Thesis Submitted in Conformity with the Requirements for the Degree of Doctor of Philosophy Department of Immunology University of Toronto

Molecular Requirements of Class Switch Recombination

Alexanda Ka-Shing Ling

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Immunology University of Toronto

Molecular Requirements of Class Switch Recombination

Alexanda Ka-Shing Ling

Doctor of Philosophy

Department of Immunology University of Toronto

2020 Abstract To meet the depth and breadth of antigen space brought to bear by potential pathogens, adaptive immunity requires somatic diversification of antigen receptors through programmed DNA damage and concomitant recombination and repair of antigen receptor loci. This biological fact is conserved in all vertebrate lineages and is elaborated upon with secondary phenomena such as class switch recombination (CSR). Together, these recombinatorial adaptive immune processes are recapitulated billions of times over in each organism, mostly without negative consequences such as undesirable mutagenesis and genomic instability. This juxtaposition of damage and repair in a physiological model has been felicitous to the understanding of DNA repair more generally, and insights into the basis of adaptive immunity and DNA repair have played off together symbiotically. In that tradition, this manuscript endeavours to examine the molecular requirements of class switch recombination (CSR) along two complementary axes. Firstly, although it is known that double-stranded breaks (DSB) are substrates necessary for CSR, it is unknown how the structure and polarity of DSBs affects downstream DNA repair. Using the Cas9 enzyme and related variants, I generated DSBs of defined structures near the switch regions of the immunoglobulin heavy chain locus (Igh) and examined the resultant CSR frequency and repair junction characteristics. I found that single-stranded breaks upwards of 250 bp apart on opposite strands could resolve as DSBs in CSR. Moreover, 5′ DSBs were more preferred CSR substrates

ii than 3′ DSBs, although both types of staggered DSBs had a higher level of resection and microhomology usage at the repair junction than blunt DSBs. Taken together, these observations suggest that DSB structure influences the type of repair effected as well as the frequency of productive CSR. I also investigated the effect of the newly-identified shieldin complex downstream of 53BP1 on concluding CSR. Shld2–/– B cells have profoundly impaired CSR, although B cell development and V(D)J recombination were not grossly affected. Additionally, I observed an increased Iglo population in shieldin-deficient switching B cells that was the result of extreme resection impairing Ig expression, rather than increased inversional recombination, and congruent with the proposed function of shieldin.

iii

Acknowledgments

Dulce et Decorum est pro Scientia laborat,

autem nihil est fructum plus

quam familia et amici convivens.

Table of Contents

Acknowledgments...... iv

Table of Contents ...... v

List of Tables ...... vii

List of Figures ...... viii

List of Abbreviations ...... x

Chapter 1 INTRODUCTION ...... 1

Origins of Adaptive Immunity ...... 1

1.1 Humoral immunity ...... 1

1.2 Antigen receptor diversification ...... 2

1.2.1 V(D)J recombination ...... 3

1.2.2 Lymphocyte development ...... 5

1.2.3 Gene conversion...... 7

1.2.4 Somatic hypermutation ...... 8

1.2.5 Class switch recombination ...... 9

1.3 DNA Damage Response and Repair ...... 13

1.3.1 DNA damage detection and signaling ...... 13

1.3.2 Base excision repair ...... 15

1.3.3 Mismatch repair ...... 15

1.3.4 Non-homologous end joining ...... 16

1.3.5 Alternative end joining ...... 19

1.3.6 Homologous recombination ...... 19

1.4 Prokaryotic adaptive immunity ...... 20

1.5 CSR and NHEJ repair as mutually reinforcing paradigms of inquiry ...... 20

Chapter 2 DOUBLE-STRANDED DNA BREAK POLARITY SKEWS REPAIR PATHWAY CHOICE DURING CLASS-SWITCH RECOMBINATION ...... 21

Abstract ...... 22

2.1 Introduction ...... 22

2.2 Results ...... 23

2.2.1 Cas9- and nickase-mediated DSBs induce CSR ...... 23

2.2.2 CSR has a preference for 5′ DSBs over 3′ DSBs ...... 27

2.2.3 Staggered DSBs promote alternative end-joining during CSR...... 27

2.3 Discussion ...... 31

2.4 Experimental Methods ...... 33

Chapter 3 THE SHIELDIN COMPLEX PROMOTES CLASS SWITCH RECOMBINATION ...... 36

Abstract ...... 37

3.1 Introduction ...... 37

3.2 Results ...... 39

3.2.1 Lymphocyte development and B cell populations are largely unaffected by Shld2 deficiency ...... 39

3.2.2 SHLD2 is necessary for CSR ...... 42

3.2.3 The shieldin complex and other NHEJ factors exhibit an Iglo population upon CSR ...... 45

3.2.4 The Iglo population is not the result of increased inversional recombination ...... 49

3.2.5 CSR in 53bp1–/– and Shld2–/–/– CH12 cells leads to aberrant recombination involving deletions in the acceptor constant region ...... 51

3.3 Discussion ...... 57

3.4 Experimental Methods ...... 58

Conclusions ...... 61

References ...... 63

Appendices ...... 85

List of Tables

Appendix Table 1. Oligonucleotides used in Chapter 2 for Cas9-mediated CSR and gene perturbation...... 90

Appendix Table 2. Primers used in Chapter 3...... 104

Appendix Table 3. Cas9 sgRNA used in Chapter 3...... 107

vii

List of Figures

Figure 1. Overview of class switch recombination...... 12

Figure 2. Overview of factors and signaling in NHEJ ...... 18

Figure 3. Cas9-mediated DSBs can recapitulate CSR and associated recombination...... 25

Figure 4. The μ–α junctions derived from 5′ DSB intermediates have increased resection...... 30

Figure 5. SHLD2 does not affect lymphocyte development or V(D)J recombination...... 40

Figure 6. SHLD2-deficient mice have defects in class switch recombination...... 44

Figure 7. Shld2-deficient B cells and B cells deficient in other NHEJ factors exhibit an Iglo population upon CSR induction...... 46

Figure 8. Reduced Ig expression in 53BP1, SHLD2, and SHLD3-deficient CH12 cells is permanent and dependent on CSR...... 48

Figure 9. Inversional recombination in CH12 and ex vivo B cells are rare events...... 50

Figure 10. Loss of Ig cell-surface expression in CH12 cells is accompanied by aberrant IgA transcripts...... 53

Figure 11. Loss of Ig expression in CH12 cells is accompanied by large deletions within the IgA constant region...... 55

Appendix Figure 1. Cas9-mediated DSBs induce CSR in multiple cell lines...... 85

Appendix Figure 2. Non-canonical lesions mediate switching...... 87

Appendix Figure 3. Nucleolytic resection prior to ligation following Cas9 DSBs...... 88

Appendix Figure 4. Generation of Aid–/– Lig4–/– CH12 cells...... 89

Appendix Figure 5. Gating strategies for the assessment of B and T cell populations in WT and Shld2−/− mice...... 93

viii

Appendix Figure 6. Sterile transcript or AID protein levels were unaffected in Shld2−/− B cells. 94

Appendix Figure 7. Germinal center B cell frequency is not affected by SHLD2-deficiency...... 95

Appendix Figure 8. SHLD2- and 53BP1-deficiency has no effect on Cas9-mediated switching. 97

Appendix Figure 9. Genotypes of novel mutant CH12 cells generated for this study...... 98

Appendix Figure 10. CSR induces a permanent loss of Ig expression in CH12 cells...... 99

Appendix Figure 11. DC-DDPCR assay validation ...... 101

Appendix Figure 12. Flow Cytometry analysis of Iglo WT, 53bp1−/− and Shld2−/−/− CH12 subclones...... 102

Appendix Figure 13. Repair junctions in Iglo cells have characteristics of alternative end joining...... 103

List of Abbreviations

53BP1 Tumor suppressor p53-binding protein 1

AID Activation-induced cytidine deaminase

AP Apurinic/apyrimidinic site

ATM Ataxia-telangiectasia mutated

ATR ataxia telangiectasia and Rad3-related

BCR B cell receptor

BER Base excision repair

BM Bone marrow

CDR Complementarity-determining region

CRISPR Clustered Regular Interspaced Short Palindromic Repeats

CSR Class switch recombination

DNA-PKcs DNA-dependent serine/threonine protein kinase, catalytic subunit

DSB Double-stranded DNA break

DZ Dark zone (of the germinal centre)

EXO1 Exonuclease I

GC Germinal centre

GLT Germline transcript

Ig Immunoglobulin

IgH Immunoglobulin heavy chain

IgL Immunoglobulin light chain

LIG4 Ligase IV

LZ Light zone (of the germinal centre)

MDC1 Mediator of DNA damage checkpoint 1

MMR Mismatch repair

MRN MRE11-RAD50-NBS1

MLH MutL homologue

MSH MutS homologue

NBS1 Nijmegen breakage syndrome 1

PARP Poly(ADP-ribose) polymerase

PCNA Proliferating cell nuclear antigen

PIKK Phosphoinositide-3-kinase-related protein kinase

PTIP PAX transcription activation domain interacting protein

RAG Recombination activating gene

RIF1 Rap1 interacting factor

RPA Replication protein A

SCID Severe combined immunodeficiency

SHLD Shieldin

SHM Somatic hypermutation

SSB Single-stranded DNA break

ssDNA Single-stranded DNA

TCR T cell receptor

V(D)J Variable (Diversity) Joining

WRCY (W = A or T, R = purine, Y = pyrimidine)

XLF XRCC4-like factor

XRCC4 X-ray repair cross-complementation group 4

xii

Chapter 1 INTRODUCTION Origins of Adaptive Immunity 1.1 Humoral immunity

First described by Paul Ehrlich at the turn of the 20th century, humoral immunity is the phenomenon by which specific immunity is conferred against microorganisms, environmental toxins, or noxious or benign stimuli by molecules in extracellular fluids. The principal agent underwriting humoral immunity are antibodies or immunoglobulins (Ig). Ig are composed of a homodimer of heavy polypeptide chains (IgH), each of which is associated with a light chain (IgL) and together form a bivalent Y-shaped protein complex. The IgH chain is composed of four or five Ig domains, depending on the isotype, and can be further sub-characterized into an N-terminal variable domain and (three or four) C-terminal constant domains. The IgL chain is shorter but similarly composed of a N-terminal variable region domain and a C-terminal constant domain. The variable domains of the IgH and IgL chains can contact antigen through three variable peptide loops known as complementarity-determining regions (CDR), and the confluence of different CDR on both IgH and IgL together determine antigen specificity. Ig bind specifically to macromolecular antigens such as proteins and carbohydrates, or even smaller moieties and species. Subsequent to binding, Ig can mediate immunity through several different modes such as direct neutralization, activation of downstream molecular factors such as complement, or directing the activation of effector immune cells such as macrophages or mast cells. The specific effector functions of a given Ig molecule are largely determined by the constant regions of its isotype, which controls multimerization, secretion to specific body compartments, and binding to immuno- modulatory receptors and other effector proteins.

Ig are made by immune cells known as B cells and can be made in great quantities in response to an immune challenge such as infection or vaccination. As such, humoral immunity mediated by B cells constitute an important pillar of adaptive immunity; the other complementary pillar is cell- mediated immunity mediated by T cells. As a class of immune cells known as lymphocytes, B and T cells form the adaptive immune system and can respond to novel unencountered antigens by

1 2 generating a diverse antigen receptor repertoire, and subsequently proliferating and executing effector functions in response to binding to cognate antigen.

The adaptive immune system is complemented and supported by the innate immune system, which responds to conserved microbial- or damage-associated molecular patterns with germline-encoded receptors. Broadly speaking, the main distinguishing characteristic between adaptive immunity and innate immunity respectively is whether the immune receptor specificity is generated de novo in the somatic genome (section 1.2), or whether the receptors have hard-encoded specificity in the germline genome.

1.2 Antigen receptor diversification

It is an axiom that there is a functionally infinite number of substances or antigens that can be acted upon by the immune system. To meet this challenge, the adaptive immune system must also have a great diversity of receptors that can recognize many of these potential antigens. The discovery of V(D)J recombination by Susumu Tonegawa illustrated that antigen receptor diversity is generated de novo in the somatic genome rather than through evolution in the germline genome [1]. Similarly in other phyla such as avians and jawless fish, primary antigen receptor diversification also occurs through somatic recombination, although through the parallel process of gene conversion (section 1.2.3) [2]. From these data, adaptive immunity appears to be an emergent phenomenon from combinatorial processes that, by necessity, require DNA damage and repair in somatic immune cells. In humans and mice, V(D)J recombination generates B cell receptors (the secreted form being antibodies/Ig) as well as T cell receptors, which define the function of B and T cells respectively in humoral immunity and cell-mediated immunity.

Additionally for B cells, there are related programmed mutagenic phenomena that occur in the secondary immune response, principally in the germinal centre of secondary lymphoid organs, such as somatic hypermutation (SHM, section 1.2.4) and class switch recombination (CSR, section 1.2.5). These secondary responses allow the finetuning of the adaptive immune response by either selecting for higher-affinity antibodies in the case of SHM, or for different classes of antibodies more suited for a particular type of antigen/pathogen or bodily compartment. It certainly seems clear that an understanding of the processes of programmed DNA damage and repair in these above phenomena are important to understanding the basis of adaptive immunity, and in turn these adaptive immune phenomena are excellent models for investigating the myriad DNA repair

3 processes that maintain genomic integrity. Indeed, the evolutionary origin of certain DNA repair proteins seems to coincide with the origin of vertebrate animals and adaptive immunity, underlining the deep interconnectedness of both physiological processes [3], [4].

1.2.1 V(D)J recombination

The B cell receptor (BCR) is constituted of an IgH and an IgL chain (κ or λ) composed of Ig domains; the T cell receptor (TCR) is similarly constituted of Ig domains with two polypeptide chains, α and β or γ and δ. Both the BCR and TCR contact antigen through an amino-terminal variable domain, which is assembled through V(D)J recombination. Specifically, there are scores of variable gene segments (V), diversity gene segments (D), and joining gene segments (J) in the antigen receptor loci, although the BCR light chain (IgL) loci as well as TCR α and γ loci lack D segments. These gene segments are flanked by recombination signal sequences (RSS) composed of a conserved heptamer and nonamer sequence, which are separated by either a 12 or 23 bp spacer sequence [1]. These RSS are targeted by the endonuclease complex RAG1/RAG2 [5], [6], which nicks 5′ of the heptamer sequence to generate a 3′ hydroxyl group on the coding end of the gene segment [7]. This 3′ hydroxyl can then mediate a nucleophilic attack onto the opposing strand and covalently bind to the phosphoryl group of the directly opposite nucleotide to create a hairpin double-stranded break (DSB) on the coding end. This process also simultaneously generates a blunt DSB on the signal end. This hairpin DSB is then the target of the Artemis endonuclease [8], which can variably nick and open the hairpin end, generating DSB substrates that can be recombined with DSBs created at another RSS. Specifically, a DSB created at one RSS can only recombine with a DSB created at a second RSS of a differing spacer length, conforming to what is known as the 12/23 rule, and this process is enforced by RAG by physical synapsis of 12RSS and 23RSS [9], [10].

This 12/23 rule allows for a programmed, sequential, and predictable recombination and joining of V, D, and J segments to each other, rather than homotypic joins like a V to V segment. Firstly, a random D and J segment are joined in the immunoglobulin heavy chain (Igh) locus in B cells, or TCRβ (Tcrb) in T cells. Subsequently, a random V segment is joined with the combined DJ segment and this combined VDJ exon is expressed with the rest of the immunoglobulin M (Ighm) or immunoglobulin D (Ighd) constant region exons, depending on transcript splicing; a similar V to DJ joining also occurs in T cells at the Tcrb locus. If the VDJ exon was recombined in-frame

4 and without any premature stop codons, then the Ig heavy chain can homodimerize and further associate with two surrogate light chains composed of VpreB and λ5 proteins, forming the pre- BCR; similarly, the TCRβ chains can associate with preTα and form the pre-TCR. At this point, developing pre-B and pre-T cells undergo a selection step that selects for cells that can express a functional receptor at the cell surface. The pre-BCR and pre-TCR can then transduce ligand- independent signals through associated signaling subunits (e.g. Igα/Igβ associates with BCR) that induce cell proliferation and down-regulation and dilution of RAG1/2 such that continued Igh or Tcrb recombination is halted. This process generally prevents more than one functional recombined IgH or TCRβ chain from being expressed and is known as allelic exclusion. Eventually, signals from the pre-BCR/pre-TCR cease due to dilution from cell proliferation and result the re-expression of RAG1/2, allowing for the recombination of the κ and λ light chains (Igk and Igl) or TCRα loci. This proceeds in a similar fashion as before, although V gene segments are apposed directly next to a J segment due to these loci not having D segments. The rearrangement of TCRγ (Tcrg) and TCRδ (Tcrd) proceeds atypically from the canonical hierarchical steps outlined above, chiefly through simultaneous rearrangement temporally with Tcrb and lack of a pre-TCR selection step [11].

To recapitulate, the repertoire diversity of the antigen receptors comes, in part, from the stochastic and combinatorial assembly of V(D)J segments on two different polypeptide chains. However, additional diversity can also arise at the granular level of the DSB. Specifically, when the hairpin end generated by RAG1/2 is opened by Artemis/DNA-PKcs, the scission adds a variable number of “palindromic” or P nucleotides to the DSB end. Moreover, this open DSB can consequently then be the target of error-prone polymerases such as terminal deoxynucleotidyl transferase (TdT) that add non-templated “N” nucleotides [12], [13], or nucleases that remove nucleotides. The junctional variability introduced by P and N nucleotides, in conjunction with the combinatorial diversity of V(D)J segments, accounts for the ability of the adaptive immune system to generate a wide antigen receptor repertoire capable of binding a vast number of potential antigens. These enzymatic processes of nucleolytic resection and polymerase fill-in also happen in the context of DSB repair through the non-homologous end-joining pathway (NHEJ, section 1.3.4) and alternative end-joining pathway (AEJ, section 1.3.5), which both underwrite V(D)J recombination. The importance of end joining repair to V(D)J recombination can be illustrated by the profound lack of Ig, and concomitant immunodeficiency, presented by mice and humans with defects in

Artemis and DNA-PKcs [14]–[17]; similar forms of severe combined immunodeficiency (SCID) can manifest with defects in other DNA repair factors.

1.2.2 Lymphocyte development

Definitive hematopoietic stem cells arise in the mammalian embryo from the aorta-gonad- mesonephros region and populate the hematopoietic compartments of the fetal liver and bone marrow (BM) [18]. From these stem cells, more differentiated precursors arise and in turn give rise to all the cells of the hematopoietic compartment, including B and T lymphocytes. B cell development occurs primarily in the fetal liver early in development, with the BM eventually taking over the role later in development [19]. After birth, the BM is generally the only compartment for lifelong B cell development, although some mammalian species have unique gut- associated lymphoid tissues that supplement or replace the BM as the site of B cell development [2]. B cell development can be divided into stages based on the stepwise genetic rearrangement of the Igh and then Igl loci in B cell precursors (section 1.2.1), and these fractions can be traced by the expression of certain cell-surface markers [20]. Of these fractions, typically fraction D (large pre-B cells undergoing pre-BCR selection and dividing into small pre-B cells) represents the most important developmental checkpoint as defects in receptor signaling or DNA repair will manifest phenotypes at this stage.

T cells similarly derive from hematopoietic stem cells, and pre-T cells with concomitant rearrangement of Tcrb or Tcrd loci can be found in the fetal liver, but T cell development by in large occurs in the thymus. T cell development can be divided into stages based on the step-wise genetic rearrangement of the Tcrb/Tcra loci in T cell precursors, and these fractions can be traced by the expression of certain cell-surface markers. Of these fractions, typically fraction DN3 (pre- T cells undergoing pre-TCR selection) represents the most important developmental checkpoint.

Within the pool of immature B cells (Hardy fraction E) and immature T cells (CD4+ CD8+ double positive cells), negative selection of the antigen receptor repertoire occurs. This is otherwise known as the central tolerance checkpoint. While a broad repertoire capable of recognizing unknown and potentially noxious antigens and pathogens is an important feature of adaptive immunity, the same repertoire would be maladaptive were it to also target endogenous self- antigens. This dilemma necessitates mechanisms to prevent such self-reactive immature lymphocytes from joining the peripheral population. Of the immature B cells that are self-reactive,

6 receptor editing is usually the predominant mechanism of central tolerance. Specifically, immature B cells that can multivalently bind self-antigen will trigger BCR internalization and loss of BCR tonic signaling, the result which is thought to relieve repression of RAG expression. RAG and related V(D)J rearrangement machinery in these cells mediate additional VL-JL rearrangement of Igk/Igl loci until either a non-self reactive BCR is created with a novel IgL chain and pre-existing IgH chain, or until additional IgL rearrangement options are exhausted. B cells in the latter case will either undergo either deletion or anergy where B cells become unresponsive to cognate antigen and migrate out of the BM. Although there have been some reports of receptor editing for T cells, more especially for peripheral T cells, it does not seem to be as important of a diversion mechanism for self-reactive immature T cells; instead, these cells die or become anergic.

These mature lymphocytes that survive development can then migrate to secondary lymphoid tissues, such as the spleen or lymph node, where they may differentiate into different functional subsets. B cell subsets are best defined in mouse models, and their orthologous counterparts in humans is often unclear. It is evident, however, that these B cell subsets vary in terms of the compartments they occupy and their contributions to immunity. One such subset are follicular B cells, which migrate continuously through the blood and B cell areas of lymph nodes and the spleen. These follicular B cells present processed peptides, derived from antigen bound by BCR and internalized, to follicular helper T cells at the border of B cell areas in lymph nodes and spleen. If a T cell recognizes a processed antigen on the surface of a follicular B-cell, it can then activate the B-cell via costimulatory receptors to proliferate and develop into either short-lived antibody- secreting plasma cells or long-lived quiescent memory B cells. Those memory B cells that encounter cognate antigen again in a secondary immune response can activate and proliferate much more quickly. Activated follicular B cells can also enter into the germinal centre of the lymph node to undergo affinity maturation of the BCR (section 1.2.4) or Ig isotype switching (section 1.2.5). Another subset of B cells are marginal zone (MZ) B cells, which are innate-like cells located in the marginal sinus of the spleen. These cells appear to provide quick T cell independent antibody responses to blood borne antigens of microbial origin, and generally have a lower BCR signalling threshold for activation. MZ B cells also have a role in transporting antigen in the form of immune complexes from the marginal sinus to follicular B-cells in the spleen. A third subset of B cells are B-1 B-cells which reside in the pleuro-peritoneal cavities. B-1 B cells are innate-like immune cells,

7 like MZ B-cells, and are capable of T-cell independent activation and antibody responses to microbial antigens.

1.2.3 Gene conversion

While V(D)J recombination is responsible for most of the diversity of the antigen receptor repertoire in mice and humans, alternative mechanisms exist in other jawed vertebrates [2]. One such of the more well characterized examples of this is the avian lineage. Although V(D)J recombination is conserved in birds, it tends to generate a more limited repertoire as the Ig loci do not have much variety in V, D, and J segments. Instead, a diverse cluster of pseudogene V segments upstream of the rearranged IgH and IgL exons provide template DNA for gene conversion to copy and overwrite patches of DNA in the rearranged V(D)J exons; these gene conversion tracts range from a few bp to more than 200 bp [21]. Iterative rounds of copy-and-paste then form the basis of Ig diversity in birds. This process of gene conversion is initiated by the enzyme activation-induced cytidine deaminase (AID), which deaminates deoxycytidine residues into deoxyuridine preferentially in the context of WRCY (W = A or T, R = purine, Y = pyrimidine) motifs in single-stranded DNA, an activity that occurs transiently during transcription [22], [23]. The resultant uracil-guanosine mismatches are then the target of base excision repair (BER, section 1.3.2) and mismatch repair (MMR, section 1.3.3) pathways that result in single-stranded DNA breaks (SSB) [24]. If these SSB are closely apposed and on opposing strands, they can result in double-stranded DNA breaks (DSB). Both SSB and DSB can be acted upon by the gene conversion repair pathway, itself a subset of homologous recombination repair (HR, section 1.3.6) [25]–[28].

Certain other jawed vertebrates such as rabbits, swine, and ruminant mammals also undergo diversification of their BCR repertoire through gene conversion initiated through AID activity, which supports the hypothesis that immunoglobulin gene conversion was a basal trait in jawed vertebrates that was subsequently lost in murine and hominid lineages. Somewhat tangentially, jawless fish (lampreys and hagfish) evolved a convergent antigen receptor system that has deep parallels with the BCR/TCR paradigm in jawed vertebrates. Specifically, the antigen receptors in jawless fish are constituted of leucine rich repeat (LRR) domains, rather than Ig domains, and are known as variable lymphocyte receptors (VLR). These VLR are assembled with a gene conversion mechanism initiated by cytidine deaminases with homology to AID, and this process iteratively copies-and-pastes 10-30 bp tracts from variable pseudogenes upstream and downstream into the

VLR cassette. Moreover, different types of VLRs are recombined at different loci, and expressed by different populations of immune cells that seemingly have deep functional parallels to B and T cells if not actual homology. Thus, gene conversion as a method of antigen receptor repertoire diversification may be more fundamental than currently appreciated, and although not in the scope of the work presented in this manuscript, nevertheless gene conversion-mediated adaptive immunity has been an important model for insights into gene conversion and homologous recombination.

1.2.4 Somatic hypermutation

The primary response by the adaptive immune response to a novel antigen usually lags behind the innate immune response on the scale of days and weeks, in part because activated B and T cells require time to proliferate and differentiate into more specialized effector cell types. B cells and T cells will typically encounter antigen in secondary lymphoid tissue such as lymph nodes, Peyer’s patches, and the spleen—tissues that accumulate antigen from the lymphatic system, gut lumen, and blood respectively. These secondary lymphoid tissues are structured into mostly segregated B cell and T cell zones by chemokine gradients, which guide and regulate B and T cell interaction. B cells in the follicle (B cell zone) that can bind to available antigen will internalize the antigen and become activated, and then upregulate chemokine receptors and migrate towards the T cell zone. At the B and T cell area interface, B cells present peptide antigens, derived from internalized macromolecular antigen, to T cells; if T cells can bind to peptide antigens on the surface of a B cell, the T cell can further activate the B cell through CD40:CD40L interactions. Activated B cells can differentiate into short-lived plasma cells that secrete low-affinity immunoglobulin M (IgM) or into quiescent memory B cells (see also section 1.2.2). Activated B cells can also seed the germinal centre (GC), a specialized niche within the B cell follicle, with light zone (LZ) and dark zone (DZ) microanatomical areas. It is within the GC that affinity maturation of the B cell antigen receptor repertoire occurs.

Affinity maturation is the process where GC B cells compete for limited antigen, with B cells with higher affinity for antigen being able present peptide antigens to follicular T cells and subsequently receive activation signals to clonally expand. Implicit in affinity maturation is the requirement for a diverse repertoire of Ig with a spectrum of affinity for antigen, which is where somatic hypermutation (SHM) plays a central role. Within rapidly dividing DZ B cells, SHM is initiated

9 by the enzyme AID, which deaminates deoxycytidine residues to form deoxyuridine in WRCY motifs in the actively transcribed V region (see also section 1.2.3) [29]. These deoxyuridine residues do not base pair with the opposing deoxyguanidine residues, resulting in lesions that can be repaired by BER and MMR pathways (section 1.3.2, 1.3.3) or mutations if cell division occurs before repair. Briefly, these excision repair pathways cut out the deoxyuridine lesions, generating abasic sites as well as single-stranded tracts from 1-10 bp (BER) as well as longer tracts >10 bp (MMR). These tracts can then be filled in by DNA polymerases. Uniquely in the context of SHM however, the BER and MMR pathways recruit error-prone polymerases that can bypass AID- mediated damage more readily than high-fidelity polymerases—a side effect of this is a much higher rate of mutagenic synthesis. These mutations can therefore generate additional V region diversity, especially those mutations at CDR rather than the flanking framework sequence. B cells migrating from the DZ to the LZ are then subject to the selection and expansion process for high- affinity binders. During a secondary immune response triggered by re-exposure to an antigen, long-lived memory B cells that have undergone affinity maturation will differentiate into plasma cells and secrete high-affinity Ig, allowing for a more rapid and robust humoral response.

1.2.5 Class switch recombination

In addition to SHM, activated B cells in the germinal centre can undergo class switch recombination (CSR) [29]. CSR is a process where, in response to cytokine and receptor signals in the milieu, B cells can change the constant region of their BCR and Ig from (typically) IgM to either IgG, IgE, or IgA, whilst keeping the same antigen-binding variable domain. This has functional consequences in that the various different Ig isotypes have differing effector functions. For example, while IgM antibodies are typically low affinity and characteristic of a primary immune response, IgM monomers can multimerize into pentamers and hexamers and therefore bind to multimeric antigen with higher avidity. Moreover, IgM pentamers are bound by the complement factor C1q, which can trigger the classical complement pathway that releases inflammatory signals to recruit immune cells, as well as triggering the formation of pores on target membranes. B cells do not typically switch to IgD, as it is co-expressed on IgM+ cells by alternative splicing of the VDJ exon with either the IgM or IgD constant regions; however, CSR to IgD can occur at low levels due to the existence of a cryptic switch region between IgM and IgD constant regions [30], and IgD seems to play a role in mucosal immunity as well as modulating IgM signaling [31]. IgG antibodies are most abundant isotypes in serum, and typically are higher

10 affinity antibodies than IgM. IgG can also bind to immuno-modulatory Fcγ receptors on immune cells such as macrophages or natural killer cells to direct activation (e.g. antibody-directed phagocytosis or cytotoxicity) or signaling inhibition. IgG is also transported across the placental barrier by the neonatal Fc receptor to protect the developing fetus with maternal IgG. IgE is bound by the Fcε receptor on mast cells, eosinophils, and basophils, and is the isotype associated with the allergic response towards parasites. Finally, IgA monomers can form dimers that can be transported across epithelial barriers into mucosal compartments such as the gastrointestinal tract, respiratory system, and genitalia, and have important roles in controlling and regulating the microbial ecosystem in those compartments.

Like SHM, CSR is also initiated by AID, but rather than targeting the V region, AID targets switch regions [29]. Switch regions are several kilobases of GC-rich tandem repeats upstream of Igh constant regions [32]. These switch regions are therefore rich in WRCY motifs, and similar to gene conversion and SHM, AID-mediated dU:dG mismatches are targeted by BER and MMR pathways resulting in single-stranded breaks. As AID targets ssDNA, transcription through the switch region is necessary—these sterile or germline transcripts (GLT) start from a promoter upstream of the switch region, transcribe through an I exon, intronic switch region, and downstream constant region, after which the I exon and constant region exons splice together to form the mature RNA transcript. Because of the GC-rich nature of the switch region, the resultant pre-mRNA transcript hybridizes strongly to the transcribed strand, generating a R-loop structure of a double-stranded RNA:DNA hybrid and single-stranded non-transcribed DNA [33]. This structure is relatively long- lived, allowing for AID to target the single-stranded non-transcribed DNA strand. However, AID can target both the transcribed and non-transcribed strands equally, which leaves open the question as to how the transcribed strand may become single-stranded and accessible to AID. Several plausible models exist to explain this, such as R-loop collapse or transient ssDNA resulting from negative supercoiling by the passage of RNA polymerase [34]–[36].

Another longstanding question is how AID is specifically targeted to the switch region during CSR. Although AID can mediate deamination events elsewhere in the genome [37]–[42], with rare consequences such as chromosomal translocations [43], AID-mediated damage occurs at several orders of magnitude more frequently at Ig loci. One clue comes from the fact that AID associates with stalled RNA polymerase II at the switch regions, particularly through the interaction of AID with SPT5 and PAF [44], [45]. Moreover, AID can interact with guanine quadruplex secondary

11 structures on RNA, which occur frequently in G-rich intronic switch RNA—in fact, AID can bind to switch RNA and is guided to switch region DNA in a sequence-dependent manner [46]. These observations highlight the close association between programmed AID-mediated DNA damage and the switch regions. In any event, AID-mediated damage leads to the generation of SSB on both strands of the switch regions. If these SSB are somewhat apposed and on opposing strands, they can result in DSBs, the necessary substrate of CSR [47]. Most of these DSBs manifest as staggered DSBs with overhangs of ssDNA due to the random nature of AID activity in the switch region [47], although further enzymatic processing through nucleases or polymerases may still occur between DSB generation and final end joining [4].

For successful CSR, long-range deletional recombination between DSBs in the upstream donor switch region and the downstream acceptor switch region is required; much like V(D)J recombination, the intervening region between the switch regions is looped out and formed into an excision circle. However, alternative recombinatorial outcomes can also result from the CSR reaction, apart from rare unintended chromosomal translocations. For example, while most switching events occur on the same chromosome (i.e. in cis), around 7% of switching events occur between homologous chromosomes (i.e. in trans) [48], [49]. Moreover, while most CSR outcomes result in long-range deletional recombination between switch regions, a small proportion of recombination events result in the inversion of the intervening region between switch regions [50]. This bias towards productive deletional recombination may reflect topological organization in the Igh locus that allows distal DSBs to synapse and consequently join together. Indeed, there are reports that transcriptional elements in the Igh locus control the co-localization and synapsis of switch regions in response to CSR induction [51], [52], and the NHEJ-associated protein 53BP1 may also play a role in Igh topological organization [50], [53]–[56].

Figure 1. Overview of class switch recombination. (A) B cells in the germinal centre can respond to signals in the local milieu and switch the isotype of their BCR and Ig from IgM to either IgG (4 classes in mice), IgE, or IgA. (B) Schematic showing the induction of damage and ultimate DSBs in the switch regions by AID, and downstream long- range end joining for canonical CSR (deletional recombination) and inversional recombination.

1.3 DNA Damage Response and Repair

1.3.1 DNA damage detection and signaling

All genomes are constantly subjected to many types of DNA damage from both endogenous and exogenous sources, and the damage can present as damage to the nucleobases (e.g. deamination, depurination, oxidation) or damage to the sugar-phosphate backbone (e.g. SSB or DSB). The spectrum of possible insults to DNA necessitates many different enzymes and pathways to repair specific types of damage. Of these types of DNA damage, DSBs are the most genotoxic as DSBs inhibit genomic replication and can result in cell death if left unrepaired; moreover, DSB repair can result in mutagenic insertion, deletion, or translocation events that threaten genomic integrity. DSBs can occur as a result of many different factors, including programmed DNA damaging enzymes RAG1/2 and AID (section 1.2), but also from reactive oxygen species generated by metabolism, DNA replication, or environmental agents such as ionizing radiation. The wide spectrum of DNA damage also requires different attuned pathways for sensing and signaling for repair, but DSB sensing and signaling will be the focus in the context of adaptive immunity.

All the important DSB sensing pathways rely on one of three phosphoinositide-3-kinase-related protein kinases (PIKK) to transduce and amplify the DSB signaling response, which have distinct but often overlapping and complementary domains of function. One such PIKK is the ataxia- telangiectasia mutated (ATM) protein. ATM is recruited by the MRN complex, composed of the nuclease MRE11, DSB-binding RAD50 protein, and the scaffold protein Nijmegen breakage syndrome 1 (NBS1), which rapidly localizes to the site of DSBs [57]. Besides sensing DSBs, the MRN complex plays a role in tethering and synapsis of DSB ends together, as well as processing DSBs for repair; notably, MRE11 exonuclease activity functions only in the 3′→5′ direction, with 5′→3′ nucleolysis performed by downstream nucleases such as CtIP and EXO1. After recruitment by MRN, ATM can then autoactivate by phosphorylation and consequently phosphorylate serine/threonine residues on many proteins relating to DNA repair and cell cycle checkpoint signaling. One critical target is the histone H2AX, a variant of histone H2A. Phosphorylated H2AX (γH2AX) can then recruit mediator of DNA damage checkpoint 1 (MDC1), which can in turn recruit more MRN and ATM, and thereby initiates positive feedback and the recruitment of MRN, ATM, MDC1, and γH2AX emanating up to one megabase away from DSB termini. ATM- phosphorylated MDC1 also initiates recruitment of the E3 ubiquitin ligase RNF8 and a cascade of

14 ubiquitination events that ultimately leads to a histone H2A poly-ubiquitination motif that can recruit tumor suppressor p53-binding protein 1 (53BP1), the master regulator of NHEJ (section 1.3.4) [58]–[60]. ATM phosphorylation of 53BP1 allows for the association of PAX transcription activation domain interacting protein (PTIP) and Rap1 interacting factor (RIF1) [61]–[66], the latter of which seems largely responsible for recruiting effectors responsible for DSB end protection (section 1.3.5). 53BP1 can be removed from DSB sites by BRCA1, the master regulator of HR (section 1.3.6) [63]. Therefore, DSB sensing by MRN and signaling initiated by ATM are important to both major pathways of DSB repair, NHEJ and HR.

Another important sensor of DSB ends is the KU70/KU80 heterodimer (KU), a core factor in NHEJ conserved across all eukaryotic organisms. KU forms a toroidal complex that slides along dsDNA until it finds a DSB, whereupon it likely changes conformation and can recruit the PIKK DNA-PKcs. DNA-PKcs, like ATM, will autoactivate by phosphorylation and phosphorylate many proteins relating to DNA repair, including the core NHEJ factors of KU, XRCC4, XLF, and LIG4, as well as H2AX and 53BP1. These phosphorylation events serve to stabilize protein-protein interactions, especially between core NHEJ factors, but also serve for further downstream recruitment of repair factors (e.g. through the 53BP1 axis). Although KU is typically first at DSB termini, the MRN complex can arrive subsequently and evict KU through the nucleolytic activity of MRE11 [67], [68]; more generally, KU and MRN functions can work at cross-purposes, as KU is associated with NHEJ, whereas MRN activity can instigate nucleolysis and the generation of long tracts of 3′ ssDNA necessary for HR.

The third PIKK relevant to DSB signaling is ATM and RAD3-related (ATR), which is typically recruited to sites of replication stress by ATR interacting protein (ATRIP). Specifically, ATR- ATRIP is recruited to long tracts of ssDNA protected by replication protein A (RPA). Like ATM and DNA-PKcs, ATR has overlapping functionality with ATM and DNA-PKcs, but it likely plays a minor role with respect to programmed somatic recombination in adaptive immunity [69]–[71].

While SSBs are not as genotoxic as DSBs, SSBs occur at a much higher frequency and likewise require robust responses and repair. SSB sensing and signaling is mediated by poly(ADP-ribose) polymerases (PARP), a large family of enzymes that can quickly localize to SSB (and DSBs) and add long chains of ADP-ribose to itself as well as proteins proximal to the break [72]. These PAR modifications can subsequently recruit excision repair and NHEJ factors such as XRCC1 and

PALF [73]–[76]. PARP1 is the most well-characterized and significant PARP, and has been also implicated in AEJ (section 1.3.5) [77]–[79]; indeed, PARP1 and KU compete at DSBs [80]–[85]. PARP1 activity may also affect repair by modifying the chromatin surrounding a break [86].

1.3.2 Base excision repair

Nucleobases can be damaged by oxidation, deamination, and alkylation reactions that, while not necessarily DNA helix distorting, nevertheless require repair to insure genomic fidelity. This is accomplished by base excision repair (BER), a highly conserved pathway amongst prokaryotes and eukaryotes. The primary effectors are DNA glycosylases, which scan DNA for specific damaged bases and remove them, leaving an apurinic/apyrimidinic (AP) site; in the context of deoxyuridine residues generated by AID, UNG glycosylase are primarily responsible for removal in antigen receptor diversification [87], with a minor contribution from SMUG1 [88], [89]. Secondarily, if the glycosylase does not also have AP lyase activity (e.g. UNG), an AP endonuclease cleaves the sugar-phosphate backbone at the AP site, generating an SSB with a single nucleotide gap [90]. Other enzymatic reactions can also occur at the ends that are more complicated and not immediately easily ligated together [91]. From this point, there are two slightly divergent conclusions, short patch and long patch repair. For short patch repair, polymerase β fills in the gap and XRCC1/DNA ligase III seals the sugar-phosphate backbone; if the ends are blocked and not immediately ligatable, long patch repair recruits more processive polymerases that synthesize and displace a flap of ssDNA that can subsequently be cleaved by a flap endonuclease and ligated by DNA ligase I.

1.3.3 Mismatch repair

Like BER, mismatch repair (MMR) is an excision repair pathway conserved in prokaryotes and eukaryotes. However, rather than targeting damaged nucleobases per se, MMR targets helix- distorting base pair mismatches that typically arise from errant DNA replication. Mammalian MMR is first carried out by MutS homologues (MSH) heterodimers, principally either MSH2/MSH6 or MSH2/MSH3, that recognize and bind to single base pair and longer mismatches respectively. These MSH heterodimers have ATPase activity that allows them to slide on DNA and facilitate mismatch excision. MutL homologue (MLH) heterodimers, principally either MLH1/PMS2 or MLH1/MLH3, can bind to the MSH heterodimer; the MLH1/PMS2 heterodimer is known to have DNA nicking activity [92]. However, unlike the paradigm in bacteria, the precise

16 details on how mismatch excision is carried out remains to be fully resolved, although molecular partners such as the 5′→3′ exonuclease I (EXO1) and the replication sliding clamp proliferating cell nuclear antigen (PCNA) are known to participate [93], [94]. The 3′→5′ exonuclease activity of polymerase δ and polymerase ε may also help with excision, in addition to their role in filling the gap left by excision [95]; however in the context of somatic hypermutation (section 1.2.4), non-canonical MMR occurs by the recruitment of error-prone polymerases. Ligation by DNA ligase I after DNA re-synthesis completes MMR.

1.3.4 Non-homologous end joining

Although NHEJ is largely a eukaryotic phenomenon, unlike HR which is found both in prokaryotes and eukaryotes, NHEJ is the predominant DSB repair pathway in vertebrates, especially in post- mitotic cells. The conserved core factors are KU70/KU80, x-ray repair cross-complementation 4 (XRCC4), XRCC4-like factor (XLF), and DNA ligase IV (LIG4), as these proteins are also conserved in yeast. As mentioned previously (section 1.3.1), DSBs are detected by KU, which then recruits DNA-PKcs to phosphorylate repair factors (section 1.3.1); KU also protects DSB ends from nuclease resectioning and immobilizes and holds DSB ends together to aid ligation [96]– [99]. End-processing of the DSB by nucleases (e.g. Artemis, MRN) and polymerases (e.g. TdT, pol λ, pol μ) is often necessary to generate blunt DSBs, which are the preferred substrates for NHEJ [4]. KU also recruits the XRCC4/LIG4 complex responsible for sealing DSBs. The function of XLF is more mysterious, but it seems to stimulate the activity of XRCC4/LIG4 [100]–[102], and XRCC4 and XLF form filaments that can bridge DSBs and aid ligation [103]–[107].

However, accessory NHEJ factors in vertebrates include DNA-PKcs, paralogue of XRCC4 and XLF (PAXX), and 53BP1, which are somewhat dispensable depending on the circumstances—for example, XLF and PAXX knockout mice develop relatively normally compared to the embryonic lethality of Lig4–/– and Xrcc4–/–, and both proteins are not required for V(D)J recombination unless as a double knockout or combined with another deficiency such as ATM [108]–[114]. XLF function in V(D)J recombination also seems to overlap H2AX and 53BP1, as double deficiency results in a more profound block [114]–[116]. In addition, unlike XLF, PAXX is not required for CSR [111]. The differential requirements of V(D)J recombination and CSR for NHEJ factors is also illustrative of certain unique idiosyncrasies. For example, RAG1/2 specifically shuttles the DSBs it generates into NHEJ [117]–[119], likely reflecting the intrinsic ability of RAG1/2 to

17 synapse RSS together [10]. As a result, although some measure of alternative end joining (AEJ) (section 1.3.5) occurs during V(D)J recombination in the context of NHEJ deficiency [118], [119], V(D)J recombination is almost wholly dependent on core NHEJ factors. By contrast, DSBs in CSR are not sequestered by RAG1/2 and therefore are more dependent on accessory NHEJ factors such as XLF and the 53BP1 [108], [120]–[123]; CSR also exhibits a higher frequency of AEJ, in the context of NHEJ deficiency, than V(D)J recombination [124]–[126].

Another layer of regulation on top of the core NHEJ factors is mediated by 53BP1 and its downstream effectors. 53BP1 is a large scaffold protein with no apparent enzymatic activity, but instead seems to mediate its important role in NHEJ through interactions with its protein-binding motifs such as PIKK-phosphorylated serine/threonine residues [127]. In particular, 53BP1 is thought to promote NHEJ through the suppression of nucleolytic resectioning of DSBs and may help synapse DSB ends together through 53BP1 oligomerization [53], [64], [66], [128], [129]. As mentioned in section 1.3.1, 53BP1 is recruited to poly-ubiquitinated H2A on lysine 15 (H2AK15ub) following a DSB signaling response [58]–[60], where it can then recruit PTIP and RIF1 [61]–[66]. PTIP can promote CSR by instigating the chromatin mark H3K4me3 and corresponding upregulated transcription at switch regions, but PTIP also has as yet ill-defined function at DSB termini in inhibiting resection [130]–[132]. However, RIF1 seems to be mostly responsible for mediating the suppression of resectioning by recruiting the Shieldin complex composed of REV7, SHLD1, SHLD2, and SHLD3 [3], [133]–[139]. In particular, SHLD2 has three tandem oligonucleotide/oligosaccharide‐binding (OB) folds that bind to ssDNA and protect it from nucleolytic resection [135], [137]. Corresponding to their importance in NHEJ, 53BP1 and shieldin deficiency deeply impair CSR [3], [122], [123], [135], [137], [138]. Interestingly, 53BP1 and shieldin deficiency does not much affect V(D)J recombination except in the most distal end- joins or unless combined with XLF deficiency [133], [140], [141], reflecting again the idiosyncratic relationship between V(D)J recombination and NHEJ. Besides suppression DNA resection, 53BP1 may also aid NHEJ by juxtaposing DSBs together through 53BP1 oligomerization and/or influencing genomic topology, which may be more significant for long- range end joining such as in CSR or distal V(D)J recombination [50], [53]–[56], [142]. The role of the shieldin complex is also examined in Chapter 3.

Figure 2. Overview of factors and signaling in NHEJ While not exhaustive, this cartoon shows both the core NHEJ complex as well as the signal cascade that leads to the ultimate recruitment of the shieldin complex.

1.3.5 Alternative end joining

In the absence of NHEJ, some residual end joining can still be detected in both yeast and mammalian cells [143], [144]. These observations led to the idea of an error-prone alternative end joining pathway (AEJ) that can repair DSBs that cannot be immediately repaired by NHEJ. However, because AEJ is still largely defined by end joining not mediated by NHEJ, it remains unclear as to whether AEJ is a single unitary pathway, a collection of separate processes, or an artefactual phenomenon arising from specific genetic models. Several candidate factors in AEJ have been identified such as DNA ligase I, ligase III, PARP1, MRE11, CtIP, and polymerase θ (pol θ) [78], [80], [145]–[160]. In many instances, AEJ seems to incorporate “microhomology” in the repair junction—essentially short tracts (2-20 bp) of similar sequence proximal to both DSB ends that are exposed by nucleases (e.g. MRE11, CtIP) and brought together in a sequence- dependent manner. With respect to the latter, pol θ is thought to be the causative factor in annealing 3′ overhangs and tethering DSB ends together in AEJ [149], [150], [161]–[164]. Bolstering the idea that AEJ, or at least pol θ mediated end joining, is a bona fide repair pathway, cells and tumours with a compound defect in NHEJ and pol θ or HR and pol θ are more sick and less viable than individual pathway deficiencies [148], [149], [152], [165]. Additionally, pol θ seems to compete with HR factors RPA and RAD51 [152], [161], and putative AEJ factors PARP1 and MRE11 compete with the NHEJ factor KU [67], [68], [80]–[85].

1.3.6 Homologous recombination

Homologous recombination (HR) is a templated and synthesis dependent DSB repair pathway conserved in prokaryotes and eukaryotes. The process of HR first starts with the generation of long 3′ ssDNA overhangs through the activity of nucleases, initially with some limited resection by MRE11 and CtIP, and then more processive 5′→3′ resection through EXO1 and associated helicases [166]–[169]. These 3′ ssDNA tails are first coated and protected by RPA, which is then replaced by RAD51 (with the help of RAD52), the primary factor in homology search for a template to mediate repair. The RAD51 filaments then initiates strand invasion by displacing the non-template strand and annealing to the template strand in a sequence-dependent manner followed by DNA synthesis. Subsequently according to the DSBR model of HR, the second 3′ ssDNA end is captured and templated synthesis and ligation fills in the gap, forming two crossed- over intermediates known as Holliday junctions [170]. This intermediate is then separated either by being dissolved by helicases or resolved by endonucleases that recognize Holliday junctions.

An alternative model known as synthesis-dependent strand annealing (SDSA) posits that after strand invasion and extension, the newly synthesized strand is displaced from the template and anneals back to the other 3′ ssDNA tail, avoiding a crossed-over intermediate. Gene conversion is a subtype of HR and involves the unidirectional copying and replacing from template to the locus being repaired. Gene conversion can occur between sister chromatids and homologous chromosomes, but in the context of Ig diversification, it typically occurs between homologous V region or LRR segments (section 1.2.3).

1.4 Prokaryotic adaptive immunity

CRISPR (clustered regularly interspaced short palindromic repeats) – Cas systems are a diverse set of adaptive immune systems in prokaryotes against bacteriophages and other mobile genetic elements [171]. These CRISPR-Cas systems work by recognizing DNA or RNA from these parasitic elements in a sequence-dependent manner, as a result of past exposure, and interfering with these elements typically by nucleolytic cleavage. While an extensive review of prokaryotic adaptive immunity is not within the scope of this thesis, CRISPR-Cas systems have been adapted biotechnologically. Specifically, Cas9 from type II systems has been adapted to easily target double-stranded DNA in genomes and mediate cleavage for the purposes of genome-editing [172]. Within the work presented here, CRISPR-Cas has been leveraged to generate defined DNA damage substrates to study downstream repair (Chapter 2) as well as genetic knockouts in cell lines and mice; however, novel applications of CRISPR-Cas systems will doubtlessly aid future research programs.

1.5 CSR and NHEJ repair as mutually reinforcing paradigms of inquiry

Aside from meiotic recombination, adaptive immunity is the only known set of phenomena to incorporate programmed DNA damage and concomitant repair. V(D)J recombination and CSR are the two phenomena within adaptive immunity where NHEJ is relevant, and as V(D)J recombination often includes exceptions with respect to NHEJ repair, CSR is typically the physiological system used for insight into NHEJ more broadly. Advancements in understanding in both domains often further insight into the other. Thus modestly put, this thesis endeavours to investigate the granular, molecular, and mechanistic details of CSR on the structural level of DNA (Chapter 2) and factors effectuating repair (Chapter 3).

Chapter 2 DOUBLE-STRANDED DNA BREAK POLARITY SKEWS REPAIR PATHWAY CHOICE DURING CLASS-SWITCH RECOMBINATION

Excerpted and adapted from: Double-stranded DNA break polarity skews repair pathway choice during intrachromosomal and interchromosomal recombination. Alexanda K Ling, Clare C So, Michael X Le, Audrey Y Chen, Lisa Hung, Alberto Martin. Proc Natl Acad Sci U S A, 2018 Mar 13;115(11):2800-2805.

21 22

Abstract

Activation-induced cytidine deaminase (AID) inflicts DNA damage at immunoglobulin genes to initiate class switch recombination and chromosomal translocations. However, the DNA lesions formed during these processes retain an element of randomness, and hence, knowledge of the relationship between specific DNA lesions and AID-mediated processes remains incomplete. To identify necessary and sufficient DNA lesions in class switch recombination (CSR), the Cas9 endonuclease and nickase variants were employed to program DNA lesions at a higher degree of predictability than achievable with conventional induction of CSR. Here we show that Cas9- mediated nicks separated by up to 250 nucleotides on opposite strands can mediate CSR. Staggered double stranded breaks (DSBs) result in more end resection and junctional microhomology than blunt DSBs. These data indicate that DSB with 5′ overhangs skew intra- and interchromosomal end-joining towards AEJ. In addition to lending potential insight to AID-mediated phenomena, this work has broader carryover implications in DNA repair and lymphomagenesis.

2.1 Introduction

During the germinal center reaction, AID catalyzes DNA lesions at the immunoglobulin heavy chain (Igh) locus that ultimately become substrates for class switch recombination (CSR) [29]. AID deaminates deoxycytidines to produce deoxyuridines at repetitive sequences upstream of immunoglobulin constant regions known as switch regions [22], [23], which are engaged by base excision repair (BER) and mismatch repair (MMR) pathways to generate nicks or longer gaps [24]. These nicks and gaps on opposite DNA strands give rise to double-stranded DNA breaks (DSBs) in donor and acceptor switch regions [47], which are then synapsed and ligated together by nonhomologous end-joining (NHEJ) or alternative end-joining (AEJ) [53], [124], [173], [174], thereby completing CSR and introducing the expression of a new immunoglobulin isotype.

There are two fundamental and longstanding problems associated with studying the molecular mechanism of AID-dependent recombination events. First, switch regions are composed of long, repetitive stretches of AID hotspot sequences, making it difficult to precisely know the cleavage sites and insertional and/or deletional events prior to final ligation. Second, because AID initiates nicks on both strands of switch regions [87], [175], the resulting DSB intermediates represent a spectrum of polarities (i.e. blunt, 5′, or 3′ overhang) and varying lengths of overhanging single- stranded DNA. Furthermore, it is unknown how far apart nicks on opposite DNA strands can be

23 while still resolving as a DSB, or which types of DSBs are necessary, sufficient, or preferred for CSR and chromosomal translocations. Moreover, it is unknown whether NHEJ and AEJ favour repair of a specific subset of these lesions and what conditions lead to the usage of one pathway over the other.

While there exist model systems that examine CSR by providing transgenic substrates [129], [176]–[178], these systems are limited in recapitulating the spectrum of DNA lesions initiated by AID. In this report, we apply CRISPR-Cas9 technology to generate specific DNA lesions at AID target loci in order to model AID-induced events in vivo and characterize DSB repair factors that promote or inhibit CSR. We demonstrate that DSB polarity influences the frequency of CSR and microhomology usage at recombination junctions.

2.2 Results

2.2.1 Cas9- and nickase-mediated DSBs induce CSR

To characterize the specific DNA lesions that are sufficient to induce CSR, we used CRISPR-Cas9 gene editing to generate specific DSBs in the CH12F3-2 mouse B cell line (hereafter referred to as CH12 cells). These Cas9-generated DNA lesions mimic DSBs that arise downstream of AID- mediated deamination and base excision. We designed single guide RNAs (sgRNAs) targeting upstream of switch region μ (S′μ) and downstream of switch region α (S′α) to avoid the repetitive elements within the switch regions, while still targeting regions of AID-mediated CSR [32], [179]. These regions have been shown to be physiological sites of AID activity during CSR [175], [179], [180]. To eliminate any contributions by AID to lesions in S′μ, we used Aicda (Aid)–/– CH12 cells [181]. Aid–/– CH12 cells transiently transfected with S′μ and S′α sgRNA undergo Cas9-mediated switching from IgM to IgA, or to other isotypes by changing the acceptor switch region sgRNA (Figure 3A, S1A). Cas9-mediated switching to IgA can be induced in different mouse B cell lines (Appendix Figure 1B,C). Multiple DSBs in the S′μ and S′α region increased Cas9-mediated switching to IgA (Appendix Figure 1D), supporting a report that a multiplicity of breaks within switch regions increases CSR [178].

To determine whether the Cas9-mediated switching can recapitulate other properties of CSR, we assessed two other phenomena associated with AID-mediated CSR. First, since approximately 7% of IgA CSR events occur between homologous chromosomes in trans [48], [49], we examined

24 whether Cas9-mediated switching can likewise occur in trans. The native KpnI and StuI restriction sites in the 3′ UTR of Igha on the same chromosome with the rearranged V(D)J exon were “scarred” with Cas9 to generate a cell line where only the unrearranged homologous Igha allele maintained these restriction sites (Figure 3B). After inducing Cas9-mediated switching in these clones and performing reverse transcription PCR of the Igha 3′ UTR followed by restriction digest, we observed that ~2% of Cas9-mediated switching occurred in trans (Figure 3B, Appendix Figure 1E). Second, although AID-mediated CSR preferentially favours deletional recombination of DSBs in donor and acceptor switch regions leading to productive immunoglobulin expression, recombination may also instead lead to inversion of the intervening region resulting in loss of immunoglobulin expression (Figure 3C, Appendix Figure 1F) [50]. In contrast to AID-mediated CSR (Figure 3D), we found that Cas9-mediated switching had a similar proportion of Iglo (i.e. inversional recombination) to IgA+ (i.e. deletional recombination) cells (Figure 3E), in seeming agreement with a previous report examining I-SceI-mediated switching [50]. We tried to confirm that Iglo cells in Cas9-mediated switching truly represented inversional recombination by PCR amplification of the predicted inversional junction using primers 5′ of S′μ and S′α sgRNA target sites. We saw a stronger PCR signal corresponding to the predicted inversional junction in sorted Iglo compared to IgA+ cells (Appendix Figure 1G); however, this PCR-based assay may not accurately assess the true frequency of inversional recombination (Chapter 3). Together, these data suggest that Cas9-mediated switching recapitulates some properties of AID-mediated CSR.

Figure 3. Cas9-mediated DSBs can recapitulate CSR and associated recombination. (A) Percentage of switched Aid−/− CH12 cells when transfected with S′μ1 sgRNA only or in combination with a sgRNA targeting the relevant acceptor switch region. (B, Left) Schematic of two Igh alleles present in CH12 cells. The rearranged allele (VDJ) has been edited using CRISPR- Cas9 to lack KpnI and StuI restriction sites in the 3′ UTR downstream of Cα exon 3 (the “scarred” allele), while the unrearranged allele (WT) retains these restriction sites. (B, Right) To detect trans CSR, KpnI and StuI sites in the 3′ UTR of Igha were scarred with Cas9 on the V(D)J-rearranged chromosome. After inducing CSR and amplifying the 3′ UTR of IgA from cDNA, restriction digestion was performed, followed by electrophoresis and band intensity quantitation (Appendix Figure 1E). (C) Schematics showing the unswitched Igh locus (Top), the deletional CSR that leads to IgA expression (Middle), and the “inversional” CSR that leads to loss of Ig expression (Bottom). (D) Percentage of WT CH12 cells that have undergone IgA+ CSR, or “inversion” of the

Sμ–Sα intervening region (Iglo) at 48 h after stimulation with CIT (αCD-40, IL-4, and TGF-β) (Appendix Figure 1F,G). (E) Percentage of Aid−/− CH12 cells that have undergone deletional recombination (IgA+) or inversion of the S′μ-S′α region (Iglo). ΔCμ is a sgRNA that targets Cμ exon 1, leading to loss of IgM expression, and serves as a positive control for Ig loss. (F) Schematic of Igh locus and sgRNA target regions in S′μ and S′α. The distance between the Cas9 cleavage site in sense- and antisense-targeting sgRNAs in nucleotides is denoted. (G) Percentage of Aid−/− CH12 cells that have undergone CSR to IgA on transfection with two pairs of sgRNAs and Cas9D10A to generate 5′ DSBs in S′μ and S′α. Combinations of sgRNAs used in each experiment are denoted by colored blocks below and correspond to the sgRNAs depicted in A. (H) Same as G, except that one or two sgRNAs have been removed from the transfection. (I) Same as G, except that CasN863A was used to generate 3′ DSBs in S′μ and S′α. (J) Ratio of IgA-switched Aid−/− CH12 cells transfected with Cas9D10A to those transfected with Cas9N863A, with the sgRNA combinations used held constant. pX330, empty vector control. Error bars represent standard deviation.

2.2.2 CSR has a preference for 5′ DSBs over 3′ DSBs

AID deaminates dC to dU, which is then engaged by UNG to generate an abasic site that is cleaved by AP endonuclease to induce a nick in the DNA (section 1.3.2). To model AID-mediated nicks in vitro, we used the Cas9 RuvC-null (D10A) or the HNH-null (N863A) nickases together with a pair of proximal sgRNAs to create precise staggered DSBs with a 5′ (5′ DSB) or 3′ overhang (3′ DSB), respectively [182]–[184]. To test whether Cas9-induced 5′ DSBs can mediate switching, we transfected two S′μ and two S′α sgRNA-Cas9D10A vectors simultaneously into Aid–/– CH12 cells to produce 5′ DSBs with various overhang lengths (Figure 3F). S′μ and S′α overhangs do not share significant sequence similarity with each other. 5′ DSBs induced switching up to 3% (Figure 3G), a 3-fold lower frequency than that achieved with wild-type Cas9 (Figure 3A), perhaps reflecting the preference for a blunt DSB over a staggered DSB during CSR. Surprisingly, a pair of sgRNAs creating a 248 nt 5′ overhang was sufficient for Cas9-induced switching (Figure 3G). Removal of any one sgRNA from transfection greatly diminished Cas9-induced switching (Figure 3H), suggesting that paired opposing nicks indeed resolve as staggered DSBs to complete the reaction. Similarly, targeting sgRNA to only one strand, in mimicry of the processive catalysis of AID in vitro [185], yielded little switching (Appendix Figure 2). 3′ DSBs created by Cas9N863A can also mediate switching (Figure 3I). Although Cas9N863A has similar enzymatic activity to Cas9D10A [186], we observed a 5-fold reduction in Cas9-mediated switching using Cas9N863A compared to Cas9D10A while keeping sgRNA targets and overhang lengths constant in S′μ and S′α (35 and 24 nt, respectively) (Figure 3J). As the overhang length is increased, 5′ DSBs persist as better substrates for Cas9-mediated switching to IgA than 3′ DSBs (Figure 3J). These results suggest the role of a 5′ to 3′ exo- or endonuclease that converts distal nicks into staggered DSBs (Appendix Figure 3A; Discussion). We conclude that staggered DSBs of different polarities and with nicks separated by as much as ~250 nts can induce switching, although 5′ DSBs are preferred over 3′ DSBs.

2.2.3 Staggered DSBs promote alternative end-joining during CSR

DSBs created during CSR are joined by NHEJ and the AEJ pathways, the latter of which is often characterized by repair junctions with increased resection and microhomology usage compared to junctions repaired by NHEJ [129]. Since AID initiates staggered DSBs [47], we hypothesized that DSBs with different end polarities and different lengths of ssDNA overhang may be resolved by different DNA repair pathways in order to complete CSR. Hence, we sequenced individual Cas9

28 and nickase-mediated S′μ-S′α junctions from Aid–/– CH12 cells to quantify resection and microhomology usage. Blunt DSBs generated by Cas9 produced junctions with low levels of resection (Figure 4A, Appendix Figure 3B). Using the same sgRNA pairs to create a 35 and 24 nt overhang in S′μ and S′α, respectively, S′μ-S′α junctions arising from 5′ DSB intermediates exhibited a higher median resection than 3′ DSBs, although the distributions of resection were not statistically different (Figure 4A, Appendix Figure 3C). The same trends were observed when the overhang length in S′μ was increased from 35 nt to 98 nt (Figure 4A, Appendix Figure 3C). Both 5′ and 3′ DSBs, regardless of overhang length, were resected significantly more than blunt DSBs during Cas9-mediated switching.

We next measured the amount of microhomology usage in S′μ-S′α junctions. 5′ and 3′ DSBs, regardless of overhang length, led to junctions with increased microhomology compared to those arising from blunt DSBs, suggesting the predominance of AEJ in repair of staggered DSBs (Figure 4B). We observed a positive correlation between resection and microhomology usage in S′μ-S′α junctions arising from blunt DSBs in Aid–/– CH12 cells, suggesting that blunt DSBs in NHEJ- proficient cells can also produce junctions that are characteristic of AEJ (Figure 4C), although to a lesser degree than that achieved with 5′ DSBs or 3′ DSBs.

Our data thus far suggests that 5′ and 3′ DSBs bias end-joining towards AEJ during Cas9-mediated switching as compared to blunt DSBs. To determine whether 5′ and 3′ DSBs are indeed preferential substrates for AEJ compared to blunt DSBs, we examined Cas9-mediated switching using blunt, 5′, and 3′ DSBs in Aid–/– CH12 cells deficient for DNA ligase IV (LIG4), a core component of NHEJ (Appendix Figure 4, section 1.3.4) [125], [187]. We reasoned that if the disparity in μ–α resection and microhomology between junctions from blunt versus 5′ or 3′ DSBs in Aid–/– cells was no longer observed in Aid–/– Lig4–/– cells, then repair of 5′ and 3′ DSBs in NHEJ-proficient cells may truly be driven by AEJ. Resection could not previously be measured from AID-mediated CSR junctions due to the random targeting of switch region deoxycytidines by AID. Resection at μ–α switch junctions was increased in LIG4-deficient cells across blunt, 5′, and 3′ DSB intermediates (Figure 4D, Appendix Figure 3D). As expected, microhomology usage at μ–α switch junctions increased across blunt, 5′, and 3′ DSB intermediates (Figure 4E), consistent with increased microhomology usage at μ–α switch junctions in AID-initiated CSR in Lig4–/– CH12 cells [125], [181]. Importantly, we no longer observed the disparity in μ–α resection and microhomology between junctions from blunt versus 5′ or 3′ DSBs in Aid–/– Lig4–/– cells.

Collectively, these data suggest that 5′ and 3′ DSBs skew end-joining towards AEJ during Cas9- mediated switching in contrast to blunt DSBs.

Figure 4. The μ–α junctions derived from 5′ DSB intermediates have increased resection. (A) Total resection of S′μ and S′α (Top) and microhomology use (Bottom) at μ–α junctions derived from Aid−/− CH12 cells transfected with various sgRNA combinations and Cas9 nickase variants. The median resection length is denoted by a black line, and the blue line denotes the sum of overhang lengths generated by Cas9D10A and CasN863A in the S′μ and S′α. (B) Scatterplot of microhomology use from junctions derived from different DSB intermediates, with mean microhomology indicated. (C) Spearman’s correlation between total resection and microhomology use in μ–α junctions induced by blunt junctions in Aid−/− CH12 cells. (D) Same as A except that μ–α junctions were derived from Aid−/− Lig4−/− CH12 cells. (E) Same as B except that μ–α junctions were derived from Aid−/− Lig4−/− CH12 cells. ***P < 0.001. ns, not significant.

2.3 Discussion

In this report, we used CRISPR-Cas9 to study the DNA lesions required for CSR. By modeling these AID-dependent recombination events via Cas9-induced blunt, 5′, and 3′ DSB intermediates, we found that 5′ DSBs induced Cas9-mediated switching at greater frequencies than 3′ DSBs. 5′ DSBs functioned as better substrates for Cas9-mediated switching than 3′ DSBs as overhang length was increased. Importantly, the Cas9 nickase-induced overhangs used in this study may more accurately mimic those generated in vivo by AID.

These results also suggest that a high density of AID-induced mutations at switch regions is not required for CSR, especially if distal nicks are oriented to lead to 5′ DSBs. The tolerance for distal nicks as precursors to staggered DSBs are complemented by a previous finding that nicks 900 nucleotides apart can induce homologous recombination (HR) [188]. The preference for 5′ over 3′ DSBs as substrates for Cas9-mediated switching also implicates a 5′ to 3′ exonuclease or endonuclease to facilitate the generation of a DNA end (Appendix Figure 3A), since nicks separated by long distances may not necessarily spontaneously “melt” into a staggered DSB. This preference for staggered DSBs as substrates for AEJ is supported by a previous report demonstrating that knockdown of AID leading to reduced deamination and lower density of nicks in the switch regions is correlated with enriched AEJ activity [179]. Mechanistically, we speculate that overhangs would be a poorer substrate for the KU complex and therefore predispose to AEJ; indeed, cell cycle regulator of NHEJ (CYREN) can bind to KU at DSBs with ssDNA overhangs and inhibit NHEJ [189].

Our results also establish a role for DSB end polarity in determining DNA repair pathway choice, in addition to known cell cycle influences [63], [190]. The DNA damage response to different prescribed DNA lesions has been investigated in vitro [191], [192] and most recently in vivo in yeast [193]. We illustrate that 5′ and 3′ DSBs undergo increased resection relative to blunt DSBs and lead to recombination with increased microhomology. It is not immediately clear why 3′ DSBs are poorer substrates for CSR compared to 5′ DSBs, especially as DSBs can change end polarity by the nuclease processing before final ligation. Interestingly, this aversion for 3′ DSBs is not limited to NHEJ-associated phenomena, as Cas9 nickase-generated 3′ DSBs induce HR less frequently than 5′ DSBs in a GFP reporter assay [188]. These results suggest that end-joining pathway choice is influenced by DSB polarity.

The varied degree of nucleolytic resection observed at μ–α junctions derived from 5′ DSBs suggests different simultaneous modes of resection prior to ligation. For example, resection equal to the sum total of the partner overhangs suggests the removal of the overhangs by a flap endonuclease, yielding a blunt DSB for ligation. In contrast, resection less than the sum total of the partner overhangs implicates filling in of the ssDNA overhang by a polymerase prior to ligation. Resection beyond the sum total of the 5′ DSB overhangs suggests the conversion of 5′ DSBs into 3′ DSBs prior to repair. Our results suggest that 5′ DSBs in wild-type cells are repaired by all three of these mechanisms leading to CSR or chromosomal translocations, while complete resection of the overhang and beyond dominates repair of 5′ DSBs in NHEJ-deficient cells. While our model system effectively captures the initial substrates and final state of end-joining, future experiments are required to observe the likely iterative process of resection and filling prior to final ligation.

This study examines molecular characteristics of end-joining from only productive recombination events, i.e. successful Cas9-mediated switching or chromosomal translocations. Non-productive repair outcomes at a single target locus such as indel formation or faithful repair by NHEJ or HR are also likely outcomes and are the focus of previous studies [186], [194], [195]. HR factors such as BRCA1 and RAD52 have been shown to inhibit CSR [196], [197]. It is possible that HR- mediated repair may compete for DSB ends at the expense of CSR. Indeed, RAD52 was shown to compete with KU70/80 for binding to a blunt Sμ DNA probe in an electrophoretic mobility shift assay [197]. Since our study selectively analyzed DNA repair leading to productive CSR, we did not investigate the role of HR in inhibiting CSR. It would be an interesting follow-up study to determine how specific DSB ends are preferential substrates for CSR versus HR.

We acknowledge that Cas9 does not recapitulate all aspects of the role(s) AID plays in CSR. DSB formation in vivo requires AID-mediated deoxycytidine deamination followed by uracil excision by UNG and nicking of the phosphodiester backbone by AP endonuclease [87], [90]. Therefore, our model examines DNA repair leading to CSR at least two enzymatic steps downstream of AID activity. Additionally, AID activity is regulated by the cell cycle [198], [199], whereas Cas9 activity in our model system is not cell cycle restricted. Moreover, it is possible that AID might act as a DNA repair factor scaffold as previously suggested [200], which Cas9 is unlikely to do. Future work along these lines could proceed by substituting the Cas9 nickases used in this study

33 with catalytically-dead Cas9 fused to a cytidine deaminase [201], [202], and/or catalytically inactivating AID to preserve any scaffolding function.

We also recognize that there may be slight differences in enzymatic activity between wild-type Cas9, RuvC-null, and HNH-null Cas9 that could affect our results. Cas9D10A generated indels with paired sgRNAs at a similar frequency to wild-type Cas9 with either individual sgRNA, although indel formation is a poor proxy for cutting efficiency [183]. While RuvC-null and HNH- null Cas9 were previously demonstrated to have similar nickase activity [182], [186], Cas9 was shown to release target and nontarget strands asymmetrically and may potentially bias accessibility of repair in a strand-dependent fashion [203]. Deciphering the potential idiosyncratic influences of nucleases on modelling AID-mediated phenomena will require additional work, as well as improved understanding and implementation of CRISPR technology.

In conclusion, our study delineates the changing DNA damage response to varying DSB structures and highlights the ability of DSB structure to bias end-joining between NHEJ and AEJ pathways. Altogether, a greater understanding of how diverse DNA lesions are differentially repaired will improve our understanding of the mechanism of physiological processes required for adaptive immunity and protective mechanisms that maintain genomic integrity from diverse threats.

2.4 Experimental Methods

Cell culture and transfection All murine B cell lines were cultured in RPMI 1640 supplemented with 2 mM L-glutamine, 10% FBS, 5% NCTC-109, 0.5 mM β-mercaptoethanol, and penicillin/streptomycin and cultured at 37 °C/5% CO2. CH12 cells were generally electroporated with 4-5 µg of each relevant plasmid, unless otherwise indicated, in 4 mm cuvettes with an exponential wave at 325 V, 975 µF, and ∞ Ω in a Bio-Rad GenePulser Xcell. Cells were harvested 3 days post-transient transfection for CSR and translocation analysis.

CRISPR-Cas9 vector construction All sgRNA used are summarised in Appendix Table 1. sgRNAs were cloned into CRISPR-Cas9 vectors as previously described [204]. All sgRNAs used in this study were chosen based on favorable predicted on- and off-target activity [205]. SpCas9 bearing an N863A mutation was

34 obtained from Feng Zhang (MIT) and substituted into pX330 to generate a Cas9N863A CRISPR vector.

CRISPR-mediated gene knockout in CH12 cells Aid–/– Lig4–/– CH12 clones 1 and 2 were generated by electroporating Aid–/– CH12 cells, provided by Kefei Yu (MSU) with 4 ug of Lig4_G2 (Appendix Table 1) targeting exon 2 of murine Lig4 and expanded in complete RPMI media. Individual clones were generated by single cell dilution in 96w tissue culture plates. Genomic DNA was harvested from single clones 7 days post-plating by proteinase K digest (20 μg/mL) at 55 °C for 1 hour followed by heat inactivation at 95 °C for 15 minutes. Mutant clones were screened using the mismatch cleavage assay, as described before [195]. Candidate clones exhibiting mismatch cleavage products were verified by quantitative PCR and sequencing at The Centre for Applied Genomics (TCAG) (Toronto, Ontario, Canada).

Flow cytometry and fluorescence activated cell sorting (FACS) CH12 cells were typically stained 3-7 days post electroporation with relevant antibodies (IgA-PE, Southern Biotech cat. 1040-09; IgG1-PE, BD Biosciences clone A85-1; IgG2a-PE, Southern Biotech cat. 1080-09S; IgG2b-PE, Southern Biotech cat. 1090-09S; IgG3-FITC, BD Biosciences clone R40-82; IgE-FITC, BD Biosciences clone R35-72) and acquired on a FACSCalibur. Cells were sorted on a BD FACSAria IIu (Faculty of Medicine Flow Cytometry Facility, University of Toronto).

Molecular cloning of junction and variable Ig sequences S′µ-S′α switch junctions were amplified using primers S′µ_seq_F2 and S′α_seq_R2 (Appendix Table 1) and Q5 polymerase (NEB), then A-tailed using Taq polymerase. All amplified sequences were gel purified, cloned into pGEM-T Easy (Promega), and sequenced at The Centre for Applied Genomics (Toronto, ON, Canada).

Inversional and trans recombination screening Inversional recombination following CSR was detected with S′µ_seq_F2 and S′α_1.1 (Appendix Table 1). With respect to trans CSR, Aid–/– CH12 were transfected either with Igha3UTR_R2.3 or Igha3UTR_R3.10-pX330 to ablate StuI and KpnI sites in the 3′ UTR of IgA. Heterozygous clones were then transfected with single S′µ and S′α CRISPR constructs to induce switching, and RNA was harvested 72 hrs post-electroporation. cDNA synthesis was performed with 1 µg total RNA with Maxima H Minus reverse transcriptase (Thermo Fisher) according to manufacturer’s

35 instructions, and the 3′ UTR of IgA was amplified from diluted cDNA and column purified. Subsequently, 1 µg of PCR was digested with 10 units of relevant restriction enzyme (New England Biolabs), and electrophoresed restriction fragments were quantified with Quantity One software (Bio-Rad). Heterozygote clones were sorted on the basis of digest efficiency as either retaining a restriction site in cis with the rearranged V(D)J exon or retaining the restriction site in trans on the homologous chromosome. Only the latter clones were included in the trans CSR analysis.

Statistics Unless otherwise indicated in the figure legend, all data were analyzed using the Mann-Whitney test using GraphPad Prism 6. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. ns = not significant. All error bars represent standard deviations.

Chapter 3 THE SHIELDIN COMPLEX PROMOTES CLASS SWITCH RECOMBINATION

Excerpted and adapted from: SHLD2 promotes class switch recombination by preventing inactivating deletions within the Igh locus. Alexanda K Ling, Meagan Munro, Natasha Chaudhary, Conglei Li, Maribel Berru, Brendan Wu, Daniel Durocher, Alberto Martin. EMBO Reports, 2020 Jun 17;21(8):e49823.

Abstract

Adaptive immunity requires programmed DNA damage and concomitant repair to generate antigen receptor diversity and fine-tune the immune response. These phenomena are also well- suited for studying the various DNA repair pathways responsible for maintaining genome integrity. In particular, the humoral response mediated by B cells requires the non-homologous end joining (NHEJ) repair pathway for V(D)J recombination and class switch recombination (CSR) of the immunoglobulin loci. The shieldin complex, composed of SHLD1, SHLD2, SHLD3 and REV7, lies downstream of 53BP1-RIF1 regulated double-stranded break (DSB) DNA repair and acts to inhibit DNA resection. Thus, similar to deficiency in 53BP1, deficiency in shieldin leads to a strong defect in AID-mediated CSR in the CH12F3-2 B cell line. Here we show that Shld2−/− mice have defective CSR. Lymphocyte development and RAG1/2 mediated recombination were unaffected in Shld2−/− mice and CH12 cells. Interestingly, a significant fraction of Shld2−/− primary B cells and 53BP1-, shieldin- and other NHEJ-deficient CH12 cells exhibit a low-to-no expression of immunoglobulin upon induction of CSR. In CH12 cells, these Ig-low cells do not recover Ig expression after induction of CSR suggesting a permanent suppression of Ig expression. This loss of Ig expression is not due to inversional recombination of the Igh locus as previously reported for 53bp1–/– B cells. Rather, we find that Ig-low 53bp1−/− CH12 cells have undergone conventional deletional recombination coupled with hyper-resection into the IgA constant region, consistent with a role of the 53BP1-RIF1-shieldin axis in inhibiting DNA resection. Collectively, these data show that SHLD2 is critical for CSR in vivo by inhibiting DNA end resection and thereby promoting NHEJ and CSR.

3.1 Introduction

Across multiple phyla, adaptive immunity requires programmed DNA damage and concomitant repair to generate antigen receptor diversity and fine-tune the immune response. These phenomena are also particularly well-suited for studying the various DNA repair pathways responsible for maintaining genome integrity. In particular, the humoral response mediated by B cells requires the non-homologous end-joining (NHEJ) repair pathway for V(D)J recombination and class switch recombination (CSR) of the immunoglobulin (Ig) loci. Activation-induced cytidine deaminase (AID) plays a central role in humoral immunity by inducing somatic hypermutation (SHM) and CSR that function to increase the antibody affinity and alter the antibody isotype respectively [29].

AID accomplishes these processes by deaminating dC to produce dU within the immunoglobulin DNA encoding the variable and switch regions. At this point both processes diverge: damage produced by AID at the variable region is engaged by base excision repair and mismatch repair pathways leading to small mutations that may potentially increase the affinity of the antibody to antigen. By contrast, damage produced by AID at the switch µ and downstream switch regions (Sγ, Sε, and Sα) is engaged by the same DNA repair pathways, with the distinction that these lesions are converted to double-stranded DNA breaks (DSBs), followed by synapsis and ligation of distant ends through NHEJ or the poorly defined alternative end joining (AEJ). Although it is unclear what specific molecular events govern the distinct outcomes that lead to SHM versus CSR, it is possible that the switch regions have evolved to facilitate DSB formation through their repetitive sequence structure, or their likelihood of forming secondary structures such as guanine quadruplexes or R-loops.

The joining reaction during CSR requires the NHEJ pathway. One member, 53BP1, plays a key role in CSR by inhibiting DNA end resection [53], [129], [196], thereby shuttling the repair of DSBs towards NHEJ instead of the homologous recombination (HR) pathway. Downstream of 53BP1 lies RIF1 that further facilitates DNA end-protection and is necessary for CSR [62]–[66]. The recently identified shieldin complex, that lies immediately downstream of the 53BP1-RIF1 axis, is necessary for CSR, NHEJ, and telomere protection [3], [133]–[139]. Shieldin is composed of REV7, SHLD1, SHLD2, and SHLD3. SHLD2 has three OB-fold domains that bind to single- stranded DNA, much like RPA1 and POT1 proteins, and appears to be the factor proximally responsible for the end-protection role ascribed to the 53BP1 pathway [135], [137]. Some shieldin components have been demonstrated to promote CSR in B cell lines [3], [135], and 53BP1/RIF1/REV7 deficient mice have also been observed to have profoundly impaired CSR [64], [66], [122], [123], [134], [136]. By contrast, the importance of SHLD1, SHLD2, and SHLD3 on CSR in mice has not been tested. In this report, we show that SHLD2-deficiency impairs CSR in mice but has no impact on early B cell development and V(D)J recombination. We further show that Shld2−/− primary B cells, as well as 53BP1-, shieldin- and other NHEJ-deficient CH12F3-2 cells (hereafter referred to as CH12 cells) exhibit a significant population with low to no expression of Ig upon induction of CSR. This effect is permanent as cells do not recover Ig expression, and this Ig-low population does not appear to be the result of inversional recombination, even in the context of 53BP1 deficiency. Instead, our analysis shows that a large proportion of these Ig-low

39 cells have undergone CSR to IgA, however with major deletions of the Ighm and Igha loci that lead to loss of Cα constant region exons. These data suggest that 53BP1- or shieldin-deficiency does not lead to reduced recombination at the DNA level per se, but rather to a relative increase in non-productive CSR that inactivates the Igh locus. These results are consistent with the role of the Shieldin complex and 53BP1 in suppressing resection of DNA.

3.2 Results

3.2.1 Lymphocyte development and B cell populations are largely unaffected by Shld2 deficiency

Shld2 knockout mice (Shld2em_del/em_del, referred hereafter as Shld2–/–) were generated by microinjection of Cas9 ribonucleoprotein complexes targeting exon 4 in the C57BL/6 background, and a founder mouse with a large out-of-frame deletion (248 bp) was generated (Figure 5A). In Shld2−/− mice, there was no apparent block in B and T cell development in the bone marrow and thymus respectively, particularly at the Hardy fraction C or DN3 populations corresponding to the pre-BCR and pre-TCR selection stages (Figure 5B,E, Appendix Figure 5A,D). Moreover, marginal zone and follicular B cells in the spleen, as well as B1 cells in the peritoneal cavity, were unaffected by SHLD2-deficiency (Figure 5C,D, Appendix Figure 5B,C). This apparent lack of a defect in lymphocyte development suggests that V(D)J recombination mediated by the RAG1/2 recombinase is unaffected by SHLD2-deficiency. To test this notion, we transduced A70.2 INV-4 cell line with CRISPR-Cas9 lentivirus targeting the 53bp1, Shld1, Shld2, Shld3, and Lig4 genes. In this cell line, imatinib-induced RAG-mediated recombination of a genomically-integrated artificial substrate results in GFP expression [206]. In the bulk edited A70.2 cells, despite similar indel penetrance using the various CRISPR constructs (Appendix Figure 5E), the only defect in GFP expression was in cells transduced with Lig4-targeting sgRNA (Figure 5F). Hence, SHLD- deficiency does not impact V(D)J recombination, consistent with what was observed in the Mb1cre/+ Rev7fl/fl mice [136].

Figure 5. SHLD2 does not affect lymphocyte development or V(D)J recombination.

(A) Schematic of the Shld2 exon 4 showing the 3 sgRNAs used to produce a 251 base pair deletion in one founder line that leads to a frameshift and usage of a premature stop codon. (B) Characterization of the various B cell progenitor fractions in the bone marrow of 4 wildtype and 4 Shld2−/− littermate controls. Gating strategy for measuring these populations is shown in Appendix Figure 5A. (C) Characterization of the indicated immature and mature splenic B cell populations. Gating strategy for measuring these populations is shown in Appendix Figure 5B. (D) Characterization of the indicated peritoneal B cell populations. Gating strategy for measuring these populations is shown in Appendix Figure 5C. (E) Characterization of the indicated thymic T cell populations. Gating strategy for measuring these populations is shown in Figure S1D. (F) Left panel: Schematic of the A70.2 INV-4 cell line strategy to induce RAG1/2-mediated recombination using imatinib. Right panel: A70.2 INV-4 cell lines were transduced with lentiviruses encoding the lentiCRISPRv2 expressing sgRNAs against the 53bp1, Shld1, Shld2, Shld3, and Lig4 genes. Guide RNA targeting chicken AID was used as a negative control (Ctrl). Cells were selected with puromycin and treated with 3 μM imatinib for 4 days after which GFP frequency was measured (mean ± SD of 3 biological replicates). The insertion-deletion (indel) penetrance as measured by TIDE analysis of sequence for each of these sgRNA constructs is shown in Appendix Figure 5E, and the baseline GFP frequency prior to imatinib stimulation is shown in Appendix Figure 5F. sgRNA sequences used are shown in Appendix Table 3.

3.2.2 SHLD2 is necessary for CSR

To determine whether SHLD2 functions in CSR in vivo, we first examined the steady-state levels of serum Ig isotypes. We found that the levels of IgM in the serum of unimmunized Shld2−/− mice were normal, but IgG2b and IgG3 isotypes had reduced concentrations relative to wildtype (Figure 6A). Interestingly, IgA levels seemed to be elevated in Shld2−/− animals, perhaps pointing to a reduced dependence of IgA CSR on NHEJ as previously reported [207]. To further test the role of SHLD2 in CSR, we purified splenic B cells and induced them to switch to various isotypes using different stimulation cocktails. We found that ex vivo CSR to all tested Ig isotypes show a clear impairment of Shld2−/− B cells relative to WT (Figure 6B), and this impairment was not due to a decrease in AID protein or sterile transcript expression (Appendix Figure 6A,B). In addition, using an experimental system in which class switching is mediated by Cas9 [208], we found that neither 53BP1- nor SHLD2-deficiency affected this Cas9-mediated CSR. (Appendix Figure 8). This finding is similar to previous results involving 53BP1 deficiency and I-SceI mediated CSR [129], pointing to a unique relationship between 53BP1/Shieldin function and AID-mediated DNA damage.

To examine the effect of SHLD2 on antigen-specific CSR, mice were immunized with 4-Hydroxy- 3-nitrophenylacetyl Chicken Gamma Globulin (NP-CGG). Shld2−/− mice had reduced NP-specific serum of the IgG1, IgG2a, IgG2b, and IgG3 classes relative to WT, as well as a trend to increased NP-specific IgM (Figure 6C). In some cases, the NP-specific serum Ig defect in Shld2−/− mice was intermediate between WT and 53bp1−/− animals. In addition, Shld2−/− splenic NP-specific IgG1 secreting cells were ~7-fold reduced compared to WT, and congruent with 53bp1−/− results (Figure 6D). Moreover, there was no apparent difference in splenic germinal center B cell frequency before or after NP-CGG immunization, suggesting that the Shld2−/− CSR defect is on the molecular level of end-joining (Appendix Figure 7). All together, these data show a critical function of Shld2 in CSR in mice, supporting previous findings in the CH12 B cell line [3], [135], [136], [138].

A ns

200 ns 250 400 300 2000 L

ns L L

L ** **

m m

m 180 300 /

/ 1500

m g

200 200 /



 

160 ] 200 

] 1000

]

3 1

150 2 100

A G

140 G 100 G

g 500

[

[ [ 120 100 [ 0 0 0 /- /- /- /- - T - T - T - T - -/ 2 2 2 T W d W 2 W d W d 2 l ld l l W ld h h h h h S S S S S

B IgG1 IgG3 IgG2b IgE IgA

15 3 2.5 3 1.5

s l

l **** s

l **** **** **** ****

l e

2.0 l

c 10 2 1.0

2 c

+ 1.5

E A

G 1.0

g g

5 1 G 1 0.5

I 0.5

% % 0 0 % 0.0 0 0.0

T O T O T O T O T O W K W K W K W K W K

2.0 2.5 3 4 -NP IgG1 -NP IgG2a -NP IgG2b C -NP30 IgM 30 30 30 2.0 1.5 3

m 2 n

1.5 0

0 1.0 2 4 1.0

s 1 b

A 0.5 1 0.5

0.0 0.0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 /1 /3 /6 2 4 8 6 2 /1 /3 /6 2 4 8 6 2 /1 /3 /6 2 4 8 6 2 /1 /3 /6 2 4 8 6 2 1 1 1 /1 /2 /4 /9 9 1 1 1 /1 /2 /4 /9 9 1 1 1 /1 /2 /4 /9 9 1 1 1 /1 /2 /4 /9 9 1 1 1 1 /1 1 1 1 1 /1 1 1 1 1 /1 1 1 1 1 /1 1 1 1 1

4 1.0 1.5 -NP IgG3 -NP IgE -NP IgA 30 30 30 53bp1-/- 0.8 3

m 1.0 -/- n

0.6 Shld2 0

0 2 4

0.4 WT s

b 0.5

A 1 0.2

0 0.0 0.0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 /1 /3 /6 2 4 8 6 2 /1 /3 /6 2 4 8 6 2 /1 /3 /6 2 4 8 6 2 1 1 1 /1 /2 /4 /9 9 1 1 1 /1 /2 /4 /9 9 1 1 1 /1 /2 /4 /9 9 1 1 1 1 /1 1 1 1 1 /1 1 1 1 1 /1

1 1 1

l l

D e 400

g * **

o i

i 300

e 200

G e

g 100

N 0 - - -

 / / T - - 2 1 W d p l b h 3 S 5

Figure 6. SHLD2-deficient mice have defects in class switch recombination. (A) Concentration of the various indicated isotypes in the serum of 6-8 week old unimmunized WT and Shld2−/− mice. Values are mean concentration ± SD of 4 biological replicates; ** P ≤ 0.01, unpaired two-tailed t-test. (B) B cells were purified from spleens from WT and Shld2−/− mice, and stimulated to undergo CSR to the various indicated isotypes using different stimulation cocktails. Cells were then analyzed by flow cytometry for expression of the various indicated isotypes and the percent expression of each isotype are reported. Values are mean frequency ± SD of 4 biological replicates; **** P ≤ 0.0001, unpaired two-tailed t-test. (C) WT, 53bp1−/−, and Shld2−/− mice were immunized with NP-CGG and the serum was withdrawn 2 weeks post immunization and serial dilutions were subjected to ELISA analysis for NP-specific antibodies of the indicated isotypes. Values are mean absorbance ± SD of 4 biological replicates. (D) WT, 53bp1−/−, and Shld2−/− mice were immunized with NP-CGG, spleens were isolated and anti-IgG secreting cells were enumerated by the ELISPOT assay. Values are mean frequency ± SD of 4 biological replicates, except for Shld2–/– (n=3); * P ≤ 0.05, ** P ≤ 0.01, unpaired two-tailed t test.

3.2.3 The shieldin complex and other NHEJ factors exhibit an Iglo population upon CSR

In carrying out the ex vivo CSR of WT and Shld2−/− B cells, we observed that Shld2−/− B cells stimulated with LPS and IL4 showed an increased proportion of IgMlo IgG1lo cells relative to WT cells at day 6 post stimulation (Figure 7A). This phenomenon was also recapitulated in 53bp1−/−, Shld1−/−, Shld2−/−/−, and Shld3−/− CH12 B cells at three days after stimulating with the CIT cocktail (anti-CD40, IL4, TGFβ) (Figure 7B). We also found that CH12 cells deficient in other NHEJ factors such as KU70, KU80, DNA-PKcs, XLF, XRCC4, and LIG4 (but not PAXX) have reduced CSR to IgA and an increased Iglo population (Figure 7B, C).

The presence of this Iglo population was temporary and was reduced by 9 days post stimulation in ex vivo B cells (Figure 7A) and 7 days post CIT stimulation in 53bp1−/−, Shld2−/−/−, and Shld3−/− CH12 B cells (Figure 8A), as well as in CH12 cells deficient in SHLD1, KU70, KU80, XLF, XRCC4, and LIG4 (Appendix Figure 9A). DSB formation has been shown to transiently reduce expression of a gene, a process mediated by ATM [209]. Hence, cells deficient in NHEJ factors might have DSBs that persist longer than in WT cells, and thus impose a longer-lasting decrease in Ig expression. To test whether this Iglo population was exhibiting a transient decrease in Ig expression or was being diluted from the population due to increased death and/or decreased proliferation, WT and mutant CH12 cells were stimulated with CIT for 3 days, and IgM+ IgA–, IgM– IgA+, and IgMlo IgAlo populations from WT, 53bp1−/−, Shld2−/−/−, and Shld3−/− CH12 cells were sorted out and re-cultured for 5 or 12 days. All three populations from all CH12 genotypes largely maintained their Ig expression phenotype at the point of sorting (Figure 8B, Appendix Figure 9B). These data suggest that CSR induces a permanent reduced Ig expression in 53bp1−/− and Shldnull CH12 cells.

Figure 7. Shld2-deficient B cells and B cells deficient in other NHEJ factors exhibit an Iglo population upon CSR induction. (A) WT and Shld2−/− B cells were purified from spleens and stimulated with LPS + IL4 and examined for IgM and IgG1 expression 3, 6, and 9 days post-stimulation by flow cytometry. Representative plots are shown for both WT and Shld2−/− B cells 6-days post stimulation. The graph plots show proportion of IgG1+ and Iglo cells, mean ± SD from 6 biological replicates; ** P ≤ 0.01, *** P ≤ 0.001, two-way ANOVA with post hoc Dunnett’s test. (B) WT CH12 cells, as well as two each of 53BP1-, SHLD1-, SHLD2-, and SHLD3-deficient clones generated previously, as well as a LIG4-deficient CH12 clone were subjected to CSR induction with the CIT cocktail and measured for both IgM and IgA expression by flow cytometry. Representative flow plots for WT and 53bp1−/− CH12 cells are shown at day 3 post CIT stimulation. The graph plots show proportion

47 of IgA+ and Iglo CH12 cells, mean ± SD from 3 biological replicates; * P ≤ 0.05, **** P ≤ 0.0001, two-way ANOVA with post hoc Dunnett’s test. The letters below the x-axis represent the clone codes for the 53BP1, SHLD1, SHLD2, and SHLD3-deficient CH12 clones. One LIG4-deficient CH12 clone was also used. (C) The Xlf, Ku70, Ku80, Xrcc4, and Paxx genes were knocked out in CH12 cells by CRISPR, and two independent clones each were analyzed as in Figure 7B with mean ± 3 biological replicates; * P ≤ 0.05, **** P ≤ 0.0001, two-way ANOVA with post hoc Dunnett’s test.

Figure 8. Reduced Ig expression in 53BP1, SHLD2, and SHLD3-deficient CH12 cells is permanent and dependent on CSR. (A) WT, and two independent clones each of 53bp1−/−, Shld2−/−/−, and Shld3−/− CH12 cells were stimulated with CIT and analyzed by flow cytometry for IgM and IgA expression. The Iglo population were reduced after 7 days in culture. Values are mean ± SD from 3 biological replicates; * P ≤ 0.05, ** P ≤ 0.01, **** P ≤ 0.0001, two-way ANOVA with post hoc Dunnett’s test. (B) WT, and two independent clones each of 53bp1−/−, Shld2−/−/−, and Shld3−/− CH12 clones were stimulated with CIT for 3 days. The IgM+, IgA+, and Iglo populations were sorted and reanalyzed for expression of IgM and IgA 5 days post sort as well as 12 days post sort (see Appendix Figure 10B). Shown on bar graphs are sorted IgM+, IgA+, and Iglo populations (each column, 1 technical replicate) from WT and mutant CH12 clones, and the percent of cells expressing IgM, IgA, or low for both isotypes (Iglo) after 5 days of culture post-sort. A representative flow plot of 53bp1−/− CH12 3 days post CIT treatment is shown.

3.2.4 The Iglo population is not the result of increased inversional recombination

The finding that CSR-induction permanently induces generates an Iglo population in 53bp1−/− and Shldnull CH12 cells suggests that the Igh locus has been inactivated leading to loss of IgH expression. One possibility that can explain this observation is the increased inversional recombination observed in 53bp1−/− cells during CSR [50]. Inversional recombination during CSR is predicted to lead to loss of Ig expression (Figure 1). To assess whether the Iglo population may be the result of inversional recombination, a variation of digestion-circularization (DC) PCR was performed. Genomic DNA was digested with EcoRI and circularized by dilution and self-ligation. Subsequently, deletional (canonical CSR) as well as inversional recombination at the Igh locus was probed by digital droplet PCR (DDPCR) with primers corresponding to circularized deletional or inversional fragments, and compared to an internal processing control at Chrnb1. Reassuringly, the DC-DDPCR assay demonstrated reduced deletional recombination events in 53bp1−/−, Shld2−/−/−, and Shld3−/− CH12 cells (Figure 9A), mirroring the effects observed by flow cytometry (Figure 7B). Similarly, deletional recombination events were reduced in Shld2–/– ex vivo B cells stimulated to switch to IgG1 and IgA, as compared to WT (Figure 9C,D). However, inversional recombination events were extremely rare (< 1% frequency) in both stimulated CH12 and ex vivo B cells (Figure 9B,C,D). Importantly, inversional events are not detected in unstimulated cells (Appendix Figure 11A), and the primers used to detect inversional DC products do not amplify non-specifically (Appendix Figure 11B-E), which suggests that the DC-DDPCR assay is a reliable detector of inversional recombination.

Figure 9. Inversional recombination in CH12 and ex vivo B cells are rare events. (A) CH12 cells and derivative clones mutated for 53bp1–/– (TA, T1), Shld2–/–/– (F2, F3), and Shld3– /– (C1, C3) were stimulated with α-CD40, IL4, and TGFβ for 3 days and genomic DNA was digested with EcoRI, circularized, and assayed with ddPCR for deletional recombination (i.e. canonical CSR) events. (B) Like (A) but assayed with ddPCR for inversional recombination events. (C) Splenic B cells were stimulated with IL4 for 6 days and assayed for IgG1 deletional and inversional recombination by ddPCR. (D) Splenic B cells were stimulated with IL4, anti-IgD dextran, TGFβ, and IL5 and assayed for IgA deletional and inversional recombination by ddPCR. **** P ≤ 0.0001, two-way ANOVA with post hoc Dunnett’s test.

3.2.5 CSR in 53bp1–/– and Shld2–/–/– CH12 cells leads to aberrant recombination involving deletions in the acceptor constant region

The finding that CSR-induction generates a persistent Iglo population in 53bp1−/− and Shldnull CH12 cells suggests that the Igh locus has been inactivated. One possible explanation for this phenomenon is that 53bp1−/− and other NHEJ-deficient CH12 cells exhibited increased DNA resection during CSR, leading to loss of sequences that allow for surface expression of IgH. To test this notion, Iglo cells from WT, 53bp1−/−, and Shld2−/−/– CH12 cells were sorted, and subcloned, and expression for IgM and IgA was re-assessed by flow cytometry to confirm that Ig expression remained negative or low (Appendix Figure 12). To test whether expression of IgM or IgA was affected at the mRNA level, we used primers that amplified the entire coding region of the IgM and IgA transcript from the leader sequence to stop codon. For the WT Iglo subclones, we found that most expressed IgA at the mRNA level, however, 50% (4/8) of tested subclones only expressed “truncated” (~0.7 kb) IgA cDNAs relative to the full-length IgA mRNA (~1.6 kb); the positive control also has a truncated IgA cDNA band likely corresponding to a mis-spliced product (Figure 10A). For the 53bp1−/− Iglo subclones, 72% (18/25) of tested subclones only expressed truncated IgA cDNAs (~0.7 or 1 kb); a small fraction of subclones expressed neither IgM or IgA cDNA, and at least one clone (D01) appeared oligoclonal by expressing both IgM and IgA transcripts (Figure 10A). Likewise, Shld2−/−/– Iglo subclones also expressed truncated or no IgA transcripts, as compared with Shld2−/−/– IgA+ subclones which expressed the full-length IgA transcript (Figure 10B).

To determine the nature of these truncated IgA cDNAs, we sequenced the IgA amplicons for select WT Iglo CH12 subclones. The analysis showed that these subclones expressed IgA mRNAs that were variably missing exons Cα1, Cα2, and Cα3 (Figure 10C). These data suggest that many of the Iglo clones in WT, 53bp1−/−, and Shld2−/−/– CH12 cells had developed deletions during CSR that either deleted or affected splicing of IgA constant region Cα exons to the variable region VDJ exon.

To confirm these deletions at the DNA level, we extracted DNA from these subclones and assayed for Ighm and Igha deletions on the assumption that most of the Iglo subclones had undergone recombination. Using a long-range PCR assay, we attempted to amplify a region between the Ighm Iμ exon and the Igha M exon downstream of Cα3, with the maximal size of the amplicon expected to be at ~12kb (Figure 11A). Almost all the subclones, both Iglo and IgA+, yielded long-range

52 amplicons and confirmed that Iglo cells have undergone recombination (Figure 11B,C). In many cases, there was more than one amplicon from each subclone, which is likely due to recombination of both the productive and non-productive (without the rearranged VDJ exon) Igh alleles; alternatively but not mutually exclusively, some of the subclones may have been or become oligoclonal at the point of or after subcloning. Moreover, many of the amplicons from 53bp1−/− and Shld2–/–/– Iglo subclones were shorter than amplicons from Shld2–/–/– IgA+ subclones, suggesting that the Iglo amplicons had large deletions within the IgH region (Figure 11B,C). To assess the extent of these deletions, some of these long-range amplicons were sequenced (Figure 11D). Most of the sequenced amplicons had deletions of Cα exons, although 12% had no deletion of coding sequence, which may be due to the capture of both productive and non-productive Igh alleles. Interestingly, only 6% of the Ighm-Igha junctions were direct joins, with the other junctions either incorporating microhomology or sequence insertion (Appendix Figure 13), suggesting that many of these junctions are the product of alternative end-joining. All together, these data show that 53BP1- and SHLD2-deficiency leads to aberrant CSR that involves loss of coding sequence in the acceptor constant region. These data thereby provide an explanation for the increased loss of Ig-expression in NHEJ-deficient B cells undergoing CSR, which supports the role of the 53BP1- RIF1-shieldin axis in inhibiting DNA resection.

Figure 10. Loss of Ig cell-surface expression in CH12 cells is accompanied by aberrant IgA transcripts. (A) mRNA expression analysis of IgM and IgA was carried out by RT-PCR for Iglo subclones from WT and two independent 53bp1−/− CH12 (TA and T1) clones. The Iglo subclones derived from the 53bp1−/− CH12 TA clone are listed as D01-D18, while the Iglo subclones derived from the 53bp1−/− CH12 T1 clone are listed as F03-F24. Gels show the RT-PCR analysis for IgM cDNA (top), IgA cDNA (middle), and AID cDNA (bottom) as control. Appendix Figure 12 shows the expression of IgA and IgM by flow cytometry for each of these subclones. (B) Representative IgA cDNA sequence analysis from WT Iglo clones. (C) mRNA expression analysis as in A of Iglo

54 subclones of two independent Shld2–/–/– (F2 and F3) clones, and IgA+ subclones from Shld2–/–/– clone F3. The Iglo subclones derived from the Shld2–/–/– CH12 F2 clone are listed as H03-H40, while the Iglo subclones derived from the Shld2–/–/–CH12 F3 clone are listed as J02-J109.

Figure 11. Loss of Ig expression in CH12 cells is accompanied by large deletions within the IgA constant region. (A) Schematic of the Ighm and Igha loci juxtaposed by their respective switch regions, along with primer binding sites for long-range PCR. (B) Long-range PCR of Iglo subclones from WT and two independent 53bp1–/– CH12 (TA and T1) clones. The Iglo subclones derived from the 53bp1−/− CH12 TA clone are listed as D01-D18, while the Iglo subclones derived from the 53bp1−/− CH12

T1 clone are listed as F03-F24. Gels show long-range amplicons (top) and AID genomic DNA PCR control (bottom). (C) Long-range PCR as in B of Iglo subclones of two independent Shld2–/– /– (F2 and F3) clones (left 2 panels), and IgA+ subclones from Shld2–/–/– clone F3. (D) Sequence analysis of a representative subset of long range-PCR amplicons. Symbol annotations: (*) inverted sequence, (◊) rearranged fragment order, (†) amplified with the EμF primer rather than IμF, (‡) undamaged Cα coding sequence.

3.3 Discussion

The recently characterized shieldin complex composed of SHLD1, SHLD2, SHLD3, and REV7 suppresses end-resection during DSB repair that would otherwise attenuate NHEJ and CSR. However, most of the characterization of the shieldin complex has been accomplished in cell lines, with the exception of REV7 [136]. In this report, we generated Shld2−/− mice and report that SHLD2-deficiency largely recapitulated 53BP1-deficiency in terms of a defect in CSR and largely normal lymphocyte development and V(D)J recombination [122], [123]. This work then establishes the shieldin complex as an essential factor in the CSR process in vivo.

We observed that murine Shld2−/− B cells, in addition to CH12 cells deficient in 53BP1, SHLD2, and other NHEJ repair factors, exhibited an increase in an Iglo population during CSR induction. Two initial possibilities struck us as potential explanatory mechanisms: first, these cells may be temporarily decreasing Ig expression due to unrepaired DNA damage in the switch regions, a process that might be regulated by ATM [209]. Alternatively, CSR induced an increase in non- productive recombination events in SHLD2- and other NHEJ-deficient B cells [50], [210]. Indeed, we observed that most Iglo clones from 53BP1-deficient CH12 cells were the product of non- functional CSR into the Igha locus, leading to the production of transcripts that were missing IgA constant region exons. These recombination events also occur in WT CH12 cells, but because the frequency of Iglo cells is reduced in these cells, the frequency of such non-productive CSR events is minimal in WT B cells. Interestingly, the aggregate frequency IgA+ and Iglo populations together in WT, SHLD2-deficient, and other NHEJ-deficient B cells is strikingly similar (Figure 7B,C). This observation suggests that recombination at the DNA level per se is not necessarily reduced in 53BP1- and SHLD2-deficient cells, but a larger proportion of such recombination events are non-functional in repair-deficient cells due to excessive resectioning and loss of coding sequence. Additionally, these non-functional recombination events may be biased towards alternative end- joining as a last-ditch repair attempt (Appendix Figure 13).

Inversional recombination has previously been observed to be increased in 53bp1−/− B cells during CSR. Inversional recombination during CSR is predicted to lead to loss of Ig expression as constant region coding sequences downstream of variable region VDJ exon would be inverted and incapable of being translated. Curiously, however, we find that Iglo clones from 53BP1-deficient CH12 cells are largely due to loss of IgA constant region sequences during CSR (Figure 11), and

58 not due to inversional recombination (Figure 9). These data suggest that inversional recombination may occur at frequencies lower than previously reported. Nevertheless, we conclude that the shieldin complex promotes CSR by inhibiting excessive DNA end resection, and the presence of a significant Iglo population during CSR may be a useful proxy indicator for increased DNA resection in the Igh locus.

3.4 Experimental Methods Mouse generation and husbandry Shld2−/− mice were generated by injecting Cas9 ribonucleoprotein complexes and single guide RNA(s) with spacer sequences of GTCCACTAGTCATATCACTC, TCTTTGGAAGTTCCGAACGC and TCACGATGTCCTGTCGGCTC targeting ENSMUSE00000618046 into C57Bl/6 embryos, resulting in a 251-bp del Chr14:34268371 to 34268621, as well as an insertion of GGA. These mice were backcrossed for several generations to the C57Bl/6 background. Shld2+/− mice were bred with Shld2+/− to generate Shld2−/− and WT littermates for experimental use (sex-matched; ~6-8 weeks old). 53bp1+/− mice were bred with 53bp1+/− to generate 53bp1−/− and WT littermates for experimental use (sex-matched; ~ 8 weeks old). All mice were maintained under pathogen-free conditions. The experimental procedures were approved by the Animal Care Committee of University of Toronto.

Spleen, bone marrow, thymus, and peritoneal cavity profiling Single-cell suspensions of spleen, bone marrow, thymus, and peritoneal cavity was prepared from Shld2−/− or WT littermate mice (6-8 weeks old). The cells were resuspended in staining buffer (PBS + 2% FBS) and incubated with mouse Fc blocker (2.4G2 mAb). As previously described [207], splenocytes were stained with rat anti-CD45R/B220 APC (RA3–6B2; Southern Biotech; 1/150 dilution), rat anti-IgM eFluor450 (II/41; eBioscience; 1/150 dilution), rat anti-CD93 PE (AA4.1; eBioscience; 1/50 dilution), rat anti-CD23 PerCP-eFluor710 (B3B4; eBioscience; 1/100 dilution), and rat anti-CD21/CD35 PE-Cy7 (8D9; eBioscience; 1/100 dilution); BM cells were stained with anti-CD45R/B220 APC, anti-IgM eFluor450, rat anti-IgD PerCP-eFluor710 (11-26c; eBioscience; 1/100 dilution), rat anti-CD43 PE-Cy7 (S7; BD Pharmingen; 1/150 dilution), rat anti- CD24 FITC (30-F1; eBioscience; 1/100 dilution), and rat ant- BP-1 PE (6C3; BioLegend; 1/33 dilution); peritoneal cells were stained with anti-CD45R/B220 APC, anti-CD23 PerCP-eFluor710, and rat anti-CD5 PE (53-7.3; eBioscience; 1/100 dilution); thymocytes were stained with anti-

CD25 FITC, anti-CD44 Super Bright 600, anti-CD4 Alexa Fluor 700, anti-CD8α PE, anti-TCRβ APC, and anti-TCRγδ. Stained cells were acquired on an LSR II (BD Biosciences) and analyzed by FlowJo software.

Ex vivo CSR Splenic B cells were purified from Shld2−/− mice (6-8 weeks old), and age- and sex- matched littermate controls, using the EasySep Mouse B-cell Isolation Kit (Stemcell Technologies) and stimulated in vitro as described previously [207]. The stained cells were acquired on an LSR II (BD Biosciences) and analyzed by FlowJo software.

NP-specific ELISA and ELISPOT assay Shld2−/− or WT littermates (6-8 weeks old) and 53bp1−/− mice (10-12 weeks) were intraperitoneally immunized with 100 μg of 1 mg/mL NP20-CGG in PBS (Biosearch) precipitated with an equal volume of Imject Alum (Thermo) according to manufacturer instructions. At day 21 post immunization, sera and spleens were harvested and subjected to ELISA and ELISPOT as previously described [207].

In vitro RAG1/2 and Cas9 induced switching 2x105 puromycin-resistant A70.2 INV-4 cells transduced with lentiCRISPRv2 (Addgene #52961) were stimulated as previously described for four days (34). Aid−/− CH12 cells were electroporated with sgRNA/Cas9 vectors targeting upstream of Sμ and Sα as previously described [181].

Cell culture and CRISPR/Cas9 editing Gene targeting and CH12 cells was performed as described previously [208]. The CH12 parental clone I (referred to above as WT CH12) was used to delete the following genes using CRISPR: Lig4, Xlf, Ku70, Ku80, Xrcc4, and Paxx using sgRNA sequences listed in Appendix Table 3. Verification of gene targeting was accomplished by sequencing. The 53bp1-, Shld1-, Shld2-, and Shld3-deficient CH12 clones were previously generated (23).

PCR and qPCR

RNA was isolated with TRIzol (Thermo Fisher) and cDNA was synthesized with Maxima H Minus reverse transcriptase (Thermo Fisher) according to manufacturer’s instructions. Quantitative PCR was performed with qPCRBIO SyGreen Blue mix (PCR Biosystems) and CFX384 Real-Time PCR Detection System (Bio-Rad) according to manufacturers’ instructions. The primers used for all reactions are listed in Appendix Table 2. Long range PCRs was accomplished using Platinum SuperFi II DNA polymerase (Invitrogen) using the manufacturer’s recommendations.

Digestion-circularization/digital-droplet PCR

Digestion and subsequent circularization was performed with 5 μg genomic DNA was digested overnight with 20 units of EcoRI-HF (New England Biolabs) in 100 μL reaction volume at 37 °C; digested DNA was then diluted to 10 ng/μL in a 500 μL ligation reaction with 800 units of T4 ligase (New England Biolabs) and incubated overnight at 16°C. This digested-circularized DNA was phenol-chloroform extracted and then added to a QX200 ddPCR EvaGreen supermix reaction (Bio-Rad) and emulsified with an AutoDG (Bio-Rad) according to manufacturer’s instructions. The digital-droplet PCR reactions were then cycled on a C1000 Thermal Cycler (Bio-Rad) and analyzed with a QX200 Droplet Reader (Bio-Rad). Chrnb1 DC events were detected with Chrnb1 F2 and Chrnb1 R2 (Appendix Table 2); IgA deletional DC events were detected with IgA_del F2 and IgA_del R2; IgA inversional DC events were detected with IgA_invA F2 and IgA_del R1; IgG1 deletional DC events were detected with IgG1_del F4 and IgG1_del R4; and IgG1 inverrsional DC events were detected with IgG1_invA F4 and IgG1_invA R4.

Statistics All analyses were performed on GraphPad Prism. For Student’s t tests, and two-way analysis of variance (ANOVA), p values of 0.05 or less were considered significant: *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. All error bars represent standard deviations.

Conclusions

At first glance, the mutually reinforcing paradigms of CSR and NHEJ appear straightforward, with most broad strokes of insight already applied on canvas. However, as work presented in the preceding chapters demonstrates, the granular details at the level of DSB structure (Chapter 2) and DSB repair factors (Chapter 3) are of significant import and warranting of continued and further study.

In Chapter 2, I used CRISPR-Cas9 to study the influence of DSB end polarity and overhang length on the repair outcomes during CSR. I found that 5′ DSBs functioned as better substrates for Cas9- mediated switching than 3′ DSBs, especially as overhang length was increased. Moreover, staggered DSBs seem to be resected more by nucleases, as well as incorporate more junctional microhomology, than blunt DSBs—these data suggest that staggered DSBs are more prone to being repaired by AEJ as compared to blunt DSBs. The enzymology underwriting the differential repair outcomes between structurally different DSBs during CSR is not known, and this line of inquiry would be an interesting and natural extension of the work presented in Chapter 2. Moreover, novel CRISPR-Cas9 technologies incorporating base editors may more faithfully mimic DNA lesions generated during CSR and allow for a closer examination of the BER and MMR pathways that lead to DSBs. This latter point would also help disentangle the potential differential activity of the RuvC and HNH domains in Cas9 nickases as well as any idiosyncrasies of Cas9-DNA binding that may influence downstream repair.

In Chapter 3, I characterized the function of the shieldin complex in primary murine B cells and confirmed its importance in CSR, as a profound switching defect follows from shieldin deficiency. Moreover, I confirmed that shieldin is largely dispensable for V(D)J recombination, in line with previous literature on 53BP1 and REV7. I also observed that a significant proportion of NHEJ- deficient B cells, while undergoing CSR, lose expression of Ig. This loss is not due to non- canonical inversion of the Igh locus, as previous literature suggests, but rather completed deletional recombination accompanied with hyper nucleolytic resection and scarring of the downstream acceptor constant region. This excessive resection in the context of shieldin deficiency is likely due to its function in protecting DSB ends; in the context of NHEJ-deficiency more broadly, slow and inefficient long-range end joining likely allows more opportunities for nuclease activity. The phenomenon of Iglo cells is therefore a useful proxy for resection and should be characterized in

62 conjunction with productive switching in future work examining the molecular effectors in CSR. For example, a genome-wide CRISPR-Cas9 screen for novel genes involved in CSR would have candidate hits under-represented in switched cells and also potentially over-represented in Iglo cells, thus providing more statistical power and lowering the frequency of false positive hits. Additionally, more work will be necessary to confirm that this loss of Ig in B cell lines also occurs on the sequence level in primary B cells, although this will likely entail the use of single-cell sequencing or sophisticated high-throughput DSB end-sequencing.

The significance of the work presented here can be seen as a manifestation of the continuing influence of molecular biology on other domains—that is to say, the specific details on the level of DNA and DSB structure, as well as the molecular players effecting repair, matter to the cellular and physiological phenomenon of adaptive immunity. No description of adaptive immunity would be sufficient, for example, without a description of adaptive immune receptor ontogeny and the unusual recombinatorial gymnastics that occur in the genome. Additionally, inasmuch the details of DNA repair inform and contextualize the broader phenomenon of adaptive immunity, the programmed DNA damage and concomitant repair that occurs within adaptive immunity provides excellent model systems for studying DNA repair. Doubtlessly, the synergy between the two fields will continue fruitfully in the future.

References

[1] S. Tonegawa, “Somatic generation of antibody diversity,” Nature, vol. 302, no. 5909, pp. 575–581, Apr. 1983.

[2] M. Hirano, S. Das, P. Guo, and M. D. Cooper, “The Evolution of Adaptive Immunity in Vertebrates,” in Advances in immunology, vol. 109, 2011, pp. 125–157.

[3] R. Gupta et al., “DNA Repair Network Analysis Reveals Shieldin as a Key Regulator of NHEJ and PARP Inhibitor Sensitivity,” Cell, vol. 173, no. 4, pp. 972-988.e23, May 2018.

[4] M. R. Lieber, “The mechanism of double-strand DNA break repair by the nonhomologous DNA end-joining pathway.,” Annu. Rev. Biochem., vol. 79, pp. 181–211, Jan. 2010.

[5] D. G. Schatz, M. A. Oettinger, and D. Baltimore, “The V(D)J recombination activating gene, RAG-1,” Cell, vol. 59, no. 6, pp. 1035–1048, Dec. 1989.

[6] M. Oettinger, D. Schatz, C. Gorka, and D. Baltimore, “RAG-1 and RAG-2, adjacent genes that synergistically activate V(D)J recombination,” Science (80-. )., vol. 248, no. 4962, pp. 1517–1523, Jun. 1990.

[7] J. F. McBlane et al., “Cleavage at a V(D)J recombination signal requires only RAG1 and RAG2 proteins and occurs in two steps,” Cell, vol. 83, no. 3, pp. 387–395, Nov. 1995.

[8] Y. Ma, U. Pannicke, K. Schwarz, and M. R. Lieber, “Hairpin opening and overhang processing by an Artemis/DNA-dependent protein kinase complex in nonhomologous end joining and V(D)J recombination.,” Cell, vol. 108, no. 6, pp. 781–94, Mar. 2002.

[9] D. C. van Gent, D. A. Ramsden, and M. Gellert, “The RAG1 and RAG2 proteins establish the 12/23 rule in V(D)J recombination.,” Cell, vol. 85, no. 1, pp. 107–13, Apr. 1996.

[10] K. Hiom and M. Gellert, “Assembly of a 12/23 paired signal complex: a critical control point in V(D)J recombination.,” Mol. Cell, vol. 1, no. 7, pp. 1011–9, Jun. 1998.

[11] Y. Chien, C. Meyer, and M. Bonneville, “γδ T Cells: First Line of Defense and Beyond,” Annu. Rev. Immunol., vol. 32, no. 1, pp. 121–155, Mar. 2014.

[12] T. Komori, A. Okada, V. Stewart, and F. Alt, “Lack of N regions in antigen receptor variable region genes of TdT-deficient lymphocytes,” Science (80-. )., vol. 261, no. 5125, pp. 1171–1175, Aug. 1993.

[13] S. Gilfillan, A. Dierich, M. Lemeur, C. Benoist, and D. Mathis, “Mice lacking TdT: mature animals with an immature lymphocyte repertoire,” Science (80-. )., vol. 261, no. 5125, pp. 1175–1178, Aug. 1993.

[14] S. Rooney et al., “Leaky Scid phenotype associated with defective V(D)J coding end processing in Artemis-deficient mice.,” Mol. Cell, vol. 10, no. 6, pp. 1379–90, Dec. 2002.

[15] S. Rooney et al., “Defective DNA Repair and Increased Genomic Instability in Artemis- deficient Murine Cells,” J. Exp. Med., vol. 197, no. 5, pp. 553–565, Mar. 2003.

[16] C. Kirchgessner et al., “DNA-dependent kinase (p350) as a candidate gene for the murine SCID defect,” Science (80-. )., vol. 267, no. 5201, pp. 1178–1183, Feb. 1995.

[17] T. Blunt et al., “Defective DNA-dependent protein kinase activity is linked to V(D)J recombination and DNA repair defects associated with the murine scid mutation,” Cell, vol. 80, no. 5, pp. 813–823, Mar. 1995.

[18] A. Medvinsky and E. Dzierzak, “Definitive Hematopoiesis Is Autonomously Initiated by the AGM Region,” Cell, vol. 86, no. 6, pp. 897–906, Sep. 1996.

[19] T. W. LeBien and T. F. Tedder, “B lymphocytes: how they develop and function,” Blood, vol. 112, no. 5, pp. 1570–1580, Sep. 2008.

[20] R. R. Hardy, C. E. Carmack, S. A. Shinton, J. D. Kemp, and K. Hayakawa, “Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow.,” J. Exp. Med., vol. 173, no. 5, pp. 1213–1225, May 1991.

[21] W. T. McCormack and C. B. Thompson, “Chicken IgL variable region gene conversions display pseudogene donor preference and 5’ to 3’ polarity.,” Genes Dev., vol. 4, no. 4, pp. 548–558, Apr. 1990.

[22] S. K. Petersen-Mahrt, R. S. Harris, and M. S. Neuberger, “AID mutates E. coli suggesting

a DNA deamination mechanism for antibody diversification.,” Nature, vol. 418, no. 6893, pp. 99–103, Jul. 2002.

[23] R. Bransteitter, P. Pham, M. D. Scharff, and M. F. Goodman, “Activation-induced cytidine deaminase deaminates deoxycytidine on single-stranded DNA but requires the action of RNase,” Proc. Natl. Acad. Sci., vol. 100, no. 7, pp. 4102–4107, Mar. 2003.

[24] C. Rada, J. M. Di Noia, and M. S. Neuberger, “Mismatch Recognition and Uracil Excision Provide Complementary Paths to Both Ig Switching and the A/T-Focused Phase of Somatic Mutation,” Mol. Cell, vol. 16, no. 2, pp. 163–171, Oct. 2004.

[25] E. S. Tang and A. Martin, “NHEJ-deficient DT40 cells have increased levels of immunoglobulin gene conversion: evidence for a double strand break intermediate.,” Nucleic Acids Res., vol. 34, no. 21, pp. 6345–51, Dec. 2006.

[26] M. Nakahara et al., “Genetic evidence for single-strand lesions initiating Nbs1-dependent homologous recombination in diversification of Ig v in chicken B lymphocytes.,” PLoS Genet., vol. 5, no. 1, p. e1000356, Jan. 2009.

[27] L. Davis and N. Maizels, “Homology-directed repair of DNA nicks via pathways distinct from canonical double-strand break repair.,” Proc. Natl. Acad. Sci. U. S. A., vol. 111, no. 10, pp. E924-32, Mar. 2014.

[28] G. Bastianello and H. Arakawa, “A double-strand break can trigger immunoglobulin gene conversion,” Nucleic Acids Res., vol. 45, no. 1, pp. 231–243, Jan. 2017.

[29] M. Muramatsu, K. Kinoshita, S. Fagarasan, S. Yamada, Y. Shinkai, and T. Honjo, “Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme.,” Cell, vol. 102, no. 5, Sep. 2000.

[30] C. Arpin et al., “The Normal Counterpart of IgD Myeloma Cells in Germinal Center Displays Extensively Mutated IgVH Gene, Cμ–Cδ Switch, and λ Light Chain Expression,” J. Exp. Med., vol. 187, no. 8, pp. 1169–1178, Apr. 1998.

[31] C. Gutzeit, K. Chen, and A. Cerutti, “The enigmatic function of IgD: some answers at last.,” Eur. J. Immunol., vol. 48, no. 7, pp. 1101–1113, 2018.

[32] T. M. . M. Luby, C. E. . E. Schrader, J. . Stavnezer, and E. . c Selsing, “The μ switch region tandem repeats are important, but not required, for antibody class switch recombination,” J. Exp. Med., vol. 193, no. 2, pp. 159–168, 2001.

[33] K. Yu, F. Chedin, C.-L. Hsieh, T. E. Wilson, and M. R. Lieber, “R-loops at immunoglobulin class switch regions in the chromosomes of stimulated B cells.,” Nat. Immunol., vol. 4, no. 5, pp. 442–51, May 2003.

[34] U. Basu et al., “The RNA Exosome Targets the AID Cytidine Deaminase to Both Strands of Transcribed Duplex DNA Substrates,” Cell, vol. 144, no. 3, pp. 353–363, Feb. 2011.

[35] P. Kodgire, P. Mukkawar, S. Ratnam, T. E. Martin, and U. Storb, “Changes in RNA polymerase II progression influence somatic hypermutation of Ig-related genes by AID,” J. Exp. Med., vol. 210, no. 7, pp. 1481–1492, Jul. 2013.

[36] J.-Y. Parsa et al., “Negative Supercoiling Creates Single-Stranded Patches of DNA That Are Substrates for AID–Mediated Mutagenesis,” PLoS Genet., vol. 8, no. 2, p. e1002518, Feb. 2012.

[37] H. M. Shen, A. Peters, B. Baron, X. Zhu, and U. Storb, “Mutation of BCL-6 Gene in Normal B Cells by the Process of Somatic Hypermutation of Ig Genes,” Science (80-. )., vol. 280, no. 5370, pp. 1750–1752, Jun. 1998.

[38] M. Müschen, D. Re, B. Jungnickel, V. Diehl, K. Rajewsky, and R. Küppers, “Somatic Mutation of the Cd95 Gene in Human B Cells as a Side-Effect of the Germinal Center Reaction,” J. Exp. Med., vol. 192, no. 12, pp. 1833–1840, Dec. 2000.

[39] L. Pasqualucci et al., “BCL-6 mutations in normal germinal center B cells: evidence of somatic hypermutation acting outside Ig loci.,” Proc. Natl. Acad. Sci. U. S. A., vol. 95, no. 20, pp. 11816–21, Sep. 1998.

[40] A. Martin and M. D. Scharff, “Somatic hypermutation of the AID transgene in B and non- B cells,” Proc. Natl. Acad. Sci., vol. 99, no. 19, pp. 12304–12308, Sep. 2002.

[41] M. S. Gordon, C. M. Kanegai, J. R. Doerr, and R. Wall, “Somatic hypermutation of the B cell receptor genes B29 (Igbeta, CD79b) and mb1 (Igalpha, CD79a).,” Proc. Natl. Acad.

Sci. U. S. A., vol. 100, no. 7, pp. 4126–31, Apr. 2003.

[42] M. Liu et al., “Two levels of protection for the B cell genome during somatic hypermutation,” Nature, vol. 451, no. 7180, pp. 841–845, Feb. 2008.

[43] A. R. Ramiro et al., “AID is required for c-myc/IgH chromosome translocations in vivo.,” Cell, vol. 118, no. 4, pp. 431–8, Aug. 2004.

[44] R. Pavri et al., “Activation-induced cytidine deaminase targets DNA at sites of RNA polymerase II stalling by interaction with Spt5.,” Cell, vol. 143, no. 1, pp. 122–33, Oct. 2010.

[45] K. L. Willmann et al., “A role for the RNA pol II–associated PAF complex in AID- induced immune diversification,” J. Exp. Med., vol. 209, no. 11, pp. 2099–2111, Oct. 2012.

[46] S. Zheng, B. Q. Vuong, B. Vaidyanathan, J.-Y. Lin, F.-T. Huang, and J. Chaudhuri, “Non- coding RNA Generated following Lariat Debranching Mediates Targeting of AID to DNA,” Cell, vol. 161, no. 4, pp. 762–773, May 2015.

[47] J. S. Rush, S. D. Fugmann, and D. G. Schatz, “Staggered AID-dependent DNA double strand breaks are the predominant DNA lesions targeted to S mu in Ig class switch recombination.,” Int. Immunol., vol. 16, no. 4, pp. 549–57, Apr. 2004.

[48] M. Kingzette, H. Spieker-Polet, P. C. Yam, S. K. Zhai, and K. L. Knight, “Trans- chromosomal recombination within the Ig heavy chain switch region in B lymphocytes.,” Proc. Natl. Acad. Sci. U. S. A., vol. 95, no. 20, pp. 11840–5, Sep. 1998.

[49] S. Reynaud, L. Delpy, L. Fleury, H.-L. Dougier, C. Sirac, and M. Cogné, “Interallelic class switch recombination contributes significantly to class switching in mouse B cells.,” J. Immunol., vol. 174, no. 10, pp. 6176–83, May 2005.

[50] J. Dong et al., “Orientation-specific joining of AID-initiated DNA breaks promotes antibody class switching,” Nature, vol. 525, no. 7567, pp. 134–139, Aug. 2015.

[51] R. Wuerffel et al., “S-S synapsis during class switch recombination is promoted by

distantly located transcriptional elements and activation-induced deaminase.,” Immunity, vol. 27, no. 5, pp. 711–22, Nov. 2007.

[52] X. Zhang, Y. Zhang, Z. Ba, N. Kyritsis, R. Casellas, and F. W. Alt, “Fundamental roles of chromatin loop extrusion in antibody class switching.,” Nature, vol. 575, no. 7782, pp. 385–389, Nov. 2019.

[53] A. Bothmer et al., “Regulation of DNA end joining, resection, and immunoglobulin class switch recombination by 53BP1.,” Mol. Cell, vol. 42, no. 3, pp. 319–29, May 2011.

[54] P. P. Rocha et al., “A Damage-Independent Role for 53BP1 that Impacts Break Order and Igh Architecture during Class Switch Recombination,” Cell Rep., vol. 16, no. 1, pp. 48– 55, Jun. 2016.

[55] S. Feldman, R. Wuerffel, I. Achour, L. Wang, P. B. Carpenter, and A. L. Kenter, “53BP1 Contributes to Igh Locus Chromatin Topology during Class Switch Recombination.,” J. Immunol., vol. 198, no. 6, pp. 2434–2444, Mar. 2017.

[56] F. Ochs et al., “Stabilization of chromatin topology safeguards genome integrity,” Nature, vol. 574, no. 7779, pp. 571–574, Oct. 2019.

[57] J. Falck, J. Coates, and S. P. Jackson, “Conserved modes of recruitment of ATM, ATR and DNA-PKcs to sites of DNA damage,” Nature, vol. 434, no. 7033, pp. 605–611, Mar. 2005.

[58] C. Doil et al., “RNF168 Binds and Amplifies Ubiquitin Conjugates on Damaged Chromosomes to Allow Accumulation of Repair Proteins,” Cell, vol. 136, no. 3, pp. 435– 446, Feb. 2009.

[59] G. S. Stewart et al., “The RIDDLE Syndrome Protein Mediates a Ubiquitin-Dependent Signaling Cascade at Sites of DNA Damage,” Cell, vol. 136, no. 3, pp. 420–434, Feb. 2009.

[60] F. Mattiroli et al., “RNF168 Ubiquitinates K13-15 on H2A/H2AX to Drive DNA Damage Signaling,” Cell, vol. 150, no. 6, pp. 1182–1195, Sep. 2012.

[61] I. M. Munoz, P. A. Jowsey, R. Toth, and J. Rouse, “Phospho-epitope binding by the BRCT domains of hPTIP controls multiple aspects of the cellular response to DNA damage,” Nucleic Acids Res., vol. 35, no. 16, pp. 5312–5322, Aug. 2007.

[62] M. Zimmermann, F. Lottersberger, S. B. Buonomo, A. Sfeir, and T. de Lange, “53BP1 Regulates DSB Repair Using Rif1 to Control 5’ End Resection,” Science (80-. )., vol. 339, no. 6120, pp. 700–704, Feb. 2013.

[63] C. Escribano-Díaz et al., “A cell cycle-dependent regulatory circuit composed of 53BP1- RIF1 and BRCA1-CtIP controls DNA repair pathway choice.,” Mol. Cell, vol. 49, no. 5, pp. 872–83, Mar. 2013.

[64] J. R. Chapman et al., “RIF1 Is Essential for 53BP1-Dependent Nonhomologous End Joining and Suppression of DNA Double-Strand Break Resection,” Mol. Cell, vol. 49, no. 5, pp. 858–871, Mar. 2013.

[65] L. Feng, K.-W. Fong, J. Wang, W. Wang, and J. Chen, “RIF1 Counteracts BRCA1- mediated End Resection during DNA Repair,” J. Biol. Chem., vol. 288, no. 16, pp. 11135– 11143, Apr. 2013.

[66] M. Di Virgilio et al., “Rif1 Prevents Resection of DNA Breaks and Promotes Immunoglobulin Class Switching,” Science (80-. )., vol. 339, no. 6120, pp. 711–715, Feb. 2013.

[67] L. R. Myler et al., “Single-Molecule Imaging Reveals How Mre11-Rad50-Nbs1 Initiates DNA Break Repair,” Mol. Cell, vol. 67, no. 5, pp. 891-898.e4, Sep. 2017.

[68] P. Chanut, S. Britton, J. Coates, S. P. Jackson, and P. Calsou, “Coordinated nuclease activities counteract Ku at single-ended DNA double-strand breaks,” Nat. Commun., vol. 7, p. 12889, Sep. 2016.

[69] E. J. Gapud et al., “Ataxia telangiectasia mutated (Atm) and DNA-PKcs kinases have overlapping activities during chromosomal signal joint formation,” Proc. Natl. Acad. Sci., vol. 108, no. 5, pp. 2022–2027, Feb. 2011.

[70] E. Callén et al., “Essential Role for DNA-PKcs in DNA Double-Strand Break Repair and

Apoptosis in ATM-Deficient Lymphocytes,” Mol. Cell, vol. 34, no. 3, pp. 285–297, May 2009.

[71] S. Zha et al., “Ataxia telangiectasia-mutated protein and DNA-dependent protein kinase have complementary V(D)J recombination functions,” Proc. Natl. Acad. Sci., vol. 108, no. 5, pp. 2028–2033, Feb. 2011.

[72] M. S. Satoh and T. Lindahl, “Role of poly(ADP-ribose) formation in DNA repair,” Nature, vol. 356, no. 6367, pp. 356–358, Mar. 1992.

[73] S. F. El-Khamisy, M. Masutani, H. Suzuki, and K. W. Caldecott, “A requirement for PARP-1 for the assembly or stability of XRCC1 nuclear foci at sites of oxidative DNA damage,” Nucleic Acids Res., vol. 31, no. 19, pp. 5526–5533, Oct. 2003.

[74] S. Okano, L. Lan, K. W. Caldecott, T. Mori, and A. Yasui, “Spatial and temporal cellular responses to single-strand breaks in human cells.,” Mol. Cell. Biol., vol. 23, no. 11, pp. 3974–81, Jun. 2003.

[75] S. Bekker-Jensen et al., “Human Xip1 (C2orf13) is a novel regulator of cellular responses to DNA strand breaks.,” J. Biol. Chem., vol. 282, no. 27, pp. 19638–43, Jul. 2007.

[76] S. Kanno et al., “A novel human AP endonuclease with conserved zinc-finger-like motifs involved in DNA strand break responses,” EMBO J., vol. 26, no. 8, pp. 2094–2103, Apr. 2007.

[77] W. Y. Mansour, T. Rhein, and J. Dahm-Daphi, “The alternative end-joining pathway for repair of DNA double-strand breaks requires PARP1 but is not dependent upon microhomologies,” Nucleic Acids Res., vol. 38, no. 18, pp. 6065–6077, Oct. 2010.

[78] M. Audebert, B. Salles, and P. Calsou, “Involvement of poly(ADP-ribose) polymerase-1 and XRCC1/DNA ligase III in an alternative route for DNA double-strand breaks rejoining.,” J. Biol. Chem., vol. 279, no. 53, pp. 55117–26, Dec. 2004.

[79] I. Robert, F. Dantzer, and B. Reina-San-Martin, “Parp1 facilitates alternative NHEJ, whereas Parp2 suppresses IgH/c-myc translocations during immunoglobulin class switch recombination,” J. Exp. Med., vol. 206, no. 5, pp. 1047–1056, May 2009.

[80] M. Wang et al., “PARP-1 and Ku compete for repair of DNA double strand breaks by distinct NHEJ pathways.,” Nucleic Acids Res., vol. 34, no. 21, pp. 6170–82, Dec. 2006.

[81] M. N. Paddock, A. T. Bauman, R. Higdon, E. Kolker, S. Takeda, and A. M. Scharenberg, “Competition between PARP-1 and Ku70 control the decision between high-fidelity and mutagenic DNA repair,” DNA Repair (Amst)., vol. 10, no. 3, pp. 338–343, Mar. 2011.

[82] H. Hochegger et al., “Parp-1 protects homologous recombination from interference by Ku and Ligase IV in vertebrate cells,” EMBO J., vol. 25, no. 6, pp. 1305–1314, Mar. 2006.

[83] G. Yang et al., “Super-resolution imaging identifies PARP1 and the Ku complex acting as DNA double-strand break sensors.,” Nucleic Acids Res., vol. 46, no. 7, pp. 3446–3457, 2018.

[84] M. Bétermier et al., “Is Non-Homologous End-Joining Really an Inherently Error-Prone Process?,” PLoS Genet., vol. 10, no. 1, p. e1004086, Jan. 2014.

[85] Q. Cheng et al., “Ku counteracts mobilization of PARP1 and MRN in chromatin damaged with DNA double-strand breaks,” Nucleic Acids Res., vol. 39, no. 22, pp. 9605–9619, Dec. 2011.

[86] A. Ray Chaudhuri and A. Nussenzweig, “The multifaceted roles of PARP1 in DNA repair and chromatin remodelling,” Nat. Rev. Mol. Cell Biol., vol. 18, no. 10, pp. 610–621, Oct. 2017.

[87] C. E. Schrader, E. K. Linehan, S. N. Mochegova, R. T. Woodland, and J. Stavnezer, “Inducible DNA breaks in Ig S regions are dependent on AID and UNG.,” J. Exp. Med., vol. 202, no. 4, pp. 561–8, Aug. 2005.

[88] C. Rada, G. T. Williams, H. Nilsen, D. E. Barnes, T. Lindahl, and M. S. Neuberger, “Immunoglobulin isotype switching is inhibited and somatic hypermutation perturbed in UNG-deficient mice.,” Curr. Biol., vol. 12, no. 20, pp. 1748–55, Oct. 2002.

[89] K. Imai et al., “Human uracil–DNA glycosylase deficiency associated with profoundly impaired immunoglobulin class-switch recombination,” Nat. Immunol., vol. 4, no. 10, pp. 1023–1028, Oct. 2003.

[90] J. E. J. Guikema et al., “APE1- and APE2-dependent DNA breaks in immunoglobulin class switch recombination.,” J. Exp. Med., vol. 204, no. 12, pp. 3017–26, Nov. 2007.

[91] G. L. Dianov and U. Hübscher, “Mammalian base excision repair: the forgotten archangel.,” Nucleic Acids Res., vol. 41, no. 6, pp. 3483–90, Apr. 2013.

[92] F. A. Kadyrov, L. Dzantiev, N. Constantin, and P. Modrich, “Endonucleolytic Function of MutLα in Human Mismatch Repair,” Cell, vol. 126, no. 2, pp. 297–308, Jul. 2006.

[93] J. Genschel, L. R. Bazemore, and P. Modrich, “Human Exonuclease I Is Required for 5′ and 3′ Mismatch Repair,” J. Biol. Chem., vol. 277, no. 15, pp. 13302–13311, Apr. 2002.

[94] A. Umar et al., “Requirement for PCNA in DNA mismatch repair at a step preceding DNA resynthesis.,” Cell, vol. 87, no. 1, pp. 65–73, Oct. 1996.

[95] H. T. Tran, D. A. Gordenin, and M. A. Resnick, “The 3’-->5’ exonucleases of DNA polymerases delta and epsilon and the 5’-->3’ exonuclease Exo1 have major roles in postreplication mutation avoidance in Saccharomyces cerevisiae.,” Mol. Cell. Biol., vol. 19, no. 3, pp. 2000–7, Mar. 1999.

[96] E. Soutoglou et al., “Positional stability of single double-strand breaks in mammalian cells,” Nat. Cell Biol., vol. 9, no. 6, pp. 675–682, Jun. 2007.

[97] M. Clerici, D. Mantiero, I. Guerini, G. Lucchini, and M. P. Longhese, “The Yku70– Yku80 complex contributes to regulate double‐strand break processing and checkpoint activation during the cell cycle,” EMBO Rep., vol. 9, no. 8, pp. 810–818, Aug. 2008.

[98] E. Y. Shim et al., “Saccharomyces cerevisiae Mre11/Rad50/Xrs2 and Ku proteins regulate association of Exo1 and Dna2 with DNA breaks,” EMBO J., vol. 29, no. 19, pp. 3370– 3380, Oct. 2010.

[99] E. P. Mimitou and L. S. Symington, “Ku prevents Exo1 and Sgs1-dependent resection of DNA ends in the absence of a functional MRX complex or Sae2,” EMBO J., vol. 29, no. 19, pp. 3358–3369, Oct. 2010.

[100] H. Lu, U. Pannicke, K. Schwarz, and M. R. Lieber, “Length-dependent Binding of Human

XLF to DNA and Stimulation of XRCC4·DNA Ligase IV Activity,” J. Biol. Chem., vol. 282, no. 15, pp. 11155–11162, Apr. 2007.

[101] C. J. Tsai, S. A. Kim, and G. Chu, “Cernunnos/XLF promotes the ligation of mismatched and noncohesive DNA ends,” Proc. Natl. Acad. Sci., vol. 104, no. 19, pp. 7851–7856, May 2007.

[102] P. Hentges et al., “Evolutionary and Functional Conservation of the DNA Non- homologous End-joining Protein, XLF/Cernunnos,” J. Biol. Chem., vol. 281, no. 49, pp. 37517–37526, Dec. 2006.

[103] Q. Wu, T. Ochi, D. Matak-Vinkovic, C. V. Robinson, D. Y. Chirgadze, and T. L. Blundell, “Non-homologous end-joining partners in a helical dance: structural studies of XLF–XRCC4 interactions,” Biochem. Soc. Trans., vol. 39, no. 5, pp. 1387–1392, Oct. 2011.

[104] M. Hammel et al., “XRCC4 protein interactions with XRCC4-like factor (XLF) create an extended grooved scaffold for DNA ligation and double strand break repair.,” J. Biol. Chem., vol. 286, no. 37, pp. 32638–50, Sep. 2011.

[105] V. Ropars et al., “Structural characterization of filaments formed by human Xrcc4- Cernunnos/XLF complex involved in nonhomologous DNA end-joining,” Proc. Natl. Acad. Sci., vol. 108, no. 31, pp. 12663–12668, Aug. 2011.

[106] S. N. Andres, A. Vergnes, D. Ristic, C. Wyman, M. Modesti, and M. Junop, “A human XRCC4–XLF complex bridges DNA,” Nucleic Acids Res., vol. 40, no. 4, pp. 1868–1878, Feb. 2012.

[107] S. Roy et al., “XRCC4/XLF Interaction Is Variably Required for DNA Repair and Is Not Required for Ligase IV Stimulation.,” Mol. Cell. Biol., vol. 35, no. 17, pp. 3017–28, Sep. 2015.

[108] V. Kumar, F. W. Alt, and R. L. Frock, “PAXX and XLF DNA repair factors are functionally redundant in joining DNA breaks in a G1-arrested progenitor B-cell line,” Proc. Natl. Acad. Sci., vol. 113, no. 38, pp. 10619–10624, Sep. 2016.

[109] C. Lescale et al., “Specific Roles of XRCC4 Paralogs PAXX and XLF during V(D)J Recombination,” Cell Rep., vol. 16, no. 11, pp. 2967–2979, Sep. 2016.

[110] S. K. Tadi et al., “PAXX Is an Accessory c-NHEJ Factor that Associates with Ku70 and Has Overlapping Functions with XLF,” Cell Rep., vol. 17, no. 2, pp. 541–555, Oct. 2016.

[111] X. Liu, Z. Shao, W. Jiang, B. J. Lee, and S. Zha, “PAXX promotes KU accumulation at DNA breaks and is essential for end-joining in XLF-deficient mice,” Nat. Commun., vol. 8, no. 1, p. 13816, Apr. 2017.

[112] P. J. Hung et al., “Deficiency of XLF and PAXX prevents DNA double-strand break repair by non-homologous end joining in lymphocytes,” Cell Cycle, vol. 16, no. 3, pp. 286–295, Feb. 2017.

[113] G. Balmus et al., “Synthetic lethality between PAXX and XLF in mammalian development,” Genes Dev., vol. 30, no. 19, pp. 2152–2157, Oct. 2016.

[114] V. Oksenych et al., “Functional redundancy between the XLF and DNA-PKcs DNA repair factors in V(D)J recombination and nonhomologous DNA end joining,” Proc. Natl. Acad. Sci., vol. 110, no. 6, pp. 2234–2239, Feb. 2013.

[115] S. Zha et al., “ATM damage response and XLF repair factor are functionally redundant in joining DNA breaks,” Nature, vol. 469, no. 7329, pp. 250–254, Jan. 2011.

[116] X. Liu, W. Jiang, R. L. Dubois, K. Yamamoto, Z. Wolner, and S. Zha, “Overlapping functions between XLF repair protein and 53BP1 DNA damage response factor in end joining and lymphocyte development.,” Proc. Natl. Acad. Sci. U. S. A., vol. 109, no. 10, pp. 3903–8, Mar. 2012.

[117] G. S. Lee, M. B. Neiditch, S. S. Salus, and D. B. Roth, “RAG proteins shepherd double- strand breaks to a specific pathway, suppressing error-prone repair, but RAG nicking initiates homologous recombination.,” Cell, vol. 117, no. 2, pp. 171–84, Apr. 2004.

[118] B. Corneo et al., “Rag mutations reveal robust alternative end joining,” Nature, vol. 449, no. 7161, pp. 483–486, Sep. 2007.

[119] X. Cui and K. Meek, “Linking double-stranded DNA breaks to the recombination activating gene complex directs repair to the nonhomologous end-joining pathway,” Proc. Natl. Acad. Sci., vol. 104, no. 43, pp. 17046–17051, Oct. 2007.

[120] C. Lescale et al., “RAG2 and XLF/Cernunnos interplay reveals a novel role for the RAG complex in DNA repair,” Nat. Commun., vol. 7, no. 1, p. 10529, Apr. 2016.

[121] G. Li et al., “Lymphocyte-Specific Compensation for XLF/Cernunnos End-Joining Functions in V(D)J Recombination,” Mol. Cell, vol. 31, no. 5, pp. 631–640, Sep. 2008.

[122] J. P. Manis, J. C. Morales, Z. Xia, J. L. Kutok, F. W. Alt, and P. B. Carpenter, “53BP1 links DNA damage-response pathways to immunoglobulin heavy chain class-switch recombination,” Nat. Immunol., vol. 5, no. 5, pp. 481–487, May 2004.

[123] I. M. Ward et al., “53BP1 is required for class switch recombination,” J. Cell Biol., vol. 165, no. 4, pp. 459–464, May 2004.

[124] C. T. Yan et al., “IgH class switching and translocations use a robust non-classical end- joining pathway,” Nature, vol. 449, no. 7161, pp. 478–482, Aug. 2007.

[125] L. Han and K. Yu, “Altered kinetics of nonhomologous end joining and class switch recombination in ligase IV-deficient B cells.,” J. Exp. Med., vol. 205, no. 12, pp. 2745–53, Nov. 2008.

[126] C. Boboila et al., “Alternative end-joining catalyzes class switch recombination in the absence of both Ku70 and DNA ligase 4.,” J. Exp. Med., vol. 207, no. 2, pp. 417–27, Feb. 2010.

[127] Z. Mirman and T. de Lange, “53BP1: a DSB escort.,” Genes Dev., vol. 34, no. 1–2, pp. 7– 23, Jan. 2020.

[128] F. Lottersberger, A. Bothmer, D. F. Robbiani, M. C. Nussenzweig, T. de Lange, and T. de Lange, “Role of 53BP1 oligomerization in regulating double-strand break repair.,” Proc. Natl. Acad. Sci. U. S. A., vol. 110, no. 6, pp. 2146–51, Feb. 2013.

[129] A. Bothmer, D. F. Robbiani, N. Feldhahn, A. Gazumyan, A. Nussenzweig, and M. C.

Nussenzweig, “53BP1 regulates DNA resection and the choice between classical and alternative end joining during class switch recombination.,” J. Exp. Med., vol. 207, no. 4, pp. 855–65, Apr. 2010.

[130] J. A. Daniel et al., “PTIP Promotes Chromatin Changes Critical for Immunoglobulin Class Switch Recombination,” Science (80-. )., vol. 329, no. 5994, pp. 917–923, Aug. 2010.

[131] K. R. Schwab, S. R. Patel, and G. R. Dressler, “Role of PTIP in class switch recombination and long-range chromatin interactions at the immunoglobulin heavy chain locus.,” Mol. Cell. Biol., vol. 31, no. 7, pp. 1503–11, Apr. 2011.

[132] E. Callen et al., “53BP1 Mediates Productive and Mutagenic DNA Repair through Distinct Phosphoprotein Interactions,” Cell, vol. 153, no. 6, pp. 1266–1280, Jun. 2013.

[133] G. Xu et al., “REV7 counteracts DNA double-strand break resection and affects PARP inhibition,” Nature, vol. 521, no. 7553, pp. 541–544, May 2015.

[134] V. Boersma et al., “MAD2L2 controls DNA repair at telomeres and DNA breaks by inhibiting 5′ end resection,” Nature, vol. 521, no. 7553, pp. 537–540, May 2015.

[135] S. M. Noordermeer et al., “The shieldin complex mediates 53BP1-dependent DNA repair,” Nature, vol. 560, no. 7716, pp. 117–121, Aug. 2018.

[136] H. Ghezraoui et al., “53BP1 cooperation with the REV7–shieldin complex underpins DNA structure-specific NHEJ,” Nature, vol. 560, no. 7716, pp. 122–127, Aug. 2018.

[137] H. Dev et al., “Shieldin complex promotes DNA end-joining and counters homologous recombination in BRCA1-null cells,” Nat. Cell Biol., vol. 20, no. 8, pp. 954–965, Aug. 2018.

[138] S. Findlay et al., “SHLD 2/ FAM 35A co‐operates with REV 7 to coordinate DNA double‐strand break repair pathway choice,” EMBO J., vol. 37, no. 18, Sep. 2018.

[139] S. Gao et al., “An OB-fold complex controls the repair pathways for DNA double-strand breaks,” Nat. Commun., vol. 9, no. 1, p. 3925, Dec. 2018.

[140] S. Difilippantonio et al., “53BP1 facilitates long-range DNA end-joining during V(D)J recombination,” Nature, vol. 456, no. 7221, pp. 529–533, Nov. 2008.

[141] V. Oksenych et al., “Functional redundancy between repair factor XLF and damage response mediator 53BP1 in V(D)J recombination and DNA repair.,” Proc. Natl. Acad. Sci. U. S. A., vol. 109, no. 7, pp. 2455–60, Feb. 2012.

[142] S. Kilic et al., “Phase separation of 53 BP 1 determines liquid‐like behavior of DNA repair compartments,” EMBO J., vol. 38, no. 16, p. e101379, Aug. 2019.

[143] S. J. Boulton and S. P. Jackson, “Saccharomyces cerevisiae Ku70 potentiates illegitimate DNA double-strand break repair and serves as a barrier to error-prone DNA repair pathways.,” EMBO J., vol. 15, no. 18, pp. 5093–103, Sep. 1996.

[144] E. B. Kabotyanski, L. Gomelsky, J.-O. Han, D. B. Roth, and T. D. Stamato, “Double- strand break repair in Ku86- and XRCC4-deficient cells,” Nucleic Acids Res., vol. 26, no. 23, pp. 5333–5342, Dec. 1998.

[145] L. N. Truong et al., “Microhomology-mediated End Joining and Homologous Recombination share the initial end resection step to repair DNA double-strand breaks in mammalian cells.,” Proc. Natl. Acad. Sci. U. S. A., vol. 110, no. 19, pp. 7720–5, May 2013.

[146] K. Beagan et al., “Drosophila DNA polymerase theta utilizes both helicase-like and polymerase domains during microhomology-mediated end joining and interstrand crosslink repair.,” PLoS Genet., vol. 13, no. 5, p. e1006813, May 2017.

[147] S. H. Chan, A. M. Yu, and M. McVey, “Dual roles for DNA polymerase theta in alternative end-joining repair of double-strand breaks in Drosophila.,” PLoS Genet., vol. 6, no. 7, p. e1001005, Jul. 2010.

[148] P. A. Mateos-Gomez, F. Gong, N. Nair, K. M. Miller, E. Lazzerini-Denchi, and A. Sfeir, “Mammalian polymerase θ promotes alternative NHEJ and suppresses recombination,” Nature, vol. 518, no. 7538, pp. 254–7, Feb. 2015.

[149] D. W. Wyatt et al., “Essential Roles for Polymerase θ-Mediated End Joining in the Repair

of Chromosome Breaks.,” Mol. Cell, vol. 63, no. 4, pp. 662–73, Aug. 2016.

[150] K. E. Zahn, A. M. Averill, P. Aller, R. D. Wood, and S. Doublié, “Human DNA polymerase θ grasps the primer terminus to mediate DNA repair.,” Nat. Struct. Mol. Biol., vol. 22, no. 4, pp. 304–11, Mar. 2015.

[151] M. J. Yousefzadeh et al., “Mechanism of suppression of chromosomal instability by DNA polymerase POLQ.,” PLoS Genet., vol. 10, no. 10, p. e1004654, Oct. 2014.

[152] R. Ceccaldi et al., “Homologous-recombination-deficient tumours are dependent on Polθ- mediated repair,” Nature, vol. 518, no. 7538, pp. 258–262, Feb. 2015.

[153] M. Audebert, B. Salles, M. Weinfeld, and P. Calsou, “Involvement of polynucleotide kinase in a poly(ADP-ribose) polymerase-1-dependent DNA double-strand breaks rejoining pathway.,” J. Mol. Biol., vol. 356, no. 2, pp. 257–65, Feb. 2006.

[154] H. Wang et al., “DNA ligase III as a candidate component of backup pathways of nonhomologous end joining.,” Cancer Res., vol. 65, no. 10, pp. 4020–30, May 2005.

[155] L. Liang et al., “Human DNA ligases I and III, but not ligase IV, are required for microhomology-mediated end joining of DNA double-strand breaks.,” Nucleic Acids Res., vol. 36, no. 10, pp. 3297–310, Jun. 2008.

[156] A. Xie, A. Kwok, and R. Scully, “Role of mammalian Mre11 in classical and alternative nonhomologous end joining.,” Nat. Struct. Mol. Biol., vol. 16, no. 8, pp. 814–8, Aug. 2009.

[157] E. Rass, A. Grabarz, I. Plo, J. Gautier, P. Bertrand, and B. S. Lopez, “Role of Mre11 in chromosomal nonhomologous end joining in mammalian cells.,” Nat. Struct. Mol. Biol., vol. 16, no. 8, pp. 819–24, Aug. 2009.

[158] M. Lee-Theilen, A. J. Matthews, D. Kelly, S. Zheng, and J. Chaudhuri, “CtIP promotes microhomology-mediated alternative end joining during class-switch recombination,” Nat. Struct. Mol. Biol., vol. 18, no. 1, pp. 75–79, Dec. 2010.

[159] M. H. Yun and K. Hiom, “CtIP-BRCA1 modulates the choice of DNA double-strand-

break repair pathway throughout the cell cycle.,” Nature, vol. 459, no. 7245, pp. 460–3, May 2009.

[160] Y. Zhang and M. Jasin, “An essential role for CtIP in chromosomal translocation formation through an alternative end-joining pathway.,” Nat. Struct. Mol. Biol., vol. 18, no. 1, pp. 80–4, Jan. 2011.

[161] P. A. Mateos-Gomez et al., “The helicase domain of Polθ counteracts RPA to promote alt- NHEJ.,” Nat. Struct. Mol. Biol., Oct. 2017.

[162] T. Kent, P. A. Mateos-Gomez, A. Sfeir, and R. T. Pomerantz, “Polymerase θ is a robust terminal transferase that oscillates between three different mechanisms during end- joining.,” Elife, vol. 5, 2016.

[163] T. Kent, G. Chandramouly, S. M. McDevitt, A. Y. Ozdemir, and R. T. Pomerantz, “Mechanism of microhomology-mediated end-joining promoted by human DNA polymerase θ,” Nat. Struct. Mol. Biol., vol. 22, no. 3, pp. 230–237, Feb. 2015.

[164] S. J. Black et al., “Molecular basis of microhomology-mediated end-joining by purified full-length Polθ,” Nat. Commun., vol. 10, no. 1, p. 4423, Dec. 2019.

[165] W. Feng et al., “Genetic determinants of cellular addiction to DNA polymerase theta,” Nat. Commun., vol. 10, no. 1, p. 4286, Dec. 2019.

[166] E. P. Mimitou and L. S. Symington, “Sae2, Exo1 and Sgs1 collaborate in DNA double- strand break processing,” Nature, vol. 455, no. 7214, pp. 770–774, Oct. 2008.

[167] A. Shibata et al., “DNA double-strand break repair pathway choice is directed by distinct MRE11 nuclease activities.,” Mol. Cell, vol. 53, no. 1, pp. 7–18, Jan. 2014.

[168] Z. Zhu, W.-H. Chung, E. Y. Shim, S. E. Lee, and G. Ira, “Sgs1 Helicase and Two Nucleases Dna2 and Exo1 Resect DNA Double-Strand Break Ends,” Cell, vol. 134, no. 6, pp. 981–994, Sep. 2008.

[169] S. Gravel, J. R. Chapman, C. Magill, and S. P. Jackson, “DNA helicases Sgs1 and BLM promote DNA double-strand break resection.,” Genes Dev., vol. 22, no. 20, pp. 2767–72,

Oct. 2008.

[170] J. W. Szostak, T. L. Orr-Weaver, R. J. Rothstein, and F. W. Stahl, “The double-strand- break repair model for recombination,” Cell, vol. 33, no. 1, pp. 25–35, May 1983.

[171] K. S. Makarova et al., “Evolutionary classification of CRISPR–Cas systems: a burst of class 2 and derived variants,” Nat. Rev. Microbiol., vol. 18, no. 2, pp. 67–83, Feb. 2020.

[172] H. Wang, M. La Russa, and L. S. Qi, “CRISPR/Cas9 in Genome Editing and Beyond,” Annu. Rev. Biochem., vol. 85, no. 1, pp. 227–264, Jun. 2016.

[173] M. R. Ehrenstein, C. Rada, A. M. Jones, C. Milstein, and M. S. Neuberger, “Switch junction sequences in PMS2-deficient mice reveal a microhomology-mediated mechanism of Ig class switch recombination.,” Proc. Natl. Acad. Sci. U. S. A., vol. 98, no. 25, pp. 14553–8, Dec. 2001.

[174] Q. Pan, C. Petit-Frére, A. Lähdesmäki, H. Gregorek, K. H. Chrzanowska, and L. Hammarström, “Alternative end joining during switch recombination in patients with ataxia-telangiectasia.,” Eur. J. Immunol., vol. 32, no. 5, pp. 1300–8, May 2002.

[175] K. Xue, C. Rada, and M. S. Neuberger, “The in vivo pattern of AID targeting to immunoglobulin switch regions deduced from mutation spectra in msh2-/- ung-/- mice.,” J. Exp. Med., vol. 203, no. 9, pp. 2085–94, Sep. 2006.

[176] A. A. Zarrin et al., “Antibody class switching mediated by yeast endonuclease-generated DNA breaks.,” Science, vol. 315, no. 5810, pp. 377–81, Jan. 2007.

[177] A. Bothmer et al., “Mechanism of DNA resection during intrachromosomal recombination and immunoglobulin class switching.,” J. Exp. Med., vol. 210, no. 1, pp. 115–23, Jan. 2013.

[178] M. Gostissa et al., “IgH class switching exploits a general property of two DNA breaks to be joined in cis over long chromosomal distances.,” Proc. Natl. Acad. Sci. U. S. A., vol. 111, no. 7, pp. 2644–9, Feb. 2014.

[179] E. M. Cortizas, A. Zahn, M. E. Hajjar, A.-M. Patenaude, J. M. Di Noia, and R. E. Verdun,

“Alternative end-joining and classical nonhomologous end-joining pathways repair different types of double-strand breaks during class-switch recombination.,” J. Immunol., vol. 191, no. 11, pp. 5751–63, Dec. 2013.

[180] S. Longerich and U. Storb, “The contested role of uracil DNA glycosylase in immunoglobulin gene diversification,” Trends Genet., vol. 21, no. 5, pp. 253–256, May 2005.

[181] S. Ramachandran et al., “The SAGA Deubiquitination Module Promotes DNA Repair and Class Switch Recombination through ATM and DNAPK-Mediated γH2AX Formation.,” Cell Rep., vol. 15, no. 7, pp. 1554–65, May 2016.

[182] G. Gasiunas, R. Barrangou, P. Horvath, and V. Siksnys, “Cas9-crRNA ribonucleoprotein complex mediates specific DNA cleavage for adaptive immunity in bacteria.,” Proc. Natl. Acad. Sci. U. S. A., vol. 109, no. 39, pp. E2579-86, Sep. 2012.

[183] F. A. Ran et al., “Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity.,” Cell, vol. 154, no. 6, pp. 1380–9, Sep. 2013.

[184] B. Shen et al., “Efficient genome modification by CRISPR-Cas9 nickase with minimal off-target effects.,” Nat. Methods, vol. 11, no. 4, pp. 399–402, Apr. 2014.

[185] P. Pham, R. Bransteitter, J. Petruska, and M. F. Goodman, “Processive AID-catalysed cytosine deamination on single-stranded DNA simulates somatic hypermutation.,” Nature, vol. 424, no. 6944, pp. 103–7, Jul. 2003.

[186] A. Bothmer et al., “Characterization of the interplay between DNA repair and CRISPR/Cas9-induced DNA lesions at an endogenous locus,” Nat. Commun., vol. 8, no. May 2016, p. 13905, Jan. 2017.

[187] Q. Pan-Hammarström et al., “Impact of DNA ligase IV on nonhomologous end joining pathways during class switch recombination in human cells.,” J. Exp. Med., vol. 201, no. 2, pp. 189–94, Jan. 2005.

[188] L. E. M. Vriend et al., “Distinct genetic control of homologous recombination repair of Cas9-induced double-strand breaks, nicks and paired nicks.,” Nucleic Acids Res., p.

gkw179, Mar. 2016.

[189] N. Arnoult et al., “Regulation of DNA repair pathway choice in S and G2 phases by the NHEJ inhibitor CYREN.,” Nature, vol. 549, no. 7673, pp. 548–552, Sep. 2017.

[190] J. R. Chapman, M. R. G. Taylor, and S. J. Boulton, “Playing the End Game: DNA Double-Strand Break Repair Pathway Choice,” Mol. Cell, vol. 47, no. 4, pp. 497–510, 2012.

[191] H. H. Y. Chang et al., “Different DNA End Configurations Dictate Which NHEJ Components Are Most Important for Joining Efficiency.,” J. Biol. Chem., vol. 291, no. 47, pp. 24377–24389, Nov. 2016.

[192] D. A. Reid et al., “Bridging of double-stranded breaks by the nonhomologous end-joining ligation complex is modulated by DNA end chemistry.,” Nucleic Acids Res., pp. gkw1221-, Dec. 2016.

[193] Z. Liang, S. Sunder, S. Nallasivam, and T. E. Wilson, “Overhang polarity of chromosomal double-strand breaks impacts kinetics and fidelity of yeast non-homologous end joining.,” Nucleic Acids Res., vol. 44, no. 6, pp. 2769–81, Apr. 2016.

[194] J. M. Daley and T. E. Wilson, “Rejoining of DNA double-strand breaks as a function of overhang length.,” Mol. Cell. Biol., vol. 25, no. 3, pp. 896–906, Feb. 2005.

[195] M. van Overbeek et al., “DNA Repair Profiling Reveals Nonrandom Outcomes at Cas9- Mediated Breaks,” Mol. Cell, vol. 63, no. 4, pp. 633–646, 2016.

[196] S. F. Bunting et al., “53BP1 inhibits homologous recombination in Brca1-deficient cells by blocking resection of DNA breaks.,” Cell, vol. 141, no. 2, pp. 243–54, Apr. 2010.

[197] H. Zan et al., “Rad52 competes with Ku70/Ku86 for binding to S-region DSB ends to modulate antibody class-switch DNA recombination.,” Nat. Commun., vol. 8, p. 14244, Feb. 2017.

[198] Q. Wang et al., “The cell cycle restricts activation-induced cytidine deaminase activity to early G1.,” J. Exp. Med., vol. 214, no. 1, pp. 49–58, 2017.

[199] Q. Le et al., “Cell Cycle Regulates Nuclear Stability of AID and Determines the Cellular Response to AID,” PLOS Genet., vol. 11, no. 9, p. e1005411, Sep. 2015.

[200] M. Larijani, A. P. Petrov, O. Kolenchenko, M. Berru, S. N. Krylov, and A. Martin, “AID Associates with Single-Stranded DNA with High Affinity and a Long Complex Half-Life in a Sequence-Independent Manner,” Mol. Cell. Biol., vol. 27, no. 1, pp. 20–30, Jan. 2007.

[201] A. C. Komor, Y. B. Kim, M. S. Packer, J. A. Zuris, and D. R. Liu, “Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage,” Nature, vol. 533, no. 7603, pp. 420–424, 2016.

[202] K. Nishida et al., “Targeted nucleotide editing using hybrid prokaryotic and vertebrate adaptive immune systems,” Science (80-. )., vol. 353, no. 6305, pp. aaf8729–aaf8729, Sep. 2016.

[203] C. D. Richardson, G. J. Ray, M. A. DeWitt, G. L. Curie, and J. E. Corn, “Enhancing homology-directed genome editing by catalytically active and inactive CRISPR-Cas9 using asymmetric donor DNA.,” Nat. Biotechnol., Jan. 2016.

[204] F. A. Ran, P. D. Hsu, J. Wright, V. Agarwala, D. A. Scott, and F. Zhang, “Genome engineering using the CRISPR-Cas9 system.,” Nat. Protoc., vol. 8, no. 11, pp. 2281–308, Nov. 2013.

[205] J. G. Doench et al., “Rational design of highly active sgRNAs for CRISPR-Cas9– mediated gene inactivation,” Nat. Biotechnol., vol. 32, no. 12, pp. 1262–7, Sep. 2014.

[206] A. L. Bredemeyer et al., “ATM stabilizes DNA double-strand-break complexes during V(D)J recombination,” Nature, vol. 442, no. 7101, pp. 466–470, Jul. 2006.

[207] C. Li et al., “The H2B deubiquitinase Usp22 promotes antibody class switch recombination by facilitating non-homologous end joining,” Nat. Commun., vol. 9, no. 1, p. 1006, Dec. 2018.

[208] A. K. Ling, C. C. So, M. X. Le, A. Y. Chen, L. Hung, and A. Martin, “Double-stranded DNA break polarity skews repair pathway choice during intrachromosomal and interchromosomal recombination,” Proc. Natl. Acad. Sci., vol. 115, no. 11, pp. 2800–

2805, Mar. 2018.

[209] S. M. Harding, J. A. Boiarsky, and R. A. Greenberg, “ATM Dependent Silencing Links Nucleolar Chromatin Reorganization to DNA Damage Recognition,” Cell Rep., vol. 13, no. 2, pp. 251–259, Oct. 2015.

[210] R. A. Panchakshari et al., “DNA double-strand break response factors influence end- joining features of IgH class switch and general translocation junctions.,” Proc. Natl. Acad. Sci. U. S. A., vol. 115, no. 4, pp. 762–767, 2018.

Appendices

Appendix Figure 1. Cas9-mediated DSBs induce CSR in multiple cell lines.

85 86

(A) Flow cytometry plots illustrating CSR to the indicated Ig isotypes by Aid−/− CH12 cells transfected with Cas9 and sgRNAs targeting S′μ and the relevant acceptor switch region. pX330, empty vector control (Cas9-expressing only, no sgRNA). (B) Schematic of Igh locus and sgRNA target regions in S′μ and S′α, showing the percentages of Aid−/− CH12, WEHI-279, M12.4.1, and A20 murine B cell lines that have undergone CSR to IgA after transfection with the indicated pairs of S′μ- and S′α-targeting sgRNAs. (C) Flow cytometry plots illustrating CSR to IgG3 or IgG1 by WEHI-279 cells transfected with Cas9 and S′μor S’γ-targeting sgRNAs, respectively. (D) Percentage of IgA- switched Aid−/− CH12 cells on transfection with sgRNAs targeting S′μ and S′α. The mass of each sgRNA plasmid used in each transfection is indicated in the table below. Data were analyzed using one-way ANOVA with Bonferroni post hoc testing. (E) A ∼1-kb region surrounding the 3′ UTR downstream of the Cα3 exon was amplified by reverse-transcription PCR and digested with KpnI in Aid−/− CH12 cells that had undergone successful Cas9-mediated CSR to IgA. Each lane is denoted by the status of the KpnI restriction site in the VDJ-rearranged and - unrearranged alleles. The 600-bp band in lane 3 indicates that CSR has occurred in trans. Quantification of independent clonal replicates of KpnI and StuI digests are shown in Figure 3D. (F) Flow cytometry plots showing Igκ and IgA expression following CIT stimulation and Cas9- mediated CSR in wild-type and Aid−/− CH12 cells. Igκlo cells represent an inversional recombination of DSBs created in S′μ. ΔCμ is a sgRNA that targets Cμ exon 1, leading to loss of IgM expression, and serves as a positive control for loss of surface Ig. (G) Gel electrophoresis image showing detection of inversional recombination by PCR. PCR amplification of inversion recombination (Figure 3C) from bulk, sorted IgA+, and sorted Igκlo CH12 cells stimulated with CIT or Aid−/− CH12 cells induced to undergo Cas9-mediated CSR to IgA. Two replicates are shown (1 and 2). The center lane contains a 100-bp DNA ladder. The control amplicon represents amplification of Eny2 exon 1 (Appendix Table 1). Error bars represent SD. *P < 0.05, **P <0.01, ***P < 0.001.

Appendix Figure 2. Non-canonical lesions mediate switching. (A) A single Cas9 break in S′μ can mediate switching in CH12 cells. (B) Schematic of S′μ (upstream of Sμ) and S′α (downstream of Sα) loci in Igh and sgRNA targeting either the top or the bottom strands. (C) Percentage of Aid−/− CH12 cells that have undergone CSR to IgA on transfection with sgRNA targeting the top strand and Cas9D10A to generate same-strand nicks in S′μ and S′α. Combination of sgRNAs used in each experiment are denoted by colored blocks below and correspond to the sgRNAs depicted in B. (D) Same as C, but using sgRNA targeting the bottom strand. pX330 is the empty vector control.

Appendix Figure 3. Nucleolytic resection prior to ligation following Cas9 DSBs.

(A) AID-induced dU are converted to nicks or gaps with either Ung-1/APE or the mismatch repair pathway (MMR) leading to 5’DSBs or 3’DSBs. 5′ to 3′ exonuclease-dependent resection of distal nicks or gaps is expected to lead to production of DNA ends with 5′ DSBs but not 3′ DSBs with long overhangs. Arrows depict resection in the 5′ to 3′ direction. Resection of S′μ versus S′α in μ-α junctions derived from Aid–/– CH12 cells induced to undergo nickase-mediated CSR via DSBs with a 35 bp overhang in S′μ and a (B) 24 bp or (C) 98 bp overhang in S′α. (D) Resection of S′μ versus S′α in μ-α junctions derived from Aid–/–Lig4–/– CH12 cells induced to undergo nickase-mediated CSR via DSBs with a 35 bp overhang in S′μ and a 24 bp overhang in S′α.

Appendix Figure 4. Generation of Aid–/– Lig4–/– CH12 cells. (A) Aid–/– CH12 cells were transfected with LigG2 sgRNA and single clones were obtained. Lig4 alleles from Aid–/– Lig4–/– CH12 clone 1 were sequenced and shown to possess frame-shift mutations. (B) qPCR analysis showed a reduced level of LIG4 mRNA in Aid–/–Lig4–/– CH12 clones 1 and 2 compared to Aid–/– CH12 parental cells.

Appendix Table 1. Oligonucleotides used in Chapter 2 for Cas9-mediated CSR and gene perturbation.

→ Category Oligonucleotide name Type Sequence (5' 3') S’μ1 sgRNA CAGGATTGCCTTCTTAGCCT S’μ2 sgRNA GATACCATTCTTTAACAACC S’μ3 sgRNA AAGGACAGTGCTTAGATCCG S’μ4 sgRNA GTTGAGGCCAGCAGGTCGGC S’μ5 sgRNA AGTTAGTCCAGCCGACCTGC S’μ6 sgRNA GTTGAGAGCCCTAGTAAGCG S’μ7 sgRNA CTGGGCCGCTAAGCTAAACT S'α1 sgRNA GCAGTGGACCCAAAGACGAG S'α2 sgRNA GCTTGGAAGTTACACTGGCG Cas9- S'α3 sgRNA GGACCAGAACTGGTGGGTCT mediated S'α4 sgRNA GTTCTGGTCCTCCTAACCCT CSR S'α5 sgRNA TGGAGCGCTAGACTGCTCAG S'α6 sgRNA CCTTGTGAAAGACTACCTGC S'α7 sgRNA TGTTGGTCAGCTCCGACTGC S’mu_seq_F2 Primer TTTGAGTACCGTTGTCTGGG S’alpha_seq_R2 Primer CAGGTCACATTCATCGTGCC S’γ1 sgRNA TCTCCACTGTTTGCTTAACC S’γ2b sgRNA CCACATTAGCACTATTAGGG S’γ3 sgRNA TCAGGGCCTGTCTAACCCAC S’ε sgRNA CTGATTCCTTACACCCCGCC Gene Lig4_G2 sgRNA TTTCTGTATCCGTTCTAGTG knockout Inversional Igha3UTR_R2.3 (StuI) sgRNA GTCTTAGCTCTGAAACGTGG Cas9- Igha3UTR_R3.10 sgRNA CGTTGGGCAGGGTAAACTCA mediated (KpnI) switching C sgRNA CAAGTACCTAGCCACCTCGC Igha_3UTR_F Primer CCTACTGAGCTTGTTCTACA

90 91

Trans Cas9- mediated Igha_3UTR_R Primer ACATTCCACTACACATTGTT switching

A Bone marrow gating strategy

B Spleen gating strategy

C Peritoneal cavity gating strategy

D Thymus gating strategy

e 100 40 c

E n F

s a

80 l r

l 30

e c

n +

e 20

P p

l e G 10

d 20

n i

0 0 %

1 2 1 1 2 1 2 1 2 1 2 1 1 2 1 2 1 2 1 2

2 3

d d

l l

h h

S S

5 5

Appendix Figure 5. Gating strategies for the assessment of B and T cell populations in WT and Shld2−/− mice. (A) Gating strategies used to quantitate Hardy fractions A-F of progenitor B cells in bone marrows of WT and Shld2−/− mice. (B) Gating strategies used to quantitate immature, mature, T1, T2, MZ, and FO B cell populations in the spleen of WT and Shld2−/− mice. (C) Gating strategies used to quantitate B1, B2, B1a, B1b populations in the peritoneal cavity of WT and Shld2−/− mice. (D) Gating strategies used to quantitate various thymocyte populations in WT and Shld2−/− mice. (E) Insertion-deletion (indel) penetrance was measured by TIDE sequencing for the lentiCRISPRv2 constructs expressing the indicated sgRNAs targeting the 53bp1, Shld1, Shld2, Shld3, and Lig4 genes. (F) Baseline GFP frequency of bulk gene-edited A70.2 cells prior to imatinib stimulation (Figure 5F).

Appendix Figure 6. Sterile transcript or AID protein levels were unaffected in Shld2−/− B cells. (A) Purified splenic B cells from WT and Shld2−/− mice were unstimulated or stimulated for 2 days with LPS + IL4, and germline (sterile) transcripts for Iμ (left panel) and Iγ1 (right panel) were quantitated by qPCR and compared to HPRT mRNA levels. Data is shown relative to WT, which is set at 1. (B) Lysates of purified splenic B cells from WT and Shld2−/− mice that were stimulated for 3 days with LPS + IL4 and subjected to western blot analyses for AID and the internal control β-actin.

Appendix Figure 7. Germinal center B cell frequency is not affected by SHLD2-deficiency. (A) Germinal B cell (GL-7+ Fas+) frequency relative to all B cells (B220+) in the spleen in unimmunized mice; mean ± SD of 3 or 4 biological replicates, ns P ≥ 0.05, unpaired two-tailed t- test. (B) As in A, but at day 10 post NP-CGG immunization; mean ± SD of 4 or 5 biological replicates, ns P ≥ 0.05, unpaired two-tailed t-test. (C) As in A, but at day 21 post NP-CGG

96 immunization; mean ± SD of 3 or 4 biological replicates, ns P ≥ 0.05, unpaired two-tailed t-test. (D) Representative flow plots of fluorescence minus one (FMO) controls and day 21 data points.

Appendix Figure 8. SHLD2- and 53BP1-deficiency has no effect on Cas9-mediated switching. (A) Schematic of the mouse immunoglobulin heavy chain locus and Cas9/sgRNA targeting for CSR. sgRNAs (depicted as arrows) were designed to target 5′ of Sμ and 3′ of Sα. Recombination between DSBs generated at Sμ and Sα results in class switching to IgA. The schematic is not to scale. (B) Cas9-induced switching was carried out on Aid−/−, Aid−/− Lig4−/−, Aid−/− 53bp1−/−, and Aid−/− Shld2−/−/− CH12 clones and switching to IgA was measured 3 days post transfection. The percent of Iglo cells is also reported. Transfection with the empty vector pX330 served as negative control. Values are mean frequency ± SD of 3 biological replicates; * P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.001, **** P ≤ 0.0001, two-way ANOVA with post hoc Dunnett’s test.

Xlf #1 (586/A05) 1 V S Q H L I H P L M G V S L A L Q S H V R E L A A L L R M K D L E I Q A Y Q E S G A WT GTCTCTCAGCATTTGATTCATCCTCTGATGGGTGTGAGCCTGGCACTGCAGAGTCATGTGAGGGAGCTAGCAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT |||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Allele 1 GTCTCTCAGCATCTGATTCATCCT----TGGGTGTGAGCCTGGCACTGCAGAGTCATGTGAGGGAGCTAGCAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT c.25_28delCTGA

V S Q H L I H P L M G V S L A L Q S H V R E L A A L L R M K D L E I Q A Y Q E S G A WT GTCTCTCAGCATTTGATTCATCCTCTGATGGGTGTGAGCCTGGCACTGCAGAGTCATGTGAGGGAGCTAGCAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT |||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Allele 2 GTCTCTCAGCATTTGATTCA------ATGGGTGTGAGCCTGGCACTGCAGAGTCATGTGAGGGAGCTAGCAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT c.21_27delTCCTCTG

Xlf #2 (586/B10) 1 V S Q H L I H P L M G V S L A L Q S H V R E L A A L L R M K D L E I Q A Y Q E S G A WT GTCTCTCAGCATTTGATTCATCCTCTGATGGGTGTGAGCCTGGCACTGCAGAGTCATGTGAGGGAGCTAGCAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT |||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||| Allele 1 GTCTCT------CAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT c.7_87del

V S Q H L I H P L M G V S L A L Q S H V R E L A A L L R M K D L E I Q A Y Q E S G A WT GTCTCTCAGCATTTGATTCATCCTCTG-ATGGGTGTGAGCCTGGCACTGCAGAGTCATGTGAGGGAGCTAGCAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT ||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Allele 2 GTCTCTCAGCATTTGATTCATCCTGTGAATGGGTGTGAGCCTGGCACTGCAGAGTCATGTGAGGGAGCTAGCAGCATTGCTTCGGATGAAGGACCTTGAGATCCAGGCCTACCAGGAGAGTGGGGCT c.27_28insA

Xrcc4 #1 (999/15) 1 M E R K V S R I Y L A S E P N V P Y F L Q V S W E R T I G S G F V I T L T D G H S A W WT ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACCTTATTTTCTGCAAGTGTCTTGGGAGAGAACAATAGGATCCGGCTTTGTTATTACACTTACTGACGGCCATTCAGCCTGG ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||| Allele 1 ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACCTTATTTTCTGCAAGTGTCTTGGGAGAGAACAATAG-ATCCGGCTTTGTTATTACACTTACTGACGGCCATTCAGCCTGG c.86delG

M E R K V S R I Y L A S E P N V P Y F L Q V S W E R T I G S G F V I T L T D G H S A W WT ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACCTTATTTTCTGCAAGTGTCTTGGGAGAGAACAATAGGATCCGGCTTTGTTATTACACTTACTGACGGCCATTCAGCCTGG |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||| Allele 2 ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACCTTATTTTCTGCAAGTGTCTTGGGAGAGAACAAGA—-ATCCGGCTTTGTTATTACACTTACTGACGGCCATTCAGCCTGG c.83_86delinsGA

Xrcc4 #2 (999/31) 1 M E R K V S R I Y L A S E P N V P Y F L Q V S W E R T I G S G F V I T L T D G H S A W WT ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACCTTATTTTCTGCAAGTGTCTTGGGAGAGAACAATAGGATCCGGCTTTGTTATTACACTTACTGACGGCCATTCAGCCTGG ||||||||||||||||||||||||||||||||||||||||||||||||| || ||| ||| |||||||||| |||||||||||||||||||||||||||||||||||||||| Allele 1 ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACTTTTTTTACTGGAAGTGTCATGA------CGGCTTTGTTATTACACTTACTGACGGCCATTCAGCCTGG c.72_89delinsA

M E R K V S R I Y L A S E P N V P Y F L Q V S W E R T I G S G F V I T L T D G H S A W WT ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACCTTATTTTCTGCAAGTGTCTTGGGAGAGAACAATAGGATCCGGCTTTGTTATTACACTTACTGACGGCCATTCAGCCTGG ||||||||||||||||||||||||||||||||||||||||||||||||||| | |||| | ||||| || Allele 2 ATGGAAAGGAAAGTAAGCAGAATCTATCTTGCTTCTGAACCCAACGTACCTGAAATTCTTCTAGTGTATTA------TTATTACACTTACTGACGGCCATTCAGCCTGG c.71_97delinsA

Paxx #1 (275/8) 1 M A P P L L S L P L C I L P P G S G S P R L V C Y C E R D S G G D G D R D D F N L WT ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGCTACTGCGAGCGGGATAGTGGTGGAGACGGGGACCGCGACGACTTCAACCTC ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| | |||||||| || ||| ||||||||||| Allele 1 ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGGTGCTGCGAGCCGGGTAG------ACTTCAACCTC c.90_112del

M A P P L L S L P L C I L P P G S G S P R L V C Y C E R D S G G D G D R D D F N L WT ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGCTACTGCGAGCGGGATAGTGGTGGAGACGGGGACCGCGACGACTTCAACCTC |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| | | | ||| | | |||||||||||||||||||||||||||||||||| Allele 2 ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGCAAACGGGGACGGCA-ACTGGTGGAGACGGGGACCGCGACAACTTCAACCTC c.73_89delinsAAACGGGGACGGCAAC

Paxx #2 (275/14) 1 M A P P L L S L P L C I L P P G S G S P R L V C Y C E R D S G G D G D R D D F N L WT ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGCTACTGCGAGCGGGATAGTGGTGGAGACGGGGACCGCGACGACTTCAACCTC ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| || |||||||||||||||||||||||||||||||||| Allele 1 ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGCTACTGCGAGAGG-----TGGTGGAGACGGGGACCGCGACGACTTCAACCTC c.85-89delGATAG

M A P P L L S L P L C I L P P G S G S P R L V C Y C E R D S G G D G D R D D F N L WT ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGCTACTGCGAGCGGGATAGTGGTGGAGACGGGGACCGCGACGACTTCAACCTC |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||| Allele 2 ATGGCTCCTCCGTTGTTGTCGCTGCCGCTTTGTATTCTGCCGCCGGGTTCGGGCTCCCCCCGCCTGGTGTGCTACTGCGAGC------CGGGGACCGCGACGACTTCAACCTC c.83_98delGGGATAGTGGTGGAGA

Ku70 #1 (864/31) n/a

Ku70 #2 (980/6) WT AGAGGATCATGCTGTTCACCAATGAAGACGACCCCCATGGCCGTGACAGTGCTAAAGCCAGCCGGGCCAGGACCAAAGCCAGCGACCTCCGGGACACTGGTGGGCACTTC ||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Allele 1 AGAGGATCATGCTGTTCACCAATGAAGACGA-CCCCATGGCCGTTACAGTGCTAAAGCCAGCCGGGCCAGGAGCAAAGCCAGCGACCTCCGGGACACTGGTGGGCACTTC c.485delC

Allele 2 n/a

WT Ku70 #1 #2

Appendix Figure 9. Genotypes of novel mutant CH12 cells generated for this study. WT CH12 cells were used to knockout the indicated genes through CRISPR-Cas9 using the sgRNAs listed in Appendix Table 3. Sequencing results for each of these mutants are shown.

Appendix Figure 10. CSR induces a permanent loss of Ig expression in CH12 cells. (A) The indicated NHEJ-mutant CH12 clones were stimulated with CIT and analyzed by flow cytometry for IgM and IgA expression at day 3 and 7. (B) WT, 53bp1−/−, Shld2−/−/−, and Shld3−/−

100

CH12 clones were stimulated with CIT for 3 days. The IgM+, IgA+, and Iglo populations were sorted and reanalyzed for expression of IgM and IgA 12 days post sort. Shown on bar graphs are sorted IgM+, IgA+, and Iglo populations (each column) from WT and mutant CH12 clones, and the percent of cells expressing IgM, IgA, or low for both isotypes (Iglo) after 5 days of culture post- sort.

101

Appendix Figure 11. DC-DDPCR assay validation (A) Unstimulated CH12 clones were assayed for inversional DC events as in Figure 9. (B) PCR with a gradient of annealing temperatures with Chrnb1 DC primers on digested-circularized genomic DNA. (C) Like B, but with IgA deletional DC primers. (D) Like B, but with IgA inversional DC primers. (E) DC-PCR with IgG1 deletional and inversional primers with an annealing temperature of 60 °C.

102

Appendix Figure 12. Flow Cytometry analysis of Iglo WT, 53bp1−/− and Shld2−/−/− CH12 subclones. Iglo cells from WT, 53bp1-/- and Shld2−/−/− CH12 cells were sorted and subcloned and reanalyzed for expression of IgM and IgA by flow cytometry. As positive controls, expression of IgA and IgM are shown for two specific subclones that are IgA or IgM positive.

103

Appendix Figure 13. Repair junctions in Iglo cells have characteristics of alternative end joining. (A) Categorization of Ighm-Igha junctions from sequences in Figure 11D, separated into junctions incorporating DNA insertion, microhomology, or neither in the form of direct joins. (B) Microhomology distribution of Ighm-Igha junctions from sequences in Figure 11D, except for those with insertions (26 independent sequences from one experiment). Median and lower/upper quartiles defined by box, and whiskers represent minimum and maximum values.

104

Appendix Table 2. Primers used in Chapter 3.

Name Sequence

5′ Iμ germ line transcript GAACATGCTGGTTGGTGGTT

3′ Iμ germ line transcript TCACACAGAGCATGTGGACT

5′ Iγ1 germ line transcript CAGGTTGAGAGAACCAAGGAAG

3′ Cγ1 germ line transcript AGGGTCACCATGGAGTTAGT

5′ Iα germ line transcript GGGACAAGAGTCTGCGAGAA

3′ Cα germ line transcript TCAGGCAGCCGATTATCACT

5′ HPRT CCCAGCGTCGTGATTAGC

3′ HPRT GGAATAAACACTTTTTCCAAAT

F5μ GAACATGCTGGTTGGTGGTT

F2μ CAGTCCACATGCTCTGTGTG

SμR CGGCCCAGCTCATTCCAGTTCATTACAG

F3.4α GTTCTGGTCCTCCTAACCCT

R2α GGGTCTGAAGAGGACACAAATAC

R3α CCTGGAGTTGGGTTATGTCTTC

F-AID CAACTCAGACCGCTCTCTCC

R-AID GTGCAGCTTTCCTTTACCCAAC

Chrnb1 F2 CACACAAACCACTAAACTACTCAC

Chrnb1 R2 CCCACCCATCCTAACACTTT

105

Chrnb1 F3 CCACTAAACTACTCACTATCTGTTGT

Chrnb1 R3 ATCCTAACACTTTGGTGATAGAGG

IgA_del F1 ACTGGGAAGGATGCAGTGC

IgA_del R1 TGCTTCGGGTATTGGAAAAGA

IgA_del F2 GGGAGCTGTCTTCACCTGG

IgA_del R2 TGGAGACCAATAATCAGAGGGA

IgA_del F3 AGGATGCAGTGCAGAAGAAA

IgA_del R3 GGAGACCAATAATCAGAGGGAAG

IgA_invA F1 GAGACAAAGGGCTGTAGGAAA

IgA_invA F2 ATGCCTGTCTGTCTCCTTATTC

IgA_invA F3 AAAGGCTGGGTTCTGTCTG

IgA_invA F4 TCTGTCTCCTTATTCTTCAGAGA

IgG1_del F4 GTTACAGGTCAAGGCTGAGTAG

IgG1_del R4 CACTGTAAATGCTTCGGGTATTG

IgG1_del F5 CTGGGTAGGTTACAGGTCAAG

IgG1_del R5 GACCAATAATCAGAGGGAAGAATAATAG

IgG1_del F6 GGGATAACAAGGCTAAGAACAC

IgG1_del R6 GAATTGAATGGAGACCAATAATCAG

IgG1_invA F1 AATGGTGCCCAACATGGA

106

IgG1_invA R1 GGAGACCAATAATCAGAGGGAAG

IgG1_invA F2 GGGCATATTTGTGATGGAGAGA

IgG1_invA R2 CACTGTAAATGCTTCGGGTATTG

IgG1_invA F3 GCCCAACATGGAGCTTTGA

IgG1_invA R3 GAGACCAATAATCAGAGGGAAGAATAATAG

IgG1_invA F4 GATGGAGAGATTTACTGCAACA

IgG1_invA R4 GAATTGAATGGAGACCAATAATCAG

107

Appendix Table 3. Cas9 sgRNA used in Chapter 3.

Name Sequence

Xlf GCCAGGCTCACACCCATCAG

Ku70 #1 TCCGAGACACGGTTGGCCAT

Ku70 #2 ACCAATGAAGACGACCCCCA

Ku80 AATGGCGAGCCTGGCGAGAG

Xrcc4 TGGGAGAGAACAATAGGATC

Paxx TACTGCGAGCGGGATAGTGG

Lig4 #1 TTTCTGTATCCGTTCTAGTG

Lig4 #2 ACAAAGATGGCGCGCTGTAC

Ctrl #1 (G. gallus Aicda) GTAGGAAGAGAACCTCCACA

Ctrl #2 (G. gallus Aicda) CTACTTCTGTGAAGATCGCA

Shld1 ACACACCGCGGGTAGATCCA

Shld2 #1 AACCTGAGTGATATGACTAG

Shld2 #2 ACGTTTTGACGACTTCTGTG

Shld3 #1 GACTCATCGTATGGAAACCA

Shld3 #2 GGAAGTTTGGACTCATCGTA

Shld3 #3 AGTGAAGGAGCAGACCAATG

53bp1 #1 CAGTTGGTGACCACTAACTC

53bp1 #2 TTCTAGCCCGCTATCTGATG

108