Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A functional and prognostic study of the transcription factor SIX1 in melanoma Graziano, Laura 2017

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2018_february_graziano_laura.pdf [ 2.37MB ]
JSON: 24-1.0362582.json
JSON-LD: 24-1.0362582-ld.json
RDF/XML (Pretty): 24-1.0362582-rdf.xml
RDF/JSON: 24-1.0362582-rdf.json
Turtle: 24-1.0362582-turtle.txt
N-Triples: 24-1.0362582-rdf-ntriples.txt
Original Record: 24-1.0362582-source.json
Full Text

Full Text

  A FUNCTIONAL AND PROGNOSTIC STUDY OF THE TRANSCRIPTION FACTOR SIX1 IN MELANOMA by  Laura Graziano  B.Sc., The University of Guelph, 2014  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)  December 2017  © Laura Graziano, 2017    ii Abstract This study focuses on melanoma, a cancer of pigment-producing cells within the skin (melanocytes) which insidiously spreads (metastasizes) throughout the body. This devastating disease often prevents patients from living beyond five years from diagnosis. Treatment options for melanoma are inept, which has necessitated a demand for research on the molecular mechanisms of melanoma to aid in the development of novel therapeutic targets and treatments.   The gene Sineoculis homeobox homolog 1 (SIX1) encodes for a homeoprotein transcription factor which is an important developmental regulator during embryogenesis. A preliminary microarray analysis found that SIX1 was upregulated in melanoma. The purpose of this study was to pioneer the assessment of SIX1 in melanoma, including its role in the metastatic functions of melanoma in vitro as well as its clinical relevance. The specific aims included: (1) to assess SIX1 expression in melanoma cell lines and clinical samples; (2) to investigate the functional significance and pathogenic role of Six1 in vitro; and (3) to evaluate Six1’s clinical significance in relation to prognosis and clinicopathological characteristics.  Cell lines and clinical samples were tested for SIX1 transcript level using qPCR and protein level using Western Blotting. Next, two malignant melanoma cell lines were transfected with shRNA plasmids to generate stable clones with low Six1 expression. These altered cell lines were used to study the functional implications of Six1 expression disparities in vitro. Furthermore, immunohistochemistry against Six1 was performed on a tissue microarray of 438 melanoma patient biopsies.   iii  We discovered that SIX1 was overexpressed in melanoma cell lines and clinical samples. Six1 knockdown cells demonstrated diminished cell growth and proliferation, increased apoptosis, and decreased migration and invasion. We also found that a profound nuclear to cytoplasmic shift of Six1 accompanied melanoma progression and correlated with poor five-year survival. Higher cytoplasmic Six1 was associated with increased tumor thickness, lower nuclear Six1 with ulceration and histological satellitosis, and both with advanced AJCC stages and nodular melanoma.  To summarize, these results hint that Six1 may play a role in the execution of metastatic functions in melanoma. Furthermore, Six1 presents as a tentative candidate for a prognostic marker in patient biopsies.                 iv Lay Summary Melanoma is the most dangerous type of skin cancer involving cells that are responsible for skin colour known as melanocytes. Melanoma cells are harmful because they grow uncontrollably and spread (metastasize) throughout the body. Many patients die as treatment options are limited due to a lack of knowledge of this disease. In this study, we used melanoma cells and experimented on them in a petri dish to try to figure out why these cells are special: why do they grow so fast and how do they spread? We focused on a tiny molecule within these cells known as a protein named Six1 to see if partially stopping Six1 from doing its job would minimize these unhealthy activities. We found that inhibition of Six1 may reduce the detrimental effects of melanoma and that its location within the cell may be useful when guesstimating the severity of melanoma patient condition.               v Preface This thesis is an original intellectual product from the author L. Graziano. No collaborations or publications have arisen from this research. Figures 1.1, 1.2, 1.3, 1.4, 2.1 and 2.2 are used with permission from applicable sources.  Dr. Yabin Cheng constructed the tissue microarray, its patient database, and performed the immunohistochemistry in this project. Dr. Mingwan Su provided the clinical sample RNA.                  vi   Table of Contents Abstract .................................................................................................................................... ii Lay Summary ......................................................................................................................... iv Preface ...................................................................................................................................... v Table of Contents ................................................................................................................... vi List of Tables .......................................................................................................................... ix List of Figures .......................................................................................................................... x List of Abbreviations ............................................................................................................ xii Acknowledgements............................................................................................................. xviii Dedication ............................................................................................................................. xix 1 Introduction .......................................................................................................................... 1 1.1. Melanoma: an overview ............................................................................................................... 1 1.1.1. An introduction to skin cancer .............................................................................................. 1 1.1.2. Biology of melanoma ............................................................................................................ 2 1.1.3. Contributing factors to melanoma acquisition and progression ............................................ 2 1.1.4. Melanoma progression .......................................................................................................... 4 1.1.5. Clinical staging of melanoma and histological subclassification........................................ 12 1.2. Current metastatic melanoma therapeutic interventions ............................................................ 14 1.2.1. Biomarkers of melanoma .................................................................................................... 18 1.3. The SIX1 gene and Six1 protein ................................................................................................ 20 1.3.1. Background ......................................................................................................................... 20 1.3.2. Structure .............................................................................................................................. 21 1.3.3. Function and molecular mechanisms .................................................................................. 22   vii 1.4. SIX1 in developmental disorders ............................................................................................... 27 1.5. SIX1 in cancer ............................................................................................................................ 28 1.6. Thesis outline ............................................................................................................................. 29 1.6.1. Rationale.............................................................................................................................. 29 1.6.2. Research hypotheses ........................................................................................................... 30 1.6.3. Specific aims ....................................................................................................................... 31 2 Materials and Methods ...................................................................................................... 32 2.1. Established metastatic melanoma cell lines and growth conditions (cell culture) ..................... 32 2.2. RNA extraction and reverse transcription .................................................................................. 32 2.3. Quantitative real-time polymerase chain reaction (qRT-PCR) .................................................. 33 2.4. Protein extraction and Western blotting ..................................................................................... 34 2.5. Tissue microarray (TMA) .......................................................................................................... 36 2.5.1. Source of biopsies ............................................................................................................... 36 2.5.2. Clinicopathological characteristics of patients .................................................................... 36 2.6. Tissue microarray (TMA) immunohistochemistry .................................................................... 37 2.7. Immunohistochemistry of tissue microarray (TMA) ................................................................. 37 2.8. Evaluation of TMA immunostaining ......................................................................................... 38 2.9. Puromycin kill curve .................................................................................................................. 39 2.10. Establishing stable transfection SIX1 knockdown clones of metastatic melanoma cell lines with reduced Six1 protein expression ............................................................................................... 39 2.10.1. shRNA plasmid preparation and acquisition from glycerol stock and subsequent transfection and gene knockdown ................................................................................................. 43 2.10.2. Ethanol precipitation ......................................................................................................... 44 2.10.3. Generation of stably transfected cells ............................................................................... 44 2.11. Cell growth assay ..................................................................................................................... 45 2.12. Cell proliferation assay............................................................................................................. 46 2.13. Cell migration assay ................................................................................................................. 47 2.14. Cell invasion assay ................................................................................................................... 48 2.14.1. Corning BioCoat matrigel invasion chamber .................................................................... 48 2.14.2. Measurement of cell invasion ........................................................................................... 49 2.15. Statistical analysis .................................................................................................................... 50 3 Results ................................................................................................................................. 51 3.1. Determination of SIX1 mRNA expression in established cell lines and clinical samples ......... 51 3.2. Six1 protein levels are increased in metastatic melanoma cell lines .......................................... 52   viii 3.3. Western blot to assess Six1 shRNA stable knockdown in A375 and RPMI-7951 .................... 53 3.4. In vitro functional studies following stable knockdown of Six1................................................ 54 3.4.1. Cell growth and Six1 ........................................................................................................... 54 3.4.2. Cell proliferation and Six1 .................................................................................................. 55 3.4.3. Apoptosis of melanoma cells and Six1 ............................................................................... 56 3.4.4. Migration of melanoma cells and Six1 ............................................................................... 58 3.4.5. Invasion and Six1 in melanoma cells .................................................................................. 60 3.5. The clinical relevance of Six1 in melanoma .............................................................................. 62 3.5.1. Six1 and melanoma prognosis............................................................................................. 63 3.5.2. Six1’s association with lesion type ..................................................................................... 63 3.5.3. Six1’s association with clinicopathological parameters ...................................................... 64 4 Discussion ............................................................................................................................ 70 4.1. Limitations and future directions ............................................................................................... 87 5 Summary and Conclusion ................................................................................................. 94 Bibliography .......................................................................................................................... 96         ix List of Tables Table 2.1 Primers for qPCR .................................................................................................... 33 Table 2.2 List of primary antibodies for Western blot ............................................................ 35 Table 2.3 Oligonucleotides encoding the shRNAs ................................................................. 40 Table 2.4 Description of pLKO.1 plasmid cloning vector elements ...................................... 42 Table 2.5 Timecourse of post-transfection assays .................................................................. 45 Table 3.1 Nuclear Six1 expression and clinicopathological characteristics ........................... 66 Table 3.2 Cytoplasmic Six1 expression and clinicopathological characteristics .................... 68                  x List of Figures Figure 1.1 Biological events in the progression of melanoma as described by the Clark model. .............................................................................................................................. 12 Figure 1.2 Visual representation of the SIX1 gene ................................................................. 22 Figure 1.3 Diagrammatic depiction of the transcription factor homeoprotein Six1. .............. 22 Figure 1.4 A diagram depicting the Six-Eya-Dach-regulated activation of target genes on DNA in the nucleus. ........................................................................................................ 27 Figure 2.1 Map of pLKO.1 plasmid containing a shRNA insert used for stable transfections. ......................................................................................................................................... 41 Figure 2.2 Diagram and details of the shRNA insert .............................................................. 43 Figure 3.1 SIX1 mRNA levels are increased in metastatic melanoma cell lines and clinical tissues. ............................................................................................................................. 52 Figure 3.2 Six1 protein levels are increased in metastatic melanoma cell lines. .................... 53 Figure 3.3 Six1 knockdown was achieved via shRNA plasmid transfection in metastatic melanoma cell lines. ........................................................................................................ 54 Figure 3.4 Six1 inhibition reduces cell growth of metastatic melanoma cells in vitro. .......... 55 Figure 3.5 Proliferation declines in metastatic melanoma cells with reduced Six1 protein. .. 56 Figure 3.6 Six1 knockdown has minor or no effect on apoptosis proteins in metastatic melanoma cell line A375. ............................................................................................... 57 Figure 3.7 Six1 knockdown restricts metastatic melanoma cell migration. ........................... 59 Figure 3.8 Invasive ability of malignant melanoma cells is inhibited by Six1 knockdown. .. 61 Figure 3.9 Six1 protein staining in tissue microarray. ............................................................ 62   xi Figure 3.10 Kaplan-Meier survival analyses of melanoma patients regarding Six1 expression levels. .............................................................................................................................. 63 Figure 3.11 Six1 protein expression shifts from the nucleus to the cytoplasm during melanoma progression. ................................................................................................... 64                      xii List of Abbreviations Abbreviation Definition A375 Metastatic melanoma human cell line ATCC AJCC American Joint Committee on Cancer AKT Alpha serine/threonine-protein kinase  APC Anaphase-promoting complex ATF2 Activating transcription factor 2 ATP Adenosine triphosphate  BCC Basal cell carcinoma BP1 Beta Protein 1  BRAF Murine sarcoma viral (v-raf) oncogene homolog B1 (gene); B-Raf proto-oncogene serine/threonine-protein kinase (protein) BrdU 5-bromo-2’-deoxyuridine c-Myc Proto-oncogene c-Myc; V-Myc myelocytomatosis viral oncogene homolog CBP CREB-binding protein   xiii ChIP-seq  Chromatin immunoprecipitation assays with sequencing CK2 Casein kinase II CRAF Normal cellular RAF gene; RAF proto-oncogene serine/threonine-protein kinase DACH1 Dachshund family transcription factor 1 DMEM Dulbecco’s modified eagle medium DN  Dysplastic nevi EMT  Epithelial to mesenchymal transition ERK Extracellular signal-regulated kinases EYA Eyes absent gene Ezh2 Enhancer of zeste homolog 2 FACS Fluorescence-activated call sorting FasL Fas ligand FBS Fetal bovine serum FDA Food and Drug Administration GAPDH  Glyceraldehyde-3-phosphate dehydrogenase   xiv GBX2 Gastrulation Brain Homeobox 2 Gdnf Glial cell line-derived neurotrophic factor H&E Hematoxylin and eosin HD IL-2  High dose interleukin 2 IF Immunofluorescence IFN- Interferon-alpha IHC Immunohistochemistry IL Interleukin kDa Kilodalton KIT KIT proto-oncogene receptor tyrosine kinase KRAS Kirsten rat sarcoma viral oncogene homolog LDH Lactate dehydrogenase  MAPK Mitogen-activated protein kinases MC1R Melanocortin receptor 1 miRNA microRNA   xv MIS  Melanoma in situ MITF Melanogenesis associated transcription factor mRNA Messenger RNA NN  Normal nevi P53 Tumor suppressor p53 PBS Phosphase buffered saline PD-1 Programmed cell death protein 1 PD-L1 Programmed cell death protein-ligand 1  PELP1 Proline-, glutamic acid-, and leucine-rich protein-1 PIC Protease inhibitor cocktail PM Primary melanoma PM  Primary melanoma pTNM Pathological tumor node metastasis qPCR Quantitative polymerase chain reaction RGP Radial growth phase   xvi RIME Rapid immunoprecipitation mass spectrometry of endogenous proteins RIPA buffer  Radioimmunoprecipitation assay buffer Rpm  Rotations per minute RPMI-7951  Metastatic melanoma human cell line ATCC RT Room temperature RT-PCR Reverse transcription-polymerase chain reaction SC Satellite cell; Stem cell SCC Squamous cell carcinoma SD SIX domain SDS Sodium dodecyl sulfate sh1, sh2  shRNA knockdown cells shCTRL   shRNA Control (no knockdown) shRNA Short hairpin RNA siRNA  Short interfering RNA SIX1 Sineoculis homeobox homolog 1 gene   xvii Six1 Six homeobox 1 protein SMA Smooth muscle actin StDev Standard deviation TERT Telomerase reverse transcriptase TGF- Transforming growth factor beta TMA Tissue microarray UVR Ultraviolet radiation VGP Vertical growth phase  WNT Wingless-type mammary tumor virus integration-site family ZEB1 Zinc finger E-box binding homeobox 1 MSH Alpha-melanocyte-stimulating hormone         xviii Acknowledgements I would like to thank my master’s supervisor, Dr. Youwen Zhou, for providing guidance and prompting critical thinking during my time at UBC. I also would like to thank my committee members: Dr. Vincent Duronio and Dr. Kevin McElwee for their guidance and input on my project. Additionally, I thank our lab manager, Dr. Mingwan Su, for being there to answer my many technical experimental questions. I owe many thanks to the laboratory training I received from Dr. Yabin Cheng who patiently taught me many laboratory techniques. I would also like to thank the members of the Molecular Medicine Lab for supporting me in my studies, including Xue Zhang, Rayeheh Bahar, and Dr. Yuanshen Huang. Also, I extend my sincerest gratitude to Dr. Gang Wang and Dr. Magdalena Martinka for contributing their pathology expertise to my project. Furthermore, I would like to thank my fellow classmates and friends for their moral support during my thesis work: Eva Yap, Patrick Coulombe, and Mia Divac. Finally, I would like to thank my amazing family: my mother, Laurie Columbus, for her never-ending support; my father, Ken Graziano, for rooting for me; my sister Katie Graziano and my dog Molly. Finally, I would not have been able to get through my master’s without the support of my wonderful husband, Oliver Majewski. Thank you for always being there for me and asking me intricate questions about molecular biology that I would not have thought to ask myself.        xix Dedication         I dedicate this thesis to my Uncle Kevin Columbus,  Grandmother Joanne Columbus and  Grandfather Chris Columbus.  Thank you for inspiring me to pursue my dreams.             1 1 Introduction 1.1. Melanoma: an overview 1.1.1. An introduction to skin cancer Sun exposure is a part of our daily lives; but throughout history and until recently, we rarely considered it a threat to our well-being. This view has now shifted and there is focus on the cumulative effects and compounded risks sun exposure contributes to disease acquisition. One such disease is skin cancer, which constitutes 40% of cancer cases worldwide, making it the most common type of cancer (Dubas and Ingraffea, 2013). Skin cancer can be separated into three main types: basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma. Histologically, BCC, SCC and melanoma can be distinguished by the skin cell type within the epidermis from which they arise. Both BCC and SCC result from the uncontrolled proliferation of keratin-producing cells known as keratinocytes, the most common type of cell in the skin. In contrast, melanoma involves the aberrant growth and metastatic characteristics developed and accumulated by melanocytes (Dubas & Ingraffea, 2013; Miller & Mihm, 2006). Both BCC and SCC are correlated with high rates of survival after diagnosis and treatment and rarely progress to metastatic disease; however, this is not the case for melanoma (Dubas & Ingraffea, 2013). Melanoma only constitutes 4% of skin cancer diagnoses but is responsible for 80% of the deaths caused by skin cancer (Rigel et al., 2011). The American Cancer Society estimated that in 2016, 76,380 new cases were diagnosed and 10,130 deaths resulted from metastatic melanoma, making it the fifth most common cancer in men and seventh in women in the U.S. (American Cancer Society, 2016). When diagnosed at an early stage, melanoma patients have a high chance of survival. However, patients   2 diagnosed with stage IV melanoma have a poor prognosis, with the associated five-year survival rate being less than 10% (Balch et al., 2009; Rigel et al., 2011). In fact, from the time of initial stage IV diagnosis, the median survival is a mere 6-7.5 months (Rigel et al., 2011). 1.1.2. Biology of melanoma Melanocytes arise from the neural crest during development, after which they reside in the skin and to a lesser degree in the eye, anogenital tract and meninges (Shain & Bastian, 2016). The focus of this thesis will be on cutaneous melanoma, or melanoma which originates in the skin and is the most common type. The skin is made up of three layers: the upper epidermis, the dermis and the hypodermis. Melanocytes are located in the basilar epidermis—the deepest level of the upper layer of the skin (Shain & Bastian, 2016). Their rate of mitosis is less than twice per year, but they have an important job of protecting the skin from harmful ultraviolet radiation (UVR) (Shain & Bastian, 2016). Melanocytes accomplish this protection by producing a pigmented macromolecule known as melanin, responsible for hair and skin colour, which scatters and absorbs UVR (Shain & Bastian, 2016). Melanin production by melanocytes is triggered by alpha-melanocyte-stimulating hormone (MSH), which is produced by keratinocytes in response to UVR-induced DNA damage (Shain & Bastian, 2016). Stimulated melanocytes produce the melanin-containing organelles melanosomes, which are shuttled to keratinocytes via arm-like protrusions termed dendrites (Rigel et al., 2011).  1.1.3. Contributing factors to melanoma acquisition and progression The cause of melanoma, as with most cancers, is not simple or restricted to one factor or event. It can be triggered by a combination of genetic and environmental risk factors   3 (Reed et al., 2012; Meyle & Guldberg, 2009). The most prominent risk factors include genetic predispositions such as a family history of melanoma, multiple nevi (moles), and having already had melanoma (Miller & Mihm, 2006). Other risk factors include exposure to ultraviolet radiation (UVR), sun sensitivity and immunosuppression (Miller & Mihm, 2006). Although the genetic risk factors are far from being fully understood, studies have revealed that mutations in the genes cyclin-dependent kinase inhibitor 2A (CDKN2A) and cyclin-dependent kinase 4 (CDK4) are present in melanoma-prone families (Miller & Mihm, 2006). Since CDKN2A and CDK4 are tumor-suppressor genes, mutations which render them non-functional are linked to a predisposition to melanoma (Miller & Mihm, 2006). The functional significance of these genes will be discussed further below. UVR exposure, although a natural occurrence when outside on a sunny day, is a serious risk factor for melanoma. The dangers of UVR exposure include increased production of growth factors, impairment of cutaneous immune function, genetic changes given that it is a potent mutagen, and stimulated production of DNA-damaging reactive oxygen species (Miller & Mihm, 2006). Evidence from epidemiological studies suggest that intense, intermittent exposures to UVR cause genetic damage and increased risk of melanoma more-so than chronic or low-grade exposures (which may confer protection to DNA) (Miller & Mihm, 2006).  The normal tanning response of melanin production by melanocytes has been linked to the previously mentioned hormone MSH. The function of MSH is to bind to its receptor, melanocortin receptor 1 (MC1R) on melanocytes which stimulates signaling pathways to increase melanin production (Miller & Mihm, 2006). Germ-line polymorphisms have been found in MC1R genes which reduce the production of UVR-protective melanin in   4 light-skinned and redheaded individuals, thereby rendering them more susceptible to UVR damage and consequently melanoma (Miller & Mihm, 2006). 1.1.4. Melanoma progression Metastasis, or malignancy, can be defined as the spread of cancer from a primary, often isolated site in the body to new and sometimes distant sites (Miller & Mihm, 2006). Malignant melanoma is much more dangerous and has a drastically decreased survival rate and grim prognostic outlook than non-metastatic skin cancer. The morbidity and mortality associated with melanoma is due to the local invasion and metastatic spread of melanoma cells (Miller & Mihm, 2006). However, melanoma is a complex disease which most often begins from precursor lesions. As with many tumors, the lethal potential of melanoma increases over time. Therefore, an understanding of melanoma progression is important to aid in earlier diagnosis and excision of intermediate tumors by recognizing the potential for malignancy to occur. Our understanding of melanoma progression is often focused on histopathological characteristics of lesions; however, genetic and molecular pathway abnormalities are becoming useful in this endeavor.  Melanoma generally progresses throughout an individual’s lifetime when accumulation of pro-proliferative and pro-metastatic mutations occur within melanocytes. These mutations bestow this aberrant cell population with abnormal features and behaviours by exploiting normal molecular pathways to assist in growth and metastasis. There are two main ways in which this can be accomplished: downregulation of tumor suppressors and upregulation of proto-oncogenes (oncogenes, for short). Tumor suppressor genes exist to keep the growth and normal function of cells in check. Therefore, they are often downregulated in cancer cells, so that abnormal activities like exacerbated proliferation can   5 continue without disruption. Oncogenes have the opposite effect: they are present in the cell to stimulate growth and other aberrant characteristics. Therefore, oncogenes are usually upregulated in cancer and constitutively activated to promote the growth and spread of cancer cells.  Metastatic melanoma harbours many pathogenic mutations, as melanoma tumors are heterogenous cell populations due to the polyclonality of melanocytes (Shain & Bastian, 2016). Attempts have been made to summarize the complex process of melanoma progression. The evolution of melanoma will be discussed in terms of a simplified linear model, as depicted in Figure 1.1. Melanoma progression is often described as the evolution of a precursor lesion which eventually leads to metastatic melanoma, and the following will trace the path from melanocyte to metastatic melanoma based on our current knowledge. In 1984, Clark et al. described melanoma tumor progression as having six evident lesional steps: (1) the common acquired melanocytic nevus; (2) a melanocytic nevus with lentiginous melanocytic hyperplasia (aberrant differentiation); (3) a melanocytic nevus with aberrant differentiation and melanocytic nuclear atypia (dysplastic nevi); (4) the radial growth phase of primary melanoma; (5) the vertical growth phase of primary melanoma; and (6) metastatic melanoma (Clark et al., 1984). Figure 1.1 roughly depicts this progression. This Clark model of melanoma progression is now outdated and there are problems associated with it due to the diagnostic difficulty of intermediate lesions, given that they present both benign and malignant characteristics (Elder, 2016). Nonetheless, it is useful in that it provides a simplistic outline of a complex process. We will begin with melanocytic nevi, commonly known as moles. They are benign proliferations of melanocytes which often develop during the first 20 years of life and rarely   6 progress to melanoma. As mentioned previously, melanocytes rarely divide; therefore, for a nevus to form, an initiating mutation must occur in a melanocyte to stimulate proliferation. Common somatic mutations which stimulate growth occur in N-RAS (15%) and BRAF (50%) genes (Miller & Mihm, 2006). These abnormal proteins cause constitutive activation of the ERK-MAPK pathway (Miller & Mihm, 2006). For example, a mutation in BRAFV600E provides constitutive activation of the MAPK pathway, even in the absence of upstream signaling. ERK kinases activate transcription factors which activate genes such as CD1 which are involved in the promotion of proliferation and cell-cycle progression (Miller & Mihm, 2006). Observations of melanocytic nevi suggest that only a single driver mutation is required to initiate formation of a nevus, such as BRAFV600E (Shain & Bastian, 2016).  Mutated melanocytes proliferate for a limited time, forming a nevus, before entering a cellular senescence-like state. Some melanocytes within a nevus have been shown to proliferate, but this growth is offset by apoptosis due to oncogenic stress and immune system regulation by chronic lymphocytic infiltrates (Shain & Bastian, 2016). Melanocytic nevi do not pose a threat at this stage. The next lesion, although the subject of much debate and ambiguity, is the dysplastic nevus (DN). Dysplasia is defined as the abnormal growth or development of a tissue (Shain & Bastian, 2016). DN are intermediate neoplasms which microscopically present atypical cellular or architectural features (Shain & Bastian, 2016). They are enlarged flat nevi displaying variable pigmentation, asymmetry and/or irregular borders (Shain & Bastian, 2016). The presence of multiple driver mutations has been documented in DN, including BRAFV600E, BRAFV600K, BRAFK601E and NRAS (Shain & Bastian, 2016). DN also contain mutations in the telomerase promoter TERT, which enriches TERT expression   7 (Nagore et al., 2016). The TERT mutation has been linked to the promotion of malignant transformation (Liu et al., 2016). DN do not always progress to melanoma; however, studies have revealed that the presence of multiple DN increases an individual’s risk for developing melanoma in the future (Shain & Bastian, 2016). Following DN is melanoma in situ (MIS). MIS is melanoma which is confined to the epidermis with melanocytes having enlarged nuclei and proliferating in an irregular pattern (Shain & Bastian, 2016). Many of the same mutations are found in MIS as in DN, albeit at a higher frequency. There are a larger number of TERT mutations and, unlike benign lesions, NRAS and other driver mutations outweigh BRAFV600E mutations (Shain et al, 2015). Mutations in NF-1 are common in older patients (Shain & Bastian, 2016). However, MIS are still dependent on the environment within the epidermis for survival and growth. MIS proliferation is thought to be controlled by immune cells or cell-autonomous mechanisms (Shain & Bastian, 2016). MIS must acquire additional genetic alterations to leave the epidermis and become invasive. For a melanoma to become invasive, it may also have to escape immune surveillance (Shain & Bastian, 2016). The next stage of melanoma remains within the skin, but begins to invade the dermis. There are two types of growth that have been described, the radial growth phase (RGP) and the vertical growth phase (VGP). Metastatic cells first begin to break the boundaries of the basement membrane in the RGP and spread outward parallel to the surface of the skin, creating a plaque (Miller & Mihm, 2006). During the RGP, although some cells can invade and survive in the dermis, they are not tumorigenic at this stage (Miller & Mihm, 2006). Melanoma cells do not become fully established beyond the basement membrane within the dermis until the VGP commences, which is the downward growth of cells. VGP is when   8 tumorigenicity occurs in the dermis, along with mitogenicity (Miller & Mihm, 2006). It is possible for the RGP to be skipped, as is the case for nodular melanoma which is the most aggressive type of melanoma.  On top of the already established driver mutations of TERT and MAPK pathways, a mutation found to first occur at this point in melanoma progression is the bi-allelic inactivation of CDKN2A (Shain & Bastian, 2016). CDNK2A encodes for tumor suppressors p16 and p14ARF which are meant to block the cell cycle transition from G1 to S-phase and control cellular proliferation (Shain & Bastian, 2016). Without these tumor suppressors functioning, the G1/S checkpoint of the cell cycle is overridden by invasive melanoma, thereby promoting uninhibited cellular proliferation. Invasive melanoma is also associated with mutations in the SWI/SNF chromatin-remodeling complex (Shain & Bastian, 2016). The function of this complex remains unclear, but it is thought to maintain genomic integrity; without proper functioning of SWI/SNF, extensive chromosomal aberrations occur (Shain & Bastian, 2016). In late-stage primary melanoma, mutations in the tumor suppressors p53 and PTEN are known to occur. However, p53 mutations are not as frequent in melanoma as in other cancers. Scientists have reasoned that mutation of p53 may be unnecessary, since mutation in CDKN2A renders p14ARF inactive, which normally functions to inhibit MDM2, the major ubiquitin ligase that degrades p53 (Miller & Mihm, 2006). When p14ARF is not present to sequester MDM2, p53 is degraded and its tumor suppressive role is inhibited (Miller & Mihm, 2006). A more common mutation in late-stage primary melanoma is that of the phosphatase PTEN which normally inhibits AKT activation by inhibiting growth factors signaling (Miller & Mihm, 2006). When PTEN is deleted and not functioning properly,   9 active AKT levels increase, which prolongs cell survival by inactivating the pro-apoptotic protein BAD and increasing cell proliferation by upregulating cyclin D1 expression (Miller & Mihm, 2006). Growth of cancer cells is important; however, the characteristic that makes melanoma so lethal is its ability to metastasize to different areas of the body. The beginning of this process involves the transition from the radial growth phase (RGP) to the vertical growth phase (VGP)—once melanocytes have begun to move beyond their normal location within the epidermis. To accomplish this “break out”, several cellular alternations must occur, in a process termed “epithelial to mesenchymal transition” (EMT). The loss of epithelial cellular characteristics and gain of mesenchymal, multipotent stromal cell characteristics bestows cells with invasive and migratory capacity. Melanocytes are polarized epithelial cells within the epidermis that are normally bound to each other via tight junctions, adherens junctions and desmosomes (Pearlman et al., 2017). The purpose of these cell-cell connections is to promote the integrity of the epidermis and to anchor melanocytes in place. Invasion and spread of melanoma cells occurs when there is a modification of normal cell adhesion. E-cadherin is a multifunctional transmembrane protein which is expressed by healthy melanocytes (Miller & Mihm, 2006). It connects melanocytes to each other and keratinocytes by cell-cell contacts and forms connections with the actin cytoskeleton, which inhibits proliferation, suppresses melanoma markers and promotes branching of melanocytes (Miller & Mihm, 2006). Normal melanocyte structure and function is reliant upon the epithelial histologic features apical-basolateral polarization, cell-cell adhesion and basement membrane integrity (Pearlman et al., 2017). The first two are controlled by E-cadherins. The basement membrane is a complex   10 extracellular matrix that separates the epidermis from the dermis and is composed of major glycoproteins such as collagen (Pearlman et al., 2017). Melanocytes are specialized and differentiated cells meant to live and function within the epidermis, and the basement membrane is the barrier that keeps them in their proper place. During EMT, these epithelial histologic features are lost.  One of the major markers of EMT is the loss of E-cadherin (Miller & Mihm, 2006). Without E-cadherin, cell-cell adhesion and polarity of melanocytes is lost. At this stage, melanocytes lose their attachment to and orientation within the epidermis. An increase in N-cadherin also occurs, which is a mesenchymal protein that promotes migration as melanocytes can now interact with N-cadherin-expressing cells, such as dermal fibroblasts and the vascular endothelium (Miller & Mihm, 2006). This is also a property that supports transendothelial migration. Finally, for melanocytes to be able to leave the epidermis and enter the dermis, the basement membrane must be penetrated. This is thought to be accomplished in part by the expression of alphaVbeta3 integrin (Miller & Mihm, 2006). This integrin activates the enzyme metalloproteinase-2 which functions in degradation of the collagen in the basement membrane (Miller & Mihm, 2006). It has also been implicated in the reorganization of the cytoskeleton to promote motility of melanoma cells (Miller & Mihm, 2006). These alterations in protein expression bestow this aberrant cell population with the ability to break free from the confines of the basement membrane of their microenvironment, the epidermis. At this point, the melanoma is now considered invasive, and the risk of further complications and mortality increases. The depth of invasion of the primary tumor, known as tumor thickness, is a major predictor of melanoma survival.   11 Another protein known to be linked to the metastatic potential of melanoma cell is catenin. One of its roles is to link the intracellular regions of cadherins to the actin cytoskeleton (Miller & Mihm, 2006). Normally, catenin is phosphorylated by GSK3and targeted for degradation in the cytoplasm (Miller & Mihm, 2006). However, wingless-type mammary tumor virus integration-site family (WNT) proteins block GSK3and allow -catenin to translocate to the nucleus and transduce signals which increase survival and proliferation by upregulating MITF and cyclin D1 (Miller & Mihm, 2006). This pathway is known to be activated in melanoma, and sometimes -catenin can be mutated to increase nuclear localization and promote these cancerous signals in melanoma cells (Miller & Mihm, 2006). The final stage of melanoma is the metastatic variety, and there are three classifications of metastases: locoregional, regional and distant (Elder, 2016). The first, locoregional, is known as satellite or in-transit metastases, characterized by distance from the primary tumor site (5cm+) (Elder, 2016). Metastases is termed regional when it has spread further and is found in the lymph nodes (Elder, 2016). The third classification is characterized by the dissemination and establishment of malignant melanoma cells in distant organs such as the lungs, liver and brain (Elder, 2016; Miller & Mihm, 2006). At this point, melanoma cells have developed the ability to live beyond the dermis and often travel to foreign environments by way of the lymphatic system or bloodstream. They then seed themselves into a new organ and/or tissue, take up residence and proliferate to form metastases. To add to the complexity of melanoma, it is important to note that some melanomas have been observed "de novo", meaning there is no apparent primary tumor from which the   12 metastatic lesions arose (Weatherhead et al., 2007). The genes and proteins mentioned only scratch the surface of melanoma progression, as our knowledge continues to expand and improve each year. The scientific community is just beginning to explore the inner workings of metastatic melanoma. Uncovering additional genes and proteins and determining their functional significance in melanoma will aid in our understanding of this multifaceted process and perhaps reveal potential targets for treatment.  Figure 1.1 Biological events in the progression of melanoma as described by the Clark model. [Reproduced with permission from Miller & Mihm, 2006, Copyright Massachusetts Medical Society.] 1.1.5. Clinical staging of melanoma and histological subclassification The linear Clark model is a simplification of the path from precursor lesion to metastatic melanoma. These histopathological stages are useful when visualizing the process of melanoma evolution, but it is uncommon for melanomas to follow this exact progression. It is now known that the progression is much more complex and variable, depending on the subtype of melanoma (Shain & Bastian, 2016). The American Joint Committee on Cancer   13 (AJCC) has developed a new staging system that does not place as much emphasis on the Clark model. The AJCC utilizes a system of staging for cutaneous melanoma which encompasses three categories termed TNM staging. The first is T, which describes localized or primary melanoma (stage I and II) in terms of tumor thickness and ulceration, which at this point are the dominant independent predictors of survival (Rigel et al., 2011). A thicker primary tumor measured in millimeters with an absence of intact dermis (ulceration) is correlated with a worse prognosis (Rigel et al., 2011). Once the melanoma has advanced and spread beyond the skin and into lymph nodes, is it termed stage III, which is when N comes in to play. N defines the number of metastatic nodes and nodal metastatic burden (Rigel et al., 2011). At the fourth and final stage of melanoma, M is used, which designates the site of metastases (Rigel et al., 2011). The location of metastases dictates the prognosis, with increasing disease severity from distant skin, subcutaneous and nodal metastases, to lung metastases and to all other visceral metastases (Rigel et al., 2011). Additionally, serum lactate dehydrogenase (LDH) is an independent adverse predictor of survival in stage IV metastatic melanoma (Rigel et al., 2011). The importance of LDH stems from the hypoxic environment of melanoma cells. The production of adenosine triphosphase (ATP) by oxidative phosphorylation cannot occur in low oxygen environments. Instead, melanoma cells have been known to utilize LDH to convert pyruvate to lactate for energy production (Palmer at el., 2011). In addition to AJCC staging, melanomas are often histologically subclassified. There are four main morphologic types of cutaneous melanoma – superficial spreading (SSM), nodular (NM), lentigo maligna (LMM) and acral lentiginous (ALM). SSM accounts for   14 approximately 70% of all diagnosed melanomas and is the most common type of melanoma in Caucasian populations (Gray-Schopfer et al., 2007). SSM is flat in appearance with a predominant radial growth phase and is associated with incidences of severe sunburns early in life (Egger et al., 2013; Gray-Schopfer et al., 2007). NM is named appropriately as it presents as an elevated nodule on the skin which is sometimes devoid of pigment (Gray-Schopfer et al., 2007). It is considered the most aggressive form of melanoma, as it lacks a radial growth phase, rapidly progressing to the vertical growth phase sometimes without a preexisting lesion (Kalkhoran et al., 2010). LMM arises from premalignant melanoma in situ (or lentigo maligna); it is slow growing and presents as a patch of discoloured skin that changes shape and colour over time, eventually becoming invasive (McCourt et al., 2014). LMM is linked with lifetime chronic sun exposure as it is commonly found on sun-exposed skin in older adults (Gray-Schopfer et al., 2007). ALM is the most common type of melanoma in darker skinned populations, accounting for 50% of cases (Gray-Schopfer et al., 2007). The most common locations for ALM occurrence are the soles of the feet, the nail bed or the palms of the hands (Gray-Schopfer et al., 2007). 1.2. Current metastatic melanoma therapeutic interventions On a positive note, an increased awareness of public health related to skin cancer and improvements in screening and early detection have led to an increase in five-year survival rate. However, this does not seem to stop the incidence and number of deaths caused by melanoma from increasing (Rigel et al., 2011). Unfortunately, there are few effective treatments available that allow for extended survival of metastatic melanoma patients. While less aggressive forms of skin cancer (such as BCC, SCC and early-stage melanoma) can simply be removed via surgical excision (i.e. Mohs surgery) or similar methods, late-stage   15 and metastatic melanoma are problematic and ultimately life-threatening. This makes the treatment considerably more troublesome.  When detected and diagnosed in its early stages, melanoma has a favourable prognosis. Melanoma in situ (stage 0) and stage I and II are locoregional tumors that are still located within the skin and are relatively thin. They can often be removed either via surgical excision, electrodessication and curettage (tissue eradication by electric current and removal by scraping with a curette), or cryosurgery (tissue eradication by freezing) (Balch et al, 2009). Furthermore, certain topical medication and radiation therapy may be utilized. Diagnosis and treatment begins by removal and microscopic examination of the suspect cells. The primary melanoma growth and surrounding normal tissue, and sometimes a sentinel lymph node, is biopsied and the stage of cancer subsequently determined via pathological analysis (Balch et al, 2009).  If melanoma cells are found to have spread to a lymph node, this is now considered stage III metastatic melanoma, and the tumor has progressed to a more aggressive form of disease. Once the melanoma has become malignant in nature, further and more extensive treatment options must be considered and executed immediately in an attempt to prevent further spread. The current approved treatment plan for melanoma consists of a combination of surgical excision and traditional cytotoxic chemotherapy, as well as more recently approved immune-based and targeted therapies (American Cancer Society, 2016; Lovly et al., 2016). Cytotoxic chemotherapy is an all-encompassing, general cancer therapy. It is not specific, targeted or personalized, but can be used to treat many cancer types. Chemotherapy works by annihilating any cell in the body that has a high rate of growth- i.e. cancer cells.    16 A common chemotherapeutic agent that is used to combat metastatic melanoma is Dacarbazine (DITC). DITC is a DNA alkylating agent has long been used as the primary chemotherapeutic drug of choice in patients with metastatic or unresectable melanoma (Bedikian et al, 2006; Lui et al, 2007). However, chemotherapy is not selective for only cancerous cells and exerts its destructive effects throughout the entire body. The cells these drugs target for eradication also include normal, healthy cells within the body that divide often such as: hair follicle cells, gastrointestinal cells, and immune cells. Since the impact of chemotherapy is not confined to one bodily area or cell type, this treatment option produces an abundance of adverse effects. Cytotoxic chemotherapy response rates are quite low ranging from 10-15%, do not provide long-term patient survival, and are often used palliatively (Bhatia et al., 2010; Johnson & Sosman et al., 2015).  Another common type of therapy approved by the FDA for metastatic melanoma treatment are immune-based therapies such as high dose interleukin 2 (HD IL-2) and interferon-alpha (IFN-. These are naturally occurring molecules known as cytokines which are involved in the regulation of the immune system. IL-2, normally secreted by activated CD4+ lymphocytes, does not have a direct effect on cancer cells. Instead, it acts on immune cells (i.e. cytotoxic T cells, natural killer (NK) cells, macrophages) to stimulate their proliferation and function to attack melanoma cells (Bhatia et al., 2009). IFN- has similar effects to IL-2 on immune cells, along with direct anti-proliferative effects on tumor cells (Rigel et al., 2011). Although these therapies were at first thought to be promising alternatives to chemotherapy, they produced similar response rates and were unsuccessful at improving the overall survival for metastatic melanoma patients (Eggermont & Schadendorf, 2009). These compounds also resulted in severe adverse effects, limiting their use to patients   17 who are young and healthy enough to withstand this intense treatment (Johnson & Sosman, 2015). It is worth noting that HD IL-2 has achieved durable complete responses in five to 8% of patients with long-term disease-free survival in patients who can tolerate the toxicity (Johnson & Sosman, 2015).  The successes of these first-generation immunotherapies and the lessons learned from them encouraged researchers to explore more about the immune system in melanoma. The hot topic in the melanoma therapy world over the past few years has been the development of new immunotherapies and immune checkpoint inhibitors. Tumor cells are known to evade the immune system by overexpressing cytotoxic T lymphocyte antigen-4 (CTLA-4) and/or programmed death-1 receptor/ligand (PD-1/PD-L1). These ligands and receptors inhibit T cell response, an important immune system defense against cancer. The next-generation immunotherapies are antibodies which work to inhibit these ligands and receptors, and have improved patient outcomes by ultimately initiating and boosting an antitumor immune response (Johnson & Sosman, 2015).  Recently, scientists and physicians have achieved breakthroughs in targeted therapy for metastatic melanoma by focusing on the genetics of the disease. Important genes that have been studied over the past decade and that are defined at the molecular level by recurrent “driver” mutations include: BRAF, MEK (MAP2K1), NRAS, NF1 and KIT (Johnson & Sosman, 2015; Lovly et al., 2016; Shain & Bastian, 2016). Driver mutations in genes such as these often perpetuate tumorigenesis by causing constitutive activation of aberrant signaling proteins (Lovly et al., 2016).  Approximately 50% of metastatic melanoma patients contain BRAF mutations, making therapies targeting the BRAF protein an effective treatment option (Miller & Mihm,   18 2006). In fact, new targeted therapies were approved by the FDA in 2011 which target the mutant BRAF protein directly known as Vemurafenib and Dabrafenib, and the drug Trametnib for the downstream MEK protein (Jang & Atkins, 2013). Although these targeted drugs are promising, the anti-tumor activity does not last beyond 6-8 months when disease progression resumes due to acquired resistance (Jang & Atkins, 2013; Robert et al., 2015). To delay the onset of acquired resistance to such therapies, combination protocols have been created (Day and Siu, 2016; Robert et al., 2015). For example, a therapeutic regimen which combines BRAF and MEK inhibitors has been proven to be more effective and result in improved response rates and survival, as opposed to single-agent BRAF inhibitors (Johnson & Sosman, 2015; Robert et al., 2015). The downside of these targeted genetic and immunological therapies is that they do not work for all melanoma patients. Some patients’ melanoma does not harbor these specific mutations or rely on the evasion of the immune system to survive and persist within the human body. To improve the efficacy of melanoma treatment and expand treatment options for patients, scientists and physicians have directed their focus and research efforts towards the genetic signature of metastatic melanoma. Expanding the scientific knowledge of the genetics and molecular mechanisms of melanoma is necessary to uncover more treatment options.  1.2.1. Biomarkers of melanoma Many genes have been and will continue to be examined and tested to determine their usefulness for metastatic melanoma patients as both potential targets for treatment and prognostic biomarkers. It is also hoped that which therapy to choose and monitoring response to therapy will be aided by biomarkers once more become known and tested (Rigel et al.,   19 2011). Biomarkers are factors present within a tumor or host which correlate with patient prognosis and biological behaviour of the cancer (Gogas et al., 2009). Biomarkers have the potential to aid in detection, prognostication, and monitoring of disease progression (Palmer et al., 2011).  Current examples of histopathological biomarkers approved by the AJCC in melanoma for prognosis are tumor thickness, ulceration, and mitotic activity. Another is S100B, an immunohistochemical marker commonly used for initial histopathological diagnosis of melanoma which was used as a control in this study for the TMA analysis (Levine & Fisher, 2014). Recently, serologic LDH was found to be an independent prognostic factor and therefore useful biomarker in patients with late-stage melanoma (Palmer et al., 2011).   Additionally, the BRAFV600E mutation is currently being used as a biomarker to determine if melanoma patients will benefit from BRAF-targeted treatments (Spagnolo et al., 2015). To qualify for the use of drug treatments targeting BRAF, patients must be positive for the BRAFV600 mutation (Spagnolo et al., 2015). This is yet another use of biomarkers: to narrow down treatment options for patients by predicting if a treatment will be beneficial for an individual. Unfortunately, as with treatment, the benefit of these prognostic indicators varies due to the heterogeneity of melanoma.  The elusive nature and resistance of melanoma to treatment accentuates the need to develop novel therapeutic strategies for metastatic melanomas. Cutaneous melanoma is the solid tumor type with the highest frequency of genetic mutations, and its complex mutational landscape has prompted the pursuit of targeted therapies (Davar et al., 2013). A wider picture including more genes and molecular interactions is required to further understand this disease   20 to advance treatments and prognostic techniques. In a sea of possibilities, this thesis work chose to investigate and focus on one gene in metastatic melanoma: Sineoculis homeobox homolog 1 (SIX1). 1.3. The SIX1 gene and Six1 protein  Our interests first turned to the gene SIX1 when the Molecular Medicine Lab’s transcriptome microarray found it to be upregulated in metastatic melanoma cell lines and clinical samples in comparison to normal melanocytes, skin and nevi. The focus of this study will be on SIX1 and its function in melanoma, and this section will focus on the background, structure and normal function of the SIX1 gene and Six1 protein. 1.3.1. Background The sine oculus homeobox (SIX) gene family is made up of six genes (SIX 1 to 6) in humans and are highly conserved throughout evolution. This gene family was originally discovered in Drosophila melanogaster, from which it was cloned in 1994 (Cheyette et al., 1994). These genes were found to be essential for compound-eye formation and are a part of the retinal determination network in Drosophilia (Cheyette et al., 1994; Ohto et al., 1999). Only three genes are present in Drosophilia, whereas vertebrates have six genes, suggesting that the genes diverged from so to SIX1 and SIX2, optix to SIX3 and SIX6, and DSix4 to Six4 and Six5 (Kumar, 2009). The SIX genes have been studied in species such as the human, mouse, chicken, frog, zebrafish, nematode and fly (Kawakami et al., 2000). However, they are most well characterized in mice. Homeobox genes are master regulators of events during embryonic development. They encode for homeoproteins which function as transcription factors (Hu et al., 2008). Their functional roles include the control of body plan specification and cell fate   21 determination (Haria & Naora, 2014; Hu et al., 2008). The SIX genes function during embryonic development and have been shown to maintain the differentiated state of tissues (Boucher et al., 2000). We will be focusing on the human gene SIX1, which is a homolog of the gene Drosophila sine oculis (So) (Wu et al., 2014).  Limited expression analysis of SIX1 is available in humans, and so most information available details the murine homolog of SIX1. The principal expression sites of Six1 in mouse embryos are in the otic vesicles, nasal placodes, branchial arches, Rathke’s pouch, dorsal root ganglia, somites, nephrogenic cords and limb mesenchyme (Kawakami et al., 2000). Following the completion of organ development, SIX1 is downregulated (Ford et al., 1998; Patrick et al., 2013). SIX1 is not normally expressed in adult tissues, with the exception of normal skeletal muscle, salivary gland, lung, trachea, and kidney (Ford et al., 1998).  1.3.2. Structure By using a rodent-human somatic cell hybrid panel, it was determined that SIX1 is located within a cluster of related genes on chromosome 14 (Boucher et al., 1996). The SIX1 gene is composed of two exons that are 833 and 543 base pairs respectively, and a single 2,052 base pair intron. Its coding region transcript total length is 1,376 base pairs (Figure 1.2) (Ford et al., 2005). The Six1 protein consists of two evolutionarily conserved domains: the SIX domain (SD) and nucleic acid recognition homeodomain (HD) (Figure 1.3). The SD binds protein and is involved in protein-protein interactions, while the HD binds DNA once in the nucleus (Hu et al., 2008). The SD may also contribute to DNA binding specificity (Kawakami et al., 2000). The Six1 protein has a total length of 284 amino acids and molecular weight of 32   22 kDa. The SD is an N-terminal 115 amino acid domain and the HD is a 60 amino acid six-type helix-turn-helix DNA-binding homeodomain (124-183 HD). The homeodomain DNA sequence is termed the “homeobox”. Six1 is post-translationally modified via the attachment of phosphate groups (which can include a complex of 5’-phospho-DNA), making it a phosphoprotein (Ford et al., 2000). It has also been found to be a hyperphosphorylated phosphoprotein during mitosis (Ford et al., 2000).  Figure 1.2 Visual representation of the SIX1 gene The SIX1 gene is composed of two exons that are 833 and 543 base pairs respectively, and a single 2,052 base pair intron. Its coding region transcript total length is 1,376 base pairs. Start and stop codon positions are shown. Start codon: ATG (mRNA: AUG); stop codon: TAA (mRNA: UAA) [Reproduced with permission from Ford et al., 2006].   Figure 1.3 Diagrammatic depiction of the transcription factor homeoprotein Six1. The Six1 protein has a total length of 284 amino acids and molecular weight of 32 kDa. It contains an N-terminal 115 amino acid Six domain (SD) and a 60 amino acid six-type helix-turn-helix DNA-binding homeodomain (124-183 HD) [Reproduced with permission from Ford et al., 2006]. 1.3.3. Function and molecular mechanisms SIX1 is a human homeobox gene that encodes for a transcription factor, which plays a key role in normal embryogenesis (Wu et al., 2015). The normal activity of the Six1 protein   23 takes place in the nucleus, where it forms a transcription factor complex with eyes absent (EYA1-4) proteins to transcribe additional genes into messenger RNA (mRNA) and ultimately create new protein when prompted to by the cell (Blevins et al., 2015). EYA proteins are nuclear cofactors of Six1 which possess tyrosine and serine-threonine phosphatase activity (Li et al., 2003; Xu, 2013). Six1 does not have an intrinsic activation domain, therefore Six1’s function of transcriptional activity mediation is normally dependent on the presence of EYA proteins (Blevins et al., 2015; Patrick et al., 2013). A study of Six1’s crystal structure discovered that Six1 and EYA binding is mainly reliant on a single amphipathic -helix on Six1 which binds to a hydrophobic cleft on EYA (Patrick et al., 2013).  The Six1 protein contains a nuclear localization signal which guides the complex to the nucleus via active translocation after binding to EYA via its protein-binding SD (Li et al., 2003). In the absence of EYA proteins, Six1 is bound by DACH1 which prevents target gene expression by recruiting a co-repressor complex of Six1 (Li et al., 2003). DACH1 repression of Six1 can be reversed by EYA phosphatase activity which transforms DACH1 into a co-activator (Li et al., 2003). EYA and its protein phosphatase activity is required for the recruitment of other co-activators such as CREB-binding protein (CBP) and ultimately the initiation of transcription by Six1 (Blevins et al., 2015; Li et al., 2003). This transcription complex activates genes such as those involved in cell proliferation and survival like c-Myc and Gdnf (Li et al., 2003) (Figure 1.4).  Six1 plays an instrumental role in eye development in Drosophila, the species within which it was originally discovered (Wu et al., 2015; Blevins et al., 2015). It has been shown to play a vital role in organogenesis during human embryo development, solidifying its role   24 as a developmental gene during embryogenesis after a variety of gain and loss of function mutations were performed (Wu et al., 2015; Blevins et al., 2015). Six1 has been implicated in the development of the eye, auditory system, lung, kidney, gonads, sensory organs, olfactory organs, taste organs, craniofacial structures and skeletal muscle (Wu et al., 2015; Blevins et al., 2015).  Extensive research on Six1’s role in muscle development has taken place. Six1-deficient mice displayed large muscle hypoplasia and died at birth, suggesting Six1 is important in muscle development (Laclef et al., 2003). The function of Six1 was found to be an important regulator during tissue regeneration and replenishment of the satellite cell (SC) (stem cell) pool after skeletal muscle trauma in adult tissues (Grand et al., 2012). Six1 has been shown to control the myogenic regulatory factors MyoD and Myogenin by binding to their promoters or enhancers at the MEF3 site, which is thought to account for its ability to control proper skeletal muscle regeneration (Grand et al., 2012; Wu et al., 2015). The critical regulation of SC self-renewal is linked to Six1’s regulation of the ERK-1 pathway (Grand et al., 2012). Furthermore, the Six1-Eya1 complex has also been shown to convert slow-twitch muscle fibers to fast-twitch muscle fibers by binding to the MEF3 site on the aldolase A promoter (Grifone et al., 2004; Kumar, 2009; Wu et al., 2015).  Six1 has also been implicated in the development of the auditory system. Six1+/- mutant mice exhibited hearing loss and a complete loss of Six1 caused malformations of the outer, middle and inner ears; the constituents of the inner ear did not form in these mice (Zheng et al., 2003). The Six1-EYA1 complex has been shown to control inner ear development in mammals by having regulatory power over Fgf3, Fgf20, Bmp3, Nkx5.1 and Gata3, which are all important for proper ear formation (Zheng et al., 2003). Additionally,   25 although the functional role of Six1 in eye formation is currently unknown, a genome-wide association identified SIX1 variants to correlate significantly with optic disc area and vertical cup-disc ratio (VCDR) (Ramdas et al., 2010; Wu et al., 2015). These factors are important for normal eye development, and therefore SIX1 may play a role in this process.  Moreover, Six1 is known to play a role in the formation of olfactory organs and taste organs in mice. Both Six1 and Six4 are required for the formation of the olfactory placode, olfactory epithelium and neurogenesis (Ikeda et al., 2007; Wu et al., 2015). Basal progenitors and apical progenitors are not present in Six1 deficient mice, which results in the loss of olfactory receptor neurons and disorder of the olfactory epithelium structure (Ikeda et al., 2010; Wu et al., 2015). The absence of Six1 also causes taste bud or papillae malformation as well as the ability to taste (Suzuki et al., 2010; Wu et al., 2015). Embryonic lung development was also found to require Six1, as a study provided evidence that Six1 coordinates Shh-Fgf10 signaling during lung formation (El-Hashash et al., 2014). Proper proliferation and differentiation of the lung during epithelial, mesenchymal and vascular development required the presence of Six1 (El Hashash et al., 2014).  Mouse neonates lacking Six1 were born deficient of a kidney and a thymus and had highly disorganized craniofacial structures such as the inner ear, nasal cavity, craniofacial skeleton, lacrimal and parotid glands (Laclef et al, 2003). Connections between Six1 and most of these body structures are lacking, but Six1 in kidney development is the most well-studied. In mice lacking Six1, the uretic bud does not invade into the metanephric mesenchyme, which is critical for kidney development and leads to apoptosis of the mesenchyme (Wu et al., 2015; Xu et al., 2003). Additionally, essential regulators of renal   26 development Pax2, Six2 and Sall1 were downregulated in mice lacking Six1, suggesting Six1 may control these proteins (Wu et al., 2015; Xu et al., 2003).  Due to Six1’s critical roles during development it is not surprising that Six1 has been implicated in the cell cycle. Six1 transcript and protein levels have been found to be the highest during S-phase, and it has been associated with the expression of numerous cell cycle proteins such as cyclin A1 and cyclin D1, indicating that it plays a role in cell cycle progression (Ford et al., 2000). The Six1 protein is hyperphosphorylated by the casein kinase II (CK2) protein during mitosis which downregulates its activity and interferes with the DNA binding ability of Six1 (Ford et al., 2000). It is finally degraded via ubiquitin-mediated proteolysis during late mitosis by the anaphase-promoting complex (APC) (Christensen et al., 2007). Six1 is therefore regulated to be active during the cell cycle from the G1/S boundary through early mitosis (Christensen et al., 2007).   27  Figure 1.4 A diagram depicting the Six-Eya-Dach-regulated activation of target genes on DNA in the nucleus. The recruitment of Eya phosphatase by Six1 co-recruits more co-activators such as CBP, which then allows the expression (turning “ON”) of certain target genes responsible and required for pro-metastatic abilities like survival and proliferation. On the contrary, when Eya is absent, Dach is able to inhibit the expression of these pro-growth target genes by attracting and assembling a co-repressor multiplex. [Reproduced with permission from Li et al., 2003]. 1.4. SIX1 in developmental disorders As touched upon above, SIX1 is a key gene and protein involved in embryological development during a human fetus’ period within the womb. This is a time in which a human embryo undergoes a significant number of transformations and maturation. A high level of growth is observed, along with alignment of certain germ layers and structures. Since SIX1 is so critical during embryonic development, it is not surprising that when mutations occur in the SIX1 and EYA1 complex, this results in developmental problems. These genes are known to have convoluted expression patterns within embryonic tissues; if a slight mishap occurs in the development of this highly detailed and complicated spatio-temporal gene   28 setup, issues arise (Blevins et al., 2015; Xu., 2012). This multiplex system of the Six1-EYA partners along with other genes within a human tissue is important for proper development and can result in phenotypic abnormalities if a mistake in the replication machinery occurs or other factors are out of place (Blevins et al., 2015; Xu., 2012). A simple error can cause a domino effect which disrupts downstream processes and consequently a failure in the proper execution of developmental programs results. One result of a developmental glitches in this intricate developmental scheme is the rare genetic syndrome known as Branchio-oto-renal (BOR) syndrome, which occurs in approximately 1:40000 births (Jalil et al., 2014). This syndrome is characterized by several congenital anomalies, including: a passage from the throat to the outside of the neck that is not usually there (branchial fistulae), abnormal opening, cyst or mass in the tonsil area, hearing loss and outer ear (oto) malformations such as preauricular pits and kidney (renal) hypoplasia or dysplasia (Blevins et al., 2015; Rodrigues, 2003; Smith, 1999, 2015; Wu et al., 2015). BOR is an autosomal dominant disorder caused by mutations in EYA1 and Six1 (Jalil et al., 2014). When EYA1 is mutated in BOR syndrome, the mutations are normally located within the EYA domain, within which lies the Six1 binding site (Smith, 1999, 2015). The mutations in SIX1 often disrupt the Six1-DNA binding activity and act to inhibit normal organ development (Smith, 1999, 2015).  1.5. SIX1 in cancer  Over the past two decades, investigative attention has turned towards SIX1’s role in human cancer. SIX1 was first found to be overexpressed in mammary carcinomas and to correlate strongly with metastatic breast cancer by Ford et al. in 1998. Since then, there have been numerous studies focusing on SIX1 in a number of cancers. Overexpression of both the   29 gene transcript and protein has been shown to occur in a variety of neoplasms such as colorectal cancer, hepatocellular carcinoma, rhabdomyosarcoma, pancreatic cancer and Hodgkin lymphoma (Blevins et al., 2015; Nagel et al., 2015). Furthermore, aberrant expression of the Six1 transcript and protein has been correlated with poor prognosis in numerous cancers such as colorectal, lung, cervical, ovarian and pancreatic cancers (Blevins et al., 2015). This abnormal upregulation of SIX1 across cancers led scientists to hypothesize that SIX1 may be an oncogene driving malignant phenotypes of cancer cells. In turn, this inspired numerous studies with a focus on several hallmarks of cancer. The hallmarks in which SIX1 has been found to play a role include sustaining proliferative signaling, evading growth suppressors, genome instability and mutation, resisting cell death, and activating invasion and metastasis (Blevins et al., 2015; Hanahan & Weinberg, 2011; Wu et al., 2015). Overexpression of Six1 in a mouse model of breast cancer resulted in a larger tumor burden (Coletta et al., 2004). Furthermore, multiple studies have shown that Six1 is a prognostic marker and oncogenic driver of cancer growth and metastasis in breast, colorectal, cervical, and ovarian cancer (Wu et al., 2015; Blevins et al., 2015).  These studies have proven that Six1 is a critical regulator of tumorigenesis in many types of cancer. However, Six1’s role in melanoma has not been investigated or reported.  1.6. Thesis outline  1.6.1. Rationale Melanoma patients currently present with a poor prognosis due to the aggressive nature of the disease. The low remission rate and minimal therapeutic success in metastatic melanoma patients has prompted scientists and physicians to search for novel, more effective   30 treatment options. Melanoma is a genetically complex disease, including many mutations, upregulations, downregulations and abnormal function of genes and proteins. Often the focus of new treatments is genetic in principle—such as targeting an aberrantly expressed, oncogenic gene and ultimately protein. More genetic targets need to be explored to contribute to the knowledge and understanding of the complicated molecular biology of metastatic melanoma. The more information we know about the genetics of and protein function in a cancer, the more we will be able to improve and expand treatment options specific to a disease and personalized for patients. We have chosen to investigate the role of the transcription factor Six1 in melanoma. SIX1 is an important developmental gene which is normally not expressed in adult human tissue. Studies on multiple cancers have found aberrant Six1 expression to be associated with cancerous phenotypes in vitro and in vivo. Additionally, Six1 expression has been reportedly linked to worse prognosis in various cancer types. However, Six1 has not been investigated in melanoma. A previous microarray analysis by the Molecular Medicine Lab found the SIX1 gene to be upregulated in melanoma patient biopsies in comparison to normal skin and normal nevi. Taken together, these findings led us to hypothesize that Six1 may also be involved in the pathogenesis of melanoma. Therefore, the purpose of the current study was to investigate Six1 expression levels in melanoma cell lines and clinical samples, its biological function and clinical relevance in melanoma. 1.6.2. Research hypotheses  We predict that SIX1 plays an oncogenic role in the pathogenesis of melanoma. It is predicted that Six1 is associated with pro-proliferative or pro-metastatic mechanisms in metastatic melanoma.   31 We further hypothesize that the out-of-context expression of the Six1 protein could serve as a prognostic marker for melanoma. 1.6.3. Specific aims    To test the aforementioned hypotheses in relation to the pathogenesis of melanoma, the following objectives were formulated: (1) To assess the expression of the SIX1 gene and protein in melanoma cell lines and clinical samples. (2) To investigate the functional significance and pathogenic role of Six1 in metastatic melanoma cells in vitro.   (3) To evaluate Six1 protein expression in melanoma patient biopsies and its clinical relevance in relation to prognosis and clinicopathological characteristics.               32 2 Materials and Methods 2.1. Established metastatic melanoma cell lines and growth conditions (cell culture)   The established metastatic melanoma cell lines A375 and RPMI-7951 were utilized in all functional assays as they were found to have high endogenous SIX1 expression by qPCR. Both A375 and RPMI-7951 were derived from skin tissue at a metastatic lymph node site. A375 cells came from a malignant melanoma case in a 54-year-old Caucasian female. An 18-year-old Caucasian female with malignant melanoma donated cells to establish the RPMI-7951 cell line. Both cell lines are epithelial in morphology and were purchased from American Type Culture Collection (ATCC), which noted them to be suitable transfection hosts.   The cells were initially cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% (A375) or 15% (RPMI-7951) Fetal Bovine Serum (FBS). Medium and FBS were obtained from Gibco. Following transfection with shRNA plasmids and after allowing time for the knockdown to occur and the cells to acclimate, puromycin was added to the medium at a concentration of 2μg/mL. In the case of RPMI-7951, a gradually increase of puromycin concentration was required over one week--from 0.5μg/mL to 2μg/mL, doubling each increment for a total of two increases. The cells were cultured in this medium for the duration of the functional experiments, unless otherwise specified. 2.2. RNA extraction and reverse transcription Total RNA was extracted by TRIzol (Invitrogen) reagent and Qiagen RNEasy Mini Kit (RNA Isolation Kit) following the manufacturer’s protocol. The acquired RNA concentration was then measured using a spectrophotometer at 260nm (OD260/28). Subsequently, 100ng of RNA was reverse transcribed into cDNA in a 20μL reaction using a   33 SuperScript VILO cDNA Synthesis Reverse Transcription Kit (SuperScript VILO Synthesis Kit, Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. 2.3. Quantitative real-time polymerase chain reaction (qRT-PCR) First, primer efficiency was tested with 2ng, 4ng, and 8ng of cDNA to create a standard curve, and only primers specific to the particular gene and fell within the range of 90% - 110% efficiency were accepted for use. For the actual reactions, 5ng of cDNA was loaded into each well, along with primers at the required concentration and 10μl of SYBR Select Master Mix (Thermo Fisher Scientific) to create a 20μl total mixture. Quantitative real time polymerase chain reaction (qPCR) was then performed with a real-time PCR system (StepOnePlus and StepOne software v.2.3; Applied Biosystems) following the manufacturer’s instructions. The PCR conditions were 95°C for 10 minutes, 40 cycles of 95°C for 15 seconds, 60°C for 1 minute, and followed by the melt curve step through temperature increment of 0.3 °C from 60°C to 95°C. All experiments were run in triplicate. To quantify gene expression, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an endogenous control. In this way, relative SIX1 expression levels could be expressed as mRNA copies per 1000 GAPDH copies, otherwise known as the ΔΔCT method. The primers used in qPCR are listed in Table 2.1 below. Table 2.1 Primers for qPCR Gene Forward primer (5’→3’) Reverse primer (5’→3’) SIX1 AGCAACTGGTTTAAGAACCG GGTTCTGCTTGTTGGAGGA GAPDH AAGATCATCAGCAATGCCTCC TGGACTGTGGTCATGAGTCCTT    34 2.4. Protein extraction and Western blotting Protein was extracted from cells for Western blotting analysis by directly lysing cell pellets in radioimmunoprecipitation assay (RIPA) buffer (Phosphate Buffered Saline (PBS), 0.1% SDS, 1% NP-40, 0.5% sodium deoxycholate). The mixtures were pipetted and vortexed to assist lysis of the cells. Next, sonication was used to further break down the cells and expose the protein (15amp, five strokes per sample). The lysed cells were then placed on ice for 30 mins, vortexing briefly every 10 minutes. Supernatants were collected by centrifugation at 12,000 rpm for 20 minutes, and stored at -80 °C. Protein concentration was evaluated by PierceTM BCA Protein Assay Kit (Thermo Scientific, MA, USA) according to the manufacturer’s instructions. The absorbance of protein samples at 562 nm was measured using Elx808 Absorbance Microplate Reader (BioTekk Instruments, Winooski, VT). Protein concentration was determined by creating a standard curve on a 96-well microplate from 0 to 1000 μg/mL using bovine serum albumin (BSA). Protein concentrations of the samples were then extrapolated based on standard curve readings.  Western blotting was used to evaluate protein expression. Briefly, protein lysates (~20μg/lane) were incubated at 70 °C for 10 minutes, loaded onto and separated by a 12% SDS-PAGE gel (Thermofischer/Life Technologies) and transferred to polyvinylidene difluoride (PVDF) membrane (Bio-Rad). The membranes were blocked in 5% BSA in Tris-buffered saline containing 0.05% Tween-20 (TBST) for 1 hour at room temperature, washed in TBST three times (5 minutes each), and incubated with primary antibody (Sigma Rabbit polyclonal Six1 HPA001893 1:500 dilution) overnight at 4 °C. Then the membrane was washed in TBST three times (5 minutes each), and incubated with species-specific secondary antibody (1:5000 or 1:40000 dilution) for 1 hour at room temperature. Finally, after washing   35 in TBST three times (5 minutes each), the membrane was scanned on a machine to excite and detect the fluorophore signal now attached to the protein of interest. The quantity of protein was measured against a human actin loading control. The primary and secondary antibodies are listed in Table 2.2.    Table 2.2 List of primary antibodies for Western blot    Antigen Primary Antibody Primary Antibody Dilution Secondary Antibody and Dilution Six11 Rabit Polyclonal (HPA001893, SIGMA) 1:500 Goat anti-mouse antibody (IRDye 680RD, LI-COR), 1:5000 Pro-Caspase-8 Mouse Monoclonal (9746, Cell Signalling) 1:1000 β-Actin Mouse Monoclonal (A5441, Sigma-Aldrich) 1:5000 Bcl-2 Rabbit Polyclonal (2872, Cell Signaling) 1:1000 Goat anti-rabbit antibody (IRDye 800CW, LI-COR), 1:40000 Bax Rabbit Polyclonal (2772, Cell Signaling) 1:1000 Pro-Caspase-3 Rabbit Polyclonal (9662, Cell Signaling) 1:1000 Pro-Caspase-9 Rabbit Polyclonal (9502, Cell Signaling) 1:1000                                                   1Six1 Western Blot primary antibody was also used for IHC   36 2.5. Tissue microarray (TMA)   2.5.1. Source of biopsies Dr. Yabin Cheng provided the TMA, which was constructed before this thesis work began. All the biopsies utilized within TMA were provided by the derma-pathologist Dr. Magdalena, Martinka, from Department of Pathology, Vancouver General Hospital (VGH), BC, Canada. The biopsies were obtained from the VGH archives from 1995 to 2009; the samples were reviewed, marked the selected for having clear diagnosis and intact tissue. Eventually 707 samples were collected and stored at room temperature before they were subjected to TMA construction. Since mucosal melanomas, uveal melanomas and ocular melanomas display distinct clinical courses and molecular phenotypes, and account for a relatively small proportion of melanoma patients, only more prevalent cutaneous melanomas were selected. Non-Caucasian patients were excluded due to differences in genomics, as well as differences in protein expression patterns. The information of patient survival and treatment received were retrieved from British Columbia Cancer Agency, and were updated in September, 2013. 2.5.2. Clinicopathological characteristics of patients The clinicopathological information and follow-up survival data of each selected patient was collected, followed by the establishment of a comprehensive clinical database. The clinicopathological parameters included age, sex, thickness, Clark’s level, AJCC stage, subtype, location, mitotic rate, ulceration, lympho-vascular invasion, lymphocyte response, regression, histological satelitosis, diagnose date, collection date, survival status, death date, death cause and time of follow. The clinical database was established by Dr. Yabin Cheng and double checked by two other researchers, Drs. G Chen, and Z Zhan (Cheng et al., 2014).   37 2.6. Tissue microarray (TMA) immunohistochemistry A TMA of 438 patient biopsies was utilized in this project which as previously constructed by Cheng et al. in 2014. Immunohistochemistry against Six1 protein performed on the TMA by Dr. Yabin Cheng. Polyclonal antibody specific against Six1 was used to stain tissue following hematoxylin staining. Double-blind scoring of the Six1 stained TMA was performed by a registered pathologist at Vancouver General Hospital, recording both nuclear and cytoplasmic scores. Finally, a Kaplan-Meier test was performed to determine five-year disease-specific survival. The X-tile program was utilized to find the cutoff. The data was then analyzed using SPSS 13.0 statistical software (IBM). 2.7. Immunohistochemistry of tissue microarray (TMA)  TMA slides were dewaxed at 55°C for 30 min followed by three five-minute washes in xylene. Tissues were rehydrated by a series of washes in 100%, 95% and 80% ethanol and distilled water. Antigen retrieval was performed by heating the tissues at 95°C for 30 minutes in 10 mM sodium citrate (pH 6.0). Endogenous peroxidase activity was blocked by incubation with 3% hydrogen peroxide for 20 min. Nonspecific antigens were blocked by universal blocking serum (DAKO Diagnostics, Mississauga, Ontario, Canada) for 30 minutes. The TMA slides were incubated overnight with specific antibodies at 4°C. The next day, the slides were incubated with biotin-labeled secondary antibody and streptavidin-peroxidase (DAKO Diagnostics, Mississauga, Ontario, Canada) for 30 minutes each at room temperature with light protection. The samples were developed using 3,3’-diaminobenzidine substrate (DAB, Vector Laboratories, Burlington, Ontario, Canada) and counterstained with hematoxylin. The slides were finally dehydrated and sealed with coverslips. Negative controls were performed by omitting the primary antibody for overnight incubation (Cheng et   38 al., 2014). 2.8. Evaluation of TMA immunostaining The evaluation of each biomarker staining was blindly and independently examined. Three investigators, including experienced dermatologists (Dr. Martinka and Dr. Gang Wang, Department of Pathology and Laboratory Medicine, Vancouver General Hospital), scored the staining independently to obtain a consensus score for each tissue core. Immunostaining of specific protein is defined as either cytoplasmic or nuclear, or whole staining, and graded according to both intensity and percentage of cells with positive staining. The staining intensity was scored as 0, 1+, 2+ and 3+; the percentage of stain-positive cells was also scored into 4 categories: 1 (0-25%), 2 (26-50%), 3 (51-75%), and 4 (76-100%). In the cases with a discrepancy between duplicated cores, the higher score from the two tissue cores was taken as the final score. The expression levels of a biomarker was evaluated by immunoreactive score (IRS) (Remmele & Stegner, 1987), calculated by multiplying the scores of staining intensity and the percentage of positive cells. To ensure that only the melanoma cells were scored for Six1 staining, the adjacent sections of the TMA were stained with S100 to help identify areas on the TMA cores where the melanoma cells were located. χ2 tests were applied to evaluate the statistical differences in demographic and clinical parameters, as well as in the expression levels of the biomarker. Survival time of each patient was calculated from the date of melanoma diagnosis to the death date. The correlation between protein biomarker expression and patient survival was examined by Kaplan-Meier survival curve and log-rank test. SPSS version 16.0 (SPSS Inc., Chicago, IL, USA) software was used for all analyses (Cheng et al., 2014).    39 2.9. Puromycin kill curve In order to determine the appropriate amount of puromycin to utilize during shRNA plasmid knockdown experiments, A375 and RPMI-7951 cells were seeded in separate wells within 96-well plates containing 10% FBS DMEM medium. The cells were allowed to settle overnight until 70-90% confluency was reached. The normal medium was changed to medium containing the antibiotic puromycin, at concentrations of: 0g/mL, 0.5g/mL, 1g/mL, 2g/mL, 4g/mL, 6g/mL, 8g/mL, 10g/mL. The cells were put in the incubator for two weeks, and the medium was changed every three days. The cells were observed under the microscope to determine which concentration was sufficient to kill the entire population of the well (we were seeking the lowest concentration that could accomplish this). Alternatively, the medium could be removed, the wells trypsinized, and trypan blue added; the cells could then be counted to determine the number of live cells remaining in the well. This experiment was performed in quadruplicate and repeated three times. 2.10. Establishing stable transfection SIX1 knockdown clones of metastatic melanoma cell lines with reduced Six1 protein expression Plasmid vector shRNA was utilized for SIX1 knockdown, and was ordered as a glycerol stock (sent as bacteria in agar stab). The plasmids utilized include: pLKO.1 vector containing small hairpin RNA (shRNA) inserts that specifically target human SIX1 mRNA (SHCLNG-NM_005982 Sigma Aldrich, Saint Louis, MO, USA), as well as a non-targeting shRNA control (shCTRL) (SHC002, Sigma Aldrich). This shCTRL is a 3rd generation empty backbone plasmid which is inserted the same way as the coding shRNAs via direct transfection, but in theory does not disrupt cellular processes as it is a replication-incompetent lentiviral vector. This is a convenient method of stable selection of desired cell   40 populations as following a successful introduction of this plasmid and incubation period, certain cells which contain the plasmid will thereby attain puromycin resistance from the encoded marker (Addgene, 2006). The oligonucleotides encoding the shRNAs are included in Table 2.3.  Table 2.3 Oligonucleotides encoding the shRNAs shRNA Insert MISSION® shRNA Sequence Control-sh CAACAAGATGAAGAGCACCAA Six1-sh1 CCGGAGCTTGTTTCTGGAGTTGTTTCTCGAGAAACAACTCCAGAAACAAGCTTTTTT Six1-sh2 CCGGCCAGACCAGAACTCGGTCCTTCTCGAGAAGGACCGAGTTCTGGTCTGGTTTTT Six1-sh3 CCGGCAAGAACGAGAGCGTACTCAACTCGAGTTGAGTACGCTCTCGTTCTTGTTTTT  Below is a picture detailing the pKLO.1 puro vector with shRNA construct utilized throughout this project to obtain knockdown of the Six1 protein (Figure 2.1).   41  Figure 2.1 Map of pLKO.1 plasmid containing a shRNA insert used for stable transfections. [Reproduced with permission from Addgene, 2006].   More detailed information can be found in Table 2.4 below [Addgene, 2006].           42 Table 2.4 Description of pLKO.1 plasmid cloning vector elements   [Reproduced with permission from Addgene, 2006]. A pictorial representation of the shRNA construct insert contained within pLKO.1 puro is shown in Figure 2.2 (Addgene, 2006).      Description Vector Element U6 Human U6 promoter drives RNA Polymerase III transcription for generation of shRNA transcripts. cPPT Central polypurine tract, cPPT, improves transduction efficiency by facilitating nuclear import of the vector's preintegration complex in the transduced cells. hPGK Human phosphoglycerate kinase promoter drives expression of puromycin. Puro R Puromycin resistance gene for selection of pLKO.1 plasmid in mammalian cells. sin 3'LTR 3' Self-inactivating long terminal repeat. f1 ori f1 bacterial origin of replication. Amp R Ampicillin resistance gene for selection of pLKO.1 plasmid in bacterial cells pUC ori pUC bacterial origin of replication. 5'LTR 5' long terminal repeat. RRE Rev response element for nuclear export of RNA.   43  Figure 2.2 Diagram and details of the shRNA insert The U6 promoter directs RNA Polymerase III transcription of the shRNA. The shRNA contains 21 "sense" bases that are identical to the target gene, a loop and 21 "antisense" bases that are complementary to the "sense" bases. The shRNA is followed by a polyT termination sequence for RNA Polymerase III [Reproduced with permission from Addgene, 2006]. 2.10.1. shRNA plasmid preparation and acquisition from glycerol stock and subsequent transfection and gene knockdown  The shRNA glycerol stock was diluted in Terrific Broth (TB) (without ampicillin) and put in the 37°C incubator for one hour (shaking). The shRNA glycerol stock dilution was plated on an LB plate containing ampicillin to select for the desired colonies. The plates were grown in 37°C incubator overnight (shaking). A single colony was selected from the plate and added to TB containing ampicillin (5-30mL). The bacterial culture was then shaken in the 37°C incubator overnight. The next day, a small amount of the expanded colony was plated onto an LB plate and kept in the fridge at 4°C. Plasmid DNA was extracted from the remaining shRNA bacteria in TB using the PureLinkTM HiPure Plasmid Miniprep kit (see manufacturer’s protocol, Invitrogen, K210002). The extracted DNA was prepared at the correct concentration with primers and sent away for Sanger sequencing at Genewiz. It was confirmed that the sequencing results matched the desired plasmid insert, indicating that the   44 correct plasmid was in fact present within the extracted DNA. 250-1000mL of TB containing ampicillin was inoculated by a colony from the leftover LB plate which contains the single colony confirmed by sequencing. The flask was then allowed to shake in 37°C incubator overnight. The following day, the PureLink® HiPure Plasmid Filter Maxiprep kit (see manufacturer’s protocol, Invitrogen, K2100006) was used to extract the plasmid DNA from the shRNA bacterial culture. Again, a portion of the extracted DNA was prepared and sent away for sequencing at Genewiz. It was again confirmed that the sequencing results matched the desired plasmid insert.  2.10.2. Ethanol precipitation Depending on the final concentration of DNA ready via OD microplate reader, it was in some cases necessary to concentrate the DNA using the ethanol precipitation method. First, sodium acetate pH 5.2 (final concentration of 0.3M) was added, at 1/10 the volume of the DNA sample and the solution was mixed. Next, 2 volumes of cold 100% ethanol was added (calculated after salt addition) and mixed well. The solution was then placed on ice for 20+ minutes, after which it was spun at maximum speed in a microcentrifuge for 15 minutes. The supernatant was carefully decanted and 1mL of 70% ethanol was added. The solution was then mixed, spun briefly, and the supernatant decanted carefully. The pellet was allowed to air dry and then resuspended in TE. 2.10.3. Generation of stably transfected cells shRNA plasmid DNA was transfected into each desired cell line using lipofectamine 2000. First, cells were plated in 6-well plates to 70-90% confluency, using 10% FBS DMEM. Metastatic melanoma cells were plated at a density of 2 × 105 cells/mL in 1 mL culture media in 6-well plates. Next, 15L lipofectamine 2000 was mixed with 150L optiMEM, and the   45 proper amount of plasmid DNA with optiMEM for the desired concentration. The lipofectamine and plasmid DNA solutions were mixed together and incubated for 5 minutes at room temperature. 300L of the lipofectamine-plasmid DNA solution was added to each well. Cell plates were incubated at 37°C for 72 hours. After 72 hours, a selection antibiotic (puromycin) was added to the medium to select for stably transfected cell populations at a concentration of 2 g/L in 10% FBS DMEM (A375) and 2.0 g/L in 15% FBS DMEM (RPMI-7951). The cells were transfected either by control (shCTRL) or gene-specific shRNA plasmids. Stable knockdown was confirmed by Western blotting 3-6 weeks after puromycin selection began (Table 2.5). Table 2.5 Timecourse of post-transfection assays Post-transfection assay Incubation time post-transfection Incubation time with puromycin selection Protein knockdown (Western) 4+ days 3-6+ weeks Phenotypic assay 4+ days 6-8+ weeks  2.11. Cell growth assay  CellTiter Blue® Viability Assay kit (Promega) was used to assay for cell growth. This assay uses the indicator dye resazurin to measure the metabolic capacity of cells. Viable cells are able to reduce resazurin into resorufin, while non-viable cells are unable to convert resazurin to resorufin. The compounds are distinguished based on absorbance wavelength, as resazurin has an absorbance wavelength of 600nm and resorufin has an absorbance wavelength of 570nm (Promega). First, 700 and 1000 cells (A375 and RPMI-7951,   46 respectively) were seeded in 96 well-plates with 100L of 10% FBS DMEM medium. Six replicate wells were used for each condition (Medium, Control, sh1, sh2, sh3). Four plates were prepared for the four time points (0h, 24h, 48h, 72h). The cells were allowed to sit for four hours in order to settle and adhere to the plate. CellTiter Blue® reagent was taken out of the -20°C freezer and allowed to thaw in 37°C water bath while being protected from light. Next, 20uL of the CellTiter Blue® reagent was pipetted into each well. The plate was then shaken manually for 10 seconds before being placed into the 37°C incubator. Once the plate had incubated for four hours, the plate was again shaken manually for 10 seconds and then the absorbance was read on a microplate reader at a wavelength of 570nm. This process was then repeated for each time point plate (0h, 24h, 48h, 72h). Each condition included six replicates. The average absorbance 600nm values of the culture medium background was subtracted from all absorbance 570nm values of experimental wells. 2.12. Cell proliferation assay In order to evaluate and calculate the number of cells actively synthesizing DNA, a 5-bromo-2'-deoxyuridine (BrdU) Incorporation Assay (6813, Cell Signalling) was performed. This assay uses an anti-BrdU antibody to detect the amount of BrdU which has been incorporated into cells, which is a measure of cellular proliferation. BrdU is a pyrimidine analog which is incorporated into newly synthesized DNA during the cell division. First, 1000 cells per well were plated in 96-well plates, triplicate per condition. Four plates were prepared for four time point assessments (0h, 24h, 48h, 72h). Cells were allowed to adhere to the plate by incubating them for 3-4 hours in a 37°C 5% CO2 incubator. Prepared 10X BrdU solution was then added to each well, for a final concentration of 1X. Next, cells were placed in a 37°C 5% CO2 incubator and incubated for 24 hours. Following incubation with BrdU,   47 the medium was removed and 100L of the Fixing/Denaturing Solution was added to each well. The plate was then kept at room temperature for 30 mins to allow for fixation of cells to occur. The solution was removed after 30 mins. 100L/well prepared 1X detection antibody solution was then added to each well, and the plate kept at room temperature for 1 hour. The solution was removed after one hour and each well washed three times with 1X Wash Buffer. Next, 100L prepared 1X HRP-conjugated secondary antibody solution was added to each well, and the plate kept at room temperature for 30 mins. The solution was removed after 30 mins and the plate was then washed 3 times with 1X Wash Buffer. Finally, 100L TMB Substrate was added to each well and allowed to incubate for 30 mins at room temperature. To complete the experiment, 100L of STOP Solution was added to each well containing TMB Substrate. The absorbance was read at 450nm on a microplate reader to determine BrdU incorporation. The entire process was then repeated for the remaining time points (24h, 48h, 72h). Each condition included six replicates for a single assay.  2.13. Cell migration assay A scratch assay was performed to measure cell migration. Cells were seeded in a 24-well plate and were left to grow overnight until fully confluent (control, sh1, sh2, sh3), in 10% FBS DMEM medium. The cells were monitored so that the cells did not overgrow and/or begin layering. The medium was removed the following day, and a p200 tip was used to create four scratches in a # pattern in each well. Following the scratch, the wells were washed with serum-free medium to remove any detached cells and debris. The wells were then filled with 0.1% FBS DMEM medium containing 2 g/mL of puromycin. This low serum medium was used to limit the proliferation of the cells to distinguish between proliferative and migratory closure of the monitored scratch (Liang et al., 2007). Once the   48 initial protocol was complete, the cells were placed at 37°C in a humidified cell culture incubator with 5% CO2 for a period of 48h. Pictures of the scratches were taken at exact coordinates at 0h, 24h, and 48h (Carl Zeiss MicroImaging ©, 2011). Two photos were taken for each duplicate. These pictures were then analyzed using the freely available TScratch program ( This software was used to quantify the open area of the initial scratch and subsequent scratch area at later time points, providing an immediate readout (Gebäck and Schulz et al., 2007).  2.14. Cell invasion assay  The Boyden Chamber Assay (BioCoat™ Matrigel® Invasion Chambers, #354480, Corning®) was used to mimic cell invasion in vitro. 2.14.1. Corning BioCoat matrigel invasion chamber The Matrigel Invasion Chambers (BioCoat Matrigel Invasion Chambers, #354480, Corning) were rehydrated by removing them from -20°C and adding warm DMEM to the bottom of the wells. They were allowed to rehydrate for 2 hours in a humidified tissue culture incubator, 37°C, 5% CO2 atmosphere. After rehydration, the medium was removed without disturbing the layer of Corning Matrigel Matrix on the membrane. Two control inserts and two matrigel inserts were used per condition (12 total). Following this, 15% FBS DMEM was added to the wells of the 24-well plate to serve as a chemoattractant. Sterile forceps were then used to transfer the chamber and control inserts to the wells containing the chemoattractant. It was ensured that no air bubbles were trapped beneath the membranes. Next, the cells were prepared by trypsinizing, centrifuging down, and resuspending with 0.1% FBS DMEM. 2.5x104 cells/mL were seeded in the upper compartment of each   49 chamber. The Corning BioCoat Matrigel Invasion Chambers were incubated for 22 hours in a humidified tissue culture incubator, at 37°C, 5% CO2 atmosphere.  shCTRL, sh1, sh2 and sh3 RPMI-7951 cells were seeded into both control inserts and matrigel inserts and incubated at 37°C for 24h. The same protocol was repeated from A375 shCTRL, sh1 and sh2 cell lines. Fixing and staining procedures were then performed, as described below.  2.14.2. Measurement of cell invasion Non-invading cells were removed by inserting a cotton tipped swab into the Corning BioCoat Matrigel insert and applying gentle but firm pressure while moving the tip over the membrane surface (“scrubbing”). Scrubbing was repeated with a second swab moistened with DMEM medium. Next, 0.5 mL 10% neutral buffered formalin solution was added to each chamber. It was allowed to fix the cells for 10 minutes. The fixed cells were then stained for 10 minutes with 0.5 mL 0.05% crystal violet solution in 2% ethanol (filter before use). The formalin solution and crystal violet solution in was discarded appropriately.  Next, distilled water was added to two beakers and the inserts were sequentially transferred through the two beakers of water to wash the stain off. The inserts were allowed to air dry at room temperature overnight. Stained cells were counted by taking five photos from the upper, middle, lower, left, and right sides for each insert, and cell numbers were counted using ImageJ software to calculate the percentage of cell invasion of Six1-sh cells relative to shCTRL cells. First, the average cell number of each insert was utilized to calculate % Invasion = Mean number of cells invading through Matrigel insert membrane / Mean number of cell migrating though control insert membrane ×100%. Next, the Invasion Index was calculated using: Invasion Index = % Invasion Six1-sh Cell / % Invasion shCTRL   50 Cell. The Invasion Index was the final analysis and number used in the results. Each condition was performed in biological duplicate, and repeated at least once. 2.15. Statistical analysis Statistical analyses were performed using SPSS 16 (Chicago, IL), X-tile (New Haven, CT), GraphPad Prism 5.00 (San Diego, CA), and Microsoft Excel programs. P values < 0.05 were considered to be statistically significant. Two-tailed t tests were used to compare continuous variables. Data are shown as mean ± standard deviation (StDev). The correlation between Six1 expression and five-year disease-specific mortality was evaluated using the Kaplan-Meier survival curve and log-rank test. Using X-tile software (Camp et al, 2004), the optimal cut-off points (with the lowest P value) of Six1 protein expression and/or intensity levels were determined. All statistical analyses were carried out using the SPSS version 16.0 software (SPSS Inc., Chicago, IL, USA).              51 3 Results 3.1. Determination of SIX1 mRNA expression in established cell lines and clinical samples  The Molecular Medicine Lab (MML) microarray analysis database revealed that SIX1 was upregulated in metastatic melanoma. To determine the relative expression of SIX1 mRNA in melanoma cell lines and clinical samples, qPCR was performed on a number of samples (melanocytes, metastatic melanoma cells, normal skin, normal nevi, and metastatic melanoma patient biopsies). It was verified that this overexpression held true in metastatic melanoma cell lines and malignant melanoma clinical samples (Figure 3.1 A, B). As seen in Figure 3.1 A, melanoma cell lines (maroon) contained a significantly higher level of SIX1 mRNA than melanocyte cell lines (pink) (P=0.0087). The same trend was seen in clinical samples: metastatic melanoma patient samples expressed significantly higher levels of mRNA when compared to normal nevi (Figure 3.1 B) (P<0.01). However, comparing normal skin and metastatic melanoma, the results approached but did not reach statistical significance (P>0.05; P<0.1). This trend may reach statistical significance with a larger sample size.          52 A      B  Figure 3.1 SIX1 mRNA levels are increased in metastatic melanoma cell lines and clinical tissues. (A) Comparison of SIX1 mRNA levels in cell lines; HEMC1, 2 and 3, MC0-24 and MC are melanocyte cell lines; A375, G361, SH4, RPMI-7951, WM-115, SK-MEL-1, SK-MEL-3, and SK-MEL-24 are metastatic melanoma cell lines (*, P=0.0087). (B) Comparison of SIX1 mRNA levels in patient biopsies including normal skin (NS, P=0.0923; n=4), normal nevi (NN, *, P=0.0083; n=6), and metastatic melanoma (MM; n=9) samples. Error bars were plotted using standard deviation (StDev). Statistical significance was assessed using an independent two-tailed t test. P<0.05 were considered significant. 3.2. Six1 protein levels are increased in metastatic melanoma cell lines Following confirmation of the upregulation of SIX1 mRNA levels in melanoma cell lines and clinical samples, Western blotting was performed on a select number of cell lines. (melanocytes (MC) and metastatic melanoma cells (RPMI-7951, A375) (Figure 3.2). The Western blot verified that the overexpression of mRNA levels held true when assessing Six1 protein levels in metastatic melanoma cell lines. It was found that the cell lines RPMI-7951 and A375 contained a higher quantity of Six1 protein than MC. MC did not contain any visible Six1 protein. These findings indicated that RPMI-7951 and A375 metastatic melanoma cell lines contained high Six1 protein levels, making them suitable to use as in vitro models to study the biological function of Six1 in metastatic melanoma.   53  Figure 3.2 Six1 protein levels are increased in metastatic melanoma cell lines. Western blotting showed marked Six1 protein overexpression in metastatic melanoma cell lines (RPMI-7951, A375) compared to immortalized melanocytes (MC). Beta-actin was utilized as an internal control for protein loading. Molecular weight: Six1 = 32 kDa; Beta-actin = 42 kDa. 3.3. Western blot to assess Six1 shRNA stable knockdown in A375 and RPMI-7951  One of the first steps before in vitro functional assays could be performed was achieving a reduced amount of the Six1 protein in metastatic melanoma cells. The quantification of Six1 protein in metastatic melanoma cell lines were determined via Western blotting following Six1 plasmid shRNA stable knockdown. The quantity of Six1 protein was compared between control (shCTRL) and Six1 shRNA knockdown (sh1, sh2) cell lines, using beta-actin as an internal control for protein loading. The Image J program was used to quantify protein levels. It was found that A375 sh1 and sh2 cell lines resulted in a 57% and 43% knockdown of Six1 protein, respectively (Figure 3.3 A). RPMI-7951 sh1 and sh2 cell lines achieved a Six1 protein knockdown of 53% and 61%, respectively (Figure 3.3 B). Only once this was achieved could we move on to in vitro functional assays to assess the biological effects of this knockdown. Furthermore, knockdown of the Six1 protein band validated the antibody’s specificity for Six1.       54 A      B  Figure 3.3 Six1 knockdown was achieved via shRNA plasmid transfection in metastatic melanoma cell lines. Six1 knockdown by two shRNAs (sh1, sh2, both specifically target SIX1 gene) compared to control (shCTRL, non-targeting shRNA) in metastatic melanoma cell lines A375 (A) and RPMI-7951 (B). Protein lysates were probed with antibodies against Six1 after transfected cells were selected by puromycin (2μg/mL) for 5 days. Beta-actin was included as a protein loading control. Molecular weight: Six1 = 32 kDa; Beta-actin = 42 kDa 3.4. In vitro functional studies following stable knockdown of Six1 Several functional studies manipulating melanoma cells were performed to document the effects of stable Six1 knockdown in vitro. Cell growth, proliferation and apoptosis assays were utilized to observe if Six1 knockdown had any impact on the growth and survival of metastatic melanoma cells. Migration and invasion assays were used to decipher if decreased levels of Six1 affected metastatic dissemination of melanoma cells. 3.4.1. Cell growth and Six1  We first wanted to investigate if Six1 knockdown affected cell growth. A cell growth assay was performed over a period of three days on control cells (shCTRL) and Six1 shRNA knockdown cells (sh1, sh2) to measure cell growth. In both A375 and RPMI-7951, knockdown cells showed a significant decrease in metabolic ability over the period assessed in comparison to control cells (Figure 3.4 A, B). Significant differences (P < 0.05) were   55 observed all three days with greater significance as the end point approached. While knockdown cells showed little difference in metabolic ability over three days, control cells showed a steady increase in cell metabolic activity. Similar trends were observed in both cell lines. More dramatic growth was observed in RPMI-7951, likely due to more cells being seeded than A375 (1000 vs 700, respectively).  A         B  Figure 3.4 Six1 inhibition reduces cell growth of metastatic melanoma cells in vitro. Cell growth analysis of Six1 shRNA knockdown cells (sh1, sh2) in A375 (A) and RPMI-7951 (B) resulted in a marked decrease in cell growth over a period of three days in comparison to control cells (shCTRL). The CellTiter Blue® Viability Assay was performed and the plate read from day 0 to day 3. Fold change was calculated by OD570 absorbance at each time point divided by OD570 value at 0 h to show the progression of cell growth over time. Significance was determined using an independent two-tailed t test. P<0.05 was considered significant. *, P < 0.05, **, P < 0.01, ***, P < 0.001.  The experiment was performed using six replicates per condition. Error bars indicate StDev between replicates. 3.4.2. Cell proliferation and Six1 We then wanted to determine whether the decreased growth of the Six1 knockdown cells was due in part to a difference in cell proliferation ability, and so a BrdU assay was performed. In line with the cell growth assay results, A375 and RPMI-7951 knockdown cells displayed a significantly lower proliferative rate than control cells (Figure 3.5 A, B). In   56 A375, significant differences (P < 0.05) were observed in both sh1 and sh2 all three days. In RPMI-7951, significance was observed all three days for sh1, but only day 2 and 3 for sh2. The results show that the A375 cell line has a naturally higher rate of proliferation than RPMI-7951, since the same number of cells were seeded for this experiment. A      B      Figure 3.5 Proliferation declines in metastatic melanoma cells with reduced Six1 protein. Proliferation in Six1 shRNA knockdown cells (sh1, sh2) in both A375 (A) and RPMI-7951 (B) was significantly diminished. Proliferation of cells was assessed via BrdU cell proliferation assay over a period of three days. Cells were incubated with a BrdU antibody and read on a microplate reader at 450nm to assess the amount of BrdU incorporation. Fold change was calculated by OD450 absorbance at each time point divided by OD450 value at 0 h to show the change in cell proliferation over time. Significance was determined using an independent two-tailed t test. P<0.05 was considered significant. *  P < 0.05, ** P < 0.01, *** P < 0.001.  The experiment was performed using six replicates per condition. Error bars indicate StDev between replicates. 3.4.3. Apoptosis of melanoma cells and Six1 We then used Western blot to determine the quantity of key apoptotic proteins in metastatic melanoma cell line A375. Five proteins were examined by incubating a PVDF membrane containing protein with antibodies for: Bcl-2, Bax, pro-caspase 3 and pro-caspase 9 (intracellular apoptotic pathway), and pro-caspase 8 (extracellular apoptotic pathway). It   57 was found that protein levels of bcl-2, pro-caspase 3 and pro-caspase 9 were reduced in knockdown cells (Figure 3.6 A, B). The opposite trend, an increase, was found for Bax. No difference was observed in pro-caspase 8. A  B  Figure 3.6 Six1 knockdown has minor or no effect on apoptosis proteins in metastatic melanoma cell line A375. (A) Bar graph depicting apoptosis protein level as measured by Image J from (B) Western blots of A375 cells comparing apoptosis protein levels in control (shCTRL) and Six1 shRNA (sh1, sh2) knockdown cells. shCTRL, sh1 and sh2 were assessed using specific antibodies against Bax, Bcl-2, pro-caspase 3, pro-caspase 8 and pro-caspase 9. Beta-actin was used as a loading control.   58 3.4.4. Migration of melanoma cells and Six1 We next wanted to determine what happens to the migratory ability of metastatic melanoma cells which are Six1 deficient by performing a scratch assay. As depicted in the Figure 3.7, both A375 (A, C) and RPMI-7951 (B, D) Six1 knockdown cells showed a significant decrease in migration (P < 0.01). In A375, when compared to control cells, knockdown cell wounds showed a decreased in migratory ability of approximately 68% (sh1) and 39% (sh2). In RPMI-7951, the knockdown cells had a reduced migratory ability of approximately 38% (sh1) and 54% (sh2) compared to control cells.                   59 A              B  C               D Figure 3.7 Six1 knockdown restricts metastatic melanoma cell migration. Migration of A375 (A, C) and RPMI-7951 (B, D) was studied using a scratch assay. The wound size was monitored by taking microscopic photos at specific grid locations during a timecourse of 0h and 72h for A375 (C) and 0h and 24h for RPMI-7951 (D) (until the wound closed). The dotted lines were used to provide contrast and clarify the areas lacking cells at the end of the experiment. The photos were analyzed using Image J to determine the percentage of wound closure and the data represented in bar graphs (A, B). **, P < 0.01. Experiments were performed in 24-well plates with biological duplicate wells per condition. Error bars denote StDev. Single photos are shown which are representative of the experiments. Bars represent 50m.   60 3.4.5. Invasion and Six1 in melanoma cells  Cell invasion was examined using the Boyden Chamber technique. This is a common method used to test the effects of a treatment or alteration to an adherent cancer cell line’s ability to invade, as the Matrigel Matrix imitates the basement membrane, blocking non-invasive cells from migrating through it. In this way, invasive cells can be differentiated between non-invasive cells. An Invasion Index value is calculated by comparing normal migratory ability through Control membrane inserts and capacity to invade through the Corning Matrigel Matrix and membrane.  In this way, the effects of Six1 knockdown on the melanoma cell’s ability to invade could be assessed when comparing the results for the control cells (shCTRL) and Six1 shRNA knockdown cells (sh1, sh2). In both cell lines, a significant reduction in invasion was observed with both shRNAs (Figure 3.8) (**P<0.01). For A375 cells, Six1 knockdown cells exhibited an Invasion Index of approximately 6.9% and 8.5% invasion, respectively (Figure 3.8 A, C). Similarly, RPMI-7951 Six1 knockdown cells displayed an average Invasion Index of approximately 14.2% and 5.1%, respectively (Figure 3.8 B, D).             61 A              B C                  D  Figure 3.8 Invasive ability of malignant melanoma cells is inhibited by Six1 knockdown. A Boyden chamber assay was used to test the differences in invasion ability between control cells (shCTRL) and Six1 knockdown cells (sh1, sh2) in metastatic melanoma cell lines A375 (A, C) and RPMI-7951 (B, D). (A) and (B) are bar graphs depicting the Invasion Index of each cell line. **, P < 0.01. (C) and (D) are photos of wells which were chosen to be representative of each cell line. Analysis (counting of cells) of control migration inserts and Matrigel invasion inserts was performed using Image J software. Each experiment consisted of biological duplicate wells. Error bars denote StDev. Bars represent 50m (C) and 100m (D).   62 3.5. The clinical relevance of Six1 in melanoma We then wanted to study the expression of Six1 protein in patient biopsies using a tissue microarray (TMA) consisting of 438 samples. The TMA included 19 normal nevi (NN), 32 dysplastic nevi (DN), 23 melanoma in situ (MIS), 219 primary melanoma (PM) and 145 metastatic melanoma (MM) patient samples. Immunohistochemical staining using a specific Six1 antibody was performed on the TMA slides containing formalin-fixed, paraffin-embedded human tissues. The TMA was scored for the area and intensity of Six1 protein in a double-blind setting by a registered pathologist. Separate scores were recorded for Six1 in the nucleus and cytoplasm. Six1 protein staining was brown in colour and was contrasted by subsequent hematoxylin staining (faded blue, nuclear) within the nucleus (Figure 3.9 A, B). A           B Figure 3.9 Six1 protein staining in tissue microarray. Representative images of Six1 immunohistochemical staining in human tissue biopsy TMA. (A) Nuclear Six1 protein staining in normal nevi; (B) Cytoplasmic Six1 protein staining in metastatic melanoma. Scale bars represent 50m.   63 3.5.1. Six1 and melanoma prognosis Following scoring of the TMA, we wanted to determine if Six1 protein expression correlated with melanoma patient survival. A score of zero to three was considered “low” and a score of four to 12 was considered “high” (cutoff of four). A Kaplan-Meier log-rank test was performed to determine if Six1 nuclear and cytoplasmic expression correlated with five-year disease-specific survival. The TMA pathological scoring revealed that high nuclear Six1 correlated with increased survival, whereas low nuclear Six1 correlated with decreased survival. On the contrary, high cytoplasmic Six1 correlated with decreased survival, whereas low cytoplasmic Six1 correlated with increased survival. These results were found to be significant (P = 0.001 nuclear, P = 0.019 cytoplasmic; log-rank test) (Figure 3.10). A              B  Figure 3.10 Kaplan-Meier survival analyses of melanoma patients regarding Six1 expression levels. (A) Patients with low nuclear Six1 expression have a significantly worse disease-specific five-year survival than those with high nuclear Six1 expression (P=0.001, log-rank test); (B) Patients with high cytoplasmic Six1 expression have a significantly worse disease-specific five-year survival than those with low cytoplasmic Six1 expression (P=0.019, log-rank test).  3.5.2. Six1’s association with lesion type We were then interested in whether Six1 protein expression was associated with lesion type. The types of lesions assessed, in order of melanoma progression were: normal   64 nevi (NN), dysplastic nevi (DN), melanoma in situ (MIS), primary melanoma (PM) and metastatic melanoma (MM). The same cutoff score was used as in the Kaplan-Meier log-rank test for low vs high protein expression. A 2 test was used to assess significance between Six1 low and Six1 high protein in both the nucleus and cytoplasm.  The data showed that nuclear Six1 expression decreased as melanoma progressed from normal nevi to metastatic melanoma (Figure 3.11 A). The opposite trend was observed with cytoplasmic Six1 expression; melanoma progression was associated with increased levels of cytoplasmic Six1 (Figure 3.11 B). These results were found to be significant (P<0.0001 for both, 2 test).  A            B  Figure 3.11 Six1 protein expression shifts from the nucleus to the cytoplasm during melanoma progression. (A) Reduced nuclear expression of Six1 protein correlates with melanoma progression; (B) Increased cytoplasmic expression of Six1 protein correlates with melanoma progression. A significant difference in Six1 expression was observed between lesion types in both the nucleus and cytoplasm using the 2 test (P<0.0001). NN = normal nevi, DN = dysplastic nevi, MIS = melanoma in situ, PM = primary melanoma, MM = metastatic melanoma. 3.5.3. Six1’s association with clinicopathological parameters  We were then interested to see if recorded clinicopathological characteristics in patient biopsies correlated with Six1 protein expression. The same cutoff score was used as in the Kaplan-Meier log-rank test for low vs high protein expression. A 2 test was used to   65 assess the correlation between Six1 expression and clinicopathological parameters. Both nuclear Six1 and cytoplasmic Six1 were evaluated. The clinicopathological parameters assessed were AJCC stage, age, sex, tumor thickness, mitosis, tumor-infiltrating lymphocytes, ulceration, regression, histological satellitosis, and melanoma subtype.   In our data set, nuclear Six1 was found to be significantly associated with AJCC staging, ulceration, histological satellitosis and melanoma subtype (P<0.0001, P = 0.0395, P = 0.0309, P = 0.0305, respectively, 2 test) (Table 3.1). At AJCC stage 0, 74% of patients presented with high nuclear Six1 staining, and this nuclear staining percentage decreased as the stages advanced. By AJCC stage III and IV, high nuclear staining was observed in only 20% and 24% of patients. As for ulceration, lower nuclear Six1 was associated with the presence of ulceration. Only 39% of patients with ulceration displayed high nuclear Six1, compared to 56% of patients without ulceration. Furthermore, histological satellitosis was associated with lower levels of nuclear Six1. Only 13% of patients with histological satellitosis were found to have high nuclear Six1, compared to 53% of patients without histological satellitosis. As for melanoma subtypes, nuclear Six1 was found to be more prominent in superficial spreading melanoma and lentigo maligna melanoma at 56% and 59%, respectively, compared to 34% in nodular melanoma. Cytoplasmic Six1 was found to be significantly correlated with AJCC staging, tumor thickness, and melanoma subtype (P<0.0001, P<0.0001, P=0.0062, respectively) (Table 3.2). A significant increase in cytoplasmic Six1 was observed as AJCC stages progressed from 0 to IV, the opposite of the nuclear Six1 trend. Only 48% of patients had strong Six1 staining at stage 0, compared to 89% at stage III and 76% at stage IV. For tumor thickness, a thicker tumour corresponded with an increased presence of cytoplasmic Six1, with T1 at 59%, T2 at   66 67%, T3 at 86% and T4 at 90%. Finally, 93% of nodular melanomas had high cytoplasmic Six1 staining, which is significantly higher than superficial spreading melanoma and lentigo maligna melanoma at 71% and 67%, respectively.     Table 3.1 Nuclear Six1 expression and clinicopathological characteristics Histopathologic prognosticator Total Six1 Nuclear Staining Significance (P value of χ2 test)    Six1 Low Six1 High   Lesion Type      NN 19 4(21%) 15(79%) *<0.0001 DN 32 9(28%) 23(72%) MIS 23 6(26%) 17(74%) PM 219 106(48%) 113(52%) MM 145 114(79%) 31(21%) AJCC      0 23 6(26%) 17(74%) *<0.0001 I 99 37(37%) 62(63%) II 120 69(58%) 51(43%) III 99 79(80%) 20(20%) IV 46 35(76%) 11(24%) Age      ≤ 60 202 113(56%) 89(44%) 0.3566 > 60 185 113(61%) 72(39%) Sex       Male 227 140(62%) 87(38%) 0.1463  Female 160 86(54%) 74(46%)      67 Histopathologic prognosticator Total Six1 Cytoplasmic Staining Significance (P value of  χ2 test)    Six1 Low Six1 High    Tumour Thickness (mm)       ≤1.00 (T1) 66 24(36%) 42(64%) 0.2242  1.01-2.00 (T2) 64 22(34%) 42(66%) 2.01-4.0 (T3) 51 26(51%) 25(49%)  >4.0 (T4) 61 21(34%) 40(66%) Mitosis       Present 132 70(53%) 62(47%) 0.1522  Not present 255 156(61%) 99(39%) Tumor-infiltrating lymphocytes      Not present 86 40(47%) 46(53%) 0.8694 Non-brisk 112 55(49%) 57(51%) Brisk 21 11(52%) 10(48%) Ulceration       Present 61 37(61%) 24(39%) *0.0395  Not present 159 70(44%) 89(56%) Regression       Present 14 3(21%) 11(79%) 0.0519𝛼  Not present 206 103(50%) 103(50%) Histological Satellitosis        Present 8 7(88%) 1(13%) *0.0309𝛼   Not present 211 99(47%) 112(53%) Subtype      Superficial Spreading 89 39(44%) 50(56%) *0.0305 Nodular 44 29(66%) 15(34%) Lentigo Maligna 39 16(41%) 23(59%)            *P<0.05 was considered to be statistically significant by χ2 test  or 𝛼Fisher's exact test  Total n = 438; Melanoma patients n = 387;      n < 387 is due to data unavailability       68  Table 3.2 Cytoplasmic Six1 expression and clinicopathological characteristics Histopathologic prognosticator Total Six1 Cytoplasmic Staining Significance (P value of χ2 test)    Six1 Low Six1 High   Lesion Type      NN 19 17(89%) 2(11%) *<.0001 DN 32 15(47%) 17(53%) MIS 23 12(52%) 11(48%) PM 219 49(22%) 170(78%) MM 145 22(15%) 123(85%) AJCC      0 23 12(52%) 11(48%) *<.0001 I 99 33(33%) 66(67%) II 120 16(13%) 104(87%) III 99 11(11%) 88(89%) IV 46 11(24%) 35(76%) Age      ≤ 60 202 47(23%) 155(77%) 0.431 > 60 185 36(19%) 149(81%) Sex       Male 227 44(19%) 183(81%) 0.2921  Female 160 39(24%) 121(76%) Tumour Thickness (mm)       ≤1.00 (T1) 66 27(41%) 39(59%) *<.0001  1.01-2.00 (T2) 64 21(33%) 43(67%) 2.01-4.0 (T3) 51 7(14%) 44(86%)  >4.0 (T4) 61 6(10%) 55(90%) Mitosis       Present 132 33(25%) 99(75%) 0.2733  Not present 255 50(20%) 205(80%) Tumor-infiltrating lymphocytes      Not present 86 15(17%) 71(83%) 0.2739 Non-brisk 112 30(27%) 82(73%) Brisk 21 4(19%) 17(81%) Ulceration       Present 61 11(18%) 50(82%) 0.4386  Not present 158 38(24%) 120(76%)   69 Histopathologic prognosticator Total Six1 Cytoplasmic Staining Significance (P value of χ2 test)    Six1 Low Six1 High   Regression       Present 13 3(23%) 10(77%) 1 𝛼  Not present 206 46(22%) 160(78%) Histological Satellitosis        Present 8 1(13%) 7(88%) 0.68727185 𝛼  Not present 211 48(23%) 163(77%) Subtype      Superficial Spreading 89 26(29%) 63(71%) *0.0062 Nodular 44 3(7%) 41(93%) Lentigo Maligna 39 13(33%) 26(67%)            *P<0.05 was considered to be statistically significant by χ2 test  or 𝛼Fisher's exact test  Total n = 438; Melanoma patients n = 387;      n < 387 is due to data unavailability                   70 4 Discussion The incidence of melanoma is on the rise; however, efficacious treatments for melanoma are lacking. This is largely due to the genetic heterogeneity of melanoma; it is a genetically complicated disease with a high number of mutations and abnormal gene and protein expression. This study chose to focus on SIX1, which is a homeobox gene known to play a role in embryogenesis. SIX1 has also been found to be aberrantly expressed and to promote tumorigenesis in a variety of cancers such as breast, pancreatic and colorectal. However, the expression and biological functions of Six1 have never been studied in melanoma. The first evidence of abnormal SIX1 enrichment in melanoma originated from the Molecular Medicine Lab’s previous transcriptome analysis. Abnormalities in gene and protein expression need to be investigated so that we can improve our inadequate understanding of the molecular mechanisms contributing to the metastasis and ultimately mortality of melanoma. Characterization of Six1 expression and biological function in melanoma will provide further insight into the insidious nature of this disease and perhaps facilitate the development of novel therapies and biomarkers.  In the present study, we utilized clinical samples, in vitro models and a tissue microarray of patient biopsies to investigate the expression, pathogenic function and prognostic potential of Six1 in melanoma. Through a combination of qPCR, Western blot, cell assays and immunohistochemistry, we identified Six1 as an aberrantly expressed gene and protein in metastatic melanoma. Our data provided evidence that this abnormal expression of Six1 may be involved in the pathogenic biological functions of malignant melanoma. In addition, a profound nuclear to cytoplasmic shift of Six1 accompanied melanoma progression and correlated with poor five-year survival.   71 The TMA also revealed that high cytoplasmic Six1 was associated with advanced AJCC stages, increased tumor thickness, and melanoma subtype, whereas low nuclear Six1 correlated with AJCC stages, ulceration, histological satellitosis, and melanoma subtype Our data confirmed enhanced Six1 expression in metastatic melanoma, as Six1 upregulation was a common feature shared by the majority of metastatic melanoma cell lines. Moreover, high SIX1 mRNA levels effectively differentiate metastatic melanoma from non-malignant controls (i.e. normal melanocytes, normal skin and normal nevi). Further, our protein quantification results via Western blot clearly show that Six1 protein is absent in normal melanocytes and present in two metastatic melanoma cell lines. Six1 is an embryonic gene which is not normally expressed in adult tissues. However, there are a growing number of studies reporting the upregulation of Six1 in a variety of cancers. Our findings are in line with these studies, suggesting that Six1 is being expressed out-of-context in metastatic melanoma.  Several mechanisms could contribute to SIX1 upregulation in melanoma, such as genetic instability, alterations in mRNA stability, promoter activity and epigenetic factors such as DNA methylation and histone acetylation (Reichenberger et al., 2005). The cause of SIX1 upregulation in melanoma remains undefined, however, a few mechanisms have been investigated in other cancers. Genetic instability, or the ability to maintain an abnormal number of chromosomes, is common in cancer and can cause gene amplification (Balmain, 2001). In Hodgkin’s lymphoma, a fluorescence in situ hybridization (FISH) analysis revealed that upregulation of SIX1 was likely due to copy number gains at chromosome 14q23 (Nagel et al., 2015). Gene amplification and overrepresentation was also found to be one of the causes of Six1 overexpression in breast cancer (Reichenberger et al., 2005).   72 Additionally, epigenetics has been found to play a role in many cancers, since epigenetic alterations can have a significant impact on the transcription of genes (Sharma et al., 2010). The loss of inhibitory epigenetic mechanisms, like DNA methylation, is another possible explanation for upregulated SIX1 transcription. Although upstream regulators of Six1 have yet to be identified in cancer, a study on postnatal cardiac homeostasis revealed that the Polycomb histone methyltransferase Ezh2 is responsible for the repression of Six1 in differentiating cardiac progenitors (Delgado-Olguin et al., 2012). The study found that epigenetic regulation of Six1 is required in postnatal cardiac cells to stabilize postnatal cardiac gene expression and prevent pathologic cardiac remodeling such as hypertrophy (Delgado-Olguin et al., 2012). Although this has only been demonstrated in cardiac cells, this raises the possibility that Six1 may be under epigenetic control postnatally, and aberrant expression of Six1 in melanoma could be caused by an alteration in this epigenetic control.  Furthermore, previous reports have indicated that the Six1 homeoprotein is degraded by ubiquitin-mediated proteolysis by the anaphase-promoting complex (APC, Cdh1) (Christensen et al., 2007). It is therefore possible that Six1 has been mutated so that it does not undergo degradation by the APC or the APC may be mutated itself, thereby preventing Six1 degradation. This would account for the abnormal increase in Six1 protein in melanoma, and so is another possibility. Six1 overexpression in metastatic melanoma could be contributed to by several factors simultaneously, and until more studies are done concerning the mechanism of this upregulation, we cannot draw any conclusions.  Cancers are notorious for genetically altering molecular pathways within cells, sometimes providing them with survival advantages and aiding in the acquisition and execution of metastatic phenotypes. SIX1 is not normally expressed in adult tissues, and so   73 Six1’s reemergence in cancers is suspicious. SIX1 is a homeobox gene that has been found to be important for the development of body structures during embryogenesis, indicating that SIX1 is involved in and has control over molecular mechanisms involving growth and formation. Indeed, numerous homeobox genes have been found to be reactivated in numerous cancers, such as Beta Protein 1 (BP1) in breast cancer and Gastrulation Brain Homeobox 2 (GBX2) in prostate cancer (Abate-Shen, 2002; Fu et al., 2016; Gao et al., 1998). It was this upregulation of Six1 that drove us to question whether or not Six1 played a role in the pathogenic phenotypes of metastatic melanoma. Using two metastatic melanoma cell lines, we found that SIX1 suppression by plasmid-mediated shRNA gene silencing markedly reduced cancerous phenotypes. The aggressiveness of melanoma is attributed to the ability of melanoma cells to proliferate without normal growth restrictions and metastasize to distant parts of the body via epithelial to mesenchymal transition. We first wanted to assess the relationship between Six1 and cell growth and survival. There are two main methods in which cancer cells achieve a growth and survival advantage: by increasing proliferation and decreasing apoptosis. By knocking down Six1, we could observe if and in what way these cellular processes were affected by a reduction in Six1. Our first experiment was a cell growth assay which allowed us to observe the effect of reduced Six1 protein level on melanoma cell growth. Control cells converted redox dye resazurin into end product resorufin to a greater extent than Six1 knockdown cells. In other words, a diminished level of Six1 positively correlated with a reduction in cell growth in two metastatic melanoma cell lines. Six1 knockdown was the only variable that was altered between conditions, and so we reasoned that Six1 knockdown may have caused the   74 difference in cell growth. This data suggests that the upregulation of Six1 may be associated with increased cell growth in melanoma. It is known that cancer cells have an increased growth and metabolism in comparison to normal cells (Scott et al., 2011). One study on melanoma cell metabolism found that normal melanocytes used less glucose or glutamine than melanoma cells, indicating that melanoma cells have a more active metabolism than normal melanocytes (Scott et al., 2011). Next, we decided to test if this difference in cell growth was related to proliferative ability and/or induction of apoptosis. First, a decrease in Six1 protein was accompanied by a decrease in cellular proliferation as measured by a BrdU assay. This decline in proliferation was consistent with the decline in cell growth observed in Six1 knockdown cells. These results suggested that Six1 may be responsible for enhanced proliferation of metastatic melanoma cells, implicating it as a potential oncogene. This data is consistent with previous cancer studies. In pancreatic cancer, forced overexpression of Six1 enhanced the growth of cells and knockdown of Six1 significantly reduced pancreatic cancer cell proliferation both in vitro and in vivo (Li et al., 2013).  It is still uncertain as to how Six1 is involved in these pro-proliferative functions as the interactions and downstream effects of Six1 have yet to be studied in melanoma. However, several mechanisms have been proposed by teams investigating Six1 in other neoplasms, mostly relating to cell cycle control. A study on breast cancer discovered that the DNA damage-induced G2 cell cycle checkpoint was evaded in cells overexpressing Six1, thereby promoting entrance into and progression through the S phase (Ford et al., 1998). In both pancreatic cancer and rhabdomyosarcoma, studies reported that Six1 increases cellular proliferation by upregulating cyclin D1 expression by directly regulating cyclin D1 promoter   75 activity (Li et al., 2013; Yu et al., 2006). Cyclin D1 enhances cellular proliferation and cell cycle progression as it is a regulator of the G1/S phase transition of the cell cycle (Li et al., 2013). Furthermore, a study found that Six1 reactivated cyclin A1 in breast cancer cells, which was required for Six1 pro-proliferative properties (Coletta et al., 2004). Cyclin A1 is involved in entrance into and progression through S phase and the G2/M transition, controls cellular proliferation and is also involved in DNA repair and cell survival (Coletta et al., 2008; Ji et al., 2005). Another protein that promotes proliferation, c-myc, was found to be transcriptionally regulated by Six1 in rhabdomyosarcoma (Yu et al., 2006). These studies point to Six1 acting as a cell cycle regulator and stimulating proliferation in cancer through these means, which could also be true in melanoma pending further studies. Nonetheless, our results agree with other cancer studies, supporting the notion that Six1 may have control over the proliferative ability of cancer cells.  There were some differences observed when apoptotic proteins were assessed via Western blot, via the intrinsic apoptotic pathway (Bcl-2, Bax, pro-caspase 9 and pro-caspase 3). No differences were seen in caspase 8, which is part of the extrinsic apoptotic pathway. An increase in Bax and decrease in Bcl-2 is consistent with increased apoptosis, and a decrease in pro-caspase 9 and pro-caspase 3 further supports these results. These data suggest that Six1 may affect apoptotic pathways in melanoma cells, particularly the intrinsic pathway. While searching the literature we found that this data is in line with results from other studies on Six1 and apoptosis. First, a study on gangliogenesis observed that the absence of Six1 and Six4 in mice resulted in massive apoptosis and activation of cleaved caspase 3 (Konishi et al., 2006). Another study found that more apoptotic cells were present in the   76 osteosarcoma Six1 knockdown cell group and an upregulation of cleaved caspase 3 was observed (Hua et al., 2013). Six1 knockdown in gastric cancer cells promoted mitochondrial apoptosis by repressing Bcl-2 and activating executor caspase-7 (Du et al., 2017). These studies are consistent with our findings.  However, another study on ovarian carcinoma revealed that Six1 overexpression caused resistance to TRAIL-mediated apoptosis (Behbakht et al., 2007). This links Six1 to the extrinsic apoptotic pathway, thereby conflicting with our findings in melanoma. Several reasons for this difference can be surmised. First, not all cancers express the same molecular pathway abnormalities; target genes may differ due to tissue-specific requirements and mechanisms in melanoma compared to ovarian carcinoma. Also, we did not compare TRAIL-mediated apoptosis, as the Behbakht et al. did; we only tested pro-caspase 8. Another possibility is that the knockdown level of Six1 was not sufficient to see consistent results in apoptotic pathways; apoptosis may only be partially affected by the level of knockdown that was achieved in the current study. Clearly, further investigation is required on Six1 and its effects on apoptosis in metastatic melanoma cells. The mechanisms by which Six1 controls apoptosis remain unclear at this time. A study found that Six1 regulated p53 in breast cancer, an important tumor suppressor (Towers et al., 2015). Overexpression of Six1 reduced p53 expression, which inhibited apoptosis and cell cycle arrest in breast cancer cells (Towers et al., 2015). Therefore, it is possible that Six1’s regulation of p53 expression could be responsible for its effects on apoptosis in cancer cells. However, like proliferation, no conclusions can be drawn regarding the mechanism of Six1 control over apoptosis in melanoma.    77 Embryological and developmental studies on Six1 have also determined that it is important for cell growth and survival. Studies involving Six1 knockout mice have reported the failure of organ development due to decreased cellular proliferation and increased apoptosis (Ozaki et al., 2004; Xu et al., 2003; Zheng et al., 2003). These findings suggest that Six1 is important for cellular proliferation and apoptosis, acting to promote growth and survival during embryogenesis. Therefore, it is possible that Six1 may be recapitulating its embryological role in melanoma, which will be a common theme throughout this discussion. Taken together, these data suggest that Six1 may play a role in the cell growth, proliferation and apoptosis of metastatic melanoma cells in vitro. Since a marked decrease in cell growth and proliferation and an increase in apoptosis occurred in knockdown Six1 cells, it stands to reason that Six1 knockdown could be responsible for these differences. Our findings support that Six1 overexpression in metastatic melanoma may bestow cells with a growth and survival advantage. The prognosis for melanoma decreases dramatically once metastasis has occurred. A process in which melanoma cells undergo to promote metastasis is EMT, which includes the acquisition of migratory and invasive properties. EMT is one of the hallmarks of cancer and is a major stepping stone for cancer to spread successfully from the primary site to distant sites in the body (Hanahan & Weinberg, 2011). Therefore, identifying supportive molecular components and mechanisms involved in metastasis is important for our understanding of how it occurs and will provide insight into avenues by which is can be inhibited. We wanted to know if the overexpression of Six1 was contributing to EMT properties of melanoma cells, as has been proposed in other cancers. To do so, we assessed the effects of Six1 knockdown on migration and invasion by using a scratch assay and Boyden chamber assay, respectively.   78 Our results indicate that the knockdown of Six1 negatively affects both migration and invasion of two metastatic melanoma cells in vitro. A significant decline in the migratory ability of Six1 knockdown cells led us to propose that Six1 may play a role in the cellular migration of melanoma cells. A reduction in Six1 also inhibited the ability of a cell to invade through a matrigel matrix, indicating that Six1 may also contribute to melanoma cell invasion. Therefore, our data suggests that the absence of Six1 in metastatic melanoma cells somehow inhibits their ability to migrate and invade, implying that Six1 may control molecular pathways associated with cellular migration and invasion. Our findings are consistent with Six1 EMT studies in other neoplasms. Knockdown of Six1 has contributed to decreased EMT in several cancers, such as colorectal and pancreatic cancer (Lerbs et al., 2017; Li et al., 2014). Although how Six1 contributes to EMT is unknown in melanoma, several mechanisms have been suggested in other cancers. A study on colorectal cancer found that SIX1 overexpression represses E-cadherin expression, which is implicated in EMT (Ono et al., 2012). E-cadherin normally suppresses motility and invasiveness, and so a decline in E-cadherin is associated with metastasis (Hanahan & Weinberg, 2011). Another study on Six1-overexpressing mammary glands discovered that Six1 was directly capable of driving normal adult cells to undergo EMT by gaining the mesenchymal markers ZEB1 and SMA and losing E-cadherin (McCoy et al., 2009). Six1 was also found to transcriptionally regulate Ezrin in rhabdomycosarcoma, a protein that links the plasma membrane with the cortical actin cytoskeleton of the cell (Yu et al., 2004). Ezrin regulates cytoskeletal organization and adhesion and has been connected to the metastatic spread of mammary and pancreatic adenocarcinomas (Yu et al., 2004; Yu et al., 2006). Additionally, in breast cancer, Six1 induced EMT by increasing transforming growth factor   79 beta (TGF- signaling (Micalizzi et al., 2009). Although TGF- normally has tumor suppressive properties, such as cytostasis and differentiation, it plays a dichotomous role by switching to tumor-promotion in cancer (Massague, 2012). The upregulation of TGF- by Six1 is another pathway in which Six1 has been found to promote metastasis and invasion, as TFG- is a potent stimulator of EMT (Massague, 2012). Six1 is also known to play a role in EMT during embryogenesis, which involves a significant amount of cell migration and invasion while organs and body structures are formed. Evidence to support Six1’s role in developmental EMT comes from a study on myogenesis, which found that Six1 was required for myogenic precursor cells to delaminate from the dermomyotome and migrate to the limb bud (Grifone et al., 2005; McCoy et al., 2009). Additionally, during early kidney organogenesis, Six1 is necessary for the invasion of the ureteric bud into the metanephric mesenchyme (Xu et al., 2003). Six1 has also been implicated in several EMT-promoting pathways during normal development, including the Notch pathway, Wnt/-catenin pathway and Sonic hedgehog pathway (Micalizzi et al., 2009).  Several studies have theorized that cancerous cell populations could in fact be hijacking the normal developmental pathway of Six1 and exploiting it for their gain (Coletta et al., 2004; Yu et al., 2006). It is posited that this could be accomplished, at least in part, by utilizing Six1 out-of-context. Normally, Six1 ceases its functional roles once the human embryo reaches full maturity, indicated by its reduced expression in adult tissues. But it seems that in cancer, Six1 is being re-stimulated and proceeding to function beyond its normal role in embryo development. Therefore, it is possible that Six1 is reprising these embryonic roles in metastatic melanoma, such as stimulation of growth and EMT, to promote   80 tumorigenesis. Similar to Six1, transcription factors such as Snail, Slug, Twist and Zeb1/2 have been implicated in cancer after originally being discovered by developmental genetics (Hanahan & Weinberg, 2011). These transcription factors are known to have control over the embryological EMT and migration processes, much like Six1. Subsequently, the upregulation of the SIX1 gene and protein in metastatic melanoma led us to propose that the Six1 protein may be a disease marker with prognostic value for melanoma. We tested this hypothesis and provided evidence for its potential usefulness as a prognostic tool for metastatic melanoma by performing immunohistochemistry (IHC) on a tissue microarray (TMA) made up of metastatic melanoma patient biopsies (FFPE).   In our study, we found that melanoma progression was associated with a profound shift of the Six1 protein from the nucleus to the cytoplasm. Nuclear expression of Six1 was significantly reduced in primary melanoma and metastatic melanoma in comparison to normal nevi and dysplastic nevi. Opposite to nuclear Six1, cytoplasmic expression of Six1 was significantly increased in cancerous lesions. Furthermore, our analysis of five-year disease-specific survival showed that melanoma patients in whom Six1 was excluded from the nucleus demonstrated a significantly worse prognosis than the patients who retained nuclear expression. The inverse was true for cytoplasmic Six1 staining. Based on these findings we infer that either the absence of nuclear expression or elevated cytoplasmic presence (or both) of Six1 may play a role in the progression of melanoma from benign to metastatic lesions and be a marker of poor prognosis. This difference in Six1 expression may prove useful in predicting prognosis of melanoma histologically, pending further study. These findings were further supported by observing the correlation between Six1 protein expression and clinicopathological characteristic of melanoma (Table 3.1 and 3.2).   81 We discovered that low nuclear and high cytoplasmic Six1 expression was significantly correlated with advanced AJCC stages and increased melanoma thickness. This suggests that a shift of Six1 from the nucleus to the cytoplasm may be involved in the advancement of a melanoma tumor and/or the vertical growth phase of melanoma. As AJCC stages progress and tumor thickness increases, the aggressiveness of melanoma escalates, indicating that Six1 may be linked to the pathogenesis of melanoma. This further supports Six1 as a marker of poor prognosis. Additionally, our analysis revealed the correlation between Six1 nuclear and cytoplasmic expression and subtype of melanoma. Nodular melanoma was found to be significantly correlated with a lower nuclear Six1 and higher cytoplasmic Six1 compared to the other subtypes, which could be because nodular melanoma is a more aggressive and quickly progressing form of melanoma (Crowson et al., 2006). This also fits with low nuclear Six1 and high cytoplasmic Six1 being associated with a worse prognosis. The relationship between melanoma subtype and genetic changes has not yet been established, but it is not uncommon for melanoma subtypes to differ in their genetic signature such as with BRAF and KIT mutations (Ghosh & Chin, 2009). Furthermore, a whole-genome study of the mutational landscape of melanoma discovered that melanoma subtypes involved diverse carcinogenic processes (Hayward et al., 2017). Moreover, the loss of nuclear Six1 was associated with the presence of ulceration and histological satellitosis. Both characteristics are associated with decreased survival, as increased ulceration is thought to provide angiogenic support to a tumor and histological satellitosis indicates tumor dissemination (Rigel et al., 2011). This is in line with the correlation between the loss of nuclear Six1 and reduced survival. Our findings of increased Six1 cytoplasmic staining correlating with melanoma progression and five-year disease-free survival rate are consistent with Six1 prognostic   82 studies in other cancers, such as prostate, ovarian and hepatocellular carcinoma (Behbakht et al., 2007; Ng et al., 2006; Zeng et al., 2015). For example, in ovarian carcinoma, Six1 protein expression was found to correlate with poor overall survival, advanced pathologic tumor node metastasis (pTNM) stages and venous infiltration (Ng et al., 2006). However, no other studies have reported a nuclear to cytoplasmic shift of Six1 during cancer progression. It should be noted that the studies analyzed overall Six1 staining instead of separate nuclear and cytoplasmic staining. The subcellular localization of the Six1 protein in metastatic melanoma remains ambiguous. Multiple studies such as those in breast cancer have found, following immunohistochemistry on human neoplastic tissue with an identical antibody, the Six1 protein to be present primarily in the nucleus (Blevins et al., 2015; Iwanaga et al., 2012). This is expected, considering Six1 is a transcription factor and its primary function occurs in the nucleus where it controls the transcription of genes. However, our study found Six1 to be predominantly expressed in the cytoplasm of human metastatic melanoma tissues.  In agreement with our results, several studies have been published which found predominant cytoplasmic Six1 protein expression in several cancers. In prostate cancer, Six1 IHC staining revealed its location to be mainly in the cytoplasm and perinucleus of prostate cancer cells (Zeng et al., 2015). Six1 IHC staining in breast phyllode tumours found only cytoplasmic Six1 to correlate with tumour grade (Tan et al., 2014). Furthermore, an immunofluorescence study on hepatocellular carcinoma cells revealed that the Six1 protein was mainly located in the cytoplasm (Kong et al., 2014). In addition, they noted that non-tumor liver cells expressed the Six1 protein in the nucleus (Kong et al., 2014). In pancreatic   83 ductal adenocarcinoma, IHC staining patterns of Six1 were reported as mainly cytoplasmic/perinuclear (Jin et al., 2014).  Cytoplasmic Six1 has not only been observed in cancerous tissues. A team studying the expression and activity of Six1 during embryogenesis used in situ hybridization to visualize the location of Six1. Six1 was detected in the cytoplasm during the first four weeks of embryogenesis (Fougerousse et al., 2002). After week four, Six1 accumulated in the nucleus (Fougerousse et al., 2002; Wu et al., 2012). This suggests that perhaps Six1 has a cytoplasmic function during embryogenesis. It can be speculated that metastatic melanoma cells may revert to an embryonic state causing Six1 to sequester Six1 in the cytoplasm, perhaps to perform a yet to be determined function. These conflicting results regarding Six1 subcellular localization are puzzling. No studies have been done to suggest a reason for these inconsistent accounts of Six1 localization, and few have questioned or provided an explanation for these results. It is difficult to speculate the meaning of these results, as the localization of the Six1 protein and the molecular mechanisms controlling it remain unclear. Using the current knowledge of Six1, some hypotheses can be considered. To begin, it is possible that Six1 acquired a mutation during melanoma progression which prevents it from entering the nucleus. One possibility includes a mutation in Six1’s nuclear localization signal, required for nuclear entry, which would block it from the nucleus. A study which investigated various regions of the Six1 protein using truncated versions and their distributions within the cell found that the Six1 homeodomain (HD) was required for nuclear localization, as the HD was more prominent in the nucleus whereas the SD favoured the cytoplasm (Wu et al., 2012). This suggests that the Six1 protein in metastatic melanoma   84 could have a defect in its HD which disrupts its subcellular localization to the nucleus, leaving it to function in the cytoplasm using its SD. This may be worth investigating in the future. Furthermore, Six1 could be sequestered in the cytoplasm for some other function. Although cytoplasmic functions of Six1 have not yet been determined, other transcription factors have been found to have additional functions in the cytoplasm, such as p53 and ATF2 (Green & Kroemer, 2009; Lau & Ronai, 2012). Since the Six1 protein consists of two domains, the DNA binding homeodomain (HD) and protein binding six domain (SD), it is possible that Six1 may interact with other proteins in the cytoplasm through its SD. This could be contributing to its aberrant activity in metastatic melanoma. Furthermore, the presence of Six1 in the cytoplasm during the first four weeks of embryogenesis further supports the idea of a cytoplasmic function (Fougerousse et al., 2002). Indeed, other homeoproteins have been known to partake in non-transcriptional activities such as protein stability, mRNA export and translation (Haria & Naora, 2013). In addition, altered subcellular localization of proteins have been observed in the pathogenesis of other cancers. For example, PELP1 (proline-, glutamic acid-, and leucine-rich protein-1) is a coregulator of transcription factors, and its increased cytoplasmic presence was found to modulate additional cellular processes such as activation of MAPK and AKT (Vadlamudi et al., 2005).  Additionally, when considering the technical aspects of the immunohistochemistry process, it is a possibility that the Six1 antibody was being blocked from entering the nucleus to detect the nuclear Six1 protein. This would account for the build-up of Six1 antibody in the cytoplasm, and would thereby mimic cytoplasmic staining. This exclusion of the Six1   85 antibody from the nucleus could be the result of the sample tissue matrix organization and cellular structure.  The conflicting accounts of subcellular localization of Six1 and the lack of studies pertaining to the cytoplasmic role of Six1 demonstrate that there is still much to be learned about this protein. It also raises many questions regarding Six1’s location and function to which the answers are presently unclear. Since nuclear Six1 is lost during melanoma progression, this observation supports a tumor suppressive function, as its loss is associated with a worse prognosis. However, an increase of Six1 in the cytoplasm during melanoma progression indicates that cytoplasmic Six1 may function as an oncogene, as its gain is correlated with a shorter survival time. This raises the possibility of a dual role of the Six1 protein based on subcellular localization. However, a difference in subcellular localization does not necessarily indicate that the protein is changing from tumor suppressor to tumor promotor; the shift in location may be the result of Six1 changing from a neutral to an oncogenic protein. At this point, not enough information is available to determine the answer and further study in this area is required. Nonetheless, the in vitro results from this project support that Six1 most likely functions as an oncogene in metastatic melanoma, given the reduction in cancerous phenotypes following Six1 knockdown. If Six1 were acting as a tumor suppressor in metastatic melanoma, Six1 knockdown would have promoted cancerous phenotypes. The association of increased cytoplasmic Six1 with melanoma progression, advanced AJCC staging and increasing tumor thickness also implicates it as a tumor-promoting protein. Furthermore, the upregulation of a gene in cancer usually suggests an oncogenic function, not a tumor suppressive function.  Currently, prognostic indicators of metastatic melanoma are mainly limited to   86 histopathological features. These include tumor thickness, lymph node involvement, distant metastasis, ulceration, regression, mitotic rate, lymphovascular invasion, age, histological type and tumor infiltrating lymphocytes. There is a deficit in reliable molecular markers, and none have been approved for clinical use. This is unfortunate, given that the pathogenesis of melanoma is driven by molecular events. The absence of molecular markers in the clinic impedes prognostication for metastatic melanoma patients, and so implementing knowledge of the molecular biology of melanoma in the clinic could assist in more accurate prognoses. Adding molecular markers in the clinic could aid clinicians in predicting prognosis in numerous ways. The ability to predict the aggressiveness of disease is critical for deciding the appropriate treatment plan for patients. If it were possible to use a molecular marker such as Six1 in conjunction with histopathological features to better determine the prognosis of patients, improved sensitivity of prognosis could prevent a patient from having to endure excessive toxic treatments if their melanoma is anticipated to be less aggressive. Likewise, patients afflicted with cancers harboring a molecular marker pattern with a more troubling prognosis would signal the physician to provide them with the appropriate level of treatment to combat a more aggressive form of disease. This would improve patient quality of life and put less of a strain on patient finances and the healthcare system. Furthermore, with additional study, knowledge of molecular markers such as Six1 has the potential to be utilized as a method of monitoring patient response to treatment. In conclusion, this study found for the first time that the SIX1 gene and protein expression was aberrantly expressed in metastatic melanoma. Upon Six1 knockdown, in vitro models revealed marked reductions in cell growth, proliferation, migration and invasion and a slight increase in apoptosis, suggesting a pathogenic role for Six1. Our results provide   87 evidence of a link between increased cytoplasmic Six1 protein expression, decreased nuclear protein expression and metastatic melanoma patient survival. The TMA data revealed a profound shift of Six1 expression from the nucleus to the cytoplasm during melanoma progression, and loss of nuclear and gain of cytoplasmic Six1 expression was associated with multiple histopathological characteristics. Our findings suggest that Six1 most likely functions as an oncogene, and therefore the targeting of Six1 for inhibition may be a possible therapeutic approach for metastatic melanoma. Furthermore, characterization of Six1 protein expression may serve as a prognostic marker in metastatic melanoma.  4.1. Limitations and future directions Despite the promising results presented in this study, limitations of this pilot study do exist. This pilot study also raises many questions regarding Six1 and its role in melanoma progression and pathogenesis. This section will discuss the limitations of this work and possible future directions. One limitation of this study was the utilization of in vitro human cell models, which only provide a static environment. Human cells grown in vitro are devoid of their natural environment within the human body, including but not limited to: chemoattractants and chemotaxis (with respective gradients), cell signaling molecules such as cytokines and hormones, and an abundant yet tightly controlled variety of other molecules contained within and excreted by components of the human body (Aldinucci & Colombatti, 2014; Dhawan & Richmond, 2002). Furthermore, the cell systems we used were composed of only one cell type, which does not account for the diverse cell-cell interactions that cells are normally subjected to in their natural environment (Hartung & Daston, 2009). It would be of greater accuracy and of extreme value to assess Six1’s metastatic functions in vivo in mouse tumor   88 models, which would mimic the heterogeneous, continually fluctuating environment found within the human body. In the future, it is advisable to move to melanoma tumor initiation and formation in immunodeficient mice (xenografts) and subsequent monitoring of tumor growth and metastatic dissemination. This will allow a more precise determination of the true biological behaviour and function of Six1 in metastatic melanoma. More in vitro experiments could also be performed on the Six1 knockdown melanoma cells in the future. For example, an anoikis assay would test if Six1 knockdown has an effect on anchorage-dependent cell death in melanoma cells. Other possible in vitro assays include colony formation assays and additional apoptosis assays such as Annexin V using flow cytometry and analysis of TRAIL-mediated apoptosis.  Furthermore, our TMA patient database did not include serum LDH level, sentinel lymph node biopsy status and mutations like BRAFV600, and so our study was missing important prognostic features. In the future, these should be recorded and included in a melanoma patient database as these are now important clinical features and would allow for more comparisons. Also, some patient information was not available in the clinical database. Not all patients had certain clinicopathological characteristics recorded, resulting in smaller sample sizes when comparing Six1 staining and these characteristics. This was out of our control, but something to note which reduced the power of the study.  In the future, it is necessary to expand the melanoma patient biopsy pool to determine the feasibility of Six1 as a prognostic marker. Although our TMA included a modest sample size, biomarkers must be tested at higher levels to increase the accuracy of the findings. Furthermore, our TMA analysis did not include an independent external validation cohort, which is required for statistical and clinical validity (Altman & Royston, 2000). Therefore, to   89 evaluate the performance and transportability of Six1 as a prognostic marker, testing it in a separate patient population in the future is advisable (Collins et al., 2016). Additionally, the biopsies studied only included Caucasian patient samples from Vancouver, British Columbia, which limited the assumptions we could make about Six1 and melanoma in a global context. The study of Six1 in melanoma should be expanded to include patients of different ethnicities and from different parts of the world. Furthermore, the small size of the tissue cores used for TMA construction is a major shortcoming of this technology due to the heterogeneity of tumors. The small TMA cores may not accurately reflect the staining pattern observed from the whole tissue sections. Another point to consider is the use of shRNA for gene knockdown. Although it is an improvement from siRNA as it is longer acting and does not require repeated knockdown, it is still with limitations. Specificity is known to be a limitation of shRNA technology. A lack in specificity can occur when the shRNA fails to bind to the intended target (i.e. SIX1) 100% of the time. This is problematic not only because the shRNA is not exerting its full effects to knock down the target gene fully, but because the shRNA itself can cause off-target effects. One way this is known to occur is when an antisense strand of shRNA binds imperfectly to a similar complement mRNA and proceeds to degrade it (Mockenhaupt et al., 2015). This has been shown to occur in both in vitro and in vivo (Mockenhaupt et al., 2015; Song et al., 2015; Masuda et al., 2016). Luckily, off-target effects have been found to be less dramatic than the intended target.  One way to improve this study in the future would be to upgrade shRNA to Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) technology. Although both shRNA and CRISPR aim to silence genes, CRISPR was found to be more effective than   90 shRNA when utilizing them in negative selection (dropout) screens (Evers et al., 2016). The study found that shRNA was subject to more off-target effects than CRISPR (Evers et al., 2016). Additionally, the CRISPR method resulted in less data variation than shRNA (Evers et al., 2016). Furthermore, while shRNA utilizes an endogenous, natural process within cells which may disrupt endogenous microRNAs, CRISPR does not (Boettcher & McManus, 2015). CRISPR is also a more precise mechanism of inactivating genes, given that is acts at the level of DNA, instead of post-translationally like shRNA. It is more reliable for silencing genes as it is a method of gene knockout, whereas shRNA produces variable results, sometimes only knocking down genes partially. These benefits make CRISPR a more reliable technology for gene manipulation. It is also beneficial to consider upstream and downstream components of Six1 in the future, as Six1 is likely part of many molecular pathways. Studies on Six1 in melanoma have not been published, and although we can consider that its interactions may be related to other cancers that have been studied in the past, it cannot be certain until concrete studies are done on metastatic melanoma. This is because each cancer is a distinct and unique disease, even within the same organ and/or tissue type and between individuals. A starting point would be to assess genes and proteins that Six1 has been found to interact with in other cancers, such as cyclin A1, cyclin D1, Ezrin, Zeb1 and TGF-. Additionally, RNA sequencing could be performed on control and Six1 knockdown cells. In this way, difference in gene expression could be compared to identify possible downstream targets of Six1 specific to melanoma. Furthermore, methods such as ChIP-seq (chromatin immunoprecipitation sequencing), RIME (rapid immunoprecipitation mass spectrometry of endogenous proteins), co-  91 immunoprecipitation and/or protein affinity chromatography could be performed to identify protein-DNA and protein-protein interactions (Fontaine et al., 2015). Earlier, the typical location of the Six1 protein was illustrated as a puzzling variable, considering the conflicting accounts of nuclear vs. cytoplasmic findings within the literature. We found that Six1 seems to shift from the nucleus to the cytoplasm during melanoma evolution, which is a paradox since it is a transcription factor. Location is closely related to function within a cell, and vice versa, which dictates that knowledge in this area would be useful to expand upon within melanoma. Some examples of exploratory expeditions that can be taken are related to nuclear translocation signals and nuclear pore complexes, immunocytochemistry before and after Six1 knockdown, and other molecular biology and imaging techniques to further explore Six1’s subcellular location in melanoma. It would also be interesting to extract the nuclear Six1 and cytoplasmic Six1 protein separately from cells and test them for structural and functional differences. The oncogenic activities of homeobox genes in cancers are generally believed to be a result of the embryonic, normal functions of the wild-type homeoprotein occurring in an inappropriate cellular context, rather than a mutant form of the protein (Abate-Shen, 2002). However, this should not be assumed, and it would still be worthwhile to investigate if any abnormalities in Six1 are occurring in metastatic melanoma. For example, the SIX1 gene could be sequenced in cancerous melanocytes to determine if there are any mutations in the gene. In addition, performing fluorescence in situ hybridization (FISH) would uncover any abnormalities of the SIX1 locus on chromosome 14 which may be causing the overexpression of SIX1, such as copy number irregularities and chromosomal rearrangements. It would also be useful to compare the Six1 protein structure in cancerous   92 melanocytes and normal melanocytes, as differences in protein structure could contribute to abnormal functions and behaviour, such as sequestration in the cytoplasm.  Another future direction that would enhance the understanding of Six1 in melanoma would be to perform an experiment to upregulate Six1 in the same cell lines. This is another way to test if the phenotypes observed in these experiments are due to the knockdown of Six1, as overexpression of Six1 should promote the opposite cellular phenotype. To accomplish this, an expression vector could be used to increase the amount of Six1 within cells. For example, a plasmid could be designed to contain the Six1 gene and a promotor such as a CMV-promoter to drive the production of Six1 protein. The final future direction that will be touched upon is the identification and testing of targeted therapies of Six1. Metastatic melanoma requires more effective treatments, and since Six1 has been implicated in pathogenic characteristics of melanoma in vitro, it is plausible that a therapy targeting Six1 may be beneficial. Since Six1 is an embryonic gene and is not highly expressed in most adult tissues, it has the possibility of becoming a selective and more direct target for cancer cells, leading to less adverse effects on normal tissues (Liu et al., 2015). However, transcription factors are notorious for being difficult to inhibit, considering their large surface area for protein-protein and protein-DNA interactions and their main function being in the nucleus which is not always accessible to drugs (Yeh et al., 2013). Furthermore, transcription factors generally lack hydrophobic pockets, or surface involutions, which does not allow high affinity binding by small molecules (Moellering et al., 2010). Although transcription factors present a challenge for inhibition, there are still options available.   93 One option that has proven successful in inhibiting transcription factors are the use of small molecule inhibitors. These tiny molecular weight organic compounds are still in the early stages today, but scientists have been experimenting with potential microscopic blockades of Six1 and its cofactors (Blevins et al., 2015; Fan et al., 2000; Krueger et al., 2013). The hope is that one of these infinitesimal (<900 Da) molecules will inhibit the function and subsequent biological effect of the Six1 protein, its cofactors and/or complexes (Macielag, 2012). It is hypothesized by these studies that this blockade will extinguish or lessen the observed maleficent effects of Six1 and its team of co-factors and downstream targets (Blevins et al., 2015; Fan et al., 2000; Krueger et al., 2013). Uncovering which small molecular inhibitors to test for Six1 inhibition presents as an arduous task, but can be accomplished using high throughput screening of already available synthetic compounds (Szymanski et al., 2012). This will provide a starting point by identifying candidate molecules for testing purposes. Once candidates are available, they can be tested for their specificity, potency, and mechanism of action (Szymanski et al. 2012). This can be accomplished by using assays that will analyze the biochemical interaction, structural properties and in vitro effects of these compounds (Szymanski et al., 2012). Since targeting Six1 may prove difficult, targeting its association with its cofactors could also work to inhibit Six1 function. Six1 is reliant on direct interactions with transcriptional cofactors, such as Eya proteins, to perform its regulatory functions (Bhagwat & Vakoc, 2015). Therefore, possible therapeutic targets include the surfaces of Six1 required for its interactions with cofactors (Bhagwat & Vakoc, 2015). In fact, there are already investigations being pursued focusing on targeting the Six1/Eya2 interface, having found that Eya2 is required for numerous pro-metastatic characteristics of Six1 (Patrick et al., 2013). Of   94 course, cofactors of Six1 will need to be investigated in metastatic melanoma specifically. Furthermore, the biochemical structure of Six1 could to be investigated to determine the appropriate binding locations of cofactors, such as X-ray crystallography and nuclear magnetic resonance (NMR) spectroscopy (Alberts et al., 2002). Another method of Six1 inhibition would be to examine its regulatory factors and exploit them. For example, a study focusing on microRNAs in melanoma cells found that miR-185 may in fact be linked to Six1 in melanoma. miR-185 was demonstrated to be a tumor suppressor in melanoma xenografts by inhibiting proliferation, invasion and tube formation (Greenberg et al., 2011). They link miR-185 to studies in which it was found to suppress Six1 (Greenberg et al., 2011; Imam et al., 2010; Takahashi et al., 2009). This idea is not tested and merely speculation, but it suggests that it may be worthwhile to explore miR-185 and its relevance to Six1’s function in metastatic melanoma. This presents as another way to inhibit Six1—by harnessing the power of its molecular regulators such as the potential candidate miR-185 and other microRNAs. 5 Summary and Conclusion In summary, this pilot study has determined for the first time that the transcription factor Six1 may indeed play a role in metastatic melanoma. SIX1 transcript and protein is overexpressed in metastatic melanoma, which is consistent with other neoplasms. Our data suggests that Six1 may play a role in cancerous phenotypes including the cell growth, proliferation, apoptotic, migration, and invasion abilities of metastatic melanoma cells in vitro. This suggests that it may promote metastatic characteristics and stimulate malignant cellular behaviour in melanoma cells in vitro.   95 We also determined that the Six1 protein correlated strongly with five-year disease-specific survival and melanoma progression. The TMA data revealed that a shift of Six1 protein from the nucleus to the cytoplasm correlated with melanoma progression and decreased survival. Therefore, it is possible that the Six1 protein may play a dual role of an oncogene and a tumor suppressor, depending on whether it is in the cytoplasm or the nucleus, respectively. Furthermore, high cytoplasmic Six1 was found to correlate with advanced AJCC stages, increased tumor thickness and nodular melanoma, while low nuclear Six1 was associated with advanced AJCC stages, ulceration, histological satellitosis and nodular melanoma.  This project presents a promising molecular candidate to explore further in metastatic melanoma in the form of additional Six1 in vitro and in vivo studies. Taken together, these data support that Six1 may be a player in the molecular pathways of malignant melanoma. From this study, it can be surmised that Six1 may have the potential to become a therapeutic target for inhibition in the future and may be useful as a clinical prognostic marker. New, more effective and efficient therapies are needed to combat metastatic melanoma, and so the feasibility of Six1 as a therapeutic target and prognostic marker should be further explored.          96 Bibliography  Abate-Shen, C. (2002). Deregulated homeobox gene expression in cancer: cause or consequence? Nature Reviews Cancer, 2, 777+.  Alberts, B., Johnson, A., & Lewis J et al. (2002). Molecular biology of the cell (4th ed.). New York: Garland Science. Aldinucci, D., & Colombatti, A. (2014). The inflammatory chemokine CCL5 and cancer progression. Mediators of Inflammation, 2014, 1-12. Altman, D., & Royston, P. (2000). What do we mean by validating a prognostic model? Statistics in Medicine (Vol. 19).  American Cancer Society. (2016). Cancer facts and figures 2016. Cancer Facts & Figures 2016, 1–9.  Balch, C. M., Gershenwald, J. E., Soong, S., Thompson, J. F., Atkins, M. B., Byrd, D. R., … Sondak, V. K. (2009). Final version of 2009 AJCC melanoma staging and classification. Journal of Clinical Oncology, 27(36), 6199–6206.  Balmain, A. (2001). Cancer genetics: from Boveri and Mendel to microarrays. Nat Rev Cancer, 1(1), 77–82.  Bastian, B. C. (2014). The molecular pathology of melanoma: an integrated taxonomy of melanocytic neoplasia. Annual Review of Pathology, 9, 239–71.  Behbakht, K., Qamar, L., Aldridge, C. S., Coletta, R. D., Davidson, S. A., Thorburn, A., & Ford, H. L. (2007). Six1 overexpression in ovarian carcinoma causes resistance to TRAIL-mediated apoptosis and is associated with poor survival. Cancer Research, 67(7), 3036.  Bhagwat, A. S., & Vakoc, C. R. (2015). Targeting transcription factors in cancer. Trends in Cancer, 1(1), 53–65.  Bhatia, S., Tykodi, S. S., & Thompson, J. A. (2009). Treatment of metastatic melanoma: an overview. Oncology (Williston Park, N.Y.), 23(6), 488–96.  Blevins, M. A., Kouznetsova, J., Krueger, A. B., King, R., Griner, L. M., Hu, X., … Zhao, R. (2015). Small molecule, NSC95397, inhibits the CtBP1-protein partner interaction and CtBP1-mediated transcriptional repression. Journal of Biomolecular Screening, 20(5), 663–672.    97 Blevins, M. A., Towers, C. G., Patrick, A. N., Zhao, R., & Ford, H. L. (2015). The SIX1-EYA transcriptional complex as a therapeutic target in cancer. Expert Opinion on Therapeutic Targets, 19(2), 213–225.  Boettcher, M., & McManus, M. T. (2015). Choosing the right tool for the job: RNAi, TALEN or CRISPR. Molecular Cell, 58(4), 575–585.  Boucher, C. A., Carey, N., Edwards, Y. H., Siciliano, M. J., & Johnson, K. J. (1996). Cloning of the human SIX1 gene and its assignment to chromosome 14. Genomics, 33(1), 140–142.  Bresnick, A. R., Weber, D. J., & Zimmer, D. B. (2015). S100 proteins in cancer. Nature Reviews Cancer, 15(2), 96–109.  Camp, R. L., Dolled-Filhart, M., & Rimm, D. L. (2004). X-Tile: a new bio-informatics tool for biomarker assessment and outcome-based cut-point optimization. Clinical Cancer Research, 10(21), 7252–7259.  Cheng, Y. (2014). Tissue microarray based biomarker study in human cutaneous melanoma (Doctoral dissertation). Retrieved from cIRcle. Cheyette, B. N. R., Green, P. J., Martin, K., Garren, H., Hartenstein, V., & Zipursky, S. L. (2017). The drosophila sine oculis locus encodes a homeodomain-containing protein required for the development of the entire visual system. Neuron, 12(5), 977–996.  Christensen, K. L., Brennan, J. D. G., Aldridge, C. S., & Ford, H. L. (2006). Cell cycle regulation of the human Six1 homeoprotein is mediated by APCCdh1. Oncogene, 26(23), 3406–3414.  Clark, W. H., Elder, D. E., Guerry, D., Epstein, M. N., Greene, M. H., & Van Horn, M. (1984). A study of tumor progression: the precursor lesions of superficial spreading and nodular melanoma. Human Pathology, 15(12), 1147–1165.  Cokol, M., Nair, R., & Rost, B. (2000). Finding nuclear localization signals. EMBO Reports, 1(5), 411 LP-415.  Coletta, R. D., Christensen, K. L., Micalizzi, D. S., Jedlicka, P., Varella-Garcia, M., & Ford, H. L. (2008). Six1 overexpression in mammary cells induces genomic instability and is sufficient for malignant transformation. Cancer Research, 68(7), 2204.  Coletta, R. D., Christensen, K., Reichenberger, K. J., Lamb, J., Micomonaco, D., Huang, L., … Ford, H. L. (2004). The Six1 homeoprotein stimulates tumorigenesis by reactivation of cyclin A1. Proceedings of the National Academy of Sciences of the United States of America, 101(17), 6478–83.    98 Collins, G. S., Ogundimu, E. O., & Altman, D. G. (2016). Sample size considerations for the external validation of a multivariable prognostic model: a resampling study. Statistics in Medicine, 35(2), 214–226.  Crowson, A. N., Magro, C. M., & Mihm, M. C. (2006). Prognosticators of melanoma, the melanoma report, and the sentinel lymph node. Mod Pathol, 19(S2), S71–S87.  Cullen, B. R. (2006). Enhancing and confirming the specificity of RNAi experiments. Nat Meth, 3(9), 677–681.  Damsky, W. E., Theodosakis, N., & Bosenberg, M. (2014). Melanoma metastasis: new concepts and evolving paradigms. Oncogene, 33(19), 2413–2422.  Davar, D., Lin, Y., & Kirkwood, J. M. (2015). Unfolding the mutational landscape of human melanoma. The Journal of Investigative Dermatology, 135(3), 659–62.  Day, D., & Siu, L. L. (2016). Approaches to modernize the combination drug development paradigm. Genome Medicine, 8(1), 115.  Delgado-Olguin, P., Huang, Y., Li, X., Christodoulou, D., Seidman, C. E., Seidman, J. G., … Bruneau, B. G. (2012). Epigenetic repression of cardiac progenitor gene expression by Ezh2 is required for postnatal cardiac homeostasis. Nat Genet, 44(3), 343–347.  Dhawan, P., & Richmond, A. (2002). Role of CXCL1 in tumorigenesis of melanoma. Journal of Leukocyte Biology, 72(1), 9–18.  Dimberg, L. Y., Anderson, C. K., Camidge, R., Behbakht, K., Thorburn, A., & Ford, H. L. (2013). On the TRAIL to successful cancer therapy? Predicting and counteracting resistance against TRAIL-based therapeutics. Oncogene, 32(11), 1341–50.  Du, P., Zhao, J., Wang, J., Liu, Y., Ren, H., Patel, R., … Huang, G. (2017). Sine oculis homeobox homolog 1 regulates mitochondrial apoptosis pathway via caspase-7 in gastric cancer cells. Journal of Cancer, 8(4), 636–645.  Dubas, L. E., & Ingraffea, A. (2013). Nonmelanoma skin cancer. Facial Plastic Surgery Clinics of North America, 21(1), 43–53.  Egger, M. E., Stepp, L. O., Callender, G. G., Quillo, A. R., Martin, R. C. G., Scoggins, C. R., … McMasters, K. M. (2013). Outcomes and prognostic factors in superficial spreading melanoma. The American Journal of Surgery, 206(6), 861–868.  El-Hashash, A. H. K., Al Alam, D., Turcatel, G., Rogers, O., Li, X., Bellusci, S., & Warburton, D. (2011). Six1 transcription factor is critical for coordination of epithelial, mesenchymal and vascular morphogenesis in the mammalian lung. Developmental Biology, 353(2), 242–258.    99 Elder, D. E. (2016). Melanoma progression. Pathology, 48(2), 147–154.  Evers, B., Jastrzebski, K., Heijmans, J. P. M., Grernrum, W., Beijersbergen, R. L., & Bernards, R. (2016). CRISPR knockout screening outperforms shRNA and CRISPRi in identifying essential genes. Nature Biotechnology, 34(6), 631–633.  Fan, X., Brass, L. F., Poncz, M., Spitz, F., Maire, P., & Manning, D. R. (2000). The alpha subunits of Gz and Gi interact with the eyes absent transcription cofactor Eya2, preventing its interaction with the Six class of homeodomain-containing proteins. Journal of Biological Chemistry, 275(41), 32129–32134.  Ferringer, T. (2012). Update on immunohistochemistry in melanocytic lesions. Dermatologic Clinics, 30(4), 567–579.  Fontaine, F., Overman, J., & François, M. (2015). Pharmacological manipulation of transcription factor protein-protein interactions: opportunities and obstacles. Cell Regeneration, 4, 2.  Ford, H. L., Landesman-Bollag, E., Dacwag, C. S., Stukenberg, P. T., Pardee, A. B., & Seldin, D. C. (2000). Cell cycle-regulated phosphorylation of the human SIX1 homeodomain protein. The Journal of Biological Chemistry, 275(29), 22245–54.  Ford, H. L., Kabingu, E. N., Bump, E. A., Mutter, G. L., & Pardee, A. B. (1998). Abrogation of the G(2) cell cycle checkpoint associated with overexpression of HSIX1: A possible mechanism of breast carcinogenesis. Proceedings of the National Academy of Sciences of the United States of America, 95(21), 12608–12613.  Ford, H., Patrick, A., & Varella-Garcia, M. (2006). SIX1 (sine oculis homeobox homolog 1) (mammalian). Atlas Genet Cytogenet Oncol Haematol, 10(2), 100–102. Frolova, N., Edmonds, M. D., Bodenstine, T. M., Seitz, R., Johnson, M. R., Feng, R., … Frost, A. R. (2009). A shift from nuclear to cytoplasmic breast cancer metastasis suppressor 1 expression is associated with highly proliferative estrogen receptor-negative breast cancers. Tumour Biology : The Journal of the International Society for Oncodevelopmental Biology and Medicine, 30(3), 148–159.  Fruebis, J., Tsao, T. S., Javorschi, S., Ebbets-Reed, D., Erickson, M. R., Yen, F. T., … Lodish, H. F. (2001). Proteolytic cleavage product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in muscle and causes weight loss in mice. Proceedings of the National Academy of Sciences of the United States of America, 98(4), 2005–10.  Fu, S. W., Kirolikar, S. P., Ginsburg, E., Tan, X., Schwartz, A., Simmens, S. J., … Berg, P. E. (2016). Beta protein 1 homeoprotein induces cell growth and estrogen-independent tumorigenesis by binding to the estrogen receptor in breast cancer. Oncotarget, 7(33), 53204–53216.    100 Gao, A. C., Lou, W., & Isaacs, J. T. (1998). Down-regulation of homeobox gene GBX2 expression inhibits human prostate cancer clonogenic ability and tumorigenicity. Cancer Research, 58(7), 1391.  Gebäck, T., Schulz, M. M. P., Koumoutsakos, P., & Detmar, M. (2009). TScratch: a novel and simple software tool for automated analysis of monolayer wound healing assays. BioTechniques, 46(4), 265–74.  Ghosh, P., & Chin, L. (2009). Genetics and genomics of melanoma. Expert Review of Dermatology, 4(2), 131.  Gillet, J.-P., Varma, S., & Gottesman, M. M. (2013). The clinical relevance of cancer cell lines. JNCI Journal of the National Cancer Institute, 105(7), 452–458.  Gogas, H., Eggermont, A. M. M., Hauschild, A., Hersey, P., Mohr, P., Schadendorf, D., … Dummer, R. (2009). Biomarkers in melanoma. Annals of Oncology : Official Journal of the European Society for Medical Oncology, 20 Suppl 6(Suppl 6), vi8-13.  Gray-Schopfer, V., Wellbrock, C., & Marais, R. (2007). Melanoma biology and new targeted therapy. Nature, 445(7130), 851–7.  Green, D. R., & Kroemer, G. (2009). Cytoplasmic functions of the tumor suppressor p53. Nature, 458(7242), 1127.  Greenberg, E., Hershkovitz, L., Itzhaki, O., Hajdu, S., Nemlich, Y., Ortenberg, R., … Markel, G. (2011). Regulation of cancer aggressive features in melanoma cells by microRNAs. PLOS ONE, 6(4), e18936-.  Grifone, R., Demignon, J., Houbron, C., Souil, E., Niro, C., Seller, M. J., … Maire, P. (2005). Six1 and Six4 homeoproteins are required for Pax3 and Mrf expression during myogenesis in the mouse embryo. Development, 132(9), 2235.  Grifone, R., Laclef, C., Spitz, F., Lopez, S., Demignon, J., Guidotti, J.-E., … Maire, P. (2004). Six1 and Eya1 expression can reprogram adult muscle from the slow-twitch phenotype into the fast-twitch phenotype. Molecular and Cellular Biology, 24(14), 6253–6267.  Hanahan, D., & Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Cell, 144(5), 646–674.  Haria, D., & Naora, H. (2013). Homeobox gene deregulation: impact on the hallmarks of cancer. Cancer Hallmarks, 1(2–3), 67–76.  Hartung, T., & Daston, G. (2009). Are in vitro tests suitable for regulatory use? Toxicological Sciences, 111(2), 233–237.    101 Hayward, N. K., Wilmott, J. S., Waddell, N., Johansson, P. A., Field, M. A., Nones, K., … Mann, G. J. (2017). Whole-genome landscapes of major melanoma subtypes. Nature, 545(7653), 175–180.  Hipfner, D. R., & Cohen, S. M. (2004). Connecting proliferation and apoptosis in development and disease. Nature Reviews Molecular Cell Biology, 5(10), 805–815.  Housden, B. E., & Perrimon, N. (2016). Comparing CRISPR and RNAi-based screening technologies. Nature Biotechnology, 34(6), 621–623.  Hu, S., Mamedova, A., & Hegde, R. S. (2008). DNA-binding and regulation mechanisms of the SIX family of retinal determination proteins. Biochemistry, 47(11), 3586–3594.  Hua, L., Fan, L., Aichun, W., Yongjin, Z., Qingqing, C., & Xiaojian, W. (2014). Inhibition of Six1 promotes apoptosis, suppresses proliferation, and migration of osteosarcoma cells. Tumor Biology, 35(3), 1925–1931.  Huang, Y. (2016). Clinical applications and functional characterization of TOX in cutaneous T-cell lymphoma (Doctoral dissertation). Retrieved from cIRcle.  Hussein, M. R., Haemel, A. K., & Wood, G. S. (2003). Apoptosis and melanoma: molecular mechanisms. The Journal of Pathology, 199(3), 275–288.  Igney, F. H., & Krammer, P. H. (2002). Death and anti-death: tumour resistance to apoptosis. Nature Reviews Cancer, 2(4), 277–288.  Ikeda, K., Kageyama, R., Suzuki, Y., & Kawakami, K. (2010). Six1 is indispensable for production of functional progenitor cells during olfactory epithelial development. The International Journal of Developmental Biology, 54(10), 1453–64.  Ikeda, K., Ookawara, S., Sato, S., Ando, Z., Kageyama, R., & Kawakami, K. (2007). Six1 is essential for early neurogenesis in the development of olfactory epithelium. Developmental Biology, 311(1), 53–68.  Imam, J. S., Buddavarapu, K., Lee-Chang, J. S., Ganapathy, S., Camosy, C., Chen, Y., & Rao, M. K. (2010). MicroRNA-185 suppresses tumor growth and progression by targeting the Six1 oncogene in human cancers. Oncogene, 29(35), 4971–4979.  Iwanaga, R., Wang, C.-A., Micalizzi, D. S., Harrell, J. C., Jedlicka, P., Sartorius, C. A., … Ford, H. L. (2012). Expression of Six1 in luminal breast cancers predicts poor prognosis and promotes increases in tumor initiating cells by activation of extracellular signal-regulated kinase and transforming growth factor-beta signaling pathways. Breast Cancer Research, 14(4), R100.  Jang, S., & Atkins, M. B. (2013). Which drug, and when, for patients with BRAF-mutant melanoma? The Lancet Oncology, 14(2), e60–e69.    102 Jemal, A., Siegel, R., Xu, J., & Ward, E. (2010). Cancer statistics, 2010. CA: A Cancer Journal for Clinicians, 60(5), 277–300.  Ji, P., Agrawal, S., Diederichs, S., Bäumer, N., Becker, A., Cauvet, T., … Müller-Tidow, C. (2005). Cyclin A1, the alternative A-type cyclin, contributes to G1/S cell cycle progression in somatic cells. Oncogene, 24(16), 2739–2744.  Jin, A., Xu, Y., Liu, S., Jin, T., Li, Z., Jin, H., … Lin, Z. (2014). Sineoculis homeobox homolog 1 protein overexpression as an independent biomarker for pancreatic ductal adenocarcinoma. Experimental and Molecular Pathology, 96(1), 54–60.  Johnson, D. B., & Sosman, J. A. (2015). Therapeutic advances and treatment options in metastatic melanoma. JAMA Oncology, 1(3), 380.  Justus, C. R., Leffler, N., Ruiz-Echevarria, M., & Yang, L. V. (2014). In vitro cell migration and invasion assays. Journal of Visualized Experiments, (88).  Kahlert, C., Lerbs, T., Pecqueux, M., Herpel, E., Hoffmeister, M., Jansen, L., … Koch, M. (2015). Overexpression of SIX1 is an independent prognostic marker in stage I–III colorectal cancer. International Journal of Cancer, 137(9), 2104–2113.  Kalkhoran, S., Milne, O., Zalaudek, I., & al, et. (2010). Historical, clinical, and dermoscopic characteristics of thin nodular melanoma. Archives of Dermatology, 146(3), 311–318.  Kawakami, K., Sato, S., Ozaki, H., & Ikeda, K. (2000). Six family genes: structure and function as transcription factors and their roles in development. BioEssays, 22(7), 616-26. Khan, A. A., Betel, D., Miller, M. L., Sander, C., Leslie, C. S., & Marks, D. S. (2009). Transfection of small RNAs globally perturbs gene regulation by endogenous microRNAs. Nature Biotechnology, 27(6), 549–555.  Kong, J., Zhou, X., Liu, S., Jin, T., Piao, Y., Liu, C., & Lin, Z. (2014). Overexpression of sineoculis homeobox homolog 1 predicts poor prognosis of hepatocellular carcinoma. International Journal of Clinical and Experimental Pathology, 7(6), 3018–3027.  Konishi, Y., Ikeda, K., Iwakura, Y., & Kawakami, K. (2006). Six1 and Six4 promote survival of sensory neurons during early trigeminal gangliogenesis. Brain Research, 1116(1), 93–102.  Krueger, A. B., Dehdashti, S. J., Southall, N., Marugan, J. J., Ferrer, M., Li, X., … Zhao, R. (2013). Identification of a selective small-molecule inhibitor series targeting the eyes absent 2 (Eya2) phosphatase activity. Journal of Biomolecular Screening, 18(1), 85–96.    103 La, L., Sas, L., Sem, S., Mva, D., Rj, A. J., Ls, V., & D, M. (2015). Annals of clinical pathology significance of a biomarkers immunohistochemistry panel for survival prognostic in patients with sporadic colorectal cancer. Ann Clin Pathol, 3(2).  Laclef, C., Hamard, G., Demignon, J., Souil, E., Houbron, C., & Maire, P. (2003). Altered myogenesis in Six1-deficient mice. Development, 130(10), 2239.  Laclef, C., Souil, E., Demignon, J., & Maire, P. (2003). Thymus, kidney and craniofacial abnormalities in Six1 deficient mice. Mechanisms of Development, 120(6), 669–679. Lau, E., & Ronai, Z. A. (2012). ATF2 – at the crossroad of nuclear and cytosolic functions. Journal of Cell Science, 125(12), 2815–2824.  Le Grand, F., Grifone, R., Mourikis, P., Houbron, C., Gigaud, C., Pujol, J., … Maire, P. (2012). Six1 regulates stem cell repair potential and self-renewal during skeletal muscle regeneration. The Journal of Cell Biology, 198(5), 815–832.  Lebbink, R. J., Lowe, M., Chan, T., Khine, H., Wang, X., & McManus, M. T. (2011). Polymerase II promoter strength determines efficacy of microRNA adapted shRNAs. PloS One, 6(10), e26213.  Lerbs, T., Bisht, S., Scholch, S., Pecqueux, M., Kristiansen, G., Schneider, M., … Kahlert, C. (2017). Inhibition of Six1 affects tumour invasion and the expression of cancer stem cell markers in pancreatic cancer. BMC Cancer, 17(1), 249.  Levine, D., & Fisher, D. E. (2014). Current status of diagnostic and prognostic markers in melanoma. Methods Mol Biol, 1102, 177-97.  Li, X., Oghi, K. A., Zhang, J., Krones, A., Bush, K. T., Glass, C. K., … Rosenfeld, M. G. (2003). Eya protein phosphatase activity regulates Six1-Dach-Eya transcriptional effects in mammalian organogenesis. Nature, 426(6964), 247–254.  Li, Z., Tian, T., Hu, X., Zhang, X., Li, L., Nan, F., … Zhang, M. (2014). Targeting Six1 by lentivirus-mediated RNA interference inhibits colorectal cancer cell growth and invasion. International Journal of Clinical and Experimental Pathology, 7(2), 631–639.  Li, Z., Tian, T., Lv, F., Chang, Y., Wang, X., Zhang, L., … Zhang, M. (2013). Six1 promotes proliferation of pancreatic cancer cells via upregulation of cyclin D1 expression. PLOS ONE, 8(3), e59203-.  Liang, C.-C., Park, A. Y., & Guan, J.-L. (2007). In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat. Protocols, 2(2), 329–333.  Liberali, P., Snijder, B., & Pelkmans, L. (2015). Single-cell and multivariate approaches in genetic perturbation screens. Nat Rev Genet, 16(1), 18–32.    104 Liu, C., Liu, Z., Chen, T., Zeng, W., Guo, Y., & Huang, T. (2016). TERT promoter mutation and its association with clinicopathological features and prognosis of papillary thyroid cancer: a meta-analysis, 6, 36990.  Liu, Y., Han, N., Zhou, S., Zhou, R., Yuan, X., Xu, H., … Wu, K. (2016). The DACH/EYA/SIX gene network and its role in tumor initiation and progression. International Journal of Cancer, 138(5), 1067–1075.  Lovly, C., Pao, W., Sosman, J. (2016). Molecular profiling of melanoma. My cancer genome. Retrieved from Lovly, C., Horn, L., & Pao, W. (2016). Molecular profiling of lung cancer - My cancer genome. Retrieved June 20, 2017, from Lv, H., Cui, A., Sun, F., Zhang, Y., Li, Y., Li, L., & Lin, Z. (2014). Sineoculis homeobox homolog 1 protein as an independent biomarker for gastric adenocarcinoma. Experimental and Molecular Pathology, 97(1), 74–80.  Macielag, M. J. (2012). Chemical properties of antimicrobials and their uniqueness. In Antibiotic Discovery and Development (pp. 793–820). Boston, MA: Springer US.  Massagué, J. (2008). TGFβ in cancer. Cell, 134(2), 215–230.  Masuda, T., Wan, J., Yerrabelli, A., Berlinicke, C., Kallman, A., Qian, J., & Zack, D. J. (2016). Off target, but sequence-specific, shRNA-associated trans-activation of promoter reporters in transient transfection assays. PloS One, 11(12), e0167867.  McCourt, C., Dolan, O., & Gormley, G. (2014). Malignant melanoma: a pictorial review. The Ulster Medical Journal, 83(2), 103–110.  McCoy, E. L., Iwanaga, R., Jedlicka, P., Abbey, N.-S., Chodosh, L. A., Heichman, K. A., … Ford, H. L. (2009). Six1 expands the mouse mammary epithelial stem/progenitor cell pool and induces mammary tumors that undergo epithelial-mesenchymal transition. The Journal of Clinical Investigation, 119(9), 2663–2677.  Meyle, K. D., & Guldberg, P. (2009). Genetic risk factors for melanoma. Human Genetics, 126(4), 499–510.  Micalizzi, D. S., Christensen, K. L., Jedlicka, P., Coletta, R. D., Barón, A. E., Harrell, J. C., … Ford, H. L. (2009). The Six1 homeoprotein induces human mammary carcinoma cells to undergo epithelial-mesenchymal transition and metastasis in mice through increasing TGF-β signaling. The Journal of Clinical Investigation, 119(9), 2678–2690.  Michaloglou, C., Vredeveld, L. C. W., Mooi, W. J., & Peeper, D. S. (2008). BRAFE600 in benign and malignant human tumours. Oncogene, 27(7), 877–895.    105 Miller, A. J., & Mihm, M. C. (2006). Melanoma. New England Journal of Medicine, 355(1), 51–65.  Mockenhaupt, S., Grosse, S., Rupp, D., Bartenschlager, R., & Grimm, D. (2015). Alleviation of off-target effects from vector-encoded shRNAs via codelivered RNA decoys. Proceedings of the National Academy of Sciences of the United States of America, 112(30), E4007-16.  Moellering, R. E., Cornejo, M., Davis, T. N., Del Bianco, C., Aster, J. C., Blacklow, S. C., … Bradner, J. E. (2009). Direct inhibition of the NOTCH transcription factor complex. Nature, 462(7270), 182–188.  Moore, C. B., Guthrie, E. H., Huang, M. T.-H., & Taxman, D. J. (2010). Short hairpin RNA (shRNA): design, delivery, and assessment of gene knockdown. Methods in Molecular Biology (Clifton, N.J.), 629, 141–58.  Morgens, D. W., Deans, R. M., Li, A., & Bassik, M. C. (2016). Systematic comparison of CRISPR/Cas9 and RNAi screens for essential genes. Nature Biotechnology, 34(6), 634–636. Nagel, S., Meyer, C., Kaufmann, M., Drexler, H. G., & MacLeod, R. A. F. (2015). Aberrant expression of homeobox gene SIX1 in Hodgkin lymphoma. Oncotarget, 6(37), 40112–40126.  Nagore, E., Heidenreich, B., Rachakonda, S., Garcia-Casado, Z., Requena, C., Soriano, V., … Kumar, R. (2016). TERT promoter mutations in melanoma survival. International Journal of Cancer, 139(1), 75–84.  Ng, K. T., Man, K., Sun, C. K., Lee, T. K., Poon, R. T., Lo, C.-M., & Fan, S.-T. (2006). Clinicopathological significance of homeoprotein Six1 in hepatocellular carcinoma. British Journal of Cancer, 95(8), 1050–1055.  Ohto, H., Kamada, S., Tago, K., Tominaga, S.-I., Ozaki, H., Sato, S., & Kawakami, K. (1999). Cooperation of Six and Eya in activation of their target genes through nuclear translocation of Eya. Molecular and Cellular Biology, 19(10), 6815–6824.  Ono, H., Imoto, I., Kozaki, K., Tsuda, H., Matsui, T., Kurasawa, Y., … Inazawa, J. (2012). SIX1 promotes epithelial-mesenchymal transition in colorectal cancer through ZEB1 activation. Oncogene, 31(47), 4923–4934.  Ossio, R., Roldan-Marin, R., Martinez-Said, H., Adams, D. J., & Robles-Espinoza, C. D. (2017). Melanoma: a global perspective. Nat Rev Cancer, 17(7), 393–394.  Ozaki, H., Nakamura, K., Funahashi, J., Ikeda, K., Yamada, G., Tokano, H., … Kawakami, K. (2004). Six1 controls patterning of the mouse otic vesicle. Development, 131(3), 551.    106 Pagliarini, R., Shao, W., & Sellers, W. R. (2015). Oncogene addiction: pathways of therapeutic response, resistance, and road maps toward a cure. EMBO Reports, 16(3), 280–96.  Palmer, S. R., Erickson, L. A., Ichetovkin, I., Knauer, D. J., & Markovic, S. N. (2011). Circulating serologic and molecular biomarkers in malignant melanoma. Mayo Clinic Proceedings, 86(10), 981–990.  Patrick, A. N., Cabrera, J. H., Smith, A. L., Chen, X. S., Ford, H. L., & Zhao, R. (2013). Structure-function analyses of the human SIX1–EYA2 complex reveal insights into metastasis and BOR syndrome. Nature Structural & Molecular Biology, 20(4), 447–453.  Pearlman, R. L., Montes de Oca, M. K., Pal, H. C., & Afaq, F. (2017). Potential therapeutic targets of epithelial-mesenchymal transition in melanoma. Cancer Letters, 391, 125–140.  Ramdas, W. D., van Koolwijk, L. M. E., Ikram, M. K., Jansonius, N. M., de Jong, P. T. V. M., Bergen, A. A. B., … van Duijn, C. M. (2010). A genome-wide association study of optic disc parameters. PLOS Genetics, 6(6), e1000978-.  Rao, D. D., Vorhies, J. S., Senzer, N., & Nemunaitis, J. (2009). siRNA vs. shRNA: Similarities and differences. Advanced Drug Delivery Reviews, 61(9), 746–759.  Reed, K. B., Brewer, J. D., Lohse, C. M., Bringe, K. E., Pruitt, C. N., & Gibson, L. E. (2012). Increasing incidence of melanoma among young adults: An epidemiological study in Olmsted County, Minnesota. Mayo Clinic Proceedings, 87(4), 328–334.  Reichenberger, K. J., Coletta, R. D., Schulte, A. P., Varella-Garcia, M., & Ford, H. L. (2005). Gene amplification is a mechanism of Six1 overexpression in breast cancer. Cancer Research, 65(7), 2668.  Remmele, W., & Stegner, H. E. (1987). Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue. Der Pathologe, 8(3), 138–40.  Rigel, D. S., Robinson, J. K., Ross, M., Friedman, R. J., Cockerell, C. J., Lim, H. W., … Kirkwood, J. M. (2011). Cancer of the skin. (John M Kirkwood, Ed.) (Second). Elsevier Saunders. Robert, C., Karaszewska, B., Schachter, J., Rutkowski, P., Mackiewicz, A., Stroiakovski, D., … Schadendorf, D. (2015). Improved overall survival in melanoma with combined dabrafenib and trametinib. New England Journal of Medicine, 372(1), 30–39.  Rodríguez Soriano, J. (n.d.). Branchio-oto-renal syndrome. Journal of Nephrology, 16(4), 603–5.    107 Scott, D. A., Richardson, A. D., Filipp, F. V, Knutzen, C. A., Chiang, G. G., Ronai, Z. A., … Smith, J. W. (2011). Comparative metabolic flux profiling of melanoma cell lines: beyond the Warburg effect. The Journal of Biological Chemistry, 286(49), 42626–42634.  Shain, A. H., Yeh, I., Kovalyshyn, I., Sriharan, A., Talevich, E., Gagnon, A., … Bastian, B. C. (2015). The genetic evolution of melanoma from precursor lesions. New England Journal of Medicine, 373(20), 1926–1936.  Shain, A. H., & Bastian, B. C. (2016). From melanocytes to melanomas. Nature Reviews Cancer, 16(6), 345–358.  Sharma, S., Kelly, T. K., & Jones, P. A. (2010). Epigenetics in cancer. Carcinogenesis, 31(1), 27–36.  Siegel, R. L., Miller, K. D., & Jemal, A. (2016). Cancer statistics, 2016. CA: A Cancer Journal for Clinicians, 66(1), 7–30.  Silva, M. P., Soave, D. F., Ribeiro-Silva, A., & Poletti, M. E. (2012). Trace elements as tumor biomarkers and prognostic factors in breast cancer: a study through energy dispersive x-ray fluorescence. BMC Research Notes, 5(1), 194.  Song, H.-W., Bettegowda, A., Oliver, D., Yan, W., Phan, M. H., de Rooij, D. G., … Wilkinson, M. F. (2015). shRNA off-target effects in vivo: impaired endogenous siRNA expression and spermatogenic defects. PLOS ONE, 10(3), e0118549.  Szymański, P., Markowicz, M., & Mikiciuk-Olasik, E. (2012). Adaptation of high-throughput screening in drug discovery—toxicological screening tests. International Journal of Molecular Sciences, 13(1), 427–452.  Takahashi, Y., Forrest, A. R. R., Maeno, E., Hashimoto, T., Daub, C. O., & Yasuda, J. (2009). MiR-107 and miR-185 can induce cell cycle arrest in human non small cell lung cancer cell lines. PLOS ONE, 4(8), e6677-.  Tan, W. J., Thike, A. A., Bay, B. H., & Tan, P. H. (2014). Immunohistochemical expression of homeoproteins Six1 and Pax3 in breast phyllodes tumours correlates with histological grade and clinical outcome. Histopathology, 64(6), 807–817.  Taylor, R. C., Cullen, S. P., & Martin, S. J. (2008). Apoptosis: controlled demolition at the cellular level. Nature Reviews Molecular Cell Biology, 9(3), 231–241.  Towers, C. G., Guarnieri, A. L., Micalizzi, D. S., Harrell, J. C., Gillen, A. E., Kim, J., … Ford, H. L. (2015). The Six1 oncoprotein downregulates p53 via concomitant regulation of RPL26 and microRNA-27a-3p, 6, 10077.    108 Vadlamudi, R. K., Manavathi, B., Balasenthil, S., Nair, S. S., Yang, Z., Sahin, A. A., & Kumar, R. (2005). Functional implications of altered subcellular localization of PELP1 in breast cancer cells. Cancer Research, 65(17), 7724.  Veronesi, U., Galimberti, V., Zurrida, S., Merson, M., Greco, M., Luini, A., … Dick, R. (1993). Prognostic significance of number and level of axillary node metastases in breast cancer. The Breast, 2(4), 224–228.  Vogelstein, B., Papadopoulos, N., Velculescu, V. E., Zhou, S., Diaz, L. A., Kinzler, K. W., & Kinzler, K. W. (2013). Cancer genome landscapes. Science (New York, N.Y.), 339(6127), 1546–58.  Volpe, V. O., Klufas, D. M., Hegde, U., & Grant-Kels, J. M. (2017). The new paradigm of systemic therapies for metastatic melanoma. Journal of the American Academy of Dermatology, 77(2), 356–368. Weatherhead, S. C., Haniffa, M., & Lawrence, C. M. (2007). Melanomas arising from naevi and de novo melanomas — does origin matter? British Journal of Dermatology, 156(1), 72–76.  Wu, W., Huang, R., Wu, Q., Li, P., Chen, J., Li, B., & Liu, H. (2014). The role of Six1 in the genesis of muscle cell and skeletal muscle development. International Journal of Biological Sciences, 10(9), 983–9.  Wu, W., Ren, Z., Chen, C., Liu, Y., Zhang, L., Chao, Z., … Xiong, Y. (2012). Subcellular localization of different regions of porcine Six1 gene and its expression analysis in C2C12 myoblasts. Molecular Biology Reports, 39(12), 9995–10002.  Wu, W., Ren, Z., Li, P., Yu, D., Chen, J., Huang, R., & Liu, H. (2015). Six1: A critical transcription factor in tumorigenesis. International Journal of Cancer, 136(6), 1245–1253.  Xie, R., Chung, J.-Y., Ylaya, K., Williams, R. L., Guerrero, N., Nakatsuka, N., … Hewitt, S. M. (2011). Factors influencing the degradation of archival formalin-fixed paraffin-embedded tissue sections. Journal of Histochemistry and Cytochemistry, 59(4), 356–365.  Xu, P.-X. (2013). The EYA-SO/SIX complex in development and disease. Pediatric Nephrology, 28(6), 843–854. Xu, P.-X., Zheng, W., Huang, L., Maire, P., Laclef, C., & Silvius, D. (2003). Six1 is required for the early organogenesis of mammalian kidney. Development (Cambridge, England), 130(14), 3085–3094.  Yeh, J., Toniolo, P., & Frank, D. (2013). Targeting transcription factors: promising new strategies for cancer therapy. Curr Opin Oncol, 25(6), 652–8.    109 Yu, J.-Y., DeRuiter, S. L., & Turner, D. L. (2002). RNA interference by expression of short-interfering RNAs and hairpin RNAs in mammalian cells. Proceedings of the National Academy of Sciences , 99(9), 6047–6052.  Yu, Y., Davicioni, E., Triche, T. J., & Merlino, G. (2006). The homeoprotein Six1 transcriptionally activates multiple protumorigenic genes but requires Ezrin to promote metastasis. Cancer Research, 66(4), 1982.  Zbytek, B., Carlson, J. A., Granese, J., Ross, J., Mihm, M. C., Slominski, A., & Slominski, A. (2008). Current concepts of metastasis in melanoma. Expert Review of Dermatology, 3(5), 569–585.  Zeng, J., Shi, R., Cai, C., Liu, X., Song, Y., Wei, M., & Ma, W. (2015). Increased expression of Six1 correlates with progression and prognosis of prostate cancer. Cancer Cell International, 15, 63.  Zeng, J., Shi, R., Cai, C., Liu, X., Song, Y., Wei, M., & Ma, W. (2015). Increased expression of Six1 correlates with progression and prognosis of prostate cancer. Cancer Cell International, 15, 63.  Zheng, H., Xia, P., Xu, X., Takahashi, H., & Takano, Y. (2017). The nuclear to cytoplasmic shift of ING5 protein during colorectal carcinogenesis with their distinct links to pathologic behaviors of carcinomas. Human Pathology, 42(3), 424–433.  Zheng, W., Huang, L., Wei, Z.-B., Silvius, D., Tang, B., & Xu, P.-X. (2003). The role of Six1 in mammalian auditory system development. Development, 130(17), 3989.      


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items