UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of podocalyxin in breast cancer progression and metastasis Snyder, Kimberly Ashley 2014

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

Item Metadata

Download

Media
24-ubc_2014_spring_snyder_kimberly.pdf [ 2.54MB ]
Metadata
JSON: 24-1.0072153.json
JSON-LD: 24-1.0072153-ld.json
RDF/XML (Pretty): 24-1.0072153-rdf.xml
RDF/JSON: 24-1.0072153-rdf.json
Turtle: 24-1.0072153-turtle.txt
N-Triples: 24-1.0072153-rdf-ntriples.txt
Original Record: 24-1.0072153-source.json
Full Text
24-1.0072153-fulltext.txt
Citation
24-1.0072153.ris

Full Text

      THE ROLE OF PODOCALYXIN IN BREAST CANCER PROGRESSION AND METASTASIS   by  KIMBERLY ASHLEY SNYDER  B.Sc., University of Victoria, 2010     A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES   (Experimental Medicine)      THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)      February 2014  ? Kimberly Ashley Snyder, 2014   ii ABSTRACT  The molecular heterogeneity of breast cancers makes them very challenging to diagnose and treat. While many breast tumors are curable surgically, others have an increased tendency to relapse or metastasize. Podocalyxin (gene name PODXL) is a CD34-related sialomucin that is important in regulating cell adhesion, migration, and polarity of hematopoietic progenitors and vascular endothelia. Previously, podocalyxin expression in primary breast tumor cells has been correlatively linked to poor patient survival. In addition, overexpression of podocalyxin in MCF-7 luminal breast cancer cells results in an overall increase in aggressive morphological features including: apical bulging, microvillus formation, and disrupted integrin targeting to the basolateral surface. To determine whether podocalyxin has a definitive role in breast tumor progression, I used shRNA-based vectors to silence expression of PODXL in a basal-like breast cancer cell-line, MDA.MB-231, which normally expresses high levels of endogenous podocalyxin, forms poorly polarized monolayers and tumorspheres in vitro and rapidly forms metastatic tumors in vivo. I found that podocalyxin impairs adhesion to extracellular matrices (ECM) and is critical for the formation of tumorspheres in vitro. In addition, I found that podocalyxin expression is critical for both primary tumor development and the formation of distant metastases in the lung, liver, and bone marrow in xenografted mice. These findings were corroborated by use of a mouse mammary tumor cell line (4T1) in a syngeneic model of metastatic tumor progression.  Furthermore, in collaboration with the Centre for Drug Research and Development (UBC), I evaluated the ability of candidate therapeutic antibodies to podocalyxin to block the growth and metastasis of established tumors. One candidate antibody, known as anti-PODO, was found to significantly inhibit primary tumorigenesis in a pre-clinical mouse model. In summary, in this thesis I demonstrate for the first time that podocalyxin plays a causal role in promoting the growth of solid tumors and enhancing metastasis of tumor cells to distant organs. These findings have the potential to improve treatments for breast cancer patients by providing a highly specific and well-tolerated adjuvant therapy for metastatic disease.    iii PREFACE  I have performed all experiments and generated the data and figures presented in this thesis with the following exceptions: ? shRNA-mediated knockdown of podocalyxin in MDA.MB-231 cells was performed in the lab of Dr. Shaun McColl at the University of Adelaide, Australia by Michelle Turvey.  ? Overexpression of podocalyxin in MCF-7 cells was performed by Drs. Julie Nielsen and Marcia Graves in Dr. Kelly McNagny and Dr. Calvin Roskelley?s labs at the University of British Columbia. ? Knockdown of podocalyxin in the mouse mammary tumor cell line, 4T1, was performed by Jane Cipollone in the lab of Dr. Calvin Roskelley ? Anti-podocalyxin (anti-PODO) and anti-OVA control antibodies were generated by John Babcook, Bradley Hedberg and colleagues at the Centre for Drug Research and Development, Vancouver, B.C.  ? Experimental planning and data interpretation performed with the help of Dr. Michael Hughes and Dr. Kelly McNagny. ? This work was approved by the Animal Care Committee under Certificate Numbers: A11-0035 and A12-0205                iv TABLE OF CONTENTS  ABSTRACT  ............................................................................................................. ii PREFACE  ............................................................................................................... iii TABLE OF CONTENTS  ......................................................................................... iv LIST OF TABLES  ................................................................................................. vii LIST OF FIGURES  ............................................................................................... viii LIST OF ABBREVIATIONS  .................................................................................... x ACKNOWLEDGEMENTS  .................................................................................... xiii DEDICATION  ....................................................................................................... xiv CHAPTER 1: INTRODUCTION  ............................................................................... 1 1.1 Overview of breast cancer  ................................................................................. 1 1.1.1 Classification of breast cancer  ................................................................. 1 1.2 The Metastatic cascade  ..................................................................................... 5 1.3 Breast cancer therapeutics  ................................................................................ 8  1.4 The Cancer stem cell hypothesis  ....................................................................... 8 1.4.1 Cancer stem cell markers  ........................................................................ 9 1.5 Overview of podocalyxin, a member of the CD34 family of sialomucins  ......... 11 1.5.1 Podocalyxin structure and function during normal development  ............ 11 1.5.2 Podocalyxin mediates adhesion  ............................................................ 14 1.5.3 Intracellular binding partners of podocalyxin  .......................................... 15 1.5.4 Podocalyxin and cancer  ......................................................................... 16 1.5.5 Upregulation of podocalyxin expression and breast cancer  .................. 16 1.5.6 Ectopic expression of podocalyxin in breast cancer cell lines  ............... 17 1.6 Hypothesis  ....................................................................................................... 18 1.6.1 Objectives  .............................................................................................. 19 CHAPTER 2: MATERIALS AND METHODS  ....................................................... 20 2.1 Cell culture  ....................................................................................................... 20 2.1.1 Cell line passaging  ................................................................................. 20 2.2 Transfections and generation of stable cell lines  ............................................. 20 2.2.1 Silencing PODXL expression in MDA.MB-231 breast tumor cell line  .... 20 2.2.2 Silencing mouse Podxl in 4T1 mouse mammary tumor cell line  ............ 21  v 2.2.3 Forced expression of Podxl in MCF-7 breast tumor cells  ...................... 21 2.3 Western blotting  ............................................................................................... 22 2.4 Quantitive RT-PCR  .......................................................................................... 23 2.4.1 RNA isolation and cDNA synthesis  ........................................................ 23 2.4.2 Quantitive Reverse-Transcriptase PCR  ................................................. 23 2.5 Adhesion Assay  ............................................................................................... 24 2.6 MTS proliferation assay  ................................................................................... 25 2.7 Anchorage-independent growth assays  .......................................................... 25 2.7.1 Tumorsphere assay  ............................................................................... 25 2.7.2 Colony forming assay  ............................................................................ 26 2.8 Animals  ............................................................................................................ 26 2.9 Xenograft models of breast tumourigenesis  .................................................... 27 2.9.1 Subcutaneous xenograft model of tumor development  .......................... 27 2.9.2 Competitive subcutaneous xenograft model of tumor development and  surgical resection  ............................................................................................ 27 2.9.3 Competitive orthotopic xenograft model of tumor development  ............. 28 2.9.4 Competitive experimental xenograft model of lung metastasis  .............. 28 2.10 Microscopic imaging of metastatic sites  ........................................................ 29 2.11 Preparation of single cell suspensions from tumors and metastatic sites  ..... 29 2.12 Flow cytometry  ............................................................................................... 30 2.13 Tumor histological analysis  ........................................................................... 30 2.14 Experimental syngeneic model of lung metastasis  ........................................ 30 2.15 Luciferase enzymatic assay  .......................................................................... 30 2.16 Therapeutic antibody production  ................................................................... 31 2.16.1 In vivo screening to identify candidate therapeutic antibodies  .................... 32 2.16.2 Pre-clinical mouse model to test anti-podocalyxin antibody efficacy  ... 32 2.17 Statistical analysis  ......................................................................................... 33 CHAPTER 3: RESULTS  ....................................................................................... 34 3.1 Rationale  .......................................................................................................... 34  3.2 Podocalyxin in highly expressed in aggressive breast tumor cell lines  ........... 35 3.3 Podocalyxin can be efficiently knocked down in MDA.MB-231 cells using   vi shRNA-mediated lentiviral infection  ....................................................................... 37 3.4 Podocalyxin expression has no effect on monolayer proliferation, but critical  for anchorage-independent growth  ........................................................................ 40 3.5 Loss of podocalyxin expression increases adhesion to different extracellular  matrices  ................................................................................................................. 45 3.6 Podocalyxin expression promotes primary tumor formation  ............................ 47 3.7 Podocalyxin expression promotes metastasis  ................................................. 48 3.8 Podocalyxin expression in MDA.MB-231 cells is critical for tumor regrowth  and lung metastasis  ............................................................................................... 51 3.9 Podocalyxin expression enhances metastatic potential at late stages of  metastasis, but not initial seeding of tumor cells in the lung  .................................. 54 3.10 Podocalyxin expression is also critical in a syngeneic model of metastasis .. 59 3.11 Development of therapeutic antibody against podocalyxin  ............................ 61 CHAPTER 4: DISCUSSION AND CONCLUSION  ................................................ 68 4.1 Summary  ......................................................................................................... 68 4.2 Podocalyxin functioning as an anti-adhesive protein in tumorigenesis  ............ 69 4.3 Podocalyxin facilitates anchorage-independent growth of tumor cells  ............ 72 4.4 Podocalyxin as a mediator of survival under hypoxic conditions  ..................... 74 4.5 Podocalyxin promotes primary tumor development  ......................................... 75 4.6 Podocalyxin promotes tumor invasion and migration  ...................................... 76 4.7 Podocalyxin and the tumor microenvironment  ................................................. 77 4.8 Future Directions  ............................................................................................. 79 REFERENCES  ...................................................................................................... 82 APPENDICES  ....................................................................................................... 93 Appendix A. Primary tumor development of MCF-7Podxl cells in NSG mice  ........... 93         vii LIST OF TABLES  Table 1-1 The features of the five intrinsic subtypes of breast cancer  .................... 3 Table 2-1 Thermocycler program for qPCR  .......................................................... 24                                            viii LIST OF FIGURES  Figure 1-1 The Metastatic cascade  ........................................................................ 6 Figure 1-2 Schematic of podocalyxin/NHERF/ezrin protein complex  ................... 13 Figure 3-1 Breast cancer cell lines express levels of podocalyxin, which  positively correlate with cell line aggressiveness  ................................................... 36 Figure 3-2 Podocalyxin expression can be efficiently knocked down in  MDA.MB-231 cells  ................................................................................................. 38 Figure 3-3 Podocalyxin expression returns after 14 days when removed from  antibiotic selection in vivo  ...................................................................................... 39 Figure 3-4 Proliferation is not effected by podocalyxin expression in  MDA.MB-231 cells in monolayer culture  ................................................................ 42 Figure 3-5 Podocalyxin expressing MDA.MB-231 cells form tumorspheres more  efficiently when grown in suspension culture  ......................................................... 43 Figure 3-6 Podocalyxin expression has no effect on self-renewal by serial  passage of tumorsphere cultures  .......................................................................... 44 Figure 3-7 Podocalyxin facilitates a reduction in cell-ECM adhesion  ................... 46 Figure 3-8 Podocalyxin expression promotes primary tumor formation, local  invasion, and lung metastasis  ............................................................................... 49 Figure 3-9 Podocalyxin expression is critical for local primary tumor recurrence  and micro-metastasis in MDA.MB-231 cells  .......................................................... 52 Figure 3-10 Podocalyxin expression is critical for orthotopic tumor development  in the mouse mammary fat pad  ............................................................................. 53 Figure 3-11 Podocalyxin expression has no effect on initial tumor seeding in the  lung using a model of experimental metastasis  ..................................................... 56 Figure 3-12 Podocalyxin expression increases the metastatic burden of the  lungs, liver, and bone at late stages of metastasis  ................................................ 57 Figure 3-13 Podocalyxin expression enhances metastasis of 4T1 mammary  tumor cells to the lung in a syngeneic model of experimental metastasis  ............. 60 Figure 3-14 Pre-treatment of MDA.MB-231 cells with anti-PODO antibody  significantly reduces early primary tumor development in NSG mice  .................... 64 Figure 3-15 Treatment with anti-PODO results in inhibition of primary tumor   ix development and a reduction in micrometastasis to the lung  ................................ 65 Figure 3-16 Treatment with anti-PODO antibody results in smaller tumors and  reduced metastasis to the lung  .............................................................................. 67 Figure 4-1 Model of podocalyxin functioning as an anti-adhesive  ........................ 70 Figure A-1 Podocalyxin overexpressing MCF-7 cells possess no primary tumor  growth advantage in NSG mice  ............................................................................. 93     x LIST OF ABBREVIATIONS  AF AlexaFluor? ANN axillary node negative APC allophycocyanin ATP adenosine triphosphate bp base pair BSA bovine serum albumin CAF carcinoma associated fibroblast cDNA complement DNA CFTR cystic fibrosis transmembrane conductance regulator CHO Chinese hamster ovary  ColI collagen I CXCL12 CXC-chemokine ligand 12 CXCR4 CXC-chemokine receptor 4 DMEM Dulbecco?s minimal essential medium DNA deoxyribonucleic acid DTHL Asp-Thr-His-Leu ECM extracellular matrix EDTA ethylenediaminetetraacetic acid EGFR epidermal growth factor receptor EMT epithelial-to-mesenchymal transition EPC endothelial progenitor cell ER estrogen receptor ErbB2 avian erythroblastosis oncogene B ERM ezrin-radixin-moesin ESC embryonic stem cell EtOH ethanol F-12 Ham?s F12 media FACS fluorescence activated cell sorting FAK focal adhesion kinase FBS fetal bovine serum G418 Geneticin? GFP green fluorescence protein GLUT glucose-transporter H&E hematoxylin and eosin HA hyaluronan HBSS Hank?s buffered salt solution HEPES 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid sodium salt HER-2 human epidermal growth factor receptor-2 hESCs human embryonic stem cells HIF hypoxia-inducible factor HR hormone receptor IDC invasive ductal carcinoma Ig immunoglobulin  xi ILC invasive lobular carcinoma i.p intraperitoneal i.v intravenous IVIS in vivo imaging system kDa kiloDalton KO knock out LAR luciferase assay reagent Lin lineage LVI lymphatic invasion mAB monoclonal antibody MAPK mitogen-activated protein kinase MCAM melanoma cell adhesion molecule MRCRB mouse red cell reduction buffer  MDCK Madin-Darby canine kidney MFI mean fluorescence intensity MMP matrix metalloproteinase Ms mouse MTS 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium MUC mucin n.s. non-significant NaAzide sodium azide NHERF  Na+/H+ exchanger regulatory factor  NSG NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ PAGE polyacrylamide gel electrophoresis PBS phosphate buffered saline PCLP-1 podocalyxin-like-protein 1 PCR polymerase chain reaction PDGFR platelet derived growth factor receptor PDZ PSD-95/Dlg/ZO-1 PEM polymorphic epithelial mucin PFA paraformaldehyde PgR progesterone receptor PI3K phosphoinositide-3-kinase PODXL-KD human podocalyxin knockdown Podxl-KD murine podocalyxin knockdown PVDF  polyvinylidene fluoride Rb rabbit RFP red fluorescent protein RIPA radioimmunoprecipitation assay buffer RNA ribonucleic acid RPMI Roswell Park Memorial Institute medium Scr scrambled Scr-ctrl scrambled control SD standard deviation  xii SDF1 stromal-derived factor 1 SDS sodium dodecyl sulfate SEM standard error of the mean shRNA short-hairpin interfering RNA TBS tris buffered saline TBS-T tris buffered saline + 0.05% Tween-20 TGF-? transforming growth factor beta TIC tumor initiating cell TNBC triple negative breast cancer TNF tumor necrosis factor UTR untranslated region VC vector control WT wildtype ZO-1 zonula occludens ?2AR beta-2 adrenergic receptor                       xiii ACKNOWLEDGEMENTS       I would like to take this opportunity to thank my supervisor Dr. Kelly McNagny for giving me the opportunity to study in his lab and for sharing his extensive experience with me. I would also like to thank the members of my supervisory committee, Dr. Cal Roskelley and Dr. Michael Underhill, for their guidance and helpful advice. I would like to extend thanks to John Babcook, Bradley Hedberg, Jill Brandon and the other members of Podocalyxin-Antibody team for generating the anti-podocalyxin antibodies used in this thesis.       I would especially like to thank Dr. Michael Hughes for recruiting me to the lab, for his constant support, encouragement, and exceptional research advice. I would like to express my sincerest gratitude to Dr. Frann Antignano for teaching me important research skills during my early days as a student and for letting me steal her husband away more often than I should have.       I would also like to thank the current members of the McNagny Lab, Matthew Gold, Bernard Lo, Martin Lopez, Diana Canals, Jessie Cait, and Alissa Cait for their continuous support. I would like to acknowledge the past members of the McNagny lab, Dr. Erin Debruin, Dr. Steve Maltby, Alison Hirukawa and Dr. Jami Bennett for their kindness to me. I owe a special thank you to Les Rollins, Krista Ranta, and Wei Yuan for all of their hard work looking after our mice. I would also like to thank Marcia Graves and Jane Cipollone for their helpful research advice.     I would like to thank Vivian Lam, and the past members of the Krystal Lab for making my first research experience so pleasant. I would especially like to thank Sherie, Mike, Frann, Matt, Prasadani, and Diana for their friendship, for truly making the BRC my home away from home, and for occasionally talking me down from the ledge J?.      Most importantly, I am extremely grateful for the unwavering love and support of my parents, Ric and Karen. I owe you both much more thanks than I can even express.          xiv DEDICATION For my parents, Ric and Karen CHAPTER 1: INTRODUCTION  1.1 Overview of breast cancer Breast cancer was first described by the ancient Egyptians greater than 3,500 years ago as a ?bulging tumor? of the breast for which ?there is no cure?. Since this time, medicine has classified breast cancer over a spectrum of ever-changing paradigms. Hippocrates (circa 460-370 BCE) considered breast cancer as a purely humoral disease supposedly caused by an excess of black bile (melanchole). From the mid-18th century well into the 20th century, breast cancer was considered to be a completely localized disease treated principally by radical surgery. As evidence accumulated showing that a tumor localized to the breast is capable of metastatic spread to distant sites of the body in the mid-20th century, breast cancer, at least in its lethal form, was once again considered a systemic disease (1).   Presently, breast cancer is regarded as an extremely heterogeneous group of tumors, diverse in their behavior, outcome, and therapeutic response (2). It is well known that while many breast tumors are curable by local surgery, others have an increased propensity to relapse or metastasize. Despite improvements in patient survival resulting from advancements in early diagnosis and systemic adjuvant therapy, systemic spread of tumor cells is the primary cause of death from breast cancer (3). Today, breast cancer is the most common female cancer, affecting greater than 1 million women per year worldwide and killing close to 400,000 patients every year (4).  1.1.1 Classification of breast cancer There are two major forms of breast cancer, ?ductal? and ?lobular? carcinomas, which both arise from the terminal duct lobular unit and are characterized based on their histological features (5). Ductal tumors form more glandular structures, whereas lobular tumors form less-cohesive tumors, which tend to invade in single file (6). Ductal and lobular tumors can be further classified as invasive or non-invasive, depending on whether infiltration into the surrounding stroma has occurred (?infiltrating?), or if the neoplasm remains completely localized to the site of origin within the breast (?in situ?) (5).  Invasive ductal carcinomas (IDC) are the most common making up approximately 75% of invasive breast cancers,  2 whereas invasive lobular carcinomas (ILC) make up approximately 15% of invasive breast cancers and are generally more aggressive (7).  Pathological classification of breast cancer based on characteristics such as tumor size, histological grade, and axillary lymph-node involvement correlates with clinical outcome and overall prognosis (8). In addition, breast cancers can also be classified into a number of subgroups based on expression of hormone receptors (HR) (e.g., estrogen receptor (ER) and progesterone receptor (PgR)) as well as ErbB2/Human epidermal growth factor receptor 2 (HER-2). Breast cancer has been divided into five distinct subtypes by both immunohistochemical and gene expression analysis. The five broadly classified subgroups are luminal A, luminal B, HER-2+, basal-like, and normal-breast-like subtypes of breast cancer (9-11) (Table 1-1).  Luminal A and luminal B subtypes of breast cancer are both HR-positive and possess gene expression patterns similar to the luminal epithelial cells of the breast. The luminal A subtype generally has higher levels of ER/PgR expression, whereas luminal B subtype tumors have higher levels of genes associated with proliferation and also may express HER-2. The HER-2+ subtype of breast cancer is characterized by low expression of the HRs and additional copies of the HER-2 gene.  HER-2 amplification often occurs in highly aggressive breast cancers and has become an important biomarker and ?druggable?-target for therapy (12). The basal-like subtype is also known as triple negative breast cancer (TNBC). TNBC is unique from the four subtypes because it lacks the characteristic hormone receptors, ER, PgR, and HER-2 amplification, and thus these tumors do not respond to the targeted adjuvant therapies that are used to treat the HR expressing breast cancers (4, 13). A lack of effective therapies for systemic treatment of TNBC results in a disproportionate number of deaths compared to HR or HER-2 expressing breast cancer subtypes (2).      3 Table 1-1. The features of the five intrinsic subtypes of breast cancer. Adapted from (14).  Subtype (10, 11) Receptor Status Prevalence Treatment Clinical Outcome  Luminal A ? ER/PgR++,  ? HER2 - ~42-59% ? Endocrine therapy  (tamoxifen, aromatase inhibitors).  ? Variable response to chemotherapy.  ? Generally good prognosis. ? Tumor dormancy and late recurrences (15) Luminal B ? ER/PgR+,  ? some HER2+ ~10% ? Endocrine therapy (tamoxifen, aromatase inhibitors).  ? Variable response to chemotherapy  ? Higher grade.  ? Increased proliferation rates.  ? Worse overall prognosis than luminal A.  HER-2  ? ER/PgR-,  ? HER2+  ~15% ? Targeted therapy (trastuzumab, lapatinib) ? Anthracycline-based chemotherapy  ? More likely to be high grade and node positive.  ? Generally poor prognosis.  Basal-like (Triple-negative) ? ER/PgR-  ? HER2- ? cytokeratin 5/6+ ~15% ? No response to endocrine therapy or trastuzumab (Herceptin?)  ? Sensitive to platinum-based chemotherapy and PARP inhibitors   ? Generally poor prognosis (but not uniformly poor) ? BRCA1 mutations are common  ? Common among young and African American women  Normal-breast-like ? ER/PgR+/-  ? HER-2- ? cytokeratin 5/6 and 17 ? >adipose tissue genes ? >non-epithelial genes ? <luminal-epithelial genes ~10% (16) ? Does not respond well to neoadjuvant chemotherapy (17)  ? Intermediate prognosis (16) ? Prognosis better than Basal-like breast cancers (11, 18)        4 The fifth breast cancer subtype is known as the ?normal? subtype of breast cancer. Normal-like breast cancers have a similar gene expression profile to normal breast tissue and lack ER, PgR expression and HER-2 amplification. These ?normal? breast cancers highly express many genes typical of adipose tissue and other non-epithelial cell types; they also express high levels of basal epithelial genes and low levels of luminal epithelial genes (11). It is debated whether the ?normal-like? gene expression profile of this breast cancer subtype may have stemmed from a gene expression analysis artifact caused by contamination from tissue surrounding the tumors (16, 19).    While the current classification system of human breast tumors has been important in providing the guidelines for prognostic and predictive evaluation, there are considerable limitations. For example, there remains a large degree of variation in therapeutic response and clinical outcome within subtypes. In addition, the biological features used to classify the different subtypes provide little insight into the complexities of the molecular pathways and fundamental biology that drives cancer progression (20). The heterogeneity of breast cancer cannot be adequately explained by the clinical parameters or biomarkers routinely used to diagnose and treat patients. A recent publication by Curtis et al, 2012 highlighted the importance of treating breast cancer as a highly heterogenic disease, and identified ten different breast cancer subgroups based on integrated genome and transcriptome analyses of 2,000 breast tumor samples (21). This highly impactful study showed the genetic variability within the five intrinsic subtypes, which had never before been analyzed in such an extensive and large-scale study. Because standard medical practice does not have the capacity to fully examine the genomic profile of all patient tumors, traditional methods and the biological features used to define the intrinsic subtypes (luminal A, luminal B, HER-2, basal-like, and normal-like) will continue to be used to make treatment decisions in the clinic and gain information on prognosis as it is more economically and technologically feasible. However the Curtis study brings us closer towards personalized breast cancer management and provides a framework for studying the underlying biology of the breast cancer subtypes.      5 1.2 The Metastatic cascade  Metastasis is an extremely complex process. A tumor cell must overcome a number of challenging steps, known as the ?metastatic cascade?, before it can become a clinically detectable lesion (Figure 1-1). For metastasis to occur, a tumor cell must first separate from the primary tumor and invade through the extracellular matrix and stromal cells surrounding the tumor. Following local invasion, a tumor cell must intravasate into the blood or lymphatic vessels to enter circulation. Upon entry into circulation, a tumor cell must then evade immune detection, and programmed cell death (anoikis), before arresting in a target organ. Extravasation of the tumor cell into the surrounding tissue must occur to colonize and form micrometastases at the secondary site. In order for a tissue-colonizing tumor cell to progress to a clinically detectable metastatic lesion, the tumor cell must adapt to the foreign microenvironment, while reactivating proliferative pathways, and inducing angiogenesis (22, 23). The metastatic process can be divided into early and late phases; the early phase includes the steps that precede initiation of ectopic growth in the secondary site, whereas the late phase contains all steps subsequent to initiation of growth in the new site (24).         6  Figure 1-1. The Metastatic cascade (23).   Overall, metastasis is an extremely inefficient process. Despite the fact that primary tumors are capable of shedding large numbers of cells into the vasculature every day, few true metastatic lesions eventually develop (25). The blood and lymphatic systems are highly toxic environments. Mechanical shear stress, anoikis, and cell-mediated cytotoxicity have a profound impact on the ability of tumor cells to survive in circulation. Thus, only those cells that are capable of adapting to this microenvironment can successfully disseminate throughout the body and take root in new locations to form metastases (26, 27).   The complex mechanism by which cancers metastasize is not well understood. There are several models that have been postulated to explain the metastatic process. Traditionally, metastasis was believed to be a late event in tumor progression. Leslie Fould described tumor progression as a series of steps known as the ?linear model of metastasis? (28-30).  7 This model postulates that tumor cells begin to disseminate once the primary tumor has reached a considerable size and subpopulations of tumor cells have gained advantageous genetic modifications. Tumor dissemination correlating with primary tumor size is supported by the observation that metastatic risk is reduced by the surgical resection of tumors that are less than 2 cm in diameter (31-33). The process by which subpopulations of tumor cells gain advantageous genetic modifications over time and enable these cells to metastasize and form new solid tumors at distant sites is called the ?clonal evolution theory? (34, 35). There is also evidence for a second model of metastasis known as the  ?parallel model of metastasis?, whereby tumor cells are capable of disseminating at an early stage of primary tumor progression. This model theorizes that disseminated tumor cells evolve independently from the primary tumor and that tumor cells are capable of disseminating at early stages of tumor development. In breast cancer, there is evidence to support both linear and parallel models of metastasis (31, 36).   It is not possible to accurately predict an individual?s risk of developing distant metastasis; thus, a large number of patients receive adjuvant chemotherapy. Therefore, a subset of patients unnecessarily suffers the toxic side effects of adjuvant chemotherapy. Conversely, those patients that shun adjuvant therapy due to a perceived low relative risk of metastasis may miss an opportunity to eliminate future metastatic disease at an early stage. For both of these reasons, the discovery of reliable biomarkers to facilitate early detection of tumors that possess a high likelihood to metastasize would have profound and immediate effects on prognosis and overall breast cancer survival rates. In addition, beyond their utility as biomarkers, these proteins may play a functional role in metastatic progression and their characterization may help reveal the complex mechanisms by which primary tumors progress and metastasize, thus serving as potential therapeutic targets (37). Indeed, this is the basis of many current selective chemotherapeutic agents.     8 1.3 Breast cancer therapeutics  Predictably, the biomarkers currently used to classify breast cancer subtypes have been the basis for the most effective therapeutics.  For example ER and PgR positive breast cancers can be treated with anti-estrogens, such as tamoxifen or aromatase inhibitors.  Similarly, HER-2 positive breast cancers, are treatable by the humanized monoclonal antibody, trastuzumab (Herceptin?) that targets HER-2, or by the small molecule tyrosine kinase inhibitor, Lapatinib, which blocks both epidermal growth factor receptor (EGFR) and HER-2 receptor tyrosine kinase activity (38-41). In contrast, due to a lack of therapeutic targets in TNBCs, conventional systemic chemotherapy is the only treatment option. While some TNBCs respond initially to chemotherapy, over time they frequently become resistant resulting in high rates of relapse within the first 3 years of initial treatment (42, 43). In addition to the high risk of relapse associated with this class of tumors, patients with TNBC have a significantly shorter time of survival after diagnosis of metastatic disease (13, 42). The chemotherapy resistant characteristics of TNBC cells are thought to result from the selective survival of a small population of therapy-resistant cells known as tumor-initiating cells (TICs), or cancer stem cells. The ability to treat TNBCs by targeting the TIC population would dramatically improve patient outcomes (44).   1.4 The Cancer stem cell hypothesis Accumulating evidence demonstrates that a wide variety of malignant hematopoietic and solid tumors may be driven by a small subset of TICs that have the normal biological features of stem cells (45, 46).  Adult stem cells are a rare population of undifferentiated tissue resident cells that have the unique ability to self-renew and differentiate into multiple lineages (47). The ability of tissue stem cells to self-renew is of particular interest in cancer biology because of the many parallels to immortal cell growth. Genes that permit unlimited self-renewal and the high proliferative potential of stem cells may also, when aberrantly expressed, contribute to the malignant phenotype of cancers (48). Classic models of carcinogenesis describe cancer as a random process in which any cell can transform into a cancerous cell with the right combination of mutations. In this scenario, all or most cells in a fully developed cancer are of equal malignancy and, in order to ?cure? the cancer, all of the malignant cells need to be eliminated (49). In contrast to this classic model of  9 carcinogenesis, identification of a small population of cells in the tumor bulk with the unique capacity to form tumors when orthotopically transplanted into immunocompromised mice provided evidence for a revised model for cancer progression (50). The presence of a specialized population of TICs supports the notion of cellular heterogeneity, whereby tumors form from stem or progenitor cells. Through dysregulation of their self-renewal capacity, TICs possess the characteristics of normal cells while, at the same time, retaining key stem cell properties such as their ability to self-renew and the potential to develop into any cell in the overall tumor population. In addition, TICs retain their capacity to drive continued expansion of the population of malignant cells (49). TICs that escape chemotherapy are thought to be responsible for the high rates of cancer relapse in these TNBC patients (51, 52).   1.4.1 Cancer stem cell markers It has been challenging to find satisfactory markers that selectively identify cells with tumor initiating capacity. In addition, it also appears that TIC markers differ between types of cancers. Several groups have reported that breast cancer TICs can be isolated based on their differential expression of the cell surface proteins CD44 and CD24. CD44highCD24low cells are negative for other differentiated-cell lineage markers (53) and exhibit enhanced invasive properties compared to other subpopulations in the same tumor. Notably, CD44highCD24low cells are more prevalent in aggressive basal/mesenchymal-like human breast tumor cell lines (e.g., MDA.MB-231) compared to less aggressive luminal cell lines (e.g., MCF-7) (54).   Functionally, CD44 is a member of a large family of cell adhesion molecules (CAMs) that play a role in mediating communication and adhesion between adjacent cells and between cells and the extracellular matrix (ECM). CD44, a c-type lectin/link-domain containing protein, acts as a receptor for hyaluronan (HA) through its extracellular domain and, via its intracellular kinase domain, interacts with ankyrin and ezrin/radixin/moesin (ERM) family of binding partners to reorganize the actin cytoskeleton and mediate cell adhesion and motility (55, 56). CD44 is a widely recognized marker of tumor initiating cells in a variety of cancers and its expression heralds an increased likelihood of undergoing metastasis. CD44  10 has been shown to play a role in tumor invasion and metastasis (57-59); evasion of apoptosis; and, drug resistance (60, 61). There is also evidence that inhibition of CD44 expression or function re-sensitizes cancer cells to chemotherapy and prevents cancer recurrence (62). Thus, the ability to regulate CD44 expression would be beneficial for the development of effective cancer therapies.   While CD44 is expressed on progenitor-like cells in primary human breast tumors, the GPI-anchored cell surface sialomucin, CD24 (a.k.a. Heat stable antigen) is expressed on more differentiated cells (63-66). Although CD24 is typically not expressed in the TIC population, expression of CD24 has also been shown to positively correlate with invasive breast cancers (53, 65). The functional role of CD24 in breast tumor cells is ambiguous (67). Multiple other lineage markers are used to identify tumor cells that are poorly differentiated and represent an immature population of cells. Expression markers that are common indicators of lineage (Lin) and stage include: CD2, CD3, CD10, CD16, CD18, CD31, CD64, and CD140b (53)  The percentage of CD44highCD24lowLin- cells in primary breast tumors or tumor cell lines varies greatly depending on breast tumor subtype but, in general, these highly tumorigenic cells represent a small subpopulation of the total cells (67, 68). Tumor cells that have undergone epithelial-to-mesenchymal transition (EMT) generally have a greater frequency of CD44highCD24lowLin- cells (69). In a seminal study by Al Hajj et al., only one hundred CD44highCD24lowLin- cells isolated from human breast cancer primary tumors were required to form tumors when transplanted into the mammary fat pad of immunocompromised mice. Conversely, cells with alternate cell surface marker expression when isolated from the same patients could not form tumors in the same assay even when transplanted in much higher numbers. In addition, re-isolated CD44highCD24lowLin- cells from the tumors of recipient mice could be serially passaged in secondary recipient mice to form new tumors containing additional CD44highCD24lowLin- tumorigenic cells. These secondary tumors were also comprised of a phenotypically diverse mixed population of non-tumorigenic cells phenotypically similar to the initial patient tumor (53). These studies highlight the ability to identify prospective tumorigenic cancer cells and have paved the way  11 for the elucidation of pathways that regulate TIC growth and survival. Furthermore, because these cells drive tumor development, strategies designed to target this population may lead to more effective therapies (70).  1.5 Overview of podocalyxin, a member of the CD34 family of sialomucins 1.5.1 Podocalyxin structure and function during normal development Podocalyxin (gene name PODXL), also known as, podocalyxin like protein-1 (PCLP-1), Thrombomucin, MEP21, gp135, Tra-1-80, GCTM2 is a member of the highly conserved CD34 family of cell surface sialomucins (which includes CD34 and endoglycan) (71-74). Podocalyxin is a 150-165 kDa single-pass transmembrane protein comprised of a large, highly negatively charged, glycosylated and sialylated extracellular mucin domain, a disulfide-bonded globular domain and a highly conserved cytoplasmic tail containing a C-terminal PSD-95/Dlg/ZO-1 (PDZ)-domain binding site, terminating with the sequence of aspartate-threonine-histidine-lysine (DTHL) (71, 73, 75) (Figure 1-2). This C-terminal tail has been shown to bind the PDZ-domain proteins Na+/H+-exchanger regulatory factor (NHERF) 1 and 2, which offer a mechanism to link podocalyxin to the actin cytoskeleton (76, 77). NHERF-1 and NHERF-2 are scaffold proteins that bind to the DTHL motif of podocalyxin. Structurally, NHERF-1 and NHERF-2 both consist of two tandem N-terminal PDZ domains and a C-terminal ERM-binding region. The PDZ class II domain (PDZ2) of NHERF1/2 binds to the C-terminal DTHL motif of podocalyxin (76-78). The PDZ class I domain (PDZ1) of NHERF functions to interact with other PDZ-docking binding partners such as the cystic fibrosis transmembrane conductance regulator (CFTR), ?2 adrenergic receptor (?2AR), platelet derived growth factor receptor (PDGFR), and epidermal growth factor receptor (EGFR) (79-81). The C-terminal ERM-binding domain of NHERF binds ezrin, radixin, and moesin, closely related ERM proteins which, when activated, interact with both membrane proteins and the actin cytoskeleton. Ezrin, radixin and moesin regulate cellular survival, adhesion, and migration/invasion, all of which are important during tumor development and progression (82). By mediating interactions with membrane receptors and actin binding proteins, NHERF acts to stabilize multi-protein complexes at the plasma membrane, as well as regulate ion transport and signal transduction downstream of these receptors (83).   12 During embryogenesis, podocalyxin is expressed in all three germ layers (ectoderm, mesoderm, and endoderm) most prominently at sites that define the non-adhesive boundaries between tissues (73, 84). Interestingly within the hematopoietic compartment, podocalyxin expression in embryonic development correlates with the initial establishment of sites of active hematopoiesis. Podocalyxin is first expressed by hematopoietic cells at embryonic day 7.5 (E7.5) in the murine yolk sac, the site of primitive hematopoiesis and in circulation (85). Expression of podocalyxin steadily declines in the yolk sac until E15 when hematopoiesis shifts from the yolk sac to the fetal liver.  At this stage, 75% of fetal liver cells express podocalyxin (85). As in the embryonic yolk sac, podocalyxin expression then gradually declines in the fetal liver with continued development. When hematopoiesis switches to the fetal spleen and bone marrow at late stages of fetal development, these sites also contain subpopulations that express podocalyxin, and thereafter podocalyxin expression declines to very low levels by birth (85). In normal adult bone marrow, podocalyxin expression is restricted to a rare population of hematopoietic stem/progenitor cells known as LSK cells (Lin-, Sca-1+, c-Kit+). LSK cells expressing podocalyxin are thought to represent long-term repopulating HSC population in adult bone marrow (85). Finally, although podocalyxin is not expressed on normal mature erythroid cells, podocalyxin is expressed on mouse pre-blast erythroid progenitors and immature erythroid cells known as reticulocytes in response to hemolytic anemia, and thus may play a role in the erythropoietic stress response (86).   13  Figure 1-2. Schematic of the podocalyxin/NHERF/ezrin protein complex. Structure of podocalyxin adapted from (87). The orange regions represent the extracellular mucin domain. Within this domain, the black circles are potential N-linked carbohydrates, the horizontal lines are potential O-linked carbohydrates, and the triangles are potential sialic acid residues. The yellow region is the paired cysteine globular domain, the grey region is the stalk, blue region is the transmembrane region and red is the highly conserved cytoplasmic tail. The C-terminal DTHL sequence preferentially binds to the second PDZ domain of NHERF. Ezrin can bind either directly to podocalyxin via a juxtamembrane ezrin binding site, or through NHERF proteins.    14 Podocalyxin was first identified in human renal glomerular epithelial cells, known as podocytes and gene-deletion studies in mice identified functional roles for podocalyxin in kidney development. Podocalyxin expression on podocytes is essential for the development and maintenance of the glomerular filtration apparatus of the kidney (84). The highly negatively-charged extracellular domain of podocalyxin is required for the extensive morphological changes that occur during glomerular development (84, 88). Complete, germ-line deletion of podocalyxin in mice results in improper development of the podocytes of the kidney which, in turn, leads to anuria, high blood pressure and eventual mortality within hours of birth (84). In addition to kidney podocytes, podocalyxin is also normally expressed by vascular endothelia and hematopoietic progenitor cells (89). Podocalyxin is also expressed at low levels on the luminal face of kidney tubule cells, breast ductal lumens, oviductal luminal cells, mesothelial cells, and a discrete subset of neurons in adults (90).   Despite podocalyxin being expressed on multiple cell-types throughout embryonic and adult development, its presence during the development of the glomerular filtration apparatus of the kidney remains its only known essential role. In addition to defects in the kidney filtration apparatus, complete deletion of podocalyxin in mice results in herniation of the gut into the umbilical cord (omphalocele), likely due to an inability of the gut to retract through the umbilical ring without podocalyxin acting as an anti-adhesive molecule (84). It is likely that podocalyxin expression has more subtle effects on the luminal surface of tubules and ducts during normal development and morphogenesis, such as during the formation of the embryonic aorta (91).   1.5.2 Podocalyxin and adhesion Podocalyxin is best known to function as an anti-adhesive due to its highly negatively-charged, glycosylated, extracellular mucin domain (76, 92). A number of studies have shown that podocalyxin, through its ability to modulate adhesion, is critical for normal development and functioning of the podocyte foot processes of the kidney and embryonic aorta (84, 91, 93, 94). In addition, Takeda et al. were the first to show that podocalyxin overexpression inhibits cell-cell adhesion in an expression-dependent manner in Madin-Darby canine kidney (MDCK) cells in vitro (76). Ectopic expression of podocalyxin is also  15 sufficient to induce microvilli formation on the apical surface of MDCK kidney epithelial cells and MCF-7 breast carcinoma cells (74, 92). The highly negatively-charged, glycosylated, extracellular domain of podocalyxin was found to be critical for microvillus formation, whereas, surprisingly, the presence of most of cytoplasmic tail (including the highly conserved NHERF-binding site) was dispensable (92). Subsequent studies have shown that forced overexpression of podocalyxin in OVCAR-3 ovarian carcinoma cells and MCF-7 breast carcinoma cells also leads to decreased adhesion to matrix, detachment from basement membranes, and morphological changes (95-97). Rather than overt destruction of cell-ECM contact sites, podocalyxin acts to more subtly alter adhesion by excluding ECM-adhesion complexes, such as ?1 integrin and focal-adhesion kinase (FAK) from free apical domains; by recruiting and sequestering cortical actin complexes to the apical membrane; and, subsequently, weakening adhesion of the basolateral membrane to ECM (90). This is likely to be highly significant since alterations in cell-cell and cell-endothelium adhesion occurring during neoplastic growth are known to be critical for tumor invasion and progression to metastasis (98).   1.5.3 Intracellular binding partners of podocalyxin NHERF-1 (EBP-50), NHERF-2 (E3KARP/TKA-1), and ezrin are the most extensively studied intracellular binding partners of podocalyxin. Ezrin and NHERF-1/2 function as adapter proteins and thus have the ability to connect podocalyxin to a multitude of signaling pathways (87). There is a great deal of evidence to support that changes in expression, localization, and mutations in known podocalyxin binding partners can result in increased tumorigenesis (90). For example, ezrin promotes invasion and metastasis in a number of cancers, including breast cancer (99-105). Ezrin is a member of the ERM family of proteins that act to connect integral membrane proteins, such as CD44, and intercellular adhesion molecules (ICAMs), to the actin cytoskeleton (106-108). Podocalyxin and ezrin colocalize at the apical membrane domains of podocytes and breast and prostate tumor cell lines (105, 109). In experimental animal models and in vitro, podocalyxin, via its interaction with ezrin, promotes tumor cell migration and invasion, possibly through activation of the MAPK pathway (105).    16 1.5.4 Podocalyxin and cancer Aberrant expression of podocalyxin has been implicated in many different types of cancer including germ-cell cancers, carcinomas (e.g., breast, ovarian, pancreatic, colon and bladder), malignant astrocytoma, and leukemia. A number of studies have retrospectively analyzed patient tumor samples to assess whether podocalyxin expression correlates with survival or prognosis. Protein and transcript expression analyses show that podocalyxin is often expressed on a subset of extremely aggressive epithelial cell or germ-cell tumors that have a greater propensity to metastasize (90). This data is supported by several studies using human cancer cell lines (90, 97, 105, 110-112). In addition, podocalyxin has the potential to be used as a diagnostic marker to identify primary tumors more likely to metastasize and thus demand adjuvant chemotherapy (38, 95, 110, 113, 114). Larsson et al. suggest that podocalyxin could be used as a predictive marker for adjuvant treatment of colorectal carcinoma, as patients with stage III colorectal cancer that have high levels of membranous podocalyxin expression benefited from adjuvant chemotherapy. In contrast, colorectal carcinomas with low podocalyxin expression did not benefit from this treatment (114).  Thus, detectable expression of podocalyxin on primary tumors may be an indication of those tumors that have a high likelihood to progress to form metastatic lesions and would benefit from early treatment with chemotherapeutics (114).   1.5.5 Upregulated podocalyxin expression and breast cancer Podocalyxin is normally expressed on the luminal ductal epithelial cells of the mouse and human mammary gland. In a 272 patient sample array comprised of primary breast tumor biopsy specimens, podocalyxin was highly expressed on a subset (6%) of lymph node-negative breast tumors. In these patients, podocalyxin expression is a strong, independent predictor of poor prognosis and survival: Those breast tumors that express high levels of podocalyxin tend to be more aggressive and have a higher likelihood to progress to form distant metastases (97). Forse et al. provided corroborating data that podocalyxin expression is an indicator of poor prognosis in breast carcinoma. Additionally, Forse et al. showed that podocalyxin expression is associated with those tumors with large size, high histological grade, ER and PgR negativity, and a basal-like subtype (EGFR and cytokeratin-5 positivity). Surprisingly, in this study they also showed that axillary lymph node negative  17 (ANN) patients with podocalyxin-positive tumors had better disease free survival if they showed signs of lymphatic invasion (LVI), compared to those patients with podocalyxin-negative tumors. Thus, despite podocalyxin expression and LVI both being indicators of poor prognosis on their own, when analyzed together, these authors concluded that podocalyxin expression on LVI-positive tumors is an indicator of improved disease free survival (115).  1.5.6 Ectopic expression of podocalyxin in breast cancer cell lines The human breast cancer cell line, MCF-7, which normally express low levels of endogenous podocalyxin, is minimally invasive and non-metastatic. Upon ectopic expression of podocalyxin (MCF-7Podxl), MCF-7 cells ?bulge apically?, produce apical and lateral microvilli.  These cells become less adhesive in vitro, and exhibit a delay in targeting integrins to the basolateral surface (92, 97). These studies demonstrate that MCF-7Podxl cells exhibit morphologic features of more aggressive carcinomas. In addition, MCF-7Podxl cells actively shed from confluent monolayers, have disrupted E-cadherin localization, and perturbed cell junctions between cells (97). Subsequent studies also showed that podocalyxin may contribute to an increase in invasion.  For example, MCF-7Podxl cells exhibit enhanced migration towards serum, invasion through matrix coated membranes, and up-regulated expression of the matrix metalloproteinases (MMPs) MMP1 and MMP9 (105). Recently, MCF-7Podxl cells have been shown to exhibit collective tumor migration through the mammary fat pad in RAG2M-/- mice (116). Interestingly, this increase in collective MCF-7Podxl invasion through the stroma and lymph node of the mammary gland was found to occur independently of overt epithelial-to-mesenchymal transition (EMT).  In this scenario, podocalyxin appears to drive collective tumor migration without features typical of EMT including disrupting cell-cell interactions, down regulating epithelial markers (cytokeratins or E-cadherin) or altering epithelial-like cell morphology. Thus, podocalyxin likely contributes to primary breast tumor invasion by expanding non-adhesive luminal membrane surfaces while restricting the size and function of adhesive surfaces, which may force collective migration of cohesive tumor cells without overtly inducing EMT (96, 116). Intriguingly, unlike other mucins (MUC1 (also known as polymorphic epithelial mucin (PEM)), MUC4, and MUC18 (also known as CD146 or melanoma cell adhesion molecule  18 (MCAM)) that facilitate a reduction in adhesion via an EMT-dependent mechanism when overexpressed in tumors, podocalyxin-mediated invasion may occur by an EMT-independent mechanism (96, 117-121).   It is noteworthy, however that others have reported conflicting observations. For example, Meng et al. reported that exposure to transforming growth factor beta (TGF-?), a known inducer of EMT, results in an increase in podocalyxin expression, subsequent loss of apical polarity and acquisition of a more fibroblast like spindle morphology in A549 lung adenocarcinoma cells. These effects were coupled with hallmarks of EMT including, loss of E-cadherin and an increase in vimentin expression and could be recapitulated by simple ectopic expression of podocalyxin (122). Thus, the precise role of podocalyxin in breast tumor progression is not well established.   1.6 Hypothesis  Given the evidence that podocalyxin is highly expressed on a distinct subset of tumors that are highly malignant, I postulated that podocalyxin actively contributes to the aggressive phenotype of these tumors. I hypothesized that podocalyxin expression not only marks extremely aggressive and invasive breast carcinomas, but also directly facilitates tumor progression and metastasis by modulating tumor cell adhesion, migration, invasion, or survival.     19 1.6.1 Objectives My overall objective was to determine whether podocalyxin has a direct, causal role in breast cancer tumor invasion and migration using in vivo models of tumor growth and metastasis.  Aim 1. To silence expression of PODXL in the metastatic breast cancer cell line, MDA.MB-231 and then determine if loss of podocalyxin results in morphological and functional changes in MDA.MB-231 cells in vitro by performing assays to assess: proliferation, adhesion, and anchorage-independent growth potential.  Aim 2. To determine the role of podocalyxin in tumor invasion and metastasis in vivo by transplanting PODXL-deficient cell lines (PODXL-KD) and non-deleted controls (Scr-ctrl) into highly immunocompromised NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice to assess primary tumor growth, invasion and metastasis. In addition to this xenograft cancer model, I also performed syngeneic tumor models to assess the role of podocalyxin in mouse mammary tumor 4T1 cells in immunocompetent BALB/c mice.   Aim 3. To use podocalyxin-specific antibodies generated in collaboration with UBC?s Centre for Drug Research and Development (CDRD) to test the ability of podocalyxin targeted therapies to block breast cancer tumor progression and metastasis in a pre-clinical mouse model of primary tumor growth and metastasis.    20 CHAPTER 2: MATERIALS AND METHODS 2.1 Cell culture MDA.MB-231, MCF-7, T47D and 4T1 cell lines were purchased from the American Tissue Culture Collection (ATCC, Rockville, MD, USA). MDA.MB-231 cells were cultured in RPMI (Gibco, Burlington, ON) supplemented with 5% fetal bovine serum (FBS), penicillin, and streptomycin. MCF-7 and T47D cells were cultured in DMEM (Gibco, Burlington, ON) supplemented with 10% FBS, penicillin and streptomycin. 4T1 Balb/c mouse derived mammary tumor cells were grown in DMEM (Gibco, Burlington, ON) supplemented with 10% FBS, 2mM glutamine, non-essential amino acids, penicillin, and streptomycin. SUM-149 cells were generously provided by Dr. Sandra Dunn (Child and Family Research Institute, Vancouver, B.C.) and originally purchased from Asterand Inc. (Detroit, MI). SUM-149 cells were grown in DMEM/F12 media (Invitrogen, Carlsbad, CA) supplemented with 5% FBS, bovine pancreas insulin (5?g/ml; Sigma, Oakville, ON), hydrocortisone (1?g/ml; Sigma, Oakville, ON), HEPES, penicillin, and streptomycin. All cell lines were cultured under normal cell culture conditions in 5% carbon dioxide at 37?C.  2.1.1 Cell line passaging Cell lines were passaged or harvested for experiments upon reaching 80 to 90% confluence. Adherent cells were gently washed once with pre-warmed Ca2+ and Mg2+ free, phenol red containing Hank?s buffered salt solution (HBSS) (Gibco, Burlington, ON). Cells were trypsinized for five minutes with 0.25% Trypsin/EDTA (Invitrogen, Carlsbad, CA) at 37?C. Trypsin was quenched with normal growth medium and cells were centrifuged for five minutes at 300 x g to remove residual enzyme.   2.2 Transfections and generation of stable cell lines 2.2.1 Silencing PODXL expression in MDA.MB-231 cells.  First, MDA.MB-231 cells were fluorescently labeled with GFP or RFP by infecting with retrovirus vectors pLNCX2-GFP or pLNCX2-RFP, respectively. All cell lines used were obtained from pooled cultures. Human PODXL was silenced (?knocked down?) in MDA.MB-231 breast tumor cells by lentiviral infection using pLKO.1 with either a scrambled shRNA construct (Scr-ctrl) or shRNA targeting the PODXL gene (PODXL-KD). Cells were  21 cultured under continuous antibiotic selection with puromycin (4?g/ml; Invitrogen, Burlington, ON) and G418 (1mg/ml; Calbiochem, Darmstadt, Germany). shRNA constructs were generously provided by Dr. John Wilkins (University of Manitoba, Winnipeg, MB) and the infections were performed by Michelle Turvey and Dr. Shaun McColl (University of Adelaide, Australia).    2.2.2 Silencing mouse Podxl in 4T1 mouse mammary tumor cell line Predicted shRNA sequences designed for targeted knockdown of murine podocalyxin were identified using PSI Oligomaker v1.5 freeware (http://web.mit.edu/jacks-lab/protocols/pSico.html). Three individual shRNA oligomers were each cloned into the into HpaI and XhoI sites of the pLL3.7 lentiviral vector downstream of the U6 RNA promoter. Positive clones were selected and then verified by sequencing. 4T1-luc cells were maintained under selection in G418 (400?g/ml; Calbiochem, Darmstadt, Germany). To produce lentiviral particles, 293T cells were co-transfected with 10mg of pLL3.7 and the appropriate packaging plasmids (3.5mg of pVSVg, 3.5mg of pRSV-Rev, 6.5mg of pMDLgag/pol) by calcium phosphate transfection. Lentiviral containing media was collected 36 hours post-transfection and transferred to sub-confluent 4T1 cells seeded the day before. The virus-containing media was replaced with regular growth media after 48 hours and incubated for an additional 48 hours. The cells were then harvested for mouse podocalyxin transcript and protein expression analysis. Knockdown of Podxl in 4T1 cells was performed by Jane Cipollone and Dr. Marcia Graves at the University of British Columbia.   2.2.3 Forced expression of Podxl in MCF-7 breast tumor cells Parental MCF-7 cells were transfected with either control plasmid (pIRES2-EGFP) or pIRES2 containing the full-length murine podocalyxin cDNA using DMRIE-C according to the manufacturer?s instructions (Invitrogen, Carlsbad, CA). To enrich for high levels of ectopic podocalyxin expression, genetically selected stable cell populations (MCF-7Podxl) were subjected to two independent rounds of FACS after indirect immunofluorescent labeling of murine podocalyxin using a species-specific antibody (anti-mouse PCLP-1; MBL, Nagoya, Japan). Cells transfected with the empty vector (MCF-7pIRES) were also collected  22 by FACS to control for any phenotypic changes resulting from transfection and the cell sorting process. The sorted, stable transfected populations were expanded in culture and maintained under genetic selection in DMEM/F12 supplemented with 5% FBS and gentamycin (50?g/ml; Sigma, Oakvile, ON). Overexpression of Podxl in MCF-7 cells was performed by Drs. Julie Nielsen and Marcia Graves at the University of British Columbia.   2.3 Western blotting Cells were grown to 80% confluence, rinsed twice with cold PBS and scraped on ice using a cell lifter in 200?l of RIPA lysis buffer (150mM NaCl, 50mM Tris pH 7.4, 5mM EDTA, 1.0% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulphate (SDS) and protease inhibitor cocktail (Calbiochem, Darmstadt, Germany)). Cells in lysis buffer were transferred to a 1.5ml microfuge tube on ice and rotated for 15 minutes at 4?C followed by centrifugation at 15700 x g for 10 minutes. The concentration of protein in the total cell lysate was determined by BCA assay (Pierce, Rockford, IL) according to the manufacturer?s instructions. Equal amounts of protein were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes (Biorad, Hercules, CA). Membranes were blocked for 1 hour in 10% bovine serum albumin (BSA) prepared in a 1:1 mixture of PSB and TBS. Membranes were washed in Tris (pH 7.5) Buffered Saline with 0.05% Tween (TBS-T). Membranes were incubated with primary antibody overnight at 4?C.  All primary antibodies were prepared in 2% BSA in TBS-T. Membranes were probed with mouse-anti-human podocalyxin (clone 3D3) (2?g/ml; Santa Cruz Biotechnology, Santa Cruz, CA) and GAPDH as loading control (0.5?g/ml; Ablab UBC Antibody Facility, Vancouver, BC). Membranes were washed three times for five minutes each to remove excess primary antibody. Primary antibody binding was detected by incubating membranes with anti-mouse secondary antibody coupled to IRDye? 700 (LI-COR, Lincoln, NE) for 45 minutes protected from light or HRP?conjugated rabbit anti-mouse IgG (Jackson Immunoresearch, West Grove, PA) in TBS-T. The membrane was then washed three times for five minutes each protected from light to remove excess secondary antibody and protect the signal. Protein on the membrane was visualized using a LI-COR Odyssey imaging system (LI-COR, Lincoln, NE). For western blots incubated with HRP conjugated secondary antibodies, enhanced chemiluminescence  23 reagents (ECL, Perkin Elmer, Woodbridge, ON) were applied for one minute and developed after exposing film (Kodak) to membranes.   2.4 Quantitative RT-PCR 2.4.1 RNA isolation and cDNA synthesis  Total RNA isolation was performed using TRIzol (Ambion, Rockford, IL) according to manufacturer?s instructions. TRIzol (1ml) was added directly to cells cultured to confluence in a six well plate. A 1ml pipette tip was used to rinse the cells off of the plate and the TRIzol cell lysate was transferred to a 2mL microfuge tube. Next, 200?l of chloroform (CHCl3) was added to the tubes and shaken for 15 seconds followed by centrifugation at 4?C for 15 minutes at 12000 x g. The upper-phase solution (~500?l) was transferred to new microfuge tubes without disruption of the phenol-containing lower layer or interface. An equal volume of isopropanol (500?l) was added to the upper-phase and inverted to mix. The mixture was then centrifuged at 12000 x g for 10 minutes at 4?C to pellet the RNA. The supernatant was decanted and 1ml of 75% ethanol (3:1 absolute EtOH:RNAse free water) was rinsed down the sides of the tube and then carefully removed without disruption of the pellet. Finally, the pellet was dissolved in 50?l of RNase free water. The RNA concentration was determined by NanoDrop (ThermoFisher Scientific Inc., Wilmington, DE) using the ND1000 software program. Complement DNA (cDNA) was synthesized from 2000ng RNA using the High capacity cDNA Reverse Transcription Kit (200 reactions) (Applied Biosystems, Foster City, CA) according to the manufacturers instructions.   2.4.2 Quantitative Reverse-Transcriptase PCR Quantitative polymerase chain reaction (qPCR) was performed using SYBR fast (Kapabiosystems, Woburn, MA). Samples were run in duplicate on 384 well PCR plates. Master mix for each primer set was made from 5?l Sybr fast, 2?l RNase-free water, and 1?l of 2mM primer. A total of 2?l of 500ng/?l cDNA was used for the amplification of each gene. Reactions were carried out using an ABI 7900 real-time PCR machine (Applied Biosystems, Carlsbad, CA). The PCR program in Table 2-1 was used to amplify all genes. Values were expressed relative to GAPDH.    24 PODXL 70 primer sequence:  hPODXL70 Forward: 5'-CTCACCGGGGACTACAACC-3' hPODXL70 Reverse: 5'-GCCTCCTCTAGCCACGGTA-3'   Table 2-1. Thermocycler program for qPCR. Step Temperature Time Ramping  1. Initial heating:  95?C  20 seconds Amplification                                    2. Denaturation: 3. Annealing and Extension:  Repeat for 40 cycles  95?C  1 second 60?C 20 seconds    Dissociation 4. 5. 6.  95?C  15 seconds 60?C 95?C 15 seconds 15 seconds     Data was collected at the annealing and extension stage (60?C) and during the ramp between 60?C and 95?C of the dissociation stage.    2.5 Adhesion assay Different concentrations of collagen (5?g/cm2 and 10?g/cm2) and Matrigel? (1:10 [low] and 1:5 [high] dilutions) were prepared in PBS and plated in triplicate in 96-well plates for each time-point analyzed. Plates were incubated overnight at 4?C to allow for complete coating with substrates. Sub-confluent MDA.MB-231 cells were harvested and washed twice with serum-free RPMI. Substrate solutions were aspirated and 2 x 104 MDA.MB-231 cells in serum free RPMI were plated in each well. Plates were incubated at 37?C for 15, 30, or 60 minutes. Media and non-adherent cells were aspirated at the end of each time point and wells were gently washed three times with room temperature PBS using a multichannel pipette. Adherent cells were fixed with ice-cold 95% ethanol (EtOH) for 15 minutes and stained with 0.1% crystal violet in 95% EtOH for 15 minutes. Wells were washed with  25 distilled water until clear. Dye was released from stained cells by adding 100?l of a solution of 30% acetic acid in methanol and transferred to a clean 96 well plate. Absorbance of each well was read at 590nm using a Spectra-max 190 microplate reader (Molecular Devices, Sunnyvale, CA). Uncoated and ECM coated wells with no cells were stained to control for background absorbance.   2.6 MTS proliferation assay Proliferation of monolayer cultures was determined using a MTS colorimetric assay as a surrogate to measure cell growth and viability. Scr-ctrl and PODXL-KD MDA.MB-231 cells (1 x 104 cells) were serially diluted into 96 well plates with six replicates each. A CellTiter 96? Non-Radioactive Cell Proliferation Assay (Promega, Madison, WI) kit was used to assess cell growth at 48, 72, and 96 hour time-points. Cells were incubated with tetrazolium salt, which gets converted into a formazan product that is detectable at 490nm using a Spectra-max 190 microplate reader (Molecular Devices, Sunnyvale, CA).    2.7 Anchorage-independent growth assays 2.7.1 Tumorsphere assay Sub-confluent MDA.MB-231 or MCF-7 cells were harvested by first rinsing adherent cells with 5ml Ca2+ Mg2+ free HBSS. A sterile cell scraper was used to scrape the cells gently from the dish and cells were resuspended in 5ml of complete MammoCult? media (MammoCult? Basal Medium (Stem Cell Technologies, Vancouver, BC), Complete MammoCult? media consists of a 1/10 dilution of MammoCult? proliferation supplement (Stem Cell Technologies, Vancouver, BC), 0.48?g/ml fresh hydrocortisone (Stem Cell Technologies, Vancouver, BC), and 4?g/ml heparin (Stem Cell Technologies, Vancouver, BC) in MammoCult? media. Cells were centrifuged for four minutes at 300 x g and viable cells were counted using trypan blue stain. A total of 5 x 103 MDA.MB-231 cells or 2 x 104 MCF-7 cells were seeded in triplicate into ultra-low adherent six-well plates (Corning Inc., Corning, NY) in 2ml complete MammoCult? medium. Cells were incubated at 37?C in 5% CO2 for seven days without disruption to prevent cellular aggregation. The number of tumorspheres was quantified by manually using a transparent counting grid. Only spheres larger than 60?m (approximately 15 cells) were counted. Tumorsphere forming efficiency  26 (%) was calculated by (number of tumorspheres / number of cells initially plated) x 100. Tumorsphere size was measured from representative images of tumorspheres observed using the 10X objective of a dissecting microscope. ImageJ software was used to calculate the area of the tumorspheres in each image.   Tumorspheres were passaged every seven days. Spheres from triplicate wells were pooled and centrifuged at 300 x g for five minutes. Pre-warmed trypsin-EDTA (1ml) was added to the pellet and tumorspheres were triturated for 90 seconds using a p1000 pipette tip. Trypsin was quenched by adding 5ml of 2% FBS in cold HBSS and the cells were centrifuged at 300 x g for five minutes. A viable cell count was performed using trypan blue and cells were re-plated in ultra-low adherent plates as described previously.    2.7.2 Colony forming assay For methylcellulose colony forming assays tumor cells were harvested and plated as in the tumorsphere assay described above. Cells were plated in triplicate in 0.9% MethoCult base methylcellulose medium for human cells (Stem Cell Technologies, Vancouver, BC) in complete MammoCult? medium. Cloning efficiency (%) was determined by (number of colonies / number of cells initially plated) x 100.   2.8 Animals Tumor model animal experiments were carried out using 6-12 week old female NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice obtained from the Jackson Laboratory (Cat# 005557, Bar Harbor, ME) and BALB/cJ mice obtained from the Jackson Laboratory (Cat# 000651, Bar Harbor, ME). Animals were maintained and bred in a specific pathogen-free environment and tested negative for pathogens in routine screening. All experiments were carried out at the University of British Columbia following institutional and Canadian Council on Animal Care (CCAC) guidelines approved by the University of British Columbia Committee on Animal Care.     27 2.9 Xenograft models of breast tumorigenesis 2.9.1 Subcutaneous xenograft model of tumor development For subcutaneous models of tumor development, sub-confluent MDA.MB-231 (Scr-ctrl and PODXL-KD) or MCF-7 (MCF-7pIRES and MCF-7Podxl) cells were harvested and 1 x 106 (MDA.MB-231) or 3 x 106  (MCF-7) cells resuspended in a 1:1 solution of sterile Matrigel? (BD Biosciences, Mississauga, ON) and Hank?s Balanced Salt Solution (HBSS) (Gibco, Burlington, ON) kept on ice, and then injected (subcutaneous) into the shaved right hind flank of NSG mice. MDA.MB-231 tumors were allowed to develop for 21 to 27 days and tumor size was quantified at least three to five times per week by measuring tumor dimensions (length and width) using manual calipers. MCF-7 tumors were allowed to develop for three months and tumor dimensions were measured using manual calipers as described above. Tumor volumes (mm3) were calculated using the formula ((length ? width2) / 2). Mice were sacrificed when tumors reached a volume no greater than 1.0cm3. Tumors and lungs were excised for further analysis as described below.   2.9.2 Competitive subcutaneous xenograft model of tumor development and surgical resection For competitive subcutaneous tumor experiments sub-confluent Scr-ctrlRFP and PODXL-KDGFP MDA.MB-231 cells were harvested and 1 x 106 cells were prepared in a 50:50 mix in a 1:1 solution of Matrigel? and HBSS and injected into the right hind flank of NSG mice. Primary tumors were resected when tumors reached a volume of approximately 250-500 mm3 (18 days post-implantation). Mice were given a pre-operative dose of buprenorphine (Temgesic) analgesic (0.05-0.10 mg/kg; Schering-Plough, Merck & Co, Whitehouse Station, NJ). Mice were anesthetized with ketamine (100mg/kg; Bioniche, Belleville, ON) and xylazine (10mg/kg; Bayler Inc., TO) and 1-3% isoflurane gas (to effect) was continuously administered for the duration of the surgery. Mice were placed on a heating pad for the duration of the surgical procedure and Refresh LacriLube natural tear gel (Allergan Inc., Markham, ON) was applied to their eyes to prevent drying. Mice were first checked for pedal reflexes to confirm adequate plane of anesthesia for surgery and then a small incision was made in the flank ~ 2mm from the tumor bulk.  The skin was dissected free of the tumor and the tumor was dissected en masse from adherent tissue to resect. Topical anesthetic,  28 0.5% bupivacaine hydrochloride (Marcaine?) (Hospira, Lake Forest, IL), was applied to the incision area prior to suturing. Wound closure was performed using sterile 5-0 ? inch suture needles threaded with 27? coated vicryl (polyglactin 910) absorbable suture thread (J433) (Ethicon Inc., Blue Ash, OH). Buprenorphine analgesic was provided subcutaneously every 8-12 hours as needed. Mice were monitored closely for signs of post-operative dehydration and pain. Resected tumors were retained for further analysis by flow cytometry as described below. Cages were kept partially on a heating pad for three days post-surgery. On day 38 post-resection (56 days post-transplantation) mice were sacrificed. Relapsed primary tumors that had formed at the site of excision and lungs were excised and retained for further analysis as described below.   2.9.3 Competitive orthotopic xenograft model of tumor development For competitive orthotopic tumor experiments, mice received pre-surgical dose of analgesic and then were anesthetized as described previously for surgical resections of tumors. Once plane of surgical anesthesia was confirmed, a 1 to 1.5 cm midline incision was made in the skin, being careful not to puncture the peritoneal cavity or injure the abdominal musculature. A lateral incision from the midline point was made ending between the #4 and #5 nipples. The skin was loosened from the body wall using a sterile cotton swab dipped in sterile saline solution to expose the left abdominal (#4) mammary fat pad. A 50?l Hamilton syringe and sterile 22 gauge needle was used to transplant 10?l of a 50:50 mixture of 1 x 106 Scr-ctrlRFP (or Scr-ctrlGFP) cells and PODXL-KDGFP (or PODXL-KDRFP) MDA.MB-231 cells prepared in a 1:1 solution of Matrigel? (BD Biosciences, Mississauga, ON), HBSS and 1?l of trypan blue into the right abdominal (#4) mammary fat pad of NSG mice. Wound closure and post-operative care was performed as described previously for surgical resections of tumors. Mice were sacrificed after 23 days and tumors and lungs were excised and retained for further analysis by flow cytometry as described below.   2.9.4 Competitive experimental xenograft model of lung metastases Mice were warmed under a heat lamp for 5 minutes to dilate their superficial tail veins. Mice were immobilized in a restraining device and a 50:50 mixture of between 5 x 104 and 2 x 105 Scr-ctrlRFP (or Scr-ctrlGFP) and PODXL-KDGFP (or PODXL-KDRFP) MDA.MB-231 cells in a  29 0.1% BSA + HBSS mixture was into the lateral tail vein in 100?l volume using a ? c.c insulin syringe (Beckton Dickinson, Franklin Lakes, NJ). Lungs were analyzed by flow cytometry 3, 7, or 14 days post-injection (early lung colonization/micrometastasis) as described below. In some experiments, mice were left for six weeks (late-stage metastasis/macrometastasis), at which time they were sacrificed to collect lungs, livers, and bone marrow obtained from femurs and tibias for analysis by fluorescence microscopy and flow cytometry to quantify the tumor burden at each site as described below.   2.10 Microscopic imaging of metastatic sites Fluorescent images of the lungs and livers were taken using a Leica Fluo? dissecting microscope (Leica, Solms, Germany) using QImaging? software (QImaging, Redwood City, CA) and tumor nodules on the lung were counted manually from these images.   2.11 Preparation of single cell suspensions from primary tumors and metastatic sites Solid primary tumors were minced and digested in a solution of collagenase (2mg/ml; Sigma, St Louis, MO), 2mg/mL dispase (2mg/ml; Roche, Mannheim, GER) in HBSS for one to two hours at 37?C to obtain a single cell suspension. The cellular suspension was then filtered through a 70?m cell strainer, centrifuged at 453 x g for five minutes, and red blood cells were lysed in an ammonium chloride solution (mouse red cell reduction buffer (MRCRB)). Cells were washed in PBS, passed through a 70?M strainer and then resuspended in FACS buffer (2% FBS, 2mM EDTA, PBS, 0.05% NaAzide) prior to flow cytometric analysis performed as described below.   Lungs from NSG mice that had been subcutaneously, orthotopically, or intravenously injected with tumor cells were perfused with 10mL ice-cold PBS, excised, digested and prepared for flow cytometry as for the tumors described above.   Livers were prepared for flow cytometric analysis as previously described for tumors and lungs except that 33U/mL bovine pancreas DNase I (Sigma, St Louis, MO) was added to the collagenase/dispase digestion solution.    30 Femurs and tibias were excised from mice and bone marrow was flushed with 10mL ice-cold PBS using a 26cc syringe and red cells were lysed as described above prior to flow cytometric analysis.  2.12 Flow cytometry Sub-confluent cell lines were washed once with Ca2+ and Mg2+ free HBSS (Invitrogen, Carlsbad, CA) harvested with 0.25% Trypsin/EDTA (Invitrogen, Carlsbad, CA). Cells were quenched with normal growth media and centrifuged at 300 x g for 5 minutes. Cells were washed twice with PBS and resuspended in FACS buffer (2% FBS, 2mM EDTA, PBS, 0.05% NaAzide). Cells were blocked for 15 minutes at 4?C in 5?g/ml dilution of 2.4G2 (anti-CD16/CD32) in a 96 well ?v? bottom plate. Cells were spun at 453 x g for 4 minutes and primary antibodies were incubated for 30 minutes at 4?C. Goat-anti-human podocalyxin (2?g/ml; R&D Systems, Minneapolis, MN) was used for the detection of human podocalyxin by flow cytometry. Goat-IgG (R&D Systems, Minneapolis, MN) was used as an isotype control. After primary incubation with primary antibodies, cells were washed three times in FACS buffer. Cells were then incubated with chicken-anti-goat AlexaFluor (AF) 647-coupled secondary antibody (6.6?g/ml; Invitrogen-Molecular Probes, Carlsbad, CA) for 30 minutes at 4?C. Cells were washed three times with FACS buffer and transferred to bullet tubes for flow cytometry. PE and FITC channels were used to detect RFP and GFP labeled MDA.MB-231 cells. A LSRII flow cytometry machine (BD Biosciences, Mississauga, ON) was used for all flow cytometric experiments. FlowJo? software was used to analyze all flow cytometry data (FlowJo, Treestar Inc., Ashland, OR).   2.13 Tumor histological analysis Solid tumors were fixed overnight in 10% buffered formalin, embedded in paraffin and then serially sectioned for histological analysis. Representative sections were deparaffinized and stained with hematoxylin and eosin (H&E) using standard staining procedures. Tumor specimens were fixed overnight in 10% buffered formalin (Sigma Aldrich, St. Louis, MN) and paraffin embedded in blocks. Blocks were serially sectioned at 5?m thickness onto standard glass microscope slides. Slides were rehydrated in xylene (twice, 10 minutes each), 100% ethanol (twice, 5 minutes each), 95% ethanol (2 minutes), 70% ethanol (2 minutes), and  31 distilled water (twice, 2 minutes each). Slides were stained with hematoxylin (2 minutes) and rinsed with tap water until the color of the rinse changed from red to blue. Slides were then stained with eosin solution made up in 95% ethanol (5 minutes). Slides were subsequently dehydrated in 95% ethanol (twice, 2 minutes each), 100% ethanol (twice, 2 minutes each), and xylene (twice, 10 minutes each). Permount solution (eBioscience) was used to mount specimens on slides. H&E stained sections were examined qualitatively for evidence of muscular invasion and tumor border integrity.   2.14 Experimental syngeneic model of lung metastasis BALB/c mice were warmed under a heat lamp for 5 minutes to dilate superficial tail veins. A total of 1 x 105 VC or Podxl-KD 4T1-luc cells were intravenously injected into BALB/c mice. Lungs were perfused and excised as described above.    2.15 Luciferase enzymatic assay Total luciferase activity was assayed from lungs harvested from BALB/c mice intravenously injected with luciferase expressing 4T1 cells. Lungs were homogenized in 1X cell lysis buffer (Promega, Madison, WI). Protein concentration was determined using a BCA Protein Assay kit (Pierce, Rockford, IL). Luciferase Reporter System (Promega, Madison, WI) was used to detect luciferase activity, 20?l of sample supernatant was mixed with 50?l of luciferase assay reagent (LAR) and luciferase activity was quantified using a SpectraMAX L Luminescence microplate reader (Molecular Devices, Sunnyvale, CA). Results were reported as relative light units (RLU).  2.16 Therapeutic antibody production  Note that most of the following steps were performed by personnel at the CDRD. Rabbits were immunized with MDA-MB-231 breast carcinoma cells that express high levels of glycosylated human podocalyxin on their cell surface. Individual B-cell clones were isolated from animals whose sera recognized MDA-MB-231 cell podocalyxin. Individual B-cell clone supernatants were screened against immobilized extracellular domains of podocalyxin produced by MDA-MB-231 tumor cells and HEK293 (kidney) cells by ELISA. Next, supernatants were screened against MDA-MB-231 and HEK293 cells with and  32 without podocalyxin on their surface to ensure that the antibodies recognize the native form of the molecule and that they do not exhibit non-specific binding to cells. Finally, CHO cells expressing podocalyxin and related family members CD34 and endoglycan were screened to ensure that the antibodies targeted podocalyxin specifically. By comparing affinities for podocalyxin expressed on tumor and normal cells, B cell clones that produce antibodies with favorable binding profiles to tumor cells were selected. Once screening and characterization was complete, the VH and VL regions from clones producing high affinity anti-podocalyxin antibodies were rescued by PCR and subcloned, in-frame, into expression vectors containing the constant regions of human heavy and light chains, respectively, such that humanized IgG1 molecules will be generated by and purified from myeloma cells (123).  2.16.1 In vivo screening to identify candidate therapeutic antibodies Based on affinity assessment using flow cytometry and ELISA, several anti-podocalyxin antibodies and non-specific controls (anti-OVA) were selected for efficacy screening in vivo.  To do this, 1 x 106 MDA.MB-231RFP tumor cells were incubated or 30 minutes at room temperature with 25?g of one of the anti-podocalyxin candidate antibodies or a non-specific control anti-OVA antibodies. Prior to injection, the tumor cell/antibody mixture was diluted 2:1 in Matrigel? on ice and then injected subcutaneously injected into the flank region of NSG mice. Tumor dimensions were measured every three days and tumor volumes (mm3) were calculated by ((length x width2)/2).  2.16.2 Pre-clinical mouse model to assess anti-podocalyxin antibody efficacy MDA.MB-231RFP tumor cells were pre-treated with 25?g of either anti-PODO83 or non-specific control anti-OVA peptide antibodies per 106 tumor cells for 30 minutes at room temperature. Prior to injection, the tumor cell/antibody mixture was diluted 2:1 in Matrigel? on ice. A total of 1 x 106 MDA.MB-231RFP cells were subcutaneously injected into the flank region of NSG mice. Tumor dimensions were measured every three days and tumor volumes (mm3) were calculated by ((length x width2)/2). On day 14 post-transplantation, mice were administered 100?g of antibody (4.5mg/kg) by intraperitoneal injection twice weekly. On day 27 post-transplantation, mice were anesthetized with avertin and lungs were perfused with ice cold PBS. Lungs and tumors were excised. Final tumor volumes  33 were calculated and tumors were weighed. Lungs were minced and digested in a solution of collagenase (2mg/ml; Sigma, St Louis, MO) and 2mg/mL dispase (2mg/ml; Roche, Mannheim, GER) in HBSS for one to two hours at 37?C to obtain a single cell suspension. The cellular suspension was then filtered through a 70?m cell strainer, centrifuged at 453 x g for five minutes, and red blood cell lysed in ammonium chloride solution (MRCRB). Cells were washed in PBS, filtered and resuspended in FACS buffer (2% FBS, 2mM EDTA, PBS, 0.05% NaAzide) prior to flow cytometry. The PE channel was used to detect RFP fluorescing MDA.MB-231 cells in the lung. A LSRII flow cytometry machine (BD Biosciences, Mississauga, ON) was used for all flow cytometry experiments.   2.17 Statistical analysis All data was expressed as means ? standard error of the mean (SEM). Statistical analysis was performed using Prism 5 (GraphPad Software). For comparisons Student?s t test was used. For time-dependent studies two-way ANOVA was used. P-value < 0.05 was considered to be statistically significant. P < 0.05*, P < 0.01**, P < 0.001***      34 CHAPTER 3: RESULTS 3.1 Rationale Like many solid tumors, death from breast cancer most often occurs as a result of cells that escape the confines of the primary tumor and metastasize to distant tissues. For this reason, the ability to target and eliminate systemic disease is essential for the realization of truly curative therapies. While examples of such adjuvant therapies have been developed for several types of breast cancer (e.g. hormone receptor positive (HR+) and HER-2+ tumors), triple negative breast cancers (TNBCs) are typically very challenging to treat due the lack of adjuvant therapies available and the highly aggressive nature of this particular cancer subtype (as reviewed in the Introduction (1.1.1)). My overall objectives were to delineate the mechanisms underlying primary breast tumor progression and metastasis and to identify molecular targets for the treatment of systemic breast cancer, particularly TNBCs, as this cohort of patients would greatly benefit from the characterization of additional therapeutic targets.  In a seminal study by Somasiri et al., podocalyxin expression on primary tumors was found to be an independent predictor of breast cancer progression, metastasis, and poor prognosis (97). Experiments using MCF-7 cells engineered to express high levels of murine podocalyxin (MCF-7Podxl) demonstrated that podocalyxin expression results in increased migration and invasion, altered morphogenesis, and disrupted cell-cell and cell-ECM contacts in these cells (92, 97, 105). Based on these studies and additional publications showing that podocalyxin expression on primary human epithelial tumors positively correlates with tumor malignancy, I hypothesized that podocalyxin promotes tumor progression and metastasis.   To test this hypothesis, I used two approaches. First, in a collaborative effort, using a highly aggressive human breast cancer cell line we silenced podocalyxin expression to study tumor growth and metastasis in vivo. For these experiments, we selected the breast cancer cell line MDA.MB-231; a triple negative (ER, PgR and HER-2 negative), basal-like breast cancer line that is highly invasive and efficiently forms metastases in immunocompromised mice (124). MDA.MB-231 cells express a high level of podocalyxin compared to breast  35 cancer cell lines that are less aggressive (e.g. MCF-7 cells or T47-D cells) (97). We complemented these studies by silencing podocalyxin expression in a mouse breast cancer line so we could perform similar in vivo experiments in an immune competent host. I also used the MCF-7Podxl cells previously used by our lab to confirm and add to my findings in a distinct human breast cancer cell line (92, 96, 97, 116). Secondly, I used a novel human podocalyxin-binding antibody to determine if targeting podocalyxin expressed on tumor cells could attenuate tumor growth and metastasis.   3.2 Podocalyxin is highly expressed in aggressive breast tumor cell lines To begin my study, I assessed the expression level of podocalyxin in four distinct breast cancer cell lines MDA.MB-231, SUM-149, MCF-7, and T47D. I first confirmed that the highly invasive and metastatic cell line MDA.MB-231 expresses high levels of the PODXL transcript. MDA.MB-231 cells express 4-fold more PODXL than SUM-149, MCF-7, or T47D cells (Figure 3-1A). Using Western blot, I found that MDA.MB-231 cells have the highest level of translated podocalyxin protein in whole cell lysate whereas SUM-149 cells express much less. MCF-7 and T47D lysates have little or no podocalyxin protein (Figure 3-1B). Since membranous (rather than cytoplasmic) expression of podocalyxin has been shown to more closely correlate with tumor progression in human patients (113, 114, 125), I also assessed the level of surface expressed podocalyxin protein by flow cytometry. SUM-149 cells express the highest level of surface podocalyxin with a mean fluorescence intensity (MFI) value of 192 followed by MDA.MB-231 cells (MFI = 70.4). MCF-7 and T47D cells express very low levels of surface podocalyxin with MFI values of 17.6 and 16.6, respectively (Figure 3-1C).     36  Figure 3-1. Breast cancer cell lines express levels of podocalyxin, which positively correlate with cell line aggressiveness. (A) PODXL gene expression relative to GAPDH in MDA.MB-231, SUM-149, MCF-7, and T47D breast cancer cell lines as determined by qPCR. (B) Whole cell lysates (40?g) from MDA.MB-231, SUM-149, MCF-7, and T47D breast cancer cells were resolved by SDS-PAGE on a 10% gel and analyzed by western blotting using antibodies against podocalyxin (clone 3D3) and GAPDH. (C) MDA.MB-231, SUM-149, MCF-7, and T47D cells were labeled with unconjugated goat-anti-human podocalyxin antibody followed by chicken-anti-goat AF647 secondary antibody and analyzed by flow cytometry to detect the surface expression of podocalyxin on these breast carcinoma cell lines. Data presented as histograms showing podocalyxin expression on MDA.MB-231 (orange), SUM-149 (blue), MCF-7 (black), and T47D (green) cells relative to normal goat IgG isotype control.   37 3.3 Podocalyxin can be efficiently silenced in MDA.MB-231 cells using shRNA mediated lentiviral infection Because MDA.MB-231 cells endogenously express high levels of podocalyxin, and numerous studies have shown that this cell line is highly invasive both in vitro and in vivo, we selected MDA.MB-231 cells to determine if podocalyxin functions in the regulation of the migratory and invasive functions of this cell line. First, in order to track MDA.MB-231 cells in vivo, we generated two independent MDA.MB-231 cell lines ? one set expressing RFP (MDA.MB-231RFP) and the other GFP (MDA.MB-231GFP). Next, we transfected either control (Scr-ctrl) or PODXL-silencing (PODXL-KD) shRNA into MDA.MB-231RFP or MDA.MB-231GFP cells. I found that PODXL-KDRFP cells express half the concentration of PODXL transcript as Scr-ctrlRFP cells. Whereas, PODXL-KDGFP cells express approximately one-third less PODXL than Scr-ctrlGFP cells (Figure 3-2A). Total podocalyxin protein was greatly diminished in PODXL-KD infected cell lines compared to Scr-ctrl cells (Figure 3-2B). Finally, to comprehensively assess the knockdown efficiency in these cells, I performed flow cytometry to assess the level of cell-surface podocalyxin. Podocalyxin is most efficiently attenuated in PODXL-KDGFP cells, which have 3-fold less cell surface podocalyxin compared to Scr-ctrlGFP control cells (Figure 3-2C). Likewise, podocalyxin expression in PODXL-KDRFP cells is reduced by approximately 2-fold compared to Scr-ctrlRFP control cells (Figure 3-2C). Podocalyxin expression is silenced for up to seven days in vivo without puromycin selection although podocalyxin surface expression begins to return in three days in the absence of drug selection. Consequently, in the absence of drug-selection podocalyxin expression in PODXL-KD cells returns to Scr-ctrl cell levels after 14 days in culture (Figure 3-3). Thus, stable silencing of PODXL in MDA.MB-231 cells requires continuous selection.     38  Figure 3-2. Podocalyxin expression can be efficiently knocked down in MDA.MB-231 cells. (A) PODXL gene expression relative to GAPDH in Scr-ctrl and PODXL-KD MDA.MB-231 as determined by qPCR (data shown as mean ? SEM). (B) Scr-ctrl and PODXL-KD MDA.MB-231 whole cell lysates were resolved by SDS-PAGE and analyzed by western blotting using antibodies against podocalyxin (clone 3D3). An antibody to detect ?-actin was used as a loading control. (C) Scr-ctrl and PODXL-KD MDA.MB-231 cells were labeled with unconjugated goat-anti-human podocalyxin antibody followed by chicken-anti-goat AF647 secondary antibody and analyzed by flow cytometry to detect surface expression of podocalyxin. Data presented as histograms showing podocalyxin expression on Scr-ctrlGFP cells compared to PODXL-KDRFP MDA.MB-231 cells (left histogram) and Scr-ctrlRFP compared to PODXL-KDGFP MDA.MB-231 cells (right histogram) relative to normal goat-IgG isotype control (dotted line).   39  Figure 3-3. Podocalyxin expression returns after 14 days when removed from antibiotic selection in vivo. Flow cytometry was performed using an antibody to detect podocalyxin expression on tumor cells within lungs isolated from NSG mice 3, 7, and 14 days post-intravenous injection with a 50:50 mixture of 2 x 105 Scr-ctrl and PODXL-KD MDA.MB-231 cells. Upper panel: Histograms displaying the level of surface podocalyxin expression in Scr-ctrlRFP (solid line) cells compared to PODXL-KDGFP (dashed line) cells from day 0 to day 14. Lower panel: Histograms displaying the level of surface podocalyxin expression in Scr-ctrlGFP cells (solid line) compared to PODXL-KDRFP (dashed line) cells from day 0 to day 14. The dotted line shown in Day 0 histograms represents the normal goat-IgG isotype control.          40 3.4 Podocalyxin expression has no effect on monolayer proliferation, but is critical for anchorage-independent growth.  First, to determine whether silencing podocalyxin in the aggressive breast tumor cell line MDA.MB-231 has an effect on behavior in vitro, I assessed the effect of podocalyxin expression on MDA.MB-231 proliferation under normal adherent growth conditions. I found that, in monolayer cell culture, there is no difference in proliferation between Scr-ctrl and PODXL-KD cells at 48, 72, and 96 hours after initial plating (Figure 3-4). Next, I assessed the effect of podocalyxin on anchorage-independent growth using both MammoCult? culture medium and methylcellulose semi-solid culture conditions. I found that Scr-ctrl cells are able to form tumorspheres with significantly greater efficiency than PODXL-KD cells when grown in suspension culture. Scr-ctrlGFP cells (9.8 ? 1.4%) form tumorspheres 3-fold more efficiently compared to PODXL-KDGFP cells (2.9 ? 0.6%). Similarly, I observed a 22-fold increase in tumorsphere forming efficiency of Scr-ctrlRFP cells (17 ? 2%) compared to PODXL-KDRFP cells (0.76 ? 0.11%) (Figure 3-5A). I did not observe a significant difference in tumorsphere morphology (Figure 3-5B) or size (Figure 3-5C) in Scr-ctrl and PODXL-KD spheroids. Sphere-forming efficiency of PODXL-KD and Scr-ctrl cells was confirmed using methylcellulose semi-solid culture conditions.  This was important since previous studies have suggested that podocalyxin alters the adhesive properties of cells.  Under semi-solid culture conditions I observed similar reductions in sphere-forming efficiency and thus these results cannot be explainable by aggregate formation in suspension. To complement these results I assessed tumorsphere formation in MCF-7 cells overexpressing murine podocalyxin using MammoCult? liquid culture. Forced expression of podocalyxin in MCF-7 cells resulted in a 1.4-fold increase in tumorsphere formation in MCF-7Podxl cells (2.5 ? 0.01%) compared to MCF-7pIRES control cells (1.8 ? 0.1%) (Figure 3-5D).  Next, I assessed whether podocalyxin expression has an effect on self-renewal of the tumorsphere forming population of MDA.MB-231 cells. To determine whether podocalyxin expression is required to maintain the sphere-forming population of cells after continuous culture under anchorage-independent conditions, I serially passaged tumorspheres starting with initial PODXL-KD and Scr-ctrl cultures. I found that continuous passage of suspension cultures enriches for a population of cells with enhanced tumorsphere forming potential  41 (Figure 3-6). Serially passaged PODXL-KD spheres continue to have a significantly reduced ability to form tumorspheres compared to Scr-ctrl cells across passages. From initial plating to the first passage (P0 ?? P1), I observed a non-significant increase in Scr-ctrlRFP tumorsphere forming efficiency (7.7 ? 1.0% (P0) to 8.8 ? 0.8% (P1)). However, there was a slight but significant increase in tumorsphere forming efficiency in PODXL-KDRFP cells (1.8 ? 0.08% (P0) ?? 2.6 ? 0.2% (P1)). Between P1 and P2 there was a significant increase in tumorsphere formation. Tumorsphere forming efficiency of Scr-ctrlRFP cells increased 2.8-fold (8.8 ? 0.8% (P1) ?? 25 ? 2% (P2)). Likewise, I observed a 4.3-fold increase in tumorsphere forming efficiency of PODXL-KDRFP cells (2.6 ? 0.2% (P1) ?? 11 ? 2% (P2)). Thus, podocalyxin expression does not appear to impact self-renewal of the tumorsphere forming population of MDA.MB-231 cells. Despite this, podocalyxin expression still confers a significant advantage to the ability of breast tumor cells to grow under anchorage independent conditions. I performed flow cytometry to assess whether tumorsphere culture selected for those cells that retained podocalyxin expression in the PODXL-KD cultures. Podocalyxin remained silenced in these spheres compared to Scr-ctrl spheres throughout culture in the presence of puromycin selection (data not shown).         42  Figure 3-4. Proliferation is not effected by podocalyxin expression in MDA.MB-231 cells in monolayer culture. A MTS proliferation assay was performed on Scr-ctrl and PODXL-KD MDA.MB-231 cells 48, 72, and 96 hours after initial seeding. Proliferation was quantified by the amount of formazan product detected at 490nm absorbance using a microplate reader. Level of proliferation of Scr-ctrlRFP compared to PODXL-KDRFP cells over time. (non-significant (n.s) by two-way ANOVA). All values graphed as mean ? SEM.   43   Figure 3-5. Podocalyxin expressing MDA.MB-231 cells form tumorspheres more efficiently when grown in suspension culture. (A) Scr-ctrl and PODXL-KD MDA.MB-231 cells were cultured in suspension in 0.9% MethoCult? in complete MammoCult? media in triplicate. After seven days of non-adherent culture, tumorspheres were counted and represented as tumorsphere forming efficiency. (B) Images of Scr-ctrl and PODXL-KD MDA.MB-231 tumorspheres. (C) Tumorsphere size calculated using ImageJ software (pixels). (D) MCF-7pIRES control and MCF-7Podxl cells were cultured in suspension in MammoCult? media for seven days. Tumorspheres were manually counted and results reported as tumorsphere forming efficiency. All values graphed as mean ? SEM.   44  Figure 3-6. Podocalyxin expression has no effect on self-renewal by serial passage of tumorsphere cultures. Scr-ctrlRFP and PODXL-KDRFP MDA.MB-231 cells were resuspended in MammoCult? media and grown as tumorspheres and passaged every 7 days. The number of tumorspheres were manually counted and reported as tumorsphere forming efficiency. All values graphed as mean ? SEM.             45 3.5 Loss of podocalyxin expression increases adhesion to different extracellular matrices. Generally, podocalyxin functions as an anti-adhesin through its highly glycosylated and negatively-charged, extracellular domain (76, 109). To determine the effect of podocalyxin expression on MDA.MB-231 cell adhesion, I performed in vitro adhesion assays by assessing time-dependent adhesion of Scr-ctrl and PODXL-KD MDA.MB-231 cells to a number of ECM substrates. Overall, I found that PODXL-KD cells adhere with greater efficiency to a number of substrates when compared to Scr-ctrl cells (Figure 3-7). Specifically, after 15 minutes, PODXL-KDGFP cells adhere to uncoated, [low] and [high] Matrigel? coated, and 5?g/cm2 collagen coated wells with significantly greater affinity than Scr-ctrlGFP cells. There was also a non-significant trend towards greater adhesion of PODXL-KDGFP cells to 10?g/cm2 collagen coated wells after 15 minutes (Figure 3-7A). After 30 minutes, I did not observe a significant increase in PODXL-KDGFP adhesion to any of the substrates tested, although there was a trend towards greater PODXL-KDGFP adhesion at this time-point for all conditions (Figure 3-7B). After 60 minutes, I observed a significant increase PODXL-KDGFP cell adhesion to 10?g/cm2 collagen coated wells. There was also a trend towards increased adhesion in the PODXL-KDGFP cells on uncoated, [low] and [high] Matrigel?, and 5ug/cm2 collagen wells (Figure 3-7C).    46   Figure 3-7. Podocalyxin facilitates a reduction in cell-ECM adhesion. Cell adhesion assays were performed on Scr-ctrlGFP and PODXL-KDGFP MDA.MB-231 cells by assessing level of adhesion to uncoated, [low] 1/10 Matrigel, [high] 1/5 Matrigel, 5?g/cm2 collagen and 10?g/cm2 collagen coated wells of a 96 well plate. Data is represented as absorbance of crystal violet released from the cells at 570nm. All conditions were performed in triplicate. (A) Absorbance measured 15 minutes after initial plating (B) Absorbance measured 30 minutes after initial plating (C) Absorbance measured 60 minutes after initial plating. All values graphed as mean ? SEM.   47 3.6 Podocalyxin expression promotes primary tumor formation To determine the effect of podocalyxin expression on primary tumor growth in vivo, I subcutaneously injected Scr-ctrl or PODXL-KD MDA.MB-231 cells into the flank of immunocompromised NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice. NSG mice lack B, T, and Natural killer (NK) cells and thus are unable to reject xenografted human tumor cells (126). Palpable solid tumors were detectable eight days after injection of Scr-ctrlRFP MDA.MB-231 cells and continued to rapidly increase in volume over time. Primary tumor development was significantly reduced in mice injected with PODXL-KDRFP MDA.MB-231 cells. Injection of PODXL-KDRFP cells resulted in the formation of solid tumors, which were < 250mm3 in volume by day 21 whereas Scr-ctrlRFP tumors reached sizes > 750mm3 in volume (Figure 3-8A). I obtained similar results in an independent experiment using Scr-ctrlGFP and PODXL-KDGFP MDA.MB-231 cells over a time course of 25 days. At sacrifice PODXL-KDGFP tumors were < 250mm3 whereas Scr-ctrlGFP tumors were > 500mm3 (Figure 3-8B and 3-8C). Between 21 and 37 days post-subcutaneous injection of tumor cells, PODXL-KD tumors begin to develop more rapidly correlating with my observation that expression of podocalyxin returns after approximately 14 days growth in vivo. Thus, as podocalyxin expression returns, primary tumor volume begins to steadily increase (data not shown). To more comprehensively assess the differences in tumor size between the Scr-ctrl and PODXL-KD tumors, I weighed the excised tumors immediately after excision (Figure 3-8C) and found that the Scr-ctrl solid tumors (0.73 ? 0.11 g) were 2.6-fold bigger than the PODXL-KD tumors (0.28 ? 0.06 g) (Figure 3-8D). In addition to dramatic differences in primary tumor growth and development, I found that PODXL-KD tumors were highly encapsulated and only minimally invaded the surrounding skeletal musculature based on histological staining of sections from the formalin-fixed and paraffin embedded tumors from mice sacrificed on Day 21 post-transplant (Figure 3-8E).         48 3.7 Podocalyxin expression promotes metastasis.  Following sacrifice of NSG mice bearing subcutaneous MDA.MB-231 tumors, I used flow cytometry to assess the effect of podocalyxin on lung metastasis. Although I did not observe any visible tumor nodules on the lungs I was able to detect a significant number of MDA.MB-231 cells in the lungs by flow cytometry based on expression of RFP and GFP. Moreover, I found a 12-fold increase in tumor cell numbers in the lungs of mice (e.g., micrometastases) with established Scr-ctrl derived primary tumors compared to mice with PODXL-KD derived primary tumors. Specifically, I found 3200 ? 770 Scr-ctrl tumor cells per million lung cells, whereas there were only 270 ? 160 PODXL-KD tumor cells detectable per million lung cells (Figure 3-8F).                               49 Figure 3-8     50 Figure 3-8. Podocalyxin expression promotes primary tumor formation, local invasion, and lung metastasis. A total of 1 x 106 Scr-ctrl or PODXL-KD MDA.MB-231 cells were subcutaneously injected into the flank of NSG mice. (A) Primary tumors resulting from Scr-ctrlRFP and PODXL-KDRFP MDA.MB-231 cells were monitored over 21 days and the length and width of the tumors was measured manually. Tumor volumes (mm3) were calculated by the formula: ((length x width2) / 2) and plotted against time (days) (p < 0.001 by two-way ANOVA; n = 5 for Scr-ctrlRFP group and n = 6 for PODXL-KDRFP group). (B) Primary tumor development resulting from Scr-ctrlGFP and PODXL-KDGFP MDA.MB-231 cells was monitored over 25 days and the length and width of the tumors was measured manually and volumes were plotted against time (days) (p = 0.003 by two-way ANOVA; n= 6 for both groups). (C) Representative digital images showing size of excised Scr-ctrl and PODXL-KD tumors. (D) Weight (g) of pooled Scr-ctrl and PODXL-KD tumors measured immediately after excision (p = 0.02 by unpaired Student?s t test; n = 11 and n = 5 respectively). (E) Representative 4X and 10X objective images of H&E stained Scr-ctrl and PODXL-KD primary. (F) Flow cytometry was performed on lung digests to detect the number of fluorescent tumor cells residing in the lungs. Data reported as the number of fluorescent tumor cells per 106 lung cells (p = 0.02 by unpaired Student?s t test; n = 11 and n = 5 respectively). All values graphed as mean ? SEM.                             51 3.8 Podocalyxin expression in MDA.MB-231 cells is critical for primary tumor growth and metastasis To assess primary tumor development in a competitive manner, I subcutaneously injected a 50:50 mixture Scr-ctrlRFP and PODXL-KDGFP MDA.MB-231 cells into the flank of NSG mice. When tumors were 250 - 500mm3 in volume (day 18), I resected the primary tumors surgically and analyzed the cellular composition by flow cytometry. The resected tumors had 4-fold more Scr-ctrlRFP cells than PODXL-KDGFP cells. Scr-ctrlRFP cells made up 79.6 ? 2.6% of the primary tumor, whereas PODXL-KDGFP MDA.MB-231 cells made up only 20.4 ? 2.6% (Fig 3-9A). Re-growth of palpable tumors was observed at the site of the resection 36 days after surgery in the remaining four mice. Flow cytometric analysis of these tumors revealed that, like the initial tumor, Scr-ctrlRFP cells made up the bulk of the tumor (90 ? 8%) compared to PODXL-KDGFP cells (10 ? 8%) (Figure 3-9B). Next, I assessed lung micrometastases in tumor-bearing mice that had experienced regrowth and found a non-significant trend towards a higher percentage of Scr-ctrlRFP cells over PODXL-KDGFP cells in the lung. Although not statistically significant, there were 80 ? 14% Scr-ctrlRFP cells residing in the lungs at this time point, whereas PODXL-KDGFP cells comprised 20 ? 14% of this population (Figure 3-9C).   To confirm that podocalyxin expression enhances primary tumor development in a more physiologically relevant location, I orthotopically implanted a 50:50 mixture of GFP or RFP tagged Scr-ctrl and PODXL-KD MDA.MB-231 cells into the mammary fat pad of NSG mice. After 23 days, mice were sacrificed and used flow cytometry was used to determine the prevalence of each cell type within the primary tumor as well as the lung. The results confirmed the previous findings using subcutaneous transplant of tumor cells. At resection 78.1 ? 8.4% of the primary tumor is comprised of Scr-ctrl tumor cells, whereas only 21.9 ? 8.4% of the tumor is made up of PODXL-KD cells (Figure 3-10A). In addition, I found that the Scr-ctrl MDA.MB-231 cells migrated to the lung in higher numbers, whereby 2.5-fold more of Scr-ctrl MDA.MB-231 cells are detectable in the lung 23 days post-transplant compared to PODXL-KD cells. Scr-ctrl cells comprise 71.6 ? 11.8% of the lung micrometastases, whereas PODXL-KD cells make up 28.4 ? 11.8%, however these values did not reach statistical significance (Figure 3-10B).  52  Figure 3-9. Podocalyxin expression is critical for local primary tumor recurrence and micrometastasis in MDA.MB-231 cells. A 50:50 mixture of Scr-ctrlRFP and PODXL-KDGFP cells were subcutaneously injected into the flank of NSG mice (n = 8). (A) Graph represents percent Scr-ctrlRFP cells and PODXL-KDGFP cells of total MDA.MB-231 tumor cell population within the resected primary tumor by flow cytometry (paired Student?s t test p = 0.016; n = 4). (B) Graph represents percent Scr-ctrlRFP cells and PODXL-KDGFP cells of total MDA.MB-231 tumor cell population in excised tumors after regrowth by flow cytometry (paired Student?s t test: p < 0.001; n= 4). (C) Graph represents percent Scr-ctrlRFP cells and PODXL-KDGFP cells of the total MDA.MB-231 tumor cell population within the lungs by flow cytometry (paired Student?s t test: p = 0.12; n = 4). All values graphed as mean ? SEM.                     53  Figure 3-10.  Podocalyxin expression is critical for orthotopic tumor development in the mouse mammary fat pad. A 50:50 mixture of 1 x 106 GFP or RFP tagged Scr-ctrl and PODXL-KD cells were injected into the mammary fat pads of NSG mice (n = 7). (A) Percent Scr-ctrl cells and PODXL-KD cells of the total MDA.MB-231 tumor cell population within the excised primary tumor as determined by flow cytometry (paired Student?s t test: p = 0.015; pooled n = 7). (B) Percent Scr-ctrl cells and PODXL-KD cells of the MDA.MB-231 tumor cell population within the lungs as determined by flow cytometry (paired Student?s t test: p = 0.11; n = 7). All values graphed as mean ? SEM.                        54 3.9 Podocalyxin expression enhances metastatic potential at late stages of metastasis, but not initial seeding of tumor cells in the lung.  I used an experimental model of metastasis to further assess whether podocalyxin expression has a measurable effect on metastatic potential, irrespective of the presence of primary tumor. To assess early survival of tumor cells or ?micrometastasis? in the lung, a 50:50 mixture of PODXL-KD and Scr-ctrl MDA.MB-231 cells were intravenously injected into the lateral tail vein of NSG mice. Recipient mice were sacrificed in three groups at 3, 7, and 14 days post-injection. There was no significant difference in the ratio of Scr-ctrl cells to PODXL-KD cells recovered from the lung at 3 and 7 days post-injection. However, by day 14, I observed a statistically significant increase in the proportion of Scr-ctrl cells colonizing the lung compared to PODXL-KD MDA.MB-231 tumor cells (Figure 3-11).   To assess the effect of podocalyxin expression on late stages of metastasis, I allowed metastatic lesions to form in NSG mouse lungs over six-weeks. Upon sacrifice of the mice, I quantified tumor burden by enumerating fluorescent (RFP or GFP) metastatic nodules on the surface of the lungs (Figure 3-12A). On average I found 87.1 ? 16.6 Scr-ctrl tumor nodules and 38 ? 17 PODXL-KD tumor nodules ? a 2.3 fold difference (Figure 3-12C). Upon necropsy, by fluorescent microscopy I also observed large cancerous lesions in the livers of mice. Interestingly, all visible liver metastases were derived from Scr-ctrl cells (Figure 3-12B). Subsequent flow cytometry of single-cell suspensions derived from the lungs, liver, and bone marrow further revealed a dramatic reduction in the number of PODXL-KD MDA.MB-231 cells colonizing these tissues. In the lung, there are 4.7-fold fewer PODXL-KD cells compared to Scr-ctrl tumor cells. Despite being introduced into circulation in equivalent numbers, Scr-ctrl cells represented 83 ? 2.7% of the population of tumor cells in the lung compared to 18 ? 2.7% PODXL-KD cells (Figure 3-12D). In the liver, there was a 20-fold reduction in the contribution of PODXL-KD cells (4.7 ? 1.2%) compared to Scr-ctrl tumor cells (95 ? 1.2%) (Figure 3-12E). Finally, in the bone marrow, 92 ? 3.5% of the tumor cells detected were Scr-ctrl cells and 8.0 ? 3.5% are PODXL-KD cells (Figure 3-12F).   Based on the results obtained using experimental models of metastasis, I conclude that podocalyxin has no observable role in early survival of disseminated tumor cells in  55 circulation or in the lung since equal proportions of Scr-ctrl and PODXL-KD MDA.MB-231 cells persist in the lungs for the first seven days after being introduced into circulation. However, podocalyxin-expressing MDA.MB-231 cells have a significantly greater ability to develop into overt metastatic lesions in the lung and other tissues, possibly reflecting a defect in extravasation from the blood vessels into the parenchyma of the lung or a requirement for colonizing a supportive tumor niche.  I note that the highly immunocompromised NSG mouse strain allowed for a greater number of podocalyxin knockdown MDA.MB-231 cells to form metastases in the lung upon intravenous injection. Although we obtained similar results using SCID mice, in my hands, NSG mice require injection of 8-fold fewer MDA.MB-231 cells than SCID mice to generate metastasis in multiple organs. Based on my observation that MDA.MB-231 tumor cells, including PODXL-KD tumor cells, engraft more efficiently in NSG mice than SCID mice (data not shown), we suspect that NK cells are an important in the clearance of MDA.MB-231 tumor cells.              56  Figure 3-11. Podocalyxin expression has no effect on initial tumor seeding in the lung using a model of experimental metastasis. A 50:50 mixture of 2 x 105 fluorescent Scr-ctrl and PODXL-KD MDA.MB-231 cells were intravenously injected into the tail vein of NSG mice. Mice were sacrificed 3, 7, and 14 days post-injection (n = 6). Graph represents the percent Scr-ctrl and PODXL-KD of the total population of MDA.MB-231 cells within the lungs as determined by flow cytometry. All values graphed as mean ? SEM.                            57 Figure 3-12    58 Figure 3-12. Podocalyxin expression increases the metastatic burden of the lungs, liver, and bone at late stages of metastasis. A 50:50 mixture of 5 x 104 fluorescent Scr-ctrl and PODXL-KD MDA.MB-231 cells were intravenously injected into the tail vein of NSG mice. Mice were sacrificed six weeks post-injection (A) Upper left panel: fluorescent images of lungs taken using the GFP-channel of a dissecting microscope show PODXL-KDGFP tumor nodules (n = 5). Lower left panel: fluorescent images of lungs taken using the RFP-channel to detect Scr-ctrlRFP tumor nodules. Upper right panel: fluorescent images of excised lungs taken using the GFP-channel to detect Scr-ctrlGFP tumor nodules (n = 5). Lower right panel: fluorescent images of lungs taken using the RFP-channel to detect PODXL-KDRFP tumor nodules. (B) Upper left panel: fluorescent images of the liver taken using the GFP-channel to detect PODXL-KDGFP liver metastases (n = 5). Lower left panel: fluorescent images of the liver taken using the RFP-channel to detect Scr-ctrlRFP liver metastases. Upper right panel: fluorescent images of the liver taken using the GFP-channel to detect Scr-ctrlGFP metastases (n = 3). Lower right panel: fluorescent images of the liver taken using the RFP channel to detect PODXL-KDRFP metastases. (C) Number of Scr-ctrl and PODXL-KD tumor nodules manually counted on the surface of the lungs (p < 0.001 by paired student?s t test; pooled n = 10). (D) Percent Scr-ctrl and PODXL-KD tumor cells of the total population of tumor cells within the lungs as determined by flow cytometry (p < 0.001; pooled n = 10). (E) Percent Scr-ctrl and PODXL-KD tumor cells of the total tumor population of tumor cells within the liver as determined by flow cytometry (p < 0.001 by paired student?s t test; pooled n = 10). (F) Percent Scr-ctrl and PODXL-KD tumor cells of the total tumor cell population within the bone marrow isolated from both femur and tibias as determined by flow cytometry (p < 0.001 by paired student?s t test; pooled n = 10). All values graphed as mean ? SEM.                        59 3.10 Podocalyxin expression is also critical in a syngeneic model of metastasis To assess the function of podocalyxin expression in a syngeneic model of tumor metastases, I used the highly aggressive 4T1 mouse mammary tumor cell line. 4T1 cells endogenously express podocalyxin and efficiently metastasize to organs typically affected in systemic breast cancer (127). One of the major advantages of syngeneic mouse models of cancer progression is that tumorigenesis can be assessed in the presence of an intact immune system; therefore, this model can more closely reproduce the natural progression of tumor development. First, we silenced Podxl in 4T1 cells using a Podxl-targeted shRNA vector that also stably expresses luciferase (Podxl-KD 4T1-luc cells) and I determined that surface expression of podocalyxin was silenced by 6.8 fold (Podxl MFI 46 ?? 6.7) in Podxl-KD 4T1-luc cells (Figure 3-13A). Transplant of Podxl-KD 4T1-luc cells into the mammary fat pad of Balb/c mice did not yield significantly different primary tumor volumes compared to vector control transfected 4T1 cells (VC 4T1) (data not shown). To determine whether expression of podocalyxin plays a role in metastatic potential of 4T1-luc cells, I intravenously injected VC or Podxl-KD 4T1-luc cells into the lateral tail veins of BALB/c mice. By manually counting the number of visible tumor nodules on the surface of the lungs, I observed a significant decrease (6.5-fold fewer) in the number of visible Podxl-KD tumor nodules (10 ? 3) compared to VC tumor nodules (65 ? 7) (Figure 3-13B). In addition to manually counting tumor nodules on the surface of the lungs, I used a luciferase assay as a surrogate method to assess the frequency of Podxl-KD and VC 4T1-luc cells in the lungs. Using this method, I found that luciferase activity was reduced by 2.5-fold in the lungs of mice intravenously injected with Podxl-KD cells compared to VC 4T1-luc cells (Figure 3-13C). I conclude that, in 4T1 cells, expression of podocalyxin enhances metastatic potential in a syngeneic model of lung metastasis, but does not appear to have a role in primary tumor growth in subcutaneous or mammary fat pad transplant models.         60  Figure 3-13. Podocalyxin expression enhances metastasis of 4T1 mammary tumor cells to the lung in a syngeneic model of experimental metastasis. (A) VC (black line) and Podxl-KD (dashed line) 4T1-luc cell mouse mammary tumor cells were labeled with APC conjugated rat-anti-mouse podocalyxin antibody and analyzed by flow cytometry to detect surface expression of podocalyxin. Rat IgG2B was used as an isotype control (dotted line). Data presented as histograms showing podocalyxin expression on VC cells compared to Podxl-KD 4T1-luc cells. (B) A total of 1 x 105 VC or Podxl-KD 4T1-luc cells were intravenously injected into BALB/c mice (n = 4 and n = 3 respectively). Mice were sacrificed after 14 days. Visible tumor nodules on the lung were manually counted using a dissecting microscope (unpaired Student?s t test p < 0.01). (C) Luciferase expression in the lungs from non-injected and, VC and Podxl-KD 4T1-luc injected mice was measured using a luminometer. (unpaired Student?s t test p < 0.05) All values graphed as mean ? SEM.      61 3.11 Development of a therapeutic antibody against podocalyxin In collaboration with the Centre for Drug Research and Development (CDRD) (Vancouver, B.C.) we developed an antibody to target podocalyxin on tumor cells ? either to block its function or eliminate tumor cells. Of the candidate antibodies generated by the CDRD, we selected several antibodies with a number of favorable binding profiles based on flow cytometry screening of tumor cell lines known to highly express podocalyxin (MDA.MB-231, CAOV3, A172), tumor cell lines known to express low levels of podocalyxin (MCF-7, T47D, OVCAR10), and non-tumor derived human cells known to express podocalyxin (human umbilical vascular endothelial cells (HUVEC), and human embryonic kidney 293 (HEK293) cells). Nine candidate antibodies were selected for an initial screen using an in vivo tumor model. I first assessed the ability of the candidate antibodies to inhibit tumor formation in vivo by pre-treating MDA.MB-231 cells with experimental or control antibody in vitro prior to injection into NSG mice. The prediction was that pre-treatment with antibody would allow us to detect any effects on tumor growth in a subcutaneous transplant model. Of these nine candidate antibodies, one candidate (anti-PODO) significantly delayed primary tumor development compared to the anti-OVA control antibody (anti-OVA) (Figure 3-14A).  Pre-treatment of MDA.MB-231 cells with anti-PODO (anti-PODOPT) significantly delayed tumor development in vivo for the first twelve days post-implantation compared to tumors treated with anti-OVAPT. After twelve days in vivo, anti-PODOPT treated tumors rapidly increased in volume and by day sixteen were equal in volume to anti-OVAPT treated tumors (Figure 3-14B). I hypothesize that the late stage increase in growth of the tumors treated with anti-PODOPT is a consequence of a gradual decline in antibody concentration.   Next, we scaled up production of anti-PODO antibody and attempted to repeat my initial findings in a larger cohort of mice and also to investigate the effects of anti-PODO on established tumors when administered systemically (i.p.). To do this I used a crossover experimental treatment design as outlined in Figure 3-15A. The overall effects of antibody treatment on tumor volume starting from initial transplantation to the sacrifice of recipient mice are shown in Figure 3-15B. I found that mice bearing MDA.MB-231RFP prophylactically treated with anti-PODOPT displayed delayed tumor development within the first eleven days after transplant (Figure 3-15C). As in the prophylactic screening experiment, I observed  62 that anti-PODOPT tumors reach similar volumes to anti-OVAPT tumors by day 14, likely reflecting a time when the antibody pre-treatment titers diminish. To assess the effect of antibody treatment on the development of established tumors, mice were also systemically treated with anti-OVA control or anti-PODO twice per week starting on day 14. In this experimental design, tumors pretreated with anti-OVA antibody were subsequently treated with anti-PODO once they were firmly established (day 14). I found that treatment of established subcutaneous MDA.MB-231 tumors results in complete inhibition of tumor growth. In contrast, systemic treatment of anti-OVAPT tumor bearing mice with anti-OVA control antibody had no such inhibitory effect on primary tumor development and tumors continued to rapidly increase in volume (Figure 3-15D). Likewise, systemic treatment of anti-PODOPT tumor bearing mice with systemic anti-PODO also resulted in complete inhibition of tumor growth. Conversely, systemic treatment of anti-PODOPT tumor bearing mice with anti-OVA control antibody had no such inhibitory effect on primary tumor development and these tumors continue to rapidly increase in volume (Figure 3-15E). Thus, pre-treatment of tumor cells with anti-PODOPT (prophylactic regimen) or post-treatment of established tumors (therapeutic regimen) dramatically delays tumor growth.  Representative images of excised tumors from mice systemically treated with anti-OVA control or anti-PODO reveal a clear difference in the size of the primary tumors treated with anti-PODO (Figure 3-16A and Figure 3-16D). In addition, anti-OVAPT tumors from mice systemically treated with anti-PODO weigh 5.3-fold less (0.07 ? 0.02 g vs. 0.42 ? 0.2 g) than tumors from mice systemically treated with anti-OVA antibody (Figure 3-16B). Likewise, while not reaching statistical significance, anti-PODOPT tumors from mice systemically treated with anti-PODO, weigh 3.6-fold less (0.31 ? 0.1 g vs. 0.087 ? 0.03 g) than tumors from mice treated systemically with anti-OVA antibody (Figure 3-16E).   Finally, I used flow cytometry to assess the number of RFP-positive tumor cells that had successfully moved from the site of injection in the flank to colonize the lung. I found that the number of micrometastases within the lung positively correlates with primary tumor size. Thus, tumors with the largest volumes have the greatest number of detectable tumor cells within the lungs and therapeutic treatment with anti-PODO antibody resulted in fewer  63 detectable tumor cells in the lung. In the lungs of mice possessing anti-OVAPT tumors and systemic treatment with anti-OVA control antibody, there was an average of 750 ? 400 tumor cells per million lung cells detectable upon sacrifice. In anti-OVAPT tumors treated systemically with anti-PODO, 67 ? 19 tumor cells per million lung cells were detectable (ie greater than 10-fold less) (Figures 3-16C). In the lungs of mice pre-treated with anti-PODOPT tumors systemically treated with anti-OVA control antibody, an average of 760 ? 420 tumor cells per million lung cells were detectable upon sacrifice. Finally, within the lungs of mice pretreated with anti-PODO mice and then systemically treated with anti-PODO, only 30 ? 8 tumor cells per million lung cells were detectable (greater than 20-fold less).  While the number of tumor cells within the lungs positively correlated with podocalyxin expression these results failed to reach statistical significance ? likely due to the high variability of tumor cells detected in the lung for these experiments.  In summary, therapeutic treatment of established tumors with anti-PODO antibody results in a reduction in the number of micrometastases at distal sites (lung), which correlates directly with the size of the primary tumor at the site of injection.              64                            Figure 3-14. Pre-treatment of MDA.MB-231 cells with anti-PODO antibody significantly reduces early primary tumor development in NSG mice.  A total of 1 x 106 anti-PODOPT or anti-OVAPT MDA.MB-231RFP cells were subcutaneously injected into the flanks of NSG mice. (A) Volumes of tumors pre-treated with nine candidate anti-podocalyxin antibodies (Anti-PODO and candidate antibodies C1-8) or anti-OVA control antibody (red squares) over time (days). (B) Volumes of tumors pre-treated with anti-OVA control (red squares) or anti-PODO (black triangles) antibody from initial transplantation to sacrifice.        65 Figure 3-15       66 Figure 3-15.  Treatment with anti-PODO results in inhibition of primary tumor development. A total of 1 x 106 anti-PODOPT or anti-OVAPT MDA.MB-231 cells were subcutaneously injected into the flanks of NSG mice. On day 14 post-transplantation, 100?g of anti-OVA (anti-OVAsys) or anti-PODO (anti-PODOsys) antibody was intraperitoneally (i.p.) injected twice weekly (orange stars). (A) Experimental flow chart showing treatment schedules. (B) Tumor volumes measured over time (days) for mice with anti-OVAPT tumors followed by anti-OVAsys treatment (red), anti-PODOPT tumors followed by anti-OVAsys treatment (black), anti-OVAPT tumors followed by anti-PODOsys treatment (blue) and anti-PODOPT tumors followed by anti-OVAsys treatment (green). (C) Volume of anti-OVAPT and anti-PODOPT tumors from day five to fourteen post-transplantation (p < 0.05 by two-way ANOVA; n = 5). (D) Volume of anti-OVAPT tumors treated systemically with anti-OVA control or anti-PODO antibody from primary systemic administration to sacrifice (p < 0.01 by two-way ANOVA; n = 5). (G) Volume of anti-PODOPT tumors treated systemically with anti-OVA control or anti-PODO83 antibody from primary systemic administration of antibody to sacrifice (p < 0.01 by two-way ANOVA; n = 5).        67   Figure 3-16. Treatment with anti-PODO antibody results in smaller tumors and reduced metastasis to the lung. A total of 1 x 106 anti-PODOPT or anti-OVAPT MDA.MB-231RFP cells were subcutaneously injected into the flanks of NSG mice. Starting on day 14 post-transplantation, 100?g of anti-OVA (anti-OVAsys) or anti-PODO (anti-PODOsys) antibody was intraperitoneally (i.p.) injected into NSG mice twice weekly. (A) Representative images of anti-OVAPT tumors treated with either anti-OVAsys control or anti-PODOsys. (B) Weight (g) of anti-OVAPT tumors treated with either anti-OVAsys control or anti-PODOsys antibody (C) Number of RFP-positive tumor cells per 106 lung cells of mice with anti-OVAPT tumors systemically treated with either anti-OVA control or anti-PODO antibody as detected by flow cytometry. (D) Representative images of anti-PODOPT tumors systemically treated with either anti-OVA control or anti-PODO antibody. (E) Weight of anti-PODOPT tumors systemically treated with either anti-OVA control or anti-PODO antibody. (F) Number of RFP-positive tumor cells per 106 lung cells of mice with anti-PODOPT tumors treated with either anti-OVAsys control or anti-PODOsys as detected by flow cytometry.   68 CHAPTER 4: DISCUSSION AND CONCLUSION 4.1 Summary We have previously shown that podocalyxin expression in primary breast tumor cells is correlatively linked to poor patient survival and that podocalyxin enhances the motility and invasiveness of breast cancer cell lines in vitro (90, 92, 96, 97, 105, 122). Using in vivo models of breast tumor growth and metastasis, in this thesis I show for the first time that podocalyxin has a causal role in promoting the growth of solid tumors and enhancing metastasis of tumor cells to distant organs. Specifically, I observed that silencing podocalyxin expression in MDA.MB-231 cells, an aggressive human basal breast cancer cell line, severely impairs primary tumor growth (at subcutaneous flank and mammary fat pad transplant sites) and metastasis to the lung, liver, and bone marrow in a xenograft model using immunocompromised mice. I corroborated these results using fully immunocompetent mice by silencing podocalyxin expression in mouse mammary tumor 4T1 cells and found that podocalyxin expression also enhances metastasis (but not primary tumor growth) in this syngeneic tumor model. Experiments using the competitive lung metastasis model suggest that podocalyxin promotes establishment of a hospitable niche in the lung to foster metastatic growth. Although podocalyxin expression on MDA.MB-231 cells does not alter initial ?seeding? of cells in the lung at early time points (up to 14 days), expression of podocalyxin greatly enhances subsequent establishment of tumors in the lung and other organs. Finally, we developed and screened specific antibodies to target the tumor growth and metastasis-promoting functions of podocalyxin. I found that one antibody (anti-PODO) is able to inhibit MDA.MB-231 growth and dissemination up to two weeks following a single prophylactic treatment and, importantly, can inhibit the growth and metastasis of established primary tumors upon systemic treatment. Thus, in this thesis, I have clearly established that expression of podocalyxin on breast tumor cells is sufficient to promote tumor growth, invasion of surrounding tissue and systemic dissemination. Furthermore, I have shown that it is possible to apply this knowledge to the development of a specific therapeutic antibody to target podocalyxin and block the growth and metastasis of established podocalyxin-positive tumors. Combined with applications in diagnostics, our work has the potential to improve treatments for breast cancer patients by providing a selective, potentially low-toxic adjuvant therapy.   69 Understanding the mechanisms by which podocalyxin promotes tumor growth, invasion and metastasis will maximize the clinical potential of anti-podocalyxin therapeutics. There are several processes in the metastatic cascade by which podocalyxin could promote tumor aggressiveness including: adhesion; proliferation; survival under anchorage-independent conditions; nutrient metabolism; survival in hypoxic conditions; chemotaxis and homing; interactions with the tumor microenvironment; and, tumor vascularization.  In the following sections I will consider these possibilities in the context of the data presented in this thesis.  4.2. Podocalyxin functioning as an anti-adhesive protein in tumorigenesis Typically, cancer cells are less adhesive compared to their normal tissue counterparts. In normal tissues, epithelial cells form a tight and highly organized architecture. In contrast, tumor cells generally have decreased cell-cell and cell-extracellular matrix (ECM) adhesion and this change contributes to extensive morphological and cytoskeletal abnormalities. Acquisition of motile properties corresponds with the failure of tumor cells to appropriately adhere to neighboring cells and ECM (128). Based on a multitude of publications showing that podocalyxin acts as an anti-adhesin in several normal and malignant tissues I sought to determine if podocalyxin functions as an anti-adhesive in the highly aggressive MBA.MB-231 breast cancer cell line (89, 90).   Previously it was shown that in MCF-7Podxl cells, podocalyxin expression mediates the exclusion of ?1-integrins from the free, apical surface of cells. Low levels of podocalyxin expression may actually promote efficient segregation of cell apical-basal membrane domain proteins; however if highly expressed, podocalyxin may sequester F-actin away from the basolateral surface to expand the apical membrane domain cytoskeleton and weaken basolateral adhesion (92, 97). Consistent with this model, podocalyxin expression in MDA.MB-231 (Figure 3-7) and MCF-7 cells weakens or delays adhesion of cells to ECM (96). Specifically, I found that PODXL-deficient MDA.MB-231 cells adhere to several ECM substrates more efficiently than PODXL-expressing control cells (i.e., parental MDA.MB-231 cells), particularly at early time points. Recalling that PODXL-KD cells fail to efficiently migrate from primary tumors to the lungs (Figure 3-8F), I posit that the anti-adhesive properties of podocalyxin, when highly expressed, contributes to the release of neoplastic  70 cells from the primary tumor bulk, thus facilitating metastasis (see model Figure 4-1). In addition, because membranous expression of podocalyxin is primarily observed at the invasive tumor front in tumors of colorectal, ovarian, pancreatic, and urothelial bladder cancers, it is likely that podocalyxin plays a key role in mediating collective tumor invasion by altering adhesion between the tumor and stromal environment (38, 95, 113, 114, 129).   Figure 4-1. Model of podocalyxin functioning as an anti-adhesive. Podocalyxin is expressed on the apical domain of polarized epithelia and promotes non-EMT loss of adhesion and shedding when overexpressed. Modified from McNagny et al (90).  Having confirmed that podocalyxin acts as an anti-adhesive protein in two distinct human breast cancer lines using alternative approaches (exogenous expression and gene-silencing) it would be interesting to further investigate whether podocalyxin has the ability to play a more dynamic role in adhesion, given that interactions between tumor cells in circulation and the vascular endothelium are extremely important for the extravasation of tumor cells into metastatic tissue sites. Extravasation of tumor cells into tissue from circulation can be facilitated by tethering to selectin family receptors (130). Selectins are receptors present on stimulated endothelial cells, platelets, and leukocytes that bind to a variety of O-glycosylated ligands and function to mediate transient adhesion of immune cells and tumor cells to the endothelium (131, 132). Recently, two papers showed that  71 podocalyxin modified with O-linked sialofucosyl oligosaccharides can support adhesion of colon and pancreatic carcinoma cells to E- and L- selectin (but not P-selectin) (111, 113). Consistent with our findings, podocalyxin expression in colon and pancreatic carcinoma increases the aggressive properties of tumors (113, 125, 129). This data may support a multifaceted role for podocalyxin, whereby podocalyxin is capable of acting as an anti-adhesive via its highly charged and bulky extracellular domain and, when bearing the requisite glycosylation motifs, may bind to E- and L-selectin to promote adhesion. In continuing our studies, we could use ?rolling blot? assays to determine whether podocalyxin expressed on MDA.MB-231 cells is also capable of binding to selectins. By acting as a ligand for E-selectin, podocalyxin may function to promote binding of tumor cells to the vascular endothelium and thus promote extravasation in the establishment of metastases. Additionally, as a ligand for L-selectin, podocalyxin may aid in the formation of leukocyte-tumor cell aggregates, which can form emboli and arrest in the microvasculature of distant organs (132).   Although not formally proven with the data presented in this thesis, the fact that podocalyxin is not required for seeding the lung in the competitive experimental metastasis model (Figure 3-11) suggests that adhesion to L, E-selectin is not a required component of podocalyxin?s metastasis-promoting function. However, this model, which involves a bolus injection of tumor cells, does not simulate shedding of small numbers of cells from a primary tumor into circulation. An experimental design to model the shedding from a primary tumor is complicated by the dramatically increased growth rate of podocalyxin-deficient tumors. Since metastasis from the primary tumors correlates with tumor size (Figure 3-8), I cannot conclude that podocalyxin promotes enhanced shedding or enhanced invasion from these experiments. Introduction of a greater number of PODXL-KD cells over parental cells would increase the size of PODXL-KD primary tumors and may provide a more definitive information as to whether the observed increase in lung metastasis is dependent on tumor size or if podocalyxin expressed on tumor cells facilitates shedding of tumor cells as a result of the anti-adhesive properties of podocalyxin.       72 4.3 Podocalyxin facilitates anchorage-independent growth of tumor cells I have shown in this thesis that expression of podocalyxin has no overt role in monolayer proliferation of breast tumor cells in vitro (Figure 3-4) and others have reported similar results (105). However, when MDA.MB-231 cells are cultured in anchorage-independent conditions I observed a significant increase in the number of tumorspheres when podocalyxin is expressed (Figure 3-5). Previously, the effect of podocalyxin on sphere formation under non-adherent growth conditions had been unclear. Sizemore et al. found that podocalyxin expression on MCF-7 and PC3 cells grown in semi-solid conditions had no effect on the number or size of the tumorspheres over two weeks (105). Conversely, Graves et al. observed that MCF-7Podxl cells form tumorspheres 2-fold more efficiently than parental MCF-7 cells, in addition to observing differences in sphere morphology and size after four weeks in semi-solid culture (96). One of several caveats of spheroid assays is that there is not a standardized protocol used to culture spheroids. Significant differences in the growth factor composition, volume of culture medium, cell density, culture dish surface area, and culture duration causes variability between research groups (133, 134). I suspect that such differences in spheroid culture methods are responsible for the variable data reported for the role of podocalyxin in anchorage-independent growth. My data suggests that podocalyxin is critical for growth of tumor cells in the absence of a basolateral-defining surface or other semi-solid matrix.   Anoikis is a programmed cell-death process that eliminates cells in tissues that lose contact with the extracellular matrix or neighboring cells (135, 136). Although tumors potentially release a large number of dysplastic cells into circulation, only those able to avoid anoikis and survive in anchorage-independent conditions can effectively survive in transit to form distant metastases (27). The ability for podocalyxin expressing breast tumor cells to more efficiently form spheres when cultured in suspension suggests that podocalyxin may facilitate resistance to anoikis.   Three-dimensional culture methods are commonly used to assess the relative number of tumor initiating cells (TICs) within a heterogeneous tumor population (137-140). TICs are often defined as being ?stem-like? and thus have the ability to self renew indefinitely (141,  73 142). In general, tumorsphere forming potential under non-adherent growth conditions positively correlates with the level of neoplastic growth potential in immunosuppressed mice (61, 143). I can confirm that, in this thesis, the ability of PODXL expressing tumor cells to form tumorspheres correlates with the ability of these cells to form tumors and metastasize in vivo and was a far superior predictor of malignancy in vivo than monolayer culture. I postulate that podocalyxin may function in the maintenance and/or survival of a putative TIC population responsible for tumor progression.   Podocalyxin is highly expressed on undifferentiated human embryonic stem cells (hESCs) but expression declines as hESCs move towards a more differentiated state. This observation has lead to the hypothesis that podocalyxin is a marker of ?stem-like? cells (144, 145). A recent publication from Binder and colleagues, demonstrates that podocalyxin is expressed in the population of ?stem-like? cells in glioblastoma multiforme, which are undifferentiated and grow as ?oncospheres? and that differentiation of the oncosphere population results in a significant reduction in podocalyxin expression. In addition, podocalyxin expression correlates with other markers of ?stem-like? cells including Musashi1, SOX2 and BMI1 (146). Since my data suggests that podocalyxin is critical for tumorsphere formation, further investigation of the potential role of podocalyxin in the maintenance or regulation of other common ?stemness? characteristics of tumor cells is warranted.  Self-renewal is the process by which stem cells divide either symmetrically or asymmetrically to produce more stem cells and maintain the stem cell population. Since the number of tumorspheres progressively increases in both control and PODXL-KD cultures upon serial passaging, these results suggest that podocalyxin expression is not required for self-renewal capacity. Despite finding that serial passage of spheroid cultures enriches for tumorspheres over time in both control and PODXL-KD cultures, I consistently observed fewer PODXL-KD spheres, compared to control, for each serial passage (Figure 3-6). From this, one can discern that, while podocalyxin may not completely inhibit the self renewal of tumorsphere forming cells, podocalyxin may be important for the fidelity of self renewal of MDA.MB-231 cells in suspension culture. Alternatively, since podocalyxin knockdown is not completely efficient (Figure 3-2), passaging may select for a subset of tumorsphere  74 initiating cells in the PODXL-KD line with the highest podocalyxin expression. However flow cytometric analysis of PODXL-KD tumorspheres suggests that, if this is true, then the bulk of the tumorspheres revert to the initial distribution of podocalyxin expression.    4.4 Podocalyxin as a mediator of survival under hypoxic conditions Tumor vasculature is often poorly developed: Tumor vessels are generally disorganized and leaky and are therefore inefficient at delivering oxygen and nutrients (147). For this reason, the central portion of tumors that are more than 1 to 2mm in diameter are deprived of oxygen because blood does not adequately perfuse the tumors and O2 diffusion from vessels surrounding the tumors cannot reach the central cells (148). Therefore, it is crucial that neoplastic cells adapt to a hypoxic microenvironment in order to thrive within a large primary tumor (149). Several investigators have shown that cancer cells upgrade their stem-like properties in oxygen-deprived microenvironments (150-153). Maltby et al. postulated that there might be a link between podocalyxin expression and hypoxia after finding that podocalyxin expression is dramatically upregulated in developing erythrocytes in response to anemia-associated hypoxic stress (86).  In addition, expression of podocalyxin on hematopoietic progenitor cells during embryonic development correlates closely with a period of hypoxic stress (85). Based on these observations, it would be interesting to investigate whether podocalyxin expression is upregulated on a subset of tumor cells under hypoxic conditions. There is increasing evidence that the hypoxic microenvironment within a tumor or during embryogenesis may provide essential cellular interactions and environmental signals to inhibit differentiation and preferentially maintain ?stem-like? properties of TICs or hESCs (154, 155). If podocalyxin expression is upregulated during hypoxia, it may mark the ?stem-like? tumor initiating population of cells within a tumor or even indicate that podocalyxin has a direct role in maintaining quiescence or preventing differentiation of this population.   Hypoxia induces overexpression of hypoxia-inducible factor alpha (HIF1?), a master transcription factor that activates transcription of a large number of genes, including those involved in metabolism, angiogenesis, proliferation, and metastasis (156). Hypoxia triggers a HIF-dependent switch from mitochondrial oxidative phosphorylation to aerobic glycolysis  75 to generate energy (157-159). Reliance on aerobic glycolysis is mediated by upregulation of key glycolytic enzymes and overexpression of the glucose transporter (GLUT) genes: GLUT1 and GLUT3 (160-162). Much like podocalyxin, GLUT3 expression correlates with high histological grade in endometrial and breast carcinoma (163). Podocalyxin has been shown to associate with GLUT3 in embryonal carcinoma stem cells, either directly or as a larger complex of proteins. It is possible that GLUT3 expression on embryonic stem cells and malignant cells may be regulated through interactions with podocalyxin (164). Podocalyxin may act to regulate and maintain GLUT3 expression on the ?stem-like? population of tumor cells, thus allowing these cells to generate ATP and survive in low oxygen and nutrient deprived conditions. GLUT3 expression on the TIC population of glioblastoma multiforme tumor cells suggests that GLUT3 expression confers a survival advantage to this population in nutrient-deprived microenvironments (165). Given that podocalyxin has recently been shown to also be present on glioblastoma multiforme TICs (146) and is known to associate with GLUT3 in embryonal carcinoma stem cells (164), podocalyxin may act in conjunction with GLUT3 to promote tumor progression. Although this is an intriguing hypothesis, podocalyxin and GLUT3 are currently unlinked markers on the same cell populations.    4.5 Podocalyxin promotes primary tumor development Based on studies by myself (Figure A-1) and others (96) showing that podocalyxin expression in MCF-7Podxl breast tumor cells had little or no effect on promoting primary tumor formation in immunocompromised NSG or RAG2M-/- mice, I was initially surprised to observe a remarkable difference in primary tumor formation upon knock down of podocalyxin in MDA.MB-231 cells (Figure 3-8). I attribute these results to the fact that MCF-7 and MDA.MB-231 tumor cell lines represent distinct breast cancer subtypes with differing growth rates, whereby MCF-7 cells do not grow in vivo as efficiently as MDA.MB-231 cells. This coupled with the fact that initial findings to support a role of podocalyxin in breast tumorigenesis were based on overexpression of murine podocalyxin in human breast tumor cells may have led to attenuated results obtained by Graves et al. in vivo.    76 4.6 Podocalyxin promotes tumor invasion and migration The observation that PODXL-KD MDA.MB-231 cells fail to metastasize in mice (Figure 3-8 to Figure 3-12) concords with previous studies showing that podocalyxin promotes cellular migration in vitro using wound-healing and chemotaxis assays (105, 110, 112, 122, 166). Through interactions with the intracellular binding partners, podocalyxin has been implicated in promoting migration and invasion of breast and prostate tumor cells by stimulating the PI3K-AKT and MAPK pathways (105). The podocalyxin-NHERF-ezrin axis also regulates activation of the Rho-family GTPases, RhoA and Rac1 (110, 167). Thus, through intracellular interactions with NHERF and ezrin, podocalyxin may also increase tumor cell motility. Rho GTPases are important regulators of focal adhesion assembly, cell motility, cell polarity, and cell cycle progression (168, 169). Increased Rho GTPase activity in breast cancer is associated with growth under serum deprived and anchorage independent conditions (170). A recent study by Fernandez et al. (171) suggests that interaction of the cytosolic domain of podocalyxin is necessary and sufficient to enhance migration of CHO cells and propose that podocalyxin regulates migration through interactions with Rac1 and Cdc42 GTPases. In contrast, ectodomain of podocalyxin is reported to be critical for adhesion (76, 92).  I observed that podocalyxin promotes invasion of primary MDA.MB-231 tumor cells into musculature surrounding the tumor regardless of tumor size (Figure 3-8E). This observation complements data from Graves et al., showing that MCF-7Podxl cells collectively migrate through the mammary fat pad from the primary tumor (116). Boman et al. also observed that in urothelial bladder cancer, there is a significantly greater degree of lamina propria invasion in those samples expressing a high level of podocalyxin. Tumor invasion into the lamina propria represents a serious problem when attempting to surgically treat bladder cancers, and thus podocalyxin expression may delineate patients likely to fail first-line conservative surgical treatment of the tumor (38). One mechanism of tissue invasion includes expression of matrix metalloproteinases (MMPs) to facilitate degradation of the stromal cells surrounding the primary tumor and healthy tissue. MMP1 and MMP9 are upregulated in MCF-7Podxl and PC3Podxl cells through a mechanism that requires interactions with ezrin (105). Similarly, knockdown of podocalyxin in U-87 glioblastoma cells and  77 overexpression of podocalyxin in SW1783 astrocytoma cells also result in altered expression of MMPs (112). Thus, I would expect that expression of MMPs is associated with podocalyxin expression and contributes to its invasive role in tumor progression.   4.7 Podocalyxin and the tumor microenvironment  By performing experimental metastasis assays, I found that podocalyxin expression on MDA.MB-231 cells conveys no advantage to initial colonization of the lung (between day 0 and 7). However, there was a gradual shift towards a greater proportion of PODXL-expressing tumor cells residing in the lung from day 14 onwards (Figure 3-11). One caveat of direct injection of tumor cells into the blood stream is that tumor cells become sequestered within the first capillary bed encountered upon hematogeneous dissemination, which in the case of tail vein injection is the pulmonary capillary bed (172). However, mechanical entrapment of tumor cells in the lung is not sufficient to induce development of lung metastases (173). It is accepted that tumor cells that fail to attach to the vasculature (via tumor-endothelial cell adhesion) and invade into the lung parenchyma (extravasate) will be cleared from the lung rapidly after injection (174). There is evidence to show that tumor cells within the vasculature are typically cleared within 48-72 hours (175). However, the precise timeline for extravasation of metastatic tumor cells into the tissue is largely unknown and likely is variable for different tumor cells.  There are equal proportions of parental and PODXL-KD MDA.MB-231 cells populating the lungs at 3 and 7 days; these are timepoints when I propose that cells that have not successfully extravasated into the lung parenchyma likely will have been cleared. To definitively assess whether tumor cells have extravasated at these timepoints, experiments must be performed to assess localization within the lung by fluorescence microscopy. Thus, when introduced via the hematogeneous route, podocalyxin does not convey an initial survival advantage to tumor cells in circulation, selectively direct these cells to their target organs, and likely does not enhance tumor-transendothelial adherence and extravasation ? as these steps in the metastatic-invasion cascade would normally be complete well before 7 days post-injection. Since the tissue colonizing advantage of parental MDA.MB-231 cells over PODXL-KD tumor cells is not observable until 14 days post-injection and because podocalyxin expression does not remain stably silenced in vivo beyond 14 days (Figure 3-3), It is likely that that podocalyxin  78 plays a subtle but important role early in the seeding and colonization stage of the metastatic cascade. I propose that podocalyxin directs tumor cells to a favorable microenvironment in the lung parenchyma or facilitates interactions within the pulmonary microenvironment to initiate the formation of metastatic lesions. Podocalyxin may be able to recruit or interact with inflammatory cells, vascular endothelial cells, and fibroblasts to foster tumor growth and progression. There is increasing evidence to support the notion that a subset of fibroblasts known as carcinoma-associated fibroblasts (CAFs) perform unique roles in the tumor microenvironment (176). CAFs promote tumor angiogenesis by secreting SDF1 and recruit mobilized endothelial progenitor cells (EPCs) (177). Furthermore, in addition to promoting tumor invasion through the production of diffusible factors, CAFs have been shown to lead collective invasion of carcinoma cells that retain their epithelial markers and migrate as a collective group. By secreting proteases and through mechanical force, CAFs can remodel the ECM to pave a tract for collectively invading tumor cells (178). The findings presented in this thesis showing that podocalyxin promotes tumor growth and metastasis combined with data showing that podocalyxin mediates collective migration of tumor cells (116), suggest that CAFs may contribute to the mechanism by which podocalyxin functions in breast cancer metastasis as well as other epithelial cancers.   Podocalyxin also promoted development of metastases in the liver and the bone marrow. Unpublished results from our lab suggest that podocalyxin enhances CXCR4-mediated chemotaxis of hematopoietic stem and progenitor cells (179). As the lungs, liver, and bone marrow are all prominent producers of the CXCR4 ligand, CXCL12 (also known as SDF-1), podocalyxin may act in a similar manner on tumor cells to facilitate invasion of CXCL12 rich organs (180). Additionally CAFs in the tumor microenvironment are known to secrete CXCL12 to promote growth and angiogenesis (177).  My findings using an experimental model of metastasis support a role for podocalyxin in the formation of metastases, but do not test whether it facilitates dissemination of tumor cells from a primary tumor to form distant metastases. In a metastatic model conforming more closely to human disease, I found that podocalyxin also promotes metastasis of tumor cells from solid tumors established in the subcutaneous (flank) or mammary fat pad. (Figure 3- 79 8F, Figure 3-9C, and Figure 3-10B). These experiments suggest that podocalyxin expression may play a role in tumor cell migration and colonization of the lungs.  In this model very few PODXL-KD tumor cells are detectable in the lungs even at early stages of metastatic progression. This suggests that podocalyxin may contribute to the initial release of tumor cells from an established tumor, mediate survival of circulating tumor cells, or promote migration and homing of tumor cells to the lungs. However, because PODXL-KD tumors are much smaller than control tumors, these observations may also be attributable to primary tumor size.  The results from the syngeneic model of experimental metastasis (mouse 4T1 cell injected into the tail vein of Balb/c mice) were similar to results obtained using the MDA.MB-231 cells, whereby podocalyxin enhances experimental lung metastasis (Figure 3-13). However, in this syngeneic model, I was not able to recapitulate the results we observed in xenograft solid tumor models. Specifically, podocalyxin-deficient 4T1 tumors did not grow more rapidly in the mammary fat pad of Balb/c mice. I do not know if the discrepancy in these results can be attributed to 1) differences in the properties of 4T1 and MDA.MB-231 tumors, 2) the consequence of using immunocompetent and immunodeficient mouse models or 3) both factors. Importantly, 4T1 cells lack many of the important features of human tumors, such as genetic and epigenetic complexity and heterogeneity as they are derived from an inbred mouse stain (Balb/c) (172). Similarly, there was no observable increase in primary tumor formation upon overexpression of podocalyxin in MCF-7 cells (Appendix A). Notably, 4T1 and MCF-7 are both luminal-like breast cancer cell lines that express E-cadherin, whereas MDA.MB-231 cells are classified as triple-negative, claudin-low, E-cadherin-negative basal-like tumor cell line (181, 182). Further investigation will be needed to clarify this interesting difference  4.8 Future Directions The preceding discussion outlined some of the possible avenues of investigation that may allow us to delineate the molecular mechanisms by which podocalyxin drives breast tumor progression and metastasis. One strategy to begin these follow up studies would be the re-expression of full-length and mutant podocalyxin constructs in MDA.MB-231 cells. These  80 MDA.MB-231 PODXL ?add-back? mutants could then be assayed in experimental metastasis or tumor growth models as described in this thesis. These experiments would reveal which protein domains of podocalyxin govern its influence on tumor growth or metastasis-promoting functions, as it is possible that these two functions are not coupled. Concurrently, we could use real-time microscopy imaging (e.g. Incucyte?) to determine which domains of podocalyxin enhance in vitro migration and begin to delineate the intracellular pathways regulated by podocalyxin in response to chemotactic stimuli.   It would also be interesting to determine whether podocalyxin is either selectively or more highly expressed on putative stem-like cells within the bulk population of breast tumor cells. Traditionally, these populations are identified by flow cytometry using the breast carcinoma TIC markers CD44, CD24, and EpCAM. MDA.MB-231 cells are exclusively CD44highCD24low and are nearly homogeneous. Thus they do not represent an optimal model for evaluation of TIC activity. For that reason, I propose to suppress podocalyxin expression in SUM-149 cells, a mixed luminal/basal human breast cancer cell line that express high levels of surface podocalyxin (Figure 3-1). This approach would allow us to assess any changes in the TIC population as a result of podocalyxin expression using flow cytometry, tumorsphere assays, and in vivo tumor assays. Intriguingly, preliminary studies reveal that podocalyxin is more highly expressed on the CD44highCD24high population (data not shown). Traditionally the CD44highCD24high cells are considered to be more differentiated than the CD44highCD24low population. However, stem-like populations may be more dynamic than once believed, because purportedly differentiated, non-stem-like cells are capable of being converted into stem-like cells. For example, Meyer et al. showed that in ER- and PgR-negative tumors both CD44posCD24pos and CD44posCD24neg-low cells are equally tumorigenic. In addition, the CD44posCD24neg-low population can arise from CD44posCD24pos populations. This is an indication that, at least in ER- and PgR-negative tumors, TICs may not always be restricted to the CD44posCD24neg-low population (183). In the end, functional analyses of side-by-side populations are the only informative way to address this question directly.  Finally, further insights into the function of podocalyxin may be gained by investigating the mechanism by which anti-PODO or other candidate anti-podocalyxin antibodies block tumor  81 growth and metastasis. I posit that anti-PODO antibody may inhibit angiogenesis of the primary tumor since mice treated with anti-PODO possess smaller tumors that appear to be less vascularized (by gross examination) (Figure 3-16). However, it is possible that anti-PODO may function by blocking growth signals; stimulating antibody-dependent cell cytotoxicity (ADCC) or complement-mediated cytotoxicity (CDC); or, by interfering with interactions with stromal cells (184). Depending on the mechanism, the potency of anti-PODO antibodies may be enhanced by coupling cytotoxic agents that can be delivered directly to the tumor (185). This approach would only be a viable therapy if we identify an antibody that is highly specific to tumor-expressed podocalyxin and not endogenous podocalyxin expressed by normal tissues; otherwise there is a high potential for toxic side effects in normal tissues. Given that podocalyxin is expressed on normal kidney and vascular endothelia, the safety of anti-podocalyxin antibodies will have to be thoroughly determined. My data suggests that anti-PODO antibody is a potent inhibitor of MDA.MB-231 tumor formation when used prophylactically or systemically (Figure 3-14 and Figure 3-15). Ultimately, the hope is to identify an anti-podocalyxin antibody with therapeutic potential and can be used as a frontline therapy for those patients with highly aggressive tumors. Alternatively, our more readily achievable goal is to utilize antibodies against podocalyxin diagnostically as prognostic indicator of tumor behavior, and a guide to the appropriate therapies.               82 REFERENCES  1. Olson J. Bathsheba's Breast: Women, Cancer, and History. Baltimore, M.D.: John Hopkins University Press; 2002. 2. Rakha EA, El-Sayed ME, Green AR, Lee AH, Robertson JF, Ellis IO. Prognostic markers in triple-negative breast cancer. Cancer. 2007;109:25-32. 3. Lu J, Steeg PS, Price JE, Krishnamurthy S, Mani SA, Reuben J, et al. Breast cancer metastasis: challenges and opportunities. Cancer Res. 2009;69:4951-3. 4. Gluz O, Liedtke C, Gottschalk N, Pusztai L, Nitz U, Harbeck N. Triple-negative breast cancer--current status and future directions. Ann Oncol. 2009;20:1913-27. 5. Sainsbury JR, Anderson TJ, Morgan DA. ABC of breast diseases: breast cancer. BMJ. 2000;321:745-50. 6. Rosen PP. Breast Pathology. 2 ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001. 7. Li CI, Anderson BO, Daling JR, Moe RE. Trends in incidence rates of invasive lobular and ductal breast carcinoma. JAMA. 2003;289:1421-4. 8. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology. 1991;19:403-10. 9. Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, et al. Molecular portraits of human breast tumours. Nature. 2000;406:747-52. 10. Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A. 2003;100:8418-23. 11. Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A. 2001;98:10869-74. 12. Goldhirsch A, Wood WC, Coates AS, Gelber RD, Thurlimann B, Senn HJ. Strategies for subtypes--dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol. 2011;22:1736-47. 13. De Laurentiis M, Cianniello D, Caputo R, Stanzione B, Arpino G, Cinieri S, et al. Treatment of triple negative breast cancer (TNBC): current options and future perspectives. Cancer Treat Rev. 2010;36 Suppl 3:S80-6. 14. Schnitt SJ. Classification and prognosis of invasive breast cancer: from morphology to molecular taxonomy. Mod Pathol. 2010;23 Suppl 2:S60-4. 15. Kim RS, Avivar-Valderas A, Estrada Y, Bragado P, Sosa MS, Aguirre-Ghiso JA, et al. Dormancy signatures and metastasis in estrogen receptor positive and negative breast cancer. PLoS One. 2012;7:e35569. 16. Parker JS, Mullins M, Cheang MC, Leung S, Voduc D, Vickery T, et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol. 2009;27:1160-7. 17. Rouzier R, Perou CM, Symmans WF, Ibrahim N, Cristofanilli M, Anderson K, et al. Breast cancer molecular subtypes respond differently to preoperative chemotherapy. Clin Cancer Res. 2005;11:5678-85.  83 18. Fan C, Oh DS, Wessels L, Weigelt B, Nuyten DS, Nobel AB, et al. Concordance among gene-expression-based predictors for breast cancer. N Engl J Med. 2006;355:560-9. 19. Peddi PF, Ellis MJ, Ma C. Molecular basis of triple negative breast cancer and implications for therapy. Int J Breast Cancer. 2012;2012:217185. 20. Dawson SJ, Rueda OM, Aparicio S, Caldas C. A new genome-driven integrated classification of breast cancer and its implications. EMBO J. 2013;32:617-28. 21. Curtis C, Shah SP, Chin SF, Turashvili G, Rueda OM, Dunning MJ, et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature. 2012;486:346-52. 22. Fidler IJ. The pathogenesis of cancer metastasis: the 'seed and soil' hypothesis revisited. Nat Rev Cancer. 2003;3:453-8. 23. Valastyan S, Weinberg RA. Tumor metastasis: molecular insights and evolving paradigms. Cell. 2011;147:275-92. 24. Hedley BD, Chambers AF. Tumor dormancy and metastasis. Adv Cancer Res. 2009;102:67-101. 25. Chambers AF, Groom AC, MacDonald IC. Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer. 2002;2:563-72. 26. Gassmann P, Haier J. The tumor cell-host organ interface in the early onset of metastatic organ colonisation. Clin Exp Metastasis. 2008;25:171-81. 27. Guadamillas MC, Cerezo A, Del Pozo MA. Overcoming anoikis--pathways to anchorage-independent growth in cancer. J Cell Sci. 2011;124:3189-97. 28. Farber E. The multistep nature of cancer development. Cancer Res. 1984;44:4217-23. 29. Foulds L. Multiple etiologic factors in neoplastic development. Cancer Res. 1965;25:1339-47. 30. Foulds L. Neoplastic Development: Academic Press; 1969. 31. Klein CA. Parallel progression of primary tumours and metastases. Nat Rev Cancer. 2009;9:302-12. 32. Kraljevic Pavelic S, Sedic M, Bosnjak H, Spaventi S, Pavelic K. Metastasis: new perspectives on an old problem. Mol Cancer. 2011;10:22. 33. Sethi N, Kang Y. Unravelling the complexity of metastasis - molecular understanding and targeted therapies. Nat Rev Cancer. 2011;11:735-48. 34. Greaves M, Maley CC. Clonal evolution in cancer. Nature. 2012;481:306-13. 35. Nowell PC. The clonal evolution of tumor cell populations. Science. 1976;194:23-8. 36. Talmadge JE. Clonal selection of metastasis within the life history of a tumor. Cancer Res. 2007;67:11471-5. 37. Weigelt B, Peterse JL, van 't Veer LJ. Breast cancer metastasis: markers and models. Nat Rev Cancer. 2005;5:591-602. 38. Boman K, Larsson AH, Segersten U, Kuteeva E, Johannesson H, Nodin B, et al. Membranous expression of podocalyxin-like protein is an independent factor of poor prognosis in urothelial bladder cancer. Br J Cancer. 2013;108:2321-8. 39. Konecny GE, Pegram MD, Venkatesan N, Finn R, Yang G, Rahmeh M, et al. Activity of the dual kinase inhibitor lapatinib (GW572016) against HER-2-overexpressing and trastuzumab-treated breast cancer cells. Cancer Res. 2006;66:1630-9.  84 40. Slamon D, Eiermann W, Robert N, Pienkowski T, Martin M, Press M, et al. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365:1273-83. 41. Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001;344:783-92. 42. Dent R, Trudeau M, Pritchard KI, Hanna WM, Kahn HK, Sawka CA, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res. 2007;13:4429-34. 43. Liedtke C, Mazouni C, Hess KR, Andre F, Tordai A, Mejia JA, et al. Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol. 2008;26:1275-81. 44. Stratford AL, Reipas K, Maxwell C, Dunn SE. Targeting tumour-initiating cells to improve the cure rates for triple-negative breast cancer. Expert Rev Mol Med. 2010;12:e22. 45. Clevers H. The cancer stem cell: premises, promises and challenges. Nat Med. 2011;17:313-9. 46. Rosen JM, Jordan CT. The Increasing Complexity of the Cancer Stem Cell Paradigm. Science. 2009;324:1670-3. 47. Eckfeldt CE, Mendenhall EM, Verfaillie CM. The molecular repertoire of the 'almighty' stem cell. Nat Rev Mol Cell Biol. 2005;6:726-37. 48. Lobo NA, Shimono Y, Qian D, Clarke MF. The biology of cancer stem cells. Annu Rev Cell Dev Biol. 2007;23:675-99. 49. Kakarala M, Wicha MS. Implications of the cancer stem-cell hypothesis for breast cancer prevention and therapy. J Clin Oncol. 2008;26:2813-20. 50. Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang T, Caceres-Cortes J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 1994;367:645-8. 51. Calcagno AM, Salcido CD, Gillet JP, Wu CP, Fostel JM, Mumau MD, et al. Prolonged drug selection of breast cancer cells and enrichment of cancer stem cell characteristics. J Natl Cancer Inst. 2010;102:1637-52. 52. Levina V, Marrangoni AM, DeMarco R, Gorelik E, Lokshin AE. Drug-selected human lung cancer stem cells: cytokine network, tumorigenic and metastatic properties. PLoS One. 2008;3:e3077. 53. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A. 2003;100:3983-8. 54. Sheridan C, Kishimoto H, Fuchs RK, Mehrotra S, Bhat-Nakshatri P, Turner CH, et al. CD44+/CD24- breast cancer cells exhibit enhanced invasive properties: an early step necessary for metastasis. Breast Cancer Res. 2006;8:R59. 55. Lesley J, Hyman R. CD44 structure and function. Front Biosci. 1998;3:d616-30. 56. Goodison S, Urquidi V, Tarin D. CD44 cell adhesion molecules. Mol Pathol. 1999;52:189-96. 57. Balic M, Lin H, Young L, Hawes D, Giuliano A, McNamara G, et al. Most early disseminated cancer cells detected in bone marrow of breast cancer patients have a putative breast cancer stem cell phenotype. Clin Cancer Res. 2006;12:5615-21.  85 58. Montgomery N, Hill A, McFarlane S, Neisen J, O'Grady A, Conlon S, et al. CD44 enhances invasion of basal-like breast cancer cells by upregulating serine protease and collagen-degrading enzymatic expression and activity. Breast Cancer Res. 2012;14:R84. 59. Ouhtit A, Abd Elmageed ZY, Abdraboh ME, Lioe TF, Raj MH. In vivo evidence for the role of CD44s in promoting breast cancer metastasis to the liver. Am J Pathol. 2007;171:2033-9. 60. Cain JW, Hauptschein RS, Stewart JK, Bagci T, Sahagian GG, Jay DG. Identification of CD44 as a surface biomarker for drug resistance by surface proteome signature technology. Mol Cancer Res. 2011;9:637-47. 61. Hirschhaeuser F, Menne H, Dittfeld C, West J, Mueller-Klieser W, Kunz-Schughart LA. Multicellular tumor spheroids: an underestimated tool is catching up again. J Biotechnol. 2010;148:3-15. 62. Jin L, Hope KJ, Zhai Q, Smadja-Joffe F, Dick JE. Targeting of CD44 eradicates human acute myeloid leukemic stem cells. Nat Med. 2006;12:1167-74. 63. Shipitsin M, Campbell LL, Argani P, Weremowicz S, Bloushtain-Qimron N, Yao J, et al. Molecular definition of breast tumor heterogeneity. Cancer Cell. 2007;11:259-73. 64. Fischer GF, Majdic O, Gadd S, Knapp W. Signal transduction in lymphocytic and myeloid cells via CD24, a new member of phosphoinositol-anchored membrane molecules. J Immunol. 1990;144:638-41. 65. Kristiansen G, Winzer KJ, Mayordomo E, Bellach J, Schluns K, Denkert C, et al. CD24 expression is a new prognostic marker in breast cancer. Clin Cancer Res. 2003;9:4906-13. 66. Pirruccello SJ, LeBien TW. The human B cell-associated antigen CD24 is a single chain sialoglycoprotein. J Immunol. 1986;136:3779-84. 67. Fillmore C, Kuperwasser C. Human breast cancer stem cell markers CD44 and CD24: enriching for cells with functional properties in mice or in man? Breast Cancer Res. 2007;9:303. 68. Fillmore CM, Kuperwasser C. Human breast cancer cell lines contain stem-like cells that self-renew, give rise to phenotypically diverse progeny and survive chemotherapy. Breast Cancer Res. 2008;10:R25. 69. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008;133:704-15. 70. O'Brien CA, Kreso A, Jamieson CH. Cancer stem cells and self-renewal. Clin Cancer Res. 2010;16:3113-20. 71. Kershaw DB, Thomas PE, Wharram BL, Goyal M, Wiggins JE, Whiteside CI, et al. Molecular cloning, expression, and characterization of podocalyxin-like protein 1 from rabbit as a transmembrane protein of glomerular podocytes and vascular endothelium. J Biol Chem. 1995;270:29439-46. 72. McNagny KM, Lim F, Grieser S, Graf T. Cell surface proteins of chicken hematopoietic progenitors, thrombocytes and eosinophils detected by novel monoclonal antibodies. Leukemia. 1992;6:975-84. 73. McNagny KM, Pettersson I, Rossi F, Flamme I, Shevchenko A, Mann M, et al. Thrombomucin, a novel cell surface protein that defines thrombocytes and multipotent hematopoietic progenitors. J Cell Biol. 1997;138:1395-407.  86 74. Meder D, Shevchenko A, Simons K, Fullekrug J. Gp135/podocalyxin and NHERF-2 participate in the formation of a preapical domain during polarization of MDCK cells. J Cell Biol. 2005;168:303-13. 75. Kershaw DB, Beck SG, Wharram BL, Wiggins JE, Goyal M, Thomas PE, et al. Molecular cloning and characterization of human podocalyxin-like protein. Orthologous relationship to rabbit PCLP1 and rat podocalyxin. J Biol Chem. 1997;272:15708-14. 76. Takeda T, Go WY, Orlando RA, Farquhar MG. Expression of podocalyxin inhibits cell-cell adhesion and modifies junctional properties in Madin-Darby canine kidney cells. Mol Biol Cell. 2000;11:3219-32. 77. Tan PC, Furness SG, Merkens H, Lin S, McCoy ML, Roskelley CD, et al. Na+/H+ exchanger regulatory factor-1 is a hematopoietic ligand for a subset of the CD34 family of stem cell surface proteins. Stem Cells. 2006;24:1150-61. 78. Li Y, Li J, Straight SW, Kershaw DB. PDZ domain-mediated interaction of rabbit podocalyxin and Na(+)/H(+) exchange regulatory factor-2. Am J Physiol Renal Physiol. 2002;282:F1129-39. 79. Karthikeyan S, Leung T, Ladias JA. Structural basis of the Na+/H+ exchanger regulatory factor PDZ1 interaction with the carboxyl-terminal region of the cystic fibrosis transmembrane conductance regulator. J Biol Chem. 2001;276:19683-6. 80. Karthikeyan S, Leung T, Ladias JA. Structural determinants of the Na+/H+ exchanger regulatory factor interaction with the beta 2 adrenergic and platelet-derived growth factor receptors. J Biol Chem. 2002;277:18973-8. 81. Lazar CS, Cresson CM, Lauffenburger DA, Gill GN. The Na+/H+ exchanger regulatory factor stabilizes epidermal growth factor receptors at the cell surface. Mol Biol Cell. 2004;15:5470-80. 82. McClatchey AI. Merlin and ERM proteins: unappreciated roles in cancer development? Nat Rev Cancer. 2003;3:877-83. 83. Weinman EJ. New functions for the NHERF family of proteins. J Clin Invest. 2001;108:185-6. 84. Doyonnas R, Kershaw DB, Duhme C, Merkens H, Chelliah S, Graf T, et al. Anuria, omphalocele, and perinatal lethality in mice lacking the CD34-related protein podocalyxin. J Exp Med. 2001;194:13-27. 85. Doyonnas R, Nielsen JS, Chelliah S, Drew E, Hara T, Miyajima A, et al. Podocalyxin is a CD34-related marker of murine hematopoietic stem cells and embryonic erythroid cells. Blood. 2005;105:4170-8. 86. Maltby S, Hughes MR, Zbytnuik L, Paulson RF, McNagny KM. Podocalyxin selectively marks erythroid-committed progenitors during anemic stress but is dispensable for efficient recovery. Exp Hematol. 2009;37:10-8. 87. Nielsen JS, McNagny KM. Novel functions of the CD34 family. J Cell Sci. 2008;121:3683-92. 88. Kerjaschki D, Sharkey DJ, Farquhar MG. Identification and characterization of podocalyxin--the major sialoprotein of the renal glomerular epithelial cell. J Cell Biol. 1984;98:1591-6. 89. Nielsen JS, McNagny KM. The role of podocalyxin in health and disease. J Am Soc Nephrol. 2009;20:1669-76.  87 90. McNagny KM, Hughes MR, Graves ML, DeBruin EJ, Snyder K, Cipollone J, et al. Podocalyxin in the diagnosis and treatment of cancer. In: Mohan R, editor. Advances in Cancer Management: InTech; 2012. 91. Strilic B, Kucera T, Eglinger J, Hughes MR, McNagny KM, Tsukita S, et al. The molecular basis of vascular lumen formation in the developing mouse aorta. Dev Cell. 2009;17:505-15. 92. Nielsen JS, Graves ML, Chelliah S, Vogl AW, Roskelley CD, McNagny KM. The CD34-related molecule podocalyxin is a potent inducer of microvillus formation. PLoS One. 2007;2:e237. 93. Galeano B, Klootwijk R, Manoli I, Sun M, Ciccone C, Darvish D, et al. Mutation in the key enzyme of sialic acid biosynthesis causes severe glomerular proteinuria and is rescued by N-acetylmannosamine. J Clin Invest. 2007;117:1585-94. 94. Gelberg H, Healy L, Whiteley H, Miller LA, Vimr E. In vivo enzymatic removal of alpha 2-->6-linked sialic acid from the glomerular filtration barrier results in podocyte charge alteration and glomerular injury. Lab Invest. 1996;74:907-20. 95. Cipollone JA, Graves ML, Kobel M, Kalloger SE, Poon T, Gilks CB, et al. The anti-adhesive mucin podocalyxin may help initiate the transperitoneal metastasis of high grade serous ovarian carcinoma. Clin Exp Metastasis. 2012;29:239-52. 96. Graves M. Modeling mammary epithelial polarization and the role of podocalyxin in breast tumor progression. Vancouver, B.C.: University of British Columbia; 2008. 97. Somasiri A, Nielsen JS, Makretsov N, McCoy ML, Prentice L, Gilks CB, et al. Overexpression of the anti-adhesin podocalyxin is an independent predictor of breast cancer progression. Cancer Res. 2004;64:5068-73. 98. Lafrenie RM, Buckner CA, Bewick MA. Cell adhesion and cancer: is there a potential for therapeutic intervention? Expert Opin Ther Targets. 2007;11:727-31. 99. Li Q, Gao H, Xu H, Wang X, Pan Y, Hao F, et al. Expression of ezrin correlates with malignant phenotype of lung cancer, and in vitro knockdown of ezrin reverses the aggressive biological behavior of lung cancer cells. Tumour Biol. 2012;33:1493-504. 100. Yu Y, Khan J, Khanna C, Helman L, Meltzer PS, Merlino G. Expression profiling identifies the cytoskeletal organizer ezrin and the developmental homeoprotein Six-1 as key metastatic regulators. Nat Med. 2004;10:175-81. 101. Meng Y, Lu Z, Yu S, Zhang Q, Ma Y, Chen J. Ezrin promotes invasion and metastasis of pancreatic cancer cells. J Transl Med. 2010;8:61. 102. Curto M, McClatchey AI. Ezrin...a metastatic detERMinant? Cancer Cell. 2004;5:113-4. 103. Saito S, Yamamoto H, Mukaisho K, Sato S, Higo T, Hattori T, et al. Mechanisms underlying cancer progression caused by ezrin overexpression in tongue squamous cell carcinoma. PLoS One. 2013;8:e54881. 104. Khanna C, Wan X, Bose S, Cassaday R, Olomu O, Mendoza A, et al. The membrane-cytoskeleton linker ezrin is necessary for osteosarcoma metastasis. Nat Med. 2004;10:182-6. 105. Sizemore S, Cicek M, Sizemore N, Ng KP, Casey G. Podocalyxin increases the aggressive phenotype of breast and prostate cancer cells in vitro through its interaction with ezrin. Cancer Res. 2007;67:6183-91. 106. Louvet-Vallee S. ERM proteins: from cellular architecture to cell signaling. Biol Cell. 2000;92:305-16.  88 107. Tsukita S, Yonemura S. Cortical actin organization: lessons from ERM (ezrin/radixin/moesin) proteins. J Biol Chem. 1999;274:34507-10. 108. Tsukita S, Yonemura S, Tsukita S. ERM proteins: head-to-tail regulation of actin-plasma membrane interaction. Trends Biochem Sci. 1997;22:53-8. 109. Orlando RA, Takeda T, Zak B, Schmieder S, Benoit VM, McQuistan T, et al. The glomerular epithelial cell anti-adhesin podocalyxin associates with the actin cytoskeleton through interactions with ezrin. J Am Soc Nephrol. 2001;12:1589-98. 110. Hsu YH, Lin WL, Hou YT, Pu YS, Shun CT, Chen CL, et al. Podocalyxin EBP50 ezrin molecular complex enhances the metastatic potential of renal cell carcinoma through recruiting Rac1 guanine nucleotide exchange factor ARHGEF7. Am J Pathol. 2010;176:3050-61. 111. Thomas SN, Schnaar RL, Konstantopoulos K. Podocalyxin-like protein is an E-/L-selectin ligand on colon carcinoma cells: comparative biochemical properties of selectin ligands in host and tumor cells. Am J Physiol Cell Physiol. 2009;296:C505-13. 112. Wu H, Yang L, Liao D, Chen Y, Wang W, Fang J. Podocalyxin regulates astrocytoma cell invasion and survival against temozolomide. Exp Ther Med. 2013;5:1025-9. 113. Dallas MR, Chen SH, Streppel MM, Sharma S, Maitra A, Konstantopoulos K. Sialofucosylated podocalyxin is a functional E- and L-selectin ligand expressed by metastatic pancreatic cancer cells. Am J Physiol Cell Physiol. 2012;303:C616-24. 114. Larsson A, Johansson ME, Wangefjord S, Gaber A, Nodin B, Kucharzewska P, et al. Overexpression of podocalyxin-like protein is an independent factor of poor prognosis in colorectal cancer. Br J Cancer. 2011;105:666-72. 115. Forse CL, Yilmaz YE, Pinnaduwage D, O'Malley FP, Mulligan AM, Bull SB, et al. Elevated expression of podocalyxin is associated with lymphatic invasion, basal-like phenotype, and clinical outcome in axillary lymph node-negative breast cancer. Breast Cancer Res Treat. 2013;137:709-19. 116. Graves M, Cipollone, Jane A., Austin, Pamela, Harder, Eileen, Nielsen, Julie S., Gilks,  C. Blake, McNagny, Kelly M., Roskelley, Calvin D. . Podocalyxin drives collective primary breast tumor microinvasion independent of an overt epitheial-to-mesenchymal transformation. 2013. 117. Imbert AM, Garulli C, Choquet E, Koubi M, Aurrand-Lions M, Chabannon C. CD146 expression in human breast cancer cell lines induces phenotypic and functional changes observed in Epithelial to Mesenchymal Transition. PLoS One. 2012;7:e43752. 118. Liu WF, Ji SR, Sun JJ, Zhang Y, Liu ZY, Liang AB, et al. CD146 Expression Correlates with Epithelial-Mesenchymal Transition Markers and a Poor Prognosis in Gastric Cancer. Int J Mol Sci. 2012;13:6399-406. 119. Ponnusamy MP, Lakshmanan I, Jain M, Das S, Chakraborty S, Dey P, et al. MUC4 mucin-induced epithelial to mesenchymal transition: a novel mechanism for metastasis of human ovarian cancer cells. Oncogene. 2010;29:5741-54. 120. Roy LD, Sahraei M, Subramani DB, Besmer D, Nath S, Tinder TL, et al. MUC1 enhances invasiveness of pancreatic cancer cells by inducing epithelial to mesenchymal transition. Oncogene. 2011;30:1449-59.  89 121. Zeng Q, Li W, Lu D, Wu Z, Duan H, Luo Y, et al. CD146, an epithelial-mesenchymal transition inducer, is associated with triple-negative breast cancer. Proc Natl Acad Sci U S A. 2012;109:1127-32. 122. Meng X, Ezzati P, Wilkins JA. Requirement of podocalyxin in TGF-beta induced epithelial mesenchymal transition. PLoS One. 2011;6:e18715. 123. Babcook JS, Leslie KB, Olsen OA, Salmon RA, Schrader JW. A novel strategy for generating monoclonal antibodies from single, isolated lymphocytes producing antibodies of defined specificities. Proc Natl Acad Sci U S A. 1996;93:7843-8. 124. Lacroix M, Leclercq G. Relevance of breast cancer cell lines as models for breast tumours: an update. Breast Cancer Res Treat. 2004;83:249-89. 125. Larsson AH, Nodin B, Syk I, Palmquist I, Uhlen M, Eberhard J, et al. Podocalyxin-like protein expression in primary colorectal cancer and synchronous lymph node metastases. Diagn Pathol. 2013;8:109. 126. Ito M, Hiramatsu H, Kobayashi K, Suzue K, Kawahata M, Hioki K, et al. NOD/SCID/gamma(c)(null) mouse: an excellent recipient mouse model for engraftment of human cells. Blood. 2002;100:3175-82. 127. Tao K, Fang M, Alroy J, Sahagian GG. Imagable 4T1 model for the study of late stage breast cancer. BMC Cancer. 2008;8:228. 128. Hirohashi S, Kanai Y. Cell adhesion system and human cancer morphogenesis. Cancer Sci. 2003;94:575-81. 129. Larsson A, Fridberg M, Gaber A, Nodin B, Leveen P, Jonsson G, et al. Validation of podocalyxin-like protein as a biomarker of poor prognosis in colorectal cancer. BMC Cancer. 2012;12:282. 130. Orr FW, Wang HH, Lafrenie RM, Scherbarth S, Nance DM. Interactions between cancer cells and the endothelium in metastasis. J Pathol. 2000;190:310-29. 131. Geng Y, Marshall JR, King MR. Glycomechanics of the metastatic cascade: tumor cell-endothelial cell interactions in the circulation. Ann Biomed Eng. 2012;40:790-805. 132. Konstantopoulos K, Thomas SN. Cancer cells in transit: the vascular interactions of tumor cells. Annu Rev Biomed Eng. 2009;11:177-202. 133. Chaichana K, Zamora-Berridi G, Camara-Quintana J, Quinones-Hinojosa A. Neurosphere assays: growth factors and hormone differences in tumor and nontumor studies. Stem Cells. 2006;24:2851-7. 134. Pastrana E, Silva-Vargas V, Doetsch F. Eyes wide open: a critical review of sphere-formation as an assay for stem cells. Cell Stem Cell. 2011;8:486-98. 135. Frisch SM, Francis H. Disruption of epithelial cell-matrix interactions induces apoptosis. J Cell Biol. 1994;124:619-26. 136. Frisch SM, Ruoslahti E. Integrins and anoikis. Curr Opin Cell Biol. 1997;9:701-6. 137. Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J, et al. Identification of a cancer stem cell in human brain tumors. Cancer Res. 2003;63:5821-8. 138. Grimshaw MJ, Cooper L, Papazisis K, Coleman JA, Bohnenkamp HR, Chiapero-Stanke L, et al. Mammosphere culture of metastatic breast cancer cells enriches for tumorigenic breast cancer cells. Breast Cancer Res. 2008;10:R52. 139. Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, Todaro M, Peschle C, et al. Identification and expansion of human colon-cancer-initiating cells. Nature. 2007;445:111-5.  90 140. Fang D, Nguyen TK, Leishear K, Finko R, Kulp AN, Hotz S, et al. A tumorigenic subpopulation with stem cell properties in melanomas. Cancer Res. 2005;65:9328-37. 141. Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ, et al. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev. 2003;17:1253-70. 142. Ponti D, Costa A, Zaffaroni N, Pratesi G, Petrangolini G, Coradini D, et al. Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Res. 2005;65:5506-11. 143. Kunz-Schughart LA. Multicellular tumor spheroids: intermediates between monolayer culture and in vivo tumor. Cell Biol Int. 1999;23:157-61. 144. Brandenberger R, Wei H, Zhang S, Lei S, Murage J, Fisk GJ, et al. Transcriptome characterization elucidates signaling networks that control human ES cell growth and differentiation. Nat Biotechnol. 2004;22:707-16. 145. Cai J, Chen J, Liu Y, Miura T, Luo Y, Loring JF, et al. Assessing self-renewal and differentiation in human embryonic stem cell lines. Stem Cells. 2006;24:516-30. 146. Binder ZA, Siu IM, Eberhart CG, Ap Rhys C, Bai RY, Staedtke V, et al. Podocalyxin-like protein is expressed in glioblastoma multiforme stem-like cells and is associated with poor outcome. PLoS One. 2013;8:e75945. 147. Carmeliet P, Jain RK. Principles and mechanisms of vessel normalization for cancer and other angiogenic diseases. Nat Rev Drug Discov. 2011;10:417-27. 148. Goel S, Duda DG, Xu L, Munn LL, Boucher Y, Fukumura D, et al. Normalization of the vasculature for treatment of cancer and other diseases. Physiol Rev. 2011;91:1071-121. 149. Brown JM, Giaccia AJ. The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res. 1998;58:1408-16. 150. Liang D, Ma Y, Liu J, Trope CG, Holm R, Nesland JM, et al. The hypoxic microenvironment upgrades stem-like properties of ovarian cancer cells. BMC Cancer. 2012;12:201. 151. Ma Y, Liang D, Liu J, Axcrona K, Kvalheim G, Stokke T, et al. Prostate cancer cell lines under hypoxia exhibit greater stem-like properties. PLoS One. 2011;6:e29170. 152. Soeda A, Park M, Lee D, Mintz A, Androutsellis-Theotokis A, McKay RD, et al. Hypoxia promotes expansion of the CD133-positive glioma stem cells through activation of HIF-1alpha. Oncogene. 2009;28:3949-59. 153. Louie E, Nik S, Chen JS, Schmidt M, Song B, Pacson C, et al. Identification of a stem-like cell population by exposing metastatic breast cancer cell lines to repetitive cycles of hypoxia and reoxygenation. Breast Cancer Res. 2010;12:R94. 154. Ezashi T, Das P, Roberts RM. Low O2 tensions and the prevention of differentiation of hES cells. Proc Natl Acad Sci U S A. 2005;102:4783-8. 155. Kim Y, Lin Q, Glazer PM, Yun Z. Hypoxic tumor microenvironment and cancer cell differentiation. Curr Mol Med. 2009;9:425-34. 156. Semenza GL. HIF-1: mediator of physiological and pathophysiological responses to hypoxia. J Appl Physiol (1985). 2000;88:1474-80. 157. Zhang G, Yang P, Guo P, Miele L, Sarkar FH, Wang Z, et al. Unraveling the mystery of cancer metabolism in the genesis of tumor-initiating cells and development of cancer. Biochim Biophys Acta. 2013;1836:49-59.  91 158. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis? Nat Rev Cancer. 2004;4:891-9. 159. Dang CV. Links between metabolism and cancer. Genes Dev. 2012;26:877-90. 160. Chen C, Pore N, Behrooz A, Ismail-Beigi F, Maity A. Regulation of glut1 mRNA by hypoxia-inducible factor-1. Interaction between H-ras and hypoxia. J Biol Chem. 2001;276:9519-25. 161. Denko NC. Hypoxia, HIF1 and glucose metabolism in the solid tumour. Nat Rev Cancer. 2008;8:705-13. 162. Maxwell PH, Dachs GU, Gleadle JM, Nicholls LG, Harris AL, Stratford IJ, et al. Hypoxia-inducible factor-1 modulates gene expression in solid tumors and influences both angiogenesis and tumor growth. Proc Natl Acad Sci U S A. 1997;94:8104-9. 163. Krzeslak A, Wojcik-Krowiranda K, Forma E, Jozwiak P, Romanowicz H, Bienkiewicz A, et al. Expression of GLUT1 and GLUT3 glucose transporters in endometrial and breast cancers. Pathol Oncol Res. 2012;18:721-8. 164. Schopperle WM, Lee JM, Dewolf WC. The human cancer and stem cell marker podocalyxin interacts with the glucose-3-transporter in malignant pluripotent stem cells. Biochem Biophys Res Commun. 2010;398:372-6. 165. Flavahan WA, Wu Q, Hitomi M, Rahim N, Kim Y, Sloan AE, et al. Brain tumor initiating cells adapt to restricted nutrition through preferential glucose uptake. Nat Neurosci. 2013;16:1373-82. 166. Larrucea S, Butta N, Arias-Salgado EG, Alonso-Martin S, Ayuso MS, Parrilla R. Expression of podocalyxin enhances the adherence, migration, and intercellular communication of cells. Exp Cell Res. 2008;314:2004-15. 167. Schmieder S, Nagai M, Orlando RA, Takeda T, Farquhar MG. Podocalyxin activates RhoA and induces actin reorganization through NHERF1 and Ezrin in MDCK cells. J Am Soc Nephrol. 2004;15:2289-98. 168. Burbelo P, Wellstein A, Pestell RG. Altered Rho GTPase signaling pathways in breast cancer cells. Breast Cancer Res Treat. 2004;84:43-8. 169. Tang Y, Olufemi L, Wang MT, Nie D. Role of Rho GTPases in breast cancer. Front Biosci. 2008;13:759-76. 170. Zhang X, Nie D. Rho GTPases and Breast Cancer In: Gunduz M, editor. Breast Cancer - Focusing Tumor Microenvironment, Stem cells and Metastasis: InTech; 2011. 171. Fernandez D, Horrillo A, Alquezar C, Gonzalez-Manchon C, Parrilla R, Ayuso MS. Control of cell adhesion and migration by podocalyxin. Implication of Rac1 and Cdc42. Biochem Biophys Res Commun. 2013;432:302-7. 172. Khanna C, Hunter K. Modeling metastasis in vivo. Carcinogenesis. 2005;26:513-23. 173. Glinskii OV, Huxley VH, Glinsky GV, Pienta KJ, Raz A, Glinsky VV. Mechanical entrapment is insufficient and intercellular adhesion is essential for metastatic cell arrest in distant organs. Neoplasia. 2005;7:522-7. 174. Labelle M, Hynes RO. The initial hours of metastasis: the importance of cooperative host-tumor cell interactions during hematogenous dissemination. Cancer Discov. 2012;2:1091-9. 175. Qian B, Deng Y, Im JH, Muschel RJ, Zou Y, Li J, et al. A distinct macrophage population mediates metastatic breast cancer cell extravasation, establishment and growth. PLoS One. 2009;4:e6562.  92 176. Rasanen K, Vaheri A. Activation of fibroblasts in cancer stroma. Exp Cell Res. 2010;316:2713-22. 177. Orimo A, Gupta PB, Sgroi DC, Arenzana-Seisdedos F, Delaunay T, Naeem R, et al. Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell. 2005;121:335-48. 178. Gaggioli C, Hooper S, Hidalgo-Carcedo C, Grosse R, Marshall JF, Harrington K, et al. Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells. Nat Cell Biol. 2007;9:1392-400. 179. Lo BC, Hughes MR, Tan PC, Eder A, Nielsen JS, McNagny KM. Podocalyxin enhanced CXCR4-dependent migration of hematopoietic cells. University of British Columbia; 2013. 180. Teicher BA, Fricker SP. CXCL12 (SDF-1)/CXCR4 pathway in cancer. Clin Cancer Res. 2010;16:2927-31. 181. Campbell CI, Thompson DE, Siwicky MD, Moorehead RA. Murine mammary tumor cells with a claudin-low genotype. Cancer Cell Int. 2011;11:28. 182. Kao J, Salari K, Bocanegra M, Choi YL, Girard L, Gandhi J, et al. Molecular profiling of breast cancer cell lines defines relevant tumor models and provides a resource for cancer gene discovery. PLoS One. 2009;4:e6146. 183. Meyer MJ, Fleming JM, Lin AF, Hussnain SA, Ginsburg E, Vonderhaar BK. CD44posCD49fhiCD133/2hi defines xenograft-initiating cells in estrogen receptor-negative breast cancer. Cancer Res. 2010;70:4624-33. 184. Reichert JM, Valge-Archer VE. Development trends for monoclonal antibody cancer therapeutics. Nat Rev Drug Discov. 2007;6:349-56. 185. Sliwkowski MX, Mellman I. Antibody therapeutics in cancer. Science. 2013;341:1192-8. 186. Iorns E, Drews-Elger K, Ward TM, Dean S, Clarke J, Berry D, et al. A new mouse model for the study of human breast cancer metastasis. PLoS One. 2012;7:e47995.             93 APPENDICES Appendix A. Primary tumor development of MCF-7Podxl cells in NSG mice Unpublished results by Graves et al. describe a subtle increase in tumor progression upon orthotopic injection of MCF-7Podxl cells in Rag2M-/- mice, whereby expression of podocalyxin increases collective tumor migration from the primary tumor through the mammary fat pad. Based on these results, I chose to assess whether NSG mice act as better recipients for MCF-7Podxl cells and whether use of a highly immunocompromised strain of mice would better allow us to observe a podocalyxin-dependent effect on subcutaneous primary tumor growth of MCF-7Podxl cells. I found that podocalyxin has no effect on primary tumor volume when comparing tumor development of control MCF-7pIRES cells and MCF-7Podxl cells in NSG mice (Figure A-1). However, I did confirm that MCF-7 tumor cells are able to develop tumors without estrogen supplementation in NSG mice, as is necessary in the Rag2M-/- mouse strain (186).   Figure A-1: Podocalyxin overexpressing MCF-7 cells possess no primary tumor growth advantage in NSG mice. A total of 3 x 106 MCF-7pIRES control and MCF-7Podxl cells were subcutaneously injected into the flank of NSG mice (n = 8). Tumor dimensions were measured over the course of 3 months. Tumor volumes (mm3) were calculated by (length x width2)/2 and plotted against time (p = 0.57 by two-way ANOVA). All values graphed as mean ? SEM.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0072153/manifest

Comment

Related Items