DELINEATING THE ROLE OF HEPATOCYTE NUCLEAR FACTOR 1 BETA (HNF1B) TRANSCRIPT VARIANTS IN PROSTATE CANCER
Shubhra Chandra B.Sc. (Hons), M.Sc.
Submitted in fulfillment of the requirements for the degree of
Masters of Philosophy
School of Biomedical Sciences Faculty of Health Queensland University of Technology
2019
Keywords
Genome wide association studies (GWAS), Hepatocyte nuclear factor 1 beta (HNF1B), single nucleotide polymorphisms (SNPs), PCa, splice variants.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer i
Abstract
Prostate Cancer (PCa) is the second common cause of cancer death in Australian men. Genome Wide Association Studies (GWAS) have identified more than 150 single nucleotide polymorphisms (SNPs) which are linked with increased risk of developing PCa, with most of the SNPs occurring in the non-protein coding region of the DNA. Through this approach, Hepatocyte Nuclear Factor 1 beta (HNF1B), a homeobox transcription factor, having a vital role in embryonic development of organs mainly liver, kidney, and pancreas and located at chromosome 17q12 has been reported to be a foremost risk gene for PCa susceptibility in multi-ethnic populations. There are three classical transcript variants of HNF1B namely A, B, and D, having a complete N- terminal dimersation domain, DNA-binding domains POUH and POUS and a C- terminal transactivation domain. There are other annotated transcript variants of HNF1B variants namely C, C’ which differ on the basis of the exons that are missing and code for C-terminal transactivation domains, E and I that differ on the basis of exons coding for dimersation and transactivation domains. Using in-silico analyses, we observed reduced expression of HNF1B in castrate-resistant PCa (CRPC) in comparison to prostate carcinoma. These findings suggest that HNF1B might not be participating in the later stage of PCa. So far there has not been a holistic study on HNF1B transcript variants. Therefore, we characterised the HNF1B transcript variants in a panel of prostate cell lines. From the gene and protein expression studies we observed higher expression of HNF1B transcript variants A and B in VCaP and DU145 in comparison to other prostate cell lines. As we know that variation in androgen receptor function is associated with PCa, we analysed the expression of HNF1B transcript variants in the presence of androgen treatment. There was no change in any of the HNF1B transcript variants, which highlights that HNF1B does not act in the initial androgen-dependent stage of cancer. We next examined the regulation of HNF1B transcript variant expression under docetaxel treatment, to see whether chemotherapy could affect variant expression. We observed increased expression of HNF1B transcript variants suggesting a possible role for some of these HNF1B transcript variants during chemotherapy and/or that chemotherapy affects HNF1B expression. Finally, we also demonstrated a functional effect of the HNF1B transcript variants. There was reduced expression observed with
ii Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
siRNA knockdown at the mRNA level for the transcript variants in VCaP and DU145. There was a reduction observed in cell proliferation using siRNA knockdown for transcript variant C. Next we generated stable cell lines overexpressing HNF1B transcript variants A, B and C and determined their functional effects. There was a reduction in cell proliferation in overexpressed A and B variants in comparison to vector control, whereas an opposite effect, i.e. an increase in cell proliferation was observed for overexpressed C model. Taken together, this thesis represents a comprehensive study on defining the expression patterns and some functional attributes of HNF1B transcript variants in PCa. Future studies will enable the discovery of the molecular mechanism of its actions and whether targeting these HNF1B transcript variants in cancer may prove a useful therapeutic strategy.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer iii
Table of Contents Keywords ...... i Abstract ...... ii List of Figures ...... vi List of Tables ...... viii List of Abbreviations ...... ix Statement of Original Authorship ...... xi Acknowledgements ...... 14 Chapter 1: Literature Review ...... 15 1.1 Introduction ...... 15 1.2 Role of Androgens and Androgen receptor (AR) in PCa ...... 16 1.3 Genetics and epigenetics of HNF1B ...... 18 1.3.1 Genetics of HNF1B ...... 18 1.3.2 HNF1B AND EPIGENETICS ...... 23 1.5 HNF1B and its Splice Variants ...... 24 1.6 HNF1B and its target genes ...... 27 1.7 Role of HNF1B in cancer ...... 29 1.7.1 Tumour-suppressive function ...... 29 1.7.2 Oncogenic role ...... 31 1.8 HNF1B as a biomarker ...... 32 1.9 Summary ...... 34 2. HYPOTHESIS AND AIMS ...... 34 Aim 1. Characterisation of HNF1B transcript variant expression in PCa cell lines and clinical samples...... 35 AIM 2. Role of androgens and docetaxel in regulating the expression of HNF1B transcript variants...... 35 AIM 3. Functional effects of HNF1B and its transcript variants in PCa cell lines. ………………………………………………………………………………...35 Chapter 2: Materials and Methods ...... 37 2.1 Cell culturing conditions ...... 37 2.2 RNA extraction and cDNA synthesis ...... 38 2.3 Designing variant-specific primers ...... 38 2.4 Real time (rRT) and quantitative real-time PCR (qRTPCR) ...... 38 2.6.Statistical analysis ...... 44
iv Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Chapter 3: Characterisation and expression of HNF1B transcript variants in prostate cell lines ...... 45 3.1. Introduction ...... 45 3.2. Methods ...... 45 3.3 Results ...... 46 3.3.1 Expression analysis of HNF1B using bioinformatics tool (cbioportal, GTEx, Oncomine, TCGA-Spliceseq, ChIP and RNA-seq) ...... 46 3.3.2 Quantitative expression of HNF1B transcript variants ...... 53 3.3.3 Determining the protein expression level of HNF1B transcript variants in PCa cell lines ...... 56 3.3.4 Discussion ...... 57 Chapter 4: Role of androgens and docetaxel in the expression of HNF1B transcript variants ...... 59 4.1 Introduction ...... 59 4.2 Methods ...... 60 4.2.1 . Androgen deprivation assay in LNCaP cell lines ...... 60 4.2.2 Docetaxel treatment in LNCaP cell lines ...... 60 4.3 Results ...... 61 4.3.1 Discussion ...... 67 Chapter 5: Functional effects of The HNF1B and its transcript variants on Prostate cell lines ...... 70 5.1 Introduction ...... 70 5.2 Methods ...... 70 5.2.1. Generation of loss of function (knockdown; using siRNA) and to perform functional assays for knockdown ...... 70 5.2.2 Generation of loss of function/knockdown at the mRNA level...... 71 5.2.3. Cellular proliferation assay for knockdown models ...... 72 5.2.4. To generate overexpression templates for HNF1B transcript variants A, B and C ...... 72 5.2.5. Mass Spectrometry of overexpression models for HNF1B A...... 73 5.2.6. Cellular proliferation assay for overexpression models ...... 73 5.3. Results ...... 73 5.4. Discussion ...... 83 Chapter 6: General Discussion ...... 86 Chapter 7: Conclusion ...... 90 Bibliography ...... 91 Appendices ...... 108 Appendix B ...... 108
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer v
List of Figures
Figure 1-1- Mechanism of Androgen-AR interaction in the prostate...... 16 Figure 1-2-Locus Explorer plot for the HNF1B PCa risk locus at Chr17q12...... 20 Figure 1-3- Characterised and annotated HNF1B Transcript variants ...... 26 Figure 2-1-RT-PCR Primer Scheme...... 41 Figure 2-2-qRTPCR Primer Scheme: ...... 43 Figure 2-3- N and C terminal Antibody Schema: ...... 44 Figure 3-1-HNF1B expression in various tissue samples obtained from Cbioportal: ...... 47 Figure 3-2- HNF1B Gene Expression in GTEx dataset...... 48 Figure 3-3-HNF1B expression in Bittner-Multi-Cancer Dataset...... 50 Figure 3-4-HNF1B expression in localised prostate carcinoma versus castrate resistant metastatic prostate carcinoma in the Grasso Prostate dataset...... 50 Figure 3-5 Multiple sequence alignment of HNF1B isoforms: ...... 51 Figure 3-6-Different transcripts of HNF1B as per TCGA Spliceseq: ...... 52 Figure 3-7-RNA-Seq: The data for the four different prostate cell lines are aligned with the HNF1B transcript variants...... 53 Figure 3-8-Characterisation of HNF1B transcript variants in a panel of cell lines ...... 54 Figure 3-9-The relative expression of HNF1B variants using qRTPCR: ...... 56 Figure 3-10-Western blotting carried out to identify HNF1B transcript variants at the protein level: ...... 57 Figure 4-1-Positive control used for confirming the DHT/ androgen treatment: ...... 62 Figure 4-2-HNF1B transcript variant expression in LNCaP cells in the presence of androgens and anti-androgens...... 63 Figure 4-3-Positive Control used for confirming docetaxel treatment...... 65 Figure 4-4-HNF1B transcript variants in the presence of docetaxel: ...... 66 Figure 5-1-siRNA Scheme for HNF1B transcript variants: ...... 71 Figure 5-2-Si RNA2 mediated Knockdown at the mRNA level in VCaP cells: ...... 74 Figure 5-3-siRNA Mediated Knockdown of different HNF1B transcripts in DU145 cells: ...... 75 Figure 5-4-Cell proliferation assay following siRNA Knockdown in DU145 Cells: ...... 76
vi Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Figure 5-5-Cellular proliferation following siRNA knockdown in DU145 Cells: ...... 76 Figure 5-6-PCR analysis to identify positive clones: ...... 77 Figure 5-7-Restriction enzyme analysis of positive HNF1B transcript variant clones showing correct sizes...... 78 Figure 5-8-Overexpression of HNF1B A in PC3 cells...... 80 Figure 5-9-Overexpression of HNF1B B in PC3 cells...... 80 Figure 5-10-Overexpression of HNF1B C in PC3 cells...... 81 Figure 5-11-Stable overexpression of HNF1B A in PC3 cells resulted in a decrease in proliferation...... 82 Figure 5-12-Stable Overexpression model of HNF1B B in PC3 cells resulted in a decrease in proliferation...... 82 Figure 5-13-Cell proliferation assay showing overexpression of HNF1B C’ variant which resulted in increased proliferation...... 83 Figure 7-1-Gel images for HNF1B transcript variants: ...... 108 Figure 7-2-Sequence confirmation for HNF1B transcript variants A and B ... 109 Figure 7-3-Sequence confirmed for HNF1B transcript variant C (cloned in p GEMT vector) ...... 109 Figure 7-4- Western blotting for overexpressed clones for HNF1B a: ...... 109 Figure 7-5: Ingenuity Pathways Analysis (IPA) for prediction of upstream pathways involved in HNF1B overexpression model A...... 110
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer vii
List of Tables
Table 1-1: HNF1B risk-associated SNPs identified in prostate cell lines...... 23 Table 2-1: The primers used for RT-PCR for HNF1B transcript variants...... 39 Table 2-2: RT- PCR reaction protocol ...... 40 Table 2-3: The primers used for q RT-PCR for HNF1B transcript variants ..... 42 Table 3-1: Overview of the expression of HNF1B transcript variants in prostate cell lines from the RT-PCR and RT-qPCR data...... 55
viii Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
List of Abbreviations
AR Androgen Receptor
ARE Androgen Response Elements
ATCC American Type Culture Collection
ATT Androgen Targeted Therapy
BP Base Pair
BPH Benign prostatic hyperplasia
CRPC Castration-resistant PCa
DH Dihydrotestosterone
DMSO Dimethyl sulfoxide
EMT Epithelial to Mesenchymal Transition
EGF Epidermal Growth Factor
GWAS Genome-Wide Association Studies
HCC Hepatocellular Carcinoma
HNF1B Hepatocyte Nuclear Factor 1 Beta
HNF1A Hepatocyte Nuclear Factor 1 Alpha
HPC Hepatic Progenitor Cell
ICC Cholangiocarcinoma
MAF Minor allele frequency
miRNAs microRNAs
MODY Mature-onset Diabetes of the Young
MS Mass spectrometry
μg Microgram
μl Microlitre
Ml Millilitre
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer ix
mM millimolar
NLS Nuclear Localisation Signal
PSA Prostate Specific Antigen
PTP-BL Protein Tyrosine Phosphatase
PDAC Pancreatic Ductal Adenocarcinoma
SNP Single Nucleotide Polymorphisms
T2D Type 2 Diabetes
x Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer Statement of Original Authorship
The work contained in this thesis has not been previously submitted to meet requirements for an award at this or any other higher education institution. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made.
Signature: QUT Verified Signature
Date: 28.08.2019
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer xi
Conference presentations (Poster) 1. Australian Society of Medical Research (ASMR) post graduate student conference, Brisbane, Australia [2016 & 2017] 2. Australia Prostate Cancer Conference (APCC), Melbourne, Australia [2016] 3. Brisbane Cell and Development Meeting, Brisbane, Australia [2016 & 2018] 4. IHBI Inspires postgraduate student conference, Gold Coast, Queensland, Australia [2016] 5. Translational Research Institute (TRI) post-graduate student conference, Brisbane, Australia [2017 & 2018] 6. Princess Alexandra Health (PAH) Symposium, Brisbane, Australia,[2017] 7. IHBI Inspires postgraduate student conference, Brisbane, Australia [2017] 8. 13th Indo-Australian Conference, Brisbane, Australia [2017] 9. IHBI Inspires postgraduate student conference, Brisbane, Australia [2018]-
Best poster People’s Choice Award 10. Princess Alexandra Health (PAH) Symposium, Brisbane, Australia [2018] 11. IHBI Inspires 2018, Brisbane, Australia [2018]
Conference Presentations (Oral) 1. Deutscher Akademischer Austauschdienst (DAAD) workshop: Brisbane, Australia [2016]
xii Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Acknowledgements
It gives me immense pleasure to pen down my sincere gratitude to my principal supervisor A/Prof. Jyotsna Batra for her untiring and continuous support throughout my candidature. Her mentorship has been rewarding to me. I thank you for your positive criticism and encouragement for my research. Your invaluable advice has made me a capable researcher and much wiser. My gratitude to D/Prof. Judith Clements for sharing her valued opinions, insights and zealous encouragement. Her intriguing questions have motivated me to widen my research from various perspectives.
With this, I would also like to thank Queensland University of Technology Postgraduate Research Award (QUTPRA) scholarship and QUT HDR fee waiver awarded to me, that has enabled me to carry out my research.
A big thank you to Dr. Gregor Tevz and Dr. Srilaksmi Srinivasan for their suggestions.
I want to thank A/Prof. Pamela Pollock for her stimulating discussions related to my research project.
I would also like to extend my heartfelt thanks to Clements and Batra group members for their helpful recommendations.
A special mention to all in my family: parents, sister, brother in law and my dearest niece for all the emotional and spiritual support and love they provided me all through my research and life in general.
14 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Chapter 1: Literature Review
1.1 INTRODUCTION
PCa is potentially life-threatening, and the most serious disease of the four main ailments of the prostate - PCa, prostatodynia, benign prostatic hyperplasia (BPH), and, prostatitis. It is the most frequently diagnosed cancer amongst men in Australia (excluding non-melanoma skin cancers) and accounted for nearly 23.1% of newly diagnosed cancers among males with 16,665 males detected with PCa in 2017 (AACR, 2017) . The number of estimated male deaths from PCa in 2016 was 3,398 which accounted for 12.8% of all male deaths in Australia (Australian Institute of Health and Welfare 2016). The magnitude of PCa is, therefore, wide-ranging — affecting males identified with the condition, their families and communities. Genetic influences on PCa risk have been recognised, and our understanding of the molecular genetics of the disease is improving (Barbieri & Tomlins, 2014). Family history, race and age are some of the identified risks for PCa (Crawford, 2003). PCa is one of the most heritable cancers with genetic influences projected to account for nearly 42% of the risk (Lichtenstein et al., 2000). Approximately 85-90% of PCa cases are dependent on androgens, due to which patients are placed on hormone targeted therapy directed towards the reduction of serum androgens and inhibition of the Androgen receptor (AR) (Denis & Griffiths, 2000). However, androgen ablation therapy ultimately fails, and PCa progresses to a hormone independent state (Chodak et al., 1992; Mohler et al., 1996; Sadi, Walsh, & Barrack, 1991; van der Kwast et al., 1991). Factors indicative of a genetic contribution to PCa comprise of: 1) multiple affected first-degree relatives (FDRs) by PCa, with three successive generations with PCa in the maternal or paternal lineage 2) early-onset PCa (age ≤55 years); and 3) PCa with a family history of other cancers (e.g., breast, ovarian, pancreatic)(Board., 2018). Genome-wide associations (GWA) studies have been effective in determining the susceptible loci for PCa identifying more than 100 genetic variants. Risk prediction for PCa by merging numerous SNPs is still basic and is not quite accurate for clinical use, but the model designed by Nakagawa et al (Nakagawa, Akamatsu, & Takata, 2016) captures the patients with extremely high risk of PCa. Recently, the polygenic risk score (PGRS), which is a mode of risk metric calculated on numerous single nucleotide polymorphisms, based upon effect size
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 15
approximations (i.e. linear regression from GWAS meta-analysis), which in other words is an estimate of the complete genetic risk, has become a research focus (Bahcall, 2014). Although increasingly we are learning more about the aetiology of cancer along with function of many genes, this still needs to be studied further as this will clarify their role as potential biomarkers or therapeutic targets for PCa.
1.2 Role of Androgens and Androgen receptor (AR) in PCa
The androgen receptor (AR) via activation by androgens play an important role in normal development and functioning of the prostate. AR is a ligand-dependent transcription factor that regulates the expression of certain genes. This AR binds to its cognate ligands 5α- dihydrotestosterone (DHT) and testosterone which begins male sexual development and differentiation (Tan, Li, Xu, Melcher, & Yong, 2015).
Figure 1-1- Mechanism of Androgen-AR interaction in the prostate.
Testosterone and DHT bind to AR, which stimulates the AR co-regulators (ARAs). AR further translocates to the nucleus and attaches to Androgen Response Elements (AREs) present in promoter regions of target genes to induce cell proliferation and apoptosis. There are other signal transduction pathways that can augment AR activity through phosphorylation of AR
16 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
such as TGFβ, IL-6, and IGF-1. Hsp, Heat shock protein; R, membrane receptor; P, protein phosphorylation. Figure adapted from (Chang et al, 2004).
Testosterone and DHT exert their biological function by binding to AR which leads to the transcriptional activity of AR (Figure 1-1). The transcriptional activity of AR is controlled, via the crosstalk of AR with co-regulators (participating with growth factors) and by phosphorylation of AR (Buchanan, Irvine, Coetzee, & Tilley, 2001; Cunha et al., 1987; Heinlein & Chang, 2002; Roy et al., 1999). The initiation and subsequent progression of PCa to Castrate Resistant Prostate Cancer (CRPC) is dependent on AR and its functions (Lonergan & Tindall, 2011). AR overexpression, mutations, and variants all add up to a prominent clinical importance throughout the development of CRPC (Jernberg, Bergh, & Wikstrom, 2017). Androgen Deprivation therapy (ADT) is the first line of treatment given to metastatic PCa (Coutinho, Day, Tilley, & Selth, 2016). Docetaxel has also shown to be the most important survival drug in CRPC patients with a median overall survival of 19.2 months when administered every 3 weeks (Berthold et al., 2008). New clinical studies including GETUG- AFU 15(Gravis et al., 2013), STAMPEDE (James et al., 2016) and CHAARTED (Sweeney et al., 2015) have elucidated the role of chemotherapy in hormone-sensitive scenarios. The data from these studies suggest a paradigm shift in disease treatment with researchers suggesting a patient with metastatic disease should be provided with six cycles of docetaxel in combination with hormonal therapy to increase survival.
HNF Family
Hepatocyte nuclear factors (HNFs) exist as an assembly of transcription factors that have a vital role in the expression and regulation of genes pertaining to the liver (Lau, Ng, Loo, Jasmen, & Teo, 2017). These transcription factors are not restricted to hepatocytes, as their expression has been observed in other tissues (Azmi, Bao, Gao, Mohammad, & Sarkar, 2013). There are four major families of HNFs namely HNF1, HNF3, HNF4, and HNF6. The HNF1 family members; HNF1α and HNF1β comprise of a POU-homeodomain and bind to DNA as homodimers (Mendel, Hansen, Graves, Conley, & Crabtree, 1991). Amongst the above mentioned HNFs, the HNF1 family members have been reported to have a function in cancer. HNFs function as a homo or heterodimers of two closely related proteins HNF-1A and HNF- 1B. Hepatocyte nuclear factor 1B (HNF1B, TCF2) located on chromosome 17q12 (Gudmundsson et al., 2007), belongs to the family of Pit-1, Oct-1/2, UNC-86 (POU) homeodomain-comprising transcription factors. The POU domain is a bipartite domain (POUH and POUS) and is comprised of two subunits separated by a short 15-55 amino acid amino acids
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 17
non-conserved region. The protein contains three predominant functional domains for dimersation, DNA binding and transactivation, respectively. HNF1B protein is anticipated to have an alpha helical structure which could lead to the formation of a coiled structure (De Simone et al., 1991). Wu et al. studied different domains of HNF1B in the process of nephrogenesis, identifying a dimersation domain, homeo domain and a conserved 26 amino acid segment between the POUS and POUH domains found in the B variant of HNF1B to interfere with pro-nephric development (Wu, Bohn, & Ryffel, 2004). According to the researchers in this study, the 26 amino acid segment may interact with other transcription factors or may alter the protein structure of HNF1B. This underlines the role of different domains in HNF1B splice variants along with the importance of studying individual transcripts. In humans, mutations in HNF1B have been recognised with, multisystem phenotype. HNF1B mutations were first defined as an intermittent genetic basis of maturity- onset diabetes of the young (MODY) (Horikawa et al., 1997). It includes a set of ailments that are normally identified by autosomal dominant inheritance, early onset of diabetes mellitus, commonly before 25 years of age, and pancreatic beta-cell disruption. Renal cysts were found to be the main clinical feature with HNF1B mutations and the linkage of renal cysts with diabetes led to the description of renal cysts with diabetes syndrome. Supplementary phenotypes detected in patients include urogenital tract anomalies, condensed exocrine function, anomalous liver function tests, pancreatic hypoplasia, biliary disorders, hypomagnesemia, hyperuricemia and early-onset gout (Bingham & Hattersley, 2004; Edghill, Bingham, Ellard, & Hattersley, 2006).
1.3 GENETICS AND EPIGENETICS OF HNF1B
1.3.1 Genetics of HNF1B Ever since the advent of single nucleotide polymorphism (SNPs) genotyping arrays, scientists have used genome-wide association studies (GWAS) to detect numerous loci associated with multiple diseases, including cancer. The vast majority of these SNPs were in intergenic or intronic regions (Edwards, Beesley, French, & Dunning, 2013; Welter et al., 2014). One of the susceptible regions of PCa which has lately been a region of interest to researchers includes the 17q12 region (Table 1). Hepatocyte nuclear factor 1 beta (HNF1B), along with metabolism and inflammation-related genes have been searched in this region for association with PCa prognosis (Yu, Guo, Jing, Dong, & Wei, 2015). The initial report on HNF1B association with PCa had been studied in Iceland (Gudmundsson, et al., 2007) and was later replicated in USA and UK populations (Eeles et al., 2008; Thomas et al., 2008) and had shown
18 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
two independent PCa risk-associated loci on chromosome 17q. SNP rs4430796, in HNF1B, was strongly linked to PCa risk and was the first loci to be identified for PCa (Sun et al., 2008). Another independent variant, rs11649743, found at chromosome 17q12 was established to correspond with PCa risk (Sun, et al., 2008). Widespread fine-mapping studies have been conducted on the region 17q12 and have confirmed the previously documented signals in PCa, indicating supplementary variants add to the risk of PCa (Berndt et al., 2011). The fine- mapping studies conducted by Olama et al. in the HNF1B region identified five SNPs, rs7405696, rs4430796, rs4794758, rs3094509, and rs1016990; these collectively captured more risk linked within this region (Amin Al Olama et al., 2015). Further Zhang et al. discovered the loci linked with PCa in a Northern Chinese population pointing out the AG allele on HNF1B (rs4430796, A) could be linked with high PSA (Zhang, Xu, et al., 2012)
Gudmundsson et al recognised the relationship with PCa and the A allele of the SNP rs4430796 (O.R. =1.22, P=1.4 X 10 -11) along with C allele of the SNP rs7501939 (O.R=1.19, P=4.7 X 10 -9) in a GWAS comprising of 1,501 Icelandic men with PCa and 11,290 controls. This was later studied in three case-control studies in individuals from Chicago, Spain and the Netherlands (Gudmundsson, et al., 2007). This was replicated and confirmed in a larger GWAS of PCa (Thomas, et al., 2008) with the same reported SNP rs4430796 (P= 9.58 x 10 -10 ). Further there was a stronger correlation that was found among rs4430796 (random effect model p=10 -13 ) and rs7501939 (random effect model p= 2 x 10 -11) in a two-stage GWAS of PCa (Eeles, et al., 2008) rs4430796 and rs7501939 have also been suggested play a role in early onset of PCa (before the age of 50 years) (Levin et al., 2008). There was a link established for rs4430796 with PCa in a multi-ethic sample set comprising of 2,768 incidents of PCa and 2,359 controls from a multi-ethnic cohort. Another group conducted fine-mapping of HNF1B in 10,272 PCa cases and 9,123 controls from European ancestry that had 10 case-control studies and recognised numerous SNPs (rs 4430796, rs7405696, rs11649743, and rs4794758) affecting PCa (Al Olama et al., 2014). This investigation demonstrated a multifaceted relationship between variants in the HNF1B region and PCa risk. An additional study suggested that rs757210 was linked with PCa risk (Machiela et al., 2012). A fine-mapping study of recognised risk loci amongst a European population reported for more strongly associated lead SNPs as indicators for PCa. The novel SNPs included rs11263763 in the first intron and rs718961 in the fourth intron overlapping with bio features such as Dnase hypersensitivity sites in ENCODE (DNase track) and Histone modifications in ENCODE (Histone track) . There was another SNP rs2229295 located in the 3’UTR of HNF1B, and maybe a potent candidate worthy of
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 19
further investigation (Figure 1-2) (Amin Al Olama, et al., 2015). In a first-ever study carried out in a Korean population by Kim et al (Kim et al., 2011) there were 47 SNPs from the HNF1B gene that were genotyped from 240 patients having PCa along with 223 control subjects to define the possible association of this gene with PCa development. It was found that 14 polymorphisms and 3 haplotypes showed minimal association with the risk of PCa (P = .002- .05). Out of these 14, there were 9 SNPs shown to be associated with lower risk of PCa (OR .67-.71; P = .005-.05), whereas 5 SNPs were shown to be associated with greater risk of cancer (OR 1.49-1.51; P = .002-.02). These SNPs are all confined to the intronic region of HNF1B which might affect any of the splice events, exon skipping, or might activate cryptic spliced sites, or modify the alternative spliced isoforms. According to the previous citations, it has been reported that intronic SNPs could potentially create new splice sites, and might affect the intronic splicing process altogether, ultimately affecting the synthesis or physiological activity of the protein and affecting various phenotypes of human diseases (Pagani et al., 2002).
Figure 1-2-Locus Explorer plot for the HNF1B PCa risk locus at Chr17q12.
Graphical representation of PCa risk-associated SNPs. HNF1B gene is marked in purple, for the negative strand. Red color (the bell-shaped curve on the top along with other
20 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
color-coded curves) depicts the 3’UTR. The PCa SNPs have been marked with a triangle depicting the genotyped variants whereas the circles denote the imputed ones. The rs id is the tag SNPs identified in this region. The annotated bio features (involved in multiple biological pathways and potential drug targets) within the LNCaP cell-line; HNF1B is expressed in androgen-sensitive cell line is depicted by its track (second from bottom), Dnase track (third from bottom) depicts the regulatory regions (enhancer, promoter, insulator). Histone track (fourth from bottom) depicts the post-translation modifications. (Source: https://github.com/oncogenetics/LocusExplorer).
The function of HNF1B in PCa is still under consideration. The presence of the risk alleles rs11649743 and rs3760511 has been associated with higher levels of HNF1B transcripts in tumour tissue. Both these SNPs also showed association with reduced promoter methylation (Ross-Adams et al., 2016). Functional assays identified this gene to be suppressing epithelial- to-mesenchymal transition (EMT). However, this tumour suppressive function was seen to be lost after promoter methylation and silencing of this gene (Ross-Adams, et al., 2016).
The function of HNF1B has been mentioned in relation to liver development and regulating its vital functions such as expression of clotting factors, plasma proteins, cholesterol and fat metabolism (Costa, Kalinichenko, Holterman, & Wang, 2003). HNF1B is shown to be involved in re-differentiation and dedifferentiation of the tubular epithelial cells in the kidney (Omata et al., 2016). It also has a role in ureteric bud branching and commencement of nephrogenesis. Deficiency of HNF1B leads to certain major defects markedly in the late regionalised S-shaped body, with transformed morphology, degeneration of proximo-medial subdomain and amplified apoptosis (Heliot et al., 2013). HNF1B has a role in cellular metabolism and mutations which might lead to conditions such as mature-onset diabetes of the young type 5 (MODY5) (Teo et al., 2016) , type 2 diabetes mellitus (Goda, Murase, Kasezawa, Goda, & Yamakawa-Kobayashi, 2015) and reduced glucose metabolism (Kornfeld et al., 2013). The internal deletion that is a mutation occurring within POUS domain of HNF1B, leads to MODY5 syndrome. The patients suffering from such a medical condition generally are diagnosed with pancreatic hypoplasia (Teo, et al., 2016). Previous studies have shown that SNPs (rs7501939 and rs4430796) found at HNF1B loci, being linked to type 2 diabetes (T2D) in a Chinese and Caucasian populations (Gudmundsson, et al., 2007) and have also shown to be associated with T2D in a Japanese population (Miyake et al., 2009). Very recently, there have been many studies in China which correlate to the similar HNF1B variants rs7501939 and rs4430796
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 21
(Gudmundsson, et al., 2007) and might be involved in T2D risk in the Chinese population (Han et al., 2010; Li et al., 2013; Wen et al., 2010; Zhang, Qiao, et al., 2012).
There are more studies of HNF1B’s association with diabetes mellitus through polymorphisms reported in its microRNA binding site (Goda, et al., 2015). A SNP rs2229295 in the 3’ UTR of HNF1B altered the binding of miRNAs, hsa-miR-214-5p and hsa-miR-550a-5p, which may regulate its expression. Another micro RNA miR-802, has also been previously identified as a to silence HNF1B in the liver that leads to glucose intolerance, impaired insulin signalling and promotion of hepatic gluconeogenesis (Kornfeld, et al., 2013).
SNPs at the HNF1B locus have been found to have a role in endometrial and colorectal cancer (Painter et al., 2015; Setiawan et al., 2012; Spurdle et al., 2011). Through GWAS, the SNP rs4430796 has been found to be the common variant to be linked with T2D and endometrial cancer (Spurdle, et al., 2011). A replication study in a multiethnic population identified rs4430796 associations near the HNF1B locus with endometrial cancer risk (Long et al., 2012). A 3 stage GWAS found that SNPs rs1202524 and rs1202529, which are in strong LD, were associated with endometrial cancer risk among both Chinese women and women of European ancestry (De Vivo et al., 2014). Another study which was conducted in a large case control cohort on HNF1B variants confirmed the association of SNP rs78501939 with endometrial cancer risk. It was then concluded that HNF1B SNPs are associated both with the risk of type I and II endometrial cancer (Setiawan, et al., 2012). Analysis including a more comprehensive validation phase of this GWAS has since identified an additional 6 loci associated with endometrial cancer risk at genome-wide levels of significance (De Vivo, et al., 2014; Long, et al., 2012). Further with more fine-mapping studies conducted along with the in silico and laboratory analysis of the HNF1B locus, there has been an identification of candidate variants that mediate the risk of endometrial cancer. The SNPs which were detected in this study were rs11263763 (refer to Table 1-1) along with four other SNPs and an association with HNF1B methylation (Painter, et al., 2015). It had been reported that the minor alleles of rs11263763 and rs8064454 are linked with decreased HNF1B promotor activity. This study suggested the risk at the HNF1B locus is associated with endometrial cancer and is likely to be facilitated via altered HNF1B gene expression (Painter, et al., 2015). Shen and his research group mapped variation in HNF1B with respect to epithelial ovarian cancer risk and analysed DNA methylation and expression profiles across histological subtypes (Shen et al., 2013). Diverse single-nucleotide polymorphisms were found to be linked with invasive serous (rs7405776) and clear cell (rs11651755) epithelial ovarian cancer. Risk alleles for the serous
22 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
subtype were associated with greater HNF1B-promoter methylation in these tumors. Strong linkage between HNF1B expression and CIMP (CpG island methylator phenotype) methylation was detected, and the reciprocal nature of DNA methylation at the HNF1B- promoter CpG islands, against other CpG islands across the genome, proposes that HNF1B- promoter methylation is not merely a CIMP passenger event; in fact, HNF1B expression may even contribute to the hypermethylation phenotype. Collectively, these data depict divergent roles for HNF1B in these invasive EOC subtypes: a potential gain-of-function in clear cell ovarian cancer and loss-of-function in serous ovarian cancer, emphasizing the heterogeneity of this disease (Shen, et al., 2013). In another study carried out by Kristiansen et al (Kristiansen et al., 2015) there was genome-wide association for rs7501939 located in the intronic region of the HNF1B gene, which has revealed to be associated with predisposition to germ cell tumor.
This SNP has also been highlighted to be as diabetogenic SNP (rs7501939 (HRRec= 1.44, 95% CI = 1.18–1.76, P = 0.0001)) showcasing poor overall survival of multiple myeloma patients (Rios-Tamayo et al., 2016).
Table 1-1: HNF1B risk-associated SNPs identified in prostate cell lines
SNPs ODDS P- MAJOR MINOR MINOR CELL GENOTYPE RATIO VALUE ALLELE ALLELE ALLELE LINE (OR) FREQUENCY 95%CI rs11263762 1.20 9.52E-117 A G 0.4282 DUCaP AG
22RV1 AA rs11263763 1.23 3.06E-141 A G 0.5292 PC3 AG
DU145 AA
DUCaP AG rs11649743 0.89 8.96E-33 A G 0.1945 - - rs4795218 0.90 9.93E-28 A G 0.2362 22RV1 AG
1.3.2 HNF1B AND EPIGENETICS Genetic modifications and epigenetic mechanisms have been advocated to influence carcinogenesis. Next generation sequencing has led to identification of these modifications, further explaining the role of these mutants on the epigenome and in altering cellular
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 23
characteristics (You & Jones, 2012). Bubancova et al observed an unmethylated HNF1B region correlating to a higher survival rate in patients suffering from ovarian cancer (Bubancova et al., 2017). Early-stage breast cancer has been demonstrated to be associated with epigenetic reprogramming, via the substantial degree of hypermethylated CpG islands occurring in many homeobox genes such as HNF1B (significant degree of methylation: 73%) (Tommasi, Karm, Wu, Yen, & Pfeifer, 2009). Balch et al highlighted a methylation profile to be an important diagnostic marker to predict ovarian cancer recurrence, survival and resistance to chemotherapy (Balch, Huang, Brown, & Nephew, 2004). HNF1B has several transcriptional targets, one of them being HNF4α (Casemayou et al., 2017). Terasawa et al showed improved expression of HNF4α on inducing expression of HNF1B by a methyltransferase inhibitor, suggesting a role for demethylation of HNF1B to cause changes in the hepatocyte nuclear factor network (Terasawa et al., 2006). The study also reported the role of histone deacetylation organized with methylation to be involved in tumorigenesis in ovarian cancer subgroups. Colorectal, pancreatic and gastric cell lines were observed to have HNF1B methylated regions. The study carried out by Shen et al (Shen, et al., 2013) suggested that HNF1B-promoter methylation and its expression might contribute to a hypermethylation phenotype. There was identification of causal SNPs in the HNF1B region, the SNP-HNF1B promoter DNA methylation was found to overlap with a PRC2 mark in serous ovarian cancer. Molecular signatures like these (HNF1B status or CIMP) may help classify subtypes of clear cell carcinomas. Silva et al suggested hypermethylated HNF1B to be a useful epigenetic marker for non-invasive colorectal cancer screening (Silva et al., 2013). Mutations and methylation profile of HNF1B seems to be associated with different cancers, suggesting improved study of HNF1B and its associated variants would help predict the usage of HNF1B as a biomarker in different cancers.
1.5 HNF1B AND ITS SPLICE VARIANTS
Nearly partial genes within the mammalian genomes are prone to alternative splicing, this plays a vital role in terms of its participation to give rise to intricate human proteome (El Marabti & Younis, 2018)
Genes encoding splicing factors are differentially expressed in cancer compared to normal tissues, having decreased inter-individual expression differences in cancer (Chen & Weiss, 2015). Structural transcript variation is profoundly due to alternative splicing, however vital contributions from alternative polyadenylation, RNA editing, chimeric RNAs and alternative
24 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
promoter usage shaped by structural chromosomal rearrangements or otherwise transcriptional read-through across adjacent genes. Multiple mRNAs and downstream proteins are generated by alternative splicing from a solitary gene through the presence or absence of specific exons (Sveen, Kilpinen, Ruusulehto, Lothe, & Skotheim, 2016).
Different properties of transcription factors have been shown to be changed by alternative splicing and the distinctive usage of initiation codons (Han et al., 2017; Talavera, Orozco, & de la Cruz, 2009). Deviation in expression can be produced through species-specific changes in the quantity plus nature of mRNA transcripts formed. It is a known fact that alternative mRNA splicing events are conserved amongst species; but some researchers have advocated that a substantial amount of genes which have been identified to be alternatively spliced in humans, do not form numerous transcript variants in rodents (Yeo, Van Nostrand, Holste, Poggio, & Burge, 2005). HNF1A, B and HNF4A genes process diverse isoforms (3 isoforms each in HNF1A and HNF1B; 9 in HNF4A) in man by an arrangement of differential polyadenylation sites, alternate promoter usage and alternative splicing (Bach & Yaniv, 1993; Harries, Ellard, Stride, Morgan, & Hattersley, 2006; Harries et al., 2008). HNF1A, HNF1B, and HNF4A transcription factors are present in a synchronised feedback circuit in the majority of tissues, though the specific mechanism of collective regulation might vary amongst tissues (Ferrer, 2002). The alterations among transcript variants lead to the formation of proteins with different properties. HNF1A and B function just as dimers, so even minimal amounts of the isoforms can alter overall action in vivo (Harries, Brown, & Gloyn, 2009). The HNF1B gene encodes three transcript variants, HNF1B (A), HNF1B (B) and HNF1B (C) (refer to Figure 1-3), in human beings. HNF1B (A) and HNF1B (B) variants are analogous structurally and hence could exhibit functional anomalies. The difference between HNF1B A and B variants lies in the third exon; Variant B is short of 26bp in comparison to variant A. It was observed that higher levels of HNF1B (C), a repressor molecule could lead to a decline in HNF1B activity in rodents. Data have suggested that HNF1A (B), HNF1A(C), HNF4A3 and HNF4A9 might have a function in human beta cells as their existence can alter the MODY phenotype (Harries, et al., 2006; Harries, et al., 2008). A study by Harries et al. indicated the HNF1B (C) isoform as predominant in Benign Prostate Hyperplasia (BPH), amounting to 90% of the total gene expression. However, in PCa tissues the HNF1B (B) isoform was 95% of the total HNF1B expression. HNF1B (C) accounted for only 3% while the HNF1B (A) proportion did not change (Harries, et al., 2008). HNF1B (C) variant has also been detected to negatively control the GSTA (Glutathione-S-transferase A) promoter (Romero, Ng, & Kirby, 2006).
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 25
These differences in isoform specificity to targets and transcriptional property hint at changes to the whole activity of HNF1B gene. There is another classical transcript variant D and predicted HNF1B transcript variants (unpublished data) namely C, C’, E and I (novel variant identified in our laboratory) in prostate cell lines (Figure 1-3)
Figure 1-3- Characterised and annotated HNF1B Transcript variants
HNF1B transcript variants; A, B and D are the classical transcript variants having intact domains, whereas C, C’, E and I are the annotated transcript variants lacking the exons coding for the different domains. The transcript variants differ on the basis of the exons coding for different domains; exons 1 to 2 code for the dimerisation domain, exons 2 to 4 code for DNA-binding domain POU (Pituitary specific Pit-1, octamer transcription factor proteins Oct-1 and Oct-2, the neural Unc-86 transcription factor from C.elegans), POUH domains exons 4 to 9 code for transactivation domain. The number highlighted underneath the exons codes for nucleotides and these vary for exons according to the transcript variants.
26 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
1.6 HNF1B and its target genes
To understand cellular processes, it is essential to identify the target genes affected by transcription factors. Several studies have identified the individual target genes of HNF1B in hepatic, ovarian and kidney cell lines. Hu et al. integrated data from Dragon database of genes associated with prostate cancer (DDPC) with microarray data to identify potential target genes of HNF1B such as - (BCL2 Associated Athanogene 1) BAG1, erb-b2 receptor tyrosine kinase 4 (ERBB4), Estrogen receptor 1 (ESr1), Heat Shock Protein Family D (HSPD1), Nuclear Receptor Subfamily 4 Group A Member 1 (NR4A1) and Phosphatidylinositol-4, 5- Bisphosphate 3-Kinase Catalytic Subunit Gamma (PIK3CG) (Hu et al., 2013; Kim et al., 2008). One of the target genes, UDP Glucuronosyltransferase Family 2 Member B17 (UGT2B17), has been studied in PCa, and proposed to be a major factor regulating the intracellular level of androgens in many steroid-responsive tissues (Gregory, Hansen, & Mackenzie, 2000).
Overexpression of HNF1B in pancreatic B cells has been linked to apoptosis and the cell cycle, underlying its importance in B cell growth, though the exact pathway is unclear. There is a probability of apoptosis along with differential control of the cell cycle with increased expression of HNF1B in pancreatic B cells. Protein tyrosine phosphatase-BL (PTP-BL or ptpn13), an HNF1B modulated protein in the B-cell has been identified conjointly with its function in INS-1cells (Insulin Secretin B cell lines), an insulin secreting B-cell line (Welters, Oknianska, Erdmann, Ryffel, & Morgan, 2008). Welters and Morgan observed elevated PTP- BL protein levels on subsequent induction of HNF1B expression in INS-1 Flp-In T-Rex cells (Welters et al., 2006). Amplified HNF1B protein expression leads to compromised insulin secretion, augmented apoptosis and a decline in cell proliferation in INS-1 cells (Welters, et al., 2006) but the investigation of these responses showed their differential sensitivity to PTP- BL. The results from this study indicated the Wnt signaling pathway modulates mature B-cell growth through to regulation by HNF1B (Welters, et al., 2008). Osteopontin (OPN) gene expression has been found to be elevated in Ovarian Clear Cell Carcinoma (CCC) and its expression to be associated with HNF1B overexpression (Kato & Motoyama, 2008).
OPN has HNF1B functional binding sites in its promoter stretch, validating it as a direct target gene of HNF1B (Senkel, Lucas, Klein-Hitpass, & Ryffel, 2005). OPN plays a crucial role in tumorigenicity by inhibiting apoptosis or activating matrix-degrading proteases (Rangaswami, Bulbule, & Kundu, 2006). Research on ovarian CCC has suggested HNF1B to be at the center of a functional circuit, identifying susceptible targets of HNF1B (Shigetomi, Higashiura, Kajihara, & Kobayashi, 2012). Chk1 (Checkpoint kinase 1) protein has been observed to be
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 27
stimulated in the HNF-1B-overexpressing CCC cells. The inhibition of Chk1 expression specifically chemosensitizes HNF1B-overexpressing CCC cells in vitro, underlying its importance as a novel therapeutic target in HNF1B positive cells (Shigetomi, et al., 2012). Senkel et al identified HNF1B regulated in the human embryonic kidney cell line (HEK293); eight genes were found to be upregulated in ovarian CCC in comparison to other subtypes. Increased expression of HNF1B caused deregulation of genes including SPP1, DPP4, SAH, RBPMS, CD24, NID2, LAMB1, RHOB and SOX9 in ovarian CCC (Senkel, et al., 2005). SPP1 and Dipeptidyl Peptidase 4 (DPP4) had HNF1B binding sites in their promoter stretch; identifying them as direct targets. Additional studies in ovarian CCC revealed 22 genes to be involved as downstream targets of HNF1B (Kajihara et al., 2010). The genes identified were involved in drug metabolism and liver detoxification (GLRX, GPx3, TST, SOD2, NNMT, ANXA4, UGT1A1), oxidative stress (DPPIV, ACE2, Collectrin, TFPi2, Octamer4, PAX8), signal transduction pathways (MAP3K5/ASK1, mTOR), metabolism (GLUT2, ALDOB), anti- apoptosis, cell growth and migration, survival, immune response and redox status. HNF1B has been observed to be directly regulating HNF4α in humans (Hatzis & Talianidis, 2001). Various pathways related to HNF1B need to explore which could provide new insights to developing anticancer agents targeting this HNF1B regulated circuit. Cuff et al observed HNF1B induced expression of clotting factors in tumor cells, including elements involved in clotting cascade like prothrombin, fibrinogen, factor XIII, contributing to a prothrombotic state in malignancy (Cuff et al., 2013). The findings also indicated starch and sucrose metabolism genes as HNF1B targets. Xu et al identified a putative binding site in the NNMT (Nicotinamide N- methyltransferase) gene promoter region, which is greatly expressed in papillary thyroid cancers along with cell lines (Xu, Capezzone, Xu, & Hershman, 2005). These findings suggest HNF1B to be a transcriptional activator of NNMT gene expression in some papillary thyroid cancers. Co-immunoprecipitation experiments by Choi et al indicated zyxin, a focal adhesion protein, as a novel interacting partner of HNF1B in renal epithelial cells (Choi, McNally, & Igarashi, 2013). The study established the importance of an additional LIM domain of zyxin interacting with HNF1B. Additionally, it established the role of zyxin bound to CREB Binding Protein (CBP) stimulating the transcription of HNF1B. Co-localization of zyxin with HNF1B was observed in the nucleus (Choi, et al., 2013). Increased expression of zyxin led to transcriptional activity of HNF1B, on the contrary, siRNA (RNA mediated) silencing of zyxin impeded the HNF1B dependent transcription. Expression of dominant-negative mutant HNF1B, silencing zyxin, resulted in reduced EGF-accelerated cell migration. These results suggest a novel route
28 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
for regulating HNF1B which is vital for renal epithelial differentiation (Grisanzio et al., 2012). Previous studies have reported HNF1B is needed for renal tubulogenesis by regulating the gene expression of SOCS3 (Ma et al., 2007). This pathway showcases another route through which HNF1B might control tubulogenesis throughout kidney development. Also, the initial decline in HNF1B expression is linked with the transient overexpression of one of its target genes, SOCS3, which is a requisite for renal repair.
In summary, genome-wide association studies have shown at least 100 susceptibility loci for PCa risk, including HNF1B. It has been revealed that the expression of HNF1B is related to cancer risk in numerous tumors, including prostate, endometrial, renal, ovarian, hepatocellular and pancreatic carcinoma. However, the possible pathogenic mechanisms of HNF1B in PCa and regulatory mechanisms are still less understood. Along with this, we still need to define the role of HNF1B splice variants in PCa which will lead to a better understanding of the mechanisms and might lead to new approaches in therapeutic applications.
1.7 ROLE OF HNF1B IN CANCER
Some genes, including HNF1B, have both oncogenic and tumor suppressor functions.
1.7.1 Tumour-suppressive function Pancreatic hypoplasia has been described in several individuals with HNF1B-associated diseases (Clissold, Hamilton, Hattersley, Ellard, & Bingham, 2015). Janky et al discovered that HNF1A/B are the topmost enhanced principal regulators of the genes which are expressed in the normal pancreatic tissue in comparison to the pancreatic ductal adenocarcinoma (PDAC) regulatory network (Janky et al., 2016).The study established that on immunohistochemistry staining of PDAC samples there was minimal expression of HNF1B in well-defined tumors and not any expression in six poorly defined PDAC samples. In another study, HNF1B was seen to be down-regulated in vitro in PDAC cells through a miRNA mechanism encompassing the-miR-24 and/or hsa-miR-23a (Yonemori, Kurahara, Maemura, & Natsugoe, 2017). In the proposed mechanism, HNF1B downregulation in PDAC tissue leads to loss of expression of the adhesion molecule E-cadherin, inducing epithelial-mesenchymal transition (EMT) and permitting cells to separate from cell agglomerations and migrate. Similarly, HNF1A has been shown as an important regulator of the transcriptome in pancreatic tumor tissues, proposed to be a tumor suppressor in the pancreas (Hoskins et al., 2014). Hoskins et al. observed that inducible overexpression of HNF1A in pancreatic tumor-derived cells leads to growth
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 29
inhibition, a G0/G1 cell cycle arrest, and apoptosis. Biological Pathway based investigation of GWAS data is a compatible method to ascertain clusters of genes or biological pathways supplemented with disease-linked SNPs whose singular affects may be too trivial to be identified by customary single-locus methods. Li and colleagues based on text searches and online resources compiled pathways involved in pancreatic cancer and identified a significant association between five pathways i.e. apoptosis, pancreatic development, hedgehog, Helicobacter pylori Lacto/neolacto and Th1/Th2 immune response. These significant genes involved in these pathways highlighted HNF1B along with HNF1A, HNF4G, and PDX1 as pancreatic development genes that encode important constituents of the transcriptional networks that manage embryonic development of the pancreas, and maintain homeostasis in the adult gland (Li et al., 2012). Numerous studies have found an association of HNF1B with several types of renal disease, primarily renal cysts and dysfunction; renal cysts are allied with early-onset diabetes (Edghill et al., 2008). Buchner et al indicated HNF1B’s role as a therapeutic target and a possible tumor suppressor in Renal Cell Carcinoma (RCC). HNF1B is important for the maintenance of renal tissue growth; dysregulation is linked with invasive alteration and dedifferentiation (Buchner et al., 2010). Lebrun et al identified somatic deletion of HNF1B and germline mutations in renal tumors (Lebrun et al., 2005). Chromophobe renal cell carcinoma (RCC) is a sporadic renal cancer which accounts for 4-5 % of all types of kidney cancer. Deregulation of HNF1B expression has been shown to be important in the pathogenesis of ChRCC and might aid as a worthy “classic” biomarker (Verhave, Bech, Wetzels, & Nijenhuis, 2016; Wang, Mao, Yang, & Jeng, 2013). Biallelic HNF1B inactivation in ChRCC has been associated with germline mutation and somatic gene deletion, suggesting that germline mutations in the gene might predispose to renal tumors. This study suggested HNF1B as a tumor suppressor in ChRCC under PKHD1 expression control (Rebouissou et al., 2005). Ross-Adams et al linked intronic SNPs with HNF1B expression levels and epigenetic silencing in multi-ethnic populations for both ovarian and PCas (Ross-Adams, et al., 2016). The study suggests the role of HNF1B as a pro-differentiation factor that represses EMT in normal unmethylated tissue. Once methylated, the activity of HNF1B as a tumour suppressor is lost during to development of PCa.
Yamamoto et al. established that HNF1B immunoreactivity contrasted significantly between clear cell carcinoma (CCC) and other histologies equally in the ovary and the endometrium, proposing HNF1B to be an eminent marker for differentiating CCCs from other lesions in both the ovary and the endometrium (Yamamoto et al., 2007). . Decreased expression of HNF1B
30 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
could impart drug resistance in OC and HNF1B may accomplish its drug resistance-associated functions through four pathways comprising p53 signaling, focal adhesion, ErbB signaling in addition to apoptosis (Li et al., 2014).
1.7.2 Oncogenic role HNF1B was shown to have an eminent role in cancer etiology. This gene was one of the first to be identified in clear cell ovarian cancer. As previously pointed out, the gene was shown to be associated with serous ovarian cancer at the genome wide-significant level. It has been reported to be overexpressed in clear cell tumors whereas it is silenced in serous tumors. It was revealed by Kao and colleagues that the overexpression of HNF1B is specific for ovarian CCC amongst ovarian carcinomas (Kao, Lin, Lin, Jeng, & Mao, 2012). Ablation of HNF1B expression in ovarian CCC cells lead to a substantial proliferation, while increased expression of HNF1B in a serous Ovarian cancer (OC) cell line affected cell growth (Okamoto et al., 2015)
In the study carried out by Shen et al (Shen, et al., 2013), it was highlighted that specific variants of HNF1B may have susceptibility towards clear cell carcinoma, prostate, and uterine cancer, and also highlighted that certain other variants might be responsible for diabetes (Winckler et al., 2007) and serous ovarian cancer. This was the first study which probed to see the effects of HNF1B overexpression in endometriosis, and the results were in accordance that HNF1B may have an oncogenic role in clear cell ovarian cancers which was also highlighted by Gounaris et al (Gounaris, Charnock-Jones, & Brenton, 2011), further suggesting this as to be an essential step involved in endometriosis. Matsui et al highlighted the role of ERBB2 which has been recognized as a causal factor of breast cancer. Further investigation led to the discovery that ERBB2 is activated by HNF1B, though it was observed that both these genes work independently in the transformation process. HNF1B weakly induces EMT whereas ERBB2 stimulates proliferation in the case of the cell line used in this study i.e. NMuMG cells. Further, it was highlighted that some of the tumourigenic properties i.e. anchorage-independent growth focus formation i.e. an assay used to determine the changing state of an oncogene, induction of EMT along with an invasive phenotype are all caused by the DNA-binding domain-dependent manner of HNF1B. As per the initial mention of the collaborative function of ERBB2 with HNF1B, the group anticipated that HNF1B overexpression was involved in poor metastasis survival of HER2-positive breast cancer (Matsui et al., 2016).
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 31
1.8 HNF1B AS A BIOMARKER
HNF1B has been well characterised in liver, pancreas and kidney with diagnosing its related disease a challenging task due to phenotypic variability. Among the first to identify the potential of HNF1B in hepatocellular carcinoma (HCC), Ninomiya et al. highlighted the ratio of HNF1A/HNF1B expression in HCC tissues to be higher in well-differentiated cases compared to undifferentiated and poorly differentiated cases (Ninomiya et al., 1996) . Similarly, Wang et al reported the ratio of HNF1A: HNF1B expression to be related to histologically differentiated disease (Wang et al., 1998) . The expression of HNF1B was found to be higher in differentiated HCC compared to non-cancerous tissues. These studies laid the groundwork of unwinding the role of HNF1B as a biomarker. Several studies have highlighted the role of HNF1 family members in regulating alpha-fetoprotein (AFP) promoter during hepatic development and carcinogenesis (Bois-Joyeux & Danan, 1994; Lazarevich, 2000) Immunohistochemistry studies of the expression of AFP, HNF1A and HNF1B in liver transplant patients with HCC revealed that HNF1B has been associated with serum AFP level and AFP expression. Transcriptional regulation of AFP through HNF1B may function during different stages of HCC progression following recurrence. The expression of HNF1B in tumour tissue thus can foretell relapse and death after transplantation (Shim et al., 2013). Additionally, Yu et al. investigated the expression of HNF1B with clinicopathological features and prognosis in HCC and cholangiocarcinoma (ICC) patients (Yu, et al., 2015). HNF1B expression was found to be positively correlated with the change of phenotype on recurrence in HCC, thereby showing poor prognosis. However, no correlation was found with its expression in ICC and survival. Further studies need to be conducted on developing HNF1B as a prognostic marker predicting recurrence in HCC.
Pancreatic abnormalities have been tested for HNF1B mutations and its prospectus as a biomarker being widely debated in renal hyperplasia and cysts (Clissold, Shields, Ellard, Hattersley, & Bingham, 2015) Yang and colleagues investigated the role of HNF1B in PDAC as a diagnostic marker in a large cohort of 127 primary and 17 metastatic PDACs, 47 biliary adenocarcinomas and 231 pancreaticobiliary carcinomas (Yang et al., 2018) Majority of pancreatic and biliary epithelium carcinomas had an expression of HNF1B, with statistical analysis showing 84% sensitivity and 85% predictive value overall. The expression of HNF1B was associated with tumour size and grade highlighting the potential of the gene as a biomarker in PDAC. Likewise, HNF1B was shown to be part of five gene expression signature predictive of relapse in prostate cancer patients. Kaplan-Meier analysis revealed that relapse of predictor
32 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
markers are highly useful in classification of patients into subgroups with distinct relapse-free survival after therapy (Glinsky, Glinskii, Stephenson, Hoffman, & Gerald, 2004) . Furthermore, Harries et al reported a probable role of alternatively spliced mRNA of HNF1B and MSMB genes in the cause of prostate cancer (Harries, Perry, McCullagh, & Crundwell, 2010)
Ovarian CCC has one of the worst prognosis of entire epithelial OC. Tsuchiya et al performed an oligonucleotide array technique to detect genes implicated in CCC (Tsuchiya et al., 2003). Among several upregulated genes HNF1B upregulation was investigated at mRNA and protein level. Immunohistochemical results showed that HNF1B protein had nuclear staining for clear cell carcinoma specimens on the other hand non-clear cell carcinoma specimens displayed no immunostaining for HNF1B (focal or faint staining in the nucleus). Apoptotic cell death was seen in TOV-21G and JHOC-5 ovarian CCC cell lines induced by reduction of HNF1B expression by RNA interference. Results suggested that HNF1B is an excellent CCC specific molecular marker and can also be used as molecular target therapy for ovarian clear cell carcinoma (Tsuchiya, et al., 2003) Similarly, Huang and colleagues also showed expression of HNF1B in ovarian CCC patients with a statistical specificity of 76.5% and selectivity of 85.2% (Huang, Cheng, Ji, Zhang, & Li, 2016) In a study investigating HNF1B expression in ovarian clear cell tumours, there was a clear distinction of the levels of the protein being present between clear cell carcinomas vs non clear cell carcinoma. The results indicated a strong belief in developing HNF1B as a molecular marker for ovarian CCC irrespective of benign or malignant lesions (Kato & Motoyama, 2008). Cuff et al defined HNF1B as an extensive marker for clear cell phenotypes, endorsing a mechanistic association to glycogen aggregation and thrombosis. This outcome alludes to a novel mechanism of tumour linked thrombosis centred on cancer cells directly producing clotting factors (Cuff, et al., 2013).
Hoang and colleagues advocated HNF1B and ER (oestrogen receptor) as a diagnostic panel to be deliberated for sorting out endometrioid from clear cell carcinoma besides serous carcinoma of the endometrium (Hoang et al., 2014). CCC of the urinary tract has been identified as a rare malignancy, which mimics the clear cell carcinoma of the female genital tract morphologically (Young & Scully, 1985). One of the studies described HNF1B usefulness as a biomarker in distinguishing clear cell adenocarcinomas of the bladder/urethra besides other primary tumours of the urinary bladder, specifically aggressive urothelial tumour with clear cell change (Brimo et al., 2011). Davidson et al advocated the central part of HNF1B marker in effusion diagnosis
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 33
is the disparity among CCC in addition to serous carcinoma, in arrangement with a serous carcinoma marker, like Wilms tumour 1 (WT1) (Davidson, 2014).
In a nutshell, a wide array of expression changes in different cancer and its subtypes does point towards the role of this gene as a potential biomarker. However, even after identifying its involvement in development and progression to tumour relapse, the details about the regulatory pathways still are missing, further studies which highlight this missing links can help in finally helping the potential of HNF1B to become a biomarker in different cancers.
1.9 SUMMARY
HNF1B was first reported during the end of 20th century as a potential candidate gene for MODY; extensive research indicated it as an important gene having role in tumorigenesis. HNF1B profiles in different tumours throw a light on different mechanisms governing HNF1B and its expression. GWA studies identifying HNF1B risk loci in prostate cancer to epigenetic alterations in ovarian cancer has unravelled the role of DNA methylation and risk alleles to be used as biomarkers for disease prognosis. The function of HNF1B as a tumour suppressor gene in renal cell carcinoma has highlighted its role in development of tumour. Therapeutic approach via restoring HNF1B function in ovarian cancer and understanding its role in drug resistance has broaden the scope of using HNF1B as a therapeutic target. Moreover, through current findings, there has been understanding that transcript variants might gradually become crucial for developing biomarkers and effective therapeutic strategies. Regulatory pathways and mechanisms involving HNF1B still needs to be elucidated and studied in depth to delve into the networking of HNF family. Understanding splice variants of HNF1B and their role will broaden the scope of pursuing new downstream targets; it’s associated signalling and transcriptional efficacy governing different gene sets. Further study needs to be calibrated around HNF1B and its splice variants to apprehend its ever growing importance in different cancers.
2. HYPOTHESIS AND AIMS
The underlying objective of my project is to characterise and identify the functional role of HNF1B transcript variants in PCa, and determine if any of the transcript variants are regulated by androgens or docetaxel. This hypothesis will be examined using the following related aims
34 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
AIM 1. Characterisation of HNF1B transcript variant expression in PCa cell lines and clinical samples.
• In-silico analysis using GTEx, Oncomine, TCGA-Spliceseq, ChIP, and RNA-Seq • mRNA and protein analysis using RT-PCR, qRTPCR and Western blotting.
AIM 2. Role of androgens and docetaxel in regulating the expression of HNF1B transcript variants.
• mRNA and protein analysis post androgen and docetaxel treatment using RT-PCR, qRTPCR and Western blotting.
AIM 3. Functional effects of HNF1B and its transcript variants in PCa cell lines.
• Generation of loss of function (Knockdown; using siRNA,) validating at mRNA and protein level using qRTPCR and Western blotting and performing functional assays using Incucyte in DU145 cells. • Gain of function models (Over-expression) of transcript variants A, B and C and confirmation at the mRNA level using qRTPCR along with functional assays in PC3 cells.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 35
Chapter 2: Materials and Methods
This chapter describes the materials and methods that have been used to perform the experiments described in this thesis. Any alterations and/or additions to the materials and methods specific to work described in a single chapter will be mentioned in that chapter.
2.1 Cell culturing conditions
A panel of cell lines representing PCa (DuCaP, VCaP, LNCaP, C4-2B, DU145, PC3 RWPE2), and benign prostate (RWPE1) were used in this study. All cell lines were obtained from the American Type Culture Collection (ATCC). RWPE-1 and RWPE-2 cell lines were grown in Keratinocyte-SFM (1X) (K-SFM, catalog number - 17005-042), whereas, the rest of the cell lines were grown in RPMI1640 (1X) (Life Technologies, catalog number - 11835-030) supplemented with 5% fetal bovine serum (FBS, Sigma, Catalog number - F2442). Cells were passaged at 70 – 80% confluence by washing twice with phosphate-buffered saline (PBS) and detachment of RWPE-1 and RWPE-2 cell lines was performed with TrypLE™ Select Enzyme (1X), no phenol red (Life Technologies, catalog number - 12563-011) and other cell lines were detached with Trypsin/EDTA Solution (TE) (Life Technologies, Catalog number - R-001-100). These cultures were incubated at 37°C and 5% CO2 and the media was changed every two days. The cell cultures were passaged by washing the flask with phosphate buffer saline (PBS; Life Technologies), detaching the cells with 0.05% Trypsin or Ethylenediaminetetra-acetic acid (EDTA; Life Technologies) and resuspended in fresh media. The cells were pelleted by centrifuging at 1000g for 5 minutes. The media was aspirated and the pellet was resuspended in 5 ml fresh media and pipetted thoroughly till the pellet was completely dissolved. The mixture was then transferred into a T-25cm2 flask. The cells were passaged after alternate days and transferred from a T-25cm2 flask to a T-75cm2. Once the cells were confluent enough, they were trypsinised and counted. Cryo-media was prepared containing 90% FBS, 10% dimethoxysulfoxide (DMSO). The cell solution was diluted to 1 x106 cells/ml and 1 ml of the mixture added to each cryovial. Vials were then placed in a Mr Frosty Freezing Container (Thermo Fisher Scientific, Australia) at -80°C, and transferred to a liquid nitrogen tank after 48 hours for long term storage. The vials were subsequently thawed and used for the assays as required.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 37
2.2 RNA extraction and cDNA synthesis
RNA was extracted from DuCaP, VCaP, LNCaP, C4-2B, DU145, PC3, RWPE1 and RWPE2 cells using the BIOLINE extraction kit according to the standard protocol (Bioline, Catalog number – BIO-52073). RNA concentration was measured using the NanoDropTM1000 (Thermo Scientific, Biolab, Scoresby, VIC, Australia). 1 μg of RNA was reverse transcribed to cDNA using SuperScript® III Reverse Transcriptase (Invitrogen, Catalog number - 18080- 044).
2.3 Designing variant-specific primers
RT-PCR primers were designed using the NCBI Primer design, the sequences of each transcript variant along with the annotated ones were saved in the molecular biology tool software serial cloner and the primers as desired were run using the In-silico PCR to check the exact product size and the primer set for every variant (Table 2-2). Considering most of the exons are the same for each transcript variants, in case of RT-PCR; the forward primers were designed for A and C on the second exon; and the same for B and C’. The reverse primers were the same for A and B, i.e. on exon 5 and on exon 4 for C and C’. D and E had forward primer on exon 6. The reverse primer was designed on exon 9 for D and E. Refer to (Table 2-1) for primer sequence. For qRTPCR, the forward and reverse primers were designed intron spanning; for transcript variant, A, B, C and C’ the forward primer was designed on exon 2 and 3; and the reverse primer for A and B were designed on 5th exon. The reverse primer for C and C’ was designed on exon 4. For D; it was designed intron spanning exon 6 and 9; whereas for E, it was intron spanning exon 6 and 6. The reverse primer was the same for D and E on exon 9 (Table 2-3).
2.4 Real time (rRT) and quantitative real-time PCR (qRTPCR)
RT-PCR was carried out with the cDNA from a panel of cell lines representing PCa (DuCaP, VCaP, LNCaP, C4-2B, DU145, PC3, RWPE2), benign prostate (RWPE1). Each reaction comprised of a final concentration of 1x 10x PCR Buffer,-Mg, final concentration of 1.5 mM of 50 mM MgCl2, the final concentration of 0.2 mM each of 10 nM DNTP Mix, the final concentration of 0.5 μM of 10 μm forward and reverse primer. The template DNA used was 2 μl, the Taq DNA polymerase (5U/ μl) final concentration varied from 1.0-2.5 U/rxn, the final reaction mixture was 25 μl (Table 2-2). The cycling parameters were 94˚C for 3 minutes (initial denaturation), 40 cycles of 94 ˚C for 45 seconds (denaturation), annealing for 55 ˚C (depending on primer Tm).
38 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Table 2-1: The primers used for RT-PCR for HNF1B transcript variants.
RT-PCR SEQUENCE (5’-3’) Expected Band Size
rHNF1BAF rHNF1BBF ACCTGGTACGTCAGAAAGCA A: 693 rHNF1BCF B: 615 rHNF1BC’F C: 601
rHNF1AR rHNF1BBR ACCATCAGGTGAGAGGAGAT B:615
rHNF1BCR rHNF1BC’R AGATCCGTGGCAAGAACCAG C’:523
rHNF1BDF rHNF1BEFP GTCTCAGGAGGAGGTTTGCC D: 309
rHNF1BDRP rHNF1BERP GTGGATTGTCTGAGGTGCCA E: 428
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 39
Table 2-2: RT- PCR reaction protocol
Component 25μl rxn Final conc.
10Xpcr Buffer, -Mg 2.5 1X
50 mM Mgcl2 0.75 1.5 mM
10 mM dNTP Mix 0.5 0.2 mM
10 μm forward primer 0.5 0.2 μM
10 μm reverse primer 0.5 0.2 μM
Taq DNA Polymerase 0.1 1.0-2.5U/rxn
Template DNA 2 μl 1-500 ng
Autoclave water to 25 μl -
40 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Figure 2-1-RT-PCR Primer Scheme.
HNF1B transcript variants along with the primers marked on the specific exon; the forward and reverse primers for each variant; variant A, B, C, and C’ share the same forward primer in the exon 2, the reverse primer was same for variant A and B on exon 5. The reverse primer for variant C and C’ was same and was made specifically using the 4th exon and the 3’ UTR region. The forward and reverse primers for the variants D and E on exon 6 and exon 9 were the same.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 41
Table 2-3- The primers used for q RT-PCR for HNF1B transcript variants
Q-PCR SEQUENCE (5’- 3’) EXPECTED BAND SIZE qHNF1BAF qHNF1BCF ATTCAACCAGACAGTCCAGAGT
qHNF1BAR TGTTTCCCTGCTGGCTGTAG 527
qHNF1BBF qHNF1BC’F CGAGAGATCCTCCGACAGTTC
qHNF1BBR TGTTTCCCTGCTGGCTGTAG 466
qHNF1BCR qHNF1BC’R AAGAACCAGGATGGTTGGGTTG C: 552 C’:491 qHNF1BDF GGCAATTGCACAAATGTCCTCT
qHNF1BDR GTGGATTGTCTGAGGTGCCA D: 196
qHNF1BEF GGCAATTGCACAAATGTACGC
qHNF1BER ATCGTGGGAGAGGCATTGTG E: 384
42 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Forward Primer
Reverse primer
Figure 2-2-qRTPCR Primer Scheme:
Forward and Reverse primers marked for HNF1B transcript variants; A and C transcript variants have the same forward primer designed on exon 2 and 3; the reverse primer is designed on exon 5. For transcript variant C and C’ the forward primer was designed on exon 2 and 3 and the reverse primer was designed on exon 4. The forward primer was designed on exon 6 and 9 and for variant, E was designed on exon 6 and 8. The reverse was designed on exon 9 for D and E.
2.5. Protein expression level of HNF1B isoforms in PCa cell lines
The cell lysate from PCa cell lines (VCaP, DUCaP, LNCaP, C42B, PC3, and DU145) was isolated with RIPA buffer (50mM Tri-HCl pH 7.5, 150 mM NaCl, 1% SDS, 1% Triton X-100, 1% CHAPS/IGEPAL and 1X Protease Inhibitor Cocktail) and the western blot was carried out (Odyssey® system) with an anti-HNF1B antibody for the C-terminal region [ERP6334 (2)] (ab 128912) from Abcam, Rabbit monoclonal) and another for the N-terminal region [EPR18644- 13](ab213149) from Abcam, Rabbit monoclonal) (Figure 2-3) to confirm the protein expression level of HNF1B transcript variants. Briefly, the protein concentrations were
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 43
measured using a standard BCA assay with Pierce™ BCA Protein Assay Kit (Sigma). 20 μg of proteins were loaded on to a freshly prepared 12% gel electrophoresed for 2 hours at 110v. The gel was then transferred onto a nitrocellulose membrane (Bio Trace NT, Pall Life Sciences, United States). For protein transference, the membrane was assembled in the following order; pre-wet sponge, paper with the membrane kept in transfer buffer (200 ml methanol, 700 ml water and 100 ml of 10x Transfer buffer). Membranes were subsequently blocked in Odyssey® blocking buffer (LI-COR) for 30-45 min at room temperature. The primary antibody, Anti- HNF1B antibody, was diluted in Odyssey® blocking buffer (1:1000) and incubated with the membrane overnight at 4˚C. After washing membranes four times (10 min each) with TBS- Tween, Cell Signalling conjugated secondary antibody was diluted in Odyssey buffer (1:10,000) and incubated with the membranes for 1 hour at room temperature. Membranes were washed three times (10 min each) with TBS-Tween and then scanned with the Odyssey® system (LI-COR).
Figure 2-3- N and C terminal Antibody Schema:
The N-terminal ab will bind to the 1-150 amino acid sequence (exon 1), and the C-terminal ab will bind to the 500-557 amino acid sequence (exon 9) of the A, B, D and E transcript variant.
2.6. Statistical Analysis
The experiments were carried out in triplicates for assays, the data retrieved was stored in an excel sheet which was then exported to GraphPad Prism7, a statistical analysis software. T-test was carried out to compare means of two populations and One-way ANOVA; which is usually used to calculate the mean for comparing several groups, the groups are compared to the control.
44 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Chapter 3: Characterisation and expression of HNF1B transcript variants in prostate cell lines
3.1. Introduction
HNF1B’s role is still not defined in PCa. Recent studies have shown its involvement in cancerous pathways, along with the mention of specific domains for promoting its metastatic property. Various studies by Harries (Harries, et al., 2009) and Adams et al (Ross-Adams, et al., 2016) have mentioned HNF1B as the possible gene that is altered in PCa. There were isoform shifts that were reported in one study carried out by Harries et al (Harries, et al., 2010), which highlighted the probable significance of alternative splicing and its involvement in essential disease mechanisms. Therefore, we need more clarity on the mRNA splicing of this gene and hence the first aim of this study was to characterise the HNF1B transcript variants. So far, we know that there are three characterised transcript variants A, B, and D, in prostate cancer, though there are several other annotated transcripts, which need to be defined in PCa.
3.2. Methods
The common Materials and Methods have been outlined in Chapter 2. We referred to online tools such as Cbioportal for cancer genomics (www.cbioportal.org), where we extracted the raw data for HNF1B expression in various tissue samples and plotted the same using GraphPad Prism7. Next, we referred to the Genotype-Tissue Expression (GTEx) (https://www.gtexportal.org/) project to distinguish HNF1B gene expression across organs. It is a resource database and associated tissue bank for the scientific community to study the relationship between genetic variation and gene expression in human tissues. Oncomine is a cancer microarray database along with the web-based data-mining platform, which is designed to enable discovery from genome-wide expression analyses. Oncomine comprises of 65 gene expression datasets which contain around 48 million gene expression measurements from over 4700 microarray experiment (https://www.oncomine.org/). There are other databases like TCGA spliceseq, which gives an overview of each alternative mRNA splice variant. RNA-seq
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 45
data is combined along with computational tools. There is a designated Percent Spliced (PSI) value for each splicing event. The TCGA sample reads are aligned to the splice graph, and the summary statistics are generated for each exon and splice junction. There is an intuitive graphical representation of splicing patterns, read counts along with various statistical summaries, with percent spliced in. We investigated the HNF1B transcript variants in PCa in those datasets and then to further validate this, designed specific RT and q RT-PCR primers that we're able to identify HNF1B transcript variants in prostate cell lines.
3.3 Results
3.3.1 Expression analysis of HNF1B using bioinformatics tool (cbioportal, GTEx, Oncomine, TCGA-Spliceseq, ChIP and RNA-seq)
The expression of HNF1B was inspected through the cbioportal (Figure 3-1) and GTEx-portal (Figure 3-2). HNF1B was found to be expressed in prostate tissue as well as in kidney-cortex, colon, pancreas, small intestine, stomach, bladder, liver, lung, and testis, as shown in Figure 3-1. Next, HNF1B gene expression was mined in various cancer datasets using Oncomine (microarray data). In the Bittner Multi-Cancer dataset, a microarray dataset, used to determine the expression of a gene in multiple cancers, the expression of HNF1B was compared across various cancers (Figure 3-3). In the Grasso Prostate dataset; the microarray represents the variations of aggressively treated metastatic cancer (CRPC) with primary prostate carcinoma; the expression of HNF1B in CRPC was downregulated in comparison to prostate carcinoma, having significant P-value of 0.05 and fold change of 2.0 (Figure 3-4). We also performed multiple sequence alignment of publically available isoforms, highlighting the difference in the protein sequence (Figure 3-5). There was isoform difference in A and B (183-208 absent); followed by difference in A and C (401-557 absent). Next, we surveyed for the origin of the HNF1B transcript variants in PCa using TCGA Spliceseq (projects.insilico.us.com/TCGASpliceseq) (Figure 3-6).There are seven splice events, out of which for the HNF1B transcript variant had one of them with exon skipping in exon 7 and 8 coding for the transactivation domain and another one with the same exon skipping in exon 8. An alternative acceptor site was seen in one of the variants in exon 3 coding for the DNA- binding domain, which probably explains for derivation of HNF1B variants B and C. There are other splice events, alternate donor (AD), retained intron (RI), mutually exclusive exons (ME), alternate promoter (AP), alternate terminator (AT) which could possibly explain the reason for the other annotated splice variants C,C’,E, and I. We next wanted to check the expression of
46 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
HNF1B transcript variants in prostate cell lines using RNA-seq data (kindly provided by John Lai) (Figure 3-7). The above findings highlight there is an expression of HNF1B in the prostate and the different splice variants of HNF1B could have originated as per the splice events which occur during gene expression.
Figure 3-1-HNF1B expression in various tissue samples obtained from Cbioportal:
Tissue expression levels for HNF1B (mRNA): It is expressed in PCa as well as bladder, colorectal, kidney renal cell, liver, lung, ovarian and pancreatic cancer.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 47
Figure 3-2- HNF1B Gene Expression in GTEx portal.
HNF1B mRNA expression was observed in prostate tissue as well as from kidney-cortex, EBV-transformed lymphocytes, colon, pancreas, small intestine, stomach, bladder, liver, lung, and testis.
48 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Figure 3-3-HNF1B expression in Bittner-Multi-Cancer Dataset.
HNF1B is expressed in PCa as well as various other cancers. Legend at bottom of the figure indicates each cancer type. P-value (Calculated probability) is 0.05 and the fold change indicates the change from primary (various cancers) to concluding value (2.0) (for PCa) (Source: The Oncomine™ Platform (Life Technologies, Ann Arbor, MI)).
Figure 3-4-HNF1B expression in localised prostate carcinoma versus castrate resistant metastatic prostate carcinoma in the Grasso Prostate dataset.
(0) Samples not scored (1.) Prostate Carcinoma (2) Castrate Resistant Metastatic Prostate Carcinoma (CRPC): HNF1B expression was low in castrate resistant metastatic PCa compared to localised disease (Source: The Oncomine™ Platform (Life Technologies, Ann Arbor, MI).
50 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
HNF1B A MVSKLTSLQQELLSALLSSGVTKEVLVQALEELLPSPNFGVKLETLPLSPGSGAEPDTKP HNF1B B MVSKLTSLQQELLSALLSSGVTKEVLVQALEELLPSPNFGVKLETLPLSPGSGAEPDTKP HNF1B C MVSKLTSLQQELLSALLSSGVTKEVLVQALEELLPSPNFGVKLETLPLSPGSGAEPDTKP
HNF1B A VFHTLTNGHAKGRLSGDEGSEDGDDYDTPPILKELQALNTEEAAEQRAEVDRMLSEDPWR HNF1B B VFHTLTNGHAKGRLSGDEGSEDGDDYDTPPILKELQALNTEEAAEQRAEVDRMLSEDPWR HNF1B C VFHTLTNGHAKGRLSGDEGSEDGDDYDTPPILKELQALNTEEAAEQRAEVDRMLSEDPWR
HNF1B A AAKMIKGYMQQHNIPQREVVDVTGLNQSHLSQHLNKGTPMKTQKRAALYTWYVRKQREIL HNF1B B AAKMIKGYMQQHNIPQREVVDVTGLNQSHLSQHLNKGTPMKTQKRAALYTWYVRKQREIL HNF1B C AAKMIKGYMQQHNIPQREVVDVTGLNQSHLSQHLNKGTPMKTQKRAALYTWYVRKQREIL
HNF1B A RQFNQTVQSSGNMTDKSSQDQLLFLFPEFSQQSHGPGQSDDACSEPTNKKMRRNRFKWGP HNF1B B RQF------SQQSHGPGQSDDACSEPTNKKMRRNRFKWGP HNF1B C RQFNQTVQSSGNMTDKSSQDQLLFLFPEFSQQSHGPGQSDDACSEPTNKKMRRNRFKWGP
HNF1B A ASQQILYQAYDRQKNPSKEEREALVEECNRAECLQRGVSPSKAHGLGSNLVTEVRVYNWF HNF1B B ASQQILYQAYDRQKNPSKEEREALVEECNRAECLQRGVSPSKAHGLGSNLVTEVRVYNWF HNF1B C ASQQILYQAYDRQKNPSKEEREALVEECNRAECLQRGVSPSKAHGLGSNLVTEVRVYNWF
HNF1B A ANRRKEEAFRQKLAMDAYSSNQTHSLNPLLSHGSPHHQPSSSPPNKLSGVRYSQQGNNEI HNF1B B ANRRKEEAFRQKLAMDAYSSNQTHSLNPLLSHGSPHHQPSSSPPNKLSGVRYSQQGNNEI HNF1B C ANRRKEEAFRQKLAMDAYSSNQTHSLNPLLSHGSPHHQPSSSPPNKLSG-----KQRLGL
HNF1B A TSSSTISHHGNSAMVTSQSVLQQVSP-----ASLDPGHNLLSPDGKMISVSGGGLPPVST HNF1B B TSSSTISHHGNSAMVTSQSVLQQVSP-----ASLDPGHNLLSPDGKMISVSGGGLPPVST HNF1B C TASATQPSWFLPRILSGLRVFRGANAFEMILGPLSHCQNILPWK------
HNF1B A LTNIHSLSHHNPQQSQNLIMTPLSGVMAIAQSLNTSQAQSVPVINSVAGSLAALQPVQFS HNF1B B LTNIHSLSHHNPQQSQNLIMTPLSGVMAIAQSLNTSQAQSVPVINSVAGSLAALQPVQFS HNF1B C ------
HNF1B A QQLHSPHQQPLMQQSPGSHMAQQPFMAAVTQLQNSHMYAHKQEPPQYSHTSRFPSAMVVT HNF1B B QQLHSPHQQPLMQQSPGSHMAQQPFMAAVTQLQNSHMYAHKQEPPQYSHTSRFPSAMVVT HNF1B C ------
HNF1B A DTSSISTLTNMSSSKQCPLQAW HNF1B B DTSSISTLTNMSSSKQCPLQAW HNF1B C ------
Figure 3-5 Multiple sequence alignment of HNF1B isoforms:
A, B and C isoforms highlighting the difference in the sequences. --- represents the isoform difference in A and B (183-208 missing); KQRLGL TASATQPSWFLPRILSGLRVFRGANAFEMILGPLSHCQNILPWK; represents the change from the canonical sequence in isoform C; --- represents the isoform difference in A and C (401-557 missing).
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 51
Figure 3-6-Different transcripts of HNF1B as per TCGA Spliceseq:
This database was used to gain an overview of alternative mRNA splicing in the HNF1B gene. There are seven events that take place during splicing. One of the exon skipping events was observed in exon 7 and 8, coding for the transactivation domain in the splice variant, confirming the origin of the D and E variants along with another exon-skipping occurring in exon 8 confirming the E variant. An alternate acceptor event, took place in exon 3, coding for the DNA-binding domain region, and confirming the B and C’ variant.
52 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Figure 3-7-RNA-Seq: The data for the four different prostate cell lines are aligned with the HNF1B transcript variants.
The expression of HNF1B in LNCaP cells is higher as compared to PC3 and BPH1, we see the highest peaks in the first two exons indicating higher expression for the transcript variants having the dimersation domain, whereas there was no expression in the normal human prostate cell.
3.3.2 Quantitative expression of HNF1B transcript variants We designed primers for the respective transcript variants. For RT-PCR, HNF1B transcript variants A and B are expected to have a band size of 693 and 615 bp respectively. Transcript variant C and C’ are shown to have the expected band size of 601 bp and 523 bp, whereas variant D and E are expected to have a band size of 309 bp and 428 bp respectively. There are several bands that are observed using D and E primer set, we hypothesize these might be other annotated transcript variants. These results are depicted in Figure 3-8 (refer to Figure 7-1 for the full image of gel for HNF1B transcript variants and Table 3-1). Prostate cell lines used in the study are androgen-dependent; VCaP, DUCaP and LNCaP (androgen-independent cell lines) DU145 PC3 and C42B (androgen-independent cell lines) along with normal prostate cell lines RWPE1 and RWPE2. Variant A is expressed in DUCaP, VCaP, LNCaP, and DU145, though its expression is low in PC3. The variant B was also expressed in RWPE2, DUCaP, VCaP, LNCaP, DU145, and PC3. There was no expression of the A and B variant in C42B. Variant C and C’ was expressed in DUCaP, VCaP, LNCaP,
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 53
DU145, and PC3. There was no expression for C and C’ in C42B. Variant D was expressed in RWPE1, DUCaP, LNCaP, and DU145. There was a lower expression for D in VCaP, C42B, and PC3. The expression for variant E was seen in DUCaP, VCaP, LNCaP, DU145, PC3, and very low expression was seen in C42B. The primer set for D and E also picked up the A and B variant having a band size of 623 bp in DUCaP, VCaP, LNCaP, C42B, DU145, and PC3. Though the variant A and B were not picked up by their specific primer sets in C42B. We hypothesize that there is another novel variant I which is identified in C42B, lacking the first two exons coding for dimersation domain but having the remaining exons, coding for the DNA binding domain along with transactivation domain within it.
Relative expression of the HNF1B was then assessed using q RT specific primers which showed the highest expression of HNF1B transcript variants in VCaP and DU145 (Figure 3-9).
A
B
C I 623
Figure 3-8-Characterisation of HNF1B transcript variants in a panel of cell lines
HNF1B transcript variants were confirmed in a panel of cell lines representing PCa (RWPE 2, DUCaP, VCaP, LNCaP, C42B, DU145, and PC3) and, benign prostate (RWPE1). RT-PCR product sizes that have been marked with a black arrow were
54 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
confirmed by running them on 1.2 % for HNF1B A, B (A), C, C’ (B) and 2% for HNF1B D and E (C) transcript variants agarose gel electrophoresis. There are other bands that are observed in panel (C) subsection, which were identified using the DE primer set, and which we suspect might be other variants such as I. (Refer to Figure 2-1 for the scheme of primers (N=3)).
Table 3-1. Overview of the expression of HNF1B transcript variants in prostate cell lines from the RT-PCR and RT-qPCR data.
TRANSCRIPT FORWARD REVERSE PRIMER EXPRESSION IN
VARIANT PRIMER CELL LINE
A EXON 2 EXON 5 DUCaP, VCaP, LNCaP,
DU145, PC3
B EXON 2 EXON 5 DUCaP, VCaP, LNCaP,
DU145, PC3
C EXON 2 EXON 4 DUCaP, VCaP, LNCaP,
DU145, PC3
C’ EXON 2 EXON 4 DUCaP, VCaP, LNCaP,
DU145, PC3
D EXON 6 EXON 9 RWPE1, DUCaP, VCaP,
LNCaP, C42B, DU145,
PC3
E EXON 6 EXON 9 DUCaP, VCaP, LNCaP,
C42B, DU145, PC3
I EXON 6 EXON 9 Sequence confirmed, all
cell lines
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 55
Figure 3-9-The relative expression of HNF1B variants using qRTPCR:
The relative expression correlated with RT-PCR results, showcasing endogenous expression of HNF1B transcript variants in several PCa cell lines. (N=3)
As the size of A and B variants had a very minor difference, we used varying percentages of gels to define these variants in RT-PCR. Additionally, the q RT-PCR data was checked for the melt curves for each variant, which showed individual variants and their expression levels. Both these data combined showed proper quantitative expression of all the variants.
3.3.3 Determining the protein expression level of HNF1B transcript variants in PCa cell lines
After characterizing the transcript variants in the panel of PCa cell lines, we went on to further determine the protein expression level. The antibodies used here detected both the N and C-terminal regions. VCaP and DUCaP had a higher expression for transcript variants A with an expected band size of 61 kDa and B with an expected size of 57 kDa in comparison to LNCaP and DU145 (Figure 3-10). Housekeeping protein, that was used as a control was Beta-Actin. The antibody used in the Western blotting technique could only detect HNF1B A and B transcript variants; though variant C cannot be identified since the antibody epitope may be in the transactivation domain, which is completely missing in the C variant.
56 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Figure 3-10-Western blotting carried out to identify HNF1B transcript variants at the protein level:
Lysates (20 μg) from PCa cell lines. HNF1B A, B transcript variants were detected having expected band size of 61 and 57 kDa respectively with VCaP and DUCaP showing a higher expression as compared to LNCaP and DU145. Actin was used as a housekeeping control protein having an expected band size of 45 kDa.
3.3.4 Discussion
As discussed in the literature review, there has been mention of three HNF1B transcript variants which are expressed in PCa cell lines, and describing the role of transcript variant A and B as transcriptional activators (Bach & Yaniv, 1993) in malignant tissue and transcript variant C as a transcriptional repressor predominant in benign tissue (Harries, et al., 2010). Therefore we wanted to characterize the expression of the other annotated transcript variants of HNF1B in a panel of PCa cell lines, along with determining their protein expression; the antibody used in western blotting technique could only detect HNF1B A and B transcript variants; though variant C could not be detected as it seems the antigenicity for the antibody exists in between the transactivation domain, which is completely missing in the C variant. From the in- silico results obtained, it was seen that there is an expression for HNF1B in PCa, followed by its downregulation in CRPC highlighting its potential role as a tumor suppressor. To date, the majority of the PCa cases in CRPC have been reported to be
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 57
associated with dysregulation of androgens which act via androgen receptors (Debes & Tindall, 2002). Regarding the alternative splicing events occurring in HNF1B which have been gathered from TCGA-sliceseq, these are exon skipping events in exon 7 and 8 (HNF1B D and E) coding for the transactivation domain, along with another different exon skipping in exon 8 (HNF1B E). An alternative acceptor was seen in one of the variants in exon 3 coding for the DNA-binding domain (HNF1B A and B). There are other splice events, alternate donor (AD), retained intron (RI), Mutually exclusive exons (ME), Alternate promoter (AR), Alternate terminator (AT) which could possibly explain the reason for the other annotated splice variants C, C’, E, and I. Alternative splicing has been often discussed for its crucial role in cancer (Singh & Eyras, 2017), which could possibly explain the reason for a dual role of HNF1B, i.e., as a tumor suppressor or oncogene. We next analysed the RNA-seq data for prostate cell lines, to determine the expression of each transcript variant in these cells. There was less expression observed in LNCaP and PC3 cells as compared to other cell lines. Additionally, western blot also showed less expression in LNCaP and PC3 cells, wherein PC3 cells did not show an expected protein band. The previous studies carried out for knockdown of HNF1B for the classical transcript variant A were carried out in LNCaP cells (Grisanzio, et al., 2012). Until now there have been studies that were carried out with reference to the HNF1B (A) transcript variant (Ross-Adams, et al., 2016; Shen, et al., 2013), though there has been no report so far with respect to other HNF1B transcript variants. There isn’t any published study regarding identification and functional characterisation of HNF1B transcript variants in prostate cancer. This is the primary study which deals with understanding the role of these variants in prostate cancer, moreover identification of these variants at protein level needs to be quantified using analytical techniques such as Mass spectrometry.
58 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Chapter 4: Role of androgens and docetaxel in the expression of HNF1B transcript variants
4.1 Introduction
The male sex hormone, i.e. androgens regulate the embryological progression of the prostate, advancement, and function of male reproductive organs along with the development of male secondary features. On the contrary, lack of androgens, through castration, for instance, leads to diffused atrophy primarily of the luminal epithelial cells which often results in the involution of the prostate gland. Testosterone, is the major androgen involved in circulation and Dihydrotestosterone (DHT) is the prevalent androgen in prostate tissue; these are the two most crucial androgens in men. The action for the mentioned androgens occurs via the androgen receptor (AR). Recent studies have also shown that there is a graded system of transcription factors which administers prostate cancer progression which is via androgen receptor (Wang et al., 2007) AR plays a major role in prostate cancer pathogenesis, Huggins and Hodges were the first ones to validate the receptiveness of prostate cancer towards androgen deprivation, this clearly highlighted that prostate cancer is dependent on androgen receptor activation for progression and survival, as a first line of treatment for prostate cancer patients, different drugs are administered such as some of them are; Enzalutamide which is a strong oral androgen receptor has shown to be effective in terms of survival in metastatic castration-resistant patients formerly and afterward chemotherapy. Bicalutamide is an antiandrogen that is non-steroidal and has been used extensively to treat patients with nonmetastatic or metastatic CRPC. Docetaxel has also been used over time for patients who might develop resistance with the above- mentioned drugs or others. There have been several studies carried out so far, to understand the mechanism behind the resistance towards these drug’s activity particularly, Lundon et al (Lundon et al., 2017) have cited that there are several complex changes which take place during the pro and anti-apoptotic proteins with respect to advancing of resistance to docetaxel along with changes in central signaling pathways which are controlled by transcription factors. Docetaxel; is a semi-synthetic taxoid that results in cell cycle arrest (Puente et al., 2017). Keeping this in view, the
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 59
aim of this chapter is to define if any of the HNF1B transcript variants are androgen- regulated; the treatment groups were as follows; Charcoal Strip Serum (CSS) with and without DHT (10 nM), anti-androgens Enzalutamide, Bicalutamide (10 μM) and Docetaxel, having varying conc. starting from 01.,1.0,5,10 and 20nM. In addition to the above mentioned, we also referred to our RNA-seq data results which demonstrated that LNCaP, androgen-dependent, cell line had a higher expression for HNF1B in comparison to other cell lines. On treating the cells with DHT, HNF1B was observed to be downregulated in the first four exons which correspond to variant C. Next, we wanted to observe the expression of HNF1B transcript variants after docetaxel treatment as this is the current therapy of choice for advanced, CRPC (the fatal final stage of the disease) after the failure of hormonal therapy.
4.2 Methods
The common materials and methods have been outlined in Chapter 2.
4.2.1 . Androgen deprivation assay in LNCaP cell lines An androgen deprivation assay was performed in one of the androgen-responsive cell lines, LNCaP to determine the regulation of HNF1B transcript variants expression by androgens and anti-androgens. LNCaPs (30,000cells/well) were seeded in RPMI1640 media (Life Technologies, catalog number - 11835-030) supplemented with 5% fetal calf serum (FBS) in a 6-well plate and incubated in the cell incubator for 3 days. The medium was then replaced with an androgen-depleted culture medium (RPMI1640) containing 5% charcoal-stripped serum (CSS). After 48 hours, the cells in Charcoal Striped Serum (CSS) were supplemented with 10 nM DHT with or without 10 μM of anti-androgens – bicalutamide (B) and enzalutamide (E) as well as ethanol (EtOH, vehicle control), and Dimethyl Sulfoxide (DMSO) and incubated at 37˚ C for 48 hours. RNA extraction, cDNA synthesis, and RT-qPCR were carried out as described in chapter 2.
4.2.2 Docetaxel treatment in LNCaP cell lines
Docetaxel treated samples were kindly provided by Dr. Patrick Thomas (Ghrelin Group, APCRC-Q) of varying concentrations starting from 0.1, 1.0, 5, 10, 20nM. Docetaxel was prepared in 2%FBS/RPMI. Cells were treated once they reached 60-70% confluency. The medium is replaced with docetaxel/2%FBS-RPMI1640. RNA was extracted post 48 hours
60 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
of treatment. RNA extraction, cDNA synthesis, and RT-qPCR were carried out as described in Chapter 2.
4.3 Results
Androgen/ anti-androgen treatments: RT-qPCR, to measure the relative expression levels of HNF1B transcript variants, was carried out with the LNCaP cDNA according to the treatment groups depicted in (Figure 4-1 and Figure 4-2). These comprise of CONTROL (CSS); Dihydrotestosterone (DHT); Dihydrotestosterone with Enzalutamide (DHT+ENZ); Enzalutamide (ENZ); Dihydrotestosterone with Bicalutamide (DHT+BICULT), Bicalutamide (BICULT.). The relative gene expression was determined by the comparative CT (ΔΔCT) method. RPL32 expression was used as the endogenous control. Initially, to check the validity of the treatment, a PSA primer set was used for one of the treated samples, Prostate-specific antigen (PSA) is a serine protease which is synthesised mutually by normal and malignant epithelial cells of the prostate. PSA is primarily induced by androgens and is further controlled by AR; (androgens such as testosterone and dihydrotestosterone are facilitated via the androgen receptor (AR)) and is the most vital biomarker of the diseases as it initiates and later progresses (Kim & Coetzee, 2004; Saxena et al., 2012). HNF1B transcript variants A, B, C, C’, D and E were determined in the same treated samples (Figure 4-1 and Figure 4-2). There was no change in HNF1B A and B transcript variants, under DHT conditions, though they were downregulated in the presence of treatment groups; DHT+ENZ, Enz, DHT+BICULT (A only). On the other hand, transcript variant C was upregulated under DHT, along with BICULT conditions, whereas it was downregulated in DHT+ENZ along with DHT+BICULT conditions. Similarly for C’, which was upregulated in DHT, DHT+ENZ, ENZ and BICULT. Though it was downregulated under DHT+BICULT. For transcript variant D, it was down-regulated under DHT conditions along with DHT+BICULT conditions. It was seen to be upregulated under DHT+ENZ, ENZ, and BICULT conditions although not statistically different. In the case of transcript variant E, expression was essentially unchanged in all conditions.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 61
Figure 4-1-Positive control used for confirming the DHT/ androgen treatment:
PSA is mainly induced by androgens (dihydrotestosterone/DHT) and repressed by the anti-androgens ENZ and BICULT. (Mean +/- SEM, n=3, one-way ANOVA * P <_0.01 ** P<0.01; ***P <0.001; **** P<0.0001).
62 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Figure 4-2-HNF1B transcript variant expression in LNCaP cells in the presence of androgens and anti-androgens.
LNCaP cells were treated in the presence of androgens (DHT) and anti-androgens (DHT+ENZ, ENZ, DHT+BICULT, BICULT. q RTPCR performed) (Mean +/- SEM, n=3, one-way ANOVA * P <_0.01 ** P<0.01; ***P <0.001; **** P<0.0001).
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 63
Docetaxel treatment: RT-qPCR was carried out with the LNCaP cDNA according to treatment groups comprising of (CONTROL; Ethanol) along with varying concentrations of Docetaxel treated at 0.1, 1.0, 0.5, 10 and 20nM respectively to measure the relative expression levels of HNF1B transcript variants. Initially, to check the feasibility of the treatment the PSA primer set was used to determine PSA expression for the treated samples, HNF1B transcript variants A, B, C, C’, D and E were then determined in the treated samples (Figure 4-3 and Figure 4-4)
1 .5
e C o n tro l (E tO H )
g n
a D X T (0 .1 n M ) h
c 1 .0
D X T (1 .0 n M )
d l
o D X T (5 n M )
f
e D X T (1 0 n M ) v
i 0 .5 t * a **
l D X T (2 0 n M )
e R
0 .0 ) ) ) ) ) ) H M M M M M O n n n n n t 1 0 0 0 E . . (5 ( 0 1 (1 (2 l ( ( T X T T o T T r X D X X t X D D n D D o C P S A
Figure 4-3-Positive Control used for confirming docetaxel treatment.
Low to medium expression of PSA with varying concentrations of DXT conditions. This shows that the docetaxel treatment; with different concentrations worked. (Mean +/- SEM, n=3, one-way ANOVA * P <_0.01 ** P<0.01; ***P <0.001; **** P<0.0001).
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 65
Figure 4-4-HNF1B transcript variants in the presence of docetaxel:
HNF1B A and B transcript variants showed almost the same pattern with the various concentrations of docetaxel used, with the highest expression seen from 5-20nM concentration. Transcript variants C and C’ showed the same pattern, with the highest effect at 20nM. Though there was not much difference that was observed for the D transcript variant across all doses. There was the highest expression for transcript variant E with docetaxel under 20nM conc. n=3.
66 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
4.3.1 Discussion According to the study carried out by Hu et al, there was a higher expression for HNF1B in the LNCaP cell line, along with this, they also observed higher expression levels in wild type PCa cells NR4A1, HSPD1, ERBB4 and ESR1 in HNF1B transfected cells. The group hypothesized that most likely the mechanism of action for HNF1B specifically in PCa possibly includes regulation by androgenic hormones (Hu, et al., 2013). In another study by Pihlajamaa et al, it was highlighted that AR-binding mostly takes place in murine prostate, kidney, and epididymis as verified by ChIP-seq analysis. This highlighted that in vivo AR cistromes along with their respective androgen-dependent transcription programs are highly tissued specific facilitating distinct biological pathways (Pihlajamaa et al., 2014). There has been no clarification with respect to how androgens and HNF1B variants take part in mechanisms underlying PCa. This is the first study that showcases the effect of androgens on HNF1B transcript variants. Upon further analysis of the RNA-seq data for LNCaP treated with DHT and the anti-androgen bicalutamide, we saw exon peaks from the second to the eighth exon for both the experimental groups as mentioned above, this highlights that HNF1B transcript variants might be elevated during CRPC. The same was correlated with the quantitative analysis here, the experimental groups, had a similar trend that was observed in HNF1B A and B with androgens and anti- androgens; followed by C and C’, which highlights the same trend, based on the exons which are common in the transcript variants. There was not much difference (in terms of fold change) for the transcript variants D and E within each experimental group. These results reflect that androgens are not involved directly along with HNF1B in implantation of PCa and hence neither did anti-androgens show any effect on the same. Though to introspect further, about how HNF1B could be participating in PCa; we might have to perform Androgen Receptor chromatin immunoprecipitation (AR- ChIP), through this approach we would be analysing the binding of AR on promoter/enhancer region of HNF1B and state if it’s the sole key transcription factor or other co-transcription factors (eg. FOXA1) which are responsible for its activity. Docetaxel has been used over time for patients who might develop resistance with androgen-dependent drugs as stated above. There have been several studies carried out so far, to understand the mechanism behind the resistance towards these drug’s activity particularly, Watson et al have cited that there are several complex changes which take place during the pro and anti-apoptotic proteins with respect to advancing of resistance
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 67
to docetaxel along with changes in central signalling pathways which are controlled by transcription factors. Thus we aim to determine the role of docetaxel in the expression of HNF1B. We next analysed the expression of these transcript variants with docetaxel treatment, which is a chemotherapeutic drug used at the CRPC stage. We observed a higher expression of HNF1B transcript variants with a 20nM concentration of docetaxel for all HNF1B transcript variants except for D, which suggests, the above concentration does not suppress the expression of HNF1B transcript variants. HNF1B is not suppressed by docetaxel and so could not be used as a biomarker of efficacy and whatever cellular effects HNF1B regulates, it is not affected by docetaxel either.
68 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Chapter 5: Functional effects of The HNF1B and its transcript variants on Prostate cell lines
5.1 Introduction
The functional effect of HNF1B transcript variants still needs to be defined in PCa. We used specific siRNA for knockdown of HNF1B transcript variants and determined its functional effect. Next, we generated stable cell lines overexpressing HNF1B transcript variants A, B and C and wanted to determine the functional effect of overexpressed variants on PCa cell lines.
5.2 Methods
The common materials and methods have been outlined in Chapter 2.
5.2.1. Generation of loss of function (knockdown; using siRNA) and to perform functional assays for knockdown
Once HNF1B transcript variants had been characterised, and it was observed that the highest expression for transcript variants was seen in VCaP and DU145, hence these cell lines were an ideal choice for studying knockdown effect. Exon specific siRNA were designed: siRNA pertaining to exon 2 and 3 (siRNA A/1) which is common for A, C and D and will knockdown all three; siRNA pertaining to 4 exon and 3’UTR is specific for C and C’ (siRNA C) and the siRNA pertaining to exon 5 (siRNA 2) was common to all transcript variants (Figure 5-1)
70 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer siRNA 2
siRNA A/1
siRNA C
Figure 5-1-siRNA Scheme for HNF1B transcript variants:
The position of the siRNAs has been marked with the blue (siRNA A/1, C and 3). siRNA 1 will be targeting exon 2 and 3 of transcript variant A which will also target variant C, as the exons 2 and 3 are the same. siRNAC which is the customised one specifically for the variant C and C’ considering the fourth exon along with 3’UTR. siRNA2 will be targeting exon 5 of the longer transcript variants A, B, D, and E.
5.2.2 Generation of loss of function/knockdown at the mRNA level Expression of HNF1B was highest in VCaP and DU145 and hence knockdown was carried out in these cell lines. Cells (1.2 x 10-6) were seeded in a 6-well plate were in RPMI1640 media (Life Technologies, catalog number - 11835-030) supplemented with 10% and 5% fetal calf serum (FBS). After 48 hours the cells were transfected with siRNA A/1(Life technologies, Catalogue no.AM16708A, siRNA ID: 235837) targeting exon 2 and 3; siRNA3 (Life Technologies, Catalogue no.AM16708A, siRNA ID: 235838; refer to Appendix A for siRNA sequences used in this study) targeting exon 5, (for which is common in transcript variants A, B, D and E. siRNA C was customised for C and C’ as the exons were closely shared with transcript variant A and B. The siRNA were diluted in lipofectamine 3000 reagent along with Opti-MEM Medium (1:1 ratio), which is then incubated for 10-15 mins and the complex is added to the cells. The shorter one was made considering the fourth exon and 3’ UTR region. (Figure 5-1) The cells after transfection were kept in incubation at 37◦ C for 2 days. RNA extraction, cDNA synthesis, and RT-qPCR were carried out as described in Chapter 2.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 71
5.2.3. Cellular proliferation assay for knockdown models
We assessed cell proliferation with transient HNF1B knockdown in the androgen- independent cell line, DU145 using the IncuCyte Zoom system for 5 consecutive days. Initially the seeding density of the cells was optimised. Therefore we have started with 1000 cells/well to 3000 cells/well. The siRNA A/1 chosen were targeting exon 2 and 3; which will knockdown HNF1B A and C, siRNA 2 targeting exon 5 common in HNF1B variants A, B, D and E, a customised siRNA C was used specifically targeting exon 4 and 5’ UTR.
5.2.4. To generate overexpression templates for HNF1B transcript variants A, B and C
Templates for HNF1B variants A and B variant were generated by using LNCaP treated with CSS and DHT, the gel fragments; having inserts of A and B variant; confirmed using RT-PCR primers specific for the variants were collected and gel purified using Wizard® Gel and PCR Clean-Up System (Promega) and ligated together using T4 DNA Ligase as per manufacturer’s instructions into p GEMT easy vector. For HNF1B variant C the template was generated using the normal VCaP cell line, followed by the same procedure as mentioned above. The ligation reaction was kept overnight at 4˚C. The next day these were transformed into competent Stbl3 E. coli cells on LB agar plates having ampicillin with X-gal/IPTG and the positive colonies were selected and colony crack screening was carried out. The primer set used during the colony crack screening were RT specific primers for HNF1B A (693 bp), B (615 bp) and C (601 bp) Refer to Table 2-1. Based on the sizes for A, B, and C, we next did colony purification using the Bioline purification kit (plasmid). Samples were sent for sequencing at the Australian Genome Research Facility Ltd (AGRF, University of Queensland) (Results attached in Appendix B). For transfection, PC3 cells were plated at a density of 0.5X105 cells/well (24-well cell culture plate) in 500 µl of antibiotic-free media and cultured. Transfection reagent and HNF1B transcript variants A, B, and C plasmid DNA (pcDNA3.1 (Invitrogen)) clones were mixed at a 3:1 ratio (0.5 µg of plasmid per well) and incubated at room temperature for 10 mins. Approximately 25 µl of this complex was added to each well and the cells were incubated overnight for 24-48 hours. For stable transfection, the cells were incubated in media with selection antibiotic (G-418 (Invitrogen)) for 2-3
72 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
weeks and the expression of and the expression of target gene or protein were analysed by RT-PCR and Western blot analysis respectively.
5.2.5. Mass Spectrometry of overexpression models for HNF1B A
The cell lysates for overexpression models generated in PC3 overexpressing HNFB1 A were extracted for mass spec analysis. The samples were boiled in Sodium deoxycholate buffer (1% sodium deoxycholate, 10 mM TCEP, 40 mM 2CAA, 100 mM Tris pH8) and diluted in H2O. The samples were sonicated for 15 mins to denature proteins and shear DNA. The soluble fraction was collected after a centrifugation step at 1000g for 15 mins. The sample concentration was measured by Direct Detect and samples diluted to 1 µg/µl. The samples were then digested with trypsin (0.1 μg) and incubated at 370C overnight and acidified with a final formic acid concentration of 1%. The sodium deoxycholate precipitate was removed by centrifugation at 13000g for 10 mins. The samples were then cleaned with C18 tips and resuspended in 0.05% formic acid to make 0.5 μg/μl final concentration. 1μg of protein was injected into the mass spec for analysis at MS facility at TRI. Data analysis was performed on SpectrumMil.
5.2.6. Cellular proliferation assay for overexpression models
We assessed cell proliferation for overexpressed models of HNF1B A, B and C in PC3 cells using the IncuCyte Zoom system for 5 consecutive days. The overexpression model which had the highest expression at the mRNA level was chosen for cellular proliferation.
5.3. Results
We assessed the knockdown at the mRNA level in the androgen-sensitive cell line, VCaP (Figure 5-2) for exon 5. The siRNA 2 (targeting exon 5) targeting exon 5 for the longer transcript variants A, B, D, and E was the one which showed a higher percentage of knockdown with significant p values. Knockdown for A, B, C, and D was also assessed in DU145 (Figure 5-3). There was significant knockdown with siRNA2 for A and B transcript variants, though there was no significant knockdown with either of the siRNA for variant C and D. Our preliminary data for siRNA for the shorter transcript variants namely C and C’ showed reduction with siRNA mediated knockdown.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 73
Figure 5-2-Si RNA2 mediated Knockdown at the mRNA level in VCaP cells:
Cells were seeded in RPMI1640 media supplemented with 10% FBS. The cells were then transfected with 10 μM of si RNA2 (targeting exon 5) the next day. The cells were then incubated for 48 hours at 37˚C. The following day RNA was collected and c DNA was prepared and the knockdown efficiency was monitored by RT-qPCR. (a) HNF1B A, (b) B and (c) D transcript variants showed a knockdown with significant P values whereas (d) HNF1B E transcript variant had a high standard error. The siRNA for C and C’ are still to be optimised. (Mean +/- SEM, n=3, unpaired t-test * P <_0.01, ** P<0.01; ***P <0.001; **** P<0.0001).
74 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Figure 5-3-siRNA Mediated Knockdown of different HNF1B transcripts in DU145 cells:
Cells were seeded in RPMI1640 media supplemented with 5% FBS. The cells were then transfected with 10 μM siRNA the next day and were then incubated for 48 hours at 37˚C. The following day RNA was collected and c DNA was prepared and the knockdown efficiency was monitored by RT-q PCR for (a) HNF1B A, (b) B, (c) C and (d) D transcript variants. SiRNA A/1(si1) and si RNA2 (si2; siRNA Scheme Figure 5-1) was used for knocking down A, siRNA2 gave 90% knockdown efficiency for A and B. There was significant knockdown on transcript variant C with siRNA C. We saw 70% reduction for transcript variant D with SiRNA2. (Mean +/- SEM, n=3, unpaired t-test* P <_0.01 ** P<0.01; ***P <0.001; **** P<0.0001).
Once after confirming 3000 cells/well as a cell seeding density number ideal for cell proliferation assay. We carried out cell proliferation in an androgen-independent cell line, DU145, and not in VCaP cell lines, as this cell line is difficult to manage for cell- based assay. We observed a significant increase with siRNA2 (exon 5; targeting HNF1B transcript variants A, B, D, and E) whereas an opposite trend; i.e. a significant increase was observed with siRNA C (Figure 5-4 and Figure 5-5).
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 75
1 5 0 R N A iM a x
N e g a tiv e C o n tro l
y 1 0 0 s iR N A 2 (1 )
c ** **** n
e s iR N A 2 (2 )
u l
f s iR N A 2 (3 ) n
o 5 0 C
0 0 2 0 4 0 6 0 T im e (h o u r s )
Figure 5-4-Cell proliferation assay following siRNA Knockdown in DU145 Cells:
DU145 cells (3000 cells/well) were seeded in a 96 well plate. These were then transfected with 5 pmol of siRNA2 targeted to the fifth exon. RNAiMAX and a non- targeting were used as controls. The cells were observed for 5 days in the Incucyte machine, over a period of 48 hours (as we saw the effect earlier, the results have been analysed within this time frame) in which time cells go through lag to stationary phase. There was a significant increase in cell proliferation with siRNA mediated knockdown for transcript variant A, B, D and E. (n=3) (two-way ANOVA, Dunnett’s multiple comparison test, n=3 ** P<0.01; ***P <0.001; **** P<0.0001).
1 5 0 S iR N A C (1 ) s iR N A C (2 ) s iR N A C (3 )
y 1 0 0 c
n R N A iM a x
e *** *** ** u
l N e g a tiv e C o n tro l
f n
o 5 0 C
0 0 1 2 2 4 3 6 4 8 6 0 7 2 8 4 9 6 T im e (h o u r s )
Figure 5-5-Cellular proliferation following siRNA knockdown in DU145 Cells:
76 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
There was significant reduction observed in cell proliferation with siRNA mediated knockdown for transcript variant C (siRNA C was targeting exon 4 and 3’UTR) (n=3) (two-way ANOVA, Dunnett’s multiple comparison test, n=3 ** P<0.01; ***P <0.001; **** P<0.0001). Overexpression models of HNF1B transcript variants were generated to identify the biological pathways which are regulated by HNF1B. The templates were generated for HNF1B transcript variants A, B, and C by cloning them into pGEMT Easy Vector, to differentiate amongst the variants, which was also confirmed by sequencing. After confirming the variants, these were then ligated into pcDNA3.1, which was also confirmed for the presence of HNF1B A and B clones in the vector (refer to Figure 5-6 and Figure 5-7)
Figure 5-6-PCR analysis to identify positive clones:
Confirmation of positive clones for A and B variants using RT-specific primer set in DHT A1 7 (A variant), DHT A1 8 (B variant). The red box is highlighting the A variant and the purple box highlights the B variant.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 77
Figure 5-7-Restriction enzyme analysis of positive HNF1B transcript variant clones showing correct sizes.
A B and C cloned in pcDNA3.1. (Highlighted in red for transcripts A, B and C marked images respectively; Expected Size: 1674bp, B: 1596bp and C: 1100bp). The fragments were excised and restriction digestion (using enzymes BamH1 and EcoRI) was performed to confirm their cloning into pcDNA3.1.
78 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
To determine the efficiency of transfection of the positive clones, the polyclonal population of cells were selected using Geneticin-418 media. For PC3 cells, transfection efficiency of ~50-60% was observed in our previous transfection studies in our lab using the protocol used in this study for GFP containing plasmids. However, we maintain the selected cells for 4 weeks until all native non-transfected cells were killed and we continue to use a maintenance concentration of the antibiotic (10% of the selection concentration) to inhibit the growth of any non-transfected cells if any. It is thus expected that the antibiotic was able to kill all untransfected cells. The efficiency of transfection was also determined by quantitative mRNA and protein expression analysis. (Refer to Appendix B; Figure 7-2 and Figure 7-3). There was nearly a 20 fold difference in comparison to the vector control for overexpression A model (Figure 5-8); a 27 fold change for overexpression B model (Figure 5-9) and an 11 fold difference for the overexpression C model (Figure 5-10). Protein from the HNF1B A overexpression model in PC3 cells were run on the mass spectrometer and pathway analysis of the data was done using Ingenuity Pathway Analysis (IPA). In PC3 cells 1483 proteins were identified. The pathways identified were usually the ones involved in metabolic pathways and protein processing in the endoplasmic reticulum. This experiment has been done once and will be repeated twice with the overexpression models to remove any bias for identifying pathways associated with overexpression of the HNF1B A transcript variant. The same will be done for transcript HNF1B transcript variants B and C.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 79
H N F 1 B A
2 5 ***
P c D N A 3 .1 e
g 2 0 H N F 1 B O E A
n
a
h c
1 5
d
l
o
f
e 1 0
v
i
t
a l
e 5 R
0
.1 A 3 E A O N D B c 1 P F N H
Figure 5-8-Overexpression of HNF1B A in PC3 cells.
There was nearly a 20 fold upregulation in the overexpression model for HNF1B A in comparison to the vector control; pcDNA3.1. Mean +/- SEM, n=3, unpaired t-test ***P <0.001).
H N F 1 B B
****
3 0
e p c D N A 3 .1 g
n H N F 1 B O E B
a h
c 2 0
d
l
o
f
e v
i 1 0
t
a
l
e R
0
.1 B 3 E A O N D B c 1 p F N H
Figure 5-9-Overexpression of HNF1B B in PC3 cells.
There was nearly a 27 fold change in the overexpression model for HNF1B B in comparison to the vector control; pcDNA3.1. Mean +/- SEM, n=3, unpaired t-test **** P<0.0001).
80 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
H N F 1 B C
1 5 **
e p c D N A 3 .1 g
n H N F 1 B O E C
a h
c 1 0
d
l
o
f
e v
i 5
t
a
l
e R
0
.1 C 3 A E N O D B c 1 p F N H
Figure 5-10-Overexpression of HNF1B C in PC3 cells.
There was nearly a 9 fold change in the HNF1B C overexpression model in comparison to the vector control; pcDNA3.1.Mean +/- SEM, n=3, unpaired t-test ** P<0.01
Next, we wanted to observe any functional role of the overexpression model for the A and B variants in PC3 cells; and observed a reduction in cell proliferation in comparison to empty vector pcDNA3.1. (Figure 5-11 and Figure 5-12). Though there was an opposite trend, i.e. there was an increase in proliferation, with overexpression that was observed with the overexpression model for HNF1B C (Figure 5-13). Interestingly Adams et al also observed the same trend with their overexpression models for HNF1B A and highlighted the transitioning of epithelial to a mesenchymal state of the cells (Ross-Adams, et al., 2016).
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 81
H N F 1 B O E A
8 0 P C D N A 3 .1 ** * * H N F 1 B O E A 1 (1 ) 6 0
y H N F 1 B O E A 1 (2 ) c
n H N F 1 B O E A 1 (3 ) e
u 4 0
l
f
n
o C 2 0
0 0 1 2 2 4 3 6 4 8 6 0 7 2 8 4 T im e (h o u rs )
Figure 5-11-Stable overexpression of HNF1B A in PC3 cells resulted in a decrease in proliferation.
Overexpression (OE) model of HNF1B A in PC3 cells. There was a reduction in proliferation for the A variant overexpression model in comparison to the vector control; pcDNA3.1. (two way ANOVA, Dunnett’s multiple comparison test, n=3 * P <_0.01 ** P<0.01).
H N F 1 B O E B
8 0 P c D N A 3 .1
H N F 1 B O E B 1 (1 ) 6 0 y *** H N F 1 B O E B 1 (2 )
c **** *** n
e H N F 1 B O E B 1 (3 ) u
l 4 0
f
n
o C 2 0
0 0 1 2 2 4 3 6 4 8 6 0 7 2 8 4 T im e (h o u rs )
Figure 5-12-Stable Overexpression model of HNF1B B in PC3 cells resulted in a decrease in proliferation.
There was a reduction in proliferation for the B variant overexpression (OE) model in comparison to the vector control; pcDNA3.1.(two way ANOVA, Dunnett’s multiple comparison test, n=3 ***P <0.001; **** P<0.0001).
82 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
H N F 1 B O E C
1 5 0 P c D N A 3 .1
H N F 1 B O E C 2 H N F 1 B O E C 4
y 1 0 0 c
n H N F 1 B O E C 6
e **** **** ****
u
l
f n
o 5 0 C
0 0 1 2 2 4 . 3 6 . 4 8 . 6 0 . 7 2 . 8 4 . T im e (h o u rs )
Figure 5-13-Cell proliferation assay showing overexpression of HNF1B C’ variant which resulted in increased proliferation.
Overexpression (OE) model of HNF1B C in PC3 cells. There was an increase in proliferation for the C variant overexpression model in comparison to the vector control; pcDNA3.1. (two-way ANOVA, Dunnett’s multiple comparison test, n=3 **** P<0.0001).
5.4. Discussion
After characterizing the HNF1B transcript variants in PCa cell lines, we observed that there was a higher expression for all transcript variants in VCaP, an androgen-sensitive cell line, and DU145, an androgen-independent cell line. There has not been a single report so far, which has determined the effect of knockdown for each HNF1B transcript variant in either of the above-mentioned cell lines. In the study carried out by Grisanzio et al, the siRNA for HNF1B (A) was designed to exon 5, and the knockdown was carried out in PC3 cells, an androgen-independent cell line, though there was not significant knockdown that was observed by suppressing HNF1B (Grisanzio, et al., 2012). There was a significant knockdown that we saw with the targeting of exon 5 i.e. siRNA 2 for transcript variants A, B, D and E in VCaP. This particular exon is common in all the above-mentioned transcript variants and is part of the transactivation domain, which is conserved in all these variants. The same effect was observed for variants A and B in DU145, whereas there was a reduced effect of
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 83
knockdown for variants C and D respectively at the mRNA level. Nevertheless, there was sufficient knockdown in these models to use them for functional assays. The siRNA targeting variant C was customised and designed combining exon 3 and 4. For cell proliferation, we saw a reduction only for siRNA against transcript variant C. These results highlights the effect of domains with respect to transcript variants in androgen-dependent and independent cell lines. Though it will be interesting to determine the effect of respective siRNAs on other annotated transcript variants Adams et al reported a marked reduction in cell proliferation (Ross-Adams, et al., 2016) for an overexpression A variant model. The same trend was seen in our study for overexpression of the A and B variants in PC3 cell lines, whereas the opposite effect was observed for the overexpression C variant model, wherein we could see an increased proliferation in comparison to control. This variation observed in overexpression models highlights the importance of the domains that are specific to each transcript variant, considering that HNF1B A and B have conserved dimersation,
DNA binding POUH and POUs and transactivation domain, because of which the same trend was observed for A and B. On the other hand transcript variant C completely lacks the transactivation domain, thus the increase in proliferation suggests the importance of the transactivation domain, HNFs function as a homo or heterodimers by close interaction of HNF-1A and HNF1B. So it seems that there is HNF1A binding to HNF1B via the transactivation domain and hence the opposite results for the overexpression models of HNF1B transcript variants. Ross-Adams et al reported a change in morphology such as flattened, more mesenchymal-like morphology for overexpressed PC3 cells (Ross-Adams, et al., 2016). These findings suggested loss of HNF1B expression has a fundamental role in epithelial to mesenchymal transition. Considering the Epithelial to Mesenchymal transition (EMT) transition which occurs, it plays a crucial role in the progression of metastasis further leading to drug resistance (Montanari et al., 2017). During the course of development of cancer, there is a synchronised declining and progression of altered epithelial, i.e. E-cadherin, laminin, MUC-1 and mesenchymal markers i,e. slug, snail, and vimentin, along with greater migration and certain other invasive potential, higher resistance in apoptosis along with prolonged production of extracellular matrix. As mentioned above, this may inhibit resistance to standard treatment (i.e. chemotherapy and/or androgen deprivation therapy) for treating advanced PCa. Furthermore, in our overexpression models EMT
84 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
transitioning will be validated and analysed at the transcriptional and translational level using the various EMT markers.
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 85
Chapter 6: General Discussion
HNF1B belongs to the transcription family of HNF, which mostly plays a vital role in the regulation of genes related to the liver, apart from other tissues. It was first identified as a gene with a genetic mutation in MODY. Later, through GWAS, it was found that the 17q12 region, was one of the susceptible regions for PCa. There were several reports from multiple populations that further proved the presence of risk associated loci in the same region.
Altogether the function of HNF1B in PCa still needs to be revealed. In this study, we showed the expression of HNF1B transcript variants in the panel of prostate cell lines, which has not been reported so far. The highest expression of HNF1B transcript variants was observed in the androgen-sensitive cell line VCaP and the androgen- independent cell line, DU145. We next carried out knockdown in the above mentioned cell lines using different siRNA targeting longer transcript variants i.e. A, B, D and E and shorter transcript variants (C and C’) and witnessed significant knockdown for transcript variants A,B, C and D. There was a decrease in cell proliferation which correlates with the study by Grisanzio et al, who also reported the same effect on PC3 an androgen-independent cell line, for transcript variant A (Grisanzio, et al., 2012). Interestingly, in our study, we had one of our siRNA targeting exon 5. As pointed out in our study, there was a higher expression for HNF1B transcript variants in the androgen-dependent cell line, which correlated with the study carried out by Hu et al, showing higher expression for HNF1B in the LNCaP, androgen-dependent cell line (Hu, et al., 2013).
Androgens play an important role in PCa and in CRPC and therefore we were interested to look for its effects on expression of HNF1B transcript variants, which has not been reported so far. The treatment comprised of combination of androgens (DHT) and anti-androgens/androgen deprivation (ENZ, BICULT; CSS) for every transcript variant. There was minimal effect on these transcript variants, reflecting the possibility of HNF1B and its variant in PCa aetiology, viz. possible interactions with other androgen regulated genes or via activating androgen independent pathways also defined as ‘bypass’ pathways. Next, we also observed the effect of the
86 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
chemotherapeutic drug i,e docetaxel on HNF1B transcript variants. Interestingly there was an increase in expression for all the variants excluding HNF1B transcript variant D, suggesting these variants could potentially be upregulated during this treatment, although its role at this time is unknown. We have also shown the effect of HNF1B A, B and C overexpression on cell proliferation, highlighting the importance of domains of the transcription factor. These models could now be used for other functional assays. Alternative splicing has been identified as an essential part of gene regulation. There has been a large amount of data that has shown the association of aberrant splicing in diseases such as cancer. Through this study, there has been an identification of different HNF1B splice variants in PCa, which has previously not been reported. There are numerous roles of alternative splicing in cancer; such as dysfunctional proteins, increased genomic instability, activating invasion and metastasis. Due to the presence of abnormal proteins, it helps the cancerous cells to stimulate their growth, proliferation, and tumorigenesis. It becomes challenging to tackle the splice variants, as these are not recognised by the immune system, because they have developed a complicated mechanism of action, which is difficult to be recognised by the immune system. The current research study focuses on delineating the role of HNF1B and its transcript variants in PCa. HNF1B is a transcription factor, containing different domains. The dimersation domain is important for controlling gene expression and cellular function. The DNA-binding domain attaches to the enhancer or promoter regions, the specific sequences of DNA next to regulated genes. The DNA sequences as stated above, are known as response elements. The transactivation domain, contains the binding site for other proteins for e.g. transcription co-regulators and these are referred to as activation functions. There has been an identification of variants that lack exons coding for the domains as mentioned above. We hypothesise, that any of these splice variants could potentially be participating in PCa pathogenesis via means of adhesion, migration, survival, and proliferation.
Our study provides preliminary data that gives further insight into specific HNF1B transcript variants. Researchers who focus on the understanding of tumor-specific variants are mostly to become extensively vital for the expansion of potential biomarkers thus leading to potent therapeutic strategies. Further, we could explore more about HNF1B transcription factor in terms of its regulation by using techniques such as Androgen receptor- chromatin immunoprecipitation sequencing, assay for
Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer 87
transposase-accessible chromatin sequencing and Hi-C; this will reveal us with some additional information with respect to transcription factor binding pattern along with the chromatin structure and genome organization. The previous decade in cancer research has mostly been identifying the possible mutations, next-generation sequencing of cancer genomics and exploring markers for targeting cancer treatments to susceptible patients. But still, there lies a challenge to understand the oncogenic pathway which is regulated by alternative splicing, as defining this may lead to the identification of better biomarkers for treating cancer. This is the first study that presents a holistic view on characterisation of HNF1B transcript variants followed by their functional studies in PCa and is the foundation for future research to help determine their role in PCa.
88 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Chapter 7: Conclusion
HNF1B was first reported during the end of the 20th century as a potential candidate gene for MODY; extensive genetic research indicated it as an important gene having a role in tumorigenesis. HNF1B profiles in different tumors throw light on different mechanisms governing HNF1B and its expression. GWA studies identifying HNF1B risk loci in PCa to epigenetic alterations in ovarian cancer have unraveled a potential role for this gene and its associated factors. Regulatory pathways and mechanisms involving HNF1B still need to be elucidated and studied in depth to delve into the networking of the hepatocyte nuclear factor family. Understanding the splice variants of HNF1B and their role will broaden the scope of pursuing new downstream targets, it’s associated signaling and transcriptional efficacy governing different gene sets. Further study needs to be performed around HNF1B and its splice variants to understand its ever-growing importance in different cancers.
90 Delineating the role of Hepatocyte Nuclear Factor 1 Beta (HNF1B) transcript variants in Prostate Cancer
Bibliography
AACR, A. (2017). Cancer in Australian overview of 2017. 74.
Al Olama, A. A., Kote-Jarai, Z., Berndt, S. I., Conti, D. V., Schumacher, F., Han, Y., . . . Haiman, C. A. (2014). A meta-analysis of 87,040 individuals identifies 23 new susceptibility loci for prostate cancer. Nat Genet, 46(10), 1103-1109. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/25217961. doi:10.1038/ng.3094
Amin Al Olama, A., Dadaev, T., Hazelett, D. J., Li, Q., Leongamornlert, D., Saunders, E. J., . . . Kote-Jarai, Z. (2015). Multiple novel prostate cancer susceptibility signals identified by fine-mapping of known risk loci among Europeans. Hum Mol Genet, 24(19), 5589-5602. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/26025378. doi:10.1093/hmg/ddv203
Australian Institute of Health and Welfare (2016).
Azmi, A. S., Bao, G. W., Gao, J., Mohammad, R. M., & Sarkar, F. H. (2013). Network insights into the genes regulated by hepatocyte nuclear factor 4 in response to drug induced perturbations: a review. Curr Drug Discov Technol, 10(2), 147- 154. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/23237677.
Bach, I., & Yaniv, M. (1993). More potent transcriptional activators or a transdominant inhibitor of the HNF1 homeoprotein family are generated by alternative RNA processing. EMBO J, 12(11), 4229-4242. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/7900999.
Bahcall, O. (2014). Risk prediction and population screening for breast,ovarian and prostate cancers. nature iCOGS(doi:10.1038/ngicogs.5).
Balch, C., Huang, T. H., Brown, R., & Nephew, K. P. (2004). The epigenetics of ovarian cancer drug resistance and resensitization. Am J Obstet Gynecol, 191(5), 1552-1572. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/15547525. doi:10.1016/j.ajog.2004.05.025
Barbieri, C. E., & Tomlins, S. A. (2014). The prostate cancer genome: perspectives and potential. Urol Oncol, 32(1), 53 e15-22. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24239470. doi:10.1016/j.urolonc.2013.08.025
Berndt, S. I., Sampson, J., Yeager, M., Jacobs, K. B., Wang, Z., Hutchinson, A., . . . Chanock, S. J. (2011). Large-scale fine mapping of the HNF1B locus and
Bibliography 91
prostate cancer risk. Hum Mol Genet, 20(16), 3322-3329. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21576123. doi:10.1093/hmg/ddr213
Berthold, D. R., Pond, G. R., Soban, F., de Wit, R., Eisenberger, M., & Tannock, I. F. (2008). Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer: updated survival in the TAX 327 study. J Clin Oncol, 26(2), 242-245. doi:10.1200/jco.2007.12.4008
Bingham, C., & Hattersley, A. T. (2004). Renal cysts and diabetes syndrome resulting from mutations in hepatocyte nuclear factor-1beta. Nephrol Dial Transplant, 19(11), 2703-2708. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/15496559. doi:10.1093/ndt/gfh348
Board., P. C. G. E. (2018). Genetics of Prostate Cancer (PDQ®):Health Professional Version. Bethesda (MD): National Cancer Institute (US). Retrieved from https://www.ncbi.nlm.nih.gov/books/NBK65784/.
Bois-Joyeux, B., & Danan, J. L. (1994). Members of the CAAT/enhancer-binding protein, hepatocyte nuclear factor-1 and nuclear factor-1 families can differentially modulate the activities of the rat alpha-fetoprotein promoter and enhancer. Biochem J, 301 ( Pt 1), 49-55. doi:10.1042/bj3010049
Brimo, F., Herawi, M., Sharma, R., Netto, G. J., Epstein, J. I., & Illei, P. B. (2011). Hepatocyte nuclear factor-1beta expression in clear cell adenocarcinomas of the bladder and urethra: diagnostic utility and implications for histogenesis. Hum Pathol, 42(11), 1613-1619. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/21496868. doi:10.1016/j.humpath.2011.01.007
Bubancova, I., Kovarikova, H., Laco, J., Ruszova, E., Dvorak, O., Palicka, V., & Chmelarova, M. (2017). Next-Generation Sequencing Approach in Methylation Analysis of HNF1B and GATA4 Genes: Searching for Biomarkers in Ovarian Cancer. Int J Mol Sci, 18(2). Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/28241454. doi:10.3390/ijms18020474
Buchanan, G., Irvine, R. A., Coetzee, G. A., & Tilley, W. D. (2001). Contribution of the androgen receptor to prostate cancer predisposition and progression. Cancer Metastasis Rev, 20(3-4), 207-223. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12085963.
Buchner, A., Castro, M., Hennig, A., Popp, T., Assmann, G., Stief, C. G., & Zimmermann, W. (2010). Downregulation of HNF-1B in renal cell carcinoma is associated with tumor progression and poor prognosis. Urology, 76(2), 507 e506-511. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/20538322. doi:10.1016/j.urology.2010.03.042
Casemayou, A., Fournel, A., Bagattin, A., Schanstra, J., Belliere, J., Decramer, S., . . . Faguer, S. (2017). Hepatocyte Nuclear Factor-1β Controls Mitochondrial Respiration in Renal Tubular Cells. Journal of the American Society of
92 Bibliography
Nephrology : JASN, 28(11), 3205-3217. Retrieved from PubMed.Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/28739648 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5661272/. doi:10.1681/ASN.2016050508
Chen, J., & Weiss, W. A. (2015). Alternative splicing in cancer: implications for biology and therapy. Oncogene, 34(1), 1-14. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24441040. doi:10.1038/onc.2013.570
Chodak, G. W., Kranc, D. M., Puy, L. A., Takeda, H., Johnson, K., & Chang, C. (1992). Nuclear localization of androgen receptor in heterogeneous samples of normal, hyperplastic and neoplastic human prostate. J Urol, 147(3 Pt 2), 798- 803. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/1371552.
Choi, Y. H., McNally, B. T., & Igarashi, P. (2013). Zyxin regulates migration of renal epithelial cells through activation of hepatocyte nuclear factor-1beta. Am J Physiol Renal Physiol, 305(1), F100-110. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23657850. doi:10.1152/ajprenal.00582.2012
Clissold, R., Shields, B., Ellard, S., Hattersley, A., & Bingham, C. (2015). Assessment of the HNF1B Score as a Tool to Select Patients for HNF1B Genetic Testing. Nephron, 130(2), 134-140. Retrieved from https://www.karger.com/DOI/10.1159/000398819. doi:10.1159/000398819
Clissold, R. L., Hamilton, A. J., Hattersley, A. T., Ellard, S., & Bingham, C. (2015). HNF1B-associated renal and extra-renal disease-an expanding clinical spectrum. Nat Rev Nephrol, 11(2), 102-112. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/25536396. doi:10.1038/nrneph.2014.232
Costa, R. H., Kalinichenko, V. V., Holterman, A. X., & Wang, X. (2003). Transcription factors in liver development, differentiation, and regeneration. Hepatology, 38(6), 1331-1347. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/14647040. doi:10.1016/j.hep.2003.09.034
Coutinho, I., Day, T. K., Tilley, W. D., & Selth, L. A. (2016). Androgen receptor signaling in castration-resistant prostate cancer: a lesson in persistence. Endocr Relat Cancer, 23(12), T179-T197. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/27799360. doi:10.1530/ERC-16-0422
Crawford, E. D. (2003). Epidemiology of prostate cancer. Urology, 62(6 Suppl 1), 3- 12. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/14706503.
Cuff, J., Salari, K., Clarke, N., Esheba, G. E., Forster, A. D., Huang, S., . . . Pollack, J. R. (2013). Integrative bioinformatics links HNF1B with clear cell carcinoma and tumor-associated thrombosis. PLoS One, 8(9), e74562. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24040285. doi:10.1371/journal.pone.0074562
Bibliography 93
Cunha, G. R., Donjacour, A. A., Cooke, P. S., Mee, S., Bigsby, R. M., Higgins, S. J., & Sugimura, Y. (1987). The endocrinology and developmental biology of the prostate. Endocr Rev, 8(3), 338-362. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/3308446. doi:10.1210/edrv-8-3-338
Davidson, B. (2014). Hepatocyte nuclear factor-1beta is not a specific marker of clear cell carcinoma in serous effusions. Cancer Cytopathol, 122(2), 153-158. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/24115587. doi:10.1002/cncy.21353
De Simone, V., De Magistris, L., Lazzaro, D., Gerstner, J., Monaci, P., Nicosia, A., & Cortese, R. (1991). LFB3, a heterodimer-forming homeoprotein of the LFB1 family, is expressed in specialized epithelia. Embo j, 10(6), 1435-1443.
De Vivo, I., Prescott, J., Setiawan, V. W., Olson, S. H., Wentzensen, N., Australian National Endometrial Cancer Study, G., . . . Kraft, P. (2014). Genome-wide association study of endometrial cancer in E2C2. Hum Genet, 133(2), 211-224. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24096698. doi:10.1007/s00439-013-1369-1
Debes, J. D., & Tindall, D. J. (2002). The role of androgens and the androgen receptor in prostate cancer. Cancer Lett, 187(1-2), 1-7. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12359344.
Denis, L. J., & Griffiths, K. (2000). Endocrine treatment in prostate cancer. Semin Surg Oncol, 18(1), 52-74. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10617897.
Edghill, E. L., Bingham, C., Ellard, S., & Hattersley, A. T. (2006). Mutations in hepatocyte nuclear factor-1beta and their related phenotypes. J Med Genet, 43(1), 84-90. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/15930087. doi:10.1136/jmg.2005.032854
Edghill, E. L., Oram, R. A., Owens, M., Stals, K. L., Harries, L. W., Hattersley, A. T., . . . Bingham, C. (2008). Hepatocyte nuclear factor-1beta gene deletions--a common cause of renal disease. Nephrol Dial Transplant, 23(2), 627-635. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/17971380. doi:10.1093/ndt/gfm603
Edwards, S. L., Beesley, J., French, J. D., & Dunning, A. M. (2013). Beyond GWASs: illuminating the dark road from association to function. Am J Hum Genet, 93(5), 779-797. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/24210251. doi:10.1016/j.ajhg.2013.10.012
Eeles, R. A., Kote-Jarai, Z., Giles, G. G., Olama, A. A., Guy, M., Jugurnauth, S. K., . . . Easton, D. F. (2008). Multiple newly identified loci associated with prostate
94 Bibliography
cancer susceptibility. Nat Genet, 40(3), 316-321. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/18264097. doi:10.1038/ng.90
El Marabti, E., & Younis, I. (2018). The Cancer Spliceome: Reprograming of Alternative Splicing in Cancer. Front Mol Biosci, 5, 80. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/30246013. doi:10.3389/fmolb.2018.00080
Ferrer, J. (2002). A genetic switch in pancreatic beta-cells: implications for differentiation and haploinsufficiency. Diabetes, 51(8), 2355-2362.
Glinsky, G. V., Glinskii, A. B., Stephenson, A. J., Hoffman, R. M., & Gerald, W. L. (2004). Gene expression profiling predicts clinical outcome of prostate cancer. J Clin Invest, 113(6), 913-923. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/15067324. doi:10.1172/JCI20032
Goda, N., Murase, H., Kasezawa, N., Goda, T., & Yamakawa-Kobayashi, K. (2015). Polymorphism in microRNA-binding site in HNF1B influences the susceptibility of type 2 diabetes mellitus: a population based case-control study. BMC Med Genet, 16, 75. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/26329304. doi:10.1186/s12881-015- 0219-5
Gounaris, I., Charnock-Jones, D. S., & Brenton, J. D. (2011). Ovarian clear cell carcinoma--bad endometriosis or bad endometrium? J Pathol, 225(2), 157- 160. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21898874. doi:10.1002/path.2970
Gravis, G., Fizazi, K., Joly, F., Oudard, S., Priou, F., Esterni, B., . . . Soulie, M. (2013). Androgen-deprivation therapy alone or with docetaxel in non-castrate metastatic prostate cancer (GETUG-AFU 15): a randomised, open-label, phase 3 trial. Lancet Oncol, 14(2), 149-158. doi:10.1016/s1470-2045(12)70560-0
Gregory, P. A., Hansen, A. J., & Mackenzie, P. I. (2000). Tissue specific differences in the regulation of the UDP glucuronosyltransferase 2B17 gene promoter. Pharmacogenetics, 10(9), 809-820. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11191885.
Grisanzio, C., Werner, L., Takeda, D., Awoyemi, B. C., Pomerantz, M. M., Yamada, H., . . . Freedman, M. L. (2012). Genetic and functional analyses implicate the NUDT11, HNF1B, and SLC22A3 genes in prostate cancer pathogenesis. Proc Natl Acad Sci U S A, 109(28), 11252-11257. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22730461. doi:10.1073/pnas.1200853109
Gudmundsson, J., Sulem, P., Steinthorsdottir, V., Bergthorsson, J. T., Thorleifsson, G., Manolescu, A., . . . Stefansson, K. (2007). Two variants on chromosome 17 confer prostate cancer risk, and the one in TCF2 protects against type 2 diabetes. Nat Genet, 39(8), 977-983. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/17603485. doi:10.1038/ng2062
Bibliography 95
Han, H., Braunschweig, U., Gonatopoulos-Pournatzis, T., Weatheritt, R. J., Hirsch, C. L., Ha, K. C., . . . Blencowe, B. J. (2017). Multilayered Control of Alternative Splicing Regulatory Networks by Transcription Factors. Mol Cell, 65(3), 539- 553 e537. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/28157508. doi:10.1016/j.molcel.2017.01.011
Han, X., Luo, Y., Ren, Q., Zhang, X., Wang, F., Sun, X., . . . Ji, L. (2010). Implication of genetic variants near SLC30A8, HHEX, CDKAL1, CDKN2A/B, IGF2BP2, FTO, TCF2, KCNQ1, and WFS1 in type 2 diabetes in a Chinese population. BMC Med Genet, 11, 81. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/20509872. doi:10.1186/1471-2350-11- 81
Harries, L. W., Brown, J. E., & Gloyn, A. L. (2009). Species-specific differences in the expression of the HNF1A, HNF1B and HNF4A genes. PLoS One, 4(11), e7855. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/19924231. doi:10.1371/journal.pone.0007855
Harries, L. W., Ellard, S., Stride, A., Morgan, N. G., & Hattersley, A. T. (2006). Isomers of the TCF1 gene encoding hepatocyte nuclear factor-1 alpha show differential expression in the pancreas and define the relationship between mutation position and clinical phenotype in monogenic diabetes. Hum Mol Genet, 15(14), 2216-2224. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/16760222. doi:10.1093/hmg/ddl147
Harries, L. W., Locke, J. M., Shields, B., Hanley, N. A., Hanley, K. P., Steele, A., . . . Hattersley, A. T. (2008). The diabetic phenotype in HNF4A mutation carriers is moderated by the expression of HNF4A isoforms from the P1 promoter during fetal development. Diabetes, 57(6), 1745-1752. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/18356407. doi:10.2337/db07-1742
Harries, L. W., Perry, J. R., McCullagh, P., & Crundwell, M. (2010). Alterations in LMTK2, MSMB and HNF1B gene expression are associated with the development of prostate cancer. BMC Cancer, 10, 315. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/20569440. doi:10.1186/1471-2407- 10-315
Hatzis, P., & Talianidis, I. (2001). Regulatory mechanisms controlling human hepatocyte nuclear factor 4alpha gene expression. Mol Cell Biol, 21(21), 7320- 7330. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/11585914. doi:10.1128/MCB.21.21.7320-7330.2001
Heinlein, C. A., & Chang, C. (2002). Androgen receptor (AR) coregulators: an overview. Endocr Rev, 23(2), 175-200. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11943742. doi:10.1210/edrv.23.2.0460
Heliot, C., Desgrange, A., Buisson, I., Prunskaite-Hyyrylainen, R., Shan, J., Vainio, S., . . . Cereghini, S. (2013). HNF1B controls proximal-intermediate nephron segment identity in vertebrates by regulating Notch signalling components and
96 Bibliography
Irx1/2. Development, 140(4), 873-885. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23362348. doi:10.1242/dev.086538
Hoang, L. N., Han, G., McConechy, M., Lau, S., Chow, C., Gilks, C. B., . . . Lee, C. H. (2014). Immunohistochemical characterization of prototypical endometrial clear cell carcinoma--diagnostic utility of HNF-1beta and oestrogen receptor. Histopathology, 64(4), 585-596. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/24103020. doi:10.1111/his.12286
Horikawa, Y., Iwasaki, N., Hara, M., Furuta, H., Hinokio, Y., Cockburn, B. N., . . . Bell, G. I. (1997). Mutation in hepatocyte nuclear factor-1 beta gene (TCF2) associated with MODY. Nat Genet, 17(4), 384-385. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/9398836. doi:10.1038/ng1297-384
Hoskins, J. W., Jia, J., Flandez, M., Parikh, H., Xiao, W., Collins, I., . . . Amundadottir, L. T. (2014). Transcriptome analysis of pancreatic cancer reveals a tumor suppressor function for HNF1A. Carcinogenesis, 35(12), 2670-2678. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/25233928. doi:10.1093/carcin/bgu193
Hu, Y. L., Zhong, D., Pang, F., Ning, Q. Y., Zhang, Y. Y., Li, G., . . . Mo, Z. N. (2013). HNF1b is involved in prostate cancer risk via modulating androgenic hormone effects and coordination with other genes. Genet Mol Res, 12(2), 1327-1335. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23661456. doi:10.4238/2013.April.25.4
Huang, W., Cheng, X., Ji, J., Zhang, J., & Li, Q. (2016). The Application Value of HNF-1beta Transcription Factor in the Diagnosis of Ovarian Clear Cell Carcinoma. Int J Gynecol Pathol, 35(1), 66-71. doi:10.1097/pgp.0000000000000213
James, N. D., Sydes, M. R., Clarke, N. W., Mason, M. D., Dearnaley, D. P., Spears, M. R., . . . Parmar, M. K. (2016). Addition of docetaxel, zoledronic acid, or both to first-line long-term hormone therapy in prostate cancer (STAMPEDE): survival results from an adaptive, multiarm, multistage, platform randomised controlled trial. Lancet, 387(10024), 1163-1177. doi:10.1016/s0140- 6736(15)01037-5
Janky, R., Binda, M. M., Allemeersch, J., Van den Broeck, A., Govaere, O., Swinnen, J. V., . . . Topal, B. (2016). Prognostic relevance of molecular subtypes and master regulators in pancreatic ductal adenocarcinoma. BMC Cancer, 16, 632. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/27520560. doi:10.1186/s12885-016-2540-6
Jernberg, E., Bergh, A., & Wikstrom, P. (2017). Clinical relevance of androgen receptor alterations in prostate cancer. Endocr Connect, 6(8), R146-R161. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/29030409. doi:10.1530/EC-17-0118
Bibliography 97
Kajihara, H., Yamada, Y., Kanayama, S., Furukawa, N., Noguchi, T., Haruta, S., . . . Kobayashi, H. (2010). Clear cell carcinoma of the ovary: potential pathogenic mechanisms (Review). Oncol Rep, 23(5), 1193-1203.
Kao, Y.-C., Lin, M.-C., Lin, W.-C., Jeng, Y.-M., & Mao, T.-L. (2012). Utility of hepatocyte nuclear factor-1β as a diagnostic marker in ovarian carcinomas with clear cells. Histopathology, 61(5), 760-768. Retrieved from http://dx.doi.org/10.1111/j.1365-2559.2012.04267.x. doi:10.1111/j.1365- 2559.2012.04267.x
Kato, N., & Motoyama, T. (2008). Overexpression of osteopontin in clear cell carcinoma of the ovary: close association with HNF-1beta expression. Histopathology, 52(6), 682-688. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/18393978. doi:10.1111/j.1365- 2559.2008.03006.x
Kim, H. J., Bae, J. S., Lee, J., Chang, I. H., Kim, K. D., Shin, H. D., . . . Myung, S. C. (2011). HNF1B polymorphism associated with development of prostate cancer in Korean patients. Urology, 78(4), 969 e961-966. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21982019. doi:10.1016/j.urology.2011.06.045
Kim, J., & Coetzee, G. A. (2004). Prostate specific antigen gene regulation by androgen receptor. J Cell Biochem, 93(2), 233-241. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/15368351. doi:10.1002/jcb.20228
Kim, L., Liao, J., Zhang, M., Talamonti, M., Bentrem, D., Rao, S., & Yang, G. Y. (2008). Clear cell carcinoma of the pancreas: histopathologic features and a unique biomarker: hepatocyte nuclear factor-1beta. Mod Pathol, 21(9), 1075- 1083. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/18536653. doi:10.1038/modpathol.2008.95
Kornfeld, J. W., Baitzel, C., Konner, A. C., Nicholls, H. T., Vogt, M. C., Herrmanns, K., . . . Bruning, J. C. (2013). Obesity-induced overexpression of miR-802 impairs glucose metabolism through silencing of Hnf1b. Nature, 494(7435), 111-115. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23389544. doi:10.1038/nature11793
Kristiansen, W., Karlsson, R., Rounge, T. B., Whitington, T., Andreassen, B. K., Magnusson, P. K., . . . Wiklund, F. (2015). Two new loci and gene sets related to sex determination and cancer progression are associated with susceptibility to testicular germ cell tumor. Hum Mol Genet, 24(14), 4138-4146. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/25877299. doi:10.1093/hmg/ddv129
Lau, H. H., Ng, N. H. J., Loo, L. S. W., Jasmen, J., & Teo, A. K. K. (2017). The molecular functions of hepatocyte nuclear factors in and beyond the liver. J Hepatol. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/29175243. doi:10.1016/j.jhep.2017.11.026
98 Bibliography
Lazarevich, N. L. (2000). Molecular mechanisms of alpha-fetoprotein gene expression. Biochemistry (Mosc), 65(1), 117-133.
Lebrun, G., Vasiliu, V., Bellanne-Chantelot, C., Bensman, A., Ulinski, T., Chretien, Y., & Grunfeld, J. P. (2005). Cystic kidney disease, chromophobe renal cell carcinoma and TCF2 (HNF1 beta) mutations. Nat Clin Pract Nephrol, 1(2), 115-119. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/16932376. doi:10.1038/ncpneph0054
Levin, A. M., Machiela, M. J., Zuhlke, K. A., Ray, A. M., Cooney, K. A., & Douglas, J. A. (2008). Chromosome 17q12 variants contribute to risk of early-onset prostate cancer. Cancer Res, 68(16), 6492-6495. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/18701471. doi:10.1158/0008- 5472.CAN-08-0348
Li, D., Duell, E. J., Yu, K., Risch, H. A., Olson, S. H., Kooperberg, C., . . . Stolzenberg- Solomon, R. Z. (2012). Pathway analysis of genome-wide association study data highlights pancreatic development genes as susceptibility factors for pancreatic cancer. Carcinogenesis, 33(7), 1384-1390. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22523087. doi:10.1093/carcin/bgs151
Li, H., Gan, W., Lu, L., Dong, X., Han, X., Hu, C., . . . Lin, X. (2013). A genome-wide association study identifies GRK5 and RASGRP1 as type 2 diabetes loci in Chinese Hans. Diabetes, 62(1), 291-298. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22961080. doi:10.2337/db12-0454
Li, J., Zhang, Y., Gao, Y., Cui, Y., Liu, H., Li, M., & Tian, Y. (2014). Downregulation of HNF1 homeobox B is associated with drug resistance in ovarian cancer. Oncol Rep, 32(3), 979-988. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24968817. doi:10.3892/or.2014.3297
Lichtenstein, P., Holm, N. V., Verkasalo, P. K., Iliadou, A., Kaprio, J., Koskenvuo, M., . . . Hemminki, K. (2000). Environmental and heritable factors in the causation of cancer--analyses of cohorts of twins from Sweden, Denmark, and Finland. N Engl J Med, 343(2), 78-85. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10891514. doi:10.1056/NEJM200007133430201
Lonergan, P. E., & Tindall, D. J. (2011). Androgen receptor signaling in prostate cancer development and progression. Journal of carcinogenesis, 10, 20-20. Retrieved from PubMed.Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/21886458 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3162670/. doi:10.4103/1477- 3163.83937
Long, J., Zheng, W., Xiang, Y. B., Lose, F., Thompson, D., Tomlinson, I., . . . Shu, X. O. (2012). Genome-wide association study identifies a possible susceptibility locus for endometrial cancer. Cancer Epidemiol Biomarkers Prev, 21(6), 980- 987. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22426144. doi:10.1158/1055-9965.EPI-11-1160
Bibliography 99
Lundon, D. J., Boland, A., Prencipe, M., Hurley, G., O'Neill, A., Kay, E., . . . Watson, R. W. (2017). The prognostic utility of the transcription factor SRF in docetaxel-resistant prostate cancer: in-vitro discovery and in-vivo validation. BMC Cancer, 17(1), 163. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/28249598. doi:10.1186/s12885-017- 3100-4
Ma, Z., Gong, Y., Patel, V., Karner, C. M., Fischer, E., Hiesberger, T., . . . Igarashi, P. (2007). Mutations of HNF-1beta inhibit epithelial morphogenesis through dysregulation of SOCS-3. Proc Natl Acad Sci U S A, 104(51), 20386-20391. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/18077349. doi:10.1073/pnas.0705957104
Machiela, M. J., Lindstrom, S., Allen, N. E., Haiman, C. A., Albanes, D., Barricarte, A., . . . Kraft, P. (2012). Association of type 2 diabetes susceptibility variants with advanced prostate cancer risk in the Breast and Prostate Cancer Cohort Consortium. Am J Epidemiol, 176(12), 1121-1129. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23193118. doi:10.1093/aje/kws191
Matsui, A., Fujimoto, J., Ishikawa, K., Ito, E., Goshima, N., Watanabe, S., & Semba, K. (2016). Hepatocyte nuclear factor 1 beta induces transformation and epithelial-to-mesenchymal transition. FEBS Lett, 590(8), 1211-1221. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/27001343. doi:10.1002/1873-3468.12147
Mendel, D. B., Hansen, L. P., Graves, M. K., Conley, P. B., & Crabtree, G. R. (1991). HNF-1 alpha and HNF-1 beta (vHNF-1) share dimerization and homeo domains, but not activation domains, and form heterodimers in vitro. Genes Dev, 5(6), 1042-1056. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/2044952.
Miyake, K., Yang, W., Hara, K., Yasuda, K., Horikawa, Y., Osawa, H., . . . Kasuga, M. (2009). Construction of a prediction model for type 2 diabetes mellitus in the Japanese population based on 11 genes with strong evidence of the association. J Hum Genet, 54(4), 236-241. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/19247372. doi:10.1038/jhg.2009.17
Mohler, J. L., Chen, Y., Hamil, K., Hall, S. H., Cidlowski, J. A., Wilson, E. M., . . . Sar, M. (1996). Androgen and glucocorticoid receptors in the stroma and epithelium of prostatic hyperplasia and carcinoma. Clin Cancer Res, 2(5), 889- 895. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9816246.
Montanari, M., Rossetti, S., Cavaliere, C., D'Aniello, C., Malzone, M. G., Vanacore, D., . . . Facchini, G. (2017). Epithelial-mesenchymal transition in prostate cancer: an overview. Oncotarget, 8(21), 35376-35389. Retrieved from PMC.Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5471062/. doi:10.18632/oncotarget.15686
100 Bibliography
Nakagawa, H., Akamatsu, S., & Takata, R. (2016). [Genome-wide association study(GWAS) and genetic risk of prostate cancer]. Nihon Rinsho, 74(1), 34- 39. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/26793876.
Ninomiya, T., Hayashi, Y., Saijoh, K., Ohta, K., Yoon, S., Nakabayashi, H., . . . Itoh, H. (1996). Expression ratio of hepatocyte nuclear factor-1 to variant hepatocyte nuclear factor-1 in differentiation of hepatocellular carcinoma and hepatoblastoma. J Hepatol, 25(4), 445-453.
Okamoto, T., Mandai, M., Matsumura, N., Yamaguchi, K., Kondoh, H., Amano, Y., . . . Konishi, I. (2015). Hepatocyte nuclear factor-1β (HNF-1β) promotes glucose uptake and glycolytic activity in ovarian clear cell carcinoma. Molecular Carcinogenesis, 54(1), 35-49. Retrieved from http://dx.doi.org/10.1002/mc.22072. doi:10.1002/mc.22072
Omata, M., Doke, Y., Yamada, C., Kawashima, K., Sho, R., Enomoto, K., . . . Inomata, N. (2016). Hepatocyte Nuclear Factor-1beta Induces Redifferentiation of Dedifferentiated Tubular Epithelial Cells. PLoS One, 11(5), e0154912. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/27196561. doi:10.1371/journal.pone.0154912
Pagani, F., Buratti, E., Stuani, C., Bendix, R., Dork, T., & Baralle, F. E. (2002). A new type of mutation causes a splicing defect in ATM. Nat Genet, 30(4), 426-429. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11889466. doi:10.1038/ng858
Painter, J. N., O'Mara, T. A., Batra, J., Cheng, T., Lose, F. A., Dennis, J., . . . Spurdle, A. B. (2015). Fine-mapping of the HNF1B multicancer locus identifies candidate variants that mediate endometrial cancer risk. Hum Mol Genet, 24(5), 1478-1492. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/25378557. doi:10.1093/hmg/ddu552
Pihlajamaa, P., Sahu, B., Lyly, L., Aittomaki, V., Hautaniemi, S., & Janne, O. A. (2014). Tissue-specific pioneer factors associate with androgen receptor cistromes and transcription programs. EMBO J, 33(4), 312-326. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24451200. doi:10.1002/embj.201385895
Puente, J., Grande, E., Medina, A., Maroto, P., Lainez, N., & Arranz, J. A. (2017). Docetaxel in prostate cancer: a familiar face as the new standard in a hormone- sensitive setting. Ther Adv Med Oncol, 9(5), 307-318. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/28529548. doi:10.1177/1758834017692779
Rangaswami, H., Bulbule, A., & Kundu, G. C. (2006). Osteopontin: role in cell signaling and cancer progression. Trends Cell Biol, 16(2), 79-87. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/16406521. doi:10.1016/j.tcb.2005.12.005
Bibliography 101
Rebouissou, S., Vasiliu, V., Thomas, C., Bellanne-Chantelot, C., Bui, H., Chretien, Y., . . . Zucman-Rossi, J. (2005). Germline hepatocyte nuclear factor 1alpha and 1beta mutations in renal cell carcinomas. Hum Mol Genet, 14(5), 603-614. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/15649945. doi:10.1093/hmg/ddi057
Rios-Tamayo, R., Lupianez, C. B., Campa, D., Hielscher, T., Weinhold, N., Martinez- Lopez, J., . . . Sainz, J. (2016). A common variant within the HNF1B gene is associated with overall survival of multiple myeloma patients: results from the IMMEnSE consortium and meta-analysis. Oncotarget, 7(37), 59029-59048. doi:10.18632/oncotarget.10665
Romero, L., Ng, L., & Kirby, G. M. (2006). Chemical inducers of rodent glutathione s-transferases down-regulate human GSTA1 transcription through a mechanism involving variant hepatic nuclear factor 1-C. Mol Pharmacol, 70(1), 277-286. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/16608922. doi:10.1124/mol.105.018622
Ross-Adams, H., Ball, S., Lawrenson, K., Halim, S., Russell, R., Wells, C., . . . Mills, I. G. (2016). HNF1B variants associate with promoter methylation and regulate gene networks activated in prostate and ovarian cancer. Oncotarget, 7(46), 74734-74746. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/27732966. doi:10.18632/oncotarget.12543
Roy, A. K., Lavrovsky, Y., Song, C. S., Chen, S., Jung, M. H., Velu, N. K., . . . Chatterjee, B. (1999). Regulation of androgen action. Vitam Horm, 55, 309- 352. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9949684.
Sadi, M. V., Walsh, P. C., & Barrack, E. R. (1991). Immunohistochemical study of androgen receptors in metastatic prostate cancer. Comparison of receptor content and response to hormonal therapy. Cancer, 67(12), 3057-3064. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/1710537.
Saxena, P., Trerotola, M., Wang, T., Li, J., Sayeed, A., Vanoudenhove, J., . . . Languino, L. R. (2012). PSA regulates androgen receptor expression in prostate cancer cells. Prostate, 72(7), 769-776. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21956655. doi:10.1002/pros.21482
Senkel, S., Lucas, B., Klein-Hitpass, L., & Ryffel, G. U. (2005). Identification of target genes of the transcription factor HNF1beta and HNF1alpha in a human embryonic kidney cell line. Biochim Biophys Acta, 1731(3), 179-190. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16297991. doi:10.1016/j.bbaexp.2005.10.003
Setiawan, V. W., Haessler, J., Schumacher, F., Cote, M. L., Deelman, E., Fesinmeyer, M. D., . . . Kooperberg, C. (2012). HNF1B and endometrial cancer risk: results from the PAGE study. PLoS One, 7(1), e30390. Retrieved from
102 Bibliography
https://www.ncbi.nlm.nih.gov/pubmed/22299039. doi:10.1371/journal.pone.0030390
Shen, H., Fridley, B. L., Song, H., Lawrenson, K., Cunningham, J. M., Ramus, S. J., . . . Pearce, C. L. (2013). Epigenetic analysis leads to identification of HNF1B as a subtype-specific susceptibility gene for ovarian cancer. Nat Commun, 4, 1628. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/23535649. doi:10.1038/ncomms2629
Shigetomi, H., Higashiura, Y., Kajihara, H., & Kobayashi, H. (2012). Targeted molecular therapies for ovarian cancer: an update and future perspectives (Review). Oncol Rep, 28(2), 395-408. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/22641286. doi:10.3892/or.2012.1833
Shim, J. H., Lee, H. C., Han, S., Kang, H. J., Yu, E., & Lee, S. G. (2013). Hepatocyte nuclear factor 1beta is a novel prognostic marker independent of the Milan criteria in transplantable hepatocellular carcinoma: a retrospective analysis based on tissue microarrays. Liver Transpl, 19(3), 336-345. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/23203386. doi:10.1002/lt.23584
Silva, T. D., Vidigal, V. M., Felipe, A. V., De Lima, J. M., Neto, R. A., Saad, S. S., & Forones, N. M. (2013). DNA methylation as an epigenetic biomarker in colorectal cancer. Oncology Letters, 6(6), 1687-1692. Retrieved from PMC.Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3834199/. doi:10.3892/ol.2013.1606
Singh, B., & Eyras, E. (2017). The role of alternative splicing in cancer. Transcription, 8(2), 91-98. doi:10.1080/21541264.2016.1268245
Spurdle, A. B., Thompson, D. J., Ahmed, S., Ferguson, K., Healey, C. S., O'Mara, T., . . . Easton, D. F. (2011). Genome-wide association study identifies a common variant associated with risk of endometrial cancer. Nat Genet, 43(5), 451-454. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21499250. doi:10.1038/ng.812
Sun, J., Zheng, S. L., Wiklund, F., Isaacs, S. D., Purcell, L. D., Gao, Z., . . . Xu, J. (2008). Evidence for two independent prostate cancer risk-associated loci in the HNF1B gene at 17q12. Nat Genet, 40(10), 1153-1155. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/18758462. doi:10.1038/ng.214
Sveen, A., Kilpinen, S., Ruusulehto, A., Lothe, R. A., & Skotheim, R. I. (2016). Aberrant RNA splicing in cancer; expression changes and driver mutations of splicing factor genes. Oncogene, 35(19), 2413-2427. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/26300000. doi:10.1038/onc.2015.318
Sweeney, C. J., Chen, Y. H., Carducci, M., Liu, G., Jarrard, D. F., Eisenberger, M., . . . DiPaola, R. S. (2015). Chemohormonal Therapy in Metastatic Hormone- Sensitive Prostate Cancer. N Engl J Med, 373(8), 737-746. doi:10.1056/NEJMoa1503747
Bibliography 103
Talavera, D., Orozco, M., & de la Cruz, X. (2009). Alternative splicing of transcription factors' genes: beyond the increase of proteome diversity. Comp Funct Genomics, 905894. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/19609452. doi:10.1155/2009/905894
Tan, M. H., Li, J., Xu, H. E., Melcher, K., & Yong, E. L. (2015). Androgen receptor: structure, role in prostate cancer and drug discovery. Acta Pharmacol Sin, 36(1), 3-23. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24909511. doi:10.1038/aps.2014.18
Teo, A. K., Lau, H. H., Valdez, I. A., Dirice, E., Tjora, E., Raeder, H., & Kulkarni, R. N. (2016). Early Developmental Perturbations in a Human Stem Cell Model of MODY5/HNF1B Pancreatic Hypoplasia. Stem Cell Reports, 6(3), 357-367. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/26876668. doi:10.1016/j.stemcr.2016.01.007
Terasawa, K., Toyota, M., Sagae, S., Ogi, K., Suzuki, H., Sonoda, T., . . . Tokino, T. (2006). Epigenetic inactivation of TCF2 in ovarian cancer and various cancer cell lines. Br J Cancer, 94(6), 914-921. Retrieved from http://dx.doi.org/10.1038/sj.bjc.6602984. doi:http://www.nature.com/bjc/journal/v94/n6/suppinfo/6602984s1.html
Thomas, G., Jacobs, K. B., Yeager, M., Kraft, P., Wacholder, S., Orr, N., . . . Chanock, S. J. (2008). Multiple loci identified in a genome-wide association study of prostate cancer. Nat Genet, 40(3), 310-315. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/18264096. doi:10.1038/ng.91
Tommasi, S., Karm, D. L., Wu, X., Yen, Y., & Pfeifer, G. P. (2009). Methylation of homeobox genes is a frequent and early epigenetic event in breast cancer. Breast Cancer Res, 11(1), R14. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/19250546. doi:10.1186/bcr2233
Tsuchiya, A., Sakamoto, M., Yasuda, J., Chuma, M., Ohta, T., Ohki, M., . . . Hirohashi, S. (2003). Expression profiling in ovarian clear cell carcinoma: identification of hepatocyte nuclear factor-1 beta as a molecular marker and a possible molecular target for therapy of ovarian clear cell carcinoma. Am J Pathol, 163(6), 2503-2512. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/14633622. van der Kwast, T. H., Schalken, J., Ruizeveld de Winter, J. A., van Vroonhoven, C. C., Mulder, E., Boersma, W., & Trapman, J. (1991). Androgen receptors in endocrine-therapy-resistant human prostate cancer. Int J Cancer, 48(2), 189- 193. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/1708363.
Verhave, J. C., Bech, A. P., Wetzels, J. F., & Nijenhuis, T. (2016). Hepatocyte Nuclear Factor 1beta-Associated Kidney Disease: More than Renal Cysts and Diabetes. J Am Soc Nephrol, 27(2), 345-353. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/26319241. doi:10.1681/ASN.2015050544
104 Bibliography
Wang, C. C., Mao, T. L., Yang, W. C., & Jeng, Y. M. (2013). Underexpression of hepatocyte nuclear factor-1beta in chromophobe renal cell carcinoma. Histopathology, 62(4), 589-594. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/23237209. doi:10.1111/his.12026
Wang, Q., Li, W., Liu, X. S., Carroll, J. S., Janne, O. A., Keeton, E. K., . . . Brown, M. (2007). A hierarchical network of transcription factors governs androgen receptor-dependent prostate cancer growth. Mol Cell, 27(3), 380-392. doi:10.1016/j.molcel.2007.05.041
Wang, W., Hayashi, Y., Ninomiya, T., Ohta, K., Nakabayashi, H., Tamaoki, T., & Itoh, H. (1998). Expression of HNF-1 alpha and HNF-1 beta in various histological differentiations of hepatocellular carcinoma. J Pathol, 184(3), 272-278. doi:10.1002/(sici)1096-9896(199803)184:3<272::aid- path4>3.0.co;2-k
Welter, D., MacArthur, J., Morales, J., Burdett, T., Hall, P., Junkins, H., . . . Parkinson, H. (2014). The NHGRI GWAS Catalog, a curated resource of SNP-trait associations. Nucleic Acids Res, 42(Database issue), D1001-1006. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24316577. doi:10.1093/nar/gkt1229
Welters, H. J., Oknianska, A., Erdmann, K. S., Ryffel, G. U., & Morgan, N. G. (2008). The protein tyrosine phosphatase-BL, modulates pancreatic beta-cell proliferation by interaction with the Wnt signalling pathway. J Endocrinol, 197(3), 543-552. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/18413347. doi:10.1677/JOE-07-0262
Welters, H. J., Senkel, S., Klein-Hitpass, L., Erdmann, S., Thomas, H., Harries, L. W., . . . Morgan, N. G. (2006). Conditional expression of hepatocyte nuclear factor- 1beta, the maturity-onset diabetes of the young-5 gene product, influences the viability and functional competence of pancreatic beta-cells. J Endocrinol, 190(1), 171-181. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16837621. doi:10.1677/joe.1.06768
Wen, J., Ronn, T., Olsson, A., Yang, Z., Lu, B., Du, Y., . . . Hu, R. (2010). Investigation of type 2 diabetes risk alleles support CDKN2A/B, CDKAL1, and TCF7L2 as susceptibility genes in a Han Chinese cohort. PLoS One, 5(2), e9153. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/20161779. doi:10.1371/journal.pone.0009153
Winckler, W., Weedon, M. N., Graham, R. R., McCarroll, S. A., Purcell, S., Almgren, P., . . . Altshuler, D. (2007). Evaluation of common variants in the six known maturity-onset diabetes of the young (MODY) genes for association with type 2 diabetes. Diabetes, 56(3), 685-693. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/17327436. doi:10.2337/db06-0202
Wu, G., Bohn, S., & Ryffel, G. U. (2004). The HNF1beta transcription factor has several domains involved in nephrogenesis and partially rescues Pax8/lim1-
Bibliography 105
induced kidney malformations. Eur J Biochem, 271(18), 3715-3728. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/15355349. doi:10.1111/j.1432- 1033.2004.04312.x
Xu, J., Capezzone, M., Xu, X., & Hershman, J. M. (2005). Activation of nicotinamide N-methyltransferase gene promoter by hepatocyte nuclear factor-1beta in human papillary thyroid cancer cells. Mol Endocrinol, 19(2), 527-539. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/15486044. doi:10.1210/me.2004-0215
Yamamoto, S., Tsuda, H., Aida, S., Shimazaki, H., Tamai, S., & Matsubara, O. (2007). Immunohistochemical detection of hepatocyte nuclear factor 1beta in ovarian and endometrial clear-cell adenocarcinomas and nonneoplastic endometrium. Hum Pathol, 38(7), 1074-1080. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/17442376. doi:10.1016/j.humpath.2006.12.018
Yang, M. X., Coates, R. F., Ambaye, A., Gardner, J.-A., Zubarick, R., Gao, Y., . . . Mino-Kenudson, M. (2018). Investigation of HNF-1B as a diagnostic biomarker for pancreatic ductal adenocarcinoma. Biomarker research, 6, 25- 25. Retrieved from PubMed.Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/30065837 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6062876/. doi:10.1186/s40364-018- 0139-6
Yeo, G. W., Van Nostrand, E., Holste, D., Poggio, T., & Burge, C. B. (2005). Identification and analysis of alternative splicing events conserved in human and mouse. Proc Natl Acad Sci U S A, 102(8), 2850-2855. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/15708978. doi:10.1073/pnas.0409742102
Yonemori, K., Kurahara, H., Maemura, K., & Natsugoe, S. (2017). MicroRNA in pancreatic cancer. J Hum Genet, 62(1), 33-40. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/27251005. doi:10.1038/jhg.2016.59
You, J. S., & Jones, P. A. (2012). Cancer genetics and epigenetics: two sides of the same coin? Cancer Cell, 22(1), 9-20. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/22789535. doi:10.1016/j.ccr.2012.06.008
Young, R. H., & Scully, R. E. (1985). Clear cell adenocarcinoma of the bladder and urethra. A report of three cases and review of the literature. Am J Surg Pathol, 9(11), 816-826.
Yu, D. D., Guo, S. W., Jing, Y. Y., Dong, Y. L., & Wei, L. X. (2015). A review on hepatocyte nuclear factor-1beta and tumor. Cell Biosci, 5, 58. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/26464794. doi:10.1186/s13578-015- 0049-3
106 Bibliography
Zhang, X., Qiao, H., Zhao, Y., Wang, X., Sun, H., Liu, A., . . . Fu, S. (2012). Association of single nucleotide polymorphisms in TCF2 with type 2 diabetes susceptibility in a Han Chinese population. PLoS One, 7(12), e52938. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23300827. doi:10.1371/journal.pone.0052938
Zhang, Y. R., Xu, Y., Yang, K., Liu, M., Wei, D., Zhang, Y. G., . . . Yang, Z. (2012). Association of six susceptibility Loci with prostate cancer in northern chinese men. Asian Pac J Cancer Prev, 13(12), 6273-6276. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23464444.
Bibliography 107
Appendices
Appendix A
The siRNA sequences used in this study
HNF1B siRNA Sense (5’-3’) Antisense (3’-5’)
SiRNA1 CCAGCCAGUCGGUUUUACAtt UGUAAAACCGACUGGUGGtc
SiRNA2 CUCCGACAAUUCAACCAGAtt UCUGGUUGAAUUGUCGGAGga
SiRNA3 GAAUAUACUCCCUGGAAAUU UUUCCAGGGAGUAUAUUCUU
Appendix B
Figure 7-1-Gel images for HNF1B transcript variants: HNF1B transcript variants AB, CC’, DE. The expected band sizes have been marked on the image against the transcript variant.
108 Appendices
Figure 7-2-Sequence confirmation for HNF1B transcript variants A and B
Figure 7-3-Sequence confirmed for HNF1B transcript variant C (cloned in p GEMT vector)
Figure 7-4- Western blotting for overexpressed clones for HNF1B a: Western blotting was carried out to confirm the overexpressed HNF1B A variant model at the protein level (the band highlighted with red box), we used actin as a housekeeping gene.
Appendices 109
Figure 7-5: Ingenuity Pathways Analysis (IPA) for prediction of upstream pathways involved in HNF1B overexpression model A.
Raw Data Storage:
U drive> Research>Projects>ihbi>hdc>mol_gen_students>Shubhra
110 Appendices