UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Dissecting the function of Y-box binding protein-1 (YB-1) during the development of breast cancer Davies, Alastair Henry 2013

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

Item Metadata


24-ubc_2013_spring_davies_alastair.pdf [ 40.76MB ]
JSON: 24-1.0073701.json
JSON-LD: 24-1.0073701-ld.json
RDF/XML (Pretty): 24-1.0073701-rdf.xml
RDF/JSON: 24-1.0073701-rdf.json
Turtle: 24-1.0073701-turtle.txt
N-Triples: 24-1.0073701-rdf-ntriples.txt
Original Record: 24-1.0073701-source.json
Full Text

Full Text

DISSECTING THE FUNCTION OF Y-BOX BINDING PROTEIN-1 (YB-1) DURING THE DEVELOPMENT OF BREAST CANCER  by Alastair Henry Davies B.H.Sc. (Hons), The University of Calgary, 2007  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Experimental Medicine) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2013  © Alastair Henry Davies, 2013  ABSTRACT The series of events that trigger the transformation of a normal cell into a malignant cell are unclear. One prospective driver of tumourigenesis is Y-box binding protein-1 (YB-1). BLG-YB-1 transgenic mice all form mammary tumours and the protein is expressed in over 40% of breast carcinomas. The studies presented in this thesis aimed to uncover the function of YB-1 during the transformation process. To this end, we conditionally expressed YB-1 in normal human mammary epithelial cells (HMECs). In agreement with tumours from the transgenic mice, genomic instability was manifested in the form of numerical and structural chromosomal abnormalities. To query the mechanism responsible for these phenotypes, we assessed global changes in signal transduction using an antibody microarray. Notably, we identified an increase in LIM kinase (LIMK1/2) activity that acted as a catalyst for cytokinesis failure. Subsequent cell cycle checkpoint slippage, due to cyclin E over-expression, potentiated centrosome amplification leading to mitotic spindle abnormalities. The resulting genomic instability was not stochastic but rather it increased susceptibility to cancer by promoting low-level HER2 amplification, as one example. Deeper interrogation revealed that YB-1 was capable of fully transforming HMECs. Through stabilization and upregulation of the histone acetyltransferase p300, YB-1 reprogrammed HMECs into stem/progenitor-like tumour-initiating cells (TICs). Mechanistically, the relaxation of promoter-centered chromatin permitted YB-1 to bind and transcriptionally regulate the TIC-associated genes BMI1, CD44, and CD49f. These cells formed DCIS-like luminal outgrowths in a three-dimensional model of breast acini. Over time, pressures exerted by YB-1 led to the emergence of cells expressing RSK2 and hTERT that had the capacity to form tumours in vivo. These cells were subtyped as triplenegative breast cancer (TNBC), a particularly aggressive form of the disease that is prone to relapse. We discovered that YB-1 regulates the multidrug resistance transporter ABCG2 to render the CD44+/CD49f+ TIC subpopulation refractory to traditional chemotherapy. However, these cells were responsive to RSK inhibitors, which prevent the activation of YB-1. As a whole, the studies outlined in this thesis indicate that YB-1 facilitates the genesis of TNBC through epigenetic reprogramming and targeting it has the potential to overcome drug resistance and prevent tumour recurrence.  ii  PREFACE Chapter 2: YB-1 evokes susceptibility to cancer through cytokinesis failure, mitotic dysfunction, and HER2 amplification.  Alastair Davies conceived the project, performed more than 90% of the experiments (including all microscopy and imaging, immunoblotting, siRNA and drug treatments, microtubule regrowth assays, quantification of metaphase spreads, and optimization of the GFP:YB-1 construct), analyzed all data, and prepared the manuscript in its entirety. Contributions of co-authors: •  Irene Barrett provided technical assistance with the preparation of metaphase spreads and aided in the analysis of HER2 FISH data (Figure S2.9).  •  Mary Rose Pambid assisted in the characterization of YB-1 expression in the HTRZ and HTRY cell lines (Figure S2.1A).  •  Kaiji Hu, PhD, provided technical assistance with Cellomics screening.  •  Anna L. Stratford, PhD, provided protein lysates from the MDA-MB-231 cell line treated with siRNA targeting RSK1 and RSK2 (used in Figure S2.2).  •  Spencer Freeman aided with confocal microscopy (Figure 2.1E).  •  Isabelle M. Berquin, PhD, developed the HTRZ and HTRY cell lines.  •  Steven Pelech, PhD, provided the Kinexus antibody microarray and aided in the analysis of the expression data.  •  Philip Hieter, PhD, assisted with editing the manuscript.  •  Christopher Maxwell, PhD, provided technical guidance with the microtubule regrowth assay and assisted with editing the manuscript.  •  Sandra E. Dunn, PhD, supervised the project and assisted with editing the manuscript.  A version of this chapter has been published. Davies AH, Barrett I, Pambid MR, Hu K, Stratford AL, Freeman S, Berquin IM, Pelech S, Hieter P, Maxwell C, and Dunn SE. YB-1 evokes susceptibility to cancer through cytokinesis failure, mitotic dysfunction, and HER2 amplification. Oncogene (2011).  iii  Chapter 3: YB-1 promotes tumour-initiation through chromatin remodeling leading to the development of triple-negative breast cancer.  Alastair Davies conceived the project, performed more than 90% of the experiments (including all microscopy and imaging, siRNA and drug treatments, plasmid transfections, immunoblotting, quantitative RT-PCR, mammosphere assays, soft agar assays, annexin V apoptosis assays, co-immunoprecipitations, chromatin immunoprecipitations and design of primers flanking putative YB-1 binding sites in the BMI1 promoter, preparation of samples for FACS analysis, optimization of acini culture conditions for H16N2 HMECs, optimization of HAT activity and telomerase assays, and animal monitoring), analyzed all data, and prepared the manuscript in its entirety. Contributions of co-authors: •  Kristen M. Reipas aided in the bilateral mammary fat pad injections. Alastair Davies also performed surgeries and was responsible for animal monitoring.  •  Sumaiya Islam performed the nanoString gene expression profiling, interpreted the data, and performed statistical analysis (Figures 3.6C and 3.6D). She also analyzed Affymetrix expression data from the public Gene Expression Omnibus (GEO) database (Table 3.1).  •  Abbas Fotovati, PhD, DVM, performed dissections of the mouse mammary glands and IHC staining.  •  Mary Rose Pambid preformed preliminary experiments characterizing the HTRZ and HTRY cell lines. This work was incorporated into Figures 3.1C and S3.1A.  •  Kaiji Hu, PhD, provided technical assistance with Cellomics screening.  •  Anna L. Stratford, PhD, constructed the pLenti6/V5-RSK2 expression vector.  •  Christopher Maxwell, PhD, provided technical assistance with the threedimensional basement membrane cultures.  •  Sandra E. Dunn, PhD, supervised the project and assisted with editing the manuscript.  Animal work was performed under approval by the UBC Animal Care Committee (certificate A11-0082).  iv  Chapter 4: YB-1 enhances ABCG2 expression in breast cancer to elicit chemoresistance that can be circumvented using RSK inhibitors.  Alastair Davies conceived the project, performed more than 90% of the experiments (including all immunoblotting, quantitative RT-PCR, soft agar assays, acini formation assays, annexin V apoptosis assays, chromatin immunoprecipitations and design of primers flanking the YB-1 binding site in the ABCG2 promoter, and optimization of drug treatments), analyzed all data, and prepared the manuscript in its entirety. Contributions of co-authors: •  Kaiji Hu, PhD, provided technical assistance with Cellomics screening.  •  Kristen M. Reipas assisted in the characterization of ABCG2 expression in the X43 cell line (Figure 4.5F)  •  Sandra E. Dunn, PhD, supervised the project and assisted with editing the manuscript.  v  TABLE OF CONTENTS ABSTRACT ....................................................................................................................... ii	
   PREFACE......................................................................................................................... iii	
   TABLE OF CONTENTS ................................................................................................ vi	
   LIST OF TABLES ........................................................................................................... ix	
   LIST OF FIGURES .......................................................................................................... x	
   LIST OF ABBREVIATIONS ....................................................................................... xiii	
   ACKNOWLEDGEMENTS ........................................................................................... xv	
   CHAPTER 1. INTRODUCTION .................................................................................... 1	
   The Fundamentals of Breast Cancer ................................................................ 1	
   The Hallmarks of Cancer ............................................................................ 1	
   Genomic Instability..................................................................................... 4	
   The Cell Cycle ............................................................................................ 7	
   The Centrosome and Mitotic Spindle ....................................................... 16	
   Epigenomics.............................................................................................. 20	
   The Genesis of Breast Cancer ........................................................................ 29	
   The Mammary Gland Hierarchy ............................................................... 31	
   The Cell of Origin ..................................................................................... 33	
   Premalignant Lesions ................................................................................ 36	
   Ductal Carcinoma in situ .......................................................................... 37	
   Invasive Ductal Carcinoma ....................................................................... 37	
   Molecular Classification of Breast Carcinoma ......................................... 38	
   Modeling Breast Cancer Progression........................................................ 40	
   Breast Cancer Therapies ................................................................................ 43	
   Current Therapies...................................................................................... 44	
   The Triple-Negative Paradox .................................................................... 46	
   Mechanisms of Resistance: ABC Transporters ........................................ 46	
   Intrinsic Drug Resistance: Tumour-Initiating Cells .................................. 50	
   YB-1: A Potential Regulator of Breast Cancer Progression .......................... 50	
   General Overview of YB-1 ....................................................................... 51	
   Tissue Profile of YB-1 Expression ........................................................... 53	
   Regulation of YB-1 Expression ................................................................ 53	
   YB-1 Function in Breast Cancer: Proliferation and Drug Resistance ...... 54	
   Insights into YB-1 Induction of Tumour-Initiating Cells ......................... 56	
   Clinical Correlations ................................................................................. 56	
   Hypothesis and Aims ..................................................................................... 57	
   vi  CHAPTER 2. YB-1 EVOKES SUSCEPTIBILITY TO CANCER THROUGH CYTOKINESIS FAILURE, MITOTIC DYSFUNCTION, AND HER2 AMPLIFICATION ......................................................................................................... 58	
   Overview ........................................................................................................ 58	
   Introduction .................................................................................................... 58	
   Results ............................................................................................................ 60	
   YB-1 alters the expression and activity of cell cycle proteins .................. 60	
   Cytokinesis failure primes cells for premalignant transformation ............ 64	
   Cell cycle checkpoint slippage potentiates centrosome amplification leading to aneuploidy ................................................................................ 66	
   Identification of YB-1 as a centrosomal protein ....................................... 70	
   Genomic instability arises during premalignancy to generate clones with strong tumourigenic potential ................................................................... 74	
   Discussion ...................................................................................................... 77	
   Experimental Procedures ............................................................................... 79	
   Supplementary Data ....................................................................................... 83	
   Supplementary figures .............................................................................. 83	
   Supplementary tables ................................................................................ 92	
   Overview ........................................................................................................ 95	
   Introduction .................................................................................................... 95	
   Results ............................................................................................................ 97	
   Ectopic YB-1 expression in HMECs is at physiological level ................. 97	
   HMECs acquire characteristics of stem/progenitor-like tumour-initiating cells following YB-1 expression............................................................... 98	
   Chromatin remodelling by p300 underlies the reprogramming of HMECs into TICs ................................................................................................. 100	
   Chromatin remodelling permits YB-1 to transcriptionally regulate BMI1 and enhance self-renewal capacity.......................................................... 104	
   YB-1 evokes luminal filling and invasion in a three-dimensional model of breast acini .............................................................................................. 108	
   Sustained upregulation of YB-1 leads to full transformation and tumourinitiation .................................................................................................. 110	
   YB-1 transforms HMECs into cells with a TNBC subtype .................... 115	
   Foci of histologically normal mammary epithelia overexpress YB-1 .... 117	
   Discussion .................................................................................................... 119	
    vii  3.5	
   Experimental Procedures ............................................................................. 121	
   Supplementary Data ..................................................................................... 125	
   Supplementary figures ............................................................................ 125	
   Supplementary tables .............................................................................. 136	
   CHAPTER 4. YB-1 ENHANCES ABCG2 EXPRESSION IN BREAST CANCER TO ELICIT CHEMORESISTANCE THAT CAN BE CIRCUMVENTED USING RSK INHIBITORS ....................................................................................................... 139	
   Overview ...................................................................................................... 139	
   Introduction .................................................................................................. 140	
   Results .......................................................................................................... 142	
   YB-1 is a direct transcriptional regulator of ABCG2 ............................. 142	
   The ABCG2 promoter is hyperacetylated in YB-1 expressing cells ...... 144	
   Inhibition of p300 HAT activity prevents YB-1 binding to the ABCG2 promoter .................................................................................................. 144	
   RSK inhibitors can reverse drug resistance and re-sensitize cells to ABCG2 substrates................................................................................... 147	
   RSK inhibition is sufficient to induce apoptosis in TNBC cell lines ..... 149	
   RSK inhibitors are effective in eliminating the TIC subpopulation ....... 152	
   High expression of ABCG2 is associated with clinical recurrence ........ 152	
   Discussion .................................................................................................... 154	
   Experimental Procedures ............................................................................. 156	
   Supplementary Data ..................................................................................... 159	
   Supplementary figures ............................................................................ 159	
   Supplementary tables .............................................................................. 161	
   CHAPTER 5. CONCLUDING REMARKS ............................................................... 162	
   Summary and Discussion ............................................................................. 162	
   Future Directions and Clinical Implications ................................................ 176	
   REFERENCES.............................................................................................................. 181	
    viii  LIST OF TABLES Table 1.1 Histone modifications and their effect on gene transcription. .......................... 23	
   Table 1.2 Molecular subtypes of breast carcinoma. ......................................................... 38	
   Table 1.3 Chemotherapy substrates of the ABCB1 and ABCG2 transporters. ................ 48	
   Table 2.1 Cell cycle associated proteins putatively regulated by YB-1. .......................... 62	
   Table S2.1 Proteins regulated by YB-1 based on Kinex™ microarray analysis. ............. 92	
   Table S2.2 List of antibodies, dilutions, and suppliers used for immunoblotting (IB) and immunofluorescence (IF). ......................................................................................... 93	
   Table S2.3 siRNA target sequences. ................................................................................. 94	
   Table 3.1 Genes correlated with YB-1 in a TNBC cohort. ............................................. 117	
   Table S3.1 List of antibodies, dilutions, and suppliers used for immunoblotting (IB), immunofluorescence (IF), and immunoprecipitation (IP). ..................................... 136	
   Table S3.2 Primers used for ChIP assays. ...................................................................... 137	
   Table S3.3 siRNA target sequences. ............................................................................... 138	
   Table S4.1 Primers used for ChIP assays. ...................................................................... 161	
    ix  LIST OF FIGURES Figure 1.1 Regulation of the cell cycle. .............................................................................. 9	
   Figure 1.2 Mechanisms leading to aneuploidy. ................................................................ 18	
   Figure 1.3 The breast cancer epigenome and relevant gene mutations. ........................... 21	
   Figure 1.4 Polycomb-mediated repression of the CDKN2A locus. .................................. 28	
   Figure 1.5 The stages of breast cancer progression. ......................................................... 30	
   Figure 1.6 Human mammary epithelial cell hierarchy and the respective cellular phenotypes. ............................................................................................................... 33	
   Figure 1.7 A cellular wiring diagram of the YB-1 signal transduction network. ............. 51	
   Figure 1.8 The structural domains of YB-1. ..................................................................... 52	
   Figure 2.1 YB-1 altered the activity of the centrosomal protein LIMK1/2. ..................... 63	
   Figure 2.2 Premalignancy was initiated by pLIMK1/2T508/T505 mislocalization leading to cytokinesis failure. .................................................................................................... 65	
   Figure 2.3 Centrosome amplification and aneuploidy emerged as a consequence of cytokinesis failure and cell cycle checkpoint slippage. ............................................ 68	
   Figure 2.4 pYB-1S102 localized to the centrosomes throughout the cell cycle. ................. 71	
   Figure 2.5 YB-1 altered the architecture and microtubule nucleation capacity of centrosomes by directly binding pericentrin and γ-tubulin. ..................................... 73	
   Figure 2.6 Numerical and structural chromosomal aberrations materialized as a consequence of YB-1 expression. ............................................................................. 75	
   Figure 2.7 Proposed model for how YB-1 instigates premalignancy. .............................. 76	
   Figure S2.1 Characterization of the HTRZ and HTRY cell lines. .................................... 83	
   Figure S2.2 Knockdown of RSK1/2 suppressed the phosphorylation of YB-1 and LIMK1/2. .................................................................................................................. 84	
   Figure S2.3 PLK1 localized to the cleavage furrow in HTRZ and HTRY cells. ............. 85	
   Figure S2.4 Amplification of the mother centriole was a consequence of YB-1 expression. ................................................................................................................ 86	
   Figure S2.5 Aneuploidy and centrosome amplification were dependent on YB-1 Ser-102 phosphorylation......................................................................................................... 87	
   x  Figure S2.6 Validation of YB-1 as a centrosomal protein. ............................................... 88	
   Figure S2.7 Inhibition of YB-1 Ser-102 phosphorylation promotes centrosome dysfunction................................................................................................................ 89	
   Figure S2.8 Cohesion defects were an underlying cause of chromosomal instability...... 90	
   Figure S2.9 HER2 amplification was absent in HTRZ cells but occurred in tetraploid HTRY cells. .............................................................................................................. 91	
   Figure 3.1 HMECs acquired TIC characteristics following ectopic YB-1 expression. .... 99	
   Figure 3.2 YB-1 enhanced p300 stability, nuclear localization, and HAT activity to reprogram HMECs into TICs.................................................................................. 102	
   Figure 3.3 YB-1 transcriptionally regulated BMI1 to enhance self-renewal capacity. .. 106	
   Figure 3.4 YB-1 drove luminal translocation and outgrowth in a three-dimensional model of breast acini. ......................................................................................................... 109	
   Figure 3.5 Synergism between YB-1, RSK2, and hTERT conferred complete transformation. ........................................................................................................ 114	
   Figure 3.6 HTRY-LT cells were molecularly classified as TNBC. ................................ 116	
   Figure 3.7 YB-1 staining was detected in focal areas of histologically normal tissue from breast cancer patients. ............................................................................................. 118	
   Figure S3.1 Characterization of ectopic YB-1 expression level in HTRY cells. ........... 125	
   Figure S3.2 Schematic diagrams of the BMI1, CD44, and CD49f promoters. .............. 126	
   Figure S3.3 Anacardic acid inhibited HAT activity. ...................................................... 127	
   Figure S3.4 YB-1 binds to the promoter of BMI1. ......................................................... 128	
   Figure S3.5 Upregulation of YB-1 decreased p16INK4a expression and nuclear localization. ................................................................................................................................. 129	
   Figure S3.6 TIC-associated gene expression, HAT activity, and mammosphere formation in the HTRY-LT cell lines. ..................................................................................... 130	
   Figure S3.7 HTRY-LT cells were sensitive to RSK inhibition. ..................................... 131	
   Figure S3.8 184hTERT cells acquired a malignant phenotype following stable YB-1 expression. .............................................................................................................. 133	
   Figure S3.9 HTRZ and HTRY-LT growth kinetics. ....................................................... 134	
    xi  Figure S3.10 YB-1 was expressed during breast cancer progression in vivo. ................ 135	
   Figure 4.1 YB-1 transcriptionally regulated ABCG2. .................................................... 143	
   Figure 4.2 ABCG2 transcription was dependent on promoter acetylation. .................... 146	
   Figure 4.3 RSK inhibitors reversed ABCG2-mediated multidrug resistance. ................ 148	
   Figure 4.4 Inhibiting RSK suppressed the growth of TNBC cell lines. ......................... 151	
   Figure 4.5 TICs were sensitive to RSK inhibition. ......................................................... 153	
   Figure S4.1 ABCG2 was highly expressed in HTRY-LT cells. ..................................... 159	
   Figure S4.2 HTRY-LT cell viability was reduced by luteolin........................................ 160	
   Figure 5.1 A model of YB-1-mediated tumourigenesis.................................................. 165	
    xii  LIST OF ABBREVIATIONS 5-FU AA ABC ac ADH ALDH APC/C ATM ATR BLBC BLG BMI1 BRCA CDK CFC CGH ChIP CIN CK CKI CMF CRS CSC CSD DAPI DCIS DFS DMSO DNMT DSB DXR EGFR EMT ER EV FACS FISH FTC HAT HDAC HMEC HMT hTERT HTRY  5-fluorouracil anacardic acid ATP-binding cassette acetylated atypical ductal hyperplasia aldehyde dehydrogenase anaphase-promoting complex/cyclosome ataxia telangiectasia mutated ATM and Rad3-related protein basal-like breast cancer beta-lactoglobulin B lymphoma Mo-MLV insertion region 1 homolog breast cancer susceptibility cyclin-dependent kinase change from control comparative genomic hybridization chromatin immunoprecipitation chromosomal instability cytokeratin cyclin kinase inhibitor cyclophosphamide, methotrexate, 5-fluorouracil cytoplasmic retention site cancer stem cell cold-shock domain 4’,6-diamidino-2-phenylindole ductal carcinoma in situ disease-free survival dimethyl sulfoxide DNA methyltransferase double-strand break doxorubicin epidermal growth factor receptor epithelial-mesenchymal transition estrogen receptor empty vector fluorescence-activated cell sorting fluorescence in situ hybridization fumitremorgin C histone acetyltransferase histone deacetylase human mammary epithelial cell histone methyltransferase human telomerase reverse transcriptase HMEC tetracycline-repressed YB-1  xiii  HTRZ IDC IP LIMK LY MAD MaSC MCM MDR MEF MMTV NLS NOD/SCID PcG PCG PCM PCNA PI3K PLK PR PRC PRE PTEN qRT-PCR RB RPPA rpS6 RSK RTK Scr siRNA SP TAC TDLU TGFβ TIC TMA TNBC TrxG ub uPA VEGF vHMEC WAP YB-1 YRE  HMEC tetracycline-repressed LacZ invasive ductal carcinoma immunoprecipitation LIM kinase LY294002 mitotic arrest deficient mammary stem cell minichromosome maintenance multidrug resistance mouse embryonic fibroblast mouse mammary tumour virus nuclear localization signal nonobese diabetic/severe combined immunodeficient Polycomb group primary constriction gap pericentriolar material proliferating cell nuclear antigen phosphatidyl-inositol 3-kinase polo-like kinase progesterone receptor Polycomb repressive complex Polycomb response element phosphatase and tensin homologue deleted on chromosome 10 quantitative real-time polymerase chain reaction retinoblastoma reverse phase protein microarray S6 ribosomal protein p90 ribosomal S6 kinase receptor tyrosine kinase scrambled peptide small interfering RNA side population paclitaxel, doxorubicin, cyclophosphamide terminal ductal-lobular unit transforming growth factor beta tumour-initiating cell tissue microarray triple-negative breast cancer (ER-/PR-/HER2-) trithorax group ubiquitinated urokinase-type plasminogen activator vascular endothelial growth factor variant human mammary epithelial cell whey acidic protein Y-box binding protein-1 YB-1 responsive element  xiv  ACKNOWLEDGEMENTS I would like to take this opportunity to thank the many individuals who have impacted me during this academic journey. First, Dr. Sandra Dunn, your passion for science is infectious and the past five years spent in your lab have been a period of tremendous personal growth. Your belief in me was always so much greater than I had in myself. I will always remember your remarkable ability to see the glass half full. You are, in every sense of the word, a mentor. I am also grateful to my supervisory committee Dr. Phil Hieter, Dr. Thibault Mayor, and Dr. Mike Gold. Thank-you for taking the time to provide helpful advice. It has been a privilege to work with you during my studies. In addition, Dr. Vincent Duronio and the Experimental Medicine Graduate Program have always been incredibly supportive. I would especially like to thank Cornelia Reichelsdorfer for all her hard work. Finally, I would like to acknowledge the Michael Smith Foundation for Health Research and Canadian Institutes of Health Research (Frederick Banting and Charles Best Doctoral Scholarship) for their gracious financial support throughout my degree. A large part of what made my graduate career so rewarding was all the “professional” individuals in the Dunn lab that I have been fortunate to work with and get to know personally. To Kristen, thanks for your friendship and invaluable expertise with the in vivo mouse experiments. To Mary, thanks for always being there to pick up the slack. Your help and support has been invaluable. To Sumaiya, only you could make statistics so entertaining. To Kaiji, thanks for scanning “just one more plate.” To Jessie, Karen, and Arezoo, thanks for all your support while you were in the lab and the great memories that continued even after you graduated. To Joanna, Cathy, Abbas, Rachel, and Nicole, thanks for your help with everything. It has been a pleasure getting to know you all. To Anna. You have become such an amazing friend over the years. Thanks for all the support, talks, and endless laughter. Finally, to my family. This would not have been possible without your unwavering love, support, and encouraging words.  xv  CHAPTER 1. INTRODUCTION  1.1 The Fundamentals of Breast Cancer The earliest account of cancer can be traced back to the Edwin Smith Papyrus, an ancient Egyptian medical text dating from 1600 B.C. (Breasted and Library, 1930). Despite considerable advancement in targeted therapies and screening programs, in the year 2012, nearly 186,000 individuals in Canada were diagnosed with cancer and 76,000 succumbed to the disease (CCS/SC, 2012). Lung, colorectal, and breast are the most commonly diagnosed cancers and account for a disproportionate share of mortality. In fact, approximately one out of every nine women will develop breast cancer at some time in their lives (Kelsey and Berkowitz, 1988). Cancer is not a single disease, but rather a collection of diseases all characterized by a deregulation of cellular proliferation, survival, differentiation, and migration. During the evolution of a cancer, cells undergo dynamic changes to their genome to enable oncogenes through dominant gain of function and delete tumour suppressor genes through repressive loss of function (Weinberg, 1995). This multi-step process is facilitated by aberrant tumour-stromal cell interactions and/or pre-disposing germline mutations. In some instances, cells metastasize through the vasculature to establish a secondary tumour in a foreign microenvironment (Van't Veer and Weigelt, 2003). Even with the immense diversity between individual cancers most, if not all, share a set of unifying features known as the “hallmarks of cancer” (Hanahan and Weinberg, 2000; Hanahan and Weinberg, 2011).  1.1.1 The Hallmarks of Cancer The fundamental hallmarks of cancer were proposed by Douglas Hanahan and Robert Weinberg in 2000 to act as a conceptual framework for rationalizing the complexities of neoplastic disease. They distilled the six essential cellular circuits that must be collectively deregulated to dictate malignant growth: (1) acquiring the capability for growth signal autonomy, (2) evasion of growth inhibitory signals, (3) resisting apoptotic  1  programmed cell death, (4) replicative immortality, (5) sustained angiogenesis, and (6) invasion and metastasis (Hanahan and Weinberg, 2000). In 2011, deregulation of metabolism and avoiding immune destruction were proposed as two additional hallmarks (Hanahan and Weinberg, 2011). The way in which these cellular hallmarks pertain to the genesis of breast cancer, in particular, is discussed below. Growth signal autonomy. Normal cells require external cues from their microenvironment. Mitogenic growth factors bind to cell surface receptors to initiate signaling cascades responsible for proliferation. Breast tumours frequently overexpress the epidermal growth factor (ErbB1/EGFR) and ErbB2/HER2 receptor tyrosine kinases (RTKs) (Slamon et al., 1987). As a result, these cells are hypersensitive to ambient levels of growth factor that would not normally trigger proliferation. Structural alteration of receptors can further enhance unrestrained growth by promoting ligand-independent signaling; for example, a truncated form of EGFR lacking much of its cytoplasmic domain is constitutively active (Segatto et al., 1988). Evasion of growth inhibitory signals. Within a normal tissue, anti-proliferative signals operate to maintain homeostasis. To overcome this barrier, the Retinoblastoma (RB) tumour suppressor gene is frequently mutated or deleted in breast cancer (Wang et al., 1993). The loss of RB liberates E2F transcription factors that coordinately regulate the progression from G1 to S phase of the cell cycle. In a meta-analysis of nearly 7000 tumour microarray experiments, transcriptional targets of E2F were disproportionately overrepresented in breast carcinoma underlining the importance of the loss of RB and unrestrained E2F activity in these tumours (Rhodes et al., 2005). Resisting apoptosis. Cancer cells have evolved to circumvent the normal apoptotic pathways responsible for cellular attrition. The p53 tumour suppressor gene plays a central role in inhibiting proliferation and, if necessary, eliminating abnormal cells by trans-activating cell cycle inhibitors, such as p21Waf1, and pro-apoptotic regulators, including Bax, PUMA, and NOXA (Gasco et al., 2002). Accordingly, loss-of-function mutations in p53 are detected in approximately 30% of breast cancers (Banerji et al., 2012). Apoptosis can also be triggered through engagement of the FAS death receptor  2  (CD95); however, it is often found to be down-regulated in breast cancer cell lines (Keane et al., 1996). Replicative immortality. The early work of Hayflick established that cells have a finite replicative potential (Hayflick and Moorhead, 1961). To support expansive tumour growth cancer cells must attain unlimited replicative potential. This is achieved through the activation of telomere maintenance mechanisms to stop the “mitotic clock” by preventing the erosion of telomeres, short 6 basepair repeats that protect the ends of chromosomes. Telomerase (TERT), the enzyme that maintains telomere length above a critical threshold to prevent senescence, can be detected in virtually all breast tumours (Hoos et al., 1998). Sustained angiogenesis. Oxygen and nutrients supplied by the vasculature are crucial for cell survival and function. Cancer cells must thus induce the formation of new blood vessels, a process termed angiogenesis, to support survival and expansion. Vascular endothelial growth factor (VEGF) is the primary stimulus of angiogenesis in breast tumours. Its levels are enhanced by over 7-fold in the primary tumour microenvironment relative to the surrounding normal tissue (Yoshiji et al., 1996). Moreover, the VEGF receptor, Flk-1/KDR, is highly expressed and activated in malignant mammary tissue (Kranz et al., 1999). Invasion and metastasis. During the progression of cancer, cells from the primary tumour invade through the basement membrane, disseminate into the vasculature, and form new colonies in distant tissues. Breast cancer cells preferentially metastasize to the regional lymph nodes in addition to the bone, lung, liver, and brain (Weigelt et al., 2005). These metastases are the cause of over 90% of breast cancer related deaths (Greenberg et al., 1996). To successfully metastasize, cancer cells must first degrade their surrounding extracellular matrix. Urokinase-type plasminogen activator (uPA), a serine protease that degrades matrix components, is detected at high levels in primary breast cancer and is associated with decreased disease-free survival (Foekens et al., 2000). Moreover, to fracture cell-cell adhesions and acquire a mesenchymal phenotype, breast cancer cells frequently downregulate E-cadherin (Oka et al., 1993).  3  Energy metabolism. To support uncontrolled cell proliferation cancer cells must make corresponding adjustments to energy metabolism. Cancer is fueled predominately through aerobic glycolysis, a state in which cells limit their energy metabolism to glycolysis even in the presence of oxygen (Warburg, 1956). In breast cancer, the phosphatidyl-inositol 3-kinase (PI3K)/mammalian target of rapamycin (mTOR) signaling cascade is frequently altered leading to the expression of pyruvate kinase M2 (PKM2), a key driver of aerobic glycolysis (Sun et al., 2011). Evading immune destruction. It has been well-documented in transgenic mouse models deficient for various components of the immune system that tumours grow more frequently and rapidly in immunodeficient mice relative to immunocompetent controls (Kim et al., 2007). This suggests that the immune system can contribute significantly to tumour eradication. To escape immune surveillance, breast cancer cells secrete the immunosuppressive factor transforming growth factor-β (TGFβ) to paralyze cytotoxic T lymphocytes and natural killer cells (Derynck et al., 2001).  1.1.2 Genomic Instability Acquisition of the hallmarks enumerated above depends largely on successive alterations to the genomes of neoplastic cells. In short, cancer follows the principals of Darwinian evolution. Repetitive cycles of mutation and selection ultimately yield a fully malignant phenotype. Mutations that increase a cell’s evolutionary fitness, aptly called driver mutations, enable clonal outgrowth and eventual dominance in the local tissue environment (Stratton et al., 2009). The incremental progression through increasingly malignant phenotypes over many years is reflected clinically; for example, breast cancer proceeds through a pre-invasive stage of ductal carcinoma in situ before breaching the basement membrane and metastasizing (Jones et al., 2008). At the beginning of the twentieth century, Boveri observed abnormal chromosome complements in tumour cells. It is now well established that genomic instability is a remarkably conspicuous feature of many human cancers (Lengauer et al., 1997). Most malignant cells exhibit a form of genomic instability called chromosomal instability (CIN), which refers to their high incidence of chromosome number and structure changes  4  compared to normal cells. Other less common types of genomic instability include microsatellite instability (MIN), characterized by the expansion or contraction of oligonucleotide repeats present in microsatellite sequences, as well as defective nucleotide excision repair and base excision repair (Lengauer et al., 1998). It is important to note that genomic instability is not necessary for tumourigenesis to occur, nor is it the principal driving force (Marx, 2002). Rather it is an enabling phenotype that arises though mutation in DNA repair pathways, telomere attrition, and/or deregulation of mitotic checkpoint control to give cells an advantage in terms of faster progression to malignancy. This is best exemplified by the finding that human epithelial cells can become transformed in the absence of excessive aneuploidy, chromosomal translocations, and MIN (Zimonjic et al., 2001). Conversely, H-rasV12-transformed human mammary epithelial cells (HMECs) fail to form tumours in mice despite widespread genomic instability (Dumont et al., 2009). A recent paradigm shift dictates that low-level genomic instability promotes tumour progression, but higher levels result in a so-called “mutational meltdown” causing cell death and tumour suppression (Weaver et al., 2007). Observations in cultured cells indicate that various DNA repair processes act to restore the normal nucleotide sequence following genotoxic stress with extraordinary precision. Accordingly, the rate of spontaneous mutation is very low at 2.5 x 10-8 per position per genome or, simply put, each cell accumulates one mutant gene during the lifespan of an individual (Roach et al., 2010). Taken at face value, it would appear as though cancer would be a rare disease; however, it appears at substantial frequency within the human population. To resolve this paradox it has been proposed that tumour cells exhibit a mutator phenotype, that is, they acquire increased mutability by inactivating “caretaker genes” that function in the maintenance of genomic stability (Loeb, 1991; Loeb, 2011). A second class of genes colloquially referred to as “gatekeepers” regulate cellular proliferation and are also frequent targets of mutation (Kinzler and Vogelstein, 1997). From a genetic perspective these genes behave much like tumour suppressor genes. Their loss of function accelerates the accumulation of subsequent mutations and greatly increases susceptibility to cancer.  5  The mutator hypothesis is best exemplified in hereditary breast cancers. Germline mutations in breast cancer susceptibility 1 (BRCA1) and BRCA2 predispose to the development of breast cancer. Carriers of BRCA1/2 mutations have a 50-80% lifetime risk of developing the disease (Foulkes, 2008). This is due to the essential role of BRCA1 and BRCA2 in the repair of DNA double-strand breaks (DSBs) by homologous recombination. Specifically, BRCA1-containing macro-complexes recognize DNA damage and recruit BRCA2 and RAD51 to initiate the repair (Roy et al., 2012). BRCA1 has also been proposed to mediate DSB repair via the non-homologous end-joining (NHEJ) pathway (Zhong et al., 2002) as well as activate the G1/S checkpoint by facilitating the phosphorylation of p53 on its Ser-15 residue by ataxia telangiectasia mutated (ATM)/ATM and Rad3-related protein (ATR) (Fabbro et al., 2004). Loss of fidelity in DNA repair pathways increases the rate of mutation leading to genomic instability. A second, albeit extremely rare, breast cancer susceptibility syndrome is LiFraumeni. This condition is linked to germline mutations in the TP53 tumour suppressor and DNA damage checkpoint gene. Individuals with Li-Fraumeni syndrome have a 25fold increased risk of developing breast cancer, as well as leukemia and brain cancer (Ginsburg et al., 2009). In the past few years several consortia have begun sequencing the genomes of human cancer cells providing unparalleled insight into the molecular basis of genomic instability in non-hereditary, sporadic breast cancer. In one study, Vogelstein and colleagues sequenced 18,191 genes from 11 primary breast cancers (Wood et al., 2007). Mutations were found in 1,137 genes; however, only five were found to be mutated, deleted, and/or amplified in more than 20% of the tumours analyzed. These included TP53 in addition to the classical oncogenes EGFR, H-ras, CDKN2A (encoding p16INK4a), and phosphatase and tensin homologue deleted on chromosome 10 (PTEN) (Sjoblom et al., 2006; Wood et al., 2007). A second independent study examining the genomes of 100 breast tumours found the same five genes to be mutated at significantly high frequency (Stephens et al., 2012). DNA sequencing has also identified recurrent somatic inactivating mutations in stromal antigen 2 (STAG2). This gene encodes a cohesion subunit that is required for sister chromatin cohesion and its inactivation yields aneuploidy (Solomon et al., 2011). Results  6  from these high-throughput, next-generation sequencing studies confirm the notion put forward by the mutator hypothesis, that is, the loss of caretaker and gatekeeper genes increases susceptibility to cancer and underlies disease progression. While the prevailing dogma of cancer evolution is one of gradualism, it has recently been proposed that a single catastrophic event, termed chromothripsis, can lead to massive chromosome breakage that is subsequently stitched back together by DNA repair machinery in a mosaic patchwork of genomic rearrangements (Stephens et al., 2011). It has been postulated that genome wide telomere attrition acts as a catalyst for this process (Janssen et al., 2011; Stephens et al., 2011). The resulting genomic rearrangements can advance cancer progression by promoting copy number changes, including deletion of tumour suppressor genes and amplification of oncogenes, or by juxtaposing the coding region of two genes to create an oncogenic fusion gene. In support of this model, sequencing of single cells from polygenomic breast tumours revealed three distinct subpopulations with few persistent intermediates (Navin et al., 2011). Notably, there existed significant overlap between the somatic genomic rearrangements in the primary tumours and those in the corresponding metastases suggesting that many rearrangements occurred in the primary cancers (Navin et al., 2011).  1.1.3 The Cell Cycle Genomic instability is potentiated by cell cycle deregulation. Unscheduled proliferation can increase susceptibility to the accumulation of genetic lesions. Moreover, if a cell fails to accurately and faithfully partition its genome the resulting daughter cells might inherit too many or too few chromosomes, a condition referred to as aneuploidy. This can lead to the emergence of aggressive phenotypes that facilitate tumour progression. Cell division consists two consecutive processes: Interphase, which is collectively comprised of a quiescent resting phase (G0), growth and synthesis of macromolecules (G1 phase), DNA replication (S phase), and a transitional gap (G2). This is followed by mitosis whereby the replicated chromosomes are segregated into two separate daughter cells. The basic regulation of the cell cycle has been well established by pioneering studies in yeast (Hartwell et al., 1974; Nurse et al., 1976). While the number of CDKs  7  and cyclins has increased considerably during evolution, the eukaryotic cell cycle is not conceptually too different from that of yeast. Progression through the cell cycle phases is regulated by changes in the activity of specific cyclin-dependent kinases (CDKs), with CDK2/CDK4/CDK6 and CDK1/CDK2 controlling the G1-S and G2-M transitions, respectively. These proteins generally remain at a constant level throughout the cell cycle, while their binding partners (cyclins) and post-translational modifiers undergo periodic fluctuations to regulate DNA synthesis and cell division (reviewed in (Malumbres and Barbacid, 2009)). Interestingly, a recent study in which the CDK loci were systematically knocked out of the mouse germline has shown that the interphase CDKs are not essential for cell cycle progression. Only elimination of the mitotic kinase CDK1 resulted in cell cycle arrest suggesting that compensatory activities exist among the CDKs (Santamaria et al., 2007). The concept that checkpoint pathways exist to send inhibitory signals in the event of an uncompleted cell cycle event was first proposed by Hartwell and Weinert (Hartwell and Weinert, 1989). It is now known that an intricate series of feedback loops integrate intracellular and extracellular signals to regulate three different transitions: G1-S, S-G2, and G2-M (Kastan and Bartek, 2004). By inhibiting the activity of CDK-cyclin complexes the checkpoint pathways can delay cell cycle progression. This is modulated though the ubiquitin-dependent degradation of cyclins, stoichiometric inhibition by cyclin kinase inhibitors (CKIs), and inhibitory phosphorylation. Figure 1.1 depicts the regulatory mechanisms that control the cell cycle. It is not surprising that the cell cycle machinery is deregulated in cancer. The mechanisms by which the cell cycle checkpoints are circumvented in breast cancer are discussed below. Emphasis has been placed on G1-S and mitotic checkpoint signaling due to their particular relevance to the main body of this thesis.  8  cdc20% APC/C%  cdc20%  BUBR1%  securin%  APC/C%  D% p53%  p21%  B%  GADD45% PLK1%  G0%  D%  CDK1%  cdc25C%  CDK1%  INK4%family% (p16)%  CDK6%  P% RB%  HDAC% E2F% G1%  A%  Growth' Factors'  CDK4%  RB%  E2F%  G2% E% A%  HDAC%  S% cdc25A%  CDK2%  Wee1%  p27%  CDK2%  ATM/ATR%  MDM2% DNA'Damage'  p21%  P% Chk2%  SCFSkp2%  p53% P% MDM2%  Figure 1.1 Regulation of the cell cycle. Progression through the mammalian cell cycle is governed by CDK/cyclin activity. Growth factors stimulate CDK4/cyclin D and CDK6/cyclin D to phosphorylate RB causing the release of E2F transcription factors. Cyclin E is a direct transcriptional target of E2F and complexes with CDK2 to hyperphosphorylate RB, which functions in a positive feedback loop to enhance the activity of CDK2/cyclin E complexes necessary for the G1-S transition (depicted in green). In response to DNA damage, ATM/ATR phosphorylates CHK2 and MDM2 to engage the G1-S and S-G2 checkpoints by degrading cdc25A and activating p21Waf1 (depicted in red). Entry into mitosis is dependent on CDK1/cyclin B activity that is enhanced by PLK1 and repressed by p53. The mitotic checkpoint ensures successful chromosome capture and alignment. The metaphase-anaphase transition requires BUBR1-dependent activation of the APC/C via displacement of CDC20. This results in the degradation of CDK1 and securin (depicted in blue).  9  G1-S Checkpoint During G1, cells proliferate in response to environmental stimuli and ensure their genome is ready to be replicated. Mitogenic signals activate the Ras/MAPK and Wnt/β-catenin signaling pathways to stimulate transcription of D-type cyclins, CDK4, and CDK6 by the effectors AP-1 and LEF-1 (Shtutman et al., 1999). In this way, the D-type cyclins directly integrate extracellular signaling with the cell-cycle machinery. Cyclin D1 is overexpressed in approximately 45% of primary breast cancers, thus uncoupling the cell cycle from exogenous growth factor signaling. Upregulation can be attributed to enhanced transcription as a consequence of constitutive RTK and Ras activity and in some cases gene amplification (Buckley et al., 1993). Transgenic mouse mammary tumour virus (MMTV)-cyclin D1 mice develop mammary adenocarcinomas, but this occurs with a long latency (~18 months) suggesting that cyclin D1 alone is a weak oncogene (Wang et al., 1994). Nevertheless, its expression is an absolute requirement for tumourigenesis as cyclin D1 knockout mice are protected against MMTV-neu and MMTV-ras-induced tumourigenesis (Yu et al., 2001). Cyclin D binds CDK4/6 and these complexes phosphorylate the pocket proteins (RB, p107, and p130) leading to their partial inactivation and the release of E2F transcription factors. This drives the expression of cyclin E1 and E2, both E2F target genes, to reach a maximum near the G1-S boundary. The E-type cyclins bind and activate CDK2, which further stimulates RB phosphorylation and provides a positive feedback loop for cyclin E transcription (Harbour et al., 1999). Many other E2F target genes are also essential in DNA replication and are required for progression into S-phase, including DNA polymerase-α, proliferating cell nuclear antigen (PCNA), and dihydrofolate reductase, an enzyme required for purine nucleotide biosynthesis (Bracken et al., 2004). Moreover, cmyc is an E2F target (Mudryj et al., 1990) that acts to further enhance cell proliferation by functioning as a transcription factor that regulates genes encoding proteins involved in the G1 transition, such as Cul1, CDK4, and the cdc25A phosphatase (Luscher, 2001). Cyclin E is frequently overexpressed in breast cancers and is particularly associated with the highly aggressive triple-negative subtype (Bostrom et al., 2009). This can be attributed to amplification of the cyclin E locus (Keyomarsi and Pardee, 1993) and/or  10  loss-of-function mutations in the gene encoding hCdc4, an F-box protein required for the ubiquitin-mediated turnover of cyclin E (Strohmaier et al., 2001). Constitutive expression of cyclin E in human breast epithelial cells has been associated with genomic instability (Spruck et al., 1999). This can be explained by two lines of evidence. First, cyclin E overexpression allows for an accelerated G1/S transition even in the presence of genomic abnormalities. Second, it impairs the loading of minichromosome maintenance (MCM) protein complexes onto DNA replication origins. The loss of fidelity in DNA replication can result in the accumulation of DNA lesions leading to chromosomal defects and aneuploidy (Geng et al., 2007). Interestingly, while transgenic beta-lactoglobulin (BLG)Cyclin E1 mice form hyperplastic lesions in the mammary gland, only 10% of the animals develop mammary carcinoma (Bortner and Rosenberg, 1997). This suggests that cyclin E is not a driver of the disease, but rather cooperates with other oncogenes to facilitate the escape of tumour cells from quiescence. Stimuli such as DNA damage, contact inhibition, replication senescence, and growth factor withdrawal yields the expression of the INK4 (p15, p16, p18, and p19) and Cip/Kip (p21, p27, and p57) family of CKIs to activate the G1-S checkpoint (Kastan and Bartek, 2004). For instance, in response to DNA damage ATM/ATR phosphorylates and inactivates the E3 ubiquitin ligase MDM2 to prevent the degradation of p53. The subsequent expression of p21Cip1, a direct p53 transcriptional, acts as a universal stoichiometric inhibitor of CDK/cyclin complexes. Moreover, ATM/ATR also activates the Chk2 protein kinase. When phosphorylated by Chk2, cdc25A, a phosphatase involved in activating CDK2 by opposing Wee1-mediated phosphorylation, is rapidly degraded to further enforce the G1-arrested state. Cancer cells exploit a variety of mechanisms to proliferate despite accumulating DNA damage. Apart from inactivating mutations in TP53, the F-box protein Skp2 is frequently overexpressed in estrogen receptor (ER)-negative breast cancers (Signoretti et al., 2002). This protein recognizes and targets the CKIs p21Cip1 (Bornstein et al., 2003) and p27Kip1 (Carrano et al., 1999) for degradation by the SCFSkp2 ubiquitin proteasome complex. In addition, sustained phosphorylation of p21Cip1 and p27Kip1 by the Akt oncogene impairs their nuclear import and ability to evoke a G1 arrest (Liang et al., 2002). The role of  11  p27Cip1 in protecting against mammary tumourigenesis is best exemplified by the finding that p27Cip1 haploinsufficiency enhances cell proliferation and, as a direct consequence, transgenic MMTV-neu/p27Cip1 heterozygous mice develop mammary tumours with a shorter latency compared to wild-type controls (Muraoka et al., 2002). One of the first CKIs lost during breast cancer progression is p16INK4a. This protein inhibits CDK4 and CDK6/cyclin D complexes in response to DNA damaging stimuli, including ionizing radiation, oxygen radicals, and telomere dysfunction. Under such conditions, ERK-mediated activation of Ets1/2 increases the transcription of p16INK4a (Kim and Sharpless, 2006). While post-translational regulatory mechanisms have not been well characterized, recent evidence suggests that miR-24 might play a key role in suppressing p16INK4a translation (Lal et al., 2008). The importance of p16INK4a in protecting against tumourigenesis was established by the finding that transgenic mice carrying an allele of the Ink4a/Arf/Ink4b locus exhibit a 3-fold reduction in the incidence of spontaneous cancers (Matheu et al., 2004). In further support, transgenic MMTV-p16INK4a mice are resistant to ErbB2-mediated transformation (Yang et al., 2004a). In human breast cancer, the frequency of inactivation of p16INK4a is second only to that of p53 (Geradts and Wilson, 1996). While this can partially be attributed to mutation (Prowse et al., 2003), epigenetic silencing of the INK4a locus is the predominate mode of repression (Romanov et al., 2001). This will be discussed further in section 1.1.5. S-G2 Checkpoint The intra-S phase and G2 checkpoints sense whether DNA has been faithfully replicated before cells enter mitosis. Both are controlled by the ATM/ATR signaling machinery in response to persistent DNA lesions. One of the effector mechanisms operates though the cdc25A-degradation cascade as previously described. Ultimately, the inhibition of CDK2 activity prevents replication origin firing by blocking the loading of CDC45 onto chromatin. This protein is required for the recruitment of DNA polymerase-α into assembled pre-replicative complexes (Bartek et al., 2004). ATM also phosphorylates NBS1 at Ser-343, a component of the Mre11/Rad50/NBS1 (MRN) complex that localizes in nuclear foci in response to DNA damage and links DSB recognition and repair, recruitment of ATM, and cell-cycle checkpoint signaling. Specifically, the S phase 12  checkpoint is activated in response to ATM/NBS1-mediated phosphorylation of SMC1 in a process dependent on its association with BRCA1 (Kitagawa et al., 2004). During the late stages of DNA replication, CDK1 and CDK2 bind the A-type cyclins to facilitate the transition from S phase to mitosis. Following nuclear envelope breakdown, the A-type cyclins are degraded releasing CDK1 which subsequently complexes with cyclin B to drive cells through mitosis (Malumbres and Barbacid, 2009). As such, the CDK1/cyclin B complex is the critical target of the G2 checkpoint. Its activation is restrained by ATM/ATR, CHK1/CHK2, p38-kinase-mediated subcellular sequestration, and/or inhibition of cdc25 phosphatases (Kastan and Bartek, 2004). Polo-like kinase-1 (PLK1) is often overexpressed in breast cancers where it functions as a key mediator of mitotic checkpoint inactivation though its regulation of cdc25C (Toyoshima-Morimoto et al., 2002). The maintenance of the G2 checkpoint relies largely on transcriptional programs controlled by BRCA1 and p53 leading to up-regulation of cell-cycle inhibitors such as p21Waf1, growth arrest and DNA-damage-inducible 45-α (GADD45α), and 14-3-3 proteins (Kastan and Bartek, 2004). Interestingly, tumour cells with defective p53 tend to selectively accumulate in G2 after DNA damage (Kastan et al., 1991), indicating that p53-independent mechanisms are sufficient to sustain G2 arrest. The Mitotic Checkpoint The mitotic checkpoint, also known as the spindle assembly checkpoint, is the primary safeguard against chromosome missegregation and aneuploidy. In its simplest form, this regulatory mechanism ensures the high fidelity of chromosome segregation by delaying the onset of anaphase until all chromosomes are property bi-oriented on the mitotic spindle (reviewed in (Musacchio and Salmon, 2007)). The mitotic checkpoint was initially recognized two decades ago in experiments that identified Saccharomyces cerevisiae mutants with the ability to bypass mitotic arrest in the presence of spindle poisons (Li and Murray, 1991). It was later established in mammals that this checkpoint is essential and complete inactivation yields massive chromosome missegregation, cell death, and early embryonic lethality (Kops et al., 2004).  13  The mitotic state is maintained by the high activity of CDK1/cyclin B1 complexes, which are only degraded following alignment of chromosomes on the metaphase plate. During prometaphase sister chromatids become topologically linked by ring-like cohesion complexes (Haering et al., 2008). Simultaneously, microtubules are nucleated from the centrosomes and attach at specialized proteinaceous structures on centromeric chromatin known as kinetochores. As cells proceed from prometaphase to metaphase, a signaling complex containing mitotic arrest deficient-1 (MAD1), MAD2, MPS1, BUB1, BUB3, and BUBR1 assembles at unoccupied kinetochores. This in turn leads to the generation of a diffusible inhibitory signal relayed by MAD1 and MAD2 that prevents the anaphasepromoting complex/cyclosome (APC/C) from exercising its E3 ubiquitin-ligase activity on cyclin B1 and securin (De Antoni et al., 2005). Securin is a small inhibitory chaperone of separase, the activity of which is necessary for the dissolution of cohesion complexes. As soon as the last kinetochore is attached to microtubules emanating from opposite spindle poles, the inhibitory signal is extinguished and the APC/C becomes fully activated following the displacement of its inhibitory cofactor, CDC20, by BUBR1 (Braunstein et al., 2007). The poly-ubiquitylation of securin and cyclin B1 results in their degradation by the 26S proteasome allowing for the separation of sister chromatids and execution of the metaphase-to-anaphase transition. Under normal circumstances, the mitotic checkpoint delays progression through the cell cycle in response to a single unattached kinetochore. However, if checkpoint signaling is weakened cells can initiate anaphase before all chromosomes have established proper bipolar spindle attachments. This can lead to chromosome missegregation and aneuploidy. It was originally hypothesized that mutations in mitotic checkpoint components were responsible for a weakened checkpoint. This was fueled by early reports documenting that haploinsufficiency of checkpoint genes in genetically engineered mice enhanced tumour development (Holland and Cleveland, 2009). For example, aneuploid tumours from BRCA2 knockout mice constantly exhibit mutations in BUB1 and introducing a dominant-negative mutant into mouse embryonic fibroblasts (MEFs) impairs their ability to undergo mitotic arrest (Lee et al., 1999). In human breast cancer, an extensive search uncovered inactivating mutations in MAD1L1 (encoding  14  MAD1) (Tsukasaki et al., 2001) and MAD2L1 (encoding MAD2) (Percy et al., 2000). However, these mutations were detected at very low frequency. Whereas inactivating mutations in mitotic checkpoint genes are rarely observed in breast cancer, abnormally high expression of their gene products is much more frequent. Forced overexpression of separase in diploid mouse mammary epithelial cells yields chromosome segregation errors and when transplanted into the mammary glands of mice these cells form aneuploid tumours (Zhang et al., 2008). MAD1, MAD2, BUB1, BUB3, and CDC20 were all found to be overexpressed in a panel of twelve breast cancer cell lines relative to normal MCF-10A mammary epithelial cells (Yuan et al., 2006). Notably, BUB3 is highly expressed in 80% of primary breast cancer tissues (Yuan et al., 2006) and MAD2L1 clusters with genes that have been correlated with poor disease outcome (van 't Veer et al., 2002). Finally, breast cancer specific gene-1 (BCSG1), which is expressed only in breast carcinoma cells, binds BUBR1 and causes its degradation leading to overactivation of the mitotic checkpoint (Gupta et al., 2003). Paradoxically, recent evidence implicates hyperactivation of the mitotic checkpoint in the transformation process. It has been proposed that aneuploidy is an early event in cancer evolution, which induces a quasi-stable karyotypic state that is balanced by selection towards increasingly malignant phenotypes (Nicholson and Duesberg, 2009). Prolonged activation of the mitotic checkpoint by overexpression of MAD2 in an inducible mouse model (Sotillo et al., 2007) or by using the spindle-poison nocodazole (Cimini et al., 1999) leads to lagging sister chromatids and merotelic attachments, an error where a single kinetochore attaches to microtubules that emanate from both poles of the spindle. As a result, both sister chromatids might be missegregated towards the same pole or be left in the spindle midzone and excluded from both nuclei. Eventually mitotic slippage occurs generating potentially tumourigenic aneuploid progeny. The precise mechanism of mitotic slippage remains unclear; however, it is thought to result from the degradation of cyclin B1 by the APC/C despite the activation of the mitotic checkpoint (Brito and Rieder, 2006). Interestingly, several mitotic checkpoint genes (MAD2L1, BUB3, CDC20, securin (PTTG), and cyclin B1 (CCNB1)) are direct E2F targets (Bracken et al.,  15  2004), indicating that loss of RB is not only associated with uncontrolled proliferation but also the generation of mitotic CIN.  1.1.4 The Centrosome and Mitotic Spindle As discussed, dysfunctional mitotic checkpoint signaling and defective chromosome cohesion generate aneuploidy, yet a final source arises from centrosome amplification and the assembly of multipolar mitotic spindles. The centrosome functions as the major microtubule organizing center (MTOC) responsible for nucleating and anchoring microtubules (Kellogg et al., 1994). As such, it is integral for progression through the cell cycle by determining the number, polarity, and organization of mitotic microtubules. Mammalian centrosomes consist of a pair of centrioles surrounded by a dynamic protein rich matrix known as pericentriolar material (PCM). Despite the importance of this organelle and more than 100 years of study, it was not until recently that proteomic analysis by mass-spectrometry provided insight into its composition (Andersen et al., 2003). Components include pericentrin, a protein that acts as a scaffold to anchor other centrosomal proteins; γ-tubulin, a unique member of the tubulin family that is responsible for the nucleation of microtubules; and centrin, which functions in centrosome duplication and separation. Once in each cell cycle, at the G1-S transition, the centrosome duplicates itself. Elegant studies using extracts prepared from Xenopus eggs uncovered a requirement for CDK2/cyclin E activity (Hinchcliffe et al., 1999; Lacey et al., 1999). Specifically, sustained expression of cyclin E was found to be sufficient for continued replication of centrosomes in S-phase arrested cells. This can be directly attributed to CDK2/cyclin Emediated phosphorylation of RB, as the overexpression of E2F alone was able to induce centrosome duplication (Meraldi et al., 1999). Genetic screens have implicated Aurora A and Mps1 kinase as the major downstream E2F effectors; however, the mechanism of action remains unclear (Winey and Byers, 1993). CDK2/cyclin E also directly phosphorylates nucleophosmin, an inhibitor of centriole duplication, causing it to dissociate from the PCM (Okuda et al., 2000).  16  As cells enter mitosis the polar separation of the duplicated centrosomes gives rise to a bipolar microtubule array. The mitotic spindle serves to segregate chromosomes to opposite poles of the cell and define the plane of cell division. This microtubule-based structure is comprised of dynamic αβ-tubulin heterodimers associated with a large variety of structural proteins (NUMA, TPX2, RHAMM), kinesin-like proteins (CENP-E), dynein motor proteins, and regulatory protein kinases (Aurora A, PLK1) (Nousiainen et al., 2006; Sauer et al., 2005). During metaphase, microtubules emanating from the centrosomes capture each chromosome at the kinetochore. TPX2 mediates Ran-GTP nucleation of microtubules (Gruss et al., 2002), which are then focused to the spindle midzone by NUMA and the BimC/Eg5-family of plus-end directed kinesins (Silk et al., 2009). Captured microtubules associate with the kinetochore-bound microtubule motor CENP-E to drive the poleward translocation of chromosomes (Brown et al., 1996). The integrity of the mitotic spindle is maintained by RHAMM, which crosslinks microtubules (Maxwell et al., 2003), and TPX2, which localizes Aurora A to the spindles poles where it activates the TACC-Msps complex to cap and stabilize microtubule minus-ends (Barros et al., 2005; Kufer et al., 2002). Extra centrosomes are found in over 90% of breast cancers and their presence correlates with chromosomal instability and aneuploidy (Lingle et al., 2002). It was originally proposed that cells with multiple centrosomes form a multipolar mitotic spindle to produce three of more highly aneuploid daughter cells (Nigg, 2002). However, the frequency of multipolar divisions is rare. Massive aneuploidy following multipolar division would compromise cell fitness and time-lapse imaging has revealed that the progeny of multipolar divisions are typically not viable (Ganem et al., 2009). Instead, high-resolution microscopy has exposed that cells cluster amplified centrosomes into two groups to form a pseudo-bipolar spindle. These spindles are associated with an increased frequency of merotelic attachments and lagging anaphase chromosomes (Ganem et al., 2009). Figure 1.2 depicts how a deregulated mitotic checkpoint and amplified centrosomes can yield aneuploidy.  17  Mito%c'checkpoint'  Mul%ple'centrosomes' Absent%MC% Cell%death%  Weak%MC%  Overac,ve% MC% Lagging% chromosome%  Premature% sister%% separa,on%  Furrow%regression%  2N% 2N=1%  2N+1%  Mul,polar%  Clustering% Cytokinesis%  2N% 4N%bipolar%  2N=1%  2N+1%  Figure 1.2 Mechanisms leading to aneuploidy. The normal mitotic checkpoint (MC) events from prometaphase to anaphase are depicted (blue box). An absent checkpoint results in mitotic cell death. A weak checkpoint leads to premature separation of sister chromatids and near-diploid aneuploidy. An overactive checkpoint yields lagging chromosomes and subsequent tetraploidy following furrow regression and re-entry into S phase or near-diploid aneuploidy in the event of an intervening cytokinesis. Multiple centrosomes can have similar consequences to a deregulated mitotic checkpoint (green box). A multipolar mitosis results in cell death unless centrosomes cluster, in which the probability of lagging chromosomes is high. In the first comprehensive analysis of centrosomes in human tumours, tissues from invasive breast adenocarcinomas were found to exhibit several centrosomal abnormalities including increased number, accumulation of excess PCM, supernumerary centrioles, and altered phosphorylation of centrin (Lingle et al., 1998). In addition, relative to normal breast tissue, tumour centrosomes had the unique ability to nucleate large microtubule arrays. This attribute was particularly pronounced in high-grade, undifferentiated tumours. A subsequent study demonstrated that centrosome amplification could be detected in most ductal carcinoma in situ (DCIS) lesions suggesting that it is an early event in breast tumourigenesis (Lingle et al., 2002).  18  It has recently emerged that members of the LIM kinase (LIMK) family, which includes LIMK1 and LIMK2, are important for centrosome and mitotic spindle assembly. This family of serine kinases was originally identified as regulators of actin cytoskeleton dynamics (Yang et al., 1998). More recently, the mitotic-specific activity of LIMK1 was found to be dependent upon phosphorylation of its Thr-508 residue by CDK1 (Sumi et al., 2002). Upon entry into mitosis LIMK1 is redistributed from cell-cell adhesion sites to the spindle poles (Chakrabarti et al., 2007) where it regulates microtubule disassembly by phosphorylating and inactivating tubulin polymerization-promoting protein (TPPP/p25) (Acevedo et al., 2007). This ability to oppose mitotic spindle assembly and arrest cells in mitosis may also explain why LIMK1 is sequestered to the nucleus by the cell cycle inhibitor p57Kip2 (Yokoo et al., 2003). As cells progress into telophase, LIMK1 accumulates at the cleavage furrow, an actin/myosin based structure. Together with large tumour suppressor-1 (LATS1) it controls cytokinesis by modulating actin polymerization and cytoskeletal reorganization (Yang et al., 2004b). Pioneering studies in prostate cancer have demonstrated that LIMK1 overexpression is associated with multiple centrosomes and abnormal mitotic spindles (Davila et al., 2007). Consequently, these cells exhibit delayed progression though mitosis and ultimately become multinucleated. Interestingly, the level and activity of LIMK1 is increased in invasive breast and prostate cancer cell lines suggesting that the protein might not only inflict karyotypic instability but also modulate tumour invasion (Yoshioka et al., 2003). This could be due, in part, to its induction of the serine protease urokinase-type plasminogen activator (uPA), which functions in extracellular matrix degradation (Bagheri-Yarmand et al., 2006). In tissue from prostate cancer patients, nuclear LIMK1 staining correlated with higher Gleason score, that is, the more aggressive and metastatic tumours (Davila et al., 2007). Likewise, in a breast cancer xenograft model forced expression of LIMK1 was associated with enhanced tumour growth. This could explain why LIMK1 was detected in over 75% of malignant breast tissues, but less than half of benign breast samples, on a tumour tissue microarray (TMA) (McConnell et al., 2011).  19  1.1.5 Epigenomics While initially underappreciated, recent work has established the importance of nonmutational, epigenetic mechanisms in deregulating gene expression during tumour progression (Baylin and Jones, 2011; Berdasco and Esteller, 2010; Sandoval and Esteller, 2012). Epigenetic gene regulation involves cooperation between many processes, including DNA methylation, covalent modifications of histones, and physical alterations in nucleosome positioning. In cancer, epigenetic changes can alter gene expression to give rise to a growth advantage and are selected for by the tumour cells in the same manner as mutations. An important distinction is that, unlike mutations, a single epigenetic event can function to deregulate multiple genes across a variety of pathways. Thus, epigenetic abnormalities result in a more nuanced, integrated disruption of signaling networks to foster tumour progression. Recently, evidence has emerged that genetics and epigenetics are not mutually exclusive events in cancer; they intertwine and take advantage of each other during tumourigenesis. Alterations in epigenetic mechanisms can cause the loss of function of genes that predispose the genetic mutation. For example, promoter hypermethylation is the predominant mechanism for the loss of O6-methylguanine-DNA methyltransferase (MGMT), which encodes a DNA repair gene, cyclin-dependent kinase inhibitor 2B (CDKN2B), which encodes the cell cycle inhibitor p15INK4b, and RASSF1A, which encodes an effector of the Ras oncoprotein (Baylin and Jones, 2011). On the other hand, whole exome gene sequencing of thousands of human cancers has identified many inactivating mutations in genes that control the epigenome (Barretina et al., 2012). The resulting epigenetic alterations can lead to abnormal gene expression and genomic instability, which may predispose to cancer. Figure 1.3 depicts how epigenetics regulate gene expression and the common mutations that disrupt these processes in breast cancer.  20  Nucleosome( remodeling( •  BRG1•  BAF250a•  BAF180--  DNA( methyla.on( •  DNMT1•  MBD1•  TET1-  Histone( modifica.ons( •  EZH2/Polycomb•  G9a•  p300•  HDAC4--  X( Gene-ACompac;on-of-chroma;n-  Gene-BLoosening-of-chroma;n-  Figure 1.3 The breast cancer epigenome and relevant gene mutations. Epigenetic gene regulation involves synergism between DNA methylation, nucleosome remodeling, and histone modifications. Silenced genes (for example, ‘Gene A’) are associated with gains in DNA methylation (indicated by red circles) in promoter region CpG islands, nucleosomes positioned over the transcriptional start site, and repressive histone modifications (indicated by red flags). Conversely, active genes (for example, ‘Gene B’) are hypomethylated (indicated by green circles), have an open nucleosome configuration, and acetylated histone lysines (indicated by green flags). Mutations and/or altered transcription of genes involved in modulating the epigenome have been identified in breast cancer (indicated in grey boxes) and are reviewed in (You and Jones, 2012).  The role of DNA methylation, covalent histone modifications, and nucleosome remodeling as they pertain to breast cancer are reviewed below. While discussed as individual entities it is important to consider the interrelationships between these processes. For example, DNA methylation and histone acetylation are linked so that the extent of global methylation alters the level of histone acetylation and vise versa (Nan et al., 1998). DNA methylation DNA methylation occurs predominately at CpG dinucleotides, and methylation of CpG island promoters acts as a relatively stable gene silencing mechanism (Jones and Liang, 2009). This process is regulated by a family of DNA methyltransferases (DNMTs) that 21  catalyze the transfer of methyl groups from S-adenosyl-L-methionine to the 5’ position of cytosine bases in the CpG dinucleotide. DNMT3A and DNMT3B establish DNA methylation patterns early in development, which are then maintained primarily by DNMT1 (Kinney and Pradhan, 2011). As supported by a recent structural study DNMT1 prefers hemi-methylated DNA over nonmethylated DNA as a substrate and therefore preserves methylation patterns during cell division (Song et al., 2012). While CpG islands remain relatively unmethylated in somatic cells, in cancer large stretches of DNA become abnormally methylated. This can be largely attributed to constitutive expression of DNMT1 leading to ectopic hypermethylation. Mutations in the destruction domain of DNMT1 preventing its ubiquitylation have been described in breast cancer (Agoston et al., 2005). Furthermore, gain-of-function mutations in the methylbinding domain (MBD) protein MBD2, which binds methylated CpG moieties to mediate transcriptional repression, are frequently detected in breast cancer (Sansom et al., 2007). While DNMTs have been demonstrated to be oncogenic, the enzymes that function to protect cells from becoming malignant by acting as erasers of DNA methylation are repressed. Ten eleven translocation-1 (TET1), an enzyme involved in cytosine demethylation, is downregulated in breast and prostate cancer tissue. TET1 depletion is associated with enhanced cell invasion and tumour growth in xenograft models and is correlated with poor survival rates in breast cancer patients (Hsu et al., 2012). Histone modification Nucleosomes, which are the basic building blocks of chromatin, consist of DNA wrapped around four core histones (H2A, H2B, H3, and H4) (Luger et al., 1997). Chromatin exists in two main states: less condensed and transcriptionally active euchromatin, and highly condensed and typically silent heterochromatin. Covalent modification of histone tails regulates the dynamics between these states by altering electrostatic charge and providing protein recognition sites for chromatin remodeling complexes. Histone modifications at specific lysine residues characterize genomic regulatory regions, such as active promoters which are enriched in acetylated histone H3 at lysine 9 (H3K9ac) and inactive promoters which are enriched in trimethylated histone H3 at lysine 27 (H3K27me3) (Hon et al.,  22  2009). The histone modification patterns are regulated by enzymes including histone acetyltransferases (HATs) that acetylate histone tails and deacetylases (HDACs) to remove these marks. Similarly, histone methyltransferases (HMTs) and demethylases introduce and remove methyl marks. Table 1.1 outlines common histone modifications, the associated modifying enzymes, and proposed effect on gene transcription.  Histone  Type of modification  H2AK119  H3K4  H3K9 ✔ p300  Ac ✔ MLL  Me1  H3K14  H4K20  ✔ PCAF  ✖ G9a  ✖ G9a  ✖ Pr-Set7  ✖ G9a  Me2 ✔ Set7/9  Me3  ✖ Clr4  ✖ Ezh2  ✖ Suv4-20h  ✔ Ser10 by RSK2/MSK1  Phospho Ub  H3K27  ✖ Bmi1  Table 1.1 Histone modifications and their effect on gene transcription. Specific enzymes (listed) can add modifications to histone tails to activate (✔) or repress (✖) gene transcription. Ac, acetylation; Me1, mono-methylation; Me2, di-methylation; Me3, tri-methylation; Phospho, phosphorylation; Ub, ubiquitylation. Reviewed in (Kouzarides, 2007) and (Peterson and Laniel, 2004).  Acetylation Acetylation of lysine residues on histones is almost invariably associated with active gene transcription. The neutralization of the lysine’s positive charge weakens the affinity between histones and DNA facilitating access to gene promoter regions for RNA polymerase and transcription factors. HATs are divided into three categories based on sequence similarities: Gcn5/PCAF, p300/CBP, and the MYST family (Yang, 2004). The enzymes themselves display virtually no substrate specificity; however, co-factors target the multiprotein acetylase complexes to particular lysine groups. For example, p300 preferentially acetylates lysine 9 and 18 of histone H3 and lysine 5 of histone H4, while  23  PCAF primarily acetylates lysine 14 of histone H3 (Schiltz et al., 1999; Szerlong et al., 2010; Zhong and Jin, 2009). The HAT protein p300 was initially identified in a protein interaction assay through its binding to the adenoviral-transforming protein E1A (Eckner et al., 1994). It has the unique ability to not only acetylate histone tails and relax chromatin, but also act as a scaffold for the transcription machinery via three cytosine-histidine (CH)-rich zinc finger motifs. Accordingly, p300 is indispensible for the activity of many transcription factors, the most important with regard to breast cancer being c-Jun, E2F, and p53. It was first observed in mice that EP300-/- embryos were considerably smaller than their wild-type littermates (Yao et al., 1998). This phenotype can be explained by the functional requirement for p300 in cell cycle progression. Specifically, it enhances cell proliferation by increasing the transcriptional activity of E2F toward its target genes. This is mediated though a p300-E2F complex that is stabilized following phosphorylation of the E2F transactivation domain by cyclin E/CDK2 (Morris et al., 2000). In cancer, cell proliferation could become unrestrained by the ability of p300 to stimulate MDM2mediated degradation of p53 (Grossman et al., 1998). This may explain why chimeric p300 fusion proteins are common in hematological malignancies (Ida et al., 1997). Mutations in p300, in the form of insertions and deletions, have been detected in both breast cancer cell lines and primary tissue. These mutations are largely clustered in the catalytic HAT domain and KIX domain, which mediates interaction with p53 (Gayther et al., 2000). While the role of p300 in breast cancer has not been well established, individuals with heterozygous germline mutations in EP300 or CREBBP, the gene encoding the p300 family member CREB-binding protein (CBP), are at increased risk of developing the disease (Roelfsema et al., 2005). It has been proposed that p300 is a positive regulator of breast tumourigenesis based on its requirement for the transcription of HIF-1 downstream genes (Vleugel et al., 2006), induction of VEGF (Arany et al., 1996), and opposition of BRCA1 (Fan et al., 2002). In support, p300 has recently been reported to be strongly expressed in malignant breast tissue relative to the surrounding normal epithelia (Fermento et al., 2010).  24  Methylation and Polycomb repressive complexes A second important histone modification is methylation, which is predominately regulated by the Polycomb group (PcG) and trithorax group (TrxG) class of proteins. PcG complexes include the SET-domain containing HMT zeste homologue 2 (EZH2), which methylates histone H3 at lysine 27 to initiate transcriptional repression. While not as extensively characterized, TrxG complexes also contain an HMT, mixed lineage leukemia (MLL), which methylates histone H3 at lysine 4 to establish active chromatin marks to reverse PcG-mediated repression (Mills, 2010). The relative distribution of PcG and TrxG complexes creates a “molecular tug of war” that determines the extent of gene expression. The PcG complexes are discussed in detail as they pertain to this thesis. There are two core PcG complexes each comprised of multiple subunits that work collaboratively to repress transcription. Polycomb repressive complex 2 (PRC2) recognizes Polycomb response elements (PREs) in gene promoters and establishes the histone code. Its activity is dependent on three core subunits: EZH2, embryonic ectoderm development (EED), and suppressor of zeste homologue 12 (SUZ12). EZH2 is a HMT that trimethylates histone H3 at lysine 27, a mark of repressed chromatin. EED and SUZ12 are required for the HMT activity of EZH2. Methylated H3K27 provides a binding site for PRC1. This complex contains chromobox homologue 2 (CX2), which recognizes and binds the H3K27me3 mark left by PRC2, ring finger protein 1 (RING1), which serves as an E3 ubiquitin ligase for mono-ubiquitylation of H2AK119, and BMI1, which stimulates the E3 ligase activity to dictate the extent of repression (Mills, 2010). It is believed that H2A ubiquitylation interferes with RNA polymerase loading and stability, thereby preventing overt gene expression at PcG-target loci (Stock et al., 2007). The PcG repressors were initially identified as being crucial for maintaining the inactive state of the Hox loci in developing Drosophila melanogaster embryos (Lewis, 1978). More recently, a genome wide chromatin immunoprecipitation study mapping the binding sites of PcG components in humans revealed that many of the targets encode transcription regulators that are inactive in pluripotent cells, but become expressed during differentiation (Bernstein et al., 2006). Loss-of-function studies in embryonic stem (ES) cells resulted in upregulation of tissue-specific genes, unequivocally establishing the  25  necessity of PcG proteins in maintaining ES cell pluripotency (Azuara et al., 2006; Jorgensen et al., 2006). However, it remains unclear how PcG proteins are recruited to specific sites. Emerging evidence suggests that the stem cell specific DNA binding factors Oct4, Sox2, and nanog co-occupy a subset of PcG target genes and may recruit PcGs to repress differentiation-promoting genes, although, so far, no direct interactions have been reported (Lee et al., 2006). This hypothesis is supported by data demonstrating that knock-down of Oct4 results in the loss of the PRC2-component SUZ12 from specific target promoters (Squazzo et al., 2006). In addition, long non-coding RNAs (ncRNAs), such as HOTAIR, have been shown to recruit PcG proteins to chromatin by acting as a bridge between genomic DNA and SET domains and/or chromodomains that are present in the PRC2 complex (Khalil et al., 2009; Rinn et al., 2007). Illustrative of the dynamism between epigenetic mechanisms, PRC complexes can recruit DNMTs to methylate DNA and stably repress gene expression. The HDAC sirtuin 1 (SIRT1) has been purified in these complexes (Kuzmichev et al., 2005). Interestingly, SIRT1 overexpression was reported in a comprehensive RT-PCR-based expression study of 27 primary breast cancers and 18 breast cancer cell lines (Ozdag et al., 2006). The altered composition of PRC complexes in tumours could have a major effect on changing patterns of gene expression. Due to their strong association with primitive, undifferentiated stem cells it is not surprising that Polycomb-group proteins are commonly deregulated in many cancers. Overexpression of EZH2 has been reported in several cancers, notably breast, prostate, lung, and bladder where it correlates with an increase in histone H3 lysine 27 trimethylation (Chase and Cross, 2011). On a tissue microarray containing 280 breast cancers, EZH2 was found to be elevated in invasive breast cancer relative to normal mammary epithelia (Kleer et al., 2003). In a later study, increased expression was also found in early ductal carcinoma in situ lesions (Ding et al., 2006). BMI1, an essential component of the PRC1 complex, was first identified as an oncogene that cooperates with MYC in lymphomagenesis (Haupt et al., 1991) and subsequent studies have found it to be robustly associated with many cancers. In contrast to prostate  26  and brain cancers, the importance of BMI1 in breast cancer has not been well established. However, it has been reported that BMI1 is expressed in primary breast cancer tissues and correlates with an unfavorable overall survival (Guo et al., 2011). Interestingly, BMI1 is expressed highest in relapsed breast cancer metastases (Joensuu et al., 2011). BMI1 facilitates tumourigenesis by mediating escape from cellular senescence, a state in which cells lose the ability to divide and thus undergo apoptosis. Mechanistically, BMI1 coordinately represses the CDKN2A (which encodes both p16INK4a and p14ARF) and CDKN2B (which encodes p15INK4b) loci by mediating the loading of CDC6 onto a cisacting replication origin upstream of the transcriptional start site. CDC6 recruits histone deacetylases and facilitates heterochromatinization in the promoter (Gonzalez et al., 2006). The repression of p16INK4a and p15INK4b allows for unrestrained CDK4-dependent phosphorylation of RB and, accordingly, E2F-mediated cell cycle progression. Moreover, the silencing of p14ARF, an inhibitor of MDM2-mediated ubiquitylation and degradation of p53, prevents apoptosis. In Bmi1-/- mice the self-renewal capacity of both neural and hematopoietic stem cells is dramatically compromised. These mice also exhibit increased apoptosis in lymphoid tissues. The phenotypes could be partially rescued by reintroducing p16INK4a and p14ARF, underscoring the importance of BMI1-mediated repression of these senescence and apoptosis inducers. Figure 1.4 depicts how PcG complexes affect transcription, with emphasis on the CDKN2A locus regulated by BMI1.  27  PRC1" PRC2" SUZ12& EED& EZH2& H3K27&  BMI1& RING1& CX2&  H3K27me3& H3K27me3& H2AK119ub& X"  p16INK4A&  D& CDK4/6&  Cell"cycle" arrest"  CDKN2A& CDKN2A&locus&  p14ARF&  MDM2&  p53&  Apoptosis"  Figure 1.4 Polycomb-mediated repression of the CDKN2A locus. In PcG-mediated gene repression, the PRC2 complex is recruited to PREs where it trimethylates histone H3 at lysine 27. The SET domain of EZH2 confers this activity. The H3K27me3 mark is recognized by the chromodomain of CX2 in the PRC1 complex. RING1 is an E3 ubiquitin ligase for histone H2A regulated by BMI1. Ubiquitylation represses gene transcription. Inhibition of p16INK4a and p14ARF, two genes encoded by the CDKN2A locus, facilitates cell proliferation and immunity from apoptosis.  Phosphorylation Relative to acetylation and methylation very little is known about histone phosphorylation. Stress-activated protein kinase-2 (MSK2) and p90 ribosomal S6 kinase2 (RSK2) have been shown to phosphorylate histone H3 at Ser-10 (Sassone-Corsi et al., 1999). This destabilizes higher-order compaction of the chromatin fiber to promote gene expression. H3S10 is specifically detected on nucleosomes at the promoter and/or within the body of activated immediate-early genes (Mahadevan et al., 1991). These genes are rapidly transcriptionally activated upon stimulation of the Ras-MAPK signaling pathway and include c-myc, c-fos, and uPA. This suggests that histone phosphorylation promotes rapid, but transient gene expression. Nucleosome Remodeling The term nucleosome remodeling subsumes a large number of ATP-dependent changes to nucleosome structure and positioning brought about by large, multifactorial chromatin  28  remodeling complexes. Well the fluidity of chromatin is necessary for many biological processes, such as DNA repair, it also functions to alter the accessibility of DNA sequence elements to regulatory proteins. ATP dependent chromatin remodelers are generally divided into four main families, the most important with regard to breast cancer being switch/sucrose non-fermenting (SWI/SNF). This complex remodels chromatin by changing nucleosome occupancy patterns, thereby contributing to either transcriptional activation or repression (Wilson and Roberts, 2011). Smarca4+/- (which encodes BRG1, the ATPase subunit of SWI/SNF) mice develop mammary tumours that are genetically unstable (Bultman et al., 2008). This could be related to the interaction between BRG1 and BRCA1 that is required to direct p53-mediated transcription of CKIs to elicit cell cycle arrest upon DNA damage (Bochar et al., 2000). BAF250a, another SWI/SNF subunit with a role in cell cycle repression, was reported to be mutated in breast cancers, specifically in the aggressive triple-negative subtype (Mamo et al., 2012). Finally, the BAF180 subunit is mutated in over 40% of breast cancers and this affects the ability of the SWI/SNF chromatin remodeling enzymes to regulate p53 transcriptional activity toward its target genes required for senescence, such as p21Waf1 (Xia et al., 2008).  1.2 The Genesis of Breast Cancer The human mammary gland is a branching ductal system that ends in terminal ducts with their associated acinar structures, termed terminal ductal-lobular units (TDLUs). Two general lineages of concentrically arranged epithelial cells constitute the dual-layered architecture. Luminal secretory cells border a hollow lumen, and in turn are surrounded by an outer layer of contractile myoepithelial cells that lie in direct contact with the basement membrane. In addition, there is emerging evidence that stem cells exist within the TDLU. The supportive stroma is comprised of large adipocytes interspersed with stromal fibroblasts (Smalley and Ashworth, 2003). The vast majority of breast cancers evolve from epithelial cells within TDLUs. Distinct pathological stages of the disease have been defined reflecting the multi-step nature of transformation. It is widely held that breast cancer initiates as the pre-malignant stage of  29  atypical ductal hyperplasia (ADH), progresses into the preinvasive stage of ductal carcinoma in situ (DCIS), and culminates in the potentially deadly stage of invasive ductal carcinoma (IDC) (Allred et al., 2001). ADH is characterized by increased numbers of abnormal cells within a duct or lobule; however, the lumen of these glands remains relatively hollow. The transition into a DCIS occurs as cells gain the ability to proliferate uncontrollably within the lumen resulting in distention of the acini. Once these malignant cells progress to invade through the basement membrane and into the surrounding stroma the tumour becomes classified as an IDC (Debnath and Brugge, 2005; Smalley and Ashworth, 2003). Figure 1.5 depicts the linear model of breast cancer progression and the genetic and epigenetic alterations characteristic of each stage. Mammary&Gland&  Pre-malignant&TDLU&  Ductal&carcinoma&in#situ#  Invasive&ductal&carcinoma&  !"COX!2"overexpressed" !"p16INK4A"silenced" "(epigene<c)"" "  !"HER2"overexpressed" !"p53,"RB"lost" !"Gain"of"1p" "  !"HIF!1α"overexpressed" !"E!cadherin"lost" " Genomic"instability"  Figure 1.5 The stages of breast cancer progression. The mammary gland is a branching network of ducts and lobuloalveolar structures (depicted in detail) comprised of luminal (red) and myoepithelial (purple) cells surrounded by a basement membrane (black). The TDLU is where most breast cancers initiate. It is hypothesized that a small population of pre-malignant cells reside in the normal mammary gland (shown in light blue). These cells, which are the precursors to cancer, are characterized by p16INK4a promoter hypermethylation and COX-2 expression. The acquisition of genetic lesions drives disease progression into a DCIS by promoting proliferation and resistance to apoptosis, for example. These genomically unstable cells (shown in blue) invade into the lumen of the mammary alveolus and acquire additional mutations to become increasingly malignant (shown in dark blue). Eventually, the disease becomes invasive when transformed epithelial cells breach the basement membrane, invade the surrounding stroma, and metastasize throughout the body.  30  1.2.1 The Mammary Gland Hierarchy Breast cancer is an extremely heterogeneous disease at both the histological and molecular level. Variation within a tumour occurs with respect to morphology, proliferation rate, cell surface marker expression, metastatic proclivity, and tumour regeneration capacity (Heppner, 1984). For example, studies of primary breast cancers revealed that the frequency of putative tumour-propagating cells between individual tumours varies considerably from 3 to 35% (Al-Hajj et al., 2003; Ginestier et al., 2007; Shipitsin et al., 2007). Understanding the differentiation hierarchy in the normal mammary gland is an important prerequisite to building a conceptual framework that explains breast cancer heterogeneity. Analogous to the paradigm established in the hematopoietic compartment, evidence suggests that stem cells are important for both mammary gland development and maintaining tissue homeostasis (Visvader, 2009). These stem cells undergo selfrenewal, the process by which they generate progeny identical to themselves. Stem cells can also differentiate into multipotent progenitors, which in turn yield committed progenitors and finally differentiated cells. The existence of stem cells in the mammary gland has long been postulated since this organ undergoes profound expansion during puberty and pregnancy. Historical and rather extraordinary studies by DeOme provided the first evidence of mammary stem cells (MaSCs) by transplanting mammary explants into a cleared mouse mammary fat pad and generating a fully functional mammary gland (Deome et al., 1959). Since then, accumulating evidence has further supported that existence of MaSCs, which give rise to mature epithelium of either the luminal or myoepithelial lineage via a series of lineagerestricted intermediates (Dontu et al., 2003; Stingl et al., 2006b; Visvader, 2009). A major focus of mammary gland biology has been to decipher the identity of mammary stem cells. Much enthusiasm was generated by the discovery that in mice a single cell expressing the surface marker CD24 in combination with CD29 or CD49f (also known as integrin alpha 6) could regenerate an entire mammary epithelial tree (Shackleton et al., 2006; Stingl et al., 2006a). Only recently; however, has a transplantation assay been  31  developed making it possible to explore the regenerative potential of epithelial cells in human mammary tissue. The “humanization” of the mammary fat pads in NOD/SCID mice through injection of immortalized human fibroblasts creates a stromal environment reminiscent of human breast tissue (Kuperwasser et al., 2004). Using this assay, it was demonstrated that human mammary epithelial cells with high aldehyde dehydrogenase-1 (ALDH1) activity have stem/progenitor properties (Ginestier et al., 2007). Curiously, while ALDH+ cells exhibited self-renewal capacity their differentiation was confined to the luminal lineage suggesting that these cells are not true MaSCs, but rather luminalrestricted progenitors (Eirew et al., 2012). With a refined marker selection, a subset of human breast cells was isolated and shown to have mammary regenerative capacity in vivo (Eirew et al., 2008; Lim et al., 2009). These cells, defined by high expression of CD49f and negligible expression of EpCAM (CD49fhiEpCAM-), are the most primitive thus far identified and represent a compelling candidate for the human MaSC. Through in vitro serial passaging, luminal- and myoepithelial-restricted progenitors were shown to exist downstream from MaSC/bipotent progenitors (Stingl et al., 2001). The luminal-restricted progenitors are associated with high levels of both CD49f and EpCAM. These cells also express MUC1 (Stingl et al., 1998) and KIT (Lim et al., 2009), whereas MaSCs/bipotent progenitors do not. In contrast, the unique cell surface marker expression of myoepithelial-restricted progenitor cells has yet to be elucidated. Fully differentiated luminal and myoepithelial cells are defined by expression of cytokeratin-8 (CK8)/CK18 and smooth muscle actin (SMA)/CK5/CK14, respectively (Perou et al., 2000). Figure 1.6 illustrates the epithelial cell hierarchy and the respective cellular phenotypes present in the human mammary gland.  32  Luminal2restricted% progenitor%  Luminal%cell% CK8+'CK18+'  CD49f+'EpCAM+' MUC,1+'KIT+'ALDH+' MaSC%  CD49f+'EpCAM,'  Bipotent%% progenitor% CD49f+'EpCAM,' MUC,1,'KIT,' Myoepithelial2restricted% progenitor%  Myoepithelial%cell%  CK5+'CK14+'ALDH,'  CK5+'CK14+'SMA+'  Figure 1.6 Human mammary epithelial cell hierarchy and the respective cellular phenotypes. A differentiation hierarchy exists within the human mammary gland. A stem cell sits at the apex and gives rise to a bipotent progenitor that further differentiates through lineage restricted-progenitors to give rise to a mature epithelium of luminal and myoepithelial cells. Primary cell surface markers used in the isolation of each population are shown.  1.2.2 The Cell of Origin The normal cell that acquires the first tumour-promoting mutation(s) is referred to as the cell of origin. It has been suggested that heterogeneity among human breast cancers is ascribed to their derivation from distinct epithelial cell types in the mammary gland hierarchy (Ince et al., 2007). While there has been some controversy about the nature of the cells that serve as targets for transformation, two paradigms have been developed to explain how intratumoral heterogeneity comes about: the clonal evolution model and the cancer stem cell (CSC) model. The Clonal Evolution Model The clonal evolution model states that cancer cells acquire various combinations of mutations and that genetic drift and natural selection facilitate the expansion of increasingly aggressive cells that drive tumourigenesis (Nowell, 1976). Tumour initiation takes place once multiple mutations occur within a single cell providing it with a 33  selective growth and survival advantage over adjacent normal cells. Genomic instability lays a favorable ground for the acquisition of additional mutations that accelerate tumour progression (Gao et al., 2007). These mutations can be biologically inert passengers or advantageous drivers that increase cellular fitness to yield divergent expansion of a new subpopulation. This stepwise nature of tumour progression where increasingly variant cells are born and other populations contract is the basis of heterogeneity. Perhaps the most convincing evidence for clonal evolution comes from genomic profiling of discrete tumour regions. Analysis of single cells taken from various regions within a tumour established that the majority of breast cancers are polygenomic, that is, they contain multiple clonal subpopulations that are spatially separated (Navin et al., 2011; Navin et al., 2010). In further support, genomic complexity was found to increase with progression from localized to metastatic breast cancer (Ding et al., 2010). The Cancer Stem Cell Model In contrast to clonal evolution, the CSC model dictates that tumours originate specifically in tissue stem cells or progenitor cells (Clarke et al., 2006). Stem cells are particularly enticing targets for transformation because of their inherent capacity for self-renewal and their longevity, which would allow for the sequential accumulation of genetic or epigenetic mutations necessary for tumourigenesis. Like their normal counterparts, CSCs possess self-renewal capacity to drive tumour growth, recurrence, and metastatic spread. They can also differentiate giving rise to a heterogeneous population of non-tumourigenic cancer cells that constitute the bulk of the tumour (Visvader and Lindeman, 2012). Experimental evidence supporting the cancer stem cell hypothesis first emerged in 1997 by Dicks’ group whereby they demonstrated that acute myeloid leukemia is propagated by a small population of cells capable of transferring the disease to immunocompromised mice (Bonnet and Dick, 1997). The surface markers used to isolate these cells, CD34+/CD38-, had previously been established to identify normal stem cells in the hematopoietic system (Larochelle et al., 1996). Thus, it was rationalized that leukemia stem cells capable of initiating the disease originated as normal stem cells.  34  The CSC concept was extended to solid tumours by Clarke and Wicha. In an influential study, they reported that human breast cancers contain a small population of highly tumourigenic cells bearing the surface markers CD44+/CD24- (Al-Hajj et al., 2003). Transplanting as few as 100 of these cells into NOD/SCID mice could regenerate a tumour, whereas it took thousands of cells with alternative phenotypes. Moreover, this tumourigenic subpopulation could be serially passaged and the resulting xenografts maintained the phenotypic heterogeneity of the original tumour (Al-Hajj et al., 2003; Ponti et al., 2005). These progenitor-like properties are reminiscent of normal tissue stem cells suggesting that CD44+/CD24- cells could represent transformed MaSCs. The selection of prospective markers that can identify and isolate CSCs in mammary tumours remains controversial and perplexing. This is because, by definition, CSCs should be phenotypically similar to normal MaSCs or progenitor cells, which themselves are poorly characterized. Gene expression profiling revealed that CD44+/CD24- cells are more “stem-like” and less differentiated than CD44-/CD24+ cells (Shipitsin et al., 2007). Specifically, they express putative stem cell markers and have an activated TGFβ signaling pathway, which is known to regulate human embryonic stem cell fate (James et al., 2005). Interestingly, approximately 90% of CD49f+/EpCAM- cells (which includes putative bipotent MaSCs) co-express CD44, while only 30% of the more differentiated CD49f-/EpCAM+ cells were also CD44+ (Raouf et al., 2008). This could explain why CD44-/- mice exhibit delayed ductal outgrowth and morphologically small terminal end buds (Louderbough et al., 2011). High ALDH activity was identified as the first common marker capable of isolating both normal and cancer stem/progenitor cells in the mammary gland (Ginestier et al., 2007). This supports the notion that stem/progenitor cells are targets for transformation in breast cancer. A subsequent study using Blg-Cre Brca1f/fp53+/- and K14-Cre Brca1f/fp53+/transgenic mice revealed that Blg-mediated BRCA1 loss in luminal progenitors produced cancers identical to ER-negative, basal-like human tumours. On the other hand, K14driven BRCA1 knockout in myoepithelial progenitors resulted in adenomyoepitheliomas, a rare tumour in humans (Molyneux et al., 2010). This suggests that luminal progenitors are the cell type most commonly associated with initiation of breast cancer.  35  1.2.3 Premalignant Lesions The transition from a normal mammary epithelial cell into a premalignant hyperplasia (ADH) is associated with increased growth and cytologic atypia, but the number of genetic changes remains relatively low. Merely 2% of these lesions exhibit amplification and/or overexpression of ErbB2/HER2 and no abnormalities in p53 have been reported (Allred et al., 2001). This implies that epigenetic changes may underlie premalignant lesions, creating a fertile environment for the acquisition of mutations that drive breast cancer progression. Seminal studies by Tlsty and colleagues explored breast tissue from healthy, disease-free women and identified a subpopulation of epithelial cells that exhibit silenced p16INK4a due to promoter hypermethylation (Crawford et al., 2004; Holst et al., 2003; Romanov et al., 2001). These cells, termed variant HMECs (vHMECs), were detected in over 30% of biopsies and were associated with risk of developing breast cancer (Radisky et al., 2011). When isolated and propagated in vitro, vHMECs do not exhibit a classical senescent arrest. Instead, they proliferate an additional 30 to 50 generations beyond the time that the HMEC population activates proliferative arrest (Romanov et al., 2001). Due to eroding telomeres, these cells ultimately enter telomere-based crisis and acquire chromosomal aneuploidy in addition to various other structural abnormalities seen in the earliest lesions of breast cancer (Holst et al., 2003; Romanov et al., 2001). Further characterization revealed cyclo-oxygenase-2 (COX2) to be significantly activated in the vHMEC population (Crawford et al., 2004). COX2 stimulates mammary epithelial cell growth through the biosynthesis of prostaglandins and thromboxane A2, is a positive regulator of invasion, and acts to inhibit apoptosis and immune surveillance (Singh and Lucci, 2002). Interestingly, it was found that women who frequently used NSAIDs, a class of drugs that block the COX2 enzyme pathway, experienced a significant reduction in risk of breast cancer (Harris et al., 2006). Transducing vHMECs with H-rasV12 to simulate an oncogenic hit promoted additional aberrations commonly observed during progression to malignancy, including upregulation of telomerase activity, anchorage-independent growth, and DNA methylation at the RASSF1A, p57, and MGMT gene loci (Dumont et al., 2009). Taken together, vHMECs are ideal candidates for breast cancer precursors.  36  1.2.4 Ductal Carcinoma in situ The progression from an ADH to DCIS is accompanied by considerable genetic defects and genomic instability. These lesions are divided into two categories: low grade DCIS is well differentiated and ER/PR-positive, while high grade DCIS is poorly differentiated and ER/PR-negative (Allred et al., 2008). Most notably, the RTK ErbB2/HER2 is highly expressed and/or amplified in nearly 60% of DCIS lesions (Rosenthal et al., 2002). This yields strong and constant proliferative signaling, which could fuel tumour progression. Interestingly, only 20 to 30% of IDCs overexpress ErbB2/HER2 (Slamon et al., 1987) suggesting it is a transient phenotype that facilitates unrestrained growth during early malignancy. Allelic imbalance at loci encoding cyclin D1 (11p), RB (13q), and p53 (17p) is also a defining feature of DCIS lesions (Fujii et al., 1996). Of particular interest is the high frequency of chromosome 1p gain because it contains the YB-1 locus (discussed in detail in section 1.4).  1.2.5 Invasive Ductal Carcinoma Based on comparative genomic hybridization (CGH) and loss of heterozygosity (LOH) analysis it is widely regarded that DCIS lesions are the non-obligate precursor to IDC (O'Connell et al., 1998). During the evolution of IDC cells breach the basement membrane (stage I), invade the surrounding stroma (stage II), and migrate to form micrometastases in the axillary lymph nodes (stage III) and eventually further throughout the body (stage IV). Many genes are deregulated in these tumours, including those with a role in cell proliferation (cyclin D1, p16INK4a, p27Kip1), angiogenesis (HIF-1α), and inhibition of apoptosis (MDM2, Bcl-2) (Zhao et al., 2004). However, a denominator common to over 70% of these tumours is loss of CDH1, which encodes the cell-cell adhesion molecule E-cadherin (Kowalski et al., 2003). Functional perturbations of Ecadherin and/or catenins have been associated with susceptibility to enter an epithelialmesenchymal transition (EMT). During this biologic process polarized epithelial cells undergo multiple biochemical changes to assume a mesenchymal cell phenotype, which includes enhanced migratory capacity, invasiveness, and elevated resistance to apoptosis (Kalluri and Weinberg, 2009).  37  1.2.6 Molecular Classification of Breast Carcinoma Long before the advent of modern molecular profiling techniques, histopathologists recognized that breast cancer was a heterogeneous disease though morphological observation. The development of DNA microarrays proved this heterogeneity, demonstrating that through gene expression profiling of large patient cohorts breast cancer could be classified into at least five subtypes (Herschkowitz et al., 2007; Perou et al., 2000; Sorlie et al., 2003). Expression of the estrogen receptor (ER) demarcates the main divergence point between the subtypes. There are two subtypes of ER-positive tumours (luminal A and luminal B) and three subtypes with negligible expression of ER (HER2, basal-like, and claudin-low). Each of these subtypes differs not only with respect to the genes they express, but also prognosis and treatment response (Sorlie et al., 2001). Patients with luminal breast tumours are amenable to hormone therapy and have a relatively favorable prognosis. Conversely, HER2-overepxressing and basal-like breast cancers (BLBCs) have the shortest disease-free survival. Very recently it was proposed that breast cancer could be classified into at least ten subtypes according the common genetic traits (Curtis et al., 2012). Table 1.2 outlines the characteristics of the established subtypes, which are described in more detail below.  Subtype  Immunoprofile  Luminal A  ER+, PR+/-, HER2-  Luminal B  DFS (%)  Origin  Ki67 low  83  Luminal  ER+, PR +/-, HER2+  Ki67 high, mutated p53  87  Lum pro  HER2  ER-, PR-, HER2+  Ki67 high  66  Lum pro  Basal-like  ER-, PR-, HER2-  EGFR+ and/or CK5/6+  74  Lum pro  Ki67, E-cadherin low  74  MaSC  Claudin-low ER-, PR-, HER2-  Other characteristics  Table 1.2 Molecular subtypes of breast carcinoma. The molecular subtypes of breast cancer differ with respect to gene expression, 5-year disease-free survival (DFS), and cell of origin. Lum pro, luminal progenitor; MaSC, mammary stem cell. Adapted from (Sorlie et al., 2003), (Onitilo et al., 2009), and (Visvader, 2009).  38  Luminal Subtype The majority of breast cancers (~70%) belong to the luminal subtype. The gene expression profile of these tumours is similar to the luminal epithelium of the mammary gland. They are ER-positive and typically express the progesterone receptor (PR) and the luminal cell-specific markers CK8 and CK18, in addition to several other transcription factors associated with ER activation (Perou et al., 2000). Luminal A tumours are generally low-grade and carry the best prognosis of all breast cancer subtypes (Sorlie et al., 2001; Sorlie et al., 2003). They exhibit the highest expression of ER and ERassociated genes, and low expression of HER2. In contrast, the high-grade luminal B tumours have low to moderate expression of the ER gene cluster. TP53 mutation is also much more prevalent in the luminal B subtype (Sorlie et al., 2001). HER2 Subtype Approximately 15% of breast cancers are classified as HER2-overexpressing. These tumours are characterized by high expression of several genes in the ErbB2 amplicon (17q22.24) including ErbB2 and GRB7. Furthermore, they are typically negative for ER and PR, do not express the ER-associated cluster genes, and frequently have mutated TP53. Despite advances in HER2-targeted therapies, patients diagnosed with this subtype have poor prognosis (Sorlie et al., 2003). Basal-like Subtype The BLBC subtype comprises 15 to 20% of breast cancers. The classification as basallike refers to the similarity in gene expression profiles between these tumours and normal basal/myoepithelial cells in the breast. The basal phenotype is characterized by high expression of CK5, CK17, laminin, and fatty acid binding protein 7 (Sorlie et al., 2001). Because BLBCs also lack expression of ER, PR, and HER2, they are commonly referred to as triple-negative breast cancers (TNBCs). Although there are similarities in the basal and triple-negative phenotypes, they are not strictly interchangeable. Based on gene expression profiling, 91% of triple-negative cancers are of basal-like subtype, but only 77% of BLBCs are triple-negative (Kreike et al., 2007). While lack of ER, PR, and HER2 cannot identify BLBC, the addition of CK5/6 and EGFR to the panel can resolve these tumours with 100% specificity and 76% sensitivity (Nielsen et al., 2004). 39  BLBCs are notoriously aggressive and difficult to treat due to the absence of expression of a recognized therapeutic target, such as ER or HER2. Relative to the other subtypes, patients diagnosed with these tumours have the greatest probability of recurrence within five years of the initial treatment (Dent et al., 2007). With high rates of relapse and mortality there is a great need to identify therapeutics that will improve clinical outcome. Claudin-low Subtype The claudin-low subtype was recently described following a rigorous analysis of established human and murine breast tumour data sets (Herschkowitz et al., 2007). Initially clustered with the basal-like subtype due to a lack of ER, PR, and HER2 expression and associated poor prognosis, these tumours uniquely downregulate the tight junction proteins claudin 3, 4, and 7, and the cell-cell adhesion glycoprotein E-cadherin. Furthermore, the claudin-low subtype is enriched for features associated with CSCs, such as expression of CD44 and lack of CD24 (Prat et al., 2010).  1.2.7 Modeling Breast Cancer Progression Rapid advances in next generation sequencing and DNA microarray technology have made it possible to identify genes causally involved in the development and progression of a cancer from an individual patient. Most recently, Aparicio and colleagues sequenced the genome and transcriptome of a primary ER-positive lobular breast cancer and a matched metastasis. Five of the 32 somatic coding mutations detected in the primary tumour were shared with the metastasis suggesting that they may underlie disease progression (Shah et al., 2009). Interestingly, these genes functioned in drug efflux, mitotic spindle assembly and maintenance of centrosome integrity, and DNA repair. A complementary study generated comprehensive in situ transcriptome profiles of the premalignant, preinvasive, and invasive stages of breast cancer. It was found that significant alterations in gene expression occur in ADH lesions that are maintained in the later stages of DCIS and IDC suggestive of a clonal relationship between pathological states (Ma et al., 2003). While sequencing studies have provided unique insight into the evolution of breast cancer, they only offer a snapshot of a highly dynamic process. Model systems are needed to directly evaluate the role of specific genes in tumour initiation and progression.  40  The pioneering work of Leder and Muller, who developed the first transgenic mouse model of breast cancer (Muller et al., 1988), has contributed extensively to our understanding of the genetic and molecular events that underlie the disease. However, the utility of using transgenic models for modeling human breast cancer has been subject to intense scrutiny. In 2000, the US National Institutes of Health Breast Cancer Think Tank compared the pathology between human breast tumours and those from 39 unique breast cancer transgenic mice (Cardiff et al., 2000). From this analysis it was concluded that the histology of most tumours from transgenic mice do not resemble the common types of human breast cancer, notably adenocarcinomas. Unlike human tumours, the majority of transgenic mouse tumours metastasize only to the lung, contain limited fibrosis and inflammation, and nearly all are hormone independent. Traditionally, transgenic models of breast cancer were engineered using MMTV- or whey acidic protein (WAP)-driven oncogene expression. Several limitations exist with respect to this approach. First, the promoters are mammary gland selective, but not specific. Second, the transgenes are expressed throughout the entire mammary epithelium rather than targeted to a specific cell of origin. Finally, the level of oncogene expression that is driven by these promoters cannot be tailored to match what is observed in human cancer (Vargo-Gogola and Rosen, 2007). To address these issues much effort has been directed toward developing next-generation transgenic animals with increased spatial-temporal control over transgene expression. Notably, an elegant transgenic knock-in mouse model that places activated ErbB2/HER2 under the control of the endogenous ErbB2 promoter yields mammary tumours that are molecular phenocopies of ErbB2-initiated human breast cancer (Andrechek et al., 2000; Montagna et al., 2002). In addition, Cre/loxP transgenic models have been developed that allow for the deletion or activation of genes in specific mammary cell lineages. Using K14-Cre to conditionally delete E-cadherin and p53 in myoepithelial cells results in mammary tumours that histologically resemble a human invasive carcinoma (Derksen et al., 2006). While transgenic mice can take upwards of a year to form tumours, cell culture models represent a promising approach to rapidly identify and probe the function of genes that drive breast cancer progression. Weinberg and colleagues established the first model of  41  full HMEC transformation by serial induction of SV40 early region, hTERT, and activated Ras (H-rasV12) into primary HMECs (Elenbaas et al., 2001). It was shown that SV40, which binds and inactivates p53 and RB, abolishes senescence, whereas the catalytic subunit of telomerase, hTERT, is needed to promote immortalization. Ras, which is constitutively active in many human cancers, stimulates the Raf/MEK/ERK and PI3K/AKT pathways to modulate cell proliferation, differentiation, and survival. Refinements to this transformation scheme have helped eliminate the confounding effects of oncogenic Ras to construct more relevant models. Specifically, it was demonstrated that Ras-mediated activation of the PI3K pathway increases Myc activity, which in turn represses thrombospondin-1 (Tsp-1), an anti-angiogenic factor (Watnick et al., 2003). This was shown to be an absolute requirement for cells to form tumours with unfettered growth potential as those expressing Tsp-1 could not progress beyond a 2-millimeter diameter. Moreover, expressing Akt1 and Rac1, two downstream targets of the PI3K signaling pathway, permitted anchorage independent growth in the absence of Ras (Zhao et al., 2003). HMECs transformed using SV40, hTERT, and Ras give rise to poorly differentiated carcinomas with areas of squamous differentiation when implanted into immunocompromised mice (Elenbaas et al., 2001; Ince et al., 2007). This particular tumour phenotype is extremely rare, representing less than 1% of human breast cancers, urging the need to develop more clinically relevant models. HMEC transformation in vitro is a useful platform for evaluating combinations of genes necessary for tumour initiation, yet it suffers a disadvantage in that it does not reflect the distinct stages of breast cancer progression. To map the multistage transformation process breast cancer progression series have been developed. The MCF10 series consists of a set of mammary epithelial cell lines that evolved from Ras-transformed HMECs derived from a reduction mammoplasty (Dawson et al., 1996). Cells were serially passaged through xenografts to generate a set of cell lines that progressively increase in tumourigenicity. They range in phenotype from MCF10AT cells, which are premalignant and form lesions resembling ADH and DCIS, to MCF10CA cells, which form well-defined adenocarcinomas within 15 weeks and readily metastasize to the lung (Dawson et al., 1996; Santner et al., 2001). The first patient-derived mammary tumour  42  progression series, known as the 21T series, is comprised of three cell lines all isolated from an individual with metastatic breast cancer (Band et al., 1990). Two cell lines established from the mastectomy specimen are non-tumourigenic and tumourigenic but non-metastatic, respectively. The third cell line derived from a malignant pleural effusion is both tumourigenic and metastatic. A principal limitation of the abovementioned in vitro culture studies is that the conditions are markedly different from the breast microenvironment. To recapitulate the in vivo structural organization and architecture of the glandular breast epithelium, Bissell and Brugge pioneered a three-dimensional (3D) culture model (Debnath and Brugge, 2005; Debnath et al., 2003; Lee et al., 2007). Epithelial cells embedded in a reconstituted basement membrane matrix proliferate and organize into cyst-like spheroids, commonly called acini. These structures are notable for the presence of a centrally located, hollow lumen and the polarization of cells surrounding this lumen. Early studies revealed that when breast tumour cells are compared to normal HMECs in 3D culture conditions, a distinguishing feature of the malignant cells is their inability to form organized, polar 3D acini (Petersen et al., 1992). As such, 3D cultures provide a context to identify genes that induce phenotypic changes associated with tumour progression. For example, filling of the lumen and loss of apicobasal polarity are salient features of early breast cancer that remain poorly understood. Developing breast cancer progression models in 3D culture could delineate some of the mechanisms and pathways that contribute to these processes.  1.3 Breast Cancer Therapies Molecular profiling has shifted the way breast cancer is treated. Whereas histopathological features such as tumour size, grade, and lymph node status once guided treatment, gene expression analysis is now used to select high-risk patients that would benefit from adjuvant chemotherapy (van 't Veer et al., 2002). The development of new targeted therapies, cytotoxic agents, and radiation therapy techniques has much improved clinical outcomes. Despite these advances, resistance remains a significant obstacle to the success of chemotherapy, particularly in the triple-negative/basal-like subtype.  43  1.3.1 Current Therapies The majority of breast cancers are detected by an abnormal mammogram. Breast conserving surgery followed by local radiation and adjuvant chemotherapy is the mainstay of treatment. Cytotoxic drugs used in the adjuvant setting are designed to eliminate rapidly dividing cells. These include alkylating agents (cyclophosphamide), anthracyclines (doxorubicin and epirubicin), and anti-metabolites (5-fluorouracil), which kill tumourigenic cells by DNA damage, interfering with DNA repair mechanisms, and disturbing metabolic pathways (Bange et al., 2001). One of the most potent classes of anti-cancer drugs, the taxanes (paclitaxel and docetaxel), promote hyper-stabilization of the mitotic spindle to induce apoptosis via a p53-independent mechanism (Lanni et al., 1997). Because adjuvant therapies are non-selective, side effects including nausea, vomiting, and alopecia are common. To limit acute toxicity simultaneous combination of two or more agents is often used in the management of breast cancer. This approach allows for synergistic efficacy at a reduced dose of individual drug. CMF (cyclophosphamide, methotrexate, 5-fluorouracil) and TAC (paclitaxel, doxorubicin, cyclophosphamide) are the most common combination chemotherapy regimens used in breast cancer (Lee and Nan, 2012). Molecular characterization of breast cancer has guided the transition from a “one size fits all” treatment approach to a personalized approach where tumour subtypes are targeted with distinct systemic therapies. The discovery of tamoxifen in the 1970s and trastuzumab (Herceptin®) in the 1990s revolutionized the treatment of hormone receptor (ER/PR) positive and HER2 positive breast cancer, respectively. The essential function of estrogen in stimulating malignant cell proliferation resulted in the development of tamoxifen, a selective antagonist of ER in breast tissue and the first target-directed cancer drug (Ali and Coombes, 2002). Tamoxifen itself is a prodrug with little affinity for ER, but following first pass liver metabolism 4-hydroxytamoxifen is generated which competes with estrogen for ER binding. Administered as an endocrine therapy, tamoxifen has dramatically reduced the incidence of recurrence and mortality in hormone-receptor positive tumours. Maximal benefit is obtained by treatment for five years, with a 51% reduction in relapse and 28% reduction in death that is sustained well beyond the  44  treatment period (Osborne, 1998). The success of tamoxifen has fueled the development and FDA approval of aromatase inhibitors, such as anastrozole, that block the synthesis of estrogen, and selective ER modulators (SERMs), such as raloxifene, that are antiestrogenic for the breast but estrogenic for other tissues where the protective actions of the hormone are desirable (Dunn and Ford, 2007). Trastuzumab, a humanized monoclonal antibody that binds to the extracellular domain of HER2, was an important discovery for the treatment of HER2-positive breast cancer. When combined with taxane-based chemotherapy trastuzumab confers a 50% decrease in recurrence and 40% decrease in mortality at three years of follow-up (Joensuu et al., 2006). Although trastuzumab is a very active agent in HER2-overexpressing breast cancer the majority of patients who initially respond develop resistance and relapse. To date, the mechanism of action of trastuzumab and how resistance develops is not fully understood. It has been postulated that the drug diminishes HER2 receptor signaling by mediating its internalization (Cuello et al., 2001) and/or dephosphorylation (Nagata et al., 2004). Resistance has been attributed to ligand-independent PI3K signaling as a consequence of activating PIK3CA mutations or loss of PTEN (Berns et al., 2007). As such, several next-generation drugs are being evaluated for the treatment of trastuzumab refractory cancers. The dual EGFR/HER2 tyrosine kinase inhibitor lapatinib is active in a proportion of patients with metastatic disease who have relapsed on trastuzumab therapy (Geyer et al., 2006). Most recently, the antibody-drug conjugate trastuzumab emtansine (T-DM1), which combines trastuzumab with the antimicrotubule agent DM1, was found to significantly extend survival of patients with HER2-positive metastatic breast cancer compared to those who received current standard therapy (Verma et al., 2012). In contrast to hormone receptor positive and HER2 positive breast cancers, there are currently no viable targeted therapies for the triple-negative subtype (ER-/PR-/HER2-). Accordingly, the only systemic treatment option available for these patients is standard chemotherapy with cytotoxic agents. Fortunately, several clinical trials have shown a marked chemosensitivity for these tumours, especially to neoadjuvant taxane and anthracycline-based regimens. TNBC patients derive the most substantial benefit from TAC chemotherapy with nearly 40% achieving a pathologic complete response compared  45  to only 15% of non-TNBC patients (Huober et al., 2010). However, in advanced metastatic TNBC response to chemotherapy lacks durability and the median survival of patients is only four months (Kennecke et al., 2010). Novel therapeutics designed to target key signaling molecules in TNBC are currently in clinical development. Bevacizumab, a monoclonal antibody against VEGF, has demonstrated modest efficacy in prolonging progression-free survival from 5.9 months to 11.8 months (Miller et al., 2007). A second therapeutic being explored in phase II clinical trials is the poly(ADPribosyl)ation polymerase (PARP) inhibitor BSI-201 (O'Shaughnessy et al., 2011). In addition, as EGFR is frequently over-expressed in TNBC, it follows that targeting it using tyrosine kinase inhibitors and/or monoclonal antibodies could offer therapeutic benefit. While single-agent clinical trials with cetuximab and gefitinib were disappointing, they have value when combined with taxane- and platinum-based chemotherapy (De Laurentiis et al., 2010). Finally, the finding that c-Kit is highly expressed in the TNBC/basal-like subtype makes it an attractive target (Lim et al., 2009). An inhibitor, imatinib, is currently in a phase II clinical trial for metastatic breast cancer at MD Anderson Cancer Center (NCT00338728).  1.3.2 The Triple-Negative Paradox Compared to all the other breast cancer subtypes TNBCs consistently demonstrate the highest response rates to neoadjuvant chemotherapy, yet these patients have the greatest likelihood of distant recurrence and death within five years of diagnosis (Dent et al., 2007). This surprising behavior is colloquially referred to as the “TNBC paradox.” It has been postulated that the overall poor prognosis could be attributed to a small subset of cells with intrinsic multidrug resistance (Creighton et al., 2009; Li et al., 2008).  1.3.3 Mechanisms of Resistance: ABC Transporters Failure of cancer chemotherapy is largely the result of malignant cells becoming simultaneously resistant to different classes of drugs, a trait known as multidrug resistance (MDR). There are three major mechanisms of drug resistance in cells: first, decreased uptake of water-soluble drugs, which require transporters to enter cells; second, various changes that diminish the capacity of cytotoxic drugs to kill cells,  46  including alterations in cells cycle, increased repair of DNA damage, reduced apoptosis, and activation of detoxifying enzymes; and third, increased ATP-dependent efflux of hydrophobic drugs that can easily enter cells through diffusion across the plasma membrane (Gottesman et al., 2002; Szakacs et al., 2006). Of these mechanisms, the most prominent in cultured cancer cells is enhanced drug efflux mediated by a family of energy-dependent transporters known as ATP-binding cassette (ABC) transporters. The human genome contains 48 genes encoding ABC transporters that have been divided into seven distinct subfamilies (ABCA to ABCG) based on sequence homology and domain organization (Dean et al., 2001). The most important with respect to clinical drug resistance are ABCB1 (also known as MDR1 and P-glycoprotein) and ABCG2 (also known as MXR and breast cancer resistance protein (BCRP)). These transporters are particularly overexpressed in breast cancer and can efflux a wide range of chemotherapeutic agents used to treat patients (Szakacs et al., 2006). Table 1.3 outlines the cancer-specific substrates of ABCB1/MDR1 and ABCG2/BCRP.  47  Transporter  Chemotherapy substrates Drug Class  ABCB1/MDR1  ABCG2/BCRP  Drug  Anthracyclines  Daunorubicin, Doxorubicin, Epirubicin  Epipodophyllotoxins  Etoposide, Teniposide  Kinase inhibitors  Imatinib  Vinca alkaloids  Vinblastine, Vincristine  Other  Actinomycin-D, Bisantrene, Colchicine  Anthracyclines  Doxorubicin, Epirubicin  Epipodophyllotoxins  Etoposide, Teniposide  Kinase inhibitors  Gefitinib, Imatinib, Flavopiridol  Taxanes  Docetaxel, Paclitaxel  Thiopurines  5-fluorouracil  Other  Methotrexate, Mitoxantrone, SN-38  Table 1.3 Chemotherapy substrates of the ABCB1 and ABCG2 transporters. ABC transporters efflux chemotherapeutics routinely used in the management of breast cancer. A complete list of substrates is reviewed in (Szakacs et al., 2006).  ABCB1 was the first efflux pump discovered with the ability to transport a broadspectrum of compounds central to most chemotherapy regimens, including anthracyclines, epipodophyllotoxins, and vinca alkaloids (Juliano and Ling, 1976). A meta-analysis found ABCB1 to be expressed in 41% of breast tumours and this was associated with a greater likelihood of treatment failure (Trock et al., 1997). Accordingly, most research in the last two decades has pursued reversing MDR using ABCB1 inhibitors. Despite initial optimism, clinical trials have been disappointing overall (reviewed in (Szakacs et al., 2006)). The first generation ABCB1 inhibitors, such as verapamil, were associated with unacceptable levels of toxicity. While the second generation inhibitors, such as valspodar, demonstrated increased potency and decreased toxicity they failed to improve clinical outcomes. Although third generation inhibitors have now been developed the initial results are not promising. For example, a phase II trial of women with recurrent breast cancer found no additional benefit in overall survival  48  with the inclusion of zosuquidar, a specific ABCB1 inhibitor, to docetaxel chemotherapy (Ruff et al., 2009). Taken together, these trials have led to the conclusion that the contribution of ABCB1 to clinical drug resistance might not be particularly important. While the relevance of ABCB1 to the treatment of breast cancer remains questionable, ABCG2 clearly has the potential to contribute to drug resistance in these tumours. It was initially identified as a mediator of doxorubicin resistance in MCF-7 breast cancer cells (Doyle et al., 1998). Further studies solidified the role of ABCG2 in chemoresistance by demonstrating that cells derived from abcg2 knockout mice were exquisitely sensitive to mitoxantrone (Zhou et al., 2002; Zhou et al., 2001). It has since been shown that ABCG2 is a high-capacity transporter with promiscuous substrate specificity. Similar to ABCB1, it effluxes many of the common therapeutics used in breast cancer treatment including 5fluorouracil (Yuan et al., 2009) and paclitaxel (Guo et al., 2004) in addition to more recently developed tyrosine kinase inhibitors (Chen et al., 2011). In support of its clinical importance, ABCG2 can be used as a reliable prognostic indicator to identify breast cancer patients who will derive benefit from anthracycline-based therapy (Burger et al., 2003). Curiously, in the dozens of clinical trials for ABCB1 inhibitors only the dual ABCB1/ABCG2 inhibitor cyclosporine was found to significantly improve overall survival (Szakacs et al., 2006). To date, no specific ABCG2 inhibitors are suitable for clinical studies; however, early results from the third generation dual ABCB1/ABCG2 inhibitor tariquidar have been promising (Kelly et al., 2011). ABCG2 is expressed and active in hematopoietic stem cells, but the gene is silenced in committed progenitors and mature blood cells (Scharenberg et al., 2002). Of note, abcg2 knockout mice are viable and have normal stem cell compartments (Zhou et al., 2002). In response to hedgehog signaling, Gli transcription factors enhance ABCG2 expression by direct promoter binding (Singh et al., 2011), while PI3K/Akt signaling mediates plasma membrane trafficking (Mogi et al., 2003). It has recently emerged that epigenetic changes are responsible for ABCG2 expression in refractory cancer cells. Through an elusive mechanism, progression to a drug resistant phenotype is associated with histone H3 acetylation and CpG island demethylation along the abcg2 promoter (Calcagno et al., 2008; To et al., 2006).  49  1.3.4 Intrinsic Drug Resistance: Tumour-Initiating Cells There is considerable evidence that a population of cells, termed tumour-initiating cells (TICs), have the intrinsic capacity to evade chemotherapy and serve as an unrestricted reservoir for drug-resistant tumour relapse (Creighton et al., 2009; Fillmore and Kuperwasser, 2008; Marangoni et al., 2009). It is important to note that TICs, the cellular subset with tumour-propagating capacity, are not necessarily related to CSCs, the normal stem cells that acquire the first cancer-promoting mutations. TICs can be fractionated from breast tumours based on CD44+/CD24- (Al-Hajj et al., 2003) and/or CD133+ marker expression (Wright et al., 2008) in addition to Hoechst 33342 dye efflux (HirschmannJax et al., 2004). These populations are expanded post-chemotherapy (Creighton et al., 2009) and are most frequent in the TNBC subtype (Honeth et al., 2008; Park et al., 2010), which could explain their particularly high rates of relapse. One explanation for the refractoriness of TICs is through their expression of ABC transporters. In particular, ABCG2 is responsible for the efflux of Hoechst 33342 that defines the “side-population (SP)” (Hirschmann-Jax et al., 2004). SP cells isolated from primary breast tumours exhibit increased resistance to chemotherapy, have high efflux capacity, and as few as 100 are able to generate tumours in vivo (Nakanishi et al., 2010). Interestingly, ABCG2 is also enriched in CD133+ cells (Bertolini et al., 2009) and Gli1/2 transcription factors are highly expressed in the CD44+/CD24- population (Shipitsin et al., 2007). This suggests that high levels of ABCG2 may represent a common denominator between the different TIC phenotypes. Although the expression of ABCG2 could render TICs resistant to drugs it is not the sole determinant of resistance as DNA-repair capacity and reluctance to enter apoptosis could be equally as important. Efforts to develop novel therapeutic approaches that target the TIC population should represent a research priority.  1.4 YB-1: A Potential Regulator of Breast Cancer Progression Y-box binding protein-1 (YB-1) is a transcription and translation factor belonging to the evolutionarily conserved cold-shock domain protein superfamily. It was first discovered as a protein bound to the major histocompatibility complex (MHC) class II promoter  50  (Didier et al., 1988) and shortly thereafter was found to interact with the EGFR enhancer and HER2 promoter (Sakura et al., 1988). Considerable research over the past two decades has established that YB-1 coordinates a diverse array of cellular processes including proliferation, DNA repair, drug resistance, and EMT. Figure 1.7 depicts the signal transduction cascade that stimulates YB-1 to orchestrate its many functions, which are discussed below.  RTKs% (EGFR,%HER2,%IGF1R)%  P  P  PI3K% AKT%  RAS% P  YB#1% DNA+ AP1% Repair+ PCNA% Polδ% P YB#1%  RSK1/2% YB#1%  P  MEK% ERK%  P P  P  Transla)on+  YB#1% AUG%  EMT+mRNA+(Snail1,%Twist)%  eIF4E%Complex%  Transcrip)on+ P  YB#1% CCAAT%  Prolifera)on+ Cell+Cycle+ CDC6% EGFR% Cyclin%A% HER2% Cyclin%B1% PIK3CA% %  Resistance+  TICs+  MDR1% MKNK1%  CD44% CD49f%  Figure 1.7 A cellular wiring diagram of the YB-1 signal transduction network. Engagement of receptor tyrosine kinases (RTKs) stimulates the activation of the PI3K (green) and MAPK (purple) pathways leading to the phosphorylation of YB-1 (YB-1S102) by RSK1/2 and to a lesser extent AKT. Activated YB-1 (1) functions in DNA repair by associating with damaged DNA and assembling repair complexes (red); (2) activates transcription by binding to inverted CCAAT boxes in the promoters of genes involved in proliferation, cell cycle progression, drug resistance, and maintenance of TICs (grey); and (3) activates cap-independent translation of EMT-mediating factors (turquois).  1.4.1 General Overview of YB-1 YB-1 is comprised of three unique structural domains: an alanine/proline rich N-terminal domain, which functions in trans-activation and mRNA localization; a cold-shock domain (CSD), which is essential for YB-1 affinity to nucleic acids; and a C-terminal  51  domain consisting of alternate clusters of basic and acidic amino acids (B/A repeats), which mediates protein-protein interactions (Kohno et al., 2003). The intracellular trafficking of YB-1 is governed by a non-canonical nuclear localization signal (NLS) and cytoplasmic retention site (CRS) contained within the C-terminal domain. Deletion mutants lacking the CRS are confined to the nucleus indicating the necessity of the motif for cytoplasmic localization (Bader and Vogt, 2005). Proteolytic cleavage of the CRS by the 20S proteasome in a ubiquitin- and ATP-independent manner is one mechanism responsible for YB-1 nuclear accumulation (Sorokin et al., 2005). In the nucleus, YB-1 binds to a consensus CCAAT sequence, known as a YB-1 responsive element (YRE), to activate or repress transcription (Didier et al., 1988). Various protein modifications such as phosphorylation and ubiquitylation have been shown to be critical in regulating YB-1 function. Phosphorylation at the Ser-102 residue, within the CSD, has been most extensively characterized. RSK (Stratford et al., 2008) and to a lesser extent AKT (Sutherland et al., 2005) are responsible for YB-1 Ser-102 phosphorylation, which is an absolute requirement for DNA binding activity (Wu et al., 2006). Moreover, this modification mediates the nuclear translocation of YB-1 and decreases its function as a transcriptional repressor (Evdokimova et al., 2006). Other putative phosphorylation sites on YB-1 have been identified by mass spectrometry and could have important function, notably Tyr-162 within the NLS (Rush et al., 2005). YB-1 is degraded by the 26S proteasome following ubiquitylation by the SCFFBX33 complex (Lutz et al., 2006). Figure 1.8 depicts the functional elements of the YB-1 protein. FBX33)binding)site) N  Y162*)  Cold)shock)domain)  A/P$rich) 1)  S102)  51)  Trans$acKvaKon)  CTD) 129)  DNA/RNA)Binding)  20S)cleavage)site)  NLS)  C)  CRS) 219)  324)  Protein$protein)InteracKons)  Figure 1.8 The structural domains of YB-1. YB-1 contains three domains: an A/P-rich N-terminal, a cold shock domain, and a Cterminal domain (CTD). A NLS and CRS allow for nucleo-cytoplasmic trafficking. This is mediated by proteolytic cleavage and, possibly, Y162 phosphorylation. S102 phosphorylation is required for DNA binding. YB-1 is tagged for degradation by FBX33.  52  1.4.2 Tissue Profile of YB-1 Expression YB-1 is expressed in a developmentally regulated manner with ubiquitous tissue expression throughout murine embryogenesis that begins to diminish at birth until it is largely undetectable at 24 months (Miwa et al., 2006). Its role in early development is non-redundant as YB-1 knockout mice are embryonic lethal with a runting phenotype and neural tube defects (Lu et al., 2005; Uchiumi et al., 2006). Normal human adult tissues are largely devoid of YB-1 expression. However, it has been detected in primitive cell populations, such as progenitor cells in the mammary gland (Finkbeiner et al., 2009) and neural stem cells in the subventricular zone of the brain (Fotovati et al., 2011). In stark contrast, YB-1 is highly expressed in a vast array of malignancies, for example melanomas, glioblastoma multiforme, and colorectal, prostate, and breast carcinomas (Eliseeva et al., 2011).  1.4.3 Regulation of YB-1 Expression Although the YB-1 gene has been cloned from multiple species including human (Didier et al., 1988), mouse (Shaughnessy and Wistow, 1992), and chicken (Ozer et al., 1990), very little is known about the regulatory elements and upstream factors that control its transcription. The human YB-1 gene (YBX1) has been mapped to chromosome 1p34.1, which flanks a region (1p34.2 - 1p34.4) commonly amplified in colon, lung, and breast cancers (Henderson et al., 2005). However, the YB-1 locus was not amplified in primary breast tumours or breast cancer cell lines assessed by array CGH (Shadeo and Lam, 2006; Stratford et al., 2007). An alternative mechanism of YB-1 over-expression in cancer may be through transcriptional upregulation. Sequence analysis of the 5’ regulatory region (2000 base pairs upstream of the transcriptional start site) have identified numerous Eboxes and GF-boxes that are potential binding sites for the transcription factors c-Myc and Twist. Using reporter assays, it was demonstrated that p73 stimulates the transcription of YB-1 by recruiting Myc/Max dimers to E-boxes in the promoter (Uramoto et al., 2002). Other studies have implicated Twist in regulating the expression of YB-1. Although the exact mechanisms are unclear, direct interaction between Twist and p300 is required for Twist-induced YB-1 expression (Shiota et al., 2010).  53  Conversely, p53 directly binds the N-terminal domain of Twist to repress YB-1 promoter activity (Shiota et al., 2008). Altogether, aberrant expression of c-Myc and Twist, increased p300 activity, and loss of p53 could lead to upregulation of YB-1 in cancers.  1.4.4 YB-1 Function in Breast Cancer: Proliferation and Drug Resistance YB-1 is a bona fide oncogene in breast cancer. Transgenic BLG-YB-1 mice all formed mammary carcinomas with a latency of 12 to 15 months. The histomorphology of these tumours varied, but was reminiscent of squamous cell carcinomas and adenocarcinomas of the breast. Closer examination revealed centrosome amplification and numerical chromosomal aberrations in the malignant cells (Bergmann et al., 2005). However, the underlying mechanism responsible for genomic instability and whether it acts as a driving force in YB-1-mediated tumourigenesis has not been deduced. The earliest phenotype associated with YB-1 expression was enhanced cellular proliferation. During the G1-S transition, YB-1 accumulates in the nucleus where it functions to promote cell cycle progression via transcription of cyclin A and cyclin B1 (Jurchott et al., 2003). It also functionally binds the promoter of CDC6, a canonical E2F1-regulated gene that is part of the pre-replicative complex necessary for S-phase entry and DNA replication (Basaki et al., 2010). Interestingly, the majority of promoters bound by YB-1 in breast cancer also contain E2F1/E2F binding sites (Lasham et al., 2012), suggesting the YB-1 co-regulates E2F target genes. Finally, YB-1 contributes to cell proliferation by transcriptionally activating members of the growth-promoting MAPK/Erk and PI3K/Akt signaling cascades, including EGFR (Stratford et al., 2007), HER2 (Wu et al., 2006), and PIK3CA (Astanehe et al., 2009). These findings are translated clinically as YB-1 was strongly associated with the proliferation marker PCNA in a TMA of 131 invasive ductal carcinomas (Yu et al., 2010). With an established role in cell proliferation, it is not surprising that inhibition of YB-1 using either siRNA or a cell permeable peptide could suppress the growth of breast cancer cell lines (Law et al., 2010; Stratford et al., 2007; Wu et al., 2006). Notably, fibroblasts derived from YB-1 knockout embryos displayed reduced growth and cell density, which could be rescued to wild-type levels by ectopically re-introducing YB-1  54  (Uchiumi et al., 2006). The proliferative effects of YB-1 were found to be contingent upon phosphorylation of its Ser-102 site. Mutating this residue to a non-phosphorylatable alanine (YB-1S102A) perturbs cell growth due, in part, to the inability of YB-1 to bind the EGFR and HER2 promoters (Sutherland et al., 2005; Wu et al., 2006). Likewise, inhibiting RSK using small molecule inhibitors prevents YB-1S102 activity thereby reducing proliferation and ultimately inducing apoptosis in breast cancer cells (Stratford et al., 2012). In addition to cell proliferation, YB-1 also plays a significant role in multidrug resistance. First, it is involved in the transcriptional control of MDR1, which encodes the ABCB1 MDR transporter (Bargou et al., 1997). Analysis of breast tumours following paclitaxel chemotherapy revealed the emergence of cells with nuclear localization of YB-1 and high ABCB1 expression (Fujita et al., 2005). Second, YB-1 focally regulates effectors of the MAPK pathway to stimulate its activity independent of growth factor receptor signaling. This could thwart antibody-based therapies. It has been distilled that YB-1-mediated transcription of MKNK1, a constituent of the MAPK pathway, imparts trastuzumab resistance (Astanehe et al., 2012). Both paclitaxel- and trastuzumab-resistant breast cancer cells are absolutely dependent on YB-1 to maintain their resistance phenotypes as targeted deletion of the oncogene is sufficient to re-sensitize these cells to drug (Dhillon et al., 2010; To et al., 2010). Third, YB-1 functions in the repair of DNA damage. It acts as a scaffold for assembly of base and nucleotide excision repair complexes by interacting with PCNA, DNA ligase IIIα, DNA polymerase δ, and AP endonuclease-1 (Eliseeva et al., 2011). Moreover, YB-1 itself possesses endonuclease activity and may actively excise damaged DNA. In response to the DNA-damaging agent cisplatin, YB-1 translocates to the nucleus and preferentially binds cisplatin-modified DNA to initiate nucleotide excision repair (Gaudreault et al., 2004). Fibroblasts from heterozygous YB1+/- mice demonstrate increased sensitivity to cisplatin and mitomycin C indicating that YB-1 is integral to the repair of DNA adducts (Shibahara et al., 2004). Finally, YB-1 is an anti-apoptotic protein. It inhibits the FAS-mediated cell death pathway at several points by acting as a transcriptional repressor at the FAS, BAX, and caspase 7 promoters (Eliseeva et al., 2011).  55  1.4.5 Insights into YB-1 Induction of Tumour-Initiating Cells It has recently emerged that YB-1 is intimately linked to the TIC phenotype. In a genome-wide chromatin immunoprecipitation-on-chip (COC) analysis, YB-1 was found to interact with the promoters of many TIC-associated genes, notably BMI1, CD44, CD49f, Met, c-Kit, and Notch family members (Finkbeiner et al., 2009). It was later validated that CD44 and CD49f are direct YB-1 target genes that are indispensible for the self-renewal capacity and drug resistance of TICs (To et al., 2010). As a result, targeting YB-1 directly or inhibiting its activation (YB-1S102) by suppressing RSK2 induces apoptosis in the CD44+/CD24- TIC population (Stratford et al., 2012). Transient dedifferentiation associated with EMT has been linked to the metastatic potential of TICs (Polyak and Weinberg, 2009). In Ras-transformed cells, YB-1 stimulates cap-independent translation of EMT-inducing factors including Lef-1, Snail1, Twist, and Zeb2 by directly binding the 5’-UTR of mRNAs (Evdokimova et al., 2009). Taken together, YB-1 coordinates many of the diverse phenotypes associated with TICs such as self-renewal, drug resistance, and metastatic spread.  1.4.6 Clinical Correlations YB-1 is consistently elevated in primary breast cancer tumours where it is most frequently associated with the more aggressive ER-negative subtypes (Habibi et al., 2008; Wu et al., 2006). In a cohort of over 4400 cases, YB-1 was detected in approximately 40% of invasive carcinomas and its expression was associated with relapse and poor overall survival across all breast cancer subtypes (Habibi et al., 2008). In fact, it more accurately predicted relapse than the gold-standard biomarkers ER and HER2 and, as such, has been used clinically to prospectively stratify individuals who would benefit from more aggressive high-dose chemotherapy (Gluz et al., 2009b). Patients with high YB-1 expression have a much greater probability of relapse compared to those with negligible expression (66% verses 0%) (Janz et al., 2002). More specifically, expression of YB-1 in tamoxifen-treated patients correlates with decreased breast cancer specific survival (Habibi et al., 2008).  56  1.5 Hypothesis and Aims The molecular events that define the evolution of a normal mammary epithelial cell into a fully malignant breast cancer are poorly understood. The overall objective of this research is to distill the complexities of breast tumourigenesis and identify the essential drivers of the disease that could be used as biomarkers or therapeutic targets. It is known that (1) all YB-1 transgenic mice form tumours (Bergmann et al., 2005), and (2) YB-1 is strongly associated with primary ER-negative breast cancers (Habibi et al., 2008). Therefore, the hypothesis tested in this research is whether YB-1 is a driver of the triplenegative breast cancer subtype. I developed a tetracycline-inducible in vitro model of tumour progression by ectopically expressing YB-1 in the normal human mammary epithelial cell line H16N2. This model faithfully recapitulated the genomic instability observed in tumours isolated from YB-1 transgenic mice. Thus, I characterized the underlying mechanism responsible for this phenotype and assessed whether it was stochastic or targeted toward increasing susceptibility to cancer (chapter 2). Further interrogation revealed that YB-1 expression permits HMECs to form tumours when transplanted into immunocompromised mice. Therefore, I probed the functional role of the oncogene in altering the genomic and epigenomic landscape of normal cells and identified the key signaling molecules that collectively promote full malignant transformation (chapter 3). Finally, I used the model to uncover novel mechanisms employed by YB-1 to evoke multidrug resistance, with a particular emphasis on the TIC subpopulation. The efficacies of therapeutics that target the RSK/YB-1 axis were evaluated for their ability to overcome TIC chemoresistance, which could translate into decreased clinical relapse (chapter 4).  57  CHAPTER 2. YB-1 EVOKES SUSCEPTIBILITY TO CANCER THROUGH CYTOKINESIS FAILURE, MITOTIC DYSFUNCTION, AND HER2 AMPLIFICATION  2.1 Overview Y-box binding protein-1 (YB-1) expression in the mammary gland promotes breast carcinoma that demonstrates a high degree of genomic instability. In the present study, we developed a model of premalignancy to characterize the role of this gene during breast cancer initiation and early progression. Antibody microarray technology was used to ascertain global changes in signal transduction following the conditional expression of YB-1 in human mammary epithelial cells (HMECs). Cell cycle associated proteins were frequently altered with the most dramatic being the LIM kinases (LIMK1/2). Consequently, the misexpression of LIMK1/2 was associated with cytokinesis failure that acted as a precursor to centrosome amplification. Detailed investigation revealed that YB-1 localized to the centrosome in a phosphorylation-dependent manner where it complexed with pericentrin and γ-tubulin. This was found to be essential in maintaining the structural integrity and microtubule nucleation capacity of the organelle. Prolonged exposure to YB-1 led to rampant acceleration toward tumourigenesis with the majority of cells acquiring numerical and structural chromosomal abnormalities. Overexpression of cyclin E potentiated cell cycle checkpoint slippage, a phenomenon where genomically compromised cells fail to maintain an arrested state and continue to proliferate. As malignancy further progressed, we identified a subset of cells harbouring HER2 amplification. Our results recognize YB-1 as a cancer susceptibility gene with the capacity to prime cells for tumourigenesis.  2.2 Introduction Cancer arises from the progressive evolution of a cell from normalcy through intermediate premalignant states to finally become invasive and metastasize (Hanahan  58  and Weinberg, 2000). Cells position themselves for this progression to malignancy by accumulating genetic alterations that result in the activation of oncogenes and inactivation of tumour suppressors. Studies of human mammary epithelial cells (HMECs) are beginning to provide key insight into these early genetic events to ascertain their role in fueling breast carcinogenesis (Crawford et al., 2004; Dumont et al., 2009; Romanov et al., 2001). Y-box binding protein-1 (YB-1) is a transcription/translation factor that is overexpressed in a plethora of cancers, including human breast carcinoma (40%) (Habibi et al., 2008; Wu et al., 2006). A protumourigenic role for YB-1 is supported by its ability to directly bind Y-box promoter elements of a variety of genes, notably epidermal growth factor receptor (EGFR), ErbB2/HER2, cyclin A, and cyclin B1 (Jurchott et al., 2003; Stratford et al., 2007; Wu et al., 2006). Moreover, a function for YB-1 in regulating cell cycle progression is beginning to emerge through its ability to alter the expression of genes involved at the G1/S boundary (Basaki et al., 2010; Yu et al., 2010). The transcriptional activity of YB-1 is dependent upon phosphorylation at its Ser-102 residue mediated by Akt/PKB and even more potently by p90 ribosomal S6 kinase (RSK) (Stratford et al., 2008; Sutherland et al., 2005). To establish the importance of YB-1 in malignant transformation, transgenic mice were developed where expression was targeted to the lactating mammary gland (Bergmann et al., 2005). The resulting mouse mammary tumours formed with 100% penetrance and close examination revealed substantial centrosome amplification and chromosomal instability (Bergmann et al., 2005). Given these findings concurrent with the high prevalence of YB-1 in patient tumours, we hypothesized that it plays an essential role in breast tumourigenesis. Genomic instability, in the form of alterations to chromosome number and structure, is a characteristic feature of almost all types of cancer (Holland and Cleveland, 2009; Nigg, 2002). However, whether this represents a cause or consequence of tumourigenesis remains mysterious. To address this contentious issue and begin to establish a paradigm for malignant transformation, it has become imperative to study cancer during the earliest premalignant stages, which, to date, have remained wholly uncharacterized. A large body of evidence has recently been compiled indicating that amplification of centrosomes has  59  the potential to cause mitotic defects that lead to chromosomal instability (Basto et al., 2008; Ganem et al., 2009; Nigg, 2006). According to this model, centrosome abnormalities would need to emerge early during neoplastic progression. Ultimately, through Darwinian selection, a karyotype adept at enhancing tumour progression would materialize and expand (Fujiwara et al., 2005; Shi and King, 2005). Of all sporadic breast cancer, 20 to 30% exhibit amplification of HER2 prompting us to address if this is a common feature of premalignancy that arises through targeted genomic instability preceding clonal outgrowth (Slamon et al., 1987). Herein we examined the role of YB-1 during premalignancy to uncover the molecular events that define the earliest transitions in breast cancer initiation and progression. A comprehensive understanding of these processes will usher the development of novel therapeutics that target the process, rather than the consequences, of tumourigenesis.  2.3 Results 2.3.1 YB-1 alters the expression and activity of cell cycle proteins To address the potential contribution of YB-1 in the initiation of tumourigenesis we engineered non-malignant H16N2 HMECs that conditionally expressed the gene under control of a tetracycline-inducible promoter (designated HTRY cells). HMECs containing an inducible LacZ construct served as a matched control (designated HTRZ cells). The ectopic expression level of YB-1 achieved in this model closely recapitulated that observed in established cancer cell lines (Figure S2.1A). Further characterization revealed that both cell lines were karyotypically normal (data shown below) and possessed similar telomerase activity (Figure S2.1B) deeming them genetically stable and thus amenable for investigating early transformation. To gain a global understanding into proteome remodeling following YB-1 induction we utilized the Kinexus Kinex™ antibody microarray platform, which allowed us to probe the expression and activation status (levels of phosphorylation) of over 600 proteins  60  concurrently. From this unbiased protein array, we identified 56 proteins with altered expression (Table S2.1) many of which are fundamental in regulating centrosome dynamics and the cell cycle (Table 2.1). Identification of known YB-1 transcriptional targets, including HER2, strongly supported the fidelity of the screen. We prioritized subsequent analysis on the active LIM kinases (pLIMK1/2T508/T505) as they exhibited the most profound increase in level (365%) following YB-1 induction. This correlation was validated both by immunoblotting (Figure 2.1A) and immunofluorescence staining (Figure 2.1B). In a reciprocal experiment, silencing YB-1 with siRNA in MDA-MB-231 and SUM149 breast cancer cell lines repressed the phosphorylation of LIMK1/2 at Thr508/Thr-505 with minimal effect on total LIMK1 expression (Figure 2.1C). To confirm these results, we utilized a complementary pharmacological approach for suppression of YB-1 activity using a signal transduction inhibitor to RSK. Our prior work indicated that inhibition of RSK directly impaired YB-1 phosphorylation and activity (Stratford et al., 2008). Treating MDA-MB-231 cells with the RSK inhibitor BI-D1870 yielded complete suppression of pLIMK1/2T508/T505 highlighting that pYB-1S102 was necessary to promote LIMK1/2 activation (Figure 2.1D). These data were mirrored using siRNA against RSK1 and RSK2 (Figure S2.2). Previous reports implicated LIMK1/2 as centrosomal proteins (Chakrabarti et al., 2007; Sumi et al., 2006) and, accordingly, we wanted to examine the localization in our HTRY cell model. At 96 hours post-YB-1 induction we observed punctate pLIMK1/2T508/T505 staining that included the centrosome as demonstrated by colocalization with the centrosomal marker γ-tubulin (Figure 2.1E).  61  Protein  Antibody  Localization  % CFC*  LIMK1/2  Y507+T508/Y504+T505  Centrosome  365  ZAP70  Y315+Y319  Centrosome  230  RSK1/2  S380/S386  Kinetochore  269  CDK1/2  Y15  Centrosome  161  Cdc34  Pan-specific  Mitotic spindle  148  Cofilin  Pan-specific  Centrosome  135  p53  Pan-specific  Cytosol/nucleus  118  CDK9  Pan-specific  Nucleus  116  PP2A  Pan-specific  Centrosome  102  ZAP70  Y292  Centrosome  102  CDK6  Pan-specific  Centrosome  98  PKA  Pan-specific  Centrosome  -67  Table 2.1 Cell cycle associated proteins putatively regulated by YB-1. *%CFC refers to the percent change from control (uninduced HTRY cells).  62  B"  HTRY%  HTRZ%  50%  YB#1% Ac(n%  siYB#1%#1%  Scr%  siYB#1%#1%  Control%  SUM149%  Scr%  kDa%  MDA#MB#231% Control%  C"  D" kDa%  10%μM%BI%  pLIMK1/2T508/T505%  1%μM%BI%  75%  DMSO%  LIMK1%  Control%  75%  37%  HTRY%  DAPI% pLIMK1/2T508/T505%  siYB#1%#2%  kDa%  HTRZ%  A"  50%  YB#1%  50%  YB#1%  50%  pYB#1S102%  75%  LIMK1%  75%  LIMK1%  75%  pLIMK1/2T508/505%  75%  pLIMK1/2T508/505%  50%  E"  Ac(n%  DAPI%  pLIMK1/2T508/T505%  γ#tubulin%  50%  Ac(n%  Composite%  Figure 2.1 YB-1 altered the activity of the centrosomal protein LIMK1/2. (a) Immunoblot analysis of YB-1 and LIMK1/2 in the HTRZ and HTRY cell lines. Actin served as a loading control. (b) Immunofluorescence staining of pLIMK1/2T508/T505 in HTRZ and HTRY cells. Nuclei were counterstained with DAPI. Scale bar = 20 μm. (c) Immunoblot following siRNA-mediated silencing of YB-1 (siYB-1) in MDA-MB-231 and SUM149 cell lines for 96 hours. Scrambled peptide (Scr) was a control. Actin was used as a loading control. (d) Immunoblot analysis assessing phosphorylation of YB-1 and LIMK1/2 in MDA-MB-231 cells treated with BI-D1870 (1 μM and 10 μM) for 24 hours. Actin served as a loading control. (e) Immunofluorescence analysis with antibodies targeting pLIMK1/2T508/T505 and γ-tubulin in HTRY cells. Co-localization is denoted with arrows. Nuclei were counterstained with DAPI. Scale bar = 20 μm.  63  2.3.2 Cytokinesis failure primes cells for premalignant transformation With YB-1 regulating a myriad of cell cycle associated genes, we wondered how cells from a non-malignant background would respond to expression of the gene. One of the earliest and most remarkable changes in the HTRY cells following YB-1 induction was the strikingly high incidence of multinucleated cells (Figure 2.2A). At 48 hours following YB-1 induction, which corresponded roughly to the doubling time of these cells (data not shown), 28% of HTRY cells were binucleate thus indicating a failure of cytokinesis. This was compared to only 5% of HTRZ cells (Figure 2.2B). Previous work has demonstrated that LIMK1 localizes to the cleavage furrow during cytokinesis (Sumi et al., 2006) and its overexpression is associated with polyploidy (Amano et al., 2002). Based on its role in modulating actin dynamics (Bernard, 2007), a likely explanation for the defect in cytokinesis might therefore be that deregulation of LIMK1/2 perturbs the stability of the contractile ring. In cytokinetic HTRZ cells, pLIMK1/2T508/T505 was concentrated in the junction between the two daughter cells as observed by immunofluorescence. Accordingly, F-actin was visualized, using phalloidin, along the cleavage furrow and at the nuclear periphery (Figure 2.2C). On the other hand, in cytokinetic HTRY cells, pLIMK1/2T508/T505 was strongly expressed but remained diffuse throughout the cytoplasm (Figure 2.2C). Consequently, the actin cytoskeleton failed to reposition itself for cytokinesis. Another well-established protein at the cleavage furrow, polo-like kinase 1 (PLK1), correctly localized in HTRY cells during telophase; however, the actomyosin contractile ring consistently failed to form in these cells (Figure S2.3). This was in contrast to HTRZ cells, which formed a tight contractile ring with PLK1 characteristically localized to the midzone between the dividing cells. This supports our hypothesis that YB-1 promotes cytokinesis failure through deregulation and subsequent mislocalization of LIMK1/2.  64  HTRZ"  HTRY"  DAPI" αItubulin"  B" Percent"of"binucleate"cells"  A"  35"  **"  30" 25" 20" 15" 10" 5" 0" HTRZ"  C"  Phalloidin"  pLIMK1/2T508/505"  Composite"  HTRY"  HTRZ"  DAPI"  HTRY"  Figure 2.2 Premalignancy was initiated by pLIMK1/2T508/T505 mislocalization leading to cytokinesis failure. (a) HTRZ and HTRY cells were induced with doxycycline for 48 hours and immunostained with α-tubulin antibody. Nuclei were visualized using DAPI. Arrows indicate binucleate cells. Scale bar = 20 μm. (b) Quantification of binucleate HTRZ and HTRY cells. 200 cells were assessed in three independent experiments. (c) Immunofluorescence staining with phalloidin and antibody targeting pLIMK1/2T508/T505 in cytokinetic HTRZ and HTRY cells following a 48-hour induction with doxycycline. The actomyosin contractile ring is identified with arrows. Scale bar = 20 μm.  65  2.3.3 Cell cycle checkpoint slippage potentiates centrosome amplification leading to aneuploidy Cytokinesis failure can lead to both centrosome amplification and production of tetraploid cells, which could set the stage for the development of tumour cells (Fujiwara et al., 2005). We examined whether YB-1 expression allowed for cells arrested in cytokinesis to slip through cell cycle checkpoints and re-enter mitosis with multiple centrosomes and a compromised genome. At 96 hours post-YB-1 induction, HTRY cells demonstrated centrosome amplification leading to multipolar spindle formation in mitosis (Figure 2.3A). Consequently, kinetochore bi-orientation was not established resulting in failed segregation, which manifested as lagging chromosomes at the metaphase plate and micronuclei (Figure 2.3A). Quantifying the proportion of tetraploid cells (>4N DNA content) and those with supernumerary centrosomes (>2 centrosomes) revealed a significant increase of 4.5-fold and 12.2-fold, respectively, between the HTRY and HTRZ cells (Figure 2.3B). Deeper interrogation uncovered that the amplified centrosomes contained an excess of mother, but not daughter, centrioles (Figure S2.4A). Specifically, the 1:1 mother:daughter centriole ratio observed in the HTRZ cells approached 3:1 in the HTRY cells (Figure S2.4B). Having established the importance of pYB-1S102 at the centrosome it is not surprising that the described phenotype was contingent upon YB-1 Ser-102 phosphorylation as transient expression of YB-1S102D in HTRZ cells could recapitulate the phenotype whereas YB-1S102A could not (Figure S2.5A). Moreover, the incidence of aneuploidy and centrosome amplification was most profound in YB-1S102D overexpressing MDA-MB-231 cells by a significant margin (Figure S2.5B). Due to HPV-16 E6/E7 immortalization, the p53 and retinoblastoma (RB) tumour suppressor genes are inactivated in the HTRZ and HTRY cells. To ascertain if this background was necessary to generate the abnormal phenotypes observed in HTRY cells, we transfected YB-1 into the 184hTERT cell line. The parental cells, which express no YB-1, have been extensively characterized to be chromosomally stable with a nearly normal karyotype (Raouf et al., 2005). Transient expression of YB-1 for 96-hours was sufficient to promote polyploidy and centrosome amplification (Figure 2.3C). We also  66  observed early indicators of genomic instability, including lagging chromosomes and micronuclei (Figure 2.3C). The faithful recapitulation of phenotypes observed between the HTRY and 184hTERT:YB-1 cells indicates that YB-1 expression alone is sufficient to drive genomic instability without the requirement of secondary gene deregulation. One would expect that given the centrosome amplification coupled with aneuploid DNA content that the YB-1 induced cells would be subject to cell cycle arrest. To our surprise, only 18% of HTRY cells were classified as being in G1-phase of the cell cycle based on DNA content. This was in stark contrast to 66% of HTRZ cells (Figure 2.3D), strongly supporting the notion that HTRY cells resist anti-proliferative signals and slip through the G1/S checkpoint. Consistent with our findings of a cytokinesis defect, 62% of HTRY cells were in G2/M-phase compared to 14% of HTRZ cells (Figure 2.3D). To differentiate between cells arrested at the G2 checkpoint and those that have progressed into M-phase we quantified pHistone H3S10 by immunofluorescence. In agreement with increased DNA content, 23% of HTRY cells were positive for pHistone H3S10, thus truly in mitosis, compared to 3% of HTRZ cells (Figure 2.3E). This implies that overcoming the cytokinesis defect may be a rate-limiting step in tumourigenesis. We next assessed changes in signal transduction to define a mechanism to explain the observed slippage through the G1/S checkpoint. Following 96 hours of YB-1 induction, strong expression of HER2 was detected correlative with RSK activation. Accordingly, p27Kip1, an inhibitory target of RSK and negative regulator of cyclin E/CDK2, was suppressed (Figure 2.3F). These data indicate that YB-1 deregulates the cell cycle by enhancing the MAP kinase signal transduction pathway to favour a proliferative program.  67  Amplified"" centrosomes"  MulMpolar"" spindles"  DAPI" α<tubulin" γ<tubulin"  DAPI" Pericentrin"  B"  Lagging"" chromosomes"  DAPI"  Micronuclei"  40"  HTRZ" HTRY"  **"  Percent"of"cells"  A"  30"  **"  20" 10"  DAPI"  0" >4N"DNA"  184"hTERT<EV" 184"hTERT<YB<1" *"  5"  YB<1" AcMn"  0" >4N"DNA"  Centrosome" >2"  DAPI" FLAG:YB<1" Pericentrin"  80" 60" 40" 20"  ***"  20" 15"  G0/G1"  S"  G2/M"  150"  10" 5" 0"  0"  kDa" 250"  ***"  25"  HTRZ"  HTRZ" HTRY" **"  DAPI" FLAG:YB<1"  F"  HTRY"  Percent"of"cells"  100"  Micronuclei"  DAPI" FLAG:YB<1"  E" Percent"pH3+"cells"  D"  Lagging"" chromosomes"  HTRZ"  HTRY"  HTRY"  10"  Amplified"" centrosomes"  HTRZ"  **"  Centrosome" >2"  YB<1"  Percent"of"cells"  15"  EV"  C"  pH3S10" DAPI" "  HER2" EGFR"  100"  RSK1/2/3"  100"  pRSKS380"  25" 50" 37"  p27Kip1" Cyclin"E" CDK2"  50"  pYB<1S102"  50"  YB<1" AcMn"  37"  Figure 2.3 Centrosome amplification and aneuploidy emerged as a consequence of cytokinesis failure and cell cycle checkpoint slippage. (a) Immunofluorescence staining of 96-hour induced HTRY cells with antibodies targeting microtubules (α-tubulin) and centrosomes (pericentrin and γ-tubulin). Nuclei were counterstained with DAPI. Lagging chromosomes and micronuclei are demarcated with arrows. Scale bar = 10 μm. (b) Quantification of polyploidy and centrosome amplification in 500 HTRZ and HTRY cells following a 96-hour induction. DNA content, which is proportional to the total Hoechst 33342 intensity per nucleus, was measured using an ArrayScan VTI while the number of centrosomes per cell was counted manually. The data represent a compilation of three independent experiments. (c) Quantification of polyploidy and centrosome amplification in 184hTERT cells transiently  68  transfected with FLAG-tagged YB-1 (confirmed by immunoblotting). Data was acquired from six random microscope fields and is presented as the mean and standard deviation from three independent experiments. Representative images depict genomic instability. Scale bar = 10 μm. (d) Cell cycle profile of asynchronous populations of HTRY and HTRZ cells measured by DNA content and nuclear morphology using an ArrayScan VTI. Variation between three separate experiments is indicated. (e) Quantification of HTRZ and HTRY cells positive for nuclear pHistone H3S10. Data was collected from five unique microscope fields and is presented as the mean and standard deviation from three independent experiments. Representative images are shown. Scale bar = 100 μm. (f) Immunoblot analysis of cell cycle regulatory proteins in induced HTRY cells.  69  2.3.4 Identification of YB-1 as a centrosomal protein The data described above establishes the importance of YB-1 in regulating centrosomal proteins with a role in promoting amplification of the organelle during premalignancy. On this basis, we next explored whether YB-1 was itself directly associated with the centrosome. Co-localization between pYB-1S102 and pericentrin, a centrosomal marker, was observed in both interphase and mitotic HTRY cells using immunofluorescence (Figure 2.4A). Notably, during metaphase, pYB-1S102 was expressed along the entire length of the mitotic spindle. In MDA-MB-231 cells, pYB-1S102 was found to co-localize with pericentrin confirming its association with centrosomes was not unique to our inducible system but rather extended to established cancer cell lines (Figure S2.6A). To validate YB-1 as a bona fide centrosomal protein we mapped both FLAG:YB-1 (Figure S2.6B) and GFP:YB-1 (Figure S2.6C) to the centrosome using antibody directed against the FLAG epitope and direct immunofluorescence, respectively. To ascertain whether phosphorylation was a prerequisite for YB-1 centrosomal localization, we generated MDA-MB-231 cells stably expressing FLAG:YB-1WT, FLAG:YB-1S102D, and FLAG:YB-1S102A protein (Figure 2.4B). Double immunofluorescence using anti-FLAG and anti-pericentrin antibodies revealed that the centrosomal localization of YB-1 was contingent upon phosphorylation at the Ser-102 residue. FLAG:YB-1S102D protein, which mimicked constitutively phosphorylated YB-1, was detected in 93.5% of centrosomes. In contrast, the non-phosphorylatable FLAG:YB1S102A protein failed to localize to the centrosomes (Figures 2.4C and 2.4D). Collectively, these data provide insight into the dependence on phosphorylation for the trafficking, retention, and/or function of YB-1 at the centrosome.  70  A!  pYB)1S102%  Pericentrin%  Composite%  Metaphase% Anaphase%  100%  Vinculin%  FLAG%  Pericentrin%  Composite%  YB)1WT%  ***%  YB)1S102D%  40% 20% 0%  DAPI%  YB)1S102A%  50%  FLAG%  60%  YB)1)S102A%  pYB)1S102%  80%  YB)1)S102D%  50%  D!  ***%  YB)1)WT%  YB)1%  100%  EV%  50%  Centrosome%co)localiza=on%(%)%  kDa%  C!  FLAG:YB)1S102D%  B!  EV% FLAG:YB)1WT% FLAG:YB)1S102A%  Telophase%  Mito=c%  Prophase%  Interphase%  DAPI%  Figure 2.4 pYB-1S102 localized to the centrosomes throughout the cell cycle. (a) Immunofluorescence with specific antibodies targeting pYB-1S102 and pericentrin in HTRY cells. Arrows denote centrosomal YB-1. DAPI marked the nuclei. Scale bar = 20 μm. (b) Immunoblotting confirmed ectopic expression of FLAG-tagged YB-1WT, YB1S102D, and YB-1S102A in stable MDA-MB-231 cells. (c) Quantification of FLAG and pericentrin co-localization in 250 interphase cells transiently transfected with FLAGtagged YB-1 (n = 3). (d) Immunofluorescence of representative MDA-MB-231 cells expressing FLAG-tagged YB-1. Arrows denote centrosomal YB-1. Scale bar = 10 μm.  71  To functionally examine the role of YB-1 at the centrosome we began by performing coimmunoprecipitation experiments, which revealed physical association between FLAG:YB-1 and the centrosomal proteins pericentrin and γ-tubulin (Figure 2.5A). We further queried whether LIMK1 was part of a YB-1 centrosomal complex; however, the two proteins failed to co-precipitate (data not shown). Next, we silenced YB-1 in MDAMB-231 cells using two independent siRNA sequences (siYB-1#1 and siYB-1#2), which in turn, was found to yield substantial enlargement and morphological changes of the centrosomes as visualized by immunofluorescence targeting pericentrin (Figure 2.5B). Further investigation uncovered that these centrosomes were enriched in γ-tubulin, a major component of the gamma-tubulin ring complex (Figure 2.5B). We quantified a 3.3 to 4.1-fold increase in centrosome area at 96 hours post-siYB-1 transfection relative to mock-transfected cells (Figure 2.5C). Importantly, because Ser-102 phosphorylation was a necessity for YB-1 to localize at the centrosome, we wanted to assess if the mere inhibition of protein activity would be sufficient to alter centrosome structure. In agreement with the siRNA experiments, treating cells with 1 μM or 10 μM BI-D1870 for 24 hours prompted the emergence of cells containing enlarged centrosomes with numerous γ-tubulin punctae (Figure S2.7A). Specifically, the centrosomes increased in area by up to 2.9-fold relative to the DMSO-treated controls (Figure S2.7B). In further support, we observed increased centrosomal area in MDA-MB-231 cells stably expressing YB-1S102A mutant protein (Figure S2.7C). Finally, to analyze if the changes to centrosome structure correlated with altered function we performed microtubuleregrowth assays to detect defects in centrosome-mediated microtubule nucleation and anchoring. The assay was used to assess a fundamental parameter of centrosome function, that is, the ability to regrow microtubules following depolymerization. Displacement of YB-1 from the centrosome following siRNA silencing clearly delayed microtubule-regrowth as asters of short microtubules only began to emerge after a fiveminute regrowth as opposed to one minute in mock-transfected cells (Figure 2.5D). This clearly demonstrates that loss of YB-1 perturbs normal centrosome function. Taken together, these results provide strong evidence for a previously uncharacterized, yet essential, role for YB-1 at the centrosome.  72  Flag:YB#1%  50% 250%  Pericentrin%  *%  3% 2%  Scr% siYB#1%#2%  Time%aEer%nocodazole%release%(min)% 0%  *%  Control% Scr% siYB#1%#1%  γ#tubulin% DAPI%  1%  5%  10%  Scrambled%  4%  YB#1% Ac(n%  50%  D"  5%  siYB#1%#2%  Pericentrin% DAPI%  γ#tubulin%  50%  si Y  ro l nt Co  Sc r% B# 1% # siY 1% B# 1% #2 %  0%  siYB#1%#1%  1%  %  Fold#change%in%centrosome%area%  siYB#1%#1%  γ#tubulin%  Pericentrin%  γ#tubulin%  FLAG:YB#1%  IgG%  kDa%  Input%  Immunoprecipita(on%  C"  Scrambled%  B" Pericentrin%  A"  DAPI% α#tubulin% γ#tubulin%  Figure 2.5 YB-1 altered the architecture and microtubule nucleation capacity of centrosomes by directly binding pericentrin and γ-tubulin. (a) Immunoblot showing co-immunoprecipitation with FLAG, γ-tubulin, and pericentrin antibodies in MDA-MB-231 cells stably expressing FLAG-tagged YB-1. (b) MDA-MB231 cells transfected with siRNA oligos targeting YB-1 (siYB-1#1 and siYB-1#2) were analyzed after 96 hours by immunofluorescence staining with pericentrin (upper panel) and γ-tubulin (lower panel). Immunoblotting confirmed YB-1 knockdown. Scale bar = 10 μm. (c) Centrosome area was measured using Image Pro Analyzer software and is represented relative to the non-transfected control. 100 centrosomes from G1-phase cells were measured across three independent experiments. (d) Microtubule regrowth in scrambled and siYB-1 transfected MDA-MB-231 cells. Microtubules were visualized with antibody targeting α-tubulin at 1, 5, or 10 minutes following nocodazole release. Centrosomes were visualized by γ-tubulin. Nuclei were counterstained with DAPI. Scale bar = 20 μm.  73  2.3.5 Genomic instability arises during premalignancy to generate clones with strong tumourigenic potential To better understand the aneuploidy and chromosomal instability that emerge as a consequence of centrosome amplification during premalignancy, we assessed metaphase chromosomes. The majority of uninduced HTRZ and HTRY cells, as well as induced HTRZ cells, had a normal diploid karyotype. A small subset were classified as “near diploid” (40 to 52 chromosomes; Figure 2.6A). In stark contrast, 94% of induced HTRY cells were aneuploid (Figure 2.6A). Further to these numerical abnormalities, structural chromosome aberrations were readily detected in the induced HTRY cells. Dramatic increases of ≥5.5-fold in the appearance of dicentric chromosomes and double minutes were observed in the HTRY spreads compared to those from HTRZ cells. Most strikingly, there was a 17.6-fold increase in acentric pairs and 13.2-fold increase in acentric fragments between HTRY and HTRZ spreads indicating that YB-1 promoted extensive chromosome breakage (Figure 2.6B). In addition, defective sister chromatid cohesion was commonly detected in the HTRY cells. Almost half of these cells exhibited a lack of primary constriction, identified by primary constriction gaps (PCGs) between sister chromatids at metaphase, compared to 11% of HTRZ cells (Figure S2.8). To address whether genomic instability initiated by YB-1 could promote an optimal karyotypic composition for tumourigenesis, the frequency of HER2 amplification was measured using fluorescence in situ hybridization (FISH). We uncovered that 11% of HTRY cells were positive for HER2 amplification (HER2:CEP17 >2.2). No HTRZ cells exhibited HER2 amplification (Figure 2.6C). Upon a more rigorous assessment we noted that 20% of the HTRY cell population contained low-level HER2 amplification (HER2:CEP17 between 1.5 and 2.2; Figure 2.6C). The HER2 amplification in HTRY cells was largely undetected until they reached tetraploid DNA content. At this time there was an apparent relaxation in the mechanisms safeguarding genomic stability and the number of HER2 signals began to exceed the number of centromeres indicative of gene amplification (Figure S2.9). Collectively, these data convey that a subset of premalignant cells have an amplification at the HER2 locus that could enhance their tumourigenic potential.  74  100"  ***"  ***"  80" 60" 40"  *"  20"  TDOX" HTRZ" HTRY" HTRZ"+DOX" HTRY"+DOX"  HTRZ"  HTRZ"  HTRY"  *"  0"  C! 30" 20" 10"  Percent"of"cells"  Severe" Moderate" Mild"  40"  0" HTRZ"HTRY"HTRZ"HTRY"HTRZ"HTRY"HTRZ"HTRY" Double" minute""  80" 60" 40" 20" 0"  In  Acentric" Acentric" Dicentric" pairs" fragments"  HTRZ" HTRY"  100"  50"  HTRY"  B!  HTRZ"  Near" Aneuploid" diploid"  ve te " rm ed ia te " Po siJ ve "  Diploid"  Metaphases"(%)"  +DOX" HTRY"  Ne ga J  Percent"of"metaphases"  A!  DAPI" CEP17" HER2"  HER2"amplificaJon"  Figure 2.6 Numerical and structural chromosomal aberrations materialized as a consequence of YB-1 expression. (a) Quantification of chromosomes in metaphase spreads prepared from HTRZ and HTRY cells following a 96-hour induction with doxycycline (+DOX). Uninduced cells served as controls (-DOX). Three independent experiments were preformed and representative images are shown. Scale bar = 100 μm. (b) Structural chromosomal abnormalities evaluated in 200 metaphase spreads from 96-hour induced HTRZ and HTRY cells. Three independent experiments were performed. Representative images are shown. Scale bar = 10 μm. (c) FISH analysis with labeled DNA probes to the pericentromeric region of chromosome 17 (CEP17) and to the HER2 locus (17q21.1). Following a 96-hour induction, 100 interphase cells were assessed for low-copy (HER2:CEP17 = 1.5 – 2.2) and high-copy (HER2:CEP17 > 2.2) gene amplification. Representative images are shown. Scale bar = 10 μm.  75  We conclude from our data that YB-1 activates a tumourigenic program that manifests as a cytokinesis defect and progresses toward the emergence of fully transformed cells. From this, we have proposed a model of premalignant progression (Figure 2.7).  LIMK%mislocaliza7on%% leading%to%cytokinesis% failure%  YB#1% expression% HMEC% Cyclins!  Post#mito7c%and%G1/S% cell%cycle%checkpoint% slippage%  Mul7#polar%spindles% leading%to%aneuploidy%  Clonal%selec7on%and%% expansion%of%HER2+% cells%  Figure 2.7 Proposed model for how YB-1 instigates premalignancy. Forced YB-1 expression in non-tumourigenic HMECs prompted a strong enhancement in LIMK1/2 activity that resulted in a cytokinesis defect. Concurrently, YB-1 altered signal transduction allowing cells to slip through the G1/S checkpoint. The resulting centrosome amplification led to multipolar spindles during mitosis, which promoted aneuploidy. With sustained YB-1 expression, a small population of cells containing HER2 amplification emerged.  76  2.4 Discussion In the present study we propose a distinctive model of breast cancer premalignancy whereby YB-1 enables the evolution of human mammary epithelial cells toward a tumourigenic fate. A cytokinesis defect acted as the initial trigger for transformation promoting centrosome amplification and aneuploidy, which were potentiated by cell cycle checkpoint slippage. In turn, we identified a small population of cells harbouring amplification at the HER2 locus. These studies provide significant insight into the process of tumour initiation and demonstrate how YB-1 alone can initiate a program that primes cells for tumourigenesis. Although YB-1 upregulation is well-characterized in breast cancer cell lines and advanced stage primary tumours (Habibi et al., 2008; Janz et al., 2002; Kohno et al., 2003; To et al., 2010), a role for the gene in tumour initiation and premalignant progression is unknown. Chromosomal aberrations observed in YB-1 transgenic mice prompted us to address if ectopic YB-1 expression in genetically stable HMECs acted to directly destabilize the genome as a prelude to malignancy (Bergmann et al., 2005). Here we have demonstrated that expression of the gene promoted gross alterations to the centrosomal milieu and, ultimately, led to centrosome amplification. Most notably, a strong activation of LIMK1/2 was detected at the centrosomes. Likewise, in prostate cancer LIMK is expressed at the centrosome and has been linked to chromosomal instability and metastasis (Chakrabarti et al., 2007; Davila et al., 2007; Yoshioka et al., 2003). An important finding of this study was that cytokinesis failure is the predominant mechanism for the amplification of centrosomes during premalignancy. We identified that sustained upregulation and mislocalization of active LIMK1/2 by YB-1 was sufficient to induce a cytokinesis defect. This could be attributed to enhanced actin polymerization at the cleavage furrow (Yang et al., 2004b). Given these results we propose that YB-1 causes early changes in cytokinesis and centrosomal architecture that lead to eventual chromosomal instability. In this work we have established YB-1 as a centrosomal protein. This was found to be contingent upon phosphorylation of the Ser-102 residue in the cold shock domain (CSD) implying that this domain is minimally required for centrosomal trafficking. Especially 77  interesting is the fact that the CSD is necessary for binding oligonucleotides, including RNA due to the presence of two RNP motifs (Kohno et al., 2003). As the centrosome contains an intrinsic complement of RNA (Alliegro et al., 2006) it is possible that YB-1 is involved in regulating their translation. YB-1 has already been shown to induce capindependent translation of RNA giving credence to this hypothesis (Evdokimova et al., 2009). A second function for YB-1 at the centrosome may be to mediate protein stability via physical association. It has been demonstrated that YB-1 interacts with a myriad of proteins, including PCNA, MSH2, and DNA polymerase δ, via B/A repeats residing in the C-terminal domain (Gaudreault et al., 2004; Ise et al., 1999). It is tempting to speculate that centrosomal proteins may represent an underappreciated pool with strong capacity to promote tumourigenesis by their inherent ability to directly interface with the genome. In support of this, deregulation of centrosomal proteins including Aurora A (Wang et al., 2006), BRCA1 (Scully, 2000), and PLK1 (Takai et al., 2005) all promote genomic instability with eventual cellular transformation. Studies of human mammary epithelial cells have proven effective in providing key insights into the early genetic events that fuel breast carcinogenesis (Elenbaas et al., 2001; Romanov et al., 2001; Tlsty et al., 2001). Here we report that YB-1 expression in this model leads to catastrophic genetic changes that if left unchecked could allow for the replication of cells containing vast chromosomal amplifications and rearrangements. Because YB-1 expressing HTRY cells fail to arrest following genomic destabilization it suggests that mutant cells are able to escape the necessary checkpoints needed to eliminate such renegade cells. Permissiveness through the cell cycle could relate to direct YB-1 transcriptional targets, such as CCNB1 (Jurchott et al., 2003), CDC6 (Basaki et al., 2010), PCNA (Ise et al., 1999), and TOPO2 (Shibao et al., 1999). We chose to address the possibility that YB-1 permits the expansion of cells harbouring specific amplifications common to breast cancer. Notably, we describe HER2 as being amplified in a small subset of HTRY cells. We speculate that over time this population of cells would clonally expand due to the distinct survival advantage brought about by HER2 overexpression. Moreover, this study furthers our understanding of the relationship between YB-1 and HER2 as previously described by our laboratory (Dhillon et al., 2010;  78  Wu et al., 2006). Our previous studies show that YB-1 directly binds to the HER2 promoter in cells where the gene is known to be amplified (Wu et al., 2006). One could envisage that YB-1 is permissive for allowing cells with HER2 amplification to slip through the cell cycle checkpoints. Following this YB-1 is poised to increase the expression of HER2 by binding directly to its promoter. This too may explain why YB-1 and HER2 are highly expressed in ~65% of primary breast tumours (Habibi et al., 2008). Future work will focus on identifying additional genomic rearrangements that frequently materialize during premalignancy. We report that increased expression of YB-1 as a single event is sufficient to uncouple genomic integrity and cell cycle progression during breast cancer premalignancy. In summary, our findings argue that YB-1 plays a principal role in the early evolution of cancer and thus represents a promising biomarker and therapeutic target.  2.5 Experimental Procedures Cell culture and drug treatments. H16N2 human mammary epithelial cells with tetracycline-inducible YB-1 (HTRY) or LacZ (HTRZ) were generated using the T-REx system as previously described (Band et al., 1990; Berquin et al., 2005). Briefly, cells expressing the pcDNA6/TR regulatory vector were infected with YB-1 cDNA and plated at clonal density. Colonies were isolated by ring cloning and YB-1 expression was confirmed by RT-PCR. The cells were cultured in Ham’s F12 media and induction was achieved through the addition of 1 μg/mL doxycycline (Calbiochem, Gibbstown, NJ, USA). The human mammary epithelial 184hTERT cell line (a gift from Dr. J. Carl Barrett, National Institutes of Health, Bethesda, MD) was cultured in supplemented HuMEC media (Invitrogen, Burlington, ON, Canada). LCC6, MDA-MB-231 (American Tissue Culture Collection, Manassas, VA, USA), and SUM149 (Asterand, Detroit, MI, USA) breast cancer cell lines were cultured as recommended.  79  For treatment with BI-D1870, MDA-MB-231 cells were seeded at a density of 3 x 105 cells in a six-well plate. Subsequently, cells were treated with a DMSO vehicle or BID1870 (1 μM or 10 μM) for 24 hours. Kinexus Kinex™ antibody microarray. HTRY cells were induced for 96 hours. Comparisons were made to cells not treated with doxycycline. Protein was sent to Kinexus Bioinformatics Corporation (Vancouver, BC, Canada) for hybridization and analysis using the Kinex™KAM-1.1 antibody microarray. Immunoblotting, immunoprecipitation, and immunofluorescence. Immunoblotting, immunoprecipitation, and immunofluorescence were performed as described previously (Finkbeiner et al., 2009; Stratford et al., 2008; Wu et al., 2006). Egg lysis buffer (ELB) was used to isolate protein for immunoblotting. YB-1 (50 kDa) and actin (42 kDa) were probed concomitantly on the same blot. For immunoprecipitation, cells were lysed using radioimmunoprecipitation lysis buffer (RIPA) and 500 μg of cell lysate was pre-cleared with 35 μl of protein G agarose (Sigma, St. Louis, MO, USA) prior to overnight antibody incubation. The proteins were retrieved through the addition of protein G agarose for 2 hours and eluted in 5x SDS-sample loading buffer heated to 100°C for 5 minutes. For immunofluorescence staining, antibodies were diluted in ICC buffer (10% BSA, 2% goat serum, 1% saponin in PBS), and all incubations were carried out at room temperature for one hour with three washes in PBS following each of the incubations. Cells were mounted using ProLong Gold antifade reagent containing DAPI (Invitrogen). The origin and dilutions of all antibodies used in this study are detailed in Table S2.2. Images were acquired using an Olympus FV1000 laser scanning confocal microscope, a DeltaVision personalDV live cell imaging microscope, or an Olympus BX61 epifluorescence microscope and analyzed with ImageJ 1.43 (National Institutes of Health). siRNA and plasmid transfections. Cells were transfected with 20 nM of siRNA to RSK1, RSK2, YB-1, or scrambled control using Lipofectamine RNAiMAX (Invitrogen). The siRNA target sequences are provided in Table S2.3. The FLAG-tagged YB-1WT, YB1S102A, and YB-1S102D constructs have previously been described (Finkbeiner et al., 2009; Sutherland et al., 2005). Plasmid transfections were performed using 4 μg of DNA and  80  carried out with Lipofectamine 2000 (Invitrogen). Stable transfectants were generated and selected for using G418 (400 μg/mL). The GFP:YB-1 construct (Guay et al., 2006), was transfected into MDA-MB-231 cells by electroporation with Amaxa Nucleofactor Kit V using the manufacturer’s recommendations (Lonza, Walkersville, MD, USA). Microtubule regrowth assay. siRNA transfected MDA-MB-231 cells were treated with 5 μM nocodazole for 1 hour to depolymerize all microtubules. Nocodazole was then removed by washing cells twice with DMEM. At 1, 5, and 10 minutes after regrowth the cells were fixed with 100% methanol and stained with α-tubulin and γ-tubulin antibodies. Cell cycle analysis. HTRZ and HTRY cells were seeded into 96 well plates and induced for 96 hours. The cells were subsequently fixed using 2% paraformaldehyde and stained with Hoechst 33342 (1μg/ml; Sigma) for 30 minutes. Based on nuclear morphology and total nuclear Hoechst intensity the proportion of cells in each stage of the cell cycle was analyzed by Cell Cycle Bioapplication software on a Cellomics high content screening instrument (ArrayScan®, Thermo Fisher Scientific). Chromosome spreads. Following a 96-hour induction, HTRZ and HTRY cells were treated with 0.1 μg/mL colcemid (Invitrogen) for 2 hours. Mitotic chromosomes were resuspended in hypotonic solution (75 mM KCl) for 20 minutes and fixed using methanol:glacial acetic acid (3:1), as previously described (Barber et al., 2008). Metaphase chromosomes were imaged using a Zeiss Axioplan digital imaging microscope and analyzed with Metamorph imaging software (Universal Imaging Corp, Downingtown, PA, USA). For analysis, we assessed chromosomal abnormalities based on their incidence with “mild” referring to less than five occurrences in a spread, “moderate” between five and twenty, and “severe” greater than twenty. Primary constriction gaps (PCGs) were defined as a clear separation between DAPI stained sister chromatids. The severity ranged from only one or two chromosomes in a spread exhibiting a gap (PCGI), to between three and ten chromosomes (PCGII), to no semblance of cohesion (PCGIII). HER2 FISH. Asynchronous HTRZ and HTRY cells were prepared for chromosome analysis as described above. Interphase cells were hybridized with LSI HER2 and CEP17 81  probe using the PathVysion HER2 DNA Probe Kit at the Centre for Translational and Applied Genomics (Vancouver, BC, Canada). Analysis of FISH signals was performed in 100 randomly selected cells. HER2 amplification was defined as a HER2:CEP17 ratio greater than 2.2. A HER2:CEP17 ratio <1.5 was considered negative for HER2 amplification while a ratio near the cut-off (1.5 - 2.2) was interpreted as intermediate amplification. Telomerase assay. The telomerase activity in 1 μg of cell lysate from HTRZ and HTRY cells was measured using the Quantitative Telomerase Detection Kit (Allied Biotech, Vallejo, CA, USA) following the manufacturer’s instructions. Each sample was analyzed in triplicate. A no template control and cell lysate from telomerase positive cells (MDAMB-231) were included in each experiment. Statistical analysis. Data from at least three independent experiments are reported as mean ± standard deviation. Significance was examined using a paired Student's t-test, where *P <0.05, **P <0.01, and ***P <0.001.  82  2.6 Supplementary Data  50%  pYB#1S102%  50%  YB#1%  37%  Telomerase%ac(vity% (molecules/rxn)%  B!  LCC6%  HTRY%  +DOX%  HTRZ%  HTRY%  kDa%  HTRZ%  #DOX%  SUM149%  A!  MDA#MB#231%  2.6.1 Supplementary figures  4000% 3000% 2000% 1000%  Ac(n%  0% HTRZ%  HTRY%  Figure S2.1 Characterization of the HTRZ and HTRY cell lines. (a) Immunoblot assessing YB-1 expression and activation in uninduced (-DOX) and 96hour induced (+DOX) HTRZ and HTRY cells, in addition to the SUM149, MDA-MB231, and LCC6 cell lines. Actin served as a loading control. (b) Telomerase activity assay depicting the amount of telomerase activity (molecules/reaction) in 1 μg of lysate prepared from 96-hour induced HTRZ and HTRY cells.  83  siRSK1+2%  siRSK2%  siRSK1%  Scr%  kDa% 100%  RSK1%  100%  RSK2%  50%  YB#1%  50%  pYB#1S102%  75%  LIMK1%  75%  pLIMK1/2T508/505%  100%  Vinculin%  Figure S2.2 Knockdown of RSK1/2 suppressed the phosphorylation of YB-1 and LIMK1/2. Immunoblot analysis of RSK, YB-1, and LIMK expression and activation in MDA-MB231 cells following 96-hour siRNA silencing of RSK1, RSK2, or RSK1/2. Vinculin served as a loading control.  84  Phalloidin!  PLK1!  Composite!  HTRY!  HTRZ!  DAPI!  Figure S2.3 PLK1 localized to the cleavage furrow in HTRZ and HTRY cells. Immunofluorescence staining with antibody targeting PLK1 in cytokinetic HTRZ and HTRY cells. The actomyosin contractile ring can be visualized using phalloidin. Nuclei were counterstained with DAPI. Scale bar = 20 μm.  85  HTRY%  x60%  B" 4!  Number%of%centrioles/cell%  HTRZ%  x40%  A"  Cenexin% CNTROB% DAPI%  3! 2! 1! 0! HTRZ%  HTRY%  Mother!  HTRZ%  HTRY%  Daughter!  Figure S2.4 Amplification of the mother centriole was a consequence of YB-1 expression. (a) HTRZ and HTRY cells were induced for 96 hours and subsequently analyzed by immunofluorescence using antibodies specific for mother (cenexin) and daughter (CNTROB) centrioles. Scale bar = 20 μm (x40) or 10 μm (x60). (b) Quantification of mother and daughter centrioles in 50 cells from random fields. Data is presented as the mean and standard deviation of three independent experiments.  86  B" 12" 10"  Percent"of"cells"  EV" HTRZ<WT<YB<1" HTRZ<S102D<YB<1" HTRZ<S102A<YB<1"  **"  8" 6" 4"  **" **"  **"  25"  ***" Percent"of"cells"  A"  20" 15"  **" *"  10"  **"  5"  2"  EV" MDA<MB<231<WT<YB<1" MDA<MB<231<S102D<YB<1" MDA<MB<231<S102A<YB<1"  0"  0" Mul)nucleated" Centrosome">2"  Mul)nucleated"  Centrosome">2"  Figure S2.5 Aneuploidy and centrosome amplification were dependent on YB-1 Ser102 phosphorylation. (a) Quantification of multinucleation and centrosome amplification in HTRZ cells transiently transfected with plasmid encoding wild-type (WT-YB-1) or mutant (S102DYB-1 and S102A-YB-1) YB-1. 500 cells were assessed across three independent experiments. (b) MDA-MB-231 cells were transfected with plasmid encoding YB-1 wild-type or mutant DNA and stable clones were generated. The extent of polyploidy and centrosome amplification was assessed in 500 cells from three unique experiments.  87  A"  B"  C"  DAPI*  pYB(1S102*  Pericentrin*  Composite*  DAPI*  FLAG:YB(1*  Pericentrin*  Composite*  DAPI*  GFP:YB(1*  Pericentrin*  Composite*  Figure S2.6 Validation of YB-1 as a centrosomal protein. (a) Immunofluorescence staining in MDA-MB-231 cells with antibodies targeting pYB1S102 and pericentrin. (b) Immunofluorescence staining with antibodies directed against FLAG and pericentrin in MDA-MB-231 cells stably expressing FLAG-tagged YB-1. (c) Direct immunofluorescence visualizing YB-1 in MDA-MB-231 cells transfected with full length GFP-tagged YB-1. Scale bar = 20 μm.  88  EV'  FLAG:YB61WT'  FLAG:YB61S102D'  4' **'  3' ***'  2' 1'  ' 1' μM 'B I' 10 'μ M 'B I'  '  0' DM SO  γ6tubulin' DAPI'  B!  ro l  Pericentrin' DAPI'  γ6tubulin'  C!  10'μM'BI'  nt  1'μM'BI'  Co  DMSO'  Fold6change'in'centrosome'area'  Control'  Pericentrin'  A!  FLAG:YB61S102A'  DAPI' Pericentrin'  Figure S2.7 Inhibition of YB-1 Ser-102 phosphorylation promotes centrosome dysfunction. (a) Immunofluorescence staining with antibodies targeting pericentrin (upper panel) and γ-tubulin (lower panel) in MDA-MB-231 cells treated with either a DMSO vehicle or BID1870 (1 μM or 10 μM) for 24 hours. Scale bar = 10 μm. (b) Quantification of centrosome area in MDA-MB-231 cells treated for 24 hours with DMSO or BI-D1870 (1 μM or 10 μM) was measured using Image Pro Analyzer software and is represented relative to the untreated control. 100 centrosomes from G1-phase cells were measured across three independent experiments. (c) Immunofluorescence targeting pericentrin in MDA-MB-231 cells stably expressing FLAG:YB-1WT, FLAG:YB-1S102D, or FLAG:YB1S102A protein. Scale bar = 10 μm.  89  Percent'of'metaphases'  60' Type'III'  50'  Normal'PCG'  PCG'Type'I'  PCG'Type'II'  PCG'Type'III'  Type'II'  40'  Type'I'  30' 20' 10' 0' HTRZ'  HTRY'  Figure S2.8 Cohesion defects were an underlying cause of chromosomal instability. Sister chromatid cohesion were assessed in 200 metaphase spreads from HTRZ and HTRY cells following a 96-hour induction with doxycycline. Representative images demonstrate normal primary constriction cohesion and the various classes of defective cohesion: PCGI (mild), PCGII (moderate), and PCGIII (severe). Scale bar = 10 μm.  90  B"  HTRZ"  Percent"of"cells"  100"  CEP17" HER2"  80" 60" 40" 20"  HTRY" 50" Percent"of"cells"  A"  CEP17" HER2"  40" 30" 20" 10" 0"  0" 1" 2" 3" 4" 5" 6" 7" 8" 9" 10" Number"of"signals"  1" 2"  3"  4" 5" 6" 7" 8" Number"of"signals"  9" 10"  Figure S2.9 HER2 amplification was absent in HTRZ cells but occurred in tetraploid HTRY cells. (a) HTRZ and (b) HTRY cells were induced for 96 hours and hybridized with FISH probes against the pericentromeric region of chromosome 17 (CEP17) and the HER2 locus (17q21.1). The number of signals was counted in 100 interphase cells.  91  2.6.2 Supplementary tables  Target Protein Name LIMK1/2 RSK1/2 ZAP70 PKA Cb ErbB2 (HER2) Arrestin b1 Src GSK3a/b Mnk1 IKKg/NEMO CDK1/2 AMPKb Paxillin MARK HspBP1 PKC-nu (PKN3) Cdc34 Tau eIF4E Cofilin  Phospho Site (Human)  % CFC*  Y507+T508/Y504+T505 S380/S386 Y315+Y319 S338 Y1248 Pan-specific Y529 Y279/ Y216 T209+T214 Pan-specific Y15 Pan-specific Pan-specific Pan-specific Pan-specific  365 269 230 198 196 192 177 177 174 167 161 157 154 149 149  Pan-specific Pan-specific S716 Pan-specific Pan-specific  PKCd  S664  p53 CDK9 PKR1 PP4/A’2 PKR1 Vrk1 PI3K PTEN ZIPK ZAP70 PP2A/Aa/b PKCq CDK6  Pan-specific Pan-specific T451 Pan-specific Pan-specific Pan-specific Pan-specific Pan-specific Pan-specific Y292 Pan-specific Pan-specific Pan-specific  149 148 148 142 135 125 118 116 115 114 112 111 109 103 103 102 102 101 98  Target Protein Name CASK/Lin2 GFAP CASP4 4E-BP1 Axl CAMK2d CK1g2 STAT3 CASP12 Caveolin 2 CaMK1d CAMK2b CASP7 Trail FGFR2  Phospho Site (Human) Pan-specific S8 Pan-specific S65 Pan-specific Pan-specific Pan-specific Pan-specific Pan-specific S23 Pan-specific Pan-specific Pan-specific Pan-specific Pan-specific  ANKRD3  Pan-specific  Erk4 PARP1 PKA Ca/b PKBa (Akt1) p38g MAPK (Erk6) STAT5A  Pan-specific Pan-specific Pan-specific S473  % CFC*  Pan-specific Pan-specific  -97 -86 -85 -85 -85 -83 -83 -82 -82 -81 -77 -73 -73 -72 -72 -69 -68 -68 -67 -66 -66 -65  Table S2.1 Proteins regulated by YB-1 based on Kinex™ microarray analysis. Proteins with enhanced (green) and repressed (orange) expression/activity following YB1 induction. *%CFC refers to the percent change from control (uninduced HTRY cells).  92  Antibody  Dilution  Supplier  Pan-actin  1:1000  Cell Signaling  α-tubulin  1:200  Abcam  CDK2  1:1000  Cell Signaling  Cenexin1  1:100  Abcam  CNTROB  1:100  Abcam  Cyclin E (clone HE12)  1:1000  Millipore  EGFR  1:1000  StressGen  FLAG M2  1:2000 (IB); 1:1000 (IF)  Sigma  HER2  1:200  Abcam  pHistone H3 (S10)  1:100  Cell Signaling  LIMK1  1:1000  Abcam  pLIMK1/2T508/T505  1:1000 (IB); 1:20 (IF)  Santa Cruz  p27Kip1 (C-19)  1:1000  Santa Cruz  p53 (DO-1)  1:100  Santa Cruz  Pericentrin  1:500 (IB); 1:250 (IF)  Abcam  Phalloidin (AF488-conjugate)  1:100  Invitrogen  PLK1  1:100  Sigma  RSK1  1:1000  Santa Cruz  RSK2  1:500  Santa Cruz  RSK1/2/3  1:1000  Cell Signaling  1:1000  Cell Signaling  γ-tubulin (clone D-10)  1:1000 (IB); 1:50 (IF)  Santa Cruz  YB-1  1:2000  Cell Signaling  pYB-1  1:1000 (IB); 1:200 (IF)  Cell Signaling  Vinculin  1:1000  Upstate  pRSK  S380  S102  Table S2.2 List of antibodies, dilutions, and suppliers used for immunoblotting (IB) and immunofluorescence (IF).  93  siRNA  siRNA target sequence (5’ – 3’)  Manufacturer  RSK1  CCCAACATCATCACTCTGAAA  Qiagen  RSK2  AGCGCTGAGAATGGACAGCAA  Qiagen  YB-1 #1  CCACGCAATTACCAGCAAA  Dharmacon  YB-1 #2  Hs_YBX1_1_HP Validated siRNA (#SI03019191) AATTCTCCGAACGTGTCACGT  Qiagen  Scrambled  Qiagen  Table S2.3 siRNA target sequences.  94  CHAPTER 3. YB-1 PROMOTES TUMOUR INITIATION THROUGH CHROMATIN REMODELING LEADING TO THE DEVELOPMENT OF TRIPLE-NEGATIVE BREAST CANCER  3.1 Overview The molecular events that trigger the development of triple-negative breast cancer (TNBC) are not well understood. Here, we demonstrate that Y-box binding protein-1 (YB-1) was sufficient to transform human mammary epithelial cells (HMECs). Through stabilization and up-regulation of the histone acetyltransferase p300, YB-1 orchestrated an epigenetic reprogramming to transition HMECs into stem/progenitor-like tumourinitiating cells. Specifically, altered histone acetylation patterns along the BMI1, CD44, and CD49f promoters facilitated chromatin relaxation allowing for transcriptional activation by YB-1. This enhanced self-renewal capacity thereby priming cells for malignancy. Over time, cells acquired RSK2 and hTERT providing the capacity to form tumours when transplanted in vivo that were molecularly subtyped as TNBC. Notably, YB-1 overexpression was detected in focal areas of histologically normal mammary tissue that could represent ideal candidates as precursors to breast cancer.  3.2 Introduction Next-generation genomic sequencing has revealed that a single mutation in Akt2 or KRas, for example, is sufficient to drive the initiation, maintenance, and progression of breast cancer (Stephens et al., 2012). Our group identified that Y-box binding protein-1 (YB-1) is expressed in over 40% of breast cancers, but is most commonly associated with the triple-negative (ER-/PR-/HER2-) subtype (Habibi et al., 2008). The transcription and translation factor is known to regulate a myriad of genes that promote breast cancer growth and survival, including cyclin A, cyclin B1 (Jurchott et al., 2003), EGFR (Stratford et al., 2007), MET (Finkbeiner et al., 2009), PIK3CA (Astanehe et al., 2009), and MDR1 (Bargou et al., 1997). Accordingly cancer cells, both in vitro and in vivo, are  95  physiologically dependent upon the continued overexpression of YB-1 for maintenance of their malignant phenotype (Stratford et al., 2008; Stratford et al., 2007; Sutherland et al., 2005). The activity of YB-1 is contingent upon phosphorylation of its serine-102 residue by p90 ribosomal S6 kinase (RSK) (Stratford et al., 2008) and to a lesser extent AKT (Sutherland et al., 2005). Most recently, it has been reported that activated pYB1S102 is involved in maintaining the tumour-initiating cell (TIC) population via the transactivation of CD44 and CD49f (To et al., 2010). Within the last decade, mounting evidence has supported the notion that cancer arises from a pool of stem/progenitor-like cells, colloquially referred to as TICs (Gupta et al., 2009; Molyneux et al., 2010; Reya et al., 2001; Visvader and Lindeman, 2008). These cells share important properties with classical tissue stem cells including self-renewal and multilineage differentiation capacity as well as therapeutic refractoriness (Clarke et al., 2006). Pioneering work in this field originated in studies of leukemia stem cells (Lapidot et al., 1994) and later was expanded to encompass solid tumours of the breast (Al-Hajj et al., 2003). It has since been established that a subpopulation of mammary cells defined as CD44+/CD24- express putative stem cell markers and have the capacity to initiate tumours at limiting dilution (Shipitsin et al., 2007). Genes involved in cell fate determination, immortalization, and DNA repair are epigenetically deregulated in breast cancer, the most notable being the Polycomb group (PcG) targets CDKN2A (encoding p16INK4a) (Holst et al., 2003) and HOXA9 (Reynolds et al., 2006). Specifically, the PcG protein BMI1 is a key modulator in transcriptionally repressing these loci during the transformation of human mammary epithelial cells (HMECs) (Dimri et al., 2002). This permits cells to escape from senescence and acquire stem/progenitor cell properties such as enhanced self-renewal capacity (Lessard and Sauvageau, 2003; Liu et al., 2006). Notably, a subset of HMECs detected in disease free women exhibits silenced p16INK4a and genomic instability (Holst et al., 2003; Romanov et al., 2001). In light of our previous finding that YB-1 functions as a cancer susceptibility gene by promoting genomic instability (Davies et al., 2011), we wanted to address its utility in  96  fully transforming HMECs. Here we show that expression of YB-1 in HMECs enhanced p300 activity to alter histone acetylation patterns and relax the promoter-centered chromatin of TIC-associated genes, BMI1, CD44, and CD49f. Consequently, YB-1 was able to directly bind and transcriptionally regulate BMI1 resulting in repression of p16INK4a and enhanced self-renewal capacity. Prolonged YB-1 induction led to the emergence of cells with a triple-negative breast cancer (TNBC) subtype that were fully transformed and capable of forming tumours in vivo. In addition to YB-1, these cells were discovered to overexpress RSK2 and hTERT. Our findings translated clinically as a strong positive correlation was found to exist between these three genes in TNBC patient tissue. Ominously, we identified cells with high YB-1 expression in histologically normal mammary glands from individuals who develop breast cancer that could represent ideal candidates as precursors to the disease.  3.3 Results 3.3.1 Ectopic YB-1 expression in HMECs is at physiological level To begin to profile the molecular events that govern breast cancer progression, we engineered a Tet-On YB-1 expression system into non-malignant H16N2 HMECs. Our group has extensively characterized this system previously (Davies et al., 2011). Cells conditionally expressing YB-1 under control of the tetracycline-inducible promoter were termed HMEC Tet-repressed YB-1 (HTRY), while a LacZ-expressing control cell line was designated HMEC Tet-repressed LacZ (HTRZ). Immunoblotting confirmed an increase in YB-1 following a 96-hour induction with doxycycline, as compared to HTRZ cells that expressed negligible levels (Figure S3.1A). We also detected elevated levels of RSK1, an upstream kinase responsible for activation (phosphorylation) of YB-1S102, in the HTRY cells (Figure S3.1A). Importantly, the level of ectopic YB-1 expression was similar to established breast cancer cell lines. Relative to HTRZ, HTRY cells exhibited a 2.8-fold increase in YB-1 transcript, well within the 1.8 to 7.2-fold range measured across the cancer cell lines (Figure S3.1B).  97  3.3.2 HMECs acquire characteristics of stem/progenitor-like tumourinitiating cells following YB-1 expression Genome wide ChIP-on-chip analysis preformed previously by our group identified a potential interaction between YB-1 and the BMI1 promoter (Finkbeiner et al., 2009). As expected, expression of BMI1 was elevated in HTRY cells, relative to HTRZ cells, at both the mRNA transcript (Figure 3.1A) and protein (Figure 3.1B) level. This correlated with an increase in histone H2A ubiquitylation and as a result the INK4a locus was repressed as measured by decreased p16INK4a. Consistent with our previous findings (To et al., 2010), the YB-1 transcriptional targets CD44 and CD49f were also more highly expressed in the HTRY cells. To validate that this phenomenon was not influenced by the H16N2 genetic background, we transiently transfected a FLAG:YB-1 expression vector into the normal 184hTERT and MCF10A HMECs. Both cell lines exhibited an increase in the expression of TIC-associated genes concomitant with YB-1 (Figure 3.1B). Accordingly, we questioned whether TIC marker expression translated into phenotypic changes. TICs possess stem/progenitor-like properties such as the ability to self-renew and differentiate along each of the mammary epithelial lineages. To assess self-renewal, HTRZ and HTRY cells were plated into non-adherent mammosphere assays. In stark contrast to HTRZ cells, HTRY cells formed mammospheres that could subsequently be dissociated and single cells re-passaged as new spheres for at least five generations (Figure 3.1C). In agreement with this finding, transient transfection of FLAG:YB-1 into 184hTERT cells greatly enhanced their ability to form mammospheres that could be serially passaged (Figure 3.1D). In a second set of experiments, cells from secondary mammospheres were seeded at clonogenic density onto a collagen substratum to evaluate their potential for bilineage differentiation. HTRZ and 184hTERT:EV (empty vector) cells predominately differentiated into mixed lineage colonies expressing both luminal (CK18) and basal (CK14) markers (Figure 3.1E). Interestingly, induction of YB-1 skewed cellular differentiation along the luminal lineage in both models (Figure 3.1E).  98  1"  kDa" 50"  YB51"  37"  BMI1"  25"  ubH2A"  15"  p16INK4a"  75"  CD44"  150"  CD49f"  *"  150" 100"  HTRZ" HTRY"  p16"  CD44" CD49f" HTRZ" HTRY"  **" **"  **" **"  50"  **"  0" 1"  2"  3"  4"  FLAG:YB51"  *"  EV"  *"  2"  MCF10A"  EV"  184hTERT"  HTRY"  3"  BMI1" Number"of" mammospheres"  Tet"system"  4"  0"  C"  B"  **"  FLAG:YB51"  5"  HTRZ"  Fold5change"in" transcript"  A"  Vinculin"  100"  5"  FLAG:YB51"  Passage"  80"  **"  184hTERT:EV"  **" **"  EV"  60" 40"  FLAG:YB51" Vinculin"  20" 0" 1"  2"  E"  CK18"  CK14"  184hTERT:YB51" YB51"  Number"of" mammospheres"  100"  3"  Passage"  Percent"of"colonies"  D"  100"  **"  75" *"  50" 25" 0"  HTRZ" HTRY" 184hTERT:EV" 184hTERT:YB51"  **"  **"  **" Mixed"  Luminal"  Basal"  Figure 3.1 HMECs acquired TIC characteristics following ectopic YB-1 expression. (a) qRT-PCR and (b) immunoblot analysis of TIC markers in induced HTRZ and HTRY cells. Following a 96-hour transient transfection, protein expression was measured in empty vector (EV) and FLAG:YB-1-expressing 184hTERT and MCF10A cell lines. (c) HTRZ and HTRY cells grown in mammosphere cultures and serially passaged. Doxycycline was replenished at each generation. Data represent mean ± SD (n = 3). (d) Mammosphere assay following transient transfection of EV and FLAG:YB-1 in 184hTERT cells (confirmed by immunoblotting). Data represent mean ± SD (n = 3). (e) Differentiation culture of cells from secondary mammospheres. 50 colonies per experiment were evaluated for CK18 and CK14 immunofluorescence. Scale bar = 100 μm. Data represent mean ± SD (n = 3).  99  3.3.3 Chromatin remodelling by p300 underlies the reprogramming of HMECs into TICs Work from other groups has highlighted the importance of the histone acetyltransferase protein p300 in maintaining the stem cell compartment (Chen et al., 2008; Zhong and Jin, 2009). Coupled with the fact that p300 is up-regulated in CD44-positive breast cancer cells (Bourguignon et al., 2009), we questioned whether its expression altered histone acetylation patterns to facilitate the reprogramming of HMECs into TICs. Compared to HTRZ cells, p300 protein was strongly expressed in HTRY cells (Figure 3.2A). We sought to uncover the mechanism behind p300 up-regulation in these cells, which was aided by the knowledge that YB-1 directly regulates the expression of p110α (Astanehe et al., 2009) and that signaling via the PI3K/Akt pathway stabilizes the p300 protein (Chen et al., 2004). As predicted, the catalytic subunit of PI3K (p110α) and pAktS473 were elevated in HTRY cells and treatment with the PI3K inhibitor LY294002 decreased p300 expression (Figure 3.2A). As p300 is degraded by the 26S proteasome (Poizat et al., 2000), we could partially rescue its expression using MG132 (Figure 3.2A). As p300 shuttles between the cytoplasm and nucleus (Chen et al., 2007), we next questioned whether it localized to the nuclear compartment where it functions as a histone acetyltransferase (HAT). The number of p300-positive nuclei and the intensity of staining was much greater in HTRY cells relative to HTRZ cells as visualized by immunofluorescence (Figure 3.2B). Moreover, to ensure that p300 was enzymatically active we performed a HAT assay using nuclear lysates. HAT activity was over five times higher in the HTRY cells compared to the HTRZ cells (Figure 3.2C). As expected this could be abrogated with siRNA targeting YB-1. We further discovered that inhibition of RSK1 and RSK2, the upstream kinases responsible for YB-1 phosphorylation, using siRNA or the small molecule pan-RSK inhibitor BI-D1870 also decreased HAT activity. Treatment with LY294002 had a similar effect (Figure 3.2C). Together, these data suggest that activated pYB-1S102 is required to maintain p300 activity via up-regulation of the PI3K signaling pathway. The p300 protein epigenetically regulates gene expression by acetylating lysine resides on histone proteins. Therefore, as a direct measure of p300 activity we assessed the extent 100  of histone acetylation by immunoprecipitation of histone H3. Using anti-acetylated-lysine antibody we detected enrichment in the pool of acetylated histones in HTRY cells relative to HTRZ cells (Figure 3.2D). To refine this analysis we next evaluated acetylation of histone H3 at lysine 9 (AcH3-K9) specifically along promoter-centered chromatin of TIC-associated genes. The 9th lysine residue (H3K9) of histone H3 is a preferential substrate of p300 (Szerlong et al., 2010; Zhong and Jin, 2009). DNA purified after immunoprecipitation with anti-AcH3-K9 antibody was evaluated by PCR using primers targeting the length of the BMI1, CD44, and CD49f promoters (Figure S3.2). Higher levels of promoter-associated histone H3-K9 acetylation were observed in HTRY cells compared to HTRZ cells (Figure 3.2E). Utilizing anacardic acid (AA), a potent inhibitor of p300, we demonstrate that inhibition of p300 activity prevented the reprogramming of HMECs into TICs. When HTRY cells were pre-treated with 15 μM AA prior to YB-1 induction, HAT activity was quenched by nearly 80% (Figure S3.3), which correlated with loss of BMI1, CD44, and CD49f mRNA expression (Figure 3.2F). Functionally, this yielded fewer, and smaller, primary mammospheres that could not be serially passaged (Figure 3.2G).  101  100"  E"  HTRZ"  D"  Input"  BMI1"  Ac1lysine"  15" 1.0" 3.8" 15" 1.0" 1.0"  Histone" H3"  CD49f" CD44"  kDa"  HTRZ" HTRY" HTRZ" HTRY" HTRZ" HTRY"  IP:" Input" HH3" IgG"  ChIP"a" ChIP"b" ChIP"c" ChIP"a" ChIP"b" ChIP"a" ChIP"b"  AcH31K9"  IgG"  20"μM"LY"  2"μM"BI"  1"μM"BI"  **"  DMSO"  siRSK1/2"  Scr"  0"  siYB11"  **"  200"  HTRZ"  HAT"ac=vity"(pmol/min)"  400"  HTRY"  F"  1.2" DMSO" 15"μM"AA"  0.8"  *"  0.4"  *" *"  0"  I1 CD " 44 CD " 4 18 9 S "r f " RN GA A" PD H"  250"  600"  BM  250"  HTRY"  50"  HTRZ"  pAktS473" Akt" p300" short"exp" p300" long"exp" Vinculin"  50"  Composite"  HTRY"  100"  HTRZ"  YB11" p110α"  p300"  HTRZ"  DAPI"  HTRY"  LY"+"MG"  LY"  DMSO"  HTRZ"  kDa" 50"  C"  Rela=ve"gene"expression"  B"  HTRY"  HTRY"  A"  GAPDH"  G"  15"μM"AA"  60" 40" 20" 0"  **" **"  DMSO"  DMSO"  80"  15"μM"AA"  Number"of" mammospheres"  100"  Passage"1" Passage"2"  Figure 3.2 YB-1 enhanced p300 stability, nuclear localization, and HAT activity to reprogram HMECs into TICs. (a) Immunoblot analysis of p300 and PI3K signaling components in induced HTRZ and HTRY cells treated for 24 hours with DMSO, LY294002 (LY), or LY in combination with MG132 (MG). (b) Immunofluorescence of p300 localization in induced HTRZ and HTRY cells. Arrows denote p300-positive nuclei (visualized by DAPI). Scale bar = 100 μm. (c) HAT activity in nuclear lysate from induced HTRZ and HTRY cells treated for 72 hours with DMSO, LY, BI-D1870 (BI), or siRNA targeting YB-1 and RSK1/2. Data represent mean ± SD (n = 3). (d) Histone H3 (HH3) immunoprecipitation followed by immunoblotting to measure acetylated lysine residues. (e) ChIP targeting the promoters of BMI1, CD44, and CD49f in induced HTRZ and HTRY cells. DNA templates were  102  pulled down with acetyl-histone H3 (Lys9) or nonimmune IgG antibody. GAPDH served as a control. (f) qRT-PCR analysis of mRNA transcript in adherent HTRY cells pretreated with DMSO or anacardic acid (AA) for 4 hours prior to induction. Data represent mean ± SD (n = 3). (g) HTRY cells serially passaged as mammospheres in the presence of DMSO or AA. Data represent mean ± SD (n = 3). Representative images of primary mammospheres are shown. Scale bar = 500 μm.  103  3.3.4 Chromatin remodelling permits YB-1 to transcriptionally regulate BMI1 and enhance self-renewal capacity We wondered whether p300-mediated chromatin relaxation was a prerequisite for YB-1 to bind TIC-associated gene promoters. We now show that YB-1 also binds to the BMI1 promoter in both HTRY and MDA-MB-231 breast cancer cells using conventional ChIP (Figure S3.4). Furthermore, in ChIP assays using anti-YB-1 antibody, pre-treating HTRY cells with AA prevented YB-1 binding to the BMI1 and CD49f promoters as well as one region on the CD44 promoter (Figure 3.3A). This suggests that p300-mediated chromatin remodeling is upstream of YB-1 promoter binding and gene transcription. The regulation of BMI1 by YB-1 is novel and, therefore, we sought to characterize it further. First, to confirm functional binding of YB-1 to the BMI1 promoter, we silenced the protein using two unique siRNA targeting sequences. This resulted in decreased BMI1 expression and rescue of p16INK4a in MDA-MB-231 cells (Figure 3.3B). We next questioned whether the expression of BMI1 would fluctuate concordantly with changes in YB-1 activation. HTRZ cells transiently transfected with a constitutively active YB-1 mutant (FLAG:YB-1S102D) expressed the highest level of BMI1 compared to cells expressing wild-type YB-1 (FLAG:YB-1WT) or a non-phosphorylatable mutant (FLAG:YB-1S102A). Accordingly, p16INK4a was repressed in YB-1S102D transfected cells, while conversely the YB-1S102A mutant acted as a dominant negative (Figure 3.3C). Capitalizing on the fact that YB-1 is predominately activated (phosphorylated) by RSK (Stratford et al., 2008), we inhibited the kinase using siRNA (Figure 3.3D) or 10 μM BID1870 (Figure 3.3E). In both HTRY and MDA-MB-231 cells, suppression of RSK1/2 expression and/or activity led to a decrease in activated pYB-1S102 resulting in loss of BMI1 and rescue of p16INK4a. BMI1-mediated silencing of p16INK4a has been associated with a loss of G1/S checkpoint fidelity (Molofsky et al., 2003). In accordance, only 22.4 ± 2.3% of YB-1-induced HTRY cells were in G1 compared to nearly half of uninduced cells. We were able to partially restore checkpoint activity following knockdown of BMI1 using siRNA (Figure 3.3F). This suggests that YB-1 expression potentiated G1/S checkpoint slippage via a BMI1dependent mechanism. Accordingly, we next visualized nuclear localization of p16INK4a 104  (Figure S3.5A). In contrast to uninduced HTRY cells the majority of nuclei following YB-1 induction were p16INK4a-depleated; however, its expression could be restored following treatment with BMI1 siRNA (Figure S3.5B). The abovementioned findings prompted us to question whether the enhanced self-renewal potential of HTRY cells was a direct consequence of BMI1 expression. The mammosphere forming capacity of HTRZ cells was increased from 12 ± 2 to 73.7 ± 16.1 following transient transfection with pMIN:BMI1 expression vector. Conversely, siRNAmediated silencing of BMI1 in HTRY cells decreased their ability to form mammospheres by nearly 80% (from 104 ± 18.5 to 22.7 ± 2.8) (Figure 3.3G). Taken together, these data convey that during the earliest stages of malignancy upregulation of YB-1 facilitates the transformation of HMECs into stem/progenitor-like TICs. Reprogramming occurs in two steps. First, YB-1-mediated activation of the HAT protein p300 alters the histone acetylation landscape. Subsequently, the relaxation of promoter-centered chromatin allows for activated pYB-1S102 to bind and transcriptionally regulate BMI1, CD44, and CD49f to instill TIC phenotypes, such as enhanced selfrenewal and multipotency (Figure 3.3H).  105  BI1D1870"  RSK1"  50"  pYB11S102"  RSK2"  37"  BMI1"  50"  pYB11S102"  15"  p16INK4a"  37"  BMI1"  15"  p16INK4a"  50"  Tubulin"  Vinculin"  100"  FLAG:YB11S102A"  FLAG:YB11S102D"  FLAG:YB11WT"  BMI1" ubH2A"  15" 150"  p16INK4a" Vinculin"  60" 50"  *"  40" 30"  *"  +YB11"  20" 10"  BMI1" p16INK4a" Vinculin"  Scr" siBMI1" 1YB11"  +YB11"  H" EV" pMIN:BMI1" Scr" siBMI1"  120" 100"  **"  80" 60" 40"  **"  20"  X"  HTRZ"  siBMI1"  Scr"  TIC1associated"genes"  +"YB+1"  1" YB+1" 2"  HTRY"  "  CCAAT" BMI1" p16INK4a" Vinculin" p300" Ac"  pMIN:BMI1"  0"  37"  0"  140"  pMIN:EV"  G"  Number"of"mammospheres"  100"  Percent"cells"in"G1"  F"  MDA1MB1231"  FLAG:YB11"  1YB11" Scr" siBMI1"  Vinculin"  BI1D1870"  kDa"  EV"  siYB11"#2"  150"  HTRY" DMSO"  siRSK1/2"  p16INK4a"  DMSO"  E"  MDA1MB1231"  15"  50" 25"  BMI1"  37"  ChIP"b"  kDa"  YB11"  50"  ChIP"a"  siRSK1/2"  Scr"  kDa"  ChIP"a"  HTRY"  C" siYB11"#1"  15"μM"AA"  DMSO"  DMSO"  B"  IgG"  ChIP"b"  Scr"  kDa" 100"  YB11"  ChIP"a"  Scr"  D"  15"μM"AA"  CD49f" CD44"BMI1"  DMSO"  10%"Input"  15"μM"AA"  A"  P YB11" CCAAT"  Ac"  "  BMI1".."CD44".."CD49f" ubH2A"  p16INK4a"  Figure 3.3 YB-1 transcriptionally regulated BMI1 to enhance self-renewal capacity. (a) ChIP analysis of HTRY cells pre-treated with DMSO or anacardic acid (AA) for 4 hours prior to induction. DNA templates were pulled down with YB-1 or nonimmune IgG antibody and different promoter regions (ChIP a and ChIP b) were amplified using primers flanking YB-1 binding sites in the BMI1, CD44, and CD49f promoters. (b) Immunoblot following siRNA silencing of YB-1 in MDA-MB-231 cells. Scrambled peptide (Scr) was a control. (c) Immunoblot analysis following 96-hour transient  106  transfection of FLAG-tagged YB-1 expression plasmids into HTRZ cells. (d-e) Immunoblot analysis of induced HTRY and MDA-MB-231 cells treated with (d) RSK1/2 siRNA for 96 hours or (e) BI-D1870 for 24 hours. (f) Cell cycle profile of induced HTRY cells treated with Scr or BMI1 siRNA measured using an ArrayScan VTI. Uninduced cells served as a control. Data represent mean ± SD (n = 3). (g) HTRZ cells transfected with empty vector (EV) or BMI expression plasmid and induced HTRY cells treated with Scr or BMI1 siRNA for 96 hours were grown in mammosphere cultures. Data represent mean ± SD (n = 3). (h) Model depicting the series of events that reprogram YB-1expressing HMECs into TICs.  107  3.3.5 YB-1 evokes luminal filling and invasion in a three-dimensional model of breast acini Early changes during the development of breast cancer involve the transition of normal mammary ducts into those with a hyperplastic epithelium. This is followed by degradation of the basement membrane and cellular invasion (Figure 3.4A). To model this phenomenon in vitro, we grew HTRY cells on a reconstituted basement membrane. These cells became organized as three-dimensional polarized acini structures with a hollow lumen (Figure 3.4B). Subsequent induction of YB-1 led to luminal filling mirroring a ductal carcinoma in situ (DCIS). By 8 days post-YB-1 induction, cells began to invade through the basement membrane and into the surrounding microenvironment similar to an invasive ductal carcinoma (IDC) (Figures 3.4B and 3.4C). In agreement, YB-1 expression caused HTRY cells to become invasive through Matrigel-coated Transwell chambers (Figure 3.4D). The earliest luminal outgrowths were found to express CD44, which, at later time points, was ubiquitously expressed by cells invading the luminal space (Figure 3.4E). As CD44 expression is dependent upon YB-1 serine-102 phosphorylation by RSK (To et al., 2010), we questioned whether inhibiting this activation could prevent DCIS-like luminal outgrowths. Notably, induced acini treated concomitantly with the RSK inhibitor BI-D1870 resembled the uninduced control (Figure 3.4F).  108  A"  Ductal"carcinoma" in"situ"  Normal"duct"  Invasive"ductal" carcinoma"  B"  YB41"inducNon"(days)" uninduced"  2"  4"  8"  CD49f"" "ZO41" DAPI" **" **"  100"  HTRZ" HTRY"  75"  *"  50"  **"  25" 0" 0"  2"  4"  8"  Luminal"filling"  0"  2"  4"  DAPI"  D"  8"  **"  60" Percent"invaded"cells"  Percent"of"acini"  C"  **"  40" 20" 0"  Invasive"  HTRZ"  E"  HTRY"  MDA4MB4231"  F" YB41"inducNon"(days)" 2"  4"  8"  CD44" DAPI"  Percent"hollow"acini"  uninduced"  CD49f"" ZO41" DAPI"  125" 100" 75" 50" 25"  **"  0" uninduced"  DMSO"  10"μM"BI4 D1870"  +YB41"  Figure 3.4 YB-1 drove luminal translocation and outgrowth in a three-dimensional model of breast acini. (a) The stages of breast cancer progression. (b-c) HTRY cells were grown as 3D acini on a reconstituted basement membrane. Following YB-1 induction, acinar structures were immunostained with ZO-1 (luminal) and CD49f (basolateral) antibodies and evaluated for luminal filling and invasive outgrowths at the times indicated. Data represent mean ± SD (n = 3). Scale bar = 100 μm. (d) Quantification of cellular invasion through Matrigelcoated Transwell chambers at 24 hours. Data represent mean ± SD (n = 3). (e) CD44 staining in YB-1-induced acini. Scale bar = 100 μm. (f) Acini were grown as above; however, YB-1 was induced along with the addition of DMSO or BI-D1870. At 96 hours luminal filling was evaluated. Data represent mean ± SD (n = 3). Scale bar = 50 μm. 109  3.3.6 Sustained upregulation of YB-1 leads to full transformation and tumour-initiation The ability of YB-1 to elicit phenotypes associated with malignant progression in threedimensional acini prompted us to investigate whether the oncogene could confer full transformation and ultimately tumour-initiation. To our surprise, HTRY cells failed to form colonies in soft agar suggesting that these cells are not tumourigenic (Figure 3.5A). We, therefore, questioned whether long-term YB-1 expression would yield enhanced tumourigenic potential. For 30 days, HTRY cells were grown in the presence of doxycycline to sustain YB-1 expression. We generated two cell lines in parallel, denoted as HTRY-LT #1 and #2, that acquired the capacity to form colonies in soft agar (Figure 3.5A). The expression and activity of YB-1 was higher in the HTRY-LT cell lines compared to the HTRY cells (Figures 3.5B and S3.6A). However, there was little to no significant change in BMI1, CD44, and CD49f gene expression (Figure S3.6B) or HAT activity (Figure S3.6C). The ability to form primary mammospheres was also similar between the HTRY and HTRY-LT cell lines, but they were only maintained upon serial passaging in the HTRY-LT cells (Figure S3.6D). Previous work from our group and others has uncovered that TNBCs are exquisitely dependent on RSK2 signaling (Brough et al., 2011; Stratford et al., 2012). Furthering these studies, we found that the RSK2 isoform, specifically, becomes up-regulated during the evolution of HTRY cells into HTRY-LT cells (Figure 3.5B). Telomerase (hTERT) activity was also detected in the HTRY-LT cell lines, but not in HTRZ and HTRY cells (Figure 3.5C). As MAP kinase signaling is known to regulate hTERT expression and activity (Maida et al., 2002), we reasoned that its induction could be a direct consequence of RSK2 expression. In support of this notion, siRNA-mediated silencing of RSK2 or treatment with BI-D1870 decreased telomerase activity by over 3-fold in HTRY-LT cells (Figure 3.5D). The findings above raised the question as to whether RSK inhibitors could represent a viable therapeutic approach to overcome this transformation step. To test this, we plated HTRZ and HTRY-LT cells on a collagen substratum to reflect the mammary microenvironment. Four-day treatment with ≤ 2 μM BI-D1870 inhibited the growth of 110  HTRY-LT cells by up to 70% (Figure S3.7A). Moreover, when treated for 8 days, with repeat dosing at 4 days, nearly all HTRY-LT cells were eliminated. In contrast, no overt toxicity was observed in the parental HTRZ cells (Figure S3.7B). This growth inhibitory effect translated into a significant reduction in the ability of BI-D1870-treated HTRY-LT cells to form soft agar colonies (Figure S3.7C). Interestingly, the capacity of these cells to grow as mammospheres that could be serially passaged was also abrogated by the drug (Figure S3.7D). This strongly implies that BI-D1870 is not only eradicating bulk tumour cells, but also the TIC subpopulation. Importantly, we demonstrate that treatment with BI-D1870 induced apoptosis in the HTRY-LT cell lines. A shift toward annexin Vpositive, apoptotic cells was observed beginning at 72 hours following BI-D1870 treatment. By 6 days post-treatment a significant pool (~50%) of cells were deemed apoptotic (Figure S3.7E). Finally, to confirm that BI-D1870 functioned via inhibition of RSK activity we measured a decrease in the phosphorylation of YB-1 and S6 ribosomal protein (Figure S3.7F). We endeavored to uncover the minimal combination of genes necessary for the HTRYLT cells to become fully transformed. Notably, uninduced HTRY cells acquired the ability to form soft agar colonies following YB-1/RSK2 double-transfection. Expression of YB-1 or RSK2 alone did not significantly enhance soft agar colony formation (Figure 3.5E). In a reciprocal experiment, we silenced YB-1, RSK2, and hTERT in the HTRYLT cell lines using siRNA oligonucleotides. This led to a repression in the number of soft agar colonies. Knockdown of RSK1 had only a partial effect highlighting the unique importance of the RSK2 isoform (Figure 3.5F). Taken together, these results suggest that interplay between YB-1, RSK2, and hTERT is necessary for transformation and maintenance of the malignant phenotype. One cannot exclude that the H16N2 genetic background may have influenced the transformative potential of these cells; therefore, we introduced YB-1 into a second mammary epithelial cell line, 184hTERT. Following stable transfection, 184hTERT clones emerged that expressed YB-1, RSK1, and RSK2 to a similar level observed in MDA-MB-231 cells (Figure S3.8A). Moreover, compared to cells transfected with empty vector, 184hTERT YB-1-expressing cells exhibited increased soft agar colony formation  111  (Figure S3.8B) and invasion through Matrigel-coated Transwell chambers (Figure S3.8C). Thus, the addition of RSK2 enabled YB-1-expressing cells to become fully transformed through collaboration with hTERT. To definitively assess the tumourigenicity of HTRY-LT cells in vivo, we injected them into the mammary glands of NOD/SCID mice at limiting dilutions. HTRZ cells were injected bilaterally as a control. Both HTRZ and HTRY-LT cell lines exhibited similar growth kinetics in vitro with doubling times of 55.6 and 43.5 hours, respectively (Figure S3.9). Palpable tumours were detected in mammary fat pads of mice injected with 105, 104, and 102 HTRY-LT cells with a latency of 14, 31, and 58 days, respectively (Figure 3.5G). Microscopic examination of mammary fat pads transplanted with HTRY-LT cells showed that these tumours grew as a solid mass of neoplastic cells. Glandular DCIS lesions were also evident (Figure 3.5H). In contrast, at 180 days post-injection tumours had not formed in mammary fat pads transplanted with HTRZ cells. Microscopic examination showed well-formed epithelial ductal structures (Figures 3.5G and 3.5H).  112  100"  MDA6MB6231"  HTRY6LT"#1"  HTRY6LT"#2"  HTRY"  SUM149"  **"  50" 0"  pYB61S102" RSK1"  100"  RSK2"  100"  pRSKS221/227"  100"  Vinculin"  HTRY6LT"#1" HTRY6LT"#2"  6" 5" 4" 3"  **"  2"  "BI6D1870"  0"  siRSK2"  1" Control"  **"  184hTERT"  HTRY6LT"#2"  HTRY6LT"#1"  **"  F"  G"  10,000" 100"  YB61" Vinc"  sihTERT"  siYB61"#2"  siYB61"#1"  siRSK2"  **"  Scr" siYB61#1" siYB61#2"  Scr" siRSK1" siRSK2" RSK1" RSK2" Vinc"  H"  Days"post"injec?on"  100,000"  50" 0"  YB61"+"RSK2"  RSK2"  YB61"  **"  100"  siRSK1"  20"  V5:RSK2" YB61" Vinculin"  150"  Scr"  40"  Colonies/well"  Control" YB61" RSK2" YB61+RSK2"  60"  0"  HTRY6LT"#1" HTRY6LT"#2"  200"  Control"  Colonies/well"  50" 100"  80"  HTRY6LT"  YB61"  Telomerase"ac?vity"(x104)"  **"  HTRY"  6" 5" 4" 3" 2" 1" 0"  E"  HTRZ"  50"  D"  HTRZ"  Telomerase"ac?vity"(x104)"  C"  HTRY6LT"#2"  kDa"  **"  HTRY6LT"#1"  **"  HTRY"  **"  150"  HTRZ"  B"  200"  HTRZ"  Colonies/well"  A"  HTRY6LT"  14"  30"  60"  90"  150"  5/5"  5/5"  5/5"  5/5"  5/5"  3/5"  5/5"  5/5"  5/5"  2/4"  4/4"  4/4"  100,000"  0/5"  10,000"  0/5"  100"  0/5"  HTRZ"  DCIS"  500"  Tumour"  100"  500"  113  Figure 3.5 Synergism between YB-1, RSK2, and hTERT conferred complete transformation. (a) Quantification of HTRZ, HTRY, and HTRY-LT cell growth under anchorageindependent conditions. MDA-MB-231 and SUM149 cells acted as a positive control. (b) Immunoblot assessing YB-1 and RSK expression and activation. (c) Telomerase activity in HTRZ, HTRY, and HTRY-LT cell lysate. 184hTERT cells acted as a positive control. Data represent mean ± SD (n = 3). (d) Telomerase activity in HTRY-LT cell lines following inhibition of RSK2 using siRNA or BI-D1870 for 72 hours. Data represent mean ± SD (n = 3). (e) Soft agar colony growth of HTRY cells expressing YB-1, RSK2, or YB-1 and RSK2. Uninduced cells served as the control. Data represent mean ± SD (n = 3). Immunoblotting confirmed transgene expression at 96 hours post-transfection. (f) HTRY-LT cells treated with scrambled peptide (Scr) or siRNA targeting YB-1, RSK1, RSK2, or hTERT for 48 hours prior to being transplanted into soft agar assays. Immunoblotting confirmed gene silencing. (g) The ability of HTRZ and HTRY-LT #1 cells to form palpable tumours when injected at limiting dilution bilaterally into the mammary fat pad of NOD/SCID mice (5 mice/group). (h) Hematoxylin and eosin staining of mammary fat pads transplanted with HTRZ and HTRY-LT cells. Scale bar = 100 μm or 500 μm, as indicated.  114  3.3.7 YB-1 transforms HMECs into cells with a TNBC subtype As breast cancer can be classified into distinct subtypes with differential aggressiveness and prognosis (Sorlie et al., 2001), we questioned which subtype the HTRY-LT cells represented. During YB-1 mediated transformation, the HTRY-LT cells gained expression of EGFR and lost expression of ESR1 (ER) and PGR (PR) as measured by qRT-PCR based gene expression profiling (Figure 3.6A) and immunoblotting (Figure 3.6B). Negligible level of ERBB2 (HER2) was also detected in the HTRY-LT cell lines. Taken together these findings suggest that the transformed cells represent the TNBC subtype. In further support, HTRY-LT cells showed a similar expression profile to TNBC clinical specimens in a hierarchical cluster analysis of subtype classification markers (Figure 3.6C). The physiological relevance of our model can be appreciated by the fact that YB-1 mRNA levels detected in the HTRY-LT cell line were perfectly emulated in a cohort of 83 primary TNBC patient tissues (p = 0.611) (Figure 3.6D). The levels of YB-1 were ~12-fold higher in the HTRY-LT cells compared to HTRZ cells. Likewise, TNBCs exhibit ~12-fold higher YB-1 as compared to luminal breast cancers. In summary, our data convey that HMEC transformation by YB-1 occurs in a step-wise process. During an initial pre-malignant phase, YB-1 and p300 co-operate to remodel histone acetylation patterns to promote the expression of BMI, CD44, and CD49f, which together evoke a stem/progenitor-like TIC phenotype. Over time, pressures exerted by YB-1 leads to the emergence of tumourigenic cells with a TNBC subtype that express high levels of YB-1, RSK2, and hTERT (Figure 3.6E). In support of this model, Affymetrix data from a cohort of nearly 600 TNBC patients revealed positive correlation between YB-1 and its transformation partners hTERT and RPS6KA3 (encoding RSK2), in addition to CD44 and CD49f (Table 3.1).  115  MDA4MB4231%  HTRY4LT%#2%  HTRY4LT%#1%  B"  HTRZ%  7%  HTRY4LT%#1%  6%  HTRY4LT%#2%  kDa%  5%  MDA4MB4231%  150%  EGFR%  4%  MCF47%  3%  BT474%  150%  HER2%  2%  50%  1%  100%  0% EGFR%  ERBB2%  ESR1%  BT474%  MCF47%  8%  HTRZ%  Fold4change%in%transcript%  A"  ER% PRβ% PRα%  150%  PGR%  C"  Vinculin%  D"  p = 0.611  Fold Difference in YB-1 mRNA levels  15  *  *  10  5  Cell Lines  TNBC  LT Y-  TR  H  TR  Z  Luminal H  H TR Z H HTRZ TR YLT  HTRY-LT  0  Patient Tissue  E" CD44/CD49f/BMI1% p300% RSK1% Normal%duct% YB41% %  hTERT% RSK2% YB'1"  TNBC%  HTRZ% HTRY% +%  HTRY% +%  HTRY4LT% ++%  Mammosphere%  +%  +%  +%  SoW%agar%  4%  4%  +%  Tumour4iniXaXng%  4%  4%  +%  Invasion%  Figure 3.6 HTRY-LT cells were molecularly classified as TNBC. (a) qRT-PCR and (b) immunoblot analysis of subtype biomarkers. MDA-MB-231 (TNBC), MCF-7 (ER+), and BT474 (HER2+) cells were used as controls. (c) Heat map of subtype biomarkers in HTRY-LT cells and patient breast tumours (red: high expression). (d) YB-1 transcript in HTRY-LT cells and luminal (n = 2) and TNBC (n = 83) tumour tissue measured by nanoString. (e) Depiction of the genetic and phenotypic features that define each step of YB-1-driven transformation of HMECs into a TNBC. 116  Gene  Correlation  YB-1  RPS6KA3  Spearman’s p-value rho (two-tailed) 0.213 2.4 x 10-7  CD49f  0.161  1 x 10-4  CD44  0.233  1 x 10-8  hTERT  0.113  0.006  Table 3.1 Genes correlated with YB-1 in a TNBC cohort. YB-1 mRNA levels were positively correlated to RPS6KA3 (encoding RSK2), CD44, CD49f, and hTERT expression, as assessed by Affymetrix data from 579 TNBC patients.  3.3.8 Foci of histologically normal mammary epithelia overexpress YB-1 Consistent with the findings reported here and by others (Bargou et al., 1997; Dahl et al., 2009), YB-1 was not detected in non-malignant breast epithelia obtained from reduction mammoplasties of healthy women (Figure 3.7A). However, the oncogene was highly expressed by ductal epithelial cells within sections of histologically normal, nonneoplastic mammary epithelia from individuals with breast cancer (Figure 3.7A). We also noted that activated pYB-1S102 coincides with pRSKS221/227 and CD44 in adjacent serial sections of these mammary glands (Figure 3.7B), supporting our in vitro data that this pathway likely has an important role in the early stages of breast cancer development. Moreover, YB-1 was detected in early DCIS lesions as well as invasive ductal and lobular carcinomas and lymph node metastases (Figure S3.10A), where it was coexpressed with pRSKS221/227 and CD44 in the local disease (Figure S3.10B). Finally, we specifically show intense YB-1 staining in the normal mammary epithelia of women that developed TNBC (Figure 3.7C). These data support the idea that YB-1 may have the ability to transform mammary epithelial cells in vivo leading to the development of TNBC.  117  A" Non$malignant+  C"  B"  YB$1+ Normal+“pre$malignant”+  Serial+sec=ons+ pYB$1S102+  pRSKS221/227+  CD44+  YB$1+ Malignant+  TNBC+Pa=ent+1+  TNBC+Pa=ent+6++  Histologically+normal+  Figure 3.7 YB-1 staining was detected in focal areas of histologically normal tissue from breast cancer patients. (a) Immunoperoxidase staining for YB-1 in representative mammary tissue from an individual without breast cancer (left) and histologically normal, non-neoplastic mammary epithelium from a breast cancer patient (right; n = 7). Scale bar = 500 μm. (b) Co-localization of pYB-1S102, pRSKS221/227, and CD44 staining in non-neoplastic mammary tissue from breast cancer patients. Scale bar = 100 μm (c) Histologically normal mammary epithelium from TNBC patients with high (patient 6) and low (patient 1) YB-1 expression in the malignant TNBC. Scale bar = 500 μm.  118  3.4 Discussion In the present study, we implicate YB-1 as a catalyst for the transformation of HMECs into a TNBC. Up-regulation of YB-1 promoted p300-mediated chromatin remodeling that reeducated mature HMECs into stem/progenitor-like TICs via induction of BMI1, CD44, and CD49f. However, this was not sufficient for complete transformation. Over time, pressures exerted by YB-1 promoted the emergence of cells with high expression of YB-1, RSK2, and hTERT that were tumourigenic in vivo and were molecularly classified as TNBC. Notably, strong correlation between these three genes in a cohort of TNBC patients emphasizes the relevance of our in vitro studies in emulating the disease. In addition, we discovered foci of intense YB-1 expression in histologically normal mammary epithelia from individuals who developed breast cancer that could represent putative precursors to TNBC. Royer and colleagues first demonstrated that YB-1 was expressed in primary human breast cancers, but absent in normal breast tissue (Bargou et al., 1997). Following this, a transgenic mouse model was engineered that formed tumours with 100% penetrance thereby cementing YB-1 as an oncogene with a principal role in tumourigenesis (Bergmann et al., 2005). We have now significantly furthered the field by dissecting the underlying mechanism of YB-1-mediated transformation. In this study, we found that YB-1 works in concert with p300 to facilitate the epigenetic reprogramming of HMECs into stem/progenitor-like TICs during the earliest stages of the neoplastic process. This interpretation is supported by the recent finding that YB-1 functions to promote growth and inhibit the differentiation of neural stem cells suggesting that it is a key mediator of the progenitor-differentiation switch (Fotovati et al., 2011). The de novo emergence of TICs is further corroborated by the discovery of plasticity within both normal and malignant breast cells (Chaffer et al., 2011; Gupta et al., 2011). Our work challenges the perception that cancer is simply a genetic disease driven by mutations and chromosomal abnormalities. We have presented compelling evidence that epigenetic alterations fuel the earliest stages of breast cancer progression. In support of this paradigm shift, promoter methylation and inactivation of tumour suppressor genes such as CDKN2A (Holst et al., 2003; Nuovo et al., 1999), HIC-1 (Fujii et al., 1998), and 119  RASSF1A (Dumont et al., 2009) is frequently detected in pre-invasive breast lesions. The finding that YB-1 transcriptionally regulates BMI1 is especially exciting because, apart from miR-200c (Shimono et al., 2009), it was previously unknown how its expression is controlled. The protein has a well-established role in altering histone ubiquitylation and promoter methylation to repress senescence, apoptotic, and differentiation pathways in stem/progenitor cells (Park et al., 2003). Our results suggest that these same pathways are silenced by YB-1 during early breast cancer development at least in part through BMI1. Interestingly, the reprogramming of HMECs into TICs did not confer tumourigenicity, but rather these “primed” pre-malignant cells had to acquire high expression of YB-1 in addition to RSK2 and hTERT to become fully transformed. It is tempting to speculate that YB-1, due to its ability to promote genomic instability (Davies et al., 2011), exerts selective pressures that lead to the up-regulation of these genes. Uncovering the molecular underpinnings of TNBC fueled our discovery of addiction to the RSK2/YB-1 signaling axis for initiation and progression of the disease. As a proof-ofconcept, we demonstrated that this represents a point of molecular vulnerability. The small molecule RSK inhibitor BI-D1870 prevented luminal filling of mammary acini suggesting that it could have therapeutic benefit for individuals at high risk of developing TNBC, notably BRCA1-mutation carriers (Palacios et al., 2005). Importantly, while the drug displayed efficacy in eradicating tumourigenic TNBC cells the toxicity toward normal HMECs was negligible. While RSK inhibitors have yet to advance into clinical trials, our work provides the necessary foundation to support in vivo pre-clinical studies. In summary, our study presents an attractive model to delineate how a single oncogene, YB-1, can transition a normal cell into a cancer cell. The finding that YB-1 specifically promotes the development of TNBC rationalizes why the oncogene is expressed in over 70% of clinical cases of the disease (Habibi et al., 2008). However, the ubiquitous occurrence of YB-1 in cancer and the fact that it, as well as RSK2, have been shown to mediate hematopoietic transformation (Bhullar and Sollars, 2011; Kang et al., 2007) suggest that the implications of this study extend beyond breast cancer.  120  3.5 Experimental Procedures Cell lines and treatments. H16N2 HMECs with tetracycline-inducible YB-1 (HTRY) or LacZ (HTRZ) have been previously described (Berquin et al., 2005). The HTRY-LT cell lines were developed in parallel by long-term (30-day) induction of YB-1 in HTRY cells passaged at low density. Transgene expression was achieved with 1 μg/ml doxycycline (Calbiochem, Gibbstown, NJ, USA) for 96 hours, unless otherwise noted. The normal HMECs 184hTERT (provided by Dr. Carl Barrett, National Cancer Institute) and MCF10A (ATCC, Manassas, VA, USA) were cultured in HuMEC media (Invitrogen, Burlington, Canada). The breast cancer cell lines MDA-MB-231 (ATCC) and SUM149 (Asterand, Detroit, MI, USA) were maintained in DMEM (Invitrogen) and Ham’s F12 media (Invitrogen), respectively. The inhibitors anacardic acid (Sigma-Aldrich, Oakville, Canada), BI-D1870 (synthesized by CDRD, Vancouver, Canada), LY-294002 (Calbiochem), and MG-132 (Calbiochem) were dissolved in DMSO. Plates were pre-coated with collagen (StemCell Technologies, Vancouver, Canada) prior to experiments involving BI-D1870. Quantitative real-time PCR. qRT-PCR was carried out with FAM-labeled Taqman Assays-on-Demand probes (Applied Biosystems, Carlsbad, CA, USA) on freshly isolated RNA as described (Astanehe et al., 2009). Results were analyzed with the Δ-ΔCt method normalized to TATA-binding protein (TBP) and compared to a comparator sample. Acini morphogenesis assay. Three-dimensional basement membrane cultures were setup using a well-established method (Muthuswamy et al., 2001). On Day 12, doxycycline was added at 1 μg/ml to assay media (F12 supplemented with 2% donor horse serum, 10 μg/ml insulin, 100 μg/ml hydrocortisone, 5 ng/ml EGF, and 2% Matrigel). Immunostaining of 3D cultures was carried out as previously described (Muthuswamy et al., 2001), except fixation was achieved using 100% methanol at -20°C for 2 hours. Images were acquired using a DeltaVision personal DV microscope (Applied Precision, Issaquah, WA, USA) and elaborated by softWoRx 5.0.0 deconvolution software. Immunoblotting, immunofluorescence, and immunoprecipitation. Immunoblotting, immunofluorescence, and immunoprecipitation were performed as described previously 121  (Davies et al., 2011). Images were acquired using an Olympus FV10i laser scanning confocal microscope (Richmond Hill, Canada). The origin and dilutions of antibodies are detailed in Table S3.1. Mammosphere assay. Cells were plated at a density of 2 x 104 into ultra-low attachment 6-well plates (Corning, Corning, NY, USA) in MammoCult Basal media (StemCell Technologies). Spheres with a minimum diameter of 50 μm (or >15 cells) were counted at Day 7. For serial passaging, mammospheres were collected by centrifugation at 350g for 2 minutes, dissociated with 0.25% trypsin, counted, and re-seeded. Differentiation culture conditions. Secondary mammospheres were dissociated into a single cell suspension and plated at a density of 10 viable cells/well in a collagen-coated 8-well chamber slide. After 7 days, cells were fixed in 100% methanol at -20°C for 20 minutes and stained with CK14 and CK18 primary antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA). HAT activity assay. HAT activity in 3 μg of nuclear lysate was measured using the HAT Assay Kit (Active Motif, Carlsbad, CA, USA) according to manufacturer’s instructions. Nuclear protein was isolated using the NE-PER Extraction Kit (Thermo Fisher Scientific, Rockford, IL, USA) and incubated in HAT assay buffer containing 0.5 mM acetyl-CoA and 50 μM histone H3 peptide for 15 minutes at room temperature. Fluorescence was measured at 460 nm with an excitation wavelength of 380 nm. Chromatin immunoprecipitation. Promoter complexes were isolated using the EZ-ChIP Kit (Millipore, Billerica, MA, USA) according to manufacturer’s instructions, with the following modifications. Cells were cross-linked with 1% formaldehyde for 10 minutes, rinsed with 5 mmol/L sodium butyrate (Sigma-Aldrich), and incubated in SDS cell lysis buffer for 10 minutes on ice. Chromatin was sonicated to an average length of 500bp and incubated overnight at 4°C with 1 μg of YB-1 (C-term; Epitomics, Burlingame, CA, USA), 5 μg acetyl-histone H3 (Lys9; Millipore), or isotype matched control antibody. The sequences, annealing positions, and optimized annealing temperatures of the primers used to amplify the immunoprecipitated DNA are provided in Table S3.2.  122  siRNA and plasmid transfections. Cells were transfected with 20 nM siRNA using Lipofectamine RNAiMAX (Invitrogen). The siRNA target sequences are provided in Table S3.3. Plasmid transfections (2 μg) were carried out using FuGENE HD (Roche, Laval, Canada). The 3xFLAG-YB-1WT, YB-1S102A, and YB-1S102D constructs have been described (Finkbeiner et al., 2009; Sutherland et al., 2005). pLenti6/V5-RSK2 was constructed by inserting full length RSK2 (provided by Dr. Levi Garraway, Massachusetts Institute of Technology) into the pLenti6/V5-DEST vector using Gateway Cloning. The pMIN-BMI1 construct was a gift from Dr. Sheila Singh (McMaster University). Anchorage-independent growth assay. Soft agar assays were preformed as previously described (Finkbeiner et al., 2009). Cells were embedded in 0.6% agar at a density of 5 x 103 per well in a 24-well plate. Colonies were counted at 28 days. Telomerase assay. The telomerase activity in 1 μg of total cell lysate was assessed using the Quantitative Telomerase Detection Kit (Allied Biotech, Vallejo, CA, USA) following the manufacturer’s instructions. Tumour xenografts. HTRZ and HTRY-LT #1 cells were injected bilaterally into the fourth inguinal mammary gland of 6-8 week old female immune-deficient NOD/SCID mice (Charles River, Wilmington, MA, USA). Five mice/group were injected with 105, 104, or 102 cells suspended in PBS containing 25% Matrigel and 10% trypan blue. Mice were humanely euthanized at 180 days post-injection. Histology and immunohistochemistry. A breast carcinoma progression tissue microarray of 7 patients (BrCaProg1; Cooperative Human Tissue Network, University of Virginia) and 83 fixed paraffin-embedded (FFPE) TNBC tissues (kindly provided by Dr. Anita Bane, McMaster University) were immunostained and analyzed by at least two of the authors. Immunohistochemistry was carried out via standard techniques (Fotovati et al., 2011) with YB-1 (N-term; a gift from Dr. Antony Braithwaite, University of Otago), pYB-1S102, pRSKS221/227, and CD44 antibodies diluted 1:200.  123  Gene expression profiling. RNA (100-250 ng) purified from cell lines and FFPE patient tissue (provided by Dr. Anita Bane) was used for analysis with the nCounter system, according to the manufacturer’s protocol (nanoString Technologies, Seattle, WA, USA). Complete normalized Affymetrix HG-U133A microarray data representing 579 cases of TNBC were assembled from the public Gene Expression Omnibus (GEO) database (GSE31519). Two-sided t tests determined significant differences in mean gene mRNA levels between groups of samples. Log-transformed, median centered expression values were visualized as colour maps using Java TreeView software, where appropriate. Invasion assay. 1 x 105 cells were plated onto Matrigel-coated Transwell chambers for 24 hours in F12 media against a 5% FBS gradient using a previously described protocol (Woo et al., 2007). Invaded cells were visualized using DAPI. Annexin V apoptosis assay. A single-cell suspension was incubated with PE-conjugated annexin V (1:20) on ice for 20 minutes according to manufacturer’s instructions (Annexin V:PE Apoptosis Detection Kit; BD Biosciences, Mississauga, Canada). The proportion of viable and apoptotic cells was detected by flow cytometry and analyzed using FlowJO software. Statistical analysis. All data is presented as mean ± standard deviation of at least three individual experiments. Significance was evaluated using a paired Student’s t-test, where *P < 0.05 and **P < 0.01. Statistical analyses were performed using SPSS 19.0 software.  124  3.6 Supplementary Data  4" 2"  **"  0"  " DA R 0M Y " B0 45 3" M CF HC 7" C M DA 114 0M 3" B0 4 HC 68" C1 93 SU 7" M 14 9"  YB1" AcJn"  *"  RZ  pRSKS221/227"  *"  6"  M  HT  50"  RSK1"  8"  HT  100"  Fold0change"in"YB01" transcript"  B" LCC6"  SUM149"  HTRY"  kDa" 100"  HTRZ"  A"  MDA0MB0231"  3.6.1 Supplementary figures  Figure S3.1 Characterization of ectopic YB-1 expression level in HTRY cells. (a) Immunoblot analysis of ectopic YB-1 expression level in HTRY cells following a 96hour induction compared to established breast cancer cell lines. RSK expression and activity was also measured. (b) Quantitative RT-PCR of YB-1 transcript level in induced HTRY cells and breast cancer cell lines. Gene expression is reported relative to normal HMECs (HTRZ cells).  125  BMI1$promoter$ ChIP%a%  ChIP%b%  ChIP%c%  (/4873%to% (/4509%to% /4494)% /4084)%  (/3460%to% /3037)%  Exon1%  ChIP%b%  (/2022%to% /1645)%  (/1067%to% /721)%  Exon1%  (/863%to% /615)%  CAAT% ATTG%  +1%  CD49f$promoter$ ChIP%a%  ATTG%  +1%  CD44$promoter$ ChIP%a%  CAAT%  ChIP%b%  Exon1%  (/389%to% /151)%  +1%  CAAT% ATTG%  Figure S3.2 Schematic diagrams of the BMI1, CD44, and CD49f promoters. Diagrams of human BMI1, CD44, and CD49f promoters displaying the target regions for amplification using ChIP primers. The numbers associated with each primer set denote the position of the amplicon in relation to the major transcriptional start site. Position +1 means the 5’ end of the RefSeq cDNA sequence (BMI1, NM_005180; CD44, NM_000610; CD49f, NM_000210). Putative YB-1 binding sites, an inverted CAAT box, ATTG, are identified. YB-1 binding sites on the CD44 and CD49f promoters have been previously described (To et al., 2010), while we used TFSEARCH to identify the binding sites in the human BMI1 promoter (ENSG00000168283).  126  Fold4change"in"HAT"ac>vity"  1.2" 1" 0.8" 0.6" 0.4"  **"  0.2" 0" DMSO"  15"μM"AA"  Figure S3.3 Anacardic acid inhibited HAT activity. HTRY cells were pre-treated with DMSO or anacardic acid (AA) for 4 hours prior to induction with doxycycline. HAT activity was significantly decreased (> 80%) in nuclear extracts prepared from AA treated cells. Data is presented as mean ± SD (n = 3).  127  IgG%  YB51%  HTRY%  HTRZ%  HTRY%  IgG%  10%%input%  MDA5MB5231% YB51% HTRZ%  HTRY%  HTRZ%  10%% input%  ChIP%a% ChIP%b% ChIP%c%  Figure S3.4 YB-1 binds to the promoter of BMI1. ChIP analysis targeting the promoter of BMI1 in induced HTRY and MDA-MB-231 cells. DNA templates were pulled down with either YB-1 or nonimmune IgG antibody and amplified using primers flanking putative YB-1 binding sites (termed a, b, and c). Input designates DNA isolated from cross-linked DNA/protein complexes before immunoprecipitation.  128  A"  p16INK4a%  Composite%  siBMI1%  +YB41%  Scr%  4YB41%  DAPI%  B" Percent%p164posiAve%nuclei%  60% 50% 40% 30% 20%  **%  10% 0%  Scr% 4YB41%  siBMI1% +YB41%  Figure S3.5 Upregulation of YB-1 decreased p16INK4a expression and nuclear localization. (a) Immunofluorescence with antibody targeting p16INK4a in HTRY cells induced concomitantly with scrambled peptide (Scr) or BMI1 siRNA for 96 hours. Nuclei were counterstained with DAPI. Uninduced HTRY cells (-DOX) served as a control. Scale bar = 50 μm. (b) Quantification of p16INK4a-positive nuclei. Cells in 5 random fields/experiment were analyzed. Data is presented as mean ± SD (n = 3).  129  B" 18" 15"  *"  12" 9"  *" **"  6"  *"  3"  HTRZ" HTRY" HTRY/LT"#1"  **"  6"  **"  *"  4"  *" *"  *" 2"  *"  0"  DA # /M 2" B/ 23 SU 1" M 14 9"  #1 "  T"  200" 100"  2" "#  HT R  Y/  LT  "#  1"  Y"  LT  HT R  Y/  HT R  "  0"  Number"of"mammospheres"  300"  RZ  **"  p16"  T"  **"  400"  HT  BMI1"  D"  **"  **"  500"  CD49f"  /L  /L  M 600"  CD44"  RY  HT  HT  RY  HT  HT  RZ  "  0"  C" HAT"acKvity"(pmol/min)"  Fold/change"in"transcript"  8"  **"  RY "  Fold/change"in"YB/1"transcript"  A"  200"  **"  150" 100"  **"  **"  **" **"  **"  **"  HTRZ" HTRY" HTRY/LT"#1"  **" *"  50"  **"  0" 1"  2"  3"  4"  5"  Passage"  Figure S3.6 TIC-associated gene expression, HAT activity, and mammosphere formation in the HTRY-LT cell lines. (a) Quantitative RT-PCR of YB-1 transcript in HTRZ, HTRY, and HTRY-LT cell lines. MDA-MB-231 and SUM149 cells were used as a control representative of triple-negative breast cancer. Data is presented as mean ± SD (n = 3). (b) CD44, CD49f, BMI1, and p16INK4a transcript was compared between HTRZ, HTRY, and HTRY-LT #1 cell lines using quantitative RT-PCR. (c) HAT activity was measured in nuclear extracts prepared from HTRZ, HTRY, and HTRY-LT cell lines. Data is presented as mean ± SD (n = 3). (d) HTRZ, HTRY, and HTRY-LT cells were grown in mammosphere cultures and serially passaged for five generations. Data is presented as mean ± SD (n = 3).  130  Percent"cell"viability"  HTRZ" HTRY)LT"  75" 50"  **"  **"  25"  C"  100"  HTRZ" HTRY)LT"#1" HTRY)LT"#2"  80" 60"  *"  40"  *"  20"  200"  100"  **" 50" 0"  0"  0"  DMSO"  1"  2"  5"  HTRY) LT"#1" HTRY) LT"#2"  150"  Colonies/well"  B"  100"  Percent"cell"viability"  A"  1"  10"  1"  2"  DMSO" BI)D1870"(μM)"  DMSO" BI)D1870"(μM)"  BI)D1870"(μM)"  E" 250"  DMSO"  200  200  YB)1" rpS6S235/6" Vinculin"  FL2-H: Annexin-PE  0.025  1000  0.01  8.52 1  10  2  10  3  10  FL2-H: Annexin-PE  0.045  91.2  0 0 10  4  10  1000  0.05  800  800  600  600  600  400  200  200  4.59 1  10  2  10  3  10  FL2-H: Annexin-PE  0  FSC-H  800  400  1000  0  10  10  2  3  10  FL2-H: Annexin-PE  0  10  1000  0  600  600  200  94.1  0 100  1000  5.89 1  10  2  10  3  10  FL2-H: Annexin-PE  0  FSC-H  600  FSC-H  800  200  1000  0  8.95 1  10  2  10  3  10  FL2-H: Annexin-PE  0  1000  0  800  600  600  600  200  200  92.4  0 100  7.61  101  102  103  FL2-H: Annexin-PE  104  0.055  9.88  10 1  102  103  104  FL2-H: Annexin-PE  0  83.4  10  800  400  4  10  0  0 100  4  800  400  3  200  91  10  10  400  0 100  4  2  0.32  0 10 0  4  800  400  10  FL2-H: Annexin-PE  89.7  5.03 1  800  400  8.8 1  200  94.9  10  10  400  0 0 10  4  0  200  0 0 10  4  10  FSC-H  50" 37"  3  10  FSC-H  72)hour"  FSC-H  FSC-H  BI)D1870" 1" 2" μM" pYB)1S102"  2  10  0  400  91.5  5.05 1  10  FSC-H  600  400  1000  96)hour"  siRSK2"  siRSK1"  Control"  BI)D1870" (μM)"  50"  100"  600  400  95.4  FSC-H  kDa"  600  0 0 10  6"Days"  F"  2"  2"μM"BI)D1870" 1000  0  800  1000  1"  0  800  94.9  **"  1"μM"BI)D1870" 1000  0  800  0 0 10  50" 0"  0  FSC-H  100"  FSC-H  Passage"2"  150"  48)hour"  200"  1000  FSC-H  Passage"1"  DMSO"  Number"of"mammospheres"  D"  2"  16.6  101  102  103  104  FL2-H: Annexin-PE  0  0  400 200  47.1  0 100  52.9  101  102  103  FL2-H: Annexin-PE  104  44.4  0 100  55.6  101  102  103  FL2-H: Annexin-PE  104  Figure S3.7 HTRY-LT cells were sensitive to RSK inhibition. (a-b) Viability of HTRZ and HTRY-LT cells treated with increasing concentration of BID1870 for (a) 4 days or (b) 8 days, with repeat dosing at Day 4, was assessed by Cellomics ArrayScan VTI. Hoechst 33342-stained nuclei were used as a measure of cellularity. Data is presented as mean ± SD (n = 3) and reported relative to DMSO control. (c) Soft agar colony formation of HTRY-LT cells grown in the presence of DMSO or BI-D1870. Drug was replenished weekly. Data is presented as mean ± SD (n = 3). (d) HTRY-LT #1 cells pre-treated for 72 hours with BI-D1870 or siRNA targeting RSK1 and RSK2 were grown in mammosphere conditions (passage 1). After 7 days,  131  mammospheres were serially passaged into secondary cultures absent of siRNA or drug (passage 2). Data is presented as mean ± SD (n = 3). (e) Annexin V staining of HTRYLT #1 cells treated with BI-D1870 analyzed by flow cytometry at the times indicated. A rightward shift indicates annexin-positive, apoptotic cells. (f) Immunoblot assessing the activity of the RSK substrates YB-1 and ribosomal protein S6 (rpS6) in HTRY-LT #1 cells at 96 hours post-BI-D1870 treatment.  132  184hTERT:YB31"  kDa"  184hTERT:EV"  MDA3MB3231"  A"  50"  YB31"  50"  pYB31S102"  100"  RSK1"  100"  RSK2"  100"  pRSKS221/227"  150"  Vinculin"  B"  C" 70"  Percent"invaded"cells"  Colonies/well"  16"  **"  60" 50" 40" 30" 20" 10"  14"  *"  12" 10" 8" 6" 4" 2" 0"  0" 184hTERT:EV" 184hTERT:YB31"  184hTERT:EV" 184hTERT:YB31"  Figure S3.8 184hTERT cells acquired a malignant phenotype following stable YB-1 expression. (a) Immunoblot assessing YB-1 and RSK expression and activation in 184hTERT cells stably expressing empty vector (EV) or FLAG-tagged YB-1. The MDA-MB-231 cell line was used as a control representative of triple-negative breast cancer. (b-c) Quantification of 184hTERT:EV and 184hTERT:YB-1 (b) cell growth under anchorage-independent conditions and (c) invasion through Matrigel-coated Transwell chambers. Data is presented as mean ± SD (n = 3).  133  Number"of"cells"(x105)"  5"  HTRZ" HTRY8LT"  4" 3" 2" 1" 0" 0"  24"  48"  72"  96"  Time"(hr)"  Figure S3.9 HTRZ and HTRY-LT growth kinetics. HTRZ and HTRY-LT #1 cells were plated at a density of 1.0 x 105 in a 6-well plate. At 24-hour intervals the number of viable cells was assessed by a haemocytometer-based trypan blue exclusion assay. Data is presented as mean ± SD (n = 3).  134  A"  DCIS&(high)&  H&E&  IDC&(low)&  YB@1&  IDC&(high)&  pYB@1S102&  ILC&&  pRSKS221/227&  Lymph&node&  CD44&  ILC&  IDC&(low)&  DCIS&(low)&  B"  DCIS&(low)&  Figure S3.10 YB-1 was expressed during breast cancer progression in vivo. (a) Representative IHC pictures of YB-1 expression in malignant mammary epithelium at various stages of breast cancer progression. (b) YB-1 staining coincides with pYB-1S102, pRSKS221/227, and CD44 expression in serial sections from DCIS, IDC, and ILC lesions. Scale bar = 100 μm.  135  3.6.2 Supplementary tables  Antibody α/β-tubulin BMI1 CD44 (clone EPR1013Y) CK14 (LL001) CK18 (H-80) EGFR ER (MC-20) FLAG M2 HER2 ubiquityl-Histone H2A (clone E6C5) Histone H3 (D2B12) Histone H3 (D1H2) p16INK4a (clone EP435Y) AF488-conjugated p16INK4a (F-12) p300 (N-15) PR (C-19) Integrin α6/CD49f Acetylated-lysine RSK1 (C-2) RSK2 (E-1) pRSKS221/227 pS6 ribosomal proteinS235/236 YB-1 (C-term) pYB-1S102 V5 Vinculin (clone V284) ZO-1  Dilution 1:1000 1:1000 1:1000 1:100 (IF) 1:100 (IF) 1:1000 1:1000 1:2000 1:1000 1:1000 1:50 (IP) 1:1000 (IB) 1:1000 (IB) 1:100 (IF) 1:1000 (IB); 1:100 (IF) 1:500 1:1000 (IB); 1:200 (IF) 1:1000 1:1000 1:500 1:1000 1:1000 1:1000 1:1000 1:2000 1:1000 1:200 (IF)  Supplier Cell Signaling Abcam Abcam Santa Cruz Santa Cruz StressGen Santa Cruz Sigma Abcam Millipore Cell Signaling Cell Signaling Millipore Santa Cruz Santa Cruz Santa Cruz Cell Signaling Cell Signaling Santa Cruz Santa Cruz Invitrogen Cell Signaling Epitomics Cell Signaling Invitrogen Millipore Zymed  Table S3.1 List of antibodies, dilutions, and suppliers used for immunoblotting (IB), immunofluorescence (IF), and immunoprecipitation (IP).  136  Primer set  Oligonucleotide sequencea  Amplicon sizeb  Ta (°C)c  F: 5’-GAAGGCTGTTCTTTAATTTACCAAG-3’ R: 5’-TGGAAATCCAGGACCTTTTG-3’ F: 5’-GGTCCTGGATTTCCATTCAA-3’ R: 5’-CAAGAAGAGGAGAGGGCAAA-3’ F: 5’-GAGGCGAGTTCCGAATCC-3’ R: 5’-GCTGGGAGCTGAGCGTATTA-3’  379 (nt -4873 to -4494) 425 (nt -4509 to -4084) 423 (nt -3460 to -3037)  58  F: 5’-CCCCATACCTGTAACCTCATGT-3’ R: 5’-CCAAACCCTATTATGGCTGCT-3’ F: 5’-TGCGTTTGATTTCCAAACAT-3’ R: 5’-TCCACCATCCTCTTCTCCAC-3’  377 (nt -2022 to -1645) 346 (nt -1067 to -721)  59  F: 5’-GAAGGGATAGCAAAGAAAAGAGG-3’ R: 5’-TGAATGGCTAACCTCCATGT-3’ F: 5’-CCCACTTTCACACTGATGTTCT-3’ R: 5’-ATTCCCCAAAGTGGCACTAA-3’  248 (nt -863 to -615) 238 (nt -389 to -151)  57  BMI1 A B C  58 58  CD44 A B  57  CD49f A B  59  Table S3.2 Primers used for ChIP assays. a  The sense/forward (F) and antisense/reverse (R) primer sequences  b  The genomic position is the location of the 5’ nucleotide of the primer in relation to the  major transcriptional start site c  Optimal annealing temperature (Ta) used in PCR  137  siRNA  siRNA target sequence (5’ – 3’)  Manufacturer  BMI1  ATGGGTCATCAGCAACTTCTT  Qiagen  hTERT  CCCGGTGTACGCCGAGACCAA  Qiagen  RSK1  CCCAACATCATCACTCTGAAA  Qiagen  RSK2  AGCGCTGAGAATGGACAGCAA  Qiagen  YB-1 #1  CCACGCAATTACCAGCAAA  Dharmacon  YB-1 #2  Hs_YBX1_1_HP Validated siRNA  Qiagen  Scrambled  AATTCTCCGAACGTGTCACGT  Qiagen  Table S3.3 siRNA target sequences.  138  CHAPTER 4. YB-1 ENHANCES ABCG2 EXPRESSION IN BREAST CANCER TO ELICIT CHEMORESISTANCE THAT CAN BE CIRCUMVENTED USING RSK INHIBITORS  4.1 Overview Despite advances in the treatment and management of breast cancer, recurrence remains a significant problem. Current dogma dictates that populations of drug resistant cells within a heterogeneous tumour are responsible for relapse. In this study, we demonstrate that the oncogenic transcription factor Y-box binding protein-1 (YB-1) regulates the expression of ABCG2, an ATP-binding cassette transporter involved in multidrug resistance. Introducing YB-1 into human mammary epithelial cells (HMECs) was associated with a striking increase in ABCG2 expression. Accordingly, these cells acquired resistance to chemotherapeutic substrates of ABCG2, notably 5-fluorouracil, doxorubicin, and gefitinib. We discovered that the upregulation of ABCG2 was contingent upon p300mediated promoter-centered chromatin relaxation, which allowed for subsequent YB-1 binding and transcriptional regulation. As such, multidrug resistance could be reversed by inhibition of p300 activity using anacardic acid. Moreover, repressing YB-1S102 activation using the small molecule RSK inhibitors BI-D1870 and luteolin also led to a decrease in ABCG2 expression and a re-sensitization of cells to chemotherapy. Interestingly, we found that sustained treatment with luteolin induced apoptosis in triple-negative breast cancer cell lines while displaying virtually no toxicity toward normal, non-tumourigenic HMECs. Relapse is believed to initiate from a drug resistant CD44 and CD49f doublepositive (CD44+/CD49f+) stem/progenitor-like subpopulation within a heterogeneous tumour. Sorting for these cells revealed them to have an over 2-fold increase in ABCG2 expression relative to bulk tumour cells and, as a result, they were refractory to chemotherapy. In stark contrast the CD44+/CD49f+ fraction was exquisitely sensitive to the RSK inhibitors BI-D1870 and luteolin. Therefore, targeting the RSK/YB-1 pathway inhibits ABCG2 and can be used to overcome multidrug resistance.  139  4.2 Introduction Despite recent improvements in breast cancer mortality, many patients relapse after an initial response to chemotherapy. This is especially prominent in the triple-negative breast cancer (TNBC) subtype, which is characterized by the lack of estrogen receptor, progesterone receptor, and the HER2 receptor tyrosine kinase (Gluz et al., 2009a). Accordingly, currently viable targeted therapies, such as endocrine and anti-HER2 agents, are ineffective in these cancers. While TNBC accounts for approximately 15% of all breast cancer diagnosis, it is responsible for a disproportionate share of disease-related death (Carey et al., 2006). Paradoxically, these tumours demonstrate the highest response rate to neoadjuvant chemotherapy but despite this initial responsiveness recurrence is common within five years following treatment (Carey et al., 2007; Dent et al., 2007). Several hypotheses have been proposed to explain this treatment failure and recurrence. In particular, it has been suggested that a small subpopulation of cells within a tumour, referred to as tumour initiating cells (TICs), are resistant to therapy and have the capacity to reinitiate tumour growth following treatment (Nguyen et al., 2012; Visvader and Lindeman, 2012). These cells can be isolated using cell surface marker expression by FACS. For example, in cells from primary breast cancers, the CD44+/CD24-/Linsubpopulation was enriched for TICs with the ability to form tumours in mice at limiting dilutions (Al-Hajj et al., 2003). Refinement in marker selection has established that high expression of CD44 and CD49f can reproducibly identify and isolate TICs from breast cancer cell lines and biopsies (Meyer et al., 2010). These cells have an enhanced mammosphere-forming ability, an in vitro surrogate assay of self-renewal capacity (To et al., 2010). Moreover, the TIC subpopulation appears to be relatively resistant to chemotherapy (Creighton et al., 2009; Li et al., 2008; Phillips et al., 2006) and is actually expanded following treatment (Fillmore and Kuperwasser, 2008). This could explain the high frequency of TICs in TNBC relative to other breast cancer subtypes (Honeth et al., 2008; Park et al., 2010). The unique ability of TICs to elude chemotherapy is believed to be associated with their high expression of ATP-binding cassette (ABC) drug transporters (Dean et al., 2005). Of particular importance is ABCG2, also referred to as breast cancer resistance protein 140  (BCRP). This efflux pump was originally identified in MCF-7 cells resistant to doxorubicin (Doyle et al., 1998) and has since been shown to effectively reduce the intracellular concentration of several prominent chemotherapeutic agents used in the treatment of breast cancer, including 5-fluorouracil, gefitinib, methotrexate, and paclitaxel (An and Ongkeko, 2009; Chen et al., 2011; Guo et al., 2004; Yuan et al., 2009). The strong association between ABCG2 and TICs first emerged from studies where it was found to be the key determinant of a highly tumourigenic subpopulation of cells called the side-population (Hirschmann-Jax et al., 2004; Zhou et al., 2001). Supporting its role as the primary regulator of TIC chemoresistance, ABCG2 has been associated with poor prognosis and recurrence in leukemia and a variety of solid tumours (Ross and Nakanishi, 2010). Specifically, in breast cancer ABCG2 is inversely correlated to the efficacy of anthracycline-based chemotherapy (Burger et al., 2003). It is also part of a gene signature predictive of poor response to neoadjuvant chemotherapy (Park et al., 2006). ABCG2 is most strongly associated with the TNBC subtype, which could rationalize their high rate of recurrence (Britton et al., 2012). Targeting the drug-resistant TIC subpopulation is critical to prevent tumour regeneration post-chemotherapy. With this objective in mind, our group has discovered that these cells have a strong dependency on the oncogenic transcription factor Y-box binding protein-1 (YB-1) (Dhillon et al., 2010; Fotovati et al., 2011; Stratford et al., 2012; Stratford et al., 2010; To et al., 2010). Repression of YB-1 was found to suppress CD44 and CD49f expression, mammosphere formation, and soft agar colony growth (To et al., 2010). On the other hand, overexpression yielded tumours in mice that were enriched in CD44positive cells. Both activated pYB-1S102 and p90 ribosomal S6 kinase (RSK), its potent upstream effector, are highly expressed in the CD44+/CD24- TIC population (Stratford et al., 2012). Interestingly, the loss of YB-1 sensitized these cells to paclitaxel (To et al., 2010) and trastuzumab (Dhillon et al., 2010). In light of this finding our current work addresses the mechanism by which YB-1 mediates chemoresistance, with a particular emphasis on the TIC population.  141  4.3 Results 4.3.1 YB-1 is a direct transcriptional regulator of ABCG2 To decipher the role of YB-1 in mediating drug resistance we utilized a Tet-On inducible expression system that had been previously engineered and characterized by our group (Berquin et al., 2005; Davies et al., 2011). Non-malignant, karyotypically normal H16N2 human mammary epithelial cells (HMECs) expressing YB-1 under the control of a tetracycline-inducible promoter were designated HMEC Tet-repressed YB-1 (HTRY). The ectopic expression level of the YB-1 transgene closely mimicked that observed in breast cancer cell lines and patient tissues supporting the physiological relevance of this model (see chapter 3). A second cell line conditionally expressing LacZ was referred to as HMEC Tet-repressed LacZ (HTRZ) and served as a control. In contrast to MDR1, ABCG2 has been associated with clinical drug resistance and, in particular, chemoresistance in the TIC population (Doyle and Ross, 2003). This prompted us to investigate the relationship between this multidrug efflux pump and YB-1. Following a 96-hour YB-1 induction we measured an increase of over 20-fold in ABCG2 mRNA expression by qRT-PCR (Figure 4.1A). Correspondingly, ABCG2 protein was elevated in HTRY cells compared to HTRZ cells (Figure 4.1B). These findings raised the possibility that YB-1 may be directly binding to the ABCG2 promoter to regulate gene transcription. A single YB-1 binding site, an inverted CCAAT box, was identified 273 base pairs upstream of the major transcriptional start site within the proximal ABCG2 promoter. Conventional ChIP analysis provided direct evidence of the ability of YB-1 to bind to the ABCG2 promoter in HTRY cells (Figure 4.1C). In further support, we validated this interaction in the MDA-MB-231 breast cancer cell line (Figure 4.1C).  142  kDa"  20"  ABCG2"  75"  10"  Vinculin"  100"  0" HTRY"  Distal"  Proximal"  YB/1"inducible"system"  MDA/MB/231"  HTRY"  IgG"  HTRZ"  HTRY"  YB/1"  HTRZ"  HTRY"  10%"Input"  ABCG2"  CCAAT" /273"  IgG"  C"  YB/1"  HTRZ"  HTRZ"  HTRY"  B"  **"  HTRZ"  30"  10%"input"  Fold/change"in"ABCG2" transcript"  A"  Figure 4.1 YB-1 transcriptionally regulated ABCG2. (a) Quantitative RT-PCR of ABCG2 transcript in induced HTRY and HTRZ cells. (b) Immunoblot analysis of ABCG2 in induced HTRY and HTRZ cells. Vinculin was included as a loading control. (c) ChIP targeting the ABCG2 promoter in induced HTRZ and HTRY cells as well as MDA-MB-231 cells. DNA templates were pulled down with YB-1 or nonimmune IgG antibody. The final precipitated DNA was amplified using primers flanking the putative YB-1 binding site, an inverted CCAAT box, located 273bp upstream of the ABCG2 transcriptional start site (GenBank accession no. AF151530).  143  4.3.2 The ABCG2 promoter is hyperacetylated in YB-1 expressing cells We have previously identified a role for YB-1 in epigenetic gene regulation through the activation and recruitment of the histone acetyltransferase (HAT) protein p300. Accordingly, we evaluated ABCG2 promoter-associated histone H3 acetylation in HTRZ and HTRY cells. Primer pairs targeting the proximal (A-D; nt -687 to +20) and distal (nt -1527 to -1268) promoters were designed (To et al., 2006). To attribute changes in acetylation to p300, we performed ChIP assays using an antibody to detect acetylation of histone H3 at lysine 9 (AcH3-K9). This residue is preferentially acetylated by p300 (Szerlong et al., 2010; Zhong and Jin, 2009). The ABCG2 promoter was hyperacetylated in HTRY cells relative to HTRZ cells. Specifically, we detected elevated levels of histone H3 lysine 9 acetylation along the distal promoter and regions A and D of the proximal promoter (Figure 4.2A).  4.3.3 Inhibition of p300 HAT activity prevents YB-1 binding to the ABCG2 promoter We hypothesized that YB-1 binding to the ABCG2 promoter was conditional upon chromatin relaxation facilitated by p300-mediated acetylation. To address this, HTRY cells were pre-treated with anacardic acid (AA), a potent inhibitor of p300 HAT activity (Balasubramanyam et al., 2003), prior to YB-1 induction. This hindered the ability of YB-1 to bind the ABCG2 promoter as assessed by ChIP (Figure 4.2B). Consequently, ABCG2 transcript was repressed by nearly 50% (Figure 4.2C). ABCG2 protein was also diminished (Figure 4.2D). Next, we questioned whether inhibition of HAT activity would be sufficient to increase the efficacy of the ABCG2 substrates 5-fluorouracil, gefitinib, and doxorubicin. We selected these chemotherapeutics based on the fact that they are commonly used in the treatment of TNBC (Gluz et al., 2009a). Importantly, the drug concentrations used in these studies are physiologically achievable (Diasio and Harris, 1989; Speth et al., 1988; Swaisland et al., 2005). Treating HTRY cells for 96 hours with 50 μM 5-fluorouracil, 10 μM gefitinib, or 10 nM doxorubicin had only a marginal (< 20%) effect on reducing cell viability (Figure 4.2E). In contrast, when HTRY cells were first pre-treated with AA for  144  four hours prior to the addition of drug cell viability decreased by as much as 80% (Figure 4.2E). To demonstrate that the reversal in drug resistance could be directly attributed to the loss of ABCG2 we pre-treated HTRY cells with fumitremorgin C (FTC), a specific ABCG2 inhibitor (Rabindran et al., 2000). When subsequently treated with the chemotherapeutics cell viability was reduced by over 70%, directly emulating the results observed with AA (Figure 4.2E). AA and FTC had no effect on the cells alone.  145  HTRY'  HTRZ' AA'  IgG'  DMSO'  Distal"  AcH39K9'  HTRZ'  Proximal" A" B" 9687'9579' 9428'9271' C" 91527' 91268' D" 9293'9139' 9146'+20'  HTRZ'  ABCG2"  HTRY'  10%' Input'  HTRY'  A"  Distal' A' B' C'  C"  IgG'  AA'  AA'  DMSO'  AA'  YB91'  DMSO'  10%' Input'  DMSO'  B"  Fold9change'in'ABCG2' transcript'  D'  D"  1'  **'  kDa'  0.5'  75'  ABCG2'  100'  Vinculin'  0' DMSO' 15'μM'AA'  120' 100' 80' 60'  *'  40'  15'μM'AA'  10'nM'DXR'  10'μM'GefiQnib'  50'μM'59Fu'  10'nM'DXR'  10'μM'GefiQnib'  50'μM'59Fu'  10'nM'DXR'  10'μM'GefiQnib'  50'μM'59Fu'  10'μM'FTC'  0'  15'μM'AA'  20'  DMSO'  Percent'cell'viability'  E"  10'μM'FTC'  Figure 4.2 ABCG2 transcription was dependent on promoter acetylation. (a) ChIP analysis of the ABCG2 promoter in induced HTRZ and HTRY cells using acetylated histone H3 (Lys9; AcH3-K9) or control IgG antibody. DNA templates were amplified with the designated primer pairs. (b) ChIP of HTRY cells pre-treated with DMSO or anacardic acid (AA; 15 μM) for 4 hours prior to induction. DNA templates were pulled down with YB-1 or IgG antibody and amplified using primers flanking the YB-1 binding site in the ABCG2 promoter. (c) qRT-PCR and (d) immunoblot analysis of HTRY cells pre-treated for 4 hours with DMSO or AA (15 μM) prior to induction. (e) Viability of induced HTRY cells treated with AA, fumitremorgin C (FTC), 5-fluorouracil (5-Fu), gefitinib, or doxorubicin (DXR) for 96 hours was measured by Hoechst 33342 using Cellomics ArrayScan VTI. For combination treatments, AA and FTC were added 4 hours prior to addition of drug. Data is presented as mean ± SD (n = 3) relative to DMSO.  146  4.3.4 RSK inhibitors can reverse drug resistance and re-sensitize cells to ABCG2 substrates We have previously shown that RSK phosphorylates YB-1 on its serine-102 residue, which triggers nuclear localization and activation of its DNA binding ability (Stratford et al., 2008). Coupled with our finding that YB-1 binds to the ABCG2 promoter, we rationalized that RSK inhibitors could prevent the induction of ABCG2 and reverse multidrug resistance. Treating HTRY cells with BI-D1870, a specific RSK inhibitor (Sapkota et al., 2007), yielded a loss of YB-1S102 phosphorylation and consequently ABCG2 expression (Figure 4.3A). The same effect was achieved using luteolin, a bioflavonoid that we have recently identified as a RSK inhibitor through in silico molecular docking and in vitro kinase assays (K. Reipas and S.E. Dunn, manuscript submitted for publication) (Figure 4.3A). To ascertain whether the loss of ABCG2 following RSK inhibition was sufficient to re-sensitize cells to chemotherapy, we treated HTRY cells with BI-D1870 and luteolin in combination with 5-fluorouracil, gefitinib, and doxorubicin. A 96-hour treatment with 1 μM BI-D1870 or 10 μM luteolin had only a marginal effect (~25%) on decreasing cell viability. However, when administered simultaneously with 50 μM 5-fluorouracil, 10 μM gefitinib, or 10 nM doxorubicin cell viability decreased by as much as 85% (Figure 4.3B). Taken together, we have developed a robust model to conceptualize how YB-1 mediates multidrug resistance in breast cancer. The stabilization and recruitment of p300 to the ABCG2 promoter by YB-1 yields an open chromatin conformation exposing transcription factor binding sites. Concomitantly, activation (phosphorylation) of YB-1 by RSK allows it to bind to an inverted CCAAT box within the proximal ABCG2 promoter. The resultant upregulation of ABCG2 confers chemotherapeutic resistance, which can be overcome using RSK inhibitors (Figure 4.3C).  147  *"  40"  1"μM"BI'D1870"  10"nM"DXR"  10"μM"Gefi=nib"  50"μM"5'Fu"  0"  10"nM"DXR"  20"  10"μM"Gefi=nib"  Vinculin"  60"  50"μM"5'Fu"  100"  80"  10"μM"Luteolin"  ABCG2"  100"  1"μM"BI'D1870"  75"  Percent"cell"viability"  50"  pYB'1S102"  120"  DMSO"  B"  10"μM"Luteolin"  1"μM"BI'D1870"  kDa"  DMSO"  A"  10"μM"Luteolin"  C" Luteolin" BI1D1870" AA"  RSK"  p300"  Ac" H3"  P"  5'FU" DXR" GEF" FTC"  YB'1"  ABCG2"  CCAAT"  Figure 4.3 RSK inhibitors reversed ABCG2-mediated multidrug resistance. (a) Immunoblot analysis of induced HTRY cells treated for 48 hours with DMSO, BID1870, or luteolin. (b) Viability of induced HTRY cells treated with BI-D1870 and luteolin alone or in combination with 5-Fluorouracil (5-Fu), gefitinib, or doxorubicin (DXR) for 96 hours. Hoechst 33342-stained nuclei were used as a measure of cellularity and assessed on Cellomics ArrayScan VTI. Data is presented as mean ± SD (n = 3) relative to DMSO control. (c) Schematic model of how co-operation between YB-1 and p300 elicits ABCG2-mediated multidrug resistance. Small molecule inhibitors targeting p300 (AA, anacardic acid), RSK (luteolin and BI-D1870), and ABCG2 (FTC, fumitremorgin C) can reverse drug resistance.  148  4.3.5 RSK inhibition is sufficient to induce apoptosis in TNBC cell lines The knowledge that ABCG2 protects YB-1-expressing breast cancer cells from a wide variety of chemotherapeutic drugs prompted us to investigate strategies to overcome this resistance mechanism. There is a growing body of work supporting the use of RSK inhibitors in breast cancer treatment (Stratford and Dunn, 2011). To assess the efficacy of these inhibitors under conditions that closely mimic the mammary microenvironment, we cultured HTRY cells on a reconstituted basement membrane (Matrigel). These cells became organized as three-dimensional polarized acini structures with a hollow lumen. Induction of YB-1 led to cells invading into the luminal space, a hallmark of ductal carcinoma in situ. However, luminal cell translocation could be prevented with luteolin (Figure 4.4A). Up until this point we have utilized the HTRY cell model, which mimics pre-malignant ductal carcinoma. We next asked whether RSK inhibitors would be equally effective in eradicating fully transformed cells. To address this, we used a model of TNBC generated following long-term YB-1 induction in HTRY cells (Figure 4.4B). Two cell lines, termed HTRY-LT #1 and #2, expressed YB-1 to the same degree as TNBC patient tissues, had a TNBC subtype gene signature, and were tumourigenic in vivo (see chapter 3). Concordant with YB-1, they were also found to express high levels of ABCG2 (Figure S4.1). Treating the HTRY-LT cell lines for 96 hours with a single dose of luteolin (10 and 50 μM) resulted in a modest decrease in cell viability (Figure S4.2). Therefore, we extended the treatment time to 8 days, with repeat dosing at Day 4. This yielded a reduction in HTRY-LT cell viability by nearly 90%, yet virtually no toxicity was observed in the parental, non-malignant HTRZ cells (Figure 4.4C). These findings raised the question as to whether the decrease in HTRY-LT cell viability was attributed to apoptosis. A shift toward annexin V-positive, apoptotic cells was observed at 72 hours following luteolin treatment (Figure 4.4D). With the abovementioned experiments conducted in monolayer cultures, we wanted to assess whether the growth inhibitory effects of luteolin would translate to anchorage-  149  independent conditions. HTRY-LT cells were seeded into soft agar assays and treated weekly with drug. Relative to the DMSO control, colony growth was dramatically reduced in luteolin-treated cells (Figure 4.4E). To validate that luteolin was functioning by diminishing RSK activity in HTRY-LT cells we measured decreased phosphorylation of two RSK substrates, YB-1 and S6 ribosomal protein (Figure 4.4F). As described in our previous work, decreased soft agar colony formation could also be achieved using siRNA oligos targeting RSK1/2 (see chapter 3). Together, this suggests that the observed growth inhibition can be directly attributed to the ability of luteolin to block RSK activity.  150  B" CD49f" ZOB1" DAPI"  100"  25"  HTRY"  TNBC"  HTRYBLT"  0"  YBB1"  50"μM" Luteolin"  ABCG2"  +YBB1"  D" HTRZ" HTRYBLT"#1" HTRYBLT"#2"  120" 100" 80" 60" 40" 20" 0"  Luteolin" DMSO" 1000  **" **"  600  600  600  400  400  200  200  F"  **"  50" 0" DMSO"  10"μM" 50"μM" Luteolin" Luteolin"  7.04 101  102  103  FL2-H: AnnexinV:PE  0  400 200  86.2  104  0  0 100  13.8  101  102  103  FL2-H: AnnexinV:PE  104  82.7  0 100  17.3  101  102  103  FL2-H: AnnexinV:PE  104  Luteolin" kDa"  100"  50"μM" 1000  0  800  HTRYBLT"#2"  150"  0  800  93  HTRYBLT"#1"  200"  1000  0  800  0 100  10"μM" 50"μM" Luteolin" Luteolin"  10"μM"  0  FSC-H  DMSO"  FSC-H  BYBB1"  DMSO"  Percent"cell"viability"  HTRZ"  **"  DMSO"  Colonies/well"  PreBmalignant" carcinoma"  50"  C"  E"  Normal"duct"  75"  FSC-H  Percent"hollow"acini"  A"  10" 50" μM"  50"  pYBB1S102"  50"  YBB1"  37"  rpS6S235/236"  100"  Vinculin"  Figure 4.4 Inhibiting RSK suppressed the growth of TNBC cell lines. (a) HTRY cells were grown as 3D acini on a reconstituted basement membrane for 14 days after which YB-1 was induced (+DOX) along with the addition of DMSO or luteolin. After 96 hours acini were immunostained with ZO-1 (luminal) and CD49f (basolateral) antibodies and luminal filling was quantified. Data represent mean ± SD (n = 3). Scale bar = 50 μm. (b) Depiction of YB-1-driven breast cancer progression. (c) Viability of cells treated with luteolin for 8 days, with repeat dosing at Day 4, was assessed by Hoechst 33342-stained nuclei using Cellomics ArrayScan VTI. Data represent mean ± SD (n = 3) relative to DMSO. (d) Annexin V staining of HTRY-LT #1 cells treated with luteolin analyzed by flow cytometry at 72 hours. (e) Soft agar colony formation of HTRY-LT cells grown in the presence of DMSO or luteolin. Drug was replenished weekly. Data is presented as mean ± SD (n = 3). (f) Immunoblot analysis of RSK substrate activity in HTRY-LT #1 cells treated for 72 hours with DMSO or luteolin.  151  4.3.6 RSK inhibitors are effective in eliminating the TIC subpopulation Due to the role of TICs in mediating tumour recurrence, we wanted to evaluate whether luteolin was effective in not only eradicating bulk tumour cells but also this highly tumourigenic subpopulation. HTRY-LT cells were cultured in mammosphere conditions to enrich for TICs (Ponti et al., 2005). Treatment with luteolin not only suppressed mammosphere growth, but also diminished their ability to be serially passaged (Figure 4.5A). These results could be mirrored using siRNA oligos targeting RSK1/2, as described previously (see chapter 3). This suggests that the TIC population is dependent on RSK expression and/or activity for survival. To rigorously assess the efficacy of RSK inhibitors against TICs, we fractionated HTRYLT cells into CD44+/CD49f+ TIC-enriched and CD44-/CD49f- populations by FACS. Sizable separation in CD44 and CD49f fluorescent intensity confirmed fractionation into two unique populations defined by high/high and low/low marker expression (Figure 4.5B). As expected, CD44+/CD49f+ cells had an enhanced ability to form mammospheres (Figure 4.5C) and expressed 2-fold higher CD44 (Figure 4.5D) relative to the doublenegative population. These cells also expressed a greater amount of ABCG2 (Figure 4.5D). In agreement, paclitaxel, 5-florouracil, gefitinib, and doxorubicin were relatively effective in eliminating the CD44-/CD49f- population; however, the CD44+/CD49f+ TICs were highly resistant (Figure 4.5E). In stark contrast, inhibiting the RSK/YB-1 axis by treatment with BI-D1870 or luteolin suppressed the growth of both populations (Figure 4.5E).  4.3.7 High expression of ABCG2 is associated with clinical recurrence As we found ABCG2 to be a mediator of drug resistance in TICs, we questioned its importance in patient samples. ABCG2 was highly expressed in X43 cells, a primary cell line derived from a relapsed TNBC patient (Figure 4.5F). While this finding will need to be expanded to a larger cohort, it suggests that ABCG2 could be a clinically relevant predictor of treatment failure and recurrence in TNBC.  152  % of M  60 40 20 0  0102 103 104 105 <PE-A>: CD44  B"  0102 103 104 105  0  20"  3.5"  *"  3"  E"  80 60  ABCG2"  2.5"  40  2"  20  1.5"  0 010  1"  2  0.5"  CD44+/ CD49f+"  80" 60"  10 40" 104 105  **"  3  <FITC-A>: CD49f  20"  RSK"inhibitors" X43"  MDA7MB7231"  CD447" CD44+" CD49f7" CD49f+"  CD447/ CD49f7"  100"  0"  0"  0"  120"  CD44"  50"μM"Luteolin"  100  0102 103 104 105 <FITC-A>: CD49f CD49f"  5"nM"Paclitaxel"  40"  50"μM" Luteolin"  DMSO"  **"  60"  SUM149"  40  CD44"  Percent"cell"viability"  80"  F"  60  <PE-A>: CD44  D"  hi"  2"μM"BI7D1870"  0  10"μM" Luteolin"  lo"  20  0"  % of Max  Number"of"mammospheres"  40 20  50"  CD447"" CD44+"" CD49f7" CD49f+"  kDa"  60  80  10"nM"DXR"  **"  hi"  10"μM"Gefi<nib"  % of Max  150"  DMSO"  C"  80  Passage"2"  100"  lo"  50"μM"57Fu"  200"  100  100  Passage"1"  % of Max  250"  Fold7change"in"transcript"  Number"of"mammospheres"  A"  75"  ABCG2"  37"  Ac<n"  Figure 4.5 TICs were sensitive to RSK inhibition. (a) Mammosphere cultures of HTRY-LT #1 cells grown in the presence of luteolin (passage 1). After 7 days, mammospheres were serially passaged into secondary cultures absent of drug (passage 2). Data is presented as mean ± SD (n = 3). (b) FACS histograms depicting CD44 and CD49f signal separation between the CD44/CD49f double-positive and double-negative populations. (c) Mammosphere assay of sorted HTRY-LT #1 cells. (d) qRT-PCR analysis of CD44 and ABCG2 in sorted HTRY-LT #1 cells. (e) HTRY-LT #1 cells were fractionated into CD44+/CD49f+ and CD44-/CD49f- populations by FACS and treated for 96 hours with paclitaxel, 5-fluorouracil (5-Fu), gefitinib, doxorubicin (DXR), BI-D1870, and luteolin. Cell viability was measured by Hoechst 33342 using Cellomics ArrayScan VTI. Data is presented as mean ± SD (n = 3). (f) Immunoblot analysis of ABCG2 in a primary cell line derived from a relapsed TNBC patient (X43).  153  4.4 Discussion Expression of YB-1 is associated with poor prognosis and relapse in a multitude of cancers, including breast (Habibi et al., 2008), prostrate (Gimenez-Bonafe et al., 2004), ovarian (Kamura et al., 1999), and brain (Faury et al., 2007). This has historically been attributed to intrinsic multidrug resistance conferred by YB-1 controlling multidrug resistance protein-1 (MDR1/Pgp) gene transcription (Bargou et al., 1997). However, it has since been shown that MDR1 has little prognostic value in predicting relapse (Larkin et al., 2004) and inhibitors have demonstrated virtually no benefit in clinical trials (Szakacs et al., 2006). As such, the molecular underpinnings responsible for the strong correlation between YB-1 and relapse are not understood. In the present study, we discovered a relationship between YB-1 and ABCG2 principal in evoking multidrug resistance in TNBC. Ectopic expression of YB-1 in HMECs protected them against a variety of chemotherapeutics, including 5-fluorouracil, doxorubicin, and gefitinib. We provide compelling evidence that YB-1 facilitates drug resistance via transcriptional regulation of ABCG2 that is conditional upon p300-mediated promoter hyperacetylation. Interestingly, our previous study showed treatment with paclitaxel yielded nuclear localization of activated YB-1 in TNBC cell lines resulting in therapeutic refractoriness (To et al., 2010). One possible explanation for this is that chemotherapy stimulates YB-1 transcription of ABCG2 to reduce intracellular drug concentration. This ability to protect cells could rationalize why nuclear YB-1, in particular, is such a strong prognostic marker for poor survival (Dahl et al., 2009; Habibi et al., 2008; Maciejczyk et al., 2012). Our findings likely transcend breast cancer as YB-1 has also been implicated in mediating resistance to doxorubicin in multiple myeloma (Chatterjee et al., 2008) and cisplatin in ovarian carcinoma (Yahata et al., 2002). Women diagnosed with primary TNBC commonly receive CMF (cyclophosphamide, methotrexate, 5- fluorouracil) chemotherapy, yet over 50% relapse within four years (Demicheli et al., 2005). As both methotrexate and 5-fluorouracil are ABCG2 substrates it is conceivable that the multidrug transporter is protecting a subpopulation of cells that ultimately regenerate the tumour post-chemotherapy. In support, we demonstrate that ABCG2 is enriched in TICs relative to the bulk tumour population. Moreover, we 154  provide clinical evidence that ABCG2 is highly expressed in tumours that relapse. Complementing these findings, ABCG2 has been described to drive resistance in a subset of cells responsible for the reemergence of breast tumour growth following tamoxifen treatment (Selever et al., 2011). Thus, we suggest that ABCG2 may be a useful prognostic marker to predict the risk of relapse in patients with TNBC. One potential strategy to prevent relapse is to target the ABCG2-overexpressing TIC population. However, to date, no specific ABCG2 inhibitors have advanced into clinical trials (Dean et al., 2005). This is largely related to toxicity associated with these compounds do to the ubiquitous nature of ABCG2 in hematopoietic stem cells, the liver, and intestines (Maliepaard et al., 2001). To circumvent these side effects we rationalized that targeting YB-1 activation by RSK could serve as a surrogate for ABCG2 inhibitors. YB-1 and RSK are especially attractive therapeutic targets because they are both expressed at a high frequency in TNBC, but are absent in normal tissue (Dahl et al., 2009; Stratford et al., 2012). Inhibiting the activation of YB-1 using small molecule RSK inhibitors was sufficient to not only reverse drug resistance, but also induce apoptosis in TNBC cell lines. Notably, both BI-D1870 and luteolin were highly effective in eliminating the ABCG2-enriched, TIC subpopulation suggesting that they could help curb relapse. This finding is collaborated by a study in which luteolin had an antiproliferative effect in ABCG2-overexpressing cell lines (Rao et al., 2012). Our prior work supports RSK inhibition as the primary mode of action. Despite substantial in vitro evidence demonstrating the efficacy of RSK inhibitors in restraining tumourigenesis, they have yet to progress into clinical trials (Stratford and Dunn, 2011). In an effort to expedite these compounds into the clinic we report that repositioning luteolin represents a straightforward and feasible therapeutic option for patients with TNBC. In conclusion, we discovered that the emergence of drug resistance in TNBC is mediated, in part, by the direct regulation of ABCG2 by YB-1. To contend with recurrence we propose that RSK inhibitors could reverse drug resistance by inhibiting YB-1 from binding to the ABCG2 promoter. The unique ability of these drugs to target the TIC subpopulation might provide sustained remission.  155  4.5 Experimental Procedures Cell lines. H16N2 HMECs expressing YB-1 (HTRY) or LacZ (HTRZ) under the control of a tetracycline-inducible promoter were generated as previously described (Berquin et al., 2005). Transgene expression was achieved by supplementing media with 1 μg/ml doxycycline (Calbiochem, Gibbstown, NJ, USA). Unless otherwise specified, cells were induced for 96 hours prior to assessment of biological/biochemical effects or drug treatments. The HTRY-LT cell lines (#1 and #2) were generated by long-term (30-day) induction of YB-1 in HTRY cells and have been described previously (see chapter 3). Cells were cultured in Ham’s F-12 media (Gibco, Burlington, Canada) supplemented with 5 μg/ml insulin (Sigma-Aldrich, Oakville, Canada), 1 μg/ml hydrocortisone (SigmaAldrich), 10 mM HEPES (Sigma-Aldrich), and 5% fetal bovine serum (FBS; Gibco). MDA-MB-231 (ATCC, Manassas, VA, USA) cells were cultured in DMEM (Gibco) containing 10% FBS. The X43 primary TNBC cell line (a generous gift from Dr. John Hassell, McMaster University) was cultured in RPMI (Gibco) containing 10% FBS, 100 units/ml penicillin, 100 units/ml streptomycin and 0.5 μg/ml amphotericin B. Drug treatments. Cells were seeded at a density of 5 x 103 in 96-well plates pre-coated with collagen (StemCell Technologies, Vancouver, Canada) prior to treatment with 5fluorouracil (Sigma-Aldrich), anacardic acid (Sigma-Aldrich), BI-D1870 (synthesized by CDRD, Vancouver, Canada), doxorubicin hydrochloride (Sigma-Aldrich), fumitremorgin C (Sigma-Aldrich), gefitinib (AstraZeneca, Mississauga, Canada), luteolin (SigmaAldrich), and paclitaxel (Sigma-Aldrich). Quantitative PCR. RNA was isolated using an RNeasy mini kit (Qiagen, Mississauga, Canada) and reverse transcribed using SuperScript III (Invitrogen, Burlington, Canada). Pre-designed Assays-on-Demand labeled with a FAM reporter (Applied Biosystems, Streetsville, Canada) were used to detect ABCG2 and YB-1. Results were analyzed using the Δ-ΔCt method normalized to the endogenous control TBP. Immunoblotting. Proteins were harvested in egg lysis buffer (ELB) and immunoblotting analysis was performed as previously described (Wu et al., 2006). Antibodies used were as follows: anti-ABCG2 (clone 5D3; Millipore, Billerica, MA, USA), anti-actin (Cell  156  Signaling, Danvers, MA, USA), anti-phospho-S6 ribosomal proteinS235/236 (Cell Signaling), anti-YB-1 (C-term; Epitomics, Burlingame, CA, USA), anti-phospho-YB1S102 (Cell Signaling), and anti-vinculin (Millipore). Chromatin immunoprecipitation. ChIP was performed using an EZ-ChIP kit (Millipore). Cells (2 x 106 per treatment) were cross-linked in 1% formaldehyde for 10 minutes, washed in phosphate-buffered saline (PBS) containing 5 mmol/L sodium butyrate (Sigma-Aldrich) and protease inhibitors, and resuspended in SDS lysis buffer. Samples were sonicated to an average size of 200-600 bp and incubated overnight at 4°C with 1 μg of anti-YB-1 (C-term; Epitomics), 5 μg anti-acetyl-histone H3 (Lys9; Millipore), or isotype matched control IgG antibody. Immunoprecipitated DNA was purified as per the manufacturer’s instructions (Millipore) and analyzed by PCR using the specific primers provided in Table S4.1. Acini morphogenesis assay. Three-dimensional (3D) culture on a reconstituted basement membrane was performed as described previously (Muthuswamy et al., 2001). Assay media (F12 supplemented with 2% donor horse serum, 10 μg/ml insulin, 100 μg/ml hydrocortisone, 5 ng/ml EGF, and 2% Matrigel) was replaced every 4 days. Cells were cultured for 14 days prior to the addition of drug. Immunostaining was carried out as previously described (Lee et al., 2007), except fixation was achieved using 100% methanol at -20°C for 2 hours. Imaging was performed on a DeltaVision personal DV microscope (Applied Precision, Issaquah, WA, USA) and elaborated by DeltaVision softWoRx 5.0.0 deconvolution software. Annexin V apoptosis assay. A single-cell suspension was stained with PE-conjugated annexin V (1:20) on ice for 20 minutes according to manufacturer’s instructions (Annexin V:PE Apoptosis Detection Kit; BD Biosciences, Mississauga, Canada). Annexin V-negative viable cells and annexin V-positive apoptotic cells (20,000 events) were detected by flow cytometry and analyzed using FlowJO (Tree Star Inc., Ashland, OR, USA).  157  Soft-agar anchorage-independent growth assay. Cells were embedded in 0.6% agar at a density of 5 x 103 per well in a 24-well plate as described previously (Finkbeiner et al., 2009). Drug treatments were replenished weekly. Colonies were quantified at 28 days. Mammosphere assay. Cells were washed through a 40 μm filter (BD Biosciences) to obtain a single-cell suspension and seeded at 2 x 104 per well into ultra-low attachment 6well culture plates (Corning, Corning, NY, USA) in freshly supplemented MammoCult Basal media (StemCell Technologies). Spheres with a minimum diameter of 50 μm (or >15 cells) were counted at 7 days. For self-renewal serial passaging experiments, primary mammospheres were collected by centrifugation at 350g for 2 minutes, dissociated with 0.25% trypsin, counted, and re-seeded. FACS analysis. Single cells were suspended in FACS buffer (2% FBS and 5 mM EDTA in PBS) and stained with PE-conjugated CD44 (BD Biosciences) and FITC-conjugated CD49f (BD Biosciences) as previously described (To et al., 2010). The top 10% CD44+/CD49f+ and CD44-/CD49f- populations were sorted out using a FACSCalibur flow cytometer. Statistical analysis. Data are represented as mean ± standard deviation of at least three independent experiments. Statistical significance was evaluated by a paired, 2-tailed Student’s t-test. P values *P < 0.05 and **P < 0.01 were considered significant.  158  4.6 Supplementary Data 4.6.1 Supplementary figures  Fold-change"in"transcript"  40"  **"  35" 30"  ABCG2"  **"  25" 20"  YB-1"  **"  15" 10"  *"  5" 0" HTRZ"  HTRY"  HTRY-LT"  Figure S4.1 ABCG2 was highly expressed in HTRY-LT cells. Quantitative RT-PCR comparing YB-1 and ABCG2 transcript between HTRZ, HTRY (96-hour induction), and HTRY-LT #1 cell lines. Data is presented as mean ± SD (n = 3).  159  Percent"cell"viability"  120"  HTRZ" HTRYCLT"  100" 80" 60" 40"  **"  20" 0" DMSO"  10"μM" Luteolin"  50"μM" Luteolin"  Figure S4.2 HTRY-LT cell viability was reduced by luteolin. Viability of HTRY-LT #1 cells treated with a single dose of luteolin (10 and 50 μM) for 96 hours was assessed using Cellomics ArrayScan VTI. A counter-screen was performed using HTRZ cells. Hoechst 33342-stained nuclei were used as a measure of cellularity. Data is presented as mean ± SD (n = 3) and reported relative to DMSO control.  160  4.6.2 Supplementary tables  Primer set  Oligonucleotide sequencea  Amplicon sizeb  Ta (°C)c  CCAAT Box  118 (nt -329 to -211) 259 (nt -1527 to -1268) 108 (nt -687 to -579) 157 (nt -428 to -271)  59  B  F: 5’-GGCCAGTGACGGCGACCAAA-3’ R: 5’-CTCCCGCCTCCGGGATCGAA-3’ F: 5’-CTCCTCCTGTAGTGCCTTCAGATCTTGCT-3’ R: 5’-TTGCAAATGACCCGAGATCCCACCA-3’ F: 5’-GATGCAGCAGGTAGATGTTGGGA-3’ R: 5’-TGTGCAATATTCCGATGGTGTGGA-3’ F: 5’-CCATTCACCAGAAACCACCCATTT-3’ R: 5’-GCTCATTGGGCTGATCAGTACCT-3’  C  F: 5’-AGGTACTGATCAGCCCAATGAGC-3’ R: 5’-TGAGCCGCCAGCAGGACT-3’  154 (nt -293 to -139)  55  D  F: 5’-AGTCCTGCTGGCGGCTCA-3’ R: 5’-GCCAGAGCTGAACGCAGTGG-3’  166 (nt -146 to +20)  55  Distal A  55 55 55  Table S4.1 Primers used for ChIP assays. a  The sense/forward (F) and antisense/reverse (R) primer sequences  b  The genomic position is the location of the 5’ nucleotide of the primer in relation to the  major transcriptional start site (GenBank accession no. AF151530) c  Optimal annealing temperature (Ta) used in PCR  161  CHAPTER 5. CONCLUDING REMARKS  5.1 Summary and Discussion Despite the apparent complexity of the breast cancer phenotype next-generation sequencing studies indicate that very few changes, perhaps as few as one, in the genome are required to liberate neoplastic cells from the homeostatic mechanisms that govern normal cells (Stephens et al., 2012). It has been well established that in hereditary breast cancer inherited mutations in the BRCA1 and BRCA2 tumour-suppressor genes predispose to the disease. Individuals carrying mutations in one of these genes have a 50 to 80% lifetime risk of developing breast cancer (Foulkes, 2008). Conversely, the driver mutations causally involved in the initiation and progression of sporadic breast cancer have yet to be comprehensively explored. The majority of these tumours arise from epithelial cells and genetic analysis of malignant cells obtained from patients has provided clues as to the gene mutations that may increase susceptibility to breast cancer. Loss-of-function mutations in TP53 occur in over 30% of tumours (Banerji et al., 2012). In addition, breast cancers commonly harbour mutations that function to deregulate the retinoblastoma protein pathway, including loss of RB (Wang et al., 1993) and p16INK4a (Geradts and Wilson, 1996) as well as overexpression of cyclin D1 (Buckley et al., 1993). The Ras-signaling pathway is also a frequent target for alterations, which most notably occur via amplification or overexpression of the ErbB2/HER2 (Slamon et al., 1987) and PIK3CA genes (Banerji et al., 2012). Finally, constitutive or deregulated c-Myc expression is detected in nearly 20% of breast tumours (Berns et al., 1992). Whereas individual mutations have been catalogued in many breast cancers, tumours bear multiple genetic lesions. As such, it is difficult to decipher the genes that are instrumental in the initiation and progression of tumourigenesis from those that are simply inert passengers. Y-box binding protein-1 is an oncogenic transcription and translation factor that functions as a node of integration to coordinate many diverse cellular processes essential for tumourigenesis. Originally associated with proliferation in breast cancer cells due to its interaction with the regulatory elements of EGFR, HER2 (Sakura et al., 1988), and c-  162  Myc (Kolluri and Kinniburgh, 1991), many growth-promoting genes have since been identified as YB-1 targets, including cyclin A, cyclin B1 (Jurchott et al., 2003), and Topo2 (Shibao et al., 1999). YB-1 is also involved in drug resistance through transcriptional regulation of MDR1 (Bargou et al., 1997) and assembly of nucleotide excision repair complexes following DNA damage by cisplatin, for example (Gaudreault et al., 2004). Moreover, YB-1 activates translation of Snail1 and Twist to mediate an EMT leading to enhanced cell invasion and metastatic potential (Evdokimova et al., 2009). Most recently, it has been intimately linked to the tumour-initiating cell phenotype through regulation of CD44 and CD49f (To et al., 2010). The widespread function of YB-1 makes it an ideal candidate for facilitating the initiation, progression, and maintenance of breast cancer. Despite its well-characterized role in established tumours, the functional importance of YB-1 during the initiation and early progression of breast cancer has not been thoroughly investigated. Nevertheless, two lines of evidence suggest that it may function as a driver of the disease. First, targeted expression of YB-1 in the mammary gland of mice resulted in tumour formation with 100% penetrance (Bergmann et al., 2005). In addition, using a tumour tissue microarray our group has shown YB-1 to be overexpressed in 40% of invasive breast tumours where it served as a highly significant independent predictor of poor patient outcome (Habibi et al., 2008). The purpose of this project was to query the mechanism underlying YB-1-mediated cellular transformation with the goal of identifying novel biomarkers and targets for breast cancer. The studies presented in this thesis nominate YB-1 as a bona fide oncogene that functions as a principal driver of breast tumourigenesis. In chapter two, we ectopically expressed YB-1 in normal human mammary epithelial cells which, in agreement with tumours from YB-1 transgenic mice, resulted in cytokinesis failure that acted as a catalyst for genomic instability (Davies et al., 2011). This increased susceptibility for tumour development and, in chapter three, we discovered that YB-1 expression was sufficient to fully transform normal cells. Interrogation of the mechanism underlying tumourigenesis revealed that HMECs were reeducated into cells with stem/progenitor-like properties through plasticity that arose via p300-dependent epigenetic remodeling. The complete transformation of these “primed”, pre-malignant cells was dependent on synergism  163  between YB-1, RSK2, and hTERT. Subtyping the malignant cells exposed that they possessed the molecular signature of a TNBC, which are particularly aggressive cancers that are prone to relapse. Therefore, in chapter four we used our model of YB-1-driven tumourigenesis to uncover that YB-1 transcriptionally regulates the multidrug resistance transporter ABCG2. This mode of drug resistance which was found to be central to instilling broad-spectrum chemoresistance in TNBCs. Repressing YB-1 activation (phosphorylation) using small molecule inhibitors targeting RSK represented a point of molecular vulnerability effective in eliminating tumourigenic cells including the classically resistant CD44-positive TIC population. Figure 5.1 depicts our model of how YB-1 mediates the conversion of HMECs into a TNBC. Taken together, the significance of this work lies in the unparalleled insight it provides into the mechanism by which a single oncogene, YB-1, can dictate the transformation of HMECs specifically into a TNBC. Notably, these studies were the first to associate YB-1 with epigenetic reprogramming as well as ABCG2-mediated multidrug resistance. An important contribution of this research is the powerful discovery platform we have engineered for future testing of biomarkers and therapeutic modalities for TNBC.  164  1$  CCAAT$  p300$  +$YB01$ ABCG2$ YB01$ BMI1$ X, P CD44$ dedifferen<a<on$ 2$ YB01$ CD49f$ CCAAT$  ABCG2$ +$RSK2,$hTERT$ BMI1$ CD44$ clonal$selec<on$ CD49f$ TNBC,  Mature,HMEC, Progenitor)like,cell,  Figure 5.1 A model of YB-1-mediated tumourigenesis. Upregulation of YB-1 in a mature HMEC (as a result of c-Myc amplification, for example) facilitates dedifferentiation into a stem/progenitor state. Reprogramming occurs in two steps. YB-1-mediated activation of p300 alters histone acetylation patterns (step 1). This allows for YB-1 binding and transcriptional activation of the stem/progenitorassociated genes ABCG2, BMI1, CD44, and CD49f, which together instill drug resistance, self-renewal capacity, and multipotency (step 2). These cells subsequently acquire expression of RSK2 and hTERT, due to genomic instability and pressures exerted by YB-1, which confers complete transformation into a TNBC. Red flags, repressive histone modifications; Green flags, acetylated histone lysines.  The normal human mammary epithelial H16N2 cell line provided a suitable background for delineating the function and consequences of YB-1 in transformation. These cells are particularly relevant for studying breast cancer because they are of luminal origin and thus correspond to the lineage where most mammary tumours are believed to arise (Molyneux et al., 2010). Immortalization was achieved using the HPV-16 E6 and E7 viral proteins, which target p53 and RB for degradation, respectively (Band et al., 1990). It is important to note that E6 and E7 are not sufficient to directly transform cells and lowlevels are not associated with aneuploidy or chromosomal rearrangements (Moody and Laimins, 2010). Nevertheless, we verified the genomic integrity of the H16N2 HMECs by performing metaphase spreads and found them to be karyotypically normal (chapter 2). Through collaboration with Dr. Isabelle Berquin we engineered a variant H16N2 cell line that carried the full-length human YB-1 gene under the control of a tetracyclineinducible promoter (Berquin et al., 2005). Expression of YB-1 in the parental HMECs 165  was negligible, while the ectopic level achieved following tetracycline-induction was within the range detected in established TNBC cell lines (chapters 2 and 3). Moreover, the level of YB-1 expression was found to correspond to that measured in primary TNBC patient tissue confirming the physiological relevance of our model (chapter 3). The work in this thesis utilized a Tet-On expression system, that is, the introduction of doxycycline to the system initiated the transcription of YB-1. A particular concern with tetracycline-controlled transcriptional activation is background, or leaky, gene expression in the absence of induction. While we observed minute YB-1 expression in uninduced HTRY cells, it was insufficient to elicit genomic instability and aneuploidy (chapter 2). In addition, previous work from our group has demonstrated that uninduced HTRY cells express negligible levels of the YB-1 transcriptional targets CD44 (To et al., 2010) and p110α (Astanehe et al., 2009). Nevertheless, tighter control of YB-1 expression could be achieved using the Tet-On 3G system which utilizes prokaryotic promoter elements to prevent endogenous transcription factors from binding and causing leaky gene expression. Establishing a Tet-Off system, where doxycycline is the repressing agent, could represent an alternative approach to reducing basal leakiness as it has exceedingly tighter control on transgene expression compared to Tet-On systems (Meyer-Ficca et al., 2004). Herein we have developed a transformation scheme where the introduction of YB-1 was capable of transforming HMECs to a tumourigenic state. We note, however, that other combinations of genetic alterations have previously been reported to convert HMECs into carcinoma cells (Elenbaas et al., 2001; Seger et al., 2002; Watnick et al., 2003; Zhao et al., 2003). The current gold standard dictates that hTERT in combination with SV40 large T antigen (LT), small t antigen (st), and H-rasV12 suffices to transform HMECs (Elenbaas et al., 2001; Ince et al., 2007). We point out that, in contrast to other models, our experimental approach more accurately mimics spontaneously arising human breast cancers in a number of respects. First, it relies on the introduction of a transforming gene, YB-1, at an expression level that closely approximates the situation detected in patient tumour tissue. Moreover, YB-1 is commonly disrupted in over 40% of breast cancers (Habibi et al., 2008), while H-ras, for example, is overexpressed in merely 5% of cases  166  (Clark and Der, 1995). Lastly, our model has been able to phenocopy many aspects of naturally occurring human adenocarcinomas, the most common type of breast tumour, including their invasive behavior and expression of CK18 (Chu and Weiss, 2002). A caveat of our experimental system was that the H16N2 cell line used as a background for YB-1-mediated transformation had pre-existing genetic aberrations due to immortalization with HPV-16. This raised the possibility that a particular cellular background could dictate the transformative potential of YB-1. To address this concern we ectopically expressed YB-1 in the normal mammary epithelial 184hTERT and MCF10A cell lines, immortalized by hTERT or the spontaneous loss of the p16 locus associated with a t(3;9) translocation (Brenner and Aldaz, 1995; Raouf et al., 2005). Irrespective of the cellular background YB-1 was capable of instigating phenotypes associated with malignant progression, including genomic instability (chapter 2), in addition to invasion, dedifferentiation, and expression of TIC markers (chapter 3). Still, we cannot ignore the prospect that immortalization was a requirement for transformation. An important future experiment will be to transduce primary HMECs isolated from reduction mammoplasties (which could be obtained through collaboration with Dr. Connie Eaves, BC Cancer Agency) with lentiviral vectors driving YB-1 expression to query if it is capable of fully transforming these normal cells. It is worth noting that we found YB-1 activates hTERT in H16N2 cells, suggesting that it could have the capacity to immortalize primary cells through this mechanism. One of the most striking phenotypes to emerge following expression of YB-1 in HMECs was genomic instability in the form of both structural and numerical chromosomal aberrations (chapter 2). The same phenomenon was observed in tumours from YB-1 transgenic mice giving credence to our model’s value in recapitulating in vivo tumour progression (Bergmann et al., 2005). However, previous work has failed to identify the molecular basis for the emergence of genomic instability and whether it is a cause or consequence of YB-1-mediated transformation (Bergmann et al., 2005). To address this gap in knowledge, we used a reverse phase protein microarray (RPPA) to uncover the earliest changes in signal transduction following YB-1 expression. Notably, we validated a nearly five-fold increase in the activation of LIM kinase-1 and -2 (pLIMK1/2T508/T505).  167  The level of total protein remained constant implying that this change in activity was not simply the result of enhanced gene transcription. LIMK1 is phosphorylated by CDK1 (Sumi et al., 2002), which is a direct transcriptional target of YB-1 (Lasham et al., 2012). Therefore, we can postulate that YB-1 functions via CDK1 to indirectly activate LIMK1. Ultimately, unrestrained LIMK activity acted as a stimulus for cytokinesis failure likely due to excess actin polymerization that mechanically impeded the formation of the contractile ring (Yang et al., 2004b). Under normal conditions, cytokinesis failure results in cell cycle arrest and eventual apoptosis (Andreassen et al., 2001). To overcome these obstacles and allow for the expansion of genomically compromised cells we discovered that YB-1 facilitated the elimination of important cell cycle checkpoint controls by promoting overexpression of cyclin E (chapter 2) and repression of p16INK4a (chapter 3). Failure to properly regulate cyclin E-associated kinase activity and p16INK4a has been associated with accelerated Sphase entry and genomic instability (McDermott et al., 2006; Spruck et al., 1999). The latter is largely the result of uncontrolled centrosome duplication leading to a multipolar mitosis and karyotypic instability. Most recently, it was discovered that cyclin E is required for loading MCM complexes onto chromatin, which allows cells to escape from quiescence and re-enter the cell cycle (Geng et al., 2007). The work presented in this thesis does not address the mechanism responsible for the increased abundance of cyclin E following YB-1 induction. We can speculate that the regulation is indirect as the cyclin E promoter is devoid of YREs. However, the gene is the target of the E2F1 transcription factor (Ohtani et al., 1995), which is regulated by YB-1 (Lasham et al., 2012), thereby indirectly placing cyclin E transcription under the control of YB-1. Overexpression could also be attributed to YB-1 antagonizing TGFβ and thus counteracting its ability to target CDK2/cyclin E complexes for degradation (Nagahara et al., 1999). It is interesting to note that cyclin E is highly expressed in TNBCs suggesting that it could be important for increasing susceptibility to this subtype of cancer not only in our model but also in patient tumours (Bostrom et al., 2009). The loss of p16INK4a in YB-1-expressing HMECs was particularly intriguing. Breast tissue from healthy, disease-free women has been shown to contain a population of cells that  168  have silenced p16INK4a due to hypermethylation of promoter sequences (Holst et al., 2003). Moreover, our group has found YB-1 to be expressed in a subset of normal mammary cells in patients who develop cancer (chapter 3). It is tempting to speculate that YB-1 overexpression coincides with focal areas of p16INK4a hypermethylation in vivo, creating ideal candidates as precursors to breast cancer. In support of this hypothesis, we discovered that YB-1 transcriptionally regulates BMI1, a Polycomb-group protein that acts as a potent negative regulator of the INK4a locus. BMI1 functions in the epigenetic modulation of gene expression through both histone H2A ubiquitylation and promoter methylation (Kallin et al., 2009). While our data demonstrate an increase in the pool of ubiquitylated histone H2A, the methylation status of the CpG island in the p16INK4a promoter remains to be measured by pyrosequencing. Regardless of the precise mechanism of silencing the endpoint was loss of p16INK4a and, accordingly, reduced fidelity of the G1-S cell cycle checkpoint. The removal of p16INK4a has also been implicated in the upregulation and recruitment of the Polycomb repressor EZH2 to HOXA9, a locus expressed in normal breast development but epigenetically silenced in breast cancer (Reynolds et al., 2006). This rationalizes our finding that YB-1 is highly correlated with BMI1 (r = 0.245, p = 0.031) and EZH2 (r = 0.518, p < 0.001) in TNBC patient samples (S. Islam and S.E. Dunn, unpublished). Thus it is conceivable that YB-1, acting via BMI1-mediated silencing of p16INK4a activity, instills a non-random pattern of Polycomb-mediated gene silencing that reprograms transcription to skew cells toward a TNBC fate. A comprehensive, genome-wide profile of the epigenetic changes controlled by YB-1 could be investigated using Illumina methylation arrays. In addition to the deregulation of cell-cycle regulatory factors, we also identified a propensity for the repression of pro-apoptotic proteins following YB-1 induction. Notably, caspase-7, caspase-12, and TNF-related apoptosis-inducing ligand (TRAIL) were down-regulated in our RPPA analysis. These findings remain to be validated by immunoblotting. However, we can speculate that YB-1 may act as a transcriptional repressor at the promoters of pro-apoptotic genes. This is supported by the observation that YB-1 binds the caspase-7 promoter by ChIP (Finkbeiner et al., 2009) and its inhibition yields increased caspase-7 gene expression (Lasham et al., 2012). Taken  169  together, our results provide insight into how YB-1-induced HMECs could evade apoptosis and continue to proliferate despite widespread genomic instability. Unexpectedly, we found that over 30% of YB-1-expressing HMECs carried a low-level HER2 amplification (chapter 2). This discovery is remarkable because it suggests that the chromosomal instability that arose due to YB-1 was not stochastic, but rather targeted toward increasing susceptibility to cancer development. Notably, overexpression of HER2, but not c-Myc, Akt1, or cyclin D1, elicited clonal outgrowths in acini cultures, a defining feature of neoplastic progression in early-stage breast tumours (Leung and Brugge, 2012). Together with our findings, this suggests that enhanced MAP kinase signaling via the HER2/EGFR-RSK-YB1 axis may be an essential prerequisite for malignant progression. It will be of particular interest to further interrogate the genome of these cells using spectral karyotyping and DNA sequencing to identify if a propensity exists for additional rearrangements. For example, given that RSK2 is necessary for complete transformation it is possible that genomic instability could contribute to its expression. This is supported by the notion that the RSK2 locus (Xp22.2) is a frequent target for mutations in breast cancer (Wood et al., 2007). Considerable controversy exists with respect to the importance of genomic instability for driving tumour growth and whether or not it is an initiating event in tumourigenesis (Negrini et al., 2010; Sieber et al., 2003). Our data convey that genomic instability is an enabling phenotype that underlies tumour progression by enhancing cellular fitness, but it is not synonymous with transformation. For example, while upregulation of YB-1 led to luminal cell translocation and outgrowth in organotypic acinar structures these cells were unable to form colonies in soft agar, a hallmark of full malignant transformation. This was despite overt genomic instability (chapter 3). In agreement, HMECs transduced with H-rasV12 also display genomic instability yet they are unable to form tumours in vivo (Dumont et al., 2009). This suggests that specific genetic lesions must arise through natural selection to complete the transformation process. Sustained culturing of HMECs in the presence of YB-1 resulted in the emergence of cells with upregulated RSK2 and hTERT expression. These cells had the capacity to grow  170  under anchorage-independent conditions and form small tumours when injected into the mammary gland of NOD/SCID mice (chapter 3). We postulate that tumour growth was restrained by the mouse immune system and/or foreign mammary microenvironment. Accordingly, the severely immunocompromised NOD/SCID IL2R gamma null (NSG) mouse strain may support greater engraftment. In addition, humanizing the mouse mammary fat pad by transplanting immortalized human fibroblasts prior to injection of transformed mammary epithelial cells may better recapitulate the tumour-stroma interactions that spur tumour growth. As the control HMECs were non-tumourigenic in vivo, the malignant cells most likely arose due to pressures exerted by YB-1 rather than through the physiologic pressures of in vitro passaging. One important line of future investigation will be to assess the heterogeneity that exists within the YB-1-induced HMEC population to delineate how these cells clonally evolve. To pinpoint the earliest molecular alterations imparted by YB-1 that are causative in tumour progression we capitalized on the fact that three-dimensional acini cultures allow for long-term monitoring at single-cell resolution. Notably, we discovered that a cell expressing CD44, a marker of the TIC phenotype, escaped from the epithelial layer to drive the luminal outgrowths that are characteristic of a DCIS (chapter 3). YB-1 is a known transcriptional regulator of CD44 (To et al., 2010), which can subsequently assemble a signaling complex with ErbB2/HER2 to provoke enhanced migration and resistance to apoptosis (Ghatak et al., 2005). CD44 also activates the invasion proteases cathepsin K, MT1-MMP, and uPA, which could perturb cell-matrix interactions to further aid in cell translocation to the lumen (Montgomery et al., 2012). It is not difficult to envision that the intimate co-operation between YB-1 and CD44 may define the earliest stages of malignancy and facilitate tumour progression. Moreover, it is intriguing that the CD44-positive phenotype is enriched in basal-like/TNBCs (Honeth et al., 2008) suggesting that the commitment of YB-1-expressing cells to this subtype may transpire early in tumourigenesis. The realization that YB-1-induced HMECs were reprogrammed into cells with a TIC phenotype ignited the possibility that YB-1 acts as a mechanistic bridge between epigenetic regulation and oncogenic transcription. Bidirectional interconversion between  171  committed progenitors and multipotent stem cells can be ascribed to both normal and tumourigenic breast epithelial cells (Chaffer et al., 2011; Gupta et al., 2011). However, the factors that regulate this plasticity have remained elusive. We present the striking finding that in immortalized HMECs the gain of YB-1 caused a sequential process, p300mediated histone acetylation followed by transcriptional activation, in genes critical for differentiation and cell fate determination, BMI1, CD44, and CD49f (chapter 3). It is important to note that in this thesis YB-1 promoter occupancy was equated with transcriptional regulation. This was detected by ChIP and could be further elaborated using electrophoretic mobility shift assays. To confirm whether YB-1 binding is functional, luciferase reporter assays could be employed to provide an in vitro measure of transcriptional activity at the ABCG2 and BMI1 promoters. Subsequent mutation of the YB-1 binding site(s) could verify that YB-1 binding serves to activate transcription. Our results suggest a model whereby expression of YB-1 stimulates p300 activity, which in turn relaxes promoter-centered chromatin ultimately allowing for YB-1 binding and transcriptional activation at loci involved in cell-fate processes. Specifically, CD44 and CD49f prevent terminal differentiation (Pham et al., 2011; Yu et al., 2012), while BMI1 enhances self-renewal capacity (Liu et al., 2006). Enrichment in p300-mediated acetylation is a common feature of embryonic stem cells that diminishes when they are forced to differentiate (Krejci et al., 2009). Accordingly, by upregulating p300, YB-1 may be reactivating a dormant stem cell program by altering histone acetylation patterns. This notion is supported by our finding that YB-1-induced HMECs acquired the capacity for self-renewal and bipotential differentiation. Together, these data convey YB-1 as a central regulator in the dedifferentiation toward a more progenitor-like phenotype. The fact that inhibition of p300 activity blocked transcription at the BMI1, CD44, and CD49f gene loci places histone acetylation upstream of YB-1 promoter binding. Thus, for the first time, we have demonstrated that YB-1 controls the progenitor-differentiation switch in HMECs in a process that is strictly dependent on epigenetic remodeling by p300. Our model provides an elegant mechanistic rationale for the observation that tumour cells share many of the same characteristics as stem cells (Ben-Porath et al., 2008; Wong et al., 2008). This is especially evident in the TNBC subtype, which is enriched in  172  undifferentiated cells that exhibit progenitor-like properties (Honeth et al., 2008). Based on our finding that YB-1 epigenetically regulates “stemness” it is likely not a coincidence that the oncogene is expressed in over 70% of TNBCs (Habibi et al., 2008). In order to address if YB-1 can reeducate mature HMECs into TNBCs we looked at the expression of differentiation markers at various stages of tumour development (chapter 3). Normal HMECs expressed the expected markers of differentiation (such as ER and PR), but did not express stem/progenitor cell markers (such as BMI1, CD44, and CD49f). However, following YB-1 induction the expression of differentiation markers decreased concomitantly with an increase in the expression of stem/progenitor markers, suggesting that the cells undergo dedifferentiation. The fully transformed cells were classified as TNBC using a clinical panel of breast cancer subtyping markers. Notably, YB-1 expression has been detected in normal luminal progenitors (Finkbeiner et al., 2009), the cell population from which TNBCs arise (Visvader, 2009), suggesting that it may function in maintaining the undifferentiated, stem-like state of these cells. This interpretation is supported by a recent report linking the loss of YB-1 to neural stem cell differentiation (Fotovati et al., 2011). It will be of particular interest to dissect if YB-1 expression yields expansion of the luminal progenitor population, based on gain of CD61 and loss of Gata-3 marker expression (Asselin-Labat et al., 2007), which could act as an unrestricted pool for the development of TNBC. In agreement with other studies (Lim et al., 2009; Molyneux et al., 2010; Shipitsin et al., 2007; Visvader, 2009), our findings suggest that breast cancers probably arise from various types of undifferentiated progenitor or stem cells. However, this process is likely more complex then initially perceived as HMECs exhibit plasticity (Chaffer et al., 2011; Gupta et al., 2011) and thus may have an underappreciated role in breast cancer development. We propose that these mature cells may be the recipient of the initial mutation and serve as the cell of origin. Specifically, we show that YB-1 expression promoted the epigenetic dedifferentiation of HMECs to generate a stem/progenitor state that initiated tumour progression. In support of the cancer stem cell hypothesis, over time these cells acquired genetic lesions, for example upregulation of RSK2 and hTERT, to give rise to TNBCs.  173  The prominent role of YB-1 in the genesis of TNBC prompted us to address its utility in maintaining the progression of the disease. Known as the “triple-negative paradox” these cancers are extremely responsive to neoadjuvant chemotherapy, yet they readily relapse within five years of initial treatment (Dent et al., 2007). To better understand the basis of this phenomenon we identified, for the first time, that YB-1 transcriptionally regulates the multidrug resistance transporter ABCG2. This was found to be contingent upon p300mediated promoter-centered histone acetylation (chapter 4). ABCG2 has been strongly associated with stem/progenitor cell populations providing further credence to the dogma that YB-1 instills a progenitor-like state by altering histone acetylation patterns (Zhou et al., 2001). We discovered that upregulation of ABCG2 rendered TNBC cells refractory to taxane and anthracycline chemotherapy. Notably, this mechanism of multidrug resistance was found to be particularly active in the CD44+/CD49f+ TIC population, which is postulated to serve as a reservoir for regenerating the drug-resistant tumour (Creighton et al., 2009; Fillmore and Kuperwasser, 2008; Marangoni et al., 2009). While we identified ABCG2 as the principal determinant of drug resistance, it has also been shown that YB-1 regulates the MDR1 drug transporter (Bargou et al., 1997). However, MDR1 has little prognostic value in predicting relapse (Larkin et al., 2004) and inhibitors have displayed virtually no benefit to relapse-free survival in clinical trials (Szakacs et al., 2006). Together with our data, this suggests that the contribution of ABCG2 may be more important to clinical drug resistance. The model we developed of TNBC is an ideal platform for testing therapeutics because the parental non-transformed HMECs are a true comparator. It has been previously shown that inhibition of RSK, the predominate upstream kinase responsible for activating YB-1 via Ser-102 phosphorylation (Stratford et al., 2008), induces apoptosis in breast cancer cell lines (Stratford et al., 2012). Our current research furthered this line of investigation by evaluating the efficacy of small molecule RSK inhibitors (chapter 4). These included BI-D1870 and luteolin, an off patent bioflavonoid identified by our group as a RSK inhibitor through molecular docking and in vitro kinase assays. Incredibly, we found these compounds to be highly selective toward TNBC cell lines, with negligible toxicity in the parental HMECs. As the TIC population is specifically believed to be  174  responsible for TNBC recurrence post-chemotherapy, we assessed the ability of RSK inhibitors to eliminate these highly tumourigenic cells. Of note, in a departure from the arbitrary CD44+/CD24- marker expression commonly used to isolate TICs, we defined these cells as CD44+/CD49f+ based on the recent knowledge that these markers share commonality with normal human mammary progenitor/stem cells (Meyer et al., 2010). In stark contrast to traditional chemotherapeutics, RSK inhibition was effective in eliminating the TIC population as well as the bulk non-TIC tumour cells. This is particularly important as we have demonstrated that any YB-1-expressing cell not eradicated has the plasticity to dedifferentiate and spawn cells with tumour initiating capacity de novo. Although our results are promising, the future challenge will be to validate the efficacy of RSK inhibitors in vivo and in carefully controlled clinical trials. If confirmed, the transition of luteolin, particularly, into a therapeutic for TNBC should be straightforward and feasible. The experimental model of TNBC developed in this thesis was contingent upon ectopic YB-1 expression in HMECs, which begs the question as to how it becomes activated during tumourigenesis. Notably, c-Myc is a known transcriptional activator of YB-1 (Uramoto et al., 2002) that is often amplified in TNBC (Turner et al., 2004). In a feedforward loop YB-1 could subsequently increase c-Myc translation (Bommert et al., 2012) to further enhance its own expression. Bioinformatics analysis of the YB-1 promoter also identified putative binding sites for the transcription factor Snail (Wu et al., 2007). This is intriguing because it could directly link YB-1 to the EMT, which has been closely associated with the gain of stem/progenitor cell properties (Mani et al., 2008). In addition to transcriptional regulation, the YB-1 locus (1p34.1) may be the target for mutations or amplification. This area flanks a region (1p34.2 - 1p34.4) commonly amplified in colon, lung, and breast cancers (Henderson et al., 2005). While amplification was not detected in primary TNBCs in one analysis, this was based on a small cohort and used lowresolution CGH arrays (Stratford et al., 2007). Next-generation sequencing will be necessary to detect small insertions and deletions as well as translocations at the YB-1 locus. It is tempting to speculate that mutations resulting in the constitutive activation of YB-1 could be a common feature among TNBCs.  175  5.2 Future Directions and Clinical Implications The present findings described in this thesis were based on in vitro transformation and, thus, it will be important to transition this research into mouse models. While we have convincingly demonstrated that YB-1 can fully transform HMECs into malignant TNBC cells capable of forming small tumours when orthotopically injected into the mammary fat pad of NOD/SCID mice (chapter 3), cell culture systems may not reflect the physiological fate of cells in their native environment. Accordingly, it will be necessary to employ a transgenic approach to fully assess the requirement for YB-1 in facilitating the development of TNBC within the natural microenvironment. Our collaborator Dr. Anna Mandinova (at Harvard Medical School/Broad Institute) has recently developed a Cre/loxP transgenic model that will allow us to ablate YB-1 expression in the mammary epithelium. We propose to cross mice bearing the floxed YB-1 gene with MMTV-Cre mice to tissue-specifically knockdown YB-1 in the mammary gland. Subsequently, we will cross the MMTV-Cre YB-1fl/fl mice with C3(1)/SV-40 T-antigen (Tag) mice. The C3(1)/Tag mice are particularly suited for this study. They all develop invasive mammary carcinomas with a short latency (~16 weeks) that most closely approximate human TNBC based on gene expression profiling and other important biological features, such as metastasis to the lung (Green et al., 2000; Herschkowitz et al., 2007). Thus, if YB-1 is principal to the genesis of TNBC, repressing its expression should decrease the frequency of tumours in the C3(1)/Tag transgenic mice. We could further dissect the cell of origin using lineage-specific cytokeratin promoter-driven Cre recombinase (K18-Cre and K14Cre) to specifically knockdown YB-1 in luminal and myoepithelial cells. We established that YB-1 reprograms mature HMECs into progenitor-like cells that serve as precursors to TNBC in vitro. Most likely these cells resemble luminal epithelial progenitors as TNBCs have been shown to originate from this lineage (Molyneux et al., 2010; Visvader, 2009). A recently developed inducible genetic lineage tracing strategy will allow us to study the dynamics of the luminal cell population in vivo (Van Keymeulen et al., 2011). In this elegant system, the administration of tamoxifen yields the activation of a yellow fluorescent protein (YFP) reporter by the K8 promoter in luminal progenitor cells. Accordingly, we can cross the K8-CreER/Rosa-YFP mice with  176  BLG-YB-1 mice that target YB-1 to the mammary epithelium. If our hypothesis is correct and YB-1 provokes an expansion of luminal-restricted progenitors we will witness an increase in YFP-positive cells following transgenic YB-1 expression. Our findings hold implications for the development of anticancer therapeutics. As YB-1 can force the dedifferentiation of HMECs into progenitor-like TICs, then targeting the TIC populations will, on its own, be unlikely to yield durable clinical results. This is because the eradication of existing TIC populations could potentially be followed by their regeneration from non-TICs within the tumour leading to recurrence. In support of this interpretation, our group has previously reported that YB-1 activates the transcription of TIC-associated genes following chemotherapy exposure (To et al., 2010). Accordingly, because it bridges the gap between the TIC and non-TIC populations, targeting YB-1 could represent a promising therapeutic strategy. This approach is particularly attractive because YB-1 is not expressed in normal adult tissues but is upregulated in tumours (Bargou et al., 1997), which are absolutely dependent on its function for growth and survival (chapters 3 and 4). However, translating these findings into the clinic has been hampered by the lack of YB-1 small molecule inhibitors. Nevertheless, a promising therapeutic approach could involve in vivo delivery of siRNAs formulated into lipid nanoparticles. Successful pre-clinical trials of this technology in nonhuman primates (Semple et al., 2010) coupled with advances in the long-term stability of siRNA in serum (over 45 days with Stealth-modified YB-1 siRNA (Lasham et al., 2012)) could make targeted delivery of YB-1-siRNA a reality in the near future. Alternatively, suicide gene therapy using the YB-1 promoter to drive expression of a toxic gene, such as caspase-3 or the dominant negative YB-1 S102A, could represent another approach for selectively killing cancer cells. In addition to directly targeting YB-1, repressing its activation (Ser-102 phosphorylation) using small-molecule RSK inhibitors has shown immense promise in eradicating malignant cells in vitro (Astanehe et al., 2012; Dhillon et al., 2010; Stratford et al., 2012). Complementing these studies, we have shown that RSK inhibitors demonstrate efficacy against both the TIC and non-TIC populations within a TNBC (chapter 4). Furthermore, due to the necessity of activated pYB-1S102 for maintaining the progenitor-like state  177  (chapter 3), these inhibitors block the perpetual cycle of dedifferentiation that could replenish the TIC population. Another important benefit of targeting RSK is the apparent safety window. RSK inhibitors are not overtly toxic to normal HMECs (chapter 4) and hematopoietic stem cells (Stratford et al., 2012) at doses that effectively eliminate malignant cells. Moreover, RSK2 knockout mice are viable and, apart from being approximately 10% smaller, are near exact phenocopies of their wild-type littermates (Dufresne et al., 2001). In light of these observations, we believe RSK inhibitors are ideally suited to advance into pre-clinical trials. In collaboration with Dr. Emma Guns (at the Vancouver Prostate Centre), we propose to evaluate the pharmacokinetics of these compounds and transition them into mouse models. However, a disadvantage of the current generation inhibitors is that they are not RSK isoform specific (Sapkota et al., 2007). While RSK3 and RSK4 have been described as tumour suppressor genes (Bignone et al., 2007; Thakur et al., 2008), our work dictates that RSK2 is an absolute requirement for HMEC transformation (chapter 3). Together with the fact that it is most strongly associated with TNBC (Brough et al., 2011) and its inhibition triggers apoptosis in the TIC population (Stratford et al., 2012), provides a strong rationale for selectively targeting RSK2. The recent solving of its crystal structure (Malakhova et al., 2009) should expedite the discovery of RSK2 specific inhibitors. Recently, there has been much enthusiasm about the prospect of developing epigenetic therapies for the treatment of breast cancer. However, in consideration of our findings, great caution should be exercised when contemplating this therapeutic strategy because it could tip the balance in favour of an undifferentiated stem/progenitor state. For example, the histone deacetylase inhibitor panobinostat, which increases global histone acetylation, has been associated with growth arrest and apoptosis in breast cancer cell lines (Tate et al., 2012). At first these results appear to be promising, but further interrogation revealed that the drug promotes dedifferentiation based on the loss of ER, PR, and Gata-3 (Tate et al., 2012). This study, together with our data, presents a compelling case against the clinical use of HDAC inhibitors. We believe a more rationale approach will involve targeting specific epigenetic machinery to improve therapeutic efficacy and reduce toxicity. Although we have not formally shown this in our model, it has been reported  178  that the Polycomb protein EZH2 cooperates with BMI1 to regulate the self-renewal ability and tumour-initiation capacity of the TIC population (Suva et al., 2009). Notably, small molecule inhibitors developed against EZH2 have exhibited an extremely powerful anti-lymphoma effect in human cell line xenografts (Knutson et al., 2012; McCabe et al., 2012). Therefore, these compounds could represent a promising targeted epigenetic therapy for breast cancer that, in addition to eradicating the bulk tumour, might also promote the extinction of the EZH2/BMI1-dependent TIC population. The studies described in this thesis have shown the importance of YB-1 in the initiation and progression of sporadic breast cancer. Curiously, high expression of YB-1 has also been documented to be a common feature of BRCA1-mutated hereditary breast cancer (Sorlie et al., 2003). In striking similarity to the phenotypes observed in YB-1 transformed HMECs, BRCA1-mutated tumours exhibit genomic instability and lack expression of ER, PR, and HER2 (Foulkes, 2007). Moreover, they harbour an expansion of the luminal progenitor population, which is the probable target for basal-like/TNBC tumours (Lim et al., 2009). Based on our findings, this is most likely attributed to dedifferentiation mediated by YB-1. Taken together, we propose that the upregulation of YB-1, rather than loss of BRCA1, is the driving force that predisposes to hereditary breast cancer. This has important clinical implications because it suggests that YB-1 inhibitors could be effective in preventing the onset of cancer in BRCA1-mutation carriers. This possibility would be important to test, and could easily be tested first in vitro using BRCA1-deficient cell lines, and later by crossing BRCA1-null mice with the Cre/loxP YB-1 transgenic mice described above. Although our studies have focused on breast cancer, the implications of this work may span numerous malignancies. YB-1 is also highly expressed and correlates with poor patient prognosis in carcinomas of the colon, lung, ovary, and prostate, in addition to melanoma, large B-cell lymphoma, and multiple myeloma (Eliseeva et al., 2011). Notably, YB-1 has been associated with glioblastoma multiforme, where it functions to promote the proliferation and inhibit differentiation of brain tumour-initiating cells (Fotovati et al., 2011). This implies that the function of YB-1 in controlling cell fate is not specific to breast cancer. It will be of interest to ascertain if p300-mediated histone  179  acetylation patterning is a conserved mechanism that underlies cellular dedifferentiation and tumourigenicity. Finally, because YB-1 has been implicated in many different malignancies its validity as a therapeutic target likely extends beyond breast cancer. The present findings outlined in this thesis hold the implication that patient- and tissuespecific stem cells may one day be created in vitro via forced YB-1 expression in a patient’s own terminally differentiated epithelial cells. The ability to reprogram adult cells into stem cells could be important for regenerative therapies. Currently, transduction of differentiated somatic cells with Oct3/4, Sox2, Klf4, and c-Myc can generate induced pluripotent stem (iPS) cells that are similar to embryonic stem cells in terms of morphology, proliferation, and gene expression (Takahashi et al., 2007; Takahashi and Yamanaka, 2006). The efficacy of this process, however, is low and mice derived from iPS cells often develop tumours or die in utero (Okita et al., 2007). We discovered YB-1 to be a reliable and efficient factor for transforming mature HMECs into stem/progenitor cells. It is tempting to speculate that by fine-tuning the level and timing of YB-1 expression we could achieve cellular reprogramming without the subsequent cancer. In summary, YB-1 expression in human mammary epithelial cells acts as a catalyst to facilitate the development of TNBC. Through recruitment of the HAT protein p300, YB1 modifies chromatin structure so that it can bind and upregulate “stemness” genes. This promotes dedifferentiation and expands the pool of stem/progenitor-like cells that are targets for transformation. Because of the core function of YB-1 in bridging the stem and non-stem populations, targeting it represents a viable therapeutic strategy to eradicate all tumourigenic cells and likely prevent relapse. Together, our work emphasizes the importance of YB-1 in cellular reprogramming and transformation and has implications not only for cancer biology but also stem cell biology.  180  REFERENCES  Acevedo, K., Li, R., Soo, P., Suryadinata, R., Sarcevic, B., Valova, V. A., Graham, M. E., Robinson, P. J., and Bernard, O. (2007). The phosphorylation of p25/TPPP by LIM kinase 1 inhibits its ability to assemble microtubules. Experimental cell research 313, 4091-4106. Agoston, A. T., Argani, P., Yegnasubramanian, S., De Marzo, A. M., Ansari-Lari, M. A., Hicks, J. L., Davidson, N. E., and Nelson, W. G. (2005). Increased protein stability causes DNA methyltransferase 1 dysregulation in breast cancer. The Journal of biological chemistry 280, 18302-18310. Al-Hajj, M., Wicha, M. S., Benito-Hernandez, A., Morrison, S. J., and Clarke, M. F. (2003). Prospective identification of tumorigenic breast cancer cells. Proceedings of the National Academy of Sciences of the United States of America 100, 3983-3988. Ali, S., and Coombes, R. C. (2002). Endocrine-responsive breast cancer and strategies for combating resistance. Nature reviews Cancer 2, 101-112. Alliegro, M. C., Alliegro, M. A., and Palazzo, R. E. (2006). Centrosome-associated RNA in surf clam oocytes. Proceedings of the National Academy of Sciences of the United States of America 103, 9034-9038. Allred, D. C., Mohsin, S. K., and Fuqua, S. A. (2001). Histological and biological evolution of human premalignant breast disease. Endocrine-related cancer 8, 47-61. Allred, D. C., Wu, Y., Mao, S., Nagtegaal, I. D., Lee, S., Perou, C. M., Mohsin, S. K., O'Connell, P., Tsimelzon, A., and Medina, D. (2008). Ductal carcinoma in situ and the emergence of diversity during breast cancer evolution. Clinical cancer research : an official journal of the American Association for Cancer Research 14, 370-378. Amano, T., Kaji, N., Ohashi, K., and Mizuno, K. (2002). Mitosis-specific activation of LIM motif-containing protein kinase and roles of cofilin phosphorylation and dephosphorylation in mitosis. The Journal of biological chemistry 277, 22093-22102. An, Y., and Ongkeko, W. M. (2009). ABCG2: the key to chemoresistance in cancer stem cells? Expert Opin Drug Metab Toxicol 5, 1529-1542. Andersen, J. S., Wilkinson, C. J., Mayor, T., Mortensen, P., Nigg, E. A., and Mann, M. (2003). Proteomic characterization of the human centrosome by protein correlation profiling. Nature 426, 570-574. Andreassen, P. R., Lohez, O. D., Lacroix, F. B., and Margolis, R. L. (2001). Tetraploid state induces p53-dependent arrest of nontransformed mammalian cells in G1. Molecular biology of the cell 12, 1315-1328.  181  Andrechek, E. R., Hardy, W. R., Siegel, P. M., Rudnicki, M. A., Cardiff, R. D., and Muller, W. J. (2000). Amplification of the neu/erbB-2 oncogene in a mouse model of mammary tumorigenesis. Proceedings of the National Academy of Sciences of the United States of America 97, 3444-3449. Arany, Z., Huang, L. E., Eckner, R., Bhattacharya, S., Jiang, C., Goldberg, M. A., Bunn, H. F., and Livingston, D. M. (1996). An essential role for p300/CBP in the cellular response to hypoxia. Proceedings of the National Academy of Sciences of the United States of America 93, 12969-12973. Asselin-Labat, M. L., Sutherland, K. D., Barker, H., Thomas, R., Shackleton, M., Forrest, N. C., Hartley, L., Robb, L., Grosveld, F. G., van der Wees, J., et al. (2007). Gata-3 is an essential regulator of mammary-gland morphogenesis and luminal-cell differentiation. Nature cell biology 9, 201-209. Astanehe, A., Finkbeiner, M. R., Hojabrpour, P., To, K., Fotovati, A., Shadeo, A., Stratford, A. L., Lam, W. L., Berquin, I. M., Duronio, V., and Dunn, S. E. (2009). The transcriptional induction of PIK3CA in tumor cells is dependent on the oncoprotein Ybox binding protein-1. Oncogene 28, 2406-2418. Astanehe, A., Finkbeiner, M. R., Krzywinski, M., Fotovati, A., Dhillon, J., Berquin, I. M., Mills, G. B., Marra, M. A., and Dunn, S. E. (2012). MKNK1 is a YB-1 target gene responsible for imparting trastuzumab resistance and can be blocked by RSK inhibition. Oncogene 31, 4434-4446. Azuara, V., Perry, P., Sauer, S., Spivakov, M., Jorgensen, H. F., John, R. M., Gouti, M., Casanova, M., Warnes, G., Merkenschlager, M., and Fisher, A. G. (2006). Chromatin signatures of pluripotent cell lines. Nature cell biology 8, 532-538. Bader, A. G., and Vogt, P. K. (2005). Inhibition of protein synthesis by Y box-binding protein 1 blocks oncogenic cell transformation. Molecular and cellular biology 25, 20952106. Bagheri-Yarmand, R., Mazumdar, A., Sahin, A. A., and Kumar, R. (2006). LIM kinase 1 increases tumor metastasis of human breast cancer cells via regulation of the urokinasetype plasminogen activator system. International journal of cancer Journal international du cancer 118, 2703-2710. Balasubramanyam, K., Swaminathan, V., Ranganathan, A., and Kundu, T. K. (2003). Small molecule modulators of histone acetyltransferase p300. The Journal of biological chemistry 278, 19134-19140. Band, V., Zajchowski, D., Swisshelm, K., Trask, D., Kulesa, V., Cohen, C., Connolly, J., and Sager, R. (1990). Tumor progression in four mammary epithelial cell lines derived from the same patient. Cancer research 50, 7351-7357. Banerji, S., Cibulskis, K., Rangel-Escareno, C., Brown, K. K., Carter, S. L., Frederick, A. M., Lawrence, M. S., Sivachenko, A. Y., Sougnez, C., Zou, L., et al. (2012). Sequence 182  analysis of mutations and translocations across breast cancer subtypes. Nature 486, 405409. Bange, J., Zwick, E., and Ullrich, A. (2001). Molecular targets for breast cancer therapy and prevention. Nature medicine 7, 548-552. Barber, T. D., McManus, K., Yuen, K. W., Reis, M., Parmigiani, G., Shen, D., Barrett, I., Nouhi, Y., Spencer, F., Markowitz, S., et al. (2008). Chromatid cohesion defects may underlie chromosome instability in human colorectal cancers. Proceedings of the National Academy of Sciences of the United States of America 105, 3443-3448. Bargou, R. C., Jurchott, K., Wagener, C., Bergmann, S., Metzner, S., Bommert, K., Mapara, M. Y., Winzer, K. J., Dietel, M., Dorken, B., and Royer, H. D. (1997). Nuclear localization and increased levels of transcription factor YB-1 in primary human breast cancers are associated with intrinsic MDR1 gene expression. Nature medicine 3, 447450. Barretina, J., Caponigro, G., Stransky, N., Venkatesan, K., Margolin, A. A., Kim, S., Wilson, C. J., Lehar, J., Kryukov, G. V., Sonkin, D., et al. (2012). The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature 483, 603-607. Barros, T. P., Kinoshita, K., Hyman, A. A., and Raff, J. W. (2005). Aurora A activates DTACC-Msps complexes exclusively at centrosomes to stabilize centrosomal microtubules. The Journal of cell biology 170, 1039-1046. Bartek, J., Lukas, C., and Lukas, J. (2004). Checking on DNA damage in S phase. Nature reviews Molecular cell biology 5, 792-804. Basaki, Y., Taguchi, K., Izumi, H., Murakami, Y., Kubo, T., Hosoi, F., Watari, K., Nakano, K., Kawaguchi, H., Ohno, S., et al. (2010). Y-box binding protein-1 (YB-1) promotes cell cycle progression through CDC6-dependent pathway in human cancer cells. European journal of cancer 46, 954-965. Basto, R., Brunk, K., Vinadogrova, T., Peel, N., Franz, A., Khodjakov, A., and Raff, J. W. (2008). Centrosome amplification can initiate tumorigenesis in flies. Cell 133, 10321042. Baylin, S. B., and Jones, P. A. (2011). A decade of exploring the cancer epigenome biological and translational implications. Nature reviews Cancer 11, 726-734. Ben-Porath, I., Thomson, M. W., Carey, V. J., Ge, R., Bell, G. W., Regev, A., and Weinberg, R. A. (2008). An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nature genetics 40, 499-507. Berdasco, M., and Esteller, M. (2010). Aberrant epigenetic landscape in cancer: how cellular identity goes awry. Developmental cell 19, 698-711.  183  Bergmann, S., Royer-Pokora, B., Fietze, E., Jurchott, K., Hildebrandt, B., Trost, D., Leenders, F., Claude, J. C., Theuring, F., Bargou, R., et al. (2005). YB-1 provokes breast cancer through the induction of chromosomal instability that emerges from mitotic failure and centrosome amplification. Cancer research 65, 4078-4087. Bernard, O. (2007). Lim kinases, regulators of actin dynamics. The international journal of biochemistry & cell biology 39, 1071-1076. Berns, E. M., Klijn, J. G., van Putten, W. L., van Staveren, I. L., Portengen, H., and Foekens, J. A. (1992). c-myc amplification is a better prognostic factor than HER2/neu amplification in primary breast cancer. Cancer research 52, 1107-1113. Berns, K., Horlings, H. M., Hennessy, B. T., Madiredjo, M., Hijmans, E. M., Beelen, K., Linn, S. C., Gonzalez-Angulo, A. M., Stemke-Hale, K., Hauptmann, M., et al. (2007). A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Cancer cell 12, 395-402. Bernstein, B. E., Mikkelsen, T. S., Xie, X., Kamal, M., Huebert, D. J., Cuff, J., Fry, B., Meissner, A., Wernig, M., Plath, K., et al. (2006). A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125, 315-326. Berquin, I. M., Pang, B., Dziubinski, M. L., Scott, L. M., Chen, Y. Q., Nolan, G. P., and Ethier, S. P. (2005). Y-box-binding protein 1 confers EGF independence to human mammary epithelial cells. Oncogene 24, 3177-3186. Bertolini, G., Roz, L., Perego, P., Tortoreto, M., Fontanella, E., Gatti, L., Pratesi, G., Fabbri, A., Andriani, F., Tinelli, S., et al. (2009). Highly tumorigenic lung cancer CD133+ cells display stem-like features and are spared by cisplatin treatment. Proceedings of the National Academy of Sciences of the United States of America 106, 16281-16286. Bhullar, J., and Sollars, V. E. (2011). YBX1 expression and function in early hematopoiesis and leukemic cells. Immunogenetics 63, 337-350. Bignone, P. A., Lee, K. Y., Liu, Y., Emilion, G., Finch, J., Soosay, A. E., Charnock, F. M., Beck, S., Dunham, I., Mungall, A. J., and Ganesan, T. S. (2007). RPS6KA2, a putative tumour suppressor gene at 6q27 in sporadic epithelial ovarian cancer. Oncogene 26, 683-700. Bochar, D. A., Wang, L., Beniya, H., Kinev, A., Xue, Y., Lane, W. S., Wang, W., Kashanchi, F., and Shiekhattar, R. (2000). BRCA1 is associated with a human SWI/SNFrelated complex: linking chromatin remodeling to breast cancer. Cell 102, 257-265. Bommert, K. S., Effenberger, M., Leich, E., Kuspert, M., Murphy, D., Langer, C., Moll, R., Janz, S., Mottok, A., Weissbach, S., et al. (2012). The feed-forward loop between YB-1 and MYC is essential for multiple myeloma cell survival. Leukemia : official journal of the Leukemia Society of America, Leukemia Research Fund, UK.  184  Bonnet, D., and Dick, J. E. (1997). Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nature medicine 3, 730737. Bornstein, G., Bloom, J., Sitry-Shevah, D., Nakayama, K., Pagano, M., and Hershko, A. (2003). Role of the SCFSkp2 ubiquitin ligase in the degradation of p21Cip1 in S phase. The Journal of biological chemistry 278, 25752-25757. Bortner, D. M., and Rosenberg, M. P. (1997). Induction of mammary gland hyperplasia and carcinomas in transgenic mice expressing human cyclin E. Molecular and cellular biology 17, 453-459. Bostrom, P., Soderstrom, M., Palokangas, T., Vahlberg, T., Collan, Y., Carpen, O., and Hirsimaki, P. (2009). Analysis of cyclins A, B1, D1 and E in breast cancer in relation to tumour grade and other prognostic factors. BMC research notes 2, 140. Bourguignon, L. Y., Xia, W., and Wong, G. (2009). Hyaluronan-mediated CD44 interaction with p300 and SIRT1 regulates beta-catenin signaling and NFkappaB-specific transcription activity leading to MDR1 and Bcl-xL gene expression and chemoresistance in breast tumor cells. The Journal of biological chemistry 284, 2657-2671. Bracken, A. P., Ciro, M., Cocito, A., and Helin, K. (2004). E2F target genes: unraveling the biology. Trends in biochemical sciences 29, 409-417. Braunstein, I., Miniowitz, S., Moshe, Y., and Hershko, A. (2007). Inhibitory factors associated with anaphase-promoting complex/cylosome in mitotic checkpoint. Proceedings of the National Academy of Sciences of the United States of America 104, 4870-4875. Breasted, J. H., and Library, N.-Y. H. S. (1930). The Edwin Smith Surgical Papyrus: Published in Facsimile and Hieroglyphic Transliteration with Translation and Commentary in Two Volumes: Chic. U.P.). Brenner, A. J., and Aldaz, C. M. (1995). Chromosome 9p allelic loss and p16/CDKN2 in breast cancer and evidence of p16 inactivation in immortal breast epithelial cells. Cancer research 55, 2892-2895. Brito, D. A., and Rieder, C. L. (2006). Mitotic checkpoint slippage in humans occurs via cyclin B destruction in the presence of an active checkpoint. Current biology : CB 16, 1194-1200. Britton, K. M., Eyre, R., Harvey, I. J., Stemke-Hale, K., Browell, D., Lennard, T. W., and Meeson, A. P. (2012). Breast cancer, side population cells and ABCG2 expression. Cancer Lett 323, 97-105. Brough, R., Frankum, J. R., Sims, D., Mackay, A., Mendes-Pereira, A. M., Bajrami, I., Costa-Cabral, S., Rafiq, R., Ahmad, A. S., Cerone, M. A., et al. (2011). Functional viability profiles of breast cancer. Cancer Discov 1, 260-273. 185  Brown, K. D., Wood, K. W., and Cleveland, D. W. (1996). The kinesin-like protein CENP-E is kinetochore-associated throughout poleward chromosome segregation during anaphase-A. Journal of cell science 109 ( Pt 5), 961-969. Buckley, M. F., Sweeney, K. J., Hamilton, J. A., Sini, R. L., Manning, D. L., Nicholson, R. I., deFazio, A., Watts, C. K., Musgrove, E. A., and Sutherland, R. L. (1993). Expression and amplification of cyclin genes in human breast cancer. Oncogene 8, 21272133. Bultman, S. J., Herschkowitz, J. I., Godfrey, V., Gebuhr, T. C., Yaniv, M., Perou, C. M., and Magnuson, T. (2008). Characterization of mammary tumors from Brg1 heterozygous mice. Oncogene 27, 460-468. Burger, H., Foekens, J. A., Look, M. P., Meijer-van Gelder, M. E., Klijn, J. G., Wiemer, E. A., Stoter, G., and Nooter, K. (2003). RNA expression of breast cancer resistance protein, lung resistance-related protein, multidrug resistance-associated proteins 1 and 2, and multidrug resistance gene 1 in breast cancer: correlation with chemotherapeutic response. Clinical cancer research : an official journal of the American Association for Cancer Research 9, 827-836. Calcagno, A. M., Fostel, J. M., To, K. K., Salcido, C. D., Martin, S. E., Chewning, K. J., Wu, C. P., Varticovski, L., Bates, S. E., Caplen, N. J., and Ambudkar, S. V. (2008). Single-step doxorubicin-selected cancer cells overexpress the ABCG2 drug transporter through epigenetic changes. British journal of cancer 98, 1515-1524. Cardiff, R. D., Anver, M. R., Gusterson, B. A., Hennighausen, L., Jensen, R. A., Merino, M. J., Rehm, S., Russo, J., Tavassoli, F. A., Wakefield, L. M., et al. (2000). The mammary pathology of genetically engineered mice: the consensus report and recommendations from the Annapolis meeting. Oncogene 19, 968-988. Carey, L. A., Dees, E. C., Sawyer, L., Gatti, L., Moore, D. T., Collichio, F., Ollila, D. W., Sartor, C. I., Graham, M. L., and Perou, C. M. (2007). The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clinical cancer research : an official journal of the American Association for Cancer Research 13, 2329-2334. Carey, L. A., Perou, C. M., Livasy, C. A., Dressler, L. G., Cowan, D., Conway, K., Karaca, G., Troester, M. A., Tse, C. K., Edmiston, S., et al. (2006). Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 295, 2492-2502. Carrano, A. C., Eytan, E., Hershko, A., and Pagano, M. (1999). SKP2 is required for ubiquitin-mediated degradation of the CDK inhibitor p27. Nature cell biology 1, 193-199. Chaffer, C. L., Brueckmann, I., Scheel, C., Kaestli, A. J., Wiggins, P. A., Rodrigues, L. O., Brooks, M., Reinhardt, F., Su, Y., Polyak, K., et al. (2011). Normal and neoplastic nonstem cells can spontaneously convert to a stem-like state. Proceedings of the National Academy of Sciences of the United States of America 108, 7950-7955.  186  Chakrabarti, R., Jones, J. L., Oelschlager, D. K., Tapia, T., Tousson, A., and Grizzle, W. E. (2007). Phosphorylated LIM kinases colocalize with gamma-tubulin in centrosomes during early stages of mitosis. Cell cycle 6, 2944-2952. Chase, A., and Cross, N. C. (2011). Aberrations of EZH2 in cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 17, 26132618. Chatterjee, M., Rancso, C., Stuhmer, T., Eckstein, N., Andrulis, M., Gerecke, C., Lorentz, H., Royer, H. D., and Bargou, R. C. (2008). The Y-box binding protein YB-1 is associated with progressive disease and mediates survival and drug resistance in multiple myeloma. Blood 111, 3714-3722. Chen, J., Halappanavar, S., Th' ng, J. P., and Li, Q. (2007). Ubiquitin-dependent distribution of the transcriptional coactivator p300 in cytoplasmic inclusion bodies. Epigenetics : official journal of the DNA Methylation Society 2, 92-99. Chen, J., Halappanavar, S. S., St-Germain, J. R., Tsang, B. K., and Li, Q. (2004). Role of Akt/protein kinase B in the activity of transcriptional coactivator p300. Cell Mol Life Sci 61, 1675-1683. Chen, X., Xu, H., Yuan, P., Fang, F., Huss, M., Vega, V. B., Wong, E., Orlov, Y. L., Zhang, W., Jiang, J., et al. (2008). Integration of external signaling pathways with the core transcriptional network in embryonic stem cells. Cell 133, 1106-1117. Chen, Y. J., Huang, W. C., Wei, Y. L., Hsu, S. C., Yuan, P., Lin, H. Y., Wistuba, II, Lee, J. J., Yen, C. J., Su, W. C., et al. (2011). Elevated BCRP/ABCG2 expression confers acquired resistance to gefitinib in wild-type EGFR-expressing cells. PloS one 6, e21428. Chu, P. G., and Weiss, L. M. (2002). Keratin expression in human tissues and neoplasms. Histopathology 40, 403-439. Cimini, D., Tanzarella, C., and Degrassi, F. (1999). Differences in malsegregation rates obtained by scoring ana-telophases or binucleate cells. Mutagenesis 14, 563-568. Clark, G. J., and Der, C. J. (1995). Aberrant function of the Ras signal transduction pathway in human breast cancer. Breast cancer research and treatment 35, 133-144. Clarke, M. F., Dick, J. E., Dirks, P. B., Eaves, C. J., Jamieson, C. H., Jones, D. L., Visvader, J., Weissman, I. L., and Wahl, G. M. (2006). Cancer stem cells--perspectives on current status and future directions: AACR Workshop on cancer stem cells. Cancer research 66, 9339-9344. Crawford, Y. G., Gauthier, M. L., Joubel, A., Mantei, K., Kozakiewicz, K., Afshari, C. A., and Tlsty, T. D. (2004). Histologically normal human mammary epithelia with silenced p16(INK4a) overexpress COX-2, promoting a premalignant program. Cancer cell 5, 263-273.  187  Creighton, C. J., Li, X., Landis, M., Dixon, J. M., Neumeister, V. M., Sjolund, A., Rimm, D. L., Wong, H., Rodriguez, A., Herschkowitz, J. I., et al. (2009). Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proceedings of the National Academy of Sciences of the United States of America 106, 13820-13825. Cuello, M., Ettenberg, S. A., Clark, A. S., Keane, M. M., Posner, R. H., Nau, M. M., Dennis, P. A., and Lipkowitz, S. (2001). Down-regulation of the erbB-2 receptor by trastuzumab (herceptin) enhances tumor necrosis factor-related apoptosis-inducing ligand-mediated apoptosis in breast and ovarian cancer cell lines that overexpress erbB-2. Cancer research 61, 4892-4900. Curtis, C., Shah, S. P., Chin, S. F., Turashvili, G., Rueda, O. M., Dunning, M. J., Speed, D., Lynch, A. G., Samarajiwa, S., Yuan, Y., et al. (2012). The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 486, 346-352. Dahl, E., En-Nia, A., Wiesmann, F., Krings, R., Djudjaj, S., Breuer, E., Fuchs, T., Wild, P. J., Hartmann, A., Dunn, S. E., and Mertens, P. R. (2009). Nuclear detection of Y-box protein-1 (YB-1) closely associates with progesterone receptor negativity and is a strong adverse survival factor in human breast cancer. BMC cancer 9, 410. Davies, A. H., Barrett, I., Pambid, M. R., Hu, K., Stratford, A. L., Freeman, S., Berquin, I. M., Pelech, S., Hieter, P., Maxwell, C., and Dunn, S. E. (2011). YB-1 evokes susceptibility to cancer through cytokinesis failure, mitotic dysfunction and HER2 amplification. Oncogene 30, 3649-3660. Davila, M., Jhala, D., Ghosh, D., Grizzle, W. E., and Chakrabarti, R. (2007). Expression of LIM kinase 1 is associated with reversible G1/S phase arrest, chromosomal instability and prostate cancer. Molecular cancer 6, 40. Dawson, P. J., Wolman, S. R., Tait, L., Heppner, G. H., and Miller, F. R. (1996). MCF10AT: a model for the evolution of cancer from proliferative breast disease. The American journal of pathology 148, 313-319. De Antoni, A., Pearson, C. G., Cimini, D., Canman, J. C., Sala, V., Nezi, L., Mapelli, M., Sironi, L., Faretta, M., Salmon, E. D., and Musacchio, A. (2005). The Mad1/Mad2 complex as a template for Mad2 activation in the spindle assembly checkpoint. Current biology : CB 15, 214-225. De Laurentiis, M., Cianniello, D., Caputo, R., Stanzione, B., Arpino, G., Cinieri, S., Lorusso, V., and De Placido, S. (2010). Treatment of triple negative breast cancer (TNBC): current options and future perspectives. Cancer treatment reviews 36 Suppl 3, S80-86. Dean, M., Fojo, T., and Bates, S. (2005). Tumour stem cells and drug resistance. Nature reviews Cancer 5, 275-284.  188  Dean, M., Rzhetsky, A., and Allikmets, R. (2001). The human ATP-binding cassette (ABC) transporter superfamily. Genome research 11, 1156-1166. Debnath, J., and Brugge, J. S. (2005). Modelling glandular epithelial cancers in threedimensional cultures. Nature reviews Cancer 5, 675-688. Debnath, J., Muthuswamy, S. K., and Brugge, J. S. (2003). Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures. Methods 30, 256-268. Demicheli, R., Miceli, R., Moliterni, A., Zambetti, M., Hrushesky, W. J., Retsky, M. W., Valagussa, P., and Bonadonna, G. (2005). Breast cancer recurrence dynamics following adjuvant CMF is consistent with tumor dormancy and mastectomy-driven acceleration of the metastatic process. Ann Oncol 16, 1449-1457. Dent, R., Trudeau, M., Pritchard, K. I., Hanna, W. M., Kahn, H. K., Sawka, C. A., Lickley, L. A., Rawlinson, E., Sun, P., and Narod, S. A. (2007). Triple-negative breast cancer: clinical features and patterns of recurrence. Clinical cancer research : an official journal of the American Association for Cancer Research 13, 4429-4434. Deome, K. B., Faulkin, L. J., Jr., Bern, H. A., and Blair, P. B. (1959). Development of mammary tumors from hyperplastic alveolar nodules transplanted into gland-free mammary fat pads of female C3H mice. Cancer research 19, 515-520. Derksen, P. W., Liu, X., Saridin, F., van der Gulden, H., Zevenhoven, J., Evers, B., van Beijnum, J. R., Griffioen, A. W., Vink, J., Krimpenfort, P., et al. (2006). Somatic inactivation of E-cadherin and p53 in mice leads to metastatic lobular mammary carcinoma through induction of anoikis resistance and angiogenesis. Cancer cell 10, 437449. Derynck, R., Akhurst, R. J., and Balmain, A. (2001). TGF-beta signaling in tumor suppression and cancer progression. Nature genetics 29, 117-129. Dhillon, J., Astanehe, A., Lee, C., Fotovati, A., Hu, K., and Dunn, S. E. (2010). The expression of activated Y-box binding protein-1 serine 102 mediates trastuzumab resistance in breast cancer cells by increasing CD44+ cells. Oncogene 29, 6294-6300. Diasio, R. B., and Harris, B. E. (1989). Clinical pharmacology of 5-fluorouracil. Clinical pharmacokinetics 16, 215-237. Didier, D. K., Schiffenbauer, J., Woulfe, S. L., Zacheis, M., and Schwartz, B. D. (1988). Characterization of the cDNA encoding a protein binding to the major histocompatibility complex class II Y box. Proceedings of the National Academy of Sciences of the United States of America 85, 7322-7326. Dimri, G. P., Martinez, J. L., Jacobs, J. J., Keblusek, P., Itahana, K., Van Lohuizen, M., Campisi, J., Wazer, D. E., and Band, V. (2002). The Bmi-1 oncogene induces telomerase  189  activity and immortalizes human mammary epithelial cells. Cancer research 62, 47364745. Ding, L., Ellis, M. J., Li, S., Larson, D. E., Chen, K., Wallis, J. W., Harris, C. C., McLellan, M. D., Fulton, R. S., Fulton, L. L., et al. (2010). Genome remodelling in a basal-like breast cancer metastasis and xenograft. Nature 464, 999-1005. Ding, L., Erdmann, C., Chinnaiyan, A. M., Merajver, S. D., and Kleer, C. G. (2006). Identification of EZH2 as a molecular marker for a precancerous state in morphologically normal breast tissues. Cancer research 66, 4095-4099. Dontu, G., Abdallah, W. M., Foley, J. M., Jackson, K. W., Clarke, M. F., Kawamura, M. J., and Wicha, M. S. (2003). In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes & development 17, 1253-1270. Doyle, L., and Ross, D. D. (2003). Multidrug resistance mediated by the breast cancer resistance protein BCRP (ABCG2). Oncogene 22, 7340-7358. Doyle, L. A., Yang, W., Abruzzo, L. V., Krogmann, T., Gao, Y., Rishi, A. K., and Ross, D. D. (1998). A multidrug resistance transporter from human MCF-7 breast cancer cells. Proceedings of the National Academy of Sciences of the United States of America 95, 15665-15670. Dufresne, S. D., Bjorbaek, C., El-Haschimi, K., Zhao, Y., Aschenbach, W. G., Moller, D. E., and Goodyear, L. J. (2001). Altered extracellular signal-regulated kinase signaling and glycogen metabolism in skeletal muscle from p90 ribosomal S6 kinase 2 knockout mice. Molecular and cellular biology 21, 81-87. Dumont, N., Crawford, Y. G., Sigaroudinia, M., Nagrani, S. S., Wilson, M. B., Buehring, G. C., Turashvili, G., Aparicio, S., Gauthier, M. L., Fordyce, C. A., et al. (2009). Human mammary cancer progression model recapitulates methylation events associated with breast premalignancy. Breast cancer research : BCR 11, R87. Dunn, B. K., and Ford, L. G. (2007). Hormonal interventions to prevent hormonal cancers: breast and prostate cancers. European journal of cancer prevention : the official journal of the European Cancer Prevention Organisation 16, 232-242. Eckner, R., Ewen, M. E., Newsome, D., Gerdes, M., DeCaprio, J. A., Lawrence, J. B., and Livingston, D. M. (1994). Molecular cloning and functional analysis of the adenovirus E1A-associated 300-kD protein (p300) reveals a protein with properties of a transcriptional adaptor. Genes & development 8, 869-884. Eirew, P., Kannan, N., Knapp, D. J., Vaillant, F., Emerman, J. T., Lindeman, G. J., Visvader, J. E., and Eaves, C. J. (2012). Aldehyde dehydrogenase activity is a biomarker of primitive normal human mammary luminal cells. Stem cells 30, 344-348.  190  Eirew, P., Stingl, J., Raouf, A., Turashvili, G., Aparicio, S., Emerman, J. T., and Eaves, C. J. (2008). A method for quantifying normal human mammary epithelial stem cells with in vivo regenerative ability. Nature medicine 14, 1384-1389. Elenbaas, B., Spirio, L., Koerner, F., Fleming, M. D., Zimonjic, D. B., Donaher, J. L., Popescu, N. C., Hahn, W. C., and Weinberg, R. A. (2001). Human breast cancer cells generated by oncogenic transformation of primary mammary epithelial cells. Genes & development 15, 50-65. Eliseeva, I. A., Kim, E. R., Guryanov, S. G., Ovchinnikov, L. P., and Lyabin, D. N. (2011). Y-box-binding protein 1 (YB-1) and its functions. Biochemistry Biokhimiia 76, 1402-1433. Evdokimova, V., Ruzanov, P., Anglesio, M. S., Sorokin, A. V., Ovchinnikov, L. P., Buckley, J., Triche, T. J., Sonenberg, N., and Sorensen, P. H. (2006). Akt-mediated YB-1 phosphorylation activates translation of silent mRNA species. Molecular and cellular biology 26, 277-292. Evdokimova, V., Tognon, C., Ng, T., Ruzanov, P., Melnyk, N., Fink, D., Sorokin, A., Ovchinnikov, L. P., Davicioni, E., Triche, T. J., and Sorensen, P. H. (2009). Translational activation of snail1 and other developmentally regulated transcription factors by YB-1 promotes an epithelial-mesenchymal transition. Cancer cell 15, 402-415. Fabbro, M., Savage, K., Hobson, K., Deans, A. J., Powell, S. N., McArthur, G. A., and Khanna, K. K. (2004). BRCA1-BARD1 complexes are required for p53Ser-15 phosphorylation and a G1/S arrest following ionizing radiation-induced DNA damage. The Journal of biological chemistry 279, 31251-31258. Fan, S., Ma, Y. X., Wang, C., Yuan, R. Q., Meng, Q., Wang, J. A., Erdos, M., Goldberg, I. D., Webb, P., Kushner, P. J., et al. (2002). p300 Modulates the BRCA1 inhibition of estrogen receptor activity. Cancer research 62, 141-151. Faury, D., Nantel, A., Dunn, S. E., Guiot, M. C., Haque, T., Hauser, P., Garami, M., Bognar, L., Hanzely, Z., Liberski, P. P., et al. (2007). Molecular profiling identifies prognostic subgroups of pediatric glioblastoma and shows increased YB-1 expression in tumors. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 25, 1196-1208. Fermento, M. E., Gandini, N. A., Lang, C. A., Perez, J. E., Maturi, H. V., Curino, A. C., and Facchinetti, M. M. (2010). Intracellular distribution of p300 and its differential recruitment to aggresomes in breast cancer. Experimental and molecular pathology 88, 256-264. Fillmore, C. M., and Kuperwasser, C. (2008). Human breast cancer cell lines contain stem-like cells that self-renew, give rise to phenotypically diverse progeny and survive chemotherapy. Breast cancer research : BCR 10, R25.  191  Finkbeiner, M. R., Astanehe, A., To, K., Fotovati, A., Davies, A. H., Zhao, Y., Jiang, H., Stratford, A. L., Shadeo, A., Boccaccio, C., et al. (2009). Profiling YB-1 target genes uncovers a new mechanism for MET receptor regulation in normal and malignant human mammary cells. Oncogene 28, 1421-1431. Foekens, J. A., Peters, H. A., Look, M. P., Portengen, H., Schmitt, M., Kramer, M. D., Brunner, N., Janicke, F., Meijer-van Gelder, M. E., Henzen-Logmans, S. C., et al. (2000). The urokinase system of plasminogen activation and prognosis in 2780 breast cancer patients. Cancer research 60, 636-643. Fotovati, A., Abu-Ali, S., Wang, P. S., Deleyrolle, L. P., Lee, C., Triscott, J., Chen, J. Y., Franciosi, S., Nakamura, Y., Sugita, Y., et al. (2011). YB-1 bridges neural stem cells and brain tumor-initiating cells via its roles in differentiation and cell growth. Cancer research 71, 5569-5578. Foulkes, W. D. (2007). Clinically relevant biology of hereditary breast cancer. Seminars in oncology 34, 379-383. Foulkes, W. D. (2008). Inherited susceptibility to common cancers. The New England journal of medicine 359, 2143-2153. Fujii, H., Biel, M. A., Zhou, W., Weitzman, S. A., Baylin, S. B., and Gabrielson, E. (1998). Methylation of the HIC-1 candidate tumor suppressor gene in human breast cancer. Oncogene 16, 2159-2164. Fujii, H., Szumel, R., Marsh, C., Zhou, W., and Gabrielson, E. (1996). Genetic progression, histological grade, and allelic loss in ductal carcinoma in situ of the breast. Cancer research 56, 5260-5265. Fujita, T., Ito, K., Izumi, H., Kimura, M., Sano, M., Nakagomi, H., Maeno, K., Hama, Y., Shingu, K., Tsuchiya, S., et al. (2005). Increased nuclear localization of transcription factor Y-box binding protein 1 accompanied by up-regulation of P-glycoprotein in breast cancer pretreated with paclitaxel. Clinical cancer research : an official journal of the American Association for Cancer Research 11, 8837-8844. Fujiwara, T., Bandi, M., Nitta, M., Ivanova, E. V., Bronson, R. T., and Pellman, D. (2005). Cytokinesis failure generating tetraploids promotes tumorigenesis in p53-null cells. Nature 437, 1043-1047. Ganem, N. J., Godinho, S. A., and Pellman, D. (2009). A mechanism linking extra centrosomes to chromosomal instability. Nature 460, 278-282. Gao, C., Furge, K., Koeman, J., Dykema, K., Su, Y., Cutler, M. L., Werts, A., Haak, P., and Vande Woude, G. F. (2007). Chromosome instability, chromosome transcriptome, and clonal evolution of tumor cell populations. Proceedings of the National Academy of Sciences of the United States of America 104, 8995-9000.  192  Gasco, M., Shami, S., and Crook, T. (2002). The p53 pathway in breast cancer. Breast cancer research : BCR 4, 70-76. Gaudreault, I., Guay, D., and Lebel, M. (2004). YB-1 promotes strand separation in vitro of duplex DNA containing either mispaired bases or cisplatin modifications, exhibits endonucleolytic activities and binds several DNA repair proteins. Nucleic acids research 32, 316-327. Gayther, S. A., Batley, S. J., Linger, L., Bannister, A., Thorpe, K., Chin, S. F., Daigo, Y., Russell, P., Wilson, A., Sowter, H. M., et al. (2000). Mutations truncating the EP300 acetylase in human cancers. Nature genetics 24, 300-303. Geng, Y., Lee, Y. M., Welcker, M., Swanger, J., Zagozdzon, A., Winer, J. D., Roberts, J. M., Kaldis, P., Clurman, B. E., and Sicinski, P. (2007). Kinase-independent function of cyclin E. Molecular cell 25, 127-139. Geradts, J., and Wilson, P. A. (1996). High frequency of aberrant p16(INK4A) expression in human breast cancer. The American journal of pathology 149, 15-20. Geyer, C. E., Forster, J., Lindquist, D., Chan, S., Romieu, C. G., Pienkowski, T., JagielloGruszfeld, A., Crown, J., Chan, A., Kaufman, B., et al. (2006). Lapatinib plus capecitabine for HER2-positive advanced breast cancer. The New England journal of medicine 355, 2733-2743. Ghatak, S., Misra, S., and Toole, B. P. (2005). Hyaluronan constitutively regulates ErbB2 phosphorylation and signaling complex formation in carcinoma cells. The Journal of biological chemistry 280, 8875-8883. Gimenez-Bonafe, P., Fedoruk, M. N., Whitmore, T. G., Akbari, M., Ralph, J. L., Ettinger, S., Gleave, M. E., and Nelson, C. C. (2004). YB-1 is upregulated during prostate cancer tumor progression and increases P-glycoprotein activity. Prostate 59, 337-349. Ginestier, C., Hur, M. H., Charafe-Jauffret, E., Monville, F., Dutcher, J., Brown, M., Jacquemier, J., Viens, P., Kleer, C. G., Liu, S., et al. (2007). ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell stem cell 1, 555-567. Ginsburg, O. M., Akbari, M. R., Aziz, Z., Young, R., Lynch, H., Ghadirian, P., Robidoux, A., Londono, J., Vasquez, G., Gomes, M., et al. (2009). The prevalence of germ-line TP53 mutations in women diagnosed with breast cancer before age 30. Familial cancer 8, 563-567. Gluz, O., Liedtke, C., Gottschalk, N., Pusztai, L., Nitz, U., and Harbeck, N. (2009a). Triple-negative breast cancer--current status and future directions. Ann Oncol 20, 19131927. Gluz, O., Mengele, K., Schmitt, M., Kates, R., Diallo-Danebrock, R., Neff, F., Royer, H. D., Eckstein, N., Mohrmann, S., Ting, E., et al. (2009b). Y-box-binding protein YB-1 193  identifies high-risk patients with primary breast cancer benefiting from rapidly cycled tandem high-dose adjuvant chemotherapy. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 27, 6144-6151. Gonzalez, S., Klatt, P., Delgado, S., Conde, E., Lopez-Rios, F., Sanchez-Cespedes, M., Mendez, J., Antequera, F., and Serrano, M. (2006). Oncogenic activity of Cdc6 through repression of the INK4/ARF locus. Nature 440, 702-706. Gottesman, M. M., Fojo, T., and Bates, S. E. (2002). Multidrug resistance in cancer: role of ATP-dependent transporters. Nature reviews Cancer 2, 48-58. Green, J. E., Shibata, M. A., Yoshidome, K., Liu, M. L., Jorcyk, C., Anver, M. R., Wigginton, J., Wiltrout, R., Shibata, E., Kaczmarczyk, S., et al. (2000). The C3(1)/SV40 T-antigen transgenic mouse model of mammary cancer: ductal epithelial cell targeting with multistage progression to carcinoma. Oncogene 19, 1020-1027. Greenberg, P. A., Hortobagyi, G. N., Smith, T. L., Ziegler, L. D., Frye, D. K., and Buzdar, A. U. (1996). Long-term follow-up of patients with complete remission following combination chemotherapy for metastatic breast cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 14, 2197-2205. Grossman, S. R., Perez, M., Kung, A. L., Joseph, M., Mansur, C., Xiao, Z. X., Kumar, S., Howley, P. M., and Livingston, D. M. (1998). p300/MDM2 complexes participate in MDM2-mediated p53 degradation. Molecular cell 2, 405-415. Gruss, O. J., Wittmann, M., Yokoyama, H., Pepperkok, R., Kufer, T., Sillje, H., Karsenti, E., Mattaj, I. W., and Vernos, I. (2002). Chromosome-induced microtubule assembly mediated by TPX2 is required for spindle formation in HeLa cells. Nature cell biology 4, 871-879. Guay, D., Gaudreault, I., Massip, L., and Lebel, M. (2006). Formation of a nuclear complex containing the p53 tumor suppressor, YB-1, and the Werner syndrome gene product in cells treated with UV light. The international journal of biochemistry & cell biology 38, 1300-1313. Guo, B., Villeneuve, D. J., Hembruff, S. L., Kirwan, A. F., Blais, D. E., Bonin, M., and Parissenti, A. M. (2004). Cross-resistance studies of isogenic drug-resistant breast tumor cell lines support recent clinical evidence suggesting that sensitivity to paclitaxel may be strongly compromised by prior doxorubicin exposure. Breast cancer research and treatment 85, 31-51. Guo, B. H., Feng, Y., Zhang, R., Xu, L. H., Li, M. Z., Kung, H. F., Song, L. B., and Zeng, M. S. (2011). Bmi-1 promotes invasion and metastasis, and its elevated expression is correlated with an advanced stage of breast cancer. Molecular cancer 10, 10. Gupta, A., Inaba, S., Wong, O. K., Fang, G., and Liu, J. (2003). Breast cancer-specific gene 1 interacts with the mitotic checkpoint kinase BubR1. Oncogene 22, 7593-7599.  194  Gupta, P. B., Chaffer, C. L., and Weinberg, R. A. (2009). Cancer stem cells: mirage or reality? Nature medicine 15, 1010-1012. Gupta, P. B., Fillmore, C. M., Jiang, G., Shapira, S. D., Tao, K., Kuperwasser, C., and Lander, E. S. (2011). Stochastic state transitions give rise to phenotypic equilibrium in populations of cancer cells. Cell 146, 633-644. Habibi, G., Leung, S., Law, J. H., Gelmon, K., Masoudi, H., Turbin, D., Pollak, M., Nielsen, T. O., Huntsman, D., and Dunn, S. E. (2008). Redefining prognostic factors for breast cancer: YB-1 is a stronger predictor of relapse and disease-specific survival than estrogen receptor or HER-2 across all tumor subtypes. Breast cancer research : BCR 10, R86. Haering, C. H., Farcas, A. M., Arumugam, P., Metson, J., and Nasmyth, K. (2008). The cohesin ring concatenates sister DNA molecules. Nature 454, 297-301. Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57-70. Hanahan, D., and Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674. Harbour, J. W., Luo, R. X., Dei Santi, A., Postigo, A. A., and Dean, D. C. (1999). Cdk phosphorylation triggers sequential intramolecular interactions that progressively block Rb functions as cells move through G1. Cell 98, 859-869. Harris, R. E., Beebe-Donk, J., and Alshafie, G. A. (2006). Reduction in the risk of human breast cancer by selective cyclooxygenase-2 (COX-2) inhibitors. BMC cancer 6, 27. Hartwell, L. H., Culotti, J., Pringle, J. R., and Reid, B. J. (1974). Genetic control of the cell division cycle in yeast. Science 183, 46-51. Hartwell, L. H., and Weinert, T. A. (1989). Checkpoints: controls that ensure the order of cell cycle events. Science 246, 629-634. Haupt, Y., Alexander, W. S., Barri, G., Klinken, S. P., and Adams, J. M. (1991). Novel zinc finger gene implicated as myc collaborator by retrovirally accelerated lymphomagenesis in E mu-myc transgenic mice. Cell 65, 753-763. Hayflick, L., and Moorhead, P. S. (1961). The serial cultivation of human diploid cell strains. Experimental cell research 25, 585-621. Henderson, L. J., Coe, B. P., Lee, E. H., Girard, L., Gazdar, A. F., Minna, J. D., Lam, S., MacAulay, C., and Lam, W. L. (2005). Genomic and gene expression profiling of minute alterations of chromosome arm 1p in small-cell lung carcinoma cells. British journal of cancer 92, 1553-1560. Heppner, G. H. (1984). Tumor heterogeneity. Cancer research 44, 2259-2265.  195  Herschkowitz, J. I., Simin, K., Weigman, V. J., Mikaelian, I., Usary, J., Hu, Z., Rasmussen, K. E., Jones, L. P., Assefnia, S., Chandrasekharan, S., et al. (2007). Identification of conserved gene expression features between murine mammary carcinoma models and human breast tumors. Genome biology 8, R76. Hinchcliffe, E. H., Li, C., Thompson, E. A., Maller, J. L., and Sluder, G. (1999). Requirement of Cdk2-cyclin E activity for repeated centrosome reproduction in Xenopus egg extracts. Science 283, 851-854. Hirschmann-Jax, C., Foster, A. E., Wulf, G. G., Nuchtern, J. G., Jax, T. W., Gobel, U., Goodell, M. A., and Brenner, M. K. (2004). A distinct "side population" of cells with high drug efflux capacity in human tumor cells. Proceedings of the National Academy of Sciences of the United States of America 101, 14228-14233. Holland, A. J., and Cleveland, D. W. (2009). Boveri revisited: chromosomal instability, aneuploidy and tumorigenesis. Nature reviews Molecular cell biology 10, 478-487. Holst, C. R., Nuovo, G. J., Esteller, M., Chew, K., Baylin, S. B., Herman, J. G., and Tlsty, T. D. (2003). Methylation of p16(INK4a) promoters occurs in vivo in histologically normal human mammary epithelia. Cancer research 63, 1596-1601. Hon, G. C., Hawkins, R. D., and Ren, B. (2009). Predictive chromatin signatures in the mammalian genome. Human molecular genetics 18, R195-201. Honeth, G., Bendahl, P. O., Ringner, M., Saal, L. H., Gruvberger-Saal, S. K., Lovgren, K., Grabau, D., Ferno, M., Borg, A., and Hegardt, C. (2008). The CD44+/CD24phenotype is enriched in basal-like breast tumors. Breast cancer research : BCR 10, R53. Hoos, A., Hepp, H. H., Kaul, S., Ahlert, T., Bastert, G., and Wallwiener, D. (1998). Telomerase activity correlates with tumor aggressiveness and reflects therapy effect in breast cancer. International journal of cancer Journal international du cancer 79, 8-12. Hsu, C. H., Peng, K. L., Kang, M. L., Chen, Y. R., Yang, Y. C., Tsai, C. H., Chu, C. S., Jeng, Y. M., Chen, Y. T., Lin, F. M., et al. (2012). TET1 Suppresses Cancer Invasion by Activating the Tissue Inhibitors of Metalloproteinases. Cell reports 2, 568-579. Huober, J., von Minckwitz, G., Denkert, C., Tesch, H., Weiss, E., Zahm, D. M., Belau, A., Khandan, F., Hauschild, M., Thomssen, C., et al. (2010). Effect of neoadjuvant anthracycline-taxane-based chemotherapy in different biological breast cancer phenotypes: overall results from the GeparTrio study. Breast cancer research and treatment 124, 133-140. Ida, K., Kitabayashi, I., Taki, T., Taniwaki, M., Noro, K., Yamamoto, M., Ohki, M., and Hayashi, Y. (1997). Adenoviral E1A-associated protein p300 is involved in acute myeloid leukemia with t(11;22)(q23;q13). Blood 90, 4699-4704.  196  Ince, T. A., Richardson, A. L., Bell, G. W., Saitoh, M., Godar, S., Karnoub, A. E., Iglehart, J. D., and Weinberg, R. A. (2007). Transformation of different human breast epithelial cell types leads to distinct tumor phenotypes. Cancer cell 12, 160-170. Ise, T., Nagatani, G., Imamura, T., Kato, K., Takano, H., Nomoto, M., Izumi, H., Ohmori, H., Okamoto, T., Ohga, T., et al. (1999). Transcription factor Y-box binding protein 1 binds preferentially to cisplatin-modified DNA and interacts with proliferating cell nuclear antigen. Cancer research 59, 342-346. James, D., Levine, A. J., Besser, D., and Hemmati-Brivanlou, A. (2005). TGFbeta/activin/nodal signaling is necessary for the maintenance of pluripotency in human embryonic stem cells. Development 132, 1273-1282. Janssen, A., van der Burg, M., Szuhai, K., Kops, G. J., and Medema, R. H. (2011). Chromosome segregation errors as a cause of DNA damage and structural chromosome aberrations. Science 333, 1895-1898. Janz, M., Harbeck, N., Dettmar, P., Berger, U., Schmidt, A., Jurchott, K., Schmitt, M., and Royer, H. D. (2002). Y-box factor YB-1 predicts drug resistance and patient outcome in breast cancer independent of clinically relevant tumor biologic factors HER2, uPA and PAI-1. International journal of cancer Journal international du cancer 97, 278-282. Joensuu, H., Kellokumpu-Lehtinen, P. L., Bono, P., Alanko, T., Kataja, V., Asola, R., Utriainen, T., Kokko, R., Hemminki, A., Tarkkanen, M., et al. (2006). Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. The New England journal of medicine 354, 809-820. Joensuu, K., Hagstrom, J., Leidenius, M., Haglund, C., Andersson, L. C., Sariola, H., and Heikkila, P. (2011). Bmi-1, c-myc, and Snail expression in primary breast cancers and their metastases--elevated Bmi-1 expression in late breast cancer relapses. Virchows Archiv : an international journal of pathology 459, 31-39. Jones, P. A., and Liang, G. (2009). Rethinking how DNA methylation patterns are maintained. Nature reviews Genetics 10, 805-811. Jones, S., Chen, W. D., Parmigiani, G., Diehl, F., Beerenwinkel, N., Antal, T., Traulsen, A., Nowak, M. A., Siegel, C., Velculescu, V. E., et al. (2008). Comparative lesion sequencing provides insights into tumor evolution. Proceedings of the National Academy of Sciences of the United States of America 105, 4283-4288. Jorgensen, H. F., Giadrossi, S., Casanova, M., Endoh, M., Koseki, H., Brockdorff, N., and Fisher, A. G. (2006). Stem cells primed for action: polycomb repressive complexes restrain the expression of lineage-specific regulators in embryonic stem cells. Cell cycle 5, 1411-1414. Juliano, R. L., and Ling, V. (1976). A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochimica et biophysica acta 455, 152-162.  197  Jurchott, K., Bergmann, S., Stein, U., Walther, W., Janz, M., Manni, I., Piaggio, G., Fietze, E., Dietel, M., and Royer, H. D. (2003). YB-1 as a cell cycle-regulated transcription factor facilitating cyclin A and cyclin B1 gene expression. The Journal of biological chemistry 278, 27988-27996. Kallin, E. M., Cao, R., Jothi, R., Xia, K., Cui, K., Zhao, K., and Zhang, Y. (2009). Genome-wide uH2A localization analysis highlights Bmi1-dependent deposition of the mark at repressed genes. PLoS genetics 5, e1000506. Kalluri, R., and Weinberg, R. A. (2009). The basics of epithelial-mesenchymal transition. The Journal of clinical investigation 119, 1420-1428. Kamura, T., Yahata, H., Amada, S., Ogawa, S., Sonoda, T., Kobayashi, H., Mitsumoto, M., Kohno, K., Kuwano, M., and Nakano, H. (1999). Is nuclear expression of Y boxbinding protein-1 a new prognostic factor in ovarian serous adenocarcinoma? Cancer 85, 2450-2454. Kang, S., Dong, S., Gu, T. L., Guo, A., Cohen, M. S., Lonial, S., Khoury, H. J., Fabbro, D., Gilliland, D. G., Bergsagel, P. L., et al. (2007). FGFR3 activates RSK2 to mediate hematopoietic transformation through tyrosine phosphorylation of RSK2 and activation of the MEK/ERK pathway. Cancer cell 12, 201-214. Kastan, M. B., and Bartek, J. (2004). Cell-cycle checkpoints and cancer. Nature 432, 316323. Kastan, M. B., Onyekwere, O., Sidransky, D., Vogelstein, B., and Craig, R. W. (1991). Participation of p53 protein in the cellular response to DNA damage. Cancer research 51, 6304-6311. Keane, M. M., Ettenberg, S. A., Lowrey, G. A., Russell, E. K., and Lipkowitz, S. (1996). Fas expression and function in normal and malignant breast cell lines. Cancer research 56, 4791-4798. Kellogg, D. R., Moritz, M., and Alberts, B. M. (1994). The centrosome and cellular organization. Annual review of biochemistry 63, 639-674. Kelly, R. J., Draper, D., Chen, C. C., Robey, R. W., Figg, W. D., Piekarz, R. L., Chen, X., Gardner, E. R., Balis, F. M., Venkatesan, A. M., et al. (2011). A pharmacodynamic study of docetaxel in combination with the P-glycoprotein antagonist tariquidar (XR9576) in patients with lung, ovarian, and cervical cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 17, 569-580. Kelsey, J. L., and Berkowitz, G. S. (1988). Breast cancer epidemiology. Cancer research 48, 5615-5623. Kennecke, H., Yerushalmi, R., Woods, R., Cheang, M. C., Voduc, D., Speers, C. H., Nielsen, T. O., and Gelmon, K. (2010). Metastatic behavior of breast cancer subtypes.  198  Journal of clinical oncology : official journal of the American Society of Clinical Oncology 28, 3271-3277. Keyomarsi, K., and Pardee, A. B. (1993). Redundant cyclin overexpression and gene amplification in breast cancer cells. Proceedings of the National Academy of Sciences of the United States of America 90, 1112-1116. Khalil, A. M., Guttman, M., Huarte, M., Garber, M., Raj, A., Rivea Morales, D., Thomas, K., Presser, A., Bernstein, B. E., van Oudenaarden, A., et al. (2009). Many human large intergenic noncoding RNAs associate with chromatin-modifying complexes and affect gene expression. Proceedings of the National Academy of Sciences of the United States of America 106, 11667-11672. Kim, R., Emi, M., and Tanabe, K. (2007). Cancer immunoediting from immune surveillance to immune escape. Immunology 121, 1-14. Kim, W. Y., and Sharpless, N. E. (2006). The regulation of INK4/ARF in cancer and aging. Cell 127, 265-275. Kinney, S. R., and Pradhan, S. (2011). Regulation of expression and activity of DNA (cytosine-5) methyltransferases in mammalian cells. Progress in molecular biology and translational science 101, 311-333. Kinzler, K. W., and Vogelstein, B. (1997). Cancer-susceptibility genes. Gatekeepers and caretakers. Nature 386, 761, 763. Kitagawa, R., Bakkenist, C. J., McKinnon, P. J., and Kastan, M. B. (2004). Phosphorylation of SMC1 is a critical downstream event in the ATM-NBS1-BRCA1 pathway. Genes & development 18, 1423-1438. Kleer, C. G., Cao, Q., Varambally, S., Shen, R., Ota, I., Tomlins, S. A., Ghosh, D., Sewalt, R. G., Otte, A. P., Hayes, D. F., et al. (2003). EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells. Proceedings of the National Academy of Sciences of the United States of America 100, 11606-11611. Knutson, S. K., Wigle, T. J., Warholic, N. M., Sneeringer, C. J., Allain, C. J., Klaus, C. R., Sacks, J. D., Raimondi, A., Majer, C. R., Song, J., et al. (2012). A selective inhibitor of EZH2 blocks H3K27 methylation and kills mutant lymphoma cells. Nature chemical biology 8, 890-896. Kohno, K., Izumi, H., Uchiumi, T., Ashizuka, M., and Kuwano, M. (2003). The pleiotropic functions of the Y-box-binding protein, YB-1. BioEssays : news and reviews in molecular, cellular and developmental biology 25, 691-698. Kolluri, R., and Kinniburgh, A. J. (1991). Full length cDNA sequence encoding a nuclease-sensitive element DNA binding protein. Nucleic acids research 19, 4771.  199  Kops, G. J., Foltz, D. R., and Cleveland, D. W. (2004). Lethality to human cancer cells through massive chromosome loss by inhibition of the mitotic checkpoint. Proceedings of the National Academy of Sciences of the United States of America 101, 8699-8704. Kouzarides, T. (2007). Chromatin modifications and their function. Cell 128, 693-705. Kowalski, P. J., Rubin, M. A., and Kleer, C. G. (2003). E-cadherin expression in primary carcinomas of the breast and its distant metastases. Breast cancer research : BCR 5, R217-222. Kranz, A., Mattfeldt, T., and Waltenberger, J. (1999). Molecular mediators of tumor angiogenesis: enhanced expression and activation of vascular endothelial growth factor receptor KDR in primary breast cancer. International journal of cancer Journal international du cancer 84, 293-298. Kreike, B., van Kouwenhove, M., Horlings, H., Weigelt, B., Peterse, H., Bartelink, H., and van de Vijver, M. J. (2007). Gene expression profiling and histopathological characterization of triple-negative/basal-like breast carcinomas. Breast cancer research : BCR 9, R65. Krejci, J., Uhlirova, R., Galiova, G., Kozubek, S., Smigova, J., and Bartova, E. (2009). Genome-wide reduction in H3K9 acetylation during human embryonic stem cell differentiation. J Cell Physiol 219, 677-687. Kufer, T. A., Sillje, H. H., Korner, R., Gruss, O. J., Meraldi, P., and Nigg, E. A. (2002). Human TPX2 is required for targeting Aurora-A kinase to the spindle. The Journal of cell biology 158, 617-623. Kuperwasser, C., Chavarria, T., Wu, M., Magrane, G., Gray, J. W., Carey, L., Richardson, A., and Weinberg, R. A. (2004). Reconstruction of functionally normal and malignant human breast tissues in mice. Proceedings of the National Academy of Sciences of the United States of America 101, 4966-4971. Kuzmichev, A., Margueron, R., Vaquero, A., Preissner, T. S., Scher, M., Kirmizis, A., Ouyang, X., Brockdorff, N., Abate-Shen, C., Farnham, P., and Reinberg, D. (2005). Composition and histone substrates of polycomb repressive group complexes change during cellular differentiation. Proceedings of the National Academy of Sciences of the United States of America 102, 1859-1864. Lacey, K. R., Jackson, P. K., and Stearns, T. (1999). Cyclin-dependent kinase control of centrosome duplication. Proceedings of the National Academy of Sciences of the United States of America 96, 2817-2822. Lal, A., Kim, H. H., Abdelmohsen, K., Kuwano, Y., Pullmann, R., Jr., Srikantan, S., Subrahmanyam, R., Martindale, J. L., Yang, X., Ahmed, F., et al. (2008). p16(INK4a) translation suppressed by miR-24. PloS one 3, e1864.  200  Lanni, J. S., Lowe, S. W., Licitra, E. J., Liu, J. O., and Jacks, T. (1997). p53-independent apoptosis induced by paclitaxel through an indirect mechanism. Proceedings of the National Academy of Sciences of the United States of America 94, 9679-9683. Lapidot, T., Sirard, C., Vormoor, J., Murdoch, B., Hoang, T., Caceres-Cortes, J., Minden, M., Paterson, B., Caligiuri, M. A., and Dick, J. E. (1994). A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature 367, 645-648. Larkin, A., O'Driscoll, L., Kennedy, S., Purcell, R., Moran, E., Crown, J., Parkinson, M., and Clynes, M. (2004). Investigation of MRP-1 protein and MDR-1 P-glycoprotein expression in invasive breast cancer: a prognostic study. International journal of cancer Journal international du cancer 112, 286-294. Larochelle, A., Vormoor, J., Hanenberg, H., Wang, J. C., Bhatia, M., Lapidot, T., Moritz, T., Murdoch, B., Xiao, X. L., Kato, I., et al. (1996). Identification of primitive human hematopoietic cells capable of repopulating NOD/SCID mouse bone marrow: implications for gene therapy. Nature medicine 2, 1329-1337. Lasham, A., Samuel, W., Cao, H., Patel, R., Mehta, R., Stern, J. L., Reid, G., Woolley, A. G., Miller, L. D., Black, M. A., et al. (2012). YB-1, the E2F pathway, and regulation of tumor cell growth. Journal of the National Cancer Institute 104, 133-146. Law, J. H., Li, Y., To, K., Wang, M., Astanehe, A., Lambie, K., Dhillon, J., Jones, S. J., Gleave, M. E., Eaves, C. J., and Dunn, S. E. (2010). Molecular decoy to the Y-box binding protein-1 suppresses the growth of breast and prostate cancer cells whilst sparing normal cell viability. PloS one 5. Lee, G. Y., Kenny, P. A., Lee, E. H., and Bissell, M. J. (2007). Three-dimensional culture models of normal and malignant breast epithelial cells. Nat Methods 4, 359-365. Lee, H., Trainer, A. H., Friedman, L. S., Thistlethwaite, F. C., Evans, M. J., Ponder, B. A., and Venkitaraman, A. R. (1999). Mitotic checkpoint inactivation fosters transformation in cells lacking the breast cancer susceptibility gene, Brca2. Molecular cell 4, 1-10. Lee, J. H., and Nan, A. (2012). Combination drug delivery approaches in metastatic breast cancer. Journal of drug delivery 2012, 915375. Lee, T. I., Jenner, R. G., Boyer, L. A., Guenther, M. G., Levine, S. S., Kumar, R. M., Chevalier, B., Johnstone, S. E., Cole, M. F., Isono, K., et al. (2006). Control of developmental regulators by Polycomb in human embryonic stem cells. Cell 125, 301313. Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1997). Genetic instability in colorectal cancers. Nature 386, 623-627. Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1998). Genetic instabilities in human cancers. Nature 396, 643-649. 201  Lessard, J., and Sauvageau, G. (2003). Bmi-1 determines the proliferative capacity of normal and leukaemic stem cells. Nature 423, 255-260. Leung, C. T., and Brugge, J. S. (2012). Outgrowth of single oncogene-expressing cells from suppressive epithelial environments. Nature 482, 410-413. Lewis, E. B. (1978). A gene complex controlling segmentation in Drosophila. Nature 276, 565-570. Li, R., and Murray, A. W. (1991). Feedback control of mitosis in budding yeast. Cell 66, 519-531. Li, X., Lewis, M. T., Huang, J., Gutierrez, C., Osborne, C. K., Wu, M. F., Hilsenbeck, S. G., Pavlick, A., Zhang, X., Chamness, G. C., et al. (2008). Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. Journal of the National Cancer Institute 100, 672-679. Liang, J., Zubovitz, J., Petrocelli, T., Kotchetkov, R., Connor, M. K., Han, K., Lee, J. H., Ciarallo, S., Catzavelos, C., Beniston, R., et al. (2002). PKB/Akt phosphorylates p27, impairs nuclear import of p27 and opposes p27-mediated G1 arrest. Nature medicine 8, 1153-1160. Lim, E., Vaillant, F., Wu, D., Forrest, N. C., Pal, B., Hart, A. H., Asselin-Labat, M. L., Gyorki, D. E., Ward, T., Partanen, A., et al. (2009). Aberrant luminal progenitors as the candidate target population for basal tumor development in BRCA1 mutation carriers. Nature medicine 15, 907-913. Lingle, W. L., Barrett, S. L., Negron, V. C., D'Assoro, A. B., Boeneman, K., Liu, W., Whitehead, C. M., Reynolds, C., and Salisbury, J. L. (2002). Centrosome amplification drives chromosomal instability in breast tumor development. Proceedings of the National Academy of Sciences of the United States of America 99, 1978-1983. Lingle, W. L., Lutz, W. H., Ingle, J. N., Maihle, N. J., and Salisbury, J. L. (1998). Centrosome hypertrophy in human breast tumors: implications for genomic stability and cell polarity. Proceedings of the National Academy of Sciences of the United States of America 95, 2950-2955. Liu, S., Dontu, G., Mantle, I. D., Patel, S., Ahn, N. S., Jackson, K. W., Suri, P., and Wicha, M. S. (2006). Hedgehog signaling and Bmi-1 regulate self-renewal of normal and malignant human mammary stem cells. Cancer research 66, 6063-6071. Loeb, L. A. (1991). Mutator phenotype may be required for multistage carcinogenesis. Cancer research 51, 3075-3079. Loeb, L. A. (2011). Human cancers express mutator phenotypes: origin, consequences and targeting. Nature reviews Cancer 11, 450-457.  202  Louderbough, J. M., Brown, J. A., Nagle, R. B., and Schroeder, J. A. (2011). CD44 Promotes Epithelial Mammary Gland Development and Exhibits Altered Localization during Cancer Progression. Genes & cancer 2, 771-781. Lu, Z. H., Books, J. T., and Ley, T. J. (2005). YB-1 is important for late-stage embryonic development, optimal cellular stress responses, and the prevention of premature senescence. Molecular and cellular biology 25, 4625-4637. Luger, K., Mader, A. W., Richmond, R. K., Sargent, D. F., and Richmond, T. J. (1997). Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389, 251260. Luscher, B. (2001). Function and regulation of the transcription factors of the Myc/Max/Mad network. Gene 277, 1-14. Lutz, M., Wempe, F., Bahr, I., Zopf, D., and von Melchner, H. (2006). Proteasomal degradation of the multifunctional regulator YB-1 is mediated by an F-Box protein induced during programmed cell death. FEBS letters 580, 3921-3930. Ma, X. J., Salunga, R., Tuggle, J. T., Gaudet, J., Enright, E., McQuary, P., Payette, T., Pistone, M., Stecker, K., Zhang, B. M., et al. (2003). Gene expression profiles of human breast cancer progression. Proceedings of the National Academy of Sciences of the United States of America 100, 5974-5979. Maciejczyk, A., Szelachowska, J., Ekiert, M., Matkowski, R., Halon, A., Lage, H., and Surowiak, P. (2012). Elevated nuclear YB1 expression is associated with poor survival of patients with early breast cancer. Anticancer Res 32, 3177-3184. Mahadevan, L. C., Willis, A. C., and Barratt, M. J. (1991). Rapid histone H3 phosphorylation in response to growth factors, phorbol esters, okadaic acid, and protein synthesis inhibitors. Cell 65, 775-783. Maida, Y., Kyo, S., Kanaya, T., Wang, Z., Yatabe, N., Tanaka, M., Nakamura, M., Ohmichi, M., Gotoh, N., Murakami, S., and Inoue, M. (2002). Direct activation of telomerase by EGF through Ets-mediated transactivation of TERT via MAP kinase signaling pathway. Oncogene 21, 4071-4079. Malakhova, M., Kurinov, I., Liu, K., Zheng, D., D'Angelo, I., Shim, J. H., Steinman, V., Bode, A. M., and Dong, Z. (2009). Structural diversity of the active N-terminal kinase domain of p90 ribosomal S6 kinase 2. PloS one 4, e8044. Maliepaard, M., Scheffer, G. L., Faneyte, I. F., van Gastelen, M. A., Pijnenborg, A. C., Schinkel, A. H., van De Vijver, M. J., Scheper, R. J., and Schellens, J. H. (2001). Subcellular localization and distribution of the breast cancer resistance protein transporter in normal human tissues. Cancer research 61, 3458-3464. Malumbres, M., and Barbacid, M. (2009). Cell cycle, CDKs and cancer: a changing paradigm. Nature reviews Cancer 9, 153-166. 203  Mamo, A., Cavallone, L., Tuzmen, S., Chabot, C., Ferrario, C., Hassan, S., Edgren, H., Kallioniemi, O., Aleynikova, O., Przybytkowski, E., et al. (2012). An integrated genomic approach identifies ARID1A as a candidate tumor-suppressor gene in breast cancer. Oncogene 31, 2090-2100. Mani, S. A., Guo, W., Liao, M. J., Eaton, E. N., Ayyanan, A., Zhou, A. Y., Brooks, M., Reinhard, F., Zhang, C. C., Shipitsin, M., et al. (2008). The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133, 704-715. Marangoni, E., Lecomte, N., Durand, L., de Pinieux, G., Decaudin, D., Chomienne, C., Smadja-Joffe, F., and Poupon, M. F. (2009). CD44 targeting reduces tumour growth and prevents post-chemotherapy relapse of human breast cancers xenografts. British journal of cancer 100, 918-922. Marx, J. (2002). Debate surges over the origins of genomic defects in cancer. Science 297, 544-546. Matheu, A., Pantoja, C., Efeyan, A., Criado, L. M., Martin-Caballero, J., Flores, J. M., Klatt, P., and Serrano, M. (2004). Increased gene dosage of Ink4a/Arf results in cancer resistance and normal aging. Genes & development 18, 2736-2746. Maxwell, C. A., Keats, J. J., Crainie, M., Sun, X., Yen, T., Shibuya, E., Hendzel, M., Chan, G., and Pilarski, L. M. (2003). RHAMM is a centrosomal protein that interacts with dynein and maintains spindle pole stability. Molecular biology of the cell 14, 22622276. McCabe, M. T., Ott, H. M., Ganji, G., Korenchuk, S., Thompson, C., Van Aller, G. S., Liu, Y., Graves, A. P., Iii, A. D., Diaz, E., et al. (2012). EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2-activating mutations. Nature. McConnell, B. V., Koto, K., and Gutierrez-Hartmann, A. (2011). Nuclear and cytoplasmic LIMK1 enhances human breast cancer progression. Molecular cancer 10, 75. McDermott, K. M., Zhang, J., Holst, C. R., Kozakiewicz, B. K., Singla, V., and Tlsty, T. D. (2006). p16(INK4a) prevents centrosome dysfunction and genomic instability in primary cells. PLoS biology 4, e51. Meraldi, P., Lukas, J., Fry, A. M., Bartek, J., and Nigg, E. A. (1999). Centrosome duplication in mammalian somatic cells requires E2F and Cdk2-cyclin A. Nature cell biology 1, 88-93. Meyer, M. J., Fleming, J. M., Lin, A. F., Hussnain, S. A., Ginsburg, E., and Vonderhaar, B. K. (2010). CD44posCD49fhiCD133/2hi defines xenograft-initiating cells in estrogen receptor-negative breast cancer. Cancer research 70, 4624-4633. Meyer-Ficca, M. L., Meyer, R. G., Kaiser, H., Brack, A. R., Kandolf, R., and Kupper, J. H. (2004). Comparative analysis of inducible expression systems in transient transfection studies. Analytical biochemistry 334, 9-19. 204  Miller, K., Wang, M., Gralow, J., Dickler, M., Cobleigh, M., Perez, E. A., Shenkier, T., Cella, D., and Davidson, N. E. (2007). Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. The New England journal of medicine 357, 26662676. Mills, A. A. (2010). Throwing the cancer switch: reciprocal roles of polycomb and trithorax proteins. Nature reviews Cancer 10, 669-682. Miwa, A., Higuchi, T., and Kobayashi, S. (2006). Expression and polysome association of YB-1 in various tissues at different stages in the lifespan of mice. Biochimica et biophysica acta 1760, 1675-1681. Mogi, M., Yang, J., Lambert, J. F., Colvin, G. A., Shiojima, I., Skurk, C., Summer, R., Fine, A., Quesenberry, P. J., and Walsh, K. (2003). Akt signaling regulates side population cell phenotype via Bcrp1 translocation. The Journal of biological chemistry 278, 39068-39075. Molofsky, A. V., Pardal, R., Iwashita, T., Park, I. K., Clarke, M. F., and Morrison, S. J. (2003). Bmi-1 dependence distinguishes neural stem cell self-renewal from progenitor proliferation. Nature 425, 962-967. Molyneux, G., Geyer, F. C., Magnay, F. A., McCarthy, A., Kendrick, H., Natrajan, R., Mackay, A., Grigoriadis, A., Tutt, A., Ashworth, A., et al. (2010). BRCA1 basal-like breast cancers originate from luminal epithelial progenitors and not from basal stem cells. Cell stem cell 7, 403-417. Montagna, C., Andrechek, E. R., Padilla-Nash, H., Muller, W. J., and Ried, T. (2002). Centrosome abnormalities, recurring deletions of chromosome 4, and genomic amplification of HER2/neu define mouse mammary gland adenocarcinomas induced by mutant HER2/neu. Oncogene 21, 890-898. Montgomery, N., Hill, A., McFarlane, S., Neisen, J., O'Grady, A., Conlon, S., Jirstrom, K., Kay, E. W., and Waugh, D. J. (2012). CD44 enhances invasion of basal-like breast cancer cells by upregulating serine protease and collagen-degrading enzymatic expression and activity. Breast cancer research : BCR 14, R84. Moody, C. A., and Laimins, L. A. (2010). Human papillomavirus oncoproteins: pathways to transformation. Nature reviews Cancer 10, 550-560. Morris, L., Allen, K. E., and La Thangue, N. B. (2000). Regulation of E2F transcription by cyclin E-Cdk2 kinase mediated through p300/CBP co-activators. Nature cell biology 2, 232-239. Mudryj, M., Hiebert, S. W., and Nevins, J. R. (1990). A role for the adenovirus inducible E2F transcription factor in a proliferation dependent signal transduction pathway. The EMBO journal 9, 2179-2184.  205  Muller, W. J., Sinn, E., Pattengale, P. K., Wallace, R., and Leder, P. (1988). Single-step induction of mammary adenocarcinoma in transgenic mice bearing the activated c-neu oncogene. Cell 54, 105-115. Muraoka, R. S., Lenferink, A. E., Law, B., Hamilton, E., Brantley, D. M., Roebuck, L. R., and Arteaga, C. L. (2002). ErbB2/Neu-induced, cyclin D1-dependent transformation is accelerated in p27-haploinsufficient mammary epithelial cells but impaired in p27-null cells. Molecular and cellular biology 22, 2204-2219. Musacchio, A., and Salmon, E. D. (2007). The spindle-assembly checkpoint in space and time. Nature reviews Molecular cell biology 8, 379-393. Muthuswamy, S. K., Li, D., Lelievre, S., Bissell, M. J., and Brugge, J. S. (2001). ErbB2, but not ErbB1, reinitiates proliferation and induces luminal repopulation in epithelial acini. Nature cell biology 3, 785-792. Nagahara, H., Ezhevsky, S. A., Vocero-Akbani, A. M., Kaldis, P., Solomon, M. J., and Dowdy, S. F. (1999). Transforming growth factor beta targeted inactivation of cyclin E:cyclin-dependent kinase 2 (Cdk2) complexes by inhibition of Cdk2 activating kinase activity. Proceedings of the National Academy of Sciences of the United States of America 96, 14961-14966. Nagata, Y., Lan, K. H., Zhou, X., Tan, M., Esteva, F. J., Sahin, A. A., Klos, K. S., Li, P., Monia, B. P., Nguyen, N. T., et al. (2004). PTEN activation contributes to tumor inhibition by trastuzumab, and loss of PTEN predicts trastuzumab resistance in patients. Cancer cell 6, 117-127. Nakanishi, T., Chumsri, S., Khakpour, N., Brodie, A. H., Leyland-Jones, B., Hamburger, A. W., Ross, D. D., and Burger, A. M. (2010). Side-population cells in luminal-type breast cancer have tumour-initiating cell properties, and are regulated by HER2 expression and signalling. British journal of cancer 102, 815-826. Nan, X., Ng, H. H., Johnson, C. A., Laherty, C. D., Turner, B. M., Eisenman, R. N., and Bird, A. (1998). Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature 393, 386-389. Navin, N., Kendall, J., Troge, J., Andrews, P., Rodgers, L., McIndoo, J., Cook, K., Stepansky, A., Levy, D., Esposito, D., et al. (2011). Tumour evolution inferred by singlecell sequencing. Nature 472, 90-94. Navin, N., Krasnitz, A., Rodgers, L., Cook, K., Meth, J., Kendall, J., Riggs, M., Eberling, Y., Troge, J., Grubor, V., et al. (2010). Inferring tumor progression from genomic heterogeneity. Genome research 20, 68-80. Negrini, S., Gorgoulis, V. G., and Halazonetis, T. D. (2010). Genomic instability--an evolving hallmark of cancer. Nature reviews Molecular cell biology 11, 220-228.  206  Nguyen, L. V., Vanner, R., Dirks, P., and Eaves, C. J. (2012). Cancer stem cells: an evolving concept. Nature reviews Cancer 12, 133-143. Nicholson, J. M., and Duesberg, P. (2009). On the karyotypic origin and evolution of cancer cells. Cancer genetics and cytogenetics 194, 96-110. Nielsen, T. O., Hsu, F. D., Jensen, K., Cheang, M., Karaca, G., Hu, Z., HernandezBoussard, T., Livasy, C., Cowan, D., Dressler, L., et al. (2004). Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clinical cancer research : an official journal of the American Association for Cancer Research 10, 5367-5374. Nigg, E. A. (2002). Centrosome aberrations: cause or consequence of cancer progression? Nature reviews Cancer 2, 815-825. Nigg, E. A. (2006). Origins and consequences of centrosome aberrations in human cancers. International journal of cancer Journal international du cancer 119, 2717-2723. Nousiainen, M., Sillje, H. H., Sauer, G., Nigg, E. A., and Korner, R. (2006). Phosphoproteome analysis of the human mitotic spindle. Proceedings of the National Academy of Sciences of the United States of America 103, 5391-5396. Nowell, P. C. (1976). The clonal evolution of tumor cell populations. Science 194, 23-28. Nuovo, G. J., Plaia, T. W., Belinsky, S. A., Baylin, S. B., and Herman, J. G. (1999). In situ detection of the hypermethylation-induced inactivation of the p16 gene as an early event in oncogenesis. Proceedings of the National Academy of Sciences of the United States of America 96, 12754-12759. Nurse, P., Thuriaux, P., and Nasmyth, K. (1976). Genetic control of the cell division cycle in the fission yeast Schizosaccharomyces pombe. Molecular & general genetics : MGG 146, 167-178. O'Connell, P., Pekkel, V., Fuqua, S. A., Osborne, C. K., Clark, G. M., and Allred, D. C. (1998). Analysis of loss of heterozygosity in 399 premalignant breast lesions at 15 genetic loci. Journal of the National Cancer Institute 90, 697-703. O'Shaughnessy, J., Osborne, C., Pippen, J. E., Yoffe, M., Patt, D., Rocha, C., Koo, I. C., Sherman, B. M., and Bradley, C. (2011). Iniparib plus chemotherapy in metastatic triplenegative breast cancer. The New England journal of medicine 364, 205-214. Ohtani, K., DeGregori, J., and Nevins, J. R. (1995). Regulation of the cyclin E gene by transcription factor E2F1. Proceedings of the National Academy of Sciences of the United States of America 92, 12146-12150. Oka, H., Shiozaki, H., Kobayashi, K., Inoue, M., Tahara, H., Kobayashi, T., Takatsuka, Y., Matsuyoshi, N., Hirano, S., Takeichi, M., and et al. (1993). Expression of E-cadherin  207  cell adhesion molecules in human breast cancer tissues and its relationship to metastasis. Cancer research 53, 1696-1701. Okita, K., Ichisaka, T., and Yamanaka, S. (2007). Generation of germline-competent induced pluripotent stem cells. Nature 448, 313-317. Okuda, M., Horn, H. F., Tarapore, P., Tokuyama, Y., Smulian, A. G., Chan, P. K., Knudsen, E. S., Hofmann, I. A., Snyder, J. D., Bove, K. E., and Fukasawa, K. (2000). Nucleophosmin/B23 is a target of CDK2/cyclin E in centrosome duplication. Cell 103, 127-140. Onitilo, A. A., Engel, J. M., Greenlee, R. T., and Mukesh, B. N. (2009). Breast cancer subtypes based on ER/PR and Her2 expression: comparison of clinicopathologic features and survival. Clinical medicine & research 7, 4-13. Osborne, C. K. (1998). Tamoxifen in the treatment of breast cancer. The New England journal of medicine 339, 1609-1618. Ozdag, H., Teschendorff, A. E., Ahmed, A. A., Hyland, S. J., Blenkiron, C., Bobrow, L., Veerakumarasivam, A., Burtt, G., Subkhankulova, T., Arends, M. J., et al. (2006). Differential expression of selected histone modifier genes in human solid cancers. BMC genomics 7, 90. Ozer, J., Faber, M., Chalkley, R., and Sealy, L. (1990). Isolation and characterization of a cDNA clone for the CCAAT transcription factor EFIA reveals a novel structural motif. The Journal of biological chemistry 265, 22143-22152. Palacios, J., Honrado, E., Osorio, A., Cazorla, A., Sarrio, D., Barroso, A., Rodriguez, S., Cigudosa, J. C., Diez, O., Alonso, C., et al. (2005). Phenotypic characterization of BRCA1 and BRCA2 tumors based in a tissue microarray study with 37 immunohistochemical markers. Breast cancer research and treatment 90, 5-14. Park, I. K., Qian, D., Kiel, M., Becker, M. W., Pihalja, M., Weissman, I. L., Morrison, S. J., and Clarke, M. F. (2003). Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature 423, 302-305. Park, S., Shimizu, C., Shimoyama, T., Takeda, M., Ando, M., Kohno, T., Katsumata, N., Kang, Y. K., Nishio, K., and Fujiwara, Y. (2006). Gene expression profiling of ATPbinding cassette (ABC) transporters as a predictor of the pathologic response to neoadjuvant chemotherapy in breast cancer patients. Breast cancer research and treatment 99, 9-17. Park, S. Y., Lee, H. E., Li, H., Shipitsin, M., Gelman, R., and Polyak, K. (2010). Heterogeneity for stem cell-related markers according to tumor subtype and histologic stage in breast cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 16, 876-887.  208  Percy, M. J., Myrie, K. A., Neeley, C. K., Azim, J. N., Ethier, S. P., and Petty, E. M. (2000). Expression and mutational analyses of the human MAD2L1 gene in breast cancer cells. Genes, chromosomes & cancer 29, 356-362. Perou, C. M., Sorlie, T., Eisen, M. B., van de Rijn, M., Jeffrey, S. S., Rees, C. A., Pollack, J. R., Ross, D. T., Johnsen, H., Akslen, L. A., et al. (2000). Molecular portraits of human breast tumours. Nature 406, 747-752. Petersen, O. W., Ronnov-Jessen, L., Howlett, A. R., and Bissell, M. J. (1992). Interaction with basement membrane serves to rapidly distinguish growth and differentiation pattern of normal and malignant human breast epithelial cells. Proceedings of the National Academy of Sciences of the United States of America 89, 9064-9068. Peterson, C. L., and Laniel, M. A. (2004). Histones and histone modifications. Current biology : CB 14, R546-551. Pham, P. V., Phan, N. L., Nguyen, N. T., Truong, N. H., Duong, T. T., Le, D. V., Truong, K. D., and Phan, N. K. (2011). Differentiation of breast cancer stem cells by knockdown of CD44: promising differentiation therapy. Journal of translational medicine 9, 209. Phillips, T. M., McBride, W. H., and Pajonk, F. (2006). The response of CD24(/low)/CD44+ breast cancer-initiating cells to radiation. Journal of the National Cancer Institute 98, 1777-1785. Poizat, C., Sartorelli, V., Chung, G., Kloner, R. A., and Kedes, L. (2000). Proteasomemediated degradation of the coactivator p300 impairs cardiac transcription. Molecular and cellular biology 20, 8643-8654. Polyak, K., and Weinberg, R. A. (2009). Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nature reviews Cancer 9, 265-273. Ponti, D., Costa, A., Zaffaroni, N., Pratesi, G., Petrangolini, G., Coradini, D., Pilotti, S., Pierotti, M. A., and Daidone, M. G. (2005). Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer research 65, 5506-5511. Prat, A., Parker, J. S., Karginova, O., Fan, C., Livasy, C., Herschkowitz, J. I., He, X., and Perou, C. M. (2010). Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast cancer. Breast cancer research : BCR 12, R68. Prowse, A. H., Schultz, D. C., Guo, S., Vanderveer, L., Dangel, J., Bove, B., Cairns, P., Daly, M., and Godwin, A. K. (2003). Identification of a splice acceptor site mutation in p16INK4A/p14ARF within a breast cancer, melanoma, neurofibroma prone kindred. Journal of medical genetics 40, e102. Rabindran, S. K., Ross, D. D., Doyle, L. A., Yang, W., and Greenberger, L. M. (2000). Fumitremorgin C reverses multidrug resistance in cells transfected with the breast cancer resistance protein. Cancer research 60, 47-50. 209  Radisky, D. C., Santisteban, M., Berman, H. K., Gauthier, M. L., Frost, M. H., Reynolds, C. A., Vierkant, R. A., Pankratz, V. S., Visscher, D. W., Tlsty, T. D., and Hartmann, L. C. (2011). p16(INK4a) expression and breast cancer risk in women with atypical hyperplasia. Cancer prevention research 4, 1953-1960. Rao, P. S., Satelli, A., Moridani, M., Jenkins, M., and Rao, U. S. (2012). Luteolin induces apoptosis in multidrug resistant cancer cells without affecting the drug transporter function: involvement of cell line-specific apoptotic mechanisms. International journal of cancer Journal international du cancer 130, 2703-2714. Raouf, A., Brown, L., Vrcelj, N., To, K., Kwok, W., Huntsman, D., and Eaves, C. J. (2005). Genomic instability of human mammary epithelial cells overexpressing a truncated form of EMSY. Journal of the National Cancer Institute 97, 1302-1306. Raouf, A., Zhao, Y., To, K., Stingl, J., Delaney, A., Barbara, M., Iscove, N., Jones, S., McKinney, S., Emerman, J., et al. (2008). Transcriptome analysis of the normal human mammary cell commitment and differentiation process. Cell stem cell 3, 109-118. Reya, T., Morrison, S. J., Clarke, M. F., and Weissman, I. L. (2001). Stem cells, cancer, and cancer stem cells. Nature 414, 105-111. Reynolds, P. A., Sigaroudinia, M., Zardo, G., Wilson, M. B., Benton, G. M., Miller, C. J., Hong, C., Fridlyand, J., Costello, J. F., and Tlsty, T. D. (2006). Tumor suppressor p16INK4A regulates polycomb-mediated DNA hypermethylation in human mammary epithelial cells. The Journal of biological chemistry 281, 24790-24802. Rhodes, D. R., Kalyana-Sundaram, S., Mahavisno, V., Barrette, T. R., Ghosh, D., and Chinnaiyan, A. M. (2005). Mining for regulatory programs in the cancer transcriptome. Nature genetics 37, 579-583. Rinn, J. L., Kertesz, M., Wang, J. K., Squazzo, S. L., Xu, X., Brugmann, S. A., Goodnough, L. H., Helms, J. A., Farnham, P. J., Segal, E., and Chang, H. Y. (2007). Functional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell 129, 1311-1323. Roach, J. C., Glusman, G., Smit, A. F., Huff, C. D., Hubley, R., Shannon, P. T., Rowen, L., Pant, K. P., Goodman, N., Bamshad, M., et al. (2010). Analysis of genetic inheritance in a family quartet by whole-genome sequencing. Science 328, 636-639. Roelfsema, J. H., White, S. J., Ariyurek, Y., Bartholdi, D., Niedrist, D., Papadia, F., Bacino, C. A., den Dunnen, J. T., van Ommen, G. J., Breuning, M. H., et al. (2005). Genetic heterogeneity in Rubinstein-Taybi syndrome: mutations in both the CBP and EP300 genes cause disease. American journal of human genetics 76, 572-580. Romanov, S. R., Kozakiewicz, B. K., Holst, C. R., Stampfer, M. R., Haupt, L. M., and Tlsty, T. D. (2001). Normal human mammary epithelial cells spontaneously escape senescence and acquire genomic changes. Nature 409, 633-637.  210  Rosenthal, S. I., Depowski, P. L., Sheehan, C. E., and Ross, J. S. (2002). Comparison of HER-2/neu oncogene amplification detected by fluorescence in situ hybridization in lobular and ductal breast cancer. Applied immunohistochemistry & molecular morphology : AIMM / official publication of the Society for Applied Immunohistochemistry 10, 40-46. Ross, D. D., and Nakanishi, T. (2010). Impact of breast cancer resistance protein on cancer treatment outcomes. Methods Mol Biol 596, 251-290. Roy, R., Chun, J., and Powell, S. N. (2012). BRCA1 and BRCA2: different roles in a common pathway of genome protection. Nature reviews Cancer 12, 68-78. Ruff, P., Vorobiof, D. A., Jordaan, J. P., Demetriou, G. S., Moodley, S. D., Nosworthy, A. L., Werner, I. D., Raats, J., and Burgess, L. J. (2009). A randomized, placebocontrolled, double-blind phase 2 study of docetaxel compared to docetaxel plus zosuquidar (LY335979) in women with metastatic or locally recurrent breast cancer who have received one prior chemotherapy regimen. Cancer chemotherapy and pharmacology 64, 763-768. Rush, J., Moritz, A., Lee, K. A., Guo, A., Goss, V. L., Spek, E. J., Zhang, H., Zha, X. M., Polakiewicz, R. D., and Comb, M. J. (2005). Immunoaffinity profiling of tyrosine phosphorylation in cancer cells. Nature biotechnology 23, 94-101. Sakura, H., Maekawa, T., Imamoto, F., Yasuda, K., and Ishii, S. (1988). Two human genes isolated by a novel method encode DNA-binding proteins containing a common region of homology. Gene 73, 499-507. Sandoval, J., and Esteller, M. (2012). Cancer epigenomics: beyond genomics. Current opinion in genetics & development 22, 50-55. Sansom, O. J., Maddison, K., and Clarke, A. R. (2007). Mechanisms of disease: methylbinding domain proteins as potential therapeutic targets in cancer. Nature clinical practice Oncology 4, 305-315. Santamaria, D., Barriere, C., Cerqueira, A., Hunt, S., Tardy, C., Newton, K., Caceres, J. F., Dubus, P., Malumbres, M., and Barbacid, M. (2007). Cdk1 is sufficient to drive the mammalian cell cycle. Nature 448, 811-815. Santner, S. J., Dawson, P. J., Tait, L., Soule, H. D., Eliason, J., Mohamed, A. N., Wolman, S. R., Heppner, G. H., and Miller, F. R. (2001). Malignant MCF10CA1 cell lines derived from premalignant human breast epithelial MCF10AT cells. Breast cancer research and treatment 65, 101-110. Sapkota, G. P., Cummings, L., Newell, F. S., Armstrong, C., Bain, J., Frodin, M., Grauert, M., Hoffmann, M., Schnapp, G., Steegmaier, M., et al. (2007). BI-D1870 is a specific inhibitor of the p90 RSK (ribosomal S6 kinase) isoforms in vitro and in vivo. Biochem J 401, 29-38.  211  Sassone-Corsi, P., Mizzen, C. A., Cheung, P., Crosio, C., Monaco, L., Jacquot, S., Hanauer, A., and Allis, C. D. (1999). Requirement of Rsk-2 for epidermal growth factoractivated phosphorylation of histone H3. Science 285, 886-891. Sauer, G., Korner, R., Hanisch, A., Ries, A., Nigg, E. A., and Sillje, H. H. (2005). Proteome analysis of the human mitotic spindle. Molecular & cellular proteomics : MCP 4, 35-43. Scharenberg, C. W., Harkey, M. A., and Torok-Storb, B. (2002). The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 99, 507-512. Schiltz, R. L., Mizzen, C. A., Vassilev, A., Cook, R. G., Allis, C. D., and Nakatani, Y. (1999). Overlapping but distinct patterns of histone acetylation by the human coactivators p300 and PCAF within nucleosomal substrates. The Journal of biological chemistry 274, 1189-1192. Scully, R. (2000). Role of BRCA gene dysfunction in breast and ovarian cancer predisposition. Breast cancer research : BCR 2, 324-330. Segatto, O., King, C. R., Pierce, J. H., Di Fiore, P. P., and Aaronson, S. A. (1988). Different structural alterations upregulate in vitro tyrosine kinase activity and transforming potency of the erbB-2 gene. Molecular and cellular biology 8, 5570-5574. Seger, Y. R., Garcia-Cao, M., Piccinin, S., Cunsolo, C. L., Doglioni, C., Blasco, M. A., Hannon, G. J., and Maestro, R. (2002). Transformation of normal human cells in the absence of telomerase activation. Cancer cell 2, 401-413. Selever, J., Gu, G., Lewis, M. T., Beyer, A., Herynk, M. H., Covington, K. R., Tsimelzon, A., Dontu, G., Provost, P., Di Pietro, A., et al. (2011). Dicer-mediated upregulation of BCRP confers tamoxifen resistance in human breast cancer cells. Clinical cancer research : an official journal of the American Association for Cancer Research 17, 6510-6521. Semple, S. C., Akinc, A., Chen, J., Sandhu, A. P., Mui, B. L., Cho, C. K., Sah, D. W., Stebbing, D., Crosley, E. J., Yaworski, E., et al. (2010). Rational design of cationic lipids for siRNA delivery. Nature biotechnology 28, 172-176. Shackleton, M., Vaillant, F., Simpson, K. J., Stingl, J., Smyth, G. K., Asselin-Labat, M. L., Wu, L., Lindeman, G. J., and Visvader, J. E. (2006). Generation of a functional mammary gland from a single stem cell. Nature 439, 84-88. Shadeo, A., and Lam, W. L. (2006). Comprehensive copy number profiles of breast cancer cell model genomes. Breast cancer research : BCR 8, R9. Shah, S. P., Morin, R. D., Khattra, J., Prentice, L., Pugh, T., Burleigh, A., Delaney, A., Gelmon, K., Guliany, R., Senz, J., et al. (2009). Mutational evolution in a lobular breast tumour profiled at single nucleotide resolution. Nature 461, 809-813. 212  Shaughnessy, M., and Wistow, G. (1992). Absence of MHC gene expression in lens and cloning of dbpB/YB-1, a DNA-binding protein expressed in mouse lens. Current eye research 11, 175-181. Shi, Q., and King, R. W. (2005). Chromosome nondisjunction yields tetraploid rather than aneuploid cells in human cell lines. Nature 437, 1038-1042. Shibahara, K., Uchiumi, T., Fukuda, T., Kura, S., Tominaga, Y., Maehara, Y., Kohno, K., Nakabeppu, Y., Tsuzuki, T., and Kuwano, M. (2004). Targeted disruption of one allele of the Y-box binding protein-1 (YB-1) gene in mouse embryonic stem cells and increased sensitivity to cisplatin and mitomycin C. Cancer science 95, 348-353. Shibao, K., Takano, H., Nakayama, Y., Okazaki, K., Nagata, N., Izumi, H., Uchiumi, T., Kuwano, M., Kohno, K., and Itoh, H. (1999). Enhanced coexpression of YB-1 and DNA topoisomerase II alpha genes in human colorectal carcinomas. International journal of cancer Journal international du cancer 83, 732-737. Shimono, Y., Zabala, M., Cho, R. W., Lobo, N., Dalerba, P., Qian, D., Diehn, M., Liu, H., Panula, S. P., Chiao, E., et al. (2009). Downregulation of miRNA-200c links breast cancer stem cells with normal stem cells. Cell 138, 592-603. Shiota, M., Izumi, H., Onitsuka, T., Miyamoto, N., Kashiwagi, E., Kidani, A., Yokomizo, A., Naito, S., and Kohno, K. (2008). Twist promotes tumor cell growth through YB-1 expression. Cancer research 68, 98-105. Shiota, M., Yokomizo, A., Tada, Y., Uchiumi, T., Inokuchi, J., Tatsugami, K., Kuroiwa, K., Yamamoto, K., Seki, N., and Naito, S. (2010). P300/CBP-associated factor regulates Y-box binding protein-1 expression and promotes cancer cell growth, cancer invasion and drug resistance. Cancer science 101, 1797-1806. Shipitsin, M., Campbell, L. L., Argani, P., Weremowicz, S., Bloushtain-Qimron, N., Yao, J., Nikolskaya, T., Serebryiskaya, T., Beroukhim, R., Hu, M., et al. (2007). Molecular definition of breast tumor heterogeneity. Cancer cell 11, 259-273. Shtutman, M., Zhurinsky, J., Simcha, I., Albanese, C., D'Amico, M., Pestell, R., and BenZe'ev, A. (1999). The cyclin D1 gene is a target of the beta-catenin/LEF-1 pathway. Proceedings of the National Academy of Sciences of the United States of America 96, 5522-5527. Sieber, O. M., Heinimann, K., and Tomlinson, I. P. (2003). Genomic instability--the engine of tumorigenesis? Nature reviews Cancer 3, 701-708. Signoretti, S., Di Marcotullio, L., Richardson, A., Ramaswamy, S., Isaac, B., Rue, M., Monti, F., Loda, M., and Pagano, M. (2002). Oncogenic role of the ubiquitin ligase subunit Skp2 in human breast cancer. The Journal of clinical investigation 110, 633-641.  213  Silk, A. D., Holland, A. J., and Cleveland, D. W. (2009). Requirements for NuMA in maintenance and establishment of mammalian spindle poles. The Journal of cell biology 184, 677-690. Singh, B., and Lucci, A. (2002). Role of cyclooxygenase-2 in breast cancer. The Journal of surgical research 108, 173-179. Singh, R. R., Kunkalla, K., Qu, C., Schlette, E., Neelapu, S. S., Samaniego, F., and Vega, F. (2011). ABCG2 is a direct transcriptional target of hedgehog signaling and involved in stroma-induced drug tolerance in diffuse large B-cell lymphoma. Oncogene 30, 48744886. Sjoblom, T., Jones, S., Wood, L. D., Parsons, D. W., Lin, J., Barber, T. D., Mandelker, D., Leary, R. J., Ptak, J., Silliman, N., et al. (2006). The consensus coding sequences of human breast and colorectal cancers. Science 314, 268-274. Slamon, D. J., Clark, G. M., Wong, S. G., Levin, W. J., Ullrich, A., and McGuire, W. L. (1987). Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235, 177-182. Smalley, M., and Ashworth, A. (2003). Stem cells and breast cancer: A field in transit. Nature reviews Cancer 3, 832-844. Solomon, D. A., Kim, T., Diaz-Martinez, L. A., Fair, J., Elkahloun, A. G., Harris, B. T., Toretsky, J. A., Rosenberg, S. A., Shukla, N., Ladanyi, M., et al. (2011). Mutational inactivation of STAG2 causes aneuploidy in human cancer. Science 333, 1039-1043. Song, J., Teplova, M., Ishibe-Murakami, S., and Patel, D. J. (2012). Structure-based mechanistic insights into DNMT1-mediated maintenance DNA methylation. Science 335, 709-712. Sorlie, T., Perou, C. M., Tibshirani, R., Aas, T., Geisler, S., Johnsen, H., Hastie, T., Eisen, M. B., van de Rijn, M., Jeffrey, S. S., et al. (2001). Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proceedings of the National Academy of Sciences of the United States of America 98, 10869-10874. Sorlie, T., Tibshirani, R., Parker, J., Hastie, T., Marron, J. S., Nobel, A., Deng, S., Johnsen, H., Pesich, R., Geisler, S., et al. (2003). Repeated observation of breast tumor subtypes in independent gene expression data sets. Proceedings of the National Academy of Sciences of the United States of America 100, 8418-8423. Sorokin, A. V., Selyutina, A. A., Skabkin, M. A., Guryanov, S. G., Nazimov, I. V., Richard, C., Th'ng, J., Yau, J., Sorensen, P. H., Ovchinnikov, L. P., and Evdokimova, V. (2005). Proteasome-mediated cleavage of the Y-box-binding protein 1 is linked to DNAdamage stress response. The EMBO journal 24, 3602-3612.  214  Sotillo, R., Hernando, E., Diaz-Rodriguez, E., Teruya-Feldstein, J., Cordon-Cardo, C., Lowe, S. W., and Benezra, R. (2007). Mad2 overexpression promotes aneuploidy and tumorigenesis in mice. Cancer cell 11, 9-23. Speth, P. A., van Hoesel, Q. G., and Haanen, C. (1988). Clinical pharmacokinetics of doxorubicin. Clinical pharmacokinetics 15, 15-31. Spruck, C. H., Won, K. A., and Reed, S. I. (1999). Deregulated cyclin E induces chromosome instability. Nature 401, 297-300. Squazzo, S. L., O'Geen, H., Komashko, V. M., Krig, S. R., Jin, V. X., Jang, S. W., Margueron, R., Reinberg, D., Green, R., and Farnham, P. J. (2006). Suz12 binds to silenced regions of the genome in a cell-type-specific manner. Genome research 16, 890900. Stephens, P. J., Greenman, C. D., Fu, B., Yang, F., Bignell, G. R., Mudie, L. J., Pleasance, E. D., Lau, K. W., Beare, D., Stebbings, L. A., et al. (2011). Massive genomic rearrangement acquired in a single catastrophic event during cancer development. Cell 144, 27-40. Stephens, P. J., Tarpey, P. S., Davies, H., Van Loo, P., Greenman, C., Wedge, D. C., NikZainal, S., Martin, S., Varela, I., Bignell, G. R., et al. (2012). The landscape of cancer genes and mutational processes in breast cancer. Nature 486, 400-404. Stingl, J., Eaves, C. J., Kuusk, U., and Emerman, J. T. (1998). Phenotypic and functional characterization in vitro of a multipotent epithelial cell present in the normal adult human breast. Differentiation; research in biological diversity 63, 201-213. Stingl, J., Eaves, C. J., Zandieh, I., and Emerman, J. T. (2001). Characterization of bipotent mammary epithelial progenitor cells in normal adult human breast tissue. Breast cancer research and treatment 67, 93-109. Stingl, J., Eirew, P., Ricketson, I., Shackleton, M., Vaillant, F., Choi, D., Li, H. I., and Eaves, C. J. (2006a). Purification and unique properties of mammary epithelial stem cells. Nature 439, 993-997. Stingl, J., Raouf, A., Eirew, P., and Eaves, C. J. (2006b). Deciphering the mammary epithelial cell hierarchy. Cell cycle 5, 1519-1522. Stock, J. K., Giadrossi, S., Casanova, M., Brookes, E., Vidal, M., Koseki, H., Brockdorff, N., Fisher, A. G., and Pombo, A. (2007). Ring1-mediated ubiquitination of H2A restrains poised RNA polymerase II at bivalent genes in mouse ES cells. Nature cell biology 9, 1428-1435. Stratford, A. L., and Dunn, S. E. (2011). The promise and challenges of targeting RSK for the treatment of cancer. Expert Opin Ther Targets 15, 1-4.  215  Stratford, A. L., Fry, C. J., Desilets, C., Davies, A. H., Cho, Y. Y., Li, Y., Dong, Z., Berquin, I. M., Roux, P. P., and Dunn, S. E. (2008). Y-box binding protein-1 serine 102 is a downstream target of p90 ribosomal S6 kinase in basal-like breast cancer cells. Breast cancer research : BCR 10, R99. Stratford, A. L., Habibi, G., Astanehe, A., Jiang, H., Hu, K., Park, E., Shadeo, A., Buys, T. P., Lam, W., Pugh, T., et al. (2007). Epidermal growth factor receptor (EGFR) is transcriptionally induced by the Y-box binding protein-1 (YB-1) and can be inhibited with Iressa in basal-like breast cancer, providing a potential target for therapy. Breast cancer research : BCR 9, R61. Stratford, A. L., Reipas, K., Hu, K., Fotovati, A., Brough, R., Frankum, J., Takhar, M., Watson, P., Ashworth, A., Lord, C. J., et al. (2012). Targeting p90 Ribosomal S6 Kinase (RSK) Eliminates Tumor-Initiating Cells by Inactivating Y-Box Binding Protein-1 (YB1) in Triple-Negative Breast Cancers. Stem cells. Stratford, A. L., Reipas, K., Maxwell, C., and Dunn, S. E. (2010). Targeting tumourinitiating cells to improve the cure rates for triple-negative breast cancer. Expert Rev Mol Med 12, e22. Stratton, M. R., Campbell, P. J., and Futreal, P. A. (2009). The cancer genome. Nature 458, 719-724. Strohmaier, H., Spruck, C. H., Kaiser, P., Won, K. A., Sangfelt, O., and Reed, S. I. (2001). Human F-box protein hCdc4 targets cyclin E for proteolysis and is mutated in a breast cancer cell line. Nature 413, 316-322. Sumi, T., Hashigasako, A., Matsumoto, K., and Nakamura, T. (2006). Different activity regulation and subcellular localization of LIMK1 and LIMK2 during cell cycle transition. Experimental cell research 312, 1021-1030. Sumi, T., Matsumoto, K., and Nakamura, T. (2002). Mitosis-dependent phosphorylation and activation of LIM-kinase 1. Biochemical and biophysical research communications 290, 1315-1320. Sun, Q., Chen, X., Ma, J., Peng, H., Wang, F., Zha, X., Wang, Y., Jing, Y., Yang, H., Chen, R., et al. (2011). Mammalian target of rapamycin up-regulation of pyruvate kinase isoenzyme type M2 is critical for aerobic glycolysis and tumor growth. Proceedings of the National Academy of Sciences of the United States of America 108, 4129-4134. Sutherland, B. W., Kucab, J., Wu, J., Lee, C., Cheang, M. C., Yorida, E., Turbin, D., Dedhar, S., Nelson, C., Pollak, M., et al. (2005). Akt phosphorylates the Y-box binding protein 1 at Ser102 located in the cold shock domain and affects the anchorageindependent growth of breast cancer cells. Oncogene 24, 4281-4292. Suva, M. L., Riggi, N., Janiszewska, M., Radovanovic, I., Provero, P., Stehle, J. C., Baumer, K., Le Bitoux, M. A., Marino, D., Cironi, L., et al. (2009). EZH2 is essential for glioblastoma cancer stem cell maintenance. Cancer research 69, 9211-9218. 216  Swaisland, H. C., Smith, R. P., Laight, A., Kerr, D. J., Ranson, M., Wilder-Smith, C. H., and Duvauchelle, T. (2005). Single-dose clinical pharmacokinetic studies of gefitinib. Clinical pharmacokinetics 44, 1165-1177. Szakacs, G., Paterson, J. K., Ludwig, J. A., Booth-Genthe, C., and Gottesman, M. M. (2006). Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5, 219-234. Szerlong, H. J., Prenni, J. E., Nyborg, J. K., and Hansen, J. C. (2010). Activatordependent p300 acetylation of chromatin in vitro: enhancement of transcription by disruption of repressive nucleosome-nucleosome interactions. The Journal of biological chemistry 285, 31954-31964. Takahashi, K., Tanabe, K., Ohnuki, M., Narita, M., Ichisaka, T., Tomoda, K., and Yamanaka, S. (2007). Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861-872. Takahashi, K., and Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663-676. Takai, N., Hamanaka, R., Yoshimatsu, J., and Miyakawa, I. (2005). Polo-like kinases (Plks) and cancer. Oncogene 24, 287-291. Tate, C. R., Rhodes, L. V., Segar, H. C., Driver, J. L., Pounder, F. N., Burow, M. E., and Collins-Burow, B. M. (2012). Targeting triple-negative breast cancer cells with the histone deacetylase inhibitor panobinostat. Breast cancer research : BCR 14, R79. Thakur, A., Sun, Y., Bollig, A., Wu, J., Biliran, H., Banerjee, S., Sarkar, F. H., and Liao, D. J. (2008). Anti-invasive and antimetastatic activities of ribosomal protein S6 kinase 4 in breast cancer cells. Clinical cancer research : an official journal of the American Association for Cancer Research 14, 4427-4436. Tlsty, T. D., Romanov, S. R., Kozakiewicz, B. K., Holst, C. R., Haupt, L. M., and Crawford, Y. G. (2001). Loss of chromosomal integrity in human mammary epithelial cells subsequent to escape from senescence. Journal of mammary gland biology and neoplasia 6, 235-243. To, K., Fotovati, A., Reipas, K. M., Law, J. H., Hu, K., Wang, J., Astanehe, A., Davies, A. H., Lee, L., Stratford, A. L., et al. (2010). Y-box binding protein-1 induces the expression of CD44 and CD49f leading to enhanced self-renewal, mammosphere growth, and drug resistance. Cancer research 70, 2840-2851. To, K. K., Zhan, Z., and Bates, S. E. (2006). Aberrant promoter methylation of the ABCG2 gene in renal carcinoma. Molecular and cellular biology 26, 8572-8585. Toyoshima-Morimoto, F., Taniguchi, E., and Nishida, E. (2002). Plk1 promotes nuclear translocation of human Cdc25C during prophase. EMBO reports 3, 341-348.  217  Trock, B. J., Leonessa, F., and Clarke, R. (1997). Multidrug resistance in breast cancer: a meta-analysis of MDR1/gp170 expression and its possible functional significance. Journal of the National Cancer Institute 89, 917-931. Tsukasaki, K., Miller, C. W., Greenspun, E., Eshaghian, S., Kawabata, H., Fujimoto, T., Tomonaga, M., Sawyers, C., Said, J. W., and Koeffler, H. P. (2001). Mutations in the mitotic check point gene, MAD1L1, in human cancers. Oncogene 20, 3301-3305. Turner, N., Tutt, A., and Ashworth, A. (2004). Hallmarks of 'BRCAness' in sporadic cancers. Nature reviews Cancer 4, 814-819. Uchiumi, T., Fotovati, A., Sasaguri, T., Shibahara, K., Shimada, T., Fukuda, T., Nakamura, T., Izumi, H., Tsuzuki, T., Kuwano, M., and Kohno, K. (2006). YB-1 is important for an early stage embryonic development: neural tube formation and cell proliferation. The Journal of biological chemistry 281, 40440-40449. Uramoto, H., Izumi, H., Ise, T., Tada, M., Uchiumi, T., Kuwano, M., Yasumoto, K., Funa, K., and Kohno, K. (2002). p73 Interacts with c-Myc to regulate Y-box-binding protein-1 expression. The Journal of biological chemistry 277, 31694-31702. van 't Veer, L. J., Dai, H., van de Vijver, M. J., He, Y. D., Hart, A. A., Mao, M., Peterse, H. L., van der Kooy, K., Marton, M. J., Witteveen, A. T., et al. (2002). Gene expression profiling predicts clinical outcome of breast cancer. Nature 415, 530-536. Van Keymeulen, A., Rocha, A. S., Ousset, M., Beck, B., Bouvencourt, G., Rock, J., Sharma, N., Dekoninck, S., and Blanpain, C. (2011). Distinct stem cells contribute to mammary gland development and maintenance. Nature 479, 189-193. Van't Veer, L. J., and Weigelt, B. (2003). Road map to metastasis. Nature medicine 9, 999-1000. Vargo-Gogola, T., and Rosen, J. M. (2007). Modelling breast cancer: one size does not fit all. Nature reviews Cancer 7, 659-672. Verma, S., Miles, D., Gianni, L., Krop, I. E., Welslau, M., Baselga, J., Pegram, M., Oh, D. Y., Dieras, V., Guardino, E., et al. (2012). Trastuzumab emtansine for HER2-positive advanced breast cancer. The New England journal of medicine 367, 1783-1791. Visvader, J. E. (2009). Keeping abreast of the mammary epithelial hierarchy and breast tumorigenesis. Genes & development 23, 2563-2577. Visvader, J. E., and Lindeman, G. J. (2008). Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nature reviews Cancer 8, 755-768. Visvader, J. E., and Lindeman, G. J. (2012). Cancer stem cells: current status and evolving complexities. Cell stem cell 10, 717-728.  218  Vleugel, M. M., Shvarts, D., van der Wall, E., and van Diest, P. J. (2006). p300 and p53 levels determine activation of HIF-1 downstream targets in invasive breast cancer. Human pathology 37, 1085-1092. Wang, N. P., To, H., Lee, W. H., and Lee, E. Y. (1993). Tumor suppressor activity of RB and p53 genes in human breast carcinoma cells. Oncogene 8, 279-288. Wang, T. C., Cardiff, R. D., Zukerberg, L., Lees, E., Arnold, A., and Schmidt, E. V. (1994). Mammary hyperplasia and carcinoma in MMTV-cyclin D1 transgenic mice. Nature 369, 669-671. Wang, X., Zhou, Y. X., Qiao, W., Tominaga, Y., Ouchi, M., Ouchi, T., and Deng, C. X. (2006). Overexpression of aurora kinase A in mouse mammary epithelium induces genetic instability preceding mammary tumor formation. Oncogene 25, 7148-7158. Warburg, O. (1956). On respiratory impairment in cancer cells. Science 124, 269-270. Watnick, R. S., Cheng, Y. N., Rangarajan, A., Ince, T. A., and Weinberg, R. A. (2003). Ras modulates Myc activity to repress thrombospondin-1 expression and increase tumor angiogenesis. Cancer cell 3, 219-231. Weaver, B. A., Silk, A. D., Montagna, C., Verdier-Pinard, P., and Cleveland, D. W. (2007). Aneuploidy acts both oncogenically and as a tumor suppressor. Cancer cell 11, 25-36. Weigelt, B., Peterse, J. L., and van 't Veer, L. J. (2005). Breast cancer metastasis: markers and models. Nature reviews Cancer 5, 591-602. Weinberg, R. A. (1995). The molecular basis of oncogenes and tumor suppressor genes. Annals of the New York Academy of Sciences 758, 331-338. Wilson, B. G., and Roberts, C. W. (2011). SWI/SNF nucleosome remodellers and cancer. Nature reviews Cancer 11, 481-492. Winey, M., and Byers, B. (1993). Assembly and functions of the spindle pole body in budding yeast. Trends in genetics : TIG 9, 300-304. Wong, D. J., Liu, H., Ridky, T. W., Cassarino, D., Segal, E., and Chang, H. Y. (2008). Module map of stem cell genes guides creation of epithelial cancer stem cells. Cell stem cell 2, 333-344. Woo, M. M., Salamanca, C. M., Minor, A., and Auersperg, N. (2007). An improved assay to quantitate the invasiveness of cells in modified Boyden chambers. In Vitro Cell Dev Biol Anim 43, 7-9. Wood, L. D., Parsons, D. W., Jones, S., Lin, J., Sjoblom, T., Leary, R. J., Shen, D., Boca, S. M., Barber, T., Ptak, J., et al. (2007). The genomic landscapes of human breast and colorectal cancers. Science 318, 1108-1113.  219  Wright, M. H., Calcagno, A. M., Salcido, C. D., Carlson, M. D., Ambudkar, S. V., and Varticovski, L. (2008). Brca1 breast tumors contain distinct CD44+/CD24- and CD133+ cells with cancer stem cell characteristics. Breast cancer research : BCR 10, R10. Wu, J., Lee, C., Yokom, D., Jiang, H., Cheang, M. C., Yorida, E., Turbin, D., Berquin, I. M., Mertens, P. R., Iftner, T., et al. (2006). Disruption of the Y-box binding protein-1 results in suppression of the epidermal growth factor receptor and HER-2. Cancer research 66, 4872-4879. Wu, J., Stratford, A. L., Astanehe, A., and Dunn, S. E. (2007). YB-1 is a transcription/translation factor that orchestrates the oncogenome by hardwiring signal transduction to gene expression. Translational Oncogenomics 2, 49-65. Xia, W., Nagase, S., Montia, A. G., Kalachikov, S. M., Keniry, M., Su, T., Memeo, L., Hibshoosh, H., and Parsons, R. (2008). BAF180 is a critical regulator of p21 induction and a tumor suppressor mutated in breast cancer. Cancer research 68, 1667-1674. Yahata, H., Kobayashi, H., Kamura, T., Amada, S., Hirakawa, T., Kohno, K., Kuwano, M., and Nakano, H. (2002). Increased nuclear localization of transcription factor YB-1 in acquired cisplatin-resistant ovarian cancer. J Cancer Res Clin Oncol 128, 621-626. Yang, C., Ionescu-Tiba, V., Burns, K., Gadd, M., Zukerberg, L., Louis, D. N., Sgroi, D., and Schmidt, E. V. (2004a). The role of the cyclin D1-dependent kinases in ErbB2mediated breast cancer. The American journal of pathology 164, 1031-1038. Yang, N., Higuchi, O., Ohashi, K., Nagata, K., Wada, A., Kangawa, K., Nishida, E., and Mizuno, K. (1998). Cofilin phosphorylation by LIM-kinase 1 and its role in Racmediated actin reorganization. Nature 393, 809-812. Yang, X., Yu, K., Hao, Y., Li, D. M., Stewart, R., Insogna, K. L., and Xu, T. (2004b). LATS1 tumour suppressor affects cytokinesis by inhibiting LIMK1. Nature cell biology 6, 609-617. Yang, X. J. (2004). The diverse superfamily of lysine acetyltransferases and their roles in leukemia and other diseases. Nucleic acids research 32, 959-976. Yao, T. P., Oh, S. P., Fuchs, M., Zhou, N. D., Ch'ng, L. E., Newsome, D., Bronson, R. T., Li, E., Livingston, D. M., and Eckner, R. (1998). Gene dosage-dependent embryonic development and proliferation defects in mice lacking the transcriptional integrator p300. Cell 93, 361-372. Yokoo, T., Toyoshima, H., Miura, M., Wang, Y., Iida, K. T., Suzuki, H., Sone, H., Shimano, H., Gotoda, T., Nishimori, S., et al. (2003). p57Kip2 regulates actin dynamics by binding and translocating LIM-kinase 1 to the nucleus. The Journal of biological chemistry 278, 52919-52923.  220  Yoshiji, H., Gomez, D. E., Shibuya, M., and Thorgeirsson, U. P. (1996). Expression of vascular endothelial growth factor, its receptor, and other angiogenic factors in human breast cancer. Cancer research 56, 2013-2016. Yoshioka, K., Foletta, V., Bernard, O., and Itoh, K. (2003). A role for LIM kinase in cancer invasion. Proceedings of the National Academy of Sciences of the United States of America 100, 7247-7252. You, J. S., and Jones, P. A. (2012). Cancer genetics and epigenetics: two sides of the same coin? Cancer cell 22, 9-20. Yu, K. R., Yang, S. R., Jung, J. W., Kim, H., Ko, K., Han, D. W., Park, S. B., Choi, S. W., Kang, S. K., Scholer, H., and Kang, K. S. (2012). CD49f enhances multipotency and maintains stemness through the direct regulation of OCT4 and SOX2. Stem cells 30, 876887. Yu, Q., Geng, Y., and Sicinski, P. (2001). Specific protection against breast cancers by cyclin D1 ablation. Nature 411, 1017-1021. Yu, Y. N., Yip, G. W., Tan, P. H., Thike, A. A., Matsumoto, K., Tsujimoto, M., and Bay, B. H. (2010). Y-box binding protein 1 is up-regulated in proliferative breast cancer and its inhibition deregulates the cell cycle. International journal of oncology 37, 483-492. Yuan, B., Xu, Y., Woo, J. H., Wang, Y., Bae, Y. K., Yoon, D. S., Wersto, R. P., Tully, E., Wilsbach, K., and Gabrielson, E. (2006). Increased expression of mitotic checkpoint genes in breast cancer cells with chromosomal instability. Clinical cancer research : an official journal of the American Association for Cancer Research 12, 405-410. Yuan, J., Lv, H., Peng, B., Wang, C., Yu, Y., and He, Z. (2009). Role of BCRP as a biomarker for predicting resistance to 5-fluorouracil in breast cancer. Cancer chemotherapy and pharmacology 63, 1103-1110. Zhang, N., Ge, G., Meyer, R., Sethi, S., Basu, D., Pradhan, S., Zhao, Y. J., Li, X. N., Cai, W. W., El-Naggar, A. K., et al. (2008). Overexpression of Separase induces aneuploidy and mammary tumorigenesis. Proceedings of the National Academy of Sciences of the United States of America 105, 13033-13038. Zhao, H., Langerod, A., Ji, Y., Nowels, K. W., Nesland, J. M., Tibshirani, R., Bukholm, I. K., Karesen, R., Botstein, D., Borresen-Dale, A. L., and Jeffrey, S. S. (2004). Different gene expression patterns in invasive lobular and ductal carcinomas of the breast. Molecular biology of the cell 15, 2523-2536. Zhao, J. J., Gjoerup, O. V., Subramanian, R. R., Cheng, Y., Chen, W., Roberts, T. M., and Hahn, W. C. (2003). Human mammary epithelial cell transformation through the activation of phosphatidylinositol 3-kinase. Cancer cell 3, 483-495.  221  Zhong, Q., Boyer, T. G., Chen, P. L., and Lee, W. H. (2002). Deficient nonhomologous end-joining activity in cell-free extracts from Brca1-null fibroblasts. Cancer research 62, 3966-3970. Zhong, X., and Jin, Y. (2009). Critical roles of coactivator p300 in mouse embryonic stem cell differentiation and Nanog expression. The Journal of biological chemistry 284, 9168-9175. Zhou, S., Morris, J. J., Barnes, Y., Lan, L., Schuetz, J. D., and Sorrentino, B. P. (2002). Bcrp1 gene expression is required for normal numbers of side population stem cells in mice, and confers relative protection to mitoxantrone in hematopoietic cells in vivo. Proceedings of the National Academy of Sciences of the United States of America 99, 12339-12344. Zhou, S., Schuetz, J. D., Bunting, K. D., Colapietro, A. M., Sampath, J., Morris, J. J., Lagutina, I., Grosveld, G. C., Osawa, M., Nakauchi, H., and Sorrentino, B. P. (2001). The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nature medicine 7, 1028-1034. Zimonjic, D., Brooks, M. W., Popescu, N., Weinberg, R. A., and Hahn, W. C. (2001). Derivation of human tumor cells in vitro without widespread genomic instability. Cancer research 61, 8838-8844.  222  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items