The Mechanisms of DNA Double Strand Break Repair and Mismatch Recognition a Dissertation

Home , DnaE

The Mechanisms of DNA Double Strand Break Repair and Mismatch Recognition A Dissertation

Presented to

The Faculty of the Graduate School of Arts and Sciences Brandeis University

Graduate Program in Molecular and Cell Biology

James E. Haber, Advisor

In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

Danielle Nicole Gallagher

February 2021

The signed version of this form is on file in the Graduate School of Arts and Sciences. This dissertation, directed and approved by Danielle Gallagher’s Committee, has been accepted and approved by the Faculty of Brandeis University in partial fulfillment of the requirements for the degree of:

DOCTOR OF PHILOSOPHY

Eric Chasalow, Dean Graduate School of Arts and Sciences

Dissertation Committee:

James E. Haber, Biology Dept.

Lizbeth Hedstrom, Biology and Chemistry Dept.

Susan T. Lovett, Biology Dept.

Sue Jinks-Robertson, Cell and Molecular Biology Dept., Duke University

2021

iii Acknowledgements

I cannot accurately express my gratitude to all of the wonderful mentors that I have had, not only throughout graduate school, but throughout my life. Among these, I would specifically like to thank Bill (Woody) Woodrum and Lori Woodrum, 4H leaders in my home community who have encouraged me since I was 11 years old to believe in myself and pursue higher education, even when it seemed like an impossibility.

To my wonderful Haber lab, thank you for such a supportive and stimulating environment. To Neal and Miyuki, thank you for keeping the lab running smoothly. To Gonen

Memisoglu, David Waterman, Brenda Lemos-Waterman, and Annette Beach, thank you for your patience and mentorship when I first joined the lab. A very special thank you to Jim Haber for being such an incredible advisor. Your passion and kindness has not only made me a better scientist, but a better mentor for others.

I would like to thank my family for always supporting me and my goals, even when they do not fully understand them. To my mom, Barbara Gallagher, thank you for a lifetime of encouraging words and unparalleled kindness. To my father, John Gallagher, thank you for teaching me the value of hard work and independence. Thank you to my big brothers, Jimmy

Chiapuzio and David Murphy, for all of the love and support - even when we fight.

To Matthew Pack, one of the best people I have ever met, thank you for being my best friend since 6th grade. Your unremitting friendship and support throughout my life means more to me than I could ever express with words. You’re my person.

To my adopted classmates, Brenda Lemos-Waterman, Laura Laranjo, Chloe Greppi, Rylie

Walsh and Meghan Harris, thank you for being my support system throughout this crazy ride.

For my father, John Franklin Gallagher III who passed on during my time at Brandeis

Tanyán yá yo. Tókša akhé wanchíyankin kte.

v Abstract The Mechanisms of DNA Double Strand Break Repair and Mismatch Recognition

A dissertation presented to the Faculty of the Graduate School of Arts and Sciences of Brandeis University Waltham, Massachusetts

By Danielle Nicole Gallagher

DNA double-strand breaks (DSBs) are among the most toxic forms of DNA damage and pose a severe threat to genomic integrity. As such, the cell has evolved highly coordinated and complex mechanisms to repair such lesions. Improper repair of DSBs can lead to chromosomal duplication, chromosomal deletions, and chromosomal translocations, all of which are hallmarks of human cancers. Previous work in DNA damage has focused on the mechanistic characterization of how cells repair DNA DSBs, and my research builds on this foundation. In this work, I study repair of DNA DSBs made with the site-specific nucleases HO and CRISPR/Cas9 to show repair via single-stranded DNA, a common method of gene editing, utilizes an uncharacterized Rad51-independent DNA repair pathway that is dependent on Rad52, Srs2, and the MRX complex, but independent of other canonical repair proteins. We also find that Rad59 plays a significant role in the process by alleviating Rad51’s inhibition of Rad52 via the Rad51 homologs. Furthermore, genome-wide genetic screening suggests that this pathway also utilizes proteins that are traditionally components of nucleotide excision repair and telomere recombination. Additionally, I investigate the effect of heterologous templates during the more conventional DSB repair pathway, gene conversion. Here, we show that there is an inherent asymmetry in DSB repair, as correction of mismatches templated upstream of the DSB are mechanistically different than those templated downstream of the DSB. While the activity of polymerase 훿 is primarily responsible for incorporating mismatches into the recipient locus on the left side of the DSB, mismatches templated on the right side of the DSB are primarily corrected via components of the mismatch repair pathway (MMR). These corrections patterns, however, are heavily influenced by the nature of the DSB itself. Collectively, these results highlight the immense complexity of DNA DSB repair and offer insights into the mechanisms of DNA repair, as well as to the field of genome engineering.

Table of Contents

Chapter 1: Introduction ...... 1

1.1 – Types of Site-Specific Endonucleases ...... 1 HO-Endonuclease ...... 1 I-SceI ...... 3 CRISPR-Cas9 ...... 4

1.2 – Types of Double Strand Break Repair ...... 6 Nonhomolgous End Joining (NHEJ) ...... 7 Homologous Recombination (HR) ...... 9 5’ to 3’ Resection ...... 9 Single Strand Annealing ...... 11 Formation of the Rad51 Filament and the Homology Search ...... 11 Break-Induced Replication (BIR) ...... 13 Rad51-Independent BIR ...... 14 DSB repair via Gene Conversion ...... 15 Double Holliday Junction ...... 16 Synthesis-Dependent Strand-Annealing ...... 16

1.3 – Mismatch Repair Pathway ...... 17 Mismatch Repair in HR...... 22

References ...... 26

Chapter 2: A Rad51-Independent Pathway Promotes Gene Editing

Abstract ...... 41

Introduction...... 42

Results...... 45 Single stranded template repair is Rad51-independent ...... 45 Rad59 regulates Rad52-mediated strand annealing ...... 49 Pol is responsible for target-adjacent mutagenesis ...... 52 Additional genetic requirements of SSTR depend on template design...... 55 Genetic requirements of SSTR are identical using Cas9 ...... 57 Mismatch repair acts differently at the 5’ and 3’ ends of the ssODN ...... 60

vii Effect of modifying the 5’ and 3’ ends of the ssODN...... 64 SSTR is specific to single-stranded DNA ...... 66

Discussion ...... 68

Methods ...... 74 Strain List ...... 77

References ...... 81

Chapter 3: A Genome-Wide Screen to Characterize a Novel Rad51-Independent Pathway Utilized in Genome Engineering

Abstract ...... 89

Introduction ...... 90

Results...... 93

Discussion ...... 97

Methods ...... 99 Strain List ...... 102

References ...... 135

Chapter 4: The Asymmetry of Mismatch Correction of Divergent Substrates Following Double-Strand Break Repair

Abstract ...... 140

Introduction ...... 141

Results...... 146 Symmetrical mismatches do not affect mismatch incorporation patterns ..152 Pattern of mismatch incorporation is influence by endonuclease cleavage .158

Discussion ...... 160

Methods ...... 163 Strain List ...... 165

viii

References ...... 169

Chapter 5: Future Directions ...... 175

ix Table of Figures

Chapter 1 Figure 1.1 - Arrangement of the mating type locus...... 2 Figure 1.2 - I-SceI mediated gene drive ...... 3 Figure 1.3 - Cas9-mediated DNA cleavage ...... 4 Figure 1.4 - Regulation of repair pathway choice ...... 7 Figure 1.5 - Pathways of DSB repair ...... 9 Figure 1.6 - Models of eukaryotic mismatch repair ...... 21 Figure 1.7 - Clipping of nonhomologous tails ...... 24

Chapter 2 Figure 2.1 - SSTR is a novel form of Rad51-independent DSB repair ...... 47 Figure 2.2 - Rad59 alleviates the inhibition of Rad51 on Rad52’s annealing activity ...... 50 Figure 2.3 - Rad59 interacts with Rad51 via the Rad51 homolog Rad55 ...... 52 Figure 2.4 - Target-adjacent mutagenesis is dependent on Pol ...... 53 Figure 2.5 - The genetic requirements of SSTR are dependent upon template design ...... 56 Figure 2.6 - SSTR initiated by retron-Cas9 utilizes a Rad51-independent repair pathway ...59 Figure 2.7 - Gene editing events are dependent on activity of MSH2 ...... 62 Figure 2.8 - Model of heteroduplex formation and incorporation of mismatches ...... 65 Figure 2.9 - SSTR is distinct from DSTR ...... 67 Figure 2.10 - Model of SSTR ...... 69

Chapter 3 Figure 3.1 - Retron generated single stranded DNA ...... 95 Figure 3.2 - Increasing ssDNA donor length does not increase SSTR efficiency ...... 96

Chapter 4 Figure 4.1 - Basic gene conversion set-up ...... 146 Figure 4.2 - Mismatches and MMR machinery influence GC repair outcome ...... 148 Figure 4.3 -Effect of MMR and DNA repair mutants on viability ...... 149 Figure 4.4 - Sequencing shows different mechanisms for mismatch correction ...... 151 Figure 4.5 - Strain reconstruction with symmetrical mismatches ...... 153 Figure 4.6 - Effect of DNA repair mutants on viability and mismatch correction ...... 155 Figure 4.7 - Presence of a nonhomologous tail has different effects on GC ...... 157 Figure 4.8 - Pattern of mismatch incorporation is heavily influenced by DSB ...... 159

x Table of Tables

Chapter 2 Table 2.1 - Strains used in these experiments ...... 77 Table 2.2 - Plasmids used in these experiments ...... 80

Chapter 3 Table 3.1 - SSTR screen hits ...... 98 Table 3.2 - Strains used in these experiments ...... 102 Table 3.3 - Plasmids used in these experiments ...... 102 Table 3.4 - L1 gene drop-outs extended data table...... 103 Table 3.5 - L2 gene drop-outs extended data table...... 119

Chapter 4 Table 4.1 - Strains used in these experiments ...... 165 Table 4.2 - Plasmids used in these experiments ...... 168

xi Chapter 1: Introduction

The accurate transfer of genetic information is the fundamental component of life, allowing for the genetic code to be transmitted from one generation to the next. For billions of years,

DNA-based life has existed on Earth, all the while the genetic code has been assaulted by an onslaught of both endogenous and exogenous sources of damage, including, but certainly not limited to, reactive oxygen species, ionizing radiation, and the implicit stress of DNA replication.

The DNA double strand break (DSB) is one of the most lethal forms of DNA damage that can occur within the cell. Repair of these deleterious lesions is vital to maintain genomic integrity, while inadequate repair of DSBs can result in chromosomal deletions, duplications, and nonreciprocal translocations – all of which are hallmarks of cancer. Cells have evolved two primary pathways to repair DNA DSBs: 1) nonhomologous end joining (NHEJ), a highly mutagenic form of repair usually resulting in small base pair (bp) insertions or deletions (indels), and 2) homologous recombination (HR), where the DSB is repaired using homologous sequences as a template for repair. The mechanisms of DNA DSB repair have been elucidated using a variety of tools, both in vitro and in vivo; however, the majority of these assays make use of site-specific endonucleases to create a DSB.

1.1– Site-Specific Endonucleases

HO Endonuclease

Saccharomyces cerevisiae have an endogenous endonuclease, HO, that recognizes and cleaves a specific 24 bp sequence on chromosome 3, leaving 4-nt 3’ overhangs (Nickloff et al.,

1986; Butler et al., 2004). Normally, repair of an HO-induced DSB occurs through gene

1 conversion, a form of HR discussed in-depth below, using one of two donors HML and HMRa,

located >100 kb away (Figure 1.1.). These donor sequences are normally maintained via highly

positioned nucleosomes, rendering them transcriptionally inactive, and prevent cutting of the

HO-recognition sequence in each donor. Cleavage in the donors is generally only seen when

their silencing, via the Sir2 histone deacetylase, is disabled (Lee and Haber, 2015), although

more sensitive assays have revealed transient unsilencing during DSB repair (Sieverman and

Rine, 2018). When HO is placed under a galactose-inducible promoter, cleavage of the

recognition sequence is quite rapid, with >90% of cells being cleaved in 30 minutes (Jensen et

al., 1984; Lee et al., 2015). Since the HO-recognition sequence can be inserted into any

euchromatic region in the nucleus and maintain the same level of cleavage efficiency, it has

been used to characterize many facets of DSB repair in budding yeast. However, very little` use

of HO has been made in metazoans.

Figure 1.1 - Arrangement of the mating type locus. The MAT locus is divided into five regions, W, X, Y, Z1 and Z2 (as shown). The two mating type alleles (a and 훼) differ by approximately 700 bp. The recombination enhancer (RE) promotes HML as the donor in MATa cells. This region is silenced in MAT훼 cells. A) Gene conversion from MATa to MAT훼. B) Gene conversion from MAT훼 to MATa. Figure modified and adapted from Haber, 2016.

2 I-SceI Nuclease

I-SceI has been characterized extensively in vitro, and has thus been used much more extensively in metazoans (Perrin et al., 1993; Joshi et al., 2011; Prieto, et al., 2016). The I-SceI endonuclease has an 18-bp recognition sequence. When placed under the same galactose- inducible promoter, I-SceI has a reasonable cutting efficiency, with DSBs observed via Southern blot 60-90 minutes after induction (Plessis et al., 1992). However, like HO, I-SceI makes 4-nt 3’- overhanging ends (Perrin et al., 1993). I-SceI is endogenously found in the mitochondria of budding yeast, encoded within an intron designated omega (+) within the 21S rDNA. When an

+ strain is crossed to an - strain, there is a very efficient conversion of the - mitochondrial genome to a sequence containing the intron (Figure 1.2) (Jacquier et al., 1985; Zinn et al.,

1985). The conversion happens by site-specific cleavage of I-SceI within the - 21S rDNA sequence and then utilizing the + as a repair template to copy via homologous recombination.

I-SceI cannot cleave the inserted + sequence,

preventing further I-SceI DSBs. This system was the

first well-studied example of gene drive.

I-SceI was “domesticated” for further use in

other organisms by optimizing the coding sequence

of the gene and expressing it in the nucleus so that Figure 1.2 - I-SceI mediated gene drive. When a + yeast mates with a - strain, I-SceI, encoded within the intron, cleaves its target site in the - locus and promotes repair from the + copy. The exact mechanism of this has not been well- characterized. Figure modified and adapted from Gallagher and Haber, 2018.

3 it was dependent on traditional cytoplasmic protein synthesis (Colleaux et al., 1986). In mammalian systems, I-SceI became the primary technique to induce site-specific DSBs and to study the mechanisms of DNA repair.

CRISPR-Cas9

Recently, a form of adaptive immunity was discovered in the bacteria Streptococcus pyogenes. In bacterial cells, this system contains DNA sequences specific to phages and other foreign DNA and is flanked by repeated regulatory sequences (Clustered Regularly Interspaced

Short Palindromic Repeats – CRISPR). These incorporated foreign sequences serve as a guide

(gRNA) to a CRISPR associated protein (Cas9 in S. pyogenes and the protein used throughout this work), to induce site-specific DNA DSBs (Barrangou et al., 2007; Brouns et al., 2013). Once a

Figure 1.3 - Cas9-mediated DNA cleavage. Cas9 has two distinct lobes, the recognition lobe consisting of 3 recognition domains (shown as Rec1, Rec2 and Rec3), and the nuclease lobe, containing the HNH domain, the RuvC domain, and the variable C-terminal domain (CTD) (not shown). Upon locating a PAM sequence (shown in orange), and gRNA recognition (shown in blue) the CTD restructures, placing the HNH domain in position to make the first cleavage of the target strand (dark green triangle). Cleavage by the HNH domain causes a conformational change, placing RuvC in position to cleave the nontarget strand (light green triangle) (Chen et al., 2014; Jinek et al., 2014; Jiang et al., 2016). Figure adapted from crystal structures shown in Nishimasu et al., 2014.

4 20-nt gRNA has been loaded into the Cas9 protein, the nuclease scans the DNA for a protospacer adjacent motif (PAM), which is a 5’-NGG-3’ in this system where N can be any of the four nucleotides (Gasiunas et al., 2012; Jinek et al., 2012). Once a PAM is identified, Cas9 melts the surrounding DNA to hybridize the unique gRNA sequence to the exposed genomic ssDNA sequences (Figure 1.3) (Nishimasu et al., 2014; Sternberg et al., 2014). Once perfect complementarity between the genomic DNA and the gRNA has been found, Cas9 undergoes a conformational change to cleave the strand complementary to the gRNA sequence, called the target strand, via the HNH domain, followed by cleavage of the non-target strand via the RuvC domain (Figure 1.3) (Sternberg et al., 2015; Jiang et al., 2016; Shibata et al., 2017).

The versatility of the CRISPR-Cas9 has completely revolutionized gene editing and has rapidly been adapted to create designer DSBs in most model systems (Shen et al., 2013; Mali et al., 2013; DiCarlo et al., 2013). Inducing a site-specific DSB in the presence of either a single- stranded (ssDNA) or double-stranded (dsDNA) donor template allows for a wide range of gene editing – insertions of designer sequences, deletions of genomic regions, and correction of mutations (Ran et al., 2013; Doudna and Charpentier, 2014; Hsu et al., 2014). In the absence of a repair template, a Cas9 DSB is repaired via NHEJ, usually resulting in small +/- 1 bp indels; one prominent indel is a +1 templated insertion resulting from Cas9 cleavage yielding a 1-nt 5’ overhang rather than a blunt end (Lemos et al., 2018). This has become a common practice to inactivate genes, and a common readout in genome-wide screens in mammalian systems

(Shalem et al., 2014; Wang et al., 2014; Bertomeu et al., 2018; Han et al., 2018; Adelmann et al.,

2019). Additionally, CRISPR-Cas9 has already been used to introduce mutations in human embryos, some of which have been brought to full term, despite the highly contested ethical

5 implications (Ma et al., 2017; Cyranoski and Ledford, 2018; Dzau et al., 2018; Lander et al.,

2019). Although in the future CRISPR-Cas9 has the potential to correct the mutations that cause numerous genetic diseases, there is much more work that needs to determine the mechanisms of gene editing, allowing us to prevent potentially harmful outcomes and off-target effects

(Zeng et al., 2018). Off-target effects of gene editing are still highly contested, with some research claims that there are numerous unintended DSBs that occur when using Cas9, while others revel in its specificity (Zhang et al., 2015; Kleinstiver et al., 2016; Iyer et al., 2018; Lin and

Wong et al., 2018).

1.2 – Types of DSB Repair

As previously stated, DSBs can be repaired by either NHEJ or one of several forms of HR, each requiring a distinct set of proteins (reviewed in Haber et al., 2016; Chang et al., 2017;

Wright et al., 2018). Unlike NHEJ, HR requires a homologous donor template. Though the preferred pathway of repair in Saccharomyces cerevisiae, the model organism used throughout the studies discussed below, mammalian cells rarely repair a DSB via homologous recombination, instead predominately repairing DSBs via NHEJ (Heyer et al., 2010). HR only occurs in late S phase or in G2 phase when there are sister chromatids available and when cyclin-dependent kinases allow for resection to take place (Figure 1.4) (Aylon et al., 2004; Ira et al., 2004). Since most spontaneous DSBs occur during replication, DSB repair by HR during G1 is not normally necessary.

Figure 1.4 - Regulation of Repair Pathway Choice. DSB repair is heavily regulated by the cell cycle. The three primary modes of DSB repair are outlined. If the Ku heterodimer, a complex required for NHEJ, dissociates from the DSB ends, there is initiation of end resection. Although this process can occur throughout the cell cycle, it is particularly prominent in G1 when resection is blocked (Ira et al., 2004). Microhomology mediated end joining (MMEJ), possibly facilitated short-range end resection done by the MRX complex, results when exceedingly small regions of homology are used to ligate broken ends, resulting in small indels. After the initiation of S-phase, the activity of Cdk1, the master regulator of the cell cycle, is higher and the rate of end resection increases (Aylon et al., 2004). Long-range resection fully commits the cell to homologous recombination.

Non-homologous End Joining (NHEJ)

NHEJ repair involves direct re-ligation of the broken DSB ends and, as it does not require

a homologous template, it is not believed to be restricted to a specific phase of the cell cycle

(Figure 1.5A). One of the key protein complexes that distinguishes NHEJ repair is the Ku70-Ku80

heterodimer which forms a ring-shaped structure around the DNA helix (Mari et al., 2006;

Zhang et al., 2001). Recognition of broken DNA ends by Ku is the first step in NHEJ repair, and

the heterodimer has been shown to localize to the ends within seconds of laser generated DSBs

7 (Blier et al., 1993). In budding yeast, it is not known whether the MRX (Mre11-Rad51-Xrs2) or

Ku binds the DSB end first, but in mammalian cells the MRX homologue MRN (Mre11-Rad50-

Nbs1) is known to bind after Ku (Stracker et al., 2004). However, in both yeast and mammalian systems, MRX/N tethers both ends of the DNA break, holding the chromosome together

(Ghodke and Muniyappa, 2013).

Once Ku and MRX are bound and protecting the DSB ends from exonucleases, specific ligases are recruited to the damaged site. In S. cerevisiae, DNA ligase IV (Dnl4) and its associated

Lif1 bind Ku80 and Xrs2, aligning the broken ends though the available overhangs, and ligate

(Zhang et al., 2007). An accessory protein, Nej1, is also a required component of NHEJ (Frank-

Vaillant and Marcand, 2001; Kegel et al., 2001; Ooi et al., 2001; Valencia et al., 2001). Nej1 binds to both Lif1 and Dnl4, aiding in stabilization of the DSB (Deshpande and Wilson, 2007).

In mammalian systems the process is more complex, where a phosphatidylinositol-3 kinase-like-kinase DNA-PKcs binds to Ku70-Ku80, causing the Ku complex to translocate inward, activating the DNA-PKcs kinase activity. This activity recruits Artemis, a nuclease protein that processes the DSB ends of chemical modification or mismatching overhangs (Douglas et al.,

2005; Drouet et al., 2006). After processing DNA ligase IV (Lig4) and XRCC4 are recruited to ligate the chromosomal ends. It is worth noting that Lig4 and XRCC4 can be recruited independently of DNA-PKcs and directly interact with Ku to repair the DSB, bypassing the processing step by Artemis (Drouet et al., 2005). Similar to Nej1 in yeast, the human ortholog

XLF binds to XRCC4-Lig4 to stabilize the break and promote NHEJ (Ahnesorg et al., 2006).

8 Figure 1.5 - Pathways of DSB repair. Grey arrow denote traditionally Rad51-dependent mechanisms of HR, while green arrows signify Rad51-independent mechanisms of repair. A- C) Highly mutagenic forms of DNA repair that result in the loss of genomic integrity. D) Strand invasion into a sister chromatid or ectopic site leads to loss of heterozygosity in diploids, and in haploid organisms BIR can result in a highly mutagenic nonreciprocal translocation. E-F) Two mechanisms of gene conversion, thought to be the DSB repair pathway with the highest fidelity. E) Double Holliday junction involves a second strand invasion event known as second- end capture, which can result in a crossover event, possibly resulting in severe mutagenesis. F) Thought to be the less mutagenic, and predominant form of gene conversion, strand invasion repairs across the break and then reanneals to the broken chromosome. Figure modified and adapted from Mehta and Haber, 2014.

Homologous Recombination (HR)

5’ to 3’ Resection of the DSB ends

The first step in HR is degradation of the DSB ends to create long 3’-ssDNA tails, by 5’ to

3’ end resection. This end-resection step is conserved in all domains of life, but here I will discuss the end-processing machinery involved and its regulation in eukaryotic cells.

9 Through many genetic studies in Saccharomyces cerevisiae, the MRX complex, Sae2,

Exo1, RPA, Sgs1 and Dna2 were identified as key factors for 5’ to 3’ end resection. Their activities are conserved in mammalian systems, with the functional orthologs of MRX being

MRN, Sae2 being CtIP, and Sgs1 being BLM. Sgs1 and Dna2 also form a complex with Top3 and

Rmi1, while mammals have an additional component Rmi2 (Nimonkar et al., 2011; Peterson et al., 2011; Karanja et al., 2012). A widely accepted model on initiation of end resection is that the 5’ ends internal to the DSB undergo endonucleolytic cleavage by the MRX complex and

Sae2, which allow 3’ to 5’ resection of the nicked strand and creates short 3’-ended tails (Ivanov et al., 1994; Gravel et al., 2008; Mimitou and Symington, 2008, 2009; Cannavo and Cejka, 2014;

Cannavo, et al., 2019). The short 3’ ssDNA tails are then extended via two parallel pathways: one dependent on the 5’-3’ exonuclease Exo1, and the other dependent on the helicase/endonuclease complex of Sgs1-Top3-Rmi1 as well as the Dna2 endonuclease

(reviewed in Symington, 2014). The formation of these ssDNA tails can be followed by Southern blotting, which has shown that resection in budding yeast occurs at approximately 1 nt/s

(Fishman-Lobel and Haber, 1992). However, in the absence of a repair template, resection has been shown to continue up to 50 kb away from the site of the DSB (Zhu et al., 2008). It should be noted that the 3’ ssDNA tail is extremely stable, as there are apparently no 3’ to 5’ ssDNA exonucleases in the yeast nucleus. However, harmful secondary structures that might also be cleaved are prevented by the binding of replication protein A (RPA) (Chen et al., 2013).

10 Single-Strand Annealing

Single-strand annealing (SSA) only occurs in repetitive regions of DNA, however, these repeats can be as short as 15-nt, often referred to as microhomology mediated end joining

(MMEJ) (Figure1.5B and 1.5C) (Villarreal et al., 2012). Resection following a DSB exposes complementary sequences, which can anneal to one another and the intervening region clipped away. SSA is considered to be a highly mutagenic form of DSB repair because it results in a deletion where only a single copy of the repeated sequence is retained. The single stranded tails between the two repeats are removed by the Rad1-Rad10 endonuclease (XPF-ERCC1 in mammalian cells) (Sugawara et al., 1997). After tail-clipping the remaining gaps are filled in by

DNA synthesis and ligation.

SSA does not have a strand invasion step, and instead requires the strand annealing activity of Rad52, which is aided by the Rad52-homolog Rad59 (Ivanov et al., 1996; Sugawara et al., 2000). This pathway is also able to occur when the regions of homology are spaced 50kb apart, and thus resulting in a 50kb deletion of genomic DNA (Jain et al., 2009).

Formation of the Rad51 Filament and the Homology Search

Once the 3’-ended ssDNA tail has been generated, the Rad51 protein can bind to form a nucleoprotein filament that is capable of searching the genome for a homologous template.

Once a homologous sequence has been found, Rad51 facilitates strand invasion into that region to initiate repair. Both in vitro and in vivo studies have shown that a single strand binding protein, RPA, first binds to the resected DSB ends (Sugawara et al., 2003; Wang and Haber,

2004; Wolner et al., 2003). RPA is then displaced by Rad51 in a process that requires the

11 mediator Rad52 and the Rad51 paralogs, Rad55-Rad57 (Sugiyama et al., 2002). There is also evidence that that Rad55-Rad57 is incorporated into the filament itself, however, HR can proceed in rad55∆ and rad57∆ strains (Liu et al., 2011).

Much of our understanding of Rad51’s mechanism of action comes from in vitro studies with the bacterial homolog RecA. RecA and Rad51 contain a ssDNA binding site and a double- stranded DNA binding site (dsDNA) . Each RecA protein binds 3-nt of DNA and stretches the helix to 150% of its normal length (Chen et al., 2008; Cloud et al., 2012). In vitro studies have shown that RecA and Rad51 require eight consecutive base-pairings for homology recognition, however, in vivo data has shown that Rad51 only requires 5 consecutive base-pairs of homology for HR to proceed when the total homology length is about 100 bp (Lee et al., 2015; Anand et al., 2017).

Once the Rad51 filament has formed, it can search the genome for a homologous sequence to use as a template for DSB repair. Chromatin immunoprecipitation assays (ChIP) demonstrate that the filament preferentially explores intrachromosomal sites, most likely through a series of collisions that is dictated by the persistence length of the filament, rather than through a sliding mechanism (Renkawitz et al., 2013). This is supported by in vivo studies that show that the kinetics of intrachromosomal repair are significantly faster than interchromosomal repair events (Lee et al., 2016). However, once a donor sequence is located, base pairing between the invading ssDNA with the dsDNA donor can create a displacement loop (D-loop), allowing for the initiation of new DNA synthesis (White and Haber, 1990; Wang et al., 2004).

12 DSB Repair via Break-Induced Replication

Break-induced replication (BIR) occurs when only one end of the DSB shares homology with a donor, such as in collapsed replication forks and the lengthening of telomeres (Doksani and de Lange, 2014). During BIR, strand invasion results in a unidirectional replication fork capable of copying all of the sequences distal to the site of homology (Figure 1.5D) (Davis and

Symington, 2004). Therefore, this repair pathway typically happens close to the telomere and results in loss of heterozygosity, and, when the donor is located on a different chromosome, a nonreciprocal translocation.

The initial steps of BIR, including the homology search and strand invasion, are identical to those of gene conversion (GC) (discussed below) (Davis and Symington, 2004). However, initiating BIR also requires all of the replication factors and all three major DNA polymerases required for leading and lagging strand synthesis (Pol훿, Pol휀, and Pol⍺) (Lydeard et al., 2007,

2010).

Despite its similarity in machinery, BIR is drastically more mutagenic than normal DNA replication (Deem et al., 2011; Anand et al., 2017). Also, BIR shows a high frequency of template switching, indicating instability in the structure (Smith et al., 2007; Anand et al., 2014).

BIR also requires Pol32, a subunit of Pol훿 that increases its processivity, as well as the Pif1 helicase (Lydeard et al., 2007; Wilson et al., 2013). BIR is also significantly decreased by the dominant-negative PCNA (proliferating cell nuclear antigen) mutation pol30-FF248,249AA, which has no effect on normal DNA replication or GC (Lydeard, et al., 2010). PCNA itself is an essential gene and a processivity factor of DNA polymerase 훿 (Prelich et al., 1987; Krishna et al.,

13 1994). Whether these differences are required for the establishment of the repair replication fork or maintaining the D-loop structure is not known.

The kinetics of BIR also provide another distinction from GC, as it shows a dramatic delay in the initiation of new DNA synthesis. This delay is the result of the replication execution checkpoint (REC), which monitors whether only one or both ends of a DSB are able to pair with a donor sequence within a single homologous locus in the proper orientation (Jain et al., 2009).

The kinetics of strand invasion in both a GC and BIR set-up are identical, so the REC acts before the initiation of new DNA synthesis. This delay can still be seen in GC set ups where pairing of the DSB ends with a donor sequence is separated by more than 5 kb (Malkova et al., 2005; Jain et al., 2009; Mehta et al., 2017). This strong kinetic barrier helps promote GC repair following a

DSB.

1. Rad51-Independent BIR

Although less prominent, there are BIR pathways that are Rad51-independent.

Using a donor sequence on a plasmid, this pathway has been characterized to also be

independent of the Rad51 homologs, Rad55 and Rad57. However, Rad52 and the Rad54

homolog Rdh54/Tid1 plays an important role in this pathway, as well as the MRX

complex (Signon et al., 2001; Ira and Haber, 2002). Surprisingly, this pathway only

requires 30 bp of homology to initiate strand invasion, whereas the same assay shows

that Rad51-dependent BIR requires 100 bp (Ira and Haber, 2002). It has also been

observed that yeast cells lacking telomerase (est1Δ or tlc1Δ) can still maintain telomeres

through HR requiring Rad52 (Lundblad and Blackburn, 1993). There are two

independent telomere maintenance pathways, Type I events require canonical DSB

14 repair proteins (Rad51, Rad52, Rad54 and Rad57) (Le et al., 1999; Teng and Zakian,

1999). This pathway frequently involves subtelomereic Y’ sequences. However, Type II

events are Rad51-independent, requiring Rad52, Rad59, MRX and Sgs1 (Le et al, 1999;

Teng et al., 2000; Chen et al., 2001; Cohen and Sinclair et al., 2001; Johnson et al., 2001).

During these events the TG1-3 telomere sequences themselves are extended, either

through intertelomere recombination or via rolling circle replication.

DSB repair via Gene Conversion

In Saccharomyces, MAT switching is best characterized example of gene conversion (GC)

(Figure 1.1). During this process, new DNA synthesis is carried out primarily by polymerase 훿

(White and Haber, 1990; Wang et al., 2004; Hicks et al., 2010). The newly copied strand is displaced and can anneal with the resected ssDNA sequences of the second DSB end, completing the GC event (Hicks et al., 2011; Mehta et al., 2017). Although GC is considered the most faithful pathway to repair a DSB, the mutation rate is still approximately 1000 times higher than that of normal replication (Holbeck et al., 1997; Rattray et al., 2002; Hicks et al.,

2010). Many of these errors are due to polymerase 훿, including slippage events resulting in frameshifts in homonucleotide sequences, but also more dramatic events such as interchromosomal template switching (Hicks et al., 2010; Tsaponina et al., 2014). Additionally, approximately 10%-20% of interchromosomal GCs are accompanied by cross-overs (COs)

(Pâques and Haber 1999).

Two different mechanisms, both of which are extensively supported by experimental data, can explain GC events.

15 1. The Double Holliday Junction Mechanism

In the double Holliday junction (dHJ) model (Resnick, 1976; Szostak et al., 1983),

after the 3’ ended ssDNA within the Rad51 filament undergoes strand invasion and

creates a D-loop, the D-loop is extended by the initiation new DNA synthesis of the 3’

end so that the ssDNA of the opposite side of the DSB can anneal, forming a dHJ

intermediate (Figure 1.5E). Alternatively, both ends of the DSB can undergo strand

invasion and initiate new DNA synthesis, also resulting in a dHJ intermediate (Resnick,

1976; Szostak et al., 1983). A HJ can be resolved via cleavage by an HJ resolvase (Yen1

and Mus81 in Saccharomyces) which, depending on which pair of strand is cleaved, can

result in a noncrossover (NCO) or a CO (Figure 1.5E). Alternatively, dHJs can be

“dissolved” by the BLM helicase/Top3 complex, resulting exclusively in NCOs (Wu and

Hickson, 2003; Ira et al., 2003).

There is extensive evidence for the formation of dHJs. This intermediate structure

has been visualized in meiotic cells (Bzymek et al., 2010). There is also genetic evidence

of dHJs. In ectopic recombination assays where components of the “dissolvases” have

been deleted (Sgs1-Top3-Rmi1), there is an increase in crossover events from 4% to 12%

(Ira et al., 2003). The proportion of COs increases further when another 3’ to 5’ helicase,

Mph1, is also deleted, suggesting that Mph1 has a role in shifting GC events away from

the dHJ recombination pathway (Prakash et al., 2009)

2. Synthesis-Dependent Strand-Annealing Mechanism

Synthesis-dependent strand-annealing (SDSA) (Gloor et al., 1991) begins with D-loop

formation, which is followed by new DNA synthesis initiated by Pol훿 or Pol휀 from the 3’

16 end of the invading strand (Figure 1.5F) (Lydeard et al., 2007; Li et al, 2009, Sebesta et

al., 2011). Pol⍺ primase is not required for SDSA (Wang et al., 2004). SDSA is a unique

form of DNA DSB repair because it is not a semi-conservative process, in which the

newly synthesized strand remains paired with the donor template. Instead, the newly

copied strand is displaced from the migrating replication bubble and pairs with the 3’

ssDNA on the other side of the DSB – a process which results in a NCO product. Studies

have shown that mutations that occur during GC DSB repair are only seen in the

recipient locus, whereas the donor locus remains unchanged (Pâques and Haber, 1999).

Additionally, density transfer experiments have shown that all of the newly synthesized

DNA (“heavy” DNA) was at the recipient locus, whereas the donor sequence remained

unaltered (“light” DNA) (Ira et al., 2006). Both helicases Sgs1 and Mph1 (mentioned

above) appear to have a role in SDSA, in addition to another helicase Srs2, whose

deletion is associated with a 70% drop in viability (Ira et al., 2003). Survivors of the DSB

in these experiments show a much high proportion of CO events, indicating that Srs2 is

required for SDSA, but not required for completion of the dHJ pathway (Ira et al., 2003)

1.3 – Mismatch Repair Pathway

The mismatch repair pathway (MMR) plays a key role in maintaining genomic integrity by repairing base-pair mismatches and indels in the DNA that result from DNA replication, unfaithful DNA DSB repair, or chemical mutagenesis. The mechanism of MMR is best characterized in Escherichia coli (reviewed in Iyer et al., 2006). In E. coli, DNA mismatches are first recognized by the homodimer MutS (Su and Modrich, 1986). Once bound, MutS recruits

17 MutL, which moves down the DNA until it reaches a GATC site (Grilley et al., 1989; Langle-

Rouault et al., 1987; Lahue et al., 1987; Au et al., 1992). Some bacteria can distinguish the original vs newly synthesized DNA strand through a deoxyadenine methylase (dam) that adds a methyl group to the N6 position of the adenine base within the palindromic GATC sequence

(Marinus and Morris, 1973; Geier and Modrich, 1979). After DNA replication occurs in a fully methylated GA*TC/CTA*G sequence, the newly copied strand has not yet been methylated, resulting in a hemimethylated sequence that allows MutH to distinguish between the original strand and the newly synthesized strand containing the mismatch. Dam methylase follows close behind the replication fork to methylate the newly synthesized GATC sequences, meaning that the MMR machinery must act quickly (Schlagman et al., 1986; Campbell and Kleckner, 1990).

Once MutL reaches a hemimethylated GATC site, the site-specific MutH endonuclease is recruited and nicks the newly synthesized, unmethylated strand (Welsh et al., 1987). Once bound, MutL also recruits UvrD, a 3’-5’ helicase which displaces the newly synthesized region of

DNA following the MutH generated nick (Hall et al., 1998; Yamaguchi et al., 1998). The strand containing the mismatched DNA is then degraded by one of several nucleases: ExoVII, RecJ,

ExoI, and ExoX (Cooper et al., 1993; Burdett et al., 2001). The gap is then filled in and the mismatch corrected by DNA polymerase III re-copying the original strand and the nicked site repaired by DNA ligase (Lahue et al., 1989).

The proofreading capacity of the DNA polymerases (polymerases 훿 and 휀) is the first line of defense against base-pair mismatches. Adding a mis-paired nucleotide causes a conformational change in the polymerase as the newly-copied strand is moved to the exonuclease domain (Simon et al., 1991; Morrison et al., 1991). Once in this new

18 conformational position, the polymerase can sense mismatched base-pairs, and, if necessary, remove the offending base by shifting conformations again so that the 3’ DNA end inserts the correct base (Joyce and Steitz, 1994; Swan et a., 2009; Patel et al., 1991) . In E. coli, removing the proofreading capacity of PolIII increases the rate of mutation over 1000-fold (Schaaper,

1993). Similarly, when a pol3-01 allele is used in Saccharomyces cerevisiae, a proofreading- defective allele of polymerase 훿 is elevated 10-100 fold (Morrison et al., 1993; Tran et al.,

1999). Although polymerase proofreading is highly effective, the errors that remain are still too high to maintain genomic stability, and thus the low rates of mutagenesis are due to MMR machinery following the replication fork.

Despite the in-depth mechanistic detail known about the E. coli MMR pathway, it is substantially different from MMR in most other bacteria and eukaryotes – most of which do not have a MutH homolog nor use DNA methylation to distinguish newly synthesized DNA. The use of hemimethylated DNA to differentiate parental vs non-parental strand is limited to E. coli and other closely related gammaproteobacteria, and how this distinction is made in other organisms is not understood (Hiraga et al., 2000; Lobner-Olesen et al., 2005). However, the roles of MutS and MutL, followed by excision and gap-repair are universally conserved.

In eukaryotic systems, there are multiple MutS homologs, which all show functional asymmetry in the heterodimer (i.e. Msh2-Msh6 or Msh2-Msh3) with one subunit directly recognizing the mismatched DNA bases and the other subunit interacting with the DNA backbone (Figure 1.6) (Lamers et al., 2000; Obmolova et al., 2000). In Saccharomyces cerevisiae, as well as in mammals, Msh2-Msh6 and Msh2-Msh3 are MutS homologs with distinctly different recognition patterns. Msh6 recognizes base-base mispairings and small 1-2 nucleotide

19 indels. It is worth noting that C:C mismatches are poorly recognized (Marsischky et al., 1996;

Marsischky and Kolodner, 1999). Msh3 is primarily responsible for recognizing large indels (1-14 nt), but can, to a lesser extent, also recognize A:A, C:C, and T:G mismatches (Habraken et al.,

1996; Harrington and Kolodner, 2007). These differences are due to changes in the binding domain, as a chimeric Msh6 with a mispair-binding domain of Msh3 shows an Msh3 mismatch recognition specificity (Shell et al., 2007). Msh2 is a common subunit between these two heterodimers, but it does not directly interact with the mismatched base-pairs. However, msh2Δ results in a complete loss of MMR (Marsischky et al., 1996). Binding of the Msh2-Msh6 or Msh2-Msh3 primes the heterodimer to undergo a conformational change that recruits a

MutL homolog. Upon binding ATP, the MutS homolog can also form a clamp that slides away from the DNA mismatch (Gradia et al., 1999).

The primary MutL heterodimer in Saccharomyces is the Mlh1-Pms1 complex (Mlh1-

Pms2 in mammals) (Prolla et al., 1994). The complex is recruited to the mispaired DNA in an

ATP-dependent manner via both Msh2-Msh6 and Msh2-Msh3 (Mendillo et al., 2009). Mlh1-

Pms1, unlike its E. coli MutL homolog, has an endonuclease function that is promoted by both

PCNA, Msh2-Msh6, and presumably Msh2-Msh3, and mutating the active site of this domain disables MMR in vivo (Figure 1.6) (Kadyrov et al., 2007). Eukaryotic systems contain additional

MutL homologs, Mlh1-Mlh3, which have roles in meiosis and in CAG expansions (Hyun-Min et al., 2008), and Mlh1-Mlh2 (known as Mlh1-Pms1 in mammals), which appears to be a nonessential accessory factory in MMR (Campbell et al., 2014).

Figure 1.6 - Models of eukaryotic mismatch repair. Following mispair recognition by the Msh2- Msh6 heterodimer (or Msh2-Msh3, not shown), Mlh1-Pms1 is recruited, followed by recruitment of PCNA and endonucleolytic cleavage of the newly synthesized strand. Two separate models have been proposed after the first cleavage event by PCNA-activated Mlh1-Pms1. 1) Shown on the left, the nick 5’ of the mismatched bases initiates strand displacement by DNA polymerase 훿. The 5’ flap is cleaved and the template strand is recopied (Kadyrov et al., 2009). 2) Shown on the right, Mlh1-Pms1, stimulated by PCNA, undergoes multiple rounds of endonucleolytic cleavage, leading to degradation of the DNA surrounding the mispair. The resulting gap is filled in by DNA polymerase (Goellner et al., 2014). These two models may not be mutually exclusive.

As previously stated, hemimethylated GATC sites are the critical signal in the E. coli

MMR pathway, but this mechanism does not exist in most other organisms. However, a number

of studies have shown that MMR is still linked to replication and suggest that intermediates of

DNA replication may themselves be the signal for strand differentiation. Msh6 and Msh3 can

both bind to PCNA, suggesting that the MutS homologs interact with PCNA located at

replication forks, or are potentially left on the DNA following replication (Figure 1.6) (Clark et

21 al., 2000; Kleczkowska et al., 2001). Live cell imaging has shown that the Msh2-Msh6 heterodimer colocalizes with components of the replication fork in PCNA-Msh6 interaction- dependent fashion (Hombauer et al., 2011). Mlh1-Pms1, however, does not colocalize with the replication machinery, but approximately 10% of cells do form foci, which requires the presence of Msh2-Msh6 (Hombauer et al., 2011). Together this suggests that Msh2-Msh6 is a constitutive component of replication that detects mismatches left by polymerases and that

Mlh1-Pms1 are MMR intermediates that require multiple rounds of loading by the MutS homolog in response to the presence of mispaired bases. The question still remains on how eukaryotes distinguish the parental strand from the newly synthesized strand.

Mismatch repair in Homologous Recombination

The mismatch repair pathway plays a large role in HR by suppressing recombination with divergent substrates, known as homeologous recombination (Datta et al., 1996). However, the precise mechanisms and molecular triggers of this pathway remain largely unknown. There has been extensive research into what distinguishes a homologous sequence from a homeologous sequence. Studies in bacteria have shown that a 0.3% sequence divergence will reduce recombination rates (Roberts and Cohan, 1993). In Saccharomyces cerevisiae, a single mismatch in a recombination tract of 300 bp (0.3% divergence) reduces recombination threefold (Datta et al., 1997). As the sequence divergence increases to approximately 10%, the impact of MMR proteins in suppressing recombination is increasingly important for faithful DNA repair, however, recombination between sequences with >10% divergence is largely unaffected

22 by MMR. These data suggest that template rejection of greatly divergent sequences are most likely aborted before MMR machinery is recruited (Li et al., 2006).

The MMR machinery is required to both recognize and process heteroduplex DNA following DNA DSB repair using homeologous templates. The key proteins in these processes are Msh2-Msh6, which shows an antirecombination activity several times greater than other key MMR proteins such as Pms1 or Mlh1, indicating that it orchestrates a secondary pathway to

Sgs1 to antagonize homeologous recombination (Datta et al., 1996; Welz-Voegele and Jinks-

Robertson, 2008). The role of MMR in heteroduplex rejection has been extensively studied in

SSA models because it provides an easy and straightforward mechanism for generating heteroduplexes in vivo. In these studies, heteroduplex rejection depends on the Msh2-Msh6 heterodimer, which recognizes the mismatches, as well as Sgs1, which presumably unwinds the

DNA helix (Sugawara et al., 2004). Heteroduplex rejection is also significantly reduced in mlh1Δ mutants and in a pms1Δ mlh2Δ mlh3Δ triple mutant, although to a much lesser extent than that seen in an msh6Δ strain (Sugawara et al., 2004). These results reinforce the idea that heteroduplex rejection requires Msh2-Msh6, as well as one of three other heterodimers (Mlh1-

Pms1, Mlh1-Mlh2, Mlh1-Mlh3) (Sugawara et al., 2004). It should be noted that heteroduplex rejection in SSA is likely to be very different from heteroduplex rejection in a D-loop structure generated during GC or BIR. In these processes, the heteroduplex regions of the D-loop likely remain coated with Rad51, unlike the Rad51-free heteroduplex DNA generated in SSA systems

(Li and Heyer, 2009). In vivo studies have shown that during BIR, the Rad51-filament will still utilize a repair template that has 12% divergence for recombination (Anand et al., 2017).

Additionally, heterologous GC tracts in Saccharomyces with 2% sequence divergence suggest

23 that Msh6 and Mlh1 have distinct anti-

recombination activities, as Msh6 acts in repair

events that require donor engagements from both

ends of the DSB, while Mlh1 only affects SDSA

events (Hum and Jinks-Robertson, 2019). This has

led to a model where Msh2-Msh6 dismantles

heteroduplex intermediates prior to extension of

the 3’-end, and Mlh1 heterodimers destroy SDSA Figure 1.7 - Clipping of nonhomologous tails. A) After resection and annealing of intermediates after DNA synthesis has initiated complementary sequences, the 3’ nonhomologous tails generated during (Hum and Jinks-Robertson, 2019). SSA, or in some instances of BIR and GC, must be removed. Rad1-Rad10 cleavage The MMR machinery also plays a role in the (red scissors) allows the new 3’ ends to serve as primers for completion of the processing of nonhomologous tails (Figure 1.7). In repair event. Modified and adapted from Fishman-Lobell and Haber, 1992. B) SSA, 3’ flap removal depends on the Msh2-Msh3 Proteins required for 3’ flap excision. Modified and adapted from Li et al., heterodimer, which both recognizes and stabilizes 2008. heteroduplex structures with flaps (Figure 1.7) (Pâques and Haber 1997; Sugawara et al., 1997).

Once recognized, these flaps are typically removed by the Rad1-Rad10 complex, in conjunction with Saw1 and Slx4 (Figure 1.7) (Tomkinson et al., 1993; Li et al., 2008). The flap removal function of the Msh2-Msh3 heterodimer appears to be independent of Msh6, Mlh1, ad Pms1

(Pâques and Haber, 1997; Sugawara et al., 1997). Surprisingly, the impact of heterology on DSB repair is heavily influenced by the presence of a nonhomologous tail. When a nonhomologous tail is not present, Msh2 does not discourage homeologous recombination. However, when a 3’

24 tail is present on the DSB end, Msh2 promotes the rejection of the mismatched substrate as a suitable template for repair (Anand et al., 2017).

In this work I focus on the mechanisms that contribute to the asymmetry of DSB repair.

Specifically, I use site-specific nucleases to characterize a noncanonical, Rad51-independent

DSB repair pathway that promotes gene editing. We find that templated gene edits are heavily influenced by whether they are templated 5’ or 3’ of the DSB, with 5’ edits being predominantly incorporated by Msh2 and the MMR pathway, while 3’ edits are predominately fixed into the genome by the activity of Polymerase 훿. I also examine the effects of a heterologous template on the more canonical gene conversion pathway. Here we also see asymmetry in DSB repair – with a donor containing mismatches to the 5’ end of the DSB more likely to cause rejection of the donor template than its counterpart containing mismatches to the 3’ end of the DSB. Even more, we see that these heterologies are incorporated into the recipient locus differently, as mismatches 5’ of the DSB are fixed into the recipient locus through the activities of Pol훿, and those on the 3’ end through the MMR pathway. Collectively, these data highlight the complexity in DNA DSB repair and offer new insights on the differential processing of the DSB ends.

25 References

Adelmann, C.H., Wang, T., Sabatini, D.M., and Lander, E.S. (2019). Genome-Wide CRISPR/Cas9 Screening for Identification of Cancer Genes in Cell Lines. Methods Mol Biol 1907, 125-136.

Ahnesorg, P., Smith, P., and Jackson, S.P. (2006). XLF interacts with the XRCC4-DNA ligase IV complex to promote DNA nonhomologous end-joining. Cell 124, 301-313.

Anand, R., Beach, A., Li, K., and Haber, J. (2017). Rad51-mediated double-strand break repair and mismatch correction of divergent substrates. Nature 544, 377-380.

Anand, R.P., Tsaponina, O., Greenwell, P.W., Lee, C.S., Du, W, Petes, T.D., and Haber, J.E. (2014). Chromosome rearrangements via template switching between diverged repeated sequences. Genes Dev 28(21), 2394–2406.

Au, K.G., Welsh, K., and Modrich, P. (1992). Initiation of methyl-directed mismatch repair. J. Biol Chem 267, 12142–12148.

Aylon, Y., Liefshitz, B., and Kupiec, M. (2004). The CDK regulates repair of double-strand breaks by homologous recombination during the cell cycle. EMBO J 23, 4868–4875.

Barrangou, R., Fremaux, C., Deveau, H., Richards, M., Boyaval, P., Moineau, S., Romero, D.A., and Horvath, P. (2007). CRISPR provides acquired resistance against viruses in prokaryotes. Science 315, 1709-1712.

Bertomeu, T., Coulombe-Huntington, J., Chatr-Aryamontri, A., Bourdages, K.G., Coyaud, E., Raught, B., Xia, Y., and Tyers, M. (2018). A High-Resolution Genome-Wide CRISPR/Cas9 Viability Screen Reveals Structural Features and Contextual Diversity of the Human Cell- Essential Proteome. Mol Cell Biol 38.

Blier, P.R., Griffith, A.J., Craft, J., and Hardin, J.A. (1993). Binding of Ku protein to DNA. Measurement of affinity for ends and demonstration of binding to nicks. J Biol Chem 268, 7594- 7601.

Brouns, S.J., Jore, M.M., Lundgren, M., Westra, E.R., Slijkhuis, R.J., Snijders, A.P., Dickman, M.J., Makarova, K.S., Koonin, E.V., and van der Oost, J. (2008). Small CRISPR RNAs guide antiviral defense in prokaryotes. Science 321, 960-964.

Burdett, V., Baitinger, C., Viswanathan, M., Lovett, S.T., and Modrich, P. (2001). In vivo requirement for RecJ, ExoVII, ExoI, and ExoX in methyl-directed mismatch repair. Proc Natl Acad Sci USA 98, 6765–6770.

26 Butler, G., Kenny, C., Fagan, A., Kurischko, C., Gaillardin, C., and Wolfe, K. H. (2004). Evolution of the MAT locus and its HO endonuclease in yeast species. Proc Natl Acad Sci USA 101, 1632− 1637.

Bzymek, M., Thayer, N.H., Oh, S.D., Kleckner, N., and Hunter, N. (2010). Double Holliday junctions are intermediates of DNA break repair. Nature 464, 937–941.

Campbell, C.S., Hombauer, H., Srivatsan, A., Bowen, N., Gries, K., Desai, A., Putnam, C.D., and Kolodner, R.D. (2014). Mlh2 is an accessory factor for DNA mismatch repair in Saccharomyces cerevisiae. PLoS Genet 10, e1004327.

Campbell, J.L., and Kleckner, N. (1990). E. coli oriC and the dnaA gene promoter are sequestered from dam methyltransferase following the passage of the chromosomal replication fork. Cell 62, 967– 979.

Cannavo, E., and Cejka, P. (2014). Sae2 promotes dsDNA endonuclease activity within Mre11- Rad50-Xrs2 to resect DNA breaks. Nature 514, 122-125.

Cannavo, E., Reginato, G., and Cejka, P. (2019). Stepwise 5’ DNA end-specific resection of DNA breaks by the Mre11-Rad50-Xrs2 and Sae2 nuclease ensemble. Proc Natl Acad Sci USA 116(12), 5505-5513.

Chang, H.H.Y., Pannunzio, N.R., Adachi, N., and Lieber, M.R. (2017). Non-homologous DNA end joining and alternative pathways to double-strand break repair. Nat Rev Mol Cell Biol 18, 495- 506.

Chen, H., Choi, J., and Bailey, S. (2014). Cut site selection by the two nuclease domains of the Cas9 RNA-guided endonuclease. J Biol Chem 289, 13284-13294.

Chen, H., Lisby, M., and Symington, L.S. (2013). RPA coordinates DNA end resection and prevents formation of DNA hair- pins. Mol Cell 50, 589–600.

Chen, Q., Ijpma, A., and Greider, C.W. (2001). Two survivor pathways that allow growth in the absence of telomerase are generated by distinct telomere recombination events. Mol Cell Biol 21, 1819–1827.

Chen, Z., Yang, H., and Pavletich, N.P. (2008). Mechanism of homologous recombination from the RecA-ssDNA/dsDNA structures. Nature 453, 484–489.

Clark, A.B., Valle, F., Drotschmann, K, Gary, R.K., and Kunkel, T.A. (2000). Functional interaction of proliferating cell nuclear antigen with MSH2–MSH6 and MSH2–MSH3 complexes. J Biol Chem 275, 36498–36501.

27 Cloud, V.,Chan, Y.L.,Grubb, J., Budke, B., and Bishop, D.K. (2012). Rad51 is an accessory actor for Dmc1-mediated joint molecule formation during meiosis. Science 337, 1222–25.

Cohen, H., and Sinclair, D.A. (2001). Recombination-mediated lengthening of terminal telomeric repeats requires the Sgs1 DNA helicase. Proc Natl Acad Sci USA 98, 3174–3179.

Colleaux, L., d’Auriol, L., Betermier, M., Cottarel, G., Jacquier, A., Galibert, F., and Dujon, B. (1986). Universal code equivalent of a yeast mitochondrial intron reading frame is expressed into E. coli as a specific double strand endonuclease. Cell 44, 521−533.

Cooper, D.L., Lahue, R.S., and Modrich, P. (1993). Methyl-directed mismatch repair is bidirectional. J Biol Chem 268, 11823–11829.

Cyranoski, D., and Ledford, H. (2018). Genome-edited baby claim provokes international outcry. Nature 563, 607-608.

Datta, A., Adjiri, A., New, L., Crouse, G.F., and Jinks-Robertson, S. (1996). Mitotic crossovers between diverged sequences are regulated by mismatch repair proteins in Saccharomyces cerevisiae. Mol Cell Biol 16, 1085–1093.

Datta, A., Hendrix, M., Lipsitch, M., and Jinks-Robertson, S. (1997). Dual roles for DNA sequence identity and the mismatch repair system in the regulation of mitotic crossing-over in yeast. Proc Natl Acad Sci USA 97, 9757–9762.

Davis, A.P., and Symington, L.S. (2004). RAD51-dependent break- induced replication in yeast. Mol Cell Biol 24, 2344 – 2351.

Deem, A., Keszthelyi, A., Blackgrove, T., Vayl, A., Coffey, B., Mathur, R., Chabes, A., and Malkova, A. (2011). Break-induced replication is highly inaccurate. PLoS Biol 9, e1000594.

Deshpande, R.A., and Wilson, T.E. (2007). Modes of interaction among yeast Nej1, Lif1 and Dnl4 proteins and comparison to human XLF, XRCC4 and Lig4. DNA Repair 6, 1507- 1516.

DiCarlo, J.E., Norville, J.E., Mali, P., Rios, X., Aach, J., and Church, G.M. (2013). Genome engineering in Saccharomyces cerevisiae using CRISPR-Cas systems. Nucleic Acids Res 41, 4336- 4343.

Doksani Y, and de Lange T. (2014). The role of DSB repair pathways at functional and dysfunctional telomeres. Cold Spring Harb Perspect Biol, a016576.

Doudna, J.A., and Charpentier, E. (2014). Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science 346, 1258096.

28 Douglas, P., Gupta, S., Morrice, N., Meek, K., and Lees-Miller, S.P. (2005). DNA-PK-dependent phosphorylation of Ku70/80 is not required for non-homologous end joining. DNA Repair 4, 1006-1018.

Drouet, J., Delteil, C., Lefrancois, J., Concannon, P., Salles, B., and Calsou, P. (2005). DNA- dependent protein kinase and XRCC4-DNA ligase IV mobilization in the cell in response to DNA double strand breaks. J Biol Chem 280, 7060-7069.

Drouet, J., Frit, P., Delteil, C., de Villartay, J.P., Salles, B., and Calsou, P. (2006). Interplay between Ku, Artemis, and the DNA-dependent protein kinase catalytic subunit at DNA ends. J Biol Chem 281, 27784-27793

Dzau, V.J., McNutt, M., and Bai, C. (2018). Wake-up call from Hong Kong. Science 362, 1215.

Fishman-Lobell, J., and Haber, J.E. (1992). Removal of nonhomologous DNA ends in double- strand break recombination: the role of the yeast ultraviolet repair gene RAD1. Science 258(5081), 480–484.

Frank-Vaillant, M., and Marcand, S. (2001). NHEJ regulation by mating type is exercised through a novel protein, Lif2p, essential to the ligase IV pathway. Genes Dev 15, 3005-3012.

Gasiunas, G., Barrangou, R., Horvath, P., and Siksnys, V. (2012). Cas9-crRNA ribonucleoprotein complex mediates specific DNA cleavage for adaptive immunity in bacteria. Proc Natl Acad Sci U S A 109, E2579-2586.

Geier, G.E., and Modrich, P. (1979). Recognition sequence of the dam methylase of Escherichia coli K12 and mode of cleavage of DpnI endonuclease. J Biol Chem 254(4), 1408-1413.

Ghodke, I., and Muniyappa, K. (2013). Processing of DNA double-stranded breaks and intermediates of recombination and repair by Saccharomyces cerevisiae Mre11 and its stimulation by Rad50, Xrs2, and Sae2 proteins. J Biol Chem 288, 11273-11286

Gloor, G.B., Nassif, N.A., Johnson-Schlitz, D.M., Preston, C.R., and Engels, W.R. (1991). Targeted gene replacement in Drosophila via P element-induced gap repair. Science 253, 1110 – 1117.

Goellner, E.M., Smith, C.E., Campbell, C.S., Hombauer, H., Desai, A., Putnam, C.D., and Kolodner, R.D. (2014). PCNA and Msh2-Msh6 activate an Mlh1-Pms1 endonuclease pathway required for Exo-independent mismatch repair. Mol Cell 55, 291-304.

Gradia, S., Subramanian, D., Wilson, T., Acharya, S., Makhov, A., Griffith, J., and Fishel, R. (1999). hMSH2–hMSH6 forms a hydrolysis-independent sliding clamp on mismatched DNA. Mol Cell 3, 255–261.

29 Gravel, S., Chapman, J.R., Magill, C., and Jackson, S.P. (2008). DNA helicases Sgs1 and BLM promote DNA double-strand break resection. Genes Dev 22, 2767–2772.

Grilley, M., Welsh, K.M., Su, S.S., and Modrich, P. (1989). Isolation and characterization of the Escherichia coli mutL gene product. J Biol Chem 264, 1000–1004.

Haber, J.E. (2016). A Life Investigating Pathways That Repair Broken Chromosomes. Annu Rev Genet 50, 1-28

Habraken, Y., Sung, P., Prakash, L., and Prakash, S. (1996). Binding of insertion/deletion DNA mismatches by the heterodimer of yeast mismatch repair proteins MSH2 and MSH3. Curr Biol 6, 1185–1187.

Hall, M.C., Jordan, J.R., and Matson, S.W. (1998). Evidence for a physical interaction between the Escherichia coli methyl-directed mismatch repair proteins MutL and UvrD. EMBO J 17, 1535–1541.

Han, J., Perez, J.T., Chen, C., Li, Y., Benitez, A., Kandasamy, M., Lee, Y., Andrade, J., tenOever, B., and Manicassamy, B. (2018). Genome-wide CRISPR/Cas9 Screen Identifies Host Factors Essential for Influenza Virus Replication. Cell Rep 23, 596-607.

Harrington, J.M., and Kolodner, R.D.(2007). Saccharomyces cerevisiae Msh2–Msh3 acts in repair of base–base mispairs. Mol Cell Biol 27, 6546–6554.

Heyer, W.D., Ehmsen, K.T., and Liu, J. (2010). Regulation of homologous recombination in eukaryotes. Annu Rev Genet 44, 113-139.

Hicks, W.M., Kim, M., and Haber, J.E. (2010). Increased mutagenesis and unique mutation signature associated with mitotic gene conversion. Science 329, 82-85.

Hicks, W.M., Yamaguchi, M., and Haber, J.E. (2011). Real-time analysis of double-strand DNA break repair by homologous recombination. Proc Natl Acad Sci USA 108, 3108-3115.

Hiraga, S., Ichinose, C., Onogi, T., Niki, H., Yamazoe, M. (2000). Bidirectional migration of SeqA- bound hemimethylated DNA clusters and pairing of oriC copies in Escherichia coli. Genes Cells 5, 327–341.

Holbeck, S.L., and Strathern, J.N. (1997). A role for REV3 in mutagenesis during double-strand break repair in Saccharomyces cerevisiae. Genetics 147, 1017–24.

Hombauer, H., Srivatsan, A., Putnam, C.D., and Kolodner, R.D. (2011). Mismatch repair, but not heteroduplex rejection, is temporally coupled to DNA replication. Science 334, 1713– 1716.

30 Hsu, P.D., Lander, E.S., and Zhang, F. (2014). Development and applications of CRISPR-Cas9 for genome engineering. Cell 157, 1262-1278.

Hum, Y.F., and Jinks-Robertson, S. (2019). Mismatch recognition and subsequent processing have distinct effects on mitotic recombination intermediates and outcomes in yeast. Nucleic Acids Res 47(9), 4554-4568.

Hyun-Min, K., Narayanan, V., Mieczkowski, P.A., Petes, T.D., Krasilnikova, M.M., Mirkin, S.M., and Lobachev K.S. (2008). Chromosome fragility at GAA tracts in yeast depends on repeat orientation and requires mismatch repair. EMBO J 27, 2896-2906.

Ira, G., and Haber, J.E. (2002). Characterization of Rad51-independent break-induced replication that acts preferentially with short homologous sequences. Mol Cell Biol 22, 6384-6392.

Ira, G., Malkova, A., Liberi, G., Foiani, M., and Haber, J.E. (2003). Srs2 and Sgs1-Top3 suppress crossovers during double-strand break repair in yeast. Cell 115, 401–411.

Ira, G., Pellicioli, A., Balijja, A., Wang, X., Fiorani, S., Carotenuto, W., Liberi, G., Bressan, D., Wan, L., Hollingsworth, N.M., et al. (2004). DNA end resection, homologous recombination and DNA damage checkpoint activation require CDK1. Nature 431, 1011–1017.

Ira, G., Satory, D., and Haber, J.E. (2006). Conservative inheritance of newly synthesized DNA in double-strand break-induced gene conversion. Mol Cell Biol 26, 9424–9429.

Ivanov, E.L., Sugawara, N., Fishman-Lobell, J., and Haber, J.E. (1996). Genetic requirements for the single-strand annealing pathway of double-strand break repair in Saccharomyces cerevisiae. Genetics 142, 693–704.

Ivanov, E.L., Sugawara, N., White, C.I., Fabre, F., and Haber, J.E. (1994). Mutations in XRS2 and RAD50 delay but do not prevent mating-type switching in Saccharomyces cerevisiae. Mol Cell Biol 14, 3414–3425.

Iyer, R.R., Pluciennik, A., Burdett, V., and Modrich, P.L. (2006) DNA mismatch repair: functions and mechanisms. Chem Rev 106, 302–323.

Iyer, V., Boroviak, K., Thomas, M., Doe, B., Riva, L., Ryder, E., and Adams, D.J. (2018). No unexpected CRISPR-Cas9 off-target activity revealed by trio sequencing of gene-edited mice. PLoS Genet 14, e1007503.

Jacquier, A., and Dujon, B. (1985). An intron-encoded protein is active in a gene conversion process that spreads an intron into a mitochondrial gene. Cell 41, 383−394.

31 Jain, S., Sugawara, N., Lydeard, J., Vaze, M., Tanguy Le Gac, N., and Haber, J.E. (2009). A recombination execution checkpoint regulates the choice of homologous recombination pathway during DNA double-strand break repair. Genes Dev 23, 291–303.

Jensen, R., Sprague, G. F., and Herskowitz, I. (1983). Regulation of yeast mating-type interconversion: feedback control of HO gene expression by the mating-type locus. Proc. Natl. Acad. Sci. U. S. A. 80, 3035−3039.

Jiang, F., Taylor, D.W., Chen, J.S., Kornfeld, J.E., Zhou, K., Thompson, A.J., Nogales, E., and Doudna, J.A. (2016). Structures of a CRISPR-Cas9 R-loop complex primed for DNA cleavage. Science 351, 867-871

Jinek, M., Chylinski, K., Fonfara, I., Hauer, M., Doudna, J.A., and Charpentier, E. (2012). A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816-821

Jinek, M., Jiang, F., Taylor, D.W., Sternberg, S.H., Kaya, E., et al. (2014). Structures of Cas9 endonucleases reveal RNA-mediated conformational activation. Science 343(6176), 816-821.

Johnson, F.B., Marciniak, R.A., McVey, M., Stewart, S.A., Hahn, W.C., and Guarente, L. (2001). The Saccharomyces cerevisiae WRN homolog Sgs1p participates in telomere maintenance in cells lacking telomerase. EMBO J 20, 905–913.

Joyce, C.M., and Steitz, T.A. (1994). Function and structure relationships in DNA polymerases. Annu Rev Biochem 63, 888-822

Kadyrov, F.A., Genschel, J., Fang, Y., Penland, E., Edelmann, W., and Modrich, P. (2009). A possible mechanism for exonuclease 1-independent eukaryotic mismatch repair. Proc Natl Acad Sci USA 106, 8495-8500.

Kadyrov, F.A., Holmes, S.F., Arana, M.E., Lukianova, O.A., O'Donnell, M., Kunkel, T.A., and Modrich, P. (2007). Saccharomyces cerevisiae MutLalpha is a mismatch repair endonuclease. J Biol Chem 282, 37181–37190.

Karanja, K.K., Cox, S.W., Duxin, J.P., Stewart, S.A., and Campbell, J.L. (2012). DNA2 and EXO1 in replication-coupled, homology-directed repair and in the interplay between HDR and the FA/BRCA network. Cell Cycle 11, 3983–3996.

Kegel, A., Sjostrand, J.O., and Astrom, S.U. (2001). Nej1p, a cell type-specific regulator of nonhomologous end joining in yeast. Curr Biol 11, 1611-1617.

Kleczkowska, H.E., Marra, G., Lettieri, T., and Jiricny, J. (2001). hMSH3 and hMSH6 interact with PCNA and colocalize with it to replication foci. Genes Dev 15, 724–736.

32 Kleinstiver, B.P., Pattanayak, V., Prew, M.S., Tsai, S.Q., Nguyen, N.T., Zheng, Z., and Joung, J.K. (2016). High-fidelity CRISPR–Cas9 nucleases with no detectable genome-wide off-target effects. Nature 529, 490.

Krishna, T.S., Kong, X.P., Gary, S., Burgers, P.M., and Kuriyan, J. (1994). Crystal structure of the eukaryotic DNA polymerase processivity factor PCNA. Cell 79, 1233–1243.

Lahue, R.S., Au, K.G., and Modrich, P. (1989). DNA mismatch correction in a defined system. Science 245, 160–164.

Lahue, R.S., Su, S.S., and Modrich, P. (1987) Requirement for d(GATC) sequences in Escherichia coli mutHLS mismatch correction. Proc Natl Acad Sci USA 84, 1482–1486.

Lamers, M.H., Perrakis, A., Enzlin, J.H., Winterwerp, H.H., de Wind, N., and Sixma, T.K. (2000). The crystal structure of DNA mismatch repair protein MutS binding to a G × T mismatch. Nature 407, 711–717.

Lander, E.S., Baylis, F., Zhang, F., Charpentier, E., Berg, P., Bourgain, C., Friedrich, B., Joung, J.K., Li, J., Liu, D., et al. (2019). Adopt a moratorium on heritable genome editing. Nature 567, 165- 168.

Langle-Rouault, F., Maenhaut-Michel, G., and Radman, M. (1987). GATC sequences, DNA nicks and the MutH function in Escherichia coli mismatch repair. EMBO J 6, 1121–1127.

Le, S., Moore, J.K., Haber, J.E., and Greider, C.W. (1999). RAD50 and RAD51 define two pathways that collaborate to maintain telomeres in the absence of telomerase. Genetics 152, 143–152.

Lee, C. S., and Haber, J. E. (2015). Mating-type Gene Switching in Saccharomyces cerevisiae. Microbiol. Spectrum 3, MDNA3-0013- 2014.

Lee, C.S., Wang, R.W., Chang, H.H., Capurso, D., Segal, M.R., and Haber, J.E. (2016). Chromosome position determines the success of double-strand break repair. PNAS 113, E146– 54.

Lee, J.Y., Terakawa, T., Qi, Z., Steinfeld, J.B., Redding, S., Kwon, Y., Gaines, W.A., Zhao, W., Sung, P., and Greene, E.C. (2015). DNA Recombination. Base triplet stepping by the Rad51/RecA family of recombinases. Science 349, 977–81.

Lemos, B.R., Kaplan, A.C., Bae, J.E., Ferrazzoli, A.E., Kuo, J., Anand, R.P., Waterman, D.P., and Haber, J.E. (2018). CRISPR/Cas9 cleavages in budding yeast reveal templated insertions and strand-specific insertion/deletion profiles. Proc Natl Acad Sci USA 115, E2040-E2047.

33 Li, F., Dong, J., Pan, X., Oum, J.H., Boeke, J.D., Lee, S.E. (2008). Microarray-based genetic screen defines Saw1, a gene required for Rad1/Rad10-dependent processing of recombination intermediates. Mol Cell 30(3), 325-335

Li, X., and Heyer, W.D. (2009). Rad54 controls access to the invading 3’-OH end after Rad51- mediated DNA strand invasion in homologous recombination in Saccharomyces cerevisiae. Nucleic Acids Res 37(2), 638-646.

Li, X., Stith, C.M., Burgers, P.M., and Heyer, W.D. (2009). PCNA is required for initiation of recombination-associated DNA synthesis by DNA polymerase d. Mol Cell 36, 704 – 713.

Li, Z., Peled, J.U., Zhao, C., Svetlanov, Ronai, D., Cohen, P.E., and Scharff, M.D. (2006). A role for Mlh3 in somatic hypermutation. DNA Repair 5(6), 675-682.

Lin, J., and Wong, K.C. (2018). Off-target predictions in CRISPR-Cas9 gene editing using deep learning. Bioinformatics 34, i656-i663.

Liu, J., Renault, L., Veaute, X., Fabre, F., Stahlberg, H., and Heyer, W.D. (2011). Rad51 paralogues Rad55-Rad57 balance the antirecombinase Srs2 in Rad51 filament formation. Nature 479, 245–48.

Lobner-Olesen, A., Skovgaard, O., and Marinus, M.G. (2005). Dam methylation: coordinating cellular processes. Curr Opin Microbiol 8, 154–160.

Lundblad, V., and Blackburn, E.H. (1993). An alternative pathway for yeast telomere maintenance rescues est1 senescence. Cell 73, 347–360.

Lydeard, J.R., Jain, S., Yamaguchi, M., and Haber, J.E. (2007). Break- induced replication and telomerase-independent telomere maintenance require Pol32. Nature 448, 820–823.

Lydeard, J.R., Lipkin-Moore, Z., Sheu, Y.J., Stillman, B., Burgers, P.M., Haber, J.E. (2010). Break- induced replication requires all essential DNA replication factors except those specific for pre- RC assembly. Genes Dev 24, 1133–1144.

Ma, J.L., Kim, E.M., Haber, J.E., and Lee, S.E. (2003). Yeast Mre11 and Rad1 proteins define a Ku- independent mechanism to repair double-strand breaks lacking overlapping end sequences. Mol Cell Biol 23, 8820-8828.

Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E., Church, G.M. (2013). RNA-guided human genome engineering via Cas9. Science 339(6121), 823–826.

Malkova, A., Naylor, M.L., Yamaguchi, M., Ira, G., and Haber, J.E. (2005). RAD51-dependent break-induced replication differs in kinetics and checkpoint responses from RAD51- mediated gene conversion. Mol Cell Biol 25, 933–944.

34 Mari, P.O., Florea, B.I., Persengiev, S.P., Verkaik, N.S., Bruggenwirth, H.T., Modesti, M., Giglia- Mari, G., Bezstarosti, K., Demmers, J.A., Luider, T.M., et al. (2006). Dynamic assembly of end- joining complexes requires interaction between Ku70/80 and XRCC4. Proc Natl Acad Sci U S A 103, 18597-18602.

Marinus, M. and Morris, N.R. (1973). Isolation of deoxyribonucleic acid methylase mutants of Escherichia coli K-12. J Bacteriol 114(3), 1143-1150.

Marsischky, G.T., and Kolodner, R.D. (1999). Biochemical characterization of the interaction between the Saccharomyces cerevisiae MSH2–MSH6 complex and mispaired bases in DNA. J Biol Chem 274, 26668–26682.

Marsischky, G.T., Filosi, N., Kane, M.F., and Kolodner, R. (1996). Redundancy of Saccharomyces cerevisiae MSH3 and MSH6 in MSH2-dependent mismatch repair. Genes Dev 10, 407–420.

Mehta, A., Beach, A., and Haber, J.E. (2017). Homology Requirements and Competition between Gene Conversion and Break-Induced Replication during Double-Strand Break Repair. Mol Cell 65, 515-526 e513.

Mendillo, M.L., Mazur, D.J., and Kolodner, R.D. (2005). Analysis of the interaction between the Saccharomyces cerevisiae MSH2–MSH6 and MLH1–PMS1 complexes with DNA using a reversible DNA end- blocking system. J Biol Chem 280, 22245–22257.

Mimitou, E.P., and Symington, L.S. (2008). Sae2, Exo1 and Sgs1 collaborate in DNA double- strand break processing. Nature 455, 770–774.

Mimitou, E.P., and Symington, L.S. (2009). DNA end resection: Many nucleases make light work. DNA Repair 8, 983 – 995.

Morrison, A., Bell, J.B., Kunkel, T.A., and Sugino, A. (1991). Eukaryotic DNA polymerase amino acids sequence required for 3’-5’ exonuclease activity. Proc Natl Acad Sci USA 88(21), 9473- 9477.

Morrison, A., Johnsono, A.L., Johnston, L.H., and Sugino, A. (1993). Pathways correcting DNA replication errors in Saccharomyces cerevisiae. EMBO 12(4), 1467-1473.

Nickoloff, J. A., Chen, E. Y., and Heffron, F. (1986) A 24-base- pair DNA sequence from the MAT locus stimulates intergenic recombination in yeast. Proc. Natl. Acad. Sci. U. S. A. 83, 7831−7835.

35 Nimonkar, A.V., Genschel, J., Kinoshita, E., Polaczek, P., Campbell, J.L., Wyman, C., Modrich, P., and Kowalczykowski, S.C. (2011). BLM-DNA2-RPA-MRN and EXO1-BLM-RPA-MRN constitute two DNA end resection machineries for human DNA break repair. Genes Dev 25, 350-362.

Nishimasu, H., Ran, F.A., Hsu, P.D., Konermann, S., Shehata, S.I., Dohmae, N., Ishitani, R., Zhang, F., and Nureki, O. (2014). Crystal structure of Cas9 in complex with guide RNA and target DNA. Cell 156, 935-949.

Obmolova, G., Ban, C., Hsieh, P., and Yang, W. (2000). Crystal structures of mismatch repair protein MutS and its complex with a substrate DNA. Nature 407, 703–710.

Ooi, S.L., and Boeke, J.D. (2001). A DNA microarray-based genetic screen for nonhomologous end-joining mutants in Saccharomyces cerevisiae. Science 294, 2552–56.

Pâques, F, and Haber, J.E. (1997). Two pathways for removal of nonhomologous DNA ends during double-strand break repair in Saccharomyces cerevisiae. Mol Cell Biol 17(11), 6765-6771.

Pâques, F., and Haber, J.E. (1999). Multiple pathways of recombination induced by double- strand breaks in Saccharomyces cerevisiae. Microbiol Mol Biol Rev 63, 349-404.

Patel, S.S., Wong, I., and Johnson, K.A. (1991). Pre-steady-state kinetic analysis of processive DNA replication including complete characterization of an exonuclease-deficient mutant. Biochemistry 30(2), 511-525.

Perrin, A., Buckle, M., and Dujon, B. (1993). Asymmetrical recognition and activity of the I-SceI endonuclease on its site and on intron-exon junctions. EMBO J 12, 2939−2947.

Peterson, S.E., Li, Y., Chait, B.T., Gottesman, M.E., Baer, R., and Gautier, J. (2011). Cdk1 uncouples CtIP-dependent resection and Rad51 filament formation during M-phase double- strand break repair. J Cell Biol 194, 705–720.

Prakash, R., Satory, D., Dray, E., Papusha, A., Scheller, J., Kramer, W., Krejci, L., Klein, H., Haber, J.E., Sung, P., et al. (2009). Yeast Mph1 helicase dissociates Rad51-made D-loops: Implications for crossover control in mitotic recombination. Genes Dev 23, 67–79.

Prelich, G., Tan, C.K., Kostura, M., Mathews, M.B., So, A.G., Downey, K.M., and Stillman, B. (1987). Functional identity of proliferating cell nuclear antigen and a DNA polymerase-delta auxiliary protein. Nature 326, 517–520.

Prolla, T.A., Pang, Q., Alani, E., Kolodner, R.D., and Liskay, R.M. (1994). MLH1, PMS1, and MSH2 interactions during the initiation of DNA mismatch repair in yeast. Science 265, 1091–1093.

Ran, F.A., Hsu, P.D., Wright, J., Agarwala, V., Scott, D.A., and Zhang, F. (2013). Genome engineering using the CRISPR-Cas9 system. Nat Protoc 8, 2281-2308.

36 Rattray, A.J., Shafer, B.K., McGill, C.B., and Strathern, J.N. (2002). The roles of REV3 and RAD57 in double-strand-break- repair-induced mutagenesis of Saccharomyces cerevisiae. Genetics 162(3), 1063–1077.

Renkawitz, J., Lademann, C.A., Kalocsay, M., and Jentsch, S. (2013). Monitoring homology search during DNA double-strand break repair in vivo. Mol Cell 50, 261–72.

Resnick, M.A. (1976). The repair of double-strand breaks in DNA; a model involving recombination. J Theor Biol 59, 97–106.

Roberts, M.S., and Cohan, F.M. (1993). The effect of DNA sequence divergence on sexual isolation in Bacillus. Genetics 132(2), 401-408.

Schaaper RM (1993) The mutational specificity of two Escherichia coli dnaE antimutator alleles as determined from lacI mutation spectra. Genetics 134, 1031-1038.

Schlagman, S.L., Hattman, S., and Marinus, M.G. (1986). Direct role of the Escherichia coli Dam DNA methyltransferase in methylation-directed mismatch repair. J Bacteriol 165, 896–900.

Sebesta, M., Burkovics, P., Haracska, L., and Krejci, L. (2011). Reconstitution of DNA repair synthesis in vitro and the role of polymerase and helicase activities. DNA Repair 10, 567 – 576.

Shalem, O., Sanjana, N.E., Hartenian, E., Shi, X., Scott, D.A., Mikkelson, T., Heckl, D., Ebert, B.L., Root, D.E., Doench, J.G., et al. (2014). Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343, 84-87.

Shell, S.S., Putnam, C.D., and Kolodner, R.D. (2007). The N terminus of Saccharomyces cerevisiae Msh6 is an unstructured tether to PCNA. Mol Cell 26, 565–578.

Shen, B., Zhang, J., Wu, H., Wang, J., Ma, K., Li, Z., Zhang, X., Zhang, P., and Huang, X. (2013). Generation of gene-modified mice via Cas9/RNA-mediated gene targeting. Cell Res 23, 720-723.

Shibata, M., Nishimasu, H., Kodera, N., Hirano, S., Ando, T., Uchihashi, T., and Nureki, O. (2017). Real-space and real-time dynamics of CRISPR-Cas9 visualized by high-speed atomic force microscopy. Nat Commun 8, 1430.

Sieverman, K.J., and Rine, J. (2018). Impact of homologous recombination on silent chromatin in Saccharomyces cerevisiae. Genetics 208(3), 1099-1113.

Signon, L., Malkova, A., Naylor, M.L., Klein, H., and Haber, J.E. (2001) Genetic requirements for RAD51- and RAD54- independent break-induced replication repair of a chromosomal double- strand break. Mol Cell Biol 21(6): 2048–2056.

37 Simon, M., Giot, L., and Faye, G. (1991). The 3’ to 5’ exonuclease activity located in the DNA polymerase delta subunit of Saccharomyces cerevisiae required for accurate replication. EMBO J 10(8), 2165-2170.

Smith, C.E., Llorente, B., and Symington, L.S. (2007). Template switching during break-induced replication. Nature 447, 102–105.

Sternberg S.H., LaFrance B, Kaplan M, Doudna J.A. (2015). Conformational control of DNA target cleavage by CRISPR-Cas9. Nature 527, 110-113.

Sternberg, S.H., Redding, S., Jinek, M., Greene, E.C., and Doudna, J.A. (2014). DNA interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature 507, 62-67.

Stracker, T.H., Theunissen, J.W., Morales, M., and Petrini, J.H. (2004). The Mre11 complex and the metabolism of chromosome breaks: the importance of communicating and holding things together. DNA Repair 3, 845-854.

Su, S.S., and Modrich, P. (1986). Escherichia coli mutS-encoded protein binds to mismatched DNA base pairs. Proc Natl Acad Sci USA 83, 5057–5061.

Sugawara, N., Goldfarb, T., Studamire, B., Alani, E., and Haber, J.E. (2004). Heteroduplex rejections during single-strand annealing requires Sgs1 helicase and mismatch repair proteins Msh2 and MSh6 but not Pms1. Proc Natl Acad Sci USA 101(25), 9315-9320.

Sugawara, N., Ira, G., and Haber, J.E. (2000). DNA length dependence of the single-strand annealing pathway and the role of Saccharomyces cerevisiae RAD59 in double-strand break repair. Mol Cell Biol 20, 5300–5309.

Sugawara, N., Pâques, F., Colaiacovo, M., and Haber, J.E. (1997). Role of Saccharomyces cerevisiae Msh2 and Msh3 repair proteins in double-strand break-induced recombination. Proc Natl Acad Sci 94, 9214–921.

Sugawara, N., Wang, X., and Haber, J.E. (2003). In vivo roles of Rad52, Rad54, and Rad55 proteins in Rad51-mediated recombination. Mol Cell 12, 209–19.

Sugiyama, T., and Kowalczykowski, S.C. (2002). Rad52 protein associates with RPA-ssDNA to accelerate Rad51-mediated displacement of RPA and presynaptic complex formation. J Biol Chem 19, 19.

Swan, M.K., Johnson, R.E., Prakash, L., Prakash, S., and Aggarwal, A.K. (2009). Structural basis of high-fidelity DNA synthesis by yeast DNA polymerase delta. Nat Struct Mol Biol 16(9), 979-986.

Symington, L.S. (2014). End resection at double-strand breaks: mechanism and regulation. Cold Spring Harb Perspect Biol 6(8): a016436.

38 Szostak, J.W., Orr-Weaver, T.L., Rothstein, R.J., and Stahl, F.W. (1983). The double-strand-break repair model for recombination. Cell 33, 25–35.

Teng, S. C., and Zakian, V.A. (1999). Telomere-telomere recombination is an efficient bypass pathway for telomere maintenance in Saccharomyces cerevisiae. Mol Cell Biol 19, 8083–8093.

Teng, S.C., Chang, J., McCowan, B., and Zakian, V.A. (2000). Telomerase-independent lengthening of yeast telomeres occurs by an abrupt Rad50p- dependent, Rif-inhibited recombinational process. Mol Cell 6, 947–952.

Tomkinson, A.E., Bardwell, A.J., Bardwell, L., Tappe, N.J., and Friedberg, E.C. (1993). Yeast DNA repair and recombination proteins Rad1 and Rad10 constitute a single-stranded-DNA endonuclease. Nature 362, 860–862.

Tran, H.T., Degtyareva, N.P., Gordenin, D.A., and Resnick, M.A. (1999). Genetic factors affecting the impact of DNA polymerase delta proofreading activity on mutation avoidance in yeast. Genetics 152(1), 47-59.

Tsaponina, O., and Haber, J.E. (2014). Frequent Interchromosomal Template Switches during Gene Conversion in S. cerevisiae. Mol Cell 55(4), 615–625.

Valencia, M., Bentele, M., Vaze, M.B., Herrmann, G., Kraus, E., Lee, S.E., Schar, P., and Haber, J.E. (2001). NEJ1 controls non-homologous end joining in Saccharomyces cerevisiae. Nature 414, 666-669.

Villarreal, D.D., Lee, K., Deem, A., Shim, E.Y., Malkova, A., and Lee, S.E. (2012). Microhomology directs diverse DNA break repair pathways and chromosomal translocations. PLoS Genet 8, e1003026.

Wang, T., Wei, J.J., Sabatini, D.M., and Lander, E.S. (2014). Genetic screens in human cells using the CRISPR-Cas9 system. Science 343, 80-84.

Wang, X., and Haber, J.E. (2004). Role of Saccharomyces single-stranded DNA-binding protein RPA in the strand invasion step of double-strand break repair. PLOS Biol 2, 104–111.

Wang, X., Ira, G., Tercero, J.A., Holmes, A.M., Diffley, J.F., and Haber, J.E. (2004). Role of DNA replication proteins in double-strand break-induced recombination in Saccharomyces cerevisiae. Mol Cell Biol 24, 6891–6899.

Welsh, K.M., Lu, A.L., Clark, S., and Modrich P. (1987). Isolation and characterization of the Escherichia coli mutH gene product. J Biol Chem 262, 15624–15629.

Welz-Voegele, C., and Jinks-Robertson, S. (2008). Sequence divergence impedes crossover more than noncrossover events during mitotic gap repair in yeast. Genetics 179(3), 1251-1262.

39 White, C.I., and Haber, J.E. (1990). Intermediates of recombination during mating type switching in Saccharomyces cerevisiae. EMBO J 9, 663–673.

Wilson, M.A., Kwon, Y., Xu, Y., Chung, W.H., Chi, P., Niu, H., Mayle, R., Chen, X., Malkova, A., Sung, P., et al. (2013). Pif1 helicase and Pol훿 promote recombination-coupled DNA synthesis via bubble migration. Nature 502, 393–396.

Wolner, B., van Komen, S., Sung, P., and Peterson, C.L. (2003). Recruitment of the recombinational repair machinery to a DNA double-strand break in yeast. Mol Cell 12, 221–232.

Wright, W.D., Shah, S.S., and Heyer, W.D. (2018). Homologous recombination and the repair of DNA double-strand breaks. J Biol Chem 293, 10524-10535.

Wu, L., and Hickson, I.D. (2003). The Bloom’s syndrome helicase suppresses crossing over during homologous recombination. Nature 426, 870–874.

Yamaguchi, M., Dao, V., and Modrich, P. (1998). MutS and MutL activate DNA helicase II in a mismatch- dependent manner. J Biol Chem 273, 9197–9201.

Zeng, Y., Li, J., Li, G., Huang, S., Yu, W., Zhang, Y., Chen, D., Chen, J., Liu, J., and Huang, X. (2018). Correction of the Marfan Syndrome Pathogenic FBN1 Mutation by Base Editing in Human Cells and Heterozygous Embryos. Mol Ther 26, 2631-2637.

Zhang, X.H., Tee, L.Y., Wang, X.G., Huang, Q.S., and Yang, S.H. (2015). Off-target Effects in CRISPR/Cas9-mediated Genome Engineering. Mol Ther Nucleic Acids 4, e264.

Zhang, Y., Hefferin, M.L., Chen, L., Shim, E.Y., Tseng, H.M., Kwon, Y., Sung, P., Lee, S.E., and Tomkinson, A.E. (2007). Role of Dnl4-Lif1 in nonhomologous end-joining repair complex assembly and suppression of homologous recombination. Nat Struct Mol Biol 14, 639-646.

Zhang, Z., Zhu, L., Lin, D., Chen, F., Chen, D.J., and Chen, Y. (2001). The three-dimensional structure of the C-terminal DNA-binding domain of human Ku70. J Biol Chem 276, 38231- 38236.

Zhu, Z., Chung, W.H., Shim, E.Y., Lee, S.E., and Ira, G. (2008). Sgs1 helicase and two nucleases Dna2 and Exo1 resect DNA double-strand break ends. Cell 134, 981–994.

Zinn, A. R., and Butow, R. A. (1985). Nonreciprocal exchange between alleles of the yeast mitochondrial 21S rRNA gene: kinetics and the involvement of a double-strand break. Cell 40, 887−895.

40 Chapter 2 – A Rad51-Indpendent Pathway Promotes Gene Editing

Abstract

The Rad51/RecA family of recombinases perform a critical function in typical repair of double- strand breaks (DSBs): strand invasion of a resected DSB end into a homologous double-stranded

DNA (dsDNA) template sequence to initiate repair. However, repair of a DSB using single- stranded DNA (ssDNA) as a template, a common method of CRISPR/Cas9- mediated gene editing, is Rad51-independent. We have analyzed the genetic requirements for these Rad51- independent events in Saccharomyces cerevisiae by creating a DSB with the site-specific HO endonuclease and repairing the DSB with 80-nt single-stranded oligonucleotides (ssODNs), and confirmed these results by Cas9-mediated DSBs in combination with a bacterial retron system that produces ssDNA templates in vivo. We show that single-strand template repair (SSTR), is dependent on Rad52, Rad59, Srs2 and the Mre11-Rad50-Xrs2 (MRX) complex, but unlike other

Rad51-independent recombination events, independent of Rdh54. We show that Rad59 acts to alleviate the inhibition of Rad51 on Rad52’s strand annealing activity via the Rad51 homolog

Rad55. Gene editing is Rad51-dependent when double-stranded oligonucleotides of the same size and sequence are introduced as templates. The assimilation of mismatches during gene editing is dependent on the activity of Msh2, which acts very differently on the 3’ side of the ssODN which can anneal directly to the resected DSB end compared to the 5’ end. In addition,

DNA polymerase Polδ’s 3’ to 5’ proofreading activity frequently excises a mismatch very close to the 3’ end of the template. We further report that SSTR is accompanied by as much as a 600- fold increase in mutations in regions adjacent to the sequences directly undergoing repair.

These DNA polymerase ζ-dependent mutations may compromise the accuracy of gene editing.

41 Summary

This chapter was published in (Gallagher et al., 2020). Nhung Pham in Grzegorz Ira’s lab contributed to data concerning the role of Rad59 in SSA in Figure 2. The manuscript was authored by Danielle N. Gallagher and James E. Haber. Danielle Gallagher performed data analysis, designed primers and constructed all strains and plasmids. Experiments in Figures 1 and 4 were performed with the help of Annie M. Tsai and experiments in Figure 3 were done with the help of Abigail N. Janto. Additional unpublished data have been included in this chapter.

This project was partially funded by the National Science Foundation under the

Graduate Research Fellowship Program, awarded to Danielle N. Gallagher, Fellow 2017240285.

Introduction

DSBs are repaired through one of two pathways: homologous recombination (HR) or nonhomologous end joining (NHEJ). Both classical NHEJ and microhomology-mediated end joining (MMEJ) involve DNA ligase-mediated joining of the broken chromosome ends, which usually results in small insertions or deletions (indels) at the junction (Moore and Haber, 1996;

Boulton and Jackson, 1996; Ma et al., 2003; McVey and Lee, 2008; Waters et al., 2014; Lemos et al., 2018). HR is a less mutagenic form of DSB repair, as it makes use of a homologous sequence as a donor template for repair. The template can be located on a sister chromatid, a homologous chromosome, or at an ectopic site. The majority of HR events are dependent on a core group of proteins, including the Rad51 strand-exchange protein that is responsible for homology recognition and initiating strand invasion into a double-stranded DNA (dsDNA)

42 template (Haber, 2016). In budding yeast, Rad51 interacts with and is assisted by several key recombination proteins, including the mediator Rad52 and the Rad51 paralogs, Rad55 and

Rad57, as well as the chromatin remodeler, Rad54 (Hays et al., 1995; Jiang et al., 1996; Clever et al., 1997; Sung et al., 1997; Petukhova et al., 1998; Mazin et al., 2000). Rad52 also plays a critical role in later steps of DSB repair, facilitating second-end capture of the DNA polymerase- extended repair intermediate (Lao et al., 2008).

However, some DSB repair events, though still requiring Rad52, are Rad51-independent.

The best studied mechanism is single-strand annealing (SSA), where homologous sequences flanking a DSB are rendered single-stranded by 5’ to 3’ exonucleases and then annealed, creating genomic deletions (Ozenberger and Roeder, 1991; Fishman-Lobell and Haber, 1992;

McDonald and Rothstein, 1994; Ivanov et al., 1996). SSA requires the Rad52 paralog Rad59, especially when the size of the flanking homologous regions is small (Sugawara et al., 2000). A second Rad51-independent mechanism involves break-induced replication (BIR) (Malkova et al.,

1996; Signon et al., 2001). Rad51-independent BIR is also independent of Rad54, Rad55, and

Rad57; however, Rad59 and a paralog of Rad54 called Rdh54/Tid1 assume important roles.

Rad51-independent BIR also requires the Mre11-Rad50-Xrs2 complex, whereas Rad51- mediated events and SSA are merely delayed by the absence of these proteins (Ivanov et al.,

1996). A third Rad51-independent pathway, another form of BIR, operates to maintain telomeres in the absence of telomerase (known as Type II events). Here too, Rad59 and the

MRX complex, as well as Rad52, are necessary, whereas the Type I Rad51-dependent telomere maintenance pathway does not require either Rad59 or MRX (Le et al., 1999; Teng et al., 2000;

Chen et al., 2001). Both Rad51-dependent and Rad51-independent forms of telomere

43 maintenance require the nonessential DNA polymerase 훿 subunit, Pol32, as do other BIR events

(Lydeard et al., 2007) Similarly, DSB repair by intramolecular gene conversion involving short

(33-bp) regions of homology is inhibited by Rad51 and is dependent on the MRX complex,

Rad59, and Rdh54 (Ira and Haber, 2002). Rad51-independent BIR pathways are also dependent on the Srs2 helicase that antagonizes loading of Rad51 onto resected DSB ends (Signon et al.,

2001; Sasanuma et al., 2013).

Use of the RNA-guided CRISPR/Cas9 endonuclease has revolutionized gene editing in eukaryotic systems ranging from yeast to mammals (Shen et al., 2013; Mali et al., 2013; DiCarlo et al., 2013). Guided endonucleases are programmed to create site-specific DSBs that can be repaired by providing a homologous template (Gallagher and Haber, 2018; Knott and Doudna,

2018). One approach that has been shown to be an efficient method of gene editing in a variety of eukaryotic systems is to introduce short single-stranded oligodeoxynucleotides

(ssODN) as a donor template (Storici et al., 2006; Davis and Maizels, 2016; Bothmer et al., 2017;

Paix et al., 2017).

Here we have examined the genetic requirements for single-strand template repair

(SSTR) in budding yeast, using two different systems: 1) an inducible HO endonuclease and an

80-nucleotide (nt) ssODN as a template for DSB repair, and 2) an optimized bacterial retron system to produce ssDNA templates in vivo with a targeted Cas9-mediated DSB. We confirm that in budding yeast, as in other eukaryotes, SSTR is a Rad51-independent mechanism, but show that this pathway is distinct from the previously-described Rad51-independent recombination pathways. SSTR depends on Rad52, Rad59, and Srs2 proteins, as well as the MRX complex, but is independent of Rdh54/Tid1. Surprisingly, deleting both Rad51 and Rad59 or

44 deleting Rad51 suppresses rad59Δ defect. We show a similar suppression of rad59Δ by rad51Δ in SSA. We conclude that Rad59 prevents Rad51 from inhibiting Rad52-mediated strand annealing. In contrast, this novel form of repair is specific to ssDNA, as dsDNA templates of the same size and sequence use a canonical Rad51-dependent process.

By analyzing the fate of mismatches between the ssODN and the target DNA, we show that the mismatches at the 5’ and 3’ ends of the template are differently incorporated into the gene-edited product. We show that both the MSH2 mismatch repair (MMR) protein and the 3’ to 5’ exonuclease activity of DNA Pol훿 play important roles in resolving heteroduplex DNA.

Finally, we demonstrate that SSTR is accompanied by as much as a 600-fold increase in mutations in the 1-kb region adjacent to the site of gene editing. These mutations are dependent on the error-prone DNA polymerase  that fills in single-stranded regions generated during DSB repair.

Results

Single Stranded Template Repair is Rad51-Independent

Gene editing using ssODNs in yeast is limited both by the efficiency of DSB initiation and by the efficiency of transformation to introduce the ssDNA template. As a model for DSB- induced gene editing, we used a galactose-inducible HO endonuclease to create a site-specific

DNA break at the MAT locus of chromosome 3, coupled with the introduction of the ssDNA template by transformation. In this strain, both HML and HMR donors have been deleted, so that a DSB that can only be repaired via NHEJ unless an ectopic donor is provided (Moore and

Haber, 1996). When HO is continually expressed, imprecise NHEJ repair occurs in approximately

45 2 x 10-3 cells, distinguishable by indels in the cleavage site that prevent further HO activity.

Repair by homologous recombination can be accomplished by introducing an 80-nt ssODN as a repair template. Cells were transformed with an ssODN template and then plated onto media containing galactose, which rapidly induces an HO-mediated DSB (Hicks et al., 2011). The template contains 37-nt of perfect homology to each end of the DSB, surrounding a 6-nt XhoI restriction site (Figure 2.1A). SSTR leads to the disruption of the HO cleavage site by the insertion of the XhoI site, whose presence can be confirmed by an XhoI digest of a PCR product spanning the region (Figure 2.1B). In WT cells, we achieve an editing efficiency of 75-90% among in DSB survivors, with the remaining survivors repaired via NHEJ; these indels are eliminated by mutants such as mre11Δ, rad50Δ, and yku70Δ that are known to be required for

NHEJ (Fig. 2.1C). However, survival in this assay is quite low, with an average survival rate of

2.8%; this low efficiency reflects limitations in transforming ssODN, as shown below.

We applied this assay to determine which recombination factors are required for SSTR.

Consistent with previous results in both budding yeast and metazoans (Davis and Maizels, 2014;

Keskin et al., 2014; Bothmer et al., 2017; Paix et al., 2017), SSTR proved to be Rad51- independent (Figure 2.1C). Furthermore, SSTR was significantly inhibited when Rad51 was overexpressed from an ADH1 promoter on a multicopy 2휇 plasmid. SSTR proved to be independent of the Rad51 paralog, Rad55, and the Rad54 translocase/chromatin remodeler, both of which are required for most DSB repair events that involve dsDNA templates. As expected, SSTR was dependent on the single-strand annealing protein Rad52, with essentially all rad52∆ survivors resulting from NHEJ events. There was also a significant reduction of SSTR

46 Figure 2.1 - SSTR is a novel form of Rad51-independent DSB repair. A) A DSB was created at the MAT locus via induction of a galactose-inducible HO endonuclease. A ssODN with 37-nt homology on either side of an XhoI site (yellow) provides a template for SSTR, resulting in the insertion of an XhoI restriction site into the MAT locus. B) Proportion of SSTR and NHEJ events in various mutants are determined via PCR of the MAT locus, followed by an XhoI digest. C) Viability determined by plate counts from galactose-containing media (induction media) over YEPD (non- induction media). D) Effect of double mutants on SSTR. E) Effect of checkpoints mutants on SSTR. * p < 0.01, comparing mutant’s to WT. Error bars refer to standard error of the mean. WT and rad51Δ n=9, all other mutants n=3 for SSTR.

in the absence of the Rad52 paralog Rad59, as well as less profound reduction in a strain lacking the helicase Sgs1, although sgs1Δ did not prove to be statistically significant in this assay

(Figure 2.1C). However, since this assay compares a large number of mutants, we used a strict statistical p-value of 0.01 and sgs1Δ has a p-value of 0.014 when looking at the total cell

47 viability. If we compare only SSTR events, sgs1Δ has a p-value of 0.02 compared to WT, still not meeting our cut-off for statistical significance.

In previous studies of Rad51-independent recombination, excluding SSA which is also

Rad51-independent, both Rdh54 and the MRX complex were required (Signon et al., 2001; Ira and Haber, 2002). For SSTR, while MRX is required, Rdh54 is not, distinguishing SSTR from previously studied Rad51-independent pathways. SSTR is also independent of Pol32. It is also notable that while the MRX complex is required, Sae2 – which often functions in conjunction with MRX in regulating end-resection but not in DSB end-tethering or other functions (Ferrari et al., 2015; Symington, 2016; Cassani et al., 2018) – is not. A sae2Δ strain is still capable of both

SSTR and NHEJ.

SSTR also requires the Srs2 helicase (Figure 2.1C). One major function of the Srs2 helicase is to act as an anti-recombination factor by stripping Rad51 from the ssDNA tails formed after resection (Krejci et al., 2003; Veaute et al., 2003); thus, srs2Δ might mimic the inhibition of SSTR that is seen when Rad51 is overexpressed. Indeed, deleting RAD51 suppressed srs2Δ’s defect in SSTR (Figure 2.1D). Moreover, deleting RDH54 suppressed the defect in srs2Δ. Although these results might suggest that Rdh54 acts in the same pathway as

Rad51, we do not believe this to be the case since their deletions behave differently in other genetic combinations (see below).

We also examined the role of several genes that are involved in the 5’ to 3’ resection of

DSB ends: the Exo1 exonuclease and the Sgs1-Rmi1-Top3-Dna2 helicase/endonuclease complex

(Mimitou and Symington, 2008; Zhu et al., 2008). Deleting either Sgs1 or Exo1 had no significant effect on the efficiency of XhoI insertion, and neither did the sgs1Δ exo1Δ double mutant

48 (Figure 2.1D). These results suggest that the MRX complex can provide sufficient end resection to allow SSTR involving the homologous 37 nt of the ssODN donor. Deleting the Fun30 SWI/SNF chromatin remodeler has also been shown to strongly retard 5’ to 3’ resection of DSB ends

(Chen et al., 2012; Costelloe et al., 2012; Eapen et al., 2012), but we found a significant increase in SSTR. We note that in previous research using human cancer lines, SSTR was dependent on proteins in the Fanconi anemia (FA) pathway (Richardson et al., 2018). The helicase function of

Mph1 is the only homolog of the FA pathway found in yeast, but Mph1 does not appear to play a role in SSTR in our system (Figure 2.1C).

It is also worth noting that components of the yeast DNA damage response (DDR), the

ATR homolog Mec1 and the ATM homolog Tel1, which is largely dispensable in the DNA DSB response, are not required for SSTR (Figure 2.1E) (Mallory and Petes,2000; Mantiero et al.,

2007; Gobbini et al., 2013). The role that the checkpoint plays in SSTR remains to be further studied.

Rad59 regulates Rad52-mediated strand annealing

Given that Rad59 has an important role in SSTR, we further examined the genetic interaction between Rad51 and Rad59. Previous biochemical studies (Wu et al., 2008) had suggested that Rad59 might mediate the ability of the Rad52 protein to facilitate single-strand annealing, which is the first step in SSTR (Figure 2.1A), specifically by modulating the inhibitory effect of Rad51 on Rad52’s strand annealing activity. Indeed, deleting Rad51 suppressed the inhibition of SSTR by rad59Δ and increased the rate of SSTR significantly higher than observed in WT cells or in the absence of Rad51 (Figure 2.2A). We then examined a separation-of-

49 function mutation of Rad52, rad52-R70A, that is proficient for loading of Rad51 onto ssDNA but fails to carry out strand annealing (Shi et al., 2009). SSTR is severely impaired in a rad52-R70A mutant, similar to rad59Δ’s phenotype (Figure 2.1B; Figure 2.2A). However, deletion of Rad51 did not suppress rad52-R70A, supporting the conclusion that Rad59 affects Rad51’s modulation of Rad52-mediated strand annealing activity. Since deletion of RAD51 in a rad59Δ or a rdh54Δ

Figure 2.2 - Rad59 alleviates the inhibition of Rad51 on Rad52’s annealing activity. A) Viability of single and double mutants following SSTR, determined by plating on galactose- containing media. B) An HO-induced DSB results in repair via SSA between partial leu2 gene repeats located 5 kb apart on the left arm of chromosome 3. C) Representative southern blots showing DSB repair products by SSA in WT and indicated mutants. D) Viability of mutants on galactose-containing plates, where HO DSBs are repaired via SSA (mean + SD; n=3). Welch’s t-test was used to determine the p-value. E) Graphs show quantitative densitometric analysis of repair efficiency by 6 hr compared to WT (mean + SD; n=3). Welch’s t-test was used to determine the p-value.

50 background had a marked increase on the cells ability to undergo SSTR, we also asked if rad59Δ rdh54Δ might show a similar increase in SSTR. However, rad59Δ is not suppressed by rdh54Δ

(Figure 2.2A).

To confirm that Rad59 affects the strand-annealing function of Rad52, we turned to a well-characterized SSA system, in which a DSB promotes formation of a deletion between flanking repeated sequences (Vaze et al., 2002). Previous studies have shown that Rad59 is important in SSA, especially when the length of the flanking homologous repeats is short, below a few hundred base pairs (Sugawara et al., 2000). We used an HO-induced DSB within a leu2 gene, which is repaired by SSA with a direct “U2” repeat (1.3 kb) located 5 kb away (Figure

2.2B), resulting in a chromosomal deletion of the sequences located between the two repeats, as well as one of the partial copies of the leu2 gene. Like SSTR, SSA is severely impaired in a rad52-R70A mutant (Figure 2.2C). Deletion of Rad59 reduced the efficiency of SSA of the 1.3-kb repeats to approximately 30% (Figure 2.2D). Deletion of Rad51 by itself only had a mild negative effect on SSA, but deletion of Rad51 partially suppressed the SSA deficiency of rad59Δ

(Figure 2.2D). As with SSTR, this suppression was not observed Figure 2.3 - Rad59 interacts with Rad51 via the Rad51 in either rad52-R70A rad51Δ or rad52-R70A rad59Δ, again homolog Rad55. A,B) Viability of single and double mutants following SSTR, determined by plating galactose-containing media. Significance determined using t-test with Welch’s correction. * p < 0.01, comparing ot WT. Error bars refer to standard error of the mean. WT and rad51Δ n=9, all other mutants n=3 for SSTR, n=3

51 suggesting that Rad59 specifically plays a role in the modulation by Rad51 of Rad52-mediated strand annealing; however we did not see the same large increase in the rad51Δ rad59Δ that we do in SSTR, suggesting that Rad59 might have an additional role in SSTR initiated by the HO- endonuclease or when there is limited homology in the donor sequence.

This modulation is partially due to interaction with the Rad51 homolog, Rad55-Rad57, as a rad55Δ strain shows a partial rescue of the rad59Δ phenotype (Figure 2.3A). The deletion of

Rad55, or components of another Rad51 homolog – the SHU complex – have no effect on ability of SSTR on their own (Figure 2.3B). Therefore, the Rad55-Rad57 heterodimer, and potentially the SHU complex, are not directly involved in SSTR, but work specifically to deter

Rad59’s inhibition of Rad52. The role of the SHU complex remains to be studied.

Pol is responsible for target-adjacent mutagenesis

Previous research has suggested that DSB repair that involves DNA resection is highly mutagenic because ssDNA regions created during resection must be filled in once HR has completed (Rattray et al., 2002; Sinha et al., 2017; Lee et al., 2019). Gap-filling, either by DNA polymerase 훿 or by translesion DNA polymerases such as Pol have been shown to be responsible for mutation rates 1000-fold over background spontaneous mutation rates in canonical DSB repair (Holbeck and Strathern, 1997; Hicks et al., 2010). Since SSTR is a form of

HR and likely involves extensive end-resection and gap-filling, it seemed possible that there are significant off-target effects that have not previously been considered. To examine this possibility, the yeast URA3 gene within an MX cassette was inserted 200 bp centromere- proximal to the HO cleavage site, such that the URA3 sequences themselves are approximately

52 400 bp beyond the 37 nt of homology shared between the DSB end and the ssODN template.

These cells were then targeted in the same XhoI insertion assay previously described (Fig. 2.4A).

SSTR survivors were collected and then replica-plated onto 5-fluroorotic acid media (FOA), which selects for ura3 mutants (Boeke et al., 1987). Compared to the spontaneous mutation rate of URA3 mutations (3.5 x 10-7), determined by fluctuation analysis (Gillet-Markowska et al.,

2015), there was an almost 600-fold increase in ura3 mutants after gene editing events (Fig.

Figure 2.4 - Target-adjacent mutagenesis is dependent on Pol. A) Effect of SSTR on a URA3 gene integrated near the site of HO cleavage. Mutations in URA3 following SSTR were collected by replica plating survivors on galactose media onto 5-FOA medium. B) The increase in mutation rate in SSTR over the spontaneous mutation rate (determined by a fluctuation analysis) was determined at the indicated locations surrounding the DSB site. C) Effect of deleting Rev1 and Rev3 components of Pol. D) Cell viability following SSTR plating assay. Significance determined using an unpaired t-test with Welch’s correction, comparing mutants to WT. Error bars refer to standard error of the mean. Spontaneous mutation rate determine by fluctuation analysis, n=10. SSTR mutation rate, n=3.

53 2.4B). However, this increased mutagenesis was confined to a region close to the area of the

DSB, as the rate of mutagenesis dropped significantly as the URA3 marker was inserted further upstream. For a site that was approximately 650 bp upstream of the DSB, there was an approximate 50-fold increase in mutagenesis, whereas at 2.2 kb there was only an 8-fold increase. There was a nearly 200-fold increase in ura3 mutations when the URA3 marker was located 550 bp downstream of the DSB. We could not investigate the effects of moving the ura3 marker further downstream of the DSB given that Taf2, an essential gene, is located 3’ of the HO cleavage site. There is a statistically significant difference between integrating URA3 upstream and downstream of the HO cleavage site. It is possible that this difference is simply due to the increased distance, however, it raises the possibility the two sides of the DSB engage different repair machinery, discussed below.

The source of the frequent ura3 mutations appears to be dependent on the error-prone

DNA polymerase Pol, as deleting either the Rev1 or Rev3 components of Pol, resulted in very few ura3 mutants (Figure 2.4C). However, neither deletion of Rev1 nor Rev3 affected cell viability, indicating that other, less mutagenic mechanisms can be used for gap-filling in the absence of Pol (Figure 2.4D). When using dsDNA of the same size and sequence, we found an approximately 100-fold increase in target-adjacent mutagenesis (Figure 2.4C). These mutagenic events were still dependent on the activities Pol These data suggest that adjacent off-target effects of gene editing pose a danger that should be ruled out in selecting gene-editing events.

54 Additional Genetic Requirements of SSTR Depend on Template Design

To test if changing the design of the ssODN donor altered the genetic requirements of

SSTR, we used a ssODN similar to that described in Figure 2.1A, except that the 37 nt of homology on each side of the XhoI site were each targeted to sequences that are 500 bp from the DSB; thus, successful gene editing via the 80-nt ssODN creates a 1-kb deletion flanking the

XhoI site (Figure 2.5A). Successful SSTR should then only occur after extensive 5’ to 3’ resection of the DSB ends. Repair efficiency, as measured by viability, using this donor template was significantly lower than the ssODN that simply incorporated an XhoI restriction site

(approximately 1% compared to 3%) (Figure 2.5B).

The core recombination requirements of SSTR for this configuration were the same as those seen with the simple XhoI insertion, as gene editing was independent of Rad51, Rad54,

Rad55, Rdh54, and Sae2, but still dependent on Rad52, Rad59, Srs2, and the MRX complex

(Figure 2.5C). As before, srs2Δ was suppressed by both rad51Δ and rdh54Δ, and there were still substantial increases in rad51Δ rad59Δ and rad51Δ rdh54Δ compared to wildtype or rad51∆

(Figure 2.5C and 2.5D). Moreover, using an ssODN that creates a large deletion imposes additional requirements. For the ssODN to pair with a resected DSB end, there must be extensive 5’ to 3’ resection. While deleting either the Sgs1 or Exo1 individually had no significant impact, the double mutant sgs1Δ exo1Δ abolished SSTR (Figure 2.5D). This result stands in contrast to the lack of effect of the double mutant in the simple incorporation of the

XhoI site and emphasizes the need for long-range 5’ to 3’ resection (Figure 2.1D). The effect of blocking long-range resection in sgs1∆ exo1∆ was not mimicked by deleting Fun30, whereas in

55 previous studies examining 5’ to 3’ resection of DSB ends fun30∆ significantly slowed resection similar to sgs1∆ exo1∆ (Chen et al., 2012; Costelloe et al., 2012; Eapen et al., 2012). Previously, deleting Fun30 was shown to protect double-stranded DNA fragments used in “ends-out”

Figure 2.5 - The genetic requirements of SSTR are dependent upon ssODN template design. A) SSTR using a ssODN with 37-nt homologies located 500 bp on either side of the HO-induced DSB. SSTR results in a 1-kb deletion and the incorporation of the XhoI restriction site into the MAT locus, which can be screened via PCR and restriction digest with XhoI. B) cell viability following SSTR with an 80-nt SSTR template that creates a 6-bp insertion versus a 1-kb deletion. C) Viability determined by galactose-induction plate counts over YEPD plate counts. Proportion of SSTR and NHEJ events in various recombination mutants determined via PCR. D) Effect of double mutants on deletions created by SSTR. Significance determined using an unpaired t-test with Welch’s correction comparing mutant’s to WT, * p < 0.01, ** p < 0.001, *** p < 0.0001. Error bars refer to the standard error of the mean. WT and rad51Δ n=9, all other mutants n=3.

56 transformation (Chen et al., 2012), but in the present scenario, the transformed DNA is single- stranded. Whether Fun30 also affects the stability of ssODNs is not clear.

Another requirement in the deletion assay was for Rad1, and presumably Rad10, which together act as a 3’ flap endonuclease that can remove the 3’-ended 500-nt nonhomologous tail that would be created by annealing the ssODN to its complementary strand (Fishman-Lobell and Haber, 1992; Bardwell et al., 1994) (Fig. 4A). Rad1 was not needed in the simple XhoI insertion assay (Figure 2.1C).

Genetic Requirements of SSTR are Identical Using Cas9

Although it was convenient to survey many mutations using the highly efficient and easily inducible HO endonuclease, we confirmed that the same genetic requirements apply when a DSB is created by CRISPR/Cas9. To overcome the low efficiency of transforming ssODNs into yeast and to better screen Cas9-mediated gene editing events, we turned to a modified version of the CRISPEY system to produce ssDNA templates in vivo (Sharon et al., 2018). This system utilizes a yeast-optimized E. coli retron system, Ec86, to generate designer ssDNA sequences in vivo. Retrons are natural DNA elements encoding for a reverse transcriptase (RT) that acts on a specific consensus sequence to generate single stranded DNA products (Hsu et al., 1990; Hsu et al., 1992; Shimamoto et al., 1993). These ssDNA products are covalently tethered to their template RNA by the RT, however after reverse transcription the RNA template is degraded (Miyata et al., 1992; Mirochnitchenko et al., 1994). The CRISPEY system utilizes a chimeric RNA of Ec86 joined to the gRNA scaffolding of Cas9 at the 3’ end (Sharon et al., 2018). By integrating a yeast-optimized galactose-inducible Cas9 and retron (RT) onto

57 chromosome 15, and using a CEN/ARS plasmid containing a galactose-inducible gRNA linked to the retron donor template (Figure 2.6A; 2.6B), we were able to achieve high efficiency of Cas9- mediated gene editing at two different chromosomal locations, within the MAT locus near the

HO cleavage site, and at a 5-bp insertion in the lys5 locus (Figure 2.6C). Compared to the <3% of cells that properly inserted the XhoI site at MAT with an HO-induced DSB and a transformed ssODN template, the retron system yielded efficiencies of >20%. At lys5, successful SSTR via an

80-nt retron-generated ssDNA donor carrying the wild type LYS5 sequence resulted in Lys+ recombinants that were recovered at a rate of 34%, compared to the <1% of lysine prototrophic events in cells with the same Cas9-induced DSB, but in the absence of the retron donor to provide an ssDNA repair template (Figure 2.6C, 2.6D, and 2.6E).

To test whether the genetic components of SSTR were the same with a Cas9-induced

DSB, we introduced gene deletions in this strain background. The requirements were generally the same as for HO-induced SSTR events, being independent of Rad51, Rad55, and Fun30, but still dependent on Rad52, Rad59, Srs2, and the MRX complex (Figure 2.6D, 2.6E). With the retron system and Cas9 endonuclease, the double mutants rad51Δ rad59Δ and rad51Δ rdh54Δ did not show the same significant increase in SSTR above WT levels as we observed with the

HO-endonuclease. This difference could be explained by several different reasons. First, the 3’ end of the retron ssDNA is covalently linked to the gRNA, so the template itself may limit access to the repair machinery from the 3’ end. In addition, the tethering of the template to Cas9 could change the dynamics of the homology search (Roy et al., 2018). There

58 Figure 2.6 - SSTR initiated by retron-Cas9 utilizes a Rad51-indepent repair pathway. A) Galactose-inducible, yeast optimized spCas9 was introduced into the trp1 locus along with a galactose-inducible, yeast optimized retron, Ec86 (labelled RT). Upon galactose induction apo- Cas9 and the retron are transcribed. The blue region of the ssDonor (single-stranded donor)

59 Figure 2.6 - continued. is the donor sequence to repair the DSB break, while the red region refers to the 34-bp consensus region that the retron binds to on the mRNA transcript to initate reverse transcription, and the yellow region represents the termination sequence. The ssDonor and the gRNA are constitutively active. Galactose-induction results in either an irreparable DSB (shown on the left), since no donor is encoded in the strain, or repair of Cas9 cleavage via the reverse transcribed retron system. B) The retron system utilizes a modified Cas9 gRNA that tethers the ssDNA donor template to the RNA scaffolding of the Cas9 protein. Successful SSTR at the MAT locus results in the insertion of an XhoI restriction site, as in Figure 2.1. SSTR at the lys5 locus repairs a 5-bp insertion in the lys5 locus, resulting in Lys+ recombinants. C) Efficiency of the Retron-Cas9 system at two chromosomal locations. Cells were plated onto URA- plates with dextrose (non-induction) and URA- with galactose (induction) media. At MAT, the percent gene editing was determined by PCR and XhoI digest of induction survivors as described in Methods. At lys5, the percentage of gene editing was determined by replica plating URA-Gal survivors onto Lys- media. The resulting plate count over plate counts of URA- non-induction media results in % gene editing. D) Effect of recombination mutants on retron-Cas9 SSTR gene editing. After induction of the retron system on galactose-containing media, survivors were replica plates to Lys- media. The frequency of Lys+ colonies was calculated as a percentage of total cells plated and normalized to wild type. E) Cas9-retron system shows similar genetic requirements as HO and MAT locus. Significance was determined using two-tailed t-tests compared to WT, using the two-stage Benjamini, Krieger, and Yekutieli false discovery rate approach [88], * p < 0.01, ** p < 0.001, *** p < 0.0001. Error bars refer to the standard error of the mean. n=3. rad59Δ compared to rad59Δ rad51Δ p = 0.009

could be other differences as well, as Cas9 may stay bound to DNA after cleavage, although how long it remains bound in vivo is not clear (Nickloff et al., 1986; Shibata et al., 2017).

Mismatch repair acts differently at the 5’ and 3’ ends of the ssODN

How SSTR occurs is still not fully understood. One question concerns the fate of the ssDNA template strand itself. Another concerns the fate of mismatches between the template strand and the complementary single-stranded DSB end. We used the same ssODN described in

Fig. 1, but now using ssODNs that contained mismatches in the donor sequence. One donor had

4 mismatches 5’ to the XhoI site, while the second had 4 heterologies on the 3’ side, spaced

60 every 9-nt (Figure 2.7A). We note that mismatch position 1 is only 2-nt away from the end of the ssODN. We observed that there was a significant decrease in gene editing with 4 mismatches on either the 5’ or the 3’ side, compared to the fully homologous template, although the effect was more pronounced on the 3’ side (Figure 2.7B). This difference may reflect the fact that the initial annealing steps in SSTR can only happen on the side 3’ to the XhoI site and may be quite different from the capture of the second end.

In studies of SSA and DSB-mediated gene conversion, the inhibitory effects of a small percentage of mismatches could be overcome by deleting Sgs1 or components of the mismatch repair system (Spell and Jinks-Robertson, 2004; Sugawara et al., 2004). Here, however, deleting

Sgs1 did not suppress the effect of the four mismatches. In fact, with mismatches on the 3’ side of the XhoI site, the majority of survivors in sgs1Δ were NHEJ events. Deleting Sgs1 also consistently reduced SSTR in the fully homologous case, though not statistically significantly in any one assay (Figure 1 and 5).

Deleting the mismatch repair gene, MSH2, did not suppress the reduced level of SSTR in the templates carrying 4 mismatches (Figure 2.7B), but there was a notable change in the inheritance of these mismatches (Figure 2.7C). We analyzed the DNA sequences of 23 SSTR events in both wild type and msh2∆ strains for each of the ssODNs carrying mismatches (Figure

2.7C). In wild type strains, a majority of XhoI insertions were accompanied by co-inheritance of

3 of the 4 mismatches. However, in an msh2Δ strain, the majority of SSTR events using the 5’ mismatch ssODN template showed heteroduplex tracts, as evidenced by the presence of both

61 Figure 2.7 - Gene editing events are dependent on the activity of MSH2. A) Experimental set- up to determine the effect of mismatches on SSTR. Mismatches are spaced every 9 nt from the XhoI site in the center of the template. Mismatch 1 is located only 2 nt from the terminus of the ssODN. SSTR events are determined by incorporation of an XhoI restriction site. B) Significance determined using an unpaired t-test compared to WT of the same template, * p < 0.01. n=3. C) Inheritance of mismatches in wild-type (left) and msh2 (right) strains. The inheritance of markers was determined separately by sequencing SSTR survivors using ssODNs with mismatches 5’ or 3’ of the XhoI site. The different outcomes were grouped together for comparison. D) Sequencing n=23 per template per genetic background. E) Viability test of cells ability to survive DSB repair via SSTR with a template containing a chemical modification on the terminus of the ssODN. Significance determined using an unpaired t-test with wild-type levels using a template that does not have chemically modified ends, * p < 0.01. Error bars refer to standard error of the mean. n=3.

the chromosomal and mutant alleles at these sites when DNA from single transformant colonies was sequenced. On the 3’ side, msh2Δ eliminated the great majority of events in

62 which the mutations in the ssODN were inherited into the gene-edited product. These results extend the conclusions reached by Harmsen et al. studying SSTR in mammalian cells, where the absence of mismatch repair largely prevented incorporation of heterologies on the 3’ half of the ssODN, while not preventing their assimilation in the 5’ half (Harmsen et al., 2018). In their studies of mammalian cells, it was not possible to detect the presence of unrepaired heteroduplex DNA, as we show in Figure 2.7C.

We noted that mismatches located 2-nt from either end of the ssODN were only rarely assimilated into the gene-edited product (Figure 2.7C). In our recent study of break-induced replication (BIR), we discovered that heteroduplex DNA created by strand invasion was corrected (i.e. mutations were assimilated into the BIR product) in a strongly polar fashion from the 3’ invading end (Bai and Symington, 2003). Moreover, these corrections of the heteroduplex were orchestrated by the 3’ to 5’ exonuclease (proofreading) activity of DNA polymerase , which removed up to 40 nt from the invading end and replaced them by copying the template. Incorporations of the mismatched base from the template was almost completely abrogated by eliminating the proofreading activity of DNA polymerase 훿 (pol3-01). Here, using the XhoI insertion ssODN with 4 mismatches on one side or the other, we found that the overall-incorporation of mismatches was unaffected by proofreading-defective mutations in

Pol휀 (pol2-4) or Pol훿 (pol3-01), with one notable exception: the mutation 2-nt from the 3’ end of the template was incorporated at a very high level in pol3-01 mutants (Figure 2.7D). These data suggest that Pol훿 can be loaded not only onto the 3’ end of the chromosomal DSB, where it initiates copying of the rest of the ssODN template, but can also be recruited by the 3’ DNA end of the ssODN itself and then chew back the 3’ end. It is possible that Pol훿 might also extend

63 this end of the ssODN template and raises the possibility that the ssODN itself could be incorporated into the gene-edited product in some cases. The pol3-01 mutation did not affect the assimilation of the most terminal mismatch on the 5’ end of the ssODN (Figure 2.7D).

Effect of modifying the 5’ and 3’ ends of the ssODN

Work by Harmsen et al. in mammalian cells also showed that blocking the ends of the template strand reduced SSTR, suggesting that the ssDNA strand itself might be more than a simple template that anneals with a DSB end and is then copied by a DNA polymerase

(Harmsen et al., 2018). If the ssODN might be assimilated into the product, then blocking access to either 5’ or 3’ end of the template might affect its usage. We used the XhoI insertion ssODN with complete homology to the chromosomal site, except that these templates were chemically modified with either an inverted thymine at the 5’ end, or an inverted dideoxy-thymine on the

3’ end. These chemical modifications should prevent ligation of the ssODN into chromosomal

DNA, since the terminal thymines cannot pair with the resected chromosomal DNA, and should block extension of the 3’ end by Pol훿 unless the block is excised. There was no significant difference between either modified and unmodified donor templates in wildtype cells or in a pol3-01 strain. An alternative way that a modified 3’ end nucleotide might be removed would be through the use of the Rad1-Rad10 flap endonuclease. Indeed, we found a modest reduction in SSTR with the 3’ block in a rad1Δ strain when compared to wildtype cells (Figure 2.7E). These results indicate that the template might sometimes be ligated into the repaired product, or that the inverted thymine causes increased rejection of the ssODN as a suitable repair template.

64 We propose a model for SSTR where edits templated by the 5’ and the 3’ are incorporated through different mechanisms. Incorporation of mismatches on the 3’ end of the

XhoI site should occur only during the time that the resected DSB end has paired with the

Figure 2.8 - Model of heteroduplex formation and incorporation of mismatches during SSTR. Fate of an ssODN with 4 mismatches either 5’ to the XhoI site (A) or 3’ (B). After the DSB is created and resected only the strand on the right can anneal to the template. This annealing creates a heteroduplex that may be repaired by the mismatch repair machinery including Msh2. Heterologies close to the 3’ end of the invading strand, but also at the 3’ end of the ssODN, can be excised by the 3’ to 5’ exonuclease activity of DNA polymerase . Only if the heteroduplex is converted to the template strand genotype will these mismatches be incorporated into the SSTR product (B). Mismatches 5’ to the XhoI site will be obligately copied by DNA polymerase after strand invasion (A). The dissociation of the newly copied strand allows it to anneal with the resected second end of the DSB, creating an obligate heteroduplex. Dissociation of the newly copied strand may occur without copying the end of the ssODN template. Heteroduplex DNA may then be corrected to the genotype of the donor template or left as unrepaired heteroduplex. In the absence of Msh2, most outcomes will have heteroduplex to the left of the XhoI site but no incorporation to the right, resulting in sectored colonies.

65 donor, to prime DNA polymerase to copy the template strand (Figure 2.8). On the 5’ side, however, the initial copying of the template and its subsequent annealing to the second DSB end should obligately produce heteroduplex DNA that will be resolved by mismatch repair to be fully mutant or fully wild type, dependent on the activity of Msh2 (Figure 2.8). We suggest that the failure to incorporate edits located close to the 5’ end into the final product might occur if the DNA polymerase copying the template dissociates before it has copied the last several nucleotides, so this site is not incorporated as heteroduplex DNA involving the second DSB end; alternatively, there could be a Msh2- and DNA polymerase proofreading-independent mechanism that corrects the heteroduplex in favor of the chromosomal sequence (Figure 2.8).

SSTR is Specific to Single-Stranded DNA

Our SSTR assays employ templates containing only 37-nt of homology on either side of the DSB. We wanted to know if this noncanonical repair pathway is specific to ssDNA, or might also apply to repair with dsDNA templates with the same limited homology. We annealed complementary 80-nt ssDNA oligonucleotides to create a dsDNA template with free ends that had 37-bp of perfect homology flanking each side of a 6-bp XhoI restriction site. After duplexing, the pool of dsDNA template was treated with S1 nuclease to degrade any remaining non-duplexed ssDNA. We transformed the template into cells using the same protocol used with the ssDNA templates (Figure 2.9A). Double-stranded template repair (DSTR) proved to be more than five times as efficient as SSTR (Figure 2.9B). With this short dsDNA template, the repair process shifted to a Rad51-dependent event, now also requiring Rad54, Rad55 and

Rdh54, but still dependent on the MRX complex, Srs2 and Rad59 (Figure 2.9C).

66 Figure 2.9 - SSTR is distinct from DSTR. A) Successful DSTR results in the incorporation of the XhoI restriction site into the MAT locus, which is screened for via PCR followed by an XhoI restriction digest. B, C, D) Viability tests to determine strains ability to undergo DSB repair with designate 80-nt ssDNA or 80-bp dsDNA template. Viability determined via plate counts of galactose- induction media over YEPD non-induction media. Significance determined using a paired t-test with Welch’s correction, * p < 0.01, ** p < 0.001. E) Sequencing of MAT locus following repair from each of two 80-bp templates, with 4 mismatches at the 5’ or 3’ side of the XhoI site (n=24 in each case).

With DSTR templates carrying 4 mismatches on either side, the efficiency of repair dropped markedly compared to wildtype, similar to what we see with SSTR (Figure 2.9D). We note that because both the resected 3’ end of the DSB on the left or the right could initiate strand invasion, and the template itself is double-stranded, we cannot account for the difference in the effect of the mismatches on the upstream or the downstream side in the same terms we used with the SSTR substrates. It is possible that the two ends of the DSB do not participate equally in initiating DSTR. This possibility is supported by sequencing data, which show that in wildtype cells a mismatch 9 bp away from the upstream side of the DSB will be incorporated into the genome in 78% of survivors, while a mismatch 9 bp away from the

67 downstream side of the DSB will only be incorporated in 48% of survivors (Figure 2.9E).

Additionally, the pol3-01 proofreading-defective allele of Pol훿 shows a higher rate of mismatch incorporation on the downstream side of the DSB compared to wildtype cells, indicating that

Pol훿 may still be resecting and replacing the 3’ end of the annealed template strand. Given the apparent symmetry of this DSTR system, it is difficult to explain the different results on the upstream and the downstream sides of the template.

Figure 2.10 - Model of SSTR. 1) MRX binds to the ends of the DSB, initiating end resection. 2) Rad51 binds to ssDNA tails. 3) The 3’ end of the ssDNA template anneals to complementary resected DNA, dependent on Rad52. A mismatch in homology is shown in red. 4) A DNA polymerase synthesizes across the template. 5) Rad52 anneals the newly synthesized and displaced DNA to the 5’-3’ resected second end of the DSB. 6) Msh2 corrects heteroduplexes that form, fixing edits into the genome. 7) Second-strand synthesis occurs. 8) Gaps are filled in by translesion Pol.

Discussion

Although CRISPR/Cas9 has made it possible to generate specific changes to the genome in many organisms, budding yeast still serves as an important resource to determine the mechanism of gene editing, and thus to optimize experimental design. We show that gene editing using ssODN templates utilizes a novel pathway of DSB repair that is independent of the

68 canonical DSB repair proteins Rad51, Rad54, Rad55 and Rdh54/Tid1, but still depends on

Rad52, Rad59, Srs2, and the MRX/MRN complex. When the ssODN templates are designed to create a large deletion, Rad1-Rad10 flap endonuclease and the long-range resection machinery also become essential. We have confirmed that the genetic requirements of SSTR are generally the same in HO-mediated SSTR as with Cas9-mediated events. Moreover, the mechanism of

SSTR is specific to ssDNA templates, as gene editing using a dsDNA template of the same size and sequence (with 37 bp homology to each DSB end) switches to a Rad51-dependent mechanism that is nevertheless distinct from that involving gene conversion between long regions of homology, most notably by its requirement for Rdh54 and for the MRX complex.

Although Rad51-independent BIR events require Pol32, it also is not necessary in SSTR, possibly because of the length of new DNA synthesis can be accomplished without additional processivity factors. The role of Sgs1 remains to be investigated.

We have also determined an important role of the Rad52 paralog Rad59, which apparently acts to alleviate the inhibition of Rad51 on Rad52’s ability to anneal ssDNA tails.

Previous studies have shown that Rad51 impairs Rad52’s ability to anneal DNA strands, but this can be overcome by Rad59 (Anand et al., 2017a). In both SSTR and SSA, deleting Rad51 suppresses rad59. We note that suppression of rad59 by rad51 is quite a different relationship than is seen in spontaneous recombination between chromosomal regions, where rad51 rad59 is much more severe than rad59 or rad51 alone (Bai and Symington, 1996).

In some of our assays the efficiency of SSTR is significantly greater for rad51 rad59 than for rad51 alone; this result suggests that Rad59 may impair SSTR in other ways than simply modulating Rad51. Rad59 interacts directly with Rad52 and may form heteromeric rings (Davis

69 and Symington, 2003; Cortes-Ledesma et al., 2004); possibly without Rad59, Rad52’s annealing activity is intrinsically greater. We note that Rad59 is also important in DSTR with very short homology, while its role in DSB repair involving chromosomal regions with much longer homology is not critical (Signon et al., 2001). There are two distinct strand-annealing steps in

SSTR: the initial annealing between the resected end of the DSB and the 3’ end of the ssODN template, followed by second-end capture in which the newly copied strand anneals to the other resected end of the DSB. In DSTR, Rad51 is apparently needed for the initial invasion of the resected end into the dsDNA template, but there would still be a second-end capture process where Rad59 might be critical when the homology is very short (in our experiments, only 37 nt). Similarly, the Srs2 helicase is essential for SSTR (and DSTR). The role of Srs2 as an anti-recombinase to displace Rad51 appears to account for its importance, as rad51 suppresses srs2.

We do not yet understand the role of RDH54. In other contexts, HO endonuclease- induced Rad51-independent recombination, involving either intrachromosomal plasmid recombination or interchromosomal BIR events, requires Rdh54 (Ira and Haber, 2002); but in

SSTR and DSTR, Rdh54 is not required. Recent studies have suggested that a major role for

Rdh54 is in controlling the size of a strand invasion D-loop (Piazza et al., 2019). However, in

SSTR there is no D-loop but only an annealed structure between the 3’ end of the ssODN and the resected DSB end. How such strand invasions occur in Rad51-indpendent events remains unknown, but apparently Rdh54 is important in facilitating this event. Why rdh54 suppresses srs2 in SSTR also remains unclear. Without Srs2, more Rad51 will be loaded onto the ssDNA end of the DSB, and potentially also on the ssODN, and stabilization of the Rad51 filament

70 structure when the regions are short may depend on Rdh54. Rdh54 also plays a novel role in interchromosomal template switching during DSB repair, where a partially copied strand of

DNA jumps from one template to another (Anand et al., 2014; Anand et al., 2017a). These secondary jumps are impaired in rdh54 but simple gene conversion events, copying one ectopic template, are not. Possibly the dissociation of the newly copied first strand in SSTR from its short template (Figure 2.8 and 2.10) requires this template-jumping activity.

SSTR shares some features with other HR pathways that involve short regions of homology (Le et al., 1999; Ira and Haber, 2002), such as the need for the MRX complex. In DSB repair events that involve longer (>200 bp) homology, deleting components of MRX delays but does not diminish HO-induced DSB repair (Ivanov et al., 1992); yet with short substrates MRX plays a central role, both for SSTR and DSTR. Yeast MRX has been implicated in many early steps in DSB repair (Gobbini et al., 2018); it is required for most NHEJ events, can bridge DSB ends, promote short 3’-ended ssDNA ends by 3’ to 5’ resection from a nick, promote the loading of

DSB-associated cohesin, and more. How it is implicated in SSTR and DSTR is not yet clear.

Whatever steps require MRX, they do not need Sae2.

As expected, the presence of mismatches in the ssDNA template reduces the efficiency of SSTR. However, the degree of inhibition for the level of heterology we used – 4 mismatches in a 37-nt region (~11% divergence) – was only about 4-fold, quite different from the greater than the 700-fold reduction seen between dsDNA inverted repeat substrates with similar levels of heterology, but where recombination is blocked only when both Rad51 and Rad59 are deleted (Chen and Jinks-Robertson, 1999; Spell and Jinks-Robertson, 2003). Moreover, we were surprised that the effect of these heterologies was not suppressed by deleting either

71 Msh2 or the helicase Sgs1, as previous studies have shown that recombination between divergent sequences – both gene conversions between dsDNA sequences and SSA – is markedly improved by deleting Sgs1 and components of mismatch repair (Nickloff et al., 1986; Roy et al.,

2018). For Sgs1, the 11% level of heterology is greater than that studied in SSA (3%) yet comparable to spontaneous inverted repeats (9%), but it is possible that Sgs1 cannot respond to such a level of divergence. However, we note that the ability of Sgs1 and Msh2/Msh6 to discourage recombination of heterologous sequences is much greater when the DSB end contains a nonhomologous tail versus ends that do not have such sequences (as in the cases studied here) (Spell and Jinks-Robertson, 2004). Alternatively, Sgs1 may play a specific role in

SSTR that has not yet been revealed.

Incorporation of mismatches templated by the ssODN into the genome occurs in an

Msh2-dependent manner. SSTR provides an unambiguous way to distinguish between the initial strand annealing event with the one DSB end that is complementary to the ssODN and the subsequent events that lead to second end capture and the completion of DSB repair.

Confronted with 4 mismatches in the ssODN on either side of the DSB, the cell efficiently incorporates these heterologies into the gene-edited product, but by two distinctly different processes. The initial annealing of the resected end with the template produces a heteroduplex

DNA that should be short-lived, until DNA polymerase extends the 3’ end and the newly copied strand dissociates and anneals with the second resected end. Only during this short-lived annealing step can mismatch repair transfer the heterology to the strand that will be incorporated into the final gene-edited product (Figure 2.8). We have previously shown that

72 similar events occur rapidly during an HO-induced gene conversion (MAT switching) and depend on mismatch repair machinery (Haber, 1993).

Although most heterologies in the ssODN are readily incorporated, a mismatch very close to the 3’ end of the ssODN is usually not incorporated, unless the 3’ to 5’ proofreading activity of DNA Polymerase  is eliminated. We demonstrated a similar type of proofreading in

BIR, where the 3’ strand invading into a duplex DNA donor is resected (Anand et al., 2017a). By the same token, 3’ to 5’ resection of the DSB end will assure that a heterology close to that end

(close to the XhoI site) could be incorporated without the need for Msh2.

Once the annealed end is extended, copying the 5’ side of the ssODN, all of the heterologies on that side will be copied; but when this newly-copied strand anneals with the second DSB end, there will be an obligate heteroduplex. Hence, on this side, in the absence of

Msh2, we recover sectored colonies. The failure to incorporate the 5’-most heterology may reflect dissociation of the newly copied strand before it reaches the end of the template.

These considerations lead us to the model of SSTR shown in Figure 2.8 and 2.10. After a

DSB, the cell initiates end resection, forming ssDNA tails. This process may require MRX proteins to create 3’-ended tails. In the absence of this complex it is possible that neither Exo1 nor Sgs1-Rmi1-Top3-Dna2 can act soon enough to permit use of the ssDNA template before its degradation; however, MRX is required even when ssDNA templates are generated by the retron system, which presumably has the capacity to create ssDNA continually while under induction. Strand annealing depends on Rad52 and on the action of Rad59 to thwart a Rad51- mediated inhibition of Rad52’s annealing activity. Pol is also engaged in its proofreading mode to remove heterologies near either the 3’ end of the DSB or the template. In SSTR, the degree

73 of resection of the ssODN is quite limited, as only a marker 2-nt from the end is affected by this

“chewing back”. When strand invasion occurs during BIR, the 3’ to 5’ excision can extend up to about 40 bases (Anand et al., 2017a). Once the first end is annealed and extended, the newly synthesized DNA must dissociate from the ssODN and anneal to the second end of the DSB, thus bridging both sides of the break and creating heteroduplex DNA that is subject to mismatch repair. The final filling-in of resected ssDNA regions appears to be carried out by the translesion polymerase Pol or other redundant polymerases. When SSTR is used to create a large deletion, long-range resection is required and the Rad1-Rad10 flap endonuclease becomes essential.

Finally, we have also determined that SSTR is a highly mutagenic event. Until now, the off target-effects of gene editing have primarily been thought of as a result of non-specific cutting of the endonuclease. Here we found that the regions adjacent to the site of SSTR have a significantly higher rate of mutation compared to the spontaneous rate. Whether this is due to filling-in of the gaps created by long-rage resection following the DSB, or damage accumulated in the ssDNA remains to be studied (Yang et al., 2008). However, it is clear that gene-edited products should be screened for these potential mutations.

Methods

Parental Strain. JKM179 (ho MATα hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2-3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO) was used as the parental strain in these experiments. This strain lacks the HML and HMR donor sequences that would allow repair of a DSB at MAT by gene conversion, and thus all repair occurs through NHEJ or via the provided ssODN or dsODN.

74 ORFs were deleted by replacing the target gene with a prototrophic or an antibiotic-resistance marker via the high-efficiency transformation procedure of S. cerevisiae with PCR fragments

(Rothstein, 1983). A list of all strains used is provided in Table 2.1. Point mutations in POL2 and

POL3 were made via Cas9 and an ssODN template. gRNAs were ligated into a BplI digested site in a backbone that contains a constitutively active Cas9 and either an HPH or LEU2-marker

(bRA89 and bRA90, respectively) (Anand et al., 2017b). Plasmids were verified by sequencing

(GENEWIZ) and transformed as previously described (Anand et al., 2017b). Plasmids are listed in

Table 2.2.

Retron Plasmid Construction. pZS165, a yeast centromeric plasmid marked with ura3 for the galactose inducible expression of the retron-guide chimeric RNA with a flanking HH-HDV ribozyme (Sharon et al., 2018) obtained from Addgene. gBlocks were designed that contained a gRNA and donor sequence and were cloned into the NotI-digested pZS165 backbone using the

NEBuilder HiFi DNA Assembly Cloning kit. Integration was verified by sequencing (GENEWIZ).

SSTR Viability Using HO endonuclease. Strains were grown overnight in selective media, and were then diluted into 50 mL of YEPD and grown for 3 hours. Cells were then pelleted, washed with dH2O, and resuspended in 0.1M LiAc. After pelleting, 25 휇L of 100 휇M ssODN, 25 uL of TE,

25 휇L of 2 mM salmon sperm DNA, 240 휇L of 50% PEG, and 36 휇L of 1M LiAc was added to the pellet and vortexed. Reactions were incubated at 30oC for 30 minutes, followed by a 20-min incubation at 42oC. Cells were then diluted 1000-fold and plated onto YEPD and YEP-Galactose

(YEP-Gal) media. YEPD plates were grown at 30oC for 2 days, and YEP-Gal plates were incubated at 30oC for 3 days. Colonies were then counted to obtain average viability. SSTR vs NHEJ events

75 were determined by pooling survivors and amplifying the MAT locus via PCR. Following amplification, PCR products were digested with the XhoI nuclease and quantified via gel electrophoresis.

SSTR Viability Using Cas9/Retron System. Cas9 and the Ec86 retron were integrated into trp1 on chromosome 15 into strain JKM179 that had ade3::GAL::HO deleted by restoring ADE3. After the plasmid containing the gRNA and Ec86 donor sequence was introduced, strains were resuspended in dH2O and plated onto uracil drop-out media or uracil drop-out media containing galactose (URA-Gal). URA plates were incubated at 30oC for 3 days, and URA-Gal plates were incubated at 30oC for 4 days. After incubation, plates were counted to obtain viability. URA-Gal plates were then replica plated to lysine drop-out media to obtain SSTR levels.

DNA Sequence Analysis. Using primers flanking the region of interest, PCR was used to amplify

DNA from surviving colonies. PCR products were purified and Sanger-sequenced by GENEWIZ.

The sequences were analyzed using Serial Cloner 2-6-1.

Statistical Analysis. All statistical analysis was performed using GraphPad Prism 8 software.

76 Table 2.1. Strains used in these experiments

Strain Genotype Notes

DG_24 JKM179; rad59::URA3

DG_25 JKM179; rad51::URA3

DG_26 JKM179; rdh54::KAN

DG_27 JKM179; sgs1::URA3

DG_29 JKM179; srs2::KAN

DG_30 JKM179; pJH627

DG_31 JKM179; rad50::KAN

DG_32 JKM179; mre11::KAN

DG_33 JKM179; rad52::KAN

DG_37 JKM179; rad54::NAT

DG_38 JKM179; exo1::NAT

DG_39 JKM179; sae2::NAT

DG_40 JKM179; fun30::KAN

DG_41 JKM179; mph1::NAT

DG_42 JKM179; rad55::NAT

DG_43 JKM179; pms2::KAN

DG_44 JKM179; rad1::NAT

DG_45 JKM179; pol2-4

DG_46 JKM179; pol3-01

77 DG_47 JKM179: rad51::URA3; rad59::KAN

DG_48 JKM179: rdh54::KAN; rad59::URA3

DG_49 JKM179: rdh54::KAN; rad51::URA3

DG_50 JKM179: rad51::URA3; srs2::NAT

DG_51 JKM179: rdh54::KAN; srs2::NAT

DG_52 JKM179: sgs1::URA3; exo1::NAT

DG_78 ho MATα hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2-3,112 lys5 trp1::Gal::spCas9::Gal::Ec86::trp1 ura3-52 DG_79 DG_78; rad51::HPH; pDG396

DG_80 DG_78; rad52::KAN; pDG396

DG_81 DG_78; rad55::NAT; pDG396

DG_82 DG_78; rad59::NAT; pDG396

DG_84 DG_78; rad50::NAT; pDG396

DG_85 DG_78; rdh54::NAT; pDG396

DG_86 DG_78; srs2::NAT; pDG396

DG_87 DG_78; fun30::NAT; pDG396

DG_88 DG_78; rad51::HPH; pDG397

DG_89 DG_78; rad52::KAN; pDG397

DG_90 DG_78; rad55::NAT; pDG397

DG_91 DG_78; rad59::NAT; pDG397

DG_92 DG_78; rad50::NAT; pDG397

DG_93 DG_78; rdh54::NAT; pDG397

78 DG_94 DG_78; srs2::NAT; pDG397

DG_95 DG_78; fun30::NAT; pDG397

DG_104 JKM179: URA3-MX URA3-MX integrated 200 bp upstream HO cut site DG_105 DG_104; rev1::NAT

DG_106 DG_104; rev3::NAT

DG_109 DG_78; rdh54::KAN; rad51::HPH; pDG396

DG_110 DG_78; rdh54::KAN; rad51::HPH; pDG397

DG_112 DG_78; rad59::NAT rad51::HPH; pDG396

DG_113 DG_78; rad59::NAT rad51::HPH; pDG397

DG_115 JKM179; mre11::KAN; ku70::NAT

DG_116 JKM179: URA3-MX URA3-MX integrated 200 bp downstream HO cut site DG_117 JKM179: URA3-MX URA3-MX integrated 500 bp upstream HO cut site DG_118 JKM179: URA3-MX URA3-MX integrated 1000 bp upstream HO cut site DG_119 JKM179: URA3-MX URA3-MX integrated 1500 bp upstream HO cut site DG_120 JKM179: URA3-MX URA3-MX integrated 2000 bp upstream HO cut site yMV45 ho hml::ADE1 MATa::hisG hmr::ADE1 leu2::leu2(Asp718- (Vaze et. al., 2002) SalI)-URA3-pBR332-MATa ade3::GAL::HO ade1 lys5 ura3- 52 trp1::hisG NP_706 yMV45; rad51::KANMX

NP_558 yMV45; rad59::NATMX

79 NP_707 yMV45; rad51::KANMX rad59::NATMX

NP_591 yMV45; rad52-R70A

NP_708 yMV45; rad51::KANMX rad52-R70A

NP_610 yMV45; rad59::NATMX rad52-R70A

NP_709 yMV45; rad51::KANMX rad59::NATMX rad52-R70A

NP_867 yMV45; rad52::HPHMX

Table 2.2. Plasmids used in these experiments

Plasmid Description pDG_344 pZS165 with gRNA to target MAT⍺ (CACGCGGACAAAATGCAGCA) with 80 nt retron donor sequence to repair Cas9 DSB (TCTGCTCGCTGAAGAATGGCACGCGGACAAAATGCActcgagGCACGGAATATGGG ACTACTTCGCGCAACAGTATAATA) pDG_396 pZS165 with gRNA to target lys5 (ATGAGTTTACGTTCGAGGCG) with no retron donor sequence pDG_397 pZS165 with gRNA to target lys5 (ATGAGTTTACGTTCGAGGCG) with 80 nt retron donor sequence to repair Cas9 DSB (TTCAAGAGGATATACTCGCGGATGAGTTTACGTTCGAGGCATTAATGAGAACTTTG CCATTGGCGTCTCAAGCCAGAATC) pRA_114 Cas9 vector used to make pol3-01 (guide TCCTTTGATATCGAGTGT GC) pRA_124 Cas9 vector used to make pol2-4 allele (guide TATCAAATGCCATTAC CACA)

PL_634 Cas9 vector used to make rad52-R70A allele (guide ACTCTCTTGGAGATATACTC)

80 References

Anand, R., Beach, A., Li, K., and Haber, J.E. (2017). Rad51-mediated double-strand break repair and mismatch correction of divergent substrates. Nature 544(7650), 377-380.

Anand, R., Memisoglu, G., and Haber, J.E. (2017). Cas9-mediated gene editing in Saccharomyces cerevisiae. Protoc Exch.

Anand, R.P., Tsaponina, O., Greenwell, P.W., et al. (2014). Chromosome rearrangements via template switching between diverged repeated sequences. Genes Dev 28(21), 2394-2406.

Bai, Y., and Symington, L.S. (1996). A Rad52 homolog is required for RAD51-independent mitotic recombination in Saccharomyces cerevisiae. Genes Dev 10(16), 2025–2037.

Bardwell, A.J., Bardwell, L., Tomkinson, A.E., Friedberg, E.C. (1994) Specific cleavage of model recombination and repair intermediates by the yeast Rad1-Rad10 DNA endonuclease. Science 265(5181), 2082-2085.

Boeke, J.D., Trueheart, J., Natsoulis, G., and Fink, G.R. (1987). 5-fluoroorotic acid as a selective agent in yeast molecular genetics. Methods Enzymol 154, 164-175.

Bothmer, A., Phadke, T., Barrera, L.A., Margulies, C.M., Lee, C.S., Buquicchio, F., Moss, S., Abdulkerim, H.S., Selleck, W., Jayaram, H., Myer, V.E., and Cotta-Ramusino, C. (2017). Characterization of the interplay between DNA repair and CRISPR/Cas9-induced DNA lesions at an endogenous locus. Nat Commun 8, 13905

Boulton, S.J. and Jackson, S.P. (1996). Saccharomyces cerevisiae Ku70 potentiates illegitimate DNA double-strand break repair and serves as a barrier to error-prone DNA repair pathways. EMBO J 15(18), 5093-5103.

Cassani, C., Gobbini, E., Vertemara, J., Wang, W., Marsella, A., Sung, P., Tisi, R., Zampella, G., and Longhese, M.P. (2018). Structurally distinct Mre11 domains mediate MRX functions in resection, end-tethering and DNA damage resistance. Nucleic Acids Res 46(6), 2990-3008.

Chen, W., and Jinks-Robertson, S. (1999). The role of the mismatch repair machinery in regulating mitotic and meiotic recombination between diverged sequences in yeast. Genetics 151(4), 1299-1313.

Chen, X., Cui, D., Papusha, A., Zhang, X., Chu, C.D., Tang, J., Chen, K., Pan, X., and Ira, G. (2012).

81 The Fun30 nucleosome remodeler promotes resection of DNA double-strand break ends. Nature 484(7417): 576-580.

Clever, B., Interthal, H., Schmuckli-Maurer, J., King, J., Sigrist, M., and Heyer, W.D. (1997). Recombinational repair in yeast: functional interactions between Rad51 and Rad54 proteins. EMBO J 16(9), 2535–2544.

Cortes-Ledesma, F., Malagon, F., and Aguilera, A. (2004). A novel yeast mutation, rad52-L89F, causes a specific defect in Rad51-independent recombination that correlates with a reduced ability of Rad520L89F to interact with Rad59. Genetics 168(1), 553-557.

Costelloe, T., Louge, R., Tomimatsu, N., Mukherjee, B., Martini, E., Khadaroo, B., Dubois. K., Wiegant, W.W., Thierry, A., Burma, S., et al. (2012). The yeast Fun30 and human SMARCAD1 chromatin remodelers promote DNA end resection. Nature 489(7417), 581-584.

Davis, A.P., and Symington, L.S. (2003). The Rad52-Rad59 complex interacts with Rad51 and replication protein A. DNA Repair 2(10), 1127-1134.

Davis, L., and Maizels, N. (2014). Homology-directed repair of DNA nicks via pathways distinct from canonical double-strand break repair. Proc Natl Acad Sci USA 111(10), 924-932.

Davis, L., and Maizels, N. (2016). Two distinct pathways support gene correction by single- stranded donors at DNA nicks. Cell Rep 17(7), 1872-1881.

DiCarlo, J.E., Norville, J.E., Mali, P., Rios, X., Aach, J., and Church, G.M. (2013). Genome engineering in Saccharomyces cerevisiae using CRISPR-Cas systems. Nucleic Acids Res 41(7), 4336-4343.

Eapen, V.V., Sugawara, N., Tsabar, M., Wu, W.H., and Haber, J.E. (2012). The Saccharomyces cerevisiae chromatin remodeler Fun30 regulates DNA end resection and checkpoint deactivation. Mol Cell Biol 32(22), 4727-4740.

Ferrari, M., Dibitetto, D., De Gregorio, G., Eapen, V.V., Rawal, C.C., Lazzro, F., Tsabar, M., Marini, F., Haber, J.E., and Pellicioli, A. (2015). Functional interplay between the 53BP1-ortholog Rad9 and the Mre11 complex regulates resection, end-tethering and repair of a double strand break. PLoS Genet 11(1)

Fishman-Lobell, J., and Haber, J.E. (1992). Removal of nonhomologous DNA ends in double- strand break recombination: the role of the yeast ultraviolet repair gene RAD1. Science 258(5081), 480–484.

Gallagher, D.N., and Haber, J.E. (2018). Repair of a site-specific DNA cleavage: old-school l lessons for Cas9-mediated gene editing. ACS Chem Biol 13(2), 397-405.

82 Gillet-Markowska, A., Louvel, G., and Fischer, G. (2015). bz-rates: a web-tool to estimate mutation rates from fluctuation analysis. G3 5(11), 2323-2327.

Gobbini, E., Cassani, C., Vertemara, J., Wang, W., Mambretti, F., Casari, E., Sung, P., Tisi, R., Zampella, G., and Longhese, M.P. (2018). The MRX complex regulates Exo1 resection activity by altering DNA end structure. EMBO J 37(16), e98588.

Gobbini, E., Cesena, D., Galbiati, A., Lockhart, A., and Longhese, M.P. (2013). Interplays between ATM/Tel1 and ATR/Mec1 in sensing and signaling DNA double-strand breaks. DNA Repair 12, 791-799.

Haber, J.E. (2016). A life investigating pathways that repair broken chromosomes. Annu Rev Genet 50, 1-28.

Haber, J.E., Ray, B.L., Kolb, J.M., White, C.I. (1993). Rapid kinetics of mismatch repair of heteroduplex DNA that is formed during recombination in yeast. Proc Natl Acad Sci USA 90(8), 3363-3367.

Harmsen, T., Klaasen, S., van de Vrugt, H., and Te Riele, H. (2018). DNA mismatch repair and oligonucleotide end-protection promote base-pair substitution distal from a CRISPR/Cas9- induced DNA break. Nucleic Acids Res 46(6), 2945-2955.

Hays, S.L., Firmenich, A.A., and Berg, P. (1995). Complex formation in yeast double-strand break repair: participation of Rad51, Rad52, Rad55, and Rad57 proteins. Proc Natl Acad Sci USA 92(15), 6925–6929.

Hicks, W.M., Kim, M., and Haber, J.E. (2010). Increased mutagenesis and unique mutation signature associated with mitotic gene conversion. Science 329(5987), 82-85.

Hicks, W.M., Yamaguchi, M., and Haber, J.E. (2011). Real-time analysis of double-strand DNA break repair by homologous recombination. Proc Natl Acad Sci USA 108(8), 3108–3115.

Holbeck, S.L., and Strathern, J.N. (1997). A role for Rev3 in mutagenesis during double-strand break repair in Saccharomyces cerevisiae. Genetics 147(3), 1017-1024.

Hsu, M.Y., Eagle, S.G., Inouye, M., and Inouye, S. (1992). Cell-free synthesis of the branched RNA-linked msDNA from retron-Ec67 of Escherichia coli. J Biol Chem 267(20), 13823–13829.

Hsu, M.Y., Inouye, M., and Inouye, S. (1990). Retron for the 67-base multicopy single-stranded DNA from Escherichia coli: a potential transposable element encoding both reverse transcriptase and Dam methylase functions. Proc Natl Acad Sci USA 87(23), 9454–9458.

Ira, G., and Haber, J.E. (2002). Characterization of RAD51-independent break-induced replication that acts preferentially with short homologous sequences. Mol Cell Biol 22(18),

83 6384-6392.

Ivanov, E.L., Korolev, V.G., and Fabre, F. (1992). XRS2, a DNA repair gene of Saccharomyces cerevisiae, is needed for meitotic recombination. Genetics 132(3), 651-664.

Jiang, H., Xie, Y., Houston, P., Stemke-Hale, K., Mortensen, U.H., Rothstein, R., and Kodadek, T. (1996). Direct association between the yeast Rad51 and Rad54 recombination proteins. J Biol Chem 271(52), 33181–33186.

Keskin, H., Shen, Y., Huang, F., Patel, M., Yang, T., Ashley, K., Mazin, A.V., and Storici, F. (2014). Transcript-RNA-templated DNA recombination and repair. Nature 515(7527), 436-439.

Knott, G.J., and Doudna, J.A. (2018). CRISPR-Cas guides the future of genetic engineering. Science 361(6405), 866-869.

Krejci, L., Van Komen, S., Li, Y., Villemain, J., Reddy, M.S., Klein, H., Ellenberger, T., and Sung, P. (2003). DNA helicase Srs2 disrupts the Rad51 presynaptic filament. Nature 423(6937), 305-309.

Lao, J.P., Oh, S.D., Shinohara, M., Shinohara, A., and Hunter, N. (2008). Rad52 promotes postinvasion steps of meiotic double-strand-break repair. Mol Cell 29(4), 517-524.

Le, S., Moore, J.K., Haber, J.E., and Greider, C.W. (1999). Rad50 and Rad51 define two pathways that collaborate to maintain telomeres in the absence of telomerase. Genetics 152(1), 143-152.

Lee, K., Ji, J.H., Yoon, K., Che, J., Seol, J.H., Lee, S.E., and Shim, E.Y. (2019). Microhomology selection for microhomology mediated end joining in Saccharomyces cerevisiae. Genes 10(4), E284.

Lydeard, J.R., Jain, S., Yamaguchi, M., and Haber, J.E. (2007). Break-induced replication and telomerase-independent telomere maintenance require Pol32. Nature 448(7155): 820-823.

Ma, J.L., Kim, E.M., and Haber, J.E. (2003). Yeast Mre11 and Rad1 proteins define a Ku- independent mechanism to repair double-strand breaks lacking overlapping end sequences. Mol Cell Biol 23, 8820-8828.

84 Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E., and Church, G.M. (2013). RNA-guided human genome engineering via Cas9. Science 339(6121): 823-826.

Malkova, A., Ross, L., Dawson, D., Hoekstra, M.F., and Haber, J.E. (1996). Meiotic recombination initiated by a double-strand break in rad50 delta yeast cells otherwise unable to initiate meiotic recombination. Genetics 143(2), 741–754.

Mallory, J.C., and Petes, T.D. (2000). Protein kinase activity of Tel1p and Mec1p, two Saccharomyces cerevisiae proteins related to the human ATM protein kinase. Proc Natl Acad Sci USA 97, 13749-13754.

Mantiero, D., Clerici, M., Lucchini, G., and Longhese M.P. (2007). Dual role for Saccharomyces cerevisiae Tel1 in the checkpoint in response to double-strand breaks. EMBO Rep 8, 380-387.

Mazin, A.V., Bornarth, C.J., Solinger, J.A., Heyer, W.D., and Kowalczykowski, S.C. (2000). Rad54 protein is targeted to pairing loci by the Rad51 nucleoprotein filament. Mol Cell 6(3), 583–592.

McDonald, J.P., and Rothstein, R. (1994). Unrepaired heteroduplex DNA in Saccharomyces cerevisiae is decreased in Rad1 Rad52-independent recombination. Genetics 137(2), 393-405.

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

Mimitou, E.P., and Symington, L.S. (2008)., Sae2, Exo1 and Sgs1 collaborate in DNA double- strand break processing. Nature 455(7214), 770-774.

Mirochnitchenko, O., Inouye, S., and Inouye, M. (1994). Production of single- stranded DNA in mammalian cells by means of a bacterial retron. J Biol Chem 269(4), 2380–2383.

Miyata, S., Ohshima, A., Inouye, S., Inouye, M. (1992). In vivo production of a stable single- stranded cDNA in Saccharomyces cerevisiae by means of a bacterial retron. Proc Natl Acad Sci USA 89(13), 5735–5739.

Moore, J.K., and Haber, J.E. (1996). Cell cycle and genetic requirements of two pathways of nonhomologous end-joining repair of double-strand breaks in Saccharomyces cerevisiae. Mol Cell Biol 16(5), 2164–2173.

Nickoloff, J.A., Chen, E.Y., and Heffron, F.A. (1986). 24-base-pair DNA sequence from the MAT locus stimulates intergenic recombination in yeast. Proc Natl Acad Sci USA 83(20), 7831-7835.

Ozenberger, B.A. and Roeder, G.S. (1991). A unique pathway of double-strand break repair operates in tandemly repeated genes. Mol Cell Biol 11(3), 1222-1231.

Paix, A., Folkmann, A., and Seydoux, G. (2017). Precision genome editing using CRISPR-Cas9 and

85 linear repair templates in C. elegans. Methods 15, 86-93.

Petukhova, G., Stratton, S., and Sung, P. (1998). Catalysis of homologous DNA pairing by yeast Rad51 and Rad54 proteins. Nature 393(6680), 91–94.

Piazza, A., Shah, S.S., Wright, W.D., Gore, S.K., Koszul, R., and Heyer, W.D. (2019). Dynamic Processing of Displacement Loops during Recombinational DNA Repair. Mol Cell 73(6), 1255- 1266.e4.

Rattray, A.J., Shafer, B.K., McGill, C.B., and Strathern, J.N. (2002). The roles of REV3 and RAD57 in double-strand-break-repair-induced mutagenesis of Saccharomyces cerevisiae. Genetics 162(3), 1063-1077.

Richardson, C.D., Kazane, K.R., Feng, S.J., Zelin, E., Bray, N.L., Schafer, A.J., Floor, S.N., and Corn, J.E. (2018). CRISPR-Cas9 genome editing in human cells occurs via the Fanconi anemia pathway. Nat Genet 50(8), 1132-1139.

Rothstein, R.J. (1983) One-step gene disruption in yeast. Methods Enzymol 101, 202-211. Roy, K., Smith, J., Vonesch, S. et al. (2018). Multiplexed precision genome editing with trackable genomic barcodes in yeast. Nat Biotechnol 36, 512-520.

Sasanuma, H., Furihata, Y., Shinohara, M., and Shinohara, A. (2013). Remodeling of the Rad51 DNA strand-exchange protein by the Srs2 helicase. Genetics 194(4), 859-872.

Sharon, E., Chen, S.A., Khosla, N.M., Smith, J.D., Pritchard, J.K., and Fraser, H.B. (2018) Functional genetic variants revealed by massively parallel precise genome editing. Cell 175(2), 544-557.

Shen, B., Zhang, J., Wu, H., Wang, J., Ma, K., Li, Z., Zhang, X., Zhang, P., and Huang, X. (2013). Generation of gene-modified mice via Cas9/RNA-mediated gene targeting. Cell Res 23(5): 720- 723.

Shi, I., Hallwyl, S.C., Seong, C., Mortensen, U., Rothstein, R., and Sung, P. (2009) Role of the Rad52 amino-terminal DNA binding activity in DNA strand capture in homologous recombination. J Biol Chem 284(48), 33275-33284.

Shibata, M., Nishimasu, H., Kodera, N., Hirano, S., Ando, T., Uchihashi, T., Nureki, O. (2017). Real-space and real-time dynamics of CRISPR-Cas9 visualized by high-speed atomic force microscopy. Nat Commun 8(1), 1430.

Shimamoto, T., Hsu, M.Y., Inouye, S., and Inouye, M. (1993). Reverse transcriptases from bacterial retrons require specific secondary structures at the 5’-end of the template for the cDNA priming reaction. J Biol Chem 268(4), 2684–2692.

86 Signon, L., Malkova, A., Naylor, M.L., Klein, H., and Haber, J.E. (2001). Genetic requirements for RAD51- and RAD54-independent break-induced replication repair of a chromosomal double- strand break. Mol Cell Biol 21(6), 2048-2056.

Sinha, S., Li, F., Villarreal, D., Shim, J.H., Yoon, S., Myung, K., Shim, E.Y., and Lee, S.E. (2017). Microhomology-mediated end joining induces hypermutagenesis at breakpoint junctions. PLOS Genet 13(4), e1006714.

Spell, R.M., and Jinks-Robertson, S. (2003). Role of mismatch repair in the fidelity of RAD51- and RAD59-dependent recombination in Saccharomyces cerevisiae. Genetics 165(4), 1733-1744.

Spell, R.M., and Jinks-Robertson, S. (2004). Examination of the roles of Sgs1 and Srs2 helicase in the enforcement of recombination fidelity in Saccharomyces cerevisiae. Genetics 168(4), 1855- 1865.

Storici, F., Snipe, J.R., Chan, G.K., Gordenin, D.A., and Resnick, M.A. (2006). Conservative repair of a chromosomal single-strand DNA through two steps of annealing. Mol Cell Biol 26(20), 7654-7657.

Sugawara, N., Goldfarb, T., Studamire, B., Alani, E., and Haber, J.E. (2004). Heteroduplex rejection during single-strand annealing requires Sgs1 helicase and mismatch repair proteins Msh2 and Msh6 but not Pms1. Proc Natl Acad Sci USA 101(25), 9315-9320.

Sung, P. (1997). Yeast Rad55 and Rad57 proteins form a heterodimer that functions with replication protein A to promote DNA strand exchange by Rad51 recombinase. Genes Dev 11(9), 1111–1121.

Symington, L.S. (2016). Mechanism and regulation of DNA end resection in eukaryotes. Crit Rev Biochem Mol Biol 51(3), 195-212.

Teng, S.C., Chang, J., McCowan, B., and Zakian, V.A. (2000). Telomerase-independent lengthening of yeast telomeres occurs by an abrupt Rad50p-dependent Rif-inhibited recombinational process. Mol Cell 6(4), 947-952.

Tsaponina, O., and Haber, J.E. (2014). Frequent Interchromosomal Template Switches during Gene Conversion in S. cerevisiae. Mol Cell 55(4), 615-625.

Vaze, M.B., Pellicioli, A., Lee, S.E., Ira, G., Liberi, G., Arbel-Eden, A., Foiani, M., and Haber, J.E. (2002). Recovery from checkpoint-mediated arrest after repair of a double-strand break requires Srs2 helicase. Mol Cell 10(2), 373-385.

87 Veaute, X., Jeusset, J., Soustelle, C., Kowalczykowski, S.C., Le Cam, E., and Fabre, F. (2003). The Srs2 helicase prevents recombination by disrupting Rad51 nucleoprotein filaments. Nature 423(6937), 309-312.

Waters, C.A., Strande, N.T., Pryor, J.M., Strom, C.N., Mieczkowski, P., Burkhalter, M.D., Oh, S., Qaqish, B.F., Moore, D.T., Hendrickson, E.A., and Ramsden, D.A. (2014). The fidelity of the ligation step determines how ends are resolved during non-homologous end joining. Nat Commun 5, 4286.

Wu, Y., Kantake, N., Sugiyama, T., Kowalczykowski, S.C. (2008). Rad51 protein controls Rad52- mediated DNA annealing. J Biol Chem 283(21), 14883-14892.

Zhu, Z., Chung, W.H., Shim, E.Y., Lee, S.E., and Ira, G. (2008). Sgs1 helicase and two nucleases Dna2 and Exo1 resect DNA double-strand break ends. Cell 134(6), 981-994.

88 Chapter 3 – A genome-wide screen to characterize a novel Rad51- independent pathway utilized in genome engineering

Abstract

The DNA double-strand break (DSB) is one of the most lethal forms of damage that can occur to the genome. The cell has evolved two main repair pathways in order to cope with DNA

DSBs: homologous recombination (HR), which repairs the break using a homologous segment of

DNA as a repair template, and nonhomologous end joining (NHEJ), which simply ligates the ends of the break back together. HR is a less mutagenic form of DNA repair that is usually dependent on the extensively characterized protein Rad51, which has the capability to recognize homologous regions of DNA. However, we have recently shown that genome editing using the CRISPR/Cas9 system with a single stranded DNA (ssDNA) template utilizes a noncanonical DNA repair pathway termed single-strand template repair (SSTR) that is independent of, and even suppressed by, Rad51. Although this pathway requires some canonical repair proteins, such as the Rad52, Srs2 and the MRX complex, this pathway is largely uncharacterized. Here we have used transposon-mediated mutagenesis to create a library of mutants with a two-component CRISPR/Cas9-retron system to screen for components of SSTR.

This drop-out screen utilizes a Cas9 targeting a nonfunctional lys5 gene, resulting in a nonfunctional lysine synthesis pathway, with an 80-nt ssDNA donor generated in vivo by an optimized E. coli retron system to correct lys5, resulting in prototrophic events. This screen identified predominantly components of the nucleotide excision repair (NER) pathway and components of telomeric recombination as potential factors of SSTR.

89 Summary

This chapter contains unpublished work that was done in collaboration with the

Schacherer lab at the University of Strasbourg. Conceptualization of this project was done by

Danielle N. Gallagher, James E. Haber, and Joseph Schacherer. Illumina Hi-Seq sequencing and analysis was done by Albert Yu, a graduate student of Michael Roshbash’s lab. Experiments were performed and data analysis was done by Danielle N. Gallagher.

This project was partially funded by the National Science Foundation under the

Graduate Research Opportunities Worldwide (GROW) program and the French Ministry of

Higher Education, Research and Innovation (MESRI), awarded to Danielle N. Gallagher, NSF

GRFP Fellow 2017240285.

Introduction

The accurate transfer of genetic information is a fundamental component of life, allowing the genetic code to be transmitted from one generation to the next. For billions of years, DNA-based life has existed on Earth, all the while the genetic code has been assaulted by both endogenous and exogenous sources of damage, including reactive oxygen species, ionizing radiation, and the implicit stress of DNA replication. The DNA double-strand break (DSB) is one of the most lethal forms of DNA damage that can occur within the cell. Repair of these deleterious lesions is vital to maintain genomic integrity, and improper DNA repair can result in chromosomal deletions, duplications, and non-reciprocal translocations – all of which are hallmarks of human cancers. Cells have evolved two primary pathways to repair DNA DSBs: 1) non-homologous end joining (NHEJ), a highly mutagenic form of repair usually resulting in small

90 base pair (bp) insertions or deletions (indels), and 2) homologous recombination (HR), where the DSB is repaired using homologous sequences as a template for repair (reviewed in Haber,

2016).

A vast amount of research has been done on the mechanisms of DNA DSB repair in the model organism of budding yeast, Saccharomyces cerevisiae. One potent tool in understanding

DSB repair is the endogenous site-specific endonuclease HO, which normally catalyzes the switching of the yeast mating type (Strathern et al., 1982). This occurs through an HR pathway – specifically gene conversion – that is entirely dependent on the protein Rad51, the eukaryotic homologue of the E. coli RecA (Mehta et al., 2017; Sung, 1994). Study of HO-induced DSBs has led to the definition of several distinct pathways of HR (gene conversion and break-induced replication) and NHEJ (both “classic” and microhomology-mediated end-joining) (reviewed in

Gallagher and Haber, 2018).

However, DSBs are not always undesirable, as they can be exploited for gene editing.

This has become particularly apparent since the discovery of the CRISPR/Cas9 system, which has completely revolutionized gene editing in eukaryotic organisms. Discovered in the bacteria

Streptococcus pyogenes as a form of adaptive immunity, the Cas9 system is a simple, yet ideal, programmable nuclease with the twist of relying on RNA:DNA rather than DNA:DNA recognition

(Jinek et al., 2012). Due to the simplicity and low cost of providing a short, single RNA sequence in a guide RNA (gRNA), CRISPR/Cas9 has been rapidly adapted to work in most model organisms and is actively being studied for use in pharmaceutical, biomedical, and ecological applications

(Cong et al., 2013; Mali et al., 2013; Jiang et al., 2015; Hammond et al., 2016; Kyrou et al.,

2018).

91 The CRISPR/Cas9 system offers a high degree of specificity – although off-target effects are still heavily debated – and extremely simple construction, as it utilizes a 20-nucleotide (nt) sequence adjacent to a 3-nt NGG sequence that acts as the gRNA. Coupled together with the

Cas9 protein, the gRNA directs the nuclease to the complementary sequence in the genome, where it creates a DSB. In haploid budding yeast, as in mammals, this break can be repaired by

NHEJ, usually resulting in a nonfunctional gene product. However, if the Cas9 is introduced along with a segment of template DNA, whether single or double stranded, the resident DNA sequences can be edited by HR to create insertions, deletions or base-pair substitutions, depending upon the template design. Despite widespread use of this system across multiple fields of study, the mechanism of how templated Cas9-mediated DSBs are repaired is not yet completely understood. This is fundamental to understanding research utilizing CRISPR/Cas9, especially considering its potential in the medical field.

As previously stated, Rad51 is responsible for homology recognition in most HR processes and is required for repair of chromosomal DSBs using a double-stranded DNA template located on a homologous chromosome or at an ectopic site. However, previous data studying organisms as diverse as yeast, nematodes and mammals, have discovered that Cas9- mediated repair with a single-stranded DNA (ssDNA) template (hereafter referred to as Single

Stranded Template Repair, or SSTR) is a Rad51-independent mechanism (Davis et al., 2014;

Gallagher et al., 2020). In the previous chapter, we showed that SSTR is dependent on few canonical DSB repair proteins: 1) Rad52, which is required for its strand-annealing function, 2)

Rad59, a Rad52 paralog in which we demonstrated a novel function in Rad51 interference, 3)

Srs2, a helicase and Rad51 antagonist, 4) Rad50 and Mre11, both of which are components of

92 the MRX complex and potentially play a role in repair pathway choice and short-range resection

(Lao et al., 2008; Sasanuma et al., 2013; Cassani et al., 2018; Gobbini et al., 2018; Gallagher et al., 2020).

Here we used transposon-mediated mutagenesis and a retron-Cas9 system to conduct a genome-wide screen of components of SSTR. Interestingly, major hits from the screen include components of the nucleotide excision repair (NER) pathway, as well as components of telomeric lengthening via recombination. These hits are currently being individually verified.

Results

Rather than undergoing a traditional yeast-screening approach using the Saccharomyces deletion collection, we decided to do a drop-out screen using our own collection of mutants generated using the Hermes transposon. The Hermes transposon (HTn) is a hAT family member

(known as hobo from Drosophila and Ac from maize) that has been introduced into the genomes of S. cerevisiae for functional genomics studies (Gangadharan et al., 2010;

Arensburger et al., 2011). Although the sites of integration of the Hermes transposon are not completely random, favoring 5’-nTnnnnAn-3’ sites in DNA regions with low nucleosome occupancy, the massive numbers of these sites allow for extensive quasi-random coverage across the yeast genome (Gangadharan et al., 2010; Guo et al., 2013; Hickman et al., 2018). In a single plasmid construct, a hygromycin-resistance marker was inserted inside of the Hermes transposon, while transposase was put under the control of a galactose-inducible promoter.

Following galactose induction in a saturated culture, we see approximately 16 different integration sites per gene across the genome (unpublished data, Schacherer lab). We can

93 observe these integration sites in both nonessential genes, as well as the C-terminus of some essential genes (extended data table 3.1 and 3.2). This screening strategy has multiple advantages over a traditional yeast screen using the deletion collection. There are inherent issues with the collection, simply due to the techniques available when it was made. First, the collection is incomplete, as a few hundred genes were missed in the first annotation of the

Saccharomyces cerevisiae genome that was used as the basis for construction of the deletion collection (Grunenfelder and Winzeler, 2002). Second, and more importantly, there are known issues with multiple strains in the collection where the transformation procedure had secondary mutagenic events. Thirdly, almost 8% of the strains have a compensatory aneuploidy, usually involving an ancient duplicated gene located elsewhere in the genome

(Hughes et al., 2000). Finally, analysis of experimental results using the deletion collection is more convoluted, due to multiple mutations in the barcodes that were added to annotate the different strains making up the collection (Eason et al., 2004; Smith et al., 2009). By using the genomic sequence adjacent to the site of integration as a “tag” and though deep sequencing of our libraries, we avoided the problems presented by the yeast deletion collection.

As a background for these libraries, we utilized a haploid strain, yDG168 that contains a

5-bp insertion in lys5, a frameshift mutation that causes a nonfunctional lysine synthesis pathway, as well as a galactose (gal) inducible apo-Cas9 integrated on chromosome 15. Next to apo-Cas9, we also integrated a gal-inducible yeast-optimized retron from E. coli, Ec86

(Gallagher et al., 2020). Retrons are natural DNA elements that encode for a reverse transcriptase (RT) that act on a specific consensus sequence to generate single-stranded DNA

(ssDNA) products (Hsu et al., 1990; Hsu et al., 1992; Shimamoto et al., 1993) (Figure 3.1).

94 However, the gRNA to direct Cas9-

cleavage and the retron ssDNA

donor sequence containing the

necessary 34-bp consensus sequence

were integrated onto a plasmid.

Figure 3.1 - Retron generated single stranded DNA. Therefore, even upon induction, After RNA polymerase transcribes the ssDonor region (shown in blue) and its consensus region (CR, shown in strains without the complementary red), the retron (Rt, shown in purple) binds to the consensus region of the mRNA transcript. After binding, plasmid could not cause DSBs or the retron reverse transcribes the mRNA to single- stranded DNA (ssDNA). produce pools of ssDNA in vivo. We also deleted Nej1 in this strain to prevent cells from undergoing NHEJ and confounding results.

To ensure optimal SSTR we tried three versions of the secondary plasmid where the length of the ssDNA to be generated increased (Figure 3.2). We observed no significant increase between an 80-nt, 100-nt, or 120-nt donor. Moving forward, we used the plasmid containing the 80-nt donor to keep consistency with previously published work (Gallagher et al., 2020).

To generate our mutant libraries, we transformed the plasmid containing the Gal- inducible Hermes transposon system into yDG168. Two colonies containing the plasmid were grown to saturation in parallel and the galactose was added to induce transposition. After 3 days in induction media, these cultures were pelleted and frozen to serve as reference libraries

(L1 and L2).

95 After creating our libraries, the

next step was to add the second-

component of the Cas9-retron system

to the cells, the plasmid containing the

Cas9 gRNA and the retron donor

sequence. It is important to note that

after activating transposase, the initial

Figure 3.2 - Increasing ssDNA donor does not increase plasmid is destroyed. In order to SSTR efficiency. Three different plasmids were tested for optimal SSTR. There is not significant difference in ensure full coverage of mutants, three viability between different length donors. However, Fun30 is only inhibitory with the shorter ssDNA donor. pellets of each library were We see very few prototrophic events without the retron donor (shown in blue). transformed with the URA-marked

plasmid containing a Cas9 gRNA targeting the 5-bp insertion of lys5 in our strain background

and an ssDNA donor template that would restore the lys5 reading frame using lithium acetate

and salmon sperm as a single stranded carrier (Gietz and Schiestl, 1991). Following

transformation, the cells were plated onto URA drop-out media to select for mutants that had

taken up the plasmid. Ura+ colonies from each transformation were combined, pelleted, and

frozen down to serve as our second reference libraries (L1_Retron and L2_Retron).

Following pooling Ura+ colonies, a pellet from each libraries was resuspended and grown

in URA drop-out media containing galactose to induce the Cas9-retron system. Now that both

components of the system were present in the cell, cleavage would be directed to the

nonfunctional lys5 locus and the frameshift mutation restored via SSTR using the retron-

generated ssDNA as a template for repair. After 24 hours of growth in induction media, cells

96 were plated onto lysine drop-out media. This process was repeated three times to ensure full coverage of the mutant library. Lys+ from all inductions were pooled together and frozen down as our final library (L1_Lys and L2_Lys). It should be noted that we saw very few prototrophic events in the absence of the ssDNA donor template (Figure 2.2).

Libraries from every step in the process were sequenced using Illumina Hi-seq. Genes that dropped out of pools L1_Lys and L2_Lys but were present in the L1_Retron and L2_Retron pools were considered potential components of SSTR. The top hits from this screen are listed in

Table 3.1. A full list of gene drop-outs from each library can be found in Extended Data Tables 1 and 2.

Discussion

Interestingly, many of the top hits are components of NER or the pathway for telomere lengthening via recombination. A previous screen done in mammalian systems showed that

CRISPR-Cas9 mediated gene editing occurred via the Fanconi anemia pathway, however, the only yeast homolog of this pathway is Mph1, which we have shown is not required for SSTR in

Saccharomyces (Richardson et al., 2018; Gallagher et al., 2020). Yet, the next biggest hits from the mammalian screen were components of the NER pathway (unpublished data from the Corn lab). These data indicate possible conservation of the mechanism of SSTR between yeast and mammalian systems.

Additionally, a Rad51-indpendent form of break-induced replication (BIR) is known to maintain telomere length in the absence of telomerase (known as Type II events) (Lundblad and

Blackburn, 1993; Le et al., 1999). These events are also known to require Rad52, Rad59, and the

97 Table 3.1 Gene Function Asf1 Nucleosome assembly factor; null has defects in telomere localization (Singer et al., 1998; Schwabish and Struhl, 2006) Csm4 Chromosome segregation in meiosis; meiotic telomere clustering (Conrad et al., 2008 Elp6 RecA-like ATPase; component of the elongator complex (Glatt et al., 2012) Fyv6 Uncharacterized protein implicated in regulation of DSB repair (NHEJ) (Wilson, 2002) Gmc2/ Complex required for crossover events in meiosis (both components were hits in Ecm11 screen); nuclear presence in mitotic cells (Comino et al., 2001; Zavec et al., 2004; Humphreyes et al., 2013) Hnt3 DNA 5’ adenylate hydrolase; potential NER protein; involved in DNA damage response (Daley et al., 2010; Tkach et al., 2012) Ies5 Subunit of Ino80; required for telomere maintenance via recombination (Hu et al., 2013) Irc13 Unknown function; null has increased Rad52 foci formation (Alvaro et al., 2007) Lrp1 Nuclear exosome-associated nucleic acid binding protein; Deletion affects maintenance of telomere length; Mammalian homolog involved in DNA repair/recombination (C1D) (Erdemir et al., 2002; Askree et al., 2004; Stead et al., 2007) Mgt1 DNA repair methyltransferase (Sassanfar and Samson, 1990) Mhf2 MHF histone-fold complex; Mammalian homolog interacts with DNA and FANCM (Mph1 ortholog) to repair damaged DNA (Yan et al., 2010) Nse4 Component of SMC-SMC6 complex; removal of X-shaped DNA structures during replication and repair (Hu, 2005; Bermudez-Lopez et al., 2010) Ntg1 DNA N-glycosylase and apurinic/apyrimidinic lyase; involved in BER (Alseth et al, 1999) Pcc1 Component of EKC/KEOPS protein complex; required for tRNA modification and telomeric TG1-3 recombination; Bud32 is another complex member and screen hit (Hu et al., 2013) Rad26 Involved in transcription-coupled NER (Lee et al., 2002) Rad6 Ubiquitin transferase; DNA damage checkpoint; meiotic DNA DSB formation; telomere maintenance via recombination; mitotic HR and NHEJ (reviewed in Game and Chernikova, 2009) Rim1 Mitochondrial ssDNA binding protein; previous screening has showed decreased meiotic recombination (Van Dyck et al., 1992; Su and Mitchell, 1993) Rsc58 Component of RSC chromatin remodeling complex; functions in establishing sister chromatid cohesion; involved in telomere maintenance (Sanders et al., 2002) Sld2 ssDNA origin-binding and annealing protein; required for initiation of DNA replication; synthetically lethal with Dpb11-1 (Kamimura et al., 1998; Kanter and Kaplan, 2011) Tfb2 Subunit of TFIIH and NER factor 3 complexes; required for NER (Prakash and Prakash, 2000) Tpp1 DNA 3’-phosphatase; Involved in NER and BER (Vance and Wilson, 2001)

98 MRX complex – all of which are also required for SSTR (Le et al., 1999; Teng et al., 2000; Chen et al., 2001). This raises the possibility that the cell sees the DSB and the ssDNA donor template as a telomere and is “extending” the resected ssDNA in favor of the ssDNA template provided.

This hypothesis is supported by previous studies that Type II telomere recombination events only require 30 bp of homology to initiate strand invasion, and previous studies have also shown that extending the homology of ssDNA donor templates decreases the efficiency of gene editing (Ira and Haber, 2002; Paix et al., 2017).

There were also a significant number of hits that map to genes involved in RNA processing. These hits could be the result of using the retron system to generate ssDNA in vivo.

It is also possible that SSTR is actually utilizing a pathway in place to repair DNA DSBs via their own RNA transcripts, as proposed by the Storici lab (Storici et al., 2006; Keskin et al., 2014).

These hits are currently being individually verified using the Cas9-retron system described, as well as by using the endogenous HO-endonuclease combined with an ssODN to insert the 6-bp XhoI restriction site into the MAT locus of chromosome 3 (Gallagher et al.,

2020).

Methods

Parental Strain. The strain used in this screen, yDG123 was made by transforming a yeast- optimized spCas9 and yeast-optimized Ec86 into the trp1 locus of JKM179 (ho MATα hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2-3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO) that had restored the ade3 locus, thus deleting Gal::HO, was used as the parental strain in this screen.

Nej1 was deleted with a kanamycin-resistance marker to prevent NHEJ. Thus, DSB repair can

99 only occur via provided template. gRNA/ssDonor plasmids were transformed via lithium acetate and salmon sperm high-efficiency transformation (Gietz and Schiestl, 1991).

NEBuilder HiFi DNA Assembly Cloning kit. Integration was verified by sequencing (GENEWIZ).

SSTR Viability Using Cas9/Retron System. After the plasmid containing the gRNA and Ec86 donor sequence was introduced to yDG123, strains were resuspended in dH2O and plated onto uracil drop-out media or uracil drop-out media containing galactose (URA-Gal). URA plates were incubated at 30oC for 3 days, and URA-Gal plates were incubated at 30oC for 4 days. After incubation, plates were counted to obtain viability. URA-Gal plates were then replica plated to lysine drop-out media to obtain SSTR levels.

Mutagenesis via Transposon Saturation. After ploidy was confirmed by staining and cytometry with the BD Accuri C6 Plus System, pDG401, a plasmid containing a galactose-inducible hygromycin-marked Hermes transposon, was transformed into yDG123. Two colonies containing the plasmid were grown in parallel in 30 mL of YEPD-Hygromycin (HPH) and grown at 30o for 24 hours, after which cultures were diluted to an OD of 0.05 in 50 mL of YEP-

Galactose with HPH. Cultures were grown for 72 hours to reach saturation. Cultures were pelleted, washed, and resuspended in 100 mL of YEPD and grown for 24 hours to lose any remaining plasmid. Cells were then resuspended in YEPD-HPH and grown for 24 hours to keep

100 only cells that had integrated the transposon. Cells were pelleted and frozen down as mutant libraries L1 and L2. After mutagenesis, high-efficiency lithium-acetate and salmon sperm based transformation was used to transform pDG397 into the library, and plated onto URA drop-out media to select for cells that had taken up the plasmid (Gietz and Schiestl, 1991). This was repeated for each library 3 times to ensure coverage. URA+ colonies were combined, resuspend, and frozen down (L1_Retron and L2_Retron). Cells were then resuspended and the

CRISPR-Cas9 and retron system induced in 100 mL URA-galactose drop-out media and grown for 24 hours. After induction and growth, cells were plated onto lysine drop-out media. This process was repeated 3 times to ensure coverage, and Lys+ cells were combined and frozen down (L1_Lys and L2_Lys).

DNA Library Preparation and Sequencing. Genomic DNA from each library (L1, L2, L1_Retron,

L2_Retron, L1_Lys, and L2_Lys) was prepped using Lucigen’s MasterPure Yeast DNA Purification

Kit and digested in separate reactions with DpnII and NlaIII. Following digestion, fragments were circularized and the transposon-genome junctions were amplified using Q5 PCR and outward-facing primers internal to the Hermes transposon. Amplicons were prepped for sequencing using the Nextera DNA Flex Library Prep kit with commercial Tn5 transposase. After adapter and index sequences were added, libraries were sequenced using the Illumina HiSeq

2500 system.

Statistical Analysis. All statistical analysis was performed using GraphPad Prism 8 software.

101 Table 3.2. Strains used in these experiments

Strain Description yDG123 ho MATα hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2-3,112 lys5 trp1::hisG′ ura3-52 trp1::spCas9::Ec86::trp1

Table 3.3. Plasmids used in these experiments

Plasmid Description pDG_396 pZS165 with gRNA to target lys5 (ATGAGTTTACGTTCGAGGCG) with no retron donor sequence pDG_397 pZS165 with gRNA to target lys5 (ATGAGTTTACGTTCGAGGCG) with 80-nt retron donor sequence to repair Cas9 DSB (TTCAAGAGGATATACTCGCGGATGAGTTTACGTTCGAGGCATTAATGAGAACTTTG CCATTGGCGTCTCAAGCCAGAATC) pDG_398 pZS165 with gRNA to target lys5 (ATGAGTTTACGTTCGAGGCG) with 100-nt retron donor sequence to repair Cas9 DSB (GTTGTTGAAATTCAAGAGGATATACTCGCGGATGAGTTTACGTTCGAGGCATTAATGA GAACTTTGCCATTGGCGTCTCAAGCCAGAATCCTCAATAAAA pDG_399 pZS165 with gRNA to target lys5 (ATGAGTTTACGTTCGAGGCG) with 120-nt retron donor sequence to repair Cas9 DSB (CTGTTTTCCTGTTGTTGAAATTCAAGAGGATATACTCGCGGATGAGTTTACGTTCGAG GCATTAATGAGAACTTTGCCATTGGCGTCTCAAGCCAGAATCCTCAATAAAAAGGAAA CCCG pDG_401 URA3; gal-inducible Hermes transposon; transposase

102 Table 3.4 – L1 gene drop outs, extended data Gene Systematic Name Gene Standard Name Gene Name RPR1 rpr1 RNase P Ribonucleoprotein RUF23 ruf23 RNA of Unknown Function snR10 snr10 Small Nucleolar RNA snR128 snr128 Small Nucleolar RNA snR17a snr17A Small Nucleolar RNA snR19 snr19 Small Nuclear RNA snR191 snr191 Small Nucleolar RNA snR34 snr34 Small Nucleolar RNA snR39B snr39B Small Nucleolar RNA snR41 snr41 Small Nucleolar RNA snR43 snr43 Small Nucleolar RNA snR61 snr61 Small Nucleolar RNA snR63 snr63 Small Nucleolar RNA snR65 snr65 Small Nucleolar RNA snR66 snr66 Small Nucleolar RNA snR79 snr79 Small Nucleolar RNA snR7-L snr7-L Small Nuclear RNA snR7-S snr7-S Small Nuclear RNA snR80 snr80 Small Nucleolar RNA snR81 snr81 Small Nucleolar RNA snR85 snr85 Small Nucleolar RNA snR87 snr87 Small Nucleolar RNA snR9 snr9 Small Nucleolar RNA SRG1 srg1 SER3 Regulatory Gene YAL068C pau8 seriPAUperin YBL003C hta2 Histone h Two A YBL008W-A "" "" YBL018C pop8 Processing Of Precursor RNAs YBL029C-A "" "" YBL044W "" "" YBL071W-A kti11 Kluveromyces lactis Toxin Insensitive YBL090W mrp21 Mitochondrial Ribosomal Protein YBR009C hhf1 Histone H Four YBR010W hht1 Histone H Three YBR016W "" "" YBR022W poa1 Phosphatase Of ADP-ribose 1"-phosphate YBR032W "" "" YBR040W fig1 Factor-Induced Gene YBR046C zta1 ZeTA-crystallin YBR047W fmp23 Found in Mitochondrial Proteome

103 YBR052C rfs1 Rad55 (Fifty-five) Suppressor YBR070C alg14 Asparagine Linked Glycosylation YBR071W "" "" YBR072C-A "" "" YBR077C slm4 Synthetic Lethal with Mss4 YBR085C-A "" "" YBR089C-A nhp6B Non-Histone Protein YBR090C "" "" YBR091C tim12 Translocase of the Inner Membrane YBR096W "" "" YBR111W-A sus1 Sl gene Upstream of ySa1 YBR120C cbp6 Cytochrome B Protein synthesis YBR139W atg42 AuTophaGy YBR149W ara1 D-ARAbinose dehydrogenase YBR156C sli15 Synthetically Lethal with Ipl1 YBR171W sec66 SECretory YBR188C ntc20 Prp19p (NineTeen)-associated Complex YBR196C-A "" "" YBR196C-B "" "" YBR210W erv15 ER Vesicle Protein YBR233W-A dad3 Duo1 And Dam1 interacting YBR242W "" "" YBR255C-A rcf3 Respiratory superComplex Factor Mitochondrial Ribosomal Protein, Large YBR268W mrpl37 subunit YBR302C cos2 COnserved Sequence YCL001W-A "" "" YCL001W-B "" "" YCL035C grx1 GlutaRedoXin YCL042W "" "" YCL044C mgr1 Mitochondrial Genome Required YCR002C cdc10 Cell Division Cycle Mitochondrial Ribosomal Protein, Large YCR003W mrpl32 subunit YCR020C pet18 PETite colonies YCR024C-B "" "" YCR025C "" "" YCR028C-A rim1 Replication In Mitochondria YCR035C rrp43 Ribosomal RNA Processing YCR044C per1 protein Processing in the ER Exit from rapamycin-induced GrOwth YCR075W-A ego2 arrest

104 YCR086W csm1 Chromosome Segregation in Meiosis YCR099C "" "" YCR101C "" "" YCR104W pau3 seriPAUperin family YCR108C "" "" YDL007C-A "" "" YDL018C erp3 Emp24p/Erv25p Related Protein YDL045C fad1 FAD synthetase YDL045W-A mrp10 Mitochondrial Ribosomal Protein YDL046W npc2 Niemann Pick type C homolog YDL047W sit4 Suppressor of Initiation of Transcription YDL051W lhp1 La-Homologous Protein YDL067C cox9 Cytochrome c OXidase Yeast Endoplasmic reticulum YDL072C yet3 Transmembrane protein YDL081C rpp1A Ribosomal Protein P1 Alpha YDL088C asm4 Anti-Suppressor in Multicopy YDL092W srp14 Signal Recognition Particle YDL105W nse4 Non-SMC Element YDL120W yfh1 Yeast Frataxin Homolog YDL123W sna4 Sensitivity to NA+ YDL130W rpp1B Ribosomal Protein P1 Beta YDL135C rdi1 Rho GDP Dissociation Inhibitor Suppressor of Chromosome YDL139C scm3 Missegregation YDL157C "" "" YDL160C-A mhf2 Mph1-associated Histone-Fold protein YDL173W par32 Phosphorylated After Rapamycin YDL189W rbs1 RNA-Binding Suppressor of PAS kinase O-6-MethylGuanine-DNA YDL200C mgt1 methylTransferase YDL201W trm8 Transfer RNA Methyltransferase YDL212W shr3 Super high Histidine Resistant YDL219W dtd1 D-Tyr-tRNA(Tyr) Deacylase YDL232W ost4 OligoSaccharylTransferase YDL235C ypd1 tYrosine Phosphatase Dependent YDL241W "" "" YDL245C hxt15 HeXose Transporter YDR002W yrb1 Yeast Ran Binder YDR003W rcr2 Resistance to Congo Red YDR003W-A "" "" YDR004W rad57 RADiation sensitive

105 YDR016C dad1 Duo1 And Dam1 interacting YDR022C atg31 AuTophaGy related YDR029W "" "" Mitochondrial Intermembrane space YDR031W mix14 CX(n)C motif protein YDR034W-B "" "" YDR042C "" "" Negative Regulator of Glucose-repressed YDR043C nrg1 genes YDR055W pst1 Protoplasts-SecreTed YDR059C ubc5 UBiquitin-Conjugating YDR063W aim7 Altered Inheritance rate of Mitochondria YDR068W dos2 "" YDR070C fmp16 Found in Mitochondrial Proteome YDR073W snf11 Sucrose NonFermenting YDR079C-A tfb5 "" YDR084C tvp23 Tlg2-Vesicle Protein YDR090C ilt1 Ionic Liquid Tolerance YDR102C "" "" YDR119W-A cox26 "" YDR139C rub1 Related to UBiquitin YDR157W "" "" YDR168W cdc37 Cell Division Cycle YDR169C-A "" "" YDR179W-A nvj3 Nucleus-Vacuole Junction YDR181C sas4 Something About Silencing YDR194W-A "" "" YDR197W cbs2 Cytochrome B Synthesis YDR209C "" "" YDR210W "" "" YDR225W hta1 Histone h Two A Protein 1 of Cleavage and polyadenylation YDR228C pcf11 Factor I YDR233C rtn1 ReTiculoN-like YDR246W trs23 TRapp Subunit YDR248C "" "" YDR272W glo2 GLyOxalase Mitochondrial Glutaredoxin-like Protein of YDR286C mgp12 12 kDa YDR315C ipk1 Inositol Polyphosphate Kinase YDR344C "" "" YDR366C mor1 Mitochondrial mORphology affecting

106 YDR373W frq1 FReQuenin homolog YDR374C pho92 PHOsphate metabolism YDR391C "" "" YDR412W rrp17 Ribosomal RNA Processing YDR438W thi74 THIamine regulon YDR446W ecm11 ExtraCellular Mutant YDR454C guk1 GUanylate Kinase Mitochondrial Ribosomal Protein, Large YDR462W mrpl28 subunit YDR529C qcr7 ubiQuinol-cytochrome C oxidoReductase YDR532C kre28 "" Translocase of the Inner Mitochondrial YEL020W-A tim9 membrane YEL033W mtc7 Maintenance of Telomere Capping YEL048C tca17 TRAPP Complex Associated protein YEL051W vma8 Vacuolar Membrane Atpase YEL057C sdd1 Suppressor of Degenerative Death YEL059C-A som1 SOrting Mitochondrial YER011W tir1 TIp1-Related YER026C cho1 CHOline requiring YER030W chz1 Chaperone for Htz1/H2A-H2B dimer YER034W "" "" YER049W tpa1 Termination and PolyAdenylation YER074W rps24A Ribosomal Protein of the Small subunit YER085C "" "" YER092W ies5 Ino Eighty Subunit YER101C ast2 ATPase STabilizing YER126C nsa2 Nop Seven Associated YER127W lcp5 Lethal with Conditional Pap1 YER145C-A "" "" YER156C myg1 "" YER165W pab1 Poly(A) Binding protein YER177W bmh1 Brain Modulosignalin Homolog WW domain containing protein interacting YFL010C wwm1 with Metacaspase YFL015C "" "" YFR001W loc1 LOCalization of mRNA YFR003C ypi1 Yeast Phosphatase Inhibitor YFR032C-B min10 mitochondrial MINi protein of 10 kDa YFR033C qcr6 ubiQuinol-cytochrome C oxidoReductase YFR054C "" "" YGL007W brp1 ""

107 YGL030W rpl30 Ribosomal Protein of the Large subunit YGL032C aga2 a-AGglutinin YGL039W "" "" YGL058W rad6 RADiation sensitive YGL061C duo1 Death Upon Overproduction YGL068W mnp1 Mitochondrial-Nucleoid Protein YGL070C rpb9 RNA Polymerase B YGL079W kxd1 KxDL homolog YGL080W mpc1 Mitochondrial Pyruvate Carrier YGL081W "" "" YGL096W tos8 Target Of Sbf YGL101W ygk1 "" YGL175C sae2 Sporulation in the Absence of spo Eleven YGL187C cox4 Cytochrome c OXidase YGL191W cox13 Cytochrome c OXidase YGL194C-A "" "" YGL198W yip4 Ypt-Interacting Protein YGL222C edc1 Enhancer of mRNA DeCapping YGL224C sdt1 Suppressor of Disruption of TFIIS YGL226C-A ost5 OligoSaccharylTransferase YGL226W mtc3 Maintenance of Telomere Capping YGL248W pde1 PhosphoDiEsterase YGL258W vel1 VELum formation YGL259W yps5 YaPSin YGR001C efm5 Elongation Factor Methyltransferase YGR018C "" "" YGR028W msp1 Mitochondrial Sorting of Proteins YGR035W-A "" "" YGR038W orm1 "" YGR039W "" "" YGR055W mup1 Methionine UPtake YGR092W dbf2 DumbBell Former YGR102C gtf1 Glutaminyl Transamidase subunit F YGR106C voa1 V0 Assembly protein YGR149W gpc1 GlyceroPhosphoCholine acyltransferase YGR159C nsr1 "" YGR161C rts3 "" YGR161W-C "" "" YGR168C pex35 PEroXin YGR169C-A lso2 Late-annotated Small Open reading frame YGR174C cbp4 Cytochrome B mRNA Processing YGR183C qcr9 ubiQuinol-cytochrome C oxidoReductase

108 YGR192C tdh3 Triose-phosphate DeHydrogenase YGR207C cir1 Changed Intracellular Redox state YGR225W ama1 Activator of Meiotic APC/C YGR230W bns1 Bypasses Need for Spo12p YGR263C say1 Steryl Acetyl hYdrolase YGR271C-A efg1 Exit From G1 YGR273C "" "" YGR280C pxr1 PinX1-Related gene YGR282C bgl2 Beta-GLucanase YGR285C zuo1 ZUOtin YHL018W mco14 Mitochondrial Class One protein of 14 kDa YHL031C gos1 GOlgi Snare YHL037C "" "" YHL042W "" "" YHR001W osh7 OxySterol binding protein Homolog YHR014W spo13 SPOrulation YHR039C-A vma10 "" YHR053C cup1-1 Cu, copper, CUPrum YHR057C cpr2 Cyclosporin A-sensitive Proline Rotamase YHR081W lrp1 Like RrP6 YHR086W-A "" "" YHR089C gar1 Glycine Arginine Rich YHR130C "" "" YHR133C nsg1 "" YHR152W spo12 SPOrulation YHR175W ctr2 Copper TRansport YHR175W-A "" "" YHR180W "" "" YHR185C pfs1 Prospore Formation at Spindles YHR213W-A "" "" YIL010W dot5 Disruptor Of Telomeric silencing YIL012W "" "" YIL015W bar1 BARrier to the alpha factor response APical growth revealed by Quantitative YIL040W apq12 morphological analysis YIL051C mmf1 Mitochondrial Matrix Factor YIL053W gpp1 Glycerol-3-Phosphate Phosphatase Arginine methyltransferase-Interacting YIL079C air1 RING finger protein YIL086C "" "" YIL087C aim19 Altered Inheritance rate of Mitochondria YIL097W fyv10 Function required for Yeast Viability

109 YIL102C "" "" YIL102C-A "" "" YIL117C prm5 Pheromone-Regulated Membrane protein YIL127C rrt14 Regulator of rDNA Transcription YIL133C rpl16A Ribosomal Protein of the Large subunit YIL138C tpm2 TroPoMyosin YIL139C rev7 REVersionless YIL156W-B mco8 Mitochondrial Class One protein of 8 kDa YIR008C pri1 DNA PRImase YIR021W-A "" "" YIR042C "" "" YJL011C rpc17 RNA Polymerase C YJL028W "" "" YJL030W mad2 Mitotic Arrest-Deficient YJL043W "" "" YJL077W-B "" "" YJL088W arg3 ARGinine requiring YJL089W sip4 SNF1-Interacting Protein Presequence translocase-Associated YJL104W pam16 Motor YJL115W asf1 Anti-Silencing Function YJL124C lsm1 Like SM YJL127C-B mco6 Mitochondrial Class One protein of 6 kDa YJL145W sfh5 Sec Fourteen Homolog YJL160C pir5 Protein with Internal Repeats YJL177W rpl17B Ribosomal Protein of the Large subunit YJL178C atg27 AuTophaGy related YJL179W pfd1 PreFolDin YJL185C atg36 AuTophaGy related YJL203W prp21 Pre-mRNA Processing YJL222W vth2 Vps Ten Homolog YJR007W sui2 SUppressor of Initiator codon YJR044C vps55 Vacuolar Protein Sorting YJR045C ssc1 Stress-Seventy subfamily C YJR063W rpa12 RNA Polymerase A YJR069C ham1 6-n-HydroxylAMinopurine sensitive YJR085C tmh11 TMem14 Homolog of 11 kDa YJR086W ste18 STErile YJR102C vps25 Vacuolar Protein Sorting YJR112W-A "" "" YJR146W "" "" YJR151W-A "" ""

110 YJR155W aad10 Aryl-Alcohol Dehydrogenase YKL002W did4 Doa4-Independent Degradation YKL003C mrp17 Mitochondrial Ribosomal Protein YKL023C-A min9 mitochondrial MINi protein of 9 kDa YKL024C ura6 URAcil requiring YKL042W spc42 Spindle Pole Component YKL044W mmo1 Mini Mitochondria ORF YKL049C cse4 Chromosome SEgregation YKL055C oar1 3-Oxoacyl-[Acyl-carrier-protein] Reductase YKL063C "" "" YKL077W psg1 Pma1 Stabilization in the Golgi YKL082C rrp14 Ribosomal RNA Processing YKL085W mdh1 Malate DeHydrogenase YKL096W cwp1 Cell Wall Protein YKL096W-A cwp2 Cell Wall Protein YKL102C "" "" YKL108W sld2 Synthetically Lethal with Dpb11-1 YKL119C vph2 Vacuolar pH YKL122C srp21 Signal Recognition Particle Mitochondrial Ribosomal Protein, Large YKL138C mrpl31 subunit YKL138C-A hsk3 Helper of ASK1 YKL159C rcn1 Regulator of CalciNeurin YKL160W elf1 ELongation Factor Mitochondrial Ribosomal Protein, Large YKL170W mrpl38 subunit YKL183C-A "" "" YKL196C ykt6 "" YKR002W pap1 Poly(A) Polymerase YKR015C "" "" YKR035W-A did2 Doa4-Independent Degradation YKR046C pln1 PeriLipiN YKR048C nap1 Nucleosome Assembly Protein YKR049C fmp46 Found in Mitochondrial Proteome YKR051W hfl1 Has Fused Lysosomes YKR052C mrs4 Mitochondrial RNA Splicing YKR073C "" "" Polarized growth Chromatin-associated YKR095W-A pcc1 Controller YLL009C cox17 Cytochrome c OXidase YLL018C-A cox19 Cytochrome c OXidase YLL027W isa1 Iron Sulfur Assembly

111 YLL038C ent4 Epsin N-Terminal homology YLL046C rnp1 RiboNucleoProtein YLL052C aqy2 AQuaporin from Yeast YLL053C "" "" Presequence translocase-Associated YLR008C pam18 Motor YLR009W rlp24 Ribosomal-Like Protein YLR010C ten1 TElomeric pathways with STn1 YLR011W lot6 LOw Temperature-responsive YLR012C "" "" YLR013W gat3 "" YLR017W meu1 Multicopy Enhancer of UAS2 YLR021W irc25 Increased Recombination Centers YLR026C sed5 Suppressor of Erd2 Deletion YLR031W "" "" YLR033W rsc58 Remodel the Structure of Chromatin YLR036C "" "" Negative regulator of the Filamentous YLR042C nfg1 Growth MAPK pathway YLR043C trx1 ThioRedoXin YLR064W per33 Pore and ER protein, 33 kDa YLR068W fyv7 Function required for Yeast Viability YLR074C bud20 BUD site selection YLR099W-A mim2 "" YLR104W lcl2 Long Chronological Lifespan 2 YLR109W ahp1 Alkyl HydroPeroxide reductase YLR111W "" "" YLR132C usb1 U Six Biogenesis YLR145W rmp1 RNase MRP Protein YLR146C spe4 SPErmidine auxotroph YLR155C asp3-1 ASParaginase YLR157W-D "" "" YLR167W rps31 Ribosomal Protein of the Small subunit YLR211C atg38 AuTophaGy related YLR215C cdc123 Cell Division Cycle YLR243W gpn3 Gly-Pro-Asn (N) motif YLR250W ssp120 Saccharomyces Secretory Protein YLR251W sym1 Stress-inducible Yeast Mpv17 YLR254C ndl1 NuDeL homolog YLR261C vps63 Vacuolar Protein Sorting YLR262C-A tma7 Translation Machinery Associated YLR268W sec22 SECretory

112 YLR287C "" "" YLR298C yhc1 Yeast Homolog of human U1C YLR323C cwc24 Complexed With Cef1p YLR333C rps25B Ribosomal Protein of the Small subunit YLR344W rpl26A Ribosomal Protein of the Large subunit YLR346C cis1 CItrinin Sensitive knockout YLR363C nmd4 Nonsense-Mediated mRNA Decay YLR365W "" "" YLR390W-A ccw14 Covalently linked Cell Wall protein YLR406C rpl31B Ribosomal Protein of the Large subunit YLR445W gmc2 Grand Meiotic recombination Cluster YML001W ypt7 Yeast Protein Two YML007C-A min4 mitochondrial MINi protein of 4 kDa YML018C "" "" YML022W apt1 Adenine PhosphoribosylTransferase YML026C rps18B Ribosomal Protein of the Small subunit YML047C prm6 Pheromone-Regulated Membrane protein YML048W gsf2 Glucose Signaling Factor YML074C fpr3 Fk 506-sensitive Proline Rotamase YML079W "" "" YML129C cox14 Cytochrome c OXidase YMR041C ara2 ARAbinose YMR052W far3 Factor ARrest YMR069W nat4 N-AcetylTransferase YMR070W mot3 Modifier of Transcription YMR074C sdd2 Suppressor of Degenerative Death YMR081C isf1 Increasing Suppression Factor YMR082C "" "" YMR084W "" "" YMR103C "" "" YMR105W-A "" "" YMR107W spg4 Stationary Phase Gene YMR114C "" "" YMR122C "" "" YMR138W cin4 Chromosome INstability YMR156C tpp1 Three Prime Phosphatase YMR161W hlj1 HomoLogous to E. coli dnaJ protein YMR175W-A "" "" YMR181C "" "" YMR183C sso2 Supressor of Sec One Alpha1-proteinase inhibitor-Degradation YMR184W add37 Deficient

113 YMR195W icy1 Interacting with the CYtoskeleton YMR230W-A "" "" YMR242W-A "" "" YMR247W-A "" "" YMR252C mlo1 Mitochondrially LOcalized protein YMR255W gfd1 Good For Dbp5p Mitochondrial Ribosomal Protein, Large YMR286W mrpl33 subunit YMR295C "" "" YMR312W elp6 ELongator Protein YMR320W "" "" YMR321C "" "" YMR325W pau19 seriPAUperin YNL030W hhf2 Histone H Four YNL031C hht2 Histone H Three YNL044W yip3 Ypt-Interacting Protein YNL050C "" "" YNL055C por1 PORin YNL079C tpm1 TroPoMyosin MItochondrial contact site and Cristae YNL100W mic27 organizing system YNL111C cyb5 CYtochrome B YNL119W ncs2 Needs Cla4 to Survive YNL122C mrp35 Mitochondrial Ribosomal Protein YNL130C cpt1 CholinePhosphoTransferase YNL130C-A dgr1 2-Deoxy-Glucose Resistant 1 YNL133C fyv6 Function required for Yeast Viability YNL135C fpr1 Fk 506-sensitive Proline Rotamase YNL148C alf1 ALpha tubulin Folding YNL153C gim3 Gene Involved in Microtubule biogenesis YNL158W pga1 Processing of Gas1p and ALP YNL159C asi2 Amino acid Sensor-Independent YNL160W ygp1 Yeast GlycoProtein YNL184C "" "" YNL200C nnr1 Nicotinamide Nucleotide Repair YNL232W csl4 Cep1 Synthetic Lethal YNL234W "" "" YNL245C cwc25 Complexed With Cef1p YNL255C gis2 GIg Suppressor YNL274C gor1 GlyOxylate Reductase YNL277W-A "" "" YNL281W hch1 High-Copy Hsp90 suppressor

114 YNL320W "" "" YNR004W swm2 Synthetic With mud2-delta YNR010W cse2 Chromosome SEgregation YNR024W mpp6 M-Phase Phosphoprotein 6 homolog YNR034W sol1 Suppressor Of Los1-1 Exit from rapamycin-induced GrOwth YNR034W-A ego4 arrest Mitochondrial Ribosomal Protein, Small YNR036C mrps12 subunit YNR046W trm112 TRna Methyltransferase YNR077C "" "" YOL001W pho80 PHOsphate metabolism YOL005C rpb11 RNA Polymerase B YOL024W "" "" YOL042W ngl1 "" YOL052C spe2 SPErmidine auxotroph YOL067C rtg1 ReTroGrade regulation YOL077W-A atp19 ATP synthase YOL085C "" "" YOL086W-A mhf1 Mph1-associated Histone-Fold protein YOL088C mpd2 Multicopy suppressor of PDI1 deletion YOL101C izh4 Implicated in Zinc Homeostasis YOL102C tpt1 tRNA 2'-PhosphoTransferase YOL111C mdy2 Mating-Deficient Yeast YOL129W vps68 Vacuolar Protein Sorting YOL135C med7 MEDiator complex YOL139C cdc33 Cell Division Cycle YOL149W dcp1 mRNA DeCaPping YOL162W "" "" YOR002W alg6 Asparagine-Linked Glycosylation YOR003W ysp3 Yeast Subtilisin-like Protease III YOR011W-A "" "" YOR015W "" "" YOR020W-A mco10 Mitochondrial Class One protein of 10 kDa YOR021C sfm1 Spout Family Methyltransferase 1 YOR060C sld7 Synthetic Lethality with Dpb11-24 YOR072W "" "" YOR094W arf3 ADP-Ribosylation Factor YOR097C "" "" YOR101W ras1 homologous to RAS proto-oncogene YOR104W pin2 Psi+ INducibility

115 Regulator of heterotrimeric G protein YOR107W rgs2 Signaling YOR131C "" "" YOR142W lsc1 Ligase of Succinyl-CoA YOR145C pno1 Partner of NOb1 YOR148C spp2 Suppressor of PrP Mitochondrial Ribosomal Protein, Large YOR150W mrpl23 subunit YOR161C-C "" "" Diadenosine and Diphosphoinositol YOR163W ddp1 Polyphosphate phosphohydrolase YOR185C gsp2 Genetic Suppressor of Prp20-1 YOR194C toa1 "" YOR210W rpb10 RNA Polymerase B YOR214C spr2 SPorulation-Regulated YOR215C aim41 Altered Inheritance of Mitochondria YOR226C isu2 IscU homolog YOR232W mge1 Mitochondrial GrpE YOR235W irc13 Increased Recombination Centers YOR258W hnt3 Histidine triad NucleoTide-binding YOR262W gpn2 Gly-Pro-Asn (N) motif YOR265W rbl2 Rescues Beta-tubulin Lethality YOR266W pnt1 PeNTamidine resistance YOR285W rdl1 RhoDanese-Like protein YOR288C mpd1 Multicopy suppressor of PDI1 deletion YOR293C-A "" "" YOR295W uaf30 Upstream Activation Factor subunit YOR304C-A bil1 Bud6-Interacting Ligand YOR313C sps4 SPorulation Specific trancript YOR320C gnt1 GlcNAc Transferase YOR367W scp1 S. cerevisiae CalPonin YOR376W "" "" YOR376W-A "" "" YOR377W atf1 AcetylTransFerase YOR382W fit2 Facilitator of Iron Transport YOR383C fit3 Facilitator of Iron Transport YOR387C "" "" YPL011C taf3 TATA binding protein-Associated Factor Mitochondrial Ribosomal Protein, Small YPL013C mrps16 subunit YPL034W "" "" YPL046C elc1 ELongin C

116 YPL047W sgf11 SaGa associated Factor 11kDa YPL056C lcl1 Long Chronological Lifespan YPL059W grx5 GlutaRedoXin YPL070W muk1 coMpUtationally-linked to Kap95 YPL090C rps6A Ribosomal Protein of the Small subunit YPL096C-A eri1 ER-associated Ras Inhibitor YPL106C sse1 Stress Seventy subfamily E YPL108W "" "" YPL122C tfb2 Transcription Factor B subunit 2 YPL131W rpl5 Ribosomal Protein of the Large subunit YPL143W rpl33A Ribosomal Protein of the Large subunit YPL145C kes1 KrE11-1 Suppressor YPL149W atg5 AuTophaGy related YPL154C pep4 carboxyPEPtidase Y-deficient YPL156C prm4 Pheromone-Regulated Membrane protein YPL159C pet20 PETite colonies YPL162C "" "" YPL166W atg29 AuTophaGy related YPL175W spt14 SuPpressor of Ty YPL189C-A coa2 Cytochrome Oxidase Assembly YPL200W csm4 Chromosome Segregation in Meiosis YPL203W tpk2 Takashi's Protein Kinase Genes de Respuesta a Estres (spanish for YPL223C gre1 stress responsive genes) YPL225W "" "" YPL232W sso1 Supressor of Sec One YPL234C vma11 Vacuolar Membrane Atpase YPL237W sui3 SUppressor of Initiator codon YPL249C-A rpl36B Ribosomal Protein of the Large subunit YPL261C "" "" YPL271W atp15 ATP synthase YPL278C "" "" YPR009W sut2 Sterol UpTake YPR027C "" "" YPR037C erv2 Essential for Respiration and Viability YPR046W mcm16 MiniChromosome Maintenance YPR064W "" "" YPR065W rox1 Regulation by OXygen YPR071W "" "" YPR098C tmh18 TMem205 Homolog of 18 kDa YPR113W pis1 Phosphatidyl Inositol Synthase YPR137W rrp9 Ribosomal RNA Processing

117 YPR167C met16 METhionine requiring YPR168W nut2 Negative regulation of URS Two YPR170W-B "" "" YPR174C csa1 Cdc5 SPB Anchor YPR188C mlc2 Myo1p Light Chain Histone and other Protein YPR193C hpa2 Acetyltransferase YPR195C "" "" ZOD1 zod1 Zone Of Disparity "" Denotes Uncharacterized/Unnamed Genes

118 Table 3.5 – L2 gene drop outs, extended data Gene Systematic Name Gene Standard Name Gene Name RPR1 rpr1 RNase P Ribonucleoprotein RUF23 ruf23 RNA of Unknown Function snR10 snr10 Small Nucleolar RNA snR128 snr128 Small Nucleolar RNA snR17a snr17A Small Nucleolar RNA snR19 snr19 Small Nuclear RNA snR191 snr191 Small Nucleolar RNA snR34 snr34 Small Nucleolar RNA snR39B snr39B Small Nucleolar RNA snR41 snr41 Small Nucleolar RNA snR43 snr43 Small Nucleolar RNA snR61 snr61 Small Nucleolar RNA snR63 snr63 Small Nucleolar RNA snR65 snr65 Small Nucleolar RNA snR66 snr66 Small Nucleolar RNA snR79 snr79 Small Nucleolar RNA snR7-L snr7-L Small Nuclear RNA snR7-S snr7-S Small Nuclear RNA snR80 snr80 Small Nucleolar RNA snR81 snr81 Small Nucleolar RNA snR85 snr85 Small Nucleolar RNA snR87 snr87 Small Nucleolar RNA snR9 snr9 Small Nucleolar RNA SRG1 srg1 SER3 Regulatory Gene YAL068C pau8 seriPAUperin YBL003C hta2 Histone h Two A YBL008W-A "" "" YBL018C pop8 Processing Of Precursor RNAs YBL029C-A "" "" YBL044W "" "" YBL071W-A kti11 Kluveromyces lactis Toxin Insensitive YBL090W mrp21 Mitochondrial Ribosomal Protein YBR009C hhf1 Histone H Four YBR010W hhr1 Histone H Three YBR016W "" "" YBR022W poa1 Phosphatase Of ADP-ribose 1"-phosphate YBR032W "" "" YBR040W fig1 Factor-Induced Gene YBR046C zta1 ZeTA-crystallin YBR047W fmp23 Found in Mitochondrial Proteome

119 YBR052C rfs1 Rad55 (Fifty-five) Suppressor YBR070C alg14 Asparagine Linked Glycosylation YBR071W "" "" YBR072C-A "" "" YBR077C slm4 Synthetic Lethal with Mss4 YBR085C-A "" "" YBR089C-A nhp6B Non-Histone Protein YBR090C "" "" YBR091C tim12 Translocase of the Inner Membrane YBR096W "" "" YBR111W-A sus1 Sl gene Upstream of ySa1 YBR120C cbp6 Cytochrome B Protein synthesis YBR139W atg42 AuTophaGy YBR149W ara1 D-ARAbinose dehydrogenase YBR156C sli15 Synthetically Lethal with Ipl1 YBR171W sec66 SECretory YBR188C ntc20 Prp19p (NineTeen)-associated Complex YBR196C-A "" "" YBR196C-B "" "" YBR210W erv15 ER Vesicle Protein YBR233W-A dad3 Duo1 And Dam1 interacting YBR242W "" "" YBR255C-A rcf3 Respiratory superComplex Factor Mitochondrial Ribosomal Protein, Large YBR268W mrpl37 subunit YBR302C cos2 COnserved Sequence YCL001W-A "" "" YCL001W-B "" "" YCL035C grx1 GlutaRedoXin YCL042W "" "" YCL044C mgr1 Mitochondrial Genome Required YCR002C cdc10 Cell Division Cycle Mitochondrial Ribosomal Protein, Large YCR003W mrpl32 subunit YCR020C pet18 PETite colonies YCR024C-B "" "" YCR025C "" "" YCR028C-A rim1 Replication In Mitochondria YCR035C rrp43 Ribosomal RNA Processing YCR044C per1 protein Processing in the ER Exit from rapamycin-induced GrOwth YCR075W-A ego2 arrest

120 YCR086W csm1 Chromosome Segregation in Meiosis YCR099C "" "" YCR101C "" "" YCR104W pau3 seriPAUperin family YCR108C "" "" YDL007C-A "" "" YDL018C erp3 Emp24p/Erv25p Related Protein YDL045C fad1 FAD synthetase YDL045W-A mrp10 Mitochondrial Ribosomal Protein YDL046W npc2 Niemann Pick type C homolog YDL047W sit4 Suppressor of Initiation of Transcription YDL051W lhp1 La-Homologous Protein YDL067C cox9 Cytochrome c OXidase Yeast Endoplasmic reticulum YDL072C yet3 Transmembrane protein YDL081C rpp1A Ribosomal Protein P1 Alpha YDL088C asm4 Anti-Suppressor in Multicopy YDL092W srp14 Signal Recognition Particle YDL105W nse4 Non-SMC Element YDL120W yfh1 Yeast Frataxin Homolog YDL123W sna4 Sensitivity to NA+ YDL130W spp1B Ribosomal Protein P1 Beta YDL135C rdi1 Rho GDP Dissociation Inhibitor Suppressor of Chromosome YDL139C scm3 Missegregation YDL157C "" "" YDL160C-A mhf2 Mph1-associated Histone-Fold protein YDL173W par32 Phosphorylated After Rapamycin YDL189W rbs1 RNA-Binding Suppressor of PAS kinase O-6-MethylGuanine-DNA YDL200C mgt1 methylTransferase YDL201W trm8 Transfer RNA Methyltransferase YDL212W shr3 Super high Histidine Resistant YDL219W dtd1 D-Tyr-tRNA(Tyr) Deacylase YDL232W ost4 OligoSaccharylTransferase YDL235C ypd1 tYrosine Phosphatase Dependent YDL241W "" "" YDL245C hxt15 HeXose Transporter YDR002W yrb1 Yeast Ran Binder YDR003W rcr2 Resistance to Congo Red YDR003W-A "" "" YDR004W rad57 RADiation sensitive

121 YDR016C dad1 Duo1 And Dam1 interacting YDR022C atg31 AuTophaGy related YDR029W "" "" Mitochondrial Intermembrane space YDR031W mix14 CX(n)C motif protein YDR034W-B "" "" YDR042C "" "" Negative Regulator of Glucose-repressed YDR043C nrg1 genes YDR055W pst1 Protoplasts-SecreTed YDR059C ubc5 UBiquitin-Conjugating YDR063W aim7 Altered Inheritance rate of Mitochondria YDR068W dos2 "" YDR070C fmp16 Found in Mitochondrial Proteome YDR073W snf11 Sucrose NonFermenting YDR079C-A tfb5 "" YDR084C tvp23 Tlg2-Vesicle Protein YDR090C ilt1 Ionic Liquid Tolerance YDR102C "" "" YDR119W-A cox26 "" YDR139C rub1 Related to UBiquitin YDR157W "" "" YDR168W cdc37 Cell Division Cycle YDR169C-A "" "" YDR179W-A nvj3 Nucleus-Vacuole Junction YDR181C sas4 Something About Silencing YDR194W-A "" "" YDR197W cbs2 Cytochrome B Synthesis YDR209C "" "" YDR210W "" "" YDR225W hta1 Histone h Two A Protein 1 of Cleavage and polyadenylation YDR228C pcf11 Factor I YDR233C rtn1 ReTiculoN-like YDR246W trs23 TRapp Subunit YDR248C "" "" YDR272W glo2 GLyOxalase Mitochondrial Glutaredoxin-like Protein of YDR286C mgp12 12 kDa YDR315C ipk1 Inositol Polyphosphate Kinase YDR344C "" "" YDR366C mor1 Mitochondrial mORphology affecting

122 YDR373W frq1 FReQuenin homolog YDR374C pho92 PHOsphate metabolism YDR391C "" "" YDR412W rrp17 Ribosomal RNA Processing YDR438W thi74 THIamine regulon YDR446W ecm11 ExtraCellular Mutant YDR454C guk1 GUanylate Kinase Mitochondrial Ribosomal Protein, Large YDR462W mrpl28 subunit YDR529C qcr7 ubiQuinol-cytochrome C oxidoReductase YDR532C kre28 "" Translocase of the Inner Mitochondrial YEL020W-A tim9 membrane YEL033W mtc7 Maintenance of Telomere Capping YEL048C tca17 TRAPP Complex Associated protein YEL051W vma8 Vacuolar Membrane Atpase YEL057C sdd1 Suppressor of Degenerative Death YEL059C-A som1 SOrting Mitochondrial YER011W tir1 TIp1-Related YER026C cho1 CHOline requiring YER030W chz1 Chaperone for Htz1/H2A-H2B dimer YER034W "" "" YER049W tpa1 Termination and PolyAdenylation YER074W rps24A Ribosomal Protein of the Small subunit YER085C "" "" YER092W ies5 Ino Eighty Subunit YER101C ast2 ATPase STabilizing YER126C nsa2 Nop Seven Associated YER127W lcp5 Lethal with Conditional Pap1 YER145C-A "" "" YER156C myg1 "" YER165W pab1 Poly(A) Binding protein YER177W bmh1 Brain Modulosignalin Homolog WW domain containing protein interacting YFL010C wwm1 with Metacaspase YFL015C "" "" YFR001W loc1 LOCalization of mRNA YFR003C ypi1 Yeast Phosphatase Inhibitor YFR032C-B min10 mitochondrial MINi protein of 10 kDa YFR033C qcr6 ubiQuinol-cytochrome C oxidoReductase YFR054C "" "" YGL007W brp1 ""

123 YGL030W rpl30 Ribosomal Protein of the Large subunit YGL032C aga2 a-AGglutinin YGL039W "" "" YGL058W rad6 RADiation sensitive YGL061C duo1 Death Upon Overproduction YGL068W mnp1 Mitochondrial-Nucleoid Protein YGL070C rpb9 RNA Polymerase B YGL079W kxd1 KxDL homolog YGL080W mpc1 Mitochondrial Pyruvate Carrier YGL081W "" "" YGL096W tos8 Target Of Sbf YGL101W ygk1 "" YGL175C sae2 Sporulation in the Absence of spo Eleven YGL187C cox4 Cytochrome c OXidase YGL191W cox13 Cytochrome c OXidase YGL194C-A "" "" YGL198W yip4 Ypt-Interacting Protein YGL222C edc1 Enhancer of mRNA DeCapping YGL224C sdt1 Suppressor of Disruption of TFIIS YGL226C-A ost5 OligoSaccharylTransferase YGL226W mtc3 Maintenance of Telomere Capping YGL248W pde1 PhosphoDiEsterase YGL258W vel1 VELum formation YGL259W yps5 YaPSin YGR001C efm5 Elongation Factor Methyltransferase YGR018C "" "" YGR028W msp1 Mitochondrial Sorting of Proteins YGR035W-A "" "" YGR038W orm1 "" YGR039W "" "" YGR055W mup1 Methionine UPtake YGR092W dbf2 DumbBell Former YGR102C gtf1 Glutaminyl Transamidase subunit F YGR106C voa1 V0 Assembly protein YGR149W gpc1 GlyceroPhosphoCholine acyltransferase YGR159C nsr1 "" YGR161C rts3 "" YGR161W-C "" "" YGR168C pex35 PEroXin YGR169C-A lso2 Late-annotated Small Open reading frame YGR174C cbp4 Cytochrome B mRNA Processing YGR183C qcr9 ubiQuinol-cytochrome C oxidoReductase

124 YGR192C tdh3 Triose-phosphate DeHydrogenase YGR207C cir1 Changed Intracellular Redox state YGR225W ama1 Activator of Meiotic APC/C YGR230W bns1 Bypasses Need for Spo12p YGR263C say1 Steryl Acetyl hYdrolase YGR271C-A efg1 Exit From G1 YGR273C "" "" YGR280C pxr1 PinX1-Related gene YGR282C bgl2 Beta-GLucanase YGR285C zuo1 ZUOtin YHL018W mco14 Mitochondrial Class One protein of 14 kDa YHL031C gos1 GOlgi Snare YHL037C "" "" YHL042W "" "" YHR001W osh7 OxySterol binding protein Homolog YHR014W spo13 SPOrulation YHR039C-A vma10 "" YHR053C cup1-1 Cu, copper, CUPrum YHR057C cpr2 Cyclosporin A-sensitive Proline Rotamase YHR081W lrp1 Like RrP6 YHR086W-A "" "" YHR089C gar1 Glycine Arginine Rich YHR130C "" "" YHR133C nsg1 "" YHR152W spo12 SPOrulation YHR175W ctr2 Copper TRansport YHR175W-A "" "" YHR180W "" "" YHR185C pfs1 Prospore Formation at Spindles YHR213W-A "" "" YIL010W dot5 Disruptor Of Telomeric silencing YIL012W "" "" YIL015W bar1 BARrier to the alpha factor response APical growth revealed by Quantitative YIL040W apq12 morphological analysis YIL051C mmf1 Mitochondrial Matrix Factor YIL053W gpp1 Glycerol-3-Phosphate Phosphatase Arginine methyltransferase-Interacting YIL079C air1 RING finger protein YIL086C "" "" YIL087C aim19 Altered Inheritance rate of Mitochondria YIL097W fyv10 Function required for Yeast Viability

125 YIL102C "" "" YIL102C-A "" "" YIL117C prm5 Pheromone-Regulated Membrane protein YIL127C rrt14 Regulator of rDNA Transcription YIL133C rpl16A Ribosomal Protein of the Large subunit YIL138C tpm2 TroPoMyosin YIL139C rev7 REVersionless YIL156W-B mco8 Mitochondrial Class One protein of 8 kDa YIR008C pri1 DNA PRImase YIR021W-A "" "" YIR042C "" "" YJL011C rpc17 RNA Polymerase C YJL028W "" "" YJL030W mad2 Mitotic Arrest-Deficient YJL043W "" "" YJL077W-B "" "" YJL088W arg3 ARGinine requiring YJL089W sip4 SNF1-Interacting Protein Presequence translocase-Associated YJL104W pam16 Motor YJL115W asf1 Anti-Silencing Function YJL124C lsm1 Like SM YJL127C-B mco6 Mitochondrial Class One protein of 6 kDa YJL145W sfh5 Sec Fourteen Homolog YJL160C pir5 Protein with Internal Repeats YJL177W rpl17B Ribosomal Protein of the Large subunit YJL178C atg27 AuTophaGy related YJL179W pfd1 PreFolDin YJL185C atg36 AuTophaGy related YJL203W prp21 Pre-mRNA Processing YJL222W vth2 Vps Ten Homolog YJR007W sui2 SUppressor of Initiator codon YJR044C vos55 Vacuolar Protein Sorting YJR045C ssc1 Stress-Seventy subfamily C YJR063W rpa12 RNA Polymerase A YJR069C ham1 6-n-HydroxylAMinopurine sensitive YJR085C tmh11 TMem14 Homolog of 11 kDa YJR086W ste18 STErile YJR102C vps25 Vacuolar Protein Sorting YJR112W-A "" "" YJR146W "" "" YJR151W-A "" ""

126 YJR155W aad10 Aryl-Alcohol Dehydrogenase YKL002W did4 Doa4-Independent Degradation YKL003C mrp17 Mitochondrial Ribosomal Protein YKL023C-A min9 mitochondrial MINi protein of 9 kDa YKL024C ura6 URAcil requiring YKL042W spc42 Spindle Pole Component YKL044W mmo1 Mini Mitochondria ORF YKL049C cse4 Chromosome SEgregation YKL055C oar1 3-Oxoacyl-[Acyl-carrier-protein] Reductase YKL063C "" "" YKL077W psg1 Pma1 Stabilization in the Golgi YKL082C rrp14 Ribosomal RNA Processing YKL085W mdh1 Malate DeHydrogenase YKL096W cwp1 Cell Wall Protein YKL096W-A cwp2 Cell Wall Protein YKL102C "" "" YKL108W sld2 Synthetically Lethal with Dpb11-1 YKL119C vph2 Vacuolar pH YKL122C srp21 Signal Recognition Particle Mitochondrial Ribosomal Protein, Large YKL138C mrpl31 subunit YKL138C-A hsk3 Helper of ASK1 YKL159C rcn1 Regulator of CalciNeurin YKL160W elf1 ELongation Factor Mitochondrial Ribosomal Protein, Large YKL170W mrpl38 subunit YKL183C-A "" "" YKL196C ykt6 "" YKR002W pap1 Poly(A) Polymerase YKR015C "" "" YKR035W-A did2 Doa4-Independent Degradation YKR046C pln1 PeriLipiN YKR048C nap1 Nucleosome Assembly Protein YKR049C fmp46 Found in Mitochondrial Proteome YKR051W hfl1 Has Fused Lysosomes YKR052C mrs4 Mitochondrial RNA Splicing YKR073C "" "" Polarized growth Chromatin-associated YKR095W-A pcc1 Controller YLL009C cox17 Cytochrome c OXidase YLL018C-A cox19 Cytochrome c OXidase YLL027W isa1 Iron Sulfur Assembly

127 YLL038C ent4 Epsin N-Terminal homology YLL046C rnp1 RiboNucleoProtein YLL052C aqy2 AQuaporin from Yeast YLL053C "" "" Presequence translocase-Associated YLR008C pam18 Motor YLR009W rlp24 Ribosomal-Like Protein YLR010C ten1 TElomeric pathways with STn1 YLR011W lot6 LOw Temperature-responsive YLR012C "" "" YLR013W gat3 "" YLR017W meu1 Multicopy Enhancer of UAS2 YLR021W irc25 Increased Recombination Centers YLR026C sed5 Suppressor of Erd2 Deletion YLR031W "" "" YLR033W rsc58 Remodel the Structure of Chromatin YLR036C "" "" Negative regulator of the Filamentous YLR042C nfg1 Growth MAPK pathway YLR043C trx1 ThioRedoXin YLR064W per33 Pore and ER protein, 33 kDa YLR068W fyv7 Function required for Yeast Viability YLR074C bud20 BUD site selection YLR099W-A mim2 "" YLR104W lcl2 Long Chronological Lifespan 2 YLR109W ahp1 Alkyl HydroPeroxide reductase YLR111W "" "" YLR132C usb1 U Six Biogenesis YLR145W rmp1 RNase MRP Protein YLR146C spe4 SPErmidine auxotroph YLR155C asp3-1 ASParaginase YLR157W-D "" "" YLR167W rps31 Ribosomal Protein of the Small subunit YLR211C atg38 AuTophaGy related YLR215C cdc123 Cell Division Cycle YLR243W gpn3 Gly-Pro-Asn (N) motif YLR250W ssp120 Saccharomyces Secretory Protein YLR251W sym1 Stress-inducible Yeast Mpv17 YLR254C ndl1 NuDeL homolog YLR261C vps63 Vacuolar Protein Sorting YLR262C-A tma7 Translation Machinery Associated YLR268W sec22 SECretory

128 YLR287C "" "" YLR298C yhc1 Yeast Homolog of human U1C YLR323C cwc24 Complexed With Cef1p YLR333C rps25B Ribosomal Protein of the Small subunit YLR344W rpl26A Ribosomal Protein of the Large subunit YLR346C cis1 CItrinin Sensitive knockout YLR363C nmd4 Nonsense-Mediated mRNA Decay YLR365W "" "" YLR390W-A ccq14 Covalently linked Cell Wall protein YLR406C rpl31B Ribosomal Protein of the Large subunit YLR445W gmc2 Grand Meiotic recombination Cluster YML001W ypt7 Yeast Protein Two YML007C-A min4 mitochondrial MINi protein of 4 kDa YML018C "" "" YML022W apt1 Adenine PhosphoribosylTransferase YML026C rps18B Ribosomal Protein of the Small subunit YML047C prm6 Pheromone-Regulated Membrane protein YML048W gsf2 Glucose Signaling Factor YML074C fpr3 Fk 506-sensitive Proline Rotamase YML079W "" "" YML129C cox14 Cytochrome c OXidase YMR041C ara2 ARAbinose YMR052W far3 Factor ARrest YMR069W nat4 N-AcetylTransferase YMR070W mot3 Modifier of Transcription YMR074C sdd2 Suppressor of Degenerative Death YMR081C isf1 Increasing Suppression Factor YMR082C "" "" YMR084W "" "" YMR103C "" "" YMR105W-A "" "" YMR107W spg4 Stationary Phase Gene YMR114C "" "" YMR122C "" "" YMR138W cin4 Chromosome INstability YMR156C tpp1 Three Prime Phosphatase YMR161W hlj1 HomoLogous to E. coli dnaJ protein YMR175W-A "" "" YMR181C "" "" YMR183C sso2 Supressor of Sec One Alpha1-proteinase inhibitor-Degradation YMR184W add37 Deficient

129 YMR195W icy1 Interacting with the CYtoskeleton YMR230W-A "" "" YMR242W-A "" "" YMR247W-A "" "" YMR252C mlo1 Mitochondrially LOcalized protein YMR255W gfd1 Good For Dbp5p Mitochondrial Ribosomal Protein, Large YMR286W mrpl33 subunit YMR295C "" "" YMR312W elp6 ELongator Protein YMR320W "" "" YMR321C "" "" YMR325W pau19 seriPAUperin YNL030W hhf2 Histone H Four YNL031C hht2 Histone H Three YNL044W yip3 Ypt-Interacting Protein YNL050C "" "" YNL055C por1 PORin YNL079C tpm1 TroPoMyosin MItochondrial contact site and Cristae YNL100W mic27 organizing system YNL111C cyb5 CYtochrome B YNL119W ncs2 Needs Cla4 to Survive YNL122C mrp35 Mitochondrial Ribosomal Protein YNL130C cpt1 CholinePhosphoTransferase YNL130C-A dgr1 2-Deoxy-Glucose Resistant 1 YNL133C fyv6 Function required for Yeast Viability YNL135C fpr1 Fk 506-sensitive Proline Rotamase YNL148C alf1 ALpha tubulin Folding YNL153C gim3 Gene Involved in Microtubule biogenesis YNL158W pga1 Processing of Gas1p and ALP YNL159C asi2 Amino acid Sensor-Independent YNL160W ygp1 Yeast GlycoProtein YNL184C "" "" YNL200C nnr1 Nicotinamide Nucleotide Repair YNL232W csl4 Cep1 Synthetic Lethal YNL234W "" "" YNL245C cwc25 Complexed With Cef1p YNL255C gis2 GIg Suppressor YNL274C gor1 GlyOxylate Reductase YNL277W-A "" "" YNL281W hch1 High-Copy Hsp90 suppressor

130 YNL320W "" "" YNR004W swm2 Synthetic With mud2-delta YNR010W cse2 Chromosome SEgregation YNR024W mpp6 M-Phase Phosphoprotein 6 homolog YNR034W sol1 Suppressor Of Los1-1 Exit from rapamycin-induced GrOwth YNR034W-A ego4 arrest Mitochondrial Ribosomal Protein, Small YNR036C mrps12 subunit YNR046W trm112 TRna Methyltransferase YNR077C "" "" YOL001W pho80 PHOsphate metabolism YOL005C rpb11 RNA Polymerase B YOL024W "" "" YOL042W ngl1 "" YOL052C spe2 SPErmidine auxotroph YOL067C rtg1 ReTroGrade regulation YOL077W-A atp19 ATP synthase YOL085C "" "" YOL086W-A mhf1 Mph1-associated Histone-Fold protein YOL088C mpd2 Multicopy suppressor of PDI1 deletion YOL101C izh4 Implicated in Zinc Homeostasis YOL102C tpt1 tRNA 2'-PhosphoTransferase YOL111C mdy2 Mating-Deficient Yeast YOL129W vps68 Vacuolar Protein Sorting YOL135C med7 MEDiator complex YOL139C cdc33 Cell Division Cycle YOL149W dcp1 mRNA DeCaPping YOL162W "" "" YOR002W alg6 Asparagine-Linked Glycosylation YOR003W ysp3 Yeast Subtilisin-like Protease III YOR011W-A "" "" YOR015W "" "" YOR020W-A mco10 Mitochondrial Class One protein of 10 kDa YOR021C sfm1 Spout Family Methyltransferase 1 YOR060C sld7 Synthetic Lethality with Dpb11-24 YOR072W "" "" YOR094W arf3 ADP-Ribosylation Factor YOR097C "" "" YOR101W ras1 homologous to RAS proto-oncogene YOR104W pin2 Psi+ INducibility

131 Regulator of heterotrimeric G protein YOR107W rgs2 Signaling YOR131C "" "" YOR142W lsc1 Ligase of Succinyl-CoA YOR145C pno1 Partner of NOb1 YOR148C spp2 Suppressor of PrP Mitochondrial Ribosomal Protein, Large YOR150W mrpl23 subunit YOR161C-C "" "" Diadenosine and Diphosphoinositol YOR163W ddp1 Polyphosphate phosphohydrolase YOR185C gsp2 Genetic Suppressor of Prp20-1 YOR194C toa1 "" YOR210W rpb10 RNA Polymerase B YOR214C spr2 SPorulation-Regulated YOR215C aim41 Altered Inheritance of Mitochondria YOR226C isu2 IscU homolog YOR232W mge1 Mitochondrial GrpE YOR235W irc13 Increased Recombination Centers YOR258W hnt3 Histidine triad NucleoTide-binding YOR262W gpn2 Gly-Pro-Asn (N) motif YOR265W rbl2 Rescues Beta-tubulin Lethality YOR266W pnt1 PeNTamidine resistance YOR285W rdl1 RhoDanese-Like protein YOR288C mpd1 Multicopy suppressor of PDI1 deletion YOR293C-A "" "" YOR295W uaf30 Upstream Activation Factor subunit YOR304C-A bil1 Bud6-Interacting Ligand YOR313C sps4 SPorulation Specific trancript YOR320C gnt1 GlcNAc Transferase YOR367W scp1 S. cerevisiae CalPonin YOR376W "" "" YOR376W-A "" "" YOR377W atf1 AcetylTransFerase YOR382W fit2 Facilitator of Iron Transport YOR383C fit3 Facilitator of Iron Transport YOR387C "" "" YPL011C taf3 TATA binding protein-Associated Factor Mitochondrial Ribosomal Protein, Small YPL013C mrps16 subunit YPL034W "" "" YPL046C elc1 ELongin C

132 YPL047W sgf11 SaGa associated Factor 11kDa YPL056C lcl1 Long Chronological Lifespan YPL059W grx5 GlutaRedoXin YPL070W muk1 coMpUtationally-linked to Kap95 YPL090C rps6A Ribosomal Protein of the Small subunit YPL096C-A eri1 ER-associated Ras Inhibitor YPL106C sse1 Stress Seventy subfamily E YPL108W "" "" YPL122C tfb2 Transcription Factor B subunit 2 YPL131W rpl5 Ribosomal Protein of the Large subunit YPL143W rpl33A Ribosomal Protein of the Large subunit YPL145C kes1 KrE11-1 Suppressor YPL149W atg5 AuTophaGy related YPL154C pep4 carboxyPEPtidase Y-deficient YPL156C prm4 Pheromone-Regulated Membrane protein YPL159C pet20 PETite colonies YPL162C "" "" YPL166W atg29 AuTophaGy related YPL175W spt14 SuPpressor of Ty YPL189C-A coa2 Cytochrome Oxidase Assembly YPL200W csm4 Chromosome Segregation in Meiosis YPL203W tpk2 Takashi's Protein Kinase Genes de Respuesta a Estres (spanish for YPL223C gre1 stress responsive genes) YPL225W "" "" YPL232W sso1 Supressor of Sec One YPL234C vma11 Vacuolar Membrane Atpase YPL237W sui3 SUppressor of Initiator codon YPL249C-A rpl36B Ribosomal Protein of the Large subunit YPL261C "" "" YPL271W atp15 ATP synthase YPL278C "" "" YPR009W sut2 Sterol UpTake YPR027C "" "" YPR037C erv2 Essential for Respiration and Viability YPR046W mcm16 MiniChromosome Maintenance YPR064W "" "" YPR065W rox1 Regulation by OXygen YPR071W "" "" YPR098C tmh18 TMem205 Homolog of 18 kDa YPR113W pis1 Phosphatidyl Inositol Synthase YPR137W rrp9 Ribosomal RNA Processing

133 YPR167C met16 METhionine requiring YPR168W nut2 Negative regulation of URS Two YPR170W-B "" "" YPR174C csa1 Cdc5 SPB Anchor YPR188C mlc2 Myo1p Light Chain Histone and other Protein YPR193C hpa2 Acetyltransferase YPR195C "" "" ZOD1 zod1 Zone Of Disparity "" Denotes Uncharacterized/Unnamed Genes

134 References

Alseth, I., Eide, L., Pirovano, M., Rognes, T., Seeberg, E., and Bjoras, M. (1999). The Saccharomyces cerevisiae homologues of endonuclease III from Escherichia coli, Ntg1 and Ntg2, are both required for efficient repair of spontaneous and induced oxidative DNA damage in yeast. Mol Cell Biol 19, 3779-3787.

Alvaro, D., Lisby, M., and Rothstein, R. (2007). Genome-wide analysis of Rad52 foci reveals diverse mechanisms impacting recombination. PLoS Genet 3, e228.

Arensburger, P., Hice, R.H., Zhou, L., Smith, R.C., Tom, A.C., Wright, J.A., Knapp, J., O’Brochta, D.A., Craig, N.L., and Atkinson. (2011). Phylogenetic and functional characterization of the hAT transposon superfamily. Genetics 188, 45-57.

Askree, S.H., Yehuda, T., Smolikov, S., Gurevich, R., Hawk, J., Coker, C., Krauskopf, A., Kupiec, M., and McEachern, M.J. (2004). A genome-wide screen for Saccharomyces cerevisiae deletion mutants that affect telomere length. Proc Natl Acad Sci USA 101, 8658-8663.

Bermudez-Lopez, M., Ceschia, A., de Piccoli, G., Colomina, N., Pasero, P., Aragon, L, and Torres- Rosell, J. (2010). The Smc5/6 complex is required for dissolution of DNA-mediated sister chromatid linkages. Nucleic Acids Res 38, 6502-0512.

Comino, A., Mancek, M., and Komel, R. (2001). Ecm11 is located in the cell nucleus throughout mitosis. Pflugers Arch 442, 179-181

Conrad, M.N., Lee, C.Y., Chao, G., Shinohara, M., Kosaka, H., Shinohara, A., Conchello, J.A., and Dresser, M.E. (2008). Rapid telomere movement in meiotic prophase is promoted by NDJ1, MPS3, and CSM4 and is modulated by recombination. Cell 133, 1175-1187.

Daley, J.M., Wilson, T.E., and Ramotar, D. (2010). Genetic interactions between HNT3/Aprataxin and RAD27/FEN1 suggest parallel pathways for 5’ end processing during base excision repair. DNA Repair 9, 690-699.

Eason, R.G., Pourmand, N., Tongprasit, W., Herman, Z.S., Anthony, K., Jejelowo, O., Davis, R.W., and Stolc, V. (2004). Characterization of synthetic DNA bar codes in Saccharomyces cerevisiae gene-deletion strains. Proc Natl Acad Sci USA 101, 11046-11051.

Erdemir, T., Bilican B., Cagatay, T., Goding, C.R., and Yavuzer, U. (2002). Saccharomyces cerevisiae C1D is implicated in both non-homologous DNA end joining and homologous recombination. Mol Microbiol 46, 947-957.

135

Gallagher, D.N., Pham, N., Tsai, A.M., Janto, A.N., Choi, J., Ira, G., and Haber, J.E. (2020). A Rad51-independent pathway promotes single-strand template repair in gene editing. PLoS Genet 16(10), e1008689.

Game, J.C., and Chernikova, S.B. (2009). The role of RAD6 in recombinational repair, checkpoints and meiosis via histone medication. DNA Repair 8, 470-482.

Gangadharan, S., Mularoni, L., Fain-Thornton, J., Wheelan, S.J., and Craig, N.L. (2010). DNA transposon Hermes inserts into DNA in nucleosome-free regions in vivo. Proc Natl Sci USA 107, 21966-21972.

Gietz, R.D., and Schiestl, R.H. (1991). Applications of high efficiency lithium acetate

Glatt, S., Letoquart, J., Faux, C., Taylor, N.M., Seraphin, B., and Muller C.W. (2012). The Elongator subcomplex Elp456 is a hexameric RecA-like ATPase. Nat Struct Mol Biol 19, 314-320.

Grunenfelder, B., and Winzeler, E.A. (2002) Treasures and traps in genome-wide data sets: case examples from yeast. Nat Rev Genet 3, 653-661.

Guo, Y., Min Park, J., Cui, B., Humes E., Gangadharan, S., Hung, S., FitzGerald, P.C., Hoe, K.L., Grewal, S.S., Craig, N.L., and Levin, H.L. (2013). Integration profiling of gene function with dense maps of transposon integration. Genetics 195, 599-609.

Hickman, A.B., Ewis, H.E., Li, X., Knapp, J.A., Laver, T., Doss, A.L., Tolun, G, Steven, A.C., Grishaev, A., Bax, A., et al. (2014). Structural basis of hAT transposon end recognition by Hermes, an octameric DNA transposase from Musca domestica. Cell 158, 353-367.

Hsu, M.Y., Eagle, S.G., Inouye, M., and Inouye, S. (1992). Cell-free synthesis of the branched RNA-linked msDNA from retron-Ec67 of Escherichia coli. J Biol Chem 267(20), 13823–13829.

Hu, B., Liao, C., Millson, S.H., Mollapour, M., Prodromou, C., Pearl, L.H., Piper, P.W., and Panaretou, B. (2005). Qri2/Nse4, a component of the essential Smc5/6 DNA repair complex. Mol Microbiol 55, 1735-1750.

Hu, Y., Tang, H.B., Liu, N.N., Tong, X.J., Dang, W., Duan, Y.M., Fu, X.H., Zhang, Y., Peng, J., Meng, F.L., et al. (2013). Telomerase-null survivor screening identifies novel telomere recombination regulators. PLoS Genet 9, e1003208.

136 Hughes, T.R., Roberts, C.J., Dai, H., Jones, A.R., Meyer, M.R., Slade, D., Burchard, J., Dow, S., Ward, T.R., Kidd, M.J., et al. (2000) Widespread aneuploidy revealed by DNA microarray expression profiling. Nat Genet 25, 333-337.

Humphreyes, N., Leung, W.K., Argunhan, B., Terentyev, Y., Dvorackova, M., and Tsubouchi, H. (2013). The Ecm11-Gmc2 complex promotes synaptonemal complex formation through assembly of transverse filaments in budding yeast. PLoS Genet 9, e1003194.

Kamimura, Y., Masumoto, H., Sugino, A., and Araki, H. (1998). Sld2, which interacts with Dpb11 in Saccharomyces cerevisiae, is required for chromosomal DNA replication. Mol Cell Biol 18, 6102-6109.

Kanter, D.M., and Kaplan, D.L. (2011). Sld2 binds to origin single-stranded DNA and stimulates DNA annealing. Nucleic Acids Res 39, 2580-2592.

Keskin, H., Shen, Y., Huang, F., Patel, M., Yang, T., Ashley, K., Mazin, A.V., and Storici, F. (2014). Transcript-RNA-templated DNA recombination and repair. Nature 515(7527), 436-439.

Kyrou, K., Hammond, A.M., Galizi, R., Kranjc, N., Burt, A., Beaghton, A.K., Nolan, T., and Crisanti A. (2018) A CRISPR-Cas9 gene drive targeting doublesex causes complete population suppression in caged Anopheles gambiae mosquitos. Nat Biotechnol 36, 1062-1066.

Le, S., Moore, J.K., Haber, J.E., and Greider, C.W. (1999). Rad50 and Rad51 define two pathways that collaborate to maintain telomeres in the absence of telomerase. Genetics 152(1), 143-152.

Lee, S.K., Yu, S.L., Prakash, L., and Prakash, S. (2002). Yeast RAD26, a homolog of the human CSB gene, functions independently of nucleotide excision repair and base excision repair in promoting transcription through damaged bases. Mol Cell Biol 22, 4383-4389.

Lundblad, V., and Blackburn, E.H. (1993). An alternative pathway for yeast telomere maintenance rescues est1 senescence. Cell 73, 347–360.

Paix, A., Folkmann, A., and Seydoux, G. (2017). Precision genome editing using CRISPR-Cas9 and linear repair templates in C. elegans. Methods 15, 86-93.

Prakash, S., and Prakash, L. (2000). Nucleotide excision repair in yeast. Mutat Res 451, 13-24. Richardson, C.D., Kazane, K.R., Feng, S.J., Zelin, E., Bray, N.L., Schafer, A.J., Floor, S.N., and Corn, J.E. (2018). CRISPR-Cas9 genome editing in human cells occurs via the Fanconi anemia pathway. Nat Genet 50(8), 1132-1139.

Sanders, S.L., Jennings, J., Canutescu, A., Link, A.J., and Weil, P.A. (2002). Proteomics of the eukaryotic transcription machinery: identification of proteins associated with components of yeast TFIID by multidimensional mass spectrometry. Mol Cell Biol 22, 4723-4738.

137 Sassanfar, M., and Samson, L. (1990). Identification and preliminary characterization of an o6- methylguanine DNA repair methyltransferase in the yeast Saccharomyces cerevisiae. J Biol Chem 265, 20-25.

Schwabish, M.A., and Struhl, K. (2006). Asf1 mediates histone eviction and deposition during elongation by RNA polymerase II. Mol Cell 22, 415-422.

Sharon, E., Chen, S.A., Khosla, N.M., Smith, J.D., Pritchard, J.K., and Fraser, H.B. (2018) Functional genetic variants revealed by massively parallel precise genome editing. Cell 175(2), 544-557.

Singer, M.S., Kahana, A., Wolf, A.J., Meisinger, L.L., Peterson, S.E., Goggin, C., Mahowald, M., and Gottschling, D.E. (1998). Identification of high-copy disruptors of telomeric silencing in Saccharomyces cerevisiae. Genetics 150, 613-632.

Smith, A.M., Heisler, L.E., Mellor, J., Kaper, F., Thompson, M.J., Chee, M., Roth, F.P., Giaever, G., and Nislow, C. (2009). Quantitative phenotyping via deep barcode sequencing. Genome Res 19, 1836-1842.

Stead, J.A., Costello, J.L., Livingstone, M.J., and Mitchell, P. (2007). The PMC2NT domain of the catalytic exosome subunit Rrp6p provides the interface for binding with its cofactor Rrp47p, a nucleic acid-binding protein. Nucleic Acids Res 35, 5556-5567.

Storici, F., Snipe, J.R., Chan, G.K., Gordenin, D.A., and Resnick, M.A. (2006). Conservative repair of a chromosomal single-strand DNA through two steps of annealing. Mol Cell Biol 26(20), 7654-7657

Su, S.S., and Mitchell, A.P. (1993). Molecular characterization of the yeast meiotic regulatory gene RIM1. Nucleic Acids Res 21, 3789-3797.

Tkach, J.M., Yimit, A., Lee, A.Y., Riffle, M., Costanzo, M, Jaschob, D., Hendry, J.A., Ou, J., Moffat, J, Boone, C., et al. (2012). Dissecting DNA damage response pathways by analyzing protein localization and abundance changes during DNA replication stress. Nat Cell Biol 14, 966-976. transformation of intact yeast cells using single-stranded nucleic acids as carrier. Yeast 7, 253- 263.

138 Van Dyck, E., Foury, F., Stillman, B., and Brill, S.J. (1992). A single-stranded DNA binding protein required for mitochondrial DNA replication in S. cerevisiae is homologous to E. coli SSB. EMBO J 11, 3421-3430.

Vance, J.R., and Wilson, T.E. (2001). Uncoupling of 3’-phosphatase and 5’-kinase functions in budding yeast. Characterization of Saccharomyces cerevisiae DNA 3’-phosphatase (TPP1). J Biol Chem 276, 15073-15081.

Wilson, T.E. (2002). A genomics-based screen for yeast mutants with an altered recombination/end-joining repair ratio. Genetics 162, 677-688.

Yan, Z., Delannoy, M., Ling, C., Daee, D., Osman, F., Muniandy, P.A., Shen, X., Ooostra, A.B., Du, H., and Steltenpool, J. (2010). A histone-fold complex and FANCM form a conserved DNA- remodeling complex to maintain genome stability. Mol Cell 37, 865-878.

Zavec, A.B., Lesnik, U., Komel, R., and Comino, A. (2004). Saccharomyces cerevisiae gene ECM11 is a positive effector of meiosis. FEMS Microbiol Lett 241, 193-199.

139 Chapter 4 – The asymmetry of mismatch correction of divergent substrates following double-strand break repair

Abstract

Faithful DNA double-strand break (DSB) repair is an integral part of maintaining genome stability. Gene conversion (GC) is a common form of DSB repair that occurs when both ends of the DSB have homology to a suitable repair template. Here, using the model organism

Saccharomyces cerevisiae, we show that a heterologous template with mismatches spaced every 9-bp on the centromere-proximal (left arm) of the break shows a significant decrease in successful GC repair, while the same mismatch density on the right arm of the DSB has no significant effect on repair. However, we show that mismatches on the telomere-proximal

(right) arm of the DSB prevent faithful second-end capture, causing mutations in the region adjacent to the GC tract. Furthermore, sequencing of successful GC events show that the mechanism of mismatch correction is dependent on its position with respect to the DSB. While the activity of polymerase 훿 is primarily responsible for incorporating mismatches into the recipient locus on the left side of the DSB, mismatches templated on the right side of the DSB are primarily corrected via components of the mismatch repair (MMR) pathway. Finally, we show that this asymmetry is heavily influenced by the DSB itself, as we see different correction patterns when different endonucleases cleave the same GC substrate.

140 Summary

This chapter includes unpublished data that is based on published work showing the effect of a heterologous template on break-induced replication (Anand et al., 2017). The initial experiments in this chapter (Figures 1-4) were performed with the help of Annie M. Tsai.

Introduction

The genome is constantly under assault from an array of both endogenous and exogenous sources of DNA damaging agents, including, but certainly not limited to, metabolic reactions, radiation, and replication stress (reviewed in Hoeijmakers, 2009; reviewed in

Niedernohofer et al., 2018). Of the many types of DNA damage, the DNA double strand break

(DSB) is a particularly toxic lesion and requires complex mechanisms to repair. DSBs can be repaired in two general ways. One such way is an error-prone mechanism known as nonhomologous end-joining (NHEJ), which involves the direct ligation of the broken chromosome ends and results in small insertions/deletions at the repair junction (reviewed in

Symington and Gautier, 2011). Homologous recombination (HR) is a less-error prone pathway that uses a sister chromatid or a homologous sequence as a template to restore broken DNA

(Haber, 2016).

HR that uses a sister chromatid to repair a DSB generally has no genetic consequence, while DSB repair that uses a homologous sequence typically results in the loss of heterozygosity. One type of HR is break-induced replication (BIR), which occurs when only one end of the DSB shares homology with a template. In this context strand invasion results in a unidirectional replication fork that results in a nonreciprocal translocation of the chromosome

141 end (Malkova et al., 1996; Davis and Symington, 2004). Gene conversion (GC) is a more common mechanism of HR that involves the nonreciprocal transfer of genetic information between homologous sequences when homology is shared between a donor and both ends of the DSB. There are two generally accepted models of gene conversion: synthesis-dependent strand-annealing (SDSA) and the double Holliday junction (dHJ). SDSA involves a migrating replication bubble that, after synthesis across the homologous region, results in the pairing of the newly copied strand with the resected 3’ ssDNA on the other side of the DSB, resulting in a non-crossover (NCO) event (Nassif et al., 1994). The dHJ model involves a displacement loop (D- loop) that can be extended by new DNA synthesis from the 3’ end of the invading strand so that the ssDNA of the opposite DSB end can anneal, forming a dHJ intermediate structure (Resnick,

1976; Szostak et al., 1983). After forming, the dHJ can be resolved by one of several dHJ resolvases, yielding both crossover (CO) and NCO events. Alternatively, dHJs can be dissolved to yield exclusively NCO outcomes (Wu and Hickson, 2003). In budding yeast, 10%-20% of interchromosomal mitotic GC events are accompanied by a CO, whereas ectopic recombination with short homologous regions results in significantly fewer COs (Pâques and Haber 1999; Ira et al., 2004).

Both the SDSA and dHJ model share an intermediate step which is the formation of heteroduplex DNA (hDNA). Heteroduplex DNA must be repaired by components of the mismatch repair pathway (MMR) to either restore the original DNA sequence, or convert the broken DNA locus to duplicate the sequence found in the homologous donor region. Although less mechanistically understood than the E. coli MMR pathway, the eukaryotic process has conserved the same basic steps: 1) mismatch recognition by MutS homologs, 2) recruitment of

142 MutL homologs, 3) excision of the region containing the mispaired nucleotides, and 4) gap repair via new DNA synthesis (Harfe and Jinks-Robertson, 1999). However, how eukaryotes recognize the newly synthesized strand from the template strand is not understood (Lobner-

Olesen et al., 2005). Saccharomyces cerevisiae has two MutS homologs. Msh2-Msh6 is a heterodimer that recognizes base-base mismatches, as well as small indels of 1-2 nucleotides

(Miret et al., 1993; Acharya et al., 1996; Marsischky et al., 1999). Msh2-Msh3 is the second heterodimer that recognizes some base mismatches, specifically A:A, C:C, and T:G, but primarily recognizes small indels that cause loop structures to form (Habraken et al., 1996; Palombo et al., 1996; Srivatsan et al., 1996; Harrington and Kolodner, 2007). While Msh2 is the common subunit in these heterodimers, it does not directly interact with the mispaired bases and is believed to be involved in the recruitment of the MutL homologs, as its deletion completely ablates MMR (Reenan and Kolodner, 1992; Marsischky et al., 1996). Yeast’s primary MutL homologs involved in repair of heteroduplex DNA are the Mlh1 and Pms1, which form a heterodimer required for the majority of MMR (Prolla et al., 1994). Mlh1-Mlh3 and Mlh1-Mlh2 appear to play non-essential accessory roles in promoting mitotic MMR (Campbell et al., 2014).

As previously stated, it is not well understood how eukaryotic organisms differentiate between the newly synthesized and the template strand during MMR. There is evidence that

Msh6 and Msh3 bind to a processivity factor of polymerase 훿, PCNA (proliferating cell nuclear antigen), during replication (Prelich et al., 1987; Clark et al., 2000; Flores-Rozas et al., 2000).

Previous research has indicated that PCNA plays an important role in both BIR and in gene conversion at the MAT locus of Saccharomyces cerevisiae, however, its involvement in MMR

143 during DSB repair has not been analyzed in detail (Holmes and Haber, 1999; Lydeard et al.,

2010).

Although differentiation of parental vs. newly-synthesized strand is not well understood, studies have shown that approximately 80% of heteroduplex DNA (hetDNA) intermediates are efficiently repaired by MMR during GC (Hum and Jinks-Robertson, 2017). This is done in part by

DNA polymerase 훿, as altering the proofreading capacity of pol훿 significantly shortens the length of hetDNA (Guo et al., 2017). The length of the hetDNA tract formed during repair heavily influences the repair outcome, as hetDNA tracts associated with CO events are longer than those associated with NCOs (Mitchel et al., 2010; Guo et al., 2017). CO outcomes are also increased when MMR, specifically mismatch binding via Msh6, and presumably Msh3, is compromised (Hum and Jinks-Robertson, 2019). However, deletion of mlh1, which is plays a role in mismatch processing and not recognition, primarily affected the directionality of hetDNA in NCO events with no discernable difference in CO outcomes (Hum and Jinks-Robertson, 2019).

This is supported by previous data showing that mlh1Δ strains show a strong bias for unidirectional hetDNA tracts, suggesting that MLH1 primarily acts on hetDNA that is only transiently present during SDSA events, vs hetDNA formed during dHJ GC events (Hum and

Jinks-Robertson, 2017). Together these data suggest a model where the anti-recombination activity of MMR proteins is dependent on whether one or both ends of the broken chromosome are able to engage in the donor sequence. This model also suggests that shorter hetDNA tracts are more likely to escape the anti-recombination activity of MMR proteins (Hum and Jinks-Robertson, 2019)

144 The key step in most HR pathways is the search for homologous DNA via the Rad51

(RecA homolog) nucleoprotein filament. Each monomer of Rad51 binds three nucleotides of single stranded DNA, which then probes the genome for a region of homologous DNA (Chen et al., 2008). In vitro analysis has shown that Rad51 requires eight perfectly matched bases for strand exchange to proceed and for the recruitment factor to initiate new DNA synthesis

(Danilowicz et al., 2015; Lee et al., 2015). However, an in vivo study has shown that DSB repair via BIR can occur when there is a base-pair mismatch every six nucleotides along the Rad51 nucleoprotein filament – indicating that in cells only five consecutive nucleotides of homology is required for a template to be utilized for BIR repair when the entire substrate is 108 bp (Anand et al., 2017). Furthermore, this study showed that correction of the resulting heteroduplex during strand invasion was principally dependent on the proof-reading capacity of polymerase

훿 (Anand et al., 2017). Given that BIR primarily occurs when there is homology available on one side of the DSB, we decided to examine the effect that heterologous templates have during mitotic gene conversion which occurs when both ends of the DSB can initiate strand invasion and interact with a donor sequence. My data shown here indicate that there is an inherent asymmetry in the mechanism of MMR associated with DSB repair. Additionally, while GC- mediated DSB repair is similar to that previously shown in BIR, it is not identical.

145 Results

Previous research has shown that the Rad51 recombinase allows extensive heterologies

in a donor template during break-induced replication (Anand et al., 2017). Given that break-

induced replication is an atypical form of double strand break repair (DSB), we wanted to

analyze the effect that a heterologous donor template has during gene conversion (GC). To do

this we designed substrates that contained mismatches spaced every 9 base pairs (bp) on either

Figure 4.1 - Basic gene conversion set-up. A) Four different base strains were constructed. The recipient sequence with an HO recognition site was integrated into the CAN1 locus of chromosome 5. A donor sequence for repair was integrated approximately 300 bp upstream of CHD1 on chromosome 5. Donor sequences had 100% homology (yDG_53), a mismatch every 9 bp on the centromere-proximal side of the DSB (yDG_54), a mismatch every 9 bp on the telomere-proximal side of the DSB (yDG_55), or a mismatch spaced every 9 bp across the entire donor sequence (yDG_56). B) Viability determined by plating assay - CFU’s (colony forming units) on induction YEP-Galacose media divided by CFU’s on non-induction YEPD media. C) NAT- sensitive galactose-survivors divided by total number of galactose survivors. Significance determined by t-test. *p > 0.05, **p > 0.01, ***p > 0.001. Error bars refer to standard error of the mean.

146 the left arm of the DSB with perfect homology on the right arm, perfect homology on the left arm with a mismatch every 9 bp on the right arm, or a mismatch every 9 bp throughout the donor template (Figure 4.1A). As the removal of a nonhomologous tail has shown to be a confounding variable when measuring the success of DSB repair, our strains were designed so that following cleavage via the HO-endonuclease neither side of the DSB contains a tail that must be excised from the invading strand before copying the heterologous region (Fishman-

Lobell & Haber, 1992; Colaiacovo et al., 1999; Studamire et al., 1999; Anand et al., 2017).

Retention of a nourseothricin (NAT) drug resistance gene, located immediately adjacent to the

3’ end of the GC tract, indicated repair of the DSB via either GC or nonhomologous end-joining

(NHEJ), while sensitivity to NAT following DSB repair indicated either BIR or SDSA where improper second-end capture has disrupted expression of the NAT resistance cassette. GC events can be confirmed by PCR amplification of the locus containing the HO cut site and subsequent digest with HindIII, as the donor sequence contains a restriction site that disrupts the HO recognition sequence.

The GC tract in this system is fairly short, only 168 bp of homology on either side of the

DSB making the total donor tract 336 bp, but we still see 24% viability following DSB repair

(Figure 4.1B). However, there is a significant drop in viability when mismatches are present on the left arm (13%), on the right arm (16%), and an even more drastic drop when mismatches are located on both arms of the GC tract (7%). (Figure 4.1B). In our strain that contains a donor with 100% homology, only 0.5% of the survivors lost resistance to NAT, and there was no significant increase when mismatches were located on the left arm of the GC tract (Figure

4.1C). However, there is a significant increase in NAT sensitive DSB-survivors when mismatches

147 are located on the right arm of the GC tract or on both arms of the tract, with 3.8% and 4.8% of

survivors gaining NAT sensitivity respectively. This result suggests that heterologies located on

the right arm of the GC tract have a negative effect on second end capture, pushing the cell to

repair via break-induced replication, or causing extension of the D-loop beyond the region of

homology. However, we believe that these events are failed SDSA events that use

microhomologies to anneal to the opposite end of the DSB, disrupting expression of the NAT

cassette (discussed below).

Although it is generally thought that SDSA is the predominant mechanism of gene

conversion, there are significant data indicating that heteroduplex DNA has a strong effect on

CO repair outcomes (Pâques and Haber

1999; Ira et al., 2006; Guo et al., 2017;

Hum and Jinks-Robertson, 2019). For

this reason, we decided to screen

survivors for crossovers via 2 different

PCR reactions (Figure 4.2A). For each Figure 4.2 - Mismatches and MMR machinery strain we screened 120 DSB-survivors. influence GC repair outcome. A) Crossover events were determined by two PCR reaction. One reaction We observed that repair in our WT included a primer upstream of the donor integration site in the GCG1 locus and a reverse primer internal to strain with 100% homology resulted in the NAT resistance cassette (shown in purple arrows). The second confirmation reaction used primers 17.5% rate of COs (Figure 4.2B). There internal to the NPR2 locus upstream of the recipient integration site with a reverse primer downstream of was no statistically significant difference the donor in the CHD1 locus. B) 120 successful GC events were PCR amplified using both primer sets in between this strain and the strain with 4.2A. C) Crossover events in a 100% homology donor are increased in an msh2Δ null strain. Significance mismatches on the left arm (24.2%), but determined by t-test. **p > 0.01.

148 there is a significant decrease in crossovers when mismatches are located on the right arm of the GC tract (4.2%). However, when there were mismatches located on both sides of the DSB only 10% of survivors repaired via CO, which is not statistically significant from the 100% homology donor (17.5%) via chi square analysis (Figure 4.2B). We then asked how deleting

MSH2 would affect repair and mismatch assimilation. Interestingly, when Msh2 is deleted, we see that DSB repair by COs increases significantly to 34.2% in our 100% homology donor strain, similar to previous reports (Fleck et al., 1999; Harrington and Kolodner, 2007; Goellner et al.,

2014) (Figure 4.2C). While previous research indicates that mutations in MSH2 promote

crossover events between divergent

substrates that retain 91% homology and

not strains with higher rates of

heterologies (Datta et al., 1996), these

data indicate the MMR pathway also

prevents recombination between donors

with limited regions of homology and that

MSH2 is heavily influenced by mismatch

density.

We next decided to test how

Figure 4.3 - Effect of MMR and DNA repair different DNA repair and MMR mutants mutants on viability. A) Viability determined by CFU on galactose-containing induction media affected the viability of these strains. We divided by growth on non-induction media. B) Loss of NAT resistance determined by replica tested genes known to play a role in plating galactose-survivor’s to plates containing NAT. Error bars refer to standard error of the MMR, polymerases, resolvases, and mean.

149 components of nucleotide excision repair (Figure 4.3A). None of these mutants, other than pol32Δ, a subunit of DNA polymerase 훿 that is known to aid in its processivity during BIR

(Lydeard et al., 2007; Jain et al., 2009), consistently caused a significant decrease in cell survival following a DSB across all donor backgrounds. However, it is worth noting that mus81Δ in strains with donors containing mismatches on one or the other side of the DSB show a significant decrease in viability when compared to WT strains with the same donor sequence.

The same does not hold true for strains with a perfect homology donor and a donor that contains mismatches across the whole gene conversion tract. Loss of the NAT resistance marker followed the same trend as previous observed, with mismatches on the right arm and on both arms consistently resulting in increased drug sensitivity (Figure 4.3B). Previously, Anand et al. showed that most of the assimilation of mismatches into BIR products required the proofreading of DNA polymerase 훿 which apparently resects the 3’ end, removing the heterologies and then copying from the donor template (Anand et al., 2017). However, there is a significant increase in NAT sensitivity among survivors with a proofreading-defective, 3’-5’ exonuclease dead allele DNA pol 훿 strains when compared against WT strains (Figure 4.3B). This could be a result of the decreased processivity of the polymerase.

Sequencing of individual survivors showed that there is an inherent asymmetry in the incorporation of mismatches following DSB repair. We sequenced between 30 and 40 individual colonies following induction of the HO-endonuclease in strains that contained mismatches on one or the other side of the DSB. These results are grouped and shown side by side for clarity in comparison (Figure 4.4A and 4.4B). These data consistently show us that there is an inherent asymmetry in DSB repair, even in WT strains. The first mismatch on the left arm of the GC tract

150 – located 9 bp upstream of the DSB - is incorporated into the recipient 98% of the time.

Meanwhile, the first mismatch on the right arm of the GC tract – located 9 bp downstream of

the DSB – is only incorporated into the recipient locus 57% of the time (Figure 4.4A and 4.4B).

This pattern is preserved in the strain that contains mismatches spaced every 9 bp on both

arms of the GC tract (Figure 4.4C).

Figure 4.4 - Sequencing shows different mechanisms for mismatch correction on the left and right sides of the DSB. A, B) Recipient locus of galactose-induced, NAT-resistant colonies were PCR amplified and individually sequenced. 30-44 individual sequences from strains with mismatches locatedindividua on either the left or right side of the DSB were analyzed separately and grouped together for clarity in comparing data. Uncorrected mismatches (evidenced by heteroduplex formation and sectored colonies) are shown as incorporated. C) Sequencing of recipient locus in galactose-induced, NAT-resistant colonies when mismatches are located on both sides of the DSB. Uncorrected mismatches (evidenced by heteroduplex formation and sectored colonies are shown as incorporated D) Sequencing of msh2Δ (top panel) and msh6Δ (bottom panel) showed primarily heteroduplex GC tracts. Each line represents the sequence of a single DSB-survivor colony. Strains with mismatches located on either the left or the right arm are grouped together for comparison. Uncorrected heteroduplex tracts are shown in blue.

151 These sequencing data show that mismatches incorporated into the recipient were significantly reduced in a pol3-01 strain, however, only when the mismatch is located on the left arm of the DSB (Figure 4.4A). The opposite is observed in the GC tracts when mismatches were located on the right arm of the DSB, where a Wilcoxon signed-rank test shows a significant increase in mismatch incorporation templated on the right arm of the DSB (p < 0.0001) (Figure

4.4A). Previously published results show that during BIR, mismatches are never incorporated into the recipient locus when they are located further than 50-bp upstream of the DSB (Anand et al., 2017). Here, however, we see mismatches are assimilated all the way to the very ends of the GC tract on either side of the DSB - 165 bp away from the DSB in this system - are incorporated into the recipient. Similar to BIR however, we see that mismatches tend to be incorporated in a strongly polar fashion. For example, if mismatch 15 was incorporated into the recipient, mismatches 1-14 were almost always incorporated as well (Figure 4.4D).

We also sequenced strains deleted for msh6, which forms a heterodimer with Msh2 and is involved in mismatch detection (Harfe and Jinks-Robertson, 1999). Sequencing of msh6Δ GC events show that there is no significant effect, determined by Wilcoxon-rank sum test, in mismatch incorporation when mismatches are located on the left arm of the GC tract.

However, there is a significant increase in mismatch incorporation when templated on the right side of the DSB. This is in contradiction to sequencing of an msh2Δ strain, which still shows no significant difference from WT on mismatch incorporation on the left arm of the break, but a significantly lower rate of incorporation of mismatches templated on the right side of the DSB.

However, closer analysis of the sequencing data shows us extensive tracts of heteroduplex DNA on the right side of the DSB, evidenced by the formation of sectored colonies. (Figure 4.4D).

152 Symmetrical mismatches do not affect mismatch incorporation patterns

Since there are different mismatches located on each arm of the DSB, we decided to reconstruct the GC tracts so that the same mismatches on each arm were identical, i.e. if the mismatch in position 1 of the left arm is an A:G mismatch, the mismatch on position 1 of the

right arm will be a A:G

mismatch (Figure 4.5A and

4.5B). These strains also have a

slightly longer regions of

homology, totaling 360 bp and

20 mismatches spaced 9 bp

apart on each arm of the GC

tract. In these strains, we see Figure 4.5 - Strain reconstruction with symmetrical mismatches. A) Strains were reconstructed so that the 34.4% viability when there is a mismatch in each position (position number shown in red) was the same on both the right and left side of the DSB. B) donor with perfect homology. Four different base strains were constructed by integrating the recipient sequence with an HO recognition site and the We see a statistically significant donor sequences were integrated into the same position as in previous strains (Figure 4.1A). Donor sequences had drop in viability to 14.4% when 100% homology (yDG_126), a mismatch every 9 bp on the centromere-proximal side of the DSB (yDG_127), a mismatches are located on the mismatch every 9 bp on the telomere-proximal side of the DSB (yDG_128), or a mismatch spaced every 9 bp across the left arm of the GC tract, and an entire donor sequence (yDG_129). A kanamycin-resistance is approximately 2 kb downstream of GC recipient even more drastic decrease to sequence. C) Viability determined by plating assay - CFU’s (colony forming units) on induction YEP-Galacose media 8.9% survival when mismatches divided by CFU’s on non-induction YEPD media. C) KAN- resistant galactose-survivors divided by total number of are located on both arms of the galactose survivors. Significance determined by t-test. **p > 0.01, Error bars refer to standard error of the mean. GC tract (Figure 4.5C). Despite a

153 drop in viability to 23.1% viability when mismatches are located on the right arm of the GC tract, this is not statistically significantly different from the donor with perfect homology

(determined by t-test) (Figure 4.5C).

Similar to the previous strains used, these strains contain a kanamycin (KAN) drug- resistance gene to identify cells that have repaired the DSB via either GC or NHEJ, however, this marker is located almost 2 kb downstream of the GC tract (Figure 4.5B). Therefore, survivors that lose resistance to KAN must have repaired via BIR. Indeed, we see that the vast majority of cells ( > 95%) retain KAN resistance in all donor conditions (Figure 4.5D). Comparing with our previous data (Figure 4.3B), these data most likely indicate that the increase in NAT sensitivity previously observed is most likely due to mismatches on the right arm obstructing proper second end capture following new strand synthesis (Figure 4.3B and 4.5D). Although these improper SDSA events are most likely mediated through a microhomology junction, further sequencing is required to confirm this hypothesis.

Again, we wanted to see how mismatches in the donor sequence affected viability in different mutant backgrounds. We examined polymerase mutants, resolvases, and the MutL homology Mlh1. Additionally, we looked at an rdh54/tid1 deletion, as it has recently been implicated in maintaining D-loops (Piazza et al., 2019). Rdh54 is a member of the SWI2/SNF2 family of helicase-like chromatin-remodelers and is conserved in eukaryotes (Flaus and Owen-

Hughes, 2011). In Saccharomyces cerevisiae it is well known for its interactions with Dmc1 during meiotic recombination, but its role in somatic recombination is less-well understood

(Lee et al., 2001; Nimonkar et al., 2007; Wright and Heyer, 2014). Indeed, we see that even with our perfect homology there is a significant decrease in viability compared to the WT strain

154 (Figure 4.4A). When mismatches are added to either arm of the GC tract, DSB survival drops to

1%-3% of cells plated. The only other mutants where we consistently see a decrease in cell

viability from the WT is again a pol32Δ mutant, which is known to reduce the processivity of

polymerase 훿.

We next wanted to analyze the mismatch incorporation pattern in some of these strains

to see how they differed from the asymmetrical GC tract previously analyzed. Analysis of the

recipient GC tract in individual colonies shows us the same asymmetry we previously observed

(Figure 4.6B). In a wildtype

strain, mismatches located on

the left arm of the DSB are more

likely to be incorporated into

the recipient locus than those

same mismatches located on

the right arm of the GC tract (p

value < 0.0001, Wilcoxon rank

test).

Figure 4.6 - Effect of DNA repair mutants on viability and Additionally, we again see mismatch incorporation A) Viability determined by CFU on galactose-containing induction media divided by growth on that mismatch incorporation on non-induction media. B) Recipient locus of galactose- induced, KAN-resistant colonies were PCR amplified and the left arm is primarily due to individually sequenced. 30-44 individual sequences from strains with mismatches located on either the left or right proofreading capacity of side of the DSB were analyzed separately and grouped together for clarity in comparing data. Uncorrected polymerase 훿, as when its mismatches (evidenced by heteroduplex formation and sectored colonies) are shown as incorporated. Error bars exonuclease capabilities are refer to standard error of the mean.

155 removed there is a significant reduction in mismatch incorporation. However, the opposite is true when mismatches are located on the right arm of the DSB, as we see an increase in mismatch incorporation when the same pol3-01 allele is used. This pattern is identical to that previously observed (Figure 4.4A). Moreover, pol32Δ shows a significant difference in mismatch incorporation on opposite sides of the GC tract. On the right arm, deletion of Pol32 results in a drastic decrease in mismatch incorporation, as mismatches 18 bp from the DSB are never incorporated into the recipient locus. This is in contrast to the left arm of the tract, where mismatches located 120 bp away from the DSB are incorporated into the recipient. Another key difference between the two arms of the GC tract is the role of Rdh54. Although rdh54Δ causes a significant reduction in viability in either background, survivors with heterologies on the left arm of the GC tract show a significant reduction in mismatch incorporation, while survivor’s with heterologies in the right arm of the GC tract show a significant increase in mismatch incorporation. Unlike the other mutants examined here, a proof-reading defective allele of polymerase 휀 (pol2-4) shows a significant reduction in mismatch incorporation when mismatches are located on either arm of the DSB. DNA polymerase 휀 has not previously been implicated in a prominent role during DSB repair or MMR.

Previous research has shown that the presence of a nonhomologous tail causes a significant reduction in BIR (Anand et al., 2017). To observe the effect of a nonhomologous tail in GC, we designed new donor tracts that contained perfect homology, but a 9-nt tail on either the left arm of the GC tract, the right arm of the GC tract, or a tail on both arms of the GC tract.

For comparison, we also constructed donor templates that had mismatches spaced every 9-bp on both arms of the GC tract with a 9-nt tail on either the left arm, right arm or both arms

156 Figure 4.7 - Presence of a nonhomologous tail has different effects on the left and right sides of the DSB A) Strains were constructed so that the a 9-nt nonhomologous tail was located on either the right, left or both sides of the DSB. The donor in these strains had either 100% homology to the recipient locus or a mismatch spaced every 9- bp across the donor sequence. B) Viability was determined by plating assay - CFU’s (colony forming units) on induction YEP-Galacose media divided by CFU’s on non-induction YEPD media. C) Recipient locus of galactose-induced, KAN- resistant colonies were PCR amplified and individually sequenced. 30-44 individual sequences from strains with mismatches located on either the left or right side of the DSB were analyzed separately and grouped together for clarity in comparing data. Uncorrected mismatches (evidenced by heteroduplex formation and sectored colonies) are shown as incorporated. *p > 0.05, Error bars refer to standard error of the mean. (Figure 4.7A). In contrast to BIR, we see a significant increase when a tail is located on the left arm of the DSB when there is perfect homology (Figure 4.7B). However, we see that a nonhomologous tail has no effect on viability when located on right arm or when present on both arms when there is perfect homology. Yet, when a nonhomologous tails is added to strains that have mismatches spaced every 9-bp across the donor sequence, there is a significant decrease in viability when compared to a heterologous donor with no mismatches (Figure

4.7B). Sequencing of the DSB survivors of these strains show us that the presence of a tail had differing effects on mismatch incorporation. When the tail is located on either the left side or the right side, we see that that same arm has a higher rate of mismatch incorporation (Figure

4.7C). However, when there is a tail that must be excised from both arms of the GC tract, we see that the rate of mismatch incorporation on the left increases, while the rate of mismatch

157 incorporation on the right arm of the DSB decreases. This provides further evidence that the presence of a nonhomologous tail recruits components of the MMR pathway, as well as the inherent asymmetry of DSB repair (Figure 4.7C). The asymmetry present in mismatch incorporation when tails are present on both sides of the DSB could be due to our previous observations that MMR plays a significant role in incorporating mismatches templated on the right side of the DSB compared to the left side.

Pattern of mismatch incorporation is influenced by endonuclease cleavage

One possibility to explain the asymmetry that we observe in mismatch incorporation is due to bias of the DNA cleavage itself. It is possible that after cleavage the rate of resection differs on one side of the DSB compared to the other. To examine this possibility, we remade the strains with the same set-up previously used (Figure 4.5B and 4.8A), except we reversed the

24-bp HO recognition sequence within the GC recipient. Interestingly, in these strains we no longer see the same asymmetry observed in mismatch incorporation at the recipient locus

(Figure 4.8B). Cleavage at the WT HO cut site or the reverse HO cut site will still result in 4-nt 3’ overhangs, but it possible that the HO endonuclease sits at the site of the DSB following cleavage for a unknown period of time (Nickloff et al., 1986).

To further investigate the influence of the endonuclease on the mismatch incorporation pattern of GC events, we turned to the programmable Cas9 endonuclease (Mali et al., 2013;

DiCarlo et al., 2013). We modified the GC recipient locus to create two PAM sites so that a gRNA designed to target the Watson or the Crick strand would result in cleavage in the same position (Figure 4.8C). Unlike HO which results in 4-nt 3’ overhangs, Cas9 cleavage results in

158 blunt ends or 1-nt overhangs

after cleavage (Nickloff et al.,

1986; Lemos et al., 2018).

However, it is known that Cas9

remains bound to DNA

following cleavage, although

the length of time it remains

bound is still heavily debated

(Sternberg et al., 2014; Shibata

et al., 2017). These strains

show a similar viability to our

strains with an HO cut site, Figure 4.8 - Pattern of mismatch incorporation is heavily influenced by DSB A) The strains in Figure 4.5B were however sequencing of GC reconstructed with a reverse HO cut site. B,D) Recipient locus of galactose-induced, KAN-resistant colonies were PCR survivor’s following Cas9 amplified and individually sequenced. 30-44 individual sequences from strains with mismatches located on either cleavage shows contrasting the left or right side of the DSB were analyzed separately and grouped together for clarity in comparing data. Uncorrected results with those following an mismatches (evidenced by heteroduplex formation and sectored colonies) are shown as incorporated. Error bars HO-induced DSB. We now see refer to standard error of the mean. C) The strains in Figure 4.5B were reconstructed with two PAM sites so that a gRNA that mismatches templated on targeting either the Watson or the Crick strand results in cleavage at the same position. the right arm of the DSB are significantly more likely to be incorporated into the recipient locus than those templated on the left arm of the DSB when using a gRNA targeting the Watson strand (Figure 4.8D). However, when using a gRNA to target the Crick strand, we again see that there is no significant

159 difference in mismatch incorporation between the two sides of the DSB, comparable to the patterns observed when we reversed the HO cut site (Figure 4.8D). Since Cas9 remains bound 5’ of the PAM site following DNA cleavage, the opposing mismatch incorporation patterns that we observe could be explained by the HO-endonuclease binding 3’ of the cleavage site. However, it is curious that a Cas9 with a gRNA targeting the Crick strand and reversing the HO cut site eliminate the asymmetry in mismatch correction. This could be due to the rate of resection being influenced by the DSB itself, but remains a crucial point for further study.

Discussion

Previous research on the asymmetry of DSB repair has focused on how it influences the outcome of repair: CO vs NCO. These data provide additional experimental evidence of asymmetry in DSB repair, as well as highlight the different mechanisms involved in MMR on one side of the DSB or the other. In both of our GC systems, sequencing individual DSB survivors shows that mismatches located the same distance from the DSB are more likely to be incorporated when they are on the left arm of the GC tract compared to the right arm of the GC tract. Furthermore, the mechanisms in which these mismatches are corrected and fixed into the recipient locus occur through two independent methods. Similar to what has previously been observed in BIR, we see that the mismatches located on the left, centromere-proximal side of the DSB are typically corrected in favor of the donor locus via the exonuclease capabilities of polymerase 훿. However, mismatch correction on the right, telomere-proximal side of the DSB break is dependent on the traditional MMR machinery. However, when mispaired bases are located on the right arm of the GC tract, we see an increase when the same

160 proofreading defective allele of polymerase 훿. Furthermore, we see that mismatch incorporation on the right side of the DSB is significantly reduced in an msh2Δ, while there is no significant effect on the left side of the break (Figure 4.4B). These data contradict sequencing results in an msh6Δ strain, where we see that 50% of survivors with mismatches templated on the right side of the break result in heteroduplex DNA that is left unresolved. This could be due to an inability to recruit Mlh1 which is involved in the processing of DNA mispairs. Msh2 forms multiple heterodimers, so it is also possible that Msh2-Msh3 also plays a role in correcting mismatches in this GC tract and is potentially the cause of the different phenotypes of msh2Δ and msh6Δ in this assay (Lamers et al., 2000; Obmolova et al., 2000). It has also been shown that Msh6 directly interacts with PCNA to facilitate MMR, and blocking this interacts results in uncorrected hetDNA (Clark et al., 2000; Kleczkowska et al., 2001; Kadyrov et al., 2007). It is therefore possible that ablating this interaction in an msh6Δ is enough to reduce mismatch correction, resulting in primarily uncorrected heteroduplex DNA, while an msh2Δ strain completely abolishes MMR altogether. Another possibility involves the role of the Sgs1/BLM helicase, which has been shown to restrict HR between divergent sequences, and potentially acts preferentially on one side of the DSB versus the other (Spell and Jinks-Robertson, 2004;

Sugawara et al., 2004). The role of Sgs1, as well as that of the Mph1 helicase, which has been shown to heavily influence GC repair outcomes, in these strains remains an important, yet unexplored area (Mitchel et al., 2013; Piazza et al., 2019).

Furthermore, the DNA-dependent ATPase and DNA translocase Rdh54/Tid1, while causing a significant decrease in DSB survivability in all strains containing heterologous donors, shows significant asymmetry in its role on either side of the DSB. An rdh54Δ strain causes a

161 significant restoration of hetDNA to favor the original strand vs the strand newly copied strand from the heterologous donor on the left side of the DSB, while having the opposite effect on sequences located on the right side of the DSB. Rdh54 has been implicated in maintaining D- loops, but it has not been implicated in MMR before. This result most likely reflects a property of the structures formed during gene conversion. Together, these data suggest that the left arm of the GC tract is more likely to initiate strand invasion. Mispaired bases would have to be corrected during this initial invasion step where a D-loop is formed. DNA synthesis would occur across the junction and the right arm of the GC tract, which – in the SDSA model – would eventually anneal to the opposite DSB arm. Only at this point would mismatched bases located on the right arm be recognized and processed by the MMR machinery. This aligns with the differences we see between strains concerning the retention of the drug resistance marker downstream of the GC tract. In the strains utilized in Figures 4.1-4.4, the NAT resistance marker is immediately adjacent to the right arm of the GC tract, whereas the strains used in figures 4.5-

4.8 have a KAN resistance marker located further downstream of the GC tract. We see that in strains containing the NAT marker are more likely to gain drug sensitivity when mismatches are located on the right arm of the GC tract, most likely indicating an issue with second end capture following new strand synthesis. Loss of KAN resistance is not observed when the resistance marker is moved further downstream of the GC tract.

Additionally, we see that there is a significant drop in CO events when mismatches are located on the right arm of the DSB compared to strains with perfect homology donors or mismatches located on the left arm of the GC tract. This observation is not easily explained, however many studies out of Sue Jinks-Robertson’s and Frank Stahl’s labs have implicated the

162 MMR has a role determining a CO or an NCO outcome. This could be explained by a bias in the roles of Sgs1 or Mph1 on the left or the right side of the DSB, given that deletion of either of these helicase’s have previously been shown to cause distinctly different CO vs NCO outcome ratios (Mitchel et al., 2013).

Together these data show an inherent asymmetry in the mechanism of DSB repair that has not previously been considered.

Methods

Parental Strain. yRA111 (MATa::DEL.HOcs::hisG ura3D851 trp1DEL.63 leu2DEL.::KAN hmlDEL.::hisG HMR::ADE3 ade3::GAL::HO can1DEL::UR intron_SD :: HOcs::NAT) was used as the parental strain in these experiments (Anand et al., 2017). This strain lacks the HML, HMR, and the MAT locus so that all homology to the HO cut site was removed from the strain. The 336 bp region upstream of the NAT promoter was used as the recipient sequence. Donor sequences were integrated approximately 200 bp upstream of chd1. ORFs were deleted by replacing the target gene with a prototrophic or an antibiotic-resistance marker via the high-efficiency transformation procedure of S. cerevisiae with PCR fragments (Rothstein, 1983). A list of all strains used is provided in Table 4.1. Point mutations in POL2 and POL3 were made via Cas9 and an ssODN template. gRNAs were ligated into a BplI digested site in a backbone that contains a constitutively active Cas9 and either an HPH or LEU2-marker (bRA89 and bRA90, respectively)

(Anand et al., 2017b). Plasmids were verified by sequencing (GENEWIZ) and transformed as previously described (Anand et al., 2017b). Plasmids are listed in table 4.2

163 GC Viability Assay Using HO endonuclease. Strains were grown overnight in selective media and then counted and diluted to plate approximately 200 cells onto YEPD and YEP-Gal plates.

YEPD plates were grown at 30oC for 2 days, and YEP-Gal plates were incubated at 30oC for 3 days. After counting CFUs, YEP-Gal plates were replica plated to either NAT (Figures 4.1 - 4.4) or

KAN (Figures 4.5 - 4.8) to isolate GC and NHEJ events. GC events were confirmed by PCR amplifying recipient locus and digestion with HindIII restriction enzyme.

DNA Sequence Analysis. Using primers flanking the recipient locus were used to PCR amplify

DNA from surviving colonies. PCR products were purified and Sanger-sequenced by GENEWIZ.

The sequences were analyzed using Geneious software.

Statistical Analysis. All statistical analysis was performed using GraphPad Prism 8 software.

164 Table 4.1. Strains used in these experiments

Strain Genotype Notes

DG_53 yRA111 with DG_59 integrated upstream of chd1

DG_54 yRA111 with DG_60 integrated upstream of chd1

DG_55 yRA111 with DG_61 integrated upstream of chd1

DG_56 yRA111 with DG_62 integrated upstream of chd1

DG_57 DG_53; msh2::HPH

DG_58 DG_54; msh2::HPH

DG_59 DG_55; msh2::HPH

DG_60 DG_56; msh2::HPH

DG_61 DG_53; msh6::HPH

DG_62 DG_54; msh6::HPH

DG_63 DG_55; msh6::HPH

DG_64 DG_56; msh6::HPH

DG_65 DG_53; mlh1::HPH

DG_66 DG_54; mlh1::HPH

DG_67 DG_55; mlh1::HPH

DG_68 DG_56 mlh1::HPH

DG_69 DG_53; pol2-4

DG_70 DG_54; pol2-4

DG_71 DG_55; pol2-4

165

DG_72 DG_56; pol2-4

DG_73 DG_53; pol3-01

DG_74 DG_54; pol3-01

DG_75 DG_55; pol3-01

DG_76 DG_56; pol3-01 ho MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO Parental strain is DG_126 DG_228 integrated into can1; DG_229 integrated 200 JKM170 bp upstream of chd1 ho MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO Parental strain is DG_127 DG_228 integrated into can1; DG_230 integrated 200 JKM170 bp upstream of chd1 ho MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO Parental strain is DG_128 DG_228 integrated into can1; DG_231 integrated 200 JKM170 bp upstream of chd1 ho MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO Parental strain is DG_129 DG_228 integrated into can1; DG_232 integrated 200 JKM170 bp upstream of chd1

DG_130 yDG_126; pol2-4

DG_131 yDG_127; pol2-4

DG_132 yDG_128; pol2-4

DG_133 yDG_129; pol2-4

DG_134 yDG_126; pol3-01

DG_135 yDG_127; pol3-01

DG_136 yDG_128; pol3-01

DG_137 yDG_129; pol3-01

166

DG_138 yDG_126; pol32Δ

DG_139 yDG_127; pol32Δ

DG_140 yDG_128; pol32Δ

DG_141 yDG_129; pol32Δ

DG_142 yDG_126; rdh54Δ

DG_143 yDG_127; rdh54Δ

DG_144 yDG_128; rdh54Δ

DG_145 yDG_129; rdh54Δ

DG_146 yDG_126; mlh1Δ

DG_147 yDG_127; mlh1Δ

DG_148 yDG_128; mlh1Δ

DG_149 yDG_129; mlh1Δ

DG_150 yDG_126; mus81Δ

DG_151 yDG_127; mus81Δ

DG_152 yDG_128; mus81Δ

DG_153 yDG_129; mus81Δ

DG_154 yDG_126; yen1Δ

DG_155 yDG_127; yen1Δ

DG_156 yDG_128; yen1Δ

DG_157 yDG_129; yen1Δ hoΔ MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- 100% homology donor; DG_160 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO 9-nt tail on left arm

167 DG_228 integrated into can1; DG_328 integrated 200 bp upstream of chd1 hoΔ MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO 100% homology donor; DG_161 DG_228 integrated into can1; DG_329 integrated 200 9-nt tail on right arm bp upstream of chd1 hoΔ MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO 100% homology donor; DG_162 DG_228 integrated into can1; DG_330 integrated 200 9-nt tail on both arms bp upstream of chd1 hoΔ MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- Mismatch spaced every 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO 9-nt along donor DNA DG_163 DG_228 integrated into can1; DG_331 integrated 200 sequence; 9-nt tail on bp upstream of chd1 left arm hoΔ MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- Mismatch spaced every 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO 9-nt along donor DNA DG_164 DG_228 integrated into can1; DG_332 integrated 200 sequence; 9-nt tail on bp upstream of chd1 right arm hoΔ MAT::DEL hmlΔ::ADE1 hmrΔ::ADE1 ade1-100 leu2- Mismatch spaced every 3,112 lys5 trp1::hisG′ ura3-52 ade3::GAL::HO 9-nt along donor DNA DG_165 DG_228 integrated into can1; DG_333 integrated 200 sequence; 9-nt tail on bp upstream of chd1 both arms

Table 4.2. Plasmids used in these experiments

Plasmid Description pRA_114 Cas9 vector used to make pol3-01 (guide TCCTTTGATATCGAGTGT GC) pRA_124 Cas9 vector used to make pol2-4 allele (guide TATCAAATGCCATTAC CACA)

168 References

Acharya, S., Wilson, T., Gradia, S., Kane, M.F., Guerrette, S., Marsischky, G.T., Kolodner, R., and Fishel. R. (1996). hMSH2 forms specific mispair-binding complexes with hMSH3 and hMSH6. Proc Natl Acad Sci USA 93, 13629–13634.

Anand, R., Beach, A., Li, K., and Haber, J. (2017). Rad51-mediated double-strand break repair and mismatch correction of divergent substrates. Nature 544, 377-380.

Anand, R., Memisoglu, G., and Haber, J.E. (2017). Cas9-mediated gene editing in Saccharomyces cerevisiae. Protoc Exch.

Colaiacovo, M.P., Pâques, F., and Haber, J.E. (1999). Removal of one nonhomologous DNA end during gene conversion by a RAD1- and MSH2-indepenent pathway. Genetics 151, 1409-1423.

Danilowicz, C., Yang, D., Kelley, C., Prévost, C., and Prentiss, M. (2015). The poor homology stringency in the heteroduplex allows strand exchange to incorporate desirable mismatches without sacrificing recognition in vivo. Nucleic Acids Res 43, 6473-6485.

Davis, A.P., and Symington, L.S. (2004). RAD51-dependent break- induced replication in yeast. Mol Cell Biol 24, 2344 – 2351.

DiCarlo, J.E., Norville, J.E., Mali, P., Rios, X., Aach, J., and Church, G.M. (2013). Genome engineering in Saccharomyces cerevisiae using CRISPR-Cas systems. Nucleic Acids Res 41, 4336- 4343.

Fishman-Lobell, J., and Haber, J.E. (1992). Removal of nonhomologous DNA ends in double- strand break recombination: the role of the yeast ultraviolet repair gene RAD1. Science 258(5081), 480–484.

Flaus, A., and Owen-Hughes, T. (2011). Mechanisms for ATP-dependent chromatin remodelling: the means to the end. FEBS J 278(19), 3579-3595.

169 Fleck, O., Lehmann, E., Schar, P., and Kohli, J. (1999). Involvement of nucleotide-excision repair in msh2 pms1-independent mismatch repair. Nat Genet 21, 314-317.

Flores-Rozas, H., Clark, D., and Kolodner, R.D. (2000). Proliferating cell nuclear antigen and Msh2p–Msh6p interact to form an active mispair recognition complex. Nat Genet 26, 375–378.

Goellner, E.M., Smith, C.E., Campbell, C.S., Hombauer, H., Desai, A., Putnam, C.D., and Kolodner, R.D. (2014). PCNA and Msh2–Msh6 activate an Mlh1–Pms1 endonuclease pathway required for Exo1-independent mismatch repair. Mol Cell 55, 291–304.

Guo, X., Hum, L.F., Lehner, K, and Jinks-Robertson, S. (2017). Regulation of hetDNA length during mitotic double-strand break repair in yeast. Mol Cell 67, 539-549.

Haber, J.E. (2016). A Life Investigating Pathways That Repair Broken Chromosomes. Annu Rev Genet 50, 1-28

Habraken, Y., Sung, P., Prakash, L., and Prakash, S. (1996). Binding of insertion/deletion DNA mismatches by the heterodimer of yeast mismatch repair proteins MSH2 and MSH3. Curr Biol 6, 1185–1187.

Harfe, B.D., and Jinks-Robertson, S. (1999). Removal of frameshift intermediates by mismatch repair proteins in Saccharomyces cerevisiae. Mol Cell Biol 19, 4766-4773.

Harrington, J.M., and Kolodner, R.D. (2007). Saccharomyces cerevisiae Msh2–Msh3 acts in repair of base–base mispairs. Mol Cell Biol 27, 6546–6554.

Harrington, J.M., and Kolodner, R.D.(2007). Saccharomyces cerevisiae Msh2–Msh3 acts in repair of base–base mispairs. Mol Cell Biol 27, 6546–6554.

Hoeijmakers, J.H.J. (2009). DNA damage, aging, and cancer. N Engl J Med 361, 1475–1485.

Holmes, A.M., and Haber, J.E. (1999). Double-strand break repair in yeast requires both leading and lagging strand DNA polymerases. Cell 96, 415-414.

Hum, Y.F., and Jinks-Robertson, S. (2017). Mitotic gene conversion tracts associated with repair of a defined double-strand break in Saccharomyces cerevisiae. Genetics 207, 115-128.

Ira, G., Satory, D., and Haber, J.E. (2006). Conservative inheritance of newly synthesized DNA in double-strand break-induced gene conversion. Mol Cell Biol 26, 9424–9429.

170 Jain, S., Sugawara, N., Lydeard, J., Vaze, M., Tanguy Le Gac, N., and Haber, J.E. (2009). A recombination execution checkpoint regulates the choice of homologous recombination pathway during DNA double-strand break repair. Genes Dev 23, 291–303.

Kleczkowska, H.E., Marra, G., Lettieri, T., and Jiricny, J. (2001). hMSH3 and hMSH6 interact with PCNA and colocalize with it to replication foci. Genes Dev 15, 724–736.

Lau, P.J., Flores-Rozas, H., and Kolodner, R.D. (2002). Isolation and characterization of new proliferating cell nuclear antigen (POL30) mutator mutants that are defective in DNA mismatch repair. Mol Cell Biol 22, 6669–6680.

Lee, S.E., Pellicioli, A., Malkova, A., Foiani, M., and Haber, J.E. (2001). The Saccharomyces recombination protein Tid1p is required for adaptation from G2/M arrest induced by a double- strand break. Current Biology 11(13), 1053-1057.

Lobner-Olesen, A., Skovgaard, O., and Marinus, M.G. (2005). Dam methylation: coordinating cellular processes. Curr Opin Microbiol 8, 154–160.

Lydeard, J.R., Jain, S., Yamaguchi, M., and Haber, J.E. (2007). Break- induced replication and telomerase-independent telomere maintenance require Pol32. Nature 448, 820–823.

Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E., Church, G.M. (2013). RNA-guided human genome engineering via Cas9. Science 339(6121), 823–826.

171 Malkova, A., Naylor, M.L., Yamaguchi, M., Ira, G., and Haber, J.E. (2005). RAD51-dependent break-induced replication differs in kinetics and checkpoint responses from RAD51- mediated gene conversion. Mol Cell Biol 25, 933–944.

Marsischky, G.T., Filosi, N., Kane, M.F., and Kolodner, R. (1996). Redundancy of Saccharomyces cerevisiae MSH3 and MSH6 in MSH2-dependent mismatch repair. Genes Dev 10, 407–420.

Miret, J.J., Milla, M.G., and Lahue, R.S. (1993). Characterization of a DNA mismatch-binding activity in yeast extracts. J Biol Chem 268, 3507–3513.

Mitchel, K., Lehner, K., and Jinks-Robertson, S. (2013). Heteroduplex DNA position defines the roles of Sgs1, Srs2, and Mph1 helicases in promoting distinct recombination outcomes. PLoS Genet 9, e1003340.

Mitchel, K., Zhang, H., Welz-Voegele, C., and Jinks-Robertson, S. (2010). Molecular structures of crossover and noncrossover intermediates during gap repair in yeast: implications for recombination. Mol Cell 38, 211-222.

Nassif, N., Penney, J., Pal., S., Engels, W.R., and Gloor, G.B. (1994). Efficient copying of nonhomologous sequences from ectopic sites via P-element-induced gap repair. Mol Cell Biol 14, 1613-1625.

Nickoloff, J.A., Chen, E.Y., and Heffron, F. (1986). A 24-base- pair DNA sequence from the MAT locus stimulates intergenic recombination in yeast. Proc Natl Acad Sci USA 83, 7831−7835.

Niedernhofer, L.J., Gurkar, A.U., Wang, Y., Vijg, J., Hoeijmakers, J.H.J., and Robbins, P.D. (2018). Nuclear genomic instability and aging. Annu Rev Biochem 87, 295-322.

Nimonkar, A.V., Amitani, I., Baskin, R. J., and Kowalczykowski, SC. (2007). Single-molecule imaging of Tid1/Rdh54, a Rad54 homolog that translocates on duplex DNA and can disrupt joint molecules. J Biol Chem 282, 30776-30784.

Obmolova, G., Ban, C., Hsieh, P., and Yang, W. (2000). Crystal structures of mismatch repair protein MutS and its complex with a substrate DNA. Nature 407, 703–710.

Palombo, F., Iaccarino, I., Nakajima, E., Ikejima, M., Shimada, T., and Jiricny, J. (1996). hMutSbeta, a heterodimer of hMSH2 and hMSH3, binds to insertion/deletion loops in DNA. Curr Biol 6, 1181–1184.

172 Pâques, F., and Haber, J.E. (1999). Multiple pathways of recombination induced by double- strand breaks in Saccharomyces cerevisiae. Microbiol Mol Biol Rev 63, 349-404.

Piazza, A., Shah, S.S., Wright, W.D., Gore, S.K., Koszul, R., and Heyer, W.D. (2019). Dynamic Processing of Displacement Loops during Recombinational DNA Repair. Mol Cell 73(6), 1255- 1266.e4.

Prolla, T.A., Pang, Q., Alani, E., Kolodner, R.D., and Liskay, R.M. (1994). MLH1, PMS1, and MSH2 interactions during the initiation of DNA mismatch repair in yeast. Science 265, 1091–1093.

Reenan, R,A., and Kolodner, R.D. (1992). Isolation and characterization of two Saccharomyces cerevisiae genes encoding homologs of the bacterial HexA and MutS mismatch repair proteins. Genetics 132, 963–973.

Resnick, M.A. (1976). The repair of double-strand breaks in DNA; a model involving recombination. J Theor Biol 59, 97–106.

Spell, R.M., and Jinks-Robertson, S. (2004). Examination of the roles of Sgs1 and Srs2 helicase in the enforcement of recombination fidelity in Saccharomyces cerevisiae. Genetics 168, 1855- 1865.

Srivatsan, A., Bowen, N., and Kolodner, R.D. (2014). Mispair-specific recruitment of the Mlh1– Pms1 complex identifies repair substrates of the Saccharomyces cerevisiae Msh2–Msh3 complex. J Biol Chem 289, 9352–9364.

Sternberg, S.H., Redding, S., Jinek, M., Greene, E.C., and Doudna, J.A. (2014). DNA interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature 507, 62-67.

Studamire, B., Price, G., Sugawara, N., Haber, J.E., and Alani, E. (1999). Separation-of-function mutations in Saccharomyces cerevisiae MSH2 that confer mismatch repair defects but do not affect nonhomologous-tail removal during recombination. Mol Cell Biol 19, 7558-7567.

173 Sugawara, N., Goldfarb, T., Studamire, B., Alani, E., and Haber, J.E. (2004). Heteroduplex rejection during single-strand annealing requires Sgs1 helicase and mismatch repair proteins Msh2 and Msh6 but not Pms1. Proc Natl Acad Sci USA 101, 9315-9320.

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

Szostak, J.W., Orr-Weaver, T.L., Rothstein, R.J., and Stahl, F.W. (1983). The double-strand-break repair model for recombination. Cell 33, 25–35.

Wright, W.D., and Heyer, W.D. (2014). Rad54 functions as a heteroduplex DNA pump modulated by its DNA substrates and Rad51 during D loop formation. Mol Cell 53(3), 420-432.

Wu, L., and Hickson, I.D. (2003). The Bloom’s syndrome helicase suppresses crossing over during homologous recombination. Nature 426, 870–874.

174 Chapter 5: Future Directions

Mechanisms of CRISPR-Cas9 mediated gene editing with ssDNA

Using CRISPR-Cas9 to engineer DNA has only occurred in the past seven years. During this short time, we have learned a great deal about how this system works and many ways to optimize its efficiency. In chapter 2, I share our published work on the mechanism of single- strand template repair (SSTR) following a DNA double strand break (DSB) with either the HO- endonuclease with a single-stranded oligonucleotide (ssODN) as a donor template, as well as with a Cas9-mediated DSB with a ssDNA donor generated in vivo by an optimized bacterial retron system.

There are several follow-up studies that would add to our understanding of the mechanism of SSTR. One important area that we did not study are the kinetics of SSTR. Since

SSTR is not affected by a mec1Δ or tel1Δ, it is most likely a very rapid process. Understanding the kinetics of SSTR would have a significant impact on the potential of using CRISPR in a therapeutic setting.

In our published study we also observed a unknown function of Rad59 in alleviating

Rad51’s inhibition of Rad52’s single strand annealing activity. We have since observed that this interaction is partially through a Rad51 homolog, the Rad55-Rad57 heterodimer, as a rad55Δrad59Δ double mutant partially rescues the rad59Δ phenotype. The next step would be to make double mutants of Rad59 and components of the SHU complex, another Rad51

175 homolog. If this mutant also shows partial rescue of the rad59Δ phenotype, the next question would be if a triple mutant of rad55Δ shu1Δ rad59Δ shows full rescue of the rad59Δ phenotype.

Additionally, the potential effect of transcription of Cas9-mediated gene editing is an important area to study. Preliminary data suggest that cells have a higher rate of editing efficiency if a region is transcriptionally active at the cleaved locus. Active transcription in a region would provide better access and binding of the ssDNA template to targeted region.

A genome-wide screen to find components of SSTR

In chapter three, I outlined a screen that we performed to find potential components of

SSTR. The top hits from this screen, outlined in Table 3.1, are primarily involved in nucleotide excision repair and telomere extension via recombination. These hits are currently being individually verified with both the retron system utilized in the screen, as well as with the HO- endonuclease and an 80-nt ssODN (assay described in Figure 2.1A and 2.1B).

Since many of these hits affect components of telomeric recombination, it raises the possibility that the cell views the DSB and ssODN as a possible telomere. It would be interesting to collect survivors of SSTR and do next-generation sequence to determine if telomeric sequences have been inserted at the site of the DSB. Since the readout of our retron assay is a lysine prototrophic event, and that of our HO-ssODN assay is incorporation of a XhoI restriction site, insertion of telomeric sequences at the site of the DSB would be overlooked in our assays.

176 The asymmetry of mismatch correction of divergent substrates

The most likely explanation for the asymmetry of DSB repair is that one strand of broken chromosome is more likely to initiate strand invasion into a homologous region than the other.

In order to test this hypothesis, a time course would need to be done using these mismatched strains to determine if mismatches on the left arm or on the right arm of the GC tract caused a delay in DSB repair. It is also possible that this would not be able to be determined via a time course, if the difference in completion of the GC event was not significant enough to be determined by qPCR or Southern blot. If this shows to be true, it is possible that using donors that contained a higher density of mismatches could potentially show us that mismatches spaced closer together on the left side of the DSB could not undergo GC, while donors that had the same mismatches on the right arm still could.

Another interesting possibility that these data present is the role of the Sgs1 and Mph1 helicases in short-tract gene conversion. The role of these helicases could provide an explanation for the different mechanisms of mismatch correction that we observe in hetDNA tracts on the opposite sides of the DSB. These mutants could also provide insight into the apparent failed SDSA events observed in our strains with the NAT-resistance cassette located immediately adjacent to the GC recipient tract (Figure 4.1 and 4.3). It is also important to collect NAT-sensitive mutants following DSB repair and sequence the junctions to determine where second-end capture happens when mismatches are located on the right arm of the DSB.

Microhomology-mediated jumps like these have not been extensively characterized in the context of gene conversion.

177 Finally, we need to compare the mismatch correction patterns of an msh6Δ strain to a msh3Δ to confirm the role of MMR in mismatch correction on the right side the break. Although

Msh2-Msh3 is thought to primarily recognize small loop structures, it also recognizes some base-base mispairings and could potentially explain the difference observed in our msh2Δ and msh6Δ strains.

178