with ADT (42). The incidence of impotence varies from one study to another, ranging from 30% to 70% (43-45). Currently, the use of PDE-5 inhibitors could be helpful in a subset of patients and thus are being commonly prescribed.
Hormonal Therapy (HT) or Androgen Deprivation Therapy (ADT) is the third major option for primary prostate cancer. This treatment is usually combined with RT. Its implications can be dated back to 1972 when Huggins and Hodges performed orchiectomies on patients with prostate cancer and found a dramatic reduction in the levels of alkaline phosphatase (46). ADT can be initiated as neoadjuvant, post-operative adjuvant or in combination with RT (27). Apart from surgical castration, androgen deprivation can be achieved by a number of medications targeting the androgen receptor (AR) pathway at different levels. They include: Luteinizing hormone-releasing hormone (LHRH) modulators such as Goserelin and Leuprolide, anti- androgens such as Flutamide, Bicalutamide and more recently Enzalutamide and the androgen synthesis inhibitor, Abiraterone.
In the majority of patients with primary PC, chemical androgen ablation as neoadjuvant or in combination with other treatment modalities leads to a dramatic reduction in the serum
12
levels of PSA, which is considered a marker of response (47-49). Although there is debate about timing, regimen and duration of ADT, combining ADT with RT significantly improves outcomes such as progression free survival and decreased rates of metastasis (50,51). More recently, the combination of Docetaxel and ADT ("chemohormonal therapy") has been studied, showing impressive outcome in clinical trials (52). This has led to recommendations favouring chemohormonal therapy in hormone-sensitive metastatic prostate cancer. Considering the adverse effects of ADT and its frequent impact on quality of life, the benefits and harms of such therapies need to be carefully assessed.
Adverse effects due to ADT dramatically alter patients’ quality of life and this is primarily a result of significant inhibition of androgens. LHRH agonists such as leuprolide induce a "flare" effect due to the initial rise of testosterone production. Depending on the stage of the disease and the presence of metastasis, this could have serious consequences on survival.
Cases with uretheral obstruction or spinal cord compression have also been reported (53,54).
Addition of a nonsteroidal anti-androgen such as bicalutamide may help alleviate the flare phenomenon. Among common adverse effects of anti-androgens are general pain and hot flashes of various intensity (55). Impotence has also been reported to be common in patients undergoing
ADT for obvious reasons and so affects one's sexual life (56). Loss of muscle mass and gynecomastia are among a relatively common group of effects of such medications, according to the drugs' monographs, and also include psychosocial complications among men with PC (57).
Two potentially fatal adverse effects of ADT include bone loss and cardiovascular dysfunction. Osteoporosis exposes patients to increased risk of bone fractures as androgens are key to bone homeostasis and this effect of ADT is often overlooked in prostate cancer patients
13
(58-60). Currently, upon initiation of ADT, calcium and vitamin D supplementation is recommended.
Cardiovascular complications are another fatal consequence of ADT. This is of particular importance in patients with pre-existing cardiovascular or metabolic disorders making prostate cancer management and the risk-benefit assessment challenging. ADT affects several cardiovascular risk factors such as obesity and dyslipidemia. Cases with heart failure, stroke, myocardial infarction and arrhythmias have been reported to increase with use of various anti- androgens (54,55,61). In a population-based study, use of LHRH agonists were associated with
44%, 16% and 11% increase risk of diabetes, coronary artery disease and myocardial infarction, respectively (62). Based on such concerns, the FDA issued a warning for cardiovascular risks of
ADT and there is general caution among urologists, as discussed in American Urology
Association meeting (AUA 2014) (63,64).
In addition to quality of life complications, the ultimate consequence of ADT is that almost all patients will develop the resistance to such therapies, a stage of PC known as castrate resistant prostate cancer (CRPC). CRPC consists of a variety of clinical states, from localized to metastatic PC (Fig.4) (65). Historically, the only treatment at this stage has been taxanes such as
Docetaxel (66). More recently, the approvals of Enzalutamide and Abiraterone have expanded treatment options. However, they are not curative and resistance eventually develops leaving
CRPC incurable. There are ongoing investigations to elucidate mechanisms of CRPC emergence and biomarker discovery indicating the emergence of resistance. Despite the nature of androgen independence in CRPC, research has shown that the AR pathway is still active and remains central in prostate cancer growth at this stage. Multiple additional molecular mechanisms have been examined, as will be discussed below.
14
Figure 4. Clinical course of prostate cancer and possible treatment options in each stage. Upon diagnosis, most prostate cancers are confined within the organ. Depending on the choice of treatment, the localized disease progresses to either metastatic castrate-sensitive disease or remains organ-confined but resistance occurs. Treatment at this point involves anti-androgen therapy, eventually leading to CRPC and metastasis to other sites such as bone, lymph node and visceral organs.
15
1.4. Molecular mechanisms of prostate cancer progression
1.4.1. Androgen receptor (AR) pathway
Considering the fact that one form of prostate cancer treatment consists of different types of castration, it is no surprise that the AR pathway is central in PC progression. AR is a steroid nuclear receptor, whose function is highly dependent on its localization to the nucleus. Nuclear receptors, upon binding of ligand and translocation to the nucleus, act as transcription factors and bind their respective response elements, altering gene expression. Steroid hormone receptors primarily reside in the cytoplasm and are translocated to the nucleus upon binding to their respective ligands. Once in the nucleus, they form homodimers and bind to hormone response elements (HRE) to regulate expression of target genes. The HRE consists of an inverted sequence three nucleotides apart, which is recognized by the homodimerized receptors in a head- to-head fashion.
The AR protein consists of 919 amino acids (~110 kDa) encoding different domains that account for its functions including ligand, co-regulator and DNA binding domains (Figure 5).
The N terminal domain (NTD) serves as a regulatory domain and it is the least conserved domain across nuclear receptors. The activation function 1 (AF-1) domain is thought of as a transcriptional activator of AR and is located within the NTD. It is independent of ligand binding properties and mediates the co-regulator functions of the AR (67). The NTD appears to regulate transcriptional activity of the AR by an interaction with the C terminal domain in the presence of androgen (68). There are reports that members of co-repressor complexes can bind the NTD leading to transcriptional repression (69). The glutamine repeat sequence located in the NTD accounts for several AR polymorphisms, which is one of the mechanisms responsible for interracial differences in PC prevalence (70).
16
The DNA binding domain (DBD) of the AR is the most highly conserved domain across steroid nuclear receptors, which partially explains the functional overlap among them. This domain consists of two zinc fingers. The first one has P-box sequence, which is responsible for recognition of the androgen response element (ARE). The D-box sequence in the second zinc finger plays a key role in receptor dimerization. The C terminal end of the DBD and the hinge region encode the AR nuclear localization signal (NLS) domain. It is now known that the binding of ligand causes a conformational change leading to exposure of the NLS and nuclear translocation. Three lysines in the hinge region are subject to acetylation by lysine acetyltransferases (KATs) such as TIP60 and P300, promoting nuclear localization (71,72).
The C terminal part of the AR protein contains the ligand binding domain (LBD). Within this domain, there is the activation function 2 (AF-2) domain, which is mostly responsible for providing a surface for co-activator binding. Upon binding of ligand, LBD undergoes a conformational change, which stabilizes the ligand-bound receptor and exposes the AF-2 surface for co-activator binding. It is also involved in the intramolecular N/C interactions (73).
17
Figure 5. The structure of the AR protein. AR protein consists of 919 aminoacids. NTD and AF-1 are involved in co-regulator functions.
DBD consists of two zinc fingers with DNA binding and homodimerization properties. Hinge region is an unstructured region of AR partially containing the NLS. LBD is responsible for ligand binding and intramolecular N/A interactions. (NTD: N-terminal domain, DBD: DNA binding domain, H: Hinge region, LBD: Ligand binding domain, AF: Activation Function domains)
18
The AR pathway involves many other proteins and complexes known as co-regulators, which consist of co-activators and co-repressors. The term co-regulator is less functionally defined and co-regulators include a wide range of proteins with functions in AR post- translational modifications, chromatin remodelling and gene transcription among others. They are typically studied by ARE-driven reporter assays and protein-protein interactions in different cell types. As will be discussed in 1.4.2, co-activators and co-repressors are of wide interest due to their contribution in PC development into CRPC.
Steroid receptor co-activator 1-3 (SRC1-3) proteins were among the first reported to be involved in the AR pathway. They act as scaffolding proteins, to recruit other proteins such as
KATs and methyltransferases. SRC1 knock out mice are interesting in that they are viable but the hormone dependent organs such as breast and prostate were affected, and their growth in response to steroids was hampered (74). SRC2 knockout mice exhibit a lack of normal spermatogenesis and affect AR primarily in the testis (75,76). SRC3-/AR+ mice develop significantly smaller prostate tumors than SRC3+/AR+ mice. SRC3 knockout in transgenic adenocarcinoma mouse prostate (TRAMP) mice completely abrogated neuroendocrine-like PC
(which has CRPC properties) indicating its significant role in this type of PC (77).
KATs modify key cellular functions via direct post-translational modifications of proteins and transcription factors or acetylating histones resulting in chromatin modification. Hence, they affect the AR pathway at various stages. P300 is the first KAT reported to acetylate AR on its hinge region, promoting nuclear localization and AR activity and reducing the recruitment of co- repressors (72,78). It has also been shown to be critical for IL-6 mediated AR activation (79).
Similarly, P/CAF also acetylates the hinge region of AR promoting its activity (78). About the same time, work by Dr. Craig Robson revealed TIP60, an essential KAT, as a co-activator of
19
hormone receptors including AR. Their further studies showed that TIP60 acetylates AR and they form a complex with HDAC1 on the PSA promoter that is promoted by the androgen
(71,80-82). Later on, another group found that TIP60 acetylates the same sites as P300 and
P/CAF (on the AR hinge region) and that the acetylation modulates AR nuclear translocation regardless of the ligand (83).
Histone methyltransferases such as CARM1 are also known as AR co-activators.
CARM1 can promote AR transcriptional activity but it is dependent on expression of SRC1 or 2 and was therefore referred to as a "secondary" co-activator (84). G9a is also reported by one study to have AR activation capabilities (85). Demethylases such as LSD1 (86) and JARID1B
(87) have also been reported to have activating properties, suggesting the complex dynamics of histones modifications in affecting the AR pathway.
Efficient turnover of AR protein is required for its functionality. Inhibition of proteosome activity by MG-132 decreases AR activity as determined by androgen responsive regulation of
PSA transcription (88). E6AP ubiquitin ligase has a significant role in AR function as the knock out mouse develops a smaller prostate (89). It is reported consistently in a number of studies that
AR can reside on gene promoters but it is less transcriptionally active when proteasome activity is blocked indicating that AR turnover is required for proper gene regulation (88,89). Similar to dynamic histone methylation, deubiquitination of histones has been reported to regulate AR transactivation (90). Other known co-activators of AR that will not be discussed further here include phosphatases and kinases (91), SUMO ligases and proteases (92) and DNA repair genes
(93).
AR co-repressors are less well-defined than co-activators. Two major co-repressor complexes are nuclear receptor co-repressor 1 and 2 (NCoR1, NCoR2 a.k.a SMRT), which
20
typically recruit HDACs to deacetylate histones or interfere with receptor interactions on gene promoters (94,95). Although conceptually challenging, co-repressor recruitment or lack of co- activator recruitment have been proposed as a mechanism of action of anti-androgens such as bicalutamide (96-98).
1.4.2. Mechanisms of Castrate Resistant Prostate Cancer (CRPC) progression
Through extensive investigations, it is now clear that AR activity is retained in the majority of
CRPC cases. Other aberrations such as TP53, PTEN and ERG fusion also occur and contribute to the advanced stages of the PC. Since the major focus of this dissertation is the AR pathway, its alterations will be discussed in more detail in this section followed by other major genomic events in CRPC.
Despite an excellent initial response to ADT, almost all prostate tumours develop into
CRPC stage. Yet, even at this stage most tumours rely on AR for cellular survival and proliferation. This phenomenon might be explained by a version of the "oncogene addiction" model, where most tumour cells are "addicted" to AR and would undergo biological adaptations to keep the AR pathway active regardless of the environment.
Although it is probable that other oncogenic pathways will become activated in CRPC independent of AR, current studies point to a still-critical role of AR in a majority of prostate cancer types. Many mechanisms have been proposed for the (hyper)activation of AR in CRPC as summarized in Figure 6. But there is lack of understanding as to when and how a tumour cell chooses one (or more) mechanism over the other and how to prevent, or at least predict, the course of progression for better and more efficient prostate cancer management.
21
Figure 6. Summary of AR aberrations in CRPC. A number of mechanisms are proposed for development of CRPC. AR mutations in LBD result in AR responsiveness to other hormones. AR variants lacking LBD are reported to be constitutively active driving CRPC independent of androgens. Amplification of the AR gene occurs in ~30% of CRPC cases. Increase in intratumour synthesis of androgens can activate AR while the physiological androgens are blocked with anti-androgens. Finally, overexpression of
AR co-activators contributes to development of CRPC.
22
One of the many mechanisms of AR activation in CRPC is the occurrence of mutations that alter the LBD and allow other hormones to be used as AR agonists (99). Not surprisingly,
AR mutations are concentrated mostly within the LBD as noted by the cbioportal website (Figure
7). For example, one of the well-known mutations is T877A in prostate cancer, which expands
AR responsiveness to different molecules such as estrogen, adrenal androgens and flutamide-a classical anti-androgen (100). Mechanisms for how such mutations increase AR activity include stabilizing the ligand-bound receptor, increasing co-activator recruitment and, as is the case by
T877A, increasing the ability of the LBD pocket to bind more diverse ligands (101-103).
23
Figure 7. Point mutations of the human AR gene. The point mutations on the AR protein domains are depicted. Data is from the TCGA prostate adenocarcinoma cohorts by cbioportal.com. All of the point mutations are within the prostate tissue and whether they affect other tissues are not known. The indicated domains are according to protein families database (PFAM). (Androgen_recep: AR amino acids 6-447, zf-C4: zinc finger, Hormone_recep: LBD of AR amino acids 689-879)
24
Another common event affecting AR activity is expression of variants lacking the LBD and therefore becoming constitutively active. Two clinically relevant AR variants are AR-V7 and AR-567es, which are associated with poorer prognosis (104,105). More recently, expression of AR-V7 is shown to occur in Enzalutamide resistant PC xenograft models (106). It is expressed at very low levels but a rapid induction was reported upon castration in xenografts, which might account for basal AR activity and cellular survival in the context of castration (107). The contribution of these deletion mutants to CRPC biology is still under investigation and AR antagonists targeting the DNA binding domain have been generated and studied with the idea of inhibiting both full length AR and its variants (108).
In 1995, a Finnish group reported AR locus amplification in 30% (7 out of 23 samples) of recurrent prostate cancer after ADT (109). Although the number of specimens was small and they reported very high intratumor variation, it was the first report to indicate AR amplification as a possible mechanism for CRPC. The rate of AR amplification has been consistent in other studies since then (110-112). According to public datasets, AR is amplified in 45-54% of cases
(cbioportal.com). In addition, it has been found that overexpression of AR indeed leads to emergence of CRPC and anti-androgen resistance. The overexpression of AR could convert the effects of bicalutamide on gene expression to ones similar to R1881, an androgen analog, and alter co-regulator recruitment in favour of activating the AR pathway (113,114).
An additional mechanism of regulating AR activity in CRPC is increased production of androgens by tumour cells. The idea of intraprostatic conversion of adrenal androgens to more potent AR ligands such as DHT came from early studies where the use of ketoconazole led to a greater decline in intracellular DHT than flutamide (115). Hence, one can speculate that alterations of enzymes involved in androgen synthesis and metabolism may contribute to CRPC.
25
One enzyme responsible for converting testosterone to DHT, 5-alpha reductase, was reported to be up-regulated in prostate cancer and therefore its antagonist, finasteride or dutasteride has been indicated as treatments in early prostate cancer (116,117). Abiraterone is an FDA approved
CYP17A1 inhibitor for metastatic CRPC, which inhibits production of DHEA (118). Its favourable clinical outcome and the emergence of abiraterone resistant tumours highlight the key role of the androgen synthesis pathway in the development of CRPC.
As mentioned in 1.4.1, a growing list of proteins are known as repressors or activators of the AR pathway and, therefore, it is not surprising that aberrations in such proteins can tip the balance of cell and tissue homeostasis in favour of cancer. A number of AR co-activators such as
SRC1, P300 and TIP60 are up-regulated in prostate cancer and contribute to progression to
CRPC. Other than the ones described in 1.4.1 such as KATs and SRCs, some proteins exert their activation roles through direct effects on AR protein. These include members of the heat shock protein family such as HSP70 and HSP90. They are often described as nonclassical AR co- activators with epigenetic regulators affecting receptor activity on gene promoters described as classical co-activators (119). HSPs are involved in stabilizing the AR and its nuclear translocation and compounds such as HSP27 and HSP90 inhibitors targeting these proteins have been tested in PC (120,121). ARA70 is another co-activator of AR that is involved in conformational stability in cytoplasm and nuclear translocation (122). The disruption of co- activators in PC is an active area of research particularly as several of these co-activators act as scaffolding proteins. The recent effort to target SRC-3 is an example of such attempts to target co-activators (123).
Additional genomic events have been shown to contribute to PC progression and CRPC with some of them implicated in CRPC as well as primary prostate cancer. However, which
26
genomic event appears earlier or later in the course of disease is not clear yet. Like other cancer types, alteration of TP53 is common in primary prostate cancer and CRPC. Inactivation of TP53 in the mouse prostate leads to prostatic intraepithelial neoplasia and in combination with RB or
PTEN deletion, induces development of lethal prostate cancer (124,125). Mutations of TP53 are more frequently reported in CRPC compared to primary prostate cancer (126-129).
One of the genomic events that happen exclusively in prostate cancer is TMPRSS2-ERG fusion. It appears in about half of prostate cancer patients, which is a result of fusion between the androgen regulated promoter of TMPRSS2 and ERG (130). The mice with ERG overexpression and PTEN LOH developed high grade prostatic neoplasia, which was consistent with the finding of concurrent ERG fusion and PTEN loss in prostate cancer patients (131). PTEN abnormalities leading to hyper-activation of an oncogenic pathway are common in human cancers and PC is not an exception. Deletion of PTEN in mouse prostate is enough to initiate and progress PC that is sensitive to anti-androgens (132). This implies that PTEN might be an early event and other events such as ERG fusion occur later. However, in a genomics study of 57 matched PC samples, PTEN loss appeared to be a late event (133).
To sum up, a recent multicentre study of 150 CRPC cases conclusively described the critical genomic phenomena in CRPC. The alterations of AR and TP53 happened more frequently in CRPC than primary PC. Other aberrations that were previously known such as loss of PTEN and gain of TMPRSS2-ERG fusion gene were also observed in this study. In addition,
BRCA1, BRCA2 and ATM alterations were reported to be ~19% more frequent in CRPC compared to primary PC. Other novel alterations revealed by this study include APC, beta- catenin and BRAF mutations, which calls for closer examinations regarding their roles in CRPC
(126).
27
1.5. The INhibitor of growth (ING) family of epigenetic regulators
The first member of the ING family of proteins, ING1, was identified, cloned and named by our group in 1996, the overexpression of which inhibited breast cancer cell growth (134). Four other members, ING2-5 were subsequently cloned and reported, which was expedited by the completion of the Human Genome Project (135-137). There were two consistent findings across these early studies:
1) Overexpression of INGs acted as a negative regulator of cell growth in mammalian cells. This finding alongside the studies in cancers such as breast, gastric and lymphoma (138-141) led to classification of INGs as "type II tumour suppressors", a group of tumour suppressors that are frequently dysregulated in cancer, but not necessarily mutated (142,143).
2) All INGs contain a well-conserved plant homeodomain (PHD). The PHD is a domain commonly found in chromatin remodellers, which recognizes histone mark for open chromatin
H3K4me3 (144-146). As will be mentioned below, once INGs bind H3K4me3, associated KATs and KDACs are recruited as epigenetic regulators, affecting chromatin structure and consequently altering gene expression.
Studies in other model organisms such as Saccharomyces cerevisiae and Xenopus laevis not only revealed detailed mechanisms of INGs but also elucidated that they are evolutionarily conserved (147,148). Indeed, when we did a phylogenetic study of 5 INGs across numerous species, we found that INGs are conserved from yeast (three INGs) to vertebrates (five INGs), with ING1 and 2, and ING4 and 5 being closely related and ING3 being more evolutionarily distant (149).
In further attempts to address INGs functions, Loewith et al found that Yng2 (the yeast homolog of ING3) interacts with Esa1, which is the yeast homolog of TIP60 KAT, suggesting a
28
role for INGs in histone acetylation (148). ING1 was interestingly reported to associate with
Sin3A KDAC complex (150). Although INGs were reported in different studies as part of chromatin remodeling complexes, it was a comprehensive biochemical study by the Cote laboratory that coined them as essential stoichiometric members of KDAC (ING1 and 2) and
KAT (ING3-5) complexes (151,152). We now understand that INGs, through their PHD, "read"
H3K4me3 and are involved in recruitment and retention of KDAC and KAT complexes, which
"edit" the histone code, thereby remodelling chromatin and affecting cellular functions through alteration of gene expression. In addition, they are essential members of their respective complexes and therefore can contribute to epigenetic regulation by virtue of making the complexes more efficient and/or facilitating protein-protein interactions.
Structurally, INGs contain multiple domains with different functions (149) (Figure 8).
Apart from the PHD described above, ING1-5 proteins also contain one or two nuclear localization signals (NLS), which accounts for their dominant localization in the nucleus. Within the ING1 NLS, there are nucleolar translocation sequences (NTS), which play a role in UV- induced apoptosis by ING1 in fibroblasts (153). All INGs also have a unique novel conserved region, which is now known to be responsible for the interaction of ING proteins with Lamin A.
Hence, it is called the lamin interacting domain (LID), which helps anchor ING1 in the nucleus to act as an epigenetic regulator (154). The polybasic region (PBR) in ING1 and 2 is key in binding phosphoinositides and interacting with Ubiquitin (155,156). The partial bromodomain
(PBD) that overlaps with the LID in two isoforms of ING1 is known to interact with HDAC1 and HDAC2 (157). Apart from ING1, all INGs contain a leucine zipper like (LZL) domain near the N-terminus. The LZL of ING2 was reported to be required for interaction with P53, nucleotide excision repair as well as muscle differentiation (158,159). ING4 has been reported to
29
homodimerize through its LZL and this dimerization is essential for maximal apoptosis induction, though this dimerization seems unlikely in other LZL containing INGs according to their structural similarities with ING4 (160).
30
Figure 8. ING family splicing isoforms and domains. While some domains of INGs are characterized experimentally, others are proposed based on sequence homology. PHD is the most well-characterized domain with the primary function of binding H3K4me3. We identified LID in ING1, but it has not been experimentally confirmed in other INGs. What is not depicted in the figure is the nucleolar translocation sequence within NLS of ING1. LZL: Leucine zipper like domain, LID: Lamin interacting domain, NLS: Nuclear localization signal, PBD: Partial bromodomain, PBR: Polybasic region, PHD, Plant homeodomain. (From (161) with permission)
31
INGs have been studied in various cancer types where they appear to fulfil the criteria of type II tumour suppressors. ING1 was reported to be down regulated in breast, brain, lung and lymphoid cancers (138,140,162,163). In esophageal squamous cell cancer, mutations were found in the PHD and NLS domains (164). Consistently, mutations in the PHD domain of ING1 were also found in head and neck cancers (165). Mislocalization of ING1 was reported in melanoma, childhood leukemia and brain tumours (166-168). Similarly, other INGs were down regulated or lost their nuclear localization in various types of cancer (169-171). However, some studies reported no mutations or changes in the levels of INGs or even increased levels of INGs, which suggested differential functions of the INGs depending on the context, such as in different tissues and genetic background. For example, although numbers were small and the characteristics of tumour and normal samples were unknown, ING1 was reported to be intact in colorectal carcinoma (172). Campos et al reported overexpression of ING1 in melanoma cell lines (173), and ING2 mRNA levels, appeared to be increased in colon cancer and were shown to regulate invasion (174).
Studies of ING knockout mice further elucidated the diverse functions of this family of proteins. ING1 knock out mouse exhibited spontaneous follicular B cell lymphomas, behavioral abnormalities, hypersensitivity to DNA damage and were smaller in size compared to wild-type
(175). ING2 knockout mouse, however, develops histiocytic sarcomas and were infertile as a result of abnormal spermatogenesis and increased apoptosis in the testes (176). Given the fact that ING1 and 2 are both stoichiometric members of the Sin3A complex, it is speculated that
ING1 and 2 double knockout is lethal during early embryogenesis. ING3 knockout mouse has not been reported. However, ongoing studies by our collaborator suggest that it is also embryonically lethal. In contrast, ING4 knockout mice exhibit a relatively mild phenotype. With
32
no spontaneous tumour formation and weight changes, they have high levels of cytokine production and show reduced innate immunity (177). Similar to ING1 and 2, there is a possibility of compensation for loss of ING4 by ING5 in these mice. ING5 knockout mice have not been reported.
These studies have pointed to a wide range of functions that contribute to ING proteins' function as tumour suppressors. Being a tumour suppressor and members of chromatin remodelling complexes, "INGology" was first focused on the regulation of apoptosis and gene expression with other functions being discovered later by us and several other investigators as noted below.
1) Apoptosis
Overexpression studies have identified the role of INGs in the induction of apoptosis. Infection of cells with adenovirus expressing ING1b increases apoptosis in gliomas (178). Interaction of endogenous ING1b with PCNA is UV-inducible and is important for efficient UV-induced apoptosis. ING1b also translocates to the nucleolus in response to UV, facilitating the apoptotic process (153,179). Through its N terminus, ING1b also induces expression of HSP70, which is involved in apoptosis by TNA-alpha (180). ING1 is also involved in apoptosis induced by TSA in Glioblastoma cell lines (181). ING2 was reported to regulate proliferation and survival, which is dependent on P53, similar to ING3 (135,136). One particularly intriguing finding in apoptosis regulation by INGs is that ING2 is the target of the HDAC inhibitor SAHA, and its dissociation from its complex reduces Sin3A complex recruitment to the p21 promoter that functions in inducing apoptosis (182).
In some of the overexpression studies, adenovirus infection was used as a method of gene delivery. Adenoviral infection usually increases the levels of expression dramatically due to
33
strong viral promoters. Considering the fact that INGs are stoichiometric members of KAT and
KDAC complexes and exert their functions in cooperation with their respective complexes, the overexpression studies might point to functions of INGs that may not be physiologically relevant. This might explain the discrepancy among reciprocal studies. However, the viral delivery of ING1b has been proposed as a novel anti-cancer therapy. For example, we showed that overexpression of ING1b in head and neck squamous carcinoma cell lines increases cell death upon irradiation treatment, which further suggests its potential as a localized viral therapy in head and neck cancers to boost the effectiveness of RT (183).
2) Chromatin remodelling
Through recognizing histone marks and directing their associated KAT and KDAC complexes, chromatin remodelling is one of the major functions of INGs. This function is intercalated with other functions of INGs including those during the development process. Other than functioning within the Sin3A complex, ING1 was shown to interact with P300 and mediate acetylation (184).
ING1 is also shown to mediate DNA demethylation by interacting with and directing GADD45a to H3K4me3 (185). ING2 is involved in muscle differentiation independent of its PHD and
ING4 induction was reported to be essential for prostate epithelial differentiation, which was dependent on its PHD (159,186). ING1 and 2 also interact with Alien, a corepressor of E2F1 and are involved in Alien-mediated gene repression (187).
3) Aging and senescence
The longer isoform of ING1, ING1a, is pivotal in regulating senescence. We reported that its levels increase dramatically as primary fibroblasts undergo senescence (188). Overexpression of
ING1a induced senescence within 48 hours (189). One of the mechanisms of ING1a-induced
34
senescence is through regulation of endocytosis by inducing expression of intersectin 2, which inhibits the RB-E2F pathway, resulting in cellular aging (190).
4) Cancer cell invasion and metastasis
We showed that diminished levels of ING1 in breast cancer correlate with reduced metastasis free survival and that overexpression of ING1 completely eliminated the metastasis in vivo (191). by contrast, ING2 was reported to be elevated in colon cancer and regulated by NF-kB, promoting invasion (174). ING4 overexpression could inhibit MYC induced mammary hyperplasia and an ING4 mutant lacking the PHD increased metastasis and decreased time to tumour formation (192). By regulating IL-6 via the NF-kB pathway, ING4 might also be involved in melanoma angiogenesis (193). In breast cancer, ING4 protein was degraded by the
SCF (JFK) complex and its levels were negatively associated with JFK E3 ligase levels, subsequently promoting angiogenesis and metastasis in vivo (194).
5) Developmental and stem cell biology
The effects of INGs in embryonic development are relatively understudied. Epigenetics plays a significant role in pluripotency and it is highly probable that INGs, by the virtue of their respective complexes, are of critical importance. ING3 is important in oocyte function and division and was recognized as one of the key factors in oocyte reprogramming (195,196).
Through a genetic screen, Mulder et al identified the key chromatin remodellers responsible in epidermal differentiation, one of which was ING5 (197).
6) Hormone Receptor regulation
Induction of ING2 by triiodothyronine in Xenopus laevis was the first indication of the role of the INGs in hormone receptor biology (147). ING1b is known to regulate estrogen receptor alpha through its ligand binding domain (198). More recently, ING1b was reported as an AR co-
35
repressor, inhibiting LNCaP cell growth and inducing senescence (199). ING2 was also reported to function as an AR co-repressor, the expression of which can be regulated by ING1, and induces cellular senescence (200).
1.5.1. ING3 is a member of the TIP60 complex
ING3 is the third member of the ING family of epigenetic regulators described. As mentioned above, it is evolutionarily unique among other INGs. Its locus (7q31) also differs from the other four ING genes that are located near the telomeric regions of the chromosomes (13q34, 4q35,
12p13, 2q37 for ING1, 2, 4 and 5, respectively). It is an essential member of TIP60 KAT complex. We found that knockout of its yeast homolog, Yng2 exhibit severe growth defects.
This evidence, together with lethal phenotype of Esa1 (yeast homolog of TIP60), indicates a crucial function for the NuA4 KAT complex in yeast (148,201). The NuA4/TIP60 complex is conserved from yeast to mammalian cells and the small subcomplex consisting of TIP60, ING3 and EPC1 - known as piccolo NuA4 in yeast, is sufficient to have HAT activity (151). We also found that ING3 deletion mutants in C.elegans are sensitive to IR induced apoptosis in germ cells (202). More recently, the specific recognition of H3K4me3 by the ING3 PHD was reported to be similar to the behaviour of the ING2 PHD further indicating a classical function as an epigenetic regulator for ING3 (146,203).
Like other members of the ING family, ING3 has been studied in a number of cancers.
Reduction in ING3 protein is reported in melanoma (171). However, the antibody used for detecting the ING3 protein was, unfortunately, not sufficiently optimized or characterized. The only control in this study was a no primary antibody control. It was also reported that ING3 was involved in UV-induced apoptosis in melanoma cells and that the reduction of ING3 is dependent on its ubiquitination by Skp2 E3 ligase (204,205). About 50% of head and neck
36
cancer patients exhibit lower expression of ING3 mRNA compared to the matched normal samples in two separate studies (206,207). An immunohistochemical study performed in colorectal mucosa, adenomas and carcinomas reported that ING3 protein levels were reduced from normal lesions to carcinomas (208). Apart from the caveat of the antibody being used not properly recognizing ING3, the levels of ING3 protein were not correlated to any clinical parameters, nor were biologically relevant mechanisms proposed to explain the results presented.
The study of ING3 in growth regulation is currently ongoing and different functions have been suspected in different tissues. As mentioned above, Yng2 knockout in yeast results in growth defects. In the present study, we show that ING3 expression is higher in proliferating tissues such as skin epidermis and small intestine (209). Among all INGs, it was found that only
ING3 levels increased upon treatment of rat cardiomyocytes with Angiotensin II or
Phenylephrine, two stimulators of cardiac hypertrophy and that the overexpression of ING3 leads to increase in mTOR signalling leading to cardiac overgrowth (210).
ING3 was also reported to be one of the most dysregulated genes in prostate cancer and its knockdown decreased DU145 PC cell line invasion (211,212). Through an unbiased discovery based siRNA screening, one independent study attempted to identify genes, the knock down of which can act synergistically with androgen deprived conditions in the VCaP cell line.
ING3 was one of the genes identified in this study as assayed by increased apoptosis or diminished proliferation (213).
1.6. Hypothesis and specific aims
Based on the literature discussed above, ING3 is an epigenetic regulator with a diverse range of functions depending on the tissue type, cancer type and other molecular contexts. Antibodies have been widely used in molecular biology and are considered as crucial tools for deciphering
37
biological mechanisms. However, rigorous controls and optimization are infrequently performed, leading to non-reproducible and sometimes misleading reports. For the first part of this study, therefore, we focus on the validation and optimization of a new monoclonal antibody for ING3 and profile normal human tissues for ING3 protein levels and compare with the available literature.
In the second part of this dissertation, we hypothesized that ING3 could contribute to PC development and progression by virtue of being an essential member of the TIP60 KAT complex. We proposed that since TIP60 is a known co-regulator of AR, ING3 could affect AR protein by promoting its interaction with TIP60. We also tested whether ING3 could function independent of its chromatin remodeling properties.
Finally, based on published data and our preliminary results, we tested ING3 as a novel prognostic biomarker in PC that might aid clinicians in predicting tumor aggressiveness.
Considering the challenge of PC management in the geriatric population and the financial burden resulting from overdiagnosis and overtreatment in this disease, our hope is that this study might be useful in introducing a valuable biomarker for prostate cancer prognosis.
38
CHAPTER TWO: MATERIALS AND METHODS
39
2.1.Cell culture and transfections
LNCaP, VCaP, PC3, DU145 and HEK293T cell lines were purchased from the ATCC. The C4-2 cell line was a kind gift from Dr. Martin Gleave. LNCaP, C4-2 and PC3 were grown in RPMI media supplemented with 10% FBS. DU145, VCaP and HEK293T cells were grown in DMEM supplemented with 10% FBS. For androgen-deprivation, cells were incubated with media supplemented with 5% charcoal stripped FBS (CSS) (InVitrogen) for 48 hr. Mibolerone (MB)
(Toronto Research Chemicals) was used as an androgen analog at concentrations of 1-10 nM.
The pCMV-3myc-AR plasmid was a gift from Dr. Marja Nevalainen. pCIN4-FLAG-HA-TIP60 was a gift from Dr. Wei Gu. AR mutants were a kind gift of Dr Craig Robson. HEK293T cells were transfected using TransIT 293 reagent (Mirus) and PCa cells were transfected using
Lipofectamine LTX (InVitrogen).
2.2. Generation of ING3 mouse monoclonal antibody
Four female BALB/c mice were each injected intraperitoneally with 10 mg of bacterially expressed ING3-GST mixed with Complete Freund's Adjuvant (CFA) initially, and with two subsequent injections at two week intervals using the antigen mixed with incomplete Freund's
Adjuvant (IFA). Three to four days after the third injection, one or two mice were sacrificed, and their spleen cells were fused with the myeloma cell line Sp2/mIL6 using Polyethylene Glycol
1500. The fused cells were plated into 96-well culture plates and when colonies formed, the supernatants were screened by enzyme-linked immunoabsorbent assay (ELISA) to detect positive clones. Briefly, 96-well plates (Nunc-Immuno, Thermo Scientific) were coated with either 1 mg ING3-GST/ml in carbonate buffer, pH 9.2 or 1 mg GST/ml carbonate buffer, pH 9.2 and incubated at 37°C for 2 hours or at 4°C overnight. Plates were then washed three times and non-specific binding sites were blocked with 1% BSA/PBS, pH 7.4 at 37°C for 30 minutes.
40
Supernatants from candidate clones were added to plates previously washed, and were then incubated at 37°C for 1 hour. Goat anti-mouse IgG-horseradish peroxidase (HRP) (Jackson
Laboratories) was added as secondary antibody at a 1:1000 dilution and plates were incubated at
37°C for 30 minutes. Plates were washed three times, ABTS-peroxidase substrate (Mandel scientific, Inc) was added to wells and incubated at 37°C for 30 minutes, and then the absorbance was read using a plate reader at 405 nm. Selected clones were then subcloned by limiting dilution, as previously described (214).
2.3. Cloning and viral preparations
ING3 coding sequence was cloned in the pCDNA3.1 mammalian expression vector fusing an
HA tag on the C-terminus. Deletion mutants were made based on sequence analysis, in silico sequence alignment and available literature regarding function.
Three shRNA sequences against ING3 (shING3) were obtained from the RNAi codex (215). One scrambled shRNA (shCtrl) was generated with a similar GC content to the shING3s and used as a control. The sequences were cloned in pINDUCER10, a doxycycline (Dox) inducible lentiviral vector (216). HEK293T cells were co-transfected with helper and envelope plasmids and pINDUCER10 using TransIT-293 reagent. Supernatants were collected at 48 and 72 hr post- transfection, filtered using 0.45 mm filters and concentrated using spin columns (Amicon,
Millipore). Cells were infected using the concentrated virus plus 8 mg/ml polybrene (Sigma) in serum free media. One day post-infection, media were changed to complete media with or without 100 ng/ml Dox. C4-2 cells were infected with concentrated inducible lentivirus encoding either shCtrl or shING3. After addition of Dox, cells were sorted keying on RFP expression.
Cells were kept in puromycin as the selection marker and knock-down efficiency was confirmed by western blotting. A polyclonal pool of infected cells was used for the experiments.
41
2.4. SDS-PAGE and western blotting
After washing twice with ice-cold PBS, cells were lysed in 50 mM TRIS pH 7.5, 150 mM NaCl,
1 mM EDTA, 1% Triton X-100 supplemented with protease inhibitor cocktail (Roche). After centrifugation and addition of Laemmli's sample buffer, samples were boiled for 5 minutes and electrophoresed. Nitrocellulose membranes (Pall Inc.) were used for transfer and membranes were blotted with in-house mouse ING3 2A2 monoclonal antibody, mouse anti-beta-actin (Santa
Cruz), rabbit anti-HA tag (Santa Cruz) or rabbit anti-GAPDH (Cell Signaling), as indicated in the figures. As secondary antibody, horse radish peroxidase-conjugated secondary antibodies for mouse or rabbit (Millipore) were used. ECL reagent (Millipore) was used to visualize protein bands on X-ray film (Kodak).
2.5. Immunoprecipitation (IP)
For IP, 1x107 cells were lysed at 4°C using lysis buffer (50 mM TrisHCL, 150 mM NaCl, 1 mM
EDTA, 1% Triton X-100, 10 mM NaF, 10 mM Na pyrophostate, 1 mM Na β-Glycerophosphate) supplemented with protease inhibitors (cOmplete, Roche). After centrifugation, a fraction of each cell lysate was taken as input. Antibodies including anti-HA (Roche), anti-Myc (Sigma), anti-acetyl lysine (Cell Signaling or Santa Cruz), anti-ING3 (Kerafast), anti-AR (N-20, Santa
Cruz) or anti-TIP60 (C-7, Santa Cruz) were crosslinked to beads (GE Healthcare) and used for
IP. After incubation with lysates, beads were washed with ice-cold IP buffer three times,
Laemmli sample buffer was added and precipitates were boiled for 10 minutes and used for western blotting. For acetylation studies 0.2% SDS was added to lysis buffer and IP wash buffer to eliminate protein-protein interactions.
2.6. In vitro acetylation assay
HEK293T cells were transfected with GFP, ING3-HA or AR-Myc and 24 hr post-transfection,
42
cells were lysed at 4°C as per the IP protocol. After clearing by centrifugation, GFP- or ING3- transfected cell lysates were incubated with anti-HA and AR-transfected cell lysate was incubated with either anti-Myc or normal rabbit IgG (Santa Cruz) overnight at 4°C. 20 µl of protein A bead slurry was added to each AR immuniprecipitate and incubated at 4°C for 2 hr.
HA immunoprecipitates were washed multiple times with non-denaturing IP buffer, whereas
Myc immunoprecipitates were washed with IP buffer containing 0.1% SDS to minimize pull down of AR-interacting proteins. To perform in vitro acetylation assays, IP samples were washed once in HAT buffer (50 mM Tris Cl pH 8.0, 10% glycerol, 0.1 mM EDTA, 10 mM butyric acid, 2 µM TSA) supplemented with protease inhibitor cocktail. HA-IP samples were mixed with protein A beads or control rabbit IgG IP which served as a negative control. The AR-
IP sample was divided equally into two tubes and mixed with HA-IP samples from either GFP- or ING3-transfected cell lysates. Following addition of 1 mM Acetyl Coenzyme A (Lithium salt,
Sigma), all tubes were incubated at 30°C for 1 hour with occasional shaking. Reactions were terminated by adding 4X Laemmli sample buffer and samples were subject to western blotting with anti-acetyl-K and subsequently anti-AR. To confirm the presence of TIP60 in HA-IPs, a fraction of HA-IP samples and pre-IP samples were run on separate gel and probed with anti-HA and anti-TIP60.
2.7. Luciferase assay
Luciferase reporter assays. 5x104 HEK293T cells were plated in 24 well plates and transfected with plasmids as indicated in the text, together with AR3-tkk-LUC (a gift from Dr. Paul
Rennie)(217), pCMV-3myc-AR and a CMV-β galactosidase (PBL3-β-gal) construct as an internal control (a gift from Dr. Shirin Bonni). Luciferase assays were performed as described
(218). Briefly, one day after transfection, cells were washed with PBS and lysed using reporter 43
lysis buffer (Promega) for one hour at room temperature (RT) and frozen at -20°C overnight. 30
µl of lysates were transferred in triplicate to 96-well plates and luminescence was detected using a Berthold luminometer. Β-gal staining was used as an internal control (218). Plates were incubated at 37°C for 15 minutes and the optical density at 600 nm was measured.
2.8. Chromatin immunoprecipitation (ChIP) and quantitative PCR (qPCR)
The effects of ING3 knock down on AR recruitment to the FKBP5 ARE were determined by performing ChIP assays (219) using 3x107 C4-2 cells transfected with either siCtrl or siING3 for
48 hr in media supplemented with 5% Charcoal stripped serum (CSS), with or without 10 nM
MB. Cells were cross-linked using 1% formaldehyde for 10 min and quenched with 0.125 M glycine. Cells were then lysed in 1 ml ChIP lysis buffer (150 mM NaCl, 50 mM Tris, pH 8.0, 1%
Triton X-100, 0.1% deoxycholate, 1 mM EDTA plus protease inhibitors) and sonicated, eight times for 12 seconds each on ice. After centrifugation the supernatants were immunoprecipiated with rabbit anti-AR (N-20, Santa Cruz) or rabbit control IgG overnight at 4°C and incubated with protein A Beads (GE Healthcare) for 2 h at 4°C. Immunoprecipitates were washed with
ChIP lysis buffer (500 mM NaCl, 10 mM Tris-HCl, pH 8.0, 250 mM LiCl, 0.5% NP-40, 0.5% sodium deoxycholate, 1 mM EDTA) and TE buffer (10 mM Tris and 1 mM EDTA, pH 8.0) and eluted using 1% SDS. The IP and input samples were then reverse cross-linked using NaCl at
65°C overnight and DNA was isolated by ethanol precipitation after phenol-chloroform extraction. Binding of AR to the ARE was tested using qPCR. Primer sequences used were:
FKBP5 ARE6/7: Fwd 5'-CCCCCCTATTTTAATCGGAGTAC-3' and Rev 5'-
TTTTGAAGAGCACAGAACACCCT-3', Non-specific Fwd 5'-
GGTCAGGTTTTGGTTGAGGA-3' and Rev 5'-CAAGCACAGTGAGGGAGACA-3'. TRIzol
44
and an Omniscript Reverse Transcription kit (Qiagen) were used for isolation of total RNA and generating cDNA. Real-time PCR was performed using Maxima SYBR Green Mastermix
(Fermentas) with an Applied Biosystems 7900HT PCR system. The qPCR primer sequences for qPCR assay are listed in Table 1.
45
Table 1. List of primers for qPCR experiments Gene Name Forward primer (5’-3’) Reverse primer (5’-3’)
Maspin CTACTTTGTTGGCAAGTGGATGAA ACTGGTTTGGTGTCTGTCTTGTTG
TMPRSS2 CTGGTGGCTGATAGGGGAT GTCTGCCCTCATTTGTCGAT
KLK2 AGCCTGCCAAGATCACAGAT GGAAGAACTCCTCTGGTTCG
FASN AGGATCACAGGGACAACCTG ACTCCACAGGTGGGAACAAG
FKBP5 TCCCTCGAATGCAACTCTCT GCCACATCTCTGCAGTCAAA
PSA AGGTCAGCCACAGCTTCCCA GGGCAGGTCCATGACCTTCA
ING3 CAGCCAGTGAACAATCACCAT CAGCACAGACACGTTCCTCT
GUSB CGTCCCACCTAGAATCTGCT TTGCTCACAAAGGTCACAGG
NKX3.1 GTACCTGTCGGCCCCTGAACG GCTGTTATACACGGAGACCAGG
Actin GAACCCTAAGGCCAACCGTGA AGGAAGAGGATGCGGCAGTGG
Snail ACTGCAACAAGGAATACCTCAG GCACTGGTACTTCTTGACATCTG
Twist1 GTCCGCAGTCTTACGAGGAG GCTTGAGGGTCTGAATCTTGCT
Ecadherin ATTTTTCCCTCGACACCCGAT TCCCAGGCGTAGACCAAGA
46
2.9. Immunofluorescence
Cells were grown on coverslips and fixed with 4% paraformaldehyde in PBS for 10 min at RT and permeabilized using 0.1% Triton X-100 (Millipore) in PBS for 10 min. Fixed cells were blocked using 5% BSA for 1 hour at RT. Ki67 antibody (Dako) or AR antibody (N-20, Santa
Cruz) were used at a dilution of 1:200 in PBS plus 5% BSA for 2 hr at RT, washed with PBS and incubated with Alexa-488 goat anti-mouse secondary antibody (1:1000 in PBS/5% BSA) for an hour at RT in the dark. Cells were then stained with Hoechst 33258 (Sigma), mounted on slides and analyzed using an Axiovert 200 microscope.
2.10. Alamar Blue metabolic survival assay
LNCaP, PC3 and DU145 cells were transfected with siCtrl or siING3 and 1x104 cells were seeded in 24 well plates. Cells were washed twice and stained with 0.1% crystal violet for 15 min at RT followed by extensive washing. Alamar Blue assays were performed to estimate cell proliferation according to the manufacturer's protocol. Cell proliferation was also monitored by seeding cells in 96-well plates in parallel and counting cells at the indicated times using a Celigo
Cell Cytometer (Cyntellect).
2.11. Colony forming assay
Agar (0.5%) was prepared in RPMI containing 10% FBS and 1 ml was poured into each well of
24-well plates to form a bottom layer. 1x104 cells were then mixed with RPMI-20% FBS containing 0.3% agarose and poured on top of the bottom layer. Colonies were analyzed using an inverted microscope 10 days after seeding. Colony diameters were measured using ImageJ software and colony volumes were calculated (4/3 πr3).
2.12. Patient cohort
Prostate biopsies from a cohort of 312 patients were collected by Dr. Bismar at Rockyview
47
General Hospital. A total of 256 patients were diagnosed with various stages of PC. Two tissue microarray (TMA) slides were constructed with each biopsy in duplicate. ERG and AR expression were determined by immunohistochemistry (211). Table 2 shows the characteristics of patients in this cohort and the randomly derived datasets (see results section).
48
Table 2. Patient characteristics in prostate cancer cohort and the derived datasets Characteristic Derivation Dataset Validation Dataset Combined
n=133 n=123 n=256
Age at diagnosis 55-97 54-96 54-97
Survived: 107 Survived: 94 Survived: 201
Death PC-Related: 23 PC-Related: 22 PC-Related: 45
Missing: 3 Missing: 7 Missing: 10
PCA: 100 PCA: 96 PCA: 196
CRPC CRPC: 32 CRPC: 27 CRPC: 59
Missing: 1 Missing: 1 Missing: 2
=< 7 : 63 =< 7 : 60 =< 7 : 123
Gleason Score > 7 : 67 > 7 : 59 > 7 : 126
Missing: 3 Missing: 4 Missing: 7
Low: 60 Low: 57 Low: 117 AR expression High: 73 High: 66 High: 139
= 0: 101 = 0: 88 = 0: 189
ERG expression > 0: 29 > 0: 33 > 0: 62
Missing: 3 Missing: 2 Missing: 5
< 1.66: 83 < 1.66: 72 < 1.66: 155 ING3 expression > 1.66: 50 > 1.66: 51 > 1.66: 101
49
2.13. Immunohistochemistry and automated immunofluorescence
Four µm thick sections were cut from TMA blocks and deparaffinized in xylene, rinsed in ethanol, and rehydrated. Heat-induced epitope retrieval was performed by heating slides to
121°C at pH 6 in Target Retrieval Solution (Dako) for 3 minutes in a decloaking chamber
(Biocare Medical). Slides were stained using a Dako Autostainer. Endogenous peroxidase activity was quenched with peroxidase block (10 min, Dako) followed by a 15 min protein block
(Signal Stain, Cell Signaling, Danvers, MA, USA) to minimize non-specific antibody binding.
Slides were washed with TBST and incubated at RT for 60 minutes with Signal Stain protein block containing a 1:1500 dilution of ING3 mouse mAb and a 1:100 dilution of anti-pan- cytokeratin rabbit polyclonal antibody (Dako). Secondary reagents were incubated at RT for 60 min: ready-to-use goat anti-mouse antibody conjugated to a horseradish peroxidase-decorated dextran polymer backbone from the DAKO EnVision+ system (Dako) and 1:200 dilution of
Alexa-555 conjugated goat anti-rabbit antibody (InVitrogen). Slides were washed with TBST and incubated for 5 minutes with the TSA-Plus Cy5 tyramide signal amplification reagent
(Perkin Elmer). After three washes in TBST, slides were mounted with ProLong® Gold anti- fade mounting medium containing DAPI and stored at 4°C overnight before scanning.
Automated image acquisition used an Aperio Scanscope FL 8/10-bit monochrome TDI line- image capture camera with filters specific for DAPI, Cy3 (Alexa-555) to define the tumor cytosolic compartment based on cytokeratin, and Cy5 for ING3. Images were analysed using the
AQUAnalysis® program, version 2.3.4.1 as described (220). All images were processed using optimal threshold values and all subsequent image manipulations involved only image information from masked areas. Scores were based on percent area of the nuclear mask that was also positive for ING3 based on the ING3 mask.
50
2.14.Transwell migration assay
Transwell inserts of 8 μm pore size (Corning) were placed in 24-well plates and 500 μl of
RPMI+20% FBS was added to bottom of the plates. After transfection with siRNA and treatment with 1 nM MB for 48 hr (where indicated), 5x104 LNCaP cells were seeded on top of the transwell inserts and supplemented with 250 μl of RPMI+CSS or RPMI+1 nM MB. The inserts were fixed at the indicated time points with 4% paraformaldehyde and methanol and stained with crystal violet. For quantification, 6 random fields were chosen and the cells were counted in a single-blinded fashion using Photoshop CS5.
2.15. Wound healing assay
C4-2 cells stably infected with shING3 or shCtrl were plated at 80% confluency in 6-well plates.
Doxycycline was added to induce the expression of shRNA and cells were grown in RPMI supplemented with 10 nM MB for the duration of the experiment. A 200 μl sterile tip was used to wound monolayers. At the indicated time points, images were taken from the same fields across the course of the experiment. To quantify migration, images were processed in Photoshop
CS5 and the clear region in the initially circumscribed areas (delineated by the red lines in Figure
7E), were measured. The percentage of healed wound was then calculated using the following formula: 1-(surface area in one field)/(surface area of the same field at first day).
2.16. Statistical analysis
All experiments were done in triplicate. Each patient's tissue samples were punched in duplicate on TMA slides. Graphpad Prism was used for graphs, statistical analyses such as standard error calculations, confidence intervals, student's t-tests and ANOVA statistics. SPSS statistics software was used for analyzing TMA results including Kaplan Meier and Cox proportional hazard analyses. Log rank test was used to analyze the significance of survival data. Unless
51
otherwise stated, data represent Mean +/- SEM. Throughout this study and wherever possible, levels of significance were shown as asterisks on the graphs (* <0.05, ** <0.01, *** <0.001).
52
CHAPTER THREE: RESULTS
53
3.1. Higher ING3 correlates with proliferation
3.1.1. Characterization of ING3 antibody
Lysates from HEK293 cells expressing ING1-5 (221) were run on polyacrylamide gels, transferred to nitrocellulose and membranes were probed with the 2A2 monoclonal antibody followed by a secondary HRP-conjugated rabbit anti-mouse secondary and chemiluminescent development. As shown in Figure 9A, the antibody showing the best titre (2A2) recognized overexpressed human ING3, but not overexpressed INGs 1,2,4 or 5. To confirm that it also could recognize ING3 expressed in human cells, HEK293 cells were transfected with pCDNA-ING3 plasmid or pcDNA as a control, and lysates were blotted with ING3 monoclonal antibody. Cells undergoing mock transfection showed that the antibody recognized endogenous ING3 while the signal increased dramatically upon overexpression of ING3 (Figure 9B). Next, we compared the specificity of our antibody with a commercially available antibody. HEK293 cells were co- transfected with pcDNA-ING3 and either scrambled siRNA or siING3. As shown in Figure 9C,
2A2 recognized the ING3 band, and this band decreased in response to increasing amounts of siING3, whereas the commercial antibody detected a band that did not change in intensity in response to ING3 knockdown. To further test if 2A2 was specific for ING3, we undertook immunoprecipitation-western blot analysis using non-denatured HA-tagged ING3 protein. As seen in Figure 9D, this antibody recognized HA-tagged ING3 in immunoprecipitates, and cleared lysates showed no signal as noted in the blot of post-IP samples.
54
Figure 9. Characterization of a new ING3 monoclonal antibody. A) HEK293 lysates expressing ING1-5 were used in western blotting to test specificity of the antibody with lanes 1-5 corresponding to ING1-5. The expression of all INGs was confirmed previously (221). B) HEK293 cells were transfected with ING3 construct and served as the positive control for ING3 antibody. C) HEK293 cells were co-transfected with ING3 and either scrambled siRNA or siING3 (0.5 to 2 pmol). Two antibodies were used to detect the effects of siRNA. D) Immunoprecipitation of lysates from HEK293 cells transfected with HA-ING3 shows that 2A2 recognizes the ING3 since it is precipitated and recognized by an anti-HA antibody in the subsequent blot.
55
Since we have previously determined that INGs are well-conserved in organisms from human to yeast (149), we asked if our new antibody recognized ING3 in other vertebrates. To test if 2A2 could recognize mouse ING3, we blotted for ING3 expression in different mouse tissues. As shown in Figure 10A, the antibody recognizes a band in mouse that co-migrates with overexpressed human HA-ING3, and protein levels of ING3 appear to be higher in lung and spleen compared to other mouse tissues. This observation is relatively consistent with a previous study (210), in which a different commercial antibody was used. Of note, however, is the presence of two bands in most murine tissues, which might represent the existence of multiple splicing isoforms and/or post-translationally modified forms of murine ING3 (See section 4.1).
In order to confirm the detection of mouse ING3, we transfected BALB-3T3 and NIH-3T3 mouse cells with siCtrl or siING3. As shown in Figure 10B, a major band and a weaker band were detected, both of which were reduced upon siING3 transfection indicating the detection of the mouse ING3 band.
56
Figure 10. ING3 protein levels in normal mouse tissues. A) Detection of endogenous ING3 protein in lysates of different mouse tissues. Untransfected
HEK293 cells and HEK293 cells transfected with ING3 construct served as negative and positive controls, respectively. To ensure equal protein loading, blots were stained with Amido black. B) BALB-3T3 and NIH-3T3 mouse cells were transfected with siCtrl or siING3. After 48 hr of incubation, cells were lysed and lysates were subjected to western blotting.
57
The specificity of 2A2 was next tested using indirect immunofluorescence. As shown in
Figure 11, when ING3 antibody was used at a 1:100 dilution, it recognized overexpressed protein in HEK293 cells (row A) as well as endogenous ING3 in untransfected cells (row B). In order to further examine specificity, we incubated the antibody with fixed and permeabilized cells overexpressing ING3 or GFP prior to staining. As shown in figure 11D and E, the antibody incubated with ING3 expressing cells was almost completely unable to recognize the ING3 signal compared to the control antibody. This indicates that the antibody was blocked, which serves as another control for specificity. Moreover, when we knocked down ING3 protein using siRNA, the staining was dramatically diminished compared to scrambled siRNA (Figure 11F and
G) further confirming the specificity of ING3 staining. These data indicate that monoclonal antibody 2A2 specifically recognizes both denatured and formalin fixed human and murine
ING3.
58
Figure 11. Indirect immunofluorescence using the 2A2 ING3 antibody.
59
Staining is shown for A) HEK293 cells co-transfected with ING3 plus GFP expression constructs, B) HEK293 cells transfected with GFP only and C) HEK293 cells stained with secondary antibody only as a negative control. To further examine the specificity, HEK293 cells co-transfected with ING3 plus GFP were stained with blocked ING3 antibody using fixed
HEK293 cell pellet overexpressing ING3 (D) or GFP (E). HEK293 cells were also transfected with scrambled siRNA (F) or siING3 (G) and stained for endogenous ING3.
60
3.1.2. ING3 protein profiling in normal human tissues
We stained ING3 in various commercially available normal human tissues. The staining pattern in human hematopoietic tissues including bone marrow, lymphatic tissue, thymus and spleen are shown in Figure 12. ING3 staining was relatively high in bone marrow suggesting that
ING3 expression may be high in blood progenitor cells (12A). By comparison, ING3 expression was lower and more sporadic in thymus, lymphatic tissue and spleen (12B-D). As shown in
Figure 13, ING3 is also expressed to varying degrees in epithelial cells, particularly in small intestine epithelium (Figure 13A) and skin epidermis (Figure 13B), both of which proliferate rapidly.
In contrast to the high levels of ING3 seen in small intestine and skin epithelia, ING3 expression was moderate in lung epithelium (Figure 13C), while cells in breast, prostate, esophagus, colon and rectum showed only sporadic expression of ING3 protein (Figures 13D-H).
A low level of expression in prostate is consistent with higher, dysregulated levels of ING3 seen in prostate cancer (222), further suggesting a role for ING3 in cell proliferation. Representative images of other human tissues stained with the ING3 2A2 antibody, including stomach, ovary, pancreas, cerebrum, kidney, liver, testis and uterine, are shown in Figure 14.
In order to quantitate the proportion of cells showing detectable staining for ING3 in the various human tissues examined, three individuals scored random fields of cells within each micrograph in a double blind experimental design. The average scores for each of the tissues are shown in Figure 15, which underscores the fact that bone, small intestine and skin show the highest levels of staining, while spleen, lung, kidney and testis show intermediate levels and other tissues show very few cells expressing detectable levels of ING3. These observations identify a correlation between ING3 levels and the predicted growth rate of cells within different
61
tissues, where tissues that contain a larger proportion of growing cells also show a greater proportion of staining for ING3.
62
Figure 12. ING3 staining by 2A2 in blood cells. Representative images of staining in A) bone marrow B) lymphatic tissue C) thymus and D) and spleen are shown. All sections from different human tissues were stained under identical conditions using the same reagents and the same timing for staining, washing and other manipulations. Counterstaining and photography of images was also done under the same conditions for all samples.
63
Figure 13. ING3 staining by 2A2 in human epithelium. Representative images of ING3 staining in human epithelial tissues including A) small intestine,
B) skin, C) lung, D) breast, E) prostate, F) esophagus, G) colon and H) rectum.
64
Figure 14. ING3 staining pattern in normal human tissues.
65
Figure 15. Quantification of ING3 staining in normal human tissues. The graph shows the percentage of cells in each tissue staining for ING3. Numbers were determined in a double blind experimental protocol and averages and standard deviations were calculated. Large error bars in the same samples indicate marginal levels of ING3 that were near the limits of detection. Of note, a pitfall of this method of quantification is that the intensity of
ING3 expression in different tissues was not taken into consideration, which may be of biological significance.
66
The correlation between predicted growth rate of cells in tissues and the levels of ING3 seen in the human tissue samples, suggested that ING3 expression might be largely repressed in quiescent cells compared to growing cells. To test this idea we examined ING3 levels in hTERT-
HME and RWPE-1 cell lines, which are immortalized breast and prostate epithelial cell lines.
Cells were deprived of growth factors for 48 hr (to induce replicative quiescence) and serum containing growth factors were subsequently added to the media for 24 hr. As shown in Figure
16, subjecting lysates from quiescent and growing cells to western blot analyses confirmed that
ING3 is indeed expressed in growing cells but expression is significantly less in quiescent cells.
67
Figure 16. ING3 expression in growing and quiescent human epithelial cells. Human mammary fibroblasts (HME) immortalized with hTERT and prostate epithelial cells
(RWPE-1) immortalized with HPV18 were serum-starved for 48 hr and released by addition of serum. Cells were harvested 24 hr later and cell lysates were subject to western blot analysis.
68
3.1.3. ING3 promoter analysis
To acquire more insight into the signal transduction pathways that influence ING3 might be involved in, we estimated the bounds of the predicted ING3 core promoter sequence using the
Transcriptional Regulatory Element Database (TRED) (223) and analyzed this for possible transcription factor binding sites using TFSEARCH
(www.cbrc.jp/research/db/TFSEARCH.html) (224) and TFBIND (www.tfbind.hgc.jp) (225).
Table 3 shows the top ten transcription factors predicted to bind ING3 promoter based on the consensus sequence homology. Figure 17 indicates sites for the Runt-Related transcription factor
1 (RUNX1) (226,227), which showed 0.95 similarity to the consensus binding sequence as noted in Table 3, and plays an important role in hematopoietic cell development and that might be responsible for driving ING3 expression. Consistent with RUNX1 driving ING3 expression, analysis of the BloodChIP database revealed binding of RUNX1 and TCF7 to the ING3 promoter in CD34 blood progenitor cells, megakaryocytes, as well as SKNO-1 cells (Figure 18)
(228). Data mining of available ChIP-seq studies also confirmed that RUNX1 binds the ING3 promoter (229,230).
69
Table 3. Ten transcription factors predicted to bind the ING3 promoter
MATRIX Acc. no. column shows the transcription factor matrix ID from TRANSFAC R.3.4.
Percentage similarity depicts similarity between a sequence on ING3 promoter and the consensus sequence for the associated transcription factor. Only the top ten transcription factors predicted by both TFSEARCH and TFBIND are shown.
70
Figure 17. ING3 promoter and location of RUNX1 binding sites. The predicted promoter sequence was retrieved from TRED database The numbers above each binding site shows percentage similarity as reported by TFBIND to the RUNX1 consensus sequence. The red box shows the transcription start site.
71
Figure 18. BloodChIP database search for factors binding the ING3 locus. BloodChIP database is a collection of transcription factor bindings to different types of blood cells as determined in ChIP-seq studies. Peaks corresponding to RUNX1 and TCF7 transcription factor binding near the beginning of the ING3 gene are shown. Markers of open chromatin such as H3K4me3 and H3K27ac are shown and suggest the open chromatin conformation near ING3 transcription start site.
72
3.2. ING3 acts as a co-activator of AR
3.2.1.ING3 levels correlate with AR levels
Tissue microarray slides consisting of 265 PC patient tissue samples were stained with
ING3 monoclonal antibody (209) and were analyzed using automated quantitative immunofluorescence analysis (AQUA). As shown in Figure 19A, ING3 protein scores significantly correlated with AR levels. We also analyzed RNAseq reads of ING3 in a prostate adenocarcinoma cohort (n=550, TCGA Research Network, http://cancergenome.nih.gov/) and again found that ING3 levels significantly correlated with AR mRNA levels (Fig 19B).
73
Figure 19. ING3 levels correlate with androgen receptor (AR) levels in patient samples. A) Representative images of patient samples and their ING3 staining (red) that were classified as having high or low AR levels. Pancytokeratin (PCK) was used to mark the epithelial versus stromal cells in tissue cores. The graph shows ING3 AQUA score in the patient subgroups based on their AR expression (Mann Whitney Test ***P<0.001). AR expression of samples was determined by immunohistochemistry (data not shown) (211) B) A prostate adenocarcinoma cohort was retrieved from TCGA data portal. Levels of ING3 and AR mRNA was plotted in
Log2 scale.
74
ING3 protein and mRNA levels were variable in different AR-positive PC cell lines and were highest in VCaP (Fig 20A and 20B). We next asked if levels of ING3 respond to AR activity. VCaP, LNCaP and C4-2 cells were grown in media supplemented with charcoal- stripped serum for 48 hr and then treated with the androgen analog mibolerone (MB), the anti- androgen bicalutamide (Bic) or ethanol. C4-2 is an LNCaP subline isolated from androgen deprived animals and represents an advanced stage of prostate cancer with inducible levels of
AR (231). As shown in Figure 20C, both AR and ING3 protein levels increased in response to
MB and decreased in response to anti-androgen. The exception was C4-2 cells, where AR and
ING3 levels were low in unstimulated cells, and did not decrease in response to bicalutamide.
This may be due to the resistance of C4-2 to anti-androgens (231). Increased ING3 protein did not occur to a significant degree as a consequence of transcriptional induction as shown in the left panel of Figure 20D, using concentrations of MB that were effective in inducing expression of known AR-regulated genes (Fig 20D, right panel). To determine if ING3 protein was stabilized in response to androgen, de novo protein synthesis was blocked in LNCaP cells using cycloheximide. ING3 was stabilized in the presence of MB compared to cells grown in charcoal- stripped media, with its estimated half-life increasing from 1 hour to 4 hr (Fig 20E). These data indicate that ING3 protein is stabilized by androgen, which is consistent with the generally higher ING3 protein levels seen in AR-positive cells.
75
Figure 20. ING3 levels correlate with AR levels. A) Lysates from three AR-positive prostate cancer cell lines were subject to western blotting with antibodies against ING3, GAPDH and actin. B) mRNA levels of ING3 were normalized to
76
actin in three prostate cancer cell lines. C) LNCaP, C4-2 and VCaP cells were grown in media with charcoal stripped serum (CSS) for 48 hr and treated with mibolerone (MB) or bicalutamide
(Bic). Protein loadings were adjusted for each cell type to detect low protein expression. Levels of ING3 and AR were then visualized by western blotting with actin used as a loading control.
D) qRT-PCR study of ING3 in LNCaP cells after treatment with increasing concentrations of
MB. The left graph shows mRNA levels of ING3 in response to MB. The right graph shows mRNA levels of seven androgen regulated genes in response to mibolerone. E) A cycloheximide experiment using LNCaP cells grown in the presence or absence of MB to estimate ING3 protein half life.
77
3.2.2. ING3 activates the AR pathway
To ask if ING3 also affected AR activity, we transfected LNCaP cells with HA-tagged
ING3 expression construct and measured levels of the androgen regulated PSA gene. ING3 increased the levels of PSA in LNCaP cells (Fig 21A). Conversely, in C4-2 cells, efficient knock down of ING3 decreased PSA levels (Fig 21A). We next measured mRNA levels of the androgen-responsive genes PSA, TMPRSS2 and FKBP5 in response to ING3 overexpression.
ING3 selectively potentiated the effects of MB in LNCaP and C4-2 cells (Fig 21B). In the case of FKBP5, mRNA levels were affected individually by androgen treatment and ING3 overexpression. Knockdown of ING3 in LNCaP cells by ~90% (Fig 21C) decreased the PSA, and to a greater extent, FKBP5 responsiveness to MB (Fig 21D).
78
Figure 21. ING3 promotes androgen-induced gene expression. A) LNCaP cells transfected with empty vector or HA-ING3 were harvested 24hr later and lysates were blotted. Lysates of C4-2 cells transfected with siCtrl or siING3 for 48 hr were blotted for the indicated proteins. B) LNCaP or C4-2 cells transfected with GFP or ING3 constructs were
79
grown for 24hr +/- 10 nM MB and were tested for PSA, TMPRSS2 and FKBP5 expression
(ANOVA **P<0.01, ***P<0.001). C & D) LNCaP cells transfected with siCtrl or siING3 for
24hr were treated with 10 nM MB for 24 hr. Levels of androgen-regulated genes were assessed by qPCR (ANOVA ***P<0.001).
80
Because the FKBP5 gene was regulated most dynamically by ING3 and it functions in the AR pathway as one of the genes mostly regulated by androgens (232), we tested the effects of ING3 on AR recruitment to the FKBP5 androgen response element (ARE). MB increased AR recruitment dramatically, and this effect was abrogated upon ING3 knockdown with siRNA (Fig
22).
81
Figure 22. ING3 regulates AR recruitment to the FKBP5 gene androgen response element. C4-2 cells transfected with siCtrl or siING3 for 24hr were untreated or treated with 10 nM MB for 24hr. ChIP assays using α-AR antibody and AR-bound DNA used primers specific for an
ARE on the FKBP5 gene (t test *P<0.05).
82
ING3 differentially affected the ability of MB to induce the expression of different genes, probably resulting from different chromatin environments near the ARE elements of the genes.
To reduce this variable, we used an ARE-driven luciferase reporter construct in transient assays.
We first tested the efficiency of the luciferase reporter. As shown in Figure 23, when HEK293T cells where not transfected with Luciferase reporter construct (LUC), the recorded luminescence intensity was close to background in the presence or absence of MB, indicating that these cells are AR-negative and LUC is not activated by other mechanisms. When co-transfected with AR expression plasmid, however, the recorded intensity was increased dramatically in the presence of MB.
As shown in Figure 24A, knockdown of ING3 reduced the ability of MB to maximally induce expression from the ARE-driven construct while ING3 overexpression enhanced expression (Fig 24B). A time-course experiment with 1 nM MB (Fig 24C) indicates that ING3 affects AR transactivation as early as 2 hr after MB treatment and suggests ING3 directly functions in AR pathway.
ING3 targets TIP60 to actively transcribing genes via the ING3 PHD binding the
H3K4Me3 mark (233), to affect nearby chromatin structure. However, since AR activation by acetylation is thought to occur in the cytoplasm and induce AR translocation to the nucleus, we asked if the ING3 PHD was required for AR activation. A PHD deletion mutant was fully capable of stimulating the ARE-driven reporter in the absence of MB (Fig 24D), suggesting that activation of the AR by ING3 did not require the chromatin targeting capability of ING3.
83
Figure 23. Validation of an ARE-driven luciferase reporter system. HEK293T cells were transfected with 0.5 µg of the indicated plasmids along with β-gal plasmid as an internal control with or without 1 nM MB. After 48 hr, cells were lysed and luminescence was determined.
84
Figure 24. ING3 promotes AR transactivation. A) HEK293T cells were co-transfected with 1 µg AR3-tkk-LUC, 0.2 µg β-gal, 0.1 µg GFP and
20 nM of either siCtrl or siING3 for 48 hr +/- MB for the final 24 hr. LUC reporter activity was normalized to b-Gal (t Test **P<0.01). B) Cells were co-transfected with 1 µg AR3-tkk-LUC,
0.2 µg β-gal, 1 µg AR and either empty vector or indicated amounts of ING3 expression plasmid for 48 hr +/- MB (***P<0.001). C) Cells were co-transfected with 1 µg AR3-tkk-LUC, 0.2 µg β- gal, 1 µg AR and 0.2 µg of either empty vector (Ctrl) or full-length ING3 expression plasmid. 1 nM MB was added for the indicated time points (t Test *P<0.05, ***P<0.001). D) Cells were co- transfected with 1 µg AR3-tkk-LUC, 0.2 µg β-gal, 1 µg AR and either empty vector, full-length
85
ING3 expression plasmid or PHD deletion mutant in CSS medium for 48 hr. Levels of HA-ING3 expression were verified by western blotting and ARE-driven reporter expression is shown in response to full length and PHD-deleted ING3 (t Test *P<0.05).
86
3.2.3. ING3 interacts with AR
ING3 co-precipitated with AR, and MB increased their interaction, since higher levels of
AR protein were immunoprecipitated with ING3 after MB treatment of HEK293T cells transfected with HA-tagged ING3 and myc-tagged AR (Fig 25A). Addition of ethidium bromide
(EtBr) to the IP buffer did not alter this interaction (234), suggesting it was not dependent on
DNA (Fig 25B). Immunoprecipitation of endogenous ING3 in LNCaP cells +/- MB showed that
AR was detected in ING3 immunoprecipitates from MB-stimulated cells, and higher levels of
ING3 were pulled down in MB-treated samples (Fig 25C). This may be due to MB altering
ING3 protein complex conformation leading to increased exposure of the ING3 epitope to the antibody.
87
Figure 25. ING3 interacts with AR. A) HEK293T cells were co-transfected with 1 µg of Myc tagged AR expression construct and 1
µg of either empty vector or HA-tagged ING3 +/- 10 nM MB. ING3 was pulled down with HA- affinity beads and precipitates were blotted with α-AR. B) To determine the effects of DNA on the interaction, co-immunoprecipitations were repeated with addition of ethidium bromide
(EtBr). ING3 was precipitated using HA-affinity beads. C) Endogenous AR-ING3 interaction was assessed by co-precipitation in LNCaP cells +/- 10 nM MB for 24 hr. ING3 immunoprecipitation used α-ING3 covalently crosslinked to protein G beads. Increased amounts of ING3 protein were pulled down when cells were treated with MB, perhaps due to MB inducing conformational changes to expose the ING3 epitope recognized by the monoclonal antibody.
88
We next asked if the interaction of ING3 and AR was exclusively nuclear or also occurred in the cytoplasm. As shown in Figure 26, HEK293T cells were co-transfected with HA-
ING3 and AR plasmids in the absence of MB. Prior to IP, nuclear and cytosolic fractions were separated using the REAP protocol (235,236). We found that ING3 and AR interact in the cytoplasm in the absence of MB. Our results, however cannot exclude the existence of nuclear interaction as little unbound AR protein resides in nucleus.
89
Figure 26. ING3 and AR interact in the cytoplasm. HEK293T cells were co-transfected with AR and HA-ING3 in the absence of MB. 48 hr later, nuclear and cytosolic fractions were isolated and were subjected to IP using anti HA beads.
Western blotting was performed to detect the interaction. Lamin A and alpha-tubulin were used on input samples as nuclear and cytosolic markers, respectively.
90
To identify the AR domain that interacted with ING3, the AR deletion constructs shown in Figure 27A, were co-transfected with ING3 expression construct into HEK293 cells. As shown in Figure 27A, the N terminal domain of AR did not interact with ING3, whereas the N terminal 625 amino acids including the DNA binding domain, did. The hinge or ligand binding domains of the AR did not alter the interaction (data not shown). Full length and PHD-deleted versions of ING3 showed that interaction between the AR and ING3 or TIP60 and ING3 was not dependent on the PHD (Fig 27B).
91
Figure 27. ING3 interacts with the DNA-binding domain of AR independent of the PHD. A) Deletion constructs used to map domains required for interaction. Co-precipitation study using HEK293T cells co-transfected with AR deletion mutants and HA-ING3. FLAG beads were used to immunoprecipitate AR constructs and ING3 was detected in IP samples by western blotting. B) ING3 deletion mutant lacking the PHD. Co-precipitation using HEK293T cells co- transfected with full length AR, ING3 and an ING3 deletion mutant. HA beads were used to
92
immunoprecipitate ING3 constructs. The interacting AR and endogenous TIP60 were detected by western blotting with their respective antibodies.
93
3.2.4. ING3 promotes TIP60 function and interaction with AR
Immunoprecipitates of endogenous TIP60 recovered AR in an MB-sensitive manner as previously reported (Fig 28A). When cells were co-transfected with increasing amounts of ING3, increasing amounts of AR protein were immunoprecipitated with HA-tagged TIP60 and vice versa (Fig 28B). ING3 increased association between AR and TIP60 by ~8-fold (Fig 28B) while siING3 reduced association between AR and TIP60 by ~3-fold (Fig 28C). An ARE luciferase assay showed that siING3 largely abrogated TIP60 effects on AR transactivation in the absence or presence of MB (Fig 28D), consistent with ING3 facilitating interaction between TIP60 and
AR.
94
Figure 28. ING3 promotes TIP60-AR interaction.
95
A) TIP60 was immunoprecipitated in C4-2 cells grown +/- 10 nM MB for 24 hr. B) Cells were co-transfected with AR, TIP60 and increasing amounts of ING3 plasmid + 10 nM MB for 24 hr.
AR or TIP60 were immunoprecipitated using α-AR or HA-affinity beads, respectively. The graph shows the average ratio of AR:HA-TIP60 in three independent experiments (t Test
**P<0.01. C) C4-2 cells transfected with siCtrl or siING3 were treated for 24 hr with MB, immunoprecipitated with α-TIP60 and blotted with α-TIP60 or α-AR. The graph shows the average of TIP60:AR ratio (t Test *P<0.05). D) ARE-reporter activity of cells co-transfected with 1 µg AR3-tkk-LUC, 0.2 µg β-gal, 0.2 µg AR, 50 ng of TIP60 and 20 nM of either siCtrl or siING3 for 48 hr +/- 1 nM MB for the final 24 hr (t Test *P<0.05).
96
3.2.5. ING3 promotes AR acetylation and nuclear translocation
Since TIP60 acetylates AR (71), we asked whether ING3, a member of the TIP60 complex, affected AR acetylation. Overexpression of ING3 increased AR acetylation, as estimated by probing AR immunoprecipitates for acetyl lysine (Fig 29A). The in vitro acetylation assay shown in Figure 29B in which co-immunoprecipitated HA-ING3-TIP60 was added to AR-IP samples plus acetyl CoA, confirmed that ING3 complexes acetylated the AR. In the complementary experiment shown in Figure 29C, knockdown of ING3 dramatically decreased the ability of TIP60 to acetylate AR. Transfection with AR acetylation mutants K630R and K632/33R (72) completely abrogated the effect of ING3 compared to transfection with wildtype AR further indicating that AR activation by ING3 is dependent on the acetylation of
AR (Fig 29D).
97
98
Figure 29. ING3 affects TIP60-mediated acetylation of AR. A) HEK293T cells were co-transfected with 1 µg of AR construct or vector +/- 0.2 µg ING3 and were immunoprecipitated with α-Myc and AR acetylation was determined. The graph shows the average ratio of Ac-AR:total AR (t Test *P<0.05). B &C) HEK293T cells were co-transfected with AR-Myc, ING3-HA or GFP (A) or with AR-Myc, siCtrl or siING3 (B). AR was immunoprecipitated with α-Myc. ING3 was immunoprecipitated with α-HA in GFP or ING3- transfected cells (B). TIP60 was immunoprecipitated in siCtrl or siING3-transfected cells (C). In vitro acetylation assays were performed using 1 mM of Acetyl CoA. C) The graph (right panel) shows the average ratio of Ac-AR:total AR (t Test *P<0.05). D) Cells were co-transfected with 1
µg AR3-tkk-LUC, 0.2 µg β-gal, 0.2 µg of either vector or ING3 and 0.2 µg of either wildtype or mutant AR constructs for 48 hr +/- 0.1 nM MB for 24 hr.
Since it has been reported that acetylation of AR contributes to its nuclear translocation
(72), we tested whether ING3 affects AR localization. Overexpression of ING3 in LNCaP cells increased AR staining intensity in nuclei (Fig 30), while its knockdown inhibited MB-induced translocation of the AR to the nucleus as shown by immunofluorescence experiments and the fractionation assay done using the REAP protocol (Fig 31A-C) (235,236).
99
Figure 30. ING3 overexpression alters AR localization. LNCaP cells were transfected with GFP or HA-ING3, left untreated or treated with 10 nM MB for 4 hr and stained with α-AR and α-HA antibodies. DAPI was used to stain DNA to indicate nuclei.
100
Figure 31. ING3 knockdown reduces AR nuclear translocation. A & B) Cells transfected with siRNA for 48 hr were treated with ethanol or 10 nM MB for 24 hr and were fixed and stained for DNA with DAPI and for AR. Three fields of 35 cells each were examined visually for nuclear staining of AR using a blind experimental design (t Test
***P<0.001). C) LNCaP cells were transfected with siCtrl or siING3 for 48 hr. After 2 hr of treatment with 1 nM MB, nuclear and cytoplasmic fractions were blotted to detect the indicated proteins. Lamin A and α-tubulin was used as nuclear and cytoplasmic markers, respectively.
101
3.2.6. ING3 affects prostate cancer proliferation and migration
Since higher expression of ING3 correlated with shorter survival times (see section 3.3) , we asked if ING3 promoted growth by knocking down ING3 in LNCaP, PC3 and DU145 cells.
Staining cells 10 days later with crystal violet showed that knockdown decreased proliferation of all cancer lines tested (Fig 32).
102
Figure 32. ING3 affects prostate cancer cell proliferation. LNCaP, DU145 and PC3 were seeded at equal density, transfected with siCtrl or siING3 and 10 days later were fixed and stained with Crystal Violet to visualize all cells.
103
Alamar Blue assays confirmed that proliferation rates decreased upon ING3 knockdown in LNCaP and C4-2 cells (Fig 33A). This was corroborated in separate experiments by counting
LNCaP cells 3 and 7 days post-transfection of siING3 (Fig 33B). The possibly more physiologically relevant colony forming capability of PC cells in soft agar was also reproducibly reduced upon ING3 knockdown, with average calculated volumes of siING3-transfected colonies being less than half of those for cells transfected with siCtrl (313±50 vs. 153±16 mm3,
Fig 33C). Cells transfected with siING3 or siCtrl were stained for the presence of Ki-67 and confirmed that ING3 knockdown reduced proliferation (Fig 33D). LNCaP cells infected with doxycycline-inducible lentiviral shING3 also showed reduced Ki67 staining compared to shCtrl
(Fig 33E).
104
Figure 33. ING3 affects proliferation in AR-positive prostate cancer cells. A) LNCaP and C4-2 were grown in media supplemented with 5% CSS for the indicated times and cellular proliferation was assessed using Alamar Blue (t Test *P<0.05, **P<0.01). B)
LNCaP cells were transfected with either siCtrl or siING3 for 48 hr and seeded at equal density in 96-well plates in media supplemented with 5% CSS. Cells were counted at the indicated times
(t Test ***P<0.001). C) 48 hr after transfection with siCtrl or siING3, LNCaP cells were seeded at equal density in 24 well plates in media containing 20% FBS and 0.3% agarose and incubated to allow gels to solidify. Photomicrographs were taken after 10 days and the diameter of colonies
105
was measured. Representative images of siCtrl and siING3-transfected colonies are shown. D)
Immunofluorescence images of LNCaP cells transfected with either siCtrl or siING3 for 48 hr and stained for Ki67 (green) as a proliferation marker. E) LNCaP cells were infected with shCtrl or shING3 lentiviral particles at low multiplicity of infection. Dox was added to induce the expression of shRNA and RFP. Cells were then grown in CSS medium for 72 hr and stained for
Ki67 as a proliferation marker. RFP expression was used as a marker of infected cells. Arrows highlight infected cells that generally did (shCtrl) or did not (shING3) stain for Ki67. The shown infected cells are representation of minimum of 30 cells per condition.
106
To further ask how higher ING3 levels might increase PC recurrence and decrease survival, we examined cell motility. LNCaP cells transfected with siCtrl or siING3 and treated with 1 nM MB, were stained with Alexa 568-conjugated phalloidin to show filopodia, a measure of cell motility (highlighted by arrows in Fig 34). We then determined the effect of ING3 on cell mobility using a transwell migration assay. Transwell migration assays in the absence and presence of MB (Fig 35A & B) and wound-healing assays (Fig 35C) confirmed that ING3 knockdown reduced cell migration.
107
Figure 34. ING3 knockdown reduces AR-mediated filopodia formation. LNCaP cells were transfected with siCtrl or siING3 and treated with 1 nM MB for 72 hr, then fixed and stained with Texas Red-conjugated phalloidin and DAPI for nuclei. Arrows highlight actin projections consistent with filopodia formation. The cells shown here are representation of minimum of twenty cells per condition analyzed.
108
Figure 35. ING3 knockdown reduces prostate cancer cell migration. A) LNCaP cells were transfected with either siCtrl or siING3 and where indicated, treated with 1 nM MB for the times indicated. Transwell migration assays were performed as mentioned in the
Materials and Methods. Representative images are shown. B) Images were taken from six random fields for each condition and cells were counted manually on a computer using a blind
109
experimental protocol (t Test *P<0.05, **P<0.01). C) Wounds were made in monolayers of C4-2 cells stably expressing either shCtrl or shING3 in the presence of 10 nM MB and Dox to induce shRNA expression. Wounds were then allowed to heal during a course of four days. Images were taken from the same fields for each condition. Percentage of healed wound was then calculated based on pixels observed in each condition.
110
3.3. ING3 is a novel prognostic biomarker for prostate cancer
3.3.1. ING3 levels correlate with Gleason score in low AR subgroup
Examination of ING3 levels and clinico-pathological parameters of our PC cohort indicated a significant trend towards higher ING3 level with increasing Gleason score in a subset of patients with low expression of AR (Fig 36A). Similarly, using TCGA prostate cohort data we found that ING3 levels in low AR subgroup of patients correlated with the higher Gleason score
(Fig 36B).
111
Figure 36. ING3 levels correlate with Gleason score in AR low subset. A) ING3 protein scores in patient samples with low AR levels were graphed based on the
Gleason Score (GS) (Mann Whitney test) (n= 48, 56, 133 for GS <7, GS =7 and GS >7, respectively) (Mann Whitney Test **P<0.01 and *P<0.05). B) Transformed mRNA levels of
ING3 from TCGA prostate adenocarcinoma cohort in the low AR subgroup of patients (AR mRNA levels < 1969, which is the mean of AR levels in normal adjacent tissues.) was graphed based on the Gleason Score (Mann Whitney test) (n= 35, 162, 149 for GS <7, GS =7 and GS >7, respectively) (Mann Whitney Test **P<0.01).
112
3.3.2. ING3 levels predict prostate cancer patient overall survival and recurrence
We next asked if ING3 had any prognostic value in prostate cancer and to address this, we split the cohort into derivation and validation datasets (237). This allowed us to modify our statistical model in the derivation cohort and test it on another set of patients. We tested the validity of these two datasets by performing a Kaplan-Meier analysis using Gleason score as the known predictor (Fig 37).
113
Figure 37. Validation of datasets based on Gleason score. Kaplan Meier survival curves using Gleason score as a known prognostic marker in derivation and validation datasets (Log rank test).
114
ING3 AQUA scores were dichotomized using a 1.66 cutoff point based on the derivation dataset, for testing in the validation dataset. Analysis by Kaplan Meier based on cohort AR status showed that higher ING3 levels correlated strongly with poorer overall survival in patients with low AR levels (Fig 38). In low AR patients, ING3 predicted survival better than Gleason score (p
= 0.010 vs p=0.099 in the validation dataset), but Gleason score predicted survival better in patients with high AR levels (Data not shown).
115
Figure 38. Kaplan Meier curves for patient overall survival based on ING3 levels. Kaplan Meier survival curves in our prostate cancer patient cohort in the low AR subgroup in derivation and validation datasets (Log rank test).
116
ING3 levels were also useful in predicting the hazard function using Cox proportional hazard analysis (238). Factors such as AR levels, age, Gleason score, occurrence of CRPC and
ERG were taken into consideration in the multivariate analyses (Table 3 & 4). The contribution of ING3 in predicting hazard function was significant in both tested datasets with hazard ratios of 3.309 and 2.571, respectively. Testing the regression coefficients of Cox models on the two datasets indicated that ING3 coefficients were not significantly different, implying that ING3 contributes similarly to the outcome prediction and was independent of patient dataset.
Perhaps more clinically relevant, when Cox regression analyses +/-ING3 were performed and the likelihood ratios (LR) compared, ING3 significantly improved the Cox model in prediction of the hazard function (ΔLR= 5.075, p-value=0.024273, ΔLR= 3.941, p-value= 0.047123 for derivation and validation datasets, respectively). This shows that addition of ING3 protein levels to currently used clinical and pathological parameters would provide more accurate prognosis in primary PC. As a complementary study, we stratified the TCGA prostate cohort based on the maximum mRNA expression of ING3 in normal adjacent tissues. We investigated the correlation of ING3 levels with biochemical recurrence rate, as determined by the rise of PSA after surgery or ADT. Consistent with our TMA study, we found higher mRNA levels of ING3 were correlated with more rapid recurrence (Fig 39).
117
Table 3. Cox proportional hazard model for derivation dataset Hazard Covariate Coefficient SE p-value 95% CI Ratio
ING3 1.197 0.52 0.021 3.309 1.193-9.173
Age -0.36 0.031 0.247 0.965 0.908-1.025
CRPC 1.778 0.541 0.001 5.915 2.047-17.090
Gleason Score 2.426 0.731 0.001 11.31 2.699-47.393
AR expression -0.314 0.532 0.555 0.73 0.257-2.073
ERG expression -1.768 0.697 0.011 0.171 0.044-0.669
The listed clinical and pathological parameters were used along with ING3 as variables for multivariate analysis in the derivation dataset. ING3 is a significant marker, which improves the
Cox model to predict hazard function.
118
Table 4. Cox proportional hazard model for validation dataset Covariate Coefficient SE p-value Hazard Ratio 95% CI
ING3 0.944 0.471 0.045 2.571 1.022-5.468
Age 0.016 0.026 0.538 1.016 0.966-1.069
CRPC 1.067 0.494 0.031 2.906 1.102-7.659
Gleason Score 1.709 0.714 0.017 5.524 1.363-22.383
AR expression 0.549 0.534 0.305 1.731 0.607-4.934
ERG expression 0.556 0.535 0.299 1.743 0.611-4.976
Similar to Table 3, multivariate analysis was performed with known clinico-pathological parameters along with ING3 in the validation dataset. ING3 retains its significant contribution to the Cox model, indicating its cohort-independent effect.
119 Survival of Data 1:Survival proportions Cutoff is 601 which is mean of ING3 raw_count in normal adjacent tissue
Time: Days to biochemical recurrence-first
100 Low ING3 High ING3 80
60
survival 40
20 p = 0.007
Biochemical recurrence free free recurrence Biochemical 0 0 500 1000 1500 2000 Time to biochemical recurrence (Days)
Figure 39. Kaplan Meier curve for patient biochemical recurrence rate based on ING3 levels. TCGA RNAseq prostate adenocarcinoma cohort was stratified as low and high ING3 using the average mRNA levels of ING3 in normal adjacent tissue samples. Kaplan Meier analysis was performed for biochemical recurrence rate as determined by the rise of PSA levels above castrate levels. Consistent with our study on ING3 protein levels, higher levels of ING3 mRNA positively correlate with biochemical recurrence rate (Log rank test).
120
CHAPTER FOUR: DISCUSSION
121
4.1. ING3 profiling in normal human tissues
Antibodies have been always a concern in molecular biology, due to the frequency of non-specific binding and variable signal-to-noise ratio. Particularly with less studied and less abundant proteins, this concern is more dramatic. Our initial attempts to examine ING3 levels using a commercial antibody were problematic and therefore we decided to generate a new monoclonal antibody for ING3 and tested it in various settings to confirm the specificity. Indeed, many studies have reported on the ineffectiveness of numerous antibodies and lack the proper controls to detect the protein or proteins that the antibody is thought to recognize (239-242).
While this is less of a problem in western blotting since protein size can be confirmed, when an antibody recognizes proteins other than its target, interpretation of immunofluorescence, immunohistochemistry and immunoprecipitation results can be very misleading (242,243). We have been previously successful in generating well characterized monoclonal antibodies (CAbs1-
9) against INGs (221,244,245), which have been used extensively in numerous applications such as western blotting, immunoprecipitation and tissue microarray (TMA) staining (190,191). These antibodies are commercially available for other investigators.
We generated and optimized a new mouse monoclonal antibody against ING3. We confirmed its specificity toward ING3 across 5 INGs indicating a lack of cross-reactivity. Using
HEK293T cell lysates transfected with siCtrl or siING3, we confirmed the observed band as
ING3 with the predicted molecular weight. We also performed western blot analyses using lysates from mouse tissues. The ING3 band came as doublets and the lower band was significantly weaker, particularly in mouse cell line lysates (see the very faint lower band in
NIH-3T3 in Fig. 2B). Knockdown of ING3 in mouse cell lines confirmed that the bands were indeed mING3. The appearance of doublet in most mouse tissues suggests a post-translational
122
modification (246) or a shorter isoform in mouse but not in human transcriptome. In fact, according to NCBI gene database, there are two transcript variants of mouse ING3 encoding two proteins with 410 and 421 amino acids that do not exist in human ING3 and may therefore explain the doublet seen in mouse tissues and cells. The ING3 knockout mouse is embryonically lethal and mouse embryonic fibroblasts are challenging to isolate (personal communications), which makes the proper controls for this experiment difficult to obtain.
Our immunofluorescence results with ING3 overexpression indicated that the antibody recognizes the overexpressed protein properly. Staining was absent in our no primary antibody controls and when the antibody was preabsorbed with overexpressed ING3. When we knocked down ING3 and performed immunostaining, almost all staining was eliminated.
Using this new antibody, we present the first analysis of ING3 protein staining in normal human tissues and find that expression of ING3 strongly correlates with rapid cell growth. The distribution of expression we find is distinct from levels of ING3 mRNA previously reported in human and rodent tissues (136,210), which might reflect active post-transcriptional/translational regulation of ING3 protein levels. This is consistent with the fact that ING3 undergoes various post-translational modifications (246) that could affect protein half-life under different growth conditions. We then compared our results with the results published in the Human Protein Atlas
(http://www.proteinatlas.org/ENSG00000071243-ING3/tissue#gene_information). The staining of duodenum and skin were reported to be weak but more than almost all other tissues (excluding lymph nodes) which is consistent with our study. Albeit weak, the staining of lung epithelium and prostate seem to be more intense than what we found. Despite the very high mRNA expression of ING3 in bone marrow, no staining was reported in this tissue in the Human Protein
Atlas. The antibody used in this study was a Rabbit polyclonal antibody, which might explain the
123
discrepancies between these two studies. The datasheet for that antibody (Sigma Aldrich
HPA067388) was not very informative and the conclusion for data reliability is stated as
"uncertain" in the human protein atlas, which raises the question of whether their antibody was appropriate to use in IHC.
Using our monoclonal antibody, we found that ING3 staining is the most intense in skin epithelium and small intestine. Relatively high ING3 levels in skin might reflect a role for ING3 in UV-induced apoptosis and the loss of ING3 reported to occur in melanoma (171,204).
Recently, ING3 was also identified as one of the genes that play a role in pigmentation in
Holstein cattle (247), which is consistent with our finding of high ING3 levels in skin. It would be interesting to co-stain skin samples with ING3 and a marker of melanocytes to determine whether ING3 is expressed exclusively in melanocytes. Loss of the high levels of ING3 we observe in the small intestine might also contribute to colorectal cancer progression since a significant loss of ING3 has been reported in this form of cancer (208), although it will be important to determine whether reports of loss of ING3 in a particular cancer are reproducible given the discrepancies between ING3 mRNA and protein levels.
In our analyses, some discrepancies were found among tissue replicates. Here we showed representative images of each tissue. However, one core of esophagus and one core of lymph node stained at very high levels (Data not shown). The fact that most of the tissues were from autopsies and thus pathological data was lacking, we could not track such phenomena to any particular physiological or pathological events. However, this suggests that further investigation with regard to the role of ING3 in such tissues under certain conditions may lead to a better understanding of the function of ING3 in particular diseases.
We note that ING3 levels correlate well with the degree of predicted growth rate in
124
human tissues examined, with growth rate positively correlated with ING3 levels. This relationship is also seen in normal epithelial cells in culture. These data are consistent with ING3 playing a positive role in regulating cell growth, perhaps during S phase, as part of the
Tip60/NuA4 HAT complex (152) that is required for the transcriptional activation of histone promoters needed for replication and proper nucleosomal packaging (248). Our findings here are consistent with our previous report that the knockout of yng2, the yeast homolog of ING3 causes major growth defects in yeast (148). Recently ING3 was shown to be correlated with cardiomyocyte hypertrophy also consistent with ING3 playing a role in cell growth (210). Our collaborator's finding that ING3 knockout in the mouse is embryonically lethal and the size of the embryo is significantly smaller than wildtype, further suggests a pivotal role for ING3 in embryonic and cellular proliferation in early stages of development.
Our in silico analysis of the predicted ING3 promoter suggested binding of several transcription factors involved in various cellular processes. CREB was previously reported to regulate the transcription of ING3 in C4-2 prostate cancer cells (249), which is consistent with our analysis. The other predicted transcription factors need to be experimentally tested as the computational algorithms do not account for flanking sequence nor for non-consensus binding elements. However, the emergence of available public datasets such as NCBI GEO provides a useful tool for hypothesis generation and experimental design. We looked at the available gene expression and ChIP datasets for transcription factors of interest. The binding of RUNX1 on the
ING3 promoter was found by data-mining of public ChIP-seq data on different blood cells. High
ING3 staining in bone marrow (but not in more differentiated hematopoietic tissues) correlates well with the role of RUNX1 in hematopoietic progenitor cells. RUNX1 plays a critical role in hematopoiesis, as its homozygous deletion in mice causes severe defects in fetal liver
125
hematopoiesis leading to death at E12.5 (250). Whether ING3 is indeed involved in hematopoietic development through RUNX1 remains to be addressed in depth.
The top predicted transcription factor binding site in the ING3 promoter was E2F2. The
E2F family play key roles in regulating the cell cycle. We looked at available microarray data between small intestine villi and crypts of either wildtype or e2f1-3 knockout mice. Using the
GEO2R tool from NCBI, we noted a reduction of ing3 in e2f knockout mice compared to wildtype (251). Thus, it is likely that ING3 is involved as an effector in the E2F pathway, particularly in the small intestine that undergoes rapid proliferation, which is consistent with high levels of staining we observed. As a well-established regulator of the cell cycle that directly affects growth, this observation may link the function of ING3 and ING1a, the latter of which inhibits cell growth by affecting Rb (190).
In summary, we show the normal distribution of ING3 protein expression in representative human tissues, using a well-characterized and validated immunological reagent.
We also note that ING3 levels correlate well with the degree of growth rate in normal human tissues suggesting, together with its correlation with key genes such as E2F2 and RUNX1, a role in regulating cellular proliferation.
4.2. ING3 is a co-activator of AR in prostate cancer
In the present study, we found that the protein levels of ING3 and AR correlate in a cohort of prostate cancer patient biopsies. Consistently, mRNA levels of ING3 and AR from
TCGA data portal correlate significantly. Using three AR positive cell lines, LNCaP, C4-2 and
VCaP, we found ING3 protein is highest in VCaP, which is a highly metastatic prostate cancer cell line derived from vertebrae metastasis. It is also considered as a metastatic model of prostate cancer in which the AR is amplified and fusion of the ERG gene to TMPRSS2 androgen-
126
regulated promoter, which often occurs in prostate cancer, is present (252,253). Higher levels of
ING3 in this cell line were consistent with a previous report showing higher levels of ING3 in
ERG positive prostate cancer samples (212).
We altered the levels and function of AR protein by treatment with androgen analog
(MB) or anti-androgen (Bic). Unlike the other two cell lines, ING3 levels did not decrease in response to Bic in C4-2 cells. C4-2 is a derivative of LNCaP cells that was isolated and grown under conditions of androgen ablation. Although these cells still retain AR activity and responsiveness, they are resistant to anti-androgen treatment. Unlike the canonical pathway, the
NTD of AR was shown to be critical for AR activation and resistance to anti-androgens in C4-2 cells (254).
Consistent with the androgen responsive gene database (255), we found that ING3 was not transcriptionally regulated by AR. Also, in our analysis of the ING3 promoter, no androgen response element was found. In addition, ING3 localization in the nucleus and cytoplasm was not altered in LNCaP cells treated with androgen (data not shown) and similar finding was reported regarding the ING1 repressor role in AR pathway (199). Although ING3 was not affected through these mechanisms, we did find that ING3 protein is stabilized in the presence of androgen in LNCaP cells. This suggested that ING3 might be post-translationally modified in the context of activated AR. Many residues of ING3 have been found in mass spectrometry screens
(www.phosphosite.org) to be phosphorylated or acetylated. Ubiquitination of ING3 was also reported previously to be increased in melanoma cells by the SKP2 E3 ligase (205). Since both phosphorylation and acetylation can affect ubiquitination and thus protein stability, such modifications are likely to be involved in regulating the stability of the protein in prostate cancer, but further experiments are needed to address the exact mechanisms governing ING3 protein
127
stability.
In contrast to relatively modest effects of the AR on ING3, increasing and decreasing levels of ING3 showed that ING3 positively regulates AR activity, and that ING3 potentiates the effects of androgen on AR activity as estimated by activating expression of both endogenous
AR-regulated genes and an ARE-driven reporter construct. ING3 strongly affected FKBP5 levels in C4-2 and LNCaP and affected PSA in C4-2 cells. FKBP5 is one of the genes highly regulated by AR activation (256). In the absence of androgen, it forms a complex with AR and other proteins such as HSP70 and 90 to stabilize AR, keeping it ready for ligand to bind (257). It is shown to interact with AR in LNCaP cells and increase AR activity as determined by luciferase assay (258). Our ChIP results showed that ING3 affects AR recruitment to the FKBP5 ARE.
Based on these observations, we propose is that ING3 in the cytoplasm affects AR acetylation
(mentioned below), which in turn alters gene expression dependent on AR activation. Although our study on AR activation supports this hypothesis, the possibility of a distinct and possibly epigenetic role for nuclear ING3 in regulating FKBP5 AR-regulated expression cannot be overlooked and requires more investigation.
ING3 co-immunoprecipitated with AR and this interaction was enhanced by androgen.
Addition of ethidium bromide did not alter the interaction, indicating that it is not DNA- dependent -a common pitfall in interpretation of co-IP studies. This indicates that the interaction is either a direct physical interaction or the presence of ING3 and AR in a protein complex causes the co-IP. When we treated LNCaP cells with MB and immunoprecipitated ING3, significantly more ING3 was pulled down than in the absence of MB. We speculate this may be due to MB altering the conformation of the ING3 protein complex, leading to increased exposure of the ING3 epitope to the antibody. Since ING3 and AR (in the presence of its ligand) are
128
localized primarily in the nucleus, we sought to determine the cellular compartment in which the interaction happens. Using HEK293T cells transfected with ING3 and AR and in the absence of
MB, we found that the two proteins could interact in the cytoplasm. Our result, however does not exclude the possibility of a nuclear interaction in the active AR context. Rather, it indicates an interesting role of ING3 1) independent of its chromatin-remodelling function and 2) upstream in the AR pathway in the cytoplasm.
Using AR and ING3 deletion mutants, we found that ING3 interacts with the DNA binding domain (DBD) of AR and this interaction is not dependent on the PHD domain of ING3.
The AR DBD consists of two zinc fingers that are conserved across nuclear receptors. The first zinc finger contains a sequence called the P-box which is responsible for ARE recognition, while the second one contains a D-box, which is responsible for homo-dimerization (259). The AR nuclear localization signal (NLS) is located in DBD and hinge region, and activation leads to it becoming acetylated, leading to a conformational change that exposes the NLS. We previously found that ING1b co-immunoprecipitated with estrogen receptor alpha and modulated its activity via an Activation Function (AF2) domain (198). AF2 is a co-activator binding pocket and is critical for nuclear receptors to function properly. In addition, more recently ING1b was reported to act as a co-repressor of AR (199). These findings collectively indicate roles of the ING family in Nuclear Receptor biology.
The interaction between the AR and TIP60 has been previously reported. We were able to reproduce and confirm that this interaction occurs in LNCaP cells. ING3 increased this interaction in a dose-dependent manner and knockdown of ING3 decreased the precipitation of
AR and TIP60. However, we also found that knock down of ING3 slightly decreases the TIP60 protein levels, which is likely due to alteration of its stability. Similarly, overexpression of ING3
129
slightly increased the TIP60 protein levels. This observation was found in both IP samples and lysates, which is consistent with previous studies indicating ING3 functioning as an essential member of the TIP60 complex (151). When we corrected for this effect by normalization, we still found significant change in the amount of interaction. Hence, our results indicate that ING3 promotes the interaction of TIP60 and AR but also suggests critical contributions to the stability of the complex and its interactions with other regulatory molecules.
In contrast to studies showing that the ING proteins serve to target lysine acetyltransferase (KAT) and lysine deacetylase (KDAC) complexes to the H3K4Me3 mark via interaction of their PHD form of zinc finger (146,233,260), we observed that a C-terminal- deletion mutant of ING3 lacking the PHD increases AR transactivation to levels similar to seen with wild type ING3. A similar lack of function for the ING2 PHD domain in the differentiation of C2C12 myoblasts into myotubes was previously noted (159). While ING2 induced the expression of myogenein and subsequently promoted differentiation, a deletion construct of
ING2 lacking the PHD was active in transcriptional activation and was actually more effective in inducing differentiation. These cases of ING3 and ING2 being active in regulating the activity of their respective acetylation (TIP60) and deacetylation (Sin3A) complexes when lacking their
PHDs indicate that cellular levels of the INGs affect two separate pools of substrates for their complexes, chromatin associated and chromatin non-associated. Studies of ING4 also supports this idea since the ING4 PHD domain was reported to be necessary for prostate epithelial cell differentiation but not cell-death induction by overexpression (186).
AR acetylation by different KATs such as TIP60 and P300 is believed to be a rate- limiting step for AR nuclear translocation. As mentioned above, this acetylation is needed for the
NLS to be exposed and initiates the translocation process (71,72,83,261). Through a series of
130
experiments, we found that ING3 promotes AR acetylation and its effect on AR is diminished by mutating the lysines on the hinge region of AR. This effect was reflected by altered rates of AR translocation. Knockdown of ING3 delayed AR nuclear translocation in response to MB and overexpression of ING3 increased nuclear AR staining. These effects are all consistent with
ING3 contributing to regulate the localization of the AR, helping direct it to the promoters of
AR-regulated genes and to potentiate its effects on transcription
We found that ING3 knockdown significantly decreased the proliferation of prostate cancer cells. This effect was more dramatic on AR-positive cells such as LNCaP. In PC3 and
DU145 AR-negative cell lines, ING3 also induced a morphological change; the formed colonies were more disperse and heterogeneous. This effect was not observed with LNCaP cells, which suggests that the functional role(s) of ING3 may be context-dependent. Saeed et al. performed high throughput screening using VCaP cells, in which they knocked down genes using over 4000 pairs of siRNAs and grew the cells under normal and androgen-deprived conditions. They then asked which genes were negatively related to cellular proliferation or positively associated with apoptosis, specifically under androgen-deprived conditions. One of the genes they identified that had these effects was ING3 (213). This study is consistent with our findings with LNCaP and
C4-2 cells under androgen-deprived conditions, since knockdown of ING3 in these cells and under androgen-deprived conditions significantly reduced proliferation. The rationale behind using androgen deprivation for such experiments is its importance to clinical relevance. Current treatments for primary prostate cancer involve either surgical or chemical castration, which has a very good initial response rate. However, the tumors generally recur within 12-18 months following castration. Numerous studies have found that androgen deprivation therapy causes tumor selection in which the AR remains active, or hyperactivated by numerous mechanisms
131
including the overexpression of its co-activators. We propose at this point that ING3 constitutes a previously unknown co-activator, contributing to activation of AR pathway at minimal androgen levels, consequently impacting prostate cancer proliferation and thereby correlating with patient survival. The degree to which ING3 might be involved in androgen dependent proliferation requires more experiments, possibly using androgens as well as anti-androgens.
One good experiment to further test this proposal might be to stably overexpress ING3 in prostate cancer cells and treat these cells with anti-androgen. This could not be achieved in our setting due to the low transfection efficiency of LNCaP cells, but generation of regulatable lentiviral expression constructs may help overcome the transfection issue.
ING3 knockdown decreases androgen-induced migration as determined by wound healing and transwell assays. While the effects observed using these assays could be partially due to effects on proliferation, the reduction in the formation of filopodia suggests an alteration in migration process in addition to proliferation. Figure 40 summarizes our results regarding the interaction and effects of ING3 on the AR pathway in PC cells and how they fit into the current understanding of major aspects of the AR pathway.
132
Figure 40. Model of ING3 function in the AR pathway. After AR binds its ligand (DHT or MB), it dissociates from its co-chaperone complex (FKBP5,
HSP90, p23) and interacts with ING3 that directs AR acetylation by TIP60, promoting nuclear
133
translocation. Nuclear AR regulates gene expression promoting prostate cancer survival and proliferation. ING3 promotes AR recruitment to the FKBP5 ARE (and a subset of other genes) and increases FKBP5 mRNA levels. This generates a positive feedback loop making more
FKBP5 available to bind AR, thereby stabilizing the complex, which promotes additional binding to ligand. This pathway promotes proliferation and migration of AR-responsive cells, and in particular prostate cancer cells.
134
4.3. ING3 as a novel biomarker for prostate cancer
Since current diagnostic methods can fail to accurately predict aggressive versus latent forms of prostate cancer and ING3 was linked to AR levels and activity, we asked if ING3 levels could be used as a novel biomarker. One of the current pitfalls of biomarker discovery is the lack of validation in many biomarker studies. This is reflected by the dramatic difference between the number of proposed biomarkers in the literature and the number of biomarkers clinically used.
This ratio is actually less than 0.1% in 2011 (262,263). For this reason and because collection of additional patient specimens was not feasible in the timeframe of our study, we decided to use statistical methods to validate our biomarker findings (237). We split our patient data randomly into two datasets and confirmed its validity by using Gleason score as a conventional biomarker for prognosis. This allowed us to study ING3 as a biomarker in a cohort-independent way. We defined the ING3 cutoff point in the derivation dataset and tested it on the validation dataset. We found that the proposed cutoff point for ING3 was valid and independent of the patient data, and that the addition of ING3 to current pathological parameters such as ERG status and Gleason score significantly improved predictability of patient prognosis. The cutoff for ING3 was also observable by manually screening through the cores, which suggests that it may be applicable in regular pathology based laboratories.
In order to encourage more complete and transparent reporting in biomarker discovery, guidelines and checklists have been recommended to avoid misinterpretation of the data and to have the relevant information available for other researchers (264,265). Despite the fact that our introduction of ING3 as a candidate prostate cancer biomarker is in the early phases of research and the present study focuses primarily on the molecular mechanism underlying the role of ING3 in AR pathway, when possible, we adhered to recommendations of reporting the results and
135
statistical analyses. For example, we analyzed our data based on known markers such as Gleason score, and compared results to data obtained when adding ING3 as an additional marker. For our multivariate analyses, we reported hazard ratios together with confidence intervals for all covariates in the hazard model, which provides a sufficient level of transparency for other researchers to assess the value of ING3 as a prognostic marker.
Our analyses of primary prostate tumors showed that high levels of ING3 predicted poorer outcome in patients with low AR levels. A similar trend was observed when analyzing recurrence rate using TCGA data stratified based on AR levels (Log rank test, p= 0.009 in low
AR group based on the mean of AR in normal adjacent tissues, data not shown). This indicates that higher ING3 levels can compensate for low AR levels by activating AR, promoting PC growth. In the context of AR hyperactivation, ING3 may not be required in the process and may primarily function through gene regulation, for example by reducing apoptosis upon RSK- mediated suppression (249). Indeed, in our TMA study the mean ING3 AQUA score was significantly higher in CRPC than in primary prostate cancer, but only in the low AR subset of tissue cores (Mann Whitney test, p=0.006, data not shown). Epithelial-mesenchymal transition
(EMT) and invasion of PC cells were recently reported to be dependent on the levels of AR with lower levels of AR promoting androgen-induced EMT (266). This is in line with our findings that ING3 modulated cell migration and that higher ING3 levels correlated with poorer outcome in the subset of patients having tumors with low levels AR. However, these data need to be interpreted with caution, since effects of ING3 on migration may be influenced by effects on cell growth.
We recently found that high levels of ING3 also correlate with poorer survival in ERG negative PC (211). Although the interplay between ERG and AR remains unclear, several studies
136
have suggested that the ERG fusion protein inhibits AR expression and activity at several loci
(267,268) supporting the idea that ING3 can potentiate the activity of the AR pathway, particularly when AR inhibitory factors such as ERG are absent.
Our in vitro experiments indicate a role for ING3 in prostate cancer cellular proliferation, primarily under androgen-deprived conditions. We were unable to test this in our retrospective cohort; a subset of patients were undergone ADT and this might change the biology and function of ING3. To test whether ING3 can be a predictive biomarker, a prospective cohort is needed, where biopsies are taken from patients with primary prostate cancer before chemical castration.
With available follow up clinical data, we would then be able to determine whether levels of
ING3 can predict the aggressiveness of prostate cancer. This is of importance to investigate, as distinguishing between aggressive and latent prostate tumors is a current clinical challenge.
Many patients with latent disease undergo radical prostatectomy and this has adverse physiological and psychological consequences and also causes a financial burden on the medical system.
In this dissertation, we generated and validated a monoclonal antibody for ING3 and studied the value of ING3 protein as a novel biomarker in prostate cancer. Determining the protein levels of ING3 might also be useful in other diseases and contexts. For example,
Merenuik et al performed an siRNA-based screening study to identify genes that are synthetically lethal with polynucleotide kinase/phosphatase (PNKP) inhibitors. While knock down of ING3 together with scrambled siRNA showed marginal effect of cellular proliferation, knock down of both ING3 and PNKP resulted in a dramatic reduction in cellular proliferation
(269). Therefore, with our novel antibody and our methods, one could imagine, determining the
137
protein levels of ING3 could be clinically useful and serve as a surrogate to stratify patients that may benefit from therapeutics such as PNKP inhibitors.
In conclusion, in the present dissertation, we have generated and optimized a new antibody for ING3 protein and shown that ING3 regulates the AR pathway in prostate cancer, primarily by virtue of being a member of TIP60 complex. It promotes TIP60-AR interaction and affects acetylation of AR, a necessary step in activating the AR pathway. We studied and validated ING3 as a novel prognostic biomarker and found that it could dramatically improve our prediction of overall survival in prostate cancer. ING3, like other ING proteins, have been previously reported as a tumor suppressor in various types of cancer (161). To our knowledge, this is the first study indicating a role for ING3 as a proto-oncogene in prostate cancer, particularly in the AR-positive cells.
138
References
1. Adams J. The case of scirrhous of the prostate gland with corresponding affliction of the lymphatic glands in the lumbar region and in the pelvis. Lancet 1853;1:393. 2. Miller KD, Siegel RL, Lin CC, Mariotto AB, Kramer JL, Rowland JH, et al. Cancer treatment and survivorship statistics, 2016. CA: a cancer journal for clinicians 2016. 3. Irvine RA, Yu MC, Ross RK, Coetzee GA. The CAG and GGC microsatellites of the androgen receptor gene are in linkage disequilibrium in men with prostate cancer. Cancer Res 1995;55(9):1937-40. 4. Hsing AW, Gao YT, Wu G, Wang X, Deng J, Chen YL, et al. Polymorphic CAG and GGN repeat lengths in the androgen receptor gene and prostate cancer risk: a population- based case-control study in China. Cancer Res 2000;60(18):5111-6. 5. Zeigler-Johnson CM, Spangler E, Jalloh M, Gueye SM, Rennert H, Rebbeck TR. Genetic susceptibility to prostate cancer in men of African descent: implications for global disparities in incidence and outcomes. Can J Urol 2008;15(1):3872-82. 6. Hemminki K, Czene K. Attributable risks of familial cancer from the Family-Cancer Database. Cancer Epidemiol Biomarkers Prev 2002;11(12):1638-44. 7. Edwards SM, Kote-Jarai Z, Meitz J, Hamoudi R, Hope Q, Osin P, et al. Two percent of men with early-onset prostate cancer harbor germline mutations in the BRCA2 gene. Am J Hum Genet 2003;72(1):1-12. 8. Goh CL, Schumacher FR, Easton D, Muir K, Henderson B, Kote-Jarai Z, et al. Genetic variants associated with predisposition to prostate cancer and potential clinical implications. Journal of internal medicine 2012;271(4):353-65. 9. Hoffmann TJ, Van Den Eeden SK, Sakoda LC, Jorgenson E, Habel LA, Graff RE, et al. A large multiethnic genome-wide association study of prostate cancer identifies novel risk variants and substantial ethnic differences. Cancer discovery 2015;5(8):878-91. 10. Eeles RA, Olama AA, Benlloch S, Saunders EJ, Leongamornlert DA, Tymrakiewicz M, et al. Identification of 23 new prostate cancer susceptibility loci using the iCOGS custom genotyping array. Nat Genet 2013;45(4):385-91, 91e1-2. 11. Thomas G, Jacobs KB, Yeager M, Kraft P, Wacholder S, Orr N, et al. Multiple loci identified in a genome-wide association study of prostate cancer. Nat Genet 2008;40(3):310-5. 12. Allott EH, Masko EM, Freedland SJ. Obesity and prostate cancer: weighing the evidence. Eur Urol 2013;63(5):800-9. 13. Kenfield SA, Stampfer MJ, Giovannucci E, Chan JM. Physical activity and survival after prostate cancer diagnosis in the health professionals follow-up study. J Clin Oncol 2011;29(6):726-32. 14. Kenfield SA, Stampfer MJ, Chan JM, Giovannucci E. Smoking and prostate cancer survival and recurrence. Jama 2011;305(24):2548-55. 15. Jones MR, Joshu CE, Kanarek N, Navas-Acien A, Richardson KA, Platz EA. Cigarette Smoking and Prostate Cancer Mortality in Four US States, 1999-2010. Preventing chronic disease 2016;13:E51. 16. Shui IM, Wong CJ, Zhao S, Kolb S, Ebot EM, Geybels MS, et al. Prostate tumor DNA methylation is associated with cigarette smoking and adverse prostate cancer outcomes. Cancer 2016.
139
17. Jacobs ET, Kohler LN, Kunihiro AG, Jurutka PW. Vitamin D and Colorectal, Breast, and Prostate Cancers: A Review of the Epidemiological Evidence. J Cancer 2016;7(3):232- 40. 18. Xu Y, He B, Pan Y, Deng Q, Sun H, Li R, et al. Systematic review and meta-analysis on vitamin D receptor polymorphisms and cancer risk. Tumour Biol 2014;35(5):4153-69. 19. Stamey TA, Yang N, Hay AR, McNeal JE, Freiha FS, Redwine E. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med 1987;317(15):909-16. 20. Scosyrev E, Wu G, Mohile S, Messing EM. Prostate-specific antigen screening for prostate cancer and the risk of overt metastatic disease at presentation: analysis of trends over time. Cancer 2012;118(23):5768-76. 21. Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med 2003;349(3):215-24. 22. Andriole GL, Crawford ED, Grubb RL, 3rd, Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med 2009;360(13):1310-9. 23. Andriole GL, Crawford ED, Grubb RL, 3rd, Buys SS, Chia D, Church TR, et al. Prostate cancer screening in the randomized Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial: mortality results after 13 years of follow-up. J Natl Cancer Inst 2012;104(2):125-32. 24. Wolf AM, Wender RC, Etzioni RB, Thompson IM, D'Amico AV, Volk RJ, et al. American Cancer Society guideline for the early detection of prostate cancer: update 2010. CA: a cancer journal for clinicians 2010;60(2):70-98. 25. Jemal A, Fedewa SA, Ma J, Siegel R, Lin CC, Brawley O, et al. Prostate Cancer Incidence and PSA Testing Patterns in Relation to USPSTF Screening Recommendations. Jama 2015;314(19):2054-61. 26. Cui T, Kovell RC, Terlecki RP. Is it time to abandon the digital rectal examination? Lessons from the PLCO Cancer Screening Trial and peer-reviewed literature. Current medical research and opinion 2016:1-22. 27. Kantoff P, Carroll PR, D'Amico AV. Prostate cancer : principles and practice. Philadelphia: Lippincott Williams & Wilkins; 2002. xv, 748 p. p. 28. Epstein JI, Allsbrook WC, Jr., Amin MB, Egevad LL. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. The American journal of surgical pathology 2005;29(9):1228-42. 29. Harnden P, Shelley MD, Coles B, Staffurth J, Mason MD. Should the Gleason grading system for prostate cancer be modified to account for high-grade tertiary components? A systematic review and meta-analysis. Lancet Oncol 2007;8(5):411-9. 30. Steinberg DM, Sauvageot J, Piantadosi S, Epstein JI. Correlation of prostate needle biopsy and radical prostatectomy Gleason grade in academic and community settings. The American journal of surgical pathology 1997;21(5):566-76. 31. Allsbrook WC, Jr., Mangold KA, Johnson MH, Lane RB, Lane CG, Amin MB, et al. Interobserver reproducibility of Gleason grading of prostatic carcinoma: urologic pathologists. Hum Pathol 2001;32(1):74-80.
140
32. Epstein JI, Zelefsky MJ, Sjoberg DD, Nelson JB, Egevad L, Magi-Galluzzi C, et al. A Contemporary Prostate Cancer Grading System: A Validated Alternative to the Gleason Score. Eur Urol 2016;69(3):428-35. 33. Epstein JI, Egevad L, Amin MB, Delahunt B, Srigley JR, Humphrey PA. The 2014 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma: Definition of Grading Patterns and Proposal for a New Grading System. The American journal of surgical pathology 2016;40(2):244-52. 34. Stephenson AJ, Kattan MW, Eastham JA, Bianco FJ, Jr., Yossepowitch O, Vickers AJ, et al. Prostate cancer-specific mortality after radical prostatectomy for patients treated in the prostate-specific antigen era. J Clin Oncol 2009;27(26):4300-5. 35. Eggener SE, Scardino PT, Walsh PC, Han M, Partin AW, Trock BJ, et al. Predicting 15- year prostate cancer specific mortality after radical prostatectomy. J Urol 2011;185(3):869-75. 36. Coelho RF, Rocco B, Patel MB, Orvieto MA, Chauhan S, Ficarra V, et al. Retropubic, laparoscopic, and robot-assisted radical prostatectomy: a critical review of outcomes reported by high-volume centers. Journal of endourology / Endourological Society 2010;24(12):2003-15. 37. Murphy GP, Mettlin C, Menck H, Winchester DP, Davidson AM. National patterns of prostate cancer treatment by radical prostatectomy: results of a survey by the American College of Surgeons Commission on Cancer. J Urol 1994;152(5 Pt 2):1817-9. 38. Penson DF, McLerran D, Feng Z, Li L, Albertsen PC, Gilliland FD, et al. 5-year urinary and sexual outcomes after radical prostatectomy: results from the Prostate Cancer Outcomes Study. J Urol 2008;179(5 Suppl):S40-4. 39. Kuban DA, Tucker SL, Dong L, Starkschall G, Huang EH, Cheung MR, et al. Long-term results of the M. D. Anderson randomized dose-escalation trial for prostate cancer. International journal of radiation oncology, biology, physics 2008;70(1):67-74. 40. Zietman AL, Bae K, Slater JD, Shipley WU, Efstathiou JA, Coen JJ, et al. Randomized trial comparing conventional-dose with high-dose conformal radiation therapy in early- stage adenocarcinoma of the prostate: long-term results from proton radiation oncology group/american college of radiology 95-09. J Clin Oncol 2010;28(7):1106-11. 41. Zelefsky MJ, Pei X, Chou JF, Schechter M, Kollmeier M, Cox B, et al. Dose escalation for prostate cancer radiotherapy: predictors of long-term biochemical tumor control and distant metastases-free survival outcomes. Eur Urol 2011;60(6):1133-9. 42. Zelefsky MJ, Cowen D, Fuks Z, Shike M, Burman C, Jackson A, et al. Long term tolerance of high dose three-dimensional conformal radiotherapy in patients with localized prostate carcinoma. Cancer 1999;85(11):2460-8. 43. Incrocci L. Radiotherapy for prostate cancer and sexual health. Translational andrology and urology 2015;4(2):124-30. 44. Incrocci L. Sexual function after external-beam radiotherapy for prostate cancer: what do we know? Crit Rev Oncol Hematol 2006;57(2):165-73. 45. Pinkawa M, Gagel B, Piroth MD, Fischedick K, Asadpour B, Kehl M, et al. Erectile dysfunction after external beam radiotherapy for prostate cancer. Eur Urol 2009;55(1):227-34.
141
46. Huggins C, Hodges CV. Studies on prostatic cancer. I. The effect of castration, of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. CA: a cancer journal for clinicians 1972;22(4):232-40. 47. Gleave ME, Goldenberg SL, Jones EC, Bruchovsky N, Sullivan LD. Biochemical and pathological effects of 8 months of neoadjuvant androgen withdrawal therapy before radical prostatectomy in patients with clinically confined prostate cancer. J Urol 1996;155(1):213-9. 48. Fair WR, Cookson MS, Stroumbakis N, Cohen D, Aprikian AG, Wang Y, et al. The indications, rationale, and results of neoadjuvant androgen deprivation in the treatment of prostatic cancer: Memorial Sloan-Kettering Cancer Center results. Urology 1997;49(3A Suppl):46-55. 49. Gleave ME, Bruchovsky N, Moore MJ, Venner P. Prostate cancer: 9. Treatment of advanced disease. CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne 1999;160(2):225-32. 50. Lawton CA, Winter K, Murray K, Machtay M, Mesic JB, Hanks GE, et al. Updated results of the phase III Radiation Therapy Oncology Group (RTOG) trial 85-31 evaluating the potential benefit of androgen suppression following standard radiation therapy for unfavorable prognosis carcinoma of the prostate. International journal of radiation oncology, biology, physics 2001;49(4):937-46. 51. Tomita N, Soga N, Ogura Y, Hayashi N, Kageyama T, Ito M, et al. High-dose radiotherapy with helical tomotherapy and long-term androgen deprivation therapy for prostate cancer: 5-year outcomes. J Cancer Res Clin Oncol 2016;142(7):1609-19. 52. Sweeney CJ, Chen YH, Carducci M, Liu G, Jarrard DF, Eisenberger M, et al. Chemohormonal Therapy in Metastatic Hormone-Sensitive Prostate Cancer. N Engl J Med 2015;373(8):737-46. 53. Debruyne FM, Denis L, Lunglmayer G, Mahler C, Newling DW, Richards B, et al. Long- term therapy with a depot luteinizing hormone-releasing hormone analogue (Zoladex) in patients with advanced prostatic carcinoma. J Urol 1988;140(4):775-7. 54. Zoladex. In RxTx. 142 63. Levine GN, D'Amico AV, Berger P, Clark PE, Eckel RH, Keating NL, et al. Androgen- deprivation therapy in prostate cancer and cardiovascular risk: a science advisory from the American Heart Association, American Cancer Society, and American Urological Association: endorsed by the American Society for Radiation Oncology. Circulation 2010;121(6):833-40. 64. New trends in androgen deprivation therapy: Summary of key research presented at AUA 2014. Canadian Urological Association journal = Journal de l'Association des urologues du Canada 2014;8(7-8 Suppl 4):S134-6. 65. Scher HI, Halabi S, Tannock I, Morris M, Sternberg CN, Carducci MA, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol 2008;26(7):1148-59. 66. Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi KN, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004;351(15):1502-12. 67. Alen P, Claessens F, Verhoeven G, Rombauts W, Peeters B. The androgen receptor amino-terminal domain plays a key role in p160 coactivator-stimulated gene transcription. Mol Cell Biol 1999;19(9):6085-97. 68. Wong CI, Zhou ZX, Sar M, Wilson EM. Steroid requirement for androgen receptor dimerization and DNA binding. Modulation by intramolecular interactions between the NH2-terminal and steroid-binding domains. J Biol Chem 1993;268(25):19004-12. 69. Zhu P, Baek SH, Bourk EM, Ohgi KA, Garcia-Bassets I, Sanjo H, et al. Macrophage/cancer cell interactions mediate hormone resistance by a nuclear receptor derepression pathway. Cell 2006;124(3):615-29. 70. Edwards A, Hammond HA, Jin L, Caskey CT, Chakraborty R. Genetic variation at five trimeric and tetrameric tandem repeat loci in four human population groups. Genomics 1992;12(2):241-53. 71. Gaughan L, Logan IR, Cook S, Neal DE, Robson CN. Tip60 and histone deacetylase 1 regulate androgen receptor activity through changes to the acetylation status of the receptor. J Biol Chem 2002;277(29):25904-13. 72. Fu M, Rao M, Wang C, Sakamaki T, Wang J, Di Vizio D, et al. Acetylation of androgen receptor enhances coactivator binding and promotes prostate cancer cell growth. Mol Cell Biol 2003;23(23):8563-75. 73. van Royen ME, Cunha SM, Brink MC, Mattern KA, Nigg AL, Dubbink HJ, et al. Compartmentalization of androgen receptor protein-protein interactions in living cells. J Cell Biol 2007;177(1):63-72. 74. Xu J, Qiu Y, DeMayo FJ, Tsai SY, Tsai MJ, O'Malley BW. Partial hormone resistance in mice with disruption of the steroid receptor coactivator-1 (SRC-1) gene. Science 1998;279(5358):1922-5. 75. Ye X, Han SJ, Tsai SY, DeMayo FJ, Xu J, Tsai MJ, et al. Roles of steroid receptor coactivator (SRC)-1 and transcriptional intermediary factor (TIF) 2 in androgen receptor activity in mice. Proc Natl Acad Sci U S A 2005;102(27):9487-92. 76. Gehin M, Mark M, Dennefeld C, Dierich A, Gronemeyer H, Chambon P. The function of TIF2/GRIP1 in mouse reproduction is distinct from those of SRC-1 and p/CIP. Mol Cell Biol 2002;22(16):5923-37. 143 77. Tien JC, Liao L, Liu Y, Liu Z, Lee DK, Wang F, et al. The steroid receptor coactivator-3 is required for developing neuroendocrine tumor in the mouse prostate. International journal of biological sciences 2014;10(10):1116-27. 78. Fu M, Wang C, Reutens AT, Wang J, Angeletti RH, Siconolfi-Baez L, et al. p300 and p300/cAMP-response element-binding protein-associated factor acetylate the androgen receptor at sites governing hormone-dependent transactivation. J Biol Chem 2000;275(27):20853-60. 79. Debes JD, Schmidt LJ, Huang H, Tindall DJ. p300 mediates androgen-independent transactivation of the androgen receptor by interleukin 6. Cancer Res 2002;62(20):5632- 6. 80. Brady ME, Ozanne DM, Gaughan L, Waite I, Cook S, Neal DE, et al. Tip60 is a nuclear hormone receptor coactivator. J Biol Chem 1999;274(25):17599-604. 81. Halkidou K, Gnanapragasam VJ, Mehta PB, Logan IR, Brady ME, Cook S, et al. Expression of Tip60, an androgen receptor coactivator, and its role in prostate cancer development. Oncogene 2003;22(16):2466-77. 82. Gaughan L, Brady ME, Cook S, Neal DE, Robson CN. Tip60 is a co-activator specific for class I nuclear hormone receptors. J Biol Chem 2001;276(50):46841-8. 83. Shiota M, Yokomizo A, Masubuchi D, Tada Y, Inokuchi J, Eto M, et al. Tip60 promotes prostate cancer cell proliferation by translocation of androgen receptor into the nucleus. Prostate 2010;70(5):540-54. 84. Chen D, Ma H, Hong H, Koh SS, Huang SM, Schurter BT, et al. Regulation of transcription by a protein methyltransferase. Science 1999;284(5423):2174-7. 85. Lee DY, Northrop JP, Kuo MH, Stallcup MR. Histone H3 lysine 9 methyltransferase G9a is a transcriptional coactivator for nuclear receptors. J Biol Chem 2006;281(13):8476-85. 86. Metzger E, Wissmann M, Yin N, Muller JM, Schneider R, Peters AH, et al. LSD1 demethylates repressive histone marks to promote androgen-receptor-dependent transcription. Nature 2005;437(7057):436-9. 87. Xiang Y, Zhu Z, Han G, Ye X, Xu B, Peng Z, et al. JARID1B is a histone H3 lysine 4 demethylase up-regulated in prostate cancer. Proc Natl Acad Sci U S A 2007;104(49):19226-31. 88. Kang Z, Pirskanen A, Janne OA, Palvimo JJ. Involvement of proteasome in the dynamic assembly of the androgen receptor transcription complex. J Biol Chem 2002;277(50):48366-71. 89. Khan OY, Fu G, Ismail A, Srinivasan S, Cao X, Tu Y, et al. Multifunction steroid receptor coactivator, E6-associated protein, is involved in development of the prostate gland. Molecular endocrinology (Baltimore, Md) 2006;20(3):544-59. 90. Zhu P, Zhou W, Wang J, Puc J, Ohgi KA, Erdjument-Bromage H, et al. A histone H2A deubiquitinase complex coordinating histone acetylation and H1 dissociation in transcriptional regulation. Mol Cell 2007;27(4):609-21. 91. Guo Z, Dai B, Jiang T, Xu K, Xie Y, Kim O, et al. Regulation of androgen receptor activity by tyrosine phosphorylation. Cancer Cell 2006;10(4):309-19. 92. Poukka H, Aarnisalo P, Karvonen U, Palvimo JJ, Janne OA. Ubc9 interacts with the androgen receptor and activates receptor-dependent transcription. J Biol Chem 1999;274(27):19441-6. 144 93. Polkinghorn WR, Parker JS, Lee MX, Kass EM, Spratt DE, Iaquinta PJ, et al. Androgen receptor signaling regulates DNA repair in prostate cancers. Cancer discovery 2013;3(11):1245-53. 94. Agoulnik IU, Krause WC, Bingman WE, 3rd, Rahman HT, Amrikachi M, Ayala GE, et al. Repressors of androgen and progesterone receptor action. J Biol Chem 2003;278(33):31136-48. 95. Hu X, Lazar MA. Transcriptional repression by nuclear hormone receptors. Trends in endocrinology and metabolism: TEM 2000;11(1):6-10. 96. Yoon HG, Wong J. The corepressors silencing mediator of retinoid and thyroid hormone receptor and nuclear receptor corepressor are involved in agonist- and antagonist- regulated transcription by androgen receptor. Molecular endocrinology (Baltimore, Md) 2006;20(5):1048-60. 97. Hodgson MC, Astapova I, Hollenberg AN, Balk SP. Activity of androgen receptor antagonist bicalutamide in prostate cancer cells is independent of NCoR and SMRT corepressors. Cancer Res 2007;67(17):8388-95. 98. Masiello D, Cheng S, Bubley GJ, Lu ML, Balk SP. Bicalutamide functions as an androgen receptor antagonist by assembly of a transcriptionally inactive receptor. J Biol Chem 2002;277(29):26321-6. 99. Scher HI, Buchanan G, Gerald W, Butler LM, Tilley WD. Targeting the androgen receptor: improving outcomes for castration-resistant prostate cancer. Endocr Relat Cancer 2004;11(3):459-76. 100. Veldscholte J, Berrevoets CA, Brinkmann AO, Grootegoed JA, Mulder E. Anti- androgens and the mutated androgen receptor of LNCaP cells: differential effects on binding affinity, heat-shock protein interaction, and transcription activation. Biochemistry 1992;31(8):2393-9. 101. Han G, Foster BA, Mistry S, Buchanan G, Harris JM, Tilley WD, et al. Hormone status selects for spontaneous somatic androgen receptor variants that demonstrate specific ligand and cofactor dependent activities in autochthonous prostate cancer. J Biol Chem 2001;276(14):11204-13. 102. He B, Gampe RT, Jr., Hnat AT, Faggart JL, Minges JT, French FS, et al. Probing the functional link between androgen receptor coactivator and ligand-binding sites in prostate cancer and androgen insensitivity. J Biol Chem 2006;281(10):6648-63. 103. Bohl CE, Wu Z, Miller DD, Bell CE, Dalton JT. Crystal structure of the T877A human androgen receptor ligand-binding domain complexed to cyproterone acetate provides insight for ligand-induced conformational changes and structure-based drug design. J Biol Chem 2007;282(18):13648-55. 104. Guo Z, Yang X, Sun F, Jiang R, Linn DE, Chen H, et al. A novel androgen receptor splice variant is up-regulated during prostate cancer progression and promotes androgen depletion-resistant growth. Cancer Res 2009;69(6):2305-13. 105. Sun S, Sprenger CC, Vessella RL, Haugk K, Soriano K, Mostaghel EA, et al. Castration resistance in human prostate cancer is conferred by a frequently occurring androgen receptor splice variant. J Clin Invest 2010;120(8):2715-30. 106. Li Y, Chan SC, Brand LJ, Hwang TH, Silverstein KA, Dehm SM. Androgen receptor splice variants mediate enzalutamide resistance in castration-resistant prostate cancer cell lines. Cancer Res 2013;73(2):483-9. 145 107. Yu Z, Chen S, Sowalsky AG, Voznesensky OS, Mostaghel EA, Nelson PS, et al. Rapid induction of androgen receptor splice variants by androgen deprivation in prostate cancer. Clin Cancer Res 2014;20(6):1590-600. 108. Dalal K, Roshan-Moniri M, Sharma A, Li H, Ban F, Hassona MD, et al. Selectively targeting the DNA-binding domain of the androgen receptor as a prospective therapy for prostate cancer. J Biol Chem 2014;289(38):26417-29. 109. Visakorpi T, Hyytinen E, Koivisto P, Tanner M, Keinanen R, Palmberg C, et al. In vivo amplification of the androgen receptor gene and progression of human prostate cancer. Nat Genet 1995;9(4):401-6. 110. Linja MJ, Savinainen KJ, Saramaki OR, Tammela TL, Vessella RL, Visakorpi T. Amplification and overexpression of androgen receptor gene in hormone-refractory prostate cancer. Cancer Res 2001;61(9):3550-5. 111. Ford OH, 3rd, Gregory CW, Kim D, Smitherman AB, Mohler JL. Androgen receptor gene amplification and protein expression in recurrent prostate cancer. J Urol 2003;170(5):1817-21. 112. Bubendorf L, Kononen J, Koivisto P, Schraml P, Moch H, Gasser TC, et al. Survey of gene amplifications during prostate cancer progression by high-throughout fluorescence in situ hybridization on tissue microarrays. Cancer Res 1999;59(4):803-6. 113. Holzbeierlein J, Lal P, LaTulippe E, Smith A, Satagopan J, Zhang L, et al. Gene expression analysis of human prostate carcinoma during hormonal therapy identifies androgen-responsive genes and mechanisms of therapy resistance. Am J Pathol 2004;164(1):217-27. 114. Chen CD, Welsbie DS, Tran C, Baek SH, Chen R, Vessella R, et al. Molecular determinants of resistance to antiandrogen therapy. Nat Med 2004;10(1):33-9. 115. Geller J, Albert J. Effects of castration compared with total androgen blockade on tissue dihydrotestosterone (DHT) concentration in benign prostatic hyperplasia (BPH). Urological research 1987;15(3):151-3. 116. Andriole GL, Humphrey P, Ray P, Gleave ME, Trachtenberg J, Thomas LN, et al. Effect of the dual 5alpha-reductase inhibitor dutasteride on markers of tumor regression in prostate cancer. J Urol 2004;172(3):915-9. 117. Iczkowski KA, Qiu J, Qian J, Somerville MC, Rittmaster RS, Andriole GL, et al. The dual 5-alpha-reductase inhibitor dutasteride induces atrophic changes and decreases relative cancer volume in human prostate. Urology 2005;65(1):76-82. 118. Zytiga. In RxTx. 146 122. Hu YC, Yeh S, Yeh SD, Sampson ER, Huang J, Li P, et al. Functional domain and motif analyses of androgen receptor coregulator ARA70 and its differential expression in prostate cancer. J Biol Chem 2004;279(32):33438-46. 123. Song X, Chen J, Zhao M, Zhang C, Yu Y, Lonard DM, et al. Development of potent small-molecule inhibitors to drug the undruggable steroid receptor coactivator-3. Proc Natl Acad Sci U S A 2016;113(18):4970-5. 124. Zhou Z, Flesken-Nikitin A, Corney DC, Wang W, Goodrich DW, Roy-Burman P, et al. Synergy of p53 and Rb deficiency in a conditional mouse model for metastatic prostate cancer. Cancer Res 2006;66(16):7889-98. 125. Chen Z, Trotman LC, Shaffer D, Lin HK, Dotan ZA, Niki M, et al. Crucial role of p53- dependent cellular senescence in suppression of Pten-deficient tumorigenesis. Nature 2005;436(7051):725-30. 126. Robinson D, Van Allen EM, Wu YM, Schultz N, Lonigro RJ, Mosquera JM, et al. Integrative clinical genomics of advanced prostate cancer. Cell 2015;161(5):1215-28. 127. Eastham JA, Stapleton AM, Gousse AE, Timme TL, Yang G, Slawin KM, et al. Association of p53 mutations with metastatic prostate cancer. Clin Cancer Res 1995;1(10):1111-8. 128. Meyers FJ, Gumerlock PH, Chi SG, Borchers H, Deitch AD, deVere White RW. Very frequent p53 mutations in metastatic prostate carcinoma and in matched primary tumors. Cancer 1998;83(12):2534-9. 129. Grignon DJ, Caplan R, Sarkar FH, Lawton CA, Hammond EH, Pilepich MV, et al. p53 status and prognosis of locally advanced prostatic adenocarcinoma: a study based on RTOG 8610. J Natl Cancer Inst 1997;89(2):158-65. 130. Tomlins SA, Rhodes DR, Perner S, Dhanasekaran SM, Mehra R, Sun XW, et al. Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 2005;310(5748):644-8. 131. Carver BS, Tran J, Gopalan A, Chen Z, Shaikh S, Carracedo A, et al. Aberrant ERG expression cooperates with loss of PTEN to promote cancer progression in the prostate. Nat Genet 2009;41(5):619-24. 132. Wang S, Gao J, Lei Q, Rozengurt N, Pritchard C, Jiao J, et al. Prostate-specific deletion of the murine Pten tumor suppressor gene leads to metastatic prostate cancer. Cancer Cell 2003;4(3):209-21. 133. Baca SC, Prandi D, Lawrence MS, Mosquera JM, Romanel A, Drier Y, et al. Punctuated evolution of prostate cancer genomes. Cell 2013;153(3):666-77. 134. Garkavtsev I, Kazarov A, Gudkov A, Riabowol K. Suppression of the novel growth inhibitor p33ING1 promotes neoplastic transformation. Nat Genet 1996;14(4):415-20. 135. Nagashima M, Shiseki M, Miura K, Hagiwara K, Linke SP, Pedeux R, et al. DNA damage-inducible gene p33ING2 negatively regulates cell proliferation through acetylation of p53. Proc Natl Acad Sci U S A 2001;98(17):9671-6. 136. Nagashima M, Shiseki M, Pedeux RM, Okamura S, Kitahama-Shiseki M, Miura K, et al. A novel PHD-finger motif protein, p47ING3, modulates p53-mediated transcription, cell cycle control, and apoptosis. Oncogene 2003;22(3):343-50. 137. Shiseki M, Nagashima M, Pedeux RM, Kitahama-Shiseki M, Miura K, Okamura S, et al. p29ING4 and p28ING5 bind to p53 and p300, and enhance p53 activity. Cancer Res 2003;63(10):2373-8. 147 138. Toyama T, Iwase H, Watson P, Muzik H, Saettler E, Magliocco A, et al. Suppression of ING1 expression in sporadic breast cancer. Oncogene 1999;18(37):5187-93. 139. Tokunaga E, Maehara Y, Oki E, Kitamura K, Kakeji Y, Ohno S, et al. Diminished expression of ING1 mRNA and the correlation with p53 expression in breast cancers. Cancer Lett 2000;152(1):15-22. 140. Ohmori M, Nagai M, Tasaka T, Koeffler HP, Toyama T, Riabowol K, et al. Decreased expression of p33ING1 mRNA in lymphoid malignancies. Am J Hematol 1999;62(2):118-9. 141. Oki E, Maehara Y, Tokunaga E, Kakeji Y, Sugimachi K. Reduced expression of p33(ING1) and the relationship with p53 expression in human gastric cancer. Cancer Lett 1999;147(1-2):157-62. 142. Sager R. Expression genetics in cancer: shifting the focus from DNA to RNA. Proc Natl Acad Sci U S A 1997;94(3):952-5. 143. Feng X, Hara Y, Riabowol K. Different HATS of the ING1 gene family. Trends Cell Biol 2002;12(11):532-8. 144. Li H, Ilin S, Wang W, Duncan EM, Wysocka J, Allis CD, et al. Molecular basis for site- specific read-out of histone H3K4me3 by the BPTF PHD finger of NURF. Nature 2006;442(7098):91-5. 145. Matthews AG, Kuo AJ, Ramon-Maiques S, Han S, Champagne KS, Ivanov D, et al. RAG2 PHD finger couples histone H3 lysine 4 trimethylation with V(D)J recombination. Nature 2007;450(7172):1106-10. 146. Pena PV, Davrazou F, Shi X, Walter KL, Verkhusha VV, Gozani O, et al. Molecular mechanism of histone H3K4me3 recognition by plant homeodomain of ING2. Nature 2006;442(7098):100-3. 147. Wagner MJ, Gogela-Spehar M, Skirrow RC, Johnston RN, Riabowol K, Helbing CC. Expression of novel ING variants is regulated by thyroid hormone in the Xenopus laevis tadpole. J Biol Chem 2001;276(50):47013-20. 148. Loewith R, Meijer M, Lees-Miller SP, Riabowol K, Young D. Three yeast proteins related to the human candidate tumor suppressor p33(ING1) are associated with histone acetyltransferase activities. Mol Cell Biol 2000;20(11):3807-16. 149. He GH, Helbing CC, Wagner MJ, Sensen CW, Riabowol K. Phylogenetic analysis of the ING family of PHD finger proteins. Mol Biol Evol 2005;22(1):104-16. 150. Skowyra D, Zeremski M, Neznanov N, Li M, Choi Y, Uesugi M, et al. Differential association of products of alternative transcripts of the candidate tumor suppressor ING1 with the mSin3/HDAC1 transcriptional corepressor complex. J Biol Chem 2001;276(12):8734-9. 151. Doyon Y, Selleck W, Lane WS, Tan S, Cote J. Structural and functional conservation of the NuA4 histone acetyltransferase complex from yeast to humans. Mol Cell Biol 2004;24(5):1884-96. 152. Doyon Y, Cayrou C, Ullah M, Landry AJ, Cote V, Selleck W, et al. ING tumor suppressor proteins are critical regulators of chromatin acetylation required for genome expression and perpetuation. Mol Cell 2006;21(1):51-64. 153. Scott M, Boisvert FM, Vieyra D, Johnston RN, Bazett-Jones DP, Riabowol K. UV induces nucleolar translocation of ING1 through two distinct nucleolar targeting sequences. Nucleic Acids Res 2001;29(10):2052-8. 148 154. Han X, Feng X, Rattner JB, Smith H, Bose P, Suzuki K, et al. Tethering by lamin A stabilizes and targets the ING1 tumour suppressor. Nat Cell Biol 2008;10(11):1333-40. 155. Gozani O, Karuman P, Jones DR, Ivanov D, Cha J, Lugovskoy AA, et al. The PHD finger of the chromatin-associated protein ING2 functions as a nuclear phosphoinositide receptor. Cell 2003;114(1):99-111. 156. Thalappilly S, Feng X, Pastyryeva S, Suzuki K, Muruve D, Larocque D, et al. The p53 tumor suppressor is stabilized by inhibitor of growth 1 (ING1) by blocking polyubiquitination. PLoS One 2011;6(6):e21065. 157. Kuzmichev A, Zhang Y, Erdjument-Bromage H, Tempst P, Reinberg D. Role of the Sin3-histone deacetylase complex in growth regulation by the candidate tumor suppressor p33(ING1). Mol Cell Biol 2002;22(3):835-48. 158. Wang Y, Wang J, Li G. Leucine zipper-like domain is required for tumor suppressor ING2-mediated nucleotide excision repair and apoptosis. FEBS Lett 2006;580(16):3787- 93. 159. Eapen SA, Netherton SJ, Sarker KP, Deng L, Chan A, Riabowol K, et al. Identification of a novel function for the chromatin remodeling protein ING2 in muscle differentiation. PLoS One 2012;7(7):e40684. 160. Culurgioni S, Munoz IG, Moreno A, Palacios A, Villate M, Palmero I, et al. Crystal structure of inhibitor of growth 4 (ING4) dimerization domain reveals functional organization of ING family of chromatin-binding proteins. J Biol Chem 2012;287(14):10876-84. 161. Tallen G, Riabowol K. Keep-ING balance: tumor suppression by epigenetic regulation. FEBS Lett 2014;588(16):2728-42. 162. Kameyama K, Huang CL, Liu D, Masuya D, Nakashima T, Sumitomo S, et al. Reduced ING1b gene expression plays an important role in carcinogenesis of non-small cell lung cancer patients. Clin Cancer Res 2003;9(13):4926-34. 163. Tallen G, Kaiser I, Krabbe S, Lass U, Hartmann C, Henze G, et al. No ING1 mutations in human brain tumours but reduced expression in high malignancy grades of astrocytoma. Int J Cancer 2004;109(3):476-9. 164. Chen L, Matsubara N, Yoshino T, Nagasaka T, Hoshizima N, Shirakawa Y, et al. Genetic alterations of candidate tumor suppressor ING1 in human esophageal squamous cell cancer. Cancer Res 2001;61(11):4345-9. 165. Gunduz M, Ouchida M, Fukushima K, Hanafusa H, Etani T, Nishioka S, et al. Genomic structure of the human ING1 gene and tumor-specific mutations detected in head and neck squamous cell carcinomas. Cancer Res 2000;60(12):3143-6. 166. Nouman GS, Anderson JJ, Mathers ME, Leonard N, Crosier S, Lunec J, et al. Nuclear to cytoplasmic compartment shift of the p33ING1b tumour suppressor protein is associated with malignancy in melanocytic lesions. Histopathology 2002;40(4):360-6. 167. Nouman GS, Anderson JJ, Wood KM, Lunec J, Hall AG, Reid MM, et al. Loss of nuclear expression of the p33(ING1b) inhibitor of growth protein in childhood acute lymphoblastic leukaemia. J Clin Pathol 2002;55(8):596-601. 168. Vieyra D, Senger DL, Toyama T, Muzik H, Brasher PM, Johnston RN, et al. Altered subcellular localization and low frequency of mutations of ING1 in human brain tumors. Clin Cancer Res 2003;9(16 Pt 1):5952-61. 149 169. Lu F, Dai DL, Martinka M, Ho V, Li G. Nuclear ING2 expression is reduced in human cutaneous melanomas. Br J Cancer 2006;95(1):80-6. 170. Gunduz M, Nagatsuka H, Demircan K, Gunduz E, Cengiz B, Ouchida M, et al. Frequent deletion and down-regulation of ING4, a candidate tumor suppressor gene at 12p13, in head and neck squamous cell carcinomas. Gene 2005;356:109-17. 171. Wang Y, Dai DL, Martinka M, Li G. Prognostic significance of nuclear ING3 expression in human cutaneous melanoma. Clin Cancer Res 2007;13(14):4111-6. 172. Sarela AI, Farmery SM, Markham AF, Guillou PJ. The candidate tumour suppressor gene, ING1, is retained in colorectal carcinomas. Eur J Cancer 1999;35(8):1264-7. 173. Campos EI, Cheung KJ, Jr., Murray A, Li S, Li G. The novel tumour suppressor gene ING1 is overexpressed in human melanoma cell lines. The British journal of dermatology 2002;146(4):574-80. 174. Kumamoto K, Fujita K, Kurotani R, Saito M, Unoki M, Hagiwara N, et al. ING2 is upregulated in colon cancer and increases invasion by enhanced MMP13 expression. Int J Cancer 2009;125(6):1306-15. 175. Coles AH, Liang H, Zhu Z, Marfella CG, Kang J, Imbalzano AN, et al. Deletion of p37Ing1 in mice reveals a p53-independent role for Ing1 in the suppression of cell proliferation, apoptosis, and tumorigenesis. Cancer Res 2007;67(5):2054-61. 176. Saito M, Kumamoto K, Robles AI, Horikawa I, Furusato B, Okamura S, et al. Targeted disruption of Ing2 results in defective spermatogenesis and development of soft-tissue sarcomas. PLoS One 2010;5(11):e15541. 177. Coles AH, Gannon H, Cerny A, Kurt-Jones E, Jones SN. Inhibitor of growth-4 promotes IkappaB promoter activation to suppress NF-kappaB signaling and innate immunity. Proc Natl Acad Sci U S A 2010;107(25):11423-8. 178. Shinoura N, Muramatsu Y, Nishimura M, Yoshida Y, Saito A, Yokoyama T, et al. Adenovirus-mediated transfer of p33ING1 with p53 drastically augments apoptosis in gliomas. Cancer Res 1999;59(21):5521-8. 179. Scott M, Bonnefin P, Vieyra D, Boisvert FM, Young D, Bazett-Jones DP, et al. UV- induced binding of ING1 to PCNA regulates the induction of apoptosis. J Cell Sci 2001;114(Pt 19):3455-62. 180. Feng X, Bonni S, Riabowol K. HSP70 induction by ING proteins sensitizes cells to tumor necrosis factor alpha receptor-mediated apoptosis. Mol Cell Biol 2006;26(24):9244-55. 181. Tamannai M, Farhangi S, Truss M, Sinn B, Wurm R, Bose P, et al. The inhibitor of growth 1 (ING1) is involved in trichostatin A-induced apoptosis and caspase 3 signaling in p53-deficient glioblastoma cells. Oncol Res 2010;18(10):469-80. 182. Smith KT, Martin-Brown SA, Florens L, Washburn MP, Workman JL. Deacetylase inhibitors dissociate the histone-targeting ING2 subunit from the Sin3 complex. Chem Biol 2010;17(1):65-74. 183. Bose P, Thakur SS, Brockton NT, Klimowicz AC, Kornaga E, Nakoneshny SC, et al. Tumor cell apoptosis mediated by cytoplasmic ING1 is associated with improved survival in oral squamous cell carcinoma patients. Oncotarget 2014;5(10):3210-9. 184. Vieyra D, Loewith R, Scott M, Bonnefin P, Boisvert FM, Cheema P, et al. Human ING1 proteins differentially regulate histone acetylation. J Biol Chem 2002;277(33):29832-9. 150 185. Schafer A, Karaulanov E, Stapf U, Doderlein G, Niehrs C. Ing1 functions in DNA demethylation by directing Gadd45a to H3K4me3. Genes Dev 2013;27(3):261-73. 186. Berger PL, Frank SB, Schulz VV, Nollet EA, Edick MJ, Holly B, et al. Transient induction of ING4 by Myc drives prostate epithelial cell differentiation and its disruption drives prostate tumorigenesis. Cancer Res 2014;74(12):3357-68. 187. Fegers I, Kob R, Eckey M, Schmidt O, Goeman F, Papaioannou M, et al. The tumor suppressors p33ING1 and p33ING2 interact with alien in vivo and enhance alien- mediated gene silencing. Journal of proteome research 2007;6(11):4182-8. 188. Soliman MA, Berardi P, Pastyryeva S, Bonnefin P, Feng X, Colina A, et al. ING1a expression increases during replicative senescence and induces a senescent phenotype. Aging Cell 2008;7(6):783-94. 189. Rajarajacholan UK, Riabowol K. Aging with ING: a comparative study of different forms of stress induced premature senescence. Oncotarget 2015;6(33):34118-27. 190. Rajarajacholan UK, Thalappilly S, Riabowol K. The ING1a tumor suppressor regulates endocytosis to induce cellular senescence via the Rb-E2F pathway. PLoS biology 2013;11(3):e1001502. 191. Thakur S, Singla AK, Chen J, Tran U, Yang Y, Salazar C, et al. Reduced ING1 levels in breast cancer promotes metastasis. Oncotarget 2014;5(12):4244-56. 192. Kim S, Welm AL, Bishop JM. A dominant mutant allele of the ING4 tumor suppressor found in human cancer cells exacerbates MYC-initiated mouse mammary tumorigenesis. Cancer Res 2010;70(12):5155-62. 193. Li J, Li G. Cell cycle regulator ING4 is a suppressor of melanoma angiogenesis that is regulated by the metastasis suppressor BRMS1. Cancer Res 2010;70(24):10445-53. 194. Yan R, He L, Li Z, Han X, Liang J, Si W, et al. SCF(JFK) is a bona fide E3 ligase for ING4 and a potent promoter of the angiogenesis and metastasis of breast cancer. Genes Dev 2015;29(6):672-85. 195. Suzuki S, Nozawa Y, Tsukamoto S, Kaneko T, Imai H, Minami N. ING3 is essential for asymmetric cell division during mouse oocyte maturation. PLoS One 2013;8(9):e74749. 196. Awe JP, Byrne JA. Identifying candidate oocyte reprogramming factors using cross- species global transcriptional analysis. Cellular reprogramming 2013;15(2):126-33. 197. Mulder KW, Wang X, Escriu C, Ito Y, Schwarz RF, Gillis J, et al. Diverse epigenetic strategies interact to control epidermal differentiation. Nat Cell Biol 2012;14(7):753-63. 198. Toyama T, Iwase H, Yamashita H, Hara Y, Sugiura H, Zhang Z, et al. p33(ING1b) stimulates the transcriptional activity of the estrogen receptor alpha via its activation function (AF) 2 domain. J Steroid Biochem Mol Biol 2003;87(1):57-63. 199. Esmaeili M, Jennek S, Ludwig S, Klitzsch A, Kraft F, Melle C, et al. The tumor suppressor ING1b is a novel corepressor for the androgen receptor and induces cellular senescence in prostate cancer cells. Journal of molecular cell biology 2016. 200. Esmaeili M, Pungsrinont T, Schaefer A, Baniahmad A. A novel crosstalk between the tumor suppressors ING1 and ING2 regulates androgen receptor signaling. Journal of molecular medicine (Berlin, Germany) 2016. 201. Smith ER, Eisen A, Gu W, Sattah M, Pannuti A, Zhou J, et al. ESA1 is a histone acetyltransferase that is essential for growth in yeast. Proc Natl Acad Sci U S A 1998;95(7):3561-5. 151 202. Luo J, Shah S, Riabowol K, Mains PE. The Caenorhabditis elegans ing-3 gene regulates ionizing radiation-induced germ-cell apoptosis in a p53-associated pathway. Genetics 2009;181(2):473-82. 203. Kim S, Natesan S, Cornilescu G, Carlson S, Tonelli M, McClurg UL, et al. Mechanism of Histone H3K4me3 Recognition by the Plant Homeodomain of Inhibitor of Growth 3. 2016. 204. Wang Y, Li G. ING3 promotes UV-induced apoptosis via Fas/caspase-8 pathway in melanoma cells. J Biol Chem 2006;281(17):11887-93. 205. Chen G, Wang Y, Garate M, Zhou J, Li G. The tumor suppressor ING3 is degraded by SCF(Skp2)-mediated ubiquitin-proteasome system. Oncogene 2010;29(10):1498-508. 206. Gunduz M, Ouchida M, Fukushima K, Ito S, Jitsumori Y, Nakashima T, et al. Allelic loss and reduced expression of the ING3, a candidate tumor suppressor gene at 7q31, in human head and neck cancers. Oncogene 2002;21(28):4462-70. 207. Gunduz M, Beder LB, Gunduz E, Nagatsuka H, Fukushima K, Pehlivan D, et al. Downregulation of ING3 mRNA expression predicts poor prognosis in head and neck cancer. Cancer Sci 2008;99(3):531-8. 208. Gou WF, Sun HZ, Zhao S, Niu ZF, Mao XY, Takano Y, et al. Downregulated inhibitor of growth 3 (ING3) expression during colorectal carcinogenesis. The Indian journal of medical research 2014;139(4):561-7. 209. Nabbi A, Almami A, Thakur S, Suzuki K, Boland D, Bismar TA, et al. ING3 protein expression profiling in normal human tissues suggest its role in cellular growth and self- renewal. Eur J Cell Biol 2015;94(5):214-22. 210. Wang J, Liu Z, Feng X, Gao S, Xu S, Liu P. Tumor suppressor gene ING3 induces cardiomyocyte hypertrophy via inhibition of AMPK and activation of p38 MAPK signaling. Archives of biochemistry and biophysics 2014;562:22-30. 211. Almami A, Hegazy SA, Nabbi A, Alshalalfa M, Salman A, Abou-Ouf H, et al. ING3 is associated with increased cell invasion and lethal outcome in ERG-negative prostate cancer patients. Tumour Biol 2016. 212. Bismar TA, Alshalalfa M, Petersen LF, Teng LH, Gerke T, Bakkar A, et al. Interrogation of ERG gene rearrangements in prostate cancer identifies a prognostic 10-gene signature with relevant implication to patients' clinical outcome. BJU Int 2014;113(2):309-19. 213. Saeed K, Ostling P, Bjorkman M, Mirtti T, Alanen K, Vesterinen T, et al. Androgen receptor-interacting protein HSPBAP1 facilitates growth of prostate cancer cells in androgen-deficient conditions. Int J Cancer 2015;136(11):2535-45. 214. Zhang C. Hybridoma technology for the generation of monoclonal antibodies. Methods Mol Biol 2012;901:117-35. 215. Olson A, Sheth N, Lee JS, Hannon G, Sachidanandam R. RNAi Codex: a portal/database for short-hairpin RNA (shRNA) gene-silencing constructs. Nucleic Acids Res 2006;34(Database issue):D153-7. 216. Meerbrey KL, Hu G, Kessler JD, Roarty K, Li MZ, Fang JE, et al. The pINDUCER lentiviral toolkit for inducible RNA interference in vitro and in vivo. Proc Natl Acad Sci U S A 2011;108(9):3665-70. 217. Snoek R, Rennie PS, Kasper S, Matusik RJ, Bruchovsky N. Induction of cell-free, in vitro transcription by recombinant androgen receptor peptides. J Steroid Biochem Mol Biol 1996;59(3-4):243-50. 152 218. Mulholland DJ, Cox M, Read J, Rennie P, Nelson C. Androgen responsiveness of Renilla luciferase reporter vectors is promoter, transgene, and cell line dependent. Prostate 2004;59(2):115-9. 219. Yan Y, Chen H, Costa M. Chromatin immunoprecipitation assays. Methods Mol Biol 2004;287:9-19. 220. Camp RL, Chung GG, Rimm DL. Automated subcellular localization and quantification of protein expression in tissue microarrays. Nat Med 2002;8(11):1323-7. 221. Suzuki K, Boland D, Gong W, Riabowol K. Domain recognition of the ING1 tumor suppressor by a panel of monoclonal antibodies. Hybridoma (2005) 2011;30(3):239-45. 222. Bismar TA, Alshalalfa M, Petersen LF, Teng LH, Gerke T, Bakkar A, et al. Interrogation of ERG gene rearrangements in prostate cancer identifies a prognostic 10-gene signature with relevant implication to patients' clinical outcome. BJU Int 2013. 223. Jiang C, Xuan Z, Zhao F, Zhang MQ. TRED: a transcriptional regulatory element database, new entries and other development. Nucleic Acids Res 2007;35(Database issue):D137-40. 224. Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE, Kel OV, et al. Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nucleic Acids Res 1998;26(1):362-7. 225. Tsunoda T, Takagi T. Estimating transcription factor bindability on DNA. Bioinformatics 1999;15(7-8):622-30. 226. Mathelier A, Zhao X, Zhang AW, Parcy F, Worsley-Hunt R, Arenillas DJ, et al. JASPAR 2014: an extensively expanded and updated open-access database of transcription factor binding profiles. Nucleic Acids Res 2014;42(Database issue):D142-7. 227. Sandelin A, Alkema W, Engstrom P, Wasserman WW, Lenhard B. JASPAR: an open- access database for eukaryotic transcription factor binding profiles. Nucleic Acids Res 2004;32(Database issue):D91-4. 228. Chacon D, Beck D, Perera D, Wong JW, Pimanda JE. BloodChIP: a database of comparative genome-wide transcription factor binding profiles in human blood cells. Nucleic Acids Res 2014;42(Database issue):D172-7. 229. Wu JQ, Seay M, Schulz VP, Hariharan M, Tuck D, Lian J, et al. Tcf7 is an important regulator of the switch of self-renewal and differentiation in a multipotential hematopoietic cell line. PLoS Genet 2012;8(3):e1002565. 230. Tanaka Y, Joshi A, Wilson NK, Kinston S, Nishikawa S, Gottgens B. The transcriptional programme controlled by Runx1 during early embryonic blood development. Developmental biology 2012;366(2):404-19. 231. Wu HC, Hsieh JT, Gleave ME, Brown NM, Pathak S, Chung LW. Derivation of androgen-independent human LNCaP prostatic cancer cell sublines: role of bone stromal cells. Int J Cancer 1994;57(3):406-12. 232. Jaaskelainen T, Makkonen H, Palvimo JJ. Steroid up-regulation of FKBP51 and its role in hormone signaling. Curr Opin Pharmacol 2011;11(4):326-31. 233. Pena PV, Hom RA, Hung T, Lin H, Kuo AJ, Wong RP, et al. Histone H3K4me3 binding is required for the DNA repair and apoptotic activities of ING1 tumor suppressor. Journal of molecular biology 2008;380(2):303-12. 234. Lai JS, Herr W. Ethidium bromide provides a simple tool for identifying genuine DNA- independent protein associations. Proc Natl Acad Sci U S A 1992;89(15):6958-62. 153 235. Suzuki K, Bose P, Leong-Quong RY, Fujita DJ, Riabowol K. REAP: A two minute cell fractionation method. BMC Res Notes 2010;3:294. 236. Nabbi A, Riabowol K. Rapid Isolation of Nuclei from Cells In Vitro. Cold Spring Harbor protocols 2015;2015(8):pdb.prot083733. 237. Katz MH. Multivariable Analysis: a practical guide for clinicians and public health researchers. New york: cambridge university press; 2011. 238. Cox DR. Regression models and life tables. J R Stat Soc 1972;34:187-220. 239. Voskuil J. Commercial antibodies and their validation. F1000Research 2014;3:232. 240. Yu W, Hill WG. Lack of specificity shown by P2Y6 receptor antibodies. Naunyn- Schmiedeberg's archives of pharmacology 2013;386(10):885-91. 241. Talmont F, Mouledous L. Evaluation of commercial antibodies against human sphingosine-1-phosphate receptor 1. Naunyn-Schmiedeberg's archives of pharmacology 2014;387(5):427-31. 242. Herrera M, Sparks MA, Alfonso-Pecchio AR, Harrison-Bernard LM, Coffman TM. Lack of specificity of commercial antibodies leads to misidentification of angiotensin type 1 receptor protein. Hypertension 2013;61(1):253-8. 243. Egelhofer TA, Minoda A, Klugman S, Lee K, Kolasinska-Zwierz P, Alekseyenko AA, et al. An assessment of histone-modification antibody quality. Nature structural & molecular biology 2011;18(1):91-3. 244. Garkavtsev I, Boland D, Mai J, Wilson H, Veillette C, Riabowol K. Specific monoclonal antibody raised against the p33ING1 tumor suppressor. Hybridoma 1997;16(6):537-40. 245. Boland D, Olineck V, Bonnefin P, Vieyra D, Parr E, Riabowol K. A panel of CAb antibodies recognize endogenous and ectopically expressed ING1 protein. Hybridoma 2000;19(2):161-5. 246. Satpathy S, Nabbi A, Riabowol K. RegulatING chromatin regulators: post-translational modification of the ING family of epigenetic regulators. Biochem J 2013;450(3):433-42. 247. Fan Y, Wang P, Fu W, Dong T, Qi C, Liu L, et al. Genome-wide association study for pigmentation traits in Chinese Holstein population. Animal genetics 2014;45(5):740-4. 248. DeRan M, Pulvino M, Greene E, Su C, Zhao J. Transcriptional activation of histone genes requires NPAT-dependent recruitment of TRRAP-Tip60 complex to histone promoters during the G1/S phase transition. Mol Cell Biol 2008;28(1):435-47. 249. Yu G, Lee YC, Cheng CJ, Wu CF, Song JH, Gallick GE, et al. RSK promotes prostate cancer progression in bone through ING3, CKAP2, and PTK6-mediated cell survival. Mol Cancer Res 2015;13(2):348-57. 250. Okuda T, van Deursen J, Hiebert SW, Grosveld G, Downing JR. AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell 1996;84(2):321-30. 251. Chong JL, Wenzel PL, Saenz-Robles MT, Nair V, Ferrey A, Hagan JP, et al. E2f1-3 switch from activators in progenitor cells to repressors in differentiating cells. Nature 2009;462(7275):930-4. 252. Korenchuk S, Lehr JE, L MC, Lee YG, Whitney S, Vessella R, et al. VCaP, a cell-based model system of human prostate cancer. In vivo (Athens, Greece) 2001;15(2):163-8. 253. Sun C, Dobi A, Mohamed A, Li H, Thangapazham RL, Furusato B, et al. TMPRSS2- ERG fusion, a common genomic alteration in prostate cancer activates C-MYC and abrogates prostate epithelial differentiation. Oncogene 2008;27(40):5348-53. 154 254. Dehm SM, Tindall DJ. Ligand-independent androgen receptor activity is activation function-2-independent and resistant to antiandrogens in androgen refractory prostate cancer cells. J Biol Chem 2006;281(38):27882-93. 255. Jiang M, Ma Y, Chen C, Fu X, Yang S, Li X, et al. Androgen-responsive gene database: integrated knowledge on androgen-responsive genes. Molecular endocrinology (Baltimore, Md) 2009;23(11):1927-33. 256. Makkonen H, Kauhanen M, Paakinaho V, Jaaskelainen T, Palvimo JJ. Long-range activation of FKBP51 transcription by the androgen receptor via distal intronic enhancers. Nucleic Acids Res 2009;37(12):4135-48. 257. Ni L, Yang CS, Gioeli D, Frierson H, Toft DO, Paschal BM. FKBP51 promotes assembly of the Hsp90 chaperone complex and regulates androgen receptor signaling in prostate cancer cells. Mol Cell Biol 2010;30(5):1243-53. 258. Febbo PG, Lowenberg M, Thorner AR, Brown M, Loda M, Golub TR. Androgen mediated regulation and functional implications of fkbp51 expression in prostate cancer. J Urol 2005;173(5):1772-7. 259. Claessens F, Denayer S, Van Tilborgh N, Kerkhofs S, Helsen C, Haelens A. Diverse roles of androgen receptor (AR) domains in AR-mediated signaling. Nuclear receptor signaling 2008;6:e008. 260. Shi X, Hong T, Walter KL, Ewalt M, Michishita E, Hung T, et al. ING2 PHD domain links histone H3 lysine 4 methylation to active gene repression. Nature 2006;442(7098):96-9. 261. Wang C, Tian L, Popov VM, Pestell RG. Acetylation and nuclear receptor action. J Steroid Biochem Mol Biol 2011;123(3-5):91-100. 262. Poste G. Bring on the biomarkers. Nature 2011;469(7329):156-7. 263. Drucker E, Krapfenbauer K. Pitfalls and limitations in translation from biomarker discovery to clinical utility in predictive and personalised medicine. The EPMA journal 2013;4(1):7. 264. McShane LM, Altman DG, Sauerbrei W, Taube SE, Gion M, Clark GM. Reporting recommendations for tumor marker prognostic studies. J Clin Oncol 2005;23(36):9067- 72. 265. Altman DG, McShane LM, Sauerbrei W, Taube SE. Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK): explanation and elaboration. PLoS medicine 2012;9(5):e1001216. 266. Zhu ML, Kyprianou N. Role of androgens and the androgen receptor in epithelial- mesenchymal transition and invasion of prostate cancer cells. Faseb j 2010;24(3):769-77. 267. Yu J, Yu J, Mani RS, Cao Q, Brenner CJ, Cao X, et al. An integrated network of androgen receptor, polycomb, and TMPRSS2-ERG gene fusions in prostate cancer progression. Cancer Cell 2010;17(5):443-54. 268. Mounir Z, Korn JM, Westerling T, Lin F, Kirby CA, Schirle M, et al. ERG signaling in prostate cancer is driven through PRMT5-dependent methylation of the Androgen Receptor. eLife 2016;5. 269. Mereniuk TR, Maranchuk RA, Schindler A, Penner-Chea J, Freschauf GK, Hegazy S, et al. Genetic screening for synthetic lethal partners of polynucleotide kinase/phosphatase: potential for targeting SHP-1-depleted cancers. Cancer Res 2012;72(22):5934-44. 155 APPENDIX A: PERMISSIONS TO REUSE 156 8/29/2016 RightsLink Printable License WOLTERS KLUWER HEALTH LICENSE TERMS AND CONDITIONS Aug 29, 2016 This Agreement between Arash Nabbi ("You") and Wolters Kluwer Health ("Wolters Kluwer Health") consists of your license details and the terms and conditions provided by Wolters Kluwer Health and Copyright Clearance Center. License Number 3938350542509 License date Aug 29, 2016 Licensed Content Publisher Wolters Kluwer Health Licensed Content Publication WK Health Book Licensed Content Title Textbook of Therapeutics Licensed Content Author Richard A Helms, David J Quan, Licensed Content Date 2006 Type of Use Dissertation/Thesis Requestor type academic/educational Format print and electronic Portion figures/tables/illustrations Number of 1 figures/tables/illustrations The ID numbers of the Figure 100.2 figures/tables/illustrations are... Will you be translating? no Reusing current or a current edition previous edition Circulation/distribution 20 Order reference number Title of your thesis / Role of ING3 epigenetic regulator in prostate cancer dissertation Expected completion date Sep 2016 Estimated size (number of 180 pages) Requestor Location Arash Nabbi 3330 Hospital Dr NW Calgary, AB T2N4N1 Canada Attn: Arash Nabbi Billing Type Invoice Billing Address Arash Nabbi 3330 Hospital Dr NW Calgary, AB T2N4N1 https://s100.copyright.com/AppDispatchServlet 1/4 7/4/2016 RightsLink Printable License JOHN WILEY AND SONS LICENSE TERMS AND CONDITIONS Jul 04, 2016 This Agreement between Arash Nabbi ("You") and John Wiley and Sons ("John Wiley and Sons") consists of your license details and the terms and conditions provided by John Wiley and Sons and Copyright Clearance Center. License Number 3902141160430 License date Jul 04, 2016 Licensed Content Publisher John Wiley and Sons Licensed Content Publication FEBS Letters Licensed Content Title Keep‐ING balance: Tumor suppression by epigenetic regulation Licensed Content Author Gesche Tallen,Karl Riabowol Licensed Content Date Mar 14, 2014 Licensed Content Pages 15 Type of use Dissertation/Thesis Requestor type University/Academic Format Print and electronic Portion Figure/table Number of figures/tables 1 Original Wiley figure/table Figure 1 number(s) Will you be translating? No Title of your thesis / Role of ING3 epigenetic regulator in prostate cancer dissertation Expected completion date Sep 2016 Expected size (number of 180 pages) Requestor Location Arash Nabbi 3330 Hospital Dr NW Calgary, AB T2N4N1 Canada Attn: Arash Nabbi Publisher Tax ID EU826007151 Billing Type Invoice Billing Address Arash Nabbi 3330 Hospital Dr NW Calgary, AB T2N4N1 Canada Attn: Arash Nabbi Total 0.00 USD Terms and Conditions https://s100.copyright.com/AppDispatchServlet 1/5 6/13/2016 RightsLink Printable License ELSEVIER LICENSE TERMS AND CONDITIONS Jun 13, 2016 This Agreement between Arash Nabbi ("You") and Elsevier ("Elsevier") consists of your license details and the terms and conditions provided by Elsevier and Copyright Clearance Center. License Number 3887141047914 License date Jun 13, 2016 Licensed Content Publisher Elsevier Licensed Content Publication European Journal of Cell Biology Licensed Content Title ING3 protein expression profiling in normal human tissues suggest its role in cellular growth and self-renewal Licensed Content Author Arash Nabbi,Amal Almami,Satbir Thakur,Keiko Suzuki,Donna Boland,Tarek A. Bismar,Karl Riabowol Licensed Content Date May 2015 Licensed Content Volume 94 Number Licensed Content Issue 5 Number Licensed Content Pages 9 Start Page 214 End Page 222 Type of Use reuse in a thesis/dissertation Intended publisher of new other work Portion full article Format both print and electronic Are you the author of this Yes Elsevier article? Will you be translating? No Order reference number Title of your Role of ING3 epigenetic regulator in prostate cancer thesis/dissertation Expected completion date Sep 2016 Estimated size (number of 180 pages) Elsevier VAT number GB 494 6272 12 Requestor Location Arash Nabbi 3330 Hospital Dr NW Calgary, AB T2N4N1 Canada Attn: Arash Nabbi https://s100.copyright.com/AppDispatchServlet 1/6 6/13/2016 RightsLink Printable License ELSEVIER LICENSE TERMS AND CONDITIONS Jun 13, 2016 This Agreement between Arash Nabbi ("You") and Elsevier ("Elsevier") consists of your license details and the terms and conditions provided by Elsevier and Copyright Clearance Center. License Number 3887140845200 License date Jun 13, 2016 Licensed Content Publisher Elsevier Licensed Content Publication The Lancet Oncology Licensed Content Title Should the Gleason grading system for prostate cancer be modified to account for high-grade tertiary components? A systematic review and meta-analysis Licensed Content Author Patricia Harnden,Mike D Shelley,Bernadette Coles,John Staffurth,Malcom D Mason Licensed Content Date May 2007 Licensed Content Volume 8 Number Licensed Content Issue 5 Number Licensed Content Pages 9 Start Page 411 End Page 419 Type of Use reuse in a thesis/dissertation Portion figures/tables/illustrations Number of 1 figures/tables/illustrations Format both print and electronic Are you the author of this No Elsevier article? Will you be translating? No Order reference number Original figure numbers Figure 1 Title of your Role of ING3 epigenetic regulator in prostate cancer thesis/dissertation Expected completion date Sep 2016 Estimated size (number of 180 pages) Elsevier VAT number GB 494 6272 12 Requestor Location Arash Nabbi 3330 Hospital Dr NW Calgary, AB T2N4N1 https://s100.copyright.com/AppDispatchServlet 1/6