UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of the membrane scaffolding protein KAI1 in human cutaneous melanoma Tang, Yun 2016

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

Item Metadata

Download

Media
24-ubc_2016_may_tang_yun.pdf [ 4.23MB ]
Metadata
JSON: 24-1.0224025.json
JSON-LD: 24-1.0224025-ld.json
RDF/XML (Pretty): 24-1.0224025-rdf.xml
RDF/JSON: 24-1.0224025-rdf.json
Turtle: 24-1.0224025-turtle.txt
N-Triples: 24-1.0224025-rdf-ntriples.txt
Original Record: 24-1.0224025-source.json
Full Text
24-1.0224025-fulltext.txt
Citation
24-1.0224025.ris

Full Text

THE ROLE OF THE MEMBRANE SCAFFOLDING PROTEIN KAI1 IN HUMAN CUTANEOUS MELANOMA  by  Yun Tang  BSc. The University of British Columbia, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Experimental Medicine)   THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)    February 2016  © Yun Tang, 2016  ii Abstract Cutaneous melanoma remains to be not only one of the most deadly among all skin cancers, but it’s one of the most deadly of all cancers in general. It is crucial to have an early detection of the disease despite lacking any effective treating options while therapeutic strategies of later stages of melanoma have yet to be discovered. Meanwhile, the mechanism modulating the progression of melanoma is still not well understood. In this study, we investigated KAI1’s role during metastasis regulation in human melanoma. We proposed the tumor suppressor function of KAI1 was directly correlated with KAI1 expression and showed the loss of KAI1 expression in melanoma patient samples significantly correlated with poorer patient survival. Furthermore, forced KAI1 expression was shown to suppress melanoma cell migration and invasion primarily through its regulation of another tumor suppressor gene: inhibitor of growth 4 (ING4). Moreover, KAI1 expression significantly suppressed melanoma angiogenesis by reducing HUVEC cell growth and tubular structure formation. In fact, KAI1’s regulation on angiogenesis was associated with the modulation of IL-6 and VEGF expression. Additionally, we investigated the mechanistic pathway between KAI1 and ING4 and found that KAI1 suppressed Akt phosphorylation through the regulation of EGFR and VEGFR phosphorylation. Meanwhile, the semaphorin 3C (SEMA3C) protein had been identified as an oncogene that induced cancer cell migration and invasion. In this study, we found that SEMA3C was also able to induce melanoma angiogenesis observed in the elevated HUVEC growth and tube formation. Furthermore, we showed that KAI1 expression suppressed SEMA3C-induced melanoma angiogenesis whereas KAI1 knockdown rescued the SEAM3C-suppressed melanoma angiogenesis.   iii According to our study, we have illustrated a regulatory pathway of KAI1 on the regulation of melanoma metastasis which involves the regulation of the PI3K/Akt pathway and the tumor suppressor gene ING4. Also, the restoration of KAI1 expression was shown to significantly suppress melanoma cancer cell migration, invasion and angiogenesis. Taken together, KAI1 was a potential diagnostic marker for advanced melanoma and the restoration of KAI1 expression might shed light on new therapeutic approaches for treating cutaneous human melanoma.   iv Preface Contributions 1. A version of chapter 3 has been published [Tang Y, Cheng Y, Martinka M, Ong, C, Li G. Prognostic Significance of KAI1/CD82 in Human Melanoma and Its Role in Cell Migration and Invasion through the Regulation of ING4. Carcinogenesis. 2014 Jan]. I was the major investigator, designed and performed most of the experiments, and prepared the manuscript. Dr. G. Li provided facilities, research materials, and contributed to experimental design and manuscript preparation. Dr. Y. Cheng and Dr. C. Ong contributed in the revision of the manuscript. Dr. M. Martinka assisted with the scoring of tissue microarray. 2. A version of chapter 4 has been published [Tang Y, Madhuri B, Cheng Y, Lu J, Li G, Ong C. The Role of the Metastasis Suppressor gene KAI1 in Melanoma Angiogenesis. Pigment Cell & Melanoma Research. 2015 Jul]. I was the major investigator, designed and performed most of the experiments, and prepared the manuscript. Dr. G. Li and Dr. Ong C provided facilities, research materials, and contributed to experimental design and manuscript preparation. Dr. Madhuri B, Dr. Y. Cheng and Dr. J. Lu provided the raw data and contributed in the design of the study. 3. A version of chapter 5 has been prepared [Tang Y, Madhuri B, Cheng Y, Lu J, Ong C. The Role of KAI1 and SEMA3C in the Regulation of VEGFR2 Phosphorylation.]. I was the major investigator, designed and performed most of the experiments, and prepared the manuscript. Dr. G. Li and Dr. C. Ong provided facilities, research materials, and contributed to experimental design and manuscript preparation. Dr. Madhuri B, Dr. Y.  v Cheng and Dr. J. Lu provided some of the raw data and contributed in the design of the study.   List of publications: 1. Tang Y, Cheng Y, Martinka M, Ong CJ, Li G (2014). Prognostic significance of KAI1/CD82 in human melanoma and its role in cell migration and invasion through the regulation of ING4. Carcinogenesis 35: 86-95.   2. Tang Y, Bhandaru M, Cheng Y, Lu J, Li G, Ong CJ (2015). The Role of the Metastasis Suppressor gene KAI1 in Melanoma Angiogenesis. Pigment cell & melanoma research.  3. Tang Y, Bhandaru M, Cheng Y, Lu J, Ong CJ. The Role of KAI1 and SEMA3C in the Regulation of VEGFR2 Phosphorylation. Manuscript is under preparation.   Ethics certificate: The use of human skin tissues in this study was approved by the Clinical Research Ethics Board of University of British Columbia (certificate number is H09-01321).  vi Table of Contents  Abstract .............................................................................................................................. ii Preface ............................................................................................................................... iv Table of Contents .............................................................................................................. vi List of Tables .......................................................................................................................x List of Figures .................................................................................................................... xi List of Abbreviations ....................................................................................................... xiii Acknowledgements ...........................................................................................................xvi Dedication ...................................................................................................................... xviii Chapter 1: Introduction ......................................................................................................1 1.1 Malignant Melanoma ................................................................................................... 1 1.1.1 The Development of Melanoma ............................................................................... 1 1.1.2 Epidemiology of Melanoma ..................................................................................... 3 1.1.3 Etiology of Melanoma .............................................................................................. 4 1.1.4 Staging and Subtypes of Melanoma .......................................................................... 7 1.1.5 Mutation and Immune Targeted Therapeutic Strategies ............................................ 8 1.2 Cancer Migration and Invasion ................................................................................... 12 1.2.1 Cell Migration Mechanism ..................................................................................... 12 1.2.2 Tumor Invasion Mechanism ................................................................................... 14 1.2.3 Types of Transition ................................................................................................ 15 1.3 Angiogenesis .............................................................................................................. 17 1.3.1 Angiogenesis and Cancer ....................................................................................... 18  vii 1.3.2 Angiogenic Factors ................................................................................................ 19 1.3.3 Anti-angiogenic Therapies ..................................................................................... 20 1.4 The Tetraspanin Superfamily ...................................................................................... 21 1.4.1 Structure of Tetraspanins ........................................................................................ 22 1.4.2 Regulation of Tetraspanin Expression .................................................................... 23 1.4.3 Diverse Functions of Tetraspanin Proteins .............................................................. 24 1.4.4 Protein Interactions of the Tetraspanin Web ........................................................... 25 1.5 The Tumor Suppressor KAI1 ...................................................................................... 27 1.5.1 Gene Construct and Mutations................................................................................ 27 1.5.2 Structure of Tetraspanin KAI1................................................................................ 27 1.5.3 KAI1’s Regulation of Cancer Progression .............................................................. 28 1.5.4 Regulation of KAI1 Expression .............................................................................. 31 1.6 Objective and Hypotheses .......................................................................................... 32 Chapter 2: Material and Methods .................................................................................... 34 2.1 Immunohistochemistry ............................................................................................... 34 2.2 Evaluation of TMA Immunostaining .......................................................................... 34 2.3 Statistical Analyses .................................................................................................... 35 2.4 Cell Lines and Cell Culture ........................................................................................ 35 2.5 Protein Extraction and Western Blot ........................................................................... 36 2.6 ELISA ........................................................................................................................ 36 2.7 Expression Plasmids, siRNA and Transfection............................................................ 37 2.8 Reverse Transcription and Real-time Quantitative Polymerase Chain Reaction ........... 37 2.9 Wound Healing Assay ................................................................................................ 38 2.10 Cell Invasion Assay .................................................................................................... 38 2.11 Zymography ............................................................................................................... 39 2.12 Immunofluorescent Staining ....................................................................................... 39  viii 2.13 Sulphorhodamine B (SRB) Assay ............................................................................... 40 2.14 Tube Formation Assay of HUVECs ............................................................................ 41 2.15 FACS Analysis ........................................................................................................... 41 2.16 In Vivo Matrigel Plug Assay and Immunofluorescent Staining .................................... 41 Chapter 3: Prognostic Significance of KAI1/CD82 in Human Melanoma and Its Role in Cell Migration and Invasion through the Regulation of ING4 ........................................ 43 3.1 Background and Rationale .......................................................................................... 43 3.2 Results ....................................................................................................................... 44 3.2.1 Reduced KAI1 Expression Was Correlated With Melanoma Progression ................ 44 3.2.2 Decreased KAI1 Eexpression Correlated with a Worse Patient Survival ................. 49 3.2.3 KAI1 Expression Reduced Melanoma Cell Migration and Stress Fibre Formation .. 53 3.2.4 KAI1 Regulated Cell Invasion and MMP2 Activity ................................................ 57 3.2.5 KAI1 as an Upstream Regulator of ING4 ............................................................... 59 3.3 Discussion .................................................................................................................. 63 Chapter 4: The Role of the Metastasis Suppressor gene KAI1 in Melanoma Angiogenesis ...................................................................................................................... 68 4.1 Background and Rationale .......................................................................................... 68 4.2 Results ....................................................................................................................... 69 4.2.1 KAI1 Expression in Melanoma Cells Suppressed the Growth and Tubular Structure Formation of HUVECs ....................................................................................................... 69 4.2.2 KAI1 Regulated HUVEC Growth through the Expression of IL-6 and VEGF ......... 76 4.2.3 KAI1 Suppresses Melanoma Angiogenesis through ING4 ...................................... 79 4.2.4 KAI1 Expression in Melanoma Cells Inhibited Angiogenesis in vivo ...................... 83 4.2.5 KAI1 was an Upstream Regulator of a Serine/threonine Kinase Akt ....................... 86 4.3 Discussion .................................................................................................................. 88  ix Chapter 5: The Role of KAI1 and SEMA3C in the Regulation of VEGFR2 Phosphorylation................................................................................................................. 93 5.1 Background and Rationale .......................................................................................... 93 5.2 Results ....................................................................................................................... 94 5.2.1 SEMA3C Expression Induced HUVEC Growth and Tubular Structure Formation through the Regulation of IL-6 and VEGF .......................................................................... 94 5.2.2 SEMA3C Upregulated Melanoma Angiogenesis in vivo ......................................... 99 5.2.3 SEMA3C Expression Upregulated VEGFR2 Phosphorylation .............................. 101 5.2.4 KAI1’s Regulation on SEMA3C in VEGFR2 Phosphorylation, HUVEC Growth and Tube Formation ................................................................................................................ 104 5.3 Discussion ................................................................................................................ 106 Chapter 6: Conclusions ................................................................................................... 110 6.1 Summary of Findings ............................................................................................... 110 6.2 Limitations of The Study and Future Directions ........................................................ 113 Bibliography .................................................................................................................... 116    x List of Tables Table 3.1 KAI1 Staining and Clinicopathologic Characteristics of Melanoma Patients 48 Table 3.2 Univariate Cox Proportional Regression Analysis on 5-year Overall and Disease-Specific Survival of 262 Primary and 155 Metastatic Melanoma Patients ........ 52 Table 3.3 Multivariate Cox Regression Analysis on 5-year Overall and Disease-Specific Survival of all Melanoma Patients .................................................................................... 53     xi List of Figures Figure 3.1 KAI1 expression was reduced in advanced human melanoma ...................... 46 Figure 3.2 KAI1 expression level was downregulated in multiple human melanoma cell lines compare to the melanocyte ....................................................................................... 47 Figure 3.3 KAI1 expression is associated with 5-year survival of melanoma patients ... 50 Figure 3.4 KAI1 regulated melanoma cell migration ....................................................... 55 Figure 3.5 KAI1 regulated stress fiber formation via ROCK.......................................... 56 Figure 3.6 KAI1 regulates melanoma cell invasion and the activity of MMP-2 ............. 58 Figure 3.7 Regulation of ING4 by KAI1 and its effect on melanoma cell migration ...... 61 Figure 4.1 KAI1 expression in melanoma cells inhibited HUVEC growth and tube formation ........................................................................................................................... 71 Figure 4.2 KAI1 expression in Mel624 and A375 cell lines inhibited tube formation .... 73 Figure 4.3 Effect of KAI1 on cell cycle ............................................................................. 75 Figure 4.4 KAI1 expression in melanoma cells suppressed IL-6 and VEGF .................. 78 Figure 4.5 KAI1 suppressed HUVEC growth and tube formation through the regulation of ING4............................................................................................................. 81 Figure 4.6 KAI1 expression in melanoma cells suppressed IL-6 and VEGF through the regulation of ING4............................................................................................................. 82 Figure 4.7 Expression of KAI1 and ING4 mRNA in matrigel plugs was determined by qRT-PCR ........................................................................................................................... 84 Figure 4.8 Forced KAI1 expression in MMRU melanoma cell line inhibited blood vessel formation in vivo through the regulation of ING4 ........................................................... 85 Figure 4.9 KAI1 overexpression suppressed Akt phosphorylation ................................. 87  xii Figure 5.1 SEMA3C expression in melanoma cells induced HUVEC growth and tube formation ........................................................................................................................... 96 Figure 5.2 SEMA3C expression in melanoma cells upregulated IL-6 and VEGF .......... 98 Figure 5.3 Forced SEMA3C expression in MMRU melanoma cell line induced blood vessel formation in vivo ................................................................................................... 100 Figure 5.4 SEMA3C induced while KAI1 suppressed VEGFR2 phosphorylation ....... 103 Figure 5.5 KAI1 regulated SEMA3C-induced VEGFR2 phosphorylation and HUVEC growth .............................................................................................................................. 105     xiii List of Abbreviations Abbreviations Definition   Akt  Thymoma viral proto-oncogene/ protein kinase B ALK Anaplastic lymphoma kinase ALM Acral lentiginous melanoma AML Acute myeloid leukemia  AP-2 Activating protein 2 APAF Apoptosis protease activating factor APC Anaphase-promoting complex/cyclosome ARF Alternate open reading frame ARID1 AT-rich interactive domain-containing protein 1 Arp2/3 Actin-related proteins 2/3 ATF Activating transcription factor 2 Bak Bcl-2 homologous antagonist/killer Bax Bcl-2 -associated X protein Bcl B-cell lymphoma Bcl-xL B-cell lymphoma-extra large bFGF Basic fibroblast growth factor BMI Polycomb ring finger oncogene BMP Bone morphogenesis protein BRAF v-Raf murine sarcoma viral oncogene homolog B1 BRMS1 Breast cancer metastasis-suppressor 1 CAN Common Acquired Nevi CD Cluster of Differentiation Cdc42 Cell division control protein 42 homolog CDK Cyclin-dependent kinase CDK4 Cyclin-dependent kinase 4 CDKN2A Cyclin-dependent kinase inhibitor 2A CEACAM-1 Carcinoembryonic antigen-related cell adhesion molecule 1 COX-2 Cyclooxygenase-2 CPD Cdc phospho-degrons CTLA-4 Cytotoxic T-lymphocyte-associated antigen 4  CXCR4 C-X-C chemokine receptor type 4 CYLD Cylindromatosis DSS Disease-specific survival E1A Early region 1A ECM Extracellular matrix EGF Epidermal growth factor EGFR Epidermal growth factor receptor EML4 Echinoderm microtubule-associated protein-like 4  EMT Epithelial-to-mesenchymal transition   xiv ERK Extracellular signal-regulated kinase ETV1 ETS translocation variant 1 EZH2 Histone-lysine N-methyltransferase FAK Focal adhesion kinase FOXP3 Forkhead box p3 GAP GTPases activating protein GDP Guanosine diphosphate GEF Guanine-nucleotide exchange factor GNAQ Guanine nucleotide-binding protein G(q) subunit alpha GSK3 Glycogen synthase kinase 3 GTP Guanosine-5'-triphosphate HER2 Epidermal growth factor receptor 2 HIF1 Hypoxia-inducible factor 1 HLA Human leukocyte antigens HMGA2 High-mobility group AT-hook 2 hTERT Human telomerase reverse transcriptase IC50 Half maximal inhibitory concentration ICAM1 Intercellular adhesion molecule 1 ID Inhibitor of DNA-binding 1 IDH1 Isocitrate dehydrogenase 1 (NADP+) soluble IDO Indoleamine 2,3-dioxygenase IHC Immunohistochemistry IRF4 Interferon regulatory factor 4 IRS Immunoreactive score JNK Jun N-terminus kinase  KLF Krüppel-like family L1CAM L1 cell adhesion molecule LDH Lactate dehydrogenase LEL Large extracellular loop LMM lentigo meligna melanoma LRR Leucine-rich repeats MAGE Melanoma-associated antigen 3 MAPK Mitogen-activated protein kinase MART-1 Prostate-specific antigen MC1R Melanocortin-1 receptor MCAM Melanoma cell adhesion molecule MDM2 Mouse double minute 2 homolog  MEK (MKK) Mitogen-activated protein kinase kinase MET Mesenchymal–epithelial transition MHC Major histocompatibility complex MIA Melanoma-inhibiting activity MITF Microphthalmia-associated transcription factor MLC Myosin light chain  xv MLCK Myosin light-chain kinase MLCP Myosin-light-chain phosphatase MMP Matrix metalloproteinase mTOR Mammalian target of rapamycin NFκB Nuclear factor kappa B NM Nodular melanoma Notch1-IC Notch1 intracellular domain NRAS Neuroblastoma RAS viral oncogene homolog OS Overall survival PAK1 p21 protein-activated kinase 1 PCNA Porliferating cell nuclear antigen PD1 programmed cell death 1  PFS Progression-free survival PI3K Phosphoinositide 3-kinase PPP6C Serine/threonine-protein phosphatase 6 catalytic subunit PTEN Phosphatidylinositol-3,4,5-trisphosphate 3-phosphatase Rb Retinoblastoma protein Rho Ras homolog gene family ROCK Rho-associated, coiled-coil containing protein kinase 1 RUNX3 Runt-related transcription factor 3 SCF Skip1-Cul1-F-box protein SEL Small Extracellular Loop SEMA Semaphorin SGK1 Serum- and glucocorticoid-inducible kinase 1 Skp2 S-phase kinase-associated protein 2 SPARC Secreted protein acidic and rich in cysteine SRB Sulphorhodamine B Src Proto-oncogene tyrosine-protein kinase Src SREBP1 Sterol regulatory element-binding transcription factor 1 SSM Superficial spreading melanoma STAT Signal transducer and activator of transcription TMA Tissue microarray TGFβ Transforming growth factor-β TGIF1 TGFβ-induced factor 1  TLR4 Toll-like receptor 4 TM Transmembrane domains UV Ultraviolet  VEGF Vascular endothelial growth factor WASP/WAVE Wiskott-Aldrich syndrome protein WD40 Tryptophan-aspartic acid depeptide 40 WNT Wingless-type MMTV integration site family  xvi Acknowledgements I would like to first show my gratitude to my supervisors Dr. Christopher Ong and Dr. Gang Li who have encouraged and supported me throughout my Ph.D. study. Dr. Ong and Dr. Li both inspired me to develop critical analytic skills and helped me gain expertise and professional standpoints in the research field. I want to especially express my appreciation to Dr. Ong for his tremendous support and help on my manuscript preparation and submission as well as on the completion of my PhD thesis. I would also like to thank my supervisory committee Dr. Vincent Duronio, Dr. Youwen Zhou, and Dr. Williams Jia for their support and constructive input throughout my graduate study.  I want to specially thank Dr. Magdalena Martinka, our faithful dermatopathologist at Vancouver General Hospital, for her support and tremendous input during the evaluation of the issue microarrays.  She has supplied us with all of the patient biopsy samples and helped us review all our tissue microarray staining. She has also answered countless questions and provided us with valuable comments and advice for our projects.  I would also like to thank Dr. Yabin Cheng, Dr. Ronald Wong, Dr. Jing Lu, Dr. Venus Guo, Dr. Mehdi Jafarnejad, Dr. Rotte Anand, and Dr. Bhandaru Madhuri for their kind help and selfless contributions through my graduate study, as well as the assistance and companionship provided from Reza Safaee, Shahram Khosravi, Cecilia Sjoestroem, and Alan Yip. I would like to thank all the colleagues and volunteers at the Jack Bell Research Centre and Research Pavilion of Vancouver General Hospital for the friendly support and assistance.  I would also like to show my gratefulness to Dr. Harvey Lui, Leanne Li, Karen Ng and other staff from the Department of Dermatology and Skin Sciences of University of British Columbia. Also, I would like to thank Dr. Vincent Duronio and Cornelia  xvii Reichelsdorfer from the Experimental Medicine Program for providing such a wonderful program at the UBC as well as the immense support provided to me during my graduate years at the University. I am honored to have been the recipient of the NSERC graduate student award funded by the Natural Sciences and Engineering Research Council of Canada. This work was also generously supported by grants from Canadian Institutes of Health Research, Canadian Cancer Society Research Institute (CCSRI), and Canadian Dermatology Foundation to Dr. G. Li and Dr. Y. Zhou, as well as the grants awarded to Dr. C. Ong from the Terry Fox Research Institute, Canadian Research Society, the NIH Pacific Northwest Prostate SPORE, Prostate Cancer Canada, and National Centers of Excellence of Canada.  xviii Dedication This thesis is dedicated to Jacqueline Ren, for her kindness and companionship, for her unconditioned love, and for her endless support throughout the years.    1 Chapter 1: Introduction 1.1 Malignant Melanoma  1.1.1 The Development of Melanoma Lying between the layers of neural and non-neural ectoderm, the neural crest cells are highly migratory embryonic cells that later develop into the melanocytes. (Erickson & Reedy, 1998). Melanocytes reside along the basal layer of the epidermis as well as in the hair bulb, eyes and ears (Worobec & Solomon, 1978). They can produce a colored pigment referred to as melanin with their  organelles referred to as melanosomes (Orlow, 1995). The production of melanin pigmentation is stimulated by keratinocytes that exist within close proximity of the melanocytes. There are on average about 6 times more keratinocytes than melanocytes (Vancoillie et al, 1999). The melanin produced forms a protective layer on the top of the basal cells to decrease the destructive effect caused by ultraviolet radiation that can potentially cause DNA damage and cancer (Hermanns et al, 2000). At the earlier stages of the development of neural crest cells, bone morphogenesis protein (BMP) signaling pathway is crucial, which later leads to the epithelial-mesenchymal transition (EMT) of the neural crest and causes the cells to migrate. Furthermore, Snail/Slug transcription factors are able to suppress E-cadherin expression that assists neural crest cells detachment and migration (Gajavelli et al, 2004). The Notch protein also plays important function during neural crest development where it can induce the growth of neural crest as well as maintaining the survival of mature melanocyte (Nickoloff et al, 2005). In order for the full maturation of melanocytes, neural crest is first developed into melanoblast by the regulation of Wnt  2 signaling. Wnt can also induce the expression of beta-catenin which contributes to the final development of melanocytes (Yamada et al, 2013).  In order to maintain the development of melanocytes, several proteins have been shown to have important roles during melanocyte growth, migration and function. The microphthalmia-associated transcription factor (MITF) is primarily responsible for maintenance of the survival of melanocyte precursor cells melanoblasts (Baxter & Pavan, 2003).  It can also induce the expression of the pro-survival gene B-cell lymphoma 2 (Bcl2) to prevent melanocyte apoptosis as well as upregulate the production of melanin pigmentation (Lekmine & Salti, 2008; Saladi et al, 2013). Furthermore, the receptor tyrosine kinase KIT and its ligand have also been shown to regulate melanocyte where it controls the migration and apoptosis of the melanocytes. A loss of KIT results in a significant loss of melanocyte production and pigmentation (Grichnik et al, 1996). As mentioned earlier, Snail/Slug is able to suppress E-cadherin during early neural crest development. It has been shown to be regulated by Sox10 where an induction of Sox10 causes an increased Slug expression and the development of melanocytes (Murisier et al, 2007).  Genetic mutations that may occur in any of the pathways during melanocyte development, contribute directly to the formation of melanoma. When mutations occur in melanocytes, they can advance and spread, which initially results in the formation of the common acquired nevi or common mole (Clark & Tucker, 1998). At the earlier stages where the tumor is benign, the proliferation of melanocytes is normally confined and restricted to the epidermis and the dermis, which are referred to as a junctional nevus and dermal nevus, respectively (Hurwitz & Buckel, 1997). Further mutations can cause the nevus to progress into a radial growth phase and enter the intraepidermal growth period.  After the radial  3 growth phase, these mutated cells enter the vertical growth phase that is extremely invasive and highly metastatic (Ciarletta et al, 2011; Laga & Murphy, 2010). Though this is the projected sequential process of melanoma progression, not all melanocytes become cancerous in this order.  Some cases of melanoma can progress directly into the malignant stage and this is the reason behind the extremely poor prognosis for malignant  melanoma patients (Clark & Tucker, 1998).  1.1.2 Epidemiology of Melanoma With a low occurrence of only 5% among all cutaneous skin tumors, malignant melanoma, however,  is responsible for over 80% of lethal skin cancer with an average 5 year survival of less than 15% (Radovic-Kovacevic et al, 1997). Reports have shown that the occurrence of melanoma has been increasing steadily for the last several decades and it remains to be one of the most lethal malignancies (Siegel et al, 2014). In 2015, there are over one million melanoma patients in the USA; and more than 70,000 new cases are expected to be diagnosed in the upcoming year, of which over 9000 of the patients may die from this disease. It has also been found to be the fifth most common cancer for men and seventh most common cancer in women (American Cancer Society, 2015). Among all the occurrence rates around the globe, Australia has the highest incidence rates whereas the occurrence rate in the Asian population is among the lowest (Iannacone et al, 2015).  There is an equal chance for men and women to obtain this disease, but the patient’s susceptibility to melanoma has been found to increase with age. Also, some typical sites of the disease usually occur near the trunk, extremities, and near the head and neck of the patients (Karakousis & Driscoll, 1995). Melanoma is more common in light skin populations  4 with a significant lower incidence rates found in darker skin type populations (Clairwood et al, 2014). A history of melanoma in the family, an excess exposure of UV radiation, a suppressed immune system, or a previous disease occurrence all contribute to the risk factors that predispose the patient to melanoma (Nagore et al, 2010). With such a low prognosis level, early detection is important as the 5-year survival for patients with confined melanoma is more than 90%. However, the survival rate significantly decreases as disease progresses, of which the survival rate is less than 60% when there is a regional spread and only 15% after metastasis (American Cancer Society, 2015).   1.1.3 Etiology of Melanoma After years of study, numbers of mechanisms and signaling pathways have been identified as the key drivers for the transformation of the normal melanocytes into the invasive melanoma cancer cells. Besides the genetic alterations caused by the environmental effects, inheritance that causes DNA mutation, is also responsible for the suppression of tumor suppressor genes as well as induction of oncogenes. Some well-known tumor suppressors whose functions are lost during melanoma progression have been identified such as the CDKN2A gene as well as PTEN. Conversely, the Ras/Raf/MEK/ERK pathway has been shown to be frequently upregulated. Furthermore, the expression of growth factors such as EGF, IGF, or TGF, which are activators of the ERK pathway by transducing signal through the receptor tyrosine kinases located on the membrane, are also elevated during advanced melanoma (Fecher et al, 2008). Together, these mutational events result in the manifestation of the histological changes at different phases during the development of the disease.  5 It has been reported that mutations in the ERK pathway are found in about half of the melanoma diagnosis (Sumimoto et al, 2006). Mutations in this pathway directly contribute to its activation. In particular, NRAS mutation is found to be associated with 15% of all melanomas and BRAF mutation is found to be in half of all melanomas (Saldanha et al, 2006). Among the various types of  somatic mutations found in BRAF, the most common one that contributes to over 80% of all the mutations is known to be the V600E mutation, which is a substitution mutation of valine with glutamic acid (Kumar et al, 2007). In comparison, the second most observed BRAF mutation is the valine to lysine mutation V600K, which only contributes to 5% of all BRAF mutations (Sosman et al, 2012). The BRAF V600E mutation has been shown to significantly increase the kinase activity of BRAF, yet this mutation is found in many human nevi (Liu et al, 2007). However, the histological resemblance of human melanoma can be observed when the activity of p53 is lost after the BRAF V600E mutation (Potu et al, 2014). Furthermore, additional studies revealing the possible relationship between BRAF V600E and PTEN suggest that PTEN loss is an important factor during BRAF mutation-induced melanoma progression as well as BRAF inhibitor resistance (Paraiso et al, 2011). As mentioned earlier MITF is a transcription factor that is usually overexpressed in melanoma and is responsible for maintaining tumor cell growth and survival.  Recent discoveries have further correlated MITF with BRAF where BRAF is able to significantly upregulate the expression of MITF (Wellbrock et al, 2008).  As a result, the sole activation of the ERK pathway is insufficient to result in the initiation of melanoma. Rather, multiple triggering events amplify the effect and produce the deteriorating result.   6 One of the tumor suppressors whose function is lost in melanoma is the CDKN2A gene, which accounts for a third of the mutations in melanoma. The CDKN2A locus has been identified to encode two tumor suppressor genes referred to as INK4a and ARF (Abdel-Rahman et al, 2011). A suppression of INK4a results in the upregulation of cyclinD and cdk4 which can lead to the uncontrollable progression of cell cycle (Quelle et al, 1995). Furthermore, mutations in cdk4 and cyclinD have also been found to render them insensitive to the suppression by INK4a (Buecher et al, 2010). Meanwhile, though the p53 tumor suppressor has rarely been found during melanoma advancement, the frequent downregulation of ARF has been identified which can repress p53’s function during melanoma prevention (Piepkorn, 2000). Therefore, it is likely that the inactivation of CDKN2A prevents the cell to initiate senescence and stimulates melanoma progression (Baker et al, 2008).   In addition to the mutation of the tumor suppressor CDKN2A, another commonly mutated tumor suppressor is referred to as PTEN. PTEN mutation has been found to occur in about 15% of advanced melanomas. It has been found to be responsible for the regulation of another signaling pathway that is responsible for maintaining cell survival referred to as the phosphoinositide-3-kinase (PI3K)/Akt signaling pathway (Yajima et al, 2012). PI3K activates phosphatidylinositol phosphate (PIP3) which in turn activates Akt through phosphorylation at Ser473. The activated Akt has been found to be responsible for cell cycle progression, cell proliferation, migration, invasion and angiogenesis (Sinnberg et al, 2009). The tumor suppressor function of PTEN is to keep the PIP3 expression levels low in order to suppress the activity of Akt (Stemke-Hale et al, 2008). When Akt is downregulated, the Bcl-2 antagonist of cell death (BAD) protein is no longer suppressed which leads to the  7 restoration of cell apoptosis. Furthermore, cyclinD1 expression is also lost after the inactivation of Akt which leads to the decrease in cell proliferation (Cantley & Neel, 1999). Finally, findings have revealed that when PTEN expression is restored, the ability of the cells to become cancerous is significantly decreased, and the inactivation of Akt also reduces the growth of tumor in melanoma cells (Stahl et al, 2003).   1.1.4 Staging and Subtypes of Melanoma The American Joint Committee on Cancer (AJCC) system has been commonly acknowledged as the standard staging system for melanoma (Balch et al, 2001). Four separate subgroups are used to differentiate individual melanoma patients, namely the evaluation of the primary tumor thickness (T), the incidence of regional lymph nodes metastases (N), as well as the distant metastases (M). The system also includes tumor thickness and ulceration as additional criteria that offer more staging information for primary melanoma (Balch et al, 2001). Later revision of the AJCC system also recommends the addition of cancer cell mitotic activity for the examination of thin melanoma. Also, the sentinel lymph node biopsy (SLNB) is important and beneficial when designing the adjuvant therapy for stage I melanomas (Balch et al, 2009). Furthermore, LDH is an enzyme found in nearly all living organisms, and it acts as an indicator for cellular metabolism. In normal cells, LDH level is usually low. However, its level is elevated in various cancers, and thus it can be used as an indicator for the presence of cancer. While it provides little information on the identity of the cancer, the elevated LDH level in the serum nevertheless provides a good predictor for stage IV melanoma (Palmer et al, 2011).   8 Currently, malignant melanoma has been divided into four separate subtypes, namely the superficial spreading melanoma (SSM), the nodular melanoma (NM), the lentigo maligna melanoma (LMM), and the acral lentiginous melanoma (ALM) (Gray-Schopfer et al, 2007). SSM is the most common melanoma that accounts for 50 to 70% of melanomas. It occurs at various sites throughout the body and is not limited to certain age groups. Pigmentation and irregular borders are what usually observed in this type of melanoma (Gray-Schopfer et al, 2007). NM has been found to be absent from the radial growth phase, and it has been found to associate with around 20% of all melanomas. NM is considered to be very aggressive, and it grows more rapidly along with considerable increment of tumor thickness (Gray-Schopfer et al, 2007). LMM usually occurs at the site of frequent sun-exposure, such as the face and upper extremities. Lentigo maligna is non-invasive but a precursor to melanoma. A 5% chance exists for lentigo maligna to progress into lentigo maligna melanoma. The invasive LMM is usually darkly pigmented with nodule and raised skin patch is observed with brown pigmented lesion with irregular shapes (Gray-Schopfer et al, 2007).  ALM accounts for 5 to 10% of melanomas and is usually found in people with darker skin. It is usually found on the palms, the nail bed, and oral mucosa with dark pigmentation. However, sun exposure is not to be consider as a factor as it is rarely found in Caucasians or in people with lighter skin tone (Gray-Schopfer et al, 2007).     1.1.5 Mutation and Immune Targeted Therapeutic Strategies Since mutations in ERK pathway occur in about a half of all melanomas, it is a very important target for discovering melanoma treatments (Shah & Dronca, 2014). Furthermore, the discovery of BRAF V600E mutation in over 60% of all melanomas also suggests that it  9 can be a very useful therapeutic target (Davies et al, 2002). Since MEK is the downstream target of BRAF, it can also be a important target for developing therapeutic strategy towards treating advanced melanoma (Meier et al, 2005). Asides from targeting ERK pathway, the understanding of the human T-cell immune responses is also important in investigating the pathophysiology of melanoma. During an immune response, the antigen-presenting cells (APCs) are able to target the receptors located at the surface of the T-cells. At the same time, B7 protein expressed by APCs will also bind to CD28 receptor that is located on the T-cell (Denfeld et al, 1995). However, B7 protein on an APC, which in this case can be a cancer cell, is able to bind to the CTLA-4 receptor on the T-cell. Furthermore, programmed cell death 1 ligand (PD-L1) located on the tumor tissue can also bind to another T-cell receptor referred to as PD-1. Both binding interactions lead to the release of an inhibitory signal that causes the inactivation of the T-cells (Duraiswamy et al, 2014). Therefore, targeting the members of the ERK pathway or the T-cell receptors can be very useful in developing new approaches for melanoma treatments. Vemurafenib was the first BRAF inhibitor that was approved by the FDA in 2011. It was developed so that it only targets the V600E mutated form of BRAF mutation that can further suppress the activation of ERK (Bollag et al, 2010). During the clinical trials of vemurafenib, a 50% clinical response rate was observed and the average progression free survival (PFS) was around 5 months with a median overall survival of 13 months. Some adverse effects of the vemurafenib treatments included rashes, photosensitivity, fatigue, arthralgia, nausea, diarrhea, headache, and vomiting (Chapman et al, 2011). Also, it should be noted that in BRAF wild-type tumors and in normal cells, vemurafenib treatment was able to cause an upregulation of the ERK pathway (Joseph et al, 2010). Furthermore, some  10 patients also developed cutaneous squamous-cell carcinoma or keratoacanthoma after receiving the vemurafenib treatment (Chapman et al, 2011).  Another BRAF inhibitor that was approved by the FDA in 2013 was referred to as dabrafenib. It is a reversible ATP inhibitor that targets specifically the BRAF V600E mutation. Besides a shorter half-life than vemurafenib, it is very similar to the vemurafenib with regard to their modes of action and pharmacodynamic effect (Hauschild et al, 2012). Thus, darafenib is considered to be less toxic and more manageable. Dabrafenib treatment has been found to have a response rate similar to that of vemurafenib. The average PFS with the dabrafenib treatment was also about 5 months with a better median overall survival of 18 months. Besides a rarer occurrence of photosensitivity during the treatments, similar number of patients developed cutaneous squamous-cell carcinoma or keratoacanthoma. Other adverse effects included fatigue, headache and arthralgia (Hauschild et al, 2012).  A more effective BRAF inhibitor is currently under development that is also targeting the BRAF V600 mutations. This inhibitor, labeled as LGX818, has been found to have a better inhibitory function on the ERK pathway compared to either vemurafenib or dabrafenib. Moreover, the response rate has been reported to be around 60% (Anforth et al, 2015). Some of the adverse effects observed in vemurafenib or dabrafenib treatments were less severe or unlikely after the LGX818 treatment (Menzies & Long, 2013).  Unfortunately, resistance to BRAF inhibitors will start to develop within 8 months of the treatments. In particular, the elevation in cyclinD1 expression directly contributes to the resistance towards BRAF inhibitors (Sullivan & Flaherty, 2013). Furthermore, PTEN loss has also been shown to cause resistance to BRAF inhibitor treatments (Paraiso et al, 2011). In additional to the use of BRAF inhibitor, the MEK inhibitor has also been shown to be  11 useful in treating metastatic melanoma with the BRAF V600 mutations. The MEK inhibitor, trametinib, is a potent MEK1 and MEK2 inhibitor that has been approved by the FDA on 2013. The overall response rate of the treatment is lower compared to both vemurafenib and dabrafenib at 22% with PFS of 4.8 months (Sausville, 2012). A major difference between trametinib and vemurafenib was that no cutaneous squamous-cell carcinomas were observed during the trametinib clinical trials (Flaherty et al, 2012).  As mentioned earlier, the binding of CTLA-4 with B7 protein initiate an inhibitory signal that deactivates the T-cell immune response. As a result, a monoclonal antibody was designed against the CTLA-4 receptor referred to as Ipilimumab and was approved by the FDA on 2011. According to results of the clinical trials, the ipilimumab treatment has a response rate of 10% with a median overall survival of 10 months (Robert et al, 2014). Meanwhile, another monoclonal antibody referred to as Nivolumab was approved by the FDA on 2014 that was able to target PD-1 receptor of the T-cell. It has been shown to have a better clinical effect and with less toxic effects (Robert et al, 2015).  The challenge remains to come up with the best sequence and combination of treatment options to treat metastatic melanoma. Various combinations of protein targeted inhibitors and immune targeted inhibitors have been proposed, and various clinical trials are underway (Chapman et al, 2014). Furthermore, other therapeutic strategies such as targeting the angiogenesis pathways or the PI3K/Akt pathway are also being developed (Gray-Schopfer et al, 2007). The effectiveness of these new approaches is yet to be investigated; but continuous studies following the etiology of melanoma are destined in the future for the discovery of more novel therapeutic approaches to treat patients with unresectable melanoma.   12 1.2 Cancer Migration and Invasion Migration and invasion occurs all the time in normal cells during various physiological processes such as embryonic morphogenesis and wound healing. This is not that different from the processes that cancer cells utilise to metastasize. Cancer cells adapt the ability to migrate and invade in order to travel to a distant organ and then proliferate (Chambers et al, 2002). For a migrating cell, it is able to modify its shape and rigidity to interact with the extracellular matrix (ECM). The ECM can thus act as the anchor upon which the cells can attach. The contraction of the cell can then pull itself forward. These simple steps are then continuously repeated to result in the forward motion of the migrating cell (Friedl & Brocker, 2000).   1.2.1 Cell Migration Mechanism The first step during cell migration is the formation of actin filaments in the cytoplasm of the cell that can push the membrane outwards. During this process, actin filaments can bind to the actin-nucleating ARP2/3 complex, which then bind to the Wiscott-Aldrich syndrome protein (WASP). This complex can further recruit other actin filaments to create a multi-actin branching network that makes up the leading edge of the migrating cell (Carlier et al, 1999). Previous research found that the WASP complex is activated by the small GTPase of the Rho family such as RAC, Rho and CDC42 (Abe et al, 2003). Meanwhile, PIPs that are generated by PI3K are shown to activate the small GTPase such as CDC42 (Chang et al, 2005).. CDC42 is then attached to the PIPs which in turn bind and activate WASP. Together with ARP2/3 and polymerized actin filament chains, WASP forms a recruitment complex that creates a multiple actin filament network at the protrusion of the  13 migrating cell. The actin filament complex is then recruited by the actin-binding proteins, such as vinculin and paxillin, and binds to the integrins located on the membrane. These integrins can then bind to the ECM ligands and produce multiple focal contacts at the leading edge of the migrating cell (Miyamoto et al, 1995). Furthermore, focal adhesion kinase (FAK) as well as talin and tensin are also required during the recruitment of actin filament complex and the actin-binding proteins to the integrins at the focal contacts (Lawson et al, 2012).   In order for the cell to migrate through the rigid extracellular network, the ECM has to be cleaved by proteases. Surface proteases such as MT1-MMP and MMP1 concentrate at the leading edge and cleave the ECM components such as collagen and fibronectin, as well as pro-MMPs (Visse & Nagase, 2003). When pro-MMPs are cleaved, they become the active soluble MMPS such as MMP2. After the cleavage of ECM components, MMP2 can then access these fragments and further degrade them (Deryugina et al, 1998). As the leading edge of the migrating cell pushes through the degraded ECM, the cell starts to contract in the direction of the migration. This is done through the contraction of actomyosin, which begins with the binding of the active myosin II to the actin filaments (Wilson et al, 2010). The contraction is executed by Ca+2 and calmodulin-dependent myosin light-chain kinase (MLCK) that phosphorylates the myosin light chain (MLC). Conversely, this contraction is suppressed by the myosin light chain phosphatase (MLCP). Meanwhile, Rho-associated protein kinase (ROCK), activated by Rho, is able to suppress the activity of the MLCPase and assist actomyosin activity during cell contraction (Chen et al, 2002). Finally, any focal contacts at the trailing edge of the cell are detached. Sheddase, a membrane-bound enzyme that cleaves extracellular domain of a transmembrane protein, disassembles the extracellular contacts through proteolytic cleavage of adhesion receptors (Moss & Lambert, 2002).  14 Meanwhile, cytoplasmic protease calpain separates the cytoplasmic actin/ARP2/3 complex from the integrins and other adhesion receptors. The disassembled membrane adhesion receptors are then recycled through internalization and transported to the leading edge (Pfaff et al, 1999).    1.2.2 Tumor Invasion Mechanism The early observation of tumor metastasis is the evidence of a localized cancer cell invasion as cancer cells work their way around the ECM network either individually or collectively. The individual tumor cell migration is primarily regulated by integrins and proteases, while collective tumor cell migration usually retains the intracellular junctions and is regulated through cadherins and various adhesion receptors. The epithelial-mesenchymal transition is used to describe the process where normal epithelial cells are converted to cancerous migrating cells (Kopfstein & Christofori, 2006). Two types of invasion mechanisms have been identified as the mesenchymal and amoeboid modes of migration. And these two modes of migration are interchangeable depending on the environmental stimulus.  The mesenchymal mode of migration is characterised by the protrusion of actin filaments at the leading edge of the cell, as well as the establishment of focal adhesion and proteolytic remodelling of the extracellular matrix. The mesenchymal cells migrate in the extracellular space following the five-step migration cycle discussed earlier. They are normally found in the cancers such as fibrosarcomas or gliomas where the cells are primarily a part of the connective tissue (Young et al, 1995). The morphology of the cells that invade according to the mesenchymal mode of migration is usually characterized by having shape of  15 fibroblast-like spindles. Ttheir migrating speed is relatively slow, which is about 1um/min (Friedl et al, 1998). Furthermore, their adhesion and migration phenotypes are closely regulated by integrins for focal contacts, as well as MMPs and other proteases for the degradation of the ECM (d'Ortho et al, 1998; Muller et al, 1997).  Finally, the mesenchymal migration depends on the actomyosin-mediated contraction, which is regulated by Rho, ROCK and MLCK (Zondag et al, 2000). Therefore, by suppressing the activity of integrins, Rho, and MLCK, the mesenchymal mode of migration can be weakened. The amoeboid cells were characterized as roundish or ellipsoid in shape (Enterline & Coman, 1950). Their migration behaviours are independent of the adhesive mechanism as opposed to the observations made in mesenchymal cells. Their dependence on integrins and proteases is also low, since focal contacts are dispensable during amoeboid migration (Schienbein & Gruler, 1995; Yumura et al, 1984). Thus, amoeboid migration is mainly regulated by the Rho/ROCK signalling pathway which is responsible for the actomyosin contractility (Wyckoff et al, 2006). The amoeboid movement is characterized as a propulsion mechanism where the rounded, blebby cells are pushed in the forward direction with little to no adhesion to the substrates. These cells usually have a higher migration speed that can be as fast as 20μm/min (Fackler & Grosse, 2008). Some types of cancer cells, such as those found in lymphoma, small-cell lung carcinoma or small-cell prostate cancer, retain the amoeboid movement (Carragher et al, 2006; Kim et al, 2014; Wolf et al, 2003b).   1.2.3 Types of Transition In order for the primary tumor cells to break away from the epithelial layer and gain invasive properties, they have to go through the process referred to as the epithelial- 16 mensenchymal transition (EMT). It is the most important process for tumor cells to develop invasiveness characteristics (Chaffer & Weinberg, 2011). This transition causes the cancer cells to become detached and dispersed and results in the loss of cell-to-cell junctions and adhesions (Thiery, 2002).  The loss of expression of the tumor suppressor gene E-cadherin, which is responsible for cell-cell adhesion, is the essential step towards the progression of EMT (Kalluri & Weinberg, 2009). During an event of EMT, E-cadherin is suppressed by various transcription factors such as Snail1/2, TWIST, and fork-head box protein C2 (FOXC2) (Yang & Weinberg, 2008). Alternatively, various signalling pathways are able to activate Snail during EMT. For example, the ERK pathway has been shown to induce the expression of Snail as well as the Wnt/beta-catenin pathway (Jiang et al, 2007; Xie et al, 2004). Furthermore, transforming growth factor beta (TGF-beta) has also been associated with the process of EMT by activating the expression of Snail (Xu et al, 2009). Conversely, grainyhead-like protein 2 homologue (GRHL2) and ETS-related transcription factors 5 (ELF5) are able to suppress EMT by promoting mensenchymal-epithelial transition (MET) (Chakrabarti et al, 2012; Cieply et al, 2013). Finally, the well-studied tumor suppressor p53 has also been shown to suppress EMT through the regulation of microRNAs such as miR34 and miR200. Both of these microRNAs are upregulated by p53 and they have been shown to also suppress Snail expression during the suppression of EMT (Chang et al, 2011; Kim et al, 2011).  Under different situations and various circumstances, tumor cells undergo transitions between mesenchymal migration and amoeboid migration, which is referred to as the mesenchymal-amoeboid or the amoeboid-mesenchymal transition (MAT/AMT).  Cancer cells undertake morphological changes, integrin re-localization, or actin filament  17 rearrangements during the transition in order to adapt to various environmental factors (Pankova et al, 2010). As discussed earlier, Rho/ROCK signalling pathway is crucial for actomyosin contractility that is fundamental during the amoeboid migration. Therefore, the suppression of the Rho/ROCK pathway induces the AMT (Sahai & Marshall, 2003). Conversely, the regulator of WASP during actin filament polymerization, CDC42, has been associated with the promotion of MAT. Furthermore, CDC42 has also been shown to induce the phosphorylation of MLC during the actomyosin contractility (Gadea et al, 2008). As mentioned earlier, the pericellular proteolysis is very important during the mesenchymal migration, and the abolishment of the cell surface proteases, such as the MMPS, can essentially suppress the ability to remodel or degrade ECM. However, since tumor cells are not confined or trapped within the extracellular network, they proceed to the amoeboid transition to overcome this limitation (Wolf et al, 2003a). Furthermore, since integrins are required during mesenchymal migration for the construction of focal contacts that give rise to the resultant spindle-shaped morphology, the disruption of the integrins’ function forces the tumor cells to adapt to the rounded shapes. Thus, tumor cells that undergo the MAT have a reduced expression of their cell surface integrins, as well as a reduced FAK phosphorylation level (Carragher et al, 2006). Therefore, the regulation of actomyosin contractility, proteolytic activity, and ECM integrity are considered to be important factors that contribute to the selection of migration modes undertaken by cancer cells.    1.3 Angiogenesis With the ability to spread to local tissues and travel to distant organs, cancer cells are deadly and extremely life threatening. Angiogenesis is a process that provides fundamental  18 supplies to the tumor cells that help them to grow and proliferate. In order to translocate to a distant site, angiogenesis also assists the tumor cells to penetrate through blood vessels and enter the circulation system. This is underlying reason behind how tumor cells are able to spread with a remarkable rate of efficiency (Folkman, 1971).   1.3.1 Angiogenesis and Cancer Angiogenesis is known to be the process in which neovessels are formed from pre-existing vessels where they act as the basic source to provide cancer cells with oxygen and nutrients as well as a direct route for cancer metastasis (Folkman, 1971). In fact, when tumor cells were infused with or without a blood circulation, a significant smaller tumor diameter was observed when there was limited blood supply compared to a constant blood supply that permitted angiogenesis (Muthukkaruppan et al, 1982). Furthermore, previous findings also revealed that with the lack of blood circulation and angiogenesis, tumor cells were found to undergo apoptosis (Holmgren et al, 1995). During an onset of angiogenesis, endothelial cells stimulated by angiogenic factors are first relocated to the site of injury. These cells proliferate at the injury site where there is already tissue destruction and hypoxia in order to stabilize the condition. Meanwhile, angiogenic factors are continuously being released to further amplify the effect (Jakobsson, 1994). However, even in the absence of injuries, angiogenesis is also stimulated by tumor cells that manipulate various regulatory proteins that affect angiogenic factors. It has been noted that regulation of both activators and inhibitors of the angiogenic factors are required during the onset of angiogensis (Dameron et al, 1994).  Some notable angiogenic factors such as vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) have all contributed to the rapid onset of  19 angiogenesis responses in human melanoma (Danielsen & Rofstad, 1998). Other known activators of angiogenesis include, but are not limited to, the cytokines, proteases and certain oncogenes (Gruss et al, 2003; Rak et al, 2000).   1.3.2 Angiogenic Factors VEGF has been identified as a potent angiogenic inducer. Both VEGF and its natural receptor (VEGFR) are observed in tumor cells and nearby stroma cells where they perform vital functions during the regulation of angiogenesis and neovascularization (Folkman, 2002). When there is a shortage of blood supply near the tumor cells, or the tumor cells have grown beyond the limit of the oxygen supply referred to as hypoxia, angiogenic responses would initiate the upregulation of VEGF through the activation of hypoxia-inducible factor-1a (HIF-1a) (Forsythe et al, 1996). After the binding of VEGF to its receptor VEGFR2, the activated endothelial cells are able to induce the expression of MMPs. As discussed earlier, MMPs induce the degradation of the ECM which in turn assists the endothelial cells to migrate to the site of neovessel formation (Heo et al, 2010). During the development of the newly formed blood vessels, angiotensins are required for maintaining the structural integrity of the blood vessels. Besides governing neovessel formation, angiogensins also induces the expression of VEGF to further stimulate blood vessel formation (Otani et al, 2000). Another angiogenesis activator, referred to as bFGF, is released through exocytosis from the endoplasmic reticulum. Following the degradation of the ECM by MMPs, bFGF expression is significantly increased, since a significant amount of bFGF is found at the extracellular matrix, to induce the endothelial cell proliferation and vascular structure formation (Cross & Claesson-Welsh, 2001). Furthermore, the placental growth factor (PIGF) is able to combine  20 with VEGF to form a heterodimer, and together they can bind to VEGFR2 on the endothelial cells to initiate angiogensis responses (Eriksson et al, 2002).  Meanwhile, PIGF can also bind to neuropilin receptors to induce angiogenesis without the need to form a heterodimer with VEGF (Zins et al, 2013). Interleukin-6 (IL-6) is another very important angiogenic factor, which has multiple other biological functions, including the regulation of inflammation, proliferation, immune responses, and tumorigenesis (Guo et al, 2012). IL-6 can bind to the IL-6 receptor alpha-chain (IL-6R or gp80), which then recruit the beta-chain (gp130). This in turn causes the homodimerization of gp130, and the activated gp130 is able to initiate the singling pathway of Janus-activated kinases/signal transducers and activators of transcription (Jak/STAT), Ras/ERK, and the PI3K/Akt pathways (Hibi et al, 1996). IL-6 has also been shown to induce the expression of VEGF during angiogenic responses (Huang et al, 2004).   1.3.3 Anti-angiogenic Therapies Anti-angiogenic therapy ensures the destruction of tumor vasculature and marks the death of the tumor cells but with minimal adverse effects. This is because during normal physiological processes, angiogenesis normally does not have a big influence other than the female reproductive cycle and the wound healing process. Thus, anti-angiogeneic therapy is likely to only affect the recovery of injuries for the patients receiving the treatments (Bodnar, 2014). The targets of the anti-angiogenic treatments include the extracellular matrix and metalloproteinase (Rasmussen & McCann, 1997), the growth factors and their receptors (Ferrara et al, 2005), and the integrin receptors (Chen et al, 2008). For targeting the metalloproteinase, small molecules are made to target the active site of MMPs to inhibit their functions (Skotnicki et al, 1999). Meanwhile, monoclonal antibodies and soluble forms of  21 receptors can be constructed to bind growth factors and inhibit their activities. An example of such a monoclonal antibody that has been approved by the FDA is referred to as bevacizumab (Avastin), which can block the activities of VEGF (Gerber & Ferrara, 2005). At the same time, small molecule inhibitors have also been made to target the growth factor receptors to suppress their tyrosine kinase activities (Drevs et al, 2004). Sorafenib is an FDA approved small molecular inhibitor that targets multiple tyrosine kinases including VEGFR (Barrascout et al, 2010). Finally, monoclonal antibodies have been designed to block the binding site of integrin receptors, such as the monoclonal anti-alphavbeta3 integrin antibody (Abegrin) for treating metastatic cancers, and it is currently being tested in clinical trials (Zhang et al, 2007). Interestingly, however, anti-angiogenesis treatments can sometimes increase tumor hypoxia which can stimulate more-aggressive cancer cells to relocate to a new environment. As a result, the cancer cells become more aggressive and metastatic which decreases the effective of chemotherapy. An alternative approach, referred to as pro-angiogenesis treatments, increases blood vessel density and blood flow which leads to the improvement of chemotherapy delivery. Increased angiogenesis also increases blood vessel dilation and leakiness which causes the reduction of cancer cell growth and metastasis (Wong et al, 2015).  1.4 The Tetraspanin Superfamily The tetraspanins are proteins that reside on the membrane of a eukaryotic cell. They have four transmembrane domains, two intracellular domains with an N and a C terminus, and two extracellular domains. Members of the tetraspanin family have various functions  22 including the regulation of cell development, growth, motility, immune responses, metastasis suppression, and signal transduction (Hemler, 2005).   1.4.1 Structure of Tetraspanins  What distinguish the tetraspanins from other membrane proteins with four transmembrane domains is that all tetraspanins share a number of conserved domains.  The four hydrophobic transmembrane domains (TM1-TM4) are embedded within the membrane. The extracellular domain of tetraspanins are composed of two loops with one longer than the other, which are denoted as the small extracellular loop (SEL or EC1) and the large extracellular loop (LEL or EC2). Meanwhile, the intracellular domain contains the N and C terminal of the polypeptide chain (Seigneuret et al, 2001). Most tetraspanins are known to have cysteine residues in the EC2 domain that contains the highly conversed cysteine-cysteine-glycine motif. Other regions contain the N-linked glycosylation sites in the EC2 or palmitoylation in the TM region (Kitadokoro et al, 2001). Meanwhile, antibody binding tetraspanins have been found to be sensitive to reducing agents, which reveal the presence of disulfide bonding located at EC2 binding site (Hemler, 2005). Finally, polar amino acid residues that form structures similar to that of ion channels are present in the transmembrane domains. This marks the possibility that tetraspanins may have functions similar to proteins such as ion channels. However, the existence of the polar amino acid residues in those tetraspanins may also be involved in maintaining structural integrity and protein stability (Horejsi & Vlcek, 1991).      23 1.4.2 Regulation of Tetraspanin Expression A common feature among most tetraspanin genes is the absence of a TATA box for transcription initiation (Le Naour et al, 1996). In fact, the promoter regions of the tetraspanins are found to be G-C rich and contain up to three different binding sites that can be targeted by various transcription factors such as the activator protein 1 (AP-1), Sp1, or ETF (Hotta et al, 1992). Furthermore, some post-translational modifications in tetraspanins include glycosylation or acylation. Various glycosylation sites have been found such as in the EC1 or EC2 (Tominaga et al, 2014), whereas some tetraspanins do not get glycosylated altogether as observed in cluster of differentiation 81 (CD81) (Rothwangl et al, 2008).  Previous studies have revealed some tetraspanins to have relationship with tumor progression. For example, CD9 expression is lost in various cancers such as human ovarian, breast and lung cancers (Funakoshi et al, 2003; Hwang et al, 2012; Kischel et al, 2012). Also, the reduced CD9 expression has been reported to be inversely correlated with patient survival in breast carcinoma (Miyake et al, 1996). The restoration of CD9 has been shown to suppress cancer cell motility and metastasis (Ovalle et al, 2007). CD82, which has been originally found to be a putative metastasis suppressor in prostate cancer, has also been reported in other cancer types including melanoma (Lijovic et al, 2002; Tang et al, 2014b). Furthermore, CD81 has also been reported to upregulate liver cancer cell proliferation by inducing the ERK/MAPK pathway (Carloni et al, 2004). Moreover, not only is CD63 expression absent in the late stages of melanomas which suggest its tumor suppressive functionality, the reduced CD63 expression has also been found to be related to platelet lysosome and dense granule deficiencies which can cause platelet abnormality (Jang & Lee, 2003; Nishibori et al, 1993).   24 1.4.3 Diverse Functions of Tetraspanin Proteins Research in the field of tetraspanins has only been ongoing for less than 20 years. So far, there have been 34 tetraspanins in mammals discovered with 33 of them identified in human (Hemler, 2005). Tetraspanin proteins have diverse functions including the regulation of cell adhesion, motility, proliferation, cancer metastasis, immune responses, and acting as a cell surface receptor (Hemler, 2001). The tetraspanin CD151 have been shown to have important functions in the regulation of cell adhesion, growth, and motility. It has been found to interact with integrins or with other tetraspanins located on the membrane. The tetraspanin CD53 has been identified as a leukocyte surface antigen (Korinek & Horejsi, 1993). It has important function during signal transduction within the T-cells and natural killer cells, and it has also been found to regulate cell growth and proliferation. CD53 deficiency has been linked with immunodeficiency induced diseases, where the patients are more susceptible to bacteria and virus infections (Rasmussen et al, 1994). CD37 has been found to be highly expressed in mature B-cells. CD37 has been found to be essential for  B-cell survival as well as T cell proliferation, antibody production, and dendritic cell migration (Knobeloch et al, 2000). With its suppressed expression in T-cells, macrophages, and natural killer cells, it has been investigated and found to be a potential target to treat  B-cell malignancies, where it is found to be suitable for immunotherapy (Palomba & Younes, 2013). Tetraspanin CD81 is one of the most studied tetraspanins,, and it is generally being expressed in endothelial, epithelial, and hematopoietic cells (Lin et al, 2011; Rohlena et al, 2009; Yanez-Mo et al, 2001). CD81 has been shown to interact with another tetraspanin protein, referred to as CD9, during the regulation of muscle cell fusion and myotube maintenance (Charrin et al, 2013). Also, during the infection of the hepatitis C virus (HCV),  25 CD81 has been reported to regulate the attachment and internalization of HCV (Bartosch et al, 2003). Furthermore, HIV-1 group-specific antigen (gag) protein has been shown to interact with CD81 within the tetraspanin-enriched microdomain on the T-cells during viral infection. When HIV-1 producing cells receive anti-CD81 antibodies or CD81 siRNA treatments that suppress CD81 expression, the level of HIV-1 protein released is significantly reduced and results in a suppressed infectivity (Grigorov et al, 2009). The tetraspanin-enriched microdomain that HIV-1 used for virus assembly and release was found to contain the tetraspanin CD63 protein. Once again, by silencing CD63, a significant decrease in the production of HIV-1 was observed (Fu et al, 2015). Furthermore, CD63 has also been reported to influence the activity of VEGFR thereby activating downstream signalling pathways as well as the proliferation of endothelial cells (Tugues et al, 2013). Recent studies also revealed the oncogenic function of CD63, where it has been shown to induce ovarian cancer cell invasion (Kobayashi et al, 2014). Another well studied tetraspanin, referred to as CD82 or KAI1, has been reported to be downregulated in multiple cancers (Liu & Zhang, 2006). It has been shown to be activated by p53 during transcription regulation, and the loss of KAI1 in advanced cancers is directly correlated with a poorer patient survival (Shinohara et al, 2001). As will be discussed in greater detail below, the functions of KAI1 are a major focus of my studies.  1.4.4 Protein Interactions of the Tetraspanin Web Tetraspanins have been reported to associate with other tetraspanins as well as many other membrane proteins such as cell surface integrins, some Ig superfamily members, and growth factor receptors and even their ligands. This microdomain consisting of multiple  26 tetraspanins and various membrane proteins is being described as the tetraspanin web (Levy & Shoham, 2005). For example, CD151 has been reported to interact with the α3β1 integrin using covalent crosslinking. This interaction allows CD151 to mediate the regulation on cell morphology and motility through the α3β1 integrin (Winterwood et al, 2006). Furthermore, CD9 and CD81 tetraspanins have also been shown to have strong covalent crosslinks with the transmembrane immunoglobulin superfamily (IgSF) proteins EWI-2 and EWI-F. It is believed that the functional relevance of the association is related to oocyte fertilization, cell fusion, and the growth factors receptor signalling cascade (Kolesnikova et al, 2009). CD9 can also interact with various ligands of EGF receptor including TGFα and heparin-binding EGF (HB-EGF). This association is able to increase the efficiency of the ligand binding and receptor activities (Nakamura et al, 1995; Shi et al, 2000). Besides modulating the agonists of the growth factor receptors, tetraspanins can also associate with the growth factor receptors directly. CD9, CD63 and CD81 have been reported to form a complex with the c-kit receptor tyrosine kinase and induce the phosphorylation of c-kit. Thus, the tetraspanin complex is believed to stimulate the recruitment of phosphatases and tyrosine kinases to the tetraspanin network (Anzai et al, 2002). Also, CD9 has also been shown to increase the ligand binding efficiency of HB-EGF after the association with its transmembrane receptor (Nishida et al, 2000). Furthermore, tetraspanin CD82 has been shown to associate with the EGFR to deactivate EGFR’s activity. As a result, the downstream signalling pathways of EGFR are repressed along with a decreased migratory rate and metastasis activity of the cancer cell (Odintsova et al, 2013).    27 1.5 The Tumor Suppressor KAI1 KAI1, also known as CD82, is a member of the tetraspanin superfamily, which is located on the surface of the cell and is able to form the tetraspanin microdomain with various other tetraspanins (Rubinstein et al, 1996). It has been shown to have important function during tumor progression through its regulation of tumor migration, invasion, and metastasis. In addition, decreased KAI1 expression has been correlated with a poorer metastatic patient survival in esophageal squamous cell carcinoma (Miyazaki et al, 2000).   1.5.1 Gene Construct and Mutations KAI1 is located on chromosome 11p11.2 with 10 exons and 9 introns that make up this 80kb gene (Dong et al, 1995). The KAI1 protein contains 267 amino acids, which includes 4 hydrophobic domains that span the plasma membrane. One isoform of KAI1 protein has been observed to have only 242 residues, but its functional differences are yet to be described (Mooez et al, 2011). Similar to other tetraspanins, KAI1 does not have a TATA box at the promoter region, yet contains binding sites for various transcription factors such as SP1 and AP2 (Dong et al, 1995). KAI1 has been found to be rarely mutated and that even with the existence of mutations, there is minimal effect on KAI1’s regulation and function (Miyazaki et al, 2000). As shown in ovarian carcinoma, a single mutation at codon 241 has been discovered in both normal and tumor cells (Liu et al, 2000).  1.5.2 Structure of Tetraspanin KAI1 KAI1 resembles the typical shape of a tetraspanin protein. It has four transmembrane domains and two extracellular domains, SEL and LEL. Similar to the TM domains in other  28 tetraspanins, the TM domains of KAI1 also contain polar residues. They are located within the TM interface where they can form hydrogen bonds and maintain structure stability during the event of dimerization (Bari et al, 2009). The TM1 domain of KAI1 has been reported to contain the cell membrane localization signal, of which a construct o KAI1 missing TM1 is not able to be transported from the endoplasmic reticulum to the cell surface (Cannon & Cresswell, 2001). The LEL of KAI1 contains the highly conserved sequence of six consecutive cysteine amino acids. The key element of the cysteine residues is their contribution to disulfide bond formation, which is important for the structural integrity of the LEL (Hemler, 2001).  N-glycosylation is a common feature in many tetraspanins, and it has also been observed in KAI1 (Ono et al, 2000). Furthermore, the SEL of KAI1 has been reported to act as a supportive structure which helps the folding of the LEL (Masciopinto et al, 2001). Meanwhile, the cytoplasmic domains of KAI1 have been found to be relatively short with 11 residues at the N-terminus and 15 residues at the C-terminus (Zhou et al, 2004). Furthermore, since the cytoplasmic domain of KAI1 contains palmitoylation sites, acylation is observed at the N-terminus that primarily help to stabilize the short cytoplasmic domain on the membrane as well as to assist tetraspanin-tetraspanin associations during the formation of a tetraspanin microdomain (Stipp et al, 2003).    1.5.3 KAI1’s Regulation of Cancer Progression KAI1 was originally found as a tumor suppressor in prostate cancer cells. Since then, various studies revealed that KAI1 had important functions during the regulation of cancer migration, invasion, and metastasis. The role of KAI1 has been observed in a wide range of cutaneous cancers such as prostate, breast, lung, thyroid, skin, and liver cancers.  29 Furthermore, the loss of KAI1 expression has frequently being associated with a poor prognosis in many advanced cancers (Dong et al, 1995; Ito et al, 2003; Malik et al, 2009; Mine et al, 2015; Tang et al, 2014b; Zheng et al, 2007). Furthermore, KAI1 has also been reported to induce apoptosis through reactive oxygen intermediates (Schoenfeld et al, 2004). Due to the functional similarities among tetraspanin proteins, KAI1’s regulation of cancer progression also involves the utilization of the tetraspanin web, where interaction and regulation of various other cell surface proteins can be observed (Bari et al, 2009).  KAI1 has been shown to interact with various integrins during its regulation of cell adhesion and migration (Sridhar & Miranti, 2006). For example, KAI1 has been reported to suppress the activity of Src kinase which is activated by integrin signalling (Jee et al, 2003). KAI1 has also been shown to inhibit the activity and the formation of the p130CAS-Crk complex that is responsible for the regulation of cell motility (Zhang et al, 2003a). Since Rho GTPases are activated by the p130CAS-Crk complex, KAI1 can indirectly suppress the activity of Rho GTPases and reduce the migratory rate of cancer cells through the regulation of cytoplasmic actin organization (DeMali et al, 2003). EGF and EGF receptor (EGFR) signalling pathway, which is responsible for cell proliferation and migration, has been shown to be frequently elevated in advanced cancers (Yarden & Sliwkowski, 2001). However, KAI1 has been shown to associate with EGFR and induce the inactivation and a rapid internalization of EGFR, which in turn causes the attenuation of the EGFR signaling pathway (Odintsova et al, 2013). As mentioned earlier, KAI1 can form strong covalent crosslinks with the transmembrane immunoglobulin superfamily (IgSF) proteins EWI-2 and EWI-F. EWI-2 has been shown to have functional relevance with regard to oocyte fertilization, cell fusion, cell migration, and the growth factor receptor signalling cascade (Kolesnikova et al, 2009).  30 As a result, the association of EWI-2 and KAI1 has also been reported to effectively suppress cancer cell migration (Zhang et al, 2003b).  As cell surface proteins can move freely in the cell membrane where they are frequently recruited to the site of activation for the regulation of cell motility, the regulation of membrane protein distribution is important for cancer cell migration. KAI1 has been shown to be involved during the activation of urokinase-type plasminogen activator receptor (uPAR). The uPAR protein has been reported to have important functions during cancer cell migration, adhesion and metastasis. Thus, the suppressed activity of uPAR due to KAI1 regulation significantly suppresses its ligand binding ability and significantly decreases the plasminogen activity (Bass et al, 2005). Besides KAI1’s ability to cause the internalization of EGFR, KAI1 is able to internalize α6 integrins. This causes a significant decrease in α6 integrin-induced cell adhesion activities (He et al, 2005). Recent studies also revealed an interaction of KAI1’s C-terminal domain and another transmembrane protein, also with four TM domains, but not belonging to the tetraspanin superfamily, referred to as KAI1 C-terminal interacting tetraspanin (KITENIN). This association was shown to inhibit KAI1’s tumor suppressive function. In fact, KITENIN overexpression has been shown to promote cancer cell migration, invasion, and metastasis in various cancer models (Lee et al, 2004). As a result, the regulation of KAI1’s activity or even gene expression is crucial during cancer progression, and the restoration of KAI1’s function is an important benchmark for the development of new therapeutic strategies.      31 1.5.4 Regulation of KAI1 Expression As discussed earlier, the KAI1 gene does not have the TATA box at the promoter region, yet it contains binding sites for various transcription factors such as SP1 and AP2 for the regulation of protein expression (Dong et al, 1995). The transcription factor p53 has also been shown to induce the expression of KAI1. And the loss of p53 expression can directly induce the downregulation of KAI1 expression during the progression of cancer metastasis (Mashimo et al, 1998). Furthermore, the transcription factor JunB has been shown to induce KAI1 expression. JunB and p53 have a synergic effect on KAI1 expression, of which the loss of the activity or expression of the two transcription factors directly results in the decreased expression of KAI1 (Marreiros et al, 2005). The p50 subunit of NFκB has also been shown to have important regulatory functions on KAI1’s transcription regulation. However, the p65 subunit of NFκB does not seem to incorporate with the promoter of KAI1. In fact, the regulation of p50 depends on its dimerization and the co-factors that are recruited along with p50 (Shinohara et al, 2001). For example, while the association of p50 and Tip60/Fe65 can upregulate KAI1 expression, its interaction with N-CoR/TAB2/HDAC3 complex leads to the suppression of KAI1 expression (Baek et al, 2002; Telese et al, 2005). Besides being regulated by transcription factors, KAI1’s expression can also be modulated by cytokines. The cytokines that are responsible for the regulation of immune and inflammatory responses, such as IL-6 and tumor necrosis factors (TNF), have also been shown to indirectly induce the expression of KAI1 (Baek et al, 2002; Shinohara et al, 2001). Transcription regulation of KAI1 is a complex process involving various activators and repressors. Thus, the restoration of KAI1 expression is crucial for suppressing cancer metastasis.    32 1.6 Objective and Hypotheses KAI1 was previously identified as a metastasis suppressor in prostate cancer (Dong et al, 1995), yet its role during cancer progression and metastasis was still not clear.  In our study, we were trying to investigate the potential physiological role of KAI1 in melanoma prognosis and cancer progression. We performed immunohistochemistry on tissue microarrays to investigate whether differences in KAI1 staining could be detected at different stages of melanoma. Also, we overexpressed KAI1 and knocked down KAI1 to study the cellular responses of melanoma cells in tissue culture by using in vitro functional analyses. According to the results, we found that KAI1 was crucial during melanoma metastasis and had important function during the regulation of melanoma cell migration and invasion. We hypothesized that KAI1 was an independent diagnostic marker and a key player during melanoma progression. During this investigation, we also showed that KAI1 was able to regulate another tumor suppress known as ING4.   Since angiogenesis was required for cancer metastasis (Folkman, 2002), we next investigated KAI1’s role during melanoma angiogenesis. By culturing HUVECs in the conditioned media collected from melanoma cells treated with KAI1 overexpression or knockdown, we were able to investigate KAI1’s regulation on HUVEC growth. We hypothesized that KAI1 was able to suppress melanoma angiogenesis through the regulation of ING4. We also investigated KAI1’s regulation on EGFR and VEGFR since KAI1 was a scaffolding membrane protein that was able to recruit other member proteins.  Since the secreted SEMA3C protein had previously been associated with tumor cell migration and angiogenesis (Goshima et al, 2002), we investigated SEMA3C’s role during melanoma angiogenesis. We performed in vitro and in vivo analyses after forced SEMA3C  33 expressing or knockdown and found that SEMA3C had an important regulatory function during melanoma angiogenesis. Since KAI1 expression was frequently lost in cancer cells and forced KAI1 expression suppressed melanoma angiogenesis, we proposed that KAI1 overexpression was able to abolish the SEMA3C-induced melanoma angiogenesis. We performed a dual overexpression of KAI1 and SEMA3C and collected the conditioned media to investigate HUVEC growth and tubular structure formation. Conversely, a co-knockdown of KAI1 and SEMA3C was also performed. Collectively, we discovered that KAI1 was able to suppress SEMA3C-induced angiogenesis as well as VEGFR2 phosphorylation.     34 Chapter 2: Material and Methods 2.1 Immunohistochemistry Tissue microarray construction and immunohistochemistry staining were performed as described previously (Dai et al, 2005). We selected and marked the most representative tumor area on the hematoxylin and eosin-stained slide. The primary rabbit anti-KAI1 antibody (1:1000 dilution, Novus Biologicals, Littleton, CO, USA) and the biotin-labeled secondary antibody (DAKO Diagnostics, Glostrup, Denmark) were used. Negative controls were included following the same procedure except that the KAI1 antibody was replaced with DAKO antibody diluent during the primary antibody incubation. The staining intensity and percentage of KAI1-positive cells were evaluated in a blinded manner by three independent observers (including a dermatopathologist) simultaneously, and a consensus score was reached for each core. KAI1 staining intensity was scored as 0 (negative), 1 (weak), 2 (moderate), and 3 (strong). The percentage of KAI1-positive cells was scored into four 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. ImageJ software (NIH, Bethesda, MD) was also used as a semi-automated scoring method. Images were captured for each sample and analyzed by ImageJ software to obtain the relative staining intensity.  2.2 Evaluation of TMA Immunostaining The level of KAI1 staining was evaluated by immunoreactive score (Remmele & Stegner, 1987), which is calculated by multiplying the scores of staining intensity and the score of the  35 percentage of positive cells. Based on the immunoreactive score (IRS), KAI1 staining pattern was defined as weak (IRS: 0–6) and strong (IRS: 8–12).   2.3 Statistical Analyses The SPSS version 16.0 software (SPSS, Chicago, IL, USA) was used for the statistical analysis and all tests of statistical significance were two-sided. The 2 test was used to compare the KAI1 staining level in different melanocytic lesions, as well as the correlation between KAI1 staining and the clinicopathological parameters of the melanoma patients, including AJCC stage, age, gender, tumor thickness, ulceration, histological subtype and tumor location. The survival time used was from the date of melanoma diagnosis until the date of death or the last follow-up. The correlation between KAI1 expression and patient survival was analyzed using the Kaplan–Meier survival curve and log-rank test. Univariate and multivariate Cox proportional hazards regression models were performed to estimate the crude HRs or adjusted HRs with 95% CIs. p values of less than 0.05 were considered statistically significant.  2.4 Cell Lines and Cell Culture The human melanoma cell lines MMRU and MMLH were developed by Massachusetts General Hospital, and Harvard Medical School. MMRU and MMLH cell lines were derived from cervical lymph nodes and visceral (mental) metastasis respectively, and they were obtained by explant culture. Both melanoma cell lines were metastatic melanoma cell lines (Byers et al, 1991). Melanoma cell lines A375 and Mel624 were purchased from the American Type Culture Collection (Manassas, VA). All the melanoma cell lines were  36 cultured in the Dulbecco’s modified Eagle’s medium (Hyclone, South Logan, UT) and supplemented with 10% fetal bovine serum (Hyclone). Human umbilical vein endothelial cells (HUVECs) were cultured in endothelial cell medium (ECM) (ScienCell, Carlsbad, CA, USA). All cells were maintained in 5% CO2 at 37°C.  2.5 Protein Extraction and Western Blot Cells were harvested and washed with PBS twice and the whole-cell proteins were extracted as described previously (Wang & Li, 2006). After protein concentration was determined by protein assay (Bio-Rad, Mississauga, Ontario, Canada), western blot analysis was performed as descried previously (Chin et al, 2005). The following antibodies were used for Western blot: rabbit anti-KAI1 (1:1000; Novus Biologicals), mouse anti-Flag (1:1000; Applied Biological Materials), mouse anti-IkB (1:1000; Cell Signalling), rabbit anti-p65 (1:1000, Cell Signalling), mouse anti-BRMS1 (1:1000) (Phadke et al, 2008), rabbit anti-ING4 (1:1000; Sigma) and β-Actin (1:10000; ImmuneChem Pharmaceuticals Inc, Burnaby, BC, Canada). Infrared IR dye-labeled secondary antibody IRDye 680 or IRDye 800 was applied to the blot for 1 hour at room temperature and then blots were scanned using Odyssey Infrared Imaging system (LI-COR Biosciences).   2.6 ELISA The secreted IL-6 protein level in the conditioned muedium (CM) was measured with the human IL-6 ELISA kit (eBioscience, San Diego, CA) following the manufacturer’s instructions.    37 2.7 Expression Plasmids, siRNA and Transfection Flag-tagged KAI1 was kindly provided by Dr. K. Watabe (Southern Illinois University School of Medicine, Springfield, Illinois). HA-ING4 and miING4 were constructed as described previously (Li & Li, 2010). For transfection, melanoma cells were grown to ~65% confluency and then transiently transfect with Flag-KAI1, HA-ING4, shKAI1 (Santa Cruz Biotechnology, Santa Cruz, CA), or miING4 plasmids by Effectene reagent (Qiagen, Mississauga, ON) according to the instructions provided by the manufacturer. The cells were collected for use 48h after transfection. For siRNA transfection, cells were transfected with Silenfect transfection reagent (Bio-Rad, Mississauga, Ontario, Canada) when the cells were ~50% confluent. The cells were incubated with either nonspecific control siRNA (si-Ctrl) or KAI1, p65, and ING4 specific siRNAs (Qiagen, Mississauga, Ontario, Canada).  2.8 Reverse Transcription and Real-time Quantitative Polymerase Chain Reaction Total RNA was obtained by using Trizol extraction (Invitrogen, Carlsbad, CA) and reverse transcribed into cDNA by using SuperScript First-Strand Synthesis System (Invitrogen) following the protocol provided by the manufacturer. Real-time reverse transcription-PCR was performed with SYBR Green Master mix system (Applied Biosystem, Carlsbad, CA). The sequences of human KAI1 primers were 5’-GCTCATGGGCTTCCTGGGCT (forward) and 5’-GAGCTCAGTCACGATGCCGC (reverse). The primers for human ING4 were 5’-CACAGACCTGGCCCGTTTT (forward) and 5’-AGTCCGGCCTTTCTTTTTGC (reverse). The primers for GAPDH were 5’-GAAGGCTGGGGCTCATTT (forward) and 5’-CAGGAGGCATTGCTGATGAT (reverse).   38 2.9 Wound Healing Assay The cells were seeded into 6-well plates and transfected with either Flag-KAI1 plasmid or control plasmid and either 2 si-KAI1s or si-Ctrl. After 24 hours, the cells were cultured in fresh medium for 24 hours and treated with 10 μg/mL of mitomycin C (Sigma) for 2 hours. After being washed with PBS, a standard 2-μL pipette tip was used to scratch across the wells. The detached cells were removed by gentle washing of PBS twice. The gaps were monitored using microscope and five pictures were taken from each gap. Pictures were taken at the same spot after 18 hours. Each experimental group was repeated three times.  2.10 Cell Invasion Assay Boyden chamber assay was used to perform the cell invasion analysis which was previously described (Wong et al, 2007). In brief, 20 μL of 5 mg/mL matrigel (BD Biosciences, Mississauga, Canada) in serum-free medium was added to the upper compartment of the transwell culture chamber (polycarbonate membrane containing 8.0 μm pores). 4 × 104 cells were suspended in 250 μL of serum-free medium and seeded onto the upper chamber. The transwells were then placed in a 24-well plate that contains 750 μL of complete medium. The cells were treated with 10 μg/mL of mitomycin C (Sigma) for 2 hours followed by an incubation of 24 hours at 37°C. Then, the cells were fixed with 10% trichloroacetic acid at 4°C for 1 hour. After the removal of non-invaded cells, cells were incubated with 0.5% crystal violet for 2 hours at room temperature. Cells were destained with 30% acetic acid that was tested for absorbance at 590nm.    39 2.11 Zymography The protocol used for zymography assay was described previously (Wong et al, 2007). Cells were transfected with Flag-KAI1 plasmid or control plasmid, and 2 si-KAI1s or si-Ctrl. Serum-free medium was used to starve the cells overnight. The conditioned medium obtained was concentrated with YM-3 Centricon membranes (Millipore, Billerica, MA). Proteins (10 μg) were loaded on a 10% polyacrylamide gel with 0.1% gelatin (Sigma) added. After electrophoresis, the gel was incubated in Triton X-100 (20 mmol/L Tris-HCl [pH 8.0], 150 mmole/L NaCl, 5 mmol/L CaCl2 and 2.5% Triton X-100] exchange buffer for 30 minutes. The gel was then washed with incubation buffer (exchange buffer lacking Triton X-100) thrice, and then with incubation buffer at 37°C overnight. To stain the gel, 0.5% Coomassie blue R250 (Sigma) was used for 1 hour and the gel was destained using a solution containing 30% methanol and 10% acetic acid for 1 hour. Gelatinolytic activity was visualized by the clearing on the gel. Recombinant MMP-2 (Millipore, Billerica, MA) was used as a positive control.   2.12 Immunofluorescent Staining Cells were transfected with Flag-KAI1 plasmid or control plasmid, and 2 si-KAI1s or si-Ctrl and subcultured onto coverslips in six-well plates. The cells were serum starved overnight and stimulated with complete medium containing 10% fetal bovine serum for 30 minutes. ROCK inhibitor Y27632 (10 μmol/L) was used together with serum free medium for the designated experiment groups and applied to cells overnight. The cells were fixed with fixation solution (1% paraformaldehyde and 0.5% Triton X-100 in PBS) for 20 minutes at 4°C. The slips were incubated with bovine serum albumin for 1 hour, then with rabbit anti- 40 KAI1 antibody (1:1000) for 1 hour, followed by Cy2-conjugated goat anti-rabbit antibody (1:2000; Jackson ImmunoResearch, West Grove, PA) for 1 hour. The cells were then stained with phalloidin–rhodamine (1 U per coverslip; Invitrogen) for 30 min. Finally, one drop of mounting media was added to each coverslip and incubated for 30 minutes and the cells were visualized under a florescent microscope. Pictures were taken using a cooled mono 12-bit Retiga-Ex camera equipped with Northern Eclipse imaging software. 10 pictures were taken for each slide and subjected to analysis using ImageJ software to obtain the relative cellular fluorescent intensity.  2.13 Sulphorhodamine B (SRB) Assay Melanoma cell lines were cultured in six-well plates with ~70% confluency using fresh serum-free medium for 24h, and 1mL of the CM was collected. HUVECs were seeded in 96-well plates at 5 ×103 per well and cultured in fresh ECM for 24 h, and then in 100 μl of CM for another 24h. At each time point, HUVEC were fixed with 10% trichloroacetic acid, stained with 0.4% SRB in 1% acetic acid, and then destained with 1% acetic acid. HUVEC density was quantified by dissolving bound dye in 10 mM Tris (pH 10.5) followed by colorimetric determination at 550 nm. The initial time point (0 day) was measured by fixing cells immediately after they had been cultured in CM for 24h. Subsequent time points were measured by fixing cells 1, 2 and 3 days later. Relative rates of cell growth were calculated as a ratio of the cell density at each time point over the cell density at 0 day.     41 2.14 Tube Formation Assay of HUVECs The tube formation assay was conducted as described previously (Wani et al, 2011). Melanoma cell lines were cultured in six-well plates with ~70% confluency with fresh serum-free medium for 24h, and 1mL of the CM was collected. MatrigelTM (BD Biosciences, Mississauga, Ontario, Canada) was coated on 96-well plates at 37°C for 2h prior to the addition of 2 × 104 HUVECs suspended in 100μl of CM.  After 10 hours of incubation at 37°C, HUVECs tube formation was examined by photography under a microscope. The tubular structures were counted in randomly selected fields.   2.15 FACS Analysis Melanoma cell lines was cultured in six-well plates with ~70% confluency using fresh serum-free medium for 24h, and the CM was collected. HUVECs were cultured in fresh ECM for 24 h, then in CM for another 24h, and finally collected by trypsinization. HUVECs were then pelleted using centrifugation at 500 × g for 5 min, which were resuspended in 1 ml of hypotonic fluorochrome buffer (0.1% Triton X-100, 0.1% sodium citrate) containing 25 μg/ml RNase A and 50 μg/ml propidium iodide (PI) (Sigma). After incubation at 4°C in dark for 1 h, samples were analyzed by EPICS XL-MCL flow cytometer (Beckman Coulter, Miami, FL).  2.16 In Vivo Matrigel Plug Assay and Immunofluorescent Staining The matrigel plug assay was conducted as previous (Wani et al, 2011). Melanoma cell lines were transiently transfected with shKAI1 or miING4 for 48h, and transiently transfected with Flag-KAI1 or HA-ING4 for 24h. Six groups of cells (Ctrl/miCtrl, KAI1/miCtrl,  42 KAI1/miING4, shCtrl/HA, shKAI1/HA, and shKAI1/HA-ING4) containing 4 × 105 cells were suspended in 100 μl of phosphate-buffered saline and mixed with 300 μl of matrigel. The cell-containing matrigel was implanted subcutaneously into the flanks of a 5-week-old male severe combined immunodeficiency mouse (three mice per group). The severe combined immunodeficiency (SCID) mouse is characterized by its inability to initiate or sustain an appropriate immune response that is usually due to the absence or abnormal T and B lymphocytes. The mice were sacrificed after 10 days of injection and the implanted matrigel plugs were excised, photographed immediately and embedded in Tissue-Tek® O.C.T. compound (Sakura Finetek, Torrance,k CA) stored at -80°C for 1h. The frozen samples were cut into 5 μm sections using the CM1850 cryostat (Leica Microsystems, Concord, Ontario, Canada) and secured on glass slides. The slides were stained with anti-mouse CD31, anti-human IL-6 and anti-human VEGF antibodies (Santa Cruz). 4’,6-diamidino-2-phenylindole (DAPI, Sigma) was used to indicate the overall cell density of each matrigel plug. Confocal images were taken using Zeizz AIM software on the Zeiss LSM 510 confocal microscope (Zeiss, Toronto, Ontario, Canada) with excitation at 488 nm (fluorescein isothiocyanate) and 340 nm (DAPI). Relative fluorescent intensity was analyzed using ImageJ software (NIH, Bethesda, MD).        43 Chapter 3: Prognostic Significance of KAI1/CD82 in Human Melanoma and Its Role in Cell Migration and Invasion through the Regulation of ING4 3.1 Background and Rationale Cutaneous melanoma is a type of skin cancer that results from the abnormal proliferation of epidermal melanocytes (Houghton & Polsky, 2002). Its occurrence has increased steadily for the past several decades and continues to be one of the most deadly cutaneous malignancies (Siegel et al, 2014). In 2014, there were over one million melanoma patients in the USA, and more than 76,100 new cases were expected to be diagnosed, of which 9,710 of them may die (American Cancer Society, 2015). Surgical excision is normally used to treat early melanomas; but around 20% of the patients will progress to metastatic melanoma (Balch et al, 2001). Furthermore, melanoma is also very likely to metastasize since it has a very high capability to invade to nearby organs (Houghton & Polsky, 2002). In fact, patients with metastatic melanoma have a very poor prognosis with an average survival of only 8 months (Jemal et al, 2010). Therefore, the 5 year survival rate for melanoma patients is less than 5% if metastasis has occurred (Tsao & Sober, 2005). However, after years of laboratory and clinical studies, there are still few effective treatments of the disease due to its ability to resist conventional cancer therapies (Soengas & Lowe, 2003). Furthermore, the fundamental mechanisms driving melanoma progression and metastasis are still not well recognized (Marquez-Rodas et al, 2011).  KAI1 was originally identified as a putative metastasis suppressor in prostate cancer and it maps on human chromosome 11p11.2 (Dong et al, 1995). KAI1 belongs to the transmembrane 4 superfamily since it has 4 membrane-spanning segments and a single  44 extracellular domain that makes up a 267 amino acid protein (Wright & Tomlinson, 1994). Furthermore, KAI1 mRNA level and protein expression are frequently down-regulated in many types of metastatic cancer (Dong et al, 1996). It has also been reported that KAI1 was able to regulate c-Met and Src kinase to suppress integrin-induced invasion (Sridhar & Miranti, 2006). However, the underlying mechanism of KAI1 regulation is still unclear.  In order to further investigate the role of KAI1 in melanoma progression and prognosis, we used tissue microarray and immunohistochemistry to evaluate KAI1 expression in different stages of human melanocytic lesions. We also performed in-vitro functional and mechanistic studies of KAI1 to investigate its role in melanoma migration, invasion and regulation.   3.2 Results 3.2.1 Reduced KAI1 Expression Was Correlated With Melanoma Progression Membrane and cytoplasm staining were used for the observation of KAI1. As seen in Figure 3.1, different intensities of staining were observed in nevi and melanoma biopsies with a decrease in staining intensity observed for metastatic melanoma (Figure 3.1a-h). In particular, strong KAI1 staining decreased from 71% in common acquired nevi to 60% in dysplastic nevi.  A further decrease was observed for the primary (38%) and metastatic melanomas (10%) (Figure 3.1i). The p-values used to compare staining intensity between different groups were all less than 0.05, which represented significant differences in staining among different stages of melanocytic lesions. In particular, we found significant differences in KAI1 staining between dysplastic nevi and primary melanoma (P = 1.8x10-4) and between primary melanoma and metastatic melanoma (P = 9.4x10-15). In order to show KAI1  45 expression was downregulated in multiple human melanoma cell lines, we performed the western blot analysis and showed that KAI1 expression was reduced in seven additional melanoma cell lines compared to the positive control, the human melanocyte (Figure 3.2). Therefore, these data suggested that reduced KAI1 expression was a key factor during melanoma progression. According to the results from staining and clinicopathologic characteristics of melanoma patient samples shown in Table 3.1, KAI1 staining was not correlated with age, gender, anatomic sites, and tumor subtype in primary melanoma. However, KAI1 staining was significantly correlated with tumor thickness and ulceration. Furthermore, a strong KAI1 staining was observed in 47% of the cases, of which the tumor thickness of the samples were less than or equal to 2mm while only 29% of the cases with tumors thicker than 2mm (P = 0.035, χ2 test). Furthermore, a strong KAI1 staining was observed in 44% of the patients without ulceration and only 18% of whom with ulceration (P = 0.001, χ2 test) (Table 3.1).   46  Figure 3.1 KAI1 expression was reduced in advanced human melanoma  (a-h) Representative images of KAI1 immunohistochemical staining in human melanocytic lesions. (a) and (e), Strong KAI1 staining in CAN; (b) and (f), moderate KAI1 staining in DN; (c) and (f), weak KAI1 staining PM; D and H, negative KAI1 staining in MM. Bar = 50 μm. (i) KAI1 expression is significantly reduced when comparing common acquired nevi and dysplastic nevi (P = 0.04, χ2 test), dysplastic nevi and primary melanoma (P = 1.8×10-4, χ2 test), and primary melanoma and metastatic melanoma (P = 9.4×10-15, χ2 test). CAN, common acquired nevi; DN, dysplastic nevi; PM, primary melanoma; MM, metastatic melanoma.    47  Figure 3.2 KAI1 expression level was downregulated in multiple human melanoma cell lines compare to the melanocyte Western blot was performed to analyze the expression of KAI1 in different human melanoma cell lines. The human melanocytes was the positive control.    48 Table 3.1 KAI1 Staining and Clinicopathologic Characteristics of Melanoma Patients Variables KAI1 Staining    Weak,    No. (%)    Strong,    No. (%) Total Pa Primary melanoma         Age, y             ≤60 70 (56.5) 54 (43.5) 124 0.090       >60 93 (67.4) 45 (32.6) 138     Sex             Male 90 (62.5) 54 (37.5) 144 0.975       Female 73 (61.9) 45 (38.1) 118     Tumor thickness, mm           ≤1 26 (54.2) 22 (45.8) 48 0.035       1.01-2.00 40 (51.9) 37 (48.1) 77        2.01-4.00 38 (65.5) 20 (34.5) 58        >4.00 56 (74.7) 19 (25.3) 75     Ulceration          Absent 114 (56.4) 88 (43.6) 202 0.001       Present 49 (81.7) 11 (18.3) 60     Subtype             Lentigo maligna 24 (66.7) 12 (33.3) 36 0.490b       Superficial spreading 53 (55.2) 43 (44.8) 96        Nodular 34 (66.7) 17 (33.3) 51        Unspecified 52 (65.8) 27 (34.2) 79     Sitec             Sun-protected 126 (62.4) 76 (37.6) 202 0.956       Sun-exposed 37 (61.7) 23 (38.3) 60  Metastatic melanoma        Age, y             ≤59 75 (98.7) 1 (1.30) 76 0.096       >59 70 (92.1) 9 (11.4) 79     Sex             Male 100 (98.0) 2 (2.00) 102 0.494       Female 52 (98.1) 1 (1.89) 53     AJCC stage            I 66 (52.0) 61 (48.0) 127 1.39x10-15       II 97 (71.9) 38 (28.1) 135        III 62 (95.4) 3 (4.55) 65        IV 86 (95.6) 4 (4.40) 90  AJCC indicates American Joint Committee on Cancer. a Chi-square test. b Comparison of the lentigo maligna group with all the other groups. c Sun-protected sites: truck, arm, leg and feet; sun-exposed sites: head and neck.  49 3.2.2 Decreased KAI1 Eexpression Correlated with a Worse Patient Survival In order to investigate the correlation between KAI1 expression and melanoma patient survival, Kaplan-Meier survival curves were used. According to Figure 4.3a, a reduced KAI1 expression significantly correlated with a poorer 5-year overall (P = 1.2 × 10-7, χ2 test) and disease-specific survival (P = 3.6 × 10-7, χ2 test) of all melanoma patients. Furthermore, the 5-year disease-specific survival rate dropped from 79% for the patients with strong KAI1 expression to 50% for those with negative KAI1 expression (Figure 4.3a). In primary melanoma patients (n = 262), KAI1 expression was also closely associated with overall (P = 0.003) and disease-specific 5-year survival (P = 0.007) (Figure 4.3b). Similar pattern was also observed in metastatic melanoma patients (n = 155) where reduced KAI1 expression significantly associated with a poorer overall (P = 0.012) and disease-specific 5-year survival (P = 0.015) (Figure 4.3c). In order to estimate the crude hazard ratios (HRs) of KAI1 expression and clinicopathological variables on patient survival, the univariate Cox proportional regression analysis was used. In primary melanoma patients, age of the patient, tumor thickness, ulceration, and KAI1 expression were significantly associated with both overall and disease-specific 5-year survival (Table 4.2). Additionally, KAI1 staining was also significantly associated with 5-year overall and disease-specific survival in metastatic melanoma patients (Table 4.2). In order to determine KAI1 expression as an independent marker for melanoma prognosis, a multivariate Cox regression analysis was used. As shown in Table 4.3, KAI1 expression was indeed an independent prognostic factor for the 5-year overall (HR, 0.59; 95% CI, 0.36-0.96; P = 0.029) and disease-specific (HR, 0.60; 95% CI, 0.37-0.97; P = 0.037) survival of the Melanoma patients.   50  Figure 3.3 KAI1 expression is associated with 5-year survival of melanoma patients 262 primary melanoma patients and 155 metastatic melanoma patients were analyzed. (a) Reduced KAI1 expression is correlated with poor overall and disease-specific 5-year survival in all melanoma patients (P = 1.2×10-7 and 3.6×10-7, respectively, log-rank test). (b) Reduced KAI1 expression is correlated with poor overall and disease-specific 5-year survival in primary melanoma patients (P = 0.003 and 0.007, respectively, log-rank test). (c) Reduced KAI1 expression is correlated with poor overall and disease-specific 5-year survival in  51 metastatic melanoma patients (P = 0.012 and 0.015, respectively, log-rank test). Cum., cumulative.    52 Table 3.2 Univariate Cox Proportional Regression Analysis on 5-year Overall and Disease-Specific Survival of 262 Primary and 155 Metastatic Melanoma Patients Variable    Patients, Overall Survival Disease-Specific Survival     No. (%) Deaths Deaths Rate, % HR  (95% Cl) P Deaths Deaths Rate, % HR  (95% Cl) P Primary melanoma             Age, y                ≤60     106 (40.5) 16 15.1 1.00 0.001 15 14.2 1.00 0.001      >60     156 (59.5) 54 34.6 2.61 (1.50-4.57)  52 33.3 2.69 (1.51-4.78)     Sex                 Male     144 (55.0) 36 25.0 1.00 0.482 36 25.0 1.00 0.758       Female     118 (45.0) 34 28.8 1.18 (0.74-1.89)  31 26.3 1.08 (0.67-1.74)     Tumor thickness, mm               ≤2.0     141 (53.8) 19 13.5 1.00 <.001 18 12.8 1.00 <.001       >2.0     121 (46.2) 51 42.1 3.87 (2.28-6.56)  49 40.5 3.93 (2.29-6.75)     Ulceration                Absent 202 (77.1) 37 18.3 1.00 <.001 36 17.8 1.00 <.001       Present 60 (22.9) 33 55.0 0.25 (0.16-0.40)  31 51.7 0.26 (0.16-0.42)     Sitea                 Sun-protected 155 (59.2) 43 27.7 1.00 0.538 40 25.8 1.00 0.752       Sun-exposed 107 (40.8) 27 25.2 0.86 (0.53-1.39)  27 25.2 0.92 (0.57-1.51)     KAI1                 Weak 163 (62.2) 54 33.1 1.00 0.004 51 31.3 1.00 0.008       Strong 99 (37.8) 16 16.2 0.44 (0.25-0.78)  16 16.2 0.47 (0.27-0.82)             Metastatic melanoma            Age, y                 ≤59 76 (49.0) 55 72.4 1.00 0.969 54 71.1 1.00 0.965       >59 79 (51.0) 57 72.2 1.00 (0.70-1.46)  55 69.6 0.99 (0.68-1.44)     Sex                 Male 102 (65.8) 72 70.6 1.00 0.296 69 67.6 1.00 0.215       Female 53 (34.2) 40 75.5 1.23 (0.84-1.81)  40 75.5 1.28 (0.87-1.89)     KAI1                 Weak 140 (90.3) 105 75.0 1.00 0.016 103 73.6 1.00 0.019       Strong 15 (9.70) 6 40.0 0.36 (0.16-0.83)  6 40.0 0.37 (0.16-0.85)  HR indicates hazard ratio; Cl, confidence interval. a Sun-protected sites: truck, arm, leg and feet; sun-exposed sites: head and neck.    53 Table 3.3 Multivariate Cox Regression Analysis on 5-year Overall and Disease-Specific Survival of all Melanoma Patients Variables Overall Survival Disease-Specific Survival  βa SE HR 95% Cl Pb βa SE HR 95% Cl Pb    Age 0.322 0.153 1.38 1.02-1.86 0.035 0.342 0.154 1.41 1.04-1.90 0.027    Sex 0.209 0.152 1.23 0.91-1.66 0.170 0.214 0.154 1.24 0.92-1.67 0.163    AJCC 1.387 0.167 4.00 2.88-5.55 1.19x0-16 1.389 0.169 4.01 2.88-5.58 1.97x10-16    KAI1 -0.534 0.245 0.59 0.36-0.96 0.029 -0.513 0.246 0.60 0.37-0.97 0.037 SE indicates standard error of β; HR, hazard ratio; Cl, confidence interval. a β: regression coefficient. b Coding of variables was as follows. Age was coded as 1, ≤60 years, and 2, >60. Gender was coded as 1, male; and 2, female. AJCC stage was coded as 1, stage I-II; and 2, stage III-IV. KAI1 expression was coded as 1, weak (0-6); and 2, strong (8-12).   3.2.3 KAI1 Expression Reduced Melanoma Cell Migration and Stress Fibre Formation As cancer cell migration was very important in tumor metastasis, the role of KAI1 in melanoma cell migration was investigated using wound healing assay. We transfected MMRU cells with Flag-KAI1 or control plasmids and the gaps were photographed by microscopy. As shown in Figure 4.4b, KAI1 significantly inhibited MMRU cell migration. Conversely, when KAI1 was knocked down, cell migration was significantly enhanced (Figure 4.3e). The observed effect on cell migration was not due to cell proliferation because the cells were treated with mitomycin C for 2 hours to inhibit cell growth. Furthermore, previous report showed that ROCK activity and F-actin stress fibre formation were closely related to cell migration (Gao et al, 2006). In order to further investigate the role of KAI1 on melanoma cell migration, we performed stress fibre formation assay. KAI1 was overexpressed or knocked down in MMRU cells and the cells were starved overnight with serum depletion. Cells were then stimulated with complete serum for 30min before subjected to rhodamin-conjugated phalloidin staining to show the stress fiber formation (Figure 4.5a and c). The F-actin staining representing the stress fiber formation was reduced at least 50%  54 when KAI1 was overexpressed (Figure 4.5b). Conversely, the F-actin staining was increased by 2-fold in cells with KAI1 knockdown (KD) (Figure 4.5d). However, when Rock inhibitor Y27632 was used to treat the cells transfected with control siRNA or KAI1 siRNA, the formation of stress fiber was abrogated compared to the non-treated cells (Figure 4.5c). Taken together, KAI1 was able to regulate melanoma cancer cell migration through the regulation of stress fiber formation. Also, these results suggested that ROCK activity was required during F-actin stress fiber formation.     55  Figure 3.4 KAI1 regulated melanoma cell migration (a, f) Western blot analysis of KAI1 expression. Flag antibody was used Figure 3A and KAI1 antibody was used for Figure 3F. (b, e) Representative images of the effect of KAI1 overexpression or knockdown on melanoma cell migration. Twenty-four hours after transfection with si-KAI1-1, si-KAI1-2 or control siRNA, MMRU cells were transfected with KAI1 or control plasmids for 24 h. Then the monolayer of MMRU cells was scratched. The gaps were observed by microscopy and photographed. (c, f) The migrated cells during wound healing were counted randomly in five fields of each group, and each group was repeated three times. Columns, mean; bars, standard deviation. ***P<0.001. Ctrl, control; CV, control vector; si, small interfering.    56  Figure 3.5 KAI1 regulated stress fiber formation via ROCK (a, c) The effect of KAI1 overexpression or knockdown on melanoma cell stress fiber formation. MMRU Cells were transfected with Flag-KAI1, control plasmid, si-KAI1-1, siKAI1-2, or control siRNA, respectively, followed by serum starvation overnight and serum stimulation for 30 min. For ROCK inhibitor treatment, 10 μmol/L Y27632 in serum-free medium was added to the cells after serum starvation overnight and incubated for 2 h, then the cells were incubated with complete medium containing 10% fetal bovine serum with 10 μmol/L Y27632 for 30 min. Magnification, 400×. (b, d) Quantitation of (a) and (c), respectively. RI, ROCK inhibitor Y27632. Columns, mean; bars, standard deviation. ***P<0.001.  57 3.2.4 KAI1 Regulated Cell Invasion and MMP2 Activity In order to investigate the role of KAI1 in melanoma cell invasion, the Boyden chamber invasion assay was performed. We found that forced KAI1 expression significantly inhibited cell invasion by at least 80% (Figure 4.6a and b), while KAI1 knockdown enhanced cell invasion by at least 2-fold using two independent siRNAs (Figure 4.6c and d). Furthermore, zymography assay was used to evaluate MMP-2 activity after KAI1 overexpression or KD.  As shown in Figure 4.6e, MMP-2 gelatinolytic activity was increased at least 2.5-fold in KAI1-KD cells compared to the control, whereas MMP-2 gelatinolytic activities were significantly reduced in KAI1-overexpressing cells compared to the control.      58  Figure 3.6 KAI1 regulates melanoma cell invasion and the activity of MMP-2 (a, c) The effect of KAI1 overexpression or knockdown on melanoma cell invasion using Transwell culture chamber. MMRU cells were transfected with Flag-KAI1, control plasmid, si-KAI1-1, siKAI1-2, or control siRNA, respectively. The cells were seeded on to matrigel with serum-free medium, incubated for 24 h at 37°C, stained with crystal violet and quantified. (b, d) Quantitation of (a) and (c), respectively. The experiment was done in triplicate wells. (e) KAI1 inhibits the activity of MMP2 in MMRU cells by performing the zymography assay. (f) Quantitation of (e). The experiment was repeated three times. Columns, mean; bars, standard deviation. ***P<0.001.   59 3.2.5 KAI1 as an Upstream Regulator of ING4 Previous studies showed that inhibitor of growth (ING) family proteins may have a crucial role in melanoma tumorigenesis. It was shown that reduced ING4 expression directly correlated with poor 5-year survival of melanoma patients and that ING4 also had an important role in melanoma cell migration and invasion (Li et al, 2008). Thus, we asked if KAI1 and ING4 mediated melanoma cell migration and invasion in the same signalling pathway. Real-time quantitative PCR assay was used to measure the expression of ING4 mRNA level after KAI1 overexpression or knockdown. As shown in Figure 4.7a, KAI1-KD inhibited the expression of ING4 mRNA, whereas KAI1 overexpression increased the mRNA level of ING4. A similar pattern was also observed at the protein level where an increase in ING4 protein expression was observed when KAI1 was overexpressed, while KAI1-KD reduced ING4 protein level (Figure 4.7b). Furthermore, ING4 was previously shown to be regulated by BRMS1 (Li & Li, 2010). Thus, ING4 may be regulated by KAI1 through the regulation of BRMS1. Furthermore, a recent study showed that phosphorylation of RelA/p65 repressed BRMS1 transcriptionally (Liu et al, 2012). Therefore, it was possible that KAI1 was able to regulate ING4 through p65. As shown in Figure 4.7b, overexpression of KAI1 inhibited the phosphorylation of p65, which resulted in the upregulation of BRMS1 and ING4 expression. Conversely, after KAI1 was knocked down, the phosphorylation of p65 was upregulated, which resulted in the suppression of BRMS1 and ING4 expression. To confirm the regulation of ING4 by KAI1, we knocked down both KAI1 and p65. As shown in Figure 4.7c, ING4 expression was upregulated after p65 KD even when KAI1 was also knocked down. To further investigate how KAI1 regulated melanoma cell migration through ING4, wound healing assay was performed with KAI1 KD and ING4 overexpressed or vice  60 versa. As expected, there were less cells migrated into the gaps when ING4 was overexpressed after KAI1 KD compared to the control (Figure 4.7e). Conversely, KAI1-overexpressing cells with ING4-KD resulted in more cells migrated into the gaps compared to the control (Figure 4.7f). Thus, we were able to show that KAI1 was an upstream regulator of ING4 and it regulated melanoma cell migration in an ING4-dependent manner.   61  Figure 3.7 Regulation of ING4 by KAI1 and its effect on melanoma cell migration (a) Quantitative reverse transcription-PCR analysis was done by transfection cell with Flag-KAI1, control plasmid, si-KAI1-1, siKAI1-2, or control siRNA, respectively. Expression of ING4 mRNAs was measured using real-time quantitative PCR and normalized with GAPDH  62 as loading control. (b) Western blot analysis of ING4 expression after the cells were transfected with Flag-KAI1, control plasmid, si-KAI1-1, siKAI1-2, or control siRNA, respectively. (c) Western blot analysis of ING4 expression after cells were transfect with control siRNA and si-KAI1-1, si-p65 alone or together. (d) KAI1/p65/ING4 signaling pathway in melanoma migration. (e) The effect of reduced KAI1 expression on cell migration was reduced by forced ING4 expression. The cells were transfected with siKAI1 or control siRNA, then after 48 hours, the cells were transfected with HA-ING4 or HA control plasmid. Monolayer MMRU cells were scratched and the gap was monitored by microscopy and photographed. (f) The reduction in cell migration by the forced KAI1 expression was compensated by ING4 knockdown. The cells were transfected with KAI1 or control plasmid, then after 24 h, the cells were transfected with mi-ING4 or control plasmid. (g, h) Quantitation of (e) and (f), respectively. The migrated cells during wound healing were counted randomly in five fields of each group, and each group was repeated three times. Columns, mean; bars, standard deviation. ***P<0.001. si, small interfering; Ctrl, control; HA, hemagglutnin; mi, microRNA.    63 3.3 Discussion KAI1 was originally recognized as a metastasis suppressor for prostate cancer. However, many evidences had shown that during the progression of many solid tumors, KAI1 was considered a wide-spectrum invasion and metastasis suppressor (Dong et al, 1995; Hemler, 2001; Stipp et al, 2003). In this study we examined 38 common acquired nevi, 78 dysplastic nevi, 262 primary melanoma, and 155 metastatic melanoma cases using the TMA analysis. We also investigated the role of KAI1 during melanoma cell migration and invasion since these steps are critical in cancer progression. Our TMA staining revealed that KAI1 staining was mainly in the cytoplasm (Fig. 1A), which was consistent with other reports (Guo et al, 2000; Rotterud et al, 2007). According to the TMA staining, statistical analysis revealed a significant reduction in cytoplasmic KAI1 expression in metastatic melanoma compared to other three stages of melanocytic lesions according to the tissue microarray immunoreactivity scores. Since there were significant differences in staining intensity between dysplastic nevi, primary melanoma, and metastatic melanoma, reduced KAI1 expression was a key factor during melanoma progression and metastasis. Our data also showed that reduced KAI1 expression was significantly correlated with tumor thickness, AJCC stage, age, and patient survival. Thus, KAI1 expression was involved in both the tumor growth and melanoma metastasis. We also showed a significant correlation between cytoplasmic KAI1 staining with the 5-year disease-specific and overall survival of melanoma patients, of which higher KAI1 staining resulted in a better patient survival (Figure 3.2). Similar results had also been found in non-small cell lung cancer (Adachi et al, 1996) and prostate cancer (Liu et al, 2011a) where KAI1 staining was significantly correlated with the survival of the patients. Furthermore, KAI1 was able to significantly suppress melanoma cell migration (Figure 3.3)  64 and cell invasion (Figure 3.5), which were considered to be important events when cancer progressed to metastasis (Gupta & Massague, 2006). Therefore, our data suggested that KAI1 was an effective melanoma tumor suppressor.  Previous study showed that KAI1 was able to inhibit melanoma cell migration by attenuating the integrin signalling such as Crk or Rho small GTPases signalling pathways (Tsuda et al, 2004; Zhang et al, 2003a). In particular, the RhoA-ROCK signalling pathway was able to regulate actin cytoskeleton during the process of stress fiber formation (Kiss et al, 1997). In our study, we showed a significant decrease in F-actin formation when KAI1 was overexpressed, whereas a significant increase in the stress fiber formation when KAI1 was knocked down (Figure 3.4). Stress fiber formation was considered to be a very important event during cell motility and metastasis (Yamazaki et al, 2005), and the regulation of KAI1 on stress fiber formation revealed that KAI1 was an very effective metastasis suppressor. In order to show that ROCK was required for the regulation on stress fiber formation by KAI1, we treated the cell with the ROCK-inhibitor, Y27632.  As shown in Figure 3.4c, the use of ROCK inhibitor abrogated the upregulation of F-actin formation caused by KAI1-KD cell and resulted in a significant reduction on the F-actin stress fiber formation. As a result, this revealed a close involvement of KAI1 during RhoA/ROCK-mediated stress fiber formation and more importantly the key function of KAI1 in the regulation of melanoma cell migration.  Not only has KAI1 suppressed melanoma cell migration, it has also been shown to have an important role in melanoma cell invasion (Liu et al, 2011b). As shown by Liu et al., a reduction of cell invasion after KAI1 overexpression was observed in pancreatic cancer cells. In our study, we also showed a decrease in cell invasion after forced KAI1 expression in human melanoma cell lines (Figure 4.5). Furthermore, it had been shown that MMP-2 was  65 closely related to cancer cell invasion (Zheng et al, 2006). According to our previous immunohistochemistry results, MMP-2 positive cells were also significantly increased in different stages of melanocytic lesions (Vaisanen et al, 1996). Thus, it was important to show that KAI1 was able to regulate MMP-2 activity. By using zymography assay, we found that KAI1 overexpression lead to a reduction on gelatinolytic activity of MMP-2. Conversely, the activity of MMP-2 was upregulated when KAI1 was knocked down compared to the control (Figure 3.6). Since MMP-2 activity was representative of cancer cell invasion, it should correlate directly with tumor thickness. This was consistent with the staining pattern of KAI1 that showed a negative correlation between KAI1 expression and tumor thickness (Table 3.1). Additionally, primary melanoma patients with weak KAI1 staining also had lower  5-year survival compared to those with strong KAI1 staining. We had so far established that KAI1 regulated melanoma cell migration through the regulation of stress fiber formation and regulated invasion through MMP-2 activity. In order to further investigate KAI1’s regulation on melanoma progression, we proposed a novel regulatory pathway where KAI1 was able to regulate another tumor suppressor gene referred to as inhibitor of growth (ING) 4. Previous report showed that forced ING4 expression suppressed cell colony forming efficiency, the cell population in S phase, and the inhibition on cell apoptosis (Shiseki et al, 2003). ING4 was also shown to be regulated by BRMS1 (Phadke et al, 2008). Furthermore, recent discovery also revealed that BRMS1 was transcriptionally suppressed by p65 (Liu et al, 2012). Thus, it was likely that KAI1 regulated ING4 through the regulation of p65. By using real-time PCR, we showed a significant decrease in ING4 mRNA level after KAI1-KD while an increase in ING4 mRNA was found after forced KAI1 expression (Figure 3.6a). Furthermore, we also showed that KAI1 was able  66 to regulate ING4 at the protein level according to our western blot result (Figure 3.6b). Additionally, we also showed that KAI1 suppressed the phosphorylation of p65 (Figure 3.6b). In order to confirm this effect, we knocked down both p65 and KAI1 and discovered a restoration of ING4 expression (Figure 3.6c). In order to show the importance of ING4 during KAI1’s regulation on melanoma cell migration, ING4 was overexpressed while KAI1 was knocked down. According to Figure 3.6e, a significant reduction in cell migration was observed by using the cell migration assay (Figure 3.6e). Conversely, a significant increase in cell migration was observed when KAI1 was overexpressed while ING4 was knocked down. Therefore, we concluded that KAI1 was an upstream regulator of ING4 where it regulated ING4 at the transcriptional level and protein level. Previous reports showed that KAI1 suppressed metastasis through the inhibition of cancer cell migration and invasion by regulating KAI1 associated proteins such as integrin and epidermal growth factor receptor (EGFR) (Lee et al, 2003; Mannion et al, 1996; Nakamura et al, 2000; Ono et al, 1999; Zhang et al, 2001). And since EGFR was able to activate phosphoinositide-3 kinase (PI3K)/Akt pathway (Kallergi et al, 2008), which then triggered the activation of p65 (Madrid et al, 2001), it was possible that KAI1 regulated p65 through EGFR.   Originally, KAI1 was identified as a tumor suppressor in prostate cancer where it had been shown to have the ability to inhibit cell motility and invasiveness (Wright & Tomlinson, 1994). Since then, KAI1 had been reported to be a putative marker for metastatic potential in human bladder cancer (Yu et al, 1997). It was then found in breast cancer (Yang et al, 1997), non-small cell lung cancer (Tokuhara et al, 2001), and ovarian carcinoma (Liu et al, 2000). KAI1 had very important function in various signal transduction pathways; and it was rarely mutated and frequently down-regulated. It has been showed that various transcription factors  67 such as β-catenin, Reptin (Kim et al, 2005), NcoR, TAB2, and HDAC3 (Baek et al, 2002). In addition to this transcriptional regulation, KAI1 was also regulated by ubiquitin ligases. Previous report showed that KAI1 was targeted for degradation by an ubiquitin ligase referred to as gp78, which promoted sarcoma metastasis (Tsai et al, 2007). Therefore, the ubiquitin pathway could be potentially targeted for cancer treatment. Furthermore, the utilization of various proteasome inhibitors could also be used as a clinical therapeutic approach. Finally, according to the multivariate Cox regression analysis, decreased KAI1 expression was statistically significant as well as independent of other clinical parameters including age, gender and AJCC stages, which highlighted the importance of KAI1 in melanoma prognosis.  Our study revealed that the reduced KAI1 expression was significantly correlated the poorer 5-year patient survival in advanced melanomas. KAI1’s ability to suppress melanoma cell migration and invasion also demonstrated its important function with regard to melanoma patient survival and tumor progression. Therefore, KAI1 could be used as a promising prognostic marker and a potential therapeutic target for treating malignant melanoma.      68 Chapter 4: The Role of the Metastasis Suppressor gene KAI1 in Melanoma Angiogenesis 4.1 Background and Rationale In order for cancer cells to proliferate, migrate, invade, as well as metastasize, an adequate supply of nutrients and oxygen as well as the removal of toxic and waste products are essential and compulsory (Folkman, 2002). The ability for cancer to spread to distant organs makes it especially life threatening; and metastasized tumor can result in a significant reduction on patient survival. For cancer cells to metastasize, they need to penetrate through the blood vessels and travel across the intravascular system to arrive at a new location where they are permitted to proliferate (Folkman, 1974). As vascular network growth is vital during cancer progression, the study of blood vessel formation is especially important. Furthermore, since angiogenesis is crucial during proliferation and metastasis of many solid tumors, regulating and manipulating cancer angiogenesis is a powerful tactic against cancer progression (Qin et al, 2012; Weis & Cheresh, 2011). Therefore, identifying and manipulating various angiogenesis factors and their activators are both crucial and compulsory for the exploration of cancer therapeutics. Angiogenesis in melanoma further increased the lethality of melanoma and dramatically decreased patient survival (Mahabeleshwar & Byzova, 2007). Especially during the vertical growth phase of melanoma, high antigenic activity was crucial during melanoma metastasis (Elder et al, 1996). Previously KAI1 was shown to downregulate melanoma cell migration and invasion. (Tang et al, 2014b). Yet, its role in melanoma angiogenesis was still unclear.  In order to investigate the KAI1’s ability to regulate melanoma angiogenesis, we performed human vascular endothelial cell (HUVEC) growth assay and tubular structure  69 formation assay. Furthermore, since KAI1 regulated ING4 during melanoma cell migration and invasion, it was also likely that KAI1’s regulation on ING4 was also evident during melanoma angiogenesis. Finally, we also performed in vivo matrigel plug assay to study KAI1’s role in angiogenesis and neovessel formation in a mouse model. Thus, these findings highlighted the importance of KAI1 in melanoma angiogenesis and proposed KAI1 as a potential therapeutic target for melanoma treatment.   4.2 Results 4.2.1 KAI1 Expression in Melanoma Cells Suppressed the Growth and Tubular Structure Formation of HUVECs KAI1 was overexpressed or knocked down in MMRU melanoma cell line with the transfection efficiency of 75% estimated according to the pEGFP-N1 plasmid transfection (Wang et al, 2006). The western blot results illustrated the KAI1 expression after KAI1 overexpression and knockdown (Figure 4.1a). The conditioned media (CM) was then collected and used for HUVEC growth assay or tubular formation assay. As shown in Figure 4.1b, KAI1 overexpression reduced HUVEC growth by about 50% percent after 24h and 48h compared to the control. Conversely, KAI1-KD increased HUVEC growth by about 2-fold after 24h and 48h (Figure 4.1b). Furthermore, the average number of tubular structure formations of HUVEC was reduced by 75% in CM with KAI1 overexpression compared to the control. And tube formation was increased by about 2-fold in CM with KAI1-KD compared to the control (Figure 4.1c, d). This experiment was also repeated using two more melanoma cell lines, Mel624 and A375, where similar results were obtained (Figure 4.2). In order to show that KAI1’s regulation on HUVEC growth was not due to apoptosis of  70 HUVEC, we performed the Fluorescence-activated cell sorting (FACS) analysis. As shown in Figure 4.3a, HUVEC growing in KAI1-overexpressed CM had about the same amount of subG1 population.  The same was observed with KAI1-KD (Figure 4.3a). Furthermore, in the KAI1 overexpression group, there was ignificantly more G1 population and had significantly less S/G2 population compared to the control (Figure 4.3a top). Conversely, HUVEC growing in shKAI1-expressed CM had a significantly less G1 population and more S/G2 population compared to the control (Figure 4.3a bottom). The quantification was shown as bar graphs in Figure 4.3b. Therefore, KAI expression did not cause HUVEC apoptosis but only affect the rate of the HUVEC growth.     71  Figure 4.1 KAI1 expression in melanoma cells inhibited HUVEC growth and tube formation MMRU cells were transiently transfected with Flag-KAI1 to overexpression KAI1 and shKAI1 to silence KAI1 expression, and transfected with control vectors (CV) or scrambled shRNA (shCtrl) for the controls. The conditioned medium (CM) collected was used for HUVEC growth or tube formation assay. (a) KAI1 protein expression of forced KAI1 expression (top) and KAI1 knockdown (bottom) in MMRU cells was confirmed using Western blot. (b) the CM was prepared from MMRU cells transfected with KAI1, shKAI1 or respective controls for the study of HUVEC growth after 1, 2 and 3 days. Data was presented as means ± SD from 3 independent experiments. (c) the CM was prepared from MMRU cells  72 transfected with KAI1, shKAI1 or respective controls for detection of the tube formation of HUVECs. (d) the number of tubes formed in each field was counted in 5 random fields. **, P < 0.01; ***, P < 0.001; Student’s t test.     73  Figure 4.2 KAI1 expression in Mel624 and A375 cell lines inhibited tube formation Mel624 and A375 cell lines were transiently transfected with Flag-KAI1 to overexpression KAI1 and shKAI1 to silence KAI1 expression, and transfected with control vectors (CV) or scrambled shRNA (shCtrl) for the controls. The conditioned medium (CM) collected was used for tube formation assay. (a-b) KAI1 protein expression of forced KAI1 expression (top) and KAI1 knockdown (bottom) in Mel624 (a) and in A375 (b) was confirmed using Western blot. (c-d) the CM was prepared from Mel624 (c) or A375 (d) transfected with  74 KAI1, shKAI1 or respective controls for tube formation assay. (e-f) the number of tubes formed in each field was counted in 5 random fields. **, P < 0.01; ***, P < 0.001; Student’s t test.     75  Figure 4.3 Effect of KAI1 on cell cycle MMRU cells were transiently transfected with Flag-KAI1 to overexpression KAI1 and shKAI1 to silence KAI1 expression, and transfected with control vectors (CV) or scrambled shRNA (shCtrl) for the controls. HUVECs were stimulated to enter cell cycle using CM collected. (a) cell distribution into the cell cycle phase subG1, G1, S, and G2/M. (b) percentage of the cell number in each cell cycle phase.  **, P < 0.01; Student’s t test.   76 4.2.2 KAI1 Regulated HUVEC Growth through the Expression of IL-6 and VEGF According to our previous finding, the upregulation of IL-6 and VEGF was directly resulted from the increase in the regulation of angiogenesis (Wani et al, 2011).  Therefore, we then investigated KAI1’s regulation on HUVEC growth through the regulation of IL-6 and VEGF. According to the real-time RT-PCR results, KAI1 overexpression reduced IL-6 mRNA by 50% whereas KAI1-KD increased IL-6 mRNA by 2-fold compared to controls (Figure 4.4a). The ELISA assay was then used to investigate IL-6 protein level. As shown in Figure 4.4b, KAI1 overexpression reduced IL-6 protein level in CM by 50% whereas KAI1-KD increased IL-6 protein level in CM for more than 3-fold in MMRU melanoma cell line.  Moreover, in order to show that KAI1 was able to regulate IL-6 during HUVEC growth, IL-6 induction and IL-6 antibody blocking treatments were introduced to the HUVEC. As shown in Figure 4.2c, the addition of 0.8 ng/mL of recombinant IL-6 to the CM obtained from MMRU with KAI1 overexpression reduced the growth of HUVEC after 24h and 48h. Conversely, the addition of IL-6 antibody in the CM abolished the increased HUVEC growth due to KAI1-KD after 24h and 48h (Figure 4.4d). In addition to IL-6, VEGF was also an important factor for angiogenesis. In Figure 2e, we were able to show that KAI1 overexpression suppressed the mRNA level of VEGF whereas KAI1-KD increased the mRNA level in MMRU cells. We also showed that VEGF protein level was also suppressed by KAI1 overexpression and upregulated by KAI1-KD according to the western blot analysis (Figure 4.4f). Furthermore, we repeated the functional analyses using VEGF recombinant protein and antibody. As shown in Figure 4.2g, the addition of recombinant VEGF (0.10mg/mL) increased HUVEC growth in the CM obtained from MMRU cells with forced  77 KAI1 expression after 24h and 48h. Conversely, the addition of 0.25 mg/mL bevacizumab, a VEGF inhibitor, suppressed HUVEC growth after 24h and 48h (Figure 4.4h).      78  Figure 4.4 KAI1 expression in melanoma cells suppressed IL-6 and VEGF MMRU cells were transfected with Flag-KAI1, shKAI1, CV or shCtrl. (a-b) KAI1 expression suppressed IL-6 expression. IL-6 mRNA and secreted protein levels in CM were determined by qRT-PCR (a) and ELISA (b). (c) CM was prepared from MMRU cells transfected with KAI1 or control vector. Addition of IL-6 recombinant protein (0.8 ng/mL) to the CM rescued the HUVEC growth after 24h and 48h. (d) CM was prepared from MMRU cells transfected with shKAI1 or scrambled shRNA. Introduction of IL-6 antibody (320 ng/mL) in the CM suppressed the HUVEC growth after 24h and 48h. (e) VEGF mRNA level in MMRU after KAI1 overexpression or KD was determined by qRT-PCR. (f) Western blot  79 analysis of MMRU whole cell lysate showed that KAI1 suppressed VEGF protein expression. (g) the CM was prepared from MMRU cells transfected with KAI1 or control vector. Addition of VEGF recombinant protein (0.10mg/mL) to the CM rescued the HUVEC growth after 24h and 48h. (h) the CM was prepared from MMRU cells transfected with shKAI1 or scrambled shRNA. Introduction of bevacizumab (0.25 mg/mL) in the CM suppressed the HUVEC growth after 24h and 48h. **, P < 0.01; ***, P < 0.001; Student’s t test.  4.2.3 KAI1 Suppresses Melanoma Angiogenesis through ING4 The tumor suppressor ING4 protein was previously shown to also inhibit angiogenesis (Li & Li, 2010). And since ING4 was regulated by KAI1, it was likely that KAI1 was able to regulate melanoma angiogenesis through the regulation of ING4. As seen inn Figure 4.5a, when ING4 was knocked down while KAI1 was overexpressed in MMRU cells, the rate of proliferation of HUVEC was significantly increased. Conversely, when ING4 was overexpressed and KAI1 was knocked down, the average HUVEC growth was decreased (Figure 4.5b). We further investigated this regulation through the study of tubular structure formation assay. According to Figure 4.5c, the number of tubes formed was increased when ING4 was knocked down and KAI1 was overexpressed. And the opposite was observed where tube formation was decreased when ING4 was overexpressed and KAI1 was knocked down (Figure 4.5d). Therefore, ING4 expression was required during KAI1’s regulation on melanoma angiogenesis.  Since ING4 was a downstream target of KAI1 and that KAI1 was able to regulate angiogenesis through the regulation of IL-6 and VEGF, we investigated KAI1’s regulation  80 on IL-6 and VEGF in association with ING4. As shown in Figure 4.5b, when KAI1 was overexpressed while ING4 was knocked down in MMRU cells, the IL-6 mRNA level increased about 2-fold compared to the control and more than 4 folds of increase was observed if compared to KAI1 overexpression group. Conversely, a 2-fold reduction in IL-6 mRNA level was observed with ING4 overexpression and KAI1 KD compared to control (Figure 4.5b). Furthermore, we also investigated the IL-6 protein level in the CM, where a similar pattern was obtained (Figure 4.5c-d). Finally, experiments were repeated for the detection of VEGF mRNA and VEGF protein levels. And according to Figure 4.5e and g, ING4 KD was able to rescue the reduction of VEGF mRNA and protein expression caused by KAI1 overexpression. Conversely, ING4 overexpression was able to inhibit the upregulation of VEGF mRNA and protein level caused by KAI1 KD (Figure 4.5g-f).      81  Figure 4.5 KAI1 suppressed HUVEC growth and tube formation through the regulation of ING4 (a-d) MMRU cells were transfected with Flag-KAI1, shKAI1, or in combination with HA-ING4 or miING4. CV, miRNA control (miCtrl), shCtrl, and HA control (HA) were the corresponding contros. The CM was collected for the study of HUVEC growth (a and b) and tube formation assay (c and d). (e and f) the number of tubes formed per field was counted in 5 random fields. **, P < 0.01; ***, P < 0.001; Student’s t test.     82  Figure 4.6 KAI1 expression in melanoma cells suppressed IL-6 and VEGF through the regulation of ING4 MMRU cells were transfected with Flag-KAI1, shKAI1, or in combination with HA-ING4 or miING4. CV, miCtrl, shCtrl, and HA were the corresponding controls. (a-d) KAI1 expression suppresses IL-6 expression through ING4. IL-6 mRNA levels were determined by qRT-PCR (a and b) and secreted protein levels in CM ELISA (b and d). (e-g) KAI1,  83 through the regulation of ING4, suppressed VEGF mRNA level determined by qRT-PCR (e and f) and VEGF protein expression according to Western blot analysis (g). The quantification of western blots was analyzed by ImageJ (NIH, Bethesda, MD, USA). ***, P < 0.001; Student’s t test.  4.2.4 KAI1 Expression in Melanoma Cells Inhibited Angiogenesis in vivo In order to show that KAI1 was able to regulate neovessel formation in a mouse model, we performed in vivo matrigel plug assay using MMRU melanoma cell line. The mRNA level of KAI1 and ING4 in the matrigel plugs was confirmed using qRT-PCR (Figure 4.7). Based on visual examinations, KAI1 overexpression resulted in significantly less vascularization in the matrigel plugs compared to the control while KAI1 KD caused significantly more blood vessel formation (Figure 4.8a). Furthermore, KAI1’s regulation on ING4 during angiogenesis was also observed in the matrigel plug assay. As seen Figure 4.8a, ING4 KD was able to rescue the inhibition of blood vessel formation triggered by KAI1 overexpression, whereas ING4 overexpression inhibited the upregulation of blood vessel formation due to KAI1 KD. As seen Figure 4.8b and c, the average staining intensity of CD31 positive cells in KAI1 KD plugs was about 3 folds less than the control plugs according to the immunofluorescence analysis. Conversely, a 3-fold increase in staining intensity was observed when KAI1 was knocked down (Figure 4.8b and c). Also, the reduction in the staining of IL-6 and VEGF in the matrigel plugs further confirmed that KAI1 was able to suppress angiogenesis by downregulating the expression of IL-6 and VEGF. (Figure 4.8b, left panel). Once again, ING4 was required for KAI1’s regulation on angiogenesis where ING4 KD restored the inhibition resulted from KAI1 overexpression,  84 while ING4 overexpression inhibited the upregulation of fluorescent intensity resulted from KAI1 KD (Figure 5b and c).    Figure 4.7 Expression of KAI1 and ING4 mRNA in matrigel plugs was determined by qRT-PCR (a) KAI1 mRNA expression in matrigel plugs. (b) ING4 mRNA expression in matrigel plugs. ***, compare to CV + miCtrl, P < 0.001; ###, compare to shCtrl + HA, P < 0.001; Student’s t test.    85  Figure 4.8 Forced KAI1 expression in MMRU melanoma cell line inhibited blood vessel formation in vivo through the regulation of ING4 MMRU cells were transfected with Flag-KAI1, shKAI1, or in combination with HA-ING4 or miING4. CV, miCtrl, shCtrl, and HA were the corresponding control vectors. (a) Photographs of matrigel plugs excised from severe combined immunodeficiency mice after 10 days of growth in vivo. (b) CD31, IL-6 and VEGF expression in matrigel plugs were examined by immunofluorescence staining. DAPI staining indicated the overall cell density in each matrigel plugs. (c) Relative fluorescent intensity of CD31, IL-6, and VEGF was  86 analyzed in 5 random fields for each group using ImageJ. ***, compare to first columns; ###, compare to fourth columns; P < 0.001; Student’s t test.  4.2.5 KAI1 was an Upstream Regulator of a Serine/threonine Kinase Akt We have already established that KAI1 regulates ING4 expression through RelA/p65 (Tang et al, 2014b). Since p65 was regulated by Akt (Bai et al, 2009), we next investigated if  KAI1 was able to regulate ING4 through Akt. According to the result of the western blot analysis, KAI1 overexpression was able to increase the phosphorylation of Akt (Figure 4.9a).  Previous studies revealed that Akt was regulated by EGFR (Galbaugh et al, 2006; Lu et al, 2007; Wang et al, 2000). Thus, in order to investigate KAI1’s regulation on Akt, we tested EGFR phosphorylation (Tyr 1086) level after KAI1 overexpression and KD. To further study the regulation of KAI1 on Akt, we used a potent Akt inhibitor, LY294002, and showed that it was able to rescue the inhibition on p-p65 after KAI KD (Figure 4.9b).  Furthermore, the addition of LY294002 also restored ING4 expression in the presence of KAI1 KD. Next, we investigated if Akt inhibitor was able to suppress the upregulation of IL-6 and VEGF after KAI1 KD. As seen in the RT-PCR results in Figure 4.9c, IL-6 and VEGF mRNA levels were reduced at least 4 folds with the addition of LY294002 compared to non-treated cells. Furthermore, LY294002 was also able to reduce IL-6 and VEGF after KAI1 KD (Figure 4.9c).  Finally, LY294002 was shown to inhibit VEGF expression when KAI1 was knocked down (Figure 4.9d). Therefore, we were able to conclude that KAI1 regulated angiogenesis factors IL6 and VEGF through the regulation of Akt. Taken together, we proposed a pathway for KAI1’s regulation on angiogenesis, of which KAI1 suppressed the phosphorylation of  87 Akt through the regulation of EGFR and this in turn released the inhibition on ING4 from p65, which resulted in an upregulation of IL-6 and VEGF.    Figure 4.9 KAI1 overexpression suppressed Akt phosphorylation (a) MMRU cells were transfected with Flag-KAI1, shKAI1 or respective controls. The Akt phosphorylation level, ING4 expression, and EGFR phosphorylation level were confirmed by Western blot analysis from the whole cell lysate. (b) Cells were treated with DMSO or 20μM  88 of LY294002 for 48h after transfection. LY294002 treatment was able to suppress KAI1 KD-mediated upregulation of p65 phosphorylation and rescue the KAI1 KD induced ING4 suppression. (c) LY294002 treatment in MMRU cell line was able to suppress KAI1 KD-induced upregulation of IL-6 mRNA, IL-6 protein secreted in CM, and VEGF mRNA levels. (d) LY294002 treatment in MMRU cell line suppressed KAI1 KD-induced upregulation of VEGF protein expression.   4.3 Discussion  KAI1, a cell surface glycoprotein belonged to the tetraspanin 4 superfamily, was  responsible for suppressing cancer invasion and metastasis in a number of solid tumors, such as prostate, breast, and lung cancer (Dong et al, 1995; Hemler, 2001; Ichikawa et al, 1992; Liu et al, 2001; Stipp et al, 2003). We have already shown that KAI1 was able to suppress melanoma cell migration and invasion (Tang et al, 2014b). Since angiogenesis was an important step that supplied tumor cells with nutrient and a route to metastasize after invasion, we then investigated the role of KAI1 on melanoma angiogenesis in this study. According to our data, we showed that KAI1 was able to suppress melanoma angiogenesis in vitro and in vivo by downregulating important angiogenic factors IL-6 and VEGF. Furthermore, we also showed that KAI1 regulated melanoma angiogenesis through the regulation of Akt and EGFR.   Besides proliferation, migration, and invasion, cancer cell angiogenesis was also a critical step during tumor metastasis in various types of cancer such as breast, lung, prostate, as well as melanoma (Czubayko et al, 1996; Weidner, 1995; Weidner et al, 1991). Furthermore, individual processes including endothelial cell proliferation, migration, and  89 blood vessel formation were also amongst the important steps during angiogenesis (Kalluri, 2003; Mamou et al, 2006). Thus, the inadequate suppression on angiogenesis due to KAI1 downregulation was vital during melanoma progression and metastasis. In our study, we showed that the growth and the tubular structure formation of HUVECs were inhibited when cultured in the CM collected from melanoma cells with KAI1 overexpression. Conversely, the endothelial cell growth and tubular structure formation were significantly enhanced when KAI1 was knocked down (Figure 4.1b-d). By using the in vivo matrigel plugs assay, we also showed that forced KAI1 expression in melanoma cells was able to inhibit the supportive vasculature while KAI1 KD enhanced it (Figure 4.8). It had been shown before that angiogenesis resulted in the penetration of capillaries that caused the growth of the invasive tumor and an increased in tumor thickness (Folkman, 2002). Furthermore, we had also found that KAI expression in melanoma was significantly reduced during melanoma progression, and this reduction directly resulted in an increased in melanoma tumor thickness and a poorer patient survival (Tang et al, 2014b). Thus, downregulation of KAI1 not only contributed in increased melanoma tumor invasion and metastasis, but also in melanoma angiogenesis, which directly resulted in a poor melanoma patient survival.   Previously we have found that ING4 downregulated IL-6 (Li & Li, 2010) while regulated by KAI1 (Tang et al, 2014b). Furthermore, IL-6 was also shown to regulate the expression of VEGF (Cohen et al, 1996; Hatakeyama et al, 2007; Tzeng et al, 2013). Therefore, it was likely that KAI1 suppressed melanoma angiogenesis by downregulating  IL-6 and VEGF through the regulation of ING4. According Figure 4.4, KAI1 expression was able to decrease, yet KAI1 KD was able to increase IL-6 and VEGF expression level. Furthermore, the use of recombinant IL-6 was able to rescued KAI1’s suppression on  90 HUVEC growth, while IL-6 antibody abolished KAI1’s upregulation on HUVEC growth (Figure 4.4c, d). Similar resulted were also obtained when recombinant VEGF or VEGF antibody were used (Figure 4.4g, h). Thus, we concluded that IL-6 and VEGF were required during KAI’s regulation on melanoma angiogenesis. It was shown previously that KAI was able to regulate ING4; and in order to show this regulation also applied to the regulation of melanoma angiogenesis, we overexpressed ING4 or knocked down ING4 while knocked down KAI1 and overexpressed KAI1, respectively to investigate HUVEC growth as well as IL-6 and VEGF expression. As shown Figure 4.5, our data revealed that ING4 overexpression was able to abolish the upregulation of HUVEC growth and tubular structure formation after KAI1 KD. Conversely, ING4 KD rescued the suppression of HUVEC growth and tube formation caused by forced KAI1 expression (Figure 4.5b, d). Furthermore, similar regulation of KAI1 on ING4 was also revealed during the regulation of IL-6 and VEGF expression (Figure 4.6b, d, f, and g). This effect was further confirmed according the matrigel plugs assay, of which ING4 overexpression reduced the formation of supportive vasculature in the matrigel plugs under the influence of KAI1 KD. Conversely, ING4 KD rescued the inhibition on vascularization due to KAI1 overexpression (Figure 4.8). Taken together, these results revealed that KAI1 suppressed ING4 expression and resulted in a downregulation of IL-6 and VEGF expression and a decreased proliferation of HUVEC.    We had already established KAI1’s regulation on ING4 through RelA/p65 (Campbell et al, 2004; Tang et al, 2014b). P65 was shown to suppress transcription by increasing methylation of the promoter of the target gene (Wehbe et al, 2006). Furthermore, p65 was shown to be regulated by p-Akt (Bai et al, 2009). Since Akt was also shown to have important role during the regulation of angiogenesis, we proposed that KAI1 collaborated  91 with Akt during angiogenesis modulation (Dimmeler & Zeiher, 2000; Karar & Maity, 2011; Shiojima & Walsh, 2002). According to Figure 4.9, KAI1 significantly reduced Akt phosphorylation, while KAI1 KD increased Akt phosphorylation. Moreover, the inhibition of ING4 expression due to KAI1 KD was rescued using a potent Akt inhibitor, referred to as the LY294002 (Figure 4.9b). Furthermore, LY294002 also abolished IL-6 and VEGF expression even after KAI1 KD (Figure 4.9c and d). Additionally, Akt’s activity was also shown to be triggered by different transmembrane receptors such as ingetrins, cytokine receptros, or G-protein-coupled receptros, which was able to induce the production of phosphatidylinositol 3,4,5 triphosphates by phosphoinositide 3-kinase (PI3K) (Tang et al, 2014a). Meanwhile, previous reports indicated that KAI1 was able to suppress cancer metastasis through the interaction of with other membrane proteins, including integrins and growth factors (Mannion et al, 1996; Nakamura et al, 2000; Ono et al, 1999; Zhang et al, 2001). Furthermore, recent discovery also showed that KAI1 controlled the ubiquitylation of endothelial growth factor receptor (EGFR) and activated protein kinase C, which ultimately lead to the attenuation of EGFR signalling pathway (Odintsova et al, 2013). Therefore, KAI1’s regulation on Akt may be due to EGFR suppression as Akt was previously shown to be regulated by EGFR (Galbaugh et al, 2006; Lu et al, 2007; Wang et al, 2000). In fact, our data revealed that KAI1 overexpression suppressed EGFR phosphorylation, while KAI1 KD increased EGFR phosphorylation (Figure 4.9a).  In summary, our study illustrated KAI1’s regulation on human melanoma angiogenesis by suppressing IL-6 and VEGF expression. We also showed that KAI1 suppressed Akt phosphorylation through the regulation of p-EGFR. As a result, our results revealed a better understanding with regard to the role of KAI1 on melanoma angiogenesis  92 and that the restoration of KAI1 expression would be an effective strategy towards antiangiogenesis therapy for human melanoma.      93 Chapter 5: The Role of KAI1 and SEMA3C in the Regulation of VEGFR2 Phosphorylation 5.1 Background and Rationale The class-3 semaphorins were a class of secreted proteins originally characterized as the axon guidance factors (Huber et al, 2003). In particular, semaphoring 3C (SEMA3C) was a member of this highly conserved protein family that primarily functioned as axon guidance for neuron development in the central and peripheral nervous system (Castellani & Rougon, 2002; Cohen et al, 2005). Semaphorins was found to be involved in numerous pathways such as proliferation, adhesion, migration, and invasion (Goshima et al, 2002). As oppose to other members of the class-3 semaphorins that suppressed migration, invasion, and angiogenesis, sema3C upregulated cancer invasion and metastatic (Bagnard et al, 2001; Galani et al, 2002).  Class-3 semaphorins were also found to bind to receptor family of neuropilin that was able to associate with plexin receptors to propagate signal (Tamagnone et al, 1999). Furthermore, recent studies revealed that neuropilin was able to interact with VEGFR2 during the activation of angiogenesis, which suggested that SEMA3C may also be involved in the regulation of cancer angiogenesis (Soker et al, 1998).  Angiogenic factor VEGF was also shown to act through the tyrosine kinase receptors VEGFR2, which then activated the phosphatidylinositol 3-kinase (PI3K)/Akt pathway that was primarily responsible for cell proliferation and survival(Gerber et al, 1998). VEGF2 was also shown to mediate the activation extracellular signal-regulated kinase 1 and 2 (ERK1 and ERK2) and Src/focal adhesion kinase (FAK) pathway, which played important roles in cell motility, vascular permeability and actin remodeling (Abedi & Zachary, 1997; Gliki et al, 2001). Although  94 recent studies had shown that SEMA3C may be important in cancer progression (Neufeld et al, 2005), its role was still not well-studied.  In order to further investigate the SEMA3C’s regulation on melanoma angiogenesis, we performed HUVEC growth and tube formation assays while overexpressing or knocking-down SEMA3C. Since KAI1 suppressed melanoma angiogenesis, we also investigated the role of KAI1 during SEMA3C induced angiogenesis. Therefore, these findings highlighted the significance of SEMA3C during the regulation melanoma angiogenesis and its association with the tumor suppressor KAI1.  5.2 Results 5.2.1 SEMA3C Expression Induced HUVEC Growth and Tubular Structure Formation through the Regulation of IL-6 and VEGF In order to investigate the role of SEMA3C in melanoma cell angiogenesis, we used MMRU melanoma cell line that was transiently transfected with either SEMA3C plasmid to overexpress SEMA3C or shSEMA3C shRNA to knockdown SEMA3C. The expression was confirmed using Western blot analysis (Figure 5.1a). The conditioned media (CM) was collected for each cell line and used for either HUVEC growth assay or tubular formation assay. Compared to the negative control, SEMA3C overexpression resulted in 2-fold increase in HUVEC cell growth after 3 days. On the other hand, SEMA3C KD showed a 50% reduction on HUVEC cell growth (Figure 5.1b). The average number of tubular structure formation was also increased by 2 folds in CM when SEMA3C was overexpressed and reduced by 80% in CM with SEMA3C KD compared to the control (Figure 5.1c, d).   95 It had been shown that angiogenesis was closely associated with an upregulation in IL-6 and VEGF expression (Wani et al, 2011).  So, it is necessary to test the effect of SEMA3C on the expression of IL-6 and VEGF. The real-time RT-PCR data revealed that SEMA3C overexpression in melanoma cell line increased IL-6 mRNA level by about 2 folds and SEMA3C KD reduced IL-6 mRNA level by about 40% compared to the negative control (Figure 5.2a). By using the ELISA assay, we found that SEMA3C overexpression increased IL-6 protein level in CM by more than 2 folds and that SEMA3C-SD reduced IL-6 protein by about 50% (Figure 5.2b). Furthermore, the regulation of HUVEC growth by SEMA3C through the regulation of IL-6 expression was illustrated using IL-6 induction and IL-6 antibody blocking treatments. As shown in Figure 5.2c, the addition of IL-6 antibody in the CM from the forced SEMA3C expression group abolished the elevated HUVEC growth after 24h and 48h. Conversely, the addition of 0.8 ng/mL of recombinant IL-6 in the CM from cells overexpressed with shSEMA3C rescued the inhibition effect on HUVEC growth triggered by the SEMA3C KD (Figure 5.2d). Moreover, we were able to show that SEMA3C overexpression upregulated the mRNA expression of VEGF, whereas SEMA3C KD suppressed the mRNA level of VEGF (Figure 5.2e). Similar results were obtained after the western blot analysis where SEMA3C was able to induce VEGF protein expression  (Figure 5.2f).   96  Figure 5.1 SEMA3C expression in melanoma cells induced HUVEC growth and tube formation MMRU cells were transiently transfected with SEMA3C to overexpression SEMA3C and shSEMA3C to silence SEMA3C expression, and transfected with empty vector and shRNA control (Ctrl) for the negative control. The conditioned medium (CM) collected was used for HUVEC growth or tube formation assay. (a) the expression of SEMA3C was confirmed using Western blot. (b) SEMA3C overexpression induced while SEMA3C KD reduced the growth of HUVECs. Results shown were presented as means ± SD from 3 independent experiments. (c) SEMA3C overexpression induced while SEMA3C KD reduced the  97 formation of tubular structure of HUVECs. (d) the number of tubes formed in each of the 5 randomly chosen fields was counted. **, P < 0.01; ***, P < 0.001; Student’s t test.     98  Figure 5.2 SEMA3C expression in melanoma cells upregulated IL-6 and VEGF (a and b) SEMA3C expression induced IL-6 expression. IL-6 mRNA and secreted protein levels in CM were determined by qRT-PCR (a) and ELISA (b). (c) IL-6 antibody (320 ng/mL) neutralized the IL-6 in the CM suppressed the HUVEC growth after SEMA3C overexpression. (d) IL-6 recombinant protein (0.8 ng/mL) was added to CM to rescue the HUVEC growth suppression caused by SEMA3C KD. shCtrl, scrambled shRNA. (e) SEMA3C induced VEGF mRNA level shown by qRT-PCR. (f) SEMA3C induced VEGF protein level determined by Western blot analysis. **, P < 0.01; ***, P < 0.001; Student’s t test.      99 5.2.2 SEMA3C Upregulated Melanoma Angiogenesis in vivo In order to show that SEMA3C was able to regulate neovessel formation in a mouse model, we conducted the in vivo matrigel plug assay by overexpressing SEMA3C plasmid to overexpress SEMA3C and shSEMA3C to knockdown SEMA3C. Upon visual examination, we noted that SEMA3C expression resulted in a significantly more vascularization in the matrigel plugs while a significant less vascularization was observed with SEMA3C was knocked down (Figure 5.3a). The immunofluorescence analysis revealed that the intensity of the CD31 positive cells when SEMA3C was overexpressed increased more than 3 folds whereas SEMA3C KD showed a 3-fold reduction in the fluorescent intensity compareed to the negative control (Figure 5.3b and c). Furthermore, IL-6 and VEGF fluorescence staining also confirmed that SEMA3C was able to induce angiogenesis by upregulating the expression of IL-6 and VEGF (Figure 5.3b and c).   100  Figure 5.3 Forced SEMA3C expression in MMRU melanoma cell line induced blood vessel formation in vivo (a) Photographs were taken for the matrigel plugs excised from severe combined immunodeficiency mice after 10 days of growth in vivo. (b) Immunofluorescence staining was used to determine CD31, IL-6 and VEGF expression in matrigel plugs. DAPI staining was used to show the overall cell density in each matigel plugs. (c) Relative fluorescent  101 intensity was analyzed in 5 random fields of each group by using ImageJ. ***, P < 0.001; Student’s t test.  5.2.3 SEMA3C Expression Upregulated VEGFR2 Phosphorylation Since phosphatidylinositol 3-kinase (PI3K) pathway has been shown to play an important role in different cellular functions including angiogenesis, we asked if SEMA3C can regulate Akt phosphorylation, which is one important downstream target of the PI3K pathway (Jiang & Liu, 2009). According to Figure 5.4a, SEMA3C overexpression was able to induce the phosphorylation of Akt at Ser 473, which was suppressed by SEMA3C KD. Since the activation of VEGFR2 was important for the signaling of the PI3K/Akt pathway (Zhang et al, 2014), we asked whether SEMA3C was able to regulate the phosphorylation of VEGFR2. As shown in Figure 5.4a, forced SEMA3C expression increased the phosphorylation of VEGFR2 at Tyr 1175, whereas SEMA3C KD suppressed the phosphorylation of VEGFR2. Furthermore, VEGFR inhibitor Sorafenib was able to abolish the effect of SEMA3C overexpression on the phosphorylation of Akt (Figure 5.4a). In order to show that the phosphorylation of VEGFR2 was primarily due to SEMA3C, we treated the cells with 600nM of SEMA3C recombinant protein and performed a time point analysis. We found that there was an increased in VEGFR phosphorylation after 30 minutes of treatment and that VEGF expression was increased much later time (Figure 5.4b and c As shown in Figure 5.4d, when we blocked VEGF using a VEGF antibody Bevacizumab, SEMA3C overexpression was able to induce VEGFR2 phosphorylation even after the Bevacizumab treatment. Thus, SEMA3C overexpression was able to induce the phosphorylation of VEGFR2 even when VEGF was suppressed. As we had already revealed that KAI1  102 suppressed melanoma angiogenesis, we wanted to see if KAI1 was able to regulate VEGFR2 phosphorylation during its regulation on angiogenesis. As shown in Figure 5.4e, our result showed that KAI1 expression was able to suppress the phosphorylation of VEGFR2. Conversely, the phosphorylation of VEGFR2 was upregulated when KAI1 was knocked down (Figure 5.4e). Taken together, we showed that VEGFR2 phosphorylation was regulated by both SEMA3C and KAI1.    103  Figure 5.4 SEMA3C induced while KAI1 suppressed VEGFR2 phosphorylation (a) MMRU cells were treated with DMSO or Sorafenib (10 μM) for 24h. The VEGFR2 and Akt phosphorylation levels were obtained using Western Blot analysis. Sorafenib treatment was able to suppress SEMA3C-mediated upregulation of Akt phosphorylation. (b) cells were treated with 600nM of SEMA3C recombinant protein at for 0h, 1/2h, 1h, 2h, 4h, or 8h to show different levels of VEGFR2 phosphorylation and VEGF expression at each time point. (c) The quantification of p-VEGFR2 and VEGF for the Western blot analysis for (b). (d) cells were treated with SEMA3C recombinant protein (600nM) and Bevacizumab  104 (250μg/mL) for 24h. Western blot analysis was used to evaluate the phosphorylation of VEGFR2. (e) KAI1 was overexpressed or knocked down to evaluate the phosphorylation of VEGFR2. **, P < 0.01; ***, P < 0.001; Student’s t test.  5.2.4 KAI1’s Regulation on SEMA3C in VEGFR2 Phosphorylation, HUVEC Growth and Tube Formation Since in addition to SEMA3C’s regulation on VEGFR2 phosphorylation, KAI1 was also shown to regulate VEGFR2 phosphorylation, we then investigated the phosphorylation of VEGFR2 after forced expression of both SEMA3C and KAI1. As seen in Figure 5.5a, our Western blot result showed that forced KAI1 expression was able to suppress VEGFR2 phosphorylation after SEMA3C overexpression. Conversely, KAI1 KD rescued in suppression of VEGFR2 phosphorylation due to SEMA3C KD (Figure 5.5a). The quantification of Western blots shown was analyzed by using the ImageJ image processing software (NIH, Bethesda, MD, USA). To further study KAI1’s regulation on SEMA3C during the regulation of angiogenesis, we performed the HUVEC growth assay. We found that KAI1 expression was able to abolish SEMA3C induced HUVEC growth whereas KAI1 KD rescued the suppression of HUVEC growth caused by SEMA3C KD (Figure 5.5b). We further investigated this regulation through the use of tube formation assay. According to Figure 5.5c, the number of tubular structure formed was increased when KAI1 was overexpressed even SEMA3C was overexpressed. And the opposite was observed where tube formation was decreased when KAI1 was knocked down even SEMA3C was knocked down (Figure 5.5d). Taken together, we showed that KAI1 regulated SEMA3C’s regulation on melanoma angiogenesis.  105   Figure 5.5 KAI1 regulated SEMA3C-induced VEGFR2 phosphorylation and HUVEC growth (a) VEGFR2 phosphorylation was evaluated after SEMA3C and KAI1 overexpression or knockdown. The quantification of western blots was analyzed by ImageJ (NIH, Bethesda, MD, USA). (b) the evaluation of HUVEC growth in CM collected from cells with SEMA3C and KAI1 overexpression or knockdown. ***, compared to CV; ###, compared to shCtrl.  (c and d) the investigation of tubular structure formation of HUVEC cultured in CM from  106 cells with SEMA3C and KAI1 overexpression or knockdown.  Results shown were presented as means ± SD from 3 independent experiments. (e and f) the number of tubes formed in each of the 5 randomly chosen fields was counted. The quantification of western blots was analyzed by Image J (NIH, Bethesda, MD, USA). **, P < 0.01; ***, P < 0.001; ##, P < 0.01; ###, P < 0.001; Student’s t test.  5.3 Discussion Many members of the semaphorins class 3 family, such as SEMA3A, B, D and G, were shown to be inhibitors of cell migration and cancer progression, which were considered to have antitumor functionalities (Miyato et al, 2012). SEMA3C, however, was associated with the promotion of cancer cell migration and invasion in various metastatic tumors (Esselens et al, 2010). Increased expression of SEMA3C was also reported in lung cancer, gastric cancer, and breast cancer (Martin-Satue & Blanco, 1999; Miyato et al, 2012; Neufeld et al, 2005). Recent study also showed that increased SEMA3C expression was closed related to the increase cancer cell migration and invasion in prostate cancer (Herman & Meadows, 2007). Furthermore, an increased SEMA3C expression in ovarian cancer was also associated with a poorer patient survival (Galani et al, 2002). Moreoever, previous findings revealed that SEMA3C also upregulated endothelial cell proliferation, migration and tube formation in vitro, which exposed the possibility of SEMA3C’s role during cancer angiogenesis (Banu et al, 2006). One of the key players during the regulation of angiogenesis was the receptor for the angiogneic factor VEGF referred to as VEGFR2 which was shown to regulate multiple cellular activities such as survival, proliferation, migration, invasion and gene expression (Gille et al, 2001). Past findings also related VEGFR2 with the phosphoinositide 3-kinase  107 (PI3K) signaling pathway was one of the main signal transduction pathways for VEGFR2 to control cell survival and proliferation (Cantley, 2002). Akt was shown to be one of many protein kinases that would bind to phosphatidylinositol-3,4,5-triphosphate (PIP3) which was activated by PI3K. The activated Akt then was able to initiate its regulation on cell proliferation, progression and survival (Rameh & Cantley, 1999). Therefore, it was possible that SEMA3C was able to regulate VEGFR2 and tumor angiogenesis.  In our study, we showed that the growth and the tubular structure formation of HUVECs were significantly upregulated when they were cultured in the CM collected from melanoma cells with SEMA3C overexpression. Conversely, the HUVEC growth and tube formation were significantly suppressed when SEMA3C was knocked down (Figure 5.1). According the in vivo matrigel plug assay, forced SEMA3C expression in melanoma cells was able to induce the supportive vasculature that resulted in an increase in blood vessel formation, whereas the suppression of neovessel formation was observed after SEMA3C was knocked down (Figure 5.3). The expression of angiogenic factors IL-6 and VEGF was showed to be elevated during the upregulation of angiogenesis (Wani et al, 2011). According to our findings, we showed that SEMA3C expression upregulated IL-6 and VEGF mRNA expression and protein level whereas SEMA3C KD suppressed IL-6 and VEGF (Figure 5.2). Moreover, the antibody blocking assay further supported SEMA3C’s regulation on IL-6 during melanoma angiogenesis. As seen in Figure 5.2c, IL-6 antibody treatment inhibited the upregulation of HUVEC growth due to SEMA3C overexpression. Conversely, the addition of recombinant IL-6 protein rescued the inhibition of HUVEC growth caused by SEMA3C KD (Figure 5.2d). Taken together, these results revealed that SEMA3C expression resulted in an upregulation of IL-6 and VEGF expression which induced the proliferation of HUVEC.   108  Since VEGFR2 was shown to have important role during the regulation of angiogenesis, we next investigated SEMA3C’s role in regulating VEGFR2 phosphorylation. As shown in Figure 5.4a, forced SEMA3C expression upregulated VEGFR2 phosphorylation whereas SEMA3C KD suppressed VEGFR2 phosphorylation. As Akt was shown to be regulated by VEGFR2 (Cantley, 2002), SEMA3C overexpression induced Akt phosphorylation through VEGFR2, while SEMA3C KD suppressed Akt phosphorylation (Figure 5.4a). By using the kinase inhibitor sorafenib that suppressed VEGFR2 phosphorylation, it was able to suppress the upregulation of Akt phosphorylation caused by SEMA3C overexpression (Figure 5.4a).  The time point analysis also revealed that SEMA3C was able to elevate VEGFR2 phosphorylation faster than the production of VEGF (Figure 5.4b). It was further confirmed by introducing VEGF inhibitor, bevacizumab, where VEGFR2 phosphorylation was upregulated after SEMA3C overexpression even in the presence of VEGF inhibitor bevacizumab (Figure 5.4d).  Previously, we had established that KAI1 was able to suppress melanoma angiogenesis, and in this study we further showed that KAI1 expression was able to suppress VEGFR2 phosphorylation (Figure 5.4e).  In order to see if KAI1 was able to control SEMA3C’s regulation on VEGFR2 phosphorylation, we overexpressed KAI1 while SEMA3C was overexpressed. As shown in Figure 5.5a, KAI1 expression suppressed SEMA3C induced VEGFR2 phosphorylation while KAI1 KD rescued the inhibition of VEGFR2 phosphorylation due to SEMA3C KD. This finding was further confirmed by examine the HUVEC growth where KAI1 expression abolished SEMA3C induced HUVEC growth, whereas KAI1 KD upregulated the HUVEC growth even after SEMA3C KD (Figure 5.5b). Similar patterns were observed in the tubular structure formation of HUVEC (Figure 5.5c and d). Taken together, KAI1 overexpression was able to  109 downregulate SEMA3C’s effect on VEGFR2 phosphorylation and SEMA3C induced HUVEC growth and tube formation.   In summary, our data for the first time revealed SEMA3C’s role in the upregulation of melanoma angiogenesis through the regulation of IL-6 and VEGF. We also showed that SEMA3C elevated Akt phosphorylation through the regulation of VEGFR2. Furthermore, we showed that KAI1 was able to regulate SEMA3C induced melanoma angiogenesis through VEGFR2 phosphorylation, HUVEC growth and tube formation. As a result, our data revealed a better understanding of SEMA3C’s function in melanoma angiogenesis as well as KAI1’s role in regulating SEMA3C induced angiogenesis. Therefore, the inhibition of SEMA3C’s activity or the reestablishment of KAI1’s function would be effective treatment methods towards anti-angiogenic therapy.    110 Chapter 6: Conclusions 6.1 Summary of Findings In chapter 3, we investigated the prognostic significance of KAI1, which was a tetraspanin membrane protein that was responsible for the recruitment of various other membrane proteins to form a scaffold. By using the tissue microarray technology, we were able to show a significance correlation between a weaker KAI1 staining and a poorer 5-year patient survival. Furthermore, a weak KAI1 staining was significantly correlated with increased tumor thickness and the presence of ulceration. Finally, by using multivariate Cox regression analysis, we showed that KAI1 could be represented as an independent diagnostic marker. In order to further analyze the function of KAI1 expression in melanoma cell lines, in vitro functional experiments were performed. The wound healing assay revealed that KAI1 expression suppressed melanoma cell migration while KAI1 KD induced melanoma cell migration. This regulation was further investigated to show a correlation with ROCK activity and F-actin stress fibre formation, of which a significant reduction in F-actin staining was observed with KAI1 overexpression whereas stress fibre formation was induced after KAI KD. We also explored KAI1’s regulation on melanoma invasion by using the Boyden Chamber assay. According to our data, forced KAI1 expression resulted in a significantly less invaded cell compared to the control. Conversely, KAI1 KD caused a significant increase in the number of cell that invaded through the membrane. It was further confirmed that the regulation of KAI1 on melanoma cell invasion was through the regulation of MMP2 activity, of which a significant loss in the MMP2’s gelatinolytic activity was observed after KAI1 overexpression. Finally, we were able to show for the first time that KAI1 induced the expression of another tumor suppressor gene referred to as ING4. In particular, we showed  111 that ING4 was regulated by KAI1 through the regulation of p65 phosphorylation. Therefore, the reactivation of KAI1 and the subsequent suppression of p65 phosphorylation may be a potential therapeutic strategy in future melanoma treatments. Also, our result indicated that KAI1 was an important and independent diagnostic marker in melanoma where the loss of KAI1 expression directly contributed to the decreased 5-year patient survival. Meanwhile, since angiogenesis is another vital phase during cancer cell proliferation and metastasis, the regulation and manipulation of angiogenesis is viewed as a prevailing approach against cancer progression. As a result, we continued to investigate the role of KAI1 and its regulation on melanoma angiogenesis. We found a significantly suppression of HUVEC growth using the condition medium with forced KAI1 expression whereas an upregulation was observed with KAI1 KD. Furthermore, we also investigated KAI1’s regulation on the tubular structure formation of HUVEC since the tubular structure formed by HUVECs usually represented the later stages of the angiogenic process where endothelial cells were differentiated (Goodwin, 2007). According to our findings, KAI1 expression suppressed HUVEC tube formation while KAI1 KD restored the tubular structure formation of HUVEC. Moreover, the angiogenic factors IL-6 and VEGF were also suppressed by KAI1 expression. Since it had been shown that ING4 was regulated by KAI1, we investigated KAI1’s regulation on ING4 during melanoma angiogenesis. Our findings revealed that during the regulation of HUVEC growth, tube formation, and in vivo neovessel formation, ING4’s activity was required during KAI1’s regulation on melanoma angiogenesis. Furthermore, we showed that KAI1’s regulation of ING4 was through the regulation of the PI3K/Akt pathway, of which KAI1 expression suppressed EGFR phosphorylation and resulted in a downregulation of Akt phosphorylation. It was further confirmed by using the  112 Akt inhibitor, LY294002, where it was able to abolish the upregulation of p65 phosphorylation even after KAI1 KD and resulted in the restoration of ING4 expression. Taken together, our data revealed KAI1’s role in melanoma angiogenesis through its regulation on IL-6 and VEGF. This was further confirmed according to the in vivo matrigel plug assay. Also, KAI1 expression was able to suppress EGFR phosphorylation and subsequently the phosphorylation of Akt and resulted in the suppression of p65 phosphorylation. As a result, these data provided additional understanding towards KAI1’s regulation on ING4 expression and melanoma progression. In chapter 5, we investigated the role of SEMA3C that was previously shown to provide axon guidance for neuron development in the central and peripheral nervous system. Since recent discoveries had linked SEMA3C with many advanced cancers and their progression, we then examined its function in human melanoma and in particular, in angiogenesis regulation. Our results revealed that SEMA3C expression induced the growth of HUVEC as well as their tubular structure formation. Also, IL-6 and VEGF expression was elevated after SEMA3C overexpression. SEMA3C was also shown to induce blood vessel formation in the matrigel plugs in vivo. Since VEGFR2 was one of the key players during the regulation of angiogenesis that was important during the regulation of various cellular activities, we examined the regulation of VEGFR2 by SEMA3C. According to our findings, forced SEMA3C expression elevated the phosphorylation of SEMA3C, which in turn activated the downstream PI3K/Akt pathway shown by the upregulation of p-Akt. Meanwhile, KAI1 was shown to also regulate EGFR phosphorylation. Since EGFR and VEGFR share some structural and functional similarities, as well as some common downstream pathways, we investigated KAI1’s regulation on VEGFR2 phosphorylation. Our  113 data revealed that KAI1 overexpression suppressed the phosphorylation of VEGFR2 while KAI1 KD induced the phosphorylation. We also found that KAI1 expression was able to abolish SEMA3C-induced phosphorylation of VEGFR2. Conversely, KAI1 KD rescued the suppression of VEGFR2 phosphorylation, even after SEMA3C KD. The results from our in vitro functional analyses further confirmed KAI1’s regulation on SEMA3C.  Based on our findings, we suggest that KAI1 is a key player regulating melanoma progression and angiogenesis. Not only might it be used as an independent diagnostic marker, it also has important role in controlling melanoma cell migration, invasion, and angiogenesis. Restoration of KAI1’s activity could potentially serve as a valuable therapeutic strategy for future melanoma treatments.  6.2 Limitations of The Study and Future Directions In this study, we investigated the role of KAI1 in melanoma progression by using TMA analysis. However, the scoring method was subjective, since the examination of immunostained tissue microarrays and classification of the staining was by visual comparison using immunoreactive score. We tried to minimize the subjectivity in our analysis by performing the experiment with 3 individuals in a blinded manner. And if there was any discrepancy between the scores, a higher score was taken. The scores ranged from 0-3 representing the intensities and from 1-4 based upon the percentage of cells showing the staining (0-25, 26-50, 51-75 and 76-100%), and the final score was based on the product of the intensity score and the percent staining score.  An automated analysis using computer software could be used to remove any subjective factors. However, this might be highly unfavorable due to its inability to distinguish between tumor and non-tumor samples in the  114 microarray since there is often normal skin tissue around the melanoma tissue sample. Thus, a trained pathologist or an experienced researcher is required to identify the lesion within a biopsy, and a semi-automated analysis using internal controls could be more convincing. The pictures of the stained samples can be captured and the non-tumor samples can then eliminated prior to analysis. Finally, we could quantify the intensity staining using image analysis software to obtain the relative staining intensity of the TMA. The results obtained could be used to confirm and improve the results obtained manually.  We also discovered that KAI1 expression suppressed melanoma cell migration and invasion in vitro. Yet, it was unclear whether KAI1’s function as a metastasis suppressor of melanoma was observable in vivo. Thus, the xenograft models or tail vain injection tumor metastasis assays would be useful for future studies in order to address this matter. Also, we showed that KAI1 was able to suppress melanoma angiogenesis in vivo by using matrigel plug assay. The advantage of this assay was the relative ease of the technique and the robust results acquisition. However, some limitations of this method included nonspecific immune responses that caused the release of angiogenic cytokines, inflammation responses, or unwanted matrix depositions (Staton et al, 2009). All of these could potentially interfere with our data interpretation. However, the use of immune deficient mice was beneficial to minimize the effects. In sum, besides the use of multiple in vitro angiogenesis assays, more than one in vivo analysis could be performed in order to consolidate the findings. Again, the study of xenografts analysis with the subcutaneous implantation of tumour cells was a powerful technique to investigate tumor metastasis and angiogenesis in vivo and enhance the likelihood of obtaining meaningful outcomes.   115 We had shown KAI1’s regulation on EGFR and VEGFR together with the activity of SEMA3C. However, a detailed mechanism of this regulation is not apparent. It was shown previously that KAI1 was able to recruit multiple membrane proteins including EGFR to the tetraspanin-enriched microdomains that suppressed EGFR phosphorylation and downstream signaling (Odintsova et al, 2013). Therefore, it was likely that KAI1 recruited the SEMA3C and plexin/neuropilin complexes as well as the VEGF or EGF receptor, by which KAI1 prevented the interaction between the plexin/neuropilin complexes with the receptor tyrosine kinase to achieve the suppressive functionality. Other proteins may also be involved or recruited to the tetraspanin microdomain. Therefore, addition experiments are required in order to fully understand KAI1’s role in the regulation of melanoma progression and in the scaffolding complex.   116 Bibliography Abdel-Rahman MH, Pilarski R, Massengill JB, Christopher BN, Noss R, Davidorf FH (2011) Melanoma candidate genes CDKN2A/p16/INK4A, p14ARF, and CDK4 sequencing in patients with uveal melanoma with relative high-risk for hereditary cancer predisposition. Melanoma research 21: 175-179  Abe T, Kato M, Miki H, Takenawa T, Endo T (2003) Small GTPase Tc10 and its homologue RhoT induce N-WASP-mediated long process formation and neurite outgrowth. J Cell Sci 116: 155-168  Abedi H, Zachary I (1997) Vascular endothelial growth factor stimulates tyrosine phosphorylation and recruitment to new focal adhesions of focal adhesion kinase and paxillin in endothelial cells. The Journal of biological chemistry 272: 15442-15451  Adachi M, Taki T, Ieki Y, Huang CL, Higashiyama M, Miyake M (1996) Correlation of KAI1/CD82 gene expression with good prognosis in patients with non-small cell lung cancer. Cancer Res 56: 1751-1755  Anforth RM, Carlos GR, Scolyer RA, Chou S, Fernandez-Penas P (2015) Eruptive naevi in a patient treated with LGX818 for BRAF mutant metastatic melanoma. Melanoma research 25: 91-94  Anzai N, Lee Y, Youn BS, Fukuda S, Kim YJ, Mantel C, Akashi M, Broxmeyer HE (2002) C-kit associated with the transmembrane 4 superfamily proteins constitutes a functionally distinct subunit in human hematopoietic progenitors. Blood 99: 4413-4421  American Cancer Society (2015) Cancer Fact and Figures 2015, American Cancer Society, 2015.  Baek SH, Ohgi KA, Rose DW, Koo EH, Glass CK, Rosenfeld MG (2002) Exchange of N-CoR corepressor and Tip60 coactivator complexes links gene expression by NF-kappaB and beta-amyloid precursor protein. Cell 110: 55-67  Bagnard D, Vaillant C, Khuth ST, Dufay N, Lohrum M, Puschel AW, Belin MF, Bolz J, Thomasset N (2001) Semaphorin 3A-vascular endothelial growth factor-165 balance mediates migration and apoptosis of neural progenitor cells by the recruitment of shared receptor. The Journal of neuroscience : the official journal of the Society for Neuroscience 21: 3332-3341  Bai D, Ueno L, Vogt PK (2009) Akt-mediated regulation of NFkappaB and the essentialness of NFkappaB for the oncogenicity of PI3K and Akt. International journal of cancer Journal international du cancer 125: 2863-2870  Baker DJ, Jin F, van Deursen JM (2008) The yin and yang of the Cdkn2a locus in senescence and aging. Cell Cycle 7: 2795-2802  117  Balch CM, Buzaid AC, Soong SJ, Atkins MB, Cascinelli N, Coit DG, Fleming ID, Gershenwald JE, Houghton A, Jr., Kirkwood JM, McMasters KM, Mihm MF, Morton DL, Reintgen DS, Ross MI, Sober A, Thompson JA, Thompson JF (2001) Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 19: 3635-3648  Balch CM, Morton DL, Gershenwald JE, McMasters KM, Nieweg OE, Powell B, Ross MI, Sondak VK, Thompson JF (2009) Sentinel node biopsy and standard of care for melanoma. Journal of the American Academy of Dermatology 60: 872-875  Banu N, Teichman J, Dunlap-Brown M, Villegas G, Tufro A (2006) Semaphorin 3C regulates endothelial cell function by increasing integrin activity. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 20: 2150-2152  Bari R, Zhang YH, Zhang F, Wang NX, Stipp CS, Zheng JJ, Zhang XA (2009) Transmembrane interactions are needed for KAI1/CD82-mediated suppression of cancer invasion and metastasis. The American journal of pathology 174: 647-660  Barrascout E, Medioni J, Scotte F, Ayllon J, Mejean A, Cuenod CA, Tartour E, Elaidi R, Oudard S (2010) [Angiogenesis inhibition: review of the activity of sorafenib, sunitinib and bevacizumab]. Bulletin du cancer 97: 29-43  Bartosch B, Vitelli A, Granier C, Goujon C, Dubuisson J, Pascale S, Scarselli E, Cortese R, Nicosia A, Cosset FL (2003) Cell entry of hepatitis C virus requires a set of co-receptors that include the CD81 tetraspanin and the SR-B1 scavenger receptor. The Journal of biological chemistry 278: 41624-41630  Bass R, Werner F, Odintsova E, Sugiura T, Berditchevski F, Ellis V (2005) Regulation of urokinase receptor proteolytic function by the tetraspanin CD82. The Journal of biological chemistry 280: 14811-14818  Baxter LL, Pavan WJ (2003) Pmel17 expression is Mitf-dependent and reveals cranial melanoblast migration during murine development. Gene expression patterns : GEP 3: 703-707  Bodnar RJ (2014) Anti-Angiogenic Drugs: Involvement in Cutaneous Side Effects and Wound-Healing Complication. Advances in wound care 3: 635-646  Bollag G, Hirth P, Tsai J, Zhang J, Ibrahim PN, Cho H, Spevak W, Zhang C, Zhang Y, Habets G, Burton EA, Wong B, Tsang G, West BL, Powell B, Shellooe R, Marimuthu A, Nguyen H, Zhang KY, Artis DR, Schlessinger J, Su F, Higgins B, Iyer R, D'Andrea K, Koehler A, Stumm M, Lin PS, Lee RJ, Grippo J, Puzanov I, Kim KB, Ribas A, McArthur GA, Sosman JA, Chapman PB, Flaherty KT, Xu X, Nathanson KL, Nolop K (2010) Clinical  118 efficacy of a RAF inhibitor needs broad target blockade in BRAF-mutant melanoma. Nature 467: 596-599  Buecher B, Gauthier-Villars M, Desjardins L, Lumbroso-Le Rouic L, Levy C, De Pauw A, Bombled J, Tirapo C, Houdayer C, Bressac-de Paillerets B, Stoppa-Lyonnet D (2010) Contribution of CDKN2A/P16 ( INK4A ), P14 (ARF), CDK4 and BRCA1/2 germline mutations in individuals with suspected genetic predisposition to uveal melanoma. Fam Cancer 9: 663-667  Byers HR, Etoh T, Doherty JR, Sober AJ, Mihm MC, Jr. (1991) Cell migration and actin organization in cultured human primary, recurrent cutaneous and metastatic melanoma. Time-lapse and image analysis. The American journal of pathology 139: 423-435  Campbell KJ, Rocha S, Perkins ND (2004) Active repression of antiapoptotic gene expression by RelA(p65) NF-kappa B. Mol Cell 13: 853-865  Cannon KS, Cresswell P (2001) Quality control of transmembrane domain assembly in the tetraspanin CD82. The EMBO journal 20: 2443-2453  Cantley LC (2002) The phosphoinositide 3-kinase pathway. Science 296: 1655-1657  Cantley LC, Neel BG (1999) New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway. Proceedings of the National Academy of Sciences of the United States of America 96: 4240-4245  Carlier MF, Ducruix A, Pantaloni D (1999) Signalling to actin: the Cdc42-N-WASP-Arp2/3 connection. Chem Biol 6: R235-240  Carloni V, Mazzocca A, Ravichandran KS (2004) Tetraspanin CD81 is linked to ERK/MAPKinase signaling by Shc in liver tumor cells. Oncogene 23: 1566-1574  Carragher NO, Walker SM, Scott Carragher LA, Harris F, Sawyer TK, Brunton VG, Ozanne BW, Frame MC (2006) Calpain 2 and Src dependence distinguishes mesenchymal and amoeboid modes of tumour cell invasion: a link to integrin function. Oncogene 25: 5726-5740  Castellani V, Rougon G (2002) Control of semaphorin signaling. Curr Opin Neurobiol 12: 532-541  Chaffer CL, Weinberg RA (2011) A perspective on cancer cell metastasis. Science 331: 1559-1564  Chakrabarti R, Hwang J, Andres Blanco M, Wei Y, Lukacisin M, Romano RA, Smalley K, Liu S, Yang Q, Ibrahim T, Mercatali L, Amadori D, Haffty BG, Sinha S, Kang Y (2012) Elf5 inhibits the epithelial-mesenchymal transition in mammary gland development and breast cancer metastasis by transcriptionally repressing Snail2. Nat Cell Biol 14: 1212-1222  119  Chambers AF, Groom AC, MacDonald IC (2002) Dissemination and growth of cancer cells in metastatic sites. Nature reviews Cancer 2: 563-572  Chang CJ, Chao CH, Xia W, Yang JY, Xiong Y, Li CW, Yu WH, Rehman SK, Hsu JL, Lee HH, Liu M, Chen CT, Yu D, Hung MC (2011) p53 regulates epithelial-mesenchymal transition and stem cell properties through modulating miRNAs. Nat Cell Biol 13: 317-323  Chang JC, Chang HH, Lin CT, Lo SJ (2005) The integrin alpha6beta1 modulation of PI3K and Cdc42 activities induces dynamic filopodium formation in human platelets. J Biomed Sci 12: 881-898  Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J, Dummer R, Garbe C, Testori A, Maio M, Hogg D, Lorigan P, Lebbe C, Jouary T, Schadendorf D, Ribas A, O'Day SJ, Sosman JA, Kirkwood JM, Eggermont AM, Dreno B, Nolop K, Li J, Nelson B, Hou J, Lee RJ, Flaherty KT, McArthur GA (2011) Improved survival with vemurafenib in melanoma with BRAF V600E mutation. The New England journal of medicine 364: 2507-2516  Chapman PB, Solit DB, Rosen N (2014) Combination of RAF and MEK inhibition for the treatment of BRAF-mutated melanoma: feedback is not encouraged. Cancer Cell 26: 603-604  Charrin S, Latil M, Soave S, Polesskaya A, Chretien F, Boucheix C, Rubinstein E (2013) Normal muscle regeneration requires tight control of muscle cell fusion by tetraspanins CD9 and CD81. Nature communications 4: 1674  Chen BH, Tzen JT, Bresnick AR, Chen HC (2002) Roles of Rho-associated kinase and myosin light chain kinase in morphological and migratory defects of focal adhesion kinase-null cells. The Journal of biological chemistry 277: 33857-33863  Chen Q, Manning CD, Millar H, McCabe FL, Ferrante C, Sharp C, Shahied-Arruda L, Doshi P, Nakada MT, Anderson GM (2008) CNTO 95, a fully human anti alphav integrin antibody, inhibits cell signaling, migration, invasion, and spontaneous metastasis of human breast cancer cells. Clin Exp Metastasis 25: 139-148  Chin MY, Ng KC, Li G (2005) The novel tumor suppressor p33ING2 enhances UVB-induced apoptosis in human melanoma cells. Experimental cell research 304: 531-543  Ciarletta P, Foret L, Ben Amar M (2011) The radial growth phase of malignant melanoma: multi-phase modelling, numerical simulations and linear stability analysis. Journal of the Royal Society, Interface / the Royal Society 8: 345-368  Cieply B, Farris J, Denvir J, Ford HL, Frisch SM (2013) Epithelial-mesenchymal transition and tumor suppression are controlled by a reciprocal feedback loop between ZEB1 and Grainyhead-like-2. Cancer Res 73: 6299-6309  120  Clairwood M, Ricketts J, Grant-Kels J, Gonsalves L (2014) Melanoma in skin of color in Connecticut: an analysis of melanoma incidence and stage at diagnosis in non-Hispanic blacks, non-Hispanic whites, and Hispanics. Int J Dermatol 53: 425-433  Clark WH, Jr., Tucker MA (1998) Problems with lesions related to the development of malignant melanoma: common nevi, dysplastic nevi, malignant melanoma in situ, and radial growth phase malignant melanoma. Human pathology 29: 8-14  Cohen S, Funkelstein L, Livet J, Rougon G, Henderson CE, Castellani V, Mann F (2005) A semaphorin code defines subpopulations of spinal motor neurons during mouse development. The European journal of neuroscience 21: 1767-1776  Cohen T, Nahari D, Cerem LW, Neufeld G, Levi BZ (1996) Interleukin 6 induces the expression of vascular endothelial growth factor. The Journal of biological chemistry 271: 736-741  Cross MJ, Claesson-Welsh L (2001) FGF and VEGF function in angiogenesis: signalling pathways, biological responses and therapeutic inhibition. Trends Pharmacol Sci 22: 201-207  Czubayko F, Schulte AM, Berchem GJ, Wellstein A (1996) Melanoma angiogenesis and metastasis modulated by ribozyme targeting of the secreted growth factor pleiotrophin. Proceedings of the National Academy of Sciences of the United States of America 93: 14753-14758  d'Ortho MP, Stanton H, Butler M, Atkinson SJ, Murphy G, Hembry RM (1998) MT1-MMP on the cell surface causes focal degradation of gelatin films. FEBS letters 421: 159-164  Dai DL, Martinka M, Li G (2005) Prognostic significance of activated Akt expression in melanoma: a clinicopathologic study of 292 cases. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 23: 1473-1482  Dameron KM, Volpert OV, Tainsky MA, Bouck N (1994) Control of angiogenesis in fibroblasts by p53 regulation of thrombospondin-1. Science 265: 1582-1584  Danielsen T, Rofstad EK (1998) VEGF, bFGF and EGF in the angiogenesis of human melanoma xenografts. International journal of cancer Journal international du cancer 76: 836-841  Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J, Woffendin H, Garnett MJ, Bottomley W, Davis N, Dicks E, Ewing R, Floyd Y, Gray K, Hall S, Hawes R, Hughes J, Kosmidou V, Menzies A, Mould C, Parker A, Stevens C, Watt S, Hooper S, Wilson R, Jayatilake H, Gusterson BA, Cooper C, Shipley J, Hargrave D, Pritchard-Jones K, Maitland N, Chenevix-Trench G, Riggins GJ, Bigner DD, Palmieri G, Cossu A, Flanagan A, Nicholson A, Ho JW, Leung SY, Yuen ST, Weber BL, Seigler HF, Darrow TL, Paterson H,  121 Marais R, Marshall CJ, Wooster R, Stratton MR, Futreal PA (2002) Mutations of the BRAF gene in human cancer. Nature 417: 949-954  DeMali KA, Wennerberg K, Burridge K (2003) Integrin signaling to the actin cytoskeleton. Curr Opin Cell Biol 15: 572-582  Denfeld RW, Dietrich A, Wuttig C, Tanczos E, Weiss JM, Vanscheidt W, Schopf E, Simon JC (1995) In situ expression of B7 and CD28 receptor families in human malignant melanoma: relevance for T-cell-mediated anti-tumor immunity. International journal of cancer Journal international du cancer 62: 259-265  Deryugina EI, Bourdon MA, Reisfeld RA, Strongin A (1998) Remodeling of collagen matrix by human tumor cells requires activation and cell surface association of matrix metalloproteinase-2. Cancer Res 58: 3743-3750  Dimmeler S, Zeiher AM (2000) Akt takes center stage in angiogenesis signaling. Circ Res 86: 4-5  Dong JT, Lamb PW, Rinker-Schaeffer CW, Vukanovic J, Ichikawa T, Isaacs JT, Barrett JC (1995) KAI1, a metastasis suppressor gene for prostate cancer on human chromosome 11p11.2. Science 268: 884-886  Dong JT, Suzuki H, Pin SS, Bova GS, Schalken JA, Isaacs WB, Barrett JC, Isaacs JT (1996) Down-regulation of the KAI1 metastasis suppressor gene during the progression of human prostatic cancer infrequently involves gene mutation or allelic loss. Cancer Res 56: 4387-4390  Drevs J, Konerding MA, Wolloscheck T, Wedge SR, Ryan AJ, Ogilvie DJ, Esser N (2004) The VEGF receptor tyrosine kinase inhibitor, ZD6474, inhibits angiogenesis and affects microvascular architecture within an orthotopically implanted renal cell carcinoma. Angiogenesis 7: 347-354  Duraiswamy J, Freeman GJ, Coukos G (2014) Dual blockade of PD-1 and CTLA-4 combined with tumor vaccine effectively restores T-cell rejection function in tumors--response. Cancer Res 74: 633-634; discussion 635  Elder DE, Van Belle P, Elenitsas R, Halpern A, Guerry D (1996) Neoplastic progression and prognosis in melanoma. Seminars in cutaneous medicine and surgery 15: 336-348  Enterline HT, Coman DR (1950) The ameboid motility of human and animal neoplastic cells. Cancer 3: 1033-1038  Erickson CA, Reedy MV (1998) Neural crest development: the interplay between morphogenesis and cell differentiation. Current topics in developmental biology 40: 177-209   122 Eriksson A, Cao R, Pawliuk R, Berg SM, Tsang M, Zhou D, Fleet C, Tritsaris K, Dissing S, Leboulch P, Cao Y (2002) Placenta growth factor-1 antagonizes VEGF-induced angiogenesis and tumor growth by the formation of functionally inactive PlGF-1/VEGF heterodimers. Cancer Cell 1: 99-108  Esselens C, Malapeira J, Colome N, Casal C, Rodriguez-Manzaneque JC, Canals F, Arribas J (2010) The cleavage of semaphorin 3C induced by ADAMTS1 promotes cell migration. The Journal of biological chemistry 285: 2463-2473  Fackler OT, Grosse R (2008) Cell motility through plasma membrane blebbing. The Journal of cell biology 181: 879-884  Fecher LA, Amaravadi RK, Flaherty KT (2008) The MAPK pathway in melanoma. Current opinion in oncology 20: 183-189  Ferrara N, Hillan KJ, Novotny W (2005) Bevacizumab (Avastin), a humanized anti-VEGF monoclonal antibody for cancer therapy. Biochem Biophys Res Commun 333: 328-335  Flaherty KT, Infante JR, Daud A, Gonzalez R, Kefford RF, Sosman J, Hamid O, Schuchter L, Cebon J, Ibrahim N, Kudchadkar R, Burris HA, 3rd, Falchook G, Algazi A, Lewis K, Long GV, Puzanov I, Lebowitz P, Singh A, Little S, Sun P, Allred A, Ouellet D, Kim KB, Patel K, Weber J (2012) Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. The New England journal of medicine 367: 1694-1703  Folkman J (1971) Tumor angiogenesis: therapeutic implications. The New England journal of medicine 285: 1182-1186  Folkman J (1974) Proceedings: Tumor angiogenesis factor. Cancer Res 34: 2109-2113  Folkman J (2002) Role of angiogenesis in tumor growth and metastasis. Semin Oncol 29: 15-18  Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, Koos RD, Semenza GL (1996) Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol 16: 4604-4613  Friedl P, Brocker EB (2000) The biology of cell locomotion within three-dimensional extracellular matrix. Cellular and molecular life sciences : CMLS 57: 41-64  Friedl P, Zanker KS, Brocker EB (1998) Cell migration strategies in 3-D extracellular matrix: differences in morphology, cell matrix interactions, and integrin function. Microsc Res Tech 43: 369-378  Fu E, Pan L, Xie Y, Mu D, Liu W, Jin F, Bai X (2015) Tetraspanin CD63 is a regulator of HIV-1 replication. International journal of clinical and experimental pathology 8: 1184-1198  123  Funakoshi T, Tachibana I, Hoshida Y, Kimura H, Takeda Y, Kijima T, Nishino K, Goto H, Yoneda T, Kumagai T, Osaki T, Hayashi S, Aozasa K, Kawase I (2003) Expression of tetraspanins in human lung cancer cells: frequent downregulation of CD9 and its contribution to cell motility in small cell lung cancer. Oncogene 22: 674-687  Gadea G, Sanz-Moreno V, Self A, Godi A, Marshall CJ (2008) DOCK10-mediated Cdc42 activation is necessary for amoeboid invasion of melanoma cells. Current biology : CB 18: 1456-1465  Gajavelli S, Wood PM, Pennica D, Whittemore SR, Tsoulfas P (2004) BMP signaling initiates a neural crest differentiation program in embryonic rat CNS stem cells. Exp Neurol 188: 205-223  Galani E, Sgouros J, Petropoulou C, Janinis J, Aravantinos G, Dionysiou-Asteriou D, Skarlos D, Gonos E (2002) Correlation of MDR-1, nm23-H1 and H Sema E gene expression with histopathological findings and clinical outcome in ovarian and breast cancer patients. Anticancer Res 22: 2275-2280  Galbaugh T, Cerrito MG, Jose CC, Cutler ML (2006) EGF-induced activation of Akt results in mTOR-dependent p70S6 kinase phosphorylation and inhibition of HC11 cell lactogenic differentiation. BMC Cell Biol 7: 34  Gao K, Dai DL, Martinka M, Li G (2006) Prognostic significance of nuclear factor-kappaB p105/p50 in human melanoma and its role in cell migration. Cancer Res 66: 8382-8388  Gerber HP, Ferrara N (2005) Pharmacology and pharmacodynamics of bevacizumab as monotherapy or in combination with cytotoxic therapy in preclinical studies. Cancer Res 65: 671-680  Gerber HP, McMurtrey A, Kowalski J, Yan M, Keyt BA, Dixit V, Ferrara N (1998) Vascular endothelial growth factor regulates endothelial cell survival through the phosphatidylinositol 3'-kinase/Akt signal transduction pathway. Requirement for Flk-1/KDR activation. The Journal of biological chemistry 273: 30336-30343  Gille H, Kowalski J, Li B, LeCouter J, Moffat B, Zioncheck TF, Pelletier N, Ferrara N (2001) Analysis of biological effects and signaling properties of Flt-1 (VEGFR-1) and KDR (VEGFR-2). A reassessment using novel receptor-specific vascular endothelial growth factor mutants. The Journal of biological chemistry 276: 3222-3230  Gliki G, Abu-Ghazaleh R, Jezequel S, Wheeler-Jones C, Zachary I (2001) Vascular endothelial growth factor-induced prostacyclin production is mediated by a protein kinase C (PKC)-dependent activation of extracellular signal-regulated protein kinases 1 and 2 involving PKC-delta and by mobilization of intracellular Ca2+. The Biochemical journal 353: 503-512   124 Goodwin AM (2007) In vitro assays of angiogenesis for assessment of angiogenic and anti-angiogenic agents. Microvascular research 74: 172-183  Goshima Y, Ito T, Sasaki Y, Nakamura F (2002) Semaphorins as signals for cell repulsion and invasion. The Journal of clinical investigation 109: 993-998  Gray-Schopfer V, Wellbrock C, Marais R (2007) Melanoma biology and new targeted therapy. Nature 445: 851-857  Grichnik JM, Ali WN, Burch JA, Byers JD, Garcia CA, Clark RE, Shea CR (1996) KIT expression reveals a population of precursor melanocytes in human skin. The Journal of investigative dermatology 106: 967-971  Grigorov B, Attuil-Audenis V, Perugi F, Nedelec M, Watson S, Pique C, Darlix JL, Conjeaud H, Muriaux D (2009) A role for CD81 on the late steps of HIV-1 replication in a chronically infected T cell line. Retrovirology 6: 28  Gruss CJ, Satyamoorthy K, Berking C, Lininger J, Nesbit M, Schaider H, Liu ZJ, Oka M, Hsu MY, Shirakawa T, Li G, Bogenrieder T, Carmeliet P, El-Deiry WS, Eck SL, Rao JS, Baker AH, Bennet JT, Crombleholme TM, Velazquez O, Karmacharya J, Margolis DJ, Wilson JM, Detmar M, Skobe M, Robbins PD, Buck C, Herlyn M (2003) Stroma formation and angiogenesis by overexpression of growth factors, cytokines, and proteolytic enzymes in human skin grafted to SCID mice. The Journal of investigative dermatology 120: 683-692  Guo XZ, Friess H, Shao XD, Liu MP, Xia YT, Xu JH, Buchler MW (2000) KAI1 gene is differently expressed in papillary and pancreatic cancer: influence on metastasis. World journal of gastroenterology : WJG 6: 866-871  Guo Y, Xu F, Lu T, Duan Z, Zhang Z (2012) Interleukin-6 signaling pathway in targeted therapy for cancer. Cancer treatment reviews 38: 904-910  Gupta GP, Massague J (2006) Cancer metastasis: building a framework. Cell 127: 679-695  Hatakeyama M, Imaizumi T, Sakaki H, Yoshida H, Tanaka H, Kimura H, Fukuda I, Satoh K (2007) Interleukin-1 induces the expression of vascular endothelial growth factor in human pericardial mesothelial cells. Heart and vessels 22: 123-127  Hauschild A, Grob JJ, Demidov LV, Jouary T, Gutzmer R, Millward M, Rutkowski P, Blank CU, Miller WH, Jr., Kaempgen E, Martin-Algarra S, Karaszewska B, Mauch C, Chiarion-Sileni V, Martin AM, Swann S, Haney P, Mirakhur B, Guckert ME, Goodman V, Chapman PB (2012) Dabrafenib in BRAF-mutated metastatic melanoma: a multicentre, open-label, phase 3 randomised controlled trial. Lancet 380: 358-365  He B, Liu L, Cook GA, Grgurevich S, Jennings LK, Zhang XA (2005) Tetraspanin CD82 attenuates cellular morphogenesis through down-regulating integrin alpha6-mediated cell adhesion. The Journal of biological chemistry 280: 3346-3354  125  Hemler ME (2001) Specific tetraspanin functions. The Journal of cell biology 155: 1103-1107  Hemler ME (2005) Tetraspanin functions and associated microdomains. Nature reviews Molecular cell biology 6: 801-811  Heo SH, Choi YJ, Ryoo HM, Cho JY (2010) Expression profiling of ETS and MMP factors in VEGF-activated endothelial cells: role of MMP-10 in VEGF-induced angiogenesis. Journal of cellular physiology 224: 734-742  Herman JG, Meadows GG (2007) Increased class 3 semaphorin expression modulates the invasive and adhesive properties of prostate cancer cells. Int J Oncol 30: 1231-1238  Hermanns JF, Petit L, Martalo O, Pierard-Franchimont C, Cauwenbergh G, Pierard GE (2000) Unraveling the patterns of subclinical pheomelanin-enriched facial hyperpigmentation: effect of depigmenting agents. Dermatology 201: 118-122  Hibi M, Nakajima K, Hirano T (1996) IL-6 cytokine family and signal transduction: a model of the cytokine system. J Mol Med (Berl) 74: 1-12  Holmgren L, O'Reilly MS, Folkman J (1995) Dormancy of micrometastases: balanced proliferation and apoptosis in the presence of angiogenesis suppression. Nat Med 1: 149-153  Horejsi V, Vlcek C (1991) Novel structurally distinct family of leucocyte surface glycoproteins including CD9, CD37, CD53 and CD63. FEBS letters 288: 1-4  Hotta H, Miyamoto H, Hara I, Takahashi N, Homma M (1992) Genomic structure of the ME491/CD63 antigen gene and functional analysis of the 5'-flanking regulatory sequences. Biochem Biophys Res Commun 185: 436-442  Houghton AN, Polsky D (2002) Focus on melanoma. Cancer Cell 2: 275-278  Huang SP, Wu MS, Shun CT, Wang HP, Lin MT, Kuo ML, Lin JT (2004) Interleukin-6 increases vascular endothelial growth factor and angiogenesis in gastric carcinoma. J Biomed Sci 11: 517-527  Huber AB, Kolodkin AL, Ginty DD, Cloutier JF (2003) Signaling at the growth cone: ligand-receptor complexes and the control of axon growth and guidance. Annu Rev Neurosci 26: 509-563  Hurwitz RM, Buckel LJ (1997) Superficial congenital compound melanocytic nevus. Another pitfall in the diagnosis of malignant melanoma. Dermatologic surgery : official publication for American Society for Dermatologic Surgery [et al] 23: 897-900   126 Hwang JR, Jo K, Lee Y, Sung BJ, Park YW, Lee JH (2012) Upregulation of CD9 in ovarian cancer is related to the induction of TNF-alpha gene expression and constitutive NF-kappaB activation. Carcinogenesis 33: 77-83  Iannacone MR, Youlden DR, Baade PD, Aitken JF, Green AC (2015) Melanoma incidence trends and survival in adolescents and young adults in Queensland, Australia. International journal of cancer Journal international du cancer 136: 603-609  Ichikawa T, Ichikawa Y, Dong J, Hawkins AL, Griffin CA, Isaacs WB, Oshimura M, Barrett JC, Isaacs JT (1992) Localization of metastasis suppressor gene(s) for prostatic cancer to the short arm of human chromosome 11. Cancer Res 52: 3486-3490  Ito Y, Yoshida H, Uruno T, Nakano K, Takamura Y, Miya A, Kobayashi K, Yokozawa T, Matsuzuka F, Kuma K, Miyauchi A (2003) KAI1 expression in thyroid neoplasms: its linkage with clinicopathologic features in papillary carcinoma. Pathology, research and practice 199: 79-83  Jakobsson AE (1994) Angiogenesis induced by mast cell secretion in rat peritoneal connective tissue is a process of three phases. Microvascular research 47: 252-269  Jang HI, Lee H (2003) A decrease in the expression of CD63 tetraspanin protein elevates invasive potential of human melanoma cells. Experimental & molecular medicine 35: 317-323  Jee B, Jin K, Hahn JH, Song HG, Lee H (2003) Metastasis-suppressor KAI1/CD82 induces homotypic aggregation of human prostate cancer cells through Src-dependent pathway. Experimental & molecular medicine 35: 30-37  Jemal A, Siegel R, Xu J, Ward E (2010) Cancer statistics, 2010. CA: a cancer journal for clinicians 60: 277-300  Jiang BH, Liu LZ (2009) PI3K/PTEN signaling in angiogenesis and tumorigenesis. Adv Cancer Res 102: 19-65  Jiang YG, Luo Y, He DL, Li X, Zhang LL, Peng T, Li MC, Lin YH (2007) Role of Wnt/beta-catenin signaling pathway in epithelial-mesenchymal transition of human prostate cancer induced by hypoxia-inducible factor-1alpha. International journal of urology : official journal of the Japanese Urological Association 14: 1034-1039  Joseph EW, Pratilas CA, Poulikakos PI, Tadi M, Wang W, Taylor BS, Halilovic E, Persaud Y, Xing F, Viale A, Tsai J, Chapman PB, Bollag G, Solit DB, Rosen N (2010) The RAF inhibitor PLX4032 inhibits ERK signaling and tumor cell proliferation in a V600E BRAF-selective manner. Proceedings of the National Academy of Sciences of the United States of America 107: 14903-14908   127 Kallergi G, Agelaki S, Kalykaki A, Stournaras C, Mavroudis D, Georgoulias V (2008) Phosphorylated EGFR and PI3K/Akt signaling kinases are expressed in circulating tumor cells of breast cancer patients. Breast cancer research : BCR 10: R80  Kalluri R (2003) Basement membranes: structure, assembly and role in tumour angiogenesis. Nature reviews Cancer 3: 422-433  Kalluri R, Weinberg RA (2009) The basics of epithelial-mesenchymal transition. The Journal of clinical investigation 119: 1420-1428  Karakousis CP, Driscoll DL (1995) Prognostic parameters in localised melanoma: gender versus anatomical location. Eur J Cancer 31A: 320-324  Karar J, Maity A (2011) PI3K/AKT/mTOR Pathway in Angiogenesis. Frontiers in molecular neuroscience 4: 51  Kim J, Morley S, Le M, Bedoret D, Umetsu DT, Di Vizio D, Freeman MR (2014) Enhanced shedding of extracellular vesicles from amoeboid prostate cancer cells: potential effects on the tumor microenvironment. Cancer biology & therapy 15: 409-418  Kim JH, Kim B, Cai L, Choi HJ, Ohgi KA, Tran C, Chen C, Chung CH, Huber O, Rose DW, Sawyers CL, Rosenfeld MG, Baek SH (2005) Transcriptional regulation of a metastasis suppressor gene by Tip60 and beta-catenin complexes. Nature 434: 921-926  Kim NH, Kim HS, Li XY, Lee I, Choi HS, Kang SE, Cha SY, Ryu JK, Yoon D, Fearon ER, Rowe RG, Lee S, Maher CA, Weiss SJ, Yook JI (2011) A p53/miRNA-34 axis regulates Snail1-dependent cancer cell epithelial-mesenchymal transition. The Journal of cell biology 195: 417-433  Kischel P, Bellahcene A, Deux B, Lamour V, Dobson R, E DEP, Clezardin P, Castronovo V (2012) Overexpression of CD9 in human breast cancer cells promotes the development of bone metastases. Anticancer Res 32: 5211-5220  Kiss C, Li J, Szeles A, Gizatullin RZ, Kashuba VI, Lushnikova T, Protopopov AI, Kelve M, Kiss H, Kholodnyuk ID, Imreh S, Klein G, Zabarovsky ER (1997) Assignment of the ARHA and GPX1 genes to human chromosome bands 3p21.3 by in situ hybridization and with somatic cell hybrids. Cytogenet Cell Genet 79: 228-230  Kitadokoro K, Bordo D, Galli G, Petracca R, Falugi F, Abrignani S, Grandi G, Bolognesi M (2001) CD81 extracellular domain 3D structure: insight into the tetraspanin superfamily structural motifs. The EMBO journal 20: 12-18  Knobeloch KP, Wright MD, Ochsenbein AF, Liesenfeld O, Lohler J, Zinkernagel RM, Horak I, Orinska Z (2000) Targeted inactivation of the tetraspanin CD37 impairs T-cell-dependent B-cell response under suboptimal costimulatory conditions. Mol Cell Biol 20: 5363-5369   128 Kobayashi M, Salomon C, Tapia J, Illanes SE, Mitchell MD, Rice GE (2014) Ovarian cancer cell invasiveness is associated with discordant exosomal sequestration of Let-7 miRNA and miR-200. Journal of translational medicine 12: 4  Kolesnikova TV, Kazarov AR, Lemieux ME, Lafleur MA, Kesari S, Kung AL, Hemler ME (2009) Glioblastoma inhibition by cell surface immunoglobulin protein EWI-2, in vitro and in vivo. Neoplasia 11: 77-86, 74p following 86  Kopfstein L, Christofori G (2006) Metastasis: cell-autonomous mechanisms versus contributions by the tumor microenvironment. Cellular and molecular life sciences : CMLS 63: 449-468  Korinek V, Horejsi V (1993) Genomic structure of the human CD53 gene. Immunogenetics 38: 272-279  Kumar SM, Yu H, Edwards R, Chen L, Kazianis S, Brafford P, Acs G, Herlyn M, Xu X (2007) Mutant V600E BRAF increases hypoxia inducible factor-1alpha expression in melanoma. Cancer Res 67: 3177-3184  Laga AC, Murphy GF (2010) Cellular heterogeneity in vertical growth phase melanoma. Archives of pathology & laboratory medicine 134: 1750-1757  Lawson C, Lim ST, Uryu S, Chen XL, Calderwood DA, Schlaepfer DD (2012) FAK promotes recruitment of talin to nascent adhesions to control cell motility. The Journal of cell biology 196: 223-232  Le Naour F, Prenant M, Francastel C, Rubinstein E, Uzan G, Boucheix C (1996) Transcriptional regulation of the human CD9 gene: characterization of the 5'-flanking region. Oncogene 13: 481-486  Lee JH, Park SR, Chay KO, Seo YW, Kook H, Ahn KY, Kim YJ, Kim KK (2004) KAI1 COOH-terminal interacting tetraspanin (KITENIN), a member of the tetraspanin family, interacts with KAI1, a tumor metastasis suppressor, and enhances metastasis of cancer. Cancer Res 64: 4235-4243  Lee JH, Seo YW, Park SR, Kim YJ, Kim KK (2003) Expression of a splice variant of KAI1, a tumor metastasis suppressor gene, influences tumor invasion and progression. Cancer Res 63: 7247-7255  Lekmine F, Salti GI (2008) Induction of microphthalmia transcription factor (Mitf) by forskolin and stimulation of melanin release in UISO-Mel-6 cells. The Journal of surgical research 149: 27-30  Levy S, Shoham T (2005) Protein-protein interactions in the tetraspanin web. Physiology (Bethesda) 20: 218-224   129 Li J, Li G (2010) Cell cycle regulator ING4 is a suppressor of melanoma angiogenesis that is regulated by the metastasis suppressor BRMS1. Cancer Res 70: 10445-10453  Li J, Martinka M, Li G (2008) Role of ING4 in human melanoma cell migration, invasion and patient survival. Carcinogenesis 29: 1373-1379  Lijovic M, Somers G, Frauman AG (2002) KAI1/CD82 protein expression in primary prostate cancer and in BPH associated with cancer. Cancer Detect Prev 26: 69-77  Lin KK, Rossi L, Boles NC, Hall BE, George TC, Goodell MA (2011) CD81 is essential for the re-entry of hematopoietic stem cells to quiescence following stress-induced proliferation via deactivation of the Akt pathway. PLoS Biol 9: e1001148  Liu FS, Chen JT, Dong JT, Hsieh YT, Lin AJ, Ho ES, Hung MJ, Lu CH (2001) KAI1 metastasis suppressor gene is frequently down-regulated in cervical carcinoma. The American journal of pathology 159: 1629-1634  Liu FS, Dong JT, Chen JT, Hsieh YT, Ho ES, Hung MJ (2000) Frequent down-regulation and lack of mutation of the KAI1 metastasis suppressor gene in epithelial ovarian carcinoma. Gynecologic oncology 78: 10-15  Liu W, Iiizumi-Gairani M, Okuda H, Kobayashi A, Watabe M, Pai SK, Pandey PR, Xing F, Fukuda K, Modur V, Hirota S, Suzuki K, Chiba T, Endo M, Sugai T, Watabe K (2011a) KAI1 gene is engaged in NDRG1 gene-mediated metastasis suppression through the ATF3-NFkappaB complex in human prostate cancer. The Journal of biological chemistry 286: 18949-18959  Liu W, Kelly JW, Trivett M, Murray WK, Dowling JP, Wolfe R, Mason G, Magee J, Angel C, Dobrovic A, McArthur GA (2007) Distinct clinical and pathological features are associated with the BRAF(T1799A(V600E)) mutation in primary melanoma. The Journal of investigative dermatology 127: 900-905  Liu WM, Zhang XA (2006) KAI1/CD82, a tumor metastasis suppressor. Cancer Lett 240: 183-194  Liu X, Guo XZ, Zhang WW, Lu ZZ, Zhang QW, Duan HF, Wang LS (2011b) KAI1 inhibits HGF-induced invasion of pancreatic cancer by sphingosine kinase activity. Hepatobiliary & pancreatic diseases international : HBPD INT 10: 201-208  Liu Y, Mayo MW, Nagji AS, Smith PW, Ramsey CS, Li D, Jones DR (2012) Phosphorylation of RelA/p65 promotes DNMT-1 recruitment to chromatin and represses transcription of the tumor metastasis suppressor gene BRMS1. Oncogene 31: 1143-1154  Lu J, Rho O, Wilker E, Beltran L, Digiovanni J (2007) Activation of epidermal akt by diverse mouse skin tumor promoters. Molecular cancer research : MCR 5: 1342-1352   130 Madrid LV, Mayo MW, Reuther JY, Baldwin AS, Jr. (2001) Akt stimulates the transactivation potential of the RelA/p65 Subunit of NF-kappa B through utilization of the Ikappa B kinase and activation of the mitogen-activated protein kinase p38. The Journal of biological chemistry 276: 18934-18940  Mahabeleshwar GH, Byzova TV (2007) Angiogenesis in melanoma. Semin Oncol 34: 555-565  Malik FA, Sanders AJ, Kayani MA, Jiang WG (2009) Effect of expressional alteration of KAI1 on breast cancer cell growth, adhesion, migration and invasion. Cancer Genomics Proteomics 6: 205-213  Mamou F, May KS, Schipper MJ, Gill N, Kariapper MS, Nair BM, Brewer G, Normolle D, Khan MK (2006) Tetrathiomolybdate blocks bFGF- but not VEGF-induced incipient angiogenesis in vitro. Anticancer Res 26: 1753-1758  Mannion BA, Berditchevski F, Kraeft SK, Chen LB, Hemler ME (1996) Transmembrane-4 superfamily proteins CD81 (TAPA-1), CD82, CD63, and CD53 specifically associated with integrin alpha 4 beta 1 (CD49d/CD29). J Immunol 157: 2039-2047  Marquez-Rodas I, Martin Algarra S, Aviles Izquierdo JA, Custodio Cabello S, Martin M (2011) A new era in the treatment of melanoma: from biology to clinical practice. Clinical & translational oncology : official publication of the Federation of Spanish Oncology Societies and of the National Cancer Institute of Mexico 13: 787-792  Marreiros A, Dudgeon K, Dao V, Grimm MO, Czolij R, Crossley M, Jackson P (2005) KAI1 promoter activity is dependent on p53, junB and AP2: evidence for a possible mechanism underlying loss of KAI1 expression in cancer cells. Oncogene 24: 637-649  Martin-Satue M, Blanco J (1999) Identification of semaphorin E gene expression in metastatic human lung adenocarcinoma cells by mRNA differential display. Journal of surgical oncology 72: 18-23  Masciopinto F, Campagnoli S, Abrignani S, Uematsu Y, Pileri P (2001) The small extracellular loop of CD81 is necessary for optimal surface expression of the large loop, a putative HCV receptor. Virus Res 80: 1-10  Mashimo T, Watabe M, Hirota S, Hosobe S, Miura K, Tegtmeyer PJ, Rinker-Shaeffer CW, Watabe K (1998) The expression of the KAI1 gene, a tumor metastasis suppressor, is directly activated by p53. Proceedings of the National Academy of Sciences of the United States of America 95: 11307-11311  Meier F, Schittek B, Busch S, Garbe C, Smalley K, Satyamoorthy K, Li G, Herlyn M (2005) The RAS/RAF/MEK/ERK and PI3K/AKT signaling pathways present molecular targets for the effective treatment of advanced melanoma. Frontiers in bioscience : a journal and virtual library 10: 2986-3001  131  Menzies AM, Long GV (2013) New combinations and immunotherapies for melanoma: latest evidence and clinical utility. Therapeutic advances in medical oncology 5: 278-285  Mine M, Yamaguchi K, Sugiura T, Chigita S, Yoshihama N, Yoshihama R, Hiyake N, Kobayashi Y, Mori Y (2015) miR-203 Inhibits Frizzled-2 Expression via CD82/KAI1 Expression in Human Lung Carcinoma Cells. PloS one 10: e0131350  Miyake M, Nakano K, Itoi SI, Koh T, Taki T (1996) Motility-related protein-1 (MRP-1/CD9) reduction as a factor of poor prognosis in breast cancer. Cancer Res 56: 1244-1249  Miyamoto S, Teramoto H, Coso OA, Gutkind JS, Burbelo PD, Akiyama SK, Yamada KM (1995) Integrin function: molecular hierarchies of cytoskeletal and signaling molecules. The Journal of cell biology 131: 791-805  Miyato H, Tsuno NH, Kitayama J (2012) Semaphorin 3C is involved in the progression of gastric cancer. Cancer Sci 103: 1961-1966  Miyazaki T, Kato H, Shitara Y, Yoshikawa M, Tajima K, Masuda N, Shouji H, Tsukada K, Nakajima T, Kuwano H (2000) Mutation and expression of the metastasis suppressor gene KAI1 in esophageal squamous cell carcinoma. Cancer 89: 955-962  Mooez S, Malik FA, Kayani MA, Rashid R, Zahid A, Khan A (2011) Expressional alterations and transcript isoforms of metastasis suppressor genes (KAI1 and KiSS1) in breast cancer patients. Asian Pacific journal of cancer prevention : APJCP 12: 2785-2791  Moss ML, Lambert MH (2002) Shedding of membrane proteins by ADAM family proteases. Essays in biochemistry 38: 141-153  Muller U, Wang D, Denda S, Meneses JJ, Pedersen RA, Reichardt LF (1997) Integrin alpha8beta1 is critically important for epithelial-mesenchymal interactions during kidney morphogenesis. Cell 88: 603-613  Murisier F, Guichard S, Beermann F (2007) The tyrosinase enhancer is activated by Sox10 and Mitf in mouse melanocytes. Pigment cell research / sponsored by the European Society for Pigment Cell Research and the International Pigment Cell Society 20: 173-184  Muthukkaruppan VR, Kubai L, Auerbach R (1982) Tumor-induced neovascularization in the mouse eye. J Natl Cancer Inst 69: 699-708  Nagore E, Hueso L, Botella-Estrada R, Alfaro-Rubio A, Serna I, Guallar J, Gonzalez I, Ribes I, Guillen C (2010) Smoking, sun exposure, number of nevi and previous neoplasias are risk factors for melanoma in older patients (60 years and over). Journal of the European Academy of Dermatology and Venereology : JEADV 24: 50-57   132 Nakamura K, Iwamoto R, Mekada E (1995) Membrane-anchored heparin-binding EGF-like growth factor (HB-EGF) and diphtheria toxin receptor-associated protein (DRAP27)/CD9 form a complex with integrin alpha 3 beta 1 at cell-cell contact sites. The Journal of cell biology 129: 1691-1705  Nakamura K, Mitamura T, Takahashi T, Kobayashi T, Mekada E (2000) Importance of the major extracellular domain of CD9 and the epidermal growth factor (EGF)-like domain of heparin-binding EGF-like growth factor for up-regulation of binding and activity. The Journal of biological chemistry 275: 18284-18290  Neufeld G, Shraga-Heled N, Lange T, Guttmann-Raviv N, Herzog Y, Kessler O (2005) Semaphorins in cancer. Frontiers in bioscience : a journal and virtual library 10: 751-760  Nickoloff BJ, Hendrix MJ, Pollock PM, Trent JM, Miele L, Qin JZ (2005) Notch and NOXA-related pathways in melanoma cells. The journal of investigative dermatology Symposium proceedings / the Society for Investigative Dermatology, Inc [and] European Society for Dermatological Research 10: 95-104  Nishibori M, Cham B, McNicol A, Shalev A, Jain N, Gerrard JM (1993) The protein CD63 is in platelet dense granules, is deficient in a patient with Hermansky-Pudlak syndrome, and appears identical to granulophysin. The Journal of clinical investigation 91: 1775-1782  Nishida M, Miyagawa J, Yamashita S, Higashiyama S, Nakata A, Ouchi N, Tamura R, Yamamori K, Kihara S, Taniguchi N, Matsuzawa Y (2000) Localization of CD9, an enhancer protein for proheparin-binding epidermal growth factor-like growth factor, in human atherosclerotic plaques: possible involvement of juxtacrine growth mechanism on smooth muscle cell proliferation. Arterioscler Thromb Vasc Biol 20: 1236-1243  Odintsova E, van Niel G, Conjeaud H, Raposo G, Iwamoto R, Mekada E, Berditchevski F (2013) Metastasis suppressor tetraspanin CD82/KAI1 regulates ubiquitylation of epidermal growth factor receptor. The Journal of biological chemistry 288: 26323-26334  Ono M, Handa K, Withers DA, Hakomori S (1999) Motility inhibition and apoptosis are induced by metastasis-suppressing gene product CD82 and its analogue CD9, with concurrent glycosylation. Cancer Res 59: 2335-2339  Ono M, Handa K, Withers DA, Hakomori S (2000) Glycosylation effect on membrane domain (GEM) involved in cell adhesion and motility: a preliminary note on functional alpha3, alpha5-CD82 glycosylation complex in ldlD 14 cells. Biochem Biophys Res Commun 279: 744-750  Orlow SJ (1995) Melanosomes are specialized members of the lysosomal lineage of organelles. The Journal of investigative dermatology 105: 3-7   133 Otani A, Takagi H, Oh H, Suzuma K, Matsumura M, Ikeda E, Honda Y (2000) Angiotensin II-stimulated vascular endothelial growth factor expression in bovine retinal pericytes. Investigative ophthalmology & visual science 41: 1192-1199  Ovalle S, Gutierrez-Lopez MD, Olmo N, Turnay J, Lizarbe MA, Majano P, Molina-Jimenez F, Lopez-Cabrera M, Yanez-Mo M, Sanchez-Madrid F, Cabanas C (2007) The tetraspanin CD9 inhibits the proliferation and tumorigenicity of human colon carcinoma cells. International journal of cancer Journal international du cancer 121: 2140-2152  Palmer SR, Erickson LA, Ichetovkin I, Knauer DJ, Markovic SN (2011) Circulating serologic and molecular biomarkers in malignant melanoma. Mayo Clinic proceedings 86: 981-990  Palomba ML, Younes A (2013) In the spotlight: a novel CD37 antibody-drug conjugate. Blood 122: 3397-3398  Pankova K, Rosel D, Novotny M, Brabek J (2010) The molecular mechanisms of transition between mesenchymal and amoeboid invasiveness in tumor cells. Cellular and molecular life sciences : CMLS 67: 63-71  Paraiso KH, Xiang Y, Rebecca VW, Abel EV, Chen YA, Munko AC, Wood E, Fedorenko IV, Sondak VK, Anderson AR, Ribas A, Palma MD, Nathanson KL, Koomen JM, Messina JL, Smalley KS (2011) PTEN loss confers BRAF inhibitor resistance to melanoma cells through the suppression of BIM expression. Cancer Res 71: 2750-2760  Pfaff M, Du X, Ginsberg MH (1999) Calpain cleavage of integrin beta cytoplasmic domains. FEBS letters 460: 17-22  Phadke PA, Vaidya KS, Nash KT, Hurst DR, Welch DR (2008) BRMS1 suppresses breast cancer experimental metastasis to multiple organs by inhibiting several steps of the metastatic process. The American journal of pathology 172: 809-817  Piepkorn M (2000) Melanoma genetics: an update with focus on the CDKN2A(p16)/ARF tumor suppressors. Journal of the American Academy of Dermatology 42: 705-722; quiz 723-706  Potu H, Peterson LF, Pal A, Verhaegen M, Cao J, Talpaz M, Donato NJ (2014) Usp5 links suppression of p53 and FAS levels in melanoma to the BRAF pathway. Oncotarget 5: 5559-5569  Qin L, Bromberg-White JL, Qian CN (2012) Opportunities and challenges in tumor angiogenesis research: back and forth between bench and bed. Adv Cancer Res 113: 191-239  Quelle DE, Zindy F, Ashmun RA, Sherr CJ (1995) Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell 83: 993-1000  134  Radovic-Kovacevic V, Pekmezovic T, Adanja B, Jarebinski M, Marinkovic J, Tomin R (1997) [Survival analysis in patients with cutaneous malignant melanoma]. Srpski arhiv za celokupno lekarstvo 125: 132-137  Rak J, Yu JL, Klement G, Kerbel RS (2000) Oncogenes and angiogenesis: signaling three-dimensional tumor growth. The journal of investigative dermatology Symposium proceedings / the Society for Investigative Dermatology, Inc [and] European Society for Dermatological Research 5: 24-33  Rameh LE, Cantley LC (1999) The role of phosphoinositide 3-kinase lipid products in cell function. The Journal of biological chemistry 274: 8347-8350  Rasmussen AM, Blomhoff HK, Stokke T, Horejsi V, Smeland EB (1994) Cross-linking of CD53 promotes activation of resting human B lymphocytes. J Immunol 153: 4997-5007  Rasmussen HS, McCann PP (1997) Matrix metalloproteinase inhibition as a novel anticancer strategy: a review with special focus on batimastat and marimastat. Pharmacol Ther 75: 69-75  Remmele W, Stegner HE (1987) [Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue]. Pathologe 8: 138-140  Robert C, Long GV, Brady B, Dutriaux C, Maio M, Mortier L, Hassel JC, Rutkowski P, McNeil C, Kalinka-Warzocha E, Savage KJ, Hernberg MM, Lebbe C, Charles J, Mihalcioiu C, Chiarion-Sileni V, Mauch C, Cognetti F, Arance A, Schmidt H, Schadendorf D, Gogas H, Lundgren-Eriksson L, Horak C, Sharkey B, Waxman IM, Atkinson V, Ascierto PA (2015) Nivolumab in previously untreated melanoma without BRAF mutation. The New England journal of medicine 372: 320-330  Robert C, Ribas A, Wolchok JD, Hodi FS, Hamid O, Kefford R, Weber JS, Joshua AM, Hwu WJ, Gangadhar TC, Patnaik A, Dronca R, Zarour H, Joseph RW, Boasberg P, Chmielowski B, Mateus C, Postow MA, Gergich K, Elassaiss-Schaap J, Li XN, Iannone R, Ebbinghaus SW, Kang SP, Daud A (2014) Anti-programmed-death-receptor-1 treatment with pembrolizumab in ipilimumab-refractory advanced melanoma: a randomised dose-comparison cohort of a phase 1 trial. Lancet 384: 1109-1117  Rohlena J, Volger OL, van Buul JD, Hekking LH, van Gils JM, Bonta PI, Fontijn RD, Post JA, Hordijk PL, Horrevoets AJ (2009) Endothelial CD81 is a marker of early human atherosclerotic plaques and facilitates monocyte adhesion. Cardiovascular research 81: 187-196  Rothwangl KB, Manicassamy B, Uprichard SL, Rong L (2008) Dissecting the role of putative CD81 binding regions of E2 in mediating HCV entry: putative CD81 binding region 1 is not involved in CD81 binding. Virology journal 5: 46  135  Rotterud R, Fossa SD, Nesland JM (2007) Protein networking in bladder cancer: immunoreactivity for FGFR3, EGFR, ERBB2, KAI1, PTEN, and RAS in normal and malignant urothelium. Histology and histopathology 22: 349-363  Rubinstein E, Le Naour F, Lagaudriere-Gesbert C, Billard M, Conjeaud H, Boucheix C (1996) CD9, CD63, CD81, and CD82 are components of a surface tetraspan network connected to HLA-DR and VLA integrins. European journal of immunology 26: 2657-2665  Sahai E, Marshall CJ (2003) Differing modes of tumour cell invasion have distinct requirements for Rho/ROCK signalling and extracellular proteolysis. Nat Cell Biol 5: 711-719  Saladi SV, Wong PG, Trivedi AR, Marathe HG, Keenen B, Aras S, Liew ZQ, Setaluri V, de la Serna IL (2013) BRG1 promotes survival of UV-irradiated melanoma cells by cooperating with MITF to activate the melanoma inhibitor of apoptosis gene. Pigment cell & melanoma research 26: 377-391  Saldanha G, Potter L, Daforno P, Pringle JH (2006) Cutaneous melanoma subtypes show different BRAF and NRAS mutation frequencies. Clinical cancer research : an official journal of the American Association for Cancer Research 12: 4499-4505  Sausville EA (2012) Promises from trametinib in RAF active tumors. The New England journal of medicine 367: 171-172  Schienbein M, Gruler H (1995) Chemical amplifier, self-ignition mechanism, and amoeboid cell migration. Physical review E, Statistical physics, plasmas, fluids, and related interdisciplinary topics 52: 4183-4197  Schoenfeld N, Bauer MK, Grimm S (2004) The metastasis suppressor gene C33/CD82/KAI1 induces apoptosis through reactive oxygen intermediates. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 18: 158-160  Seigneuret M, Delaguillaumie A, Lagaudriere-Gesbert C, Conjeaud H (2001) Structure of the tetraspanin main extracellular domain. A partially conserved fold with a structurally variable domain insertion. The Journal of biological chemistry 276: 40055-40064  Shah DJ, Dronca RS (2014) Latest advances in chemotherapeutic, targeted, and immune approaches in the treatment of metastatic melanoma. Mayo Clinic proceedings 89: 504-519  Shi W, Fan H, Shum L, Derynck R (2000) The tetraspanin CD9 associates with transmembrane TGF-alpha and regulates TGF-alpha-induced EGF receptor activation and cell proliferation. The Journal of cell biology 148: 591-602   136 Shinohara T, Miki T, Nishimura N, Nokihara H, Hamada H, Mukaida N, Sone S (2001) Nuclear factor-kappaB-dependent expression of metastasis suppressor KAI1/CD82 gene in lung cancer cell lines expressing mutant p53. Cancer Res 61: 673-678  Shiojima I, Walsh K (2002) Role of Akt signaling in vascular homeostasis and angiogenesis. Circ Res 90: 1243-1250  Shiseki M, Nagashima M, Pedeux RM, Kitahama-Shiseki M, Miura K, Okamura S, Onogi H, Higashimoto Y, Appella E, Yokota J, Harris CC (2003) p29ING4 and p28ING5 bind to p53 and p300, and enhance p53 activity. Cancer Res 63: 2373-2378  Siegel R, Ma J, Zou Z, Jemal A (2014) Cancer statistics, 2014. CA: a cancer journal for clinicians 64: 9-29  Sinnberg T, Lasithiotakis K, Niessner H, Schittek B, Flaherty KT, Kulms D, Maczey E, Campos M, Gogel J, Garbe C, Meier F (2009) Inhibition of PI3K-AKT-mTOR signaling sensitizes melanoma cells to cisplatin and temozolomide. The Journal of investigative dermatology 129: 1500-1515  Skotnicki JS, Zask A, Nelson FC, Albright JD, Levin JI (1999) Design and synthetic considerations of matrix metalloproteinase inhibitors. Ann N Y Acad Sci 878: 61-72  Soengas MS, Lowe SW (2003) Apoptosis and melanoma chemoresistance. Oncogene 22: 3138-3151  Soker S, Takashima S, Miao HQ, Neufeld G, Klagsbrun M (1998) Neuropilin-1 is expressed by endothelial and tumor cells as an isoform-specific receptor for vascular endothelial growth factor. Cell 92: 735-745  Sosman JA, Kim KB, Schuchter L, Gonzalez R, Pavlick AC, Weber JS, McArthur GA, Hutson TE, Moschos SJ, Flaherty KT, Hersey P, Kefford R, Lawrence D, Puzanov I, Lewis KD, Amaravadi RK, Chmielowski B, Lawrence HJ, Shyr Y, Ye F, Li J, Nolop KB, Lee RJ, Joe AK, Ribas A (2012) Survival in BRAF V600-mutant advanced melanoma treated with vemurafenib. The New England journal of medicine 366: 707-714  Sridhar SC, Miranti CK (2006) Tetraspanin KAI1/CD82 suppresses invasion by inhibiting integrin-dependent crosstalk with c-Met receptor and Src kinases. Oncogene 25: 2367-2378  Stahl JM, Cheung M, Sharma A, Trivedi NR, Shanmugam S, Robertson GP (2003) Loss of PTEN promotes tumor development in malignant melanoma. Cancer Res 63: 2881-2890  Staton CA, Reed MW, Brown NJ (2009) A critical analysis of current in vitro and in vivo angiogenesis assays. International journal of experimental pathology 90: 195-221  Stemke-Hale K, Gonzalez-Angulo AM, Lluch A, Neve RM, Kuo WL, Davies M, Carey M, Hu Z, Guan Y, Sahin A, Symmans WF, Pusztai L, Nolden LK, Horlings H, Berns K, Hung  137 MC, van de Vijver MJ, Valero V, Gray JW, Bernards R, Mills GB, Hennessy BT (2008) An integrative genomic and proteomic analysis of PIK3CA, PTEN, and AKT mutations in breast cancer. Cancer Res 68: 6084-6091  Stipp CS, Kolesnikova TV, Hemler ME (2003) Functional domains in tetraspanin proteins. Trends Biochem Sci 28: 106-112  Sullivan RJ, Flaherty KT (2013) Resistance to BRAF-targeted therapy in melanoma. Eur J Cancer 49: 1297-1304  Sumimoto H, Imabayashi F, Iwata T, Kawakami Y (2006) The BRAF-MAPK signaling pathway is essential for cancer-immune evasion in human melanoma cells. The Journal of experimental medicine 203: 1651-1656  Tamagnone L, Artigiani S, Chen H, He Z, Ming GI, Song H, Chedotal A, Winberg ML, Goodman CS, Poo M, Tessier-Lavigne M, Comoglio PM (1999) Plexins are a large family of receptors for transmembrane, secreted, and GPI-anchored semaphorins in vertebrates. Cell 99: 71-80  Tang JF, Wen Q, Sun J, Zhang WM, Zhu HL (2014a) Advances in the Researches on the Biological Activities and Inhibitors of Phosphatidylinositol 3-kinase. Anti-cancer agents in medicinal chemistry 14: 673-687  Tang Y, Cheng Y, Martinka M, Ong CJ, Li G (2014b) Prognostic significance of KAI1/CD82 in human melanoma and its role in cell migration and invasion through the regulation of ING4. Carcinogenesis 35: 86-95  Telese F, Bruni P, Donizetti A, Gianni D, D'Ambrosio C, Scaloni A, Zambrano N, Rosenfeld MG, Russo T (2005) Transcription regulation by the adaptor protein Fe65 and the nucleosome assembly factor SET. EMBO reports 6: 77-82  Thiery JP (2002) Epithelial-mesenchymal transitions in tumour progression. Nature reviews Cancer 2: 442-454  Tokuhara T, Hasegawa H, Hattori N, Ishida H, Taki T, Tachibana S, Sasaki S, Miyake M (2001) Clinical significance of CD151 gene expression in non-small cell lung cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 7: 4109-4114  Tominaga N, Hagiwara K, Kosaka N, Honma K, Nakagama H, Ochiya T (2014) RPN2-mediated glycosylation of tetraspanin CD63 regulates breast cancer cell malignancy. Molecular cancer 13: 134  Tsai YC, Mendoza A, Mariano JM, Zhou M, Kostova Z, Chen B, Veenstra T, Hewitt SM, Helman LJ, Khanna C, Weissman AM (2007) The ubiquitin ligase gp78 promotes sarcoma metastasis by targeting KAI1 for degradation. Nat Med 13: 1504-1509  138  Tsao H, Sober AJ (2005) Melanoma treatment update. Dermatologic clinics 23: 323-333  Tsuda M, Makino Y, Iwahara T, Nishihara H, Sawa H, Nagashima K, Hanafusa H, Tanaka S (2004) Crk associates with ERM proteins and promotes cell motility toward hyaluronic acid. The Journal of biological chemistry 279: 46843-46850  Tugues S, Honjo S, Konig C, Padhan N, Kroon J, Gualandi L, Li X, Barkefors I, Thijssen VL, Griffioen AW, Claesson-Welsh L (2013) Tetraspanin CD63 promotes vascular endothelial growth factor receptor 2-beta1 integrin complex formation, thereby regulating activation and downstream signaling in endothelial cells in vitro and in vivo. The Journal of biological chemistry 288: 19060-19071  Tzeng HE, Tsai CH, Chang ZL, Su CM, Wang SW, Hwang WL, Tang CH (2013) Interleukin-6 induces vascular endothelial growth factor expression and promotes angiogenesis through apoptosis signal-regulating kinase 1 in human osteosarcoma. Biochem Pharmacol 85: 531-540  Vaisanen A, Tuominen H, Kallioinen M, Turpeenniemi-Hujanen T (1996) Matrix metalloproteinase-2 (72 kD type IV collagenase) expression occurs in the early stage of human melanocytic tumour progression and may have prognostic value. The Journal of pathology 180: 283-289  Vancoillie G, Lambert J, Nayaert JM (1999) Melanocyte biology and its implications for the clinician. European journal of dermatology : EJD 9: 241-251  Visse R, Nagase H (2003) Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res 92: 827-839  Wang J, Chin MY, Li G (2006) The novel tumor suppressor p33ING2 enhances nucleotide excision repair via inducement of histone H4 acetylation and chromatin relaxation. Cancer Res 66: 1906-1911  Wang X, McCullough KD, Franke TF, Holbrook NJ (2000) Epidermal growth factor receptor-dependent Akt activation by oxidative stress enhances cell survival. The Journal of biological chemistry 275: 14624-14631  Wang Y, Li G (2006) ING3 promotes UV-induced apoptosis via Fas/caspase-8 pathway in melanoma cells. The Journal of biological chemistry 281: 11887-11893  Wani AA, Jafarnejad SM, Zhou J, Li G (2011) Integrin-linked kinase regulates melanoma angiogenesis by activating NF-kappaB/interleukin-6 signaling pathway. Oncogene 30: 2778-2788   139 Wehbe H, Henson R, Meng F, Mize-Berge J, Patel T (2006) Interleukin-6 contributes to growth in cholangiocarcinoma cells by aberrant promoter methylation and gene expression. Cancer Res 66: 10517-10524  Weidner N (1995) Current pathologic methods for measuring intratumoral microvessel density within breast carcinoma and other solid tumors. Breast cancer research and treatment 36: 169-180  Weidner N, Semple JP, Welch WR, Folkman J (1991) Tumor angiogenesis and metastasis--correlation in invasive breast carcinoma. The New England journal of medicine 324: 1-8  Weis SM, Cheresh DA (2011) Tumor angiogenesis: molecular pathways and therapeutic targets. Nat Med 17: 1359-1370  Wellbrock C, Rana S, Paterson H, Pickersgill H, Brummelkamp T, Marais R (2008) Oncogenic BRAF regulates melanoma proliferation through the lineage specific factor MITF. PloS one 3: e2734  Wilson CA, Tsuchida MA, Allen GM, Barnhart EL, Applegate KT, Yam PT, Ji L, Keren K, Danuser G, Theriot JA (2010) Myosin II contributes to cell-scale actin network treadmilling through network disassembly. Nature 465: 373-377  Winterwood NE, Varzavand A, Meland MN, Ashman LK, Stipp CS (2006) A critical role for tetraspanin CD151 in alpha3beta1 and alpha6beta4 integrin-dependent tumor cell functions on laminin-5. Molecular biology of the cell 17: 2707-2721  Wolf K, Mazo I, Leung H, Engelke K, von Andrian UH, Deryugina EI, Strongin AY, Brocker EB, Friedl P (2003a) Compensation mechanism in tumor cell migration: mesenchymal-amoeboid transition after blocking of pericellular proteolysis. The Journal of cell biology 160: 267-277  Wolf K, Muller R, Borgmann S, Brocker EB, Friedl P (2003b) Amoeboid shape change and contact guidance: T-lymphocyte crawling through fibrillar collagen is independent of matrix remodeling by MMPs and other proteases. Blood 102: 3262-3269  Wong PP, Demircioglu F, Ghazaly E, Alrawashdeh W, Stratford MR, Scudamore CL, Cereser B, Crnogorac-Jurcevic T, McDonald S, Elia G, Hagemann T, Kocher HM, Hodivala-Dilke KM (2015) Dual-action combination therapy enhances angiogenesis while reducing tumor growth and spread. Cancer Cell 27: 123-137  Wong RP, Ng P, Dedhar S, Li G (2007) The role of integrin-linked kinase in melanoma cell migration, invasion, and tumor growth. Mol Cancer Ther 6: 1692-1700  Worobec SM, Solomon LM (1978) Structure and function of the skin. Major problems in clinical pediatrics 19: 7-27   140 Wright MD, Tomlinson MG (1994) The ins and outs of the transmembrane 4 superfamily. Immunol Today 15: 588-594  Wyckoff JB, Pinner SE, Gschmeissner S, Condeelis JS, Sahai E (2006) ROCK- and myosin-dependent matrix deformation enables protease-independent tumor-cell invasion in vivo. Current biology : CB 16: 1515-1523  Xie L, Law BK, Chytil AM, Brown KA, Aakre ME, Moses HL (2004) Activation of the Erk pathway is required for TGF-beta1-induced EMT in vitro. Neoplasia 6: 603-610  Xu J, Lamouille S, Derynck R (2009) TGF-beta-induced epithelial to mesenchymal transition. Cell Res 19: 156-172  Yajima I, Kumasaka MY, Thang ND, Goto Y, Takeda K, Yamanoshita O, Iida M, Ohgami N, Tamura H, Kawamoto Y, Kato M (2012) RAS/RAF/MEK/ERK and PI3K/PTEN/AKT Signaling in Malignant Melanoma Progression and Therapy. Dermatology research and practice 2012: 354191  Yamada T, Hasegawa S, Inoue Y, Date Y, Yamamoto N, Mizutani H, Nakata S, Matsunaga K, Akamatsu H (2013) Wnt/beta-catenin and kit signaling sequentially regulate melanocyte stem cell differentiation in UVB-induced epidermal pigmentation. The Journal of investigative dermatology 133: 2753-2762  Yamazaki D, Kurisu S, Takenawa T (2005) Regulation of cancer cell motility through actin reorganization. Cancer Sci 96: 379-386  Yanez-Mo M, Tejedor R, Rousselle P, Sanchez -Madrid F (2001) Tetraspanins in intercellular adhesion of polarized epithelial cells: spatial and functional relationship to integrins and cadherins. J Cell Sci 114: 577-587  Yang J, Weinberg RA (2008) Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell 14: 818-829  Yang X, Welch DR, Phillips KK, Weissman BE, Wei LL (1997) KAI1, a putative marker for metastatic potential in human breast cancer. Cancer Lett 119: 149-155  Yarden Y, Sliwkowski MX (2001) Untangling the ErbB signalling network. Nature reviews Molecular cell biology 2: 127-137  Young HE, Mancini ML, Wright RP, Smith JC, Black AC, Jr., Reagan CR, Lucas PA (1995) Mesenchymal stem cells reside within the connective tissues of many organs. Developmental dynamics : an official publication of the American Association of Anatomists 202: 137-144  Yu Y, Yang JL, Markovic B, Jackson P, Yardley G, Barrett J, Russell PJ (1997) Loss of KAI1 messenger RNA expression in both high-grade and invasive human bladder cancers.  141 Clinical cancer research : an official journal of the American Association for Cancer Research 3: 1045-1049  Yumura S, Mori H, Fukui Y (1984) Localization of actin and myosin for the study of ameboid movement in Dictyostelium using improved immunofluorescence. The Journal of cell biology 99: 894-899  Zhang D, Pier T, McNeel DG, Wilding G, Friedl A (2007) Effects of a monoclonal anti-alphavbeta3 integrin antibody on blood vessels - a pharmacodynamic study. Investigational new drugs 25: 49-55  Zhang G, Ye X, Ji D, Zhang H, Sun F, Shang C, Zhang Y, Wu E, Wang F, Wu F, Tian H, Liu X, Chen L, Liu K, Wang Y, Liu H, Zhang W, Guan Y, Wang Q, Zhao X, Wan X (2014) Inhibition of lung tumor growth by targeting EGFR/VEGFR-Akt/NF-kappaB pathways with novel theanine derivatives. Oncotarget 5: 8528-8543  Zhang XA, Bontrager AL, Hemler ME (2001) Transmembrane-4 superfamily proteins associate with activated protein kinase C (PKC) and link PKC to specific beta(1) integrins. The Journal of biological chemistry 276: 25005-25013  Zhang XA, He B, Zhou B, Liu L (2003a) Requirement of the p130CAS-Crk coupling for metastasis suppressor KAI1/CD82-mediated inhibition of cell migration. The Journal of biological chemistry 278: 27319-27328  Zhang XA, Lane WS, Charrin S, Rubinstein E, Liu L (2003b) EWI2/PGRL associates with the metastasis suppressor KAI1/CD82 and inhibits the migration of prostate cancer cells. Cancer Res 63: 2665-2674  Zheng H, Takahashi H, Murai Y, Cui Z, Nomoto K, Niwa H, Tsuneyama K, Takano Y (2006) Expressions of MMP-2, MMP-9 and VEGF are closely linked to growth, invasion, metastasis and angiogenesis of gastric carcinoma. Anticancer Res 26: 3579-3583  Zheng H, Tsuneyama K, Cheng C, Takahashi H, Cui Z, Nomoto K, Murai Y, Takano Y (2007) Expression of KAI1 and tenascin, and microvessel density are closely correlated with liver metastasis of gastrointestinal adenocarcinoma. J Clin Pathol 60: 50-56  Zhou B, Liu L, Reddivari M, Zhang XA (2004) The palmitoylation of metastasis suppressor KAI1/CD82 is important for its motility- and invasiveness-inhibitory activity. Cancer Res 64: 7455-7463  Zins K, Thomas A, Lucas T, Sioud M, Aharinejad S, Abraham D (2013) Inhibition of stromal PlGF suppresses the growth of prostate cancer xenografts. International journal of molecular sciences 14: 17958-17971   142 Zondag GC, Evers EE, ten Klooster JP, Janssen L, van der Kammen RA, Collard JG (2000) Oncogenic Ras downregulates Rac activity, which leads to increased Rho activity and epithelial-mesenchymal transition. The Journal of cell biology 149: 775-782    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0224025/manifest

Comment

Related Items