• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

The role of STAT5 in early B cell development and leukemia

Dissertation
Author: Mark Joseph Lueth Willette
Abstract:
Acute lymphoblastic leukemia (ALL) is one of the most common forms of cancer in children, yet the mechanisms driving its development remain incompletely defined. Due to the prevalence of ALL in children and its poor prognosis in adults, furthering our understanding of the underlying factors driving this disease is important for controlling a significant source of morbidity and mortality in patients. The purpose of the experiments described in this thesis was to describe the interplay between important B cell development factors in driving B cell ALL. Our lab previously reported that expression of a constitutively active form of STAT5 (STAT5b-CA) during B cell development leads to a 1-2% incidence of ALL in mice. In order to describe the mechanism driving leukemogenesis, the role of STAT5 in ALL was investigated. Elevated STAT5 activity was found in human ALL and predicted treatment response in high-risk patients. In mice, STAT5b-CA-driven ALL was found to lack expression of pre-BCR signaling factors and STAT5 cooperated with a loss of pre-BCR signaling factors to drive pre-B ALL. This cooperation was due to effects of STAT5 other than up-regulation of survival signals. Instead, STAT5 was shown to repress a set of NF-κB target genes, and these genes were further down-regulated in STAT5b-CA-driven leukemia when pre-BCR signaling was also impaired. These findings provide a point of cross-talk between STAT5 and pre-BCR signaling in B cell development. Furthermore, loss of NF-κB cooperated with constitutively active STAT5 to induce leukemia, implicating for the first time NF-κB as a tumor suppressor in ALL. Thus, herein is described a novel model of pre-B ALL which mimics disease in humans and provides insight into the mechanisms regulating both normal and aberrant B lymphopoiesis.

vi Table of Contents

Acknowledgements i Dedication iii Abstract iv Table of Contents vi List of Tables vii List of Figures viii List of Abbreviations x Author Contribution xii Chapter I: Introduction 1 Chapter II: Constitutively active STAT5 promotes development of pre-B cell 28 acute lymphoblastic leukemia

Chapter III: Constitutively active STAT5 cooperates with loss of pre-BCR 52 signaling to promote ALL

Chapter IV: An NF-κB pathway suppresses STAT5-dependent pre-B cell 96 leukemia

Chapter V: Summary and Model 115

References 124

vii List of Tables

Table Title Page

Table 3.1 Top aberrantly expressed genes in leukemic mice. 78 Table 4.1 Genes regulated by antagonistic cross-talk between 107 STAT5b-CA and NF-κB.

viii List of Figures

Figure Title Page

1.1 B cell development in the mouse 3 1.2 Activation of STAT5 by IL7R signaling 11 1.3 Pre-BCR signaling 16 2.1 Spontaneous tumors in Stat5b-CA mice 35 2.2 Genetic analysis of Stat5b-CA leukemia 37 2.3 Cyclin D2 and Myc expression in Stat5b-CA ALL 40 2.4 Total STAT5 protein in human ALL patients 43 2.5 Level of phosphorylated STAT5 inversely correlates 45 with imatanib response in ALL patients

2.6 Distinct STAT5 phosphorylation in different cytogenetic 47 subsets of ALL

3.1 Loss of pre-BCR signaling factors cooperates with 60 STAT5b-CA to induce pre-B ALL

3.2 Pre-BCR signaling-deficient leukemic mice are 63 characterized by high numbers of lymphocytes in the blood and enlarged lymphoid organs

3.3 Cell surface expression of phenotypic markers on ALL 65 cells

3.4 Cytogenetic analysis of STAT5b-CA-driven ALL 67

3.5 Loss of Ebf1 or Pax5 cooperates with STAT5b-CA to 71 induce B ALL.

3.6 Cd79a expression in ALL subsets 73

ix Figure Title Page

3.7 Microarray analysis of tumor samples from indicated mice 76

3.8 Canonical pathways differentially regulated in tumor 81 samples

3.9 Expression of TNF/TNFR superfamily genes 85

3.10 STAT5b-CA does not induce ALL solely by increasing 88 proliferation or survival of pre-B cells

4.1 NF-κB1 acts as a tumor suppressor to prevent STAT5b- 102 CA-driven ALL development

4.2 Proportion of NF-κB1 target genes that show synergistic 105 deregulation between STAT5b-CA and Blnk or PKCβ deficiency

4.3 Ikzf3 expression is inhibited by STAT5 110

5.1 Model of STAT5 and pre-BCR signaling cross-talk in 119 pre-B ALL

x List of Abbreviations

ALL Acute lymphoblastic leukemia ASNase Asparaginase ASNS Asparagine synthetase B ALL B-lineage acute lymphoblastic leukemia Blnk B cell linker BM Bone marrow BrdU Bromodeoxyuridine Btk Bruton’s tyrosine kinase CCND2 Cyclin D2 CLP Common lymphoid progenitor EBF Early B cell factor ELP Early lymphoid progenitor Eµ Immunoglobulin µ enhancer FISH Fluorescent in situ hybridization FOXO Forkhead box family O GAS γ interferon activated sequence HAT Histone acetyl transferase HDAC Histone deacetylase HSC Hematopoietic stem cell IgH Immunoglobulin heavy chain Ikzf1/3 Ikaros family zinc finger protein 1/3 IL7 Interleukin-7 IL7R Interleukin-7 receptor IRF4/8 Interferon regulatory factor 4/8 ITAM Immunoreceptor tyrosine-based activation motif Jak Janus kinase LC Immunoglobulin light chain LN Lymph node MPP Multipotent progenitor NF-κB Nuclear factor κ light chain enhancer of activated B cells Pax5 Paired box protein 5 PKCβ Protein kinase C β pre-BCR pre-B cell receptor pSTAT5 phosphorylated STAT5 RAG1/2 Recombination activating gene 1/2 SH2 Src-homology 2 SKY Spectral karyotyping SLC Surrogate light chain STAT5 Signal transducer and activator of transcription 5 STAT5b-CA Constitutively active STAT5b TNFRSF Tumor necrosis factor receptor superfamily VDJ Immunoglobulin variable, distal, and joining regions

xi XID X-linked immunodeficiency γc Common γ chain of the interleukin-2 receptor µHC Immunoglobulin µ heavy chain

xii Author Contribution

The studies described in this thesis were performed by myself with significant assistance from Lynn Heltemes Harris and Laura Ramsey, except where noted as follows. Mouse analysis, tissue isolation, and survival curves were completed by myself, while Lynn assisted with much of the flow cytometric analysis. Control samples for microarray were procured by Laura Ramsey with help from Nisha Shah and the Flow Cytometry Facility at the University of Minnesota, and Laura performed the original microarray analyses in Chapters II and III along with Thearith Koeuth. Analyses of microarray results for NF-κB target gene expression in Chapter IV were done by myself. Reverse-phase protein analysis was done by Steven Kornblau, E. Shannon Neeley, Nianxiang Zhang, and Yi Qiu. Cytogenetic analysis was performed by LeAnn Oseth in the University of Minnesota Cytogenetics Core. Michael Farrar oversaw and directed this project. This thesis was written by myself; Chapters II and III are modified from an unpublished manuscript prepared with assistance from Lynn Heltemes Harris, Laura Ramsey, and Michael Farrar.

1

CHAPTER I

Introduction

2 B cell Development

The mammalian immune system consists of a complex integration of cellular and molecular factors that cooperate to protect the organism against invading foreign pathogens. One of the key cell types involved in protection from pathogens is the B cell. These antibody-producing lymphocytes are an integral part of the humoral immune response, but their abnormal growth and activity can lead to autoimmunity and cancer. Because of this, stringent developmental regulation is important. B cell development occurs in the bone marrow (BM) where progenitor cells progress through phenotypically distinct stages until they become immature B cells and are released into the periphery [1, 2]. These developmental stages are characterized and identified by variable expression of cell surface markers and transcription factors and rely on specific signals for proper development (Figure 1). Interruptions in these signals result in blocks at associated developmental stages. This system acts as a series of checkpoints that allow only cells with normal signaling function to be released into the periphery. B cell development begins with multipotent hematopoietic stem cells (HSC) which give rise to multipotent progenitor (MPP) cells. Both HSCs and MPPs express the surface markers c-Kit and Sca-1, but are considered lineage negative, meaning they lack markers specific for various hematopoietic subsets. MPPs, however, lose their self-renewal capacity. Those that express Flt3 and begin transcribing the lymphocyte-specific recombination activating genes 1 and 2 (rag1/2) are called early lymphoid progenitors (ELP) [3]. ELPs that begin expressing the interleukin-7 receptor (IL7R) are designated as common lymphoid progenitors (CLPs), which have

3

Figure 1

HSC MPP CLP pre- pro-B pro-B Large pre-B Small pre-B Immature B Flt3 c-Kit Sca-1 Rag1/2 IL-7R B220 CD19 CD43 IgM pre-BCR BCR D H -J H V H -DJ H il7r -/- Stat5 -/- blnk -/- btk -/- pkc -/- pax5 -/- ebf -/- V L -J L

4 Figure 1. B cell development in the mouse. B cell progenitors progress through distinct developmental stages within the bone marrow. Below the cells are listed phenotypic markers and the stages at which they are expressed. Yellow bars represent stages at which the associated knock-out mice have developmental blocks. HSC, hematopoietic stem cell; MPP, multipotent progenitor; CLP, common lymphoid progenitor.

5 the potential to develop into any of the lymphoid-derived cell types, but lack myeloid lineage potential [4]. The point at which a lymphocyte progenitor first acquires B- lineage markers is at the transition from CLP to pre-pro-B cell, where the cell begins expressing the B cell-specific surface marker B220; however, these cells still retain some non-B lineage potential. The next stage – the pro-B cell – is when the cells first become restricted to the B lineage. Pro-B cells are further subdivided into early and late pro-B stages, where the cells begin expressing CD19 and BP-1, respectively. During the pro-B stage, cells begin rearranging the immunoglobulin heavy chain (Igh) locus by expressing RAG1/2. Following successful rearrangement of Igh, a pre- B cell receptor (pre-BCR) is expressed on the cell and thus defines the pre-B cell. If the pre-BCR is able to signal properly, the cell undergoes several rounds of proliferation, followed by down-regulation of the pre-BCR, rearrangement of the immunoglobulin light chain (Igl) locus, and eventual expression of a B cell receptor (BCR). This expression of a fully formed BCR, or surface IgM, characterizes the immature B cell. If a proper, non-self-reactive BCR is expressed, the immature B cell then migrates out of the BM into the peripheral lymphoid organs.

Control of B cell development The first events that occur during B cell differentiation include the expression of the transcription factors Ikaros and PU.1. Together, these two appear to promote B-lineage differentiation through the expression of the IL7R and Flt3. IL7R signaling is absolutely necessary for B lymphopoiesis. This requirement was first demonstrated by use of anti-IL7 antibodies and later in IL7R α chain (IL7Rα)-deficient mice [5-8].

6 In these studies, lack of IL7R signaling resulted in a severe deficiency of B cells with a block occurring at the pre-pro-B cell stage. IL7R signaling at the pre-pro-B stage occurs via activation of signal transducer and activator of transcription 5 (STAT5; IL7R signaling through STAT5 is discussed in detail in the next section), which controls expression of early B cell factor (EBF), important for driving cells down the B cell lineage. Not surprisingly, ebf-deficient mice lack pre-pro-B cells, similar to Il7r -/- mice [9]. EBF controls expression of multiple genes critical for B cell development, including pax5, cd19, and mb-1. While surface expression of CD19 first characterizes the pro-B stage and continues throughout B cell development, Pax5 acts as a B lineage “commitment” factor by inhibiting multilineage potential through inhibition of non-B cell genes. Busslinger and colleagues demonstrated this by showing that pax5 -/- pro-B cells readily differentiated into various non-lymphoid progenitors. Pax5 -/- pro-B cells also efficiently transcribed genes encoding macrophage colony stimulating factor receptor (MCSFR) and myeloperoxidase (MPO) (myeloid), GATA-1 (erythroid), perforin (NK cell) and pTa (T lymphoid); moreover, transcription of these genes was inhibited following retroviral transduction of wild type pax5 [10]. Although Pax5 restricts cells to the B lineage, it is also important for promoting further differentiation. For example, Pax5 regulates expression of multiple critical B cell-specific factors, including CD19, and pax5 -/-

mice have a block from the early to late pro-B stage [11]. In addition, expression of the pre-BCR that characterizes pre-B cells requires rearrangement of the variable, distal, and joining (VDJ) regions of the Igh locus. Pax5 has been shown to play a role in V-DJ rearrangement by controlling Igh locus contraction [12]; additionally, Pax5

7 regulates expression of λ5, an integral part of the pre-BCR complex [13]. Thus, Ebf and Pax5 are essential for differentiation from a CLP to a pre-B cell. Following IgH rearrangement, a pre-BCR is expressed on the cell surface. Functional signaling through the pre-BCR results in a brief proliferative burst followed by immunoglobulin light chain (IgL) rearrangement to form a complete BCR with the IgH, and progression to the immature B cell stage. Pre-BCR signaling is described in detail below.

STAT5 The STAT proteins are a family of transcription factors that mediate cytokine signaling. Structurally, they consist of N-terminal, coiled-coil, DNA binding, linker, Src-homolgy 2 (SH2), transcriptional activation, and C-terminal domains. These domains allow STATs to be recruited to phosphorylated tyrosines on cytokine receptors (via the SH2 domain), become activated, bind to DNA target sequences (via the DNA binding domain), and control gene expression (via the transcriptional activation domain). One of the STAT proteins, STAT5, is involved in B cell development. It was first identified in 1994 by Groner and colleagues as a transcription factor expressed in mammary glands that regulated prolactin-induced transcription. Because of this role and apparent tissue specificity, it was named mammary gland factor (MGF), although the name was changed to STAT5 after determining that it shared homology with other known STAT family members [14]. Not long after this identification, three groups independently demonstrated that additional cytokine receptors signal via two

8 distinct STAT5 isoforms, called STAT5a and STAT5b [15-17]. Because of its identification in mammary cells, and its responsiveness to prolactin, the original studies examining the importance of STAT5 used mice deficient for either STAT5a or STAT5b, and focused mainly on mammary gland lactation and sexual development. Stat5a -/- and Stat5b -/- mice have distinct phenotypes, with STAT5a- deficiency resulting in defective mammary tissue development, and STAT5b- deficiency having growth and sexual development problems [18, 19]. Because of this, it was believed that the two isoforms had distinct roles in development. However, STAT5 was shown to be activated by numerous cytokines involved in lymphocyte development and survival, including IL7 [20], so studies by Warren Leonard’s group focused on effects of STAT5a deficiency on immune cells [21]. They found that Stat5a -/- had no decrease in thymic cellularity, and only a very modest decrease in total splenocytes. The overall ratio of B and T cells appeared normal [21]. Because of the close homology of STAT5a and STAT5b, this modest effect was thought to be due to redundant roles for the two isoforms in lymphocyte development. This possibility was addressed by Ilhe and colleagues by targeted disruption of both Stat5a and Stat5b genes. Their approach targeted the first coding exon in order to produce a protein-null phenotype (Stat5a/b Δ N/ Δ N ) [22]. This approach provided evidence that the two STAT5 isoforms do serve redundant functions in multiple tissues, and doubly-deficient mice have severe defects in these tissues, but the overall effects on B cell development were subtle. Stat5a/b Δ N/ Δ N mice had a decreased percentage of leukocytes but increased neutrophils, and bone marrow progenitors had an impaired ability to develop into pre-B cells in a colony formation

9 assay, while Stat5a Δ N/ Δ N and Stat5b Δ N/ Δ N mice did not, but no defects were detected in peripheral B cells [22]. In contrast, further characterization of these mice by Sexl et al. revealed a decrease in pro- and pre-B cells in the bone marrow, but no change in B cells in the spleen and lymph nodes [23]. It was already known that Il7r -/- mice have a complete block in B lymphopoiesis at the pro-B cell stage [6], so the finding that Stat5a/b Δ N/ Δ N mice have only subtle B cell defects suggested that STAT5-independent IL7R signaling could also promote B cell development. However, our own findings showed that constitutive STAT5b activation was sufficient for rescuing the B cell defect seen in Il7r -/- mice [24]. The ability of STAT5 to correct this severe defect was difficult to reconcile with the notion that loss of both STAT5a and STAT5b has only subtle effects on B lymphopoiesis. About a year later, studies by Hennighausen and colleagues addressed the role of STAT5 in conditional mammary-specific deletions. Using a knockout approach where the entire span of the closely linked STAT5a and STAT5b genes was floxed, followed by Cre-mediated deletion (referred to here as Stat5a/b -/- ), they found that complete loss of STAT5 during embryogenesis resulted in nearly complete embryonic lethality, suggesting that there is a functional difference between Stat5a/b Δ N/ Δ N and Stat5a/b -/- [25]. Indeed, they found that Stat5a/b Δ N/ Δ N mice produce an N-terminally truncated protein which could still be phosphorylated by cytokine stimulation. Further analysis of B lymphopoiesis using the Stat5a/b -/- mice by separate groups demonstrated much stronger phenotype closely mimicking Il7r -/-

mice with absent B cells and a complete developmental block at the pre-pro- to pro-B transition [26, 27]. These studies demonstrated that STAT5a and STAT5b play

10 redundant roles in B lymphopoiesis, but expression of at least one isoform is essential.

Activation of STAT5 in developing B cells occurs via IL7R signaling (Figure 2). The IL7R exists as a heterodimer of the IL7Rα chain and the IL2R common gamma chain (γc). Binding of the receptor with its ligand IL7 brings the two chains in close proximity. Janus kinases 1 and 3 (Jak1/3) associated with the cytoplasmic tail of both chains (IL7Rα and γc, respectively) are also brought together, allowing the two kinases to phosphorylate tyrosine residues within the IL7Rα chain. This phosphorylation creates a docking site for STAT5 molecules via the latter’s SH2 domain. Once recruited to the IL7Rα, STAT5 is phosphorylated by Jak1 and Jak3 on tyrosine residue 694 or 699 of STAT5a and STAT5b, respectively. Phosphorylated STAT5 (pSTAT5) then dissociates from the IL7Rα chain and forms dimers in which the SH2 domain of one molecule binds the phosphorylated tyrosine of the other, and vice versa. The dimerized STAT5 then translocates to the nucleus where it regulates a variety of target genes, via direct binding to γ interferon activated sequences (GAS) characterized by the TTCxxxGAA motif. As a transcription factor, STAT5 relies on histone acetyltransferase (HAT) co-activators to regulate transcription of target genes. STAT5 interaction with HATs such as p300 and CBP has been demonstrated by Pfitzner et al. to induce transcription of numerous targets via interaction through STAT5’s transactivation domain [28]. Likewise, STAT5 interaction with the co- repressor SMRT inhibits transcription of target genes, although this interaction occurs via the C-terminal coiled-coil domain, in contrast to HATs [29]. SMRT recruits histone deacetylases (HDACs), which inhibit transcription by preventing chromatin

11

Figure 2

Y449 Jak3 Jak1 P Il-7 Il-7R C Y449 Jak3 Jak1 P Il-7 Il-7R C STAT5 P Y449 Jak3 Jak1 P Il-7 Il-7R C STAT5 P STAT5 P a b c

12 Figure 2. Activation of STAT5 by IL7R signaling. Schematic of the IL7R signaling cascade. (a) The IL7R, consisting of the IL7Rα chain and the IL2R γc chain, binds its ligand, IL7. The associated Jaks cross- phosphorylate each other as well as tyrosine 449 (Y449) on the IL7Rα chain. (b) STAT5 binds to phosphorylated Y449 via its SH2 domain, bringing it in proximity to the Jaks, which can then phosphorylate STAT5. (c) Phosphorylated STAT5 dissociates from the IL7Rα chain, and forms a homodimer via the SH2 domain of another STAT5 molecule. STAT5 is now able to translocate to the nucleus and regulate transcription of target genes. IL-7, interleukin-7; IL-7R α, interleukin-7 receptor α chain; γC, interleukin-2 receptor common γ chain; P, phosphate group.

13 accessibility; intriguingly, however, HDAC recruitment appears to be important for promoting STAT5 target transcription as well. Our own lab has demonstrated that HDACs are required for STAT5-dependent expression of foxp3 in regulatory T cells (Tregs). In these studies, treatment with either the pan-HDAC inhibitor trichostatin A or the HDAC2/3-specific inhibitor apicidin prevented the conversion of Foxp3 -

progenitors into Foxp3 + Tregs [30]. In B cell progenitors, this requirement for HDACs in STAT5-dependent transcription was demonstrated by Xu et al. They showed that expression of id-1 requires deacetylation of C/EBPβ by HDAC1 recruited by STAT5 [31]. Together, these studies demonstrate that STAT5 can both induce and repress transcription of target genes by interacting with HATs and HDACs. In vivo STAT5 target genes have not been definitively identified, but it is known that STAT5 activation results in increased expression of bclxl, pax5, ebf, ccnd2 (Cyclin D2), and myc genes [24, 32-34]. Much interest has been focused on the specific mechanism by which STAT5 regulates B cell development. Although STAT5 activation can promote expression of bclxl [35], constitutive expression of the pro-survival factor Bcl-2 is unable to rescue B cell development at any stage in Il7r -/-

mice [36]. In contrast, STAT5 can also up-regulate the B lineage-specific factors ebf and pax5 [7, 35, 37], suggesting that it is involved in driving B cell differentiation, as opposed to simply promoting survival of progenitors. This distinction is often referred to as an “instructive” rather than “permissive” role for STAT5. This model has recently been called into question by studies utilizing a cre gene that is knocked into an endogenous rag1 allele, resulting in deletion of floxed Stat5a/b beginning in MPPs

14 with complete deletion at the pro-B stage [38]. In these mice pro-B and later stages are severely decreased; however, the investigators were able to overcome this block by expression of Bcl-2, suggesting that during early B cell development, STAT5 plays a “permissive” rather than “instructive” role. Despite this, it appears that the Bcl-2 transgene is not sufficient to restore the pre-B cell compartment. This is contrasted with our data indicating that constitutively active STAT5 can rescue the pre- and immature-B cell compartments in Il7r -/- mice [24]. Thus, although STAT5- mediated survival may be sufficient for the generation of pro-B cells, further differentiation past this stage relies on additional effects of STAT5.

Pre-BCR Signaling During the pro-B cell stage, rearrangement of the Igh locus initiates with Rag- mediated D-J and subsequent V-DJ recombination and results in expression of a µ heavy chain (µHC). µHC is expressed on the cell surface with a surrogate light chain (SLC) consisting of the VpreB and λ5 subunits, and together these proteins make up the pre-BCR, expression of which characterizes the pre-B cell stage. Pre-BCR expression is required for differentiation from pro- to pre-B cell, as evidenced in mice lacking any of the pre-BCR subunits. These mice all have increased numbers of pro- B cells and in the case of µMT -/- mice, lack peripheral B cells [39-41]. The pre-BCR is associated with the signaling molecules Igα and Igβ which mediate signaling through the pre-BCR via the kinase Syk as well as Src-family kinases such as Blk, Fyn, and Lyn [42]. Although lack of pre-BCR expression inhibits B cell development, it is the ability of the receptor to signal that is essential for

15 differentiation. This can be seen in Igα and Igβ-deficient mice, which have a similar phenotype to µMT -/- , vpreb -/- , and λ5 -/- mice [43]. So pre-BCR signaling is important for B lymphopoiesis, but exactly how signals initiate remains controversial. Studies have shown that pre-BCR can bind to antigen expressed on bone marrow stromal cells [44, 45], but these ligands are not required for B cell development in vitro [46]. In contrast, other studies showed that membrane localization of the cytoplasmic tails of Igα and Igβ was sufficient for signaling, even without the µHC, providing evidence for tonic signaling [47]. One model posits that tonic signaling is possible due to the ability of the pre-BCR to self-oligomerize independent of ligand via λ5 [48]. However, the SLC is not required for pre-BCR signaling as it can be substituted with a conventional light chain (LC) [49]. Further complicating this question is the recent finding that the SLC can recognize multiple molecular structures, including DNA, insulin, lipopolysaccharide, and more [50]. Regardless of how signaling initiates, activation of downstream effectors is important for differentiation. Following activation of Blk, Fyn, and Lyn, phosphorylation of PI3K occurs, initiating a signaling cascade leading to activation of AKT and eventual proliferation via control of the cell cycle [51] (Figure 3). In addition to PI3K, phosphorylation of the adaptor molecule Blnk also occurs. Blnk is a 65-kDa protein that acts as a scaffold protein for SH2-containing molecules. Blnk is complexed with the linker molecule Grb2, coupling it to the Ras/Raf/MEK/ERK pathway, but it has also been shown to interact with Bruton’s tyrosine kinase (Btk) to promote activation of PLCγ2 and ultimately NF-κB [52, 53]. This Btk-dependent activation of PLCγ2 has been shown to activate the

16

Figure 3

PI3K Akt Syk Lyn pre-BCR µHC VpreB 5 Ig/ PLC Blnk Btk Grb2 PKC Carma1 Malt1 Bcl10 NF-B P Proliferation Ras Raf Mek Erk Differentiation PtdIns(4,5)P 2 DAG

17 Figure 3. Pre-BCR signaling. Schematic of signaling through the pre-BCR. Tyrosine kinase activity of Syk and Lyn phosphorylates immunoreceptor tyrosine-based activation motifs (ITAMs) in the cytoplasmic tails of Igα and Igβ, leading to the recruitment of Blnk. Phosphorylation of Blnk by Syk provides docking sites for SH2 domain-containing effectors Btk, PLCγ, and Grb2. These effectors can then activate separate pathways downstream of Blnk, leading to proliferation or differentiation. Blnk-independent activation of PI3K by Syk also leads to proliferation.

18 Ras/Raf/MEK/ERK pathway as well, suggesting that loss of either Blnk or Btk can result in deficient signaling through this pathway [54]. Signaling through the pre-BCR acts as a checkpoint in B lymphopoiesis to ensure the generation of cells carrying a functional HC. A functional pre-BCR signal halts HC rearrangement, initiates several rounds of proliferation, and down-regulates IL7R and pre-BCR expression, resulting in a self-limiting expansion. Next, RAG activity is reinitiated, and recombination of the LC begins. However, if a signaling- competent pre-BCR is not expressed, RAG-mediated Igh recombination continues and a newly-rearranged µHC is produced. The mechanisms by which these different outcomes are controlled remains an area of intense interest, but it is thought that the two distinct pre-BCR pathways – Blnk-dependent and Blnk-independent – separately control differentiation and proliferation, respectively. It is currently thought that proliferation is initiated by AKT activation via the PI3K-dependent pathway, whereas Blnk-dependent signaling inhibits proliferation, down-regulates IL7R expression, and drives LC rearrangement [51]. Evidence for this role for comes from blnk -/- mice, which have increased numbers of pre-B cells and enhanced proliferative capacity; additionally, re-expression of Blnk in these mice resulted in down-regulation not only of pre-BCR expression, but PKB signaling as well [55]. Blnk-dependent signaling is also important for down-regulating pre-BCR expression and initiating LC rearrangement. Two targets of NF-κB downstream of Blnk are the interferon regulatory factors 4 and 8 (IRF4/8). These factors play redundant roles in governing B cell differentiation by regulating expression of the key developmental factors Ikaros and Aiolos [56-59]. These related zinc finger transcriptions factors are

Full document contains 158 pages
Abstract: Acute lymphoblastic leukemia (ALL) is one of the most common forms of cancer in children, yet the mechanisms driving its development remain incompletely defined. Due to the prevalence of ALL in children and its poor prognosis in adults, furthering our understanding of the underlying factors driving this disease is important for controlling a significant source of morbidity and mortality in patients. The purpose of the experiments described in this thesis was to describe the interplay between important B cell development factors in driving B cell ALL. Our lab previously reported that expression of a constitutively active form of STAT5 (STAT5b-CA) during B cell development leads to a 1-2% incidence of ALL in mice. In order to describe the mechanism driving leukemogenesis, the role of STAT5 in ALL was investigated. Elevated STAT5 activity was found in human ALL and predicted treatment response in high-risk patients. In mice, STAT5b-CA-driven ALL was found to lack expression of pre-BCR signaling factors and STAT5 cooperated with a loss of pre-BCR signaling factors to drive pre-B ALL. This cooperation was due to effects of STAT5 other than up-regulation of survival signals. Instead, STAT5 was shown to repress a set of NF-κB target genes, and these genes were further down-regulated in STAT5b-CA-driven leukemia when pre-BCR signaling was also impaired. These findings provide a point of cross-talk between STAT5 and pre-BCR signaling in B cell development. Furthermore, loss of NF-κB cooperated with constitutively active STAT5 to induce leukemia, implicating for the first time NF-κB as a tumor suppressor in ALL. Thus, herein is described a novel model of pre-B ALL which mimics disease in humans and provides insight into the mechanisms regulating both normal and aberrant B lymphopoiesis.