UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Clinical application and functional characterization of TOX in cutaneous T-cell lymphoma Huang, Yuanshen 2016

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

Item Metadata


24-ubc_2016_may_huang_yuanshen.pdf [ 3.99MB ]
JSON: 24-1.0300011.json
JSON-LD: 24-1.0300011-ld.json
RDF/XML (Pretty): 24-1.0300011-rdf.xml
RDF/JSON: 24-1.0300011-rdf.json
Turtle: 24-1.0300011-turtle.txt
N-Triples: 24-1.0300011-rdf-ntriples.txt
Original Record: 24-1.0300011-source.json
Full Text

Full Text

CLINICAL APPLICATION AND FUNCTIONAL CHARACTERIZATION OF TOX IN CUTANEOUS T-CELL LYMPHOMA  by  Yuanshen Huang  B. Med, Peking University Health Science Center, 2007 MD, Peking University Health Science Center, 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)    April 2016  © Yuanshen Huang, 2016  ii Abstract Cutaneous T-cell lymphoma (CTCL) is a group of lymphoproliferative disorders consisting of two main subtypes: mycosis fungoides (MF) and Sézary syndrome (SS). Due to the lack of robust histological markers, it remains a challenge to establish an accurate diagnosis and offer long term prognostication for CTCL. In addition, the molecular pathogenesis of CTCL is only partially understood. Previously our group discovered that early stage MF skin biopsies contained ectopic expression of TOX gene, which is essential for the early development of CD4+ T cells but normally is switched off in mature CD4+ T cells in the peripheral tissues. The objectives of my thesis research are to evaluate if TOX can be used to improve CTCL diagnosis and prognostication, and to characterize the functional role of TOX in the pathogenesis of CTCL.   Using skin biopsies and clinical databases from Vancouver, Beijing and Boston, I confirmed that TOX expression levels were significantly upregulated in the full spectrum of MF and in SS. In addition, as a diagnostic marker, high TOX expression levels differentiated CTCL from non-CTCL controls with good sensitivity and specificity. Furthermore, as a prognostic marker, high TOX mRNA levels correlated with increased risks of disease progression and disease-specific mortality in MF, and increased risks of disease-specific mortality in SS. I also investigated the functional role of TOX in CTCL pathogenesis using multiple CTCL cell lines and a mouse xenograft model. TOX knockdown in three CTCL cell lines led to markedly increased apoptosis, reduced cell proliferation, and impaired tumorigenic ability. These effects were partially mediated by increased expression of two cell cycle regulators, CDKN1B and CDKN1C. In addition, transcriptome analysis between TOX-suppressed cells  iii and control CTCL cells uncovered additional potential molecules downstream of TOX, such as tumor suppressors FOXO3 and HBP1.  Our results provide strong evidence that aberrant activation of TOX can serve as a diagnostic and prognostic biomarker for CTCL. Further, we demonstrated that TOX plays a crucial oncogenic role in CTCL pathogenesis, partially through regulating transcription of CDKN1B, CDKN1C and other downstream genes. Therefore TOX and/or its downstream genes may be promising therapeutic targets for CTCL.  iv Preface 1. A version of chapter 3 has been published. [Huang Y], Litvinov IV, Wang Y, Su MW, Tu P, Jiang X, Kupper TS, Dutz JP, Sasseville D, Zhou Y. Thymocyte selection-associated high mobility group box gene (TOX) is aberrantly over-expressed in mycosis fungoides and correlates with poor prognosis. Oncotarget. 2014 Jun 30;5(12):4418-25. I collected and analyzed the data, prepared most of the figures/tables and the manuscript under the supervision of Drs. Y. Zhou and X. Jiang. Drs. Y. Wang, P. Tu, Y. Zhou and JP. Dutz provided clinical samples in MF cohort 1. Drs. IV. Litvinov and TS. Kupper provided TOX mRNA expression levels and clinical outcome data for MF patient cohort 2. Dr. MW. Su assisted with skin biopsy processing, extracted RNA and performed quantitative PCR for TOX expression in MF patient cohort 1. Drs. Youwen Zhou and Denis Sasseville contributed to the conception and design of the study.  2. A version of chapter 4 and some of the results in Chapter 3, including section 3.2.2, 3.2.5, and part of section 3.2.3 have been published. [Huang Y], Su MW, Jiang X, Zhou Y. (2015) Evidence of an oncogenic role of aberrant TOX activation in cutaneous T-cell lymphoma. Blood. 2015 Feb 26;125(9):1435-43.  I designed and performed the majority of the experiments, analyzed all the data, and prepared the manuscript. Dr. MW. Su contributed to experimental design and data analysis, extracted RNA and performed quantitative PCR for TOX expression in CTCL patients and control participants, and provided intellectual input on the manuscript. Drs. Y. Zhou and X. Jiang provided facilities and research materials, conceived and designed the study, and contributed to manuscript preparation and revision.     v Ethics certificate: The use of primary human samples in this study was approved by the Clinical Research Ethics Board of University of British Columbia (certificate numbers are H12-02653 and C98-0493). The mouse work involved in this study was performed under UBC Animal Care protocol A11-0005 (Dr. Xiaoyan Jiang). Mice were housed in the Animal Resource Centre, British Columbia Cancer Research Centre.   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 ............................................................................................................. xvii Dedication ............................................................................................................................. xix Chapter 1: Introduction ......................................................................................................... 1 1.1 T-cell Biology and Function ................................................................................................... 1 1.2 T-cell Development................................................................................................................. 5 1.3 Cutaneous T-cell Lymphoma (CTCL) .................................................................................... 9 1.3.1 Mycosis Fungoides (MF) .............................................................................................. 10 1.3.2 Sézary Syndrome (SS) .................................................................................................. 13 1.3.3 Disease Pathogenesis ..................................................................................................... 16 1.3.4 Chromosomal, Genetic and Molecular Findings of CTCL ........................................... 18 1.3.5 Biomarkers of CTCL ..................................................................................................... 21 1.3.6 Treatment of CTCL ....................................................................................................... 24 1.4 TOX, a Critical T-cell Regulator .......................................................................................... 26 1.4.1 TOX Structure and Expression ...................................................................................... 27 1.4.2 TOX Involvement in T-cell Development .................................................................... 31 1.4.3 TOX beyond T-cell Development ................................................................................. 34  vii 1.4.4 TOX in T-cell Malignancies .......................................................................................... 35 TOX in T-cell Acute Lymphoblastic Leukemia (T-ALL) ................................. 35 TOX in MF ........................................................................................................ 36 1.5 Thesis Outline ....................................................................................................................... 36 1.5.1 Research Hypothesis ..................................................................................................... 36 1.5.2 Specific Aims ................................................................................................................ 37 Chapter 2: Materials and Methods ..................................................................................... 39 2.1 Primary Human Samples ...................................................................................................... 39 2.2 Cell Lines and Cell Culture ................................................................................................... 48 2.3 Lentiviral Vectors ................................................................................................................. 49 2.4 Production of Lentiviral Particles ......................................................................................... 51 2.5 Generation of Stably Transduced Cells................................................................................. 51 2.6 Protein Extraction and Western Blotting .............................................................................. 52 2.7 RNA Extraction and Reverse Transcription ......................................................................... 53 2.8 Real-time Quantitative Polymerase Chain Reaction ............................................................. 53 2.9 Cell Viability Assay .............................................................................................................. 56 2.10 Colony-forming Cell Assay .................................................................................................. 56 2.11 Activation of CD4+ T Cells ................................................................................................... 56 2.12 Fluorescence-activated Cell Sorting (FACS) Analysis ......................................................... 57 2.13 Apoptosis Assay .................................................................................................................... 57 2.14 5-bromo-2'-deoxyuridine (BrdU) Incorporation Assay ........................................................ 57 2.15 Immunofluorescence (IF) Staining ....................................................................................... 58 2.16 Animals and Tumor Formation Assay .................................................................................. 58 2.17 Microarray Analysis .............................................................................................................. 59 2.18 Statistical Analysis ................................................................................................................ 60  viii Chapter 3: The Potential of TOX as a Disease Marker for CTCL .................................. 62 3.1 Background and Rationale .................................................................................................... 62 3.2 Results ................................................................................................................................... 63 3.2.1 TOX Is Upregulated in Skin Biopsies from MF Patients of All Stages ........................ 63 3.2.2 TOX Is Upregulated in Primary Sézary Cells and CTCL Cell Lines ............................ 67 3.2.3 Increased TOX Levels Differentiate CTCL from Benign Inflammatory Dermatoses .. 72 3.2.4 High TOX Levels Correlate with Increased Risk of Disease Progression and Disease-specific Mortality in MF Patients ............................................................................................... 74 3.2.5 High TOX Levels Correlate with Increased Disease-specific Mortality in SS ............. 76 3.3 Discussion ............................................................................................................................. 78 Chapter 4: The Role of TOX in the Development of CTCL ............................................. 82 4.1 Background and Rationale .................................................................................................... 82 4.2 Results ................................................................................................................................... 83 4.2.1 Reduced TOX in CTCL Cells Results in Marked Reduction in Cellular Proliferation and Colony Formation in Vitro .................................................................................................. 83 4.2.2 TOX Inhibition Abolishes or Halts Tumor Formation by CTCL Cells in Vivo ............ 86 4.2.3 Reduced TOX Sensitizes CTCL Cells to Apoptosis ..................................................... 89 4.2.4 Reduced TOX Causes Cell-cycle Progression Disruption in CTCL Cells .................... 91 4.2.5 TOX Reduction Restores Cell Cycle Regulators, CDKN1B and CDKN1C ................. 93 4.2.6 Inhibition of CDKN1B or CDKN1C in TOX-suppressed CTCL Cells Partially Rescues the Growth Inhibition by TOX Suppression .............................................................................. 95 4.2.7 Gene Expression Profiling in TOX-suppressed CTCL Cells ........................................ 98 4.2.8 TOX Upregulation Is Unlikely a Result of TCR Activation in Mature CD4+ T Cells 107 4.3 Discussion ........................................................................................................................... 109 Chapter 5: Summary and Future Directions ................................................................... 114  ix 5.1 Summary ............................................................................................................................. 114 5.2 Significance and Limitations .............................................................................................. 118 5.3 Future Directions ................................................................................................................ 121 Bibliography ........................................................................................................................ 125 Appendix: Additional Publications ................................................................................... 150    x  List of Tables Table 2.1 Summary of demographics for patients with MF (n = 113) ................................... 41 Table 2.2 Summary of demographics for patients with SS (n = 12)....................................... 41 Table 2.3 Demographics and clinical characteristics of individual MF patients (n = 113) .... 42 Table 2.4 Demographics and clinical characteristics of individual SS patients (n = 12) ....... 47 Table 2.5 Oligonucleotides encoding the shRNAs ................................................................. 50 Table 2.6 Primary antibodies used in Western blotting .......................................................... 53 Table 2.7 Primers for qPCR .................................................................................................... 55 Table 3.1 Multivariate analyses of disease progression and disease-specific survival in mycosis fungoides subjects using COX proportional hazards model ..................................... 75 Table 4.1 qPCR validation of the 22 differentially expressed genes identified by gene expression profiling .............................................................................................................. 101 Table 4.2 Biological processes related to the top differentially expressed genes in TOX-suppressed CTCL cells as identified by Genomatrix software ............................................. 102    xi List of Figures Figure 1.1 Overview of T-cell development in the thymus. ..................................................... 8 Figure 1.2 Clinical manifestations of mycosis fungoides and Sézary syndrome ................... 15 Figure 1.3 The TOX subfamily of HMG-box proteins. .......................................................... 30 Figure 1.4 Mode of TOX expression during T-cell development. ......................................... 33 Figure 3.1 TOX mRNA levels are increased in MF skin biopsies. ........................................ 65 Figure 3.2 Ectopic TOX protein is detected in CD4+ T lymphocytes in MF skin lesions, but absent in BID. ......................................................................................................................... 66 Figure 3.3 TOX mRNA levels are increased in peripheral blood CD4+ T cells from SS patients. ................................................................................................................................... 68 Figure 3.4 TOX protein levels are increased in SS CD4+ T cells and CTCL cell lines. ........ 70 Figure 3.5 TOX expression is elevated in multiple CTCL cell lines. ..................................... 71 Figure 3.6 Increased TOX mRNA levels differentiate CTCL from non-CTCL. .................... 73 Figure 3.7 Higher TOX mRNA levels correlate with worse clinical outcome in MF. ........... 75 Figure 3.8 Higher TOX mRNA levels correlate with increased disease-specific mortality in SS ............................................................................................................................................ 77 Figure 4.1 TOX inhibition confers growth disadvantage to CTCL cells. ............................... 84 Figure 4.2 TOX suppression decreases the clonogenic ability of CTCL cells in long term culture. .................................................................................................................................... 85 Figure 4.3 TOX inhibition impairs the tumor-forming ability of CTCL cells in vivo. ........... 87 Figure 4.4 TOX suppression leads to increased apoptosis and caspase activation in CTCL cells. ........................................................................................................................................ 90 Figure 4.5 TOX suppression leads to cell cycle arrest. ........................................................... 92  xii Figure 4.6 TOX suppression leads to elevated cell cycle repressors. ..................................... 94 Figure 4.7 Knockdown of CDKN1B or CDKN1C reverses the growth inhibition of TOX-sh cells. ........................................................................................................................................ 96 Figure 4.8 Identification of transcriptional targets of TOX in CTCL. .................................. 100 Figure 4.9 Top biological processes enriched in the genes differentially expressed upon TOX knockdown as identified by DAVID Functional Annotation Clustering analysis. ............... 103 Figure 4.10 TOX depletion restores SMAD3 expression. .................................................... 105 Figure 4.11 SMAD3 suppression in TOX-sh Hut78 cells fails to reverse the growth inhibition associated with TOX knockdown. ........................................................................................ 106 Figure 4.12 TOX mRNA levels are highly suppressed by TCR activation in normal CD4+ T cells, but not in CTCL cells. ................................................................................................. 108 Figure 5.1 Mode of TOX expression in normal T cells and in CTCL. ................................. 116 Figure 5.2 Proposed model of the role of TOX in CTCL. .................................................... 117  xiii List of Abbreviations Abbreviation Definition   7-AAD 7-aminoactinomycin D ACTB Beta-actin  AHI-1 Abelson helper integration site 1 AICD Activation-induced cell death AIRE Autoimmune regulator APC Antigen presenting cell ARID1A AT rich interactive domain 1A AUC Area under the curve BCL11A B-cell lymphoma/leukemia 11A BCL2 B-cell lymphoma 2 BCL7A B-cell lymphoma/leukemia 7A BID Benign inflammatory dermatoses BrdU 5-bromo-2'-deoxyuridine BTBD7 BTB (POZ) domain containing 7 CARD11 Caspase recruitment domain family, member 11 CCL Chemokine (C-C motif) ligand  CCR4 CC-chemokine receptor 4 CD3EAP CD3e molecule, epsilon associated protein CDK Cyclin-dependent kinase CDKN Cyclin-dependent kinase inhibitor c-FLIP CASP8 and FADD-like apoptosis regulator CGH Comparative genomic hybridization CGREF1 Cell growth regulator with EF-hand domain 1 ChIP-seq Chromatin immunoprecipitation assays with sequencing CI Confidence interval CLA Cutaneous lymphocyte antigen CRTC1   Creb regulated transcription coactivator 1 CTCL Cutaneous T-cell lymphoma  CTLA-4 Cytotoxic T lymphocyte antigen-4  CTR Control DAPI 4',6-Diamidino-2-phenylindole DC Dendritic cell DD Double dull DMEM Dulbecco's modified eagle medium DN Double negative DNM3 Dynamin 3 DP Double positive  xiv DPP4 Dipeptidyl-peptidase 4 E2F1 E2F transcription factor 1 EORTC European Organization for Research and Treatment of Cancer EPHA4 EPH receptor a4 ERCC2 Excision repair cross-complementation group 2  EZH2 Enhancer of zeste 2 polycomb repressive complex 2 subunit FACS Fluorescence-activated cell sorting FAM117A Family with sequence similarity 117, member A FasL Fas ligand FBS Fetal bovine serum FDA Food and Drug Administration FITC Fluorescein isothiocyanate FKBP4  FK506 binding protein 4 FOXO3 Forkhead box O3 FOXP1 Forkhead box P1 GAPDH Glyceraldehyde-3-phosphate dehydrogenase GATA3 GATA binding protein 3 GZMB  Granzyme B H&E Hematoxylin and eosin HBP1  High mobility group box transcription factor 1 HDACi Histone deacetylase inhibitor HMG High mobility group HP Healthy participant HR Hazard ratio HS Healthy skin IF Immunofluorescence IFN Interferon Ig Immunoglobin IHC Immunohistochemistry IL Interleukin ISCL International Society for Cutaneous Lymphomas ITM2B  Integral membrane protein 2B JAK Janus kinase kb Kilobase kDa Kilodalton KIR3DL2 Killer cell immunoglobulin-like receptor, three domains, long cytoplasmic tail, 2 KRAS Kirsten rat sarcoma viral oncogene homolog LDH Lactate dehydrogenase LTi Lymphoid tissue inducer MED13L Mediator complex subunit 13-like  xv MEF2C Myocyte enhancer factor 2C MF Mycosis fungoides MGAT4A Mannosyl (alpha-1,3-)-glycoprotein beta-1,4-N-acetylglucosaminyltransferase, isozyme A MHC Major histocompatibility miRNA microRNA MLL2 Mixed-lineage leukemia protein 2 MRPL12 Mitochondrial ribosomal protein L12 MYBL1 V-myb avian myeloblastosis viral oncogene homolog-like 1 Myc V-myc avian myelocytomatosis viral oncogene homolog NEDD4L Neural precursor cell expressed, developmentally down-regulated 4-like, E3 ubiquitin  NFAT Nuclear factor of activated T-cells NF-κB Nuclear factor of kappa light polypeptide gene enhancer in B-cells NHEJ Non-homologous end joining NK Natural killer NLS Nuclear localization sequence NOD Non-obese diabetic NRAS Neuroblastoma ras viral (v-ras) oncogene homolog NSG NOD/SCID interleukin-2 receptor gamma chain deficiency PBMC Peripheral blood mononuclear cell PBS Phosphate buffered saline PD-1 Programmed death-1 PE Phycoerythrin PEI Polyethylenimine PI Propidium iodide PIC Protease inhibitor cocktail PLCG1 Phospholipase C, gamma 1 PLS3 Plastin 3 PMA Phorbol 12-myristate 13-acetate PMSF Phenylmethylsulfonyl fluoride PSB Phosphorylation solubilization buffer PSMB5   Proteasome subunit beta 5 PTPRCAP Protein tyrosine phosphatase, receptor type, c-associated protein qPCR Quantitative polymerase chain reaction RAB3A RAB3A, member RAS oncogene family RNAi RNA interference RNF20 Ring finger protein 20, E3 ubiquitin protein ligase ROC Receiver operating characteristic RPMI Rosewell park memorial institute RPS6KA1 Ribosomal protein S6 kinase, 90kDa, polypeptide 1  xvi RT Room temperature RUNX3 Runt-related transcription factor 3 SATB1 Special AT-rich sequence binding protein 1 SCID Severe-combined inmmunodeficiency SDS Sodium dodecyl sulfate SEM Standard error of the mean SETD1A SET domain containing 1A shRNA Small hairpin RNA siRNA Small interfering RNA SLC26A11 Solute carrier family 26 (anion exchanger), member 11 SMAD Mothers against decapentaplegic homolog SP Single positive SS Sézary syndrome STAT Signal transducer and activator of transcription T-ALL T-cell acute lymphoblastic leukemia  TBST Tris-buffered saline Tween 20 TCR T cell receptor  Tg Transgenic TGFBR2 Transforming growth factor, beta receptor II TGF-β Transforming growth factor-β Th T helper THAP1  THAP domain containing, apoptosis associated protein 1 TIAM2 T-cell lymphoma invasion and metastasis 2 TNFR2 Tumor necrosis factor receptor 2 TNFRSF1B Tumor necrosis factor receptor superfamily, member 1B TNFSF11 Tumor necrosis factor ligand superfamily, member 11 TNFSF7 Tumor necrosis factor ligand superfamily member 7 TNMB Tumor-Node-Metastasis-Blood  TOX Thymocyte selection-associated high mobility group box TP53 Tumor protein p53 Treg T regulatory cells TWIST1 Twist family bHLH transcription factor 1 VSV-G Vesicular stomatitis virus glycoprotein WHO World Health Organization XPO5 Exportin 5 ZEB1 Zinc finger E-box binding homeobox 1   xvii Acknowledgements I would like to thank my supervisor Dr. Youwen Zhou, for giving me the opportunity to work in his laboratory and for the endless support throughout my graduate studies. He inspired me to think critically, act responsibly, and pursue my scientific interests in the face of challenges. My special gratitude goes to my co-supervisor, Dr. Xiaoyan Jiang, for generously accepting me into her lab family, where I gained scientific knowledge, expertise, and broad perspectives in cancer research. Her guidance, encouragement and scientific support over the years were essential for the completion of my thesis research. I would also like to express my sincere appreciation to my committee members, Drs. Jan Dutz, Gang Li, and Christian Steidl, for their constructive suggestions and continuous guidance throughout my graduate studies.   I would like to acknowledge our collaborators Drs. Ivan Litvinov, Denis Sassevile, Thomas Kupper, Yang Wang, and Ping Tu for their assistance in providing clinical samples and clinical databases for CTCL patients.  Many colleagues and friends in the Zhou lab and the Jiang lab have helped and inspired me along the way. I wish to thank Dr. Ming-Wan Su, Dr. Shengquan Zhang, Dr. Yang Wang, Helena Wang, Dr. Kevin Lin, Dr. Min Chen, Dr. Sharmin Esmailzadeh, Dr. Katharina Rothe for inspiring discussions and for always being willing to lend a helpful hand in time of need. I also wish to thank Dr. Yaohua Zhang, Richard Yu, Dr. Guohong Zhang, Dr. Yabin Cheng, Rayeheh Bahar, Xue Zhang, Laura Graziano, Josephine Leung, Leon Lin, Will Liu, Damian Liu, and many others for generously sharing ideas, expertise and reagents. I also wish to thank my colleagues, friends, and fellow students in the Terry Fox Laboratory, Dr. Wei Wei, Dr. Ping Xiang, Dr. Shu-Huei Tsai, Courteney Lai, Victoria Garside, Yuyin Yi, and many others for always being willing to lend a helping hand when I need it.   xviii My graduate studies would not have been smooth and successful without the support and assistance of Dr. Harvey Lui, Karen Ng, Leanne Li, Bay Gumboc and other administrative staff from the Department of Dermatology and Skin Science, and Dr. Vincent Duronio, Cornelia Reichelsdorfer, Virginia Grosman, and other staff from the Experimental Medicine Program. I am honored to be the recipient of several awards, including the following: Vanier Canada Graduate Scholarship, Canadian Institutes of Health Research-Skin Research Training Centre scholarship, UBC Faculty of Medicine Graduate Award, Canada Graduate Scholarships-Michael Smith Foreign Study Supplement, Four Year Fellowships (FYF) For PhD Students, a travel award to the 21st International Pigment Cell Conference, an abstract achievement award to the 56th American Society of Hematology annual meeting, a Rising Star Scholarship to the 23rd World Congress of Dermatology, and a UBC International Tuition Award.  This work was also generously supported by grants from Canadian Institutes of Health Research to Dr. Youwen Zhou and Dr. Xiaoyan Jiang, Canadian Dermatology Foundation to Dr. Youwen Zhou, and Canadian Melanoma Foundation to Dr. Xiaoyan Jiang.       xix               To my dearest parents and sister for their love and support.     1 Chapter 1: Introduction 1.1 T-cell Biology and Function The key function of the immune system is to protect against extrinsic infectious organisms and intrinsic abnormalities. The human immune system has two branches, innate immunity and adaptive immunity. Innate immunity serves as the first-line defense when pathogens are encountered. It initiates immediate response in a non-specific manner, but does not offer long-lasting immunity to the host (Alam & Gorska, 2003; Larosa & Orange, 2008). Natural killer (NK) cells are important players in the innate immune system. During evolution, vertebrates developed a more sophisticated immune response, the adaptive response, which responds to the offending pathogen in an antigen-specific way using receptors with enormous diversity, and which forms long-term immunological memory (Larosa & Orange, 2008). T cells and B cells are critical players in adaptive immunity, defending respectively against intracellular (through cytotoxic T cells) and extracellular pathogens (through antibodies) (Alam & Gorska, 2003).    Activation and proliferation of naïve T cells start with their interaction with antigen presenting cells (APCs), a process called priming. T cells can only recognize specific antigens processed and presented by the major histocompatibility (MHC) molecules on the surface of APCs. There are two classes of MHCs, MHC I and MHC II.  MHC class I is present in all nucleated cells, and MHC class II is expressed by professional APCs, such as dendritic cells (DCs), B cells, and monocyte/macrophages. CD4+ T cells bind to antigens presented by MHC class II molecules, while CD8+ T cells bind to antigens presented by MHC I molecules (Alam & Gorska, 2003; Larosa & Orange, 2008). In addition to T cell receptor (TCR)-MHC-antigen interactions, coreceptors (CD4 or CD8) and costimulatory  2 signals through CD28 (on T cells) binding to B7 (on APCs) are required for naïve T cell activation, leading to interleukin (IL)-2 production by the activated T cells and subsequent clonal proliferation. Without costimulation, the T cells become anergic, demonstrated by the inability to produce IL-2 or proliferate, even with subsequent constimulatory signals (Larosa & Orange, 2008). Once naïve T cells become primed, they further differentiate into effector T cell subtypes.  CD8+ T effector cells, upon priming, exert cytotoxicity by two major mechanisms. First, they secrete cytotoxic proteins perforin and granzyme B. Perforin forms pores in the cell membrane and allows granzyme B into the cytoplasm, which leads to both caspase-dependent and caspase-independent apoptosis (Alam & Gorska, 2003). Second, activated CD8+ T cells may bind to Fas receptor on the target cells through their Fas ligand (FasL), leading to rapid apoptosis of the target cells (Russell & Ley, 2002). Unlike CD8+ T cells, primed naïve CD4+ T cells further differentiate into several subsets of cells with distinctive cytokine production profiles and functions, with T helper (Th)1, Th2, Th17, and T regulatory cells (Treg) being the major subtypes (Weaver et al, 2006; Zhu & Paul, 2008). Th1 cells produce interferon (IFN)-γ and IL-2, important in activating macrophages to fight against intracellular pathogens and establishing immunological memory in CD4+ T cells (Paul & Seder, 1994; Zhu & Paul, 2008). Th2 cells secrete IL-4, IL-5, IL-10, IL-13, and mediate defense mechanism against extracellular parasites (Larosa & Orange, 2008). Specifically, IL-4 promotes B cell proliferation and antibody production, IL-5 facilitates eosinophil recruitment (Coffman et al, 1989), and IL-10 inhibits both dendritic cells and Th1 cells (Fiorentino et al, 1989; Moore et al, 2001). Th17 is a relatively newly identified subset of effector T cell subset that mediates defense against extracellular organisms as well as  3 autoimmunity, through secretion of IL-17A, IL-17F, IL-6, IL-21, and IL-22 (Bettelli et al, 2007; Weaver et al, 2006). Finally, Treg cells produce transforming growth factor-β (TGF-β), IL-10, IL-35, and function as negative regulators of the immune response and important mediators for peripheral tolerance (Luckheeram et al, 2012; Zhu & Paul, 2008).      Upon elimination of the offending pathogen, the immune response is attenuated to restore homeostasis through several mechanisms, including upregulation of inhibitory receptors, such as cytotoxic T lymphocyte antigen-4 (CTLA-4) and programmed death-1 (PD-1) (Parry et al, 2005; Walunas et al, 1994), and clonal deletion by the activation-induced cell death (AICD) mechanism (Green et al, 2003). Despite clearance of the majority of activated lymphocytes, some antigen-specific T cells manage to survive and become memory T cells that are capable of mounting quicker and stronger response against the same pathogen in the second attack. There are two types of memory T cells, central memory T cells and effector memory T cells, distinguished by their surface markers and homing properties (Kaech et al, 2002). Central memory T cells express CD45RO, CD62L, and CCR7, and home to the lymph nodes. Effector memory T cells express CD45RO, but lack CD62L and CCR7, and circulate to the peripheral tissue where they provide constant surveillance and mount a rapid effector response when re-challenged (Alam & Gorska, 2003). Taken together, each of the T cell subtypes carries out specialized functions and they work together to mount and terminate an effective adaptive immunity response. Dysfunctional T cells may lead to inadequate defense against invading pathogens, inability to eliminate transformed cells, or reactivity against self-antigens, leading to uncontrolled infections, malignancies, or autoimmunity, respectively. Therefore, a normal T-cell  4 development process is crucial in eliminating abnormal thymocytes and ensuring only properly functional T cells are allowed to mature and exit the thymus.  T cells play an essential role in defending exogenous pathogens and maintaining long-term memory in epithelial barrier tissues, including skin, lung, brain, gastrointestinal tract, and reproductive tracts (Mackay et al, 2013; Wakim et al, 2012). Previously, tissue-tropic T cells were thought to be primarily residing in the circulation and they were only recruited to the tissues with infections in times of need. Upon the clearance of intruding pathogens, the T cells recruited either underwent apoptosis or returned to the circulation, leaving very few T cells in the peripheral tissues where pathogens were encountered (Clark, 2015). However, this idea was challenged by a series of experiments in the last two decades. It was observed that antigen-specific CD8+ T cells remained long-term memory cells in the peripheral tissues after Listeria and vesicular stomatitis virus infection (Masopust et al, 2001). Skin grafts of non-lesional skin from patients with psoriasis, when transplanted onto immunodeficient mice, generated active psoriatic skin lesions, showing strong evidence of the presence of resident pathogenic T cells in the non-lesional skin of psoriatic patients (Boyman et al, 2004). Subsequent studies conducted in human lung, gastrointestinal system, reproductive system, and peritoneum demonstrated the existence of a plethora of antigen-specific resident memory T cells (TRM) (Booth et al, 2014; Okhrimenko et al, 2014; Purwar et al, 2011; Roberts et al, 2009; Turner et al, 2014).  This notion of long-lived tissue-tropic TRM cells was further supported by a recent observation by Clark et al that the skin surface of a healthy adult was populated by around 20 billion T cells, which was almost twice as much as those circulating in the peripheral blood. Under normal conditions, the vast majority (98%) of skin-tropic T cells resided in the skin,  5 and the remaining 2% located in the circulation (Clark et al, 2006). Furthermore, Watanabe et al. identified four populations of resident and recirculating T cells in the human skin, including: 1). CD103+ TRM , 2). CD103- TRM, 3). CCR7+/L-selectin+ central memory T cells (TCM), and 4) CCR7+/L-selectin- migratory memory T cells (TMM), each displaying distinct features in locations, cytokine production profiles and functional activities (Watanabe et al, 2015). Further characterization of these resident and recirculating T cells would shed light on the healthy immune response, as well as pathological immune responses seen in inflammatory and autoimmune conditions.  1.2 T-cell Development T-cell development takes place in the thymus. The thymus consists of two lobes, each with numerous lobules made of central medulla and surrounding cortex. The hematopoietic precursor cells enter the thymus at the cortico-medullary junction, and start their journey migrating towards the outer cortex first (Ciofani & Zuniga-Pflucker, 2007). These cells lack the expression of both CD4 and CD8 surface markers, and thus are called double negative (DN) cells. DN cells undergo four developmental stages (DN1-DN4) in the cortex (Godfrey et al, 1993), become double positive (DP) cells, and finally become single positive (SP) cells in the medulla before thymic exit to the circulation and peripheral tissues. T-cell development involves multiple dynamic steps whereby thymocyte progenitors migrate, mature and undergo stringent selections (Figure 1.1). There are several critical checkpoints during T-cell development. The earliest checkpoint the thymocytes face is β-selection, which starts during DN3 stage as the immature cells rearrange TCR-β chain genes and express preTCR on the surface. Recombination of TCR gene segments (α, β, γ, and δ) begins at the β, γ, and δ gene loci.  6 A minority of thymocytes complete successful TCRγδ gene rearrangement and commit to γδ T-cell lineage and leave the thymus with expression of CD4 or CD8 (Lauritsen et al, 2006). However, the majority of the thymocytes will express TCR-β chain, and subsequently pair with the surrogate primitive α receptor (pre-TCRα) to form the pre-TCR (Larosa & Orange, 2008). This process is crucial for further differentiation and expansion of thymocytes, known as β-selection. Without successful pairing with pre-Tα, the cells will die by neglect (Starr et al, 2003). The β-selected thymocytes further mature to the DN4 stage and start migrating inwards to the medulla. At this point, CD4 and CD8 are both upregulated, and the thymocytes obtain a DP phenotype, accompanied by the rearrangement of the α loci and the generation of mature TCRαβ. The expression of functional TCRαβ allows for interaction with peptide-bound MHC molecules on the surface of thymic epithelial cells, which is needed for the second checkpoint, positive selection. DP cells that interact with peptide-MHC complexes with intermediate affinity are selected to live and undergo further differentiation, while those that do not are eliminated by apoptosis (Klein et al, 2009; Larosa & Orange, 2008). The surviving DP cells, upon TCR engagement, temporarily downregualate both CD4 and CD8 expression, and become a transitional subset of cells called double dull (DD) thymocytes (CD4loCD8lo) (Adlam & Siu, 2003). A second transitional stage follows, characterized by the CD4 molecule being re-expressed, and the cells become CD4+CD8lo. These cells then commit to either CD4 SP or CD8 SP lineage, migrate to the medulla, where they are tested by the final checkpoint in T-cell development, negative selection. SP cells that react too strongly to the self-antigens presented by MHC molecules pose a significant risk of autoimmunity, and are eliminated by apoptosis (Lauritsen et al, 2006). The autoimmune regulator (AIRE) gene, a transcription factor that is  7 responsible for expressing tissue-specific antigens in the thymic epithelial cells, is critical in establishing central tolerance and preventing the T cells from attacking the “self” tissues (Anderson et al, 2002).  Collectively, the stringent selection checkpoints during the development of thymocytes safeguard the quality of T cell production, thus only those T cells that are functionally capable of fighting against foreign intruders or intrinsic abnormal cells will mature, and the autoreactive T cells are effectively eliminated to prevent autoimmunity.  8  ThymusDN1 DN2 DN3 DN4 DP DDCD4+CD8loCD4+CD8+BloodCD4+CD8+β-selectionPositiveselectionNegativeselection Figure 1.1 Overview of T-cell development in the thymus. The progenitor DN cells enter the thymus and undergo four developmental stages, DN1-DN4. TCR gene rearrangement starts during DN2 stage and completes at DN3, when the cells undergo β-selection. These cells continue through DN4 and proceed to the DP stage, then undergo TCR-α rearrangement and fully express functional TCR. DP cells that interact with MHC molecules with sufficient affinity will be selected (positive selection). After two transitional stages, DD and CD4+CD8lo, the cells further commit to either CD4+ or CD8+ T cell lineages. The single positive CD4+ or CD8+ T cells then undergo negative selection, a process that eliminates those cells that react too strongly with self-peptides presented by MHC class II (recognized by CD4+ T cells) or MHC class I (recognized by CD8+ T cells). Finally, the thymocytes that successfully undergo all the selection process fully mature and exit the thymus.     9  1.3 Cutaneous T-cell Lymphoma (CTCL)  CTCL is a group of lymphoproliferative disorders characterized by the homing and accumulation of malignant T lymphocytes in the skin. CTCL is classified as cutaneous T cell and NK/T cell lymphomas in the 2005 WHO-EORTC (World Health Organization-European Organization for Research and Treatment of Cancer) classification (Willemze et al, 2005), and as mature T-cell and NK-cell neoplasms in the 2008 WHO classification of tumors of hematopoietic and lymphoid tissues (Campo et al, 2011). According to a recent study in the US, the age-adjusted incidence of CTCL is 6.4 cases per million and continues to increase by 2.9 cases per million per decade during 1973-2002, making it the second most common extranodal, non-Hodgkin lymphoma (Criscione & Weinstock, 2007). Currently, it is estimated that there are about 3000 new CTCL cases in the US per year, and a rough estimation of 25000 to 50000 people in the US are afflicted with CTCL (Wong et al, 2011).  CTCL primarily affects the elderly (median age of onset 55-60 years) and the incidence increases rapidly with age, although it can occasionally be seen in children and adolescents (Kim et al, 2003a; van Doorn et al, 2000; Wain et al, 2003; Zackheim et al, 1999). CTCL tends to affect males more than females (male-to-female ratio ~ 2:1), and 50% higher incidence is observed in black people, compared to white people (Criscione & Weinstock, 2007; Kim et al, 2003a).  The etiology of CTCL is unknown, but has been linked to chronic antigenic stimulation (Girardi et al, 2004). Other etiologic factors include bacterial (staphylococcus aureus) and virus (Epstein-Barr virus and cytomegalovirus) infection (Herne et al, 2003; Talpur et al, 2008), immunosuppression (Ravat et al, 2006), and occupational exposure  10 (glass, pottery, ceramics) (Morales-Suarez-Varela et al, 2004), but there is minimal evidence supporting direct causative effects.  CTCL is a highly heterogeneous disease presenting with diverse clinical manifestations, histopathological features, and disease courses. While many entities are included in the scope of CTCL, the two most common clinical variants of CTCL are mycosis fungoides (MF) and Sézary syndrome (SS), together accounting for 65-75% of total CTCL cases (Bradford et al, 2009; Willemze et al, 2005). These two disease entities are closely related while displaying distinct clinical as well as molecular characteristics. A recent study suggested that MF and SS arise from different subsets of memory T cells. MF cells are deficient in both L-selectin and CCR7 expression, a feature of TRM. In contrast, SS cells resemble TCM cells in that they express both L-selectin and CCR7, which is consistent with their presence in both skin and circulation (Campbell et al, 2010).    Currently, the staging of CTCL patients relies on the Tumor-Node-Metastasis-Blood (TNMB) system, first proposed in 1978 at a National Cancer Institute Workshop (Lamberg & Bunn, 1979), and revised in 2007 by the International Society for Cutaneous Lymphomas (ISCL)/EORTC. The staging system incorporates the extent of skin lesion (T1-T4), involvement of lymph nodes (N0-N3), visceral metastasis (M0-M1), and presence of Sézary cells in the peripheral blood (B0-B2). Clinical stages are determined on these 4 components (Olsen et al, 2007).  1.3.1 Mycosis Fungoides (MF) MF is by far the most common type of CTCL, accounting for 50-72% of CTCL (Criscione & Weinstock, 2007; Willemze et al, 2005). MF has various clinical presentations (Figure 1.2), including patches, infiltrating plaques, tumors and erythroderma. Most MF patients (~70%)  11 present with early stage (stage IA to IIA, patch and plaque) lesions, classically distributed in sun-protected areas such as the buttocks, trunk, and proximal extremities (Talpur et al, 2012). In 10-20% of the patients, the patches may evolve to plaques and tumors (Agar et al, 2010). However, more infiltrating tumors and erythroderma are the presenting lesions in about one third of MF patients, without previous patches or plaques (Kim et al, 2003a).  Morphologically, the skin-infiltrating malignant MF cells are small- to medium- sized lymphocytes with hyperconvoluted cerebriform nuclei (Willemze et al, 2005). Immunophenotypically, these cells are predominantly CD4+CD45RO+ mature memory T cells, although CD8+ CTCL cases have been infrequently reported, mostly in children (Crowley et al, 1998). Both types of cells also express cutaneous lymphocyte antigen (CLA) and CC-chemokine receptor 4 (CCR4) (Kim et al, 2005). Ancillary findings in CTCL include loss of certain T-cell surface markers, such as CD2, CD3, CD5, and CD7 (Willemze et al, 2005).   In addition to compatible clinical manifestations, the diagnosis of MF depends on histological confirmation on a lesional biopsy. Classic MF displays two characteristics: (1). Epidermotropism, defined by the band-like infiltration of atypical lymphocytes into the epidermis without epidermal edema; (2). Pautrier’s microabcess, characterized by several atypical lymphocytes clustering around Langerhans cells, the skin antigen presenting cells. Although Pautrier’s microabcess is the hallmark of CTCL, it is only present in 25% of cases (Willemze et al, 2005). In addition, the lesions have much elevated CD4+ to CD8+ T lymphocyte ratios (>4:1) (Kim et al, 2005).  It is often difficult to accurately diagnose MF, especially in its early stages, which resemble many benign inflammatory dermatoses in clinical, histopathological, and molecular  12 features (Kim et al, 2005). Currently, the diagnosis of MF relies on a combination of clinical-pathological parameters in four categories (ISCL criteria, 2005) (Pimpinelli et al, 2005): (1) clinical presentations; (2) histopathologic features; (3) presence of clonal TCR gene rearrangement; (4) immunopathologic features. The diagnosis of MF requires 4 points of the above criteria. Further modifications to this system have been proposed by Ferrara et al. (Ferrara et al, 2008). While this diagnostic system offers an improved standardization, it may take a long period of time for clinicians to fully evaluate its validity (Zhang et al, 2012). It is noteworthy that the features included in the ISCL criteria are not universally or exclusively present in MF. For example, the loss of surface T-cell antigens (CD3, CD5, CD7, and CD26) has been observed in CTCL (Robson, 2010). However, reduced expression of CD5, CD7, and CD26 may be seen in reactive inflammatory T cells in benign skin conditions (Murphy et al, 2002). In addition, the confirmatory diagnostic feature for MF, TCR clonality, can be seen in patients with benign inflammatory dermatoses (BID), such as lichen planus and pityriasis lichenoides chronica (Dereure et al, 2000a; Lukowsky et al, 2000; Schiller et al, 2000). Finally, the markers proposed in the ISCL immunopathologic criteria are all negative markers: MF is considered more likely with the absence of cellular markers, such as CD2, CD3, CD5, and CD7. There has been no positive marker in clinical use for MF.  MF is typically an indolent disease, which tend to progress slowly over many years or even decades (Jawed et al, 2014a). Patients with limited (<10% total body area) patch/plaques have a life expectancy similar to that of the age-, race- and sex-adjusted general population. However, disease-specific survival rates drop as stages advance. For example, 5-year disease-specific survival rates were around 100% for patients with stage-IA  13 MF (<10% skin involvement), about 50% for patients with tumor(s), and as low as around 20% for patients with visceral involvement (Agar et al, 2010).  Currently, the prognosis of MF patients depends primarily on the clinical stage. It has been reported that prognostic factors for increased mortality include old age, elevated lactate dehydrogenase (LDH), large cell transformation, and folliculotropic MF, although most of these studies were of relatively small sample sizes and applied only univariate models (Agar et al, 2010; Diamandidou et al, 1998; Diamandidou et al, 1999; Kim et al, 2003a; Talpur et al, 2012). At present there is no molecular marker with prognostic value being used clinically. Having a highly specific and sensitive molecular marker for MF would greatly facilitate patient stratification according to risk of disease progression and mortality, and would allow appropriate treatment regimens to be administered to improve patient outcomes.  1.3.2 Sézary Syndrome (SS) SS is a leukemia variant of CTCL, which accounts for 2.5% of CTCL cases (Criscione & Weinstock, 2007). SS often develops abruptly without precedent lesions. The classic clinical triad of SS consists of lymphoadenopathy, generalized erythroderma, and the presence of Sézary cells in the circulation. Clinically, SS may present as mild erythema or widespread erythroderma (Figure 1.2D), often accompanied by severe pruritus. Extensive exfoliation, edema, hyperkeratosis in the palms and soles, alopecia, and nail dystrophy are frequently seen (Wieselthier & Koh, 1990).  The histological features of SS may be similar to those of MF, but could be non-specific in up to one-third of the patients. Also, more monotonous lymphocyte infiltration and less or absent epidermotropism is often seen in SS biopsies (Trotter et al, 1997). Immunophenotyping shows malignant CD3+CD4+CD8- cells in the skin, the affected lymph  14 nodes, and the peripheral blood. Loss of CD7 and CD26 is frequently seen in circulating Sézary cells (Vonderheid et al, 2002).           The diagnosis of SS is not straightforward. Exfoliative erythroderma with pruritus can be attributed to aging in the elderly, and many benign skin conditions, such as psoriasis, drug eruptions, generalized seborrheic dermatitis etc., may have similar skin appearances with SS (Jawed et al, 2014a). The histological diagnostic feature for Sézary cells is the presence of cerebriform nucleus. However, this morphological feature can be difficult to differentiate from the enlarged, and often multi-lobular, nuclear structure of T lymphocytes after activation. Furthermore, while the absence of CD7 and CD26 is supportive of a diagnosis of SS, similar immunophenotypic features may be seen in BID (Harmon et al, 1996; Jones et al, 2001; Ortonne et al, 2006). Finally, the TCR clonality diagnostic criteria proposed by the ISCL may not only be seen in SS, but also in patients with systemic sclerosis and healthy people (Marie et al, 2005; Vonderheid et al, 2002).  Collectively, the overlapping features in the clinical, histopathological, and molecular findings between SS and BID poses a significant challenge in SS diagnosis. Novel markers with good sensitivity and specificity for the identification of Sézary cells would enhance the diagnostic accuracy of SS.  Unlike most MF cases, SS has an aggressive clinical course, with only 24% 5-year survival rate and a mean survival time of 2-4 years (Scarisbrick et al, 2001; Willemze et al, 2005), despite the use of multiple treatment modalities. Most SS patients die of opportunistic infections due to severe immunosuppression. More effective treatments with minimal toxicities are needed.   15  A B C D  Figure 1.2 Clinical manifestations of mycosis fungoides and Sézary syndrome (A) Patch-stage mycosis fungoies. (B) Plaque-stage mycosis fungoies. (C) Tumor-stage mycosis fungoies. Ulceration is frequently present. (D) Erythroderma in a patient with Sézary syndrome. Over 80% of his body surface area is affected. Photos courtesy of Dr. Youwen Zhou.    16  1.3.3 Disease Pathogenesis  The development of CTCL is a multistep process involving deregulated cytokine production, host immune response, and skin-homing of malignant T cells.  In early CTCL lesions there are sparse malignant CD4+ T cells in the skin and the host antitumor immune response is strong. There is an abundance of cytotoxic CD8+ T cells, which share a Th1 cytokine profile (IL-2, IL-12, IFN-γ) and play an important role in cell-mediated immunity (Saed et al, 1994). As the disease progresses, the infiltrating malignant CD4+ T cells continue to accumulate, and the Th2 cytokines they produce (IL-4, IL-5, IL-10, IL-13) gradually predominate in the skin, causing a Th1 to Th2 profile switch in more advanced stages (Wong et al, 2011). The cytokine pattern change is also evidenced by gene expression analysis in the peripheral blood mononuclear cell (PBMC) in various stages of MF (Chong et al, 2008). The Th2-specific transcription factors, such as GATA3 and JunB, are over-expressed in SS (Kari et al, 2003).  The increase in Th2 cytokine production is accompanied by a profound decrease in DC and host antitumor response. The increase in tumor burden of malignant T cells in the circulation has been associated with a decrease in both myeloid DCs and plasmatoid DCs, and a decline in the cytokines they produce (IL-12 and IFN-α) (Wysocka et al, 2002). A decline in myeloid DCs could also result in reduced IL-15 and IL-18, both important stimulating agents for IFN-γ and Th1 responses. A marked reduction in cellular immunity has been observed during disease progression from early to more advanced CTCL. Both the number and activity of CD56+ NK cells declines during disease progression. Similarly, circulating tumor burden inversely correlates with the number of CD8+ T cells in the peripheral blood (Yoo et al, 2001). In addition, certain immunosuppressive molecules are  17 overexpressed in CTCL, including PD-1, FasL, and CTLA-4 (Ni et al, 2001; Wong et al, 2006; Zhang et al, 2012), highlighting the potential of harnessing these molecules for therapeutic purpose. Furthermore, reduced complexity of TCR repertoire occurs across all stages of CTCL, especially in later stages (Yawalkar et al, 2003). Together, these findings highlight the profound immunosuppression which occurs in advanced CTCL patients, who are prone to opportunistic infections and developing secondary skin cancers (Axelrod et al, 1992; Evans et al, 2004).    Other immunological abnormalities which occurs during disease progression include the development of eosinophilia and increased serum Immunoglobin (Ig)E levels (Molin et al, 1978; Vowels et al, 1992), which may be attributed to the elevated IL5 levels and might contribute to the intense pruritus experienced by CTCL patients (Suchin et al, 2001). Other cytokines involved in the immunopathogenesis of CTCL have recently been identified, including IL-17A/F, IL-21, IL-22, produced by Th17 cells, a newly discovered CD4+ T cell subset.  IL-17A, IL-17F, and IL-17 receptor were up-regulated in MF skin lesions, and IL-17A was increased in the peripheral blood of SS patient. The elevation of IL-17 was under the regulation of the Janus kinase/Signal transducer and activator of transcription 3 (JAK/STAT3) pathway. Importantly, high IL-17F expression levels correlated with increased risk of disease progression (Krejsgaard et al, 2013).  What initiates the skin-directed homing of the malignant T cells remains elusive, but it could be partially due to their expression of certain cytokine receptors. Of particular importance in mediating the migration of malignant T cells are CLA, CCR4, and CCR10 (Kim et al, 2005). Endothelial cells express E-selectin and chemokine (C-C motif) ligand (CCL) 17, ligands for CLA and CCR4 respectively, allowing malignant T cells to migrate  18 through the blood vessels. CCL17, which is also produced by activated keratinocytes and dendritic cells, and CCL27 (ligand for CCR10), which is produced by keratinocytes, further attract the malignant T cells to migrate into the dermis and epidermis (Hwang et al, 2008). In advanced stages, such as tumor stage MF, the malignant cells lose the skin-homing CCR4 and gain expression of CCR7, a critical receptor for lymphatic migration, leading to less epidermotropism and more systemic involvement (Kallinich et al, 2003).  1.3.4 Chromosomal, Genetic and Molecular Findings of CTCL Unlike myeloid leukemias and B-cell lymphomas, CTCL has not been linked with a distinctive hallmark chromosomal dislocation. A few genetic studies were conducted in MF and SS in search for recurrent chromosomal alterations in CTCL, and revealed genomic gain or loss in certain chromosomes (chromosomes 1, 6, 7, 8, 9, 10, and 17) (Izykowska & Przybylski, 2011; Mohr et al, 1996). MF and SS showed different patterns of chromosomal abnormalities. Using comparative genomic hybridization (CGH) technique, MF showed gain of 1q, 7p or 7q and loss of 9p21, while SS showed gain of 8q and17q, and loss of 10p and 17p (Laharanne et al, 2010). The genes involved in these genomic regions have not been characterized. Gain of certain oncogenes was identified, such as Myc and STAT3 (Vermeer et al, 2008). However, these studies have not yielded consistent findings, and no abnormal gene hotspot has been uncovered. Instead, the involvement of many chromosomal regions highlights the underlying genomic instability in CTCL (Wong et al, 2011).  Abnormal epigenetic changes have been shown in many cancers, including CTCL. Epigenetic CGH in CTCL revealed methylation in several tumor suppressor genes, such as TP73, cyclin-dependent kinase inhibitor (CDKN)2A (p16), and CDKN2B (p15) (van Doorn et al, 2005). In support of this notion, mutations involving the epigenetic regulation  19 machinery, including MLL2, SETD1A, and RNF20, have recently been identified (McGirt et al, 2015).  Mutation profiling in CTCL has provided insights into abnormal pathways leading to malignant transformation. By profiling 396 somatic mutations in 33 genes in 90 CTCL samples, Kiessling et al. identified oncogenic mutations in KRAS and NRAS in 4 samples, hinting at the RAS pathway being disrupted in CTCL development (Kiessling et al, 2011). However, the mutations seemed to be a rare event. Recent development of high throughput sequencing has allowed for more detailed and thorough assessment of the mutation landscape in cancer cells. Whole exome sequencing in 11 CTCL samples has identified recurrent point mutations in TNFRSF1B in 18% of samples. TNFRSF1B encodes the tumor necrosis factor receptor, TNFR2, and the expression of a specific mutation TNFR2Thr337Ile led to activation of non-canonical nuclear factor of kappa light polypeptide gene enhancer in B-cells (NF-κB) pathway. In addition, the same group identified a recurrent CTLA4-CD28 fusion, and Jurkat cells (an acute T-cell leukemia cell line) expressing this fusion product proliferated faster than the control cells, supporting the oncogenic role of this novel fusion protein (Ungewickell et al, 2015). Genomic profiling in 37 SS patients uncovered profound disruptions in cell cycle control and T-cell development pathways, with frequent mutations found in TP53, CARD11, CCR4, PLCG1, CDKN2A, ARID1A, RPS6KA1 and ZEB1 (Wang et al, 2015). Whole exome sequencing in 42 CTCL (SS = 25, other CTCL = 17) discovered a list of mutated genes involved in epigenetic regulation and TCR stimulation (da Silva Almeida et al, 2015). Kiel et al found recurrent mutations in epigenetic modifiers and JAK/STAT genes, as well as numerous novel fusion genes in SS, including MYBL1-TOX and EZH2-FOXP1 (Kiel et al, 2015). Another recent study applied whole genome  20 sequencing in 5 MF tumors and normal skin revealed activating mutations in JAK3, which is part of the IL2 signaling pathway. Taken together, these genomic studies revealed aberrant signaling pathways and provided promising therapeutic targets for CTCL.  Dysregulated cell cycle control has been implicated in CTCL disease development. Decreased expression of several important cell cycle regulators, including p14, p15, p16, and RB1, have been found in CTCL (Mao et al, 2006; Navas et al, 2000; Scarisbrick et al, 2002). Further, multiple cell cycle regulators, such as CDKN2A and CDKN1B, have been found deleted or mutated in CTCL samples by recent sequencing studies (da Silva Almeida et al, 2015; Wang et al, 2015). Apoptosis resistance is another well-characterized feature in CTCL cells. Fas, one member of the death receptors, mediates apoptosis, and Fas reduction or defects have been associated with apoptosis resistance and disease aggressiveness (Contassot et al, 2008; Dereure et al, 2002; Dereure et al, 2000b; Wu et al, 2009). Potential mechanisms for Fas defects include gene mutations (Dereure et al, 2002), a non-functional splice variant (van Doorn et al, 2002), and promoter hypermethylation (Jones et al, 2010). However, only a small number of cases have been studied. Another contributing factor for apoptosis resistance was overexpression of c-FLIP, an inhibitor of death-receptor mediated apoptosis (Contassot et al, 2008). Recently, the loss of a tumor suppressor gene, BIN1, has been shown to lead to c-FLIP upregulation in CTCL (Esmailzadeh et al, 2015), uncovering new molecular mechanism of the apoptosis resistance in CTCL. Little is known about other pathways involved in apoptosis regulation. The exact molecular mechanisms leading to the cell cycle control and apoptosis defects in CTCL are not completely understood.     21 1.3.5 Biomarkers of CTCL Biomarkers can aid accurate diagnosis and guide appropriate management for the patients. In addition, specific biomarkers can provide valuable insights into disease pathogenesis, and pinpoint potential therapeutic targets. In the last decade, much effort has been devoted to the identification of useful biomarkers for CTCL. Due to the disease rarity, most studies were performed on a small number of cases, and progress in biomarker identification has been slow.  Transcriptome analysis is a powerful high throughput method that assesses gene expression profiles and allows for efficient identification of differentially expressed genes in CTCL samples compared to control cells. Among the transcriptional signatures identified by different groups, little overlap was seen, possibly due to the inherent heterogeneity of CTCL samples, the fact that various platforms were applied, differences in the types of samples used (skin biopsies versus peripheral blood), and differences in the control samples selected (Dulmage & Geskin, 2013; Wong et al, 2011).     Despite these limitations, a list of 20 gene changes has been consistently identified by more than two gene expression profiling studies. These include PLS3, DNM3, NEDD4L, KIR3DL2, TWIST1, CD52, PTPRCAP, JUNB, TNFSF7, TNFSF11, TOX, VCAN, EPHA4, BCL11A, SATB1, CD26, STAT3, STAT4, BCL2, and TGFBR2 (Dulmage & Geskin, 2013). Among these changes, some are shared by MF and SS, such as TWIST1, CD52, PTPRCAP, JUNB, and TOX, while others are present in only MF or SS (Dulmage & Geskin, 2013). Some genes showed differential expression patterns in MF, such as EPHA4 and STAT4, suggesting their involvement in disease pathogenesis is stage-dependent, but the factors affecting their expression change remain to be characterized.   22 To highlight the core pathways disrupted in CTCL, Dulmage et al (Dulmage & Geskin, 2013) applied Ingenuity Pathway Analysis to 12 recent gene expression studies in CTCL, which revealed pathway enrichment in the following four areas: (1). AICD, evidenced by decreased Fas, FasL, and SATB1 (positive regulator of FasL) (Wang et al, 2011; Wu et al, 2009; Zoi-Toli et al, 2000). In addition, negative apoptosis regulators were found to be elevated, such as JUNB (negative regulator of FasL) and NF-κB (Lotem & Sachs, 1998; Sors et al, 2006); (2). Immune activation and Th2 predominance, supported by activation of Th2-specific transcription factors, JUNB and GATA3 (Gibson et al, 2013; Hartenstein et al, 2002), and loss of Th1-specific transcription factor, STAT4 (Nishikomori et al, 2002; van Doorn et al, 2004). JUNB also controls NFAT, which mediates T-cell activation and immune response (Bueno et al, 2002). (3). TGF-β signaling, highlighted by the loss of TGF-β receptor II protein, encoded by TGFBR2, a tumor suppressor gene with antiproliferative and proapoptotic activity, as well as loss of its downstream mediators, SMAD3 and SMAD7 (van Doorn et al, 2004); and (4). TNF signaling pathway, evidenced by increased expression of TNFSF7/CD70, TNFSF11/RANKL, and TNFSF5/CD40L, which are involved in T-cell proliferation, immune activation and T-cell migration (Storz et al, 2001; van Doorn et al, 2004). These core pathways are closely related and may collectively contribute to malignant cell transformation.     Aberrant expression of microRNAs (miRNAs) has been implicated in many cancers (Hayes et al, 2014). Similarly, a number of differentially expressed miRNAs have been uncovered in CTCL by microarray analysis. Upregulation of miR-21 was identified in SS, and it is regulated by STAT3 through promoter binding (van der Fits et al, 2011). MiR-155 elevation was found in CTCL (van Kester et al, 2011), which was driven by STAT5 that was  23 aberrantly turned on in the tumor cells (Kopp et al, 2013). Another overexpressed miRNA, miR-122, reduced sensitivities of CTCL cells to chemotherapy (Manfe et al, 2012). Downregulation of miR-342, miR-17-5q was reported in SS, the correction of which sensitized CTCL cells to apoptosis (Ballabio et al, 2010). MiR-223, which regulates several oncogenes (E2F1, MEF2C, and TOX), was decreased in MF (McGirt et al, 2014). Interestingly, a combination of 5 miRNAs (miR-326, miR-663b, miR-711, miR-203, miR-205) could distinguish CTCL from benign skin conditions with >90% accuracy, and could have diagnostic value (Ralfkiaer et al, 2011). These studies highlight the importance of miRNAs in CTCL pathogenesis and offer potential therapeutic opportunities.  Interestingly, certain physical properties of the malignant T cells can also serve as an identification marker for CTCL. Clark et al reported that a unique population of T cells with increased forward and side scatter, the high-scatter T cells, could be used as a biomarker for CTCL and an indicator for response to treatment (Clark et al, 2011).  Molecular prognostic biomarkers are less studied than diagnostic markers, partially due to limited CTCL cohorts with long-term clinical follow up data. Several molecular markers have been described to be of potential prognostic value. For MF prognosis, persistence of the same CD4+ T cell clone over time in skin biopsies correlated with an aggressive disease course (Vega et al, 2002). Loss of BCL7A expression predicted aggressive disease course in patients with early stage CTCL (Litvinov et al, 2013). Downregulation of CDKN1C was associated with worse prognosis in CTCL in a set of 60 patients (Litvinov et al, 2012). Specifically for tumor stage MF, presence of chromosomal alterations on 9p21, 8q24, 10q26qter and 1q21-1q22 often indicates a poor prognosis (Salgado et al, 2010; van Doorn et al, 2009). Molecular prognostic markers for SS are largely  24 lacking. For SS, the presence of the same peripheral T cell clone indicated poor prognosis (Fraser-Andrews et al, 2000). A list of 14 microRNAs, including miR-21, may predict patient survival in SS (Narducci et al, 2011). Using a combination of 10 genes, Kari et al. can predict SS patient survival regardless of tumor burden (Kari et al, 2003). The clinical utility of these markers needs further evaluation in larger studies, and to date there is no molecular prognostic marker used in the clinical setting.  1.3.6 Treatment of CTCL CTCL is rarely curable, and the purpose of treatment is to relieve symptoms, prevent progression, and improve the quality of life. Despite the plethora of treatments being introduced and improvement of care, it remains very difficult to treat CTCL in the advanced stages, and the prognosis is poor.  In early stage MF (IA-IIA), topical treatments are the mainstay modalities, including topical corticosteroids, nitrogen mustard, and retinoids. Topical corticosteroids are the most frequently used treatment for early MF, and it is a common adjunct modality in more advanced stages. However, response was not sustainable after drug discontinuation (Zackheim et al, 1998). Nitrogen mustard is an alkylating agent, and when applied topically, it works by affecting T cell interactions with keratinocytes and Langerhans cells (Kim et al, 2003b). Topical nitrogen mustard is associated with common cutaneous side effects (burning, allergy, irritation) and increased risk of developing nonmelanoma skin cancers (de Quatrebarbes et al, 2005; Lindahl et al, 2013). Retinoids are immunomodulating agents affecting cell differentiation and inducing apoptosis (Querfeld et al, 2006). Topical retinoids commonly used include bexarotene and tazarotene. Other modalities commonly used in early stage MF include phototherapy, in the form of either 8-methyoxypsoralen with UVA  25 irradiation (PUVA) or UVB irradiation, and total skin electron beam therapy (TSEBT) (Jawed et al, 2014b). For refractory early stage MF, a combination of phototherapy and either low-dose IFN-α or bexarotene may be used (Jawed et al, 2014b).     In advanced MF and SS, treatment is more challenging, and generally requires a combination of modalities. A combination of two of three modalities (IFN-α, retinoids, and phototherapy) may be applied (Jawed et al, 2014b). Extracorporeal photopheresis (ECP) is primarily useful in erythrodermic CTCL, but complete response rates were merely around 20% (Dani & Knobler, 2009; Duvic et al, 2003; Duvic et al, 1996; Edelson, 1999; Knobler et al, 2012). Histone deacetylase inhibitors (HDACis), include vorinostat and romidepsin, were approved by the US Food and Drug Administration (FDA) for refractory advanced stage CTCL (Jawed et al, 2014b). Systematic chemotherapy is usually reserved for patients with rapidly progressing advanced CTCL. Single-agent chemotherapies include antifolates (methotrexate and pralatraxate), gemcitabine, and pegylated doxorubicin (Dummer et al, 2012; Duvic et al, 2006; Izbicka et al, 2009). Multi-agent regimens include cyclophosphamide, doxorubicin, vincristine, and prednisone (Olsen et al, 2011). Systematic toxicity is the main issue with chemotherapy. Finally, stem cell transplantation may be curative in refractory advanced CTCL cases, although no large case studies are available (Duarte et al, 2008; Duvic et al, 2010).         In the search for more effective therapies for CTCL, some experimental therapies have been explored, such as Toll-like receptor agonists (Kim et al, 2010), proteasome inhibitors (eg. bortezomib) (Zinzani et al, 2007), and CCR4 blocking antibodies (Chang et al, 2012; Yano et al, 2007), but the therapeutic benefits and long-term side effects of these agents are still under investigation. Since CTLA-4 and PD-1 are upregulated in CTCL (Wong  26 et al, 2006; Zhang et al, 2012), they also represent promising therapeutic targets. Recently, immune-checkpoint blocking therapies targeting CTLA-4 and/or PD-1 pathways using antibodies have emerged as a promising antitumor approach, with durable response observed in both solid and hematological malignancies, such as metastatic melanoma, colorectal cancer, lymphoma, and leukemia (Ansell et al, 2015; Berger et al, 2008; Brahmer et al, 2010; Larkin et al, 2015). Ipilimumab (anti-CTLA-4), the first checkpoint inhibitor, was approved by the US FDA in 2010 for advanced melanoma. More recently, pembrolizumab and nivolumab, two anti-PD-1 antibodies, were approved by US FDA for advanced melanoma in 2014 (Ascierto & Marincola, 2015). Studies are warranted to evaluate the clinical benefits and safety of immune-checkpoint blocking therapies in CTCL, especially for advanced stage and treatment-refractory patients. To date, the treatment for advanced stage CTCL is largely unsatisfactory, and novel effective therapies with minimal toxicity are needed.  1.4 TOX, a Critical T-cell Regulator A member of the high mobility group (HMG)-box family of proteins, thymocyte associated HMG box protein (TOX) was first discovered in 2002 as a murine homologue of the protein encoded by the human gene KIAA0808, initially isolated from human brain tissue (Nagase et al, 1998). The mouse and human homologues shared about 90% nucleotide and amino acid sequences identity (Wilkinson et al, 2002). HMG-box proteins are important gene expression regulators which act by modifying chromatin structures (Bustin, 1999). The DNA-binding motif HMG box was first identified in non-histone chromosomal protein HMGB1, and consists of 70-80 amino acids (Baxevanis & Landsman, 1995). Upon binding to the minor groove of the DNA helix, HMG-box mediates DNA structural changes and potentially allows for the binding of other regulatory factors (Bewley et al, 1998). Interestingly, the HMG box  27 may also mediate protein-protein interactions (Aidinis et al, 1999), further contributing to its functional diversity in regulating gene expression. TOX plays multiple roles in the immune system, most notably in regulating T-cell development (Aliahmad et al, 2012).  1.4.1 TOX Structure and Expression The human TOX gene is located on chromosome 8 (8q12.1) and consists of 9 exons spanning over 300 kilobase (kb) (O'Flaherty & Kaye, 2003). So far a single human TOX transcript of 4131 bases has been identified, according to the ensembl database (ENSG00000198846). The TOX protein is 526 amino acids in length, with a molecular weight of 57 kDa. In addition, a 63 kDa form has also been detected, which is thought to be due to post-translational modification (Wilkinson et al, 2002). Indeed, the TOX protein contains several predicted phosphorylation sites (Artegiani et al, 2015).  The TOX protein contains four important domains (Figure 1.3), including: 1). the centrally located HMG box domain (for DNA-binding), 2). an N-terminal domain with transactivation activity, 3). a lysine-rich potential nuclear localization sequence (NLS) immediately upstream of the HMG box, and 4). a C-terminal domain rich in proline and glutamine, which may be involved in protein-protein interactions (Aliahmad et al, 2012; O'Flaherty & Kaye, 2003). As expected, TOX is primarily located in the nucleus, as demonstrated by immunofluorescence staining in both T-cell lines and the infiltrating T cells seen in human skin biopsies (Aliahmad & Kaye, 2006; Zhang et al, 2012). Structural analysis of the HMG box component of TOX showed an L-shaped structure formed by three α-helices, a common feature of HMG box proteins (Aliahmad et al, 2012). Generally, HMG box family members bind to DNA in either a sequence-dependent manner (eg. transcription factor LEF-1) or a structure-dependent manner (eg. HMGB proteins). Sequence analysis of  28 TOX revealed that it most likely binds to DNA in a structure-dependent fashion (O'Flaherty & Kaye, 2003). In line with this prediction, TOX is able to bind to distorted DNA structures such as cisplatinated DNA (Aliahmad et al, 2012).  TOX defines a small family of 4 members, including TOX, TOX2, TOX3, and TOX4, characterized by a nearly identical HMG box and other shared features (Figure 1.3). TOX family members are highly conserved in both mouse and human. The HMG box and NLS domains of these 4 family members are well conserved, and the remaining amino acids shared about 20-30% identical sequences (O'Flaherty & Kaye, 2003). The N-terminus of these four proteins have around 30-40% sequence identity, whereas the C-terminal regions show significant variations (Aliahmad et al, 2012), suggesting differences in gene functions. While TOX’s role in the immune system is best-characterized, the functions of the other three family members are largely unknown. Previous studies reported that TOX2 may be involved in the reproductive system (Kajitani et al, 2004). TOX3 has been implicated in breast cancer susceptibility (Easton et al, 2007a) and neuronal survival (Dittmer et al, 2011). Finally, TOX4 was shown to be associated with a phosphatase complex regulating chromatin structure and cell cycle progression (Lee et al, 2010). TOX4 can also interact with platinated DNA, pointing to its potential role in mediating cellular response after chemotherapy (Bounaix Morand du Puch et al, 2011).   In the mouse, TOX mRNA is most abundant in the thymus, followed by the liver and brain, with low or absent expression in other tissues (Wilkinson et al, 2002). Similarly, TOX is highly abundant in the human thymus (Su et al, 2002). TOX was initially found to be upregulated in DP thymocytes after TCR stimulation for 6 hours, indicating its involvement in positive selection (Wilkinson et al, 2002). Using a mouse model, in vivo studies supported  29 this notion, in that TOX was highly expressed in CD69+ (ie. activated) DP T cells as well as in CD4+CD8lo transitional cells. Further exploration of its expression in the earliest stages of thymocyte development revealed that TOX was also elevated during β-selection during the DN3 stage (Wilkinson et al, 2002). The expression of TOX is stage-dependent, since by the end of positive selection, TOX is effectively turned-off as T cells commit to the CD4 or CD8 lineage, and remains lowly expressed in the peripheral tissues after T cells have exited the thymus (Wilkinson et al, 2002).    30   TOXTOX2TOX3TOX4Transactivation NLS HMG-box (DNA binding)Interaction? Figure 1.3 The TOX subfamily of HMG-box proteins.  TOX contains an HMG-box DNA-binding domain that is highly conserved in three additional proteins TOX2, TOX3, and TOX4. An adjacent lysine-rich region may serve as the nuclear localization signal (NLS) as well as influence the interaction with DNA. The N-terminal domains of these proteins show approximately 30–40% sequence identity and have transactivation activity, while the C-terminal domains differ greatly between family members. Modified from Aliahmad et al. 2012, with permission to reprint.   31  1.4.2 TOX Involvement in T-cell Development Initially, TOX was found to be specifically upregulated during β-selection and positive selection stages in developing thymocytes (Figure 1.4) (Wilkinson et al, 2002). Since then, both transgenic and gene knockout strategies have been applied in mouse models to uncover the role of TOX in thymocyte development.   In TOX-transgenic (TOX-Tg) mice, driven by the lck proximal promoter, forced expression of TOX during the DN stage was able to initiate coreceptor changes, such as CD4 and CD8 upregulation. Overexpression of TOX in the DP stage led to up-regulation of RUNX3, a CD4 gene silencer, and downregulation of CD4. TOX-Tg mice showed a much expanded population of CD8 SP thymocytes, and a reduced CD4 SP population. This lineage commitment bias was linked to reduced effectiveness of TCR signaling, which favors CD8 lineage (Wilkinson et al, 2002). Interestingly, calcineurin signaling was shown to be upstream of TOX expression in this process (Aliahmad et al, 2004). It should be noted that although more CD8 SP thymocytes were produced in the thymuses of TOX-Tg mice, their maturation was impaired, in that they were unable to exit the thymus, and they failed to express some of the markers normally present on wildtype CD8 SP cells (Aliahmad et al, 2004). This suggests that factors other than TOX are needed to fully initiate the signaling pathway leading to CD8 SP thymocyte maturation and thymic exit.  In a TOX knockout (TOX-/-) mouse model, the results were somewhat surprising since here CD8 SP thymocytes developed normally, while CD4 SP lineages failed to form (Aliahmad & Kaye, 2008). The production of all CD4 T lineage cells was blocked, including T helper, FOXP3+ T regulatory, and NK T cells. All of these lineages are required to go through a DP stage and then positive selection. This developmental block happened at a  32 specific stage, as positive selection was successfully initiated, but later on the DD CD4loCD8lo populations failed to progress to the CD4+CD8lo transitional stage. CD8 SP thymocytes, however, were able to develop, migrate to the peripheral lymphoid tissue and carry out cytolylic functions (Aliahmad & Kaye, 2008). These results suggest that there is an alternative pathway for CD8 SP thymocytes to develop, other than going through the CD4+CD8lo transitional stage. Another interesting finding in TOX-/- mice was that β-selection was not affected, as DN thymocytes were able to move on to the DP stage, indicating β-selection is not TOX-dependent. One possible explanation for this result is that TOX2 is closely related to TOX, and is also highly expressed in the thymus (Su et al, 2002), therefore it could have overlapping functions with TOX in the thymus and compensate for the deficiency of TOX. This hypothesis remains to be tested.  The discrepancy between the TOX-Tg and TOX-/- mouse models suggested that TOX may have dual effects and/or context-dependent effects during thymocyte development. TOX may play a role in both the transition from DD to CD4+CD8lo and the CD4+CD8lo to CD8 SP ‘coreceptor reversal’ process (Brugnera et al, 2000). In agreement with this notion, IL-7 receptor was overexpressed in the CD8 SP thymocytes in TOX-Tg mice (Aliahmad & Kaye, 2006). IL-7 signaling contributes to the ‘coreceptor reversal’ process (Brugnera et al, 2000), during which CD8SP thymocytes develop from the CD4+CD8lo cells. The exact mechanisms underlying TOX’s role in T-cell lineage commitment warrant further investigation.    33  ThymusDN1 DN2 DN3 DN4 DP DDCD4+CD8loCD4+CD8+BloodCD4+CD8+β-selectionPositiveselectionNegativeselectionTOX TOX  Figure 1.4 Mode of TOX expression during T-cell development. TOX is transiently upregulated during β-selection and positive selection. However, as T cells mature beyond the DP stage, it is gradually downregulated prior to T cells leaving the thymus. Subsequently, TOX remains at low levels in circulating T cells as well as those found in the peripheral lymphoid tissues.     34  1.4.3 TOX beyond T-cell Development In addition to its indispensable role in T-cell development, TOX appears to be a multifunctional player in NK cell, lymphoid tissue inducer (LTi), and neuron development.  In their TOX-/- mouse, Aliahmad et al. observed a complete absence of lymph nodes, and a drastic reduction in Peyer’s patches (both number and size). These deficiencies were due to the lack of LTi cells, which are vital for lymphoid tissue organogenesis (Aliahmad et al, 2010). In addition, the development of NK lineage cells was severely impaired. TOX is normally highly expressed in both immature and mature NK cells in the bone marrow. However, without TOX, the progenitor NK cells failed to progress to the immature NK cell stage (Aliahmad et al, 2010). Previously, two other transcription regulators, Ikaros and Id2, have also been shown to be involved in the development of both LTi and NK cells (Wang et al, 1996; Yokota et al, 1999). Intriguingly, Id2 expression was significantly reduced in NK cells from the TOX-/- mice, suggesting that TOX could act as an upstream regulator of Id2 expression. However, TOX knockdown experiments in human cord blood CD34+ cells using small interfering RNA (siRNA) showed no change in Id2 expression, possibly highlighting potential mechanistic differences in the role of TOX between human and murine immune systems.  TOX is also highly expressed in the human brain, suggesting a role in the nervous system. Indeed, TOX was recently identified as a multifunctional player for brain development, neural stem cell differentiation, and neural cell outgrowth (Artegiani et al, 2015). Furthermore, using the DNA adenine methytransferase identification method, Artegiani et al. demonstrated several downstream targets of TOX, including Sox2, Tbr2, and Prox1, among others (Artegiani et al, 2015). These results emphasize that TOX is a versatile  35 regulator in multiple developmental steps, and provide a springboard for future studies on the mechanistic role of TOX in cell fate.    1.4.4 TOX in T-cell Malignancies Given TOX’s important role in the immune system, including all T-cell lineages, NK cells, and lymph node organogenesis, it is not surprising that if abnormally expressed, TOX may lead to malignant transformation. In fact, forced TOX expression caused increased IL-7 receptor expression, amplifying a signaling pathway that promotes cell survival and proliferation (Aliahmad & Kaye, 2006). In addition, the oncogenic potential of TOX can be inferred by analogy with its family member, TOX3, which confers breast cancer susceptibility and associates with bone metastasis (Stacey et al, 2007). TOX in T-cell Acute Lymphoblastic Leukemia (T-ALL) TOX was genetically amplified and overexpressed in human T-cell Acute Lymphoblastic Leukemia (T-ALL) (Maser et al, 2007). Using a zebrafish transgenic model, Lobbardi et al. demonstrated that overexpressing TOX has a synergistic effect with both Myc and Notch in inducing T-ALL (Lobbardi et al, 2014). Further, overexpression of TOX was shown to inhibit a DNA repair protein complex KU70/KU80, which is essential in initiating Non-Homologous End Joining (NHEJ) repair (Walker et al, 2001), thus inhibiting DNA repair and promoting genomic instability in T-ALL (Lobbardi et al, 2014). In the context of thymocyte development, TOX is first up-regulated during β-selection, when TCR rearrangement of the β,γ, and δ gene loci begins, which involves the NHEJ mechanism in segment joining. Possibly the activation of TOX at this stage could lead to inhibition of the NHEJ process and prolonged DNA repair time, promoting accurate recombination of segments that are located  36 distantly (Lobbardi et al, 2014). However, this mechanism is abnormally activated in T-ALL, contributing to accumulation of mutations and a subsequent pre-malignant state. TOX in MF The first evidence of TOX overexpression in MF came from our observation in early stage MF, where strong nuclear TOX staining was observed in CD4+ T cells in MF lesional skin. In contrast, little to no TOX staining was found in the CD4+ T cells from BID (Zhang et al, 2012). McGirt et al. observed enhanced TOX expression in both early and advanced stage MF skin lesions, and demonstrated TOX as a downstream target of miR-223 (McGirt et al, 2014). However, functional characterization of the role of TOX in CTCL has not been conducted, and whether TOX plays a role in the malignant transformation in CTCL is unknown.  1.5 Thesis Outline 1.5.1 Research Hypothesis The early recognition and accurate diagnosis of CTCL is essential for improving patient outcomes. However, due to the lack of robust molecular markers in this heterogeneous group of diseases, it is difficult to establish a diagnosis and start appropriate treatment. Moreover, the clinical course of CTCL is highly variable, and so far there is no molecular marker available to aid in prognosis and guide patient management. The marked enhancement of TOX expression that we observed in early MF lesions, but not in any normal skin or BID lesions, led us to propose that Aberrantly expressed TOX could serve as a diagnostic and prognostic marker for CTCL.  Although considerable progress has been made into the molecular pathogenesis of CTCL, much is still unknown about what leads to uncontrolled proliferation and apoptosis  37 resistance in CTCL cells. Abundant evidence supports TOX playing a critical role in the regulation of thymocyte development, including affecting CD4 and CD8 lineage commitment, during which TOX suppresses the threshold of TCR signaling. However, TOX is downregulated before T cells exit the thymus, and it normally remains minimally expressed in the peripheral lymphoid tissues (Wilkinson et al, 2002). Thus the increased expression of TOX in CTCL is likely to be pathogenic. Interestingly, TOX3, one of the TOX family members, has been implicated in breast cancer susceptibility and bone metastasis, suggesting the oncogenic potential of TOX family members (Easton et al, 2007b; Stacey et al, 2007). We therefore hypothesize that TOX overexpression plays an oncogenic role in the pathogenesis of CTCL. 1.5.2 Specific Aims Within the scope of this thesis we sought to fill knowledge gap with respect to molecular markers and disease pathogenesis of CTCL. I proposed two specific aims to test the above hypotheses. Specific Aim 1. To evaluate the potential value of TOX as a diagnostic and prognostic marker for CTCL.  In Chapter 3, we assembled a panel of CTCL cases, measured TOX mRNA and protein levels, and correlated TOX levels with tumor burden and long term clinical outcomes in CTCL. We formulated the following sub-aims: 1. To test if TOX is overexpressed in the full-spectrum of MF and in peripheral CD4+ T cells in SS patients; 2. To examine if TOX transcript levels can differentiate CTCL from non-CTCL cases in large sample cohorts;  38 3. To assess if TOX transcript levels could be correlated with CTCL prognosis, including disease progression and disease-specific mortality Specific Aim 2. To investigate the pathogenic role of TOX overexpression in the development of CTCL. In Chapter 4, I systematically examined the biological and molecular effects of TOX knockdown in patient-derived CTCL cell lines, which show high TOX expression, and explored the mechanism of TOX’s action by identifying its downstream mediators. This section has the following sub-aims:  1. To assess if specific knockdown of TOX in CTCL cell lines has an effect on cellular phenotypes, including cell viability, colony formation ability, and in vivo tumor formation ability; 2. To investigate major molecular pathways affected by TOX knockdown in CTCL cell lines, including apoptosis and cell cycle progression; 3. To explore potential downstream molecules of TOX in the context of CTCL, and validate their functional significance in the face of TOX loss; 4. To explore potential mechanisms that are responsible for ectopic TOX activation in CTCL    39 Chapter 2: Materials and Methods 2.1 Primary Human Samples Skin biopsies of 113 MF patients were included in this study (see Table 2.1 for summary, and Table 2.3 for individual information). These patients were prospectively collected as two cohorts from three clinical centers. Cohort 1 (n = 54) was recruited from the Skin Lymphoma Clinic of the British Columbia Cancer Agency (BCCA, Vancouver, Canada, n = 26), and Peking University First Hospital (Beijing, China, n = 28) over the period 2008-2012 (Huang et al, 2014). Cohort 2 (n = 59) was a well-characterized cohort collected from Harvard University with institutional approval, for which long-term clinical outcome data (up to 6 years) was available. (Litvinov et al, 2010; Litvinov et al, 2012; Litvinov et al, 2013; Shin et al, 2007). For controls, 25 individuals with BID (psoriasis, n = 9; chronic dermatitis, n = 13; pityriasis rubra pilaris, n = 3) and 11 volunteers with normal healthy skin (HS) were recruited during the same interval from the outpatient dermatology clinic of the University of British Columbia (UBC, Vancouver, Canada). Full-thickness lesional skin or normal healthy skin samples were harvested by 3mm punches, which were immediately immersed in RNAlater, and stored at -20°C before RNA extraction (Zhang et al, 2012).  Peripheral blood samples from 12 SS patients (see Table 2.2 for summary, and Table 2.4 for individual information), 18 patients with BID (psoriasis, n = 8; rosacea, n = 5; vitiligo, n = 5), and 9 normal participants were collected from the Skin Lymphoma Clinic, BCCA, and the Skin Care Center of Vancouver General Hospital with informed consent, and with approval from the Clinical Ethics Board of University of British Columbia (H12-02653 and C98-0493). The leukemic Sézary cells and normal blood cells were purified by negative selection with monoclonal antibodies directed against granulocytes, B cells, CD8+ and CD7+  40 T cells using a Rosette Sep kit (StemCell Technologies, Vancouver, BC, Canada). Over 90% purity of desired cells was confirmed by FACS analysis of most patient samples, using phycoerythrin (PE)-conjugated anti-CD7 antibody and fluorescein isothiocyanate (FITC)-conjugated anti-CD4 antibody (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA), as described previously (Su et al, 2003) (Ringrose et al, 2006).  The diagnosis and clinical staging were established according to the diagnostic criteria of CTCL (Olsen et al, 2007), based on the clinical, histopathologic and immunopathologic features of each patient. Human T-cell leukemia virus-1 status was tested and confirmed to be negative in all patients. Every participant provided informed consent. The procedures used in this study were approved by the Research Ethics Board of the University of British Columbia, in accordance with the Declaration of Helsinki principles (H12-02653 and C98-0493).    41 Table 2.1 Summary of demographics for patients with MF (n = 113)   Table 2.2 Summary of demographics for patients with SS (n = 12) Demographics No % Sex Male 6 50 Female 6 50 Age at diagnosis (years) Median 69.5 Range 43-82 Race Caucasian 11 92 Japanese 1 8 TCR Clonality Yes 7 58 No 5 42 Demographics  Cohort 1 (n=54) Cohort 2 (n=59) Sex     Male 40 (74%) 36 (61%)     Female 14 (26%) 23 (39%) Age at diagnosis (years)     Median (range) 50 (28-85) 62 (26-91) Race     Caucasian 17 (31%) 59 (100%)     Chinese 31 (57%)      Other 6 (11%)  Clinical stage     I 31 (57%) 42 (71%)     II 16 (30%) 4 (7%)     III 5 (9%) 5 (8%)     IV 2 (4%) 8 (14%)  42 Table 2.3 Demographics and clinical characteristics of individual MF patients (n = 113) Samples ID Centre Eth Sex# Age Lesion type  Stage TOX level* Death Status Total F/U (mo)  Time to 1st Progression (mo)a Time to Death (mo) Cause of Death  MF 00 Van Cau M 53 Plaque I 30.13 NA NA NA NA NA MF 01 Van Cau F 74 Patch I 11.03 NA NA NA NA NA MF 03 Van Chi M 43 Patch I 7.17 NA NA NA NA NA MF 05 Van Cau M 46 Plaque I 41.08 NA NA NA NA NA MF 06 Van Cau F 64 Plaque I 13.19 NA NA NA NA NA MF 07 Van Cau F 46 Plaque I 38.76 NA NA NA NA NA MF 08 Van Ind M 30 Plaque I 16.55 NA NA NA NA NA MF 09 Van Cau F 60 Plaque I 15.88 NA NA NA NA NA MF04A Van Cau M 78 Plaque I 7.60 NA NA NA NA NA MF04B Van Cau M 78 Plaque I 10.07 NA NA NA NA NA MF10 Van Cau F 55 Plaque IV 3.26 NA NA NA NA NA MF11 Van Cau M 30 Plaque I 1.18 NA NA NA NA NA MF12 Van Cau M 74 Plaque II 6.12 NA NA NA NA NA MF13 Van Ind M 49 Plaque II 20.01 NA NA NA NA NA MF14 Van Ind M 47 Plaque III 29.04 NA NA NA NA NA MF15 Van Asian F 40 Plaque I 81.95 NA NA NA NA NA MF16 Van Ind M 40 Plaque II 24.99 NA NA NA NA NA MF17 Van Cau M 60 Patch I 29.38 NA NA NA NA NA MF19F Van Chi M 54 Plaque III 3.44 NA NA NA NA NA MF19N Van Chi M 54 Tumor (LCT) III 7.95 NA NA NA NA NA MF20 Van Asian F 85 Plaque I 12.48 NA NA NA NA NA MF21 Van Cau M na Patch I 29.24 NA NA NA NA NA  43 Samples ID Centre Eth Sex# Age Lesion type  Stage TOX level* Death Status Total F/U (mo)  Time to 1st Progression (mo)a Time to Death (mo) Cause of Death  MF22(1) Van Cau M na Patch I 4.51 NA NA NA NA NA MF22(2) Van Cau M na Patch I 9.15 NA NA NA NA NA MF23 Van Cau M 70 Patch I 4.53 NA NA NA NA NA MF24 Van Cau M 48 Patch I 11.86 NA NA NA NA NA CTCL009 BJ Chi M 30 Plaque I 42.92 NA NA NA NA NA CTCL016 BJ Chi M 28 Plaque I 47.64 NA NA NA NA NA CTCL021 BJ Chi M 61 Plaque III 102.12 NA NA NA NA NA CTCL023 BJ Chi M 52 Plaque I 34.21 NA NA NA NA NA CTCL024 BJ Chi F 64 Plaque II 183.16 NA NA NA NA NA CTCL025 BJ Chi F 47 Plaque I 29.28 NA NA NA NA NA CTCL028 BJ Chi M 55 Plaque II 76.42 NA NA NA NA NA CTCL033 BJ Chi M 52 Plaque III 4.74 NA NA NA NA NA CTCL037P BJ Chi F 44 Tumor II 19.27 NA NA NA NA NA CTCL037T BJ Chi F 44 Tumor II 82.76 NA NA NA NA NA CTCL038 BJ Chi F 48 Patch I 20.74 NA NA NA NA NA CTCL039 BJ Chi M 67 SS IV 21.54 NA NA NA NA NA CTCL040(1) BJ Chi M 58 Plaque II 159.84 NA NA NA NA NA CTCL040(2) BJ Chi M 58 Plaque II 203.36 NA NA NA NA NA CTCL041 BJ Chi F 37 Patch I 42.84 NA NA NA NA NA CTCL042 BJ Chi M 47 Tumor II 90.45 NA NA NA NA NA CTCL058 BJ Chi M 52 Tumor II 133.89 NA NA NA NA NA CTCL060 BJ Chi M 82 Plaque I 78.89 NA NA NA NA NA CTCL066 BJ Chi M 42 Patch I 10.16 NA NA NA NA NA CTCL069 BJ Chi M 37 Plaque II 71.23 NA NA NA NA NA  44 Samples ID Centre Eth Sex# Age Lesion type  Stage TOX level* Death Status Total F/U (mo)  Time to 1st Progression (mo)a Time to Death (mo) Cause of Death  CTCL082 BJ Chi M 40 Plaque I 32.51 NA NA NA NA NA CTCL089 BJ Chi M 33 Patch I 8.65 NA NA NA NA NA CTCL093 BJ Chi F 43 Patch II 48.78 NA NA NA NA NA CTCL104 BJ Chi M 74 Plaque I 23.14 NA NA NA NA NA CTCL107 BJ Chi M 53 Tumor II 331.75 NA NA NA NA NA CTCL112 BJ Chi M 28 Tumor (LCT) II 126.25 NA NA NA NA NA CTCL116 BJ Chi M 35 Tumor (LCT) II 18.19 NA NA NA NA NA CTCL124 BJ Chi M 50 Plaque I 951.92 NA NA NA NA NA M1 MB Cau F 50 NA I 5.65 Alive 40       M10 MB Cau F 56 NA I 1.32 Alive 52       M11 MB Cau F 72 NA I 35.51 Alive 27       M12 MB Cau M 77 NA III 0.80 Alive 65       M13 MB Cau M 80 NA I 2.54 Alive 66       M14 MB Cau M 64 NA I 5.22 Alive 4       M15 MB Cau M 65 NA I 8.22 Alive 53       M16 MB Cau M 62 NA IV 10.30 Dead 5 1 5 CTCL M17 MB Cau F 79 NA IV 4.62 Dead 36 3 36 CTCL M18 MB Cau M 72 NA I 1.81 Alive 41       M19 MB Cau M 62 NA I 1.48 Alive 5 5     M2 MB Cau F 79 NA II 3.64 Dead 14 14 14 CTCL M20 MB Cau F 72 NA III 3.88 Alive 3 3     M21 MB Cau F 66 NA I 1.40 Alive 18       M22 MB Cau F 73 NA IV 13.97 Dead 4 1 4 CTCL  45 Samples ID Centre Eth Sex# Age Lesion type  Stage TOX level* Death Status Total F/U (mo)  Time to 1st Progression (mo)a Time to Death (mo) Cause of Death  M23 MB Cau M 62 NA I 5.25 Alive 63       M24 MB Cau F 65 NA III 26.99 Alive 28 28     M25 MB Cau M 91 NA I 5.07 Dead 32      Other M27 MB Cau M 90 NA I 3.08 Dead 11      Other M28 MB Cau M 58 NA I 3.55 Alive 61       M29 MB Cau M 50 NA I 6.59 Alive 60       M30 MB Cau F 30 NA I 6.27 Alive 61       M31 MB Cau M 36 NA I 5.75 Alive 59       M33 MB Cau M 42 NA I 5.25 Alive 55       M34 MB Cau F 70 NA I 7.69 Alive 59       M35 MB Cau F 90 NA IV 105.93 Dead 48 48 48 CTCL M36 MB Cau M 68 NA II 38.51 Alive 38       M37 MB Cau F 75 NA I 3.99 Alive 40 40     M38 MB Cau M 55 NA I 23.22 Alive 54       M40 MB Cau M 62 NA I 75.80 Alive 8 8     M41 MB Cau F 84 NA I 4.50 Alive 45 45     M42 MB Cau F 54 NA I 4.17 Alive 53       M43 MB Cau M 63 NA I 3.13 Alive 10       M44 MB Cau M 56 NA I 23.07 Dead 36 12 36 CTCL M45 MB Cau F 40 NA I 3.66 Alive 53       M46 MB Cau M 58 NA I 2.33 Alive 4 4     M47 MB Cau F 74 NA IV 4.23 Alive 52       M48 MB Cau M 63 NA III 39.22 Alive 9 9     M49 MB Cau F 52 NA I 13.59 Alive 50        46 Samples ID Centre Eth Sex# Age Lesion type  Stage TOX level* Death Status Total F/U (mo)  Time to 1st Progression (mo)a Time to Death (mo) Cause of Death  M50A MB Cau M 35 NA I 62.65 Alive 47       M51 MB Cau M 48 NA II 41.14 Dead 49 48 49 CTCL M52 MB Cau M 62 NA I 37.09 Alive 31 31     M53 MB Cau F 52 NA IV 91.99 Dead 10 9 10 CTCL M54 MB Cau F 66 NA IV 38.05 Dead 31 31 31 CTCL M55 MB Cau F 26 NA I 12.85 Alive 45       M56 MB Cau M 43 NA I 5.66 Alive 47       M57 MB Cau M 71 NA IV 149.00 Dead 4 4 4 CTCL M58 MB Cau M 80 NA I 13.59 Alive 45       M59 MB Cau M 37 NA II 8.51 Dead 19 12 19 CTCL M5A MB Cau M 69 NA I 8.62 Alive 21       M5E MB Cau M 68 NA I 2.27 Alive 21       M60 MB Cau M 35 NA I 41.92 Alive 27       M61 MB Cau M 57 NA I 16.58 Alive 44       M62 MB Cau F 48 NA I 2.89 Alive 43       M63 MB Cau M 42 NA I 97.34 Alive 38       M64 MB Cau F 47 NA I 16.58 Alive 51       M65 MB Cau M 55 NA III 37.22 Alive 40       M8 MB Cau M 51 NA I 0.01 Alive 65       M9 MB Cau M 51 NA I 12.38 Alive 64        MB = Montreal/Boston; Van = Vancouver; BJ = Beijing; Ind = South Asian Indian; Cau = Caucasian; Chi = Chinese; Eth = ethnicity;  # M = male; F = female; LCT = large cell transformation; F/U = follow up; mo = months; * = TOX mRNA level per 1000 ACTB;  a = time between diagnosis and the first noted progression towards higher clinical stage; NA = not available  47 Table 2.4 Demographics and clinical characteristics of individual SS patients (n = 12) ID Eth Sex# Age TOX level* Death Status Total F/U (mo) Time to Death (mo) Cause of Death  Sézary cells (%) CD4+ cells (%) CD4+ CD7- cells (%) CD8+ cells (%) CD4/CD8 Ratio TCR clonality SS1 Cau M 78 349.47 Dead 39 39 CTCL >5% 0.95 0.94 0.01 94 Yes SS3  Jap M 73 15.80 Dead 120 120 CTCL >5% 0.54 0.27 0.26 2.1 No SS4 Cau F 66 83.44 Alive 152     >5% 0.94 0.84 0.04 24 No SS5 Cau M 82 565.70 Dead 47 47 CTCL >5% 0.95 0.94 0.01 94 Yes SS6 Cau M 59 98.73 Dead 51 51 CTCL >5% NA NA NA NA No SS7 Cau F 77 52.85 Dead 14 14 CTCL >5% 0.88 0.8 0.01 88 No SS8 Cau M 52 18.60 Dead 34 34 Other >5%  0.85 0.03 0.08 10.6 Yes SS9 Cau F 82 145.29 Dead 27 27 CTCL >5% 80 0.68 4 20 Yes SS10  Cau F 74 166.05 Dead 20 20 CTCL >5% NA NA NA NA No SS11 Cau M 60 227.79 Dead 6 6 CTCL >5% 0.95 0.79 0.01 95 Yes SS12 Cau F 43 218.62 Alive 15     >5% 0.61 0.76 0.046 13.1 Yes SS13 Cau F 51 129.94 Alive 15     >5% 0.604 0.5 0.307 1.8 Yes  SS patients were all from Vancouver centre.  Eth = ethnicity; Cau = Caucasian; Jap = Japanese; # M = male; F = female; * = TOX mRNA level per 1000 GAPDH; F/U = follow up; mo = months; NA = not available  48 2.2 Cell Lines and Cell Culture Human CTCL cell line Hut78, HH (ATCC no. TIB-161, CRL-2105), SZ4 (a generous gift from Dr. James Herman at Johns Hopkins University), as well as a panel of CTCL cell lines (H9, Myla, Mac2A, PB2B, MJ, Sez4, and SeAx, a generous gift from Dr. Ivan Litvinov at the University of Ottawa) were used in this study. Generally, cells were cultured in Rosewell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, 0.1 mg/mL streptomycin, and 10−4 M β-mercaptoethanol (STEMCELL Technologies, Vancouver, BC, Canada), at 37°C in a humidified cell culture incubator with 5% CO2. Sez4 and SeAx wells were cultured in the above culture media with the addition of 5 ng/mL recombinant human IL-2 and IL-4 (R&D Systems, Minneapolis, MN). For culture of transduced CTCL cells, vector-specific selection antibiotics (Invitrogen, Burlington, ON, Canada) were added to the above medium (puromycin for pLKO.1 puro, hygromycin for pLKO.1 hygro, and blasticidin for pLKO.1 blast). The working concentrations for selection antibiotics were determined by kill curves according to the manufacturer’s instructions. Briefly, untransduced cells were plated at 1 × 105 cells/mL and a series of concentrations of antibiotics were added to the culture media. The minimum dosage that killed all of the cells within 5-7 days was determined as the working concentration for each cell line. Using this method, the working concentration of puromycin for all three cell lines was determined to be 1 ug/mL, the working concentrations of hygromycin for Hut78 and HH were 750 μg/mL and 200 μg/mL, respectively, and the working concentrations of blasticidin for Hut78 and HH were 15 μg/mL and 10 μg/mL, respectively. Culture media for the cells was changed every 2-3 days, and the cell culture density was maintained at less at 1 × 106 cells/mL.   49 2.3 Lentiviral Vectors For TOX knockdown, pLKO.1 vector containing small hairpin RNA (shRNA) inserts that specifically target human TOX (SHCLNG-XM_376776, Sigma Aldrich, Saint Louis, MO, USA) were purchased, as well as a non-targeting shRNA control (SHC002, Sigma Aldrich). For CDKN1B and CDKN1C knockdown, the pLKO.1 hygro vector with U6 promoter was used (Addgene plasmid # 24150, a kind gift from Bob Weinberg). Briefly, annealed oligos containing shRNA sequences specifically targeting human CDKN1B and CDKN1C were ligated into the AgeI and EcoRI cloning sites on the pLKO.1 hygro vector using the Rapid DNA Dephos and Ligation Kit (Roche), followed by transformation of the ligation products into One Shot® Stbl3TM Chemically Competent E. coli (Invitrogen) following the manufacturer’s instructions. The transformation mixtures were plated on ampicillin (100 μg/mL) selection plates overnight at 37°C, positive colonies were selected, expanded and verified by Sanger sequencing (the NAPS Unit, Michael Smith Laboratories, UBC). For SMAD3 knockdown, the pLKO.1 blast vector containing the U6 promoter was used (Addgene plasmid # 26655, a gift from Keith Mostov) (Bryant et al, 2010), and shRNA sequences specific for SMAD3 were cloned into the vector using the same method descrined above, and verified by Sanger sequencing. The oligonucleotides encoding the shRNAs are included in Table 2.5.  50  Table 2.5 Oligonucleotides encoding the shRNAs shRNA insert  Sequence  Control-sh  CAACAAGATGAAGAGCACCAA TOX-sh1 CCCTGAAATCACAGTCTCCAA TOX-sh2 CGACTATCAGACTATTATCAA CDKN1B-sh1 GTAGGATAAGTGAAATGGATA CDKN1B-sh2 GCGCAAGTGGAATTTCGATTT CDKN1C-sh1 GTAAAGCTTTAAGAGTCATTT CDKN1C-sh2 GAACCGCTGGGATTACGACTT SMAD3-sh1  GCCTCAGTGACAGCGCTATTT  SMAD3-sh2  GGATTGAGCTGCACCTGAATG   51  2.4 Production of Lentiviral Particles The polyethylenimine (PEI, Polyscience, PA, USA) transfection method was used to generate levtiviral particles. One day before transfection, 6 × 106 293 T cells were seeded onto each 10-cm Falcon® tissue culture dish in Dulbecco's Modified Eagle Medium (DMEM, HyClone, Utah, USA) with 10% FBS. Prior to transfection, all reagents were brought to room temperature (RT), and culture media was changed with 4.5 mL fresh media added to each plate. For each plate, a mixture of 40 μL PEI (1 μg/μL) and 13.5 μg DNA were added, including 6 μg lentiviral plasmid DNA, 3.9 μg ∆R, 1.5 μg REV, and 2.1 μg vesicular stomatitis virus glycoprotein (VSV-G) envelope constructs. The ∆R, REV, and VSVG plasmids were gifts from P. Leboulch (Harvard University, Boston, MA, USA). Forty-eight hours post-transfection, viral supernatant was harvested, filtered through 0.45 μm low protein binding filters, and concentrated by ultracentrifugation at 25,000 rpm for 90 mins. Another 5 mL fresh media was added to each plate, and the viral supernatant was harvested again after a further 24 hours (72 hours post-transfection) using the same procedure mentioned above. The concentrated virus was stored at -80 °C until use. Generally, viral supernatant was concentrated 175 fold from 35 mL to 200 μL.    2.5 Generation of Stably Transduced Cells CTCL cells were plated at a density of 2 × 105 cells/mL in 1 mL culture media in 24-well plates. For each plate, 5 μL of the concentrated virus was added, followed by the addition of 5 μL of 2 mg/mL protamine sulfate. Cell mixtures were incubated at 37 °C for 18-24 hours, washed with Dulbecco’s Phosphate Buffered Saline (PBS) (StemCell Technologies, Vancouver, BC) two times, and re-suspended in fresh culture media. After 24 hours, selection antibiotics were added to the cells to select stably transduced cell populations. Bulk  52 populations transduced either by control (CTR) or gene-specific shRNA viruses were used. Stable knockdown was confirmed by Western blotting 5-7 days after selection was started.  2.6 Protein Extraction and Western Blotting Cells were pelleted, re-suspended in lysis buffer (1 μL for 3 × 104 cells), and incubated at 4 °C for an hour. The protein lysis buffer (per 1 mL) consisted of 10 μL 10% SDS, 100 μL of 10% NP-40 Alternative Protein Grade Detergent (Calbiochem, Gibbstown, NJ), 10 μL 200mM phenylmethylsulfonyl fluoride (PMSF) (Sigma-Aldrich, Oakville, ON), 10 μL protease inhibitor cocktail (PIC) (Sigma-Aldrich, Oakville, ON), and 870 μL phosphorylation solubilization buffer (PSB). Supernatants were collected by centrifugation at 12,000 rpm for 10 minutes, and stored at -80 °C.  Protein concentration was evaluated by Pierce™ BCA Protein Assay Kit (Thermo Scientific, MA, USA) according to the manufacturer’s instructions. The absorbance of protein samples at 562 nm was measured using Elx808 Absorbance Microplate Reader (BioTekk Instruments, Winooski, VT).  Western blotting was used to evaluate protein expression. Briefly, protein lysates (20-30 μg/lane) were incubated at 70 °C for 10 minutes, separated on 10%, 12% or 15% sodium dodecyl sulfate (SDS)-polyacrylamide gels and blotted onto Immobilon-P polyvinylidene fluoride 0.45-μm membrane (Millipore, Billerica, MA). The membranes were blocked in 5% skim milk in Tris-buffered saline containing 0.05% Tween-20 (TBST) for 1 hour at RT, washed in TBST for two times (5 minutes each), and incubated with primary antibody overnight at 4 °C. Then the membrane was washed in TBST two times (10 minutes each), and incubated with horseradish peroxidase–conjugated secondary antibody (1: 2000 dilution) for 1 hour at RT. Finally, after washing in TBST for three times (10 minutes each), the membrane was incubated with enhanced chemiluminescence reagent for 1 minute, and  53 exposed on KODAK® BioMax® XAR autoradiography film (VWR, Mississauga, ON, Canada) to visualize the proteins. The quantity of protein was measured against a human actin loading control. The primary and secondary antibodies are listed in Table 2.6.  Table 2.6 Primary antibodies used in Western blotting Antibody  Host  Dilution  Supplier (Catalogue#)  TOX  Rabbit (polyclonal)  1:500  Sigma Aldrich (HPA018322)  p27 (C-19)  Rabbit (polyclonal)  1:100  Santa Cruz (sc-528)  p57 (C-20)  Rabbit (polyclonal)  1:100  Santa Cruz (sc-1040)  Caspase 3  Rabbit (polyclonal)  1:500  Cell Signaling Technology (9662)  Caspase 9  Rabbit (polyclonal)  1:250  Cell Signaling Technology (9502)  SMAD3  Rabbit (monoclonal)  1:1000  Cell Signaling Technology (9523)  Actin (AC-40)  Mouse (monoclonal )  1: 2000  Sigma Aldrich (A3853)    2.7 RNA Extraction and Reverse Transcription Total RNA was extracted by TRIzol (Invitrogen) or QIAzol (Qiagen) reagent following the manufacturer’s protocol. RNA was dissolved in RNAse free water (60 μL for 1× 106 cells), and RNA concentration was measured by the nanodrop ND-100 spectrophotometer at 260 nm. Subsequently, 100 ng of RNA was reverse transcribed into cDNA in a 20-μL reaction using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Carlsbad, CA, USA) according to the manufacturer’s protocol.   2.8 Real-time Quantitative Polymerase Chain Reaction Real-time quantitative polymerase chain reaction (qPCR) was performed using Fast SYBR® Green Master Mix (Applied Biosystems). Each reaction of 12 μL consisted of 6 μL Fast SYBR® Green Master Mix reagent, 1 μL cDNA, 0.25 μL or 0.36 μL 10-μM gene-specific primers, and RNAse free water (to make up 12 μL). Reactions were run on the 7500 Real  54 Time PCR System (Applied Biosystems) following the manufacturer’s instructions. Each primer set was tested for specificity and primer efficiency (90%-110%) prior to use. All experiments were run in duplicate or triplicate. To quantify gene expression, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or beta-actin (ACTB) was used as an endogenous control, and gene expression levels were expressed as mRNA copies per 1000 GAPDH or ACTB copies in all experiments, as described (Huang et al, 2014; Huang et al, 2015). The primers used in qPCR are listed in Table 2.7.  55  Table 2.7 Primers for qPCR Gene  Forward primer (5’→3’)  Reverse primer (5’→3’)  TOX  GTGCAGAAATCCTCCCCCAC  TTTGTCCCTCTGCATGCCC  CD4  AGCAGAGCGGATGTCTCAGATC  AAACCGGTGAGGACACTGGC  CDKN1A  TGTCCGTCAGAACCCATGC  AAAGTCGAAGTTCCATCGCTC  CDKN1B  TAATTGGGGCTCCGGCTAACT  TGCAGGTCGCTTCCTTATTCC  CDKN1C  GCGGCGATCAAGAAGCTGT  GCTTGGCGAAGAAATCGGAGA  CDK2  GTACCTCCCCTGGATGAAGAT  CGAAATCCGCTTGTTAGGGTC  CDK4  TCAGCACAGTTCGTGAGGTG  GTCCATCAGCCGGACAACAT  SMAD3  TGGACGCAGGTTCTCCAAAC  CCGGCTCGCAGTAGGTAAC  FOXO3  CGGACAAACGGCTCACTCT  GGACCCGCATGAATCGACTAT  HBP1  GCATTCACAAGGGCTATGGTT  ACAGACTCGCCAAATGATACAC  CRTC1   TCAGTGGACAAACACGGACG  CAGGGCGGAGTCAGAATTG  THAP1   TACTAATGCCGCCTCTTCAGA  TGCTGTTCTAGCTGATGAATCCT  MED13L   TTGTCCGACCCTACGAAAAGG  GTGCTGGGCAATCTCCACA  ITM2B   TACAAACTGCAACGCAGAGAA  AAATGCCGAATTGCGAAACAA  TIAM2   TACCACCTGACGGAAGCACTA  ACACGGTCCCATAATCCTCATA  BTBD7   AAAGCCCACAAGGCTGTTATT  TCTTCACCAGTTCGTATCCTCC  FAM117A   CTGCCTCCCCTCGACCTAAT  CAAACACACGCACTTTCTCAC  SLC26A11   CCGCTGTTCGACACCAAGAT  AGGATGCCGTACTGCACCT  TNFRSF12A   CTCGCCCACTCATCATTCAT  TTGTGGTTGGAGGAGCTTG  CGREF1   ACGATGACAGTGTTAATCCTGC  CCTAGTCCCTTTAGGTAGCTCTG  PSMB5   AGGAACGCATCTCTGTAGCAG  AGGGCCTCTCTTATCCCAGC  ERCC2   GTCGATGGGAAATGCCACAG  GTCATCCAGGTTGTAGATGCC  FKBP4   GAAGGCGTGCTGAAGGTCAT  TGCCATCTAATAGCCAGCCAG  CD3EAP   AGATACGGAGCTGTGGCTTAT  CCCATTGAAGCATTCTGGGG  MRPL12   ATCCCCATAGCGAAAGAACGG  GGACGAGGTTGATGCCTTGG  XPO5   CTGGGTGTCTATGAGTCACATCA  TTCCGGTCTTCCAACTTGCC  MGAT4A   AAAATCCATGTAAACCCACCTGC  AGTCTCCAGCTATCGGTGTGA  RAB3A   CCATCTATCGCAACGACAAGAG  CCATAGCGCCCCGGTAGTA  GZMB   TACCATTGAGTTGTGCGTGGG  GCCATTGTTTCGTCCATAGGAGA  GAPDH  CGCTCTCTGCTCCTCCTGTT  CCATGGTGTCTGAGCGATGT  ACTB  CTGGAACGGTGAAGGTGACA  AAGGGACTTCCTGTAACAATGCA   56  2.9 Cell Viability Assay  Viable cells were determined using trypan blue exclusion method. Cells were plated in 12-well plates containing 2 mL culture medium at a density of 1.5 × 105 cells/mL or 2.5 × 105 cells/mL, and incubated at 37 °C for 24, 48, 72, and 96 hours. For single TOX knockdown cells, puromycin (1 μg/mL) was added to RPMI 1640 culture medium. For co-knockdown of TOX and CDKN1B or CDKN1C, both puromycin (1 μg/mL) and hygromycin (750 μg/mL for Hut78; 200 μg/mL for HH) were added to the culture medium. At each time point, viable cell number in each well was determined. A minimum of three biological replicates were performed for each cell line, and each cell line was plated in duplicate or triplicate wells in a single assay.  2.10 Colony-forming Cell Assay Colony-forming cell (CFC) assay was performed in fetal calf serum-containing methylcellulose (H4230; StemCell Technologies) with 1 μg/mL puromycin. First, the semi-solid culture mixture was generated by adding 20 mL Iscove’s Media (StemCell Technologies) to 80 mL methylcellulose. Then, 600 cells were re-suspended evenly to 3 mL culture mixture, and plated into CFC plates in duplicate (1.2 mL each). Cells were cultured at 37 °C for 12-14 days, before colonies were counted in each plate using standard scoring criteria (Kennah et al, 2009; Ringrose et al, 2006). Colony size was determined as big (>500 cells), medium (50-500 cells), and small (20-50 cells).  2.11 Activation of CD4+ T Cells Peripheral blood CD4+ T cells (StemCell Technologies) were activated either by addition of 25ng/mL phorbol 12-myristate 13-acetate (PMA) and 50ng/mL ionomycin (Sigma Aldrich), or by CD3 (mouse anti-human) antibody with or without CD28 (mouse anti-human)  57 antibodies (BD Biosciences, Mississauga, ON, Canada), as described previously (Herold et al, 2003; Klemke et al, 2009; Lenardo, 1991; Wang et al, 2011). To account for non-specific binding, an isotope control antibody (BD Biosciences) was included.    2.12 Fluorescence-activated Cell Sorting (FACS) Analysis FACS analysis was performed on a FACSCanto II or a Fortessa cell analyzer (BD Biosciences). Data was analyzed with FlowJo 7.6 software (Tree Star, Ashland, OR, USA). For human peripheral blood, cells in the lymphocyte gate were used. Rabbit anti-human TOX (PE-conjugated, eBioscience, San Diego, CA), and mouse anti-human CD4 (APC-eFluor 780 -conjugated, eBiosience) were used to assess CD4 and TOX status.  2.13 Apoptosis Assay Apoptosis assay was performed using an apoptosis detection kit (BD Biosciences) according to the manufacturer’s instructions. Total apoptotic cell numbers were calculated as total annexin V-positive cells.  2.14 5-bromo-2'-deoxyuridine (BrdU) Incorporation Assay To determine the percentage and nature of the cells that are actively synthesizing DNA, BrdU incorporation assays were performed using a FITC BrdU Flow Kit (BD Biosicences) according to the manufacturer’s protocol. Briefly, transduced CTCL cells were pulsed with 10 μM BrdU at 37 °C (45 minutes for Hut78; 2 hours for HH and SZ4). During this time, BrdU, which is an analogue of thymidine, was incorporated into newly synthesized DNA by cells that are actively progressing through DNA synthesis phase. The incorporated BrdU was then measured by FITC-conjugated anit-BrdU antibodies. In addition, 7-Aminoactinomycin D (7-AAD), which is a DNA-labeling fluorescence dye, was used to determine cell position in the cell cycle by 7-AAD intensities. Cells in the G2/M phase will have a 7-AAD intensity  58 twice as much as that of cells in the G0/G1, whereas apoptotic cells (sub-G1) will have lower 7-AAD intensities.    2.15 Immunofluorescence (IF) Staining Skin biopsies and suspension cells were assessed by immunofluorescence (IF) staining. For skin biopsies, cryosections of MF lesions and control skin were fixed with 4% paraformaldehyde, permeabilized, and blocked. Then the slides were incubated with primary antibodies against TOX (rabbit polyclonal, Sigma Aldrich) and human CD4 (mouse monoclonal, Dako, Mississauga, ON, Canada) overnight at 4 °C, washed, and incubated with goat-anti-rabbit Alexa-594-conjugated secondary antibody (red; Life Technologies, Burlington, ON, Canada) and goat-anti-mouse Alexa-488-conjugated secondary antibody (green; Life Technologies). Finally, cell nuclei were counterstained with 4',6-Diamidino-2-phenylindole (DAPI). For CTCL cells and CD4+ T cells, 1 × 106 cells were pelleted, fixed, washed, and re-suspended in 50 μL PBS. Then 10 μL of the cell mixture was smeared onto one slide, blocked, and stained using the method as described above.  Slides were viewed under a Zeiss AxioVert 200M inverted fluorescence microscope (Carl Zeiss AG, Jena, Germany), and images were collected and processed using the Zeiss AxioVision 4.8 image acquisition and processing software (Carl Zeiss AG). Brightness and contrast were adjusted consistently across all images, as reported previously (Huang et al, 2015).    2.16 Animals and Tumor Formation Assay Non-obese diabetic (NOD)/severe-combined inmmunodeficiency (SCID) interleukin-2 receptor gamma chain deficiency (NSG) mice were bred and maintained at the British Columbia Cancer Research Center Animal Resource Centre. All procedures were performed according to the experimental protocol approved by the University of British Columbia  59 Animal Care Committee. NSG mice at 8 weeks of age were injected subcutaneously on both flanks with 1 × 106 transduced Hut78 or HH (CTR = 6; TOXsh1 = 6; TOXsh2 = 6). Due to rapid cell death induced by TOX knockdown, transduced SZ4 cells could not accumulate enough cells to be assayed in the same manner. Mice were closely monitored for local tumor formation 3 times a week. When tumor formed, the tumor length and width were measured with a glide caliper, and tumor volume was calculated using the formula (length [mm] × width [mm]2) /2 (Krejsgaard et al, 2010).  To examine the histopathological features, mouse tissues were fixed in 10% formalin, paraffin embedded, and stained with hematoxylin and eosin (H&E). Furthermore, immunohistochemistry (IHC) staining was performed on the mouse tissues using rabbit monoclonal anti-human CD3 antibody (SP7; Spring Bioscience, Pleasanton, CA) and UltraMap anti-rabbit horseradish peroxidase-conjugated secondary antibody (Ventana Medical Systems, Tucson, AZ). H&E and IHC staining was performed by Centre for Translational and Applied Genomics (Vancouver, BC, Canada). Histological images were acquired by a Zeiss Axioplan 2 Imaging Microscope (Carl Zeiss).  2.17 Microarray Analysis Total RNA was extracted from transduced CTCL samples (Hut78-CTR = 3, Hut78-TOXsh = 3, HH-CTR = 3, HH-TOXsh = 3) using TRIzol method (Invitrogen). A total of 100 ng RNA was reverse-transcribed into cDNA and linearly amplified by in vitro transcription using the Low RNA Input Linear Amplification Kit (Agilent Technologies, Mississauga, Canada). Each microarray was hybridized with each amplified cDNA labeled with Cy5 or Cy3. Hybridizations were performed on SurePrint G3 Human Gene Expression 8x60K v2 Microarray slides (G4858A-039494; Agilent Technologies), and the slides were scanned and  60 quantified. Red and green processed signals were then normalized and analyzed as previously described (Zhang et al, 2012). Microarray data were deposited at NCBI GEO database (accession number: GSE57122). Genomatix Pathway System (GePS, Genomatix software suite v3.0) and the Database for Annotation, Visualization and Integrated Discovery (DAVID, v6.7) software programs were applied to evaluate enriched biological processes (ranked by enrichment scores and Benjamini adjusted P values) in the differentially expressed genes upon TOX suppression (Huang da et al, 2009a; Huang da et al, 2009b). 2.18 Statistical Analysis Statistical analyses were performed using SPSS 14 (Chicago, IL), GraphPad Prism 5.00 (San Diego, CA), X-tile (New Haven, CT), and Microsoft Excel programs. P values < 0.05 were considered to be statistically significant. Two-tailed t tests were used to compare continuous variables. Data are shown as mean ± standard error of mean (SEM). Receiver operating characteristic (ROC) method was used to determine whether TOX transcript levels could differentiate CTCL from non-CTCL (BID and healthy participants). Disease progression and disease-specific mortality were evaluated using the Kaplan-Meier curve and log-rank test. Using X-tile software (Camp et al, 2004), the optimal cut-off points (with the lowest P value) of TOX mRNA expression level were determined to be 8.2 for MF, and 129.9 for SS. Therefore, in our analysis, TOX high and TOX low groups were defined by the TOX expression levels higher and lower than 8.2 and 129.9 for MF and SS, respectively.  Disease progression in MF patients was considered to be present when disease progressed to more advanced clinical stages and/or death due to MF (Litvinov et al, 2013). Individuals with multiple progressions (n > 1) were considered to have n progression events. Survival time was defined as the duration from date of presentation and sample collection to  61 death. A two-step approach was adopted when evaluating the association of TOX expression levels and disease outcome (progression and disease-specific mortality) in MF as reported (Huang et al, 2014). First, univariate analysis using the log-rank test was used to determine whether TOX mRNA levels correlated with disease outcome. Second, multivariate analyses using COX proportional hazards regression were used to evaluate the following prognostic factors: stage at diagnosis, TOX mRNA levels, age, and sex.      62 Chapter 3: The Potential of TOX as a Disease Marker for CTCL  3.1 Background and Rationale  CTCL is a highly heterogeneous disease with a wide variety of clinical and histopathological features. The two most common subtypes of CTCL are MF and SS (Bradford et al, 2009). Classical MF presents with erythematous patches and plaques with scales, typically on sun-protected areas (trunk, buttocks, and proximal extremities) (Kim et al, 2005). Tumors and erythroderma are seen in advanced MF stages. SS is considered the leukemic stage of CTCL with circulating Sézary cells (malignant lymphocytes with hyper-convoluted, cerebriform nuclei) in the peripheral blood (Kim et al, 2005).  It is challenging to reach a diagnosis for MF, particularly in its early stage, which contains only a small percentage of malignant cells admixed with a large number of reactive CD4+ T cells, therefore lacking the fully-developed characteristics of classical MF. Further, early MF mimics clinical, histopathologic, and even molecular (including TCR clonality) features of common benign inflammatory skin conditions (Glusac, 2003; Keehn et al, 2007; Olsen et al, 2007; Pimpinelli et al, 2005). It is therefore common to have diagnostic delays in MF, which sometimes may exceed a decade (Arai et al, 1991). The diagnostic challenges are not unique to early stage MF. Patients with late stage diseases, such as SS and erythrodermic CTCL, are also difficult to diagnose since there are no characteristic positive identification markers for circulating CTCL cells.  Another challenge in managing CTCL patients stems from the difficulty to predict long term clinical outcome. MF typically has an indolent clinical course. However, approximately 9% of CTCL patients with limited patch and plaque disease, as well as 24% of patients with extensive patches or plaques, will develop end-stage disease, such as SS, which  63 has a high mortality (Kim et al, 1995; Kim et al, 1999). Identifying MF patients with high risk of disease progression or disease-specific mortality has valuable implications in that these patients may be managed more aggressively and monitored more frequently to slow or halt disease advancement. However, robust prognostic markers for CTCL are generally lacking, which hinders the proper prognostication.  In previous attempts to identify disease-specific markers in CTCL, our group performed microarray analysis in early MF, and discovered a set of 19 genes that were highly elevated in early MF skin lesions, but not in benign control skins. Among these, TOX appeared to be the most enriched gene (Zhang et al, 2012). IF and IHC staining in skin biopsies confirmed strong nuclear TOX staining in a proportion of CD4+ T cells in MF, but not in BID lesions. This observation was subsequently validated in early as well as more advanced MF in a small patient population (n = 15) (McGirt et al, 2014). It remains to be examined if TOX is also upregulated in SS, the advanced CTCL. Based on these findings, we hypothesized that TOX is aberrantly activated in the full spectrum of CTCL (from various stages of MF to SS), and that TOX may be used as a marker for CTCL.  3.2 Results 3.2.1 TOX Is Upregulated in Skin Biopsies from MF Patients of All Stages To investigate the presence and extent of TOX overexpression in the full spectrum of MF, we assembled 113 individuals with MF, 25 with BID, and 11 with healthy skin (HS). TOX mRNA expression levels (mean ± SEM) were measured by qPCR. BID and HS were combined to form the non-MF controls (3.58 ± 0.75) as their TOX expression levels were not significantly different (3.77 ± 0.94 versus 3.15 ± 1.24, P = 0.70). MF cohort 1 (n = 54) had much increased TOX mRNA levels (64.21 ± 18.69), compared with the controls (P = 0.002).  64 Furthermore, an independent MF cohort 2 also showed much higher TOX mRNA level than that in the non-MF controls (20.64 ± 3.90 versus 3.58 ± 0.75, P < 0.0001) (Figure 3.1A).   The observed increase reflected TOX mRNA levels in a mixture of cells, and may be due to a simple accumulation of the CD4+ T cells in the MF lesional skin. To test this, we normalized the TOX mRNA levels to CD4 mRNA levels. Upon normalization, TOX/CD4 mRNA ratio remained highly increased in MF skin lesions, compared with non-MF controls (0.542 ± 0.182 versus 0.062 ± 0.014, P = 0.011) (Figure 3.1B). Next, we asked whether TOX levels were indicative of tumor burden in the skin. To test this, TOX levels were obtained in skin lesions with various morphologies, including patch (low tumor burden), plaque and tumor (high tumor burden). As shown in Figure 3.1C, thicker lesions (plaques and tumors) displayed higher TOX levels (74.91 ± 29.72 in plaques and 101.3 ± 37.23 in tumors) than those of the patches (18.31 ± 4.10, P = 0.018). Similar trend can be observed when TOX levels were normalized to CD4 levels, in that TOX/CD4 ratios in thicker lesions (0.348 ± 0.065 in plaques and 1.870 ± 1.145 in tumors) were higher than those in thinner lesions (0.216 ± 0.055 in patches, P = 0.088) (Figure 3.1D).  In addition to mRNA levels, we evaluated the TOX protein levels in MF skin biopsies by IF staining, using BID lesions as controls. While numerous CD4+ T cells can be seen in BID lesions, rarely was TOX detectable. In contrast, we observed strong TOX staining in MF biopsies, primarily in the nuclei of the CD4+ T lymphocytes. Furthermore, thicker MF skin lesions (plaques and tumors) demonstrated stronger TOX signals than patch lesions, which is in line with the mRNA expression results (Figure 3.2). 65  HSBIDMF cohort 1MF cohort 20501001505001000*****NSNSTOX mRNA levelHSBIDMF cohort*NSTOX/CD4 ratioAPatchPlaqueTumor0501001502005001000*NSTOX mRNA levelC DPatchPlaqueTumor0123810P = 0.088NSTOX/CD4 ratioB Figure 3.1 TOX mRNA levels are increased in MF skin biopsies. (A) TOX mRNA in MF samples compared with BID and HS. **, P = 0.002, ***, P < 0.0001. (B) The ratios of TOX mRNA/CD4 mRNA in samples of MF, BID and HS. *, P = 0.011. (C) TOX mRNA in different types of MF lesions. *, P = 0.018. (D) The ratios of TOX mRNA/CD4 mRNA in different types of MF lesions. Horizontal bars denote the mean and SEM for each sample type analyzed. Two tailed t tests were used for comparison. NS=not significant. 66  TOX CD4 TOX + CD4+DAPIMF PatchMF PlaqueMFTumorBID Figure 3.2 Ectopic TOX protein is detected in CD4+ T lymphocytes in MF skin lesions, but absent in BID. BID (shown here is chronic dermatitis) and MF skin biopsies were stained with antibodies against TOX (red, Alexa Fluor® 594) and CD4 (green, Alexa Fluor® 488). DAPI was used to stain the nuclei of cells. Dotted lines indicate the basement membranes in the skin. Insets: magnification from representative areas. Bars = 20 μm.   67  3.2.2 TOX Is Upregulated in Primary Sézary Cells and CTCL Cell Lines We next moved on to characterize the expression of TOX in SS patients. By qPCR, the TOX mRNA levels were quantified in the CD4+CD7- T cells purified from the peripheral blood of 12 SS patients, CD4+ T cells from 18 BID patients, and 9 healthy participants (HP). BID and HP CD4+ T cells expressed similar levels of TOX (33.66 ± 3.17 versus 44.75 ± 7.64, P = 0.21), therefore they were combined as the non-SS controls (37.36 ± 3.38). Compared with non-SS, SS CD4+CD7- T cells expressed much increased TOX levels (172.70 ± 45.20, P = 0.012) (Figure 3.3A). As TCR clonality has been correlated with worse clinical outcome in SS patients (Assaf et al, 2005; Fierro et al, 2010), we assessed TOX expression levels in SS patients with and without TCR clonality. As shown in Figure 3.3B, among the 12 SS patients, SS with TCR clonality (n = 7, 236.5 ± 67.03) expressed higher TOX mRNA levels, compared with those without TCR clonality (n = 5, 83.37 ± 25.07, P = 0.070). Although this trend did not reach statistical significance, it may with a larger sample size.     68  Clonal Non-Clonal0100200300400500600P = 0.070TOX mRNA level per1000GAPDHSS HP PSO ROS VT0100200300400500600P = 0.012TOX mRNA level per1000GAPDHA B  Figure 3.3 TOX mRNA levels are increased in peripheral blood CD4+ T cells from SS patients.  (A) TOX mRNA levels in peripheral blood CD4+ T cells from SS (n = 12), healthy participant (HP, n = 9), and BID (PSO, psoriasis, n = 8; ROS, rosacea, n = 5; VT, vitiligo, n = 5). (B) TOX mRNA levels in CD4+ T cells from SS patients with clonality (n = 7), compared with SS without clonality (n = 5).     69  To determine if the high TOX expression was retained in CTCL cell lines, TOX protein levels were examined in a panel of CTCL cell lines by Western blotting. In line with previous report, we detected two bands of TOX, 57 kDa and 63 kDa, which represent the predicted and the post-translationally modified form of TOX (Wilkinson et al, 2002). We found that 10 out of 11 CTCL cell lines expressed increased TOX protein, compared to normal CD4+ T cells, and the TOX quantities in the majority of CTCL cell lines were comparable to that of clinical Sézary cells (SS5) (Figure 3.4A and B). This finding demonstrated that highly deregulated expression of TOX was found not only in MF skin biopsy specimens as we previously reported (Zhang et al, 2012) but also in Sézary syndrome, the advanced stage of CTCL. Of note, Hut102, derived from MF (early CTCL), expressed much lower TOX protein compared to the more advanced CTCL cell lines (Figure 3.4B), derived from either SS (Hut78, SZ4) or non-MF/SS aggressive CTCL (HH), suggesting that overexpressed TOX may contribute to the disease progression. FACS analysis further confirmed the TOX increase in the peripheral CD4+ T cells from a patient with SS (SS12) and in CTCL cell lines (Hut78, HH, and SZ4), while normal CD4+ T cells displayed negative TOX staining (Figure 3.4C). Finally, IF staining using TOX and CD4 antibodies demonstrated strong nuclear TOX staining, in CTCL cell lines (Hut78, HH, and SZ4). In contrast, normal CD4+ T cells showed no obvious TOX staining (Figure 3.5). These findings indicate that CTCL cell lines that retain high TOX levels may be used as appropriate in vitro models to study the biological function of TOX in CTCL.   70  ATOXActin62 kb57 kb42 kbB CActinTOX 62 kb57 kb42 kbTOXCountNegative controlNormal CD4SS12 CD4Hut78HHSZ4 Figure 3.4 TOX protein levels are increased in SS CD4+ T cells and CTCL cell lines. (A) Western blotting showed marked TOX overexpression in the majority of CTCL cell lines, compared to peripheral CD4+ T cells from an HP (Ctr). (B) Western blotting demonstrated TOX increase in 4 CTCL cell lines, as well as CD4+ T cells from the peripheral blood of a patient with SS (SS5), compared to peripheral CD4+ T cells from an HP (Ctr 1) or a patient with BID (Ctr 2). (C) FACS analysis showed elevated TOX intensities in the peripheral CD4+ T cells from a patient with SS (SS12) and in the CTCL cell lines (Hut78, HH, and SZ4), whereas normal CD4+ T cells displayed negative TOX staining. The fluorescence-minus-one control was used as a negative control.    71  Hut78 HH SZ4Normal CD4+ T cellsDAPITOXCD4Merged  Figure 3.5 TOX expression is elevated in multiple CTCL cell lines. Immunoflurescence (IF) staining using TOX (red) and CD4 (green) antibodies showed strong nuclear TOX staining in CTCL cells (Hut78, HH, and SZ4), but not in normal CD4+ T cells. DAPI was used to stain the nuclei. For each cell line, 1 million cells were mounted onto the slide and subjected to staining. Bars represent 25uM.     72 3.2.3 Increased TOX Levels Differentiate CTCL from Benign Inflammatory Dermatoses Given the marked increase of TOX in both MF and SS, we asked whether TOX levels differentiate CTCL patients from patients with non-CTCL (BID and HP) subjects. In all MF patients (n = 113), ROC curve analysis (Figure 3.6A) showed an area under the curve (AUC) value of 0.87 (95% confidence interval 0.8–0.94, P < 0.0001), indicating that TOX mRNA levels had good discriminative power for MF as a whole. ROC curve also showed the tradeoff between sensitivity and specificity at different cutpoints. At a cut-off point of 2.99, TOX mRNA levels had a specificity of 90.3% and a sensitivity of 75.0% for MF.  We next evaluated whether TOX mRNA levels could help with the discrimination between early stage MF and non-MF controls, given that it is most challenging to accurately diagnose early MF in the clinical setting. ROC curve analysis in 73 cases of stage-I MF demonstrated good discriminatory power with an AUC value of 0.85 (95% confidence interval 0.77-0.92, P < 0.0001, Figure 3.6B). At a cut-off point of 2.99, TOX mRNA levels had a specificity of 86.3% and a sensitivity of 75.0% for stage-I MF.  Last, we applied ROC curve analysis to assess whether TOX mRNA levels may differentiate SS from non-SS control CD4+ T cells. An AUC of 0.82 (95% confidence interval 0.62-1.03, P = 0.001, Figure 3.6C) was obtained, suggesting that TOX mRNA levels may help improve the diagnosis of SS. When the cut-off point was set at 52.77, the specificity and sensitivity for TOX mRNA to differentiate SS were 85.2% and 83.3% respectively.   73  A BC0 20 40 60 80 100020406080100AUC = 0.82P = 0.001SS (n = 12)100% - Specificity%Sensitivity%0 20 40 60 80 100020406080100AUC = 0.85P < 0.0001Stage-I MF (n = 73)100% - Specificity%Sensitivity%0 20 40 60 80 100020406080100AUC = 0.87P < 0.0001100% - Specificity%Sensitivity%MF (n =113) Figure 3.6 Increased TOX mRNA levels differentiate CTCL from non-CTCL.  (A) ROC curve analysis on MF as a whole (n = 113) and non-MF (n = 36) skin biopsies. (B) ROC curve analysis on stage-I MF (n = 73) and non-MF (n = 36) skin biopsies. (C) ROC curve analysis on Sézary cells (n = 12) and non-SS control CD4+ T cells (n = 27). AUC= area under the curve.    74 3.2.4 High TOX Levels Correlate with Increased Risk of Disease Progression and Disease-specific Mortality in MF Patients Having shown that TOX may be used as a potential diagnostic marker for CTCL, we speculated that varied TOX mRNA levels may also be correlated with divergent clinical outcome. For this purpose, we performed Kaplan-Meier curve in MF cohort 2 patients, for whom long-term (up to 6 years; median follow up time = 45 months) follow-up data were available. We found that high TOX mRNA levels in skin biopsies were associated with increased probability of disease progressing into a higher clinical stage during the follow up period (P = 0.003, Figure 3.7A), indicating its association with disease aggressiveness. In contrast, cases with low TOX expression levels had very low tendency to progress clinically during the follow up time. We next applied multivariate analysis to assess if TOX levels remained independent outcome indicator after adjusting other factors, including clinical stage, age, and sex. TOX expression remained significantly associated with disease progression (hazard ratio (HR) = 2.62, P = 0.028, Table 3.1).   Furthermore, increased disease-specific mortality was observed in cases with higher TOX mRNA levels (P = 0.008, Figure 3.7B). Multivariate analysis defined high TOX as an independent predictor of worse survival (HR = 6.16, P = 0.027, Table 3.1) after adjusting for other variables, including clinical stage, age, and sex. Similar to prior reports (Agar et al, 2010; Imam et al, 2013), clinical stage was shown to be an independent predictor of prognosis in our study.  Taken together, these findings supported that high TOX levels could help define the subset of MF patients with increased risk of disease progression and disease-specific mortality.   75  A B0 20 40 60 80020406080100TOX high [38]TOX low [32]P = 0.003Time (months)Progression-free status (%)0 20 40 60 80406080100TOX high [28]TOX low [31]P = 0.008Time (months)Disease-specific survival (%)  Figure 3.7 Higher TOX mRNA levels correlate with worse clinical outcome in MF. Kaplan-Meier curves showed the relationship between TOX mRNA levels and (A) disease progression in MF patients; (B) disease-specific mortality in MF patients. Log-rank tests were used.   Table 3.1 Multivariate analyses of disease progression and disease-specific survival in mycosis fungoides subjects using COX proportional hazards model Prognostic Factor  Progression  Disease-Specific  Survival HR 95.0% CI P  HR 95.0% CI P TOX  2.62 1.1 to 6.2 0.028  6.16 1.2 to 30.8 0.027 Stage  3.57 1.6 to 7.9 0.002  5.31 1.2 to 23.5 0.028 Age  1.02 1.0 to 1.0 0.246  1.00 1.0 to 1.0 0.989 Sex  1.02 0.5 to 2.3 0.960  0.55 0.1 to 2.1 0.386  Abbreviations: HR = hazard ratio; CI = confidence interval  76  3.2.5 High TOX Levels Correlate with Increased Disease-specific Mortality in SS Next, we evaluated if high TOX levels influence the clinical outcome of SS patients. Kaplan-Meier curve revealed a strong correlation between higher TOX mRNA levels and increased SS-related death (P = 0.039, log-rank test hazard ratio = 5.68, Figure 3.8). The median survival time in SS patients with higher TOX mRNA levels (> 129.9, n = 6) was 27 months, which was much shorter than that of SS with lower TOX mRNA levels (<129.9, n = 6, 120 months). This finding suggested that increased TOX levels may serve as a prognostic predictor for SS.   77   Figure 3.8 Higher TOX mRNA levels correlate with increased disease-specific mortality in SS Kaplan-Meier curve showed the relationship between TOX mRNA levels and disease-specific mortality in SS patients (n = 12). Log-rank test was used.  0 50 100 150020406080100TOX highTOX lowP = 0.039Time (months)Disease-specific survival (%) 78  3.3 Discussion In the current study, we characterized the expression levels of TOX in a panel of MF and SS patients, and evaluated its potential as a disease marker to aid CTCL diagnosis and prognostication. The first evidence of abnormal TOX enrichment in MF came from our previous transcriptome analysis on early MF, which demonstrated that MF CD4+ T cells possessed high TOX, at both mRNA and protein levels, compared to the CD4+ T cells in BID (Zhang et al, 2012). In the research performed as part of this thesis research, I found that aberrant expression of TOX is a feature persisting in the entire spectrum of CTCL, including advanced stages of MF and SS.   TOX is not induced by TCR signaling in mature T cells (Wilkinson et al, 2002). Hence its upregulation in CTCL cells is unlikely to be a result of T cell activation. Rather, this ectopic activation of TOX in MF and SS led us to propose that TOX is a disease marker with diagnostic and/or prognostic value for CTCL.   We tested this hypothesis and provided evidence for its potential usefulness as a diagnostic and prognostic marker for CTCL. Our data not only confirmed enhanced TOX expression in early MF, but also showed that TOX aberrant activation is a common feature shared by the entire spectrum of MF and SS. Moreover, high TOX mRNA levels effectively differentiate MF (for both MF as a whole or stage-I MF alone) from non-MF controls, and SS from non-SS controls, with high sensitivity and specificity. High TOX mRNA levels also demonstrated prognostic value for MF in that they were strongly associated with increased risk of disease progression and disease-specific mortality in MF patients. Furthermore, we detected higher TOX levels in Sézary cells with positive TCR clonality, as compared with those without TCR clonality. Since TCR clonality, when present, is closely correlated with  79 worse clinical outcome of SS patients (Assaf et al, 2005; Fierro et al, 2010), we wondered whether TOX transcript levels are linked to disease outcome. Indeed, higher TOX expression levels in SS patients correlated with increased risk of SS-related death. MF clinically resembles the far more common, but clinically benign skin inflammatory diseases such as chronic dermatitis, psoriasis, and cutaneous reactions to drugs. Differentiating MF from these benign conditions is difficult, largely due to the lack of well-defined molecular markers in the clinical setting. Although CD2, CD3, CD5, and CD7 deficiency is included in ISCL criteria to define early MF, the loss of CD2, CD3, and/or CD5 in T cells is only 10% sensitive, despite its 100% specificity. CD7 deficiency is about 40% sensitive and 80% specific in general (Pimpinelli et al, 2005). Therefore better markers with higher sensitivity and specificity are needed. Although a small number of markers have been reported for MF skin biopsies, including loss of CD13, ectopic expression of BLK gene, microRNAs (including miR-155, miR-203 and miR-205) (Bernier et al, 2007; Krejsgaard et al, 2009; Ralfkiaer et al, 2011), BCL7A (Carbone et al, 2008; Litvinov et al, 2013), AHI1 (Litvinov et al, 2012), and CD158K/KIR3DL2 in transformed advanced MF (Ortonne et al, 2012), few of these markers were tested in multicenter studies, or used in a clinical setting.  Diagnosing SS could also be challenging, due to the lack of robust identification markers. While several novel biomarkers have been proposed in previous reports, there is a lack of consensus, which may partially be explained by the small sample sizes, various platforms used, and different type of samples (Wong et al, 2011). Some of the biomarkers for SS include increased PLS3 (Su et al, 2003), DNM3, NEDD4L (Booken et al, 2008),  KIR3DL2 (Poszepczynska-Guigne et al, 2004), EPHA4, TWIST1 (van Doorn et al, 2004), JUNB (Mao et al, 2004), TNFSF11 (Booken et al, 2008); decreased SATB1 (Wang et al,  80 2011), DPP4, STAT4, and TGFBR2 (van Doorn et al, 2004). These markers will need to be tested in larger samples and multicentre setting before their clinical utility can be determined.  The prognostic indicators for CTCL are mainly clinical (stage, extracutaneous presentations, and age) or histopathological factors (large cell transformation, presence of CD8+ T cells) (Kim et al, 2005). There is currently no molecular prognostic marker for CTCL in clinical use, which hinders the establishment of accurate prognostication for CTCL patients.  In light of these, the current multi-center study consisting of patients with diverse ethnic origins demonstrated considerable potential of TOX to improve diagnosis and management for CTCL. TOX levels showed good discriminatory power for both MF and SS, which may be helpful in defining CTCL even in its earliest stage when other diagnostic criteria are not met. Kaplan-Meier curve analysis demonstrated that MF lesions, including early MF lesions, that contain no or low TOX mRNA expression had little tendency of disease progression or MF-related death. TOX as a marker can identify these low-risk patients for whom conservative management perhaps would be adequate rather than being subjected to more toxic treatments, such as topical nitrogen mustard, or carmustine. In contrast, MF patients with high levels of TOX expression may benefit from early and more aggressive treatment to prevent disease progression and to reduce MF-related mortality. Similarly, SS patients with high TOX levels shall be considered to be at high risk of SS-related death, and may benefit for more frequent monitoring and more aggressive management.  Despite the predictive potential of TOX for CTCL demonstrated in this study, caution needs to be taken while interpreting its clinical usefulness. Due to the rarity of this disease,  81 only a moderate sample size of 113 for MF and a small sample size of 12 for SS can be reached from three study centres from three countries. Additional confirmation in other centres is warranted to evaluate the true clinical relevance of TOX for CTCL diagnosis and prognostication.  Notwithstanding this limitation, the multicenter nature of our study underscores the consistency of TOX up-regulation in patients from diverse ethnical and geographic backgrounds. Furthermore, TOX as a marker is highly robust and user-friendly, since it can be detected by a number of routine dermatopathologic tools, such as qPCR, IF, and IHC staining.  Given TOX’s critical regulatory role in CD4+ T cell development and its aberrant overexpression in the majority of CTCL, it is highly likely that TOX activation plays a pathogenic role in CTCL. Further studies are therefore warranted to evaluate if TOX ectopic expression contributes to the development of CTCL. If proven to be true, TOX may emerge as a novel therapeutic target for MF in the future. It is noteworthy that TOX overexpression was retained in the majority of CTCL cell lines, including Hut78, HH, and SZ4, indicating that these cell lines could serve as appropriate in vitro models to study the biological functions of TOX in CTCL.  In summary, TOX ectopic expression is readily and frequently detected in the malignant CD4+ T cells in MF skin biopsies and Sézary cells, including the most challenging early MF. Moreover, increased TOX expression levels define a group of CTCL patients with worse long term clinical outcome. Therefore, characterization of TOX expression status of CTCL patients may be valuable not only for diagnostic confirmation but also for guiding CTCL management in the future.  82 Chapter 4: The Role of TOX in the Development of CTCL  4.1 Background and Rationale Although genetic studies have yielded some valuable insights, the exact pathogenesis of CTCL remains poorly understood. Recent studies into the disease mechanism of CTCL have suggested involvement of apoptosis resistance (Sors et al, 2006; Wang et al, 2011; Wu et al, 2009) and uncontrolled cell cycle progression (Mao et al, 2006; Scarisbrick et al, 2002). Our recent study identified marked TOX overexpression in CTCL cells, compared to CD4+ T cells in BID lesions. In Chapter 3, we further confirmed this feature in the full spectrum of MF and SS. The presence of TOX activation in the earliest stage of MF suggested that TOX may be involved in disease initiation, and the observation that TOX expression levels paralleled tumor burden (thicker lesion contained more TOX) pointed to the possibility that TOX may contribute to disease progression as well.  TOX is a critical player in thymocyte development. Deficiency of TOX in mouse studies showed profound disruption of normal CD4+ T-cell development, and absence of lymph nodes and Peyer’s patches in the small intestine (Aliahmad et al, 2010; Aliahmad & Kaye, 2008). Given TOX’s important functions in T cell development, we speculated that TOX plays a pathogenic role in the disease development of CTCL, a lymphoproliferative disorder primarily of CD4+ T cell origin. In the current study, we performed comprehensive in vitro and in vivo studies to characterize the biological significance of TOX aberration activation in CTCL oncogenesis.      83 4.2 Results 4.2.1 Reduced TOX in CTCL Cells Results in Marked Reduction in Cellular Proliferation and Colony Formation in Vitro To investigate whether knockdown of TOX expression affected cell growth of CTCL cells, three CTCL cells lines (Hut78, HH, and SZ4) were transduced with lentiviruses containing either a non-targeting control sequences (CTR) or each of two shRNA sequences against human TOX. Knockdown of TOX (>90%) by two shRNA constructs (sh-1 and sh-2) was confirmed in all three CTCL cell lines by Western blotting analysis (Figure 4.1A). Viability assay (4 day duration, Figure 4.1B) showed that stable suppression of TOX resulted in significantly reduced cell growth in all three cell lines, compared with control cells (2~4-fold, P < 0.001). This growth inhibition was most dramatic in SZ4 cells, which rapidly died with TOX depletion.  In addition to short term liquid culture, we subjected the transduced cells to semi-solid long term culture to assess the effect of TOX suppression on the colony-forming ability of CTCL cells. TOX suppression resulted in marked reduction in CFC output in all three cell lines. In addition to reduction in colony numbers, the colonies generated by TOX-sh cells were much smaller and more dispersed than those generated by the control cells (Figure 4.2).   84  0 1 2 3 4020406080CTRTOX-sh***SZ4Day0 1 2 3 4050100150200250***Hut78Daycell number/*10^40 1 2 3 4010203040506070***HHDayTOXHH SZ4Hut78ActinCTR   sh-1   sh-2ABCTR   sh-1   sh-2 CTR   sh-1   sh-262 5742kDa Figure 4.1 TOX inhibition confers growth disadvantage to CTCL cells. (A) TOX knockdown by 2 shRNAs (sh-1 and sh-2, both specifically target TOX gene), compared to control (CTR) CTCL (Hut78, HH, and SZ4) cells transduced by a non-targeting shRNA. Protein lysates were probed with antibodies against TOX after transduced cells were selected by puromycin (1 μg/mL) for 5 days. Actin was included as a protein loading control. (B) TOX-sh cells (transduced by sh-1 or sh-2) generated significantly reduced viable cells over a period of 4 days for all three CTCL cell lines (Hut78, HH, and SZ4), compared to CTR cells. On day 0, a total of 2.5 × 105 cells were plated in 2 mL of full RPMI media, and viable cell numbers were determined every 24 hours by the trypan blue exclusion method. ***, P < 0.001. Error bars denote SEM. Data shown are representative of at least 3 independent experiments.   85  CTR TOX-sh0102030405060***HHCTR TOX-sh020406080100120**Hut78Colony number/300input cellsCTR TOX-sh03691215BigMediumSmall***SZ4 Figure 4.2 TOX suppression decreases the clonogenic ability of CTCL cells in long term culture.  TOX-sh CTCL cells (Hut78, HH, and SZ4) cells generated fewer and smaller colonies in 3-dimentional culture, compared to CTR cells. Three hundred cells were plated in 1.2 mL methycellulose media with 1 μg/mL puromycin. On day 12, the colony numbers and morphology were examined and recorded. Numbers of colonies of different sizes (big, medium, and small) were depicted. Photographs showing representative colonies were also shown. Original magnification × 10. **, P < 0.01 and ***, P < 0.001. Error bars indicate SEM. Data shown are from 3 independent experiments, and duplicates were performed for each experiment.     86  4.2.2 TOX Inhibition Abolishes or Halts Tumor Formation by CTCL Cells in Vivo To investigate whether deregulation of TOX in CTCL cells contributes to their ability to induce tumor in vivo, we injected transduced CTCL cells (Hut78 and HH) into NSG mice and monitored local tumor formation (TOX knockdown resulted in rapid death of SZ4 cells, which could not be assayed in the same fashion). Mice injected with Hut78 CTR or HH CTR (n = 6) cells started to form local tumors within 13 days. The tumors grew progressively and reached the maximum allowed tumor volume of 1700 mm3 (by 23 days, and 30 days respectively). In contrast, only 1 out of 6 mice injected with Hut78 TOX-sh cells, and 5 out of 6 mice injected with HH TOX-sh cells formed tumors, which were significantly smaller than the control tumors (Figure 4.3A and Bi-ii). Local tumors formed by CTR cells showed abnormal mitosis and more extensive destruction of subcutaneous tissues, compared to TOX-sh cells (H&E staining, Figure 4.3Biii). The infiltrating tumor cells were positive for human CD3, confirming their origin in human helper T cells (Figure 4.3Biv). These findings indicated that stable suppression of TOX expression profoundly impaired the oncogenic ability of CTCL cells in vivo.      87 Hut780 7 13 16 18 21 2305001000150020002500**Days post injectionTumor Volume/mm3HH0 7 13 16 18 21 23 25 28 3005001000150020002500CTRTOX-sh*Days post injectionTumor Volume/mm3CTR TOX-sh CTRHut78 HHiii. H&E iv. CD3ABii. i .TOX-sh  Figure 4.3 TOX inhibition impairs the tumor-forming ability of CTCL cells in vivo.  88 (A) Tumor volume comparison between mice injected with 1 million TOX-sh cells (n = 12) and control cells (n = 6) of Hut78 and HH. Mice were monitored for local tumor formation three times a week, and tumor sizes were measured with a glide caliper. Tumor volume was calculated using the formula V = (length [mm] × width [mm]2) ÷ 2. *, P < 0.05, **, P < 0.01. Error bars indicate SEM. (B) Gross appearances of mice representative of each injection group (i), appearances of tumor or normal skin (in mice injected with Hut78 TOX-sh, which did not form tumor) from representative mice (ii), H&E staining of tissue sections from corresponding tumors or normal skin (iii), and immunohistochemistry staining of human CD3 from corresponding tumors or normal skin (iv). Bars represent 1 cm (Bi-ii), 200 μm (Biii-iv), and 25 μm (Biii-iv insets).   89  4.2.3 Reduced TOX Sensitizes CTCL Cells to Apoptosis  Next we sought to elucidate the mechanisms through which TOX silencing exerts its anti-proliferative effect. The growth inhibition effect of TOX could be due to sensitization of CTCL cells to apoptosis, cell cycle arrest, or both. Resistance to apoptosis has been recognized by one of the major defects in CTCL cells (Contassot et al, 2008; Dereure et al, 2000b; Klemke et al, 2009; Wu et al, 2009). We first assessed whether the apoptosis machinery was affected. With TOX knockdown, transduced CTCL cells had increased apoptotic cells demonstrated by annexinV positivity (Figure 4.4A and B). We further examined the caspase (full-length and cleaved) levels in CTCL cells before and after TOX suppression. On the molecular level, TOX knockdown led to activation of caspase 9 and caspase 3, which are involved in apoptosis initiation and execution in intrinsic apoptosis pathway (Figure 4.4C). The activation was more obvious in HH and SZ4 cells, consistent with the more dramatic apoptosis sensitization phenotype.  90  BAAnnexin VPIHut78HHCTR TOX-shSZ4Hut78 HH SZ401020304050CTRTOX-sh** ** **Apoptotic cellpercentage (%)CTOXActinHH SZ4Hut78Full Cas3CleavedCas 3Full Cas9CleavedCas 962 5742473735351719kDaCTR  sh-1 sh-2 CTR  sh-1 sh-2 CTR  sh-1 sh-22.12 2.0694.2 0.79100  101 102 103 104    100101102103104%%%%1.66 6.0788.4 3.848.61 7.0281.8 2.577.75 24.657.0 10.60.59 3.3271.8 24.30.6 9.1444.9 45.4Early apoptoticLate apoptotic  Figure 4.4 TOX suppression leads to increased apoptosis and caspase activation in CTCL cells.  (A) FACS analysis of apoptotic (annexin V-positive) cell populations in TOX-sh CTCL cells (Hut78, HH, and SZ4), compared to control (CTR) cells. (B) Percentages of apoptotic cells in TOX-sh CTCL cells (Hut78, HH, and SZ4), compared to CTR cells. **, P < 0.01. Error bars indicate SEM. Data depicted are representative of at least 4 independent experiments. (C) Comparison of caspase 9 and caspase 3 levels (both full-length and cleaved) in TOX-sh (sh-1 and sh-2) and CTR CTCL cells. Cas, caspase; PI, propidium iodide.  91  4.2.4 Reduced TOX Causes Cell-cycle Progression Disruption in CTCL Cells We next assessed cell cycle progression, another key process governing cellular growth. Abnormal cell cycle progression has been implicated in the pathogenesis of CTCL (Gallardo et al, 2004; Mao et al, 2006). We subjected TOX-suppressed and control CTCL cells to BrdU and 7-AAD staining, which displayed the cell distribution in different cell cycle phases. BrdU positive population denotes cells in the S phase, which actively take in BrdU for DNA synthesis. BrdU negative populations include cells in the sub-G1, G0/G1, and G2M phases. While cells in G2M phase have twice as much DNA quantities (measured by 7-AAD intensities) as those in the G0/G1 phase, cells in the sub-G1 phase possess less DNA content than those in the G0/G1 phase. In all three cell lines, TOX inhibition led to increased number of cells in the sub-G1 (apoptotic) and G0/G1 stages, with a concomitant decrease of cells in the S and G2M stages (Figure 4.5), indicating that blocking of TOX markedly affected cell cycle control at both G1-S and S-G2M transitions.   92  CTR TOX-sh7-AADBrdUAHut78HHSZ4Sub-G1 G0/G1 S G2M0102030405060* *** * ***Percentage (%)Sub-G1 G0/G1 S G2M0102030405060* ** * *CTRTOX shPercentage (%)Sub-G1 G0/G1 S G2M0102030405060* * * ***Percentage (%)BHut78HHSZ4S(%)G2M(%)G0-G1(%)Sub-G1(%)01021031041050  50K100K150K 200K   37.40.867.4350.728.67.814.7555.638.72.7111.341.  Figure 4.5 TOX suppression leads to cell cycle arrest. (A) FACS analysis of cell proliferation by BrdU incorporation in TOX-sh Hut78, HH, and SZ4 cells compared to control (CTR) cells. For BrdU incorporation assay, transduced Hut78 cells were pulsed with 10 μM BrdU for 45 minutes, whereas transduced HH and SZ4 cells were pulsed for 2 hours. (B) Cell cycle distribution in TOX-sh cells compared to CTR cells. *P < 0.05, **P < 0.01, and ***P < 0.001. Error bars indicate SEM. Data depicted are representative of at least 3 independent experiments.   93  4.2.5 TOX Reduction Restores Cell Cycle Regulators, CDKN1B and CDKN1C In light of the marked restoration of cell cycle progression by TOX suppression, we asked whether key cell cycle regulators were affected by TOX. We evaluated the expression changes of several critical cell cycle regulators, including CDK2, CDK4, CDKN1A, CDKN1B, and CNKN1C. TOX suppression increased the expression of CDKN1B and CDKN1C in all three cell lines (Figure 4.6A), without consistently changing the levels of CDK2, CDK4, and CDKN1A (data not shown). This was confirmed by Western blotting, which showed that p27 protein, encoded by CDKN1B, and p57 protein, encoded by CDKN1C, were also upregulated after TOX knockdown (Figure 4.6B).   94  AHut78 HH SZ40123456 ** ** **CDKN1BExpression levelrelative to CTRHut78 HH SZ40123456 *CDKN1C* *CTRTOX-shBp27p57ActinHH SZ4Hut78kDa27 5742CTR sh-1 sh-2 CTR sh-1 sh-2 CTR sh-1 sh-242Actin  Figure 4.6 TOX suppression leads to elevated cell cycle repressors. (A) CDKN1B and CDKN1C transcript levels in TOX-sh and CTR. qPCR was performed using primers specific for CDKN1B, CDKN1C, and GAPDH mRNA. The level of CDKN1B and CDKN1C mRNA was normalized to that of GAPDH and is depicted as the fold change compared to CTR cells. *P < 0.05, **P < 0.01, and ***P < 0.001. Error bars indicate SEM. Data depicted are representative of at least 3 independent experiments. (B) p27 and p57 protein levels in TOX-sh and CTR cells in three CTCL cell lines (Hut78, HH, and SZ4).   95  4.2.6 Inhibition of CDKN1B or CDKN1C in TOX-suppressed CTCL Cells Partially Rescues the Growth Inhibition by TOX Suppression To examine whether the enhanced p27 and p57 proteins are responsible for the growth arrest in CTCL cells upon TOX suppression, we performed additional stable knockdown of CDKN1B and CDKN1C in TOX-suppressed CTCL cells (Figure 4.7A and B). Due to cytotoxicity of second virus infection in SZ4 TOX-sh cells, which were rapidly fatal, we were only able to obtain double knockdown in Hut78 and HH cells.  Stable knockdown of either CDKN1B or CDKN1C effectively reversed the proliferation suppression by TOX knockdown in Hut78 TOX-sh (Figure 4.7C) and HH TOX-sh cells (Figure 4.7D). However, neither CDKN1B nor CDKN1C knockdown was able to fully reverse the TOX suppression phenotypes. These results indicate that p27 and p57 are responsible, at least in part, for the growth inhibitory effect of TOX suppression in CTCL cells.  96  B27 42 HH+       +       +– +       –– – +57 42 kDaHH+       +       +– +       –– – +CAp27ActinHut78TOXshCDKN1Bsh-1CDKN1Bsh-2+       +       +– +       –– – +p57ActinHut78TOXshCDKN1Bsh-1CDKN1Bsh-2+       +       +– +       –– – +Hut780 1 2 3 4020406080100*CTRTOX-sh+p57-shTOX-sh+p27-shTOX-sh**Daycell number/*10^4D HH0 1 2 3 401020304050*CTRTOX-sh+p57-shTOX-sh+p27-shTOX-sh***Daycell number/*10^4kDa Figure 4.7 Knockdown of CDKN1B or CDKN1C reverses the growth inhibition of TOX-sh cells.  (A) CDKN1B knockdown by 2 shRNAs (sh-1 and sh-2) in addition to TOX knockdown in Hut78 and HH cells. Transduced cells were selected by hygromycin (750 μg/mL for Hut78 and 200 μg/mL for HH) for 7 days before the protein lysates were probed with antibodies against p27 and actin proteins. (B) CDKN1C knockdown in addition to TOX knockdown in Hut78 and HH cells. Transduced cells were selected by hygromycin for 7 days before western blot analysis was performed to examine p57 protein levels. (C) Cosilencing of TOX with CDKN1B or CDKN1C led to increased proliferative rate in Hut78 cells. A total of 1.5 × 105 cells were cultured in 2 mL of full RPMI media for 4 days, and viable cells were  97 determined each day by the trypan blue exclusion method. (D) Cosilencing of TOX with CDKN1B or CDKN1C led to increased proliferative rate in HH cells. Experiments were conducted in the same way as for Hut78 cells. *P < 0.05, **P < 0.01, and ***P < 0.001. Error bars indicate SEM. Data depicted are representative of at least 3 independent experiments.  98  4.2.7 Gene Expression Profiling in TOX-suppressed CTCL Cells The partial rescue of growth disadvantage yielded by p27 and p57 led us to reason that there are additional mediators downstream of TOX that are also responsible for its effect in CTCL.  Since TOX likely acts as an architectural factor capable of modulating chromatin structure (Wilkinson et al, 2002), we hypothesized that TOX exerted its oncogenic effect in CTCL by transcriptionally regulating, either directly or indirectly, a network of downstream molecules involved in key cellular processes, such as cell cycle and apoptosis. In order to identify these molecules, we performed microarray analysis in RNA samples from TOX-sh cells (n = 3) and CTR cells (n = 3) of both Hut78 and HH cell lines. A total of 22 differentially expressed genes, including several candidate tumor suppressors (SMAD3, FOXO3, HBP1, THAP1, and MED13L) (Angus & Nevins, 2012; Karube et al, 2011; Letterio, 2005; Paulson et al, 2007; Yao et al, 2005), were identified (Figure 4.8) based on: (1). fold change > 1.5; (2). corrected P value < 0.05 (by Benjamini and Hochberg adjustment); (3). both upregulated or both downregulated in Hut78 and HH. We went on to validate these genes by qPCR. Except for three genes that were undetectable (SLC26A11, RAB3A, and GZMB), the remaining 19 differentially expressed candidate genes identified in TOX-suppressed cells were evaluated by qPCR in RNA samples from control cells (n = 3) and TOX-sh cells (n = 6), including the RNA samples used in microarray analysis. The majority of differential expression changes were validated in both transduced Hut78 and HH cells (Table 4.1). Pathway analysis of the candidate genes by Genomatrix software (Table 4.2) and DAVID software (Figure 4.9) revealed activation of apoptosis and cell cycle arrest. Not surprisingly, TOX inhibition greatly altered the transcription activities, in line with its role as a transcription regulator (Figure 4.9B). These results further supported the biological effects  99 of TOX knockdown in CTCL cells, and provided insights into its potential downstream mediators.    100  CTR sh (n=6) TOX-sh (n=6) Gene Putative functionSMAD3 Signal transducer and transcriptional modulatorFOXO3 Transcriptional activator which triggers apoptosisHBP1 Transcriptional repressorTHAP1 Transcription regulatorMED13L Transcription regulatorCRTC1 Transcriptional coactivatorITM2B Protease inhibitorTIAM2 GDP-dissociation stimulatorBTBD7 Function unclearFAM117A Function unclearSLC26A11 Anion exchangerTNFRSF12A Apoptosis regulatorCGREF1 Cell-cell adhesionPSMB5 Proteasome subunitERCC2 ATP-dependent 5'-3' DNA helicaseFKBP4 Immunoregulation in lymphcytesCD3EAP RNA polymerase I and preformed TCR componentMRPL12 Mitochondrial ribosome componentXPO5 Nuclear export receptor complex componentMGAT4A GlycosyltransferaseRAB3A ExocytosisGZMB Caspase activationLow High0 0.4 0.8 1.2 1.6 2 Figure 4.8 Identification of transcriptional targets of TOX in CTCL.   Heatmap of upregulated (red) and downregulated (green) genes upon TOX suppression. The top 22 differentially-expressed genes were selected based on the following criteria: (1). fold change > 1.5; (2). P value < 0.05; (3). both upregulated or both downregulated in Hut78 and HH.   101  Table 4.1 qPCR validation of the 22 differentially expressed genes identified by gene expression profiling   Transcriptome   qPCR validation   Hut78   HH   Hut78   HH  Gene symbol  Fold  P value   Fold  P value   Fold  P value   Fold  P value  Up-regulated  SMAD3  2.3  0.008   7.6  0.033   2.8  0.001   12.5  0.000  FOXO3  1.6  0.015   1.5  0.027   1.9  0.035   2.0  0.002  HBP1  1.6  0.002   2.0  0.031   2.1  0.005   2.8  0.000  THAP1  1.8  0.008   1.7  0.003   2.1  0.039   1.7  0.002  MED13L  1.7  0.025   1.7  0.039   2.5  0.034   2.4  0.001  CRTC1  2.0  0.046   1.8  0.001   2.6  0.051   2.0  0.006  ITM2B  1.6  0.035   2.4  0.025   2.2  0.043   3.3  0.000  TIAM2  1.6  0.039   2.3  0.028   1.8  0.047   3.0  0.000  BTBD7  1.8  0.024   1.6  0.031   3.0  0.041   2.5  0.001  FAM117A  1.6  0.003   1.5  0.003   2.2  0.007   2.3  0.000  SLC26A11  1.6  0.048   1.8  0.006   ND  ND   ND  ND  Down-regulated  TNFRSF12A  1.8  0.017   2.2  0.016   1.6  0.052   1.5  0.001  CGREF1  1.8  0.044   2.6  0.032   1.3  0.009   2.0  0.001  PSMB5  1.8  0.042   2.0  0.049   1.4  0.180   1.6  0.013  ERCC2  1.6  0.042   1.6  0.014   1.2  0.170   1.3  0.009  FKBP4  2.1  0.036   1.6  0.017   1.1  0.380   2.2  0.007  CD3EAP  1.5  0.044   1.7  0.010   1.2  0.210   1.6  0.035  MRPL12  1.6  0.047   1.5  0.000   0.9  0.690   1.5  0.018  XPO5  1.6  0.032   1.7  0.017   1.0  0.910   1.0  0.800  MGAT4A  1.6  0.005   1.9  0.048   0.9  0.760   1.1  0.510  RAB3A  2.0  0.047   2.0  0.040   ND  ND   ND  ND  GZMB  3.0  0.018   1.8  0.046   ND  ND   ND  ND  ND=not detectable  102  Table 4.2 Biological processes related to the top differentially expressed genes in TOX-suppressed CTCL cells as identified by Genomatrix software  Biological processes  P value  List of observed genes  Induction of apoptosis  4.85e-7  GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Induction of programmed cell death  5.02e-7  GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Positive regulation of apoptotic process  6.13e-6  GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Positive regulation of programmed cell death  6.50e-6  GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Positive regulation of cell death  7.45e-6  GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Regulation of apoptotic process  7.69e-5  PSMB5, GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Regulation of programmed cell death  8.12e-5  PSMB5, GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Regulation of cell death  9.73e-5  PSMB5, GZMB, ITM2B, FOXO3, TIAM2, SMAD3, TNFRSF12A, ERCC2  Cell cycle arrest  1.29e-4  PSMB5, SMAD3, HBP1, CGREF1, THAP1, ERCC2  Regulation of cell cycle  2.41e-4  PSMB5, SMAD3, HBP1, CGREF1, ERCC2     103 FOXO3SMAD3    ERCC2TIAM2 ITM2BTNFRSF12AGZMBCRTC1PSMB5GO:0006915~apoptosisGO:0012501~programmed cell deathGO:0008219~cell deathGO:0016265~deathGO:0012502~induction of programmed cell deathGO:0006917~induction of apoptosisGO:0043065~positive regulation of apoptosisGO:0043068~positive regulation of programmed cell deathGO:0010942~positive regulation of cell deathGO:0042981~regulation of apoptosisGO:0043067~regulation of programmed cell deathGO:0010941~regulation of cell deathGO:0005829~cytosolGO:0045944~positive regulation of transcription from RNA polymerase II promoterGO:0016563~transcription activator activityGO:0045597~positive regulation of cell differentiationGO:0051094~positive regulation of developmental process Apoptosis  FOXO3ERCC2SMAD3    CRTC1      MRPL12MED13LTHAP1  HBP1CD3EAPPSMB5XPO5FKBP4GO:0006350~transcriptionGO:0045449~regulation of transcriptionnucleusGO:0006355~regulation of transcription, DNA-dependentGO:0051252~regulation of RNA metabolic processTranscriptionGO:0030528~transcription regulator activitytranscription regulationGO:0010604~positive regulation of macromolecule metabolic processGO:0051254~positive regulation of RNA metabolic processGO:0045935~positive regulation of nucleobase, nucleoside, nucleotide and nucleic acid metabolic processGO:0051173~positive regulation of nitrogen compound metabolic processGO:0010557~positive regulation of macromolecule biosynthetic processGO:0045893~positive regulation of transcription, DNA-dependentGO:0010628~positive regulation of gene expressionGO:0031328~positive regulation of cellular biosynthetic processGO:0009891~positive regulation of biosynthetic processGO:0045941~positive regulation of transcriptionGO:0006357~regulation of transcription from RNA polymerase II promoterGO:0045944~positive regulation of transcription from RNA polymerase II promoterGO:0016563~transcription activator activityCGREF1HBP1  SMAD3    PSMB5THAP1 GO:0007050~cell cycle arrestGO:0022402~cell cycle processGO:0007049~cell cycleA. B. C.  Figure 4.9 Top biological processes enriched in the genes differentially expressed upon TOX knockdown as identified by DAVID Functional Annotation Clustering analysis.  (A) Annotation cluster 1, enrichment score = 2.33. (B) Annotation cluster 2, enrichment score = 1.94. (C) Annotation cluster 3, enrichment score = 1.7.  104 The genes identified by microarray could be either functionally important for TOX’s effect or simply by-standers without significant biological consequences. Validation of the gene function in the face of TOX absence may offer us insight into its significance in mediating TOX’s effect. SMAD3 is the most upregulated gene upon TOX suppression. As a tumor suppressor, SMAD3 is an important mediator in TGFβ signalling pathway, and plays a key role in the suppression of leukemogenesis. Loss of SMAD3 and dysregulated TGFβ signalling was reported in SS (Dulmage & Geskin, 2013; Lee et al, 2012; Letterio, 2005; van Doorn et al, 2004). Therefore SMAD3 appeared to be a promising candidate gene. We confirmed the upregulation of SMAD3 on both transcript and protein levels in TOX-suppressed cells in three CTCL cell lines (Figure 4.10). Next, SMAD3 was silenced by lentivirus-mediated shRNA (sh-1 and sh-2) approach in Hut78 cell line (Figure 4.11). Due to the rapidly fatal effect of second virus infection in SZ4 cells, SMAD3 knockdown was not assayed in SZ4 cell line. In addition, SMAD3 could not be effectively suppressed in HH cell line using the same shRNAs for Hut78. The reason remained unknown, but possible explanations include mutations in SMAD3 genes where the shRNAs target. Nonetheless, Hut78 cells co-transduced by TOX-sh virus and SMAD3-sh virus were subjected to viability assay. Compared to control (TOX-sh) cells, additional SMAD3 knockdown had similar number of viable cells over 4 days, indicating that TOX’s growth inhibitory effect was not directly mediated by SMAD3 in CTCL cells.   105         SMAD3Hut78 CTRHut78 TOX-shHH CTRHH TOX-shSZ4 CTRSZ4 TOX-sh01234121518*** *** *SMAD3 mRNA level/1000GAPDHBATOXSMAD3Hut78ActinCTR  sh-1  sh-2HHCTR  sh-1  sh-2SZ4CTR  sh-1  sh-263 42 kDa57 52  Figure 4.10 TOX depletion restores SMAD3 expression. (A) SMAD3 transcript levels in TOX-sh and CTR cells. qPCR was performed using primers specific for SMAD3 and GAPDH mRNA. The level of SMAD3 mRNA was normalized to that of GAPDH and is depicted as the fold change compared to CTR cells. *P < 0.05, and ***P < 0.001. Error bars indicate SEM. Data depicted are representative of 3 independent experiments. (B) SMAD3 protein levels in TOX-sh (sh-1 and sh-2) cells. Actin was included as a loading control.   106  SMAD3ActinTOX-shSMAD3sh-1+      +      +- +      -0 1 2 3 40204060TOX-shTOX-sh+SMAD-shNSDaycell number/*10^4SMAD3sh-2 - - +Hut78BA Figure 4.11 SMAD3 suppression in TOX-sh Hut78 cells fails to reverse the growth inhibition associated with TOX knockdown.  (A) SMAD3 knockdown by 2 shRNA constructs (sh-1 and sh-2) in addition to TOX knockdown in Hut78 cells. Transduced Hut78 cells were selected by blasticidin (15 μg/mL) for 7 days before the protein lysates were probed with antibodies against SMAD3 and actin proteins. (B) Cosilencing of SMAD3 and TOX (TOX-sh+SMAD-sh) yielded similar viable cell numbers to the control cells with TOX suppression alone (TOX-sh) over a period of 4 days. NS = not significant.  107  4.2.8 TOX Upregulation Is Unlikely a Result of TCR Activation in Mature CD4+ T Cells It remains elusive how TOX becomes activated in CTCL. TOX has been shown to be rapidly induced in developing thymocytes by pre-TCR activation (Wilkinson et al, 2002). In addition, lymphocyte activation is involved in the pathogenesis of CTCL (Wood, 1995). Therefore we asked whether the aberrant TOX activation observed in CTCL was simply an indication of TCR activation. In order to test this, normal CD4+ T cells were stimulated by PMA and ionomycin, or CD3 with/without CD28, which mimic TCR activation. Surprisingly, a dramatic reduction of TOX expression levels was observed upon TCR activation, regardless of stimulation strength (Figure 4.12A and B). Further, the reduction of TOX expression by TCR activation seemed to be much less in Hut78 cell line or lost in HH and SZ4 cell lines ( Figure 4.12B). Therefore TOX upregulation in CTCL is unlikely a result of TCR stimulation in normal CD4+ T cells.    108  CD4 Hut78 HH SZ40. 5ug/mlCD3+CD28 5ug/mlP < 0.001 P = 0.01 NS NSTOX mRNA levelsrelative to control0. 1ug/mlCD3 1ug/mlCTRP = 0.001P = 0.03P = 0.02TOX mRNA levelsrelative to controlA B  Figure 4.12 TOX mRNA levels are highly suppressed by TCR activation in normal CD4+ T cells, but not in CTCL cells. (A) TOX mRNA levels in resting or activated normal CD4+ T cells. For TCR activation, normal CD4+ T cells were incubated with 25 ng/ml PMA and 50 ng/ml ionomycin for 24 hours, or with CD3 (±CD28) for 72 hours at 1ug/ml concentration. (B) TOX mRNA levels in resting or activated normal CD4+ T cells and CTCL cell lines. Normal CD4+ T cells or CTCL cell lines (Hut78, HH, SZ4) were incubated with either isotope control (CTR) or CD3 (±CD28) at 5ug/ml for 72 hours before RNA harvest and qPCR. Two tailed t tests were used for comparison. Error bars indicate SEM. Data depicted here are representative of at least 3 independent experiments. NS = not significant.   109  4.3 Discussion TOX is a member of the evolutionarily conserved HMG box family and a key regulatory nuclear protein in the development of CD4+ T cells, NK cells, and lymphoid tissue inducer cells.(Aliahmad et al, 2010; Aliahmad et al, 2011; Aliahmad et al, 2004; Aliahmad et al, 2012; Wilkinson et al, 2002) The roles of TOX in immune system development are relatively well-characterized since its discovery (Wilkinson et al, 2002). However, TOX’s role in human cancer has not been reported. Based on our findings, TOX has emerged as a key player, when expressed ectopically, in driving malignant cell transformation in CTCL. The molecular pathogenesis of SS has been a topic of active investigation in the past decade. However, the molecular nature of SS cells has only started to be understood. Aberrations in signal transducers and transcription factors have been reported in SS studies, such as Jun B, JunD, TGFBR2, STAT3, STAT4, CDKN1C, CTLA-4, GATA3, AHI-1, SATB1, PD-1, and NOTCH1 (Gibson et al, 2013; Kamstrup et al, 2010; Kari et al, 2003; Lauenborg et al, 2010; Lee et al, 2012; Mao et al, 2008; Ringrose et al, 2006; Wang et al, 2011; Zhang et al, 2012). Several of these dysregulated genes (TGFBR2, GATA3, STAT3, SATB1, and NOTCH1) are involved in the signal cascades governing T cell development (Burute et al, 2012; Ho et al, 2009; Li & Flavell, 2008; Naito et al, 2011; Stritesky et al, 2011). Our recent observation of aberrant TOX upregulation in MF and SS prompted us to ask whether TOX contributes to the development of CTCL.        This possibility was tested and confirmed in the current study using cultured CTCL cells and a xenograft mouse model. In three CTCL cell lines, TOX suppression by lentivirus-mediated shRNA gene silencing markedly normalized the CTCL cells’ resistance to  110 apoptosis and abnormal cell cycle progression, the two best-characterized features of CTCL cells (Dulmage & Geskin, 2013; Kim et al, 2005; Mao et al, 2006; Nihal et al, 2014). In agreement with in vitro data, subcutaneous injection of CTCL cells into NSG mice confirmed this proliferative disadvantage in that the ability of TOX-suppressed CTCL cells to induce local tumors was significantly impaired or even abolished. In aggregate, these findings demonstrated the biological importance of sustained TOX activation to the leukemic activities of CTCL cells.  The exact mechanism by which TOX exerts its biological functions is unclear. To explore the possible mechanisms, we employed several lines of experiments to understand the downstream molecular events of TOX. First, we examined key regulators of cell cycle control molecules p27 (CDKN1B) and p57 (CDKN1C), given the profound impact TOX knockdown has on cell-cycle progression (Figure 4.5).   We found that both CDKN1B and CDKN1C were increased upon TOX suppression, both at the mRNA level and at the protein level. CDKN1B and CDKN1C were frequently lost in many lymphoid malignancies, including SS (Borriello et al, 2011; da Silva Almeida et al, 2015; Kennah et al, 2009; Lim et al, 2002). Moreover, a recent study reported that CDKN1C deficiency was associated with poor outcome in CTCL patients (Litvinov et al, 2012). Intriguingly, in addition to their role in cell cycle control, a dual role in apoptosis has been documented for both p27 and p57 (Besson et al, 2008; Lloyd et al, 1999; Samuelsson et al, 2002; Vlachos et al, 2007). Therefore they appear to be promising mediators of the dual effects of TOX on apoptosis and cell cycling. Indeed, additional stable knockdown of CDKN1B or CDKN1C increased growth of CTCL cells in the face of loss of TOX, thus reversing the effect of TOX  111 knockdown. These data suggest that TOX’s leukemic activities in CTCL, at least in part, rely on repressing the transcription of CDKN1B and CDKN1C. Given that silencing of neither gene was able to fully restore uncontrolled cell growth, it is likely that additional partners are involved in mediating the functions of TOX, which warrants future studies.  It is unknown how TOX regulate its downstream players. CDKN1B transcription can be activated by the binding of several transcription factors, although the changes of CDKN1B are largely regulated post-translationally (Chu et al, 2008). CDKN1C is an imprinted gene rich in CGP islands near its putative transcription start site. Its transcription is under control of numerous pathways, with transcription control, epigenetic control, and microRNA regulation being the best characterized (Algar et al, 2009; Borriello et al, 2011; Fornari et al, 2008; Kim et al, 2009; Topark-Ngarm et al, 2006; Yang et al, 2009). Several mechanisms are possible as to how TOX regulates CDKN1B and CDKN1C, including direct binding to their promoters or modulating local chromatin structure and multi-protein complexes formation, a mechanism generally employed by HMG box proteins (Wilkinson et al, 2002). Another possibility is through miRNAs. It has been shown that miR-221 and miR-222 regulate both p27 and p57 in hepatocellular carcinoma and sporadic ovarian carcinoma (Fornari et al, 2008; Wurz et al, 2010). Therefore TOX may exert its regulatory effect through control the expression of these miRNAs. Further experiments in the future may help clarify the precise mechanism of how TOX regulates p27 and p57. In addition to examining the effect of TOX on the transcription of cell cycle control genes, we also examined how TOX changes gene transcription on a global scale by performing transcrptiome profiling after TOX knockdown. First, gene expression profiling was performed on two CTCL cell lines (Hut78 and HH), comparing the transcriptome with  112 and without TOX suppression. The altered genes taken together point to a profound role of TOX in the regulation of apoptosis and cell cycle control genes. The top 22 differentially expressed include several tumor suppressors, including SMAD3, FOXO3, and HBP1. Among these, SMAD3 was the most significantly upregulated gene. Subsequent knockdown experiments of SMAD3 in the face of TOX reduction revealed no proliferation difference between cells with TOX and SMAD3 cosilencing and TOX silencing alone. The fact that SMAD3 knockdown failed to reverse the proliferation reduction in the absence of TOX indicated that SMAD3 alone cannot explain the effect of TOX on CTCL proliferation. Additional genes must be involved. Future experiments are needed to uncover these genes.  It is unclear how TOX becomes activated in CTCL. Several lines of evidence suggested potential triggers for its activation, including epigenetic changes and TCR signalling. McGirt et al recently reported miR-223, a microRNA decreased in CTCL, was a negative regulator of TOX (McGirt et al, 2014). Another line of evidence came from rapid induction of TOX by TCR signalling in the developing thymocytes (Wilkinson et al, 2002). However, we observed a reduction of TOX expression in mature CD4+ T cell following cellular activation, indicating that the regulatory role of TCR signalling on TOX expression is stage- and cellular context-dependent. Furthermore, this rapid reduction of TOX was impaired or absent in CTCL cell lines. As TCR signalling pathway was found disrupted in CTCL (Lee et al, 2012), it is possible that CTCL cells turned on abnormal TCR signalling that led to TOX induction. Interestingly, transgenic mice studies showed that TOX expression affects the sensitivity of thymocytes to TCR signalling, adding another layer of complexity to the TOX-TCR feedback loop (Wilkinson et al, 2002). Finally, the possibility  113 of gain-of-function genomic mutations in TOX transcription regulation elements remains to be tested in future studies.  In conclusion, we provide strong evidence that TOX is highly expressed in CTCL and exerts transforming activities in CTCL. Upon TOX silencing, CTCL cells’ growth was markedly slowed down or halted both in vitro and in vivo, due to cell cycle arrest and apoptosis sensitization. This effect is partially mediated by two key CDK inhibitors, CDKN1B and CDKN1C. Correction of TOX and/or CDKN1B/CDKN1C, therefore, may be an appropriate therapeutic approach to harness the uncontrolled tumor growth.    114 Chapter 5: Summary and Future Directions 5.1 Summary In this study, we sought to narrow the knowledge gaps in the clinical management and disease pathogenesis of CTCL. First, there is no highly specific and sensitive molecular diagnostic marker for CTCL used in the clinical setting. The early diagnosis of CTCL is thus difficult due to the clinical or even pathological resemblance to other benign inflammatory skin conditions. Second, molecular markers with prognostic value are largely lacking for predicting the long-term outcome of CTCL patients. Therefore it remains difficult to determine the appropriate therapeutic regimens according to the risk of disease progression or mortality. Finally, the molecular pathogenesis of CTCL remains poorly understood, hindering the development of novel effective therapeutic agents for this incurable disease. The results obtained from this research contributed knowledge in all these areas.  In Chapter 3, we examined the expression of TOX, a gene our group discovered to be upregulated in early MF skin lesions compared to BID skin (Zhang et al, 2012), in the entire spectrum of MF and SS. We confirmed that aberrant TOX activation is a feature shared by MF and SS. Further, we found that TOX, when expressed at high levels, differentiates CTCL from non-CTCL cases with good sensitivity and specificity.  TOX plays a pivotal role in thymocyte development and lineage commitment (Aliahmad et al, 2012). In Chapter 4, we characterized the biological role of TOX in the development of CTCL. Using extensive in vitro and in vivo assays, we present evidence that TOX plays an oncogenic role in mediating the malignant behaviours of CTCL cells, including apoptosis resistance and uncontrolled cell cycle progression (Figure 5.1). Further exploration of the downstream molecules of TOX uncovered two critical cell cycle  115 regulators, CDKN1B and CDKN1C, as mediators of the proliferative effect of TOX in CTCL cells (Figure 5.2). Therefore TOX and its downstream molecules may be promising therapeutic targets for CTCL.   116  ThymusDP DDCD4+CD8loCD4+CD8+BloodCD4+CD8+PositiveselectionNegativeselectionTOXSkinHomeostasisCD4+ CD4+DP DDCD4+CD8loCD4+CD8+CD4+CD8+PositiveselectionNegativeselectionTOXCD4+ CD4+ABTOXCD4+CD4+CD4+CD4+CD4+CD4+CD4+CD4+CD4+CD4+  Figure 5.1 Mode of TOX expression in normal T cells and in CTCL. (A) Normally, TOX is upregulated during positive selection, then gradually downregulated before mature T cells exit the thymus, and remain poorly expressed in the peripheral tissues. The proliferation and apoptosis for CD4+ T cells remain in a homeostatic state. (B) In CTCL, TOX is overespressed in the CD4+ T cells in the peripheral tissues, either due to failure to shut down TOX before thymic exit or re-activation of expression. The aberrant TOX expression leads to increased cell cycling and reduced sensitivity of apoptosis, together contributing to uncontrolled expansion of the malignant CD4+ T cells.    117  TOXCDKN1BCDKN1CCyclin A/E-CDK2Cyclin D-CDK4Cell cycle progressionPositive regulation Negative regulation ApoptosisCaspaseactivationOthers?Cytoplasm Nuclear  Figure 5.2 Proposed model of the role of TOX in CTCL. Aberrant TOX activation leads to suppression of two cyclin-dependent kinase inhibitors, CDKN1B (encoding p27) and CDKN1C (encoding p57), and possibly other downstream mediators. Reduced inhibition of cyclin-CDK complexes by p27 and p57 results in accelerated cell cycle progression. In addition, reduction of p27 and p57 inhibits caspase activation and therefore contributes to apoptosis resistance. As a result, CTCL cells exhibit enhanced cellular growth and tumor forming ability.        118 5.2 Significance and Limitations This research contributed new knowledge in several aspects. (1). TOX is an easy to use, sensitive and specific diagnostic marker for CTCL.  On the clinical application side, our data showed that TOX has important roles in CTCL clinical management, on two levels. On the one hand, it is an important diagnostic marker. Based on transcript levels, when TOX positivity is used as marker using qPCR, it has 85-90% specificity and 75-83% sensitivity for distinguishing CTCL from non-CTCL cases. This is one of the few markers available for CTCL. Based on protein levels as measured by IHC staining, TOX is a sensitive marker, detecting abnormal or malignant CD4+ T cells in skin biopsies and in the peripheral blood mononuclear cells. This is the only available positive protein marker for both MF and SS cells. Based on this, a novel CTCL diagnostic test (seeTCL) has been developed. This should have implications in the future for clinical diagnosis of CTCL patients. The diagnostic value of TOX in CTCL is especially important for early-stage MF, which is exceedingly similar to chronic benign dermatitis. We have demonstrated in this thesis that TOX positivity is very helpful for discrimination between these two conditions.  Even for detecting circulating CTCL cells, TOX demonstrates clinical utility. We found that using a nuclear staining protocol, TOX staining could specifically identify a small group of TOX-high cells only in patients with CTCL. This suggests that TOX could be a component of future peripheral immune phenotyping analysis of all CTCL patients. (2). TOX can be used to help guide management of patients with CTCL. In addition to diagnosis, TOX also has value in the prognostication of CTCL, in that the higher it is, the worse patient will do in the future. In early MF cases, if the TOX marker is  119 negative, the patients essentially behave in a benign fashion--that is, they have a very low risk of progressing to higher disease stages, or dying from MF, suggesting TOX could be used to stratify patients into two groups: 1). a TOX high group, needing aggressive management, and 2). a TOX low group, who only need conservative treatment and close monitoring, thus avoiding exposure to potentially toxic or dangerous treatments such as cytotoxic drugs, ionizing radiation, or UV exposure. (3). TOX overexpression provides a new model of CTCL pathogenesis.  On the biological implications side, our data expanded the known biological functions of TOX protein. Previously, TOX was known to be an essential regulator of early T cell development. However, beyond that, how it contributes to human disease processes was unknown. Our research for the first time uncovered what happens when TOX is dysregulated. Specifically, when TOX fails to be downregulated or silenced in mature T cells, as it should be, there is uncontrollable proliferation of the resultant T cells. Therefore TOX may be a promising target for the development of novel CTCL treatments. To our knowledge, this is the first time that TOX has been shown to be involved in a major human disease.    (4). TOX provides a novel and improved tool for purification of CTCL malignant cells for further experimental research. TOX staining provides a useful tool for purifying more homogenous populations of malignant cells from CTCL patients, enabling more detailed molecular and genetic characterization of the disease. At the present time, this is a difficult or impossible task, especially for early MF, where infiltrating malignant MF cells are mixed with large numbers of benign reactive CD4+ T cells.   120 Despite the promising results of TOX being a useful biomarker for CTCL, certain limitations need to be taken into consideration. First, due to the rarity of disease, only a moderate number (n = 113) of MF and a small number (n = 12) of SS subjects were assembled and evaluated. For disease progression and mortality analysis, only data for MF cohort 2 (n = 59) were available. The true clinical usefulness of TOX as a disease marker will need to be verified in other clinical centres and in larger cohorts with long term follow up data. Second, in multivariate analysis, only stage, age, and sex could be analyzed with TOX expression, based on what clinical data was available for MF cohort 2. Several other clinical parameters, such as LDH level, ethnicity, and large cell transformation, which also have prognostic implications, could not be included in our analysis. In future studies, it will be of interest to evaluate the independent prognostic value of TOX expression levels after inclusion of these additional parameters.  It should also be pointed out that in our attempt to explore the molecules downstream of TOX, two CTCL cell lines (Hut78 and HH) were subjected to gene expression profiling before and after TOX knockdown. It is well accepted that the gene expression profiles in cancer cell lines often differ from those in the clinical samples (Leupin et al, 2006). For example, PLS3 was found to be overexpressed in SS, but has been lost in Hut78 cells (Wang et al, 2011). Therefore important genes regulated by TOX in the clinical context may be missed in this cell line based approach. Furthermore, the sample sizes of cell lines (for each cell line, TOX-sh = 3, CTR = 3) used in our microarray analysis were small, which may have limited power in identifying truly differentially expressed genes.     121 5.3 Future Directions While there are still many unanswered questions needing additional investigation, the following future projects merit focused discussions and exploration in the near term. (1). Incorporating TOX marker into patient management in routine diagnostic dermatopathology labs: Given the diagnostic and prognostic value of TOX for CTCL, it could be incorporated into clinical practice. On the diagnostic side, TOX can be added to the routine IHC staining panel used on skin biopsies (CD2, CD3, CD5, CD7 etc) from all suspected MF patients, and to the FACS immunophenotyping panel (CD4, CD7, CD8, CD26 etc) used to analyze PBMC samples from all suspected SS patients. When clinical, histological and immunophenotyping features are inadequate to accurately establish a diagnosis of CTCL, having high TOX expression would strongly support the diagnosis. On the prognostic side, for a diagnosed CTCL patient, if TOX expression in the lesion/PBMC sample is high, the patient is at increased risk of disease progression and disease-specific mortality, and therefore more aggressive treatments are warranted. In contrast, for patients with low TOX expression levels, their tendency to experience disease progression or disease-specific death is low, therefore conservative management or observation alone should be considered to minimize treatment-associated toxicities.   (2). Clarification of the mechanism of action of TOX in promoting CTCL pathogenesis: Two aspects can be pursued to further our understanding of the mechanism of action of TOX. First, what are the downstream mediators and protein interaction partners of TOX? Second, how does TOX regulate its downstream players?   122 While further validation and functional study is needed to characterize the role of the differentially expressed genes discovered by microarray analysis upon TOX knockdown (as described in Chapter 4), combining chromatin immunoprecipitation assays with sequencing (ChIP-seq) could be performed to identify potential DNA fragments bound by TOX protein. In addition, TOX, as a member of the HMG box proteins, is capable of recruiting and forming a multi-protein complex (Wilkinson et al, 2002). To identify the TOX-interacting proteins, a yeast two hybrid assay could be performed, using a library that consists of all the proteins that are expressed in CTCL cells.  In terms of the potential mechanisms of TOX regulating its downstream molecules, it is likely that TOX binds to the promoter of its target genes. While TOX contains a single HMG box, sequence alignment comparison by O'Flaherty et al. predicted that TOX fits best into the sequence-independent (structure-dependent) group of HMG box proteins (O'Flaherty & Kaye, 2003). The structure profiles recognized by TOX remain to be identified. Since the computational prediction of transcription factor-DNA binding relies on the sequence-specificity (Kummerfeld & Teichmann, 2006), it is currently difficult to predict the DNA structures which might bind TOX. Another possibility is that TOX regulates its target genes by modulateing local chromatin structure, a mechanism generally employed by HMG box proteins (Wilkinson et al, 2002). An epigenetic profiling study in CTCL cells with and without TOX suppression would allow us to understand the DNA methylation and histone modification patterns affected by TOX expression. Finally, TOX may regulate its targets by modifying microRNA expression. A microRNA microarray analysis could unveil potential microRNAs regulated by TOX, and offer valuable insights into CTCL pathogenesis.  (3). Exploration of the mechanism of abnormal TOX activation in CTCL:   123 Little is known about how TOX becomes aberrantly expressed in CTCL. Potential triggers for TOX activation include dysregulated microRNA and TCR signalling. McGirt et al. recently reported miR-223, a microRNA decreased in CTCL, was a negative regulator of TOX (McGirt et al, 2014). Another line of evidence has come from the observed rapid induction of TOX by TCR signalling in developing thymocytes. Although we did not observe TOX upregulation by TCR activation in mature CD4+ T cells, it is possible that CTCL cells have turned on abnormal TCR signalling that leads to TOX induction, as TCR signalling was found to be disrupted in CTCL (Chung et al, 2011; Lee et al, 2012; Nikolova et al, 2002). Finally, the possibility of gain-of-function genomic mutations remains to be tested. (4). Identification of targeted therapies for selective inhibition of TOX:  Therapeutically targeting transcription factors, which can modulate large number of gene, is an area of intense study and challenges. The strong oncogenic ability of TOX observed in CTCL gives us great confidence that it could serve as a promising drug target for developing therapies for CTCL, which currently have limited effectiveness in advanced disease. TOX, as a transcription factor, is mainly located in the nucleus. It is hard to target transcription factors, in part due to their large surface interaction area and largely nuclear localization, which hinders their accessibility by drugs in the circulation (Yeh et al, 2013). Nonetheless, recent advancements have provided us new strategies to target these key factors.  Small molecule inhibitors have shown promising results in targeting transcription factors, such as MDM2-p53 and Myc (Fletcher & Prochownik, 2015; Vassilev et al, 2004). One powerful approach to identify small molecule inhibitors is by high throughput screening of a large number of synthetic compounds using biological assays (Redell & Tweardy, 2006). For example, the Developmental Therapeutics Program of the National Cancer Institute  124 (http://dctd.cancer.gov/ProgramPages/dtp/default.htm) maintains a drug bank with over 400,000 candidate drugs, and could facilitate discovery of potential small molecule inhibitors of TOX. Once lead compounds are identified, further modifications can be made to enhance the stability, solubility, and binding affinity.  Antisense oligonucleotides and RNA interference (RNAi), using either shRNA or siRNA, offers another promising platform for normalizing deregulated transcription factors. An antisense oligonucleotide targeting c-Myc has been administered to human subjects systemically, without obvious toxicity (Iversen et al, 2003). Promising in vivo results have been shown using siRNA to suppress transcription factors, including STAT3 and NF-κB (Gao et al, 2005; Guo et al, 2004). However, stability and in vivo delivery systems remain a challenge for this platform.  Another possible way to alter the expression of TOX is by altering its regulatory factors. As TOX is negatively regulated by miR-223, a miR-223 mimic might be used to attenuate TOX expression. Similar to siRNA therapeutics, delivery of microRNA into cells appears to be a major hurdle in drug development (Garzon et al, 2010). Interestingly, TOX has been shown to be under the regulation of the calcineurin signaling pathway in the developing thymocyte and brain (Aliahmad et al, 2004; Artegiani et al, 2015). Calcineurin inhibitors, such as cyclosporine A and tacrolimus may be promising agents to suppress TOX expression in CTCL.  Collectively, these proposed studies would not only offer us a better understanding of the mechanisms of action employed by TOX and the key events leading to its abnormal activation in CTCL, but also provide us opportunities to develop more effective therapies for CTCL.   125 Bibliography  Adlam M, Siu G (2003) Hierarchical interactions control CD4 gene expression during thymocyte development. Immunity 18: 173-184  Agar NS, Wedgeworth E, Crichton S, Mitchell TJ, Cox M, Ferreira S, Robson A, Calonje E, Stefanato CM, Wain EM, Wilkins B, Fields PA, Dean A, Webb K, Scarisbrick J, Morris S, Whittaker SJ (2010) Survival outcomes and prognostic factors in mycosis fungoides/Sezary syndrome: validation of the revised International Society for Cutaneous Lymphomas/European Organisation for Research and Treatment of Cancer staging proposal. J Clin Oncol 28: 4730-4739  Aidinis V, Bonaldi T, Beltrame M, Santagata S, Bianchi ME, Spanopoulou E (1999) The RAG1 homeodomain recruits HMG1 and HMG2 to facilitate recombination signal sequence binding and to enhance the intrinsic DNA-bending activity of RAG1-RAG2. Molecular and cellular biology 19: 6532-6542  Alam R, Gorska M (2003) 3. Lymphocytes. The Journal of allergy and clinical immunology 111: S476-485  Algar EM, Muscat A, Dagar V, Rickert C, Chow CW, Biegel JA, Ekert PG, Saffery R, Craig J, Johnstone RW, Ashley DM (2009) Imprinted CDKN1C is a tumor suppressor in rhabdoid tumor and activated by restoration of SMARCB1 and histone deacetylase inhibitors. PLoS One 4: e4482  Aliahmad P, de la Torre B, Kaye J (2010) Shared dependence on the DNA-binding factor TOX for the development of lymphoid tissue-inducer cell and NK cell lineages. Nat Immunol 11: 945-952  Aliahmad P, Kadavallore A, de la Torre B, Kappes D, Kaye J (2011) TOX is required for development of the CD4 T cell lineage gene program. J Immunol 187: 5931-5940  Aliahmad P, Kaye J (2006) Commitment issues: linking positive selection signals and lineage diversification in the thymus. Immunological reviews 209: 253-273  Aliahmad P, Kaye J (2008) Development of all CD4 T lineages requires nuclear factor TOX. J Exp Med 205: 245-256  Aliahmad P, O'Flaherty E, Han P, Goularte OD, Wilkinson B, Satake M, Molkentin JD, Kaye J (2004) TOX provides a link between calcineurin activation and CD8 lineage commitment. J Exp Med 199: 1089-1099  Aliahmad P, Seksenyan A, Kaye J (2012) The many roles of TOX in the immune system. Curr Opin Immunol 24: 173-177   126 Anderson MS, Venanzi ES, Klein L, Chen Z, Berzins SP, Turley SJ, von Boehmer H, Bronson R, Dierich A, Benoist C, Mathis D (2002) Projection of an immunological self shadow within the thymus by the aire protein. Science 298: 1395-1401  Angus SP, Nevins JR (2012) A role for Mediator complex subunit MED13L in Rb/E2F-induced growth arrest. Oncogene 31: 4709-4717  Ansell SM, Lesokhin AM, Borrello I, Halwani A, Scott EC, Gutierrez M, Schuster SJ, Millenson MM, Cattry D, Freeman GJ, Rodig SJ, Chapuy B, Ligon AH, Zhu L, Grosso JF, Kim SY, Timmerman JM, Shipp MA, Armand P (2015) PD-1 blockade with nivolumab in relapsed or refractory Hodgkin's lymphoma. N Engl J Med 372: 311-319  Arai E, Katayama I, Ishihara K (1991) Mycosis fungoides and Sezary syndrome in Japan. Clinicopathologic study of 107 autopsy cases. Pathol Res Pract 187: 451-457  Artegiani B, de Jesus Domingues AM, Bragado Alonso S, Brandl E, Massalini S, Dahl A, Calegari F (2015) Tox: a multifunctional transcription factor and novel regulator of mammalian corticogenesis. EMBO J 34: 896-910  Ascierto PA, Marincola FM (2015) 2015: The Year of Anti-PD-1/PD-L1s Against Melanoma and Beyond. EBioMedicine 2: 92-93  Assaf C, Hummel M, Steinhoff M, Geilen CC, Orawa H, Stein H, Orfanos CE (2005) Early TCR-beta and TCR-gamma PCR detection of T-cell clonality indicates minimal tumor disease in lymph nodes of cutaneous T-cell lymphoma: diagnostic and prognostic implications. Blood 105: 503-510  Axelrod PI, Lorber B, Vonderheid EC (1992) Infections complicating mycosis fungoides and Sezary syndrome. Jama 267: 1354-1358  Ballabio E, Mitchell T, van Kester MS, Taylor S, Dunlop HM, Chi J, Tosi I, Vermeer MH, Tramonti D, Saunders NJ, Boultwood J, Wainscoat JS, Pezzella F, Whittaker SJ, Tensen CP, Hatton CS, Lawrie CH (2010) MicroRNA expression in Sezary syndrome: identification, function, and diagnostic potential. Blood 116: 1105-1113  Baxevanis AD, Landsman D (1995) The HMG-1 box protein family: classification and functional relationships. Nucleic Acids Res 23: 1604-1613  Berger R, Rotem-Yehudar R, Slama G, Landes S, Kneller A, Leiba M, Koren-Michowitz M, Shimoni A, Nagler A (2008) Phase I safety and pharmacokinetic study of CT-011, a humanized antibody interacting with PD-1, in patients with advanced hematologic malignancies. Clin Cancer Res 14: 3044-3051  Bernier C, Nguyen JM, Quereux G, Renault JJ, Bureau B, Dreno B (2007) CD13 and TCR clone: markers of early mycosis fungoides. Acta Derm Venereol 87: 155-159   127 Besson A, Dowdy SF, Roberts JM (2008) CDK inhibitors: cell cycle regulators and beyond. Dev Cell 14: 159-169  Bettelli E, Oukka M, Kuchroo VK (2007) T(H)-17 cells in the circle of immunity and autoimmunity. Nat Immunol 8: 345-350  Bewley CA, Gronenborn AM, Clore GM (1998) Minor groove-binding architectural proteins: structure, function, and DNA recognition. Annual review of biophysics and biomolecular structure 27: 105-131  Booken N, Gratchev A, Utikal J, Weiss C, Yu X, Qadoumi M, Schmuth M, Sepp N, Nashan D, Rass K, Tuting T, Assaf C, Dippel E, Stadler R, Klemke CD, Goerdt S (2008) Sezary syndrome is a unique cutaneous T-cell lymphoma as identified by an expanded gene signature including diagnostic marker molecules CDO1 and DNM3. Leukemia 22: 393-399  Booth JS, Toapanta FR, Salerno-Goncalves R, Patil S, Kader HA, Safta AM, Czinn SJ, Greenwald BD, Sztein MB (2014) Characterization and functional properties of gastric tissue-resident memory T cells from children, adults, and the elderly. Front Immunol 5: 294  Borriello A, Caldarelli I, Bencivenga D, Criscuolo M, Cucciolla V, Tramontano A, Oliva A, Perrotta S, Della Ragione F (2011) p57(Kip2) and cancer: time for a critical appraisal. Molecular cancer research : MCR 9: 1269-1284  Bounaix Morand du Puch C, Barbier E, Kraut A, Coute Y, Fuchs J, Buhot A, Livache T, Seve M, Favier A, Douki T, Gasparutto D, Sauvaigo S, Breton J (2011) TOX4 and its binding partners recognize DNA adducts generated by platinum anticancer drugs. Archives of biochemistry and biophysics 507: 296-303  Boyman O, Hefti HP, Conrad C, Nickoloff BJ, Suter M, Nestle FO (2004) Spontaneous development of psoriasis in a new animal model shows an essential role for resident T cells and tumor necrosis factor-alpha. J Exp Med 199: 731-736  Bradford PT, Devesa SS, Anderson WF, Toro JR (2009) Cutaneous lymphoma incidence patterns in the United States: a population-based study of 3884 cases. Blood 113: 5064-5073  Brahmer JR, Drake CG, Wollner I, Powderly JD, Picus J, Sharfman WH, Stankevich E, Pons A, Salay TM, McMiller TL, Gilson MM, Wang C, Selby M, Taube JM, Anders R, Chen L, Korman AJ, Pardoll DM, Lowy I, Topalian SL (2010) Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity, pharmacodynamics, and immunologic correlates. J Clin Oncol 28: 3167-3175  Brugnera E, Bhandoola A, Cibotti R, Yu Q, Guinter TI, Yamashita Y, Sharrow SO, Singer A (2000) Coreceptor reversal in the thymus: signaled CD4+8+ thymocytes initially terminate CD8 transcription even when differentiating into CD8+ T cells. Immunity 13: 59-71   128 Bryant DM, Datta A, Rodriguez-Fraticelli AE, Peranen J, Martin-Belmonte F, Mostov KE (2010) A molecular network for de novo generation of the apical surface and lumen. Nat Cell Biol 12: 1035-1045  Bueno OF, Brandt EB, Rothenberg ME, Molkentin JD (2002) Defective T cell development and function in calcineurin A beta -deficient mice. Proceedings of the National Academy of Sciences of the United States of America 99: 9398-9403  Burute M, Gottimukkala K, Galande S (2012) Chromatin organizer SATB1 is an important determinant of T-cell differentiation. Immunol Cell Biol 90: 852-859  Bustin M (1999) Regulation of DNA-dependent activities by the functional motifs of the high-mobility-group chromosomal proteins. Molecular and cellular biology 19: 5237-5246  Camp RL, Dolled-Filhart M, Rimm DL (2004) X-tile: a new bio-informatics tool for biomarker assessment and outcome-based cut-point optimization. Clin Cancer Res 10: 7252-7259  Campbell JJ, Clark RA, Watanabe R, Kupper TS (2010) Sezary syndrome and mycosis fungoides arise from distinct T-cell subsets: a biologic rationale for their distinct clinical behaviors. Blood 116: 767-771  Campo E, Swerdlow SH, Harris NL, Pileri S, Stein H, Jaffe ES (2011) The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood 117: 5019-5032  Carbone A, Bernardini L, Valenzano F, Bottillo I, De Simone C, Capizzi R, Capalbo A, Romano F, Novelli A, Dallapiccola B, Amerio P (2008) Array-based comparative genomic hybridization in early-stage mycosis fungoides: recurrent deletion of tumor suppressor genes BCL7A, SMAC/DIABLO, and RHOF. Genes Chromosomes Cancer 47: 1067-1075  Chang DK, Sui J, Geng S, Muvaffak A, Bai M, Fuhlbrigge RC, Lo A, Yammanuru A, Hubbard L, Sheehan J, Campbell JJ, Zhu Q, Kupper TS, Marasco WA (2012) Humanization of an anti-CCR4 antibody that kills cutaneous T-cell lymphoma cells and abrogates suppression by T-regulatory cells. Mol Cancer Ther 11: 2451-2461  Chong BF, Wilson AJ, Gibson HM, Hafner MS, Luo Y, Hedgcock CJ, Wong HK (2008) Immune function abnormalities in peripheral blood mononuclear cell cytokine expression differentiates stages of cutaneous T-cell lymphoma/mycosis fungoides. Clin Cancer Res 14: 646-653  Chu IM, Hengst L, Slingerland JM (2008) The Cdk inhibitor p27 in human cancer: prognostic potential and relevance to anticancer therapy. Nat Rev Cancer 8: 253-267  Chung JS, Shiue LH, Duvic M, Pandya A, Cruz PD, Jr., Ariizumi K (2011) Sezary syndrome cells overexpress syndecan-4 bearing distinct heparan sulfate moieties that suppress T-cell  129 activation by binding DC-HIL and trapping TGF-beta on the cell surface. Blood 117: 3382-3390  Ciofani M, Zuniga-Pflucker JC (2007) The thymus as an inductive site for T lymphopoiesis. Annual review of cell and developmental biology 23: 463-493  Clark RA (2015) Resident memory T cells in human health and disease. Sci Transl Med 7: 269rv261  Clark RA, Chong B, Mirchandani N, Brinster NK, Yamanaka K, Dowgiert RK, Kupper TS (2006) The vast majority of CLA+ T cells are resident in normal skin. Journal of immunology 176: 4431-4439  Clark RA, Shackelton JB, Watanabe R, Calarese A, Yamanaka K, Campbell JJ, Teague JE, Kuo HP, Hijnen D, Kupper TS (2011) High-scatter T cells: a reliable biomarker for malignant T cells in cutaneous T-cell lymphoma. Blood 117: 1966-1976  Coffman RL, Seymour BW, Hudak S, Jackson J, Rennick D (1989) Antibody to interleukin-5 inhibits helminth-induced eosinophilia in mice. Science 245: 308-310  Contassot E, Kerl K, Roques S, Shane R, Gaide O, Dupuis M, Rook AH, French LE (2008) Resistance to FasL and tumor necrosis factor-related apoptosis-inducing ligand-mediated apoptosis in Sezary syndrome T-cells associated with impaired death receptor and FLICE-inhibitory protein expression. Blood 111: 4780-4787  Criscione VD, Weinstock MA (2007) Incidence of cutaneous T-cell lymphoma in the United States, 1973-2002. Arch Dermatol 143: 854-859  Crowley JJ, Nikko A, Varghese A, Hoppe RT, Kim YH (1998) Mycosis fungoides in young patients: clinical characteristics and outcome. J Am Acad Dermatol 38: 696-701  da Silva Almeida AC, Abate F, Khiabanian H, Martinez-Escala E, Guitart J, Tensen CP, Vermeer MH, Rabadan R, Ferrando A, Palomero T (2015) The mutational landscape of cutaneous T cell lymphoma and Sezary syndrome. Nat Genet  Dani T, Knobler R (2009) Extracorporeal photoimmunotherapy-photopheresis. Front Biosci (Landmark Ed) 14: 4769-4777  de Quatrebarbes J, Esteve E, Bagot M, Bernard P, Beylot-Barry M, Delaunay M, D'Incan M, Souteyrand P, Vaillant L, Cordel N, Courville P, Joly P (2005) Treatment of early-stage mycosis fungoides with twice-weekly applications of mechlorethamine and topical corticosteroids: a prospective study. Arch Dermatol 141: 1117-1120  Dereure O, Levi E, Kadin ME (2000a) T-Cell clonality in pityriasis lichenoides et varioliformis acuta: a heteroduplex analysis of 20 cases. Arch Dermatol 136: 1483-1486   130 Dereure O, Levi E, Vonderheid EC, Kadin ME (2002) Infrequent Fas mutations but no Bax or p53 mutations in early mycosis fungoides: a possible mechanism for the accumulation of malignant T lymphocytes in the skin. J Invest Dermatol 118: 949-956  Dereure O, Portales P, Clot J, Guilhou JJ (2000b) Decreased expression of Fas (APO-1/CD95) on peripheral blood CD4+ T lymphocytes in cutaneous T-cell lymphomas. Br J Dermatol 143: 1205-1210  Diamandidou E, Colome-Grimmer M, Fayad L, Duvic M, Kurzrock R (1998) Transformation of mycosis fungoides/Sezary syndrome: clinical characteristics and prognosis. Blood 92: 1150-1159  Diamandidou E, Colome M, Fayad L, Duvic M, Kurzrock R (1999) Prognostic factor analysis in mycosis fungoides/Sezary syndrome. J Am Acad Dermatol 40: 914-924  Dittmer S, Kovacs Z, Yuan SH, Siszler G, Kogl M, Summer H, Geerts A, Golz S, Shioda T, Methner A (2011) TOX3 is a neuronal survival factor that induces transcription depending on the presence of CITED1 or phosphorylated CREB in the transcriptionally active complex. Journal of cell science 124: 252-260  Duarte RF, Schmitz N, Servitje O, Sureda A (2008) Haematopoietic stem cell transplantation for patients with primary cutaneous T-cell lymphoma. Bone Marrow Transplant 41: 597-604  Dulmage BO, Geskin LJ (2013) Lessons learned from gene expression profiling of cutaneous T-cell lymphoma. Br J Dermatol 169: 1188-1197  Dummer R, Quaglino P, Becker JC, Hasan B, Karrasch M, Whittaker S, Morris S, Weichenthal M, Stadler R, Bagot M, Cozzio A, Bernengo MG, Knobler R (2012) Prospective international multicenter phase II trial of intravenous pegylated liposomal doxorubicin monochemotherapy in patients with stage IIB, IVA, or IVB advanced mycosis fungoides: final results from EORTC 21012. J Clin Oncol 30: 4091-4097  Duvic M, Chiao N, Talpur R (2003) Extracorporeal photopheresis for the treatment of cutaneous T-cell lymphoma. J Cutan Med Surg 7: 3-7  Duvic M, Donato M, Dabaja B, Richmond H, Singh L, Wei W, Acholonu S, Khouri I, Champlin R, Hosing C (2010) Total skin electron beam and non-myeloablative allogeneic hematopoietic stem-cell transplantation in advanced mycosis fungoides and Sezary syndrome. J Clin Oncol 28: 2365-2372  Duvic M, Hester JP, Lemak NA (1996) Photopheresis therapy for cutaneous T-cell lymphoma. J Am Acad Dermatol 35: 573-579  Duvic M, Talpur R, Wen S, Kurzrock R, David CL, Apisarnthanarax N (2006) Phase II evaluation of gemcitabine monotherapy for cutaneous T-cell lymphoma. Clin Lymphoma Myeloma 7: 51-58  131  Easton DF, Pooley KA, Dunning AM, Pharoah PD, Thompson D, Ballinger DG, Struewing JP, Morrison J, Field H, Luben R, Wareham N, Ahmed S, Healey CS, Bowman R, collaborators S, Meyer KB, Haiman CA, Kolonel LK, Henderson BE, Le Marchand L, Brennan P, Sangrajrang S, Gaborieau V, Odefrey F, Shen CY, Wu PE, Wang HC, Eccles D, Evans DG, Peto J, Fletcher O, Johnson N, Seal S, Stratton MR, Rahman N, Chenevix-Trench G, Bojesen SE, Nordestgaard BG, Axelsson CK, Garcia-Closas M, Brinton L, Chanock S, Lissowska J, Peplonska B, Nevanlinna H, Fagerholm R, Eerola H, Kang D, Yoo KY, Noh DY, Ahn SH, Hunter DJ, Hankinson SE, Cox DG, Hall P, Wedren S, Liu J, Low YL, Bogdanova N, Schurmann P, Dork T, Tollenaar RA, Jacobi CE, Devilee P, Klijn JG, Sigurdson AJ, Doody MM, Alexander BH, Zhang J, Cox A, Brock IW, MacPherson G, Reed MW, Couch FJ, Goode EL, Olson JE, Meijers-Heijboer H, van den Ouweland A, Uitterlinden A, Rivadeneira F, Milne RL, Ribas G, Gonzalez-Neira A, Benitez J, Hopper JL, McCredie M, Southey M, Giles GG, Schroen C, Justenhoven C, Brauch H, Hamann U, Ko YD, Spurdle AB, Beesley J, Chen X, kConFab, Group AM, Mannermaa A, Kosma VM, Kataja V, Hartikainen J, Day NE, Cox DR, Ponder BA (2007a) Genome-wide association study identifies novel breast cancer susceptibility loci. Nature 447: 1087-1093  Easton DF, Pooley KA, Dunning AM, Pharoah PD, Thompson D, Ballinger DG, Struewing JP, Morrison J, Field H, Luben R, Wareham N, Ahmed S, Healey CS, Bowman R, Meyer KB, Haiman CA, Kolonel LK, Henderson BE, Le Marchand L, Brennan P, Sangrajrang S, Gaborieau V, Odefrey F, Shen CY, Wu PE, Wang HC, Eccles D, Evans DG, Peto J, Fletcher O, Johnson N, Seal S, Stratton MR, Rahman N, Chenevix-Trench G, Bojesen SE, Nordestgaard BG, Axelsson CK, Garcia-Closas M, Brinton L, Chanock S, Lissowska J, Peplonska B, Nevanlinna H, Fagerholm R, Eerola H, Kang D, Yoo KY, Noh DY, Ahn SH, Hunter DJ, Hankinson SE, Cox DG, Hall P, Wedren S, Liu J, Low YL, Bogdanova N, Schurmann P, Dork T, Tollenaar RA, Jacobi CE, Devilee P, Klijn JG, Sigurdson AJ, Doody MM, Alexander BH, Zhang J, Cox A, Brock IW, MacPherson G, Reed MW, Couch FJ, Goode EL, Olson JE, Meijers-Heijboer H, van den Ouweland A, Uitterlinden A, Rivadeneira F, Milne RL, Ribas G, Gonzalez-Neira A, Benitez J, Hopper JL, McCredie M, Southey M, Giles GG, Schroen C, Justenhoven C, Brauch H, Hamann U, Ko YD, Spurdle AB, Beesley J, Chen X, Mannermaa A, Kosma VM, Kataja V, Hartikainen J, Day NE, Cox DR, Ponder BA (2007b) Genome-wide association study identifies novel breast cancer susceptibility loci. Nature 447: 1087-1093  Edelson RL (1999) Sezary syndrome, cutaneous T-cell lymphoma, and extracorporeal photopheresis. Arch Dermatol 135: 600-601  Esmailzadeh S, Huang Y, Su MW, Zhou Y, Jiang X (2015) BIN1 tumor suppressor regulates Fas/Fas ligand-mediated apoptosis through c-FLIP in cutaneous T-cell lymphoma. Leukemia 29: 1402-1413  Evans AV, Scarisbrick JJ, Child FJ, Acland KM, Whittaker SJ, Russell-Jones R (2004) Cutaneous malignant melanoma in association with mycosis fungoides. J Am Acad Dermatol 50: 701-705   132 Ferrara G, Di Blasi A, Zalaudek I, Argenziano G, Cerroni L (2008) Regarding the algorithm for the diagnosis of early mycosis fungoides proposed by the International Society for Cutaneous Lymphomas: suggestions from routine histopathology practice. J Cutan Pathol 35: 549-553  Fierro MT, Ponti R, Titli S, Bonello L, Comessatti A, Novelli M, Fava P, di Celle PF, Quaglino P, Bernengo MG (2010) TCRgamma-chain gene rearrangement by GeneScan: incidence and significance of clonal heterogeneity in Sezary syndrome. J Invest Dermatol 130: 2312-2319  Fiorentino DF, Bond MW, Mosmann TR (1989) Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Th1 clones. J Exp Med 170: 2081-2095  Fletcher S, Prochownik EV (2015) Small-molecule inhibitors of the Myc oncoprotein. Biochim Biophys Acta 1849: 525-543  Fornari F, Gramantieri L, Ferracin M, Veronese A, Sabbioni S, Calin GA, Grazi GL, Giovannini C, Croce CM, Bolondi L, Negrini M (2008) MiR-221 controls CDKN1C/p57 and CDKN1B/p27 expression in human hepatocellular carcinoma. Oncogene 27: 5651-5661  Fraser-Andrews EA, Woolford AJ, Russell-Jones R, Seed PT, Whittaker SJ (2000) Detection of a peripheral blood T cell clone is an independent prognostic marker in mycosis fungoides. J Invest Dermatol 114: 117-121  Gallardo F, Esteller M, Pujol RM, Costa C, Estrach T, Servitje O (2004) Methylation status of the p15, p16 and MGMT promoter genes in primary cutaneous T-cell lymphomas. Haematologica 89: 1401-1403  Gao L, Zhang L, Hu J, Li F, Shao Y, Zhao D, Kalvakolanu DV, Kopecko DJ, Zhao X, Xu DQ (2005) Down-regulation of signal transducer and activator of transcription 3 expression using vector-based small interfering RNAs suppresses growth of human prostate tumor in vivo. Clin Cancer Res 11: 6333-6341  Garzon R, Marcucci G, Croce CM (2010) Targeting microRNAs in cancer: rationale, strategies and challenges. Nature reviews Drug discovery 9: 775-789  Gibson HM, Mishra A, Chan DV, Hake TS, Porcu P, Wong HK (2013) Impaired proteasome function activates GATA3 in T cells and upregulates CTLA-4: relevance for Sezary syndrome. J Invest Dermatol 133: 249-257  Girardi M, Heald PW, Wilson LD (2004) The pathogenesis of mycosis fungoides. N Engl J Med 350: 1978-1988  Glusac EJ (2003) Criterion by criterion, mycosis fungoides. Am J Dermatopathol 25: 264-269  133  Godfrey DI, Kennedy J, Suda T, Zlotnik A (1993) A developmental pathway involving four phenotypically and functionally distinct subsets of CD3-CD4-CD8- triple-negative adult mouse thymocytes defined by CD44 and CD25 expression. Journal of immunology 150: 4244-4252  Green DR, Droin N, Pinkoski M (2003) Activation-induced cell death in T cells. Immunological reviews 193: 70-81  Guo J, Verma UN, Gaynor RB, Frenkel EP, Becerra CR (2004) Enhanced chemosensitivity to irinotecan by RNA interference-mediated down-regulation of the nuclear factor-kappaB p65 subunit. Clin Cancer Res 10: 3333-3341  Harmon CB, Witzig TE, Katzmann JA, Pittelkow MR (1996) Detection of circulating T cells with CD4+CD7- immunophenotype in patients with benign and malignant lymphoproliferative dermatoses. J Am Acad Dermatol 35: 404-410  Hartenstein B, Teurich S, Hess J, Schenkel J, Schorpp-Kistner M, Angel P (2002) Th2 cell-specific cytokine expression and allergen-induced airway inflammation depend on JunB. EMBO J 21: 6321-6329  Hayes J, Peruzzi PP, Lawler S (2014) MicroRNAs in cancer: biomarkers, functions and therapy. Trends in molecular medicine 20: 460-469  Herne KL, Talpur R, Breuer-McHam J, Champlin R, Duvic M (2003) Cytomegalovirus seropositivity is significantly associated with mycosis fungoides and Sezary syndrome. Blood 101: 2132-2136  Herold KC, Burton JB, Francois F, Poumian-Ruiz E, Glandt M, Bluestone JA (2003) Activation of human T cells by FcR nonbinding anti-CD3 mAb, hOKT3gamma1(Ala-Ala). J Clin Invest 111: 409-418  Ho IC, Tai TS, Pai SY (2009) GATA3 and the T-cell lineage: essential functions before and after T-helper-2-cell differentiation. Nat Rev Immunol 9: 125-135  Huang da W, Sherman BT, Lempicki RA (2009a) Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 37: 1-13  Huang da W, Sherman BT, Lempicki RA (2009b) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4: 44-57  Huang Y, Litvinov IV, Wang Y, Su MW, Tu P, Jiang X, Kupper TS, Dutz JP, Sasseville D, Zhou Y (2014) Thymocyte selection-associated high mobility group box gene (TOX) is aberrantly over-expressed in mycosis fungoides and correlates with poor prognosis. Oncotarget 5: 4418-4425   134 Huang Y, Su MW, Jiang X, Zhou Y (2015) Evidence of an oncogenic role of aberrant TOX activation in cutaneous T-cell lymphoma. Blood 125: 1435-1443  Hwang ST, Janik JE, Jaffe ES, Wilson WH (2008) Mycosis fungoides and Sezary syndrome. Lancet 371: 945-957  Imam MH, Shenoy PJ, Flowers CR, Phillips A, Lechowicz MJ (2013) Incidence and survival patterns of cutaneous T-cell lymphomas in the United States. Leuk Lymphoma 54: 752-759  Iversen PL, Arora V, Acker AJ, Mason DH, Devi GR (2003) Efficacy of antisense morpholino oligomer targeted to c-myc in prostate cancer xenograft murine model and a Phase I safety study in humans. Clin Cancer Res 9: 2510-2519  Izbicka E, Diaz A, Streeper R, Wick M, Campos D, Steffen R, Saunders M (2009) Distinct mechanistic activity profile of pralatrexate in comparison to other antifolates in in vitro and in vivo models of human cancers. Cancer Chemother Pharmacol 64: 993-999  Izykowska K, Przybylski GK (2011) Genetic alterations in Sezary syndrome. Leuk Lymphoma 52: 745-753  Jawed SI, Myskowski PL, Horwitz S, Moskowitz A, Querfeld C (2014a) Primary cutaneous T-cell lymphoma (mycosis fungoides and Sezary syndrome): part I. Diagnosis: clinical and histopathologic features and new molecular and biologic markers. J Am Acad Dermatol 70: 205 e201-216; quiz 221-202  Jawed SI, Myskowski PL, Horwitz S, Moskowitz A, Querfeld C (2014b) Primary cutaneous T-cell lymphoma (mycosis fungoides and Sezary syndrome): part II. Prognosis, management, and future directions. J Am Acad Dermatol 70: 223 e221-217; quiz 240-222  Jones CL, Wain EM, Chu CC, Tosi I, Foster R, McKenzie RC, Whittaker SJ, Mitchell TJ (2010) Downregulation of Fas gene expression in Sezary syndrome is associated with promoter hypermethylation. J Invest Dermatol 130: 1116-1125  Jones D, Dang NH, Duvic M, Washington LT, Huh YO (2001) Absence of CD26 expression is a useful marker for diagnosis of T-cell lymphoma in peripheral blood. Am J Clin Pathol 115: 885-892  Kaech SM, Wherry EJ, Ahmed R (2002) Effector and memory T-cell differentiation: implications for vaccine development. Nature reviews Immunology 2: 251-262  Kajitani T, Mizutani T, Yamada K, Yazawa T, Sekiguchi T, Yoshino M, Kawata H, Miyamoto K (2004) Cloning and characterization of granulosa cell high-mobility group (HMG)-box protein-1, a novel HMG-box transcriptional regulator strongly expressed in rat ovarian granulosa cells. Endocrinology 145: 2307-2318   135 Kallinich T, Muche JM, Qin S, Sterry W, Audring H, Kroczek RA (2003) Chemokine receptor expression on neoplastic and reactive T cells in the skin at different stages of mycosis fungoides. J Invest Dermatol 121: 1045-1052  Kamstrup MR, Gjerdrum LM, Biskup E, Lauenborg BT, Ralfkiaer E, Woetmann A, Odum N, Gniadecki R (2010) Notch1 as a potential therapeutic target in cutaneous T-cell lymphoma. Blood 116: 2504-2512  Kari L, Loboda A, Nebozhyn M, Rook AH, Vonderheid EC, Nichols C, Virok D, Chang C, Horng WH, Johnston J, Wysocka M, Showe MK, Showe LC (2003) Classification and prediction of survival in patients with the leukemic phase of cutaneous T cell lymphoma. J Exp Med 197: 1477-1488  Karube K, Nakagawa M, Tsuzuki S, Takeuchi I, Honma K, Nakashima Y, Shimizu N, Ko YH, Morishima Y, Ohshima K, Nakamura S, Seto M (2011) Identification of FOXO3 and PRDM1 as tumor-suppressor gene candidates in NK-cell neoplasms by genomic and functional analyses. Blood 118: 3195-3204  Keehn CA, Belongie IP, Shistik G, Fenske NA, Glass LF (2007) The diagnosis, staging, and treatment options for mycosis fungoides. Cancer Control 14: 102-111  Kennah E, Ringrose A, Zhou LL, Esmailzadeh S, Qian H, Su MW, Zhou Y, Jiang X (2009) Identification of tyrosine kinase, HCK, and tumor suppressor, BIN1, as potential mediators of AHI-1 oncogene in primary and transformed CTCL cells. Blood 113: 4646-4655  Kiel MJ, Sahasrabuddhe AA, Rolland DC, Velusamy T, Chung F, Schaller M, Bailey NG, Betz BL, Miranda RN, Porcu P, Byrd JC, Jeffrey Medeiros L, Kunkel SL, Bahler DW, Lim MS, Elenitoba-Johnson KS (2015) Genomic analyses reveal recurrent mutations in epigenetic modifiers and the JAK-STAT pathway in Sezary syndrome. Nature communications 6: 8470  Kiessling MK, Oberholzer PA, Mondal C, Karpova MB, Zipser MC, Lin WM, Girardi M, Macconaill LE, Kehoe SM, Hatton C, French LE, Garraway LA, Polier G, Suss D, Klemke CD, Krammer PH, Gulow K, Dummer R (2011) High-throughput mutation profiling of CTCL samples reveals KRAS and NRAS mutations sensitizing tumors toward inhibition of the RAS/RAF/MEK signaling cascade. Blood 117: 2433-2440  Kim EJ, Hess S, Richardson SK, Newton S, Showe LC, Benoit BM, Ubriani R, Vittorio CC, Junkins-Hopkins JM, Wysocka M, Rook AH (2005) Immunopathogenesis and therapy of cutaneous T cell lymphoma. J Clin Invest 115: 798-812  Kim YH, Bishop K, Varghese A, Hoppe RT (1995) Prognostic factors in erythrodermic mycosis fungoides and the Sezary syndrome. Arch Dermatol 131: 1003-1008   136 Kim YH, Chow S, Varghese A, Hoppe RT (1999) Clinical characteristics and long-term outcome of patients with generalized patch and/or plaque (T2) mycosis fungoides. Arch Dermatol 135: 26-32  Kim YH, Girardi M, Duvic M, Kuzel T, Link BK, Pinter-Brown L, Rook AH (2010) Phase I trial of a Toll-like receptor 9 agonist, PF-3512676 (CPG 7909), in patients with treatment-refractory, cutaneous T-cell lymphoma. J Am Acad Dermatol 63: 975-983  Kim YH, Liu HL, Mraz-Gernhard S, Varghese A, Hoppe RT (2003a) Long-term outcome of 525 patients with mycosis fungoides and Sezary syndrome: clinical prognostic factors and risk for disease progression. Arch Dermatol 139: 857-866  Kim YH, Martinez G, Varghese A, Hoppe RT (2003b) Topical nitrogen mustard in the management of mycosis fungoides: update of the Stanford experience. Arch Dermatol 139: 165-173  Kim YK, Yu J, Han TS, Park SY, Namkoong B, Kim DH, Hur K, Yoo MW, Lee HJ, Yang HK, Kim VN (2009) Functional links between clustered microRNAs: suppression of cell-cycle inhibitors by microRNA clusters in gastric cancer. Nucleic Acids Res 37: 1672-1681  Klein L, Hinterberger M, Wirnsberger G, Kyewski B (2009) Antigen presentation in the thymus for positive selection and central tolerance induction. Nature reviews Immunology 9: 833-844  Klemke CD, Brenner D, Weiss EM, Schmidt M, Leverkus M, Gulow K, Krammer PH (2009) Lack of T-cell receptor-induced signaling is crucial for CD95 ligand up-regulation and protects cutaneous T-cell lymphoma cells from activation-induced cell death. Cancer Res 69: 4175-4183  Knobler R, Duvic M, Querfeld C, Straus D, Horwitz S, Zain J, Foss F, Kuzel T, Campbell K, Geskin L (2012) Long-term follow-up and survival of cutaneous T-cell lymphoma patients treated with extracorporeal photopheresis. Photodermatol Photoimmunol Photomed 28: 250-257  Kopp KL, Ralfkiaer U, Gjerdrum LM, Helvad R, Pedersen IH, Litman T, Jonson L, Hagedorn PH, Krejsgaard T, Gniadecki R, Bonefeld CM, Skov L, Geisler C, Wasik MA, Ralfkiaer E, Odum N, Woetmann A (2013) STAT5-mediated expression of oncogenic miR-155 in cutaneous T-cell lymphoma. Cell cycle 12: 1939-1947  Krejsgaard T, Kopp K, Ralfkiaer E, Willumsgaard AE, Eriksen KW, Labuda T, Rasmussen S, Mathiesen AM, Geisler C, Lauenborg B, Becker JC, Zhang Q, Wasik MA, Odum N, Woetmann A (2010) A novel xenograft model of cutaneous T-cell lymphoma. Exp Dermatol 19: 1096-1102   137 Krejsgaard T, Litvinov IV, Wang Y, Xia L, Willerslev-Olsen A, Koralov SB, Kopp KL, Bonefeld CM, Wasik MA, Geisler C, Woetmann A, Zhou Y, Sasseville D, Odum N (2013) Elucidating the role of interleukin-17F in cutaneous T-cell lymphoma. Blood 122: 943-950  Krejsgaard T, Vetter-Kauczok CS, Woetmann A, Kneitz H, Eriksen KW, Lovato P, Zhang Q, Wasik MA, Geisler C, Ralfkiaer E, Becker JC, Odum N (2009) Ectopic expression of B-lymphoid kinase in cutaneous T-cell lymphoma. Blood 113: 5896-5904  Kummerfeld SK, Teichmann SA (2006) DBD: a transcription factor prediction database. Nucleic Acids Res 34: D74-81  Laharanne E, Oumouhou N, Bonnet F, Carlotti M, Gentil C, Chevret E, Jouary T, Longy M, Vergier B, Beylot-Barry M, Merlio JP (2010) Genome-wide analysis of cutaneous T-cell lymphomas identifies three clinically relevant classes. J Invest Dermatol 130: 1707-1718  Lamberg SI, Bunn PA, Jr. (1979) Cutaneous T-cell lymphomas. Summary of the Mycosis Fungoides Cooperative Group-National Cancer Institute Workshop. Arch Dermatol 115: 1103-1105  Larkin J, Chiarion-Sileni V, Gonzalez R, Grob JJ, Cowey CL, Lao CD, Schadendorf D, Dummer R, Smylie M, Rutkowski P, Ferrucci PF, Hill A, Wagstaff J, Carlino MS, Haanen JB, Maio M, Marquez-Rodas I, McArthur GA, Ascierto PA, Long GV, Callahan MK, Postow MA, Grossmann K, Sznol M, Dreno B, Bastholt L, Yang A, Rollin LM, Horak C, Hodi FS, Wolchok JD (2015) Combined Nivolumab and Ipilimumab or Monotherapy in Untreated Melanoma. N Engl J Med 373: 23-34  Larosa DF, Orange JS (2008) 1. Lymphocytes. The Journal of allergy and clinical immunology 121: S364-369; quiz S412  Lauenborg B, Kopp K, Krejsgaard T, Eriksen KW, Geisler C, Dabelsteen S, Gniadecki R, Zhang Q, Wasik MA, Woetmann A, Odum N (2010) Programmed cell death-10 enhances proliferation and protects malignant T cells from apoptosis. APMIS 118: 719-728  Lauritsen JP, Haks MC, Lefebvre JM, Kappes DJ, Wiest DL (2006) Recent insights into the signals that control alphabeta/gammadelta-lineage fate. Immunological reviews 209: 176-190  Lee CS, Ungewickell A, Bhaduri A, Qu K, Webster DE, Armstrong R, Weng WK, Aros CJ, Mah A, Chen RO, Lin M, Sundram U, Chang HY, Kretz M, Kim YH, Khavari PA (2012) Transcriptome sequencing in Sezary syndrome identifies Sezary cell and mycosis fungoides-associated lncRNAs and novel transcripts. Blood 120: 3288-3297  Lee JH, You J, Dobrota E, Skalnik DG (2010) Identification and characterization of a novel human PP1 phosphatase complex. The Journal of biological chemistry 285: 24466-24476  Lenardo MJ (1991) Interleukin-2 programs mouse alpha beta T lymphocytes for apoptosis. Nature 353: 858-861  138  Letterio JJ (2005) TGF-beta signaling in T cells: roles in lymphoid and epithelial neoplasia. Oncogene 24: 5701-5712  Leupin N, Kuhn A, Hugli B, Grob TJ, Jaggi R, Tobler A, Delorenzi M, Fey MF (2006) Gene expression profiling reveals consistent differences between clinical samples of human leukaemias and their model cell lines. Br J Haematol 135: 520-523  Li MO, Flavell RA (2008) TGF-beta: a master of all T cell trades. Cell 134: 392-404  Lim MS, Adamson A, Lin Z, Perez-Ordonez B, Jordan RC, Tripp S, Perkins SL, Elenitoba-Johnson KS (2002) Expression of Skp2, a p27(Kip1) ubiquitin ligase, in malignant lymphoma: correlation with p27(Kip1) and proliferation index. Blood 100: 2950-2956  Lindahl LM, Fenger-Gron M, Iversen L (2013) Topical nitrogen mustard therapy in patients with mycosis fungoides or parapsoriasis. J Eur Acad Dermatol Venereol 27: 163-168  Litvinov IV, Jones DA, Sasseville D, Kupper TS (2010) Transcriptional profiles predict disease outcome in patients with cutaneous T-cell lymphoma. Clin Cancer Res 16: 2106-2114  Litvinov IV, Kupper TS, Sasseville D (2012) The role of AHI1 and CDKN1C in cutaneous T-cell lymphoma progression. Exp Dermatol 21: 964-966  Litvinov IV, Zhou Y, Kupper TS, Sasseville D (2013) Loss of BCL7A expression correlates with poor disease prognosis in patients with early-stage cutaneous T-cell lymphoma. Leuk Lymphoma 54: 653-654  Lloyd RV, Erickson LA, Jin L, Kulig E, Qian X, Cheville JC, Scheithauer BW (1999) p27kip1: a multifunctional cyclin-dependent kinase inhibitor with prognostic significance in human cancers. Am J Pathol 154: 313-323  Lobbardi R, Pinder J, Martinez-Pastor B, Blackburn JS, Abdelfattah N, Toiber D, De Waard M, Asara JM, Mostoslavsky R, Dellaire G, Langenau DM (2014) Thymocyte Selection-Associated High-Mobility Group Box Protein (TOX) Induces Genomic Instability in T-Cell Acute Lymphoblastic Leukemia. Blood 124: 475-475  Lotem J, Sachs L (1998) Different mechanisms for suppression of apoptosis by cytokines and calcium mobilizing compounds. Proceedings of the National Academy of Sciences of the United States of America 95: 4601-4606  Luckheeram RV, Zhou R, Verma AD, Xia B (2012) CD4(+)T cells: differentiation and functions. Clinical & developmental immunology 2012: 925135   139 Lukowsky A, Muche JM, Sterry W, Audring H (2000) Detection of expanded T cell clones in skin biopsy samples of patients with lichen sclerosus et atrophicus by T cell receptor-gamma polymerase chain reaction assays. J Invest Dermatol 115: 254-259  Mackay LK, Rahimpour A, Ma JZ, Collins N, Stock AT, Hafon ML, Vega-Ramos J, Lauzurica P, Mueller SN, Stefanovic T, Tscharke DC, Heath WR, Inouye M, Carbone FR, Gebhardt T (2013) The developmental pathway for CD103(+)CD8+ tissue-resident memory T cells of skin. Nat Immunol 14: 1294-1301  Manfe V, Biskup E, Rosbjerg A, Kamstrup M, Skov AG, Lerche CM, Lauenborg BT, Odum N, Gniadecki R (2012) miR-122 regulates p53/Akt signalling and the chemotherapy-induced apoptosis in cutaneous T-cell lymphoma. PloS one 7: e29541  Mao X, Orchard G, Lillington DM, Child FJ, Vonderheid EC, Nowell PC, Bagot M, Bensussan A, Russell-Jones R, Young BD, Whittaker SJ (2004) BCL2 and JUNB abnormalities in primary cutaneous lymphomas. Br J Dermatol 151: 546-556  Mao X, Orchard G, Mitchell TJ, Oyama N, Russell-Jones R, Vermeer MH, Willemze R, van Doorn R, Tensen CP, Young BD, Whittaker SJ (2008) A genomic and expression study of AP-1 in primary cutaneous T-cell lymphoma: evidence for dysregulated expression of JUNB and JUND in MF and SS. J Cutan Pathol 35: 899-910  Mao X, Orchard G, Vonderheid EC, Nowell PC, Bagot M, Bensussan A, Russell-Jones R, Young BD, Whittaker SJ (2006) Heterogeneous abnormalities of CCND1 and RB1 in primary cutaneous T-Cell lymphomas suggesting impaired cell cycle control in disease pathogenesis. J Invest Dermatol 126: 1388-1395  Marie I, Cordel N, Lenormand B, Hellot MF, Levesque H, Courtois H, Joly P (2005) Clonal T cells in the blood of patients with systemic sclerosis. Arch Dermatol 141: 88-89  Maser RS, Choudhury B, Campbell PJ, Feng B, Wong KK, Protopopov A, O'Neil J, Gutierrez A, Ivanova E, Perna I, Lin E, Mani V, Jiang S, McNamara K, Zaghlul S, Edkins S, Stevens C, Brennan C, Martin ES, Wiedemeyer R, Kabbarah O, Nogueira C, Histen G, Aster J, Mansour M, Duke V, Foroni L, Fielding AK, Goldstone AH, Rowe JM, Wang YA, Look AT, Stratton MR, Chin L, Futreal PA, DePinho RA (2007) Chromosomally unstable mouse tumours have genomic alterations similar to diverse human cancers. Nature 447: 966-971  Masopust D, Vezys V, Marzo AL, Lefrancois L (2001) Preferential localization of effector memory cells in nonlymphoid tissue. Science 291: 2413-2417  McGirt LY, Adams CM, Baerenwald DA, Zwerner JP, Zic JA, Eischen CM (2014) miR-223 Regulates Cell Growth and Targets Proto-Oncogenes in Mycosis Fungoides/Cutaneous T-Cell Lymphoma. J Invest Dermatol 134: 1101-1107   140 McGirt LY, Jia P, Baerenwald DA, Duszynski RJ, Dahlman KB, Zic JA, Zwerner JP, Hucks D, Dave U, Zhao Z, Eischen CM (2015) Whole-genome sequencing reveals oncogenic mutations in mycosis fungoides. Blood 126: 508-519  Mohr B, Illmer T, Oelschlagel U, Nowak R, Holig K, Paaz U, Kroschinsky F, Schuler U, Ehninger G (1996) Complex cytogenetic and immunophenotypic aberrations in a patient with Sezary syndrome. Cancer genetics and cytogenetics 90: 33-36  Molin L, Thomsen K, Volden G (1978) Serum IgE in mycosis fungoides. British medical journal 1: 920-921  Moore KW, de Waal Malefyt R, Coffman RL, O'Garra A (2001) Interleukin-10 and the interleukin-10 receptor. Annual review of immunology 19: 683-765  Morales-Suarez-Varela MM, Olsen J, Johansen P, Kaerlev L, Guenel P, Arveux P, Wingren G, Hardell L, Ahrens W, Stang A, Llopis A, Merletti F, Aurrekoetxea JJ, Masala G (2004) Occupational risk factors for mycosis fungoides: a European multicenter case-control study. Journal of occupational and environmental medicine / American College of Occupational and Environmental Medicine 46: 205-211  Murphy M, Fullen D, Carlson JA (2002) Low CD7 expression in benign and malignant cutaneous lymphocytic infiltrates: experience with an antibody reactive with paraffin-embedded tissue. Am J Dermatopathol 24: 6-16  Nagase T, Ishikawa K, Suyama M, Kikuno R, Miyajima N, Tanaka A, Kotani H, Nomura N, Ohara O (1998) Prediction of the coding sequences of unidentified human genes. XI. The complete sequences of 100 new cDNA clones from brain which code for large proteins in vitro. DNA research : an international journal for rapid publication of reports on genes and genomes 5: 277-286  Naito T, Tanaka H, Naoe Y, Taniuchi I (2011) Transcriptional control of T-cell development. Int Immunol 23: 661-668  Narducci MG, Arcelli D, Picchio MC, Lazzeri C, Pagani E, Sampogna F, Scala E, Fadda P, Cristofoletti C, Facchiano A, Frontani M, Monopoli A, Ferracin M, Negrini M, Lombardo GA, Caprini E, Russo G (2011) MicroRNA profiling reveals that miR-21, miR486 and miR-214 are upregulated and involved in cell survival in Sezary syndrome. Cell Death Dis 2: e151  Navas IC, Ortiz-Romero PL, Villuendas R, Martinez P, Garcia C, Gomez E, Rodriguez JL, Garcia D, Vanaclocha F, Iglesias L, Piris MA, Algara P (2000) p16(INK4a) gene alterations are frequent in lesions of mycosis fungoides. The American journal of pathology 156: 1565-1572   141 Ni X, Hazarika P, Zhang C, Talpur R, Duvic M (2001) Fas ligand expression by neoplastic T lymphocytes mediates elimination of CD8+ cytotoxic T lymphocytes in mycosis fungoides: a potential mechanism of tumor immune escape? Clin Cancer Res 7: 2682-2692  Nihal M, Ahmad N, Wood GS (2014) SIRT1 is upregulated in cutaneous T-cell lymphoma, and its inhibition induces growth arrest and apoptosis. Cell Cycle 13: 632-640  Nikolova M, Musette P, Bagot M, Boumsell L, Bensussan A (2002) Engagement of ILT2/CD85j in Sezary syndrome cells inhibits their CD3/TCR signaling. Blood 100: 1019-1025  Nishikomori R, Usui T, Wu CY, Morinobu A, O'Shea JJ, Strober W (2002) Activated STAT4 has an essential role in Th1 differentiation and proliferation that is independent of its role in the maintenance of IL-12R beta 2 chain expression and signaling. Journal of immunology 169: 4388-4398  O'Flaherty E, Kaye J (2003) TOX defines a conserved subfamily of HMG-box proteins. BMC genomics 4: 13  Okhrimenko A, Grun JR, Westendorf K, Fang Z, Reinke S, von Roth P, Wassilew G, Kuhl AA, Kudernatsch R, Demski S, Scheibenbogen C, Tokoyoda K, McGrath MA, Raftery MJ, Schonrich G, Serra A, Chang HD, Radbruch A, Dong J (2014) Human memory T cells from the bone marrow are resting and maintain long-lasting systemic memory. Proceedings of the National Academy of Sciences of the United States of America 111: 9229-9234  Olsen E, Vonderheid E, Pimpinelli N, Willemze R, Kim Y, Knobler R, Zackheim H, Duvic M, Estrach T, Lamberg S, Wood G, Dummer R, Ranki A, Burg G, Heald P, Pittelkow M, Bernengo MG, Sterry W, Laroche L, Trautinger F, Whittaker S (2007) Revisions to the staging and classification of mycosis fungoides and Sezary syndrome: a proposal of the International Society for Cutaneous Lymphomas (ISCL) and the cutaneous lymphoma task force of the European Organization of Research and Treatment of Cancer (EORTC). Blood 110: 1713-1722  Olsen EA, Rook AH, Zic J, Kim Y, Porcu P, Querfeld C, Wood G, Demierre MF, Pittelkow M, Wilson LD, Pinter-Brown L, Advani R, Parker S, Kim EJ, Junkins-Hopkins JM, Foss F, Cacchio P, Duvic M (2011) Sezary syndrome: immunopathogenesis, literature review of therapeutic options, and recommendations for therapy by the United States Cutaneous Lymphoma Consortium (USCLC). J Am Acad Dermatol 64: 352-404  Ortonne N, Huet D, Gaudez C, Marie-Cardine A, Schiavon V, Bagot M, Musette P, Bensussan A (2006) Significance of circulating T-cell clones in Sezary syndrome. Blood 107: 4030-4038  Ortonne N, Le Gouvello S, Tabak R, Marie-Cardine A, Setiao J, Berrehar F, Nghe-Tang A, Martin N, Bagot M, Bensussan A (2012) CD158k/KIR3DL2 and NKp46 are frequently expressed in transformed mycosis fungoides. Exp Dermatol 21: 461-463  142  Parry RV, Chemnitz JM, Frauwirth KA, Lanfranco AR, Braunstein I, Kobayashi SV, Linsley PS, Thompson CB, Riley JL (2005) CTLA-4 and PD-1 receptors inhibit T-cell activation by distinct mechanisms. Molecular and cellular biology 25: 9543-9553  Paul WE, Seder RA (1994) Lymphocyte responses and cytokines. Cell 76: 241-251  Paulson KE, Rieger-Christ K, McDevitt MA, Kuperwasser C, Kim J, Unanue VE, Zhang X, Hu M, Ruthazer R, Berasi SP, Huang CY, Giri D, Kaufman S, Dugan JM, Blum J, Netto G, Wazer DE, Summerhayes IC, Yee AS (2007) Alterations of the HBP1 transcriptional repressor are associated with invasive breast cancer. Cancer Res 67: 6136-6145  Pimpinelli N, Olsen EA, Santucci M, Vonderheid E, Haeffner AC, Stevens S, Burg G, Cerroni L, Dreno B, Glusac E, Guitart J, Heald PW, Kempf W, Knobler R, Lessin S, Sander C, Smoller BS, Telang G, Whittaker S, Iwatsuki K, Obitz E, Takigawa M, Turner ML, Wood GS (2005) Defining early mycosis fungoides. J Am Acad Dermatol 53: 1053-1063  Poszepczynska-Guigne E, Schiavon V, D'Incan M, Echchakir H, Musette P, Ortonne N, Boumsell L, Moretta A, Bensussan A, Bagot M (2004) CD158k/KIR3DL2 is a new phenotypic marker of Sezary cells: relevance for the diagnosis and follow-up of Sezary syndrome. J Invest Dermatol 122: 820-823  Purwar R, Campbell J, Murphy G, Richards WG, Clark RA, Kupper TS (2011) Resident memory T cells (T(RM)) are abundant in human lung: diversity, function, and antigen specificity. PloS one 6: e16245  Querfeld C, Nagelli LV, Rosen ST, Kuzel TM, Guitart J (2006) Bexarotene in the treatment of cutaneous T-cell lymphoma. Expert Opin Pharmacother 7: 907-915  Ralfkiaer U, Hagedorn PH, Bangsgaard N, Lovendorf MB, Ahler CB, Svensson L, Kopp KL, Vennegaard MT, Lauenborg B, Zibert JR, Krejsgaard T, Bonefeld CM, Sokilde R, Gjerdrum LM, Labuda T, Mathiesen AM, Gronbaek K, Wasik MA, Sokolowska-Wojdylo M, Queille-Roussel C, Gniadecki R, Ralfkiaer E, Geisler C, Litman T, Woetmann A, Glue C, Ropke MA, Skov L, Odum N (2011) Diagnostic microRNA profiling in cutaneous T-cell lymphoma (CTCL). Blood 118: 5891-5900  Ravat FE, Spittle MF, Russell-Jones R (2006) Primary cutaneous T-cell lymphoma occurring after organ transplantation. J Am Acad Dermatol 54: 668-675  Redell MS, Tweardy DJ (2006) Targeting transcription factors in cancer: Challenges and evolving strategies. Drug Discov Today Technol 3: 261-267  Ringrose A, Zhou Y, Pang E, Zhou L, Lin AE, Sheng G, Li XJ, Weng A, Su MW, Pittelkow MR, Jiang X (2006) Evidence for an oncogenic role of AHI-1 in Sezary syndrome, a leukemic variant of human cutaneous T-cell lymphomas. Leukemia 20: 1593-1601   143 Roberts GW, Baird D, Gallagher K, Jones RE, Pepper CJ, Williams JD, Topley N (2009) Functional effector memory T cells enrich the peritoneal cavity of patients treated with peritoneal dialysis. J Am Soc Nephrol 20: 1895-1900  Robson A (2010) Immunocytochemistry and the diagnosis of cutaneous lymphoma. Histopathology 56: 71-90  Russell JH, Ley TJ (2002) Lymphocyte-mediated cytotoxicity. Annual review of immunology 20: 323-370  Saed G, Fivenson DP, Naidu Y, Nickoloff BJ (1994) Mycosis fungoides exhibits a Th1-type cell-mediated cytokine profile whereas Sezary syndrome expresses a Th2-type profile. J Invest Dermatol 103: 29-33  Salgado R, Servitje O, Gallardo F, Vermeer MH, Ortiz-Romero PL, Karpova MB, Zipser MC, Muniesa C, Garcia-Muret MP, Estrach T, Salido M, Sanchez-Schmidt J, Herrera M, Romagosa V, Suela J, Ferreira BI, Cigudosa JC, Barranco C, Serrano S, Dummer R, Tensen CP, Sole F, Pujol RM, Espinet B (2010) Oligonucleotide array-CGH identifies genomic subgroups and prognostic markers for tumor stage mycosis fungoides. J Invest Dermatol 130: 1126-1135  Samuelsson MK, Pazirandeh A, Okret S (2002) A pro-apoptotic effect of the CDK inhibitor p57(Kip2) on staurosporine-induced apoptosis in HeLa cells. Biochem Biophys Res Commun 296: 702-709  Scarisbrick JJ, Whittaker S, Evans AV, Fraser-Andrews EA, Child FJ, Dean A, Russell-Jones R (2001) Prognostic significance of tumor burden in the blood of patients with erythrodermic primary cutaneous T-cell lymphoma. Blood 97: 624-630  Scarisbrick JJ, Woolford AJ, Calonje E, Photiou A, Ferreira S, Orchard G, Russell-Jones R, Whittaker SJ (2002) Frequent abnormalities of the p15 and p16 genes in mycosis fungoides and sezary syndrome. J Invest Dermatol 118: 493-499  Schiller PI, Flaig MJ, Puchta U, Kind P, Sander CA (2000) Detection of clonal T cells in lichen planus. Arch Dermatol Res 292: 568-569  Shin J, Monti S, Aires DJ, Duvic M, Golub T, Jones DA, Kupper TS (2007) Lesional gene expression profiling in cutaneous T-cell lymphoma reveals natural clusters associated with disease outcome. Blood 110: 3015-3027  Sors A, Jean-Louis F, Pellet C, Laroche L, Dubertret L, Courtois G, Bachelez H, Michel L (2006) Down-regulating constitutive activation of the NF-kappaB canonical pathway overcomes the resistance of cutaneous T-cell lymphoma to apoptosis. Blood 107: 2354-2363  Stacey SN, Manolescu A, Sulem P, Rafnar T, Gudmundsson J, Gudjonsson SA, Masson G, Jakobsdottir M, Thorlacius S, Helgason A, Aben KK, Strobbe LJ, Albers-Akkers MT,  144 Swinkels DW, Henderson BE, Kolonel LN, Le Marchand L, Millastre E, Andres R, Godino J, Garcia-Prats MD, Polo E, Tres A, Mouy M, Saemundsdottir J, Backman VM, Gudmundsson L, Kristjansson K, Bergthorsson JT, Kostic J, Frigge ML, Geller F, Gudbjartsson D, Sigurdsson H, Jonsdottir T, Hrafnkelsson J, Johannsson J, Sveinsson T, Myrdal G, Grimsson HN, Jonsson T, von Holst S, Werelius B, Margolin S, Lindblom A, Mayordomo JI, Haiman CA, Kiemeney LA, Johannsson OT, Gulcher JR, Thorsteinsdottir U, Kong A, Stefansson K (2007) Common variants on chromosomes 2q35 and 16q12 confer susceptibility to estrogen receptor-positive breast cancer. Nat Genet 39: 865-869  Starr TK, Jameson SC, Hogquist KA (2003) Positive and negative selection of T cells. Annual review of immunology 21: 139-176  Storz M, Zepter K, Kamarashev J, Dummer R, Burg G, Haffner AC (2001) Coexpression of CD40 and CD40 ligand in cutaneous T-cell lymphoma (mycosis fungoides). Cancer Res 61: 452-454  Stritesky GL, Muthukrishnan R, Sehra S, Goswami R, Pham D, Travers J, Nguyen ET, Levy DE, Kaplan MH (2011) The transcription factor STAT3 is required for T helper 2 cell development. Immunity 34: 39-49  Su AI, Cooke MP, Ching KA, Hakak Y, Walker JR, Wiltshire T, Orth AP, Vega RG, Sapinoso LM, Moqrich A, Patapoutian A, Hampton GM, Schultz PG, Hogenesch JB (2002) Large-scale analysis of the human and mouse transcriptomes. Proceedings of the National Academy of Sciences of the United States of America 99: 4465-4470  Su MW, Dorocicz I, Dragowska WH, Ho V, Li G, Voss N, Gascoyne R, Zhou Y (2003) Aberrant expression of T-plastin in Sezary cells. Cancer Res 63: 7122-7127  Suchin KR, Cassin M, Gottleib SL, Sood S, Cucchiara AJ, Vonderheid EC, Rook AH (2001) Increased interleukin 5 production in eosinophilic Sezary syndrome: regulation by interferon alfa and interleukin 12. J Am Acad Dermatol 44: 28-32  Talpur R, Bassett R, Duvic M (2008) Prevalence and treatment of Staphylococcus aureus colonization in patients with mycosis fungoides and Sezary syndrome. Br J Dermatol 159: 105-112  Talpur R, Singh L, Daulat S, Liu P, Seyfer S, Trynosky T, Wei W, Duvic M (2012) Long-term outcomes of 1,263 patients with mycosis fungoides and Sezary syndrome from 1982 to 2009. Clin Cancer Res 18: 5051-5060  Topark-Ngarm A, Golonzhka O, Peterson VJ, Barrett B, Jr., Martinez B, Crofoot K, Filtz TM, Leid M (2006) CTIP2 associates with the NuRD complex on the promoter of p57KIP2, a newly identified CTIP2 target gene. J Biol Chem 281: 32272-32283   145 Trotter MJ, Whittaker SJ, Orchard GE, Smith NP (1997) Cutaneous histopathology of Sezary syndrome: a study of 41 cases with a proven circulating T-cell clone. J Cutan Pathol 24: 286-291  Turner DL, Bickham KL, Thome JJ, Kim CY, D'Ovidio F, Wherry EJ, Farber DL (2014) Lung niches for the generation and maintenance of tissue-resident memory T cells. Mucosal Immunol 7: 501-510  Ungewickell A, Bhaduri A, Rios E, Reuter J, Lee CS, Mah A, Zehnder A, Ohgami R, Kulkarni S, Armstrong R, Weng WK, Gratzinger D, Tavallaee M, Rook A, Snyder M, Kim Y, Khavari PA (2015) Genomic analysis of mycosis fungoides and Sezary syndrome identifies recurrent alterations in TNFR2. Nat Genet 47: 1056-1060  van der Fits L, van Kester MS, Qin Y, Out-Luiting JJ, Smit F, Zoutman WH, Willemze R, Tensen CP, Vermeer MH (2011) MicroRNA-21 expression in CD4+ T cells is regulated by STAT3 and is pathologically involved in Sezary syndrome. J Invest Dermatol 131: 762-768  van Doorn R, Dijkman R, Vermeer MH, Out-Luiting JJ, van der Raaij-Helmer EM, Willemze R, Tensen CP (2004) Aberrant expression of the tyrosine kinase receptor EphA4 and the transcription factor twist in Sezary syndrome identified by gene expression analysis. Cancer Res 64: 5578-5586  van Doorn R, Dijkman R, Vermeer MH, Starink TM, Willemze R, Tensen CP (2002) A novel splice variant of the Fas gene in patients with cutaneous T-cell lymphoma. Cancer Res 62: 5389-5392  van Doorn R, Van Haselen CW, van Voorst Vader PC, Geerts ML, Heule F, de Rie M, Steijlen PM, Dekker SK, van Vloten WA, Willemze R (2000) Mycosis fungoides: disease evolution and prognosis of 309 Dutch patients. Arch Dermatol 136: 504-510  van Doorn R, van Kester MS, Dijkman R, Vermeer MH, Mulder AA, Szuhai K, Knijnenburg J, Boer JM, Willemze R, Tensen CP (2009) Oncogenomic analysis of mycosis fungoides reveals major differences with Sezary syndrome. Blood 113: 127-136  van Doorn R, Zoutman WH, Dijkman R, de Menezes RX, Commandeur S, Mulder AA, van der Velden PA, Vermeer MH, Willemze R, Yan PS, Huang TH, Tensen CP (2005) Epigenetic profiling of cutaneous T-cell lymphoma: promoter hypermethylation of multiple tumor suppressor genes including BCL7a, PTPRG, and p73. J Clin Oncol 23: 3886-3896  van Kester MS, Ballabio E, Benner MF, Chen XH, Saunders NJ, van der Fits L, van Doorn R, Vermeer MH, Willemze R, Tensen CP, Lawrie CH (2011) miRNA expression profiling of mycosis fungoides. Molecular oncology 5: 273-280  Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic Z, Kong N, Kammlott U, Lukacs C, Klein C, Fotouhi N, Liu EA (2004) In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science 303: 844-848  146  Vega F, Luthra R, Medeiros LJ, Dunmire V, Lee SJ, Duvic M, Jones D (2002) Clonal heterogeneity in mycosis fungoides and its relationship to clinical course. Blood 100: 3369-3373  Vermeer MH, van Doorn R, Dijkman R, Mao X, Whittaker S, van Voorst Vader PC, Gerritsen MJ, Geerts ML, Gellrich S, Soderberg O, Leuchowius KJ, Landegren U, Out-Luiting JJ, Knijnenburg J, Ijszenga M, Szuhai K, Willemze R, Tensen CP (2008) Novel and highly recurrent chromosomal alterations in Sezary syndrome. Cancer Res 68: 2689-2698  Vlachos P, Nyman U, Hajji N, Joseph B (2007) The cell cycle inhibitor p57(Kip2) promotes cell death via the mitochondrial apoptotic pathway. Cell Death Differ 14: 1497-1507  Vonderheid EC, Bernengo MG, Burg G, Duvic M, Heald P, Laroche L, Olsen E, Pittelkow M, Russell-Jones R, Takigawa M, Willemze R (2002) Update on erythrodermic cutaneous T-cell lymphoma: report of the International Society for Cutaneous Lymphomas. J Am Acad Dermatol 46: 95-106  Vowels BR, Cassin M, Vonderheid EC, Rook AH (1992) Aberrant cytokine production by Sezary syndrome patients: cytokine secretion pattern resembles murine Th2 cells. J Invest Dermatol 99: 90-94  Wain EM, Orchard GE, Whittaker SJ, Spittle MSMF, Russell-Jones R (2003) Outcome in 34 patients with juvenile-onset mycosis fungoides: a clinical, immunophenotypic, and molecular study. Cancer 98: 2282-2290  Wakim LM, Woodward-Davis A, Liu R, Hu Y, Villadangos J, Smyth G, Bevan MJ (2012) The molecular signature of tissue resident memory CD8 T cells isolated from the brain. Journal of immunology 189: 3462-3471  Walker JR, Corpina RA, Goldberg J (2001) Structure of the Ku heterodimer bound to DNA and its implications for double-strand break repair. Nature 412: 607-614  Walunas TL, Lenschow DJ, Bakker CY, Linsley PS, Freeman GJ, Green JM, Thompson CB, Bluestone JA (1994) CTLA-4 can function as a negative regulator of T cell activation. Immunity 1: 405-413  Wang JH, Nichogiannopoulou A, Wu L, Sun L, Sharpe AH, Bigby M, Georgopoulos K (1996) Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity 5: 537-549  Wang L, Ni X, Covington KR, Yang BY, Shiu J, Zhang X, Xi L, Meng Q, Langridge T, Drummond J, Donehower LA, Doddapaneni H, Muzny DM, Gibbs RA, Wheeler DA, Duvic M (2015) Genomic profiling of Sezary syndrome identifies alterations of key T cell signaling and differentiation genes. Nat Genet   147 Wang Y, Su M, Zhou LL, Tu P, Zhang X, Jiang X, Zhou Y (2011) Deficiency of SATB1 expression in Sezary cells causes apoptosis resistance by regulating FasL/CD95L transcription. Blood 117: 3826-3835  Watanabe R, Gehad A, Yang C, Scott LL, Teague JE, Schlapbach C, Elco CP, Huang V, Matos TR, Kupper TS, Clark RA (2015) Human skin is protected by four functionally and phenotypically discrete populations of resident and recirculating memory T cells. Sci Transl Med 7: 279ra239  Weaver CT, Harrington LE, Mangan PR, Gavrieli M, Murphy KM (2006) Th17: an effector CD4 T cell lineage with regulatory T cell ties. Immunity 24: 677-688  Wieselthier JS, Koh HK (1990) Sezary syndrome: diagnosis, prognosis, and critical review of treatment options. J Am Acad Dermatol 22: 381-401  Wilkinson B, Chen JY, Han P, Rufner KM, Goularte OD, Kaye J (2002) TOX: an HMG box protein implicated in the regulation of thymocyte selection. Nat Immunol 3: 272-280  Willemze R, Jaffe ES, Burg G, Cerroni L, Berti E, Swerdlow SH, Ralfkiaer E, Chimenti S, Diaz-Perez JL, Duncan LM, Grange F, Harris NL, Kempf W, Kerl H, Kurrer M, Knobler R, Pimpinelli N, Sander C, Santucci M, Sterry W, Vermeer MH, Wechsler J, Whittaker S, Meijer CJ (2005) WHO-EORTC classification for cutaneous lymphomas. Blood 105: 3768-3785  Wong HK, Mishra A, Hake T, Porcu P (2011) Evolving insights in the pathogenesis and therapy of cutaneous T-cell lymphoma (mycosis fungoides and Sezary syndrome). Br J Haematol 155: 150-166  Wong HK, Wilson AJ, Gibson HM, Hafner MS, Hedgcock CJ, Berger CL, Edelson RL, Lim HW (2006) Increased expression of CTLA-4 in malignant T-cells from patients with mycosis fungoides -- cutaneous T cell lymphoma. J Invest Dermatol 126: 212-219  Wood GS (1995) Lymphocyte activation in cutaneous T-cell lymphoma. J Invest Dermatol 105: 105S-109S  Wu J, Nihal M, Siddiqui J, Vonderheid EC, Wood GS (2009) Low FAS/CD95 expression by CTCL correlates with reduced sensitivity to apoptosis that can be restored by FAS upregulation. J Invest Dermatol 129: 1165-1173  Wurz K, Garcia RL, Goff BA, Mitchell PS, Lee JH, Tewari M, Swisher EM (2010) MiR-221 and MiR-222 alterations in sporadic ovarian carcinoma: Relationship to CDKN1B, CDKNIC and overall survival. Genes Chromosomes Cancer 49: 577-584  Wysocka M, Zaki MH, French LE, Chehimi J, Shapiro M, Everetts SE, McGinnis KS, Montaner L, Rook AH (2002) Sezary syndrome patients demonstrate a defect in dendritic  148 cell populations: effects of CD40 ligand and treatment with GM-CSF on dendritic cell numbers and the production of cytokines. Blood 100: 3287-3294  Yang X, Karuturi RK, Sun F, Aau M, Yu K, Shao R, Miller LD, Tan PB, Yu Q (2009) CDKN1C (p57) is a direct target of EZH2 and suppressed by multiple epigenetic mechanisms in breast cancer cells. PLoS One 4: e5011  Yano H, Ishida T, Inagaki A, Ishii T, Ding J, Kusumoto S, Komatsu H, Iida S, Inagaki H, Ueda R (2007) Defucosylated anti CC chemokine receptor 4 monoclonal antibody combined with immunomodulatory cytokines: a novel immunotherapy for aggressive/refractory Mycosis fungoides and Sezary syndrome. Clin Cancer Res 13: 6494-6500  Yao CJ, Works K, Romagnoli PA, Austin GE (2005) Effects of overexpression of HBP1 upon growth and differentiation of leukemic myeloid cells. Leukemia 19: 1958-1968  Yawalkar N, Ferenczi K, Jones DA, Yamanaka K, Suh KY, Sadat S, Kupper TS (2003) Profound loss of T-cell receptor repertoire complexity in cutaneous T-cell lymphoma. Blood 102: 4059-4066  Yeh JE, Toniolo PA, Frank DA (2013) Targeting transcription factors: promising new strategies for cancer therapy. Curr Opin Oncol 25: 652-658  Yokota Y, Mansouri A, Mori S, Sugawara S, Adachi S, Nishikawa S, Gruss P (1999) Development of peripheral lymphoid organs and natural killer cells depends on the helix-loop-helix inhibitor Id2. Nature 397: 702-706  Yoo EK, Cassin M, Lessin SR, Rook AH (2001) Complete molecular remission during biologic response modifier therapy for Sezary syndrome is associated with enhanced helper T type 1 cytokine production and natural killer cell activity. J Am Acad Dermatol 45: 208-216  Zackheim HS, Amin S, Kashani-Sabet M, McMillan A (1999) Prognosis in cutaneous T-cell lymphoma by skin stage: long-term survival in 489 patients. J Am Acad Dermatol 40: 418-425  Zackheim HS, Kashani-Sabet M, Amin S (1998) Topical corticosteroids for mycosis fungoides. Experience in 79 patients. Arch Dermatol 134: 949-954  Zhang Y, Wang Y, Yu R, Huang Y, Su M, Xiao C, Martinka M, Dutz JP, Zhang X, Zheng Z, Zhou Y (2012) Molecular markers of early-stage mycosis fungoides. J Invest Dermatol 132: 1698-1706  Zhu J, Paul WE (2008) CD4 T cells: fates, functions, and faults. Blood 112: 1557-1569  Zinzani PL, Musuraca G, Tani M, Stefoni V, Marchi E, Fina M, Pellegrini C, Alinari L, Derenzini E, de Vivo A, Sabattini E, Pileri S, Baccarani M (2007) Phase II trial of  149 proteasome inhibitor bortezomib in patients with relapsed or refractory cutaneous T-cell lymphoma. J Clin Oncol 25: 4293-4297  Zoi-Toli O, Vermeer MH, De Vries E, Van Beek P, Meijer CJ, Willemze R (2000) Expression of Fas and Fas-ligand in primary cutaneous T-cell lymphoma (CTCL): association between lack of Fas expression and aggressive types of CTCL. Br J Dermatol 143: 313-319      150 Appendix: Additional Publications (as a result of research conducted during thesis research period, but not presented in the thesis) 1. Esmailzadeh S, Huang Y, Su MW, Zhou Y, Jiang X. BIN1 tumor suppressor regulates Fas/Fas ligand-mediated apoptosis through c-FLIP in cutaneous T-cell lymphoma. Leukemia. 2015 Jun;29(6):1402-13.  2. Litvinov IV, Cordeiro B, Fredholm S, Ødum N, Zargham H, Huang Y, Zhou Y, Pehr K, Kupper TS, Woetmann A, Sasseville D. Analysis of STAT4 expression in cutaneous T-cell lymphoma (CTCL) patients and patient-derived cell lines. Cell Cycle. 2014;13(18):2975-82. 3. Litvinov IV, Cordeiro B, Huang Y, Zargham H, Pehr K, Doré MA, Gilbert M, Zhou Y, Kupper TS, Sasseville D. Ectopic expression of cancer-testis antigens in cutaneous T-cell lymphoma patients. Clin Cancer Res. 2014 Jul 15;20(14):3799-808. 4. Mai R, Cheng Y, Huang Y, Zhang G. Esophageal squamous cell carcinoma and gastric cardia adenocarcinoma shared susceptibility locus in PLCE1: a meta-analysis. PLoS One. 2013 Jul 18;8(7):e69214. 5. Yu R, Broady R, Huang Y, Wang Y, Yu J, Gao M, Levings M, Wei S, Zhang S, Xu A, Su M, Dutz J, Zhang X, Zhou Y. Transcriptome analysis reveals markers of aberrantly activated innate immunity in vitiligo lesional and non-lesional skin. PLoS One. 2012;7(12):e51040. 6. Yu R, Huang Y, Zhang X, Zhou Y. Potential role of neurogenic inflammatory factors in the pathogenesis of vitiligo. J Cutan Med Surg. 2012 Jul-Aug;16(4):230-44.  151 7. Zhang Y, Wang Y, Yu R, Huang Y, Su M, Xiao C, Martinka M, Dutz JP, Zhang X, Zheng Z, Zhou Y. Molecular markers of early-stage mycosis fungoides. J Invest Dermatol. 2012 Jun;132(6):1698-706. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items