MECHANISTIC LINK BETWEEN DNA DAMAGE RESPONSE SIGNALING AND IMMUNE ACTIVATION

APPROVED by SUPERVISORY COMMITTEE

Asaithamby Aroumougame, Ph.D.

Michael Story, Ph.D.

Jerry Shay, Ph.D.

Hesham Sadek, M.D.,Ph.D.

i

MECHANISTIC LINK BETWEEN DNA DAMAGE RESPONSE SIGNALING AND IMMUNE ACTIVATION

by

SOUPARNO BHATTACHARYA

THESIS Presented to the Faculty of the Graduate School of Biomedical Sciences The University of Texas Southwestern Medical Center at Dallas In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY The University of Texas Southwestern Medical Center at Dallas Dallas, Texas December 2018 ii

DEDICATION

This is dedicated to my precious wife, Shreya Endapally. Your love and words of great encouragement inspire me every day of my life. You are much more than a “wife” to me; you are my best friend and confidante.

iii

ACKNOWLEDGEMENTS

I would like to thank my mentor, Dr. Asaithamby Aroumougame for his patience, support and excellent guidance. The knowledge that I have acquired from working under his supervision is priceless.

I was also lucky to have distinguished scientists on my thesis committee. I appreciate their great suggestions and insight into my project.

I would like to acknowledge all the former and current members of Aroumougame lab for providing a truly cooperative and collaborative environment. Specifically, Drs. Salim Abdisalaam, Shibani Mukherjee and Fengtao Su. In addition, none of this work would have been possible without the assistance from scientists and staff at Brookhaven National Laboratory. Special thanks to the administrative staff of the Division of Molecular Radiation Biology (MRB), especially, Lisa Gray, Jennifer Staples, and Cheryl Hoppe. I also wish to thank Dr. David Chen, Dr. Anthony Davis, and Dr. Ward Wakeland for their cooperation in the use of their lab facilities.

To all my friends from UT Southwestern and the great bunch of wonderful people whom I met in Dallas, thank you for your emotional support. I know that I will have a friend in each one of you for life.

Finally, I am in debt to my wonderful parents, brother, and my in-laws. I wish you could see deep in my heart for its hard to find just the right words. Thank you for supporting me in every decision that I made and doing more than your best to help me achieve my aims every step of the way. I know that it was hard on you having me live halfway across the world, but I have no doubt that it was worth it. I strive to make all of you proud.

My beloved sister-in-law, Pooja, and my nephews, Roop, Tapur and Tupur, not a day has passed where I did not think about you. I just hope that my experience has instilled in you the importance of pursuing your dreams no matter how hard it may be.

iv

SELECTED PUBLICATIONS

Bhattacharya, S., Kalayarasan Srinivasan, Salim Abdisalaam, Fengtao Su, Edward K. Wakeland, Shibani Mukherjee and Aroumougame Asaithamby, Rad51 Interconnects between DNA Replication, DNA Repair and Immunity, Nucleic Acids Res. 2017 May 5;45(8):4590-4605. (PMID 28334891)

Bhattacharya, S., Salim Abdisalaam, Kalayarasan Srinivasan, Debapriya Sinha, Shibani Mukherjee and Aroumougame Asaithamby, NBS1 counters cCAS mediated premature Senescence (in revision)

Su, F., Bhattacharya, S., Abdisalaam S, Mukherjee S, Yajima H, Yang Y, Mishra R, Srinivasan K, Ghose S, Chen DJ, Yannone SM, Asaithamby A. et al., Replication stress induced site-specific phosphorylation targets WRN to the ubiquitin-proteasome pathway. Oncotarget, 2016. 7(1): p. 46-65. (PMID 26695548).

Su, F., Mukherjee S, Yang Y, Mori E, Bhattacharya, S., Kobayashi J, Yannone SM, Chen DJ, Asaithamby A. Nonenzymatic role for WRN in preserving nascent DNA strands after replication stress. Cell Rep, 2014. 9(4): p. 1387-401. (PMID 25456133)

Mukherjee, S., Abdisalaam, S., Bhattacharya, S., and Aroumougame Asaithamby, Mechanistic Link between DNA Damage Sensing, Repairing and Signaling Factors and Immune Signaling. Advances in Chemistry and Structural Biology (accepted; in press)

Bhattacharya, S., & Asaithamby, A., Repurposing DNA repair factors to eradicate tumor cells upon radiotherapy. Translational Cancer Research (2017).

Bhattacharya, S. & A. Asaithamby, Ionizing radiation and heart risks. Semin Cell Dev Biol, 2016. 58: p. 14-25. (PMID 26849909)

v

CONTENTS

Section Page List of figures and tables 6 Abbreviations 8 Abstract 11 Chapter 1 Introduction 1.1 Major DNA repair pathways 13 1.2 at the intersection of DDR and immune signaling 14 I Double Strand Break repair (DSBR) II Base excision repair (BER) 14 III Nucleotide excision repair (NER) 15 IV Interstrand crosslinks (ICL) 16 1.3 Mechanisms of cross talk between DDR and immune signaling 25 1.4 Consequences of DDR Mediated Immune signaling 28 1.5 Research goals Chapter 2 RAD51 Interconnects Between DNA Replication, DNA Repair and Immunity 2.1 Summary 34 2.2 Background 34 2.3 Materials and methods 36 2.4 Results 40 2.5 Discussion 49 2.6 Figures and tables 52 Chapter 2 NBS1 Counters cGAS-Mediated Premature Senescence 3.1 Summary 70 3.2 Background 70 3.3 Materials and methods 72 3.4 Results 77 3.5 Discussion 85 5.6 Figures & tables 89 Chapter 4 Conclusion and future perspectives 110 4.1 Conclusion 4.2 Future directions

Bibliography 116

vi

LIST OF FIGURES & TABLES

Page Chapter 1 Figure 1 Crosstalk between DDR and immune signaling Figure 2 Overview of research goals presented in the dissertation Table 1 Major DNA repair pathways

Chapter 2 Figure 3 Characterization of HT1080 FUCCI cells 52 Figure 4 Depletion of RAD51 up-regulates innate immune response pathway 53 Figure 5 Excessive DNA accumulates in the cytosol of RAD51-depleted cells 55 Figure 6 Accumulation of nuclear-derived DNA in the cytosol activates STING 57 in RAD51-depleted cells Figure 7 RAD51 blocks the MRE11-mediated degradation of nascent DNA 59 strands upon irradiation Figure 8 Inhibition of MRE11 activity blocks the expression of 61 innate immune response genes in RAD51-depleted cells Figure 9 RAD51 is critical for DSB repair in S/G2 cells and 62 stability maintenance Figure 10 Heat map of significantly altered genes in RAD51-proficient and - 63 depleted cells after radiation Figure 11 MRE11 deficient cells do not accumulate DNA in cytosol 64 Figure 12 RAD51-depleted cells are sensitive to STAT3 inhibitor 65 Figure 13 RAD51 interconnects between replication fork processing, DSB 66 repair, and innate immune responses Table 2 List of primers used for qRT-PCR in chapter 2 67 Table 3 Parameters of DSB dissolution kinetics in figure 7 68

Chapter 3 Figure 14 Preparation of stable cell lines 89 Figure 15 Telomeric DSBs trigger immune signaling and initiate premature 91 senescence Figure 16 Dysfunctional telomeres initiate innate immune signaling and 93 premature senescence phenotype Figure 17 cGAS recruitment is restricted to a fraction of cytosolic chromatin 95 fragments in response to dysfunctional telomeres Figure 18 NBS1 counters cGAS recruitment to cytosolic chromatin fragments in 98 response to dysfunctional telomeres

1 LIST OF FIGURES & TABLES

Figure 19 cGAS is responsible for the initiation of premature senescence 100 phenotype in response to dysfunctional telomeres Figure 20 Premature senescence is independent of cell type and telomere length 101

Figure 21 G2 to Mitotic phase progression in response to dysfunctional 102 telomeres is critical for micronuclei formation and premature senescence Figure 22 Condensed chromatin structure restricts cGAS binding to cytosolic 103 chromatin fragments: Figure 23 ATM plays a role in the binding of cGAS to cytosolic chromatin 104 fragments Figure 24 MRE11 is dispensable for cGAS recruitment to cytosolic chromatin 105 fragments Figure 25 Inhibition of CGAS binding to DNA attenuates inflammation 106 Figure 26 Induction of telomeric DSBs do not cause apoptosis 106 Figure 27 Graphical model of how telomere dysfunction induces premature 107 senescence Table 4 List of used in chapter 3 108 Table 5 List of plasmids and RT-PCR primers used in chapter 3 109

Chapter 4

Figure 28 RAD51 and BRCA1 deficiency results in activation of immune signaling in 114 response to genotoxic stress Figure 29 DN TRF2 expression in MC38 cells 114

Figure 30 Overview of m6ATracer-DAM-ID technique.

Figure 31 Simplified schematics of ongoing/ future experimental plans

2 ABBREVIATIONS USED

Abbreviations Full form 6-Thio-DG 6-thio-2'-deoxyguanosine 8-OxoG 8-Oxoguanine AGS Aicardi–Goutieres syndrome AIM2 Absent in melanoma APC Antigen presenting cells ATLD Ataxia-telangiectasia-like disorder ATM ataxia-telangiectasia mutated ATR ataxia telangiectasia and Rad3-related protein BER Base excision repair BRCA 1, 2 Breast cancer susceptibility 1 and 2 BSA Bovine serum albumin CAD caspase activated DNase CARD9 Caspase recruitment domain-containing protein 9 CDK Cyclin-dependent Kinase CDKN2A cyclin-dependent kinase inhibitor 2A cGAMP Cyclic guanosine monophosphate–adenosine monophosphate cGAS Cyclic GMP-AMP synthase CHK1, 2 Checkpoint kinase 1, 2 COFS Cerebro-Oculo-Facio-Skeletal syndrome COS CPT Camptothecin CSB Cockayne syndrome B CSK Cytoskeletal DAI DNA-dependent activator of IFN-regulatory factors DAM DNA-adenine-methyltransferase DAM-ID DNA adenine methyltransferase identification DC Dendritic cell DD Degradation domain DDR DNA damage response DDX41 DEAD-box 41 DN Dominant Negative DNA Deoxy ribonucleic acid DNA PK DNA-dependent protein kinase DNA PKcs DNA-dependent protein kinase catalytic subunit DR Direct repair DSB Double-strand break DTT Dithiothreitol

3 Abbreviations Full form EME 1, 2 Essential meiotic 1, 2 ER Estrogen receptor ERCC4 Excision repair cross complementation group 4 Exo1 Exonuclease1 FA Fanconi anaemia H2AX H2A histone family member X HER2 Human epidermal growth factor receptor 2 HJ Holliday junctions HRR Homologus recombination repair HZE High charge (Z) and high energy (E) hTERT Human reverse transcriptase HU Hydroxyurea IFI-16 gamma inducible protein 16 IFN Interferon IFNAR The interferon-α/β receptor IKK IkB kinase IL Interleukin iNOS Inducible nitric oxide synthase IRF IFN-regulatory factor LET Linear energy transfer MDC1 Mediator of DNA damage checkpoint protein 1 MGMT O6-methylguanine-DNA-methyltransferase MLH1 MutL homolog 1 MMR Mismatch repair MRE11 Meiotic Recombination 11 Homolog A MRN MRE11-RAD50-NBS1 complex MSH2 mutS homolog 2 MUS81 MMS and UV-sensitive Protein 81 NBS1 Nijmegen breakage syndrome gene1 NEMO NF-kB essential modulator NER Nucleotide excision repair NF-kB Nuclear factor kappa B NF-kB Nuclear factor kappa-light-chain-enhancer of activated B cells NHEJ non-homologous end-joining NK Natural killer OGG1 8-Oxoguanine DNA glycosylase-1 PARP poly(ADP-ribose) PBS Phosphate buffer saline PCNA Proliferating cell nuclear antigen PFA Paraformaldehyde

4 PH3 Phospho histone 3 Abbreviations Full form PMSF Phenylmethylsulfonyl fluoride Pol Polymerase POT1 Protection of Telomeres 1 PR Progesterone receptor RAP1 Repressor/activator protein 1 RNA ribonucleic acid ROS Reactive oxygen species RPA SARD Systemic autoimmune rheumatic disease SASP Senescence-associated secretory phenotype SEM Standard Error of Mean sg RNA Small guide RNA sh RNA Short hairpin RNA SLE Systemic lupus erythematosus SSB Single strand breaks STAT Signal transducer and activator of transcription STDEV Standard Deviation STING Stimulator of interferon genes TBK1 Tank binding kinase 1 TFIIH Transcription factor II H TIN2 TRF1-interacting nuclear protein 2 TLR Toll like receptor TMEM173 Transmembrane Protein 173 TNF Tumor necrosis factor TREX1 Three primerepairexonuclease1 TRF1, 2 Telomeric Repeat Factor 1, 2 WRN Werner XLF XRCC4-like factor XP XRCC4 X-ray repair cross-complementing protein 4

5

MECHANISTIC LINK BETWEEN DNA DAMAGE RESPONSE SIGNALING AND IMMUNE ACTIVATION

Souparno Bhattacharya, M.S. The University of Texas Southwestern Medical Center at Dallas, 2018

Asaithamby Aromougame, Ph.D.

Abstract:

Proper maintenance of an intact genome is crucial for cellular homeostasis. To combat threats posed by DNA damage, cells have evolved sophisticated mechanisms – collectively termed as the DNA-damage response (DDR) signaling –, which detect DNA lesions, signal their presence, and promote their repair. Contribution of proper DDR signaling in not just confined to prevention of genomic instability and , as emerging evidence indicates crosstalk exists at different levels between DDR signaling machinery and our immune system. In my dissertation work, using innovative models and techniques, I deciphered how RAD51, a protein normally associated with repair and replication of DNA, regulates innate immune response. Besides detection and repair of damaged DNA, proper DDR signaling also enables checkpoint activation, which prevents cell cycle progression with unrepaired DNA lesions. In my thesis work, I have proved how failure to arrest cells in the G2-M boundary after genotoxic stress, leads to generation of micronuclei, present in the cytoplasm and subsequent immune activation. Work emanating from my thesis projects will add to the growing body of literature showing how different DDR factors’ roles in modulating immune signaling are most often a consequence of their inherent ability to sense, repair and signal in response to DNA damage. Finally, our improving understanding of DDR has already provided new avenues for disease management (e.g. Use of PARP inhibitors in treating BRCA mutant tumors). A more precise understanding of mechanisms by which DDR factors are involved in regulation of cellular immunity can also be exploited to redirect the immune system for both preventing and treating variety of human pathologies including cancer, autoimmune diseases and age related disorders.

6

Chapter 1

Introduction

7 1.1 Major DNA repair pathways:

DNA damage is a biological process that affects human health in many ways. Eukaryotic cells accrue DNA damage as result of endogenous DNA metabolic activities, such as replication and recombination events, or environmental exposures to ionizing radiation, ultraviolet light and chemical mutagens. It is estimated that this damage load may amount to 104–105 DNA lesions per mammalian cell per day 1. DNA lesions, if not repaired properly, can lead to cytotoxic, mutagenic, and carcinogenic effects 2-4. Cells respond to DNA breaks by initiating a series of signaling pathways collectively known as DNA damage response (DDR) signaling. DDR contributes to the timely repair of DNA breaks, transient cell cycle arrest, transcriptional and post-transcriptional activation of a wide array of genes, and, under certain circumstances, programmed cell death. Cancer is the leading cause of morbidity and mortality worldwide. According to recent estimation by the International Agency for Research on Cancer (IARC), in 2012, 8.2 million people worldwide died of cancer, with 14.1 million new cases diagnosed 5. Deficiencies in DNA repair factors and pathways are a feature common to many cancers 6. Germline mutations in DNA repair genes, such as ataxia telangiectasia mutated (ATM), Nibrin (NBS1), (WRN), (FA), BRCA1, and BRCA2, give rise to cancer-prone disease syndromes 7-11. In addition to germ line mutations in DNA repair genes, many cancers show impaired DDR and defective checkpoints either because of somatic mutations (ATM, TP53, and cyclin-dependent kinase inhibitor 2A (CDKN2A), BRCA1/2 being the most frequently mutated genes) or because of aberrant expression of DDR factors 12. Furthermore, DNA replication stress as a result of oncogene activation can also cause constitutive activation of the DDR and tumor progression 13. Faithful repair of DNA lesions is the single most important step in DDR signaling and is critical for cellular survival and prevention of genomic instability. DNA lesions, depending on the damage source, are of different types and include base modifications, helix distortions, single-strand breaks (SSBs), and double-strand breaks (DSBs) 15. For example, 1 Gy of γ-radiation induces around 850 pyrimidine lesions, 450 purine lesions, 1,000 SSBs, and 20–40 DSBs in a mammalian cell 16. To counter this, cells have developed distinct yet overlapping mechanisms for repairing different types of DNA lesions (Table 1). Non-homologous end joining (NHEJ) and homologous recombination (HR) repair DSBs. Mismatch repair (MMR) corrects replication related errors while base excision repair and nucleotide excision repair pathways reverses base modifications and helix distorting lesions, respectively.

8 1.2 At the intersection of DDR and immune signaling

The immune system, comprising of innate and adaptive immune response, is responsible for protecting organisms against extracellular and intracellular pathogens, but also against damaged cells. DNA breaks generated during viral integration in host genome or lytic replication triggers an inflammatory response 17. IFI-16 and IRF1, commonly associated with host antiviral response, are also involved in maintenance of genomic stability, through interacting with BRCA1 or regulating expression of Fanconi Anemia J (FA J), respectively 17,18. These results suggest there exists a cross talk between DDR and immune signaling. In the last decade, tremendous progress have been made in identifying how proteins involved in DNA damage sensing, repairing or other DDR signaling events actively engage in regulation of immune signaling. This crosstalk between DDR and immune signaling in important not only in protecting hosts from pathogens but also in maintenance of cellular homeostasis. Different components of the DDR including DNA damage sensors, transducer kinases, and effectors are involved in regulation of immune signaling. In the following section, first, I will discuss some of the key proteins, both DNA damage sensors and DNA repair proteins, which are involved in immune signaling (Fig1). Secondly, I will discuss four broad mechanisms by which DDR signaling can influence inflammation and immune response.

I. Double Strand Break repair (DSBR)

DSBs are the most harmful type of DNA damage incurred by the cells. In addition to exposure to radiation or chemical agents, ROS produced as metabolic by products can also induce DSBS. In mammalian cells, two main pathways repair DSB: NHEJ and HRR. a. MRN complex: In both NHEJ and HRR , the Mre11-Rad50-NBS1 (MRN) complex plays important roles in detection and signaling of DSBs. MRE11 performs endo- and exo- activities against a variety of ss (single strand) DNA and ds (doubsle strand) DNA substrates 19-22. MRE11 participates in nuclease activities required in NHEJ and HR mediated repair of DSBs and have multiple reported roles in regulation of immune signaling. MRE11 physically senses endogenous ds-DNA in the cytoplasm, resulting in the activation of STING and IRF3 and production of type I IFN 23. MRE11 also functions at perturbed replication forks after DNA damage and evidence shows that lack of BRCA1/2, FA factors, RAD51, WRN and XPG1 leads to MRE11-mediated excessive processing of newly replicated genome and MRE11- mediated degradation of newly replicated genome contributes to the accumulation of fragmented self-DNA in the cytoplasm, culminating in the activation of STING-mediated innate immune signaling 24. Thus,

9 MRE11 plays important roles not only in directly recognizing exogenous dsDNA but also in releasing self- DNA into the cytoplasm, resulting in the initiation of STING-dependent signaling. RAD50, the second component of MRN complex, is also involved in regulating immune activation. Upon encountering cytosolic dsDNA, RAD50 form a complex with adaptor protein CARD9 in dendritic cells, which are the main APCs responsible for activating adaptive immune system 25. Subsequently, formation of dsDNA–RAD50–CARD9 complex formation activates NF-κβ pathway, resulting in increased IL-1 production 25. The final component of MRN complex is NBS1. NBS1 functions as a DNA damage sensor and signal transducer in response to DNA damage. NBS1 deficiency results in Nijmegen breakage syndrome (NBS), which is associated with immunodeficiency, microcephaly, growth retardation, high frequency of lymphoid malignancies and an aging phenotype 26. Recent studies have highlighted the role of NBS1 in immune signaling. Mice with hypomorphic Nbs1 alleles exhibit an impaired inflammatory response 27. Another study found that loss of NBS1 leads to exacerbated inflammation 28. Furthermore, recent work from my lab revealed that NBS1 regulates immune signaling signaling by counteracting cGAS binding to cytosolic chromatin fragments. NBS1 lacks any known enzymatic activities, but its physical presence on the cytosolic chromatin fragments suffices to suppress the cGAS-mediated cytosolic DNA sensing pathway. I will discuss the results of this study in further detail in the chapter 3. Efforts in our lab are underway to identify the mechanism of how NBS1 counters cGAS binding to cytosolic chromatin fragments.

b. ATM and ATR kinase: Recognition of DSBs by MRN complex is crucial for recruiting and catalyzing auto-phosphorylation of the ATM (ataxia telangiectasia-mutated) kinase 29. Phosphorylated ATM then leads to the activation, phosphorylation and amplification of a cascade of downstream effector proteins, , such as H2AX, MDC1, BRCA1, Chk2 and , which leads to DNA repair, cell cycle arrest, apoptosis and other downstream processes in the DDR 30. The ATR kinase pathway is activated in response to single- strand DNA breaks but can augment ATM activity and is mediated through the Chk1 kinase 31. In the context of immune signaling, ATM has multiple roles. ATM is essential for immunoglobulin and T cell receptor gene rearrangements during development 32 and ATM deficiency results in abnormal T and B lymphocyte counts and deficient responses 33. Activated ATM associates with NEMO, the regulatory subunit of IkB kinase (IKK) and triggers NF-κB-dependent gene expression. In humans, loss of functional ATM results in Ataxia -telangiectasia (AT), a complex cancer-prone neurodegenerative disease. Individuals with AT fail to mount sufficient antibacterial immune response due to defects in their adaptive immune system 34. Mechanistically, increased oxidative stress in face of ATM deficiency inhibits inflammasome activation, which results in diminished production of IL1β, making atm -/- mice and AT

10 patients susceptible to bacterial infections. In contrast, AT patients are able to mount quick antiviral immune response and often develop autoimmune syndromes. Mechanistically, unrepaired DNA lesions accruing from loss of ATM cause genomic DNA fragments to leak into the cytoplasm, where they activate innate immune sensors. This results in an enhanced innate resistance to viral infection and a hyper-responsiveness to a variety of innate immune stimuli 35. The involvement of ATR kinase, activated by presence of single stranded DNA overhangs in the nucleus, in immune signaling is limited. Recent evidence suggests early phases Vaccinia virus 36 and Herpes Simplex Virus 1 (HSV-1) 37 infection involves activation of cytoplasmic ATR and recruitment of Chk1 and other components of canonical ATR pathways, to cytoplasmic DNA viral factories. Furthermore, inhibition of ATR kinase with VE-821 reduces accumulation of cytoplasmic dsDNA after DNA damage, suggesting a role of ATR mediated DDR signaling in cytosolic DNA mediated immune response 38. c. Non-Homologous End Joining (NHEJ) factors: NHEJ is the predominant DSB repair pathway in G0/G1 cells but remains active throughout the cell cycle. The main goal of NHEJ-mediated repair is fast rejoining of two broken DNA ends with minimal end processing, regardless of . However, NHEJ-mediated repair results in the loss of up to 20 nucleotides from either side of the DSB junction, making it error-prone 39. Classical NHEJ-mediated DNA repair consists of four steps: 1. Recognition and binding of broken DNA ends by Ku70/Ku80 heterodimer; 2. Recruitment and stabilization of the NHEJ factors like DNA-PK catalytic subunit (DNA-PKcs) and Artemis at the DSB for bridging the DNA ends; 3. Processing of the DNA ends to make them suitable for ligation, which is mediated by different exo- and endo-; 4. Synthesis of new nucleotides by DNA µ and λ; and finally, 5. ligation of the two severed DNA ends by ligase IV/XRCC4/XLF complex 40. DNA-PK is a member of the PIKK family of protein kinases and is expressed in almost all mammalian cells 41.. DNA-PK forms a functional complex with KU70 and KU80 42, which is essential for the NHEJ-mediated repair of DSBs. Besides its function in DSB repair, DNA-PKcs also phosphorylates a wide variety of proteins that function in various cellular processes. DNA-PK complex perform critical functions in development of adaptive immune system. DNA-PKcs is required for V(D)J recombination, which utilizes the NHEJ pathway to promote antigen diversity in the mammalian immune system and germline mutations targeting DNA-PKcs lead to severe combined immunodeficiency (SCID) 43. Multiple reports suggest that DNA-PKcs is also critical for mounting innate immune responses to DNA viruses in fibroblasts 44-46. Mechanistically, DNA-PK binds to pathogen derived DNA in cytoplasm, resulting in IRF-3–mediated transcription of multiple cytokine and chemokine genes independently of DNA-PKcs kinase activity 44. Furthermore, KU70, an integral component of DNA PK

11 complex induces acts as a cytosolic DNA sensor and induces IFN-λ1 production by activating IRF1 and IRF7 in response to cytosolic DNA 45,47. d. Homologous Recombination Repair (HRR) factors: HRR is the predominant DSB repair pathway in the S and G2 phases of the cell cycle, where a homologous sequence of the sister chromatid acts as a template for DNA synthesis and gap filling in an error-free manner. Unrepaired damaged bases and SSBs that are converted into DSBs during replication and require HRR to repair 48,49. Additionally, evidence suggests that the RAD51-mediated HRR pathway is critical for processing the complex DSBs induced by charged particles, including protons and carbon (12C) ions 50-52. HRR consists of three steps: first, during the pre-synaptic step, DSBs are recognized and resected to generate single-strand DNA overhangs. RPA binding to these overhangs prevent degradation and self-annealing. RAD51 and other associated proteins then bind to the DNA ends. Next, during the synaptic step, the resultant nucleoprotein filament invades the sister chromatid or homologous chromosome in search of homologous regions to form heteroduplex DNA. New nucleotides are added to the 3’ end of the invading strand by polymerase η using the sister strand as a template. An extended D-loop is formed which captures the second end of the break resulting in formation of Holliday junctions. Once formed, the HJ undergo branch migration generating varying lengths of heteroduplex DNA. The final post-synaptic step involves resolving the HJ and migration of repaired ends migrate toward each other to restore the duplex DNA 53. RAD51 plays central roles in DSB repair and replication fork processing. RAD51, as a recombinase, catalyzes the core reactions of homologous recombination mediated repair of DSBs 54. Apart from its DSB repair functions, RAD51 is also essential for replication fork restart when a replication fork encounters DNA damage 55. In doing so, RAD51 prevents accumulation of replication-associated DSBs 56 and maintain genome stability. Germ-line mutations in the RAD51 lead to embryonic death 57. Two recent studies indicate RAD51 is involved in suppressing DNA damage mediated immune response. A study by Wolf et al. found that RAD51 along with RPA prevents the release of short nuclear DNA fragments into the cytosol by binding to these DNA fragments, thereby preventing unwanted immune signaling 58. Recently, our group have identified a role for RAD51 in suppressing innate immune signaling in response to radiation induced DNA damage 24. RAD51 is recruited to the sites of perturbed replication forks and DSBs, which blocks the excess exonuclease activity of MRE11 on the newly replicated genome and on DSB repair, respectively. Consequently, this limits the accumulation of self-DNA in the cytosol and prevents the initiation of STING-mediated innate immune signaling. I will discuss the findings of this study in detail in chapter 2. BRCA1/2 are multifunctional proteins that physically interacts, both directly and indirectly, with specific partner proteins including RAD51 59. BRCA1/2 are critical for maintaining genome stability largely

12 because of its role in HR-mediated DSB repair and replication fork stability. In the absence of BRCA1, cells repair DSBs by NHEJ, resulting in increased chromosomal instability and genomic alterations 60. Importantly, BRCA1/2 deficiency interferes with RAD51 function; causing MRE11-mediated degradation of newly replicated genome 7,61,62. The knowledge regarding involvement of BRCA1/2 in immune signaling is currently limited. BRCA1-mutated breast tumors harbor a significantly higher number of CD3+ and CD8+ tumor infiltrating lymphocytes (TILs), as well as higher expression of PD-1 and PD-L1 in tumor-associated immune cells, than BRCA1-proficient tumors 63,64. Additionally, BRCA1 plays a hitherto unidentified role as a cofactor to IFI16 in the nuclear innate sensing of foreign DNA and subsequent assembly and cytoplasmic distribution of stable IFI16-inflammasomes leading into IL-1β production, as well as the induction of IFN-β via cytoplasmic signaling through IFI16, STING, TBK1 and IRF3 65. Another study has identified activation of STING-mediated chemokine production in response to endogenous or exogenous S-phase specific DNA damage in vitro, resulting in an inflammatory microenvironment in BRCA1-mutant breast tumors 66. However, the initial events responsible for triggering immune signaling in BRCA1-mutant tumors remain to be identified. BRCA1 regulates key effectors that control the G2/M cell cycle checkpoint and is involved in regulating the onset of mitosis 67 and preliminary results in our lab shows deficiency in BRCA proteins result in formation of cytosolic chromatin fragments during mitosis, which can trigger cGAS-STING mediated immune response (unpublished data). The XPF/MUS81 endonuclease plays a role in resolving recombination intermediates during DNA repair after inter-strand cross-links, replication fork collapse and DSBs. The encoded protein associates with one of two closely related endonuclease (EME1 or EME2) to form a complex that processes DNA secondary structures. MUS81 suppresses chromosomal instability arising from stalled replication forks by cleaving potentially detrimental DNA structures 68. Homozygous mus81-/- mice have significant meiotic defects, including failure to repair a subset of DSBs and develops spontaneous tumors. Ho et al. found that, like MRE11, MUS81 mediated cleavage of genomic DNA results in accumulation of fragmented self-DNA in the cytoplasm, including non-B DNA structures, repetitive sequences, DNA lesions, R-loops and common fragile sites. The resultant fragmented DNA leads to STING-dependent expression of type I IFNs and chemokines 69 and results in heightened host immune response against tumor cells.

II. Base Excision repair (BER)

BER corrects SSBs and non-helix-distorting small base lesions that arise from oxidation, deamination and alkylation. There are two sub pathways: short-patch BER, where only one nucleotide is replaced, usually takes place during the G1 phase; and long-patch BER, where 2–13 nucleotides are replaced in the S/G2 phase of the cell cycle. The repair mechanism consists of five key enzymatic steps that involve excision of

13 the damaged base by DNA glycosylase, nicking of the damaged DNA strand by AP endonuclease, removal of the sugar fragment by , filling of the gap by DNA polymerase, and finally, sealing of the nick by DNA ligase 70,71. PARP1 acts as a sensor for base damage. It is also responsible for recruiting additional repair factors to damaged bases. PARP-1 regulates the expression of NF-κB-dependent proinflammatory mediators such as TNFα, IL-6, and iNOS by inducing the translocation of NF-κB into the nucleus upon DNA damage 72. Once damaged bases are recognized, DNA glyocosylases excise damaged bases. One such glycosylase, OGG1, positively regulates expression, nuclear translocation of NF-κB and subsequent production of inflammatory cytokines and chemokines 73. After removal of damaged bases by DNA glycosylases, AP create nicks in the phosphodiester backbone. Apurinic/ apyrimidinic endonuclease 1 (APE1) accounts for over 95% of the total AP endonuclease activity in human cell lines and, besides DNA repair, also performs functions such as redox regulation (mediated through a separate redox domain) and transcriptional regulation of other genes. APE1 is also involved in regulation of multiple transcription factors involved in the pro inflammatory NF-κB pathway and this APE1 mediated activation of NF-κB pathways is important for production of inflammatory cytokines like TNF-α and IL-8 74-76.

III. Nucleotide excision repair (NER)

NER is the main pathway responsible for the removal of bulky DNA lesions induced by UV irradiation, chemical adducts, intrastrand crosslinks caused by drugs such as cisplatin, and ROS-generated cyclopurines 77. NER has two sub-pathways, global genomic NER (GG-NER), which can occur anywhere in the genome and transcription coupled NER (TC-NER), which prioritizes repair of lesions in the transcribed strand of active genes 78. The first step in the GC-NER process is damage sensing by XPC, further stabilized by its association with two other proteins, RAD23B and centrin2. This is followed by recruitment of UV-DDB to the damage site. In case of TC-NER, RNA polymerase stalled at a lesion in collaboration with TC-NER specific factors CSA, CSB, and XAB2 initiates the process 78. Following damage recognition, TFIIH (transcription initiation factor IIH) complex is recruited to the lesion sites. The XPA subunit of TFIIH complex unwinds DNA helix in 5’-3’ direction and is important for damage verification step. XPA, together with RPA, then recruits other proteins to the damage site, including two endonucleases (XPG and XPF- ERCC1). XPF creates an incision 5′ to the lesion and XPG cuts the damaged strand 3’ to the lesion, which excises the lesion within a 22–30 nucleotide-long strand. Following the 5’ incision, Pol δ, Pol κ or Pol ɛ are recruited for gap-filling DNA synthesis, using the intact complementary strand as a template77. Deficiencies in NER pathway results in rare autosomal disease like Xeroderma pigmentosum (XP), Cockayne syndrome (CS), and (TTD), which often has increased cancer risk 79. Furthermore, germline

14 mutation in NER related ERCC genes cause features of premature aging 80. Recent studies have found tantalizing evidence linking NER pathway with immune activation. Insertion of a PGBD3 piggyBac transposon in the intron of CSB gene, which is involved in TC-NER, results in formation of an additional splice variant CSB-PGBD3 81. Expression of CSB-PGDB3 alone in CSB null cells results in excess production of type I IFN and co expression of CSB and CSB-PGBD3 in necessary for maintaining homeostasis 82. Furthermore, patients carrying mutations in NER factors CSB, XPB and XPD mount a constitutive type-I IFN response and shows clinical features comparable with those seen in the autoimmune disorder AGS 83 84. These results along with conserved nature of PGBD3 transposase open reading frame in all primates indicates precise activation of NER pathways is necessary for antiviral response as well as prevention of self-DNA mediated autoimmunity.

IV. Interstrand crosslinks (ICL)

Exposure to alkylating agents like mitomycin C results in covalent linkage between two DNA strands, commonly known as ICL. ICLs are particularly toxic because they involve both strands of DNA, blocking the essential metabolic processes like replication and transcription, which requires translocation along the DNA. Repair of ICLs are particularly problematic for cells, as it requires repairing the damage on both DNA strands 85. Defects in ICL repair results in chromosome instability syndromes, of which Fanconi anemia (FA) is most well studied. FA is characterized by congenital abnormalities, progressive bone marrow failure and a highly elevated risk of hematological and squamous cell cancer 86. In addition to ICL repair, FA family of proteins play roles in replication, recombination, repair and recovery in response to a variety of DNA damaging agents. A central event in the FA pathway is the mono-ubiquitination of FANCD2 and FANCI upon DNA damage. This modification is mediated by the FA core complex, which consists of eight proteins (FANCA, B, C, E, F, G, L and M) 87,88. Once ubiquitinated, FANCD2 and FANCI functionally interact with downstream FA proteins such as FANCD1/BRCA2, FANCN/PALB2, FANCJ/BRIP1, FANCP/SLX4 and RAD51C, and their associated protein, BRCA1 89-92. Additionally, The FA pathway is critical for repairing clustered DSBs specifically in the S/G2 phase of the cell cycle, mainly because of defective RPA2 and RAD51 recruitment to the sites of DSBs 93. In addition, replication fork processing is also defective in the absence of FANCD2. Consequently, FANCD2-deficient cells exhibit high levels of anaphase bridge formation and defective chromosome segregation. Additionally, like BRCA1,2 and RAD51 the FA pathway is critical for blocking MRE11-mediated degradation of replication forks 62.

Recent evidence suggests that FA genes function in the selective autophagy of genetically distinct viruses, in mitochondrial quality control and in preventing inflammasome activation is response to

15 mitochondrial ROS 94. In vitro stimulated FA bone marrow shows elevated levels TNFα and IFN-γ 95. In our lab, we are currently evaluating the mechanism by which FA proteins regulate immune signaling.

1.3 Mechanisms of cross talk between DDR and immune signaling

The innate immune system employs pattern recognition receptors (PRRs) as first line of host defense, as they recognize the various pathogen- or danger-associated molecular patterns like self and foreign nucleic acid in the cytoplasm. Once receptor-ligand engagement takes place, cells mount an immune response, by expressing pro-inflammatory cytokines such as IL-1β, TNFα, type I IFN or upregulating ligands like NKG2D, which expressed on NK cells 96. Presence of self or foreign nucleic acid in the cytoplasm, both RNA and DNA, are sensed by specific PRRs, which upon binding activate downstream signaling events. For example, TLR9 specifically recognizes bacterial DNA and synthetic oligodeoxynucleotides, which contain unmethylated CpG dinucleotides 97. Once activated, TLR9 translocate from ER to endosomes and start downstream signaling cascade NF-κB and IFN-regulatory factor (IRF) 7, and induce TNF-α, IL-1, IL- 6 and type I IFN production 98,99. In contrast to endosomal DNA, cytosolic DNA sensing involves a different group of PRRs. Over the years, multiple cytoplasmic DNA sensors have been reported including, DAI 100, AIM2 101-103, RNA polymerase III 104, IFI-16 105, DDX41 106, Ku70 45, DNA-PK 44, MRE11 23 and cGAS 107. Among the above-mentioned factors, cGAS is considered to be the most important. cGAS catalyzes the formation of 2'-5'-cGAMP, an atypical cyclic di-nucleotide second messenger that binds and activates stimulator of interferon genes (STING), resulting in the recruitment of TBK1, the activation of the transcription factor IRF3 and the trans-activation of innate immune response genes, including type I Interferon cytokines (IFN-I) 107-109. Another key factor is STING, encoded by the TMEM173 gene, a signaling molecule associated with the endoplasmic reticulum (ER) that is essential for controlling the transcription of type I IFNs and pro-inflammatory cytokines. STING, which is downstream of cGAS and many other cytoplasmic DNA sensors, is phosphorylated and forms intracellular clusters after recognizing 2'-5'-cGAMP released by cGAS 110,111. The regulated activation of the cGAS-STING–mediated cytosolic DNA sensing pathway is essential for antiviral as well as DNA damage mediated immune response. Pre- clinical studies have found that dying tumor cells exposed to DNA damage accumulate cytosolic DNA, which activates the cGAS-STING pathway 112,113. STING activation is critical for production of type I IFN, which in turn recruits Dendritic cells (DC) to the tumor. Activated DCs act as APCs and are capable of presenting tumor antigens to T cells, which primes the T cells to selectively kill tumor cells 112,113. As discussed above, presence of DNA in unusual locations, such as the cytoplasm or the endosomes, can trigger immune response, as DNA is normally located in the nucleus of eukaryotic cells 114-117. The DNA can be foreign in origin, like in case of viral infections, or it can be the organism’s self-

16 DNA generated by both endogenous and exogenous genotoxic stress. Below are some of the major mechanisms by which defective DNA repair, replication events or DDR signaling contribute to the accumulation, sensing and removal of cytosolic DNA, which in turn regulate immune signaling (Figure 2). a. Direct Sensing of DNA in the Cytoplasm: The role of DNA damage sensors in detecting aberrant DNA structures is not only restricted to the nucleus. Though most DDR factors are localized in the nucleus, a small portion of these factors resides in the cytoplasm. For example, cytoplasmic MRE11 senses dsDNA in the cytoplasm, leading to the activation of a STING-dependent type 1 IFN response 23. KU70 recognizes cytosolic DNA and induces the production of IFN-λ1 (a member of Type-III IFN) rather than Type-I IFN. This induction is mediated via the activation of IFN regulatory factor IRF-1 and IRF-7 45. Similarly, DNA- PK binds to cytoplasmic DNA, resulting in STING-TBK1–mediated activation of type I IFN and cytokine and chemokine genes 44. So, DNA damage sensors can also recognize foreign and self-DNA in the cytosol and activate downstream cytosolic DNA sensing pathway-mediated immune signaling. b. Degradation of Replication Forks by Nucleases: Replication fork instability due to nuclease- mediated degradation of newly replicated genomic DNA can contribute to self-DNA accumulation in the cytoplasm. Recent evidence suggests that defects in BRCA1/2, FANCA and WRN lead to MRE11- mediated degradation of newly replicated genomic DNA 7,61,62. Our group has previously identified MRE11-mediated processing of newly replicated genome in the absence of RAD51 leads to the accumulation of genomic DNA in the cytoplasm 24. Similarly, MUS81 causes the accumulation of fragmented non-B DNA structures, repetitive sequences, DNA lesions, R-loops and common fragile sites in the cytoplasm in response to replication stress 69. Accumulation fragmented self-DNA in the cytoplasm triggers STING-mediated innate immune signaling. Thus, defects in factors that regulate the enzymatic activities of MRE11 and MUS81 in response to genotoxic stress lead to self-DNA accumulation, resulting in the activation of the cytosolic DNA sensing pathway-mediated immune signaling pathway. Recent reports also suggest (BLM) helicase and exonuclease 1 (EXO1), two other proteins involved in DNA resection during HR mediated DSB repair process contributes to generation of cytosolic single strand DNA fragments, which results in activated type I IFN signaling 118. c. Failure to Retain Fragmented Self-DNA within the Nucleus: Cells can produce fragmented DNA within the nucleus as a byproduct of DNA replication and repair in response to both exogenous and endogenous genotoxic stress. Multiple DDR factors bind to these fragmented DNA because of their inherent non-DNA sequence-specific interaction characteristics. As a result, the DNA fragments are retained within the nuclei, preventing their export or release into the cytoplasm. For example, RPA2 and

17 RAD51 prevent the release of short nuclear DNA fragments into the cytosol by binding to these fragments 58. Defects in DNA binding factors within the nucleus or excess production of fragmented DNA, lead to self-DNA accumulating in the cytosol, culminating in the activation of cytosolic DNA sensing pathway-mediated immune signaling. d. Generation of Chromatin Fragments Due to Defective G2/M Checkpoint: Besides DNA repair, activation of cell cycle checkpoints is a vital part in the DDR signaling. The cell cycle in eukaryotic cells consists of four phases, gap (G1), synthesis (S), G2, and mitosis (M), and one phase outside the cell cycle, G0. DNA damage checkpoints are defined as pauses in the cell cycle in response to DNA damage. These temporary pauses allow the damage to be repaired before cell division resumes. Proteins like CHK1 and CHK2 that accumulate at the damage site typically activate the checkpoint and halt cell growth at the G1/S or G2/M boundaries.

Activation of G2/M checkpoint in response to DNA damage results in the timely processing of damaged DNA before mitosis. However, if release from G2 arrest occurs before the DNA damage repair is complete or progression of cells with fused chromosome from G2 to mitotic phase, micronuclei will form upon cytokinesis. Recent findings have provided mechanistic insights into how the accumulation of chromatin fragments in the form of micronuclei and the subsequent rupture of the nuclear envelope initiate the cGAS-STING–mediated cytosolic DNA sensing pathway 119-124. Deficiencies in DDR factors, including ATM and BRCA1/2, with known functions in cell-cycle checkpoint activation in response to genotoxic stress lead to chromatin fragment accumulation in the cytoplasm. Therefore, defective DSB repair together with checkpoint release before the completion of DNA repair contribute to the formation of micronuclei and the activation of the cGAS-mediated cytosolic DNA sensing pathway. e. Failure to clear fragmented DNA from cytoplasm: To prevent unwarranted activation of cytoplasmic nucleic acid sensing pathway and subsequent immune activation, cells employ fast and effective to destroy endogenous cytoplasmic DNA, coming from pathogen infection, apoptosis or DNA replication. Among them, DNase II is one of most studied candidate that degrades DNA derived from pathogens or apoptotic cells following engulfment by macrophages. Indeed, deficiency in caspase activated DNase (CAD) and DNase II (DNase II-/-CAD-/-) results in accumulation of DNA in the thymus of mouse embryo, resulting in heightened IFN-β production 125. TREX1 is another major exonuclease in mammalian cytoplasm, and it acts on both ssDNA 126,127 and dsDNA 128. Originally thought to be a proofreading for DNA polymerases α and β, TREX1’s 3’→5’ DNA exonuclease function is essential for clearing fragments of DNA from the cytoplasm 116,129. The major function of TREX1 is to maintain the host’s innate immune tolerance to self-DNA as DNA

18 accumulation in TREX1-deficient cells due to viral infection or as a byproduct of DNA repair stimulates an immune response 116,129. TREX1-deficient mice exhibit profound systemic inflammation affecting multiple organs, elevated autoantibody production and inflammatory myocarditis early in age 116,130. Mutations that disrupt the mouse’s TREX1 DNase activity lead to the accumulation of self-DNA in the cytosol, which activates the cGAS-STING–mediated type I IFN response and systemic inflammation. The inflammation and mortality of TREX1-null mice can be genetically rescued by depleting cGas−/−, sting−/−, ifnar1−/−, or irf3−/− 116,131-134. Additionally, pharmacological inhibition of TBK1, a key serine/threonine kinase that phosphorylates STING and IRF3, alleviates autoimmune disease phenotypes and increases overall survival of TREX1−/−mice 135. Clinically, carriers of TREX1 Mutations develop a spectrum of autoimmune disorders, including Aicardi–Goutieres syndrome, familial chilblain lupus and retinal vasculopathy with cerebral leukodystrophy, and systemic lupus erythematosus 136-138.

1.4 Outcomes of DDR Mediated Immune signaling

Defects in DDR factors trigger multitude of cellular phenotypes, including auto-inflammatory disease, cellular senescence and cancer. Genotoxic agents serve as initial triggers for immune signaling activation is a consequence of inefficient DNA repair, replication stress, defective DDR signaling response or erroneous cell cycle progression. a. DNA damage mediated immune Signaling and Cancer progression: The major function of DDR signaling is to repair damaged DNA, promote genomic stability, and suppress tumorigenesis. Mechanistically, lack of DDR factors leads to erroneous DNA repair, replication fork instability, cell-cycle checkpoint defects and genomic instability in response to genotoxic stress. Recent evidence suggests that host immune response play an important role in regulating malignant tumorigenesis. Recent pre-clinical studies have found that dying tumor cells exposed to RT accumulate cytosolic DNA, which activates the stimulator of interferon genes (STING) pathway 112,113. STING activation is critical for the expression of inflammatory cytokines and chemokines, specifically type I interferon (IFN), which in turn recruits DCs to the tumor and activates them. Activated DCs are capable of presenting the tumor antigens to T cells, which primes the T cells to selectively kill tumor cells 112,113. Thus, radiation can activate the cell-mediated immune response against tumors by enhancing tumor DNA delivery to DCs. Recently, Matsunaga and colleagues showed that carbon particle radiation coupled with DC-based immunotherapy resulted in control of primary tumors in immunocompetent mice 139. Further, these challenged animals developed long lasting CD8+ T-cell-mediated anti-tumor immunity, which prevented recurrence of the tumors. They could not replicate these results in immune-deficient nude mice, however, thus highlighting the importance of the

19 immune system in controlling the tumor 139. In this regard, it is not surprising that cytosolic DNA sensing pathway acts as an intrinsic barrier to tumorigenesis140. For example, MUS81-dependent accumulation of genomic DNA in the cytoplasm triggers STING-mediated type 1 IFNs but and rejects prostate cancer 58. Ionizing radiation-induced DNA damage in B16 cells serves as a vaccine that promotes STING-dependent tumor rejection in combination with immune checkpoint blockage122. Similarly, decreased expression of STING and cGAS is associated with poor cancer prognosis124,141. However, multiple reports also suggest that cytosolic DNA mediated immune signaling promotes tumor growth and metastasis 110. For example, recent studies indicate the activation and importance of DNA cytosolic sensing pathways in cancer cells: the cGAS-STING signaling pathway in BRCA1-mutant breast cancer cells and triple negative breast cancer cells 66,118; and cytosolic DNA response in a variety of metastatic cancers 142. Thus, it is possible DDR mediated immune signaling is a double-edged sword. On one hand, acute activation of cytokines and chemokines help remove cancer cells (anti-tumor); but, on the other hand, chronic and persistent activation of inflammatory cytokines may promote carcinogenesis (pro-tumor). A detailed understanding of the factors that determine whether the immune system eliminates cancer cells or fuels tumorigenesis will help in developing novel therapeutic approaches to both prevent and treat cancers. b. DNA damage mediated immune Signaling and Cellular Senescence: Senescence is an intrinsic cellular response that results in irreversible cell-cycle arrest and plays a critical role in both aging and tumor suppression. Though the triggers for cellular senescence are manifold including telomere shortening, oxidative damage, and oncogenic signaling, the activation of DDR is a common mechanism that is critical for inducing and maintaining a senescence phenotype143. Previously, increased inflammation, especially type I IFN response have been established as a driver for DNA damage mediated cellular senescence144,145. Recent evidence show apresence of cytosolic DNA or chromatin fragments and their recognition by cGAS plays an essential role in initiating the signals required for premature senescence 58,120,121,124. Mechanistically, this is interesting because because cGAS recognition of cytosolic DNA fragments and the subsequent activation of innate immune signaling precede the appearance of the senescence phenotype, and depleting cGAS or inhibiting its activity suffices to ameliorate premature senescence in the presence of cytosolic chromatin fragments (discussed in chapter 3). Thus, the initial activation of the cGAS-mediated cytosolic DNA sensing pathway is sufficient to establish a senescence phenotype. Reports suggest that IL- 8 and IL-6 are known to feed back to the secreting cells to reinforce senescence signaling 146,147, known as senescence-associated secretory phenotype (SASP). It is likely that cGAS-mediated expression of pro- inflammatory factors, such as IFN-β and IL-6, and IL-8, in response to defective DDR signaling could serve as a paracrine signal that establishes a senescence phenotype, caused by genotoxic stress. Establishment of SASP is considered a tumor suppressive component of senescence by maintaining immune surveillance

20 and clearance of tumor cells120,148. However, multiple reports suggest SASP can also induce deleterious changes in the tissue microenvironment resulting in proliferation, transformation and metastasis of epithelial cells, thus promoting tumorigenesis149-152. Such conflicting functions of senescence and the role of the cGAS–STING pathway in these functions needs to be further clarified in future studies. c. Auto inflammatory disease: Defective DDR signaling can result in autoimmune disorders. Analysis of sera from systemic autoimmune rheumatic disease (SARD) patients revealed presence of autoantibodies against multiple DNA repair proteins including Ku, DNA-PKcs, PARP, WRN and Mre11 153. Additional studies showed that cell lines from patients with systemic lupus erythematosus (SLE) have a defective DSB repair 154. Presence of self- DNA in cytoplasm, as result of defective DDR signaling, is a key source of development of auto inflammatory disease. Recent insights in cytosolic DNA mediated inflammatory signaling suggest there are two major mechanisms responsible for the development of autoinflammatory disease: First, aberrant activation of the cytosolic DNA sensing pathway due to defective DDR factors. For example, ATM deficiency leads to autoinflammation and elevated levels of circulating type I IFN in AT patients, as a result accumulation of unreapaired DNA lesions in the cytoplasm 155,156. Intriguingly, depleting either cGAS or STING attenuated autoinflammatory disease in ATM-deficient mice 35,110,157. Similarly, breast and ovarian cancer patients with RAD51C (RAD51 paralog) mutation were found to have autoimmune features such as antinuclear antibodies, autoimmune thyroid disease or an IFN signature in their blood 58. Similarly, Deficiency in ERCC1-XPF, an NER endonuclease, results in upregulated TNF, IL and NF-kB signaling, leading to premature aging and early death in mice 158,159. In humans, mutations in ERCC1 and ERCC8/ ERCC6 results in developmental disorders called Cerebro-Oculo-Facio-Skeletal syndrome (COFS) and Cockayne syndrome (CS), respectively. Patients with COFS and CS often present similar autoimmune clinical features as AGS 160-162. The second mechanism is loss- or gain-of-function mutations in cytosolic DNA degradation enzymes and the cytosolic DNA sensing pathway that can result in development of autoimmune diseases. For example, Mutations in genes involved in nucleic acid metabolism, including TREX1, rnaseH2a, rnaseH2b and samhd1, cause AGS. Similar to human AGS patients with TREX1 deficiency, -/- mice also develop an inflammatory phenotype 163. Importantly, genetic depletion of cGAS or STING in trex1-/- mice rescues autoinflammatory symptoms associated with Trex1 deficiency 131-134. Thus, constitutive activation of the cytosolic DNA sensing pathway due to defective DNA damage sensors and nuclease factors contributes to the development of autoinflammatory disease.

21 1.5 Research goals:

Management of DNA lesions is one of the fundamental problems faced by cells. Roughly, each cell in our body are confronted with approximately 104–105 lesions per day and failure to preserve genomic integrity play a causal role in developmental disorders, cancer and age related pathologies. DNA damage response (DDR) is a collective term, which encompasses DNA damage sensing, repairing and cell cycle checkpoint pathways. Initially, the primary importance of DDR were thought to be confined in suppressing genomic instability in response to genotoxic stress, as inherited disorders in DDR signaling factors results in cancer pre disposition. In addition to cancer and developmental disorders, research in the last decade has identified several DNA repair factors in regulation of immune signaling as well. For example, DNA-PK-Ku70 complex and MRE11 are involved in sensing foreign and endogenous DNA in the cytoplasm, respectively and DNA repair function of ATM is required to prevent chronic activation of antiviral immune response and autoimmune disorders. In my dissertation work, I wanted to expand our knowledge on this front; are DDR events and immune signaling isolated events or they are integrated in a global network of events, with an ultimate goal of maintaining homeostasis. I wanted to know besides these handfuls, what are the other DNA repair factors involved in regulation of immune activation and if so, what is the underlying mechanism; is it an extension of their DNA repair function or independent of it. For example, in my lab we have a long-standing interest in factors involved homologous recombination mediated DSB repair. RAD51, a recombinase, is one of the most important factors in the HRR pathway. RAD51, a multifunctional protein, also plays a central role in DNA replication, and may be involved in tumorigenesis, as tumors often upregulate or downregulate RAD51 expression. To determine if RAD51 is involved in immune signaling, we exposed RAD51 deficient cells to ionizing radiation and tested whether RAD51 deficiency results in differential expression of immune signaling genes. Once we observed changes in immune related gene expression, we wanted to explore the relevant mechanism through which RAD51 deficiency results in elevated immune signaling. Radiation exposure induces DNA lesion in a sequence non-specific manner. I wanted to understand how cells respond to loci-specific DNA damage. One special area of interest in my lab are telomeres, which are the capping structures present at the end of eukaryotic and are composed of long stretches of TTAGGG repeat sequences and associated proteins (shelterin complex). I was especially interested in understanding how cells respond to dysfunctional telomeres because of two reasons, i. telomeres are important for maintaining overall chromosomal integrity and ii. Previous literature have indicated that telomeric DNA damages are refractory to DNA repair machinery. Reports suggested inability to repair telomeric DNA damage is one of the primary reasons why cells undergo premature senescence, when exposed to genotoxic stress. In my research, I used three different ways to induce telomere dysfunction and

22 subsequently tried to understand the molecular choreography of the events leading to cellular senescence. To my surprise, telomere dysfunction resulted in increased immune signaling and I asked whether defective DDR signaling contributes to that. Furthermore, I also explored the relevant pathways by which this immune signaling is mediated and how it contributes to senescence; and finally what are innate mechanisms by which cells try to prevent unwarranted immune activation and premature senescence.

Fig 1. Major DNA repair pathways and DDR factors involved in immune signaling.

23

Fig 2. Overview of research goals presented in this dissertation.

24 Table 1

Type of DNA lesions Source DNA damage repair pathway

Bulky and helix distorting DNA Ultraviolet (UV) rays and chemical Nucleotide excision repair (NER), Fanconi Anaemia (FA) pathway adducts mutagens like cisplatin

Base excision repair (BER).In case of impaired BER, the SSBs are Damaged bases and non helix Ionizing radiation and reactive oxygen converted to DSBs, which are repaired by HRR during S and G2 distorting DNA lesions species phase. This synthetic lethal relationship is exploited by using PARP inhibitors to kill HRR deficient tumors.

Ionizing radiation and reactive oxygen Single-strand breaks Single-strand break repair species

Base mismatches, incorrect Replication errors Mismatch repair (MMR) insertion and deletion of bases

Alkylating agents like temozolomide, Direct DNA repair (DR). In absence of MGMT, the main enzyme in O6 methylguanine adducts reactive oxygen and nitrogen species. the DR pathway, MMR and NER can process these lesions.

DSBs are repaired either by Non-homologous end joining (NHEJ) or by Homologous recombination repair (HRR) in a cell cycle Ionizing radiation, reactive oxygen Double strand breaks specific manner. Some overlap exists in between HRR and NHEJ species and replication stress pathways. DSBs which are not processed by NHEJ can be processed by HRR during S and G2 phase.

Replication events and DNA interstrand crosslinks Nucleotide excision repair (NER) and Fanconi Anaemia (FA) chemotherapeutic drugs like Mitomycin (ICLs) pathway C

Table 1: Major DNA repair pathways in eukaryotic cells

25

Chapter 2 RAD51 Interconnects Between DNA Replication, DNA Repair and Immunity

26 2.1 Summary:

RAD51, a multifunctional protein, plays a central role in DNA replication and homologous recombination repair of DSBs, and may be involved in tumorigenesis. We identified a novel role for RAD51 in innate immune response signaling. Defects in RAD51 lead to the accumulation of self-DNA in the cytoplasm, triggering a STING-mediated innate immune response after replication stress and DNA damage. Mechanistically, in absence of RAD51, the unprotected newly replicated genome is degraded by the exonuclease activity of MRE11, and the fragmented nascent DNA accumulates in the cytosol, initiating an innate immune response. Thus, our data suggest that in addition to playing roles in homologous recombination-mediated DNA double-strand break repair and replication fork processing, RAD51 is also implicated in the suppression of innate immunity. Thus, our study reveals a previously uncharacterized role of RAD51 in initiating immune signaling, placing it at the hub of new interconnections between DNA replication, DNA repair, and immunity.

2.2 Background:

Each of the ~1013 cells in the human body receives tens of thousands of DNA lesions per day 6. The maintenance of an intact genome is crucial for cellular homeostasis. Alterations in the pathways involved in the processing of stalled or collapsed replication forks, DNA repair cause genome instability, and chromosomal rearrangements, which are hallmarks of cancer cells. RAD51 is one of multiple factors involved in faithful DNA replication, repair, and recombination55,164. During double-strand break (DSB) repair, RAD51 catalyzes the core reactions of homologous recombination (HR), including strand invasion into duplex DNA and the pairing of homologous DNA strands, enabling strand exchange 165. Depending on the nature of energy deposition, Ionizing radiation (IR) are categorized as low (e.g., γ- and X-rays) and high [i.e., high charge and energy (HZE), e.g. Fe and Si radiation)] linear energy transfer (LET). IR induce a plethora of DNA damage, and the damage complexity increases with an increase in the LET of the radiation 166-168. High-LET radiation induces complex DNA damage, a unique class of DNA lesions that includes two or more individual lesions within one or two helical turns of the DNA caused by the passage of a single radiation track 166-168. These lesions can be abasic sites (apurinic/apyrimidinic sites or APs), damaged bases (oxidized purines or pyrimidines), single-strand breaks (SSBs), or double-strand breaks (DSBs) 169. Theoretical modeling and experimental evidence suggest that 70-90% high LET IR induced breaks are clustered 3,170. In mammalian cells, DSBs are mainly repaired by two distinct pathways, NHEJ and HRR 171,172. NHEJ and HRR pathways coordinate but function differentially in response to different types of DNA lesions and in different cell cycle phases. Evidence suggest that fragmented DNA

27 induced by high-LET HZE particles prevent Ku70/80 binding to DNA, which is the initial step for NHEJ, resulting in the inhibition of NHEJ-mediated clustered DSBs repair 52In contrast, it has been seen that HR repair factors are heavily involved in the processing of clustered DSBs 50. Further, the complexity of DNA damage also appears to be a critical factor in facilitating DNA end-resection 173, a crucial step in HR repair. Thus, taken together these reports imply that RAD51, a central player in HR repair pathway is fundamental to the repair of clustered DSBs. In addition to DSB repair, RAD51 also plays a role in various replication fork processes. RAD51 enables replication restart when a replication fork encounters DNA damage 55. Recent evidence indicates that RAD51 also prevents MRE11-mediated degradation of newly replicated genome after replication stress 7,62. Furthermore, RAD51 promotes cell survival following replication stress and prevents the accumulation of replication-associated DSBs 56 and genome instability. Although germ-line mutations in the Rad51 lead to embryonic death57, a precisely regulated amount of RAD51 is crucial for normal cellular functions. Multiple human tumors exhibit varying expression levels of RAD51, deleterious mutations in the protein, or defects in other tumor suppressors, such as BRCA1, BRCA2, Fanconi anemia (FA) factors 174,175. Overexpression of RAD51 due to increased transcription reduces methylation and stabilization of the protein and may cause chromosomal amplifications, deletions, and translocations resulting in a loss of heterozygosity and aneuploidy. These events can lead to cancer development and progression to metastasis 176. In contrast, down-regulation of RAD51 has been reported in many tumors 177. Despite these reports, the precise mechanism by which RAD51 suppresses or promote carcinogenesis is not clear.

Carcinogenesis is a multistage process resulting from a cumulative malfunctioning of DNA replication, DSB repair, and immune signaling. Chronic stimulation of the innate immune system can cause tumorigenesis 178,179. A number of recent studies show that DNA repair and replication factors play a role in regulation of innate immune response. For example, cells deficient in the DNA repair factor ataxia- telangiectasia mutated (ATM) causes increased cytosolic self-DNA accumulation, leading to increased inflammation 180. Similarly, MRE11, a DSB sensor protein, recognizes cytosolic DNA and initiates innate immune response signaling 23. In addition, the DNA structure-specific endonuclease MUS81, which cleaves DNA structures at stalled replication forks, also mediates the stimulator of interferon genes (STING)- dependent activation of immune signaling 69. It was recently discovered that FA proteins are involved in cellular immunity 94. Moreover, RPA2 and RAD51 were shown to protect the cytosol from the accumulation of self-DNA 58. These findings indicate the involvement of DNA repair and replication factors in immunity in addition to their known DNA repair and replication functions. Importantly, mutations in the

28 majority of these genes lead to cancer-prone disorders. However, whether defective RAD51 functions contribute to tumorigenesis through the activation of the innate immune system is still unknown.

We report a novel role of RAD51 in immunity in addition to its known functions in DSB repair and replication fork processing. We discovered that the down-regulation of RAD51 leads to the upregulation of innate immune response pathway genes upon DNA damage and replication stress induced by irradiation. In the absence of RAD51, the newly replicated genome is degraded by the exonuclease activity of MRE11. We also showed that these degraded nascent DNA fragments are exported to the cytoplasm, triggering innate immune response signaling. Our study reveals a previously unidentified role of RAD51 in triggering an innate immune response, placing this protein at the hub of new interconnections between DNA replication, DNA repair, and immunity.

2.3 Material and Methods

Cell lines and culture conditions: HT1080 cells were obtained from ATCC and maintained in Minimum Essential Medium (MEM) alpha supplemented with 10% fetal bovine serum, 100 mg/ml streptomycin sulfate, and 100 U/ml penicillin. To establish the stable expression of cell cycle markers (HT1080-FUCCI), HT1080-EYFP-53BP1 cells 181 were transduced with lentivirus carrying G1 [mCherry-hCdt1(30/120)] and S/G2 [AmCyan-hGeminin(1/110)] phase markers (Fig 3). Stable HT1080-FUCCI cells were selected using zeocin (1μg/ml). To down-regulate the expression of RAD51, HT1080-FUCCI cells were transfected with a mammalian expression plasmid containing tetracycline-inducible Rad51 shRNA 50. Stable lines were selected using hygromycin (75 μg/ml). To down-regulate RAD51 protein levels, cells were cultured in the presence of doxycycline (DOX, 1μg/mL) for 72 h. MCF10A cells stably expressing scrambled and Rad51 shRNA were kind gift from Dr. Shiaw-Yih Lin (MD Anderson Cancer center) and were maintained in mammary epithelial basal medium (LONZA, CC-3151) as described previously 182. 4T1 cells were maintained in RPMI-1640 media supplemented with 10% fetal bovine serum, 100 mg/ml streptomycin sulfate, and 100 U/ml penicillin 183. To down-regulate the expression of RAD51, 4T1 cells were transfected with a mammalian expression plasmid containing tetracycline-inducible Rad51 shRNA and selected using puromycin (1µg/ml). MRE11-defective ATLD cells were maintained in MEM-alpha supplemented with 10% fetal bovine serum, 100 mg/ml streptomycin sulfate, and 100 U/ml penicillin 7. To down-regulate RAD51 expression, ATLD cells were transduced with lentivirus carrying shRad51 (Sigma-Aldrich). Stable ATLD-shRad51 cells were selected using puromycin (1µg/ml). All cells were maintained at 37°C in a humidified 5% CO2 incubator. Irradiation and dosimetry: DNA lesions and replication stress were induced by exposing the cells to high-LET iron (Fe) particles generated at the NASA Space Radiation Laboratory in Brookhaven National

29 Laboratory, Long Island, NY. The energy of the used Fe-particles was 1 GeV/nucleon and the dose rate ranged from 50 to 200 cGy/min. The linear energy transfer (LET) of the Fe-particles was 150 keV/µm. The residual ranges of the beams were determined before each experiment and used to calculate the track- averaged LET values 184. Drugs and antibodies: The following reagents were used: Doxycycline (Sigma-Aldrich, D9891), MRE11 inhibitor [5-(4-hydroxybenzylidene)-2-iminothiazolidin-4-one, (Sigma-Aldrich, M9948)], IdU (Sigma- Aldrich, I7125), CldU (Sigma-Aldrich, C6891) and Histone deacetylase inhibitor, suberoylanilide hydroxamic acid, (Tocris Biosciences, 4652). Mouse monoclonal anti-γH2AX (JBW301; EMD Millipore, 07-164), mouse monoclonal anti-RPA2 (Ab-3; EMD Millipore, NA19L), rabbit polyclonal anti-RAD51 (H-92; Santa Cruz, sc-8349), rabbit polyclonal anti-STING (Novus Biologicals, 2-24683), anti-phospho- STING (Ser366; Cell Signaling, 85735), rabbit monoclonal anti-phospho-TBK1 (Ser172; D52C2; Cell Signaling, 5483), rabbit polyclonal anti-phospho-STAT3 (Tyr705; Cell Signaling, 9131), rabbit monoclonal anti-CASPASE-3 (8G10; Cell Signaling, 9665), mouse monoclonal anti-PARP1 (F2; Santa Cruz, 8007), mouse monoclonal anti-γ-Tubulin (GTU-88; Sigma-Aldrich, T6557), mouse monoclonal anti- BrdU (B44; BD Biosciences, 347580) and rat monoclonal anti-BrdU (BU1/75-ICR1; Novus Biologicals, NB500-169) antibodies. Fluorescent conjugated secondary antibodies Alexa488, Alexa555, Alexa633, were purchased from Molecular Probes (Invitrogen). Drug treatment: For inducible depletion of RAD51, HT1080 and 4T1 cells were grown in presence of 1µg/mL doxycycline for 72 h and then exposed to radiation. For pharmacological depletion of RAD51, HT1080 cells were grown in presence of 2.0 µM suberoylanilide hydroxamic acid (SAHA) for 36 h and then allowed to recover for an additional 12 h before irradiation. For inhibition of MRE11’s exonuclease function, cells were treated with 25 µM mirin 1 h prior to irradiation and the cells were maintained in mirin- containing medium until the end of the experiment. For inducible depletion of RAD51, ATLD cells were grown in presence of 1µg/mL doxycycline for 72 h and then exposed to radiation. For the pharmacological depletion of RAD51, ATLD cells were grown in the in presence of 10 µM SAHA for 48 h, allowed to recover for 24 h and then exposed to radiation. Indirect immunofluorescence staining: About 5 x 104 cells were seeded in a six-well plate containing cover glasses, incubated for 24 h, treated with doxycycline (DOX, 1μg/mL) for 72 h, and exposed to 1-2 Gy Fe-particles. Cells were fixed with 4% paraformaldehyde at room temperature for 20 min at different post-IR times and treated for indirect immunofluorescence as described previously 181. For the detection of RPA2 and RAD51 foci, cells were washed three times in PBS and incubated in extraction buffer (10 mM

HEPES, pH 7.4, 300 mM sucrose, 100 mM NaCl, 3 mM MgCl2, 0.1% Triton X-100) on ice for 10 min before fixation. For immunostaining, cells were permeabilized in Triton X-100 (0.5% in PBS) on ice for 5 min, washed three times with PBS, incubated in blocking solution (5% goat serum in PBS) at room

30 temperature for 60 min, and incubated with primary antibodies (diluted in 5% goat serum) at room temperature for another 3 h. Subsequently, cells were washed with 1% BSA in PBS, incubated with appropriate secondary antibodies (1:1000 in 2.5% goat serum, 1% BSA, and PBS) at room temperature for 60 min, washed five times with 1% BSA, and mounted with mounting medium containing DAPI (4',6- diamidino-2-phenylindole, Vectashield). For cytoplasmic BrdU detection, exponentially grown cells were labeled with 20 μM BrdU for 18- 20 h in regular growth medium, replaced with regular medium without BrdU, irradiated, and fixed with 80% methanol (in PBS) on ice for 20 min. Fixed cells were washed three times with cold PBS, incubated in blocking solution (5% goat serum in PBS) at room temperature for 1 h, and treated with anti-BrdU antibody (mouse) overnight at 4oC. Cells were washed with 1% BSA in PBS, incubated with the Alexa 555 secondary antibody at room temperature for 1 h, washed five times with 1% BSA, and mounted with mounting medium containing DAPI (Vectashield). Image acquisition and foci counting: Images were captured using an LSM 510 Meta laser scanning confocal microscope with a 63 × 1.4 NA Plan-Apochromat oil immersion objective. Images were taken at z-sections (15-20 sections) of 0.35µm intervals using 405-nm (DAPI), 457-nm (AmCyan), 488-nm (Alexa 488), 514-nm (EYFP), 543-nm (mCherry/Alexa 555), and 633-nm (Alexa 633) lasers. The tube current of the 488-nm argon laser was set at 6.1 A. The laser power was typically set to 3-5% transmission with the pinhole opened at 1 Airy unit. To count γH2AX foci in different cell cycle phases, we developed an algorithm in MatLab (MathWorks, MA) and used it to count foci in G1 and S/G2 phases. The quantification of foci was conducted from images of 100-200 cells for each time point from three to four independent experiments 181. Whole-cell extract preparation and western blotting: Whole-cell extracts were prepared in radio- immunoprecipitation assay (RIPA) buffer according to a published procedure185. The protein concentration was measured by the bicinchoninic acid (BCA) assay. Aliquots containing 50-150 μg protein were resolved by 8-15% SDS-PAGE, transferred onto nitrocellulose or PVDF membrane, and incubated with the indicated antibodies either at room temperature for 4 h or overnight at 4oC 7,11. Sub-cellular fractionation: The cytoplasmic extract was isolated following a previously published methodology180. Briefly, cell pellets were re-suspended in 10 mM of HEPES (pH 7.9), 10 mM KCl, 1.5 mM MgCl2, 0.34 M Sucrose, 10% glycerol, 0.1% Triton X-100, protease inhibitors cocktail (Roche, 11836170001), and incubated on ice for 10 minutes. The nuclei were separated from the cytoplasmic extract by low-speed centrifugation (1500 relative centrifugal force) at 4oC for 5 minutes. The supernatant containing the cytosolic extract was either immediately used for single-strand (ssDNA) and double-strand (dsDNA) DNA quantification or stored at -80°C until use.

31 Quantification of cytosolic single-strand (ssDNA) and double-strand (dsDNA) DNA: ssDNA and dsDNA in the cytosolic fractions were quantified using Quant-iT OliGreen and PicoGreen Assay Kits (O11492 and P7589 kits, respectively; Invitrogen). Briefly, black 96 well plates (Phenix Research Product, MPG-655076) were first loaded with 98 μl of 1x Trypsin-EDTA in each well and then with 2 μl of cytosolic extracts. Subsequently, 100 μl of 2x OliGreen or PicoGreen dye was added into each well. This mixture was incubated in the dark for 5 min, and the fluorescence intensity was measured at an excitation wavelength of 480 nm and an emission wavelength of 520 nm using a Spectrofluorometer (Perkin Elmer 2030 Multilabel Reader, Victor X3). Standard curves were generated for each experiment using standard M13 primer and Lamda DNA for ssDNA and dsDNA diluted in cytosolic extraction buffer, respectively. The final ssDNA and dsDNA concentration were calculated by normalizing against cytosolic protein concentration and presented as fold change relative to sham treated controls or picogram DNA per microgram of cytosolic protein. Microarray: Total RNA was isolated from mock-treated and irradiated cells using the Qiagen RNeasy kit (217004, Qiagen), according to manufacturer's instructions. The RNA quality was determined using the Experion system (Bio Rad). Samples were processed and hybridized to Illumina Human HT12v4 arrays (Illumina, Inc.) using standard Illumina protocols. The slides were scanned on an Illumina Beadstation (Illumina, Inc.; Genomics Core, UT Southwestern Medical Center). Microarray data processing and pathway analysis: The methods for data normalization and analysis were based on the use of “internal standards” 186 that characterize some aspects of the system’s behavior such as technical variability 187,188. The comparison between these methods with other normalization and analysis procedures was previously published 189. The two-step normalization procedure and the associative analysis functions were implemented in MatLab (Mathworks, MA) and are available from the authors upon request. These algorithms can also be obtained from an R package diffGeneAnalysis, available as part of Bioconductor packages (http://www.bioconductor.org/packages/2.5/bioc/html/diffGeneAnalysis.html). Heatmaps were generated with the Spotfire Decision Site 9 (TIBCO, Palo Alto, CA) with gene subsets created from the list of significant genes. Quantitative real-time polymerase chain reaction (qRT-PCR): cDNA was synthesized from 1-3 μg of RNA using SuperScript III Reverse Transcriptase (18-080-051 Fischer Scientific) in a total volume of 20 µl, according to manufacturer’s instructions. The cDNA was subjected to qRT-PCR for a number of genes using the primer sets (Table 2), CFX96 Touch Real-Time PCR Detection System (Bio Rad) and iTaq Universal SYBR Green Supermix (Bio Rad, 1725121), according to manufacturer’s instructions. The relative gene expression was determined by the ΔΔCT method. The difference in cycle times ΔCT was determined as the difference between the tested gene of interest and the reference housekeeping β-actin gene. The ΔΔCT was obtained by finding the difference between the groups. The fold change (FC) was

32 calculated as FC=2–ΔΔCT. All primers were purchased from Invitrogen. qRT-PCR assays were carried out in triplicates for each sample. The mean value was used for the calculation of the mRNA expression levels and presented as fold change relative to respective sham treated controls. DNA fiber assay: DNA fiber assay was performed as described previously 7,190. Briefly, 2.5 x 105 cells were labeled with IdU (150 μM) for 30 min, washed four times with warm PBS, labeled with or without CldU, treated with and without radiation (1 Gy), and recovered for either 30 min or 5 h. After three washes with warm PBS, both labeled and unlabeled cells were trypsinized and mixed at a 1:15 ratio (labeled: unlabeled). Both cell types were lysed on a clean glass slide in 20 µl of lysis buffer (0.5% SDS, 50 mM EDTA, and 200 mM Tris-HCl pH7.4) for 8 min; slides were tilted slightly (~15° angle) to help the DNA spread slowly. Subsequently, slides were immunostained with anti-BrdU antibodies, and the DNA fiber lengths were measured using Axiovision Software. Cell survival assay: Cell survival was measured using a colony formation assay. About 1000-20000 cells were seeded in triplicates in 6-well plates, incubated for 6-8 h before IR, and irradiated with 0.5 Gy. Immediately after irradiation, cells were allowed to form colonies for 8-10 days. Colonies were washed once with 1X PBS, fixed with 70% ethanol for 10 min, and stained with 0.5% crystal violet dissolved in 25% methanol and 75% water. Colonies composed of more than 50 cells were scored as having grown from a single surviving cell. Survival graphs were generated from three replicate wells with colony numbers normalized to sham-treated controls. Metaphase spreads preparation: Chromosome aberrations were carried out as described previously3. Sixteen hours after exposing cells to 1 Gy irradiation, chromosome preparations were made by accumulating metaphases in the presence of 0.1 mg/mL colcemid (Irvine Scientific) for 6 h. In the case of experiments with mirin, the medium was replaced with either fresh medium containing 25 μM mirin or DMSO 1 h before radiation. The cells were maintained in mirin-containing medium until the end of the experiment. Chromosome images were taken using an Olympus microscope (100X objective) equipped with an Image Spot camera (Spot Imaging Solutions). Chromosome aberrations were scored as described previously3. Statistical analysis: Data are expressed as means ±SEM or STDEV of three to four independent experiments. The Student’s t test was performed on all values and p<0.05 was considered statistically significant. GraphPad Prism (version 7.0) was used to create the graphs. SigmaPlot (version 12.5) was used to calculate foci dissolution kinetics.

33 2.4. Results:

RAD51-depletion leads to the up-regulation of innate immune response genes: that RAD51 plays a central role in homologous recombination (HR)-mediated DNA double-strand break (DSB) repair and replication fork processing. To gain new insight into the function of RAD51 outside DSB repair and replication fork processing, we conducted gene expression profiling of RAD51-proficient and –depleted cells (Fig. 4A) by microarray after DNA damage and replication stress. We used high-linear energy transfer radiation to induce both DNA lesions and replication stress 3,190. We found that expression levels of a number of genes were significantly (p<0.05) altered in irradiated RAD51-proficient and –depleted cells as compared to corresponding mock-treated cells (Fig. 10A). Subsequently, we performed analysis on differentially expressed gene sets using the Ingenuity Pathway Analysis (IPA) software. Our data indicated that the expression of genes known to function in DNA replication, recombination, and repair, and in the cell cycle were significantly altered in both irradiated RAD51-proficient and -depleted cells as compared to their corresponding mock-irradiated cells (data not shown). Interestingly, IPA analysis revealed that the expression of many genes with known functions in innate immunity were up-regulated significantly in irradiated RAD51-depleted cells as compared with irradiated RAD51-proficient cells (Figs. 4B-C and 10B). These results revealed that the depletion of RAD51 results in the up-regulation of innate immune response genes upon irradiation. Subsequently, we validated a set of six genes that showed elevated expression patterns in the microarray by quantitative real-time polymerase chain reaction (qRT-PCR), using mRNA purified from mock-treated and irradiated RAD51-proficient and -depleted cells. Similar to the microarray results, expression levels of IL-6, CSF2, CXCR4, TNF-α, CMKLR1, and TLR9 expression were 12, 10, 5, 3.2, 6, and 3.2 folds, respectively significantly higher in irradiated RAD51-depleted cells than in irradiated RAD51-proficient cells (p<0.05-0.008; Figs. 4D-I). Taken together, these results clearly indicate that the depletion of RAD51 leads to the up-regulation of many genes involved in the innate immune response pathway following irradiation.

RAD51-depleted cells accumulate elevated levels of cytosolic DNA: The mechanism underlying the up-regulation of innate immune response pathway genes in RAD51-depleted cells upon irradiation is unknown. An innate immune response is known to be triggered by cytosolic DNA 113. To identify whether the up-regulation of innate immune response pathway genes in response to irradiation in RAD51-depleted cells is due to the elevated amount of cytosolic DNA, we quantified single- (ssDNA) and double-stand (dsDNA) DNA in the cytoplasmic fractions of RAD51-proficient and -depleted cells at different intervals after irradiation. The amount of cytosolic ssDNA (12.83±3.51 and 12.91±3.86 pg per µg protein in mock and irradiated cells, respectively; Fig. 5A) and dsDNA (4.0±0.03 and 3.95±0.42 pg per µg protein in mock

34 and irradiated cells, respectively; Fig. 5B) was similar in mock and irradiated RAD51-proficient cells. In contrast, the amount of cytosolic ssDNA (12.19± 3.18 and 19.91±3.05 pg per µg protein in mock and irradiated cells, respectively; p=0.0006; Fig. 5A) and dsDNA (4.4±0.38 and 9.37±0.67 pg per µg protein in mock and irradiated cells, respectively; p=0.0004; Fig. 5B) was significantly higher in RAD51-depleted cells as compared to RAD51-proficient cells 8 h after irradiation. These results clearly demonstrate that the depletion of RAD51 leads to elevated levels of cytosolic DNA in response to irradiation. To further rule out the possibility that the elevated levels of cytosolic DNA and activation of innate immune response genes in irradiated RAD51-depleted cells are not due to cells undergoing apoptosis, we checked the level of cleaved products of CASPASE-3 and PARP-1 by western blotting at different times after radiation exposure. Cleaved CASPASE-3 and fragmented PARP-1 are recognized biomarkers for cells undergoing apoptosis 191,192. Cleaved PARP-1 and CASPASE-3 levels were not different between irradiated RAD51-proficient and –depleted cells at any time after irradiation (Fig. 5C). Overall, these results indicate that the elevated level of cytosolic DNA in RAD51-depleted cells upon irradiation is not due to cells undergoing apoptosis. Next, we verified whether the accumulation of cytosolic DNA is unique to RAD51-depletion in HT1080 cells or it can also occur in other cell types lacking RAD51. First, we used a previously published MCF10A cells stably expressing scrambled (shscr) and Rad51-specific shRNA 182 and examined the levels of cytosolic DNA (Fig. 5F, inset). As in RAD51 depleted HT1080 cells, the amount of cytosolic ssDNA and dsDNA (36.75±5.01 and 5.63±0.51 pg per µg protein ssDNA and dsDNA, respectively; p<0.002- 0.0002) was significantly higher in irradiated MCF10A-shRad51 cells relative to irradiated MCF10A-shscr cells (22.7±3.37 and 3.54±0.66 pg per μg protein ssDNA and dsDNA, respectively; Figs. 5D-E). Importantly, elevated levels of cytosolic DNA correlated well with the increased expression of innate immune response genes (TNF-α, TLR9 and CMKLR1) in irradiated RAD51-depleted cells (p<0.002- 0.0001; Fig. 5F). Thus, these results confirmed that the observation on the activation of innate immune response in RAD51-depleted cells upon radiation is not limited to one cell type. To confirm that the role of RAD51 in innate immunity is not limited to human cells, we exposed mouse mammary carcinoma cells (4T1) stably expressing tetracycline inducible mouse shRad51 to radiation. Similar to human cells, the cytosolic ssDNA levels were significantly elevated in 4T1-depleted cells 8 h after irradiation (12.96±0.75 and 21.1±1.1 pg per µg protein in mock and irradiated cells, respectively, p<0.0005) as compared with irradiated RAD51-proficient 4T1 cells (18.44±2.61 and 17.84±0.76 pg per µg protein, in mock and irradiated cells, respectively, Figs. 5G-H). Similarly, the cytosolic dsDNA levels were also significantly elevated in irradiated 4T1-depleted cells (4.79±0.45 and 7.11±0.39 pg per µg protein in mock and irradiated cells, respectively, p<0.02) as compared with irradiated RAD51-proficient 4T1 cells (6.99±0.88 and 6.96±0.29 pg per ug protein in mock and irradiated cells,

35 respectively, Figs. 5G-H). Importantly, elevated levels of cytosolic DNA correlated well with the increased levels of innate immune response genes (IL-6, CSF2, TLR9 and TNF-α) expression in 4T1-depleted cells upon irradiation (p<0.007-0.0006; Fig. 5I). Thus, RAD51 is involved in the suppression of cytosolic DNA accumulation in response to radiation in different mammalian cell types. In addition to the genetic depletion of RAD51 in different mammalian cells, we pharmacologically down-regulated RAD51 levels in HT1080 cells using SAHA 193. SAHA treatment of HT1080 cells with SAHA reduced the levels of RAD51 expression in HT1080 cells (Fig. 5L, inset). Subsequently, we measured the levels of both cytosolic DNA and expression of innate immune response genes after irradiation. Similarly to the genetic knock-down of RAD51, SAHA treated irradiated cells showed increased amount of ssDNA (37.2±4.17 and 58.07±5.9 pg per µg protein in SAHA treated control and irradiated cells, respectively; p<0.002) and dsDNA (8.6± 0.78 and 12.6±0.87 pg per µg protein in SAHA treated control and irradiated cells, respectively; p<0.007) relative to cytoplasmic ssDNA and dsDNA levels in sham treated HT1080 cells exposed to radiation (12.91±3.86 and 3.95±0.42 pg per µg protein ssDNA and dsDNA, respectively; Figs. 5J-K). Elevated levels of cytosolic DNA correlated with the increased levels of innate immune response gene (IL-6, CSF2, TLR9 and TNF-α) expression in SAHA- treated HT1080 cells upon irradiation relative to SAHA treated control cells (p<0.02-0.0001; Fig. 5I). Overall, RAD51 is involved in the suppression of innate immune response upon irradiation.

Nuclear-derived self-DNA accumulates in the cytosol of RAD51-depleted cells: To identify the source of cytoplasmic DNA and to distinguish the cell-cycle-dependent accumulation of cytosolic DNA, we labeled the genomic DNA of HT1080-FUCCI cells with 5-bromo-2'-deoxyuridine (BrdU) for 18-20 h, followed by irradiation. Subsequently, we immunostained the cells with anti-BrdU antibodies under non- denaturing conditions and quantified the mean BrdU fluorescent intensity (MFI) per cell. Quantification of cell cycle specific cytoplasmic BrdU intensity (Fig. 6A) showed that the cytoplasmic BrdU intensity was comparable between G1 and S/G2 cells in both non-irradiated RAD51-proficient and [3.2+0.38 and 4.28+0.09 MFI per G1 and S/G2 cell, respectively) and RAD51-depleted cells (5.0+0.83 and 5.11+0.08 MFI per G1 and S/G2 cell, respectively, Fig. 6B). We observed a slight increase in the cytoplasmic BrdU intensity after radiation in RAD51-proficient cells, but there was no significant difference in the cytoplasmic BrdU intensity between G1 and S/G2 cells after irradiation [5.08+0.18 and 5.40+0.15 MFI per G1 and S/G2 cell, respectively] cells (Fig. 6B). In contrast, we observed a highly significant increase in cytoplasmic BrdU intensity in RAD51-depleted cells after radiation relative to irradiated RAD51-proficient cells (5.24+0.02 and 10.92+0.08 MFI per cell, p<0.0001). Furthermore, the cytoplasmic BrdU fluorescent intensity in irradiated RAD51-depleted S/G2 cells (12.27±0.06 MFI per cell; p<0.0006) was significantly higher relative to irradiated RAD51-proficient S/G2 (5.41±0.14 MFI per cell) cells (Fig. 6B). More

36 importantly, cytoplasmic BrdU fluorescent intensity significantly increased in RAD51-depleted S/G2 (12.27±0.06 MFI per cell) cells than in G1 cells (9.57±0.09 MFI per cell; p<0.001) after irradiation (Fig. 6B). Thus, cytoplasmic DNA is derived from the nuclear DNA and RAD51-depleted S/G2 cells harbor a major fraction of cytosolic self-DNA upon irradiation.

Cytosolic self-DNA activates the stimulator of interferon gene (STING) signaling in RAD51- depleted cells: Evidence suggests that the innate immune response resulting from cytoplasmic nucleic acid is mediated by STING activation, including clustering and phosphorylation 23,180. It has been reported that activated STING forms higher order clusters 194. To verify whether the presence of self-DNA in the cytoplasm triggers STING clustering, we conducted immunostaining with anti-STING antibody (Fig. 6C). No significant changes in STING clustering were observed in both mock- and irradiated RAD51-proficient cell percentages (Fig. 5D). In contrast, STING clustering percentages were significantly higher in irradiated RAD51-depleted cells (p<0.002) than in irradiated RAD51-proficient cells (Fig. 6D). To further confirm whether the presence of self-DNA in the cytoplasm activates STING, we evaluated STING phosphorylation by Western blotting. We detected strong STING phosphorylation in RAD51-depleted cells as early as 4 h after irradiation which further increased at 8 h and remained stable at later time points (Fig. 6E, left panel). In contrast, the level of STING phosphorylation was much delayed and visibly weak in irradiated RAD51- proficient cells (Fig. 6E, left panel). Thus, the accumulation of self-DNA in the cytosol activates STING in cells lacking RAD51 upon irradiation. Tank binding kinase-1 (TBK1) is known to transmit immune signals downstream of STING in response to cytoplasmic DNA 195,196. To determine whether an increased STING activity in RAD51- depleted cells stimulates TBK1 activity, we monitored TBK1 phosphorylation by Western blotting. We detected TBK1 phosphorylation as early as 8 h and thereafter observed increased phosphorylation levels in irradiated RAD51-depleted cells (Fig. 6E, right panel). In contrast, RAD51-proficient cells only showed a minimal increase in TBK1 phosphorylation after irradiation. Furthermore, TBK1 activates STAT3 leading to the upregulation of IL-6 197. To further dissect the innate immune signaling network downstream of the STING-TBK1 axis, we verified STAT3 phosphorylation in RAD51-proficient and -depleted cells after irradiation. We detected low levels of STAT3 phosphorylation in irradiated RAD51-proficeint cells (Fig. 6E, right panel). In contrast, STAT3 phosphorylation was higher in irradiated RAD51-depleted cells than in irradiated RAD51-proficient cells (Fig. 5E). Thus, the activation of STING-TBK1 leads to the activation of STAT3 in response to self-DNA in RAD51-depleted cells. Overall, these results revealed that the presence of self-DNA in the cytoplasm initiates STING-mediated innate immune response signaling.

37 RAD51 blocks the MRE11-mediated degradation of nascent DNA strands in response to DNA damage: Nuclear DNA accumulates in the cytoplasm as a result of DNA damage, replication stress 110,129, or replication intermediates 116. RAD51 is involved in replication fork progression, efficient restart, and stability in response to replication stress 7,55. In addition, RAD51 also plays a role in the faithful repair of DSBs. Therefore, the source of self-DNA in the cytosol can originate from defective replication fork processing or defective DSB repair in RAD51-depleted cells. To determine the role of RAD51 in replication fork processing in response to irradiation, we first verified whether RAD51 is recruited to the sites of replication. Most RAD51 foci juxtaposed with EdU foci, representing replication forks, in irradiated- but not in mock-treated cells (Fig. 7A). Similarly, a major fraction of γH2AX foci, a surrogate marker for DSBs, juxtaposed with the RAD51 foci upon irradiation (Fig. 7A). Thus, these results suggest that RAD51 is localized at the sites of both replication forks and DSBs in response to irradiation. Second, we evaluated replication fork progression, stalling, new origin firing, and stability in RAD51-proficient and -depleted cells after irradiation, using a single-molecule DNA fiber technique 7,11,190. Replication fork lengths (determined using CIdU-labeling) were comparable between mock-treated RAD51-proficient and RAD51- depleted cells (5.08+0.06 and 5.01+0.10 μm in RAD51-proficient and RAD51-depleted cells, respectively; Fig. 7B). In contrast, replication fork lengths (labeled with IdU) in irradiated RAD51-depleted cells were significantly shorter than those in irradiated RAD51-proficient cells (4.62+0.04 and 2.89+0.07 μm in RAD51-proficient and -depleted cells, respectively; p=0.006, Fig. 7B). These results indicate that RAD51 is important for replication fork progression in response to irradiation. Subsequently, we verified the efficiency of replication fork restart in irradiated cells. In irradiated RAD51-depleted cells, most DNA fibers (73.42%; p=0.02) had both IdU and CldU tracts, whereas 79.47% fibers contained both IdU and CldU in irradiated RAD51-proficient cells (Fig. 7C). These results indicate that a greater proportion of replication forks failed to restart in RAD51-depleted cells in response to irradiation when compared to cells expressing RAD51. Next, we investigated new origin firing and replication fork stalling in RAD51-proficient and - depleted cells after irradiation. The number of DNA fibers only containing CldU tracts, representing new origins of replication, was significantly higher in RAD51-depleted cells than in irradiated RAD51- proficient cells (75.50% and 42.66% in RAD51-depleted and -proficient cells, respectively, p=0.005; Fig. 7D). In addition, we observed a significantly higher percentage of DNA fibers only containing IdU tracts, which represent stalled forks in irradiated RAD51-depleted cells as compared to irradiated RAD51- proficient cells (26.58% and 20.53% in RAD51-depleted and –proficient cells, respectively, p=0.02; Fig. 7E). Taken together, these results suggest that RAD51 is important for the suppression of new origin firing and replication fork stalling after irradiation.

38 Previous studies have reported that RAD51 is important for the stability of perturbed replication forks in response to replication stress 7,61,62. To verify whether RAD51 stabilizes nascent DNA strands, we labeled the replicating DNA with IdU for 30 min, irradiated it, and measured the lengths of IdU-labeled DNA fibers 5 h after exposure, as previously described 7. Under these conditions, the IdU tract lengths were slightly shorter in mock-treated RAD51-depleted cells relative to mock-treated RAD51-proficient cells; however, this difference is not statistically significant (5.13±0.01 and 5.03±0.03 μm, respectively; Fig. 7F). Similarly, the IdU tract lengths were comparable between irradiated and mock-treated RAD51-proficient cells (5.13±0.01 and 4.86±0.03 μm, respectively). In contrast, the IdU tract lengths were significantly shorter in irradiated RAD51-depleted cells than in mock-treated RAD51-proficient cells (2.71±0.05 and 4.86±0.03 μm, respectively, p=0.0003; Fig. 7F). This result indicated the degradation of IdU labeled tracts in RAD51-depleted cells. Thus, in addition to playing roles in replication fork progression and efficient restart, RAD51 is also involved in the maintenance of nascent DNA strands in response to irradiation. Previous studies have confirmed that MRE11 degrades nascent DNA strands in the absence of BRCA2, FA factors, WRN, and RAD51 in response to replication stress 7,61,62,164. Therefore, we hypothesized that MRE11 is the nuclease that degrades nascent DNA strands in the absence of RAD51 during radiation-induced replication stress. MRE11 has 3′–5′ exonuclease and endonuclease activities 198. Therefore, we first examined the direction of nascent DNA strand degradation in RAD51-depleted cells by sequentially labeling the replicating DNA, first with ldU and then with CIdU. We observed the shortening of recently replicated DNA (i.e., CIdU-labeled DNA) in irradiated RAD51-depleted cells (5.12±0.02 and 3.25±0.03 μm, RAD51-proficient and -depleted cells, respectively), suggesting that the nascent DNA strands were degraded in a 3′ to a 5′ direction (Figure 7G). Thus, the degradation of nascent DNA strands in RAD51 down-regulated cells is mediated by the exonuclease activity of MRE11. To confirm these results, we inhibited the exonuclease activity of MRE11 with mirin 199 and examined nascent DNA strand lengths. The IdU tract lengths were not shortened in irradiated cells pre-treated with mirin (Fig. 7H). However, The DNA fiber lengths in irradiation+ mirin-treated RAD51-depleted cells were comparable to those of mock+ mirin-treated RAD51-depleted cells (4.87±0.03 and 5.37±0.42 μm, respectively; Fig. 7H). It is worth noting that the lack of significant degradation of nascent DNA strands in mock-treated RAD51- depleted cells could be due to insufficient activation of MRE11 exonuclease activity, including post- translational modification of MRE11 and MRE11-RAD50-NBS1 (MRN) complex formation, in the absence of any exogenous (or sufficient) replication stress. However, further experimental evidences are required to confirm this notion. Overall, these degraded nascent DNA strands accumulate in the cytosol, triggering an innate immune response in RAD51-depleted cells after irradiation.

39 Blocking of the MRE11-mediated degradation of newly replicated genome attenuates cytosolic self-DNA accumulation in RAD51-depleted cells: To confirm that the MRE11-processed newly replicated genome accumulates in the cytosol of RAD51-depleted cells, we pre-treated cells with mirin followed by irradiation. Subsequently, we quantified the amount of ssDNA and dsDNA in the cytosolic fraction. We observed a significant reduction in the cytosolic ssDNA amount in mirin+radiation treated RAD51-depleted cells as compared to irradiated RAD51-depleted cells 8 h post-irradiation (19.91±3.05 and 14.36±0.65 pg per µg protein in radiation alone and radiation+mirin treated RAD51-depleted cells, respectively; p<0.0002; Fig. 8A). Similarly, the amount of cytosolic dsDNA was also significantly reduced in mirin+radiation treated RAD51-depleted cells as compared to irradiated RAD51-depleted cells (9.35±0.68 and 5.51±0.43 pg per µg protein in radiation alone and radiation+mirin treated RAD51-depleted cells, respectively; p<0.0002; Fig. 8B). Furthermore, the amount of cytosolic ssDNA and dsDNA in radiation+mirin treated RAD51-depleted cells was slightly higher than the mock-radiation+mirin treated RAD51-depleted cells. Thus, the excessive nuclease activity of MRE11 on the newly replicated genome in the absence of RAD51 contributes to the majority of cytosolic self-DNA accumulation upon replication stress induced by irradiation. In addition to the pharmacological inhibition of MRE11 in RAD51-depleted HT1080 cells, we confirmed the contribution of MRE11-mediated degradation of nascent DNA strands in the accumulation of cytosolic DNA in the absence of RAD51 by both pharmacological (SAHA) and genetic (shRNA) down- regulation of RAD51 expression in MRE11-deficient ATLD cells 7. We found that neither the genetic (shRad51) nor the pharmacological (SAHA)-mediated down-regulation of RAD51 significantly increased cytosolic-DNA in the absence of MRE11 upon irradiation (Figs. 11A-C). These results indicate that the RAD51-mediated blocking of MRE11-dependent degradation of nascent DNA suppresses the innate immune response upon irradiation. To further validate the influence of MRE11 on cytosolic self-DNA accumulation, we measured cytoplasmic BrdU levels after inhibiting MRE11 exonuclease activity. The cytoplasmic BrdU fluorescent intensity was unaltered in mirin+radiation treated RAD51-proficient cells as compared to irradiated RAD51-proficient cells (5.24±0.02 and 4.68±0.11 MFI per cell for radiation alone and mirin+irradiation samples, respectively; Fig. 8C). On the other hand, mirin+irradiation treated RAD51-depleted cells exhibited significantly reduced levels of cytoplasmic BrdU signal (7.46±0.57 MFI per cell; p<0.0001) as compared to irradiated RAD51-depleted cells (10.92±0.08 MFI per cell). Cell cycle specific analysis of BrdU intensity in irradiated RAD51-depleted cells further revealed compared to G1 cells (9.57±0.09 and 7.0±1.38 MFI per G1 cell for irradiation alone and mirin+irradiation samples, respectively) mirin treated RAD51-depleted S/G2 cells showed reduced cytoplasmic BrdU levels relative to control treated irradiated S/G2 cells (12.26±0.06 and 7.9±0.23 MFI per S/G2 cell for radiation alone and mirin+ irradiation samples,

40 respectively, p<0.05). Collectively, MRE11-mediated degradation of the newly replicated genome is partially responsible for the increased amount of self-DNA in the cytosol of RAD51-depleted cells. However, it is still unclear whether the attenuation of cytosolic self-DNA by the targeted inhibition of MRE11 exonuclease activity can actually limit the expression of innate immune response genes. Hence, we measured expression levels of IL-6, CSF2, and TLR9 by qRT-PCR. These three genes showed maximum upregulation in irradiated RAD51-depleted cells but were significantly reduced in mirin+ radiation-treated RAD51-depleted cells as compared to irradiated only RAD51-depleted cells (p<0.02- 0.002; Figs. 8D-F). Taken together, our data suggest that a significant proportion of cytoplasmic self-DNA comes from the MRE11-mediated excessive processing of nascent DNA strands, contributing to the initiation of innate immunity in RAD51-depleted cells in response to irradiation.

DSBs are difficult to repair in RAD51 down-regulated S/G2 cells: Although the amount of self- DNA is decreased in mirin-treated RAD51-depleted cells, the mirin treatment did not reduce cytosolic self- DNA to the basal level. This observation suggests that apart from processed nascent DNA strand, other cellular sources can contribute to the self-DNA accumulation in the cytoplasm. Furthermore, evidence indicates that in the absence of ATM protein, unrepaired DNA lesions contribute to the increased amount of cytosolic self-DNA, activating a STING-mediated innate immune response 180. Therefore, we hypothesized that a lack of RAD51 may lead to defective DSBs repair. Because the RAD51-dependent HR- mediated DSB repair pathway is involved in S/G2 phase of the cell cycle, the unrepaired DSBs may persist in S/G2 phase of RAD51-depleted cells. To verify this, we created a new cell line, 200, that expresses two different fluorescent markers depending on the cell cycle phase: G1 cells are red and S/G2 are cyan (Fig 3). Next, we enumerated γH2AX foci dissolution kinetics in G1 and S/G2 phases of HT1080-FUCCI cells 190, using an algorithm developed in MatLab. We detected γH2AX foci in G1 and S/G2 RAD51-proficient and RAD51-depleted cells at different time points after irradiation (Figs. 9A-B and Table 3). The levels of γH2AX foci were comparable between G1-phase RAD51-proficient and G1-phase RAD51-deficent cells 24 h after irradiation (3.56 and 4.54% in RAD51-proficient and -depleted cells, respectively (Fig. 9C and Table 3). In contrast, the percentages of persistent γH2AX foci were significantly elevated in RAD51- depleted S/G2 cells as compared to RAD51-proficient cells 24 h after irradiation (62.96+5.17% and 12.68%, respectively, p<0.001; Fig. 9C and Table 3). Thus, these results suggest that RAD51-depleted S/G2 cells cannot fully repair the DSBs induced by irradiation. Also, similarly to ATM, defective DSB repair in RAD51-depleted cells may partially contribute to the elevated levels of cytosolic self-DNA. To investigate the consequences of defects in DNA replication and DSB repair in RAD51-depleted cells, we evaluated their cellular phenotype. First, we examined cellular survival by colony formation assay. Similar to a previous report 50, RAD51-depleted cells were more sensitive to irradiation than RAD51-

41 proficient cells (Fig. 9D). Additionally, pre-treatment of RAD51-depleted cells with mirin partially rescued irradiation-induced cell survival (Fig. 9D). These results imply that RAD51 together with the exonuclease activity of MRE11 influences cellular survival upon irradiation. Furthermore, we noticed that inhibition of STAT3 significantly decreased the survival of RAD51-depeleted cells relative to RAD51-proficient cells (Fig. 12), suggesting a role for STAT3 in enhancing the sensitivity of RAD51-depleted cells to ionizing radiation. Second, we investigated chromosome instability. Conventional chromosome analysis of metaphase spreads revealed that the levels of chromosomal aberrations per mitotic cells were significantly elevated in irradiated RAD51-depleted cells as compared to the number of aberrations in irradiated RAD51-proficient cells (Fig. 9E). The average number of aberrations per irradiated RAD51-depleted mitotic cells was 5.18±0.42 (p<0.007), as compared to only 1.56+0.04 in irradiated RAD51-proficient mitotic cells. The number of chromatid-type aberrations was significantly elevated in irradiated RAD51-depleted mitotic cells (2.94+0.32, p=0.008) as compared to irradiated RAD51-proficient mitotic cells (0.38+0.04; Fig. 9E). MRE11-mediated degradation of a newly replicated genome in the absence of RAD51 contributes to self- DNA accumulation in the cytosol. To delineate the role of MRE11-exonuclase activity in chromosome stability maintenance, we inhibited the exonuclease activity of MRE11 in irradiated cells and examined chromosome aberrations (Fig. 9E). We noticed that the suppression of exonuclease activity of MRE11 resulted in a significant reduction of gross-chromosomal aberrations in irradiated RAD51-depleted mitotic cells (2.11+0.042; p=0.01) as compared to irradiated RAD51-depleted mitotic cells (Fig. 9E). Overall, these results demonstrate the role of RAD51 in the suppression of chromosomal instability.

2.5 Discussion

We identified the role of RAD51 in innate immune signaling in response to DNA damage and replication stress. RAD51 is recruited to the sites of DSBs, facilitating their repair in S/G2 phase cells. In addition, RAD51 is also recruited to the perturbed replication forks, preventing MRE11-mediated excessive processing of newly replicated genomes. Replication fork degradation combined with defects in DSB repair leads to the accumulation of self-DNA in the cytosol, resulting in the initiation of a STING-mediated innate immune response in cells lacking RAD51. Thus, the coordinated activities of RAD51 in DSB repair, replication fork maintenance, and innate immune response signaling provide new insights into carcinogenesis associated with defective RAD51 functions. We found that defects in RAD51 affect innate immune response signaling upon DNA damage and replication stress. This result was unexpected because there is no evidence of the RAD51 involvement in innate immunity upon irradiation. However, this phenomenon is not limited to RAD51 but has also been reported in other cell types defective in DNA repair and replication factors. For example, the DNA damage

42 sensor MRE11, plays important roles in the recognition of dsDNA and the initiation of STING-dependent immune response signaling, though this is not the case for its nuclease activity 23. The loss of ATM culminates in enhanced constitutive production of type I , elevated expression of different pattern recognition receptors, and their downstream signaling partners, which together contribute to priming the innate immune system 180. In addition, the DNA structure-specific endonuclease MUS81 that cleaves DNA structures at stalled replication forks also mediates the STING-dependent activation of innate immune signaling 69. FA proteins were recently identified as key players in immunity 94. Furthermore, RPA2 and RAD51 prevent the release of short nuclear DNA fragments into the cytosol by binding to these DNA fragments thereby preventing immune response signaling 58. Thus, DNA repair and replication factors participate in immune signaling outside their known DNA repair and replication functions. Our results suggest two potential mechanisms by which RAD51-deficiency may lead to cytosolic self-DNA accumulation: i. faulty replication fork maintenance and ii. defective DSB repair. We found that the excessive nuclease activity of MRE11 on the newly replicated genome in RAD51 down-regulated cells is a major source of cytosolic-DNA in response to replication stress induced by irradiation. The excessive processing of newly replicated genome by MRE11 in RAD51-depleted cells is not unique to radiation- induced replication stress but has also been observed in response to agents such as hydroxyurea, camptothecin, and gemcitabine 7,61,62. ii. Additionally, RAD51 is the key factor involved in HR-mediated DSB repair in S/G2 cells and its down-regulation results in DSB defects in S/G2 phase cells. These findings suggest that defects associated with DSB repair could be an additional source of cytosolic self-DNA accumulation in RAD51-depleted cells, similarly to the accumulation of cytosolic self-DNA in ATM defective cells 180. However, more studies are needed to identify the mechanism by which irreparable DSBs contribute to the cytosolic self-DNA accumulation in RAD51-depleted cells. Dysregulation of RAD51 levels and defects associated with RAD51-interacting proteins are known to cause cancer. Carcinogenesis is a multistage process resulting from a cumulative malfunctioning of DNA replication, DSB repair, and immune signaling. Defective DSB repair and replication fork processing may cause genetic and epigenetic mutations, initiating cell transformation and cancer. Our classic chromosome analysis revealed that the levels of chromosomal aberrations were significantly elevated in irradiated RAD51-depleted cells as compared to RAD51-proficient cells. Similarly to BRCA2 defective cells after replication fork stalling 62, the inhibition of the MRE11 exonuclease activity leads to a reduction in the number of chromosomal aberrations in irradiated RAD51-depleted cells, suggesting that maintenance of nascent DNA strands is critical for the prevention of chromosomal instability. Although the irradiation of RAD51-depleted cells compromises cell survival, some cells with chromosomal aberrations can still enter mitosis. Every subsequent round of replication is expected to increase the overall mutation level in surviving cells; a damaged genome can provide an opportunity for genomic rearrangement and can increase

43 genomic instability, leading to genetic changes required to initiate cell transformation and cancer. Activation of the immune system can promote tumor progression by inducing tissue remodeling, supporting angiogenesis, and providing growth factors to the tumor microenvironment that sustains proliferative signaling. This activation also stimulates survival factors that limit cell death, proangiogenic factors, extracellular matrix-modifying enzymes that facilitate angiogenesis, invasion, and metastasis, and triggers inductive signals that lead to the activation of epithelial-mesenchymal transition 201. Our study suggests that RAD51 may function as a tumor suppressor by facilitating faithful DSB repair and replication forks processing, preventing immune signaling. Additional in vivo experiments are required to study the contribution of RAD51 in tumor progression upon replication stress and DNA damage. We propose a model that represents the interplay between RAD51 and immune signaling in response to DNA damage and replication stress (Fig. 13). RAD51 is recruited to the sites of perturbed replication forks and DSBs, resulting in the blockage of the excess exonuclease activity of MRE11 on the newly replicated genome and DSB repair. Consequently, this activity limits the accumulation of self-DNA in the cytosol and prevents the initiation of STING-mediated innate immune response signaling. Thus, RAD51 plays a direct role in DNA replication and DSBs repair and is indirectly involved in immune signaling.

44 2.6 Figures & tables:

Figure 3: Characterization of HT1080-FUCCI cells: A. Schematics showing different fluorescent probes to identify DSB foci in G1 and G2 phase cells. B. Expression of mCherry (G1) and AmCyan (S/G2) fluorescence proteins. C-F. Flow-cytometric analysis of HT1080-FUCCI cells showing profile of different cell cycle stages (C), DNA content (D), pH3 population (E) and identification of M-phase cells using FUCCI colors.

45

Figure 4: Depletion of RAD51 up-regulates innate immune response pathway genes:

A. shRNA-mediated depletion of RAD51 expression in HT1080 cells: HT1080 cells stably integrating tetracycline-inducible Rad51 shRNA were treated with 1 µg/ml doxycycline (DOX) for 72 h. Subsequently, cells were exposed to 1 Gy of radiation (IR), collected at pre-established time points after irradiation. Total

46 cell lysates (50 µg) were separated on 8% SDS-PAGE and probed with anti-RAD51 and anti-Ku80 (loading control) antibodies. U- without DOX treatment. B. Heat map of significantly altered innate immune response pathway genes in RAD51-proficient and -depleted cells 4 and 8 h after irradiation. C. Graph shows fold changes in gene expression in irradiated (8 h) cells normalized to gene expression values in corresponding mock-treated cells. Exponentially growing RAD51-proficient and -depleted HT1080 cells were either mock-treated or irradiated with 1 Gy. Total RNA was prepared at indicated times after irradiation and analyzed for gene expression profiling using Human HT12v4 Arrays. The heat map for innate immune response network genes was generated with gene subsets created from the list of significant innate immune response genes using Spotfire Decision Site 9. D-I. Differences in expression levels of innate immune response pathway genes measured by quantitative real-time polymerase chain reaction (qRT-PCR): RAD51-proficient and -depleted HT1080 cells were irradiated (IR) with 1 Gy and total RNA was prepared 8 h after irradiation. mRNAs were converted into cDNA and the levels of IL-6 (D), CSF2 (E), CXCR4 (F), TNF-α (G), CMKLR1 (H) and TLR9 (I) mRNA were quantified by qRT-PCR. Error bars represent the SEM from three independent experiments; *p<0.05; **p<0.01.

47

Figure 5: Excessive DNA accumulates in the cytosol of RAD51-depleted cells: A-B. Quantification of single-strand (ssDNA) and double-strand (dsDNA) DNA in the cytosol (cyto): RAD51-proficient and -depleted HT1080 cells were irradiated with 2 Gy and were harvested at the indicated times. Subsequently, cells underwent sub-cellular fractionation and the amount of cytosolic ssDNA (A) and dsDNA (B) were quantified using OliGreen and PicoGreen Quant-iT reagents, respectively. Bars represent fold changes in the cytoplasmic DNA concentration relative to RAD51-proficient mock- treated samples. Error bars represent the SEM from four independent experiments; **p<0.01; ***p<0.001; ****p<0.0001. C. Western blots show lack of apoptosis-mediated cleavage of CASPASE-3 and PARP-1

48 in RAD51-proficient and –depleted HT1080 cells following irradiation: RAD51-proficient and -depleted HT1080 cells were exposed to 2 Gy of radiation and cells were harvested at indicated time points after irradiation. Total cell lysates (50-100 µg) were separated on 8-15% SDS-PAGE and probed with anti- CASPASE-3, anti-PARP1, and anti-Ku80 (loading control) antibodies. RAD51-depleted HT1080 cells treated with 1µM camptothecin (CPT) for 18 h was used as a positive control for apoptosis. M-mock- irradiated; arrows indicate cleaved CASPASE-1/PARP-1. D-L. Accumulation of ssDNA and dsDNA in the cytoplasm, and expression of innate immune response genes in RAD51-proficient and –depleted MCF10A (D-F), 4T1 (G-I) and HT1080+SAHA cells (J-L). The bars represent the fold changes in the cytoplasmic DNA concentration and immune response genes relative to respective mock-treated controls. Cells were either mock- or exposed to 2 Gy of radiation (IR), cytosolic fraction and total RNA was prepared at 8 (MCF10A), 16 (HT1080+SAHA) and 24 (4T1) h after irradiation, and ssDNA, dsDNA and the levels of innate immune response genes were quantified as described in materials and methods. Error bars represent the STDEV from four different experiments from two independent sets; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. SAHA-suberoylanilide hydroxamic acid; U- without SAHA treatment.

49

Figure 6: Accumulation of nuclear-derived DNA in the cytosol activates STING in RAD51-depleted cells: A-B. Representative images show accumulation of nuclear-derived DNA in the cytoplasm of mock- and irradiated (IR) G1 and S/G2 phase RAD51-proficient and –depleted cells (A). Quantification of the cell cycle-dependent cytoplasmic BrdU signal normalized to mock-treated RAD51-proficient samples (B). RAD51-proficient and -depleted HT1080-FUCCI cells were labeled with BrdU for 18-20 h, irradiated with 2 Gy and fixed with ice cold 80% methanol in PBS 8 h after irradiation. Cells were immunostained with an

50 anti-BrdU antibody under non-denaturing conditions. Subsequently, cells were imaged using a LSM510 confocal microscope, and the mean fluorescence BrdU signal in the cytoplasm of G1 and S/G2 phase cells was quantified using ZEN 2009 (version 6,0,0303) Software (Carl Zeiss, Jena, Germany). Bars represent mean cytoplasmic BrdU intensity per cell relative to respective control G1 and S/G2 cells. More than 150 cells were used for quantification in each condition. Error bars represent the STDEV from two-four independent experiments; Scale bars are 10 µm; *p<0.05. C-D. Representative images show STING clustering in the cytoplasm of mock- and irradiated RAD51-proficient and –depleted cells 8 h after irradiation (C). Quantification of percentage of cells with STING clustering signal relative to the total number of counted cells (D). RAD51-proficient and -depleted HT1080 cells were irradiated with 2 Gy and fixed with 4% PFA at indicated times after irradiation. Subsequently, cells were immunostained with an anti-STING antibody, imaged using a LSM510 confocal microscope; the STING clustering signal in the cytoplasm was quantified using Imaris Software (Bitplane). More than 200 cells were used for quantification in each condition. Error bars represent the STDEV from three independent experiments; Scale bars are 20 µm; **p<0.01; ***p<0.001. E. Representative Western blots show phosphorylation of STING, TBK, and STAT3 in RAD51-proficient and –depleted cells after irradiation. RAD51-proficient and -depleted HT1080 cells were irradiated with 2 Gy radiation and harvested at the indicated times. Total cell lysates (100-150 µg) were separated on 8-10% SDS-PAGE and probed with the indicated antibodies and the anti-Ku80 antibody (loading control).

51

52 Figure 7: RAD51 blocks the MRE11-mediated degradation of nascent DNA strands upon irradiation: A. Representative images show the co-localization of RAD51 foci with EdU and γH2AX foci. HT1080 cells were pulse-labeled with EdU for 30 min and immediately irradiated with 1Gy, fixed with 4% PFA 4 h after irradiation, and immunostained with anti-RAD51 and anti-γH2AX antibodies. EdU was detected using the Click-IT reaction. Scale bars are 5 µm. B. Replication fork progression was reduced in RAD51- depleted cells in response to irradiation. DNA fiber length distributions in RAD51-proficient and -depleted cells are shown before and after irradiation. Cells were labeled with IdU for 30 min, treated with and without 1Gy radiation, and labeled with CIdU for another 30 min. DNA fibers were immunostained with anti-BrdU (rat and mouse) antibodies. Images were captured using a fluorescence microscope and IdU (before) CldU (after) lengths were measured using Axiovison Software. More than 200 DNA fibers were evaluated in each sample. Each data point is the average of three independent experiments. C-E. Replication forks stall in RAD51-depleted cells after irradiation. Percentages of replication fork restarts in irradiated RAD51- proficient and -depleted cells relative to mock-treated cells were evaluated using the {(IdU→CldU)/[IdU+(IdU→CldU)] formula (C). Percentage changes in new origin firing in irradiated RAD51-proficient and -depleted cells as compared to mock-treated cells were calculated using the {CldU/[CldU+(IdU→CldU)] formula (D). Percentage changes in replication forks stalling in RAD51- proficient and -depleted cells as compared to mock-irradiated cells were evaluated using the {IdU/[IdU+(IdU→CldU)] formula (E). More than 200 DNA fibers were evaluated in each sample. Each data point is the average of two independent experiments. Error bars represent the STDEV; *p<0.05; **p<0.01. F. Nascent DNA strands were shortened in RAD51-depleted cells. Replicating DNA in RAD51-proficient and -depleted cells was labeled with IdU for 30 min and irradiated with 1 Gy. DNA fibers were immunostained with anti-BrdU (mouse) antibodies. DNA fiber images were captured using a fluorescence microscope, and IdU tract lengths were measured using the Axiovision Software. The frequency distributions of the lengths of more than 100 DNA fibers were calculated from three independent experiments in each group. G. Newly replicated DNA was shortened in the absence of RAD51. Shortening of newly synthesized DNA fiber length distributions in RAD51-proficient and -depleted cells 5 h after irradiation. Cells were labeled with IdU for 30 min and then with CIdU for another 30 min. Subsequently, cells were irradiated with 1 Gy. Five hours after irradiation, DNA fibers were immunostained with anti-BrdU (rat and mouse) antibodies, images were captured using a fluorescence microscope, and CldU tract lengths were measured using the Axiovison Software. More than 200 DNA fibers were evaluated in each sample. Each data point is the average of three independent experiments. H. RAD51 blocks the MRE11-mediated degradation of nascent DNA strands in response to irradiation. RAD51-proficient and -depleted cells were labeled with IdU for 30 min. Cells were pre-treated

53 with or without 100 µM of the MRE11 exonuclease inhibitor (mirin), irradiated with 1 Gy. Five hours after irradiation, DNA fibers were immunostained with anti-BrdU (mouse). Images were captured using a fluorescence microscope and ldU tract lengths were measured using the Axiovison Software. More than 100 DNA fibers were evaluated in each sample. Each data point is the average of two independent experiments.

Figure 8: Inhibition of MRE11 exonuclease activity blocks the expression of innate immune response genes in RAD51-depleted cells: A-B. The quantification of ssDNA and dsDNA DNA in the cytosol. RAD51-proficient and -depleted cells were pre-treated with mirin (25 µM), irradiated with 2 Gy and harvested 8 h after irradiation. Subsequently, cells were subjected to sub-cellular fractionation and the amount of cytosolic ssDNA (A) and dsDNA (B) were quantified using OliGreen and PicoGreen Quant-iT reagents, respectively. The bars represent the changes in the cytoplasmic DNA concentration relative to respective mock-treated samples. Error bars represent STDEV from four independent experiments; ***p<0.001. C. The quantification of the cytoplasmic BrdU signal normalized to mock-treated RAD51-proficient samples. RAD51-proficient and - depleted HT1080-FUCCI cells were labeled with BrdU for 18-20 h, pre-treated with mirin (25 µM), irradiated with 2 Gy and fixed with ice cold 80% methanol in PBS 8 h after irradiation. Cells were immunostained with an anti-BrdU antibody under non-denaturing conditions. Subsequently, cells were imaged using a LSM510 confocal microscope and the BrdU signal in the cytoplasm of G1 and S/G2 phase cells was quantified using the ZEN 2009 (version 6,0,0303) Software (Carl Zeiss, Jena, Germany). Bars represent mean cytoplasmic BrdU fluorescence intensity per cell relative to respective mirin-treated control

54 G1 and S/G2 cells. More than 150 cells were used for quantification in each condition. Error bars represent the STDEV from two-four independent experiments; *p<0.05. D-F. Down-regulation of IL-6, CSF2, and TLR9 expression in cells pre-treated with the MRE11 exonuclease inhibitor. RAD51-proficient and - depleted cells were pre-treated with mirin (25 µM) and irradiated with 1 Gy; total RNA was prepared 8 h after irradiation. mRNAs were converted into cDNA and the levels of IL-6 (D), CSF2 (E), and TLR9 (F) mRNA were quantified by qRT-PCR. Error bars represent the SEM from three-four independent experiments; *p<0.05; **p<0.01.

Figure 9: RAD51 is critical for DSB repair in S/G2 cells and chromosome stability maintenance: A-C. DSBs persist in RAD51-depleted S/G2 cells: Representative images show appearance and disappearance of γH2AX foci in G1 (A) and S/G2 (B) cells. Cell cycle-dependent γH2AX foci dissolution kinetics in RAD51-proficient and -depleted HT1080 cells stably expressing two different cell cycle markers, mCherry (G1) and AmCyan (S/G2) (C). Cells were irradiated with 1 Gy and immunostained with anti-γH2AX at the indicated times after irradiation. Cells were imaged using a confocal microscope and γH2AX foci in 100-120 red and Cyan fluorescent cells representing G1 and S/G2 phases, respectively were counted using the Matlab software (Mathworks, MA). Error bars represent the STDEV from three-four independent experiments. Scale bars are 5 µm. D. RAD51-depleted cells are sensitive to irradiation. RAD51-proficient and -depleted cells plated in six well plates were exposed to 0.5 Gy of radiation with or

55 without mirin treatment and cell survival was analyzed by a colony formation assay. Colonies were fixed and counted 8-10 days after irradiation. The relative survival efficiencies were plotted. The error bars represent the STDEV calculated from triplicate wells; **p<0.01. E. RAD51 suppresses chromosome instability upon irradiation. Number of chromatid and chromosome-type aberrations in mock- and - irradiated RAD51-proficient and -depleted cells pre-treated with and without mirin. Exponentially growing cells were either mock-treated or irradiated with 1 Gy, and the metaphase chromosomes spreads were prepared 16 h after treatment. Chromosomal aberrations in more than 100 metaphase spreads were scored from two to four independent experiments in each group. Error bars represent the STDEV calculated from two-four independent experiments. *p<0.05.

Figure 10: A. Heat map of significantly altered genes in RAD51-proficient and -depleted cells at 4 and 8 h after radiation. Exponentially growing RAD51-proficient and -depleted HT1080 cells were either mock-treated or irradiated with 1 Gy. Total RNA was prepared at 4 and 8 h after irradiation and analyzed for gene expression profiling using Human HT12v4 Arrays. A heat map was generated from the list of significantly (p<0.05) altered genes that showed >1.5 fold up- and down-regulation using Spotfire Decision Site 9. B. Ingenuity pathway analysis showing activation of innate immune response signaling pathway in RAD51-depleted cells 8 h after DNA damage.

56

Figure 11: MRE11 deficient cells do not accumulate DNA in cytosol Representative Western blots show suberoylanilide hydroxamic acid (SAHA) and shRad51-mediated down-regulation of RAD51 in MRE11-deficient ATLD cells (A). The quantification of ssDNA and dsDNA DNA in the cytosol of MRE11-deficient cells with and without RAD51 (B and C). RAD51 was down- regulated either by shRNA or SAHA treatment in MRE11-deficient ATLD cells and then irradiated with 2 Gy and harvested at indicated times after irradiation. Subsequently, cells were subjected to sub-cellular fractionation and the amount of cytosolic ssDNA and dsDNA were quantified using OliGreen and PicoGreen Quant-iT reagents, respectively. The bars represent the changes in the cytoplasmic DNA concentration relative to respective mock-treated samples. Error bars represent STDEV from two independent experiments.

57

Figure 12: RAD51-depleted cells are sensitive to STAT3 inhibitor: RAD51-proficient and -depleted cells plated in six well plates were treated with 10 and 25 µM STAT3 inhibitor for 24 hours and cell survival was analyzed by a colony formation assay. Colonies were fixed and counted 8-10 days after the treatment. The relative survival efficiencies were plotted. The error bars represent the STDEV calculated from triplicate wells; ***p=0.0004.

58

Figure 13. RAD51 interconnects between replication fork processing, DSB repair, and innate immune responses. The model depicts the mechanism of innate immunity initiation in RAD51-depleted cells due to defective replication fork processing and DSB repair.

59 Table: S1

Sl. No Species Gene Name Forward primer sequence (5'-3') Reverse primer sequence (5'-3')

1 Human IL-6 CCTTCGGTCCAGTTGCCTTCT GCATTTGTGGTTGGGTCA

2 Human TLR9 CGCCCTGCACCCGCTGTCTCT CGGGGTGCTGCCATGGAGAAG

3 Human TLR4 CGAGGAAGAGAAGACACCAGT CATCATCCTCACTGCTTCTGT

4 Human CXCR4 TGACGGACAAGTACAGGCTGC CCAGAAGGGAAGCGTGATGA

5 Human CSF2 AGCATGTGAATGCCATCCAG AGGGGATGACAAGCAGAAAG

6 Human TNFa AG TGACAAGCCTGTAGCCC GCAATGATCCCAAAGTAGACC

7 Human CMKLR1 CACCTGCTGCTCAGACTGAA GTGAGAGAAGGATAGCCGCC

8 Human HDAC9 CACCTGCTGCTCAGACTGAA GTGAGAGAAGGATAGCCGCC

9 Human CXCL2 TCACCTCAAGAACATCCAAAGTGTG CTTCAGGAACAGCCACCAATAAGC

10 Mouse IL-6 AGGATACCACTCCCAACAGACCT GTATGAACAACGATGATGCACTTG

11 Mouse TLR9 ACTGAGCACCCCTGCTTCTA AGATTAGTCAGCGGCAGGAA

12 Mouse CSF2 CCTGGGCATTGTGGTCTACAG TCAAAGAAGCCCTGAACCTCC

Table 2: List of primers used for qRT-PCR in chapter 2.

60 Table S2 Co efficients Cell and Cell Cycle Parameters Values Std Error a 55.12 4.40 b 0.02 0.01 Rad51-proficient-total c 59.00 4.30 d 0.45 0.08 a 28.69 6.01 b 1.11 0.38 Rad51-deficient-total c 83.14 1.88 d 0.01 0.00 a 43.63 11.31 b 0.00 0.01 Rad51-proficient-G1 phase c 70.04 10.26 d 0.33 0.11 a 66.25 6.05 b 0.06 0.01 Rad51-proficient-S/G2 phase c 63.00 11.41 d 1.12 0.40 a 52.43 12.87 b 0.21 0.09 Rad51-deficient-G1 phase c 52.83 14.38 d 0.00 0.01 a 25.57 3.45 b 0.98 0.24 Rad51-deficient-S/G2 phase c 84.05 1.30 d 0.00 0.00 Table 3: Parameters of DSB dissolution kinetics in figure 8.

Equation: y=ae-bx + ce-dx; b and d are the rate constants; a and c are the coefficients; a is percentage of γH2AX foci that are eliminated with fast kinetics; c is percentage of γH2AX foci that are eliminated with slow kinetics; d is the rate constant for the slow kinetics; b is the rate constant for the fast kinetics; x is time (in hours).

61

Chapter 3 NBS1 Counters cGAS-Mediated Premature Senescence

62 3.1 Summary: Persistent DNA damage response signaling (DDR) is a common critical mechanism that establishes and maintains cellular senescence phenotype. Dysfunctional telomeres cause defective DDR signaling and cellular senescence, but the link between these events remains unclear. Here, we show that chromosome mis-segregation due to defective cell cycle checkpoints in response to dysfunctional telomeres creates a preponderance of chromatin fragments in the cytosol, leading to a premature senescence phenotype. This phenomenon is dependent on telomere shortening but is mediated by cGAS (cyclic GMP-AMP synthase) recognizing a sub-set of cytosolic chromatin fragments and then activating the STING (stimulator of interferon genes) cytosolic DNA-sensing pathway and downstream interferon signaling. Significantly, genetic and pharmacological manipulation of cGAS not only attenuates immune signaling but also prevents premature cellular senescence in response to dysfunctional telomeres. Paradoxically, we found that NBS1, a well-known DSB sensing and repair factor, counters cGAS binding to the cytosolic chromatin fragments and the subsequent activation of immune signaling and the premature senescence phenotype. Thus, our study uncovers a previously unidentified intrinsic cellular regulatory mechanism, specifically the NBS1 signaling axis, that prevents the erroneous activation of cGAS in response to cytosolic chromatin fragments produced during various DNA metabolic activities and, ultimately, unwarranted premature senescence.

3.2 Introduction:

Chromosome ends are composed of long stretches of TTAGGG repeat sequences and associated proteins that work together to form a cap, called a telomere 202,203. The telomere-specific protein complex, known as shelterin, is composed of six molecules (TRF1, TRF2, TIN2, RAP1, TPP1 and POT1) 203. Shelterin complex protects chromosome ends against degradation, fusion, recombination and recognition by the DNA damage machinery. Defects in shelterin components and telomerase directly and adversely affect telomere stability204. Shelterin components act to prevent activation of the canonical DDR signaling at the telomeres. An un-capped telomere are recognized as single- or double-stranded DNA damage, leading to activation of DNA damage response (DDR) signaling. DDR activation can result in chromosome fusions, and subsequent progression through the cell cycle causing breakage-bridge-fusion cycles, eventually leading to premature senescence, cell death or genome instability 205.

As discussed earlier, DSBs pose a serious threat to cell viability and genome stability. In mammalian cells, majority of DSBs are repaired by NHEJ and HRR pathways 171,172. Additionally, an alternative end-joining pathway is activated when classical NHEJ and HR repair pathways fail 206,207. Despite our extensive knowledge about DSB repair pathways in eukaryotic cells, not much work has been done to determine how cells respond to telomeric DSBs. Earlier reports indicate that telomeric DNA

63 repeats, if damaged, are irreparable 208-210. However, emerging evidence indicates that reparability of telomeric DNA damage depends on cellular growth conditions. Exponentially proliferating cells employ HRR to repair telomeric DSBs, while senescent cells are incapable of repairing telomeric DNA lesions 211.

Lack of suitable model systems is one of major limitations in understanding how cells respond to teloemric DNA damage or dysfunctional telomeres. Telomere uncapping by disruption of the shelterin complex is one of the major approaches used to study the mechanisms of telomere maintenance. Additionally, Isce1 endonuclease-based introduction of DSBs in the sub-telomeric regions has been used to study the repairing ability of sub-telomeric DSBs 212. Recently, TRF1 fused with the Fokl enzyme has been used to induce DSBs in the telomere213. Additionally, persistent DSBs associated with telomeres several days after ionizing radiation exposure has also been used to study telomere DNA damage responses 208-210. Moreover, laser micro-irradiation has been used to cause damage to telomeres 214. Recently, a KillerRed chromophore fused to TRF1 has been used to cause oxidative damage-mediated telomere dysfunction215. Based on the results of these studies, it is clear that loss or dysfunction of shelterin, or damage to the telomeric DNA repeat sequences, causes telomere dysfunction, resulting in failure to protect chromosome ends and the defective activation of DNA damage response signaling (DDR) 216,217. Additionally, activation of the homologous recombination and non-homologous end joining pathways in response to dysfunctional telomeres results in genomic instability and cellular senescence 218,219. Yet, the link between defective DDR activation, chromosome instability and cellular senescence in response to dysfunctional telomeres remains unclear. Recent findings have provided mechanist insights into how genomic instability due to defective DDR signaling triggers immune signaling. Cyclic guanosine monophosphate (GMP)-adenosine monophosphate (AMP) synthase (cGAS), a cytosolic DNA sensor, detects cytosolic DNA as a danger-associated molecular pattern (DAMP) and initiates immune responses 107. Several reports suggest that the accumulation of chromatin fragments in the form of micronuclei due to improper DDR signaling and the subsequent rupture of the nuclear envelope underlies cGAS activation 119- 124. As result of cGAS activation in response to cytosolic chromatin fragments, cells either become cancer or undergo irreversible cell cycle arrest, i.e., cellular senescence220-224. Yet, it is unclear what factors determine whether cells undergo senescence or become cancerous. Therefore, additional factors, such as a high-affinity for chromatin fragments, may compete with cGAS to dictate the pathway’s cellular fate.

The DDR is composed of DNA damage sensing (MRE11/RAD50/NBS1) and the activation of upstream (ATM and ATR) and downstream (CHK1/2) kinases. Among these factors, NBS1, protein defective in Nijmegen breakage syndrome (NBS), is a multifunctional protein that senses DNA double- strand breaks (DSBs) and plays critical roles in DSB repair, DDR signaling and genome stability

64 maintenance. The human NBS1 protein consists of 754 amino acids and contains two functional domains at the N-terminus (1–196 amino acids) and the C-terminus (665–693 amino acids). The N-terminal sequence consists of a forkhead-associated (FHA) domain (20–108 amino acids) and a BRCA1 C-terminus (BRCT1: 111–197 amino acids; BRCT2: 219–327 amino acids) domain. Both the FHA and BRCT1/2 domains directly interact with the phosphorylated H2AX (γH2AX) independently of MRE11 and BRCA1 225. The C-terminus of NBS1 contains an MRE11-binding region (682–693 amino acids) 226 and an ATM- binding region (734–754 amino acids) 227. The NBS1-ATM interaction results in the phosphorylation of NBS1 in response to DSBs both in vitro and in vivo228,229. In addition, NBS1 has been implicated in inflammatory signaling and premature senescence27,28,230,231. However, how NBS1 connects genome instability, immune signaling and premature senescence has not yet been identified.

In this study, we have developed novel approaches to cause dysfunctional telomeres. Additionally, we provide novel evidence that NBS1, previously known for its functions in DNA damage sensing, repair and signaling, is involved in innate immune signaling by counteracting cGAS binding to cytosolic chromatin fragments. Furthermore, we show that cells that harbor fused chromosomes in response to dysfunctional telomeres progress to mitosis, due to defective G2-M checkpoint, leading to the accumulation of chromatin fragments in the cytoplasm. Within the cytoplasm, both NBS1 and cGAS compete for the chromatin fragments harboring DSBs, and the FHA domain containing NBS1 is preferred over cGAS. Although the molecular mechanism by which NBS1 and cGAS compete with each other is unknown, removing NBS1 enhances the interaction of cGAS with cytoplasmic chromatin fragments, which culminates in the activation of cGAS-mediated innate immune signaling and premature senescence. Thus, the antagonism between cGAS and NBS1 constitutes a crucial regulatory mechanism for cGAS signaling and functions to suppress unwarranted premature senescence.

3.3 Materials and Methods:

Cell lines: HT1080, U2OS, HeLa and IMR90 cells were obtained from the American Type Culture Collection (ATCC; USA). Human telomerase reverse transcriptase (hTERT) immortalized 82-6 fibroblasts and ATLD cells were described previously 7. NBS (NBS1-deficient) and NBS cells complemented with different NBS1 constructs were described previously 7,232 (Figs 14C). All cell lines were grown in standard

tissue culture conditions at 5% CO2 and maintained in Dulbecco modified eagle medium supplemented with 10% fetal bovine serum, 2 mM glutamine and 0.1 mM nonessential amino acids. To make stable HT1080+dd-Cas9/SgTelomere RNA and U2OS+dd-Cas9/SgTelomere RNA cell lines, HT1080 and U2OS cells were first infected with lentivirus carrying tetracycline-inducible dd-Cas9 and then placed under

65 puromycin selection (0.5 and 2.5 µg/ml for HT1080 and U2OS cells, respectively). Stable clones were isolated, and dd-Cas9 expression was evaluated by western blotting and immunostaining using anti-FLAG antibody after treating cells with Doxycycline (500 ng/ml) and Shield1 (1 µM) for 24 hours. Then, stable clones positive for dd-Cas9 were infected with lentivirus harboring sgTelomere 233 and mCherry fluorescent protein and then plated with a single cell in each well of a 96-well plate. Wells containing single colonies were selected, and clones positive for mCherry were further validated (Figs 14A). To make a stable HT1080-dominant negative (DN)-TRF2 (45 to 453 aa regions of TRF2) cell line, HT1080 cells were infected with lentivirus carrying tetracycline inducible DN-TRF2 and then placed under puromycin selection (0.5 µg/ml). Stable clones were isolated and treated with Doxycycline (1µg/ml) for 72 hours, and both western blotting and immunostaining using anti-FLAG antibodies verified the expression of DN-TRF2 (Figs 13B). To make a stable HT1080-shcGAS RNA cell line, HT1080-sgTelomere-dd-Cas9 cells were infected with pooled lentiviral particles carrying cGAS-specific shRNA (Sigma) and hygromycin resistance gene. Cells were then placed under hygromycin selection (150 µg/ml). Stable clones were isolated, and the expression of cGAS was verified by western blotting (Fig. 14D).

DNA manipulation and construction of the expression vectors: Standard molecular biology procedures were used to make all mammalian expression plasmids. A panel of human cGAS-specific shRNAs was purchased from Sigma (TRCN0000128706, TRCN0000128310, TRCN0000149984, TRCN0000146282, TRCN0000150010). pCW-57.1-Cas9 (#50661) and sgRNA targeting telomeric DNA (#51024) were purchased from Addgene. Degradation domain (DD-domain) was added to the N-terminal of Cas9 by PCR amplification followed by ligation into the Nhe1 sites of pCW57.1-Cas9. The dominant negative TRF2 fragment corresponding to the 45-453 amino acid region of TRF2 was PCR amplified using TRF2 cDNA and cloned into Nhe1/Sal1 sites of pCW-57.1 (Addgene; #41393). We confirmed the sequence of all constructs prior to use. All FLAG-tagged protein constructs are N-terminal fusions.

Lentiviral production: 293T cells were co-transfected with highly purified expression plasmids and pPLP2/pLP1/pVSVG using lipofectamine 2000 (Invitrogen), according to manufacturers’ instructions. Cell culture supernatant containing viral particles were collected 72 hours after the transfection, filtered through 0.22 microns PVDF membrane filters (Millipore) and the flow through was used for infecting the cells.

Chemicals: We used CDK1 inhibitor RO3306 (Tocris, #4181), 6-thio-dG (Sigma, #1296), MRE11- exonuclease inhibitor mirin (Sigma, #475954), doxycycline (Sigma, #D9891), cGAS inhibitor RU.521 (AOBIOUS, #37877), and Shield1 (Cheminpharma, #S1-0005).

Doxycycline and Shield1 treatment: a. To induce DSBs in the telomeric DNA in HT1080 or U2OS- sgTelomere-dd-Cas9 cells, we co-treated cells with 500 ng/ml doxycycline (Sigma) and 1µM Shield1

66 (Cheminpharma) for 24 hours. Cells were then washed three times with warm PBS and allowed to recover in regular growth media, and samples were collected at different time points. b. To induce DN-TRF2 expression, we treated HT1080-DN-TRF2 cells with 1000 ng/ml doxycycline (Sigma) for 72 hours. Cells were then washed three times with warm PBS and allowed to recover in regular growth media, and samples were collected at different post-DOX withdrawal (w/d) time points.

6-thio-dG treatment: Twenty-four hours after plating, cells were first treated with 3-5 µM 6-thio-dG for 72 hours, washed three times in warm PBS, and then cultured in drug free medium for an additional 0-10 days. Samples were collected at different post-6-thio-dG recovery times.

CDK1 inhibitor treatment: Exponentially growing cells were co-treated with 1 µM RO3306 dissolved in DMSO, doxycycline (500 ng/ml) and Shield1 (1µM) for 24 hours. Cells were then washed three times with PBS and cultured in growth medium containing 1 µM CDK1 inhibitor, and samples were collected at different post-treatment times.

Mirin treatment: Thirty-six hours after seeding, cells were first treated with 10 µM mirin for 72 hours together with 6-thio-dG, then 6-thio-dG was removed and cultured in mirin-containing medium for 0-10 days. Samples were collected at different times post-telomeric DNA damage. cGAS inhibitor (RU.521) treatment: Cells were treated with 1 µM RU.521 dissolved in DMSO for the entire duration of the experiment. Fresh drug was added to the medium for every 48 hours for continuous inhibition of cGAS.

Ionizing radiation: Exponentially growing cells were exposed to 2-5 Gy γ-ray using a γ-irradiator (Mark 1 irradiator, JL Shepherd & Associates), as described previously.3

Cell extracts and western blotting: Whole-cell extracts were prepared by suspending cell pellets in RIPA buffer containing protease (PMSF, aprotinin, leupeptin, pepstatin A, DTT; all at 1:1000 dilutions) and inhibitors (Na3VO4 at 1:500 and NAF at 1:200 dilution) on ice for 30 minutes followed by centrifugation at 14,000 RPM for 30 minutes at 4oC to remove insoluble material. Thirty to one hundred micrograms of whole-cell extracts were resolved by 6-12% SDS-PAGE, transferred onto Polyvinylidene difluoride (PVDF) membranes and reacted with different antibodies.

Antibodies used in this study: The primary antibodies used in this chapter are listed in Table 4. For western blotting, HRP conjugated goat anti-rabbit and anti-mouse secondary antibodies were purchased from BioRad and used at 1:1000 dilution in 5% BSA or milk. For indirect immunofluorescence staining, fluorescent conjugated secondary antibodies Alexa-488, Alexa-555 and Alexa-633 were purchased from Molecular Probes (Invitrogen) and used at 1:1000 dilution.

67 Indirect immunostaining: Approximately 0.5-1 × 105 cells were plated in a six well plate containing cover glasses and incubated for 36 hours. Cells were treated with various chemicals for different time periods, as described above. Cells were fixed with 4% PFA for 20 minutes at room temperature at different post-treatment times and subjected to indirect immunofluorescence, as described previously.3 Briefly, cells were permeabilized in Triton X-100 (0.5% in PBS) on ice for 5 minutes, washed three times with PBS, incubated in blocking solution (5% goat serum in PBS) at room temperature for 60 minutes, and then incubated with primary antibodies (diluted in 5% goat serum) at room temperature for another 3 hours or at 4oC overnight. Then, cells were washed with wash buffer (1% BSA in PBS), incubated with appropriate secondary antibodies (1:1000 in 2.5% goat serum, 1% BSA, and PBS) at room temperature for 60 minutes, washed five times with 1% BSA, and mounted with mounting medium containing DAPI (Vectashield).

Telomere Immunofluorescence-FISH: Cells grown on coverslips were fixed in 4% PFA for 20 minutes at room temperature and then subjected to indirect immunostaining, as described above. Immediately after incubation with a secondary antibody, coverslips were washed, re-fixed in 4% PFA for 10 minutes, washed in ethanol series solutions (70%, 90% and 100%, 5 minutes each), and denatured with hybridization mixture [(70% formamide, 1% blocking reagent, 10 mM Tris-HCl pH 7.4 and 0.5 µM Cy3 labeled TelC PNA probe (PNA Bio; #F1002] for 6 minutes at 80°C, followed by incubation in a 37oC dark humidified chamber for 4 hours. Samples were then washed with Wash Solution I (70% formamide, 0.1% BSA, 10mM Tris-HCl pH 7.4, 2 times at a 15-minute interval) and Wash Solution II (water, 150mM NaCl, 0.07% between 20 and 100mM Tris-HCl pH 7.4, 3 times at 10 minute intervals). Samples were then mounted with DAPI-containing mounting medium, and images were acquired using a LSM510 Meta confocal microscope (Zeiss).

Image acquisition and foci dissolution kinetics assay: Images were captured using an LSM 510 Meta laser scanning confocal microscope with a 63X1.4 NA Plan-Apochromatic oil immersion objective. Images were taken at z-sections (15-20 sections) of 0.35-µm intervals using the 488-nm (EGFP and Alexa 488), 543-nm (Alexa 555), 633-nm (Alexa 633) and 405-nm (for DAPI) lasers. The tube current of the 488-nm argon laser was set at 6.1 A. The laser power was typically set to 3-5% transmission with the pinhole opened to 1-2 Airy units. To count foci, the z-sections were assembled using the Imaris software and analyzed as described previously 3. For counting foci, we utilized the spot detection function of the Imaris software. We quantified foci from images of 100-150 cells for each time point from two to three independent experiments.

Micronuclei imaging and quantification: Cells fixed with 4% PFA were mounted with DAPI- containing mounting solution (Vectashield). Images were acquired using an Axio-Invert (Zeiss) microscope using the DAPI channel, and the exposure time ranged from 400 to 500 µ seconds per frame. Micronuclei

68 in 300-400 cells per experimental condition in two to three independent experiments were calculated in a blinded fashion using ImageJ (NIH).

Metaphase spreads preparation: Chromosome aberrations were carried out as described previously3. Twenty-four to seventy-two hours after inducing dysfunctional telomeres, we prepared chromosomes by accumulating metaphases in the presence of 0.1 mg/mL colcemid (Irvine Scientific) for 4 hours. Cells were trypsinized and washed once with PBS and incubated in 10 mL of hypotonic solution (0.075 M KCl) at 37°C for 15 minutes. Cells were pre-fixed with 1/10 volume of ice-cold methanol-acetic acid (3:1 ratio) in hypotonic solution and then centrifuged at 1000 rpm at 4°C for 5 minutes. Then, the cells were fixed with methanol-acetic acid (3:1 ratio) on ice for 30 minutes and kept at −20°C until used. Cells were dropped onto pre-cleaned slides, stained with 5% Giemsa stain (KaryoMAX, GIBCO) at room temperature for 4 minutes, washed with dH2O, and then mounted using Vectashield mounting medium. Images were taken using an Olympus microscope (100X objective) equipped with an Image Spot camera (Spot Imaging Solutions), and chromosome aberrations in 150-200 metaphase spreads were scored 3.

Senescence assay: Freshly fixed samples were subjected to β-galactosidase (β-gal) staining using β-gal staining kit (Cell Signaling) following manufacturer’s instructions. Random images were acquired using the 10X objective of a KEYONCE Microscope in a blinded manner. We counted in a blinded fashion the number of β-gal positive cells in 20-30 random fields consisting of 1000-5000 total cells per experimental condition in two to three independent experiments.

Quantitative real-time polymerase chain reaction (qRT-PCR): cDNA was synthesized from 1-3 μg of total RNA using SuperScript III Reverse Transcriptase (18-080-051; Fischer Scientific) in a total volume of 20 µl, according to manufacturer’s instructions. The cDNA was subjected to qRT-PCR for several genes using the primer sets (Table 5), CFX96 Touch Real-Time PCR Detection System (Bio Rad) and iTaq Universal SYBR Green Supermix (Bio Rad; #1725121,), according to manufacturer’s instructions. Relative gene expression was determined by the ΔΔCT method. The difference in cycle times, ΔCT, was determined as the difference between the tested gene of interest and the reference housekeeping β-actin gene. We then obtained ΔΔCT by finding the difference between the groups. The fold change (FC) was calculated as FC=2–ΔΔCT. All primers were purchased from Invitrogen. qRT-PCR assays were carried out in triplicates for each sample, and the mean value was used to calculate mRNA expression levels.

Flow cytometry: Cells were harvested from 10 cm tissue culture dishes at indicated times after the induction of telomeric DSBs using trypsin-EDTA and washed in PBS before fixation by a drop-wise addition of ice-cold 70% ethanol. After fixation on ice (or storage at −20°C), cells were washed once in PBS and permeabilized in 0.5% Triton X-100/PBS solution on ice for 10 minutes. Cells were washed three

69 times in PBS and incubated in blocking solution (5% goat serum in PBS) at room temperature for 60 minutes, then incubated with anti-phospho-histone H3 (Serine 10, EMD Millipore) antibody at room temperature for another 4 hours. After washing twice in PBS, cells were incubated in Alexa 488 conjugated secondary antibody (1:1000 in 2.5% goat serum, 1% BSA, and PBS) at room temperature for 60 minutes. Pellets were washed in 1% BSA+PBS washing solution three times and re-suspended in PBS containing 200 µg/mL far red DNA stain (Thermo Fisher Scientific, #10348) and 100 µg/mL RNAse A (Roche) for 30 minutes in the dark. Flow cytometry was performed on an Amnis Flow cytometer (Millipore) and analyzed using Amnis software. Single cells and G1/S/G2 peaks were manually gated. The gates for phospho-H3 positive cells were chosen by comparing with a population in which the antibody was omitted during processing.

Statistical analysis: Data are expressed as means ±SEM or STDEV of at least two to four independent experiments. The Student t-test was performed on all values, and p<0.05 was considered statistically significant. GraphPad Prism (version 7.0) was used to create the graphs.

3.4 Results

Accumulation of micronuclei due to defective G2/M checkpoint correlate with immune signaling and premature senescence in response to telomeric DSBs: To induce DSBs in telomeric DNA, we used a novel telomere-directed CRISPR/Cas9 genome editing system 234. This two-component system consists of a single-guide RNA (sgRNA) and a Cas9 nuclease (Fig. 15A). The sgRNA (sgTelomere) contains a 22-nucleotide telomere targeting sequence with improved sgRNA-Cas9 assembly characteristics233. This well-characterized sgRNA has been used previously to study telomere dynamics in living human cells233. Cas9 is an endonuclease with two active DNA cutting sites, one for each strand of the helix234. The sgTelomere directs the Cas9 nuclease to telomeric DNA. To induce regulated DSBs within the telomeric DNA, we fused Cas9 to FK506 binding protein 12 (FKBP or dd-degradation domain) to produce dose-dependent, small-molecule-regulated post-translational stabilization. In the presence of Shield1 – a small (750 Da), membrane-permeant, stabilizing ligand – the dd-tagged Cas9 (dd-Cas9) is stabilized (protected from proteasomal degradation) and accumulates in the cell. Ligand-dependent stabilization occurs very quickly 235. Additionally, Cas9 is controlled by a tetracycline inducible (TetO) promoter. Hence, adding Shield1 to cells treated with Doxycycline (DOX) stabilizes Cas9, and withdrawing Shield1 results in the rapid degradation of Cas9 (Figs. 15A top panel and 14A). Upon binding to the sgTelomere, Cas9 generates DSBs specifically in telomeric DNA (Fig. 15A, bottom panels). Thus, we created a novel cell line to induce DSBs in the telomeric DNA.

70 Evidence indicates that telomeric DSBs (T-DSBs) are more refractory to repair than non-telomeric DSBs 208. To test telomeres’ ability to repair DSBs, we utilized our newly developed sgTelomere RNA+dd- Cas9 system to induce DSBs in the telomeric DNA. We found the maximum number (20-25 DSBs/cell) of DSBs in the telomeric region 24 hours after adding DOX+Shield1 into the culture medium (Fig. 14B). We observed that most (~70%) of these DSBs were repaired within 24 hours, but 6-8 DSBs persisted even 72 hours after the withdrawal of DOX+Shield1 (Fig. 15B). Thus, under our experimental conditions, a small fraction of telomeric DSBs were irreparable.

Robust activation G2/M checkpoint after DNA damage allows cells to repair DSBs before entering mitosis and it has an estimated threshold of 10-20 DSBs below which cells can enter mitosis 3,236,237. We found that the number of irreparable telomeric DSBs is below the threshold required for the G2/M checkpoint activation. Therefore, we systematically examined the induction and maintenance of the G2/M checkpoint by western blotting and flow cytometry. We found that phosphorylation of CHK1/2, crucial kinases involved in G2/M checkpoint maintenance, was defective in response to T-DSBs but not after ionizing radiation (Fig. 15C, top panel), suggesting that the impaired G2/M checkpoint is specific to telomeric DSBs. Further examination of G2/M checkpoint using the anti-phospho-histone 3 (pH3) antibody followed by FACS analysis revealed that cells harboring unrepaired T-DSBs progressed to mitosis similar to mock samples (Fig. 15C, bottom panel). Thus, irreparable telomeric DSBs lead to imperfect G2/M checkpoint activation.

Reports suggest that failure to arrest cells in G2/M phase converts unrepaired DNA damage to chromosomal aberrations during the G2 to M phase transition 3,236). To investigate the consequences of imperfect to G2/M checkpoint activation in response to irreparable telomeric DSBs, we evaluated chromosome instability. Conventional chromosome analysis of metaphase spreads revealed significantly higher levels of chromosome aberrations, specifically dicentric chromosomal (chromosome-chromosome fusion) aberrations, in cells with telomeric DSBs than in mock-treated cells (Fig. 15D). These results clearly demonstrate that defective G2/M checkpoint in response to telomeric DSBs generate chromosomal aberrations.

The presence of fused (dicentric) chromosomes leads to the generation of micronuclei as the cell cycle progresses into the mitotic phase. To determine the fate of fused chromosomes after cytokinesis, we examined the frequency of micronuclei formation after the induction of telomeric DSBs. We found that the frequency of micronuclei positive cells was significantly higher in cells with telomeric DSBs than in control cells (Fig. 15E). Furthermore, we observed that the number of micronuclei increased gradually, peaked at 24 hours and remained elevated even 72 hours after the removal of DOX+Shield1 (Fig. 15E). Overall, these results reveal that defective telomeric DSBs repair lead to end-to-end chromosome fusions due to defective

71 G2/M checkpoint activation together with improper chromosome segregations, resulting in the generation of micronuclei upon cytokinesis.

To determine if defective DSBs in the telomeric DNA lead to cellular senescence, we counted the number of β-galactosidase (β-gal) positive cells, a well-accepted marker for the senescence phenotype. We found that the number of β-gal positive cells had significantly increased 5-10 days after the induction of DSBs in the telomeric DNA compared to mock-treated cells. Thus, induction of DSBs in the telomeric DNA significantly increases cellular senescence (Fig. 15F). Because evidence suggests that telomere shortening is a major driver of cellular senescence, we examined telomere length in cells with damaged telomeric DNA by Telomere Shortest Length Assay (TeSLA), as described previously 238. Surprisingly, we found that the telomere lengths were comparable in cells with and without telomeric DSBs (Fig. 20A). These results suggest that the observed cellular senescence in response to telomeric DNA damage is not due to telomere shortening, which reveals that other pathway(s) contribute to premature senescence in response to telomeric DNA damage.

Besides serving as a marker for genomic instability, micronuclei can also trigger immune signaling 120-124. We wanted to determine if the increased micronuclei levels resulting from defective telomeric DSBs repair activate immune signaling. We found that phosphorylation of IRF3, STAT1, TBK1 and STING, accepted markers for cytosolic DNA-mediated immune signaling pathway, increased gradually after the induction of T-DSBs (Fig. 15G). Similarly, expression of a panel of innate immune genes also increased gradually after the generation of T-DSBs (Fig. 15H). We further corroborated our results showing that the micronuclei-mediated activation of immune signaling is not specific to one cell type by testing our notion on U2OS cells stably expressing SgTelomere RNA and dd-Cas9 (Figs. 20B-D). Then, we assessed the generation of micronuclei and the activation of immune signaling in U2OS cells. As with the HT1080 cells, inducing DSBs in the telomeres not only increased the number of micronuclei (Fig. 20B) but also activated proteins involved in immune signaling as well as the expression of inflammatory cytokine genes (Figs. 20C-D). Taken together, we concluded that micronuclei generation correlates positively with the activation of immune signaling, and increased immune signaling could well be the reason for premature senescence in cells with telomeric DSBs.

The progression of G2/M phase cells with defective telomeric DSBs repair results in the generation of micronuclei, culminating in immune signaling-mediated premature senescence. Therefore, as reported previously 122, we hypothesize that preventing the progression of G2/M cells with unrepaired or misrepaired T-DSBs to the mitotic phase might prevent premature senescence. To verify this notion, we used a selective small-molecule inhibitor of CDK1, RO-3306, that reversibly arrests cells at the G2/M border of the cell cycle 239. As with a previous report 122, blocking G2 to M progression did not affect telomeric DSB induction

72 (Fig. 20A) but prevented telomeric DSB-mediated micronuclei formation (Fig. 20B) and inflammatory signaling (Figs. 20C-D) in HT1080+sgTel RNA+dd-Cas9 cells upon induction of telomeric DSBs. Furthermore, blocking G2 to M-phase progression also prevented the onset of premature senescence in cells with telomeric DNA damage (Fig. 20E). These results clearly demonstrate that a defective G2/M checkpoint in response to telomeric DSBs is a major contributor to micronuclei generation, which results in immune signaling and premature senescence.

Dysfunctional telomeres result in premature senescence through cytosolic chromatin fragments: Next, we wanted to determine whether the cytosolic chromatin fragment-mediated immune signaling and cellular senescence that we observed with sgTelomere RNA+dd-Cas9-induced telomeric DSBs occurred in other telomere dysfunction models. First, we stably expressed a tetracycline-inducible dominant negative form of TRF2, TRF2ΔBΔM (DN-TRF2), in HT1080 cells (Fig. 14B). DN-TRF2 expression has been shown to induce telomere dysfunction 240. Similar to previous findings 240, inducing DN-TRF2 expression triggered telomeric DNA damage-induced foci, known as TIFs (Fig. 16A). As with sgTelomere+dd-Cas9-induced DSBs, we observed significantly higher levels of dicentric chromosomes and micronuclei generation in cells expressing DN-TRF2 than in mock-treated cells (Figs. 16B-C). Importantly, we also found that the presence of micronuclei in the cytoplasm of DN-TRF2-expressing cells correlated well with activation of STING, IRF3, STAT1 and TBK1, as well as elevated expression of innate immune genes (Figs. 16D-E). As expected, a significantly higher proportion of DN-TRF2-expressing cells than mock-treated cells exhibited β-gal staining (Fig. 16F). Thus, micronuclei-mediated immune signaling contributes to premature senescence upon telomere dysfunction induced by DN-TRF2 expression.

To further validate these findings, we induced dysfunctional telomeres using the newly reported nucleoside analogue, 6-thio-2′-deoxyguanosine (6-thio-dG) 241-243. 6-thio-dG is recognized by telomerase and incorporated into de novo–synthesized telomeres. This alters the overall chemistry, structure and function of the shelterin complex (such as G-quadruplex forming properties and protein recognition) 244,245, leading to their recognition as telomeric DNA damage signals, but almost exclusively in cells expressing telomerase 241. As reported previously, treating cells with 6-thio-dG (3 µM) caused TIF in the telomeres (Fig. 16G). As with cells expressing sgTelomere+dd-Cas9 and DN-TRF2, we observed significantly elevated levels of micronuclei formation in 6-thio-dG-treated cells compared to mock-treated cells (Fig. 16H). Importantly, increased micronuclei formation correlated well with activation of STING, IRF3, STAT1 and TBK1 and expression of immune signaling genes in 6-thio-dG-treated cells compared to mock- treated cells (Figs. 16I-J). As seen in other experiments, a significantly higher proportion of 6-thio-dG (3 µM)-treated cells than mock-treated cells underwent senescence (Fig. 16K).

73 To demonstrate that this phenomenon is not limited to transformed cells or one cell type, we induced telomeric DNA damage in hTERT-immortalized human skin fibroblasts by 6-thio-dG treatment. As seen in HT1080 cells, telomeric DNA damage not only increased the number of micronuclei but also activated immune signaling and premature senescence in hTERT-immortalized fibroblasts (Figs. 16L-N). Overall, these results indicate that cytosolic chromatin fragments resulting from dysfunctional telomeres, irrespective of the mechanism that caused telomere dysfunction or the type of cells used, initiates immune signaling, culminating in premature senescence.

To verify that this premature senescence phenotype caused by cytosolic chromatin fragments is distinct from replicative exhaustion-mediated senescence, we assessed micronuclei formation and immune signaling in primary fetal lung fibroblasts (IMR90) undergoing replicative senescence, a widely used model for replicative senescence caused by progressive telomere shortening 246. We noticed that the frequency of senescence increased significantly after culturing the cells for 3 weeks (Fig. 16O). However, neither the number of micronuclei nor the expression of innate immune signaling pathway genes changed substantially between early and late passage IMR90 cells (Figs. 16P-Q). Thus, our results clearly indicate that cytosolic chromatin fragments generated by dysfunctional telomeres cause premature senescence, which is distinct from replicative senescence caused by telomere shortening.

Only a sub-set of cytosolic chromatin fragments recruit cGAS: Emerging evidence suggests that cGAS senses cytoplasmic chromatin fragments, which activates downstream immune signaling and cellular senescence. Therefore, we examined the presence of cGAS in the chromatin fragments in the cytoplasm. As with previous reports, we noticed clear localization of cGAS in the cytosolic chromatin fragments in cells with dysfunctional telomeres (Figs. 17A-C). Surprisingly, we consistently observed that only a sub- set, not all cytosolic chromatin fragments were positive for cGAS, regardless of how we caused dysfunctional telomeres. This raises the question of why only certain cytosolic chromatin fragments harbor cGAS.

Evidence indicates that after micronuclei are formed, they are initially coated with a nuclear membrane, and rupturing this nuclear membrane releases chromatin into the cytosol. This phenomenon is required to recruit cGAS to the cytosolic chromatin fragments and to activate STING. Therefore, we examined the relationship between cGAS, nuclear envelope rupture and STING activation by co- immunostaining after allowing cells to accumulate cytosolic chromatin fragments in response to dysfunctional telomeres. We found that most cGAS-bound cytosolic chromatin fragments did not have a Lamin A/C coating (Fig. 17D) but had activated STING in the cytosol (Fig. 17E). Yet, we found another population of cytosolic chromatin fragment that was neither Lamin A/C - nor cGAS-positive. This

74 observation raises the question of whether other factors also negatively regulate cGAS binding to cytosolic chromatin fragments.

Besides cGAS, we noticed the γH2AX signal, a well-known surrogate marker for DSBs, in a major fraction of cytosolic chromatin fragments (Figs. 17A-C), suggesting the occurrence of DSBs in these chromatin fragments. Interestingly, cytosolic chromatin fragments that were negative for Lamin A/C coating were also positive for γH2AX foci. Therefore, factors known to be recruited to the sites of DSBs might play a negative role in recruiting cGAS to cytosolic chromatin fragments in response to dysfunctional telomeres.

Because several DNA damage sensing, repair and signaling factors, including MRE11/ RAD50/ NBS1 (MRN complex), MDC1, 53BP1 and ATM, are recruited to the sites of DSBs, it is reasonable to speculate that one of these factors might negatively regulate the recruitment of cGAS to cytosolic chromatin fragments containing DSBs. We examined the localization of MDC1, ATM, 53BP1, MRE11, and NBS1 to the cytosolic chromatin fragments and found that 18-35% of micronuclei were positive for MDC1, MRE11, pATM, and NBS1 in response to dysfunctional telomeres (Figs. 17F-I). However, we were unable to detect any 53BP1 foci in the cytosolic chromatin fragments formed as a result of dysfunctional telomeres (data not shown). Interestingly, we noticed co-localization of ATM and MDC1 with cGAS in the cytoplasmic chromatin fragments (Figs. 17F-G). In contrast, we rarely detected co-localization of NBS1 and MRE11 with cGAS in the same cytoplasmic chromatin fragment (Figs. 17H-I), indicating a mutually exclusive relationship between NBS1/MRE11 and cGAS with regard to binding to cytosolic chromatin fragments

We further observed that some cytosolic chromatin fragments possessed neither cGAS nor any of the DSB sensing/repair/signaling factors examined, suggesting that additional factors modulate cGAS binding to cytosolic chromatin fragments. It is well known that chromatin compaction limits DNA metabolic factors’ access to the DNA. Therefore, we examined the status of chromatin compaction and cGAS binding to cytosolic chromatin fragments. To our surprise, we found that cytosolic chromatin fragments that contain heterochromatic chromatin marker (HP1α) are completely devoid of cGAS (Fig. 17J). Furthermore, relaxing compact chromatin using a histone deacetylase (HDAC) inhibitor, vorinostat, did not affect micronuclei formation (Fig. 22), but increased the levels of cGAS-bound micronuclei (Fig. 17K). Taken together, these results suggest that chromatin organization and the presence or absence of certain DSB sensors might regulate cGAS’s access to cytosolic chromatin fragments in response to dysfunctional telomeres.

NBS1 negatively regulates the interaction of cGAS with cytosolic chromatin fragments: Our finding that NBS1 and cGAS do not co-localize onto cytosolic chromatin fragments prompted us to investigate NBS1’s involvement in cGAS binding to these fragments using NBS1-deficient NBS cells and

75 NBS cells complemented with wild type and mutant NBS1. Initially, we noticed significantly higher frequency of micronuclei in NBS cells relative to NBS cells complemented with wild type NBS1 (NBS+WT NBS1) in response to dysfunctional telomeres (Fig. 18A). Subsequently, we found that cytosolic chromatin fragments in cells lacking NBS1 had significantly higher number of cGAS than NBS+WT NBS1 cells (Fig. 18B). To rule out the possibility that the increased cGAS occupancy on the cytosolic chromatin fragments is due to a defective nuclear membrane, we examined the presence of nuclear envelope coating on micronuclei in NBS and NBS+WT NBS1 cells. We found no statistically significant difference in the levels of nuclear envelope coating between NBS1-deficient and NBS+WT NBS1 cells (Fig. 18C). Additionally, similarly to cGAS, NBS1 was excluded from the heterochromatin marker positive micronuclei (Fig. 18D). Furthermore, immune signaling activation, as assessed by both western blotting (Fig. 18E) and qRT-PCR (Fig. 18F), was markedly higher in NBS-deficient cells than in NBS+WT-NBS1 cells in response to dysfunctional telomeres. Ultimately, these increased levels of immune signaling resulted in higher incidences of β-gal positive NBS1-deficient cells than NBS+WT NBS1 cells (Fig. 18G). Thus, NBS1 is clearly involved in negatively regulating cGAS binding to chromatin fragments.

To determine whether the physical location of NBS1 is a factor that keeps cGAS from binding to cytosolic chromatin fragments, we used ΔFHA (FE3; 1-82 aa deleted) and FL-NBS1 harboring two mis- sense point mutations (Gly 27 Asp and Arg 28 Asp; FHA-2D) in the FHA domain (Fig. 18H) to prevent recruitment of NBS1 to DSB sites via interaction with γH2AX 225,226. We found that the levels of cGAS- positive cytosolic chromatin fragments were significantly higher in NBS cells expressing ΔFHA and FHA- 2D NBS1 than in NBS+WT-NBS1 cells (Fig. 18I). Thus, the physical presence of NBS1 on the cytosolic chromatin fragments negatively regulates cGAS recruitment to those fragments, but the exact mechanism is not known.

NBS1 stability at damaged DNA is regulated by ATM-mediated phosphorylation 247. Evidence shows that NBS1 recruits ATM to DSB sites via its C-terminal regions (703 to 754 aa) 248, resulting in NBS1 phosphorylation as well as NBS1 stabilization at DSB sites 228,247. We initially examined co- localizing of NBS1 and ATM with cytosolic chromatin fragments. We found that a major percentage of cytosolic chromatin fragments were positive for both phosphorylated ATM (S1981) and NBS1 (Fig. 23A). To further explore how ATM-mediated NBS1 stabilization at chromatin fragments affect cGAS binding to cytosolic chromatin fragments, we used NBS cells stably expressing NBS1 and lacking ATM recruitment domain (Δ703-754 or ΔATM). We found that the number of cGAS-positive micronuclei was significantly increased in NBS cells stably expressing Δ703-754 (Fig. 18J). To further validate role of ATM in modulating cGAS binding to cytosolic chromatin fragments, we stably depleted ATM in HeLa cells using ATM-specific shRNA. We found that shRNA mediated ATM depletion did not affect micronuclei

76 formation in HeLa cells (Fig. 23B), but ATM-depleted HeLa cells showed increased levels of cGAS bound to cytosolic chromatin fragments (Fig. 23C) and enhanced IRF3 activation in response to dysfunctional telomeres (Fig. 23D). Overall, our results indicate that ATM indirectly regulates the binding of cGAS onto the cytosolic chromatin fragments via its phosphorylation-mediated stabilization of NBS1 onto the cytosolic chromatin fragments.

In addition to recruiting ATM to DSB sites, NBS1 also recruits MRE11 through its C-terminal 682 to 693 aa region. We found that expression of the NBS1-ΔMRE11 (truncation involving hMRE11 binding domain; 682 to 693 aa) construct in NBS cells could not alter cGAS binding to cytosolic chromatin fragments (Fig. 18J), suggesting no role for MRE11 in cGAS dynamics at the cytosolic chromatin fragments.

To rule out MRE11’s involvement in the accumulation of cGAS onto the cytosolic chromatin fragments, first, we depleted MRE11 in HT1080 cells using MRE11-specific shRNA (Fig. 24A). We found that MRE11-depletion (and MRE11-deficiency or inhibition of exonuclease function of MRE11) did not alter the number of micronuclei that formed in response to dysfunctional telomeres (Fig. 24B). Additionally, the accumulation of cGAS in the cytosolic chromatin fragments was not affected in MRE11- depleted cells (Figs. 24C). Furthermore, similar to a previous report 23, relative to MRE11-proficient HT1080 cells, the expression of immune signaling genes was marginally reduced in the MRE11 depleted cells with dysfunctional telomeres (Fig. 24F-G). Second, we used MRE11-deficient ATLD (Ataxia- telangiectasia-like disorder) cells to further verify the role of MRE11 in cGAS binding to cytosolic chromatin fragments. ATLD cells expressed c-terminally truncated MRE11 7. We noticed only a small fraction of the micronuclei were positive for cGAS in the micronuclei of ATLD cells in response to dysfunctional telomeres (Fig. 24D). Furthermore, both cytosolic DNA sensing signaling and the expression of innate immune pathway genes were not altered in MRE11-defective ALTD cells (Fig. 24H). Finally, since MRE11 possesses both exonuclease and endonuclease activities, we inhibited the exonuclease activity of MRE11 using mirin to further validate that blocking exonuclease activity of MRE11 augments cGAS accumulation in the micronuclei. As with MRE11 depletion, pre-treating cells with mirin and then inducing telomeric DNA damage did not alter cGAS recruitment to the micronuclei (Fig. 24E). Additionally, mirin treatment marginally reduced the expression of genes involved in immune pathways in response to dysfunctional telomeres (Fig. 24H). Taken together, our data strongly indicate that neither MRE11 nor its exonuclease activity is responsible for the binding of cGAS to cytosolic chromatin fragments.

Next, to determine whether NBS1 and cGAS compete for cytosolic chromatin fragments, we overexpressed full-length (FL) and 1-200 amino acid region containing FHA and BRCT domains (NBS11-

200aa) of NBS1 in NBS1-WT HT1080+sgTelomere +dd-Cas9 cells (Fig. 18K) and examined cGAS

77 recruitment to cytosolic chromatin fragments To our surprise, we noticed reduced levels of cGAS, but not micronuclei number (Fig. 18L), in cells overexpressing FL-NBS1 and NBS11-200aa in response to dysfunctional telomeres (Fig. 18M). Taken together, these results reveal that the physical stability of NBS1 is critical for blocking cGAS’s access to the cytoplasmic chromatin fragments. cGAS is essential for premature senescence in cells with telomere dysfunction: Regardless of competition between NBS1 and cGAS for the cytosolic chromatin fragments in eliciting premature cellular senescence, cGAS binds to a small fraction of chromatin fragments. To better define the involvement of cGAS in initiating premature senescence in response to cytoplasmic chromatin fragments caused by dysfunctional telomeres, we depleted cGAS using shcGAS RNA (Fig. 14D) and caused dysfunctional telomeres by treating cGAS-depleted cells with 6-thio-dG. cGAS depletion did not affect the formation of micronuclei following telomeric DNA damage (Fig. 19A). However, cGAS-depleted cells exhibited reduced activation of IRF3, TBK1 and STAT1 compared to cGAS-proficient cells (Fig. 19B). Furthermore, cGAS-depleted cells failed to upregulate the expression of immune signaling genes following telomeric DNA damage (Fig. 19C). Reduced immune signaling resulting from cGAS depletion prevented micronuclei-mediated premature senescence in response to dysfunctional telomeres (Fig. 19D). To validate these results, we inhibited cGAS binding to DNA using a recently identified small molecule inhibitor, RU.521 249. RU.521 treatment did not affect micronuclei formation after 6-thio-dG treatment (Fig. 19E) but similar to cGAS-depleted cells, RU.521-treated cells showed attenuated activation of IRF3 and STAT1 (Fig. 25) and reduced expression of immune signaling genes compared to mock-treated cells (Fig. 19F). Furthermore, the proportion of β-gal staining positive cells was also significantly reduced when cells were simultaneously exposed to RU.521 and 6-thio-dG (Fig. 19G). Taken together, the cGAS-mediated cytosolic DNA-sensing pathway is necessary to initiate premature senescence in response to dysfunctional telomeres.

3.5 Discussion

Cellular senescence restricts unlimited cell proliferation and plays a critical role in both aging and tumor suppression. We can distinguish two types of cellular senescence: replicative, which depends on telomere shortening, and stress-induced premature senescence, which does not depend on telomere shortening. Of the many stimuli that may cause premature senescence, this study provides strong evidence that the accumulation of cytosolic chromatin fragments due to dysfunctional telomeres initiates the cGAS-mediated cytosolic DNA sensing pathway and the premature senescence phenotype. Additionally, this study identified the involvement of NBS1, a DNA damage sensing, repair and signaling factor, in regulating cGAS-mediated cytosolic DNA sensing pathways and premature senescence. Ultimately, the data presented in this study further support the idea that a telomere shortening-independent mechanism contributes to the premature senescence phenotype.

78 It is generally understood that most cancer and tissue culture cells have overcome cellular senescence and achieved immortality. However, the present study shows that cancer and hTERT immortalized cells can be induced to enter cellular senescence independently of telomere shortening when the cGAS-mediated DNA sensing pathway senses cytosolic chromatin fragments in response to dysfunctional telomeres. Notably, the amount of telomeric DNA-damage required to trigger premature senescence is substantially lower than that required to induce cell death (Fig. 26). This is because the cell’s inability to accurately process dysfunctional telomeres, combined with an improperly functioning cell-cycle checkpoint mechanism, leads to the accumulation of cytosolic chromatin fragments, culminating in the activation of the cGAS-mediated cytosolic DNA sensing pathway-dependent premature senescence. Consequently, suppressing cell cycle progression reduces cellular senescence even in cells that harbor high levels of dysfunctional telomeres (Fig. 21E). Similarly, it has been shown that genotoxic stress rapidly activates G2/M checkpoint arrest; however, release from G2 arrest occurs before DSB repair is complete together with chromosome-chromosome fusions, resulting in micronuclei formation and activating cGAS- mediated cytosolic DNA sensing signaling 3,122. Thus, dysfunctional telomeres together with imperfect checkpoint activation contribute to micronuclei formation, cGAS-mediated immune signaling activation and a telomere shortening-independent premature senescence phenotype.

Our study and previous findings 120,121,124 strongly show that cGAS plays an essential role in initiating the signals required for telomere shortening-independent premature senescence. However, the question of how cGAS establishes the senescence phenotype remains unanswered 161. This is intriguing because cGAS recognition of cytosolic DNA fragments and the subsequent activation of innate immune signaling precede the appearance of a senescence phenotype. On the other hand, depleting cGAS or inhibiting its activity suffices to ameliorate premature senescence in the presence of cytosolic chromatin fragments due to dysfunctional telomeres. These results indicate that the initial activation of the cGAS- mediated cytosolic DNA sensing pathway suffices to establish a senescence phenotype. Reports suggest that IL-8 and IL-6 feed back to the secreting cells to reinforce senescence signaling 146,147. Therefore, as suggested previously 161,250, cGAS-mediated expression of pro-inflammatory factors, such as IFN-β, IL-6 and IL-8, in response to dysfunctional telomeres could serve as a paracrine signal that is required to establish a senescence phenotype.

Previous studies have reported the accumulation of cGAS in the nuclear envelope ruptured micronuclei, i.e. cytosolic chromatin fragments 120-124; however, no information is available on whether cGAS binds to all or only a sub-set of cytosolic chromatin fragments. In this study, we provide substantial evidence that only a sub-set of cytosolic chromatin fragments is bound to cGAS. Furthermore, this study

79 together with previous findings provides evidence of multiple regulatory mechanisms for avoiding erroneous cGAS activation: 1. cGAS has weak DNA binding activity (kD~20 µM) and requires additional co-factors to interact with DNA 251; 2. chromatin architecture regulates the efficiency of cGAS’s interaction with cytosolic chromatin fragments; and 3. cGAS competes with other DNA damage sensing factors. In this regard, we provide novel evidence that the physical presence of NBS1 hijacks cytosolic chromatin fragments from cGAS, which suggests a novel role for this well studied DDR factor in regulating unwarranted activation of the cGAS-mediated cytosolic DNA sensing pathway.

NBS1 was long thought to mainly sense DSBs and transmit DNA damage signaling to regulate genome integrity and cell fates. Our current findings reveal that NBS1 has an additional biological function: it negatively regulates inflammatory signaling by counteracting cGAS binding to cytosolic chromatin fragments. NBS1 lacks any known enzymatic activities, but its physical presence on the cytosolic chromatin fragments suffices to suppress the cGAS-mediated cytosolic DNA sensing pathway. However, it is not clear how NBS1 counters cGAS binding to cytosolic chromatin fragments. We can speculate two major possibilities: 1. NBS1’s affinity to cytosolic chromatin fragments may be stronger than that of cGAS; or 2. NBS1’s interaction with the damaged chromatin may alter the organization of cytosolic chromatin fragments, which might prevent cGAS from binding to chromatin fragments. Yet, the reason why cGAS recognizes a small number of chromatin fragments in the presence of a functional NBS1 is unclear. 1. This could be because of NBS1’s stability due to either its post-translational modification or its interacting partners, as a result of which, NBS1 might have a weaker affinity than cGAS for cytosolic chromatin fragments, or the abundance of NBS1 might be reduced in certain cell populations. 2. The nature of cytosolic chromatin fragments might also explain why cGAS recognizes a small number of these fragments in NBS1-proficient cells. For example, the efficiency of cGAS activation is very high for oxidatively modified self-DNA 107,252, which might serve as an ideal substrate for cGAS binding but not for the DSB sensor, NBS1. Nevertheless, additional experimental evidence is required to precisely determine how NBS1 counters cGAS binding to cytosolic chromatin fragments.

Another major finding of our study is that NBS1 exploits one of its interacting partners, ATM, to stabilize at the cytosolic chromatin fragments and counter cGAS binding to these fragments. It has been shown that the loss of ATM culminates in enhanced constitutive production of type I interferons and elevated expression of different pattern recognition receptors and their downstream signaling partners, which together contribute to priming the innate immune system 35. Thus, our findings further support an indirect role for ATM in cytosolic chromatin fragment-mediated immune signaling through phosphorylation-dependent stabilization of NBS1, though, intriguingly, our study excludes the involvement

80 of another NBS1 interacting partner and a previously known cytosolic DNA sensor, MRE11 23. Therefore, NBS1 partners with ATM for its own binding stability on the cytosolic chromatin fragments and to counter cGAS-mediated cytosolic DNA sensing and immune signaling.

We combine the data presented in this manuscript in the form of a model in figure 26. The progression of cells with fused chromosomes to mitosis due to defective G2/M checkpoint in response to dysfunctional telomeres leads to the accumulation of cytosolic chromatin fragments. NBS1, with a high affinity for damaged chromatin, binds to chromatin fragments, which limits the access of cGAS to all cytosolic chromatin fragments. However, limited binding of cGAS to the cytosolic chromatin fragments triggers innate immune signaling, which initiates and establishes a telomere shortening-independent premature senescence phenotype. Finally, NBS1 binding to damaged chromatin fragments in the cytoplasm not only attenuates cGAS binding to these fragments, but it also regulates unwarranted cellular senescence.

In conclusion, our data not only provide evidence for a previously undefined role for NBS1, but they also support the idea that mammalian cells have intrinsically developed multiple signaling mechanisms, specifically the NBS1-dependent DNA sensing axis, to avert erroneous cGAS sensing of self- DNA in the cytosol to prevent unwarranted cellular outcomes, including premature senescence. Thus, NBS1 serves as a negative regulator of the cGAS-mediated cytosolic DNA sensing pathway with important implications for cellular homeostasis and human health.

81 3.6 Figures & tables:

Figure 14: Preparation of stable cell lines.

A. Schematics showing sgTelomere RNA and dd-Cas9 expressing plasmid constructs used for inducing DSBs in the telomeric region (top two panels). Constitutively expressed sgTelomere RNA binds to telomeric DNA; addition of Doxycycline (DOX) induces expression of dd-Cas9, and addition of Shield1 stabilizes dd-Cas9. Subsequently, dd-Cas9 binds to sgTelomere RNA and induces DSBs in the telomeric DNA. TRE-tetracycline responsive element (top). Stable clones were selected based on FLAG- ddcas9 and mcherry-sgtelomere RNA expression (bottom). B. Western blot shows expression of D/N TRF2 in mock- and doxycycline treated HT1080 cells stably expressing doxycycline inducible D/N TRF2 plasmid. Cells were harvested 72 hour after the DOX treatment and total cell lysate was subjected to western blotting and subsequently probed with anti-FLAG and anti-TRF2 antibodies. C. Wetsern blot shows expression of NBS1 in NBS and NBS mutant cells complemented with different NBS1 constructs. D. Western blot shows small hairpin (sh) cGAS RNA mediated stable depletion of cGAS in HT1080 cells. Clone numbers 1 and 5 showing maximum cGAS depletion was used for further experiments. UT-untansfected; THP1 cell lysate was used as a positive control for cGAS expression.

82

83 Figure 15: Telomeric DSBs trigger immune signaling and initiate premature senescence.

A. Concurrent expression of sgTelomere RNA and Cas9 induces DSBs in the telomeric region. Western blot shows expression of dd-Cas9 during and after doxycycline (DOX) and Shield1 treatment in HT1080 cells stably expressing sgTelomere RNA and DOX inducible dd-Cas9 (top). Indirect immunostaining show induction of DSBs in the telomeric regions. HT1080 cells were immunostained with DSB marker, H2AX, and concomitantly with telomere markers, anti-TRF2 antibody (middle) and telomeric DNA probe (telC PNA probe; bottom). Representative images are shown. B. Telomeric-DSBs are partially reparable. Western blot shows expression of dd-Cas9 (top), representative images show presence of H2AX foci (middle) and the bar graph shows kinetics of telomeric DSBs disappearance (bottom) at pre- and different post-doxycycline+Shield1 withdrawal times. C. Telomeric DSBs are less efficient in G2/M check-point activation. Western blot shows extent of activation of ATM, CHK1 and CHK2 kinases (top) and the graph shows flow cytometry analysis of phosphorylated Histone 3 (pH3) positive mitotic-phase cells (bottom) at different mock- and post-DOX+Shield1 treatment times. Ionizing radiation was used as a positive control for checking the activation of checkpoint proteins. D. Induction of telomeric DSBs causes chromosome- chromosome fusions. Representative images show normal and fused chromosomes in mock- (i) and DOX+Shield1 (ii) treated HT1080+sgTelomere+dd-Cas9 cells (top). Bar graph shows frequency of fused chromosomes in mock- and at different DOX+Shield1 withdrawal times in HT1080+sgTelomere+dd-Cas9 cells (bottom). E. Telomeric DSBs causes micronuclei formation. Bar graph shows frequency of micronuclei at different post-DOX+Shield1 treatment times and corresponding mock-treated HT1080+sgTelomere+dd-Cas9 cells. F. Telomeric DSBs trigger premature senescence phenotype. Bar graph shows frequency of β-gal staining positive cells at seven days after the DOX+Shield1 treatment. G- H. Immune signaling is activated in response to telomeric DSBs. Western blots show increased activation of proteins involved in immune signaling (G) and the graph shows increased expression of genes involved in immune pathway (H) in response to telomeric DSBs. Error bars in all the graphs represent the STDEV/ SEM from three-four independent experiments. *P (range) < 0.05-0.0001.

84

Figure 16: Dysfunctional telomeres initiate innate immune signaling and premature senescence phenotype.

A. Expression of dominant negative (D/N) TRF2 causes telomere dysfunction. Indirect immunostaining shows induction of TIFs in cells expressing D/N TRF2. HT1080 cells stably expressing D/N-TRF2 were immunostained with TIF markers, H2AX and 53BP1, and concomitantly with telomere marker, telomeric DNA probe (telC PNA probe). B-C. Expression of D/N-TRF2 leads to chromosome-chromosome fusions and micronuclei formation. Bar graph shows increased frequency of chromosome-chromosome fusions (C) and micronuclei formation (D) in DOX- but not in mock (M)-treated HT1080 cells stably expressing D/N- TRF2. D-E. Innate immune signaling is activated in response to D/N-TRF2 expression. Western blot shows increased activation of proteins involved in innate immune signaling (D) and graph shows increased expression of genes involved in immune signaling (E) in DOX-treated HT1080 cells stably expressing D/N- TRF2 relative to mock-treated same cells. T-36 represents 36 hour of after DOX addition or prior to doxycycline removal. F. Dysfunctional telomeres cause premature senescence. Bar graph shows frequency of β-gal stained HT1080 cells stably expressing D/N-TRF2 cells at 15 days after +/- DOX treatment. G. 6- thio-dG treatment induces dysfunctional telomeres. Indirect immunostaining shows induction of TIFs in HT1080 cells 48 hour after 3µM 6-thio-dG treatment. Cells were immunostained with TIFs markers, H2AX and 53BP1, and concomitantly with telC PNA probe. H. 6-thio-dG treatment results in micronuclei formation in HT1080 cells. Bar graph shows increased frequency of micronuclei formation in 6-thio-dG

85 treated HT1080 cells relative to mock-treated cells. I-K. 6-thio-dG induced dysfunctional telomeres activate innate immune signaling. Western blot shows increased activation of proteins involved in innate immune signaling (I), graph shows increased expression of genes involved in immune pathway (J) and elevated levels of β-gal staining (K) in 6-thio-dG (dG) treated HT1080 cells relative to mock (M)-treated HT1080 cells. L-N. Induction of dysfunctional telomeres in hTERT immortalized human skin fibroblasts activate proteins involved in innate immune signaling (L), elevated expression of immune signaling genes (M) and increased levels of premature senescence phenotype (N). O-Q. Replicative senescence is independent of immune signaling activation. Graph shows frequency of β-gal staining (O), micronuclei frequency (P) and expression of innate immune pathway genes (Q) in IMR90 cells undergoing replicative senescence. Error bars in all the graphs represent STDEV/ SEM from two-four independent experiments. *P (range) < 0.05-0.001

86

87 Figure 17: cGAS recruitment is restricted to a fraction of cytosolic chromatin fragments in response to dysfunctional telomeres.

A-C. A major fraction of cytosolic chromatin fragments harbor DSBs (H2AX) but only a subset of chromatin fragments possesses cGAS. Representative images show co-localization of cGAS and H2AX in micronuclei (left panels) and the graph shows the frequency of micronuclei harboring either H2AX, cGAS or both (right panels) in HT1080 cells stably expressing sgTelomere+dd-Cas9 (A), D/N TRF2 (B) and treated with 3 µM 6-thio-dG for 72 hour (C). D. Not all micronuclei with ruptured nuclear envelope recruit cGAS. Representative images show presence or absence of cGAS, LaminA/C, both or none in cytosolic chromatin fragments (left). Bar graph shows the frequency of cytosolic chromatin fragments harboring either LaminA/C, cGAS, LaminA/C and cGAS or none in mock- and DOX+Shield1 treated HT1080+sgTelomere+dd-Cas9 cells (right). E. Lack of nuclear envelope (NE) around micronuclei correlates with STING activation in the cytosol. Representative images show presence of STING speckles in the cytosol of LaminA/C negative micronuclei harboring cells (top). Pie chart shows frequency of cytosolic chromatin fragments harboring either LaminA/C, STING, LaminA/C and STING or none in mock- and DOX+Shield1 treated HT1080+sgTelomere RNA+dd-Cas9 cells (bottom). F. MDC1 is recruited to cytosolic chromatin fragments and a fraction of it co-localizes with cGAS. Representative images show co-localization of MDC1 and cGAS in cytosolic chromatin fragments (left). Bar graph shows the frequency of cytosolic chromatin fragments harboring either MDC1, cGAS or both in cells expressing D/N TRF2 and treated with doxycycline for 72 hour (right). G. Cytosolic chromatin fragments harbor phosphorylated ATM (pATM, S1981) and a major proportion of them harbor both pATM and cGAS. Representative images show presence of pATM and cGAS in cytosolic chromatin fragments (left). Bar graph shows the frequency of cytosolic chromatin fragments harboring either pATM, cGAS or both in cells expressing D/N TRF2 and treated with doxycycline for 72 hour (right) H. MRE11 is recruited to cytosolic chromatin fragments but rarely co-localizes with cGAS. Representative images show presence of MRE11 and cGAS in cytosolic chromatin fragments (left). Bar graph shows the frequency of cytosolic chromatin fragments harboring either MRE11, cGAS or both n cells expressing D/N TRF2 and treated with doxycycline for 72 hour (right). I. NBS1 is recruited to cytosolic chromatin fragments but excluded from cytosolic chromatin fragments harboring cGAS. Representative images show presence of NBS1 and H2AX (left) and only cGAS but no NBS1 in cytosolic chromatin fragments (middle). Bar graph shows the frequency of cytosolic chromatin fragments harboring either NBS1, cGAS or both in HT1080 cells expressing D/N TRF2 and treated with doxycycline for 72 hour (right). J. cGAS is excluded from cytosolic fragments containing heterochromatin marker. Representative images show presence of heterochromatin marker (HP1) and cGAS in cytosolic chromatin fragments (left). Bar graph shows frequency of cytosolic

88 chromatin fragments harboring either HP1, cGAS or both in HT1080 cells expressing D/N TRF2 and treated with doxycycline for 72 hour (right). K. Chromatin compactness dictates access of cGAS to cytosolic chromatin fragments. Bar graph shows proportion of cytosolic chromatin fragments harboring either H2AX, cGAS or both in HT1080+D/N-TRF2 cells treated with and without Vorinostat (250 nM) and doxycycline for 72 hour. Error bars in all the graphs represent STDEV/ SEM from two-three independent experiments. *P < 0.05.

89

Figure 18: NBS1 counters cGAS recruitment to cytosolic chromatin fragments in response to dysfunctional telomeres.

A. Micronuclei number is increased in NBS1-deficient cells. Bar graph shows increased frequency of micronuclei formation in NBS1-deficient NBS cells relative to NBS cells complemented with wild type

90 (WT) NBS1 in response to dysfunctional telomeres. B. Elevated levels of cGAS recruitment to cytosolic chromatin fragments in NBS1-deficient cells. Bar graph shows the percentage of cytosolic chromatin fragments harboring either H2AX, cGAS or both in NBS1-deficient and -proficient cells in response to dysfunctional telomeres. C. Increased proportion of cGAS positive cytosolic chromatin fragments in NBS1- deficient cells is not due to defective nuclear envelope coating of micronuclei. Bar graph shows the percentage of cytosolic chromatin fragments harboring either LaminA/C, cGAS, LaminA/C and cGAS or none in NBS1-deficient and -proficient cells. D. NBS1 is excluded from heterochromatin containing cytosolic chromatin fragments. Bar graph shows the frequency of cytosolic chromatin fragments containing either heterochromatin marker (H3K9me3), NBS1 or both in HT1080 cells stably expressing D/N TRF2 treated with doxycycline for 72 hour (right). E-G. NBS1 suppresses cytosolic DNA sensing pathway mediated premature senescence. Western blots show increased activation of proteins involved in immune signaling (E), Graph shows increased expression of genes involved in immune pathway (F) and elevated levels of β-gal staining (G) in NBS cells relative to NBS+NBS1 cells in response to dysfunctional telomeres caused by 6-thio-dG (dG). H. Schematics of different NBS1 mutant constructs used in this study. FHA- forkhead-associated domain; BRCT-BRCA1/2 C-terminus domain; AIM-ATM-MRE11 interaction domain. I-J. NBS1 binding to cytosolic chromatin fragments is required for the blocking of cGAS to cytosolic chromatin fragments. Bar graph shows the frequency of micronuclei formation (I) and the percentage of cytosolic chromatin fragments containing either H2AX, cGAS or both in NBS cells reconstituted with different NBS1 mutants (J) in response to dysfunctional telomeres caused by 6-thio-dG.

K-M. Over expression (O/E) of full length (FL) and FHA-BRCA domain (NBS11-200aa) NBS1 in NBS1- proficient cells reduce the occupancy of cGAS onto cytosolic chromatin fragments. Western blot shows

O/E of FL (left) and NBS11-200aa (right) NBS1 in HT1080 cells stably expressing DN-TRF2 (K); Graph shows the percentages of micronuclei (L) and the levels of H2AX, cGAS or both recruitment to cytosolic chromatin fragments (M) in HT1080-DN-TRF2 cells O/E FL and NBS11-200aa NBS1 in response to dysfunctional telomeres. Error bars in all the graphs represent the STDEV/ SEM from two independent experiments. *P (range) <0.001-0.0001.

91

Figure 19: cGAS is responsible for the initiation of premature senescence phenotype in response to dysfunctional telomeres.

A. Depletion of cGAS does not alter micronuclei formation in response to dysfunctional telomeres. Bar graph shows frequency of micronuclei formation in cGAS-depleted cells relative to scrambled shRNA expressing (cGAS-proficient) cells treated with +/- 3 µM 6-thio-dG. B-C. Depletion of cGAS attenuates immune signaling in response to dysfunctional telomeres. Western blots show reduced activation of proteins involved in immune signaling (B) and graph shows reduced expression of genes involved in immune pathway in cGAS-depleted cells (C) relative to cGAS-proficient cells in response to dysfunctional telomeres. D. cGAS is critical for the suppression of initiation of premature senescence phenotype. Bar graph shows levels of β-gal positive cGAS-proficient and –depleted cells at 10 days after the induction of dysfunctional telomeres. E-G. Inhibition of DNA binding activity of cGAS using a small molecule inhibitor, RU.521, reduces immune signaling and premature senescence phenotype in response to dysfunctional telomeres. Bar graph shows frequency of micronuclei formation in RU.521 treated cells with and without the induction of dysfunctional telomeres (E). Bar graph shows decreased expression of innate

92 immune genes (F) and reduced level of β-gal staining (G) in HT1080 cells treated with cGAS inhibitor followed by 6-thio-dG treatment. Error bars represent the STDEV/SEM from two-three independent experiments. *P (range) < 0.05-0.0001.

Figure 20: Premature senescence is independent of cell type and telomere length.

A. Schematics showing sgTelomere RNA and dd-Cas9 expressing plasmid constructs used for inducing DSBs in the telomeric region (top two panels). Constitutively expressed sgTelomere RNA binds to telomeric DNA; addition of Doxycycline (DOX) induces expression of dd-Cas9, and addition of Shield1 stabilizes dd-Cas9. Subsequently, dd-Cas9 binds to sgTelomere RNA and induces DSBs in the telomeric DNA. TRE-tetracycline responsive element. B. Telomeric DSBs do not cause telomere length shortening. Representative gel images show distribution of differently sized telomeres in HT1080 cells co-expressing dd-Cas9 and sgTelomere RNA with and without DOX and Shield1 treatment (top). Bar graph shows mean length of telomeres in mock- and DOX and Shield1 treated HT1080+sgTelomere+dd-Cas9 cells at indicated time points (bottom). C-E. Telomeric DSBs trigger innate immune signaling in U2OS cells. Bar

93 graph shows frequency of micronuclei formation at 48 hours following the induction of telomeric DSBs (C), western blot shows increased activation of proteins involved in innate immune signaling (D) and the bar graph shows increased expression of immune pathway genes (E) in U2OS cells stably co-expressing dd-Cas9 and sgTelomere RNA. Error bars represent the STDEV/ SEM from two independent experiments. *P (range) < 0.05-0.01.

Figure 21: G2 to Mitotic phase progression in response to dysfunctional telomeres is critical for micronuclei formation and premature senescence.

A. Representative images show induction of telomere DSBs, as assessed by H2AX foci, in DMSO, DOX+Shield1+DMSO and DOX+Shield1+CDK1 inhibitor (RO-3306) treated HT1080 cells co-expressing

94 sgTelomere RNA and dd-Cas9. B. Micronuclei formation is significantly reduced when cells are prevented from entering mitosis after telomeric DNA damage. HT1080 cells co-expressing dd-Cas9 and sgTelomere RNA were treated with CDK1 inhibitor (1 µM, RO-3306) before inducing telomeric DSBs. Bar graph shows frequency of micronuclei formation in HT1080+sgTelomere RNA+dd-Cas9 cells with telomeric DSBs at different treatment conditions and times. C-E. CDK1 inhibition attenuates immune signaling and senescence after telomeric DNA damage. Western blot shows attenuated activation of proteins involved in innate immune signaling (C) which correlated well with attenuated expression of innate immune genes (D) and β-gal staining (E) in RO-3306 +/- DOX+Shield1 treated HT1080+sgTelomere RNA+dd-Cas9 cells relative to DMSO +/- DOX+Shield1 treated cells. Error bars represent the STDEV/SEM from two independent experiments. *P (range) < 0.05-0.001; ns-not significant.

) 6 0

%

(

i

e

l c

u 4 0

n

o

r

c

i

m

h

t 2 0

i

w

s

l

l e

C 0 D o x D O X + V o rin o s ta t

D /N T R F 2

Figure 22: Condensed chromatin structure restricts cGAS binding to cytosolic chromatin fragments: A. Western blot shows expression of D/N TRF2 in mock- and doxycycline treated HT1080 cells stably expressing doxycycline inducible D/N TRF2 plasmid. Cells were harvested 72 hour after the DOX treatment and total cell lysate was subjected to western blotting and subsequently probed with anti-FLAG and anti-TRF2 antibodies. B. Histone deacetylase (HDAC) inhibitor, Vorinostat, does not impact micronuclei formation. Bar graph shows frequency of micronuclei formation in D/N TRF2 expressing HT1080 cells in presence or absence of 250 nM Vorinostat, and treated with doxycycline for 72 hour. Error bars represent the STDEV from two independent experiments.

95

Figure 23: ATM plays a role in the binding of cGAS to cytosolic chromatin fragments. A. NBS1 co-localizes with ATM in a major fraction of the cytosolic chromatin fragments. Representative images show co-localization of NBS1 and phosphorylated ATM (S1981; left panel) and the bar graph shows the frequency of cytosolic chromatin fragments harboring either NBS1, pATM or both in HT1080 cells expressing D/N TRF2. B-D. ATM plays a role in cGAS recruitment to cytosolic chromatin fragments. Bar graph shows no effect on the frequency of micronuclei formation in shRNA mediated ATM depleted cells relative to scr shRNA expressing HeLa cells (B). Bar graph shows elevated levels of cGAS positive

96 cytosolic chromatin fragments (C) and western blots show increased activation of IRF3 (D) in ATM- depleted HeLa cells relative to scr shRNA expressing HeLa cells. Error bars represent the STDEV/SEM from two independent experiments. *P (range) < 0.05

Figure 24: MRE11 is dispensable for cGAS recruitment to cytosolic chromatin fragments. A-E. Deficiency in MRE11 and blocking of its exonuclease activity increases micronuclei number but not the innate immune signaling in response to dysfunctional telomeres. Western blot shows MRE11 expression in HT1080 cells stably expressing shMRE11 RNA and ALTD cells (A). Bar graph shows frequency of micronuclei formation in MRE11-depleted (HT1080+shMRE11) and MRE11-deficient (ATLD) and MRE11–exonuclease inhibited (HT1080+mirin) cells in response to dysfunctional telomeres (B). Bar graph shows the percentage of cytosolic chromatin fragments containing cGAS in MRE11-depleted HT1080 cells (C) MRE11-deficient ATLD cells (D), and Mirin-treated HT1080 cells (E) upon induction of dysfunctional telomeres. F-H. Innate immune signaling after telomere dysfunction is marginally reduced in MRE11 defective cells. Bar graph shows expression of immune pathway genes in MRE11-proficient HT1080 cells (F), MRE11-depleted HT1080 cells (G), and MRE11-deficient ATLD cells and mirin treated HT1080 cells (H) in response to dysfunctional telomeres. Error bars represent the STDEV/SEM from two-three independent experiments. *P (range) < 0.05-0.0001.

97

Figure 25: Inhibition of CGAS binding to DNA attenuates inflammation:

A. Western blot shows small hairpin (sh) cGAS RNA mediated stable depletion of cGAS in HT1080 cells. Clone numbers 1 and 5 showing maximum cGAS depletion was used for further experiments. UT- untansfected; THP1 cell lysate was used as a positive control for cGAS expression. B. Inhibition of cGAS using small molecule inhibitor, RU.521, attenuates immune signaling in response to dysfunctional telomeres. Western blots show reduced activation of proteins involved in immune signaling in in HT1080 cells treated with cGAS inhibitor followed by 6-thio-dG treatment for 72 hour.

98

Figure 26: Induction of telomeric DSBs do not cause apoptosis. A. Western blot shows expression of total and cleaved caspase 3 (a well-known marker for apoptosis) at different time intervals after Doxycycline and Shield1 treatment in HT1080 cells co-expressing dd-Cas9 and sgTelomere RNA. HT1080 cells treated with 2 µM camptothecin (CPT) continuously for 24 hours were used as a positive control for apoptosis. B. 6-thio-dG treatment has a marginal effect on cell survival. Cell survival was analyzed by a colony formation assay. Bar graph shows proportion of surviving HT1080 cells 7 days after 3 µM 6-thio-dG treatment relative to DMSO treated cells. The relative survival efficiencies were plotted. The error bars represent the SD calculated from triplicate wells.

99

Figure 27. Graphical model of how telomere dysfunction induce premature senescence

The model depicts both cGAS-dependent cytosolic DNA sensing pathway-mediated activation of immune signaling and subsequent initiation of premature senescence phenotype following telomere dysfunction; and NBS1 countering of cGAS binding to cytosolic chromatin fragments to prevent unwarranted initiation of premature senescence phenotype.

100 Sl. No. Primary Antibodies Catalog # Vendor Application (dilution)

1 Mouse monoclonal anti-TRF2 100-56506 Novus Bio WB (1:500)

2 Rabbit polyclonal anti-STING 24683 Novus Bio WB (1:500)

3 Rabbit polyclonal anti-TRF2 NA gift from Karlsensder lab IF (1:100)

4 Mouse monoclonal anti-FLAG F3165 Sigma WB (1:2000); IF (1:500)

5 Mouse monoclonal anti-γ-Tubulin (GTu88) T6557 Sigma WB (1:50000)

6 Rabbit polyclonal anti-phospho STING (Ser366) 85735 Cell Signaling WB (1:700)

7 Rabbit monoclonal anti-phospho-IRF3 (Ser396;4D4G) 4947 Cell Signaling WB (1:500)

8 Rabbit monoclonal anti-phospho-TBK1 (Ser172;D52C2) 5483 Cell Signaling WB (1:500)

9 Rabbit polyclonal anti-phospho-CHK1 (Ser317) 2344 Cell Signaling WB (1:1000)

10 Rabbit monoclonal anti-phospho-Histone H2AX (Ser139;20E3) 9718 Cell Signaling IF (1:1000)

11 Rabbit polyclonal anti-phospho-CHK2 (Thr68) 2661 Cell Signaling WB (1:1000)

12 Rabbit monoclonal anti-cGAS (D1D3G) 15102 Cell Signaling WB (1:200); IF (1:100)

13 Rabbit monoclonal anti-caspase-3 (8G10) 9665 Cell Signaling WB (1:500)

14 Rabbit polyclonal anti-Histone H3 (acetyl K9) 4441 Abcam IF (1:4000)

15 Mouse monoclonal anti-MDC1 (MDC1-50) 50003 Abcam IF (1:400)

16 Rabbit polyclonal anti-phospho-ATM (Ser1981; EP1890Y) 81292 Abcam WB (1:10000); IF (1:500)

17 Mouse monoclonal anti-phospho Stat1 p84/p91 (PSM1) 51700 Santacruz WB (1:200)

18 Mouse monoclonal anti-Lamin A/C (636) 7292 Santacruz IF (1:300)

19 Rabbit polyclonal 53BP1 (H-300) 22760 Santacruz IF (1:500)

20 Mouse monoclonal anti-NBS1 (B5) 515069 Santacruz WB (1:500); IF (1:100)

21 Mouse monoclonal anti-MRE11 (12D7) 70212 Gentex WB (1:1000); IF (1:1000)

22 Mouse monoclonal anti-ATM (2C1) 70103 Gentex WB (1:5000)

23 Mouse monoclonal anti-phospho-Histone H2AX (Ser139; JBW301) 05-636 Millipore IF (1:1000)

24 Mouse monoclonal anti-Ku70 NA homemade WB (1:1000)

25 Mouse monoclonal anti-Ku80 NA homemade WB (1:1000)

26 Mouse monoclonal anti-HP1α 130446 Santacruz IF (1:100) 27 Rabbit polyclonal anti-phospho-Histone H3 (Ser10) 06-570 Millipore Flow cytometry (1:1000)

Table 4: List of antibodies used in chapter 3

101 Serial # Primer Name Sequence (5'-3') 1 TRF2-Nhe1-F GCTAGCCCACCATGGCGGGAGG GGCGG

GTCGACTTAAGATCCCTTGTCGTCATCGTCCTTGTAGTCCTTGTCGTCATCGTCCTTG 2 TRF2-Flag-SalI-R TAGTCGTTCATGCCAAGTCTTTTCATGG

3 TRF2-45AA_Nhe1-F GCTAGCCCACCATGGCACGGCTGGAAGAGG

GTCGACTTAAGATCCCTTGTCGTCATCGTCCTTGTAGTCCTTGTCGTCATCGTCCTTG 4 TRF2-453AA_Flag-SalI-R TAGTCGTTTTCTACAGTCCACTTCTGCTTTTTTG

5 cGAS-Xho1_R GTCGACTCAAAATTCATCAAAAACTGG

6 DD-Nhe1-F GCTAGCATGGGAGTGCAGG

7 DD-Nhe1--XhoI-R GCTAGCCTCGAGTTCGGTTTTAGAAGCTCC

8 human IFNα Fwd (5'-3') AACTCCCCTGATGAATGCGG

9 human IFNα Rev (5'-3') TAGCAGGGGTGAGAGTCTTTG

10 human IFNβ Fwd (5'-3') CAACTTGCTTGGATTCCTACAAAG

11 human IFNβ Rev (5'-3') TATTCAAGCCTCCCATTCAATTG

12 human TLR9 Fwd (5'-3') CGCCCTGCACCCGCTGTCTCT

13 human TLR9 Rev (5'-3') CGGGGTGCTGCCATGGAGAAG

14 human CCL5 Fwd (5'-3') CCAGCAGTCGTCTTTGTCAC

15 human CCL5 Rev (5'-3') CTCTGGGTTGGCACACACTT

16 human CXCL10 Fwd (5'-3') AGGAACCTCCAGTCTCAGCA

17 human CXCL10 Rev (5'-3') CAAAATTGGCTTGCAGGAAT

18 human IFIT1 Fwd (5'-3') CCTCCTTGGGTTCGTCTACA

19 human IFIT1 Rev (5'-3') GGCTGATATCTGGGTGCCTA

20 human IFIT3 Fwd (5'-3') GAAGGAACTGGGCCGCCTGCTAAG

21 human IFIT3 Rev (5'-3') GCCCTGGCCCATTTCCTCACTACC

22 human ISG56 Fwd (5'-3') TTGATGACGATGAAATGCCTGA

23 human ISG56 Rev (5'-3') CAGGTCACCAGACTCCTCAC

24 human ISG54 Fwd (5'-3') AAGCACCTCAAAGGGCAAAAC

25 human ISG54 Rev (5'-3') TCGGCCCATGTGATAGTAGAC

26 human IL6 Fwd (5'-3') CCTTCGGTCCAGTTGCCTTCT

27 human IL6 Rev (5'-3') GCATTTGTGGTTGGGTCA

28 human CSF2 Fwd (5'-3') AGCATGTGAATGCCATCCAG

29 human CSF2 Rev (5'-3') AGGGGATGACAAGCAGAAAG

30 human β-Actin Fwd (5'-3') TCGTCGACAACGGCTCCGGCATGT 31 human β-Actin Rev (5'-3') CCAGCCAGGTCCAGACGCAGGAT

Plasmid Name Company Plasmid Number pSLQ1651-sgTelomere(F+E) Addgene 51024 pCW-Cas9 Addgene 50661 pLV.MRE11i Addgene 15565 pTRIP-CMV-tagRFP-FLAG-cGAS Addgene 86676 pTRIP-CMV-GFP-FLAG-cGAS Addgene 86675

Table 5: List of plasmids and RT-PCR primers used in chapter 3

102

Chapter 4 Conclusion and Future Perspectives

103 4.1 Conclusion

For 25 years, research on DDR machinery focused on understanding its functions and regulations in DNA damage sensing, repair and signaling. All of these activities are essential for cellular homeostasis as defects in DDR signaling leads to a multitude of pathologies, including developmental disorders and cancer susceptibility. In the last few years, research has firmly established that DDR signaling play important roles in regulating cells immune response as well. Deficiencies in DNA damage sensor, effector and repair proteins or insufficient checkpoint activation results in inflammatory response, which in one hand is beneficial and confers protection from pathogens and surveillance against tumors, but can also result in premature aging, autoimmune disorders or tumor metastasis and survival. My research in graduate school adds to growing body of literature suggesting DNA repair factors and proper DDR signaling are intimately connected with immune signaling network. As described in chapter 2, we have found that RAD51, a protein with established roles in DNA repair and replication, is involved in suppression of innate immune signaling. RAD51 deficient cells, when exposed to genotoxic stress like IR, accumulates fragmented ss and ds DNA in the cytoplasm, leading to activation of STING and upregulation of downstream innate immune genes, including inflammatory cytokines and chemokines. RAD51 is recruited to the sites of DSBs, facilitating their repair in S/G2 phase cells. In addition, RAD51 also localizes to perturbed replication forks and prevents MRE11-mediated excessive processing of newly replicated genomes. Mechanistically, we found that replication fork degradation combined with defects in DSB repair in cells lacking RAD51 leads to the accumulation of self- DNA in the cytosol, resulting in the initiation of a STING-mediated innate immune response. Thus, the coordinated activities of RAD51 in DSB repair and replication fork maintenance are connected with its role in regulation of innate immune response signaling and this may provide new therapeutic opportunities for treating RAD51 proficient tumors by inhibiting RAD51 and boosting the host’s antitumor immunity. DDR signaling is not only limited to detection and repair of DNA lesions. Activation of checkpoints in response to DNA damage constitutes an important arm of proper DDR signaling. In chapter 3, we have used telomere dysfunction as a model to study how defective DDR signaling results in premature cellular senescence. We observed aberrant DDR signaling events in response to dysfunctional telomeres; normally, genotoxic stress rapidly activates G2/M checkpoint arrest, which prevents cells with damaged DNA from entering mitosis. However, cells with damaged telomeric DNA are released from G2 arrest before repair is complete, which results in chromosome fusions and micronuclei formation. Micronuclei, because they are in the cytoplasm, are recognized by cGAS, leading to activation of immune signaling and premature senescence. Thus, dysfunctional telomeres together with imperfect checkpoint activation, contributes to cytosolic DNA mediated immune activation, that is sufficient to induce cellular senescence. Surprisingly,

104 NBS1, another well-known DSB sensing and repair factor, counters cGAS binding to the cytosolic chromatin fragments and prevent unwarranted immune activation and premature senescence.

Taken together, results from my work indicate that DDR factors’ roles in modulating immune signaling are not direct but are rather a consequence of their inherent ability to sense, repair and signal in response to DNA damage. Thus, in addition to their known DNA repair and replication functions, DDR factors help maintain homeostasis and prevent unwarranted activation of immune signaling. A more precise understanding of the mechanisms by which different DDR factors function in immune signaling can be exploited to redirect the immune system for both preventing and treating autoimmunity, cellular senescence and cancer in humans.

4.2 Future direction:

In chapter 2, I have presented evidence that cells lacking RAD51, when exposed to high LET radiation, accumulate DNA fragments in cytoplasm leading to increased expression of innate immune genes. Continuation of that work revealed cytosolic DNA mediated immune activation is not specific to IR induced DNA damage, as exposing RAD51 deficient cells to other DNA damaging agents like HU or Gemcitabine also induced an immune response (Fig 28A, B). I also saw increased accumulation of ds DNA in the cytoplasm of RAD51 deficient cells after gemcitabine treatment (Fig 28C). Now, we want to know if there is a translational significance to these findings. RAD51, is over-expressed in many human cancers, including breast, bladder, prostate, pancreas, soft tissue sarcoma, and lung 253,254. Over-expression of RAD51 causes chromosomal amplifications, deletions, and translocations, resulting in a loss of heterozygosity and aneuploidy, which impacts tumor progression and metastasis 176,255. Clinically, high RAD51 expression in patient tumors is associated with aggressive pathologic features 256,257 and unfavorable treatment outcomes 258. Small molecule RAD51 inhibitors have been shown to enhance tumor cell killing when combined with other treatment modalities in pre-clinical models, but their clinical applications remain limited. It will be interesting to see, whether depletion or pharmacological inhibition of RAD51 can be combined with DNA damaging agents like IR or gemcitabine to hone the host’s immune system for selective killing of tumor cells. To answer this, we have used a doxycycline inducible RAD51 shRNA to deplete RAD51 in mouse MC38 colon adenocarcinoma cells (Fig 28D) and we are going to implant RAD51 depleted MC138 cells in syngeneic mice. After the tumors become palpable in size (6-8 mm diameter), we will expose these mice to radiation or treat with gemcitabine to determine the effect of RAD51 depletion in inducing antitumor immunity and reducing tumor growth. Recent reports also suggest cytosolic DNA media immune response is important for prolonged survival and metastasis of tumors and

105 in that regard, it is important to have an open mind while evaluating the effect of RAD51 inhibition on treatment outcome. In the light of these findings with RAD51, another protein, which sparked my interest, is BRCA1. Mutation in BRCA1 gene increases the risk of breast, ovarian, fallopian tube and peritoneal cancer 259. BRCA1-mutant breast cancers are also highly aggressive, at least one-third of them are triple negative breast cancers (i.e., ER, PR and HER-2 negative) and associated with a poor prognosis. BRCA1 is a multifunctional protein that engages in numerous direct and indirect physical interactions with specific partner proteins.59 BRCA1 is critical for genome stability maintenance, largely due to its role in HR- mediated DSBs. In the absence of BRCA1, cells utilize non-homologous end-joining, resulting in increased chromosomal instability and a higher degree of genomic alterations.3 Additionally, BRCA1 contributes to replication fork stability, cell cycle checkpoint activation, transcription regulation, heterochromatin maintenance, mitotic spindle formation, RNA splicing control, and estrogen metabolism 60,260-265. Interestingly, BRCA1 helps recruit RAD51 to sites of DNA damage,24 and it is possible that BRCA1 deficiency results in activation of immune signaling. Preliminary data from our lab indicates that BRCA mutant HCC1937 cells accumulate ss and ds-DNA in the cytoplasm (Fig 28E), which results in increased expression of innate immune genes (Fig 28F). Furthermore, as BRCA1 is also involved in cell cycle checkpoint activation, we also observed increased micronuclei formation in BRCA1 deficient cells, even in the absence of any genotoxic stress, and recruitment of cGAS to these micronuclei. As mentioned earlier, chronic activation of cytosolic DNA- mediated immune response is pro-tumorigenic and contributes to survival and metastasis of tumors. It remains to be seen whether persistent immune signaling as result of accumulation of cytosolic DNA and chromatin fragments signaling, contributes to aggressiveness of BRCA1 deficient tumors. Our data shows dysfunctional telomeres result in formation of micronuclei or chromatin fragments. Micronuclei are present in the cytoplasm, and are recognized as cytosolic chromatin fragments by cGAS- STING, the cytosolic DNA recognition machinery. This initiates a cytosolic DNA mediated immune response and upregulation of type I IFN signaling, resulting in premature senescence. We are currently exploring the effect of telomere dysfunction and micronuclei formation in vivo, especially whether induction of telomere dysfunction result in antitumor immune response and tumor shrinkage. To test this, we have generated MC38 colon adenocarcinoma cells, which express DN TRF2 upon doxycycline treatment (Fig 29). We are going to implant DN TRF2 expressing MC38 cells in syngeneic C57BL6 mice and once the tumors become palpable, we will start doxycycline treatment to induce telomere dysfunction. It will be interesting to see if DN TRF2 expression results in increased inflammation and senescence in vivo as well (Fig 31). However, we are also open to the idea that we may see the opposite effect. As discussed in chapter 1, to reinforce senescence signaling 146,147, senescent cells acquire a senescence-associated

106 secretory phenotype (SASP), that promote tumor progression by inducing changes in the tissue microenvironment resulting in proliferation, transformation and metastasis of epithelial cells 149-152. In addition, prolonged cytosolic DNA mediated immune response contributes to survival and metastasis of tumors and it is possible that DN TRF2 expressing tumors will be more aggressive in nature. Interestingly, regardless of the method I used to cause telomere dysfunction (Cas9-sg telomere, DN TRF2 expression, or 6-Thio-DG treatment), only a subset of cells (~20-25%) undergo senescence. I am interested in knowing what happens to the cells that do not undergo senescence and continues to proliferate. Do they adapt to live with chromosome fusions and if so, what is the long-term consequence of that? Another possibility we have to consider is that some tumor cells may initially undergo senescence but can also escape later. This is of particular interest, as recent work by Milanovic and colleagues have identified, that genotoxic stress induced senescence is not irreversible as previously thought and cancer cells escaping senescence often have a gene expression signature that confers “stemness” to these escaping cells resulting in higher tumor initiating potential 266. This phenomenon is cell autonomous and different from SASP. I am currently testing whether cells, which do not undergo senescence in response to damaged or dysfunctional telomeres, have increased tumorigenic potential. After inducing initial telomere dysfunction, I have allowed the cells to grow for 45 days post treatment to form colonies. I am screening these individual clones for chromosomal aberration/ aneuploidy, persistent DDR signaling (pATM, pATR, and pCHK1), persistent immune activation (pIRF3, pTBK1, pSTAT1, pNF-κB), presence of EMT markers (changes in E-cadherin and vimentin expression), WNT signaling markers (β-catenin) and finally their ability for anchorage independent growth. Based on these results, I will select individual clones for RNA sequencing analysis to get information on their gene expression profile (Fig 31). In chapter 3, I have shown deficiencies in DNA repair pathways coupled with inadequate activation of G2-M checkpoint is the major mechanism underlying the formation of micronuclei. However, the question I want to address next is what happens to micronuclei after formation. Based on literature, there are five major possibilities for the fate of micronuclei: degradation, reincorporation in the main nucleus, extrusion, persistence and premature chromothripsis. We want to know if the micronuclei formation is completely random. The alternative hypothesis is there are some sites of chromosome that are more likely to break and be entrapped inside micronuclei. I am particularly interested in understanding the reincorporation pathway, as this would mean that micronuclei are not only an marker of genomic instability, but could possibly a direct cause, when reincorporated into the main nucleus, either during interphase or in next mitotic phase 267-271. In addition, are micronuclei reintegrated at random locations or there are certain homing regions in the genome they tend to prefer for reintegration? To answer these questions, we have recently developed the m6ATracer-DAM-ID technique 272,273, where we have used cGAS and nuclease dead mutant TREX1 to trace micronuclei (Fig 30A, B). Previously, the m6ATracing technique has been used to

107 study the interaction of genome with nuclear lamina. This technique is based on the adenine methylation (m6A) mark left on DNA by a DNA-adenine-methyltransferase (DAM) from Escherichia coli. When cGAS/ TREX1 fused with DAM are recruited to the micronuclei in the cytoplasm, DAM -cGAS/ TREX1 will methylate the adenine residue (m6A) surrounding DNA in the micronuclei. As the m6A-DNA is a stable mark, (no eukaryotic cellular machinery available can remove it), the signal is retained even after cGAS and TREX1 dissociates from the processed DNA. We have optimized a method by which we can separate micronuclei from macronuclei containing cellular fraction. by using a methylation- specific PCR amplification of DNA 274, derived from micronuclei fraction, followed by next generation sequencing, we will get information whether specific sites of chromosomes are more likely to end up inside micronuclei. Similarly, this technique can also be used to PCR amplify DNA from macronuclei fraction to get information about the possible loci in the main nuclei where micronuclei are reintegrated (Fig 31). I have summarized the work plan of my current ongoing and future experiments in form of a simple schematic diagram (Fig 31).

108 Figures:

Figure 28: RAD51 and BRCA1 deficiency results in activation of immune signaling in response to genotoxic stress. Graph shows qRT-PCR analysis of amplification of innate immune pathway genes in RAD51-depleted cells but not in RAD51-proficient cells after gemcitabine (5 µM, 24 h; GEM), HU (4 mM, 24 h) and IR (1 Gy, 8 h) treatment (A). Graph shows quantification of cytokines in cell culture medium of mock- and GEM- treated (5 µM, 24 h), RAD51-proficient and –depleted HT1080 cells (B). Gel picture shows 10 ng of cytosolic DNA purified from mock- and GEM-treated (24 h) RAD51-proficient and -depleted HT1080 cells (C). Western blot shows depletion of RAD51 after 72h of doxycycline (1µg/ml) in MC38 cells stably expressing RAD51 sg RNA (D) Graph shows accumulation of ss and ds- DNA in cytoplasmic fraction (E) and mRNA expression of innate immune genes (F) of HCC1937 and BRCA1 complemented HCC1937 cells at different time intervals after exposure to 1Gy radiation.

Figure 29: DN TRF2 expression in MC38 cells.

109 MC38 cells were infected with lentivral particles expressing DN TRF2 plasmid and selected in puromycin for 7 days. Western blot shows induction of DN TRF2 after 72h of doxycycline (1µg/ml) exposure in different MC38 clones stably expressing DN TRF2 plasmid.

Figure 30: Overview of m6ATracer-DAM-ID technique. Schematics showing DAM alone and DAM cGAS/ TREX1 expressing plasmid constructs used for the m6ATracer-DAM-ID technique (A). Western blot shows expression of FLAG-cGAS DAM after 24h of doxycycline (500 ng/ml) and shield (1µM) treatment (B). Gel image shows PCR amplified products from cells expressing DAM alone or DAM-cGAS expressing plasmids (C).

110

Figure 31: Simplified schematics of ongoing/ future experimental plans. Schematics of long-term culture of HT1080 cells after telomere dysfunction (A). Flow chart shows experimental plan for in vivo experiments with RAD51 depleted and DN TRF2 expressing MC38 cells (B). Schematics of m6ATracer-DAM-ID approach to explore the origin and fate of micronuclei (C).

111 Bibliography

1 Swenberg, J. A. et al. Endogenous versus exogenous DNA adducts: their role in carcinogenesis, epidemiology, and risk assessment. Toxicol Sci 120 Suppl 1, S130-145, doi:10.1093/toxsci/kfq371 (2011). 2 Asaithamby, A. & Chen, D. J. Mechanism of cluster DNA damage repair in response to high-atomic number and energy particles radiation. Mutation research 711, 87-99, doi:10.1016/j.mrfmmm.2010.11.002 (2011). 3 Asaithamby, A., Hu, B. & Chen, D. J. Unrepaired clustered DNA lesions induce chromosome breakage in human cells. Proceedings of the National Academy of Sciences of the United States of America 108, 8293-8298, doi:10.1073/pnas.1016045108 (2011). 4 Asaithamby, A. et al. Repair of HZE-particle-induced DNA double-strand breaks in normal human fibroblasts. Radiation research 169, 437-446, doi:10.1667/RR1165.1 (2008). 5 Bray, F., Ren, J. S., Masuyer, E. & Ferlay, J. Global estimates of cancer prevalence for 27 sites in the adult population in 2008. International journal of cancer 132, 1133-1145, doi:10.1002/ijc.27711 (2013). 6 Jackson, S. P. & Bartek, J. The DNA-damage response in human biology and disease. Nature 461, 1071-1078, doi:10.1038/nature08467 (2009). 7 Su, F. et al. Nonenzymatic role for WRN in preserving nascent DNA strands after replication stress. Cell Rep 9, 1387-1401, doi:10.1016/j.celrep.2014.10.025 (2014). 8 Mathew, C. G. Fanconi anaemia genes and susceptibility to cancer. Oncogene 25, 5875- 5884, doi:10.1038/sj.onc.1209878 (2006). 9 Roy, R., Chun, J. & Powell, S. N. BRCA1 and BRCA2: different roles in a common pathway of genome protection. Nature reviews. Cancer 12, 68-78, doi:10.1038/nrc3181 (2011). 10 Shiloh, Y. ATM and related protein kinases: safeguarding genome integrity. Nature reviews. Cancer 3, 155-168, doi:10.1038/nrc1011 (2003). 11 Su, F. et al. Replication stress induced site-specific phosphorylation targets WRN to the ubiquitin-proteasome pathway. Oncotarget 7, 46-65, doi:10.18632/oncotarget.6659 (2016). 12 Begg, A. C., Stewart, F. A. & Vens, C. Strategies to improve radiotherapy with targeted drugs. Nature reviews. Cancer 11, 239-253, doi:10.1038/nrc3007 (2011). 13 Jeggo, P. A., Pearl, L. H. & Carr, A. M. DNA repair, genome stability and cancer: a historical perspective. Nature reviews. Cancer 16, 35-42, doi:10.1038/nrc.2015.4 (2016). 14 Nikjoo, H., Uehara, S., Wilson, W. E., Hoshi, M. & Goodhead, D. T. Track structure in radiation biology: theory and applications. International journal of radiation biology 73, 355-364 (1998). 15 Cadet, J. et al. Hydroxyl radicals and DNA base damage. Mutat Res 424, 9-21 (1999). 16 Lomax, M. E., Folkes, L. K. & O'Neill, P. Biological consequences of radiation-induced DNA damage: relevance to radiotherapy. Clinical oncology 25, 578-585, doi:10.1016/j.clon.2013.06.007 (2013). 17 Brzostek-Racine, S., Gordon, C., Van Scoy, S. & Reich, N. C. The DNA damage response induces IFN. J Immunol 187, 5336-5345, doi:10.4049/jimmunol.1100040 (2011).

112 18 Aglipay, J. A. et al. A member of the Pyrin family, IFI16, is a novel BRCA1-associated protein involved in the p53-mediated apoptosis pathway. Oncogene 22, 8931-8938, doi:10.1038/sj.onc.1207057 (2003). 19 Hopkins, B. B. & Paull, T. T. The P. furiosus Mre11/Rad50 complex promotes 5’ strand resection at a DNA double-strand break. Cell 135, 250-260, doi:10.1016/j.cell.2008.09.054 (2008). 20 Williams, R. S. et al. Mre11 Dimers Coordinate DNA End Bridging and Nuclease Processing in Double-Strand-Break Repair. Cell 135, 97-109, doi:10.1016/j.cell.2008.08.017 (2008). 21 de Jager, M. et al. DNA-binding and strand-annealing activities of human Mre11: implications for its roles in DNA double-strand break repair pathways. Nucleic Acids Research 29, 1317-1325 (2001). 22 Usui, T. et al. Complex Formation and Functional Versatility of Mre11 of Budding Yeast in Recombination. Cell 95, 705-716, doi:http://doi.org/10.1016/S0092-8674(00)81640-2 (1998). 23 Kondo, T. et al. DNA damage sensor MRE11 recognizes cytosolic double-stranded DNA and induces type I interferon by regulating STING trafficking. Proc Natl Acad Sci U S A 110, 2969-2974, doi:10.1073/pnas.1222694110 (2013). 24 Bhattacharya, S. et al. RAD51 interconnects between DNA replication, DNA repair and immunity. Nucleic Acids Res 45, 4590-4605, doi:10.1093/nar/gkx126 (2017). 25 Roth, S. et al. Rad50-CARD9 interactions link cytosolic DNA sensing to IL-1beta production. Nat Immunol 15, 538-545, doi:10.1038/ni.2888 (2014). 26 Tauchi, H., Matsuura, S., Kobayashi, J., Sakamoto, S. & Komatsu, K. Nijmegen breakage syndrome gene, NBS1, and molecular links to factors for genome stability. Oncogene 21, 8967-8980, doi:10.1038/sj.onc.1206136 (2002). 27 Pereira-Lopes, S. et al. NBS1 is required for macrophage homeostasis and functional activity in mice. Blood 126, 2502-2510, doi:10.1182/blood-2015-04-637371 (2015). 28 Prochazkova, J. et al. DNA Repair Cofactors ATMIN and NBS1 Are Required to Suppress T Cell Activation. PLoS genetics 11, e1005645, doi:10.1371/journal.pgen.1005645 (2015). 29 Lamarche, B. J., Orazio, N. I. & Weitzman, M. D. The MRN complex in double-strand break repair and telomere maintenance. FEBS Lett 584, 3682-3695, doi:10.1016/j.febslet.2010.07.029 (2010). 30 Marechal, A. & Zou, L. DNA damage sensing by the ATM and ATR kinases. Cold Spring Harb Perspect Biol 5, doi:10.1101/cshperspect.a012716 (2013). 31 Cimprich, K. A. & Cortez, D. ATR: an essential regulator of genome integrity. Nat Rev Mol Cell Biol 9, 616-627, doi:10.1038/nrm2450 (2008). 32 Bredemeyer, A. L. et al. ATM stabilizes DNA double-strand-break complexes during V(D)J recombination. Nature 442, 466-470, doi:10.1038/nature04866 (2006). 33 McKinnon, P. J. ATM and the molecular pathogenesis of ataxia telangiectasia. Annual review of pathology 7, 303-321, doi:10.1146/annurev-pathol-011811-132509 (2012). 34 Erttmann, S. F. et al. Loss of the DNA Damage Repair Kinase ATM Impairs Inflammasome-Dependent Anti-Bacterial Innate Immunity. Immunity 45, 106-118, doi:10.1016/j.immuni.2016.06.018 (2016).

113 35 Hartlova, A. et al. DNA damage primes the type I interferon system via the cytosolic DNA sensor STING to promote anti-microbial innate immunity. Immunity 42, 332-343, doi:10.1016/j.immuni.2015.01.012 (2015). 36 Postigo, A., Ramsden, A. E., Howell, M. & Way, M. Cytoplasmic ATR Activation Promotes Vaccinia Virus Genome Replication. Cell Rep 19, 1022-1032, doi:10.1016/j.celrep.2017.04.025 (2017). 37 Edwards, T. G., Bloom, D. C. & Fisher, C. The ATM and Rad3-Related (ATR) Protein Kinase Pathway Is Activated by Herpes Simplex Virus 1 and Required for Efficient Viral Replication. J Virol 92, doi:10.1128/JVI.01884-17 (2018). 38 Shen, Y. J. et al. Genome-derived cytosolic DNA mediates type I interferon-dependent rejection of B cell lymphoma cells. Cell Rep 11, 460-473, doi:10.1016/j.celrep.2015.03.041 (2015). 39 Helleday, T., Petermann, E., Lundin, C., Hodgson, B. & Sharma, R. A. DNA repair pathways as targets for cancer therapy. Nature reviews. Cancer 8, 193-204, doi:10.1038/nrc2342 (2008). 40 Davis, A. J. & Chen, D. J. DNA double strand break repair via non-homologous end- joining. Translational cancer research 2, 130-143, doi:10.3978/j.issn.2218- 676X.2013.04.02 (2013). 41 Hartley, K. O. et al. DNA-dependent protein kinase catalytic subunit: a relative of phosphatidylinositol 3-kinase and the ataxia telangiectasia gene product. Cell 82, 849- 856 (1995). 42 Uematsu, N. et al. Autophosphorylation of DNA-PKCS regulates its dynamics at DNA double-strand breaks. The Journal of cell biology 177, 219-229, doi:10.1083/jcb.200608077 (2007). 43 van der Burg, M. et al. A DNA-PKcs mutation in a radiosensitive T-B- SCID patient inhibits Artemis activation and nonhomologous end-joining. The Journal of clinical investigation 119, 91-98, doi:10.1172/JCI37141 (2009). 44 Ferguson, B. J., Mansur, D. S., Peters, N. E., Ren, H. & Smith, G. L. DNA-PK is a DNA sensor for IRF-3-dependent innate immunity. Elife 1, e00047, doi:10.7554/eLife.00047 (2012). 45 Zhang, X. et al. Cutting edge: Ku70 is a novel cytosolic DNA sensor that induces type III rather than type I IFN. J Immunol 186, 4541-4545, doi:10.4049/jimmunol.1003389 (2011). 46 Monroe, K. M. et al. IFI16 DNA sensor is required for death of lymphoid CD4 T cells abortively infected with HIV. Science 343, 428-432, doi:10.1126/science.1243640 (2014). 47 Nakad, R. & Schumacher, B. DNA Damage Response and Immune Defense: Links and Mechanisms. Front Genet 7, 147, doi:10.3389/fgene.2016.00147 (2016). 48 Groth, P. et al. Homologous recombination repairs secondary replication induced DNA double-strand breaks after ionizing radiation. Nucleic acids research 40, 6585-6594, doi:10.1093/nar/gks315 (2012). 49 Harper, J. V., Anderson, J. A. & O'Neill, P. Radiation induced DNA DSBs: Contribution from stalled replication forks? DNA repair 9, 907-913, doi:10.1016/j.dnarep.2010.06.002 (2010).

114 50 Zafar, F., Seidler, S. B., Kronenberg, A., Schild, D. & Wiese, C. Homologous recombination contributes to the repair of DNA double-strand breaks induced by high-energy iron ions. Radiation research 173, 27-39, doi:10.1667/RR1910.1 (2010). 51 Gerelchuluun, A. et al. The major DNA repair pathway after both proton and carbon-ion radiation is NHEJ, but the HR pathway is more relevant in carbon ions. Radiation research 183, 345-356, doi:10.1667/RR13904.1 (2015). 52 Wang, H., Wang, X., Zhang, P. & Wang, Y. The Ku-dependent non-homologous end- joining but not other repair pathway is inhibited by high linear energy transfer ionizing radiation. DNA Repair (Amst) 7, 725-733, doi:10.1016/j.dnarep.2008.01.010 (2008). 53 Hartlerode, A. J. & Scully, R. Mechanisms of double-strand break repair in somatic mammalian cells. The Biochemical journal 423, 157-168, doi:10.1042/BJ20090942 (2009). 54 Godin, S. K., Sullivan, M. R. & Bernstein, K. A. Novel insights into RAD51 activity and regulation during homologous recombination and DNA replication. Biochemistry and cell biology = Biochimie et biologie cellulaire 94, 407-418, doi:10.1139/bcb-2016-0012 (2016). 55 Petermann, E., Orta, M. L., Issaeva, N., Schultz, N. & Helleday, T. Hydroxyurea-stalled replication forks become progressively inactivated and require two different RAD51- mediated pathways for restart and repair. Mol Cell 37, 492-502, doi:10.1016/j.molcel.2010.01.021 (2010). 56 Lundin, C. et al. RAD51 is involved in repair of damage associated with DNA replication in mammalian cells. J Mol Biol 328, 521-535, doi:Doi 10.1016/S0022-2836(03)00313-9 (2003). 57 Tsuzuki, T. et al. Targeted disruption of the Rad51 gene leads to lethality in embryonic mice. Proc Natl Acad Sci U S A 93, 6236-6240 (1996). 58 Wolf, C. et al. RPA and Rad51 constitute a cell intrinsic mechanism to protect the cytosol from self DNA. Nat Commun 7, 11752, doi:10.1038/ncomms11752 (2016). 59 Huen, M. S., Sy, S. M. & Chen, J. BRCA1 and its toolbox for the maintenance of genome integrity. Nature reviews. Molecular cell biology 11, 138-148, doi:10.1038/nrm2831 (2010). 60 Venkitaraman, A. R. Cancer suppression by the chromosome custodians, BRCA1 and BRCA2. Science 343, 1470-1475, doi:10.1126/science.1252230 (2014). 61 Schlacher, K. et al. Double-strand break repair-independent role for BRCA2 in blocking stalled replication fork degradation by MRE11. Cell 145, 529-542, doi:10.1016/j.cell.2011.03.041 (2011). 62 Schlacher, K., Wu, H. & Jasin, M. A distinct replication fork protection pathway connects Fanconi anemia tumor suppressors to RAD51-BRCA1/2. Cancer Cell 22, 106-116, doi:10.1016/j.ccr.2012.05.015 (2012). 63 Strickland, K. C. et al. Association and prognostic significance of BRCA1/2-mutation status with neoantigen load, number of tumor-infiltrating lymphocytes and expression of PD-1/PD-L1 in high grade serous ovarian cancer. Oncotarget 7, 13587-13598, doi:10.18632/oncotarget.7277 (2016).

115 64 Lakhani, S. R. et al. Multifactorial analysis of differences between sporadic breast cancers and cancers involving BRCA1 and BRCA2 mutations. Journal of the National Cancer Institute 90, 1138-1145 (1998). 65 Dutta, D. et al. BRCA1 Regulates IFI16 Mediated Nuclear Innate Sensing of Herpes Viral DNA and Subsequent Induction of the Innate Inflammasome and Interferon-beta Responses. PLoS Pathog 11, e1005030, doi:10.1371/journal.ppat.1005030 (2015). 66 Parkes, E. E. et al. Activation of STING-Dependent Innate Immune Signaling By S-Phase- Specific DNA Damage in Breast Cancer. Journal of the National Cancer Institute 109, doi:10.1093/jnci/djw199 (2017). 67 Yarden, R. I., Pardo-Reoyo, S., Sgagias, M., Cowan, K. H. & Brody, L. C. BRCA1 regulates the G2/M checkpoint by activating Chk1 kinase upon DNA damage. Nature genetics 30, 285-289, doi:10.1038/ng837 (2002). 68 Ciccia, A., McDonald, N. & West, S. C. Structural and functional relationships of the XPF/MUS81 family of proteins. Annu Rev Biochem 77, 259-287, doi:10.1146/annurev.biochem.77.070306.102408 (2008). 69 Ho, S. S. et al. The DNA Structure-Specific Endonuclease MUS81 Mediates DNA Sensor STING-Dependent Host Rejection of Prostate Cancer Cells. Immunity 44, 1177-1189, doi:10.1016/j.immuni.2016.04.010 (2016). 70 Kim, Y. J. & Wilson, D. M., 3rd. Overview of base excision repair biochemistry. Current molecular pharmacology 5, 3-13 (2012). 71 Robertson, A. B., Klungland, A., Rognes, T. & Leiros, I. DNA repair in mammalian cells: Base excision repair: the long and short of it. Cell Mol Life Sci 66, 981-993, doi:10.1007/s00018-009-8736-z (2009). 72 Mangerich, A. & Burkle, A. Pleiotropic cellular functions of PARP1 in longevity and aging: genome maintenance meets inflammation. Oxid Med Cell Longev 2012, 321653, doi:10.1155/2012/321653 (2012). 73 Mabley, J. G. et al. Potential role for 8-oxoguanine DNA glycosylase in regulating inflammation. FASEB J 19, 290-292, doi:10.1096/fj.04-2278fje (2005). 74 Fung, H. & Demple, B. A vital role for Ape1/Ref1 protein in repairing spontaneous DNA damage in human cells. Mol Cell 17, 463-470, doi:10.1016/j.molcel.2004.12.029 (2005). 75 Nishi, T. et al. Spatial redox regulation of a critical cysteine residue of NF-kappa B in vivo. J Biol Chem 277, 44548-44556, doi:10.1074/jbc.M202970200 (2002). 76 Lee, H. M. et al. Apurinic/apyrimidinic endonuclease 1 is a key modulator of keratinocyte inflammatory responses. J Immunol 183, 6839-6848, doi:10.4049/jimmunol.0901856 (2009). 77 Marteijn, J. A., Lans, H., Vermeulen, W. & Hoeijmakers, J. H. Understanding nucleotide excision repair and its roles in cancer and ageing. Nat Rev Mol Cell Biol 15, 465-481, doi:10.1038/nrm3822 (2014). 78 Scharer, O. D. Nucleotide excision repair in eukaryotes. Cold Spring Harb Perspect Biol 5, a012609, doi:10.1101/cshperspect.a012609 (2013). 79 Diderich, K., Alanazi, M. & Hoeijmakers, J. H. Premature aging and cancer in nucleotide excision repair-disorders. DNA Repair (Amst) 10, 772-780, doi:10.1016/j.dnarep.2011.04.025 (2011).

116 80 Vermeij, W. P., Hoeijmakers, J. H. & Pothof, J. Genome Integrity in Aging: Human Syndromes, Mouse Models, and Therapeutic Options. Annu Rev Pharmacol Toxicol 56, 427-445, doi:10.1146/annurev-pharmtox-010814-124316 (2016). 81 Weiner, A. M. & Gray, L. T. What role (if any) does the highly conserved CSB-PGBD3 fusion protein play in Cockayne syndrome? Mech Ageing Dev 134, 225-233, doi:10.1016/j.mad.2013.01.001 (2013). 82 Chatzinikolaou, G., Karakasilioti, I. & Garinis, G. A. DNA damage and innate immunity: links and trade-offs. Trends Immunol 35, 429-435, doi:10.1016/j.it.2014.06.003 (2014). 83 Brooks, P. J., Cheng, T. F. & Cooper, L. Do all of the neurologic diseases in patients with DNA repair gene mutations result from the accumulation of DNA damage? DNA Repair (Amst) 7, 834-848, doi:10.1016/j.dnarep.2008.01.017 (2008). 84 Crow, Y. J. & Livingston, J. H. Aicardi-Goutieres syndrome: an important Mendelian mimic of congenital infection. Dev Med Child Neurol 50, 410-416, doi:10.1111/j.1469- 8749.2008.02062.x (2008). 85 Wang, L. C. & Gautier, J. The Fanconi anemia pathway and ICL repair: implications for cancer therapy. Crit Rev Biochem Mol Biol 45, 424-439, doi:10.3109/10409238.2010.502166 (2010). 86 Fanconi, G. Familial constitutional panmyelocytopathy, Fanconi's anemia (F.A.). I. Clinical aspects. Seminars in hematology 4, 233-240 (1967). 87 Garcia-Higuera, I. et al. Interaction of the Fanconi anemia proteins and BRCA1 in a common pathway. Molecular cell 7, 249-262 (2001). 88 Smogorzewska, A. et al. Identification of the FANCI protein, a monoubiquitinated FANCD2 paralog required for DNA repair. Cell 129, 289-301, doi:10.1016/j.cell.2007.03.009 (2007). 89 Bunting, S. F. & Nussenzweig, A. Dangerous liaisons: Fanconi anemia and toxic nonhomologous end joining in DNA crosslink repair. Mol Cell 39, 164-166, doi:S1097- 2765(10)00538-1 [pii] 10.1016/j.molcel.2010.07.016 (2010). 90 Kee, Y. & D'Andrea, A. D. Expanded roles of the Fanconi anemia pathway in preserving genomic stability. Genes & development 24, 1680-1694, doi:10.1101/gad.1955310 (2010). 91 Moldovan, G. L. & D'Andrea, A. D. How the fanconi anemia pathway guards the genome. Annu Rev Genet 43, 223-249, doi:10.1146/annurev-genet-102108-134222 (2009). 92 Crossan, G. P. & Patel, K. J. The Fanconi anaemia pathway orchestrates incisions at sites of crosslinked DNA. The Journal of pathology 226, 326-337, doi:10.1002/path.3002 (2012). 93 Zhu, J. et al. FANCD2 influences replication fork processes and genome stability in response to clustered DSBs. Cell cycle 14, 1809-1822, doi:10.1080/15384101.2015.1036210 (2015). 94 Sumpter, R., Jr. et al. Fanconi Anemia Proteins Function in Mitophagy and Immunity. Cell 165, 867-881, doi:10.1016/j.cell.2016.04.006 (2016). 95 Dufour, C. et al. TNF-alpha and IFN-gamma are overexpressed in the bone marrow of Fanconi anemia patients and TNF-alpha suppresses erythropoiesis in vitro. Blood 102, 2053-2059, doi:10.1182/blood-2003-01-0114 (2003).

117 96 Cerboni, C. et al. The DNA Damage Response: A Common Pathway in the Regulation of NKG2D and DNAM-1 Ligand Expression in Normal, Infected, and Cancer Cells. Front Immunol 4, 508, doi:10.3389/fimmu.2013.00508 (2014). 97 Lund, J., Sato, A., Akira, S., Medzhitov, R. & Iwasaki, A. Toll-like receptor 9-mediated recognition of Herpes simplex virus-2 by plasmacytoid dendritic cells. J Exp Med 198, 513-520, doi:10.1084/jem.20030162 (2003). 98 Leifer, C. A. et al. TLR9 is localized in the endoplasmic reticulum prior to stimulation. J Immunol 173, 1179-1183 (2004). 99 Takeuchi, O. & Akira, S. Pattern recognition receptors and inflammation. Cell 140, 805- 820, doi:10.1016/j.cell.2010.01.022 (2010). 100 Takaoka, A. et al. DAI (DLM-1/ZBP1) is a cytosolic DNA sensor and an activator of innate immune response. Nature 448, 501-505, doi:10.1038/nature06013 (2007). 101 Burckstummer, T. et al. An orthogonal proteomic-genomic screen identifies AIM2 as a cytoplasmic DNA sensor for the inflammasome. Nat Immunol 10, 266-272, doi:10.1038/ni.1702 (2009). 102 Fernandes-Alnemri, T., Yu, J. W., Datta, P., Wu, J. & Alnemri, E. S. AIM2 activates the inflammasome and cell death in response to cytoplasmic DNA. Nature 458, 509-513, doi:10.1038/nature07710 (2009). 103 Hornung, V. et al. AIM2 recognizes cytosolic dsDNA and forms a caspase-1-activating inflammasome with ASC. Nature 458, 514-518, doi:10.1038/nature07725 (2009). 104 Chiu, Y. H., Macmillan, J. B. & Chen, Z. J. RNA polymerase III detects cytosolic DNA and induces type I interferons through the RIG-I pathway. Cell 138, 576-591, doi:10.1016/j.cell.2009.06.015 (2009). 105 Unterholzner, L. et al. IFI16 is an innate immune sensor for intracellular DNA. Nat Immunol 11, 997-1004, doi:10.1038/ni.1932 (2010). 106 Zhang, Z. et al. The helicase DDX41 senses intracellular DNA mediated by the adaptor STING in dendritic cells. Nat Immunol 12, 959-965, doi:10.1038/ni.2091 (2011). 107 Sun, L., Wu, J., Du, F., Chen, X. & Chen, Z. J. Cyclic GMP-AMP synthase is a cytosolic DNA sensor that activates the type I interferon pathway. Science 339, 786-791, doi:10.1126/science.1232458 (2013). 108 Li, X. D. et al. Pivotal roles of cGAS-cGAMP signaling in antiviral defense and immune adjuvant effects. Science 341, 1390-1394, doi:10.1126/science.1244040 (2013). 109 Wu, J. et al. Cyclic GMP-AMP is an endogenous second messenger in innate immune signaling by cytosolic DNA. Science 339, 826-830, doi:10.1126/science.1229963 (2013). 110 Ahn, J. et al. Inflammation-driven carcinogenesis is mediated through STING. Nat Commun 5, 5166, doi:10.1038/ncomms6166 (2014). 111 Ishikawa, H., Ma, Z. & Barber, G. N. STING regulates intracellular DNA-mediated, type I interferon-dependent innate immunity. Nature 461, 788-792, doi:10.1038/nature08476 (2009). 112 Woo, S. R. et al. STING-dependent cytosolic DNA sensing mediates innate immune recognition of immunogenic tumors. Immunity 41, 830-842, doi:10.1016/j.immuni.2014.10.017 (2014).

118 113 Deng, L. et al. STING-Dependent Cytosolic DNA Sensing Promotes Radiation-Induced Type I Interferon-Dependent Antitumor Immunity in Immunogenic Tumors. Immunity 41, 843-852, doi:10.1016/j.immuni.2014.10.019 (2014). 114 Lund, L. H. et al. Signal sequence deletion and fusion to tetanus toxoid epitope augment antitumor immune responses to a human carcinoembryonic antigen (CEA) plasmid DNA vaccine in a murine test system. Cancer Gene Ther 10, 365-376, doi:10.1038/sj.cgt.7700574 (2003). 115 Kerur, N. et al. IFI16 acts as a nuclear pathogen sensor to induce the inflammasome in response to Kaposi Sarcoma-associated herpesvirus infection. Cell Host Microbe 9, 363- 375, doi:10.1016/j.chom.2011.04.008 (2011). 116 Stetson, D. B., Ko, J. S., Heidmann, T. & Medzhitov, R. Trex1 prevents cell-intrinsic initiation of autoimmunity. Cell 134, 587-598, doi:10.1016/j.cell.2008.06.032 (2008). 117 Ishii, K. J. & Akira, S. Innate immune recognition of, and regulation by, DNA. Trends Immunol 27, 525-532, doi:10.1016/j.it.2006.09.002 (2006). 118 Erdal, E., Haider, S., Rehwinkel, J., Harris, A. L. & McHugh, P. J. A prosurvival DNA damage-induced cytoplasmic interferon response is mediated by end resection factors and is limited by Trex1. Genes Dev 31, 353-369, doi:10.1101/gad.289769.116 (2017). 119 Bartsch, K. et al. Absence of RNase H2 triggers generation of immunogenic micronuclei removed by autophagy. Human molecular genetics 26, 3960-3972, doi:10.1093/hmg/ddx283 (2017). 120 Dou, Z. et al. Cytoplasmic chromatin triggers inflammation in senescence and cancer. Nature 550, 402-406, doi:10.1038/nature24050 (2017). 121 Gluck, S. et al. Innate immune sensing of cytosolic chromatin fragments through cGAS promotes senescence. Nature cell biology 19, 1061-1070, doi:10.1038/ncb3586 (2017). 122 Harding, S. M. et al. Mitotic progression following DNA damage enables pattern recognition within micronuclei. Nature 548, 466-470, doi:10.1038/nature23470 (2017). 123 Mackenzie, K. J. et al. cGAS surveillance of micronuclei links genome instability to innate immunity. Nature 548, 461-465, doi:10.1038/nature23449 (2017). 124 Yang, H., Wang, H., Ren, J., Chen, Q. & Chen, Z. J. cGAS is essential for cellular senescence. Proc Natl Acad Sci U S A 114, E4612-E4620, doi:10.1073/pnas.1705499114 (2017). 125 Kawane, K. et al. Impaired thymic development in mouse embryos deficient in apoptotic DNA degradation. Nat Immunol 4, 138-144, doi:10.1038/ni881 (2003). 126 Hoss, M. et al. A human DNA editing enzyme homologous to the Escherichia coli DnaQ/MutD protein. The EMBO journal 18, 3868-3875, doi:10.1093/emboj/18.13.3868 (1999). 127 Mazur, D. J. & Perrino, F. W. Identification and expression of the TREX1 and TREX2 cDNA sequences encoding mammalian 3'-->5' . The Journal of biological chemistry 274, 19655-19660 (1999). 128 Grieves, J. L. et al. Exonuclease TREX1 degrades double-stranded DNA to prevent spontaneous lupus-like inflammatory disease. Proceedings of the National Academy of Sciences of the United States of America 112, 5117-5122, doi:10.1073/pnas.1423804112 (2015).

119 129 Yang, Y. G., Lindahl, T. & Barnes, D. E. Trex1 exonuclease degrades ssDNA to prevent chronic checkpoint activation and autoimmune disease. Cell 131, 873-886, doi:10.1016/j.cell.2007.10.017 (2007). 130 Morita, M. et al. Gene-targeted mice lacking the Trex1 (DNase III) 3'-->5' DNA exonuclease develop inflammatory myocarditis. Mol Cell Biol 24, 6719-6727, doi:10.1128/MCB.24.15.6719-6727.2004 (2004). 131 Gall, A. et al. Autoimmunity initiates in nonhematopoietic cells and progresses via lymphocytes in an interferon-dependent autoimmune disease. Immunity 36, 120-131, doi:10.1016/j.immuni.2011.11.018 (2012). 132 Ahn, J., Ruiz, P. & Barber, G. N. Intrinsic self-DNA triggers inflammatory disease dependent on STING. J Immunol 193, 4634-4642, doi:10.4049/jimmunol.1401337 (2014). 133 Gao, D. et al. Activation of cyclic GMP-AMP synthase by self-DNA causes autoimmune diseases. Proc Natl Acad Sci U S A 112, E5699-5705, doi:10.1073/pnas.1516465112 (2015). 134 Gray, E. E., Treuting, P. M., Woodward, J. J. & Stetson, D. B. Cutting Edge: cGAS Is Required for Lethal Autoimmune Disease in the Trex1-Deficient Mouse Model of Aicardi-Goutieres Syndrome. J Immunol 195, 1939-1943, doi:10.4049/jimmunol.1500969 (2015). 135 Hasan, M. et al. Cytosolic Nuclease TREX1 Regulates Oligosaccharyltransferase Activity Independent of Nuclease Activity to Suppress Immune Activation. Immunity 43, 463- 474, doi:10.1016/j.immuni.2015.07.022 (2015). 136 Lee-Kirsch, M. A., Wolf, C. & Gunther, C. Aicardi-Goutieres syndrome: a model disease for systemic autoimmunity. Clin Exp Immunol 175, 17-24, doi:10.1111/cei.12160 (2014). 137 Rice, G. et al. Heterozygous mutations in TREX1 cause familial chilblain lupus and dominant Aicardi-Goutieres syndrome. Am J Hum Genet 80, 811-815, doi:10.1086/513443 (2007). 138 DiFrancesco, J. C. et al. TREX1 C-terminal frameshift mutations in the systemic variant of retinal vasculopathy with cerebral leukodystrophy. Neurol Sci 36, 323-330, doi:10.1007/s10072-014-1944-9 (2015). 139 Matsunaga, A. et al. Carbon-ion beam treatment induces systemic antitumor immunity against murine squamous cell carcinoma. Cancer 116, 3740-3748, doi:10.1002/cncr.25134 (2010). 140 Ng, K. W., Marshall, E. A., Bell, J. C. & Lam, W. L. cGAS-STING and Cancer: Dichotomous Roles in Tumor Immunity and Development. Trends Immunol 39, 44-54, doi:10.1016/j.it.2017.07.013 (2018). 141 Song, S. et al. Decreased expression of STING predicts poor prognosis in patients with gastric cancer. Sci Rep 7, 39858, doi:10.1038/srep39858 (2017). 142 Bakhoum, S. F. et al. Chromosomal instability drives metastasis through a cytosolic DNA response. Nature 553, 467-472, doi:10.1038/nature25432 (2018). 143 Fumagalli, M. & d'Adda di Fagagna, F. SASPense and DDRama in cancer and ageing. Nature cell biology 11, 921-923, doi:10.1038/ncb0809-921 (2009). 144 Yu, Q. et al. DNA-damage-induced type I interferon promotes senescence and inhibits stem cell function. Cell Rep 11, 785-797, doi:10.1016/j.celrep.2015.03.069 (2015).

120 145 Moiseeva, O., Mallette, F. A., Mukhopadhyay, U. K., Moores, A. & Ferbeyre, G. DNA damage signaling and p53-dependent senescence after prolonged beta-interferon stimulation. Mol Biol Cell 17, 1583-1592, doi:10.1091/mbc.e05-09-0858 (2006). 146 Acosta, J. C. et al. Chemokine signaling via the CXCR2 receptor reinforces senescence. Cell 133, 1006-1018, doi:10.1016/j.cell.2008.03.038 (2008). 147 Kuilman, T. et al. Oncogene-induced senescence relayed by an interleukin-dependent inflammatory network. Cell 133, 1019-1031, doi:10.1016/j.cell.2008.03.039 (2008). 148 Krizhanovsky, V. et al. Senescence of activated stellate cells limits liver fibrosis. Cell 134, 657-667, doi:10.1016/j.cell.2008.06.049 (2008). 149 Krtolica, A., Parrinello, S., Lockett, S., Desprez, P. Y. & Campisi, J. Senescent fibroblasts promote epithelial cell growth and tumorigenesis: a link between cancer and aging. Proc Natl Acad Sci U S A 98, 12072-12077, doi:10.1073/pnas.211053698 (2001). 150 Coppe, J. P., Desprez, P. Y., Krtolica, A. & Campisi, J. The senescence-associated secretory phenotype: the dark side of tumor suppression. Annu Rev Pathol 5, 99-118, doi:10.1146/annurev-pathol-121808-102144 (2010). 151 Coppe, J. P., Kauser, K., Campisi, J. & Beausejour, C. M. Secretion of vascular endothelial growth factor by primary human fibroblasts at senescence. J Biol Chem 281, 29568- 29574, doi:10.1074/jbc.M603307200 (2006). 152 Coppe, J. P. et al. Senescence-associated secretory phenotypes reveal cell- nonautonomous functions of oncogenic RAS and the p53 tumor suppressor. PLoS Biol 6, 2853-2868, doi:10.1371/journal.pbio.0060301 (2008). 153 Schild-Poulter, C. et al. Association of autoantibodies with Ku and DNA repair proteins in connective tissue diseases. Rheumatology (Oxford) 47, 165-171, doi:10.1093/rheumatology/kem338 (2008). 154 Davies, R. C. et al. Defective DNA double-strand break repair in pediatric systemic lupus erythematosus. Arthritis Rheum 64, 568-578, doi:10.1002/art.33334 (2012). 155 Siddoo-Atwal, C., Haas, A. L. & Rosin, M. P. Elevation of interferon beta-inducible proteins in ataxia telangiectasia cells. Cancer research 56, 443-447 (1996). 156 Sugihara, T., Murano, H., Nakamura, M., Ichinohe, K. & Tanaka, K. Activation of interferon-stimulated genes by gamma-ray irradiation independently of the ataxia telangiectasia mutated-p53 pathway. Molecular cancer research : MCR 9, 476-484, doi:10.1158/1541-7786.MCR-10-0358 (2011). 157 Lan, Y. Y., Londono, D., Bouley, R., Rooney, M. S. & Hacohen, N. Dnase2a deficiency uncovers lysosomal clearance of damaged nuclear DNA via autophagy. Cell Rep 9, 180- 192, doi:10.1016/j.celrep.2014.08.074 (2014). 158 Karakasilioti, I. et al. DNA damage triggers a chronic autoinflammatory response, leading to fat depletion in NER progeria. Cell Metab 18, 403-415, doi:10.1016/j.cmet.2013.08.011 (2013). 159 Niedernhofer, L. J. et al. A new progeroid syndrome reveals that genotoxic stress suppresses the somatotroph axis. Nature 444, 1038-1043, doi:10.1038/nature05456 (2006). 160 Henning, K. A. et al. The Cockayne syndrome group A gene encodes a WD repeat protein that interacts with CSB protein and a subunit of RNA polymerase II TFIIH. Cell 82, 555- 564 (1995).

121 161 Li, T. & Chen, Z. J. The cGAS-cGAMP-STING pathway connects DNA damage to inflammation, senescence, and cancer. J Exp Med 215, 1287-1299, doi:10.1084/jem.20180139 (2018). 162 Troelstra, C. et al. ERCC6, a member of a subfamily of putative , is involved in Cockayne's syndrome and preferential repair of active genes. Cell 71, 939-953 (1992). 163 Crow, Y. J. et al. Mutations in genes encoding H2 subunits cause Aicardi- Goutieres syndrome and mimic congenital viral brain infection. Nature genetics 38, 910- 916, doi:10.1038/ng1842 (2006). 164 Hashimoto, Y., Ray Chaudhuri, A., Lopes, M. & Costanzo, V. Rad51 protects nascent DNA from Mre11-dependent degradation and promotes continuous DNA synthesis. Nat Struct Mol Biol 17, 1305-1311, doi:10.1038/nsmb.1927 (2010). 165 Krishna, S. et al. Mre11 and Ku regulation of double-strand break repair by gene conversion and break-induced replication. DNA repair 6, 797-808, doi:10.1016/j.dnarep.2007.01.006 (2007). 166 Nikjoo, H., O'Neill, P., Wilson, W. E. & Goodhead, D. T. Computational approach for determining the spectrum of DNA damage induced by ionizing radiation. Radiation research 156, 577-583 (2001). 167 Goodhead, D. T. Initial events in the cellular effects of ionizing radiations: clustered damage in DNA. International journal of radiation biology 65, 7-17 (1994). 168 Ward, J. F. DNA damage produced by ionizing radiation in mammalian cells: identities, mechanisms of formation, and reparability. Progress in nucleic acid research and molecular biology 35, 95-125 (1988). 169 Blaisdell, J. O., Harrison, L. & Wallace, S. S. Base excision repair processing of radiation- induced clustered DNA lesions. Radiat Prot Dosimetry 97, 25-31 (2001). 170 Nikjoo, H., O'Neill, P., Terrissol, M. & Goodhead, D. T. Quantitative modelling of DNA damage using Monte Carlo track structure method. Radiation and environmental biophysics 38, 31-38 (1999). 171 Kanaar, R., Hoeijmakers, J. H. & van Gent, D. C. Molecular mechanisms of DNA double strand break repair. Trends Cell Biol 8, 483-489 (1998). 172 Jackson, S. P. Sensing and repairing DNA double-strand breaks. Carcinogenesis 23, 687- 696 (2002). 173 Yajima, H. et al. The complexity of DNA double strand breaks is a critical factor enhancing end-resection. DNA repair 12, 936-946, doi:10.1016/j.dnarep.2013.08.009 (2013). 174 Silva, M. C. et al. RAD51 variant proteins from human lung and kidney tumors exhibit DNA strand exchange defects. DNA repair 42, 44-55, doi:10.1016/j.dnarep.2016.02.008 (2016). 175 Prakash, R., Zhang, Y., Feng, W. & Jasin, M. Homologous recombination and human health: the roles of BRCA1, BRCA2, and associated proteins. Cold Spring Harbor perspectives in biology 7, a016600, doi:10.1101/cshperspect.a016600 (2015). 176 Schild, D. & Wiese, C. Overexpression of RAD51 suppresses recombination defects: a possible mechanism to reverse genomic instability. Nucleic acids research 38, 1061- 1070, doi:10.1093/nar/gkp1063 (2010).

122 177 Yoshikawa, K. et al. Abnormal expression of BRCA1 and BRCA1-interactive DNA-repair proteins in breast carcinomas. International journal of cancer 88, 28-36 (2000). 178 Nowarski, R., Gagliani, N., Huber, S. & Flavell, R. A. Innate immune cells in inflammation and cancer. Cancer immunology research 1, 77-84, doi:10.1158/2326-6066.CIR-13-0081 (2013). 179 Goldszmid, R. S. & Trinchieri, G. The price of immunity. Nature immunology 13, 932-938, doi:10.1038/ni.2422 (2012). 180 Hartlova, A. et al. DNA damage primes the type I interferon system via the cytosolic DNA sensor STING to promote anti-microbial innate immunity. Immunity 42, 332-343, doi:10.1016/j.immuni.2015.01.012 (2015). 181 Asaithamby, A. & Chen, D. J. Cellular responses to DNA double-strand breaks after low- dose gamma-irradiation. Nucleic acids research 37, 3912-3923, doi:10.1093/nar/gkp237 (2009). 182 Peng, G. et al. Genome-wide transcriptome profiling of homologous recombination DNA repair. Nature communications 5, 3361, doi:10.1038/ncomms4361 (2014). 183 Takeshima, T. et al. Key role for neutrophils in radiation-induced antitumor immune responses: Potentiation with G-CSF. Proceedings of the National Academy of Sciences of the United States of America 113, 11300-11305, doi:10.1073/pnas.1613187113 (2016). 184 Asaithamby, A. et al. Irreparable complex DNA double-strand breaks induce chromosome breakage in organotypic three-dimensional human lung epithelial cell culture. Nucleic acids research 39, 5474-5488, doi:10.1093/nar/gkr149 (2011). 185 Puente, B. N. et al. The oxygen-rich postnatal environment induces cardiomyocyte cell- cycle arrest through DNA damage response. Cell 157, 565-579, doi:10.1016/j.cell.2014.03.032 (2014). 186 Rocke, D. M. & Durbin, B. A model for measurement error for gene expression arrays. J Comput Biol 8, 557-569, doi:10.1089/106652701753307485 (2001). 187 Dozmorov, I. & Centola, M. An associative analysis of gene expression array data. Bioinformatics 19, 204-211 (2003). 188 Dozmorov, I. & Lefkovits, I. Internal standard-based analysis of microarray data. Part 1: analysis of differential gene expressions. Nucleic Acids Res 37, 6323-6339, doi:gkp706 [pii] 10.1093/nar/gkp706 (2009). 189 Dozmorov, M. G., Guthridge, J. M., Hurst, R. E. & Dozmorov, I. M. A comprehensive and universal method for assessing the performance of differential gene expression analyses. PloS one 5, doi:10.1371/journal.pone.0012657 (2010). 190 Zhu, J. et al. FANCD2 influences replication fork processes and genome stability in response to clustered DSBs. Cell cycle 14, 1809-1822, doi:10.1080/15384101.2015.1036210 (2015). 191 Utz, P. J. & Anderson, P. Life and death decisions: regulation of apoptosis by proteolysis of signaling molecules. Cell death and differentiation 7, 589-602, doi:10.1038/sj.cdd.4400696 (2000). 192 Boulares, A. H. et al. Role of poly(ADP-ribose) polymerase (PARP) cleavage in apoptosis. Caspase 3-resistant PARP mutant increases rates of apoptosis in transfected cells. The Journal of biological chemistry 274, 22932-22940 (1999).

123 193 Chen, X. et al. Suberoylanilide hydroxamic acid as a radiosensitizer through modulation of RAD51 protein and inhibition of homology-directed repair in multiple myeloma. Mol Cancer Res 10, 1052-1064, doi:10.1158/1541-7786.MCR-11-0587 (2012). 194 Liu, Y. P. et al. Endoplasmic reticulum stress regulates the innate immunity critical transcription factor IRF3. Journal of immunology 189, 4630-4639, doi:10.4049/jimmunol.1102737 (2012). 195 Tanaka, Y. & Chen, Z. J. STING specifies IRF3 phosphorylation by TBK1 in the cytosolic DNA signaling pathway. Science signaling 5, ra20, doi:10.1126/scisignal.2002521 (2012). 196 Abe, T. & Barber, G. N. Cytosolic-DNA-mediated, STING-dependent proinflammatory gene induction necessitates canonical NF-kappaB activation through TBK1. Journal of virology 88, 5328-5341, doi:10.1128/JVI.00037-14 (2014). 197 Liu, S. et al. Phosphorylation of innate immune adaptor proteins MAVS, STING, and TRIF induces IRF3 activation. Science 347, aaa2630, doi:10.1126/science.aaa2630 (2015). 198 Shibata, A. et al. DNA double-strand break repair pathway choice is directed by distinct MRE11 nuclease activities. Molecular cell 53, 7-18, doi:10.1016/j.molcel.2013.11.003 (2014). 199 Dupre, A. et al. A forward chemical genetic screen reveals an inhibitor of the Mre11- Rad50-Nbs1 complex. Nature chemical biology 4, 119-125, doi:10.1038/nchembio.63 (2008). 200 Sakaue-Sawano, A. et al. Visualizing spatiotemporal dynamics of multicellular cell-cycle progression. Cell 132, 487-498, doi:10.1016/j.cell.2007.12.033 (2008). 201 Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646- 674, doi:10.1016/j.cell.2011.02.013 (2011). 202 Blackburn, E. H. Switching and signaling at the telomere. Cell 106, 661-673 (2001). 203 de Lange, T. Shelterin: the protein complex that shapes and safeguards human telomeres. Genes & development 19, 2100-2110, doi:10.1101/gad.1346005 (2005). 204 O'Sullivan, R. J. & Karlseder, J. Telomeres: protecting chromosomes against genome instability. Nature reviews. Molecular cell biology 11, 171-181, doi:10.1038/nrm2848 (2010). 205 Sherr, C. J. Principles of tumor suppression. Cell 116, 235-246 (2004). 206 Schipler, A. & Iliakis, G. DNA double-strand-break complexity levels and their possible contributions to the probability for error-prone processing and repair pathway choice. Nucleic acids research 41, 7589-7605, doi:10.1093/nar/gkt556 (2013). 207 Soni, A. et al. Requirement for Parp-1 and DNA ligases 1 or 3 but not of Xrcc1 in chromosomal translocation formation by backup end joining. Nucleic acids research 42, 6380-6392, doi:10.1093/nar/gku298 (2014). 208 Fumagalli, M. et al. Telomeric DNA damage is irreparable and causes persistent DNA- damage-response activation. Nature cell biology 14, 355-365, doi:10.1038/ncb2466 (2012). 209 Hewitt, G. et al. Telomeres are favoured targets of a persistent DNA damage response in ageing and stress-induced senescence. Nature communications 3, 708, doi:10.1038/ncomms1708 (2012).

124 210 Rochette, P. J. & Brash, D. E. Human telomeres are hypersensitive to UV-induced DNA Damage and refractory to repair. PLoS genetics 6, e1000926, doi:10.1371/journal.pgen.1000926 (2010). 211 Mao, P. et al. Homologous recombination-dependent repair of telomeric DSBs in proliferating human cells. Nat Commun 7, 12154, doi:10.1038/ncomms12154 (2016). 212 Miller, D., Reynolds, G. E., Mejia, R., Stark, J. M. & Murnane, J. P. Subtelomeric regions in mammalian cells are deficient in DNA double-strand break repair. DNA repair 10, 536- 544, doi:10.1016/j.dnarep.2011.03.001 (2011). 213 Cho, N. W., Dilley, R. L., Lampson, M. A. & Greenberg, R. A. Interchromosomal Homology Searches Drive Directional ALT Telomere Movement and Synapsis. Cell 159, 108-121, doi:10.1016/j.cell.2014.08.030 (2014). 214 Silva, B. A., Stambaugh, J. R., Yokomori, K., Shah, J. V. & Berns, M. W. DNA damage to a single chromosome end delays anaphase onset. The Journal of biological chemistry 289, 22771-22784, doi:10.1074/jbc.M113.535955 (2014). 215 Sun, L. et al. Targeted DNA damage at individual telomeres disrupts their integrity and triggers cell death. Nucleic acids research, doi:10.1093/nar/gkv598 (2015). 216 d'Adda di Fagagna, F. et al. A DNA damage checkpoint response in telomere-initiated senescence. Nature 426, 194-198, doi:10.1038/nature02118 (2003). 217 Takai, H., Smogorzewska, A. & de Lange, T. DNA damage foci at dysfunctional telomeres. Current biology : CB 13, 1549-1556 (2003). 218 de Lange, T. How telomeres solve the end-protection problem. Science 326, 948-952, doi:10.1126/science.1170633 (2009). 219 Denchi, E. L. Give me a break: how telomeres suppress the DNA damage response. DNA repair 8, 1118-1126, doi:10.1016/j.dnarep.2009.04.013 (2009). 220 Kranzusch, P. J., Lee, A. S., Berger, J. M. & Doudna, J. A. Structure of human cGAS reveals a conserved family of second-messenger enzymes in innate immunity. Cell reports 3, 1362-1368, doi:10.1016/j.celrep.2013.05.008 (2013). 221 Kranzusch, P. J. & Vance, R. E. cGAS dimerization entangles DNA recognition. Immunity 39, 992-994, doi:10.1016/j.immuni.2013.11.012 (2013). 222 Civril, F. et al. Structural mechanism of cytosolic DNA sensing by cGAS. Nature 498, 332- 337, doi:10.1038/nature12305 (2013). 223 Li, X. et al. Cyclic GMP-AMP synthase is activated by double-stranded DNA-induced oligomerization. Immunity 39, 1019-1031, doi:10.1016/j.immuni.2013.10.019 (2013). 224 Zhang, X. et al. The cytosolic DNA sensor cGAS forms an oligomeric complex with DNA and undergoes switch-like conformational changes in the activation loop. Cell reports 6, 421-430, doi:10.1016/j.celrep.2014.01.003 (2014). 225 Kobayashi, J. et al. NBS1 localizes to gamma-H2AX foci through interaction with the FHA/BRCT domain. Current biology : CB 12, 1846-1851 (2002). 226 Tauchi, H. et al. The forkhead-associated domain of NBS1 is essential for nuclear foci formation after irradiation but not essential for hRAD50[middle dot]hMRE11[middle dot]NBS1 complex DNA repair activity. The Journal of biological chemistry 276, 12-15, doi:10.1074/jbc.C000578200 (2001).

125 227 Bakkenist, C. J. & Kastan, M. B. DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature 421, 499-506, doi:10.1038/nature01368 (2003). 228 Lim, D. S. et al. ATM phosphorylates p95/nbs1 in an S-phase checkpoint pathway. Nature 404, 613-617, doi:10.1038/35007091 (2000). 229 Wu, X. et al. ATM phosphorylation of Nijmegen breakage syndrome protein is required in a DNA damage response. Nature 405, 477-482, doi:10.1038/35013089 (2000). 230 Gao, R., Singh, R., Kaul, Z., Kaul, S. C. & Wadhwa, R. Targeting of DNA Damage Signaling Pathway Induced Senescence and Reduced Migration of Cancer cells. The journals of gerontology. Series A, Biological sciences and medical sciences 70, 701-713, doi:10.1093/gerona/glu019 (2015). 231 Khongkow, P. et al. FOXM1 targets NBS1 to regulate DNA damage-induced senescence and epirubicin resistance. Oncogene 33, 4144-4155, doi:10.1038/onc.2013.457 (2014). 232 Sakamoto, S. et al. Homologous recombination repair is regulated by domains at the N- and C-terminus of NBS1 and is dissociated with ATM functions. Oncogene 26, 6002- 6009, doi:10.1038/sj.onc.1210428 (2007). 233 Chen, B. et al. Dynamic imaging of genomic loci in living human cells by an optimized CRISPR/Cas system. Cell 155, 1479-1491, doi:10.1016/j.cell.2013.12.001 (2013). 234 Jinek, M. et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816-821, doi:10.1126/science.1225829 (2012). 235 Banaszynski, L. A., Chen, L. C., Maynard-Smith, L. A., Ooi, A. G. & Wandless, T. J. A rapid, reversible, and tunable method to regulate protein function in living cells using synthetic small molecules. Cell 126, 995-1004, doi:10.1016/j.cell.2006.07.025 (2006). 236 Deckbar, D. et al. Chromosome breakage after G2 checkpoint release. The Journal of cell biology 176, 749-755, doi:10.1083/jcb.200612047 (2007). 237 Terzoudi, G. I., Manola, K. N., Pantelias, G. E. & Iliakis, G. Checkpoint abrogation in G2 compromises repair of chromosomal breaks in ataxia telangiectasia cells. Cancer Res 65, 11292-11296, doi:10.1158/0008-5472.CAN-05-2148 (2005). 238 Lai, T. P. et al. A method for measuring the distribution of the shortest telomeres in cells and tissues. Nat Commun 8, 1356, doi:10.1038/s41467-017-01291-z (2017). 239 Vassilev, L. T. et al. Selective small-molecule inhibitor reveals critical mitotic functions of human CDK1. Proceedings of the National Academy of Sciences of the United States of America 103, 10660-10665, doi:10.1073/pnas.0600447103 (2006). 240 van Steensel, B., Smogorzewska, A. & de Lange, T. TRF2 protects human telomeres from end-to-end fusions. Cell 92, 401-413 (1998). 241 Mender, I., Gryaznov, S., Dikmen, Z. G., Wright, W. E. & Shay, J. W. Induction of telomere dysfunction mediated by the telomerase substrate precursor 6-thio-2'- deoxyguanosine. Cancer Discov 5, 82-95, doi:10.1158/2159-8290.CD-14-0609 (2015). 242 Mender, I. et al. Telomerase-Mediated Strategy for Overcoming Non-Small Cell Lung Cancer Targeted Therapy and Chemotherapy Resistance. Neoplasia 20, 826-837, doi:10.1016/j.neo.2018.06.002 (2018). 243 Sengupta, S. et al. Induced Telomere Damage to Treat Telomerase Expressing Therapy- Resistant Pediatric Brain Tumors. Molecular cancer therapeutics 17, 1504-1514, doi:10.1158/1535-7163.MCT-17-0792 (2018).

126 244 Csaszar, K., Spackova, N., Stefl, R., Sponer, J. & Leontis, N. B. Molecular dynamics of the frame-shifting pseudoknot from beet western yellows virus: the role of non-Watson- Crick base-pairing, ordered hydration, cation binding and base mutations on stability and unfolding. J Mol Biol 313, 1073-1091, doi:10.1006/jmbi.2001.5100 (2001). 245 Stefl, R., Spackova, N., Berger, I., Koca, J. & Sponer, J. Molecular dynamics of DNA quadruplex molecules containing inosine, 6-thioguanine and 6-thiopurine. Biophys J 80, 455-468, doi:10.1016/S0006-3495(01)76028-6 (2001). 246 Sherwood, S. W., Rush, D., Ellsworth, J. L. & Schimke, R. T. Defining cellular senescence in IMR-90 cells: a flow cytometric analysis. Proc Natl Acad Sci U S A 85, 9086-9090 (1988). 247 Zhao, S. et al. Functional link between ataxia-telangiectasia and Nijmegen breakage syndrome gene products. Nature 405, 473-477, doi:10.1038/35013083 (2000). 248 Falck, J., Coates, J. & Jackson, S. P. Conserved modes of recruitment of ATM, ATR and DNA-PKcs to sites of DNA damage. Nature 434, 605-611, doi:10.1038/nature03442 (2005). 249 Vincent, J. et al. Small molecule inhibition of cGAS reduces interferon expression in primary macrophages from autoimmune mice. Nature communications 8, 750, doi:10.1038/s41467-017-00833-9 (2017). 250 Umbreit, N. T. & Pellman, D. Genome jail-break triggers lockdown. Nature 550, 340-341 (2017). 251 Yoh, S. M. et al. PQBP1 Is a Proximal Sensor of the cGAS-Dependent Innate Response to HIV-1. Cell 161, 1293-1305, doi:10.1016/j.cell.2015.04.050 (2015). 252 Gehrke, N. et al. Oxidative damage of DNA confers resistance to cytosolic nuclease TREX1 degradation and potentiates STING-dependent immune sensing. Immunity 39, 482-495, doi:10.1016/j.immuni.2013.08.004 (2013). 253 Klein, H. L. The consequences of Rad51 overexpression for normal and tumor cells. DNA repair 7, 686-693, doi:10.1016/j.dnarep.2007.12.008 (2008). 254 Hine, C. M., Seluanov, A. & Gorbunova, V. Use of the Rad51 promoter for targeted anti- cancer therapy. Proceedings of the National Academy of Sciences of the United States of America 105, 20810-20815, doi:10.1073/pnas.0807990106 (2008). 255 Wiegmans, A. P. et al. Rad51 supports triple negative breast cancer metastasis. Oncotarget 5, 3261-3272, doi:10.18632/oncotarget.1923 (2014). 256 Maacke, H. et al. Over-expression of wild-type Rad51 correlates with histological grading of invasive ductal breast cancer. International journal of cancer 88, 907-913 (2000). 257 Mitra, A. et al. Overexpression of RAD51 occurs in aggressive prostatic cancer. Histopathology 55, 696-704, doi:10.1111/j.1365-2559.2009.03448.x (2009). 258 Connell, P. P. et al. Pilot study examining tumor expression of RAD51 and clinical outcomes in human head cancers. International journal of oncology 28, 1113-1119 (2006). 259 Antoniou, A. et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case Series unselected for family history: a combined analysis of 22 studies. American journal of human genetics 72, 1117-1130, doi:10.1086/375033 (2003).

127 260 Gardini, A., Baillat, D., Cesaroni, M. & Shiekhattar, R. Genome-wide analysis reveals a role for BRCA1 and PALB2 in transcriptional co-activation. The EMBO journal 33, 890- 905, doi:10.1002/embj.201385567 (2014). 261 Mullan, P. B., Quinn, J. E. & Harkin, D. P. The role of BRCA1 in transcriptional regulation and cell cycle control. Oncogene 25, 5854-5863, doi:10.1038/sj.onc.1209872 (2006). 262 Savage, K. I. et al. BRCA1 deficiency exacerbates estrogen-induced DNA damage and genomic instability. Cancer research 74, 2773-2784, doi:10.1158/0008-5472.CAN-13- 2611 (2014). 263 Zhu, Q. et al. BRCA1 tumour suppression occurs via heterochromatin-mediated silencing. Nature 477, 179-184, doi:10.1038/nature10371 (2011). 264 Pathania, S. et al. BRCA1 is required for postreplication repair after UV-induced DNA damage. Molecular cell 44, 235-251, doi:10.1016/j.molcel.2011.09.002 (2011). 265 Savage, K. I. et al. Identification of a BRCA1-mRNA splicing complex required for efficient DNA repair and maintenance of genomic stability. Molecular cell 54, 445-459, doi:10.1016/j.molcel.2014.03.021 (2014). 266 Milanovic, M. et al. Senescence-associated reprogramming promotes cancer stemness. Nature 553, 96-100, doi:10.1038/nature25167 (2018). 267 Crasta, K. et al. DNA breaks and chromosome pulverization from errors in mitosis. Nature 482, 53-58, doi:10.1038/nature10802 (2012). 268 Holland, A. J. & Cleveland, D. W. Chromoanagenesis and cancer: mechanisms and consequences of localized, complex chromosomal rearrangements. Nat Med 18, 1630- 1638, doi:10.1038/nm.2988 (2012). 269 Huang, Y. et al. The fate of micronucleated cells post X-irradiation detected by live cell imaging. DNA Repair (Amst) 10, 629-638, doi:10.1016/j.dnarep.2011.04.010 (2011). 270 Huang, Y. et al. Lagging chromosomes entrapped in micronuclei are not 'lost' by cells. Cell Res 22, 932-935, doi:10.1038/cr.2012.26 (2012). 271 Zhang, C. Z. et al. Chromothripsis from DNA damage in micronuclei. Nature 522, 179- 184, doi:10.1038/nature14493 (2015). 272 Kind, J. et al. Single-cell dynamics of genome-nuclear lamina interactions. Cell 153, 178- 192, doi:10.1016/j.cell.2013.02.028 (2013). 273 Kind, J. et al. Genome-wide Maps of Nuclear Lamina Interactions in Single Human Cells. Cell, doi:10.1016/j.cell.2015.08.040 (2015). 274 Vogel, M. J., Peric-Hupkes, D. & van Steensel, B. Detection of in vivo protein-DNA interactions using DamID in mammalian cells. Nat Protoc 2, 1467-1478, doi:10.1038/nprot.2007.148 (2007).

128

129