The Pennsylvania State University

The Graduate School

College of Medicine

MECHANISM OF DRUG ACTION OF THE SPECIFIC CK2 INHIBITOR CX-4945 IN

ACUTE MYELOID LEUKEMIA

A Dissertation in

Biomedical Sciences

by

Sadie Lynne Steffens

 2015 Sadie Lynne Steffens

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

December 2015

The dissertation of Sadie Lynne Steffens was reviewed and approved* by the following:

Sinisa Dovat Physician, Associate Professor of Pediatrics, Pharmacology, & Biochemistry Director, Translational Research – Four Diamonds Pediatric Cancer Research Center Dissertation Advisor Chair of Committee

Barbara A. Miller Physician, Professor of Pediatrics Chief, Division of Pediatric Hematology/Oncology

Sergei A. Grigoryev Professor of Biochemistry and Molecular Biology

Jong K. Yun Associate Professor of Pharmacology

Ralph L. Keil Associate Professor of Biochemistry and Molecular Biology Chair, Biomedical Sciences Graduate Program

*Signatures are on file in the Graduate School

ii

ABSTRACT

Acute myeloid leukemia (AML) is a malignant disease of the myeloid line of blood cells and is characterized by the rapid growth of abnormal white blood cells that accumulate in the bone marrow and interfere with the production of normal blood cells.

Cytarabine and other currently available treatments for acute myeloid leukemia are highly toxic and insufficient, as more than half of all AML patients develop resistance to chemotherapeutic agents. Since AML often affects older people who are less tolerant of chemotherapy, there is need for novel, targeted, less toxic drugs in order to improve survival for this disease.

Casein Kinase II (CK2) is a pro-oncogenic serine/threonine kinase that is essential for cellular proliferation. Overexpression or increased CK2 activity is associated with various types of human malignancies. In hematopoietic cells, increased CK2 expression is associated with malignant transformation and development of leukemia. Increased

CK2 activity is associated with a poor prognosis in AML. Targeted inhibition of CK2 with a novel, specific inhibitor produced a strong anti-leukemia effect in vitro and in pre- clinical models. However, the mechanism through which CK2 inhibitors exert an anti- leukemia effect is unknown. The goal of our project is to identify the mechanism of the therapeutic activity of CK2 inhibition using CX-4945 in AML.

Ikaros is a zinc finger, DNA-binding protein that is encoded by the IKZF1 gene and acts as a tumor suppressor in hematopoietic malignancies. Deletion or functional inactivation of Ikaros is associated with development of high-risk acute lymphoblastic leukemia (ALL) as well as AML. Previously published data showed that CK2 directly

iii

phosphorylates Ikaros at multiple evolutionarily-conserved sites. The CK2-mediated phosphorylation of Ikaros results in reduced DNA-binding affinity and loss of Ikaros function as a transcriptional regulator of gene expression. Inhibition of CK2 restores

Ikaros function as a tumor suppressor and produces an anti-leukemia effect in ALL.

Based on these data, we hypothesized that one of the mechanisms of therapeutic action of

CK2 inhibitors in AML involves restoration of Ikaros function as a transcriptional regulator of genes involved in malignant transformation.

Genome-wide binding studies using chromatin immunoprecipitation coupled with next-generation sequencing (ChIP-Seq) demonstrated that inhibition of CK2 via CX-

4945 in U937 and primary AML cells enhances Ikaros binding affinity at the promoter regions of its target genes. We used gain-of-function and loss-of-function experiments to determine how Ikaros regulates several novel target genes involved in malignant transformation and drug resistance. Results demonstrated that Ikaros directly represses transcription of the BCL2A1 gene, which promotes leukemogenesis and has an anti- apoptotic function. We show that the ability of Ikaros to repress transcription of BCL2A1 is impaired in AML, overexpression of BCL2A1 is a negative prognostic marker, and anti-apoptotic mechanisms contribute to resistance to chemotherapy. Repression of

BCL2A1 results in increased apoptosis. The ability of Ikaros to repress BCL2A1 transcription is impaired by CK2. Inhibition of CK2 via CX-4945 increases Ikaros binding at the BCL2A1 promoter, resulting in transcriptional repression of BCL2A1.

Increased Ikaros binding to the BCL2A1 promoter is associated with formation of repressive chromatin that is characterized by the loss of the positive marker H3K4me2

iv

and increase in the negative marks H3K9me3 and H3K27me3. Since BCL2A1 has an anti-apoptotic function, we tested whether CK2 inhibition increases susceptibility of

AML cells to apoptosis. Results showed that CK2 inhibition increases apoptosis of AML cells and that the CK2 inhibitor CX-4945 has synergistic therapeutic effects in combination with a standard drug for AML, Doxorubicin. These data indicate a new

Ikaros target gene, one mechanism for the therapeutic activity of CK2 inhibition, and a novel combination treatment for AML.

Similar functional experiments were performed on four additional Ikaros target genes that were identified by ChIP-Seq (MTHFR, CDA, DLX1, and DLX2). Results demonstrated that CK2 impairs transcriptional regulation of these genes by Ikaros in

AML. Treatment with CK2 inhibitor restores Ikaros-mediated regulation of these genes.

Since two of the newly-identified Ikaros target genes are known to be involved in drug resistance, the therapeutic effect of the CK2 inhibitor CX-4945 has been tested in combination with additional chemotherapeutic agents, and results showed a synergistic effect of these combination treatments in AML cells.

Finally, a systems biology approach was used to determine the effect of CK2 inhibition on the epigenome and transcriptome of AML cells. The analysis revealed that

CK2 inhibition results in alterations in the epigenomic signature of AML cells. The prominent changes involved alteration of enhancer and super-enhancer landscapes, which were associated with transcriptional regulation of many genes that are critical for cellular proliferation.

v

In summary, our results demonstrate that the therapeutic effect of CK2 inhibition in AML cells involves restoration of Ikaros function as a tumor suppressor and transcriptional regulator. Our results have identified novel pathways that are regulated by

Ikaros as well as an epigenomic landscape that is regulated by CK2. These data led to the development of novel combination treatments for AML which showed synergy when tested on AML cells. Our results provide a mechanistic rationale for development of novel, targeted treatments for AML.

vi

TABLE OF CONTENTS

List of Figures ...... xi

List of Tables ...... xiv

List of Abbreviations ...... xv

Chapter 1 Introduction and Literature Review ...... 1

Introduction ...... 2 Current Treatments for AML ...... 8 Casein Kinase 2 as a Drug Target in AML ...... 9 The CK2 Inhibitor CX-4945 as a Potential Treatment for AML ...... 11 Ikaros as a Tumor Suppressor in Leukemia ...... 12 Ikaros in Development ...... 15 Regulation of Ikaros by CK2 ...... 17 The Role of Epigenetics in Differentiation ...... 20 Dissertation Goals and Hypotheses ...... 21 Models ...... 23 Ikaros Target Genes Identified by ChIP-Seq ...... 24 Tables ...... 27 Figures ...... 41 References ...... 47

Chapter 2 Therapeutic Efficacy of the Specific CK2 Inhibitor CX-4945 in AML Involves Transcriptional Repression of BCL2A1 by Ikaros ...... 57

Abstract ...... 58 Introduction ...... 59 Experimental Procedures ...... 61 Cells, Cell Culture, and Reagents ...... 61 ChIP-Seq experiments ...... 61 Antibodies ...... 62 Cell Proliferation Assay ...... 63 DNA Constructs and shRNA ...... 63 Retroviral Gene Transfer and Cell Sorting ...... 64 Luciferase Assay ...... 64 Quantitative Chromatin Immunoprecipitation ChIP (qChIP) ...... 65 Quantitative Reverse-Transcriptase PCR (qRT-PCR) ...... 65 Biochemical Experiments ...... 66 CK2 Kinase Activity ...... 66 Ikaros and CK2 shRNA Knockdown ...... 67 vii

Drug Synergy Analysis with Calcusyn Software ...... 67 Results...... 68 Treatment with CX-4945 enhances Ikaros binding to the promoter of BCL2A1 in AML cells...... 68 Ikaros represses transcription of BCL2A1 ...... 70 Casein Kinase II (CK2) positively regulates transcription of BCL2A1 ...... 71 CK2 inhibits Ikaros-mediated transcriptional repression of BCL2A1 ...... 72 Inhibition of CK2 alters the epigenetic signature at the BCL2A1 promoter ...... 74 Inhibition of CK2 induces apoptosis and shows synergistic effects with doxorubicin in U937 cells ...... 75 Discussion ...... 77 Figures ...... 80 References ...... 88

Chapter 3 Ikaros and CK2 Regulate Transcription of MTHFR and CDA ...... 93

Introduction ...... 94 Experimental Procedures ...... 96 Cell Culture ...... 96 DNA Constructs ...... 97 Transfection ...... 97 Gene Expression Analysis by qRT-PCR ...... 97 Chemicals ...... 98 Antibodies ...... 98 qChIP ...... 99 ChIP-Seq ...... 99 Luciferase Assay ...... 100 Ikaros and CK2 shRNA Knockdown ...... 101 Cell Proliferation Assays ...... 101 Drug Synergy Analysis with Calcusyn Software ...... 102 Annotation of ChIP-Seq Peaks ...... 102 Distribution of peaks in different gene regulatory regions ...... 102 Relation of Ikaros, HDAC1 and Histone Modification Peaks ...... 103 Results...... 104 Ikaros Binds to the Promoters of Several Genes Implicated in AML and Drug Resistance ...... 104 Gain-of-Function: Ikaros Represses the Transcription of MTHFR and CDA ...... 105 Loss-of-Function: Knockdown of Ikaros Results in Increased Transcription of MTHFR and CDA ...... 106 Molecular Inhibition of CK2 Results in Transcriptional Repression of Ikaros Target Genes in a Manner Similar to Ikaros Overexpression ..... 107 Pharmacological Inhibition of CK2 with CX-4945 Represses Ikaros Target Genes in AML ...... 108 viii

Inhibition of CK2 Enhances the Ability of Ikaros to Bind to the Promoters of its Target Genes ...... 109 Transcriptional Repression of Ikaros Target Genes Following CK2 Inhibition Requires Ikaros Activity ...... 109 Epigenetic Changes Related to Ikaros Target Genes ...... 110 Ikaros Represses Transcription of its Target Genes via HDAC1- dependent and -Independent Mechanisms ...... 112 CX-4945 and Combination Therapy ...... 113 CX-4945 Synergizes with Methotrexate ...... 113 CX-4945 Synergizes with Cytarabine ...... 115 Discussion ...... 116 Figures ...... 119 References ...... 133

Chapter 4 Ikaros and CK2 Regulate Transcription of DLX1 and DLX2 ...... 137

Introduction ...... 138 Experimental Procedures ...... 139 Cell Culture ...... 139 DNA Constructs and shRNA ...... 139 Transfection ...... 140 Gene Expression Analysis by qRT-PCR ...... 140 Antibodies and Chemicals ...... 141 qChIP ...... 141 ChIP-Seq ...... 141 Luciferase Assay ...... 142 Annotation of ChIP-Seq Peaks ...... 143 Distribution of peaks in different gene regulatory regions ...... 143 Results...... 144 Ikaros Binds to the Promoters of DLX1 and DLX2 ...... 144 Gain-of-Function: Ikaros Represses the Transcription of DLX1 ...... 144 Loss-of-Function: Knockdown of Ikaros Results in Increased Transcription of DLX1 and DLX2 ...... 146 Molecular Inhibition of CK2 Results in Transcriptional Repression of Ikaros Target Genes in a Manner Similar to Ikaros Overexpression ..... 146 Transcriptional Repression of Ikaros Target Genes Following CK2 Inhibition Requires Ikaros Activity ...... 147 Ikaros Represses Transcription of DLX1 via an HDAC-dependent Mechanism ...... 148 Discussion ...... 149 Figures ...... 150 References ...... 157

Chapter 5 Treatment with the CK2 Inhibitor CX-4945 Alters the Epigenetic Landscape in U937 Cells ...... 160 ix

Introduction ...... 161 Experimental Procedures ...... 165 Gene Expression Analysis ...... 165 ChIP-Seq Data Analysis ...... 165 Super Enhancer Analysis ...... 166 Enhancer Analysis ...... 166 Results...... 167 Transcriptome Analysis of U937 Cells by RNA-Seq ...... 167 Additional Epigenetic Changes ...... 170 Discussion ...... 171 Tables ...... 172 Figures ...... 173 References ...... 180

Chapter 6 Overall Discussion ...... 182

Summary ...... 183 Proposed Model ...... 184 Discussion ...... 186 Figures ...... 190

x

LIST OF FIGURES

Figure 1.1. The involvement of CK2 in cellular pathways that promote cell survival...... 41

Figure 1.2. Ikaros can be phosphorylated at several different CK2 consensus sequences...... 42

Figure 1.3. Modification of histone tails is important for regulation of gene expression...... 43

Figure 1.4. Cytotoxicity assay for CX-4945 in U937 cells...... 44

Figure 1.5. Comparison of unique genes bound by Ikaros in untreated and CX- 4945 treated U937 cells by ChIP-Seq analysis...... 45

Figure 1.6. Ikaros binding peaks are redistributed to the transcription start site (TSS) following CX-4945 treatment...... 46

Figure 2.1. Treatment with CX-4945 enhances Ikaros binding to the promoter of BCL2A1 in AML cells...... 80

Figure 2.2. Ikaros represses the transcription of BCL2A1...... 82

Figure 2.3. CK2 inhibits Ikaros-mediated repression of BCL2A1...... 83

Figure 2.4. Inhibition of CK2 enhances the ability of Ikaros to bind to the promoter of BCL2A1...... 84

Figure 2.5. Treatment with CX-4945 promotes heterochromatin formation at the promoter of BCL2A1...... 85

Figure 2.6. Apoptosis is increased in U937 following treatment with CX- 4945...... 86

Figure 2.7. Model representing regulation of BCL2A1 by the CK2-Ikaros axis in leukemia...... 87

Figure 3.1. Ikaros binds to the promoters of MTHFR and CDA in U937 cells...... 119

Figure 3.2. Ikaros binds to the promoters of MTHFR and CDA in Pheresis #4 primary AML cells...... 120

Figure 3.3. Ikaros directly represses the transcription of MTHFR and CDA...... 121

xi

Figure 3.4. Ikaros overexpression represses the transcription of MTHFR and CDA...... 122

Figure 3.5. Knockdown of Ikaros increases the transcription of MTHFR and CDA...... 123

Figure 3.6. Knockdown of CK2 decreases the transcription of MTHFR and CDA...... 124

Figure 3.7. Pharmacological inhibition of CK2 decreases the transcription of MTHFR and CDA in U937 cells...... 125

Figure 3.8. Pharmacological inhibition of CK2 decreases the transcription of MTHFR and CDA in Primary AML cells...... 126

Figure 3.9. Inhibition of CK2 enhances the ability of Ikaros to bind to the promoters of MTHFR and CDA in U937 Cells...... 127

Figure 3.10. Inhibition of CK2 enhances the ability of Ikaros to bind to the promoters of MTHFR and CDA in Pheresis #4 primary AML cells...... 128

Figure 3.11. Transcriptional Repression of MTHFR and CDA Following CK2 Inhibition Requires Ikaros Activity...... 129

Figure 3.12. Treatment with CX-4945 promotes heterochromatin formation at the promoter of CDA...... 130

Figure 3.13. CX-4945 Synergizes with methotrexate in U937 cells...... 131

Figure 3.14. CX-4945 Synergizes with cytarabine in U937 cells...... 132

Figure 4.1. Ikaros binds to the promoters of DLX1 and DLX2 in U937 cells...... 150

Figure 4.2. Ikaros directly represses the transcription of DLX1...... 151

Figure 4.3. Ikaros overexpression represses the transcription of DLX1 and DLX2...... 152

Figure 4.4. Knockdown of Ikaros increases the transcription of DLX1 and DLX2...... 153

Figure 4.5. Knockdown of CK2 decreases the transcription of DLX1 and DLX2...... 154

Figure 4.6. Transcriptional Repression of DLX1 and DLX2 Following CK2 Inhibition Requires Ikaros Activity...... 155

xii

Figure 4.7. Inhibition of histone deacetylases prevents the Ikaros-mediated repression of DLX1...... 156

Figure 5.1. Transcriptome Analysis...... 173

Figure 5.2. Alterations in enhancer activity after CX-4945 treatment...... 174

Figure 5.3. Change in enhancer activities are correlated with gene expression changes...... 175

Figure 5.4. Treatment of U937 cells with CX-4945 leads to differential enrichment of positive histone modifications at induced and silenced enhancers...... 176

Figure 5.5. CX-4945 treatment alters H3K27ac super-enriched regions (i.e. super-enhancers)...... 177

Figure 5.6. Peri-telomeric enrichment of histone H1 in U937 cells after CX-4945 Treatment...... 179

Figure 6.1. Proposed overall model...... 190

xiii

LIST OF TABLES

Table 1.1: World Health Organization classification of AML...... 27

Table 1.2: French-American-British (FAB) staging system of AML...... 28

Table 1.3: Chemical properties of CX-4945...... 29

Table 1.4: List of unique gene targets identified by ChIP-Seq of Ikaros in U937 cells...... 30

Table 3.1: Primers used for target gene studies...... 98

Table 4.1: Primers used for target gene studies...... 140

Table 5.1: Top differentially expressed genes after CX-4945 treatment, sorted by absolute difference...... 172

xiv

LIST OF ABBREVIATIONS

AKT protein kinase B

ALL acute lymphoblastic leukemia

AML acute myelogenous leukemia ara-C cytarabine

ARC apoptosis repressor with caspase recruiting domain

BCL2A1 BCL2-related protein A1

BCL-XL B-cell lymphoma-extra large

BID BH3 interacting domain death agonist

CDA cytidine deaminase

CEBPA CCAAT/enhancer binding protein (C/EBP), alpha

ChIP chromatin immunoprecipitation

ChIP-Seq chromatin immunoprecipitation with deep sequencing

CI combination index

CK2 casein kinase II

CK2α casein kinase II catalytic subunit

CtBP c-terminal binding protein

CX-4945 5-(3-chlorophenylamino) benzo[c][2,6]naphthyridine-8-carboxylic acid

DLX1 distal-less homeobox 1

DLX2 distal-less homeobox 2

FAB French-American-British staging system

FLT3 fms-related tyrosine kinase 3 xv

H3K14ac histone 3 lysine 14 acetylation

H3K27ac histone 3 lysine 27 acetylation

H3K27me3 histone 3 lysine 27 trimethylation

H3K36me3 histone 3 lysine 36 trimethylation

H3K4me1 histone 3 lysine 4 monomethylation

H3K4me2 histone 3 lysine 4 dimethylation

H3K4me3 histone 3 lysine 4 trimethylation

H3K56ac histone 3 lysine 56 acetylation

H3K64me3 histone 3 lysine 64 trimethylation

H3K79me2 histone 3 lysine 79 dimethylation

H3K79me3 histone 3 lysine 79 trimethylation

H3K9ac histone 3 lysine 9 acetylation

H3K9me3 histone 3 lysine 9 trimethylation

H3S10-P histone 3 serine 10 phosphorylation

H4K16ac histone 4 lysine 16 acetylation

H4K20me2 histone 4 lysine 20 dimethylation

H4K20me3 histone 4 lysine 20 trimethylation

H4K5ac histone 4 lysine 5 acetylation

HDAC histone deacetylase

HIF-1 hypoxia-inducible factor 1

HOX homeobox genes

JAK janus kinase

xvi

MAPK mitogen-activated protein kinase mRNA messenger ribonucleic acid

MTHFR methylenetetrahydrofolate reductase

MTX methotrexate

NF-κB nuclear factor kappa-light-chain-enhancer of activated B cells

NPM1 nucleophosmin 1

NuRD nucleosome remodeling and deacetylase complex

PC-HC pericentric heterochromatin

PI3K phosphoinositide 3-kinase

PML promyelocytic leukemia protein

Pol-II RNA polymerase II

PTEN phosphatase and tensin homolog qChIP quantitative chromatin immunoprecipitation shRNA small hairpin ribonucleic acid

STAT signal transducer or activator of transcription

SWI/SNF switching defective/sucrose nonfermenting

TdT terminal deoxynucleotidyl transferase

WHO World Health Organization

Wnt wingless-related integration site

xvii

Chapter 1

Introduction and Literature Review

1

Introduction

Acute myeloid leukemia (AML) is a malignant disease of the myeloid line of blood cells and is characterized by the rapid growth of abnormal white blood cells that accumulate in the bone marrow and peripheral blood and interfere with the production of normal blood cells (1,2). Alterations at the genetic and epigenetic levels result in disturbance of normal processes, especially self-renewal, proliferation, and differentiation

(3-6). AML is the most common acute leukemia affecting adults, and its incidence increases with age. Replacement of normal bone marrow with leukemic cells results in a drop in red blood cells, platelets, and normal white blood cells. Resulting symptoms include fatigue, shortness of breath, easy bruising and bleeding, and increased risk of infection. Several risk factors and chromosomal abnormalities have been identified, but for AMLs without cytogenetic abnormalities, the specific cause is not clear. As an acute leukemia, AML progresses rapidly and is typically fatal within weeks or months if left untreated.

AML is a heterogeneous disease with several subtypes, and treatment and prognosis varies greatly among these subtypes. In particular, cytogenetic abnormalities can be used to stratify patients according to disease severity (7). There are two classification systems used during diagnosis of AML to stratify patient prognosis. The

World Health Organization (WHO) classification focuses on translocations and dysplasia and divides AML into seven subtypes : 1) AML with recurrent genetic abnormalities, such as specific chromosomal changes, 2) AML with multilineage dysplasia, in which more than one myeloid cell type is involved, 3) AML related to previous chemotherapy,

2

4) AML that is not otherwise categorized, 5) myeloid sarcoma, 6) myeloid proliferations related to Down Syndrome, and 7) blastic plasmacytoid dendritic cell neoplasms (Table

1.1) (8). The French-American-British (FAB) staging system is based on morphology

(see Table 1.2). For patients that do not have cytogenetic abnormalities, FLT3, NPM1, and CEBPA mutational analysis can be used for risk stratification. By karyotyping and molecular profiling, patients can be subdivided into one of three broad risk groups including AML with favorable, intermediate, or unfavorable cytogenetic risk (9).

AML is treated initially with chemotherapy aimed at inducing a remission, and patients may go on to receive additional chemotherapy or a hematopoietic stem cell transplant, if required. However, currently available treatments for AML are insufficient, as more than half of all AML patients develop disease that is resistant to chemotherapy

(10). Despite these known classifications, many patients lack one of these prognostic indicators, and even patients that can be assigned to a group exhibit heterogeneity in outcome within the group. The identification of further somatic mutations or biological markers to aid in patient classification is necessary. Although several recurrent mutations have been identified, these findings have not been validated in a clinical setting.

Casein Kinase 2 (CK2) is a ubiquitous and constitutively expressed serine/threonine kinase that has been observed to have enhanced activity in many human cancers, including AML (11-13). The structure of CK2 is a tetramer comprised of two regulatory beta subunits and two alpha subunits which contain the catalytic kinase domain (13). Involvement of CK2 in many cellular pathways that promote cell survival is well documented (Figure 1.1) (14). The use of CK2 inhibitors such as TBB has been shown to induce apoptosis and prevent the proliferation of cancer cells, and antisense 3

inhibitors of CK2α, the catalytic subunit of CK2, can induce apoptosis in a xenograft mouse model of prostate cancer (15,16). The catalytic subunit of CK2 has also been identified as a negative prognostic marker in AML, and induced overexpression of CK2α in U937 cells, a cell line from the myeloid lineage, corresponds with an increase in phosphorylated-Akt and activation of the phosphatidylinositol 3-kinase pathway.

Treatment with a CK2 inhibitor is able to reverse these effects (17).

In addition to its implication in AML, upregulation of the catalytic subunit of

CK2 has been shown to promote the development of thymic lymphomas in mice (18).

Transfection of cancer cells with the catalytic subunit of CK2 results in a protective effect against treatment with apoptosis-promoting drugs such as etoposide. In contrast, the presence of the regulatory subunit of CK2 does not provide protection from apoptosis upon etoposide treatment (19). These observations indicate an anti-apoptotic role for

CK2 as well as a role in the mediation of cell growth and proliferation. Another example of the cancer-promoting capability of CK2 is its involvement in the Wnt signaling pathway. CK2 phosphorylates beta-catenin, resulting in its stabilization and ultimate translocation to the nucleus to activate genes involved in cell proliferation and growth

(20). These observations indicate that CK2α plays an important role in oncogenesis of myeloid leukemia and that CK2 inhibitors should be considered as potential anticancer drugs for the treatment of myeloid leukemia.

An obstacle to the use of CK2 inhibitors for cancer treatment is the possibility of off-target effects, as CK2 is known to have more than 300 substrates. CK2 is ubiquitously expressed and essential for cell viability in all cell types. However, in application, inhibition of CK2 in cancer cells and xenograft mouse models appears to 4

have little effect on normal cells (21). Treatment with an antisense CK2α oligonucleotide has been shown to potently induce apoptosis in cancer cells and to have an anti-tumor effect in xenograft mouse models of prostate cancer and squamous cell carcinoma while leaving normal cells relatively unaffected (22). The inhibition of CK2 appears to be most effective in cells with overexpression of CK2 activity and not in non-proliferating cells with normal levels of CK2 expression.

Many oncogenic signaling pathways rely on CK2, and CK2 has been shown to be involved in promoting abnormal survival, degradation of tumor suppressors, upregulation of DNA repair, positive cell cycle control, enhancement of the transforming potential of oncogenes, and also counteracting anti-tumor agents. This signaling occurs through multiple pathways, including the JAK/STAT pathway, AKT/PI3K pathway, NF-κB pathway, Wnt signaling pathway, and MAPK pathway. In addition, CK2 inhibition has been shown to decrease HIF-1 (23,24).

Of particular interest is the ability of CK2 to phosphorylate and regulate the tumor suppressor Ikaros. Ikaros is a transcription factor that was first identified by its ability to bind to and activate the enhancer of a gene known as an early marker of T cell differentiation (25). The ability of the Ikaros gene to encode a diverse family of zinc finger transcription factors that have been shown to be crucial for the development of the hematopoietic and immune lineages has identified it as a master regulator of lymphocyte maturation (25,26). Ikaros regulates the expression of its target genes via direct binding to regulatory genetic elements. Ikaros localizes to pericentromeric heterochromatin (PC-

HC) and can act as either an activator or a repressor of transcription by associating with various complexes involved in chromatin remodeling (27). Gene targets of Ikaros have 5

been identified in mouse T-cell and pre-B cell development by ChIP-Seq (28,29). By utilizing ChIP-Seq in acute lymphoblastic leukemia (ALL), our lab has found that, with the help of histone deacetylase 1 (HDAC1), Ikaros can regulate gene expression by recruitment of distinct histone markers to the promoters of its target genes (30).

Mutations that lead to Ikaros inactivation result in abrogation of proper lymphocyte differentiation, and loss of functional Ikaros has been shown to result in the development of leukemia and lymphoma (31). Additionally, Ikaros has been implicated in cell cycle control by acting as a negative regulator of the G1-S transition. The DNA- binding ability of Ikaros during mitosis is dependent upon its phosphorylation state (32).

Phosphorylation of Ikaros by CK2 has been shown to decrease the DNA-binding affinity of Ikaros, impacting the ability of Ikaros to regulate cell cycle progression (33). Eleven evolutionarily-conserved phosphorylation sites have been identified on the Ikaros protein that are located within consensus motifs for CK2, and reversible phosphorylation at these residues regulates the subcellular localization of Ikaros in addition to its DNA-binding affinity (Figure 1.2) (34).

The localization of Ikaros to pericentromeric heterochromatin is regulated by phosphorylation (27,34). In vivo phosphopeptide mapping of Ikaros determined that

Ikaros is phosphorylated at multiple sites, many of which contain consensus sequences for CK2. Phosphorylated amino acids within Ikaros are evolutionarily conserved and similar across multiple cell types, indicating that Ikaros is phosphorylated by kinases that are common to multiple cell types and that phosphorylation is an important mechanism for the regulation of Ikaros. Phosphorylation of Ikaros by CK2 regulates the DNA binding ability and nuclear localization of Ikaros (34). This evidence identifies CK2 as 6

an enzyme that directly controls Ikaros function and demonstrates that phosphorylation via CK2 can directly regulate the subcellular localization and chromatin remodeling function of Ikaros.

Defects in the Ikaros gene have also been observed in myeloid leukemia. Aberrant myelopoiesis has been observed in both mice homozygous and heterozygous for Ikaros mutations (31,35). Studies with several mouse models have implicated Ikaros as a tumor suppressor in leukemia. Mice with a heterozygous germline deletion of the DNA-binding domain of Ikaros rapidly develop T-cell lymphoma as a result of the production of dominant negative Ikaros isoforms that act to suppress the function of full-length Ikaros

(36). Additional evidence supports a role for Ikaros gene defects in many hematopoietic malignancies, including B-cell ALL (37) and AML (38). Gene targets of Ikaros include regulators of proliferation, cell cycle progression, and myeloid differentiation, several of which are proto-oncogenes (33,37,38).

Taken together, CK2 is known to phosphorylate and impair the function of the leukemia suppressor Ikaros and has oncogenic implications in many cancer cells, including AML, where it has been identified as an unfavorable prognostic marker. We hypothesize that CK2 works through inactivation of Ikaros to promote the progression of leukemia. CK2 inhibition could be important as a treatment for AML, and the work within this dissertation is focused on the drug mechanism of that inhibition. Inhibition of

CK2 works to enhance the function of Ikaros as a tumor suppressor for the prevention of cancer progression in AML. To fully understand the roles that Ikaros and other transcription factors play in AML, a more complete understanding of CK2 is required.

7

Current Treatments for AML

The first phase of treatment for AML, called remission induction, is aimed at attaining remission. Intensity of treatment is determined by the age and health of the patient at diagnosis but usually involves treatment with two chemotherapy drugs, cytarabine (ara-C) and an anthracycline such as daunorubicin. The combination of 3 days of daunorubicin followed by 7 days of cytarabine is considered the standard of care for induction treatment (10,39). In addition to inducing death in leukemia cells, most normal bone marrow cells are also destroyed by the process of remission induction. As a result, patients are generally very ill during this phase and experience extremely low blood counts. Several weeks after the completion of chemotherapy, a bone marrow biopsy will be performed to verify that there are few bone marrow cells and less than 10% blasts.

Normal bone marrow cells will eventually return and proliferate. At this point, an ideal biopsy would indicate that the leukemia is in remission. If leukemia cells are still present in the bone marrow, additional chemotherapy or a stem cell transplant may be recommended. Induction is considered successful if remission is achieved.

Because remission induction does not destroy all of the leukemia cells, consolidation, or post-remission therapy, is necessary to prevent relapse. Here, several cycles of high-dose ara-C is considered the standard of care. Another treatment option for post-remission therapy is stem cell transplant. While stem cell transplant has been shown to reduce the risk of relapse in comparison to standard chemotherapy, serious complications are more likely and treatment is associated with increased risk of death

(40).

8

The combination of daunorubicin with cytarabine has been used for decades in the treatment of AML. However, intensification of this treatment may result in more remission and improve overall survival for patients. Use of an increased dose or duration of cytarabine or replacement of daunorubicin with an alternative drug have been shown to have little impact on overall survival (40). Methotrexate (MTX) treatment, although beneficial as a treatment for Acute Lymphoblastic Leukemia (ALL), has not been included in standard AML therapy because AML has been thought to be MTX-resistant

(41,42). However, some subtypes of AML, including AML-M5, have been shown to be sensitive to MTX treatment (42). A combination therapy that could overcome MTX resistance in AML would be useful as an alternative treatment for AML. There have been few new drugs approved for use in AML in the last 20 years, and new drug targets and drug combinations are necessary to improve the current standard of care.

Casein Kinase 2 as a Drug Target in AML

The protein kinase CK2 is a tetrameric complex consisting of two regulatory beta subunits and two catalytic subunits, CK2α and CK2α’. Ubiquitously expressed in all cell types, CK2 is able to phosphorylate serine and threonine residues that are in close proximity to acidic amino acids (11-13). A minimal consensus binding sequence for

CK2 has been identified (Ser-Xaa-Xaa-Acidic, with the acidic residue being Glu, Asp, pSer or pTyr) that is distinct from any other protein kinase (43). While presence of the minimal consensus binding sequence may suggest potential targets of CK2, there are sites

9

that are efficiently phosphorylated by CK2 that do not contain this sequence. Likewise, presence of the sequence does not necessarily indicate efficient phosphorylation, as tertiary structure may modulate phosphorylation efficiency.

Overexpression of CK2 has been observed in many cancers. Phosphorylation of anti-apoptotic molecules such as the tumor suppressors PTEN and PML, ARC, and BH3

BID by CK2 contributes to protection from apoptosis (44-46). CK2 is known to regulate both the intrinsic and extrinsic apoptotic pathways. This role of CK2 is in addition to its involvement in pathways that support cell survival, including upregulation of PI3K/AKT,

NF-κB, JAK/STAT, and Wnt/β-catenin signaling (23,24). CK2 is also involved in the regulation of DNA repair and has been shown to facilitate repair of both single and double stranded DNA breaks (47). For these reasons, it appears that the role of CK2 overexpression in tumorigenesis is to promote cells that are apoptosis-resistant and inclined to high rates of DNA damage repair and cell proliferation.

The role of CK2 is less defined in blood malignancies, although CK2 overexpression has been observed. Recently, CK2 has been shown to sustain AML cell survival. Treatment of both AML cell lines and primary AML cells with specific CK2 inhibitors results in activation of apoptotic pathways and cell growth arrest (48). It has been shown that the anti-apoptotic role of CK2 in AML is involved in the prevention of drug-induced apoptosis in AML cells. Inhibition of CK2 by the inhibitor CX-4945 or siRNAs results in a synergistic increase in the cytotoxic effects of daunorubicin treatment in AML cells (48).

One study determined that CK2α overexpression occurs in AML cell lines and in a considerable proportion of primary leukemic blasts from AML patients. Rates of 10

disease-free and overall survival were significantly decreased in patients with normal karyotype AML that had high levels of CK2α expression (17). Overexpression of CK2α results in increased levels of pAkt/PKB and Bcl-2 in U937 cells and primary blasts, and treatment with a selective CK2 inhibitor is able to reduce these molecules in a dose- dependent manner (17). The CK2 inhibitor apigenin was able to increase apoptosis preferentially in AML cell lines and primary blasts expressing high levels of CK2α; and sensitivity to apigenin in cell lines with low levels of CK2α was increased with induced overexpression of CK2α (17). These data show that CK2 acts as an important regulator of AML cell survival and response to chemotherapy.

The CK2 Inhibitor CX-4945 as a Potential Treatment for AML

CX-4945 [5-(3-chlorophenylamino) benzo[c][2,6]naphthyridine-8-carboxylic acid] is a novel, orally available small molecule ATP-competitive inhibitor of CK2. The chemical properties of CX-4945 are listed in Table 1.3.

Currently undergoing clinical trials in solid tumors as well as in relapsed multiple myeloma patients, CX-4945 is the first orally available CK2 inhibitor in clinical trials for cancer (49). CX-4945 suppresses the phosphorylation of Akt as well as promotes cell cycle arrest and apoptosis in vitro. CX-4945 was well tolerated in xenograft mice and exhibited a dose-dependent antitumor activity (16,31). Drug resistance is a problem in cancer treatment, but CX-4945 has been shown to synergize with chemotherapeutics such as gemcitabine and cisplatin by blocking the DNA repair response (21).

11

It is likely that the next effective treatment for AML will involve drug combination therapy. Drug combination therapy has the potential to increase therapeutic efficacy, reduce toxicity, and decrease drug resistance. Using mechanistic rationale, combinations can be designed that target multiple molecules in the same pathway to produce a maximum response to treatment.

Ikaros as a Tumor Suppressor in Leukemia

Five genes make up the Ikaros zinc finger family, and the Ikaros gene IKZF1 was the first to be discovered. Consistent among all family members is the presence of two sets of conserved C2H2 Krupple-like zinc finger DNA-binding domains (50,51). The

C2H2 zinc finger motif consists of two cysteines and two histidines that coordinate with a zinc atom to form a DNA-binding domain (52). Zinc finger proteins are characterized by multiple zinc finger motifs arranged in tandem that allow for DNA binding or for interactions with other proteins (52-54). Members of the Ikaros family consist of two sets of zinc finger domains with distinct functions. One set of zinc finger domains is found at the amino (N)-terminus and the other set is located at the carboxy (C)-terminus (Figure

1.2). The N-terminus of Ikaros family proteins is comprised of four zinc fingers (ZF1-4) that function in DNA-binding and recognize the specific DNA consensus sequence

GGAAA (26). At the C-terminus, two additional zinc fingers (ZF5-6) facilitate interactions with other proteins. The activity of ZF5 and ZF6 allows for the dimerization

12

or multimerization of Ikaros with Ikaros family members or other proteins with similar protein interaction domains (50,55).

The seven coding exons of the Ikaros gene can be alternatively spliced to form several Ikaros isoforms (26,55). Six unique Ikaros isoforms (Ik1-Ik6) have been observed in both mice and humans (25,26,55,56). In addition, several short isoforms have been identified that lack one or all of the DNA-binding zinc finger motifs as a result of deletions or insertions that disrupt normal splicing patterns (57,58). These short isoforms have impaired DNA-binding ability and have been linked to leukemia progression

(59,60). The presence of shortened Ikaros isoforms has been observed at low levels in normal blood and bone marrow cells (61-64). While Ikaros isoform Ik-1 is the most commonly expressed Ikaros isoform in the lymphoid lineage, an additional isoform known as Ik-X is predominant in myeloid cells. Ik-X is expressed in early mouse and human hematopoietic precursors and is most similar to Ik-3, although the addition of exon 6 makes the Ik-x isoform larger (64). The interaction of various Ikaros isoforms allows for the creation of a diverse array of Ikaros complexes that are able to regulate

DNA-binding and specificity, activities that determine the subcellular localization of

Ikaros and ultimately its ability to regulate transcription of its target genes (56,65,66).

All Ikaros isoforms contain the C-terminal dimerization domain, but there is variation in the number of N-terminal DNA-binding zinc fingers amongst isoforms (57).

In general, C2H2 zinc finger proteins require at least two zinc fingers arranged sequentially for stable DNA binding (56,67). Similarly, Ikaros isoforms that contain ZF2 and ZF3 are able to bind to the core consensus Ikaros binding site, and ZF1 and ZF4 are not essential for DNA binding. ZF1 and ZF4 are involved in regulating the binding of 13

Ikaros by increasing the affinity of DNA binding at specific Ikaros target genes (26,68-

70). In particular, ZF4 is responsible for the binding of Ikaros to cells in the B and T lineage (70). Variety amongst the zinc fingers in the DNA-binding domain of Ikaros isoforms allows Ikaros to function as a transcriptional regulator with diverse roles during development and differentiation. Because dimerization is essential for Ikaros function, the presence of the dimerization domain zinc fingers ZF5 and ZF6 in all Ikaros isoforms is not surprising. Mutations that disrupt the dimerization domain result in a substantial decrease in Ikaros DNA-binding affinity and, consequently, a reduction in the ability of

Ikaros to function as a transcriptional regulator (57).

Ikaros was first identified because of its ability to bind to and activate the enhancer of an early marker of T cell differentiation. Ikaros has been identified as a master regulator of lymphocyte maturation because of its involvement in multiple lymphocyte-specific pathways (25). The Ikaros protein regulates gene transcription and chromatin remodeling, and is a master regulator of hematopoiesis and an established tumor suppressor (26,71). Alteration of Ikaros activity, such as haploinsufficiency, is sufficient to promote malignancy in human hematopoietic cells. Ikaros activity can be lost due to deletion or expression of dominant-negative Ikaros isoforms (72). Monosomy

7 or deletion of the p arm or chromosome 7, which contains the Ikaros gene locus, was identified as a recurrent chromosomal aberration in patients with secondary AML, and

Ikaros has also been identified as a poor prognostic marker in secondary AML (73).

Ikaros regulates its target genes by direct binding to regulatory genetic elements such as promotors and enhancers. The core consensus binding sequence for Ikaros is

GGGA or GGAAAA (51). Ikaros is able to recruit its target genes to regions of 14

heterochromatin through multimerization and interaction with other proteins such as histone deacetylases (HDACs) (34,68). Although Ikaros has been shown to both activate and repress its target genes, repression of its targets is most common and often occurs through association with HDAC1. Whether Ikaros acts as a repressor or activator of transcription is determined by its association with various complexes involved in chromatin remodeling.

Ikaros in Development

Studies with several mouse models have implicated Ikaros as a tumor suppressor in leukemia. Mice with a homozygous deletion of the DNA-binding domains in the N- terminus of the Ikaros gene result in the formation of dominant-negative Ikaros isoforms

(35). Homozygous deletion of the Ikaros C-terminus results in rapid degradation of

Ikaros due to its inability to form higher order structures (31). Several additional Ikaros mouse models have been identified, including two heterozygous strains with mutation of the N-terminus or C-terminus of the Ikaros gene. Mice with a heterozygous germline deletion of the DNA-binding domain produce dominant negative Ikaros isoforms that act to suppress the function of full-length Ikaros (74,75).

These transgenic Ikaros mouse models have identified Ikaros as a master regulator of hematopoiesis. Homozygous mutation of the N-terminal region of Ikaros results in deficient lymphopoiesis, as evidenced by a lack of mature lymphocytes in these mice (35). Disruption of erythropoiesis and myelopoiesis was also observed in these

15

mice. However, mice with a heterozygous mutation of the N-terminal region of Ikaros appeared to have normal surface markers at early stages but rapidly develop T-cell lymphoma as a result of the production of dominant negative Ikaros isoforms (36). These results suggest that a minimum threshold of Ikaros activity is necessary to regulate proliferation of the lymphocyte lineate and suggest that the presence of one wild-type

Ikaros allele may be sufficient (36). The expression of Ikaros throughout hematopoietic development coupled with observations from these mouse models suggest that Ikaros is necessary for lymphocyte differentiation at early stages of mouse development, but loss of Ikaros in later stages of development can be associated with malignant transformation

(25,35,36).

Homozygous mutation of the C-terminus of Ikaros results in rapid degradation that produces an effectively Ikaros-null mouse model. A complete absence of B-cells was observed in these mice. T-cells were absent in fetal mice but present shortly after birth, suggesting that additional Ikaros-independent mechanisms are responsible for differentiation of T-cells in the adult mouse (31). These mouse models provide evidence for the involvement of Ikaros in multiple hematopoietic pathways and support the identification of Ikaros as a master regulator of hematopoiesis throughout both early and late stages of mouse development.

16

Regulation of Ikaros by CK2

Ikaros is abundantly expressed in most hematopoietic cells, and Ikaros protein levels are consistent throughout the cell cycle, observations that suggest reversible, post- translational modifications as a means of regulating Ikaros. Prevention of the G2/M transition by vinblastine treatment impairs the DNA-binding ability of Ikaros, but phosphatase treatment of the arrested cells is able to dramatically increase the DNA- binding ability of Ikaros, as observed by electromobility gel shifts (32). All three evolutionarily conserved linker regions located in the DNA-binding domain of Ikaros have been shown to be phosphorylated during mitosis, and this phosphorylation is responsible for the loss of pericentromeric localization of Ikaros during the G2/M transition (32). These results indicate a cell cycle-specific regulation of Ikaros via phosphorylation within the conserved linker sequences, resulting in inactivation of Ikaros during mitosis.

Two additional regions of Ikaros that can be phosphorylated have also been identified; serine 63 within exon 4 and several amino acids upstream of the C-terminal zinc-fingers within exon 8 (33). Analysis of the sequence identified these regions as potential CK2 phosphorylation sites, and treatment with kinase-specific inhibitors determined that CK2 is the primary kinase responsible for phosphorylation of Ikaros at these sites. GSK3 and cdk were also identified as kinases able to phosphorylate Ikaros at exon 8. Ikaros was unphosphorylated in cells arrested in late G1 phase, but was phosphorylated in the transition to S phase (33). Ikaros negatively regulates the G1/S transition, but phosphorylation of Ikaros within exon 8 at this phase prevents Ikaros from 17

acting as an inhibitor by decreasing the DNA-binding ability of Ikaros. These studies provide further evidence for cell-cycle dependent phosphorylation of Ikaros.

Dephosphorylated Ikaros is active and maintains DNA-binding affinity, while phosphorylation of Ikaros abolishes its DNA-binding ability. Changes in the phosphorylation state of Ikaros have implications for cell cycle progression, especially at the transition from G1 to S phase.

In vivo phosphopeptide mapping and mass spectrometry identified four additional biologically relevant Ikaros phosphorylation sites at positions 13, 23, 101, and 294 (34).

Gel shift and confocal microscopy experiments with Ikaros phosphomutants demonstrated that phosphorylation at amino acids 13 and 294 decreased the binding of

Ikaros at γ satellite sequences as well as the regulatory regions of developmentally significant genes. Dephosphorylation of Ikaros was important for localization at pericentromeric heterochromatin (PC-HC) (34). CK2 is able to directly phosphorylate

Ikaros at these important regulatory residues, another indication that phosphorylation by

CK2 is able to regulate the ability of Ikaros to bind DNA and that this mechanism is important for normal thymocyte differentiation (34).

Phosphorylation is the major mechanism for regulation of Ikaros. The reversible nature of phosphorylation allows for the interaction of kinases and phosphatases that are able to phosphorylate and dephosphorylate Ikaros at specific amino acids. Radiolabelling has identified Protein Phosphatase 1 (PP1) as the primary phosphatase that dephosphorylates Ikaros (27). An Ikaros mutant in which the PP1 recognition motif is disrupted by changing both valine 465 and phenylalanine 467 to alanine (known as IK-

465/7A) had decreased DNA-binding ability and a loss of localization to pericentromeric 18

heterochromatin when compared to the wild-type Ikaros protein. Decreased DNA- binding ability and changes in the subcellular localization of the Ikaros protein when the

PP1 consensus site is disrupted suggest an important regulatory role for the interaction between PP1 and Ikaros. These results suggest that the decreased DNA-binding ability of

Ikaros in the absence of interaction with PP1 can be attributed to a lack of Ikaros dephosphorylation by PP1 and further suggest that the CK2 phosphorylation sites may be targets for dephosphorylation by PP1. In this way, phosphorylation of Ikaros is regulated via direct interaction with CK2 and PP1 and is crucial for Ikaros function (27).

Ikaros activity is regulated post-translationally by a number of mechanisms. First,

Ikaros is able to form higher order complexes with a variety of Ikaros isoforms and other

Ikaros family members, including shorter dominant negative isoforms unable to bind

DNA. Second, phosphorylation of Ikaros is regulated by CK2 and PP1 (27,32-34).

Phosphorylation of Ikaros by CK2 occurs at several specific amino acids and regulates its

DNA-binding ability and subcellular localization (27,32-34). Furthermore, phosphorylation by CK2 prevents Ikaros from regulating genes involved in differentiation. When compared to wild-type Ikaros, phosphoresistant Ikaros mutants are better able to bind to the promoter region of the Ikaros target gene TdT, a key gene involved in T-cell differentiation. Ikaros undergoes changes in phosphorylation state during T-cell differentiation, and dephosphorylation of Ikaros during differentiation results in increased Ikaros binding at the TdT promoter and ultimately transcriptional repression of TdT (77). These results indicate that phosphorylation of Ikaros via CK2 is one mechanism that regulates the function of Ikaros in normal hematopoiesis.

19

The Role of Epigenetics in Differentiation

All cell types share the same genetic material yet are able to differentiate into specialized cells with distinct functions. This phenomenon is accomplished by the establishment of lineage specific transcription factors that guide the activation and repression of genes in a lineage-specific manner early in development. With minor exceptions, all differentiation processes are triggered through epigenetic mechanisms.

Epigenetics is comprised of all heritable changes in gene expression and chromatin organization that are independent of the DNA itself, a dynamic system with powerful implications for the regulation of cellular processes. Epigenetics regulates the functional aspects of all genes (78).

In the cell, DNA exists in a condensed state consisting of histones and DNA, called chromatin. Post-translational modification of histone proteins at their amino- terminal tail regulates the structure and accessibility of chromatin. Histone tails can be modified by acetylation, methylation, ubiquitination, phosphorylation and sumoylation.

These modifications determine the accessibility of the DNA to transcription factors and the ability of genes in that region to be expressed. Binding of protein complexes to DNA is regulated by the presence or absence of these modifications, with some protein complexes being drawn to defined combinations of histone modifications (Figure 1.3)

(79).

Chromatin remodeling is critically important for the regulation of developmental processes. Complexes involved in chromatin remodeling often incorporate cell-type- specific subunits to promote distinct, cell-type-specific roles. In addition to their roles in 20

cell fate, proliferation, and specification, chromatin remodeling complexes have also been implicated in oncogenesis (80).

Ikaros has been shown to associate with several protein complexes that are involved in chromatin remodeling and histone deacetylation (81-83). Repression of

Ikaros target genes occurs through association with HDAC-containing complexes such as

NuRD, Sin3A and Sin3B or through interaction with the transcriptional corepressor

CtBP. Activation occurs through the association of Ikaros with the SWI/SNF complex.

The association of Ikaros with these complexes is an example of how lineage-specific transcription factors can affect epigenetic regulation of gene expression in a cell-type specific manner and provides a mechanism by which Ikaros is able to regulate differentiation in hematopoietic cells.

Dissertation Goals and Hypotheses

Although Ikaros has been implicated as a tumor suppressor in several lymphoid leukemias, and CK2 has been associated with many cancers, the proposition that Ikaros acts as a tumor suppressor in AML and that CK2 plays an important role in AML is novel. Clinically, most deletions of Ikaros occur as a loss of a single allele, and identification of the regulatory pathways involved is essential for the development of treatments to enhance the functional ability of the remaining Ikaros allele (72,84). Our use of the specific CK2 inhibitor CX-4945 is an innovative approach that allows for treatment of patients with diverse Ikaros characterization. If CK2 is responsible for

21

phosphorylation and subsequent regulation of Ikaros, treatment with CX-4945 may serve to enhance the DNA-binding ability of the wild-type Ikaros allele in leukemias that are heterozygous for Ikaros mutations, thereby increasing the tumor suppressor activity of

Ikaros in AML. Use of CX-4945 as a treatment may also provide a benefit for patients with two functional Ikaros alleles. In leukemias where overexpression of CK2 impairs the functional activity of Ikaros, treatment with CX-4945 may be able to increase the tumor suppressor activity of Ikaros even in patients with two wild type alleles.

Taken together, CK2 is known to phosphorylate and impair the function of the leukemia suppressor Ikaros and has oncogenic implications in many cancer cells, including AML, where it has been identified as an unfavorable prognostic marker. These observations have led us to hypothesize that the anti-leukemic effect of treatment with

CX-4945 occurs through Ikaros. Inhibition of CK2 restores the DNA-binding ability of

Ikaros and leads to enhanced regulation of Ikaros target genes with roles in cell cycle progression, myeloid differentiation, and cell death. Previous studies suggest that CK2 inhibition can be an effective approach for treatment of AML (17,21,85). However, the use of CK2 inhibitors has not been tested in human AML. CX-4945 is the most potent and specific CK2 inhibitor and is currently the only CK2 inhibitor being tested in Phase I clinical trials in humans.

The goal of this dissertation is to determine the therapeutic effect of the specific

CK2 inhibitor CX-4945 in AML and to identify the mechanism by which CX-4945 exerts its therapeutic activity. We therefore hypothesize that: 1) CK2 inhibition by CX-4945 has a potent anti-leukemia effect in AML. Therapeutic activity of CX-4945 will be tested in

AML cell lines and in AML primary cells. 2) The mechanism of therapeutic activity of 22

CX-4945 involves the restoration of the tumor suppressor function of Ikaros. We hypothesize that Ikaros functions as a regulator of transcription but is impaired in AML due to high activity of CK2.

Models

The work within this dissertation utilized AML cell lines and primary patient cells. For this study, the human monocytic cell line U937 was selected, which expresses full-length Ikaros (86). Although the cell line HL-60 is often used for myeloid studies, we chose the U937 cell line for two reasons: 1) Clinical relevance: HL-60 is an APML cell line, and APML is currently one of the most treatable forms of leukemia (87,88).

The U937 cell line most closely resembles monocytoid AML (M5 of the FAB classification), which often carries a poor prognosis. 2) Poor Ikaros DNA-binding in HL-

60 cells: we have observed that Ikaros exhibits poor binding in HL-60 cells, which makes comparison of changes in Ikaros DNA-binding affinity following CK2 inhibition difficult and could produce an exaggerated effect of CK2 inhibition on Ikaros DNA-binding in these cells. Pheresis #4 cells are primary cells isolated from a child with AML who presented with high blood cell count and required blood exchange (pheresis). In these studies, we will be evaluating the efficacy of CX-4945, a specific CK2 inhibitor that is currently undergoing clinical trials. In the future, this work will be continued in xenograft mice.

23

Ikaros Target Genes Identified by ChIP-Seq

To identify Ikaros target genes, ChIP-Seq of Ikaros was performed in untreated

U937 cells and in U937 cells treated with 10 µM CX-4945 for 72 hours. This concentration was chosen because it was identified as the IC50 for U937 cells after three days of treatment (Figure 1.4). After chromatin immunoprecipitation (ChIP) and library preparation, samples were sequenced from both ends and the sequencing reads were mapped to a reference genome. Protein-DNA binding regions were identified as regions with a larger number of reads than expected. Ikaros peak annotation was accomplished with the use of CisGenome software, and sequence reads aligned to a reference genome were converted into BAR files. Parameters for the detection of peaks were set, and all peaks called were required to have a number of reads at least two-fold higher than the background and a False Discovery Rate (FDR) less than 0.05. The criteria for identifying

Ikaros peaks was extremely stringent to prevent the identification of false positives.

In untreated U937 cells, 2620 Ikaros binding peaks were identified. Of these peaks, 657 unique gene targets were identified (Figure 1.5, Table 1.4). After CX-4945 treatment, 3412 Ikaros binding peaks were identified by ChIP-Seq distributed amongst

2149 unique genes. Between the two data sets, 328 genes were bound by Ikaros in both untreated and CX-4945 treated U937 cells. The global distribution of the Ikaros peaks indicated that after CX-4945 treatment, binding of Ikaros to the transcription start sites

(TSS) of genes is greatly increased (Figure 1.6).

Of the many potential Ikaros target genes identified by ChIP-Seq, we selected several genes that are important in leukemia progression. Several genes with 24

implications in drug resistance were identified as Ikaros targets by ChIP-Seq. One Ikaros target gene, cytidine deaminase (CDA) is involved in the metabolic inactivation of the chemotherapeutic agents gemcitabine and cytarabine (ara-C). Patients with a CDA mutation experience higher toxicity and adverse side effects during chemotherapy.

Mutations in CDA have also been associated with decreased overall survival and response to chemotherapy as well as increased post-induction treatment mortality in patients with

AML (89). ChIP-Seq also identified Ikaros binding in the promoter region of methylenetetrahydrofolate reductase (MTHFR), a gene involved in folate metabolism.

This pathway is important for the proper function of DNA synthesis and repair. Improper activation of MTHFR could affect the sensitivity of leukemia cells to treatment with methotrexate (MTX), which targets the folate metabolism pathway (90).

BCL2-Related Protein A1 (BCL2A1) is a member of the BCL-2 family that was identified as a potential Ikaros target gene by ChIP-Seq. This family of proteins is well known for their involvement in anti-apoptotic pathways with implications in development and tumorigenesis. BCL2A1 is able to reduce the release of pro-apoptotic cytochrome c from mitochondria and to block caspase activation. Furthermore, overexpression of

BCL2A1 has been shown to promote hematopoietic transformation and prevent apoptosis in vivo and should be considered a proto-oncogene for myeloid leukemogenesis (91).

Binding of Ikaros to the promoters of Distal-Less Homeobox 1 and 2 (DLX1 and

DLX2) was also observed via ChIP-Seq. Both DLX1 and DLX2 are downstream targets of fms-like tyrosine kinase-3 (FLT3), which is frequently mutated in AML and is a negative prognostic marker. Activation of DLX1 and DLX2 by FLT3 has been linked to

25

the MAPK/ERK signaling pathway, which provides a potential mechanism for the involvement of Ikaros in regulation of this pathway (92).

26

Tables

Table 1.1: World Health Organization classification of AML. (Modified from (93)). AML with recurrent genetic abnormalities:  AML with t(8;21)(q22;q22); RUNX1-RUNX1T1  AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBEB-MYH11  Acute promyelocytic leukemia (APL) with t(15;17)(q22;q12); PML-RARA  AML with t(9;11)(p22;q23); MLLT3-MLL  AML with t(6;9)(p23;q34); DEK-NUP214  AML with inv(3)(q21q26.2) or t(3;3)(q21;q26.2); RPN1-EVI1  AML (megakaryoblastic) with t(1;22)(p13;q13); RBM15-MKL1  Provisional entity: AML with mutated NPM1  Provisional entity: AML with mutated CEBPA AML with myelodysplasia-related change Therapy-related myeloid neoplasms AML, not otherwise specified:  Undifferentiated AML (M0)  AML with minimal differentiation (M1)  AML without maturation (M2)  AML with maturation (M2)  Acute myelomonocytic leukemia (M3)  Acute monoblastic/monocytic leukemia (M4)  Acute erythroid leukemia (M5)  Pure erythroid leukemia (M6)  Erythroleukemia, erythroid/myeloid (M6)  Acute megakaryoblastic leukemia (M7)  Acute basophilic leukemia  Acute panmyelosis with myelofibrosis Myeloid sarcoma Myeloid proliferations related to Down syndrome:  Transient abnormal myelopoiesis  Myeloid leukemia associated with Down syndrome Blastic plasmacytoid dendritic cell neoplasm

27

Table 1.2: French-American-British (FAB) staging system of AML. (Modified from (93)) FAB subtype Name Adult AML patients (%) M0 Undifferentiated acute myeloblastic leukemia 5%

M1 Acute myeloblastic leukemia with minimal maturation 15%

M2 Acute myeloblastic leukemia with maturation 25%

M3 Acute promyelocytic leukemia 10%

M4 Acute myelomonocytic leukemia 20%

M4eos Acute myelomonocytic leukemia with eosinophilia 5%

M5 Acute monocytic leukemia 10%

M6 Acute erythroid leukemia 5%

M7 Acute megakaryocytic leukemia 5%

28

Table 1.3: Chemical properties of CX-4945 (94).

CHEMICAL NAME 5-(3-chloroanilino)benzo[c][2,6]naphthyridine-8- carboxylic acid MOLECULAR FORMULA C19H12ClN3O2 MOLECULAR WEIGHT 349.7 g/mol SOLUBILITY Soluble in DMSO > 10 mM Ki VALUE 0.223 ± 0.011 nM STRUCTURE

29

Table 1.4: List of unique gene targets identified by ChIP-Seq of Ikaros in U937 cells.

IK Untreated C9ORF106 DUSP23 HUNK NAT15 RAMP3 SV2B A2LD1 C9ORF5 DYM HYDIN NBPF1 RAPGEF4 SYT15 ABCC10 CA12 DYNC1I1 ICAM1 NBPF10 RASA3 SYT2 ABCC12 CALM3 DYNC1I2 ID1 NBPF15 RASGRF1 TAF1B ABCG1 CAMK2B DZIP1 IER3 NCK2 RASSF2 TAOK2 ABLIM2 CAMK2N1 EBF2 IFNAR1 NCRNA00028 RASSF5 TBC1D2B ACAT2 CAPN13 EBF4 IFNGR1 NCRNA00164 RB1CC1 TBC1D3F ACOXL CAPN2 EFCAB8 IGSF21 NEFL RBCK1 TBCC ADAMTS17 CCDC149 EIF2C4 IKBKB NEGR1 RCBTB2 TBL1X ADAMTSL3 CCDC9 EIF3A IKZF1 NENF REM1 TCF15 ADD2 CCT6P1 EIF3L IL10 NEURL3 REXO1L2P TCF4 ADNP2 CD33 ELL IL11 NFIA RFX7 TDRD3 ADORA1 CD93 ELOF1 IL16 NFIX RGMA TFCP2L1 AGBL1 CDC27 EMILIN1 IL1R2 NLGN1 RGS8 TFDP1 AGPS CDC2L5 EML2 IL6R NOD1 RHBDD2 TFEB AHDC1 CDC5L EML6 IMPAD1 NOL10 RHBG TGS1 AIG1 CDCA7 EPHA2 INHBB NOTCH3 RNF112 THBD ALK CDH20 EPS15L1 INPP4A NOX5 RNF24 THBS2 ANGPT4 CDK5RAP1 ERICH1 INTS4 NPHP4 RNPEP THSD4 ANKRD10 CHODL ESPNP IPP NPTN RP51022P6.2 TIFAB ANKRD13B CHRNA9 ESR1 ITGA11 NSFL1C RPH3AL TM4SF4 ANKRD44 CHST12 ETS1 ITGA6 NUFIP1 RPTN TM9SF4 AP3D1 CIAO1 EYA4 IYD NWD1 RRBP1 TMC3 APBA2 CKAP5 FAHD2B JMJD5 NXT1 RREB1 TMCO3 APC2 CLDN10 FAM103A1 KBTBD8 OBFC2A RSPH1 TMED3 ARGLU1 CLDN23 FAM10A4 KCND3 OBSCN RSPO4 TMEFF2 ARHGAP30 CLEC4M FAM110A KCNJ12 OCEL1 RUNX2 TMEM106A ARHGEF10 CLYBL FAM123C KHDRBS3 OGDH RYBP TMEM121 ARHGEF10L CMKLR1 FAM136A KIAA0146 ONECUT1 RYR3 TMEM18 ARHGEF18 CNDP2 FAM155A KIAA1199 ONECUT2 S100P TMEM51 ARHGEF2 CNNM4 FAM20C KIAA1383 OPRD1 SBK2 TMTC4 ARHGEF7 CNR1 FAM27L KIAA1875 OR4C46 SCN11A TNIP2 ARNT2 CNTNAP2 FAM41C KIF14 OR4F15 SDC3 TNIP3 ASXL1 COG3 FAM66E KIF3B OR4F6 SDCBP2 TNRC18 ATG4C COL22A1 FAM70B KIF6 OR5W2 SDK1 TP53TG3B ATP11A COL23A1 FAM72D KIR2DS4 OXGR1 SEC22B TPP2 ATP4A COL4A1 FANK1 KIR3DL3 PABPCP2 SECISBP2L TPTE ATP7B COL4A2 FBXO15 KLF2 PAQR8 SELENBP1 TPX2 ATP9B COMMD7 FBXO42 KLHL29 PARN SERINC2 TRAM2 AXL COX4I2 FEM1B LACTB PAX7 SERPINC1 TRERF1 B2M CPNE5 FHAD1 LBXCOR1 PCCA SF3B1 TRIB3 B3GNT2 CPXM1 FLJ37307 LCTL PCDH8 SFRP1 TRIM44 BAGE CROCC FNDC1 LELP1 PCID2 SHKBP1 TRIM62 BASE CROCCL1 FOXA2 LGR6 PCTK3 SIAH3 TRIO BCL2 CRYAA FOXB1 LILRA3 PDE11A SIDT2 TRNP1 BCL2L1 CSGALNACT1 FOXD2 LIN7B PDE1A SIGLEC1 TSEN2 BCL8 CST1 FOXK1 LINGO1 PDE4A SIRPA TSPYL3 BDH1 CST11 FOXP4 LMX1A PDE4DIP SIRPB2 TSSC1 30

BLNK CST2 FOXS1 LRCH1 PDGFRL SIRPD TTLL9 BNC1 CST3 FRG1 LRFN2 PDK1 SKA2 TTN BPIL1 CST4 FRG1B LRPAP1 PDRG1 SLC10A2 TUBB4Q BPIL3 CST5 FRG2C LRRC2 PDXK SLC13A3 U2AF2 BRE CST8 FYCO1 LRRN2 PDYN SLC23A2 UBASH3A BSN CST9 GALNT13 MAN1A1 PEPD SLC25A12 UBR2 BTN2A1 CST9L GALNT2 MAN2A2 PIF1 SLC25A38 UMODL1 C10ORF93 CSTL1 GGCT MAP1D PKP1 SLC27A1 UNC45B C12ORF26 CSTT GGTLC1 MAP2K3 PLA2G5 SLC28A1 UQCRC1 C13ORF16 CT45A1 GLDN MAPRE1 PLAGL2 SLC2A7 USP14 C13ORF18 CUL1 GLI2 MBNL2 PLCL1 SLC41A3 UST C13ORF35 CWH43 GLIS1 MCF2L PLEKHM2 SLC6A19 UTP14C C15ORF26 CYP19A1 GLS MCM3 PLUNC SLC6A6 VAV1 C15ORF27 CYTL1 GNB5 MCM6 PMF1 SLCO5A1 VAV3 C17ORF51 CYTSB GNG7 MED26 PMS2L11 SMAD3 VEGFA C1ORF21 DCLK3 GPATCH1 MEGF11 POFUT1 SMAD6 VGLL4 C1ORF38 DCUN1D2 GPC5 MEMO1 POLR1C SMOC2 VIPR1 C1ORF96 DDX1 GPHN MFN2 POTEA SMOX VPS28 C20ORF112 DEFB115 GPI MIR663 POTEB SNRPA1 WDR75 C20ORF114 DEFB116 GPR156 MLN PPP1R16A SNTA1 WIPF1 C20ORF160 DEFB118 GPR176 MMEL1 PPP4R1L SNTG2 XKR4 C20ORF185 DEFB119 GPRIN2 MRPL14 PPYR1 SNX16 XKR7 C20ORF186 DEFB123 GRHL3 MSH6 PRAGMIN SNX33 XPO6 C20ORF30 DEFB132 GRIK3 MSTN PRDM16 SOCS6 ZAK C20ORF56 DFFB GRINA MSTP2 PRIM2 SORCS2 ZAP70 C20ORF70 DHRS12 GRM4 MTHFS PRKCB SOX1 ZBTB17 C20ORF71 DHX34 GTF2F2 MTIF3 PRKDC SOX13 ZBTB7A C21ORF121 DIAPH3 GULP1 MTUS1 PROZ SOX8 ZDHHC14 C21ORF82 DIP2C GYPC MUC12 PRSS38 SPAG4L ZFP82 C2ORF43 DLEU2 GZF1 MUC5B PRTG SPATS1 ZFYVE27 C2ORF48 DLEU7 HAT1 MUC6 PRUNE2 SPERT ZNF167 C2ORF71 DLX1 HCK MYCBP2 PSMD14 SPN ZNF229 C5AR1 DLX2 HECW2 MYL3 PSTPIP1 SPRY3 ZNF254 C6ORF132 DNAH7 HGF MYLK2 PTPN4 SSBP3 ZNF35 C6ORF208 DNAJC27 HIVEP3 MYO16 PTPRA SSTR4 ZNF358 C6ORF223 DNMT3A HM13 MYO1B PTPRN2 ST3GAL1 ZNF595 C6ORF25 DNMT3B HPCAL1 MYO1D PTPRU ST8SIA5 ZNF629 C6ORF35 DOCK3 HS3ST4 MYO5A PUM2 STAG1 ZNF648 C7ORF10 DOCK9 HS6ST3 MYO5B PUS10 STAT4 ZNF707 C7ORF47 DOT1L HSH2D MYO5C RAB20 STK35 ZNF716 C8ORF42 DSCAM HSPA12B MYT1L RAB23 STK39 ZNF717 C8ORF71 DTNB HSPG2 NALCN RAB38 STRN ZSCAN20 C8ORF86 DUSP15 HTR3C NAPB RAD51L1 SULT1A1

Both C2ORF48 DIP2C GRHL3 MSTN PTPRN2 ST3GAL1 ABCC10 C5AR1 DLEU2 GRIK3 MSTP2 RAB20 STK35 ABCG1 C6ORF132 DLEU7 GRM4 MUC12 RAD51L1 STK39 ABLIM2 C6ORF223 DLX1 GZF1 MYLK2 RAPGEF4 STRN ADD2 C6ORF35 DLX2 HAT1 MYO16 RASA3 SYT15 ALK C7ORF47 DNAH7 HCK MYO1B RASGRF1 SYT2 31

ANGPT4 C9ORF106 DNMT3B HECW2 MYO5A RASSF2 TAOK2 ANKRD10 CA12 DOCK3 HIVEP3 MYO5C RASSF5 TBC1D2B ANKRD44 CAMK2B DOCK9 HM13 NALCN RBCK1 TBCC AP3D1 CCDC149 DSCAM HPCAL1 NAPB RCBTB2 TCF15 APBA2 CCT6P1 DTNB HS3ST4 NBPF1 REM1 TDRD3 ARGLU1 CD33 DUSP15 HS6ST3 NCRNA00164 RNF24 TFDP1 ARHGEF10L CD93 DYM HSPA12B NEURL3 RP5-1022P6.2 TFEB ARHGEF18 CDC5L DYNC1I2 HYDIN NFIA RPH3AL THSD4 ARHGEF7 CDCA7 EBF4 ICAM1 NOL10 RSPO4 TM9SF4 ARNT2 CHRNA9 EFCAB8 ID1 NOTCH3 RUNX2 TMC3 ASXL1 CNDP2 ELL IER3 NOX5 RYBP TMED3 ATG4C CNTNAP2 EML2 IFNAR1 NSFL1C SDCBP2 TMEM18 ATP11A COG3 EPHA2 IGSF21 NXT1 SDK1 TNRC18 ATP7B COL22A1 EPS15L1 IKZF1 OBFC2A SEC22B TPP2 ATP9B COL4A1 ERICH1 IL16 OBSCN SECISBP2L TPTE B3GNT2 COL4A2 ESPNP IL1R2 OXGR1 SF3B1 TPX2 BAGE COMMD7 FAM110A IL6R PABPCP2 SIAH3 TRAM2 BASE COX4I2 FAM155A IMPAD1 PAQR8 SIGLEC1 TRERF1 BCL2L1 CPNE5 FAM20C IPP PCCA SIRPA TRIB3 BLNK CROCC FAM70B ITGA11 PCDH8 SIRPB2 TRIO BNC1 CROCCL1 FEM1B KCNJ12 PDE11A SKA2 TTLL9 BPIL1 CST1 FHAD1 KIAA1199 PDE4A SLC10A2 TTN BPIL3 CST11 FLJ37307 KIF6 PDE4DIP SLC23A2 TUBB4Q BRE CST2 FOXA2 KIR3DL3 PDRG1 SLC25A12 U2AF2 C13ORF16 CST3 FOXB1 KLF2 PDXK SLC25A38 UBR2 C13ORF18 CST4 FOXP4 KLHL29 PEPD SLC27A1 VAV1 C15ORF26 CST5 FOXS1 LBXCOR1 PIF1 SLC28A1 VAV3 C15ORF27 CST9 FRG1 LINGO1 PLCL1 SLC6A6 VEGFA C17ORF51 CST9L FRG1B LRCH1 PLUNC SMAD3 XKR7 C1ORF38 CSTT FRG2C LRPAP1 POFUT1 SMAD6 ZAK C20ORF112 CWH43 FYCO1 MAP1D PPYR1 SMOX ZBTB17 C20ORF114 CYP19A1 GALNT2 MAP2K3 PRDM16 SNTA1 ZBTB7A C20ORF160 DCUN1D2 GGTLC1 MAPRE1 PRIM2 SORCS2 ZDHHC14 C20ORF185 DDX1 GLI2 MCF2L PRKCB SOX1 ZFYVE27 C20ORF186 DEFB115 GLS MED26 PRTG SOX13 ZNF254 C20ORF30 DEFB116 GNB5 MEGF11 PRUNE2 SPAG4L ZNF595 C20ORF56 DEFB119 GNG7 MMEL1 PSMD14 SPN ZNF629 C20ORF70 DFFB GPI MRPL14 PTPN4 SSBP3 ZNF717 C20ORF71 DHRS12 GPRIN2 MSH6 PTPRA SSTR4 ZSCAN20 C21ORF82

CX-4945 Treated A4GALT AADAC ABCC10 ABCC2 ABCC4 ABCG1 ABHD5 ABLIM2 ABR ACBD5 ACBD6 ACN9 ACO2 ACOT11 ACOX3 ACSBG1 ACTL8 ACTL9 ACTN1 ACTR3 ADAM10 ADAM8 ADAM9 ADAMTS7 ADARB1 ADARB2 ADC ADCK1 ADCY3 ADD2 ADD3 ADIPOR1 ADORA3 ADPGK ADRA1D AFF3 AGBL4 AGRN AHRR AJAP1 AK5 AKAP13 AKIRIN1 AKNA AKR1A1 AKT1S1 AKT3

32

ALAS1 ALDH16A1 ALDH1A2 ALDH1A3 ALDH4A1 ALG8 ALK ALKBH7 AMD1 AMPD2 ANAPC1 ANGPT4 ANK1 ANKLE2 ANKRD10 ANKRD33B ANKRD44 ANKS1A ANTXR1 ANXA2 AP2A1 AP2S1 AP3D1 AP3S2 AP4E1 APBA2 APBB2 APCDD1 APH1B ARC ARF3 ARGLU1 ARHGAP11B ARHGAP19 ARHGAP25 ARHGAP26 ARHGEF1 ARHGEF10L ARHGEF11 ARHGEF16 ARHGEF18 ARHGEF19 ARHGEF3 ARHGEF7 ARID1A ARID3B ARID5A ARL11 ARMC5 ARMC6 ARNT2 ARPC1A ARPC5 ARPP19 ARRDC2 ASAP3 ASB4 ASF1B ASXL1 ASXL2 ATG4C ATG7 ATL2 ATP10A ATP11A ATP1A3 ATP2B2 ATP4B ATP5H ATP5J ATP6V0A2 ATP6V1B2 ATP6V1D ATP6V1E2 ATP7B ATP8B2 ATP9B ATXN7L3 AURKAIP1 AZIN1 B3GALNT2 B3GNT2 B4GALT1 BAGE BAI2 BAK1 BANP BASE BAT2 BBC3 BCAS2 BCKDHA BCL11A BCL2A1 BCL2L1 BCL2L14 BCL9 BCLAF1 BET1 BIN1 BIN3 BIRC6 BLK BLM BLNK BMF BMP2 BMP5 BNC1 BNIP1 BNIP2 BNIP3L BPI BPIL1 BPIL3 BRD2 BRE BRP44 BRUNOL6 BSDC1 BSG BTBD9 BTD BTF3L1 BYSL C10ORF12 C10ORF131 C10ORF28 C10ORF32 C10ORF55 C10ORF75 C11ORF45 C12ORF34 C13ORF16 C13ORF18 C14ORF147 C14ORF180 C14ORF80 C15ORF26 C15ORF27 C15ORF29 C15ORF33 C15ORF37 C15ORF39 C15ORF40 C15ORF41 C15ORF44 C15ORF50 C15ORF52 C15ORF59 C15ORF61 C16ORF5 C17ORF51 C18ORF1 C18ORF18 C18ORF22 C18ORF54 C18ORF55 C19ORF20 C19ORF23 C19ORF24 C19ORF39 C19ORF44 C1D C1ORF109 C1ORF115 C1ORF127 C1ORF159 C1ORF174 C1ORF190 C1ORF198 C1ORF231 C1ORF38 C1ORF49 C1ORF52 C1ORF58 C1ORF63 C1ORF71 C1ORF86 C1ORF95 C20ORF111 C20ORF112 C20ORF114 C20ORF141 C20ORF160 C20ORF185 C20ORF186 C20ORF194 C20ORF196 C20ORF197 C20ORF30 C20ORF54 C20ORF56 C20ORF7 C20ORF70 C20ORF71 C20ORF94 C20ORF96 C21ORF33 C21ORF34 C21ORF57 C21ORF58 C21ORF59 C21ORF82 C2ORF18 C2ORF34 C2ORF42 C2ORF48 C2ORF58 C2ORF60 C2ORF64 C2ORF76 C2ORF86 C2ORF90 C3ORF20 C3P1 C4ORF26 C4ORF34 C4ORF41 C4ORF44 C4ORF52 C5AR1 C5ORF28 C6ORF10 C6ORF103 C6ORF118 C6ORF127 C6ORF129 C6ORF130 C6ORF132 C6ORF138 C6ORF153 C6ORF223 C6ORF35 C7ORF11 C7ORF41 C7ORF47 C7ORF50 C7ORF55 C7ORF64 C7ORF72 C8ORF40 C8ORF46 C8ORF55 C8ORF75 C9ORF106 C9ORF30 C9ORF6 C9ORF91 CA12 CA14 CAB39L CACNA1C CACNA1D CACNA2D1 CACNA2D2 CACNB4 CALML4 CALN1 CALR CAMK2B CAMSAP1 CAMTA1 CAP1 CAPG CAPN11 CAPZA1 CARD8 CARKD CARM1 CARS2 CASQ2 CATSPERG CBLB CBX4 CCDC114 CCDC12 CCDC123 CCDC126 CCDC149 CCDC150 CCDC21 CCDC33 CCDC36 CCDC80 CCDC88A CCDC90A

33

CCDC94 CCDC96 CCDC97 CCND3 CCNE1 CCPG1 CCR2 CCR5 CCT4 CCT6P1 CD1C CD200R1L CD276 CD300A CD33 CD34 CD82 CD93 CDA CDAN1 CDC16 CDC25A CDC25B CDC42BPA CDC42EP3 CDC5L CDC73 CDCA7 CDGAP CDH17 CDK7 CDKN1A CEACAM1 CEACAM16 CEACAM21 CEACAM4 CEACAM6 CEBPA CEP350 CEP76 CERKL CFLAR CGB7 CHD2 CHD7 CHDH CHERP CHMP2A CHP CHRNA3 CHRNA9 CHRNB3 CHRNB4 CHST13 CHST8 CIB2 CIC CIDECP CILP CISH CITED2 CKAP2 CLASP2 CLC CLCA1 CLCA2 CLDND2 CLIP2 CLK3 CLN5 CLN8 CLPS CNDP2 CNN2 CNPY3 CNTNAP2 COBL COG3 COL15A1 COL18A1 COL22A1 COL27A1 COL4A1 COL4A2 COL6A1 COL6A2 COMMD4 COMMD7 COQ10B COQ5 CORO2B COX4I1 COX4I2 CPA3 CPAMD8 CPEB1 CPLX1 CPN1 CPNE5 CPSF1 CPZ CR2 CRABP1 CRCP CREB5 CREG2 CRLS1 CRMP1 CROCC CROCCL1 CRTC2 CRYZL1 CSK CSNK1G1 CSPG4 CST1 CST11 CST2 CST3 CST4 CST5 CST9 CST9L CSTB CSTT CTDP1 CTDSPL2 CTF1 CTNNB1 CTSE CTSH CTTNBP2NL CUEDC1 CUL4A CUL7 CUL9 CUX1 CWH43 CXCL13 CXCL2 CXCR4 CYBRD1 CYC1 CYP11A1 CYP19A1 CYP1B1 CYP26B1 CYP2A6 CYP39A1 CYP4F11 CYP4F22 CYSLTR2 CYTH4 CYYR1 DAB1 DAG1 DAGLB DAPK2 DARS DBI DCAF13 DCAF6 DCAF8 DCP1A DCUN1D2 DDAH1 DDB2 DDC DDIT4 DDRGK1 DDX1 DDX18 DECR1 DEDD2 DEFB110 DEFB115 DEFB116 DEFB119 DEFB121 DEFB124 DEFB129 DENND3 DENND4A DERL1 DFFB DGCR6 DGCR8 DGKQ DHDDS DHRS12 DIP2C DIRAS1 DIS3L DKFZP434J02 DKFZP566F0 DKFZP586I14 DLEC1 DLEU2 DLEU7 DLX1 DLX2 DMXL2 DNAH1 DNAH7 DNAJC11 DNAJC17 DNAJC3 DNASE1L3 DNMT3B DOCK3 DOCK9 DOK7 DONSON DPH2 DPP8 DPP9 DPRX DPYSL2 DSCAM DSCR3 DSTYK DTL DTNB DTWD1 DTX3L DUSP15 DUSP5 DUSP7 DUT DUX4 DVL1 DYM DYNC1I2 DYRK1A DYSF DYX1C1 E2F2 E2F6 EBF4 ECE1 ECHDC2 EDEM1 EDN2 EEF2K EEPD1 EFCAB2 EFCAB8 EFHD2 EFNB2 EFR3A EGLN2 EHBP1 EHD2 EHD4 EIF2AK1 EIF2B5 EIF2S2 ELFN1 ELL ELL3 ELMO1 ELOVL5 ELP3 ELP4 EML2 EML4 EMR2 ENPP5 ENTPD3 EPB41 EPHA2 EPHA6 EPHX1 EPM2AIP1 EPS15 EPS15L1 ERBB2IP ERCC1 ERCC5 ERI1 ERICH1 ERO1LB ESCO1 ESPNP ESYT2 ETFA ETFB

34

ETHE1 ETS2 ETV3 EXD1 EXO1 EXOG EXOSC10 EXOSC4 EYA3 EYS F10 F5 FABP5 FAH FAM100A FAM108C1 FAM110A FAM113A FAM120B FAM124A FAM125A FAM131C FAM155A FAM156B FAM167A FAM171B FAM173A FAM188B FAM189A1 FAM194B FAM20C FAM3B FAM3D FAM46C FAM49A FAM53A FAM5B FAM63B FAM69B FAM70B FAM71A FAM82A1 FAM82A2 FAM89A FAM91A1 FAM98A FARP1 FASN FBLN2 FBN1 FBXL22 FBXO2 FBXO22 FBXO25 FBXO32 FBXO9 FCAMR FCER2 FCGRT FCRLB FEM1B FERD3L FERMT1 FGD2 FGD5 FGF14 FGFR3 FGFRL1 FGL1 FGR FHAD1 FHL3 FKBP14 FKBP1A FKBP5 FKBP7 FLI1 FLJ22536 FLJ36777 FLJ37307 FLJ42393 FLJ43860 FLOT1 FLT1 FMNL2 FNDC3A FOSL2 FOXA2 FOXB1 FOXI2 FOXJ3 FOXK2 FOXN2 FOXP1 FOXP4 FOXS1 FPR3 FRG1 FRG1B FRG2C FRMD1 FRMD4A FRS3 FSHR FSIP1 FSTL3 FTL FXYD7 FYCO1 FZD1 FZR1 GADD45B GALK2 GALNT14 GALNT2 GALNT5 GAS5 GATAD2A GATSL3 GBA GBAP GCC2 GCM1 GCOM1 GFRA2 GGPS1 GGTLC1 GH1 GIPC1 GIPR GJB4 GKN2 GLCCI1 GLDC GLI2 GLI3 GLI4 GLRB GLS GLT25D1 GLTSCR1 GLUL GNB5 GNG7 GOLGA3 GOLGA6D GP1BB GPC6 GPI GPN2 GPR111 GPR116 GPR126 GPR132 GPR137B GPR141 GPR155 GPR180 GPR20 GPR78 GPRIN2 GPT2 GPX1 GPX4 GRHL2 GRHL3 GRID2IP GRIK3 GRIN2D GRINL1A GRM4 GRN GSDMD GSTA1 GSTA2 GTF2A2 GTF2F1 GTF2IRD1 GTF3A GTF3C3 GTPBP3 GUCA1A GUCY1B2 GUK1 GUSB GZF1 HACL1 HADHA HAT1 HAUS5 HBS1L HCCA2 HCG18 HCG27 HCK HCN4 HDAC1 HDGF HDGF2 HEATR1 HEATR5B HECA HECW1 HECW2 HES1 HHIPL2 HIBCH HIP1 HIST1H2AK HIST1H2BL HIVEP2 HIVEP3 HLA-B HM13 HMCN1 HMG20A HMGN2 HNRNPA2B1 HNRNPA3 HNRNPL HNRNPUL1 HOMER2 HOXA11 HOXD1 HOXD11 HOXD3 HPCAL1 HPSE2 HS1BP3 HS3ST4 HS6ST1 HS6ST3 HSPA12B HSPA1B HSPBP1 HSPE1 HTRA3 HTT HYDIN ICA1 ICAM1 ICAM4 ICK ICOSLG ID1 IER2 IER3 IFFO2 IFI16 IFI30 IFIH1 IFNAR1 IFNAR2 IGDCC3 IGFBP3 IGSF21 IKZF1 IL12B IL12RB1 IL16 IL17RA IL1F7 IL1R2 IL1RL1 IL1RL2 IL28RA IL6R IMPA2 IMPAD1 ING1 INO80C INO80E INPP1 INTS6 IP6K3 IPO4 IPO5 IPP IRAK2 IRF2 IRF2BP2 IRS2 ITGA11 ITGA9 ITGAL ITGB2 ITPA

35

ITPKB ITPR1 ITPR3 ITPRIP JAG1 JAK1 JPH3 JSRP1 JUN KCNF1 KCNG2 KCNH1 KCNJ12 KCNJ15 KCNJ5 KCNJ6 KCNK12 KCNN3 KCNN4 KCTD12 KCTD15 KDM1A KDM4B KDM6B KDSR KIAA0196 KIAA0232 KIAA0240 KIAA0427 KIAA0895 KIAA0907 KIAA0913 KIAA1024 KIAA1199 KIAA1429 KIAA1704 KIAA1715 KIAA1919 KIAA1949 KIAA1967 KIAA2018 KIF21B KIF22 KIF26A KIF26B KIF6 KIFAP3 KIR3DL3 KLC1 KLF10 KLF12 KLF13 KLF17 KLF2 KLHDC9 KLHL18 KLHL29 KLK11 KLK13 KPNA3 KREMEN1 KRT15 KRT7 KRTAP11-1 KRTCAP2 LAD1 LAGE3 LAIR1 LAMA5 LAMB3 LAMP1 LAPTM5 LARP4B LARP6 LASP1 LASS4 LBH LBR LBXCOR1 LDHAL6B LDLR LDLRAP1 LECT1 LEO1 LFNG LGTN LHFPL4 LHFPL5 LIAS LIG4 LILRA5 LIMD1 LIMD2 LIMK1 LINGO1 LIPC LIX1L LMCD1 LMO7 LMX1B LOH12CR1 LONRF2 LOXL1 LOXL2 LPA LPAL2 LPHN1 LPIN1 LPPR3 LRCH1 LRG1 LRP1 LRP1B LRPAP1 LRRC1 LRRC37A3 LRRC3B LRRC40 LRRC55 LSM10 LSM14B LSM5 LYPLAL1 LYZL4 LZTS2 MAD1L1 MAD2L1BP MAEA MAGEL2 MAGI1 MALAT1 MALL MAN1C1 MANSC1 MAP1D MAP1LC3B MAP1LC3C MAP1S MAP2K1 MAP2K3 MAP2K5 MAP2K7 MAP3K5 MAP4K3 MAPK13 MAPK4 MAPK6 MAPKAPK3 MAPRE1 MAPRE2 MARCO MARK1 MARS2 MAST3 MBD3 MBD5 MBOAT7 MBP MCF2L MCM7 MCM8 MCTP2 MDFI MDM4 MEA1 MED20 MED23 MED26 MED29 MED4 MEF2D MEGF11 MEGF6 MEPE MERTK MESDC1 MESDC2 METTL13 METTL9 MEX3A MFSD10 MGA MGAT5 MGC13005 MGC15885 MGC27382 MIB1 MICA MID1IP1 MIDN MIIP MIR146B MIR148A MIR17HG MIR184 MIR196B MIR23A MIR34A MIR802 MIXL1 MLL3 MLLT1 MLPH MMADHC MMEL1 MNS1 MOBKL2A MORC3 MORF4L1 MORN1 MPEG1 MR1 MRFAP1L1 MRPL14 MRPL17 MRPL30 MRPL54 MRPL9 MRPS12 MRPS14 MRPS18A MRPS18B MRPS24 MRPS6 MRVI1 MS4A15 MSH5 MSH6 MSRA MSTN MSTP2 MSX1 MTCH1 MTFR1 MTHFD1L MTHFR MTIF2 MTMR14 MTMR15 MTMR7 MTOR MTX2 MUC1 MUC12 MUC2 MUC20 MUM1 MUS81 MX2 MXI1 MYADM MYADML MYCL1 MYCN MYD88 MYH9 MYL10 MYLK2 MYO16 MYO1B MYO5A MYO5C MYOM2 NACC2 NAGK NALCN NANP NAPA NAPB NARS NAV1 NBN NBPF1 NBPF3 NCALD NCAPG2 NCDN NCF2 NCR2 NCRNA00051 NCRNA00095 NCRNA00111 NCRNA00164 NDUFA13 NDUFS5 NDUFS7 NEDD4

36

NEDD4L NEU1 NEURL NEURL3 NFATC1 NFE2L2 NFIA NFKB2 NFKBIA NFKBID NFKBIE NFKBIL1 NFYA NGLY1 NHLH2 NID1 NISCH NKAPL NKG7 NKTR NLRP12 NLRP3 NMNAT2 NOC4L NOL10 NOLC1 NOP56 NOP58 NOSIP NOTCH2 NOTCH2NL NOTCH3 NOTCH4 NOX5 NPAS2 NPC1L1 NQO2 NR1D2 NR1H2 NR2C2AP NRG1 NRSN2 NSFL1C NUAK2 NUDT1 NUDT15 NUDT19 NUF2 NUP85 NXT1 OAZ1 OBFC1 OBFC2A OBSCN ODZ3 OLA1 OLFM2 OLFM4 OLFML2B OLFML3 OLIG1 OPRM1 OPTC OR6K6 OR9Q1 OSBP OSBPL10 OSM OTOF OXGR1 PABPC1 PABPCP2 PACS2 PADI1 PADI4 PAIP2B PAK1IP1 PANK1 PANK4 PAQR8 PARD6G PARP6 PAX2 PBRM1 PCBP1 PCCA PCDH8 PCDH9 PCGEM1 PCIF1 PCM1 PCMT1 PCNT PCNXL2 PCSK9 PDCD11 PDCL3 PDE11A PDE4A PDE4DIP PDE6H PDE7B PDGFB PDIA6 PDRG1 PDXDC1 PDXK PDZK1IP1 PDZRN3 PEBP4 PEPD PEX3 PFKFB4 PGLYRP3 PHACTR3 PHF15 PHGDH PHLDA3 PHTF2 PIAS3 PIBF1 PIF1 PIGG PIH1D1 PIK3CD PIK3R5 PITX3 PKD1L1 PKD2L1 PLA2G2F PLA2G4A PLA2G4F PLB1 PLCD1 PLCL1 PLEKHF1 PLEKHF2 PLEKHG1 PLEKHH2 PLEKHJ1 PLEKHO1 PLEKHO2 PLK1S1 PLK3 PLTP PLUNC PLXNA2 PLXNB1 PMAIP1 PML PMS1 PMS2L3 PMVK PNMA2 POFUT1 POGZ POLH POLR1B POLRMT POMGNT1 POR POU2F1 PPAP2B PPAPDC3 PPARD PPARG PPARGC1A PPCDC PPIB PPIL2 PPM1B PPM1H PPP1CB PPP1R8 PPP1R9B PPP2R1A PPP2R2D PPP3R1 PPRC1 PPYR1 PRAMEF2 PRC1 PRDM16 PRELP PREX1 PRIM2 PRKACA PRKCB PRKCD PRKCE PRKCSH PRKCZ PRMT2 PRND PRNP PROK2 PROM2 PRPF40B PRPS1L1 PRR14 PRR20C PRR5L PRTG PRUNE2 PSCA PSG10 PSG8 PSMA4 PSMD14 PSMF1 PSMG1 PSPH PTAFR PTCD1 PTCHD2 PTDSS2 PTGFRN PTGIR PTGIS PTK2B PTK7 PTMA PTP4A2 PTPN1 PTPN12 PTPN4 PTPRA PTPRG PTPRN2 PUM1 PURB PUS7 PVR PVT1 PXDN PYCRL QKI QSOX1 RAB10 RAB11A RAB20 RAB27A RAB33B RAB3A RAB3B RAB42 RAB43 RAB4A RAB8A RAB8B RAD23B RAD51L1 RAF1 RALA RALB RANBP3 RAP1A RAP1GAP RAP2A RAPGEF4 RARB RASA3 RASGRF1 RASIP1 RASSF2 RASSF5 RB1 RBBP6 RBBP8 RBCK1 RBM12B RBM14 RBM16 RBM27 RBM47 RBMS2 RBPJ RCAN1 RCAN2 RCBTB2 RCN2 RCSD1 RDH5 REL REM1 REPS2 RER1

37

RERE RETN RFC2 RFT1 RFX2 RFX5 RGL1 RGS12 RGSL1 RHCG RHOC RHPN1 RIC8B RIMS2 RIOK3 RIPK4 RNASE3 RNASEH2B RNF123 RNF126 RNF144A RNF149 RNF167 RNF187 RNF19B RNF216 RNF24 RNF5P1 RORA RP5-1022P RPA3 RPH3AL RPL18AP3 RPL23 RPL28 RPL7L1 RPLP1 RPPH1 RPRD2 RPS10P7 RPS15 RPS16 RPS20 RPS6KA1 RPS6KA2 RPS9 RPTOR RRAGC RRM2 RRP12 RSL24D1 RSPO4 RTEL1 RTF1 RTTN RUNDC2C RUNX1 RUNX2 RUNX3 RXRA RYBP S100A10 S100A11 S100A3 S100A9 SACM1L SAP130 SARS2 SBDSP SBNO2 SBSN SCAF1 SCAMP1 SCAMP2 SCAMP4 SCAMP5 SCARB1 SCCPDH SCG3 SCN10A SCRN1 SCRT2 SCUBE3 SDCBP2 SDCCAG8 SDHB SDK1 SEC22B SECISBP2L SEMA3B SEMA3F SEMA4A SEMA6D SEMA7A SENP7 SEPN1 SERPINB8 SERPINE1 SERPINE3 SERTAD1 SETD1A SETD4 SF3B1 SF3B2 SFN SFRS2IP SFRS7 SGMS1 SGTA SH2B2 SH3BP2 SH3BP5 SH3GL1 SH3GL3 SH3PXD2A SH3RF3 SH3TC1 SH3TC2 SHARPIN SHC2 SHC4 SHQ1 SIAH2 SIAH3 SIGLEC1 SIGLEC7 SIK1 SIN3A SIPA1L3 SIRPA SIRPB2 SIRPG SIRT6 SIX2 SIX5 SKA2 SKI SLA SLAIN1 SLAMF1 SLC10A2 SLC17A7 SLC19A1 SLC1A2 SLC1A3 SLC1A4 SLC1A5 SLC22A2 SLC23A2 SLC25A12 SLC25A33 SLC25A35 SLC25A37 SLC25A38 SLC26A8 SLC27A1 SLC27A2 SLC28A1 SLC29A1 SLC2A1 SLC2A5 SLC30A6 SLC33A1 SLC35B2 SLC35C1 SLC35E1 SLC35F5 SLC39A10 SLC39A3 SLC41A1 SLC43A3 SLC45A1 SLC45A4 SLC4A1AP SLC4A7 SLC5A5 SLC6A11 SLC6A16 SLC6A20 SLC6A6 SLC7A10 SLC7A5P1 SLC8A2 SLC9A8 SLC9A9 SLIT1 SLITRK1 SLMAP SMAD2 SMAD3 SMAD6 SMAP2 SMARCC1 SMEK2 SMG7 SMOX SMYD3 SNHG5 SNIP SNORD115 SNORD116 SNPH SNRNP40 SNRPA SNRPB SNRPE SNRPN SNTA1 SNTB1 SNUPN SNX10 SNX22 SNX8 SOCS5 SORBS1 SORCS2 SOS1 SOX1 SOX13 SOX7 SP3 SPAG17 SPAG4L SPAST SPATA12 SPATC1 SPEN SPG11 SPG21 SPINT1 SPN SPNS1 SPOPL SPPL2A SPRY1 SPRY2 SPSB1 SQRDL SRBD1 SRCAP SRF SRGAP3 SRI SRPK1 SRRM3 SSBP1 SSBP3 SSC5D SSH1 SSTR4 ST18 ST20 ST3GAL1 ST3GAL3 ST6GAL2 ST7L STAB1 STARD13 STARD3 STARD5 STAT1 STC2 STEAP3 STK24 STK32B STK35 STK39 STRA6 STRN STX4 STX7 STXBP2 STXBP5 SUCLA2 SULF2 SULT1C4 SUMF2 SUPT3H SUPT5H SUZ12P SYF2 SYT12 SYT15 SYT2 SYT7 SYTL3 TACC3

38

TACSTD2 TADA1 TAF12 TANK TAOK2 TAPBP TARP TBC1D14 TBC1D20 TBC1D21 TBC1D2B TBC1D4 TBCC TBKBP1 TBX15 TBX6 TCF15 TCF3 TCP1 TDRD3 TDRD5 TECR TEDDM1 TESSP2 TEX2 TEX264 TEX9 TFAP2E TFB2M TFDP1 TFEB TFF3 TFRC TG TGDS TGFA TGFB1 TGM2 TGM3 TGOLN2 THAP10 THBS3 THRAP3 THRB THSD1 THSD1P THSD4 THUMPD3 TIA1 TICAM1 TIMM13 TINAG TIPIN TJAP1 TKT TLCD1 TLE3 TLL2 TLN2 TLR5 TLX1 TM6SF1 TM7SF4 TM9SF3 TM9SF4 TMC2 TMC3 TMCO1 TMED3 TMEM106B TMEM116 TMEM130 TMEM131 TMEM132B TMEM17 TMEM18 TMEM181 TMEM201 TMEM63B TMEM67 TMEM80 TMEM85 TMEM87A TMEM87B TMEM90B TMEM97 TMOD2 TMOD3 TNFAIP8L2 TNFAIP8L3 TNFRSF11A TNFRSF14 TNFRSF21 TNFRSF8 TNFSF12 TNFSF4 TNFSF9 TNNI3 TNPO1 TNR TNRC18 TNRC6A TOMM20 TOMM6 TOP1MT TOR1AIP1 TOR3A TP73 TPD52L2 TPM1 TPP2 TPRG1L TPST1 TPT1 TPTE TPX2 TRA2A TRAF3IP3 TRAF4 TRAIP TRAM2 TRAPPC3 TREM1 TREM2 TREML2 TREML3 TREML4 TRERF1 TRIB1 TRIB3 TRIM27 TRIM28 TRIM54 TRIM8 TRIO TRIP4 TRMT1 TRPM1 TRPM2 TRPM7 TRPV3 TRRAP TSEN34 TSHZ3 TSPAN1 TSPAN17 TSPAN18 TTBK1 TTC13 TTC15 TTC21B TTC3 TTC30A TTC32 TTC35 TTC7A TTLL10 TTLL9 TTN TTYH3 TUBB4Q TXLNA TXNDC12 TXNDC9 TYK2 TYRO3P TYROBP U2AF2 UACA UBA2 UBA7 UBAC2 UBD UBE2M UBE2Q1 UBE2QP1 UBE2U UBE4B UBIAD1 UBN2 UBP1 UBR2 UBR3 UBR4 UBR5 UBTD1 UBXN2A UCK2 UCP2 UGGT2 UHRF2 ULK3 UNC5CL UPP1 UROD USMG5 USP16 USP17 USP20 USP24 USP3 USP4 USP48 USP49 USP8 USP9X USPL1 VAC14 VASH2 VASP VAV1 VAV3 VAX2 VCPIP1 VEGFA VIPR2 VIT VKORC1 VPS11 VPS13D VPS53 VPS54 WAC WARS2 WBSCR28 WDFY2 WDR1 WDR27 WDR45L WDR51A WDR6 WDR60 WDR61 WDR7 WDR73 WDR74 WDR82 WDR88 WDYHV1 WNT2B WNT7A WRB XAF1 XCL2 XCR1 XKR6 XKR7 XRN2 YPEL3 YTHDF1 YWHAZ YY1AP1 ZAK ZBTB17 ZBTB2 ZBTB48 ZBTB7A ZBTB7B ZC3H12A ZC3H12C ZC3H13 ZC3H15 ZC3H3 ZC3H6 ZCCHC11 ZCCHC2 ZCCHC4 ZDHHC14 ZFAND2A ZFAND6 ZFHX3 ZFP36 ZFP36L2 ZFYVE26 ZFYVE27 ZFYVE28 ZKSCAN2 ZMPSTE24 ZMYND8 ZNF114 ZNF193 ZNF230 ZNF236 ZNF25 ZNF251 ZNF252 ZNF254

39

ZNF263 ZNF283 ZNF320 ZNF33A ZNF343 ZNF445 ZNF446 ZNF45 ZNF48 ZNF550 ZNF579 ZNF583 ZNF595 ZNF627 ZNF629 ZNF643 ZNF646 ZNF676 ZNF683 ZNF687 ZNF692 ZNF697 ZNF7 ZNF710 ZNF717 ZNF747 ZNF773 ZNF780A ZNF799 ZNF8 ZNF804A ZNF816A ZNF828 ZNF83 ZNF831 ZNF843 ZNF93 ZNF98 ZNHIT1 ZSCAN20 ZSCAN22

40

Figures

Figure 1.1. The involvement of CK2 in cellular pathways that promote cell survival. (a) CK2 is involved in the MAPK/p53 and NF-κB pathways to promote cell survival in response to stress. (b) CK2 promotes the Wnt/β-catenin pathway and inactivates Bid to promote cell survival. (c) CK2 upregulates the Ras growth factor pathway to promote cell survival. (Modified from (14)).

41

N-terminal C-terminal

Figure 1.2. Ikaros can be phosphorylated at several different CK2 consensus sequences. CK2 phosphorylates Ikaros and regulates its function and subcellular localization. Depicted here are phosphorylation sites on the Ikaros protein that are phosphorylated by CK2 (amino acids 13, 23, 63, 101, 294, and 389-398). (Modified from (95)).

42

Figure 1.3. Modification of histone tails is important for regulation of gene expression. Histone tails can be modified by acetylation, methylation, ubiquination, sumoylation, and phosphorylation. Of particular interest in this dissertation are the effects of acetylation and methylation of histone tails. (Modified from (79)).

43

Treatment of U937 cells with CX-4945 for 72h 1.2 1 0.8 0.6 0.4 0.2 0

Figure 1.4. Cytotoxicity assay for CX-4945 in U937 cells. U937 cells were treated with various concentrations of CX-4945 for 72 hours and cellular proliferation was measured by spectrophotometry using WST1 reagent.

44

genes genes genes

Figure 1.5. Comparison of unique genes bound by Ikaros in untreated and CX-4945 treated U937 cells by ChIP-Seq analysis. Ikaros peaks identified by ChIP-Seq were mapped to gene regions and duplicates were removed to determine the number of unique genes bound by Ikaros in each condition. U937 cells were treated with 10 µM CX-4945 for 72 hours.

45

Ikaros Peak Distribution Relative to Transcription Start Sites 160 140 120 100 80 60 40 20 0 -5000 -4000 -3000 -2000 -1000 0 1000 2000 3000 4000 5000 Position to TSS (bp)

WT CX-4945

Figure 1.6. Ikaros binding peaks are redistributed to the transcription start site (TSS) following CX-4945 treatment. Comparison of Ikaros binding at the TSS in untreated and CX-4945 treated cells. ChIP-Seq revealed that treatment of U937 cells with 10 µM CX-4945 for 72 hours resulted in a redistribution of Ikaros to the TSS of genes.

46

References

1. Mrózek, K., Heerema, N. A., and Bloomfield, C. D. (2004) Cytogenetics in acute leukemia. Blood Rev 18, 115-136 2. Marcucci, G., Mrózek, K., and Bloomfield, C. D. (2005) Molecular heterogeneity and prognostic biomarkers in adults with acute myeloid leukemia and normal cytogenetics. Curr Opin Hematol 12, 68-75 3. Gilliland, D. G., and Tallman, M. S. (2002) Focus on acute leukemias. Cancer Cell 1, 417-420 4. Dash, A., and Gilliland, D. G. (2001) Molecular genetics of acute myeloid leukaemia. Best Pract Res Clin Haematol 14, 49-64 5. Castilla, L. H., Garrett, L., Adya, N., Orlic, D., Dutra, A., Anderson, S., Owens, J., Eckhaus, M., Bodine, D., and Liu, P. P. (1999) The fusion gene Cbfb-MYH11 blocks myeloid differentiation and predisposes mice to acute myelomonocytic leukaemia. Nat Genet 23, 144-146 6. Mehrotra, B., George, T. I., Kavanau, K., Avet-Loiseau, H., Moore, D., Willman, C. L., Slovak, M. L., Atwater, S., Head, D. R., and Pallavicini, M. G. (1995) Cytogenetically aberrant cells in the stem cell compartment (CD34+lin-) in acute myeloid leukemia. Blood 86, 1139-1147 7. Slovak, M. L., Kopecky, K. J., Cassileth, P. A., Harrington, D. H., Theil, K. S., Mohamed, A., Paietta, E., Willman, C. L., Head, D. R., Rowe, J. M., Forman, S. J., and Appelbaum, F. R. (2000) Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 96, 4075-4083 8. Vardiman, J. W., Harris, N. L., and Brunning, R. D. (2002) The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 100, 2292-2302 9. Greenberg, P., Cox, C., LeBeau, M. M., Fenaux, P., Morel, P., Sanz, G., Sanz, M., Vallespi, T., Hamblin, T., Oscier, D., Ohyashiki, K., Toyama, K., Aul, C., Mufti, G., and Bennett, J. (1997) International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 89, 2079-2088 10. Robak, T., and Wierzbowska, A. (2009) Current and emerging therapies for acute myeloid leukemia. Clin Ther 31 Pt 2, 2349-2370 47

11. Münstermann, U., Fritz, G., Seitz, G., Lu, Y. P., Schneider, H. R., and Issinger, O. G. (1990) Casein kinase II is elevated in solid human tumours and rapidly proliferating non- neoplastic tissue. Eur J Biochem 189, 251-257 12. Faust, R. A., Niehans, G., Gapany, M., Hoistad, D., Knapp, D., Cherwitz, D., Davis, A., Adams, G. L., and Ahmed, K. (1999) Subcellular immunolocalization of protein kinase CK2 in normal and carcinoma cells. Int J Biochem Cell Biol 31, 941-949 13. Allende, J. E., and Allende, C. C. (1995) Protein kinases. 4. Protein kinase CK2: an enzyme with multiple substrates and a puzzling regulation. FASEB J 9, 313-323 14. Ahmed, K., Gerber, D. A., and Cochet, C. (2002) Joining the cell survival squad: an emerging role for protein kinase CK2. Trends Cell Biol 12, 226-230 15. Ruzzene, M., Penzo, D., and Pinna, L. A. (2002) Protein kinase CK2 inhibitor 4,5,6,7- tetrabromobenzotriazole (TBB) induces apoptosis and caspase-dependent degradation of haematopoietic lineage cell-specific protein 1 (HS1) in Jurkat cells. Biochem J 364, 41-47 16. Slaton, J. W., Unger, G. M., Sloper, D. T., Davis, A. T., and Ahmed, K. (2004) Induction of apoptosis by antisense CK2 in human prostate cancer xenograft model. Mol Cancer Res 2, 712-721 17. Kim, J. S., Eom, J. I., Cheong, J. W., Choi, A. J., Lee, J. K., Yang, W. I., and Min, Y. H. (2007) Protein kinase CK2alpha as an unfavorable prognostic marker and novel therapeutic target in acute myeloid leukemia. Clin Cancer Res 13, 1019-1028 18. Landesman-Bollag, E., Channavajhala, P. L., Cardiff, R. D., and Seldin, D. C. (1998) p53 deficiency and misexpression of protein kinase CK2alpha collaborate in the development of thymic lymphomas in mice. Oncogene 16, 2965-2974 19. Guo, C., Yu, S., Davis, A. T., Wang, H., Green, J. E., and Ahmed, K. (2001) A potential role of nuclear matrix-associated protein kinase CK2 in protection against drug-induced apoptosis in cancer cells. J Biol Chem 276, 5992-5999 20. Song, D. H., Sussman, D. J., and Seldin, D. C. (2000) Endogenous protein kinase CK2 participates in Wnt signaling in mammary epithelial cells. J Biol Chem 275, 23790-23797 21. Siddiqui-Jain, A., Bliesath, J., Macalino, D., Omori, M., Huser, N., Streiner, N., Ho, C. B., Anderes, K., Proffitt, C., O'Brien, S. E., Lim, J. K., Von Hoff, D. D., Ryckman, D. M., Rice, W. G., and Drygin, D. (2012) CK2 inhibitor CX-4945 suppresses DNA repair response triggered by DNA-targeted anticancer drugs and augments efficacy: mechanistic rationale for drug combination therapy. Mol Cancer Ther 11, 994-1005

48

22. Ahmad, K. A., Wang, G., Slaton, J., Unger, G., and Ahmed, K. (2005) Targeting CK2 for cancer therapy. Anticancer Drugs 16, 1037-1043 23. Unger, G. M., Davis, A. T., Slaton, J. W., and Ahmed, K. (2004) Protein kinase CK2 as regulator of cell survival: implications for cancer therapy. Curr Cancer Drug Targets 4, 77-84 24. Turowec, J. P., Duncan, J. S., French, A. C., Gyenis, L., St Denis, N. A., Vilk, G., and Litchfield, D. W. (2010) Protein kinase CK2 is a constitutively active enzyme that promotes cell survival: strategies to identify CK2 substrates and manipulate its activity in mammalian cells. Methods Enzymol 484, 471-493 25. Georgopoulos, K., Moore, D. D., and Derfler, B. (1992) Ikaros, an early lymphoid- specific transcription factor and a putative mediator for T cell commitment. Science 258, 808-812 26. Molnár, A., and Georgopoulos, K. (1994) The Ikaros gene encodes a family of functionally diverse zinc finger DNA-binding proteins. Mol Cell Biol 14, 8292-8303 27. Popescu, M., Gurel, Z., Ronni, T., Song, C., Hung, K. Y., Payne, K. J., and Dovat, S. (2009) Ikaros stability and pericentromeric localization are regulated by protein phosphatase 1. J Biol Chem 284, 13869-13880 28. Ferreirós-Vidal, I., Carroll, T., Taylor, B., Terry, A., Liang, Z., Bruno, L., Dharmalingam, G., Khadayate, S., Cobb, B. S., Smale, S. T., Spivakov, M., Srivastava, P., Petretto, E., Fisher, A. G., and Merkenschlager, M. (2013) Genome-wide identification of Ikaros targets elucidates its contribution to mouse B-cell lineage specification and pre-B-cell differentiation. Blood 121, 1769-1782 29. Zhang, J., Jackson, A. F., Naito, T., Dose, M., Seavitt, J., Liu, F., Heller, E. J., Kashiwagi, M., Yoshida, T., Gounari, F., Petrie, H. T., and Georgopoulos, K. (2012) Harnessing of the nucleosome-remodeling-deacetylase complex controls lymphocyte development and prevents leukemogenesis. Nat Immunol 13, 86-94 30. Song, C., Gowda, C., Pan, X., Ding, Y., Tong, Y., Tan, B. H., Wang, H., Muthusami, S., Ge, Z., Sachdev, M., Amin, S. G., Desai, D., Gowda, K., Gowda, R., Robertson, G. P., Schjerven, H., Muschen, M., Payne, K. J., and Dovat, S. (2015) Targeting casein kinase II restores Ikaros tumor suppressor activity and demonstrates therapeutic efficacy in high- risk leukemia. Blood

49

31. Wang, J. H., Nichogiannopoulou, A., Wu, L., Sun, L., Sharpe, A. H., Bigby, M., and Georgopoulos, K. (1996) Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity 5, 537-549 32. Dovat, S., Ronni, T., Russell, D., Ferrini, R., Cobb, B. S., and Smale, S. T. (2002) A common mechanism for mitotic inactivation of C2H2 zinc finger DNA-binding domains. Genes Dev 16, 2985-2990 33. Gómez-del Arco, P., Maki, K., and Georgopoulos, K. (2004) Phosphorylation controls Ikaros's ability to negatively regulate the G(1)-S transition. Mol Cell Biol 24, 2797-2807 34. Gurel, Z., Ronni, T., Ho, S., Kuchar, J., Payne, K. J., Turk, C. W., and Dovat, S. (2008) Recruitment of ikaros to pericentromeric heterochromatin is regulated by phosphorylation. J Biol Chem 283, 8291-8300 35. Georgopoulos, K., Bigby, M., Wang, J. H., Molnar, A., Wu, P., Winandy, S., and Sharpe, A. (1994) The Ikaros gene is required for the development of all lymphoid lineages. Cell 79, 143-156 36. Winandy, S., Wu, P., and Georgopoulos, K. (1995) A dominant mutation in the Ikaros gene leads to rapid development of leukemia and lymphoma. Cell 83, 289-299 37. Nakase, K., Ishimaru, F., Avitahl, N., Dansako, H., Matsuo, K., Fujii, K., Sezaki, N., Nakayama, H., Yano, T., Fukuda, S., Imajoh, K., Takeuchi, M., Miyata, A., Hara, M., Yasukawa, M., Takahashi, I., Taguchi, H., Matsue, K., Nakao, S., Niho, Y., Takenaka, K., Shinagawa, K., Ikeda, K., Niiya, K., and Harada, M. (2000) Dominant negative isoform of the Ikaros gene in patients with adult B-cell acute lymphoblastic leukemia. Cancer Res 60, 4062-4065 38. Yagi, T., Hibi, S., Takanashi, M., Kano, G., Tabata, Y., Imamura, T., Inaba, T., Morimoto, A., Todo, S., and Imashuku, S. (2002) High frequency of Ikaros isoform 6 expression in acute myelomonocytic and monocytic leukemias: implications for up- regulation of the antiapoptotic protein Bcl-XL in leukemogenesis. Blood 99, 1350-1355 39. Burnett, A., Wetzler, M., and Löwenberg, B. (2011) Therapeutic advances in acute myeloid leukemia. J Clin Oncol 29, 487-494 40. Bishop, J. F., Matthews, J. P., Young, G. A., Szer, J., Gillett, A., Joshua, D., Bradstock, K., Enno, A., Wolf, M. M., Fox, R., Cobcroft, R., Herrmann, R., Van Der Weyden, M., Lowenthal, R. M., Page, F., Garson, O. M., and Juneja, S. (1996) A randomized study of high-dose cytarabine in induction in acute myeloid leukemia. Blood 87, 1710-1717

50

41. Moghrabi, A., Levy, D. E., Asselin, B., Barr, R., Clavell, L., Hurwitz, C., Samson, Y., Schorin, M., Dalton, V. K., Lipshultz, S. E., Neuberg, D. S., Gelber, R. D., Cohen, H. J., Sallan, S. E., and Silverman, L. B. (2007) Results of the Dana-Farber Cancer Institute ALL Consortium Protocol 95-01 for children with acute lymphoblastic leukemia. Blood 109, 896-904 42. Rots, M. G., Pieters, R., Jansen, G., Kaspers, G. J., Van Zantwijk, C. H., Noordhuis, P., Voorn, D. A., Van Wering, E. R., Creutzig, U., Veerman, A. J., and Peters, G. J. (2001) A possible role for methotrexate in the treatment of childhood acute myeloid leukaemia, in particular for acute monocytic leukaemia. Eur J Cancer 37, 492-498 43. Salvi, M., Sarno, S., Cesaro, L., Nakamura, H., and Pinna, L. A. (2009) Extraordinary pleiotropy of protein kinase CK2 revealed by weblogo phosphoproteome analysis. Biochim Biophys Acta 1793, 847-859 44. Scaglioni, P. P., Yung, T. M., Choi, S., Choi, S. C., Baldini, C., Konstantinidou, G., and Pandolfi, P. P. (2008) CK2 mediates phosphorylation and ubiquitin-mediated degradation of the PML tumor suppressor. Mol Cell Biochem 316, 149-154 45. Li, P. F., Li, J., Müller, E. C., Otto, A., Dietz, R., and von Harsdorf, R. (2002) Phosphorylation by protein kinase CK2: a signaling switch for the caspase-inhibiting protein ARC. Mol Cell 10, 247-258 46. Desagher, S., Osen-Sand, A., Montessuit, S., Magnenat, E., Vilbois, F., Hochmann, A., Journot, L., Antonsson, B., and Martinou, J. C. (2001) Phosphorylation of bid by casein kinases I and II regulates its cleavage by caspase 8. Mol Cell 8, 601-611 47. Toczyski, D. P., Galgoczy, D. J., and Hartwell, L. H. (1997) CDC5 and CKII control adaptation to the yeast DNA damage checkpoint. Cell 90, 1097-1106 48. Quotti Tubi, L., Gurrieri, C., Brancalion, A., Bonaldi, L., Bertorelle, R., Manni, S., Pavan, L., Lessi, F., Zambello, R., Trentin, L., Adami, F., Ruzzene, M., Pinna, L. A., Semenzato, G., and Piazza, F. (2013) Inhibition of protein kinase CK2 with the clinical- grade small ATP-competitive compound CX-4945 or by RNA interference unveils its role in acute myeloid leukemia cell survival, p53-dependent apoptosis and daunorubicin- induced cytotoxicity. J Hematol Oncol 6, 78 49. Pierre, F., Chua, P. C., O'Brien, S. E., Siddiqui-Jain, A., Bourbon, P., Haddach, M., Michaux, J., Nagasawa, J., Schwaebe, M. K., Stefan, E., Vialettes, A., Whitten, J. P., Chen, T. K., Darjania, L., Stansfield, R., Bliesath, J., Drygin, D., Ho, C., Omori, M., Proffitt, C., Streiner, N., Rice, W. G., Ryckman, D. M., and Anderes, K. (2011) Pre- 51

clinical characterization of CX-4945, a potent and selective small molecule inhibitor of CK2 for the treatment of cancer. Mol Cell Biochem 356, 37-43 50. Hahm, K., Cobb, B. S., McCarty, A. S., Brown, K. E., Klug, C. A., Lee, R., Akashi, K., Weissman, I. L., Fisher, A. G., and Smale, S. T. (1998) Helios, a T cell-restricted Ikaros family member that quantitatively associates with Ikaros at centromeric heterochromatin. Genes Dev 12, 782-796 51. Kelley, C. M., Ikeda, T., Koipally, J., Avitahl, N., Wu, L., Georgopoulos, K., and Morgan, B. A. (1998) Helios, a novel dimerization partner of Ikaros expressed in the earliest hematopoietic progenitors. Curr Biol 8, 508-515 52. Evans, R. M., and Hollenberg, S. M. (1988) Zinc fingers: gilt by association. Cell 52, 1-3 53. Gamsjaeger, R., Liew, C. K., Loughlin, F. E., Crossley, M., and Mackay, J. P. (2007) Sticky fingers: zinc-fingers as protein-recognition motifs. Trends Biochem Sci 32, 63-70 54. Leon, O., and Roth, M. (2000) Zinc fingers: DNA binding and protein-protein interactions. Biol Res 33, 21-30 55. Hahm, K., Ernst, P., Lo, K., Kim, G. S., Turck, C., and Smale, S. T. (1994) The lymphoid transcription factor LyF-1 is encoded by specific, alternatively spliced mRNAs derived from the Ikaros gene. Mol Cell Biol 14, 7111-7123 56. Klug, C. A., Morrison, S. J., Masek, M., Hahm, K., Smale, S. T., and Weissman, I. L. (1998) Hematopoietic stem cells and lymphoid progenitors express different Ikaros isoforms, and Ikaros is localized to heterochromatin in immature lymphocytes. Proc Natl Acad Sci U S A 95, 657-662 57. Sun, L., Liu, A., and Georgopoulos, K. (1996) Zinc finger-mediated protein interactions modulate Ikaros activity, a molecular control of lymphocyte development. EMBO J 15, 5358-5369 58. Sun, L., Heerema, N., Crotty, L., Wu, X., Navara, C., Vassilev, A., Sensel, M., Reaman, G. H., and Uckun, F. M. (1999) Expression of dominant-negative and mutant isoforms of the antileukemic transcription factor Ikaros in infant acute lymphoblastic leukemia. Proc Natl Acad Sci U S A 96, 680-685 59. Sun, L., Goodman, P. A., Wood, C. M., Crotty, M. L., Sensel, M., Sather, H., Navara, C., Nachman, J., Steinherz, P. G., Gaynon, P. S., Seibel, N., Vassilev, A., Juran, B. D., Reaman, G. H., and Uckun, F. M. (1999) Expression of aberrantly spliced oncogenic ikaros isoforms in childhood acute lymphoblastic leukemia. J Clin Oncol 17, 3753-3766

52

60. Sun, L., Crotty, M. L., Sensel, M., Sather, H., Navara, C., Nachman, J., Steinherz, P. G., Gaynon, P. S., Seibel, N., Mao, C., Vassilev, A., Reaman, G. H., and Uckun, F. M. (1999) Expression of dominant-negative Ikaros isoforms in T-cell acute lymphoblastic leukemia. Clin Cancer Res 5, 2112-2120 61. Nakayama, H., Ishimaru, F., Avitahl, N., Sezaki, N., Fujii, N., Nakase, K., Ninomiya, Y., Harashima, A., Minowada, J., Tsuchiyama, J., Imajoh, K., Tsubota, T., Fukuda, S., Sezaki, T., Kojima, K., Hara, M., Takimoto, H., Yorimitsu, S., Takahashi, I., Miyata, A., Taniguchi, S., Tokunaga, Y., Gondo, H., Niho, Y., and Harada, M. (1999) Decreases in Ikaros activity correlate with blast crisis in patients with chronic myelogenous leukemia. Cancer Res 59, 3931-3934 62. Olivero, S., Maroc, C., Beillard, E., Gabert, J., Nietfeld, W., Chabannon, C., and Tonnelle, C. (2000) Detection of different Ikaros isoforms in human leukaemias using real-time quantitative polymerase chain reaction. Br J Haematol 110, 826-830 63. Francis, O. L., Payne, J. L., Su, R. J., and Payne, K. J. (2011) Regulator of myeloid differentiation and function: The secret life of Ikaros. World J Biol Chem 2, 119-125 64. Payne, K. J., Nicolas, J. H., Zhu, J. Y., Barsky, L. W., and Crooks, G. M. (2001) Cutting edge: predominant expression of a novel Ikaros isoform in normal human hemopoiesis. J Immunol 167, 1867-1870 65. Morgan, B., Sun, L., Avitahl, N., Andrikopoulos, K., Ikeda, T., Gonzales, E., Wu, P., Neben, S., and Georgopoulos, K. (1997) Aiolos, a lymphoid restricted transcription factor that interacts with Ikaros to regulate lymphocyte differentiation. EMBO J 16, 2004-2013 66. Li, Z., Perez-Casellas, L. A., Savic, A., Song, C., and Dovat, S. (2011) Ikaros isoforms: The saga continues. World J Biol Chem 2, 140-145 67. Wolfe, S. A., Nekludova, L., and Pabo, C. O. (2000) DNA recognition by Cys2His2 zinc finger proteins. Annu Rev Biophys Biomol Struct 29, 183-212 68. Cobb, B. S., Morales-Alcelay, S., Kleiger, G., Brown, K. E., Fisher, A. G., and Smale, S. T. (2000) Targeting of Ikaros to pericentromeric heterochromatin by direct DNA binding. Genes Dev 14, 2146-2160 69. Koipally, J., Heller, E. J., Seavitt, J. R., and Georgopoulos, K. (2002) Unconventional potentiation of gene expression by Ikaros. J Biol Chem 277, 13007-13015 70. Schjerven, H., McLaughlin, J., Arenzana, T. L., Frietze, S., Cheng, D., Wadsworth, S. E., Lawson, G. W., Bensinger, S. J., Farnham, P. J., Witte, O. N., and Smale, S. T. (2013)

53

Selective regulation of lymphopoiesis and leukemogenesis by individual zinc fingers of Ikaros. Nat Immunol 14, 1073-1083 71. Georgopoulos, K. (2002) Haematopoietic cell-fate decisions, chromatin regulation and ikaros. Nat Rev Immunol 2, 162-174 72. Kastner, P., Dupuis, A., Gaub, M. P., Herbrecht, R., Lutz, P., and Chan, S. (2013) Function of Ikaros as a tumor suppressor in B cell acute lymphoblastic leukemia. Am J Blood Res 3, 1-13 73. Milosevic, J. D., Puda, A., Malcovati, L., Berg, T., Hofbauer, M., Stukalov, A., Klampfl, T., Harutyunyan, A. S., Gisslinger, H., Gisslinger, B., Burjanivova, T., Rumi, E., Pietra, D., Elena, C., Vannucchi, A. M., Doubek, M., Dvorakova, D., Robesova, B., Wieser, R., Koller, E., Suvajdzic, N., Tomin, D., Tosic, N., Colinge, J., Racil, Z., Steurer, M., Pavlovic, S., Cazzola, M., and Kralovics, R. (2012) Clinical significance of genetic aberrations in secondary acute myeloid leukemia. Am J Hematol 87, 1010-1016 74. Kirstetter, P., Thomas, M., Dierich, A., Kastner, P., and Chan, S. (2002) Ikaros is critical for B cell differentiation and function. Eur J Immunol 32, 720-730 75. Papathanasiou, P., Perkins, A. C., Cobb, B. S., Ferrini, R., Sridharan, R., Hoyne, G. F., Nelms, K. A., Smale, S. T., and Goodnow, C. C. (2003) Widespread failure of hematolymphoid differentiation caused by a recessive niche-filling allele of the Ikaros transcription factor. Immunity 19, 131-144 76. Nichogiannopoulou, A., Trevisan, M., Neben, S., Friedrich, C., and Georgopoulos, K. (1999) Defects in hemopoietic stem cell activity in Ikaros mutant mice. J Exp Med 190, 1201-1214 77. Trinh, L. A., Ferrini, R., Cobb, B. S., Weinmann, A. S., Hahm, K., Ernst, P., Garraway, I. P., Merkenschlager, M., and Smale, S. T. (2001) Down-regulation of TDT transcription in CD4(+)CD8(+) thymocytes by Ikaros proteins in direct competition with an Ets activator. Genes Dev 15, 1817-1832 78. Arney, K. L., and Fisher, A. G. (2004) Epigenetic aspects of differentiation. J Cell Sci 117, 4355-4363 79. Rice, J. C., and Allis, C. D. (2001) Histone methylation versus histone acetylation: new insights into epigenetic regulation. Curr Opin Cell Biol 13, 263-273 80. Wilson, B. G., Helming, K. C., Wang, X., Kim, Y., Vazquez, F., Jagani, Z., Hahn, W. C., and Roberts, C. W. (2014) Residual complexes containing SMARCA2 (BRM) underlie the oncogenic drive of SMARCA4 (BRG1) mutation. Mol Cell Biol 34, 1136-1144 54

81. Kim, J., Sif, S., Jones, B., Jackson, A., Koipally, J., Heller, E., Winandy, S., Viel, A., Sawyer, A., Ikeda, T., Kingston, R., and Georgopoulos, K. (1999) Ikaros DNA-binding proteins direct formation of chromatin remodeling complexes in lymphocytes. Immunity 10, 345-355 82. Koipally, J., and Georgopoulos, K. (2000) Ikaros interactions with CtBP reveal a repression mechanism that is independent of histone deacetylase activity. J Biol Chem 275, 19594-19602 83. Koipally, J., and Georgopoulos, K. (2002) Ikaros-CtIP interactions do not require C- terminal binding protein and participate in a deacetylase-independent mode of repression. J Biol Chem 277, 23143-23149 84. Marçais, A., Jeannet, R., Hernandez, L., Soulier, J., Sigaux, F., Chan, S., and Kastner, P. (2010) Genetic inactivation of Ikaros is a rare event in human T-ALL. Leuk Res 34, 426- 429 85. Dovat, S. (2011) Ikaros in hematopoiesis and leukemia. World J Biol Chem 2, 105-107 86. Matulić, M., Paradzik, M., Puskarić, B. J., Stipić, J., and Antica, M. (2010) Analysis of Ikaros family splicing variants in human hematopoietic lineages. Coll Antropol 34, 59-62 87. Tallman, M. S., and Altman, J. K. (2008) Curative strategies in acute promyelocytic leukemia. Hematology Am Soc Hematol Educ Program, 391-399 88. Adès, L., Guerci, A., Raffoux, E., Sanz, M., Chevallier, P., Lapusan, S., Recher, C., Thomas, X., Rayon, C., Castaigne, S., Tournilhac, O., de Botton, S., Ifrah, N., Cahn, J. Y., Solary, E., Gardin, C., Fegeux, N., Bordessoule, D., Ferrant, A., Meyer-Monard, S., Vey, N., Dombret, H., Degos, L., Chevret, S., Fenaux, P., and Group, E. A. (2010) Very long-term outcome of acute promyelocytic leukemia after treatment with all-trans retinoic acid and chemotherapy: the European APL Group experience. Blood 115, 1690-1696 89. Bhatla, D., Gerbing, R. B., Alonzo, T. A., Conner, H., Ross, J. A., Meshinchi, S., Zhai, X., Zamzow, T., Mehta, P. A., Geiger, H., Perentesis, J., and Davies, S. M. (2009) Cytidine deaminase genotype and toxicity of cytosine arabinoside therapy in children with acute myeloid leukemia. Br J Haematol 144, 388-394 90. Schwahn, B., and Rozen, R. (2001) Polymorphisms in the methylenetetrahydrofolate reductase gene: clinical consequences. Am J Pharmacogenomics 1, 189-201 91. Métais, J. Y., Winkler, T., Geyer, J. T., Calado, R. T., Aplan, P. D., Eckhaus, M. A., and Dunbar, C. E. (2012) BCL2A1a over-expression in murine hematopoietic stem and

55

progenitor cells decreases apoptosis and results in hematopoietic transformation. PLoS One 7, e48267 92. Starkova, J., Gadgil, S., Qiu, Y. H., Zhang, N., Hermanova, I., Kornblau, S. M., and Drabkin, H. A. (2011) Up-regulation of homeodomain genes, DLX1 and DLX2, by FLT3 signaling. Haematologica 96, 820-828 93. Seiter, K. (2013) Acute Myeloid Leukemia Staging. Medscape 94. APExBIO. CK2 Inhibitor CX-4945 (Silmitasertib). 95. Song, C., Li, Z., Erbe, A. K., Savic, A., and Dovat, S. (2011) Regulation of Ikaros function by casein kinase 2 and protein phosphatase 1. World J Biol Chem 2, 126-131

56

Chapter 2

Therapeutic Efficacy of the Specific CK2 Inhibitor CX-4945 in AML Involves Transcriptional Repression of BCL2A1 by Ikaros

57

Abstract

Casein Kinase II (CK2) is overexpressed in many cancers, including acute myeloid leukemia (AML). CK2 has been shown to regulate the DNA-binding and subcellular localization of the tumor suppressor Ikaros (IKZF1). Mutation or loss of functional Ikaros has been associated with a poor prognosis in AML. Here, we report that

Ikaros, in complex with the histone deacetylase HDAC1, represses transcription of the anti-apoptotic gene BCL2A1. In AML, overexpression of BCL2A1 is a negative prognostic marker, and anti-apoptotic mechanisms contribute to resistance to chemotherapy. Repression of BCL2A1 results in increased apoptosis. The ability of

Ikaros to repress BCL2A1 transcription is impaired by CK2. Inhibition of CK2 increases

Ikaros binding at the BCL2A1 promoter, corresponding with loss of the positive marker

H3K4me2 and increase in the negative marks H3K9me3 and H3K27me3, signatures of a repressive chromatin environment. Our data suggest that direct repression of BCL2A1 by

Ikaros corresponds with epigenetic changes and is one mechanism by which Ikaros acts as a tumor suppressor in AML. Overexpression of CK2 in leukemia prevents this regulation, and CK2 inhibitors are able to restore Ikaros-mediated repression of BCL2A1 and promote apoptosis. Furthermore, the specific CK2 inhibitor CX-4945 synergizes with the chemotherapy drug doxorubicin. These data identify BCL2A1 as a therapeutic target in AML and provide mechanistic rationale for the use of CK2 inhibitors in combination with current drugs for the treatment of AML.

58

Introduction

Acute myeloid leukemia (AML) is a malignant disease of the myeloid blood lineage and is characterized by the rapid growth of abnormal cells that accumulate in the bone marrow and interfere with the production of normal hematopoietic cells (1,2). Alterations at the genetic and epigenetic levels result in perturbations of normal processes, especially self-renewal, proliferation, and differentiation (3-6). AML is the most common acute leukemia affecting adults, and its incidence increases with age. AML is treated initially with chemotherapy aimed at inducing a remission, and patients may go on to receive additional chemotherapy or a hematopoietic stem cell transplant. However, currently available treatments for acute myeloid leukemia are insufficient, as more than half of all

AML patients develop disease that is resistant to chemotherapy (7). Thus there is a need for novel, targeted therapy for AML.

Casein Kinase II (CK2) is a pro-oncogenic serine/threonine kinase shown to have enhanced activity in various types of human malignancies (8-10). In hematopoietic malignancies, CK2 overexpression has been observed in both B-ALL and T-ALL, suggesting its oncogenic role in these diseases (11,12). In addition, overexpression of the

CK2 catalytic subunit in the T-cell lineage in transgenic mice resulted in the development of T-ALL (13). Several studies have demonstrated a role for CK2 in AML. Overexpression of the CK2 catalytic subunit, CK2α, has been reported in AML cell lines and in a considerable proportion of primary leukemic blasts from AML patients (14). Rates of disease-free and overall survival were significantly decreased in patients with normal karyotype AML that showed high levels of CK2α expression (14). Overexpression of

59

CK2α resulted in increased levels of pAkt/PKB in primary AML cells, and treatment with a selective CK2 inhibitor inhibited this pathway in a dose-dependent manner (14). The

CK2 inhibitor, apigenin, was able to increase apoptosis preferentially in AML cell lines and primary blasts expressing high levels of CK2α (14). Inhibition of CK2 by the inhibitor

CX-4945 or siRNAs has been shown to produce strong cytotoxic effects in AML cells (15).

These data demonstrate that CK2 acts as an important regulator of AML cell survival and response to chemotherapy. Together, these results suggest that CK2 is an important therapeutic target in AML, and suggest that the use of CK2-specific inhibitors can be an effective treatment for this disease. However, the mechanisms through which CK2 promotes cell proliferation and survival in AML are still unknown, precluding the design of targeted combination treatment. Recently, CK2 has been shown to exert its oncogenic activity by inhibiting the tumor suppressor function of Ikaros (IKZF1) in B-cell acute lymphoblastic leukemia (B-ALL). Treatment with CK2 inhibitors restored the tumor suppressor activity of Ikaros resulting in a cytotoxic effect (16). Since Ikaros has been shown to act as a tumor suppressor in AML, we hypothesized that CK2 inhibitors can exert therapeutic effects in AML by enhancing Ikaros tumor suppressor function. In this report, we analyze the effect of CK2 inhibition on the transcriptional regulation of Ikaros target genes in AML. Our results demonstrate that the therapeutic efficacy of targeted CK2 inhibition involves enhanced Ikaros-mediated transcriptional repression of the BCL2A1 gene, resulting in increased apoptosis in AML cells. Our data also show that CK2 inhibition shows synergistic effects in combination with the cytotoxic drug, Doxorubicin, against

AML cells.

60

Experimental Procedures

Cells, Cell Culture, and Reagents

U937 cells were obtained from the American Type Culture Collection (ATCC),

Manassas, VA. Cells were cultured in RPMI 1640 growth medium (CellGro) supplemented with 10% fetal bovine serum (FBS) (HyClone). HEK-293T cells were cultured in DMEM (CellGro) supplemented with 10% FBS. Pheresis #4 primary cells were obtained from a patient with AML from Loma Linda University (Loma Linda, CA).

Cells were incubated at 37°C in a humidified atmosphere of 5% CO2. DMSO and

Doxorubicin (DOXO) were purchased from Sigma-Aldrich (St Louis, MO). CX-4945 was provided by Cylene Pharmaceuticals (San Diego, CA).

ChIP-Seq experiments

ChIP-Seq assays were performed as previously described and as described below (17,18).

For ChIP-Seq libraries, 2x106 U937 cells were treated with 10 µM CX-4945 for 72 hours, cross-linked for 20 min in PBS containing 1% formaldehyde, and the reaction was stopped by adding 11/20 volume of 2.5M glycine. Cell pellets were flash frozen and stored at −80°C. Chromatin was fragmented using a Bioruptor (Diagenode) for 30 min on ice (30s pulses, 30s pauses) to produce fragments ~200-400bp in size. Goat anti-rabbit

IgG Dynabeads (Invitrogen) were coated with affinity purified anti-Ikaros antibody and incubated with chromatin overnight on a rotator at 4°C. Protein/DNA complexes were captured with a Magnetic Particle Concentrator (Invitrogen). Crosslinks were reversed 61

and samples were treated with proteinase K and RNaseA. DNA was recovered using the

QIAquick PCR Purification kit (QIAGEN). ChIP-Seq libraries were created using 18 cycles of amplification with the ChIP-Seq DNA sample prep kit (Illumina). Libraries were validated using the Agilent Technologies 2100 Bioanalyzer. ChIP-Seq libraries were sequenced at the High Throughput Genomics Center at University of Washington,

Seattle and at the Core Facility at Penn State College of Medicine, Hershey, PA.

Sequence fastq files were aligned to the UCSC human genome assembly HG19 using the

Eland application (Illumina), allowing no more than two mismatches per sequence. Only sequences aligning uniquely to the human genome were used to identify peaks. Peak calling was achieved using Cisgenome2.0 and SISSRS. Sequence reads that have been aligned to a reference genome were converted into BAR files. Parameters for the detection of peaks were set, and all peaks called were required to have a number of reads at least two-fold higher than the background and a False Discovery Rate (FDR) less than

0.05. The criteria for identifying Ikaros peaks was extremely stringent to prevent the identification of false positives.

Antibodies

The antibody used for chromatin immunoprecipitation of Ikaros has been described previously (19). Other antibodies used were as follows: HDAC1 (Abcam ab7028);

H3K4me2 (Millipore 07-030); H3K9me3 (Abcam ab8898); H3K27me3 (Millipore 07-

449); anti-Rabbit IgG (ab46540, Abcam). All antibodies used for ChIP-Seq were Encode validated antibodies.

62

Cell Proliferation Assay

Proliferation assays with CX-4945 and Doxorubicin were performed as described previously (20). Briefly, cells in logarithmic growth phase were harvested and resuspended to a final concentration of 1 x 105 cells/mL of fresh medium and treated with

CX-4945, Doxorubicin, or a combination thereof. 104 cells were plated per well in triplicate in a 96-well clear bottom plate (Costar 3603) and incubated for two days at

37°C in a humidified incubator with 5% CO2. On day 2, WST-1 reagent (Roche,

11644807001) was added at 10μl/well and cells were allowed to incubate for an additional 4h at 37°C. Absorbance was measured at 440/690nm using a BioTek Synergy

Mx plate reader.

DNA Constructs and shRNA

Luciferase reporter constructs for the pLightSwitch-Rom empty vector and for the promoter region of BCL2A1 which contained the Ikaros binding peak identified by ChIP-

Seq were purchased from SwitchGear Genomics. The pMSCV bicistronic retroviral vector (MIG vector) and the pMSCV bicistronic retroviral vector encoding wild-type human HA-tagged Ikaros (IKZF1) which contains a 5’ long-terminal-repeat-driven

Ikaros, internal ribosome entry side (IRES), and enhanced green fluorescent protein

(EGFP) were described previously (21).

63

Retroviral Gene Transfer and Cell Sorting

Retroviruses were produced by transient transfection in amphotropic packaging HEK-

293T cell lines as described previously (22). U937 cells were plated in 24-well plates at

4x105 cells/well and suspended in retroviral supernatants with 12 μg/mL polybrene and centrifuged at 1,400 g, at 32 °C, for 2 hrs. The cells were then suspended in fresh 10%

FBS RPMI 1640 and cultured at 37°C, 5% CO2 for 3 days. The cells were isolated by

Ficoll separation and the GFP (+) cells were sorted using a FACSAria High speed sorter

(Becton Dickinson). Sorted cells were further cultured followed by RNA isolation.

Luciferase Assay

Luciferase assays were performed using the LightSwitch Luciferase Assay System

(SwitchGear Genomics). HEK293T cells were seeded into 24-well plates. After 24 hours, cells were transiently transfected with 0.15 μg of indicated promoter reporter constructs or pROM vector and 0.15 μg of pcDNA3.1-Ikaros or pcDNA3.1 vector in triplicate for each group using lipofectamine 2000 (Invitrogen) according to manufacturer’s instructions. Twenty four hours after transfection, cells were lysed in 100 μl of

LightSwitch Assay Solution (SwitchGear Genomics) and rocked at RT for 30min according to the manufacturer’s instructions. Lysates were measured by luminometer

(Promega GloMax 20/20 Luminometer). Luciferase activities were calculated as fold change relative to vector only cells and normalized to pcDNA3.1 vector readings. All transfection and reporter assays were performed in triplicate.

64

Quantitative Chromatin Immunoprecipitation ChIP (qChIP) qChIP assays were performed as reported previously (17). Experiments were performed with a minimum of two biological replicates. Real-time PCR was performed using

PerfeCtaTM Sybr Green FastMix (Quanta Biosciences) and a StepOne Plus real-time

PCR system (Applied Biosystems) according to manufacturer’s instructions. Enrichment of the ChIP’d DNA was normalized to a non-specific IgG control. The comparative Ct method was used for quantification and fold enrichment was calculated using the formula: 2-CTsample/2-CTinput, where CT equals the threshold cycle number. Primers for BCL2A1: 5’ –TCTCAGCACATTGCCTCAAC– 3’; 5’ –

AGTCCTGAGCCAGCCTGTAA – 3’.

Quantitative Reverse-Transcriptase PCR (qRT-PCR)

Total RNA was isolated from cells using the QIAshredder and RNeasy Mini Kit

(QIAGEN). Complementary DNA (cDNA) was generated from 1 μg total RNA using

Superscript First-Strand Synthesis System (Invitrogen). qPCR was performed using a

StepOne Plus real-time PCR machine (Applied Biosystems) with PerfeCta SYBR Green

FastMix (Quanta Biosciences) and primers for BCL2A1 (Forward: 5’ –

TCTCAGCACATTGCCTCAAC– 3’; Reverse: 5’ – AGTCCTGAGCCAGCCTGTAA –

3). Values were normalized to 18s RNA and the relative expression values were determined by the 2-ΔΔCt method.

65

Biochemical Experiments

Nuclear extraction and Western blot were performed as described previously (22,23).

Band intensities were quantified by ImageJ software. The protein expression level as percentages of control signals (% control) in each blot was used to correct for variations between blots.

CK2 Kinase Activity

U937 cells were treated with 20 µM CX-4945 for 72 hours. The last four hours of treatment, untreated and CX-4945-treated U937 cells were labeled with radioactive orthophosphate (19,24). Cells were washed twice with phosphate-free RPMI 1640 medium and incubated for 4 hours with 0.5 mCi/ml [32P]-orthophosphate (NEN) in phosphate-free medium. Cells were collected by centrifugation, and nuclear extract was obtained. The equal aliquote of nuclear extracts were incubated with anti-IK-CTS antibodies for 1 hr at 4oC and the resulting immunocomplexes were absorbed to protein

G-Sepharose (Pharmacia), washed four times with solubilizing buffer, separated by SDS-

PAGE, transferred to a nylon membrane and subjected to autoradiography. Radioactivity of excised bands corresponding to 3 largest Ikaros isoforms – IK-H, IK-1 and IK-2 was counted on scintillation counter. CK2 kinase assay was performed using -GTP on whole cell lysate of untreated and CX-treated U937 cells. An equal, small aliquote of nuclear extract of treated and untreated U937 cells were subjected to Western blot – to document

66

that the equal amount of protein in both treated and untreated cells were used for immunoprecipitation.

Ikaros and CK2 shRNA Knockdown shRNA constructs for human Ikaros (IKZF1) and human CK2 (CSNK2A1) in a GFP vector (pGFP-V-RS) were purchased from Origene. U937 cells were transiently transfected with 3.0 μg of plasmids per well in 24-well plates using the Neon

Transfection System (Invitrogen). After transfection for 1 day, U937 cells with transfection efficiency ranges from ~80% (green cells) and more than 95% cell viability were further treated with 10μM CX-4945 or non-treatment control (0.01% DMSO) for 2 days. The cells were then harvested for total RNA isolation and total lysate extraction.

The 29-mer scrambled shRNA pGFPV-RS vector was used as a control. The knockdown of Ikaros and CK2 was confirmed by qPCR (19).

Drug Synergy Analysis with Calcusyn Software

Drug synergy for combinations of CX-4945 with doxorubicin were demonstrated by using the Chou-Talalay method to determine the combination index (CI) using Calcusyn software (25,26). CI values are illustrated graphically for each combination with an isobologram. Values <0.85 are synergistic, 0.9-1.1 are additive, and >1.1 are antagonistic

(27).

67

Results

Treatment with CX-4945 enhances Ikaros binding to the promoter of BCL2A1 in AML cells.

The effect of CK2 inhibition on Ikaros function in AML was studied in U937 cells – a human myelomonocytic leukemia cell line. U937 cells were treated with CX-

4945, a highly specific CK2 inhibitor that has shown therapeutic efficacy against a variety of malignancies in preclinical models (28). Treatment with 20 µM CX-4945 for

72 hrs resulted in strong inhibition of CK2 activity (4-fold) as evidenced by a kinase assay performed using nuclear extracts of CX-4945-treated and control cells (Figure

2.1A). The phosphorylation of Ikaros following 72 hours of treatment with CK2 inhibitor was determined by in vivo labeling with orthophosphate 32P. Following in vivo labeling,

Ikaros proteins were immunoprecipitated with anti-Ikaros antibodies and autoradiographed. The radioactive signal directly reflects the amount of phosphorus incorporated into the Ikaros protein, and thus the phosphorylation status of Ikaros. Ikaros content was normalized by Western blot (Figure 2.1B left panel). Results showed that

CK2 inhibition is associated with a severe reduction of Ikaros phosphorylation (5-fold), as evidenced by the difference in signal in autoradiography of Ikaros in CX-4945-treated cells vs. control (Figure 2.1B right panels).

The effect of CK2 inhibition on global Ikaros DNA binding in U937 cells was analyzed using chromatin immunoprecipitation followed by next-generation sequencing

(ChIP-Seq) with anti-Ikaros antibodies in CX-4945-treated U937 and control cells. Ikaros

68

binding site detection (ChIP-Seq peaks) and annotation was accomplished using

CisGenome software. The distribution of Ikaros binding related to transcriptional start sites (TSS) of the genes demonstrated a striking difference between CX-4945-treated and control U937 cells (Figure 2.1C). In control U937 cells, Ikaros binding shows no preference toward particular regions in relation to the TSS. However, following CX-4945 treatment, Ikaros predominantly binds to the regions within +/- 1kb of the TSS of its target genes (Figure 2.1C). These data suggest that dephosphorylation of Ikaros following

CK2 inhibition by CX-4945 treatment enhances the DNA-binding affinity of Ikaros toward promoters of its target genes in U937 cells.

The analysis of Ikaros binding, genome-wide, identified a large number of Ikaros target genes (data not shown). Since our goal is to determine the mechanisms by which

CK2 inhibition regulates Ikaros tumor suppressor function in AML, we have focused our studies on Ikaros binding at promoters of target genes that are known to be involved in malignant transformation. Results identified BCL2-Related Protein A1 (BCL2A1) as one of the Ikaros target genes. Analysis of ChIP-Seq data showed that Ikaros binds to the promoter of the BCL2A1 gene with high affinity in CX-4945-treated cells, but with poor affinity to the same region in control cells (Figure 2.1D). These data suggest that CK2 inhibition regulates Ikaros binding to the BCL2A1 promoter in U937 cells. Since BCL2A1 expression has been associated with malignant transformation and regulation of apoptosis, we performed a functional analysis of the role of Ikaros and CK2 in regulating

BCL2A1 expression in AML (29).

69

Ikaros represses transcription of BCL2A1

To determine the functional significance of Ikaros binding to the promoter region of BCL2A1, we performed both gain-of-function and loss-of-function experiments. First, we used the luciferase reporter assay to analyze the direct effect of Ikaros binding to the

BCL2A1 promoter region on BCL2A1 transcription. We used a plasmid containing the luciferase gene downstream of the promoter region of BCL2A1 that was identified by

ChIP-Seq as an Ikaros binding site. The luciferase reporter assay was performed in HEK-

293T cells because they do not express endogenous Ikaros and the phosphorylation status of Ikaros in these cells is similar to that in hematopoietic cells. HEK-293T cells co- transfected with a promoter construct and an empty vector control showed strong luciferase activity, indicating that the promoter constructs alone could stimulate increased expression of the luciferase gene. Cells co-transfected with a plasmid containing full- length human Ikaros in addition to the promoter construct showed a 3-fold reduction in luciferase activity when compared to cells with the promoter construct and an empty vector (Figure 2.2A). These data demonstrate that Ikaros binding at the promoter of the

BCL2A1 gene can directly repress its transcription.

Next, we tested the effect of Ikaros on BCL2A1 transcription in U937 AML cells.

U937 cells were transduced with retrovirus containing wild type Ikaros (Figure 2.2B, left panel) or with empty retroviral vector (negative control). The effect of Ikaros overexpression on BCL2A1 transcription was studied using qRT-PCR. Results showed that Ikaros overexpression results in significantly decreased transcription of BCL2A1 as

70

compared to control cells transduced with retroviral vector (Figure 2.2B, right panel).

These data suggest that Ikaros represses transcription of BCL2A1 in AML cells.

To study how the loss of Ikaros function affects transcription of BCL2A1 in leukemia, we targeted Ikaros with shRNA (Figure 2.2C, left panel). U937 cells were transfected with Ikaros shRNA or with the empty vector (negative control) and changes in BCL2A1 transcription were assessed by qRT-PCR. The results showed that U937 cells transfected with Ikaros shRNA showed a 2-fold increase in transcription of BCL2A1, when compared to controls (Figure 2.2C, right panel). These results provide further evidence that Ikaros represses transcription of BCL2A1 in leukemia.

Taken together, gain-of-function and loss-of-function experiments provide evidence that Ikaros acts as a transcriptional repressor of BCL2A1 in U937 cells.

Casein Kinase II (CK2) positively regulates transcription of BCL2A1

We tested the effect of CK2 on BCL2A1 transcription in AML. First, we determined the effect of CK2 inhibition on BCL2A1 transcription in U937 cells.

Molecular inhibition of CK2 function using shRNA against the CK2 catalytic subunit,

CK2 alpha, resulted in reduced transcription of BCL2A1 (as measured by qRT-PCR)

(Figure 2.3A). The effect of pharmacological inhibition of CK2 on BCL2A1 transcription in U937 cells was tested using CX-4945. The results showed that treatment with CX-

4945 decreases transcription of BCL2A1 in a dose-dependent manner in U937 cells

(Figure 2.3B). The effect of CX-4945 treatment on BCL2A1 transcription was also tested in primary AML cells. Results demonstrate that CK2 inhibition following CX-4945

71

treatment results in reduced transcription of BCL2A1 as measured by qRT-PCR (Figure

2.3C). Thus, both molecular and pharmacological inhibition of CK2 in U937 cells and primary AML cells showed a similar effect – decreased transcription of BCL2A1– suggesting that CK2 promotes BCL2A1 transcription.

These experiments suggest that CK2 activity in AML cells positively regulates transcription of BCL2A1 and that treatment of AML cells with CK2 inhibitors results in reduced expression of BCL2A1.

CK2 inhibits Ikaros-mediated transcriptional repression of BCL2A1

The experiments described above show that Ikaros functions as a transcriptional repressor of BCL2A1, and that CK2 activity positively regulates transcription of BCL2A1 in AML cells. Our main hypothesis is that increased CK2 activity in AML cells interferes with Ikaros function as a transcriptional regulator and that the therapeutic effect of CK2 inhibitors involves restoration (or enhancement) of Ikaros-mediated transcriptional regulation of its target genes. We tested whether inhibition of CK2 function affects the ability of Ikaros to bind the BCL2A1 promoter. U937 cells were treated with CX-4945, and Ikaros DNA-binding affinity toward the promoter of BCL2A1 was compared to untreated cells by quantitative chromatin immunoprecipitation (qChIP). Results show that in untreated U937 cells, Ikaros binds to the BCL2A1 promoter with a low affinity as evidenced by Ikaros enrichment of slightly less than 2-fold over the background as measured byqChIP (Figure 2.4A, lanes 1-2). The inhibition of CK2 following CX-4945 treatment results in a strong increase in DNA-binding affinity of Ikaros to the BCL2A1

72

promoter (Figure 2.4A, lanes 3-4). Similar results were observed in primary AML cells:

Ikaros bound the BCL2A1 promoter poorly, and inhibition of CK2 restored high affinity

Ikaros binding to the BCL2A1 promoter in primary AML cells (Figure 2.4B).

The above data show that CK2 reduces Ikaros binding to the BCL2A1 promoter and that treatment with CK2 inhibitor enhances Ikaros DNA-binding affinity to this promoter. Next, we tested whether Ikaros expression is essential for the transcriptional repression of the BCL2A1 gene following CK2 inhibition. U937 cells were transfected with scrambled or Ikaros shRNA. The effect of CK2 inhibition on the transcription of the

BCL2A1 gene was compared by qRT-PCR in cells with reduced Ikaros expression and in control cells. Results show that the CK2 inhibitor CX-4945 represses transcription of

BCL2A1 in cells treated with scrambled shRNA (Figure 2.4C bar 2 vs. bar 1). However, in cells with shRNA knock-down of Ikaros, the ability of CX-4945 to repress BCL2A1 is abolished, as compared to control cells with scrambled shRNA (Figure 2.4C, bar 3 as compared to bar 1). These results suggest that the transcriptional repression of the

BCL2A1 gene following CK2 inhibition requires Ikaros function as a transcriptional repressor.

Overall, the results presented in Figure 2.4 suggest that CK2 inhibition in U937 cells enhances Ikaros function as a transcriptional regulator of BCL2A1 and that CK2 inhibition represses BCL2A1 transcription via Ikaros.

73

Inhibition of CK2 alters the epigenetic signature at the BCL2A1 promoter

Ikaros has been shown to regulate transcription of its target genes via chromatin remodeling by inducing epigenetic changes around the transcriptional start site (TSS)

(30-34). Since inhibition of CK2 activity results in repression of BCL2A1 via Ikaros, we tested whether repression of BCL2A1 is achieved through epigenetic mechanisms. First, since it has been shown that Ikaros can bind and form a complex with histone deacetylase

1 (HDAC1), we performed ChIP-Seq with HDAC1 antibodies in CX-4945-treated and control U937 cells. In addition, we performed ChIP-Seq of CX-4945-treated U937 cells and control U937 cells using antibodies for two types of histone modification markers: a) a marker that is associated with open chromatin and positive regulation of transcription – dimethylation of lysine 4 of Histone H3 (H3K4me2); and b) markers that are associated with the formation of repressive chromatin – trimethylation of lysine 9 of histone H3

(H3K9me3) and trimethylation of Lysine 27 of histone H3 (H3K27me3) (35,36). We analyzed the epigenetic landscape around the BCL2A1 promoter in control and CX-4945- treated U937 cells.

Our data show that the epigenetic signature around the BCL2A1 promoter undergoes substantial epigenetic changes following CX-4945 treatment. First we found that HDAC1 exhibits strong occupancy of the BCL2A1 promoter in U937 cells following

CX-4945 treatment (two strong peaks), while control cells do not show any HDAC1 occupancy (Figure 2.5). Interestingly one of the HDAC1 peaks that is observed in CX-

4945-treated cells is located in close proximity to the Ikaros peak that is also characteristic of CX-4945-treated U937 cells, suggesting that Ikaros might recruit

74

HDAC1 to the BCL2A1 promoter in U937 cells following CX-4945 treatment. Second, while untreated cells do not show the presence of markers of repressive chromatin,

H3K9me3 and H3K27me3, both of these histone modifications exhibit strong occupancy at the BCL2A1 promoter in CX-4945-treated U937 cells. Third, the epigenetic marker of open chromatin also associated with positive regulation of gene transcription, H3K4me2, showed a strong presence at the promoter of the BCL2A1 gene in untreated U937 cells.

Following treatment of U937 cells with CX-4945, the presence of the H3K4me2 marker was not detected at the BCL2A1 promoter.

Overall, these data show that the treatment of U937 cells with CX-4945 is associated with recruitment of HDAC1 to the promoter of the BCL2A1 gene, formation of repressive chromatin with the addition of H3K9me3 and H3K27me3 marks, and the loss of the positive chromatin marker H3K4me2. Both recruitment of HDAC1 and changes in the epigenetic signature occur in close proximity to the Ikaros binding site, suggesting that Ikaros binding at the promoter of BCL2A1 after CX-4945 treatment results in epigenetic changes associated with the formation of repressive chromatin and transcriptional repression of BCL2A1.

Inhibition of CK2 induces apoptosis and shows synergistic effects with doxorubicin in U937 cells

The BCL2A1 gene is a member of the BCL2 gene family and functions to suppress apoptosis (29). The regulation of apoptosis is achieved by a balance in expression and function of pro-apoptotic genes and anti-apoptotic genes. The increased

75

expression of anti-apoptotic genes is associated with malignant transformation and increased resistance to chemotherapy (37). Since CK2 inhibition is associated with transcriptional repression of BCL2A1 in U937 cells, we tested whether CK2 inhibition induces apoptosis in U937 cells and/or whether it increases sensitivity of these cells to chemotherapy. To see if apoptosis was increased in U937 cells after CX-4945 treatment,

U937 cells were treated with 10 µM CX-4945 for 48 hours and then stained with

Annexin V PE and 7-AAD to detect apoptotic cells. Flow cytometry analysis showed that apoptosis was increased after CX-4945 treatment when compared to the untreated control (Figure 2.6A). Similar results were obtained in primary AML cells following CX-

4945 treatment (Figure 2.6B).

Next, we determined whether treatment with the specific CK2 inhibitor, CX-4945, could enhance the sensitivity of AML cells to doxorubicin when used in combination.

Cell proliferation assays were carried out in U937 cells to examine the toxicity of combination treatment with CX-4945 and doxorubicin. Cells were treated with multiple concentrations of CX-4945 or doxorubicin alone or with combinations of CX-4945 with doxorubicin. Cellular proliferation was measured by spectrophotometry using WST1 reagent.

Synergistic concentrations were observed in U937 cells treated with a combination of

CX-4945 and doxorubicin. Two doses of CX-4945 in combination with 50 nM doxorubicin met the CalcuSyn criteria for drug synergy (Figure 2.6). A combination index (CI) value, as calculated from the dose response curves, less than 0.85 is considered synergistic (27). CX-4945 doses of 14 µM and 16 µM in combination with 50 nM doxorubicin had synergistic CI values of 0.797 and 0.756, respectively. Although not 76

meeting the criteria for synergy, additional doses of 5 µM, 10 µM, and 20 µM CX-4945 exhibited an additive effect in combination with 50 nM doxorubicin with CI values around 0.9.

These data demonstrate that CK2 inhibitors induce apoptosis and increase sensitivity of U937 cells to Doxorubicin. Since BCL2A1 is an anti-apoptotic protein, these data suggest that one of the mechanisms that contributes to the effects of CK2 inhibition involves Ikaros-mediated repression of BCL2A1.

Discussion

BCL2A1 is a member of the BCL-2 family, a family of proteins that is well known for its involvement in anti-apoptotic pathways with implications in development and tumorigenesis (29). BCL2A1 is primarily expressed in the hematopoietic system and functions to support leukocyte survival by sequestering pro-apoptotic B-cell lymphoma 2

(BCL2) proteins (38,39). BCL2A1 is able to reduce the release of pro-apoptotic cytochrome c from mitochondria and to block caspase activation (29). Overexpression of

BCL2A1 has been observed in many human cancers, including hematological malignancies such as ALL, chronic lymphocytic leukemia (CLL), chronic myelogenous leukemia (CML) and several types of lymphomas, as well as solid tumors such as melanoma, breast, colon, and stomach cancer (38,40,41). The overexpression of BCL2A1 has been shown to promote hematopoietic transformation and prevent apoptosis in vivo and should be considered a proto-oncogene for myeloid leukemogenesis (29). Here, we used a gain-of-function and loss-of-function approach to demonstrate that Ikaros

77

represses transcription of the BCL2A1 gene in U937 cells. The direct repression of

BCL2A1 by Ikaros suggests that one mechanism of Ikaros tumor suppression involves the negative regulation of BCL2A1 expression. These data reveal a novel tumor suppressor function for Ikaros in AML.

The increased expression and activity of CK2 in AML and its inverse correlation with survival make this enzyme an attractive target for AML therapy. Since CK2 is a highly-expressed kinase that has over 300 substrates (42), the mechanism of therapeutic action of CK2 inhibitors is unknown. It has been shown that CK2 directly phosphorylates the Ikaros protein at multiple residues (19,24,43,44). The primary effect of CK2- mediated phosphorylation of Ikaros appears to be a reduction in the ability of Ikaros to bind DNA and thus the loss of Ikaros ability to function as a transcriptional regulator.

The functional studies presented here demonstrate that BCL2A1 expression is regulated by the CK2-Ikaros axis. Due to the large number of CK2 substrates, we cannot exclude the possibility that CK2 inhibition might result in repression of BCL2A1 via additional mechanism(s) that are independent of Ikaros activity. However, CK2 inhibition did not result in BCL2A1 repression in cells in which Ikaros was knocked down by shRNA.

These data support the hypothesis that Ikaros activity is a major requirement for BCL2A1 repression following CK2 inhibition in leukemia. Our results showing that the CK2-

Ikaros axis regulates BCL2A1 expression reveal a novel mechanism that functions in regulating malignant transformation and drug resistance in AML and provide a mechanistic rationale for the use of CK2 inhibitors in targeted therapy for AML. The synergism observed with the use of the CK2 inhibitor CX-4945 and Doxorubicin provides support for in vivo testing of these drugs in preclinical models of AML. 78

The CK2-Ikaros axis has recently been reported to regulate transcription of a large set of genes that promote cell cycle progression and the PI3K pathway in B-ALL (16).

This report is the first evidence that the CK2-Ikaros axis controls the expression of a gene that can regulate both malignant transformation and apoptosis in AML. Based on the data reported here, we present a model for the regulation of drug resistance in AML by the

CK2-Ikaros axis (Figure 2.7). As we identify additional target genes of CK2-Ikaros in

AML, we anticipate that this model will become more complex over time.

In summary, we identify BCL2A1 as a novel target of the Ikaros tumor suppressor in AML. We show that the CK2-Ikaros signaling pathway regulates transcription of

BCL2A1 and that treatment with CK2 inhibitors results in BCL2A1 repression and apoptosis. Our results demonstrate synergistic cytotoxic activity of the CK2 inhibitor CX-

4945 with doxorubicin in U937 cells, suggesting that CK2 inhibitors could be incorporated into some existing therapeutic regimens for AML treatment. The presented data provide support for the use of CK2 inhibitors as targeted therapy for AML and further testing of therapeutic efficacy of CX-4945 in preclinical models of AML.

79

Figures

of Peaks

Number

Log2 FoldLog2 Change

Figure 2.1. Treatment with CX-4945 enhances Ikaros binding to the promoter of BCL2A1 in AML cells. A, CK2 kinase assay was performed using -GTP on whole cell lysates of U937 cells treated with 20 µM CX-4945 for 72 hours or left untreated – results 80

are values from 3 replicates. B, An equal aliquot of nuclear extract of treated and untreated U937 cells were subjected to Western blot to document that the equal amount of protein in both treated and untreated cells were used for immunoprecipitation (left panel) Radioactivity of excised bands corresponding to 3 largest Ikaros isoforms – IK-H, IK-1 and IK-2 (middle panel) was counted on scintillation counter – results represent values from 3 replicates (right panel). C, Ikaros peaks are redistributed to the transcription start site (TSS) following treatment with 10 µM CX-4945 for 72 hours. Comparison of Ikaros binding at the TSS in untreated and CX-4945 treated cells. D, Ikaros binding to the promoter of BCL2A1 in U937 cells is increased after treatment with 10 µM CX-4945. Y-axis represents log 2 fold change enrichment of Ikaros binding. (**p<0.01).

81

Figure 2.2. Ikaros represses the transcription of BCL2A1. A, Representative experiment showing repression of the BCL2A1 luciferase promoter construct by the Ikaros-expressing vector pcDNA3.1-IK in comparison to pcDNA3.1 empty vector control in HEK-293T cells. Luciferase activity was normalized to pcDNA3.1 and pROM empty vector controls. Data is presented as the average +/- SD. B, U937 cells were infected with either empty vector (MIG) or human Ikaros (MIG-Ikaros) and sorted based on expression of GFP (left panel). Relative transcription of BCL2A1 measured by qRT- PCR (right panel). Data are averages +/- SD. CERS1 shown as negative control. C, U937 cells were transfected with either scrambled shRNA (negative control) or Ikaros specific shRNA (Ikaros shRNA) (left panel). Relative transcription of BCL2A1 measured by qRT- PCR (right panel). CERS1 represented as a negative control. Data are averages +/- SD. (**p<0.01).

82

Figure 2.3. CK2 inhibits Ikaros-mediated repression of BCL2A1. A, U937 cells transfected with either scrambled shRNA (negative control) or CK2-specific shRNA (left panel). Effect of CK2 knockdown on gene expression was analyzed by qRT-PCR (right panel). CERS1 shown as negative control. B, Pharmacological inhibition of CK2 decreases the transcription of BCL2A1 in U937 cells. qRT-PCR analysis of RNA from U937 cells treated for 48 hours with various concentrations of CX-4945. SGMS1 is represented as a negative control. C, Pharmacological inhibition of CK2 decreases the transcription of BCL2A1 in Primary AML cells. qRT-PCR analysis of RNA from Pheresis #4 cells treated for 48 hours with various concentrations of CX-4945. SGMS1 is represented as a negative control. (**p<0.01).

83

Figure 2.4. Inhibition of CK2 enhances the ability of Ikaros to bind to the promoter of BCL2A1. A, qRT-PCR indicates increased Ikaros binding at the promoter of BCL2A1 in U937 cells after treatment with 10 µM CX-4945 for 72 hours. IgG – non-specific IgG control antibody. Ikaros – affinity-purified anti-Ikaros antibody. SIRT6 represented as a negative control. B, qRT-PCR indicates increased Ikaros binding at the promoter of BCL2A1 in primary AML cells after treatment with 20 µM CX-4945 for 72 hours. IgG – non-specific IgG control antibody. Ikaros – affinity-purified anti-Ikaros antibody. SIRT6 represented as a negative control. C, Effect of Ikaros knockdown on change in BCL2A1 expression following CK2 inhibition with CX-4945. qRT-PCR analysis of RNA from U937 cells transfected with either scrambled shRNA (negative control) or Ikaros-specific shRNA and treated for 24hr with CK2 inhibitor. CERS1 represented as negative control. Ikaros expression shows knockdown of Ikaros after transfection with Ikaros shRNA.

84

Figure 2.5. Treatment with CX-4945 promotes heterochromatin formation at the promoter of BCL2A1. Epigenetic changes at the BCL2A1 promoter after CX-4945 treatment correspond with Ikaros binding peak. Decrease in H3K4me2 and increase in H3K9me3 and H3K27me3 are observed in close proximity to the Ikaros binding peak after treatment with CX-4945. Y-axis represents the log 2 fold change.

85

Figure 2.6. Apoptosis is increased in U937 following treatment with CX-4945. Upper right quadrant: late apoptotic cells; lower right quadrant: early apoptotic cells; upper left quadrant: dead cells and debris; lower left quadrant: live cells. A, Treatment with 10 µM CX-4945 for 48 hours increases apoptosis in U937 cells. B, Treatment with 10 µM CX- 4945 for 48 hours increases apoptosis in Pheresis #4 primary AML cells. C, CX-4945 synergizes with doxorubicin in U937 cells. CalcuSyn software was used to determine synergistic combinations of CX-4945 and doxorubicin. 3: CI = 0.797; 4: CI = 0.756.

86

Figure 2.7. Proposed model representing regulation of BCL2A1 by the CK2-Ikaros axis in leukemia. High levels of CK2 expression result in a permissive chromatin and high expression of BCL2A1. Ikaros binding is inhibited by phosphorylation (top). Inhibition of CK2 promotes the formation of a repressive chromatin environment. Dephosphorylation of Ikaros allows for Ikaros to bind the upstream regulatory element of BCL2A1 and repress its transcription.

87

References

1. Mrózek, K., Heerema, N. A., and Bloomfield, C. D. (2004) Cytogenetics in acute leukemia. Blood Rev 18, 115-136 2. Marcucci, G., Mrózek, K., and Bloomfield, C. D. (2005) Molecular heterogeneity and prognostic biomarkers in adults with acute myeloid leukemia and normal cytogenetics. Curr Opin Hematol 12, 68-75 3. Gilliland, D. G., and Tallman, M. S. (2002) Focus on acute leukemias. Cancer Cell 1, 417-420 4. Dash, A., and Gilliland, D. G. (2001) Molecular genetics of acute myeloid leukaemia. Best Pract Res Clin Haematol 14, 49-64 5. Castilla, L. H., Garrett, L., Adya, N., Orlic, D., Dutra, A., Anderson, S., Owens, J., Eckhaus, M., Bodine, D., and Liu, P. P. (1999) The fusion gene Cbfb-MYH11 blocks myeloid differentiation and predisposes mice to acute myelomonocytic leukaemia. Nat Genet 23, 144-146 6. Mehrotra, B., George, T. I., Kavanau, K., Avet-Loiseau, H., Moore, D., Willman, C. L., Slovak, M. L., Atwater, S., Head, D. R., and Pallavicini, M. G. (1995) Cytogenetically aberrant cells in the stem cell compartment (CD34+lin-) in acute myeloid leukemia. Blood 86, 1139-1147 7. Robak, T., and Wierzbowska, A. (2009) Current and emerging therapies for acute myeloid leukemia. Clin Ther 31 Pt 2, 2349-2370 8. Münstermann, U., Fritz, G., Seitz, G., Lu, Y. P., Schneider, H. R., and Issinger, O. G. (1990) Casein kinase II is elevated in solid human tumours and rapidly proliferating non- neoplastic tissue. Eur J Biochem 189, 251-257 9. Faust, R. A., Niehans, G., Gapany, M., Hoistad, D., Knapp, D., Cherwitz, D., Davis, A., Adams, G. L., and Ahmed, K. (1999) Subcellular immunolocalization of protein kinase CK2 in normal and carcinoma cells. Int J Biochem Cell Biol 31, 941-949 10. Allende, J. E., and Allende, C. C. (1995) Protein kinases. 4. Protein kinase CK2: an enzyme with multiple substrates and a puzzling regulation. FASEB J 9, 313-323 11. Gomes, A. M., Soares, M. V., Ribeiro, P., Caldas, J., Póvoa, V., Martins, L. R., Melão, A., Serra-Caetano, A., de Sousa, A. B., Lacerda, J. F., and Barata, J. T. (2014) Adult B- cell acute lymphoblastic leukemia cells display decreased PTEN activity and constitutive

88

hyperactivation of PI3K/Akt pathway despite high PTEN protein levels. Haematologica 99, 1062-1068 12. Silva, A., Yunes, J. A., Cardoso, B. A., Martins, L. R., Jotta, P. Y., Abecasis, M., Nowill, A. E., Leslie, N. R., Cardoso, A. A., and Barata, J. T. (2008) PTEN posttranslational inactivation and hyperactivation of the PI3K/Akt pathway sustain primary T cell leukemia viability. J Clin Invest 118, 3762-3774 13. Kelliher, M. A., Seldin, D. C., and Leder, P. (1996) Tal-1 induces T cell acute lymphoblastic leukemia accelerated by casein kinase IIalpha. EMBO J 15, 5160-5166 14. Kim, J. S., Eom, J. I., Cheong, J. W., Choi, A. J., Lee, J. K., Yang, W. I., and Min, Y. H. (2007) Protein kinase CK2alpha as an unfavorable prognostic marker and novel therapeutic target in acute myeloid leukemia. Clin Cancer Res 13, 1019-1028 15. Quotti Tubi, L., Gurrieri, C., Brancalion, A., Bonaldi, L., Bertorelle, R., Manni, S., Pavan, L., Lessi, F., Zambello, R., Trentin, L., Adami, F., Ruzzene, M., Pinna, L. A., Semenzato, G., and Piazza, F. (2013) Inhibition of protein kinase CK2 with the clinical- grade small ATP-competitive compound CX-4945 or by RNA interference unveils its role in acute myeloid leukemia cell survival, p53-dependent apoptosis and daunorubicin- induced cytotoxicity. J Hematol Oncol 6, 78 16. Song, C., Gowda, C., Pan, X., Ding, Y., Tong, Y., Tan, B. H., Wang, H., Muthusami, S., Ge, Z., Sachdev, M., Amin, S. G., Desai, D., Gowda, K., Gowda, R., Robertson, G. P., Schjerven, H., Muschen, M., Payne, K. J., and Dovat, S. (2015) Targeting casein kinase II restores Ikaros tumor suppressor activity and demonstrates therapeutic efficacy in high- risk leukemia. Blood 17. Fujiwara, T., O'Geen, H., Keles, S., Blahnik, K., Linnemann, A. K., Kang, Y. A., Choi, K., Farnham, P. J., and Bresnick, E. H. (2009) Discovering hematopoietic mechanisms through genome-wide analysis of GATA factor chromatin occupancy. Mol Cell 36, 667- 681 18. Wang, Z., Zang, C., Rosenfeld, J. A., Schones, D. E., Barski, A., Cuddapah, S., Cui, K., Roh, T. Y., Peng, W., Zhang, M. Q., and Zhao, K. (2008) Combinatorial patterns of histone acetylations and methylations in the human genome. Nat Genet 40, 897-903 19. Dovat, S., Ronni, T., Russell, D., Ferrini, R., Cobb, B. S., and Smale, S. T. (2002) A common mechanism for mitotic inactivation of C2H2 zinc finger DNA-binding domains. Genes Dev 16, 2985-2990

89

20. Sayegh, J., Cao, J., Zou, M. R., Morales, A., Blair, L. P., Norcia, M., Hoyer, D., Tackett, A. J., Merkel, J. S., and Yan, Q. (2013) Identification of small molecule inhibitors of Jumonji AT-rich interactive domain 1B (JARID1B) histone demethylase by a sensitive high throughput screen. J Biol Chem 288, 9408-9417 21. Hawley, R. G., Lieu, F. H., Fong, A. Z., and Hawley, T. S. (1994) Versatile retroviral vectors for potential use in gene therapy. Gene Ther 1, 136-138 22. Li, Z., Song, C., Ouyang, H., Lai, L., Payne, K. J., and Dovat, S. (2012) Cell cycle- specific function of Ikaros in human leukemia. Pediatr Blood Cancer 59, 69-76 23. Ronni, T., Payne, K. J., Ho, S., Bradley, M. N., Dorsam, G., and Dovat, S. (2007) Human Ikaros function in activated T cells is regulated by coordinated expression of its largest isoforms. J Biol Chem 282, 2538-2547 24. Gurel, Z., Ronni, T., Ho, S., Kuchar, J., Payne, K. J., Turk, C. W., and Dovat, S. (2008) Recruitment of ikaros to pericentromeric heterochromatin is regulated by phosphorylation. J Biol Chem 283, 8291-8300 25. Greco, W. R., Bravo, G., and Parsons, J. C. (1995) The search for synergy: a critical review from a response surface perspective. Pharmacol Rev 47, 331-385 26. Berenbaum, M. C. (1985) The expected effect of a combination of agents: the general solution. J Theor Biol 114, 413-431 27. Gowda, R., Jones, N. R., Banerjee, S., and Robertson, G. P. (2013) Use of Nanotechnology to Develop Multi-Drug Inhibitors For Cancer Therapy. J Nanomed Nanotechnol 4 28. Pierre, F., Chua, P. C., O'Brien, S. E., Siddiqui-Jain, A., Bourbon, P., Haddach, M., Michaux, J., Nagasawa, J., Schwaebe, M. K., Stefan, E., Vialettes, A., Whitten, J. P., Chen, T. K., Darjania, L., Stansfield, R., Bliesath, J., Drygin, D., Ho, C., Omori, M., Proffitt, C., Streiner, N., Rice, W. G., Ryckman, D. M., and Anderes, K. (2011) Pre- clinical characterization of CX-4945, a potent and selective small molecule inhibitor of CK2 for the treatment of cancer. Mol Cell Biochem 356, 37-43 29. Métais, J. Y., Winkler, T., Geyer, J. T., Calado, R. T., Aplan, P. D., Eckhaus, M. A., and Dunbar, C. E. (2012) BCL2A1a over-expression in murine hematopoietic stem and progenitor cells decreases apoptosis and results in hematopoietic transformation. PLoS One 7, e48267

90

30. Koipally, J., and Georgopoulos, K. (2000) Ikaros interactions with CtBP reveal a repression mechanism that is independent of histone deacetylase activity. J Biol Chem 275, 19594-19602 31. Koipally, J., and Georgopoulos, K. (2002) Ikaros-CtIP interactions do not require C- terminal binding protein and participate in a deacetylase-independent mode of repression. J Biol Chem 277, 23143-23149 32. Brown, K. E., Guest, S. S., Smale, S. T., Hahm, K., Merkenschlager, M., and Fisher, A. G. (1997) Association of transcriptionally silent genes with Ikaros complexes at centromeric heterochromatin. Cell 91, 845-854 33. Ernst, P., Hahm, K., and Smale, S. T. (1993) Both LyF-1 and an Ets protein interact with a critical promoter element in the murine terminal transferase gene. Mol Cell Biol 13, 2982-2992 34. Trinh, L. A., Ferrini, R., Cobb, B. S., Weinmann, A. S., Hahm, K., Ernst, P., Garraway, I. P., Merkenschlager, M., and Smale, S. T. (2001) Down-regulation of TDT transcription in CD4(+)CD8(+) thymocytes by Ikaros proteins in direct competition with an Ets activator. Genes Dev 15, 1817-1832 35. Meshorer, E., and Misteli, T. (2006) Chromatin in pluripotent embryonic stem cells and differentiation. Nat Rev Mol Cell Biol 7, 540-546 36. Hawkins, R. D., Hon, G. C., Lee, L. K., Ngo, Q., Lister, R., Pelizzola, M., Edsall, L. E., Kuan, S., Luu, Y., Klugman, S., Antosiewicz-Bourget, J., Ye, Z., Espinoza, C., Agarwahl, S., Shen, L., Ruotti, V., Wang, W., Stewart, R., Thomson, J. A., Ecker, J. R., and Ren, B. (2010) Distinct epigenomic landscapes of pluripotent and lineage-committed human cells. Cell Stem Cell 6, 479-491 37. Pommier, Y., Sordet, O., Antony, S., Hayward, R. L., and Kohn, K. W. (2004) Apoptosis defects and chemotherapy resistance: molecular interaction maps and networks. Oncogene 23, 2934-2949 38. Vogler, M. (2012) BCL2A1: the underdog in the BCL2 family. Cell Death Differ 19, 67- 74 39. Akatsuka, Y., Nishida, T., Kondo, E., Miyazaki, M., Taji, H., Iida, H., Tsujimura, K., Yazaki, M., Naoe, T., Morishima, Y., Kodera, Y., Kuzushima, K., and Takahashi, T. (2003) Identification of a polymorphic gene, BCL2A1, encoding two novel hematopoietic lineage-specific minor histocompatibility antigens. J Exp Med 197, 1489- 1500 91

40. Nagy, B., Lundán, T., Larramendy, M. L., Aalto, Y., Zhu, Y., Niini, T., Edgren, H., Ferrer, A., Vilpo, J., Elonen, E., Vettenranta, K., Franssila, K., and Knuutila, S. (2003) Abnormal expression of apoptosis-related genes in haematological malignancies: overexpression of MYC is poor prognostic sign in mantle cell lymphoma. Br J Haematol 120, 434-441 41. Brinkmann, K., and Kashkar, H. (2014) Targeting the mitochondrial apoptotic pathway: a preferred approach in hematologic malignancies? Cell Death Dis 5, e1098 42. Pinna, L. A. (2002) Protein kinase CK2: a challenge to canons. J Cell Sci 115, 3873-3878 43. Gómez-del Arco, P., Maki, K., and Georgopoulos, K. (2004) Phosphorylation controls Ikaros's ability to negatively regulate the G(1)-S transition. Mol Cell Biol 24, 2797-2807 44. Popescu, M., Gurel, Z., Ronni, T., Song, C., Hung, K. Y., Payne, K. J., and Dovat, S. (2009) Ikaros stability and pericentromeric localization are regulated by protein phosphatase 1. J Biol Chem 284, 13869-13880

92

Chapter 3

Ikaros and CK2 Regulate Transcription of MTHFR and CDA

93

Introduction

Methotrexate (MTX) is one of the most versatile and widely administered anticancer drugs. A benefit to the use of methotrexate as a cancer therapy is that, unlike doxorubicin, methotrexate does not exhibit long term toxicity (1). Methotrexate as a drug for the treatment of acute lymphoblastic leukemia (ALL) has greatly improved patient outcome (2). However, approximately 20 percent of ALL patients have leukemia that is not susceptible to methotrexate cytotoxicity. Both inherent and acquired resistance to methotrexate have been implicated in lack of response to methotrexate treatment (3).

After administration of the drug, methotrexate gains entry to cells through the reduced folate carrier (RFC) and is polyglutamylated. Both methotrexate and its metabolites inhibit dihydrofolate reductase (DHFR) (4). This inhibition results in a decrease of tetrahydrofolate (THF) regeneration that promotes THF deficiency. THF deficiency leads to the inhibition of purines and thymidine synthesis, inhibition of DNA replication, and ultimately cell death (4).

Several genetic variants have been identified that affect the response to methotrexate therapy. The two genes that are most well studied in methotrexate resistance are methylenetetrahydrofolate reductase (MTHFR) and thymidylate synthase

(TS) (5). MTHFR facilitates methylation reactions and is responsible for conversion of the major circulating form of folate, 5-methyltetrahydrofolate. Alterations in the enzymatic activity of MTHFR result in an imbalance in the folate pools and has implications in the sensitivity of cells to methotrexate treatment (6).

94

Although often used as a treatment for ALL, Methotrexate has been shown to have little therapeutic effect in the treatment of AML. Development of drug resistance is one possible explanation for the poor response of AML to MTX treatment (7,8). Despite the fact that some promising results have been achieved with novel antifolates, the most significant improvements in cancer chemotherapy have been based on optimization of available treatments rather than the development of new drugs. Combination treatments that can increase the sensitivity of AML to MTX treatment would be of great benefit to patients.

Cytarabine together with an anthracycline such as doxorubicin have been considered the standard of care for the treatment of AML for several decades (9).

Cytarabine acts as an antimetabolite and is highly active against leukemia. Specifically, cytarabine is a pyrimidine nucleoside analog that inhibits DNA synthesis and thus halts the cell cycle at S phase (10). Therefore, rapidly dividing cells are most affected.

Intracellular metabolism of cytarabine to cytosine arabinoside triphosphate allows for the damage of DNA by multiple mechanisms. The metabolite inhibits alpha-DNA polymerase, inhibits DNA repair, and incorporates into DNA, which prevents the addition of purines or pyrimidines to the elongating DNA (10). Inefficient cellular uptake, reduction of activating enzymes, and increased degradation of the active metabolite have all been described as methods for the development of resistance to cytarabine (11). Cytidine deaminase (CDA) deaminates cytarabine to produce an inactive metabolite (12). While this is a normal process that occurs with cytarabine treatment and renders a small portion of the active drug inactive, increased exposure of

AML cells to cytarabine desensitizes cells and promotes the metabolism of cytarabine by 95

CDA. The decreased efficacy of cytarabine for the treatment of carcinomas has been associated with higher basal CDA activity (12). In AML, patient CDA status can be used as a marker for predicting outcome after treatment and to determine adequate dosing to optimize the efficacy to toxicity balance (13). Alterations to the administration of cytarabine have been examined in order to identify a more efficacious treatment regimen to increase the efficacy of cytarabine, but none have proven to be significantly better than current cytarabine protocols (14). Combination therapies that act to inhibit CDA activity could provide a solution to this problem and increase the efficacy of current cytarabine treatment regimens.

We have identified genes involved in both methotrexate (MTHFR) and cytarabine

(CDA) resistance as binding targets of Ikaros via ChIP-Seq. To further examine the relationship between Ikaros and these drug resistance genes and whether Ikaros is responsible for the regulation of these genes, the following studies were carried out.

Experimental Procedures

Cell Culture

U937 cells were obtained from the American Type Culture Collection (ATCC),

Manassas, VA. Cells were cultured in RPMI 1640 growth medium (CellGro) supplemented with 10% fetal bovine serum (FBS) (HyClone). HEK-293T cells were cultured in DMEM (CellGro) supplemented with 10% FBS. Pheresis #4 primary cells were obtained from a patient with AML from Loma Linda University (Loma Linda, CA). 96

DNA Constructs

Luciferase reporter constructs for the promoters of MTHFR and CDA and the pLightSwitch-Rom empty vector were purchased from SwitchGear Genomics. Each promoter construct contained the Ikaros binding peak identified by ChIP-Seq. The pMSCV bicistronic retroviral vector (MIG vector) and the pMSCV bicistronic retroviral vector encoding wild-type human HA-tagged Ikaros (IKZF1) which contains a 5’ long- terminal-repeat-driven Ikaros, internal ribosome entry side (IRES), and enhanced green fluorescent protein (EGFP) were described previously (15).

Transfection

HEK-293T cells were transfected using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. U937 cells were transfected using the Neon transfection system

(Invitrogen) according to the manufacturer’s protocol

Gene Expression Analysis by qRT-PCR

Total RNA was isolated from cells using the QIAshredder and RNeasy Mini Kit

(QIAGEN). Complementary DNA (cDNA) was generated from 1 μg total RNA using

Superscript First-Strand Synthesis System (Invitrogen). qRT-PCR was performed using a

StepOne Plus real-time PCR machine (Applied Biosystems) with PerfeCTa SYBR Green

FastMix (Quanta Biosciences) and primers for the indicated genes (primer pairs listed in

97

Table 3.1). Values were normalized to 18s RNA and the relative expression values were determined by the 2-ΔΔCt method.

Table 3.1: Primers used for target gene studies

Gene Forward Primer Reverse Primer

MTHFR 5’ – CCTCCAACGAGACCTGTGG – 3’ 5’ – GTGTCAGGTTGCTGGAGAGG – 3’

CDA 5’ – TGAGGAAGGAAGTTGGATGG – 3’ 5’ – AGCCTCAGTAAGGTGCCAGA – 3’

Chemicals

DMSO, TSA, and Methotrexate (MTX) were purchased from Sigma-Aldrich (St Louis,

MO). Cytarabine was purchased from VWR International (Radnor, PA). CX-4945 was provided by Cylene Pharmaceuticals (San Diego, CA).

Antibodies

The antibody used for chromatin immunoprecipitation of Ikaros has been described previously (16). Other antibodies used were as follows: HDAC1 (Abcam ab7028);

H3K4me2 (Millipore 07-030); H3K4me3 (Abcam ab8580); H3K9me3 (Abcam ab8898);

H3K27me3 (Millipore 07-449); anti-Rabbit IgG (ab46540, Abcam). All antibodies used for ChIP-Seq were Encode validated antibodies.

98

qChIP qChIP assays were performed as reported previously (17). Experiments were performed with a minimum of two biological replicates. qRT-PCR was performed using PerfeCTa

Sybr Green FastMix (Quanta Biosciences) and a StepOne Plus real-time PCR system

(Applied Biosystems) according to the manufacturer’s instructions. Enrichment of the

ChIP’d DNA was normalized to a non-specific IgG control. The comparative Ct method was used for quantification, and fold enrichment was calculated using the formula: 2-

CTsample/2-CTinput, where CT equals the threshold cycle number. Primer Pairs Listed in Table 3.1.

ChIP-Seq

ChIP-Seq assays were performed as previously described (17,18) and as described below.

For ChIP-Seq libraries, 2x106 U937 cells were cross-linked for 20 min in PBS containing

1% formaldehyde and the reaction was stopped by adding glycine. Cell pellets were flash frozen and stored at −80°C. Chromatin was fragmented using a Bioruptor (Diagenode) for 30 min on ice (30s pulses, 30s pauses) to produce fragments ~200-400bp in size. Goat anti-rabbit IgG Dynabeads (Invitrogen) were coated with affinity purified anti-Ikaros antibody and incubated with chromatin overnight on a rotator at 4°C. Protein/DNA complexes were captured with a Magnetic Particle Concentrator (Invitrogen). Crosslinks were reversed and samples were treated with proteinase K and RNaseA. DNA was recovered using the QIAquick PCR Purification kit (QIAGEN). ChIP-Seq libraries were created using 18 cycles of amplification with ChIPseq DNA sample prep kit (Illumina). 99

Libraries were validated using the Agilent Technologies 2100 Bioanalyzer. ChIP-Seq libraries were sequenced at the High Throughput Genomics Center at University of

Washington, Seattle and at the Core Facility at Penn State College of Medicine, Hershey,

PA. Sequence fastq files were aligned to the UCSC human genome assembly HG19 using the Eland application (Illumina), allowing no more than two mismatches per sequence.

Only sequences aligning uniquely to the human genome were used to identify peaks.

Peaks calling was achieved using Cisgenome2.0 and SISSRS.

Luciferase Assay

Luciferase assays were performed using the LightSwitch Luciferase Assay System

(SwitchGear Genomics). HEK-293T cells were seeded into 24-well plates. After 24 hours, cells were transiently transfected with 0.15 μg of indicated promoter reporter constructs or pROM vector and 0.15 μg of pcDNA3.1-Ikaros or pcDNA3.1 vector in triplicate for each group using lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. Twenty-four hours after transfection, cells were lysed in 100

μl of LightSwitch Assay Solution (SwitchGear Genomics) and rocked at RT for 30min according to the manufacturer’s instructions. Lysates were measured by luminometer

(Promega GloMax 20/20 Luminometer). Luciferase activities were calculated as fold change relative to vector only cells and normalized to pcDNA3.1 vector readings. All transfection and reporter assays were performed in triplicate. If HDAC inhibitors were used, 0.5nM TSA was added 4-6 hours post transfection and cells were incubated for 24 hours prior to reading.

100

Ikaros and CK2 shRNA Knockdown shRNA constructs for human Ikaros (IKZF1) and human CK2 (CSNK2A1) in a GFP vector (pGFP-V-RS) were purchased from Origene. U937 cells were transiently transfected with 3.0μg of plasmids per well in 24-well plates using the Neon Transfection

System (Invitrogen). After transfection for 1 day, U937 cells with transfection efficiency ranges from ~80% (green cells) and more than 95% cell viability were further treated with 10μM CX-4945 or Nontreatment control (0.01% DMSO) for 2days. The cells were then harvested for total RNA isolation and total lysate extraction. The 29-mer scrambled shRNA pGFPV-RS vector was used as a control. The knockdown of Ikaros and CK2 was confirmed by qPCR (16).

Cell Proliferation Assays

Proliferation assays with CX-4945, MTX and ara-C were performed as described previously (19). Briefly, cells in logarithmic growth phase were harvested and resuspended to a final concentration of 1 x 105 cells/mL of fresh medium and treated with

CX-4945, MTX, ara-C, or a combination thereof. 104 cells were plated per well in triplicate in a 96-well clear bottom plate (Costar 3603) and incubated for two days at

37°C in a humidified incubator with 5% CO2. On day 2, WST-1 reagent (Roche,

11644807001) was added at 10μl/well and cells were allowed to incubate for an additional 4h at 37°C. Absorbance was measured at 440/690nm using a BioTek Synergy

Mx plate reader.

101

Drug Synergy Analysis with Calcusyn Software

Drug synergy for combinations of CX-4945 with ara-C, and MTX were demonstrated by using the Chou-Talalay method to determine the combination index (CI) using Calcusyn software (20,21). CI values are illustrated graphically for each combination with an isobologram. Values <0.85 are synergistic, 0.9-1.1 are additive, and >1.1 are antagonistic

(22).

Annotation of ChIP-Seq Peaks

Peak annotation with associated genes was retrieved using a Perl script. Peak information was input from the CisGenome output peak files in “COD” format. The nearest gene to each peak and associated locations (gene start, gene end, 5’UTR, 3’UTR, exon and intron) in the human genome was identified using the CisGenome built-in program

“refgene_getnearestgene” and “refgene_getlocationsummary” with parameters r = 0

(TSS-up, TES-down), up=100000 (bp) and down=100000 (bp), respectively. The gene names from HGNC and expression levels from RNA-Seq analysis were also input. The output data file includes the integrated information of peaks, genes and gene expression levels.

Distribution of peaks in different gene regulatory regions

The number of peaks in different gene regulatory regions (e.g. promoter, intergenic regions, etc.) was calculated using a Perl script. The peaks located at the TSS

102

(Transcriptional Start Site) and TES (Transcriptional End Site) were also counted by a

Perl script program. In this program, the maximum number of peaks at a position was also recorded. The percentage of the number of peaks at each position over the maximum number of peaks at a position was computed.

Relation of Ikaros, HDAC1 and Histone Modification Peaks

IK-HDAC1 peaks were defined by at least 1 bp overlap between the Ikaros and the

HDAC1 peaks. The peaks of histone modification markers were mapped onto the Ikaros-

HDAC1 peaks. Peaks were classified into the following groups for further study:

Enhancer region (-3 to -10 kb upstream of TSS); Promoter regions (from -3kb to +2kb from TSS); Intragene regions (+2kb from TSS to -1 kb from TES); TES regions (-1kb to

1kb from TES); Intergene regions (+1 kb from TES of a gene to -10 kb from TSS of next gene)

103

Results

Ikaros Binds to the Promoters of Several Genes Implicated in AML and Drug Resistance

To confirm that Ikaros binds to the promoters of MTHFR and CDA, as indicated by our ChIP-Seq, we performed quantitative chromatin immunoprecipitation (qChIP) of

Ikaros in U937 cells and also in Pheresis #4 primary AML patient cells. While ChIP-Seq is an excellent screening tool to identify target genes of transcriptional factors, the ability to measure the DNA-binding affinity toward specific sites is accomplished by qChIP.

Control reactions with a non-specific IgG antibody failed to precipitate any significant protein-DNA complexes containing these promoters of interest. Binding ability at least two-fold greater than the IgG reactions was considered to be significant binding. qRT-

PCR confirmed that Ikaros binds to the promoters of MTHFR and CDA, and that binding was at least two-fold greater than the IgG control reactions for each gene (Figure 3.1).

To determine whether binding of Ikaros to these genes is clinically relevant, we also investigated Ikaros binding in Pheresis #4 primary AML cells by qChIP. Because of the clinical importance of MTHFR and CDA and their roles in drug resistance, we assessed the ability of Ikaros to bind to their promoters in cells from an AML patient.

Similar to our qChIP in U937 cells, both promoters were shown to be bound by Ikaros, and binding at these promoters was at least two-fold greater than the IgG control (Figure

3.2).

104

These results indicate that Ikaros is able to bind to the promoters of these genes in vivo and indicates a potential role for Ikaros in regulation of MTHFR and CDA in both an

AML cell line and in primary AML cells.

Gain-of-Function: Ikaros Represses the Transcription of MTHFR and CDA

To investigate the functional significance of the observed Ikaros binding at the promoter regions of MTHFR and CDA, we performed both gain-of-function and loss-of- function experiments. First, we performed luciferase reporter assays to determine whether Ikaros is able to directly regulate expression of these genes. Plasmids containing the promoter regions of MTHFR or CDA were purchased from SwitchGear Genomics.

Each construct contained the region of the promoter in which an Ikaros binding peak had been observed by ChIP-Seq. HEK-293T cells transfected with a promoter construct and an empty vector control showed strong luciferase activity, indicating that the promoter constructs alone could stimulate increased expression of the luciferase gene. Cells transfected with a promoter construct in addition to a plasmid containing full length human Ikaros had a decrease in luciferase activity of about 10-fold when compared to cells with a promoter construct and an empty vector (Figure 3.3). The decrease in luciferase activity was significant, indicating that Ikaros is able to directly bind to the promoters of MTHFR and CDA and repress their transcription.

The luciferase reporter assays indicated that Ikaros is able to directly repress the transcription of these genes. However, HEK-293T cells have no endogenous expression of Ikaros, and the use of promoter constructs creates a system in which the direct effect

105

on target gene transcription is measured with minimal or no chromatin remodeling machinery. In contrast, the native system that occurs in AML consists of endogenous

Ikaros and target genes in a complex regulatory environment that includes chromatin remodeling, limitations to chromatin accessibility, and the full-length promoters of each gene. Next, we wanted to determine whether increased Ikaros expression is able to affect the transcription of MTHFR and CDA in AML cells. To do this, we overexpressed Ikaros in U937 cells. Cells were infected with retrovirus containing wild type Ikaros (MIG-IK) or an empty retroviral control (MIG Vector). Both vectors contained GFP. Sorting based on GFP expression identified successfully infected cells, and the expression of the Ikaros target genes was analyzed using qRT-PCR. Our results indicated that cells with overexpressed Ikaros have decreased expression of MTHFR and CDA (Figure 3.4). A 4- fold reduction in MTHFR expression and a 3-fold reduction in CDA expression was observed when Ikaros was overexpressed. As expected, the gene repression observed in

AML cells was less extreme than the repression observed in HEK-293T cells. These experimental conditions are more physiologically relevant to AML and indicate a complex regulatory environment governing the expression of these target genes in AML.

Loss-of-Function: Knockdown of Ikaros Results in Increased Transcription of MTHFR and CDA

To ensure that this decrease in expression was occurring through Ikaros, we performed Ikaros loss-of-function experiments. U937 cells were transfected with Ikaros- specific shRNA or scrambled shRNA to determine whether Ikaros is responsible for the

106

observed gene repression. Again, changes in gene expression were analyzed by qRT-

PCR from isolated RNA. An increase in expression of the Ikaros target genes MTHFR and CDA was observed in U937 cells transfected with Ikaros shRNA in comparison to the scrambled control, indicating that loss of functional Ikaros results in increased expression of these gene targets (Figure 3.5).

Taken together, these results demonstrate by both gain-of-function and loss-of- function experiments that Ikaros represses transcription of MTHFR and CDA.

Molecular Inhibition of CK2 Results in Transcriptional Repression of Ikaros Target Genes in a Manner Similar to Ikaros Overexpression

Previous studies have established that phosphorylation of Ikaros is one of the major mechanism that regulates its function (23,24). The phosphorylation state of Ikaros has important implications for both the function and localization of Ikaros.

Phosphorylation of Ikaros by CK2 has been shown to decrease the affinity of Ikaros for

DNA in addition to regulating its subcellular localization. Dysregulation of CK2 has been reported in leukemia, which suggests that the DNA-binding ability of Ikaros may also be impaired in leukemia.

In order to determine whether high expression levels of CK2 in leukemia are interfering with the ability of Ikaros to regulate its target genes, we asked whether inhibition of CK2 would affect Ikaros target genes in the same way as Ikaros overexpression. To evaluate this question, we used CK2-specific shRNA to inhibit the function of CK2 in U937 cells (molecular inhibition) and analyzed RNA transcription by

107

qRT-PCR. A decrease in expression of MTHFR and CDA of about 10-fold was observed in cells transfected with CK2-specific shRNA when compared to cells that received a scrambled control (Figure 3.6). The observed decrease in transcription of the Ikaros target genes after CK2 inhibition was similar to the decrease observed when Ikaros was overexpressed in U937 cells. Therefore, these results indicate that inhibition of CK2 results in a decrease in gene transcription of MTHFR and CDA and suggest that overexpression of CK2 in leukemia cells may act to promote expression of these drug resistance genes.

Pharmacological Inhibition of CK2 with CX-4945 Represses Ikaros Target Genes in AML

To determine the effect of the specific CK2 inhibitor CX-4945 on the expression of Ikaros target genes, U937 and Pheresis #4 cells were treated with several concentrations of CX-4945 (pharmacological inhibition) for 48 hours, and RNA was extracted and analyzed by qRT-PCR. In U937 cells, the expression of MTHFR and CDA was repressed in a dose-dependent manner following treatment with CX-4945 (Figure

3.7). Likewise, a dose-dependent repression of gene expression was observed for

MTHFR and CDA in Pheresis #4 primary AML cells (Figure 3.8). Interestingly, the observed repression of MTHFR and CDA was even greater in the primary AML cells than in U937 cells, suggesting that this repression might have biological relevance.

108

Inhibition of CK2 Enhances the Ability of Ikaros to Bind to the Promoters of its Target Genes

To determine whether binding of Ikaros to the promoters of MTHFR and CDA was increased after CK2 inhibition, we performed qChIP in U937 cells that were treated with CX-4945. Our results indicate that Ikaros is able to bind to the promoters of interest in the untreated cells and that treatment with CX-4945 increases the binding of Ikaros at the promoters of its target genes (Figure 3.9). We also assessed the influence of CX-

4945 treatment on Ikaros binding in the Pheresis #4 Primary AML cells. Ikaros binding was increased at the promoters of MTHFR and CDA after CX-4945 treatment (Figure

3.10). Thus, we show that the binding affinity of Ikaros is increased toward two clinically relevant genes with implications in drug resistance after inhibition of CK2.

Transcriptional Repression of Ikaros Target Genes Following CK2 Inhibition Requires Ikaros Activity

CK2 has a large number of substrates. Thus, a possible explanation of the above results is that CK2 inhibition results in transcriptional repression of MTHFR and CDA via a pathway that does not involve Ikaros. To determine whether Ikaros activity is essential for the observed changes in gene expression of MTHFR and CDA after CK2 inhibition,

Ikaros was knocked down with Ikaros-specific shRNA in U937 cells, and RNA levels were measured using qRT-PCR. The expression level of cells containing only control shRNA was determined, and the effect of CK2 inhibition on the transcription of our target genes in all experimental shRNA experiments were compared to the control. Cells 109

that received a control shRNA and CX-4945 treatment showed a significant decrease in gene expression (Figure 3.11). Knockdown of Ikaros with an Ikaros-specific shRNA was able to restore expression of MTHFR and CDA in cells treated with CX-4945, indicating that the ability of CK2 inhibition to repress transcription of our target genes was lost when Ikaros was knocked down. These results indicate that activity of Ikaros is essential for the repression of MTHFR and CDA and suggest that repression of these genes by

Ikaros is responsible for the decrease in gene expression observed following CK2 inhibition.

Taken together, these findings provide strong evidence that phosphorylation of

Ikaros by CK2 decreases Ikaros activity as a transcriptional repressor. Furthermore, these data suggest that inhibition of CK2 could enhance the repression of Ikaros target genes in leukemia.

Epigenetic Changes Related to Ikaros Target Genes

To understand the activity of Ikaros in AML and to identify whether global epigenetic changes occur in U937 cells after CX-4945 treatment, we performed ChIP-Seq in U937 cells that were untreated or treated with CX-4945. In addition to identifying many Ikaros target genes, we observed an overall increase in Ikaros binding after treatment with CX-4945. Because Ikaros is known to associate with HDAC1, we also assessed the global binding of HDAC1 by ChIP-Seq. Ikaros has been shown to regulate its target genes by facilitating chromatin remodeling. In order to assess whether global epigenetic changes occur along with the observed increase in binding of Ikaros after CX-

110

4945 treatment, we performed ChIP-Seq of 18 histone modifications that are known to be associated with active chromatin, repressive chromatin, or heterochromatin. We also performed ChIP-Seq of Histone H1. We observed global epigenetic changes that strongly suggest that the function of Ikaros as a tumor suppressor is enhanced after treatment with CX-4945. Changes in the binding of HDAC1 suggest that treatment with

CX-4945 results in more repressive chromatin.

As stem cells differentiate they leave a highly malleable epigenetic state and enter a more restrictive, lineage appropriate epigenetic state. Bivalent chromatin, which contains both activating and repressive histone modifications within close proximity, is common in stem cells, and is often indicated by the presence of the repressive marker

H3K27me3 and the active marker H3K4me3. Upon differentiation, cells display increased levels of heterochromatin (25). Lineage-committed cells exhibit vast increases in H3K27me3 and H3K9me3, another negative marker, at the promoters of developmental genes, with many promoters being marked by both modifications simultaneously (26). Our ChIP-Seq indicated that Ikaros binding peaks at the promoter of CDA correlate with increased H3K27me3 (Figure 3.12). These results suggest that

Ikaros might play a role in the differentiation of myeloid cells by recruitment of its target genes to heterochromatin.

When looking at positive chromatin markers in the promoter region of CDA near

Ikaros binding peaks, we observed little change in the presence of H3K4me3, a known marker of positive chromatin (not shown). However, we observed a substantial decrease in the level of H3K4me2 that corresponds with the Ikaros binding peaks at the promoters

(Figure 3.12). Taken together with the increase of heterochromatin in these regions, 111

these observations indicate that chromatin is becoming more closed, even if H3K4me3 is still present at the promoter.

Ikaros Represses Transcription of its Target Genes via HDAC1-dependent and - Independent Mechanisms

Increased HDAC1 binding was observed in the promoter region of several Ikaros target genes. This increased binding suggests that these genes may be regulated by

Ikaros in association with HDAC1. Interestingly, increased binding of HDAC1 does not appear to be promoting deacetylation of the promoter, but rather, Ikaros and HDAC1 together are promoting heterochromatin formation in these target genes.

Because Ikaros has been known to associate with several complexes involved in chromatin remodeling, including HDAC1, and HDAC1 binding was increased at the promoters of several of our genes of interest, we investigated whether histone deacetylases are involved in the Ikaros-mediated repression of MTHFR and CDA. We performed luciferase assays with and without the histone deacetylase inhibitor TSA in

HEK-293T cells. While treatment with TSA eliminated the Ikaros-mediated repression of MTHFR, the Ikaros-mediated repression of CDA was only partly restored after TSA treatment. These results indicate that repression of MTHFR by Ikaros requires histone deacetylase activity. Repression of CDA by Ikaros appears to occur through a different mechanism. Alternatively, the slight restoration of gene expression after histone deacetylase inhibition could suggest that HDACs are involved in the regulation of CDA but that other factors are also involved. Further analysis of the relationship between

112

Ikaros and histone deacetylases in the regulation of these genes is necessary to elucidate the mechanism.

CX-4945 and Combination Therapy

Mutations and deletions of Ikaros have been observed in myeloid neoplasms, and dominant-negative Ikaros has been shown to induce AML in a humanized xenograft mouse model (27-30). Expression changes of proteins involved in drug metabolism are an established mechanism of drug resistance, and cancer cells are known to employ compensatory mechanisms to contribute to the development of drug resistance (31,32).

Therefore, combination therapies that inhibit multiple parts of the same pathway could provide a means of circumventing drug resistance to chemotherapy. Combination treatment that includes inhibition of CK2 may enhance the activity of the remaining wild- type Ikaros allele in patients that have an Ikaros deletion or mutation.

CX-4945 Synergizes with Methotrexate

We have shown that CK2 inhibition can enhance the ability of Ikaros to regulate expression of MTHFR, a key enzyme involved in folate metabolism. Altered enzymatic activity of MTHFR results in changes to the available folate pools and increased accumulation of the major form of folate. These mechanisms alter the effectiveness of antifolate drugs and provide a mechanism by which cells can develop resistance to methotrexate. Repression of MTHFR by Ikaros is impaired when CK2 is overexpressed,

113

allowing for increased folate production. For this reason, we sought to assess whether treatment with the specific CK2 inhibitor CX-4945 could enhance the sensitivity of AML cells to methotrexate when used in combination. Cell proliferation assays were carried out in U937 cells to examine the toxicity of combination treatment with CX-4945 and methotrexate. Cells were treated with multiple concentrations of CX-4945 or methotrexate alone or with combinations of CX-4945 and methotrexate. Cellular proliferation was measured by spectrophotometry using WST1 reagent.

Synergistic concentrations were observed in U937 cells treated with a combination of CX-4945 and methotrexate. Five doses of CX-4945 in combination with

5 nM methotrexate met the CalcuSyn criteria for drug synergy (Figure 3.13). CX-4945 doses of 6 µM, 10 µM, 14 µM, 16 µM, and 20 µM in combination with 5 nM methotrexate had synergistic CI values of 0.759, 0.344, 0.146, 0.217, and 0.226, respectively. Lower CI values indicate increased synergy.

These results indicate that CK2 inhibition enhances the cytotoxic effect of methotrexate and that combinations of CX-4945 and methotrexate have greater therapeutic efficacy than either methotrexate or CX-4945 treatment alone. The CI values calculated by the CalcuSyn software indicate that the specific CK2 inhibitor CX-4945 is able to work synergistically with methotrexate and that combination of these drugs could potentially be used to sensitize AML patients to methotrexate treatment.

114

CX-4945 Synergizes with Cytarabine

We have shown that CK2 inhibition can enhance the ability of Ikaros to regulate expression of CDA, whose enzymatic activity is directly related to cytarabine inactivation. Altered enzymatic activity of CDA results in the inactivation of a larger portion of cytarabine and leads to an increased resistance to cytarabine treatment.

Repression of CDA by Ikaros is impaired when CK2 is overexpressed, allowing for increased inactivation of cytarabine. For this reason, we sought to assess whether treatment with the specific CK2 inhibitor CX-4945 could enhance the sensitivity of AML cells to cytarabine when used in combination. Cell proliferation assays were carried out in U937 cells to examine the toxicity of combination treatment with CX-4945 and cytarabine. Cells were treated with multiple concentrations of CX-4945 or cytarabine alone or with combinations of CX-4945 and cytarabine. Cellular proliferation was measured by spectrophotometry using WST1 reagent.

Synergistic concentrations were observed in U937 cells treated with a combination of CX-4945 and cytarabine. Two doses of CX-4945 in combination with 20 nM cytarabine met the CalcuSyn criteria for drug synergy (Figure 3.14). CX-4945 doses of 10 µM and 20 µM in combination with 20 nM cytarabine had synergistic CI values of

0.66 and 0.72, respectively.

These results indicate that CX-4945 treatment enhances the cytotoxic effect of cytarabine and that combinations of CX-4945 and cytarabine have greater therapeutic efficacy than either cytarabine or CX-4945 treatment alone. The CI values calculated by the CalcuSyn software indicate that the specific CK2 inhibitor CX-4945 is able to work

115

synergistically with cytarabine and that a combination of these drugs could potentially be used to increase the efficacy of cytarabine for the treatment of AML.

Discussion

Increasing the intensity of current chemotherapy regimens is limited due to toxicity. Therefore, improvements in outcome will largely depend on the development of novel targeted therapies along with rationally designed combination treatments. We have shown here that Ikaros represses the transcription of MTHFR and CDA, genes with translational significance due to their roles in the development of resistance to commonly used chemotherapeutic drugs. Our data show that the Ikaros/CK2 axis is responsible for the regulation of these genes and provides a model by which to understand the poor prognosis associated with overexpression of CK2 in AML. Decreased Ikaros binding as a result of Ikaros mutation, overexpression of CK2, or the simultaneous occurrence of both prevents the repression of these genes, an event that is required for normal hematopoietic function. Dysregulation of these genes results in increased expression, which enables the cell to evade the toxic effects of standard chemotherapeutic drugs and results in the proliferation of drug resistant cells. The ability of CK2 to regulate the tumor suppressor function of Ikaros has been well characterized, but the clinical significance of this regulation has remained unclear. Here we show that Ikaros functions as a tumor suppressor in AML and that Ikaros regulates genes with potential clinical relevance.

116

The current standard chemotherapy agents for the treatment of AML have been used for decades and are largely effective for achieving remission. Not all leukemia cells are sensitive to the initial induction therapy, however, which results in a sub-population of leukemic cells that remains in the bone marrow and increases the risk of relapse. In general, these cells survive by employing mechanisms to circumvent drug toxicity.

When relapse occurs, the proliferation of drug resistant cells dominates and contributes to poor patient prognosis. New treatment regimens are needed to combat drug resistance and to prevent relapse, and combination treatments will provide the most likely candidates. In addition to heterogeneity of tumors, many pathways and mechanisms have been implicated in the development of drug resistance. These include decreased apoptosis, mutation or overexpression of drug targets, increased drug efflux, metabolic inactivation, and increased DNA repair. Our knowledge of these mechanisms allows for the design and characterization of drugs that target one or more of these pathways as a means to increase the efficacy of chemotherapeutic drugs. Here, we identified a role for the Ikaros/CK2 axis in the regulation of folate metabolism and cytarabine metabolism through the regulation of the genes MTHFR, and CDA.

MTHFR is one of the most well studied genes involved in MTX resistance.

Patient response to MTX treatment varies widely and can be partly attributed to genetic differences affecting metabolism of folate and MTX. While a significant percentage of all leukemia patients do not respond to MTX, in AML, most patients are considered to be resistant to MTX treatment. Here we show that the Ikaros/CK2 axis is able to repress the expression of MTHFR. Combination treatment of methotrexate and CX-4945 is able to sensitize U937 AML cells to methotrexate treatment. Further studies should be done in 117

xenograft mice to determine whether combination treatments of CX-4945 and methotrexate could provide a means for using methotrexate successfully for the treatment of AML.

CDA has a major impact on ara-C pharmacokinetics by degrading ara-C to its inactive metabolite, and levels of CDA expression have prognostic significance in AML treatment. We have identified CDA as a direct target of Ikaros and suggest that the decreased regulatory ability of Ikaros in AML contributes to CDA overexpression, providing a potential mechanism to explain the poor prognosis associated with both CDA and Ikaros in AML. Cytarabine has been established as the standard of care for AML treatment, and our data suggest that the therapeutic efficacy of cytarabine could be increased in combination with the specific CK2 inhibitor CX-4945. These findings have profound implications for the future of treatment in AML.

Our data suggest that inhibition of CK2 could be used as an effective means to sensitize AML to current drug treatments. Inhibition of CK2 would be useful as a treatment for patients with Ikaros haploinsufficiency by promoting the function of the remaining wild type allele. Even patients with no defect in Ikaros could benefit from combination treatments with CK2 inhibitors, as reduction of CK2 overexpression promotes normal Ikaros function.

Here we show that loss of Ikaros results in upregulation of MTHFR and CDA, genes with implications for drug resistance in AML. Inhibition of CK2 with CX-4945 enhances the DNA-binding ability of Ikaros and increases the repression of these drug resistance genes in an Ikaros-dependent manner. These results provide a mechanism by which loss of Ikaros function can contribute to drug resistance in AML. 118

Figures

qChIP of Ikaros in U937 Cells

8 ** 7

6 IgG 5 Ikaros 4

3 *

2 Relative Relative FoldEnrichment

1

0 MTHFR CDA SIRT6

Figure 3.1. Ikaros binds to the promoters of MTHFR and CDA in U937 cells. qChIP followed by qRT-PCR indicates that Ikaros binds to the promoters of MTHFR and CDA in U937 cells after treatment with 10 µM CX-4945 for 72 hours. IgG – non-specific IgG control antibody. Ikaros – affinity-purified anti-Ikaros antibody. SIRT6 represented as a negative control. (*p<0.05; **p<0.01)

119

qChIP of Ikaros in Pheresis #4 Cells 3 ** 2.5 **

2 IgG Anti-Ikaros 1.5

FoldEnrichment 1

0.5

0 MTHFR CDA ANAPC1

Figure 3.2. Ikaros binds to the promoters of MTHFR and CDA in Pheresis #4 primary AML cells. qChIP followed by qRT-PCR indicates that Ikaros binds to the promoters of MTHFR and CDA in Pheresis #4 primary AML cells after treatment with 10 µM CX-4945 for 72 hours. IgG – non-specific IgG control antibody. Ikaros – affinity- purified anti-Ikaros antibody. ANAPC1 represented as a negative control. (**p<0.01).

120

Luciferase Assay in HEK-293T Cells 1.2

1

Activity 0.8

0.6 pcDNA

Luciferase IK

0.4 Relative 0.2 ** ** 0 MTHFR CDA

Figure 3.3. Ikaros directly represses the transcription of MTHFR and CDA. Representative experiments showing repression of the MTHFR and CDA luciferase promoter constructs by the Ikaros-expressing vector pcDNA3.1-IK (black) in comparison to pcDNA3.1 empty vector control (grey) in HEK-293T cells. Luciferase activity was normalized to pcDNA3.1 and pROM empty vector controls. Data is presented as the average +/- SD. (**p<0.01).

121

Overexpression of Ikaros in Ikaros U937 Cells Overexpression 1.2 Mig Vector 30 Mig-Ikaros Mig

Vector PCR)

1 PCR) 25

- - Mig- Ikaros 0.8 20

0.6 15

0.4 ** 10 Relative Relative Expression (qRT ** Relative Expression (qRT 0.2 5

0 0 MTHFR CDA CERS1 Ikaros

Figure 3.4. Ikaros overexpression represses the transcription of MTHFR and CDA. Relative transcription of target genes measured by qRT-PCR after U937 cells were infected with either empty vector (Mig Vector) or human Ikaros (Mig-Ikaros) and sorted based on expression of GFP. CERS1 represented as a negative control. Data are averages +/- SD. (**p<0.01).

122

Ikaros Knockdown in U937 Ikaros Knockdown Cells 1.2 2 Scram-shRNA ** 1.8 ** Ikaros shRNA 1 Scram-

shRNA

PCR) PCR) - - 1.6 Ikaros 0.8 1.4 shRNA 1.2 0.6 1 0.8 0.4

0.6

Relative Relative Expression (qRT Relative Expression (qRT 0.4 0.2 0.2 0 0 MTHFR CDA CERS1 Ikaros

Figure 3.5. Knockdown of Ikaros increases the transcription of MTHFR and CDA. Relative transcription of target genes measured by qRT-PCR after U937 cells were transfected with either scrambled shRNA (negative control) or Ikaros-specific shRNA (Ikaros shRNA). CERS1 represented as a negative control. Ikaros gene expression levels indicate a 2-fold knockdown of Ikaros after shRNA transfection. Gene transcription was analyzed by qRT-PCR. Data are averages +/- SD. (**p<0.01).

123

CK2 Knockdown in U937 Cells CK2 Knockdown

1.2 scram-shRNA 1.2 CK2 shRNA 1 1 Scram-

shRNA

PCR) PCR)

- - CK2- 0.8 0.8 shRNA

0.6 0.6

0.4 0.4

Relative Relative Expression (qRT Relative Expression (qRT 0.2 0.2 ** ** 0 0 MTHFR CDA CERS1 CK2

Figure 3.6. Knockdown of CK2 decreases the transcription of MTHFR and CDA. U937 cells were transfected with either scrambled shRNA (negative control) or CK2- specific shRNA (CK2 shRNA). CERS1 represented as a negative control. CK2 expression levels show that CK2 is knocked down after transfection with CK2 shRNA. Effect of CK2 knockdown on target gene expression was analyzed by qRT-PCR, standardized to 18S RNA and presented as the average +/- SD. (**p<0.01).

124

CX-4945 Treatment in U937 Cells 1.4 DMSO Control 10 uM CX-4945 1.2 20 uM CX-4945

PCR) 40 uM CX-4945 - 1

0.8 ** 0.6 ** ** 0.4

Relative Relative Expression (qRT ** 0.2 **

0 MTHFR CDA SGMS1

Figure 3.7. Pharmacological inhibition of CK2 decreases the transcription of MTHFR and CDA in U937 cells. qRT-PCR analysis of RNA from U937 cells treated for 48 hours with various concentrations of CX-4945. SGMS1 is represented as a negative control. (p<0.01).

125

CX-4945 Treatment in Primary AML Cells 1.2

1

PCR) - 0.8 DMSO Control

0.6 10 uM CX-4945 20 uM CX-4945

0.4 40 uM CX-4945

Relative Relative Expression (qRT ** ** 0.2 ** ** ** ** 0 MTHFR CDA SGMS1

Figure 3.8. Pharmacological inhibition of CK2 decreases the transcription of MTHFR and CDA in Primary AML cells. qRT-PCR analysis of RNA from Pheresis #4 cells treated for 48h with various concentrations of CX-4945. SGMS1 is represented as a negative control. (**p<0.01).

126

qChIP of Ikaros in U937 Cells 45

40 **

35 ** 30 IgG - WT 25 Ikaros - WT 20 IgG - CX-4945 Ikaros - CX-4945

FoldEnrichment 15

10 ** 5 *

0 MTHFR CDA

Figure 3.9. Inhibition of CK2 enhances the ability of Ikaros to bind to the promoters of MTHFR and CDA in U937 Cells. qRT-PCR indicates increased Ikaros binding at the promoter of MTHFR and CDA after treatment with 10 µM CX-4945 for 72 hours. IgG – non-specific IgG control antibody. Ikaros – affinity-purified anti-Ikaros antibody. SIRT6 represented as a negative control. (*p<0.05; **p<0.01).

127

qChIP of Ikaros in Pheresis #4 Cells 14 ** 12

IgG - DMSO 10 IK - DMSO

8 IgG - 40 uM CX-4945 IK - 40 uM CX-4945

6 ** Fold Fold Enrichment 4 ** ** 2

0 MTHFR CDA ANAPC1

Figure 3.10. Inhibition of CK2 enhances the ability of Ikaros to bind to the promoters of MTHFR and CDA in Pheresis #4 primary AML cells. qRT-PCR indicates increased Ikaros binding at the promoter of MTHFR and CDA after treatment with 10 µM CX-4945 for 72 hours. IgG – non-specific IgG control antibody. Ikaros – affinity-purified anti-Ikaros antibody. SIRT6 represented as a negative control. (**p<0.01).

128

Ikaros Blocks the Effect of CK2 Inhibition on Target Gene Expression

1.6 scram shRNA ** scram shRNA + CX-4945

1.4 PCR)

- IK-shRNA + CX-4945 1.2

1 **

0.8

0.6 ** **

Relative Relative Expression (qRT 0.4

0.2

0 Ikaros MTHFR CDA CERS1

Figure 3.11. Transcriptional Repression of MTHFR and CDA Following CK2 Inhibition Requires Ikaros Activity. Effect of Ikaros knockdown on changes in gene expression of MTHFR and CDA following CK2 inhibition with CX-4945. qRT-PCR analysis of RNA from U937 cells transfected with either scrambled shRNA (negative control) or Ikaros-specific shRNA (IK-shRNA) and treated for 24hr with 20 µM CX- 4945. CERS1 represented as a negative control. Ikaros expression shows knockdown of Ikaros after transfection with Ikaros shRNA. Compared to scram-shRNA: **p<0.01; compared to scram-shRNA + CX-4945: **p<0.01.

129

Figure 3.12. Treatment with CX-4945 promotes heterochromatin formation at the promoter of CDA. Epigenetic changes at the CDA promoter after CX-4945 treatment correspond with Ikaros binding peak. Decrease in H3K4me2 and increase in H3K9me3 and H3K27me3 are observed in close proximity to the Ikaros binding peak after treatment with CX-4945. U937 cells were treated for 72 hours with 10 µM CX-4945 and then subjected to ChIP-Seq. Y-axis represents the log 2 fold change.

130

Figure 3.13. CX-4945 Synergizes with methotrexate in U937 cells. Cell proliferation assays showed synergy between several concentrations of CX-4945 and 5 nM MTX in U937 cells. Cells were treated with various combinations for 48 hours before being assayed with WST1 reagent. CalcuSyn software was used to determine synergistic combinations of CX-4945 and methotrexate. 1: CI = 0.759; 3: CI = 0.344; 4: CI = 0.146; 5: CI = 0.217; 6: CI = 0.226

131

Figure 3.14. CX-4945 Synergizes with cytarabine in U937 cells. Cell proliferation assays showed synergy between several concentrations of CX-4945 and 20 nM cytarabine in U937 cells. Cells were treated with various combinations for 48 hours before being assayed with WST1 reagent. CalcuSyn software was used to determine synergistic combinations of CX-4945 and cytarabine. 3: CI = 0.66; 6: CI = 0.72.

132

References

1. Salliot, C., and van der Heijde, D. (2009) Long-term safety of methotrexate monotherapy in patients with rheumatoid arthritis: a systematic literature research. Ann Rheum Dis 68, 1100-1104 2. Escherich, G., Horstmann, M. A., Zimmermann, M., Janka-Schaub, G. E., and group, C. s. (2010) Cooperative study group for childhood acute lymphoblastic leukaemia (COALL): long-term results of trials 82,85,89,92 and 97. Leukemia 24, 298-308 3. Wojtuszkiewicz, A., Peters, G. J., van Woerden, N. L., Dubbelman, B., Escherich, G., Schmiegelow, K., Sonneveld, E., Pieters, R., van de Ven, P. M., Jansen, G., Assaraf, Y. G., Kaspers, G. J., and Cloos, J. (2015) Methotrexate resistance in relation to treatment outcome in childhood acute lymphoblastic leukemia. J Hematol Oncol 8, 61 4. Fotoohi, A. K., and Albertioni, F. (2008) Mechanisms of antifolate resistance and methotrexate efficacy in leukemia cells. Leuk Lymphoma 49, 410-426 5. Ulrich, C. M., Robien, K., and Sparks, R. (2002) Pharmacogenetics and folate metabolism -- a promising direction. Pharmacogenomics 3, 299-313 6. Krajinovic, M., Lemieux-Blanchard, E., Chiasson, S., Primeau, M., Costea, I., and Moghrabi, A. (2004) Role of polymorphisms in MTHFR and MTHFD1 genes in the outcome of childhood acute lymphoblastic leukemia. Pharmacogenomics J 4, 66-72 7. Bertino, J. R., Sawicki, W. L., Cashmore, A. R., Cadman, E. C., and Skeel, R. T. (1977) Natural resistance to methotrexate in human acute nonlymphocytic leukemia. Cancer Treat Rep 61, 667-673 8. Dedhar, S., Hartley, D., Fitz-Gibbons, D., Phillips, G., and Goldie, J. H. (1985) Heterogeneity in the specific activity and methotrexate sensitivity of dihydrofolate reductase from blast cells of acute myelogenous leukemia patients. J Clin Oncol 3, 1545- 1552 9. Teuffel, O., Leibundgut, K., Lehrnbecher, T., Alonzo, T. A., Beyene, J., and Sung, L. (2013) Anthracyclines during induction therapy in acute myeloid leukaemia: a systematic review and meta-analysis. Br J Haematol 161, 192-203 10. Ewald, B., Sampath, D., and Plunkett, W. (2008) Nucleoside analogs: molecular mechanisms signaling cell death. Oncogene 27, 6522-6537

133

11. Galmarini, C. M., Mackey, J. R., and Dumontet, C. (2001) Nucleoside analogues: mechanisms of drug resistance and reversal strategies. Leukemia 15, 875-890 12. Ohta, T., Hori, H., Ogawa, M., Miyahara, M., Kawasaki, H., Taniguchi, N., and Komada, Y. (2004) Impact of cytidine deaminase activity on intrinsic resistance to cytarabine in carcinoma cells. Oncol Rep 12, 1115-1120 13. Mahlknecht, U., Dransfeld, C. L., Bulut, N., Kramer, M., Thiede, C., Ehninger, G., and Schaich, M. (2009) SNP analyses in cytarabine metabolizing enzymes in AML patients and their impact on treatment response and patient survival: identification of CDA SNP C-451T as an independent prognostic parameter for survival. Leukemia 23, 1929-1932 14. Momparler, R. L. (2013) Optimization of cytarabine (ARA-C) therapy for acute myeloid leukemia. Exp Hematol Oncol 2, 20 15. Hawley, R. G., Lieu, F. H., Fong, A. Z., and Hawley, T. S. (1994) Versatile retroviral vectors for potential use in gene therapy. Gene Ther 1, 136-138 16. Dovat, S., Ronni, T., Russell, D., Ferrini, R., Cobb, B. S., and Smale, S. T. (2002) A common mechanism for mitotic inactivation of C2H2 zinc finger DNA-binding domains. Genes Dev 16, 2985-2990 17. Fujiwara, T., O'Geen, H., Keles, S., Blahnik, K., Linnemann, A. K., Kang, Y. A., Choi, K., Farnham, P. J., and Bresnick, E. H. (2009) Discovering hematopoietic mechanisms through genome-wide analysis of GATA factor chromatin occupancy. Mol Cell 36, 667- 681 18. Wang, Z., Zang, C., Rosenfeld, J. A., Schones, D. E., Barski, A., Cuddapah, S., Cui, K., Roh, T. Y., Peng, W., Zhang, M. Q., and Zhao, K. (2008) Combinatorial patterns of histone acetylations and methylations in the human genome. Nat Genet 40, 897-903 19. Sayegh, J., Cao, J., Zou, M. R., Morales, A., Blair, L. P., Norcia, M., Hoyer, D., Tackett, A. J., Merkel, J. S., and Yan, Q. (2013) Identification of small molecule inhibitors of Jumonji AT-rich interactive domain 1B (JARID1B) histone demethylase by a sensitive high throughput screen. J Biol Chem 288, 9408-9417 20. Greco, W. R., Bravo, G., and Parsons, J. C. (1995) The search for synergy: a critical review from a response surface perspective. Pharmacol Rev 47, 331-385 21. Berenbaum, M. C. (1985) The expected effect of a combination of agents: the general solution. J Theor Biol 114, 413-431

134

22. Gowda, R., Jones, N. R., Banerjee, S., and Robertson, G. P. (2013) Use of Nanotechnology to Develop Multi-Drug Inhibitors For Cancer Therapy. J Nanomed Nanotechnol 4 23. Gurel, Z., Ronni, T., Ho, S., Kuchar, J., Payne, K. J., Turk, C. W., and Dovat, S. (2008) Recruitment of ikaros to pericentromeric heterochromatin is regulated by phosphorylation. J Biol Chem 283, 8291-8300 24. Popescu, M., Gurel, Z., Ronni, T., Song, C., Hung, K. Y., Payne, K. J., and Dovat, S. (2009) Ikaros stability and pericentromeric localization are regulated by protein phosphatase 1. J Biol Chem 284, 13869-13880 25. Meshorer, E., and Misteli, T. (2006) Chromatin in pluripotent embryonic stem cells and differentiation. Nat Rev Mol Cell Biol 7, 540-546 26. Hawkins, R. D., Hon, G. C., Lee, L. K., Ngo, Q., Lister, R., Pelizzola, M., Edsall, L. E., Kuan, S., Luu, Y., Klugman, S., Antosiewicz-Bourget, J., Ye, Z., Espinoza, C., Agarwahl, S., Shen, L., Ruotti, V., Wang, W., Stewart, R., Thomson, J. A., Ecker, J. R., and Ren, B. (2010) Distinct epigenomic landscapes of pluripotent and lineage-committed human cells. Cell Stem Cell 6, 479-491 27. Theocharides, A. P., Dobson, S. M., Laurenti, E., Notta, F., Voisin, V., Cheng, P. Y., Yuan, J. S., Guidos, C. J., Minden, M. D., Mullighan, C. G., Torlakovic, E., and Dick, J. E. (2015) Dominant-negative Ikaros cooperates with BCR-ABL1 to induce human acute myeloid leukemia in xenografts. Leukemia 29, 177-187 28. Grossmann, V., Kohlmann, A., Zenger, M., Schindela, S., Eder, C., Weissmann, S., Schnittger, S., Kern, W., Müller, M. C., Hochhaus, A., Haferlach, T., and Haferlach, C. (2011) A deep-sequencing study of chronic myeloid leukemia patients in blast crisis (BC- CML) detects mutations in 76.9% of cases. Leukemia 25, 557-560 29. Jäger, R., Gisslinger, H., Passamonti, F., Rumi, E., Berg, T., Gisslinger, B., Pietra, D., Harutyunyan, A., Klampfl, T., Olcaydu, D., Cazzola, M., and Kralovics, R. (2010) Deletions of the transcription factor Ikaros in myeloproliferative neoplasms. Leukemia 24, 1290-1298 30. Nacheva, E. P., Grace, C. D., Brazma, D., Gancheva, K., Howard-Reeves, J., Rai, L., Gale, R. E., Linch, D. C., Hills, R. K., Russell, N., Burnett, A. K., and Kottaridis, P. D. (2013) Does BCR/ABL1 positive acute myeloid leukaemia exist? Br J Haematol 161, 541-550

135

31. Longley, D. B., and Johnston, P. G. (2005) Molecular mechanisms of drug resistance. J Pathol 205, 275-292 32. Holohan, C., Van Schaeybroeck, S., Longley, D. B., and Johnston, P. G. (2013) Cancer drug resistance: an evolving paradigm. Nat Rev Cancer 13, 714-726

136

Chapter 4

Ikaros and CK2 Regulate Transcription of DLX1 and DLX2

137

Introduction

Homeobox (HOX) genes encode transcription factors that share a characteristic protein fold and regulate differentiation and cellular proliferation during development (1).

Dysregulation of HOX genes has been linked to the development of leukemia. Various common chromosomal translocations join the nucleoporin gene NUP98 with several

HOX genes (2-5). MLL, which functions to regulate HOX levels, is involved in many chromosomal translocations in both AML and ALL. Rearrangement of MLL results in the overexpression of HOX genes such as HOXA9 (6,7). Overexpression of HOX genes has been linked to myeloid proliferation and differentiation (8-11). AML patients with high levels of HOX expression generally have intermediate-risk cytogenetics as well as high levels of fms-like tyrosine kinase-3 (FLT3) or higher incidence of FLT3 mutation

(12,13). These observations suggest that dysregulation of HOX genes might contribute to malignant transformation in AML.

Expression of FLT3 is essential during early hematopoiesis and affects differentiation of hematopoietic cells (14). FLT3 is a tyrosine kinase whose downstream targets include RAS, STAT5 and PI3K (15,16). Constitutive activation of FLT3 has been observed in 30% of AML, and the internal tandem duplication (ITD) mutation of FLT3 is associated with greater incidence of relapse in AML (17-20). FLT3 is known to regulate the activity of the homeobox transcription factors DLX1 and DLX2 (21). This relationship was been observed in AML cell lines and in samples from patients with

AML, and the mechanistic link between FLT3 and DLX1 expression likely involves

MAPK signaling (21). 138

This mechanistic link between FLT3 and DLX1/2 suggests that changes in expression of the DLX genes could affect the regulation of FLT3. Because of their involvement in FLT3 activation and the identification of Ikaros binding peaks at the promoters of DLX1 and DLX2 by ChIP-Seq, we examined the Ikaros/CK2 axis and its involvement in regulation of DLX1 and DLX2 in AML.

Experimental Procedures

Cell Culture

U937 cells were obtained from the American Type Culture Collection (ATCC),

Manassas, VA. Cells were cultured in RPMI 1640 growth medium (CellGro) supplemented with 10% fetal bovine serum (FBS) (HyClone). HEK-293T cells were cultured in DMEM (CellGro) supplemented with 10% FBS.

DNA Constructs and shRNA

Luciferase reporter constructs for the promoter of DLX1 and the pLightSwitch-Rom empty vector were purchased from SwitchGear Genomics. The DLX1 promoter construct contained the Ikaros binding peak identified by ChIP-Seq. The pMSCV bicistronic retroviral vector (MIG vector) and the pMSCV bicistronic retroviral vector encoding wild-type human HA-tagged Ikaros (IKZF1) which contains a 5’ long-terminal-repeat-

139

driven Ikaros, internal ribosome entry side (IRES), and enhanced green fluorescent protein (EGFP) were described previously (22).

Transfection

HEK-293T cells were transfected using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. U937 cells were transfected using the Neon transfection system

(Invitrogen) according to manufacturer’s protocol

Gene Expression Analysis by qRT-PCR

Total RNA was isolated from cells using the QIAshredder and RNeasy Mini Kit

(QIAGEN). Complementary DNA (cDNA) was generated from 1 μg total RNA using

Superscript First-Strand Synthesis System (Invitrogen). qRT-PCR was performed using a

StepOne Plus real-time PCR machine (Applied Biosystems) with PerfeCTa SYBR Green

FastMix (Quanta Biosciences) and primers for the indicated genes (primer pairs listed in

Table 4.1). Values were normalized to 18s RNA and the relative expression values were determined by the 2-ΔΔCt method.

Table 4.1: Primers used for target gene studies

Gene Forward Primer Reverse Primer

DLX1 5’ – AATTGGGTTCCTTCCTGTCC – 3’ 5’ – GGCTGTTGAGACTTTCTGGC – 3’

DLX2 5’ – ATTACATTGGCTGCTGGAGG – 3’ 5’ – AGACGGGAAAGAGCAGAGGT – 3’

140

Antibodies and Chemicals

The antibody used for chromatin immunoprecipitation of Ikaros has been described previously (23). The anti-Rabbit IgG was purchased from Abcam (ab46540). CX-4945 was provided by Cylene Pharmaceuticals (San Diego, CA).

qChIP qChIP assays were performed as reported previously (24). Experiments were performed with a minimum of two biological replicates. Real-time PCR was performed using

PerfeCTa Sybr Green FastMix (Quanta Biosciences) and a StepOne Plus real-time PCR system (Applied Biosystems) according to manufacturer’s instructions. Enrichment of the

ChIP’d DNA was normalized to a non-specific IgG control. The comparative Ct method was used for quantification and fold enrichment was calculated using the formula: 2-

CTsample/2-CTinput, where CT equals the threshold cycle number. Primer Pairs Listed in Table 4.1.

ChIP-Seq

ChIP-Seq assays were performed as previously described (24,25) and as described below.

For ChIP-Seq libraries, 2x106 U937 cells were cross-linked for 20 min in PBS containing

1% formaldehyde and reaction the was stopped by adding glycine. Cell pellets were flash frozen and stored at −80°C. Chromatin was fragmented using a Bioruptor (Diagenode) for 30 min on ice (30s pulses, 30s pauses) to produce fragments ~200-400bp in size. Goat

141

anti-rabbit IgG Dynabeads (Invitrogen) were coated with affinity purified anti-Ikaros antibody and incubated with chromatin overnight on a rotator at 4°C. Protein/DNA complexes were captured with a Magnetic Particle Concentrator (Invitrogen). Crosslinks were reversed and samples were treated with proteinase K and RNaseA. DNA was recovered using the QIAquick PCR Purification kit (QIAGEN). ChIP-Seq libraries were created using 18 cycles of amplification with ChIPseq DNA sample prep kit (Illumina).

Libraries were validated using the Agilent Technologies 2100 Bioanalyzer. ChIP-Seq libraries were sequenced at the High Throughput Genomics Center at University of

Washington, Seattle and at the Core Facility at Penn State College of Medicine, Hershey,

PA. Sequence fastq files were aligned to the UCSC human genome assembly HG19 using the Eland application (Illumina), allowing no more than two mismatches per sequence.

Only sequences aligning uniquely to the human genome were used to identify peaks.

Peaks calling was achieved using Cisgenome2.0 and SISSRS.

Luciferase Assay

Luciferase assays were performed using the LightSwitch Luciferase Assay System

(SwitchGear Genomics). HEK-293T cells were seeded into 24-well plates. After 24 hours, cells were transiently transfected with 0.15 μg of indicated promoter reporter constructs or pROM vector and 0.15 μg of pcDNA3.1-Ikaros or pcDNA3.1 vector in triplicate for each group using lipofectamine 2000 (Invitrogen) according to manufacturer’s instructions. Twenty-four hours after transfection, cells were lysed in 100

μl of LightSwitch Assay Solution (SwitchGear Genomics) and rocked at RT for 30min

142

according to the manufacturer’s instructions. Lysates were measured by luminometer

(Promega GloMax 20/20 Luminometer). Luciferase activities were calculated as fold change relative to vector only cells and normalized to pcDNA3.1 vector readings. All transfection and reporter assays were performed in triplicate. If HDAC inhibitors were used, 0.5nM TSA was added 4-6 hours post transfection and cells were incubated for 24 hours prior to reading.

Annotation of ChIP-Seq Peaks

Peak annotation with associated genes was retrieved using a Perl script. Peak information was input from the CisGenome output peak files in “COD” format. The nearest gene to each peak and associated locations (gene start, gene end, 5’UTR, 3’UTR, exon and intron) in the human genome was identified using the CisGenome built-in program

“refgene_getnearestgene” and “refgene_getlocationsummary” with parameters r = 0

(TSS-up, TES-down), up=100000 (bp) and down=100000 (bp), respectively. The gene names from HGNC and expression levels from RNA-Seq analysis were also input. The output data file includes the integrated information of peaks, genes and gene expression levels.

Distribution of peaks in different gene regulatory regions

The number of peaks in different gene regulatory regions (e.g. promoter, intergenic regions, etc.) was calculated using a Perl script. The peaks located at the TSS

143

(Transcriptional Start Site) and TES (Transcriptional End Site) were also counted by a

Perl script program. In this program, the maximum number of peaks at a position was also recorded. The percentage of the number of peaks at each position over the maximum number of peaks at a position was computed.

Results

Ikaros Binds to the Promoters of DLX1 and DLX2

To confirm that Ikaros binds to the promoters of DLX1 and DLX2, as indicated by our ChIP-Seq, we performed quantitative chromatin immunoprecipitation (qChIP) of

Ikaros in U937 AML cells. Binding ability at least two-fold greater than the IgG control reactions was considered to be significant binding. qRT-PCR confirmed that Ikaros binds to the promoters of DLX1 and DLX2, and that binding was at least two-fold greater than the IgG control reactions for each gene (Figure 4.1). These results show that Ikaros is able to bind to the promoters of these genes in U937 cells and indicates a potential role for Ikaros in regulation of DLX1 and DLX2 in AML.

Gain-of-Function: Ikaros Represses the Transcription of DLX1

To investigate the functional significance of the observed Ikaros binding to the promoter region of DLX1, we performed both gain-of-function and loss-of-function experiments. First, we performed luciferase reporter assays to determine whether

144

expression of DLX1 is regulated by direct binding of Ikaros. A plasmid containing the region of the DLX1 promoter in which an Ikaros binding peak had been observed by

ChIP-Seq was purchased from SwitchGear Genomics. HEK-293T cells transfected with the DLX1 promoter construct and an empty vector control showed strong luciferase activity, indicating that the promoter alone could stimulate expression of the luciferase gene. Cells transfected with the promoter construct in addition to a plasmid containing full length human Ikaros had a significant decrease in luciferase activity when compared to cells with the promoter construct and an empty vector (Figure 4.2). The decrease in luciferase activity was significant, indicating that Ikaros is able to directly bind to the promoter of DLX1 and repress its transcription.

Next, we wanted to determine whether increased Ikaros expression is able to affect the transcription of DLX1 and DLX2 in AML cells. To do this, we overexpressed

Ikaros in U937 cells. Cells were infected with retrovirus containing wild type Ikaros

(MIG-IK) or an empty retroviral control (MIG vector). Both vectors contained GFP.

Sorting based on GFP expression identified successfully infected cells, and the expression of the Ikaros target genes was analyzed using qRT-PCR. Our results indicated that cells with overexpressed Ikaros have decreased expression of DLX1 and DLX2

(Figure 4.3).

145

Loss-of-Function: Knockdown of Ikaros Results in Increased Transcription of DLX1 and DLX2

To determine whether the decrease in expression of DLX1 and DLX2 was occurring through Ikaros, we performed Ikaros loss-of-function experiments. Cells were transfected with Ikaros-specific shRNA or scrambled shRNA to determine whether inhibition of Ikaros would prevent the gene repression observed when Ikaros was overexpressed. Again, changes in gene expression were analyzed by qRT-PCR from isolated RNA. An increase in expression of the Ikaros target genes DLX1 and DLX2 was observed in U937 cells transfected with Ikaros shRNA in comparison to the scrambled control, indicating that loss of functional Ikaros results in increased expression of these gene targets (Figure 4.4).

Taken together, these results demonstrate by both gain-of-function and loss-of- function experiments that Ikaros has the ability to regulate the transcription of DLX1 and

DLX2.

Molecular Inhibition of CK2 Results in Transcriptional Repression of Ikaros Target Genes in a Manner Similar to Ikaros Overexpression

Previous studies have established that phosphorylation of Ikaros is a major mechanism for its regulation, responsible for both the function and localization of Ikaros.

Phosphorylation of Ikaros by CK2 has been shown to decrease the affinity of Ikaros for

DNA in addition to regulating its subcellular localization. Deregulation of CK2 has been

146

reported in leukemia, which suggests that the DNA-binding ability of Ikaros may also be impaired in leukemia.

In order to determine whether high expression levels of CK2 in leukemia cells is interfering with the ability of Ikaros to regulate its target genes, we asked whether inhibition of CK2 would affect Ikaros target genes in the same way as Ikaros overexpression. To evaluate this question, we used CK2-specific shRNA to inhibit the function of CK2 in U937 cells and analyzed RNA by qRT-PCR. A significant decrease was observed in expression of DLX1 and DLX2 in cells transfected with CK2-specific shRNA when compared to cells that received a scrambled control (Figure 4.5). The observed decrease in transcription of the Ikaros target genes after CK2 inhibition was similar to the decrease observed when Ikaros was overexpressed in U937 cells.

Therefore, these results suggest that the observed decrease in gene transcription is a direct result of the inhibition of CK2 and suggest that overexpression of CK2 in leukemia cells can interfere with the ability of Ikaros to regulate its gene targets.

Transcriptional Repression of Ikaros Target Genes Following CK2 Inhibition Requires Ikaros Activity

To determine whether Ikaros activity is essential for the observed changes in gene expression of DLX1 and DLX2 after CK2 inhibition, Ikaros was knocked down with an

Ikaros-specific shRNA in U937 cells and RNA levels were measured using qRT-PCR.

Cells that received a control shRNA and CX-4945 treatment showed a significant decrease in expression of DLX1 and DLX2 (Figure 4.6). Knockdown of Ikaros with an

147

Ikaros-specific shRNA was able to restore the gene expression in cells treated with CX-

4945, indicating that the ability of CK2 inhibition to repress the transcription of our target genes was lost when Ikaros was knocked down. These results indicate that activity of

Ikaros is essential for the repression of its target genes and is suggest that restoration of

Ikaros function is a potential mechanism to explain the decrease in expression of DLX1 and DLX2 following CK2 inhibition.

Ikaros Represses Transcription of DLX1 via an HDAC-dependent Mechanism

Because Ikaros has been known to associate with several complexes involved in chromatin remodeling, including HDAC1, and HDAC1 binding was increased at the promoters of several of our genes of interest, we investigated whether histone deacetylases are involved in the Ikaros-mediated repression of DLX1. We performed luciferase assays with and without the histone deacetylase inhibitor TSA in HEK-293T cells. Treatment with TSA eliminated the Ikaros-mediated repression of DLX1 (Figure

4.7). These results indicate that repression of DLX1 by Ikaros requires histone deacetylase activity. Further analysis of the relationship between Ikaros and histone deacetylases, especially HDAC1, in the regulation of DLX1 is necessary to elucidate the mechanism of Ikaros-mediated repression.

148

Discussion

We have identified that Ikaros regulates the homeodomain genes DLX1 and

DLX2. While their clinical relevance is less clear than our studies with BCL2A1,

MTHFR, and CDA, expression of these genes could have implications for patient prognosis. FLT3, an upstream kinase that is able to phosphorylate and regulate DLX1 and DLX2 has been identified as a negative prognostic marker in AML. Overexpression of FLT3 leads to increased expression of genes downstream in the pathway, including

DLX1 and DLX2. We have shown here that inhibition of CK2 with CX-4945 is able to increase the Ikaros-mediated repression of DLX1 and DLX2 and could provide a means of restoration of normal function downstream of FLT3.

149

Figures

qChIP of Ikaros in U937 Cells

9 ** 8

7 IgG 6 Ikaros 5

4

FoldEnrichment 3 * 2

1

0 DLX1 DLX2 SIRT6

Figure 4.1. Ikaros binds to the promoters of DLX1 and DLX2 in U937 cells. qChIP followed by qRT-PCR indicates that Ikaros binds to the promoters of DLX1 and DLX2 in U937 cells after treatment with 10 µM CX-4945 for 72 hours. IgG – non-specific IgG control antibody. Ikaros – affinity-purified anti-Ikaros antibody. SIRT6 represented as a negative control. (*p<0.05; **p<0.01).

150

Luciferase Assay in HEK-293T Cells 1.2 pcDNA 1 IK 0.8

0.6

0.4 **

0.2 Relative Relative Luciferase Activity 0 DLX1

Figure 4.2. Ikaros directly represses the transcription of DLX1. Representative experiment showing repression of the DLX1 luciferase promoter construct by the Ikaros- expressing vector pcDNA3.1-IK (black) in comparison to pcDNA3.1 empty vector control (grey) in HEK-293T cells. Luciferase activity was normalized to pcDNA3.1 and pROM empty vector controls. Data is presented as the average +/- SD. (*p<0.05).

151

Overexpression of Ikaros in Ikaros U937 Cells Overexpression 1.2 Mig Vector 30 Mig-Ikaros Mig

Vector PCR)

1 PCR) 25

- - Mig- Ikaros 0.8 20

0.6 15

0.4 10 Relative Relative Expression (qRT ** Relative Expression (qRT 0.2 ** 5

0 0 DLX1 DLX2 CERS1 Ikaros

Figure 4.3. Ikaros overexpression represses the transcription of DLX1 and DLX2. Relative transcription of target genes measured by qRT-PCR after U937 cells were infected with either empty vector (MIG) or human Ikaros (MIG-Ikaros) and sorted based on expression of GFP. Data are averages +/- SD. CERS1 is represented as a negative control. (**p<0.01).

152

Ikaros Knockdown in U937 Ikaros Knockdown Cells 1.2 3 Scram-shRNA Ikaros shRNA 1 Scram-

** shRNA PCR)

PCR) 2.5

- - Ikaros 0.8 shRNA 2 ** 0.6 1.5 ** 0.4

1

Relative Relative Expression (qRT Relative Expression (qRT 0.5 0.2

0 0 DLX1 DLX2 CERS1 Ikaros

Figure 4.4. Knockdown of Ikaros increases the transcription of DLX1 and DLX2. Relative transcription of target genes measured after U937 cells were transfected with either scrambled shRNA (negative control) or Ikaros specific shRNA (Ikaros shRNA). CERS1 is represented as a negative control. Knockdown of Ikaros was 2 fold. Gene transcription was analyzed by qRT-PCR. Data are averages +/- SD. (**p<0.01).

153

CK2 Knockdown in U937 Cells CK2 Knockdown

1.2 scram-shRNA 1.2 CK2 shRNA 1 1 Scram-

shRNA

PCR) PCR)

- - CK2- 0.8 0.8 shRNA

0.6 0.6 **

0.4 0.4

Relative Relative Expression (qRT Relative Expression (qRT 0.2 ** 0.2

0 0 DLX1 DLX2 CERS1 CK2

Figure 4.5. Knockdown of CK2 decreases the transcription of DLX1 and DLX2. U937 cells transfected with either scrambled shRNA (negative control) or CK2-specific shRNA. CERS1 is represented as a negative control. CK2 expression levels verify that partial knockdown of CK2 is achieved after transfection with CK2 shRNA. Effect of CK2 knockdown on target gene expression was analyzed by qRT-PCR, standardized to 18S RNA and presented as the average + SD. (**p<0.01).

154

Ikaros Blocks the Effect of CK2 Inhibition on Target Gene Expression 1.6 ** scram shRNA

1.4 scram shRNA + CX-4945 PCR)

- IK-shRNA + CX-4945 1.2

1

0.8 **

0.6 ** **

Relative Relative Expression (qRT 0.4

0.2

0 Ikaros DLX1 DLX2 CERS1

Figure 4.6. Transcriptional Repression of DLX1 and DLX2 Following CK2 Inhibition Requires Ikaros Activity. Effect of Ikaros knockdown on changes in gene expression following CK2 inhibition with CX-4945. qRT-PCR analysis of RNA from U937 cells transfected with either scrambled shRNA (negative control) or Ikaros-specific shRNA (IK-shRNA) and treated for 24hr with 20 µM CX-4945. Compared to scram- shRNA: **p<0.01; compared to scram-shRNA + CX-4945: **p<0.01.

155

Luciferase Assay in HEK-293T Cells 1.2

1

0.8 pcDNA (-TSA)

0.6 IK (-TSA) pcDNA (+TSA)

0.4 IK (+TSA)

** Relative Relative Luciferase Activity

0.2

0 DLX1

Figure 4.7. Inhibition of histone deacetylases prevents the Ikaros-mediated repression of DLX1. Representative experiments showing repression of the DLX1 luciferase promoter construct by the Ikaros-expressing vector pcDNA3.1-IK (second from left) in comparison to pcDNA3.1 empty vector control (left bar) in HEK-293T cells. Treatment with TSA abolishes this repression (right bar). Luciferase activity was normalized to pcDNA3.1 and pROM empty vector controls. Data is presented as the average +/- SD. (**p<0.01).

156

References

1. Apiou, F., Flagiello, D., Cillo, C., Malfoy, B., Poupon, M. F., and Dutrillaux, B. (1996) Fine mapping of human HOX gene clusters. Cytogenet Cell Genet 73, 114-115 2. Fujino, T., Suzuki, A., Ito, Y., Ohyashiki, K., Hatano, Y., Miura, I., and Nakamura, T. (2002) Single-translocation and double-chimeric transcripts: detection of NUP98- HOXA9 in myeloid leukemias with HOXA11 or HOXA13 breaks of the chromosomal translocation t(7;11)(p15;p15). Blood 99, 1428-1433 3. Lahortiga, I., Belloni, E., Vázquez, I., Agirre, X., Larrayoz, M. J., Vizmanos, J. L., Valgañón, M., Zudaire, I., Sáez, B., Mateos, M. C., Di Fiore, P. P., Calasanz, M. J., and Odero, M. D. (2005) NUP98 is fused to HOXA9 in a variant complex t(7;11;13;17) in a patient with AML-M2. Cancer Genet Cytogenet 157, 151-156 4. Nakamura, T. (2005) NUP98 fusion in human leukemia: dysregulation of the nuclear pore and homeodomain proteins. Int J Hematol 82, 21-27 5. Nishiyama, M., Arai, Y., Tsunematsu, Y., Kobayashi, H., Asami, K., Yabe, M., Kato, S., Oda, M., Eguchi, H., Ohki, M., and Kaneko, Y. (1999) 11p15 translocations involving the NUP98 gene in childhood therapy-related acute myeloid leukemia/myelodysplastic syndrome. Genes Chromosomes Cancer 26, 215-220 6. Hanson, R. D., Hess, J. L., Yu, B. D., Ernst, P., van Lohuizen, M., Berns, A., van der Lugt, N. M., Shashikant, C. S., Ruddle, F. H., Seto, M., and Korsmeyer, S. J. (1999) Mammalian Trithorax and polycomb-group homologues are antagonistic regulators of homeotic development. Proc Natl Acad Sci U S A 96, 14372-14377 7. Rozovskaia, T., Feinstein, E., Mor, O., Foa, R., Blechman, J., Nakamura, T., Croce, C. M., Cimino, G., and Canaani, E. (2001) Upregulation of Meis1 and HoxA9 in acute lymphocytic leukemias with the t(4 : 11) abnormality. Oncogene 20, 874-878 8. Crooks, G. M., Fuller, J., Petersen, D., Izadi, P., Malik, P., Pattengale, P. K., Kohn, D. B., and Gasson, J. C. (1999) Constitutive HOXA5 expression inhibits erythropoiesis and increases myelopoiesis from human hematopoietic progenitors. Blood 94, 519-528 9. Sauvageau, G., Thorsteinsdottir, U., Hough, M. R., Hugo, P., Lawrence, H. J., Largman, C., and Humphries, R. K. (1997) Overexpression of HOXB3 in hematopoietic cells causes defective lymphoid development and progressive myeloproliferation. Immunity 6, 13-22 157

10. Antonchuk, J., Sauvageau, G., and Humphries, R. K. (2001) HOXB4 overexpression mediates very rapid stem cell regeneration and competitive hematopoietic repopulation. Exp Hematol 29, 1125-1134 11. Kyba, M., Perlingeiro, R. C., and Daley, G. Q. (2002) HoxB4 confers definitive lymphoid-myeloid engraftment potential on embryonic stem cell and yolk sac hematopoietic progenitors. Cell 109, 29-37 12. Drabkin, H. A., Parsy, C., Ferguson, K., Guilhot, F., Lacotte, L., Roy, L., Zeng, C., Baron, A., Hunger, S. P., Varella-Garcia, M., Gemmill, R., Brizard, F., Brizard, A., and Roche, J. (2002) Quantitative HOX expression in chromosomally defined subsets of acute myelogenous leukemia. Leukemia 16, 186-195 13. Roche, J., Zeng, C., Barón, A., Gadgil, S., Gemmill, R. M., Tigaud, I., Thomas, X., and Drabkin, H. A. (2004) Hox expression in AML identifies a distinct subset of patients with intermediate cytogenetics. Leukemia 18, 1059-1063 14. Kottaridis, P. D., Gale, R. E., and Linch, D. C. (2003) Flt3 mutations and leukaemia. Br J Haematol 122, 523-538 15. Zhang, S., Fukuda, S., Lee, Y., Hangoc, G., Cooper, S., Spolski, R., Leonard, W. J., and Broxmeyer, H. E. (2000) Essential role of signal transducer and activator of transcription (Stat)5a but not Stat5b for Flt3-dependent signaling. J Exp Med 192, 719-728 16. Rosnet, O., Bühring, H. J., deLapeyrière, O., Beslu, N., Lavagna, C., Marchetto, S., Rappold, I., Drexler, H. G., Birg, F., Rottapel, R., Hannum, C., Dubreuil, P., and Birnbaum, D. (1996) Expression and signal transduction of the FLT3 tyrosine kinase receptor. Acta Haematol 95, 218-223 17. Stirewalt, D. L., and Radich, J. P. (2003) The role of FLT3 in haematopoietic malignancies. Nat Rev Cancer 3, 650-665 18. Gale, R. E., Hills, R., Kottaridis, P. D., Srirangan, S., Wheatley, K., Burnett, A. K., and Linch, D. C. (2005) No evidence that FLT3 status should be considered as an indicator for transplantation in acute myeloid leukemia (AML): an analysis of 1135 patients, excluding acute promyelocytic leukemia, from the UK MRC AML10 and 12 trials. Blood 106, 3658-3665 19. Gale, R. E., Hills, R., Pizzey, A. R., Kottaridis, P. D., Swirsky, D., Gilkes, A. F., Nugent, E., Mills, K. I., Wheatley, K., Solomon, E., Burnett, A. K., Linch, D. C., Grimwade, D., and Party, N. A. L. W. (2005) Relationship between FLT3 mutation status, biologic

158

characteristics, and response to targeted therapy in acute promyelocytic leukemia. Blood 106, 3768-3776 20. Levis, M., and Small, D. (2003) FLT3: ITDoes matter in leukemia. Leukemia 17, 1738- 1752 21. Starkova, J., Gadgil, S., Qiu, Y. H., Zhang, N., Hermanova, I., Kornblau, S. M., and Drabkin, H. A. (2011) Up-regulation of homeodomain genes, DLX1 and DLX2, by FLT3 signaling. Haematologica 96, 820-828 22. Hawley, R. G., Lieu, F. H., Fong, A. Z., and Hawley, T. S. (1994) Versatile retroviral vectors for potential use in gene therapy. Gene Ther 1, 136-138 23. Dovat, S., Ronni, T., Russell, D., Ferrini, R., Cobb, B. S., and Smale, S. T. (2002) A common mechanism for mitotic inactivation of C2H2 zinc finger DNA-binding domains. Genes Dev 16, 2985-2990 24. Fujiwara, T., O'Geen, H., Keles, S., Blahnik, K., Linnemann, A. K., Kang, Y. A., Choi, K., Farnham, P. J., and Bresnick, E. H. (2009) Discovering hematopoietic mechanisms through genome-wide analysis of GATA factor chromatin occupancy. Mol Cell 36, 667- 681 25. Wang, Z., Zang, C., Rosenfeld, J. A., Schones, D. E., Barski, A., Cuddapah, S., Cui, K., Roh, T. Y., Peng, W., Zhang, M. Q., and Zhao, K. (2008) Combinatorial patterns of histone acetylations and methylations in the human genome. Nat Genet 40, 897-903

159

Chapter 5

Treatment with the CK2 Inhibitor CX-4945 Alters the Epigenetic Landscape in U937 Cells

160

Introduction

Advances in high-throughput DNA sequencing technologies have greatly increased our ability to measure variation in gene regulation and gene expression and their relationship to human disease (1). These approaches can be employed to improve disease detection and diagnosis, understand disease progression, identify risk factors, and target therapy in a patient-specific manner with the use of biomarkers, approaches with significant implications for clinical care. RNA sequencing is able to provide detailed transcriptome analysis and allow for detection of novel transcripts and alternative splicing. By RNA-Seq, identification of pathways involved in development and differentiation, as well as discovery of elements involved in regulation of these pathways, is possible (2).

One such regulatory sequence of importance to these analyses is the enhancer, a

DNA sequence with complex roles in the control of gene expression (3). Enhancers play numerous roles in the regulation of genes associated with cell commitment, differentiation, and development and contribute to nuclear organization as well as transcription (4). Long-range association of enhancers with promoters has been identified, and the activity of enhancers has been shown to regulate promoter activity during development. These promoter-enhancer interactions can be highly specific and vary greatly across the genome in different cell types (5). In general, enhancers average a few hundred base pairs in length and can be occupied by multiple transcription factors.

Enhancers can be identified on a global scale by using specific histone modifications, particularly H3K4me1, as an indicator of putative enhancer regions (6). In this way, 161

between 400,000 and 1.4 million enhancer regions have been identified, although the number of enhancers that are active in a given cell type is generally in the tens of thousands (7). Differential activity of enhancers in different cells can provide a cell type- specific means of gene regulation that relies on the influence of cell-specific and master regulatory transcription factors to bring about cellular differentiation as well as other processes.

Analysis of enhancers bound by the master transcription factors Oct4, Sox2, and

Nanog along with the coactivator complex Mediator identified large clusters of constituent enhancers that are densely occupied by the master regulators, which were named super-enhancers. Unlike enhancers, which occupy a few hundred base pairs in length, super-enhancers can occupy as much as 50kb (8). In addition to size differences, super-enhancers also differ from typical enhancers in transcription factor density, enrichment of histone modifications, transcription activation activity, and sensitivity to slight variation in cellular conditions. Super-enhancers in differentiated cells tend to contain more cell type-specific master transcription factors, especially at regions that define cell identity. Typical enhancers have been shown to be enriched with the histone modifications H3K4me1 and H3K27ac, but super-enhancers show enrichment of these histone modifications that exceed typical enhancers by at least one order of magnitude

(8). These high levels can be attributed both to the length of the domain and also to the occupancy density of the constituent enhancers. Super-enhancers are further distinguished from typical enhancers by increased enrichment of Klf4 and Esrrb, transcription factors with important roles in embryonic stem cell gene expression programs and the reprogramming of somatic cells to induced pluripotent stem cells (8). 162

Enhancers have the ability to loop and associate with nearby genes, interactions that typically occur within 50 kb of the enhancer but can also occur at a distance of several megabases (9,10). By assigning super-enhancers to genes based on proximity, a list of super-enhancer-associated genes was identified. Analysis of gene ontology functional categories identified transcription factor genes as the highest class associated with super-enhancers (8). On the contrary, housekeeping genes were not found to be associated with super-enhancers. Genes which were associated with super-enhancers were expressed at higher levels than genes associated with typical enhancers. Taken together, these findings suggest that super-enhancers are able to drive high expression of their neighboring genes, particularly genes that encode transcription factors (8).

Enhancer function is often dependent upon the interaction of multiple transcription factors and coactivators. Because of the ability to facilitate these interactions, enhancers that contain many transcription factor binding sites can be more sensitive to changes in the availability of transcription factor than enhancers with fewer binding sites (11,12). Reducing levels of the transcription factor Oct4 or the levels of the coactivator Mediator by shRNA lead to earlier and more pronounced reduction of super- enhancer-associated genes than those associated with typical enhancers (8). These results indicate that although super-enhancers have the ability to drive high expression of their associated genes, super-enhancers are also more sensitive to slight perturbations in transcription factor availability.

In differentiated cells, analysis of the master transcription factor PU.1 revealed that genome-wide occupancy of Mediator and PU.1 is highly correlated. In murine pro-B cells, constituent enhancers within super-enhancer regions were enriched with PU.1 163

sequence motifs when compared to typical enhancer regions (8). Other important regulators of B cell identity, such as Ebf1, E2A, and Foxo1 were also enriched in the super-enhancer domains. Additional cell types in which master transcription factors are well defined were also studied, and large domains with clusters of enhancers bound by these master regulators of cell fate were identified in mouse myotubes, T helper cells, and macrophages (8). The identified super-enhancer regions were also associated with genes important to the development and differentiation of these cells. These results suggest that super-enhancers have the ability to regulate cell identity and developmental processes by associating with key transcription factors in a cell-specific manner. Super-enhancers tend to be associated with genes controlling unique cellular identities, while typical enhancers can promote expression of both cell type-specific genes and genes that are active across multiple cell types (8).

Key events responsible for the development of cell identity have been shown to be driven by the binding of specific transcription factors to constituent enhancers in super-enhancer domains. The large number of transcription factor binding sites present in these super-enhancer domains makes these regions more sensitive to changes in availability of even one transcription factor. Since we know that CK2 is able to regulate

Ikaros, a master regulator of hematopoiesis, via phosphorylation and thus decrease its ability to bind DNA, we hypothesized that inhibition of CK2 would result in changes to the super-enhancer landscape in U937 cells.

164

Experimental Procedures

Gene Expression Analysis

RNA-Seq data consisted of three biological replicates with two conditions, U937 cells with or without three days of 10 µM CX-4945 treatment. Single-read RNA-Seq fastq files were aligned using an alignment pipeline by Sean Mahony's group with an hg19 reference genome. The alignment pipeline for RNA-Seq utilizes tophat (version 2.0.11) with flags, -r 100 --no-coverage-search (Trapnell, 2014). Subsequently, the aligned bam files were processed to count the number of reads that mapped to the exons using featureCounts (version 1.4.6), a highly efficient program to count mapped reads for genomic features. The gene ID list for hg19 was obtained from the UCSC genome browser and referenced in the featureCounts tool. The count files were processed to list gene IDs and raw read counts. Three replicates were merged to a single column and used for the subsequent analyses. The top ranked up-regulated and down-regulated genes were determined by sorting the gene list with more than 100 number of reads by fold change. Gene ontology analysis was performed by using DAVID on genes with more than 2-fold differences with more than 10 number of reads (13,14).

ChIP-Seq Data Analysis

ChIP-Seq data consisted of two biological replicates each for two conditions, U937 cells with or without 10 µMCX-4945 treatment for 72 hours. ChIP-Seq fastq files were

165

aligned using an alignment pipeline by Sean Mahony's group using an hg19 reference genome. The alignment pipelines for ChIP-Seq used bowtie (version 1.0.1) with options

-q --best --strata -m 1 --chunkmbs 1024. The mapped reads were available in BAM and

BED format for analyses. The histone modifications data included H3K4me1, H3K27ac, and H3K27me3.

Super Enhancer Analysis

We located super enhancers by identifying regions with high levels of H3K27ac clusters

(8). Briefly, H3K27ac enriched regions outside of 2kb from TSS that occurred within

12.5 kb of each other were stitched together to form a single, larger domain (8,15). The stitched regions spanning more than two TSS regions were excluded from the analysis.

H3K27ac domains were ranked by H3K27ac signal and the tangent of the curve was used to define two enriched domains. This analysis identified 559 super-enhancers in untreated U937 cells and 711 super enhancers in CX-4945 treated U937 cells. We assigned genes that are within 10kb of the super enhancer using GREAT (16).

Enhancer Analysis

We performed edgeR analysis to identify regions with differential H3K27ac enrichments between the two conditions (17). Four-hundred base pair regions flanking H3K4me1

ChIP-Seq peaks for WT and CX-4945 were used to count the number of H3K27ac reads.

EdgeR analysis identified 909 regions with a reduction in H3K27ac signal after CX-4945

166

treatment and 1,212 regions with an increase in H3K27ac signal after CX-4945 treatment.

A p-value of 0.05 was used to determine differential enrichment. For each site that is differentially enriched between WT and CX-4945, as well as all sites that are tested, we assigned genes within 10kb of the enriched H3K27ac sites using GREAT (16). For each list, we performed Welch two sample t-tests comparing each list of genes.

Results

Transcriptome Analysis of U937 Cells by RNA-Seq

To determine changes in global gene expression that occur as a result of CX-4945 treatment, U937 cells were treated with 10 µM CX-4945 for 72 hours or left untreated.

RNA was isolated and RNA-Seq was performed to obtain the transcriptomes of the treated and untreated U937 cells. The software EdgeR was used to identify genes that were differentially expressed between the treated and untreated cells (Table 5.1). These genes were sorted by their absolute differences (fold change) in expression, and heat maps displaying patterns of differentially expressed genes were prepared (Figure 5.1).

Of the differentially expressed genes identified, gene ontology (GO) enrichment identified a significant upregulation of genes involved in regulation of apoptosis, transcription factor binding, and the inflammatory response in cells treated with CX-4945

(Figure 5.1, top right). GO enrichment identified ribosome biogenesis and translation to be downregulated after CX-4945 treatment, among other pathways (Figure 5.1, bottom right). Of note, the gene CDKN1A, which codes for the tumor suppressor protein p21, 167

was identified among significantly upregulated genes, and FLT3, a gene associated with poor prognosis in AML, was identified to be significantly downregulated. The center panel of Figure 5.1 shows a comparison of the enrichment in RNA reads between WT and CX-4945 treated U937 cells for two upregulated and two downregulated genes identified by transcriptome analysis.

Enhancers are frequently marked by H3K4me1, with further classification as active, poised, or inactive. Active enhancers are dynamic participants in gene expression, whereas poised enhancers participate in gene expression in response to various cellular stimuli such as differentiation cues. Active enhancers exhibit high levels of H3K27ac, whereas inactive enhancers are marked by low levels of H3K27ac (18). We looked for enhancers that changed states after CX-4945 treatment and identified 1,634 positions that transitioned from active to inactive (we called these silenced enhancers), 51,678 positions that remained unchanged (51,236 that had low levels of H3K27me3 and 443 that had high levels of H3K27me3), and 2,572 positions that transitioned from inactive to active

(induced enhancers) as a result of CX-4945 treatment (Figure 5.2). H3K27ac is lost in silenced enhancers and elevated in induced enhancers as a result of CX-4945 treatment.

Analysis of genes associated with active and inactive enhancers revealed a significant difference in gene expression between genes associated with active enhancers and genes associated with inactive enhancers (Figure 5.3). To define associated genes, we identified all genes whose transcription start sites fell within 50 kb of an active or an inactive enhancer. The average log fold change in gene expression for genes associated with inactive enhancers was -0.679, while the average log fold change in gene expression for genes associated with active enhancers was 0.949. These changes in gene expression 168

were significant in comparison to the average gene expression of all genes. In addition, enhancers with differential H3K27acetylation contain motifs of key hematopoiesis factors including Ikaros, PU.1, and Runx1 (Figure 5.3, right). Treatment of U937 cells with CX-4945 leads to differential enrichment of positive histone modifications at induced and silenced enhancers (Figure 5.4). H3K4me2 as well as acetylation marks on

H3 and H4 are increased in induced enhancers in comparison to silenced enhancers.

Super-enhancers, are important for regulating genes critical for cell type-specific identity. In addition to being marked by significant levels of H3K4me1 and H3K27ac, super-enhancers can extend across significant regions of chromatin, have domains of transcription factor binding sites, and are marked by significant amounts of H3K4me1 and H3K27ac modification. We identified super-enhancers in CX-4945 treated and untreated U937 cells by finding regions with the highest levels of clustered H3K27- acetylated chromatin. H3K27ac enriched regions outside of 2kb from a TSS that occurred within 12.5 kb of each other were stitched together to form a single larger domain. The stitched regions spanning more than two TSS were excluded from the analysis. H3K27ac domains were ranked by H3K27ac signal and a tangent of the curve was used to define two enriched domains. By using this method, we were able to identify 559 super- enhancers in wild-type U937 cells and 711 super-enhancers in CX-4945 treated cells

(Figure 5.5, left). Representative super-enhancer regions are shown (Figure 5.5, right).

From this analysis, we can distinguish the subsets of enhancers that are super-enhancers only in the wild-type or only in the CX-4945 treated cells. When examining super- enhancers found only in CX-4945 treated cells, we identified 62 super-enhancer- associated genes that are significantly upregulated. Sixteen genes were found to be 169

significantly downregulated at super-enhancers found only in wild-type cells (data not shown). The difference in average gene expression between all enhancers and super- enhancers was significant (Figure 5.5, bottom).

Additional Epigenetic Changes

We also performed ChIP-Seq of Histone H1 in untreated and CX-4945 treated

U937 cells. Histone H1 is a linker histone that is thought to be involved in higher level chromatin remodeling (19). Linker histones are a far more dynamic component of chromatin than core histones (20). Phosphorylation of Histone H1 during interphase decreases the affinity of H1 for chromatin to promote chromatin decondensation and transcription. During mitosis, dephosphorylation of Histone H1 promotes mitotic chromosome condensation (21).

Our data surprisingly showed that following CX-4945 treatment, Histone H1 tends to localize around the transcription start site (9) as opposed to being distributed evenly throughout the genome as it is in untreated U937 cells (data not shown). We also observed that Histone H1 tends to localize at peri-telomeric regions after CX-4945 treatment (Figure 5.6). Although the significance of this change in Histone H1 localization is unknown, it is clear that the DNA binding of Histone H1 is redistributed in

U937 cells after treatment with the CK2 inhibitor CX-4945.

170

Discussion

While further analysis is necessary to discern the physiological significance of these enhancer or super-enhancer associated genes and their changes in expression after

CK2 inhibition, it is clear that the super-enhancer landscape in U937 cells is drastically altered after CX-4945 treatment. Many of the genes and pathways associated with these super-enhancer domains are relevant in malignant transformation in general and in AML in particular and could shed light on how reorganization of the epigenetic landscape contributes to the progression of AML.

Understanding the epigenetic changes that occur after CX-4945 treatment in AML is a long-term and complex undertaking that requires further study. However, it is clear that treatment results in global epigenetic changes and that the function of Ikaros as a transcriptional factor may have some influence on this process. Rather than identifying a global regulatory mechanism that is relevant to all genes, it is more likely that the epigenetic regulation of Ikaros target genes is gene-specific or that other factors, such as tertiary structure, affect the distribution of histone modifications at the promoters of

Ikaros target genes.

171

Tables

Table 5.1. Top differentially expressed genes after CX-4945 treatment, sorted by absolute difference

Gene WT counts CX counts Absolute difference Log fold change FLNA 420844 153553 267291 -1.45454885 FTH1 301175 35947 265228 -3.06665874 SOD2 182262 8000 174262 -4.509869993 P4HB 56171 202073 145902 1.846979177 SCD 9052 151076 142024 4.060894097 EEF1A1 387822 527109 139287 0.44270668 IL8 136307 682 135625 -7.642872199 ENO1 57233 185859 128626 1.699289414 CTSD 187117 311217 124100 0.733980232 MMP1 118813 632 118181 -7.554552424 RPL4 69008 178076 109068 1.367657566 GAPDH 78993 179251 100258 1.182184448 GLO1 23009 112307 89298 2.287177664 RPS2 46116 133726 87610 1.535940705 PTMA 26464 108571 82107 2.036535757 ALDOA 38817 119056 80239 1.616879801 IL1B 80423 557 79866 -7.173787016 GPI 32615 108677 76062 1.736439112 RPLP0 55456 125457 70001 1.177777505 RPL3 42347 109085 66738 1.365121061

172

Figures

Figure 5.1. Transcriptome Analysis. Treatment with CX-4945 increased expression of tumor suppressor genes, activates apoptotic signaling cascades, and negatively regulates ribosomal biogenesis in AML.

173

Figure 5.2. Alterations in enhancer activity after CX-4945 treatment. WT (left) vs. CX-4945 treated (right) enhancer signatures. H3K4me1 marks enhancers, H3K27ac marks active enhancers, and H3K27me3 is used as a negative control. H3K27ac is lost in silenced enhancers and elevated in induced enhancers as a result of CX-4945 treatment.

174

Figure 5.3. Change in enhancer activities are correlated with gene expression changes. (Left) Expression of log2 fold change of gene expression proximal to silenced and induced enhancers. Welch two sample t-test indicates significant changes in gene expression. Gene expression is significantly decreased in silenced enhancers and significantly increased in induced enhancers when compared to unchanged enhancers. (Right) Enhancers with differential H3K27acetylation contain motifs of key hematopoiesis factors including Ikaros, PU.1, and Runx1.

175

Figure 5.4. Treatment of U937 cells with CX-4945 leads to differential enrichment of positive histone modifications at induced and silenced enhancers. H3K4me2 as well as acetylation marks on H3 and H4 are increased in induced enhancers in comparison to silenced enhancers.

176

Figure 5.5. CX-4945 treatment alters H3K27ac super-enriched regions (i.e. super- enhancers). Super-enhancers were identified by finding regions with the highest levels of clustered H3K27-acetylated chromatin. H3K27ac enriched regions outside of 2kb from a TSS that occurred within 12.5 kb of each other were stitched together to form a single larger domain. The stitched regions spanning more than two TSS were excluded from the analysis. H3K27ac domains were ranked by H3K27ac signal and a tangent of the curve was used to define two enriched domains. (Left) Distribution of normalized H3K27ac ChIP-Seq signal across all enhancers in WT and CX-4945 treated U937 cells. 559 super- enhancers were identified in wild-type U937 cells and 711 super-enhancers were

177

identified in CX-4945 treated cells. (Right) Representative super-enhancer regions are shown for each group. (Bottom) The difference in average gene expression between all enhancers and super-enhancers was significant.

178

Figure 5.6. Peri-telomeric enrichment of histone H1 in U937 cells after CX-4945 Treatment. Histone H1 localizes to peri-telomeric chromatin in U937 cells after treatment with 10 µM CX-4945 for 72 hours. Red bars show the distribution of Histone H1 after CX-4945 treatment. Blue bars represent Histone H1 in untreated U937 cells.

179

References

1. Sankaran, V. G., and Gallagher, P. G. (2013) Applications of high-throughput DNA sequencing to benign hematology. Blood 122, 3575-3582 2. McGettigan, P. A. (2013) Transcriptomics in the RNA-Seq era. Curr Opin Chem Biol 17, 4-11 3. Bulger, M., and Groudine, M. (2010) Enhancers: the abundance and function of regulatory sequences beyond promoters. Dev Biol 339, 250-257 4. Bulger, M., and Groudine, M. (2011) Functional and mechanistic diversity of distal transcription enhancers. Cell 144, 327-339 5. Zhang, Y., Wong, C. H., Birnbaum, R. Y., Li, G., Favaro, R., Ngan, C. Y., Lim, J., Tai, E., Poh, H. M., Wong, E., Mulawadi, F. H., Sung, W. K., Nicolis, S., Ahituv, N., Ruan, Y., and Wei, C. L. (2013) Chromatin connectivity maps reveal dynamic promoter- enhancer long-range associations. Nature 504, 306-310 6. Yip, K. Y., Cheng, C., Bhardwaj, N., Brown, J. B., Leng, J., Kundaje, A., Rozowsky, J., Birney, E., Bickel, P., Snyder, M., and Gerstein, M. (2012) Classification of human genomic regions based on experimentally determined binding sites of more than 100 transcription-related factors. Genome Biol 13, R48 7. Consortium, E. P. (2012) An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57-74 8. Whyte, W. A., Orlando, D. A., Hnisz, D., Abraham, B. J., Lin, C. Y., Kagey, M. H., Rahl, P. B., Lee, T. I., and Young, R. A. (2013) Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell 153, 307-319 9. Sanyal, A., Lajoie, B. R., Jain, G., and Dekker, J. (2012) The long-range interaction landscape of gene promoters. Nature 489, 109-113 10. Ong, C. T., and Corces, V. G. (2011) Enhancer function: new insights into the regulation of tissue-specific gene expression. Nat Rev Genet 12, 283-293 11. Griggs, D. W., and Johnston, M. (1991) Regulated expression of the GAL4 activator gene in yeast provides a sensitive genetic switch for glucose repression. Proc Natl Acad Sci U S A 88, 8597-8601 12. Giniger, E., and Ptashne, M. (1988) Cooperative DNA binding of the yeast transcriptional activator GAL4. Proc Natl Acad Sci U S A 85, 382-386 180

13. Huang, d. W., Sherman, B. T., and Lempicki, R. A. (2009) Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 37, 1-13 14. Huang, d. W., Sherman, B. T., and Lempicki, R. A. (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4, 44-57 15. Lovén, J., Hoke, H. A., Lin, C. Y., Lau, A., Orlando, D. A., Vakoc, C. R., Bradner, J. E., Lee, T. I., and Young, R. A. (2013) Selective inhibition of tumor oncogenes by disruption of super-enhancers. Cell 153, 320-334 16. McLean, C. Y., Bristor, D., Hiller, M., Clarke, S. L., Schaar, B. T., Lowe, C. B., Wenger, A. M., and Bejerano, G. (2010) GREAT improves functional interpretation of cis- regulatory regions. Nat Biotechnol 28, 495-501 17. Anders, S., and Huber, W. (2010) Differential expression analysis for sequence count data. Genome Biol 11, R106 18. Creyghton, M. P., Cheng, A. W., Welstead, G. G., Kooistra, T., Carey, B. W., Steine, E. J., Hanna, J., Lodato, M. A., Frampton, G. M., Sharp, P. A., Boyer, L. A., Young, R. A., and Jaenisch, R. (2010) Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc Natl Acad Sci U S A 107, 21931-21936 19. Annalisa, I., and Robert, S. (2015) The role of linker histone H1 modifications in the regulation of gene expression and chromatin dynamics. Biochim Biophys Acta 20. Misteli, T., Gunjan, A., Hock, R., Bustin, M., and Brown, D. T. (2000) Dynamic binding of histone H1 to chromatin in living cells. Nature 408, 877-881 21. Freedman, B. S., and Heald, R. (2010) Functional comparison of H1 histones in Xenopus reveals isoform-specific regulation by Cdk1 and RanGTP. Curr Biol 20, 1048-1052

181

Chapter 6

Overall Discussion

182

Summary

In this report we determine one mechanism of drug action for the specific CK2 inhibitor CX-4945 in AML, which occurs partly through the CK2/Ikaros axis. We show that Ikaros binds to the promoters of several genes important in drug resistance or malignancy, where it promotes the repression of these target genes and formation of a repressive chromatin state. Overexpression of Ikaros in the monocytic cell line U937 results in a significant decrease in the expression of MTHFR, CDA, BCL2A1, DLX1, and

DLX2. Furthermore, knockdown of Ikaros by shRNA results in increased expression of the Ikaros target genes, indicating by both gain-of-function and loss-of-function experiments that Ikaros can regulate the transcription of several genes with important implications in AML. We demonstrate that molecular and pharmacological inhibition of

CK2 represses MTHFR, CDA, and BCL2A1 gene expression in a manner similar to overexpression of Ikaros. Inhibition of CK2 enhances the binding of Ikaros to the promoters of these target genes in an AML cell line and in primary AML cells.

Furthermore, the transcriptional repression of Ikaros target genes following CK2 inhibition requires Ikaros activity, as evidenced by CX-4945 treatment in combination with Ikaros knockdown. We also show by cell proliferation assays that the CK2 inhibitor

CX-4945 produces synergistic effects when used in combination with doxorubicin, methotrexate, or cytarabine.

Finally, we assessed the effect of CK2 inhibition by CX-4945 on the epigenome and transcriptome of AML cells. The analysis revealed that CK2 inhibition results in alterations in the epigenomic signature of AML cells. The prominent changes involved 183

alteration of enhancer and super-enhancer landscapes, which were associated with transcriptional regulation of many genes that are critical for cellular proliferation.

In summary, our results demonstrate that the therapeutic effect of CK2 inhibition in AML cells involves restoration of Ikaros function as a tumor suppressor and transcriptional regulator. Our results have identified novel pathways that are regulated by

Ikaros as well as an epigenomic landscape that is regulated by CK2. These data led to the development of novel combination treatments for AML which showed synergy when tested on AML cells. Our results provide a mechanistic rationale for development of novel, targeted treatments for AML.

Proposed Model

We have shown here that Ikaros represses the transcription of BCL2A1, MTHFR,

CDA, DLX1, and DLX2, genes with translational significance due to their roles in the development of drug resistance and/or malignant transformation in leukemia. Our data show that the Ikaros/CK2 axis is responsible for the regulation of these genes and provides a model by which to understand the poor prognosis associated with overexpression of CK2 in AML. Decreased Ikaros binding as a result of Ikaros mutation, overexpression of CK2, or the simultaneous occurrence of both prevents the repression of these genes, an event that is required for normal hematopoietic function. Dysregulation of these genes results in increased expression. In the case of BCL2A1, MTHFR, and CDA, increased expression enables the cell to evade the toxic effects of standard

184

chemotherapeutic drugs and results in the proliferation of drug resistant cells. The ability of CK2 to regulate the tumor suppressor function of Ikaros has been well characterized, but the clinical significance of this regulation has remained unclear. Here we show that

Ikaros functions as a tumor suppressor in AML and that Ikaros regulates genes with potential clinical relevance.

New treatments for AML are necessary to improve patient response and overall survival. The heterogeneity of the disease affords the opportunity to use several biomarkers to identify appropriate treatments on a patient-by-patient basis. Our use of the specific CK2 inhibitor CX-4945 is an innovative approach that allows for treatment of patients with diverse Ikaros characterization. Here we present mechanistic data that will help to understand, at least in part, the diverse roles that Ikaros plays in regulation of drug resistance and tumorigenesis in AML. Our data suggest a model of leukemia where loss or mutation of Ikaros is compounded by increased CK2 activity, resulting in loss of functional Ikaros (Figure 6.1). Inactivation of Ikaros disrupts the repression of Ikaros target genes and promotes the development of drug resistance and leukemogenesis.

Impairment of Ikaros results in increased expression of BCL2A1, MTHFR, CDA, DLX1, and DLX2. Increased expression of BCL2A1 promotes anti-apoptotic pathways that can be utilized to promote drug resistance. Changes in MTHFR transcription disrupt the folate metabolism and response to methotrexate treatment. Overexpression of CDA results in faster clearing of cytarabine, the current standard of care treatment for AML, from the body and decreases its therapeutic effect. The DLX1 and DLX2 homeodomain genes are downstream targets of FLT3, a kinase whose overexpression is associated with poor prognosis in AML. Restoration of functional Ikaros could provide a mechanism for 185

restoring normal signaling downstream of FLT3 by acting as a repressor of DLX1 and

DLX2. Treatment with CX-4945, a specific CK2 inhibitor, is able to restore Ikaros- mediated repression of these target genes by increasing the binding of Ikaros to the promoters of its target genes.

Discussion

AML is a heterogeneous disease with several subtypes, and treatment and prognosis varies greatly among these subtypes. Despite these known classifications, many patients lack one of these prognostic indicators, and even patients that can be assigned to a group exhibit heterogeneity in outcome within the group. The identification of further somatic mutations or biological markers to aid in patient classification is necessary. The work in this dissertation suggests that analysis of Ikaros status and CK2 expression levels might be able to improve risk stratification at diagnosis.

Further analysis of the Ikaros-mediated repression of BCL2A1, MTHFR, CDA, and DLX1/2 is necessary in order to determine additional factors involved in the regulation of these genes, including chromatin remodeling complexes that might be involved. A better understanding of how Ikaros phosphorylation affects interactions between Ikaros and its known binding partners will also improve our understanding of this mechanism of regulation. Whether combination treatments with CX-4945 will be effective in xenograph mouse models remains to be seen.

186

Because CK2 has hundreds of cellular targets, inhibition of CK2 with inhibitors such as CX-4945 has the potential for significant toxicity. Although we have not observed toxic side effects in mice treated with CX-4945, the possibility exists that treatment with CX-4945 in humans may present unacceptable levels of toxicity. In phase

I clinical trials, the adverse side effects associated with CX-4945 treatment were mild to moderate, indicating that CX-4945 is safe for human use. In fact, research suggests that cells with higher expression of the CK2 catalytic subunit CK2α are more susceptible to

CX-4945 treatment. Because CK2α levels are elevated in several cancers, including

AML, a context dependent inhibition of CK2 results in cancer cells being more vulnerable to CX-4945 than normal cells. The correlation between effectiveness of CX-

4945 treatment and expression of CK2α suggests that the level of CK2α expression could be used as an indicator for the effectiveness of CK2 inhibitors toward individual patients in a personalized medicine approach.

Our results suggest that screening for Ikaros status at diagnosis may be useful to help stratify patient risk and the likelihood of patients to respond well to treatment with

CK2 inhibitors. Loss or mutation of Ikaros is correlated with poor prognosis in ALL.

Further study is necessary to determine whether a similar poor prognosis is associated with Ikaros mutation in AML. Our data provide evidence that Ikaros is involved in the regulation of several genes with potential roles in drug resistance. Additional analysis of patients with mutations in these genes could provide further mechanistic data to support the role of Ikaros in the regulation of drug resistance.

In AML, mutation or deletion of Ikaros is a rare event. Our data provide a framework for understanding how the tumor suppressor function of Ikaros can be 187

impaired even in AML that lacks an Ikaros mutation. Overexpression of CK2 results in hypophosphorylated Ikaros with impaired DNA-binding ability. Inhibition of CK2 can restore the ability of Ikaros to act as a tumor suppressor and negatively regulate its target genes. In this scenario, the functionality of Ikaros is limited even in the absence of a mutation or deletion of Ikaros, but CK2 inhibition is able to restore Ikaros function.

Treatment with CK2 inhibitors such as CX-4945 could also be used as potential treatments for AML with a mutation or deletion of Ikaros. In this case, the function of the remaining wild-type allele of Ikaros can be enhanced by the use of CK2 inhibitors. Based on the mechanism described here, we would not expect treatment with CX-4945 to be effective in patients with a homozygous Ikaros mutation or deletion. However, we cannot rule out the effect of CK2 inhibition on other pathways, and the possibility exists that

CX-4945 promotes an antitumor effect through additional pathways that do not involve

Ikaros.

In childhood AML, expression of Ikaros isoform 6 (Ik6) has been observed in cases of M4 and M5 AML but not in other AML subtypes. The shortened Ik6 isoform has a reduced number of Ikaros DNA-binding domains, affording the possibility of the suppression of functional of full-length Ikaros isoforms in a dominant-negative fashion.

These observations suggest that pediatric M4 and M5 AML might be more susceptible to

CX-4945 treatment because of compromised Ikaros function that results from the expression of dominant-negative Ikaros isoforms. In fact, overexpression of BCL2A1 has been observed in the AML subtypes M4 and M5, further suggesting that the M4 and M5 subtypes of AML might be more sensitive to treatment with CX-4945 than other AML subtypes. 188

Further work is necessary to determine whether CK2 inhibitors like CX-4945 can serve as effective treatments for AML in combination with standard of care chemotherapy drugs. Analysis of additional Ikaros target genes identified by ChIP-Seq in

AML could provide additional pathways that can be targeted with additional chemotherapy drugs in combination with CX-4945. The data presented here suggest that

CX-4945 produces a therapeutic effect in AML, and that the Ikaros/CK2 axis is one pathway responsible for the observed cytotoxicity of CX-4945 in AML. While we cannot discount the contribution of other pathways affected by CK2 inhibition, the involvement of Ikaros and CK2 suggest that screening patients for mutations in Ikaros or overexpression of CK2α at diagnosis could improve the current AML classification system and provide personalized combination treatments for AML.

189

Figures

Figure 6.1. Proposed overall model. CK2 inhibits Ikaros. Treatment with CX-4945 inhibits CK2 and thus restores the DNA-binding and functional aspects of Ikaros and promotes redistribution of HDAC1 and repressive chromatin, resulting in repression of overall gene expression. When Ikaros DNA-binding ability is restored, Ikaros is able to repress its target genes and thereby prevent tumorigenesis.

190

VITA Sadie Lynne Steffens

EDUCATION 2015 Doctor of Philosophy in Biomedical Sciences Pennsylvania State University College of Medicine, Hershey, PA 2011 Bachelor of Science in Molecular Biology Grove City College, Grove City, PA

TEACHING EXPERIENCE, PROFESSIONAL DEVELOPMENT, & AWARDS 2015 Adjunct Professor of Microbiology at HACC, Lancaster, PA 2015 Genetics Teaching Assistant at Johns Hopkins Center for Talented Youth, Baltimore, MD 2015 1st Place Blog Award, Lions Talk Science, Penn State University College of Medicine, Hershey, PA 2014 New Instructor Orientation Course, Penn State University College of Medicine, Hershey, PA 2014 St. Baldrick’s Summer Fellow, Penn State University College of Medicine, Hershey, PA

PUBLICATIONS S. Steffens, C. Song, X. Pan, and S. Dovat. Therapeutic efficacy of the specific CK2 inhibitor CX-4945 in AML involves transcriptional repression of BCL2A1 by Ikaros (2015). JBC (To be submitted)

Z. Ge, J. Liu, R. Zhang, X. Guo, J. Xu, M. Li, C. Qiao, S. Steffens, and S. Dovat. Overexpression of CRLF2 co-exists with IKZF1 deletion in adult acute lymphoblastic leukemia (2015). Acta Haematologica (Submitted).

ABSTRACTS AND PRESENTATIONS S. Steffens, C. Song, X. Pan, and S. Dovat. “CK2 Inhibitors as a Treatment for Acute Myeloid Leukemia.” Poster presentation for the Graduate Research Forum, 2015, Hershey, PA.

C. Song, X. Pan, Z. Li, Y. Ding, S. Muthusami, C. Gowda, Z. Ge, S. Steffens, B. Tan, and S. Dovat. “Epigenetic control of signature gene expression in the inducible differentiation of acute promyelocytic leukemia cells.” Poster presentation for the AACR Annual Meeting, 2015, Philadelphia, PA.

X. Pan, C. Song, Y. Ding, S. Steffens, and S. Dovat. “Identification of Binding Site Enrichment of Ikaros in Leukemia.” Poster presentation for Pediatric Research Day, 2014, Hershey, PA.

S. Steffens, C. Song, X. Pan, and S. Dovat. “The Role of Ikaros in Myeloid Differentiation and Malignancy.” Poster presentation for the Graduate Research Forum, 2014, Hershey, PA.

X. Pan, C. Song, S. Steffens, and S. Dovat. “Identification of Binding Site Enrichment of Ikaros in Leukemia with a new data analysis pipeline.” Poster presentation for the 18th Annual International Conference on Research in Computational Molecular Biology, 2014, Pittsburgh, PA.*