The Pennsylvania State University The Graduate School Department of Cellular and Molecular Biology

CHARACTERIZATION OF THE

DEUBIQUITINATING ENZYME CYLD, A NOVEL

TARGET OF IκB KINASE REGULATING IMMUNE

FUNCTION

A Thesis in

Cellular and Molecular Biology

By

William Reiley

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

May 2005 The thesis of William W. Reiley was reviewed and approved* by the following:

Shao-Cong Sun Professor of Microbiology and Immunology Thesis Advisor Chair of Committee

Robert Bonneau Associate Professor of Microbiology and Immunology

Vincent Chau Professor of Cellular and Molecular Physiology

Neil Christensen Associate Professor of Pathology

David Spector Associate Professor of Microbiology and Immunology

Henry Donahue Professor of Orthopedics and Rehabilitation Director, Department of Cell and Molecular Biology Graduate Program

* Signatures are on file in the Graduate School iii

ABSTRACT

Since its beginnings in 1986, the NF-κB has been implicated in the regulation of diverse biological processes ranging from development to immune responses. The pleotropic biological role of NF-κB is explained by its involvement in the regulation of growth factors, immune receptors and ligands, and apoptosis inhibitors.

Knock-out mice studies have further defined the specific function of individual NF-κB members in regulating different aspects of immune function as well as the development of lymphoid and nonlymphoid tissues. In concert with functional studies, a vast amount of knowledge has been gained about how NF-κB is activated, how it is regulated and the which it in turn regulates. The activation of NF-κB occurs through a highly ordered pathway whereby the activation of a large kinase complex, IκB kinase (IKK), leads to the phosphorylation of IκB, its subsequent ubiquitination, proteasome-mediated degradation, and the concurrent release and nuclear translocation of NF-κB. IKK is clearly a cornerstone in the signaling pathway leading to NF-κB activation.

A major missing link in the NF-κB signaling pathway is the mechanism that connects IKK to different upstream signals. It is generally believed that IKK may be activated by distinct upstream kinases in response to different stimuli. Indeed, a number of upstream kinases have been shown to activate IKK; these include members of the

MAP3 kinase family, such as MEKK1, MEKK3, NIK and TAK1. Further, the physiological role of these upstream kinases in mediating -specific IKK activation is becoming increasingly recognized. Connecting the MAP3Ks to the cell surface receptors, especially those belonging to the TNF receptor (TNFR) superfamily, are the adaptor TRAFs. TRAFs may activate the MAP3Ks via a novel signaling iv mechanism involving ubiquitination, which occurs through -63 linkages that do not target degradation but appear to specifically lead to signal activation. Emerging evidence suggests that TRAF-mediated ubiquitination is important for the activation of both IKK and the c-Jun N-terminal kinase (JNK) by various TNFR members, as well as certain other immune receptors, such as the toll-like receptors and IL-1 receptors. It is thus apparent that the ubiquitination activity of TRAFs must be under tight control.

Recent biochemical studies have identified a negative regulator of TRAF- ubiquitination. This factor, CYLD, was originally discovered as a tumor suppressor that is mutated in cylindromatosis, a predisposition to tumors of skin appendages. CYLD belongs to the deubiquitination enzyme (DUB) family, which digests chains.

At least in transfected cells, CYLD inhibited the ubiquitination of two major TRAF members, TRAF2 and TRAF6, as well as the ubiquitination of a downstream target,

IKKγ. Consistent with its DUB activity, CYLD negatively regulates the activation of

NF-κB by TRAFs and TNFRs. These in vitro studies provide an insight into the tumor suppressor function of CYLD and shed light on the signaling role of this DUB.

However, the physiological function of CYLD in the immune system and other biological systems remain unclear. Hence, the work described in this thesis was aimed at understanding the physiological role of CYLD. A number of unique findings are presented here:

1) CYLD is a non-redundant DUB of TRAF2. Although initial reports suggest

that overexpressed CYLD inhibits the ubiquitination of transfected TRAF2, it

has remained unclear whether CYLD indeed serves as a critical regulator of

TRAF2 under endogenous conditions. By RNA interference-(RNAi)- v

mediated CYLD knock down, we demonstrated that the loss of endogenous

CYLD resulted in constitutive ubiquitination of endogenous TRAF2. This

finding suggest that CYLD is a key DUB that controls the state of TRAF2

ubiquitination.

2) CYLD is a negative regulator of both IKK and JNK. Our CYLD knock down

studies reveal that the loss of endogenous CYLD caused hyperactivation of

not only IKK, but also JNK. Further, the ability of CYLD to regulate JNK

was specific for immune receptors and involved the upstream kinase MKK7,

but not MKK4.

3) The DUB function of CYLD is negatively regulated by its phosphorylation.

CYLD becomes transiently phosphorylated along with signal-induced

activation of JNK and IKK. We mapped the phosphorylation sites of CYLD

to seven . Interestingly, the phosphorylation of CYLD appears to

involve IKK, and this posttranslational modification caused inactivation of

CYLDs ability to deubiquitinate TRAF2. Consequently, CYLD mutants that

are defective in phosphorylation prevents signal-induced TRAF2

ubiquitination and a loss in the negative regulation of JNK and IKK. These

findings suggest a model where CYLD phosphorylation serves as a

mechanism to inactivate its DUB function, thus allowing transient activation

of TRAF2 ubiquitination and initiation of downstream signaling pathways.

4) Establishment of CYLD as an immune regulator. In order to understand the

physiological functions of CYLD, we generated CYLD knockout mice.

Although these mutant mice had no gross abnormalities in growth or survival, vi

they displayed clear defects in the immune system. First, the number of

peripheral T cells was significantly reduced in the spleens of CYLD -/- mice.

This defect was at least partially due to a defect in thymocyte development

from the double-positive to the single-positive stage, which was more

pronounced for CD4+ T cells. Second, CYLD -/- T cells were hyper-

responsive to TCR stimulation, producing markedly larger amounts of

cytokines than control T cells. These findings suggest that CYLD is a

negative regulator of T cell activation, where it is required for T cell

development in the thymus. Since it is the step of negative selection that

appears to be affected by the loss of CYLD, the CYLD deficiency may cause

abnormal TCR signaling, resulting in uncontrolled T cell deletion during this

stage of development.

In summary, the work presented here 1) defines a signaling function for CYLD in regulating TRAF ubiquitination and activation of NF-κB and JNK, 2) elucidates a mechanism of CYLD regulation, which involves its site specific phosphorylation and 3) demonstrates a role for CYLD in the regulation of T cell development and activation. These results provide significant insight into the physiological function of CYLD as well as the mechanism of its regulation. These findings also provide an example for how a DUB participates in signal regulation of the immune system. vii

TABLE OF CONTENTS

Page LIST OF FIGURES………………………………………………………………….. ix LIST OF ABBREVIATIONS………………………………………………………...x ACKNOWLEDGMENTS…………………………………………………………….xiii

CHAPTER I. LITERATURE REVIEW……………………………………………..1 1.1 The Two Branches of the Immune System…………………. 2 1.2 NF-κB Pathway Regulates Both the Innate and Adaptive Immune Functions………….. 5 1.3 NF-κB Family and Structure………………………………...5 1.4 Inhibitors of κB………………………………………….….. 8 1.5 Mechanism of NF-κB Activation……………………………10 1.6 The IκB kinase (IKK)………………………………………..14 1.7 Receptors Mediating Activation of IKK and NF-κB……….. 22 1.7.1 Toll-like Receptors…….……………………22 1.7.2 Antigen Receptors……...…………………...29 1.7.3 Tumor Necrosis Factor Receptor Superfamily……………………… 36 1.8 Tumor Necrosis Factor Receptor Associated Factors (TRAFs)…………………………….….. 40 1.9 Role of TRAF2s in Activation of JNK and IKK…………….42 1.10 The Role of TRAF6 in Activation of JNK and IKK………... 43 1.11 Ubiquitination and Deubiquitination……………….………..45 1.12 CYLD a Novel DUB Involved in Regulation of of Signal Transduction……………………………………… 47

CHAPTER II. NEGATIVE REGULATION OF JNK SIGNALING BY THE TUMOR SUPPRESSOR CYLD………………………..,.. 52 Abstract……………………………………………………………... 53 Introduction…………………………………………………………. 54 Materials and Methods……………………………………………… 56 Results………………………………………………………………. 58 Discussion…………………...……………………………………… 63 Acknowledgments………………………………………………...… 66

CHAPTER III. REGULATION OF THE CYLD BY IKKγ-DEPENDENT PHOSPHORYLATION …………………...………………………..76 Abstract………………………………………………………………77 Introduction……………………………………………………….… 78 Materials and Methods……………………………………………… 80 Results………………………………………………………………. 84 viii

Discussion…………………………………………………………... 93 Acknowledgments…………………………………………………... 97

CHAPTER IV. REGULATION OF T CELL DEVELOPMENT AND ACTIVATION BY THE DEUBIQUITINATION ENZYME CYLD………………………….……………………….. 116 Abstract……………………………………………………………... 117 Introduction……………………………………………………….… 118 Materials and Methods……………………………………………… 121 Results………………………………………………………………. 124 Discussion…………………………………………………………... 130 Acknowledgments…………………………………………………... 134

CHAPTER IV. OVERVIEW AND DISCUSSION……….………………………... 146 5.1 Overview of Major Findings…………………………………147 5.2 Discussion………………………………………………....…148 5.2.1 CYLD Phosphorylation: Involvement of IKK and Another Kinase…………….….. 148 5.2.2 Post-Translational Modification of CYLD: A Mechanism of Inhibition…...…………….150 5.2.3 CYLD in T Cells: A Positive or Negative Regulator………………..…..…153 5.2.4 CYLD Regulation of Peripheral T cell Activity……………………..……..… 154 5.2.5 CYLD Regulation of Macrophage and B cells in the Spleen………………………...155

CHAPTER IV. FUTURE DIRECTIONS………………….………………………...157 6.1 Regulation of Phosphorylation……………………………....158 6.2 Domains of CYLD………………………………………..… 159 6.3 In Vivo Physiological Role of CYLD in T Cell Development……………………………………… 160 6.4 In Vivo Physiological Role of CYLD in T cell Activation……………………………………….….162

BIBLIOGRAPHY……………………….………………….………………………...165 ix

LIST OF FIGURES

Figure Page 1.1 Structure of NF-κB family members………………………………………… 7 1.2 Structure of IκB family members……………………………………………. 9 1.3 Mechanism of IκB phosphorylation and degradation………………………...12 1.4 The canonical and noncanonical NF-κB pathways………………………….. 19 1.5 Signaling through Toll-like receptors………………………………………... 25 1.6 MyD88 dependent and independent pathways lead to Activation of early or late NF-κB activation……………………………….... 28 1.7 Antigen-mediated NF-κB activation in T cells……………………………….31 1.8 Signaling through TNFR……………………………………………………...39 2.1 Characterization of CYLD antibody and siRNA……………………………..68 2.2 CYLD knockdown by RNAi results in hyperactivation of JNK, but not IKK, in TNFα-stimulated cells……………………………...71 2.3 CYLD knockdown has no effect on JNK activation by a stress agent and promotes MKK7 activation by TNFα…………………….... 73 2.4 CYLD knockdown promotes activation of JNK by diverse stimuli and IKK by selective stimuli………………………………....75 3.1 CYLD knockdown results in constitutive ubiquitination of TRAF2………....99 3.2 CYLD is phosphorylated in response to diverse cellular stimuli……………. 101 3.3 CYLD phosphorylation is mediated by IKK……………………………….... 104 3.4 Site-specific phosphorylation of CYLD by IKK…………………………….. 106 3.5 418 of CYLD is phosphorylated in vivo……………………………... 108 3.6 CYLD phosphorylation is required for signal-induced TRAF2 ubiquitination and optimal JNK activation………………………………...…111 3.7 Signaling function of the phosphorylation-deficient mutant of CYLD……....113 3.8 Apoptosis induction by TNFα……………………………………………….. 115 4.1 targeting of CYLD…………………………………………………….. 136 4.2 Reduction in T cell numbers in CYLD -/- mice…………………………….. 138 4.3 Loss of Naïve T cells in the spleen of CYLD -/- mice………………………. 141 4.4 CYLD -/- mice display defects in thymocyte development…………………..143 4.5 T cell activation as measured by proliferation and cytokine production…….. 145 5.1 Signal induced phosphorylation and subsequent inactivation of The deubiquitinating activity of CYLD……………………………………….152 x

LIST OF ABBREVIATIONS

AICD Activation Induced Cell death AP-1 Activator Protein 1 APC Antigen Presenting Cells ASK1 Apoptosis Signal-Regulated Kinase BCR B Cell Receptors β-TrCP β Transducin Repeat Containing Protein BAFF B Cell Activating Factor Bcl-10 B-Cell Lymphoma 10 CARD Caspase Recruitment Domain CARMA1 Caspase Recruitment Domain Membrane-Associated Guanylate Kinase, MAGUK, Protein 1 CAP-Gly Cytoskeleton-Associated Protein-Glycine cDNA Complementary DNA cIAP Cellular Protein c-FLIP Fas-Assoicated Death Domain-Like Interleukin-1 β-Converting Enzyme Inhibitory Protein CFSE Carboxy-Fluorescein Diacetate, Succinimidyl Ester coIP Co-Immunoprecipitation DAG Diacyl Glycerol DC Dendritic Cells DD Death Domain DMEM Dulbecco’s Modified Eagle’s Medium DNA Deoxyribonucleic Acid DTT Dithiothreitol DUB Deubiquitinating Enzyme ECL Enhanced Chemiluminescence EMSA Electrophoresis Mobility Shift Assay ERK Extracellular Regulated Kinase FADD Fas Associated Death Domain GAPDH Glutaraldehyde-3-Phosphate Dehydrogenase GCK Germinal Center Kinase GCKR Germinal Center Kinase Receptor GEF Guanine Nucleotide Exchange Factor GRR Glycine Rich Region GST Glutathion S-Transferase HA Hemagglutinin HCl Hydrochloric Acid HEPES N-(2-hydroxyethyl) Piperazine-N’ –(2-ehanesulfonic acid) HRP Horseradish Peroxidase IB Immunoblot IFNγ Interferon γ Ig Immunoglobulin IκB Inhibitor of κB xi

IKK IκB Kinase IL-1 Interleukin-1 IL-1R Interleukin-1 Receptors IP Immunoprecipitation

IP3 Inositol 1,4,5-Triphosphate IRAK Interleukin-1-Receptor-Associated Kinase IRF3 IFN-Regulatory Factor 3 ITAM Immunoreceptor -Based Activation Motif JNK c-Jun N-Terminal Kinase kDa Kilodalton L Liter LAT Linker for Activation of T-Cells LFA1 Leukocyte Function-Associated Antigen 1 LN Lymph Node LPS Lipopolysaccharide LT Lymphotoxin MAGUK Membrane-Associated Guanylate Kinase MAP Mitogen Activation Protein MAP3K MAP kinase kinase kinase MEF Mouse Embryonic Fibroblast MEKK1 Mitogen-Activated Protein Kinase Kinase Kinase 1 MEKK3 Mitogen-Activated Protein Kinase Kinase Kinase 3 mL Milliliter mRNA Messenger RNA MyD88 Myeloid Differentiation Primary-Response Protein 88 MALT1 Mucosa-Associated Lymphoid Tissue Lymphoma Translocation Protein 1 MCP-1 Monocyte Chemoattractant Protein 1 MHC Major-Histocompatibility Complex MIP-1 Macrophage Inflammatory Protein 1 NEMO NF-κB Essential Modulator NES Nuclear Export Signal NFAT Nuclear Factor of Activated T-Cells NF-κB Nuclear Factor Kappa B NIK NF-κB Inducing Kinase NLS Nuclear Localization Sequence NP-40 Nonidet P-40 PEST Sequences Rich in P, E, D, S, and T PKC Protein Kinase C PLC-γ Phospholipase C-γ PMA Phorbol Myristic Acid PRR Pattern Recognition Receptors RAG Recombination-Activating Enzyme RHD Rel domain RING Really Interesting New Gene RIP1 Receptor-Interacting Protein 1 xii

RNA Ribonucleic Acid RNAi RNA Interference RT-PCR Reverse Transcriptase-Polymerase Chain Reaction SCF Skp1-Cullin1-F Box SDS Sodium Dodecyl Sulfate SFK Src-Family Kinase SH3 Src-Homology 3 siRNA Small Inhibitory RNA SLP-76 SH2-Domain-Containing Leukocyte Protein of 76-kDa SMAC Supra-Molecular Activation Complex TAK1 TGF-β Activating Kinase TCR T Cell Receptors

TH1 T Helper 1 TH2 T Helper 2 TIR Toll/IL TIRAP TIR-Associated Protein TLR Toll-Like Receptors TNFα Tumor Necrosis Factor α TNFR Tumor Necrosis Factor Receptor -1R TRADD TNFR Associated Death Domain TRAF Tumor Necrosis Factor Receptor Associated Factor TRAM Toll-Receptor-Associated Molecule TRIF Toll-Receptor-Associated Activator of Interferon UBC Ubiquitin Conjugating Enzyme UBP Ubiquitin Processing Protease UCH Ubiquitin C-Terminal Hydrolases µCi Microcurie µg Microgram µL Microliter ZAP-70 Zeta-Chain-Associated Protein 70 xiii

ACKNOWLEDGMENTS

While writing this thesis, I could not help but realize that this great accomplishment was not made alone. Without the support of those who walk endlessly in the shadows, I would have not been able to achieve all that I have. Therefore, I wish to thank all those who have helped me to reach this point, and only hope that I have repaid their help in some form.

I first wish to humbly thank my advisor, Dr. Sun. It has been a seemingly endless road that we have taken together over the past five years. I cannot begin to express my deep gratitude for his willingness to give me the opportunity to explore new ideas and theories. Dr. Sun has a passion for science that he passes onto his students. His door was always open to me and no matter what data I presented to him, he always could come up with four more needed to be done. He has furthered my passion for science and given me the tools that can only continue throughout my career. Further, his wife, Dr. Mingying

Zhang was a wonderful collaborator. Without her endless support and work on the

CYLD project, many of the wonderful discoveries would not have been made.

Secondly, I wish to thank my committee members with who I have had the pleasure of many fruitful discussions. Over the course of the past five years, you have helped me to think intricately and to examine my results more critically. With their encouragement and criticism of my work, I have become a better scientist.

Next, I have to thank the past and present members of the Sun lab, especially Drs.

Edward Harhaj, Gutain Xiao, and Michael Waterfield, fellow students, Geetha Babu,

Mikyoung Chang, Wei Jin, Ato Wright, and Xeufeng Wu, and our wonderful technician

Mandy Losiewicz. The friendship that we have had over the past five years has always xiv kept me sane. I am thankful for your continual support, help and fruitful discussion. It has been an honor and a privilege to work, talk, and drink with all of you. I thank you all for being my teachers and mentors. Our friendships will never be forgotten.

I also wanted to give my deepest thanks to the kitchen and cleaning staff of the

Microbiology and Immunology department. Without their help, the work over the past five years would never have been done as quickly. In addition, I need to thank the secretaries in the Microbiology and Immunology department, without whom the paperwork and administrative duties would still be sitting on my desk. We are all extremely lucky to have these people here, allowing the work that we do to be preformed that much easier.

Finally, I need to thank my family. First, I need to thank my parents, who struggled with a little boy and helped him to become a man. These past few years have caused me to not be able to visit nearly as much as I should have. I love you and will never forget the sacrifices that you took to raise me as well as you did. Above all, I have to thank my wife, Brandee, and children, Camden and Kaylen. You have sacrificed the most over the course of graduate school. I can never really repay you for your dedication and support. I love you and thank you for always being there at my highs and my lows.

Without you, I do not think I could have ever gotten through my graduate work. I owe all my accomplishments and successes to you. 1

CHAPTER I

LITERATURE REVIEW 2

1.1 The Two Branches of the Immune System

During the course of , the immune system has developed mechanisms to combat the dangers of infectious organisms. Yet, the same evolutionary process that drives the selection for a stronger immune system drives selection for stronger or

“smarter” pathogens. To combat these pathogens, the immune system has developed many different strategies that are classified into two areas of defense: innate immunity and adaptive immunity. Innate immunity is truly an ancient mechanism of host defense, as many components of this mechanism are found in plants and lower animals, suggesting that these defense mechanisms arose before the kingdoms split [1]. Adaptive immunity is a relatively recent addition to evolution as it exists only in vertebrates.

Innate immunity provides the earliest line of defense that protects the host through preexisting immune components [2]. The cellular components of the innate immune system are nonlymphoid white blood cells, most importantly macrophages and neutrophils, as well as the epithelial cells and mucosal cells of the gut and lungs that form a physical barrier to prevent entry by most microbes. Finally innate immunity includes a newly classified set of molecules called “intrinsic immunity” within most cells that provides potent protection from viral infections [3]. Microbial detection by innate immunity is mainly mediated by a limited number of germline-encoded receptors known as pattern recognition receptors (PRR) [4]. The PRR have evolved to recognize metabolic products of pathogens that allow for the immune system to distinguish self from non-self. These innate immune receptors can be expressed on the cell surface, in intracellular compartments, or secreted into the extracellular environment [5]. Functions of PRR include activation of proinflammatory signaling pathways, phagocytosis, 3 opsonization, and induction of apoptosis. A prototypical family of the PRR are the Toll- like receptors (TLR). To date, 10 members in human, and 12 members in mice, have been identified, with each differing in their specificities and expression patterns, as well as potential target genes that are induced [6]. The first mammalian TLR identified and the one most extensively studied is TLR4. Engagement of TLR4 by its ligand, lipopolysaccharide (LPS), on macrophages, leads to activation of the macrophages, including the release of potent chemokines and cytokines allowing for the recruitment of other immune cells to the site of inflammation [6]. Accumulating evidence suggests that the innate immune recognition not only triggers the first line of defense, but is also a prerequisite for induction of adaptive immunity, especially in the development of T

helper 1 (TH-1) responses [7].

In contrast to the immediate response of innate immune cells, adaptive immunity arises three to five days after being alerted by the innate immune system. The central players of adaptive immunity are the lymphocytes: T cells and B cells, which have the power to detect diverse foreign substances or antigens. However, since the prevalence of antigen-specific lymphocytes is extremely low, clonal expansion and differentiation of these cells are essential and time-consuming steps of the adaptive immune response. The function of lymphocytes relies on their surface antigen receptors (T-cell receptors or TCR and B-cell receptors or BCR), which are encoded by somatically rearranged genes.

These adaptive immune receptors, unlike those on innate immune cells, have tremendous diversity and specificity in antigen recognition, allowing the immune system to detect virtually all existing antigens. The antigen receptors on T and B cells presumably evolved from non-rearranging immunogolubulin (Ig) superfamily members by insertion 4 of a recombination-activating enzyme (RAG) -based transposon [8] [9]. It is the somatic gene rearrangement that provides the adaptive immune system the power to recognize massive numbers of foreign antigens and to generate specific immunological memory.

Yet this wonderful diversity is also responsible for terrible diseases and immunological problems, such as autoimmunity, allergies, and rejection of tissue grafts [10] [11].

Therefore, considering the molecular mechanism underlying lymphocyte function is instrumental for rational design of more effective strategies to prevent these unwanted immune reactions.

In addition to lymphocytes, adaptive immunity also requires accessory cells, most importantly the antigen presenting cells (APCs), that are specialized to take up foreign substances and present the antigens to T lymphocytes. It is generally believed that dendritic cells (DC) are the primary APCs for initiating an adaptive immune response, although macrophages and B cells may also serve as APCs. The DCs reside in peripheral tissues. Upon encountering microbes, they engulf the microbes or microbial proteins via endocytosis and micropinocytosis. Consequently, the DCs are activated and undergo a maturation that allows them to efficiently degrade, process, and present antigens on their surface major histocompatibility complex (MHC) molecules and to provide costimulatory signals to the target T cells. The microbial activation of DCs also triggers their migration to the local lymphoid organs, where they can engage the naïve T cells. MHC-peptide complexes on the APC cell surface are recognized by a specific TCR leading to activation of the T cell. The antigen-activated T cells undergo a series of changes resulting in proliferation, acquisition of effector function, and modulation of various cell surface markers and eventually clearance of the pathogen. Elimination of the pathogen 5 can be directly accomplished through T cell-mediated cytotoxic effect or through helper functions leading to a humoral response.

1.2 NF-κB Signal Transduction Pathway Regulates Both the Innate and Adaptive

Immune Functions

As with other biological processes, the immune responses are tightly regulated by signal transduction events. Signal transduction programs cells to efficiently respond to environmental changes, such as stress and microbial infections that are detected by innate or adaptive immune receptors. Upon binding to their ligands, the immune receptors deliver a specific signal to the intracellular environment that leads to activation of multiple downstream signaling pathways. Although the signal pathways vary among different immune receptors, in all cases the consequence of signal transduction is the activation of transcription factors that mediate and other biochemical events required for specific cellular functions. One family of transcription factors that is critically involved in immune cell activation is nuclear factor kappa B (NF-κB).

Originally identified as a transcription factor regulating Ig kappa gene expression [12],

NF-κB is now known to regulate diverse axes of innate and adaptive immune responses.

Additionally, the NF-κB factors are also involved in the development and maturation of lymphocytes as well as the formation of certain lymphoid organs [13-15]. The mechanism of NF-κB activation has been extensively studied, although the upstream signaling steps that connect different immune receptors to NF-κB remain poorly defined.

1.3 NF-κB Family and Structure

NF-κB represents a family of structurally related and evolutionarily conserved proteins with five known family members in mammals: c-Rel, RelA (also known as p65), 6

RelB, NF-κB1 (p50/p105), and NF-κB2 (p52/p100) [16] (Figure 1.1). The NF-κB family members exist as homo- and heterodimers and are divided into two structural classes.

Class I is composed of NF-κB1 and NF-κB2, both of which are synthesized as larger precursor molecules (p105 and p100, respectively) that are either post-translationally or co-translationally processed to yield mature proteins (p50 and p52 respectively) [17-19].

The mature NF-κB1 (p50) and NF-κB2 (p52) proteins contain a DNA binding domain, but these molecules lack a transcriptional activation domain that is required for recruitment of other coactivators to the sites of transcription. In contrast to the class I

NF-κB family members, the class II family members c-Rel, RelA, and RelB are translated as mature proteins. Additionally, the class II family members contain both a

DNA binding domain and a transcriptional activation domain that is necessary for the

NF-κB dimers to transcribe their target genes.

A structural characteristic of the NF-κB family members is that they contain in a 300 amino acid region known as the Rel homology domain

(RHD). Work performed on the crystal structure of RHD has yielded much information about the RHD and its important role in NF-κB. The crystal structure of NF-κB dimers shows that the RHD is composed of two immunoglobulin-like domains connected by a flexible linker [20]. Furthermore, the C-terminal folds of the Ig domains of the RHD are responsible for dimerization between the NF-κB family members [21]. DNA contact is made through both Ig-domains within the RHD, however sequence specific interaction comes from loops in the N-terminal folds [22]. A nuclear localization sequence (NLS) is present at the C-terminal end in the RHD of all NF-κB family members, and is responsible for the nuclear import of NF-κB proteins. 7

Figure 1.1 Structure of NF-κB family members 8

1.4 Inhibitors of κB

Early biochemical studies revealed that the NF-κB family members are sequestered in the cytoplasm of cells and that a rapid translocation into the nucleus occurs upon cellular stimulation with various agents. This insight led to the discovery of a family of inhibitory proteins named inhibitor of κB (IκB) which complex with NF-κB dimers allowing for inhibition through cytoplasmic sequestration [23].

The binding of IκB to NF-κB masks the nuclear localization signal of NF-κB, thereby causing cytoplasmic retention of NF-κB and separating it from acting on the κB enhancer elements [24-26]. The IκB family includes seven known members: IκBα,

IκBβ, IκBθ, IκBε, Bcl-3, as well as the precursors of NF-κB1 (p105) and NF-κB2 (p100)

(Figure 1.2). A hallmark of the IκB proteins is their possession of six to seven ankyrin repeats, a structural domain required for their association with NF-κBs [26]. Differences in the number of ankyrin repeats appear to influence the specificity of an IκB for a given

NF-κB dimer [16]. Of the IκB proteins, IκBα and IκBβ were the first to be isolated and to date remain the most well studied.

IκBα is a 37 kDa molecule that consists of three domains: a 70 amino acid N- terminal region, a 205 amino acid ankyrin repeat domain and a 42 amino acid C-terminal domain containing PEST sequences (sequences rich in amino acids P, E, D, S, and T)

[27]. Deletional analysis of IκBα revealed that the N- or C-terminal domains are dispensable for NF-κB binding, whereas the ankyrin repeat domain is critical for IκBα’s ability to bind NF-κB dimers. The ankyrin repeat domain is also required for masking the nuclear localization signal (NLS) of NF-κB. In this regard, X-ray structural studies of co-crystallized IκBα with the prototypical NF-κB, p50-p65 heterodimer shows that 9

Figure 1.2 Structure of IκB family members. 10

IκBα forms a curved α-helical stack. The first two ankyrin repeats bind specifically to the C-terminal Ig-folds of the RHD near the NLS [26, 28]. Therefore, it is believed that in this tethered state the amino acids immediately preceding the NLS of p65 undergo a conformational change which appears to sterically hinder the binding of the nuclear import machinery [28]. Interestingly, within the IκBα/NF-κB(p50/p65) complex, the

NLS of p65 but not that of p50 is masked by IκBα. The exposed NLS of p50 allows the inactive NF-κB complex to inefficiently move into the nucleus. However, due to the possession of a strong nuclear export signal (NES) in IκBα, the steady state location of the IκB/NF-κB complex is cytoplasmic. Though the co-crystallized structure of

IκBα/NF-κB(p50-p65) yields important insights into how nuclear exclusion is maintained, it does not completely address how IκBα selectively masks the NLS of p65 and regulates the NLS of p50.

1.5 Mechanism of NF-κB Activation

NF-κB is activated by an extremely wide variety of stimuli, including antigens that mediate B and T cell activation, microbial products such as bacterial LPS, viral proteins, proinflammatory cytokines such as tumor necrosis factor α (TNFα) and interleukin-1 (IL-1), mitogens, as well as physical and chemical stresses. Though controversy surrounds exactly how these various stimuli activate NF-κB, particularly how the signaling pathways triggered by the different stimuli converge, a common theme does emerge. Upon stimulation with extracellular or intracellular stimuli, degradation of the prototypical IκB (IκBα) by the 26S proteasome occurs within a few minutes. NF-κB is subsequently transported from the cytoplasm to the nucleus of the cell allowing for transcription of various target genes. 11

The molecular mechanism of signal-induced IκB degradation has been well defined with IκBα (Figure 1.3). In response to cellular stimulations, IκBα rapidly becomes phosphorylated on two serines in the N terminus, serines 32 and 36 [29].

Mutation of these serines to alanines blocks the inducible phosphorylation and subsequent degradation of IκBα and prevents the release of NF-κB [30]. It is now clear that the phosphorylation of IκBα serves as a molecular trigger for its ubiquitination.

Upon phosphorylation, IκBα is recognized by an E3 termed F- box/WD40 E3RSIκB / β-TrCP [31, 32], which results in the polyubiquitination of IκBα primarily at 21 and 22 [33]. The process of ubiquitination is initiated when ubiquitin is activated by a ubiquitin-activating enzyme (E1). Activation involves the hydrolysis of ATP to form an E1-ubiquitin thioester. Next, the activated ubiquitin is transferred to a member of the ubiquitin-conjugating enzyme (E2) family. Finally, through the substrate specific ubiquitin-protein ligase (E3), ubiquitin is conjugated to the target protein by the formation of an isopeptide bond between the carboxyl terminus of ubiquitin and the ε-amino group of a lysine residue on the target protein [34]. Strong evidence demonstrates that phosphorylation of IκBα is required for its ubiquitination, since mutations of serines 32 and 36 to alanines abolishes inducible IκBα ubiquitination, whereas mutation of these serines to a phospho-mimetic residue, glutamic acid, leads to constitutive ubiquitination of IκBα. The requirement of ubiquitination of IκBα for its degradation has also been demonstrated by mutations of the ubiquitin conjugation sites, lysines 21 and 22. Mutation of these amino acids leads to a loss of both ubiquitination and degradation [33] yet not a loss of inducible phosophorylation. To date, there has 12

Figure 1.3 Mechanism of IκB phosphorylation and degradation. Numerous immune stimuli induce activation of upstream kinases, leading to IKKα and IKKβ phosphorylation and activation a trimeric IκB kinase (IKK) complex. Activated IKK then phosphorylates IκB, ubiquitination and subsequent degradation by 26S proteasome. 13 been no clear explanation as to how NF-κB is released from IκBα. Nevertheless, it seems that the ubiquitinated IκBα remains associated with NF-κB, since NF-κB cannot dissociate from IκBα or translocate to the nucleus when proteasome inhibitors block

IκBα degradation.

An interesting feature of NF-κB activation is the involvement of a negative- feedback mechanism [35]. The promoter region of the IκBα gene contains NF-κB binding elements [36] allowing for rapid and strong transcription upon NF-κB activation.

Thus, after its depletion, the intracellular pool of IκBα is quickly replenished through

NF-κB-mediated IκBα gene induction [37, 38]. The resynthesized IκBα not only prevents further nuclear import of the cytoplasmic NF-κB, but also enters the nucleus and stops the nuclear function of already activated NF-κB. Once in the nucleus, IκBα binds to NF-κB and removes NF-κB from the DNA, through an undefined mechanism, and the inactivated IκB/NF-κB complex is then exported back to the cytoplasm via the NES of

IκBα. Hence, IκBα not only serves to sequester NF-κB in resting cells, but also functions in the feedback loop that ensures a rapid but transient activation of NF-κB and induction of NF-κB responsive genes.

Though IκBα is the prototypical member of the IκB family, an extremely complex pattern of NF-κB regulation is starting to emerge. The rapid degradation of

IκBα and its resynthesis contribute to the transient activation of NF-κB by certain stimuli such as the proinflammatory cytokine, TNF-α. Yet, other agents, such as LPS, cause prolonged NF-κB activation despite the resynthesis of IκBα [39]. This persistent NF-κB activation is due to degradation of another IκB family member, IκBβ. 14

IκBβ is a 46 kDa molecule that is less well characterized than IκBα. Despite its structural homology with IκBα, IκBβ differs from IκBα in both NF-κB binding specificity and signal responses. Unlike IκBα, which targets the p50-p65 heterodimer,

IκBβ preferentially interacts with c-Rel containing NF-κB complexes [39]. Further IκBβ degradation is selectively induced by certain stimuli, such as IL-1 and LPS, but not

TNFα. In contrast to the rapid resynthesis of IκBα, the degration of IκBβ is persistent, due to the lack of dependence of IκBβ gene expression on NF-κB. It has thus become dogma that IκBβ regulates the more prolonged NF-κB activation, whereas IκBα is responsible for the more transient NF-κB activation.

1.6 The IκB Kinase (IKK)

As it became increasingly clear that a key step in NF-κB activation is the phosphorylation of IκBs, especially IκBα, much effort went into identifying the cytokine-induced protein kinase, deemed IκB kinase (IKK). Though many different kinases were suggested, only one fit the criteria of IκBα phosphorylation. IKKs kinetics of activation by various inducers maintained a preference for serines over threonines, and had the ability to phosphorylate both IκBα sites simultaneously [40]. IKK was isolated as a multi-subunit complex that specifically phosphorylates serines 32 and 36 [40]. The core IKK holoenzyme is composed of three proteins, including two catalytic subunits,

IKKα and IKKβ [41, 42], and a regulatory subunit termed IKKγ (also named NEMO)

[43-45].

IKKγ is a 46 kDa protein that lacks a catalytic domain, yet IKKγ is required for signal-dependant activation of the IKK catalytic subunits, especially IKKβ [46].

Functionally, the N-terminal half of IKKγ is responsible for its binding to IKKβ and its 15

C-terminal half is necessary for signal-dependant regulation of IKK activity. Although precisely how IKKγ regulates the activation of IKK holoenzyme remains poorly understood, it is generally believed that IKKγ is responsible for recruiting the IKK complex to upstream signaling events leading to IKK activation.

The catalytic subunits, IKKα and IKKβ, are 85 and 87 kDa proteins, respectively, with high amino acid sequence similarity. Both kinase subunits have a catalytic domain that is similar to other known serine-threonine kinases. Activation of IKK occurs through phosphorylation of two conserved serine residues located within the activation loop or “T loop” regulatory domain (serines 176 and 180 for IKKα and serines 177 and 181 for

IKKβ). It is believed that the T loop phosphorylation causes a conformational change in the IKK proteins resulting in activation of their catalytic activity [47, 48].

It is unclear, however, whether the phosphorylation of IKKα and IKKβ is mediated by upstream kinases or through their autophosphorylation. IKK can be physically recruited to various cellular receptors, such as TNFR, TLR, and TCR [49]. It is possible that the accumulation of IKK in the receptor complexes, mostly in the lipid rafts of plasma membrane, triggers its autophosphorylation and activation. Alternatively,

IKK may be phosphorylated by upstream kinases that are recruited to the receptor complexes. At least in vitro, IKKα and IKKβ serve as substrates for certain upstream kinases, such as NF-κB inducing kinase (NIK) and MEKK1 [50]. Interestingly, a recent study demonstrates that upon TCR crosslinking both NIK and the IKK holoenzyme are recruited to the lipid rafts in T cells and that IKK is phosphorylated by NIK [51].

Clearly, more studies are required to understand the mechanism of IKK activation downstream of different immune receptors. 16

Given the potent function of the IKK/NF-κB pathway in various cellular processes, such as cell growth, survival and inflammatory responses, the activity of IKK must be subject to negative regulation in unstimulated cells. Indeed, IKK is negatively regulated by two protein phosphatases, PP2A and PP1. Due to the constitutive activity of

PP2A and PP1, IKK is normally maintained in an unphosphorylated and inactive state.

Inhibition of PP2A and PP1 by the inhibitor okadaic acid leads to the autophosphorylation and activation of IKK, even within unstimulated cells [40].

Presently, the exact mechanism that leads to IKK phosphorylation is not known.

However, this finding suggests that the activity of IKK is controlled through the equilibrium between its phosphorylated and dephosphorylated states. It is likely that the receptor signals induce IKK phosphorylation by interfering with the activation of PP2A and PP1, or by stimulating the activity of an unidentified IKK kinase or IKK transphosphorylation [47].

The essential role of IKK in NF-κB activation has been confirmed through experiments using knock-out mice. Disruption of the Ikkβ or Ikkγ locus results in mid- embryonic lethality at day E13.5-E15.5, due to extensive hepatic apoptosis [52-54].

Examination of hepatocytes prior to day E11.5 shows a complete ablation of NF-κB activation by nuclear staining or signal-induced activation. Also, IKKβ-deficient mouse embryonic fibroblasts (MEFs) are completely defective in NF-κB activation induced by various stimuli, such as TNF-α, IL-1, dsRNA, and single-stranded DNA [53-56]. Like fetal liver cells, the IKKβ -/- and IKKγ -/- deficient MEFs are sensitive to apoptosis induction by TNFα [56]. Interestingly, the embryonic lethality of the IKKβ knock-out mice is at least partially due to TNFα-induced apoptosis, since this developmental defect 17 was largely overcome when the IKKβ knock-out mice were crossed to mutant mice deficient in the TNF-α or TNF receptor 1 (TNFR1) genes [56, 57]. Although these double knock-out mice overcome the embryonic lethality and develop normally, they ultimately fail to thrive and die within two to three weeks after birth due to severe opportunistic infections from normal flora [56].

The above findings underscore the importance of the NF-κB signaling pathway in immune function. The role of IKKβ in immune function has been further studied by the adoptive transfer of fetal liver stem cells from IKKβ knock-out mice into lethally irradiated wild-type mice. Within the irradiated host, the transferred fetal liver stem cells are able to develop and reconstitute the immune system, unless the stem cells carry severe genetic defects in immune cell development. Therefore, this adoptive transfer strategy allows the study of IKKβ’s function in both the development and activation of immune cells. Through the fetal liver cell transfer studies, it was shown that the IKKβ -/- radiation chimeras lack T cells and display thymic atrophy unless the IKKβ -/- fetal liver cells were co-transferred with wild-type bone marrow. Co-transfer studies revealed the

IKKβ knock-out T cells develop normally, however, they are defective in NF-κB activation and lack proliferative ability in response to T cell receptor signaling [58].

Disruption of the Ikkα locus does not result in embryonic lethality but mice ultimately die within a day after birth. These mice display severe morphological abnormalities, mostly in epidermal differentiation [59-61]. These mice have a shiny, taut, sticky skin without whiskers [61]. They display severe craniofacial defects with a lack of limb outgrowth [59]. Histological analysis of the epidermis revealed that of the basal, spinous, granular and cornified layers, IKKα knock-out mice have an extremely thick 18 basal proliferating layer with no block in differentiation [59]. Yet, within the basal layer, degradation of IκB and nuclear localization of NF-κB were not observed. Surprisingly, cells derived from IKKα knock-out mice display normal IKK and NF-κB activation in response to proinflammatory cytokines [61]. IKKα’s role in the immune system was revealed through the adoptive transfer of IKKα knock-out hematopoietic stem cells into lethally irradiated wild-type mice. The IKKα deficiency is associated with a marked reduction in B cell maturation and germinal center formation in response to T cell- dependent antigens [13].

The above mouse models not only outline a critical role for IKK in the activation of NF-κB, but also underscore the importance NF-κB in regulating normal development and the function of the immune system. Of particular interest is the finding that IKKα and IKKβ possess extremely different in vivo biological functions. In concert with these functional studies, emerging evidence suggest that IKKα and IKKβ differ in response to cellular stimuli. IKKβ is activated by most of the known NF-κB inducers, such as proinflammatory cytokines, T cell mitogens and microbial products [35, 47]. The activation of IKKβ is required for the rapid phosphorylation and degradation of IκBα as well as nuclear translocation of the prototypical forms of NF-κB. As such, the IKKβ- dependent NF-κB activation is referred to as the canonical NF-κB pathway (Figure 1.4).

Since the activation of IKKβ requires IKKγ, the canonical pathway of NF-κB activation relies on both IKKβ and IKKγ. In contrast to IKKβ, IKKα does not respond to the canonical NF-κB stimuli, but is activated rather by very select inducers such as LPS, lymphotoxin (LT), B cell activating factor (BAFF) and CD40 ligand [35, 47, 62]. Such activation rarely leads to strong phosphorylation and complete degradation of IκBα. 19

Figure 1.4 The canonical and noncanonical NF-κB pathways. Receptor-specific stimulation results in the activation of the canonical IKK complex or a noncanonical

NIK-IKKα complex, leading to IκB degradation and p100 processing, respectfully. 20

These inducers characteristically induce slow and more persistent activation of NF-κB.

Additionally, the IKKα-mediated signaling leads to the activation of specific NF-κB members, predominantly the p52/RelB dimmer. This pathway is thus named the noncanonical NF-κB pathway [63, 64] (Figure 1.4). It is now clear that the noncanonical pathway of NF-κB activation is independent of IKKβ and IKKγ or the degradation of

IκBα. Rather, this novel NF-κB pathway relies on the inducible processing of NF-κB2 precursor protein, p100. Before processing, p100 binds to and sequesters RelB in the cytoplasm. Upon its inducible processing, p100 loses its C-terminal IκB-like sequence and is converted to the mature NF-κB2 form, p52. The generated p52/RelB dimer then moves to the nucleus to exert its function. The RelB/p52 heterodimer is important for regulating chemokine gene expression in lymphoid stromal cells [65] and probably also regulates the expression of apoptosis inhibitor genes in B cells. Consistently, the noncanonical NF-κB pathway is required for the development of secondary lymphoid organs, B cell maturation and function, as well as germinal center formation [63]. These findings highlight the critical, but different roles that IKKβ and IKKα play in regulating the canonical and non-canonical pathways, adding to the complexity of NF-κB regulation and activation.

Although the core region of the canonical IKK complex is composed of two catalytic and one regulatory subunits, it is obvious that other proteins are also associated, although more weakly, with the IKK complex. Under native conditions, IKK exists in a super protein complex with a molecular mass between 700-900 kDa [41, 66]. Whereas several proteins have been suggested to be components of the 700-900 kDa complex [67,

68] [66] [42], biochemical evidence is lacking. By protein purification techniques, a 21 novel IKK-associated protein ELKS has recently been identified [69].

Coimmunoprecipitation assay confirms that ELKS is complexed with all three core components of IKK in both nontreated and treated HeLa cells [69]. ELKS was shown to be a regulatory subunit, like IKKγ, that functions by recruiting IκBα to the IKK complex.

Whether ELKS also recruits other proteins to the IKK complex has yet to be determined.

The identification of ELKS clearly provides an explanation for the large molecular mass of IKK complex. However, the sum of the masses of the four proteins (ELKS with IKK subunits) is still much smaller than 700-900 kDa. Although it is possible that the IKK complex contains additional proteins, one other possibility is that the IKK subunits are multimeric. Indeed, chemical crosslinking studies show that IKKγ exists as a dimer or trimer [44]. Since IKKα and IKKβ also form homo- or heterodimers [44, 45], it is logical to assume that self-association of the IKK subunits accounts for the formation of most, if not all, of the 700 - 900 kDa complex.

Despite the extensive studies of IKK, many questions still linger. One of these questions concerns the mechanism mediating the assembly of canonical and noncanonical

IKK complexes. The canonical IKK was isolated as a complex of IKKγ, IKKα, and

IKKβ. Since IKKα is dispensable (at least in gene knock-out studies) for the function of canonical IKK, it is unclear what role IKKα normally plays within the IKK complex. It also remains to be investigated whether IKKγ/IKKβ only complexes exists in IKKα expressing cells. Nevertheless, IKKα appears to be in a subcomplex that does not contain

IKKγ or IKKβ [70]. It is believed that the IKKα-specific complex mediates p100 phosphorylation and processing in the noncanonical NF-κB signaling pathway. Another remaining question is how cellular stimulation by the different inducers leads to 22 activation of the canonical or noncanonical pathway. The answer to this question relies on the characterization of upstream signaling molecules that connect the different IKK complexes to cell surface receptors. Recent findings, some of which were made during my thesis research, raise yet another important question of whether IKK regulates any other biochemical events besides the degradation of IκB. As will be discussed in Chapter

3, strong evidence suggests that IKK phosphorylates other proteins and mediates functions other than NF-κB activation. Understanding these remaining questions will not only provide new insights into IKK regulation, but may also uncover novel biological roles of IKK.

1.7 Receptors Mediating Activation of IKK and NF-κB

Numerous cellular receptors target activation of the IKK/NF-κB signaling pathway. The following discussion will summarize the major immune receptors that activate NF-κB.

1.7.1 Toll-Like Receptors

Toll-like receptors (TLR) are a family of evolutionarily conserved PRRs that recognize general molecular patterns present in microbial pathogens. A major downstream target of the TLR signal is NF-κB. Additionally, TLR also mediate activation of several families of MAP kinases, including JNK, ERK, and p38. These signaling pathways function cooperatively in the induction of various cytokine genes, many of which mediate early innate immune function, as well as regulate the development of adaptive immune responses.

The TLRs are type 1 integral membrane glycoproteins that contain considerable homology in the cytoplasmic region with interleukin-1 receptors (IL-1Rs) [71]. The 23 major region of homology between TLRs and IL-1Rs is the Toll/IL-1R (TIR) domain

[72], which is composed of approximately 200 amino acids forming three highly conserved regions that are necessary for their signaling function. Though these regions are necessary for the binding to downstream molecules, other sequences within the cytoplasmic tail are required for the propagation of the signaling cascade as demonstrated in C3H/HeJ mice. C3H/HeJ mice harbor a mutation in the Tlr4 locus within the cytoplasmic domain that does not disrupt the TIR domain, but completely ablates TLR4 signaling [73].

In contrast to their cytoplasmic domains, the extracellular domains of the TLRs and IL-1R are extremely diverse, with the former being composed of leucine-rich repeat motifs and the latter composed of Ig domains [71]. The number and size of the leucine- rich repeats varies among different TLR members, allowing them to recognize distinct sets of ligands. Diversification also occurs with the subcellular localization of different

TLR correlating with the location of their respective ligands [74-76].

Binding of various ligands to their specific TLR leads to dimerization and recruitment of downstream TIR domain adaptor proteins. These include the adaptor molecule myeloid differentiation primary-response protein 88 (MyD88), TIR-associated protein (TIRAP), Toll-receptor-associated activator of interferon (TRIF), and Toll- receptor-associated molecule (TRAM). MyD88 was the first to be identified as an adapter protein in IL-1R signaling that allowed for the recruitment of interleukin-1- receptor-associated kinase (IRAK), a serine-threonine kinase [77-79]. MyD88 contains two domains, an N-terminal death domain (DD) and a separate C-terminal TIR domain

[80]. Upon binding of TLRs to their specific ligand, MyD88 is recruited to the receptor 24 and forms homodimers through TIR-domain-TIR-domain interactions. Though different adaptors exist and are preferentially utilized by different TLRs, MyD88 is required for all

TLRs with the exception of TLR3 and partial TLR4 signaling (described below) [81, 82].

The activation of MyD88-dependant (TLR2 and TLR4) signaling requires another adaptor protein, TIRAP [83, 84]. A clear role for TIRAP has not been elucidated, but it is believed that TIRAP and MyD88 bind to the TLR at non-overlapping sites and also associate with each other, allowing for the formation of a heterotetrameric complex

(Figure 1.5, complex 1) [85]. Such a high order complex may be important for activating the downstream signaling proteins [85].

The DD of MyD88 is inverted in its interaction with downstream factors. One downstream factor recruited to the TLR complex by MyD88 is IRAK4, a DD-containing serine/threonine kinase required for TLR signaling (Figure 1.5, complex 2) [65]. IRAK4 aggregation within the MyD88-TLR complex is believed to cause its autoactivation and catalytic activation. The receptor-recruited and activated IRAK4 can by itself mediate moderate activation of NF-κB, through downstream signaling effectors that have not been defined. Upon aggregation of MyD88-TLR complexes, another IRAK family member, IRAK1, is quickly recruited (Figure 1.5, complex 3). IRAK4 functions as an activator of IRAK1 activity through phosphorylating residues in the C-terminus [86, 87].

Activated IRAK1 is believed to transphosphorylate residues in its own N-terminus, allowing for the recruitment of another protein, tumor necrosis factor receptor associated factor 6 (TRAF6) (Figure 1.5, complex 4). Of interest, it was reported that the intrinsic kinase activity of IRAK1 is dispensable for signaling [88], though the study failed to determine if IRAK4 could phosphorylate the residues in the N- and C-terminus as a 25 26 compensatory mechanism.

Like most other TRAF family members, TRAF6 contains two domains: a C- terminal domain that mediates self association and interacts with upstream signaling molecules and a N-terminal really interesting new gene (RING) finger domain. The

RING finger domain of TRAF6 contains E3 ubiquitin-ligase activity [89, 90]. Binding of

TRAF6 causes dissociation of the IRAK1/TRAF6 complex from the receptor and allows it to move into a specific compartment in the plasma membrane, possibly through undefined adaptor molecules (Figure 1.5, complex 5).

The IRAK1/TRAF6 complex becomes the supporting lattice for the formation of a signalsome that mediates IKK activation. In addition to IRAK1 and TRAF6, this signalsome contains TGF-β activating kinase (TAK1) [91, 92], two adaptor proteins

TAB1 [93] and TAB2 [94] or TAB3 [95], and two ubiquitin-conjugating enzymes Ubc13

[89] and Uev1A [90]. Once IRAK1/TRAF6 is released from the receptor, the TRAF6 E3 ubiquitin ligase becomes activated [89, 90], most likely through TRAF6 oligomerization and self ubiquitination (Figure 1.5, complex 5). TAK1 and TAB1 are then recruited to the IRAK1/TRAF6 complexes through the adaptor molecule TAB2 or TAB3, which are potentially redundant mediators of TAK1 activation [95]. Once recruited to the complex,

TAK1 and TAB2/TAB3 become phosphorylated by an unknown kinase leading to the dissociation of the IRAK1/TRAF6/TAK1/TAB1/TAB2-3 complex from the plasma membrane and migration into the cytosol (Figure 1.5, complex 6). In the cytosol, TAK1 undergoes K63-linked polyubiquitination, which is mediated by the E3 activity of

TRAF6 and the E2 enzymes Ubc13 and Uev1A. Furthermore, efficient TAK1 kinase activity is dependent on the adaptor TAB1, although how this adaptor enhances the 27 activity of TAK1 is not understood [93]. A missing piece of information in our understanding appears after the activation of TAK1. Although, TAK1 is essential for

IKK activation by TLR, as well as IL-1R, the underlying mechanism of TAK1-mediated

IKK activation is poorly defined [96].

The adaptor protein MyD88 was initially believed to be critical for all signaling events of TLR, as gene locus disruption of MyD88 leads to the complete ablation of

TLR-mediated cytokine production [71, 82]. Yet, a closer analysis of the MyD88- deficient mice has revealed the existence of MyD88-dependent and independent pathways. The MyD88-independent pathway occurs through two TLR members, TLR3 and TLR4, and is characterized by delayed kinetics of NF-κB activation (Figure 1.6) [97,

98]. TLR4 targets both MyD88-dependant and independent pathways whereas TLR3 solely induces the MyD88-independent pathway. In searching for adaptors mediating the

MyD88-independent pathway, two TIR-domain-containing proteins were identified,

TRIF [99, 100] and TRAM [101, 102]. TLR3 directly binds to TRIF [103], whereas

TLR4 binds to TRAM, which in turn binds to TRIF (Figure 1.6) [101, 104]. TRIF is required for the MyD88-independent activation of NF-κB. Additionally, this new TLR adaptor is also responsible for TLR3- and TLR4-induced activation of the IFN-regulatory factor 3 (IRF3) [105], a pathway that leads to induction of IFN-β gene expression [105].

The TRIF dependant activation of NF-κB does not go through the IRAK family of proteins but may act through direct interactions with the TRAF6/TAK1 complex [106,

107]. A recent report also suggests that TRIF interacts with a new kinase, called receptor-interacting protein 1 (RIP1) [108], and that this interaction in turn activates IKK 28

Figure 1.6 MyD88 dependent and independent pathways lead to the activation of early or late NF-κB activation. Activation of MyD88 independent pathway involves the adaptors TRAM and TRIF and leads to activation of IRF3 and delayed activation of NF-κB. 29

and NF-κB through a yet unidentified mechanism. The TLRs have gained much popularity over the past ten years and numerous adaptor molecules have been identified.

However, the exact mechanism of IKK activation by TLR signaling still remains poorly understood.

1.7.2 Antigen Receptors

Adaptive immunity utilizes a mechanism whereby somatic mutation and gene rearrangement allow for the production of a diverse receptor repertoire that has the potential to recognize a wide range of pathogens. The adaptive immune receptors include the TCR and BCR, which are multiprotein complexes composed of highly diverse antigen binding subunits and invariant signaling chains. The antigen-binding subunits of

TCR include α and β chains, which forms a dimer that recognizes antigenic peptides presented in the context of MHC molecules on the cell surface of APC [109]. The TCR

α/β dimer by itself can not transduce signals to the T cell, but does so through its associated signaling chains, including CD3 chains and ζ chains (Figure 1.7, complex 1)

[109]. Analogous to the TCR, the BCR complex is composed of a cell-surface immunoglobulin, which mediates specific recognition of antigens, and invariant signaling chains Igα and Igβ, that are responsible for transmembrane signaling [109]. However, unlike the MHC-restricted manner of TCR, BCR recognizes the epitopes in intact antigens. Despite this major difference in antigen recognition, the intracellular signaling events for these two antigen receptors are remarkably similar, with both involving the receptor-proximal activation of protein tyrosine kinases, recruitment of various adaptor molecules, and signal propagation leading to the activation of downstream factors, 30 including IKK and NF-κB. Hereafter, I will concentrate on the TCR signaling event that regulates IKK activity and NF-κB activation.

TCR signaling is initiated through its engagement by a specific antigen/MHC complex on APCs. The T cell/APC interaction also involves the coreceptor, CD4 or

CD8, which is recruited to the TCR complex via binding to MHC class II and MHC class

I molecules, respectively [109]. These initial molecular interactions trigger rapid signaling events that allow T cells to undergo a physical change resulting in relocation of

TCR and coreceptors to a “capping” structure [110]. The “capping” structure is composed of multiple copies of TCRs and coreceptors as well as various other surface molecules, such as leukocyte function-associated antigen 1 (LFA1), CD45, and costimulatory molecules like CD28, ICOS, 4-1BB, and CD27 [111]. These “capping” structures are organized into distinct areas within the interface and are known as supra- molecular activation complexes (SMACs). Although the regulation of SMAC formation is still not well understood, recent studies suggest the involvement of the reorganization of actin and intracellular microtubule structures.

The first detectable signaling events that occur upon TCR ligation are the activation of two Src-family kinases (SFKs), and Fyn. Lck and Fyn are 56 and 59 kDa proteins, respectively, that contain several domains in common: an N-terminal site for myristilation, a unique region, a Src-homology 3 (SH3) domain, a SH2 domain, a domain, and a C-terminal negative regulatory domain [112]. Lck is associated with the coreceptors CD4 and CD8 through a dicysteine motif in the unique region [113, 114]. Fyn largely colocalizes within mitotic structures of the cell [112].

Upon receptor engagement Fyn is recruited into lipid rafts, becoming activated by 31

Figure 1.7 Antigen-mediated NF-κB activation in T cells. Stimulation of T cells via the peptide-MHC complex induces the recruitment of adaptor proteins and protein tyrosine kinases, leading to downstream signaling events including the activation of NF-

κB, NFAT, and JNK. 32 33 unknown mechanisms. Activated Fyn then phosphorylates the CD3 invariant chains at their immunoreceptor tyrosine-based activation motifs (ITAMs) (Figure 1.7, complex 1).

Concurrently, CD4-Lck or CD8-Lck molecules are recruited to the TCR complex, where

Lck becomes activated via its transphosphorylation. The activated Lck then phosphorylates the ITAMs within the TCR invariant chains (Figure 1.7, complex 1)

[115]. Sufficient phosphorylation of the ITAMs leads to the recruitment of a Syc-family protein tyrosine kinase, zeta-chain-associated protein 70 (ZAP-70), via interaction between the SH2 domain of ZAP-70 and phosphorylated ITAMs. Lck then proceeds to phosphorylate and activate ZAP-70 (Figure 1.7, complex 2).

Biochemical studies show that the activation of ZAP-70, Lck, and Fyn occurs for

15 to 30 minutes after TCR ligation [116]. Yet, by 30 minutes the immunological synapse has not reached full maturity as visualized by “capping” and formation of

SMACs [117]. This finding suggests that the prolonged synapse formation may not facilitate the TCR signaling. However, the sustained synapse appears to be critical for subsequent signaling events, since disruption of synapse formation leads to the rapid break up of SMACs, leading to markedly reduced IL-2 production and T-cell proliferation [118]. Therefore, both early signaling and prolonged synapse formation are necessary for full effector and proliferative potential in T cells.

ZAP-70 is a central player in amplifying the TCR signal. Upon its phosphorylation within the TCR complex by Lck, the activated ZAP-70 influences downstream signaling events by phosphorylating a number of targets, including the linker for activation of T-cells (LAT) and SH2-domain-containing leukocyte protein of 76-kDa

(SLP-76) (Figure 1.7, complex 3). SLP-76 phosphorylation leads to the membrane 34 recruitment of a small amount of a guanine nucleotide exchange factor (GEF) and Vav

[119], which leads to rapid tyrosine phosphorylation [119, 120]. The Vav phosphorylation allows it to form a stable complex with SLP-76 (Figure 1.7, complex 4)

[121, 122]. This signaling step is tightly controlled by a mechanism that requires both the TCR and the CD28 costimulatory signals in naïve T cells. Though the exact mechanism of Vav–mediated NF-κB activation is not well defined, two recent works have started to unveil this mystery. The involvement of Vav through interactions with a small GTP binding protein, Rac, is necessary for the recruitment of PKCθ into the lipid rafts (Figure 1.7, complex 5) [123], which in turn is crucial for NF-κB activation [124].

In a separate study, Vav was shown to constitutively interact with IKKα. This interaction allows for membrane targeting of IKKα upon T cell costimulation by CD28

[125], although it is unclear whether this biochemical event contributes to the activation of NF-κB.

LAT is a plasma membrane protein anchored via palmitoylated cysteine residues

[126, 127]. ZAP-70-mediated LAT phosphorylation creates binding sites through its phosphorylated tyrosine residues, which then recruits an array of signaling molecules, including other adaptors and PTKs from the Src, Syk, and Tec families [128-130]. This dynamic complex then recruits and activates phospholipase C-γ (PLC-γ), a key enzyme involved in generation of the calcium and PKC signals (Figure 1.7, complex 4). Upon activation, PLC-γ cleaves membrane-bound phosphatidylinositol-(4,5)-bisphosphate to critical secondary messengers, diacyl glycerol (DAG) and inositol 1,4,5-triphosphate

(IP3) (Figure 1.7, complex 5) [131, 132]. IP3 mediates calcium mobilization from the intracellular stores in the endoplasmic reticulum resulting in activation of the 35 transcription factor, nuclear factor of activated T-cells (NFAT). DAG then mediates activation of a number of protein kinase C (PKC) members, with PKCθ being most critical for TCR-mediated activation of NF-κB (Figure 1.5, complex 5) [124, 133, 134].

Over the last two years, significant progress has been made in understanding how

PKCθ mediates the activation of NF-κB. Upon activation, PKCθ induces the assembly of a membrane-associated signaling complex, composed of scaffolding proteins termed caspase recruitment domain, CARD, membrane-associated guanylate kinase, MAGUK, protein 1 (CARMA1), B-cell lymphoma 10 (Bcl-10) and mucosa-associated lymphoid tissue lymphoma translocation protein 1 (MALT1) (Figure 1.7, complex 6). Knock-out mice models of CARMA1, Bcl-10, and MALT1 have completely ablated NF-κB activation in response to TCR ligation [135-139]. Protein interaction experiments demonstrate that CARMA1 and Bcl-10 preferentially interact via their CARD domain

[140-142] and that the CARMA1/Bcl-10 complex interacts with MALT1, primarily through Bcl-10 [135]. Activation of T cells by TCR ligation allows for the recruitment of

CARMA1, a protein that is constitutively associated with lipid rafts [140, 143], to the

TCR complex in a PKC-dependent manner. The CARMA1/Bcl-10/MALT1 complex associates with two ubiquitin-conjugating enzymes Ubc13 and Uev1A. This complex has been shown to catalyze the lysine-63 polyubiquitination of IKKγ, which in turn leads to the activation of the IKK complex through an undefined mechanism (Figure 1.7, complex

7 and 8) [144, 145]. A recent study suggests that MALT1 functions as the E3 ubiquitin ligase that directs the IKKγ ubiquitination. However, this model has been challenged by another study, which suggests that MALT1 has no intrinsic-ubiquitin ligase activity, but rather functions though association with a well-characterized ubiquitin ligase, TRAF6 36

[144, 145]. Whether TRAF6 is an essential component of TCR-mediated IKK activation remains largely elusive, although the work performed with Jurkat cell lines suggests this possibility.

1.7.3 Tumor Necrosis Factor Receptor Superfamily

The members of the TNFR superfamily play pivotal roles in numerous biological events such as survival, differentiation, as well as inflammatory responses. TNFRs exert their biological functions through triggering a number of downstream signaling pathways, including those leading to activation of NF-κB and JNK. The TNFR family members display a 25-30% sequence similarity, yet most of these homologous residues are responsible for their ability to trimerize upon receptor-ligand ligation [146].

Consistent with their diversity in ligand binding, TNFRs have little sequence homology in their extracellular domains. TNFR family members also vary, to a large extent, in their expression patterns with regard to cell, tissue types, and induciblity. Additionally, these receptors are diverse in the structure of their intracellular signaling domains, thus providing another level of diversity.

TNFRs can be classified into three groups based on their signaling properties, cytoplasmic sequences, and adaptor interactions [147]. The first group is composed of

TNFR1, DR3, DR6, TRAIL-R1, and TRAIL-R2. Ligand engagement of this group of

TNFRs leads to the recruitment of specific DD-containing adaptors, including Fas- associated death domain (FADD) and TNFR-associated death domain (TRADD). The consequence of such signaling events is the activation of the caspase cascade, leading to programmed cell death or apoptosis. The second group of TNFRs is composed of

TRAIL-R3, TRAIL-R4, and decoy-R3. These molecules typically have short 37 cytoplasmic tails and cannot provide intracellular signals, yet they effectively compete for ligand with the first group of TNFR molecules and therefore functionally inhibit signal transduction pathways leading to apoptosis. The third group of this receptor family includes TNFR2, CD40, CD30, CD27, Ox40, 4-1BB, HVEM, BAFF-R, BCMA,

TACI, LTβR, RANK, TROY, EDAR, XEDAR, and RELT. These receptors contain one or more TRAF-interacting motifs (TIMs) that allow for direct recruitment of the TRAF family of proteins, which in turn leads to the activation of NF-κB and different MAP

Kinases: JNK, p38, and ERK.

The ligands for TNFRs are from another superfamily of proteins, the TNFs.

These ligands typically exist in two forms: (1) those that are membrane-bound such as

TNFα, LTα/β, CD-40L, 4-1BBL, LIGHT, CD70, and OX40L, and (2) those that are secreted, such as TNFα, LTα, and BAFF. Both the soluble and membrane-bound forms of TNF family members form trimers. This property allows these ligands to induce the signaling functions of TNFR, involving the recruitment of TRAF proteins, which can be mediated through two mechanisms. The first mechanism involves the binding of adaptor proteins to the cytoplasmic tails of those TNFR members that contain DD sequences.

The second mechanism involves the direct association of TRAFs with the cytoplasmic tail of TNFRs, via specific TRAF-binding motifs.

The signaling mechanism of TNFRs has been extensively studied using TNFR1, one of the two receptors for TNF-α. Binding of soluble TNFα to TNFR1 leads to the trimerization and movement of the receptor complex into lipid rafts [148]. Upon movement into the lipid rafts, TNFR1 binds to TRADD and forms a scaffolding structure that is responsible for recruitment of other signaling proteins such as RIP1 and TRAF2 38

(Figure 1.8). Via association with RIP1 and TRAF2, the IKK complex is also recruited to the lipid rafts, where it is activated [149] through a yet unidentified mechanism. Based on the findings on IKK activation by TCR and TLRs, it has been postulated that lysine-63 linked polyubiquitination of IKKγ may play a role in regulating the activation of IKK by

TNFR stimulation. Indeed, the ligation of TNFR1 leads to the ubiquitination of IKKγ

[150]. Therefore, an emerging model is that upon recruitment to TNFR, TRAF2 becomes self ubiquitinated in the form of self-lysine-63-linked polyubiquitin chains [151]. The activated TRAF2 also mediates RIP1 lysine-63 polyubiquitination [152], which in turn allows for recruitment of IKKγ into the lipid rafts [149]. This recruitment then positions

IKKγ in close proximity to TRAF2, allowing for lysine-63 linked polyubiquitination.

Despite these compelling findings, direct evidence for the involvement of TRAF ubiquitination in TNFR1-mediated NF-κB activation is lacking. In fact, this theory has been challenged by a recent study showing that the TRAF2 ubiquitination is dispensable for TNF-α induced activation of NF-κB but rather plays a role in JNK activation [153].

Clearly, more work is required to assess the mechanism of NF-κB activation by TNFR1, as well as by other TNFR members.

Current data has shown that TRAF2 and TRADD interactions with the TNFR have much greater affinities than TRAF2 has for peptide motifs within TNFR [154].

Hence the activation and recruitment of TRAF2 by TNFR1 and TRADD is much stronger than by TNFR2, which contains peptide binding motifs for the recruitment of

TRAF2. This mechanism of higher affinities is necessary to suppress TNFR1-induced apoptosis in conjunction with strong activation of NF-κB and gene products, c-IAPs

[155, 156] and FLIPL [157]. The growing body of data has demonstrated that the 39 40

downstream activation of IKK requires TRAF molecules and that specific domains in different TRAF molecules are necessary for their activation.

1.8 Tumor Necrosis Factor Receptor-Associated Factors (TRAFs)

TRAF molecules are a highly conserved family of proteins found in mammals,

Drosophila [158], and even [159]. This family of proteins is considered to be critical signal transducers of TNFRs, as well as additional receptors of the innate and adaptive immune systems. Six TRAF members exist to date, each of which contain similar structural motifs. The defining domain that all TRAF molecules share in common is the presence of a TRAF domain at the C-terminus. This domain consists of two subregions consisting of a coiled-coil domain, also called TRAF-N, and a conserved TRAF-C domain [160]. The TRAF domain is critical for mediating homo- and heterotrimerization of TRAF molecules [160-162], as well as interactions of TRAFs with receptors and signaling proteins [163, 164]. With the exception of TRAF1, all

TRAF members contain a N-terminal RING finger domain followed by a zinc finger repeat domain [163, 165]. The ring finger motif, is similar to those found in various E3 ubiquitin ligases, and its functional significance will be discussed latter. As demonstrated by mutational analysis, the zinc finger repeats appear to be important in the activation of the NF-κB and JNK signaling pathways [164, 165].

In unstimulated cells, TRAF homo- or heterotrimers are predominantly dispersed in various subcellular locations within the cytoplasm of the cell [166, 167]. On activation of the cells with various inducers such as TNF-α, CD40, and LPS, TRAF molecules are recruited to the receptors via interactions with adaptor proteins or through direct interactions. The capacity of TRAF-trimers associating with their receptors is low in 41 unactivated cells [154]. Yet, upon ligand-induced receptor trimerization, the TRAF- receptor interaction is greatly enhanced due to the stronger trimer-trimer interactions.

Although the mechanism mediating the rapid recruitment of TRAFs to the membranes has not been defined, mutational studies suggest that a sequence motif PXQXT (where X can be any amino acid), is present in various TNFRs and serves as a consensus binding site for TRAF2, TRAF3 and TRAF5, [168, 169]. Another sequence motif, basic-

QXPXEX-acidic, is responsible for interactions with TRAF6 [170, 171]. The interactions of TRAFs with TNFR occur through amino acids located in the interface between the TRAF-N and TRAF-C domains, and these amino acids are highly conserved between TRAF2, TRAF3, and TRAF5 [161]. The receptor-TRAF binding leads to the formation of a signalsome in the lipid rafts, which has been demonstrated with the TNFR member, CD40 [167]. Hence, the cellular localization of TRAF molecules prior, during, and after receptor activation could very well yield insights into the regulatory mechanisms of TRAF-mediated signal transduction.

As alluded to earlier, receptor recruitment of TRAF molecules, especially TRAF2 and TRAF6, triggers their polyubiquitination. Polyubiquitination of proteins usually occurs through a lysine-48 type conjugation which involves diester bond formation between lysine 48 of one ubiquitin with the C-terminal glycine of another ubiquitin. This type of ubiquitin chains is recognized by the 26S proteasome, and thus targets proteins for degradation [172]. Interestingly, the ubiquitination of TRAF2 and TRAF6 occurs through the non-classical lysine-63 linked form. This form of ubiquitination is not linked to protein degradation, but mediates protein kinase activation [90], DNA repair, and vesicle trafficking [34]. Interestingly, the lysine-63 polyubiquitination of TRAF2 and 42

TRAF6 is believed to occur through self ubiquitination [173, 174]. This function of

TRAF2 and TRAF6 requires their RING finger domain as well as the ubiquitin- conjugating enzymes Ubc13 and Uev1A [173, 174]. Prior to cellular stimulation, TRAF2 and TRAF6 are not ubiquitinated. The signal-induced TRAF ubiquitination is likely triggered through its oligomerization and accumulation in the lipid rafts.

1.9 Role of TRAF2s in Activation of JNK and IKK

TRAF2 plays an important role in the activation of JNK and IKK by TNFRs, although this function varies among the different TNFR members. TRAF2 gene knockout studies demonstrate that this TRAF member is essential for the activation of both IKK and JNK downstream of CD40. Although necessary for activation of JNK,

TRAF2 by itself is not required for the activation of IKK/NF-κB induced by TNFα (a ligand of TNFR1 and TNFR2), as demonstrated in TRAF2 knock-out mice [175].

However the role of TRAF2 in TNFα-induced NF-κB activation appears to be due to the functional redundancy between TRAF2 and TRAF5, since both the JNK and NF-κB pathways are severely attenuated in cells deficient in both TRAF2 and TRAF5 [176].

How TRAF2 connects the TNFR signal to JNK and IKK is not well understood.

TRAF2 has been shown to interact with certain upstream kinases, such as MEKK1 [177,

178] and a member of germinal center kinase (GCK), GCKR [177, 179]. Both GCKR and MEKK1 have been shown to activate downstream signaling pathways leading to

JNK and NF-κB. Interestingly, the kinase activity of GCKR is dispensable for its activation of MEKK1, leading to the possibility that GCKR acts by recruiting MEKK1 to

TNFR which may trigger the autophosphorylation and activation of MEKK1. In this model, it is believed that the ubiquitination of TRAF2 is required for the receptor 43 recruitment of GCKR and therefore also MEKK1. Whereas the GCKR/MEKK1/TRAF2 pathway plays an important role in TNFR signaling, this pathway appears to occur only in certain cell types, such as embryonic stem cells, but not in macrophages or mouse embryo fibroblasts [180]. It has also been demonstrated that TRAF2 mediates activation of two different MAP kinase kinase kinases (MAP3Ks), apoptosis signal- regulated kinase (ASK1) [181] and TAK1 [92]. Unfortunately, the role of these

MAP3Ks in NF-κB activation has not been genetically confirmed, although strong evidence suggests their clear role in activation of JNK. Recently, another MAP3K,

MEKK3, has become a promising candidate for mediating NF-κB activation. Disruption of the MEKK3 locus in mice leads to severe defects in NF-κB activation in mouse embryo fibroblasts, stimulated with TNFα and IL-1β [182]. This finding establishes

MEKK3 as an essential upstream kinase for activation at least by certain inducers. It remains to be examined, however, whether MEKK3 is a mediator of TRAF2-induced

NF-κB activation.

1.10 The Role of TRAF6 in Activation of JNK and IKK

TRAF6 is activated in the same manner as TRAF2, although these different

TRAFs can be targeted to different receptors. Upon receptor-ligand binding the receptor becomes trimerized, in the case of TNFR, and allows for the efficient recruitment of

TRAF6 homo- and heterodimers. Recruitment of the TRAF6 molecules leads to lysine-

63 polyubiquitination of TRAF6 through self-ubiquitination [90, 174]. Compared to

TRAF2, TRAF6 has been shown more definitively to function as an E3 ubiquitin ligase.

In the presence of the E2 complex Ubc13/Uev1A, TRAF6 induces the synthesis of lysine-63 polyubiquitin chains in a RING finger dependent manner [89]. Further, the E3 44 activity of TRAF6 is not only specific for its self ubiquitination, but also necessary for lysine-63 polyubiquitination of specific target proteins, including TAK1 [89, 90, 174] and

IKKγ [145]. However, conclusive evidence for the requirement of TRAF6 in TCR signaling is lacking. In fact, the TRAF6 knock-out mice have no impaired T cell response to TCR signaling. Nevertheless, TRAF6 ubiquitination provides a novel mechanism of IKK activation, which is involved in NF-κB signaling downstream of IL-

1R/TLRs.

The mechanism by which TRAF6 ubiquitination mediates IKK activation has been studied using an in vitro model system. It has been shown that the ubiquitination of

TRAF6 serves as a molecular trigger for recruiting an IKK activating kinase, TAK1.

This step of signaling also involves the TAB1 and TAB2/3 adaptors, which mediate

TAK1 recruitment to TRAF6 [90]. The TRAF6/TAK1 complex formation is thought to lead to the oligomerization and, potentially, autoactivation of TAK1. Although precisely how TAK1 is activated and how TAK1 mediates IKK activation remain poorly understood, in vitro studies suggest that both TAK1 and the regulatory subunit IKKγ become ubiquitinated in the TRAF6 complex. The ubiquitination of TAK1 may trigger its catalytic activation, allowing this IKK kinase to phosphorylate IKKβ, triggering IKK activation. Since ubiquitination of IKKγ is important for IKK activation, it is conceivable that IKK activation by TRAF6 may involve both the TAK1-mediated phosphorylation of

IKKβ and IKKγ ubiquitination. A key question to be addressed though is how ubiquitination causes kinase activation. 45

1.11 Ubiquitination and Deubiquitination

Ubiquitination is best known as a protein modification event that targets proteins for degradation in the 26S proteasome [183]. However, recent studies have revealed additional functions for a cellular process, including endocytosis and signal transduction

[184-186]. Ubiquitination is catalyzed by the coordinated action of three types of enzymes: ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzymes (E2), and ubiquitin ligases (E3). To date, only a single E1, about 40 E2s, and over 500 E3s have been identified. The existence of a high number of E3s is consistent with their function governing the substrate specificity of ubiquitination. Emerging evidence suggests that each E3 regulates the ubiquitination of only a small number of proteins, whereas each E2 can be involved in the ubiquitination of multiple proteins.

Once an ubiquitin molecule is attached to a target protein, polyubiquitination chains are formed through an isopeptide bond between the C-terminal glycine residue of one ubiquitin and a lysine residue (usually lysine 48 or 63) of another. The lysine 48-linked polyubiquitination mediates protein degradation, whereas the lysine 63-linked ubiquitination is involved in endocytosis or signal transduction [187-189]. Additionally, monoubiquitination of proteins has been identified and shown to mediate additional cellular processes. Ubiquitination is a reversible process, involving ubiquitination and deubiquitination of proteins, it resembles the process of protein phosphorylation. The deubiquitination is mediated by a family of cysteine proteases, known as deubiquitinating enzymes (DUBs), which are specialized in their digestion of polyubiquitin chains [151,

190]. 46

DUBs are classified into two groups based on sequence homology: ubiquitin C- terminal hydrolases (UCH) and ubiquitin processing proteases (UBPs). UCH molecules contain a 230 amino acid core catalytic domain and typically cleave ubiquitin derivatives of small or short ubiquitin chains [191]. UBPs are a group of proteins that vary in size and for their core catalytic domain contain thiol proteases that are thought to be responsible for the removal of polyubiquitinated proteins. The catalytic domains of both

UCH and UBPs contain highly conserved amino acid residues, including cysteine, histidine, and aspartic acids, which are critical for their functions [192].

The overexpression of UBPs typically lowers the overall cellular levels of ubiquitinated proteins. However this observation raises the question of how the specificity of UBPs is obtained physiologically [193, 194]. UBPs are thought to have the ability to lower ubiquitinated protein levels, thus serving to either recycle or edit the effects of ubiquitinated proteins. The editing function of UBPs has been demonstrated to reverse the role of ubiquitin in proteolysis, endocytosis [195], and transcription factor function [196].

An interesting question arises of how UCHs or UBPs remove the polyubiquitin chains. These enzymes could shorten the polyubiquitin chain from its distal end by cleaving one ubiquitin molecule at a time or cleave internally to release polymeric or oligomeric ubiquitin chains. The former mechanism has been demonstrated for a UCH member, UCH37, which catalyzes the depolymerization of polyubiquitinated chains from the distal end [197]. This finding is somewhat surprising, since UCHs are thought to catalyze deubiquitination of short ubiquitinated chains instead of polyubiquitin.

Therefore, the function of DUBs can not always be predicted through sequence 47 comparisons. Another question regarding the mechanism of DUB function is how these enzymes recognize their specific substrates. One potential mechanism is that the different DUBs can directly recognize their substrates. This mechanism appears to be the case for CYLD, a DUB known to target deubiquitination of TRAF2 (see discussion below). Another possibility is that specific adaptor proteins recruit the substrate to

DUBs. This latter mechanism is analogous to the function of certain ubiquitin ligases that are complexed with substrate-binding proteins.

The regulation of DUBs has been extensively studied. Although different mechanisms may be involved, transcriptional regulation is by far the most important one.

The expression of different DUBs can be regulated in tissue-specific [198, 199] or developmental manners [200, 201], or even controlled by hormones [202, 203]. Post transcriptional mechanisms of DUB regulation have also been demonstrated, especially through the work of this thesis research. It seems clear that phosphorylation serves as a rapid and efficient mechanism to modulate the function of certain DUBs, as will be discussed in the following section.

1.12 CYLD a Novel DUB Involved in Regulation of Signal Transduction

Recently, a new DUB family member, CYLD, has been identified by different groups based on its sequence homology with DUBs and through functional DUB screening. CYLD has domains that are homologous to both UCHs and UBPs. At least in transfected cells, this new DUB inhibits the polyubiquitination of TRAF2, TRAF6, as well as IKKγ [204-206]. The connection between CYLD and IKKγ was first suggested by yeast-two hybrid studies, which identified CYLD as a protein that interacts with the effector domain of IKKγ. Further studies confirmed this interaction of CYLD with IKKγ 48 within eukaryotic cells. CYLD was also shown to inhibit NF-κB activation in transfected cells. The acronym CYLD came from its original identification as a tumor suppressor associated with familial cylindromatosis. Familial cylindromatosis, also known as turban tumor syndrome, or Brooke-Spiegler syndrome, is an autosomal dominant genetic predisposition to multiple tumors of the skin appendages [207]. The cylindromatosis patients carry germ-line mutations in one CYLD allele, but the other allele is lost in tumor cells [207, 210]. This so called loss of heterozygosity is also common to many other human tumors, especially those of epithelial and mesenchymal tissue [274]. The tight association of these patient studies provide little insight into the function of CYLD in normal cells, since patients carry a wildtype CYLD allele in all but the tumor cells

[207]. To date, the exact nature and origin of the tumor cells are not known, but their morphological appearance resembles that of eccrine or apocrine cells [208, 209]. The location of the predisposed gene for familial cylindromatosis, hence designated CYLD, is on 16q12-13 [210]. The cylindromatosis patients carry germ-line mutations in one of the CYLD alleles. The tumor development involves loss of heterozygosity in the wild-type CYLD allele, which usually appears in the second to third decade of life.

The tumor size and number are highly variable among patients. Even though patients may have very large tumor mass, malignant change with distant metastasis is unusual.

By sequence analysis, CYLD is predicted to contain four structural domains, 3 copies of a cytoskeleton-associated protein-glycine (CAP-Gly) domain, 2 copies of a -rich repeat, a metal-binding finger-like domain, and a UCH domain. The conserved CAP-Gly domain is known to be important for some proteins to act as a linker between endocytic vesicles and microtubules [211]. Recently, an NMR study preformed 49 on the third CAP-Gly domain demonstrated that the tertiary structure of this domain resembles a SH3 domain, a known signaling domain that binds to proline-rich motifs

[212]. The NMR analysis also suggested that the third CAP-Gly domain of CYLD mediates interaction with IKKγ via the proline-rich sequences of IKKγ. However, it is unclear whether the third CAP-Gly domain of CYLD is indeed required for IKKγ interaction. Proline-rich motifs are also known to mediate protein/protein interactions.

As mentioned above, proteins with proline-rich motifs may be recognized by SH3- containing proteins [213]. To date, the proline-rich motifs (or regions) of CYLD have not been defined.

The best studied domain within CYLD is the predicted deubiquitinating domain.

This domain of CYLD contains most of the conserved amino acid residues shared by different DUBs. Mutation of the conserved residues abolishes the ability of CYLD to deubiquitinate TRAF2, TRAF6, as well as IKKγ in vitro [204-206].

Coimmunoprecipitation studies suggest that CYLD physically interacts with not only

IKKγ but also with TRAF2 and probably with TRAF6, as well. Yet further work suggests that CYLD does not directly bind TRAF2 but requires an adaptor protein, TRIP, for recruitment to TRAF2 [214]. It has been suggested that CYLD preferentially deubiquitinates lysine-63-linked polyubiquitin chains, although it does display enzymatic activity for lysine-48-linked polyubiquitin chains [205]. The deubiquitination activity could be abolished in vitro with mutations in either the C-terminus of CYLD [204], portions or entire regions of the two UCH domains, or through point mutations of predicted catalytic histidine residues [205] or conserved cysteine residues [206]. Patients with cylindromatosis consistently carry CYLD gene mutations that usually lead to 50 truncations within the C-terminal UCH domains [207, 210]. Transient transfection studies reveal that the C-terminal truncations of CYLD abrogate the ability of CYLD to deubiquitinate TRAF2, TRAF6, and IKKγ. Consistent with the role of TRAF/IKKγ ubiquitination in IKK activation, overexpression of CYLD was also shown to negatively regulate NF-κB activation in reporter-gene assays. Together, these findings provide insight into the mechanism of tumor formation in cylindromatosis. However, the role of

CYLD in regulating TRAF ubiquitination and NF-κB activation needs to be confirmed using genetic approaches. Additionally, it remains to be investigated whether CYLD regulates other signaling pathways. A large part of my thesis work was targeted to fill this gap.

In summary, since the discovery of IKK, an extensive amount of knowledge about how NF-κB and IKK are regulated has emerged. However, a comprehensive understanding of this complex signaling pathway has yet to be achieved. Among the missing links are the upstream adaptor proteins and kinases that connect IKK to specific cell surface receptors. Additionally, how the IKK signaling machinery is properly controlled remains largely elusive. Given the important role of the IKK pathway in cell growth and survival, as well as inflammation, the existence of negative-regulatory mechanisms seems to be crucial to prevent its abnormal activation. The DUB molecule

CYLD appears to serve as a key negative regulator of IKK, although it remains to be investigated whether this regulation system exists under physiological conditions.

Whether the function of CYLD is specific to receptor signals also needs to be examined.

Finally, the study of IKK/NF-κB signals has now entered an era that allows investigation using physiological systems. Studies using knock-out mice will yield additional insights 51 into the physiological functions and mechanism of regulation of the IKK/NF-κB signaling network. 52

CHAPTER II

Negative Regulation of JNK Signaling by the Tumor

Suppressor CYLD

William Reiley, Minying Zhang, and Shao-Cong Sun

J. Biol. Chem. (2004) 279(53):55161-7 53

ABSTRACT

CYLD is a tumor suppressor that is mutated in familial cylindromatosis, an autosomal dominant predisposition to multiple tumors of the skin appendages. Recent studies suggest that transfected CYLD has deubiquitinating enzyme activity and inhibits the activation of transcription factor NF-κB. However, the role of endogenous CYLD in regulating remains poorly defined. Here we report a critical role for CYLD in negatively regulating the c-Jun N-terminal kinase (JNK). CYLD knockdown by RNA interference results in hyper activation of JNK by diverse immune stimuli, including

TNF-α, interleukin-1, lipopolysaccharide, and an agonistic anti-CD40 antibody. The

JNK-inhibitory function of CYLD appears to be specific for immune receptors, since the

CYLD knockdown has no significant effect on stress-induced JNK activation.

Consistently, CYLD negatively regulates the activation of MKK7, an upstream kinase known to mediate JNK activation by immune stimuli. We further demonstrate that

CYLD also negatively regulates IκB kinase (IKK), although this function of CYLD is seen in a receptor-dependent manner. These findings identify the JNK signaling pathway as a major downstream target of CYLD and suggest a receptor-dependent role of CYLD in regulating the IKK pathway. 54

INTRODUCTION

CYLD was originally identified as a tumor suppressor that is mutated in familial cylindromatosis [207], an autosomal dominant predisposition to multiple tumors of the skin appendages [215, 216]. Recent studies reveal that CYLD is a new member of the deubiquitinating enzyme (DUB) family [204-207]. Transient transfection studies suggest that CYLD inhibits the ubiquitination of certain signaling molecules, including members of the tumor necrosis factor receptor (TNFR)-associated factor (TRAF) family [204, 205,

214]. TRAFs are known as signaling adaptors of TNFR superfamily [217], but they are also involved in the signal transduction by several other immune receptors, such as toll- like receptors (TLRs), interleukin-1 receptor (IL-1R), and T-cell receptor (TCR) [218-

220]. All TRAFs, except TRAF1, contain a known to mediate protein ubiquitination [221]. Indeed, TRAF2 and TRAF6 have been shown to function as ubiquitin ligases that catalyze the synthesis of Lys63-linked polyubiquitin chains [188,

189]. This type of ubiquitination, which occurs early during a cellular response, does not target protein degradation but is important for signal transduction [189, 222-224].

Interestingly, the self-ubiquitination of TRAF2 and TRAF6 is potently inhibited by

CYLD under overexpression conditions [225-227]. Although it remains unclear whether

CYLD regulates the ubiquitination of TRAFs under endogenous conditions, these findings suggest the possibility that CYLD may function as a negative regulator of TRAF ubiquitination and activation of downstream signaling events.

Among the downstream signaling cascades activated by TRAFs are those that lead to activation of IκB kinase (IKK) and three families of MAP kinases (MAPKs): c-Jun N- 55 terminal kinase (JNK), extracellular signal responsive kinase (ERK), and p38 [217]. IKK is known as a specific activator of NF-κB, a family of inducible transcription factors regulating genes involved in immune and inflammatory responses, cell growth/survival, and oncogenesis [228, 229]. The MAPKs activate a number of transcription factors, including the ternary complex factor Elk-1 and members of the AP1 and CREB/ATF families [230]. Additionally, the MAPKs are involved in posttranscriptional regulation of gene expression [231-233]. The biological functions of JNK are particularly diverse, which include regulation of immune and inflammatory responses, cell growth, apoptosis, and tumor formation [234-236]. Activation of JNK is mediated by a kinase cascade involving MAPK kinases (MAP2K) and MAPK kinase kinases. Two MAP2Ks, MKK4 and MKK7, serve as the direct kinases of JNK. MKK7 is required for JNK activation by inflammatory cytokines, whereas MKK4 is more important for JNK activation by stress signals [237].

Recent studies suggest that activation of IKK by TRAF6 and TRAF2 involves

Lys63-linked ubiquitination [188, 189]. This signaling mechanism appears to be important for IKK activation by specific immune receptors, including IL-1R and TCR

[220]. The ubiquitination of TRAF2 has also been shown to mediate activation of JNK induced by the inflammatory cytokine TNF-α [222, 223]. A role for CYLD in NF-κB regulation is suggested by some recent studies, which reveal that CYLD inhibits the activation of an NF-κB reporter gene in transfected cells [225-227, 238]. However, it is unclear whether CYLD functions as a negative regulator of the IKK or other signaling cascades downstream of various immune receptors. 56

In the present study, we have taken the RNA interference (RNAi)-mediated gene knockdown approach to investigate the function of endogenous CYLD in the regulation of cell signaling. We demonstrate that CYLD is key negative regulator of JNK downstream of diverse immune receptors. Further, CYLD also inhibits IKK activation, but this function of CYLD is receptor dependent.

MATERIALS AND METHODS

Plasmid constructs. Human CYLD was cloned by RT-PCR and inserted into the pcDNA-HA vector [239] downstream of the HA epitope tag. CYLDR is a modified form of the pcDNA-HA-CYLD, in which the siRNA binding site was mutated (by site-directed mutagenesis) without altering the amino acid codons. Thus, the CYLDR retains the wildtype CYLD amino acid sequence but is resistant to siRNA-mediated suppression.

GST-IκBα (1-54) was constructed by inserting a DNA fragment encoding the first 54 amino acids of human IκBα and 3 copies of HA epitope tag into the pGEX-4T-3 vector

(Pharmacia). GST-cJun (1-79) encodes a GST-fusion protein containing the first 79 amino acids of cJun.

Cell culture and antibodies–Human embryonic kidney 293 cells, human cervical carcinoma HeLa cells, and human B-cell line BJAB were obtained from ATCC. 293 cells stably transfected with murine CD40 (293-CD40) were kindly provided by Dr.

Steven Ley [240]. The anti-CYLD antibody was generated by injection of rabbits with a

GST-fusion protein containing an N-terminal region of human CYLD (amino acid 136- 57

301). Anti-mouse CD40 antibody was purchased from PharMingen. The polyclonal antibodies for tubulin (TU-02), ERK (K-23), JNK1 (C-17), JNK2 (N18), p38 (C-20),

Oct1 (C-21), TRAF2 (C-20), IKKg (FL-419), and MKK4 (MEK-4 H-98) were purchased from Santa Cruz. The recombinant JNK protein, anti-MKK7, and phospho-specific antibodies recognizing activated forms of different MAPKs were purchased from Cell

Signaling Technology Inc.

RNA interference (RNAi)–Small inhibitory RNAs (siRNAs) specific for human CYLD and luciferase were synthesized by Dharmacon Research, Inc. (Lafayette, CO). The sense strand sequences of the oligonucleotides are shown below.

CYLD siRNA: AAG UAC CGA AGG GAA GUA UAG

Luciferase siRNA: AAC TTA CGC TGA GTA CTT CGA

For siRNA delivery, 293 and HeLa cells were transfected in 6-well plates with

140 pmol of siRNA using Oligofectamine (Invitrogen). At 16-24 hr following the first transfection, the cells were transfected again with the same amount of siRNA together with 300 ng carrier DNA using Lipofectamine 2000. At about 30 hr after the second transfection, the cells were used for different experiments. For stable gene knockdown using the small hairpin RNA (shRNA) technique, a double-stranded oligonucloetide corresponding to the CYLD siRNA was cloned into the pSUPER-retro-puro vector

(Oligoengine) downstream of the U6 promoter. The generated retroviral construct, named pSUPER-shCYLD, was used to produce recombinant viruses and infect the indicated cells as previously described [241]. The infected cells were enriched by 58 selection using puromycine. The bulk-infected cells were used in the experiments to avoid clonal variations.

Immunoblotting (IB), in vitro kinase assay, and electrophoresis mobility shift (EMSA) assays–Cell lysates were prepared by lysing the cells in a kinase lysis buffer and immediately subject to IB and in vitro kinase assays as described previously [242].

Activated MAPKs were analyzed by IB using phospho-specific antibodies. Nuclear

32 extracts were prepared and subjected to EMSA [243] using a P-radiolabeled high- affinity κB probe ( 5'-CAA CGG CAG GGG AAT TCC CCT CTC CTT-3') or a control probe containing the Oct-1 binding site (5’-TGT CGA ATG CAA ATC CTC TCC TT-

3’) followed by resolving the DNA-protein complexes on native 5% polyacrylamide gels.

RESULTS

CYLD is a negative regulator of JNK, but not IKK, in the TNF-α signaling pathway. To systematically analyze the role of CYLD in regulation of cell signaling, we generated a

CYLD-specific antibody. This antibody could readily detect the transfected CYLD (Fig.

1A, lane 2). Additionally, it also detected an endogenous protein band comigrating with the transfected CYLD (lane 1). This protein band, which was not detected by IB using a preimmune serum (data not shown), became more prominent when higher amounts of cell extracts were used in the IB (Fig. 1B, lanes 1 and 3). To confirm that this protein is endogenous CYLD, we performed RNA interference (RNAi) assays. The expression of this endogenous protein was markedly suppressed by a CYLD-specific siRNA (siCYLD,

Fig. 1B, lanes 2 and 4) but not by a control siRNA for luciferase (siLuc, lanes 1 and 3). 59

Similar results were obtained in 293 and HeLa cells (Fig. 1B). The CYLD antibody also detected some other proteins, but this was likely due to non-specific crossreaction, since these proteins are much smaller than the predicted size (105 kD) of CYLD and since their expression was not affected by the CYLD siRNA.

With the CYLD antibody and siRNA, we first examined the effect of CYLD knockdown on cell signaling stimulated by the proinflammatory cytokine TNF-α. In both 293 and HeLa cells, TNF-α stimulated the catalytic activity of IKK and JNK, as demonstrated by immunecomplex kinase assays (Fig. 2A and B, top two panels). JNK activation was also detected based on its site-specific phosphorylation in vivo by immunoblotting (IB) using a phosphospecific anti-JNK antibody (panel 4). In addition to

IKK and JNK, TNF-α stimulated the activation of the p38 MAPK (Fig. 2A and B, panel

6) but did not appreciably induce the activity of ERK (data not shown).

If endogenous CYLD serves as a negative regulator of TNF-α-stimulated cell signaling, the CYLD knockdown should result in hyper activation of the specific kinases under the negative control of CYLD. In this regard, IKK is particularly interesting, since

CYLD has been shown to inhibit the induction of NF-κB reporter gene by various immune receptors under transient transfection conditions [225-227, 244]. To our surprise, however, the CYLD knockdown did not promote IKK activation in the TNF-α- stimulated 293 cells (Fig. 2A, upper panel) or HeLa cells (Fig. 2B, upper panel).

Consistently, the TNF-α stimulated NF-κB DNA binding activity was not enhanced in the CYLD-knockdown cells (Fig. 2C, panel 1). Interestingly, parallel analyses using the same cells revealed that the CYLD knockdown markedly enhanced the activation of

JNK, as demonstrated by both kinase assays (Fig. 2A and B, panel 2) and phospho- 60 specific IB assays (panel 4). The loss of CYLD also caused a low basal level of JNK activation in unstimulated cells (Fig. 2A and B, panel 4, lane 4). This result was not due to variations in protein loading, since the amounts of total JNK1 and JNK2 proteins

(panel 5) as well as tubulin (bottom panel) were comparable in the different samples.

Further, the CYLD knockdown did not enhance the activation of p38 (panel 6).

To further confirm that CYLD negatively regulates JNK activation in the TNF-α signaling pathway, we generated a modified form of CYLD cDNA harboring sense mutations in the siRNA-targeting site. Although such mutations do not change the amino acid sequence of CYLD, they render the expressed CYLD mRNA resistant to siRNA- mediated destruction. As expected, this modified version of CYLD (CYLDR) was efficiently expressed even in the presence of CYLD siRNA (Fig. 2D, panel 3, lanes 5 and

6). More importantly, expression of CYLDR in the CYLD-knockdown cells greatly reduced the level of JNK activation (panel 2, compare lanes 4 and 6). Furthermore, the

CYLD reconstitution did not affect TNF-α stimulated activation of IKK (panel 1).

Together, these data demonstrate that JNK is a primary downstream target of CYLD in the TNF-α signaling pathway.

CYLD knockdown has no effect on JNK activation by a stress agent. JNK activation can be induced by both immune stimuli and stress signals, which involve different upstream signaling pathways. To assess the mechanism by which CYLD negatively regulates JNK, we examined the effect of CYLD knockdown on JNK activation by a stress stimulus, anisomycin. As expected, incubation of 293 cells with anisomycin resulted in strong activation of JNK (Fig. 3A, panel 1, lanes 1-4). Interestingly, the anisomycin-induced

JNK activation was not significantly affected by the CYLD knockdown (lanes 5-8). On 61 the other hand, analysis of TNF-α-stimulated JNK activation using the same cells revealed a marked enhancement of this cytokine-specific JNK response by CYLD knockdown (lanes 9-12). This result indicates that CYLD does not regulate the JNK signaling pathway stimulated by stress signals. Furthermore, this finding also implies that CYLD does not directly regulate JNK but targets an upstream step(s) involved in

JNK activation by TNF-α and other immune stimuli.

CYLD negatively regulates the activation of MKK7. Gene targeting studies suggest that

JNK activation by inflammatory cytokines and stress signals involves two different upstream kinases, MKK7 and MKK4, with MKK7 being critical for cytokine-induced

JNK activation [237]. As a further step to investigate the mechanism underlying CYLD- mediated JNK regulation, we examined the effect of CYLD knockdown on TNF-α- stimulated activation of MKK7 and MKK4. These two JNK kinases were isolated from the cells by IP followed by analyzing their catalytic activity by in vitro kinase assays using recombinant JNK (catalytically inactive) as substrate. As seen with JNK activation, the TNF-α-stimulated MKK7 activation was markedly enhanced in CYLD- knockdown cells (Fig. 3B, panel 1). This drastic effect was not due to the variation in the level of MKK7 protein expression (panel 2). MKK4 was also weakly induced by TNF-α

(panel 3, lanes 1-4), however, this response was not significantly enhanced in the CYLD- knockdown cells (lanes 5-8). Thus, MKK7 is an upstream target of CYLD in the JNK signaling pathway.

CYLD negatively regulates JNK activation by diverse stimuli. We next expanded our studies to investigate whether CYLD also negatively regulates JNK signaling 62 downstream of other immune receptors. One receptor of interest is CD40, which is a member of the TNFR superfamily and mediates important immune functions via activation of IKK, JNK, as well as other MAPKs. For convenient CYLD knockdown, we used a previously characterized 293 cell line stably transfected with the murine CD40 cDNA (293-CD40, [240]). As expected from the prior studies [240], the 293-CD40 cells did not exhibit significant signaling activity under unstimulated conditions (Fig. 4A, panels 1 and 2, lane 1). However, crosslinking of CD40 with its agonistic antibody resulted in activation of both IKK (Fig. 4A, panel 1) and JNK (panel 2). Consistent with the TNF-α-stimulated cells, CYLD knockdown markedly enhanced the activation of JNK in the anti-CD40-treated cells (panel 2). A detailed time-course analysis revealed that the magnitude, but not the kinetics, of JNK activation was regulated by CYLD (panel 2).

Thus, CYLD functions as a negative regulator of JNK in both the TNF-α and CD40 signaling pathways. Interestingly, a parallel kinase assay revealed that the CD40- mediated IKK activation was also enhanced upon CYLD knockdown (panel 1).

Consistent with this finding, anti-CD40 stimulated hyper activation of NF-κB in the

CYLD-knockdown cells (Fig. 4B, lanes 6-8). Parallel assays revealed that in contrast to the activation of IKK and JNK, the activation of p38 was not affected by CYLD knockdown (Fig. 4A, panel 4).

In order to extend our studies to additional cell models, we employed a retroviral vector (pSUPER-retro-puro) to express a CYLD-specific small hairpin RNA (shRNA).

This approach allows gene suppression in cells with both high and low transfection efficiencies. Infection with CYLDshRNA, but not the empty pSUPER vector, efficiently suppressed the expression of CYLD in BJAB B cells (Fig. 4C, panel 3). We also used 63 the pSUPER shRNA system to knockdown CYLD in HeLa cells (Fig. 4D, panel 3). As seen with the TNF-α and CD40 signaling pathways, CYLD knockdown greatly enhanced

JNK activation by LPS and IL-1β (Fig. 4C and D, panel 2). The IKK activation was also promoted by CYLD knockdown in LPS- and IL-1β-stimulated cells (Fig. 4C and D, panel 1), although less prominent compared to the effect on JNK activation. Thus, JNK appears to be a primary downstream target of CYLD, but IKK is also negatively regulated by CYLD downstream of certain receptors.

DISCUSSION

Tumor suppressor CYLD is a newly identified member of the DUB family.

Although CYLD has been shown to inhibit the activation of NF-κB in reporter gene assays, its precise role in regulating signal transduction downstream of different immune receptors is poorly defined. In this study, we have investigated the function of endogenous CYLD using RNAi-mediated CYLD knockdown. Our data suggest that

CYLD functions as a key negative regulator of the JNK signaling pathway downstream of diverse immune stimuli. We have also shown that CYLD negatively regulates IKK, although this function of CYLD is in a receptor-dependent manner. Consistent with the prior NF-κB reporter studies [225-227, 244], we have shown that CYLD inhibits the activation of IKK by certain cellular stimuli, including anti-CD40, LPS, and IL-1β (Fig.

4). To our surprise, however, the CYLD knockdown has no appreciable effect on the

TNF-α-stimulated activation of IKK or NF-κB (Fig. 2). This result was not due to variations in CYLD knockdown or cell stimulations, since parallel kinase assays reveal a remarkable elevation of JNK activation caused by the CYLD deficiency (Fig. 2). 64

How CYLD differentially regulates IKK and JNK is not completely understood, but one potential mechanism is attributed to the differential requirement of TRAFs in these signaling pathways. Gene knockout studies suggest that TRAF2 gene deficiency only weakly inhibits TNF-α-induced NF-κB activation but largely abolishes the activation of

JNK by TNF-α [245]. Since TRAF2 is an upstream target of CYLD [225-227], these findings are consistent with our data that CYLD inhibits the activation of JNK but not

NF-κB in TNF-α-stimulated cells (Fig. 2). Our results are also supported by two other studies that suggest an essential role for TRAF2 ubiquitination in TNF-α-stimulated activation of JNK [222, 223] but not that of IKK [223]. The non-essential role of TRAF2 in NF-κB activation by TNF-α is likely due to the functional compensation by another

TRAF molecule, TRAF5, becauseTRAF2/TRAF5 doubly deficient cells have a severe defect in NF-κB activation by TNF-α [176]. Since CYLD has no effect on TNF-α- induced NF-κB activation, it is tempting to speculate that the signaling function of

TRAF5 is either not regulated by ubiquitination or controlled by a different DUB. A recent study [246] suggests that negative regulation of IKK in the TNF-α signaling pathway is mediated by A20, which acts by deubiquitinating the RIP kinase known to be essential for TNF-α-induced NF-κB activation [247, 248]. Thus, it seems likely that downstream of the TNFR, CYLD and A20 regulate the JNK and IKK cascades by targeting deubiquitination of TRAF2 and RIP, respectively. However, the possibility that

CYLD possesses additional targets cannot be excluded. In fact, the finding that CYLD negatively regulates the activation of JNK and IKK by LPS and IL-1β implicates a role for CYLD in negatively regulating the ubiquitination of TRAF6, since TRAF6 is an essential factor for the activation of these pathways downstream of both TLR4 (receptor 65 for LPS) and IL-1R [249]. At least under overexpression conditions, the ubiquitination of TRAF6 is inhibited by CYLD [226, 227].

The finding that CYLD negatively regulates JNK, as well as IKK, provides an insight into the tumor suppressor function of CYLD. The IKK/NF-κB pathway is well- known for its involvement in cell survival and oncogenic transformation as well as immune responses [250]. Accumulating evidence suggests that JNK is also a critical factor involved in tumorigenesis [236]. The JNK signaling pathway is constitutively activated in various tumor cells [234, 251, 252] and has been shown to play an essential role in oncogenesis in a number of tumor models [253-258]. Consistent with its oncogenic function, JNK has been shown to promote cell growth and survival [259-261].

However, under certain conditions, JNK also functions as an inducer of apoptosis [262-

264]. Although the precise mechanism determining the pro- versus anti-apoptotic functions of JNK remains unclear, strong evidence suggests that prolonged activation of

JNK promotes apoptosis [265-267], whereas transient activation of JNK contributes to cell survival [260]. Of note, CYLD knockdown increases the magnitude of JNK transient activation but does not prolong the activation kinetics (Fig. 2B and 4). This finding supports the idea that CYLD deficiency promotes cell survival [225]. In addition to its functions in oncogenesis and apoptosis regulation, JNK plays an important role in regulating immune and inflammatory responses [236]. Given the important role of

CYLD in regulating JNK and IKK downstream of diverse immune receptors (Fig. 2 and

4), it is tempting to speculate that this DUB may play an important role in immune regulation. Generation of CYLD knockout mice will be important for better 66 understanding the biological function of CYLD-mediated regulation of JNK and other signaling pathways.

One interesting observation of the present study is that CYLD knockdown results in a basal level of activation of JNK as well as IKK/NF-κB (Fig. 2A and B and Fig. 4).

This result indicates that the loss of CYLD is sufficient for triggering a low level of constitutive cell signaling. However, since cell lines may secret a low amount of cytokines [268], it is also possible that the constitutive kinase activity in CYLD knockdown cells is due to the stimulatory action of endogenous cytokines. Nevertheless, these findings suggest that CYLD is a critical signaling regulator that prevents aberrant activation of JNK and IKK.

ACKNOWLEDGMENTS

We thank S. Ley for the 293-CD40 cells and the Sun lab members for fruitful discussion.

I would also like to thank Dr. Mingyin Zhang for performing EMSA assays and numerious immunoblot assays. 67

FIGURE 1. Characterization of CYLD antibody and siRNA. (A) 293 cells were transfected with either the empty vector pcDNA or an expression vector encoding HA- tagged CYLD. Around 7 µg of protein lysates were subjected to IB using anti-CYLD.

The transfected HA-CYLD, endogenous CYLD, and some non-specific protein bands are indicated. (B) 293 or HeLa cells were transfected with either the control luciferase siRNA (siLuc) or CYLD-specific siRNA (siCYLD) as described in Materials and

Methods. Around 20 µg of cell lysates were subjected to IB using anti-CYLD. The protein bands are indicated as in A. 68 69

FIGURE 2. CYLD knockdown by RNAi results in hyper activation of JNK, but not IKK, in TNF-α-stimulated cells. (A) and (B) Effect of CYLD knockdown on TNF-α-stimulated cell signaling in 293 (A) and HeLa (B) cells. Cells were transfected with siRNA targeting CYLD or the luciferase (Luc) control. The cells were either not treated (NT) or stimulated with human recombinant TNF-α (20 ng/ml) for the indicated times. IKK and

JNK were isolated by IP using anti-IKKγ and anti-JNK1 plus anti-JNK2, respectively.

The catalytic activity of IKK and JNK in the immune complexes was detected by in vitro kinase assays using GST-IκBα (1-54) and GST-cJun (1-79) substrates, respectively (top two panels). Phosphorylated substrates are indicated as P-GST-IκBα and P-GST-cJun.

The cell lysates were also subjected to IB using the indicated antibodies to monitor the efficiency of CYLD knockdown (panel 3), the in vivo phosphorylation of JNK (panel 4) and p38 (panel 6), the total protein expression of JNK1 and JNK2 (panel 5) and p38

(panel 7), and also tubulin loading control (bottom panel). (C) EMSA to detect NF-κB

DNA binding activity. The control and CYLD-suppressed 293 cells described in A were stimulated with TNF-α. Nuclear extracts were isolated and subjected to EMSA using a

32P-radiolabeled probe for NF-κB (upper panel). As a loading control, the EMSA was performed using a probe for the constitutive transcription factor Oct-1 (panel 2). Total cell lysates were prepared from an aliquot of the cells and subjected to IB using anti-

CYLD (panel 3) and anti-tubulin (bottom panel). (D) CYLD reconstitution in CYLD siRNA-transfected cells. 293 cells were transfected with either control luciferase siRNA or CYLD siRNA. In the latter case, the CYLD siRNA was cotransfected with an RNAi- resistant form of CYLD (CYLDR, lanes 5 and 6) or an empty vector (lanes 3 and 4). The 70 cells were either not treated (-) or stimulated with TNF-α for 7.5 min. Catalytic activity of IKK (panel 1) and JNK (panel 2) were analyzed by kinase assays as described in A.

The cell lysates were also subjected to IB to monitor the efficiency of CYLD suppression

(panel 3) and tubulin expression (bottom panel). 71 72

FIGURE 3. CYLD knockdown has no effect on JNK activation by a stress agent and promotes MKK7 activation by TNF-α. (A) Effect of CYLD knockdown on JNK activation by a stress agent and TNF-α. 293 cells infected with either the empty pSUPER vector or pSUPER-shCYLD were stimulated with the stress agent anisomycin

(1 µg/ml) for the indicated time periods or TNF-α (20 ng/ml) for 15 min followed by analyzing the JNK kinase activity by kinase assays (panel 1) and the expression of JNK,

CYLD, and tubulin by IB. (B) Hyper activation of MKK7 in CYLD-knockdown cells.

293 cells were transfected with control or CYLD siRNA and stimulated as indicated.

MKK7 and MKK4 were isolated by IP using anti-MKK7 and anti-MKK4, respectively.

The catalytic activity of isolated MKK7 (panel 1) and MKK4 (panel 3) were examined by kinase assays using catalytically inactive recombinant JNK as substrate. The cell lysates were also subjected to IB using anti-MKK7 (panel 2) and anti-MKK4 (bottom panel) to detect the expression level of these two JNK kinases. 73 74

FIGURE 4. CYLD knockdown promotes activation of JNK by diverse stimuli and IKK by selective stimuli. (A) CYLD knockdown results in hyper activation of JNK and IKK by anti-CD40. 293-CD40 cells were transfected with luciferase siRNA (siLuc) or CYLD siRNA (siCYLD), either not treated (NT) or stimulated with anti-CD40 antibody (5

µg/ml). The catalytic activity of IKK (panel 1) and JNK (panel 2) was measured as described in Fig. 2A. The cell lysates were also subjected to IB assays as indicated. (B)

EMSA to detect NF-κB activation by anti-CD40. Nuclear extracts were isolated from

293-CD40 cells transfected with control (luciferase) or CYLD-specific siRNA and subjected to EMSA using the κB probe. The NF-κB/DNA binding complex and free probe are indicated. (C) and (D) Effect of CYLD knockdown on the activation of JNK and IKK by LPS (C) and IL-1β (D). The indicated cells were infected with either empty pSUPER vector or pSUPER-shCYLD. The cells were stimulated with the indicated inducers followed by analyzing the catalytic activity of IKK and JNK by kinase assays

(top two panels) and the expression of CYLD, JNK, and tubulin by IB (panels 3 to 5). 75 76

Chapter III

Regulation of the Deubiquitinating Enzyme CYLD by

IKKγ-dependent Phosphorylation

William Reiley, Minying Zhang, Xuefeng Wu, Erica Granger, and

Shao-Cong Sun

Mol. Cell. Biol. (2005) in press 77

ABSTRACT

Tumor suppressor CYLD is a deubiquitinating enzyme (DUB) that inhibits the ubiquitination of key signaling molecules, including TNF receptor-associated factor 2

(TRAF2). However, how the function of CYLD is regulated remains unknown. Here we provide evidence that inducible phosphorylation of CYLD is an important mechanism of its regulation. Under normal conditions, CYLD dominantly suppresses the ubiquitination of TRAF2. In response to cellular stimuli, CYLD undergoes rapid and transient phosphorylation, which is required for signal-induced TRAF2 ubiquitination and activation of downstream signaling events. Interestingly, the CYLD phosphorylation requires IκB kinase gamma (IKKγ) and can be induced by IKK catalytic subunits. These findings suggest that CYLD serves as a novel target of IKK and that the site-specific phosphorylation of CYLD regulates its signaling function. 78

INTRODUCTION

Protein ubiquitination is an important mechanism that regulates diverse cellular processes, including protein degradation, signal transduction, immune response, control, endocytosis, and vesicle trafficking [269, 270]. This molecular mechanism is under the control of both ubiquitin enzymes and deubiquitinating enzymes (DUBs).

The DUBs are a family of cysteine proteases that specifically digest polyubiquitin chains

[271]. In addition to their house-keeping function in the regeneration of free ubiquitin molecules, the DUBs play an important regulatory role in determining the fate and function of specific ubiquitin-conjugated proteins [270]. A new member of the DUB family, CYLD [225-227], was identified as a tumor suppressor mutated in familial cylindromatosis [238], an autosomal dominant predisposition to multiple tumors of the skin appendages [215, 216]. Recent studies suggest that CYLD mediates deubiquitination of certain signaling molecules, including members of the tumor necrosis factor receptor (TNFR)-associated factor (TRAF) family [226, 227, 244].

TRAFs function as signaling adaptors of TNFR superfamily [217] as well as a number of other immune receptors, such as toll-like receptors, interleukin-1 receptor, and

T-cell receptor [218-220]. A common structural feature of TRAFs, with the exception of

TRAF1, is their possession of a ring finger domain known to mediate protein ubiquitination [221]. It has been shown that TRAF2 and TRAF6 function as ubiquitin ligases that catalyze both self-ubiquitination and the ubiquitination of specific target molecules involved in signal transduction [188, 189, 222-224]. Strong evidence suggests that the ubiquitination of TRAF2 and TRAF6 functions as a molecular trigger for 79 initiating downstream signaling events, including activation of c-Jun N-terminal kinase

(JNK) [222, 223] and IκB kinase (IKK) [189, 220].

IKK is known as a specific activator of the transcription factor NF-κB [228, 229].

The IKK complex is composed of two catalytic subunits, IKKα and IKKβ, and a regulatory subunit termed IKKγ [272]. Upon activation by various cellular stimuli, IKK phosphorylates the NF-κB inhibitor IκBα, triggering the proteolysis of IκBα and nuclear translocation of active NF-κB [272]. The canonical substrates of JNK include c-Jun,

JunD, and ATF2, components of the transcription factor AP-1 [273]. Like IKK, JNK regulates diverse biological processes, ranging from immune responses, cell growth, and apoptosis to tumor formation [234-236, 273]. A role for CYLD in IKK regulation is suggested by some recent studies, which reveal that CYLD inhibits the activation of an

NF-κB reporter gene [225-227, 238]. Using RNA interference (RNAi)-mediated CYLD knockdown, we have shown that CYLD is also a key negative regulator of JNK downstream of specific immune receptors [300]. In agreement with these findings,

CYLD inhibits the ubiquitination of both TRAF2 and TRAF6 in transfected cells [225-

227]. However, despite these important observations, the molecular mechanism regulating the function of CYLD remains unknown. In this study, we demonstrate that

CYLD plays a dominant role in suppressing the ubiquitination of endogenous TRAF2.

CYLD knockdown results in constitutive ubiquitination of TRAF2. Interestingly, signal- induced TRAF2 ubiquitination is associated with site-specific phosphorylation of CYLD, a molecular event that appears to prevent CYLD from inhibiting the ubquitination of

TRAF2. We further demonstrate that the CYLD phosphorylation is dependent on IKKγ. 80

MATERIALS AND METHODS

Plasmid constructs. Human CYLD was cloned by RT-PCR and inserted into the pcDNA-HA vector [239]. CYLD truncation mutants were generated by PCR and designated by the specific amino acid residues retained in the mutant proteins. The

CYLD mutants harboring amino acid substitutions were created by site-directed mutagenesis (Stratagene) using the wildtype CYLD expression vector as template (the primer sequences are available upon request). RNAi-resistant form of CYLD (CYLDR) and its mutants were produced by introducing sense mutations (no alteration of amino acid codons) at the siRNA-binding site. To generate the retroviral vectors expressing

CYLD and its mutants, the corresponding cDNAs were transferred from pcDNA-HA to the pCLXSN retroviral vector (provided by Dr. Inder Verma, see [275]. In some experiments, the pCLXSN vector was modified by replacing its neomycin-resistant gene with the green fluorescence protein (GFP) gene. -tagged IKKγ was generated by inserting the human IKKγ cDNA into pcDNA vector downstream of a myc epitope tag. pCLXSN-hCD40 was cloned by inserting the human CD40 cDNA into the pCLXSN retroviral vector. Expression vectors encoding HA-tagged IKKα and IKKβ [242], HA- tagged ubiquitin [70, 276], GST-IκBα(1-54) and GST-c-Jun(1-79) [300] were described previously. GST-CYLD(403-513) was cloned by inserting a DNA fragment encoding amino acid 403-513 of human CYLD into the pGEX4T-1 vector (Amersham/Pharmacia

Biotech). The point mutants of GST-CYLD(403-513) were generated by site-directed 81 mutagenesis. All the DNA constructs were sequenced at the Core Facility of Hershey

Medical Center.

Antibodies, cell lines, and other reagents. The anti-CYLD antibody was generated by injection of rabbits with a GST-fusion protein containing an N-terminal region of human CYLD (amino acid 136-301). The phospho-specific CYLD antibody

(anti-P-CYLD S418) specifically recognizes CYLD containing phosphorylated serine

418. This antibody was generated by injection of rabbits with a KLH-conjugated peptide covering amino acid 413-428 of human CYLD, in which serine 418 was phosphorylated.

Polyclonal antibodies for tubulin (TU-02), JNK1 (C-17), JNK2 (N18), TRAF2 (C-20), and IKKγ (FL-419) were purchased from Santa Cruz. HRP-conjugated HA antibody

(3F10) and human recombinant TNF-α were from Roche Molecular Biochemicals.

PMA, ionomycine, and LPS (derived from E. coli 0127:B8) were from Sigma.

Recombinant IKKα and IKKβ proteins were provided by Dr. Michael Karin.

Human embryonic kidney 293 cells, human cervical carcinoma HeLa cells, human B-cell line BJAB, and human leukemia T-cell line Jurkat were obtained from

ATCC. IKKg-deficient Jurkat cell line, JM4.5.2, and its IKKg-reconstituted derivative,

JM4.5.2-IKKg, were described previously [241, 277].

RNA interference (RNAi). Small interfering RNAs (siRNAs) specific for human

CYLD and luciferase were synthesized by Dharmacon Research, Inc. (Lafayette, CO).

The sense strand sequences of the oligonucleotides are shown below.

CYLD siRNA: AAG UAC CGA AGG GAA GUA UAG

Luciferase siRNA: AAC TTA CGC TGA GTA CTT CGA 82

For siRNA delivery, 293 and HeLa cells were transfected in 6-well plates with

140 pmol of siRNA using Oligofectamine (Invitrogen). At 16-24 hr following the first transfection, the cells were transfected again with the same amount of siRNA together with 300 ng carrier DNA using Lipofectamine 2000. At about 30 hr after the second transfection, the cells were used for different experiments.

For stable gene knockdown using the small hairpin RNA (shRNA) technique, a double-stranded oligonucloetide corresponding to the CYLD siRNA was cloned into the pSUPER-retro-puromycin vector (Oligoengine) downstream of the U6 promoter. The generated retroviral construct, named pSUPER-shCYLD, was used to produce recombinant viruses and infect the indicated cells. The infected cells were enriched by selection using puromycine. The bulk-infected cells were used in the experiments to avoid clonal variations.

Immunoblotting (IB), immunoprecipitation (IP), and in vitro kinase assays. Cell lysates were prepared by lysing the cells in a kinase lysis buffer supplemented with phosphatase inhibitors and immediately subject to IB and in vitro kinase assays as described previously [242]. To detect phoshorylated CYLD based on its band shift, the

CYLD proteins were first concentrated by IP (using anti-CYLD) followed by fractionation using 6% SDS gels and IB. CoIP was performed to determine the interaction CYLD with TRAF2 and IKKγ.

In vivo ubiquitin conjugation assays. The indicated cells were transfected in 6- well plates with pcDNA-HA-ubiquitin (0.5 µg). At 30-40 h following transfection, the 83 cells were stimulated as indicated and lysed in RIPA buffer [70] supplemented with inhibitors of ubiquitin hydrolases (20 mM NEM and 5 µM ubiquitin aldehyde). The cell lysates were immediately subjected to IP using anti-TRAF2, and the ubiquitin-conjugated

TRAF2 was analyzed by IB using HRP-conjugated anti-HA antibody.

Retroviral infection. Retroviral particles were produced using the pCLXSN vector system as previously described [241]. The infected cells were enriched by drug selection using either puromycin (for CYLD knockdown with pSUPER-retro-puro vector) or neomycine (for infection with pCLXSN vector). For reconstitution of the

Jurkat-shCYLD cells with RNAi-resistant CYLD (CYLDR) or mutant M4 (M4R), the pCLXSN(GFP) vector was used and the infected cells were enriched by fluorescence cell sorting (Core Facility of Hershey Medical Center). The bulk-infected cells were used in the experiments to avoid clonal variations.

RNase protection assay (RPA). Total cellular RNA was isolated from the indicated cells using the TRI reagent (Molecular Research Center, Inc., Cincinnati, OH).

RPA was performed using the BD RiboQuant Reagents and a custom template set according to the manufacturer’s instruction (BD Biosciences).

Luciferase reporter gene assay. Luciferase assays were performed using Jurkat T cells expressing wildtype CYLD (Jurkat-shCYLD-CYLD WTR) or a phosphorylation- defective CYLD mutant (Jurkat-shCYLD-CYLD M4R). To create these cells, Jurkat T 84 cells were first infected with pSUPER-shCYLD) to stably knockdown endogenous

CYLD. The generated Jurkat-shCYLD cells were then reconstituted, by retroviral infection, with RNAi-resistant form expression vectors encoding either the wildtype

CYLD (CYLD WTR) or its phosphorylation-deficient mutant (CYLD M4R). For luciferase assays, the cells (5 x 105) were seeded in 12-well plates and transfected using

Fugene 6 transfection reagent (Roche) with κB-TATA-luc reporter (250 ng) together with 250 ng of either an empty pCLXSN vector or the same vector containing human

CD40 cDNA (pCLXSN-CD40). After 40 hr, the cells were lysed in a cell lysis buffer

(Passive buffer) and subjected to dual luciferase assays (Promega, dual-luciferase reporter assay system).

RESULTS

CYLD knockdown results in constitutive ubiquitination of TRAF2. Recent studies suggest that TRAF2 undergoes rapid ubiquitination in response to TNF-α stimulation, which is involved in JNK activation [222, 223]. Since CYLD functions as a negative regulator of JNK in the TNF-α signaling pathway [300], we analyzed the role of CYLD in regulating the uibquitination of TRAF2 under endogenous conditions. For these studies, endogenous CYLD was knocked down in 293 cells by RNAi followed by examining the ubiquitination of endogenous TRAF2 under untreated or TNF-α- stimulated conditions. As expected, the expression level of CYLD was greatly reduced in cells transfected with a CYLD-specific siRNA (siCYLD, Fig. 1A, panel 2, lanes 4-6) but was not affected in cells transfected with a control siRNA for luciferase (siLuc, lanes

1-3). In the control cells, only a trace amount of ubiquitinated TRAF2 was detected 85 under untreated conditions (Fig. 1A, panel 1, lane 1), but the TRAF2 ubiquitination was markedly induced by TNF-α (lanes 2 and 3). Remarkably, when CYLD was knocked down by its specific siRNA, TRAF2 became costitutively ubiquitinated (lane 4).

Stimulation of these cells with TNF-α failed to further enhance TRAF2 ubiquitination

(lanes 5 and 6). Parallel studies using HeLa cells obtained similar results (Fig. 1B).

These findings suggest that CYLD plays a critical role in suppressing the in vivo ubiquitination of TRAF2 and raise the possibility that signal-mediated induction of

TRAF2 ubiquitination may involve functional modification of CYLD.

CYLD undergoes phosphorylation in response to immune stimuli. During the course of CYLD IB assays, we noticed that when the cell lysis buffer was supplemented with phosphatase inhibitors, a more slowly migrating CYLD band could be detected in

Jurkat cells stimulated with mitogens (PMA plus ionomycin) or TNF-α (Fig. 2A).

Interestingly, the slower-migrating form of CYLD could be converted to its basal form by in vitro incubation with a non-specific phosphatase (Fig. 2A, lane 13), thus revealing phosphorylation as the nature of CYLD modification. The CYLD phosphorylation was also readily detected in TNF-α-stimulated 293 (Fig. 2C) and HeLa cells (Fig. 2D) as well as in LPS-stimulated BJAB B cells (Fig. 2E). Notably, the kinetics of CYLD phosphorylation was correlated with the phosphorylation of IκBα, both appearing as early as 5 min in cells stimulated with mitogens and TNF-α (Fig. 2A, C, and D, lane 2; P-

CYLD and P-IκBα). Consistently, the activation of IKK was also tightly associated with

CYLD phosphorylation (compare Fig. 1A and Fig. B). The slightly delayed kinetics of

IκBα degradation, compared to that of IKK activation and phosphorylation of CYLD, is consistent with the involvement of post-phosphorylation steps (e.g. ubiquitination and 86 proteasome targeting) in this proteolytic event. Together with the previous finding that

CYLD physically interacts with the IKK regulatory subunit, IKKγ [226, 227], these results raise the intriguing possibility that IKK may be involved in the inducible phosphorylation of CYLD.

CYLD phosphorylation requires IKKγ and can be induced by IKK catalytic subunits. To determine the importance of IKK in CYLD phosphorylation, we analyzed this signaling event using an IKKγ-deficient Jurkat T-cell line, JM4.5.2, which was created in our laboratory [277] and has since been used in numerous studies by others.

Due to the lack of IKKγ, the IKK complex cannot be activated by various known IKK stimuli, such as mitogens and cytokines, and this defect can be rescued by expression of the exogenous IKKγ [241, 278]. As expected, mitogens failed to induce the degradation of IκBα (target of IKK) in JM4.5.2 cells (Fig. 3A, lower panel, lane 4), but this signaling defect was rescued when the mutant cells were reconstituted with exogenous IKKγ (lane

6). Remarkably, the inducible phosphorylation of CYLD was also dependent on IKKγ, which was blocked in the IKKγ-deficient cells and rescued upon IKKγ reconstitution

(Fig. 3A, upper panel). In concert with this finding, CYLD was phosphorylated in 293 cells by transfected IKK catalytic subunits, IKKβ (Fig. 3B, lane 2) or IKKα (lane 4).

Notably, optimal phosphorylation of CYLD by both IKKα and IKKβ required their cotransfection with IKKγ (lane 6 and data not shown). Since IKKγ physically interacts with CYLD [226, 227], it is likely that IKKγ is required for recruiting CYLD to the IKK catalytic subunits.

To localize the region within CYLD that is phosphorylated by IKK, CYLD mutants harboring sequential truncations were subjected to phosphorylation analysis (Fig. 3C). 87

Deletion of 419 amino acids from the N-terminus of CYLD has no significant effect on its phosphorylation (Fig. 3C, upper panel, lane 4). Interestingly, further deletion of 27 amino acids generated a CYLD mutant (447-956) that was no longer phosphorylated

(lane 6). The defect of this CYLD mutant in phosphorylation was not due to loss of its

IKKγ-binding activity, since all the truncation mutants were competent in IKKγ binding

(Fig. 3D). Thus, the phosphorylation site of CYLD is likely located between amino acids

420 and 446.

To determine whether IKK directly phosphorylates CYLD, we generated a GST-

CYLD fusion protein containing a region of CYLD (amino acids 403-513) that covers its phosphorylation site. We then performed in vitro kinase assays using the GST-CYLD substrate and IKK holoenzyme isolated from Jurkat cells or recombinant IKKs. The

GST-CYLD was not significantly phosphorylated by inactive IKK isolated from untreated cells but potently phosphorylated by the active IKK isolated from mitogen- and

TNF-α-stimulated cells (Fig. 3E, upper panel). The relative level of GST-CYLD phosphorylation was similar to that of the phosphorylation of GST-IκBα (lower panel).

The CYLD phosphorylation was specific, since the IKK complex did not phosphorylate the GST protein (Fig. 3F). Further, the GST-CYLD was also phosphorylated by recombinant IKKβ (Fig. 3G, lanes 6-10) and IKKα (data not shown, also see Fig. 4C).

Parallel dose-dependent kinase assays using GST-CYLD (Fig. 3G, lanes 6-10) and GST-

IκBα (lanes 1-5) revealed that IKK phosphorylates these two substrates with comparable efficiencies.

A serine cluster of CYLD is involved in its phosphorylation. The region of CYLD phosphorylation contains a cluster of serines (Fig. 4A, bolded), including two serine pairs 88 that exhibit homology with the IKK phosphorylation site in IκBα and IκBβ (Fig. 4A, underlined, and Fig. 4B). Since IKK is known to be a serine kinase [279], one or more of these serines are likely involved in CYLD phosphorylation. To test this hypothesis, serine-to-alanine (S/A) substitutions were introduced into the GST-CYLD construct to generate mutants harboring various S/A substitutions (Fig. 4A, M1-M4). In vitro kinase assays revealed that mutation of one pair of the serines (see Fig. 4A, M1) partially affected the CYLD phosphorylation by IKK (Fig. 4C, lane 2). Combined mutation of the two pairs of serines (M2) significantly diminished the CYLD phosphorylation (Fig.

4C, lane 3), and the phosphorylation of CYLD was completely abolished by the combined mutations of the entire serine cluster (M4, lane 5). Similar results were obtained with the IKK holoenzyme (Fig. 4C, panel 1), recombinant IKKβ (panel 2), and recombinant IKKα (panel 3).

To determine whether the result of in vitro kinase assays is consistent with the in vivo phosphorylation of CYLD, the various mutations illustrated in Fig. 4A were introduced directly into full-length CYLD, and the resulting CYLD mutants were examined for their in vivo phosphorylation by transfected IKKβ and IKKγ. Similar to the in vitro phosphorylation results, the in vivo phopshorylation of CYLD was progressively attenuated along with mutation of more serines and was completely abolished by the M4 mutation (Fig. 4D). The phosphorylation defect of M4 was not due to its loss of IKKγ- binding activity, as demonstrated by coIP assays (data not shown).

We next determined whether the serine cluster was responsible for CYLD phosphorylation induced by cellular stimuli. For these studies, we stably knocked down the endogenous CYLD in Jurkat and HeLa cells using a pSUPER retroviral vector 89 encoding CYLD-specific shRNA (shCYLD). The generated CYLD-deficient cells

(named Jurkat-shCYLD and HeLa-shCYLD) were then reconstituted with RNAi-resistant form of wildtype CYLD (WTR) or its phosphorylation-deficient mutant M4 (M4R). As expected, CYLD expression was extremely low in cells infected with pSUPER-shCYLD

(Fig. 4E, lanes 2 and 6) but could be readily detected in the control cells infected with the empty pSUPER vector (lanes 1 and 5). Further, the CYLD deficiency in the Jurkat- shCYLD and HeLa-shCYLD cells was efficiently rescued by retroviral infection with the

RNAi-resistant wildtype CYLD (WTR, lanes 3 and 7) and M4 mutant (M4R, lanes 4 and

8). In vivo phosphorylation analysis revealed that as seen with the endogenous CYLD, the exogenous wildtype CYLD was phosphorylated upon cellular stimulation by mitogens (Fig. 4F, lanes 1-3) and TNF-α (lanes 7-9). In contrast, the CYLD M4 mutant was not phosphorylated in response to either inducer (lanes 4-6 and 10-12). Thus, the

IKK phosphorylation sites within CYLD are responsible for its in vivo phosphorylation stimulated by immune stimuli.

To directly confirm the in vivo phosphorylation of CYLD at its serine cluster, we generated an antibody that recognizes CYLD only when it is phosphorylated at one of the clustered serines (serine 418). As expected, this phospho-specific CYLD antibody

(named anti-P-CYLD S418) did not react with the basal form of CYLD in untreated cells

(Fig. 5A, lane 3). More importantly, the phosphorylated CYLD was efficiently detected by this antibody in stimulated cells (lane 4). Thus, serine 418 is phosphorylated in vivo.

IB assays using a CYLD mutant harboring a serine to alanine mutation at serine 418

(S418A) showed that mutation of this phosphorylation site did not abolish the inducible 90 phosphorylation of CYLD (Fig. 5B). This result further supports our idea that multiple serines within the serine cluster likely contribute to CYLD phosphorylation.

CYLD phosphorylation serves as a critical mechanism for regulating signal- induced TRAF2 ubiquitination and JNK activation. To understand the functional significance of CYLD phosphorylation in regulating its signaling function, we first examined whether the CYLD phosphorylation affects its function in regulating TRAF2 ubiquitination. These studies were performed using the HeLa-shCYLD cells reconstituted with RNAi-resistant wildtype CYLD (WTR) or its phosphorylation- defective mutant (M4R). Upon stimulation with TNF-α, the endogenous TRAF2 was ubiquitinated in the CYLD WTR-reconstituted cells (Fig. 6A, upper panel, lanes 2 and 3).

The inducible TRAF2 ubiquitination was not due to alteration in the level of overall protein ubiquitination in the cells (lower panel, lanes 1-3). Remarkably, the inducible ubiquitination of TRAF2 was largely defective in cells reconstituted with the phosphorylation-defective CYLD mutant (Fig. 6A, upper panel, lanes 5 and 6). Thus, the

CYLD phosphorylation is required for TNF-α-stimulated TRAF2 ubiquitination.

Together with the results presented in Fig. 1, this finding indicates that the CYLD phosphorylation may serve as a mechanism to inactivate its TRAF2-deubiquitination activity. To further examine this possibility, we generated a phospho-mimetic form of the CYLD M4 mutant by substituting the phosphorylation sites (serines) with glutamic acids (named M4 S/E). The constitutive TRAF2-deubiquitinating activity of CYLD and its mutants were analyzed by transient transfection in 293 cells. As previously reported, overexpressed TRAF2 underwent self-ubiquitination (Fig. 6B, lane 1), which was inhibited by coexpression with wildtype CYLD (lane 2). Consistent with its ability to 91 block the inducible TRAF2 ubiquitination, the M4 mutant of CYLD retained the TRAF2- deubiquitinating activity (lane 3). Interestingly, however, the phospho-mimetic CYLD

(M4 S/E) completely lost its ability to inhibit TRAF2 ubiquitination (lane Fig. 6B, lane

4). Parallel coIP assays revealed that both M4 and M4 S/E remained competent in association with TRAF2 (Fig. 6C). It is thus likely that the phosphorylation of CYLD does not prevent its binding to TRAF2 but may inactivate its DUB function by other mechanisms.

We have recently shown that CYLD is a negative regulator of JNK in the TNF-α signaling pathway [300]. Based on the results described above, we reasoned that CYLD phosphorylation might also be required for optimal activation of JNK. This hypothesis was examined using the CYLD-knockdown cells reconstituted with wildtype or mutant forms of exogenous CYLD. In response to TNF-α stimulation, JNK was activated in

HeLa-shCYLD cells reconstituted with wildtype CYLD (Fig. 6D, lanes 1-3), and the level of JNK activation was comparable to the parental HeLa cells (data not shown).

Interestingly, JNK activation was significantly attenuated in the cells reconstituted with

CYLD M4 (lanes 4-6). Similar results were obtained with the Jurkat-shCYLD cells, which revealed a significant defect in JNK activation in the M4R-reconstituted cells (Fig.

6D, lanes 10-12) as compared with the WTR-reconstituted cells (lanes 7-9). These data suggest that CYLD phosphorylation is required for interrupting its negative-regulatory function in JNK activation.

Functional consequence of CYLD phosphorylation in regulating gene expression.

We next analyzed the functional consequence of CYLD phosphorylation in gene induction. Reporter gene assays were performed to assess the role of CYLD 92 phosporylation in regulating NF-κB-specific gene expression. We used CD40 as an inducer, because CYLD plays an important role in regulating CD40-mediated NF-κB signaling [300]. As expected, CD40 strongly induced the κB-dependent reporter gene expression in Jurkat cells expressing the wildtype CYLD (Fig. 7A, column 2). In contrast, the CD40-induced κB response was drastically reduced in cells expressing the phosphorylation-defective CYLD mutant M4 (column 4).

We also analyzed the expression of endogenous genes induced by TNF-α.

Although CYLD has no significant role in TNF-α-mediated NF-κB activation, it negatively regulates the TNF-α-stimulated JNK signaling pathway. RPA was performed to examine the expression of several genes known to be induced by TNF-α. Whereas most of these genes were similarly induced in cells expressing wildtype or M4 mutant of

CYLD, the induction of one gene, rantes, was markedly attenuated in the M4-expressing cells (Fig. 7B, top panel, lane 4). This result was not due to the variation in RNA amounts, since it was observed only with rantes. Further, expression of two house- keeping genes (L32 and GAPDH) suggest that the level of M4 cell RNA was even slightly higher than that of the WT cell RNA. This result is consistent with the involvement of JNK in rantes gene induction. Since rantes is a critical chemokine that mediates migration of inflammatory cells [280, 281], it is important to examine the role of CYLD in regulating inflammatory responses when CYLD knockout mice are available. 93

DISCUSSION

Protein ubiquitination is positively regulated by ubiquitin enzymes and counter regulated by DUBs. Although the ubiquin enzymes have been extensively studied, little is known how the DUBs are functionally regulated. In the present study, we have examined the molecular mechanism regulating the function of CYLD, a DUB that functions as a key negative regulator of cell signaling. Our data suggest that CYLD plays a dominant role in suppressing the ubiquitination of TRAF2, since CYLD knockdown results in constitutive TRAF2 ubiquitination (Fig. 1). We provide evidence that signal- induced TRAF2 ubiquitination and JNK activation requires phosphorylation of CYLD, which appears to serve as a mechanism that inactivates the TRAF2-deubiquitination function of CYLD. Moreover, we show that IKKγ, a key component of the canonical

IKK, is required for the inducible phosphorylation of CYLD.

The constitutive TRAF2 ubiquitination in CYLD knockdown cells explains the basal activation of JNK and IKK associated with CYLD knockdown [300]. However, since the kinase activity [300], but not the TRAF2 ubiquitination (Fig. 1), is further induced by TNF-α, it suggests that TRAF2 ubiquitination is not sufficient for triggering optimal activation of JNK or IKK. The finding that CYLD knockdown results in constitutive TRAF2 ubiquitination not only suggests a dominant function of the DUB

CYLD but also implies that the ubiquitin ligase TRAF2 may undergo spontaneous ubiquitination. This hypothesis is consistent with the finding that TRAF2 undergo self- ubiquitination in transfected cells, which is inhibited by contransfection with CYLD (ref.

[226, 227, 244] and Fig. 6B of this study). Of course, since these results were obtained with transformed cell lines, it remains to be determined whether CYLD-deficient primary 94 cells also exhibit basal TRAF2 ubiquitination activity. Nevertheless, it is conceivable that under normal conditions, the ubiquitinated TRAF2 may be rapidly deubiquitinated by CYLD. Based on this hypothesis, one can predict that induction of TRAF2 ubiquitination by cellular stimuli is achieved by either enhancing the ubiquitin ligase activity of TRAF2 or inhibiting the DUB function of CYLD. Whereas the former possibility remains to be investigated, we have obtained strong evidence that supports the latter possibility. We show that signal-induced TRAF2 ubiquitination is associated with phosphorylation of CYLD. When the endogenous CYLD is replaced with a phoshorylation-defective CYLD mutant, the inducible ubiquitination of TRAF2 is severely attenuated (Fig. 6A). A logical explanation of this finding is that phosphorylation serves as a mechanism that temporarily inactivates the DUB activity of

CYLD, thus allowing the accumulation of ubiquitin-conjugated TRAF2. In further support of this hypothesis, a phospho-mimetic form of CYLD completely lost its TRAF2- deubiquitinating function (Fig. 6B).

The functional consequence of CYLD phosphorylation remains an interesting topic of investigation, since the in vivo physiological function of CYLD has not been well defined. Nevertheless, we have shown that interruption of CYLD phosphorylation markedly inhibits CD40-mediated induction of a κB-specific reporter gene (Fig. 7A).

Further, in the TNF-α signaling pathway, the CYLD phosphorylation appears to be important for the expression of a chemokine gene, rantes (Fig. 7B). Future studies will employ the gene array technique to identify more genes under the regulation of CYLD.

A prior study suggests that CYLD participates in TNF-α-induced apoptosis [225]. We thus examined whether the phosphorylation of CYLD diminishes its proapoptotic 95 function. To our surprise, we did not observe any significant effect of CYLD knockdown or CYLD phosphorylation on TNF-α-induced apoptosis in our system (supplementary

Fig. 1). Although the precise reason for this discrepancy is not clear, we have noticed that the previous study used a different protocol for apoptosis induction, which involved pre-stimulating the cells with PMA [225]. Nevertheless, taken together with our recent finding that CYLD regulates NF-κB activation by only certain stimuli [300], our data indicate that CYLD regulates cell survival downstream of specific receptors. The IKK- mediated CYLD phosphorylation may provide a crosstalk between the canonical NF-κB inducing signals and the CYLD-specific receptors. This idea is consistent with the fact that patients with CYLD genetic deficiency only develop a specific type of benign tumor, cylindromatosis [238], instead of having global abnormalities in cell growth/survival.

Clearly, a better understanding of the functional consequence of CYLD phosphorylation requires more knowledge regarding the physiological function of CYLD, which in turn will rely on studies using an in vivo model system (e.g. CYLD knockout mice).

IKK is known as a kinase that specifically phosphorylates IκBα and related inhibitors, thereby mediating the nuclear translocation of NF-κB. One intriguing question is whether IKK has other substrates or mediates additional signaling functions.

Novel functions have indeed been identified for the noncanonical IKK component IKKα.

One such function of IKKα is regulation of epidermal differentiation and skeletal morphology, a process that does not require the catalytic activity of IKKα [282].

Another interesting function of IKKα, which requires its kinase activity, occurs in the nucleus and involves phosphorylation of the histone H3 [283-285]. This novel function of IKKα is required for transcriptional activation of NF-κB target genes. Our present 96 study suggests that IKK also regulates the function of DUB CYLD. We have found that

CYLD undergoes rapid and transient phosphorylation in cells stimulated with various known IKK inducers (Fig. 2). The CYLD phoshorylation is dependent on IKKγ, since it is completely blocked in IKKγ-deficient Jurkat T cells (Fig. 3A). Transfection and in vitro kinase assays reveal that both IKKα and IKKβ are able to phosphorylate CYLD

(Fig. 3 and 4), although it remains unclear whether one or both of these IKK catalytic subunits are essential for CYLD phosphorylation. We attempted to address this question using mouse embryonic fibroblasts (MEF) deficient in different IKK subunits.

Unfortunately, the phosphorylation of CYLD could not be detected even in wildtype

MEFs (data not shown). Currently, we do not know whether this result is due to the variations in species (the anti-CYLD antibody is for human protein) or cell types.

Nevertheless, at least in vitro, CYLD does not show preference to IKKα or IKKβ, a property that is different from IκBα and the IκB-like molecule p100, which are preferentially phosphorylated by IKKβ and IKKα, respectively [286]. Another interesting feature of CYLD phosphorylation is the critical requirement of IKKγ.

Although IKKα and IKKβ efficiently phosphorylate CYLD in vitro (Fig. 3G and 4C), induction of CYLD phosphorylation in vivo by these IKK catalytic subunits requires the assistance by IKKγ (Fig. 3B). One likely interpretation of this result is that IKKγ functions as both the regulatory subunit of IKK and an adaptor for recruiting CYLD to the IKK catalytic subunits. However, the possibility for the involvement of novel mechanisms in IKKγ function cannot be excluded. For example, IKKγ may regulate another kinase that cooperates with the classical IKK in CYLD phosphorylation.

Notwithstanding, our findings clearly establish IKKγ as an essential factor in signal- 97 induced CYLD phosphorylation. Phosphorylation of CYLD represents a novel aspect of

IKK function, since it regulates the deubiquitination function of CYLD rather than triggering the degradation of this novel substrate.

ACKNOWLEDGEMENTS

We thank M. Karin and I. M. Verma for reagents and the Sun lab members for fruitful discussion. We also thank the Core Facility of Hershey Medical Center for oligonucleotide synthesis and DNA sequencing analyses. I would like to graceiously thank Dr. Mingying Zhang for her work on TRAF2 ubiquitination and CYLD phosphorylation. I would further like to thank Xuefeng Wu for the generation of CYLD serine-to-alanine mutations and production of various GST-CYLD proteins. I would finaly like to acknowledge Erica Granger for her work on GST-CYLD kinase assays with recombanint IKK. This work was supported by research grants from the National

Institutes of Health to S.-C. S (CA094922 and AI45045) and M. Z (AI45045) and the

Four Diamond Fund at Penn State Children’s Hospital to M. Z. W.R. was supported by a predoctoral/postdoctoral training grant (5 T32CA60395-09) from the National Institutes of Health. 98

FIG. 1. CYLD knockdown results in constitutive ubiquitination of TRAF2. 293 (A) or

HeLa (B) cells were transfected using Lipofectamine 2000 with control or CYLD- specific siRNA together with pcDNA-HA-ubiquitin. The cells were stimulated with

TNF-α (50 ng/ml) and then lysed in a buffer containing inhibitors of ubiquitin hydrolases. Endogenous TRAF2 was isolated by IP using anti-TRAF2 antibody, and the ubiquitinated TRAF2 was detected by IB using an HRP-conjugated anti-HA antibody

(top panel). The intracellular level of CYLD and TRAF2 was analyzed by IB using anti-

CYLD and anti-TRAF2 antibodies (middle and bottom panels). 99 100

FIG. 2. CYLD is phosphorylated in response to diverse cellular stimuli. (A) Jurkat T cells were stimulated with either T-cell mitogens (50 ng/ml of PMA plus 1 µM of ionomycin) or TNF-α (20 ng/ml) for the indicated times. To prevent loss of the phosphorylated CYLD, the cells were lysed in a kinase lysis buffer supplemented with phosphatase inhibitors. CYLD proteins were concentrated by IP (using anti-CYLD) and then fractionated in low-percentage (6%) SDS gels in order to separated the basal and phosphorylated CYLD bands. In lanes 11-13, the CYLD immune precipitates isolated form TNF-α stimulated cells (15 min) were either left on ice (NT) or incubated at 37°C for 30 min in calf intestinal alkaline phosphatase (CIP, lane 13) or buffer control (lane

12). The phosphorylated (P-CYLD) and basal (CYLD) forms of CYLD were detected by

IB using anti-CYLD (upper panel). The cell lysates were also subjected to IB to detect

IκBα degradation (lower panel). (B) Immunecomplex kinase assays to detect the activation of IKK. Jurkat cells were stimulated with PMA plus inomycin as described in

A. IKK complex was isolated by IP using anti-IKKγ antibody and subjected to kinase assays using GST-IκBα(1-54) as substrate. The phosphorylated substrate (P-GST-IκBα) is indicated. (C)-(E) CYLD phosphorylation (upper panel) and IκBα degradation (lower panel) were analyzed in TNF-α-stimulated 293 and HeLa cells and LPS-stimulated BJAB

B cells as described in A. 101 102

FIG. 3. CYLD phosphorylation is mediated by IKK. (A) Inducible phosphorylation of

CYLD requires IKKγ. Parental Jurkat cells, IKKγ-deficient Jurkat mutant (JM4.5.2), and

IKKγ-reconstituted JM4.5.2 cells were either not treated (-) or stimulated with PMA plus ionomycin for 15 min. Cell lysates were subjected to CYLD phosphorylation and IκBα degradation analyses as described in Fig. 2A. (B) CYLD phosphorylation by transfected

IKK. 293 cells were transfected with HA-tagged CYLD together with either empty vector or expression vectors encoding IKKβ (0.5 µg), IKKα (0.5 µg), or IKKβ (0.5 µg) plus IKKγ (25 ng). CYLD phosphorylation was analyzed by IB using anti-HA antibody.

(C) Phosphorylatio of CYLD truncation mutants. CYLD truncation mutants covering different lengths of its C-terminus (indicated by the amino acid numbers) were expressed in 293 cells either in the absence (-) or presence (+) of HA-IKKβ plus HA-IKKγ.

Phosphorylation of CYLD (upper panel) and expression of IKKβ (middle panel) and

IKKγ (bottom panel) were analyzed by IB using anti-HA. Phosphorylated CYLD bands are indicated by an arrowhead. (D) CYLD/IKKγ physical interaction. Full-length (FL) or truncated forms of CYLD (tagged with HA) were coexpressed with myc-tagged IKKγ.

The IKKγ complex was isolated by IP using anti-myc followed by detecting the associated CYLD proteins by IB using anti-HA HRP (upper panel). The expression level of IKKγ was analyzed by IB using anti-myc (lower panel). Lane 1 is a negative control that was transfected with CYLD only. (E) In vitro kinase assays to demonstrate CYLD phosphorylation by IKK homoenzyme. IKK holoenzyme was isolated by IP (using anti-

IKKγ) from untreated (NT), PMA/ionomycin-stimulated (7.5 min), or TNF-α-stimulated

(7.5 min) Jurkat cells and subjected to in vitro kinase assays using GST-IκBα(1-54) 103

(lower panel) or GST-CYLD(403-513) (upper panel) as substrates. (F) In vitro kinase assays were performed using recombinant IKKβ (25 ng) and GST-CYLD(403-513) or

GST substrate. GST-CYLD(403-513), but not GST, was phosphorylated by IKKβ. (G)

CYLD phosphorylation by recombinant IKKβ. In vitro kinase assays were performed using the indicated amounts of purified IKKβ recombinant protein and GST-IκBα(1-54)

(laned 1-5) or GST-CYLD(403-513) (lanes 6-10) substrate. Autophosphorylated IKKβ

(P-IKKβ) and phosphorylated substrates are indicated. 104 105

FIG. 4. Site-specific phosphorylation of CYLD by IKK. (A) Amino acid sequence of

CYLD phosphorylation region. The putative phosphorylation sites (serines) are bolded, and two serine pairs are indicated. (B) Sequence homology between the CYLD serine pairs and the IKK phosphorylation sites within IκBα and IκBβ. (C) Phosphorylation of

CYLD mutants by IKK holoenzyme (panel 1) and recombinant IKKs (panels 2 and 3).

GST-CYLD(403-513) with wildtype phosphorylation site (WT) or the various mutations

(indicated in A) were subjected to in vitro phosphorylation assays using IKK holoenzyme isolated from mitogen-stimulated Jurkat cells (panel 1), recombinant IKKβ (panel 2), or recombinant IKKα (panel 3). The substrate amounts were monitored by IB using anti-

GST (bottom panel). (D) In vivo phosphorylation of CYLD mutants by IKK. HA-tagged full-length CYLD, either wildtype or the indicated mutants, were expressed in 293 cells in the absence (-) or presence (+) of IKKβ plus IKKγ. The phosphorylation of CYLD was analyzed by IB using anti-HA. (E) CYLD knockdown by shRNA and reconstitution.

Jurkat and HeLa cells were infected with either the empty pSUPER retroviral vector

(lanes 1 and 5) or the same vector encoding CYLD-specific shRNA (shCYLD, lanes 2 and 6). The CYLD-knockdown cells were reconstituted by infection with retroviruses encoding RNAi-resistant wildtype CYLD (WTR) or its phosphorylation-deficient mutant

M4 (M4R). Expression of CYLD and the house-keeping protein tubulin was detected by

IB using anti-CYLD and anti-tubulin, respectively. (F) Phosphorylation of CYLD by cellular stimuli. The CYLD-knockdown Jurkat (Jurkat-shCYLD) and HeLa (HeLa- shCYLD) cells reconstituted with CYLD WTR or M4R were stimulated with mitogens or

TNF-α as indicated, and the phosphorylation of CYLD was analyzed as described in Fig.

2A. Wildtype CYLD, but not CYLD M4, was phosphorylated. 106 107

FIG. 5. Serine 418 of CYLD is phosphorylated in vivo. (A) Jurkat cells were either not treated (NT) or stimulated with PMA plus ionomycin for 10 min. Cell lysates were subjected to IB using either the regular anti-CYLD antibody or a phospho-specific anti-

CYLD antibody (αP-CYLD) that recognizes CYLD with phosphorylated serine 418. As negative control, an IB was performed using the preserum of αP-CYLD. (B) 293 cells were transfected with wildtype CYLD (WT) or CYLD S418A together with IKKγ and

IKKβ as described in Fig. 3D. Cell lysates were subjected to IB using either phospho- specific anti-CYLD (αP-CYLD) or regular anti-CYLD (αCYLD) antibodies. 108 109

FIG. 6. CYLD phosphorylation is required for signal-induced TRAF2 ubiquitination and optimal JNK activation. (A) Signal-induced TRAF2 ubiquitination requires CYLD phosphorylation. HeLa-shCYLD cells reconstituted with RNAi-resistant CYLD (WTR) or

M4 (M4R) were transfected with HA-tagged ubiquitin. The cells were either not treated

(NT) or stimulated with TNF-α. Ubiquitin-conjugated TRAF2 were isolated by IP using anti-TRAF2 followed by detection by IB using anti-HA-HRP (upper panel). The level of total ubiquitinated cellular proteins was analyzed by direct IB (lower panel). (B)

Phospho-mimetic CYLD loses TRAF2-deubiquitinating activity. 293 cells were transfected with TRAF2 together with empty vector or expression vectors encoding wildtype CYLD (WT), M4, or a phospho-mimetic CYLD harboring serine/glutamic acid substitutions at the phosphorylation sites (M4 S/E). Ubiquitin-conjugated TRAF2 was isolated by IP using anti-TRAF2 and detected by IB using anti-HA-HRP (panel 1). The expression of CYLD and TRAF2 proteins were monitored by IB using anti-CYLD

(middle panel) and anti-TRAF2 (bottom panel). (C) CoIP assays to detect the association of CYLD mutants with TRAF2. 293 cells were transfected with HA-tagged TRAF2 together with either an empty vector or expression vectors encoding HA-tagged wildtype

(WT) CYLD, CYLD M4, or CYLD M4 S/E. The CYLD complexes were isolated by IP using anti-CYLD followed by detecting the associated HA-TRAF2 by IB using HRP- conjugated anti-HA (upper panel). The protein expression level was monitored by direct

IB using HRP-conjugated anti-HA (lower panels). (D) Diminished activation of JNK in cells expressing phosphorylation-defective CYLD mutant. CYLD-knockdown HeLa

(HeLa-shCYLD) or Jurkat (Jurkat-shCYLD) cells were reconstituted with the RNAi- resistant form of wildtype CYLD (WTR) or CYLD M4 mutant (M4R). Following TNF-α 110 stimulation, JNK kinase activity and expression were determined by kinase assays (upper panel) and IB (lower panel), respectively. 111 112

FIG. 7. Signaling function of the phosphorylation-deficient mutant of CYLD. (A)

Luciferase reporter gene assays to determine NF-κB activation by CD40. CYLD- knockdown Jurkat cells reconstituted with RNAi-resistant form of wildtype CYLD

(WTR) or M4 mutant (M4R) were transfected with a κB-luciferase reporter (κB-TATA- luc) and a control Renilla luciferase reporter driven by the constitutive thymidine kinase promoter (pRL-TK). The cells were also transfected with either an empty vector (-) or a cDNA expression vector encoding human CD40 (+). After 40 hr of transfection, cell lysates were prepared and subjected to dual luciferase assays. The κB-specific luciferase activity was normalized based on the control Renilla luciferase activity and presented as fold induction relative to the basal level measured in cells transfected with empty vector.

The data are representative of two independent experiments. (B) RNase protection assay to analyze cellular genes regulated by CYLD. CYLD-knockdown HeLa cells reconstituted with CYLD WTR or M4R were either not treated (-) or stimulated with TNF-

α for 30 min (+). Total RNA was isolated and subjected to RPA. 113 114

Supplemental Fig. 1. Apoptosis induction by TNF-α. The indicated cells were incubated with cycloheximide (5 µg/ml) either in the absence (-) or presence (+) of TNF-α (20 ng/ml) for 22 hr. The cells were stained with Annexin V-Alexa 568 followed by FACS analysis of apoptotic cells (Annexin V stained cells). No difference in the level of apoptosis induction was detected among the different cells. 115 116

CHAPTER 4

REGULATION OF T CELL DEVELOPMENT AND ACTIVATION BY THE DEUBIQUITINATION ENZYME CYLD 117

Abstract

CYLD is a deubiquitinating enzyme (DUB) that has recently been shown to negatively regulate the activation of nuclear factor κB (NF-κB) and c-Jun N-terminal kinase (JNK) in cell lines. However, the physiological role of CYLD remains unknown.

We show here that CYLD plays a critical role in regulating T lymphocyte development and activation. Germ-line inactivation of the CYLD gene resulted in a significant decrease in the number of CD3+ thymocytes. Furthermore, the ability of T cells to transition from double-positive (CD4+CD8+) into single-positive lymphocytes was also attenuated. The defect in thymocyte development was correlated with a decreased number of peripheral T cells found within the spleen. However, the CYLD-deficient T cells efficiently responded to TCR stimulation and surprisingly produced considerably larger amounts of the cytokines that help to control immune cell fate. Preliminary results indicates that the hyperresponsive phenotype is associated with dephosphorylation of the transcription factor NFAT. Thus, the DUB CYLD is required for proper T cell development and subsequent activation, a function that may involve negative regulation of NF-κB, JNK, and probably also NFAT. 118

Introduction

The adaptive immune system is composed of two T cell subsets, CD4+ and CD8+, which play critical but distinct roles in mediating the host response against infectious pathogens. Through the T cell receptor (TCR), CD4+ and CD8+ T cells recognize specific peptides complexed with MHC class II and class I molecules, respectively, on the surface of mature antigen presenting cells (APC). Following initial proliferation, both CD4+ and

CD8+ T cells undergo differentiation to become specific effector T cells. The CD8+ T cells differentiate to cytotoxic T cells, which are characterized for their ability to lyse infected cells. Activated cytotoxic T cells mainly produce the cytokines interferon-γ and

IL-2 and lyse target cells through the secretion of perforin and granzyme [287]. The differentiation of CD4+ is more complex as it leads to generation of two T helper cell

subsets, TH1 or TH2. Polarization toward the TH1 or TH2 subset is controlled in part by antigen dose, source of costimulation, and, importantly, the cytokine environment. IL-12 and IFN-γ (produced by APCs and natural killer cells, respectively) drive cell cells to a

TH1 phenotype, whereas IL-4 (thought to be produced by mature T cells) drives a TH2 phenotype. TH1 cells enhance activation of cytotoxic T cells through the production of

IL-2 and mediate macrophage activation through secretion of IFN-γ and the

costimulatory action of CD40 ligand. TH2 cells secret a large variety of cytokines, such as IL-4, IL-5, IL-6, IL-10, and IL-13, which mediate B cell activation and differentiation

[288, 289].

Both the initial activation and subsequent differentiation of T cells are tightly regulated by intracellular signaling pathways, including those pathways leading to the activation of MAP kinases and IκB kinase (IKK). The MAPKs include extracellular 119 signal-regulated kinase (ERK), c-Jun amino terminal kinases (JNK) and p38 [290-292].

Upon activation, these kinases cooperatively activate a number of transcription factors including AP1 and CREB/ATF. IKK is the effector kinase that mediates activation of

NF-κB, a family of transcription factors that regulate many immune response genes.

Upon activation by immune stimuli, IKK phosphorylates the NF-κB inhibitor IκB, triggering IκB degradation and nuclear translocation of active NF-κB members.

Activation of T cells leads to the initial activation of the JNK1 isoform, whereas prolonged stimulation leads to the activation of JNK2 [293]. A critical role for JNKs in

T cell activation and differentiations has been demonstrated through gene targeting studies. However, although the ERK pathway is activated during T cell activation, inhibition of ERK through genetic disruption does not significantly impair T cell activation [294].

In immune regulation the IKK / NF-κB pathway has many functions including development, activation, and peripheral maintenance of lymphocytes [14, 295-298].

Genetic disruption of any of the NF-κB family members or any of the IKK subunits in mice causes severe defects in the development of function of the immune system [299].

The majority of defects described to date arise in B cell maturation and activation, germinal center formation, and formation of lymphoid architecture [13-15]. To date, a few studies have explored the role of NF-κB family members in T cell development and activation. However, these initial reports have noted that T cells from most NF-κB knock-out mice do have defects in proliferative responses upon stimulation. Whereas

NF-κB-deficient mice display strong B cell phenotypes, mice lacking JNK1 or JNK2 show severe abnormalities in T cells. Disruption of the JNK1 or JNK2 isoforms leads to 120 no significant defect in the development of B or T cells. However, peripheral T cells derived from JNK2 -/- mice exhibit reduced production of certain cytokines, including

IL-2, IL-4, and IFNγ. Of note, the thymocyte population in these mice are resistant to apoptosis either through the normal process of negative selection or after anti CD3 antibody is injected into the mice. Thus, JNK plays an important role in CD4+ T cell differentiation and in thymocyte apoptosis. T cells from JNK1 -/- mice hyperproliferate

upon induction with various stimuli and display a preferential polarization into TH2 cells, as demonstrated by production of TH2 type cytokines (IL-4, IL-5, and IL-10). These in vivo studies together with a large body of in vitro work have clearly demonstrated that

JNK1 and JNK2 are necessary factors in regulating the apoptosis of thymocytes and the

differentiation of peripheral T cells towards the TH1 subset.

Despite the extensive studies on the function of JNK and IKK, our understanding of the regulation of these effector kinases remains far from complete. Particularly, little is known how these potent signaling factors are negatively regulated to prevent uncontrolled T cell activation. Nevertheless, recent work by us and others identified a negative regulator of IKK and JNK. This factor, CYLD, is a new member of the deubiquitinating enzyme (DUB) family. CYLD was originally identified as a protein that is mutated in patients with familial cylindromatosis, a rare tumor of the head and scalp.

Transfection studies suggest that CYLD inhibits the ubiquitination of specific signaling molecules, including TRAF2, TRAF6, and the IKK regulatory subunit, IKKγ [204-206].

In contrast to the classical type of ubiquitination, which occurs through lysine-48 linkage of ubiquitin molecules, the ubiquitination of these signaling molecules is mediated through lysine-63 linkages. This novel type of ubiquitination does not target proteins for 121 degradation, but mediates activation of kinases such as IKK. A further role for CYLD in the regulation of IKK is its ability to negatively regulate NF-κB reporter gene activity.

By RNA interference (RNAi)-mediated CYLD knock down, we demonstrated that

CYLD regulates both the IKK and JNK signaling pathways in a cell type and inducer specific manner [300]. Further, the DUB activity and negative signaling function of

CYLD are counter-regulated through its site-specific phosphorylation by IKK [301].

In order to study the physiological functions of CYLD we generated - defecient mice. The CYLD homozygous knockout (CYLD -/-) mice were born with expected Mendelian ratios and developed normally. However, these knock-out mice displayed significant abnormalities in the immune system. The spleen of CYLD -/- mice contained reduced numbers of T cells and increased number of B cells and macrophages.

Interestingly the CYLD -/- T cells retained the ability to respond to TCR stimulation, as demonstrated by their proliferation and cytokine production. In fact, the mutant T cells produced markedly more cytokines than the wild-type T cells. The peripheral T cells hyperplasia was associated with a developmental defect in the thymus. Thymocytes displayed a decreased number of CD3+ cells, and further defects were noted in the transition from double-positive to single-positive mature T cells. Taken together, these results indicate that CYLD plays an essential role in thymocyte development and negatively regulates peripheral T-cell activation.

Methods

Generation of CYLD-/- Mice. Mouse CYLD gene was isolated by PCR using Takara LA

Polymerase and genomic DNA from the C57BL6 derivative 129/SVJ mice. The PCR primers were designed to amplify two CYLD gene fragments: a 5 kB fragement, covering 122 the upstream region and part of exon 1, and a 2.75 kB fragment, covering the 3’ half of exon 1 and the entire exon 2. These fragments were cloned into the targeting vector pPNT (provided by Dr Thomas Saunders, University of Michigan), with the 5 kB and

2.75 kB arms inserted upstream and downstream of the neomycin gene, respectively.

The resulting targeting vector was linearized with NotI and electroporated into the mouse embryonic stem cell line R1. Genomic DNA was isolated from G418-resistant ES cell clones and subjected to Southern blot analysis using the probe indicated in Figure 1A.

The Southern probe, amplified by PCR, covers a region in the CYLD genomic sequence external to the targeting construct. After hybridization, this probe detected a 16.7 kB

BglII fragment of the wild-type CYLD gene and a 11.5 kB fragment of the targeted allele. Targeted ES cells were injected into blastocysts generated from breeding

C57BL/6 females to C57BL/6 X BBA/2 F1 male mice. Chimaeric male mice were crossed with C57BL/6 females to achieve germline transmission. After heterozygous matings, CYLD+/+, CYLD+/- and CYLD-/- mice were identified by Southern blot analysis. In all experiments, only littermate mice were used. All mice were maintained at the animal facilities of The Pennsylvania State University College of Medicine under specific pathogen-free conditions according to institutional guidelines.

Western Blots. Cell lysates were prepared by lysing the cells with RIPA lysis buffer and immediately subjected to IB assay as described previously [242]. The expression of

CYLD and tubulin was measured by immunoblot analysis using anti-CYLD [301] and anti-tubulin (TU-02) (Santa Cruz), respectively. 123

Immunocytometry. Single-cell suspensions of thymi and spleens were obtained from

CYLD +/+ and CYLD-/- mice. Spleen and thymi were removed, homogenized using a 7 mL tissue homogenizer (Wheaton), and spun over lymphocyte separation medium

(Cambrex) to yield mononuclear cells. These cells were stained with FITC-, PE- or

PE.CY5-conjugated antibodies reactive to CD3ε, CD4, CD8, CD25, CD95 (Fas), CD44,

CD62L, CD69, MHC class II (I-Ab), CD23, CD19, GR1 (PharMingen) and F4-80

(SerioTech). For the analysis of thymocyte precursors, cells were triple stained with

Alexa488 CD8, PE-CD3ε, PE.CY5.5-CD4. All samples were analyzed by flow cytometry using a FACScan (Becton Dickinson).

Generation of Bone Marrow-Derived Macrophages. BMDM were prepared as previously described [302]. Briefly, bone marrow was flushed from femurs of mice and plated to nontissue culture-treated petridishes (100 mm) in DMEM media supplemented with 20% fetal bovine serum, 30% L929 cell-conditioned media, and antibiotics. The growth media were changed every 2 days until the macrophages grew to confluence. The cells were lifted with 10 mM EDTA prepared in PBS supplemented with 20% fetal bovine serum and replated for further expansion or experiments.

Lymphocyte proliferation and cytokine production. 1X105 purified T cells, isolated using

MACS separation columns and anti-CD90 coated magnetic beads (Miltenyi Biotech), were placed into 96-well plates in IMDM (10% FCS + β-mercaptoethanol). T cells were stimulated with PMA (Sigma) plus ionomycin A23617, anti-CD3ε (clone 145-2C11, hamster IgG), anti-CD28 (clone 37.51, PharMingen), or ConA (Pharmacia). Cells were stimulated in five replicates for different time periods and pulsed for the last 4 h with 124

1µCi per well [3H]-thymidine (PerkinElmer Life Science). [3H]-thymidine incorporation was measured using a beta-scintillation counter (Coulter). To determine cytokine production, cell culture supernatants were assayed in triplicate for the production of IL-2,

IFNγ, IL-4, IL-10 using Ready-Set-Go ELISA (e-bioscience).

Results

Generation of CYLD Deficient Mice. To determine the physiological role of

CYLD within the immune system we generated knock-out mice. The CYLD gene was inactivated by homologous recombination using a targeting construct that deleted 138 nucleotides of exon 1 and replaced it with the neomycin cassette (Figure 1A). The knock-out construct was transfected into ES cells and clones were screened for correct homologous recombination. Chimeras were generated by injection of ES cells in

C57BL/6 blastocysts. Heterozygotes (+/-) were intercrossed to generate homozygous mutant mice (-/-), identified by Southern blot analysis of genomic DNA (Figure 1B).

Homozygous mutants were viable and born at the expected Mendelian ratios. Expression of CYLD protein was examined by Western Blot analysis of total splenocytes, bone marrow derived macrophages, B cells, and T cells from wild-type and CYLD knock-out mice (Figure 1C). No truncated CYLD polypeptide reactive with our antibody was detected (data not shown). These results indicate that the targeted mutation of the CYLD gene results in the absence of a detectable gene product.

CYLD Knock-out Mice Display Decreased Numbers of Splenic T cells. Prior work on CYLD in cell lines implicated it in negatively regulating JNK and NF-κB signaling, two pathways that are strongly tied to immune function. As JNK knock-out T cells display a protection from apoptosis and NF-κB knock-out B cells have perturbed 125 functions, we began our examination of the CYLD immune system by characterizing the cell populations found in the spleen. Spleen size, weight, and lymphocyte numbers

(Figure 2C) were similar among male and female littermate mate mice. Staining for cell surface markers revealed a marked decrease in CD3+ T cells in the spleens of CYLD -/- mice (Figure 2A). Typically, there was a 45% decrease in splenic T cell numbers (Figure

2E), however this decrease in cell number was not due to a loss of expression of the T cell receptor as visualized by the intensity of CD3 staining (Figure 2A). Since there are two major populations of T cells within the spleen, CD4+ and CD8+ cells, we questioned if the dramatic loss of T cells was specific to one or the other population. Interestingly, the relative decrease in cell number between the two populations of T cells showed a similar reduction, 50% decrease in CD4+ and 40% decrease in CD8+ cells (Figure 2F).

Next we examined whether the decrease in splenic T cell number was due to the sole loss of T cells or caused by the exclusion of T cells from the spleen by another cell type.

Concurrent with the loss of splenic T cells in CYLD-/- mice, there was also a 23% increase in the number of CD19+ B cells (Figure 2B and Figure 2D). Subsequently, all populations within the spleens of CYLD -/- mice experienced an increase in the percentage and number of cells as measured by surface staining of GR1+ (neutrophils, eosinophils, monocytes), F4-80+ (macrophages), and CD19+ (B) cells (Figure 2G and

2D, respectively). Since there was not an increase in one specific cell population within the spleens of CYLD -/- mice, a compensatory mechanism is likely occurring whereby other populations have migrated into the spleen to replace the loss of T cells. These data show that CYLD -/- mice display a dramatic decrease in the number of T cells with concurrent increases in other cell populations. 126

CYLD -/- splenic naïve T cells are greatly diminished. Upon activation of T cells, the expression of numerous cell surface markers change. The change of cell surface markers allows T cells to begin the maturation process and to home to the peripheral site of infection. To further understand the diminished number of T cells, we examined whether the T cells within the splenocytes were of a naïve phenotype and or an activated phenotype. I analyzed the surface expression of several markers found on activated T cells: CD62L (L-selectin), CD25 (IL2α chain), CD44, and CD69. As expected, the majority of T cells from wild-type mice were CD62L high, suggestive that these cells were of a naïve phenotype (Figure 3A). In contrast the CD62L high cell population is drastically reduced within the CYLD -/- T cells. However, the reduction in the naïve T cell population was not associated with an appreciable increase in the number of CD62L low (activated/memory) cells (Figure 3).

Since activated T cells typically migrate to the periphery, we reasoned that the loss of naïve T cells in the spleen could be due to their spontaneous activation and migration into the peripheral tissues. To asses this possibility, we examined the T cell content in peripheral blood. Peripheral blood from CYLD -/- mice had a reduction in the percentage of CD4+ T cells and a slightly higher percentage of CD8+ cells, as compared to the wild-type (Figure 3B). The mutant mice also contained a significant pool of double-positive T cells in the periphery. This elevated population of double-positive cells is extremely interesting and strongly suggests a defect in lymphocyte development.

CYLD is Critical for the Normal Development of T lymphocytes. T cell development starts with the influx of cells from the bone marrow into the thymus by chemokines-mediated migration. Once in the thymus, the T cells must undergo a series 127 of selection steps, which include positive selection of the cells with correctly rearranged

TCRs followed by negative selection to delete the autoreactive T lymphocytes. The events of positive and negative selection correlate with the surface expression of the CD4 and CD8 molecules. CD4-CD8- double negative T cells first undergo positive selection whereby they upregulate the surface expression of CD4/CD8 molecules and become the

CD4+CD8+ double-positive cells. Double-positive thymocytes then undergo a negative selection step and become single-positive for either the CD4 or CD8 lineage. The reduction in T cell number in CYLD -/- spleen and peripheral blood prompted us to examine whether the loss of CYLD is associated with a defect in T cell development through the surface staining of CD3, CD4, and CD8.

Since peripheral numbers of T cells were not sufficiently increased in CYLD -/- mice we began to question if the decreased numbers of T cells was due to some defect during thymic development. Since CYLD was shown to be a negative regulator of various signaling pathways I reasoned that the process of thymic selection would be impaired in CYLD -/- mice. To examine if there was a defect in a specific subpopulation in CYLD -/- mice, thymuses were isolated from either four week- or six week-old pups and stained for the surface markers CD3, CD4, and CD8 and examined by flow cytometry. Wild-type mice displayed normal numbers of CD3high thymocytes (Figure

4A), double-negative, double-positive, and single-positive subpopulations (Figure 4B).

CYLD -/- mice displayed a greatly reduced number of CD3high thymocytes (Figure 4C).

The apparent reduction in CD3high cells within the thymus of CYLD -/- mice suggests a defect in thymocyte development. Despite the reduction in CD3high thymocytes in CYLD

-/- mice, the total number of thymocytes of the CYLD -/- mice was comparable to that of 128 the control mice (data not shown). However, analysis of the subpopulations of CYLD -/-

+ + thymocytes revealed a defect in the transition from the double-positive CD4 CD8 to the single-positive CD4+ cells (Figure 4D). The relative percentages of CD8+ cells were unaffected. However, based on the decreased number of CD3high cells in the mutant mice the relative number of both CD4+ and CD8+ single-positive T cells were reduced. These data suggest that CYLD is required for the transition from DP to single-positive thymocytes, with CD4+ cells being affect more severely than CD8+ cells.

CYLD T Cells Maintain Effector Function with Hyperproduction of Effector

Cytokines. Having observed an apparent defect in the negative selection of CYLD -/- thymocytes, we began to examine whether the CYLD -/- T cells remains functional in signal-induced activation. Since, negative selection in the thymus allows for the selective removal of cells that recognize self antigens through induction of either apoptosis or functional inactivated (anergy), surviving CYLD -/- T cells could exhibit an unresponsive or anergic phenotype. To test this hypothesis we examined if CYLD -/- T cells could proliferate and produce various effector cytokines

CYLD -/- T cells were isolated from the spleen and seeded at equal densities and assayed for their proliferative ability in a series of proliferation assays. As expected, the control T cells were activated to proliferate when stimulated with anti-CD3 (Figure 3A, open bars). CD28 costimulation did not further enhance the proliferation, probably because of the high dose (10 µg/ml) of anti-CD3 antibody used. The CYLD -/- T cells exhibited a comparable level of proliferation during the first 48 hours of stimulation.

However, by 72 hours the proliferative ability of CYLD -/- T cells had diminished

(Figure 3B). These results indicate that the loss of CYLD does not affect the initial phase 129 of T cell activation, thus excluding the possibility of anergy induction. It remains unclear why the CYLD -/- T cells become hypoproliferative after prolonged stimulation.

We further examined the function of the CYLD -/- T cells by analyzing their cytokine production. Activated T cells initially produce IL-2, which is essential for T cell proliferation, but subsequently differentiate to gain the capacity to produce other cytokines. Differentiated CD8+ T cells (cytotoxic T cells) typically secrete IFNγ and IL-

2. Depending on the nature of antigens and the microenvironment, CD4 T cells

differentiate into either TH1 or TH2 cells, producing different types of cytokines. TH1 cells produce IFNγ and IL-2, whereas TH2 cells produce IL-4, IL-5, and IL-13. Cytotoxic

T cells and TH1 cells are responsible for the clearance of pathogens through direct cell

lysis and activation of macrophages. TH2 cells drive hummoral immune responses against extracellular pathogens through B cell costimulation and activation. In line with their competence in the initial stages of proliferative response, the CYLD -/- T cells were functional in producing both IL-2 and the effector cytokines.

Astonishingly, CYLD -/- T cells actually produced considerably larger amounts of cytokines than the wild-type T cells (Figure 5B). Since there is an upregulation of

both TH1 and TH2 cytokines, CYLD T cells probably have the potential to be polarized into either of the T helper populations, at least in vitro. The loss of CYLD did not alter the kinetics of cytokine induction. Taken together these results demonstrate that though the number of peripheral T cells is reduced in CYLD -/- mice, the mutant T cells are likely to be hyper-responsive in their production of both primary and effector cytokines.

Accordingly CYLD appears to be a negative regulator of cytokine induction that leads to

T cell activation. 130

CYLD is a negative regulator of the transcription factor NFAT. Efficient production of cytokines in T cells is dependent upon the transcription factor NFAT.

NFAT protein function is regulated at the level of phosphorylation, and upon T cell receptor stimulation NFAT is dephosphorylated and quickly migrates into the nucleus where it mediates transcription of target genes. Recent reports suggested that the duration and extent of NFAT activation are critical parameters that significantly affect T cell cytokine gene transcription patterns. Due to the extraordinarily high levels of cytokines that CYLD -/- T cells produced in vitro we questioned if CYLD, a known negative regulator of the NF-κB and JNK signaling pathways, could regulate NFAT activation. T cells were purified from splenocytes and stimulated with the mitogen PMA plus ionomycin or with CD3 + CD28. In wild-type nontreated T cells NFAT was found primarily in its phosphorylated state (Figure 5C lane 1). Upon stimulation of the cells

NFAT was converted by dephosphorylation into different isoforms, as expected for transcriptionally active proteins (Figure 5C lane 2). CYLD -/- nonstimulated T cells displayed a dramatic reduction in the phosphorylated form of NFAT (Figure 5C lane 4).

Stimulation of CYLD -/- T cells led to further reduction in the phosphorylation state of

NFAT compared to wild-type (compare Figure 5C lanes 2 to lane 5 and lane 3 to lane 6).

These data demonstrate that CYLD -/- T cells have the NFAT profile characteristic of increased effector function due to reduced levels of phosphorylation of the transcription factor.

Discussion

By generating CYLD -/- mice, we have identified a critical role for the DUB

CYLD in regulating T-cell development and activation. The CYLD -/- mice had a 131 greatly decreased number of T cells within the spleen and a concurrent increase in the other white blood cell populations including B cells, macrophages and granulocytes.

Both CD4+ and CD8= cell number was reduced suggesting that lineage development towards CD4+ and CD8+ cells is unlikely to be dependent on CYLD. It is currently unclear whether the increase in B cells and myeloid cells (granulocytes and macrophages) was compensation for the loss of T cells, or because of loss of negative-regulatory function of CYLD.

Our data suggest several possibilities as to how the loss of CYLD could cause reduction in T cell number. The most likely explanation is that the low number of T cells in the spleen is due to a defect in the development or maturation of T cells in the thymus.

+ Indeed, as in the spleen, the number of CD3 T cells was significantly reduced in the thymus. The process of positive and negative selection of T cells in the thymus requires engagement of the newly rearranged T cell receptor by MHC class I and class II molecules expressed on bone marrow derived antigen presenting cells (macrophages and

DC) and stromal cells of the thymus. The strength of the TCR signal is instrumental in the elimination or selection of a T cell during negative selection. For example, a strong signal is required for eliminating those T cells with autoreactivity. As CYLD is a negative regulator of the NF-κB and JNK signaling pathways, the loss of CYLD may result in elevated TCR signaling, leading to increased cell loss during the negative selection step. This hypothesis is consistent with the phenotype of JNK -/- mice, whose thymocytes are resistant to apoptosis during negative selection, even when mice are administered a strong agonist anti-CD3 [303]. In further support of this idea, the numbers of CD4+/CD3+ thymocytes is significantly reduced in the thymus of CYLD -/- 132 mice, where they are increased in JNK -/- mice. Therefore it will also be of interest to examine the MAPKs activation status during negative selection. Freshly isolated thymocytes will be stimulated with anti-CD3 and assayed for their phosphorylation status through flow cytometry and western blots. This would allow me to characterize not only if there is a difference in the activation status of various MAPKs in CYLD -/- mice but would also pinpoint which thymocyte populations are being affected during selection.

Alternatively the CYLD -/- T cells may be spontaneously activated and lost through activation-induced cell death (AICD). In support of this hypothesis, our preliminary data reveal that the transcription factor NFAT was constitutively dephosphorylated (or activated) in CYLD -/- T cells (Figure 5B). To test this hypothesis both peripheral and thymic T cells will be stimulated with anti-CD3 and assayed for their apoptosis status via Annexin V and propidium iodide. Consistent with this hypothesis, the CD62L high naïve T cell population was missing in the spleens of CYLD -/- mice

(Figure 3). However, parallel staining for the activation markers CD69, CD25 and CD44 did not reveal an accumulation of the activated T cells. It is unlikely that excess activated cells homed to peripheral blood in CYLD -/- mice, since the number of T cells in the peripheral blood was not elevated in the mutant mice. There could also exist another possibility that the decreased numbers of T cells could be due to some inherent defect during early stages of thymic development. While there could be a normal number of thymic precursors entering the thymus there could exist proliferation defects that lead to a decreased number of cells undergoing positive and negative selection. Therefore, it will be of interest to examine various early thymocyte surface markers, CD44, CD25, TCRβ, preTCRα, etc…, and characterized the different subpopulations of cells in the thymus. 133

Our in vitro T cell activation studies suggest a negative role for CYLD in regulating peripheral T cell activation in response to TCR stimulation. The CYLD -/- T cells supernatant contained remarkably larger amounts of cytokines than the control T cells. However, it is unclear whether there is more CYLD -/- T cells secreting the various cytokines or if the cells are producing more cytokines. Further investigation need to examine the intracellular levels of cytokine to answer this question. Interesting, however, the mutant T cells did not exhibit a hyperproliferative phenotype. In fact, these cells become hypoproliferative at late times after stimulation. This proliferative decrease could be due to a number of factors. First, the CYLD -/- T cells may be more sensitive to

AICD. As alluded to above, the CYLD -/- T cells seem to be partially activated, as suggested by the constitutive NFAT dephosphorylation. Since FasL is upregulated on activated T cells and is critical for the induction of AICD, it is conceivable that the activated CYLD -/- T cells may have elevated FasL expression and increased AICD during the in vitro stimulation.

A second potential reason for the lack of a correlation between proliferation and

IL-2 production in CYLD -/- T cells is that the loss of CYLD may have an anti- proliferative effect. In this regard, some NFAT members (NFAT1 and NFAT4) are known to both promote AICD and suppress T cell proliferation. Future studies will confirm whether CYLD indeed negatively regulates NFAT. Further, cell replication analysis using CFSE will directly examine whether the CYLD -/- T cells have a specific defect in cell division.

In summary, our data have clearly defined two important roles for CYLD in regulating T cell function. First, CYLD is required for proper development of T cells in 134 thymus. Second, CYLD negatively regulates the activation of peripheral T cells.

Additionally, the hyperproduction of effector cytokines by CYLD -/- T cells also has implications for the proper regulation of T cell differentiation and function. Together, these results define an important role for CYLD in immune regulation within T cells and provide an insight into the physiological functions of the protein.

Acknowledgments We thank Dr. C. Norbury and Dr. T. Schell for the sharing of reagents and ideas.

We also thank Nate Schafer of the Core Facility of Hershey Medical Center for his technical assistance during flow cytometry. As always we thank the fellow members of the Sun laboratory for their continual support and helpful discussions. 135

Figure 1. Gene targeting of CYLD. (A) Strategy for the disruption of the CYLD gene by homologous recombination. The genomic organization of the wildtype CYLD gene

(top), targeting vector (middle) and the interrupted CYLD gene (bottom) are schematically presented. The CYLD exons are diagramed as filled boxes and the neomycin-resistance gene and the herpes simplex virus thymidine kinase gene are shown as open boxes. The probe used for Southern blot analysis is indicated. (B) Genomic

DNA was isolated from CYLD +/+, CYLD +/- and CYLD-/- mice, digested with BglII, and analyzed by Southern blotting using the 5’ flanking probe shown in Fig. 1A. The

16.5 kB and 11.5 kB band corresponds to the wild-type and mutant alleles, respectfully.

(C) Western blot analysis of CYLD protein expression in wild-type and CYLD -/- splenocytes, purified T cells, B cells and primary bone marrow-derived macrophages.

Whole-cell lysates were separated by SDS-PAGE and subjected to immunoblotting with an anti-CYLD antibody (described earlier in Chapter 2) or anti-tubulin (loading control). 136 137

Figure 2. Reduction in T Cell numbers in CYLD -/- Mice. Splenocytes from wild- type and CYLD -/- mice were stained for CD3, CD4, CD8, CD19, GR1 and F4-80.

Graphs depict the average cell numbers from seven wild-type or CYLD -/- mice. (A)

FACS profiles showing the number of stained CD3+ T cells in the wild-type and mutant splenic population. (B) Total spleen cells were isolated from wild-type or CYLD -/- mice and stained with anti-PE-CD3ε and anti-FITC-CD19. The populations of B cells

(CD19+CD3-), and T cells (CD19-CD3+) are highlighted. The CD19-CD3- population represents macrophages, neutrophils, eosinophils, and mononuclear cells of the spleen.

(C) Wild-type or CYLD -/- splenocytes stained with trypan blue were counted on a hemocytometer. (D) and (E) Wild-type or CYLD -/- splenocytes were stained with anti-

CD19 (D) or anti-CD3ε (E). (F) Wild-type or CYLD -/- splenocytes were stained with anti-CD3ε, the CD3+ cells were gated, followed by analyzing the number of CD4+ and

CD8+ cells within this gated T cell population. (G) Wild-type or CYLD -/- splenocytes were stained with and anti-PE.CY5-GR1 or anti-PE.CY5-F-4-80, and the stained cells were analyzed by FACS. 138 139 140

Figure 3. Loss of Naïve T cells in the Spleen of CYLD -/- mice. Splenocytes from wild-type and CYLD -/- mice were isolated and stained for expression of CD3, CD62L,

CD69, CD25 and CD44 and analyzed by flow cytometry. Facs profiles represents, equal number of forward and side scatter events, gated on the CD62L, CD69, CD25, or the

CD44 expressing CD3+ T cells. Expression of CD62L and CD44 is high on naïve cells and low on activated T cells, whereas the expression of CD69 and CD25 is low on naïve cells and high on activated T cells. 141 142

Figure 4. CYLD -/- Mice Display Defects in Thymocyte Development. Thymocytes were isolated from six age-matched (four-week old) wild-type and CYLD -/- mice and stained for the cell surface markers CD3, CD4, and CD8. (A,B) Representative histogram plots showing the CD3 staining pattern of wild-type (A) or CYLD -/- (B) splenocytes. Percentage of CD3+ cells represents the average from six mice. (C,D)

FACS profile showing the CD4 and CD8 staining of CD3+ gated thymocytes from wild- type (C) and CYLD -/- (D) mice. The percentage of mature CD4 single-positive

(CD4+CD8-), CD8 single-positive (CD4-CD8+), and double-negative (CD4-CD8-) cells is indicated. The data represent the average. 143 144

Figure 5. T cell Activation as Measured by Proliferation and Cytokine Production.

Splenocytes were isolated from 8-week old wild-type or CYLD -/- mice. T cells, were purified using MACS column separation and anti-CD90 conjugated magnetic beads.

1X105 purified T cells were stimulated with plate bound anti-CD3ε (10µg/ml) together with soluble anti-CD28 (1 µg/ml), PMA (10 ng/ml) plus ionomycin (0.5 µM) or

Concanavalin A (2.5 µg/ml). At the indicated times following stimulation, the cells were pulsed with 3H-thymidine for a proliferation assay (A) and the cell culture supernatants were analyzed by ELISA to measure the production of IL-2, IFNγ, IL-4, and IL-10 (B).

(C) T cells were purified from four wild-type or CYLD -/- mice and stimulated with

PMA plus ionomycin (Lanes 2 and 5) or anti-CD3 plus anti-CD28 (Lanes 3 and 6).

Extracts were run on a gel and blotted with anti-NFAT1 antibody. 145 146

CHAPTER 5

OVERVIEW AND DISCUSSION 147

5.1 Overview of Major Findings

The primary goal of my thesis research was to understand the signaling function and mechanism of regulation of the DUB molecule CYLD. Therefore my hypothesis was that CYLD, a binding partner of IKK, could negatively regulate the signaling of IKK as well as other signal transduction pathways. To this end, I utilized both in vitro biochemical assays and in vivo immunological analysis to investigate understand the physiological function of CYLD and to elucidate the biochemical mechanism by which the function of CYLD is controlled. The major findings of my thesis research are summarized below:

(1) CYLD undergoes rapid phosphorylation both in cell lines and primary cells in

response to stimulation by a variety of stimuli. The phosphorylation of CYLD

appears to involve the IKK complex, as phosphorylation is blocked in cells

lacking the IKK regulatory subunits, IKKγ.

(2) CYLD phosphorylation occurs in a region that contains amino acid 403-447.

Substitution of the serines (seven) with alanines abrogates the inducible

phosphorylation of CYLD.

(3) Phosphorylation of CYLD serves as a mechanism that negatively regulates its

DUB activity. CYLD has constitutative TRAF2-deubiquitinating activity,

which is inactivated upon substitution of its phosphorylation sites (serines)

with a phospho-mimetic residue (glutamic acid). Consistently, a

phosphorylation-defective form of CYLD (harboring serine to alanine

substitutions) blocks the inducible ubiquitination of TRAF2. 148

(4) CYLD negatively regulates both the JNK and NF-κB, but the NF-κB

regulatory function is receptor- and inducer-specific.

(5) CYLD regulates both the development and activation of T cells. Naïve T cell

numbers within the spleens of CYLD -/- mice are greatly reduced. This

peripheral defect correlates with a decreased number of mature T cells in the

thymus, as well as impaired maturation of thymocytes from double-positive to

single-positive stages. Despite their reduced number, the CYLD -/- T cells are

hyperresponsive to TCR stimulation, as demonstrated by their production of

higher amounts of cytokines.

5.2 Discussion

The findings summarized above establish CYLD as a DUB that regulates signal transduction in the immune system and suggest an intriguing mechanism that controls its activity. However, a number of issues remain incompletely understood and will be discussed below.

5.2.1. CYLD phosphorylation: involvement of IKK and another kinase

I showed that CYLD phosphorylation requires the IKK regulatory subunit IKKγ.

However, it is unclear whether IKKγ functions through the known IKK catalytic subunits or participates in the regulation of a novel CYLD kinase. Although IKKα and IKKβ both phosphorylate CYLD in vitro, it remains to be determined whether one or both of these IKK catalytic subunits are indeed required for signal-induced CYLD phosphorylation. I have attempted to address this question using mouse embryonic fibroblasts (MEFs) lacking individual IKK subunits. Unfortunately, the CYLD phosphorylation does not occur in wild-type MEFs when stimulated with TNF-α. It is 149 currently unclear whether the MEFs lack an accessory protein required for IKK-mediated

CYLD phosphorylation, or these cells lack an unidentified IKKγ-regulated kinase. One alternative approach to examine the role of IKK catalytic subunits in CYLD phosphorylation is RNAi-mediated gene knockdown. By knocking down IKKα, IKKβ, or both IKK subunits in cell lines, we can determine whether the classical IKK is indeed required for the inducible phosphorylation of CYLD.

Phosphorylation site mapping studies have yielded yet another level of complexity of CYLD phosphorylation. Examination of CYLD sequence revealed two sites that appeared to be similar to consensus IKK phosphorylation sites. Mutation of these serines individually, or in combination, partially ablates phosphorylation.

Interestingly, only the loss of all seven serines resulted in the complete ablation of phosphorylation, based on loss of the slower migrating band as seen in western blots and in vitro kinase assays. Whether all seven serines become phosphorylated in vivo is under further investigation. Recently, we generated a phospho-specific antibody that can detect

CYLD when it is phosphorylated at serine 418. This antibody is extremely sensitive and specifically detects phosphorylated CYLD in both human and mouse cell lines.

Surprisingly, the phosphorylation of serine 418 was detected when Jurkat cells were stimulated with PMA and ionomycin but not with TNFα. On the other hand, when the same cell extracts were blotted with pan-CYLD antibody, the phosphorylation (as visualized by its band shift) was detected in both types of stimulated cells. This result strongly suggests that different inducers phosphorylate distinct sites on CYLD. This idea further implies that more than one kinase is involved in CYLD phosphorylation. In support of this hypothesis, the phosphorylation of serine 418 was independent of IKKγ, 150 as it could be detected in the IKKγ-deficient Jurkat cell line JM4.5.2. This finding suggests that CYLD phosphorylation may involve IKK and another kinase, which is induced by different stimuli. Clearly, more work needs to be performed to understand the mechanism of CYLD phosphorylation.

5.2.2 Post-translational Modification of CYLD: A Mechanism of Inhibition

I showed that CYLD is a constitutive DUB that suppresses the ubiquitination of

TRAF2, since RNAi-mediated CYLD knockdown was sufficient for triggering TRAF2 ubiquitination. This finding raises a question regarding how cellular signals induce

TRAF2 ubiquitination in CYLD-expressing cells. One possibility is that the signals upregulate TRAF2 ubiquitination activity, thus overriding the DUB function of CYLD.

However, this possibility is unlikely, since stimulation of CYLD-deficient cells with

TNF-α does not further enhance the constitutive ubiquitination of TRAF2. A more likely mechanism is that the cellular stimuli trigger CYLD modification, thereby temporarily inactivating its DUB activity. This model is supported by our finding that transient phosphorylation of CYLD accompanies the induction of TRAF2 ubiquitination in TNF-α stimulated cells. Interestingly, when the endogenous CYLD was replaced with a phosphorylation-defective CYLD mutant, the inducible ubiquitination of TRAF2 was largely blocked, and activation of JNK was diminished. Thus, it seems likely that CYLD phosphorylation is a posttranslational mechanism to temporarily inactivate its DUB activity, allowing new TRAF2 ubiquitination and activation of downstream signaling pathways. To my knowledge, this is the first observation of a posttranslational modification by phosphorylation in the regulation of the function of a DUB. 151

Figure 5.1 Signal induced phosphorylation and subsequent inactivation of the deubiquitinating activity of CYLD. Induction of the cell with various immune stimuli leads to the phosphorylation of CYLD by IKK. Phosphorylation leads to the inactivation of CYLD deubiquitinating activity allowing for TRAF2 to become ubiquitinated.

Ubiquitinated TRAF2 allows for the activation of down stream kinases such as

MEKK1/3 and TAK1 leading to the activation of various MAPKs. 152 153

5.2.3 CYLD in T cells: a Positive or Negative Regulator?

My studies using CYLD -/- mice demonstrate that loss of CYLD leads to the dramatic loss of naïve T cells in both spleen and thymus. This finding suggests, that

CYLD plays an essential role in T cell development. On the other hand, despite their apparent defect in development, the CYLD-/- T cells were competent in responding to

TCR signal, as demonstrated by their proliferation and production of cytokines. In fact, the CYLD -/- T cells were hyperresponsive to anti-CD3/anti-CD28 stimulation, producing markedly more cytokines that the control T cells. However, whether there are just more CYLD -/- T cells producing cytokines or if there is actually an increase in the production of cytokines needs to be examined using intracellular cytokines staining. This finding suggests that CYLD functions as a negative regulator of T-cell activation, although it is required for T-cell development. Precisely how CYLD play such a dual role, during development and during its activation by APC, in T cells requires additional studies. My hypothesis is that both functions of CYLD are dependent on its negative signaling function downstream of the TCR. As discussed in earlier sections, the defect in

T-cell development in CYLD-/- mice appears to be at the step of negative selection. This step of T-cell development involves elimination of auto-reactive T cells, which bind to self-antigens and receive a strong TCR signal required for apoptosis induction. Since

CYLD is a negative regulator of TCR signaling, the loss of CYLD may lead to heightened and abnormal deletion of T cells, which would explain the loss of mature T cells in CYLD-/- mice. The function of CYLD in peripheral T-cell activation is also due to its negative signaling function, since the CYLD deficiency results in hyper T-cell 154 response to TCR stimulation. It remains to be examined whether the peripheral T cells from CYLD-/- mice are also more sensitive to activation-induced cell death (AICD).

The defect of CYLD-/- mice in T-cell development is reminiscent of the JNK2-/- mice, although these two types of mice have opposite phenotypes. The JNK2-/- mice display a protection from apoptosis induction during thymic development. Since CYLD is a negative regulator of JNK, it is likely that the loss of CYLD leads to hyperactivation of JNK, thereby promoting T cell death during negative selection. In other words, the negative selection signal that normally would allow double-positive thymocytes to move to the single-positive stage is hyperactivated, therefore leading to the deletion of more T cells from the repertoire. Future studies will examine whether CYLD indeed negatively regulates JNK in thymocytes.

5.2.4 CYLD Regulation of Peripheral T-cell Activation

As mentioned above, the CYLD -/- T cells have secreted considerably more cytokines than the control cells when stimulated in vitro. This phenotype suggests a negative role for CYLD in regulating T cell activation. However, the loss of CYLD does not cause hyperproliferation of T cells. In fact, after 72 hours of stimulation, the proliferative ability of CYLD -/- T cells is severely decreased, despite their hyperproduction of IL-2. However, the increased levels of IL-2 found in the serum could be due to a decreased usage in the IL-2 be the CYLD -/- T cells. This defect could potentially arise from a defect in the ability of CYLD -/- T cells to upregulate the IL-2α high affinity receptor, which allows for efficient uptake of IL-2 and helps drive proliferation in T cells. Decreased uptake of IL-2 could also explain why CYLD -/- T cells fail to proliferate a latter time points. 155

This phenotype could also have two other sources. First, the CYLD -/- T cells may be more sensitive to AICD. Second, the loss of CYLD may cause an anti- proliferative effect. Of note, the opposite of these phenotypes has been observed in

NFAT1/NFAT4 double knock-out mice. Since NFAT seems to be negatively regulated by CYLD, it is possible that the loss of CYLD leads to constitutive or heightened activation of NFAT, which in turn contributes to the reduced proliferative ability and possibly enhanced AICD of CYLD -/- T cells. Another interesting correlation is that although CYLD -/- T cells have a slight decrease in proliferation with increased cytokine production, NFAT -/- T cells are hyperproliferative without the production of cytokines.

These functional correlations prompted us to examine the status of NFAT activation in

CYLD -/- T cells. Our preliminary IB analysis revealed moderate dephosphorylation of

NFAT1 in CYLD -/- T cells. Whether, NFAT is constitutively dephosphorylated requires further studies. Where as the dephosphorylation status of NFAT is one part of its regulation, the functional activation also requires activation of its nuclear partner AP1.

Since CYLD negatively regulates JNK, an activator of AP1, it is highly likely that the loss of CYLD enhances functional activation through AP1 activation.

5.2.5 CYLD regulation of Macrophages and B cells in the Spleen

Whereas the number of T cells in the spleen is dramatically decreased in CYLD -

/- mice, the number of B cells and macrophages is highly elevated (Figure 5.2 D and G).

This phenotype can be explained simply as a compensatory mechanism, we have obtained preliminary evidence suggesting that at least some of these cell types (B cells) are hyperproliferating. When stimulated with anti-IgM or LPS, the CYLD -/- B cells exhibited much higher hyperporliferation ability than wild-type control B cells. This 156 finding raises the possibility that the loss of CYLD may allow for hyper production of antibodies, leading to a more robust humoral response to specific pathogens. On the other hand, since antibody response to protein antigens requires T cells, the outcome of this axis of immune response may be more complex.

The large increase in macrophages also leads us to question whether CYLD has a negative role in macrophage activation. Using bone marrow-derived macrophages, we did not detect any significant alteration in the activation of JNK and IKK in response to a wide variety of inducers. However, when looking at the expression of various inflammatory genes by RNase protection assay, we noted a large increase in the induction of inducible nitric oxide synthase (iNOS) in CYLD -/- macrophages to TNF-α and CpG but not to LPS. This phenotype has been confirmed by western blots. Therefore, CYLD does play a negative role in regulating certain signaling pathways in macrophages.

However, at this time the physiological importance of this finding is not known. It will be of interest to examine how the increased numbers of macrophages in the spleen impact the immune response. It could well be envisioned that infection with pathogens would lead to an overproduction of cytokines and massive inflammation at the site of infection, potentially less beneficial to the host. Although there is a clear increase in the numbers of macrophages associated with CYLD deficiency, we do not know whether similar changes also occur in dendritic cell numbers. Due to the isolation technique used, I could not stain for the resident pool of dendritic cells. Therefore, future work needs to address the subpopulations of macrophages and dendritic cells exist within the spleen and peripheral lymph nodes of these mice. 157

CHAPTER 6

Future Directions 158

6.1 Regulation of Phosphorylation

Although I have made significant progress towards understanding the mechanism mediating CYLD phosphorylation, many questions are left unanswered. Presently, I know that all seven serines located within the phosphorylation region of CYLD are required for its phosphorylation in vitro or by overexpressed IKK in transfected cells.

However, it is unclear whether all these serines are actually phosphorylated by various inducers. Recently, by generating and using a phospho-specific antibody against phosphorylated serine 418, I have found that this site is not phosphorylated upon TNFα induction, but is instead phosphorylated by other inducers such as mitogens. Therefore, I am presently trying to answer which of the seven are regulated by which inducers. Could weak inducers, such as TNFα, only induce certain sites of phosphorylation within CYLD, thus allowing for a partial inhibition of the enzymatic activity? To answer this question, I need to create phospho-specific antibodies recognizing the other serine residues. These reagents will allow me to identify which serine residues become phosphorylated by which inducers, as well as to determine if there are temporal phosphorylation events occurring. I could then use the available serine-to-alanine mutations, individually or in combinations, to examine the functional significance of the individual phosphorylation sites in regulating the signaling functions of CYLD.

My data strongly implicate IKK as the kinase that phosphorylates CYLD.

However, genetic evidence for the requirement of one or both IKK catalytic subunits in

CYLD phosphorylation has yet to be demonstrated. MEFs lacking individual components of the IKK complex are not helpful for addressing this question since CYLD phosphorylation is not stimulated by TNFα in these cells. An ideal strategy is to produce 159 knock out mice with conditional T defects in one or two IKK catalytic subunits. These mice would allow us to examine the role of IKK in CYLD phosphorylation after mitogen or TCR/CD28 signaling, as well as other stimuli that induce CYLD phosphorylation in T cells.

Finally, the precise site of CYLD binding to IKKγ has not been defined through biochemical analysis. Mapping the minimal IKKγ-binding determinant within CYLD will allow us to examine whether this molecular interaction is required for CYLD phosphorylation. To determine the binding site of IKKγ in CYLD I am in the process of creating various N-terminal or C-terminal truncations. These constructs will be coexpressed with IKKγ in 293 cells and binding will be assayed by co-IP. From this initial work, smaller internal deletions can be made and assayed for their ability to bind to

IKKγ.

6.2 Domains of CYLD

In addition to the DUB catalytic domain, CYLD contains a number of other

notable structural domains, including three CAP-Gly domains and two proline rich

motifs. The function of these non-catalytic domains remains unknown. Mutational

analysis, coupled with functional rescue studies, will yield important information in

this regard.

Knowledge of other CYLD binding proteins is critical for understanding

additional mechanisms regulating the signaling function of CYLD. Therefore,

deletions of the specific domains have been made, or are presently being made. These

mutants will be introduced back into CYLD deficient cells to examine their effects on

the signaling functions of CYLD. The predicted CAP-Gly domains are important in 160 other proteins for intracellular trafficking through association with microtubule networks. Therefore, I predict that these domains are critical for the ability of CYLD to associated with other proteins that allowing for CYLD to traffic to various compartments in cells. Mutations of various CAP-Gly domains, can be used to study their subcellular localization and examine their effects on signaling, specifically JNK and IKK. However the proline rich regions are also very interesting as they are important in protein-protein interactions. Mutations of these proline rich regions in

CYLD will first be examined for their ability to negatively regulate JNK and IKK signaling. I will then examine the ability of CYLD proline rich mutants to be phosphorylated by various inducers. This could potentially elucidate the binding site of the unidentified kinase of serine 418 in CYLD. These future studies will hopefully expand our knowledge of how deubiquitinating enzymes regulate cellular processes.

6.3 In Vivo Physiological Role of CYLD in T Cell Development

My finding that CYLD -/- mice have a reduced number of CD3+ thymocytes demonstrate a physiological role for CYLD in regulating T cell development. Yet, I have not examined precisely which step(s) of the T cell development is subject to

CYLD regulation. To address this question, I will isolate thymocytes from 3 to 4 week old mice and examine cell surface markers CD44, CD25, and preTCRα specific for thymocytes at various stages of differentiation. This will allow for us to characterize if the defect in CYLD -/- thymocytes exists only during negative selection or if there are other developmental/proliferation defects. 161

Another question is whether the reduction in T cell number in CYLD -/- mice is caused by an intrinsic defect in T cell precursors or is caused by an extrinsic defect in the thymus microenvironment. This question will be addressed by a bone marrow reconstitution assay. In brief, I will sublethally irradiate a congenic BL6.SJL mouse of four weeks, and adoptively transfer bone marrow from CYLD -/- mice to reconstitute the immune system. If the defect in lymphocyte development is due to a stromal defect, transfer into the recipient should rescued CYLD -/- T cells in the reconstitution assay. However, if this defect (reduction in T cell number) occurs in the reconstitution assays, then it is likely that there is an intrinsic defect within the

CYLD -/- T cells. I favor the latter possibility. As discussed previously, it seems likely that the CYLD -/- thymocytes are more sensitive to apoptosis induction. Since

CYLD is a negative regulator of JNK, this abnormality in CYLD -/- mice could be caused by activation of the abundant JNK isoform JNK2, which is known to be involved in thymic apoptosis.

By analyzing the T cell development in these double knock-out mice, I will be able to assess whether JNK2 activation contributes to the T cell defect in CYLD -/- mice. Additionally, I will analyze whether JNK2 is hyperactivated in CYLD -/- thymocytes upon TCR stimulation. In brief, thymocytes from four week old mice will be stimulated with anti-CD3 antibodies for various times. Cells will be fixed in formaldehyde to quickly stop all cellular reactions and then stained with cell surface markers CD3, CD4, and CD8. Finally cells will be permabilized with either saponin or methanol and stained for phospho-JNK or other various MAPKs. Thymocyte populations will then be analyzed by flow cytometry and examined for the activation 162

JNK or the other MAPKs. I will also examine the apoptosis of CYLD -/- and control thymocytes by annexin V staining or by TUNEL staining utilizing a similar approach.

6.4 In Vivo Physiological Role of CYLD in T Cell Activation

My in vitro studies suggest that CYLD plays a negative role in regulating peripheral T cell activation. In response to TCR stimulation, CYLD -/- T cells produce remarkably larger amounts of cytokines, including both IL-2 and the effector

T-cell cytokines. To examine further the role of CYLD in T cell activation and effector function, I will use an in vivo approach. I will adoptively transfer equal numbers of purified wild-type and CYLD -/- T cells into a congenic host, CD45.2

(BL6.SJL), and one day later infect these mice with mouse-adapted strains, PR8 or

X31, of influenza virus. Using this system, I can examine the activation of naïve cells

and the memory response of both CD8 effector cells and CD4 TH1 cells, as measured by intracellular cytokine staining for IFNγ, IL-2, IL-4 and IL-10.

I will also examine the role of CYLD in regulating T cell proliferation using an in vivo approach. Although my in vitro studies detected a proliferation defect of CYLD

-/- T cells after 72 hours of stimulation, this phenotype could be due to the in vitro culture system. Therefore, prior to adoptively transferring wild-type or CYLD -/- T cells, I will label them with CFSE, a fluorescent dye that is equally partitioned into dividing cells, thereby allowing us to track the number of cellular divisions. This technique will enable us to examine the proliferative ability of CYLD -/- cells in response to an infection. If proliferation defects do arise, I would identify at what day the cells stop dividing. If there are proliferative defects such as that seen at 72 hour ex vivo in CYLD -/- cells, I will examine whether this defect is due to an increase in 163 apoptosis. Cells could be extracted from the mice at various days after adoptive transfer and infection and scored by annexin V or TUNEL staining.

These observations will examine more of the phenotype of CYLD -/- mice, yet, I still need to investigate the mechanism responsible for the increase in cytokine production and the reduction in proliferative ability. Once question is whether CYLD

-/- T cells display a similar phenotype to CYLD knock down Jurkat cells in respect to signaling. To this end, I will stimulate T cells ex vivo and assay intracellular activated JNK1 and JNK2 by flow cytometry and by phospho-western blots.

Whereas Jurkat cells exhibit immediate activation of JNK1/2, naïve T cells typically display activated JNK around 18 to 24 hours. Therefore, we will have to perform a time course to obtain a complete analysis of JNK activation in CYLD -/- T cells.

Once initial data is obtained from a direct ex vivo system, we can then proceed to a more in vivo approach. There is interesting correlation between CYLD -/- T cells and

NFAT1 or NFAT2 -/- T cells. Whereas CYLD -/- T cells have proliferative defects and hypersecretion of cytokines, NFAT -/- T cells typically display a reversed phenotype. Therefore, it will be interesting to see whether CYLD negatively regulates the NFAT pathways in vivo. My predictions is that CYLD regulates some upstream signaling molecule in TCR signaling that affects both NFAT and IKK activation. Targets could be c-Cbl, Cbl-b, or even PLC-γ as all molecules have been reported to be ubiquitinated. However, the role in ubiquitinate-mediated activation of these molecules has not been investigated. Therefore I hypothesize that CYLD negatively regulates the upstream signaling of one these molecules by persistently keeping them in a hypoubiquitinated state. If CYLD exerts its effects on NFAT, it 164 would predict that the mechanism of regulation by CYLD is extremely early in T cell receptor signaling, prior to PLCγ activation, defining a novel role for CYLD in the negative regulation of T cells in vivo. 165

BIBLIOGRAPHY

1. Hoffmann, J.A., et al., Phylogenetic perspectives in innate immunity. Science, 1999. 284(5418): p. 1313-8. 2. Natarajan, K., et al., Structure and function of natural killer cell receptors: multiple molecular solutions to self, nonself discrimination. Annu Rev Immunol, 2002. 20: p. 853-85. 3. Bieniasz, P.D., Intrinsic immunity: a front-line defense against viral attack. Nat Immunol, 2004. 5(11): p. 1109-15. 4. Janeway, C., Immunogenicity signals 1,2,3 ... and 0. Immunol Today, 1989. 10(9): p. 283-6. 5. Medzhitov, R. and C.A. Janeway, Jr., Innate immunity: impact on the adaptive immune response. Curr Opin Immunol, 1997. 9(1): p. 4-9. 6. Beutler, B., Inferences, questions and possibilities in Toll-like receptor signalling. Nature, 2004. 430(6996): p. 257-63. 7. Schnare, M., et al., Toll-like receptors control activation of adaptive immune responses. Nat Immunol, 2001. 2(10): p. 947-50. 8. Flajnik, M.F., Comparative analyses of immunoglobulin genes: surprises and portents. Nat Rev Immunol, 2002. 2(9): p. 688-98. 9. Thompson, C.B., New insights into V(D)J recombination and its role in the evolution of the immune system. Immunity, 1995. 3(5): p. 531-9. 10. Rosenberg, A.S., et al., Cellular basis of skin allograft rejection across a class I major histocompatibility barrier in mice depleted of CD8+ T cells in vivo. J Exp Med, 1991. 173(6): p. 1463-71. 11. Janeway, C.A., Jr., The immune system evolved to discriminate infectious nonself from noninfectious self. Immunol Today, 1992. 13(1): p. 11-6. 12. Sen, R. and D. Baltimore, Inducibility of kappa immunoglobulin enhancer- binding protein Nf-kappa B by a posttranslational mechanism. Cell, 1986. 47(6): p. 921-8. 13. Kaisho, T., et al., IkappaB kinase alpha is essential for mature B cell development and function. J Exp Med, 2001. 193(4): p. 417-26. 14. Caamano, J.H., et al., Nuclear factor (NF)-kappa B2 (p100/p52) is required for normal splenic microarchitecture and B cell-mediated immune responses. J Exp Med, 1998. 187(2): p. 185-96. 15. Snapper, C.M., et al., B cells lacking RelB are defective in proliferative responses, but undergo normal B cell maturation to Ig secretion and Ig class switching. J Exp Med, 1996. 184(4): p. 1537-41. 16. Ghosh, S., May, M.J., and Kopp, E.B., NF-kappa B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol., 1998. 16: p. 225-260. 17. Mercurio, F., et al., p105 and p98 precursor proteins play an active role in NF- kappa B-mediated signal transduction. Genes Dev, 1993. 7(4): p. 705-18. 18. Lin, L., G.N. DeMartino, and W.C. Greene, Cotranslational biogenesis of NF- kappaB p50 by the 26S proteasome. Cell, 1998. 92(6): p. 819-28. 166

19. Lin, L., G.N. DeMartino, and W.C. Greene, Cotranslational dimerization of the Rel homology domain of NF-kappaB1 generates p50-p105 heterodimers and is required for effective p50 production. Embo J, 2000. 19(17): p. 4712-22. 20. Ghosh, S., van Duyne, G., Ghosh, S., and Sigler, P.B., Structure of NF-kappa B p50 homodimer bound to a kB site. Nature, 1995. 373: p. 303-310. 21. Grimm, S. and P.A. Baeuerle, The inducible transcription factor NF-kappa B: structure-function relationship of its protein subunits. Biochem J, 1993. 290 ( Pt 2): p. 297-308. 22. Chen, F.E., et al., Crystal structure of p50/p65 heterodimer of transcription factor NF-kappaB bound to DNA. Nature, 1998. 391(6665): p. 410-3. 23. Baeuerle, P.A. and D. Baltimore, I kappa B: a specific inhibitor of the NF-kappa B transcription factor. Science, 1988. 242(4878): p. 540-6. 24. Beg, A.A., et al., I kappa B interacts with the nuclear localization sequences of the subunits of NF-kappa B: a mechanism for cytoplasmic retention. Genes Dev, 1992. 6(10): p. 1899-913. 25. Ganchi, P.A., et al., I kappa B/MAD-3 masks the nuclear localization signal of NF-kappa B p65 and requires the transactivation domain to inhibit NF-kappa B p65 DNA binding. Mol Biol Cell, 1992. 3(12): p. 1339-52. 26. Jacobs, M.D. and S.C. Harrison, Structure of an IkappaBalpha/NF-kappaB complex. Cell, 1998. 95(6): p. 749-58. 27. Sun, S., J. Elwood, and W.C. Greene, Both amino- and carboxyl-terminal sequences within I kappa B alpha regulate its inducible degradation. Mol Cell Biol, 1996. 16(3): p. 1058-65. 28. Huxford, T., et al., The crystal structure of the IkappaBalpha/NF-kappaB complex reveals mechanisms of NF-kappaB inactivation. Cell, 1998. 95(6): p. 759-70. 29. Brown, K., et al., Control of I kappa B-alpha proteolysis by site-specific, signal- induced phosphorylation. Science, 1995. 267(5203): p. 1485-8. 30. DiDonato, J.A., F. Mercurio, and M. Karin, Phosphorylation of I kappa B alpha precedes but is not sufficient for its dissociation from NF-kappa B. Mol Cell Biol, 1995. 15(3): p. 1302-11. 31. Davis, M., et al., Pseudosubstrate regulation of the SCF(beta-TrCP) ubiquitin ligase by hnRNP-U. Genes Dev, 2002. 16(4): p. 439-51. 32. Ben-Neriah, Y., Regulatory functions of ubiquitination in the immune system. Nat Immunol, 2002. 3(1): p. 20-6. 33. Scherer, D.C., et al., Signal-induced degradation of I kappa B alpha requires site- specific ubiquitination. Proc Natl Acad Sci U S A, 1995. 92(24): p. 11259-63. 34. Sun, L. and Z.J. Chen, The novel functions of ubiquitination in signaling. Curr Opin Cell Biol, 2004. 16(2): p. 119-26. 35. Karin, M., How NF-kappaB is activated: the role of the IkappaB kinase (IKK) complex. , 1999. 18(49): p. 6867-74. 36. Le Bail, O., R. Schmidt-Ullrich, and A. Israel, Promoter analysis of the gene encoding the I kappa B-alpha/MAD3 inhibitor of NF-kappa B: positive regulation by members of the rel/NF-kappa B family. Embo J, 1993. 12(13): p. 5043-9. 37. Brown, K., et al., Mutual regulation of the transcriptional activator NF-kappa B and its inhibitor, I kappa B-alpha. Proc Natl Acad Sci U S A, 1993. 90(6): p. 2532-6. 167

38. Sun, S.C., et al., NF-kappa B controls expression of inhibitor I kappa B alpha: evidence for an inducible autoregulatory pathway. Science, 1993. 259(5103): p. 1912-5. 39. Thompson, J.E., et al., I kappa B-beta regulates the persistent response in a biphasic activation of NF-kappa B. Cell, 1995. 80(4): p. 573-82. 40. DiDonato, J.A., et al., A cytokine-responsive IkappaB kinase that activates the transcription factor NF-kappaB. Nature, 1997. 388(6642): p. 548-54. 41. Zandi, E., et al., The IkappaB kinase complex (IKK) contains two kinase subunits, IKKalpha and IKKbeta, necessary for IkappaB phosphorylation and NF-kappaB activation. Cell, 1997. 91(2): p. 243-52. 42. Regnier, C.H., et al., Identification and characterization of an IkappaB kinase. Cell, 1997. 90(2): p. 373-83. 43. Mercurio, F., et al., IkappaB kinase (IKK)-associated protein 1, a common component of the heterogeneous IKK complex. Mol Cell Biol, 1999. 19(2): p. 1526-38. 44. Rothwarf, D.M., et al., IKK-gamma is an essential regulatory subunit of the IkappaB kinase complex. Nature, 1998. 395(6699): p. 297-300. 45. Yamaoka, S., et al., Complementation cloning of NEMO, a component of the IkappaB kinase complex essential for NF-kappaB activation. Cell, 1998. 93(7): p. 1231-40. 46. Li, Y., et al., Identification of a cell protein (FIP-3) as a modulator of NF-kappaB activity and as a target of an adenovirus inhibitor of tumor necrosis factor alpha- induced apoptosis. Proc Natl Acad Sci U S A, 1999. 96(3): p. 1042-7. 47. Delhase, M., et al., Positive and negative regulation of IkappaB kinase activity through IKKbeta subunit phosphorylation. Science, 1999. 284(5412): p. 309-13. 48. Johnson, L.N., M.E. Noble, and D.J. Owen, Active and inactive protein kinases: structural basis for regulation. Cell, 1996. 85(2): p. 149-58. 49. Chen, G., P. Cao, and D.V. Goeddel, TNF-induced recruitment and activation of the IKK complex require Cdc37 and . Mol Cell, 2002. 9(2): p. 401-10. 50. Ling, L., Z. Cao, and D.V. Goeddel, NF-kappaB-inducing kinase activates IKK- alpha by phosphorylation of Ser-176. Proc Natl Acad Sci U S A, 1998. 95(7): p. 3792-7. 51. Good, L., S.B. Maggirwar, and S.-C. Sun, Activation of IL-2 gene promoter by HTLV-1 Tax involves induction of NF-AT complexes bound to the CD28 responsive element. EMBO J., 1996. 15: p. 3744-3750. 52. Rudolph, D., et al., Severe liver degeneration and lack of NF-kappaB activation in NEMO/IKKgamma-deficient mice. Genes Dev, 2000. 14(7): p. 854-62. 53. Tanaka, M., et al., Embryonic lethality, liver degeneration, and impaired NF- kappa B activation in IKK-beta-deficient mice. Immunity, 1999. 10(4): p. 421-9. 54. Li, Z.W., et al., The IKKbeta subunit of IkappaB kinase (IKK) is essential for nuclear factor kappaB activation and prevention of apoptosis. J Exp Med, 1999. 189(11): p. 1839-45. 55. Chu, W.M., et al., JNK2 and IKKbeta are required for activating the innate response to viral infection. Immunity, 1999. 11(6): p. 721-31. 56. Li, Q., et al., Severe liver degeneration in mice lacking the IkappaB kinase 2 gene. Science, 1999. 284(5412): p. 321-5. 168

57. Alcamo, E., et al., Targeted mutation of TNF receptor I rescues the RelA-deficient mouse and reveals a critical role for NF-kappa B in leukocyte recruitment. J Immunol, 2001. 167(3): p. 1592-600. 58. Senftleben, U., et al., IKKbeta is essential for protecting T cells from TNFalpha- induced apoptosis. Immunity, 2001. 14(3): p. 217-30. 59. Li, Q., et al., IKK1-deficient mice exhibit abnormal development of skin and skeleton. Genes Dev, 1999. 13(10): p. 1322-8. 60. Takeda, K., et al., Limb and skin abnormalities in mice lacking IKKalpha. Science, 1999. 284(5412): p. 313-6. 61. Hu, Y., et al., IKKalpha controls formation of the epidermis independently of NF- kappaB. Nature, 2001. 410(6829): p. 710-4. 62. Cao, Y., et al., IKKalpha provides an essential link between RANK signaling and cyclin D1 expression during mammary gland development. Cell, 2001. 107(6): p. 763-75. 63. Senftleben, U., et al., Activation by IKKalpha of a second, evolutionary conserved, NF-kappa B signaling pathway. Science, 2001. 293(5534): p. 1495-9. 64. Xiao, G., E.W. Harhaj, and S.C. Sun, NF-kappaB-inducing kinase regulates the processing of NF-kappaB2 p100. Mol Cell, 2001. 7(2): p. 401-9. 65. Dejardin, E., et al., The lymphotoxin-beta receptor induces different patterns of gene expression via two NF-kappaB pathways. Immunity, 2002. 17(4): p. 525-35. 66. Mercurio, F., et al., IKK-1 and IKK-2: cytokine-activated IkappaB kinases essential for NF-kappaB activation. Science, 1997. 278(5339): p. 860-6. 67. Heilker, R., et al., All three IkappaB isoforms and most Rel family members are stably associated with the IkappaB kinase 1/2 complex. Eur J Biochem, 1999. 259(1-2): p. 253-61. 68. Cohen, L., W.J. Henzel, and P.A. Baeuerle, IKAP is a scaffold protein of the IkappaB kinase complex. Nature, 1998. 395(6699): p. 292-6. 69. Ducut Sigala, J.L., et al., Activation of transcription factor NF-kappaB requires ELKS, an IkappaB kinase regulatory subunit. Science, 2004. 304(5679): p. 1963- 7. 70. Xiao, G., et al., Retroviral oncoprotein Tax induces processing of NF- kappaB2/p100 in T cells: evidence for the involvement of IKKalpha. EMBO J., 2001. 20: p. 6805-6815. 71. Akira, S. and K. Takeda, Toll-like receptor signalling. Nat Rev Immunol, 2004. 4(7): p. 499-511. 72. Campos, M.A., et al., Activation of Toll-like receptor-2 by glycosylphosphatidylinositol anchors from a protozoan parasite. J Immunol, 2001. 167(1): p. 416-23. 73. Underhill, D.M., et al., The Toll-like receptor 2 is recruited to macrophage phagosomes and discriminates between pathogens. Nature, 1999. 401(6755): p. 811-5. 74. Meier, A., et al., Toll-like receptor (TLR) 2 and TLR4 are essential for Aspergillus-induced activation of murine macrophages. Cell Microbiol, 2003. 5(8): p. 561-70. 75. Hayashi, F., et al., The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature, 2001. 410(6832): p. 1099-103. 169

76. Zhang, D., et al., A toll-like receptor that prevents infection by uropathogenic bacteria. Science, 2004. 303(5663): p. 1522-6. 77. Burns, K., et al., MyD88, an adapter protein involved in interleukin-1 signaling. J Biol Chem, 1998. 273(20): p. 12203-9. 78. Muzio, M., et al., IRAK (Pelle) family member IRAK-2 and MyD88 as proximal mediators of IL-1 signaling. Science, 1997. 278(5343): p. 1612-5. 79. Wesche, H., et al., MyD88: an adapter that recruits IRAK to the IL-1 receptor complex. Immunity, 1997. 7(6): p. 837-47. 80. Muzio, M., et al., Toll-like receptors: a growing family of immune receptors that are differentially expressed and regulated by different leukocytes. J Leukoc Biol, 2000. 67(4): p. 450-6. 81. Sato, S., et al., A variety of microbial components induce tolerance to lipopolysaccharide by differentially affecting MyD88-dependent and -independent pathways. Int Immunol, 2002. 14(7): p. 783-91. 82. Adachi, O., et al., Targeted disruption of the MyD88 gene results in loss of IL-1- and IL-18-mediated function. Immunity, 1998. 9(1): p. 143-50. 83. Horng, T., et al., The adaptor molecule TIRAP provides signalling specificity for Toll-like receptors. Nature, 2002. 420(6913): p. 329-33. 84. Yamamoto, M., et al., Essential role for TIRAP in activation of the signalling cascade shared by TLR2 and TLR4. Nature, 2002. 420(6913): p. 324-9. 85. Dunne, A., et al., Structural complementarity of Toll/interleukin-1 receptor domains in Toll-like receptors and the adaptors Mal and MyD88. J Biol Chem, 2003. 278(42): p. 41443-51. 86. Suzuki, N., S. Suzuki, and W.C. Yeh, IRAK-4 as the central TIR signaling mediator in innate immunity. Trends Immunol, 2002. 23(10): p. 503-6. 87. Li, S., et al., IRAK-4: a novel member of the IRAK family with the properties of an IRAK-kinase. Proc Natl Acad Sci U S A, 2002. 99(8): p. 5567-72. 88. Li, X., et al., Mutant cells that do not respond to interleukin-1 (IL-1) reveal a novel role for IL-1 receptor-associated kinase. Mol Cell Biol, 1999. 19(7): p. 4643-52. 89. Deng, L., et al., Activation of the IkappaB kinase complex by TRAF6 requires a dimeric ubiquitin-conjugating enzyme complex and a unique polyubiquitin chain. Cell, 2000. 103(2): p. 351-61. 90. Wang, C., et al., TAK1 is a ubiquitin-dependent kinase of MKK and IKK. Nature, 2001. 412(6844): p. 346-51. 91. Yamaguchi, K., et al., Identification of a member of the MAPKKK family as a potential mediator of TGF-beta signal transduction. Science, 1995. 270(5244): p. 2008-11. 92. Ninomiya-Tsuji, J., et al., The kinase TAK1 can activate the NIK-I kappaB as well as the MAP kinase cascade in the IL-1 signalling pathway. Nature, 1999. 398(6724): p. 252-6. 93. Shibuya, H., et al., TAB1: an activator of the TAK1 MAPKKK in TGF-beta signal transduction. Science, 1996. 272(5265): p. 1179-82. 94. Takaesu, G., et al., TAB2, a novel adaptor protein, mediates activation of TAK1 MAPKKK by linking TAK1 to TRAF6 in the IL-1 signal transduction pathway. Mol Cell, 2000. 5(4): p. 649-58. 170

95. Ishitani, T., et al., Role of the TAB2-related protein TAB3 in IL-1 and TNF signaling. Embo J, 2003. 22(23): p. 6277-88. 96. Takaesu, G., et al., TAK1 is critical for IkappaB kinase-mediated activation of the NF-kappaB pathway. J Mol Biol, 2003. 326(1): p. 105-15. 97. Kawai, T., et al., Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity, 1999. 11(1): p. 115-22. 98. Kawai, T., et al., Lipopolysaccharide stimulates the MyD88-independent pathway and results in activation of IFN-regulatory factor 3 and the expression of a subset of lipopolysaccharide-inducible genes. J Immunol, 2001. 167(10): p. 5887-94. 99. Yamamoto, M., et al., Cutting edge: a novel Toll/IL-1 receptor domain-containing adapter that preferentially activates the IFN-beta promoter in the Toll-like receptor signaling. J Immunol, 2002. 169(12): p. 6668-72. 100. Bin, L.H., L.G. Xu, and H.B. Shu, TIRP, a novel Toll/interleukin-1 receptor (TIR) domain-containing adapter protein involved in TIR signaling. J Biol Chem, 2003. 278(27): p. 24526-32. 101. Fitzgerald, K.A., et al., LPS-TLR4 signaling to IRF-3/7 and NF-kappaB involves the toll adapters TRAM and TRIF. J Exp Med, 2003. 198(7): p. 1043-55. 102. Yamamoto, M., et al., TRAM is specifically involved in the Toll-like receptor 4- mediated MyD88-independent signaling pathway. Nat Immunol, 2003. 4(11): p. 1144-50. 103. Oshiumi, H., et al., TICAM-1, an adaptor molecule that participates in Toll-like receptor 3-mediated interferon-beta induction. Nat Immunol, 2003. 4(2): p. 161- 7. 104. Oshiumi, H., et al., TIR-containing adapter molecule (TICAM)-2, a bridging adapter recruiting to toll-like receptor 4 TICAM-1 that induces interferon-beta. J Biol Chem, 2003. 278(50): p. 49751-62. 105. Yoneyama, M., et al., Direct triggering of the type I interferon system by virus infection: activation of a transcription factor complex containing IRF-3 and CBP/p300. Embo J, 1998. 17(4): p. 1087-95. 106. Sato, S., et al., Toll/IL-1 receptor domain-containing adaptor inducing IFN-beta (TRIF) associates with TNF receptor-associated factor 6 and TANK-binding kinase 1, and activates two distinct transcription factors, NF-kappa B and IFN- regulatory factor-3, in the Toll-like receptor signaling. J Immunol, 2003. 171(8): p. 4304-10. 107. McWhirter, S.M., et al., IFN-regulatory factor 3-dependent gene expression is defective in Tbk1-deficient mouse embryonic fibroblasts. Proc Natl Acad Sci U S A, 2004. 101(1): p. 233-8. 108. Meylan, E., et al., RIP1 is an essential mediator of Toll-like receptor 3-induced NF-kappa B activation. Nat Immunol, 2004. 5(5): p. 503-7. 109. Janeway, C.A., P. Travers, and M. Walport, Immunobiology. 2001: New York: Garland. 110. Spiegel, S., et al., Direct visualization of redistribution and capping of fluorescent gangliosides on lymphocytes. J Cell Biol, 1984. 99(5): p. 1575-81. 111. Monks, C.R., et al., Three-dimensional segregation of supramolecular activation clusters in T cells. Nature, 1998. 395(6697): p. 82-6. 171

112. Palacios, E.H. and A. Weiss, Function of the Src-family kinases, Lck and Fyn, in T-cell development and activation. Oncogene, 2004. 23(48): p. 7990-8000. 113. Kim, P.W., et al., A zinc clasp structure tethers Lck to T cell coreceptors CD4 and CD8. Science, 2003. 301(5640): p. 1725-8. 114. Turner, J.M., et al., Interaction of the unique N-terminal region of tyrosine kinase p56lck with cytoplasmic domains of CD4 and CD8 is mediated by cysteine motifs. Cell, 1990. 60(5): p. 755-65. 115. Samelson, L.E., Signal transduction mediated by the T cell antigen receptor: the role of adapter proteins. Annu Rev Immunol, 2002. 20: p. 371-94. 116. Lee, K.H., et al., T cell receptor signaling precedes immunological synapse formation. Science, 2002. 295(5559): p. 1539-42. 117. Lee, K.H., et al., The immunological synapse balances T cell receptor signaling and degradation. Science, 2003. 302(5648): p. 1218-22. 118. Schrum, A.G. and L.A. Turka, The proliferative capacity of individual naive CD4(+) T cells is amplified by prolonged T cell antigen receptor triggering. J Exp Med, 2002. 196(6): p. 793-803. 119. Michel, F., et al., Fyn and ZAP-70 are required for Vav phosphorylation in T cells stimulated by antigen-presenting cells. J Biol Chem, 1998. 273(48): p. 31932-8. 120. Klasen, S., et al., Two distinct regions of the CD28 intracytoplasmic domain are involved in the tyrosine phosphorylation of Vav and GTPase activating protein- associated p62 protein. Int Immunol, 1998. 10(4): p. 481-9. 121. Tuosto, L., F. Michel, and O. Acuto, p95vav associates with tyrosine- phosphorylated SLP-76 in antigen-stimulated T cells. J Exp Med, 1996. 184(3): p. 1161-6. 122. Myung, P.S., et al., Differential requirement for SLP-76 domains in T cell development and function. Immunity, 2001. 15(6): p. 1011-26. 123. Villalba, M., et al., A novel functional interaction between Vav and PKCtheta is required for TCR-induced T cell activation. Immunity, 2000. 12(2): p. 151-60. 124. Bi, K., et al., Antigen-induced translocation of PKC-theta to membrane rafts is required for T cell activation. Nat Immunol, 2001. 2(6): p. 556-63. 125. Piccolella, E., et al., Vav-1 and the IKK alpha subunit of I kappa B kinase functionally associate to induce NF-kappa B activation in response to CD28 engagement. J Immunol, 2003. 170(6): p. 2895-903. 126. Koegl, M., et al., Palmitoylation of multiple Src-family kinases at a homologous N-terminal motif. Biochem J, 1994. 303 ( Pt 3): p. 749-53. 127. Paige, L.A., et al., Reversible palmitoylation of the protein-tyrosine kinase p56lck. J Biol Chem, 1993. 268(12): p. 8669-74. 128. Wilkinson, B., H. Wang, and C.E. Rudd, Positive and negative adaptors in T-cell signalling. Immunology, 2004. 111(4): p. 368-74. 129. Yang, W.C., et al., The role of Tec protein-tyrosine kinase in T cell signaling. J Biol Chem, 1999. 274(2): p. 607-17. 130. Tomlinson, M.G., J. Lin, and A. Weiss, Lymphocytes with a complex: adapter proteins in antigen receptor signaling. Immunol Today, 2000. 21(11): p. 584-91. 131. Astoul, E., et al., PI 3-K and T-cell activation: limitations of T-leukemic cell lines as signaling models. Trends Immunol, 2001. 22(9): p. 490-6. 172

132. Ward, S.G. and D.A. Cantrell, Phosphoinositide 3-kinases in T lymphocyte activation. Curr Opin Immunol, 2001. 13(3): p. 332-8. 133. Isakov, N. and A. Altman, Protein kinase C(theta) in T cell activation. Annu Rev Immunol, 2002. 20: p. 761-94. 134. Sun, Z., et al., PKC-theta is required for TCR-induced NF-kappaB activation in mature but not immature T lymphocytes. Nature, 2000. 404(6776): p. 402-7. 135. Ruland, J., et al., Differential requirement for Malt1 in T and B cell antigen receptor signaling. Immunity, 2003. 19(5): p. 749-58. 136. Ruefli-Brasse, A.A., D.M. French, and V.M. Dixit, Regulation of NF-kappaB- dependent lymphocyte activation and development by paracaspase. Science, 2003. 302(5650): p. 1581-4. 137. Wang, D., et al., A requirement for CARMA1 in TCR-induced NF-kappa B activation. Nat Immunol, 2002. 3(9): p. 830-5. 138. Pomerantz, J.L., E.M. Denny, and D. Baltimore, CARD11 mediates factor- specific activation of NF-kappaB by the T cell receptor complex. Embo J, 2002. 21(19): p. 5184-94. 139. Ruland, J., et al., Bcl10 is a positive regulator of antigen receptor-induced activation of NF-kappaB and neural tube closure. Cell, 2001. 104(1): p. 33-42. 140. Gaide, O., et al., CARMA1 is a critical lipid raft-associated regulator of TCR- induced NF-kappa B activation. Nat Immunol, 2002. 3(9): p. 836-43. 141. Gaide, O., et al., Carma1, a CARD-containing binding partner of Bcl10, induces Bcl10 phosphorylation and NF-kappaB activation. FEBS Lett, 2001. 496(2-3): p. 121-7. 142. Bertin, J., et al., CARD11 and CARD14 are novel caspase recruitment domain (CARD)/membrane-associated guanylate kinase (MAGUK) family members that interact with BCL10 and activate NF-kappa B. J Biol Chem, 2001. 276(15): p. 11877-82. 143. Srinivasula, S.M., et al., CLAP, a novel caspase recruitment domain-containing protein in the tumor necrosis factor receptor pathway, regulates NF-kappaB activation and apoptosis. J Biol Chem, 1999. 274(25): p. 17946-54. 144. Zhou, H., et al., Bcl10 activates the NF-kappaB pathway through ubiquitination of NEMO. Nature, 2004. 427(6970): p. 167-71. 145. Sun, L., et al., The TRAF6 ubiquitin ligase and TAK1 kinase mediate IKK activation by BCL10 and MALT1 in T lymphocytes. Mol Cell, 2004. 14(3): p. 289- 301. 146. Idriss, H.T. and J.H. Naismith, TNF alpha and the TNF receptor superfamily: structure-function relationship(s). Microsc Res Tech, 2000. 50(3): p. 184-95. 147. Locksley, R.M., N. Killeen, and M.J. Lenardo, The TNF and TNF receptor superfamilies: integrating mammalian biology. Cell, 2001. 104(4): p. 487-501. 148. Legler, D.F., et al., Recruitment of TNF receptor 1 to lipid rafts is essential for TNFalpha-mediated NF-kappaB activation. Immunity, 2003. 18(5): p. 655-64. 149. Zhang, S.Q., et al., Recruitment of the IKK signalosome to the p55 TNF receptor: RIP and A20 bind to NEMO (IKKgamma) upon receptor stimulation. Immunity, 2000. 12(3): p. 301-11. 150. Tang, E.D., et al., A role for NF-kappaB essential modifier/IkappaB kinase- gamma (NEMO/IKKgamma) ubiquitination in the activation of the IkappaB 173

kinase complex by tumor necrosis factor-alpha. J Biol Chem, 2003. 278(39): p. 37297-305. 151. Wertz, I.E., et al., De-ubiquitination and ubiquitin ligase domains of A20 downregulate NF-kappaB signalling. Nature, 2004. 430(7000): p. 694-9. 152. Lee, T.H., et al., The kinase activity of Rip1 is not required for tumor necrosis factor-alpha-induced IkappaB kinase or p38 MAP kinase activation or for the ubiquitination of Rip1 by Traf2. J Biol Chem, 2004. 279(32): p. 33185-91. 153. Habelhah, H., et al., Ubiquitination and translocation of TRAF2 is required for activation of JNK but not of p38 or NF-kappaB. Embo J, 2004. 23(2): p. 322-32. 154. Ye, H. and H. Wu, Thermodynamic characterization of the interaction between TRAF2 and tumor necrosis factor receptor peptides by isothermal titration calorimetry. Proc Natl Acad Sci U S A, 2000. 97(16): p. 8961-6. 155. Park, Y.C., et al., A novel mechanism of TRAF signaling revealed by structural and functional analyses of the TRADD-TRAF2 interaction. Cell, 2000. 101(7): p. 777-87. 156. Wang, C.Y., et al., NF-kappaB antiapoptosis: induction of TRAF1 and TRAF2 and c-IAP1 and c-IAP2 to suppress caspase-8 activation. Science, 1998. 281(5383): p. 1680-3. 157. Micheau, O. and J. Tschopp, Induction of TNF receptor I-mediated apoptosis via two sequential signaling complexes. Cell, 2003. 114(2): p. 181-90. 158. Liu, H., et al., A Drosophila TNF-receptor-associated factor (TRAF) binds the ste20 kinase Misshapen and activates Jun kinase. Curr Biol, 1999. 9(2): p. 101-4. 159. Wajant, H., F. Muhlenbeck, and P. Scheurich, Identification of a TRAF (TNF receptor-associated factor) gene in Caenorhabditis elegans. J Mol Evol, 1998. 47(6): p. 656-62. 160. Rothe, M., et al., A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumor necrosis factor receptor. Cell, 1994. 78(4): p. 681-92. 161. Park, Y.C., et al., Structural basis for self-association and receptor recognition of human TRAF2. Nature, 1999. 398(6727): p. 533-8. 162. Pullen, S.S., et al., CD40-tumor necrosis factor receptor-associated factor (TRAF) interactions: regulation of CD40 signaling through multiple TRAF binding sites and TRAF hetero-oligomerization. Biochemistry, 1998. 37(34): p. 11836-45. 163. Takeuchi, M., M. Rothe, and D.V. Goeddel, Anatomy of TRAF2. Distinct domains for nuclear factor-kappaB activation and association with tumor necrosis factor signaling proteins. J Biol Chem, 1996. 271(33): p. 19935-42. 164. Cheng, G., et al., Involvement of CRAF1, a relative of TRAF, in CD40 signaling. Science, 1995. 267(5203): p. 1494-8. 165. Rothe, M., et al., TRAF2-mediated activation of NF-kappa B by TNF receptor 2 and CD40. Science, 1995. 269(5229): p. 1424-7. 166. Mosialos, G., et al., The Epstein-Barr virus transforming protein LMP1 engages signaling proteins for the tumor necrosis factor receptor family. Cell, 1995. 80(3): p. 389-99. 174

167. Hostager, B.S., I.M. Catlett, and G.A. Bishop, Recruitment of CD40 and tumor necrosis factor receptor-associated factors 2 and 3 to membrane microdomains during CD40 signaling. J Biol Chem, 2000. 275(20): p. 15392-8. 168. Devergne, O., et al., Association of TRAF1, TRAF2, and TRAF3 with an Epstein- Barr virus LMP1 domain important for B-lymphocyte transformation: role in NF- kappaB activation. Mol Cell Biol, 1996. 16(12): p. 7098-108. 169. Darnay, B.G., et al., Characterization of the intracellular domain of receptor activator of NF-kappaB (RANK). Interaction with tumor necrosis factor receptor- associated factors and activation of NF-kappab and c-Jun N-terminal kinase. J Biol Chem, 1998. 273(32): p. 20551-5. 170. Tsukamoto, N., et al., Two differently regulated nuclear factor kappaB activation pathways triggered by the cytoplasmic tail of CD40. Proc Natl Acad Sci U S A, 1999. 96(4): p. 1234-9. 171. Darnay, B.G., et al., Activation of NF-kappaB by RANK requires tumor necrosis factor receptor-associated factor (TRAF) 6 and NF-kappaB-inducing kinase. Identification of a novel TRAF6 interaction motif. J Biol Chem, 1999. 274(12): p. 7724-31. 172. Pickart, C.M., Mechanisms underlying ubiquitination. Annu Rev Biochem, 2001. 70: p. 503-33. 173. Shi, C.S. and J.H. Kehrl, Tumor necrosis factor (TNF)-induced germinal center kinase-related (GCKR) and stress-activated protein kinase (SAPK) activation depends upon the E2/E3 complex Ubc13-Uev1A/TNF receptor-associated factor 2 (TRAF2). J Biol Chem, 2003. 278(17): p. 15429-34. 174. Yang, K., et al., The coiled-coil domain of TRAF6 is essential for its auto- ubiquitination. Biochem Biophys Res Commun, 2004. 324(1): p. 432-9. 175. Yeh, W.C., et al., Early lethality, functional NF-kappaB activation, and increased sensitivity to TNF-induced cell death in TRAF2-deficient mice. Immunity, 1997. 7(5): p. 715-25. 176. Tada, K., et al., Critical roles of TRAF2 and TRAF5 in tumor necrosis factor- induced NF-kappa B activation and protection from cell death. J Biol Chem, 2001. 276(39): p. 36530-4. 177. Shi, C.S. and J.H. Kehrl, Activation of stress-activated protein kinase/c-Jun N- terminal kinase, but not NF-kappaB, by the tumor necrosis factor (TNF) receptor 1 through a TNF receptor-associated factor 2- and germinal center kinase related-dependent pathway. J Biol Chem, 1997. 272(51): p. 32102-7. 178. Xia, Y., et al., MEK kinase 1 is critically required for c-Jun N-terminal kinase activation by proinflammatory stimuli and -induced cell migration. Proc Natl Acad Sci U S A, 2000. 97(10): p. 5243-8. 179. Yuasa, T., et al., Tumor necrosis factor signaling to stress-activated protein kinase (SAPK)/Jun NH2-terminal kinase (JNK) and p38. Germinal center kinase couples TRAF2 to mitogen-activated protein kinase/ERK kinase kinase 1 and SAPK while receptor interacting protein associates with a mitogen-activated protein kinase kinase kinase upstream of MKK6 and p38. J Biol Chem, 1998. 273(35): p. 22681-92. 175

180. Yujiri, T., et al., MEK kinase 1 gene disruption alters cell migration and c-Jun NH2-terminal kinase regulation but does not cause a measurable defect in NF- kappa B activation. Proc Natl Acad Sci U S A, 2000. 97(13): p. 7272-7. 181. Nishitoh, H., et al., ASK1 is essential for JNK/SAPK activation by TRAF2. Mol Cell, 1998. 2(3): p. 389-95. 182. Yang, J., et al., The essential role of MEKK3 in TNF-induced NF-kappaB activation. Nat Immunol, 2001. 2(7): p. 620-4. 183. Karin, M. and Y. Ben-Neriah, Phosphorylation meets ubiquitination: the control of NF-[kappa]B activity. Annu Rev Immunol, 2000. 18: p. 621-63. 184. Schmitz, C., A. Kinner, and R. Kolling, The Deubiquitinating Enzyme Ubp1 Affects Sorting of the ABC-Transporter Ste6 in the Endocytic Pathway. Mol Biol Cell, 2005. 185. Strous, G.J., et al., Ubiquitin system-dependent regulation of growth hormone receptor signal transduction. Curr Top Microbiol Immunol, 2004. 286: p. 81-118. 186. Gesbert, F., N. Sauvonnet, and A. Dautry-Varsat, Clathrin-lndependent endocytosis and signalling of interleukin 2 receptors IL-2R endocytosis and signalling. Curr Top Microbiol Immunol, 2004. 286: p. 119-48. 187. Galan, J.M. and R. Haguenauer-Tsapis, Ubiquitin Lys63 is involved in ubiquitination of a yeast plasma membrane protein. EMBO J., 1997. 16: p. 5847- 5854. 188. Deng, L., et al., Activation of the IkB kinase complex by TRAF6 requires a dimeric ubiqutitin-conjugating enzyme complex and a unique polyubiquitin chain. Cell, 2000. 103: p. 351-361. 189. Wang, C., et al., TAK1 is a ubiquitin-dependent kinase of MKK and IKK. Nature, 2001. 412: p. 346-351. 190. Niedzwiedz, W. and K.J. Patel, "Dub"bing a tumor suppressor pathway. Cancer Cell, 2005. 7(2): p. 114-5. 191. Larsen, C.N., B.A. Krantz, and K.D. Wilkinson, Substrate specificity of deubiquitinating enzymes: ubiquitin C-terminal hydrolases. Biochemistry, 1998. 37(10): p. 3358-68. 192. Baek, K.H., Conjugation and deconjugation of ubiquitin regulating the destiny of proteins. Exp Mol Med, 2003. 35(1): p. 1-7. 193. Park, K.C., et al., Antagonistic regulation of myogenesis by two deubiquitinating enzymes, UBP45 and UBP69. Proc Natl Acad Sci U S A, 2002. 99(15): p. 9733-8. 194. Gong, L., et al., Identification of a novel isopeptidase with dual specificity for ubiquitin- and NEDD8-conjugated proteins. J Biol Chem, 2000. 275(19): p. 14212-6. 195. Hicke, L., A new ticket for entry into budding vesicles-ubiquitin. Cell, 2001. 106(5): p. 527-30. 196. Salghetti, S.E., et al., Regulation of transcriptional activation domain function by ubiquitin. Science, 2001. 293(5535): p. 1651-3. 197. Lam, Y.A., et al., Editing of ubiquitin conjugates by an isopeptidase in the 26S proteasome. Nature, 1997. 385(6618): p. 737-40. 198. Wilkinson, K.D., et al., Synthesis and characterization of ubiquitin ethyl ester, a new substrate for ubiquitin carboxyl-terminal hydrolase. Biochemistry, 1986. 25: p. 6644-6649. 176

199. Liu, L.Q., et al., A novel ubiquitin-specific protease, UBP43, cloned from leukemia fusion protein AML1-ETO-expressing mice, functions in hematopoietic cell differentiation. Mol Cell Biol, 1999. 19(4): p. 3029-38. 200. al-Katib, A.M., et al., Induced expression of a ubiquitin COOH-terminal hydrolase in acute lymphoblastic leukemia. Cell Growth Differ, 1995. 6(2): p. 211-7. 201. Lin, H., et al., Divergent N-terminal sequences target an inducible testis deubiquitinating enzyme to distinct subcellular structures. Mol Cell Biol, 2000. 20(17): p. 6568-78. 202. Zhu, Y., et al., DUB-1, a deubiquitinating enzyme with growth-suppressing activity. Proc Natl Acad Sci U S A, 1996. 93(8): p. 3275-9. 203. Zhu, Y., et al., DUB-2 is a member of a novel family of cytokine-inducible deubiquitinating enzymes. J Biol Chem, 1997. 272(1): p. 51-7. 204. Trompouki, E., et al., CYLD is a deubiquitinating enzyme that negatively regulates NF-kappaB activation by TNFR family members. Nature, 2003. 424(6950): p. 793-6. 205. Kovalenko, A., et al., The tumour suppressor CYLD negatively regulates NF- kappaB signalling by deubiquitination. Nature, 2003. 424(6950): p. 801-5. 206. Brummelkamp, T.R., et al., Loss of the cylindromatosis tumour suppressor inhibits apoptosis by activating NF-kappaB. Nature, 2003. 424(6950): p. 797- 801. 207. Bignell, G.R., et al., Identification of the familial cylindromatosis tumour- suppressor gene. Nat Genet, 2000. 25(2): p. 160-5. 208. Michal, M., et al., Spiradenocylindromas of the skin: tumors with morphological features of spiradenoma and cylindroma in the same lesion: report of 12 cases. Pathol Int, 1999. 49: p. 419-425. 209. Reynes, M., et al., Ultrastructural study of cylindroma (Poncet-Spiegler tumor). Journal of Cutan Pathology, 1976. 3: p. 95-101. 210. Biggs, P.J., et al., The cylindromatosis gene (cyld1) on chromosome 16q may be the only tumour suppressor gene involved in the development of cylindromas. Oncogene, 1996. 12(6): p. 1375-7. 211. Pierre, P., et al., CLIP-170 links endocytic vesicles to microtubules. Cell, 1992. 70(6): p. 887-900. 212. Saito, K., et al., The CAP-Gly domain of CYLD associates with the proline-rich sequence in NEMO/IKKgamma. Structure (Camb), 2004. 12(9): p. 1719-28. 213. Feng, S., et al., Two binding orientations for peptides to the Src SH3 domain: development of a general model for SH3-ligand interactions. Science, 1994. 266(5188): p. 1241-7. 214. Regamey, A., et al., The tumor suppressor CYLD interacts with TRIP and regulates negatively nuclear factor kappaB activation by tumor necrosis factor. J Exp Med, 2003. 198(12): p. 1959-64. 215. Brooke, H.G., Epithelioma adenoides cysticum. Br. J. Dermatol., 1892. 4: p. 269- 287. 216. Spiegler, E., Ueber endotheliome der haut. Arch. Derm. Syph., 1899. 50: p. 163- 176. 177

217. Aggarwal, B.B., Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol, 2003. 3(9): p. 745-56. 218. Cao, Z., et al., TRAF6 is a signal transducer for interleukin-1. Nature, 1996. 383: p. 443-446. 219. Aderem, A. and R.J. Ulevitch, Toll-like receptors in the induction of the innate immune response. Nature, 2000. 406: p. 782-787. 220. Sun, L., et al., The TRAF6 ubiquitin ligase and TAK1 kinase mediate IKK activation by BCL10 and MALT1 in T lymphocytes. Mol. Cell, 2004. 14: p. 289- 301. 221. Lorick, K.L., et al., RING fingers mediate ubiquitin-conjugating enzyme (E2)- dependent ubiquitination. Proc. Natl. Acad. Sci. U S A, 1999. 96: p. 11364-11369. 222. Shi, C.S. and J.H. Kehrl, TNF-induced GCKR and SAPK activation depends upon the E2/E3 complex Ubc13-Uev1A/TRAF2. J. Biol. Chem., 2003. 278: p. 15429- 15434. 223. Habelhah, H., et al., Ubiquitination and translocation of TRAF2 is required for activation of JNK but not of p38 or NF-kappaB. EMBO J., 2003. 23: p. 322-332. 224. Jensen, L.E. and A.S. Whitehead, Ubiquitin activated tumor necrosis factor receptor associated factor-6 (TRAF6) is recycled via deubiquitination. FEBS Letters, 2003. 553: p. 190-194. 225. Brummelkamp, T.R., et al., Loss of the cylindromatosis tumour suppressor inhibits apoptosis by activating NF-kappaB. Nature, 2003. 424: p. 797-801. 226. Kovalenko, A., et al., The tumour suppressor CYLD negatively regulates NF- kappaB signalling by deubiquitination. Nature, 2003. 424: p. 801-805. 227. Trompouki, E., et al., CYLD is a deubiquitinating enzyme that negatively regulates NF-kappaB activation by TNFR family members. Nature, 2003. 424: p. 793-796. 228. Karin, M. and M. Delhase, The I kappa B kinase (IKK) and NF-kappa B: key elements of proinflammatory signalling. Semin. Immunol., 2000. 12: p. 85-98. 229. Silverman, N. and T. Maniatis, NF-κB signaling pathways in mammalian and insect innate immunity. Genes & Dev., 2001. 15: p. 2321-2342. 230. Guha, M. and N. Mackman, LPS induction of gene expression in human monocytes. Cellular Signaling, 2001. 13: p. 85-94. 231. Kontoyiannis, D., et al., Impaired on/off regulation of TNF biosynthesis in mice lacking TNF AU-rich elements: implications for joint and gut associated immunopathologies. immunity, 1999. 10: p. 387-98. 232. Kotlyarov, A., et al., MAPKAP kinase 2 is essential for LPS-induced TNF-alpha biosynthesis. Nat. Cell. Bio., 1999. 1: p. 94-97. 233. Dumitru, C.D., et al., TNF-alpha induction by LPS is regulated posttranscriptionally via a Tpl2/ERK-dependent pathway. Cell, 2000. 103: p. 1071-1083. 234. Ip, Y.T. and R.J. Davis, Signal transduction by the c-Jun N-terminal kinase (JNK)--from inflammation to development. Curr. Opin. Cell. Biol., 1998. 10: p. 205-219. 235. Lin, A. and B. Dibling, The true face of JNK activation in apoptosis. Aging Cell, 2002. 1: p. 112-116. 178

236. Manning, A.M. and R.J. Davis, Targeting JNK for therapeutic benefit: from junk to gold? Nat. Rev. Drug. Discov., 2003. 2: p. 554-565. 237. Tournier, C., et al., MKK7 is an essential component of the JNK signal transduction pathway activated by proinflammatory cytokines. Genes Dev., 2001. 15: p. 1419-1426. 238. Bignell, G.R., et al., Identification of the familial cylindromatosis tumour- suppressor gene. Nat. Genet., 2000. 25: p. 160-165. 239. Harhaj, E.W. and S.-C. Sun, Regulation of RelA subcellular localization by a putative nuclear export signal and p50. Mol. Cell. Biol., 1999. 19: p. 7088-7095. 240. Coope, H.J., et al., CD40 regulates the processing of NF-kappaB2 p100 to p52. EMBO J., 2002. 15: p. 5375-5385. 241. Rivera-Walsh, I., et al., The NF-kappa B signaling pathway is not required for Fas ligand gene induction but mediates protection from activation-induced cell death. J. Biol. Chem., 2000. 275: p. 25222-25230. 242. Uhlik, M., et al., NF-kappaB-inducing kinase and IkappaB kinase participate in human T-cell leukemia virus I Tax-mediated NF-kappaB activation. J. Biol. Chem., 1998. 273: p. 21132-21136. 243. Sun, S.-C., et al., NF-κB controls expression of inhibitor IκBα: evidence for an inducible autoregulatory pathway. Science, 1993. 259: p. 1912-1915. 244. Regamey, A., et al., The tumor suppressor CYLD interacts with TRIP and regulates negatively nuclear factor kappaB activation by tumor necrosis factor. J. Exp. Med., 2003. 198: p. 1959-1964. 245. Yeh, W.C., et al., Early lethality, functional NF-kappaB activation, and increased sensitivity to TNF-induced cell death in TRAF2-deficient mice. Immunity, 1997. 7: p. 715-725. 246. Wertz, I.E., et al., De-ubiquitination and ubiquitin ligase domains of A20 downregulate NF-kappaB signalling. Nature, 2004. 247. Ting, A.T., F.X. Pimentel-Muinos, and B. Seed, RIP mediates tumor necrosis factor receptor 1 activation of NF-κB but not Fas/APO-1-initiated apoptosis. EMBO J., 1996. 15: p. 6189-6196. 248. Kelliher, M.A., et al., The death domain kinase RIP mediates the TNF-induced NF-kappaB signal. Immunity, 1998. 8: p. 297-303. 249. Lomaga, M.A., et al., TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40 and LPS signaling. Genes & Dev., 1999. 13: p. 1015-1024. 250. Lin, A. and M. Karin, NF-kappaB in cancer: a marked target. Semin. Cancer. Biol., 2003. 13: p. 107-114. 251. Tsuiki, H., et al., Constitutively active forms of c-Jun NH2-terminal kinase are expressed in primary glial tumors. Cancer Res., 2003. 63: p. 250-255. 252. Xu, X., et al., Constitutively activated JNK is associated with HTLV-1 mediated tumorigenesis. Oncogene, 1996. 13: p. 135-142. 253. Rodrigues, G.A., M. Park, and J. Schlessinger, Activation of the JNK pathway is essential for transformation by the Met oncogene. EMBO J., 1997. 16: p. 2634- 2645. 254. Behrens, A., et al., Oncogenic transformation by ras and fos is mediated by c-Jun N-terminal phosphorylation. Oncogene, 2000. 19: p. 2657-2663. 179

255. Potapova, O., et al., c-Jun N-terminal kinase is essential for growth of human T98G glioblastoma cells. J. Biol. Chem., 2000. 275: p. 24767-24775. 256. Potapova, O., et al., Inhibition of c-Jun N-terminal kinase 2 expression suppresses growth and induces apoptosis of human tumor cells in a - dependent manner. Mol. Cell. Biol., 2000. 20: p. 1713-1722. 257. Chen, N., et al., Suppression of skin tumorigenesis in c-Jun NH(2)-terminal kinase-2-deficient mice. Cancer Res., 2001. 61: p. 3908-3912. 258. Yang, Y.M., et al., C-Jun NH(2)-terminal kinase mediates proliferation and tumor growth of human prostate carcinoma. Clin. Cancer Res., 2003. 9: p. 391-401. 259. Hess, P., et al., Survival signaling mediated by c-Jun NH2-terminal kinase in transformed B lymphoblasts. Nature Genet., 2002. 32: p. 201-205. 260. Lamb, J.A., et al., JunD mediates survival signaling by the JNK signal transduction pathway. Mol. Cell, 2003. 11: p. 1479-1489. 261. Zhang, J.Y., et al., NF-kappaB RelA opposes epidermal proliferation driven by TNFR1 and JNK. Genes Dev., 2004. 18: p. 17-22. 262. Lin, A. and B. Dibling, The true face of JNK activation in apoptosis. Aging Cell., 2003. 1: p. 112-116. 263. Liu, Z.G., Adding facets to TNF signaling. The JNK angle. Mol. Cell, 2003. 12: p. 795-796. 264. Varfolomeev, E.E. and A. Ashkenazi, Tumor necrosis factor: an apoptosis JuNKie? Cell, 2004. 116: p. 491-497. 265. Chen, Y.R., et al., The role of c-Jun N-terminal kinase (JNK) in apoptosis induced by ultraviolet C and gamma radiation. Duration of JNK activation may determine cell death and proliferation. J. Biol. Chem., 1996. 271: p. 31929-31936. 266. Guo, Y.L., et al., Correlation between sustained c-Jun N-terminal protein kinase activation and apoptosis induced by tumor necrosis factor-alpha in rat mesangial cells. J. Biol. Chem., 1998. 273: p. 4027-4034. 267. Sakon, S., et al., NF-kappaB inhibits TNF-induced accumulation of ROS that mediate prolonged MAPK activation and necrotic cell death. Embo J, 2003. 22(15): p. 3898-909. 268. Himeno, T., et al., Expression of endogenous tumor necrosis factor as a protective protein against the cytotoxicity of exogenous tumor necrosis factor. Cancer Res., 1990. 50: p. 4941-4945. 269. Hershko, A. and A. Ciechanover, The ubiquitin system. Annu. Rev. Biochem., 1998. 67: p. 425-479. 270. Fischer, J.A., Deubiquitinating enzymes: their roles in development, differentiation, and disease. Int. Rev. Cytol., 2003. 229: p. 43-72. 271. Kim, J.H., et al., Deubiquitinating enzymes as cellular regulators. J. Biochem., 2003. 134: p. 9-18. 272. Karin, M. and Y. Ben-Neriah, Phosphorylation meets ubiquitination: the control of NF-[kappa]B activity. Annu. Rev. Immunol., 2000. 18: p. 621-663. 273. Davis, R.J., Signal transduction by the JNK group of MAP kinases. Cell, 2000. 103: p. 239-252. 274. Weinberg, R.A., Tumor suppressor genes. Science, 1991. 254(5035): p. 1138-46. 275. Naviaux, R.N., et al., The pCL vector system: rapid production of helper-free high-titer, recombinant retroviruses. J. Virol., 1996. 70: p. 5701-5705. 180

276. Xiao, G., E.W. Harhaj, and S.C. Sun, NF-kappaB-inducing kinase regulates the processing of NF-kappaB2 p100. Mol. Cell., 2001. 7: p. 401-409. 277. Harhaj, E.W., et al., Somatic mutagenesis studies of NF-kappa B signaling in human T cells: evidence for an essential role of IKK gamma in NF-kappa B activation by T-cell costimulatory signals and HTLV-I Tax protein. Oncogene, 2000. 19: p. 1386-1391. 278. Rivera-Walsh, I., et al., NF-κB signaling pathway governs TRAIL gene expression and HTLV-I Tax-induced T-cell death. J. Biol. Chem., 2001. in press (published online). 279. DiDonato, J.A., et al., A cytokine-responsive IκB kinase that activates the transcription factor NF-κB. Nature, 1997. 388: p. 548-554. 280. Uchihara, J.N., et al., Transactivation of the CCL5/RANTES gene by Epstein- Barr virus latent membrane protein 1. Int J Cancer, 2004. 114(5): p. 747-755. 281. Mori, N., et al., Elevated expression of CCL5/RANTES in adult T-cell leukemia cells: possible transactivation of the CCL5 gene by human T-cell leukemia virus type I tax. Int J Cancer, 2004. 111(4): p. 548-57. 282. Sil, A.K., et al., IkappaB kinase-alpha acts in the epidermis to control skeletal and craniofacial morphogenesis. Nature, 2004. 428: p. 660-664. 283. Anest, V., et al., nucleosomal function for IkappaB kinase-alpha in NF-kappaB- dependent gene expression. Nature, 2003. 423: p. 659-663. 284. Yamamoto, Y., et al., Histone H3 phosphorylation by IKK-alpha is critical for cytokine-induced gene expression. Nature, 2003. 423: p. 655-659. 285. Yamamoto, Y. and R.B. Gaynor, IkappaB kinases: key regulators of the NF- kappaB pathway. Trends Biochem. Sci., 2004. 29: p. 72-79. 286. Senftleben, U., et al., Activation of IKKα of a second, evolutionary conserved, NF-κB signaling pathway. Science, 2001. 293: p. 1495-1499. 287. Stenger, S., et al., Granulysin: a lethal weapon of cytolytic T cells. Immunol Today, 1999. 20(9): p. 390-4. 288. Seder, R.A. and W.E. Paul, Acquisition of lymphokine-producing phenotype by CD4+ T cells. Annu Rev Immunol, 1994. 12: p. 635-73. 289. Paul, W.E. and R.A. Seder, Lymphocyte responses and cytokines. Cell, 1994. 76(2): p. 241-51. 290. Weih, F., D. Carrasco, and R. Bravo, Constitutive and inducible Rel/NF-kappa B activities in mouse thymus and spleen. Oncogene, 1994. 9(11): p. 3289-97. 291. Whitmarsh, A.J. and R.J. Davis, Transcription factor AP-1 regulation by mitogen- activated protein kinase signal transduction pathways. J Mol Med, 1996. 74(10): p. 589-607. 292. Ip, Y.T. and R.J. Davis, Signal transduction by the c-Jun N-terminal kinase (JNK)--from inflammation to development. Curr Opin Cell Biol, 1998. 10(2): p. 205-19. 293. Yang, D.D., et al., Differentiation of CD4+ T cells to Th1 cells requires MAP kinase JNK2. Immunity, 1998. 9(4): p. 575-85. 294. Alberola-Ila, J., et al., Selective requirement for MAP kinase activation in thymocyte differentiation. Nature, 1995. 373(6515): p. 620-3. 181

295. Caamano, J. and C.A. Hunter, NF-kappaB family of transcription factors: central regulators of innate and adaptive immune functions. Clin Microbiol Rev, 2002. 15(3): p. 414-29. 296. Doi, T.S., et al., NF-kappa B RelA-deficient lymphocytes: normal development of T cells and B cells, impaired production of IgA and IgG1 and reduced proliferative responses. J Exp Med, 1997. 185(5): p. 953-61. 297. Burkly, L., et al., Expression of relB is required for the development of thymic medulla and dendritic cells. Nature, 1995. 373(6514): p. 531-6. 298. Kontgen, F., et al., Mice lacking the c-rel proto-oncogene exhibit defects in lymphocyte proliferation, humoral immunity, and interleukin-2 expression. Genes Dev, 1995. 9(16): p. 1965-77. 299. Sha, W.C., et al., Targeted disruption of the p50 subunit of NF-kappa B leads to multifocal defects in immune responses. Cell, 1995. 80(2): p. 321-30. 300. Reiley, W., M. Zhang, and S.C. Sun, Negative regulation of JNK signaling by the tumor suppressor CYLD. J Biol Chem, 2004. 279(53): p. 55161-7. 301. Reiley, W., et al., Regulation of the Deubiquitinating Enzyme CYLD by IKKgamma-Dependent Phosphorylation. Mol Biol Cell, 2005. In Press. 302. Warren, M.K. and S.N. Vogel, Bone marrow-derived macrophages: development and regulation of differentiation markers by colony-stimulating factor and interferons. J. Immunol., 1985. 134: p. 982-989. 303. Sabapathy, K., et al., JNK2 is required for efficient T-cell activation and apoptosis but not for normal lymphocyte development. Curr Biol, 1999. 9(3): p. 116-25. CURRICULUM VITA William Walter Reiley

Biographical Information:

Born: 4 May, 1977

Academic Background:

1996-2000: Rochester Institute of Technology, Rochester, New York Bachelor of Science Degree in Biotechnology

2000-Present Graduate Student in Cell and Molecular Biology Program, The Pennsylvania State University College of Medicine Milton S. Hershey Medical Center, Hershey, PA

Honors and Awards:

NIH Training Grant “Virus and Cancers”

Publications:

Reiley, W., Zhang, M., Norbury, C., and Sun, SC. Regulation of T Cell Development and Activation by the Deubiquitination Enzyme CYLD. Manuscript in preparation.

Reiley, W., Zhang, M., Wu, X., Granger, E., and Sun, SC., Regulation of the Deubiquitinating Enzyme CYLD by IKKgamma-Dependent Phosphorylation. Mol Biol Cell, 2005. In Press.

Morrison, M.D., Reiley, W., Zhang, M., and Sun, SC., An atypical TRAF-binding motif of BAFF receptor mediates induction of the noncanonical NF-kB signaling pathway. J Biol Chem, 2005.

Reiley, W., M. Zhang, and S.C. Sun, Negative regulation of JNK signaling by the tumor suppressor CYLD. J Biol Chem, 2004. 279(53): p. 55161-7.

Waterfield, M., Jin, W., Reiley, W., Zhang, M., and Sun, SC., IKKb Is an essential component of the Tpl2 signaling pathway. Mol. Cell. Biol., 2004. 24(13) p. 6040-8.