An Abstract of the Dissertation Of

AN ABSTRACT OF THE DISSERTATION OF

Wisam Hussein Selman for the degree of Doctor of Philosophy in

Pharmaceutical Sciences presented on June 12, 2018.

Title: Bcl11b is at the Nexus of Wnt and MAP Kinase Signaling Pathways in

Two Stages of Thymocyte Development.

Abstract approved: ______Theresa M. Filtz,

Leukemogenesis, uncontrolled proliferation of dysfunctional, transformed immature lymphocytes, appears, especially in children, to be associated with abrogation of normal lymphopoiesis, and thymopoiesis in the case of T cell leukemias. Sequence specific transcription factors (SSTFs) are nuclear control switches that respond to cellular signals to alter cell state and cell fate by controlling gene transcription. Their activities are highly and precisely regulated by diverse means such as alterations in protein stability, post- translational modifications (PTMs) and/or protein-protein interactions. Several

SSTFs have been identified as “T cell lineage-determining factors” essential for proper maturation of thymocytes into T cells in immune system

development. Mutations or dysregulation of these T cell lineage-specifying

SSTFs are associated with leukemogenesis.

SSTFs act in coordination to regulate gene transcription in multi-protein complexes. There are millions of potential combinations of SSTFs available to fine-tune repression or activation of individual genes in response to multiple, complex extracellular and intracellular signaling inputs. To narrow our studies to SSTFs most likely to be of consequence in leukemogenesis, we focused on modifications and interactions between lineage-determining SSTFs in the thymocyte maturation process. Tcf1 is a T cell lineage-determining SSTF and has been identified as the nuclear effector of Wnt/Gsk3/β-catenin -dependent signaling. Bcl11b is a nuclear target of the T-cell receptor (TCR)-mediated

MAP Kinase (MAPK) signaling cascade, but there is no indication from the literature that an interaction between Tcf1 and Bcl11b has been investigated.

As an impetus for this investigation, another group demonstrated the ability of

Bcl11b to regulate Wnt target genes in a different system, but a molecular link between the Wnt signaling pathway and Bcl11b was not established. Through our studies in primary murine thymocytes (mostly double positive (DP) cells), we identified a previously unknown physical interaction between Bcl11b and

Tcf1, and among Bcl11B, TCF1 and β-catenin in Jurkat cells, a human T cell leukemia cell line. Our further studies in mutant Jurkat cells revealed that neither sumoylation nor up to 285 C terminal amino acids in Bcl11b are required for BCL11B to complex with TCF1 or β-catenin. We identified several thousand gene promoter regions that are co-occupied by the two factors,

suggesting that the Bcl11b-Tcf1 interaction may coordinately regulate many genes required for thymocyte maturation. Using a gene reporter assay, we identified that the Bcl2l1 gene, encoding for the anti-apoptotic protein Bcl-xl, is co-regulated by BCL11B and TCF1.

Treatment of thymocytes to mimic stimulation of the Wnt/Gsk3 signaling pathway also resulted in significant changes in the PTMs of Bcl11b, including increased sumoylation and decreased phosphorylation. Previous studies identified two kinases that are involved in modification of Bcl11b, Erk1/2 and p38 MAPK; in this study we identified a third candidate kinase, Gsk3, that may be responsible for basal phosphorylation. Inhibition of Gsk3 in thymocytes, mimicking canonical Wnt pathway activation, reduced the Tcf1-

Bcl11b interaction, suggesting a potential mechanism by which Bcl11b regulates Wnt target genes.

By investigating DN3-like P2C2 cells, a cell line from an earlier stage in thymocyte development relative to DP cells, we observed different patterns of

PTM kinetics in response to signaling pathway activation, as compared to primary thymocytes. Treatment to mimic pre-TCR/MAP kinase pathway activation in P2C2 cells mostly replicated findings in primary thymocytes in terms of BCL11B PTM kinetic changes, with the P2C2 cells having a slightly delayed time course overall. However, this delay creates a situation in which desumoylation and dephosphorylation occurs simultaneously, in contrast to the opposing modifications by sumoylation and phosphorylation observed in

DP thymocytes in response to MAPK pathway activation. Nevertheless, the

opposite regulation of phosphorylation and sumoylation that we observed in primary thymocytes is mimicked under other conditions in the DN-like cells.

Hyper-phosphorylation of Bcl11b after treatment with a broad phosphatase inhibitor coincided with near complete desumoylation. Additionally, activation of Wnt/Gsk3-dependent signaling in P2C2 cells resulted in composite dephosphorylation and sumoylation of Bcl11b. P2C2 cells displayed a higher basal level of β-catenin than primary thymocytes, and a higher ratio of long

(β-catenin responsive) to short (β-catenin independent) Tcf1 isoforms. Unlike in DP thymocytes, we did not observe an interaction between Bcl11b and

Tcf1 in the DN-like cells. Our results demonstrate that PTMs and specific interacting partners of Bcl11b in P2C2 cells are distinct from those previously characterized in DP thymocytes.

Our results contribute to the existing knowledge regarding dynamic regulation of Bcl11b PTMs, and the interaction of Bcl11b with specific transcriptional complexes during distinct stages of T cell development. These findings may contribute to a better understanding of the molecular mechanisms that regulate Bcl11b transcriptional activity in different cell types, potentially leading to novel treatments for T cell leukemias.

Bcl11b is at the Nexus of Wnt and MAP Kinase Signaling Pathways in Two Stages of Thymocyte Development.

by Wisam Hussein Selman

A DISSERTATION

submitted to

Oregon State University

in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Presented June 12, 2018 Commencement June 2019

Doctor of Philosophy dissertation of Wisam Hussein Selman presented on June 12, 2018

APPROVED:

Major professor, representing Pharmaceutical Sciences

Dean of the College of Pharmacy

Dean of the Graduate School

I understand that my dissertation will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my dissertation to any reader upon request.

Wisam Hussein Selman, Author

ACKNOWLEDGEMENTS

I would first and foremost like to express my sincere thanks to Dr.

Theresa Filtz for allowing me to pursue my PhD study in her laboratory. I would also like to thank her for her patient guidance, encouragement, kindness, and for allowing me to think and work freely and independently on this project. Dr. Filtz’s office was always open whenever I had a question about my research or whenever I needed help. Without her advice and support, this research would have never been accomplished. I really appreciate her efforts.

I would like to acknowledge my graduate committee members. I am grateful to Dr. Mark Leid for his support and his stimulating suggestions. I have enjoyed spending time in his office with him teaching me a lot about BCL11B. I would also to thank Dr. Chrissa Kioussi for her support and kindness, and for her valuable advice. She continuously provided us with mice to perform our experiments. Special thanks to Dr. Adam Alani for both his genial demeanor and for his extraordinary support during my study. I would also like to thank Dr.

Siva Kolluri for kindly agreeing to be a member on my graduate committee.

I would like to thank my sponsor in Iraq (HCED) for their financial support and for giving me the opportunity to complete my PhD study in the

United States.

I am grateful to the College of Veterinary Medicine, University of Al-

Qadisiyah, for their support and for keeping my job position for me during my

PhD study.

I would like to express my thanks and full appreciation to the College of

Pharmacy for their financial support and appreciation through awarding the

Holt Scholarship to me.

I would like also to thank Vera C. Lattier, Elahe Esfandiari, and Arun

Singh for helping with all animal work, and Dr. Arup Indra and Dr. Jane

Ishmael for sharing antibodies with us.

I am forever grateful to my parents for their prayers overseas, providing me moral and emotional support, and unconditional love. Mom and Dad, you both have been the best. I really missed your hugs in the last five years.

I am very grateful for my lovely wife Noora for her love, for sacrificing five years of her career to support me to finish my study, for waiting for me for lonely, countless long hours while I was working in the lab. It has meant so much to me. She poured her life into our children, Abdullah, Lobna, and

Mujtaba. I am proud to have such a great wife and beautiful kids in my life.

CONTRIBUTION OF AUTHORS Theresa Filtz helped to conceive and design all the studies and edited the dissertation. Wisam Selman wrote all chapters in the dissertation, and was involved in conception, design, performance, analysis and figure generation for all experiments and chapters except where explicitly noted below

Chapter 2: Walter K. Vogel, Ph.D., conceived, designed, and performed the experiments associated with Figure 2.6A, analyzed the data sets, prepared

Figure 2.6A and associated text, and created the BCL11B-crosslinked agarose. Arun Singh, Ph.D., analyzed the data sets used to create Figure

2.6B. Hua Wise, Ph.D., created the Bcl2l1-luciferase reporter construct, subcloned human BCL11B into pcDNA3 plasmid, and designed the protocol for and created the two mutant Jurkat cell lines by using CRISPR/Cas9- mediated genome editing. Alamjit K. Nagra, undergraduate researcher, performed the Bcl2l1-luciferase reporter assay with results displayed in Figure

2.6C & D, and performed transfection of HEK293T cells needed for results in

Figure 2.6B. Mark Leid, Ph.D., helped with study design and data analysis, and provided technical advice, antibody reagents and edited the chapter.

TABLE OF CONTENTS

Chapter Page

Chapter 1…………………………………………………………………………….1 General Introduction

1.1 Acute lymphoblastic leukemia…….………………………………………….2

1.2 T cell development..……………………………………………………....3

1.3 T cell receptor complexes and thymocyte development………………4

1.4 Structural differences between pre-TCR and TCR complexes……....5

1.5 Differences in intracellular signaling pathways initiated via pre-TCR and TCR complexes ………………………………………………………………..7

1.6 T cell development and MAPK signaling……………………………...10

2 Transcription Factors in T cell development ………………………….……...11

2.1 BCL11B……………………………………..……………………………….12

2.2 TCF1………………………………………………….……………………..13

3 Posttranslational modifications regulate the activity of SSTFs……………..15

3.1 Phosphorylation……………………………………………………………..17

3.2 Sumoylation………………………………………………………………….19

3.3 Ubiquitination………………………………………………………………..21

4 Research Objectives ……………………………………………………….23

5 References……………………………………………………………………….24

TABLE OF CONTENTS (Continued) Chapter Page

Chapter 2…………………………………………………………………….……..35 Alteration of Bcl11b upon stimulation of both MAP Kinase- and GSK3-dependent signaling pathways in double negative thymocytes

2.1 Abstract……………………………………………………………………..…35

2.2 Introduction……………………………………………………………………37

2.3 Methods………………………………………………………………………..41

2.4 Results…………………………………………………………………………48

2.5 Discussion……………………………………………………………………...59

2.6 Figures………………………………………………………………………....68

2.7 References……………………………………………………………………..87

Chapter 3………………………………………………………………….………..94 Alteration of Bcl11b upon stimulation of both MAP Kinase- and GSK3- dependent signaling pathways in double negative thymocytes

3.1 Abstract………………………………………………………………….……..95

3.2 Introduction……………………………………………………………………96

3.3 Methods………………………………………………………………………..99

3.4 Results………………………………………………………………………..102

3.5 Discussion…………………………………………………………………….107

3.6. Figures……………………………………………………………………….115

3.6 References….……………………………………………………………….126

Chapter 4…………………………………………………………………………129

TABLE OF CONTENTS (Continued)

General Conclusion 4.2 References……………………………………………………………………136

LIST OF FIGURES

Figure Page

1.1. Stages of T cell development.………………………………...………..…..4

1.2. Differences between pre-TCR and immature TCR receptors: …………….…7

1.3. Signaling molecules linked to pre-TCR activation ..………….……………..10

2.1. Bcl11b interacts with TCF1 in thymocytes...………………………….....69-70

2.2. Bcl11b-TCF1 interaction dynamics following stimulation of thymocytes ……………………………………………………………………………………..71-72

2.3. BCL11B can simultaneously complex with β-catenin and TCF1 in Jurkat cells……………………………………………………………………………….73-74

2.4. Dynamic alterations in Bcl11b-TCF1 complexes.………………………75-76

2.5. BCL11B sumoylation is not critical for interaction with TCF1 and β-catenin in Jurkat cells...………………………………………………………..……...…77-78

2.6. TCF1 interacts with BCL11B and inhibits its transcriptional activator activity on the Bcl2l1 promoter……………………………………………………….…79-81

2.7. Phosphorylation dynamics of Bcl11b and changes in activity of Gsk3α/β and Erk1/2 after LiCl treatment of thymocytes...………………………….....……82-83

2.8. Dynamic sumoylation and ubiquitinylation of Bcl11b after LiCl treatment in thymocytes……………………………………………………..…………..……84-86

3.1. Phosphorylation kinetics of Bcl11b and changes in the activity of Erk1/2 in P2C2 cells after stimulation with P/A………………………………………115-116

LIST OF FIGURES (Continued) Figure Page 3.2. Sumoylation kinetics of Bcl11b after stimulation with P/A and inhibition of phosphatase in P2C2 cells.…………………………………………...……117-118

3.3. Bcl11b degradation and ubiquitination in P2C2 cells with extended P/A treatment…………………………………………………………..………119-120

3.4. LiCl treatment of P2C2 cells and phosphorylation kinetics of Bcl11b, Gsk3α/β and Erk1/2 activity and accumulation of β-catenin ………………………121-122

LIST OF FIGURES (Continued)

3.5. Sumoylation (SUMO1) kinetics of Bcl11b after LiCl treatment of P2C2 cells...……………………………………………………………………………....123

3.6. Bcl11b degradation and ubiquitination in P2C2 cells following extended treatment with LiCl...……………………………………………..……………..…124

3.7. Bcl11b and Tcf1 in P2C2 cells.……………………………………………….125

LIST OF TABLES

Table Page

2.1. Sequences of oligonucleotides used for CrispR-Cas9 mutation and sequencing…………………………………………………………………………46

4.1: Treatment effects on Bcl11b PTMs……………………………………….134

Bcl11b is at the nexus of Wnt and MAP Kinase signaling pathways in two stages of thymocyte development

Chapter 1

General introduction

1.1 Acute lymphoblastic leukemia

Acute lymphoblastic leukemia (ALL) is the most common childhood malignancy, and arises from clonal expansion of malignant transformed lymphoid progenitors [1-3]. Eighty percent of ALL cases are diagnosed in children.

However, ALL in adults is considered one of the most challenging malignancies as regards to therapy with only a 40% long term, causes-free survival rate [1, 2].

T-cell acute lymphoblastic leukemia (T-ALL) represents 15-25% of all ALL cases

[4], and has a worse prognosis than B-cell lineage ALL [5, 6]. As an aggressive disease that invades multiple organs and is associated with immune deficiency, initial prognosis was historically poor [7]. Modern intensive chemotherapy regimens have vastly improved the 10 year remission rate [8, 9]. However, the long term medical consequences of chemotherapy treatment are unpredictable and include, but are not limited to, early mortality [10, 11]. Therefore, there is a critical need to develop better therapeutic agents with fewer side effects and lower lethality.

Childhood T-ALL is understood to be caused by abnormal maturation of thymocytes which leads to malignant transformation of immature T cells [12]. A detailed understanding of the molecular mechanisms that orchestrate the normal process of T cell development will enable us to better understand the aberrant events that lead to leukemogenesis and provide insights into new pharmacological target(s) with the ultimate goal of improving treatment options in

T-ALL.

1.2 T cell development

Intra-thymic T cell maturation proceeds in discrete stepwise developmental stages to produce appropriate numbers of functional mature T cells [13]. T cell development starts when thymocyte progenitors from fetal liver or from bone marrow after birth enter the thymus where differentiation is tightly controlled by cell-bound and soluble activators [14-17]. The earliest thymocyte immature progenitors lack expression of the key cell surface markers CD4 and CD8 and are therefore termed double negative (DN) thymocytes [14, 17]. The DN thymocytes are further subdivided into at least four consecutive developmental stages, subtyped based on expression of CD44 and CD25 cell surface markers: DN1

(CD44+ CD25-), DN2 (CD44+ CD25+), DN3 (CD44- CD25+), and DN4 (CD44-

CD25-)[18, 19]. Subsequent developmental stages are characterized by expression of CD4 and/or CD8 and include immature single positive (ISP, CD8+), double positive (DP, CD4+ CD8+) and single positive (SP, CD4+ or CD8+) thymocytes [18, 20, 21]. Maturation of DN to DP thymocytes is mediated by the essential differentiation checkpoint known as β-selection [22, 23]. DP thymocytes are winnowed by positive and negative selection and progress into mature CD4+ or CD8+ SP cells which are then released into the peripheral circulation [18, 20,

21] (Fig. 1.1)

Figure 1.1. Stages of T cell development. Thymocytes undergo distinct stages in development to T cells, starting with migration of bone-marrow (BM)-derived T cell progenitors (orange band) into thymus (yellow and green bands). In the thymus, cells commit to the T cell lineage, and initiate further differentiation and specification of T cell subtypes; . Next, the immature T cells distribute into the blood and lymphatic systems (periphery, pink) accompanied by final maturation and specification of the CD4 and CD8 SP subtypes. Arrows indicate stages of development in which an essential role of the BCL11B transcription factor has been established. Abbreviations for cell subtypes: B cells (B), macrophages (mɸ), dendritic cells (DC), natural killer (NK), natural killer T cells (NKT), ɣδ T cells (ɣδ), regulatory T cells (Treg), helper T cells (Th), cytotoxic T lymphocytes (CTL), double negative thymocytes (DN), intermediate single positive thymocytes (ISP), double positive thymocytes (DP), single CD4 or CD8 positive thymocytes (CD4 or CD8). Adapted from [24].

1.3 T cell receptor complexes and thymocyte development

Developmental progression of thymocytes into mature T cells is associated with changes in composition of the plasma membrane-spanning T cell receptor

(TCR) complexes [23] that respond to antigens (see Fig. 1.2). Initially, a pre-TCR

5 complex is expressed on DN3 thymocytes that is critical for development to the

DP stage [8, 25]. Next, an immature TCR complex is initially expressed on DP thymocytes, followed by the fully mature TCR complex expressed on mature αβ T cells [26]. TCR complexes are formed by heterodimerized α and β chains covalently linked by disulfide bonds and complexed non-covalently with CD3 subunits γ, δ, ε and ζ [26-29] .

1.4 Structural differences between pre-TCR and TCR complexes

Differences among the pre-, immature-, and mature TCR complexes appear to reside primarily in the α and β chains. The pre-TCR complex is composed of a non-completely rearranged TCRβ chain covalently paired with an invariant, nascent pre-TCRα chain (pre-Tα), forming a pre-Tα/non-completely rearranged TCR β heterodimer [1-3]. The immature-TCR complex is composed of a rearranged TCR β chain covalently bound to a nascent non-completely rearranged TCRα chain [30]. The mature TCR complex is composed of heterodimerized, highly variable rearranged α and β chains linked by disulfide bonds [26].

Pre-TCR signaling pathway components in DN thymocytes are similar to the TCR receptor in regards to some proximal signaling components, but there are structural differences between them [25]. Distinct from the TCR initially expressed on DP thymocytes, the murine pre-Tα chain in the pre-TCR has only one IgG-like extracellular loop compared to two loops in the TCRα chain [31]. In addition, the pre-Tα has a longer cytoplasmic tail of about 30 amino acids in

6 mouse pre-Tα versus 6 amino acids in mouse TCRα suggesting a specific role for the pre-Tα in DN thymocytes [31] (Fig. 1.2). DN cells devoid of a pre-Tα subunit are shunted from the dominant αβ T cell lineage pathway towards the γδ

T cell lineage [32, 33]. Replacement of the pre-Tα subunit with TCRα at the DN stage does not rescue pre-Tα negative DN cells from a primarily γδ T cell lineage, and instead results in increased apoptosis and decreased proliferation of DN thymocytes. These studies indicated that the pre-Tα is uniquely required for proper maturation of DN thymocytes and the TCRα is not a substitute [33].

In contrast to the TCR complex, pre-TCR complexes are expressed on the plasma membrane of immature DN thymocytes at a very low level, about 100- fold lower than expression of TCR complexes [34]. In terms of structural differences between pre-TCR and TCR complexes, besides the α subunit, the

CD3ζ subunit associates more weakly with the pre-TCR, but it is still physiologically linked to pre-TCR complex signaling [35]

Immature TCRs are composed of more CD3 γε subunit pairs and fewer

CD3 δε pairs relative to mature TCRs. Each of these non-covalent heterodimers is associated with a specific CD3-associated protein (CD3AP) and can form signaling complexes independent of a TCRα chain, called the clonotype- independent CD3 (CIC) complex. CICs are expressed in early DN thymocytes through the DP stage [36]. Upon stimulation with anti-CD3 antibody, CIC complexes are capable of transducing intracellular signals to drive maturation of

DN thymocytes to DP thymocytes both in vivo and in vitro [5, 37-39].

Figure 1.2. Differences between pre-TCR and immature TCR receptors: Structures of the pre-TCR on DN thymocytes and immature TCR on DP thymocytes. External domains are shown above the yellow bar indicating the plasma membrane. All subunits contain a single transmembrane domain. Grey bars overlaying internal structures indicate immunoreceptor tyrosine-based activation motifs (ITAMs). The cartoon illustrates covalent association of the TCRβ subunit with either invariant pre-Tα chain in the pre-Tα complex or non- completely rearranged TCRα chain in the immature TCR complex with CD3ε, δ, γ, and ζ subunits. The pre-Tα subunit in the pre-TCR has only one IgG like extracellular domain compared with two IgG-like extracellular domains of the TCRα subunit. Additionally, the pre-Tα subunit has a longer cytoplasmic tail compared to TCRα. Illustration adapted from [16, 23, 26, 31]

1.5 Differences in intracellular signaling pathways initiated via pre-TCR and

TCR complexes

Downstream of the different TCR complexes in maturing thymocytes--pre-

TCR, immature TCR and mature TCR--the intracellular signaling pathways are

8 likely wired into different networks causing divergent functions [23]. For example, pre-TCR-expressing DN3 thymocytes exist in a distinct cellular context from DP cells and respond distinctly to certain stimuli. Stimulation of DN and DP thymocytes with the same anti-CD3 antibody resulted in proliferation of DN cells but death of DP thymocytes [40]. The structural and mechanistic basis for this and other different outcomes are not well understood. Another example, protein kinase C θ (PKCθ), a Ca++-independent PKC expressed selectively in T lymphocytes [41], is critical for T cell activation through mature TCR signaling

[42]. In contrast, PKCθ does not appear to play a leading role in mediating signaling downstream of the pre-TCR and immature TCR complexes [23, 43].

The exact mechanism by which pre-TCR signaling is triggered remains elusive [25]. Unlike TCR activation, pre-TCR engagement with a specific ligand

(antigen) does not appear to be required for initiation of pre-TCR signaling [25,

44, 45]. Two alternative mechanisms have been proposed in initiation of pre-TCR signaling, both of which suggest that assembly of the pre-TCR subunits is sufficient [23, 25]. One model holds that surface expression of the pre-TCR complex is required for activation of pre-TCR signaling [25] while a competing model suggests that internal assembly is sufficient for initiation of pre-TCR signaling [25].

Upon pre-TCR activation, downstream signaling events seem to depend on activation of complex signaling molecules involving, but not limited to, recruitment of the non-receptor tyrosine kinase LCK/Fyn [23, 46-48]. LCK, which

9 plays a central role in regulating proximal signaling after pre-TCR assembly [49], phosphorylates several immunoreceptor tyrosine-based activation motifs (ITAMs) in the cytoplasmic domain of the pre-TCR ζ chain to provide docking sites for

Zap-70/Syk kinase [16, 23]. Zap-70/Syk is activated by phosphorylation after binding to phosphorylated ζ chain [23]. Activated ZAP-70/Syk recruits a guanine nucleotide exchange protein such as Vav-1 [23, 50] and phosphorylates the adapter proteins SLP-76 and LAT [23, 51]. SLP-76 activates several kinases which activate downstream signaling enzymes/pathways including phospholipase

C γ1 (PLCγ1), PKC, calcium mobilization, and the Ras-Raf –MAPK (Erk-1/2) cascade [16, 52] ( Fig. 1.3).

Figure 1.3. Signaling molecules linked to pre-TCR activation. Signal initiated by stimulation of pre-TCR complex activates (phosphorylates) several proximal kinases such as LCK, and Zap-70/Syk which activate adaptor proteins such as LAT and SLP-76 leading to activation of several kinases which in turn activate PLCγ1. Activated PLCγ1 hydrolyzes the membrane lipid PIP2 generating ++ IP3 and DAG. IP3 and DAG in turn trigger pathways which increase cytosolic Ca and activate PKC. Activated PCK initiates the Raf-Erk1/2 cascade. Activation of PKC, Erk1/2, and accumulation of cytosolic Ca++ contribute to stimulation of cellular programs. Illustration adapted from [23, 26, 53-55]

1.6 T cell development and MAPK signaling

Mitogen-activated protein kinases (MAPK) are protein serine/threonine kinases that play vital roles in transduction of extracellular signals to a network of

11 cellular response programs such as proliferation, differentiation and apoptosis

[56, 57]. Each MAPK is composed of at least three sequentially acting kinases: a

MAPK kinase kinase (MAPKKK or MEKK), a MAPK kinase (MAPKK or MEK) and

MAPK [56, 57]. In thymocytes, activation of Ras through activation of pre-TCR or

TCR signaling pathways results in recruitment of Raf1 which acts as a MAPKKK, which will then activate a series of MEK and MAPKs including, but not limited to,

Erk1/2 [52, 58]. Erk1/2 activation ultimately results in changes in gene expression through regulation of transcription factors [23, 52, 58]. Interestingly, it has been reported that PCK activation is sufficient to activate the MEK-Erk pathway independent of Ras but dependent on Raf1 [55, 58]. Erk is necessary for proliferation and survival of DN3 thymocytes, but is not required for differentiation of DN4 cells to the DP stage [59]. Later in DP cells, Erk activity is critical again in selection signals that differentiate positive and negative selection. Elevated MEK-

Erk stimulation results in negative selection, or death, whereas intermediate

MEK-Erk allows for positive selection, or survival, of DP cells in transition to SP

[59, 60]. Moreover, Erk activation favors the CD4+ SP lineage over CD8+ during development [61, 62].

2 Transcription Factors in T cell development

T cell development is strictly guided by site-specific transcription factors

(SSTF), DNA binding proteins or DNA-binding complex-associated proteins that regulate cell function or cell fate in response to cellular signaling by altering gene

12 transcription programs [63]. Mutations and abbreviations in the expression of several SSTFs are associated with leukemogenesis [64-73] linked to abnormal T- cell development.

2.1 BCL11B

The transcription factor B cell leukemia/lymphoma 11B (BCL11B) is a

C2H2-zinc finger SSTF that was originally named COUP-TF interacting protein 2

(CTIP2) [74], and plays a role in the development of several organs and tissues, including but not limited to, the immune [75-79], integumentary [80-86], and central nervous systems. [87-90]. In the immune system, BCL11B is exclusively expressed in T cells and progenitor thymocytes and plays a crucial role in maintaining proper development and cellular identity [77]. The Bcl11b gene consists of 4 exons and encodes two alternatively spliced variants in thymocytes; the long (1,2,3,4) variant is compiled from all four exons while the short (1,2,4) variant is lacking exon 3 [91]. As a haploinsufficient gene, Bcl11b is mutated or disrupted in many cases of T cell malignancy including 9-14% of human T-ALL patient samples [65, 70, 92], suggesting that BCL11B functions as a tumor suppressor in developing thymocytes or T cells [69].

Dysregulation of BCL11B has been linked to T-cell malignancies [65, 93].

High expression of BCL11B was reported to be associated with acute myeloid leukemia [94], and is considered an indicator of poor prognosis in adult T-ALL patients [92]. Moreover, BCL11B expression was higher in some T-ALL cell lines which are free from human T-cell leukemia-lymphoma virus such Jurkat, PEER,

13 and MOLT-4 cell lines [95]. On the other hand, BCL11B levels were reported to be reduced in other T-ALL patient derived cell lines which are positive for human

T-cell leukemia-lymphoma virus such as MT1, OMT, and KKI cells [95]. Ectopic expression of BCL11B resulted in significant inhibition of growth of some T-ALL cell lines such as MOLT-4, but not for others such as Jurkat cells [95]. Therefore, precise regulation of proper BCL11B levels appears to be important to protect thymocytes from malignant transformation.

BCL11B plays essential roles in T-cell maturation and maintenance of their identity [77] during at least three major checkpoints in thymocyte development: T cell commitment ,which is development of DN2 to DN3 thymocytes [96-98], β-selection [99], and positive selection ([76, 100] and see

Fig. 1.1) . BCL11B is also required for thymocyte survival [99, 100].

2.2 TCF1

T-cell factor 1 (TCF1, TCF7 gene product), is a HMG-box SSTF [101, 102] whose expression rises prior to that of BCL11B in the sequence of thymocyte maturation [103]. TCF1 is enriched at the promoters of key T-cell transcription factor genes including Bcl11b [104]. In thymocytes, TCF1 is expressed as at least 8 different splice variant isoforms [105]. All TCF1 isoforms contain an HMG- box DNA binding domain, an adjacent nuclear localization signal, interaction sites for Groucho/TLE corepressor proteins [106, 107], and a recently discovered histone deacetylase (HDAC) domain [108]. TCF1 isoforms can be broadly divided into two groups: the full-length (long) TCF1 isoforms have an N-terminal

β-catenin binding domain and thus form an active transcription complex upon interaction with β-catenin. The short TCF1 isoforms are N-terminally truncated, lack the β-catenin binding domain, and appear to repress Wnt target genes by competing with long TCF1 at DNA binding sites [72, 105, 109]. However, TCF1 short isoforms appear to do more than serve as antagonists to long TCF1, and were recently shown to support normal expression of a majority of TCF1 target genes and maturation of DN3 thymocytes independent of long TCF1 isoforms

[110].

TCF1 is considered the nuclear effector of the canonical Wnt pathway [72,

111]. Wnt signaling promotes thymocyte proliferation, differentiation, and survival

[111-113]. Abnormal activation of the Wnt signaling pathway was reported as an oncogenic event in many cases of childhood T-ALL [12]. Further, a deficiency of short TCF1 that predisposes to T-ALL development indicates its role as a tumor suppressor [72, 73, 114]. In addition to its role in T cell lineage determination and

DP cell survival [104, 115, 116], TCF1 regulates thymic T cell development at various stages including specification (transition of DN1 into DN2), commitment

(transition of DN2 into DN3; [104], and transition of ISP to DP cells [117]. TCF1 also participates in negative selection [118].

In the absence of Wnt activation, GSK3 phosphorylates β-catenin, the key mediator of this pathway, which results in proteasomal degradation of β-catenin.

Activation of Wnt signaling results in inhibition of GSK3, resulting in accumulation of dephosphorylated β-catenin and its translocation into the cell nucleus to

15 activate TCF and LEF, a partially redundant homolog of TCF1; [119] which then activate Wnt responsive genes [120]. However, TCF1 has also β-catenin- independent functions [113, 121]. For example, β-catenin-independent TCF1 association with the ATF2 transcription factor promotes the growth of hematopoietic malignant cells including Jurkat cells (a leukemic T-cell line) and human lymphoma cell lines [122].

3 Posttranslational modifications regulate the activity of SSTFs

During thymocyte development, regulation of gene expression by SSTFs occurs in a combinatorial fashion, in that coordinated interaction of multiple SSTFs is usually required for temporally and quantitatively refined gene transcription [123].

In their dynamic role, SSTFs are the nuclear effectors and critical targets of multiple signaling pathways [124, 125]. Signaling pathways modulate the activity of SSTFs most often through attachment of different combinations of frequently reversible posttranslational modifications (PTMs) [126, 127]. PTMs work to create a molecular barcode that can differentiate among and integrate incoming signals [128], thus regulating communication between extracellular physiological cues and cellular responses [128]. PTMs are biochemical modifications that consist of covalent addition of small molecule functional groups (such as phosphorylation and glycosylation), addition of polypeptide chains (such as sumoylation and ubiquitination), alteration of the chemical nature of the amino acids in a protein (such as acetylation and deamination), or cleavage by proteolysis resulting in covalent modification of the protein sequence after it has

16 been translated [129-132]. Almost 90,000 PTMs have been detected by biochemical and biophysical assays [133, 134], and about 5% of the human genome encodes enzymes that catalyze the chemical reactions to produce different PTMs [133, 135].

Given the vast and diverse functions of SSTFs in terms of highly regulated and defined activities at individual genes, and the fact that the number of genes encoding for SSTFs is much lower than expected or required in higher eukaryotes, PTMs provide complexity and diversity to SSTFs beyond what is made possible by gene transcripts [136-138]. PTMs of proteins expand the classical concept of the central dogma in molecular biology, which was established in a bacterial system where one gene gives rise to one protein with one function [136]. Therefore, unravelling the function of multiple PTMs and how they both subtly and overtly modulate the activity of SSTFs will help us better understand the mechanisms by which cellular characteristics and processes are orchestrated by a relatively limited number of SSTFs.

PTMs regulate the activity of SSTFs in many ways including, but not limited to, regulation of subcellular localization by increasing or decreasing the nuclear occupancy of SSTFs and access to DNA, controlling the interaction of

SSTFs with DNA, regulating the interactions among different combinations of

SSTFs, precisely controlling the molecular structure of SSTFs, regulating their stability, and altering transcriptional regulatory activity in epigenetic regulation of gene expression [126, 130, 137, 139].

PTMs are central to all aspects of signaling pathways linked to SSTFs.

Abbreviations of PTMs have been linked to many diseases such cancer [138], and can also be utilized as biomarkers of disease states [126]. Given the importance of decoration of SSTFs by PTMs in regulation of various cellular functions, modulations of these PTMs are emerging as drug targets [126]. For example, imatinib, which can be used in treatment of cancer and rheumatoid arthritis, acts by inhibiting phosphorylation of many SSTFs either directly, such as

STAT, or indirectly, as with c-JUN, RB1, TP73, and YAP1 [126, 140]. However, drugs that are designed and directed specifically to alter the PTMs of SSTFs are still rare [126]. More exploration of the modulation of SSTFs by PTMs is likely to yield insights that will allow for the development of more effective therapeutic interventions into diseases such as cancer.

3.1 Phosphorylation

Phosphorylation appears to be the most common PTM by which proteins are regulated [141-143], and consists of adding a γ phosphate group through an ester bond to a polar group of amino acid residues. Phosphorylation results in the creation of a negative charge and increased hydrophilicity at the modification site that frequently alters protein conformation and intermolecular interactions [126,

144].

Phosphorylation is catalyzed by protein kinases which mediate the transfer of a phosphate from ATP (or GTP) to specific amino acids in the substrate [144, 145]. Dephosphorylation is catalyzed by protein phosphatases

18 which mediate transfer of a phosphate from phosphorylated substrate to a water molecule [144-146]. The most commonly phosphorylated amino acids sites in a protein are serine (86.4%), followed by threonine (11.8%), tyrosine (1.8%), and rarely, phosphorylation of histidine and aspartate where the modification is less stable [144, 147, 148].

Many SSTFs have multiple phosphorylation sites which may act either cooperatively or antagonistically in response to different signaling pathways [126,

141]. Moreover, multi-phosphorylation sites can diversify the functional outputs which result from signaling through a specific kinase [149]. For example, ELK-1 is phosphorylated by Erk2 at eight sites located on its transcriptional activation domain. The phosphorylation pattern of these sites occurs at different rates-- slow, intermediate, and fast. Phosphorylation of fast and intermediate sites leads to recruitment of transcriptional activators to an ELK-1 complex and gene transcription. However, slow phosphorylation of other sites inhibits the transcriptional activity of the ELK-1 and provides a time-delimited check on ELK-

1 activation in response to Erk2 independent of phosphatase activities [149].

The processes of phosphorylation and dephosphorylation are complex and interconnected. A single kinase or phosphatase may simultaneously modify many substrates and may be involved in multiple distinct signaling pathways

[144]. Some phosphatases are activated by kinases in either downstream cascades mixing consecutive phosphatase and kinase steps or in negative feedback loops. For example, protein kinase A phosphorylates B56δ, the

19 regulatory subunit of protein phosphatase 2A, resulting in its activation [150].

The balance between protein phosphorylation and dephosphorylation is an essential cellular regulatory mechanism that is important for coordination of

SSTF activities in cellular process such as proliferation, differentiation, and apoptosis [126, 141, 144].

Defects in phosphorylation-dependent regulatory mechanisms are linked to many diseases including cancer [144, 151-153]. Cancer is not only a disease of genetic mutation, but also a disease of epigenetic changes such as abbreviation of signaling transduction pathways [154, 155]. For example, in chronic myeloid leukemia when an abnormal chromosomal translocation forms an aberrantly spliced chromosome called the Philadelphia chromosome, a new gene is created that encodes for a chimeric tyrosine kinase protein called BCL-

ABL. BCL-ABL, as a result of fusion of the ABL kinase with BCL, is constitutively active and stimulates uncontrolled proliferation of leukemia cells [156] in chronic myelogenous leukemia. Kinases inhibitors are used clinically to treat certain cancers. For example, Imatinib (Glivec ®) is an inhibitor of BCL-ABL kinase and is used effectively to treat chronic myeloid leukemia [157, 158].

3.2 Sumoylation

Sumoylation is a reversible, covalent attachment of a small ubiquitin-like modifier

(SUMO1-5) peptide to the ε-amino group in an acceptor lysine in a target protein

[126, 159-161]. Many sumoylated lysine residues are found within the short consensus sequence ψKx(D/E), where ψ is a large aliphatic hydrophobic amino

20 acid and X can be any amino acid [126, 162]. Sumo peptides are approximately

100 amino acids with a molecular mass of approximately 12 KDa [126, 161, 163].

Interestingly, mono-sumoylation of a protein at one site or poly-sumoylation at multiple sites can double the mass and surface area of the modified protein

[126].

Attachment of sumo moieties to a target protein occurs through many specific activating and ligating enzymes [161], starting with sumo-activating enzyme E1 which activates the sumo C terminus by an ATP dependent mechanism. Activated sumo is then transferred to a conjugating enzyme called

E2 (Ubc9), before the sumo moieties are transferred to a target protein by one of multiple sumo ligase enzymes (E3S) [162]. Poly-sumoylation of a protein may promote subsequent ubiquitinylation via a specific E3S enzyme called a sumoylated ubiquitin ligase (STUbL) [126]. Sumo-specific proteases (SENP) are responsible for maintenance of equilibrium of sumoylated/ non-sumoylated proteins by hydrolysis of sumo peptide from a sumo modified protein [164, 165].

Sumoylation is important for many cellular processes. Generally, sumoylation of SSTFs tends to enhance transcriptional repression [126, 166].

However, sumoylation can activate transcription through de-repression and attraction of transcriptional activators as seen with the SSTFs Ikaros [167],

BCL11B [168] and P53 [169]. Sumoylation has been shown in various contexts to regulate the activity, stability, subcellular localization and interactions of many

SSTFs [161, 170]. Sumoylation has also been found to regulate mitochondrial

21 activity and programmed cell death [161]. Deregulation of sumoylation have been linked to many diseases such as cancer where an increase in sumoylation is linked to carcinogenesis and metastatic processes [160, 171]. For example, increased sumoylation of reptin maintains downregulation of KAI1, a metastasis repressor gene, which enhances the invasive activity of cancer cells [172].

Interestingly, endogenous reptin is highly sumoylated in metastasized prostate cancer distinct from normal prostate cells [173]. In another example, inhibition of sumoylation suppresses the ability of the aberrant oncogenic protein Myc to promote tumorgenicity of human breast cancer cell lines [171]. Given the importance of sumoylation in regulation of multiple cellular functions and its implication in diseases, sumoylation can be considered a promising novel therapeutic target.

3.3 Ubiquitination

Ubiquitination is a cascade of enzymatic reactions involving addition of ubiquitin polypeptides to specific lysine residues in a target protein [174].

Ubiquitin is structurally related to sumo peptides [174, 175], and it is attached to a target protein by a three enzyme cascade including E1 ubiquitin-activating enzyme [176], E2 ubiquitin-conjugating enzyme [177] and distinct E3 ubiquitin ligases [178-181]. Ubiquitinylation is removed from substrates by deubiquitylases

(DUBs) [182, 183]. Ubiquitinylation may occur in different versions with either one ubiquitin molecule attached to the ε-amino group of a lysine residue to produce mono-ubiquitinylated protein, or on multiple lysine residues forming multi-

22 monoubiquitylation protein [184]. Moreover, since the ubiquitin peptide includes seven lysine residues in itself, orthogonal ubiquitin chains can form in addition to serial ubiquitinalytion through aminal terminus attachment forming polymeric ubiquitinylated chains [184]. Interestingly, ubiquitin is also a substrate for further

PTMs creating modifications such as phospho-ubiquitin and acetylated ubiquitin

[184, 185], and adopting an abundance of distinct conformations that mediate different signals with diverse biological outcomes called ubiquitin code [181, 185,

186].

Ubiquitinylation regulates many cellular processes, the best understood of which is protein degradation mediated by the 26S proteasome [139, 187].

Ubiquitinylation also regulates transcriptional activity either by direct regulation of

SSTFs in a proteosome-independent manner [188] or indirectly through proteosomal-dependent degradation of SSTFs and other corepressors or coactivators [189-191]. Ubiquitinylation also regulates intracellular transport, protein interactions and localization of many cellular protein such as SSTFs [174,

192]. Abnormal regulation of ubiquitinylation enzymes is linked to many diseases such as cancer and neurodegenerative diseases, and is also associated with developmental disorders [193]. For example, thalidomide, which is a compound that was once prescribed to manage the symptoms of morning sickness in pregnant women, is associated with severe developmental disorders in newborns such as short upper limbs [194, 195]. Thalidomide is suggested to initiate its teratogenic effect through binding to cereblon, an adaptor protein which is

23 required for E3 ubiquitin ligase to form a complex with damaged DNA binding protein 1 (DDB1) and Cul4A. This complex is responsible for selecting molecules for degradation and it is required for proper limb outgrowth. Therefore, thalidomide inhibition of ubiquitin ligase activity leads to abnormal regulation of ubiquitinylation and abnormal signaling in development [194, 195].

A wide range of cellular functions are regulated by ubiquitinylation, which encourages many researchers to investigate more and obtain a better understanding of the ubiquitinylation biochemistry to offer unique opportunities for developing new therapeutic approaches.

4 Research Objectives

In these studies, we first identified a previously unknown, dynamically regulated interaction between BCL11B and TCF1 in primary thymocytes (mostly

DP thymocytes). Also, we identified BCL11B as a novel target of the Wnt/GSK3- mediated signaling pathway, as indicated by dramatic changes in BCL11B PTMs after stimulation of the canonical Wnt pathway in primary thymocytes using LiCl, as detailed in chapter 2. In chapter 3, we characterized for the first time the alterations of BCL11B PTMs in a specific subset of DN thymocytes (DN3 cells) upon stimulation of pre-TCR/MAPK and Wnt/GSK3 mediated signaling pathways.

Based on our data in primary thymocytes and DN3 cells, we proposed that

BCL11B is a nuclear convergence point for MAPK and Wnt/GSK3 mediated signaling pathways in developing thymocytes.

5 References

1. Terwilliger, T. and M. Abdul-Hay, Acute lymphoblastic leukemia: a comprehensive review and 2017 update. Blood cancer journal, 2017. 7(6): p. e577. 2. Faderl, S., et al., Adult acute lymphoblastic leukemia. Cancer, 2010. 116(5): p. 1165- 1176. 3. Cooper, S.L. and P.A. Brown, Treatment of pediatric acute lymphoblastic leukemia. Pediatric Clinics, 2015. 62(1): p. 61-73. 4. Pui, C.-H., M.V. Relling, and J.R. Downing, Acute lymphoblastic leukemia. New England Journal of Medicine, 2004. 350(15): p. 1535-1548. 5. Levelt, C.N., et al., Restoration of early thymocyte differentiation in T-cell receptor beta- chain-deficient mutant mice by transmembrane signaling through CD3 epsilon. Proceedings of the National Academy of Sciences, 1993. 90(23): p. 11401-11405. 6. Coiffier, B., et al., Peripheral T-cell lymphomas have a worse prognosis than B-cell lymphomas: a prospective study of 361 immunophenotyped patients treated with the LNH-84 regimen. Annals of Oncology, 1990. 1(1): p. 45-50. 7. Tsukaguchi, M., THE EFFECT OF LONG‐TERM MAINTENANCE CHEMOTHERAPY FOR AGGRESSIVE ADULT T CELL LEUKEMIA/LYMPHOMA. Hematological Oncology, 2017. 35(S2): p. 394-395. 8. Aifantis, I., et al., Regulation of T‐cell progenitor survival and cell‐cycle entry by the pre‐ T‐cell receptor. Immunological reviews, 2006. 209(1): p. 159-169. 9. Pui, C.-H., et al., Extended follow-up of long-term survivors of childhood acute lymphoblastic leukemia. New England Journal of Medicine, 2003. 349(7): p. 640-649. 10. Mertens, A.C., et al., Late mortality experience in five-year survivors of childhood and adolescent cancer: the Childhood Cancer Survivor Study. Journal of Clinical Oncology, 2001. 19(13): p. 3163-3172. 11. Ness, K.K., et al., Adverse effects of treatment in childhood acute lymphoblastic leukemia: general overview and implications for long-term cardiac health. Expert review of hematology, 2011. 4(2): p. 185-197. 12. Ng, O., et al., Deregulated WNT signaling in childhood T-cell acute lymphoblastic leukemia. Blood cancer journal, 2014. 4(3): p. e192. 13. Winandy, S., et al., Pre–T cell receptor (TCR) and TCR-controlled checkpoints in T cell differentiation are set by Ikaros. Journal of Experimental Medicine, 1999. 190(8): p. 1039-1048. 14. Anderson, G., P.J. Lane, and E.J. Jenkinson, Generating intrathymic microenvironments to establish T-cell tolerance. Nature Reviews Immunology, 2007. 7(12): p. 954-963. 15. Takahama, Y., Journey through the thymus: stromal guides for T-cell development and selection. Nature Reviews Immunology, 2006. 6(2): p. 127-135. 16. Kruisbeek, A.M., et al., Branching out to gain control: how the pre-TCR is linked to multiple functions. Immunology today, 2000. 21(12): p. 637-644. 17. Shortman, K. and L. Wu, Early T lymphocyte progenitors. Annual review of immunology, 1996. 14(1): p. 29-47. 18. Godfrey, D.I., et al., A developmental pathway involving four phenotypically and functionally distinct subsets of CD3-CD4-CD8-triple-negative adult mouse thymocytes

defined by CD44 and CD25 expression. The Journal of Immunology, 1993. 150(10): p. 4244-4252. 19. Zúñiga-Pflücker, J.C. and M.J. Lenardo, Regulation of thymocyte development from immature progenitors. Current opinion in immunology, 1996. 8(2): p. 215-224. 20. Staal, F. and H. Clevers, Tcf/Lef transcription factors during T-cell development: unique and overlapping functions. The hematology journal, 2000. 1(1): p. 3-6. 21. Yui, M.A. and E.V. Rothenberg, Developmental gene networks: a triathlon on the course to T cell identity. Nature Reviews Immunology, 2014. 14(8): p. 529-545. 22. Dudley, E.C., et al., T cell receptor β chain gene rearrangement and selection during thymocyte development in adult mice. Immunity, 1994. 1(2): p. 83-93. 23. Michie, A.M. and J.C. Zúñiga-Pflücker. Regulation of thymocyte differentiation: pre-TCR signals and β-selection. in Seminars in immunology. 2002. Elsevier. 24. Rothenberg, E.V. and T. Taghon, Molecular genetics of T cell development. Annu. Rev. Immunol., 2005. 23: p. 601-649. 25. Wiest, D.L. and M. Carleton. Control of early thymocyte development by the pre-T cell receptor complex: A receptor without a ligand? in Seminars in immunology. 1999. Elsevier. 26. Parham, P., The immune system. 2014: Garland Science. 27. Birnbaum, M.E., et al., Molecular architecture of the αβ T cell receptor–CD3 complex. Proceedings of the National Academy of Sciences, 2014. 111(49): p. 17576-17581. 28. Yamasaki, S. and T. Saito, Molecular basis for pre-TCR-mediated autonomous signaling. Trends in immunology, 2007. 28(1): p. 39-43. 29. Groettrup, M., et al., T cell receptor (TCR) beta chain homodimers on the surface of immature but not mature alpha, gamma, delta chain deficient T cell lines. The EMBO journal, 1992. 11(7): p. 2735. 30. Yassai, M. and J. Gorski, Thymocyte maturation: selection for in-frame TCR α-chain rearrangement is followed by selection for shorter TCR β-chain complementarity- determining region 3. The Journal of Immunology, 2000. 165(7): p. 3706-3712. 31. Saint-Ruf, C., et al., Analysis and expression of a cloned pre-T cell receptor gene. Science, 1994: p. 1208-1212. 32. Aifantis, I., et al., A critical role for the cytoplasmic tail of pTα in T lymphocyte development. Nature immunology, 2002. 3(5): p. 483. 33. Borowski, C., et al., Pre-TCRα and TCRα are not interchangeable partners of TCRβ during T lymphocyte development. Journal of Experimental Medicine, 2004. 199(5): p. 607-615. 34. Borst, J., H. Jacobs, and G. Brouns, Composition and function of T-cell receptor and B-cell receptor complexes on precursor lymphocytes. Current opinion in immunology, 1996. 8(2): p. 181-190. 35. Van Oers, N., H. Von Boehmer, and A. Weiss, The pre-T cell receptor (TCR) complex is functionally coupled to the TCR-zeta subunit. Journal of Experimental Medicine, 1995. 182(5): p. 1585-1590. 36. Wiest, D.L., et al., Developmentally regulated expression of CD3 components independent of clonotypic T cell antigen receptor complexes on immature thymocytes. Journal of Experimental Medicine, 1994. 180(4): p. 1375-1382. 37. Jacobs, H., et al., CD3 components at the surface of pro‐T cells can mediate pre‐T cell development in vivo. European journal of immunology, 1994. 24(4): p. 934-939.

38. Shinkai, Y. and F.W. Alt, CD3ε-mediated signals rescue the development of CD4+ CD8+ thymocytes in RAG-2−/− mice in the absence of TCR β chain expression. International immunology, 1994. 6(7): p. 995-1001. 39. Levelt, C.N., A. Ehrfeld, and K. Eichmann, Regulation of thymocyte development through CD3. I. Timepoint of ligation of CD3 epsilon determines clonal deletion or induction of developmental program. Journal of Experimental Medicine, 1993. 177(3): p. 707-716. 40. Smith, C.A., et al., Antibodies to CD3/T-cell receptor complex induce death by apoptosis in immature T cells in thymic cultures. Nature, 1989. 337(6203): p. 181-184. 41. Altman, A., N. Isakov, and G. Baier, Protein kinase Cθ: a new essential superstar on the T- cell stage. Immunology today, 2000. 21(11): p. 567-573. 42. Isakov, N. and A. Altman, Protein kinase Cθ in T cell activation. Annual review of immunology, 2002. 20(1): p. 761-794. 43. Sun, Z., et al., PKC-θ is required for TCR-induced NF-κB activation in mature but not immature T lymphocytes. Nature, 2000. 404(6776): p. 402-407. 44. Grusby, M.J. and L.H. Glimcher, Immune responses in MHC class II-deficient mice. Annual review of immunology, 1995. 13(1): p. 417-435. 45. Koller, B.H., et al., Normal Development of Mice Deficient in (Beta)(2) M, MHC Class I Proteins, and CD8 (+) T Cells. Science, 1990. 248(4960): p. 1227. 46. Molina, T., et al., Profound block in thymocyte development in mice lacking p56lck. Nature, 1992. 357(6374): p. 161-164. 47. Groves, T., et al., Fyn can partially substitute for Lck in T lymphocyte development. Immunity, 1996. 5(5): p. 417-428. 48. van Oers, N.S., et al., αβ T cell development is abolished in mice lacking both Lck and Fyn protein tyrosine kinases. Immunity, 1996. 5(5): p. 429-436. 49. Levin, S., et al., A dominant-negative transgene defines a role for p56lck in thymopoiesis. The EMBO Journal, 1993. 12(4): p. 1671. 50. Turner, M., et al., A requirement for the Rho-family GTP exchange factor Vav in positive and negative selection of thymocytes. Immunity, 1997. 7(4): p. 451-460. 51. Williams, B.L., et al., Phosphorylation of Tyr319 in ZAP‐70 is required for T‐cell antigen receptor‐dependent phospholipase C‐γ1 and Ras activation. The EMBO journal, 1999. 18(7): p. 1832-1844. 52. Michie, A.M., et al., Extracellular Signal–regulated Kinase (Erk) Activation by the Pre-T Cell Receptor in Developing Thymocytes In Vivo. Journal of Experimental Medicine, 1999. 190(11): p. 1647-1656. 53. Aifantis, I., et al., Constitutive pre-TCR signaling promotes differentiation through Ca 2+ mobilization and activation of NF-κB and NFAT. Nature immunology, 2001. 2(5): p. 403. 54. Sommers, C.L., L.E. Samelson, and P.E. Love, LAT: a T lymphocyte adapter protein that couples the antigen receptor to downstream signaling pathways. Bioessays, 2004. 26(1): p. 61-67. 55. Ueda, Y., et al., Protein kinase C δ activates the MEK-Erk pathway in a manner independent of Ras and dependent on Raf. Journal of Biological Chemistry, 1996. 271(38): p. 23512-23519. 56. Cargnello, M. and P.P. Roux, Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiology and molecular biology reviews, 2011. 75(1): p. 50-83.

57. Zhang, W. and H.T. Liu, MAPK signal pathways in the regulation of cell proliferation in mammalian cells. Cell research, 2002. 12(1): p. 9. 58. Kaye, J., Regulation of T cell development in the thymus. Immunologic research, 2000. 21(2-3): p. 71-81. 59. Fischer, A.M., et al., The role of erk1 and erk2 in multiple stages of T cell development. Immunity, 2005. 23(4): p. 431-443. 60. Bommhardt, U., et al., MEK activity regulates negative selection of immature CD4+ CD8+ thymocytes. The Journal of Immunology, 2000. 164(5): p. 2326-2337. 61. Bommhardt, U., et al., Activation of the extracellular signal-related kinase/mitogen- activated protein kinase pathway discriminates CD4 versus CD8 lineage commitment in the thymus. The Journal of Immunology, 1999. 163(2): p. 715-722. 62. Sharp, L.L., et al., The influence of the MAPK pathway on T cell lineage commitment. Immunity, 1997. 7(5): p. 609-618. 63. López-Rodríguez, C., J. Aramburu, and R. Berga-Bolaños, Transcription factors and target genes of pre-TCR signaling. Cellular and molecular life sciences, 2015. 72(12): p. 2305- 2321. 64. Wakabayashi, Y., et al., Homozygous deletions and point mutations of the Rit1/Bcl11b gene in γ-ray induced mouse thymic lymphomas. Biochemical and biophysical research communications, 2003. 301(2): p. 598-603. 65. Przybylski, G., et al., Disruption of the BCL11B gene through inv (14)(q11. 2q32. 31) results in the expression of BCL11B-TRDC fusion transcripts and is associated with the absence of wild-type BCL11B transcripts in T-ALL. Leukemia, 2005. 19(2): p. 201-208. 66. O'neil, J. and A. Look, Mechanisms of transcription factor deregulation in lymphoid cell transformation. Oncogene, 2007. 26(47): p. 6838. 67. Kamimura, K., et al., Haploinsufficiency of Bcl11b for suppression of lymphomagenesis and thymocyte development. Biochemical and biophysical research communications, 2007. 355(2): p. 538-542. 68. Berman, J.N. and A. Thomas Look, Targeting transcription factors in acute leukemia in children. Current drug targets, 2007. 8(6): p. 727-737. 69. De Keersmaecker, K., et al., The TLX1 oncogene drives aneuploidy in T cell transformation. Nature medicine, 2010. 16(11): p. 1321-1327. 70. Gutierrez, A., et al., The BCL11B tumor suppressor is mutated across the major molecular subtypes of T-cell acute lymphoblastic leukemia. Blood, 2011. 118(15): p. 4169-4173. 71. Van Vlierberghe, P. and A. Ferrando, The molecular basis of T cell acute lymphoblastic leukemia. The Journal of clinical investigation, 2012. 122(10): p. 3398-3406. 72. Tiemessen, M.M., et al., The nuclear effector of Wnt-signaling, Tcf1, functions as a T- cell–specific tumor suppressor for development of lymphomas. PLoS biology, 2012. 10(11): p. e1001430. 73. Yu, S., et al., The TCF-1 and LEF-1 transcription factors have cooperative and opposing roles in T cell development and malignancy. Immunity, 2012. 37(5): p. 813-826. 74. Avram, D., et al., Isolation of a novel family of C2H2 zinc finger proteins implicated in transcriptional repression mediated by chicken ovalbumin upstream promoter transcription factor (COUP-TF) orphan nuclear receptors. Journal of Biological Chemistry, 2000. 275(14): p. 10315-10322.

75. Inoue, J., et al., Expression of TCRαβ partly rescues developmental arrest and apoptosis of αβ T cells in Bcl11b−/− mice. The Journal of Immunology, 2006. 176(10): p. 5871- 5879. 76. Kastner, P., et al., Bcl11b represses a mature T‐cell gene expression program in immature CD4+ CD8+ thymocytes. European journal of immunology, 2010. 40(8): p. 2143-2154. 77. Liu, P., P. Li, and S. Burke, Critical roles of Bcl11b in T‐cell development and maintenance of T‐cell identity. Immunological reviews, 2010. 238(1): p. 138-149. 78. Okazuka, K., et al., p53 prevents maturation of T cell development to the immature CD4− CD8+ stage in Bcl11b−/− mice. Biochemical and biophysical research communications, 2005. 328(2): p. 545-549. 79. Wakabayashi, Y., et al., Bcl11b is required for differentiation and survival of [alpha][beta] T lymphocytes. Nature immunology, 2003. 4(6): p. 533. 80. Bhattacharya, S., et al., Ablation of Ctip2/Bcl11b in Adult Epidermis Enhances TPA/UV- Induced Proliferation and Increases Susceptibility to DMBA/TPA-Induced Epidermal Carcinogenesis. Journal of Investigative Dermatology, 2017. 81. Bhattacharya, S., et al., Transcription Factor CTIP2 maintains hair follicle stem cell pool and contributes to altered expression of LHX2 and NFATC1. Journal of Investigative Dermatology, 2015. 135(11): p. 2593-2602. 82. Bollag, W.B., COUP-TF-interacting protein-2 (Ctip2): driving the coupé of sphingolipid biosynthesis in the epidermis. Journal of Investigative Dermatology, 2013. 133(3): p. 593- 594. 83. Ganguli‐Indra, G., et al., Expression of COUP‐TF‐interacting protein 2 (CTIP2) in human atopic dermatitis and allergic contact dermatitis skin. Experimental dermatology, 2009. 18(11): p. 994-996. 84. Golonzhka, O., et al., Dual role of COUP-TF-interacting protein 2 in epidermal homeostasis and permeability barrier formation. Journal of Investigative Dermatology, 2009. 129(6): p. 1459-1470. 85. Wang, Z., et al., Transcription factor Ctip2 controls epidermal lipid metabolism and regulates expression of genes involved in sphingolipid biosynthesis during skin development. Journal of Investigative Dermatology, 2013. 133(3): p. 668-676. 86. Wang, Z., et al., Selective ablation of Ctip2/Bcl11b in epidermal keratinocytes triggers atopic dermatitis-like skin inflammatory responses in adult mice. PloS one, 2012. 7(12): p. e51262. 87. Arlotta, P., et al., Neuronal subtype-specific genes that control corticospinal motor neuron development in vivo. Neuron, 2005. 45(2): p. 207-221. 88. Arlotta, P., et al., Ctip2 controls the differentiation of medium spiny neurons and the establishment of the cellular architecture of the striatum. Journal of Neuroscience, 2008. 28(3): p. 622-632. 89. Simon, R., et al., Structure‐function integrity of the adult hippocampus depends on the transcription factor Bcl11b/Ctip2. Genes, Brain and Behavior, 2016. 15(4): p. 405-419. 90. Simon, R., et al., A dual function of Bcl11b/Ctip2 in hippocampal neurogenesis. The EMBO journal, 2012. 31(13): p. 2922-2936. 91. Huang, X., X. Du, and Y. Li, The role of BCL11B in hematological malignancy. Experimental hematology & oncology, 2012. 1(1): p. 22.

92. Bartram, I., et al., Low expression of T-cell transcription factor BCL11b predicts inferior survival in adult standard risk T-cell acute lymphoblastic leukemia patients. Journal of hematology & oncology, 2014. 7(1): p. 51. 93. Huang, X., et al., Gene expression profiles in BCL11B-siRNA treated malignant T cells. Journal of hematology & oncology, 2011. 4(1): p. 23. 94. Abbas, S., et al., Integrated genome-wide genotyping and gene expression profiling reveals BCL11B as a putative oncogene in acute myeloid leukemia with 14q32 aberrations. Haematologica, 2014. 99(5): p. 848-857. 95. Kurosawa, N., et al., Reduced level of the BCL11B protein is associated with adult T-cell leukemia/lymphoma. PLoS One, 2013. 8(1): p. e55147. 96. Li, P., et al., Reprogramming of T cells to natural killer–like cells upon Bcl11b deletion. Science, 2010. 329(5987): p. 85-89. 97. Li, L., M. Leid, and E.V. Rothenberg, An early T cell lineage commitment checkpoint dependent on the transcription factor Bcl11b. Science, 2010. 329(5987): p. 89-93. 98. Ikawa, T., et al., An essential developmental checkpoint for production of the T cell lineage. Science, 2010. 329(5987): p. 93-96. 99. Wakabayashi, Y., et al., Bcl11b is required for differentiation and survival of αβ T lymphocytes. Nature immunology, 2003. 4(6): p. 533-539. 100. Albu, D.I., et al., BCL11B is required for positive selection and survival of double-positive thymocytes. Journal of Experimental Medicine, 2007. 204(12): p. 3003-3015. 101. Atcha, F.A., et al., A unique DNA binding domain converts T-cell factors into strong Wnt effectors. Molecular and cellular biology, 2007. 27(23): p. 8352-8363. 102. van Beest, M., et al., Sequence-specific high mobility group box factors recognize 10–12- base pair minor groove motifs. Journal of Biological Chemistry, 2000. 275(35): p. 27266- 27273. 103. Li, L., et al., A far downstream enhancer for murine Bcl11b controls its T-cell specific expression. Blood, 2013. 122(6): p. 902-911. 104. Weber, B.N., et al., A critical role for TCF-1 in T-lineage specification and differentiation. Nature, 2011. 476(7358): p. 63. 105. Van de Wetering, M., et al., Extensive alternative splicing and dual promoter usage generate Tcf-1 protein isoforms with differential transcription control properties. Molecular and cellular biology, 1996. 16(3): p. 745-752. 106. Arce, L., N. Yokoyama, and M. Waterman, Diversity of LEF/TCF action in development and disease. Oncogene, 2006. 25(57): p. 7492. 107. Hoppler, S. and C.L. Kavanagh, Wnt signalling: variety at the core. Journal of cell science, 2007. 120(3): p. 385-393. 108. Xing, S., et al., Tcf1 and Lef1 transcription factors establish CD8+ T cell identity through intrinsic HDAC activity. Nature immunology, 2016. 17(6): p. 695. 109. Eastman, Q. and R. Grosschedl, Regulation of LEF-1/TCF transcription factors by Wnt and other signals. Current opinion in cell biology, 1999. 11(2): p. 233-240. 110. Xu, Z., et al., Cutting Edge: β-Catenin–Interacting Tcf1 Isoforms Are Essential for Thymocyte Survival but Dispensable for Thymic Maturation Transitions. The Journal of Immunology, 2017. 198(9): p. 3404-3409. 111. Staal, F.J., T.C. Luis, and M.M. Tiemessen, WNT signalling in the immune system: WNT is spreading its wings. Nature reviews. Immunology, 2008. 8(8): p. 581.

112. Staal, F.J. and H.C. Clevers, WNT signalling and haematopoiesis: a WNT–WNT situation. Nature Reviews Immunology, 2005. 5(1): p. 21-30. 113. Xue, H.H. and D.M. Zhao, Regulation of mature T cell responses by the Wnt signaling pathway. Annals of the New York Academy of Sciences, 2012. 1247(1): p. 16-33. 114. Staal, F.J., et al., Aberrant Wnt signaling in leukemia. Cancers, 2016. 8(9): p. 78. 115. Ma, J., et al., β-Catenin/TCF-1 pathway in T cell development and differentiation. Journal of neuroimmune pharmacology, 2012. 7(4): p. 750-762. 116. Wang, R., et al., T cell factor 1 regulates thymocyte survival via a RORγt-dependent pathway. The Journal of Immunology, 2011. 187(11): p. 5964-5973. 117. Verbeek, S., et al., An HMG-box-containing T-cell factor required for thymocyte differentiation. Nature, 1995. 374(6517): p. 70-74. 118. Kovalovsky, D., et al., β-Catenin/Tcf determines the outcome of thymic selection in response to αβTCR signaling. The Journal of Immunology, 2009. 183(6): p. 3873-3884. 119. Okamura, R.M., et al., Redundant regulation of T cell differentiation and TCRα gene expression by the transcription factors LEF-1 and TCF-1. Immunity, 1998. 8(1): p. 11-20. 120. Klaus, A. and W. Birchmeier, Wnt signalling and its impact on development and cancer. Nature Reviews Cancer, 2008. 8(5): p. 387-398. 121. Steinke, F.C., et al., TCF-1 and LEF-1 act upstream of Th-POK to promote the CD4+ T cell fate and interact with Runx3 to silence Cd4 in CD8+ T cells. Nature immunology, 2014. 15(7): p. 646-656. 122. Grumolato, L., et al., β-Catenin-independent activation of TCF1/LEF1 in human hematopoietic tumor cells through interaction with ATF2 transcription factors. PLoS genetics, 2013. 9(8): p. e1003603. 123. Heinz, S. and C.K. Glass, Roles of lineage-determining transcription factors in establishing open chromatin: lessons from high-throughput studies, in Epigenetic Regulation of Lymphocyte Development. 2011, Springer. p. 1-15. 124. Tootle, T.L. and I. Rebay, Post‐translational modifications influence transcription factor activity: a view from the ETS superfamily. Bioessays, 2005. 27(3): p. 285-298. 125. Phillips, T., Regulation of transcription and gene expression in eukaryotes. Nature Education, 2008. 1(1): p. 199. 126. Filtz, T.M., W.K. Vogel, and M. Leid, Regulation of transcription factor activity by interconnected post-translational modifications. Trends in pharmacological sciences, 2014. 35(2): p. 76-85. 127. L Planey, S., R. Kumar, and J. A Arnott, Post-translational modification of transcription factors: mechanisms and potential therapeutic interventions. Current molecular pharmacology, 2013. 6(3): p. 173-182. 128. Benayoun, B.A. and R.A. Veitia, A post-translational modification code for transcription factors: sorting through a sea of signals. Trends in cell biology, 2009. 19(5): p. 189-197. 129. Walsh, C., Posttranslational modification of proteins: expanding nature's inventory. 2006: Roberts and Company Publishers. 130. Vuzman, D., Y. Hoffman, and Y. Levy, Modulating protein–DNA interactions by post- translational modifications at disordered regions, in Biocomputing 2012. 2012, World Scientific. p. 188-199. 131. Xie, H., et al., Functional anthology of intrinsic disorder. 3. Ligands, post-translational modifications, and diseases associated with intrinsically disordered proteins. Journal of proteome research, 2007. 6(5): p. 1917-1932.

132. Uversky, V.N., et al., Unfoldomics of human diseases: linking protein intrinsic disorder with diseases. BMC genomics, 2009. 10(1): p. S7. 133. Audagnotto, M. and M. Dal Peraro, Protein post-translational modifications: In silico prediction tools and molecular modeling. Computational and structural biotechnology journal, 2017. 15: p. 307-319. 134. Khoury, G.A., R.C. Baliban, and C.A. Floudas, Proteome-wide post-translational modification statistics: frequency analysis and curation of the swiss-prot database. Scientific reports, 2011. 1: p. 90. 135. Venne, A.S., et al., An improved workflow for quantitative N‐terminal charge‐based fractional diagonal chromatography (ChaFRADIC) to study proteolytic events in Arabidopsis thaliana. Proteomics, 2015. 15(14): p. 2458-2469. 136. Cain, J.A., N. Solis, and S.J. Cordwell, Beyond gene expression: the impact of protein post- translational modifications in bacteria. Journal of proteomics, 2014. 97: p. 265-286. 137. Duan, G. and D. Walther, The roles of post-translational modifications in the context of protein interaction networks. PLoS computational biology, 2015. 11(2): p. e1004049. 138. Wang, Y.-C., S.E. Peterson, and J.F. Loring, Protein post-translational modifications and regulation of pluripotency in human stem cells. Cell research, 2014. 24(2): p. 143. 139. Desterro, J., M. Rodriguez, and R. Hay, Regulation of transcription factors by protein degradation. Cellular and Molecular Life Sciences CMLS, 2000. 57(8-9): p. 1207-1219. 140. Colicelli, J., ABL tyrosine kinases: evolution of function, regulation, and specificity. Sci. Signal., 2010. 3(139): p. re6-re6. 141. Whitmarsh, A. and R. Davis, Regulation of transcription factor function by phosphorylation. Cellular and Molecular Life Sciences CMLS, 2000. 57(8-9): p. 1172- 1183. 142. Li, X., et al., Elucidating human phosphatase-substrate networks. Sci. Signal., 2013. 6(275): p. rs10-rs10. 143. Sacco, F., et al., The human phosphatase interactome: An intricate family portrait. FEBS letters, 2012. 586(17): p. 2732-2739. 144. Ardito, F., et al., The crucial role of protein phosphorylation in cell signaling and its use as targeted therapy. International journal of molecular medicine, 2017. 40(2): p. 271-280. 145. Cheng, H.-C., et al., Regulation and function of protein kinases and phosphatases. Enzyme research, 2011. 2011. 146. Barford, D., Molecular mechanisms of theprotein serine/threonine phosphatases. Trends in biochemical sciences, 1996. 21(11): p. 407-412. 147. Schwartz, P.A. and B.W. Murray, Protein kinase biochemistry and drug discovery. Bioorganic chemistry, 2011. 39(5-6): p. 192-210. 148. Nishi, H., A. Shaytan, and A.R. Panchenko, Physicochemical mechanisms of protein regulation by phosphorylation. Frontiers in genetics, 2014. 5: p. 270. 149. Mylona, A., et al., Opposing effects of Elk-1 multisite phosphorylation shape its response to Erk activation. Science, 2016. 354(6309): p. 233-237. 150. Ahn, J.-H., et al., Protein kinase A activates protein phosphatase 2A by phosphorylation of the B56δ subunit. Proceedings of the National Academy of Sciences, 2007. 104(8): p. 2979-2984. 151. Hanahan, D. and R. Weinberg, Hallmarks of Cancer: The Next Generation. C ell. 2011; 144 (5): 646-74. CrossRef PubMed Google Scholar.

152. Hynes, N.E. and G. MacDonald, ErbB receptors and signaling pathways in cancer. Current opinion in cell biology, 2009. 21(2): p. 177-184. 153. Sharma, A., et al., Rare and Novel Epidermal Growth Factor Receptor Mutations in Non—Small-Cell Lung Cancer and Lack of Clinical Response to Gefitinib in Two Cases. Journal of Thoracic Oncology, 2012. 7(5): p. 941-942. 154. Hanahan, D. and R.A. Weinberg, The hallmarks of cancer. cell, 2000. 100(1): p. 57-70. 155. Vogelstein, B. and K.W. Kinzler, Cancer genes and the pathways they control. Nature medicine, 2004. 10(8): p. 789. 156. Murphree, A.L. and W.F. Benedict, Retinoblastoma: clues to human oncogenesis. Science, 1984. 223(4640): p. 1028-1033. 157. Druker, B.J., Imatinib mesylate in the treatment of chronic myeloid leukaemia. Expert opinion on pharmacotherapy, 2003. 4(6): p. 963-971. 158. Stegmeier, F., et al., Targeted cancer therapies in the twenty‐first century: lessons from imatinib. Clinical Pharmacology & Therapeutics, 2010. 87(5): p. 543-552. 159. Liang, Y.-C., et al., SUMO5, a novel poly-SUMO isoform, regulates PML nuclear bodies. Scientific reports, 2016. 6: p. 26509. 160. Bettermann, K., et al., SUMOylation in carcinogenesis. Cancer letters, 2012. 316(2): p. 113-125. 161. Andreou, A.M. and N. Tavernarakis, SUMOylation and cell signalling. Biotechnology journal, 2009. 4(12): p. 1740-1752. 162. Johnson, E.S., Protein modification by SUMO. Annual review of biochemistry, 2004. 73(1): p. 355-382. 163. Park, H.J. and D.-J. Yun, New insights into the role of the small ubiquitin-like modifier (SUMO) in plants, in International review of cell and molecular biology. 2013, Elsevier. p. 161-209. 164. Bawa-Khalfe, T. and E.T. Yeh, The in vivo functions of desumoylating enzymes, in Conjugation and Deconjugation of Ubiquitin Family Modifiers. 2010, Springer. p. 170- 183. 165. Karami, S., et al., Novel SUMO-Protease SENP7S Regulates β-catenin Signaling and Mammary Epithelial Cell Transformation. Scientific Reports, 2017. 7: p. 46477. 166. Cubeñas-Potts, C. and M.J. Matunis, SUMO: a multifaceted modifier of chromatin structure and function. Developmental cell, 2013. 24(1): p. 1-12. 167. Gómez-del Arco, P., J. Koipally, and K. Georgopoulos, Ikaros SUMOylation: switching out of repression. Molecular and cellular biology, 2005. 25(7): p. 2688-2697. 168. Zhang, L.-j., et al., Coordinated regulation of transcription factor Bcl11b activity in thymocytes by the mitogen-activated protein kinase (MAPK) pathways and protein sumoylation. Journal of Biological Chemistry, 2012. 287(32): p. 26971-26988. 169. Rodriguez, M.S., et al., SUMO‐1 modification activates the transcriptional response of p53. The EMBO journal, 1999. 18(22): p. 6455-6461. 170. Flotho, A. and F. Melchior, Sumoylation: a regulatory protein modification in health and disease. Annual review of biochemistry, 2013. 82: p. 357-385. 171. Kessler, J.D., et al., A SUMOylation-dependent transcriptional subprogram is required for Myc-driven tumorigenesis. Science, 2012. 335(6066): p. 348-353. 172. Kim, J.H., et al., Roles of sumoylation of a reptin chromatin-remodelling complex in cancer metastasis. Nature Cell Biology, 2006. 8(6): p. 631.

173. Baek, S.H., A novel link between SUMO modification and cancer metastasis. Cell Cycle, 2006. 5(14): p. 1492-1495. 174. Gill, G., SUMO and ubiquitin in the nucleus: different functions, similar mechanisms? Genes & development, 2004. 18(17): p. 2046-2059. 175. Nie, M. and M.N. Boddy, Cooperativity of the SUMO and ubiquitin pathways in genome stability. Biomolecules, 2016. 6(1): p. 14. 176. Schulman, B.A. and J.W. Harper, Ubiquitin-like protein activation by E1 enzymes: the apex for downstream signalling pathways. Nature reviews Molecular cell biology, 2009. 10(5): p. 319. 177. Ye, Y. and M. Rape, Building ubiquitin chains: E2 enzymes at work. Nature reviews Molecular cell biology, 2009. 10(11): p. 755. 178. Deshaies, R.J. and C.A. Joazeiro, RING domain E3 ubiquitin ligases. Annual review of biochemistry, 2009. 78. 179. Smit, J.J. and T.K. Sixma, RBR E3‐ligases at work. EMBO reports, 2014. 15(2): p. 142-154. 180. Rotin, D. and S. Kumar, Physiological functions of the HECT family of ubiquitin ligases. Nature reviews Molecular cell biology, 2009. 10(6): p. 398. 181. Swatek, K.N. and D. Komander, Ubiquitin modifications. Cell research, 2016. 26(4): p. 399. 182. Hershko, A. and A. Ciechanover, The ubiquitin system. 1998, Annual Reviews 4139 El Camino Way, PO Box 10139, Palo Alto, CA 94303-0139, USA. 183. Wilkinson, K.D. Ubiquitination and deubiquitination: targeting of proteins for degradation by the proteasome. in Seminars in cell & developmental biology. 2000. Elsevier. 184. Yau, R. and M. Rape, The increasing complexity of the ubiquitin code. Nature cell biology, 2016. 18(6): p. 579. 185. Ohtake, F. and H. Tsuchiya, The emerging complexity of ubiquitin architecture. The Journal of Biochemistry, 2016. 161(2): p. 125-133. 186. Komander, D. and M. Rape, The ubiquitin code. Annual review of biochemistry, 2012. 81: p. 203-229. 187. Kulathu, Y. and D. Komander, Atypical ubiquitylation—the unexplored world of polyubiquitin beyond Lys48 and Lys63 linkages. Nature reviews Molecular cell biology, 2012. 13(8): p. 508. 188. Kaiser, P., et al., Regulation of transcription by ubiquitination without proteolysis: Cdc34/SCFMet30-mediated inactivation of the transcription factor Met4. Cell, 2000. 102(3): p. 303-314. 189. Conaway, R.C., C.S. Brower, and J.W. Conaway, Emerging roles of ubiquitin in transcription regulation. Science, 2002. 296(5571): p. 1254-1258. 190. Ciechanover, A., A. Orian, and A.L. Schwartz, Ubiquitin‐mediated proteolysis: biological regulation via destruction. Bioessays, 2000. 22(5): p. 442-451. 191. Muratani, M. and W.P. Tansey, How the ubiquitin–proteasome system controls transcription. Nature reviews Molecular cell biology, 2003. 4(3): p. 192. 192. Kerscher, O., R. Felberbaum, and M. Hochstrasser, Modification of proteins by ubiquitin and ubiquitin-like proteins. Annu. Rev. Cell Dev. Biol., 2006. 22: p. 159-180. 193. Rape, M., Ubiquitylation at the crossroads of development and disease. Nature Reviews Molecular Cell Biology, 2018. 19(1): p. 59.

194. Ito, T., H. Ando, and H. Handa, Teratogenic effects of thalidomide: molecular mechanisms. Cellular and Molecular Life Sciences, 2011. 68(9): p. 1569-1579. 195. Ito, T., et al., Identification of a primary target of thalidomide teratogenicity. science, 2010. 327(5971): p. 1345-1350.

Chapter 2

Bcl11b is at the nexus of Wnt and MAP Kinase signaling pathways in thymocytes

Wisam H. Selman, Walter K. Vogel, Hua Wise, Alamjit K. Nagra, Arun J. Singh, Mark Leid and Theresa M. Filtz

2.1 Abstract

Sequence specific transcription factors (SSTFs) are nuclear proteins that integrate cellular signals into gene expression programs that underlie the phenotypic properties of all cells. Several SSTFs have been identified as “T cell lineage-determining factors” required for proper maturation of thymocytes into T

36 cells. In the multi-step sequence of thymocyte maturation, lineage-determining

SSTFs operate in a cooperative, combinatorial fashion as they respond to multiple inputs. However, the mechanisms underlying cooperativity among lineage-determining SSTFs are not all known. Among the T cell lineage- determining SSTFs, Tcf1 is known as the nuclear effector of Wnt/Gsk3/β-catenin

-dependent signaling in T cells and Bcl11b is a known nuclear target of the T-cell receptor (TCR)-mediated MAP Kinase (MAPK) signaling cascade. In other cell types, Bcl11b has been identified as regulating Wnt target genes, but a molecular link between Wnt signaling pathway proteins and Bcl11b has not been previously established. Through co-immunoprecipitation studies, we identified a previously unknown physical interaction in primary murine thymocytes between Bcl11b and

Tcf1, and among BCL11B, TCF1 and β-catenin in Jurkat cells. CRISPR/Cas9- mediated genome editing to generate two C terminally-truncated forms of Bcl11b lacking sumoylation sites revealed that neither sumoylation of BCL11B nor up to

285 C terminal amino acids in BCL11B are required for the protein to complex with Tcf1 or β-catenin. We observed that the Tcf1–Bcl11b interaction was dynamically regulated by Wnt pathway activation, which lead to decreased association between the two SSTFs.

We identified several thousand gene promoter regions that are co- occupied by the two factors suggesting that the Bcl11b -Tcf1 interaction may coordinately regulate many genes required for thymocyte maturation. Using a

37 gene reporter assay, we identified the Bcl2l1 gene, encoding the anti-apoptotic protein Bcl-xl, as co-regulated by BCL11B and TCF1.

Stimulation of Wnt/Gsk3 signaling also resulted in significant changes in the posttranslational modifications (PTMs) of Bcl11b. Our previous studies identified two kinases that are involved in modification of Bcl11b (Erk1/2 and p38

MAPK); in this study we identified a third candidate (Gsk3). The dynamically regulated interaction between Tcf1 and Bcl11b may serve as a mechanism by which these two SSTFs integrate signals and coordinate responses to coincident or successive signals in the sequential, externally coordinated process of thymocyte maturation.

2.2 Introduction:

The combinatorial activities of SSTFs during thymic T-cell development control molecular events accompanying initiation, proliferation, differentiation and survival of thymocytes in a well-defined process to produce functional, mature αβ

T cells [1]. SSTFs are nuclear proteins that respond to cellular signals to dictate cell state and fate by controlling gene expression. The products of the target genes regulated by SSTFs in thymocytes determine proper cellular maturation by balancing cell proliferation, differentiation, survival and apoptosis [2]. When the process of maturation goes awry, as in cells where lineage-determining SSTFs are mutated, thymocytes are susceptible to uncontrolled proliferation and leukemogenesis [3-12].

T lymphocytes represent a critical cellular arm of the adaptive immune responses that are crucial for targeting and eradicating a wide range of invading pathogens [13, 14]. T-lymphocyte development is a multiple-stage process occurring within the thymus; the earliest stage is the seeding of lymphoid progenitor cells from bone marrow or fetal liver into thymus to initiate differentiation [15-17]. The earliest thymocytes lack expression of both CD4 and

CD8 cell-surface markers, and are therefore known as double-negative (DN) [15,

16]. DN thymocytes pass through at least four developmental stages that are further defined by expression of CD44 and CD25 cell surface markers: DN1

(CD44+ CD25-), DN2 (CD44+ CD25+), DN3 (CD44- CD25+), and DN4 (CD44-

CD25-). The β-selection checkpoint ensures that only DN3 thymocytes which have generated a functional T cell receptor β chain develop beyond the DN3 stage [18, 19]. Later development is defined by CD4 or CD8 expression, starting with CD8+ immature single positive (ISP) cells, progressing to CD4+ CD8+ double-positive (DP) cells, and finally passing through positive and negative selection to mature CD4+ or CD8+ single-positive (SP) cells [20-22].

This highly coordinated, sequential development of thymocytes is guided by different combinations of many SSTFs [22, 23]. Among these SSTFs, Bcl11b and Tcf1 play an essential role in maintaining proper T-cell development [24].

Bcl11b, a C2H2-zinc finger SSTF that was originally named COUP-TF interacting protein 2 (CTIP2) [25], plays a role in the development of several organs and tissues, including but not limited to, the immune [26-30], integumentary [31-37],

39 and central nervous systems [38-41]. In the immune system, Bcl11b is exclusively expressed in T cells and progenitor thymocytes, where it is essential for thymocyte survival, development and cellular identity at multiple differentiation steps [27, 28, 30, 42-46] including T-cell commitment at the DN2 stage [45, 47,

48], β-selection at the DN3 stage [30], and positive selection at the DP stage [27,

42]. As a haploinsufficient tumor suppressor gene, Bcl11b, when mutated, is associated with susceptibility to radiation-induced lymphomas in mice [6, 11] and with a substantial subset of T-cell acute lymphoblastic leukemias (T-ALL) in humans [4, 5, 10]. In mouse thymocytes, Bcl11b appears to function largely as a transcriptional repressor in the context of the NuRD (Nucleosome Remodeler and

Deacetylase) complex [27, 49]. Conversely, Bcl11b can act also as a transcriptional activator in T cells [27, 50]. Bcl11b is highly modified by post- translational modifications (PTMs), and we previously identified a dephosphorylation–sumoylation switch that is coincident with alleviation of the transcriptional repressive activity of Bcl11b at the oncogenic Id2 target gene [51].

The Bcl11b gene is a direct target of T-cell factor 1 (Tcf1, Tcf7 gene product), whose expression precedes that of Bcl11b in the sequence of thymocyte maturation [52]. Tcf1 is a HMG-box SSTF [53, 54] that is enriched at the promoters of key T-cell transcription factor genes including Bcl11b [55]. In thymocytes, Tcf1 is expressed as at least 8 different splice variant isoforms [56].

All Tcf1 isoforms contain an HMG-box DNA binding domain, an adjacent nuclear localization signal, interaction sites for Groucho/TLE corepressor proteins [57,

58], and a recently discovered histone deacetylase (HDAC) domain [59]. Tcf1 isoforms can be broadly divided into two groups: the full-length (long) Tcf1 isoforms have an N-terminal β-catenin binding domain and thus form an active transcription complex upon interaction with β-catenin; the short Tcf1 isoforms are

N-terminally truncated, lack the β-catenin binding domain, and appear to repress

Wnt target genes by competing with long Tcf1 at DNA binding sites [9, 56, 60].

However, Tcf1 short isoforms appear to do more than serve as antagonists to long Tcf1, and were recently shown to support normal expression of a majority of

Tcf1 target genes that are required for differentiation of DN3 thymocytes independent of long Tcf1 isoforms [14].

Tcf1 is considered the nuclear effector of the canonical Wnt pathway [9,

61]. Wnt signaling promotes thymocyte proliferation, differentiation, and survival

[61-63]. Abnormal activation of the Wnt signaling pathway was reported as an oncogenic event in many cases of childhood T-ALL [64]. Further, a deficiency of short Tcf1 that predisposes to T-ALL development indicates its role as a tumor suppressor [9, 12, 65]. In addition to its role in T cell lineage determination and

DP cell survival [55, 66, 67], Tcf1 regulates thymic T cell development at various stages including specification (transition of DN1 into DN2), commitment

(transition of DN2 into DN3; [55]), and transition of ISP to DP cells [68]. Tcf1 also participates in negative selection [69].

Best studied in Drosophila and Xenopus model systems, Tcf1 has been shown to interact with Groucho/TLE corepressor proteins to repress Wnt target

41 genes in the absence of canonical Wnt signaling activation [70-72]. Upon stimulation of the Wnt signaling pathway, inactivation of the constitutively active kinase Gsk3 leads to de-phosphorylation of cytosolic β-catenin and transport into the nucleus. Nuclear accumulation of β-catenin displaces the Groucho/TLE complex to associate with Tcf1 and LEF (a partially redundant homolog of Tcf1;

[73]), inducing expression of Wnt target genes [61, 74, 75]. However, Tcf1 also has β-catenin-independent functions [63, 76]. For example, Tcf1, independent of

β-catenin, associates with the ATF2 transcription factor to promote the growth of hematopoietic malignant cells, including Jurkat cells (a leukemic T-cell line) and human lymphoma cell lines [77].

While the roles of Bcl11b and Tcf1 are somewhat understood individually in thymocytes, the combinatorial effect of co-expression and putative cooperative interactions of these regulators is relatively unknown. We hypothesized that in developing thymocytes, Bcl11b might participate in a complex that also contains

Tcf1, co-regulating specific genes required for step-wise maturation of thymocytes. We tested our hypothesis through co-immunoprecipitation studies of these SSTFs in native murine thymocytes, and further probed the signaling pathways that are involved in modulation of Bcl11b PTMs in thymocytes by treatment with LiCl, a widely used Gsk3 inhibitor, to mimic Wnt signaling activation.

2.3 Methods

Cell culture

Primary murine thymocytes were isolated from 4 to 5-week-old wild-type mice and cultured at 37 °C, 5% CO2 in RPMI-1640 medium plus 2.5% fetal bovine serum (FBS) (Hyclone), 1% sodium pyruvate,1% nonessential amino acids, and penicillin/streptomycin for 5 h prior to stimulation. All procedures involving mice were conducted according to the protocol reviewed and approved by the Oregon State University Institutional Animal Care and Use Committee.

Jurkat cells were cultured as described above, but with 10% FBS which was then reduced to 2.5% for 24 h prior to harvest. HEK-293T were grown in DMEM medium augmented with 10% FBS and penicillin/ streptomycin and incubated at the same conditions described above.

Chemical stimulation

To mimic stimulation of Wnt signaling pathway, thymocytes were treated with 20mM LiCl (EMD Millipore, Darmstadt, Germany) or with water as vehicle control. In some experiments, cells were treated with 50 nM calyculin A (Cell

Signaling Technology, Danvers, MA), with 100 nM phorbol 12,13-dibutyrate and

500 nM A23187 (P/A), or with 0.1% DMSO as vehicle control.

Harvesting and cell lysis

After chemical stimulation, thymocytes were diluted in ice-cold PBS, centrifuged at 2000 × g for 5 min for collection, and then lysed by one of two methods depending on the experimental aim. First, for studies of protein-protein interactions, lysis occurred under mild conditions designed to preserve native protein complexes by using “native” lysis buffer (20 mM HEPES, pH 7.4, 250 mM

NaCl, 2 mM EDTA, 10% glycerol, and 0.5% NP-40 plus the protease inhibitor cocktail 0.1 mM PMSF; 1μg/mL pepstatin A; 5 μg/mL leupeptin; 10 μM E64). NaF

(50 mM) and Na2P2O4 (5 mM) was added to all buffers to preserve phosphorylation; N-ethylmaleimide (10 mM) also was added to all buffers to preserve sumoylation. Cells were incubated on ice for 30 minutes in the presence of lysis buffer with vortexing every 5 min and then sonicated to complete lysis and shear DNA, resulting in “native” lysates. Before processing of native lysates for immunoprecipitation studies, 1.5% of samples were taken and concentrated by chloroform/methanol extraction [78] to be used as input samples on Western blots.

For studies of posttranslational modifications (PTMs), a more stringent procedure that results in near complete denaturation and loss of protein–protein interactions utilized a rapid lysis buffer (20 mM HEPES, pH 7.4; 200 mM NaCl; 2 mM EDTA; 10% glycerol; 1% SDS, with the protease inhibitor cocktail, and compounds for preservation of sumoylation and phosphorylation as described above). Upon suspension of cells in rapid lysis buffer, samples were immediately placed in a boiling hot water bath for 15 min. Denatured lysates were sonicated to shear DNA, and 1.5% of samples were taken to be used as input samples.

The remaining denatured lysates were diluted 10-fold in rapid lysis buffer lacking

SDS but containing 1.1%Triton-X100 before proceeding to immunoprecipitation

[51, 79].

Immunoprecipitation and western blotting

To immunoprecipitate Tcf1 from native lysates, we used either anti-Tcf1 monoclonal antibody C46C7 for detection of all Tcf1 isoforms, (#2206, rabbit,

Cell Signaling, 1:100 dilution) or anti-Tcf1 monoclonal antibody C63D9 selective for long Tcf1 isoforms containing the β-catenin binding domain (#2203, rabbits,

Cell Signaling, 1:100 dilution). Immunoprecipitation of Bcl11b protein from lysates was performed with either rat anti-Bcl11b antibody (clone 25B6, Abcam,

#ab18465) or goat anti-Bcl11b antibody covalently linked to agarose beads.

Immunoprecipitates were subjected to SDS-PAGE then immunoblotting with one or more of the following antibodies: anti-Tcf1 (#2206, rabbit, Cell

Signaling, 1:1,000 dilution); anti-Bcl11b (#ab18465, rat, Abcam, 0.5 ng/ml); anti-

β-Catenin (# 8480, Cell Signaling, rabbit, 1:1,000 dilution); anti-MTA2 (#A300-

395A, rabbit, Bethyl Laboratories,1:5,000); anti-HDAC2 (#A300-705A, rabbit,

Bethyl Laboratories,1:5,000); anti-phospho-threonine/serine (# 9381,Cell

Signaling, mouse, 1:1,000 dilution); anti-SUMO-1 (# ab32058, Abcam, rabbit,1:1,000 dilution); anti-SUMO-2/3 (# ab3742, Abcam, rabbit, 1:1,000 dilution); anti-GSK-3α/β (# sc-7291, Santa Cruz, mouse, 1:500 dilution); anti- phospho-GSK-3α/β(Ser21/9) (#9331, Cell Signaling, rabbit, 1:1,000 dilution); anti-Erk 1/2 (# sc-292838, Santa Cruz, rabbit, 1:1,000 dilution); anti-phospho-

Erk1 (pT202/pY204)/phospho-Erk2 (pT185/pY187) (# ab50011 , Abcam, mouse,

1:10,000 dilution).

Immunoreactive bands were detected with the following fluorescently- tagged secondary anti-IgG antibodies (LI-COR): anti-rabbit IgG, 800 nm, 1:5,000

45 dilution; anti-rat IgG, 680 nm, 1:20,000 dilution; anti-mouse IgG, 800 nm, 1:5,000 dilution; anti-mouse IgG, 600 nm, 1:20,000 dilution; anti-rabbit IgG, 600 nm,

1:20,000 dilution. The intensities of resultant fluorescent bands were quantitated using a LI-COR Odyssey instrument and ImageStudio, version 2.1, (LI-COR) software.

CRISPR/Cas9-mediated BCL11B genome editing

To generate C-terminally truncated mutants of human BCL11B that encompass one or both of the sumoylation sites originally identified in mouse

Bcl11b, we used a CRISPR/Cas9 approach. A pair of sgRNAs to target exon 4 of

BCL11B (sgRNA1 and sgRNA3; table 2.1) were designed using the MIT CRISPR

Design Tool (http://crispr.mit.edu/) to minimize off-target effects. Each sequence was cloned into pSpCas9 (BB)-2A-GFP (pX458, Addgene vector # 48138) using the BbsI restriction enzyme site [80]. For transfection of the pX458 vectors containing the sgRNA sequences, Jurkat cells were seeded in a 6-well plate at the concentration of 4x105 cells/ml. Equimolar ratios of pX458 containing sgRNA1 and sgRNA3, or empty pX458 vector as control were mixed and 500 ng total DNA was used with TransIT®-Jurkat Transfection Reagent (Mirus Bio LLC,

Madison WI) according to the manufacturer’s instructions. The transfected cells were sorted on a Sony SH800 cell sorter (Sony Biotechnology, San Jose, CA) 48 h post transfection, and the top ~3% of GFP-positive cells were collected.

CRISPR-targeted cell clones were obtained by limiting dilution into 96 well plates

[81]. Individual clones were expanded for 2 weeks into colonies, and biallelic

46 deletion clones were identified by PCR screening of genomic DNA using PCR primers EX4-3F and EX4-3R (Table2 .1). Sequencing revealed that two C terminal BCL11B truncation mutant Jurkat cell lines were created; one mutant,

ΔC1, contained a C terminal truncation of 95 aa, encompassing K887, homologous to mouse Bcl11b sumoylation site K877, and the second mutant,

ΔC2, contained a C terminal truncation of 285 aa, encompassing K689 and

K887, homologous to both sumoylation sites defined in mouse Bcl11b, K679 and

K877.

Table 2.1. Sequences of oligonucleotides used for CrispR-Cas9 mutation and sequencing. Primer name Sequence Usage

Bcl11B-gRNA-1-FW CACCGCTAGGCGCACTGCCGCAGTA Oligo annealing to create

Bcl11B-gRNA-1-RV AAACTACTGCGGCAGTGCGCCTAGC sgRNA-1

Bcl11B-gRNA-3-FW CACCGCGAGTACACGTTCTCGGAC Oligo annealing to create

Bcl11B-gRNA-3-RV AAACGTCCGAGAACGTGTACTCGC sgRNA-3

EX4-3F AGCTGCTACTGGAGAACGAGAGCC Detect and sequence

EX4-3R GTTCTCGGACGAGTGCTCGGACGA Bcl11B deletions

Bioinformatic Data Analysis

Data from chromatin immunoprecipitation followed by high-throughput sequencing (ChIP-seq) studies were obtained from the Sequence Read Archive

(SRA) accession numbers SRP022127 [82] and SRP044930 (Bcl11b) and aligned to the mouse GRCm38/mm10 reference genome assembly using BWA

(version 0.6.2, [83]). MACS (version 2.0.10, [84]) was used to identify peaks of enriched ChIP-seq reads over input controls (q-value ≤ 0.05) and BEDTools

(version 2.25, [85]) was used to calculate peak overlaps in gene promoter regions (–1000 to +500 bp of transcription start sites). Peak overlaps were visualized as area proportional Venn diagrams generated with eulerAPE (version

3, [86]).

Luciferase Reporter Gene Assays

DNA Constructs

Mammalian expression vector pcDNA3 containing cDNA for human TCF1 was purchased from Addgene. A plasmid containing the human BCL11B (1-2-3-4 splice variant) cDNA sequence was purchased from abm Inc. by Dr. Mark Leid and gifted to us. The human BCL11B sequence was PCR amplified to include

BamHI and XbaI restriction enzyme sites at the 5’ and 3’ ends, and directionally subcloned into the corresponding multiple cloning sites of the pcDNA3 mammalian expression vector. A luciferase reporter construct was created in the pGL3 vector (Promega) by inserting a 2 kb fragment of the Bcl2l1 promoter immediately upstream of the transcription start site of the luciferase gene. All constructs were verified by DNA sequencing.

Transfection

HEK-293T cells (104 cells/μl) were seeded into 96 well plates, 100 μl per well, and transfected 24 h later using jetPRIME® transfection reagents (cat #114-07).

Cells were transfected with constant amount of reporter gene construct (37.5 ng)

48 and varying concentrations of BCL11B and/or TCF-1 pcDNA3 mammalian expression plasmids and incubated for 48 hours prior to assay. The total amount of DNA in each well was kept constant at 100 ng by supplementing with empty pcDNA3 vector as needed.

On the day of assay, medium was removed and 20 μl each of 1X PBS and

OneGlo™ Luciferase buffer from Promega were added to each well and mixed thoroughly according to manufacturer's instructions. Resulting luminescence intensity for each well was quantified using a microplate reader luminometer

(Synergy® .HT BioTek, Winooski, VT)

Statistical Analysis

Data were analyzed statistically by using Student's t-test and one-way analysis of variance (ANOVA) followed by less significance difference (LSD). A p-value ≤ 0.05 was considered statistically significant.

2.4 Results

Bcl11b interacts with Tcf1 in thymocytes

To test our hypothesis that Bcl11b and Tcf1 physically interact in thymocytes, we prepared cell extracts from primary murine thymocytes under lysis conditions that preserve protein–protein interactions. These lysates were immunoprecipitated with either rabbit IgG (control) or rabbit monoclonal antibody raised against Tcf1, and size separated by SDS-PAGE. Blots were then probed with antibodies against Tcf1and Bcl11b. Bcl11b co-immunoprecipitated with Tcf1

(Fig. 2.1A). Tcf1 is composed of several isoforms that migrate between

49 approximately 36 and 72 kDa, the longest of which contain a β-catenin binding domain. Although the migration of the highest-molecular-weight isoforms of Tcf1 were close to the IgG heavy chain bands, multiple bands corresponding to Tcf1 isoforms were clearly seen.

Bcl11b migrated as a doublet of two splice variants of approximately 130-

135 KDa in addition to higher molecular weight bands (up to nearly 250 kDa), which we previously identified as SUMO1 and SUMO2/3 adducts of Bcl11b [51].

The Bcl11b-Tcf1 interaction in thymocytes was also confirmed by reciprocal immunoprecipitation (Fig. 2.1B) using antibody against Bcl11b covalently linked to agarose beads. Tcf1 was clearly seen in the Bcl11b-immunoprecipitated samples and completely lacking at the level of detection in the IgG control lane.

We quantified the relative amount of “short” Tcf1 (36 – 45 kDa bands, lower box in Fig. 2.1B) and “long” Tcf1 (45 –72 kDa bands, upper box in Fig.

2.1B) isoforms in the Bcl11b-immunoprecipitated samples, corrected for input.

Quantification of Tcf1 long and short isoforms revealed an approximately two-fold increase, 2.25 ± 0.18 (mean ± SEM), in the ratio of short Tcf1 isoforms that immunoprecipitated with Bcl11b relative to the amount of long isoforms of Tcf1, suggesting that Bcl11b somewhat preferentially, but not exclusively interacts with the short isoforms of Tcf1.

To further test the hypothesis that Bcl11b preferentially interacts with short isoforms of Tcf1, we used a commercial antibody directed against the β-catenin binding domain selective for long Tcf1 isoforms for co-immunoprecipitation of

Bcl11b and compared to immunoprecipitates with the anti-Tcf1 antibody that detects all Tcf1 isoforms for the presence of Bcl11b. Consistent with our hypothesis, there was 2.7 ± 0.5 (mean ± SEM)-fold less Bcl11b in samples immunoprecipitated with antibody selective for Tcf1 long isoforms relative to samples immunoprecipitated with antibody against total Tcf1 (Fig.2.1C).

Dynamic change in Bcl11b–Tcf1 interaction following 24 h LiCl stimulation of thymocytes

Given the interaction of Bcl11b with Tcf1, the effector of Wnt signaling, we hypothesized that the interaction between factors was regulated by Wnt pathway activation. To mimic stimulation of the canonical Wnt signaling pathway, we treated thymoycytes with 20 mM LiCl, a widely used classical Wnt pathway activator that inhibits Gsk3 to promote stabilization and accumulation of β-catenin

([87] and Fig. 2.2C).

Prior to harvest and lysis of thymocytes as described for Figure 1, we treated cells for 8 and 24 h with LiCl or vehicle. Lysates were immunoprecipitated as described above with anti-Tcf1 antibody (directed against total Tcf1) and then immunoblotted for the presence of Tcf1 and Bcl11b. Treatment of thymocytes with LiCl resulted in a reduction of co-immunoprecipitation of Bcl11b with Tcf1 in a time-dependent manner (Fig. 2.2A). Reduced co-immunoprecipitation with LiCl treatment was confirmed by reciprocal immunoprecipitation with anti-Bcl11b antibody (Fig.2.2B). LiCl treatment of T-cells previously was shown to stimulate accumulation of β-catenin and reduce levels of short Tcf1 isoforms relative to

51 long [87]. We verified that 24 h LiCl reduced the ratio of short Tcf1 to long isoforms by 40 ± 3% (range based on quantitation of data shown in Fig. 2.2A and

2.2B, Input blots). We also observed accumulation of β-catenin with 24 h LiCl treatment (Fig. 2.2C) and no change in Gsk3 protein levels (Fig. 2.2C), coincident with the decrease in Bcl11b and Tcf1 co-immunoprecipitation.

Interestingly, there was no change in Bcl11b levels up to 24 h after LiCl treatment (Fig. 2.2A and B, input). We previously showed that stimulation of the

MAPK pathway in thymocytes using a drug combination of phorbol ester and calcium ionophore (P/A) resulted in degradation of Bcl11b after 24 h [51]. This finding reveals divergent effects of Wnt and MAPK pathways on stability of

Bcl11b in thymocytes.

BCL11B can simultaneously complex with β-catenin and TCF1 in Jurkat cells.

Decreased association between BCL11B and TCF1 upon stimulation of the canonical Wnt signaling pathway by LiCl coincided with accumulation of β- catenin. To investigate a potential role for β-catenin in dismissing BCL11B from a

TCF1 transcriptional complex, we utilized Jurkat cells, a T-cell acute lymphoblastic leukemia cell line, as a model system. Jurkat cells have a high basal level of β-catenin compared to other lymphocytes [88]. We prepared native lysates from Jurkat cells to preserve protein-protein interactions and then immunoprecipitated the lysates with either control IgG (goat origin) or anti-

BCL11B antibody (goat origin) covalently linked to agarose beads.

Immunoprecipitates were processed for size separation by SDS-PAGE, proteins immunoblotted onto nitrocellulose, and the resultant blot was probed with antibodies selective for BCL11B, β-catenin, and TCF1. Surprisingly, β-catenin

(band at 92 kDa, Fig. 2.3A) co-immunoprecipitated with BCL11B in Jurkat cells

(Fig. 2.3A). The β-catenin-BCL11B association was further confirmed by reciprocal immunoprecipitation using anti-total β-catenin antibody (rabbit origin).

BCL11B co-immunoprecipitated with β-catenin. Further, as expected, only the long isoforms of TCF1 that contain the β-catenin binding domain also co- immunoprecipitated with β-catenin. Long TCF1 migrates on SDS-PAGE near but sufficiently distinct from the IgG bands at approximately 55 kDa (Fig. 2.3B). We provide evidence for the first time that BCL11B can participate in a complex(es) with β-catenin and/or TCF1. In other words, β-catenin does not dismiss BCL11B from a complex with TCF1.

Bcl11b and Tcf1 have differing interactions with NuRD complex proteins.

To further investigate the nature of the complexes formed by Bcl11b and

Tcf1 and the effects of LiCl or P/A treatment, we searched for co-precipitation of core NuRD complex proteins. As seen in Fig. 2.4A, MTA2 and HDAC2 clearly co- precipitated with Bcl11b at levels that were largely unchanged by stimulation with either LiCl or P/A. However, these same complex proteins did not co-precipitate with Tcf1 at levels detectable by immunoblotting (Fig. 2.4B).

Co-immunoprecipitation of Bcl11b with Tcf1 is decreased with phosphatase inhibitor treatment.

We sought to investigate the effect of PTMs on the Bcl11b-Tcf1 interaction. We treated primary murine thymocytes with calyculin A, an inhibitor of class 1 and 2A phosphatases, for up to 2 h. We previously showed that calyculin A treatment resulted in hyperphosphorylation, a shift in gel mobility, and apparently complete desumoylation of Bcl11b [51]. Calyculin A treatment reduced co-immunoprecipitation of Bcl11b with Tcf1 (Fig. 2.4C), suggesting that the interaction may be dynamically regulated by kinase- or phosphatase- dependent signaling pathways.

BCL11B sumoylation is not critical for interaction with TCF1 and β-catenin in Jurkat cells.

The decrease in BCL11B interaction with TCF1 after hyperphosphorylation induced by calyculin A treatment may be a result of obliteration of sumoylation on BCL11B. Stated another way, sumoylation of

BCL11B may be required for the BCL11B-TCF1 interaction in thymocytes. To test this hypothesis, we used a CRISPR/Cas9-mediated genome editing approach to generate two mutant lines of Jurkat cells that express two different

C-terminally truncated forms of BCL11B. The two mutant forms of human

BCL11B, ΔC1 and ΔC2, are missing 95 and 285 amino acids, respectively, and missing either one or both of two lysines (K689 and K887), homologous to the sumoylation sites previously established in mouse (K679 and K877). First, we verified reduction or loss of sumoylation in the truncated forms of human

BCL11B. Jurkat cells were lysed under denaturing condition to preserve PTMs

54 and subject to SDS-PAGE after immunoprecipitation of BCL11B. The resultant blot was probed with antibodies against both total BCL11B protein and sumoylated lysine (SUMO1). Truncated BCL11B in the ΔC2 mutant sample appeared devoid of sumoylation, but BCL11B in the ΔC1 mutant sample retained some, albeit less, sumoylation relative to wild-type BCL11B in control samples

(Fig. 2.5A).

Second, we determined whether BCL11B truncation mutant immunoprecipitated with TCF1 in the mutated Jurkat cell lines. TCF1 was immunoprecipitated from native lysates of mutant Jurkat cells, followed by immunoblotting with anti-BCL11B and anti-TCF1 antibodies. Both truncated forms of BCL11B co-immunoprecipitated with TCF1, suggesting that neither the

C terminus (up to aa 609) nor sumoylation is required for the BCL11B interaction with TCF1 (Fig. 2.5B). Using similar protocols, we also demonstrated that partial or complete loss of BCL11B sumoylation or truncation of up to 285 C terminal amino acids did not affect BCL11B association with β-catenin in Jurkat cells (Fig.

2.5C).

TCF1 interacts with BCL11B and inhibits its transcriptional activator activity on Bcl2l1 promoter.

We next sought to investigate the functional relevance of the BCL11B-

TCF1 interaction in regulation of the transcription of genes that are known to be involved in thymocyte survival and differentiation, and whose dysregulation is linked to T cell leukemia. As SSTFs, an association between BCL11B and TCF1

55 may result in co-occupancy of promoters or DNA binding sites in co-regulation of genes. To assess co-occupancy, we accessed ChIP-seq data for Bcl11b binding sites from wild-type murine thymocytes collected by the Leid lab (Vogel, et al., manuscript in preparation). We compared Bcl11b ChIP-seq data to previously published ChIP-seq data for Tcf1 binding sites from DP thymocytes [82].

Examining promoter binding sites (-1000 to + 500 bp surrounding annotated transcriptional start sites), TCF1 occupied 10,389 promoter sites throughout the genome, whereas Bcl11b occupied 5358, and the two SSTFs co-occupied the promoter regions of 4127 genes, which is 77% and 40% of all Bcl11b and TCF1 sites, respectively (Fig. 2.6A).

To investigate a potential function for BCL11B-TCF1 promoter co- occupancy, we examined the Bcl2l1 gene which encodes the Bcl-xL protein, a key mediator for inhibition of apoptosis [89-91]. Bcl-xL promotes thymocyte survival, differentiation, and maturation at multiple checkpoints [92], and is an appealing target for anticancer therapy [89]. A high expression level of Bcl-xL is considered a characteristic feature in T-ALL [93] with Bcl-xL implicated in maintaining T cell malignancies. Bcl11b-/- thymocytes display a reduced level of

Bcl-xL protein [94] relative to wild-type cells. Similarly, TCF1 deficient thymocytes express a low level of Bcl-xL and show accelerated death [67, 95]. Interestingly, our ChIP-seq data analysis and tracing display co-occupancy of the Bcl2l1 gene promoter region by Bcl11b and Tcf1(Fig. 2.6B).

Given that BCL11B and TCF1 were reported to individually promote thymocyte survival through upregulation of the Bcl2l1 gene, and that both SSTFs occupy the Bcl2l1 gene promoter, we hypothesized that Bcl11b and TCF1 interaction co-regulates expression of Bcl-xL protein. Using a luciferase reporter assay, we determined the effect of BCL11B and TCF1 individually or together on the Bcl2l1 promoter after co-transfection of HEK293T cells with the gene reporter construct plus mammalian expression vectors encoding for BCL11B and/or TCF1 in a transient system. Easy to transfect, HEK293T are also reported to have a functional Wnt signaling pathway [96], and we observed by Western blot that

HEK293T cells expressed detectable levels of β-catenin without transfection (Fig.

2.6C, Input blot).

Co-immunoprecipitation studies were conducted on native cellular extracts prepared from transiently-transfected HEK293T cells, and revealed that BCL11B and TCF1 interact in HEK293T cells (Fig. 2.6C IP blot). Further, the level of expression of each SSTF did not change upon co-transfection relative to singly transfected cells (Fig. 2.6C input blot).

In the same HEK293T system, we assessed the effects of BCL11B and

TCF1 on the Bcl2l1 gene promoter. Dose response curves of increasing amounts of BCL11B or TCF1 expression vector in the presence of a constant amount of

Bcl2l1-luciferase reporter construct revealed maxima for each SSTF in the luciferase gene reporter assay (Fig. 2.6D). At the maximum, 30 ng of BCL11B construct and 20 ng of TCF1 construct induced 6.6 and 2.5-fold increases in

57 luciferase activity, respectively (Fig. 2.6D). In contrast, when TCF1 and BCL11B were co-transfected, the combined effect of BCL11B (30 ng) and TCF1 (20 ng) on the promoter region of Bcl2l1 gene was significantly less than the effect induced by BCL11B alone (Fig.2.6E), suggesting that TCF1 interferes with the transcriptional activity of BCL11B at the Bcl2l1 promoter region.

Dynamic PTMs of Bcl11b after LiCl treatment of thymocytes

To investigate the potential role of Bcl11b as a nuclear target of Wnt signaling, we assessed the effect of LiCl treatment on the PTMs of Bcl11b. As above, we stimulated primary cultures of murine thymocytes with 20 mM LiCl and harvested at varying times after treatment. Harvest involved a rapid denaturing lysis of the cells to preserve PTMs (but which destroys protein complexes) as we described previously [51, 79]. Analysis of composite Ser/Thr phosphorylation of

Bcl11b in thymocytes reveals a rapid, sustained and significant reduction in

Bcl11b phosphorylation up to 24 h after treatment with LiCl with maximum dephosphorylation at 2 h (Fig. 2.7A) . Distinct from the change in composite Bcl11b phosphorylation after stimulation of MAPK signaling [51], de-phosphorylation of

Bcl11b after LiCl treatment was not preceded by a transient increased phosphorylation.

We next sought to identify the enzyme(s) that may be involved in dephosphorylation of Bcl11b after treatment with LiCl, which is known to inhibit

Gsk3 [97, 98]. Phosphorylation of Gsk3 inhibits the activity of this enzyme [99].

To monitor GSK activity, we used antibodies selective for the inactive forms

(phospho-Ser21 in Gsk3α and phospho-Ser9 in Gsk3β [99]). We found that treatment of thymocytes with LiCl resulted in significantly increased phosphorylation (inhibition) of both Gsk3α and Gsk3β, while the total level of

Gsk3α and β was not changed (Fig. 2.7B), as expected. Our results suggest that

Gsk3 may be a candidate kinase for directly or indirectly maintaining the basal phosphorylation state of Bcl11b in unstimulated thymocytes.

Similarly, we evaluated the effect of LiCl treatment on Erk activity in primary thymocytes by monitoring for Erk1/2 phosphoryation, indicative of Erk activation. We had previously identified Erk1/2 as a component of the signaling pathway involved in regulation of Bcl11b composite phosphorylation following

P/A treatment of primary thymocytes [51]. Using antibodies specific for total

Erk1/2 or phosphorylated (active) Erk1/2 protein [100], we observed that LiCl treatment did not alter Erk1/2 activation, unlike P/A treatment (Fig. 2.7C), and ruling out an unexpected effect of LiCl on Erk.

Stimulation of Wnt/Gsk3 signaling pathway using lithium resulted in a rapid, sustained and significant increase in sumoylation (SUMO1 adduction) of

Bcl11b up to 24 h post-treatment (Fig. 2.8A) with a maximum increase at 2 h in murine thymocytes. The level of sumoylation with LiCl treatment at 2 h was greater than that observed previously with MAPK activation ([51] and data not shown). We observed similar levels of increased SUMO2/3 peptide adduction

(Fig. 2.8B). Increased sumoylation of Bcl11b after 24 h of lithium treatment did not lead to protein degradation. Further, LiCl treatment appeared to induce

59 sumoylation (SUMO1 or SUMO2/3 adduction) of only a relatively small number of proteins in thymocyte lysates (Fig 8C1&2; Gsk3 levels were visualized as loading controls for the immunoblots, and Fig 8C3). The small number of distinctly sumoylated bands in Fig. 2.8C suggested that LiCl is not a major stimulant of sumoylation in thymocytes and caused a somewhat selective effect on Bcl11b.

Stimulation of thymocytes with LiCl for up to 24 h did not produce notable changes in ubiquitination status of Bcl11b protein either (Fig. 8D).

2.5 Discussion

Association of Bcl11b with Tcf1 and dynamic regulation by the Wnt/Gsk3 pathway

A high expression level of Tcf1, and nuclear accumulation of β-catenin from an overstimulated Wnt signaling pathway, are strongly linked to development of colon cancer [101]. Half of the β-catenin direct target genes in human colon cancer cells were de-repressed after loss of one Bcl11b allele

[102]. Attenuation of Bcl11b in mouse and human colon crypt cells promoted the development of intestinal tumors partly through abnormal expression of Wnt target genes [102] . Given this information, we hypothesized an interaction, either direct or indirect in a multi-protein complex, between Tcf1, the nuclear effector of

Wnt/Gsk3-dependent signaling, and Bcl11b in thymocytes [61].

We have identified a novel and dynamic interaction between Tcf1 and

Bcl11b. In addition, we demonstrated an interaction between BCL11B and TCF1 in regulation of the Bcl2l1 gene promoter; Bcl2l1 encodes the antiapoptotic

60 protein Bcl-xL that is required for thymocytes survival. We have also identified

Bcl11b as a new nuclear target of Wnt/Gsk3-dependent signaling in thymocytes using the Gsk3 inhibitor LiCl. As expected, LiCl treatment of murine thymocytes stimulated accumulation of β-catenin, which activates Tcf1/LEF to regulate expression of Wnt target genes [70, 87, 103]. Treatment of thymocytes with LiCl for 8 to 24 h led to a reduction in the Bcl11b-Tcf1 interaction in time dependent manner, coincident with the expected reduction in the ratio of short to long Tcf1 isoforms [87].

We postulate that Bcl11b may play a role in a repressor complex with Tcf1 to occupy some Wnt target genes in the absence of β-catenin. Although we calculated a preferential reduction in the interaction between Bcl11b and short as opposed to the long Tcf1 isoforms, interaction with long Tcf1 isoforms also decreased. These results signify that Bcl11b may interact with a site common to all Tcf1 isoforms that is modified by Wnt pathway stimulation. The Bcl11b-TCF1 interaction that we observe is dynamic and reduced upon stimulation of the Wnt signaling pathway. Short Tcf1 isoforms have been postulated by others to function as natural antagonists of β-catenin-dependent Tcf1 response element binding [9].

One possible explanation for the reduction in the Bcl11b-Tcf1 interaction upon stimulation of the canonical Wnt signaling pathway is that accumulation and nuclear translocation of β-catenin may dismiss Bcl11b from a Tcf1 transcriptional complex. To test this possibility, we used Jurkat cells as a model system as they

61 have unusually high basal levels of β-catenin due to a deficiency of the dual specificity phosphatase PTEN [87, 104]. PTEN normally functions to dephosphorylate and decrease PIP3 levels [105, 106]. Reduced PTEN activity leads to increased PIP3 levels that recruit and activate the AKT kinase [107] which in turn phosphorylates (i.e. inactivates) Gsk3 [87, 108]. Increased PIP3 levels due to a deficiency of PTEN results in constitutive activation of AKT, which leads to constitutive inhibition of Gsk3 and stabilization of β-catenin [87, 88, 109].

We hypothesized that high basal levels of β-catenin in Jurkat cells would inhibit formation of a TCF1-BCL11B complex. Surprisingly, and contrary to our hypothesis, we demonstrated that β-catenin appears in an immunoprecipitated complex with BCL11B and TCF1 in Jurkat cells.

Along with increasing β-catenin levels and decreasing the Tcf1-Bcl11b interaction, LiCl treatment of thymocytes reduced composite Ser/Thr phosphorylation and enhanced sumoylation of Bcl11b over 2 to 24 h.

Additionally, de-sumoylation of Bcl11b after hyper-phosphorylation induced by phosphatase inhibitor also coincided with a decrease in the Bcl11b-TCF1 interaction. Given this information, we hypothesized that sumoylation of Bcl11b might be required for maintaining a Bcl11b-TCF1 interaction. However, our studies using mutant Jurkat cell lines designed to express truncated forms of

BCL11B with partial or complete loss of sumoylation established that sumoylation of BCL11B is not required for interaction with TCF1. Activation of Wnt/Gsk3- related signaling does not solely induce sumoylation of BCL11B, but it also

62 increases sumoylation of many other proteins. For example, TBL1-TBLR1 are sumoylated upon Wnt pathway activation which results in dissociation of these proteins from the corepressor complex (NCoR) and formation of a TBL1-TBLR1-

β-catenin complex that stimulates transcription of Wnt target genes [110]. In addition, sumoylation of β-catenin occurs with Wnt pathway activation and is necessary for β-catenin mediated transcriptional activation [111, 112].

Taken together, activation of the Wnt pathway induces sumoylation of other proteins which might interfere with the BCL11B-TCF1 association. Further,

Wnt pathway activation may alter the PTMs of TCF1 (perhaps TCF1 sumoylation) which may result in dissociation from a BCL11B complex. Finally, if not sumoylation then perhaps altered phosphorylation of BCL11B affects the

TCF1 interaction. Further investigation is required to determine the mechanism responsible for dissociation of BCL11B from TCF1 upon stimulation of Wnt/Gsk3- related signaling.

Regulation of Tcf1 activity, the main transcriptional effector of Wnt signaling in the nucleus in thymocytes [61], occurs as a result of changes in key

Tcf1 partners to activate or repress transcription of Tcf1 target genes. In the absence of Wnt signaling stimulation (i.e. low β-catenin levels), Groucho (the D. melanogaster ortholog of human TLE) binds to a specific domain in Tcf1. In the canonical pathway studied in D. melanogaster, nuclear accumulation of β-catenin after Wnt stimulation forms an enhanceosome with Tcf1 from which the repressor

Groucho is dismissed [71] to activate Wnt target genes [71, 74, 113]. However,

63 transcriptional repression of Tcf1 target genes does not appear to depend solely on Groucho/TLE binding to Tcf1. Only a subset of β-catenin/Tcf1 target genes were de-repressed in a Groucho-null Drosophila mutant, and the expected loss- of-function phenotype was not observed [114]. Moreover, in vitro studies showed that TLE1 does not bind readily to human Tcf1, displaying 100-fold lower affinity for Tcf1 than TCF3 or TCF4 [115]. Additionally, we have preliminary data on the

Tcf1 interactome suggesting that Groucho/TLE1 is very weakly associated with

Tcf1 in murine thymocytes (W. Selman, W.K. Vogel, T.M. Filtz, unpublished data). This information provides for the possibility that other protein(s) interact in a complex with Tcf1 and participate in transcriptional repression of a Groucho- independent subset of β-catenin/Tcf1 target genes in thymocytes.

Our data suggest that Bcl11b may serve as a Tcf1 co-repressor of some

Wnt target genes in thymocytes that is dismissed from the Tcf1 complex upon

Wnt pathway activation. Consistent with this hypothesis, the β-catenin/Tcf1 target genes Wisp2 and Spp1 (encoding osteopontin) were not repressed appropriately in Bcl11b knock-out mouse embryonic fibroblasts (MEFs) during differentiation to adipocytes when compared with wild-type MEFs [116]. Moreover, BCL11B is involved in regulation of Wnt/β-catenin target genes in intestine wherein attenuation of BCL11B in mouse and human colon crypt cells promoted the development of intestinal tumors partly through abnormal regulation of β-catenin signaling pathway [102]. Using a synthetic promoter containing multiple TCF1

64 binding sites in a luciferase reporter construct for assay in HEK293 cells, others observed that Bcl11b significantly attenuated TCF1 activity [102].

Bcl11b functions in association with the NuRD complex in thymocytes [49] and this association is unaffected by MAPK pathway stimulation [51]. Similarly,

Bcl11b association with NuRD core component proteins (MTA2 and HDAC2) was not affected by LiCl treatment (Fig. 2.4A). In contrast, Tcf1 did not co- immunoprecipitate MTA2 and HDAC2 at detectable levels (Fig. 2.4B). However, preliminary mass spectrometry data from our lab suggested that some components of the core NuRD complex, including MTA2, are present in the Tcf1 interactome (W. Selman and W. K. Vogel). Our immunoblot data suggest that

MTA2 levels in a Bcl11b-Tcf1 complex may be much lower than in all Bcl11b complexes. Thus, the NuRD subcomplex incorporating both Bcl11b and Tcf1 may be different than other Bcl11b-containing NuRD subcomplexes.

BCL11B and TCF1 regulate expression of the Bcl2l1 (Bcl-xL) gene

Comparing Bcl11b and Tcf1 ChIP-seq data sets revealed that Bcl11b and

Tcf1 co-occupy the promoters of a considerable number of genes, suggesting that a Bcl11b-Tcf1 complex in thymocytes may co-regulate many transcriptional targets. One of these is the Bcl2l1 gene. In HEK293T we showed that TCF1 antagonizes BCL11B activation of the Bcl2l1 promoter, despite TCF1 acting as a transcriptional activator of Bcl2L1 on its own, albeit with lower efficacy than

BCL11B. HEK293T cells express relatively high levels of β-catenin in the absence of Wnt pathway stimulation. Stabilized β-catenin upregulates expression

65 of Bcl-xL protein and enhances survival of DP thymocytes [117]. β-catenin has been assumed to act through TCF1 to activate Bcl2l1 [117]. However, we demonstrated that β-catenin binds BCL11B as well as TCF1 (Fig. 2.3A &B), and hypothesize that β-catenin activates the Bcl2l1 transcription through a TCF1 and

BCL11B in a required complex. Consistent with this hypothesis, knock-down of

BCL11B in Jurkat cells resulted in a severe reduction in Bcl-xL protein levels

[118] even in the presence of constitutively high β-catenin levels. In other words, even in the presence of TCF1, elevated β-catenin was not able to maintain expression of Bcl-xL protein in BCL11B knock-down Jurkat cells. Therefore, a

BCL11B-β-catenin complex may be involved in regulating expression of the

Bcl2l1 gene and perhaps others.

The mutually antagonistic effect of BCL11B and TCF1 on the promoter region of Bcl2l1 could be attributed to competition between BCL11B and TCF1 to bind to specific DNA binding sites or to transcriptional coactivator(s) such as β- catenin. Establishing the mechanism will require further study, but taken together, the Bcl11b-Tcf1 interaction and interplay may be part of precise mechanism for coordinating expression of many genes including, but not limited to, Bcl2l1.

Evidence for Bcl11b as a target of the Wnt signaling pathway in developing thymocytes

Both pre-TCR and Wnt-dependent cell surface signaling pathways are required for proper thymocyte development. In particular, Wnt signaling is

66 required for thymocytes to transition from the ISP to DP stage [119]. Thymocytes from Wnt3a−/− mice phenocopy Tcf1 knock-out thymocytes and exhibit arrested development at the ISP stage [119]. Thymocytes from Bcl11b-/- mice lack a functional pre-TCR and exhibit arrested development at the DN3 stage; a few thymocytes progress to ISP, but none progress beyond this stage [30].

Introduction of TCRβ or TCRαβ subunits into Bcl11b−/− mice fails to promote transition of arrested cells into DP thymocytes, indicating that reconstitution of pre-TCR signaling cannot compensate for loss of Bcl11b at the ISP to DP transition. Bcl11b−/− ISP cells never progress to the DP stage, despite normal levels of Tcf1 [30]. Notably, stabilization of β-catenin obviates the need for a pre-

TCR in development of DN3 to DP thymocytes [120]. Taken together, these observations are consistent with a role for Bcl11b as a target of Wnt signaling in the ISP to DP transition.

LiCl-induced dephosphorylation of Bcl11b, sustained for at least 24 h after treatment, suggests that Gsk3, which is inhibited by LiCl, is directly or indirectly involved in maintaining the basal phosphorylation state of Bcl11b in unstimulated thymocytes. As a putative substrate for Gsk3, Bcl11b contains three Gsk3 consensus phosphorylation sites based on homology to other validated Gsk3 substrates [121]. We published kinetic data on two of the three sites (S401 and

S762) in murine thymocytes and observed that P/A treatment decreased phosphorylation on S762 at 30–60 min after stimulation [79]. Consistent with these data and as also found by others in T cells [122, 123], we found that P/A

67 increased Gsk3 phosphorylation in thymocytes, which correlates with reduced

Gsk3 activity [99](Fig. 2.7B). These observations allow for the possibility that

Gsk3 is one of several kinases involved in the complex phosphokinetic pattern of

Bcl11b seen after P/A treatment of thymocytes.

Treatment with LiCl, in addition to decreasing phosphorylation, also resulted in a sustained increase in sumoylation of Bcl11b in DP thymocytes (both

SUMO1 and SUMO2/3 adducts). Further, LiCl-induced sumoylation of Bcl11b appears to increase to a greater extent than that seen with MAPK activation by

P/A [51]. In P/A-treated thymocytes, increased Bcl11b sumoylation appeared to result from disassociation of phosphoryation-dependent SENP proteins (SUMO proteases that hydrolyze the SUMO peptide adducts) from the Bcl11b complex

[51]. Similar mechanisms may apply to affect LiCl-dependent sumoylation of

Bcl11b. The effects of LiCl and P/A treatment on Bcl11b PTMs in thymocytes also differed in another notable way: LiCl treatment did not result in significant alteration in Bcl11b ubiquitination and did not lead to degradation of Bcl11b (Fig.

2.8C3 and D1&2). By contrast, long-term treatment with P/A increased sumoylation, ubiquitination and subsequent proteasomal degradation of Bcl11b at 6 to 8 h [51].

In conclusion our data imply that Bcl11b is a novel target of the Wnt/Gsk3 signaling pathway, and together with previous studies from our laboratories [51], suggest that Bcl11b is a nuclear convergence point for MAPK and Wnt signaling pathways in thymocytes. A better understanding of the signaling-dependent

regulation of Bcl11b and Tcf1 co-occupied target genes, and the combinatorial

effects of signaling pathways to control thymocyte development, may provide

insights into the pathophysiology of T cell leukemias and other cancers, and shed

light on possible courses of treatment.

2.6 Figures

Figure 2.1. Bcl11b interacts with Tcf1 in thymocytes. Primary cultures of murine thymocytes from 4-week-old wild-type mice were harvested using a native lysis buffer protocol to preserve protein–protein interactions. A) Lysates were separated into two fractions, and 1.5% was solubilized in SDS-PAGE sample buffer (Input) while the remainder was immunoprecipitated (IP) with rabbit IgG control (IgG) or anti-Tcf1 antibody, rabbit origin (Tcf1). Samples were size-separated by SDS-PAGE and the corresponding blots processed for immunofluorescent detection of proteins using anti-Tcf1, rabbit monoclonal (IB: Anti-Tcf1), or anti-Bcl11b, rat monoclonal (IB: anti-Bcl11b), antibodies followed by incubation with fluorescently-tagged secondary antibodies (green: anti-rabbit IgG, red: anti-rat IgG). Fluorescent signal was detected with a LI-COR Odyssey scanner and software. Tcf1 is composed of multiple splice variants that migrate between 28 and 60 kDa. Bcl11b runs as a doublet of two splice variants between 100 and 130 kDa plus a ladder of slower migrating species. B) Lysates were separated, 1.5% was solubilized SDS-PAGE sample buffer as input control (Input) while the remainder was immunoprecipitated with goat IgG control (IgG) or antibody against Bcl11b (goat origin) cross-linked to agarose beads, and processed for antibody detection as in A. The relative amounts of long (50–60 kDa) and short (28–50 kDa) Tcf1 isoforms were quantified in the immunoprecipitated samples versus input based on signal intensities in the indicated boxed regions. In the input blot, whole cell lysates were immunoblotted for Bcl11b and Tcf1 as in A. C) Lysates were separated, 1.5% was solubilized SDS-PAGE sample buffer as input control (Input) while the remainder was immunoprecipitated (IP) with rabbit IgG control (IgG), anti-Tcf1 antibody, rabbit monoclonal, that detects all Tcf1 isoforms (anti-Tcf1) or anti-Tcf1 antibody, rabbit monoclonal, directed against the β-catenin binding site and detecting only the long isoforms (anti-Tcf1, long). Samples were size-separated by SDS-PAGE and processed for fluorescent antibody detection as in A, and quantified as in B.

Figure 2.2. Bcl11b-Tcf1 interaction dynamics following stimulation of thymocytes. Primary cultures of murine thymocytes were treated with H2O (LiCl vehicle control) or LiCl for the indicated times. A) Cells were harvested as in Fig. 1 to preserve protein–protein interactions. 1.5% of lysates were size separated (Input) and the remainder was immunoprecipitated with anti-rabbit IgG (IgG) control of anti-Tcf1 antibody and processed for immunofluorescent detection of proteins (green; IB: Anti-Tcf1) and Bcl11b protein (red; IB: Anti-Bcl11b) as in Fig. 1. B) Thymocytes were treated and harvested as in A followed by immunoprecipitation with goat antibody against Bcl11b conjugated to agarose beads. Immunoblots were processed for detection of Bcl11b and Tcf1 as in A. C) 1.5% of lysates samples from A were concentrated, SDS-PAGE separated, and immunoblotted using antibodies against β-catenin (green; Anti-β-catenin) or Gsk3-α and β isoforms (red; anti-Gsk3α/β). MW markers migrated as indicated to the left of the blots.

Figure 2.3. BCL11B can simultaneously complex with β-catenin and TCF1 in Jurak cells. A) Jurkat cells were harvested and lysed under conditions to preserve protein-protein interaction as in Fig. 1A & B. Native lysate (1.5 %) was process as input and the reminder was immunoprecipitated with IgG (goat origin) as a control or anti-BCL11B antibody covalently linked to agarose beads (goat origin). After SDS-PAGE size separation, the resultant immunoblot was probed with anti-BCL11B (rat origin), anti- β-catenin antibody (rabbit origin), and anti- TCF1 (rabbit origin) followed by incubation with secondary fluorescently tagged antibodies (red: rat IgG, green: rabbit IgG) for detection of BCL11B in red and both β-catenin and TCF1 in green. β-catenin migrates at 92 kDa and both BCL11B and TCF1 run at expected molecular weights as in Fig. 1A & B. B) As in A, native lysates were divided into input control (1.5%) or immunoprecipitated (98.5%) with anti-β-catenin antibody (rabbit origin) and processed for detection of immunoreactive bands as in A. BCL11B and β-catenin migrate at expected molecular weight as in A, and long TCF1 isoforms migrate at approximately 55 kDa.

Figure 2.4. Dynamic alterations in Bcl11b-Tcf1 complexes. Wild-type murine primary thymocytes were treated as indicated with LiCl or P/A for 2 h. Thymocytes were harvested using a native lysis buffer protocol to preserve protein-protein interactions as in Fig. 1. A and B) Lysates were immunoprecipitated with rat IgG (IP: IgG, A) or rabbit IgG (IP: IgG, B) as controls, anti-Bcl11b (IP: Bcl11b) or anti-Tcf1 (IP: Tcf1) antibodies, and processed for immunoblotting with anti-Bcl11b (red), anti-Tcf1 (green), anti-MTA2 (green) and anti-HDAC2 (green) antibodies followed by fluorescently-tagged and appropriate secondary anti-IgG antibodies. The expected migration by molecular weight of each protein is indicated by placement to the right of the blots. C) Primary cultures of murine thymocytes were treated with 0.1% DMSO (calyculin A vehicle control) or 50 nM calyculin A (Cal A) as indicated time. Cells were harvested and separated into fractions (1.5%) as input or immunoprecipitated with rabbit IgG as a negative control (IgG) or anti-Tcf1 antibodies (IP: Tcf1). Samples were processed for Western blotting as in A.

Figure 2.5. BCL11B sumoylation is not critical for interaction with TCF1 and β-catenin in Jurkat cells. A) Jurkat cell lines, control (C), and C-terminal deletion mutants ΔC1 and ΔC2, were harvested and lysed with denaturing 1% SDS boiling lysis buffer to preserve PTMs. BCL11B was immunoprecipitated from the denatured cell lysates using anti-BCL11B (rat origin) followed by SDS- PAGE size separation. Immunoblot was processed for detection of total BCL11B protein (IB: anti-BCL11B antibody, rat origin) and sumoylated lysine (IB: anti- sumo1 antibody, rabbit origin). Followed by fluorescently-tagged secondary anti- IgG antibodies (red: anti- rat IgG, green: anti-rabbit IgG). B) Jurkat cells (C, ΔC1, and ΔC2) were lysed using the native lysis protocol to preserve protein-protein interactions. Native lysates were divided into two parts: Input control (1.5%) and the remainder (98.5%) were immunoprecipitated with IgG (rabbit origin) or anti- TCF1antibody (rabbit origin). The immunoblot was processed for detection of BCL11B (red) and TCF1 (green) as in Fig. 1A & B. C) Jurkat cells lines noted above were lysed and immunoprecipitated as in B and the resultant immunoblot was then processed for detection of immunoreactive bands (BCL11B in red, and β-catenin in green).

Figure 2.6. TCF1 interacts with BCL11B and inhibits its transcriptional activator activity on the Bcl2l1 promoter. A) Venn diagram showing the number of BCL11B and TCF1 ChIP-seq peaks identified in gene promoter regions (-1000 to +500 bp from transcription start sites) with numbers of co-occupied promoters shown as overlapping. B) Proximal TCF1 and BCL11B chromatin binding domains in the loci for Bcl2l1. TCF1 (purple) and BCL11B (cyan) ChIP-Seq data sets aligned to the mm10 mouse genome on the UCSC Genome Browser website. Overlap between BCL11B and TCF1 binding peaks is indicated by pink bars. The gene show extensive overlap between BCL11B and TCF1 peaks in promoter and intergenic regions. C) HEK- 293T cells were transfected with mammalian expression vector pcDNA3 as a control (C), or with the vector containing cDNA sequences for the human BCL11B 1-2-3-4 variant (B), or human TCF1 (T), or both (B+T). Forty-eight hours post-transfection, cells were harvested and lysed using native lysis buffer to preserve protein-protein interactions. Native lysates were divided in to two parts :1% of sample were processed as input and the reminder was immunoprotected with either IgG (rabbit origin) as a control or with anti-TCF1 antibody (rabbit origin). Immunoblots were processed for detection of BCL11B (IB: Anti-BCL11B antibody; rat origin), anti-TCF1 antibody (IB: Anti-TCF1; rabbit origin), and anti-β catenin antibody (rabbit origin), followed by fluorescently- tagged secondary anti-IgG antibodies (red, anti-rat; and green, anti-rabbit). D) HEK293T cells were transfected with indicated amounts of expression vectors encoding BCL11B or TCF1 and constant (37.5 ng) amount of reporter gene construct containing 2 Kb of the Bcl2l1 promoter upstream of luciferase gene. After 48 h, luciferase activity was assayed and resulted reported as fold stimulation relative zero. E) HEK293T cells were transfected with 30 ng of BCL11B expression vector and/or 20 ng of TCF1 expression vector along with the Bcl2l1 gene reporter construct as in B. The average fold increase in luciferase activity relative to zero is reported and ave ± SD (n = 4). a, b= p≤0.005 compared to zero. c= p≤0.005 compared to zero and compared to 30 ng BCL11B(b).

Figure 2.7. Phosphorylation dynamics of Bcl11b and changes in activity of Gsk3α/β and Erk1/2 after LiCl treatment of thymocytes. Primary cultured murine thymocytes were treated with LiCl for the indicated times. Cells were harvested by a denaturing lysis protocol to preserve PTMs, immunoprecipitated with antibodies against Bcl11b, rat (IB: Anti-Bcl11b), separated by SDS-PAGE and processed for immunoblotting. A1) Immunoblots blots using anti- phosphothreonine/serine, mouse monoclonal (IB: Anti-pT/S), or anti-Bcl11b, rat monoclonal (IB: Anti-Bcl11b), antibodies followed by incubation with fluorescently-tagged secondary antibodies (green: anti-mouse IgG, red: anti-rat IgG). The lower panel of three is an overlay of upper and middle blots with yellow indicating overlapping signals. A2) Quantification of fluorescent intensity in each lane shown in figure A1 by Licor Odyssey fluorescent image scanning and analysis software. Error bars indicate mean ± SD for five independent experiments, *p ≤ 0.05. B1) Cells were treated, lysed and harvested as in A1. Lysates were concentrated by chloroform /methanol extraction, size-separated by SDS-PAGE, and immunoblotting performed for detection of total Gsk3α/β protein (red; anti-Gsk3α/β), or phosphorylated Gsk3α/β (green, anti- phosphoGsk3α/β) followed by fluorescently-tagged secondary anti-IgG antibodies. The lower panel of three is an overlay as in A1. B2) Quantification of fluorescent intensity in each lane shown in figure A1 by Licor Odyssey fluorescent image scanning and analysis software. Error bars indicate mean ± range for two independent experiments. C) Thymocytes were treated, lysed and harvested; and lysates processed for immunoblotting as in B1. Immunoblots were incubated with anti-Erk1/2 (red), or anti-phosphoErk1/2 (green), antibodies with appropriately tagged fluorescent secondary antibodies. The lower panel of three is an overlay as in A1 and B1.

Figure 2.8. Dynamic sumoylation and ubiquitinylation of Bcl11b after LiCl treatment in thymocytes. Cultured thymocytes were treated with LiCl for times indicated, immunoprecipitated with anti-Bcl11b (rat origin) antibodies and processed as in Fig. 3. A1 and B1) Immunoblots were processed for detection of total Bcl11b protein (IB: Anti-Bcl11b, red, A1 and B1), SUMO1-sumoylated lysine (IB: Anti-SUMO1, green, A1), or SUMO2/3-sumoylated lysine (IB: Anti-SUMO2/3, green, B1) followed by fluorescently-tagged secondary anti-IgG antibodies. The lower panel of three is an overlay as in Fig. 3. A2 and B2) Quantification of data shown in figures A and B, respectively, by Licor Odyssey fluorescent image scanning and analysis software. Error bars indicate mean ± SD for four independent experiments each, *p ≤ 0.05. C) Cells were treated, lysed and harvested as in A and B. Lysates were concentrated by chloroform /methanol extraction (Input), size-separated by SDS-PAGE, and immunoblotting performed for detection of total Gsk3α/β protein (red; anti-Gsk3α/β), SUMO1-sumoylated lysine (C1 green; anti-SUMO1), or SUMO2/3-sumoylated lysine (C2 green, anti- SUMO2/3) followed by fluorescently-tagged secondary anti-IgG antibodies. D1) After immunoprecipitation of Bcl11b and SDS-PAGE, the corresponding blot was probed with antibodies against total Bcl11b (red) and ubiquitin (green). Error bars indicate mean ± SD for three independent experiments. D2) Quantification of data shown in D1 by Licor Odyssey fluorescent image scanning and analysis software. Error bars indicate mean ± SD for four independent experiments each. There was no significant different about the three groups.

2.7 References

1. Heinz, S. and C.K. Glass, Roles of lineage-determining transcription factors in establishing open chromatin: lessons from high-throughput studies, in Epigenetic Regulation of Lymphocyte Development. 2011, Springer. p. 1-15. 2. López-Rodríguez, C., J. Aramburu, and R. Berga-Bolaños, Transcription factors and target genes of pre-TCR signaling. Cellular and molecular life sciences, 2015. 72(12): p. 2305- 2321. 3. Berman, J.N. and A. Thomas Look, Targeting transcription factors in acute leukemia in children. Current drug targets, 2007. 8(6): p. 727-737. 4. De Keersmaecker, K., et al., The TLX1 oncogene drives aneuploidy in T cell transformation. Nature medicine, 2010. 16(11): p. 1321-1327. 5. Gutierrez, A., et al., The BCL11B tumor suppressor is mutated across the major molecular subtypes of T-cell acute lymphoblastic leukemia. Blood, 2011. 118(15): p. 4169-4173. 6. Kamimura, K., et al., Haploinsufficiency of Bcl11b for suppression of lymphomagenesis and thymocyte development. Biochemical and biophysical research communications, 2007. 355(2): p. 538-542. 7. O'neil, J. and A. Look, Mechanisms of transcription factor deregulation in lymphoid cell transformation. Oncogene, 2007. 26(47): p. 6838. 8. Przybylski, G., et al., Disruption of the BCL11B gene through inv (14)(q11. 2q32. 31) results in the expression of BCL11B-TRDC fusion transcripts and is associated with the absence of wild-type BCL11B transcripts in T-ALL. Leukemia, 2005. 19(2): p. 201. 9. Tiemessen, M.M., et al., The nuclear effector of Wnt-signaling, Tcf1, functions as a T- cell–specific tumor suppressor for development of lymphomas. PLoS biology, 2012. 10(11): p. e1001430. 10. Van Vlierberghe, P. and A. Ferrando, The molecular basis of T cell acute lymphoblastic leukemia. The Journal of clinical investigation, 2012. 122(10): p. 3398. 11. Wakabayashi, Y., et al., Homozygous deletions and point mutations of the Rit1/Bcl11b gene in γ-ray induced mouse thymic lymphomas. Biochemical and biophysical research communications, 2003. 301(2): p. 598-603. 12. Yu, S., et al., The TCF-1 and LEF-1 transcription factors have cooperative and opposing roles in T cell development and malignancy. Immunity, 2012. 37(5): p. 813-826. 13. Boehm, T. and Y. Takahama, Thymic development and selection of T Lymphocytes. 2014: Springer. 14. Xu, Z., et al., Cutting Edge: β-Catenin–Interacting Tcf1 Isoforms Are Essential for Thymocyte Survival but Dispensable for Thymic Maturation Transitions. The Journal of Immunology, 2017. 198(9): p. 3404-3409. 15. Anderson, G., P.J. Lane, and E.J. Jenkinson, Generating intrathymic microenvironments to establish T-cell tolerance. Nature reviews. Immunology, 2007. 7(12): p. 954. 16. Shortman, K. and L. Wu, Early T lymphocyte progenitors. Annual review of immunology, 1996. 14(1): p. 29-47. 17. Takahama, Y., Journey through the thymus: stromal guides for T-cell development and selection. Nature Reviews Immunology, 2006. 6(2): p. 127-135. 18. Ceredig, R. and T. Rolink, Opinion: a positive look at double-negative thymocytes. Nature reviews. Immunology, 2002. 2(11): p. 888.

19. Dudley, E.C., et al., T cell receptor β chain gene rearrangement and selection during thymocyte development in adult mice. Immunity, 1994. 1(2): p. 83-93. 20. Godfrey, D.I., et al., A developmental pathway involving four phenotypically and functionally distinct subsets of CD3-CD4-CD8-triple-negative adult mouse thymocytes defined by CD44 and CD25 expression. The Journal of Immunology, 1993. 150(10): p. 4244-4252. 21. Staal, F. and H. Clevers, Tcf/Lef transcription factors during T-cell development: unique and overlapping functions. The hematology journal, 2000. 1(1): p. 3-6. 22. Yui, M.A. and E.V. Rothenberg, Developmental gene networks: a triathlon on the course to T cell identity. Nature reviews. Immunology, 2014. 14(8): p. 529. 23. De Obaldia, M.E. and A. Bhandoola, Transcriptional regulation of innate and adaptive lymphocyte lineages. Annual review of immunology, 2015. 33: p. 607-642. 24. Rothenberg, E.V., J. Ungerbäck, and A. Champhekar, Chapter Four-Forging T-Lymphocyte Identity: Intersecting Networks of Transcriptional Control. Advances in immunology, 2016. 129: p. 109-174. 25. Avram, D., et al., Isolation of a novel family of C2H2 zinc finger proteins implicated in transcriptional repression mediated by chicken ovalbumin upstream promoter transcription factor (COUP-TF) orphan nuclear receptors. Journal of Biological Chemistry, 2000. 275(14): p. 10315-10322. 26. Inoue, J., et al., Expression of TCRαβ partly rescues developmental arrest and apoptosis of αβ T cells in Bcl11b−/− mice. The Journal of Immunology, 2006. 176(10): p. 5871- 5879. 27. Kastner, P., et al., Bcl11b represses a mature T‐cell gene expression program in immature CD4+ CD8+ thymocytes. European journal of immunology, 2010. 40(8): p. 2143-2154. 28. Liu, P., P. Li, and S. Burke, Critical roles of Bcl11b in T‐cell development and maintenance of T‐cell identity. Immunological reviews, 2010. 238(1): p. 138-149. 29. Okazuka, K., et al., p53 prevents maturation of T cell development to the immature CD4− CD8+ stage in Bcl11b−/− mice. Biochemical and biophysical research communications, 2005. 328(2): p. 545-549. 30. Wakabayashi, Y., et al., Bcl11b is required for differentiation and survival of [alpha][beta] T lymphocytes. Nature immunology, 2003. 4(6): p. 533. 31. Bhattacharya, S., et al., Ablation of Ctip2/Bcl11b in Adult Epidermis Enhances TPA/UV- Induced Proliferation and Increases Susceptibility to DMBA/TPA-Induced Epidermal Carcinogenesis. Journal of Investigative Dermatology, 2017. 32. Bhattacharya, S., et al., Transcription Factor CTIP2 maintains hair follicle stem cell pool and contributes to altered expression of LHX2 and NFATC1. Journal of Investigative Dermatology, 2015. 135(11): p. 2593-2602. 33. Bollag, W.B., COUP-TF-interacting protein-2 (Ctip2): driving the coupé of sphingolipid biosynthesis in the epidermis. Journal of Investigative Dermatology, 2013. 133(3): p. 593- 594. 34. Ganguli‐Indra, G., et al., Expression of COUP‐TF‐interacting protein 2 (CTIP2) in human atopic dermatitis and allergic contact dermatitis skin. Experimental dermatology, 2009. 18(11): p. 994-996.

35. Golonzhka, O., et al., Dual role of COUP-TF-interacting protein 2 in epidermal homeostasis and permeability barrier formation. Journal of Investigative Dermatology, 2009. 129(6): p. 1459-1470. 36. Wang, Z., et al., Transcription factor Ctip2 controls epidermal lipid metabolism and regulates expression of genes involved in sphingolipid biosynthesis during skin development. Journal of Investigative Dermatology, 2013. 133(3): p. 668-676. 37. Wang, Z., et al., Selective ablation of Ctip2/Bcl11b in epidermal keratinocytes triggers atopic dermatitis-like skin inflammatory responses in adult mice. PloS one, 2012. 7(12): p. e51262. 38. Arlotta, P., et al., Neuronal subtype-specific genes that control corticospinal motor neuron development in vivo. Neuron, 2005. 45(2): p. 207-221. 39. Arlotta, P., et al., Ctip2 controls the differentiation of medium spiny neurons and the establishment of the cellular architecture of the striatum. Journal of Neuroscience, 2008. 28(3): p. 622-632. 40. Simon, R., et al., Structure‐function integrity of the adult hippocampus depends on the transcription factor Bcl11b/Ctip2. Genes, Brain and Behavior, 2016. 15(4): p. 405-419. 41. Simon, R., et al., A dual function of Bcl11b/Ctip2 in hippocampal neurogenesis. The EMBO journal, 2012. 31(13): p. 2922-2936. 42. Albu, D.I., et al., BCL11B is required for positive selection and survival of double-positive thymocytes. Journal of Experimental Medicine, 2007. 204(12): p. 3003-3015. 43. Avram, D. and D. Califano, The multifaceted roles of Bcl11b in thymic and peripheral T cells: impact on immune diseases. The Journal of Immunology, 2014. 193(5): p. 2059- 2065. 44. Ha, V., et al., The T-ALL related gene BCL11B regulates the initial stages of human T-cell differentiation. Leukemia, 2017. 45. Li, L., M. Leid, and E.V. Rothenberg, An early T cell lineage commitment checkpoint dependent on the transcription factor Bcl11b. Science, 2010. 329(5987): p. 89-93. 46. VanValkenburgh, J., et al., Critical role of Bcl11b in suppressor function of T regulatory cells and prevention of inflammatory bowel disease. Journal of Experimental Medicine, 2011. 208(10): p. 2069-2081. 47. Ikawa, T., et al., An essential developmental checkpoint for production of the T cell lineage. Science, 2010. 329(5987): p. 93-96. 48. Li, P., et al., Reprogramming of T cells to natural killer–like cells upon Bcl11b deletion. Science, 2010. 329(5987): p. 85-89. 49. Cismasiu, V.B., et al., BCL11B functionally associates with the NuRD complex in T lymphocytes to repress targeted promoter. Oncogene, 2005. 24(45): p. 6753. 50. Cismasiu, V.B., et al., BCL11B participates in the activation of IL2 gene expression in CD4+ T lymphocytes. Blood, 2006. 108(8): p. 2695-2702. 51. Zhang, L.-j., et al., Coordinated regulation of transcription factor Bcl11b activity in thymocytes by the mitogen-activated protein kinase (MAPK) pathways and protein sumoylation. Journal of Biological Chemistry, 2012. 287(32): p. 26971-26988. 52. Li, L., et al., A far downstream enhancer for murine Bcl11b controls its T-cell specific expression. Blood, 2013. 122(6): p. 902-911. 53. Atcha, F.A., et al., A unique DNA binding domain converts T-cell factors into strong Wnt effectors. Molecular and cellular biology, 2007. 27(23): p. 8352-8363.

54. van Beest, M., et al., Sequence-specific high mobility group box factors recognize 10–12- base pair minor groove motifs. Journal of Biological Chemistry, 2000. 275(35): p. 27266- 27273. 55. Weber, B.N., et al., A critical role for TCF-1 in T-lineage specification and differentiation. Nature, 2011. 476(7358): p. 63. 56. Van de Wetering, M., et al., Extensive alternative splicing and dual promoter usage generate Tcf-1 protein isoforms with differential transcription control properties. Molecular and cellular biology, 1996. 16(3): p. 745-752. 57. Arce, L., N. Yokoyama, and M. Waterman, Diversity of LEF/TCF action in development and disease. Oncogene, 2006. 25(57): p. 7492. 58. Hoppler, S. and C.L. Kavanagh, Wnt signalling: variety at the core. Journal of cell science, 2007. 120(3): p. 385-393. 59. Xing, S., et al., Tcf1 and Lef1 transcription factors establish CD8+ T cell identity through intrinsic HDAC activity. Nature immunology, 2016. 17(6): p. 695. 60. Eastman, Q. and R. Grosschedl, Regulation of LEF-1/TCF transcription factors by Wnt and other signals. Current opinion in cell biology, 1999. 11(2): p. 233-240. 61. Staal, F.J., T.C. Luis, and M.M. Tiemessen, WNT signalling in the immune system: WNT is spreading its wings. Nature reviews. Immunology, 2008. 8(8): p. 581. 62. Staal, F.J. and H.C. Clevers, WNT signalling and haematopoiesis: a WNT–WNT situation. Nature Reviews Immunology, 2005. 5(1): p. 21-30. 63. Xue, H.H. and D.M. Zhao, Regulation of mature T cell responses by the Wnt signaling pathway. Annals of the New York Academy of Sciences, 2012. 1247(1): p. 16-33. 64. Ng, O., et al., Deregulated WNT signaling in childhood T-cell acute lymphoblastic leukemia. Blood cancer journal, 2014. 4(3): p. e192. 65. Staal, F.J., et al., Aberrant Wnt signaling in leukemia. Cancers, 2016. 8(9): p. 78. 66. Ma, J., et al., β-Catenin/TCF-1 pathway in T cell development and differentiation. Journal of neuroimmune pharmacology, 2012. 7(4): p. 750-762. 67. Wang, R., et al., T cell factor 1 regulates thymocyte survival via a RORγt-dependent pathway. The Journal of Immunology, 2011. 187(11): p. 5964-5973. 68. Verbeek, S., et al., An HMG-box-containing T-cell factor required for thymocyte differentiation. Nature, 1995. 374(6517): p. 70-74. 69. Kovalovsky, D., et al., β-Catenin/Tcf determines the outcome of thymic selection in response to αβTCR signaling. The Journal of Immunology, 2009. 183(6): p. 3873-3884. 70. Abu-Baker, A., et al., Lithium chloride attenuates cell death in oculopharyngeal muscular dystrophy by perturbing Wnt/β-catenin pathway. Cell death & disease, 2013. 4(10): p. e821. 71. Cavallo, R.A., et al., Drosophila Tcf and Groucho interact to repress Wingless signalling activity. Nature, 1998. 395(6702): p. 604. 72. Roose, J., et al., The Xenopus Wnt effector XTcf-3 interacts with Groucho-related transcriptional repressors. Nature, 1998. 395(6702): p. 608. 73. Okamura, R.M., et al., Redundant regulation of T cell differentiation and TCRα gene expression by the transcription factors LEF-1 and TCF-1. Immunity, 1998. 8(1): p. 11-20. 74. Daniels, D.L. and W.I. Weis, [beta]-catenin directly displaces Groucho/TLE repressors from Tcf/Lef in Wnt-mediated transcription activation. Nature structural & molecular biology, 2005. 12(4): p. 364.

75. MacDonald, B.T., M.V. Semenov, and X. He, SnapShot: Wnt/β-catenin signaling. Cell, 2007. 131(6): p. 1204. e1-1204. e2. 76. Steinke, F.C., et al., TCF-1 and LEF-1 act upstream of Th-POK to promote the CD4+ T cell fate and interact with Runx3 to silence Cd4 in CD8+ T cells. Nature immunology, 2014. 15(7): p. 646-656. 77. Grumolato, L., et al., β-Catenin-independent activation of TCF1/LEF1 in human hematopoietic tumor cells through interaction with ATF2 transcription factors. PLoS genetics, 2013. 9(8): p. e1003603. 78. Wessel, D. and U. Flügge, A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Analytical biochemistry, 1984. 138(1): p. 141-143. 79. Vogel, W.K., et al., Kinetic Analysis of BCL11B Multisite Phosphorylation– Dephosphorylation and Coupled Sumoylation in Primary Thymocytes by Multiple Reaction Monitoring Mass Spectroscopy. Journal of proteome research, 2014. 13(12): p. 5860-5868. 80. Ran, F.A., et al., Genome engineering using the CRISPR-Cas9 system. Nature protocols, 2013. 8(11): p. 2281. 81. Bauer, D.E., M.C. Canver, and S.H. Orkin, Generation of genomic deletions in mammalian cell lines via CRISPR/Cas9. Journal of visualized experiments: JoVE, 2015(95). 82. Dose, M., et al., β-Catenin induces T-cell transformation by promoting genomic instability. Proceedings of the National Academy of Sciences, 2014. 111(1): p. 391-396. 83. Li, H. and R. Durbin, Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics, 2009. 25(14): p. 1754-1760. 84. Zhang, Y., et al., Model-based analysis of ChIP-Seq (MACS). Genome biology, 2008. 9(9): p. R137. 85. Quinlan, A.R. and I.M. Hall, BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics, 2010. 26(6): p. 841-842. 86. Micallef, L. and P. Rodgers, eulerAPE: drawing area-proportional 3-Venn diagrams using ellipses. PloS one, 2014. 9(7): p. e101717. 87. Lovatt, M. and M.-J. Bijlmakers, Stabilisation of β-catenin downstream of T cell receptor signalling. PLoS One, 2010. 5(9): p. e12794. 88. Tsutsui, J.-i., et al., Expression of cadherin-catenin complexes in human leukemia cell lines. The Journal of Biochemistry, 1996. 120(5): p. 1034-1039. 89. Hata, A.N., J.A. Engelman, and A.C. Faber, The BCL2 family: key mediators of the apoptotic response to targeted anticancer therapeutics. Cancer discovery, 2015. 5(5): p. 475-487. 90. Boise, L.H., et al., bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptotic cell death. cell, 1993. 74(4): p. 597-608. 91. Breckenridge, D.G. and D. Xue, Regulation of mitochondrial membrane permeabilization by BCL-2 family proteins and caspases. Current opinion in cell biology, 2004. 16(6): p. 647-652. 92. Linette, G.P., et al., Bcl-2 is upregulated at the CD4+ CD8+ stage during positive selection and promotes thymocyte differentiation at several control points. Immunity, 1994. 1(3): p. 197-205.

93. Huang, X., et al., Analysis of the expression pattern of the BCL11B gene and its relatives in patients with T-cell acute lymphoblastic leukemia. Journal of hematology & oncology, 2010. 3(1): p. 44. 94. Wakabayashi, Y., et al., Bcl11b is required for differentiation and survival of αβ T lymphocytes. Nature immunology, 2003. 4(6): p. 533-539. 95. Ioannidis, V., et al., The β-catenin–TCF-1 pathway ensures CD4+ CD8+ thymocyte survival. Nature immunology, 2001. 2(8): p. 691. 96. Gujral, T.S. and G. MacBeath, A system-wide investigation of the dynamics of Wnt signaling reveals novel phases of transcriptional regulation. PloS one, 2010. 5(4): p. e10024. 97. Hedgepeth, C.M., et al., Activation of the Wnt signaling pathway: a molecular mechanism for lithium action. Developmental biology, 1997. 185(1): p. 82-91. 98. Stambolic, V., L. Ruel, and J.R. Woodgett, Lithium inhibits glycogen synthase kinase-3 activity and mimics wingless signalling in intact cells. Current biology, 1996. 6(12): p. 1664-1669. 99. Cohen, P. and S. Frame, The renaissance of GSK3. Nature reviews. Molecular cell biology, 2001. 2(10): p. 769. 100. Roskoski, R., Erk1/2 MAP kinases: structure, function, and regulation. Pharmacological research, 2012. 66(2): p. 105-143. 101. Shiokawa, D., et al., The Induction of Selected Wnt Target Genes by Tcf1 Mediates Generation of Tumorigenic Colon Stem Cells. Cell Reports, 2017. 19(5): p. 981-994. 102. Sakamaki, A., et al., Bcl11b SWI/SNF-complex subunit modulates intestinal adenoma and regeneration after γ-irradiation through Wnt/β-catenin pathway. Carcinogenesis, 2015. 36(6): p. 622-631. 103. Yang, Y., et al., Accumulation of β-catenin by lithium chloride in porcine myoblast cultures accelerates cell differentiation. Molecular biology reports, 2011. 38(3): p. 2043- 2049. 104. Shan, X., et al., Deficiency of PTEN in Jurkat T cells causes constitutive localization of Itk to the plasma membrane and hyperresponsiveness to CD3 stimulation. Molecular and cellular biology, 2000. 20(18): p. 6945-6957. 105. Chalhoub, N. and S.J. Baker, PTEN and the PI3-kinase pathway in cancer. Annual Review of Pathological Mechanical Disease, 2009. 4: p. 127-150. 106. Vazquez, F. and P. Devreotes, Regulation of PTEN function as a PIP3 gatekeeper through membrane interaction. Cell Cycle, 2006. 5(14): p. 1523-1527. 107. Carnero, A., et al., The PTEN/PI3K/AKT signalling pathway in cancer, therapeutic implications. Current cancer drug targets, 2008. 8(3): p. 187-198. 108. Lee, G., et al., Phosphoinositide 3-kinase signaling mediates β-catenin activation in intestinal epithelial stem and progenitor cells in colitis. Gastroenterology, 2010. 139(3): p. 869-881. e9. 109. Chung, E.J., et al., Regulation of leukemic cell adhesion, proliferation, and survival by β- catenin. Blood, 2002. 100(3): p. 982-990. 110. Choi, H.-K., et al., Reversible SUMOylation of TBL1-TBLR1 regulates β-catenin-mediated Wnt signaling. Molecular cell, 2011. 43(2): p. 203-216. 111. Huang, H.-J., et al., β-catenin SUMOylation is involved in the dysregulated proliferation of myeloma cells. American journal of cancer research, 2015. 5(1): p. 309.

112. Goretsky, T., et al., Sumoylation of Chromatin-Bound β-Catenin is a Necessary Step in Intestinal Stem Cell WNT/β-Catenin-Dependent Transcriptional Activation. Gastroenterology, 2017. 152(5): p. S963. 113. Mosimann, C., G. Hausmann, and K. Basler, [beta]-Catenin hits chromatin: regulation of Wnt target gene activation. Nature reviews. Molecular cell biology, 2009. 10(4): p. 276. 114. Schweizer, L., D. Nellen, and K. Basler, Requirement for Pangolin/dTCF in Drosophila wingless signaling. Proceedings of the National Academy of Sciences, 2003. 100(10): p. 5846-5851. 115. Chodaparambil, J.V., et al., Molecular functions of the TLE tetramerization domain in Wnt target gene repression. The EMBO journal, 2014. 33(7): p. 719-731. 116. Inoue, J., et al., Identification of BCL11B as a regulator of adipogenesis. Scientific reports, 2016. 6. 117. Xie, H., et al., Stabilized β-catenin extends thymocyte survival by up-regulating Bcl-xL. The Journal of Immunology, 2005. 175(12): p. 7981-7988. 118. Grabarczyk, P., et al., Inhibition of BCL11B expression leads to apoptosis of malignant but not normal mature T cells. Oncogene, 2007. 26(26): p. 3797. 119. Luis, T.C., et al., Wnt3a deficiency irreversibly impairs hematopoietic stem cell self- renewal and leads to defects in progenitor cell differentiation. Blood, 2009. 113(3): p. 546-554. 120. Gounari, F., et al., Somatic activation of β-catenin bypasses pre-TCR signaling and TCR selection in thymocyte development. Nature immunology, 2001. 2(9): p. 863-869. 121. Taelman, V.F., et al., Wnt signaling requires sequestration of glycogen synthase kinase 3 inside multivesicular endosomes. Cell, 2010. 143(7): p. 1136-1148. 122. Staal, F.J., et al., Tcf-1-mediated transcription in T lymphocytes: differential role for glycogen synthase kinase-3 in fibroblasts and T cells. International immunology, 1999. 11(3): p. 317-323. 123. Welsh, G.I., et al., T-cell activation leads to rapid stimulation of translation initiation factor eIF2B and inactivation of glycogen synthase kinase-3. Journal of Biological Chemistry, 1996. 271(19): p. 11410-11413.

Chapter 3

Alteration of Bcl11b upon stimulation of both MAP Kinase- and GSK3-dependent signaling pathways in double negative thymocytes

Wisam H. Selman, and Theresa M. Filtz

Submitted to “Biochemistry and Cell Biology Journal”

3.1 Abstract

Bcl11b is a site-specific transcription factor (SSTF) that is critical for proper thymocyte development and establishment of a functioning immune system. We have previously characterized the kinetic post-translational modifications (PTMs) of Bcl11b in double positive (DP) thymocytes during stimulation of the T cell receptor-activated MAP kinase pathway. However, the PTMs of Bcl11b in thymocytes from other developmental stages in the thymus, primarily double negative (DN) cells, have not been previously characterized. We have found that kinetic modifications of Bcl11b in DN cells are somewhat different than the patterns observed in DP cells. Distinct from DP thymocytes, phosphorylation and sumoylation of Bcl11b in DN cells were not oppositely regulated in response to activation of MAP kinase, even though hyper-phosphorylation of Bcl11b coincided with near complete desumoylation. Additionally, prolonged stimulation of the MAP kinase pathway in DN cells, unlike DP thymocytes, did not alter Bcl11b levels of sumoylation or ubiquitination, or stability. On the other hand, activation of

Wnt/Gsk3-dependent signaling in DN cells resulted in composite dephosphorylation and sumoylation of Bcl11b. We also observed higher activity of

Wnt signaling in DN cells compared to DP cells as indicated by higher β-catenin levels and a higher ratio of long (β-catenin responsive) to short (β-catenin independent) Tcf1 isoforms. Interestingly, unlike in DP thymocytes, we did not observe an interaction between Bcl11b and Tcf1 in DN-like cells. Our results

96 demonstrate that PTMs and specific interacting partners of Bcl11b in DN cells are distinct from those previously characterized in DP thymocytes. Defining the signaling pathways and regulation of SSTFs by PTMs at various stages of thymopoiesis will provide information that may be used to improve our understanding of leukemogenesis.

3.2 Introduction

Intrathymic T cell maturation proceeds in discrete stepwise developmental stages to produce appropriate numbers of functional mature T cells [1], and is associated with changes in accompanying T cell receptor (TCR) complexes [2].

The major stages of intrathymic development include specification whereby early thymic progenitor cells entering from the bone marrow are converted in response to Notch signaling to DN (double negative, i.e. lacking the CD4 and CD8 cell surface markers) cells. In sub-stages of development, DN2 cells become committed to the T cell lineage by passage to the DN3 stage. The cells then pass quickly through the β selection checkpoint to transient ISP (intermediate single positive) and onto DP (double positive, CD4+, CD8+) cells. Regulation of β selection is controlled by signals through the pre-T cell receptor (pre-TCR) complex which is expressed on DN3 thymocytes [2] . DP cells comprise the majority of intrathymic T cells and are selected by a winnowing process to remove cells that are autoreactive or inactive, with the survivors progressing to either

CD4+ or CD8+ SP (single positive) cells that migrate to the periphery for final maturation. Progression and selection is controlled by intrathymic juxtracrine and

97 paracrine signals acting mainly but not exclusively through Notch [3], Wnt [4, 5], pre-TCR [2, 6, 7] and TCR [8, 9] signaling pathways. Conversion of extracellular signals into a coordinated genetic response is critical for each phase of maturation. Breaks in communication at any stage may lead to aberrant proliferation of dysfunctional progenitors as is suspected in childhood leukemogenesis [10-12], or a shunting of cells into other pathways. (e.g. NKT or dendritic cells) [13-15].

T cell development is strictly guided by site specific transcription factors

(SSTF), which are DNA binding proteins or DNA-binding complex-associated proteins that regulate cell function in response to cellular signaling by altering gene transcription programs. Signal transduction pathways modulate the activity of transcription factors [16] through, among other means, reversible, covalent post-translational modifications (PTMs). Modulation of SSTFs by signal transduction-stimulated PTMs results in integration of external stimuli to appropriate cellular responses [17]. The transcription factor B cell leukemia/lymphoma 11B (Bcl11b, Ctip2) [18] is a highly modified SSTF on which

18 kinetically modified phosphorylation sites [19], 2 sumoylation sites, ubiquitinylation sites [20], and 4 arginine methylation sites ([21] and our unpublished data) have been identified. Bcl11b is mutated or disrupted in many cases of T cell malignancy including 9-14% of human T-ALL patient samples [22-

24], suggesting that Bcl11b functions as a tumor suppressor in developing thymocytes or T cells [25]. Bcl11b plays essential roles in thymocyte survival,

98 maturation and maintenance of identity [26] through at least three major checkpoints: T cell commitment at the DN2 to DN3 state [14, 15, 27], β selection of DN3 to DP thymocytes [28], and positive selection to SP thymocytes ([29, 30].

Pre-TCR and TCR activation stimulates, among other intracellular signals, the classical MAP kinase pathway that culminates in activation and nuclear translocation of Erk1/2 [31-33]. Activation of Wnt receptors on thymocytes leads to inhibition of Gsk3, allowing for accumulation of β-catenin followed by translocation into the nucleus [34]. Both pre-TCR MAP Kinase and Wnt-Gsk3 signaling pathways are required for proper maturation of DN to DP thymocytes [2, 4, 31, 35] as is Bcl11b [28]. We demonstrated that Bcl11b in DP thymocytes is kinetically modified by stimulation of the TCR/MAP kinase pathway [20] and Wnt-GSK3 signaling pathways (Chapter two). However, DN thymocytes contain a pre-TCR complex whose structure and intracellular signaling components are subtly different than the immature TCR in DP cells and that may signal significantly different functional outcomes. For example, engagement of the pre-T cell receptor in DN cells with anti-CD3 antibody resulted in cell proliferation, but in DP thymocytes engagement of immature TCR receptor resulted in death [36].

Given the differences between activation of pre-TCR and immature TCR in

DN and DP cells, we sought to clarify the modifications of Bcl11b in DN thymocytes in response to activation of both pathways MAP Kinase- and

Wnt/Gsk3-dependent signaling pathways.

3.3 Materials and Methods

Cell culture

P2C2 (SCID.adh2C2) [37] cells were cultured at 37 °C, 5% CO2 in RPMI-

1640 medium plus 10% fetal bovine serum (FBS, Hyclone), 1% sodium pyruvate,1% nonessential amino acids, and penicillin/streptomycin. Prior to treatments, FBS in the cell medium was reduced to 2.5% and cells incubated for 5 hours.

Cell stimulation

To mimic activation of T-cell receptor signaling pathways including the

MAPK pathways, cells were treated with a combination of 100 nM phorbol 12,13- dibutyrate and 500 nM A23187 (P/A) [20, 38], or with 0.1% DMSO as vehicle control. To mimic stimulation of Wnt signaling pathway, thymocytes were treated with 20 mM LiCl or with water as vehicle control. In experiments to induce hyperphosphorylation, cells were treated with 50 nM calyculin A (Cell Signaling

Technology, Danvers, MA) or 0.1% DMSO as vehicle control.

Preparation of cell lysates

To study the posttranslational modifications (PTMs) of Bcl11b, cultured cells were harvested and rapidly lysed by boiling for 25 min in denaturing 1% SDS buffer containing 20 mM HEPES, pH 7.4, 200 mM NaCl, 2 mM EDTA, 10% glycerol, and 1% SDS with the protease inhibitor cocktail of 0.1 mM PMSF,

1μg/mL pepstatin A, 5 μg/mL leupeptin, 10 μM E64. Included in the buffer were 50 mM NaF and 5 mM Na2P2O4 to broadly inhibit phosphatases, and 10 mM N-

100

Ethylmaleimide to preserve sumoylation. Cell lysates were sonicated and then cleared by centrifugation. Samples (1.5% of total cleared lysate) were removed and proteins were extracted by chloroform/ methanol precipitation [39] for analysis as input. The remaining lysates were diluted 10-fold in denaturing lysis buffer without SDS but including 1.1% Triton-X100 and then immunoprecipitated [19,

20].

To preserve protein-protein interactions, a non-denaturing “native” lysis buffer was used to lyse the cells that included 20 mM HEPES, pH 7.4, 250 mM

NaCl, 2 mM EDTA, 10% glycerol, and 0.5% NP-40 plus protease inhibitor cocktail and the chemical compounds used to preserve sumoylation, ubiquitination and phosphorylation as described above. P2C2 cells were incubated in native lysis buffer on ice for 30 minutes with vortexing every 5 minutes, then sonicated on ice to shear DNA. The resulting native cellular lysates were cleared by centrifugation,

1.5% were preserved for analysis as input, and the remained was used immediately for immunoprecipitation.

Immunoprecipitation and Immunoblotting

For studies of PTMs, Bcl11b was immunoprecipitated from cell lysates with

0.25 µg/ml of rat anti-Bcl11b antibody clone #25B6 (Abcam, ab18465). To study protein-protein interactions, immunoprecipitation of Bcl11b protein from native lysates was performed using 0.25 µg/ml of goat anti-Bcl11b antibody covalently linked to agarose beads (prepared as previously described in [20]). Anti-Tcf1 monoclonal antibody C46C7 (#2206, rabbit, Cell Signaling, 1:100 dilution) was

101 used for immunoprecipitation of Tcf1 from native lysates. Immunoprecipitated samples were subjected to SDS-PAGE, and immunoblotting was performed with one or more of the following selective antibodies: anti-Bcl11b (ab18465, Abcam, rat, 0.5 ng/ml); anti-Erk 1/2 (sc-292838, Santa Cruz, rabbit, 1:1,000 dilution); anti- phospho-Erk1 (pT202/pY204)/phospho-Erk2 (pT185/pY187) (ab50011, Abcam, mouse, 1:10,000 dilution); anti-phospho-threonine/serine (9381,Cell Signaling, mouse, 1:1,000 dilution); anti-SUMO-1 (ab32058, Abcam, rabbit,1:1,000 dilution); anti-mono- and polyubiquitinylated conjugates (BML-PW8810-0100, ENZO, mouse, 1:1,000 dilution); anti-Gsk-3α/β (sc-7291, Santa Cruz, mouse, 1:500 dilution); anti-phosphoserine21/9-Gsk-3α/β, (9331, Cell Signaling, rabbit, 1:1,000 dilution); and anti-β-Catenin (# 8480, Cell Signaling, rabbit, 1:1,000 dilution).

Immunoreactive bands were detected by use of the following fluorescently- tagged secondary antibodies from LI-COR: anti-rat IgG (680 nm, # 926-68076,1:

20,000 dilution); anti-rabbit IgG (680 nm, #926-68071, 1: 20,000 dilution); anti- mouse IgG (800 nm, #926-32210, 1: 5,000 dilution); anti-rabbit IgG (800 nm,

#926-32211,1: 5,000 dilution); anti-mouse IgG (680 nm, # 926-68070,1: 20,000 dilution). A LI-COR Odyssey scanner was used to visualize, quantify and analyze the intensities of resultant fluorescent bands using LI-COR software

(ImageStudio, version 2.1).

102

Statistical Analysis

Data were analyzed statistically by one-way analysis of variance (ANOVA) followed by least significance difference (LSD) comparisons with SPSS statistics software, version 24 (IBM); a p-value ≤ 0.05 was considered statistically significant.

3.4 Results

Phosphorylation kinetics of Bcl11b in P2C2 cells after stimulation with P/A.

Previous study [20] revealed that Bcl11b PTMs are modulated by the

TCR-linked MAPK signaling pathway in DP thymocytes. However, to the best of our knowledge, the kinetics and regulation of Bcl11b PTMs in DN3 thymocytes are unknown. To characterize the PTMs that regulate Bcl11b in DN3 thymocytes upon stimulation of the MAPK pathway, we used a model system for DN3-like thymocytes, P2C2 cells, and treatment with P/A for up to 2 h. Cells were harvested and rapidly lysed by boiling in denaturing 1% SDS buffer to preserve

PTMs as described previously [19, 20]. As expected, P/A treatment of P2C2 cells induced phosphorylation (activation) of Erk1/2, a key component of the MAPK pathway, and Erk1/2 phosphorylation levels remained elevated for up to 8 hours with continuous treatment (Fig. 3.1A & Fig. 3.3C). Bcl11b was phosphorylated on

Ser/Thr residues in the absence of stimulation and upon treatment with P/A, phosphorylation was rapidly, transiently and significantly increased at 7 minutes.

Ser/Thr phosphorylation then declined to significantly less than basal phosphorylation after 1 h of treatment. Finally, phosphorylation of Bcl11b

103 returned to levels indistinguishable from basal after 2 h of P/A treatment (Fig.

3.1B).

The composite phosphorylation kinetics of Bcl11b in P2C2 cells are roughly similar to the pattern that we previously observed in DP thymocytes [20].

One small difference in the time course is that dephosphorylation of Bcl11b reached a maximum after 1 h of P/A stimulation in P2C2 cells rather than 30 minutes in DP thymocytes, more similar to the kinetics of Bcl11b dephosphorylation in Jurkat cells [40].

Sumoylation kinetics of Bcl11b after stimulation with P/A and inhibition of phosphatase in P2C2 cells.

P/A stimulation induced desumoylation of Bcl11b in P2C2 cells that was statistically significant 1 to 2 h after stimulation (Fig. 3.2A). In P2C2 cells, both phosphorylation and sumoylation of Bcl11b were significantly lower than basal levels after 1 h of P/A treatment. This pattern is distinct from DP thymocytes in which sumoylation and phosphorylation were regulated in opposition (i.e., over the P/A treatment time course of 2 hours, increased sumoylation was coincident with decreased phosphorylation and vice versa) [20]. To further investigate the relationship between phosphorylation and sumoylation, we then treated P2C2 cells with 50 nM calyculin A, an inhibitor of class 1 and 2A phosphatases, for up to 1 h, resulting in hyperphosphorylation of Bcl11b and associated with a clear shift in gel mobility as seen previously. The hyperphosphorylation of Bcl11b coincided with obliteration of detectable sumoylation (Fig. 3.2B), and mimicked

104 the phosphorylation-desumoylation pattern that we previously reported in DP thymocytes [20].

Bcl11b degradation and ubiquitination in P2C2 cells with extended P/A treatment.

To determine the effect of prolonged treatment of P/A on stability and

PTMs of Bcl11b in DN3-like thymocytes, we treated P2C2 cells with P/A for up to

24 h. We observed that extended P/A stimulation of P2C2 cells is clearly different in its effect on Bcl11b compared to DP thymocytes. Prolonged treatment with P/A did not induce degradation of Bcl11b protein in P2C2 cells (Fig. 3.3A1, B1 & C), and ubiquitination and sumoylation were not changed dramatically (Fig. 3.3A

(1&2) &B (1&2)). Erk1/2 activation induced by P/A treatment decreased in time- dependent manner over 24 h in P2C2 cells (Fig. 3.3C).

LiCl treatment of P2C2 cells, and regulation of Bcl11b PTMs, kinases, and

β-catenin.

In addition to the MAP kinase pathway, Wnt signaling through a Gsk3/β- catenin-dependent signaling pathway is also required for proper development of

DN thymocytes. Inhibition of Gsk3, indicated by phosphorylation, is required for accumulation of β-catenin. LiCl is a Gsk3 inhibitor that is commonly used to mimic activation of Wnt signaling pathways. We treated P2C2 cells with 20 mM

LiCl and observed a time-dependent increase in phosphorylation of phospho-S21 in Gsk3α and phospho-S9 in Gsk3β (Fig. 3.4A1 & A2), indicative of Gsk3 inhibition [41]. Inhibition of Gsk3 by LiCl treatment also resulted in accumulation

105 of β-catenin, reaching a maximum level at 8 h and sustained to 24 h after LiCl stimulation. Interestingly, and distinct from DP thymocytes in which β -catenin was undetectable prior to GSK3 inhibition (Chapter two, Fig. 2.7C), unstimulated

DN3-like thymocytes have a relatively high level of β-catenin (Fig. 3.4A1). As expected, LiCl treatment did not affect the activity of Erk1/2 in P2C2 cells.

Next, we sought to investigate the effect of Wnt-Gsk3 pathway activation on the phosphorylation status of Bcl11b in P2C2 cells. Coincident with inhibition of Gsk3 activity, LiCl treatment for 24 h induced rapid, sustained and significant dephosphorylation of Bcl11b, reaching a maximum decrease at 8 h.

Sumoylation (SUMO1) kinetics of Bcl11b after LiCl treatment of P2C2 cells.

In addition to dephosphorylation of Bcl11b, activation of the Wnt/Gsk3 pathway by LiCl resulted in a rapid, sustained and significant increase in sumoylation of Bcl11b up to 24 h after treatment. Coincident with maximum dephosphorylation, maximum sumoylation of Bcl11b after LiCl stimulation occurred at 8 h post-stimulation (Fig. 3.5A1, A2).

Bcl11b status in P2C2 cells following extended treatment with LiCl.

Following 24 h treatment of DP thymocytes with P/A, Bcl11b sumoylation was accompanied by ubiquitinylation in preparation for proteosomal degradation

[20]. However, a significant increase in sumoylation of Bcl11b induced by prolonged treatment of P2C2 cells with LiCl did not change its ubiquitinylation status and did not appear to stimulate degradation of Bcl11b (Fig. 3.6A).

Bcl11b-Tcf1 interaction in P2C2 cells.

106

Tcf1 is a SSTF and the effector of Wnt signaling in the nucleus of thymocytes. Tcf1 exists as multiple splice variants in thymocytes (at least 8 different forms), the shorter versions of which are lacking the β -catenin binding site and are thought to have separate and likely antagonistic functions relative to the longer, β-catenin-sensitive isoforms [42-44]. Activation of Wnt signaling increases the ratio of longer to shorter Tcf1 isoforms presumably alleviating the

“antagonistic” activity of the short form that are proposed to interfere with long (β- catenin) responsive activation of Tcf1 responsive genes [43, 45].

DN3-like cells express a higher proportion of Tcf1 longer isoforms (45-72 kDa) compared to shorter “antagonistic” Tcf1 isoforms (36-45 kDa), (Fig. 3.7A).

Higher expression of both β-catenin (Fig. 3.4A1) and long activating Tcf1 isoforms

(Fig. 3.7A) indicates higher basal activity of the Wnt signaling pathway in DN3-like thymocytes compared with DP thymocytes.

A physical interaction between Bcl11b and Tcf1 was previously established in DP thymocytes where Bcl11b appears to preferentially (but not exclusively) interact with shorter Tcf1 isoforms, and this interaction is regulated by the Wnt signaling pathway (Chapter two). To determine if Bcl11b and Tcf1 are also interacting in DN3-like thymocytes, co-immunoprecipitation studies were performed using P2C2 cell lysates prepared under conditions that preserve protein-protein interactions. Native cellular lysates were immunoprecipitated with either antibody against Bcl11b or a monoclonal antibody that detects all Tcf1 isoforms. Unlike DP thymocytes, reciprocal co-immunoprecipitation studies

107 followed by western blotting analysis in P2C2 cells did not show Bcl11b-Tcf1 interaction at detectable levels either in non-stimulated cells or after stimulation of

Wnt signaling pathway using 20 mM LiCl for up to 24 h. (Fig. 3.7A&B).

3.5 Discussion

Pre-TCR-MAPK pathway regulation of Bcl11b PTMs in P2C2 cells:

Thymocyte development proceeds through stages of increasingly restricted potential, specified and strictly coordinated by numerous SSTFs [46].

SSTFs act as information nodes [47], integrating external and internal signals into gene expression programs often through regulation by PTMs [17] to control appropriate cellular development and function [17]. Bcl11b is an essential SSTF in proliferation and proper differentiation of thymocytes [28, 46] which is also highly modified by PTMs [19, 20]. Dysregulation of Bcl11b has been linked to T- cell malignancies [22, 48].

In mouse thymocytes, Bcl11b exists in association with the NuRD

(nucleosome remodeler and deacetylase) complex [18, 29] and appears to function mainly but not exclusively as a transcriptional repressor [20, 49]. Our research groups previously characterized Bcl11b PTMs and kinetics in double positive (DP) thymocytes over a stimulation time course following treatment with

P/A, a drug combination previously established to mimic mature T cell receptor activation [20]. We found that stimulation-dependent dephosphorylation- sumoylation of Bcl11b was coincident with reduction of its repressive transcriptional activity at the Id2 target oncogene through recruitment of the

108

histone acetyltransferase (HAT) p300 coactivator without replacement of NuRD

complex [20]. Additionally, extended P/A treatment then prepared Bcl11b for

ubiquitinylation and proteaosomal degradation in stimulated DP thymocytes [20].

We sought to characterize the PTMs modifying Bcl11b in DN3-like cells to better understand the regulation of this protein in an earlier thymocyte context.

Bcl11b is expressed earliest in DN2 cells. Herein, we quantitated the alteration in

Bcl11b PTMs in a DN3-like cell line following stimulation with the TCR/MAP kinase pathway activators P/A and the classic Wnt/Gsk3/β-catenin pathway activator LiCl. DN thymocytes are very scarce in the thymus, comprising about

1% of all thymocytes [50] and cannot be maintained in culture [51]. To address this issue, we identified Bcl11b PTMs in DN cells by utilizing an in vitro model system, P2C2 (SCID.adh2c2) cells, as a surrogate for DN3 cells. These cells were originally isolated from a spontaneously arising thymic lymphoma in SCID mice, which was adapted to grow in vitro [52]. P2C2 cells are newly-committed

DN3 thymocytes arrested prior to β-selection due to a defect in TCR gene rearrangement which is inherent in SCID mice [37, 52].

Stimulation of pre-TCR signaling pathways (present in DN cells) using P/A

was previously shown to result in activation of Erk1/2, a known downstream

effect of activation of the pre-TCR complex in a DN-like cell line derived from

SCID mice [31]. In P2C2 cells, activation of Erk1/2 by P/A treatment is sustained

for up to 8 h and is prolonged in P2C2 cells compared to DP thymocytes where

109

Erk1/2 activity increases up to 1 h and then decreases after stimulation with P/A

[20].

As described herein, Erk1/2 activation by P/A coincided with rapid and transient phosphorylation of Bcl11b after 7 minutes, followed by a decline in phosphorylation to less than basal before returning to a basal level of phosphorylation after 120 minutes of treatment. This effect of P/A on Bcl11b phosphorylation kinetics over 2 h directly mimics that seen with P/A treatment of

DP thymocytes [20]. As in DP thymocytes, we suggest that Bcl11b dephosphorylation in P2C2 cells after 30 minutes of P/A stimulation may be a result of Erk1/2 activation of a slower phosphatase(s) resulting in the bimodal phosphorylation followed by dephosphorylation pattern of Bcl11b in P2C2 cells.

In addition to transient phosphorylation followed by dephosphorylation of

Bcl11b, P/A treatment resulted in transient desumoylation of Bcl11b up to 2 h, followed by a return to basal sumoylation levels after 4 h of treatment in P2C2 cells. The effect of P/A on sumoylation of Bcl11b in P2C2 is different than the kinetic pattern observed in DP thymocytes. In DP thymocytes, a Bcl11b phospho- desumo switch results in opposing regulation of phosphorylation and sumoylation levels on Bcl11b, i.e., dephosphorylation of Bcl11b promotes resumoylation of this protein by decreasing interaction with SENP proteases which mediate desumoylation of BCl11B [20]. Therefore, we propose that differences between

DP and DN3 cell signaling networks, two distinct stages of thymocyte development, linked to pre-TCR and immature TCR lead to somewhat different

110 regulation and distinct kinetics of PTMs on Bcl11b. To further investigate the relationship of phosphorylation to sumoylation in P2C2 cells, we treated with calyculin A, a broad inhibitor of type 1 and 2A phosphatases, to induce hyperphosphorylation of cellular proteins including Bcl11b. Treatment of P2C2 cells with calyculin A resulted in near complete desumoylation of Bcl11b, confirming that a Bcl11b phosphorylation-desumoylation link or switch is active in

DN3-like thymocytes but not in response to P/A treatment.

Prolonged treatment of DP cells with P/A was previously shown to result in resumoylation of Bcl11b, increased ubiquitinylation at 6 to 8 h, and near complete proteosomal degradation by 24 h [20]. In contrast, treatment of P2C2 cells with P/A for 2 to 24 h resulted in little change in Bcl11b ubiquitination or sumoylation. Over the same time course, Erk1/2 activation increased and then started to decline after 8 h of P/A treatment. Further, and somewhat surprisingly,

Bcl11b remained intact (i.e., was not degraded) up to 24 h after stimulation of

P2C2 cells with P/A. The lack of degradation is consistent with the lack of change in Bcl11b ubiquitinylation status over 24 h of P/A stimulation with P/A. The reason for and consequence of the stability of Bcl11b in P2C2 cells relative to DP cells following cell stimulation is unknown.

We demonstrate that the PTMs modifying Bcl11b after stimulation with

P/A in DN thymocytes are somewhat distinct from those described in DP thymocytes, and might be attributable to differences in intracellular signaling networks between the two distinct stages of thymocyte development. Differences

111 between DN and DP cells include the structural differences in the TCR α subunits and in the signaling pathway components. Functionally, stimulation of pre-TCR signaling in DN thymocytes with anti-CD3 antibody resulted in proliferation of DN thymocytes. However, stimulation of immature TCR signaling in DP thymocytes with the same antibody resulted in increased apoptosis of DP thymocytes [36].

These different functional outcomes suggest that even though pre-TCR and immature TCR complexes are very similar in proximal signaling components, there are important differences in the signaling networks depending on stage of thymocyte development.

Wnt/Gsk3 pathway regulation of Bcl11b PTMs in P2C2 cells:

In parallel to signals induced via the pre-TCR-MAP kinase pathway, the

Wnt/Gsk3/β‐catenin signaling pathway is also required for proper differentiation of DN3 cells into DP thymocytes [4]. Deletion of β‐catenin, which is considered the central mediator of the classic Wnt/Gsk3 signaling pathway, resulted in impairment of differentiation of DN3 into DP thymocytes [53]. Interestingly, some data in the literature suggest that an interplay may exist between the pre-TCR and Wnt /Gsk3 signaling pathways during DN thymocyte development. Activation of pre-TCR signaling transiently stabilizes β‐catenin in DN thymocytes [54], potentially influencing the strength or duration of Wnt/Gsk3 pathway activation.

Conversely, the Wnt/Gsk3 pathway was also suggested to affect pre-TCR signaling where deletion of β‐catenin reduced induction of early growth factor gene expression associated with pre-TCR signaling activation [54]. Collectively,

112 the interplay and cooperation of pre-TCR and Wnt/Gsk3 pathways are required for driving proper differentiation of DN thymocytes.

Interestingly, one example of the differences in cellular context between

DN and DP thymocytes is the activity of the Wnt signaling pathway. As previously shown by others [55], DN cells have a significantly higher basal level of β‐ catenin, and a higher ratio of “activating” long Tcf1 isoforms relative to “inhibitory” short Tcf1 isoforms compared with DP thymocytes. DN thymocytes also have lower levels of Wnt inhibitory molecules such as Axin and HMG2L1 than DP cells

[55].

Accordingly, we have characterized the PTMs of Bcl11b after stimulation of the Wnt/Gsk3 signaling pathway in DN3-like cells (P2C2 cells). To mimic stimulation of Wnt/Gsk3 signaling, we treated P2C2 cells with LiCl which resulted in inhibition of Gsk3 and enhanced accumulation of β-catenin, but did not affect

Erk1/2 activity, as expected. Thus, any resultant effects of LiCl treatment on

Bcl11b were not a result of off-target effects on the MAPK pathway. Treatment of

P2C2 cells with LiCl for 24 h resulted in composite Ser/Thr de-phosphorylation and increased sumoylation of Bcl11b. Enhancement of Bcl11b sumoylation over

24 h of LiCl treatment was not associated with a change in ubiquitinylation of

Bcl11b and did not affect Bcl11b stability.

LiCl treatment reduced Bcl11b phosphorylation levels without a transient increase, which was different from the kinetics of phosphorylation following P/A treatment. Our data suggest that Gsk3 may be a kinase that maintains basal

113 phosphorylation of Bcl11b in thymocytes. Stable enhancement of Bcl11b sumoylation that coincided with dephosphorylation of the protein is further support for an intact dephosphorylation-sumoylation switch in Bcl11b at least in some contexts in P2C2 cells similarly to DP thymocytes. As we saw with P/A treatment, enhancement of Bcl11b sumoylation in P2C2 cells over long-term treatment with LiCl was not accompanied by a notable increase of Bcl11b ubiquitinylation or proteosomal degradation.

Distinct from DP thymocytes, Bcl11b did not co-immunoprecipitate with

Tcf1 in P2C2 cells at a detectable level by western blotting analysis. We have previously established that an intact Tcf1-Bcl11b interaction in primary thymocytes (predominantly DP cells) is significantly and substantially decreased upon stimulation of Wnt signaling pathway in a time-dependent manner. The lack of a detectable Tcf1-Bcl11b interaction in the DN3-like cells may be attributed to the higher basal activity of Wnt signaling pathway in these cells compared with

DP thymocytes. The mechanism by which activation of Wnt/Gsk3-dependent signaling alters the Tcf1-Bcl11b interaction is unknown but may partially result from the reduced levels of shorter Tcf1 isoforms in the P2C2 cells and/or other aspect of generally elevated Wnt pathway activation in these cells. In primary thymocytes, Bcl11b preferentially interacted with short Tcf1 isoforms, and the

DN3-like cells express a higher ratio of long versus short TCf1 isoforms, potentially resulting in a very weak or non-existent Tcf1-Bcl11b interaction.

114

In summary, in addition to pre-TCR/MAPK signaling pathway, the

Wnt/Gsk3 signaling pathway is also involved in modulation of Bcl11b PTMs in

DN3-like thymocytes. Our data suggest that Bcl11b is a nuclear effector that serves as a convergence point for coupling pre-TCR-MAPK and Wnt/Gsk3 pathways in DN3-like thymocytes.

115

3.6 Figures

116

Figure 3.1. Phosphorylation kinetics of Bcl11b and changes in the activity of Erk1/2 in P2C2 cells after stimulation with P/A. P2C2 cells were treated with 100 nM phorbol 12,13 dibutyrate and 500 nM A23187 (P/A) and harvested with denaturing 1% SDS buffer at indicated times post-stimulation (min). A) After harvest,1.5% of cellular lysate was concentrated by chloroform /methanol extraction, size-separated by SDS-PAGE and processed for immunoblot detection of total Erk1/2 and phosphorylated-Erk1/2 protein with specific antibodies (anti-total Erk1/2, rabbit; anti-phospho-Erk1 (pT202/pY204)/phospho-Erk2 (pT185/pY187), mouse) followed by fluorescently- tagged secondary anti-IgG antibodies (green anti-mouse IgG, 800 nm and red anti-rabbit IgG, 680). B1) Bcl11b was immunoprecipitated from P2C2 lysates. Immunoblots were processed for detection of total Bcl11b protein (anti-Bcl11b, rat) and phosphorylated Ser /Thr (anti-Phospho S/T, mouse) followed by fluorescently-tagged secondary anti-IgG antibodies (green: anti-mouse IgG, 800 nm; red: anti-rat IgG, 680 nm) The lower panel is an overlay of upper and middle blots with yellow indicating overlapping signals. Migration of protein molecular weight standards are indicated at left (MW kDa). B2) Data from quantification of fluorescence intensity for each sample lane from 95 to 250 kDal shown in figure 1B1 is plotted as the ratio of phosphorylated to total Bcl11b as obtained by Licor Odyssey fluorescent image instrumentation and analysis software. Error bars indicate mean ± S.D. for three independent experiments, *p ≤ 0.05.

117

118

Figure 3.2. Sumoylation kinetics of Bcl11b after stimulation with P/A and inhibition of phosphatase in P2C2 cells. Cells were treated with P/A for times indicated, and samples processed as in figure 1. A1) Immunoblots were processed for detection of total Bcl11b protein (anti-Bcl11b, rat) and sumoylated lysine (anti-SUMO1, mouse), followed by fluorescently-tagged secondary anti-IgG antibodies. The lower panel is an overlay of the other panels with yellow indicating overlapping signals. A2) Data from quantification of fluorescence intensity for each sample lane from 95 to 250 kDal shown in figure 2A1 is plotted as the ratio of phosphorylated to total Bcl11b. Error bars indicate mean ± S.D. for three independent experiments, *p ≤ 0.05. B1) Phosphorylation kinetics of Bcl11b after phosphatase inhibition in P2C2 cells. P2C2 cells were pretreated with the phosphatase inhibitor 50 nM calyculin A (Cal A) or vehicle for times indicated. Cells were harvested, lysed and Bcl11b was immunoprecipitated as in Fig. 3.1. Immunoblots were processed for detection of total protein (anti-Bcl11b; rat) or phosphorylated Ser /Thr (anti-Phospho S/T; mouse) followed by fluorescently-tagged secondary anti-IgG antibodies (green: anti-mouse IgG, 800 nm; red: anti-rat IgG, 680 nm). B2) Sumoylation kinetics of Bcl11b after phosphatase inhibition in P2C2 cells. The blot in B1 was stripped and re-immunoblotted for detection of total protein (anti-Bcl11b; rat) or sumoylated lysine (anti-SUMO1; rabbit), followed by fluorescently-tagged secondary anti-IgG antibodies (red anti-rat IgG, 680 nm; green: anti-rabbit IgG, 800 nm).

119

120

Figure 3.3. Bcl11b degradation and ubiquitination in P2C2 cells with extended P/A treatment. P2C2 cells were treated P/A for times indicated up to 24 hr. Detergent lysates were immunoprecipitated with anti-Bcl11b antibodies and processed for immunoblotting as in figure 1. A1) Immunoblots were processed for detection of total Bcl11b protein (anti-Bcl11b, rat) and ubiquitinylated Bcl11b (anti- mono- and poly-ubiquitinylated conjugate monoclonal antibody, mouse) followed by fluorescently-tagged secondary anti-IgG antibodies. A2) Quantification of data shown in figure A1 by Licor Odyssey Image Analysis software. Error bars indicate +/- range for two independent experiments. B1) The blot in figure 3A was then stripped and immunoblotted with anti-Bcl11b (rat), or anti-SUMO1 (mouse) antibodies followed by detection with appropriate fluorescently-tagged secondary anti-IgG antibodies as in figure 2A1. B2) Quantification and analysis of the fluorescently visualized data shown in figure B1 by Licor Odyssey Image scanner software. Error bars indicate +/- range for two independent experiments. C) Input fractions: Detergent lysates from P/A treated P2C2 cells (1.5% of each sample prior to immunoprecipitation) were concentrated, separated by SDS-PAGE and processed for immunoblot detection of total Bcl11b (red), phosphorylated (green) and total Erk1/2 (red) as in figure 1A.

121

Figure 3.4. LiCl treatment of P2C2 cells and phosphorylation kinetics of Bcl11b, Gsk3α/β and Erk1/2 activity and accumulation of β-catenin. P2C2 cells were treated with 20 mM LiCl or vehicle for the times indicated, and cells were harvested and lysed using denaturing 1% SDS buffer as in figure 1. A1) Input fractions: Detergent lysates from LiCl treated P2C2 cells (1.5% of each sample prior to immunoprecipitation) were concentrated, separated by SDS- PAGE and processed for immunoblot detection of total Gsk3-α and β isoforms (rat

122 antibody, red) and phosphorylated Gsk3-α and β (mouse, green), β-catenin (mouse, green); Erk1/2 (rat, red), or phosphorylated Erk1/2 (mouse, green). Molecular weights for each blot are indicated at left. A2) Quantification of fluorescent intensity of total Gsk3-α/β and phosphorylated Gsk3α/β shown in figure 4A1 by Licor Odyssey fluorescent scanning and analysis software. Error bars indicate mean ± S.D. for three independent experiments, *p ≤ 0.05. B1) Immunoblots were processed for detection of total Bcl11b (anti-Bcl11b; rat) or phosphorylated Ser /Thr (anti-Phospho S/T; mouse) followed by fluorescently- tagged secondary anti-IgG antibodies (green: anti-mouse IgG, 800 nm; red: anti- rat IgG, 680 nm, lower panel overlay of upper and middle panels with yellow indicating overlapping signals). B2) Quantification of fluorescent intensity in each lane shown in figure B1 by Licor Odyssey instrumentation and analysis software. Error bars indicate mean ± S.D. for three independent experiments, *p ≤ 0.05.

123

Figure 3.5. Sumoylation (SUMO1) kinetics of Bcl11b after LiCl treatment of P2C2 cells. P2C2 cells were treated with 20 mM LiCl for the indicated times, and cells were harvested and lysed as described in figure 1 with denaturing 1 % SDS buffer plus 10 mM N-ethylmaleimide to inhibit de-sumoylation. Bcl11b was immunoprecipitated as described in methods followed by SDS-PAGE separation and processed for immunoblotting. A1) Immunoblots were processed for detection of total Bcl11b protein (Anti-Bcl11b; rat) and sumoylated lysine (Anti-SUMO1; rabbit) followed by fluorescently-tagged secondary anti-IgG antibodies (red: anti- rat IgG, 680 nm, green: anti-rabbit IgG, 800 nm,). The lower panel of three is an overlay. Migration of protein molecular weight standards are indicated at left (MW kDa). A2) Quantification of fluorescent intensity in each lane shown in figure 5A1 by Licor Odyssey instrumentation and analysis software. Error bars indicate mean ± S.D. for four independent experiments, *p ≤ 0.05.

124

Figure 3.6. Bcl11b degradation and ubiquitination in P2C2 cells following extended treatment with LiCl. P2C2 cells were treated LiCl or vehicle for times indicated. Cells were harvested, lysed and Bcl11b was immunoprecipitated as in figure.1A. A1) Immunoblots were processed for detection of total Bcl11b (anti-Bcl11b; rat) and ubiquitinylated Bcl11b (monoclonal antibody against mono-and polyubiquitinylated conjugate; mouse) followed by fluorescently-tagged secondary anti-IgG antibodies (red anti- rat IgG, 680 nm, and green anti-mouse IgG, 800 nm). A2) Quantification of fluorescent intensity in each lane shown in figure 6A1 by Licor Odyssey instrumentation and analysis software. Error bars indicate mean ± SD for four independent experiments, *p ≤ 0.05. A3) Input samples: Prior to immunoprecipitation of samples, 1.5% of the extracted cell lysates were concentrated and size separated by SDS-PAGE. Immunoblots were processed for detection of total Bcl11b (anti-Bcl11b; rat) followed by fluorescently-tagged secondary anti-IgG antibody (red anti-rat IgG, 680 nm).

125

Figure 3.7. Bcl11b and Tcf1 in P2C2 cells: P2C2 cells were treated with 20 mM LiCl or vehicle for 24 h and harvested using native lysis buffer to preserve protein-protein interactions. A) Prior to immunoprecipitation, 1.5% of each cell lysate was taken, concentrated and separated by SDS-PAGE (input). The remaining native lysates were immunoprecipitated (IP) with rabbit IgG as control or anti-Tcf1 antibody (rabbit origin), then processed for size separation by SDS-PAGE followed by immunoblotting with anti-Tcf1 (rabbit origin) and anti-Bcl11b (rat origin). Immunoreactive bands were detected by fluorescently-tagged secondary anti-IgG antibodies (green: anti-rabbit IgG, 800 nm and red: anti-rat IgG, 680 nm). Tcf1 is expressed as multiple splice variants ranging from 36 to 72 kDa, distinguished as long isoforms (45-72 kDa) and short isoforms (36-45 kDa) indicated in the boxed regions. B) An aliquot of each lysate (1.5%) was processed as input, and the remainders were immunoprecipitated with goat IgG control or anti-Bcl11b antibody (goat origin) covalently linked to agarose beads. Samples were processed for size separation followed by immunoblotting with anti-Bcl11b and anti-Tcf1 antibodies as in A.

126

3.7 References

1. Winandy, S., et al., Pre–T cell receptor (TCR) and TCR-controlled checkpoints in T cell differentiation are set by Ikaros. Journal of Experimental Medicine, 1999. 190(8): p. 1039-1048. 2. Michie, A.M. and J.C. Zúñiga-Pflücker. Regulation of thymocyte differentiation: pre-TCR signals and β-selection. in Seminars in immunology. 2002. Elsevier. 3. Shah, D.K. and J.C. Zúñiga-Pflücker, An overview of the intrathymic intricacies of T cell development. The Journal of Immunology, 2014. 192(9): p. 4017-4023. 4. Staal, F.J., et al., Wnt signaling is required for thymocyte development and activates Tcf‐ 1 mediated transcription. European journal of immunology, 2001. 31(1): p. 285-293. 5. Goux, D., et al., Cooperating pre–T-cell receptor and TCF-1–dependent signals ensure thymocyte survival. Blood, 2005. 106(5): p. 1726-1733. 6. Aifantis, I., et al., Regulation of T‐cell progenitor survival and cell‐cycle entry by the pre‐ T‐cell receptor. Immunological reviews, 2006. 209(1): p. 159-169. 7. Wiest, D.L. and M. Carleton. Control of early thymocyte development by the pre-T cell receptor complex: A receptor without a ligand? in Seminars in immunology. 1999. Elsevier. 8. Robey, E. and B. Fowlkes, Selective events in T cell development. Annual review of immunology, 1994. 12(1): p. 675-705. 9. Parham, P., The immune system. 2014: Garland Science. 10. Ng, O., et al., Deregulated WNT signaling in childhood T-cell acute lymphoblastic leukemia. Blood cancer journal, 2014. 4(3): p. e192. 11. Tatarek, J., et al., Notch1 inhibition targets the leukemia-initiating cells in a Tal1/Lmo2 mouse model of T-ALL. Blood, 2011. 118(6): p. 1579-1590. 12. Yu, S., et al., The TCF-1 and LEF-1 transcription factors have cooperative and opposing roles in T cell development and malignancy. Immunity, 2012. 37(5): p. 813-826. 13. Borowski, C., et al., Pre-TCRα and TCRα are not interchangeable partners of TCRβ during T lymphocyte development. Journal of Experimental Medicine, 2004. 199(5): p. 607-615. 14. Li, P., et al., Reprogramming of T cells to natural killer–like cells upon Bcl11b deletion. Science, 2010. 329(5987): p. 85-89. 15. Li, L., M. Leid, and E.V. Rothenberg, An early T cell lineage commitment checkpoint dependent on the transcription factor Bcl11b. Science, 2010. 329(5987): p. 89-93. 16. Karin, M. and T. Smeal, Control of transcription factors by signal transduction pathways: the beginning of the end. Trends in biochemical sciences, 1992. 17(10): p. 418-422. 17. Filtz, T.M., W.K. Vogel, and M. Leid, Regulation of transcription factor activity by interconnected post-translational modifications. Trends in pharmacological sciences, 2014. 35(2): p. 76-85. 18. Avram, D., et al., Isolation of a novel family of C2H2 zinc finger proteins implicated in transcriptional repression mediated by chicken ovalbumin upstream promoter transcription factor (COUP-TF) orphan nuclear receptors. Journal of Biological Chemistry, 2000. 275(14): p. 10315-10322. 19. Vogel, W.K., et al., Kinetic Analysis of BCL11B Multisite Phosphorylation– Dephosphorylation and Coupled Sumoylation in Primary Thymocytes by Multiple Reaction Monitoring Mass Spectroscopy. Journal of proteome research, 2014. 13(12): p. 5860-5868.

127

20. Zhang, L.-j., et al., Coordinated regulation of transcription factor Bcl11b activity in thymocytes by the mitogen-activated protein kinase (MAPK) pathways and protein sumoylation. Journal of Biological Chemistry, 2012. 287(32): p. 26971-26988. 21. Guo, A., et al., Immunoaffinity enrichment and mass spectrometry analysis of protein methylation. Molecular & Cellular Proteomics, 2014. 13(1): p. 372-387. 22. Przybylski, G., et al., Disruption of the BCL11B gene through inv (14)(q11. 2q32. 31) results in the expression of BCL11B-TRDC fusion transcripts and is associated with the absence of wild-type BCL11B transcripts in T-ALL. Leukemia, 2005. 19(2): p. 201-208. 23. Gutierrez, A., et al., The BCL11B tumor suppressor is mutated across the major molecular subtypes of T-cell acute lymphoblastic leukemia. Blood, 2011. 118(15): p. 4169-4173. 24. Bartram, I., et al., Low expression of T-cell transcription factor BCL11b predicts inferior survival in adult standard risk T-cell acute lymphoblastic leukemia patients. Journal of hematology & oncology, 2014. 7(1): p. 51. 25. De Keersmaecker, K., et al., The TLX1 oncogene drives aneuploidy in T cell transformation. Nature medicine, 2010. 16(11): p. 1321-1327. 26. Liu, P., P. Li, and S. Burke, Critical roles of Bcl11b in T‐cell development and maintenance of T‐cell identity. Immunological reviews, 2010. 238(1): p. 138-149. 27. Ikawa, T., et al., An essential developmental checkpoint for production of the T cell lineage. Science, 2010. 329(5987): p. 93-96. 28. Wakabayashi, Y., et al., Bcl11b is required for differentiation and survival of αβ T lymphocytes. Nature immunology, 2003. 4(6): p. 533-539. 29. Kastner, P., et al., Bcl11b represses a mature T‐cell gene expression program in immature CD4+ CD8+ thymocytes. European journal of immunology, 2010. 40(8): p. 2143-2154. 30. Albu, D.I., et al., BCL11B is required for positive selection and survival of double-positive thymocytes. Journal of Experimental Medicine, 2007. 204(12): p. 3003-3015. 31. Michie, A.M., et al., Extracellular Signal–regulated Kinase (Erk) Activation by the Pre-T Cell Receptor in Developing Thymocytes In Vivo. Journal of Experimental Medicine, 1999. 190(11): p. 1647-1656. 32. Bommhardt, U., et al., MEK activity regulates negative selection of immature CD4+ CD8+ thymocytes. The Journal of Immunology, 2000. 164(5): p. 2326-2337. 33. Cargnello, M. and P.P. Roux, Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiology and molecular biology reviews, 2011. 75(1): p. 50-83. 34. Staal, F.J., T.C. Luis, and M.M. Tiemessen, WNT signalling in the immune system: WNT is spreading its wings. Nature reviews. Immunology, 2008. 8(8): p. 581. 35. Engel, I., et al., Early thymocyte development is regulated by modulation of E2A protein activity. Journal of Experimental Medicine, 2001. 194(6): p. 733-746. 36. Smith, C.A., et al., Antibodies to CD3/T-cell receptor complex induce death by apoptosis in immature T cells in thymic cultures. Nature, 1989. 337(6203): p. 181-184. 37. Li, L., et al., A far downstream enhancer for murine Bcl11b controls its T-cell specific expression. Blood, 2013. 122(6): p. 902-911. 38. Takahama, Y. and H. Nakauchi, Phorbol ester and calcium ionophore can replace TCR signals that induce positive selection of CD4 T cells. The Journal of Immunology, 1996. 157(4): p. 1508-1513.

128

39. Wessel, D. and U. Flügge, A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Analytical biochemistry, 1984. 138(1): p. 141-143. 40. Zhang, L., Deciphering the molecular mechanisms of the transcriptional regulation mediated by orphan nuclear receptor COUP-TFI and C2H2 zinc finger protein Ctip2. 2009: Oregon State University. 41. Cohen, P. and S. Frame, The renaissance of GSK3. Nature reviews Molecular cell biology, 2001. 2(10): p. 769. 42. van de Wetering, M., et al., The human T cell transcription factor-1 gene. Structure, localization, and promoter characterization. Journal of Biological Chemistry, 1992. 267(12): p. 8530-8536. 43. Tiemessen, M.M., et al., The nuclear effector of Wnt-signaling, Tcf1, functions as a T- cell–specific tumor suppressor for development of lymphomas. PLoS biology, 2012. 10(11): p. e1001430. 44. Van de Wetering, M., et al., Extensive alternative splicing and dual promoter usage generate Tcf-1 protein isoforms with differential transcription control properties. Molecular and cellular biology, 1996. 16(3): p. 745-752. 45. Lovatt, M. and M.-J. Bijlmakers, Stabilisation of β-catenin downstream of T cell receptor signalling. PLoS One, 2010. 5(9): p. e12794. 46. López-Rodríguez, C., J. Aramburu, and R. Berga-Bolaños, Transcription factors and target genes of pre-TCR signaling. Cellular and molecular life sciences, 2015. 72(12): p. 2305- 2321. 47. Tootle, T.L. and I. Rebay, Post‐translational modifications influence transcription factor activity: a view from the ETS superfamily. Bioessays, 2005. 27(3): p. 285-298. 48. Huang, X., et al., Gene expression profiles in BCL11B-siRNA treated malignant T cells. Journal of hematology & oncology, 2011. 4(1): p. 23. 49. Cismasiu, V.B., et al., BCL11B functionally associates with the NuRD complex in T lymphocytes to repress targeted promoter. Oncogene, 2005. 24(45): p. 6753. 50. Haks, M.C., et al., Pre-TCR signaling and inactivation of p53 induces crucial cell survival pathways in pre-T cells. Immunity, 1999. 11(1): p. 91-101. 51. Ceredig, R. and T. Rolink, A positive look at double-negative thymocytes. Nature Reviews Immunology, 2002. 2(11): p. 888. 52. Carleton, M., et al., Signals transduced by CD3ε, but not by surface pre-TCR complexes, are able to induce maturation of an early thymic lymphoma in vitro. The Journal of Immunology, 1999. 163(5): p. 2576-2585. 53. Xu, Y., et al., Deletion of β-catenin impairs T cell development. Nature immunology, 2003. 4(12): p. 1177. 54. Xu, M., et al., Pre-TCR-induced β-catenin facilitates traversal through β-selection. The Journal of Immunology, 2009. 182(2): p. 751-758. 55. Weerkamp, F., et al., Wnt signaling in the thymus is regulated by differential expression of intracellular signaling molecules. Proceedings of the National Academy of Sciences of the United States of America, 2006. 103(9): p. 3322-3326.

129

Chapter 4

General Conclusion

130

SSTFs play essential roles in regulation of gene expression programs which are, in turn, critically important for development of all tissues, including thymocytes. In response to multiple signaling inputs, SSTF activities are highly and precisely regulated by diverse means such as alterations in PTMs and/or protein-protein interactions. In these studies, we have identified a previously unknown, dynamically regulated interaction between Bcl11b and Tcf1, two requisite T cell lineage-determining SSTFs [1]. Our studies characterized the key

PTMs of Bcl11b in DN thymocytes for the first time and determined Bcl11b to be a novel target of Wnt/GSK3 signaling pathway in two distinct stages of thymocytes development (DN and DP thymocytes). Moreover, our studies indicate convergence of both pre-TCR and/or TCR/MAP kinase and Wnt/GSK3 pathways on the regulation of Bcl11b in thymocytes.

In DP thymocytes, we identified a novel interaction between Bcl11b and

Tcf1, the classic nuclear effector of Wnt/GSK3 signaling pathway, and this interaction was dynamically regulated upon stimulation of the signaling pathway using LiCl. Moreover, we provided evidence that in addition to TCF1, BCL11B is also able to simultaneously participate in a complex (or complexes) with β- catenin. A dynamically-regulated, physical interaction between Bcl11b-Tcf1 has not previously been identified in any cell type nor has been the association between BCL11B-β-catenin previously reported. Our finding in DP thymocytes that the Bcl11b-Tcf1 interaction is decreased upon stimulation of the canonical

Wnt signaling pathway and our data in P2C2 cells indicating a higher basal

131 activity of Wnt signaling in these cells compared to DP thymocytes prompted us to hypothesize that the Bcl11b-Tcf1 association is affected by elevated basal Wnt signaling in P2C2 cells. In agreement, we were not able to detect a Bcl11b-Tcf1 interaction in DN3-like thymocytes by western blot assay after co- immunoprecipitation. However, β-catenin levels are also elevated in unstimulated

Jurkat cells as a result of transformational mutations, and the BCL11B-TCF1 interaction, along with a BCL11B-β-catenin interaction is intact in these cells.

Thus, we cannot fully explain the lack of TCF1-BCL11B interaction in P2C2 cells solely as a result of elevated β-catenin levels. Wnt signaling pathways components, including levels of endogenous inhibitors of Wnt signaling, differ from cell type to cell type and the full array of relevant players in the different cell lines, and their impact on a TCF1-BCL11B interaction and subsequent gene regulation remains to be fully explored.

To study the role of sumoylation of BCL11B in mediating its interaction with

TCF1 and/or β-catenin, we utilized CRISPR/Cas9-mediated genome editing to generate two mutant forms of BCL11B which are lacking one or both SUMO- acceptor sites. We found that neither sumoylation nor deletion of up to 285 amino acids at the C-terminus of the BCL11B are required for the protein to complex with

TCF1 and/or β-catenin in Jurkat cells. Identification of a Bcl11b -Tcf1 interaction in native thymocytes and a BCL11B- β-catenin complex in Jurkat cells opens new

132 avenues to investigate the functional relevance of these complexes in these cells and other cell types with possible implications in cancer biology.

Analysis of available ChIP-seq data sets revealed that Bcl11b and Tcf1 co- occupy the promoters of many genes that are involved in regulation of thymocyte proliferation, differentiation and survival. We used this information to design a reporter gene assay to study the interplay between TCF1 and BCL11B on a common promoter. Individually, knockout of Tcf1 [2] and Bcl11b [3] in thymocytes have been shown to decrease BCL-xL levels, suggesting that both activate the

Bcl2l1 gene. Results from reporter gene assay studies confirmed that both SSTFs individually activate Bcl2l1 expression, although the magnitude of increase is greater with BCL11B than TCF1 in unstimulated cells. In contrast, when co- expressed, the two SSTFs behaved in a mutually antagonistic manner to regulate the Bcl2l1 gene promoter. Thus, our studies provide the first evidence that a

BCL11B-TCF1 interaction may affect BCL-xL expression, and suggest that a

BCL11B-TCF1 complex may integrate signals from different pathways to regulate cell survival in the developmental selection steps that allow for proliferation of functional but non-self-reacting T cells, for example. Next steps will be to develop an approach that allows for the gene reporter assays to be repeated in the presence of Wnt pathway stimulation. Unfortunately, 20 mM LiCl interferes with luciferase activity and is not a suitable activator in this system.

Assays for composite phosphorylation and sumoylation revealed that

Bcl11b is a target of the canonical Wnt/GSK3 pathway in DP thymocytes in

133 addition to its previously established position as a nuclear target of the TCR/MAP kinase [4] pathway (See Table 4.1). Activation of Wnt signaling coincided with sustained dephosphorylation and increased sumoylation of Bcl11b up to 24 h treatment, with a maximum at 2 h post-stimulation. Additionally, distinct from stimulation of MAP kinase previously established in DP thymocytes, sustained sumoylation of Bcl11b induced by LiCl was not associated with dramatic changes in Bcl11b ubiquitinylation and stability.

The PTMs which regulate Bcl11b in DN thymocytes are unknown. To address Bcl11b PTMs in these cells, we adopted an in vitro model, P2C2 cells, which are thymocytes arrested at the DN3 stage. The composite PTMs of Bcl11b after stimulation of pre-TCR /MAP kinase pathway in DN3-like thymocytes are similar to those previously described in DP thymocytes [4]. Specifically, phosphorylation of Bcl11b is rapidly increased after stimulation of MAP kinase pathway by P/A and then dephosphorylated (maximum at 1 h post-stimulation) before a return to basal phosphorylation levels after 2 h. Unlike DP thymocytes, de-sumoylation of Bcl11b (maximum at 1-2 h) induced after stimulation of pre-

TCR/MAP kinase pathway in P2C2 cells did not coincide with increased phosphorylation of serine or threonine residues in Bcl11b. However, hyperphosphorylation of Bcl11b induced by a phosphatase inhibitor obliterates sumoylation of Bcl11b in DN3-like thymocytes, similar to the effect of hyperphosphorylation on Bcl11b in DP thymocytes.

134

Our studies reveal that stimulation of Wnt/GSK3 in P2C2 cells using LiCl also resulted in significant changes in Bcl11b PTMs indicated by sustained dephosphorylation and increased sumoylation of the protein up to 24 h of treatment with maximum changes at 8 h after stimulation. Similar to DP thymocytes, ubiquitinylation of Bcl11b, which usually targets the protein for proteasomal degradation, did not change in P2C2 after 24 h stimulation by LiCl and did not appear to affect protein stability.

The effects of both MAPK and Wnt/Gsk3 pathways on Bcl11b PTMs in

DN3-like P2C2 cells and primary (mostly DP) thymocytes are summarized in

Table 4.1.

Table 4.1: Treatment effects on Bcl11b PTMs.

Treatment System Phosphorylation Sumoylation Ubiquitinylation P/A [4] Primary Transient increase at Transient de- Maximum at 6 h thymocytes 7 min, maximum desumoylation at 7 followed by protein phosphorylation at 30 min, re-sumoylation degradation at 8-24 min, return to basal within 30 min with a h. level at 2 h. maximum at 1h.

P/A P2C2 cells Transient increase at De-sumoylation No notable change 7 min, maximum de- maximum at 1-2 h, in protein phosphorylation at 60 return to basal level ubiquitination or min, return to basal at 4 h and stability to 24 h. level at 2 h. sustained up to 24 h. LiCl Primary De-phosphorylation Sumoylation No change in thymocytes maximum at 2 h, increase maximum ubiquitinylation or sustained to 24 h. at 2 h, sustained to stability to 24 h. 24 h. LiCl P2C2 cells De-phosphorylation Sumoylation No change in maximum at 8 h, increase maximum ubiquitinylation or sustained to 24 h. at 8 h, sustained to stability to 24 h. 24 h.

135

There are only small differences between the cell types in terms of treatment effects on Bcl11b PTMs summarized in Table 4.1, and the differences are mostly in terms of timing, with events delayed a bit in the P2C2 cells relative to primary thymocytes. Other differences between the cell types that are most striking are related to intracellular signaling events. For example, Erk1/2 activation induced by P/A was constant up to 8 h in DN3 thymocytes, whereas it increased and then decreased over 1h in DP thymocytes in response to same stimulant [4].

There is also a higher basal level of β-catenin in DN3 cells compared to DP thymocytes, implying a basal activation of Wnt pathway or perhaps a basal inhibition of Gsk3 in these cells.

Differences in Bcl11b PTMs likely impose specific functions at specific stages of thymocytes development. Sumoylation, which appears to be regulated by phosphorylation of Bcl11b in primary thymocytes, de-represses BCL11B activity at the Id2 gene in transfected HEK293T cell system [4]. Differences in

Bcl11b PTMs across distinct developmental stages may regulate differential expression of target genes critical for appropriate developmental progression.

BCL11B, TCF1, and GATA3 are considered the three T cell lineage- specifying SSTFs and act in combinatorial fashion to guide stepwise development of thymocytes into T cells [5] . Herein, we demonstrated that Bcl11b interacts with

Tcf1 in thymocytes. Others have recently identified a Bcl11b interaction with

GATA3 in CD4+ T helper cells [6]. Identifying the full and dynamic interactome of

BCL11B through distinct stages of thymocyte development and under various

136 conditions will help to decipher the molecular mechanisms that regulate normal thymocyte development and pathophysiology of transformation.

Our studies pave the way for future investigations to understand the functional relevance of the BCL11B-TCF1 interaction in transcriptional regulation of many other genes that are required for thymocyte development and how PTMs of BCL11B and its interactome might change in disease states. The potential to pharmacologically alter SSTF functions through manipulation of their PTMs and/or protein-protein interactions may provide a powerful means to undermine pathophysiological processes and restore a healthy cell state in many diseases including, but not limited to, cancer. One example of this approach is pharmacological inhibition of the interaction between P53, a tumor suppressor transcription factor that is mutated (deactivated) in many cancers [7], and MDM2, an E3 ubiquitin ligase that ubiquitinates P53 and targets it for degradation [8].

Application of a drug named nutlin-3a leads to reactivation of P53 by preventing its association with MDM2 and ubiquitinylation. In KSH-V-positive lymphoma, nutlin-3a induces cell death [9]. This and other encouraging examples suggest that targeting of BCL11B through its protein-protein interactions and PTMs may be a novel target for treatment of human diseases like leukemia.

4.2 References

1. Rothenberg, E.V., J. Ungerbäck, and A. Champhekar, Chapter Four-Forging T-Lymphocyte Identity: Intersecting Networks of Transcriptional Control. Advances in immunology, 2016. 129: p. 109-174. 2. Wang, R., et al., T cell factor 1 regulates thymocyte survival via a RORγt-dependent pathway. The Journal of Immunology, 2011. 187(11): p. 5964-5973.

137

3. Wakabayashi, Y., et al., Bcl11b is required for differentiation and survival of [alpha][beta] T lymphocytes. Nature immunology, 2003. 4(6): p. 533. 4. Zhang, L.-j., et al., Coordinated regulation of transcription factor Bcl11b activity in thymocytes by the mitogen-activated protein kinase (MAPK) pathways and protein sumoylation. Journal of Biological Chemistry, 2012. 287(32): p. 26971-26988. 5. Rothenberg, E.V., J. Ungerbäck, and A. Champhekar, Forging T-lymphocyte identity: Intersecting networks of transcriptional control, in Advances in immunology. 2016, Elsevier. p. 109-174. 6. Fang, D., et al., Bcl11b, a novel GATA3-interacting protein, suppresses Th1 while limiting Th2 cell differentiation. Journal of Experimental Medicine, 2018: p. jem. 20171127. 7. Vogelstein, B., D. Lane, and A.J. Levine, Surfing the p53 network. Nature, 2000. 408(6810): p. 307. 8. Honda, R., H. Tanaka, and H. Yasuda, Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor p53. FEBS letters, 1997. 420(1): p. 25-27. 9. Sarek, G., et al., Reactivation of the p53 pathway as a treatment modality for KSHV- induced lymphomas. The Journal of clinical investigation, 2007. 117(4): p. 1019-1028.