Identification of HTLV-1 Tax-1 and HBZ Binding Partners, and Their Role in HTLV-1 Biology and Pathogenesis

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By

Jacob Jamal Al-Saleem

Graduate Program in Molecular, Cellular and Developmental Biology

The Ohio State University

2016

Dissertation Committee:

Dr. Patrick L. Green, Ph.D, Advisor

Dr. Jesse Kwiek, Ph.D

Dr. Li Wu, Ph.D

Dr. Stefan Niewiesk, DVM, Ph.D

Copyrighted by

Jacob Jamal Al-Saleem

2016

Abstract

Human T cell Leukemia Virus type 1 (HTLV-1) is estimated to have infected 15-20 million individuals world-wide. A subgroup of infected individuals develop diseases associated with viral infection, which include adult T cell leukemia and HTLV-1 associated myelopathy/tropic spastic paraparesis. A closely related virus, HTLV-2, shares

70% nucleotide similarity with HTLV-1, but is not associated with any disease. Previous research has demonstrated that Tax-1 serves as the primary oncoprotein of HTLV-1, and its deletion or functional disruption completely ablates the transforming capacity of

HTLV-1. Tax-2 (HTLV-2) has been demonstrated to have lower transforming capabilities than Tax-1. These differences are attributed to two domains present in Tax-1 yet absent in Tax-2; the four-amino acid PDZ binding motif (PBM) and an eight-amino acid leucine zipper-like region (LZ). These domains have been demonstrated to be required for the ability of Tax-1 to activate the alternative NF-κB pathway.

Chapter Two of this dissertation analyzes the role that the alternative NF-κB pathway plays in HTLV-1 mediated T cell transformation. Analysis of the Tax-1 PBM or

LZ mutants revealed that deletion of the PBM does not inhibit activation of the alternative NF-κB pathway. We then show that the PBM domain is required for Tax-1 activation of Akt signaling, and the mechanism behind this activation involves interactions between Tax-1, DLG-1, and PTEN. The Tax-1 LZ domain is required for activation of alternative NF-κB and HTLV-1 featuring the mutated Tax-1 LZ transformed

ii T cells with an efficiency similar to wild type virus. This is the first evidence suggesting that alternative NF-κB activity is not required for in vitro cellular transformation of primary T-lymphocytes in culture.

Studies in Chapter Three focus on dissecting the mechanism for Tax-1 activation of the alternative NF-κB pathway. Binding partners of four different Tax constructs were identified, including a mutant incapable of activating alternative NF-κB. While six proteins were identified as potentially important for alternative NF-κB activation, none passed initial screening to confirm their interaction profile. Other novel interactions of

Tax-1 were analyzed, and an interaction between Tax-1 and SNX27 was discovered. We demonstrated that, through SNX27, Tax-1 regulates the localization of the HTLV-1 receptor molecule GLUT1. This is the first report describing a mechanism by which

HTLV-1 regulates its receptor molecule.

HTLV-1 expresses HBZ from the antisense strand of the viral genome. Both HBZ protein and mRNA have been implicated to promote proliferation of T cells. Studies in

Chapter Four focused on investigating the mechanism behind these proliferative effects.

Binding partners of both HBZ protein and mRNA were identified and analyzed to uncover interactions that regulate proliferation. Our preliminary experiments have identified cellular candidates that bind the HBZ protein and mRNA, but to date we have yet to identify a functional interacting partner. Further analysis is on-going.

Collectively the data in this dissertation demonstrates that activation of the alternative NF-κB pathway is dispensable for in vitro transformation of T cells, and identifies binding partners of both Tax-1 and HBZ. Tax-1 activation of Akt was found to be dependent on the PBM, and this activation may serve as a therapeutic target. The

iii interaction between Tax-1 and SNX27 could have profound implications on HTLV-1 biology as it is the first identified method of HTLV-1 regulation of receptor molecules post-entry. Several binding partners of both HBZ protein and mRNA were discovered, and follow up studies will be required to identify the role these interactions may play in

HTLV-1 biology and pathogenesis.

iv

Dedicated to Becky and Carmen

v

Acknowledgments

I want to first thank my advisor, Dr. Patrick Green. I have been extremely lucky to have worked for a trusting, understanding, and patient advisor. His guidance and motivation were crucial to my development as a scientist. Thank you for allowing me to develop into an independent scientist under your tutelage.

I would like to thank my committee members, Drs. Li Wu, Jesse Kwiek, and

Stefan Niewiesk for their advice and encouragement during my graduate studies.

I would like to thank all members of Dr. Green’s Lab, both past and present: Dr.

Han Yin, Dr. Rami Doueiri, Dr. Priya Kannian, Dr. Raj Anupam, Dr. Amanda Panfil, Dr.

Nathan Dissigner, Krissy Landes, Corey Howard, and Dr. Michael Martinez. For all the aid you have given me in experiments and for all the distractions from stress, I am forever grateful.

Much of the work done in my dissertation would not have been possible without the assistance and technical expertise of Drs. Mamuka Kvaratskhelia and Nikolozi

Shkriabai. I would like to thank Dr. Corine St. Gelais for her time spent editing my writing. I promise to ‘perform,’ not ‘preform’ my experiments in the future. I also want to thank Dr. Kate Hayes-Ozello for her editorial assistance. A special thanks to all other members of the basement and CRR labs for making my time here memorable.

vi I would like to also thank my undergraduate research advisor, Dr. Maki Asano.

Without her I would have never gone to graduate school. She was an amazing advisor, brilliant teacher, caring mother, and inspiring scientist. I miss her dearly. She may be gone, but she will never be forgotten.

To my family, thank you for all your support during my extended stay at OSU.

Your love and encouragement has propelled me throughout graduate career. And yes, I am done with school now.

Last, but definitely not least, I would like to thank my loving wife, Becky. The road has been long, but you stood by my side through it all. Whenever the challenges would seem too large you were there to center me. Whenever I had a long day in the lab, you were there when I got home to lift me up. You are my rock, and with you I am a better man. SFD4L. I love you forever and always.

vii

Vita

2005...... Lakota High School Kansas, Ohio

2010...... B.S. Molecular Genetics The Ohio State University Columbus, Ohio

2010 to present ...... Graduate Research Associate, Department of Veterinary Biosciences The Ohio State University Columbus, Ohio

Publications 1. Al-Saleem, J., Kvaratskhelia, M., Green, P.L. Methods for identifying and examining HTLV-1 HBZ post-translational modifications. In Human T- lymphotropic Viruses-Methods and Protocols, C Casoli (ed), Springer Science, New York In Press

2. Panfil, A. R., Al-Saleem, J., Howard, C. M., Mates, J. M., Kwiek, J. J., Baiocchi, R. A., Green, P. L. PRMT5 Is Upregulated in HTLV-1-Mediated T-Cell Transformation and Selective Inhibition Alters Viral Gene Expression and Infected Cell Survival. Viruses 8, 7 (2016).

3. Cherian, M., Baydoun, H., Al-Saleem, J., Shkriabai, N., Kvaratskhelia, M., Green, P., & Ratner, L. Akt Activation by Human T-Cell Leukemia Virus Tax Oncoprotein. J. Biol. Chem. jbc.M115.684746 (2015). doi:10.1074/jbc.M115.684746

4. Dissinger N., Shkriabai N., Hess S., Al-Saleem J., Kvaratskhelia M., Green PL.Identification and Characterization of HTLV-1 HBZ Post-Translational Modifications. PLoS ONE 9, e112762 (2014).

5. Panfil, A. R., Al-Saleem, J. J., & Green, P. L. Animal Models Utilized in HTLV- 1 Research. Virology Research and Treatment 2013, 49–59 (2013).

viii

Fields of Study

Major Field: Molecular, Cellular and Developmental Biology

ix

Table of Contents

Abstract ...... ii

Acknowledgments...... vi

Vita ...... viii

Publications ...... viii

Fields of Study ...... ix

Table of Contents ...... x

List of Tables ...... xvii

List of Figures ...... xviii

Chapter 1: Literature Review ...... 1

1.1 Introduction to Retroviruses ...... 1

1.2 HTLV Discovery and Epidemiology ...... 3

1.3 Pathology and Diseases Associated with HTLV-1 ...... 4

1.3.1 Adult T cell Leukemia ...... 5

1.3.1.1 Subtypes of ATL ...... 6

1.3.1.2 ATL Oncogenesis ...... 8

1.3.1.3 Prognosis and Treatments ...... 9

x 1.3.2 HTLV-1 Associated Myelopathy/Tropic Spastic Paraparesis ...... 12

1.3.3 HTLV-1 uveitis...... 14

1.3.4 Other Associated Diseases...... 15

1.4 Virus Life Cycle ...... 15

1.4.1 Receptor Attachment, Binding, Entry, and Uncoating ...... 16

1.4.2 Reverse Transcription ...... 17

1.4.2.1 Primer Binding and Minus-Strand Strong-Stop DNA Formation ...... 18

1.4.2.2 Translocation One ...... 18

1.4.2.3 Long Minus-Strand DNA Synthesis ...... 19

1.4.2.4 Initiation of Plus-Strand DNA Synthesis ...... 19

1.4.2.5 tRNA Primer Removal ...... 19

1.4.2.6 Translocation Two ...... 20

1.4.2.7 Completion of Synthesis ...... 20

1.4.3 Nuclear Entry and Integration ...... 20

1.4.4 Transcription and translation ...... 22

1.4.5 Virion Assembly and Budding ...... 24

1.4.6 Maturation ...... 25

1.5 Viral Genome and Proteins ...... 25

1.5.1 Structural Genes ...... 26

1.5.1.1 Gag ...... 26

1.5.1.2 Env ...... 27 xi 1.5.2 Enzymatic Genes ...... 27

1.5.2.1 Pro ...... 27

1.5.2.2 Pol ...... 28

1.5.3 Regulatory Genes ...... 28

1.5.3.1 Tax ...... 28

1.5.3.1.1 Functional Domains of Tax ...... 29

1.5.3.1.2 Role of Tax in Viral Transcription ...... 29

1.5.3.1.3 Role of Tax in Transformation ...... 31

1.5.3.1.4 Differences between Tax-1 and Tax-2 ...... 40

1.5.3.2 Rex ...... 41

1.5.3.3 HBZ...... 43

1.5.4 Accessory Genes ...... 45

1.5.4.1 p30...... 45

1.5.4.2 p12/p8 ...... 46

1.5.4.3 p13...... 48

1.5.4.4 p21...... 48

1.6 HTLV-1 Experimental Models ...... 48

1.6.1 Cell Culture...... 48

1.6.2 Animal Models ...... 49

1.7 Conclusions ...... 51

xii Chapter 2: Investigating the Importance of the Alternative NF‑κB and Akt Pathways in

HTLV-1-Induced Cellular Transformation ...... 59

2.1 Abstract...... 59

2.2 Introduction ...... 61

2.3 Materials and Methods ...... 63

2.3.1 Plasmids ...... 63

2.3.2 Cell Culture...... 64

2.3.3 Immunoblotting ...... 65

2.3.4 Reporter Gene Assays and p19 ELISA ...... 66

2.3.5 Producer Cell Line Generation and Verification ...... 67

2.3.6 Cellular Transformation Assay ...... 67

2.3.7 PBL line verification ...... 68

2.3.8 Co-immunoprecipitation ...... 69

2.3.9 Statistical Analysis ...... 69

2.4 Results ...... 69

2.4.1 Tax-1 225-232 was deficient for alternative NF-κB activation and Tax-1 ΔPBM

was not ...... 69

2.4.2 Tax-1 Activated Akt by Inhibiting PTEN Activity ...... 71

2.4.3 Tax 225-232 Mutation Did Not Affect Transformation ...... 73

xiii 2.4.4 PBL 225-232 generation and characterization ...... 75

2.5 Discussion ...... 75

Chapter 3: Identification of the Role of a Novel Tax-1 Binding Partner, SNX27, in

HTLV-1 Infection ...... 87

3.1 Abstract...... 87

3.2 Introduction ...... 88

3.3 Materials and Methods ...... 90

3.3.1 Plasmids ...... 90

3.3.2 Cell Culture...... 91

3.3.3 Pull Down and Co-Immunoprecipitation ...... 91

3.3.4 Mass Spectrometry ...... 92

3.3.5 Immunoblotting ...... 94

3.3.6 SNX27 Knockdown ...... 95

3.3.7 LTR Reporter Assay and p19 ELISA ...... 96

3.3.8 GLUT1 flow cytometry ...... 96

3.3.9 Statistical Analysis ...... 97

3.4 Results ...... 97

3.4.1 Generation of Mass Spectrometry data for Tax-1, Tax-2, and Mutants ...... 97

3.4.2 Verification of Mass Spectrometry Data...... 98

xiv 3.4.3 Tax-1 Interacted with SNX27 ...... 99

3.4.4 SNX27 Expression was Inversely Related to p19 Gag Release ...... 101

3.4.5 Tax-1 Overexpression Reduced GLUT1 Surface Levels ...... 102

3.5 Discussion ...... 103

Chapter 4: Studies to Determine the Mechanism of HBZ Proliferative Activity ...... 114

4.1 Abstract...... 114

4.2 Introduction ...... 115

4.3 Materials and Methods ...... 116

4.3.1 Plasmids and Cell Culture ...... 116

4.3.2 RNA Pull Down ...... 117

4.3.3 S-Tag Pull Down and Co-Immunoprecipitation ...... 117

4.3.4 Mass Spectrometry ...... 118

4.3.5 Immunoblotting ...... 120

4.4 Results ...... 120

4.4.1 Identification of HBZ Protein-Interacting Partners ...... 120

4.4.2 Identification of hbz mRNA-Interacting Partners ...... 121

4.5 Discussion ...... 122

Chapter 5: Summary and Future Directions ...... 130

5.1 Summary ...... 130

xv 5.2 Future Directions ...... 134

5.2.1 Tax-1 and the Alternative NF-κB pathway ...... 134

5.2.2 Tax-1 and the Akt Pathway ...... 135

5.2.3 Tax-1 and SNX27 ...... 135

5.2.4 HBZ protein and mRNA effects on proliferation ...... 136

5.3 Conclusion ...... 136

References ...... 138

Appendix: List of Abbreviations in Alphabetical Order ...... 172

xvi

List of Tables

Table 3.1 List of Tax Interacting Proteins Identified Via Mass Spectrometry...... 108

Table 4.1 HBZ Protein Interacting Partners ...... 126

Table 4.2 hbz RNA Interacting Partners ...... 129

xvii

List of Figures

Figure 1.1 The Generic Retrovirus Life Cycle ...... 52

Figure 1.2 The Process of Reverse Transcription ...... 53

Figure 1.3 The HTLV-1 Proviral Genome and Coding mRNAs ...... 55

Figure 1.4 Functional Domains of Tax-1 and Tax-2 ...... 56

Figure 1.5 The Classical and Alternative NF-κB Pathways ...... 57

Figure 1.6 The PI3K/Akt/mTOR pathway...... 58

Figure 2.1 Tax-1, Tax-2, and Tax-1 Mutant Proteins ...... 79

Figure 2.2 Characterization of Tax-1 Mutants ...... 80

Figure 2.3 Tax-1 activated Akt, while Tax-1 ΔPBM did not ...... 82

Figure 2.4 Generation and Characterization of HTLV-1 Tax-1 225-232 Producing Cell

Line ...... 84

Figure 2.5 HTLV-1 Tax-1 225-232 induced cellular transformation ...... 85

Figure 2.6 Characterization of Generated PBL Lines ...... 86

Figure 3.1 S-tag Tax-1 Pull Down Confirmation and Mass Spectrometry Sample

Preparation ...... 107

Figure 3.2 Verification of Candidate Tax-1 Interacting Partners ...... 109

xviii Figure 3.3 Tax-1 Interacted with SNX27, and this Interaction was Dependent on Tax-1

PBM and SNX27 PDZ Domains ...... 110

Figure 3.4 SNX27 Expression was Inversely related to HTLV-1 p19 Gag Release ...... 111

Figure 3.5 Tax-1 Reduced Cell Surface GLUT1 Levels ...... 113

Figure 4.1 Pull Down of S-tag HBZ ...... 125

Figure 4.2 HBZ did not interact with CHK1 ...... 127

Figure 4.3 hbz RNA Pull Down ...... 128

xix

Chapter 1: Literature Review

1.1 Introduction to Retroviruses

The Retroviridae family of viruses was discovered in the early 1900’s by two separate groups. A study from a Danish laboratory published in 1908 demonstrated that chicken leucosis was transmitted via a virus, while a study conducted at the Rockefeller

Institute in New York in 1911 demonstrated that chicken sarcomas were transmissible in a cell-free manner1,2. The initial discovery of oncogenic viruses in chickens later lead to many different animal retroviruses, including viruses infecting several mammals such as mice, cattle, and monkeys3,4. These viruses were shown to contain single-stranded RNA as their genetic material, although they did not behave in a manner similar to other single- stranded RNA viruses5,6. RNA viruses at the time were shown to replicate via a RNA- dependent RNA polymerase encoded by the virus, which resulted in double-stranded

RNA molecules. No such RNA-dependent RNA polymerase activity was found in retrovirus-infected cells7.

In the early 1960’s, Howard Temin combined several key observations that led to a major breakthrough in the understanding of how retroviruses replicate in cells. For example, transcription inhibitors impaired production of progeny virions; RNA isolated from virions hybridized to DNA from infected cells; and DNase treatment of cells soon after inoculation with retrovirus would block infection8-10. With this information, Dr.

Temin proposed that retroviruses converted their RNA genetic material into a DNA

1 intermediate, termed the provirus, which was integrated into the host genome to allow transcription of viral gene products11. At the time, this theory was widely dismissed because of a strict belief in the central dogma of molecular biology of

DNARNAprotein; this process was thought to be unidirectional12. However, in the

1970’s, Temin and Baltimore independently identified an RNA-dependent DNA polymerase in Rous sarcoma virus and Rauscher mouse leukemia virus, respectively13,14.

This enzyme, later named reverse transcriptase, gave credence to Temin’s provirus theory by demonstrating that retroviruses could convert their genomic RNA to a double-stranded

DNA counterpart by a process known as “reverse transcription”. Since the discovery of retroviruses, there have been many additions to the molecular biologist’s tool kit, which in turn have led to a deeper understanding of the retrovirus life cycle (discussed in section

1.4). The discovery of reverse transcriptase also bolstered molecular biology as a whole, with breakthroughs in cloning, the discovery of cellular oncogenes, and the confirmation of endogenous retroviruses all occurring thanks to this crucial discovery15-19.

The unique hallmark of the Retroviridae family is the ability to reverse the canonical flow of genetic information via reverse transcription. All retroviruses are spherical in shape, between 80-100 nm in diameter, enveloped in a lipid membrane bilayer, and contain two copies of the positive sense ssRNA genome in their core20.

Retroviruses were initially classified by the appearance of the virion core (as observed by electron microscopy) as A, B, C, or D-type21. Recently, retroviruses were reclassified into the genera Alpharetrovirus, Betaretrovirus, Gammaretrovirus, Deltaretrovirus,

Epsilonretrovirus, Lentivirus, and Spumavirus22. The first three genera, Alpha-, Beta-, and Gamma-, are known as simple retroviruses that encode only the structural and

2 enzymatic proteins that are common among all retroviruses. The remaining genera,

Delta-, Epsilon-, Lenti-, and Spuma-, are known as complex retroviruses that encode regulatory and accessory gene products in addition to the structural and enzymatic gene products20.

1.2 HTLV Discovery and Epidemiology

Upon the discovery of oncogenic retroviruses in higher mammals, the search for a human oncogenic retrovirus intensified. The first human retrovirus, discovered in 1980 by Bernard Poiesz et al., consisted of type-C retroviral particles isolated from two T cell lines, HUT102 and CTCL-3, which had been established a year apart from a single patient with cutaneous T cell lymphoma23. Independently, a Japanese group also isolated type-C retroviral particles from the MT-2 cell line, which was established from a patient with adult T cell leukemia (ATL) in 198124. The isolated viruses were later determined to be identical, and were referred to as human T cell leukemia virus type-1 (HTLV-1)25-27. A closely related virus, human T cell leukemia virus type-2 (HTLV-2), was discovered in a patient suffering from hairy cell leukemia, and was later found not to be associated with any disease28,29. In the 2000’s, two new strains of HTLV, HTLV-3 and HTLV-4, were discovered in Cameroon in African bush-meat hunters30,31. Because of the limited numbers of infected people, the pathogenic properties of these strains remain unclear.

It has been estimated that between 15-20 million people are infected with

HTLV-1 world-wide; areas with the highest prevalence are Japan, the Caribbean Islands,

Central America, South America, and Africa32. There are six subtypes of HTLV-1

(Subtypes A-F) based on sequence divergence in the envelope (Env) and the non-coding

3 long terminal repeat (LTR) regions of the viral genome33. Subtype A, the “cosmopolitan subtype,” is the most widespread HTLV-1 subtype and is found in almost all endemic areas34. To date, no subtype-specific differences in disease pathogenesis have been reported32. HTLV-1 infection is spread via three primary mechanisms: vertical transmission from mother to child (primarily via breastfeeding), sexual intercourse (most efficiently from male to female), and intravenous blood exposure (mainly by transfusion of contaminated blood or the sharing of needles or syringes)34,35.

HTLV is a Deltaretrovirus, as are simian T cell leukemia virus (STLV) and bovine leukemia virus (BLV)22. HTLV and STLV share a common ancestor and are referred to as primate T cell leukemia viruses (PTLVs)30. It is believed that PTLVs have repeatedly crossed between humans and simians and as a result HTLV-1 and STLV-1 are so closely related as to be indistinguishable phylogenetically33. PTLV-1 is predicted to have originated 27,300 ± 8,200 years ago, and the first interspecies transmission occurred approximately 16,000 ± 4,900 years ago36. However, because of the strong phylogenic similarity between STLV-1 and HTLV-1, it has not been possible to determine the presence or time of any further interspecies transmissions36.

1.3 Pathology and Diseases Associated with HTLV-1

HTLV-1 was the first human retrovirus linked to human diseases, which include

ATL, HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP), and

HTLV-associated uveitis (HU)23,24,37. Anecdotal reports linking HTLV-1 infection and arthritis or other skin ailments have not provided conclusive evidence of causality34.

Cases of ATL and HAM/TSP are found in all endemic areas of the virus, but the

4 distribution for each disease varies with geographical location. For example, the prevalence of ATL is three times higher in Japan compared to the Caribbean, whereas the prevalence of HAM/TSP incidence in Japan is one-tenth of that in the Caribbean38.

Differences also exist between the sexes, where ATL is more common in males and

HAM/TSP is more common in females39-41. It is worth noting, however, that only between 2-10% of individuals infected with HTLV-1 will eventually develop a disease, with a majority of infected individuals remaining asymptomatic during the course of their lives41-45.

1.3.1 Adult T cell Leukemia

ATL is a lymphoproliferative disorder of T cells that was first described in Japan in 1977, and later demonstrated to be associated with HTLV-1 infection24,46. The evidence that HTLV 1 causes ATL includes: ATL cells contain the HTLV-1 provirus integrated into their genome, infection of healthy T cells with HTLV-1 results in transformation of these cells, HTLV-1 antibodies can be found in up to 90% of ATL cases, and HTLV-1 virions can be isolated from ATL cell lines23,24,47-57. ATL develops in

2-8% of those infected with HTLV-1 following a 20-30 year clinical latency period58.

Several studies have shown that infection during childhood increases the risk of ATL development, however, there is evidence for ATL development in people infected during adulthood41,59.

5 1.3.1.1 Subtypes of ATL

There are four subtypes of ATL: smoldering, chronic, lymphoma, and acute58.

These subtypes were first proposed by Shimoyama and Takatsuki in 1985 because of the diversity in clinical features of ATL60,61. In 1991 Shimoyama suggested diagnostic criteria for the subtypes, and in 2008 the subtypes were formally adopted by the World

Health Organization Classification of Tumors of Hematopoietic and Lymphoid

Tissues62,63. Some have postulated that the subtypes represent different stages of disease progression64. In all cases of ATL the detection of anti-HTLV-1 antibodies in patient sera is required for the diagnosis62. The criteria for subtyping, prognosis, and other pertinent facts for the subtypes are listed below.

Smoldering

The smoldering subtype accounts for 5% of ATL cases and affected individuals have a median survival time >5 years62,64. Patients with smoldering ATL are generally asymptomatic, but may have skin and pulmonary lesions62. Smoldering ATL features normal lymphocyte counts (<4x109 per liter), and peripheral blood lymphocyte population with >5% ATL cells60,62. ATL cells are medium to large lymphocytes featuring a condensed and nodular nucleus that resembles the petals of a flower in shape, and are often referred to as “flower cells” 65. Patients with smoldering ATL may progress into the other more severe forms of disease62,64.

6 Chronic

Chronic ATL constitutes 15% of all cases; affected individuals have a median survival time of 2 years62,64. Diagnosis of chronic ATL requires an elevated lymphocyte count (>4x109 per liter), peripheral blood lymphocyte population with >5% ATL cells, and elevated lactate dehydrogenase (LDH) levels62,66. Patients may also present with skin and pulmonary lesions, similar to smoldering ATL, while lesions in the spleen, lymph node, and liver have also been observed58,62,66. Progression to acute ATL has been reported64,66.

Lymphoma

The lymphoma subtype accounts for 20% of cases; affected individuals have a median survival time of 10 months62,64. Lymphoma is unique from other ATL subtypes as it consists of lymphadenopathy but a lack of a lymphocytosis and ≤1% ATL cells in the peripheral blood lymphocyte population58,62. Diagnosis of ATL lymphoma requires a biopsy and detection of HTLV-1 provirus in diseased tissues62,67. Skin and pulmonary lesions are not common in this subtype as compared to the others62. The lymphoma subtype is considered by some to be a further sub-classification of acute ATL64.

Acute

The acute subtype of ATL accounts for 60% of cases; affected individuals have a median survival time of 6 months, which is the worst survival outcome of the subtypes62,64. To reach a diagnosis of acute ATL a patient would present with an elevated lymphocyte count (>4x109 per liter) with >5% ATL cells in the peripheral lymphocyte

7 population62. Patients may also present with hypercalcemia with or without osteolytic lesions, and elevated LDH levels67. It is common for lesions to be present in multiple organs with this subtype37,58,62.

1.3.1.2 ATL Oncogenesis

The transformation process of HTLV-1 infected cells into ATL cells is multifaceted. Some oncogenic retroviruses transform infected cells by proviral integration near proto-oncogenes, leading to the activation of these oncogenes68-70.

However, HTLV-1 has not been shown to transform in this manner71. HTLV-1 has no preference for any specific integration site overall, but a statistically significant proportion of individuals that develop disease have integrations within transcriptionally active regions71,72. A recent study suggested that the location of proviral integration may be important to HTLV-1-associated disease due to the presence of a CCCTC-binding factor (CTCF) binding site in the provirus73. It is postulated that through the CTCF binding site HTLV-1 may regulate cellular transcription up to a Mb away from the viral insertion site73.

In the absence of insertional effects, HTLV-1 relies on virally encoded accessory/regulatory genes to drive oncogenesis. The HTLV-1 accessory/regulatory proteins will be discussed in detail in section 1.5. Briefly, the HTLV-1 transcriptional activator (Tax) protein is involved in deregulating multiple cellular pathways including those involved in proliferation, apoptosis, and DNA damage recognition/repair74-76. It is through these functions that Tax initiates the process of oncogenesis; however, Tax is highly immunogenic and its expression results in a cytotoxic T lymphocyte (CTL)

8 response77,78. As a result, Tax expression is silenced in the majority of ATL patients likely through sense-strand promoter deletion, sense-strand promoter methylation, or tax mutation79-82. The only viral protein detected in all ATL cases is the HTLV-1 basic leucine zipper factor (HBZ)83. HBZ represses Tax activity, which reduces the antiviral immune response83. HBZ in both its protein and mRNA forms can drive cellular proliferation84. Via Tax, HTLV-1 establishes an environment primed for transformation, and via HBZ, HTLV-1 maintains this environment while promoting proliferation of the primed cells. Secondary events, such as inefficient DNA damage repair, genetic mutation, or epigenetic events could result in the full transformation of the HTLV-1- infected cell.

1.3.1.3 Prognosis and Treatments

The current prognosis for ATL overall is poor. For clinical purposes the disease is classified into indolent and aggressive forms85. The indolent group consists of the smoldering and chronic subtypes of ATL and has a relatively optimistic prognosis, with a median survival of 2 years or more. The aggressive group consists of the lymphoma and acute subtypes of ATL and has a poorer prognosis, with median survival of 6-10 months.

The dismal prognosis in aggressive ATL is due to a larger tumor burden (when compared to indolent disease), drug resistance, and multi-organ failure. Several factors have been linked to a poorer prognosis including age over 40 years, hypercalcemia, high levels of

LDH, and high proviral load86. In all subtypes of ATL there is a risk for opportunistic infections due to the immunocompromised state of the patient87.

9 The current treatment strategies for ATL vary based on the clinical sub- classifications of the disease and include watchful waiting, antiviral therapies, chemotherapy, monoclonal antibody therapy, and hematopoietic stem cell transplantation

(HSCT). For many patients with indolent ATL, a watchful waiting practice is used where no medical treatments are prescribed, but the patient is monitored for disease progression88. A 2010 report demonstrated that treatment with a combination of the antiretroviral zidovudine (AZT) and interferon alpha (INF-α) in patients with indolent

ATL led to a 5-year survival rate of 100% compared to 42% of patients who received chemotherapy with or without antiviral therapy, which demonstrated that AZT/INF-α treatment may be an effective option for those with indolent ATL89. AZT/INF-α treatments are also used in patients with aggressive ATL with varying effects90. In acute

ATL, treatment with AZT/INF-α improves survival, but in the lymphoma variant chemotherapy options are more effective89.

Chemotherapy treatment options for aggressive ATL include regimens such as

CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) and VCAP

(vincristine, cyclophosphamide, doxorubicin, prednisolone), AMP (doxorubicin, ranimustine, prednisolone), and VECP (vindesine, etoposide, carboplatin, prednisolone)

91,92. A 2007 clinical trial in Japan compared the efficacy of VCAP-AMP-VECP (LSG15) to CHOP. The complete remission (CR) with LSG15 was 40% compared to CHOP at

25%, which was a marked improvement93. However the median survival of patients treated with LSG15 was only 13 months compared to 11 months for CHOP, leaving much room for improvement93. A downside to chemotherapy treatment was an increase in opportunistic infections due to the already immunocompromised state of the patient94.

10 Because ATL cells display a specific phenotype (CD3+, CD4+, CD8-, CD25+) monoclonal antibody therapies have been proposed as prospective treatments95.

Expression of IL-2R (CD25) was shown to be increased in ATL cells, and was selected as a target for early monoclonal antibody treatment64,96. One study utilized a yttrium-90- labeled antibody that showed limited success, with 44% of patients achieving partial remission (PR) and 13% achieved CR95. However, the treatment was highly toxic to the hematopoietic system95. More recent studies have focused on mogamulizumab, an anti-

CCR4 monoclonal antibody97. CCR4 is normally expressed on regulatory T-cells and also is expressed on the surface of most ATL cells98. Higher levels of CCR4 expression on ATL cells have been correlated with poor prognosis99. A phase I clinical trial in Japan designed to determine the optimal dose of mogamulizumab led to two separate phase II clinical trials100. In the first phase II trial, mogamulizumab was administered to relapsed

ATL patients and resulted in an overall response rate of 50%, which was an improvement over other agents used to treat relapsed ATL101. The second phase II trial monitored the effect of LSG15 treatment either with or without mogamulizumab in patients newly diagnosed with aggressive ATL. This study found that 52% of patients administered combined LSG15 and mogamulizumab achieved CR, while those treated with LSG15 alone achieved 33% CR102. This increase in number of patients with CR was accompanied by an increase in adverse events, including: pneumonia, cytomegalovirus infection, and interstitial lung disease. While the monoclonal antibody therapies were successful, the adverse events leave much to be desired.

HSCT has been demonstrated to be successful for the treatment of aggressive non-Hodgkin’s lymphoma, and several studies have set out to determine the efficacy of

11 HSCT on ATL103. Initial studies using an autologous source of stem cells were not successful, with relapse and/or fatal complications occurring in treated individuals104.

Allogenic HSCT treatments were next explored for ATL and the first allogenic HSCT was performed in 1987105. This treatment resulted in CR, but the patient died within a year of treatment due to interstitial pneumonitis105. An allogeneic HSCT transplant to a female with ATL from her HTLV-1-negative sister resulted in CR for 23 months post transplantation with no detectable HTLV-1 in the blood106. Since this successful study, several other cases of successful allogenic HSCT have been reported107,108. Much of the work in this field today is focused on determining the best donor selection, and understanding the effects of combinational therapies.

While many treatment strategies for ATL have been developed, the overall prognosis for affected individuals is still poor. Future work will focus on how existing treatments can be further refined and combined to achieve a better response in patients.

1.3.2 HTLV-1 Associated Myelopathy/Tropic Spastic Paraparesis

Tropic spastic paraparesis (TSP) and HTLV-1-associated myelopathy (HAM) were initially described almost a century apart. In the late 19th century, Strachan described a unique form of neuritis in patients he treated in Jamaica109. In 1956, a review of 100 cases similar to the one described by Strachan was detailed in a report by

Cruickshank110. These and other cases of neuropathy were eventually grouped into the disease known as TSP111. In the late 1980’s, HAM was described by two separate groups in Africa and Japan112,113. Today, both diseases are known to be one and the same, and are collectively referred to as HAM/TSP. HTLV-1 was linked to HAM/TSP by several

12 key findings: a high percentage of patients with HAM/TSP have HTLV-1 antibodies present in their sera, and the HTLV-1 viral genome can be detected in the central nervous system (CNS) of patients with HAM/TSP114-116. While HTLV-1 is considered the causative agent of HAM/TSP, there are reports demonstrating that HTLV-2 is associated with a few cases of neurological disease similar to HAM/TSP117,118.

Clinically, HAM/TSP is a progressive neurological disorder. Initial presentation usually consists of weakness in the lower extremities followed by spasticity in the lower limbs119. Patients progressively lose the ability to walk. The average time until a unilateral walking aid is required is 6 years post diagnosis, bilateral aid is required at 13 years, and the requirement of a wheelchair is 21 years120. Other symptoms that present in patients include a loss of bladder control, constipation, and sexual dysfunction.

Pathologically, HAM/TSP is characterized by an infiltration of T cells into the grey and white matter of the thoracic spinal cord. The exact method by which HTLV-1 causes

HAM/TSP is still under investigation. One theory of how HAM/TSP develops involves autoimmunity and molecular mimicry121,122. This theory proposes that antibodies against the viral protein Tax recognize the host protein heterogeneous nuclear ribonucleic protein

A1, which is neuron-specific. It is believed that these antibodies cross the blood brain barrier and eventually lead to infiltration of T cells into the CNS, resulting in disease.

To date only a few clinical trials have attempted treatment of HAM/TSP. One trial monitored the effect of AZT, and found no clinical improvements123. Several other studies demonstrated that INF-α treatment decreased clinical symptoms in a significant proportion of those treated124,125. However, INF-α treatment led to a multitude of side effects, including some that were severe125. In the absence of a disease-based approach,

13 most current treatments are focused on the specific symptoms of HAM/TSP. Overall these treatments have not been successful, pointing to the necessity of additional basic research and clinical trials.

1.3.3 HTLV-1 uveitis

Strachan’s initial description of HAM/TSP included an ocular component to the disease that is not part of its current clinical definition109. Strachan described how patients he observed had inflammation in the eyes and eventual loss of sight. Today this is known as a separate clinical entity of HTLV-1, HTLV-1 uveitis (HU). HU, inflammation of the uvea, can lead to blindness in untreated cases126. HU was linked to HTLV-1 though several key findings: a high percentage of HTLV-1 seroprevalence in idiopathic uveitis patients, HTLV-1 proviral DNA detected in cells infiltrating the anterior chamber of the eye, and detection of HTLV-1 viral mRNA, viral protein, and virus particles in the aqueous humor44,127,128.

The major clinical symptoms of HU are blurred vision and the appearance of

‘floaters,’ which are dark shapes that obscure vision128. Other secondary symptoms include pain and itching129. HU generally presents as its own clinical entity, but has been shown to arise alongside HAM/TSP or ATL130,131. Pathologically, HU involves infiltration of CD3+ T cells into the anterior chamber of the eye, although the cells are not malignant132,133. It is proposed that the presence of HTLV-1-infected T cells leads to the secretion of cytokines that induce inflammation of the eye, resulting in HU128.

However, the manner by which HTLV-1-infected T cells infiltrate the eye is still poorly understood129. Treatments for HU are symptom-based using corticosteroids134. These

14 treatments have been successful, but long term corticosteroid treatment is ill advised129.

Additional studies focused on the application of current treatments to other HTLV-1- associated diseases on HU should be undertaken.

1.3.4 Other Associated Diseases

Several other diseases have been loosely tied to HTLV-1 infection including dermatitis, rheumatoid arthritis, and depression. Infective dermatitis is a form of childhood dermatitis, and was the first pediatric condition associated with HTLV-1135.

Having infective dermatitis in childhood has been strongly associated with HAM/TSP development in adulthood136. A survey of 113 Japanese patients with rheumatoid arthritis revealed a high seroprevalence of HTLV-1 when compared to normal blood donors137. A similar study in the United States also demonstrated a predisposition to arthritis in

HTLV-1-infected individuals138. A case-control study of asymptomatic HTLV-1 carriers in Brazil indicated a high rate of depression compared to non-carriers (39% vs 8% respectively)139. The authors of this study acknowledged that it was impossible to know if this increase in depression was due to the infection itself, or to the knowledge of life-long infection139. However, with all of these associated diseases, the epidemiological evidence of a direct link to HTLV-1 infection has been insufficient. More studies should be carried out to determine the exact role HTLV-1 plays in these diseases.

1.4 Virus Life Cycle

The HTLV-1 life cycle is similar to other retroviruses, such as HIV-1 (see Figure

1-1). Briefly, a mature virion binds to a target cell via an interaction between the HTLV-1 15 Env protein and its target receptor molecules on the cell surface. Once bound, the membrane of the virion fuses with the cellular membrane to release the viral core into the cell cytoplasm. The viral core is then uncoated, and the RNA genome is reverse transcribed into the double-stranded DNA genome, which becomes part of what is known as the pre-integration complex (PIC). The PIC is transported to the nucleus by mechanisms that are still unclear, and upon entry, integrates into the host cell genome becoming the provirus. From the integrated provirus, viral RNAs are produced to generate an array of viral proteins. These RNAs and proteins work together to assemble progeny virions at the host cell membrane. After assembly occurs, the virions bud from the surface of the host cell, and via proteolytic cleavages, mature into infectious particles to start the process anew140. The next section will describe the viral life cycle in finer detail, focusing on HTLV-1 where specific information is known, and retroviruses in general where gaps exist.

1.4.1 Receptor Attachment, Binding, Entry, and Uncoating

The retrovirus lifecycle begins with a mature virion coming into contact with the target cell. While HTLV-1 can infect CD8+ T cells, dendritic cells, B cells, and monocytes, the most efficient target of infection is activated CD4+ T cells141,142. HTLV-1 attachment and binding require the viral Env proteins and the target cell receptor molecules: glucose transporter type 1 (GLUT1), heparan sulfate proteoglycans (HSPG), and Neuropilin 1 (NRP-1)143-145. The process begins with the binding of the HTLV-1 surface unit (SU) of Env to HSPG on the cell surface, which brings SU into close proximity with NRP-1146,147. The interaction between SU and NPR-1 occurs by molecular

16 mimicry where HTLV-1 SU mimics a ligand of NRP-1, VEGF-165148. This interaction causes a conformational shift in SU that exposes the GLUT1-binding domain and allows viral SU to bind to GLUT1. This binding subsequently activates the fusogenic peptide of the transmembrane domain of HTLV-1 Env (TM), that is then inserted into the target cell membrane146-148. Through a cascade of events, the insertion of the fusogenic peptide into the cell membrane results in the formation of a pore between the virion and the target cell149. As this pore expands, the viral membrane fuses with the target cell membrane and the virion core is released into the cytoplasm149.

Virus entry occurs by either direct fusion of the virion membrane with the target cell membrane, or endocytosis of the virion into the target cell where pH-dependent fusion occurs in the endosomal compartment150. Considering HTLV-1 propensity towards cell-to-cell infection and the fact that HTLV-1 entry is not pH-dependent, direct fusion may be the prominent route of HTLV-1 entry151-154. Fusion allows release of the capsid

(CA) core into the cytoplasm, which is followed by uncoating of the CA core and the initiation of reverse transcription. The exact order of events between uncoating and reverse transcription have been under investigation, and a recently published article demonstrated that uncoating initiates after a specific stage of reverse transcription155. The exact mechanism of uncoating is poorly understood156.

1.4.2 Reverse Transcription

The distinct feature of the retrovirus lifecycle is reverse transcription, which is the process whereby the genomic RNA of the virus is converted into double-stranded DNA by the viral enzyme reverse transcriptase. The final dsDNA product differs slightly from the RNA genome because of duplications that generate the direct repeats termed the long

17 terminal repeats (LTRs) found at the 5' and 3' ends of the proviral genome. Reverse transcription can be seperated into seven steps: (1) primer binding and minus-strand strong-stop DNA formation, (2) translocation one, (3) long minus-strand DNA synthesis,

(4) initiation of plus-strand DNA synthesis, (5) tRNA primer removal, (6) translocation two, and (7) completion of synthesis157. These steps are explored in more detail in the following sections (for graphical representation see Figure 1-2).

1.4.2.1 Primer Binding and Minus-Strand Strong-Stop DNA Formation

The first step of reverse transcription requires the binding of the cellular tRNA primer to the primer binding sequence (PBS) of the HTLV-1 RNA genome. tRNAPRO was identified as the tRNA with sequence complementarity to the PBS of HTLV-1158.

Minus-strand DNA is synthesized from this tRNA primer towards the 5’ end of the

HTLV-1 genome where the U5 and R sequences will be generated. This round of DNA synthesis generates the minus-strand strong-stop DNA159.

1.4.2.2 Translocation One

The second step of reverse transcription requires the translocation of the minus- strand strong-stop DNA product from the 5’ end to the 3’ end of the RNA genome. For this to occur, the RNA:DNA duplex of the minus-strand strong-stop product must be separated. This is achieved via the RNase H activity of RT, which specifically degrades the RNA of the RNA:DNA duplex to allow the annealing of the newly synthesized R region of DNA to the complementary sequence on the 3’ end of the viral RNA160. This step may be facilitated by the viral nucleocapsid (NC) protein161. This strand transfer 18 appears to occur randomly, meaning that the minus-strand strong-stop product may bind to either copy of the RNA genome that was present in the virion162.

1.4.2.3 Long Minus-Strand DNA Synthesis

Once the translocation has occurred, the minus-strand short-stop DNA product serves as the primer for the continuation of minus-strand DNA synthesis. Synthesis will continue from the R region, through the U3 region, and terminate at the 5' end near the

PBS. As DNA is synthesized, the RNA in the RNA:DNA complex is degraded by the

RNase H function of RT.

1.4.2.4 Initiation of Plus-Strand DNA Synthesis

The primer for plus-strand DNA synthesis is a purine-rich sequence near the 3’ end of the viral RNA, known as the polypurine tract (ppt). The ppt is resistant to the

RNase H activity of RT and will remain hybridized to minus-strand DNA, allowing it to serve as the primer. Once DNA synthesis begins, the RNA copy of the ppt is removed by

RT. Plus-strand DNA synthesis will continue through the U5, R, and U3 regions of the viral genome and will terminate at the tRNA primer sequence at a modified base. This

DNA product is called the plus-strand strong-stop DNA.

1.4.2.5 tRNA Primer Removal

Once DNA complementary to the tRNA primer is synthesized, the tRNA primer is removed. This is achieved via the RNase H activity of RT and has been shown to be

19 mediated in a sequence specific manner in HIV-1163. The removal of the primer is not always complete however, and a single ribonucleotide base may remain in the final product163. The removal of the tRNA clears the way for the second translocation.

1.4.2.6 Translocation Two

The second translocation is mediated by the primer binding sequence of the plus- strand strong-stop DNA product. This sequence is complementary to the PBS of the minus-strand DNA product, and the hybridization of these two sequences constitutes the second translocation. This process results in a structure where both minus- and plus- strand DNA fragments are primed for completion. This second jump also completes the sequence duplication that generates the LTRs.

1.4.2.7 Completion of Synthesis

Both the minus- and plus-strands of DNA are synthesized to completion.

Synthesis of the minus strand results in the displacement of the plus strand from the 5’ end of the minus strand164. This continued synthesis and displacement will result in the generation of a linear double-stranded DNA product. This completes the steps of reverse transcription allowing the virus life cycle to proceed to the next step, nuclear entry.

1.4.3 Nuclear Entry and Integration

Following completion of reverse transcription, the viral genomic DNA translocates to the nucleus to be integrated into the host cell genome. These steps of the

20 viral life cycle are mediated by the viral integrase (IN) protein. IN binds to the LTRs of the viral genomic DNA in a sequence-dependent manner165. This binding and the association with other viral proteins forms the preintegration complex (PIC)140,165. For

HTLV-1, the nuclear entry step is thought to be dependent on the breakdown of the nuclear envelope during cellular mitosis166. However, a recent study, which identified a co-factor for IN, serine/threonine protein phosphatase 2A (PP2A), noted that this co- factor may allow the HTLV-1 PIC to enter the nucleus without cell division167.

Integration of the viral genome into the host cell genome can be broken into three steps: 3’ end processing, the strand transfer event, and DNA repair140. For 3’ end processing, IN will remove two nucleotides from the 3’ ends of both strands of the proviral DNA, resulting in a two-base pair168. The product of the 3’ end processing is then used in the strand transfer event, where the 3’ ends of the viral genomic DNA insert themselves into the host cell genome169. The strand transfer step is also mediated by

IN169. As a side effect of viral insertion, a nick is generated in the host cell DNA where several base pairs are displaced, which results in a duplication of the integration site sequence at both ends of the provirus170. This duplication is caused by the DNA repair event, where the displaced nucleotides are replaced. This step is mediated by the host cell

DNA repair machinery170. At the completion of these steps, the provirus is now integrated into the host cell genome.

Site-specific preferences for integration have been demonstrated for several retroviruses, suggesting that integration does not occur at random171-175. HTLV-1 integrations, however, are not strongly associated with any specific sites71. Several retroviruses have been shown to use a cellular cofactor to aid in the targeting of

21 integration, including lens epithelium-derived growth factor for HIV-1 and bromodomain and extraterminal domain proteins for MLV176,177. Recently, PP2A was identified as a cofactor for HTLV-1, but the exact role PP2A plays in HTLV-1 integration is still to be determined167.

1.4.4 Transcription and translation

After proviral integration, the retroviral lifecycle transitions from early phase to late phase. In the late phase of the retroviral life cycle, the virus becomes dependent on cellular machinery for subsequent steps of transcription and translation of viral RNA and proteins. This differs from the early phase, where the virus supplied the enzymes necessary for reverse transcription and integration. Transcription of viral mRNA is driven by the 5’ LTR, which contains the necessary promoters and enhancers, while the 3’ LTR contains a polyadenylation signal140. HTLV-1 transcription is enhanced by the Trans- activating transcriptional regulatory protein of HTLV-1, Tax. The transcriptional effects of Tax will be described in more detail in section 1.5.3. Briefly, Tax recruits the cyclic

AMP response element binding protein (CREB) to a set of conserved 21-base pair repeats in the LTR known as Tax-responsive elements (TREs)178,179. Tax does not directly interact with the TREs, but rather facilitates the binding of CREB to the TREs178,180.

CREB recruits cofactors such as p300 and CREB binding protein (CBP) to the 5’

LTR181,182. This complex of cofactors then promotes transcription and translation of viral gene products.

HTLV-1, like other complex retroviruses, features an array of spliced mRNAs used to generate the regulatory and accessory proteins. The full length viral mRNA

22 contains the Gag (group-specific antigen), Pol (polymerase: RT and IN), and Pro

(protease) open reading frames (ORFs)140. The other mRNA products are either singly- or doubly-spliced. These mRNAs contain the Env ORF along with the regulatory and accessory genes of HTLV-1140. The initial mRNA produced after integration is the doubly-spliced mRNA that encodes the Tax and post-transcriptional regulator of HTLV-1

(Rex) proteins. Tax-mediated transcription promotes production of its own doubly- spliced mRNA, generating a feedback loop. Once Rex reaches functional levels of expression, Rex facilitates the export of the unspliced and singly-spliced mRNA products needed for structural and enzymatic proteins183. Rex mediates this export via a Rex response element (RRE) present at the 3’ end of the mRNAs, and via the use of cellular machinery that includes the chromosome region maintenance interacting protein 1

(CRM1)184-187. The viral accessory protein p30 also plays a role in post-transcriptional regulation. p30 inhibits the export of doubly-spliced mRNAs (such as the tax/rex mRNA) by directly binding and retaining them in the nucleus188. Together Rex and p30 inhibit the expression/export of completely-spliced tax/rex mRNA, and in turn, by RNA redistribution, promote the expression of other viral gene products.

Once viral mRNAs have been generated and exported to the nucleus, translation of viral gene products can begin. For the accessory and regulatory gene products of

HTLV-1, this process is similar to normal cellular translation. For the structural and enzymatic proteins, however, a key difference exists. For one in every ten Gag proteins translated, a Gag-Pro-Pol fusion protein is produced140. This fusion protein is generated via two successive ribosomal frameshifts189. Towards the 3’ end of the Gag ORF a -1 frameshift occurs that allows the ribosome to continue into Pro140. Again, near the 3’ end

23 of the Pro ORF a second -1 frameshift occurs that allows the ribosome to continue reading through Pol140. This type of expression has two benefits. First, it ensures that the correct ratio of gene products are generated, and second, it simplifies virion assembly, which will be discussed next.

1.4.5 Virion Assembly and Budding

To generate an infectious virus, multiple elements must be assembled where the virus will eventually bud from the plasma membrane. These elements include two copies of the viral genomic RNA (gRNA), the tRNA primer, and several viral and cellular proteins. The entire process of assembly is facilitated by the Gag precursor protein. In

HTLV-1, Gag remains monomeric in the cytoplasm and oligomerizes upon reaching the plasma membrane190. The exact manner of how the Gag precursor protein reaches the plasma membrane is not fully understood, but it is known that the myristoylation of Gag is important for its interaction with the membrane156,191. The HTLV-1 Gag precursor protein collects the viral gRNA via a strong interaction between the matrix (MA) protein and a specific sequence in the viral gRNA, known as the packaging sequence192. The

Gag-Pro-Pol fusion protein is also recruited to the plasma membrane via interactions with the Gag precursor to ensure that all enzymatic proteins are present in the virion193. Fully processed Env trimers are concentrated on the plasma membrane at the site of eventual budding via an interaction with MA194. Gag precursor proteins and the Gag-Pro-Pol fusion protein oligomerizes via interactions between the CA and NC proteins of Gag195.

This interaction leads to the formation of a higher order structure underneath the plasma membrane, which curves the plasma membrane outward at the site of budding196,197. This

24 process continues until a sphere is left attached to the cell by a short stalk. Once the virus is freed from contact with the cell, assembly and budding are complete and the process of maturation can begin197.

1.4.6 Maturation

Maturation begins either during or immediately after viral budding from the host cell. Maturation is driven by the viral protease (Pro) which will cleave the Gag and Gag-

Pro-Pol precursor proteins into their functional components198. The Gag proteins then reorganize to form the important structural elements of the virus. MA remains associated with the plasma membrane, and maintains an interaction with Env197. NC coats the viral gRNA and is contained within the CA core along with reverse transcriptase and integrase197,198. This fully mature virion is then capable of starting the virus life cycle anew by infecting a permissive cell.

1.5 Viral Genome and Proteins

The HTLV-1 proviral genome is roughly 9 kb in length, and is flanked on the 5’ and 3’ ends by the long terminal repeats (LTRs) (see figure 1-3)199. The LTRs are exact duplicates, and are comprised of U3, R, and U5 regions and contain the promoter regions, polyadenylation signal sequences, and other regulatory sequences needed for proper transcription of the viral genome. The 5’ end of the plus-strand HTLV-1 genome encodes the structural and enzymatic gene products common among all retroviruses (Gag, Pol,

Pro, and Env). HTLV-1 also encodes regulatory and accessory genes, which are expressed from the historically termed pX region. The pX region is located 3’ of the 25 structural gene env, and contains four ORFs that are used to express the different genes200. HTLV-1 also encodes an antisense gene, hbz, generated from the minus strand of the viral genome201. The following sections will explore the gene products of HTLV-1 in detail.

1.5.1 Structural Genes

The structural gene products of HTLV-1 are Gag and Env. These proteins are common among all retroviruses and are important for maintaining the integrity of the virion structure and for interacting with the target cell.

1.5.1.1 Gag

The gag gene product is a 55 kDa precursor polyprotein generated from the unspliced genomic mRNA. As mentioned earlier, the Gag polyprotein is important for virus assembly and budding. After budding, Gag is cleaved by Pro to form three mature proteins. These proteins are CA (p24), MA (p19), and NC (p15). These three proteins have different roles and help maintain the structure of the virus. MA remains bound to the inner face of the membrane and has been postulated to bind Env. CA forms the viral core shell that encapsulates the two RNA genome copies, and the RT and IN enzymes. NC coats the genomic mRNA and may be involved in placement of the tRNA primer for reverse transcription202.

26 1.5.1.2 Env

The env gene product is a 62 kDa precursor protein (gp62) that is generated from a singly spliced mRNA. gp62 is cleaved by furin-like proteases to generate the surface

(SU or gp46) and transmembrane (TM or gp21) subunits203. The SU and TM subunits are linked via disulfide bonds to ensure that the extracellular SU will remain attached to the cell membrane via TM153. SU serves as the recognition molecule for HTLV-1, while TM is important for the fusion step of the virus life cycle, both of which were discussed earlier.

1.5.2 Enzymatic Genes

The enzymatic gene products of HTLV-1 are Pro and Pol. These proteins are common among all retroviruses, and provide the enzymes necessary for reverse transcription, integration, and maturation of the virus.

1.5.2.1 Pro

The pro gene product is generated from the unspliced genomic mRNA. Pro is generated as part of the fusion protein Gag-Pro or Gag-Pro-Pol, which are dependent on ribosomal frameshifts for their generation. Pro functions as the viral protease, which cleaves the Gag precursor protein into the mature subunits. It is believed that Pro has a cis-acting function that is utilized to cleave itself from the fusion proteins204.

27 1.5.2.2 Pol

The pol gene product is also generated from the unspliced genomic mRNA as part of the fusion protein Gag-Pro-Pol. Pol is made up of RT, RNase H, and IN. RT and

RNase H are located at the amino-terminus of Pol while IN is located at the carboxyl- terminus205. RT and RNase H are important for the generation of the DNA proviral genome, while IN plays a key role in inserting the provirus into the host cell genome.

1.5.3 Regulatory Genes

HTLV-1 is a complex retrovirus, which means that HTLV-1 expresses regulatory and accessory genes alongside structural and enzymatic genes. The regulatory gene products of HTLV-1 are Tax, Rex, and HBZ. These genes are essential for virus replication and efficient cellular transformation.

1.5.3.1 Tax

The regulatory gene tax is expressed from ORF IV of the pX region and encodes a 40 kDa protein. Tax has many functions during the course of HTLV-1 infection and pathogenesis. These functions include a role in viral transcription (briefly discussed in section 1.4), a role in transformation of cells, and a role in genome instability. Tax functional domains and functions will be explored in the following sections.

28 1.5.3.1.1 Functional Domains of Tax

Tax is a 353-amino acid protein generated from the doubly-spliced tax/rex mRNA

(see Figure 1-4)206. Tax has been extensively studied, and several functional domains have been identified. Tax features a CREB activation domain at the amino terminus of the protein through which transcription of the viral LTR is activated. Within the CREB activation domain is a zinc finger and nuclear localization signal that are required for import of Tax into the nucleus. In the central region of Tax are two leucine zipper-like regions (LZ), which are important for the ability of Tax to dimerize and interact with

DNA. The amino terminal LZ is important for Tax activation of the classical NF-κB pathway, while the carboxyl terminal LZ is important for the activation of the alternative

NF-κB pathway. Between the two LZs is a nuclear export signal. Tax also features a

CREB/activating transcription factor (ATF)-activating domain, which is important for the transactivation of CREB/ATF pathways. The final key functional domain of Tax is a

PDZ-binding motif (PBM). The PBM consists of the last four amino acids of the protein.

Via the PBM, Tax interacts with several key proteins, including DLG-1, which play a key role in cellular transformation. Deletion of the PBM from Tax has been shown to result in a delay of cellular transformation207.

1.5.3.1.2 Role of Tax in Viral Transcription

Tax binds to three 21-base pair repeats present in the 5’ LTR, named the Tax- responsive elements (TREs)208-210. The TREs consist of three domains, A, B, and C211.

The A and C domains of the TREs are C/G rich, while the B domain is similar in sequence to the cellular cyclic adenosine monophosphate (cAMP) response element

29 (CRE)211,212. This similarity in sequence has led to the TREs being referred to as the viral cAMP response element (vCRE). Dimerized Tax recruits CREB to the vCRE213. While

Tax has not been shown to interact directly with DNA, it does associate with the C/G rich sequences in the A and C domains of the TREs178,180. This association increases the specificity of the Tax:CREB complex for the vCRE. Once at the vCRE, a CREB homodimer or a CREB: ATF heterodimer will be formed214.

The next step of Tax-mediated transcription requires the recruitment of two transcriptional co-activators to the TREs, CBP and p300. Tax induces the recruitment of

CBP and p300 via the phosphorylation of CREB at serine 133, which occurs in response to binding at the vCRE214-216. CBP and p300 are two highly similar histone acetyltransferase proteins217. Once recruited to the TREs, CBP/p300 acetylate the histone tails, which results in chromatin remodeling and increases transcription218. Other proteins have been identified to play a role in Tax-mediated chromatin remodeling, for example,

Nucleosome Assembly Protein 1 (NAP1)219. NAP1 was demonstrated to be required for the release of nucleosomes from the promoter, resulting in an increase in transcription219.

Another Tax interacting protein that is important for chromatin remodeling is P300/CBP- associated factor (PCAF)220. A well-studied mutant of Tax (M47) is deficient for transcriptional activation of the viral promoter, and is also incapable of interacting with

PCAF220. While PCAF does function as a histone acetyltransferase, this function is not important for its role in Tax-mediated transcription220.

30 1.5.3.1.3 Role of Tax in Transformation

Tax plays an important role in HTLV-1-mediated cellular transformation, a key aspect of disease development. Loss of Tax by mutation or deletion renders HTLV-1 incapable of inducing cellular transformation in in vitro assays221. Conversely, Tax over expression, in the absence of other viral factors, has been shown to induce cellular transformation222. The contribution of Tax to cellular transformation is dependent on the interaction of Tax with, and dysregulation of, several important signaling pathways involved in cell survival, proliferation, and genomic stability. These include the classical and alternative NF-κB pathways, the PI3K/Akt/mTOR pathway, and several others223.

This section will detail the role of Tax in HTLV-1 mediated cellular transformation, with an emphasis on the NF-κB pathways.

A. Tax and the NF-κB Pathways

The NF-κB family of transcription factors is involved in the regulation of cell survival and proliferation; HTLV-1 Tax has been shown to activate members of the NF-

κB family. The NF-κB family of proteins includes the Rel proteins (RelA, RelB, and c-

Rel) and the NF-κB proteins (p50/p105, and p52/p100), which share an amino-terminal

Rel homology domain (RHD) and a carboxyl-terminal transactivation domain224. The

NF-κB subfamily of proteins exist as inactive precursor proteins (p105 and p100) and active truncated variants (p50 and p52). The NF-κB proteins are cleaved via proteolysis to remove Ankyrin repeats that inhibit their nuclear translocation225. Several different homodimers and heterodimers are formed between the RHDs of the NF-κB family members224. Because Rel family members lack DNA-binding ability, and NF-κB family

31 members do not feature transactivation domains, a functional NF-κB transcription factor necessarily consists of a member from the Rel and NF-κB subfamilies. The most important and well-studied of the NF-κB dimers is the p50/RelA heterodimer, which is sometimes referred to as ‘NF-κB’.

Typically, in an unstimulated cell, NF-κB dimers are retained in the cytoplasm via the ankyrin repeats of the NF-κB subfamily, which block nuclear localization of the dimers224. This mechanism renders the proteins functionally inactive. Interestingly, the

NF-κB family member p50 is the predominant variant in the cell as opposed to its precursor protein, p105. Though p50 lacks the ankyrin repeats of the precursor, it is retained in the cytoplasm through an interaction with an inhibitory protein, Inhibitor of

κB (IκB)225. Several IκB family members exist (for example IκBα and IκBβ) with distinct affinities for the different NF-κB dimers. The IκB family inhibits the NF-κB dimers by blocking nuclear localization signals or by inhibiting DNA binding activity225. Activation of NF-κB transcription is achieved via two distinct signaling cascades, the classical and alternative NF-κB pathways, both of which will be discussed below. Both NF-κB pathways involve the activation of an IκB kinase (IKK) complex. The IKK complex is made up of three proteins, two of which are catalytic subunits important for the phosphorylation IκB (IKKα and IKKβ), and a third, NF-κB essential modifier (NEMO or

IKKɣ), which serves as a scaffold for the catalytic subunits225. The IKK complex is activated by upstream IKK kinases to promote the phosphorylation and degradation of the IκB proteins225. This degradation event allows the NF-κB dimer to translocate into the nucleus to transactivate target genes225.

32 Initiation of the classical (canonical) NF-κB pathway results in the activation of the p50/RelA heterodimer, the predominant NF-κB dimer (Figure 1-5)224. This pathway is activated in response to almost all NF-κB stimulating factors, including inflammatory cytokines, antigen receptors, and double-stranded RNA225. Upon stimulation of the cell, the IKKα, IKKβ, and NEMO complex is phosphorylated and activated, which leads to the phosphorylation, ubiquitination, and eventual proteasome-mediated degradation of

IκBα. The p50/RelA heterodimer subsequently translocates to the cell nucleus to drive transcription of target genes225. Cellular functions modulated by the classical NF-κB pathway include cell survival, proliferation, and inflammation226. The classical NF-κB pathway is quickly activated in response to stimulus, with IκBα degradation occurring within minutes of stimulus recognition. The pathway is also short lived, and will be quickly repressed after stimulus is lost, due to the up regulation of IκBα expression by the pathway itself225.

Initiation of the alternative (non-canonical) NF-κB pathway results in the activation of the p52/RelB heterodimer (Figure 1-5)224. This pathway is activated by a small subset of stimuli that include Receptor Activator of NF-κB Ligand (RANKL) and several Tumor Necrosis Factor (TNF) receptor-associated factors225. It is important to note however, that any stimulus that activates the alternative NF-κB pathway will also activate the classical pathway225. Furthermore, classical NF-κB promotes alternative NF-

κB via up regulation of p100 and RelB protein expression, creating a link between these two pathways226. Upon stimulation of the cell, an IKKα homodimer is phosphorylated and activated by the NF-κB-inducing kinase (NIK)225. Activation of the IKKα homodimer leads to the phosphorylation of p100, which is usually dimerized with

33 RelB225. Upon phosphorylation and subsequent ubiquitination, p100 is cleaved via a proteasome-mediated mechanism into the active p52 protein224. This cleavage event removes the inhibitory ankryin repeats that allows the p52/RelB dimer to translocate into the nucleus and drive expression of target genes224. Cellular processes modulated by the alternative NF-κB pathway include lymphoid organogenesis and dendritic cell activation226. The alternative NF-κB pathway is slowly activated due to the kinetics involved in processing p100 to p52, and can take up to several hours for complete activation. In turn, the alternative pathway is also long-lived, with cellular transcription remaining active after stimulus is lost225.

Tax activates the classical NF-κB pathway via its interaction with, and activation of, the IKK complex. HTLV-1 infection and Tax expression in cells leads to constitutive activation of the IKK complex and continuous degradation of the IκBα protein227,228. Tax maintains this activation via interaction with NEMO where multiple IKK complexes come into close proximity and cross-activate IKKα and IKKβ229-231. Tax-induced activation of the IKK complex is increased by ubiquitylation of Tax, which promotes the translocation of Tax to the cytoplasm232. Tax also modulates NF-κB activity in the nucleus, where Tax recruits CBP/p300 to NF-κB target genes in a manner similar to Tax activation of viral transcription182,233. Tax interacts directly with RelA to promote this increase in NF-κB gene transcription182. SUMOylation of Tax has been shown to be important for both Tax nuclear localization and for the promotion of NF-κB transcription234,235.

Tax activates the alternative NF-κB pathway via its ability to induce the processing of p100 to the active p52 protein236. Interestingly, Tax-induced activation of

34 the alternative NF-κB pathway differs from normal cellular activation231. Tax promotes p100 processing in the absence of NIK activation and instead utilizes the interaction with

NEMO to promote the IKKα phosphorylation of p100231,237. Tax is belived to mediate this activity via interaction with both NEMO and p100, bringing the necessary components into close proximity231. The possibility of another cellular factor being involved in this activation is currently under investigation.

Tax activation of the NF-κB pathways contributes to HTLV-1 transformation in several different ways, including effects on proliferation, apoptosis avoidance, DNA damage response, and cell cycle checkpoint regulation. In fact, most of the oncogenic potential of Tax can be traced to upregulation of NF-κB activity. The importance of Tax- mediated NF-κB activation in transformation was demonstrated by the inability of

HTLV-1 Tax M22 mutant virions to transform primary cells in vitro221. The Tax M22 mutant is defective for classical NF-κB pathway activation238. Furthermore, treatment of

HTLV-1-infected cells with NF-κB inhibitors induces apoptosis. It has also been shown that mutations in Tax that cannot activate the alternative NF-κB pathway are not as efficient at inducing IL-2-independent growth of the CTLL-2 cell line239.

A recent study demonstrated that Tax-mediated hyperactivation of the classical

NF-κB pathway led to cellular senescence, contrary to the notion that NF-κB activity increased proliferation240. This group linked Tax-mediated senescence to increased stability and expression of cyclin-dependent kinase inhibitors, p21 and p27. Furthermore, the authors demonstrated that inhibition of NF-κB activity alleviated the senescence phenotype of Tax over-expression. It was later shown that the senescence phenotype was driven by the activation of the classical NF-κB pathway, not the alternative pathway241.

35 Taken together, these data demonstrate the need for HTLV-1-infected cells to counteract this senescence phenotype to stimulate cell proliferation.

B. Tax and the PI3K/Akt/mTOR Signaling Cascade

The PI3K/Akt/mTOR signaling cascade is involved in cell survival, growth, proliferation, and other cellular processes (Figure 1-6)242. This signaling cascade is activated at the plasma membrane by a growth factor binding to receptor tyrosine kinases, leading to the activation of the PI 3'-OH kinase (PI3K) and subsequent conversion of the

243 phosphoinositide PI-3,5-bisphosphate (PIP2) to PI3,4,5-trisphosphate (PIP3) . PIP3 binds to the Ser/Thr kinase Akt and promotes Akt phosphorylation by phosphoinositide- dependent kinase 1 (PDK1) or PDK2 at threonine 308 or serine 437, respectively244,245.

Activated Akt phosphorylates and inhibits a multitude of targets including TSC2, which is the negative regulator of mTOR246. Through mTOR and other Akt targets, this pathway promotes cell growth and survival. However, the PI3K/Akt/mTOR signaling cascade is also negatively regulated by several proteins. The PH domain leucine-rich repeat protein phosphatase (PHLPP) has been shown to dephosphorylate and inactivate Akt post- activation, while phosphatase and tensin homolog, deleted on chromosome ten (PTEN)

242,247 inhibits Akt prior to activation . PTEN promotes the hydrolysis of PIP3 to PIP2, which inhibits the phosphorylation of Akt244. The PI3K/Akt/mTOR pathway is often found dysregulated in many cancers, including ATL.

In normal T cells, the PI3K/Akt/mTOR signaling cascade is activated by IL-2 stimulation245. HTLV-1-transformed cells demonstrated activated PI3K/Akt/mTOR in the absence of IL-2 stimulation248. Furthermore, drug-induced inhibition of Akt signaling in

36 HTLV-1-transformed cells led to cell cycle arrest and apoptosis249,250. This finding demonstrated the importance of maintaining activation of this pathway in HTLV-1- transformed cells. The PI3K/Akt/mTOR signaling cascade has also been implicated in regulation of telomerase activity, where inhibition of Akt drastically reduced telomerase activity in HTLV-1-infected cells251. In addition, Akt activation in HTLV-1-infected cells has been shown to promote activation of the transcription factor activator protein-1

(AP-1)252. Through AP-1 activity, HTLV-1 possesses another method of regulating proliferation. How Akt signaling is maintained by HTLV-1 is currently under investigation. The current hypothesis involves a direct interaction between Tax and PI3K to promote Akt activation, but current work demonstrates that Tax may be involved in negative regulation of PTEN and PHLPP140.

C. Tax, Cell Cycle Control, and Proliferation

Tax uses multiple mechanisms to promote cellular proliferation, including activation of the NF-κB and Akt pathways. Tax interacts directly with the cell cycle machinery via cyclin-dependent kinase 4 (CDK4)253. This interaction activates CDK4, which promotes the phosphorylation of the tumor suppressor protein retinoblastoma (Rb), an inhibitor of the transcription factor E2F254. Phosphorylation of Rb releases E2F, which promotes the transition of the cell cycle from G1 to S phase254. Tax also promotes E2F directly via up regulation of the E2F-1 promoter255. Tax interacts with CDK2 and CDK6 and increases Cyclin D levels in infected cells, which stimulates proliferation253,256,257.

Expression of the CDK inhibitors (CKIs) p18INK4c, p19INK4d, and p27Kip1 is decreased in the presence of Tax, which promotes the activities of CDK4 and CDK6257-259. Tax also

37 promotes uncontrolled cell cycle progression via its ability to interact with, and inhibit,

Chk1 and Chk2, which drives progression through the G2/M checkpoint260-262.

Tax also regulates proliferation through several other signaling cascades. For example, Tax modulates the activity of several interleukin (IL) family members, including IL-2 and IL-13263,264. Tax activates IL-2 signaling via up-regulating the expression of the α-chain of the IL-2 receptor, and via increasing activity on the IL-2 promoter263,265. These two activities lead to a feedback loop of IL-2 stimulation and proliferation in infected cells. Tax increases the expression of IL-13 in infected cells via transactivation of the AP-1 pathway, which promotes proliferation of infected cells. Tax modulates the Jak/STAT and TGF-β signaling pathways as well264. In HTLV-1-infected cells, signal transducer of activated T cells (STAT) is hyper-activated leading to increased proliferation266. Transforming growth factor beta 1 (TGF-β1) inhibits growth of normal cells; Tax represses TGF-β1 via interactions with Smad3 and Smad4, thereby increasing the potential for cells to proliferate267,268.

D. Tax and DNA Damage

The normal cell has several mechanisms to sense and repair DNA damage, including: base excision repair (BER), nucleotide excision repair (NER), mismatch repair

(MMR), and non-homologous end-joining (NHEJ). Tax has been shown to inhibit all of these DNA damage repair mechanisms, which establishes an environment of genetic instability that contributes to oncogenesis. BER is involved in the repair of base lesions that arise in response to internal and external factors269. Tax inhibits BER by reducing polymeraseβ expression, which renders HTLV-1-infected cells incapable of repairing

38 lesions generated by UV light, quercetin, or hydrogen peroxide270,271. The NER pathway removes bulky and helix-distorting lesions from DNA272. Left unchecked, these lesions result in changes to the DNA sequence in the next replication cycle272. Tax alters the

NER pathway in two ways: via inhibition of p53 and via increases in proliferating cell nuclear antigen (PCNA) expression. While Tax does not directly affect the expression of p53, high levels of Tax have been correlated with a functionally inactive p53 protein, which is incapable of activating NER273. In normal cells, DNA replication is promoted by

PCNA but is inhibited by p21 in response to DNA damage274. When PCNA levels are increased in cells, as when Tax is present, the p21 block of DNA replication is overcome in the presence of DNA damage275. MMR detects and fixes base mismatches and insertion/deletion events generated during replication276. MMR is reduced in ATL cells via suppression of genes involved with regulation of this pathway, including the human

MutL homolog 1 and MutS homolog 2277. Cells utilize NHEJ to repair DNA double- strand breaks (DSBs). Tax inhibits DSB repair via its interaction with DNA-dependent protein kinase (DNA-PK), the sensor and signal transducer of NHEJ278. Tax activates

DNA-PK constitutively, inhibiting the ability of the enzyme to sense true DNA damage events. Tax also reduces the expression of a subunit of DNA-PK (kinase domain 80), which reduces the overall levels of functional DNA-PK278.

Tax also promotes chromosomal instability and aneuploidy in HTLV-1 infected cells279. Tax inhibits proper chromosome separation during mitosis via an interaction with the MAD1 protein280. Tax also promotes amplification of abnormal centromeres, increasing improper segregation of chromosomes during mitosis281,282. These effects of

Tax are thought to lead to the physical appearance of ATL cells: lobulated nuclei that

39 resemble flower petals. Moreover, inhibition of correct segregation of chromosomes has been shown to increase the likelihood of cell transformation283.

1.5.3.1.4 Differences between Tax-1 and Tax-2

Although HTLV-1 and HTLV-2 share 70% nucleotide similarity, they are distinct in their pathogenic outcomes. HTLV-1 causes ATL in a small subset of infected individuals, while HTLV-2 is not associated with malignancy. Many studies have sought to understand what differences exist that would reveal why infection with these viruses result in two strikingly different pathogenic outcomes. A large amount of work has focused on the Tax proteins of the two viruses, Tax-1 from HTLV-1 and Tax-2 from

HTLV-2284,285. These two proteins share 85% amino acid sequence similarity and several functional domains (Figure 1-4)206. Initial studies on the transforming potential of Tax-1 and Tax-2 demonstrated that Tax-1 was more efficient than Tax-2 in inducing IL-2 independence in the CTLL-2 cell line239,286. Mutational analysis revealed two domains of

Tax-1 not present in Tax-2 that were important for this difference in transforming activity: the PBM and the second LZ239,286.

The PBM domain of Tax-1 interacts with the PDZ domains of several cellular proteins including DLG-1, Scribble, and TIP-1287-289. DLG-1, a tumor suppressor, is inactivated via binding of the PBM domain of Tax-1, resulting in increased transformation. Deletion of the PBM of Tax-1 prevented this binding and resulted in a delay in transformation of primary peripheral blood mononuclear cells (PBMCs) in vitro207. When Tax-1 mutants lacking PBM were used in cellular transformation assays, the small subset of cells that transformed expressed low levels of DLG-1 compared to

40 wild type-transformed cells290. This finding indicated that successful transformation requires interaction between the PBM of Tax-1 and cellular proteins such as DLG-1.

Another difference between Tax-1 and Tax-2 is the fact that Tax-2 cannot activate the alternative NF-κB pathway; which has been mapped to the PBM and second LZ domains of Tax-1239,291. It is possible that transformation by HTLV-1 depends on the interaction of

Tax-1 with the PDZ domains of cellular proteins involved in cell growth signaling or the activation of the alternative NF-κB pathway. These will be further investigated in this dissertation.

1.5.3.2 Rex

The regulatory gene rex is expressed from ORF III of the pX region, and encodes a 27 kDa protein292. Rex is essential for the HTLV-1 lifecycle due to its role in post- translational gene regulation, which was mentioned briefly in the virus life cycle section

(section 1.4)183. During normal cellular gene transcription, intron-containing mRNAs are retained in the nucleus until they are fully processed or degraded. HTLV-1 expresses several gene products from mRNAs containing introns, and must overcome the default normal cellular pathways for proper gene expression. HTLV-1 achieves this abnormal mRNA regulation through Rex, which exports the unspliced or incompletely spliced viral mRNA from the nucleus to the cytoplasm. Rex mRNA export activity is tied to several features of the protein, including Rex binding to viral mRNA, Rex multimerization, and

Rex interaction with CRM1.

Rex binds to target viral mRNA via the Rex response element (RxRE) present on all HTLV-1 RNAs184. The RxRE is a 205 nucleotide sequence located in the U3 and R

41 regions of the 3’ LTR184. The secondary structure of RxRE consists of four stem loops, which are important for Rex-dependent mRNA export184. Rex interacts with this region of the RxRE via an arginine-rich RNA binding domain located at the amino-terminal end of the protein293. HTLV-1 mRNA also features a cis-acting repressive sequence (CRS) in both LTRs, which functions as a negative regulator of mRNA export294,295. The 5’ CRS inhibits the release of only unspliced mRNA, due to the fact that the CRS is removed during splicing. The 3’ CRS, however, is present in all HTLV-1 mRNAs, spliced or unspliced, which contributes to the retention of the mRNAs in the nucleus295. It was recently shown that Rex is required for the export of unspliced or singly spliced viral mRNAs but not for the doubly spliced mRNA products296. This finding was partially challenged by another group, which claims that tax/rex mRNA export is dependent on

Rex activity297.

Direct interaction of Rex with the RxRE as a single molecule has been shown, but

Rex multimerization is expected to be important for proper export of viral mRNAs298-300.

Mutants of Rex that fail to multimerize act as dominant negatives for Rex activity298.

Furthermore, Rex multimerization also can be mediated via the nuclear export receptor

CRM1, as shown by Rex mutants that failed to bind CRM1 failed to multimerize301. Rex interacts with CRM1 via its nuclear export signal, and is shuttled out of the nucleus along with the mRNAs it is exporting301. Once released from the viral mRNAs in the cytoplasm, Rex binds importinβ and is transported back to the nucleus, to begin the shuttle process again302.

Rex activity is regulated via phosphorylation, initially identified at residues serine

70, 177, and threonine 174303. A more recent study identified phosphorylation sites at

42 serine 36, 70, 97, and 106, and threonine 22, 37, and 174, but did not confirm the initial observation of phosphorylation at serine 177304. This recent study also demonstrated the phosphorylations at serine 97 and threonine 174 are necessary for Rex function304.

There are two final aspects of Rex function worth highlighting. The first is the proposed mechanism of how Rex inhibits splicing of unspliced and singly spliced mRNA products by interacting with the splicing machinery of a host cell to inhibit splicing305.

However, the exact mechanism of this action has not yet been reported. The other function is how Rex export activity affects tax/rex gene expression. By increasing the export of unspliced and singly spliced mRNAs, Rex negatively regulates the expression of tax/rex. This is beneficial for HTLV-1 infection, by lowering expression of the highly immunogenic Tax protein183.

1.5.3.3 HBZ

HTLV-1 expresses a majority of its gene products via the sense strand, but one gene product is expressed from the anti-sense strand of the provirus. HTLV-1 basic leucine zipper factor (HBZ) is the antisense gene product, and the most recently discovered HTLV-1 protein201. Hbz is expressed as two transcriptional isoforms: the unspliced and spliced variants201,306,307. These two mRNAs differ slightly in the coding region for the HBZ protein, having a nearly identical peptide sequence with the exception of the first several amino acids308. While unspliced hbz was discovered first, the spliced product was determined to be the predominant product in infected cells and ATL cells, and as such, the spliced product will be discussed herein309. HBZ is a nuclear protein composed of three functional domains: the activation domain (AD), the central domain

43 (CD), and the basic leucine zipper (bZIP)308. The AD is important for HBZ interaction with cellular transcription factors such as CBP/P300310. The CD is important for HBZ ability to promote Foxp3 expression, leading to the regulatory T cell (Treg) phenotype of infected cells311. This Treg phenotype induced by HBZ could allow ATL cells to suppress the immune system312. The bZIP domain of the protein allows HBZ to dimerize with other bZIP domain containing proteins, such as JunB and JunD313,314.

While Tax has been demonstrated to play a crucial role in cellular transformation mediated by HTLV-1, unchecked Tax activity is linked to cellular senescence and a strong immune response. HBZ counteracts these detrimental events via its ability to inhibit several functions of Tax. HBZ inhibits Tax-mediated transcription of itself and other positive sense genes via competitive binding between HBZ/CBP/p300310. This inhibition reduces the amount of Tax protein generated, and decreases the potential for immune detection. HBZ also inhibits the activity of the classical NF-κB pathway, which is activated by Tax. HBZ inhibits the DNA binding potential of RelA, while also promoting the degradation of RelA315. By lowering the activity of the classical NF-κB pathway, HBZ is counteracting the cellular senescence event induced by Tax240. It is interesting to note that transcription from the 5’ LTR is inhibited in most ATL cases, which blocks production of all sense strand genes79-82. The 3’ LTR remains functionally intact, allowing the expression of HBZ. In fact, HBZ is the only HTLV-1 gene product detected in all cases of ATL, which supports the key role of HBZ in promoting cellular transformation in HTLV-1-infected cells83.

Several other functions of HBZ have been discovered. HBZ inhibits Rex mRNA export in a manner that is still yet to be determined316. By blocking Rex, HBZ can lower

44 overall HTLV-1 virion production and promote a latent-like infection. HBZ has also been shown to promote cellular proliferation via the AP-1 and the Wnt signaling pathways317,318. In addition, shRNA-mediated knockdown of HBZ in HTLV-1 infected/transformed cell lines reduces proliferation84. Interestingly, the hbz mRNA has also been shown to induce proliferation and inhibit apoptosis via an up regulation of the anti-apoptotic gene survivin83,319. The HBZ protein inhibits apoptosis via down regulation of the pro-apoptotic gene Bim, and knockdown of HBZ in HTLV-1-infected cells leads to an increase in Bim expression320. HBZ has also been shown to induce genetic instability via promotion of DSBs321. In a transgenic mouse study, HBZ expression was shown to lead to lesions similar to those seen in some HTLV-1-infected individuals312. All of these functions of HBZ contribute to the oncogenic potential of the protein.

1.5.4 Accessory Genes

The HTLV-1 accessory gene products include p30, p12/p8, p13, and p21. These products are dispensable for in vitro transformation of human PBMCs but several are required for viral persistence as shown in a rabbit model of infection.

1.5.4.1 p30

The accessory gene p30 is expressed from a doubly spliced mRNA via ORF II of the pX region. p30 is a 241 amino acid nuclear protein that plays a key role in post- transcriptional regulation of HTLV-1 genes322,323. p30 lowers the expression of the tax/rex viral mRNA by binding to the second splice junction of the mRNA, and retaining it in the nucleus188. This function of p30 is specific for tax/rex mRNA, and results in 45 decreased viral gene expression188. This p30 activity is distinct from the post- transcriptional activity of Rex, because p30 reduces virion production while Rex promotes virion production. p30 also alters Tax transcriptional activity via competitive binding to p300, where high levels of p30 expression reduce the transcriptional activity of the TREs324. Microarray analysis revealed that p30 modulates the expression of several cellular genes involved in the regulation of apoptosis and transcription325,326.

p30 has also been demonstrated to affect cell cycle regulation and DNA damage recognition. The G2/M checkpoint is modulated by p30 expression. When p30 is present,

Chk1 phosphorylation is more prominent, leading to the phosphorylation of Cdc25c, which promotes progression of the cell cycle from G2 to M phase327. HTLV-1- immortalized cells lacking p30 treated with etoposide and camptothecin were more susceptible to apoptosis than wild type HTLV-1 immortalized cells327. p30 also modulates progression into S phase of the cell cycle by inhibiting the function of the

CDK2/CyclinE complex328. ATM is involved in recognition of DSBs, and p30 has been shown to lower the activation and expression ATM329. It is believed that p30 targets

ATM for degradation through Regɣ, a nuclear proteasome activator. Lowering of ATM expression results in a decrease in activation of p53 in response to DSBs, which helps infected cells avoid DNA damage-induced apoptosis329.

1.5.4.2 p12/p8

p12 is expressed from a singly spliced mRNA encoded by ORF 1 in the pX region. p12 is a membrane bound protein that is localized to the endoplasmic reticulum

(ER) and Golgi322,323. p12 appears to play a role in dendritic cell infection, but deletion of

46 p12 from the provirus did not alter PBMC immortalization in vitro or persistent infection in rabbits330. The major role for p12 in HTLV-1 infection is promoting persistence. p12 reduces expression of ICAM-1 and ICAM-2 on the surface of infected cells, which prevents killing by natural killer cells331. p12 can be proteolytically cleaved into a carboxyl terminal product, p8, which localizes at the cell membrane due to the removal of the ER retention signal332. p8 may mediate HTLV-1 transmission via activation of the lymphocyte function-associated antigen-1, which promotes cell-to-cell contact of T cells, and increases the potential for viral transmission332.

A recent study investigated the expression of p12 and p8 in 160 HTLV-1 infected individuals333. This study demonstrated a correlation between balanced expression of p12 and p8 with high levels of viral load, a factor that correlates with disease development.

Furthermore, this study investigated HTLV-1 molecular mutants that expressed p12 and p8 equally, p12 predominantly, or p8 predominantly. They demonstrated that p8- predominant viruses transmitted more efficiently in vitro compared to the other two mutants. In a Rhesus macaque animal model, the study demonstrated that equal expression of p12 and p8 molecular clones caused seroconversion in three of four animals, while one out of four animals seroconverted after inoculation with the p8- predominant mutant, and zero out of four animals seroconverted after inoculation with the p12-predominant mutant. Collectively, these data demonstrated that equal amounts of p12 and p8 expression are important for persistence, viral spread, and disease development333.

47 1.5.4.3 p13

p13 is expressed as a singly spliced mRNA encoded by ORF II of the pX region of the HTLV-1 genome. p13 is a mitochondrial-associated protein of 87 amino acids, which is identical to the carboxyl-terminal 87 amino acids of p30322,323. Mutations of p13 in HTLV-1 virions did not affect infectivity of the virus in rabbits, but did block efficient viral persistence334. Furthermore, p13 expression has been tied to increased reactive oxygen species production and apoptosis335,336. These findings could explain why persistent infection is not established by p13-deficient HTLV-1 in vivo, but further studies are required to establish the mechanism.

1.5.4.4 p21

p21 is expressed from ORF III of the pX region and is an isoform of Rex generated via an alternative splicing event337. Because p21 lacks the nuclear localization signal present in Rex, it is localized in the cytoplasm338. The exact role of p21 in HTLV-1 infection is not clear, but is predicted to serve as a negative regulator of Rex305.

1.6 HTLV-1 Experimental Models

1.6.1 Cell Culture

Cell-free HTLV-1 infection is inefficient and difficult to induce in vitro because of the viruses’ requirement for cell-to-cell contact to facilitate efficient infection. The current, and most successful, method for infection is via co-cultivation where a virus producer cell line is generated by transfection of a molecular clone of HTLV-1 and

48 selection for an incorporated provirus207,339-342. These cells are then lethally irradiated and co-cultivated with the target cells, usually human PBMCs. Using this assay, virus infectivity and PBMC transforming ability can be monitored in parallel in vitro. Because producer cell lines are generated by transfection of molecular clones of HTLV-1, mutations can be inserted into the molecular clone to determine effects on infectivity and transformation. These assays are useful for understanding the molecular importance of mutations or specific viral proteins. However, results of cell culture-based experiments should be confirmed in an animal model, due to the inability to replicate an immune system in a culture-based assay.

1.6.2 Animal Models

Several different animal models have been established for HTLV-1 infection including rabbits, rats, non-human primates, and mice343-345. Each model has strengths and weakness. Rabbits are used as a model of HTLV-1 infection and persistence, but not for pathogenesis, because they do not develop disease341,346. Inoculation of rabbits with producer cells that generate mutant viruses allows investigations of how mutations affect viral replication, proviral load, and alter viral persistence in the presence of a functional immune system. This information adds to the results that can be generated via the cell culture based transformation assay. Rabbits have also been used in the past for vaccine trials, and investigations of the determinants of HTLV-1 tropism347-351. While rats have been used as a model for transmission of HTLV-1, infection of the Wistar-King-

Aptekman strain leads to a spastic paraparesis phenotype352-355. This rat model is becoming important for understanding HAM/TSP-like disease development. Non-human

49 primates are utilized more infrequently, but serve as an important model for disease, infection, and vaccine efficacy studies356-361.

The mouse model is the most prominent animal model in HTLV-1 studies.

Infection of mice with HTLV-1 serves as a model of virus persistence similar as the rabbit, but also disease development362-364. Transgenic mice expressing the HTLV-1 protein Tax and HBZ have been shown to develop tumors365,366. Tax transgenic mice develop a wide array of diseases depending on the promoter used, but use of the granzyme B promoter to drive Tax expression resulted in the development of a leukemia/lymphoma syndrome similar to ATL367. Transgenic mice expressing HBZ using a CD4 promoter have also been shown to develop leukemia/lymphoma, further demonstrating the oncogenic potential of this protein312. While transgenic mouse studies allow exploration of how specific viral proteins induce disease, most studies have not been done in the context of other viral gene products. Two more recently developed mouse models have been used to investigate disease in the context of the whole virus: immune-deficient mice and humanized mice. ATL cell lines are injected into immune- deficient mice to monitor the tumorigenic potential of the cell lines. Tumors can be detected at the injection site in most mice; leukemia may develop in some mice368,369.

These studies allow researchers to test different drugs that may lower the tumor burden of

ATL. The humanized mouse model is an up and coming model for HTLV-1 studies. The advantage of humanized mice compared to immune-deficient mice is that disease development can be monitored at the start of infection, while immune-deficient mice are used to analyze the spread of existing leukemic cells370-372. This model will allow researchers to monitor the effect mutations in the virus may have on disease

50 development, and is a strong addition to the transformation assays already possible in cell culture.

1.7 Conclusions

HTLV-1 is a human retrovirus associated with several diseases, including ATL.

The oncogenic potential of the virus is tied primarily to the Tax protein, but recent studies have demonstrated HBZ also contributes to oncogenesis. The closely related virus

HTLV-2 is not associated with any human malignancies. This has led to research examining the differences between these two viruses, and how they may contribute to the different pathogenic outcomes. HTLV-1 Tax-1 activates the alternative NF-κB pathway, while HTLV-2 Tax-2 does not. The work detailed in this dissertation investigated the importance of the alternative NF-κB pathway for HTLV-1-mediated cellular transformation in vitro, and aimed to determine the mechanism of alternative NF-κB activation via Tax-1. Also described are investigations into a novel binding partner of

Tax-1 and the mechanism of how hbz mRNA promotes cellular proliferation.

51

Figure 1.1 The Generic Retrovirus Life Cycle A schematic diagram of the retrovirus lifecycle. Each step of the lifecycle is numbered. 1) A free virion moving into the vicinity of a target cell. 2) The virion attaches to the target cell via an interaction between the viral Envelope and the target cell receptor molecule. 3) The virion plasma membrane fuses to the plasma membrane of the cell, releasing the core into the cytoplasm. 4) The viral core undergoes partial uncoating upon entry to the cytoplasm. 5) Reverse transcription of the genomic viral mRNA will proceed to generate the double-stranded DNA provirus (6). 7) The preintegration complex forms and transports the provirus into the nucleus. 8) Once into the nucleus the provirus integrates into the host cell genome. 9) Once integrated, viral mRNAs are transcribed. 10) Viral proteins are translated. 11) Virus assembly and budding occur at the plasma membrane. 12) An immature virion is released from the cell. 13) The virus matures and is now capable of infecting another cell.

52

Figure 1.2 The Process of Reverse Transcription A schematic diagram of reverse transcription. RNA is shown in red and DNA is shown in blue. 1) The tRNA primer binds to the PBS. 2) Minus-strand strong-stop DNA is generated. 3) The RNA:DNA duplex is degraded by the RNase H function of RT. Continued

53 Figure 1.2 Continued 4) The minus-strand strong-stop product translocates to the 3’ end of the genome. 5) DNA synthesis continues, while RNase H degrades the RNA complexed with DNA. 6) Plus strand synthesis begins from the PPT, which is resistant to RNase H degradation. 7) The tRNA primer is removed from the nascent DNA. 8) Translocation two occurs, allowing continued synthesis of the final double-stranded DNA. 9) The completed double-stranded proviral DNA product. Figure adapted from Fields Virology140.

54

Figure 1.3 The HTLV-1 Proviral Genome and Coding mRNAs A) Schematic representation of the proviral genome. The grey lines indicate the coding regions of the genome. B) Schematic representation of the viral mRNAs generated by HTLV-1. Figure adapted from Fields Virology140.

55

Figure 1.4 Functional Domains of Tax-1 and Tax-2 Schematic representation of the functional domains of Tax-1 (above) and Tax-2 (below).

Abbreviations: NLS: Nuclear localization signal ZF: Zinc finger LZR: Leucine zipper like region PBM: PDZ binding motif.

56

Figure 1.5 The Classical and Alternative NF-κB Pathways Schematic diagram of the classical and alternative NF-κB pathways. The Classical Pathway (left) is activated via stimulation through the IKKα/IKKβ/NEMO complex. This active complex then phosphorylates IκB leading to its degradation. Once IκB is degraded, the p50/RelA heterodimer is free to translocate into the nucleus and drive transcription of its target genes. The Alternative Pathway (right) is activated via stimulation through the NIK protein, which phosphorylates and activates the IKKα homodimer. The active IKKα complex phosphorylates p100 leading to its processing to p52. The p52/RelB complex is then free to translocate into the nucleus and drive transcription of its target genes.

57

Figure 1.6 The PI3K/Akt/mTOR pathway A schematic representation of the PI3K/Akt/mTOR pathway. Growth factors (such as IL-2) bind to the receptor tyrosine kinases and activate PI3K. PI3K promotes the conversion of PIP2 to PIP3. PIP3 can activate PDK1, but PIP3 activity is also repressed by the phosphatase PTEN, which converts PIP3 to PIP2. Activated PDK1 phosphorylates Akt, thereby activating the protein. Activated Akt then inhibits TSC2, which is an inhibitor of mTOR, and mTOR becomes activated and drives proliferation.

58

Chapter 2: Investigating the Importance of the Alternative NF‑κB and Akt Pathways in HTLV-1-Induced Cellular Transformation

2.1 Abstract

HTLV-1 is a complex retrovirus that is the etiological agent of an aggressive malignancy of CD4+ T cells, known as ATL. By contrast, HTLV-2 is non-pathogenic in humans. Both viruses express a Tax protein that has transforming capabilities in both in vitro and in vivo models. HTLV-1 Tax (Tax-1) is more capable of inducing transformation then HTLV-2 Tax (Tax-2). Tax-1 activates both the classical and alternative NF-κB pathways, while Tax-2 is only capable of activating the classical NF-κB pathway.

Activation of the alternative NF-κB signaling pathway by Tax-1 was hypothesized to contribute to HTLV-1 pathogenesis; this property was genetically mapped to two domains present in Tax-1 but absent in Tax-2. We analyzed the importance of alternative

NF-κB activation in HTLV-1-mediated transformation of primary human T-lymphocytes.

During the course of this study, it was discovered that a Tax-1 protein with a four-amino acid deletion of the PDZ binding motif (PBM) activated the alternative NF-κB, contrary to previously published results. Further analysis also revealed that the Tax-1 PBM domain was important for activation of the Akt signaling cascade. Tax-1 activates Akt by inhibiting PTEN localization via a competitive binding event between Tax-1 and PTEN for interaction with the tumor suppressor, DLG-1. The second Tax-1 mutation (Tax-1

225-232) was found to be deficient for alternative NF-κB activation. HTLV-1 featuring the Tax-1 225-232 mutation transformed primary cells in vitro with an efficiency similar 59 to wild type virus. This study demonstrated that Tax-1 activated the Akt pathway via the

PBM domain while the 225-232 region of Tax-1 was required for alternative NF-κB activation. Tax-1 activation of Akt and alternative NF-κB pathways may contribute to the different pathogenic outcomes of HTLV-1 and HTLV-2.

60 2.2 Introduction

HTLV-1 is a Deltaretrovirus that was discovered over 30 years ago23,24.

Approximately 15-20 million people world-wide are infected with HTLV-1, with endemic areas in Japan, the Caribbean Islands, Central America, South America, and

Africa. A small subset of HTLV-1-infected individuals develop disease, including ATL and HAM/TSP23,24,37. ATL is a lethal disease, with the average survival time after diagnosis of less than 1 year58. ATL usually develops in individuals infected with HTLV-

1 during their childhood after a clinical latency period of 30-40 years41. Several studies have focused on how ATL develops, but the exact mechanism of oncogenesis is still unclear. HTLV-2, a closely related virus, shares 70% nucleotide sequence identity with

HTLV-1, and was initially discovered in a patient with hairy-cell leukemia28. However, further studies demonstrated that HTLV-2 is not associated with any diseases in humans29. The differences that exist between these two viruses that allows distinct pathologies are under further investigation.

HTLV-1 Tax-1 is a 40-kDa protein encoded by the pX region of the provirus that transactivates viral gene expression via recruitment of CREB to the viral promoter373,178.

Tax-1 is required for in vitro transformation of primary T cells, and its expression in transgenic mice leads to leukemia development367. Tax-1 promotes oncogenesis by deregulating the cell cycle via interactions with several cell signaling cascades, such as

NF-κB and Akt373.

The NF-κB signaling cascade is involved in regulation of cell proliferation and survival224. Two discrete NF-κB signaling pathways exist: the classical and alternative pathways225. The classical pathway is activated in response to stimulation via the

61 IKKα/IKKβ/NEMO complex. The activated IKK complex promotes the phosphorylation and eventual degradation of the IκBα protein, which releases the NF-κB transcription factor dimer, p50/RelA, to the nucleus to drive transcription225. The alternative NF-κB pathway is activated when NIK phosphorylates IKKα homodimer. The IKKα homodimer then promotes the proteolytic processing of p100 into p52, allowing translocation of the p52/RelB transcription factor dimer into the nucleus to drive transcription374. The classical NF-κB pathway is rapidly activated within minutes of stimulation, while the alternative NF-κB pathway can take up to 3 hours to reach full potential after stimulation225. The classical NF-κB response is also transient, in that it will be quickly repressed after loss of stimulation, while the alternative NF-κB pathway remains active after loss of stimulation. In many cancers, including ATL, the NF-κB pathways are deregulated225,231,375.

The Akt pathway is stimulated in T cells by IL-2 binding to a receptor on the cell surface242. This initial stimulation leads to the activation of PI3K, which converts the phosphoinositide PIP2 to PIP3243. PIP3 then promotes the phosphorylation of Akt via

PDK1. Activated Akt then phosphorylates and inhibits TSC2, an inhibitor of mTOR, which ultimately induces cellular proliferation244-246. The cellular genes PTEN and

PHLPP regulate Akt activity both before and after its activation. PTEN inhibits Akt pre- activation by converting PIP3 back to PIP2, while PHLPP inhibits Akt post-activation via its ability to dephosphorylate Akt242,247. In many cancers, including ATL, the Akt signaling cascade is constitutively activated243,248,376.

Tax-1 is involved in the activation of both NF-κB and Akt signaling236,239,377,378.

Interestingly, HTLV-2 Tax-2 (the Tax-1 homolog) is incapable of activating alternative 62 NF-κB signaling239,291. Two distinct domains present on Tax-1, which are absent on Tax-

2, have been demonstrated to be important for activation of alternative NF-κB signaling; the PDZ binding motif (PBM) and a leucine zipper-like region (LZ)239,286. Tax-1 with a mutation of the LZ (Tax-1 225-232) or deletion of the PBM (Tax-1 ΔPBM) is less capable of promoting IL-2 independence of the CTLL-2 cell line, indicating that these mutant have a lower transforming potential (Figure 2-1 illustrates the Tax-1 mutants)239,286. HTLV-1 virus that expresses a Tax protein lacking the PBM transform cells in a delayed manner compared to wild type virus207. We proposed that Tax-1 activation of the alternative NF-κB pathway is crucial for HTLV-1 oncogenesis, and investigated its role in HTLV-1-mediated cellular transformation. We also demonstrated that the Tax-1 ΔPBM mutant was not deficient for alternative NF-κB activation, but was incapable of activating the Akt signaling cascade.

2.3 Materials and Methods

2.3.1 Plasmids

The ACHneo (wtHTLV-1) and pH6neo (wtHTLV-2) molecular clones were generated previously379-381. The HTLV-1 Tax-1 mutant, ΔPBM, was also generated previously207. The HTLV-1 Tax-1 mutant 225-232 was generated by subcloning of the

Tax-1 region of wtHTLV-1 into a shuttle vector. Six rounds of site-directed mutagenesis were then performed to introduce nucleotide changes in Tax-1 (provirus numbering

7,969-GTCACGCTAACAGCCTGGCAAAAC-7,992 to 7,969-

TGTATCCAGACTGCCTGGTGTACA-7,992) which resulted in substitutions of amino acids 225-232 in Tax-1 for their Tax-2 counter parts (225-VTLTAWQN-232 to 225- 63 CIQTAWCT-232). The mutated Tax-1 225-232 region was then returned to the full length

HTLV-1 plasmid. The S-tag Tax-1, S-tag Tax-2, S-tag Tax-1ΔPBM, and S-tag Tax-1 225-

232 were generated via PCR amplification of the Tax gene from the respective molecular clone. PCR products were generated with NheI and XmaI restriction sites flanking the

Tax gene. The Tax open reading frame was then inserted into the pTriEx-4 Neo vector

(Novagen, Madison, WI) in frame with the N-terminal 6xHis-S-Tag. The empty pTriEx-4

Neo vector was used as the negative control. The HA-tagged PTEN expression vector

(800 pSG5L HA-PTEN WT) and MyrPTEN expression vector (1006 pSG5L HA-

MyrPTEN), were obtained from Addgene (Addgene, Cambridge, MA; plasmids no.

10750 and 10776, respectively)382,383. pEGFP SAP97 (GFP-DLG1), a gift from J. Miner,

Washington University, was originally made by W. Green, University of Chicago384.

2.3.2 Cell Culture

All cells were maintained in a humidified incubator at 37°C and 5% CO2.

HEK293T cells were maintained in Dulbecco's modified eagle medium (DMEM)

(Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS)

(Invitrogen), 2 mM glutamine, penicillin (100 U/mL) (Invitrogen), and streptomycin (100

μg/mL) (Invitrogen). The 729.b control and all 729.b HTLV producer cell lines were maintained in Iscove’s modified DMEM supplemented with 10% FBS, 2 mM glutamine, penicillin (100 U/mL), and streptomycin (100 μg/mL). Jurkat cells were maintained in

RPMI 1640 medium (Invitrogen) supplemented with 10% FBS, 2 mM glutamine, penicillin (100 U/mL), and streptomycin (100 μg/mL). PBL cell lines were maintained in

RPMI 1640 supplemented with 20% FBS, 2 mM glutamine, penicillin (100 U/mL), 64 streptomycin (100 μg/mL), and human interleukin-2 (hIL-2)(10 U/mL)(Roche,

Indianapolis, IN).

2.3.3 Immunoblotting

Cells were lysed in NP-40 lysis buffer (150 mM NaCl, 50 mM Tris-Cl pH 8.0, and

1% NP-40) supplemented with Complete Mini EDTA-free Protease Inhibitor (Roche).

Lysates were quantified using the Pierce BCA protein assay kit (Thermo Fisher, Waltham,

MA) and absorbance was measured using a FilterMax F5 Microplate Reader (Molecular

Devices, Sunnyvale, CA). Lysates were prepared for equal amounts of total protein, separated on Mini-PROTEAN TGX Precast 4%–20% SDS-PAGE gels (BioRad,

Hercules, CA), and transferred to nitrocellulose membranes (GE, Boston, MA).

Membranes were blocked in 5% milk in PBS supplemented with 0.1% Tween-20

(PBST). Blots were incubated with indicated primary antibody overnight, and after washing, were then incubated with secondary antibodies for one hour. Primary antibodies used: anti-S-tag (Abcam, Cambridge, UK; 1:1,000), anti-p100/p52 (EMD Millipore,

Billerica, MA; 1:1,000), anti-Actin (Abcam; 1:10,000), anti-Akt (Cell Signaling

Technologies, Danvers, MA; 1:1,000), anti-pAktT308 (Cell Signaling Technologies;

1:1,000), anti-pAkt S473 (Cell Signaling Technologies; 1:1,000), and anti-PTEN (Cell

Signaling Technologies; 1:1,000). Secondary antibodies used: goat-anti-rabbit HRP

(Santa Cruz Biotechnology, Dallas, TX; 1:5,000) and goat-anti-mouse HRP (Santa Cruz

Biotechnology; 1:5,000). Membranes were developed using Immunocruz Luminol

Reagent (Santa Cruz Biotechnology) and imaged using an Amersham Imager 600 (GE).

Densitometry was preformed via Multi Gauge software (Fuji Film, Valhalla, NY).

65

2.3.4 Reporter Gene Assays and p19 ELISA

Tax function was measured using an LTR reporter assay. Briefly, HEK293T cells were transfected using Lipofectamine (Invitrogen), according to the manufacturer’s instructions, either with the indicated Tax (400 ng of plasmid), indicated HTLV molecular proviral clone (1000 ng of plasmid), or an empty vector (either 400 ng or 1000 ng of plasmid). Transfections also included TK-renilla (RTK) (10 ng of plasmid) and LTR-1-

Luc (100 ng of plasmid). Post-transfection (24 hours for Tax expression vectors and 48 hours for molecular clones) cells were collected and lysed in passive lysis buffer

(Promega, Fitchburg, WI). Lysates were then subjected to the Dual-Glo Luciferase Assay

(Promega) according to manufacturer’s instructions, and read on a FilterMax F5

Microplate Reader (Molecular Devices). All experimental samples were performed in duplicate. Supernatants from cells transfected with molecular clones were also collected for ELISA of HTLV-1 or HTLV-2 p19 (Zeptometrix, Buffalo, NY), which was performed per the manufacturer’s instructions.

NF-κB activity was measured using a κB-Luciferase reporter plasmid. HEK293T cells were transfected with a RelA expression vector (50 ng of plasmid), RTK (10 ng of plasmid), the κB-Luc plasmid (100 ng of plasmid), and the indicated Tax expression vectors. Twenty-four hours post transfection, cells were collected and subjected to the

Dual-Glo Luciferase Assay as described above.

66 2.3.5 Producer Cell Line Generation and Verification

729.b cells were used to generate producer cell lines for the HTLV molecular clones. The 729.wtHTLV-1 (ACHneo), 729.wtHTLV-2 (pH6neo), and 729.HTLV-1 Tax-

1 ΔPBM were generated as described previously207,340. The 729.HTLV-1 Tax-1 225-232 cell line was generated as follows: 729.b cells were nucleofected (Amaxa, Cologne,

Germany) with the HTLV-1 Tax-1 225-232 molecular clone per the manufacturer’s instructions (Kit V, program X-001). Forty-eight hours post-nucleofection, cells with stable plasmid incorporation were selected with G418 (Life Technologies, Carlsbad, CA).

Cell lines that grew out were tested for p19 Gag production by ELISA. Cells producing p19 were then subjected to single cell clone selection by limiting dilution. The wells with cellular outgrowth were subjected to p19 ELISA to identify cell lines producing levels of p19 comparable to other lines to be used in this study. Genomic DNA was harvested from these lines using the DNeasy kit (Qiagen, Valencia, CA) and PCR was performed using primers for the HTLV-1 gag sequence or primers that would amplify only in the presence of the 225-232 mutation. Cells containing the mutant provirus that produced virions were then used for the cellular transformation assay.

2.3.6 Cellular Transformation Assay

Human peripheral blood mononuclear cells (PBMCs) were isolated from whole blood using Ficoll (GE) per the manufacturer’s instructions. PBMCs (2x106) were co- cultivated with 1x106 lethally irradiated 729.b, 729.wtHTLV-1, or 729.HTLV-1 Tax-1

225-232 cells in 24-well plates. The co-cultures were maintained in RPMI 1640 medium

67 supplemented with 20% FBS, 2 mM glutamine, penicillin (100 U/mL), streptomycin (100

μg/mL), and hIL-2 (10 U/mL). Cells were counted weekly by trypan-blue exclusion to determine the viable number of cells. Medium was exchanged three times per week; two of the exchanges were done in the absence of supplemented hIL-2 while the third exchange was supplemented with hIL-2 (10 U/mL). Wells that showed cell growth were expanded and maintained as an immortalized/transformed PBL cell line.

2.3.7 PBL line verification

PBL lines generated in the cellular transformation assay were assayed for p19 production using the p19 ELISA described earlier. Cell lines were then phenotyped for cell surface markers by flow cytometry. Cells were stained with anti-CD3-FITC and either anti-CD4-PE or anti-CD8-PE along with FITC and PE isotope controls (BD

Biosciences, San Diego, CA). All antibodies were used at 1:40 dilution, and staining was performed at 4°C for 30 minutes in PBS supplemented with 2% FBS (PBS.2). Cells were washed twice and resuspended in PBS.2. Cell staining was then measured on a Guava

EasyCyte flow cytometer (EMD Millipore). Data were analyzed and plotted using

FlowJo software (FlowJo, Ashland, OR).

Viral DNA and RNA from PBL lines were analyzed for the presence of the 225-232 mutation. DNA was tested as described in 2.3.5. For RNA, cells were collected and RNA harvested using RNeasy kit (Qiagen). RNA samples were then subjected to DNase treatment (Roche) and cDNA was generated using OmniScript (Qiagen) per the manufacturer’s instructions. cDNA was analyzed for the presence of the mutation using the specific primers.

68

2.3.8 Co-immunoprecipitation

HEK293T cells were lysed in buffer consisting of PBS with 10% glycerol, 1%

Igepal CA 630, 0.1% sodium deoxycholate, 10 mM sodium glycerophosphate, 10 mM sodium fluoride, and 2.5 mM sodium pyrophosphate with protease inhibitor cocktail.

Immunoprecipitation was performed with goat anti-GFP antibody (Genetex, Irvine, CA) using protein A/G-conjugated cross-linked agarose beads. Following overnight incubation of lysates with immunoglobulin and beads, the beads were washed with PBS with 0.1% sodium deoxycholate five times, and then protein was eluted by incubation at

70°C for 10 minutes in 1 x SDS sample buffer.

2.3.9 Statistical Analysis

Statistical analyses were performed using the unpaired Student's t test in GraphPad

Prism 6 (GraphPad Software, La Jolla, CA) as indicated. Statistical significance was defined as p<0.05.

2.4 Results

2.4.1 Tax-1 225-232 was deficient for alternative NF-κB activation and Tax-1 ΔPBM was not

Previous studies have demonstrated that the Tax-2, Tax-1 ΔPBM, and Tax-1 225-

232 were deficient for alternative NF-κB activation239,286. We set out to confirm mutant

Tax-1 proteins functioned as expected prior to further experiments. To determine if the

69 Tax constructs could activate transcription of the HTLV-1 LTR, we performed a LTR-Luc reporter assay in HEK293T cells. All Tax-1 constructs were shown to drive transcription of the HTLV LTR; however, Tax-2 transactivation activity was slightly lower than the other Tax-1 constructs, though this was not statistically significant (p=0.1754) (Figure 2-2

A). Lower Tax-2 activity was expected due to the lower transcriptional activity of Tax-2 and the use of the LTR from HTLV-1 for this assay342. We next wanted to determine the ability of these Tax constructs to activate the canonical NF-κB pathway using the κB-Luc reporter assay in HEK293T cells. The RelA positive control confirmed the function of our assay, while all Tax constructs were confirmed to promote activation of NF-κB transcription, in accordance with previous reports (Figure 2-2 B).

We next determined the ability of these Tax constructs to activate the alternative

NF-κB pathway. HEK293T cells were transfected with Tax expression constructs and cell lysates were analyzed by immunoblotting. Probing with the S-tag antibody demonstrated relatively equal expression of the Tax constructs (Figure 2-2 C). Probing with an antibody that recognized the NF-κB protein p100 was used to gauge activation of the alternative

NF-κB pathway. The p100 antibody detects both the full length p100 protein and the cleaved active variant p52. The ratio of p52:p100 was used to demonstrate activation of the pathway. The ratio of p52:p100 for the wild type Tax-1 transfected cells was 2.4 fold higher compared to the empty S-tag negative control, which demonstrated that Tax-1 activated the alternative NF-κB as expected (Figure 2-2 C). The p52:p100 ratio for Tax-2 and Tax-1 225-232 were similar to the negative control (1.1 and 0.9-fold respectively), which demonstrated that these Tax constructs could not activate the alternative NF-κB signaling pathway, as reported previously (Figure 2-2 C). Interestingly, the p52:p100 ratio

70 in cells that expressed Tax-1 ΔPBM was similar to cells that expressed the wild type Tax-

1 (2.4-fold higher than negative), which demonstrated the ability of this construct to promote alternative NF-κB signaling; this result contradicts previous reports (Figure 2-2

C). We further investigated the ability of Tax-1 ΔPBM to activate alternative NF-κB by performing titrations of transfected plasmid (1x, 2x, and 4x). This experiment revealed that Tax-1 ΔPBM was more efficient in activating the alternative NF-κB pathway compared to wild type Tax-1, as seen by higher p52:p100 ratios (Figure 2-2 D).

Collectively, these results demonstrated that the Tax-1 ΔPBM construct efficiently activated the alternative NF-κB signaling cascade.

2.4.2 Tax-1 Activated Akt by Inhibiting PTEN Activity

It was shown that the Tax-1 ΔPBM mutant had a reduced ability to transform primary T cells239. The inability of this mutant to activate the alternative NF-κB pathway was thought to contribute to the attenuated transformation. We demonstrated that the Tax-

1 ΔPBM construct was capable of activating the alternative NF-κB pathway, and activated this pathway at higher levels than wild type Tax-1 (Figure 2-2 D). We sought an alternative explanation for the Tax-1 ΔPBM transformation phenotype. Previous studies demonstrated that Tax-1 ΔPBM-transformed cell lines expressed low amounts of the tumor suppressor DLG-1, and that Tax-1 interacted with DLG-1 via the PBM domain290,385. DLG-1 is important for the membrane localization of PTEN, which allows

PTEN to function as an upstream repressor of the Akt/mTOR pathway242. We hypothesized that Tax-1 may activate the Akt/mTOR pathway via its interaction with

DLG-1. To determine if Tax-1 activates this pathway, the physiologically relevant Jurkat

71 cell line was transfected with Tax expression vectors. These samples were then analyzed by immunoblotting for activation of the Akt signaling pathway. Similar levels of Tax expression was confirmed using the S-tag antibody (Figure 2-3 A). Antibodies for total levels of Akt and two phosphorylation events of Akt, T308 and S473, were also used, and increased Akt phosphorylation is considered indicative of Akt activation. In the Tax-1 transfected cells, T308 phosphorylation was 1.93-fold higher than the negative control, which demonstrated that Tax-1 promotes Akt activation (Figure 2-3 A). In contrast, Tax-2 and Tax-1 ΔPBM did not increase phosphorylation at T308, with fold changes of 1 and

0.7, respectively, compared to the negative control (Figure 2-3 A). This data demonstrated that Tax-1 activation of Akt was dependent on the PBM domain of the protein. No change was seen in phosphorylation of Akt at S473, which demonstrated that

Tax-1 activation of Akt was specific to phosphorylation at T308.

To determine the role that PTEN may play in Tax-1-mediated activation of Akt,

Jurkat cells were co-transfected with a PTEN expression vector and either S-tag Tax-1 or the empty S-tag vector. Immunoblotting revealed that Tax-1 induced phosphorylation of

Akt at T308 in the presence of PTEN over-expression, with the Tax-1 sample showing

2.6-fold higher levels of phosphorylation compared to the empty vector control (Figure 2-

3 B). This experiment was repeated utilizing a PTEN expression vector with an N- terminal myristoylation acceptor motif, which has been shown to result in constitutive membrane expression of PTEN376. When these samples were analyzed, Tax-1 failed to induced T308 phosphorylation (1.1-fold above the negative control) (Figure 2-3 C).

Collectively, these results suggested that the activation of Akt by Tax-1 was dependent on

PTEN not reaching the membrane.

72 DLG-1 promotes the localization of PTEN to the plasma membrane, and Tax-1 interacts with DLG-1. Therefore, we propose that Tax-1 may inhibit this specific function of DLG-1, resulting in activation of Akt signaling. To investigate if Tax-1 competes with

PTEN for interaction with DLG-1, we performed a co-immunoprecipitation assay in

HEK293T cells. Cells were transfected with GFP-tagged DLG-1 and PTEN expression vectors along-side either S-tag Tax-1 or S-tag Tax-1 ΔPBM. Post-transfection, DLG-1 was immunoprecipitated using a GFP antibody and immunoblotting was performed.

Effective IPs of DLG-1 were demonstrated using a GFP antibody, while the S-tag antibody demonstrated that DLG-1 interacted with wild type Tax-1 but not Tax-1 ΔPBM

(Figure 2-3 D). When PTEN levels were analyzed, a decrease in PTEN complexed with

DLG-1 was observed in cells expressing wild type Tax-1 compared to those not expressing Tax-1 (Figure 2-3 D). Cells expressing Tax-1 ΔPBM had similar levels of

PTEN complexed with DLG-1 as the negative control (Figure 2-3 D). These results indicated that Tax-1 was involved in a competitive binding event with PTEN for interaction with DLG-1, and illustrated a potential mechanism for Tax-1 activation of

Akt.

2.4.3 Tax 225-232 Mutation Did Not Affect Transformation

The Tax-1 225-232 mutation has been shown to decrease the ability of Tax-1 to transform T cells239,291. However, studies have not determined the effect this mutation in the context of the intact virus. To investigate the effect of Tax-1 225-232 on HTLV-1- mediated cellular transformation, we generated an HTLV-1 Tax-1 225-232 mutant virus.

The HTLV-1 Tax-1 225-232 molecular clone and wild type HTLV-1 were tested for Tax

73 transactivation by LTR-Luc reporter assay. The two HTLV-1 molecular clones showed similar levels of LTR-Luc activity (no significant difference as measured by unpaired

Student’s t test), which demonstrated that the Tax-1 225-232 mutation did not alter transcription by Tax-1 in the context of the molecular clones (Figure 2-4 A). Supernatants were also collected from these transfected cells and subjected to p19 ELISA, which has been used as a surrogate for virion production. The two HTLV-1 molecular clones generated similar levels of p19 (no significant difference as measured by unpaired

Student’s t test), consistent with the Tax transactivation data (Figure 2-4 B). These results indicated that the Tax-1 225-232 mutation did not alter the virus transcription or virion production.

A caveat of HTLV research is the inefficiency of cell-free virus infection, with cell- to-cell contact being required for efficient transmission of the virus. To achieve this,

729.b cell lines that stably produce virus are generated for use in infection studies. 729.b cells were nucleofected with the HTLV-1 and HTLV-1 Tax-1 225-232 molecular clones and stable cell lines were generated. Cell clones generated from single cell isolations that produced similar levels of p19 were selected and screened for the Tax mutations in their genomic DNA (Figure 2-4 C and D). These producer cell lines were then used in cellular transformation assays. Freshly isolated human PBMCs were co-cultivated with lethally irradiated producer cell lines generating either wild type HTLV-1, HTLV-1 Tax 225-232, or no virus (729.b parental as a negative control). Viable cell numbers were counted every week over the course of 12 weeks (Figure 2-5). The negative control samples had no viable cells after week 8 of co-culture, while co-cultures containing both the HTLV-1 and HTLV-1 Tax 225-232 had viable cells throughout the assay with outgrowths

74 beginning in week 10. The wtHTLV-1 and HTLV-1 Tax 225-232 results were analyzed by unpaired Student’s t test analysis, and no significant difference was observed at weeks

10, 11, or 12. This result demonstrated that the Tax-1 225-232 mutation did not affect the ability of Tax-1 to transform cells in vitro.

2.4.4 PBL 225-232 generation and characterization

The cell lines that grew out of the in vitro cellular transformation assay were cultured and characterized. All established growing lines were shown to produce p19, which demonstrated that they were infected with HTLV (data for one line each shown)(Figure 2-6 A). Furthermore, the cells were phenotyped to ensure they were CD3+

CD4+ CD8- T cells as would be expected of an HTLV-1-transformed cell line. Two independent lines for both wild type HTLV-1 and HTLV-1 Tax-1 225-232 were found to be CD3+, CD4+, and CD8- (data for one line each shown)(Figure 2-6 B). To ensure that no reversion occurred during the infection and transformation process, the cell lines were tested for the presence of the mutation in both genomic DNA and RNA using the 225-

232-specific primers. The mutation was found to be intact in both cases (Figure 2-6 C and

D).

2.5 Discussion

HTLV-1 is an oncogenic virus shown to cause ATL in 5-10% of those infected58.

The closely related HTLV-2 virus is not associated with any human malignancies29.

Studies have shown that the Tax-1 protein has transforming potential, and plays a pivotal role in disease development of HTLV-1222. Comparisons of Tax-1 and Tax-2 have shown 75 that Tax-1 is significantly more efficient in transformation induction than Tax-

2239,284,286,291,386. Several past studies have sought to elucidate differences between Tax-1 and Tax-2 that result in these different transforming capabilities; eventually, the second

LZ and the PBM of Tax-1 were identified as important for efficient transformation286,291.

Interestingly, Tax-2 lacks both these domains. It was shown that mutations of the second

LZ (225-232) or deletion of the PBM (ΔPBM) resulted in the inability of Tax-1 to activate the alternative NF-κB pathway, which provides a potential explanation for the differences in transformation between HTLV-1 and HTLV-2239. We demonstrated here that the 225-232 mutation was deficient for alternative NF-κB activation, while the

ΔPBM mutation does activate alternative NF-κB signaling; contrary to previous reports239.

One possible explanation for how the ΔPBM mutation of Tax-1 causes a decrease in transformation potential, in light of its ability to activate alternative NF-κB, is its involvement in the PI3K/Akt/mTOR signaling cascade248,252. We demonstrated that Tax-1 over-expression in Jurkat cells results in an increase of Akt phosphorylation at T308, a marker for activated Akt. Deletion of the PBM from Tax-1 prevented the induction of Akt phosphorylation by Tax-1, demonstrating that the PBM region was important for Tax-1

Akt activation. To elucidate a mechanism for how Tax-1 activates this signaling cascade we investigated the Tax-1 interacting partner DLG-1, which interacts with Tax-1 through the PBM289. Previous studies showed that CTLL-2 cells transformed with Tax-1 ΔPBM demonstrated a selective pressure for low DLG-1 expression290. DLG-1 is important for

Akt regulation via promotion of plasma membrane localization of PTEN, a key inhibitor of Akt activation246. We showed that over-expression of wild type PTEN did not alter

76 Tax-1 activation of Akt, but a membrane-localized mutant of PTEN prevented Akt activation. Furthermore, we demonstrated a potential mechanism of Tax-1 activation of

Akt via competitive binding between PTEN and Tax-1 for interaction with DLG-1.

Immunoprecipitation assays revealed that Tax-1 expression reduced the interaction between PTEN and DLG-1, which would result in less PTEN at the plasma membrane and in turn, higher Akt activity. As Akt signaling is constitutively activated in ATL cell lines, this pathway should be thoroughly examined to understand the exact role it plays in

HTLV-1-mediated oncogenesis.

While the ΔPBM mutation activated the alternative NF-κB pathway, the 225-232 mutation did not. Activation of the alternative NF-κB pathway is important for Tax-1 transformation capabilities; however, studies have not investigated the role it plays in cellular transformation in the context of intact virus239. We generated molecular clones of

HTLV-1 featuring the Tax-1 225-232 mutation, and demonstrated that these constructs generated p19 Gag and transactivated the viral promoter. These results indicate that the

225-232 mutation did not alter these functions of the virus. These experiments are a qualitative assay to demonstrate that these functions are present, and the variance observed in these results was not significant. We next demonstrated that the Tax-1 225-

232 mutant virus could immortalize human PBMCs as well as wild type HTLV-1 in vitro.

Furthermore, we investigated the immortalized cell lines generated by infection with these viruses, and demonstrated that no reversions of the mutation had occurred, as had been seen in several other studies346. We also showed that these cell lines produce p19

Gag, which indicates that immortalization was associated with a successful HTLV-1 infection. The p19 production variance observed between wtHTLV-1 and HTLV-1 Tax-1

77 225-232 transformed cells cannot be directly attributed to the Tax-1 225-232 mutation. It is likely that the proviral genomes integrated into regions with varying transcriptional activity, which results in these differences. By comparing six individual cellular clones per condition we found no significant difference between wtHTLV-1 and HTLV-1 Tax-1

225-232 p19 Gag production (p=0.216). Collectively these results demonstrated that activation of the alternative NF-κB pathway by Tax-1 was not required for immortalization in vitro. Further studies should analyze the role of this pathway in an in vivo model. The humanized mouse model is maturing into a robust system for analyzing

HTLV-1 disease in vivo387. Future work will utilize the immortalized lines generated in our study with humanized mice to determine the role that alternative NF-κB pathway plays in HTLV-1 disease development.

78

Figure 2.1 Tax-1, Tax-2, and Tax-1 Mutant Proteins

Schematic diagram of the Tax-1, Tax-2, and Tax-1 mutants generated for this study.

Abbreviations: NLS: Nuclear localization signal LZR: Leucine zipper like region PBM: PDZ binding motif.

79

Figure 2.2 Characterization of Tax-1 Mutants A) A LTR-luciferase reporter assay was performed to determine the ability of the Tax constructs to drive transcription of the viral promoter. HEK293T cells were transfected with the indicated Tax construct, LTR-luc, and RTK for normalization. 24-hours post- transfection cells were collected, lysed, and analyzed by Dual Glo assay. All samples Continued

80 Figure 2.2 Continued were analyzed in duplicate and data shown is representative of three independent experiments. B) A κB–luciferase reporter assay was performed to determine the ability of the Tax constructs to activate classical NF-κB signaling. HEK293T cells were transfected with indicated the Tax construct, κB-luc, and RTK for normalization. 24-hours post- transfection cells were collected, lysed, and analyzed by Dual Glo assay. All samples were analyzed in duplicate and data shown is representative of three independent experiments. C) Immunoblotting was performed to determine the ability of the Tax constructs to activate alternative NF-κB signaling. HEK293T cells were transfected with the indicated Tax construct. 24-hours post-transfection cells were collected, lysed, and subjected to immunoblotting for S-tag (Tax), Actin, and p100/p52 (alternative NF-κB). The histogram shows the ratio of p52 to p100. Higher values represent higher activation of the alternative NF-κB pathway. Data shown is representative of two independent experiments. D) Titration of the Tax-1 ΔPBM mutant was performed to determine its ability to activate the alternative NF-κB pathway. HEK293T cells were transfected with the indicated amounts of Tax-1 or Tax-1 ΔPBM. 24-hours post-transfection cells were collected, lysed, and subjected to immunoblotting for S-tag (Tax), Actin, and p100/p52 (alternative NF-κB). The histogram shows the ratio of p52 to p100. Higher values represent higher activation of the alternative NF-κB pathway. Data shown is representative of two independent experiments. For all panels an asterisk indicates a significant increase over the negative control sample (p<0.05).

81

Figure 2.3 Tax-1 activated Akt, while Tax-1 ΔPBM did not

A) HEK293T cells were transfected with empty vector or expression plasmids for Tax-1, Tax-1 ΔPBM, or Tax-2. 24-hours post-transfection cells were collected, lysed, and subjected to immunoblotting for S-tag (Tax), Actin, total Akt, P-Akt-T308, and P-Akt- S473. The ratios of P-Akt-T308 and P-Akt-S473 to total Akt are shown as measured by densitometry. Data is representative of two independent experiments. B) Jurkat cells were co-transfected with a plasmid expressing wild type PTEN and either empty vector or Tax-1 expression plasmid. Levels of P-Akt-T308, P-Akt-S473, and total Akt were determined by immunoblot, and ratios were determined as described previously. Levels of PTEN and actin were also monitored by immunoblot. Data is representative of two independent experiments. C) Jurkat cells were transfected with an expression vector for myristoylated PTEN for constitutive membrane localization, together with either empty vector or Tax-1 expression vector and analyzed as in B. Data is representative of two independent experiments. D) HEK293T cells were co-transfected with expression plasmids for GFP-DLG-1, PTEN, and either Tax-1 or Tax-1 ΔPBM mutant. Cell Continued

82

Figure 2.3 Continued lysates were immunoprecipitated with control goat IgG or goat anti-GFP. Cell lysates and immunoprecipitates were examined by immunoblot performed with antibodies to PTEN, Tax (STag), and DLG-1 (GFP). Data is representative of three independent experiments.

83

Figure 2.4 Generation and Characterization of HTLV-1 Tax-1 225-232 Producing Cell Line A) HEK293T cells were transfected with the indicated HTLV-1 molecular clone, LTR- luc, and RTK for normalization. 48-hours post-transfection, cells were collected, lysed, and analyzed for luciferase levels. Data shown is representative of two independent experiments. B) Supernatants were collected from the transfection samples in A and were analyzed for p19 concentration by ELISA. Data shown is representative of two independent experiments. C) 729.b, 729.wtHTLV-1, and 729.HTLV-1 Tax-1 225-232 cells (106) were plated in 1 mL of medium. At 48 hours, supernatants were collected and analyzed for p19 production. Data shown is representative of three independent experiments. D) Genomic DNA from the indicated 729 lines was analyzed for the presence of either the 225-232 mutation or HTLV-1 gag by PCR. HTLV-1 and HTLV-1 Tax-1 225-232 were used as positive controls for PCR. For all panels an asterisk indicates a significant increase over the negative control sample (p<0.05).

84 2.0×107 729.B (Negative)

r

e 729.wtHTLV-1 b 1.5×107

m 729.HTLV-1 Tax-1 225-232

u

N

l l 7

e 1.0×10

C

e

l

b 6 a 5.0×10

i

V

0.0 0 2 4 6 8 10 12 Weeks

Figure 2.5 HTLV-1 Tax-1 225-232 induced cellular transformation

Human PBMCs (2x106) were co-cultivated with 1x106 lethally irradiated 729.b, 729.wtHTLV-1, or 729.HTLV-1 Tax-1 225-232 cells in 24-well plates. Cells were counted weekly using trypan-blue exclusion to determine viable numbers of cell.

85

Figure 2.6 Characterization of Generated PBL Lines

A) Established PBL lines (1x106) were plated in 1 mL of medium. 48-hours later, supernatants were collected and subjected to p19 ELISA. Data shown is representative of two independent experiments. B) PBL lines were analyzed by flow cytometry for expression of CD3, CD4, and CD8. Similar results were seen in two independent experiments. C) Genomic DNA of PBL lines was tested for the 225-232 mutation. D) RNA from the PBL lines was tested for the 225-232 mutation by PCR. The wtHTLV-1 and HTLV-1 Tax-1 225-232 plasmids were used as positive controls for PCR in both C and D.

86

Chapter 3: Identification of the Role of a Novel Tax-1 Binding Partner, SNX27, in HTLV-1 Infection

3.1 Abstract

Human T cell leukemia virus type 1 (HTLV-1) and 2 (HTLV-2) are related viruses with distinct pathologies in humans. HTLV-1 causes a CD4+ T-cell malignancy, ATL, while HTLV-2 is not associated with any diseases in humans. HTLV-1 Tax (Tax-1) activates both the classical and alternative NF-κB signaling pathways, while HTLV-2 Tax

(Tax-2) only activates classical NF-κB pathway. We postulated that activation of alternative NF-κB by Tax-1 is important for HTLV-1 pathogenesis, and that Tax-1 interacts with an unknown protein to activate this pathway. To determine the identity of the unknown protein, we performed proteomics analysis of Tax constructs that were capable and incapable of activating the alternative NF-κB pathway. We found several candidate proteins through mass spectrometry, but upon verification of interaction, these proteins were excluded from further analysis. Proteomics data were analyzed for novel interactions not associated with alternative NF-κB activation. Sorting Nexin 27 (SNX27) was identified as an interacting partner of Tax-1. This study demonstrated that Tax-1 modulated the localization of the HTLV-1 receptor GLUT1 through SNX27.

Furthermore, knockdown of SNX27 in HTLV-1-producing cells reduced the amount of virus produced. This study revealed the first known mechanism of HTLV-1 receptor modulation post-infection.

87 3.2 Introduction

Although Human T cell Leukemia Virus (HTLV)-1 and HTLV-2 are related retroviruses, only infection with HTLV-1 leads to eventual development of disease.

Determining the mechanisms involved in these disparate outcomes is of great interest for understanding how HTLV-1 infection causes disease and permitting development of interventions.

It is well established that expression of HTLV-1 Tax-1 protein is necessary for the transcription of viral genes; Tax-1 recruits CREB and p300 to the viral promoter to drive transcription388,389. Tax-1 also contributes to the oncogenic potential of HTLV-1. Tax-1 expression in transgenic mice led to a leukemia/lymphoma like disease, while over- expression of Tax-1 in the CTLL-2 cell line promoted IL-2 independent growth239,286,367.

The ability of Tax-1 to transform cells depends, in part, on activation of the classical and alternative NF-κB pathways, which promote proliferation and cell survival206.

The alternative NF-κB pathway is activated when the cellular protein p100 is cleaved to its active p52 variant, which is prompted by the activation of NIK, which in turn phosphorylates and activates a IKKα homodimer complex224,374. This dimer then phosphorylates p100 resulting in the proteolytic truncation of the protein to p52374. Once cleaved, p52, along with RelB, translocates to the nucleus to promote transcription of target genes. Tax-1 promotes the activation of this pathway independently of NIK. The current theory of activation involves Tax-1 interacting both with IKKα and p100, and through close proximity activates the alternative NF-κB pathway231,237. However, it is predicted that other proteins are involved in alternative NF-κB activation by Tax-1.

88 The HTLV-1 homolog of Tax-1, Tax-2, is less efficient for transformation than

Tax-1286. Tax-2 cannot activate the alternative NF-κB pathway239,291. Previous studies demonstrated that two domains of Tax-1 not present in Tax-2, the PBM and the second

LZ, were required for activation of alternative NF-κB signaling239,291. In Chapter Two, we demonstrated that deletion of the PBM from Tax-1 did not prevent, but alternatively, increased activation of the alternative NF-κB pathway. In this Chapter, we proposed that

Tax-1 activation of alternative NF-κB signaling was dependent on its interaction with an unidentified protein. We performed a proteomic screen of Tax-1, Tax-2, Tax-1 ΔPBM, and Tax-1 225-232 to potentially identify this protein.

During the course of this study, we identified a novel Tax-1 interacting protein,

Sorting Nexin 27 (SNX27). The sorting nexin family of proteins is involved in the retrieval and recycling of specific cargos, which prevents degradation of these proteins in the lysosome390. SNX27 is a unique member of the sorting nexin family because it features a PDZ domain, which is required to bind to its specific cargos, such as

GLUT1391,392. Previous studies showed that SNX27 knockdown resulted in a drastic redistribution of GLUT1 from the plasma membrane to the lysosome, where it was degraded392.

GLUT1 facilitates the transport of glucose across the plasma membrane of cells.

GLUT1 also is important in HTLV-1 biology because it serves as one of the three receptor molecules for the virus, in addition to NRP-1 and HSPG147. A recent study demonstrated that over-expression of GLUT1 in cells producing virus decreased infectivity dramatically, while over-expression of GLUT3, a closely related protein, had no effect393. This study also demonstrated that reduction of GLUT1 levels in virus-

89 producing cells increased infectivity393. In other retroviruses, such as HIV-1, the virus has developed mechanisms to promote the removal of receptor molecules from the surface of the cell394. These functions aid the virus in two ways: reduce the possibility of superinfection, and promote viral budding and release (which is inhibited by the presence of the receptor molecule). HTLV-1 has no known mechanism for regulating the localization of its receptor molecules. Herein, we investigated whether the interaction between Tax-1 and SNX27 would allow HTLV-1 to regulate the localization of GLUT1, its primary entry receptor.

3.3 Materials and Methods

3.3.1 Plasmids

The S-tag Tax-1, S-tag Tax-2, S-tag Tax-1ΔPBM, and S-tag Tax-1 225-232 expression plasmids were generated by PCR amplification of the Tax gene from the respective molecular clones of HTLV-1 described in Chapter 2. PCR products were generated with NheI and XmaI restriction sites flanking the Tax gene. The Tax open reading frame was then inserted into the pTriEx-4 Neo vector (Novagen) in-frame with the N-terminal 6xHis-S-Tag. The SNX27, Myc-SNX27, and Myc-SNX27 ΔPDZ plasmids were a kind gift from Dr. M. Playford (National Heart, Blood, and Lung

Institute, National Institutes of Health, Bethesda, MD). The plasmid expressing the wtHTLV-1 molecular clone (ACHneo) was generated previously379.

90 3.3.2 Cell Culture

All cells were maintained in humidified incubators at 37 °C and 5% CO2.

HEK293T cells were maintained in DMEM (Invitrogen, Carlsbad, CA) supplemented with 10% FBS (Invitrogen), 2 mM glutamine (Invitrogen), penicillin (100 U/mL)

(Invitrogen), and streptomycin (100 μg/mL) (Invitrogen). SLB-1 cells were maintained in

Iscove’s modified DMEM (Thermo Fisher Scientific, Waltham, MA) supplemented with

10% FBS, 2 mM glutamine, penicillin (100 U/mL), and streptomycin (100 μg/mL).

3.3.3 Pull Down and Co-Immunoprecipitation

S-tag pull down for mass spectrometry and interaction verification was performed as follows. Two 10-cm dishes were seeded with 2.5x106 HEK293T cells, and 24 hours later, transfected with 10 µg of indicated expression plasmid using TransIT-2020 (Mirus

Bio, Madison, WI) per the manufacturer’s instructions. At 24-hours post-transfection, cells were collected in cold PBS and cell pellets from both plates were combined. Cells were lysed using passive lysis buffer (Promega, Madison, WI) supplemented with cOmplete mini EDTA-free protease inhibitor (Roche, Indianapolis, IN). Lysates were cleared of debris by centrifugation, and were incubated with 40 µL of S-agarose

(Novagen, Madison, WI) overnight. Beads were centrifuged and washed four times with

RIPA lysis buffer (150 mM NaCl, 0.01 M Sodium pyrophosphate, 10 mM EDTA, 10 mM sodium fluoride, 50 mM Tris, 0.1% SDS, 12.8 mM deoxycholic acid, 10% glycerol and

1% NP-40 (pH 8.0)). Beads were resuspended in 1x SDS loading dye and boiled for 10

91 minutes. Supernatants were then loaded on a gel and either subjected to mass spectrometry or immunoblotting.

For co-immunoprecipitation, a 10-cm dish was seeded with 2.5x106 HEK293T cells. At 24 hours, the cells were transfected with indicated plasmids using TransIT-2020

(Mirus) per the manufacturer’s instructions. At 24-hours post-transfection, cells were collected and lysed in NP-40 lysis buffer, and were incubated with primary antibody for the protein of interest overnight. Antibodies used: anti-SNX27 (Abcam, Cambridge, UK;

2 µg), anti-Myc (Abcam; 2 µg), and Tax-1 antisera (described previously)207. The next day, Dynabeads Protein G (Thermo Fisher Scientific) were added according to the manufacturer’s instructions and incubated for 2 hours with gentle rocking. Beads were washed four times in NP40 lysis buffer (150 mM NaCl, 50 mM Tris-Cl pH 8.0, and 1%

NP-40) using magnetic separation using a 12-tube magnet (Qiagen, Valencia, CA). Beads were resuspended in 1x SDS loading dye and boiled for 10 minutes. Supernatants were subjected to immunoblotting.

3.3.4 Mass Spectrometry

S-tag pull down samples were loaded and separated on Mini-PROTEAN TGX

Precast 4%–20% SDS-PAGE gels (BioRad, Hercules, CA). The gel was then stained using GelCode Blue Stain Reagent (Thermo Fisher Scientific) per manufacturer’s instructions. Whole lanes were excised from the gel and subjected to proteomics analysis as described previously395.

92 Capillary-liquid chromatography-tandem mass spectrometry (Capillary-

LC/MS/MS) for global protein identification was performed on a Thermo Finnigan LTQ

Orbitrap mass spectrometer equipped with a microspray source (Michrom Bioresources

Inc., Auburn, CA) operated in positive ion mode. Samples were separated on a capillary column (0.2x150 mm Magic C18AQ 3 μ 200A, Michrom Bioresources Inc.) using an

UltiMateTM 3000 HPLC system (LC-Packings/Dionex Co., Sunnyvale, CA). The scan sequence of the mass spectrometer was based on the data-dependent TopTen method. The resolution of a full scan was set at 30,000 to achieve a high mass accuracy MS determination.

The RAW data files collected on the mass spectrometer were converted to mzXML and MGF files by use of MassMatrix data conversion tools (version 1.3, www.massmatrix.net/download). The resulting MGF files were searched using Mascot

Daemon (Matrix Science version 2.2.2; Boston, MA) and the database was searched against the full SwissProt database version 57.5 (471472 sequences; 167326533 residues) or NCBI database version 20091013 (9873339 sequences; 3367482728 residues).

Considered modifications (variable) were methionine oxidation and the presence of carbamidomethyl cysteine. Three missed cleavages for the enzyme were permitted with a peptide tolerance of 1.2 Da, and the MS/MS ion tolerance was 0.8 Da. Search results were compiled and visualized using the Scaffold 4 software (Proteome Software,

Portland, OR). Unweighted spectrum count and percent coverage provided semiquantitative data. Protein identifications were assigned using PeptideProphet

(Institutes for Systems Biology, Seattle, WA). Proteins with 80% confidence were accepted with a minimum of one peptide displaying a 95% threshold confidence level.

93

3.3.5 Immunoblotting

For whole cell lysate analysis, cells were lysed in NP-40 lysis buffer supplemented with cOmplete mini EDTA-free protease inhibitor (Roche). Lysate concentrations were quantified using Pierce BCA protein assay kit (Thermo Fisher) and analyzed with a

FilterMax F5 Microplate Reader (Molecular Devices, Sunnyvale. CA). Equal amounts of protein were loaded and separated on Mini-PROTEAN TGX Precast 4%–20% SDS-

PAGE gels (BioRad) and transferred to nitrocellulose membranes (GE, Pittsburgh, PA).

For co-IP and pull down, samples were loaded and separated on Mini-PROTEAN TGX

Precast 4%–20% SDS-PAGE gels (BioRad) and transferred to nitrocellulose membranes

(GE). Membranes were blocked in 5% milk in PBST. Blots were incubated with indicated primary antibody overnight and with secondary antibodies for one hour.

Primary antibodies used: anti-S-tag (Abcam; 1:1,000), anti-p100/p52 (EMD Millipore,

Bellerica, MA; 1:1,000), anti-Actin (Abcam; 1:10,000), anti-Tax-1 antisera (1:1,000), anti-NKAP (Santa Cruz Biotechnology, Dallas, TX; 1:200), anti-PAK1IP1 (Bethyl

Laboratories, Montgomery, TX; 1:1,000), anti-ASCC1 (Bethyl Laboratories; 1:1,000;), anti-DLG-1 (Sigma-Aldrich, St. Louis, MO; 1:500), anti-Scrib (Santa Cruz

Biotechnology; 1:200), anti-PRMT5 (Abcam; 1:1,000), anti-STIP1 (Abcam; 1:10,000), anti-FUBP3 (Abcam; 1:1,000), anti-SNX27 (Abcam; 1:1,000), anti-Myc (Abcam;

1:1,000), anti-GLUT1 (Abcam; 1:250). Secondary antibodies used: goat-anti-rabbit HRP

(Santa Cruz Biotechnology; 1:5,000), goat-anti-mouse HRP (Santa Cruz Biotechnology;

1:5,000) and donkey-anti-goat HRP (Santa Cruz Biotechnology; 1:5,000). VeriBlot secondary antibodies were used for co-IPs to avoid IgG detection: anti-mouse IgG

94 VeriBlot for IP (Abcam; 1:1,000) and anti-rabbit IgG VeriBlot for IP (Abcam; 1:200).

Membranes were developed using Immunocruz Luminol Reagent (Santa Cruz

Biotechnology) and imaged using an Amersham Imager 600 (GE).

3.3.6 SNX27 Knockdown

TRC lentiviral shRNA vectors targeting SNX27 and negative control shRNAs

(pLKO.1) were purchased from Dharmacon (GE). Lentiviruses were generated as described previously396. Briefly, HEK293T cells were transfected with lentiviral shRNA, an HIV Gag/Pol, and a VSV-G expression vector. Transfections were performed using

Lipofectamine 2000 (Invitrogen) per the manufacturer’s instructions. At 72-hours post- transfection, supernatant was collected and filtered through a 0.45 µm filter. The filtered supernatant was concentrated by overlaying the supernatant onto a 1 mL 25% sucrose cushion, and centrifuging at 28,000 RPM for 1.5 hours in a Sorvall SW41 swinging bucket rotor. The supernatant was discarded and the virus pellet was resuspended in 200

µL of RPMI 1640 medium by gentle shaking overnight. Cells were transduced with lentiviruses as follows: 5x105 cells were collected and resuspended in 100 µL fresh medium with 50 µL of concentrated virus (25% of virus preparation), and 8 μg/mL polybrene, and placed into a single well of a 96-well plate. The plate was spun at 2,000 x g for 2 hours. Post-centrifugation, cells were incubated at 37 °C for 1 hour. Cells were then washed and replated in 24-well plates. At 48 hours post-transduction, selections were performed by adding 1 µg/mL puromyocin to all wells. Lines that grew under selection were then tested for efficient knockdown of SNX27 by immunoblotting.

95

3.3.7 LTR Reporter Assay and p19 ELISA

The transactivation efficiency of Tax-1 was measured using the LTR-luc reporter plasmid. 5x105 HEK293T cells (scramble control and SNX27 knocked down) were seeded in 6-well dishes. The next day, cells were transfected using TransIT-2020 (Mirus) with the following plasmids: wtHTLV-1 (1,000 ng), LTR-luc (100 ng), and RTK (10 ng).

48 hours post-transfection cells were collected and lysed in passive lysis buffer

(Promega). Lysates were then subjected to the Dual-Glo Luciferase Assay (Promega) per the manufacturer’s instructions, and read in a FilterMax F5 Microplate Reader

(Molecular Devices). Supernatants were also collected and the concentration of HTLV-1 p19 Gag was measured by p19 ELISA (Zeptometrix, Buffalo, NY) performed per the manufacturer’s instructions.

3.3.8 GLUT1 flow cytometry

To measure the surface presence of GLUT1, a flow cytometry-based assay was performed. HEK293T cells (1x105) cells were stained with 2 µL of the GLUT1-RBD-

GFP ligand (Metafora Biosystems, Evry, France) for 20 minutes at 37 °C. Stained cells were washed in PBS containing 2% FBS twice then resuspended in the same solution.

Stained and unstained (negative control) cells were analyzed using a Guava flow cytometer (EMD Millipore). Data was then analyzed and plotted using FlowJo software.

96 3.3.9 Statistical Analysis

Statistical analyses were performed using the unpaired Student's t test in GraphPad

Prism 6 (GraphPad Software, La Jolla, CA) as indicated. Statistical significance was defined as p<0.05.

3.4 Results

3.4.1 Generation of Mass Spectrometry data for Tax-1, Tax-2, and Mutants

Prior to proteomic analysis of the S-tag Tax constructs, we confirmed the efficiency of the S-tag pull downs. A pilot S-tag pull down was performed in HEK293T cells transfected with either empty S-tag or S-tag Tax-1. Using the anti-S-tag antibody we demonstrated that the tagged constructs were efficiently pulled down, while the p100/p52 antibody demonstrated the interaction between Tax-1 and p100/p52 (Figure 3-1 A). This is a known binding partner for Tax-1, and since no p100/p52 was seen in the empty S-tag pull down, the specificity of our pull down assay was confirmed (Figure 3-1 A). We then set out to identify Tax-1 binding partners important for activation of the alternative NF-

κB pathway. S-tag Tax-1, S-tag Tax-2, S-tag Tax-1 ΔPBM, S-tag Tax-1 225-232, and an empty S-tag vector were transfected into HEK293T cells and lysates were subjected to S- tag pull down. Samples were separated using SDS-PAGE and Gel Code Blue staining

(representative gel in Figure 3-1 B), and whole lanes were excised from the gel to be used in our proteomics analysis. Table 3-1 shows a subset of mass spectrometry data generated in these experiments. Our data contained several known interacting partners of Tax-1, including: scribble, beta-2-syntrophin, DLG-1, MAGI1, beta-1-syntrophin, lin-7 homolog

97 C, and histone-arginine methyltransferase CARM1288,385,397-400. The presence of these interactions demonstrated the robustness of our experimental approach and data.

3.4.2 Verification of Mass Spectrometry Data

Wild type Tax-1 and Tax-1 ΔPBM activated the alternative NF-κB pathway, while

Tax-1 225-232 and Tax-2 did not. We identified Tax-1 binding partners that bound only to the Tax constructs that activate the alternative NF-κB pathway, with the assumption that this binding pattern would represent required interactions for activation. Six proteins fit this binding pattern in our mass spectrometry data (shown in bold in Table 3-1): NF-κB- activating protein (NKAP), p12-Activated protein Kinase-Interacting Protein 1

(PAK1IP1), RuvB-Like 1 (RuvBL1), Activating Signal Cointegrator 1 Complex subunit 1

(ASCC1), Stress-Induced Phosphoprotein 1 (STIP1), and Far Upstream element Binding

Protein 3 (FUBP3). We performed pull down experiments to determine that these proteins interacted specifically with wild type Tax-1 and not Tax-2 as shown in the mass spectrometry data. HEK293T cells were transfected with either S-tag Tax-1, S-tag Tax-2, or the empty S-tag vector followed by S-tag pull down (Figure 3-2 A and B). DLG-1,

Scribble, and p100/p52 are interacting partners of Tax-1 that served as positive controls to demonstrate the effectiveness of the pull downs. PRMT5 does not interact with Tax-1, and served as a negative control. We found the expected binding for all positive controls, where interactions were only detected in the presence of wild type Tax-1. The negative control demonstrated no interaction with any of the transfected constructs, as expected.

All proteins of interest bound to both Tax-1 and Tax-2. This result contradicted the mass spectrometry data, which demonstrated that these proteins interacted only with Tax

98 constructs capable of activating alternative NF-κB. Because all interacting partners were found to interact with Tax-2, we analyzed the mass spectrometry data for other possible novel interactions expected to not be involved in alternative NF-κB activation.

3.4.3 Tax-1 Interacted with SNX27

All Tax constructs that featured a PBM (Tax-1 and Tax-1 225-232) interacted with

SNX27. The interaction was confirmed by repeating the pull down assay. HEK293T cells were transfected with either the empty S-tag vector or S-tag Tax-1, and lysates were subjected to S-tag pull down. Tax-1 was successfully pulled down and demonstrated to interact with SNX27 (Figure 3-3 A). The negative control sample did not show an interaction with SNX27, which demonstrated specific binding with Tax-1 (Figure 3-3 A).

We followed the pull down assay with a co-immunoprecipitation (co-IP) assay for

SNX27. HEK293T cells were transfected with S-tag Tax-1, and 24 hours post transfection, lysates were subjected to co-IP with SNX27 antibody. SNX27 was efficiently immunoprecipitated and co-precipitated with Tax-1, while no SNX27 or Tax-1 were detected in the IgG control (Figure 3-3 B). We next investigated whether the interaction between Tax-1 and SNX27 could be detected in the absence of exogenous expression of Tax-1. We utilized the HTLV-1 transformed T cell line SLB-1 for a co-IP of

SNX27. SLB-1 lysates were subjected to co-IP using SNX27 antibody. Probing with the

SNX27 antibody revealed efficient IP of SNX27, while probing with the Tax-1 antisera demonstrated that Tax-1 co-precipitated with SNX27 (Figure 3-3 C). The IgG control demonstrated no interaction with either Tax-1 or SNX27 (Figure 3-3 C). Collectively,

99 these data confirmed the interaction between Tax-1 and SNX27; moreover, this interaction was detected at endogenous levels of Tax-1 expression.

We then investigated the domains of Tax-1 and SNX27 essential for their interaction. The mass spectrometry data demonstrated that SNX27 only interacted with

Tax-1 when the PBM domain was present; SNX27 is known to feature a PDZ domain401.

To confirm that these domains were required for the interaction, we performed an S-tag pull down using HEK293T cells transfected with either empty S-tag vector, S-tag Tax-1, or S-tag Tax-1 ΔPBM. As shown before (Figure 3-3 A), Tax-1 interacted with SNX27; however, Tax-1 ΔPBM did not complex with SNX27 as shown by the lack of an SNX27 band in the Tax-1 ΔPBM pull down lane (Figure 3-3 D). This finding demonstrated that the PBM of Tax-1 was required for its interaction with SNX27. We confirmed that the

PDZ domain of SNX27 was required for this interaction by transfection of HEK293T cells with either Myc-SNX27 wild type or Myc-SNX27 ΔPDZ along with the indicated

S-tag constructs (empty, Tax-1, or Tax-1 ΔPBM). Co-IP was performed on the lysates using a Myc antibody, which revealed efficient IP of both wild-type SNX27 and SNX27

ΔPDZ (Figure 3-3 E). The only interaction detected in this assay was wild type Tax-1 with wild type SNX27. No Tax-1 construct co-precipitated with the SNX27 ΔPDZ protein (Figure 3-3 E). Likewise, no Tax or SNX27 protein was detected in the IgG controls (Figure 3-3 E). These results demonstrated that Tax-1 and SNX27 interaction involved the PBM and PDZ domains, respectively.

100 3.4.4 SNX27 Expression was Inversely Related to p19 Gag Release

We wanted to understand the relevance of the Tax-1-SNX27 interaction to HTLV-1 biology. We first analyzed whether Tax-1 over-expression altered the steady state levels of SNX27. We transfected HEK293T cells with a titration of Tax-1 expression vector

(from 100 ng to 2000 ng), and lysates were subjected to immunoblotting. We found that over-expression of Tax-1 had no effect of SNX27 steady state levels, as shown by the probing with SNX27 antibody (Figure 3-4 A). We then analyzed whether knock down of

SNX27 affected Tax-1 LTR transcriptional activity. HEK293T cells with knock down of

SNX27 were generated via transduction of lentivirus expressing SNX27-specific shRNA.

Stable cell lines were then transfected with the wtHTLV-1 molecular clone, the LTR-luc reporter construct, and RTK. At 48 hours post transfection, samples were collected and lysed; both immunoblot and Dual Glo luciferase assays were performed. The luciferase results revealed no statistically significant difference in Tax LTR activity between the scrambled or SNX27 knockdown cell lines (Scramble to SNX27 shRNA 1 p=0.4538 and scramble to SNX27 shRNA 2 p=0.079) (Figure 3-4 B). Immunoblotting revealed that

SNX27 expression reduced as expected (Figure 3-4 B). Together, these data demonstrated that knock down of SNX27 did not alter Tax-1 transactivation of the HTLV-1 viral promoter.

Because SNX27 is important for the localization and expression of GLUT1, an

HTLV-1 receptor molecule, we hypothesized that the Tax-1-SNX27 interaction may alter this function. We proposed that when SNX27 was over-expressed in HTLV-1 producing cells, lower levels of virions would be released. To test this hypothesis we transfected

HEK293T cells with the wtHTLV molecular clone and a titration of the SNX27

101 expression plasmid. We demonstrated that as SNX27 levels increased in these cells, a dose-dependent downward trend in the concentration of p19 in the supernatant was also observed, though these results were not statistically significant (Figure 3-4 C).

Furthermore, the HTLV-1-transformed cell line SLB-1 was knocked down for SNX27 via transduction of lentivirus expressing shRNA against SNX27. SLB-1 scramble and

SNX27 knockdown cells were cultured for 48 hours and supernatant was collected and analyzed for p19 Gag by ELISA. In the sample with efficient knockdown of SNX27

(SNX27 shRNA 2), a significant increase in p19 Gag was detected (p=0.0159) (Figure 3-

4 D). Together, these data suggests that SNX27 expression levels were inversely related to p19 Gag release into the cell culture supernatant.

3.4.5 Tax-1 Overexpression Reduced GLUT1 Surface Levels

We hypothesized that Tax-1 would affect the ability of SNX27 to localize GLUT1 at the plasma membrane. To determine if SNX27 function was affected by Tax-1,

HEK293T cells were transfected with either 500 ng or 2,000 ng of empty S-tag, S-tag

Tax-1, or S-tag Tax-1 ΔPBM expression plasmids. At 24 hours post-transfection, cells were collected and stained using the GLUT1-RBD-GFP ligand, which binds to GLUT1 on the surface of cells. Stained cells were then analyzed by flow cytometry. The cells transfected with 500 ng of S-tag Tax-1 demonstrated lower amounts of surface GLUT1 compared to cells transfected with either the empty S-tag or the S-tag Tax-1 ΔPBM

(Figure 3-5 A). This decrease was more dramatic in the 2,000 ng of S-tag Tax-1 sample

(Figure 3-5 A). At 2,000 ng of transfected S-tag Tax-1 ΔPBM, a small decrease in surface

GLUT1 was observed, but was not as substantial as that seen with Tax-1. Immunoblotting

102 was performed to confirm expression of the S-tag constructs (Figure 3-5 A). HEK293T cells knocked down for SNX27 expression via shRNA were also tested for surface

GLUT1 levels. In these cells, a similar slight decrease in surface GLUT1 was detected when SNX27 was knocked down as compared to the scramble control (Figure 3-5 B).

Collectively these data demonstrated that Tax-1 reduced the ability of SNX27 to localize

GLUT1 to the plasma membrane, and that this function was dependent on the Tax-1 PBM domain.

3.5 Discussion

HTLV-1 is the etiological agent of ATL, a malignancy of CD4+ T cells with a poor prognosis in humans402. The closely related virus HTLV-2 is not associated with any human malignancies29. Both HTLV-1 and HTLV-2 express a transcriptional activator of the viral genome, known as Tax-1 and Tax-2, respectively206. Tax-1 has a higher transforming potential than Tax-2, and this difference may explain how HTLV-1 causes disease while HTLV-2 does not206. Tax-1 activates both the classical and alternative NF-

κB pathways, while Tax-2 can only activate the classical NF-κB pathway231,237.

Therefore, we proposed that activation of alternative NF-κB plays a crucial role in the pathogenesis of HTLV-1.

In Chapter 2 we demonstrated that the 225-232 domain of Tax-1 was important for the activation of alternative NF-κB, and that mutation of this domain did not alter transformation capabilities of the virus in vivo. In this Chapter, we analyzed the binding partners of Tax-1 to identify interacting proteins required for Tax-1-mediated activation of alternative NF-κB signaling. Using S-tag Tax expression vectors, multiple rounds of

103 mass spectrometry were performed to identify binding partners of the different Tax constructs. These data sets were compared, contrasted, and searched for proteins that bound only to Tax constructs capable of activating the alternative NF-κB pathway (Tax-1 wild type and Tax-1 ΔPBM) and not those incapable of alternative NF-κB activation

(Tax-1 225-232 and Tax-2). We identified six candidate proteins via this screen: NKAP,

PAK1IP1, RuvBL1, ASCC1, STIP1, and FUBP3 and found these six candidates bound both Tax-1 and Tax-2. This binding pattern did not fit our expected profile of interacting only with Tax-1. In light of this finding, we pursued a unique binding partner of Tax-1 predicted to not be involved in alternative NF-κB activation.

In the mass spectrometry data, we identified a novel Tax-1 interacting partner,

SNX27. SNX27 is a member of the snorting nexin family of proteins, which are important for their retrieval and recycling of target proteins390. SNX27 is a unique member of the sorting nexin family because of a PDZ domain at the amino terminus of the protein391. Via binding to this PDZ domain, SNX27 was able to regulate the localization and expression of GLUT1392. Previous studies demonstrated that knockdown of SNX27 resulted in reduced cell surface-bound GLUT1, and decreased total levels of

GLUT1 protein as well392,403. GLUT1 is responsible for the import of glucose, but also plays a role in the HTLV-1 life cycle, where it serves as one of three receptor molecules for HTLV-1, the others being HSPG and NRP-1147. Another well studied retrovirus, HIV-

1, down regulates the expression of its receptor molecule (CD4) post-infection394. To date, HTLV-1 has no known mechanism of regulating receptor molecules post-infection, but recent studies demonstrated that GLUT1 expression levels in virus-producing cells was inversely related to the infectivity of virus produced393. We proposed that an

104 interaction between Tax-1 and SNX27 may be the first known mechanism HTLV-1 uses to regulate receptor molecules: Tax-1-meditated inhibition of the normal function of

SNX27 in recycling GLUT1 to the plasma membrane.

We confirmed the interaction between Tax-1 and SNX27 by mass spectrometry and

S-tag pull down followed by immunoblot analysis. Subsequently, we identified the domains of Tax-1 and SNX27 required for their interaction. We then demonstrated that

Tax-1 and SNX27 interaction is dependent on their PBM and PDZ domains respectively.

We next investigated if Tax-1 binding affected the function of SNX27. We established that Tax-1 expression had no effect on the steady state levels of SNX27 (by immunoblot).

Then, we analyzed whether SNX27 knockdown affected Tax-1 transactivation. We found that there was no statistically significant difference in Tax-1 transactivation, measured by luciferase assay, between scramble shRNA cells or SNX27 shRNA cells. This result confirmed that the Tax-1-SNX27 interaction had no effect on Tax-1 transactivation. Next we evaluated the effect of SNX27 expression on p19 Gag production by ELISA. We used p19 Gag concentration in the supernatant as a surrogate for virus production. We found that SNX27 over-expression in HEK293T cells also transfected with the HTLV-1 molecular clone showed a non-statistically significant dose dependent reduction in p19

Gag concentration in the supernatant. We also demonstrated that shRNA-mediated knockdown of SNX27 in SLB-1 cells resulted in a statistically significant increase in p19 concentration in the supernatant. This last finding suggests an inverse relationship between SNX27 expression and virus production, similar to the GLUT1 relationship with infectivity.

105 Finally, we determined whether Tax-1 interfered with the ability of SNX27 to localize GLUT1 at the plasma membrane. We found that Tax-1 expression caused a decrease in GLUT1 on the surface of cells and that increasing the amount of Tax-1 expressed caused a dose-dependent decrease in surface GLUT1 levels. The Tax-1 ΔPBM construct had no effect on surface GLUT1 at low levels of expression, and only a modest effect at high levels, but not as dramatic as wild type Tax-1. We found similar decreases in surface GLUT-1 levels in HEK293T cells knocked down for SNX27 as in the cells with Tax-1 over-expression. Collectively, these data demonstrated that Tax-1 inhibited

SNX27 function, which resulted in decreased GLUT1 surface expression.

In summary, we set out to discover an interacting partner of Tax-1 required for alternative NF-κB activation. While this protein has yet to be identified, a novel interacting partner of Tax-1 was discovered, SNX27. We demonstrated that, through

SNX27, Tax-1 reduced GLUT1 localization at the cell surface. To date, there have been no reports of receptor molecule regulation post-HTLV-1 infection, and our work demonstrated that Tax-1 could regulate GLUT-1. In HIV-1, the reduction of receptor molecules affects the infectivity and pathogenesis of the virus. The identification of Tax-1 as a regulator of receptor molecules could serve as a target to inhibit viral spread. Future studies should analyze how this function of Tax-1 alters the infectivity of the virus.

106

Figure 3.1 S-tag Tax-1 Pull Down Confirmation and Mass Spectrometry Sample Preparation

A) Verification of S-tag Tax-1 pull down specificity. HEK293T cells were transfected with either an empty S-tag vector or an S-tag Tax-1 expression vector. At 24 hours post- transfection, cells were collected, lysed, and subjected to S-tag pull down. Samples were analyzed via immunoblotting for p100/p52 or S-tag. When Tax-1 was pulled down, an interaction with p100/p52 was seen, as expected from past literature. No interaction between the empty-S-tag construct and p100/p52 was seen. B) A representative Gel Code Blue stained gel submitted for mass spectrometry. Equal levels of Tax-1 and Tax-2 were analyzed by mass spectrometry.

107

Construct Pulled Down Protein Name Size (kDa) Negative Tax-1 225-232 Tax-1 Tax-2 Utrophin 394 kDa 100% (143) 100% (108) 50% (1) scribble 178 kDa 100% (104) 100% (49) Beta-2-syntrophin 58 kDa 100% (34) 100% (40) Sorting nexin-27 61 kDa 100% (37) 100% (30) Discs, large homolog 1 (DLG-1) 100 kDa 100% (59) 100% (28) p100 97 kDa 100% (40) 100% (26) 100% (2) Dystrobrevin 79 kDa 100% (27) 100% (25) DnaJ (Hsp40) 55 kDa 100% (4) 100% (6) 100% (20) 100% (10) STIP1 63 kDa 50% (1) 100% (1) 100% (20) 100% (10) Liprin-alpha-1 136 kDa 100% (3) 100% (25) 100% (16) 100% (5) MAGI1 165 kDa 100% (30) 100% (16) FUBP3 62 kDa 100% (3) 100% (5) 100% (15) 100% (6) Beta-1-syntrophin 58 kDa 100% (18) 100% (14) HLA-B associated transcript 3 119 kDa 12% (0) 99% (1) 100% (13) 100% (29) Calponin-3 36 kDa 100% (4) 100% (34) 100% (12) 80% (1) Lin-7 homolog C 22 kDa 100% (28) 100% (11) Peripheral plasma membrane protein CASK 105 kDa 100% (35) 100% (11) MAGUK p55 subfamily member 66 kDa 100% (19) 100% (11) Erbb2 interacting protein 154 kDa 100% (34) 100% (10) Histone-arginine methyltransferase CARM1 66 kDa 100% (2) 100% (3) 100% (8) Activating signal cointegrator 1 complex subunit 1 41 kDa 100% (7) La ribonucleoprotein domain family, member 5 81 kDa 99% (1) 100% (6) 100% (5) Probable methyltransferase TARBP1 182 kDa 97% (0) 100% (6) 100% (4) Alpha-1-syntrophin 54 kDa 100% (12) 100% (6) PreS1 binding protein 54 kDa 100% (3) 100% (5) 96% (1) RING finger protein unkempt 88 kDa 100% (2) 100% (5) 100% (2) RuvB-like 1 50 kDa 100% (5) p21-activated protein kinase-interacting protein 1 44 kDa 100% (4) NF-kappa-B-activating protein 47 kDa 100% (2)

Table 3.1 List of Tax Interacting Proteins Identified Via Mass Spectrometry.

The proteins found interacting with Tax are listed in the first column, and the size of the protein is listed in column 2. Columns 3-6 show the interaction data of the indicated Tax construct with the protein listed in column one. The first number represents the confidence of the interaction, while the number in parenthesis list the number of peptides from the protein in column one found in that Tax pull down sample. Proteins listed in red have been previously identified as Tax interacting proteins, demonstrating the robustness of our experimental approach and data288,385,397-400. Proteins highlighted in blue were selected as candidates required for Tax-1-mediated alternative NF-κB activation. The proteins listed are ordered based on the number of peptides found with wild type Tax-1 in decreasing order.

108

Figure 3.2 Verification of Candidate Tax-1 Interacting Partners

A) Tax-1 Interacting partners associated with alternative NF-κB activation were screened for their ability to interact with both Tax-1 and Tax-2. HEK293T cells were transfected with either empty S-tag, S-tag Tax-1, or S-tag Tax-2. At 24 hours post-transfection, cells were collected, lysed, and subjected to S-tag pull down. Samples were analyzed via immunoblotting. p100, DLG-1, and Scribble served as Tax-1 positive controls for efficient pull downs, while PRMT5 served as a negative control. NKAP, PAK1IP1, RuvBL1, and ASCC1 were the candidates analyzed, and all bound to both Tax-1 and Tax-2. B) STIP1 and FUBP3 were analyzed for interaction with Tax-1 or Tax-2. Both were found to bind to Tax-1 and Tax-2. p100 served as a positive control, while PRMT5 served as a negative control. Data shown in panels A and B are representative of three independent experiments.

109

Figure 3.3 Tax-1 Interacted with SNX27, and this Interaction was Dependent on Tax-1 PBM and SNX27 PDZ Domains

A) HEK293T cells were transfected with either empty S-tag or S-tag Tax-1 expression plasmids. At 24 hours post-transfection, cells were collected, lysed, and subjected to S- tag pull down. Samples were analyzed via immunoblotting with antibodies against SNX27 and S-tag. B) A reciprocal co-IP was performed to test the SNX27 and Tax-1 interaction. HEK293T cells were transfected with S-tag Tax-1 and 24 hours post- transfection, cells were collected, lysed, and subjected to co-IP with antibodies against SNX27. Samples were analyzed via immunoblotting with antibodies against S-tag and SNX27. C) SNX27 co-IPs were performed using lysates from the HTLV-1 transformed line SLB-1. Samples were analyzed via immunoblotting with antibodies against Tax-1 and SNX27. D) HEK293T cells were transfected with either S-tag empty, S-tag Tax-1, or S-tag Tax-2 expression vectors. At 24 hours post-transfection, cells were collected, lysed, and subjected to S-tag pull dows. Samples were analyzed via immunoblotting with antibodies against S-tag and SNX27. E) HEK293T were transfected with either Myc SNX27 WT or Myc SNX27 ΔPDZ, along with either S-tag empty, S-tag Tax-1, or S-tag Tax-2 expression vectors. At 24 hours post-transfection, cells were collected, lysed, and subjected to co-IP using Myc antibodies. Samples were analyzed via immunoblotting using antibodies against either S-tag or Myc. All data shown is representative of two independent experiments.

110

Figure 3.4 SNX27 Expression was Inversely related to HTLV-1 p19 Gag Release

A) HEK293T cells were transfected with a titration of Tax-1 expression plasmid (0-2,000 ng). At 24 hours post-transfection, cells were collected, lysed, and analyzed via immunoblotting using antibodies against SNX27, Tax-1, and Actin (to determine equal loading). Data shown is representative of two independent experiments. B) HEK293T cell lines expressing either scramble shRNA or two different shRNAs against SNX27 were generated. Three separate clones for each condition were transfected with wtHTLV- 1 or an empty expression vector. At 48 hours post-transfection, cells were collected, lysed, and analyzed via immunoblotting and Dual Glo luciferase assay. All samples performed in duplicate. C) HEK293T cells were transfected with wtHTLV-1 and a Continued

111 Figure 3.4 Continued titration of SNX27. At 48 hours post-transfection, supernatants were collected and analyzed for the concentration of p19. All samples performed in duplicate and data shown is representative of two independent experiments. D) The HTLV-1 transformed cell line SLB-1 was transduced with lentiviral constructs expressing either scramble shRNA or shRNA against SNX27. 1x106 cells were plated in 12-well plates, and 48 hours post plating cells and supernatant were collected separately. The supernatant was subjected to p19 ELISA, while the cells were lysed and analyzed via immunoblotting. All data shown is representative of two independent experiments. The statistical significance between the scramble control and SNX27 shRNA 2 is as indicated.

112

Figure 3.5 Tax-1 Reduced Cell Surface GLUT1 Levels

A) HEK293T cells were transfected with either empty S-tag, S-tag Tax-1, or S-tag Tax-2 at either 500 or 2,000 ng of plasmid. At 24 hours post-transfection, cells were collected and stained with GLUT-RBD-GFP to detect surface GLUT1. Stained cells were then analyzed via flow cytometry. Immunoblotting was also performed to determine the expression of the Tax constructs. B) HEK293T cells were transduced with lentivirus expressing either scramble shRNA or shRNA against SNX27. Cells were collected and stained with GLUT-RBD-GFP to detect surface GLUT1. Stained cells were then analyzed via flow cytometry. Immunoblotting was also performed to determine the expression of SNX27. Three separate cell lines were analyzed for each condition, and results between lines were similar (only one line shown per condition for flow cytometry).

113

Chapter 4: Studies to Determine the Mechanism of HBZ Proliferative Activity

4.1 Abstract

HTLV-1 causes an aggressive malignancy of CD4+ T cells known as ATL. While

HTLV-1 Tax-1 expression is important for initiating oncogenesis, the antisense HBZ gene product is important for viral persistence and cell proliferation. In many cases of ATL,

HBZ it is the only detectably expressed viral gene. The mechanism of how HBZ promotes cell proliferation is still under investigation. We postulated that HBZ protein and mRNA interact with an unknown cellular gene product(s) to drive cell proliferation.

To identify interacting partners of the HBZ protein, pull downs were performed, followed by mass spectrometry. The proteomics data revealed no distinct candidate that interacted with HBZ that was associated with cell proliferation. To identify binding partners of the hbz mRNA, we utilized biotinylated synthetic RNAs representing hbz wild type and the hbz SM1 mutation, which were incapable of promoting proliferation. Pull down experiments revealed interacting partners of the hbz mRNA, however, we have not identified any proteins that bound to only wild type hbz mRNA that were associated with cell proliferation. Further studies will expand upon this preliminary data elucidate how

HBZ protein and mRNA induce cell proliferation.

114 4.2 Introduction

ATL is an aggressive malignancy of CD4+ T cells that develops in 5-10% of infected individuals approximately 30-40-years after infection with HTLV-158. Patients with the acute variant of ATL have a median survival time of 6 months post-diagnosis64.

HTLV-1 expresses 10 genes from the positive sense strand of its genome, which encode the structural and enzymatic proteins required for virus formation, and the regulatory genes, tax and rex that regulate viral gene expression199. The Tax protein promotes transformation of primary cells through its dysregulation of cellular signaling pathways, and is thought to be the main oncogenic protein of the virus74-76. HTLV-1 also expresses a single gene from the anti-sense strand of its genome, hbz201. In ATL-derived cell lines, expression of the sense strand genes typically is lost via promoter methylation, promoter deletion, or mutations of the tax gene79-82. However, hbz mRNA and protein are produced in all ATL-derived cell lines, which indicate their potential importance in

HTLV-1 pathogenesis83.

HBZ has several functions, most of which inhibit Tax activities. HBZ inhibits Tax transactivation via competitive binding to CREB and p300310. HBZ also inhibits the classical NF-κB pathway, which Tax promotes, by interacting with RelA and inhibiting its DNA binding as well as promoting its proteasomal degradation315. Tax can over- stimulate the classical NF-κB pathway, which results in senescence240. HBZ counters this senescence by reducing the activation of classical NF-κB by Tax. HBZ also promotes cell proliferation, which suggests that HBZ may function as the second oncogene of HTLV-

184.

115 Hbz mRNA induces proliferation in T cells83,84. A recent study demonstrated that the first 50 nucleotides of hbz are required, and that silent mutations in the RNA abolished its effect on proliferation319. The same study also demonstrated that hbz increased the expression of survivin, which promotes cell survival and proliferation319.

The exact mechanism of how HBZ protein promotes proliferation and how hbz mRNA up-regulates genes like survivin requires elucidation.

This study investigated how HBZ protein and mRNA up-regulate cell proliferation.

We hypothesized that both the protein and mRNA of HBZ interact with unknown cellular factors to regulate cell cycle progression. To elucidate these interactions, we screened both HBZ protein and mRNA using mass spectrometry with the goal of identifying the unknown interactions required for cell proliferation.

4.3 Materials and Methods

4.3.1 Plasmids and Cell Culture

The S-tag HBZ construct was generated by insertion of the HBZ open reading frame into the pTriEx-4 Neo vector (Novagen, Madison, WI) in-frame with the 6xHis and

S-tags. All cells were maintained in humidified incubators at 37 °C and 5% CO2.

HEK293T cells were grown in DMEM (Invitrogen, Carlsbad, CA) supplemented with

10% FBS (Invitrogen), 2 mM glutamine (Invitrogen), penicillin (100 U/mL) (Invitrogen), and streptomycin (100 μg/mL) (Invitrogen). The IL-2-dependent human T cell line Kit

225 were grown in RPMI 1640 medium (Invitrogen) supplemented with 10% FBS, 2 mM

116 glutamine, penicillin (100 U/mL), streptomycin (100 μg/mL), and hIL-2(20

U/mL)(Roche, Indianapolis, IN )83.

4.3.2 RNA Pull Down

Streptavidin beads (Thermo Fisher Scientific, Waltham, MA; 30 µL) were washed three times with PD Buffer (50 mM HEPES pH 7.5, 100 mM NaCl, 10 mM MgCl2, 0.1%

NP40, 2 mM DTT). Beads were resuspended in PD buffer, and folded biotinylated synthetic RNA (IDT, Coralville, IA) was added (generated by heating 10 µg of synthetic

RNA at 65 °C for 5 minutes, and then cooling to room temperature for 2 hours). Beads and RNAs were incubated for 30 minutes with gentle rocking. Kit-225 cells (1x107) were lysed with passive lysis buffer (Promega, Madison, WI) supplemented with cOmplete mini EDTA-free protease inhibitor (Roche) via the manufacturer’s instructions. Cleared lysates were incubated with 20 µg of yeast tRNA for 30 minutes at 4 °C. RNaseOUT

(Thermo Fisher Scientific; 200 units) was added to lysates. Lysates were then added to the streptavidin beads and RNA, and incubated for 1.5 hours at 4 °C with gentle rocking.

Beads were washed five times using PD buffer. Proteins were eluted in 1 x SDS loading dye and boiled for 10 minutes. Samples were subjected to mass spectrometry.

4.3.3 S-Tag Pull Down and Co-Immunoprecipitation

S-tag pull downs for mass spectrometry and interaction verification were performed as follows. Two 10-cm dishes were seeded with 2.5x106 HEK293T cells (two plates were used per condition). At 24 hours, cells were transfected with 10 µg of either empty S-tag

117 or S-tag HBZ expression plasmid using TransIT-2020 (Mirus, Madison, WI) per the manufacturer’s instructions. At 24 hours post-transfection, cells were collected in cold

PBS; cells from both plates were combined. Cells were lysed using passive lysis buffer

(Promega) supplemented with cOmplete mini EDTA-free protease inhibitor (Roche).

Lysates were cleared of debris by centrifugation, then incubated with 40 µL of S-agarose

(Novagen) overnight. Beads were centrifuged and washed four times with RIPA lysis buffer. Beads were resuspended in 1x SDS loading dye and boiled for 10 minutes.

Samples were subjected to either mass spectrometry or immunoblotting.

For co-immunoprecipitation, a 10-cm dish was seeded with 2.5x106 HEK293T cells. At 24 hours, the cells were transfected with 10 µg of either empty S-tag or S-tag

HBZ expression plasmid using TransIT-2020 (Mirus) per the manufacturer’s instructions.

At 24 hours post transfection, cells were collected and lysed in NP-40 lysis buffer.

Lysates were incubated with antibody for CHK1 (Santa Cruz Biotechnology, Dallas, TX;

2 ug). The next day Dynabeads Protein G (Thermo Fisher) were added and incubated for

2 hours. Beads were washed four times in NP40 lysis buffer (150 mM NaCl, 50 mM Tris-

Cl pH 8.0, and 1% NP-40) using magnetic separation using a 12-tube magnet (Qiagen,

Valencia, CA). Beads were resuspended in 1x SDS loading dye and boiled for 10 minutes. Supernatants were subjected to immunoblotting.

4.3.4 Mass Spectrometry

S-tag pull down samples were loaded and separated on Mini-PROTEAN TGX

Precast 4%–20% SDS-PAGE gels (BioRad, Hercules, CA). The gel was then stained

118 using GelCode Blue Stain Reagent (Thermo Fisher Scientific) per the manufacturer’s instructions. Whole lanes were excised from the gel and subjected to proteomics analysis as described previously395.

Capillary-liquid chromatography-tandem mass spectrometry (Capillary-

LC/MS/MS) for global protein identification was performed using a Thermo Finnigan

LTQ Orbitrap mass spectrometer equipped with a microspray source (Michrom

Bioresources Inc., Auburn, CA) operated in positive ion mode. Samples were separated on a capillary column (0.2X150 mm Magic C18AQ 3 μ 200A, Michrom Bioresources

Inc.) using an UltiMateTM 3000 HPLC system (LC-Packings, Dionex Co., Sunnyvale,

CA). The scan sequence of the mass spectrometer was based on the data-dependent

TopTenTM method. The resolution of a full scan was set at 30,000 to achieve a high mass accuracy MS determination.

The RAW data files collected on the mass spectrometer were converted to mzXML and MGF files by MassMatrix data conversion tools (version 1.3, http:// www.massmatrix.net/download). The resulting MGF files were searched using Mascot

Daemon by Matrix Science version 2.2.2 (Boston, MA) and the database searched against the full SwissProt database version 57.5 (471472 sequences; 167326533 residues) or

NCBI database version 20091013 (9873339 sequences; 3367482728 residues).

Considered modifications (variable) were methionine oxidation and the presence of carbamidomethyl cysteine. Three missed cleavages for the enzyme were permitted with a peptide tolerance of 1.2 Da, and the MS/MS ion tolerance was 0.8 Da. Search results were compiled and visualized using Scaffold 4 software (Proteome Software, Portland,

OR). Unweighted spectrum count and percent coverage provided semiquantitative data

119 analyses. Protein identifications were assigned using PeptideProphet (Institute for

Systems Biology, Seattle, WA). Proteins with 80% confidence were accepted with a minimum of one peptide displaying a 95% threshold confidence level.

4.3.5 Immunoblotting

For co-IPs and pull downs, samples were loaded and separated on Mini-PROTEAN

TGX Precast 4%–20% SDS-PAGE gels (BioRad) and transferred to nitrocellulose membranes (GE, Marlborough, MA). Membranes were blocked in 5% milk in PBST.

Blots were incubated with indicated primary antibody overnight and with secondary antibodies for 1 hour. Primary antibodies used: anti-S-tag (Abcam, Cambridge, UK;

1:1,000) and anti-CHK1 (Santa Cruz Biotechnology; 1:200). Secondary antibodies used: goat-anti-rabbit HRP (Santa Cruz Biotechnology; 1:5,000) and goat-anti-mouse HRP

(Santa Cruz Biotechnology; 1:5,000). VeriBlot secondary antibodies were used for co-IPs to avoid heavy chain detection: anti-mouse IgG VeriBlot for IP (Abcam; 1:1,000).

Membranes were developed using Immunocruz Luminol Reagent (Santa Cruz

Biotechnology) and imaged using an Amersham Imager 600 (GE).

4.4 Results

4.4.1 Identification of HBZ Protein-Interacting Partners

Binding partners of HBZ protein were identified by a mass spectrometry-based proteomics screen. HEK293T cells were transfected with either empty S-tag or S-tag

HBZ expression vectors. At 24 hours post-transfection, cells were collected, lysed, and

120 subjected to S-tag pull downs. Samples were analyzed by immunoblotting to confirm the efficacy of the pull down; the results demonstrated that HBZ protein was enriched in pull down samples (Figure 4-1 A). Ponceau staining of the membrane was performed to reveal the specificity of the S-tag pull downs (Figure 4-1 B). A second round of SDS-PAGE was performed, and the gel was stained with Gel Code Blue (Figure 4-1 C). Whole lanes were excised and subjected to mass spectrometry.

A subset of the mass spectrometry data is detailed in Table 4-1. Several known interacting partners of HBZ were discovered in this screen, such as CREB and AP-

1313,404,405. Preliminary analysis of the mass spectrometry data revealed a novel interaction between CHK1 and HBZ. We set out to confirm this interaction because of its possible importance in HBZ-induced proliferation. HEK293T cells were transfected with the S-tag HBZ expression vector. At 24 hours post-transfection, cells were collected, lysed, and a co-IP assay was performed using CHK1 antibodies. No interaction between

CHK1 and HBZ was detected, which indicated that the MS data for this protein was a false positive. We currently are evaluating other potential binding partners.

4.4.2 Identification of hbz mRNA-Interacting Partners

We investigated possible interacting partners of hbz mRNA that could be associated with cell proliferation. To perform this assay, we utilized the SM1 construct of hbz mRNA as a negative control. The SM1 construct features silent mutations throughout the first 200 nucleotides of the hbz gene; while these mutations do not alter the amino acid sequence generated, they do affect the secondary structure of the hbz mRNA83. We performed RNA pull downs utilizing 87-nucleotide biotinylated synthetic RNAs from

121 wild type hbz, SM1 hbz, and the Influenza A Virus HA gene (negative control) (see Figure

4-3 A for sequence used and Figure 4-3 B for predicted secondary structures). Pull down samples were then subjected SDS-PAGE (Figure 4-3 C) followed by mass spectrometry.

A representation of the results is shown in Table 4-2. Characterization of these interaction partners is ongoing.

4.5 Discussion

HTLV-1 is an oncogenic retrovirus that causes an aggressive malignancy of CD4+

T cells in 5-10% of infected individuals64. The primary driver of HTLV-1 oncogenesis is believed to be the Tax protein; however, recent studies have suggested the antisense genome strand-encoded protein of HTLV-1, HBZ, may also have a role406. Indeed, transgenic expression of HBZ in mice has been shown to lead to leukemia312. HBZ was initially described as a regulator of Tax, because of its ability to inhibit Tax-mediated viral transcription and the classical NF-κB pathway310,315. Several studies have shown that the protein and mRNA forms of HBZ can induce cell proliferation83,84,319. The mechanisms behind both of these effects is still under investigation.

We proposed that HBZ protein interacts with an unknown cellular protein to drive proliferation of cells. We performed a proteomic screen of HBZ-interacting proteins, and our generated list of proteins included the known HBZ binding partners CREB and AP-1

(Table 4-1)313,404,405. The CHK1 protein stood out as a possible candidate because of the role CHK1 plays in cell-cycle checkpoint regulation. The activation of CHK1 in response to DNA double-strand breaks inhibits the transition between G2 and M phases of the cell cycle407. CHK1 interacts with the HTLV-1 protein Tax, and Tax inhibits the kinase

122 activity of CHK1, leading to suppression of DNA damage-induced cell cycle arrest261.

The release from cell cycle arrest allows HTLV-1-infected cells to continue to proliferate in the presence of DNA damage. Interestingly, most ATL cell lines feature methylation or mutations that result in silencing of the 5’ LTR, which prevents expression of Tax79-82.

The 3’ LTR is intact in all ATL cell lines; therefore, we postulated that HBZ may also inhibit CHK1 function and allow ATL cell lines to avoid DNA damage-induced cell cycle arrest83. When verifying the interaction between CHK1 and HBZ, however, we found that the mass spectrometry result was likely a false positive, because no consistent interaction was detectable between the two proteins. Two other identified interaction partners of

HBZ are also involved in proliferation, Activating Transcription Factor 7 (ATF7) and

Death Associated Protein 3 (DAP3)408,409. Ongoing work is focused on characterizing these interactions to determine the role they may play in HBZ proliferation.

Recently, the effect of hbz mRNA on cell proliferation was associated with an increase of transcript levels of several cell cycle regulation and anti-apoptotic genes, most importantly survivin319. How hbz mRNA increases transcription of survivin remains to be elucidated. We hypothesized that hbz mRNA interacted with an unknown cellular protein and promoted transcription of survivin. To identify this protein-mRNA interaction, we compared two hbz mRNAs, the wild type mRNA, and the SM1 mRNA. These two mRNAs encode the identical peptide, but the SM1 construct features mutations to the

RNA sequence that alter its secondary structure, which abolishes its ability to induce cell proliferation. Biotinylated synthetic RNAs were produced for wild type hbz, SM1 hbz, and a control sequence from Influenza A. These RNAs were then used in an RNA pull down assay followed by mass spectrometry, and a list of proteins found to interact with

123 the RNAs was generated (Table 4-2). Analysis of this initial preliminary data revealed no definitive interacting partner that bound only the wild type mRNA and not the SM1 or control RNAs. Further review of the data revealed a known regulator of survivin transcription, Interleukin enhancer-binding factor 3 (ILF3)410. Both control RNAs were found to interact with ILF3, but an interaction with Flu RNA is expected based on previously published reports411. Current work aims to ascertain the role ILF3 may play in hbz proliferative effect, and ultimately discover the mechanism of hbz mRNAs proliferative action.

In summary, we investigated possible interacting partners for hbz mRNA and protein that could be associated with cell proliferation. We identified a candidate binding partner for HBZ protein that was potentially important for HBZ proliferative action, but further analysis revealed this interaction was a false positive in our proteomics data. We are currently focused on the role ATF7 and DAP3 may play in HBZ induced proliferation. Pull down of the hbz mRNA did not reveal a clear candidate for an interaction responsible for the proliferative effect of hbz, but the ILF3 interaction may prove important. We are actively continuing our search for interacting partners important for HBZ proliferative effects, and will focus on how survivin transcription is regulated in an effort to elucidate this mechanism.

124

Figure 4.1 Pull Down of S-tag HBZ

A) HEK293T cells were transfected with either an empty S-tag or S-tag HBZ expression vector. Cells were collected, lysed, and subjected to S-tag pull downs. Samples were analyzed by immunoblotting to determine the efficacy of the pull downs. B) Membranes were also subjected to Ponceau staining to determine the specificity of the S-tag pull downs. C) Image of GelCode Blue-stained gel of an HBZ pull down that was used in the mass spectrometry data. Arrow indicates the HBZ protein.

125

Table 4.1 HBZ Protein Interacting Partners

The proteins found interacting with HBZ are listed in the first column, and the size of the protein is listed in column 2. Column 3 and 4 show the interaction data with HBZ and the negative control, respectively. The first number represents the confidence of the interaction, while the number in parenthesis list the number of peptides from the protein in column one found in that pull down sample. Proteins listed in red have been previously identified as HBZ-interacting proteins, demonstrating the robustness of our data310,318. Protein shaded in blue was selected as candidate associated with HBZ–induced cell proliferation. The proteins listed are ordered based on the number of peptides found with HBZ in decreasing order.

126

Figure 4.2 HBZ did not interact with CHK1

HEK293T cells were transfected with the S-tag HBZ expression vector. At 24 hours post- transfection, cells were collected, lysed, and subjected to co-IP using the CHK1 antibody. Samples were analyzed using immunoblotting. The S-tag antibody detects HBZ protein. No interaction between HBZ and CHK1 was detected. Data shown is representative of two independent experiments.

127

Figure 4.3 hbz RNA Pull Down

A) Sequences used in the RNA pull down assay; B) Predicted secondary structure of the RNAs used in this study. Predictions were generated using Mfold web server (http://unafold.rna.albany.edu). C) Image of GelCode Blue stained gel of an hbz pull down that was used for mass spectrometry analysis.

128

Table 4.2 hbz RNA Interacting Partners

The proteins that interacted with hbz RNA are listed in the first column, and the size of the protein is listed in column 2. Columns 3-5 show the interaction data with control, wild type, and SM1 RNA, respectively. The first number represents the confidence of the interaction, while the number in parenthesis list the number of peptides from the protein in column one found in that pull down sample. Protein shaded in blue plays a role in survivin transcription. The proteins listed are ordered based on the number of peptides found with wild type hbz in decreasing order.

129

Chapter 5: Summary and Future Directions

5.1 Summary

HTLV-1 is the causative agent of several diseases in humans, including ATL and

HAM/TSP402. HTLV-1 is estimated to have infected between 15-20 million people world-wide, although only 5-10% of infected individuals eventually develop diseases associated with infection32,412. Endemic areas of HTLV-1 infection include Japan, the

Caribbean Islands, Central America, South America, and Africa32. A closely related virus, HTLV-2, shares 70% nucleotide similarity with HTLV-1, but is not associated with any human malignancies28,29. Differences in pathogenesis between the two viruses have been linked to the different functional capacity of their respective Tax proteins. Tax-1

(HTLV-1) has a higher transforming potential than Tax-2 (HTLV-2), and this difference is related to two domains present in Tax-1 but absent in Tax-2: the PBM and the second

LZ239,286. Previous studies have demonstrated that Tax-1 utilizes these two domains to activate the alternative NF-κB pathway, which is an activity lacking in Tax-2239,291. This observation led us to postulate that activation of the alternative NF-κB pathway is required for HTLV-1 pathogenesis.

In Chapter Two, we investigated the requirement of the PBM and second LZ for transformation of primary T cells in vitro. While characterizing these Tax-1 mutants, we discovered that the deletion of the PBM did not prevent alternative NF-κB activation, a

130 result that was contrary to previously published reports239. Our lab demonstrated that

HTLV-1 Tax-1 mutant with a deletion of the PBM was able to transform cells, but the process was significantly delayed207. Because our data revealed that this transformation delay was not tied to activation of the alternative NF-κB pathway, we investigated other pathways that Tax-1 may regulate. Tax-1 interacts with DLG-1 though the PBM domain, and deletion of the PBM results in HTLV-1-mediated transformation of cells with low levels of DLG-1207,290. DLG-1 is important for the localization of PTEN, a known inhibitor of the Akt signaling cascade246. In light of this, we found that Tax-1 induced

Akt activation. Conversely, a Tax-1 ΔPBM construct was incapable of such activation.

We next demonstrated that activation of Akt was blocked by forced membrane localization of PTEN. Ultimately, we proposed that the mechanism of activation of Akt was a competitive binding event between Tax-1 and PTEN for interaction with DLG-1.

PTEN relies on DLG-1 for localization to the plasma membrane, and by blocking this interaction, Tax-1 promotes Akt activation. Activation of Akt could be important in

HTLV-1 pathogenesis, and could serve as a potential therapeutic target for treatment of

ATL.

Work described in Chapter Two demonstrated that Tax-1 with a mutated second LZ domain could not activate the alternative NF-κB pathway. This mutant Tax-1 was utilized in an in vitro transformation assay to investigate the ability of HTLV-1 to transform T cells in the absence of alternative NF-κB pathway activation. We demonstrated that this

HTLV-1 Tax-1 mutant transformed cells as efficiently as wild type HTLV-1. Several

PBL lines were isolated from this transformation assay and demonstrated that no reversions of the mutation during transformation. Collectively, these data demonstrated

131 that activation of the alternative NF-κB pathway was not required for transformation of primary T cells in vitro. However, in vitro transformation is not equivalent to tumorigenesis in vivo; thus, future in vivo experiments will be required to elucidate the role of this pathway in disease development.

Chapter Three expanded on Chapter Two by focusing on the mechanism behind

Tax-1 activation of the alternative NF-κB pathway. We utilized a proteomic screen of

Tax-1 binding partners to identify cellular gene product(s) required for Tax-1 activation of the alternative NF-κB pathway. We investigated the binding profile of Tax constructs both capable and incapable of activating the alternative NF-κB pathway. We identified six candidates that bound only to Tax proteins capable of activating the alternative NF-

κB pathway. During verification of these candidates, we demonstrated that these proteins interacted with Tax-2 as well as Tax-1. We predict that binding partners important for activating the alternative NF-κB pathway will bind only to Tax-1 and not Tax-2. In light of this, we analyzed our proteomics data for additional interacting partners of Tax-1

SNX27 was identified as a novel binding partner of Tax-1. SNX27 belongs to the family of sorting nexin proteins that are involved in the localization and recycling of target proteins390,413. SNX27 is a unique member of this family because of the presence of a PDZ domain in addition to the functional domains common among all sorting nexin proteins414. SNX27 is required for sustained GLUT1 localization at the plasma membrane; knockdown of SNX27 resulted in a drastic redistribution of GLUT1 from the cell surface to lysosomal compartments, where it was degraded392. GLUT1 is a cellular receptor molecule for HTLV-1 Env binding147. Several retroviruses, including HIV, have a mechanism for regulating the localization of cellular receptor molecules post-

132 infection394. To date, HTLV-1 has no known mechanism for regulating GLUT1. We postulated that Tax-1 utilizes the interaction with SNX27 to modulate GLUT1 localization and expression. After confirming the interaction between SNX27 and Tax-1, we elucidated its biological relevance. We found that as SNX27 expression in cells was increased, p19 Gag production was reduced. In agreement, when SNX27 was knocked down in the HTLV-1 transformed cell line SLB1, we observed increased p19 Gag levels.

We next investigated the effect Tax-1 over-expression on the ability of SNX27 to localize

GLUT1 to the cell surface. We found that as Tax-1 expression increased, less GLUT1 was detectable on the cell surface; this effect was a similar to SNX27 knock down.

Collectively, we believe these data demonstrated the first known mechanism of HTLV-1 regulation of its receptor molecule post-infection.

Studies outlined in Chapter Four were initiated to better understand how both HBZ protein and mRNA induce cell proliferation. Using a similar experimental design described in Chapter Three, we wanted to identify a cellular gene product(s) that interacted with HBZ protein or mRNA that was required for cell proliferation. When we analyzed the interacting partners of HBZ protein, we found an interesting candidate in

CHK1, a known regulator of cell cycle progression. However, this interaction could not be confirmed by co-immunoprecipitation experiments, which indicated that CHK1 was a false positive in our mass spectrometry experiments. We utilized RNA pull downs followed by mass spectrometry of both wild type hbz and SM1 hbz mRNA; SM1 mRNA could not promote proliferation in its RNA form83. A list of candidates was generated, but no binding partner stood out that interacted solely with wild type hbz and not the SM1

133 hbz mRNA. We will continue pursuing these interaction partners of HBZ protein and mRNA to identify the mechanism behind HBZ induction of cell proliferation.

5.2 Future Directions

5.2.1 Tax-1 and the Alternative NF-κB pathway

Future work regarding the activation of the alternative NF-κB pathway will focus on the role this pathway plays in transformation in vivo. The humanized mouse model will provide a great opportunity to analyze how mutations to HTLV-1 alter disease development. Lethally irradiated PBL HTLV-1 producer lines with a mutation of the second LZ (Tax-1 225-232; generated in Chapter Two) will be injected into humanized mice. We will track the inoculated mice for disease development and analyze the role of alternative NF-κB activation (or lack of) on in vivo pathogenesis344,372,387,415. A similar experiment could be performed using the rabbit model of virus persistence. This experiment would demonstrate the role that activation of the alternative NF-κB pathway plays in persistence of the virus post-infection in an immune competent animal207,339-341.

The exact mechanism behind Tax-1 activation of alternative NF-κB still requires elucidation. Our goal is to identify a cellular gene product(s) important for this process, and we plan to continue our search, utilizing different cell lines (such as Jurkat). Because

Tax-1 activated alternative NF-κB in the HEK293T model, a next step would be to repeat the experiments in the natural target cell of HTLV-1 infection to obtain more relevant results. Once a candidate is identified, we plan to utilize siRNA knockdowns to determine its contribution to Tax-1-mediated alternative NF-κB activation. Candidates found to be important will be further characterized for potential therapeutic targeting. 134

5.2.2 Tax-1 and the Akt Pathway

Future work regarding Tax-1-mediated activation of Akt will measure the importance of this pathway in HTLV-1-mediated cell transformation. In vitro cell transformation assays will be performed in the presence of various Akt inhibitors, such as perifosine416. This experiment will allow us to determine the importance of Akt signaling in in vitro transformation. In vivo experiments using the humanized mouse model and Akt inhibitors are also of interest, and will demonstrate how Akt inhibition affects HTLV-1 pathogenesis. These experiments will demonstrate the viability of Akt inhibition as a prophylactic measure to block HTLV-1 oncogenesis.

5.2.3 Tax-1 and SNX27

Chapter Three demonstrated that Tax-1 interacted with SNX27 to affect the localization of GLUT1. Future studies will investigate the role SNX27 plays in HTLV-1 infection. The infectivity of virions produced by HTLV-1-infected cells with either knocked down or over-expressed SNX27 will be measured. This experiment will be performed by co-cultivating the SNX27-adjusted producer cell lines with the BHK1E6 cell line, which contain an HTLV-1 LTR-driven lacZ gene. The BHK1E6 cells allow visualization of infectivity by β-galactosidase staining, and can demonstrate the effect that SNX27 levels have on infection333. Also of interest is the presence of a PBM in the

HTLV-1 Env TM protein153. There is potential that SNX27 may regulate the localization of Env through this domain, in a manner similar to GLUT1 trafficking and cell surface

135 expression. We will determine if an interaction between Env and SNX27 exists via co-IP; and, if an interaction exists, we will evaluate the role it plays in HTLV-1 biology.

5.2.4 HBZ protein and mRNA effects on proliferation

In Chapter Four we identified interacting partners of both HBZ protein and mRNA.

We will continue screening for cell protein that interacts with HBZ that is associated with cell proliferation. HBZ protein over-expression was performed in HEK293T cells, and these lysates were used for our mass spectrometry experiments. It is possible that the interacting partner of HBZ important for its proliferative effect was not present in

HEK293T. Future experiments will use a T cell line for these experiments, with Kit-225 serving as an ideal candidate, as the proliferative effect has been demonstrated in that line83,319. If over-expression of HBZ proves difficult in T cell lines, recombinant protein could be utilized in a manner similar to the hbz RNA pull downs (see section 4.3.2). The effects of hbz mRNA on cell proliferation might depend on its tertiary structure.

Experiments utilizing the full length mRNAs instead of the short RNA sequences will be performed. Since survivin is thought to be responsible for effect of the RNA on proliferation, studies will investigate known regulators of its expression. It is likely that hbz mRNA alters these regulators, leading to survivin up-regulation and ultimately, cell proliferation.

5.3 Conclusion

The data outlined in this dissertation demonstrated several key facts about HTLV-1 biology and pathogenesis, which included a mechanism of how Tax-1 activated the Akt 136 signaling cascade, identification of a novel interacting protein of Tax-1 (SNX27), and a mechanism for how Tax-1 regulated GLUT1 localization post-infection. The regulation of GLUT1 post-infection is of special interest, because it is the first documented mechanism of HTLV-1 regulation of its cell receptor. Our studies also generated protein interaction data for Tax-1, HBZ protein, and hbz RNA, which will serve as the groundwork for future studies to analyze how their interactions may modulate HTLV-1 biology.

137

References

1. Rous, P. A Sarcoma of the Fowl Transmissible by an Agent Separable from the Tumor Cells. J. Exp. Med. 13, 397–411 (1911).

2. Ellermann, V. & Bang, O. Experimentelle Leukämie bei Hühnern. (1909).

3. Bittner, J. J. Some Possible Effects of Nursing on the Mammary Gland Tumor Incidence in Mice. Science 84, 162 (1936).

4. Van der Maaten, M. J., Boothe, A. D. & Seger, C. L. Isolation of a virus from cattle with persistent lymphocytosis. J. Natl. Cancer Inst. 49, 1649–1657 (1972).

5. Epstein, M. A. Composition of the Rous virus nucleoid. Nature 181, 1808 (1958).

6. Bather, R. The nucleic acid of partially purified Rous No. I sarcoma virus. Br. J. Cancer 11, 611–619 (1957).

7. Wilson, R. G. & Bader, J. P. Viral ribonucleic acid polymerase: chick-embryo cells infected with vesicular stomatitis virus or Rous-associated virus. Biochim. Biophys. Acta 103, 549–557 (1965).

8. Temin, H. M. The Effects of Actinomycin D on Growth of Rous Sarcoma Virus in Vitro. Virology 20, 577–582 (1963).

9. Temin, H. M. Homology Between Rna From Rous Sarcoma Virous and Dna From Rous Sarcoma Virus-Infected Cells. Proc. Natl. Acad. Sci. U.S.A. 52, 323– 329 (1964).

10. Bader, J. P. The Role of Deoxyribonucleic Acid in the Synthesis of Rous Sarcoma Virus. Virology 22, 462–468 (1964).

11. Temin, H. M. Nature of the provirus of Rous sarcoma. in (1964).

12. Fisher, S. Not beyond reasonable doubt: Howard Temin's provirus hypothesis revisited. Journal of the history of biology 43, 661–696 (Springer Netherlands, 2010).

138 13. Temin, H. M. & Mizutani, S. Viral RNA-dependent DNA Polymerase: RNA- dependent DNA Polymerase in Virions of Rous Sarcoma Virus. Nature 226, 1211–1213 (1970).

14. Baltimore, D. RNA-dependent DNA polymerase in virions of RNA tumour viruses. Nature 226, 1209–1211 (1970).

15. Freeman, W. M., Walker, S. J. & Vrana, K. E. Quantitative RT-PCR: pitfalls and potential. BioTechniques 26, 112–22– 124–5 (1999).

16. Gelb, L. D., Aaronson, S. A. & Martin, M. A. Heterogeneity of Murine Leukemia Virus in vitro DNA; Detection of Viral DNA in Mammalian Cells. Science 172, 1353–1355 (1971).

17. Stehelin, D., Varmus, H. E., Bishop, J. M. & Vogt, P. K. DNA related to the transforming gene(s) of avian sarcoma viruses is present in normal avian DNA. Nature 260, 170–173 (1976).

18. Ross, J., Aviv, H., Scolnick, E. & Leder, P. In vitro synthesis of DNA complementary to purified rabbit globin mRNA (RNA-dependent DNA polymerase-reticulocyte-hemoglobin-density gradient centrifugation-oligo(dT) primer). Proc. Natl. Acad. Sci. U.S.A. 69, 264–268 (1972).

19. Coffin, J. M. & Fan, H. The Discovery of Reverse Transcriptase. Annu Rev Virol 3, annurev–virology–110615–035556 (2016).

20. Goff, S. P. in Fields Virology 2, 1424–1473 (Lippincott Williams & Wilkins, 2007).

21. Fine, D. & Schochetman, G. Type D primate retroviruses: a review. Cancer Res. 38, 3123–3139 (1978).

22. Stoye, J. P. et al. in Virus Taxonomy 477–495 (Elsevier, 2012). doi:10.1016/B978-0-12-384684-6.00044-6

23. Poiesz, B. J. et al. Detection and isolation of type C retrovirus particles from fresh and cultured lymphocytes of a patient with cutaneous T-cell lymphoma. Proc. Natl. Acad. Sci. U.S.A. 77, 7415–7419 (1980).

24. Yoshida, M., Miyoshi, I. & Hinuma, Y. Isolation and characterization of retrovirus from cell lines of human adult T-cell leukemia and its implication in the disease. Proc. Natl. Acad. Sci. U.S.A. 79, 2031–2035 (1982).

25. Watanabe, T., Seiki, M. & Yoshida, M. Retrovirus terminology. Science 222, 1178–1178 (1983).

26. Gallo, R. C., Blattner, W. A., Reitz, M. S. & Ito, Y. HTLV: the virus of adult T- 139 cell leukaemia in Japan and elsewhere. Lancet 1, 683 (1982).

27. Gallo, R. C. History of the discoveries of the first human retroviruses: HTLV-1 and HTLV-2. Oncogene 24, 5926–5930 (2005).

28. Saxon, A., Stevens, R. H., Quan, S. G. & Golde, D. W. Immunologic characterization of hairy cell leukemias in continuous culture. J. Immunol. 120, 777–782 (1978).

29. Rosenblatt, J. D. et al. Integrated human T-cell leukemia virus II genome in CD8 + T cells from a patient with ‘atypical’ hairy cell leukemia: evidence for distinct T and B cell lymphoproliferative disorders. Blood 71, 363–369 (1988).

30. Wolfe, N. D. et al. Emergence of unique primate T-lymphotropic viruses among central African bushmeat hunters. Proc. Natl. Acad. Sci. U.S.A. 102, 7994–7999 (2005).

31. Mahieux, R. & Gessain, A. The human HTLV-3 and HTLV-4 retroviruses: new members of the HTLV family. Pathol. Biol. 57, 161–166 (2009).

32. Watanabe, T. Current status of HTLV-1 infection. Int J Hematol (2011). doi:10.1007/s12185-011-0934-4

33. Slattery, J. P., Franchini, G. & Gessain, A. Genomic evolution, patterns of global dissemination, and interspecies transmission of human and simian T-cell leukemia/lymphotropic viruses. Genome Res. 9, 525–540 (1999).

34. Proietti, F. A., Carneiro-Proietti, A. B. F., Catalan-Soares, B. C. & Murphy, E. L. Global epidemiology of HTLV-I infection and associated diseases. Oncogene 24, 6058–6068 (2005).

35. Manns, A., Hisada, M. & La Grenade, L. Human T-lymphotropic virus type I infection. Lancet 353, 1951–1958 (1999).

36. Van Dooren, S., Salemi, M. & Vandamme, A. M. Dating the origin of the African human T-cell lymphotropic virus type-i (HTLV-I) subtypes. Mol. Biol. Evol. 18, 661–671 (2001).

37. Gonçalves, D. U. et al. Epidemiology, treatment, and prevention of human T-cell leukemia virus type 1-associated diseases. Clin. Microbiol. Rev. 23, 577–589 (2010).

38. Birmann, B. M. et al. Population differences in immune marker profiles associated with human T-lymphotropic virus type I infection in Japan and Jamaica. Int. J. Cancer 124, 614–621 (2009).

39. Arisawa, K. et al. Evaluation of adult T-cell leukemia/lymphoma incidence and 140 its impact on non-Hodgkin lymphoma incidence in southwestern Japan. Int. J. Cancer 85, 319–324 (2000).

40. Maloney, E. M. et al. Incidence of HTLV-I-Associated Myelopathy/Tropical Spastic Paraparesis (HAM/TSP) in Jamaica and Trinidad. JAIDS Journal of Acquired Immune Deficiency Syndromes 17, 167 (1998).

41. Murphy, E. L. et al. Modelling the risk of adult T-cell leukemia/lymphoma in persons infected with human T-lymphotropic virus type I. Int. J. Cancer 43, 250– 253 (1989).

42. Kaplan, J. E. et al. The Risk of Development of HTLV-I-Associated Myelopathy/Tropical Spastic Paraparesis Among Persons Infected with HTLV-I. JAIDS Journal of Acquired Immune Deficiency Syndromes 3, 1096 (1990).

43. Rathsam-Pinheiro, R. H. et al. Ocular lesions in HTLV-1 infected patients from Salvador, State of Bahia: the city with the highest prevalence of this infection in Brazil. Rev. Soc. Bras. Med. Trop. 42, 633–637 (2009).

44. Mochizuki, M. et al. Uveitis Associated With Human T-cell Lymphotropic Virus Type I. American Journal of Ophthalmology 114, 123–129 (1992).

45. Gessain, A. & Gout, O. Chronic myelopathy associated with human T- lymphotropic virus type I (HTLV-I). Ann. Intern. Med. 117, 933–946 (1992).

46. Uchiyama, T., Yodoi, J., Sagawa, K., Takatsuki, K. & Uchino, H. Adult T-cell leukemia: clinical and hematologic features of 16 cases. Blood 50, 481–492 (1977).

47. Yamamoto, N., Okada, M., Koyanagi, Y., Kannagi, M. & Hinuma, Y. Transformation of human leukocytes by cocultivation with an adult T cell leukemia virus producer cell line. Science 217, 737–739 (1982).

48. Merl, S. et al. Efficient transformation of previously activated and dividing T lymphocytes by human T cell leukemia-lymphoma virus. Blood 64, 967–974 (1984).

49. Duc Dodon, M. & Gazzolo, L. Loss of interleukin-2 requirement for the generation of T colonies defines an early event of human T-lymphotropic virus type I infection. Blood 69, 12–17 (1987).

50. Blattner, W. et al. Human T-Cell Leukaemia/Lymphoma Virus-Associated Lymphoreticular Neoplasia in Jamaica. Lancet 322, 61–64 (1983).

51. Blattner, W. A. in Human Retrovirology: HTLV 251–265 (1990).

52. Yoshida, M., Seiki, M., Yamaguchi, K. & Takatsuki, K. Monoclonal integration 141 of human T-cell leukemia provirus in all primary tumors of adult T-cell leukemia suggests causative role of human T-cell leukemia virus in the disease. Proc. Natl. Acad. Sci. U.S.A. 81, 2534–2537 (1984).

53. Hinuma, Y. et al. Adult T-cell leukemia: antigen in an ATL cell line and detection of antibodies to the antigen in human sera. Proc. Natl. Acad. Sci. U.S.A. 78, 6476–6480 (1981).

54. Miyoshi, I. et al. Caution against blood transfusion from donors seropositive to adult T-cell leukaemia-associated antigens. Lancet 1, 683–684 (1982).

55. Hoshino, H. et al. Establishment and characterization of 10 cell lines derived from patients with adult T-cell leukemia. Proc. Natl. Acad. Sci. U.S.A. 80, 6061– 6065 (1983).

56. Hinuma, Y. et al. Antibodies to adult t‐cell leukemia‐virus‐associated antigen (atla) in sera from patients with atl and controls in japan: A nation‐wide sero‐ epidemiologic study. International Journal of Cancer 29, 631–635 (1982).

57. Robert-Guroff, M. et al. Natural antibodies to human retrovirus HTLV in a cluster of Japanese patients with adult T cell leukemia. Science 215, 975–978 (1982).

58. Qayyum, S. & Choi, J. K. Adult T-cell leukemia/lymphoma. Arch. Pathol. Lab. Med. 138, 282–286 (2014).

59. Sibon, D. et al. Adult T-Cell Leukemia/Lymphoma in a Caucasian Patient After Sexual Transmission of Human T-Cell Lymphotropic Virus Type 1. Open Forum Infectious Diseases 2, ofv032–ofv032 (2015).

60. Shimoyama, M. Adult T-cell leukemia/lymphoma and its clinical subtypes from the viewpoints of viral etiology. Curr. Top. Microbiol. Immunol. 115, 113–125 (1985).

61. Takatsuki, K. et al. Clinical diversity in adult T-cell leukemia-lymphoma. Cancer Res. 45, 4644s–4645s (1985).

62. Shimoyama, M. Diagnostic criteria and classification of clinical subtypes of adult T-cell leukaemia-lymphoma. A report from the Lymphoma Study Group (1984- 87). Br. J. Haematol. 79, 428–437 (1991).

63. Swerdlow, S. H., International Agency for Research on CancerWorld Health Organization. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. (International Agency for Research on Cancer, 2008).

64. Ratner, L. Adult T cell leukemia lymphoma. Front. Biosci. 9, 2852–2859 (2004).

142 65. Kikuchi, M. et al. Virus associated adult T-cell leukemia (ATL) in Japan: clinical, histological and immunological studies. Hematol Oncol 4, 67–81 (1986).

66. Yoshida, N. et al. Molecular characterization of chronic-type adult T-cell leukemia/lymphoma. Cancer Res. 74, 6129–6138 (2014).

67. Duggan, D. B. et al. HTLV-I-induced lymphoma mimicking Hodgkin's disease. Diagnosis by polymerase chain reaction amplification of specific HTLV-I sequences in tumor DNA. Blood 71, 1027–1032 (1988).

68. Hayward, W. S., Neel, B. G. & Astrin, S. M. Activation of a cellular onc gene by promoter insertion in ALV-induced lymphoid leukosis. Nature 290, 475–480 (1981).

69. Payne, G. S. et al. Analysis of avian leukosis virus DNA and RNA in bursal tumors: Viral gene expression is not required for maintenance of the tumor state. Cell 23, 311–322 (1981).

70. Nusse, R. The activation of cellular oncogenes by retroviral insertion. Trends in Genetics 2, 244–247 (1986).

71. Derse, D. et al. Human T-cell leukemia virus type 1 integration target sites in the human genome: comparison with those of other retroviruses. Journal of Virology 81, 6731–6741 (2007).

72. Cook, L. B. et al. The role of HTLV-1 clonality, proviral structure, and genomic integration site in adult T-cell leukemia/lymphoma. Blood 123, 3925–3931 (2014).

73. Satou, Y. et al. The retrovirus HTLV-1 inserts an ectopic CTCF-binding site into the human genome. Proceedings of the National Academy of Sciences 113, 3054–3059 (2016).

74. Mariner, J. M., Lantz, V., Waldmann, T. A. & Azimi, N. Human T cell lymphotropic virus type I Tax activates IL-15R alpha gene expression through an NF-kappa B site. J. Immunol. 166, 2602–2609 (2001).

75. Nicot, C., Mahieux, R., Takemoto, S. & Franchini, G. Bcl-XL is up-regulated by HTLV-I and HTLV-II in vitro and in ex vivo ATLL samples. Blood 96, 275–281 (2000).

76. Chandhasin, C., Ducu, R. I., Berkovich, E., Kastan, M. B. & Marriott, S. J. Human T-cell leukemia virus type 1 tax attenuates the ATM-mediated cellular DNA damage response. Journal of Virology 82, 6952–6961 (2008).

143 77. Jacobson, S., Shida, H., McFarlin, D. E., Fauci, A. S. & Koenig, S. Circulating CD8+ cytotoxic T lymphocytes specific for HTLV-I pX in patients with HTLV-I associated neurological disease. Nature 348, 245–248 (1990).

78. Parker, C. E., Nightingale, S., Taylor, G. P., Weber, J. & Bangham, C. R. Circulating anti-Tax cytotoxic T lymphocytes from human T-cell leukemia virus type I-infected people, with and without tropical spastic paraparesis, recognize multiple epitopes simultaneously. Journal of Virology 68, 2860–2868 (1994).

79. Furukawa, Y., Kubota, R., Tara, M., Izumo, S. & Osame, M. Existence of escape mutant in HTLV-I tax during the development of adult T-cell leukemia. Blood 97, 987–993 (2001).

80. Koiwa, T. et al. 5'-long terminal repeat-selective CpG methylation of latent human T-cell leukemia virus type 1 provirus in vitro and in vivo. Journal of Virology 76, 9389–9397 (2002).

81. Takeda, S. et al. Genetic and epigenetic inactivation of tax gene in adult T-cell leukemia cells. Int. J. Cancer 109, 559–567 (2004).

82. Tamiya, S. et al. Two types of defective human T-lymphotropic virus type I provirus in adult T-cell leukemia. Blood 88, 3065–3073 (1996).

83. Satou, Y., Yasunaga, J.-I., Yoshida, M. & Matsuoka, M. HTLV-I basic leucine zipper factor gene mRNA supports proliferation of adult T cell leukemia cells. Proc. Natl. Acad. Sci. U.S.A. 103, 720–725 (2006).

84. Arnold, J., Zimmerman, B., Li, M., Lairmore, M. D. & Green, P. L. Human T- cell leukemia virus type-1 antisense-encoded gene, Hbz, promotes T-lymphocyte proliferation. Blood 112, 3788–3797 (2008).

85. Tsukasaki, K. et al. Definition, prognostic factors, treatment, and response criteria of adult T-cell leukemia-lymphoma: a proposal from an international consensus meeting. in 27, 453–459 (2009).

86. 1984–1987, L. S. G. Major prognostic factors of patients with adult T-cell leukemia-lymphoma: A cooperative study. Leuk. Res. 15, 81–90 (1991).

87. Marsh, B. J. Infectious complications of human T cell leukemia/lymphoma virus type I infection. Clin Infect Dis. 23, 138–145 (1996).

88. Takasaki, Y. et al. Long-term study of indolent adult T-cell leukemia-lymphoma. Blood 115, 4337–4343 (2010).

89. Bazarbachi, A. et al. Meta-analysis on the use of zidovudine and interferon-alfa in adult T-cell leukemia/lymphoma showing improved survival in the leukemic

144 subtypes. J. Clin. Oncol. 28, 4177–4183 (2010).

90. Gill, P. S. et al. Treatment of Adult T-Cell Leukemia–Lymphoma with a Combination of Interferon Alfa and Zidovudine. N. Engl. J. Med. 332, 1744– 1748 (1995).

91. Yamada, Y. et al. A new G-CSF-supported combination chemotherapy, LSG15, for adult T-cell leukaemia-lymphoma: Japan Clinical Oncology Group Study 9303. Br. J. Haematol. 113, 375–382 (2001).

92. Taguchi, H. et al. An Intensive Chemotherapy of Adult T-Cell Leukemia/Lymphoma: CHOP Followed by Etoposide, Vindesine, Ranimustine, and Mitoxantrone with Granulocyte Colony-Stimulating Factor Support. JAIDS Journal of Acquired Immune Deficiency Syndromes 12, 182 (1996).

93. Tsukasaki, K. et al. VCAP-AMP-VECP compared with biweekly CHOP for adult T-cell leukemia-lymphoma: Japan Clinical Oncology Group Study JCOG9801. J. Clin. Oncol. 25, 5458–5464 (2007).

94. Bazarbachi, A., Suarez, F., Fields, P. & Hermine, O. How I treat adult T-cell leukemia/lymphoma. Blood 118, 1736–1745 (2011).

95. Waldmann, T. A. et al. Radioimmunotherapy of interleukin-2R alpha-expressing adult T-cell leukemia with Yttrium-90-labeled anti-Tac. Blood 86, 4063–4075 (1995).

96. Uchiyama, T. et al. Interleukin-2 receptor (Tac antigen) expressed on adult T cell leukemia cells. J. Clin. Invest. 76, 446–453 (1985).

97. Remer, M., Al-Shamkhani, A., Glennie, M. & Johnson, P. Mogamulizumab and the treatment of CCR4-positive T-cell lymphomas. Immunotherapy 6, 1187– 1206 (2014).

98. Yoshie, O. et al. Frequent expression of CCR4 in adult T-cell leukemia and human T-cell leukemia virus type 1-transformed T cells. Blood 99, 1505–1511 (2002).

99. Ishida, T. et al. Clinical Significance of CCR4 Expression in Adult T-Cell Leukemia/Lymphoma. Clin. Cancer Res. 9, 3625–3634 (2003).

100. Yamamoto, K. et al. Phase I Study of KW-0761, a Defucosylated Humanized Anti-CCR4 Antibody, in Relapsed Patients With Adult T-Cell Leukemia- Lymphoma and Peripheral T-Cell Lymphoma. Journal of Clinical Oncology 28, 1591–1598 (2010).

101. Ishida, T. et al. Defucosylated anti-CCR4 monoclonal antibody (KW-0761) for

145 relapsed adult T-cell leukemia-lymphoma: a multicenter phase II study. J. Clin. Oncol. 30, 837–842 (2012).

102. Ishida, T. et al. Dose-intensified chemotherapy alone or in combination with mogamulizumab in newly diagnosed aggressive adult T-cell leukaemia- lymphoma: a randomized phase II study. Br. J. Haematol. 169, 672–682 (2015).

103. Philip, T. et al. Autologous Bone Marrow Transplantation as Compared with Salvage Chemotherapy in Relapses of Chemotherapy-Sensitive Non-Hodgkin's Lymphoma. N. Engl. J. Med. 333, 1540–1545 (1995).

104. Tsukasaki, K. et al. Poor outcome of autologous stem cell transplantation for adult T cell leukemia/lymphoma: a case report and review of the literature. Bone Marrow Transplantation 23, 87–89 (1999).

105. Sobue, R. et al. Treatment of adult T cell leukemia with mega-dose cyclophosphamide and total body irradiation followed by allogeneic bone marrow transplantation. Bone Marrow Transplantation 2, 441–444 (1987).

106. Borg, A. et al. Successful treatment of HTLV‐1‐associated acute adult T‐cell leukaemia lymphoma by allogeneic bone marrow transplantation. Br. J. Haematol. 94, 713–715 (1996).

107. Utsunomiya, A. et al. Improved outcome of adult T cell leukemia/lymphoma with allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplantation 27, 15–20 (2001).

108. Itonaga, H. et al. Distinct Clinical Features of Infectious Complications in Adult T Cell Leukemia/Lymphoma Patients after Allogeneic Hematopoietic Stem Cell Transplantation: A Retrospective Analysis in the Nagasaki Transplant Group. Biology of Blood and Marrow Transplantation 19, 607–615 (2013).

109. Strachan, W. H. On a form of multiple neuritis prevalent in the West Indies. Practitioner (1897).

110. Cruickshank, E. K. A neuropathic syndrome of uncertain origin; review of 100 cases. West Indian Med J 5, 147–158 (1956).

111. Mani, K. S., Mani, A. J. & Montgomery, R. D. A spastic paraplegic syndrome in South India. Journal of the Neurological Sciences 9, 179–199 (1969).

112. Gessain, A. et al. HTLV-I and tropical spastic paraparesis in Africa. Lancet 2, 698 (1986).

113. Osame, M. et al. HTLV-I associated myelopathy, a new clinical entity. Lancet 1, 1031–1032 (1986).

146 114. Gessain, A. et al. Antibodies to human T-lymphotropic virus type-I in patients with tropical spastic paraparesis. Lancet 2, 407–410 (1985).

115. Lehky, T. J. et al. Detection of human T-lymphotropic virus type I (HTLV-I) tax RNA in the central nervous system of HTLV-I-associated myelopathy/tropical spastic paraparesis patients by in situ hybridization. Ann. Neurol. 37, 167–175 (1995).

116. Iannone, R. et al. HTLV-I DNA Sequences in CNS Tissue of a Patient with Tropical Spastic Paraparesis and HTLV-I-Associated Myelopathy. JAIDS Journal of Acquired Immune Deficiency Syndromes 5, 810 (1992).

117. Harrington, W. J. et al. Spastic ataxia associated with human T‐cell lymphotrophic virus type II infection. Ann. Neurol. 33, 411–414 (1993).

118. Jacobson, S. et al. Isolation of HTLV‐II from a patient with chronic, progressive neurological disease clinically indistinguishable from HTLV‐I‐associated myelopathy/tropical spastic paraparesis. Ann. Neurol. 33, 392–396 (1993).

119. Araújo, A. Q. C. et al. HTLV-I-Associated Myelopathy/Tropical Spastic Paraparesis in Brazil: A Nationwide Survey. JAIDS Journal of Acquired Immune Deficiency Syndromes 19, 536 (1998).

120. Olindo, S. et al. Natural History of Human T-Lymphotropic Virus 1–Associated Myelopathy: A 14-Year Follow-up Study. Arch Neurol 63, 1560–1566 (2006).

121. García-Vallejo, F., Domínguez, M. C. & Tamayo, O. Autoimmunity and molecular mimicry in tropical spastic paraparesis/human T-lymphotropic virus- associated myelopathy. Brazilian Journal of Medical and Biological Research 38, 241–250 (2005).

122. Levin, M. C. et al. Autoimmunity due to molecular mimicry as a cause of neurological disease. Nature Medicine 8, 509–513 (2002).

123. Taylor, G. P. et al. Zidovudine plus lamivudine in Human T-Lymphotropic Virus type-I-associated myelopathy: a randomised trial. Retrovirology 3, 1 (2006).

124. Izumo, S. et al. Interferon-alpha is effective in HTLV-I-associated myelopathy: a multicenter, randomized, double-blind, controlled trial. Neurology 46, 1016– 1021 (1996).

125. Arimura, K. et al. Safety and efficacy of interferon-α in 167 patients with human T-cell lymphotropic virus type 1–associated myelopathy. J. Neurovirol. 13, 364– 372 (2007).

126. Forrester, J. Uveitis: pathogenesis. Lancet 338, 1498–1501 (1991).

147 127. Mochizuki, M. et al. HTLV-I and uveitis. The Lancet 339, 1110 (1992).

128. Sagawa, K. et al. Immunopathological mechanisms of human T cell lymphotropic virus type 1 (HTLV-I) uveitis. Detection of HTLV-I-infected T cells in the eye and their constitutive cytokine production. J. Clin. Invest. 95, 852–858 (1995).

129. Kamoi, K. & Mochizuki, M. HTLV-1 uveitis. Front Microbiol 3, 270 (2012).

130. Nakao, K. Clinical Course of HTLV-I–Associated Uveitis. Japanese Journal of Ophthalmology 43, 404–409 (1999).

131. Goncalves, D. U. et al. Simultaneous occurrence of HTLV-I associated myelopathy, uveitis and smouldering adult T cell leukaemia. Int J STD AIDS 10, 336–337 (1999).

132. Ono, A., Mochizuki, M., Yamaguchi, K., Miyata, N. & Watanabe, T. Immunologic and virologic characterization of the primary infiltrating cells in the aqueous humor of human T-cell leukemia virus type-1 uveitis. Accumulation of the human T-cell leukemia virus type-1-infected cells and constitutive expression of viral and interleukin-6 messenger ribonucleic acids. Invest. Ophthalmol. Vis. Sci. 38, 676–689 (1997).

133. Masuoka, K., Sagawa, K., Mochizuki, M., Oizumi, K. & Itoh, K. Polyclonal use of T-cell receptor alpha for human T-cell lymphotropic virus type 1-infected T cells. Invest. Ophthalmol. Vis. Sci. 36, 254–258 (1995).

134. Yoshimura, K. et al. Clinical and immunologic features of human T-cell lymphotropic virus type I uveitis. American Journal of Ophthalmology 116, 156– 163 (1993).

135. LaGrenade, L., Hanchard, B., Fletcher, V., Cranston, B. & Blattner, W. Infective dermatitis of Jamaican children: a marker for HTLV-I infection. Lancet 336, 1345–1347 (1990).

136. Primo, J. R. L. et al. Infective dermatitis and human T cell lymphotropic virus type 1-associated myelopathy/tropical spastic paraparesis in childhood and adolescence. Clin Infect Dis. 41, 535–541 (2005).

137. Eguchi, K. et al. High seroprevalence of anti-HTLV-I antibody in rheumatoid arthritis. Arthritis Rheum. 39, 463–466 (1996).

138. Murphy, E. L. et al. Respiratory and urinary tract infections, arthritis, and asthma associated with HTLV-I and HTLV-II infection. Emerging Infect. Dis. 10, 109– 116 (2004).

148 139. Stumpf, B. P., Carneiro-Proietti, A. B., Proietti, F. A., Rocha, F. L.Interdisciplinary HTLV Research Group. Higher rate of major depression among blood donor candidates infected with human t-cell lymphotropic virus type 1. Int J Psychiatry Med 38, 345–355 (2008).

140. Knipe, D. M. & Howley, P. M. Fields Virology. (Lippincott Williams & Wilkins, 2013).

141. Gross, C. & Thoma-Kress, A. Molecular Mechanisms of HTLV-1 Cell-to-Cell Transmission. Viruses 8, 74 (2016).

142. Richardson, J. H., Edwards, A. J., Cruickshank, J. K., Rudge, P. & Dalgleish, A. G. In vivo cellular tropism of human T-cell leukemia virus type 1. Journal of Virology 64, 5682–5687 (1990).

143. Manel, N. et al. The ubiquitous glucose transporter GLUT-1 is a receptor for HTLV. Cell 115, 449–459 (2003).

144. Piñon, J. D. et al. Human T-cell leukemia virus type 1 envelope glycoprotein gp46 interacts with cell surface heparan sulfate proteoglycans. Journal of Virology 77, 9922–9930 (2003).

145. Ghez, D. et al. Neuropilin-1 is involved in human T-cell lymphotropic virus type 1 entry. Journal of Virology 80, 6844–6854 (2006).

146. Ghez, D., Lepelletier, Y., Jones, K. S., Pique, C. & Hermine, O. Current concepts regarding the HTLV-1 receptor complex. Retrovirology 7, 99 (2010).

147. Jones, K. S. et al. Molecular aspects of HTLV-1 entry: functional domains of the HTLV-1 surface subunit (SU) and their relationships to the entry receptors. Viruses 3, 794–810 (2011).

148. Lambert, S. et al. HTLV-1 uses HSPG and neuropilin-1 for entry by molecular mimicry of VEGF165. Blood 113, 5176–5185 (2009).

149. Jin, Q., Agrawal, L., VanHorn-Ali, Z. & Alkhatib, G. Infection of CD4+ T lymphocytes by the human T cell leukemia virus type 1 is mediated by the glucose transporter GLUT-1: Evidence using antibodies specific to the receptor's large extracellular domain. Virology 349, 184–196 (2006).

150. Melikyan, G. B. Common principles and intermediates of viral protein-mediated fusion: the HIV-1 paradigm. Retrovirology 5, 1 (2008).

151. Fan, N. et al. Infection of peripheral blood mononuclear cells and cell lines by cell-free human T-cell lymphoma/leukemia virus type I. J. Clin. Microbiol. 30, 905–910 (1992).

149 152. Daenke, S. & Booth, S. HTLV-1-induced cell fusion is limited at two distinct steps in the fusion pathway after receptor binding. J Cell Sci 113 ( Pt 1), 37–44 (2000).

153. Ilinskaya, A., Heidecker, G. & Jones, K. Interaction between the HTLV-1 envelope and cellular proteins: impact on virus infection and restriction. Future Med Chem 2, 1651–1668 (2010).

154. Cosset, F.-L. & Lavillette, D. in 73, 121–183 (Elsevier, 2011).

155. Cosnefroy, O., Murray, P. J. & Bishop, K. N. HIV-1 capsid uncoating initiates after the first strand transfer of reverse transcription. Retrovirology 13, 58 (2016).

156. Martin, J. L., Maldonado, J. O., Mueller, J. D., Zhang, W. & Mansky, L. M. Molecular Studies of HTLV-1 Replication: An Update. Viruses 8, 31 (2016).

157. Gilboa, E., Mitra, S. W., Goff, S. & Baltimore, D. A detailed model of reverse transcription and tests of crucial aspects. Cell 18, 93–100 (1979).

158. Seiki, M., Hattori, S. & Yoshida, M. Human adult T-cell leukemia virus: molecular cloning of the provirus DNA and the unique terminal structure. Proc. Natl. Acad. Sci. U.S.A. 79, 6899–6902 (1982).

159. Haseltine, W. A., Kleid, D. G., Panet, A., Rothenberg, E. & Baltimore, D. Ordered transcription of RNA tumor virus genomes. Journal of Molecular Biology 106, 109–131 (1976).

160. Chen, Y. Mechanism of Minus Strand Strong Stop Transfer in HIV-1 Reverse Transcription. Journal of Biological Chemistry 278, 8006–8017 (2002).

161. Tsuchihashi, Z. & Brown, P. O. DNA strand exchange and selective DNA annealing promoted by the human immunodeficiency virus type 1 nucleocapsid protein. Journal of Virology 68, 5863–5870 (1994).

162. DeStefano, J., Ghosh, J., Prasad, B. & Raja, A. High fidelity of internal strand transfer catalyzed by human immunodeficiency virus reverse transcriptase. J. Biol. Chem. 273, 1483–1489 (1998).

163. Smith, C. M., Leon, O., Smith, J. S. & Roth, M. J. Sequence requirements for removal of tRNA by an isolated human immunodeficiency virus type 1 RNase H domain. Journal of Virology 72, 6805–6812 (1998).

164. Hottiger, M., Podust, V. N., Thimmig, R. L., McHenry, C. & Hübscher, U. Strand displacement activity of the human immunodeficiency virus type 1 reverse transcriptase heterodimer and its individual subunits. J. Biol. Chem. 269, 986–991 (1994).

150 165. Cherepanov, P., Maertens, G. N. & Hare, S. Structural insights into the retroviral DNA integration apparatus. Current Opinion in Structural Biology 21, 249–256 (2011).

166. Matreyek, K. & Engelman, A. Viral and Cellular Requirements for the Nuclear Entry of Retroviral Preintegration Nucleoprotein Complexes. Viruses 5, 2483– 2511 (2013).

167. Maertens, G. N. B′-protein phosphatase 2A is a functional binding partner of delta-retroviral integrase. Nucleic Acids Research 44, 364–376 (2016).

168. Panganiban, A. T. & Temin, H. M. The terminal nucleotides of retrovirus DNA are required for integration but not virus production. Nature 306, 155–160 (1983).

169. Engelman, A., Mizuuchi, K. & Craigie, R. HIV-1 DNA integration: mechanism of viral DNA cleavage and DNA strand transfer. Cell 67, 1211–1221 (1991).

170. Roe, T., Chow, S. A. & Brown, P. O. 3‘-end processing and kinetics of 5’-end joining during retroviral integration in vivo. Journal of Virology 71, 1334–1340 (1997).

171. Schröder, A. R. W. et al. HIV-1 integration in the human genome favors active genes and local hotspots. Cell 110, 521–529 (2002).

172. Wu, X., Li, Y., Crise, B. & Burgess, S. M. Transcription start regions in the human genome are favored targets for MLV integration. Science 300, 1749–1751 (2003).

173. Mitchell, R. S. et al. Retroviral DNA Integration: ASLV, HIV, and MLV Show Distinct Target Site Preferences. PLOS Biol 2, e234 (2004).

174. Narezkina, A. et al. Genome-Wide Analyses of Avian Sarcoma Virus Integration Sites. Journal of Virology 78, 11656–11663 (2004).

175. Nowrouzi, A. Genome-wide mapping of foamy virus vector integrations into a human cell line. J Gen Virol 87, 1339–1347 (2006).

176. Cherepanov, P. et al. HIV-1 integrase forms stable tetramers and associates with LEDGF/p75 protein in human cells. J. Biol. Chem. 278, 372–381 (2003).

177. Gupta, S. S. et al. Bromo- and extraterminal domain chromatin regulators serve as cofactors for murine leukemia virus integration. Journal of Virology 87, 12721–12736 (2013).

178. Zhao, L. J. & Giam, C. Z. Human T-cell lymphotropic virus type I (HTLV-I) transcriptional activator, Tax, enhances CREB binding to HTLV-I 21-base-pair 151 repeats by protein-protein interaction. Proc. Natl. Acad. Sci. U.S.A. 89, 7070– 7074 (1992).

179. Yoshimura, T., Fujisawa, J. & Yoshida, M. Multiple cDNA clones encoding nuclear proteins that bind to the tax-dependent enhancer of HTLV-1: all contain a leucine zipper structure and basic amino acid domain. EMBO J. 9, 2537–2542 (1990).

180. Zhao, L. J. & Giam, C. Z. Interaction of the human T-cell lymphotrophic virus type I (HTLV-I) transcriptional activator Tax with cellular factors that bind specifically to the 21-base-pair repeats in the HTLV-I enhancer. Proc. Natl. Acad. Sci. U.S.A. 88, 11445–11449 (1991).

181. Harrod, R. et al. p300 and p300/cAMP-responsive element-binding protein associated factor interact with human T-cell lymphotropic virus type-1 Tax in a multi-histone acetyltransferase/activator-enhancer complex. J. Biol. Chem. 275, 11852–11857 (2000).

182. Bex, F., Yin, M. J., Burny, A. & Gaynor, R. B. Differential transcriptional activation by human T-cell leukemia virus type 1 Tax mutants is mediated by distinct interactions with CREB binding protein and p300. Molecular and Cellular Biology 18, 2392–2405 (1998).

183. Hidaka, M., Inoue, J., Yoshida, M. & Seiki, M. Post-transcriptional regulator (rex) of HTLV-1 initiates expression of viral structural proteins but suppresses expression of regulatory proteins. EMBO J. 7, 519–523 (1988).

184. Ahmed, Y. F., Hanly, S. M., Malim, M. H., Cullen, B. R. & Greene, W. C. Structure-function analyses of the HTLV-I Rex and HIV-1 Rev RNA response elements: insights into the mechanism of Rex and Rev action. Genes & Development 4, 1014–1022 (1990).

185. Unge, T. et al. The Rex regulatory protein of human T-cell lymphotropic virus type I binds specifically to its target site within the viral RNA. Proc. Natl. Acad. Sci. U.S.A. 88, 7145–7149 (1991).

186. Zhang, X., Hakata, Y., Tanaka, Y. & Shida, H. CRM1, an RNA transporter, is a major species-specific restriction factor of human T cell leukemia virus type 1 (HTLV-1) in rat cells. Microbes and Infection 8, 851–859 (2006).

187. Hakata, Y., Yamada, M. & Shida, H. Rat CRM1 Is Responsible for the Poor Activity of Human T-Cell Leukemia Virus Type 1 Rex Protein in Rat Cells. Journal of Virology 75, 11515–11525 (2001).

188. Nicot, C. et al. HTLV-1-encoded p30II is a post-transcriptional negative regulator of viral replication. Nature Medicine 10, 197–201 (2004).

152 189. Jacks, T. & Varmus, H. E. Expression of the Rous sarcoma virus pol gene by ribosomal frameshifting. Science 230, 1237–1242 (1985).

190. Fogarty, K. H. et al. Characterization of Cytoplasmic Gag-Gag Interactions by Dual-Color Z-Scan Fluorescence Fluctuation Spectroscopy. Biophysical Journal 100, 1587–1595 (2011).

191. Henderson, L. E., Krutzsch, H. C. & Oroszlan, S. Myristyl amino-terminal acylation of murine retrovirus proteins: an unusual post-translational proteins modification. Proc. Natl. Acad. Sci. U.S.A. 80, 339–343 (1983).

192. Sun, M., Grigsby, I. F., Gorelick, R. J., Mansky, L. M. & Musier-Forsyth, K. Retrovirus-Specific Differences in Matrix and Nucleocapsid Protein-Nucleic Acid Interactions: Implications for Genomic RNA Packaging. Journal of Virology 88, 1271–1280 (2013).

193. Zhang, W., Cao, S., Martin, J. L., Mueller, J. D. & Mansky, L. M. Morphology and ultrastructure of retrovirus particles. AIMS Biophys 2, 343–369 (2015).

194. Rayne, F., Kajava, A. V., Lalanne, J. & Mamoun, R. Z. In Vivo Homodimerisation of HTLV-1 Gag and MA Gives Clues to the Retroviral Capsid and TM Envelope Protein Arrangement. Journal of Molecular Biology 343, 903–916 (2004).

195. Rayne, F., Bouamr, F., Lalanne, J. & Mamoun, R. Z. The NH2-Terminal Domain of the Human T-Cell Leukemia Virus Type 1 Capsid Protein Is Involved in Particle Formation. Journal of Virology 75, 5277–5287 (2001).

196. Lingappa, J. R., Reed, J. C., Tanaka, M., Chutiraka, K. & Robinson, B. A. How HIV-1 Gag assembles in cells: Putting together pieces of the puzzle. Virus Res. 193, 89–107 (2014).

197. Sundquist, W. I. & Krausslich, H. G. HIV-1 Assembly, Budding, and Maturation. Cold Spring Harbor Perspectives in Medicine 2, a006924–a006924 (2012).

198. Konvalinka, J., Kräusslich, H.-G. & Müller, B. Retroviral proteases and their roles in virion maturation. Virology 479-480, 403–417 (2015).

199. Franchini, G. Molecular mechanisms of human T-cell leukemia/lymphotropic virus type I infection. Blood 86, 3619–3639 (1995).

200. Johnson, J. M., Harrod, R. & Franchini, G. Molecular biology and pathogenesis of the human T-cell leukaemia/lymphotropic virus Type-1 (HTLV-1). International Journal of Experimental Pathology 82, 135–147 (2001).

153 201. Gaudray, G. et al. The complementary strand of the human T-cell leukemia virus type 1 RNA genome encodes a bZIP transcription factor that down-regulates viral transcription. Journal of Virology 76, 12813–12822 (2002).

202. Remy, E. et al. The annealing of tRNA3Lys to human immunodeficiency virus type 1 primer binding site is critically dependent on the NCp7 zinc fingers structure. J. Biol. Chem. 273, 4819–4822 (1998).

203. Pique, C., Tursz, T. & Dokhélar, M. C. Mutations introduced along the HTLV-I envelope gene result in a non-functional protein: a basis for envelope conservation? EMBO J. 9, 4243–4248 (1990).

204. Hatanaka, M. & Nam, S. H. Identification of HTLV-I gag protease and its sequential processing of the gag gene product. J. Cell. Biochem. 40, 15–30 (1989).

205. Schwartzberg, P., Colicelli, J. & Goff, S. P. Construction and analysis of deletion mutations in the pol gene of Moloney murine leukemia virus: a new viral function required for productive infection. Cell 37, 1043–1052 (1984).

206. Shirinian, M., Kfoury, Y., Dassouki, Z., El-Hajj, H. & Bazarbachi, A. Tax-1 and Tax-2 similarities and differences: focus on post-translational modifications and NF-κB activation. Front Microbiol 4, 231 (2013).

207. Xie, L., Yamamoto, B., Haoudi, A., Semmes, O. J. & Green, P. L. PDZ binding motif of HTLV-1 Tax promotes virus-mediated T-cell proliferation in vitro and persistence in vivo. Blood 107, 1980–1988 (2006).

208. Paskalis, H., Felber, B. K. & Pavlakis, G. N. Cis-acting sequences responsible for the transcriptional activation of human T-cell leukemia virus type I constitute a conditional enhancer. Proc. Natl. Acad. Sci. U.S.A. 83, 6558–6562 (1986).

209. Brady, J., Jeang, K.-T., Duvall, J. & Khoury, G. Identification of p40x- responsive regulatory sequences within the human T-cell leukemia virus type I long terminal repeat. Journal of Virology 61, 2175–2181 (1987).

210. Shimotohno, K., Takano, M., Teruuchi, T. & Miwa, M. Requirement of multiple copies of a 21-nucleotide sequence in the U3 regions of human T-cell leukemia virus type I and type II long terminal repeats for trans-acting activation of transcription. Proc. Natl. Acad. Sci. U.S.A. 83, 8112–8116 (1986).

211. Montagne, J. et al. Tax1 induction of the HTLV-I 21 bp enhancer requires cooperation between two cellular DNA-binding proteins. EMBO J. 9, 957–964 (1990).

212. Beimling, P. & Moelling, K. Direct interaction of CREB protein with 21 bp Tax-

154 response elements of HTLV-ILTR. Oncogene 7, 257–262 (1992).

213. Tie, F., Adya, N., Greene, W. C. & Giam, C. Z. Interaction of the human T- lymphotropic virus type 1 Tax dimer with CREB and the viral 21-base-pair repeat. Journal of Virology 70, 8368–8374 (1996).

214. Giebler, H. A. et al. Anchoring of CREB binding protein to the human T-cell leukemia virus type 1 promoter: a molecular mechanism of Tax transactivation. Molecular and Cellular Biology 17, 5156–5164 (1997).

215. Chrivia, J. C. et al. Phosphorylated CREB binds specifically to the nuclear protein CBP. Nature 365, 855–859 (1993).

216. Kim, Y. M. et al. Molecular Characterization of the Tax-containing HTLV-1 Enhancer Complex Reveals a Prominent Role for CREB Phosphorylation in Tax Transactivation. J. Biol. Chem. 282, 18750–18757 (2007).

217. Chan, H. M. & La Thangue, N. B. p300/CBP proteins: HATs for transcriptional bridges and scaffolds. J Cell Sci 114, 2363–2373 (2001).

218. Georges, S. A. et al. Tax recruitment of CBP/p300, via the KIX domain, reveals a potent requirement for acetyltransferase activity that is chromatin dependent and histone tail independent. Molecular and Cellular Biology 23, 3392–3404 (2003).

219. Sharma, N. & Nyborg, J. K. The coactivators CBP/p300 and the histone chaperone NAP1 promote transcription-independent nucleosome eviction at the HTLV-1 promoter. Proc. Natl. Acad. Sci. U.S.A. 105, 7959–7963 (2008).

220. Jiang, H. et al. PCAF interacts with tax and stimulates tax transactivation in a histone acetyltransferase-independent manner. Molecular and Cellular Biology 19, 8136–8145 (1999).

221. Michael D Robek, L. R. Immortalization of CD4(+) and CD8(+) T lymphocytes by human T-cell leukemia virus type 1 Tax mutants expressed in a functional molecular clone. Journal of Virology 73, 4856–4865 (1999).

222. Akagi, T., Ono, H. & Shimotohno, K. Characterization of T cells immortalized by Tax1 of human T-cell leukemia virus type 1. Blood 86, 4243–4249 (1995).

223. Grassmann, R., Aboud, M. & Jeang, K.-T. Molecular mechanisms of cellular transformation by HTLV-1 Tax. Oncogene 24, 5976–5985 (2005).

224. Gilmore, T. D. Introduction to NF-kappaB: players, pathways, perspectives. Oncogene 25, 6680–6684 (2006).

225. Scheidereit, C. I|[kappa]|B kinase complexes: gateways to NF-|[kappa]|B 155 activation and transcription. Oncogene 25, 6685–6705 (2006).

226. Hoffmann, A., Natoli, G. & Ghosh, G. Transcriptional regulation via the NF-κB signaling module. Oncogene 25, 6706–6716 (2006).

227. Geleziunas, R. et al. Human T-cell leukemia virus type 1 Tax induction of NF- kappaB involves activation of the IkappaB kinase alpha (IKKalpha) and IKKbeta cellular kinases. Molecular and Cellular Biology 18, 5157–5165 (1998).

228. Chu, Z.-L., DiDonato, J. A., Hawiger, J. & Ballard, D. W. The Tax Oncoprotein of Human T-cell Leukemia Virus Type 1 Associates with and Persistently Activates IκB Kinases Containing IKKα and IKKβ. J. Biol. Chem. 273, 15891– 15894 (1998).

229. Yamaoka, S. et al. Complementation cloning of NEMO, a component of the IkappaB kinase complex essential for NF-kappaB activation. Cell 93, 1231–1240 (1998).

230. Carter, R. S., Pennington, K. N., Ungurait, B. J. & Ballard, D. W. In Vivo Identification of Inducible Phosphoacceptors in the IKK /NEMO Subunit of Human I B Kinase. J. Biol. Chem. 278, 19642–19648 (2003).

231. Hiscott, J., Nguyen, T.-L. A., Arguello, M., Nakhaei, P. & Paz, S. Manipulation of the nuclear factor-κB pathway and the innate immune response by viruses. Oncogene 25, 6844–6867 (2006).

232. Kfoury, Y. et al. Ubiquitylated Tax targets and binds the IKK signalosome at the centrosome. Oncogene 27, 1665–1676 (2008).

233. Azran, I., Jeang, K.-T. & Aboud, M. High levels of cytoplasmic HTLV-1 Tax mutant proteins retain a Tax-NF-|[kappa]|B-CBP ternary complex in the cytoplasm. Oncogene 24, 4521–4530 (2005).

234. Lamsoul, I. et al. Exclusive Ubiquitination and Sumoylation on Overlapping Lysine Residues Mediate NF- B Activation by the Human T-Cell Leukemia Virus Tax Oncoprotein. Molecular and Cellular Biology 25, 10391–10406 (2005).

235. Nasr, R. et al. Tax ubiquitylation and sumoylation control critical cytoplasmic and nuclear steps of NF-κB activation. Blood 107, 4021–4029 (2006).

236. Xiao, G. et al. Retroviral oncoprotein Tax induces processing of NF- kappaB2/p100 in T cells: evidence for the involvement of IKKalpha. EMBO J. 20, 6805–6815 (2001).

237. Qu, Z., Qing, G., Rabson, A. & Xiao, G. Tax Deregulation of NF-κB2 p100

156 Processing Involves Both β-TrCP-dependent and -independent Mechanisms. J. Biol. Chem. 279, 44563–44572 (2004).

238. Smith, M. R. & Greene, W. C. Identification of HTLV-I tax trans-activator mutants exhibiting novel transcriptional phenotypes. Genes & Development 4, 1875–1885 (1990).

239. Higuchi, M. et al. Cooperation of NF-kappaB2/p100 activation and the PDZ domain binding motif signal in human T-cell leukemia virus type 1 (HTLV-1) Tax1 but not HTLV-2 Tax2 is crucial for interleukin-2-independent growth transformation of a T-cell line. Journal of Virology 81, 11900–11907 (2007).

240. Zhi, H. et al. NF-κB Hyper-Activation by HTLV-1 Tax Induces Cellular Senescence, but Can Be Alleviated by the Viral Anti-Sense Protein HBZ. PLoS Pathog 7, e1002025 (2011).

241. Ho, Y.-K. et al. HTLV-1 tax-induced rapid senescence is driven by the transcriptional activity of NF-κB and depends on chronically activated IKKα and p65/RelA. Journal of Virology 86, 9474–9483 (2012).

242. Georgescu, M. M. PTEN Tumor Suppressor Network in PI3K-Akt Pathway Control. Genes & Cancer 1, 1170–1177 (2011).

243. Engelman, J. A., Luo, J. & Cantley, L. C. The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat. Rev. Genet. 7, 606–619 (2006).

244. Maehama, T. & Dixon, J. E. The tumor suppressor, PTEN/MMAC1, dephosphorylates the lipid second messenger, phosphatidylinositol 3,4,5- trisphosphate. J. Biol. Chem. 273, 13375–13378 (1998).

245. Cantley, L. C. The phosphoinositide 3-kinase pathway. Science 296, 1655–1657 (2002).

246. Sotelo, N. S., Valiente, M., Gil, A. & Pulido, R. A functional network of the tumor suppressors APC, hDlg, and PTEN, that relies on recognition of specific PDZ-domains. J. Cell. Biochem. 113, 2661–2670 (2012).

247. Gao, T., Furnari, F. & Newton, A. C. PHLPP: a phosphatase that directly dephosphorylates Akt, promotes apoptosis, and suppresses tumor growth. Molecular Cell 18, 13–24 (2005).

248. Fukuda, R. I. et al. Alteration of phosphatidylinositol 3-kinase cascade in the multilobulated nuclear formation of adult T cell leukemia/lymphoma (ATLL). Proc. Natl. Acad. Sci. U.S.A. 102, 15213–15218 (2005).

157 249. Jeong, S.-J., Pise-Masison, C. A., Radonovich, M. F., Park, H. U. & Brady, J. N. Activated AKT regulates NF-κB activation, p53 inhibition and cell survival in HTLV-1-transformed cells. Oncogene 24, 6719–6728 (2005).

250. Liu, Y. et al. Phosphoinositide-3 kinase-PKB/Akt pathway activation is involved in fibroblast Rat-1 transformation by human T-cell leukemia virus type I tax. Oncogene 20, 2514–2526 (2001).

251. Bellon, M. & Nicot, C. Central role of PI3K in transcriptional activation of hTERT in HTLV-I–infected cells. Blood 112, 2946–2955 (2008).

252. Péloponèse, J.-M. & Jeang, K.-T. Role for Akt/protein kinase B and activator protein-1 in cellular proliferation induced by the human T-cell leukemia virus type 1 tax oncoprotein. J. Biol. Chem. 281, 8927–8938 (2006).

253. Schmitt, I., Rosin, O., Rohwer, P., Gossen, M. & Grassmann, R. Stimulation of cyclin-dependent kinase activity and G1- to S-phase transition in human lymphocytes by the human T-cell leukemia/lymphotropic virus type 1 Tax protein. Journal of Virology 72, 633–640 (1998).

254. Lemasson, I., Thébault, S., Sardet, C., Devaux, C. & Mesnard, J. M. Activation of E2F-mediated transcription by human T-cell leukemia virus type I Tax protein in a p16(INK4A)-negative T-cell line. J. Biol. Chem. 273, 23598–23604 (1998).

255. Ohtani, K. et al. Cell type-specific E2F activation and cell cycle progression induced by the oncogene product Tax of human T-cell leukemia virus type I. J. Biol. Chem. 275, 11154–11163 (2000).

256. Iwanaga, R. et al. Activation of the cyclin D2 and cdk6 genes through NF- kappaB is critical for cell-cycle progression induced by HTLV-I Tax. Oncogene 27, 5635–5642 (2008).

257. Iwanaga, R., Ohtani, K., Hayashi, T. & Nakamura, M. Molecular mechanism of cell cycle progression induced by the oncogene product Tax of human T-cell leukemia virus type I. Oncogene 20, 2055–2067 (2001).

258. Akagi, T., Ono, H. & Shimotohno, K. Expression of cell-cycle regulatory genes in HTLV-I infected T-cell lines: possible involvement of Tax1 in the altered expression of cyclin D2, p18Ink4 and p21Waf1/Cip1/Sdi1. Oncogene 12, 1645– 1652 (1996).

259. Suzuki, T., Narita, T., Uchida-Toita, M. & Yoshida, M. Down-regulation of the INK4 Family of Cyclin-Dependent Kinase Inhibitors by Tax Protein of HTLV-1 through Two Distinct Mechanisms. Virology 259, 384–391 (1999).

260. Haoudi, A., Daniels, R. C., Wong, E., Kupfer, G. & Semmes, O. J. Human T-cell

158 leukemia virus-I tax oncoprotein functionally targets a subnuclear complex involved in cellular DNA damage-response. J. Biol. Chem. 278, 37736–37744 (2003).

261. Park, H. U., Jeong, J.-H., Chung, J. H. & Brady, J. N. Human T-cell leukemia virus type 1 Tax interacts with Chk1 and attenuates DNA-damage induced G2 arrest mediated by Chk1. Oncogene 23, 4966–4974 (2004).

262. Park, H. U., Jeong, S.-J., Jeong, J.-H., Chung, J. H. & Brady, J. N. Human T-cell leukemia virus type 1 Tax attenuates γ-irradiation-induced apoptosis through physical interaction with Chk2. Oncogene 25, 438–447 (2005).

263. Ruben, S. et al. Cellular transcription factors and regulation of IL-2 receptor gene expression by HTLV-I tax gene product. Science 241, 89–92 (1988).

264. Waldele, K., Schneider, G., Ruckes, T. & Grassmann, R. Interleukin-13 Overexpression by Tax Transactivation: a Potential Autocrine Stimulus in Human T-Cell Leukemia Virus-Infected Lymphocytes. Journal of Virology 78, 6081–6090 (2004).

265. Curtiss, V. E., Smilde, R. & McGuire, K. L. Requirements for interleukin 2 promoter transactivation by the Tax protein of human T-cell leukemia virus type 1. Molecular and Cellular Biology 16, 3567–3575 (1996).

266. Migone, T. S. et al. Constitutively activated Jak-STAT pathway in T cells transformed with HTLV-I. Science 269, 79–81 (1995).

267. Lee, D. K., Kim, B.-C., Brady, J. N., Jeang, K.-T. & Kim, S.-J. Human T-cell lymphotropic virus type 1 tax inhibits transforming growth factor-beta signaling by blocking the association of Smad proteins with Smad-binding element. J. Biol. Chem. 277, 33766–33775 (2002).

268. Mori, N. et al. Human T-cell leukemia virus type I oncoprotein Tax represses Smad-dependent transforming growth factor beta signaling through interaction with CREB-binding protein/p300. Blood 97, 2137–2144 (2001).

269. Robertson, A. B., Klungland, A., Rognes, T. & Leiros, I. DNA Repair in Mammalian Cells. Cell. Mol. Life Sci. 66, 981–993 (2009).

270. Jeang, K.-T., Widen, S. G., Semmes, O. J. & Wilson, S. H. HTLV-I trans- activator protein, tax, is a trans-repressor of the human beta-polymerase gene. Science 247, 1082–1084 (1990).

271. Philpott, S. M. & Buehring, G. C. Defective DNA repair in cells with human T- cell leukemia/bovine leukemia viruses: role of tax gene. J. Natl. Cancer Inst. 91, 933–942 (1999).

159 272. Marteijn, J. A., Lans, H., Vermeulen, W. & Hoeijmakers, J. H. J. Understanding nucleotide excision repair and its roles in cancer and ageing. Nat. Rev. Mol. Cell Biol. 15, 465–481 (2014).

273. Schavinsky-Khrapunsky, Y., Priel, E. & Aboud, M. Dose-dependent dual effect of HTLV-1 tax oncoprotein on p53-dependent nucleotide excision repair in human T-cells. Int. J. Cancer 122, 305–316 (2007).

274. Kelman, Z. PCNA: structure, functions and interactions. Oncogene 14, 629–640 (1997).

275. Kao, S. Y. & Marriott, S. J. Disruption of nucleotide excision repair by the human T-cell leukemia virus type 1 Tax protein. Journal of Virology 73, 4299– 4304 (1999).

276. Li, G.-M. Mechanisms and functions of DNA mismatch repair. Cell Res. 18, 85– 98 (2008).

277. Morimoto, H., Tsukada, J., Kominato, Y. & Tanaka, Y. Reduced expression of human mismatch repair genes in adult T-cell leukemia. Am. J. Hematol. 78, 100– 107 (2005).

278. Durkin, S. S. et al. HTLV-1 Tax oncoprotein subverts the cellular DNA damage response via binding to DNA-dependent protein kinase. J. Biol. Chem. 283, 36311–36320 (2008).

279. Péloponèse, J.-M., Haller, K., Miyazato, A. & Jeang, K.-T. Abnormal centrosome amplification in cells through the targeting of Ran-binding protein-1 by the human T cell leukemia virus type-1 Tax oncoprotein. Proc. Natl. Acad. Sci. U.S.A. 102, 18974–18979 (2005).

280. Jin, D. Y., Spencer, F. & Jeang, K.-T. Human T cell leukemia virus type 1 oncoprotein Tax targets the human mitotic checkpoint protein MAD1. Cell 93, 81–91 (1998).

281. Nitta, T. et al. Centrosome amplification in adult T-cell leukemia and human T- cell leukemia virus type 1 Tax-induced human T cells. Cancer Sci. 97, 836–841 (2006).

282. Ching, Y.-P., Chan, S.-F., Jeang, K.-T. & Jin, D.-Y. The retroviral oncoprotein Tax targets the coiled-coil centrosomal protein TAX1BP2 to induce centrosome overduplication. Nat Cell Biol 8, 717–724 (2006).

283. Iwanaga, Y., Kasai, T., Kibler, K. & Jeang, K.-T. Characterization Of Regions In Hsmad1 Needed For Binding Hsmad2. J. Biol. Chem. 277, 31005–31013 (2002).

160 284. Higuchi, M. & Fujii, M. Distinct functions of HTLV-1 Tax1 from HTLV-2 Tax2 contribute key roles to viral pathogenesis. Retrovirology 6, 117 (2009).

285. Bertazzoni, U. et al. Intracellular Localization and Cellular Factors Interaction of HTLV-1 and HTLV-2 Tax Proteins: Similarities and Functional Differences. Viruses 3, 541–560 (2011).

286. Tsubata, C. et al. PDZ domain-binding motif of human T-cell leukemia virus type 1 Tax oncoprotein is essential for the interleukin 2 independent growth induction of a T-cell line. Retrovirology 2, 46 (2005).

287. Okajima, M. et al. Human T-cell leukemia virus type 1 Tax induces an aberrant clustering of the tumor suppressor Scribble through the PDZ domain-binding motif dependent and independent interaction. Virus Genes 37, 231–240 (2008).

288. Arpin-Andre, C. & Mesnard, J.-M. The PDZ domain-binding motif of the human T cell leukemia virus type 1 tax protein induces mislocalization of the tumor suppressor hScrib in T cells. J. Biol. Chem. 282, 33132–33141 (2007).

289. Suzuki, T., Ohsugi, Y., Uchida-Toita, M., Akiyama, T. & Yoshida, M. Tax oncoprotein of HTLV-1 binds to the human homologue of Drosophila discs large tumor suppressor protein, hDLG, and perturbs its function in cell growth control. Oncogene 18, 5967–5972 (1999).

290. Ishioka, K. et al. Inactivation of tumor suppressor Dlg1 augments transformation of a T-cell line induced by human T-cell leukemia virus type 1 Tax protein. Retrovirology 3, 71 (2006).

291. Shoji, T. et al. Identification of a novel motif responsible for the distinctive transforming activity of human T-cell leukemia virus (HTLV) type 1 Tax1 protein from HTLV-2 Tax2. Retrovirology 6, 83 (2009).

292. Inoue, J., Yoshida, M. & Seiki, M. Transcriptional (p40x) and post- transcriptional (p27x-III) regulators are required for the expression and replication of human T-cell leukemia virus type I genes. Proc. Natl. Acad. Sci. U.S.A. 84, 3653–3657 (1987).

293. Hammes, S. R. & Greene, W. C. Multiple Arginine Residues within the Basic Domain of HTLV-I Rex Are Required for Specific RNA Binding and Function. Virology 193, 41–49 (1993).

294. Seiki, M., Hikikoshi, A. & Yoshida, M. The U5 sequence is a cis-acting repressive element for genomic RNA expression of human T cell leukemia virus type I. Virology 176, 81–86 (1990).

295. King, J. A. et al. Nucleocytoplasmic transport of HTLV-1 RNA is regulated by

161 two independent LTR encoded nuclear retention elements. Oncogene 16, 3309– 3316 (1998).

296. Li, M., Kannian, P., Yin, H., Kesic, M. & Green, P. L. Human T Lymphotropic Virus Type 1 Regulatory and Accessory Gene Transcript Expression and Export Are Not Rex Dependent. AIDS Res Hum Retroviruses 28, 405–410 (2012).

297. Bai, X. T., Sinha-Datta, U., Ko, N. L., Bellon, M. & Nicot, C. Nuclear export and expression of HTLV-I tax/rex mRNA is RxRE/Rex-dependent. Journal of Virology (2012). doi:10.1128/JVI.06361-11

298. Bogerd, H. & Greene, W. C. Dominant negative mutants of human T-cell leukemia virus type I Rex and human immunodeficiency virus type 1 Rev fail to multimerize in vivo. Journal of Virology 67, 2496–2502 (1993).

299. Daly, T. J. et al. Biochemical characterization of binding of multiple HIV-1 Rev monomeric proteins to the Rev responsive element. Biochemistry 32, 10497– 10505 (1993).

300. Zemmel, R. W., Kelley, A. C., Karn, J. & Butler, J. G. P. Flexible Regions of RNA Structure Facilitate Co-operative Rev Assembly on the Rev-response Element. Journal of Molecular Biology 258, 763–777 (1996).

301. Hakata, Y., Umemoto, T., Matsushita, S. & Shida, H. Involvement of human CRM1 (exportin 1) in the export and multimerization of the Rex protein of human T-cell leukemia virus type 1. Journal of Virology 72, 6602–6607 (1998).

302. Palmeri, D. & Malim, M. H. Importin beta can mediate the nuclear import of an arginine-rich nuclear localization signal in the absence of importin alpha. Molecular and Cellular Biology 19, 1218–1225 (1999).

303. Adachi, Y. et al. Phosphorylation of the Rex protein of human T-cell leukemia virus type I. J. Biol. Chem. 267, 21977–21981 (1992).

304. Kesic, M., Doueiri, R., Ward, M., Semmes, O. J. & Green, P. L. Phosphorylation regulates human T-cell leukemia virus type 1 Rex function. Retrovirology 6, 105 (2009).

305. Nakano, K. & Watanabe, T. HTLV-1 Rex: the courier of viral messages making use of the host vehicle. Front Microbiol 3, (2012).

306. Murata, K. et al. A Novel Alternative Splicing Isoform of Human T-Cell Leukemia Virus Type 1 bZIP Factor (HBZ-SI) Targets Distinct Subnuclear Localization. Journal of Virology 80, 2495–2505 (2006).

307. Cavanagh, M.-H. et al. HTLV-I antisense transcripts initiating in the 3'LTR are

162 alternatively spliced and polyadenylated. Retrovirology 3, 15 (2006).

308. Ma, G., Yasunaga, J. & Matsuoka, M. Multifaceted functions and roles of HBZ in HTLV-1 pathogenesis. Retrovirology 13, (2016).

309. Usui, T. et al. Characteristic expression of HTLV-1 basic zipper factor (HBZ) transcripts in HTLV-1 provirus-positive cells. Retrovirology 5, 34 (2008).

310. Clerc, I. et al. An Interaction between the Human T Cell Leukemia Virus Type 1 Basic Leucine Zipper Factor (HBZ) and the KIX Domain of p300/CBP Contributes to the Down-regulation of Tax-dependent Viral Transcription by HBZ. J. Biol. Chem. 283, 23903–23913 (2008).

311. Zhao, T. et al. HTLV-1 bZIP factor enhances TGF-β signaling through p300 coactivator. Blood 118, 1865–1876 (2011).

312. Satou, Y. et al. HTLV-1 bZIP Factor Induces T-Cell Lymphoma and Systemic Inflammation In Vivo. PLoS Pathog 7, e1001274 (2011).

313. Basbous, J. et al. The HBZ Factor of Human T-cell Leukemia Virus Type I Dimerizes with Transcription Factors JunB and c-Jun and Modulates Their Transcriptional Activity. J. Biol. Chem. 278, 43620–43627 (2003).

314. Thébault, S., Basbous, J., Hivin, P., Devaux, C. & Mesnard, J.-M. HBZ interacts with JunD and stimulates its transcriptional activity. FEBS Lett. 562, 165–170 (2004).

315. Zhao, T. et al. Human T-cell leukemia virus type 1 bZIP factor selectively suppresses the classical pathway of NF-kappaB. 113, 2755–2764 (2009).

316. Philip, S., Zahoor, M. A., Zhi, H., Ho, Y.-K. & Giam, C.-Z. Regulation of human T-lymphotropic virus type I latency and reactivation by HBZ and Rex. PLoS Pathog 10, e1004040 (2014).

317. Hagiya, K., Yasunaga, J.-I., Satou, Y., Ohshima, K. & Matsuoka, M. ATF3, an HTLV-1 bZip factor binding protein, promotes proliferation of adult T-cell leukemia cells. Retrovirology 8, 19 (2011).

318. Ma, G., Yasunaga, J., Fan, J., Yanagawa, S. & Matsuoka, M. HTLV-1 bZIP factor dysregulates the Wnt pathways to support proliferation and migration of adult T-cell leukemia cells. Oncogene (2012). doi:10.1038/onc.2012.450

319. Mitobe, Y., Yasunaga, J.-I., Furuta, R. & Matsuoka, M. HTLV-1 bZIP Factor RNA and Protein Impart Distinct Functions on T-cell Proliferation and Survival. Cancer Res. 75, 4143–4152 (2015).

320. Tanaka-Nakanishi, A., Yasunaga, J.-I., Takai, K. & Matsuoka, M. HTLV-1 bZIP 163 Factor Suppresses Apoptosis by Attenuating the Function of FoxO3a and Altering its Localization. Cancer Res. 74, 188–200 (2013).

321. Vernin, C. et al. HTLV-1 bZIP factor HBZ promotes cell proliferation and genetic instability by activating oncomiRs. Cancer Res. (2014). doi:10.1158/0008-5472.CAN-13-3564

322. Koralnik, I. J. et al. Protein isoforms encoded by the pX region of human T-cell leukemia/lymphotropic virus type I. Proc. Natl. Acad. Sci. U.S.A. 89, 8813–8817 (1992).

323. Koralnik, I. J., Fullen, J. & Franchini, G. The p12I, p13II, and p30II proteins encoded by human T-cell leukemia/lymphotropic virus type I open reading frames I and II are localized in three different cellular compartments. Journal of Virology 67, 2360–2366 (1993).

324. Zhang, W. et al. Human T-lymphotropic virus type 1 p30(II) regulates gene transcription by binding CREB binding protein/p300. Journal of Virology 75, 9885–9895 (2001).

325. Michael, B. et al. Human T lymphotropic virus type-1 p30 II alters cellular gene expression to selectively enhance signaling pathways that activate T lymphocytes. Retrovirology 1, 39 (2004).

326. Taylor, J. M., Ghorbel, S. & Nicot, C. Genome wide analysis of human genes transcriptionally and post-transcriptionally regulated by the HTLV-I protein p30. BMC Genomics 2009 10:1 10, 311 (2009).

327. Datta, A. et al. Human T-lymphotropic virus type-1 p30 alters cell cycle G2 regulation of T lymphocytes to enhance cell survival. Retrovirology 4, 49 (2007).

328. Baydoun, H. H., Pancewicz, J., Bai, X. & Nicot, C. HTLV-I p30 inhibits multiple S phase entry checkpoints, decreases cyclin E-CDK2 interactions and delays cell cycle progression. Molecular Cancer 2013 12:1 9, 302 (2010).

329. Human T-lymphotropic virus type 1 p30 interacts with REGγ and ATM (Ataxia Telangiectasia Mutated) to promote cell survival. 8, A202 (2011).

330. Valeri, V. W. et al. Requirement of the human T-cell leukemia virus p12 and p30 products for infectivity of human dendritic cells and macaques but not rabbits. Blood 116, 3809–3817 (2010).

331. Banerjee, P., Feuer, G. & Barker, E. Human T-cell leukemia virus type 1 (HTLV-1) p12I down-modulates ICAM-1 and -2 and reduces adherence of natural killer cells, thereby protecting HTLV-1-infected primary CD4+ T cells from autologous natural killer cell-mediated cytotoxicity despite the reduction of

164 major histocompatibility complex class I molecules on infected cells. Journal of Virology 81, 9707–9717 (2007).

332. Van Prooyen, N. et al. Human T-cell leukemia virus type 1 p8 protein increases cellular conduits and virus transmission. Proceedings of the National Academy of Sciences 107, 20738–20743 (2010).

333. Pise-Masison, C. A. et al. Co-dependence of HTLV-1 p12 and p8 Functions in Virus Persistence. PLoS Pathog 10, e1004454 (2014).

334. Hiraragi, H. et al. Human T-lymphotropic virus type 1 mitochondrion-localizing protein p13(II) is required for viral infectivity in vivo. Journal of Virology 80, 3469–3476 (2006).

335. D'Agostino, D. M., Silic-Benussi, M., Hiraragi, H., Lairmore, M. D. & Ciminale, V. The human T-cell leukemia virus type 1 p13II protein: effects on mitochondrial function and cell growth. Cell Death Differ 12 Suppl 1, 905–915 (2005).

336. Hiraragi, H. et al. Human T-lymphotropic virus type 1 mitochondrion-localizing protein p13II sensitizes Jurkat T cells to Ras-mediated apoptosis. Journal of Virology 79, 9449–9457 (2005).

337. Kiyokawa, T. et al. p27x-III and p21x-III, proteins encoded by the pX sequence of human T-cell leukemia virus type I. Proc. Natl. Acad. Sci. U.S.A. 82, 8359– 8363 (1985).

338. Orita, S. et al. p21X mRNA is expressed as a singly spliced pX transcript from defective provirus genomes having a partial deletion of the pol-env region in human T-cell leukemia virus type 1-infected cells. Nucleic Acids Research 21, 3799–3807 (1993).

339. Yin, H. et al. Human T-cell leukemia virus type 2 antisense viral protein 2 is dispensable for in vitro immortalization but functions to repress early virus replication in vivo. Journal of Virology 86, 8412–8421 (2012).

340. Yamamoto, B. et al. Human T-cell leukemia virus type 2 post-transcriptional control protein p28 is required for viral infectivity and persistence in vivo. Retrovirology 5, 38 (2008).

341. Arnold, J. et al. Enhancement of infectivity and persistence in vivo by HBZ, a natural antisense coded protein of HTLV-1. Blood 107, 3976–3982 (2006).

342. Ye, J., Xie, L. & Green, P. L. Tax and Overlapping Rex Sequences Do Not Confer the Distinct Transformation Tropisms of Human T-Cell Leukemia Virus Types 1 and 2. Journal of Virology 77, 7728–7735 (2003).

165 343. Lairmore, M. D., Silverman, L. & Ratner, L. Animal models for human T- lymphotropic virus type 1 (HTLV-1) infection and transformation. Oncogene 24, 6005–6015 (2005).

344. Niewiesk, S. Animals Models of Human T Cell Leukemia Virus Type I Leukemogenesis. ILAR J 57, 3–11 (2016).

345. Panfil, A. R., Al-Saleem, J. J. & Green, P. L. Animal Models Utilized in HTLV- 1 Research. VRT 2013, 49–59 (2013).

346. Silverman, L. R., Phipps, A. J., Montgomery, A., Ratner, L. & Lairmore, M. D. Human T-cell lymphotropic virus type 1 open reading frame II-encoded p30II is required for in vivo replication: evidence of in vivo reversion. Journal of Virology 78, 3837–3845 (2004).

347. Conrad, S. F. et al. Immunogenicity and conformational properties of an N- linked glycosylated peptide epitope of human T-lymphotropic virus type 1 (HTLV-I). Biomed. Pept. Proteins Nucleic Acids 1, 83–92 (1995).

348. Frangione-Beebe, M., Rose, R. T., Kaumaya, P. T. & Schwendeman, S. P. Microencapsulation of a synthetic peptide epitope for HTLV-1 in biodegradable poly(D,L-lactide-co-glycolide) microspheres using a novel encapsulation technique. J Microencapsul 18, 663–677 (2001).

349. Lairmore, M. D. et al. Comparative biological responses of rabbits infected with human T-lymphotropic virus type I isolates from patients with lymphoproliferative and neurodegenerative disease. Int. J. Cancer 50, 124–130 (1992).

350. Franchini, G. et al. Highly attenuated HTLV type Ienv poxvirus vaccines induce protection against a cell-associated HTLV type I challenge in rabbits. AIDS Res Hum Retroviruses 11, 307–313 (1995).

351. Kannian, P. et al. Distinct transformation tropism exhibited by human T lymphotropic virus type 1 (HTLV-1) and HTLV-2 is the result of postinfection T cell clonal expansion. Journal of Virology 86, 3757–3766 (2012).

352. Ishiguro, N. et al. A rat model of human T lymphocyte virus type I (HTLV-I) infection. 1. Humoral antibody response, provirus integration, and HTLV-I- associated myelopathy/tropical spastic paraparesis-like myelopathy in seronegative HTLV-I carrier rats. J. Exp. Med. 176, 981–989 (1992).

353. Kushida, S. et al. HTLV-1-associated myelopathy/tropical spastic paraparesis- like rats by intravenous injection of HTLV-1-producing rabbit or human T-cell line into adult WKA rats. Jpn. J. Cancer Res. 84, 831–833 (1993).

166 354. Mizusawa, H. et al. A neuropathological study of paraparetic rats injected with HTLV-I-producing T cells. Journal of the Neurological Sciences 126, 101–108 (1994).

355. Kushida, S. et al. High incidence of HAM/TSP-like symptoms in WKA rats after administration of human T-cell leukemia virus type 1-producing cells. Journal of Virology 68, 7221–7226 (1994).

356. Miyoshi, I. et al. Infectious transmission of human T-cell leukemia virus to animals. Int. Symp. Princess Takamatsu Cancer Res. Fund 15, 121–127 (1984).

357. Nakamura, H. et al. Experimental inoculation of monkeys with autologous lymphoid cell lines immortalized by and producing human T-cell leukemia virus type-I. Int. J. Cancer 38, 867–875 (1986).

358. McGinn, T. M. et al. Association of primate T-cell lymphotropic virus infection of pig-tailed macaques with high mortality. Virology 304, 364–378 (2002).

359. Nakamura, H. et al. Protection of cynomolgus monkeys against infection by human T-cell leukemia virus type-I by immunization with viral env gene products produced in Escherichia coli. Int. J. Cancer 40, 403–407 (1987).

360. Nakamura, M. et al. Long-term persistence of protective immunity in cynomolgus monkeys immunized with a recombinant vaccinia virus expressing the human T cell leukaemia virus type I envelope gene. J Gen Virol 78, 147–152 (1997).

361. Kazanji, M. et al. Immunogenicity and Protective Efficacy of Recombinant Human T-Cell Leukemia/Lymphoma Virus Type 1 NYVAC and Naked DNA Vaccine Candidates in Squirrel Monkeys (Saimiri sciureus). Journal of Virology 75, 5939–5948 (2001).

362. Fang, J. et al. Transmission of human T-cell leukemia virus type 1 to mice. Journal of Virology 72, 3952–3957 (1998).

363. Kushida, S., Maeda, N., Fang, J., Uchida, K. & Miwa, M. Establishment of HTLV-1 carrier mice by injection with HTLV-1-producing T cells. Leukemia 11 Suppl 3, 260–262 (1997).

364. Feuer, G. et al. Establishment of human T-cell leukemia virus type I T-cell lymphomas in severe combined immunodeficient mice. Blood 82, 722–731 (1993).

365. Gao, L. et al. HTLV-1 Tax transgenic mice develop spontaneous osteolytic bone metastases prevented by osteoclast inhibition. Blood 106, 4294–4302 (2005).

167 366. Ohsugi, T. A transgenic mouse model of human T cell leukemia virus type 1- associated diseases. Front Microbiol 4, 49 (2013).

367. Grossman, W. J. et al. Development of leukemia in mice transgenic for the tax gene of human T-cell leukemia virus type I. Proc. Natl. Acad. Sci. U.S.A. 92, 1057–1061 (1995).

368. Kondo, A. et al. A model of in vivo cell proliferation of adult T-cell leukemia. Blood 82, 2501–2509 (1993).

369. Liu, Y. et al. Engraftment and tumorigenesis of HTLV-1 transformed T cell lines in SCID/bg and NOD/SCID mice. Leuk. Res. 26, 561–567 (2002).

370. Miyazato, P. et al. De novo human T-cell leukemia virus type 1 infection of human lymphocytes in NOD-SCID, common gamma-chain knockout mice. Journal of Virology 80, 10683–10691 (2006).

371. Villaudy, J. et al. HTLV-1 propels thymic human T cell development in ‘human immune system’ Rag2⁻/⁻ gamma c⁻/⁻ mice. PLoS Pathog 7, e1002231 (2011).

372. Tezuka, K. et al. An animal model of adult T-cell leukemia-humanized mice with HTLV-1 specific immunity. Blood 123, 346–355 (2013).

373. Romanelli, M. G. et al. Highlights on distinctive structural and functional properties of HTLV Tax proteins. Front Microbiol 4, 271 (2013).

374. Sun, S.-C. Non-canonical NF-κB signaling pathway. Cell Res. 21, 71–85 (2011).

375. Bassères, D. S. & Baldwin, A. S. Nuclear factor-κB and inhibitor of κB kinase pathways in oncogenic initiation and progression. Oncogene 25, 6817–6830 (2006).

376. Leslie, N. R. et al. Indirect mechanisms of carcinogenesis via downregulation of PTEN function. Advances in Enzyme Regulation 50, 112–118 (2010).

377. Olagnier, D. et al. HTLV-1 Tax-Mediated Inhibition of FOXO3a Activity Is Critical for the Persistence of Terminally Differentiated CD4+ T Cells. PLoS Pathog 10, e1004575 (2014).

378. Oteiza, A. & Mechti, N. The Human T-Cell Leukemia Virus Type 1 Oncoprotein Tax Controls Forkhead Box O4 Activity through Degradation by the Proteasome. Journal of Virology 85, 6480–6491 (2011).

379. Kimata, J. T., Wong, F. H., Wang, J. J. & Ratner, L. Construction and characterization of infectious human T-cell leukemia virus type 1 molecular clones. Virology 204, 656–664 (1994).

168 380. Chen, I. S., McLaughlin, J., Gasson, J. C., Clark, S. C. & Golde, D. W. Molecular characterization of genome of a novel human T-cell leukaemia virus. Nature 305, 502–505 (1983).

381. Ye, J., Silverman, L., Lairmore, M. D. & Green, P. L. HTLV-1 Rex is required for viral spread and persistence in vivo but is dispensable for cellular immortalization in vitro. Blood 102, 3963–3969 (2003).

382. Ramaswamy, S. et al. Regulation of G1 progression by the PTEN tumor suppressor protein is linked to inhibition of the phosphatidylinositol 3-kinase/Akt pathway. Proc. Natl. Acad. Sci. U.S.A. 96, 2110–2115 (1999).

383. Leslie, N. R. et al. Targeting mutants of PTEN reveal distinct subsets of tumour suppressor functions. Biochem. J. 357, 427–435 (2001).

384. Lin, E. I., Jeyifous, O. & Green, W. N. CASK Regulates SAP97 Conformation and Its Interactions with AMPA and NMDA Receptors. Journal of Neuroscience 33, 12067–12076 (2013).

385. Lee, S. S., Weiss, R. S. & Javier, R. T. Binding of human virus oncoproteins to hDlg/SAP97, a mammalian homolog of the Drosophila discs large tumor suppressor protein. Proc. Natl. Acad. Sci. U.S.A. 94, 6670–6675 (1997).

386. Endo, K. et al. Human T-cell leukemia virus type 2 (HTLV-2) Tax protein transforms a rat fibroblast cell line but less efficiently than HTLV-1 Tax. Journal of Virology 76, 2648–2653 (2002).

387. Banerjee, P. et al. Adult T-cell leukemia/lymphoma development in HTLV-1- infected humanized SCID mice. Blood 115, 2640–2648 (2010).

388. Currer, R. et al. HTLV tax: a fascinating multifunctional co-regulator of viral and cellular pathways. Front Microbiol 3, 406 (2012).

389. Franklin, A. A. et al. Transactivation by the human T-cell leukemia virus Tax protein is mediated through enhanced binding of activating transcription factor-2 (ATF-2) ATF-2 response and cAMP element-binding protein (CREB). J. Biol. Chem. 268, 21225–21231 (1993).

390. Cullen, P. J. Endosomal sorting and signalling: an emerging role for sorting nexins. Nat. Rev. Mol. Cell Biol. 9, 574–582 (2008).

391. Gallon, M. et al. A unique PDZ domain and arrestin-like fold interaction reveals mechanistic details of endocytic recycling by SNX27-retromer. Proceedings of the National Academy of Sciences 111, E3604–13 (2014).

392. Steinberg, F. et al. A global analysis of SNX27-retromer assembly and cargo

169 specificity reveals a function in glucose and metal ion transport. Nat Cell Biol 15, 461–471 (2013).

393. Maeda, Y. et al. Separate cellular localizations of human T-lymphotropic virus 1 (HTLV-1) Env and glucose transporter type 1 (GLUT1) are required for HTLV-1 Env-mediated fusion and infection. Journal of Virology 89, 502–511 (2015).

394. Levesque, K., Finzi, A., Binette, J. & Cohen, E. A. Role of CD4 receptor down- regulation during HIV-1 infection. Curr. HIV Res. 2, 51–59 (2004).

395. Doueiri, R. et al. Comparative host protein interactions with HTLV-1 p30 and HTLV-2 p28: insights into difference in pathobiology of human retroviruses. Retrovirology 9, 64 (2012).

396. Panfil, A. R. et al. PRMT5 Is Upregulated in HTLV-1-Mediated T-Cell Transformation and Selective Inhibition Alters Viral Gene Expression and Infected Cell Survival. Viruses 8, 7 (2016).

397. Rousset, R., Fabre, S., Desbois, C., Bantignies, F. & Jalinot, P. The C-terminus of the HTLV-1 Tax oncoprotein mediates interaction with the PDZ domain of cellular proteins. Oncogene 16, 643–654 (1998).

398. Béraud, C., Sun, S.-C., Ganchi, P., Ballard, D. W. & Greene, W. C. Human T- cell leukemia virus type I Tax associates with and is negatively regulated by the NF-kappa B2 p100 gene product: implications for viral latency. Molecular and Cellular Biology 14, 1374–1382 (1994).

399. Ohashi, M. et al. Human T-cell leukemia virus type 1 Tax oncoprotein induces and interacts with a multi-PDZ domain protein, MAGI-3. Virology 320, 52–62 (2004).

400. Jeong, S.-J. et al. Coactivator-associated arginine methyltransferase 1 enhances transcriptional activity of the human T-cell lymphotropic virus type 1 long terminal repeat through direct interaction with Tax. Journal of Virology 80, 10036–10044 (2006).

401. Pim, D., Broniarczyk, J., Bergant, M., Playford, M. P. & Banks, L. A Novel PDZ Domain Interaction Mediates the Binding between Human Papillomavirus 16 L2 and Sorting Nexin 27 and Modulates Virion Trafficking. Journal of Virology 89, 10145–10155 (2015).

402. Watanabe, T. HTLV-1-associated diseases. Int J Hematol 66, 257–278 (1997).

403. Clairfeuille, T. et al. A molecular code for endosomal recycling of phosphorylated cargos by the SNX27-retromer complex. Nat Struct Mol Biol (2016). doi:10.1038/nsmb.3290

170 404. Gaudray, G. et al. The complementary strand of the human T-cell leukemia virus type 1 RNA genome encodes a bZIP transcription factor that down-regulates viral transcription. Journal of Virology 76, 12813–12822 (2002).

405. Matsumoto, J., Ohshima, T., Isono, O. & Shimotohno, K. HTLV-1 HBZ suppresses AP-1 activity by impairing both the DNA-binding ability and the stability of c-Jun protein. Oncogene 24, 1001–1010 (2005).

406. Kataoka, K. et al. Integrated molecular analysis of adult T cell leukemia/lymphoma. Nat. Genet. 47, 1304–1315 (2015).

407. Langerak, P. & Russell, P. Regulatory networks integrating cell cycle control with DNA damage checkpoints and double-strand break repair. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 366, 3562–3571 (2011).

408. Hasegawa, H. et al. Cdk1-Mediated Phosphorylation of Human ATF7 at Thr-51 and Thr-53 Promotes Cell-Cycle Progression into M Phase. PLoS ONE 9, e116048 (2014).

409. Wazir, U. et al. The role of death-associated protein 3 in apoptosis, anoikis and human cancer. Cancer Cell International 2015 15:1 15, 39 (2015).

410. Nakamura, N. et al. Interleukin Enhancer-binding Factor 3/NF110 Is a Target of YM155, a Suppressant of Survivin. Mol. Cell Proteomics 11, M111.013243– M111.013243 (2012).

411. Shapira, S. D. et al. A physical and regulatory map of host-influenza interactions reveals pathways in H1N1 infection. Cell 139, 1255–1267 (2009).

412. Jeang, K.-T. HTLV-1 and adult T-cell leukemia: insights into viral transformation of cells 30 years after virus discovery. J. Formos. Med. Assoc. 109, 688–693 (2010).

413. Ghai, R., Mobli, M. & Collins, B. M. Measuring interactions of FERM domain- containing sorting Nexin proteins with endosomal lipids and cargo molecules. Meth. Enzymol. 534, 331–349 (2014).

414. Lunn, M.-L. et al. A unique sorting nexin regulates trafficking of potassium channels via a PDZ domain interaction. Nat. Neurosci. 10, 1249–1259 (2007).

415. Marsden, M. D. & Zack, J. A. Studies of retroviral infection in humanized mice. Virology 479-480, 297–309 (2015).

416. Gills, J. J. & Dennis, P. A. Perifosine: Update on a novel Akt inhibitor. Curr Oncol Rep 11, 102–110 (2009).

171

Appendix: List of Abbreviations in Alphabetical Order

AD Activation Domain

AMP Doxorubicin, ranimustine, prednisolone

AP-1 Activator protein-1

ASCC1 Activating signal cointergrator 1 complex subunit 1

ATF Activating transcription factor

ATF7 Activating transcription factor 7

ATL Adult T cell leukemia

AZT Azidothymidine

BER Base excision repair

BLV Bovine leukemia virus bZIP Basic leucine zipper

CA Capsid cAMP Cyclic adenosine monophosphate

CBP CREB binding protein

172 CD Central Domain

CDK4 Cyclin-dependent kinase 4

CHOP Cyclophosphamide, doxorubicin, vincristine, and prednisone

CKI CDK inhibitors

CNS Central nervous system co-IP Co-Immunoprecipitations

CR Complete remission

CREB cyclic AMP response element binding protein

CRE cAMP response element

CRM1 Chromosome region maintenance interacting protein 1

CRS cis-acting repressive sequence

CTCF CCCTC-binding factor

DAP3 Death associated protein 3

DNA-PK DNA-dependent protein kinase

Env Envelope

ER Endoplasmic reticulum

FBS Fetal bovine serum

FUBP3 Far upstream element binding protein 3

173 Gag Group-specific antigen

GLUT1 Glucose transporter type 1 gRNA Genomic RNA

HAM HTLV-1-associated meyelopathy

HBZ HTLV-1 basic leucine zipper hIL-2 Human interleukin-2

HSCT Hematopoietic stem cell transplantation

HSPG Heparin sulfate proteoglycans

HTLV Human T cell leukemia virus

HTLV-1 Human T cell leukemia virus type 1

HTLV-2 Human T cell leukemia virus type 2

HU HTLV-associated uveitis

IKK IκB Kinase

IL Interleukin

ILF3 Interleukin enhancer-binding factor 3

IN Intergrase

INF- α Interferon alpha

IκB Inhibitor of κB

174 LDH Lactose dehydrogenase

LTR Long terminal repeat

LZ Leucine-zipper like region

MA Matrix

MMR Mismatch repair

NAP1 Nucleosome assembly protein 1

NC Nucleocapsid

NEMO NF-κB essential modifier

NER Nucleotide excision repair

NF-κB Nuclear Factor κB

NHEJ Non-homologous end joining

NIK NF-κB-inducing kinase

NKAP NF-κB-activating protein

NRP-1 Neuropilion 1

ORF Open reading frame

PAK1IP1 p12-activated protein kinase-interacting protein 1

PBM PDZ-binding motif

PBMC Peripheral mononuclear blood cells

175 PBS Primer binding sequence

PBST PBS supplemented with 0.1% Tween-20

PCNA Proliferating cell nuclear antigen

PDK1 Phosphoinositide-dependent kinase 1

PDZ Post synaptic density protein

PHLPP PH domain leucine-rich repeat protein phosphatase

PI3K PI 3'-OH kinase

PIC Preintergration complex

PIP2 PI-3,5-bisphosphate

PIP3 PI3,4,5-trisphosphate

PP2A Serine/threonine protein phosphatase 2A ppt Polypurine tract

PR Partial remission

Pro Viral protease

PTEN Phosphatase and tensin homolog, deleted on chromosome ten

PTLV Primate T cell leukemia virus

Rb Retinoblastoma

RHD Rel homology domain

176 RT Reverse Transcriptase

RTK TK-renilla

RxRE Rex response element

SNX27 Sorting nexin 27

STAT Signal transducer of activated T cells

STIP1 Stress-induced phosphoprotein 1

STLV Simian T cell leukemia virus

SU Surface unit (of Envelope)

TGF β1 Transforming growth factor beta 1

TM Transmembrane (of Envelope)

TRE Tax-responsive element

Treg Regulatory T cell

TSP Tropic spastic paraparesis

VCAP Vincristine, cyclophosphamide, doxorubicin, prednisolone vCRE viral cAMP response element

VECP Vindesine, etoposide, carboplatin, prednisolone

177