ATM) in DNA Repair and Tumor Suppression

Novel Roles of Ataxia Telangiectasia Mutated (ATM) in DNA Repair and Tumor Suppression

Kenta Yamamoto

Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy under the Executive Committee of the Graduate School of Arts and Sciences

COLUMBIA UNIVERSITY

2016

ABSTRACT

Novel Roles of Ataxia Telangiectasia Mutated (ATM) in DNA repair and Tumor Suppression

Kenta Yamamoto

Mammalian cells possess a variety of different DNA repair pathways, which work together to safeguard genomic integrity upon encountering different types of DNA damage.

Among all lesions, DNA double-strand breaks (DSBs) are most toxic and, if left unrepaired, results in loss of genetic information and genomic instability- a hallmark of tumorigenesis. Ataxia

Telangiectasia Mutated (ATM) is a protein kinase, a master regulator of the DNA damage response, and is activated upon the formation of DSBs. ATM senses DNA DSBs through its accessory proteins and functions as a transducer of the DNA damage response (DDR), which entails the activation of genes involved in DNA repair, cell cycle checkpoint, and apoptosis.

Consequently, loss of ATM results in increased genomic instability and compromised checkpoint regulation. Moreover, loss of ATM has been reported in various human cancers, and

Atm-deficient mice uniformly develop thymic lymphomas, highlighting its role as a tumor suppressor.

Although ATM has been extensively studied, much of its known functions to date pertained to its kinase activity, and the structural function of ATM remains elusive. To investigate whether ATM possesses structural functions beyond its kinase activity, we generated a mouse model expressing kinase-dead (KD) ATM protein. Intriguingly, while Atm-/- are viable, AtmKD/KD and AtmKD/- mice were embryonic lethal and AtmKD/KD and AtmKD/- cells displayed greater genomic instability compared to ATM-null cells, suggesting that the presence of the ATM KD protein blocks additional DNA repair pathways that are not affected in ATM-null cells. In this context, we identified defects in homologous recombination, resolution of

Camptothecin (CPT)-induced Topoisomerase-I lesions, and replication progression specifically

in AtmKD/- cells beyond those observed in Atm-/-. Mouse model expressing KD ATM (AtmKD/-) in hematopoietic stem cells (HSCs) developed thymic lymphomas faster and more frequently than the corresponding model with the ATM-null HSCs, which was associated with increased genomic instability and loss of tumor-suppressor Pten. In collaboration with others, we showed that the majority of tumor-associated ATM mutations reported in TCGA are missense mutations and are highly enriched in the kinase domain, while Ataxia-Telangiectasia (A-T) associated germline ATM mutations are almost always truncating mutations leading to complete loss of

ATM protein. This result suggests that ATM KD protein might be expressed in a significant fraction of human cancer. These results, for the first time, identified a previously unknown phosphorylation-dependent, structural function of ATM in the maintenance of genomic integrity and tumor suppression. Furthermore, the tumorigenicity and vulnerability to particular DNA damaging agents caused by the expression of the ATM KD protein relative to the loss of ATM highlight the importance of distinguishing the types of ATM mutations in tumors, and provide novel insights into the clinical use of specific ATM kinase inhibitors, as well as the prognosis and treatments of ATM-mutated cancers.

ATM has been reported to be frequently inactivated in human B-cell lymphomas, including up to 50% Mantle Cell Lymphoma (MCL), which represents around 6% of all Non-

Hodgkins Lymphomas (NHLs). MCL is characterized by the recurrent t(11;14)(q13;q32) translocation, which juxtaposes CCND1/BCL-1 to the IGH enhancer, leading to deregulated expression of CyclinD1 (CCND1). However, CyclinD1 overexpression in B cells alone is not sufficient to induce MCL in mouse models, and the role of ATM in the suppression of B-cell lymphomas is not well understood, in part due to the lack of ATM-deficient mature B-cell lymphoma models. To address this, we generated a mouse model that combines conditional deletion of ATM specifically in early progenitor B-cells via Mb1cre, and overexpressing CyclinD1 in lymphoid cells via EµCyclinD1 transgene. While ATM loss alone resulted in the development

of indolent, clonal, mature B-cell lymphoma, combined ATM-loss and CyclinD1 overexpression accelerated and increased the incidence of B-cell lymphoma. Furthermore, ATM-loss combined with CyclinD1 overexpression led to greater genomic instability and the expansion of naïve

ATM-deficient B-cells in the spleen. This study, for the first time, developed an ATM-deficient B- cell lymphoma model and demonstrated a synergistic function of ATM and CyclinD1 in pre-GC

B-cell proliferation and lymphomagenesis. Furthermore, the mice described here provide a prototypic animal model to study the pathogenesis of human MCL, for which there are no suitable mouse models.

TABLE OF CONTENTS

LIST OF FIGURES ...... iv LIST OF TABLES ...... vi LIST OF ABBREVIATIONS & ACRONYMS ...... vii ACKNOWLEDGEMENTS ...... ix Chapter 1 - General Introduction ...... 1 ATM: master regulator of the DNA damage response ...... 2 Mammalian DNA double-strand break repair pathways ...... 5 DNA repair and lymphocyte development ...... 10 ATM as a tumor suppressor in lymphocytes ...... 13 Chapter 2 - Understanding the kinase-structure functions of ATM using a mouse model† ...... 17 Background and Significance ...... 18 Results ...... 19 AtmKD/KD, but not Atm+/KD mice, die during early embryonic development...... 19 AtmKD/- embryonic stem (ES) cells have greater genomic instability than Atm-/- cells. ...21 The effect of ATM-KD protein on lymphocyte development...... 23 Transgenic expression of an autophosphorylation-mimetic, kinase-dead double mutant ATM rescues the embryonic lethality of ATM KD mice ...... 25 Discussion ...... 30 Chapter 3 – Phosphorylation-dependent kinase-structure functions of ATM in DNA repair‡ ...... 33 Background and Significance ...... 34 Results ...... 35 ATM KD cells are hypersensitive to Camptothecin (CPT), and cannot upregulate CPT- induced Sister Chromatid Exchange (SCE) and RAD51 foci ...... 35 ATM KD cells are defective in CPT-induced DSB formation, Top-Icc Resolution ...... 39 ATM KD blocks structure-specific nuclease activity in a MRE11-dependent manner....41 Discussion ...... 43 Chapter 4 - Structure-kinase function of ATM in tumor suppression, and the mutational spectrum of ATM in human tumors‡ ...... 45 Background and Significance ...... 46 Results ...... 47

Expression of kinase-dead ATM is more oncogenic than complete loss of ATM ...... 47 Missense mutations in the kinase domain of ATM are enriched in human tumors ...... 54 Human cancer cell lines that harbor ATM missense mutations display defects in CPT induced DSBs formation and are also hypersensitive to CPT...... 55 Discussion ...... 57 Chapter 5 - Roles of ATM in the suppression of B-cell lymphomas♯ ...... 62 Background and Significance ...... 63 Results ...... 64 Mouse models with B-cell specific deletion of ATM ...... 64 B-cell specific deletion of ATM leads to B-cell autonomous lymphomas of both GC and non-GC phenotype ...... 67 CyclinD1 overexpression accelerates B-cell lymphoproliferations in MA mice and skews lymphomas towards pre-GC origin ...... 69 MAD tumors share a subset of molecular features with human MCL ...... 72 ATM deficiency enhances genomic instability and increases chromosomal fusions in CyclinD1+ B cells...... 76 Ectopic CyclinD1 expression increases naïve B-cell proliferation and rescues the B-cell lymphocytopenia in older MA mice...... 78 Discussion ...... 81 IMPLICATIONS, FUTURE DIRECTIONS ...... 84 Novel kinase-structure roles of ATM in DNA repair, lymphomagenesis ...... 84 Mutational landscape of ATM in human tumors- prognostic and therapeutic significance ....86 Functions of ATM in the suppression of B-cell lymphomas ...... 87 MATERIALS AND METHODS ...... 89 CHAPTER 2 ...... 89 Generation of ATM KD, DKDtg allele ...... 89 Derivation of Embryonic Stem Cells from Mice...... 90 Cytogenetic analysis for genomic instability ...... 90 Lymphocyte Development and Class Switch Recombination ...... 91 CHAPTER 3 ...... 92 Mouse Embryonic Fibroblasts (MEF) ...... 92 Immunofluorescence and quantification of repair protein foci ...... 92 Western and Southern Blotting ...... 93 Sister Chromatid Exchange Assay ...... 93

COMET Assay ...... 94 In-vitro Complex of Enzymes (ICE) assay ...... 94 DNA fiber assay ...... 94 CHAPTER 4 ...... 95 Large Cohort Patient Collection ...... 95 Mutation Density Estimation ...... 95 Structure Simulation Analyses ...... 96 Mice and lymphocyte analyses ...... 97 CHAPTER 5 ...... 97 Mice 97 Flow cytometry and Immunohistochemistry analyses ...... 98 Southern Blot and Western Blot Analyses ...... 98 Class Switch Recombination and FISH analyses ...... 99 Mutation analyses ...... 99 Comparative Genomic Hybridization (CGH) ...... 100 REFERENCE ...... 101

iii

LIST OF FIGURES

CHAPTER 1

Figure 1-1: The ATM Protein ...... 3 Figure 1-2: ATM-mediated DNA damage response ...... 6 Figure 1-3: Pathways of DNA double-strand break repair ...... 9

CHAPTER 2

Figure 2-1: Generation of AtmKD allele ...... 20 Figure 2-2: Proliferation defects and increased genomic instability in AtmKD/- embryonic stem cells ...... 22 Figure 2-3: ATM KD lymphocytes are catalytically inactive ...... 25 Figure 2-4: Analysis of ATM KD lymphocytes ...... 27 Figure 2-5: Characterization of Atm+/KD mice ...... 27 Figure 2-6: Generation of the autophosphorylation-mimetic, kinase-dead (DKD trangene) Atm Mouse...... 28 Figure 2-7: DKD Atm-/- mice do not express ATM protein, and develop thymic lymphomas...... 30

CHAPTER 3

Figure 3-1: Analyses of immortalized AtmKD/- and control MEFs ...... 36 Figure 3-2: AtmKD/- MEFs are hypersensitive to CPT and cannot upregulate CPT induced SCE and RAD51 foci ...... 38 Figure 3-3: AtmKD/- MEFs cannot generate CPT induced DSBs and accumulate TopI-cc ...... 40 Figure 3-4: Mre11-dependent recruitment of Atm KD protein blocks resolution of Top1cc by SLX4/Mus81/SLX1 nuclease ...... 42 Figure 3-5: Working model for ATM KD blockage of DSB formation upon Top-I conjugation.....43

CHAPTER 4

Figure 4-1: Lymphocyte development and characterization of the VKD and VN mice ...... 49 Figure 4-2: Expression of Atm KD protein alone in hematopoietic stem cells (HSCs) is more oncogenic than complete loss of Atm in mouse models ...... 51 Figure 4-3: ATM KD tumors undergo more frequent loss of Pten compared to ATM null ...... 52 Figure 4-4: VKD T-cells are hypersensitive to CPT and PARP inhibitor combination treatment .53 Figure 4-5: Cancer associated ATM mutations are enriched for kinase domain missense mutations ...... 55 Figure 4-6: Missense ATM mutations identified in human tumors affect the kinase domain ...... 57 Figure 4-7: ATM mutated human tumor lines are also hypersensitive to CPT...... 58

CHAPTER 5

Figure 5-1: Lack of B-cell lymphomas in CD21Cre+ ATMC/C and CD19Cre+ ATMC/C mouse models...... 65 Figure 5-2: B-cell specific deletion of ATM in Mb1+/CreATMC/- and Mb1+/CreATMC/-EµCylinD1+ mouse models...... 66 Figure 5-3: Clonal B-cell lymphoproliferations in 24 month plus MA and MD/D mice and representative flow cytometry analyses of MAD mice...... 68 Figure 5-4: MA and MAD mice develop clonal B-cell lymphoproliferations ...... 70 Figure 5-5: Histopathologic and immunophenotypic analyses of B-cell lymphoproliferations of MAD mice ...... 73 Figure 5-6: Comparative Genomic Hybridization (CGH) and FISH analyses of MA and MAD tumors ...... 76 Figure 5-7: Cell Cycle Analyses of stimulated B-cells from MA, MAD and MD/D mice...... 77 Figure 5-8: Increased chromosomal fusions in stimulated MAD B-cells ...... 79 Figure 5-9: CyclinD1 expression rescues progressive B-cell loss in MA mice ...... 80

IMPLICATIONS, FUTURE DIRECTIONS

Figure 0-1: ATM R3008H (R) retains protein expression and lacks kinase activity ...... 87

LIST OF TABLES

CHAPTER 1

Table 1.1: List of lymphomas with frequent ATM mutations...... 14

CHAPTER 2

Table 2.1: Kinase inactive ATM leads to early embryonic lethality ...... 21 Table 2.2: Tabulation of live-born mice between DKD Atm+/- (F1) and Atm+/- breedings ...... 29

CHAPTER 5

Table 5.1: Pathological characteristics of clonal B-cell lymphoproliferations of MD/D, MA and MAD mice ...... 71

LIST OF ABBREVIATIONS & ACRONYMS

4OHT 4-Hydroxytamoxifen dsDNA double-stranded DNA A-EJ Alternative End Joining Ethylenediamine EDTA A-T Ataxia-Telangiectasia Tetraacetate Bacterial Artificial ES Embryonic Stem [Cells] BAC Chromosomes Fluorescence Activated FACS B-Acute Lymphoblastic Cell Sorting B-ALL Leukemia FBS Fetal Bovine Serum B-Chronic Lymphocytic Fluorescent in-situ B-CLL FISH Leukemia hybridization BCR B-cell receptor GC Germinal Center BER Base-excision repair GL Germline BM Bone Marrow Human Chorionic hCG bp Base pair Gonadotropin BrdU Bromo-deoxyuridine HJ Holliday Junction Comparative Genomic Homologous CGH Hybridization HR/HDR Recombination/Homology- Directed Repair CldU Chloro-deoxyuridine HSC Hematopoietic Stem Cell CO Cross-over HU Hydroxyurea CPT Camptothecin In vitro Complex of ICE Enzymes [Assay] Class Switch CSR IdU Iodo-deoxyuridine Recombination IR [γ] Irradiation 4',6-diamidino-2- DAPI phenylindole K/Kid Kidney DDR DNA Damage Response Kb kilobase ATM KD Kinase-Dead DKD S1987D/D2880A/N2885K KO Knock-Out Mutant Transgene LE Long Exposure

LN Lymph Node Diffuse Large B-Cell DLBCL Leiden Open Variation Lymphoma LOVD Database (Leiden Dulbecco's Modified University, Netherlands) DMEM Eagle's Medium Lymphoproliferative LPD DMSO Dimethyl Sulfoxide Disorder DP Double Positive LPS Lipopolysaccharide MA Mb1+/cre AtmC/C(-)

[DNA] Double-strand Magnetic Activated Cell DSB MACS break Sorting

vii

Su(var)3-9 and Enhancer SET Mb1+/cre AtmC/C(-) of Zeste [domain] MAD EµCyclinD1+ SHM Somatic Hypermutation MCL Mantle Cell Lymphoma SP Single Positive Mb1+/cre(+) Atm+/+(-) Spl Spleen MD/D EµCyclinD1+ [DNA] Single-stranded SSB Mouse Embryonic break MEF Fibroblast(s) ssDNA single-stranded DNA Microhomology-Mediated T-Acute Lymphoblastic MMEJ T-ALL End Joining Leukemia MMR Mismatch Repair The Cancer Genome TCGA MRE11-RAD50-NBS1 Atlas (NIH) MRN complex TCR T-cell receptor NCO Non-crossover Tg Transgene ND Not Determined Topoisomerase-1 covalent Top1cc Neo Neomycin complex Non-Homologous End T-Prolymphocytic NHEJ T-PLL Joining Leukemia (B-) Non-Hodgkin's VKD Vav-cre+ AtmC/KD NHL Lymphoma VN Vav-cre+ AtmC/C(-) PB Peripheral Blood WT Wildtype Peripheral Blood PBMC Mononuclear Cells PDB Protein Databank Proline, Glutamic Acid, PEST Serine, Threonine Rich [domain] Pregnant Mare Serum PMS Gonadotropin PNA Peptide-Nucleic Acid ROS Reactive Oxygen Species

Roswell Park Medical RPMI Institute [Medium] Sister Chromatid SCE Exchange Severe Combined SCID Immunodeficiency SDS Sodium Dodecyl Sulfate SE Short Exposure

viii

ACKNOWLEDGEMENTS

This dissertation is an embodiment of a large collaborative effort made possible by numerous contributions, both direct and indirect. I am forever indebted to everyone that supported me throughout this process, and the limited space provided here does not do this justice.

Dr. Shan Zha- I simply could not have imagined having a better mentor to lead me through my grad school journey. Listening to your faculty talk in my first year, I immediately felt the passion and enthusiasm you have for your research and the vision you had as a brand new

PI establishing your up-incoming lab. After my short rotation in the lab (still full of unopened boxes and a non-functional tissue culture room) I knew that I wanted to help fulfill your vision.

From personally showing me how to properly close the PCR machine on my first day, to deriving MEFs together at 9am on New Year’s Day, and weekly Friday brainstorming sessions,

I’ve had a very unique and personal training experience that not many other graduate students can attest to. Furthermore, I am never ceased to be amazed by your tenacity in pursuing your scientific curiosities, and the staggering vastness of your knowledge in a wide variety of disciplines. I can’t thank you enough for taking me under your wings as your first graduate student. Every bit of the scientist in me today, from the way I approach experiments to my knowledge in the field, is a product of your guidance.

Thank you to the numerous collaborators that have helped bring my project to fruition-

Drs. Alessandro Vindigni, Chris Haddock, David O. Ferguson, Victor Lin, Dong Wang, Jun Xu,

Govind Bhagat, Chris J. Bakkenist, Alex Bishop, Agata Smogorzewska, Michael Reth, Elias

Hobieka, among many others that have helped in performing experiments and/or providing essential reagents.

Thesis committee members- Drs. Jean Gautier, Lorraine Symington, and Ulf Klein, for helping me guide my project along. I would also like to thank Dr. Alberto Ciccia for joining the thesis examination committee on short notice. As my projects touched on several different fields that Shan and I were not entirely familiar with, the variety of expertise and suggestions you each contributed were invaluable to the progression of my work. I must also apologize for the poor selection of refreshments I brought to the committee meetings over the years. I will always remember the stench emanating from the spoiling cauliflowers I brought to the qualifying exam.

From the Pathology program- Dr. Ron Liem and Zaia Sivo for your support throughout the graduate school process. Starting with the accommodations during the interview process, to helping file paperwork for student loan deferments, you helped create an environment where I can solely stay focused on doing science. Thanks to your tireless effort, you made me feel at home while at Columbia. Special thanks to Drs. Ulrich Hengst and Julie Canman for overseeing the weekly pathology department seminars, and for your insightful critiques on my presentations each year. I now feel more confident presenting my own work at seminars and conferences thanks to your training.

The Med-into-Grad program: Drs. Ron Liem, Howard Worman, and Steve & Patrice

Spitalnik for providing me with a unique opportunity to get a glimpse of what happens in the clinic. Through this program, I gained a new perspective on what I do bench-side as a basic scientist (and realized that I would have been a terrible physician had I chosen that path). Very special thanks to Dr. Julia Glade-Bender for guiding me through the pediatric oncology clinic during this program, and giving me the opportunity to meet and talk to patients valiantly fighting cancer. You truly are a hero.

Members of the Zha lab, old and new - I have been blessed to be able to work with an amazing group of colleagues during my time in the lab. Technicians Richard Dubois, Chen Li,

Brian Lee, and Denis Loredan; postdocs Wendy/Wenxia Jiang, Louis/Xiangyu Liu, Zhengping

Shao; fellow grad students Jennie Crowe and Lisa Sprinzen; medical residents and fellows

Serine Avagyan, Kerice Pinkney; undergrad Zack Wolner; summer students Zhifan Yang,

Prince Alam, and Jessie Xie. Thank you all for creating a fun, collaborative environment conducive to productive research. It has been a joy working in the Zha lab.

Other members in the Irving Cancer Research Center/Institute for Cancer Genetics- the rapport and the collaborative spirit in the institute have been tremendous. Many thanks go out to everyone who has helped me throughout the years.

Lastly, this section would not be complete without acknowledging the support from my family and friends. My partner Van Lee- I would not have been able to make it through graduate school in one piece without your patient, and caring support.

Chapter 1 - General Introduction

ATM: master regulator of the DNA damage response

Ataxia Telangiectasia Mutated (ATM) was initially discovered as the gene mutated in

Ataxia-Telangiectasia (A-T) syndrome- a rare, congenital neurodegenerative disorder characterized by motor neuron ataxia caused by cerebellar degeneration, and ocular telangiectasia – dilation of blood vessels in the eyes. A-T patients also display hypersensitivity to radiation, and heightened predisposition to leukemia and lymphoma (McKinnon 2012). ATM gene is located on human chromosome 11q22.3 (9qA5.3 in mice) and encodes a 3056 amino acid (3066 in mice), 370kDa, Serine/Threonine (Ser/Thr) protein kinase, which belongs to the

PI3-Kinase-like-protein-kinase (PIKK) family that also includes DNA-dependent Protein Kinase catalytic subunit (DNA-PKcs), Ataxia Telangiectasia and Rad3-related (ATR), mammalian

Target of Rapamycin (mTOR), and Suppressor of Mutagenesis in Genitalia 1 (SMG1), and

Transformation/Transcription domain-associated Protein (TRRAP). TRRAP is a transcriptional co-activator and is the only PIKK that does not have kinase activities. PIKKs all possess long, α- helical N-terminal HEAT repeats (Huntingtin, Elongation factor 3, protein phosphatase 2A, and

Yeast target of rapamycin), followed by a FAT (conserved in FRAP, ATM, TRAAP), a PI3-

Kinase like catalytic domain and a FAT-C (FAT C-terminal) domain in the C-terminus. Although the crystal structure of ATM has yet to be solved, previous published structures of DNA-PKcs and mTOR, and data from cryo-electron microscopy (CryoEM) based structural rendering of

ATM suggest that the protein is composed of an “head” and an “arm” domain, the latter of which undergoes major conformational changes upon interaction with DNA (Llorca, Rivera-Calzada et al. 2003, Sibanda, Chirgadze et al. 2010, Yang, Rudge et al. 2013)

The best characterized function of ATM is its role as the master regulator of the DNA damage response (DDR) upon DNA double-strand breaks (DSBs). At physiological state, ATM normally exists as inactive homodimers predominantly in the nucleoplasm (Bakkenist and

Kastan 2003). DNA DSBs trigger ATM activation, accompanied by trans-auto-phosphorylation

Figure 1-1: The ATM Protein

Schematic domain structure of ATM protein. Relative amino acid (A.A.) positions for the domains are shown below. Selected residues, including those that undergo post-translational modifications, are also indicated above. N- Amino-terminus, C- Carboxyl Terminus.

on Ser1981 (S1987 in mice) among others (e.g. Ser367, Ser1893 in humans) and its relocalization to the site of DSBs (Bakkenist and Kastan 2003, Kozlov, Graham et al. 2006).

Although ATM auto-phosphorylation is widely accepted as a marker for ATM activation, the exact role of ATM auto-phosphorylation is still controversial. While mutant ATM with an alanine substitution at Ser1981 could not be fully activated in human cells and fail to form stable foci at

DSB sites, transgenic mice expressing only the mutated ATM with alanine substitute at

Ser1987, and two other auto-phosphorylation sites, display no measurable defects in ATM activation (Bakkenist and Kastan 2003, Pellegrini, Celeste et al. 2006, Daniel, Pellegrini et al.

2008, So, Davis et al. 2009). ATM is also known to undergo other post-translational modifications (PTMs), including acetylation at Lys3016 by acetyltransferase TIP60 that is thought to be essential for its activation (Sun, Xu et al. 2007). Upon DSB formation, ATM is recruited to the DSBs by the MRE11-NBS1-RAD50 (MRN) complex, through direct interaction with the C-terminal tail of NBS1 (You, Chahwan et al. 2005). The MRN complex acts as a

“sensor” of DNA DSBs, and is one of the first protein complexes to localize at the break site.

Binding of MRN to the DNA allows the protein complex to undergo conformational changes that promotes the tethering of the coiled-coil and zinc-hook structure of RAD50 from opposing MRN

complexes to bring together the two DNA ends (Hopfner, Craig et al. 2002, Moreno-Herrero, de

Jager et al. 2005, Wiltzius, Hohl et al. 2005). Recruitment of ATM to DSB sites via MRN is critical for its function, as loss of function mutations in MRE11 or NBS1 cause A-T-like disorder

(ATLD) and Nijmegen breakage syndrome (NBS1), respectively, that share clinical features with

A-T syndrome. In addition to the MRN dependent activation of ATM induced by DSBs, ATM can also be activated independent of MRN by reactive oxygen species (ROS), which leads to the formation of disulfide linkage at Cys2991 between two ATM proteins to form dimers. It is thought that ROS-activated ATM could translocate to the cytoplasm and upregulate levels of NADPH in the mitochondria to relieve cellular oxidative stress (Guo, Deshpande et al. 2010, Guo, Kozlov et al. 2010).

ATM phosphorylates over 700 potential substrates (Shiloh and Ziv 2013) preferentially on Serine and Threonine residues followed by a Glutamine (S/TQ motifs). ATM substrates function in cell cycle arrest, DNA repair, chromatin condensation/relaxation and, when necessary, cell death. Most notably, ATM phosphorylates Histone H2A variant H2AX at Ser139.

H2AX comprises ~2-20% of all H2A proteins in mammalian cells (Fernandez-Capetillo, Lee et al. 2004). Phosphorylated-H2AX (termed γH2AX) serves as a docking site for numerous effector proteins (e.g. MDC1, 53BP1) involved in DDR, further promoting the recruitment and retention of factors required for DSB repair at the break site. In addition, ATM phosphorylates and activates other kinases, most notably checkpoint kinases 1 and 2 (CHK1/2), which can then phosphorylate p53 at Ser20 to promote its dissociation from MDM2- E3 ubiquitin ligase, and stabilize p53 protein to elicit checkpoint activation and apoptosis. ATM can also directly phosphorylate Ser15 and Ser46 on p53, as well as Ser395 on MDM2 to relieve MDM2-mediated p53 degradation, leading to its stabilization (Saito, Goodarzi et al. 2002, Berger, Stahl et al.

2005, Jenkins, Durell et al. 2012). ATM almost exclusively phosphorylates KRAB-associated protein 1 (KAP-1)- which promotes chromatin relaxation and better accessibility to the DSBs

(Ziv, Bielopolski et al. 2006). In the absence of ATM, γH2AX and CHK2 are phosphorylated through compensatory activities via ATR or DNA-PKcs (Brown and Baltimore 2003, Stiff,

O'Driscoll et al. 2004). Accordingly, although mice deficient for either ATM or DNA-PKcs are viable, mice lacking both are embryonic lethal and display severe impairments in DDR, DNA repair and lymphocyte development (Gurley and Kemp 2001, Callen, Jankovic et al. 2009,

Gapud, Dorsett et al. 2011, Zha, Jiang et al. 2011).

While the signaling function of ATM has been extensively studied, whether the large

ATM protein possesses any structural functions beyond its kinase activity remains elusive.

ATM-specific kinase inhibitors, including KU55933, CP466722, and KU60019 have been developed and widely used in many studies, (Hickson, Zhao et al. 2004, Rainey, Charlton et al.

2008, Golding, Rosenberg et al. 2009). Treatment of wildtype cells with these kinase inhibitors largely phenocopies the complete loss of ATM, including diminished phosphorylation of ATM- specific substrates, and radio- and chemo-sensitization of cells to DNA damaging agents

(Hickson, Zhao et al. 2004, Rainey, Charlton et al. 2008, Golding, Rosenberg et al. 2009).

However, several studies suggest that ATM kinase inhibition results in additional defects in DNA repair that were not observed in cells lacking ATM protein, indicating a possible structural function of ATM in DNA repair (White, Choi et al. 2010, Shiloh and Ziv 2013). This question will be investigated and discussed further in chapters 2-4.

Mammalian DNA double-strand break repair pathways

DNA DSBs are highly genotoxic lesions which, if left unrepaired, can result in genomic instability, cell death, and tumorigenesis. Mammalian cells consistently suffer from DSBs as result of exposure to genotoxic agents, irradiation, metabolic and oxidative stress, and as programmed events during the development of lymphocytes and germ cells. These lesions have

Figure 1-2: ATM-mediated DNA damage response Upon DSB formation, ATM localizes to the site of DNA DSBs and activate a number of downstream targets involved in DNA repair, cell cycle checkpoint regulation, chromatin relaxation, and apoptosis. Adapted from (Shiloh and Ziv 2013)

to be precisely and efficiently repaired to maintain genomic integrity. In mammalian cells, there are two major DNA DSB repair pathways: the Non-Homologous End-Joining (NHEJ) and the

Homologous Recombination (HR) pathways [Fig. 1-3]. ATM has a role in both pathways and

also contributes to the pathway choice by promoting end-resection, the first step of HR (Bakr,

Oing et al. 2015). Classical NHEJ (C-NHEJ) is a non-templated, error prone pathway that entails direct joining of DSBs. C-NHEJ functions throughout all cell cycle phases and is the predominant repair pathway in the G1 phase of the cell cycle when the homologous template is not usually available. Ku70/80 (Ku86 in human) protein complex binds to the ends of the DSB, initiate C-NHEJ, and recruit DNA-PKcs and Artemis endonuclease, which together processes the DNA ends (e.g. hairpin opening), among other functions (Lieber, Ma et al. 2003). The formation of XLF/CERNUNNOS-XRCC4-LIG4 filaments on the DNA is thought to bridge the two participating DNA ends, where LIG4 catalyzes the ligation reaction to complete the joining process (Andres, Vergnes et al. 2012). In contrast to C-NHEJ, HR is a template-mediated, error- free pathway of DSB repair. HR activity is largely restricted to S-phase and G2 phases of the cell cycle, when the sister-chromatid is available as a template (San Filippo, Sung et al. 2008).

HR is initiated by 5’ to 3’ resection of the break site by the endo and exo-nucleases, including

MRE11 with the help of CTIP (referred to as RBBP8 in the NCBI), and EXO1 (Symington and

Gautier 2011). Resection of the ends results in the exposure of 3’ single-stranded DNA (ssDNA) overhangs that are coated by ssDNA binding protein RPA. RPA is hyper-phosphorylated by

PIKKs, most notably at sites Ser4, Ser8, and Thr21 (Binz, Sheehan et al. 2004), which primes their removal from the ssDNA ends by RAD52 and BRCA1/2, and replacement by RAD51 filaments (McIlwraith, Van Dyck et al. 2000). The coating of the ssDNA by Rad51 filaments primes the free DNA end for strand invasion- allowing the free DNA end to undergo homology search. In mitotic cells, the homologous template used for repair is primarily on the sister chromatid, while duing meiosis, homologous chromosomes are often the preferred template

(Keeney 2001, San Filippo, Sung et al. 2008). The invasion and annealing of the donor DNA to its homologous sequence generates a Displacement loop (D-loop). The recapturing and reannealing of the invading strand to the original, non-template strand generate double Holliday

Junctions (HJs), which are processed by BLM/Top3 complex or by structure-specific nucleases,

Figure 1-3: Pathways of DNA double-strand break repair NHEJ and HR are the two major pathways of DNA repair in mammalian cells, with MMEJ/A-EJ constituting a third, minor pathway. Adapted from (Barlow and Rothstein 2010)

such as MUS81 SLX1and GEN1 (Ip, Rass et al. 2008, Castor, Nair et al. 2013, Sarbajna and

West 2014). Depending on the relative orientation of the cleavage at the two cross sites by the structure-specific endonuclease, HJ resolution can result in either a non-crossover event, where the chromosomal regions flanking the break sites are not exchanged, or a crossover where the resulting ends exchange sister chromatids. BLM helicase brings the two HJs in close proximity to allow Top3 to resolve it, which always results in non-crossover products (Wu and Hickson

2003, Liberi, Maffioletti et al. 2005). In cases when the second strand is not recaptured and the double HJs do not form, the invading strand can be displaced from the D-loop and re-anneal to the original, non-invading strand in a process known as synthesis-displacement strand annealing (SDSA) which does not result in a crossover. Alternatively, cells can also undergo

Break-Induced Replication (BIR), where the non-invading second end is lost, and the invading strand can continue to synthesize the new strand until it reaches the end of the chromosome or is permanently displaced from the D-loop (Llorente, Smith et al. 2008). The recently emerged

Microhomology-Mediated End Joining (MMEJ), also referred to as Alternative End-joining (A-

EJ), constitutes a third, less well-characterized DSB repair pathway. Although detailed mechanisms are still not well established, MMEJ relies on the presence of sequence microhomology flanking a DSB. These MH anneals to each when exposed upon resection, and effectively convert a DSB to two single strand breaks. As such, ligation during MMEJ is thought to be catalyzed by XRCC1/LIG3 and will result in the loss of the intervening sequences between the two MH sequences (McVey and Lee 2008).

DNA repair and lymphocyte development

Developing lymphocytes generate programmed DSBs during the assembly of antigen receptor gene via V(D)J recombination that joins the germline Variable (V), Diversity (D), and

Joining (J) gene segments to form the functional B or T cell receptor (BCR [Immunoglobulin, Ig] or TCR). This process is initiated by lymphocyte-specific recombination activating genes

(RAG)1 and 2, which recognize and bind recombine signal sequences (RSSs) - a set of conserved heptameric and nonameric sequences flanking either a 12bp or 23bp spacer sequence located immediately adjacent to each germline V, D or J segments. Efficient recombination only occurs between RSS with a 12bp spacer sequence to another with the 23bp spacer- termed the 12-23 rule (van Gent, Ramsden et al. 1996). RAG binding at the RSS results in the cleavage between RSS and the V/D/J segment that generates four ends, two “signal ends” containing the cleaved RSSs, and two “coding” ends containing the DNA encoding V, D or J segments. The two signal ends are blunted and 5’ phosphorylated, and can be readily ligated together to form signal joint, while the two coding ends are hairpinned and cannot be directly ligated to form the coding joint. Coding ends are then processed by Artemis endonuclease, which opens the hairpin to allow ligation. If the V(D)J recombination occurs in the deletional orientation (the case of most Ig and TCR genes), the two signal ends join to form an excision circle containing the intervening sequence. Due to opening of the hairpin outside of the apex by Artemis, and the terminal deoxynucleotidyl transferase (TdT) activity at the coding ends, the coding joint is imprecise. This imprecision contributes to antigen receptor gene diversification and the generation of high-affinity antigen receptors. The ligation between the two signal and coding ends (to form the signal and coding joints, respectively) is exclusively mediated by the ubiquitously expressed C-NHEJ machinery. While not essential for the process,

ATM plays critical roles in promoting efficient and precise DSB repair during V(D)J recombination. Studies in ATM-deficient v-Abl kinase transformed Pro-B cells revealed transient

accumulation of unrepaired coding ends and mis-joining of one signal end with another coding end (to form hybrid joints) (Bredemeyer, Sharma et al. 2006). In addition, ATM plays a role in promoting end resection and also promotes efficient end-ligation- a function partially complemented by XLF/CERNUNNOS in the absence of ATM (Zha, Guo et al. 2011). However,

ATM is dispensable for signal joint formation, as well as extrachromosomal V(D)J recombination in vitro (Hsieh, Arlett et al. 1993, Gapud, Dorsett et al. 2011). This selective function of ATM in coding, but not signaling join formation, maybe in part due to a role of ATM in promoting Artemis recruitment (Jiang, Crowe et al. 2015). Expression of antigen receptor gene products are essential for proper lymphocyte development, as both human patients and mice deficient in essential NHEJ genes exhibit severe-combined immunodeficiency (SCID) due to the lack of mature lymphocytes. Accordingly, the developmental blockage of lymphocytes in NHEJ- deficient mice can be rescued by the introduction of transgenes expressing pre-rearranged antigen receptor gene products (Yan, Boboila et al. 2007). In this context, ATM deficiency impairs T cell development without significantly affecting naïve B cell development, despite the accumulation of IgH breaks in ATM-deficient immature B cells (Callen, Jankovic et al. 2007)

Human and mouse encode four T-cell receptor (TCR) genes- TCRα, β, γ,δ,. TCR rearrangement begins in the thymus during the CD4-CD8- stage (double negative, DN) and lead to the generation of two functional T-cell subtypes: αβ T cells or γδ T-cells, based on the surface expression of TCRα/β or TCRγ/δ. In αβ T-cells, which comprises the majority of the T-cell population, successful V(D)J rearrangement of TCRβ is necessary for the transition from a DN3 stage to the DN4 stage, then to CD4+CD8+ (double positive, DP)(Germain 2002). Expansion and proliferation of the cells in the late DN3 and DN4 stage create a large pool of T cells before the rearrangement of the TCRα locus. At the DP stage, cells undergo rearrangement at the

TCRα locus, and successful expression of the TCRα allows the positive and negative selection in the thymus, and the transition from DP thymocytes to either CD4+CD8- or CD4-CD8+ single

positive (SP) cells. SP cells, or naïve T-cells, then exit the thymus and enter the periphery. Both

ATM-deficient mice and A-T patients display slightly reduced peripheral lymphocyte number, in part due to a partial blockade of T-cell development at the DP to SP transition due to aberrant

TCRα rearrangement (Borghesani, Alt et al. 2000, Schubert, Reichenbach et al. 2002).

Ig gene products are assembled in developing B-cells in the Bone Marrow. In

B220+CD43+ Pro-B cells, immunoglobulin heavy chain locus (IgH) undergoes ordered V(D)J recombination with the joining of the DH and JH segments first, followed by joining of selected

VH segments to DJH. The assembly of the IgH in Pro-B cells results in the expression of a pre-

BCR- composed of a rearranged IgH paired with a surrogate light chain, transition to

B220+CD43- Pre-B cells, and initiation of Ig light chains (IgL)- λ or κ (Igλ, Igκ) rearrangement.

Successful rearrangement of IgL triggers the transition from Pre-B cells to IgM+B220+ naïve B- cells, which can now migrate into the periphery. In secondary lymphoid organs, such as lymph nodes and the spleen, naïve B-cells undergo additional genomic rearrangement events in specialized structures known as Germinal Centers (GC). In GCs, B-cells undergo class switch recombination (CSR), and somatic hypermutation (SHM) in the light and dark zones, respectively. Activation-Induced Cytidine Deaminase (AID) initiate CSR by converting Cytidines

(Muramatsu, Kinoshita et al. 2000). The C to U transitions create base pair mismatches, which are recognized by the Mismatch Repair (MMR) and Base-Excision Repair (BER) pathways, including Uracil DNA Glycosylase (UNG) and Apurinic/apyrimidinic Endonuclease

(APE)(Stavnezer, Guikema et al. 2008). This process requires the concomitant transcription of the S-regions. MMR/BER recognition and excision of the converted bases in transcribed S- regions generate nicks which, if closely spaced, eventually lead to DSBs. Joining of the DSBs generated in the S-region preceding the default constant region Cµ (Sµ) with breaks in downstream S-regions (Sγ3, Sγ1, Sγ2b, Sγ2a,Sε,Sα) result in class-switching, and generation of Igs

with same antigen specificity, but different effector functions. Although CSR-associated DSBs are mainly repaired via classical NHEJ, MMEJ could support as much as 25-50% of the CSR activity in the absence of classical NHEJ (Soulas-Sprauel, Le Guyader et al. 2007, Yan, Boboila et al. 2007, Han and Yu 2008, Boboila, Yan et al. 2010). In vitro analyses of ATM-deficient B- cells showed reduced CSR efficiency upon cytokine stimulation, and accumulate AID- dependent chromosomal translocations and breaks involving the IgH locus (Lumsden, McCarty et al. 2004, Reina-San-Martin, Chen et al. 2004, Franco, Alt et al. 2006). Accordingly A-T patients display reduced serum concentration of IgA, IgG2, and IgE, (Sanal, Ozaltin et al. 1998,

Nowak-Wegrzyn, Crawford et al. 2004). SHM on the transcribed hypervariable region exons is also initiated by AID in the rapidly proliferating B-cells in the dark zone of the GC. The nicks that are generated during SHM mostly results in single nucleotide exchanges, although a small subset (~6%) of SHM-related lesions include DSBs which result in deletions and duplications

(Bross, Fukita et al. 2000, Papavasiliou and Schatz 2000, Klein and Dalla-Favera 2008).

Mutations created in the hypervariable region of the IgH and IgL further contribute to antibody diversification and the generation of higher affinity antibodies.

ATM as a tumor suppressor in lymphocytes

As lymphocytes undergo programmed DSBs that require efficient and timely repair, errors during V(D)J and CSR result in chromosomal translocations that are frequently observed in lymphoid malignancies. Recurrent translocations involving the antigen receptor loci and proto- oncogenes are a hallmark of various types of human lymphomas. Additionally, in a p53-deficient background, mice deficient for all known NHEJ factors (except XLF), develop recurrent IgH- c-

Myc translocations and amplifications, and succumbed to pro-B cell lymphomas (Frank,

Sharpless et al. 2000, Gao, Ferguson et al. 2000). In this context, V(D)J recombination

Common Type Cytogenetic ATM References Abnormalities Mutations

(Bea, Valdes-Mas et al. Mantle Cell t(11;14)(q13;q32) 50% Lymphoma (MCL) 2013)

(Gronbaek, Worm et al. Diffuse Large B- (3;14)/(q27;q32) 20% cell lymphoma 2002) (DLBCL)

(Wang, Lawrence et al. B-Chronic del(13)(q14.3) 20% Lymphocytic 2011) Leukemia (B-CLL)

t(14q11;14q32), (Stilgenbauer, Schaffner et T-Prolymphocytic 50% Leukemia (T-PLL) inv(14)(q11;32) al. 1997)

t(1;7)(p32;q34), T-Acute t(10;14)(q24;q11.2), 1% (Takeuchi, Koike et al. 1998) Lymphoblastic Leukemia (T-ALL) t(11;14)(p13;q11)

Table 1.1: List of lymphomas with frequent ATM mutations Five human lymphomas, and their common cytogenetic karyotype, with most frequently observed biallelic inactivation of ATM (ATM mutations).

associated breaks in the IgH locus on chromosome 12 translocate to a region downstream of the c-Myc locus located on chromosome 15, generating a dicentric chromosome. The newly generated dicentric chromosome undergoes break-fusion-bridge (BFB) cycles that ultimately results in the amplification of c-Myc (Zhu, Mills et al. 2002, Gostissa, Yan et al. 2009).These observations highlight the importance of proper maintenance of genomic integrity by necessary repair pathways in lymphocytes after DSB generation during rearrangement.

ATM plays a critical role in promoting efficient V(D)J recombination and CSR to prevent chromosomal translocations and genomic instability in developing lymphocytes. Consequently,

~25% of A-T patients develop B and T-lymphomas harboring recurrent translocations, and ATM is somatically mutated in various human lymphomas (Xu 1999, McKinnon 2012). Carriers for germline ATM mutations often develop mature T-cell Prolymphocytic Leukemia (T-PLL), characterized by t(14q11;14q32) or inv(14)(q11;32) translocations involving the TCR locus with the TCL1 oncogene (Pekarsky, Hallas et al. 2001). Somatic mutations of ATM have been identified in up to 50% of sporadic T-PLL and Mantle Cell Lymphoma (MCL), and 20% of CLL

[Table 1.1] (Stilgenbauer, Schaffner et al. 1997, Gronbaek, Worm et al. 2002, Bea, Valdes-Mas et al. 2013). A large number of the cases contain gross chromosomal deletions around 11q22.3 where ATM, as well as MLL2- another tumor suppressor gene in B-cells, are located, combined with the mutation on the other allele resulting in loss of heterozygosity (Gumy-Pause, Wacker et al. 2004, Cremona and Behrens 2014). In T-PLL, the majority of the mutations clustered in the region containing the PI3-Kinase domain of the protein (Stankovic, Taylor et al. 2001). In these lymphoid malignancies, deletion of ATM is believed to be an early event during the progression of lymphomagenesis due to the high prevalence of observed mutations in early stage of the disease (Jares, Colomer et al. 2007). The loss of ATM predisposes cells to aberrant rearrangements during development, which eventually leads to oncogenic translocations.

Atm-/- mice recapitulate many phenotypes of A-T patients with homozygous germline loss of ATM. They display defects in T-cell development, hypersensitivity to irradiation, and develop early onset, aggressive thymic lymphomas in ~100 days (Barlow, Hirotsune et al. 1996, Xu,

Ashley et al. 1996). ATM-deficient murine thymic lymphomas are typically CD4+CD8+ or CD4-

CD8+, CD3lo, sTCRβ-, which share surface markers with human T-ALL. They also harbor recurrent t(12:14) chromosomal translocations that results in the amplification of a ~500kb segment ~10Mb centromeric to the TCRα/δ locus (Zha, Bassing et al. 2010). Additionally, ~24%

of the thymic lymphomas carry deletions in regions surrounding Pten. Notch1 mutations and amplification occur in ~40% of the cases, recapitulating several frequent genetic lesions in human T-ALL (Zha, Bassing et al. 2010). Atm+/- heterozygous mice rarely develop lymphoid malignancies, consistent with the loss of heterozygosity observed in human patients and the need to loss both copy of ATM to abolish its tumor suppression function (Spring, Ahangari et al.

2002).

Over 400 unique ATM mutations have been identified in A-T patients or cancers. While an overwhelming majority (~89%) of the A-T mutations are truncation mutations that abrogate protein expression, a significant fraction of cancer associated ATM mutations are missense mutations that have been confirmed, or predicted to retain significant ATM expression.

However, little is known about the functional significance of these missense mutations. In addition, ATM-specific kinase inhibitors have been proposed to be used for both a chemo- and radio-sensitizing agent for the treatment various cancers. However, whether the presence of a kinase-inactivated ATM upon treatment of the cells with kinase inhibitor poses a risk remains unknown. These points will be further investigated and discussed as a part of chapters 2-4.

Chapter 2 - Understanding the kinase-structure functions of ATM using a mouse model†

†Work described in this chapter is published under:

Kinase-Dead ATM protein causes genomic instability and early embryonic lethality in mice

Kenta Yamamoto1,2,4, Yunyue Wang1,2, Wenxia Jiang1,2, Xiangyu Liu1,2, Richard L. Dubois1,2, Chyuan- Sheng Lin2, Thomas Ludwig1,2,5, Christopher J. Bakkenist6, Shan Zha1,2,3

1 Institute for Cancer Genetics, Columbia University, New York, NY 10032 2 Department of Pathology and Cell Biology, Herbert Irving Comprehensive Cancer Center, Columbia University, New York, NY 10032 3 Division of Pediatric Oncology, Department of Pediatrics, Columbia University Medical Center, New York, NY 10032 4 Graduate program of Pathobiology and Molecular Medicine, College of Physicians and Surgeons, Columbia University, New York, NY 10032 5 Current address: Department of Molecular & Cellular Biochemistry, Ohio State University Wexner Medical Center, Columbus, OH 43210 6 Departments of Radiation Oncology and Pharmacology and Chemical Biology, University of Pittsburgh School of Medicine, Hillman Cancer Center, Research Pavilion, Suite 2.6, 5117 Centre Avenue, Pittsburgh, PA 15213-1863, USA.

Journal of Cell Biology (2012) Aug 6;198(3):305-13. PMID: 22869596

Background and Significance

Recent study on ATM function has greatly benefited from the development of specific inhibitor for ATM kinase activity (Hickson, Zhao et al. 2004). While in most studies ATM kinase inhibitor reproduce the effect of ATM protein deficiency, several recent studies report additional defects in replication related DNA repair in ATM kinase inhibitor treated cells, but not in ATM null cells, suggesting potential kinase independent function of ATM protein (White, Choi et al.

2010, Gamper, Choi et al. 2012). While the overwhelming majority of germline A-T patients carry truncation mutations that lead to the complete loss of protein, many somatic mutations in human lymphomas have been identified to be missense mutations. Some mutations identified have been confirmed to have the expression of the ATM protein with abrogated activity.

However, the full functional consequence expressing a defective protein is not well understood.

In this chapter, we report the generation and characterization of mouse model with

D2880A/N2885K (corresponding to D2870A/N2875K in human) that leads to the expression of a kinase dead ATM (ATM KD) (Canman, Lim et al. 1998). Intriguingly, mice carrying AtmKD in homozygosity died during early embryonic development, despite normal embryonic development of Atm null mice (Atm-/-). We further show that Atm-/KD embryonic stem (ES) cells have proliferation defects and spontaneous genomic instability relative to Atm-/- ES cells. While the spontaneous genomic instability in Atm-/- ES cells are predominantly chromosomal breaks, consistent with the role of ATM in non-homologous end joining (NHEJ), Atm-/KD ES cells carry high levels of both chromatid and chromosomal breaks, indicating enzymatically- inactive ATM protein causes additional repair defects in post-replicative cell cycle phases. Despite increased genomic instability, Atm-/KD lymphocytes did not have defects in chromosomal V(D)J recombination and mature B cell immunoglobulin class switch recombination beyond those observed with Atm-/- alone. Together this study, for the first time, reveals an inhibitory function of enzymatically- inactive ATM protein in DNA repair, particularly in post-replicative DNA repair

beyond NHEJ. It also uncovers an essential function of ATM protein during early embryonic development.

Results

AtmKD/KD, but not Atm+/KD mice, die during early embryonic development.

To generate mice expressing a kinase-dead ATM protein we introduced the previously described D2880A/N2885K (corresponding to D2870A/N2875K in humans) (Canman, Lim et al.

1998, Bakkenist and Kastan 2003) double mutation in the sequence encoding the conserved catalytic loop [Fig. 2-1A] of the murine Atm gene along with a floxed Neomycin Resistant

(NeoR) cassette (referred to as ATM KDN for the presence of NeoR cassette) [Fig. 2-1B]. We selected the D2870A/N2875K double mutation, since it was fully characterized for normal protein expression and the absence of kinase activity (Canman, Lim et al. 1998, Bakkenist and

Kastan 2003). Six targeted clones were identified by Southern blot analyses [Fig. 2-1C] and the mutations were verified in 4 clones by genomic sequencing. Two independent targeted clones were injected for germ line transmission. Atm+/KDN chimeras were then bred with sperm specific

Protamine-Cre transgenic mice (O'Gorman, Dagenais et al. 1997) to induce recombination between the loxP sites flanking the NeoR to generate the Atm+/KD mice, in which the kinase- dead “AtmKD” allele is transcribed and expressed from the endogenous Atm promoter. Atm+/KD mice are normal in size, fertile, and have no detectable defects in lymphocyte development

[Table 2.1, Fig. 2-5]. However no AtmKD/KD mice were identified in more than 100 pups generated by breedings between Atm+/KD mice (p-value for Χ2 test =4.0x10-8) [Table 2.1], indicating embryonic lethality. This was unexpected since Atm-/- mice are viable and born at roughly Mendelian ratio. Timed heterozygous crosses further revealed that no live AtmKD/KD embryos could be found [Table 2.1] (p-value for Chi-Squared test=0.0056) even at embryonic day 9.5-10.5 (E9.5-10.5), indicating early embryonic lethality. We note that the embryonic

Figure 2-1: Generation of AtmKD allele A.) The corresponding sequence of the catalytic loop of human and mouse ATM kinase are aligned and the mutated residues are underlined. B.) The schematic diagram represents the murine Atm locus (top), KDN KD targeting vector (2nd row), targeted allele (Atm , 3rd row), and the neo-deleted mutant allele (Atm , bottom). The 5’ probe is marked as a black line. The exons and loxP sites are shown as solid boxes and open triangle respectively. The exon containing the KD mutation is marked as an open box. Restriction site designation: X=XhoI, RV=EcoRV, H=HindIII. The map is not drawn to scale. C.) Southern Blot +/KDN analyses of EcoRV-digested DNA from representative Atm +/+(WT) and Atm ES cells.

Total ATM+/+ ATM+/KD ATMKD/KD Absorbed Total Χ2

E9.5-10.5 2 11 0 3 16 0.03

Live Birth 27 45 0 N/A 72 4.2x10-6

Table 2.1: Kinase inactive ATM leads to early embryonic lethality Genotype of embryos and live pups obtained from timed breeding between ATM+/KD parents comparing to the expected Mendelian frequency, the chi-squared test p-value = 0.03 for E9.5-E10.5 embryos and =4.22x10-6 for live birth. N/A- Not Available.

lethality of kinase-dead ATM protein is not caused by a classical dominant-negative mechanism since Atm+/KD mice and cells that express both ATM-WT and ATM-KD protein have no discernable phenotype.

AtmKD/- embryonic stem (ES) cells have greater genomic instability than Atm-/- cells.

To explore the causes of developmental failure in AtmKD/KD mice, we generated AtmKD/-

ES cells and control Atm+/+and Atm-/- cells. To circumvent the embryonic lethality, we first derived AtmKDN/C ES cells. The previously characterized ATM conditional allele (AtmC) behaves as the WT allele and can be efficiently converted to a null (Atm-) allele upon Cre-mediated recombination (Zha, Sekiguchi et al. 2008, Zha, Jiang et al. 2011). We obtained six-independent

AtmKDN/C clones and three control Atm+/C clones. AtmKD/- ES cells were then generated by infecting AtmKDN/C cells with Adenovirus expressing Cre-recombinase. While all 24 Cre- transduced Atm+/C clones were genotyped as Atm+/-, indicating efficient recombination, only two of the 96 Cre-transduced AtmKDN/C clones were genotyped as AtmKD/-, implying that AtmKD/- cells have growth and/or survival defects. Western blot analysis verified that ATM-KD protein is expressed at levels comparable to WT endogenous ATM proteins in AtmKD/- ES cells [Fig. 2-2A].

Figure 2-2: Proliferation defects and increased genomic instability in AtmKD/- embryonic stem cells A.) Western blot analysis of total protein (50g) for mouse ATM (MAT3, Sigma) and α-tubulin (EMD) in WT, Atm C/KDN, Atm -/-, Atm +/- and Atm KD/- ES cells. B.) Fold increase of cell number relative to day 0, as measured by modified MTT assay (Sigma) in WT, Atm -/- and Atm KD/- ES cells under normal growth conditions. Three independent experiments were performed in two independently derived Atm KD/- ES cells lines. The figure and the error bar represent standard deviation derived from quadruplicated cultures in one representative experiment. C.) Representative metaphases with genomic instability from Atm KD/- ES cells. Yellow arrows- Chromosome breaks, Red arrows- Chromatid breaks. D.) The frequency of chromatid and chromosome breaks measured by T-FISH analyses in WT, Atm +/- and AtmKD/- ES cells. The chi-squared test p-value for the frequency of chromatid breaks between Atm +/- and AtmKD/- ES cells is 0.002 and for chromosome breaks is 0.101.

However, two independently derived AtmKD/- ES cell lines grew significantly slower than both

Atm-/- ES and Atm+/+ ES cells, which grew at comparable rates [Fig. 2-2B]

We proceeded to measure spontaneous genomic instability in Atm+/+, Atm KD/- and Atm -/-

ES cells using Telomere-FISH (T-FISH) assays (Franco, Gostissa et al. 2006, Zha, Sekiguchi et al. 2008). Under regular growth conditions, ~30% metaphases derived from AtmKD/- ES cells had

spontaneous cytogenetic abnormalities, while only18% of those derived from Atm -/- cells, and

<2% of those derived from WT cells showed such abnormalities [Fig. 2-2C, D] In contrast to the predominant chromosome breaks in Atm -/- ES cells (Franco, Gostissa et al. 2006, Zha,

Sekiguchi et al. 2008) [Fig. 2-2D] the cytogenetic aberrations in AtmKD/- ES cells were evenly distributed between chromosome (break involving both sister-chromatids) and chromatid

(breaks involving one of the two sister-chromatids) breaks, with the most dramatic increase observed in chromatid breaks [Fig. 2-2C, D]. While chromosome breaks are derived primarily from DSBs that arise during the G1 phase of cell cycle, chromatid breaks often result from

DSBs in S and G2 phases of the cell cycle, when HR is most active. Together, these results suggest that the physical presence of ATM-KD inhibits DNA repair, likely HR, in a manner that does not occur in the absence of ATM protein. Previous studies have identified both replication- related DNA repair defects and increased chromatid breaks in cells treated with selective ATM kinase inhibitors (White, Choi et al. 2010, Gamper, Choi et al. 2012) but not in ATM null cells.

While it is unclear if the effects of ATM kinase inhibitors on DNA repair are the same as those observed in AtmKD/- cells, our findings emphasize the need to distinguish the effects of ATM deletion and kinase inhibition in future studies.

The effect of ATM-KD protein on lymphocyte development.

To test the consequences of ATM-KD protein expression in somatic cells, we generated

Rosa+/ER-CreAtmC/KDN mice. The ROSA26-ERCre (Rosa+/ER-Cre) allele ubiquitously expresses a

Tamoxifen-inducible fusion protein of estrogen-receptor ligand-binding domain and Cre recombinase (ER-Cre) under the endogenous Rosa26 promoter (de Luca, Kowalski et al. 2005,

Guo, McMinn et al. 2007). Lymphocyte development was then analyzed in conditional ATM KD

(Rosa+/ER- CreAtmC/KDN) , conditional ATM null (Rosa+/ER- CreAtmC/C(-) ) and control (Rosa +/ER-

CreAtmC/+) mice 10 days after oral Tamoxifen administration, at which time Cre-mediated

recombination of the AtmC and AtmKDN alleles was readily observed in bone marrow, thymus and spleen. Western blot of total thymocytes showed that ATM KD protein is expressed at levels similar to WT protein in thymocytes [Fig. 2-3A]. Upon irradiation, the levels of phosphorylated KAP-1 and H2AX (γ-H2AX) in thymocytes purified from Tamoxifen-treated

Rosa+/ER-CreAtmC/KDN mice were comparable to those from Rosa+/ER-CreAtmC/C mice, confirming the loss of ATM kinase activity in AtmKD/- thymocytes [Fig. 2-3B]. Compared to WT controls

(182±27x106), the total thymocyte number in Tamoxifen-treated Rosa+/ER-CreAtmC/KDN mice

(60±17x106) was reduced to a level similar to Rosa+/ER-CreAtmC/C (54±12x106) and Atm-/-

(42±16x106) mice (n>=3 for each group). FACS analyses of Tamoxifen-treated Rosa+/ER-

CreAtmC/KDN mice revealed thymocyte-developmental defects, namely decreased TCRβ surface expression and decreased SP percentage in the thymus, comparable to those of Tamoxifen- treated Rosa+/ER-CreAtmC/C or Atm-/- mice (Borghesani, Alt et al. 2000) [Fig. 2-4A]. Together these data suggest that kinase-dead ATM protein partially blocks T cell development and, by extension, chromosomal V(D)J recombination, to a level comparable to the absence of ATM protein.

Similarly, B cell development in Tamoxifen-treated Rosa+/ER-CreAtmC/KDN mice was largely comparable to that of Atm -/- mice [Fig. 2-4A]. To test the effect of ATM-KD protein in CSR, we incubated partially purified (CD43-) Atm KD/- and control splenic B cells with anti-CD40 plus interleukin 4 (IL-4) for 4 days to stimulate CSR to IgG1. B-cells from Tamoxifen treated Rosa+/ER-

CreAtmC/C(-) mice performed CSR at ~40% of WT levels with increased genomic instability, similar to those of germline Atm-/- mice [Fig. 2-4A, B], confirming functional loss of ATM protein in B cells from Tamoxifen-treated Rosa+/ER-CreAtmC/C(-) mice. Nevertheless, B cells from Tamoxifen treated Rosa+/ER-CreAtmC/KDN mice underwent CSR at levels comparable to their ATM-null counterparts [Fig. 2-4A, B]. Consistent with our findings on AtmKD/- ES cells, activated B cells from Tamoxifen-treated Rosa+/ER-CreAtmC/KDN mice accumulated greater genomic instability, especially chromatid breaks, than Atm-/- B cells upon activation [Fig. 2-4C]. Together, those

Figure 2-3: ATM KD lymphocytes are catalytically inactive A.) Western blot for total ATM protein in thymocytes from Tamoxifen treated Rosa+/ER-CreAtm+/C, AtmC/C or AtmC/KDN mice and control Atm-/- mice. B.) Western blot for phosphorylated H2AX, phosphorylated KAP-1 and total KAP-1and Actin in irradiated thymocyte (5 Gy) with or without pre-incubation with 15 µM ATM kinase inhibitor (KU55955, Tocris Bioscience) or 5µM DNA-PKcs kinase inhibitor (NU7441, Tocris Bioscience). Cell lysates were collected 2 hours after irradiation. ATMi- ATM inhibitor (KU55933, pretreated 1 hour at 15µM), DNA-PKi- DNA-PK inhibitor (NU7441, pretreated 1 hour at 5µM)

results support the conclusion that the ATM-KD protein inhibits DNA repair in S and G2 phase of cell cycles.

Transgenic expression of an autophosphorylation-mimetic, kinase-dead double mutant

ATM rescues the embryonic lethality of ATM KD mice

ATM undergoes trans-autophosphorylation on Ser1981 (Ser1987), among other residues, upon activation. However, the exact function of ATM autophosphorylation is controversial. While human ATM with an alanine substitution at Ser1981 abrogated ATM activation and loss of retention at DNA damage sites, murine ATM with S1987A mutation, as well as ATM with several other phosphorylation site mutations, function normally in transgenic

Figure 2-4: Analysis of ATM KD lymphocytes A.) Representative flow cytometric analyses of total thymocytes stained with CD4, CD8, and TCRβ, and total splenocytes stained with B220 and IgM surface markers. For the B220/IgG1 marker staining, CD43- splenocytes were isolated and stimulated in culture with anti-CD40 and IL-4 for 4 days before staining. A total of five independent experiments were performed and one set of representative FACS analyses is shown. B.) Relative frequency of the IgG1+ cells among all B220+ B cells (relative to WT cells in each experiments) in Atm-/- and AtmKD/- B cells. The data represents the average and standard deviation from at least five experiments. The chi-squared test p-values are marked in the graph. C.) The frequency (per metaphase) of chromatid and chromosome breaks measured by T-FISH analyses in stimulated B cells from WT, Atm-/- mice or Tamoxifen treated Rosa+/ER-CreAtm+/C, AtmC/C or AtmC/KDN mice. The chi-squared test p-value for the frequency of chromatid breaks between Atm +/- and AtmKD/- B cells cells is 0.007 and for chromosome breaks is 0.816.

Figure 2-5: Characterization of Atm+/KD mice Representative flow cytometric charts of total thymocytes stained with CD4, CD8, and TCRβ, bone marrow with B220, and CD43, and total splenocytes with CD4, CD8, B220, and IgM. A total of four independent experiments were performed and one set of representative FACS analyses is shown.

mouse models (Bakkenist and Kastan 2003, Pellegrini, Celeste et al. 2006, Daniel, Pellegrini et al. 2008, So, Davis et al. 2009). The kinase-dead ATM does not display a classical dominant- negative phenotype, as Atm+/KD mice develop normally and do not display lymphocyte developmental defects as those observed in Atm-/KD mice [Fig. 2-5]. We hypothesized that trans- autophosphorylation of the KD protein by the wildtype protein relieves the toxic effect of the KD.

To explore this possibility, we generated mice carrying ATM with both a phosphomimetic

Figure 2-6: Generation of the autophosphorylation-mimetic, kinase-dead (DKD trangene) Atm Mouse. A.) General scheme for the generation of DKD Atm-/- mice. B.) FISH for the endogenous Atm locus and the DKD transgene using the ATM-containing BAC (RP24-122F10) as a probe on metaphases collected from ear fibroblasts from founder mice A and C. Yellow triangles indicate BAC integration sites.

S1987D mutation and the KD mutation (referred to as DKD). The genomic sequencing encoding the S1987 and D2880/N2885 sites are ~23 kbs away from each other, rending it technically difficult to generate simultaneous mutations on the same allele with sequential targeting or breeding using the conventional targeting approach. We thus employed bacterial artificial chromosome (BAC) recombineering to introduce the two mutations into an Atm-containing BAC

(RP24-122F10), and generated BAC transgenic (Tg) mice carrying the DKD allele.(Yang and

Sharan 2003, Pellegrini, Celeste et al. 2006) [Fig. 2-6A]. Another group has utilized similar method to introduce two different ATM KD alleles (e.g. D2899A and Q2740P) and test their function by crossing into Atm null backgrounds. Atm-/- Tg-ATMD2899A and Atm-/- Tg-ATMQ2740P

Tg- DKD Tg+ Founder ATM ATM ATM ATM ATM ATM Total χ² +/+ +/- -/- +/+ +/- -/- A Actual 12 16 9 6 15 10 68 0.74 Expected 8.5 17 8.5 8.5 17 8.5 C Actual 5 13 6 9 13 7 53 0.93 Expected 6.625 13.25 6.625 6.625 13.25 6.625 D Actual 11 22 7 3 7 7 57 0.03 Expected 7.125 14.25 7.125 7.125 14.25 7.125 E Actual 3 5 1 2 4 1 16 0.88 Expected 2 4 2 2 4 2 F Actual 4 4 2 3 5 0 18 0.55 Expected 2.25 4.5 2.25 2.25 4.5 2.25

Table 2.2: Tabulation of live-born mice between DKD Atm+/- (F1) and Atm+/- breedings

also died during embryonic development (Daniel, Pellegrini et al. 2012). Each founder line positive for the DKD Tg were assessed for BAC integration using FISH, and backcrossed to

Atm+/- to generate DKD Atm-/- mice [Fig. 2-6B]. 4 of the 5 transgenic lines that gave germline transmission of DKD produced viable DKD Atm-/- F2 offspring, rescuing the embryonic lethality of the KD [Table 2.2]. The rescue of the KD embryonic lethality could be due to the loss of transgene expression, which would render DKD Atm-/- mice to be the same as Atm-/-. RT-PCR, followed by KasI digestion to specifically detect DKD expression, confirmed the presence of

DKD mRNA expression in ear fibroblasts harvested from DKD Atm-/- mice [Fig. 2-7A, B].

However, detectable level of ATM protein cannot be found by Western blot [Fig. 2-7C].

Additionally, DKD Atm-/- mice succumbed to thymic lymphomas similar to those observed in Atm-

/- mice [Fig. 2-7D]. Based on these finding, we consider the possibility that the S1987D autophosphorylation-mimetic mutation on the KD protein causes the destabilization of the DKD protein, and rescues the embryonic lethality of just the KD alone. Consequently, DKD Atm-/- phenocopies Atm-/- and develop thymic lymphomas just as the Atm-/- mice

Figure 2-7: DKD Atm-/- mice do not express ATM protein, and develop thymic lymphomas A.) RT-PCR scheme for both the wildtype and transgene-introduced DKD Atm allele. The DKD transgene contains an added KasI site that allows for distinguishing between the wildtype and DKD alleles. Expected PCR, digestion size fragments are indicated in bps. B.) RT-PCR of total RNA collected from ear fibroblasts harvested from F2 progeny of the indicated genotypes. C.) Western blot of ATM and Actin as a loading control. D.) Representative FACS of thymocytes harvested from wildtype (WT) and thymic lymphoma-bearing DKD Atm-/- mice

Discussion

Past studies of ATM function in DNA repair have focused primarily on its kinase activity.

Here we and others, have shown that kinase-dead ATM proteins causes early embryonic lethality despite the normal embryonic development in the complete absence of ATM protein

(Daniel, Pellegrini et al. 2012). We further show that AtmKD/- ES and B cells accumulate much

higher levels of chromosome instability, particularly chromatid breaks, relative to ATM-null cells.

Despite the increased genomic instability, Atm KD/- lymphocytes carried out V(D)J recombination and CSR at levels comparable to ATM-null cells. Together these data suggest that while loss of

ATM kinase activity and ATM-mediated phosphorylation of downstream targets compromise both NHEJ and, to a lesser extent, HR function, the ATM-KD protein elicits additional inhibitory effects in post-replication DNA repair, potentially HR. By uncovering unexpected functions for

ATM protein in DNA repair and embryonic development, our study raises clinically relevant questions about the mechanism of ATM activation and the application of ATM kinase inhibitors.

While NHEJ plays critical roles in lymphocyte-specific DSB repair events, it is largely dispensable for embryonic development (Lieber 2010). In contrast, most HR factors are required for embryonic development. In this context the embryonic lethality of AtmKD/KD, but not Atm-/- mice may reflect an inhibitory function of ATM-KD protein in a subset of HR. Consistent with this hypothesis, the most dramatic increase of general genomic instability in AtmKD/- cells over ATM null cells is chromatid breaks, which are typically associated with defects in S and G2 phase

DNA repair.

ATM belongs to the family of PI3K-related kinase (PIKK), which also includes DNA-PKcs and ATR. While ATR is activated by RPA-coated ssDNA (Shiotani and Zou 2009, Liu, Shiotani et al. 2011), ATM and DNA-PKcs are both activated by DNA DSBs, phosphorylate an overlapping pool of substrates (e.g., H2AX, KAP1, and p53) (Callen, Jankovic et al. 2009, Zha,

Guo et al. 2011, Zha, Jiang et al. 2011) and have redundant functions during embryonic development and DNA repair (Gurley and Kemp 2001, Sekiguchi, Ferguson et al. 2001, Callen,

Jankovic et al. 2009, Gapud, Dorsett et al. 2011, Zha, Jiang et al. 2011). It is possible that the presence of an enzymatically-inactive ATM protein might prevent DNA-PKcs, and potentially other PIKKs, from phosphorylating shared substrates, thereby disrupting DNA repair to a greater extent than complete loss of ATM alone. Loss of both ATM and DNA-PKcs led to CSR

defects more severe than loss of either kinase alone (Callen, Jankovic et al. 2009), but AtmKD/- and Atm-/- B cells performed CSR at similar levels [Fig. 2-4A, B] inconsistent with inhibition of

DNA-PKcs function in Atm KD/- B cells. Moreover, H2AX and KAP-1, two shared substrates of

ATM and DNA-PKcs, are phosphorylated at comparable levels in AtmKD/- and Atm-/- B and T cells [Fig. 2-3D]. While phosphorylation of particular substrates of DNA-PKcs or DNA-PKcs itself may be inhibited in AtmKD/- cells, but not ATM-/- cells, our findings are inconsistent with global inhibition of DNA-PKcs activity in AtmKD/- cells. In this context, loss of ATM phosphorylation sites on DNA-PKcs leads to more severe defects in DNA repair than DNA-PK null mice (Zhang, Yajima et al. 2011). In addition, ATM carrying both the phosphomimetic at

Ser1987 and kinase-dead mutations lead to its destabilization and lack of protein expression.

The autophosphorylation of ATM may be a mechanism to prime the degradation of the protein after the DNA damage. As a consequence, mice heterozygous for the KD allele over wildtype have no discernable phenotype, consistent with the autophosphorylation-mediated degradation of the KD protein.

Chapter 3 – Phosphorylation-dependent kinase-structure functions of ATM in DNA repair‡

‡Work described in this chapter is submitted as a manuscript for publication under:

Expression of kinase kinase-dead ATM protein is common in human cancers, highly oncogenic and can be preferentially targeted by Topo-isomerase I inhibitors

Kenta Yamamoto1,2, Jiguang Wang3, Jun Xu5, Christopher J. Haddock6, Chen Li1, Brian J. Lee1, Denis G. Loredan1, Alessandro Vindigni6, Dong Wang5, Raul Rabadan3 and Shan Zha1,4

1 Institute for Cancer Genetics, Department of Pathology and Cell Biology, College of Physicians and Surgeons, Columbia University, New York City, NY 10032 2 Pathobiology and Molecular Medicine Graduate Program, Department of Pathology and Cell Biology, Columbia University, New York City, NY 10032 3 Department of Biomedical Informatics and Department of Systems Biology, College of Physicians & Surgeons, Columbia University, New York City, NY 10032 4 Division of Pediatric Oncology, Hematology and Stem Cell Transplantation, Department of Pediatrics, College of Physicians & Surgeons, Columbia University, New York City, NY 10032 5 Skaggs School of Pharmacy & Pharmaceutical Sciences, University of California San Diego, La Jolla, CA 92093 6 Edward A. Doisy Department of Biochemistry and Molecular Biology, Saint Louis University School of Medicine, St. Louis, MO 63104

Background and Significance

In the previous chapter, we discovered that while Atm-/- mice are viable, mice solely expressing ATM KD are embryonic lethal. In addition, we observed greater genomic instability in both ES cells and lymphocytes expressing ATM-KD, we hypothesized that the ATM KD protein is blocking DNA repair through a mechanism not observed in Atm null cells. In particular, the greater accumulation of chromatid breaks in the telomere FISH assay, and early embryonic lethality suggest additional defects in the HR pathway.

In this chapter, we further investigate the pathways of DNA repair affected by the expression of ATM KD. We discovered that ATM KD ES cells display defects in HR, and AtmKD/-

Mouse Embryonic Fibroblasts (MEFs) are more sensitive to Camptothecin (CPT) than Atm-/-

(ATM-null) counterparts. Specifically, AtmKD/- cells were unable to upregulate γH2AX, Rad51 foci and subsequent sister-chromatid exchange upon CPT challenge, suggesting a defect elicited by the ATM KD protein upstream of DSB formation upon TopI-DNA adduct formation. Accordingly,

AtmKD/- cells accumulated greater levels of DNA-bound Topoisomerase-I (TopI) covalent complexes (Top1ccs) upon CPT treatment compared to Atm-/- cells, indicative of a defect in the resolution TopI-DNA lesions generated by CPT treatment. The loss of DSB formation by ATM

KD expression occurs via the blockage of structure-specific nucleases, such as SLX4 and

MUS81, in a MRE11-dependent manner. AtmKD/- cells also displayed slower progression of replication, underlying defects in DNA replication resulting from ATM KD expression. We have thus identified novel structure-kinase roles of ATM in homologous recombination, removal of

Topoisomerase-I lesions via formation of DSB intermediates, and DNA replication.

Results

ATM KD cells are hypersensitive to Camptothecin (CPT), and cannot upregulate CPT- induced Sister Chromatid Exchange (SCE) and RAD51 foci

To understand how expression of kinase-dead ATM increases genomic instability, we compared the genotoxin sensitivity of immortalized Rosa+/CreERT2Atm+/C, Rosa+/CreERT2AtmC/- and

Rosa+/CreERT2AtmC/KD murine embryonic fibroblasts (MEFs). These cells were treated with three rounds of 4-OHT (200nM) to induce nuclear translocation of ER-fused Cre recombinase, inactivation of the conditional AtmC allele, and generation of Atm+/-, Atm-/- and AtmKD/- MEFs, respectively [Fig. 3-1A] (Yamamoto, Wang et al. 2012). All experiments were conducted on freshly deleted cells to avoid secondary alterations in genetically-unstable Atm-deficient cells.

While both AtmKD/- and Atm-/- MEFs were similarly sensitive to ionizing radiation (IR), AtmKD/- cells were especially sensitive, relative to Atm-/- and Atm+/- cells, to the Topoisomerase I inhibitor camptothecin (CPT) [Fig. 3-2B]. The specific CPT hypersensitivity of AtmKD/- cells does not appear to reflect an inability to process blocked DNA ends in general, as both Atm-/- and AtmKD/- cells were equally sensitive to the topoisomerase II inhibitor etoposide [Fig. 3-2C]. Replication defects in general is also unlikely to account for the CPT hypersensitivity, since AtmKD/- cells are not especially sensitive to hydroxyurea (HU) or aphidicolin (APH), two agents that block replication [Fig.3-2D, 3-2E]. These findings indicate that expression of kinase-dead ATM protein specifically sensitizes cells to Topo-I poisons, and suggest that Topo-I inhibitors might be prioritized for the treatment of cancers with catalytically-inactive ATM mutations.

CPT and related clinical Topo I blockers (Irinotecan or Topotecan) trap Topo I in a covalent complex (“Top1cc”) at the 3’ end of the single-strand DNA nick induced by Topo I cleavage. If not properly resolved, Top1cc prevents re-ligation of the single-strand DNA break,

Figure 3-1: Analyses of immortalized AtmKD/- and control MEFs A) Treatment scheme of immortalized Rosa-CreERT2 MEFs. B.) PCR and C.) Western blot analyses of the MEFs before and after three rounds of 4OHT treatment (200nM). The result confirmed the efficient ATM deletion and the expression of kinase-dead ATM protein. The Western blot also shows the normal expression of DNA-PKcs, Mre11 and Kap1 in the AtmKD/- cells.

interferes with transcription and replication, and ultimately elicits cellular cytotoxicity (Pommier

2006). In this regard, we showed that AtmKD/- MEFs accumulated Top1cc even at a very low dose of CPT (0.1µM) [Fig. 3-3D]. At high concentrations (15µM CPT), Atm-/- cells also accumulates Top1cc at a higher level than Atm+/+ cells [Fig. 3-3D], likely due to a kinase- independent function of ATM previously described in neuronal cells (Alagoz, Chiang et al. 2013,

Katyal, Lee et al. 2014). The antibody used for Top1cc assay was developed against a peptide that lies within 10 amino acids of the tyrosine covalently linked to DNA (Bethyl, Cat: A302-590).

Therefore, the Top1cc that accumulate in AtmKD/- cells may be comprised of intact Topo I or, more likely, a Topo I proteolytic fragment (Pommier 2006). In this context, CPT-induced degradation of intact Topo1 protein is not affected in AtmKD/- cells or by ATM kinase inhibitor

[Fig. 3-3E]. Together these data suggests that Atm KD protein blocks removal of Top1cc.

Top1cc can be removed in both replication-dependent and -independent manners [see

Fig. 3-5 for model]. Indepedent from replication, Top1cc can be removed by TDP1, which

Figure 3-2: AtmKD/- MEFs are hypersensitive to CPT and cannot upregulate CPT induced SCE and RAD51 foci Sensitivity of Atm+/-, Atm-/- and AtmKD/- immortalized MEFs (iMEFs) to A.) gamma-irradiation, B.) camptothecin (CPT), C.) Etoposide, D.) Hydroxyurea (HU), E.) Aphidocolin, and F.) Olaparib. *P<0.001.G.) Representative images and frequency of sister chromatid exchange (SCE) per chromosome for vehicle (DMSO) or CPT (1µM) treated MEFs. Red diamonds point to exchange events .H.) Representative Rad51 foci 10 hours after exposure to 5Gy of IR, and the quantification for Rad51 and I.) phosphorylated RPA (pT21). Brca1SF/SF cells are homozygous for the S1598F mutation, and are unable to recruit Rad51(Shakya, Reid et al. 2011). Scare bar~ 10µm. J.) The frequency of CPT (0.1µM, 2 hour)- induced Rad51 foci in different cells. The bar graph represents the average frequency of cells with greater than 5 foci per cell calculated across 3 independent experiments, and the error bars indicate SEM. All p- values in this figure were calculated based on two-tailed Student’s t-test assuming unequal variances.

hydrolyzes the tyrosine-DNA linkage, or by 3’ flapases (e.g., XPF, Mus81, Eme1 or SLX1/4) that excise the oligonucleotide bearing the conjugated Topo-I polypeptide (Pommier 2009, Rouse

2009), creates a single strand gap, which is filled in and ligated. During DNA replication, Top1cc is recognized upon encounting with replication forks, which promotes fork regression. The regressed fork could be substrates for structure specific endonucleases (e.g. Mus81 or SLX1/4), which cleave the fork structure and generate DSBs. The DSBs is subsequently repaired by the homology-dependent repair (HDR) The double holiday junction intermediates formed during

HDR can be dissolved by Bloom helicase (Blm) to create non-crossover (NCO) or resolved by nucleases (e.g. SLX1/4, GEN1, Mus81, Eme1) to generate both crossover (CO) and NCO products(Chan and West 2014). The CO event could be conveniently scored as a sister chromatid exchange (SCE).

In a previous study on human cancer cell lines, ATM kinase inhibitor reduced CPT induced SCEs (White, Choi et al. 2010). Consistent with this findings, we observed that ATM kinase inhibitors (K55933, 15µM) reduced CPT-induced SCEs in Atm+/-, but not Atm-/-, MEFs

[Fig. 3-2G]. Moreover we observed that CPT (1µM) induced robust SCEs in Atm+/- and Atm-/-

MEFs, but not in AtmKD/- MEFs [Fig. 3-2G]. Together these results suggest that the presence of catalytically inactive Atm protein prevents CPT-induced CO. Yet ATM kinase inhibitor did not significantly reduce SCEs in Blm-/- MEFs, indicating that ATM KD protein did not suppress CO

by skewing holiday junction resolvases (e.g., Mus81, SLX4/1 or GEN1) (White, Choi et al. 2010)

[Fig. 3-2G]. Next we measured DNA end resection, the first step of HDR, by the formation of phosphorylated-RPA (pRPA-T21) and Rad51 foci. While ionizing radiation (IR) induced phosphorylated-RPA (T21) and RAD51 foci were largely abolished in control BRCA1-deficient cells (Shakya, Reid et al. 2011), they are readily induced in both AtmKD/- and Atm-/- cells [Fig. 3-

2H, 3-2I], suggesting that Atm KD protein does not block resection or Rad51 filament formation upon DSB formation. Meanwhile we found that CPT-induced Rad51 foci were significantly reduced in AtmKD/- cells in comparison to Atm-/- or Atm+/- cells [Fig. 3-2J]. This specific defect in

CPT-induced, but not IR-induced, Rad51 foci formation in AtmKD/- cells suggests that the Atm KD protein blocks resolution of Top1cc at a step upstream of DNA resection and HDR.

ATM KD cells are defective in CPT-induced DSB formation, Top-Icc Resolution

DSB is an obligate intermediate for HDR dependent repair of Top1cc. Given the defects in CPT-induced SCEs and Rad51 focus formation in AtmKD/- cells, we measured the levels of

CPT-induced Ser139 phosphorylated H2AX (-H2AX), a marker for DNA DSBs. Two hours after

0.1µM CPT treatment, -H2AX foci and -H2AX protein accumulate in Atm+/- and Atm-/- cells, but not in AtmKD/- cells or Atm+/- cells treated with ATM kinase inhibitor [Fig. 3-3A, 3-3B]. We note that CPT-induced phosphorylation of RPA at threonine 21 (T21), a site that is primarily phosphorylated by ATR/DNA-PKcs and a marker for ssDNA (Marechal and Zou 2015), is not affected in AtmKD/- cells[Fig. 3-3B], suggesting that extended CPT-induced ssDNA could be generated independently of DSBs. Consistent with the lack of DSBs specifically, CPT (0.1µM) increased alkaline comet tail length [Fig.3-3C], an indicator of ssDNA breaks, in all three genotypes, but no significant increase in neutral comet tail length, a measure of DSBs, in

AtmKD/- cells as it did in Atm+/- and Atm-/- cells [Fig. 3-3C]. Baseline neutral comet tails are longer in AtmKD/- cells >

Figure 3-3: AtmKD/- MEFs cannot generate CPT induced DSBs and accumulate TopI-cc A.) Representative images and quantification of CPT-induced γH2AX foci (0.1µM CPT for 2 hours). The dot plot represents data collected in three biological replicates with over 200 nuclei quantified per genotype per condition. Scale bar ~10µm. B.) Western blots for γH2AX, p-RPA (T21), p-KAP1(S824) of cells treated with either Vehicle (-, DMSO) or CPT (+, 0.1µM) for 2 hours. ATM inhibitor (15µM, Ku55933) was added 1 hour prior to CPT/vehicle treatment if needed. C.) Quantification of Neutral COMET tail lengths of cells treated with Vehicle (-, DMSO) or CPT (+, 15µM) for 1 hour. Two replicates of the experiments were performed with over 50 comets quantified for each. D.) In-vitro Complex of Enzymes (ICE) assay probing for Top-I (top panel) and radiolabeled mouse genomic DNA (lower panel). The amounts (in g) of genomic DNA loaded on each row are indicated on the left of the top panel. This is the representative result from at least 3 independent experiments. E.) Western blot of whole-cell lysates collected from MEFs treated with indicated concentrations of CPT for 2 hours.

Atm-/- > Atm+/- cells [Fig. 3-3C], correlating with their levels of spontaneous genomic instability.

These results suggest that ATM KD protein blocks DSB formation upon CPT treatment.

ATM KD blocks structure-specific nuclease activity in a MRE11-dependent manner

During replication, CPT-induced DSBs can be generated by direct “run off” of the replication polymerase at the single-stranded DNA break or by endonucleases that process the stalled replication forks or reversed fork structures as previously revealed by crosslink agents

(Zhang and Walter 2014). SLX4 is a scaffold protein that interacts with three nucleases, SLX1,

Mus81, and XPF(Rouse 2009), which all have in vitro nuclease activity consistent with fork cleavage. Among them, XPF nuclease forms an obligate dimer with ERCC1. We observed that

CPT-induced DSBs are also significantly reduced in Slx4-/- MEF and, to a much lesser extent, in

Mus81-/- MEFs, but not in Ercc1-/- MEFs, suggesting that SLX4/Mus81 and potentially SLX1, but not XPF, might function in the same pathway suppressed by Atm KD protein [Fig. 3-4A].

Accordingly ATM kinase inhibitor completely abolished the residual CPT-induced -H2AX in

Slx4-/- MEFs [Fig. 3-4A]. Notably cells deficient for SLX4, Mus81, or ERCC1, all display reduced levels of CPT-induced T21p-RPA [Fig. 3-4A]. The loss of CPT-induced pRPA, but not H2AX, foci in Ercc1-/- cells, together with the reduction of CPT-induced H2AX, but not pRPA, in AtmKD/- cells, suggests that CPT can induce ssDNA breaks and DSBs through independent mechanisms. In this context, SLX1/4, Mus81 and XPF/ERCC1 all have 3’ flapase activity which can directly cleave the Top1cc to generate ssDNA gaps (Rouse 2009)[Fig. 3-4A]. But only

SLX4/SLX1 and Mus81 have the 5’ flapase activity and ability to cleave the cruciform structures and are implicated in cleavage of DNA crosslinks or Holliday junction structures (Rouse 2009,

Clauson, Scharer et al. 2013). Finally, complete deletion of Mre11 (Mre11-/-) rescued the CPT- induced DSBs in the presence of ATM kinase inhibitor [Fig. 3-4B], suggesting that the KD ATM is recruited by the MRN complex to DNA, where it blocks CPT-induced DSB formation. Notably, nuclease-deficient MRE11 is able to form a stable complex with NBS1 and Rad50 (Buis, Wu et al. 2008) but did not rescue CPT-induced DSBs in cells treated with ATM kinase inhibitor [Fig.

3-4B], implying that the presence of the MRE11 protein, but not its nuclease activity, is required

Figure 3-4: Mre11-dependent recruitment of Atm KD protein blocks resolution of Top1cc by SLX4/Mus81/SLX1 nuclease A.) Western Blots of Slx4-/-, Mus81-/- or Ercc1-/- MEFs. Cells were either treated 2 hours with Vehicle (-, DMSO) CPT (C, 0.1µM), or pre-treated 1 hour with 15µM Ku55933 prior to CPT treatment (CA). LE- Long exposure, SE- Short Exposure. B.) PCR (above) and C.) Western blots of Mre11+/-, Mre11-/- or Mre11-/nuc MEFs. Mre11+/C, Mre11C/- and Mre11C/nuc immortalized MEF was stably infected with retrovirus expressing ER-CRE- IRES-hCD2, MACS purified for hCD2+ cells and then induced with 4OHT (200nM) for 48 hours. The PCR confirmed the efficient conversion of the Mre11 conditional allele to the null allele as previously described (Buis, Wu et al. 2008). For the western blots, cells were either treated 2 hours with Vehicle (-, DMSO) CPT (C, 0.1µM), or pre-treated 1 hour with 15µM Ku55933 prior to CPT treatment (CA). F) Representative images of the DNA combing assay Cells were incubated 30 minutes with 25µM IdU, followed by 250µM of CldU along with either vehicle (DMSO) or CPT (1µM). The dot plot represents the summary of two independent experiments. *P<0.0001.

for the recruitment of ATM to CPT-induced lesions during replication. Notably, ATM inhibitors only partially reduced CPT-induced DSB in WT cells, yet completely abolished CPT-induced

DSBs in Mre11-nuclease defective cells [Fig. 3-4B], suggesting that the nuclease activity of

MRE11 may contribute to the release of ATM in the vicinity of the CPT lesions [Fig. 3-5].

Figure 3-5: Working model for ATM KD blockage of DSB formation upon Top-I conjugation In non-proliferating cells, Top1cc is removed by TDP1 or 3’ flapases to generate single stand breaks, which is repaired by PARP1 and Ligase3 mediated pathways. In replicating cells, Top1cc collision with replication forks leads to fork reversal as well as strand breakages. ATM KD protein selectively inhibits fork breakage in a MRE11 dependent way.

Discussion

In this section, we identified novel phosphorylation-dependent structural functions of

ATM in HR, the processing of Topo1 lesions, and replication. AtmKD/- MEFs displayed hypersensitivity to CPT compared to Atm+/- and Atm-/- cells due to Top1cc accumulation resulting from the loss of DSB formation upon CPT treatment. Moreover, we show that ATM KD protein, when recruited to the replication fork by MRE11, blocks SLX4/SLX1- and Mus81- mediated cleavage of Top1cc lesions during replication. These results identify novel functions of

ATM upstream of the formation of DNA DSBs, and provide additional insight into previously unknown structural functions of ATM in DNA repair.

While a direct effect of ATM KD lesions on HDR cannot be excluded, the slow fork progression in ATM KD cells might indirectly compromise HDR as measured by the DR-GFP reporter HR measured by DR-GFP is suppressed in Atm-/KD cells but not in Atm-/- cells. But phosphorylated RPA and Rad51 foci are only reduced in CPT treated, but not IR treated Atm-/KD cells. While defects in generating CPT induced DSBs could explain the lack of CPT induced

Rad51 foci and SCEs in Atm-/KD cells, why I-SceI nuclease induced “clean” DSBs could NOT be efficiently repaired by HR in the AtmKD/- cells remains elusive. One possibility is that ATM KD suppresses a later step of HR, such as the resolution of the holiday junctions by BLM and other nucleases. Notably, some of the nucleases contribute to holiday junction resolution (e.g. Mus81 and SLX1/4) are also implicated in the strand cleavage upon ICL or possibly also Top1cc blocks

(Rouse 2009, Zhang and Walter 2014). It is tempting to speculate that catalytically inactive ATM might block the same nuclease(s) to suppress CPT induced DSBs and HR. Alternatively, the slow progression of the replication forks in the Atm-/KD cells might indirectly reduce HR, as measured by DR-GFP, by reducing the number of HR compatible cells in culture.

Intermolecular auto-phosphorylation has been described for several PI3KK family members, including ATM, DNA-PKcs and ATR (Bakkenist and Kastan 2003, Dobbs, Tainer et al. 2010, Liu, Shiotani et al. 2011, Jiang, Crowe et al. 2015). Using a mouse model expressing the kinase dead DNA-PKcs protein, we showed that DNA-PKcs protein physically blocks end- ligation during NHEJ at the DNA ends in a Ku dependent manner and requires autophosphorylation for end-ligation (Jiang, Crowe et al. 2015). Here our data suggest that in a similar manner, ATM protein blocks the Top1cc induced strand cleavage at the replication fork in Mre11-dependent fashion that is regulated by its auto-phosphorylation. Our data also further suggests that Mre11 nuclease activity might have a role in releasing ATM KD from the replication fork. In this context, Sae2 and Mre11 have been found to limit the DNA damage responses elicited by Tel1 and Mec1 – yeast homolog of ATM and ATR respectively (Clerici,

Mantiero et al. 2006, Chen, Donnianni et al. 2015). Therefore it is tempting to suggest that autophosphorylation of PI3KKs might have a common function in promoting their release from their respective activation partners (Ku for DNA-PKcs, Mre11 for ATM), regulates specific repair events (NHEJ for DNA-PKcs, replication for ATM) and terminate the DNA damage responses.

In the AtmKD/- cells, this replication defect, combined with the lack of ATM-mediated checkpoint function and NHEJ defects, together leads to embryonic lethality and severe genomic instability.

Chapter 4 - Structure-kinase function of ATM in tumor suppression, and the mutational spectrum of ATM in human tumors‡

‡Work described in this chapter is submitted as a manuscript for publication under:

Expression of kinase kinase-dead ATM protein is common in human cancers, highly oncogenic and can be preferentially targeted by Topo-isomerase I inhibitors

Kenta Yamamoto1,2, Jiguang Wang3, Jun Xu5, Christopher J. Haddock6, Chen Li1, Brian J. Lee1, Denis G. Loredan1, Alessandro Vindigni6, Dong Wang5, Raul Rabadan3 and Shan Zha1,4

Background and Significance

Germline expression of ATM KD led to unexpected embryonic lethality in mice, and cells expressing ATM KD displayed hypersensitivity to CPT and defects in replication. However, due to the early embryonic lethality of ATM KD mice, whether the ATM KD still retains tumor suppression function in vivo remained undefined. To address this, we generated mice conditionally expressing ATM KD in hematopoietic stem cells (HSCs) using Vav-cre. Strikingly,

AtmKD/- mice developed aggressive thymic lymphomas with higher penetrance and kinetics compared to Atm-/- mice. Detailed genomic analyses identified more frequent deletion of PTEN in AtmKD/- thymic lymphomas compared to the null. Additionally, stimulated AtmKD/- T-cells were hypersensitive to a combination treatment of CPT and PARP inhibitor, providing insight into a therapeutic regimen to selectively treat tumors with ATM KD compared to the null. Therefore, the expression of a catalytically inactive ATM is more oncogenic than ATM null, and can be selectively treated using CPT and PARP inhibitors to exploit ATM KD blockage of Top-Icc removal via DSB formation mechanism.

Somatic mutations of ATM have been identified in human tumors, but the types of mutations and their functional consequences that occur in both germline A-T and somatic mutations identified in human tumors are not well understood. To address this, here we report that cancer-associated ATM mutations are highly enriched (>70%) for missense mutations clustering within the kinase domain, while A-T-associated germline ATM mutations are predominantly truncation mutations that abrogate protein expression. Several tumor-associated missense mutations identified within the kinase domain are predicted to affect residues directly involved in the kinase activity of ATM, suggesting the expression of KD ATM in human tumors.

Together, these findings provide a mechanism for the diverse cancer risk associated with different types of ATM mutations, and suggest that cancers expressing catalytically inactive

ATM protein could be preferentially targeted by Topoisomerase I inhibitors, a class of drugs that is already available in the clinic.

Results

Expression of kinase-dead ATM is more oncogenic than complete loss of ATM

Although Atm-null animals develop normally(Barlow, Hirotsune et al. 1996), mice harboring germline kinase-dead (KD) Atm mutations (AtmKD/KD or AtmKD/-) (equivalent to N2875K

& D2889A of human ATM, N2875K was found twice in TCGA cases) undergo embryonic lethality (Daniel, Pellegrini et al. 2012, Yamamoto, Wang et al. 2012), explaining the lack of kinase domain missense mutations in live-born A-T patients. Nevertheless, it remains unclear whether the kinase-dead ATM lesions can promote malignant development. Therefore, to compare the oncogenic potential of Atm-KD vs Atm-null mutations, we generated Vav-

Cre+AtmC/KD(VKD) and Vav-Cre+AtmC/C (VN, N for null) mice, in which the Vav-promoter drives expression of Cre-recombinase in hematopoietic stem cells (HSCs), leading to deletion of the conditional ATMC allele in all hematopoietic cells including lymphoid cells [Fig. 4-1A]. .

Thymocytes from VKD or VN mice display the classic spectrum of T cell developmental defects characteristic of ATM-/- mice; namely, reduced CD3 surface expression and partial blocks at the

CD4+CD8+ double-positive (DP) to CD4+CD8- and CD4-CD8+ single-positive (SP) T cell transitions (Borghesani, Alt et al. 2000) [Fig. 4-1B]. Similarly, VKD and VN B cells show a ~50% reduction in class switch recombination (CSR) efficiency [Fig. 4-1B], confirming efficient Atm inactivation in VKD and VN lymphocytes. Notably, the frequency of double-negative 3 (DN3) T cells is increased 1.5 fold in VKD mice, but not in VN or ATM-/- mice [Fig. 4-1B, 4-1C]. The DN3 stage is characterized by rapid proliferation after β-selection. Accordingly, telomere FISH

Figure 4-1: Lymphocyte development and characterization of the VKD and VN mice A.) Southern Blot for the Atm conditional vs deleted allele in kidney, bone marrow, splenocytes, thymocytes and activated splenic B cell (CD43-, LPS stimulated) of a representative VN mouse, indicating complete deletion of Atm in lymphoid lineage. KpnI-digested genomic DNA probed with ATMCKO3’ probe B.) FACS analyses show normal B cell development and both class switch recombination and T-cell development defects in B cells from VN and VKD mice similar to those observed in Atm-/- mice. LPS+IL-4 column show the frequency of IgG1+ B cells after the CD43- purified splenocytes were activated by LPS and IL-4 for 4.5 day. CD44 and CD25 double staining was performed on live gated Double Negative (DN) thymocytes defined as negative for surface staining of CD8, CD4, CD19, TCRɤδ and Ter119. C.) CSR efficiency. Bar graphs represent %IgG1+ cells collected at day 4.5 after LPS, IL-4 stimulation. Error bars indicate SEM. *P<0.05 between ctrl and each of the 3 other genotypes per two-tailed Student’s t-tests assuming unequal variances. D.) The total thymocyte counts from 4-8 week old VN (n=3), VKD (n=3) mice. E.) Telomere FISH analyses of metaphases spread from Con A-stimulated T-cells (3 days). Bar graphs represent the average number of cytogenetic aberration per metaphase (MP) from two independent mice per genotype. The error bars indicate SEM.

analyses revealed cytogenetic abnormalities in 37% VKD, 27% VN and 3% Atm+/+ Con A- stimulated peripheral T cells [Fig. 4-1E]. Consistent with previous reports in activated B cells

(Yamamoto, Wang et al. 2012), the most dramatic increases of the genomic instability in the

VKD T cells are chromatid breaks, suggestive of replicative/post-replicative repair defects [Fig.

4-1E]. As a result of the DN3 block, the total thymocyte number was reduced 66% in 6-week-old

VKD mice compared to age-matched VN littermates [Fig. 4-1D]. Despite reduced thymocyte numbers, VKD mice succumbed to thymic lymphomas 35.5 days earlier than littermate control

VN mice [Fig. 4-2A]. Overall 75% of VKD and 56% VN mice died of lymphomas within 400 days

[Fig. 4-2B]. While VN mice, like Atm-/- mice, uniformly succumbed to CD3low immature thymic lymphomas, 2 of the VKD mice (8%) developed B220+CD43+IgM- pro-B cell lymphomas [Fig. 4-

2B, 4-2E, 4-2F]. The thymic lymphomas from VKD and VN mice share a common immunophenotype (CD3low, CD4+CD8+ DP or immature CD8+ SP) and all contain clonally rearranged TCRβ loci [Fig. 4-2D]. Comparative genomic hybridization (CGH) on four VKD and one VN T cell lymphomas identified common copy number alterations characteristic of ATM-/- thymic lymphomas (Zha, Bassing et al. 2010, Jiang, Lee et al. 2015), including frequent trisomy

15, focal amplification at Chr14 upstream of the TCR/δ loci, and deletion of chromosome 12

Figure 4-2: Expression of Atm KD protein alone in hematopoietic stem cells (HSCs) is more oncogenic than complete loss of Atm in mouse models A.) Kaplan-Meier (K-M) survival curve of VN (n=18) and VKD (n=24) mice. The red and blue dots denote thymic lymphomas from the VN and VKD cohorts, respectively. The yellow dots denote the B cell lymphomas in the VKD cohort. The T1/2 for thymic lymphoma development is 139.5 days (VN) and 104.0 days (VKD) respectively. The asterisk marks the difference in T1/2 of thymic lymphoma development between the two cohorts. *P=0.03 per Mantel-Cox/log-rank test. B.) The overall frequency of B and T cell lymphomas in VN, VKD cohort monitored for 400 days. C.) FACS analyses of thymocytes harvested from VN (red bracket) and VKD (blue bracket) mice with thymic lymphomas. The dash line marks the level of surface TCRβ expression in control thymocytes D.) ) Southern blot analyses of thymocyte DNA digested with EcoRI and probed with TCRJβ1.6 and TCRJβ2.7 probes. Marker: Molecular Weight Marker. Ctrl: Kidney DNA harvested from a VavCre- AtmC/KD mouse, GL-Germline. E.) FACS analyses of the splenocytes from control and two VKD mice with B-cell lymphomas. F) Southern blot analyses of splenocyte DNA harvested from VKD mice with B-cell lymphomas. Ctrl: Kidney DNA harvested from a VavCre- AtmC/KD mouse, GL-Germline.

telomeric ends [Fig. 4-3A]. Also, Pten deletions, which are associated with ~25% of Atm-/- lymphoma (Zha, Bassing et al. 2010), were observed in all four VKD tumors but not in the VN tumor [Fig. 4-3A, 4-3B]. Western analyses of additional tumors revealed lack of Pten protein expression in all, but one, VKD tumor [Fig. 4-3C]. In contrast, most VN tumors retained residual

Pten expression, and the only Pten-expressing VKD tumor (6684) lacked expression of the ATM

KD protein [Fig. 4-3C]. Interestingly, Pten loss, which is known to synergize with Atm or Tp53 deficiency in lymphomagenesis, has been reported to occur in early lymphoid progenitors before

TCRβ rearrangement in Tp53-/- T cells, as 10 week thymocytes from Tp53-/- mice contain oligoclonal T cells with diverse TCRβ rearrangements yet identical PTEN deletions (Dudgeon,

Chan et al. 2014). Based on these findings, we propose that, unlike complete ATM loss, expression of catalytically-inactive ATM further increases genomic instability, leading to frequent

Pten inactivation in lymphoid progenitors and early onset lymphomagenesis.

AtmKD/- MEF cells are hypersensitive to CPT and PARP inhibitor Olaparib compared to

Atm null cells [Chapter 3, Fig. 3-2F]. Additionally, cells resolve CPT-induced Top-1ccs via two pathways- one that generates a DSB intermediate, while the other is exclusively SSB. DSB generation requires nucleases SLX4/MUS81 which is blocked in the presence of KD ATM in a

MRE11-dependent manner, while the SSB mechanism requires ERCC1 and gap filling repair

Figure 4-3: ATM KD tumors undergo more frequent loss of Pten compared to ATM null A.) CGH analyses of 1 VN, and 4 VKD tumors. The Y axis is Log ratio (base 2) of tumor/kidney from the same mice. The red arrow heads indicate trisomy 15 and the green error heads pointed to the deletion around Pten gene. Each chromosome is demarcated by gray lines, and the chromosome numbers are marked at the top of the CGH panels. B.) Expanded copy number analyses of around the Pten gene on mouse chromosome 19 (mm8: 32,779,983-32,908,796). The y-axis is the natural log ratio of tumor/kidney from the same mice. Log ratio of -0.67 (dotted line) indicate heterozygous deletion. C.) Western blot of Pten, total Atm and β-actin for Atm+/+ (WT Thy), VKD, and VN thymic lymphomas.

Figure 4-4: VKD T-cells are hypersensitive to CPT and PARP inhibitor combination treatment Relative survival of ConA-stimulated mature T-cells harvested from the spleen of control (ctrl) VN, VKD mice. The T-cells were harvested from the spleen, and continually stimulated with ConA for 72 hours. The cells were treated with drugs for 48hours after incubation with only ConA for 24 hours. (Right) EC50 plot for the VKD T-cells. Concentrations at which the relative survival was 50% for CPT and PARP inhibitor single treatments are shown on the X and Y axis, respectively. The green dot represents the concentration for both drugs at which there was 50% relative survival for the dual CPT/PARP inhibitor treatment.

mediated by LIGASE3 and PARP [Chapter3, Fig. 3-5]. We therefore hypothesized that VKD cells are hypersensitive to the inhibition of a combination of CPT treatment with PARP inhibition.

Indeed, ConA stimulated mature T-cells from VKD mice are synergistically hypersensitive to dual CPT and Olaparib treatment compared to wildtype and VN cells [Fig. 4-4]. We therefore propose that, despite the increased kinetics and incidence of tumors with ATM KD mutations compared to wildtype and ATM null, KD tumors are selectively hypersensitive to a combination of CPT and PARP inhibitor treatment, and demonstrate a therapeutic intervention method to specifically target kinase-dead ATM-expressing tumors.

Missense mutations in the kinase domain of ATM are enriched in human tumors

~87% of A-T-associated ATM lesions (447 unique mutations from >1000 cases) are truncation or frame-shift mutations [Fig.4-5A], consistent with the observation that ATM protein is undetectable in most A-T patients (Becker-Catania, Chen et al. 2000). In contrast, we found that among the 286 unique non-synonymous somatic mutations of ATM identified from 5,402 cancer cases in The Cancer Genome Atlas (TCGA), only 25% are truncation/frame-shift mutations and 72% are missense mutations [Fig.4-5A]. Furthermore, cancer-associated ATM missense mutations are clustered primarily in the kinase domain (AA2711-2962 of human

ATM), ~2.5-fold over expected [Fig. 4-5B], while both cancer- and A-T- associated truncation/frameshift mutations are distributed throughout the entire ATM protein [Fig. 4-5B].

Together, these findings suggest a significant subset of the cancer-associated ATM mutations is functionally distinct from those implicated in A-T.

To determine the functional consequences of the ATM kinase domain mutations, we performed homology modeling of human ATM kinase domain using the crystal structure of human mTOR (PDB 4JSP)(Yang, Rudge et al. 2013), another PI3K-related protein kinase, as modeling template(Yang, Rudge et al. 2013). Primary sequence alignment suggests that the kinase domain of ATM and mTOR are highly conserved with a sequence identity of 30% [Fig. 4-

6A]. The model predicted that residues K2717, D2720 and H2872 of human ATM are directly involved in ATP binding. In particular, residue H2872 binds the -phosphate of ATP, while amino acids D2870, N2875 and D2889 coordinate Mg+ metal ion, which is essential for ATM kinase activity [Fig.4-6B, 4-6C, 4-6D]. Within the ribbon structure, a subset of the cancer-associated

ATM missense mutations are highlighted in red [Fig. 4-6B, 4-6C]. Based on the structure, we conclude that many cancer associated ATM mutations, including D2720N (2 cases), D2721N

(1), L2722M (2), D2725G (1), as well as H2872R (1), N2875S/K(2), L2885H(1), L2890R (1) and

G2891R (1), are likely to abolish ATM kinase activity [Fig.4-6D]. In addition, several other

Figure 4-5: Cancer associated ATM mutations are enriched for kinase domain missense mutations A.) Classification of the unique and non-synonymous mutations reported for A-T patients in the Leiden open variable database (http://chromium.lovd.nl/LOVD2/home.php?select_db=ATM) (n=447) or identified as somatic mutations from 5402 unique cancer cases sequenced by TCGA (n=286). “Others” category includes splice-site and stop codon mutations. B.) A heatmap for the frequency of cancer-associated truncation and frameshift mutations (upper panel) or non-synonymous missense mutations (lower panel) along human the ATM polypeptide obtained using Gaussian kernel based analyses (online method). Hotter color indicates denser mutation distribution. The shadowed curve (bottom panel) reflects fold of mutation over expected mutation frequency, calculated by Gaussian kernel method. The confidence interval (shadow) is estimated by fitting binomial distribution (online method). The dash line at 1 represents the expected rate of mutations. The x-axis is the amino acid number (from 1-3056) along human ATM protein. mutations on the outer surface of the ATM kinase domain (e.g. R3008H) may also interfere with kinase activity.

Human cancer cell lines that harbor ATM missense mutations display defects in CPT induced DSBs formation and are also hypersensitive to CPT.

Based on information in the Cancer Cell Line Encyclopedia (CCLE) (Broad Institutes), we identified a panel of human cancer cell lines with potential KD mutations [Fig.4-7A]. The expression of ATM protein and the lack of ATM kinase activity were validated by Western blot

Figure 4-6: Missense ATM mutations identified in human tumors affect the kinase domain A.) Permutation analyses of actual truncation or missense mutation frequencies within the ATM gene or within the ATM kinase domain over the expected mutation frequency based on the size of the gene and the frequency of non-synonymous mutation. The dash line marks the expected mutation rate B.) The distribution of truncation/frame shift mutations in A-T patients. The blue-shaded region represents the 95% confidence interval C.) Sequence alignment of human ATM with mouse ATM and human mTOR. The sequences were aligned with ClustalW and the resulting output was rendered with ESPript. The AAs conserved between ATM and mTOR are indicated with red stars. D.) Superposition of the human ATM model (yellow) and the crystal structure of human mTOR (blue). Mutations conserved between ATM and mTOR are shown in red. The active site architecture of human ATM. ATP is shown in red, Mg2+ ion is shown in purple, the catalytic residues from ATM are shown in blue and the substrate is colored black. for ATM protein and IR induced phosphorylation of Kap-1, a prominent substrate of ATM [Fig.4-

7B]. Notably CPT (100nM) induced -H2AX was significantly reduced in those cell lines in comparison to the ATM+/+ HBL2 cells [Fig. 4-7C]. Both Granta519 and HBL2 are MCL cell lines with IgH-CyclinD1 translocation and ectopic expression of CyclinD1. Granta519 cells that express a catalytically inactive ATM are significantly more sensitive to CPT than HBL2. [Fig.4-

7D]. Together these results suggest that human cancer cell lines expressing catalytically inactive ATM also confer hypersensitivity to Topo1 inhibitors.

Discussion

Our findings reveal that cancer-associated ATM lesions are predominately missense mutations that compromise the kinase activity of ATM. We further show that expression of catalytically-inactive ATM protein is more oncogenic than simple loss of ATM. This is in part due to increased genomic instability in AtmKD/- HSCs, which leads to frequent deletion of PTEN.

Moreover, the majority of ATM mutations in human cancers reported in the TCGA are missense mutations that are highly enriched in the PI3K domain, and human tumor lines expressing mutant ATM are hypersensitive to CPT, similar to that observed in AtmKD/- MEFs and T-cells.

Our results identify crucial distinctions between ATM mutations and ATM loss in tumorigenesis,

Figure 4-7: ATM mutated human tumor lines are also hypersensitive to CPT. A.) Table with the annotated ATM mutations in human cancer cells. B.) Western blots of total cell lysates harvested from the indicated human tumor lines for ATM (upper), and pKap1 (lower) with (+) or without (-) 5Gy Irradiation (IR). C.) CPT Sensitivity assay for two Mantle Cell Lymphoma (MCL) cell lines HBL-2 (ATMwt/wt) and Granta519 (ATMR2832C/-). D.) Western blot of human tumor lines treated for 2 hours with either Vehicle (-, DMSO) CPT (C, 0.1µM), or pre-treated 1 hour with 15µM Ku55933 prior to CPT treatment (CA).

and suggest the use of Topo1 inhibitors- a class of drugs already available in the clinic, in the treatment of ATM mutated cancers.

The functional significance of many newly emerging missense mutations of ATM from cancer genomic analyses is not fully understood. Using genomic, structural modeling and mouse genetics tools, we showed that missense mutations in the ATM kinase domain leading to the expression of a kinase dead ATM protein are prevalent in human cancer, more oncogenic than complete loss of ATM, and most importantly confer a hypersensitivity to Topo1 inhibitors.

ATM is already routinely included in cancer genetic panels. Our findings suggest that patients carrying tumor with kinase inactivating mutations of ATM would likely benefit from Topo1 inhibitor, a class of drug that is already available in the clinic. In addition, our study also identified an auto-phosphorylation dependent structural function of the ATM protein during DNA replication that is “upstream” of the DSBs. We further showed that this role of ATM in replication is relatively specific for Top1cc-induced, but not to HU or Aphidicolin-induced, replication fork stalling and require Mre11 protein, but not its kinase activity. Together, this novel function of

ATM in replication expands the repertoire through which ATM suppresses tumorigenesis and has implications on the mechanism of replication dependent repair, cancer etiology and cancer treatment.

ATM suppresses tumorigenesis through its function in checkpoint activation and repair.

The role of ATM during checkpoint activation is primarily mediated by its kinase activity and is abolished by either the null or the KD mutations. ATM null cells primarily accumulate chromosomal breaks, consistent with a major role of ATM kinase in pre-replication DNA repair especially NHEJ (Franco, Gostissa et al. 2006). Accordingly, A-T patients or ATM null mice are mostly affected by lymphoid malignancies owing to the important role of NHEJ in lymphocyte specific DSB repair and somatic mutations of ATM in lymphoid malignancies (e.g. MCL, CLL and T-PLLs) including an even mixture of truncating and missense mutations (Stankovic,

Stewart et al. 2002, Bea, Valdes-Mas et al. 2013). In chapter 3, we show that ATM KD suppresses replication dependent repair of Top1cc lesion and a subset of HR in addition to

NHEJ. Defects in HR, such as mutation in BRCA1 or 2 genes, confer increased risk for breast, ovarian or pancreatic cancers, in which recurrent ATM mutations, especially frequent missenses mutations, were also identified (Roberts, Jiao et al. 2012, Cremona and Behrens 2014). Thus, our findings suggest that the variable risk to diverse type of cancers might be explained by how specific ATM mutations affect the different repair pathways. It also suggests that the oncogenic potential of the ATM mutations might correlate with the level of catalytically inactive ATM protein as higher expression of KD ATM would likely be more oncogenic, and might be modulated by interaction between the mutated ATM and MRN complex.

The selective hypersensitivity of AtmKD/- cells to Topo1 inhibitor provides a potential targeted therapy for ATM mutated cancer. This result also suggests that AtmKD/- cells would be sensitive to other drugs targeting Top1cc removal and repair pathways, including the PARPs which facilitates stand ligation and TDP1 that cleaves the Top1cc (Pommier 2009, Murai, Zhang et al. 2014). Indeed, a previous study showed that ATM mutated Grant519 cells are hypersensitive to PARP inhibitors (Williamson, Kubota et al. 2012) and we also found that Atm

KD/- T cells are hypersensitive to the combination of low dose PARP inhibitors and CPT [Fig. 4-4].

AtmKD/- cells confer selectively sensitivity to Topo1 inhibitor, but not HU and APH [Chapter 3,

Fig. 3-2D, 3-2E]. We found that ATM KD protein prevent CPT induced DSBs. In addition to

Topo1 inhibition, interstrand crosslink (ICL) agents (e.g. mitomycin C, cisplatins) also stimulate fork cleavage and generate replication dependent DSBs as the intermediates(Zhang and Walter

2014). Therefore, ICL agents might also be beneficial for patients with Atm mutated cancers.

Together these findings further expand the treatment options for patients with ATM mutated cancers. One of the next challenges would be how to reliably identify actual ATM KD mutations in the clinic, as sequence analysis is not sufficient to predict ATM function in most cases. In this context, the CPT induced -H2AX levels described here offer a convenient method to determine

the functional implication of ATM mutations, including the possible confounding effects of MRN mutations.

Chapter 5 - Roles of ATM in the suppression of B-cell lymphomas♯

♯Work described in this chapter is published under:

Early B-cell Specific Inactivation of ATM Synergizes with Ectopic CyclinD1 Expression to Promote Pre-germinal center B-cell Lymphomas in Mice

Kenta Yamamoto1,4,, Brian J. Lee1, Chen Li1, Richard L. Dubois1, Elias Hobeika5, Govind Bhagat2, and Shan Zha1,2,3

1 Institute for Cancer Genetics, 2 Department of Pathology and Cell Biology, 3 Department of Pediatrics, 4 Graduate Program for Pathobiology and Molecular Medicine, College for Physicians and Surgeons, Columbia University, New York, NY 10032 5Centre for Biological Signaling Studies BIOSS, Albert-Ludwigs-Universität Freiburg, Department of Molecular Immunology, Faculty of Biology, Albert-Ludwigs-Universität Freiburg and Max Planck Institute for Immunobiology, Stübeweg 51, 79108 Freiburg, Germany

Leukemia (2015) Jun;29(6):1414-24. PMID: 25676421

Background and Significance

Mature B-cell lymphomas represent 85% of non-Hodgkin Lymphomas (NHL) and many harbor characteristic chromosomal translocations involving the immunoglobulin (Ig) loci. ATM is frequently inactivated in human B-cell NHL. Among all B-NHLs, MCL has the highest frequency of bi-allelic inactivation of ATM (Bea, Valdes-Mas et al. 2013). MCL is a mature B-cell lymphoma composed of small to medium-sized lymphocytes with frequent (~30%) involvement of the gastrointestinal tract. The variable regions of the Ig are unmutated in the majority of MCL cases, consistent with a pre-GC origin. MCL is characterized by deregulated expression of D- type cyclins, especially CyclinD1, via the characteristic t(11;14) chromosomal translocation that joins CyclinD1 with the active Ig-heavy chain gene (IGH) promoter and enhancer. Yet, ectopic expression of wild type (WT) CyclinD1 is insufficient to induce MCL in mouse (Bodrug, Warner et al. 1994, Lovec, Grzeschiczek et al. 1994), implying additional genetic alterations are necessary. Effective treatment for MCL is not available, and animal models recapitulating molecular features of human MCL are yet to be developed. In addition, the role of ATM inactivation in B-cell lymphomagenesis is not yet well understood. This is in part due to the lack of ATM-deficient B-cell lymphoma model, as germline ATM-null mice develop early, aggressive thymic lymphomas.

In this chapter, we report that early and robust deletion of ATM in precursor/progenitor

B-cells causes cell-autonomous, clonal mature B cell lymphomas of both pre- and post-germinal center (GC) origins. Unexpectedly naïve B cell specific deletion of ATM is not sufficient to induce lymphomas in mice, highlighting the important tumor suppressor function of ATM in immature B cells. While EµCyclinD1 is not sufficient to induce lymphomas, EµCyclinD1 accelerates the kinetics and increased the incidence of clonal lymphomas in ATM-deficient B- cells and skews the lymphomas towards pre-GC derived small lymphocytic neoplasms sharing

morphological features of human MCL. This is in part due to CyclinD1-driven expansion of ATM- deficient naïve B cells with genomic instability, which promotes the deletions of additional tumor suppressor genes (i.g. Trp53, Mll2, Rb1 and Cdkn2a). Together these findings define a synergistic function of ATM and CyclinD1 in pre-germinal center B-cell proliferation and lymphomagenesis and provide a prototypic animal model to study the pathogenesis of human

MCL.

Results

Mouse models with B-cell specific deletion of ATM

To circumvent early lethality due to thymic lymphomas in germ line Atm knockout mice

(Atm-/-), we generated B-cell specific deletion of Atm using CD21Cre, CD19Cre, or Mb1+/Cre in combination with the ATM conditional allele (ATMC) (Zha, Sekiguchi et al. 2008). CD21Cre allele

(Kraus, Alimzhanov et al. 2004) mediates specific and robust ATM deletion in IgM+ naïve B-cells and CD19Cre+ATMC/C (Rickert, Rajewsky et al. 1995) results in ATM deletion ranging from 60% in bone marrow pre-B-cells to nearly 100% in naïve splenic B-cells [Fig. 5-1A] Despite efficient deletion of ATM in naïve splenic B-cells in both CD21Cre+ATMC/C and CD19Cre+ATMC/C mice as evidenced by Southern blot analyses, CSR defects, and genomic instability, none of the

CD21Cre+ATMC/C (n=23) or CD19Cre+ATMC/C (n=36) mice developed definitive B-cell lymphoproliferations in >28 month follow-up period [Fig. 5-1B, C], by which time the bone marrow samples were virtually devoid of B-cells.

Based on this observation and the postulated “early” deletion of ATM in human MCL

(Jares, Colomer et al. 2007), we focused on Mb1Cre(Hobeika, Thiemann et al. 2006), which is the earliest B-cell specific Cre allele available, that leads to specific and robust cre activation in early pro-B/pre-B-cells (Kwon, Hutter et al. 2008). We generated four cohorts, Mb1+/creATM+/+(C)

Figure 5-1: Lack of B-cell lymphomas in CD21Cre+ ATMC/C and CD19Cre+ ATMC/C mouse models. A) Southern blot analyses show efficient deletion of ATM conditional allele in naïve and stimulated B-cells from CD21Cre+Atm+/C and CD19Cre+Atm+/C mice. Total survival (D) and tumor-free survival (E) of CD21Cre+AtmC/C and CD19Cre+AtmC/C and the corresponding Atm+/C controls.

(hereafter referred to as M) Mb1+/CreAtmC/C(-)EµCyclinD1- (MA), Mb1+/cre(+)Atm+/+(C)EµCyclinD1+

(MD/D) and Mb1+/creAtmC/C(-) EµCyclinD1+ (MAD). First, we confirmed the efficient and specific deletion of the ATM gene and protein in splenic B-cells from MA mice by Southern [Fig. 5-2A] and Western blotting [Fig. 5-2B] respectively. In B-cells purified from MA mice, irradiation induced phosphorylation of Kap-1, an ATM specific substrate (Ziv, Bielopolski et al. 2006), was largely abolished confirming the loss of ATM kinase activity [Fig. 5-2C]. Meanwhile, T-cells from

MA or MAD mice were devoid of the development defects associated with ATM deficiency

(Borghesani, Alt et al. 2000) – namely reduced surface CD3/TCR expression and reduced

CD4 or CD8 single positive T-cells in the thymus- consistent with normal ATM function in T-cells from MA or MAD mice [Fig. 5-2D] Similarly, myeloid (Gr1+ or CD11b+) and erythroid (Ter119+) lineages were also unaffected in the bone marrow and spleen of MA and MAD mice [Fig. 5-2D]

Together, these data support the specific

Figure 5-2: B-cell specific deletion of ATM in Mb1+/CreATMC/- and Mb1+/CreATMC/-EµCylinD1+ mouse models A.) Southern blot analyses of the Atm locus with genomic DNA harvested from kidney (Kid), thymus (Thy), bone marrow (BM), total spleen cells (Spl), LPS/IL-4 stimulated splenic B-cells B.) and Con A stimulated splenic T-cells (T). B) Western Blot analyses for CyclinD1 and ATM in stimulated splenic B and T-cells harvested from MD, MA, and MAD mice. C.) Phosphorylation of Kap1 in LPS/IL4 stimulated B-cells (day 3) with or without irradiation (IR,10Gy). Protein lysate is collected 2 hours after IR. D.) Representative flow cytometry analyses of bone marrow (BM), spleen and thymocytes from Atm-/- as well as MA, MAD and MD mice.

and efficient deletion of ATM in developing B-cells. In the Mb1+/Cre mice, the Cre knock-in disrupts the endogenous Mb1/CD79a gene in the targeted allele (Hobeika, Thiemann et al.

2006). Since Mb1/CD79a is essential for B-cell development and Mb1/CD79a-/- B-cells arrest at the pro/pre- B-cell stage (Minegishi, Coustan-Smith et al. 1999, Pelanda, Braun et al. 2002), we also confirmed normal B-cell development and spleen cellularity in control MD/D, MA and MAD mice (all carrying heterozygous Mb1+/Cre alleles) and only used Mb1+/Cre for all breeding and final tumor cohorts [Fig. 5-2D]. Finally, ectopic expression of CyclinD1 in both B and T-cells was also verified in EµCyclinD1+ MD and MAD mice by Western blotting [Fig. 5-2B

B-cell specific deletion of ATM leads to B-cell autonomous lymphomas of both GC and non-GC phenotype

To determine the role of ATM deficiency in B-cell lymphomagenesis,We followed a cohort of MA mice with weekly palpation to identify any abnormal growths. During the 20-month follow-up, 12.5% (3/24) of MA mice developed overt splenomegaly and lymphadenopathy, with frequent intestinal involvement (67%, 2/3) [Fig. 5-4A]. Flow cytometry analyses revealed monoclonal expansion of CD19+ (often B220+) B-cells, which was further confirmed by Southern blot analyses that detected clonal IgH rearrangements in all confirmed cases [Fig. 5-3A, B]. In the same period, none of the Mb1Cre/+Atm+/+(C/+) mice developed any signs of tumors. At >20 month of age, all tumor cohorts were euthanized for end point analyses. One additional MA mouse

(2023) was found to have a B-cell lymphoma based on clonal IgH rearrangement and histology

[Fig. 5-3B, C] bringing the lifetime risk for monoclonal lymphoproliferative disease to 4/24

(16.7%) in MA mice. Histopathologic and immunophenotypic analyses revealed diffuse large B- cell lymphomas (DLBCL) of splenic or nodal origin in the MA mice, with one manifesting a GC phenotype (BCL6+, IRF4-) and three exhibiting a non-GC phenotype (BCL6+/-, IRF4+)

Figure 5-3: Clonal B-cell lymphoproliferations in 24 month plus MA and MD/D mice and representative flow cytometry analyses of MAD mice A) Representative flow cytometry analyses of spleen from tumor bearing MA and MD/D mice. B) Southern Blot analyses of DNA harvested from kidney (K) and spleen (S) from MA and MD/D mice. Asterisk denotes mouse 2594, which the DNA is partially degraded. Blue arrow heads point to clonal rearrangements. GL: Germ line. Probe for the constant region of TCRδ is used as loading control. C) Histological analyses of represented MA tumors. 2594 is GC-type DLBCL. The white pulp is expanded and effaced by large lymphocytes with centroblastic features and admixed tingible body macrophages. The neoplastic lymphocytes express B220 and BCL6 and lack IRF4 expression, the immunophenotypic features are indicative of a DLBCL of GC phenotype. 2583 is non-GC type DLBCL. A predominant and extensive red pulp infiltrate of large centroblast-like cells is present. The neoplastic lymphocytes express PAX5 and IRF4, but lack B220 and BCL6 expression (note: the white pulp component appears to express B220), the immunophenotypic features are indicative of a DLBCL of non-GC phenotype.

[Table 5.1, Fig. 5-3C]. Given the GC and non-GC immunophenotypes, we analyzed the IgH for

SHM and found evidence of SMH in 2/4 MA DLBCL, indicating a post-GC cell of origin. Based

on these data, we conclude that deletion of ATM in early pre-/pro- B-cells leads to infrequent mature B-cell lymphomas of heterogeneous cell origins with long latency. These results support a role of ATM as a tumor suppressor gene during early B cell development and also highlight the need for additional genetic hits.

CyclinD1 overexpression accelerates B-cell lymphoproliferations in MA mice and skews lymphomas towards pre-GC origin

Among all human B-NHLs, MCL, characterized by ectopic expression of CyclinD1, has the highest frequency of bi-allelic inactivation of ATM (Bea, Valdes-Mas et al. 2013). Given the early deletion of ATM during MCL development, it was proposed that ATM deficiency might increase IgH-CyclinD1 translocation to promote MCL (Bea, Valdes-Mas et al. 2013).

Alternatively, ectopic CyclinD1 expression might synergize with ATM deficiency to promote cellular proliferation at the price of genomic instability to drive B-cell lymphomagenesis. To test this, we bred MA mice with EµCyclinD1 transgenic mice to generate the MAD and control MD/D mice. At two months of age, B-cell number and B-cell development in MAD mice is comparable with MA and MD controls [Fig. 5-2D]. By 20-months of age, 3/18 (16.7%) MAD mice had developed overt splenomegaly and lymphadenopathy with occasional intestinal involvement and leukemic presentation [Table 5.1, Fig. 5-4B]. The tumor free survival of MAD mice was significantly shorter than either MA or MD/D control groups [Fig. 5-4A]. Moreover, at the end- point analyses (>20 month), 46.6% (7/15) of MAD mice showed clonal B-cell expansions by flow cytometry analyses and clonal IgH and Igκ rearrangements by Southern blotting [Fig. 5-4B, D,

E] bringing the life-time risk for B-cell lymphoproliferations in MAD mice to 55.6% (10/18). This is in sharp contrast to the control MD/D mice, which did not develop overt splenomegaly within the 20-month follow-up and only 2/12 MD/D mice showed evidence of B-cell lymphoproliferation upon histopathologic analyses at necropsy [Fig. 5-4B]. By flow cytometry, MA and MAD

Figure 5-4: MA and MAD mice develop clonal B-cell lymphoproliferations A) Tumor-free survival of MA, MAD and MD/D cohorts up to 20 month of age. Number of mice followed in each cohort is listed in the parentheses. *P<0.05 per Mantel-Cox/Log rank test for statistical significance. B) Lifetime risk for B-cell lymphoproliferations in MA, MAD and MD/D mice. The gray boxes represent tumors discovered within 20-month follow-up period and the white boxes represent tumors discovered at end-point analyses (> 20-month). C) Representative images of enlarged spleen and lymph node (from MAD mice 3468) and small intestine (right, from MA mice 2023). The black arrow heads point to the lymph nodes. D) Representative flow cytometry analyses of spleen from tumor bearing MA and MAD mice. The postulated tumor populations are indicated with a circle. E) Southern Blot analyses of DNA harvested from Kidney (K) and Spleen (S) from MAD mice. Asterisk denotes mouse 3613, in which the tumor primarily resided in bone marrow (B). Mouse 3468 has leukemia with significant infiltration in the kidney, thus the presence of clonal rearrangements. Blue arrow heads point to clonal rearrangements. Probe for the constant region of TCRδ is used as loading control. Densitometry intensity ratio of Myc/TCRδ is set at 1.0 for the kidney of 3557. GL: Germ line.

Surface Marker (FACS) IHC No. Age Genotype (M) Diagnosis Organs SHM B220 IgM Others PAX5 BCL6 IRF4

2594 MA 18 DLBCL S, Ln - +++ + PNA+ ND + - 2583 MA 20 DLBCL S ,Ln - + + + - + + 2023 MA 25 DLBCL S, Ln (1/710) - - ND + + + 3465 MA 19 DLBCL S, K (3/710) + + IgD- + - + S,Ln, PB, 3468 MAD 13 DLBCL Liv, K - +++ +++ ND - + 3710 MAD 23 LPD S - +++ +++ IgD- ND - +/- + 3557 MAD 24 LPD* S (2/710) - +++ + - +/- 3613 MAD 16 LPD*♯ S, BM - +++ +++ + - +/- 4271 MAD 21 LPD*♯ S - + +++ + - +/- 4159 MAD 21 LPD*♯ S - + +++ + - +/- + 4161 MAD 21 LPD*♯ S, Ln (1/710) - +++ CD5+ + - +/- 3560 MAD 24 LPD*♯ S - + + + - +/- 3713 MAD 17 LPD♯ S - + + CD5+ + - +/- 2569 MAD 23 LPD♯ S, Ln - +++ +++ + - +/- S, Ln, PB, + 2012 MD/D 26 DLBCL K (3/710) +++ +++ IgD- ND - + 2603 MD/D 24 LPD* S - + +++ CD5lo ND - +

Table 5.1: Pathological characteristics of clonal B-cell lymphoproliferations of MD/D, MA and MAD mice Organ involved: S-spleen, Ln-lymph node, K-kidney, PB- peripheral blood, Liv-liver, BM-Bone marrow. IRF4+/- indicates mixed IRF4 expression.*- Red Pulp Infiltration. ♯-Variable mantle and marginal zone expansion. ND-Not determined. SHM- Somatic Hypermutation in Jh4

lymphoproliferations had similar immunophenotypes, namely CD19+B220hi/loIgMhi/lo [Fig. 5-3A, 5-

4D]. The B-cell identity of B220 negative lymphoproliferations was further validated by PAX5

IHC staining [Fig. 5-3C, 5-5] All tumors were CD21/35-CD23- [Fig. 5-4D] excluding GC B-cell origin. 2/10 (20%) MAD tumors expressed CD5, a marker that is frequently associated with activated B-cells and also

expressed by human MCL [Table 5.1, Fig. 5-4D] In addition to DLBCL observed in 2/10 (20%)

MAD mice, including one exhibiting prominent red pulp involvement, a spectrum of indolent lymphoproliferations were noted in the remainder. All 8 lymphoproliferative disorders (LPDs) were characterized by variable degrees of white pulp enlargement, predominantly due to follicular mantle and marginal zone expansion without discernable GCs. Of note, 5/10 (50%)

MAD tumors showed significant infiltration of the red pulp by small to medium-sized lymphocytes, resembling the pattern of splenic involvement by human CLL and MCL [Table 5.1,

Fig. 5-5]. The immunophenotype of the DLBCL indicated a non-GC B-cell origin (BCL6-, IRF4+), as did that of the LPDs (PAX5+, BCL6-, IRF4+/- and CD138-). Consistent with these observations, while 50% (2/4) of MA lymphomas showed IgH variable region gene mutations, only 2/10 (20%) of the MAD lymphomas showed evidence of SHM. These findings suggest that

CyclinD1 overexpression synergizes with ATM deficiency to promote predominantly pre-GC B- cell lymphoproliferations recapitulating some morphological and immune phenotypes of human

MCL.

MAD tumors share a subset of molecular features with human MCL

To further characterize the genetic lesions that underlie lymphomagenesis in the MAD mice, we performed CGH analyses on 3 MAD, 1 MA and 1 MD tumors with high tumor content

(>50% by FACS). CGH identified a large number of gross chromosomal gains and losses in all tumors [Fig. 5-6A]. Specifically, two MAD tumors (3468 and 4161) display focal deletion involving the Trp53 tumor suppressor gene, including clear homozygous deletion in

Figure 5-5: Histopathologic and immunophenotypic analyses of B-cell lymphoproliferations of MAD mice Representative photomicrographs showing the spectrum of B-cell lymphoproliferations observed. (3468, H&E) DLBCL. The white pulp is markedly expanded and the architecture effaced by an infiltrate of large pleomorphic lymphocytes. The neoplastic cells express B220 (blue). Scattered small disrupted BCL6+ germinal centers are seen and the neoplastic cells are BCL6 negative but they show IRF4 expression, findings compatible with a DLBCL of non-GC phenotype. (4271, H&E) LPD with red pulp lymphocytic infiltrate. The white pulp is disrupted by expansions of the follicular mantle and marginal zones and a diffuse infiltrate of small-sized lymphocytes is present in the red pulp. The neoplastic lymphocytes show B220 (blue), PAX5 and IRF4(variable) expression and they lack BCL6 expression. (4161, H&E) DLBCL. Clusters of large lymphocytes are seen surrounding the white pulp and also infiltrating the red pulp. The neoplastic lymphocytes are B220 (blue) and BCL6 negative, but they express PAX5 and show IRF4 (variable) expression. (3713, H&E) LPD. The white pulp is variably disrupted by expansions of the follicular mantle and marginal zones with limited red pulp infiltration. The neoplastic cells express B220 (variable), PAX5 and IRF4 (variable), but they lack BCL6 expression.

tumor 3468 [Fig. 5-6B]. Cdkd2a (ink4a/p16) and Rb1 are also deleted in a subset of the MAD and MA lymphomas [Fig. 5-6B]. MLL2, a haplo-insufficient tumor suppressor gene in MCL as well as DLBCL, is also partially deleted in all tumors tested [Fig. 5-6B]. Meanwhile, Mdm2,

Cdk4, and Notch1 are moderately amplified (mostly trisome) in several MAD and MA lymphomas [Fig. 5-6B]. These changes are consistent with recurrent genetic alterations that have been characterized for human MCL (Jares, Colomer et al. 2007). Meanwhile, the CGH analyses also identified focal deletions in Ig (chromosome 4) and Igh (chromosome 12) loci indicative of B-cell origin. Fine mapping of the Igh locus showed that the deletions in all 3 MAD tumors are restricted to the V-D-J portion of the Igh locus and does not involve the switch region located downstream (centrometric of chr 12) [Fig. 5-6D], consistent with the pre-GC origin of the MAD tumors. In contrast, MA tumor 2023 shows heterozygous deletion within switch region [Fig. 5-6D]. Together with the positive staining for Bcl6, somatic hypermutation, this finding suggests that at least a subset of the MA tumors derived from GC or post-GC cells.

CGH is not able to detect reciprocal translocations. To test whether reciprocal translocation involving IgH locus occurred in any of the tumors, we performed chromosome 12 paint together with FISH using a probe that hybridizes to the telomeric end of Igh locus (upstream toward the

Vh region, 5’ IgH, 207) in two MAD (3468 and 4161) and one MA (2594) tumors, for which high quality metaphases are available. The analyses revealed heterozygous deletion, but not translocation involving the 5’IgH in MAD tumor 4161 and MA tumor 2594 [Fig. 5-6C]. FISH with c-Myc probes exclude IgH-Myc translocation in the tumors and CGH analyses suggested that c-

Myc is not grossly amplified in tested MAD or MA tumors. Finally targeted Sanger sequencing of

10 MAD and 4 MA tumors did not find mutations in the PEST domain of Notch1 or the SET domain

Figure 5-6: Comparative Genomic Hybridization (CGH) and FISH analyses of MA and MAD tumors A.) Landscape representation of CGH on MAD tumors-3468 and 4161, with green arrows indicating particular regions of interest. Chromosome (Ch.) numbers indicated at the top. The errors indicated the loci of interests (i.g. Ig, Igh and Trp53). The y-axis represents the log ratio of copy number between tumor vs normal at given probe. B.) Heatmap showing focal deletion of Trp53, Cdkn2a, MLL2 and Rb and potential trisomy of Mdm2 and Cdk4. It also shows that normal copy number of c-Myc. Different rows represent unique probes used for CGH analyses. C.) Chromosome (Ch.) 12 paint analyses combined with either Telomeric Igh FISH (BAC 207 at the Vh region) or c-Myc on MAD tumors-3468 and 4161. The diagram on the right represents the mono-allelic deletion of the Igh in tumor 4161. D.) CGH analysis of the Igh locus of MAD, MA, MD tumors. Demarcations of each region within the Igh locus are shown below. Average signal intensity of -0.6 indicates the value corresponding to heterozygous loss of the allele.

of MMSET, two genes that are mutated in ~10% human MCL (Bea, Valdes-Mas et al. 2013).

Together, these findings suggest that MAD tumors share molecular features with human MCL.

ATM deficiency enhances genomic instability and increases chromosomal fusions in

CyclinD1+ B cells.

To better understand the mechanism by which ectopic CyclinD1 expression synergizes with ATM deficiency to promote B-cell lymphoproliferations, we analyzed non-neoplastic B-cells from 6-12 week old MAD and MA mice. MCL is thought to originate from pre-GC, mantle zone

B-cells that have yet to encounter CSR and SHM (Walsh, Thorselius et al. 2003). As ATM deficiency reduces CSR efficiency (Lumsden, McCarty et al. 2004, Reina-San-Martin, Chen et al. 2004), we hypothesized that ectopic expression of CyclinD1 and ATM deficiency together might be deleterious to cells undergoing CSR, thus effectively trapping naïve B-cells at the pre-

GC stage of development/maturation. To test this hypothesis, we measured CSR in purified B- cells derived from MA and MAD mice as well as WT or Atm-/- control mice in vitro. Ectopic expression of CyclinD1 by itself did not measurably affect CSR. In vitro CSR to IgG1, measured by flow cytometry analyses, was reduced to ~50% of WT levels in MAD as well as MA and Atm-/-

Figure 5-7: Cell Cycle Analyses of stimulated B-cells from MA, MAD and MD/D mice. LPS/IL4 activated B-cells (Day 3) were pulsed in BrdU for 30 min without IR or 12 or 24hr after IR. The percentage of BrdU positive cells were plotted in A.) and the ratio of BrdU+ cells post-IR/before IR were quantified below. The percentage of mitotic cells (indicated by phosphorylated Ser10 of Histone H3 (pH3) was quantified before and 2 or 12 hr after IR (10Gy) in B.).

B-cells [Fig. 5-8A], suggesting ATM deficiency compromised CSR in WT and CyclinD1+ B cells at similar levels. Furthermore, irradiation induced G1/S and G2/M checkpoints in LPS/IL4

activated B-cells are normal in WT and CyclinD1+ B-cells and comparable in Atm-/-, MA and

MAD B cells [Fig. 5-7A, 5-7B], suggesting that ATM deficiency compromised DNA damage induced checkpoints in both WT and CyclinD1+ B-cells [Fig. 5-7A, B]. Next, we assessed the degree of genomic instability in LPS/IL4 activated B-cells from Atm-/-, MA and MAD mice by

Telomere-FISH (T-FISH) analyses. Two kinds of breaks can be quantified by the T-FISH assay: chromosomal breaks, which refer to breaks at both sister-chromatids, and chromatid breaks, which refer to breaks in only one of the two sister-chromatids [Fig. 5-8B] The frequencies of chromosomal breaks as well as chromatid breaks were comparable in MA and MAD mice and significantly higher than those observed in MD mice [Fig. 5-8B]. Moreover, the frequency of dicentric chromosomes generated by end-end fusion of chromosomal breaks was significantly higher in activated B-cells derived from MAD mice than in those from the MA or MD mice [Fig.

5-8C] suggesting possible increased proclivity of chromosomal translocations. We conclude that

ATM deficiency induces genomic instability in CyclinD1+ B cells.

Ectopic CyclinD1 expression increases naïve B-cell proliferation and rescues the B-cell lymphocytopenia in older MA mice.

Given the majority of the MAD and MA lymphomas are IgM+ and thus derived from cells that have not yet undergone CSR, we compared the naïve B-cell population in young (1-3 month old) vs mid-age (4-6 and 7-12 month old) MAD and MA mice. While all mature B-cell population are present in 6 and 12-month old MAD, MA and MD/D controls, the frequency of naïve B-cells (CD19+ cells) in spleen declined significantly faster in MA mice, recapitulating the

ATM deficient B-cell lymphocytopenia that worsens over time [Fig. 5-9A, B]. Ectopic expression of CylinD1 alone has, at most, moderate effects on splenic B-cell frequency in WT mice (MD vs control), but markedly rescued the splenic B-cell lymphocytopenia in MA mice (MAD vs MA,

Figure 5-8: Increased chromosomal fusions in stimulated MAD B-cells A.) Representative flow cytometry plots for surface IgG1 expression in LPS/IL4 stimulated B-cells and quantification. Efficiency is calculated as a percentage of IgG1+ cells relative to the control. B.) Representative images of cytogenetic abnormalities observed in T-FISH assay in stimulated splenic B- cells and the quantification. C.) Quantification of dicentric chromosomes observed stimulated B-cells derived from MD/D, MA, MAD, and Atm-/- mice. P-values were calculated using a two-tailed Student’s t- test assuming unequal variances.

[Fig. 5-9A, B]. This is intriguing, since the naïve B-cell population is thought to be the progenitor of MCL lymphomas. We then co-stained splenic sections from 6 month old tumor-free mice with

B-cell marker-B220 and proliferation marker-Ki67. Quantification of Ki67+B220+ cells revealed a significant reduction of Ki67+ frequency in B-cells from MA mice, which is rescued by ectopic expression of CyclinD1 in the MAD mice [Fig. 5-9C, D]. We conclude that

Figure 5-9: CyclinD1 expression rescues progressive B-cell loss in MA mice A.) Representative flow cytometry plots of splenic B220+CD19+/IgM+ B-cell populations in 6-month old mice. B.) Quantification of CD19+ B-cell populations from non tumor-bearing mice analyzed at the indicated time points. C.) Representative immunofluorescence (IF) images, and D.) Quantification of Ki67+ (green, nuclear staining) and B220+ (red, membrane staining) cells in spleens harvested from 7- month old non-tumor mice. Fields [MD (n=7), MA (n=15), and MAD (n=14)] selected for quantification were taken at 400x total magnification with equivalent B220+ cellularity. P-values were calculated based on individual two-tailed Student’s t-tests assuming unequal variances

ectopic expression of CyclinD1 promotes proliferation of ATM-deficient naïve B-cells with genomic instability, rescue the B-cell lymphocytopenia, and promote lymphomagenesis in pre-

GC B cells that would otherwise diminish due to ATM-deficiency.

Discussion

Given the mature B-cell phenotypes of most ATM–deficient human B-cell lymphomas and the well-characterized role of ATM during CSR, it was speculated that ATM maintains genomic stability in GC-B cells to suppress B cell lymphomas. However, despite complete ATM deletion in naïve B cells, our attempt to establish an ATM-deficient mature B-cell lymphoma model with naïve B-cell specific CD21Cre or less robust pre-B-cell specific CD19-Cre was fruitless. Mb1Cre allele is an early and robust B-cell specific Cre that leads to Cre activation in almost all pre-B- cells (in contrast to ~60% of pre-B-cells via CD19Cre). In this regard, 16.7% of MA and ~55% of

MAD mice developed indolent yet clonal, mature B-cell lymphomas [Fig. 5-4B] recapitulating the disease spectrum of human lymphomas, especially MCL. Even so, most of the MA and MAD tumors are IgM+ and thus have not undergone CSR. Fine mapping of the Igh loci with CGH probe and FISH analyses further revealed V(D)J recombination-related, but not CSR-related chromosomal alterations in MAD tumors. Together these findings highlights the critical tumor suppressor function of ATM in early progenitor or naïve B cells, possibly during V(D)J recombination, to prevent mature B cell lymphomas down the line. In this context, ATM is known to prevent the persistence and propagation of chromosome breaks originating from

V(D)J recombination in naïve B-cells(Callen, Jankovic et al. 2007). Recent studies reveal that early genetic alterations in hematopoietic stem cells could prime mature B-cell malignancies, including CLL (Kikushige, Ishikawa et al. 2011).

Despite strong evidence for a role of CyclinD1 lymphomagenesis, mice with ectopic expression of WT CyclinD1 (MD/D) rarely developed lymphomas. Genomic analyses of human

MCL suggest that ATM deletion coexist with CyclinD1 expression in nearly 50% of MCLs (Bea,

Valdes-Mas et al. 2013). Here we show that ATM deletion promotes B-cell lymphomagenesis in

EµCyclinD1 mice (Bodrug, Warner et al. 1994, Vaites, Lian et al. 2014). While ATM deficiency and the resulting genomic instability might simply promote IgH-CyclinD1 translocation in human

MCL, our study suggests that loss of ATM also promotes lymphomagenesis after ectopic expression of CyclinD1. In this context, we showed that ATM deficiency increases genomic instability and chromosomal translocations in activated CyclinD1+ B-cells [Fig. 5-8A]. This increased genomic instability in naïve B cells likely promotes the loss of additional tumor suppressor genes including Trp53, Cdkn2a, MLL2 or Rb1 that have all been implicated in human MCL to promote transformation(Bea, Valdes-Mas et al. 2013). In addition, ATM deficiency might potentiate the mitogenic function of CyclinD1 by increasing CyclinD1 protein levels. ATM negatively regulates F-Box proteins (FBX4 and FXBO31) that mediate the degradation of CyclinD1 during DNA damage responses (Santra, Wajapeyee et al. 2009,

Vaites, Lee et al. 2011). Indeed, germ line ATM deficiency accelerates lymphomas in mouse models expressing a constitutive-nuclear form of CyclinD1 (Gladden, Woolery et al. 2006,

Vaites, Lian et al. 2013).

MAD mice developed B-cell lymphomas significantly earlier and more frequently than

MA mice, suggesting that mitogenic cues generated by ectopic CyclinD1 expression synergize with ATM deficiency for B-cell transformation. In addition to its checkpoint function, ATM has important roles in DNA repair, as ATM deficiency is detrimental to primary cells due to ongoing genomic instability. Here we show that B-cell-specific deletion of ATM in MA mice leads to progressive B-cell lymphocytopenia at 6- and 12- months of age [Fig. 5-9A, B], a phenotype that is overlooked in germ-line ATM null mice due to lethal thymic lymphomas. We further showed

that EµCyclinD1 promotes the proliferation of naïve B-cells and successfully restores B-cell cellularity in older MA mice [Fig. 5-9A-D]. Based on this finding, we propose that CyclinD1 promotes B cell lymphomas by promoting proliferation of ATM naïve B cells with genomic instability, which in turn maintain the cellularity of tumor-prone ATM-deficient B-cells. There are three D-type cyclins (D1, D2 and D3) with overlapping functions during embryonic development

(Carthon, Neumann et al. 2005, Ciemerych and Sicinski 2005), but GC B-cells exclusively depend on CyclinD3 for proliferation and survival (Peled, Yu et al. 2010, Cato, Chintalapati et al.

2011), which might explain why the pre-GC cells are more susceptible to the mitogenic cues unleashed by ectopic CyclinD1.

In summary, we report the first animal model for ATM-deficient B-cell lymphomas and reveal a synergistic role of CyclinD1 expression and ATM deficiency in naïve B-cells to promote pre-GC B-cell lymphomas. The MA and MAD mature B-cell lymphomas are largely indolent, similar to the Bcl6-driven mouse lymphomas (Cattoretti, Pasqualucci et al. 2005) and distinct from Myc driven aggressive lymphomas. In this regard, MA and MAD lymphomas recapitulate the disease spectrum of human lymphomas, especially MCL (Zullo, Amengual et al. 2012), and would be an invaluable resource to better understand MCL and the clinical progression of indolent MCL to an aggressive form in the future.

IMPLICATIONS, FUTURE DIRECTIONS

Novel kinase-structure roles of ATM in DNA repair, lymphomagenesis

Much of the understanding of ATM function to date has been on its signaling function, while its structural functions, if any, remained elusive. To address this, we generated mice expressing a kinase-dead ATM protein initially to test whether catalytically inactive ATM would rescue any of the pleiotropic phenotypes associated with complete loss of ATM in DNA repair and lymphomagenesis. Strikingly, while ATM-null mice are viable, mice solely expressing KD

ATM are embryonic lethal with severe genomic instability beyond that seen in ATM null mice, suggesting the presence of the catalytically inactive ATM inhibits additional DNA repair pathways. In this context, we found that the catalytically inactive ATM protein, but not the lack of

ATM, suppresses the replication dependent strand cleavage necessary for Top1cc removal, which manifests to increased chromatid breaks, accumulation of Top1cc and hypersensitivity to

CPT as well as inhibitors for PARP that is implicated in the repair of Top1cc lesions. Strikingly, expressing ATM KD is more oncogenic than loss of ATM in mouse lymphoma models, at least in part due to frequent deletion of the Pten tumor suppressor gene in the progenitors. Based on these findings, we propose that ATMKD/- cells and, by extension ATMKD/- lymphomas, might be preferentially targeted by Topo1 inhibitors. The generation and characterization of the mouse model expressing catalytically inactivate ATM described here for the first time, revealed a phosphorylation-dependent structural function of ATM upstream of DSBs that have several implications in ATM function beyond its roles in DDR, and tumor suppression.

Several studies have identified several functions of ATM outside its role in classical DNA

DSB repair, including ROS processing, insulin signaling, hypotonic stress response, cell migration and metastasis, and cellular metabolism (Yang and Kastan 2000, Eaton, Lin et al.

2007, Guo, Deshpande et al. 2010, Guo, Kozlov et al. 2010, Shiloh and Ziv 2013, Chen, Ebelt et

al. 2015). While Cys2991 of ATM has been implicated in its ROS response (Guo, Kozlov et al.

2010), it is largely unknown whether these “non-canonical” functions of ATM only require its kinase activity, whether the AtmKD/- cells retains (or loses) these functions. AtmKD/- cells and mice described here provide a platform to answer these questions, and will contribute to further understanding the ever-expanding roles of ATM beyond the DNA damage response. ATM-null thymocytes accumulate ROS at greater levels compared to WT thymocytes and antioxidants delay the onset of thymic lymphomas in Atm-null mice via treatment with (Schubert, Erker et al.

2004, Valentin-Vega, Maclean et al. 2012, D'Souza, Parish et al. 2013). The effect of ATM KD in ROS processing or other non-canonical function of ATM are yet to be determined, and might contribute to the increased oncogenic potential associated with Atm KD mutation.

Given the prevalence of ATM missense mutations in human cancer and the hypersensitivity of AtmKD/- cells to CPT, these results suggest that we might be able to target

ATM with Topo1 inhibitors- a class of drugs that is already used in cancer therapy. In this precision medicine era, the identification of ATM mutations in tumors can serve as a marker to tailor therapeutic regimens to include Topo1 inhibitors in treatment. We are currently testing this using a murine models of activated Notch-driven AtmKD/-, AtmC/+ vs Atm-/- T cell leukemia models. Specifically, we had induced immature T cell leukemia by infecting bone marrows from

Rosa+/ERCRE AtmC/+, AtmC/C, or AtmC/KD mice with retrovirus encoding the oncogenic form of

Notch1 (ΔE-NOTCH1)(Pear, Aster et al. 1996, Schroeter, Kisslinger et al. 1998). We then transplanted the established leukemia to secondary recipients and induced Cre activation via

Tamoxifen to generate isogenic Atm+/-, Atm-/- and Atm-/KD leukemias, respectively. Once the deletion is confirmed and the clonality of leukemia is established, we will perform tertiary transplantation to new recipients with isogenic AtmC/KD vs Atm-/KD or Atm+/c vs Atm+/- leukemias and subject the recipient mice to Irinotecan- a clinically used Topoisomerase-I inhibitor. The results from these ongoing experiments will provide important preclinical data to determine whether Atm-/KD leukemia blasts could be effectively targeted by Topo1 inhibitor in vivo.

Furthermore, these ongoing experiments will uncover potential complications and information, including dosage, side effects, and in vivo efficacy of the therapy.

Mutational landscape of ATM in human tumors- prognostic and therapeutic significance

The increased oncognic potential of AtmKD/- allele and the prevalence of kinase domain missense mutation in human cancers we found here revealed a feature of highly oncogenic

ATM mutations – defined by the expression of KD ATM, and a potential indicator for Topo1 inhibitor-based therapy. But the requirement for MRE11 protein, but not its nuclease activity in how ATM KD protein suppress CPT induced DSBs, also suggest the potential genetic modifiers that might alter the phenotype of ATM KD alleles. We recently identified a hotspot mutation at

Arg3008 to R3008H or R3008C in the FATC domain. Structural analyses put the R3008 on the surface of ATM and at the end of an alpha helix that is likely important for helix stability [Fig. 1-

1]. Preliminary characterization of the mice generated with knock-in Atm R3008H (R3018H in mice, R allele) mutation confirmed that AtmR/R MEFs express near wildtype levels of ATM protein, but fail to phosphorylate KAP1 upon IR treatment, as indicated in previous studies on human lymphomas with R3008 mutations [Fig. 3-1C, 0-1] (Camacho, Hernandez et al. 2002,

Angele, Lauge et al. 2003, Austen, Barone et al. 2008). While Atm+/R mice, like the Atm+/KD mice, do not display any discernable dominant-negative phenotypes (Yamamoto, Wang et al. 2012), both AtmR/R and AtmR/- mice are small (~50%), yet born alive, in contrast to the early embryonic lethality of AtmKD/KD(KD/-) mice (Yamamoto, Wang et al. 2012). Given the R3008 residue is located outside of the catalytic core of ATM [Fig. 1-1, Fig. 4-6C, residue labeled in red], one possibility is that ATM R3008H is still a viable kinase, but the ATM R3008H cannot be activated by DSBs since the R3008H mutation abrogates its ability to interact with MRE11-NBS-RAD50 complex. Consistent with this hypothesis, the FATC domain of TEL1- the ATM homolog in

Figure 0-1: ATM R3008H (R) retains protein expression and lacks kinase activity Western blot of immortalized MEFs with the indicated genotypes with or without IR. Experiment testing two independent clones of AtmR/R and AtmR/+ MEF lines are shown.

yeast, is required for ATM recruitment to the damage site and for its activation by DNA damage

(Ogi, Goto et al. 2015). While additional work is necessary to further elucidate the consequence of the R3008H mutation in DNA repair, tumorigenesis, and treatment responses both in vitro and in vivo, these preliminary results further extend the complexity associated with ATM mutations and emphasize the importance of developing a functional-based system to classify cancer associated ATM mutations.

Functions of ATM in the suppression of B-cell lymphomas

Both germline and somatic mutation of ATM were found in human B cell lymphomas, yet several previous efforts of generating ATM deficient B cell lymphomas models have been fruitless. While Atm deletion using CD19-cre, which begins to express at the Pre-B stage, did not result in the development of B-cell lymphomas, deletion of Atm using Mb1cre, expressed at the early Pro-B cell stage, resulted in the development of clonal, mature B-cell lymphomas in

~50% of the mice, providing a functional validation for the need of early deletion of ATM during

B-cell lymphomagenesis. The ATM-deficient models described in this study recapitulate the

indolent forms of lymphomas seen in the clinic, thus serves as a suitable model to study early stage B-cell lymphomas. The synergistic effect of ATM deletion and CyclinD1 overexpression in driving pre-GC mature B-cell lymphomas, which share histopathological features with human

MCL, further provides a faithful animal model for MCL- a disease for which effective therapy is not yet available (Herrmann, Hoster et al. 2009). We further showed that this synergy is associated with increasing genomic instability and expansion of the proliferative naïve B-cell pool in the spleen, providing a mechanism for how ATM loss and CyclinD1 work together in

MCL. Missense mutation in the kinase domain of ATM is also found in MCL, [Fig. 4-5A]

(Stankovic, Stewart et al. 2002) suggesting that ATM KD may increase the overall incidence and accelerate the kinetics of B-cell lymphomas compared to the loss of ATM. Furthermore,

ATM KD-bearing B-cell lymphomas can potentially be targeted by Topo1 inhibitors for therapy.

We are currently developing a mouse model utilizing the Mb1cre to conditionally express ATM

KD in B-cells. If the ATM KD leads to earlier and increased development of B-cell lymphomas, this will provide further insight into the nature of the ATM KD mutation in promoting lymphomagenesis in the context of B-cells. Additionally, the mice generated in this study will serve as an invaluable resource to identify novel therapeutic interventions tailored for ATM mutant B-cell lymphomas and MCLs in the future.

MATERIALS AND METHODS

CHAPTER 2

Generation of ATM KD, DKDtg allele

The DNA sequence for homologous targeting at the Atm locus was generated via PCR from TC1 ES cell DNA (129 strain). A targeting construct was designed to insert a NeoR gene cassette oriented in the opposite transcriptional direction from the endogenous Atm promoter into intron 57 next to the D2880A/N2885K mutation in exon 58. A 3.5 kp 5’ arm and 5.1 kb 3’arm were PCR generated separately, cloned into pBK vector and sequenced. The 5’arm was directly subcloned into pLNTK in the desired orientation. The mutation was introduced into the 3’ arm using site-direct mutagenesis and confirmed by sequencing. The mutated 3’ was then subcloned into pLNTK. The targeting construct was then electroporated into CSL3 ES cells (129 strain) and successful targeting was determined via Southern blot analyses using EcoRV digested genomic DNA and 5’ genomic probe as outlined in [Fig.2-1B]. The WT band is ~22kb and the targeting introduced an additional EcoRV site and reduced the band to ~4.7kb. The corrected clones were confirmed with 3’ probe with EcoRV digestions. Six independent targeted clones were sequenced for the ATM KD mutation, 4 were verified with correct mutations. Two were injected for germ line transmission.

The DKD allele was generated by introducing S1987D and D2880A/N2885K mutations into BAC RP24-122F10, which contains Atm and genomic regions 18kb downstream and 50kb upstream of the stop and start codons, respectively, using a “hit and fix” BAC recombineering technique previously described (Yang and Sharan 2003). Linearized BAC containing the mutations were injected into the pronuclei of fertilized embryos and into a surrogate. Chimers were screened for the presence of transgene as previously described (Pellegrini, Celeste et al.

2006). Mice were crossed to Atm+/- for germline transmission and generation of DKD Atm-/- mice.

Mice

AtmC/C and Rosa+/ER-Cre mice were previously characterized (de Luca, Kowalski et al.

2005, Guo, McMinn et al. 2007, Zha, Sekiguchi et al. 2008, Zha, Jiang et al. 2011). To activate

Cre-recombination, Rosa+/ER-Cre mice were orally fed with Tamoxifen (5 mg/100µl with 10% ethanol in Sunflower oil per mouse) daily for two consecutive days. The mice were analyzed 10-

14 days after the 2nd treatment. All animal experiments were conducted in pathogen free facility and approved by the Institute Animal Care and Use Committee of Columbia University.

Derivation of Embryonic Stem Cells from Mice

3-4 week old female Atm+/KDN mice were super-ovulated by intraperitoneal injection of

5.0 IU pregnant mare serum gonadotropin (PMS), followed by intraperitoneal injection of 5.0 IU hCG (Sigma) 46 to 48 hours later. The females were set up with adult male Atm C/C mice (1:1 ratio) immediately after the hCG injection. Four days after the hCG injection, the mated females were sacrificed and mature blastocysts were flushed out of the uterus and plated one per well in

96 well plates containing irradiated feeders. After 9-14 days incubation, the cells were expanded to 24 well plates, at which stage, half of the cells were frozen down and the rest of the cells were used to derive DNA for genotyping. Atm KDN/C ES cells were then identified by PCR.

Cytogenetic analysis for genomic instability

ES cells, not grown on feeders, were plated 24 hrs before the addition of colcemid

(KaryoMAX Colcemid Solution, GIBCO) at 100 ng/ml for 1-2 hours. ES cells were harvested by trypsinization and metaphases were prepared as described (18, 21). T-FISH was performed as

described (32). Briefly, telomeres were stained with a Cy3-labled (CCCTAA)3 peptide nucleic acid probe (PNA, Biosynthesis Inc.) and DNA was counter-stained with DAPI-containing mounting media (Vectashield®). Images were acquired using Nikon Eclipse 80i with remote focus accessory and with Photometrics® CoolSNAP™ HQ2 camera unit with the Plan Fluor

Nikon Lens (60x and 100x/1.30 Oil, Japan) in room temperature. All images were processed with NIS-Elements AR (Ver. 3.10). Chromosome breaks were defined by loss of telomere signal from both sister-chromatids. Chromatid breaks were defined by loss of telomere signal from one of the two sister-chromatids or a clearly broken DAPI signal in the middle of one chromatid. We note that chromosomal gaps could not be reliably identified with the T-FISH assay; therefore, it is possible that the actual frequency of chromatid or chromosome breaks might be even higher than estimated.

Lymphocyte Development and Class Switch Recombination

Lymphocyte populations were analyzed by flow cytometry as described (Li, Alt et al.

2008). Isolation and activation of splenic B cells and flow cytometric assays were performed as described (Li, Alt et al. 2008). Briefly, CD43- splenic B cells were purified from total spleen using anti-mouse CD42 MACS bead following the manufacturer’s instructions (Miltenyi). To stimulate the cells for CSR, the cells were incubated with αCD40 plus IL-4 for 4 days and the percentage of CSR to IgG1 were determined by surface staining and FACS analysis (Li, Alt et al. 2008). To assay for genomic instability, Colcemid was added to a fraction of day4 stimulated B cells for 4 hour and the metaphases were prepared and stained for T-FISH as described for ES cells.

CHAPTER 3

Mouse Embryonic Fibroblasts (MEF)

MEFs were harvested at embryonic day 14.5 (E14.5) based on timed breeding and immortalized using retrovirus expression the large and small SV40 T antigen. Immortalized culture was established after three passages (1:3 dilutions) after the primary infection. To induce 4-hydroxytamoxifen (4OHT) dependent nuclear translocation of ER-Cre recombination

[Fig. 3-1] the cells were plated at low density (~30% confluence), passed three times (48h interval) in the presence of 4OHT (200nM). Complete deletion of the conditional allele was confirmed using PCR as described previously (Yamamoto, Wang et al. 2012). For drug sensitivity assays, the immortalized MEF treated for 4OHT (3 rounds) were seeded in gelatinized 96-well plates (6x103/well). Twenty-four hours after the initial seeding, the cells were treated with the compounds at indicated concentration for 48 hours. The cell number was quantified using CyQuant (Molecular Probes), a nucleotide staining, per the manufacturer’s instructions. The relative survival was calculated based on the cell number in the untreated wells.

Immunofluorescence and quantification of repair protein foci

Immunofluorescence was performed on 4% formaldehyde/PBS fixed, 1% Triton-X/PBS treated cells using the following antibodies: Rad51 (Clone PC130, 1:200, Calbiochem) phospho-

RPA pT21 (Cat. # ab109394, 1:500, Cell Signaling) γH2AX (Cat. #07-164, 1:500, EMD

Millipore), CyclinA2 (Clone CY-A1, Sigma-Aldrich). Slides were scanned on the Carl Zeiss Axio

Imager Z2 equipped with a CoolCube1 camera (Carl Zeiss, Thornwood, NY, USA). Metafer 4 software (MetaSystems, Newton, MA, USA) was used for automated quantification of Rad51, pRPA, γH2AX foci. More than 500 cells were quantified for every experimental replicate. Images were exported and processed on ISIS software (MetaSystems).

Western and Southern Blotting

Following antibodies were used for Western blot on whole cell lysate: ATM (clone MAT3,

1:5000), β-Actin (clone A5316, 1:20,000) and Vinculin (clone V284, 1:10,000) from Sigma-

Aldrich; phospho-Kap1 (Cat. #A300-767A, 1:1000) and Top-I (Cat. # A302-590, 1:1000) from

Bethyl laboratory; phospho-RPA pT21 (Cat. # ab109394, 1:5000) from Abcam; RPA (clone

RPA30, 1:5000), γH2AX (Cat. #07-164, 1:1000) and H2AX (Cat. #07-627) from EMD Millipore; phospho-Chk1 (clone 133D3, 1:500), Mre11 (Cat. #4895S, 1:1000), Kap1/TIF1β (clone

C42G12, 1:1000), and Pten (clone D4.3 XP, 1:1000) from Cell Signaling and PCNA (clone

PC10) from Santa Cruz. To confirm deletion of the ATM conditional allele in VKD and VN mouse models, Southern blot was performed on KpnI digested genomic DNA and probed with using the ATMCKO 3’ probe(Zha, Sekiguchi et al. 2008). For clonal analyses of the thymic lymphomas or the B cell lymphomas, Southern blot was performed on EcoRI-digested genomic

DNA, and blotted with Jβ1.6 Jβ2.7 probes (Khor and Sleckman 2005, Zha, Sekiguchi et al.

2008) or Jh4 probes or myc-A probe (Gostissa, Yan et al. 2009).

Sister Chromatid Exchange Assay

Cells were incubated for two doubling times (48hrs) with Bromodeoxyuridine (BrdU, Sigma-

Aldrich, 5µg/ml). During the second doubling period (the 2nd 24hr), the cells were also incubated with either vehicle (DMSO) or CPT (1µM) BrdU-incorporated sister chromatids were quenched using Hoechst33258 (50µg/ml) treatment followed by UV exposure, then stained with DAPI

(Vectashield). Slides were scanned for metaphase spreads using the Carl Zeiss Axio Imager Z2 equipped with a CoolCube1 camera (Carl Zeiss, Thornwood, NY, USA) and Metafer 4 software

(MetaSystems, Newton, MA, USA). Spreads were quantified and images exported via ISIS

software (MetaSystems). Over 40 metaphases were quantified for each genotype. A total of three independent replicates were performed.

COMET Assay

Neutral and alkaline COMET assays were performed per manufacturer’s protocol

(Trevigen). Slides were scanned for COMETs using Metafer 4 and tail lengths quantified using

ISIS.

In-vitro Complex of Enzymes (ICE) assay

Covalent complexes of TopI-DNA were detected as previously described (Nitiss 2001).

Briefly, cells were gently lysed in 1% Sarkosyl/TE solution, ultracentrifuged through a CsCl2 gradient (6M) and the resulting genomic DNA was dissolved in TE. Equivalent amounts of DNA were loaded on 0.45µm Nitrocellulose membranes, and probed for Top-1 (Cat. # A302-590,

1:1000, Bethyl). Loading controls were probed with radiolabeled total mouse genomic DNA.

DNA fiber assay

Cells were first incubated with 25µM IdU (30 min) then with additional 250µM CldU

(Sigma), trypsinized and harvested in ice-cold 1xPBS, lysed using 1% SDS/Tris-EDTA and stretched along glass coverslips. Slides were stained using primary Anti-IdU (B44, BD

Biosciences), CldU (BU1/75 (ICR1), Abcam) antibodies and corresponding secondary anti-Rat

Alexa594, and anti-mouse Alexa488 (Molecular Probes) antibodies. DNA fibers were analyzed on the Nikon Eclipse 80i microscope equipped with remote focus accessory and CoolSNAP HQ camera unit using the 60x/1.30 NA oil Plan Fluor lens. All images were processed and

quantified with NIS-Elements AR. Length of fibers were calculated with a base pair length of

3.4Å (340pm).

CHAPTER 4

Large Cohort Patient Collection

ATM somatic mutations were collected from 5,402 cancer patients from 24 tumor types of TCGA, including ACC, BRCA, CHOL, COADREAD, ESCA, HNSC, KIRC, LAML, LIHC,

MESO, PAAD, PRAD, SARC, STAD, THYM, UCS, BLCA, CESC, COAD, DLBC, GBM, KICH,

KIRP, LGG, LUAD, OV, PCPG, READ, SKCM, TGCT, UCEC, and UVM. All maf files of those tumor types were downloaded using firehose (http://gdac.broadinstitute.org). To unify the annotation format, we firstly used liftover tool (https://genome.ucsc.edu/cgi-bin/hgLiftOver) to map all coordinates to hg19, and then re-annotated variant effect with VEP

(http://www.ensembl.org/info/docs/tools/vep/) under transcript NM_000051.3. Mutation types

Missense_Mutation, In_Frame_Del, and In_Frame_Ins were classified as missense, while

Nonsense_Mutation, Frame_Shift_Del, Frame_Shift_Ins, and Splice_Site were classified as truncating mutations.

Germline ATM mutations are from A-T patients in the Leiden open variable database LOVD2

(http://chromium.lovd.nl/LOVD2/home.php?select_db=ATM). Protein mutation position was curated annotated based on the mutation effect provide by LOVD2. Start_Codon,

Frame_Shift_Del, Large_DEL, Nonsense_Mutation, Frame_Shift_Ins, Splice_Site, and

Stop_Codon are labeled as Truncating in our analysis.

Mutation Density Estimation

To measure the mutation density of different ATM positions, we applied a Gaussian kernel smoother to smooth the number of mutation of each amino acid site. If we use 푦(푥푖)(푖 =

1,2, ⋯ , 푛) to represents the number of mutations in each site 푥푖, then the Gaussian kernel smoother of 푦(푥) is defined by

푛 ∑푗=1 퐾(푥푖, 푥푗)푦(푥푗) 푦̂(푥푖) = 푛 ∑푗=1 퐾(푥푖, 푥푗)

(푥 −푥 )2 where 퐾(푥 , 푥 ) = exp⁡(− 푖 푗 ) is the Gaussian kernel, and 푏 indicates the window size (we 푖 푗 2푏2 use 80 in the current study). Missense mutations and truncating mutations were separately considered.

To estimate the expected number of mutations in ATM, we assume the ratio between silent and non-silent mutation (e.g. missense mutation) is constant, and then the frequency, as well as its confident interval, to observe a missense mutation can be estimated by fitting a binomial distribution with 푛 trials. The shadow curve was the 95% confidence interval of fold change between observation and expectation after a Bonferroni Correction.

Structure Simulation Analyses

Homology model of human ATM was generated based on the crystal structure of human mTOR(PDB 4JSP)(Yang, Rudge et al. 2013) using the ROBETTA server, setting default parameters(Kim, Chivian et al. 2004). Briefly, the amino acids 2520-3056 of human ATM was submitted to the ROBETTA server and the mTOR crystal structure was identified as template.

After align the query sequence with the parent structure, a template was generated and the variable regions were modelled in the context of the fixed template using Rosetta fragment assembly. Finally, the model was subjected to several rounds of optimization by Rosetta’s relax protocol. All structure analyses and visualizations were performed with pymol (http://pymol.org/).

Mice and lymphocyte analyses

All the mouse alleles used (AtmC, AtmKD, VavCre+, RosaERCre) have been previously described (de Boer, Williams et al. 2003, de Luca, Kowalski et al. 2005, Zha, Sekiguchi et al.

2008, Yamamoto, Wang et al. 2012),. All the animal work were approved by and performed according to the regulation of Institutional Animal Care and Use Committee (IACUC) of

Columbia University. For development analyses, VN, VKD and littermate matched control mice were euthanized at 4-6 weeks of age. Total lymphocyte counts, flow cytometry analyses and in vitro class switch recombination were performed as described previously (Li, Alt et al. 2008).

Briefly, splenic B and T-cells were isolated using Magnetic-Activated Cell Sorting (MACS) for

CD43- and CD43+ fractions, respectively, following the manufacturer’s protocol (Miltenyi Biotec).

B-cells (CD43-) were stimulated with LPS (20µg/ml) and IL-4 (20ng/ml) for 4.5 days at a cellular concentration no more than 1x106cells/ml, and the T-cells (CD43+) activated by ConA (2µg/ml) for 3.5 days (Li, Alt et al. 2008). Metaphases were collected at the end of the cytokine stimulation period after 4 hours of colcemid (100ng/ml KaryoMax, Gibco) treatment, stained with

Telomere FISH probes as previously described (Franco, Gostissa et al. 2006, Yamamoto, Wang et al. 2012). Lymphoma- bearing animals were identified based on enlarged lymph nodes (B cell lymphomas) or severely hunched postures (thymic lymphomas) and analyzed.

CHAPTER 5

Mice

ATMC/C (Zha, Sekiguchi et al. 2008), EµCyclinD1 transgenic (Bodrug, Warner et al.

1994), CD21Cre (Kraus, Alimzhanov et al. 2004), CD19Cre (Rickert, Rajewsky et al. 1995) and

Mb1Cre (Hobeika, Thiemann et al. 2006) mice were previously described. All experimental cohorts carry transgene (CD21Cre, EµCyclinD1) or Cre knock-in (Mb1Cre or CD19Cre) in

heterozygosity. All animal experiments were performed in accordance with Columbia University

Institutional Animal Care and Use Committee.

Flow cytometry and Immunohistochemistry analyses

Flow cytometry analyses were performed on single-cells suspensions from spleen, bone marrow, lymph nodes as previously described (Yamamoto, Wang et al. 2012). Antibodies include: CD43(S7), CD5(53-7.3), CD138(281.2), CD21/35(7G6), CD23(B3B4), CD11b/Mac-

1(M1/70) and IgK from BD Pharmingen (San Diego, CA); B220(RA3-6B2), CD19(eBio1D3),

CD8a(53-6.7), CD4(L3T4), CD3ε(145-2C11) and TCRβ(H57-597) from eBioscience (San Diego,

CA); Ter119 and Gr1(RB6-8C3) from BioLegend (San Diego, CA); IgM and IgD from Southern

Biotech (Birmingham, AL) and PNA from Vector Labs (Burlingame, CA). Immunohistochemistry

(IHC) staining was performed on formalin fixed, paraffin-embedded tissue sections as described previously using the following antibodies: B220, CD3, PAX5, BCL6, IRF4 and CD138 (Klein, Lia et al. 2010). Immunofluorescence on paraffin-embedded tissues was performed using antibodies against Ki67 (SP6, Abcam, Cambridge, MA), and B220 (RA-3B62, BD Pharmingen).

Images were acquired with a Nikon Eclipse 80i microscope (Nikon, Melville, NY) and processed on NIS-Elements AR 3.10 software (Nikon, Melville, NY).

Southern Blot and Western Blot Analyses

For Southern blot analyses, genomic DNA from the spleen, tumor and corresponding kidneys was digested with EcoRI (for JH4, c-Myc, constant region of TCRδ probe) KpnI (for 3’

ATM CKO probe), or HindIII (for the IgK probe) (Zha, Sekiguchi et al. 2008, Gostissa, Yan et al.

2009, Zha, Bassing et al. 2010). To avoid cross-reactivity with the EµCyclinD1 transgene that includes a fragment from JH4 to µ enhancer, we generated a new JH4 probe by PCR

amplification with the following primers: 5’-TGGTGACAATTTCAGGGTCA-3’ and 5’-

TTGAGACCGAGGCTAGATGC-3’. For Western blots, stimulated splenic B or T-cells were probed with antibodies against ATM (1:2000, MAT3, Sigma, St. Louis, MO), pKap1 (1:2000,

Bethyl, Montgomery, TX), Kap1/TIF1β (1:1000, Cell Signaling, Danvers, MA), β-Actin (1:10,000,

A5316, Sigma) and CyclinD1 (1:1000, Abcam).

Class Switch Recombination and FISH analyses

Splenic B-cells from 8-12 week old mice were purified with anti-CD43 MACS beads and stimulated with LPS/IL4 for 2-4 days as described previously (Yamamoto, Wang et al. 2012).

Metaphases were obtained at day 4.5 after stimulation and stained with Cy3 conjugated PNA probe against 3xtelomere sequence as previously described (Franco, Gostissa et al. 2006).

Tumor metaphases were harvested after 6 hours incubation in RPMI1640 with15% Fetal Bovine

Serum in the presence of colcemid (100ng/ml, KaryoMax, Gibco). Chromosome 12 paint was obtained from Applied Spectral Imaging (Carlsbad, CA) and the 5’ c-myc (C10) and 5’ IgH

(BAC207) locus-specific FISH probes were described previously (Franco, Gostissa et al. 2006).

Images were acquired with Zeiss Axio Imager Z2 equipped with a CoolCube1 camera (Carl

Zeiss, Thornwood, NY) and an automatic stage system, and processed with Isis and Metafer4 software packages (MetaSystems, Newton, MA).

Mutation analyses

Somatic hypermutation status of the IgH variable region was determined as described

(Jolly, Klix et al. 1997). Briefly, degenerative forward primers upstream of J558FR3 (5’-

CAGCCTGAC ATCTGAGGACTCTGC-3’), VH7183 (5’-GAASAMCCTGTWCCTGCAAATGASC-

3’), or VHQ52 (5’-CAGGTGCAGCTGAARCAGTCA-3’) were paired with a reverse primer

downstream of JH4 (5’-CTCCACCAGACCTTCTAGACAGC-3’). Rearrangements were defined as clonal when a unique band was obtained from a given tumor in three independent PCR reactions generated from the same primer pair and characterized/verified by genomic sequencing. The results were aligned with normal mouse IgH locus sequences for 129S (NCBI accession number: AJ851868.3) and C57BL/6J (NC_000078.6) strains respectively to identify somatic mutations in the JH4 intron.

Targeted Sanger sequencing of the NOTCH1 (PEST domain) and WHSC1/MMSET

(SET domain) hotspots were performed on PCR amplified genomic DNA using the following primer sets: Notch1:5’- CCAGCCAGTACAACCCACTA-3’ and 5’

TTCCAGCTTCCCTTCTCCTG-3’; WHSC1: 5’-GGACCTGAGTTCTGTTCCCA-3’, 5’-

ACACCACAGTCTACAGCGAA-3’; 5’-AGCCCATTGTTCTAGCTGGT-3’, 5’-

GAAGCAATGCAACACCAGGA-3’. Following primers were used to sequence the respective product for better depth and coverage: Notch1: 5’-CCTCCTTCCCAGCACAGTTA-3’, WHSC1:

5’-AACGCGTTACCCTAGACTGA-3’ and 5’-CTGTCCTCACCAGTGCTCA-3’. Sequences were aligned with uc008ivl.2 (NOTCH1) and uc012duw.1 (WHSC1) from the Mouse Genome

Browser (http://genome.ucsc.edu).

Comparative Genomic Hybridization (CGH)

2µg genomic DNA from tumor and non-tumor control (kidney or tail) from the same mouse were labeled with Cy-5 and Cy-3, respectively, using the SureTag DNA Labeling Kit

(Agilent, Santa Clara, CA) per manufacturer’s instructions. Hybridization and scanning was performed on a 244k Mouse Genome CGH microarray platform (Agilent) at the Integrated

Genomics Operation (IGO) Center at the Memorial Sloan Kettering Cancer Center (New York,

NY). Data analysis was performed on Agilent Genomic Workbench (ver. 7.0.4.0) and Microsoft

Excel.

100

REFERENCE

Alagoz, M., S. C. Chiang, A. Sharma and S. F. El-Khamisy (2013). "ATM deficiency results in accumulation of DNA-topoisomerase I covalent intermediates in neural cells." PLoS One 8(4): e58239.

Andres, S. N., A. Vergnes, D. Ristic, C. Wyman, M. Modesti and M. Junop (2012). "A human XRCC4-XLF complex bridges DNA." Nucleic Acids Res 40(4): 1868-1878.

Angele, S., A. Lauge, M. Fernet, N. Moullan, P. Beauvais, J. Couturier, D. Stoppa-Lyonnet and J. Hall (2003). "Phenotypic cellular characterization of an ataxia telangiectasia patient carrying a causal homozygous missense mutation." Hum Mutat 21(2): 169-170.

Austen, B., G. Barone, A. Reiman, P. J. Byrd, C. Baker, J. Starczynski, M. C. Nobbs, R. P. Murphy, H. Enright, E. Chaila, J. Quinn, T. Stankovic, G. Pratt and A. M. Taylor (2008). "Pathogenic ATM mutations occur rarely in a subset of multiple myeloma patients." Br J Haematol 142(6): 925-933.

Bakkenist, C. J. and M. B. Kastan (2003). "DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation." Nature 421(6922): 499-506.

Bakr, A., C. Oing, S. Kocher, K. Borgmann, I. Dornreiter, C. Petersen, E. Dikomey and W. Y. Mansour (2015). "Involvement of ATM in homologous recombination after end resection and RAD51 nucleofilament formation." Nucleic Acids Res 43(6): 3154-3166.

Barlow, C., S. Hirotsune, R. Paylor, M. Liyanage, M. Eckhaus, F. Collins, Y. Shiloh, J. N. Crawley, T. Ried, D. Tagle and A. Wynshaw-Boris (1996). "Atm-deficient mice: a paradigm of ataxia telangiectasia." Cell 86(1): 159-171.

Barlow, J. H. and R. Rothstein (2010). "Timing is everything: cell cycle control of Rad52." Cell Division 5.

Bea, S., R. Valdes-Mas, A. Navarro, I. Salaverria, D. Martin-Garcia, P. Jares, E. Gine, M. Pinyol, C. Royo, F. Nadeu, L. Conde, M. Juan, G. Clot, P. Vizan, L. Di Croce, D. A. Puente, M. Lopez-Guerra, A. Moros, G. Roue, M. Aymerich, N. Villamor, L. Colomo, A. Martinez, A. Valera, J. I. Martin-Subero, V. Amador, L. Hernandez, M. Rozman, A. Enjuanes, P. Forcada, A. Muntanola, E. M. Hartmann, M. J. Calasanz, A. Rosenwald, G. Ott, J. M. Hernandez-Rivas, W. Klapper, R. Siebert, A. Wiestner, W. H. Wilson, D. Colomer, A. Lopez-Guillermo, C. Lopez-Otin, X. S. Puente and E. Campo (2013). "Landscape of somatic mutations and clonal evolution in mantle cell lymphoma." Proc Natl Acad Sci U S A 110(45): 18250-18255.

Becker-Catania, S. G., G. Chen, M. J. Hwang, Z. Wang, X. Sun, O. Sanal, E. Bernatowska- Matuszkiewicz, L. Chessa, E. Y. Lee and R. A. Gatti (2000). "Ataxia-telangiectasia: phenotype/genotype studies of ATM protein expression, mutations, and radiosensitivity." Mol Genet Metab 70(2): 122-133.

Berger, M., N. Stahl, G. Del Sal and Y. Haupt (2005). "Mutations in proline 82 of p53 impair its activation by Pin1 and Chk2 in response to DNA damage." Molecular and Cellular Biology 25(13): 5380-5388.

101

Binz, S. K., A. M. Sheehan and M. S. Wold (2004). "Replication protein A phosphorylation and the cellular response to DNA damage." DNA Repair (Amst) 3(8-9): 1015-1024.

Boboila, C., C. Yan, D. R. Wesemann, M. Jankovic, J. H. Wang, J. Manis, A. Nussenzweig, M. Nussenzweig and F. W. Alt (2010). "Alternative end-joining catalyzes class switch recombination in the absence of both Ku70 and DNA ligase 4." J Exp Med 207(2): 417-427.

Bodrug, S. E., B. J. Warner, M. L. Bath, G. J. Lindeman, A. W. Harris and J. M. Adams (1994). "Cyclin D1 transgene impedes lymphocyte maturation and collaborates in lymphomagenesis with the myc gene." EMBO J 13(9): 2124-2130.

Borghesani, P. R., F. W. Alt, A. Bottaro, L. Davidson, S. Aksoy, G. A. Rathbun, T. M. Roberts, W. Swat, R. A. Segal and Y. Gu (2000). "Abnormal development of Purkinje cells and lymphocytes in Atm mutant mice." Proc Natl Acad Sci U S A 97(7): 3336-3341.

Bredemeyer, A. L., G. G. Sharma, C. Y. Huang, B. A. Helmink, L. M. Walker, K. C. Khor, B. Nuskey, K. E. Sullivan, T. K. Pandita, C. H. Bassing and B. P. Sleckman (2006). "ATM stabilizes DNA double-strand-break complexes during V(D)J recombination." Nature 442(7101): 466-470.

Bross, L., Y. Fukita, F. McBlane, C. Demolliere, K. Rajewsky and H. Jacobs (2000). "DNA double-strand breaks in immunoglobulin genes undergoing somatic hypermutation." Immunity 13(5): 589-597.

Brown, E. J. and D. Baltimore (2003). "Essential and dispensable roles of ATR in cell cycle arrest and genome maintenance." Genes Dev 17(5): 615-628.

Buis, J., Y. Wu, Y. Deng, J. Leddon, G. Westfield, M. Eckersdorff, J. M. Sekiguchi, S. Chang and D. O. Ferguson (2008). "Mre11 nuclease activity has essential roles in DNA repair and genomic stability distinct from ATM activation." Cell 135(1): 85-96.

Callen, E., M. Jankovic, S. Difilippantonio, J. A. Daniel, H. T. Chen, A. Celeste, M. Pellegrini, K. McBride, D. Wangsa, A. L. Bredemeyer, B. P. Sleckman, T. Ried, M. Nussenzweig and A. Nussenzweig (2007). "ATM prevents the persistence and propagation of chromosome breaks in lymphocytes." Cell. 130(1): 63-75.

Callen, E., M. Jankovic, N. Wong, S. Zha, H. T. Chen, S. Difilippantonio, M. Di Virgilio, G. Heidkamp, F. W. Alt, A. Nussenzweig and M. Nussenzweig (2009). "Essential role for DNA- PKcs in DNA double-strand break repair and apoptosis in ATM-deficient lymphocytes." Mol Cell 34(3): 285-297.

Camacho, E., L. Hernandez, S. Hernandez, F. Tort, B. Bellosillo, S. Bea, F. Bosch, E. Montserrat, A. Cardesa, P. L. Fernandez and E. Campo (2002). "ATM gene inactivation in mantle cell lymphoma mainly occurs by truncating mutations and missense mutations involving the phosphatidylinositol-3 kinase domain and is associated with increasing numbers of chromosomal imbalances." Blood 99(1): 238-244.

Canman, C. E., D. S. Lim, K. A. Cimprich, Y. Taya, K. Tamai, K. Sakaguchi, E. Appella, M. B. Kastan and J. D. Siliciano (1998). "Activation of the ATM kinase by ionizing radiation and phosphorylation of p53." Science 281(5383): 1677-1679.

102

Carthon, B. C., C. A. Neumann, M. Das, B. Pawlyk, T. Li, Y. Geng and P. Sicinski (2005). "Genetic replacement of cyclin D1 function in mouse development by cyclin D2." Mol Cell Biol 25(3): 1081-1088.

Castor, D., N. Nair, A. C. Declais, C. Lachaud, R. Toth, T. J. Macartney, D. M. Lilley, J. S. Arthur and J. Rouse (2013). "Cooperative control of holliday junction resolution and DNA repair by the SLX1 and MUS81-EME1 nucleases." Mol Cell 52(2): 221-233.

Cato, M. H., S. K. Chintalapati, I. W. Yau, S. A. Omori and R. C. Rickert (2011). "Cyclin D3 is selectively required for proliferative expansion of germinal center B cells." Mol Cell Biol 31(1): 127-137.

Cattoretti, G., L. Pasqualucci, G. Ballon, W. Tam, S. V. Nandula, Q. Shen, T. Mo, V. V. Murty and R. Dalla-Favera (2005). "Deregulated BCL6 expression recapitulates the pathogenesis of human diffuse large B cell lymphomas in mice." Cancer Cell 7(5): 445-455.

Chan, Y. W. and S. C. West (2014). "Spatial control of the GEN1 Holliday junction resolvase ensures genome stability." Nat Commun 5: 4844.

Chen, H., R. A. Donnianni, N. Handa, S. K. Deng, J. Oh, L. A. Timashev, S. C. Kowalczykowski and L. S. Symington (2015). "Sae2 promotes DNA damage resistance by removing the Mre11- Rad50-Xrs2 complex from DNA and attenuating Rad53 signaling." Proc Natl Acad Sci U S A 112(15): E1880-1887.

Chen, W. T., N. D. Ebelt, T. H. Stracker, B. Xhemalce, C. L. Van Den Berg and K. M. Miller (2015). "ATM regulation of IL-8 links oxidative stress to cancer cell migration and invasion." Elife 4.

Ciemerych, M. A. and P. Sicinski (2005). "Cell cycle in mouse development." Oncogene 24(17): 2877-2898.

Clauson, C., O. D. Scharer and L. Niedernhofer (2013). "Advances in understanding the complex mechanisms of DNA interstrand cross-link repair." Cold Spring Harb Perspect Med 3(10): a012732.

Clerici, M., D. Mantiero, G. Lucchini and M. P. Longhese (2006). "The Saccharomyces cerevisiae Sae2 protein negatively regulates DNA damage checkpoint signalling." EMBO Rep 7(2): 212-218.

Cremona, C. A. and A. Behrens (2014). "ATM signalling and cancer." Oncogene 33(26): 3351- 3360.

D'Souza, A. D., I. A. Parish, D. S. Krause, S. M. Kaech and G. S. Shadel (2013). "Reducing mitochondrial ROS improves disease-related pathology in a mouse model of ataxia- telangiectasia." Mol Ther 21(1): 42-48.

Daniel, J. A., M. Pellegrini, B. S. Lee, Z. Guo, D. Filsuf, N. Belkina, Z. You, T. T. Paull, B. P. Sleckman, L. Feigenbaum and A. Nussenzweig (2012). "Loss of ATM kinase activity leads to embryonic lethality in mice." Journal of Cell Biology.

103

Daniel, J. A., M. Pellegrini, J. H. Lee, T. T. Paull, L. Feigenbaum and A. Nussenzweig (2008). "Multiple autophosphorylation sites are dispensable for murine ATM activation in vivo." J Cell Biol 183(5): 777-783. de Boer, J., A. Williams, G. Skavdis, N. Harker, M. Coles, M. Tolaini, T. Norton, K. Williams, K. Roderick, A. J. Potocnik and D. Kioussis (2003). "Transgenic mice with hematopoietic and lymphoid specific expression of Cre." Eur J Immunol 33(2): 314-325. de Luca, C., T. J. Kowalski, Y. Zhang, J. K. Elmquist, C. Lee, M. W. Kilimann, T. Ludwig, S. M. Liu and S. C. Chua, Jr. (2005). "Complete rescue of obesity, diabetes, and infertility in db/db mice by neuron-specific LEPR-B transgenes." J Clin Invest 115(12): 3484-3493.

Dobbs, T. A., J. A. Tainer and S. P. Lees-Miller (2010). "A structural model for regulation of NHEJ by DNA-PKcs autophosphorylation." DNA Repair (Amst) 9(12): 1307-1314.

Dudgeon, C., C. Chan, W. Kang, Y. Sun, R. Emerson, H. Robins and A. J. Levine (2014). "The evolution of thymic lymphomas in p53 knockout mice." Genes Dev 28(23): 2613-2620.

Eaton, J. S., Z. P. Lin, A. C. Sartorelli, N. D. Bonawitz and G. S. Shadel (2007). "Ataxia- telangiectasia mutated kinase regulates ribonucleotide reductase and mitochondrial homeostasis." J Clin Invest 117(9): 2723-2734.

Fernandez-Capetillo, O., A. Lee, M. Nussenzweig and A. Nussenzweig (2004). "H2AX: the histone guardian of the genome." DNA Repair (Amst) 3(8-9): 959-967.

Franco, S., F. W. Alt and J. P. Manis (2006). "Pathways that suppress programmed DNA breaks from progressing to chromosomal breaks and translocations." DNA Repair (Amst) 5(9-10): 1030-1041.

Franco, S., M. Gostissa, S. Zha, D. B. Lombard, M. M. Murphy, A. A. Zarrin, C. Yan, S. Tepsuporn, J. C. Morales, M. M. Adams, Z. Lou, C. H. Bassing, J. P. Manis, J. Chen, P. B. Carpenter and F. W. Alt (2006). "H2AX prevents DNA breaks from progressing to chromosome breaks and translocations." Mol Cell 21(2): 201-214.

Frank, K. M., N. E. Sharpless, Y. Gao, J. M. Sekiguchi, D. O. Ferguson, C. Zhu, J. P. Manis, J. Horner, R. A. DePinho and F. W. Alt (2000). "DNA ligase IV deficiency in mice leads to defective neurogenesis and embryonic lethality via the p53 pathway." Mol Cell 5(6): 993-1002.

Gamper, A. M., S. Choi, Y. Matsumoto, D. Banerjee, A. E. Tomkinson and C. J. Bakkenist (2012). "ATM Protein Physically and Functionally Interacts with Proliferating Cell Nuclear Antigen to Regulate DNA Synthesis." J Biol Chem 287(15): 12445-12454.

Gao, Y., D. O. Ferguson, W. Xie, J. P. Manis, J. Sekiguchi, K. M. Frank, J. Chaudhuri, J. Horner, R. A. DePinho and F. W. Alt (2000). "Interplay of p53 and DNA-repair protein XRCC4 in tumorigenesis, genomic stability and development." Nature 404(6780): 897-900.

Gapud, E. J., Y. Dorsett, B. Yin, E. Callen, A. Bredemeyer, G. K. Mahowald, K. Q. Omi, L. M. Walker, J. J. Bednarski, P. J. McKinnon, C. H. Bassing, A. Nussenzweig and B. P. Sleckman (2011). "Ataxia telangiectasia mutated (Atm) and DNA-PKcs kinases have overlapping activities during chromosomal signal joint formation." Proc Natl Acad Sci U S A 108(5): 2022-2027.

104

Germain, R. N. (2002). "T-cell development and the CD4-CD8 lineage decision." Nat Rev Immunol 2(5): 309-322.

Gladden, A. B., R. Woolery, P. Aggarwal, M. A. Wasik and J. A. Diehl (2006). "Expression of constitutively nuclear cyclin D1 in murine lymphocytes induces B-cell lymphoma." Oncogene 25(7): 998-1007.

Golding, S. E., E. Rosenberg, N. Valerie, I. Hussaini, M. Frigerio, X. F. Cockcroft, W. Y. Chong, M. Hummersone, L. Rigoreau, K. A. Menear, M. J. O'Connor, L. F. Povirk, T. van Meter and K. Valerie (2009). "Improved ATM kinase inhibitor KU-60019 radiosensitizes glioma cells, compromises insulin, AKT and ERK prosurvival signaling, and inhibits migration and invasion." Mol Cancer Ther 8(10): 2894-2902.

Gostissa, M., C. T. Yan, J. M. Bianco, M. Cogne, E. Pinaud and F. W. Alt (2009). "Long-range oncogenic activation of Igh-c-myc translocations by the Igh 3' regulatory region." Nature 462(7274): 803-807.

Gronbaek, K., J. Worm, E. Ralfkiaer, V. Ahrenkiel, P. Hokland and P. Guldberg (2002). "ATM mutations are associated with inactivation of the ARF-TP53 tumor suppressor pathway in diffuse large B-cell lymphoma." Blood 100(4): 1430-1437.

Gumy-Pause, F., P. Wacker and A. P. Sappino (2004). "ATM gene and lymphoid malignancies." Leukemia 18(2): 238-242.

Guo, K., J. E. McMinn, T. Ludwig, Y. H. Yu, G. Yang, L. Chen, D. Loh, C. Li, S. Chua, Jr. and Y. Zhang (2007). "Disruption of peripheral leptin signaling in mice results in hyperleptinemia without associated metabolic abnormalities." Endocrinology 148(8): 3987-3997.

Guo, Z., R. Deshpande and T. T. Paull (2010). "ATM activation in the presence of oxidative stress." Cell Cycle 9(24): 4805-4811.

Guo, Z., S. Kozlov, M. F. Lavin, M. D. Person and T. T. Paull (2010). "ATM activation by oxidative stress." Science 330(6003): 517-521.

Gurley, K. E. and C. J. Kemp (2001). "Synthetic lethality between mutation in Atm and DNA- PK(cs) during murine embryogenesis." Curr Biol 11(3): 191-194.

Han, L. and K. Yu (2008). "Altered kinetics of nonhomologous end joining and class switch recombination in ligase IV-deficient B cells." J Exp Med 205(12): 2745-2753.

Herrmann, A., E. Hoster, T. Zwingers, G. Brittinger, M. Engelhard, P. Meusers, M. Reiser, R. Forstpointner, B. Metzner, N. Peter, B. Wormann, L. Trumper, M. Pfreundschuh, H. Einsele, W. Hiddemann, M. Unterhalt and M. Dreyling (2009). "Improvement of Overall Survival in Advanced Stage Mantle Cell Lymphoma." Journal of Clinical Oncology 27(4): 511-518.

Hickson, I., Y. Zhao, C. J. Richardson, S. J. Green, N. M. Martin, A. I. Orr, P. M. Reaper, S. P. Jackson, N. J. Curtin and G. C. Smith (2004). "Identification and characterization of a novel and specific inhibitor of the ataxia-telangiectasia mutated kinase ATM." Cancer Res 64(24): 9152- 9159.

105

Hobeika, E., S. Thiemann, B. Storch, H. Jumaa, P. J. Nielsen, R. Pelanda and M. Reth (2006). "Testing gene function early in the B cell lineage in mb1-cre mice." Proc Natl Acad Sci U S A 103(37): 13789-13794.

Hopfner, K. P., L. Craig, G. Moncalian, R. A. Zinkel, T. Usui, B. A. Owen, A. Karcher, B. Henderson, J. L. Bodmer, C. T. McMurray, J. P. Carney, J. H. Petrini and J. A. Tainer (2002). "The Rad50 zinc-hook is a structure joining Mre11 complexes in DNA recombination and repair." Nature 418(6897): 562-566.

Hsieh, C. L., C. F. Arlett and M. R. Lieber (1993). "V(D)J recombination in ataxia telangiectasia, Bloom's syndrome, and a DNA ligase I-associated immunodeficiency disorder." J Biol Chem 268(27): 20105-20109.

Ip, S. C., U. Rass, M. G. Blanco, H. R. Flynn, J. M. Skehel and S. C. West (2008). "Identification of Holliday junction resolvases from humans and yeast." Nature 456(7220): 357-361.

Jares, P., D. Colomer and E. Campo (2007). "Genetic and molecular pathogenesis of mantle cell lymphoma: perspectives for new targeted therapeutics." Nat Rev Cancer 7(10): 750-762.

Jenkins, L. M. M., S. R. Durell, S. J. Mazur and E. Appella (2012). "p53 N-terminal phosphorylation: a defining layer of complex regulation." Carcinogenesis 33(8): 1441-1449.

Jiang, W., J. L. Crowe, X. Liu, S. Nakajima, Y. Wang, C. Li, B. J. Lee, R. L. Dubois, C. Liu, X. Yu, L. Lan and S. Zha (2015). "Differential phosphorylation of DNA-PKcs regulates the interplay between end-processing and end-ligation during nonhomologous end-joining." Mol Cell 58(1): 172-185.

Jiang, W., B. J. Lee, C. Li, R. L. Dubois, M. Gostissa, F. W. Alt and S. Zha (2015). "Aberrant TCRdelta rearrangement underlies the T cell lymphocytopenia and t(12;14) translocation associated with ATM deficiency." Blood.

Jolly, C. J., N. Klix and M. S. Neuberger (1997). "Rapid methods for the analysis of immunoglobulin gene hypermutation: application to transgenic and gene targeted mice." Nucleic Acids Res 25(10): 1913-1919.

Katyal, S., Y. Lee, K. C. Nitiss, S. M. Downing, Y. Li, M. Shimada, J. Zhao, H. R. Russell, J. H. Petrini, J. L. Nitiss and P. J. McKinnon (2014). "Aberrant topoisomerase-1 DNA lesions are pathogenic in neurodegenerative genome instability syndromes." Nat Neurosci 17(6): 813-821.

Keeney, S. (2001). "Mechanism and control of meiotic recombination initiation." Curr Top Dev Biol 52: 1-53.

Khor, B. and B. P. Sleckman (2005). "Intra- and inter-allelic ordering of T cell receptor beta chain gene assembly." Eur J Immunol 35(3): 964-970.

Kikushige, Y., F. Ishikawa, T. Miyamoto, T. Shima, S. Urata, G. Yoshimoto, Y. Mori, T. Iino, T. Yamauchi, T. Eto, H. Niiro, H. Iwasaki, K. Takenaka and K. Akashi (2011). "Self-renewing hematopoietic stem cell is the primary target in pathogenesis of human chronic lymphocytic leukemia." Cancer Cell 20(2): 246-259.

106

Kim, D. E., D. Chivian and D. Baker (2004). "Protein structure prediction and analysis using the Robetta server." Nucleic Acids Res 32(Web Server issue): W526-531.

Klein, U. and R. Dalla-Favera (2008). "Germinal centres: role in B-cell physiology and malignancy." Nat Rev Immunol 8(1): 22-33.

Klein, U., M. Lia, M. Crespo, R. Siegel, Q. Shen, T. Mo, A. Ambesi-Impiombato, A. Califano, A. Migliazza, G. Bhagat and R. Dalla-Favera (2010). "The DLEU2/miR-15a/16-1 cluster controls B cell proliferation and its deletion leads to chronic lymphocytic leukemia." Cancer Cell 17(1): 28- 40.

Kozlov, S. V., M. E. Graham, C. Peng, P. Chen, P. J. Robinson and M. F. Lavin (2006). "Involvement of novel autophosphorylation sites in ATM activation." EMBO J 25(15): 3504-3514.

Kraus, M., M. B. Alimzhanov, N. Rajewsky and K. Rajewsky (2004). "Survival of resting mature B lymphocytes depends on BCR signaling via the Igalpha/beta heterodimer." Cell 117(6): 787- 800.

Kwon, K., C. Hutter, Q. Sun, I. Bilic, C. Cobaleda, S. Malin and M. Busslinger (2008). "Instructive role of the transcription factor E2A in early B lymphopoiesis and germinal center B cell development." Immunity 28(6): 751-762.

Li, G., F. W. Alt, H. L. Cheng, J. W. Brush, P. H. Goff, M. M. Murphy, S. Franco, Y. Zhang and S. Zha (2008). "Lymphocyte-Specific Compensation for XLF/Cernunnos End-Joining Functions in V(D)J Recombination." Mol.Cell. 31: 631-640.

Liberi, G., G. Maffioletti, C. Lucca, I. Chiolo, A. Baryshnikova, C. Cotta-Ramusino, M. Lopes, A. Pellicioli, J. E. Haber and M. Foiani (2005). "Rad51-dependent DNA structures accumulate at damaged replication forks in sgs1 mutants defective in the yeast ortholog of BLM RecQ helicase." Genes Dev 19(3): 339-350.

Lieber, M. R. (2010). "The mechanism of double-strand DNA break repair by the nonhomologous DNA end-joining pathway." Annu Rev Biochem 79: 181-211.

Lieber, M. R., Y. Ma, U. Pannicke and K. Schwarz (2003). "Mechanism and regulation of human non-homologous DNA end-joining." Nat Rev Mol Cell Biol 4(9): 712-720.

Liu, S., B. Shiotani, M. Lahiri, A. Marechal, A. Tse, C. C. Leung, J. N. Glover, X. H. Yang and L. Zou (2011). "ATR autophosphorylation as a molecular switch for checkpoint activation." Mol Cell 43(2): 192-202.

Llorca, O., A. Rivera-Calzada, J. Grantham and K. R. Willison (2003). "Electron microscopy and 3D reconstructions reveal that human ATM kinase uses an arm-like domain to clamp around double-stranded DNA." Oncogene 22(25): 3867-3874.

Llorente, B., C. E. Smith and L. S. Symington (2008). "Break-induced replication - What is it and what is it for?" Cell Cycle 7(7): 859-864.

Lovec, H., A. Grzeschiczek, M. B. Kowalski and T. Moroy (1994). "Cyclin D1/bcl-1 cooperates with myc genes in the generation of B-cell lymphoma in transgenic mice." EMBO J 13(15): 3487-3495.

107

Lumsden, J. M., T. McCarty, L. K. Petiniot, R. Shen, C. Barlow, T. A. Wynn, H. C. Morse, 3rd, P. J. Gearhart, A. Wynshaw-Boris, E. E. Max and R. J. Hodes (2004). "Immunoglobulin class switch recombination is impaired in Atm-deficient mice." J Exp Med 200(9): 1111-1121.

Marechal, A. and L. Zou (2015). "RPA-coated single-stranded DNA as a platform for post- translational modifications in the DNA damage response." Cell Res 25(1): 9-23.

McIlwraith, M. J., E. Van Dyck, J. Y. Masson, A. Z. Stasiak, A. Stasiak and S. C. West (2000). "Reconstitution of the strand invasion step of double-strand break repair using human Rad51 Rad52 and RPA proteins." Journal of Molecular Biology 304(2): 151-164.

McKinnon, P. J. (2012). "ATM and the molecular pathogenesis of ataxia telangiectasia." Annu Rev Pathol 7: 303-321.

McVey, M. and S. E. Lee (2008). "MMEJ repair of double-strand breaks (director's cut): deleted sequences and alternative endings." Trends Genet 24(11): 529-538.

Minegishi, Y., E. Coustan-Smith, L. Rapalus, F. Ersoy, D. Campana and M. E. Conley (1999). "Mutations in Igalpha (CD79a) result in a complete block in B-cell development." J Clin Invest 104(8): 1115-1121.

Moreno-Herrero, F., M. de Jager, N. H. Dekker, R. Kanaar, C. Wyman and C. Dekker (2005). "Mesoscale conformational changes in the DNA-repair complex Rad50/Mre11/Nbs1 upon binding DNA." Nature 437(7057): 440-443.

Murai, J., Y. Zhang, J. Morris, J. Ji, S. Takeda, J. H. Doroshow and Y. Pommier (2014). "Rationale for poly(ADP-ribose) polymerase (PARP) inhibitors in combination therapy with camptothecins or temozolomide based on PARP trapping versus catalytic inhibition." J Pharmacol Exp Ther 349(3): 408-416.

Muramatsu, M., K. Kinoshita, S. Fagarasan, S. Yamada, Y. Shinkai and T. Honjo (2000). "Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme." Cell 102(5): 553-563.

Nitiss, J. L. (2001). "Topoisomerase assays." Curr Protoc Pharmacol Chapter 3: Unit3 3.

Nowak-Wegrzyn, A., T. O. Crawford, J. A. Winkelstein, K. A. Carson and H. M. Lederman (2004). "Immunodeficiency and infections in ataxia-telangiectasia." Journal of Pediatrics 144(4): 505-511.

O'Gorman, S., N. A. Dagenais, M. Qian and Y. Marchuk (1997). "Protamine-Cre recombinase transgenes efficiently recombine target sequences in the male germ line of mice, but not in embryonic stem cells." Proc Natl Acad Sci U S A 94(26): 14602-14607.

Ogi, H., G. H. Goto, A. Ghosh, S. Zencir, E. Henry and K. Sugimoto (2015). "Requirement of the FATC domain of protein kinase Tel1 for localization to DNA ends and target protein recognition." Mol Biol Cell.

Papavasiliou, F. N. and D. G. Schatz (2000). "Cell-cycle-regulated DNA double-stranded breaks in somatic hypermutation of immunoglobulin genes." Nature 408(6809): 216-221.

108

Pear, W. S., J. C. Aster, M. L. Scott, R. P. Hasserjian, B. Soffer, J. Sklar and D. Baltimore (1996). "Exclusive development of T cell neoplasms in mice transplanted with bone marrow expressing activated Notch alleles." J Exp Med 183(5): 2283-2291.

Pekarsky, Y., C. Hallas and C. M. Croce (2001). "Molecular basis of mature T-cell leukemia." JAMA 286(18): 2308-2314.

Pelanda, R., U. Braun, E. Hobeika, M. C. Nussenzweig and M. Reth (2002). "B cell progenitors are arrested in maturation but have intact VDJ recombination in the absence of Ig-alpha and Ig- beta." J Immunol 169(2): 865-872.

Peled, J. U., J. J. Yu, J. Venkatesh, E. Bi, B. B. Ding, M. Krupski-Downs, R. Shaknovich, P. Sicinski, B. Diamond, M. D. Scharff and B. H. Ye (2010). "Requirement for cyclin D3 in germinal center formation and function." Cell Res 20(6): 631-646.

Pellegrini, M., A. Celeste, S. Difilippantonio, R. Guo, W. Wang, L. Feigenbaum and A. Nussenzweig (2006). "Autophosphorylation at serine 1987 is dispensable for murine Atm activation in vivo." Nature 443(7108): 222-225.

Pommier, Y. (2006). "Topoisomerase I inhibitors: camptothecins and beyond." Nat Rev Cancer 6(10): 789-802.

Pommier, Y. (2009). "DNA topoisomerase I inhibitors: chemistry, biology, and interfacial inhibition." Chem Rev 109(7): 2894-2902.

Rainey, M. D., M. E. Charlton, R. V. Stanton and M. B. Kastan (2008). "Transient inhibition of ATM kinase is sufficient to enhance cellular sensitivity to ionizing radiation." Cancer Res 68(18): 7466-7474.

Reina-San-Martin, B., H. T. Chen, A. Nussenzweig and M. C. Nussenzweig (2004). "ATM is required for efficient recombination between immunoglobulin switch regions." J.Exp.Med. 200(9): 1103-1110.

Rickert, R. C., K. Rajewsky and J. Roes (1995). "Impairment of T-cell-dependent B-cell responses and B-1 cell development in CD19-deficient mice." Nature 376(6538): 352-355.

Roberts, N. J., Y. Jiao, J. Yu, L. Kopelovich, G. M. Petersen, M. L. Bondy, S. Gallinger, A. G. Schwartz, S. Syngal, M. L. Cote, J. Axilbund, R. Schulick, S. Z. Ali, J. R. Eshleman, V. E. Velculescu, M. Goggins, B. Vogelstein, N. Papadopoulos, R. H. Hruban, K. W. Kinzler and A. P. Klein (2012). "ATM mutations in patients with hereditary pancreatic cancer." Cancer Discov 2(1): 41-46.

Rouse, J. (2009). "Control of genome stability by SLX protein complexes." Biochem Soc Trans 37(Pt 3): 495-510.

Saito, S., A. A. Goodarzi, Y. Higashimoto, Y. Noda, S. P. Lees-Miller, E. Appella and C. W. Anderson (2002). "ATM mediates phosphorylation at multiple p53 sites, including Ser(46), in response to ionizing radiation." Journal of Biological Chemistry 277(15): 12491-12494.

San Filippo, J., P. Sung and H. Klein (2008). "Mechanism of eukaryotic homologous recombination." Annu Rev Biochem 77: 229-257.

109

Sanal, O., F. Ozaltin, I. Tezcan and F. Ersoy (1998). "Serum IgD concentrations in patients with ataxia telangiectasia and with selective IgA deficiency." International Archives of Allergy and Immunology 116(3): 246-246.

Santra, M. K., N. Wajapeyee and M. R. Green (2009). "F-box protein FBXO31 mediates cyclin D1 degradation to induce G1 arrest after DNA damage." Nature 459(7247): 722-725.

Sarbajna, S. and S. C. West (2014). "Holliday junction processing enzymes as guardians of genome stability." Trends in Biochemical Sciences 39(9): 409-419.

Schroeter, E. H., J. A. Kisslinger and R. Kopan (1998). "Notch-1 signalling requires ligand- induced proteolytic release of intracellular domain." Nature 393(6683): 382-386.

Schubert, R., L. Erker, C. Barlow, H. Yakushiji, D. Larson, A. Russo, J. B. Mitchell and A. Wynshaw-Boris (2004). "Cancer chemoprevention by the antioxidant tempol in Atm-deficient mice." Hum Mol Genet 13(16): 1793-1802.

Schubert, R., J. Reichenbach and S. Zielen (2002). "Deficiencies in CD4(+) and CD8(+) T cell subsets in ataxia telangiectasia." Clinical and Experimental Immunology 129(1): 125-132.

Sekiguchi, J., D. O. Ferguson, H. T. Chen, E. M. Yang, J. Earle, K. Frank, S. Whitlow, Y. Gu, Y. Xu, A. Nussenzweig and F. W. Alt (2001). "Genetic interactions between ATM and the nonhomologous end-joining factors in genomic stability and development." Proc.Natl.Acad.Sci.U.S.A 98(6): 3243-3248.

Shakya, R., L. J. Reid, C. R. Reczek, F. Cole, D. Egli, C. S. Lin, D. G. deRooij, S. Hirsch, K. Ravi, J. B. Hicks, M. Szabolcs, M. Jasin, R. Baer and T. Ludwig (2011). "BRCA1 tumor suppression depends on BRCT phosphoprotein binding, but not its E3 ligase activity." Science 334(6055): 525-528.

Shiloh, Y. and Y. Ziv (2013). "The ATM protein kinase: regulating the cellular response to genotoxic stress, and more." Nat Rev Mol Cell Biol 14(4): 197-210.

Shiotani, B. and L. Zou (2009). "Single-stranded DNA orchestrates an ATM-to-ATR switch at DNA breaks." Mol Cell 33(5): 547-558.

Sibanda, B. L., D. Y. Chirgadze and T. L. Blundell (2010). "Crystal structure of DNA-PKcs reveals a large open-ring cradle comprised of HEAT repeats." Nature 463(7277): 118-121.

So, S., A. J. Davis and D. J. Chen (2009). "Autophosphorylation at serine 1981 stabilizes ATM at DNA damage sites." J Cell Biol 187(7): 977-990.

Soulas-Sprauel, P., G. Le Guyader, P. Rivera-Munoz, V. Abramowski, C. Olivier-Martin, C. Goujet-Zalc, P. Charneau and J. P. de Villartay (2007). "Role for DNA repair factor XRCC4 in immunoglobulin class switch recombination." J Exp Med 204(7): 1717-1727.

Spring, K., F. Ahangari, S. P. Scott, P. Waring, D. M. Purdie, P. C. Chen, K. Hourigan, J. Ramsay, P. J. McKinnon, M. Swift and M. F. Lavin (2002). "Mice heterozygous for mutation in Atm, the gene involved in ataxia-telangiectasia, have heightened susceptibility to cancer." Nat Genet 32(1): 185-190.

110

Stankovic, T., G. S. Stewart, P. Byrd, C. Fegan, P. A. Moss and A. M. Taylor (2002). "ATM mutations in sporadic lymphoid tumours." Leuk Lymphoma 43(8): 1563-1571.

Stankovic, T., A. M. Taylor, M. R. Yuille and I. Vorechovsky (2001). "Recurrent ATM mutations in T-PLL on diverse haplotypes: no support for their germline origin." Blood 97(5): 1517-1518.

Stavnezer, J., J. E. Guikema and C. E. Schrader (2008). "Mechanism and regulation of class switch recombination." Annu Rev Immunol 26: 261-292.

Stiff, T., M. O'Driscoll, N. Rief, K. Iwabuchi, M. Lobrich and P. A. Jeggo (2004). "ATM and DNA- PK function redundantly to phosphorylate H2AX after exposure to ionizing radiation." Cancer Res 64(7): 2390-2396.

Stilgenbauer, S., C. Schaffner, A. Litterst, P. Liebisch, S. Gilad, A. Bar-Shira, M. R. James, P. Lichter and H. Dohner (1997). "Biallelic mutations in the ATM gene in T-prolymphocytic leukemia." Nat Med 3(10): 1155-1159.

Sun, Y., Y. Xu, K. Roy and B. D. Price (2007). "DNA damage-induced acetylation of lysine 3016 of ATM activates ATM kinase activity." Mol Cell Biol 27(24): 8502-8509.

Symington, L. S. and J. Gautier (2011). "Double-strand break end resection and repair pathway choice." Annu Rev Genet 45: 247-271.

Takeuchi, S., M. Koike, S. Park, T. Seriu, C. R. Bartram, H. E. Taub, I. K. Williamson, J. Grewal, H. Taguchi and H. P. Koeffler (1998). "The ATM gene and susceptibility to childhood T-cell acute lymphoblastic leukaemia." Br J Haematol 103(2): 536-538.

Vaites, L. P., E. K. Lee, Z. Lian, O. Barbash, D. Roy, M. Wasik, A. J. P. Klein-Szanto, A. K. Rustgi and J. A. Diehl (2011). "The Fbx4 Tumor Suppressor Regulates Cyclin D1 Accumulation and Prevents Neoplastic Transformation." Mol Cell Biol 31(22): 4513-4523.

Vaites, L. P., Z. Lian, E. K. Lee, B. Yin, A. DeMicco, C. H. Bassing and J. A. Diehl (2014). "ATM deficiency augments constitutively nuclear cyclin D1-driven genomic instability and lymphomagenesis." Oncogene 33(1): 129-133.

Valentin-Vega, Y. A., K. H. Maclean, J. Tait-Mulder, S. Milasta, M. Steeves, F. C. Dorsey, J. L. Cleveland, D. R. Green and M. B. Kastan (2012). "Mitochondrial dysfunction in ataxia- telangiectasia." Blood 119(6): 1490-1500. van Gent, D. C., D. A. Ramsden and M. Gellert (1996). "The RAG1 and RAG2 proteins establish the 12/23 rule in V(D)J recombination." Cell 85(1): 107-113.

Walsh, S. H., M. Thorselius, A. Johnson, O. Soderberg, M. Jerkeman, E. Bjorck, I. Eriksson, U. Thunberg, O. Landgren, M. Ehinger, E. Lofvenberg, K. Wallman, G. Enblad, B. Sander, A. Porwit-MacDonald, M. Dictor, T. Olofsson, C. Sundstrom, G. Roos and R. Rosenquist (2003). "Mutated VH genes and preferential VH3-21 use define new subsets of mantle cell lymphoma." Blood 101(10): 4047-4054.

Wang, L., M. S. Lawrence, Y. Wan, P. Stojanov, C. Sougnez, K. Stevenson, L. Werner, A. Sivachenko, D. S. DeLuca, L. Zhang, W. Zhang, A. R. Vartanov, S. M. Fernandes, N. R. Goldstein, E. G. Folco, K. Cibulskis, B. Tesar, Q. L. Sievers, E. Shefler, S. Gabriel, N. Hacohen,

111

R. Reed, M. Meyerson, T. R. Golub, E. S. Lander, D. Neuberg, J. R. Brown, G. Getz and C. J. Wu (2011). "SF3B1 and other novel cancer genes in chronic lymphocytic leukemia." N Engl J Med 365(26): 2497-2506.

White, J. S., S. Choi and C. J. Bakkenist (2010). "Transient ATM kinase inhibition disrupts DNA damage-induced sister chromatid exchange." Sci Signal 3(124): ra44.

Williamson, C. T., E. Kubota, J. D. Hamill, A. Klimowicz, R. Ye, H. Muzik, M. Dean, L. Tu, D. Gilley, A. M. Magliocco, B. C. McKay, D. G. Bebb and S. P. Lees-Miller (2012). "Enhanced cytotoxicity of PARP inhibition in mantle cell lymphoma harbouring mutations in both ATM and p53." EMBO Mol Med 4(6): 515-527.

Wiltzius, J. J., M. Hohl, J. C. Fleming and J. H. Petrini (2005). "The Rad50 hook domain is a critical determinant of Mre11 complex functions." Nat Struct Mol Biol 12(5): 403-407.

Wu, L. and I. D. Hickson (2003). "The Bloom's syndrome helicase suppresses crossing over during homologous recombination." Nature 426(6968): 870-874.

Xu, Y. (1999). "ATM in lymphoid development and tumorigenesis." Adv Immunol 72: 179-189.

Xu, Y., T. Ashley, E. E. Brainerd, R. T. Bronson, M. S. Meyn and D. Baltimore (1996). "Targeted disruption of ATM leads to growth retardation, chromosomal fragmentation during meiosis, immune defects, and thymic lymphoma." Genes Dev 10(19): 2411-2422.

Yamamoto, K., Y. Wang, W. Jiang, X. Liu, R. L. Dubois, C. S. Lin, T. Ludwig, C. J. Bakkenist and S. Zha (2012). "Kinase-dead ATM protein causes genomic instability and early embryonic lethality in mice." J Cell Biol 198(3): 305-313.

Yan, C. T., C. Boboila, E. K. Souza, S. Franco, T. R. Hickernell, M. Murphy, S. Gumaste, M. Geyer, A. A. Zarrin, J. P. Manis, K. Rajewsky and F. W. Alt (2007). "IgH class switching and translocations use a robust non-classical end-joining pathway." Nature 449(7161): 478-482.

Yang, D. Q. and M. B. Kastan (2000). "Participation of ATM in insulin signalling through phosphorylation of eIF-4E-binding protein 1." Nat Cell Biol 2(12): 893-898.

Yang, H., D. G. Rudge, J. D. Koos, B. Vaidialingam, H. J. Yang and N. P. Pavletich (2013). "mTOR kinase structure, mechanism and regulation." Nature 497(7448): 217-223.

Yang, Y. and S. K. Sharan (2003). "A simple two-step, 'hit and fix' method to generate subtle mutations in BACs using short denatured PCR fragments." Nucleic Acids Res 31(15): e80.

You, Z., C. Chahwan, J. Bailis, T. Hunter and P. Russell (2005). "ATM activation and its recruitment to damaged DNA require binding to the C terminus of Nbs1." Mol Cell Biol 25(13): 5363-5379.

Zha, S., C. H. Bassing, T. Sanda, J. W. Brush, H. Patel, P. H. Goff, M. M. Murphy, S. Tepsuporn, R. A. Gatti, A. T. Look and F. W. Alt (2010). "ATM-deficient thymic lymphoma is associated with aberrant tcrd rearrangement and gene amplification." J Exp Med 207(7): 1369- 1380.

112

Zha, S., C. Guo, C. Boboila, V. Oksenych, H. L. Cheng, Y. Zhang, D. R. Wesemann, G. Yuen, H. Patel, P. H. Goff, R. L. Dubois and F. W. Alt (2011). "ATM damage response and XLF repair factor are functionally redundant in joining DNA breaks." Nature 469(7329): 250-254.

Zha, S., W. Jiang, Y. Fujiwara, H. Patel, P. H. Goff, J. W. Brush, R. L. Dubois and F. W. Alt (2011). "Ataxia telangiectasia-mutated protein and DNA-dependent protein kinase have complementary V(D)J recombination functions." Proc Natl Acad Sci U S A 108(5): 2028-2033.

Zha, S., J. Sekiguchi, J. W. Brush, C. H. Bassing and F. W. Alt (2008). "Complementary functions of ATM and H2AX in development and suppression of genomic instability." Proc.Natl.Acad.Sci.U.S.A. 105(27): 9302-9306.

Zhang, J. and J. C. Walter (2014). "Mechanism and regulation of incisions during DNA interstrand cross-link repair." DNA Repair (Amst) 19: 135-142.

Zhang, S., H. Yajima, H. Huynh, J. Zheng, E. Callen, H. T. Chen, N. Wong, S. Bunting, Y. F. Lin, M. Li, K. J. Lee, M. Story, E. Gapud, B. P. Sleckman, A. Nussenzweig, C. C. Zhang, D. J. Chen and B. P. Chen (2011). "Congenital bone marrow failure in DNA-PKcs mutant mice associated with deficiencies in DNA repair." J Cell Biol 193(2): 295-305.

Zhu, C., K. D. Mills, D. O. Ferguson, C. Lee, J. Manis, J. Fleming, Y. Gao, C. C. Morton and F. W. Alt (2002). "Unrepaired DNA breaks in p53-deficient cells lead to oncogenic gene amplification subsequent to translocations." Cell 109(7): 811-821.

Ziv, Y., D. Bielopolski, Y. Galanty, C. Lukas, Y. Taya, D. C. Schultz, J. Lukas, S. Bekker- Jensen, J. Bartek and Y. Shiloh (2006). "Chromatin relaxation in response to DNA double- strand breaks is modulated by a novel ATM- and KAP-1 dependent pathway." Nat Cell Biol 8(8): 870-876.

Zullo, K., J. E. Amengual, O. A. O'Connor and L. Scotto (2012). "Murine models in mantle cell lymphoma." Best Pract Res Clin Haematol 25(2): 153-163.

113