Formation of Dicentric and Acentric Chromosomes, by a Template Switch Mechanism, in Budding Yeast

Item Type text; Electronic Dissertation

Authors Paek, Andrew Luther

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 04/10/2021 02:16:46

Link to Item http://hdl.handle.net/10150/194260 1

FORMATION OF DICENTRIC AND ACENTRIC CHROMOSOMES, BY A TEMPLATE SWITCH MECHANISM, IN BUDDING YEAST

By Andrew L. Paek

A Dissertation Submitted to the Faculty of the DEPARTMENT OF MOLECULAR AND CELLULAR BIOLOGY In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA 2010

THE UNIVERSITY OF ARIZONA

GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Andrew Paek

Entitled Formation of Dicentric and Acentric Chromosomes, by a Template Switch Mechanism, in Budding Yeast

And recommend that it be accepted as fulfilling the dissertation requirement for the degree of Doctor of Philosophy

______Date: 10/01/2010 Ted Weinert

______Date: 10/01/2010 Kathleen Dixon

______Date: 10/01/2010 John Little

______Date: 10/01/2010 Lisa Nagy

______Date: 10/01/2010 Roy Parker

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

______Date: 10/01/2010 Dissertation Director: Ted Weinert

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Request for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the author

Signed: Andrew Paek

ACKNOWLEDGMENTS

I would like to thank my wonderful family for all the support over the years. To my parents, Don and Carol Paek, thanks for being loving parents and providing a wonderful education. To my siblings, Elizabeth McAshan and Donald Paek, thanks for your friendship and support over the years. I would also like to thank my future wife, Molly Stothert-Maurer, words cannot describe how important your presence has been throughout this whole process. You listened to my talks, you gave me support when needed, you even moved to the desert for me; I am extremely lucky to have a friend like you.

I would also like to thank my advisor Ted Weinert. Throughout my graduate career at the University of Arizona, Dr. Weinert has been both an excellent mentor and a friend. Ted was always available, especially if you wanted to discuss science and I will always appreciate that. He has a truly admirable enthusiasm for both science and life that I have always looked up to. I have learned so much from him, and will always be indebted to him for this.

I would also like to thank the other members of the Weinert lab. I always felt that working as a team was much more fun and productive than going solo, and the graduate students in the lab were always willing to collaborate with me. To Hope Jones, thanks for all the great science discussions, collaborations and most importantly the pranks. To Salma Kaochar, it was a pleasure working together with you on so many things, your intelligence and friendly personality made the lab a wonderful place to be. To Lisa Shanks, thanks for keeping the lab running and for all the fun conversations over the years. I also thank Arshed Al-Obeidi for being one of the hardest working undergraduates even until the last minute. All the undergraduates in the Weinert lab have added a lot to the group, so I would also like to thank Aly Elezaby, Cody Witham and Joseph Chao.

Special thanks goes out to the members of the JMST student group, Jay Konieczka, Tom Hartl, Michael Dellinger, Chris “Stretch” Pappas, Chris Bauer, Matt Callan and Chinedu Nworu. These guys helped out with all my talks, helped read through manuscripts, and threw some great parties. Thanks for making graduate school such a wonderful experience.

Finally I would like to thank my committee for always being there for science advice and keeping me on my toes. You have all helped me to grow intellectually and I appreciate that. I would also like to thank Andrew Capaldi, though not on my committee he always gave excellent advice and was a wonderful person to talk science with.

DEDICATION

To the memory of Justin Sparks,

who had such a strange and wonderful impact on all who were lucky enough to know him.

TABLE OF CONTENTS

LIST OF FIGURES……………………………………………………………………………………………………………09

ABSTRACT……………………………………………..……………………………………………………………………..10

AN EXPLANATION OF THE DISSERTATION FORMAT……………………………………..……………….13

CHAPTER I: INTRODUCTION……….……………………………………………………………………………….14

1.1 EXPLANATION OF THE PROBLEM AND ITS CONTEXT…………..………..……………………14

DNA replication…………………………….…………………………………………………………….14

DNA repeats……..………………………………………………………………………………………….15

Chromosome architecture………………………………..………………………………………….16

1.2 A REVIEW OF THE LITERATURE ON REPLICATION BYPASS PATHWAYS AND REPLICATION ASSOCIATED GENOME REARRANGEMENTS………………………………………….18

The many forms of genome rearrangements………………………………………………………18

General mechanisms underlying genome rearrangement…………………………………..19

The five degrees of freedom in replication fork recovery………………………..………….19

Evidence for the forks five degrees of freedom………………………………………...………..20

DNA polymerase uncoupling……………………..……………………………………………….……….22

Template-Switching…..……………………………………………………………………………….……….23

DSBs and stalled forks….…………………………………………………………………………….….……23

When fork recovery fails; role of sequence homology and colocalization…….…..…24 7

TABLE OF CONTENTS - Continued

Possible links between fork recovery failure and genome rearrangements…..…..24

CHAPTER II: PRESENT STUDIES………………………………………………………………………………..29

2.1 A BRIEF OVERVIEW OF WHAT WAS KNOWN…..………………………………………………29

2.2 A BRIEF OVERVIEW OF THE LOGIC AND KEY FINDINGS OF THE CURRENT

STUDY……………………………………………………………………………………………………..…..32

The unstable chromosomes are dicentric chromosomes…….………….….32

Dicentric chromosomes form at other inverted repeats in the yeast

genome………………………………………………………………………………………………..33

Inverted repeat fusion occurs independent of double strand break repair pathways……………………………………………………………………………………….33

Dicentric chromosome formation decreases in polymerase mutants……………………………………………………………………………….………………….35

2.3 ISOLATION AND ANALYSIS OF ADDITIONAL DICENTRIC CHROMOSOMES…38

2.4 STABILITY OF INVERTED REPEATS IN OTHER GENOMES…………..…………….…..41

2.5 FUSION OF NEARBY INVERTED REPEATS BY A REPLICATION-BASED MECHANISM LEADS TO FORMATION OF DICENTRIC AND ACENTRIC CHROMOSOMES THAT CAUSE GENOME INSTABILITY IN BUDDING YEAST………………………..…………....44 2.6 THE ROLE OF REPLICATION BYPASS PATHWAYS IN DICENTRIC CHROMOSOME FORMATION IN BUDDING YEAST………………………………………………………………..…46

TABLE OF CONTENTS - Continued

CHAPTER 3: FUTURE STUDIES...... 47

3.1 AN INTRODUCTION TO THE PROBLEM….……………….……………………….………..………47

3.2 HUMANS AND OTHER ORGANISMS HAVE MORE NUANCED REPAIR

PATHWAYS…………………………………………………………………………………………...………….48

3.3 STABILIZATION OF ACENTRIC AND DICENTRIC CHROMOSOMES……………………...50

3.4 SUMMARY………………………………………………………………………………………………………..53

REFERENCES………………………………………………………………………………………………………………55

APPENDIX A….…………………………………………………………………………………………………………..68 FUSION OF NEARBY INVERTED REPEATS BY A REPLICATION-BASED MECHANISM LEADS TO FORMATION OF DICENTRIC AND ACENTRIC CHROMOSOMES THAT CAUSE GENOME INSTABILITY IN BUDDING YEAST

APPENDIX B……………………………………………………………………………..………………………………135 THE ROLE OF REPLICATION BYPASS PATHWAYS IN DICENTRIC CHROMOSOME FORMATION IN BUDDING YEAST

LIST OF FIGURES

Figure 1.1: The replication fork’s five degrees of freedom………………………………………….27

Figure 2.1 Chromosome system to detect instability………………………………….…………………31

Figure 2.2 Strategy for the stabilization of dicentric chromosomes………………………………39

Figure 2.3 The frequency of inverted repeats grows exponentially with

increasing genome size……………………………………………………………………………….42

ABSTRACT

Chromosomal rearrangements occur in all organisms and are important both in the evolution of species and in pathology. In this dissertation I show that in Saccharomyces cerevisiae, or budding yeast, one type of chromosomal rearrangement occurs when inverted repeats fuse, likely during DNA replication by a novel mechanism termed “faulty template switching”. This fusion can lead to the formation of either a dicentric or acentric chromosome, depending on the direction of the replication fork. Dicentric chromosomes are inherently unstable due to their abnormal number of centromeres, and thus undergo additional chromosomal rearrangements and chromosome loss. Some of the work shown herein was done in collaboration with other members of the lab.

DICENTRIC CHROMOSOMES ARE FORMED BY INVERTED REPEAT FUSION

A genetic assay was previously developed that selects for colonies that arise from a cell with an unstable copy of chromosome VII. One of our goals was to identify the nature of the unstable chromosome. We showed that these unstable chromosomes are predominantly dicentric chromosomes by two separate methods. First we showed that cells which harbor a mutant version of the kinetochore protein Ctf13, which is known to stabilize dicentric chromosomes, form fewer mixed colonies. Second a PCR based assay was developed to detect a specific dicentric chromosome which would form if a pair of inverted repeats on chromosome VII fused. The PCR product was detected in liquid cultures and its sequence was consistent with formation of a dicentric chromosome. Using qPCR we showed that the frequency of dicentric chromosome formation correlated with unstable mixed colony formation. That is, cultures with a high frequency of dicentric chromosomes also showed a high frequency of mixed colony formation. From these data we conclude that inverted repeats often fuse to form unstable dicentric chromosomes. 11

INVERTED REPEAT FUSION OCCURS THROUGHOUT THE YEAST GENOME

Next, in order to determine whether this phenomenon was general, I sought evidence of dicentric chromosome formation at other inverted repeats throughout the yeast genome. I detected inverted repeat fusion at all five inverted repeats tested, which were located on four separate yeast chromosomes. Depending on the orientation of the fusion event, either an acentric or a dicentric chromosome was formed. At many inverted repeats we found evidence that both acentric and dicentric chromosomes can be formed.

GENETIC REQUIREMENTS OF INVERTED REPEAT FUSION

In order to determine the mechanism of inverted repeat fusion, I measured the frequency of dicentric chromosome formation in mutant cells defective for specific DNA replication and/or DNA repair pathways. Most chromosomal rearrangements are thought to arise after double strand breaks occur in DNA. If double strand breaks were the cause of inverted repeat fusion then one would expect fusion events to decrease in cells which are deficient for double strand break repair. Unexpectedly, I found that fusion events increased in cells deficient for double strand break repair, this implies that fusion does not require double strand breaks.

Further genetic analysis unveiled two separate mutations that decreased fusion events. These mutations provided clues for the mechanism of fusion. The first mutation was in the POL3 gene which is the catalytic subunit of Pol delta, the lagging strand DNA polymerase. The second mutation was in MGS1 a single strand-annealing protein which binds and stabilizes Pol delta. In a complex series of single, double and triple mutants (with the checkpoint protein Rad9), I identified defects in fusion as well as in other aspects of recombination. This and other data led us to propose a model whereby fusion occurs when replication forks stall and switch templates, a process we call “faulty template switching”. 12

ONGOING STUDIES

Preliminary data suggests that dicentric chromosomes form in several different locations on chromosome VII in addition to the specific dicentric described above. Currently, it is not known whether these dicentric chromosomes are formed by fusion of inverted repeats or by another mechanism. In order to determine the structure of these dicentric chromosomes, I have engineered a strain which is able to convert dicentric chromosomes into stable monocentric chromosomes while still leaving the overall chromosome structure intact. Using comparative genomic hybridization and DNA sequencing, I plan on determining the origin of these dicentric chromosomes.

In a separate project, I am working on determining how stable inverted repeat sequences are in other organisms. Through genome analysis I have quantified the frequency and number of inverted repeats in several different organisms. This analysis revealed an interesting trend; as genome size increases so does the frequency of inverted repeats. Similar observations have lead many to propose that organisms with a higher number of repetitive elements have evolved novel mechanisms to prevent rearrangements; mechanisms that are not present in organisms with few repeats. In order to test this hypothesis, I am quantifying the frequency of rearrangements at inverted repeats in E. coli, S. cerevisiae, D. melanogaster and in human tissue culture. If this hypothesis is correct, then inverted repeats in humans should be more stable than similar repeats in S. cerevisiae. 13

AN EXPLANATION OF THE DISSERTATION FORMAT

In accordance with the Manual of Thesis and Dissertations of the University of Arizona Graduate College and policies of the Graduate Program of the Department of Molecular and Cellular Biology, my work is presented here in two chapters. In the first chapter I introduce the problem and provide a review of the relevant literature. This review was published in the journal Current Opinion in Cell Biology. The second chapter contains a brief summary of what was known through previous work in the Weinert lab, a short description of the logic and key findings of the present study, and concludes with my current studies. The majority of the present study consists of two published manuscripts which are located in the appendix as follows:

Appendix A: A manuscript which was published in the journal Genes and Development in 2009. I am co-first author with another graduate student, Salma Kaochar. In the main body of the manuscript, I contributed all of the data for Figure 4, the majority of the data in Table 1, and worked out the model with the other authors outlined in Figure 7. For the supplement I provided the data for figures S1, S2, S5 and S8.

Appendix B: A manuscript which has been accepted for publication in the journal Genetics, due for publication in 2010. I am the first author of this publication and generated the majority of the data presented, as well as the models presented.

CHAPTER 1: INTRODUCTION

1.1 EXPLANATION OF THE PROBLEM AND ITS CONTEXT

All living organisms are prone to large scale changes to the structure of their genomes which are often called gross chromosomal rearrangements or GCRs. These changes include deletions, inversions, translocations and either loss or gain of entire chromosomes (ALBERTS et al. 2007). These rearrangements are extremely important as they can lead to evolutionary adaptations, speciation events, and are implicated in a number of different genetic disorders (LEE et al. 2007; STANKIEWICZ and LUPSKI 2010).

Due to the deleterious nature of GCRs, the cell has evolved numerous mechanisms which act to prevent their formation (MYUNG et al. 2001). Cells which have lost the function of these mechanisms have largely unstable genomes which have a high rate of GCR formation. Genome instability is a hallmark of cancer, and is thought to be the source of the complex phenotypes seen in tumor cells which are difficult to explain by conventional mutations (single base pair changes or single gene deletions; DUESBERG et al. 2005; HANAHAN and WEINBERG 2000). Currently it is not understood what events cause genomes to become destabilized, and what cellular pathways act to prevent these destabilizing events. In this introduction I will outline some of the major players that are thought to lead to chromosome instability: errors during DNA replication, repeat sequences, and chromosome architecture.

DNA replication

It has long been thought that DNA replication is a major source of GCR formation. Replication forks encounter many barriers during DNA replication, and must overcome these barriers in order to properly complete S-phase and enter into mitosis

(WEINERT et al. 2009). Barriers to replication forks can include protein complexes which are bound to DNA such as RNA polymerase, or molecular changes to the DNA such as 15

thymidine dimers (BREWER 1988; ZHANG and LAWRENCE 2005). In either case replication forks must either remove the barriers or bypass them in order to finish replication

(ATKINSON and MCGLYNN 2009; WEINERT et al. 2009).

Given the importance of DNA replication and the numerous barriers that impede it, it is not surprising that life has evolved several different pathways which act to overcome these barriers (WEINERT et al. 2009). These pathways are commonly referred to as replication bypass pathways and each one utilizes sequence homology in order to overcome replication barriers (see section 1.2 for detail). By utilizing sequence homology, replication bypass pathways limit the frequency of chromosomal rearrangements; using either identical sequences on a homologous chromosomes (in diploids) or on a just-replicated sister chromosome. This lowers the frequency of annealing to non-allelic sequences, and thus re-starting replication out of register.

DNA repeats

Repetitive elements make up approximately 50% of the human genome, and are prevalent in most genomes analyzed (ALBERTS et al. 2007; RICHARD et al. 2008). Duplicate genes make up a fraction of repetitive elements, the majority of repeat sequences in both the human genome and the S. cerevisiae genome are due to transposable elements (RICHARD et al. 2008). Repeat sequences pose a problem to replication bypass pathways since they utilize sequence homology for bypass (WEINERT et al. 2009). Undergoing bypass at sites which contain repeat sequences can lead to chromosomal rearrangements (LAMBERT et al. 2005). The structure of the resulting rearrangements depends upon the orientation of the repeat sequences. If the repeat sequences are in direct or head-to-tail orientation, improper annealing can lead to either deletion of sequences between the repeats, or expansion of this sequence (GOLDFLESS et al. 2006). If the repeat sequences are in inverted or head-to-head orientation, improper annealing 16

can lead to inversions, or as we show in this study, to the formation of dicentric or acentric chromosomes (BAI and SYMINGTON 1996; PAEK et al. 2009).

For the current study we focus on inverted repeats which are located in cis (meaning on the same chromosome) and are separated by 1.4 – 5.4 kb. We call these “nearby inverted repeats” to distinguish them from palindromes which are inverted repeats separated by less than 100 base pairs (RICHARD et al. 2008). Due to their close proximity the inverted repeats of palindromes can anneal to one another forming cruciform structures which have been shown to stall replication forks and induce chromosomal rearrangements (LOBACHEV et al. 2002; LOBACHEV et al. 2000; NARAYANAN et al. 2006; VOINEAGU et al. 2008). If the inverted repeats are separated by more than 100 base pairs, the frequency of cruciform formation and chromosomal rearrangements is severely reduced, making this an unlikely factor in the events in this study (LOBACHEV et al. 2000).

Chromosome architecture

Chromosome architecture, or the organization of DNA sequences on chromosomes, seems to play an important role in maintaining genome stability. When this architecture is disrupted chromosomes can become destabilized (ADMIRE et al. 2006;

PAEK et al. 2009). For example researchers who looked at monocentric chromosomal rearrangements in human tissue culture found that 10% of these rearrangements were unstable as indicated by slowed replication and an increase in chromosomal rearrangements (BREGER et al. 2005). Since these rearrangements were reciprocal and no genes were disrupted, this instability is not likely due to changes in protein levels, but rather is a consequence of changes in genome architecture. Altered monocentric chromosomes have also been shown to be unstable in S. cerevisiae (ADMIRE et al. 2006;

PAYEN et al. 2008). It is not currently known what makes these altered chromosomes unstable. 17

One egregious example of an unstable chromosome is a chromosome that contains two centromeres, a dicentric chromosome (PATHAK and MULTANI 2006). During mitosis, spindle microtubules pull on both centromeres of the dicentric chromosome and if these microtubules originate from opposite sides of the cell, can cause chromosome breakage (MCCLINTOCK 1941). In the next cell cycle, fusion of the broken chromosome ends often leads to formation of another dicentric chromosome, and the process is repeated. This process is referred to as the breakage-fusion-bridge cycle, and it is thought to be a major cause of the chromosome instability seen in cancer

(MCCLINTOCK 1941; PATHAK and MULTANI 2006).

In this study I show that, due to errors in DNA replication, nearby inverted repeats frequently fuse to form unstable dicentric and acentric chromosomes in S. cerevisiae. I also found that fusion of inverted repeats is general: fusion occurred at all five different locations tested on four separate yeast chromosomes. Through extensive genetic analysis I have shown that inverted repeat fusion is not likely to occur through a double strand break intermediate as is predicted for some other chromosomal rearrangements. Instead, the data suggests that fusion occurs when replication forks switch templates and anneal to a homologous yet non-allelic sequence, a process termed “faulty template switching”. Similar results were concurrently found by another group in the fission yeast Schizosacchromyces pombe; a yeast which is thought to have diverged from S. cerevisiae as long as it has from humans (MIZUNO et al. 2009; RUSSELL and NURSE 1986). Given the evolutionary distance of these two organisms, “faulty template switching” is likely to be extremely important in the evolution of species. Indeed, inverted repeat fusion was found to be associated with Turner syndrome, a disorder associated with unstable Y chromosomes (LANGE et al. 2009).

1.2 A REVIEW OF THE LITERATURE ON REPLICATION BYPASS PATHWAYS AND REPLICATION ASSOCIATED GENOME REARRANGEMENTS

Replication Fork’s Five Degrees of Freedom, Their Failure and Genome Rearrangements

T. Weinert, S. Kaochar, H. Jones, A. Paek. A.J. Clark

Curr Opin Cell Biol. 2009 Dec; 21(6): 778-784.

Abstract.

Genome rearrangements are important in pathology and evolution. Here we discuss the mechanisms that generate large-scale genome changes, such as translocations (called gross chromosomal rearrangements, GCRs, by the aficionados). Large-scale changes probably arise due to errors in DNA replication, resulting in either a fork- or DSB- mediated joining of two non-contiguous sequences. A key to rearrangement mechanisms is how replication forks normally overcome blockage, currently a hot research topic. When recovery fails, genome rearrangements occur. Drawing on studies mostly from S. cerevisiae, we discuss the five ways a replication fork may overcome blockage, and speculate briefly on the connections to chromosome rearrangements. (Feedback to the authors on topics discussed herein is welcome.)

The Many Forms of Genome Rearrangements

Genome instability is a broad term capturing any form of genetic change, from base pair changes, to small insertions, deletions or inversions, to larger-scale changes such as translocations, segmental duplications, and whole chromosome loss or duplication (1). Each is relevant for their implications for pathology (2-4). Here we focus on mechanisms 19

by which large-scale rearrangements may arise, drawing largely from the S. cerevisiae literature.

General mechanisms underlying genome rearrangement:

Genome rearrangements may in principle have a simple etiology; when cells replicate their DNA, replication forks stall at sites containing unusual DNA sequences that form secondary structures (5), at sites occupied by protein-DNA complexes (6, 7), or at sites of DNA damage (8, 9). Protein-DNA complexes are present either as features of gene expression (e.g. RNA polymerases; 6, 7) or as chromosome regulatory sequences (e.g. centromeres, origins of replication; 7). Regulatory and repair mechanisms attempt to extricate the stalled fork, error-free, by any of five mechanisms discussed below. When fork recovery mechanisms fail, the terminal strands in the stalled or broken fork anneal to other sequences in the genome, leading to genome rearrangements. Here we discuss what we call the forks five degrees of freedom for recovery, and discuss a few issues on how genome rearrangements might arise when specific mechanisms fail.

The five degrees of freedom in replication fork recovery:

Stalled replication forks recover by mechanisms that remain speculative (Figure 1). Excellent reviews of replication abound, more detailed than this one (10-12). Many ambiguities exist in this field. Even terms "stalled fork" and "lesion" most certainly have multiple molecular realities. There is a lot to discover.

The fork's five degrees of freedom of a stalled replication fork are shown in Figure 1. First, stalled forks may recover without fork remodeling; the offending damage is repaired or blocking protein removed (#1). Key regulators like Mec1 and Rad53 enforce this first option (10). The next four options have been called "fork collapse", generally 20

referring to the dissociation of DNA polymerases (13). The second, third and fourth options posit that a stalled fork undergoes a sister-strand annealing reaction, called a template-switch or copy choice. The idea is that template-switching allows either efficient repair/removal or replication past a lesion. (This idea, though attractive, has yet to gain direct experimental evidence) One form of template-switching (#2) involves a "behind the fork" template switch in which sister strands anneal and facilitate DNA synthesis, forming hemicatenanes (a shorter one of which may form at the initiation of replication; 14, 15), that then require resolution. Another form of template switching occurs in front of the fork (#3 and #4), and involves fork regression and reversal, without (#3) or with (#4) DNA replication. Reversal proceeds either by the reverse of regression, or by strand invasion and formation and resolution of a Holliday structure (see 12, 16). Finally, a fifth fork option involves a frank double-strand break (DSB), formed either by cleavage of the exposed single stranded DNA at the stalled fork or cleavage of the regressed fork. After the DSB is formed, fork recovery is completed by strand invasion, also not shown in Figure 1. Both fork-mediated and DSB-mediated mechanisms may require homologous recombination proteins to facilitate correct pairing (14, 16); without pairing proteins (e.g. in rad52 strand-annealing defective mutants), the primer may pair promiscuously to other sequences in the genome, leading to a genome rearrangement.

Evidence for the forks five degrees of freedom.

There is reasonable but far from incontrovertible evidence in support of these five degrees of freedom. Identifying specific DNA structures and linking specific proteins to their formation and resolution, is the current technical challenge being tackled by many in the field. Evidence on DNA structures comes from migration of replication structures in two dimensional gels (2-D gels, 17), or from electron microscopy (EM; see 8, 18). Two dimensional gels provide information about forks stall and abnormal structures. However additional biochemical treatments or needed to infer the precise description 21

of structures. For example, a key structure seen in 2-D gels is called an "X structure". One form is likely a Holliday structure, resolvable by RuvC and T4 endonuclease VII (14,16). A second non-Holliday X structure exists which may be a hemicatenane (14). (The mobility of regressed forks, #3 and #4 in Figure 1, is less certain. See discussion of T4 phage below and ref.18.) EM provides less ambiguous information on structures, yet is technically challenging for it requires purifying DNA structures formed in vivo (8,18). Proteins have been shown to associate with stalled forks (10,13,19), though there is not yet an established association of specific proteins to other fork structures in vivo.

Evidence bearing on the five degrees of freedom comes from studies of protein-induced or DNA damage induced fork stalling. Specific blocks are referred to as replication fork barriers, or RFBs. It is unclear to what extent specific RFBs represent the properties of stalled fork that cause instability. Nevertheless, each is fascinating and informative in its own right. Two studies suggest that forks may remain intact and simply proceed (option 1). In bacteria, a repressor-bound complex (20) blocked fork progression, and stalled forks did not break or form unusual structures. Fork recovery resumed promptly upon repressor removal (by adding inducer) in a recombination protein (RecA) independent fashion. (An ectopic Ter-site placed away from its normal site of replication termination in bacteria also lead to extended fork arrest; 21). In budding yeast, fork behavior has been examined using the rDNA RFB site system. A protein called Fob1 normally binds to specific RFB sites in the rDNA locus, and blocks replication in one direction only. This allows DNA replication and transcription of rDNA genes to be in the same direction, minimizing collisions. When placed outside the rDNA locus, the RFB-Fob1 system allowed the monitoring of fork behavior (22). Forks stalled briefly (for 30 minutes), but unlike HU or MMS-stalled forks, the Fob1-stalled forks did not form unusual DNA structures and did not require recombination (Rad52) or regulatory checkpoint proteins (Mec1, Rad53). Why the repressor and Fob1 stalled forks do not attempt recovery via template-switching or DSB formation is not clear to us. 22

Two other RFBs include one in fission yeast and one the bacteriophage T4. In fission yeast, the RTS system blocks replication to enforce unidirectional replication through the mating type locus (23). When two RTS1 sequences are placed in inverted orientation with an intervening spacer, and the RTS binding and fork-stalling protein Rtf1 is expressed, forks stall very efficiently, and homologous recombination frequently ensues. Fork stability does not involve checkpoint proteins. (Apparently, both RTS and Fob1 systems have evolved to avoid mechanisms that obviate the need for regulators that stabilize damage-induced stalled forks; 24,25.)

In bacteriophage T4 Long and Kreuzer describe an elegant system they call the "origin fork"; after infection, the replication origin fires, one of the two forks replicate immediately, while the other “origin fork” stalls and recovers later (26). This recovery behavior can be studied using 2-D gels and mutants of phage proteins, leading to key findings. For example, they identify on 2D gels a specific structure they propose is a regressed fork. These studies have real potential to enhance our understanding of stalled forks in eucaryotes.

Stalled forks that occur after DNA damage may be of the most biological relevance for instability, yet they are difficult to study because damage is necessarily random, and therefore not at specific DNA sites. Damage or drug treatment induces stalled forks, regressed forks, and X structures (proposed to be hemicatenanes), detectable by a combination of 2-D gels and electron microscopy (18, 26-30). Without detailing the many complex findings, Rad5 and Srs2 may be needed to form regressed forks (31,32). Rad53 prevents regression, and regulates Exo1s degradation of regressed forks that may lead to instability (29, 30). A number of proteins involved in postreplication repair (e.g. Rad18) may be involved in formation and resolution of the hemicatenanes (14,27). Progress is being made on assigning proteins and pathways to DNA structure.

DNA polymerase uncoupling. 23

Site-specific damage has been studied in the context of understanding the coupling and uncoupling between leading and lagging strand DNA polymerases. Elegant experiments in bacteria and SV40 demonstrated that leading and lagging strand synthesis, which is normally coupled, becomes uncoupled when the leading strand is blocked by damage (8,9). It seems reasonable to speculate, however, that leading and lagging strands retain some sort of coupling when stalled, at the least to minimize the amount of single stranded DNA. There is some evidence of proteins acting to couple proteins in the fork (Mrc1/Tof1 in yeast, DnaB helicase in bacteria; 19,33). The nature of coupling of stalled forks in vivo in the context of the DNA structures shown in Figure 1 remains to be explored.

Template-Switching.

There is ample evidence of template-switching in bacteria (34), and now in budding yeast as well. Smith et al showed template switching during a DSB repair reaction wherein a replication fork replicates an entire chromosome (a BIR event, for the aficionados), switching between allelic sequences present on two homologs as it does so (35). Schmidt et al found evidence for a similar though non-allelic template switch in forming genome rearrangements (36). Template-switching per se may not require extensive sequence homology. A human genome disorder consisting of multiple complex rearrangements is proposed to involve switching between short sequence (<10bp) homologies (3). Rattray et al provided evidence in budding yeast for single strand annealing reaction and polymerase extension also using short sequence homologies (37).

DSBs and stalled forks.

The extent to which frank DSBs occurs at stalled forks remains uncertain. Stalled forks in yeast and bacteria appear to break (for example, 7, 38,39) though it is unclear to us whether when DSBs occur. Perhaps DSBs represent a delayed, "nuclear option" in fork 24

recovery. Most models of instability favor a DSB intermediate, and issue addressed further below.

When fork recovery fails; role of sequence homology and colocalization.

When recovery fails, sequence homology and colocalization probably play important roles in determining which two sequences become fused. Rearrangements in budding yeast occur most frequently between regions of homology ranging from several to several hundred bases (40, 41). In mammalian cells, translocations in cancer cells frequently involve only short homologies (42), though fusions with longer stretches occur experimentally (43) and probably in genomic disorders (involving copy number variability; 4). Whether the length of homology indicates much about fork- or DSB- mediated mechanism is unclear, other than short homologies don’t require single strand annealing proteins for fusion (41, see below).

Co-localization of sequences to fuse must also be relevant. One study reported a higher frequency of colocalization of RET and ELE1 genes in thyroid cell as opposed to epithelial cells, and fusion of these two sequences is commonly found in thyroid cancer. (44). Specific sequences in yeast also colocalize and recombine (45, 46), and in bacteria sequences that are closer also recombine more frequently (47). Replication factors may influence colocalization as well (48).

Possible links between fork recovery failure and genome rearrangements

The potential goal of research in genome instability is to define the different kinds and causes of instability, and to ultimately understand underlying mechanisms. There are many genome instability systems in budding yeast (1,5,23,36, 37, 41,49,50), each exhibiting features useful in studying specific phenomena (e.g. instability of palindromes in Ref. 5). In our view there are two major unresolved questions. First, as stated as the 25

basic premise of this review, how are defects in replication fork recovery converted into rearranged genomes? For example, is the high instability of sgs1 mutants (36) due to a failure to resolve hemicatenanes (14), to process DSB ends (51), or to some other role of Sgs1? Clearly, a deeper understanding of replication fork biology should provide the answers to how instability arises.

A second major question is whether DSBs are intermediates in instability. DSBs certainly can be intermediates (5, 42,43), but whether they arise when forks stall is at present a question to resolve. One study by Pannunzio, Bailis and colleagues provides, in our view, a suggestion that DSBs may not be intermediates in at least some genome rearrangements (52). The authors divided the HIS3 gene into two DNA fragments inserted at two sites on different chromosomes, with an adjacent HO DSB cut site. The HIS3 gene fragments share 300 bases of sequence overlap, allowing their recombination to form an intact HIS3 gene. They found that spontaneously His+ cells were formed, contained translocations that fused the HIS3 fragments, as expected, and that DSBs induced by HO cleavage increased the frequency of His+ translocations. Strikingly, the genetics of the spontaneous His+ and DSB-mediated His+ events were completely different; though both events required RAD52, a key strand annealing protein, the spontaneous events had no requirement for Mre11, Rad1, Rad59 or Ku70, while the DSB-mediated events required these proteins. This suggests to us that spontaneous events are not proceeding by a frank DSB-mediated mechanism; perhaps a fork- mediated mechanism is at play.

We finish by citing another study by Payen, Fischer and colleagues that is interesting for a several reasons. They developed a genome instability assay that detects segmental duplication (41). Two genes encode identically-acting ribosomal proteins, deletion of one results in poor growth, and duplication of the existing gene (by segmental 26

duplication) restores growth. They studied the rearrangements formed, and found that segmental duplications involved either intra or intermolecular events and duplications of different sizes. (The authors favor the view that segmental duplication occurs via DSBs, though fork-mediated mechanisms seem possible as well.) Interestingly, Rad52, the strand annealing protein, was required when duplications involves two sites that were further apart (on the same chromosome or interchromosomally). Rad52 was not required for fusions that involved sequences that were closer together on the same chromosome. (As expected, Rad52-dependent fusions involved longer sequence homologies while Rad52-independent fusions involved short sequence homologies.) Rearrangements involving nearby sequences may occur more frequently and have much less stringent protein-function requirements, perhaps because they involve fork- mediated events.

Conclusion: Studies of replication fork biology and genome instability are progressing at a rapid pace. Understanding the normal replication fork recovery events is key to understanding how genome rearrangements arise. DSB or fork-mediated events drive genome rearrangements, and their relative contributions remain unclear.

Figure 1.1: THE REPLICATION FORK’S FIVE DEGREES OF FREEDOM. A replication fork encounters a lesion (in this case DNA damage, triangle), and uses one of 5 options to overcome it. The fork may simply resume replication (#1); may undergo sister-strand annealing and replication to form a hemicatenane that is resolved by a host of proteins (#2); may undergo sister-strand annealing to form a regressed fork (‘chicken foot’) that allows lesion repair (#3); may undergo sister-strand annealing to form a regressed fork that allows replication of the annealed strands, followed by regression or re-invasion, forming a Holliday structure that needs to be resolved (#4); fork undergoes strand breakage to form a double strand break (#5). The double strand break may also form from the regressed fork.

CHAPTER II: PRESENT STUDIES

2.1 A BRIEF OVERVIEW OF WHAT WAS KNOWN

A detailed account of the methods, results, and conclusions of this work is available in the publications which have been appended to this work. In this chapter I provide a brief summary of what was known before starting this project, and summarize the most important results from these publications. I conclude with a brief summary of my current projects in the lab.

Shortly after discovering the first DNA damage checkpoint gene, RAD9, my advisor Ted Weinert wanted to determine whether checkpoints influenced mitotic recombination frequencies (WEINERT and HARTWELL 1988). In order to do this he made use of a homologous recombination assay developed in the laboratory of his post doctoral adviser, Leland Hartwell (ADMIRE et al. 2006; CARSON and HARTWELL 1985). This assay made use of a haploid strain of budding yeast which contains two copies of chromosome VII, a chromosome VII disome (See Figure 2.1A). One copy of chromosome VII harbors the CAN1 gene on the left arm, which encodes arginine permease (Figure 2.1A). Arginine permease pumps arginine into the cell and in addition, will also pump canavanine a toxic analog of arginine into the cell causing cell death

(WALKER 1955). Thus cells which harbor the CAN1 gene die on media that contains canavanine, and allelic recombination frequencies can be measured by counting the number of canavanine resistant colonies that form in canavanine media.

When Dr. Weinert compared the frequency of canavanine resistant colonies in a wild-type strain to a rad9 strain he noticed two things. First, rad9 mutants contained more canavanine resistant colonies (ADMIRE et al. 2006). Second, and most important, most of the canavanine resistant colonies formed by rad9 mutants were sectored or serrated in appearance (ADMIRE et al. 2006). Further analysis revealed that these sectored colonies harbored cells with multiple (up to 6) different genotypes. Therefore 30

Dr. Weinert decided to name these sectored colonies “mixed colonies”. Since mixed colonies are founded by an individual cell, this implied that the original founder cell harbored an unstable chromosome which underwent rearrangements as the colony grew.

By studying mixed colonies members of the Weinert lab were able to show the following:

1) Using genetic markers on chromosome VII (Figure 2.1A), they found that about half of recombination events occurred in a 135 kb region on the left arm of

chromosome VII called the E2 region (ADMIRE et al. 2006).

2) Further molecular analysis revealed that events which occur in the E2 region involve a site 403kb from the left telomere, called the “403 site” (Figure 2.1B;

(ADMIRE et al. 2006). Cells in mixed colonies frequently contain a translocation which consists of fusion of the 403 site with the right arm of chromosome VII

(Figure 2.1B; (ADMIRE et al. 2006).

3) Sequencing the translocation junction revealed two fusion junctions; one junction was between inverted repeat sequences at the “403 site”, and the second junction was a fusion of one repeat in the 403 site and another repeat on

the right arm of chromosome VII (Figure 2.1B; (ADMIRE et al. 2006).

4) Perturbation of DNA replication increased the frequency of mixed colonies. This implies that the initial event which led to an unstable chromosome occurs during S-phase.

Figure 2.1 Chromosome system to detect instability. A. Two homologs of chromosome VII and mutant alleles on each allow for genetic detection of chromosome changes. The CAN1 gene has been removed from ChrV and inserted into one copy of ChrVII. Selection for the loss of the CAN1 gene allows growth of cells with any of three types of chromosome changes, including simple loss, allelic recombinants, and mixed colonies. Mixed colonies contain cells of multiple genotypes, including a specific translocation. B. Configuration of elements in the ChrVII403 site and the geometry and order of how fusion may occur. Two tRNA genes (pentagons) transcribe towards the oncoming fork and slow replication. Fusion between the two LTR sigma repeats (S2, S3), shown diagrammatically, forms a dicentric, followed by recombination between the two LTR delta sequences (D7, D11) to form the specific translocation. Adapted from Paek et al 2009.

2.2 A BRIEF OVERVIEW OF THE LOGIC AND KEY FINDINGS OF THE CURRENT STUDY

At the beginning of the current study three looming questions remained unanswered which proved critical to understanding the nature of these unstable chromosomes. First, what is the structure of the unstable chromosomes, and what caused it to be unstable? Second, are the unstable chromosomes formed on chromosome VII unique or do similar events occur on other yeast chromosomes? Third, how is the unstable chromosome formed, and how does the cell prevent these events from occurring. In this section I will highlight the approach I and other members of the lab took to answer these questions, and summarize the key results that are reported in this study.

THE UNSTABLE CHROMOSOMES ARE DICENTRIC CHROMOSOMES

In order to determine the structure of the initial unstable chromosome, another graduate student in the lab, Salma Kaochar looked at the sequence of the translocation which comes from mixed colonies. As mentioned previously there are two fusion junctions in this sequence, a fusion of inverted repeats at the “403 site”, and a fusion of a repeat sequence in the “403 site” with another repeat on the right arm of chromosome VII (See translocation in Figure 2.1B). Dr. Kaochar hypothesized that if the fusion of the inverted repeats occurred first, then this would lead to an unstable dicentric chromosome. In order to test this hypothesis she designed PCR primers that would detect fusion of these repeats, and thus would be diagnostic for presence of a dicentric chromosome. Surprisingly, she was able to detect the dicentric chromosome, in unselected cultures, and quantify it via quantitative PCR.

Next Dr. Kaochar set out to determine whether the dicentric chromosome was indeed responsible for the formation of mixed colonies. If this was the case then one would expect that cultures with more dicentric chromosomes in them would form mixed colonies. In order to test this Dr. Kaochar grew up 10 separate liquid yeast 33

cultures. For each culture a portion was plated to canavanine media to measure the frequency of mixed colony formation. With the remaining cells she extracted genomic DNA and measured the frequency of dicentric chromosome formation. She found a high degree of correlation between the frequency of mixed colony formation and dicentric chromosome formation (Spearman rank correlation coefficient .973, in Appendix A). This together with other data in Appendix A suggests that mixed colonies form when inverted repeats fuse to form dicentric chromosomes.

DICENTRIC CHROMOSOMES FORM AT OTHER INVERTED REPEATS IN THE YEAST GENOME

Given the ease at which Dr. Kaochar was able to detect dicentric chromosomes at the 403 site, I hypothesized that other inverted repeats throughout the yeast genome might also fuse. By scanning through the yeast genome I was able to find approximately 15 sites that appeared similar to the 403 site. I picked five separate locations on four different yeast chromosomes and designed primers to detect both dicentric and acentric chromosome formation at these sites. I was able to detect either dicentric or acentric chromosome formation at all sites tested, and in some sites was able to detect both. Quantitative PCR revealed that inverted repeat fusion occurred at a frequency similar to the 403 site. Together these data implied that inverted repeat fusion is a general phenomenon in the yeast genome.

INVERTED REPEAT FUSION OCCURS INDEPENDENT OF DOUBLE STRAND BREAK REPAIR PATHWAYS

The next task at hand was to determine the mechanism of inverted repeat fusion. Sequencing of acentric and dicentric fragments was somewhat revealing. First, I found that fusion events could occur in very short stretches of homology (<20 base pairs). Second, sequencing both and an acentric and a dicentric junction from a single site revealed distinct fusion junctions. This was important as this implied that fusion 34

events were not reciprocal, that is, an acentric and dicentric chromosome were likely formed independent of one another. Given that fusion required only modest amounts of homology, and fusions were non-reciprocal, I hypothesized that fusion events were not occurring by homologous recombination. Homologous recombination requires longer stretches of homology (over 250 base pairs as opposed to the 20 base pair requirement for inverted repeat fusion), and is a reciprocal mechanism (JINKS-ROBERTSON et al. 1993; KROGH and SYMINGTON 2004).

In order to test the hypothesis that inverted repeat fusion occurred independent of homologous recombination and to figure out what might cause fusion events, I took a genetic approach. If indeed fusion events required homologous recombination then one would expect mixed colonies which are formed by dicentric chromosomes to decrease in rad52 mutants which are defective for homologous recombination. In contrast, rad52 mutants (and other recombination mutants) showed an increase in mixed colony formation. This implied that dicentrics were not formed by homologous recombination, and that homologous recombination proteins act to prevent inverted repeat fusion.

What pathway is responsible for fusion events? This question proved difficult to answer. One by one both Hope Jones and I knocked out genes involved in DNA repair pathways looking for a decrease in the formation of mixed colonies. By looking at mixed colony formation in strains lacking genes in the homologous recombination (RAD52, RAD51, RAD59), break induced replication (POL32), non-homologous end joining (KU80, LIG4) and micro-homology mediated end joining pathways (LIG4) we were able to eliminate all double strand break pathways from involvement in inverted repeat fusion

(KROGH and SYMINGTON 2004; LYDEARD et al. 2007; MA et al. 2003). This led us to propose that fusion events do not require a double strand break for formation, as has been hypothesized for most chromosomal rearrangements. Instead we hypothesized that 35

fusion events occur when replication forks switch templates and anneal to non-allelic yet homologous sequences, a process we termed “faulty template switching”.

DICENTRIC CHROMOSOME FORMATION DECREASES IN POLYMERASE MUTANTS

In order to find evidence to support the hypothesis that fusion events occur by a “faulty template switching” mechanism I reasoned that mutations in DNA polymerase might prevent template switching and thus lower the frequency of fusion events. After analyzing several mutations I found two that were able to decrease both dicentric formation and mixed colony formation in a sensitized rad9 background. One mutation was a temperature sensitive point mutant in Pol delta (pol3-13), the lagging strand polymerase (GIOT et al. 1997). The other was a knockout of mgs1, a single strand annealing protein which binds to Pol delta and is thought to be involved in replication restart (VIJEH MOTLAGH et al. 2006). When combined a pol3-13 and an mgs1 mutant showed a significant 20 fold decrease in mixed colony formation, and a 5 fold decrease in dicentric chromosome formation (the discrepancy between these two numbers is discussed in detail in Appendix B). Together these data supported the hypothesis that fusion events occurred by a template switching mechanism.

Since fusion events are likely occurring by a template switching mechanism both Dr. Jones and I decided to look at how another replication bypass pathway, post- replication repair, might affect faulty template switching. The post-replication repair pathway is thought to be one of the major pathways the cell utilizes in order to bypass barriers to replication (LEE and MYUNG 2008). This pathway can be further divided into two separate sub-pathways; the translesion synthesis branch and the error-free bypass branch (LEE and MYUNG 2008). The translesion synthesis branch of the pathway bypasses replication barriers by recruiting translesion Polymerases such as Rad30 and Rev7 which are able to replicate past lesions often in an error-prone manner leading to frequent point mutations (LEE and MYUNG 2008). The error-free bypass branch is a 36

poorly understood pathway that bypasses replication barriers without causing point mutations (LEE and MYUNG 2008). By testing mutants which are defective for either translesion synthesis (rad30, rev7) or error-free bypass (rad5, ubc13) we were able to show that the error-free bypass branch, but not the translesion synthesis branch of post-replication repair acts to prevent faulty template switching events (LEE and MYUNG 2008).

To summarize, I have included the following point-by-point summary of the key findings of this study:

1) A mixed colony forms when a pair of inverted repeats at the 403 site fuses to form an unstable dicentric chromosome. This chromosome is a precursor to formation of the translocation discussed in section 2.1 (Appendix A).

2) Fusion of inverted repeats to form a dicentric chromosome is general; I was able to detect fusion events in all five inverted repeats tested, which are located on four separate chromosomes. These fusion events can lead to either dicentric or acentric chromosomes, depending on geometry (Appendix A).

3) Inverted repeat fusion requires only short stretches of homology (<20 base pairs); this implies that fusion could occur at many places in the yeast genome and highlight its importance in genome instability (Appendix A).

4) Inverted repeat fusion occurs independent of double strand break repair pathways; indeed some of these pathways act to prevent fusion events from occurring. This suggests that fusion likely does not occur through a “break and join” mechanism proposed for most chromosomal rearrangements (Appendix A).

5) Mutations in Pol delta (the lagging strand polymerase), and MGS1 (a single strand annealing protein which binds Pol delta and stabilizes replication forks) cause a decrease in fusion events. This supported a model that fusion events 37

occur when stalled replication forks attempt to bypass replication barriers by switching template strands and erroneously switch to the wrong template strand. (Appendix B).

6) Genes in the error free bypass branch of the post replication repair pathway are key regulators which act to prevent fusion events (Appendix A, B).

2.3 ISOLATION AND ANALYSIS OF ADDITIONAL DICENTRIC CHROMOSOMES

Our analysis of unstable chromosome VII formation led us to conclude that a number of unstable chromosomes form when inverted repeats in the 403 site fuse to form a dicentric chromosome. In this section I present evidence that dicentric chromosomes form at other sites on chromosome VII and I outline an ongoing project aimed at determining the structure of these dicentric chromosomes.

As discussed above, we initially investigated the inverted repeats at the 403 site because about half of all mixed colonies experience a recombination event in this region

(ADMIRE et al. 2006). The remaining mixed colonies show recombination at various places along the left arm of chromosome VII; and these mixed colonies have similar growth properties and genetic requirements as the mixed colonies formed in the E2 region. Together these data led us to propose that unstable chromosomes formed at other locations are dicentric chromosomes, and that these form by inverted repeat fusion as well.

In order to test this hypothesis I have manipulated our chromosome VII disome strain to contain a module which we call the “di-mono-izer”. The di-mono-izer converts unstable dicentric chromosomes into stable monocentric chromosomes by deleting one of the two centromeres, and forming an intact copy of the URA3 gene in the process (Figure 2.2). By plating mixed colonies onto media lacking Uracil only those cells which have deleted a centromere will survive. The rest of the chromosome structure remains intact, which allows us to analyze it and determine its structure (Figure 2.2). We hypothesize that we will be able to stabilize dicentric chromosomes formed in various sections of chromosome VII, and that the fusion junctions will likely contain inverted repeats.

Preliminary data looks promising. I have now acquired five separate Ura+ colonies, and have shown using pulse field gel electrophoresis (a technique that resolves 39

individual chromosomes on an agarose gel) that these colonies contain chromosomes with altered sizes. Four of these chromosomes appear to be formed outside of the 403 site. I am currently working on isolating additional Ura+ colonies with altered copies of chromosome VII. In order to determine the structure of these altered chromosomes I will analyze them using comparative genomic hybridization, an array based procedure which allows you to detect copy number changes in DNA content. This will allow me to predict the junction site, and thus design primers for isolation and sequencing of the junction.

Figure 2.2 Strategy for the stabilization of dicentric chromosomes. The centromere of the starting strain is flanked by direct repeats made using the URA3 gene. When dicentric chromosomes are formed, two copies of the direct repeat flanked centromeres will be present. At some frequency a set of these direct repeats will undergo a 40

recombination event represented by the X in the figure. This will cause deletion of a centromere and the formation of a stable Ura+ colony.

2.4 STABILITY OF INVERTED REPEATS IN OTHER GENOMES

Inverted repeats likely occur in most genomes, and evidence suggests that they can lead to chromosomal rearrangements and to genome disorders (LANGE et al. 2009;

VOINEAGU et al. 2008). Little is known about the frequency in which inverted repeats rearrange, especially in higher organisms. In this section I discuss my ongoing attempts to determine the frequency of rearrangements at inverted repeat sequences of various organisms.

In order to determine the impact that inverted repeat fusion might have on organisms with complex genomes, I have measured the frequency of inverted repeats in several different organisms (Figure 2.2). This analysis revealed an interesting trend, as genome size increases the frequency of inverted repeats increases exponentially. This means that organisms with relatively large genomes contain many inverted repeats. For example based on the frequencies calculated here there are roughly 550 inverted repeats in budding yeast, and 1.5 million inverted repeats in the human genome.

Given the vast number of inverted repeats in organisms with large genomes, we hypothesize that these organisms have evolved mechanism that stabilize these repeats; otherwise their genomes would undergo constant rearrangement. This hypothesis makes a prediction; the more inverted repeats an organism contains, the less frequent each individual inverted repeat leads to a rearrangement. In order to test this hypothesis I have designed a PCR based assay to detect inversions at inverted repeats. Using this assay I have been able to detect inversions in Escherichia coli, S. cerevisiae, Drosophila melanogaster and in human tissue culture cells. Using quantitative PCR, I have measured the frequency of these inversions for E. coli, S. cerevisiae, D. melanogaster and for human tissue culture cells. From this data I conclude that inversions occur quite frequently in all organisms analyzed; however it does not directly test the hypothesis that they occur less frequently in organisms with larger genomes. In 42

order to test this I plan on putting a set of inverted repeats which I have cloned on a plasmid into S. cerevisiae, D. melanogaster and human tissue culture cells. This way I can test whether a given inverted repeat inverts less frequently in organisms with larger genomes.

1000

C. Elegans Human Mouse IRs Arabadopsis per 100 Drosophila Mb Pombe Cerevisae E. coli

10 1.00E+006 1.00E+007 1.00E+008 1.00E+009 1.00E+010

Genome size

Figure 2.3. The frequency of inverted repeats grows exponentially with increasing genome size. The average number of inverted repeats per megabase of DNA is plotted on the Y-axis, while genome size is plotted on the X-axis. Error bars are given for the mouse and human genome as these frequencies are based on frequencies for only two chromosomes not the entire genome. For all other organisms the entire genome was used. Frequencies are determined using einverted, which is part of EMBOSS (RICE et al. 2000).

2.5 FUSION OF NEARBY INVERTED REPEATS BY A REPLICATION-BASED MECHANISM LEADS TO FORMATION OF DICENTRIC AND ACENTRIC CHROMOSOMES THAT CAUSE GENOME INSTABILITY IN BUDDING YEAST

Andrew L. Paek*, Salma Kaochar*, Hope Jones, Aly Elezaby, Lisa Shanks, and Ted Weinert

* First two authors contributed equally to this work.

Genes and Development 2009 23: 2861-2875

Abstract

Large-scale changes (gross chromosomal rearrangements; GCRs) are common in genomes and are often associated with pathological disorders. We report here that a specific pair of nearby inverted repeats in budding yeast fuse to form a dicentric chromosome intermediate, which then rearranges to form a translocation and other GCRs. We next show that fusion of nearby inverted repeats is general; we found that many nearby inverted repeats that are present in the yeast genome also fuse, as does a pair of synthetically constructed inverted repeats. Fusion occurs between inverted repeats that are separated by several kilobases of DNA and share >20 base pairs of homology. Finally, we show that fusion of inverted repeats, surprisingly, does not require genes involved in double-strand break (DSB) repair or genes involved in other repeat recombination events. We therefore propose that fusion may occur by a DSB- independent, DNA replication-based mechanism (which we term "faulty template switching"). Fusion of nearby inverted repeats to form dicentrics may be a major cause of instability in yeast and in other organisms.

2.6 THE ROLE OF REPLICATION BYPASS PATHWAYS IN DICENTRIC CHROMOSOME FORMATION IN BUDDING YEAST.

Andrew L. Paek, Hope Jones, Salma Kaochar, and Ted Weinert

Accepted to the journal Genetics

Abstract

Gross chromosomal rearrangements (GCRs) are large scale changes to chromosome structure, and can lead to human disease. We previously showed in Saccharomyces cerevisiae, that nearby inverted repeat sequences (~20 to 200bp of homology, separated by ~1-5kb) frequently fuse to form unstable dicentric and acentric chromosomes. Here we analyzed inverted repeat fusion in mutants of three sets of genes. First, we show that genes in the error-free post replication repair pathway prevent fusion of inverted repeats, while genes in the translesion branch have no detectable role. Second, we found that siz1 mutants, which are defective for Srs2 recruitment to replication forks, and srs2 mutants had opposite effects on instability. This may reflect separate roles for Srs2 in different phases of the cell cycle. Third, we provide evidence for a faulty template-switch model by studying mutants of DNA polymerases; defects in DNA pol delta (lagging strand polymerase) and Mgs1 (a pol delta interacting protein) lead to a defect in fusion events as well as allelic recombination. Pol delta and Mgs1 may collaborate either in strand annealing and/or DNA replication involved in fusion and allelic recombination events. Fourth, by studying genes implicated in suppression of GCRs in other studies, we found that inverted repeat fusion has a profile of genetic regulation distinct from these other major forms of GCR formation.

CHAPTER 3: FUTURE STUDIES

As discussed in previous chapters, organisms with comparatively large genome sizes tend to have a high frequency of inverted repeats (Figure 2.3). One natural question is to ask what kind of impact inverted repeat fusion has on these organisms, and what this might mean for human disease and evolution. In this chapter I hypothesize on how organisms with many inverted repeats might avoid fusion events, and how these different hypotheses can be tested in future studies.

3.1 AN INTRODUCTION TO THE PROBLEM

As shown in chapter 2, the frequency of inverted repeats in budding yeast that have the potential to undergo inverted repeat fusion, is about 40 per megabase of DNA

(Figure 2.3). In humans this frequency increases to about 400 inverted repeats per megabase (Figure 2.3). Taking into account the sizes of the perspective genomes, this means that there are roughly 500 inverted repeats in budding yeast and over one million inverted repeats in the human genome. In data presented in Appendix A, I have measured the rate of inverted repeat fusion at three different inverted repeats in budding yeast. Interestingly, all three of these sites fuse at roughly the same frequency

(1-8 fusion events per 100,000 cells). If all 500 inverted repeats in the budding yeast genome fused at this frequency we would expect that roughly five out of every 1000 yeast cells in a population have undergone a fusion event. If this same frequency were 48

to occur in human cells (with over 1 million inverted repeats), we would expect roughly ten fusion events for every human cell in a population.

Given the deleterious nature of inverted repeat fusion events, it seems natural to speculate that these back of the envelope calculations do not hold up for human cells. Based on this logic I hypothesize that organisms with a large number of inverted repeats have evolved mechanisms to prevent inverted repeat fusion. What follows is a summary of possible mechanisms that might have evolved to prevent inverted repeat fusion events and other deleterious repeat mediated chromosomal rearrangements.

3.2 HUMANS AND OTHER ORGANISMS HAVE MORE NUANCED REPAIR PATHWAYS

In chapter 1, I presented a review which outlined the evidence for five different mechanisms that cells use in order to bypass replication fork barriers. But why are there five separate mechanisms and not just one? One reasonable hypothesis is that the five different mechanisms evolved to handle different types of replication barriers. These separate pathways might be have evolved to handle different types of barriers. In organisms with a large number of inverted repeats, the existence of additional paralogous genes might enable more specialization and more efficient handling of replication fork barriers.

There is some evidence to support the hypothesis that different mechanisms for replication bypass have evolved to handle different types of barriers. Several different genetic assays have been designed for budding yeast in order to determine how 49

different repair pathways prevent chromosomal rearrangements (ADMIRE et al. 2006;

MYUNG et al. 2001a; PUTNAM et al. 2009). In the assay described in this study, rearrangements occur between inverted repeats located on the same chromosome. In another assay designed by Kolodner and colleagues, rearrangements occur between sequences with little or no homology (0-10bp), and are termed single copy rearrangements (MYUNG et al. 2001a; MYUNG et al. 2001b). When this assay is modified to include a repeat sequence, rearrangements occur between direct repeats located on separate chromosomes and are thus called duplication-mediated rearrangements

(PUTNAM et al. 2009). Interestingly, mutations in the separate replication bypass pathways have different effects on each of these three assays (PAEK et al. 2009; PUTNAM et al. 2009). For example, rad51 mutants, which are defective for the hemicatenane branch of bypass (Figure 1.1), have a 100 fold increase in inverted repeat fusion events, a fivefold decrease in duplication mediated rearrangements, and no effect on single- copy rearrangements (PAEK et al. 2009; PUTNAM et al. 2009). This is consistent with the

Rad51 protein functioning differently in different contexts.

Organisms with a large number of inverted repeats might have evolved additional, more nuanced pathways in order to bypass replication barriers. These additional pathways are more likely to “get it right”, or bypass replication barriers without erroneously annealing to nonallelic sequences. In support of this hypothesis there is a large increase in the number of genes devoted to DNA repair in higher eukaryotes (MCVEY 2010). For example there are two known paralogs of Rad51 in 50

budding yeast, and five known paralogs in human (MASSON et al. 2001). Rad51 is by no means the exception, for example there are five separate translesion polymerases in human cells, and in budding yeast there are only three (MCVEY 2010). Although it is not currently known what function the different paralogs in the various repair pathways serve, it has been shown that translesion polymerases show specificity in terms of what types of lesions they recognize (PRAKASH et al. 2005). In theory, if one were able to replace the DNA repair pathways in a human cell with budding yeast repair pathways, then this model would predict that inverted repeat fusion would be restored to the levels seen in budding yeast. A more likely test would be to stabilize inverted repeats in yeast by inserting genes from human repair pathways.

3.3 STABILIZATION OF ACENTRIC AND DICENTRIC CHROMOSOMES

Another strategy that higher organisms might employ is the stabilization of the resulting inverted repeat fusion events. If the resulting acentric or dicentric chromosome is stabilized, then cells will not undergo additional rearrangements or chromosome loss, thus limiting the amount of damage. There is a substantial amount of evidence that stabilization of acentric and dicentric chromosomes occurs in human cells and might be common after formation of these chromosomes (MEHTA et al. 2010).

There are two separate mechanisms that have been demonstrated to stabilize dicentric chromosomes in human cells. The first mechanism was proposed when researchers found that dicentric chromosomes which contain centromeres that are less 51

than 10 megabases apart from each other remain stable (LANGE et al. 2009; PAGE and

SHAFFER 1998; SULLIVAN and WILLARD 1998). There are two models which have been proposed to account for this stability. One model suggests that the proximity of the two centromeres might cause the centromeres to function as one individual centromere

(PAGE and SHAFFER 1998). In contrast another model proposes that the proximity of the two centromeres might prevent the attachment of spindle poles from the opposite sides of the cell (PAGE and SHAFFER 1998). Although this mechanism exists in budding yeast, it only occurs when centromeres are less than 5 kilobases apart and should therefore be much less effective in stabilizing yeast chromosomes (KOSHLAND et al. 1987).

The second strategy that human cells employ to stabilize dicentric chromosomes is to silence one of the two centromeres, thus rendering the chromosome functionally monocentric (HSU et al. 1975; THERMAN et al. 1974). Centromere silencing occurs by a

RNAi mechanism and requires a functional Dicer protein (KANELLOPOULOU et al. 2005). It is not clear how efficient this mechanism is in human cells. A recent study found that out of 13 independent dicentric Y chromosomes 10 of them where functionally monocentric (LANGE et al. 2009). The other 3 contained centromeres which were in close proximity so were likely to be functioning as monocentric chromosomes.

However, it is not known whether the high number of stabilized dicentrics was due to an efficient stabilization mechanism, or if this was due to a high incidence of loss of dicentric Y chromosomes which have not become functionally monocentric. 52

Stabilizing acentric chromosomes is a completely different task altogether. The predominant method for stabilizing acentric chromosomes is by forming a new centromere termed a neocentromere (MEHTA et al. 2010). Over 90 different neocentromeres on 20 different human chromosomes have been identified, indicating that this mechanism readily occurs on acentric chromosomes (MEHTA et al. 2010).

Neocentromere formation is not unique to humans, it has been observed in yeast and flies as well (MARSHALL et al. 2008). Interestingly, the location of the new centromeres is different in each of these organisms. In Schizosacchromyces pombe, or fission yeast, neocentromere formation occurs preferably at telomeres (ISHII et al. 2008). On

Drosophila neocentromere formation occurs near the lost centromere, while in humans neocentromeres form away from the original centromere yet not in telomeres (MEHTA et al. 2010). Thus different mechanisms for neocentromere formation have been proposed for each of these species (ISHII et al. 2008). It is therefore plausible that this process is more efficient in human than in yeasts, yet there is at present no evidence that this is the case.

Neocentromere formation and centromere silencing are both interesting strategies for stabilizing acentric and dicentric chromosomes respectively. However if these processes are not inhibited on normal monocentric chromosomes they can contribute to genome instability. Neocentromere formation on a monocentric chromosome could lead to a dicentric chromosome, while centromere silencing can render a monocentric chromosome acentric. How then are these processes regulated? 53

Although at this point it is not entirely clear, some models have been suggested.

Although neocentromeres contain most of the proteins of a normal centromere, they do lack CENP-B and possibly other centromere binding proteins (MEHTA et al. 2010). This might cause microtubule attachments to neocentromeres to be weaker than attachment to normal centromeres with the full complement of centromere binding proteins. Thus when microtubules from one side of the cell attach to the neocentromere and microtubules from the opposite side attach to a normal centromere, the resulting tug of war leads to loss of microtubule neocentromere attachments as opposed to a double strand break. Centromere silencing is thought to occur only when there are multiple copies of the centromere by a poorly understood yet intensely studied RNAi induced heterochromatin pathway (GREWAL 2010).

3.4 SUMMARY

There are likely to be many mechanisms which have evolved in higher eukaryotes to deal with the high number of inverted repeats contained in their genomes. If this were not the case then it would not be possible for these organisms to develop into healthy multicellular organisms. The formation of dicentric chromosomes by inverted repeat fusion has been observed in human sperm cells, which suggests these type of events might have had an impact on human evolution (LANGE et al. 2009).

In addition these events are thought to be a major cause of Turner syndrome (LANGE et al. 2009). Individuals with Turner syndrome are born with ambiguous genitalia and 54

exhibit “XO” mosaicism, a phenomenon whereby some somatic cells are XY, while others contain only an X chromosome (LANGE et al. 2009). It has thus been proposed that “XO” mosaicisim is caused by the presence of an unstable dicentric Y chromosome.

Given that inverted repeat fusion can occur in human’s it is likely to be implicated in other disease states as well. Many tumor suppressors are known to be involved in suppressing genome instability (OSBORNE et al. 2004). It is possible that loss of the multiple mechanisms that have evolved to prevent inverted repeat fusion might be one of the causes of the high frequency of chromosomal rearrangements seen in cancer cells.

REFERENCES

References For Section 1.1

ADMIRE, A., L. SHANKS, N. DANZL, M. WANG, U. WEIER et al., 2006 Cycles of chromosome instability are associated with a fragile site and are increased by defects in DNA replication and checkpoint controls in yeast. Genes Dev 20: 159-173.

ALBERTS, B., A. JOHNSON, J. KEWIS, M. RAFF, K. ROBERTS et al., 2007 Molecular Biology of the Cell Fifth Edition. Garland Science, San Francisco.

ATKINSON, J., and P. MCGLYNN, 2009 Replication fork reversal and the maintenance of genome stability. Nucleic Acids Res 37: 3475-3492.

BAI, Y., and L. S. SYMINGTON, 1996 A Rad52 homolog is required for RAD51-independent mitotic recombination in Saccharomyces cerevisiae. Genes Dev 10: 2025-2037.

BREGER, K. S., L. SMITH and M. J. THAYER, 2005 Engineering translocations with delayed replication: evidence for cis control of chromosome replication timing. Hum Mol Genet 14: 2813-2827.

BREWER, B. J., 1988 When polymerases collide: replication and the transcriptional organization of the E. coli chromosome. Cell 53: 679-686.

DUESBERG, P., R. LI, A. FABARIUS and R. HEHLMANN, 2005 The chromosomal basis of cancer. Cell Oncol 27: 293-318.

GOLDFLESS, S. J., A. S. MORAG, K. A. BELISLE, V. A. SUTERA, JR. and S. T. LOVETT, 2006 DNA repeat rearrangements mediated by DnaK-dependent replication fork repair. Mol Cell 21: 595-604.

HANAHAN, D., and R. A. WEINBERG, 2000 The hallmarks of cancer. Cell 100: 57-70.

LAMBERT, S., A. WATSON, D. M. SHEEDY, B. MARTIN and A. M. CARR, 2005 Gross chromosomal rearrangements and elevated recombination at an inducible site-specific replication fork barrier. Cell 121: 689-702.

LANGE, J., H. SKALETSKY, S. K. VAN DAALEN, S. L. EMBRY, C. M. KORVER et al., 2009 Isodicentric Y chromosomes and sex disorders as byproducts of homologous recombination that maintains palindromes. Cell 138: 855-869.

LEE, J. A., C. M. CARVALHO and J. R. LUPSKI, 2007 A DNA replication mechanism for generating nonrecurrent rearrangements associated with genomic disorders. Cell 131: 1235-1247.

LOBACHEV, K. S., D. A. GORDENIN and M. A. RESNICK, 2002 The Mre11 complex is required for repair of hairpin-capped double-strand breaks and prevention of chromosome rearrangements. Cell 108: 183-193.

LOBACHEV, K. S., J. E. STENGER, O. G. KOZYREVA, J. JURKA, D. A. GORDENIN et al., 2000 Inverted Alu repeats unstable in yeast are excluded from the human genome. Embo J 19: 3822-3830.

MCCLINTOCK, B., 1941 The Stability of Broken Ends of Chromosomes in Zea Mays. Genetics 26: 234-282.

MIZUNO, K., S. LAMBERT, G. BALDACCI, J. M. MURRAY and A. M. CARR, 2009 Nearby inverted repeats fuse to generate acentric and dicentric palindromic chromosomes by a replication template exchange mechanism. Genes Dev 23: 2876-2886.

MYUNG, K., C. CHEN and R. D. KOLODNER, 2001 Multiple pathways cooperate in the suppression of genome instability in Saccharomyces cerevisiae. Nature 411: 1073-1076.

NARAYANAN, V., P. A. MIECZKOWSKI, H. M. KIM, T. D. PETES and K. S. LOBACHEV, 2006 The pattern of gene amplification is determined by the chromosomal location of hairpin-capped breaks. Cell 125: 1283-1296.

PAEK, A. L., S. KAOCHAR, H. JONES, A. ELEZABY, L. SHANKS et al., 2009 Fusion of nearby inverted repeats by a replication-based mechanism leads to formation of dicentric and acentric chromosomes that cause genome instability in budding yeast. Genes Dev 23: 2861-2875.

PATHAK, S., and A. S. MULTANI, 2006 Aneuploidy, stem cells and cancer. EXS: 49-64.

PAYEN, C., R. KOSZUL, B. DUJON and G. FISCHER, 2008 Segmental duplications arise from Pol32-dependent repair of broken forks through two alternative replication- based mechanisms. PLoS Genet 4: e1000175.

RICHARD, G. F., A. KERREST and B. DUJON, 2008 Comparative genomics and molecular dynamics of DNA repeats in eukaryotes. Microbiol Mol Biol Rev 72: 686-727.

RUSSELL, P., and P. NURSE, 1986 Schizosaccharomyces pombe and Saccharomyces cerevisiae: a look at yeasts divided. Cell 45: 781-782.

STANKIEWICZ, P., and J. R. LUPSKI, 2010 Structural variation in the human genome and its role in disease. Annu Rev Med 61: 437-455.

VOINEAGU, I., V. NARAYANAN, K. S. LOBACHEV and S. M. MIRKIN, 2008 Replication stalling at unstable inverted repeats: interplay between DNA hairpins and fork stabilizing proteins. Proc Natl Acad Sci U S A 105: 9936-9941.

WEINERT, T., S. KAOCHAR, H. JONES, A. PAEK and A. J. CLARK, 2009 The replication fork's five degrees of freedom, their failure and genome rearrangements. Curr Opin Cell Biol 21: 778-784.

ZHANG, H., and C. W. LAWRENCE, 2005 The error-free component of the RAD6/RAD18 DNA damage tolerance pathway of budding yeast employs sister-strand recombination. Proc Natl Acad Sci U S A 102: 15954-15959.

References For Section 1.2

1. Kolodner, R.D., C.D. Putnam, and K. Myung, Maintenance of genome stability in Saccharomyces cerevisiae. Science, 2002. 297(5581): p. 552-7.

2. Sharp, A.J., et al., Discovery of previously unidentified genomic disorders from the duplication architecture of the human genome. Nat Genet, 2006. 38(9): p. 1038- 42.

3. Lee, J.A., C.M. Carvalho, and J.R. Lupski, A DNA replication mechanism for generating nonrecurrent rearrangements associated with genomic disorders. Cell, 2007. 131(7): p. 1235-47.

An oft-cited study of the structure of genome rearrangements in a human disorder. The authors invoke a template-switch model, in which DNA polymerase must "hop" 3 or 4 times between sequences up to 0.5 MB apart to sequences of little homology, to make sense of the rearrangement structures

4. Mefford, H.C. and E.E. Eichler, Duplication hotspots, rare genomic disorders, and common disease. Curr Opin Genet Dev, 2009. 19(3): p. 196-204.

5. Narayanan, V., et al., The pattern of gene amplification is determined by the chromosomal location of hairpin-capped breaks. Cell, 2006. 125(7): p. 1283-96.

6. Aguilera, A. and B. Gomez-Gonzalez, Genome instability: a mechanistic view of its causes and consequences. Nat Rev Genet, 2008. 9(3): p. 204-17.

7. Ivessa, A.S., et al., The Saccharomyces cerevisiae helicase Rrm3p facilitates replication past nonhistone protein-DNA complexes. Mol Cell, 2003. 12(6): p. 1525-36.

A classic study showing that in budding yeast, the replication apparatus may stall about once every 10 kb (if the Rrm3 DNA helicase is defective).

8. Cordeiro-Stone, M., et al., Analysis of DNA replication forks encountering a pyrimidine dimer in the template to the leading strand. J Mol Biol, 1999. 289(5): p. 1207-18.

A key study, using SV40 replication in vitro, defined plasmid templates with damage, and EM to study structures, that DNA replication can be uncoupled.

9. Pages, V. and R.P. Fuchs, Uncoupling of leading- and lagging-strand DNA replication during lesion bypass in vivo. Science, 2003. 300(5623): p. 1300-3. 59

A beautiful study using multiple technical tricks and mutants to show that, in bacteria, a lesion that blocks the leading strand, (and lagging strand--data not shown) causing uncoupling of DNA synthesis, with absolutely no detectable delay in replication.

10. Friedel, A.M., B.L. Pike, and S.M. Gasser, ATR/Mec1: coordinating fork stability and repair. Curr Opin Cell Biol, 2009. 21(2): p. 237-44.

11. Heller, R.C. and K.J. Marians, Replisome assembly and the direct restart of stalled replication forks. Nat Rev Mol Cell Biol, 2006. 7(12): p. 932-43.

12. Atkinson, J. and P. McGlynn, Replication fork reversal and the maintenance of genome stability. Nucleic Acids Res, 2009. 37(11): p. 3475-92.

An absolutely marvelous review that describes fully the issues of replication fork biology, complete with diagrams of strands going this way and that that make sense and are easy to follow.

13. Cobb, J.A., et al., DNA polymerase stabilization at stalled replication forks requires Mec1 and the RecQ helicase Sgs1. Embo J, 2003. 22(16): p. 4325-36.

The study often credited, among others, with defining what a collapsed fork does; DNA polymerases dissociate (regulated by Mec1) and Mcm helicases (regulated by Rad53). They performed ChIP experiments on synchronized cells after HU-depletion, generating stalled forks in specific regions of the genome.

14. Liberi, G., et al., Rad51-dependent DNA structures accumulate at damaged replication forks in sgs1 mutants defective in the yeast ortholog of BLM RecQ helicase. Genes Dev, 2005. 19(3): p. 339-50.

15. Lopes, M., et al., Branch migrating sister chromatid junctions form at replication origins through Rad51/Rad52-independent mechanisms. Mol Cell, 2003. 12(6): p. 1499-510.

16. Zou, H. and R. Rothstein, Holliday junctions accumulate in replication mutants via a RecA homolog-independent mechanism. Cell, 1997. 90(1): p. 87-96.

17. Friedman, K.L. and B.J. Brewer, Analysis of replication intermediates by two- dimensional agarose gel electrophoresis. Methods Enzymol, 1995. 262: p. 613- 27.

18. Sogo, J.M., M. Lopes, and M. Foiani, Fork reversal and ssDNA accumulation at stalled replication forks owing to checkpoint defects. Science, 2002. 297(5581): p. 599-602. 60

A landmark study using EM to study the stalled DNA structures from yeast cells.

19. Katou, Y., et al., S-phase checkpoint proteins Tof1 and Mrc1 form a stable replication-pausing complex. Nature, 2003. 424(6952): p. 1078-83.

20. Possoz, C., et al., Tracking of controlled Escherichia coli replication fork stalling and restart at repressor-bound DNA in vivo. Embo J, 2006. 25(11): p. 2596-604.

21. Bidnenko, V., S.D. Ehrlich, and B. Michel, Replication fork collapse at replication terminator sequences. Embo J, 2002. 21(14): p. 3898-907.

22. Calzada, A., et al., Molecular anatomy and regulation of a stable replisome at a paused eukaryotic DNA replication fork. Genes Dev, 2005. 19(16): p. 1905-19.

23. Lambert, S., et al., Gross chromosomal rearrangements and elevated recombination at an inducible site-specific replication fork barrier. Cell, 2005. 121(5): p. 689-702.

This study describes a RFB system in fission yeast that may yield deep insight into replication fork biology. In this study, the RTS system was manipulated to cause profound stalls of forks, leading to genome rearrangements.

24. Lopes, M., et al., The DNA replication checkpoint response stabilizes stalled replication forks. Nature, 2001. 412(6846): p. 557-61.

25. Tercero, J.A. and J.F. Diffley, Regulation of DNA replication fork progression through damaged DNA by the Mec1/Rad53 checkpoint. Nature, 2001. 412(6846): p. 553-7.

26. Long, D.T. and K.N. Kreuzer, Regression supports two mechanisms of fork processing in phage T4. Proc Natl Acad Sci U S A, 2008. 105(19): p. 6852-7.

A remarkable system described in this review. The "origin fork" may indeed be a very informative model for understanding fork behavior in more genetically complex eucaryotic systems. In particular, the authors find a specific 2-D gel structure that appears to be an intact regressed fork.

27. Branzei, D., F. Vanoli, and M. Foiani, SUMOylation regulates Rad18-mediated template switch. Nature, 2008. 456(7224): p. 915-20.

One of what will be many studies struggling to understand post-replication repair and their links to specific DNA structures. 61

28. Lopes, M., M. Foiani, and J.M. Sogo, Multiple mechanisms control chromosome integrity after replication fork uncoupling and restart at irreparable UV lesions. Mol Cell, 2006. 21(1): p. 15-27.

29. Cotta-Ramusino, C., et al., Exo1 processes stalled replication forks and counteracts fork reversal in checkpoint-defective cells. Mol Cell, 2005. 17(1): p. 153-9.

Together with the findings from ref 30, this EM study of DNA structures suggests how specific proteins are linked to specific DNA structures; Rad53 allows regressed fork formation, and Exo1 acts on regressed forks.

30. Segurado, M. and J.F. Diffley, Separate roles for the DNA damage checkpoint protein kinases in stabilizing DNA replication forks. Genes Dev, 2008. 22(13): p. 1816-27.

This study shows that mutations in exo1 have the remarkable ability to suppress the HU-sensitivity of rad53 mutants. This finding has at least two implications; first it suggests regulation of Exo1 by Rad53, and second it links Exo1 and Rad53 to regressed forks (see ref 29).

31. Blastyak, A., et al., Yeast Rad5 protein required for postreplication repair has a DNA helicase activity specific for replication fork regression. Mol Cell, 2007. 28(1): p. 167-75.

32. Kerrest, A., et al., SRS2 and SGS1 prevent chromosomal breaks and stabilize triplet repeats by restraining recombination. Nat Struct Mol Biol, 2009. 16(2): p. 159-67.

33. Indiani, C., et al., Translesion DNA polymerases remodel the replisome and alter the speed of the replicative helicase. Proc Natl Acad Sci U S A, 2009. 106(15): p. 6031-8.

34. Lovett, S.T., Polymerase switching in DNA replication. Mol Cell, 2007. 27(4): p. 523-6.

35. Smith, C.E., B. Llorente, and L.S. Symington, Template switching during break- induced replication. Nature, 2007. 447(7140): p. 102-5.

Study shows how DNA polymerase must switch between allelic sequences on two homologs as it carries out replication, starting from a DSB (A BIR event).

36. Schmidt, K.H., J. Wu, and R.D. Kolodner, Control of translocations between highly diverged genes by Sgs1, the Saccharomyces cerevisiae homolog of the Bloom's syndrome protein. Mol Cell Biol, 2006. 26(14): p. 5406-20. 62

37. Rattray, A.J., et al., A mechanism of palindromic gene amplification in Saccharomyces cerevisiae. Genes Dev, 2005. 19(11): p. 1390-9.

A genomic instability system induced by a DSB that generates palindromes by the single strand DNA "looping" back and pairing with small regions of homology. The mechanism was deduced from sequencing the rearrangements formed.

38. Weitao, T., et al., Dna2 helicase/nuclease causes replicative fork stalling and double-strand breaks in the ribosomal DNA of Saccharomyces cerevisiae. J Biol Chem, 2003. 278(25): p. 22513-22.

39. Seigneur, M., et al., RuvAB acts at arrested replication forks. Cell, 1998. 95(3): p. 419-30.

40. Putnam, C.D., V. Pennaneach, and R.D. Kolodner, Saccharomyces cerevisiae as a model system to define the chromosomal instability phenotype. Mol Cell Biol, 2005. 25(16): p. 7226-38.

41. Payen, C., et al., Segmental duplications arise from Pol32-dependent repair of broken forks through two alternative replication-based mechanisms. PLoS Genet, 2008. 4(9): p. e1000175.

A study of segmental duplication in budding yeast. The system is described briefly in this review. It’s a lovely study.

42. Elliott, B. and M. Jasin, Double-strand breaks and translocations in cancer. Cell Mol Life Sci, 2002. 59(2): p. 373-85.

43. Elliott, B., C. Richardson, and M. Jasin, Chromosomal translocation mechanisms at intronic alu elements in mammalian cells. Mol Cell, 2005. 17(6): p. 885-94.

44. Nikiforova, M.N., et al., Proximity of chromosomal loci that participate in radiation-induced rearrangements in human cells. Science, 2000. 290(5489): p. 138-41.

45. Lichten, M. and J.E. Haber, Position effects in ectopic and allelic mitotic recombination in Saccharomyces cerevisiae. Genetics, 1989. 123(2): p. 261-8.

46. Burgess, S.M. and N. Kleckner, Collisions between yeast chromosomal loci in vivo are governed by three layers of organization. Genes Dev, 1999. 13(14): p. 1871- 83.

47. Bi, X. and L.F. Liu, DNA rearrangement mediated by inverted repeats. Proc Natl Acad Sci U S A, 1996. 93(2): p. 819-23. 63

48. Kitamura, E., J.J. Blow, and T.U. Tanaka, Live-cell imaging reveals replication of individual replicons in eukaryotic replication factories. Cell, 2006. 125(7): p. 1297- 308.

49. Admire, A., et al., Cycles of chromosome instability are associated with a fragile site and are increased by defects in DNA replication and checkpoint controls in yeast. Genes Dev, 2006. 20(2): p. 159-73

50. Lemoine, F.J., et al., Chromosomal translocations in yeast induced by low levels of DNA polymerase a model for chromosome fragile sites. Cell, 2005. 120(5): p. 587- 98.

51. Mimitou, E.P. and L.S. Symington, Sae2, Exo1 and Sgs1 collaborate in DNA double-strand break processing. Nature, 2008. 455(7214): p. 770-4.

A soon to be classic study of how a DSB is degraded first by Sae2-directed mechanism, and then Exo1 and Sgs1.

52. Pannunzio, N.R., G.M. Manthey, and A.M. Bailis, RAD59 is required for efficient repair of simultaneous double-strand breaks resulting in translocations in Saccharomyces cerevisiae. DNA Repair (Amst), 2008. 7(5): p. 788-800.

References For Chapter 2

CARSON, M. J., and L. HARTWELL, 1985 CDC17: an essential gene that prevents telomere elongation in yeast. Cell 42: 249-257.

GIOT, L., R. CHANET, M. SIMON, C. FACCA and G. FAYE, 1997 Involvement of the yeast DNA polymerase delta in DNA repair in vivo. Genetics 146: 1239-1251.

JINKS-ROBERTSON, S., M. MICHELITCH and S. RAMCHARAN, 1993 Substrate length requirements for efficient mitotic recombination in Saccharomyces cerevisiae. Mol Cell Biol 13: 3937-3950.

KROGH, B. O., and L. S. SYMINGTON, 2004 Recombination proteins in yeast. Annu Rev Genet 38: 233-271.

LEE, K. Y., and K. MYUNG, 2008 PCNA modifications for regulation of post-replication repair pathways. Mol Cells 26: 5-11.

LYDEARD, J. R., S. JAIN, M. YAMAGUCHI and J. E. HABER, 2007 Break-induced replication and telomerase-independent telomere maintenance require Pol32. Nature 448: 820- 823.

MA, J. L., E. M. KIM, J. E. HABER and S. E. LEE, 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.

RICE, P., I. LONGDEN and A. BLEASBY, 2000 EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet 16: 276-277.

VIJEH MOTLAGH, N. D., M. SEKI, D. BRANZEI and T. ENOMOTO, 2006 Mgs1 and Rad18/Rad5/Mms2 are required for survival of Saccharomyces cerevisiae mutants with novel temperature/cold sensitive alleles of the DNA polymerase delta subunit, Pol31. DNA Repair (Amst) 5: 1459-1474. 65

WALKER, J. B., 1955 Canavanine and homoarginine as antimetabolites of arginine and lysine in yeast and algae. J Biol Chem 212: 207-215.

WEINERT, T. A., and L. H. HARTWELL, 1988 The RAD9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae. Science 241: 317-322.

References For Chapter 3

GREWAL, S. I., 2010 RNAi-dependent formation of heterochromatin and its diverse functions. Curr Opin Genet Dev 20: 134-141.

HSU, T. C., S. PATHAK and T. R. CHEN, 1975 The possibility of latent centromeres and a proposed nomenclature system for total chromosome and whole arm translocations. Cytogenet Cell Genet 15: 41-49.

ISHII, K., Y. OGIYAMA, Y. CHIKASHIGE, S. SOEJIMA, F. MASUDA et al., 2008 Heterochromatin integrity affects chromosome reorganization after centromere dysfunction. Science 321: 1088-1091.

KANELLOPOULOU, C., S. A. MULJO, A. L. KUNG, S. GANESAN, R. DRAPKIN et al., 2005 Dicer- deficient mouse embryonic stem cells are defective in differentiation and centromeric silencing. Genes Dev 19: 489-501.

KOSHLAND, D., L. RUTLEDGE, M. FITZGERALD-HAYES and L. H. HARTWELL, 1987 A genetic analysis of dicentric minichromosomes in Saccharomyces cerevisiae. Cell 48: 801-812.

MARSHALL, O. J., A. C. CHUEH, L. H. WONG and K. H. CHOO, 2008 Neocentromeres: new insights into centromere structure, disease development, and karyotype evolution. Am J Hum Genet 82: 261-282.

MASSON, J. Y., M. C. TARSOUNAS, A. Z. STASIAK, A. STASIAK, R. SHAH et al., 2001 Identification and purification of two distinct complexes containing the five RAD51 paralogs. Genes Dev 15: 3296-3307.

MCVEY, M., 2010 Strategies for DNA interstrand crosslink repair: insights from worms, flies, frogs, and slime molds. Environ Mol Mutagen 51: 646-658.

MEHTA, G. D., M. P. AGARWAL and S. K. GHOSH, 2010 Centromere identity: a challenge to be faced. Mol Genet Genomics 284: 75-94. 67

MYUNG, K., C. CHEN and R. D. KOLODNER, 2001a Multiple pathways cooperate in the suppression of genome instability in Saccharomyces cerevisiae. Nature 411: 1073-1076.

MYUNG, K., A. DATTA and R. D. KOLODNER, 2001b Suppression of spontaneous chromosomal rearrangements by S phase checkpoint functions in Saccharomyces cerevisiae. Cell 104: 397-408.

OSBORNE, C., P. WILSON and D. TRIPATHY, 2004 Oncogenes and tumor suppressor genes in breast cancer: potential diagnostic and therapeutic applications. Oncologist 9: 361-377.

PAGE, S. L., and L. G. SHAFFER, 1998 Chromosome stability is maintained by short intercentromeric distance in functionally dicentric human Robertsonian translocations. Chromosome Res 6: 115-122.

PRAKASH, S., R. E. JOHNSON and L. PRAKASH, 2005 Eukaryotic translesion synthesis DNA polymerases: specificity of structure and function. Annu Rev Biochem 74: 317- 353.

PUTNAM, C. D., T. K. HAYES and R. D. KOLODNER, 2009 Specific pathways prevent duplication-mediated genome rearrangements. Nature 460: 984-989.

RICE, P., I. LONGDEN and A. BLEASBY, 2000 EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet 16: 276-277.

SULLIVAN, B. A., and H. F. WILLARD, 1998 Stable dicentric X chromosomes with two functional centromeres. Nat Genet 20: 227-228.

THERMAN, E., G. E. SARTO and K. PATAU, 1974 Apparently isodicentric but functionally monocentric X chromosome in man. Am J Hum Genet 26: 83-92.

APPENDIX A

FUSION OF NEARBY INVERTED REPEATS BY A REPLICATION-BASED MECHANISM LEADS TO FORMATION OF DICENTRIC AND ACENTRIC CHROMOSOMES THAT CAUSE GENOME INSTABILITY IN BUDDING YEAST

Andrew L. Paek1*, Salma Kaochar1*, Hope Jones1, Aly Elezaby1, Lisa Shanks1, and Ted Weinert 1,2

1 Department of Molecular and Cellular Biology, University of Arizona, Tucson, AZ 85721, USA

2Corresponding author:

E-mail: [email protected]

Tel (520) 621-8511 Fax (520) 621-3709

* First two authors contributed equally to this work.

Abstract

Large-scale changes (gross chromosomal rearrangements; GCRs) are common in genomes and are often associated with pathological disorders. We report here that a specific pair of nearby inverted repeats in budding yeast fuse to form a dicentric chromosome intermediate, which then rearranges to form a translocation and other GCRs. We next show that fusion of nearby inverted repeats is general; we found that many nearby inverted repeats that are present in the yeast genome also fuse, as does a pair of synthetically-constructed inverted repeats. Fusion occurs between inverted repeats that are separated by several kilobases of DNA and share greater than 20 base pairs of homology. Finally, we show that fusion of inverted repeats, surprisingly, does not require genes involved in double-strand break repair nor genes involved in other repeat recombination events. We therefore propose that fusion may occur by a DSB- independent, DNA replication-based mechanism (that we term "faulty template switching"). Fusion of nearby inverted repeats to form dicentrics may be a major cause of instability in yeast and in other organisms.

Introduction

Large scale chromosome changes (gross chromosomal rearrangements; GCRs) occur in all genomes and can occur at a high frequency. Twin studies show frequent recombination between nearby (<1MB) large (>1kb) direct repeat sequences (BRUDER et al. 2008). Complex genome rearrangements cause loss or duplication of genes that affect human pathology (such as loss or gain of the PLP1 gene that affects myelination in humans; (LEE et al. 2007). Numerous large scale changes are present in cancer cells and are thought to contribute to the cancer phenotype (BIGNELL et al. 2007). Bacterial genomes also suffer large-scale rearrangements (Tillier and Collins 2000). How these changes occur is largely unknown, and of interest in understanding both pathology and genetic diversity.

The mechanisms underlying GCR formation appear complex, and may involve some combination of DNA replication, double strand breaks (DSBs), repeat sequences, and unstable chromosome intermediates. One common view is that GCR formation is initiated by telomere or DNA replication dysfunction, leading to DSBs and subsequent fusion of two DSBs. DSB fusions may occur by any of several repair mechanisms

(KOLODNER et al. 2002; MURNANE and SABATIER 2004; WEINSTOCK et al. 2006; WONG et al.

2000), or the breaks may be capped by telomere addition (HACKETT et al. 2001; PUTNAM et al. 2004; SABATIER et al. 2005) Chen and Kolodner 1999; see Mizuno et al. 2009 for discussion). Recent studies of instability in yeast are in general consistent with the conclusion that DSBs can be intermediates in instability (HACKETT and GREIDER 2003;

LEMOINE et al. 2005; MYUNG et al. 2001b; NARAYANAN et al. 2006; RATTRAY et al. 2005). The question of whether GCRs also arise from an alternative replication-based mechanism, without a DSB intermediate, remains speculative. However, replication-based 71

mechanisms have been inferred from complex rearrangements in human disorders (Lee et al. 2007), and rearrangements in bacteria (Bi and Liu 1996).

Repeat sequences present in genomes contribute to the formation of GCRs (whether a DSB is involved or not). Immediately adjacent repeat sequences, called palindromes, may form hairpin structures that are easily broken and precipitate further genome rearrangements (LOBACHEV et al. 2002). Once a damaged structure (e.g. DSB or stalled fork) is formed, repeats may provide sequence homology that directs DNA repair (DUJON

2006; SCHMIDT et al. 2006b; VOINEAGU et al. 2008; ZHOU et al. 2001). Repeats in eukaryotes are plentiful (RICHARD et al. 2008) and their recombination in mammalian cells is common (LOBACHEV et al. 2007; ZHOU et al. 2001). Pairs of nearby repeats studied in this report are of special interest since their close proximity and extensive homology may render them highly prone to rearrangements.

A third issue concerning GCR formation is whether they involve unstable chromosome intermediates. The presence of an unstable chromosome would readily explain how one cell generates progeny with many different chromosome structures, as commonly observed in cancer cells (WEINSTOCK et al. 2006). An apparently common unstable chromosome is a dicentric, first described by McClintock (1941); once formed, the attempted segregation of the two centromeres of the dicentric to opposite poles during mitosis leads to chromosome breakage and further rearrangements (called the breakage-fusion-bridge (BFB) cycle). Dicentrics and the BFB cycle have been invoked to explain the behavior of complex chromosome changes (HACKETT and GREIDER 2003; LO et al. 2002).

In this report we provide evidence that GCRs form when nearby inverted repeats fuse to form unstable dicentric intermediates. This study is an extension of a previous study describing a highly unstable region of a yeast chromosome (Chromosome VII) that underwent multiple recombination events, formed unstable chromosomes, eventually forming a specific GCR (ADMIRE et al. 2006a). We reported that this highly unstable region contains inverted repeats and tRNA genes that appear to contribute to instability, instability that was further increased when DNA replication was disrupted. How the inverted repeats underwent recombination to form unstable chromosomes was unclear, and is clarified in this study. We now demonstrate that those inverted repeats first fuse to form a dicentric chromosome, that then rearranges further to form GCRs. We show that in budding yeast, fusion of nearby inverted repeats is a general phenomenon; we report that fusion occurs between naturally-occurring and synthetically-constructed inverted repeats present at different sites in the yeast genome. We then investigate the mechanism of fusion and find, surprisingly, that fusion does not require known DSB repair and replication fork-rescue pathways, nor does it proceed by mechanisms known to mediate direct or inverted repeat recombination. We therefore suggest a replication- based mechanism of fusion of nearby inverted repeats. In an accompanying manuscript, Mizuno et al. (2009) also present evidence for fusion of nearby inverted repeats in fission yeast, and suggest that fusion events also occurs by a DSB-independent, replication-based mechanism.

Results

The genetic system that led us to the discovery of inverted repeat fusion is shown in

Figure 1A (CARSON and HARTWELL 1985b). Briefly, we constructed a haploid yeast strain that contains an extra copy of ChrVII (a disomic strain, “ChrVII” means chromosome VII). Any changes to this extra chromosome should not compromise cell viability per se. The CAN1 gene, inserted near the left telomere of the extra chromosome, provides the means of identifying cells with chromosome changes. CAN1 encodes arginine permease that imports arginine as well as the toxic analog canavanine; only cells that lose the CAN1 gene can grow in media containing canavanine (forming CanR colonies). Using multiple genetic markers, we detect three types of chromosome changes in CanR colonies: whole chromosome loss, normal allelic recombination, and other complex rearrangements that give rise to colonies with a sectored appearance (which we will refer to as ‘mixed colonies’; Figure 1A). (Point mutations in CAN1 occur more rarely; (Weinert and Hartwell 1990). Mixed CanR colonies contain cells of multiple genotypes, whereas normal CanR colonies contain cells of a single genotype (See Experimental Procedures). Previously we were able to infer that a mixed colony arises from a cell with an unstable chromosome, whereas a normal colony arises from a cell with a stable chromosome (Admire et al, 2006).

The initial focus of this study is to determine the nature of the unstable chromosome that forms mixed colonies. We use rad9 mutants here merely as a tool to study unstable chromosomes, for unstable chromosomes are formed more frequently in rad9 mutants than in wild type (Rad+) cells (ADMIRE et al. 2006a). We emphasize that unstable chromosomes are formed in other mutants as well (some at an even higher frequency than in rad9 mutants, see below.)

A major clue to solving the structure of the unstable chromosome came from the structure of a specific translocation (the "D7/D11" translocation) present only in mixed colonies (Figure 1B). Our previous analyses suggest that this translocation probably arises by multiple deletion and fusion events centered at a site 403 kb from the left end (Admire et al. 2006; Figure 1B). This ChrVII403 site contains long terminal repeat (LTR) sequences as well as tRNA genes that appear to contribute to instability (ADMIRE et al.

2006a); H.J. T.W. unpublished data). LTRs are derived from retrotransposons (KIM et al. 1998) and are present in 100's of copies in the yeast genome. (The Sigma 2 (S2) and Sigma 3 (S3) repeats are derived from Ty3 elements, are 278 and 245 bases long, share 82% identity, and are 3.3 kb apart, while the Delta 7 (D7) and Delta 11 (D11) are from Ty1, are both 331 bases, share 97% identity and are 133 kb apart.) Given the structure of the D7/D11 translocation, we suggest a relatively simple model in which one event fuses S2 and S3 inverted repeats to generate a dicentric chromosome, depicted in Figure 1B, and a second event fuses D7 and D11 direct repeats to resolve the dicentric and form the translocation. We first test this hypothesis that a dicentric intermediate is involved in ChrVII’s instability.

A specific dicentric is present in cells that form mixed colonies

We employed a PCR assay to test for the presence of the specific dicentric (Figure 2A). We reasoned that the dicentric, if formed, might be present at a low but detectable amount in a culture of cells, and the amount of dicentric might predict the frequency of CanR mixed colonies arising from that culture. We therefore grew (in rich media) ten single rad9 cells (initially containing an unrearranged ChrVII) each into ten independent cultures (first to colonies on plates, then each into liquid culture). We then performed two tests on each of these ten cultures. We isolated genomic DNA from each culture and subjected it to amplification using primers that should detect the specific dicentric chromosome (Figure 2A). We did in fact detect a specific PCR fragment of the correct 75

size and sequence if S2 and S3 inverted repeats fused to form the dicentric (see Figure S1 for sequence). Control PCR reactions failed to amplify the same DNA fragment from native ChrVII403 site sequences (data not shown, native sequences were made by PCR amplifying the ChrVII403 site). (We note that these PCR primers would amplify the same DNA fragment if it arose by an inversion event, a formal possibility we eliminate by genetic and molecular tests; Figure S2, and see below).

We next tested if the amount of the PCR product correlates with the frequency of mixed colonies in the ten cultures. If so, cultures with quantitatively more PCR product should give rise to more mixed colonies. We determined the amount of dicentrics by quantitative PCR, and the frequency of mixed colonies by a quantitative genetic assay (see Experimental Procedures). We found that in fact cultures that had more dicentrics gave rise to generally a higher frequency of mixed colonies (Figure 2B; correlation coefficient 0.97; p<0.001). The concentration of dicentrics varied between cultures, with an average of about 1 in 2,000 rad9 cells (and 1 in 100,000 cells in Rad+ cells, Figure S3).

We further tested the correlation between the amount of PCR product and instability studying strains with low instability (Rad+, wildtype cells) and high instability (rad51 mutants with a defect in homologous recombination; Figure 2B; discussed further below). Again, we found more dicentric PCR product in rad51 mutants with high instability than in Rad+ strains with low instability (Figure 2B). We observe this correlation in other mutants as well (see below). We conclude that the specific dicentric depicted in Figure 1B and Figure 2A is most likely a key intermediate to instability and to formation of the specific D7/D11 translocation shown in Figure 1B.

Genetic test confirms that dicentrics cause the instability in mixed colonies Though results from the PCR assay described above argue strongly that a specific dicentric causes instability, we cannot directly detect intact dicentric chromosomes (due to their fragility and low abundance). We therefore turned to a genetic approach to test if in fact dicentrics are intermediates in instability.

Our genetic test uses a temperature-sensitive mutant of a kinetochore protein, ctf13-30, that functions properly at 23˚C but not at 36˚C (DOHENY et al. 1993). Ctf13 connects microtubules to the kinetochore. The idea of using this mutant is as follows: when a dicentric undergoes chromosome segregation during mitosis and Ctf13 is active, the DNA will break, but if Ctf13 is inactive, the microtubule-kinetochore connection will break instead of the DNA (Figure 3A). (DNA breakage is required to form mixed colonies, since mixed colonies contain cells of multiple genotypes.) We therefore predicted that Ctf13 status will affect the frequency of mixed colonies if dicentrics are intermediates, whereas Ctf13 status will not affect the frequency of mixed colonies if dicentrics are not intermediates. To test this prediction, we grew rad9 and rad9 ctf13- 30 cells at the permissive (23˚C) or the restrictive (36˚C) temperature, determined the level of dicentrics by quantitative PCR and of instability from the frequency of mixed colonies (Fig 3B). (We found that the status of Ctf13 did not drastically affect the amount of dicentrics formed, as measured by quantitative PCR; there is a 4-fold increase in dicentrics in Ctf13-defective cells.) We found that Ctf13 status did in fact affect instability; rad9 ctf13-30 mutants had an 8-fold lower frequency of mixed colonies when grown at a restrictive temperature compared to the permissive temperature, and compared to rad9 CTF13+ strains grown at either temperature. Thus, the amount of dicentric formed was largely independent of Ctf13 activity (there is actually more dicentric in Ctf13-defective cells), but the fate of the dicentrics was dependent on Ctf13 status (fewer mixed colonies in Ctf13-defective cells). This Ctf13 result is consistent with 77

the hypothesis that the specific dicentric detected by PCR is an intermediate to instability.

We note that rad9 ctf13-30 cells have a higher frequency of chromosome loss and of allelic recombination when grown at the restrictive temperature compared to the permissive temperature, as reported previously (Figure 3B, Doheny et al. 1993). An increase in chromosome loss was expected from the model; dicentrics missegregate and are lost when the Ctf13 kinetochore connection breaks. The higher frequency of allelic recombination in ctf13-defective cells has also been reported, for monocentric chromosomes, in both ctf13- defective cells (DOHENY et al. 1993) and for cells treated with the microtubule inhibitor benomyl (Wood and Hartwell 1982). The cause of increased allelic recombination in any of these studies is unknown, but unlikely due to dicentric behavior (see discussion Figure S4).

In sum, we conclude that inverted repeats in the ChrVII403 site fuse to form a dicentric that causes the instability evident in mixed colonies. We next show that fusion of nearby inverted repeats is a general phenomenon in the yeast genome.

The pair wise fusion of many nearby inverted repeats in the yeast genome

To test if inverted repeat fusion is general, we again used a PCR-based assay. By examining the yeast genome, we find that there are about 15 sites in budding yeast that contain LTR inverted repeats flanked by tRNA genes, a geometry similar to that of the unstable ChrVII403 site (Figure 4A). We chose 5 of these sites to test for their fusion of inverted repeats. We isolated genomic DNA from cells and subjected it to PCR amplification using appropriate primers (as in Figure 2A). We developed separate sets of primers that could detect either acentrics or dicentrics, since theoretically either is equally possible (see Figure 4D). We detected DNA fragments diagnostic of acentric 78

and/or dicentric chromosomes at all five genomic sites tested (Figure 4B). To get an idea of how frequently these other nearby inverted repeats fuse, we performed quantitative PCR on two additional sites (ChrVI160, ChrX542, in addition to the ChrVII403), and found that the frequency of fusion events in Rad+ strains at all three sites was comparable (1.5, 4.7, 8.1 x 10-5 for the three sites, respectively; see Figure S5I).

By examining the DNA sequence of the fused pairs of inverted repeats, we can deduce some features on the mechanism(s) of fusion. First, we found that fusion always involves some sequence homology; fusion can occur between two sequences that share as little as 20 base pairs of homology with a one base pair mismatch (ChrV137a site, Figure S5D), or as much as 148 bases of exact homology (ChrVII403 dicentric; summarized in Figure S5H). Genetic tests described below are consistent with the involvement of only limited homology in the fusion reactions.

Second, we found that fusion of inverted repeats can happen between repeats that are as far as 5.4 kb away from each other (ChrX542 site, Figure S5H). Since this site had the largest distance between the two repeats, and still readily formed both dicentric and acentric chromosomes, it is likely that the maximum distance for fusion is larger (Figure S5I). Distance requirements for fusion are the focus of a future study.

Third, we found evidence that bears on the issue of whether one event generates both an acentric and dicentric chromosome (Figure 4C), or whether one event generates either an acentric or a dicentric chromosome, but not both (Figure 4D). We term these two options "symmetric" and "asymmetric" events, respectively. The evidence that bears on this question comes from the sequence of the fusion junctions for sites where 79

both acentric and dicentric chromosomes are formed. Surprisingly, we found that for two of them (ChrVII403 and ChrV137b) the fusion junction in the acentric product was different from the dicentric product (utilizing different bases within the repeats as depicted in Figure 4D; actual sequences in Figure S1, S5). For the ChrVII403 site, we found that 16 of 16 acentrics used the same sequence (within 8 bases of homology between S2 and S3), whereas 13 of 13 dicentrics analyzed used a different sequence (within 150 bases of homology between S2 and S3). For the third site (ChrX542), the fusion to form acentrics and dicentrics does occur in the same sequence (as in Figure 4C); so there is a formal possibility that for this site both acentrics and dicentrics are generated by a single event. Formation of acentric and dicentric chromosomes by an asymmetric event has mechanistic implications addressed in the Discussion.

Synthetic inverted repeats also fuse to form dicentrics associated with instability

To further test the generality of the fusion reaction, we generated a synthetic inverted repeat consisting of non-LTR sequences (all the events analyzed thus far contain LTR sequences). This synthetic inverted repeat also does not contain any tRNA genes that likely contribute to instability. We divided the URA3 gene into three fragments, with each pair of consecutive fragments sharing about 200 base pairs of sequence homology (designated as "UR", "RA" and "A3"; Figure 5A; Figure S6). We first inserted the URA3 modules into the ChrVII sites that form the dicentric and D7/D11 translocation (the 403 and 535 sites, respectively; see Figure 1B, Figure 5A). We allowed instability of these genetically-modified cells to occur, identified mixed colonies by selection for CanR, and analyzed cells in the mixed colonies for URA3-mediated events. Overall, we found that the URA3-mediated events were virtually identical to the LTR-mediated events. First, we found that the URA3 sequence-containing inverted repeats formed dicentric chromosomes, just as the LTR sequences S2 and S3 had (Figure 5B). Second, the URA3- 80

dicentric intermediate was frequently resolved to form a translocation (Figure S7), just as with the LTR-containing dicentric. As discussed further below, we also found that the genetic requirements for fusion of the URA3-based inverted repeats is the same as for fusion of the S2 and S3 repeats (neither reaction requires the RAD52 nor RAD59 genes; Figure 5B, and see below).

We also used the synthetic inverted repeat module to test if fusion is for some reason constrained to occur in the ChrVII403 site. We inserted the synthetic inverted repeat module into two additional ChrVII sites (the ChrVII110 and ChrVII461 sites) that do not contain LTR fragments, do not contain tRNA genes, and are in completely different regions of the chromosome (one 110 kb from the telomere, one 37 kb from the centromere). We in fact were able to identify URA3-containing dicentrics formed at both sites (Figure 5B), confirming that nearby inverted repeat fusion is a general phenomenon. These studies show that fusion of nearby inverted repeats is not restricted to types of DNA sequences, nor to sequences near tRNA genes, nor to specific regions of a chromosome. (We also identified Ura+ translocations, formed by resolution of the dicentric, from Ura-modules inserted at the ChrVII461 site but not the ChrVII110 site, a result for which we have no explanation).

Finally, we used the synthetic inverted repeat module to re-examine how much sequence homology is required to form dicentrics. We found that 60 bases of homology, but not 20 or 0, allowed dicentric formation (Figure 5B). In sum, the lengths of homology involved in LTR and URA3 inverted repeat fusion is consistent with homologies of 60 bases and greater and in some cases as few as 20bp undergoing fusion.

Major DNA repair and replication fork pathways and the mechanism of inverted repeat fusion

We next use genetic analyses to address questions of mechanism; how does the cell normally prevent such a deleterious fusion reaction, and when not prevented, how does the fusion reaction then occur? We first considered the possibility that fusion of inverted repeats occurs by the same mechanisms by which other types of repeats recombine. The mechanisms of recombination between direct repeats (Thomas and Rothstein 1989), between inverted repeats (Bai and Symington 1996) and between sister chromosomes (PAULOVICH et al. 1999) has been thoroughly studied in budding yeast. The key observations in those studies are that recombination is decreased over 100-fold by mutations in either RAD52 or RAD52 and RAD1. Rad52 is a single strand annealing protein required for most homologous recombination reactions (Krogh and Symington 2004) and Rad1 is a nuclease with roles in homologous recombination and nucleotide excision repair. We therefore determined the frequency of dicentrics and instability at the ChrVII403 site in rad52, rad1 and rad52 rad1 mutants. Strikingly, we find that inverted repeat fusions were actually increased (over 10-fold) in rad52 and rad52 rad1 double mutants compared to wildtype (Rad+) cells (Table 1; determined from dicentric PCR or mixed colonies). (Some of the fusion events that occurred in rad52 mutants maybe Rad1 dependent since the level of dicentric in rad52 rad1 double mutants was 2-fold lower by quantitative PCR (Table 1); this does not change our overall conclusion that the inverted repeat fusion is occurring by a RAD52, RAD1 independent mechanism. We also found that rad52 and rad52 rad1 also gave rise to the specific D7/D11 translocation, a product formed when the dicentric intermediate is resolved). Other mutants shown in other studies to decrease inverted repeat inversions (e.g. rad51, rad59; (Bai and Symington 1996) also lead in our study to an increase in inverted repeat fusion at the ChrVII403 site (Table 1). We have also analyzed the roles of RAD52 and RAD1 in fusion at two other sites, ChrVI160 and ChrX542, by using quantitative PCR. 82

We found that fusion between inverted repeats occurred efficiently in rad52 and rad52 rad1 mutants; fusion products are present at nearly the levels formed in wildtype cells (Figure S5I). We do not know why the ChrVII403 site is more unstable in a rad52 mutant than the other two sites. The ChrVII403 site is at a replication terminus and that might contribute to its instability and different genetic regulation; Admire et al. 2006). We therefore conclude that inverted repeat fusion occurs by a mechanism that is very different from "conventional" recombination between inverted and direct repeats (forming inversions or deletions, respectively).

To further address the mechanisms that underlie inverted repeat fusion, we have focused on the observation that fusions are increased when DNA replication forks stall (either by lowering dNTPs, by defects in a DNA helicase or by tRNA gene transcription; Admire et al. 2006; HJ, TW, unpublished data). We therefore suggest that inverted repeats fuse by a mechanism related to stalled replication forks and perhaps how they recover. To guide our studies we provide a working model of replication fork biology (Atkinson and McGlynn 2009). A stalled replication fork (top) may resume replication (not shown), undergo a DSB (left), or undergo sister-strand annealing reactions to form a regressed fork (center) or a hemicatenane (right). Some outcomes from these structures are shown, and others are possible as well (e.g. the regressed fork may be cleaved by a Holliday junction resolvase to form a DSB; (Atkinson and McGlynn 2009).

To begin to test if any of these pathways are relevant to the fusion of inverted repeats to form dicentrics, we measured the frequency of mixed colony formation in a number of mutants thought to have roles in replication fork recovery. The frequency of mixed colonies is a suitable quantitative surrogate for the frequency of dicentric formation (see Figure 2), and for some mutants we also performed quantitative dicentric PCR 83

(Table 1). Our genetic and molecular analyses uncovered several relationships between replication fork recovery pathways and nearby inverted repeat fusion. First, as noted above, we found that the Rad52 protein was not required for dicentric formation; fusion of inverted repeats is Rad52-independent. Second, we found that defects in other end joining pathways such as non homologous end joining (NHEJ), micro-homology mediated end joining (MMEJ), and single strand annealing (SSA) had little effect on instability as single mutants (Ku70, Lig4, Rad1; NHEJ/MMEJ, SSA annealing; (Paques and Haber 1999). In strains where events occur more frequently (i.e. rad9 mutants), these events do not require any of these genes (i.e. rad9 and rad9 lig4 mutants have similar phenotypes). Inverted repeat fusion also does not require either Rad50, a component of the MRX complex which is required for both NHEJ and MMEJ, nor Mus81, a prominent nuclease. Indeed fusion events increase approximately 30-fold in both of these mutants, implying that they have a prominent role in the prevention of fusion events. These results suggest that fusion of inverted repeats coming from stalled forks has little to do with NHEJ, MMEJ or SSA repair pathways.

We next tested the possible role of hemicatenane in inverted repeat fusion (see Figure 6). These structures are proposed to involve Rad51 and Rad18, among other genes. Currently it is believed that Rad18 and Rad51 promote a template switch event to facilitate formation of a hemicatenane that is resolved by Sgs1 and Top3 (Branzei et al. 2006, 2008). We found that mutations in RAD18, RAD51, RAD52 and SGS1 significantly increased instability. These results suggest two things: first, these error free recovery pathways limit nearby inverted repeat fusion, and second that these genes are not required to fuse inverted repeats.

We also tested the possible role of Srs2, a protein that removes and therefore inhibits

Rad51 from acting on single-stranded DNA (VEAUTE et al. 2003). We found that a mutation in SRS2 suppressed instability in a rad9 background ( rad9 srs2 has lower instability thanrad9). We wondered whether this suppression is due increased Rad51 activity in a srs2 mutant. We therefore compared the instability of rad9, rad9srs2, rad51, rad9rad51, rad51srs2, and rad9 srs2 rad51 triple mutant. In a srs2 mutant that is Rad51 plus instability was low while in srs2 mutant where rad51 was mutant instability was as high as rad51 single mutant; rad51 is epistatic to srs2. In sum, regulation of the activity of Rad51 by Srs2 is important in preventing nearby inverted repeat fusion.

Rad59

In addition to Rad51, the single-strand annealing protein Rad59 is thought to define a second distinct Rad52-dependent HR pathway that mediates annealing between sequences with shorter regions of homology (Spell and Jinks-Robertson 2003; Pannunzio et al. 2008). We found that strains lacking Rad59 showed an increase in mixed colony formation, and readily formed both dicentric and acentric chromosomes (Table 1, Figure S8); therefore, Rad59 acts like Rad52 and Rad51 to normally prevent inverted repeat fusion. We did make one surprising observation studying Rad59 that bears on how the dicentric is resolved. We found that Rad59 is required for resolution of the dicentric (involving recombination between D7 and D11 to from the D7/D11 translocation); neither rad59 nor rad9 rad59 double mutants generated this translocation (Table 1) while both mutants generated dicentric chromosomes. Events associated with the URA3 synthetic module also obey the same genetics as LTR repeat recombination in all respects: the “URA”-containing inverted repeats fuse to form the dicentric in the absence of RAD59, but the resolution of the dicentric to form a URA3-containing 85

translocation is reduced 200 fold in rad9 rad59 mutants (Figure 5B, compare rad9 to rad9 rad59).

Pol32 and BIR-mediated events

There is increasing evidence that links replication stress to microhomology-mediated fusion events (reviewed in Hastings et. al. 2009a). A mechanism termed microhomology mediated break induced replication (MMBIR) has been proposed to resolve stalled forks by mediating fusion events (reviewed in Hastings et. al. 2009b). Both homology-driven and microhomology-driven BIR require a non-essential subunit of DNA polymerase delta, Pol32 (Payen et al. 2008; Lydeard et al. 2007). We thus tested if Pol32 was required for inverted repeat fusion, and for resolution of the dicentric to form the translocation. We found that instability is actually increased in pol32 mutants, and in rad9 pol32 double mutants when compared to Rad+ and rad9 strains. We therefore infer that neither Pol32, nor BIR events, are involved in either fusion of inverted repeats to form the dicentric, or in resolution of the dicentric to form the D7/D11 translocation.

Conclusion of genetic analysis of inverted repeat fusion

In conclusion, our genetic studies first indicate that inverted repeat fusion occurs by a mechanism distinct from repeat recombination events previously reported in budding yeast. Second, fusion is prevented by genes proposed to act in replication fork recovery to form hemicatenanes (RAD51, RAD52, and RAD18). Third, fusion does not require genes that are involved in double strand break repair, nor in BIR, nor in hemicatenane formation (RAD1, RAD50, RAD51, RAD59, RAD52, yKU80, LIG4, POL32, and RAD18). These genetic results rule out several models, and lead us to speculate on a replication- based model addressed in the discussion. 86

DISCUSSION

In this paper, we report and discuss below two major findings. First, we find that nearby inverted repeats fuse to form either dicentric or acentric chromosomes, and that dicentric formation is linked to chromosome instability. This fusion is general; it happens at multiple sites in the yeast genome, and at a synthetic site placed at three different locations on a single chromosome. Second, fusion of nearby inverted repeats is suppressed by many DNA repair pathways, and when it does occur fusion is surprisingly independent of DSB repair and other major repair pathways. These and other results lead to a replication-based model of how fusions occur.

Inverted repeat instability in budding yeast

Inverted repeats represent potential sites of instability in both prokaryotic and eukaryotic genomes. Here, we found that nearby inverted repeats, consisting of either LTR repeats or synthetic inverted repeats, fuse to form acentric or dicentric chromosomes. (This reaction is mechanistically very different from an inversion between inverted repeats, so we have used the phrase “inverted repeat fusion”.) A search of the literature indicates that nearby inverted repeat fusion is an event that occurs in bacteria and in fission yeast as well (see below). The extent to which nearby inverted repeats destabilize the budding yeast genome is not yet clear, and will depend in part on defining sequence parameters required for fusion, parameters that may involve chromosome topology as well. We do find that the fusion events in budding yeast occur between inverted repeats with as few as about 20 bases of homology and separated by approximately 1.3 kb to 5.5 kb of DNA (Figure S5I). The frequency of inverted repeat fusion is about 1 in 100,000 wild type cells at the ChrVII403 site to 8 in 87

100,000 at the ChrX542 site. The inverted repeats in ChrVII403 do appear to cause most of that chromosome’s instability (H.J, T.W, unpublished data); other inverted repeats on ChrVII probably also cause instability, though they have not yet been tested. Sites of inverted repeats also appear to be sites of translocation events that led to speciation of Saccharomyces species (Admire et al. 2006; unpublished observation).

Inverted repeat fusion and instability in other genomes

After we identified inverted repeat fusion in budding yeast, we examined the literature seeking similar observations in other organisms. Inverted repeat fusions similar to that described in this study have been reported in two studies in bacteria and two in fission yeast. In particular, in one bacterial study nearby inverted repeats present on a plasmid were found to undergo a complex rearrangement that did not involve RecA (Rad51- ortholog; (Bi and Liu 1996), and in a second study DNA intermediates were identified that also appeared to be undergoing inverted repeat fusion (Ahmed and Podemski 1998). The geometry and RecA independence of these reactions suggest to us they may occur by a similar mechanism as the events in budding yeast (see below for discussion of mechanism). Finally a third study of instability in fission yeast also appears due to nearby inverted repeat fusion. (ALBRECHT et al. 2000) selected for the amplification of Sod2 that confers resistance to lithium chloride, and isolated strains that contained high copies of acentric chromosomes. Examination of the acentric suggests it was formed by fusion of two nearby inverted repeats. Finally, an accompanying report by Muzino et al. (2009) also shows that nearby inverted repeats can form acentric and dicentric chromosomes in fission yeast.

Similarities and differences in instability associated with palindromes and nearby inverted repeats Palindromes are a class of inverted repeat sequences separated by very few base pairs of DNA. Nearby inverted repeats can be separated by many kilobases of DNA. Both of these classes of repeat sequences lead to formation of dicentric and acentric chromosomes, but how they do so appears to differ. In budding yeast, Lobachev and colleagues have found that palindromes formed a hairpin or cruciform structure that rearranges to form acentric and dicentric chromosomes (Voineagu et al. 2008; Narayanan et al. 2006). Similarly, Lemoine and colleagues also found that palindromes cause instability, presumably also by formation of a cruciform (Lemoine et al. 2005). In Schizosaccharomyces pombe, 80-bp perfect inverted repeats form hairpin and cruciform structure and confer mitotic instability (Farah et al. 2002). Due to their unstable nature, palindromes are under-represented in the human genome (Lobachev et al. 2000). Though the nearby inverted repeat fusion events reported here also lead to the formation of dicentric and acentric chromosomes, they must do so by a different mechanisms since the repeats are separated by many kilobases of DNA and hence unlikely to form hairpin or cruciform structures (LOBACHEV et al. 2000). replication-based mechanism for fusion of nearby inverted repeats

The mechanism of inverted repeat fusion remains speculative. What we do know is that many repair and checkpoint control pathways normally suppress fusion of nearby inverted repeats (Admire et al. 2006, Figure 6, Table 1). It is also clear that fusions occur by a mechanism very different from those mediating recombination between inverted and direct repeats, events that can be stimulated by DSBs (Sugawara and Haber 1992; Bai and Symington 1996; Fasullo et al. 2001). Whatever the mechanism of inverted repeat fusion is, it should account for (1) the geometry of the fusion reaction, (2) lack of involvement of Rad52 and other repair proteins, (3) increased fusion of inverted repeats upon disruption of DNA replication, and (4) an asymmetric mechanism of fusion. 89

To account for these observations, we suggest a replication-based mechanism we call faulty template switching (Figure 7). This mechanism bears similarity to mechanisms proposed from studies in bacteria (Long and Kreuzer 2008); Atkinson and McGlynn 2009). In this model, a stalled fork fails to be rescued by error-free pathways (the Rad51, Rad18-dependent hemicatenane pathway, for example; see Figure 6), and undergoes a different template switch event (Figure 6; fork regression, replication and reversal) to form a so-called “chicken-foot” structure. We suggest that during fork restart, the nascent DNA strand may either undergo fork reversal (not shown), pair with the correct sequence and be resolved by an error-free mechanism (as summarized by Atkinson and McGlynn 2009), or pair with an incorrect, nearby sequence (faulty template switch). This chicken-foot based faulty template switch reaction appears to be Rad18 independent. (Rad18 has been proposed to be involved in hemicatenane based template switch (Fig 6; Branzei et al. 2008)). How the pairing reaction occurs is unclear, since strand annealing proteins (i.e. Rad52) is not required. A speculative strand annealing reaction without Rad52 has been recently discussed (Hastings et al. 2009). Once strand annealing has occurred, completion of DNA replication leads fairly directly to formation of a dicentric (or acentric) chromosome.

Though speculative, this model illustrates how a relatively simple error in fork restart (selection of the wrong sequence) can lead to a chromosome change. The model makes specific testable predictions; we hypothesize that faulty template switching requires a 5’-3’ exonuclease for degradation of the regressed fork, and an endonuclease to resolve the structure after annealing to non-allelic sequence. Further work linking gene function to DNA structure will allow us to further test how inverted repeats fuse. Models having a DSB as an intermediate are also plausible.

Other examples of possible template switching Models involving some form of template switching have been proposed previously. In bacteria, a faulty template switch model was proposed to explain rearrangements between direct repeats (Goldfless et al. 2006). In that study rearrangements were independent of RecA (a Rad51 ortholog), as inverted repeat fusion events reported here. (In the bacterial study, rearrangements required a gene (DnaK) known to interact with DNA polymerases.) Fusion of nearby inverted repeats in bacteria (mentioned above) also suggest a template switch mechanism ((Bi and Liu 1996); (Ahmed and Podemski 1998). In yeast, template switching can occur during BIR (break induced replication), and is thought to be due to the unstable nature of replication forks during BIR (Smith and Symington 2007, Smith and Symington 2009, lydeard and Haber 2007). Smith and Symington showed switching between allelic and non-allelic sequences, though those events require proteins involved in homologous recombination. Finally, a template switch reaction has been suggested to explain complex rearrangements in a human disorder (Lee et al. 2007). The template switch mechanism they propose involves fusion between sequences sharing only very short stretches of sequence homology, seemingly an event that would not involve strand-annealing activities (Hastings et al. 2009). The extent to which a faulty template switch model between non- allelic sequences might explain rearrangements in budding yeast remains to be determined (Payen et al. 2008; Myung and Kolodner 2001). 91

Clearly, understanding the mechanism of template switching is a technical challenge, in part given how infrequently these events occur. Recently, Muzino et al. (2009) described a system that may permit dissection of replication fork stalling and template switching. Their results bear similarities to those reported here in two aspects; they found that replication forks stall in nearby inverted repeats, and that stalling causes the fusion of those inverted repeats to form acentric and dicentric chromosomes. In addition, Muzino and colleagues (2009) fail to detect any DSBs as intermediates to fusion, and they suggest a template switch model with some similarities to that shown in Figure 7. However, the genetic requirements of those events in fission yeast seem are similar in some aspects (events are independent of Lig4, Ku70) but differ in others (85% of the events require genes involved in homologous recombination), thus the mechanistic similarities are as yet unclear. The role of genes in homologous recombination lead Muzino et al. (2009) to propose that following a template switch, a Holliday structure is formed and resolved to generate symmetric products. We do not have an explanation for the differences, though it’s possible that initiation of faulty template switch events is similar but resolution of intermediates may differ.

The role of Rad59

We also report here the surprising finding that Rad59 has a role clearly independent of Rad52. Specifically, we found that resolution of the dicentric to the D7/D11 translocation required Rad59, but not Rad52 nor Rad51 (See Figure 1B; Table 1). This result was surprising because previous genetic analyses suggested that Rad52 is required for Rad59 function (Bai and Symington 1996, Davis and Symington 2001). However, recently Pannunzio and colleagues (2008) found that chromosomal rearrangements in rad52 mutants were partially dependent on Rad59. Several other 92

studies have shown that dicentric chromosomes are often resolved by translocations involving repetitive elements (Surosky et al. 1985; Haber and Thorburn 1984; Pennaneach and Kolodner 2009). It is not clear whether these translocations require Rad59. The Rad59-dependent resolution of the dicentric chromosome will be the focus of a further study.

Broader Implications for Genome Stability in Mammalian Genomes

It is clear many eukaryotic genomes harbor nearby inverted repeats that might destabilize their chromosomes. The human genome contains a high frequency of large inverted repeats with high sequence homology (Warburton et. al. 2004). Wang and Leung (2009) estimated the frequency of inverted repeats of >30bp in length to be around 1 in 10,000bp in the human genome The extent to which inverted repeats fuse or cause instability in mammalian genomes is unknown. A recent article by Lange and colleagues found that fusion of large inverted repeats (0.3 to 2.9 Mb arms separated by 2.1 – 169 kb) on the human Y chromosome leads to the formation of dicentric chromosomes (Lange et al. 2009). Patients that harbored dicentric chromosomes exhibited spermatogenic failure, and in some cases Turner syndrome, a sex disorder that is characterized by “XO” mosaicism. The authors hypothesize that the “XO” mosaicism seen in at least some of these patients is due to an unstable dicentric Y chromosome. Interestingly it has recently been shown that individuals with Turner syndrome have a higher incidence of cancer than the general population (Schoemaker et al. 2008). It is not yet known whether the increased cancer risk is in fact due to the existence of an unstable dicentric chromosome. Also the extent to which these fusion events are caused by a faulty template switching mechanism, as opposed to a crossover event between sister chromatids, is unclear.

Acknowledgements: We thank F. Spencer for plasmids, and Sue Lovett for bringing to our attention the bacterial literature on template switching. T.W was funded by an NACRP/NCI grant and NIH 1R01 GM076186, S.K. was supported by GM008659-11A1, and A.P. was funded by NSF-IGERT grant DGE-0114420. We also thank JMST for useful discussions. H.J. was supported by Award Number F31GM087120 from the National Institute of General Medical Sciences.

Materials and Methods

Yeast Strains. Yeast strains are derivatives of the A364a strain described in previous studies and references therein (Weinert and Hartwell 1990), and are derived from TY200 or TY206 by transformation. TY200 is a ChrVII disome generated by T. Formosa, M. Carson and L. Hartwell that is MATα +/hxk2::CAN1 lys5/+ cyhr/CYHs trp5/+ leu1/+ Centromere ade6/+ +/ade3, ura3-52. TY206 harbors a rad9 null mutation, rad9::ura3, described by Weinert and Hartwell (1990). Derivatives were made by transformation (Gietz and Woods 2001) of appropriate DNA fragments, and genotypes were verified by Southern analysis, PCR, and/or by phenotype. Mutant DNA fragments were generated by PCR amplification of KanMX4-marked mutations from the Euroscarf strain using primers that flank each gene. Candidate mutants were analyzed by PCR using outside primers and restriction digests were performed in cases where mutant and wildtype alleles had similar-sized fragments. Genotyping for all strains was unambiguous. In some cases, we introduced a URA3 allele to replace the KanMX4 allele. The rad1 mutation was generated by PCR amplifying the hygromycin gene from pAG32 with primers that included 45 base pair of homology to DNA flanking the rad1 ORF. rad9 ctf13-30 alleles were generated by a “popin-popout” strategy. TY206 was transformed with HindIII- linearized pBFS218 ctf13-30 (gift of F. Spencer) under selection for Ura+. We then isolated 5-Fluoroorotic acid-resistant derivatives (“popouts”) that were screened for 94

temperature sensitivity at 37°C. Further details of strain constructions are available upon request.

The URA3 inverted repeat module was constructed as shown in Figure S5. Briefly, fragments of the URA3 gene were joined to either the Nat1 gene (RA:NatR:RU cassette) or to KanMX4 (A3:KanMX4). These cassettes were inserted into plasmids that contain ~500bp of sequence flanking sites 110, 403, 461, or 535. Each module was inserted into ChrVII by restriction digestion, transformed into cells, and inserts selected by growth on drug-containing plates. Candidate insertions were confirmed by genetic analyses. For example, the rad9 430:RA:NatR:RU cassette is linked to the CAN1 gene and resides genetically between TRP5 and CYH2, mapped by analysis of allelic recombinants. Canavanine (Can; 60ug/ml), geneticin (100 or 500ug/ml), hygromycin B (300ug/ml) and nourseothricin (Nat; 50ug/ml) were used at the concentrations given.

Chromosome instability assays

Genetic analysis of ChrVII instability was performed as described in Admire et al. (2006). Briefly, we grew cells on synthetic media lacking adenine to ensure selection for both ChrVII homologs. We then plated cells on rich media (YEPD and 2% dextrose) and after 2-3 days of growth at 30˚C, identified single colonies that contained 1-3 x 106 cells.

Synthetic media were as described (SHERMAN et al. 1986). To identify the types of chromosome changes that occurred as cells grew on rich media, we resuspended individual colonies in water, counted, and plated on selective media consisting of either synthetic media supplemented with 60ug/ml Can and all essential amino acids except arginine and serine, or the same selective media also lacking adenine (Ade). Cells plated on selective media containing Ade identify chromosome loss events as colonies that prove to be CanR Ade- after replica plating on media lacking Ade (see Figure 1A). Cells plated on selective media without Ade identifies allelic recombinants and mixed 95

colonies (both CanR Ade+). We scored the morphology of colonies as either "round", which proved generally to be stable allelic recombinants, or "sectored/mixed", which proved generally to contain cells of many different genotypes. We performed "lineage assays" to confirm that cells in allelic recombinants contained the same genotype (typically >95% of the cells had the same genotype), and that cells in mixed colonies contained multiple genotypes (see Admire et al. 2006), including chromosome loss and allelic recombinants in multiple genetic intervals shown in Figure 1A. In all strains, we verified that round and mixed colonies contained single or multiple genotypes, and corrected the frequencies for the rare strains where this was not the case. Frequencies were determined from analysis of at least 6 colonies, the average and standard deviations are shown. Statistical tests were performed using the Kruskal-Wallis method (Kruskal and Wallis 1952).

Molecular analysis of altered chromosomes.

We analyzed CanR Ade+ allelic and mixed colonies for the presence of the specific translocation. CanR Ade+ cells were grown in 2-3mls of rich media, and genomic DNA was prepared and subjected to PCR amplification using ChrVII403-535 primers (Figure S11 for coordinates). We have sequenced the 1.1 kb fragment generated by the translocation to confirm it has the recombination junction shown in Figure 1B (ADMIRE et al. 2006a). We typically analyzed 4 independent cultures by PCR to judge if the translocation was generated. For rad59 mutants that fail to make the translocation, we have analyzed over 30 mixed colonies. We also used a PCR assay to detect dicentric or acentric chromosomes formed at natural inverted repeats or in the URA module. Again, genomic DNA was isolated from cells grown in rich media, and subjected to PCR amplification using the appropriate PCR primers (Figure S11). We confirmed that the URA module generates Ura+ translocations, using both a PCR assay of the 96

recombination junction (data not shown) and pulse field gels probed with a ChrVII specific probe (YGL250; Figure S6). For pulse field gels, chromosomes were separated (Iadonato and Gnirke 1996) using conditions that optimize for separation of 1100 kb (native) and 1200 kb (translocation) chromosomes.

Real-time PCR quantification

The level of dicentric chromosome was analyzed on a LightCycler 1.0 (Roche Diagnostics, Mannheim, Germany). A 20 µl amplification mixture contained 2 µl of template DNA, 3 mM MgCl2, 5 µM primers (Figure S11) and LightCycler FastStart Master SYBR Green I Mix (Roche Diagnostics). The thermal cycling conditions consisted of 10 min at 95°C, 40 cycles 15s at 95°C, 10s at 58°C and 90s at 72°C. For quantification, standard curves were prepared by serial 15-fold dilutions of PCR purified ChrVII403dicentric DNA fragment. We can detect between 5 and 106 dicentric molecules. Crossing points, representing the PCR cycles at which products were detectable, were determined for each sample using the second derivative maximum method (LightCycler 3.5.3 software, Roche Diagnostics). Quality control was ensured by a non-template control and determination of the melting point of each PCR product. Only samples in exact concordance with the melting point of a sequence-proofed amplicon were included. In each LightCycler run, the calibrator and unknown sample were measured in duplicate. Variability of the real-time PCRs was 6.3%.

REFERENCES

Admire, A., Shanks, L., Danzl, N., Wang, M., Weier, U., Stevens, W., Hunt, E., and Weinert, T. 2006. Cycles of chromosome instability are associated with a fragile site and are increased by defects in DNA replication and checkpoint controls in yeast. Genes and Development 20, 159- 173.

Ahmed, A., and Podemski, L. 1998. Observations on template switching during DNA replication through long inverted repeats. Gene 223, 187-194.

Albrecht, E. B., Hunyady, A. B., Stark, G. R., and Patterson, T. E. 2000. Mechanisms of sod2 gene amplification in Schizosaccharomyces pombe. Mol Biol Cell 11, 873-886.

Atkinson, J., and McGlynn, P. 2009. Replication fork reversal and the maintenance of genome stability. Nucleic Acids Res 37, 3475-3492.

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

Bi, X., and Liu, L. F. 1996. DNA rearrangement mediated by inverted repeats. Proc Natl Acad Sci U S A 93, 819-823.

Bignell, G.R., et al. 2007. Architectures of somatic genomic rearrangement in human cancer amplicons at sequence-level resolution. Genome research 17, 1296-1303.

Branzei, D., Sollier, J., Liberi, G., Zhao, X., Maeda, D., Seki, M., Enomoto, T., Ohta, K., and Foiani, M. 2006. Ubc9- and MMS21-Mediated Sumoylation Counteracts Recombinogenic Events at Damaged Replication Forks.

Branzei, D., and Foiani, M. 2007. Interplay of replication checkpoints and repair proteins at stalled replication forks. DNA Repair Amst 6, 994-1003.

Branzei, D., Vanoli, F., and Foiani, M. 2008. SUMOylation regulates Rad18-mediated template switch. Nature 456, 915-920.

Branzei, D., and Foiani, M. 2009. The checkpoint response to replication stress. DNA Repair Amst.

Bruder, C.E., et al. 2008. Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet 82, 763-771.

Carson, M.J., and Hartwell, L. 1985. CDC17: an essential gene that prevents telomere elongation in yeast. Cell 42, 249-257.

Chen, C., and Kolodner, R. D. 1999. Gross chromosomal rearrangements in Saccharomyces cerevisiae replication and recombination defective mutants. Nat Genet 23, 81-85. 98

Cotta-Ramusino, C., Fachinetti, D., Lucca, C., Doksani, Y., Lopes, M., Sogo, J., and Foiani, M. 2005. Exo1 processes stalled replication forks and counteracts fork reversal in checkpoint-defective cells. Mol Cell 17, 153-159.

Doheny, K.F., Sorger, P.K., Hyman, A.A., Tugendreich, S., Spencer, F., and Hieter, P. 1993. Identification of essential components of the S. cerevisiae kinetochore. Cell 73, 761-774.

Dujon, B. 2006. Yeasts illustrate the molecular mechanisms of eukaryotic genome evolution. Trends Genet 22, 375-387.

Farah, J. A., Hartsuiker, E., Mizuno, K., Ohta, K., and Smith, G. R. 2002. A 160-bp palindrome is a Rad50.Rad32-dependent mitotic recombination hotspot in Schizosaccharomyces pombe. Genetics 161, 461-468.

Fasullo, M., Giallanza, P., Dong, Z., Cera, C., and Bennett, T. 2001. Saccharomyces cerevisiae rad51 mutants are defective in DNA damage-associated sister chromatid exchanges but exhibit increased rates of homology-directed translocations. Genetics 158, 959-972.

Gietz, R.D., and Woods, R.A. 2001. Genetic transformation of yeast. BioTechniques 30, 816-820, 822-816, 828 passim.

Goldfless, S.J., Morag, A.S., Belisle, K.A., Sutera, V.A., Jr., and Lovett, S.T. 2006. DNA repeat rearrangements mediated by DnaK-dependent replication fork repair. Molecular cell 21, 595- 604.

Haber JE, Thorburn PC. 1984. Healing of broken linear dicentric chromosomes in yeast. Genetics 106, 207-226.

Hackett, J.A., and Greider, C.W. 2003. End resection initiates genomic instability in the absence of telomerase. Mol Cell Biol 23, 8450-8461.

Hastings PJ, Lupski JR, Rosenberg SM, Ira G. 2009. Mechanisms of change in gene copy number. Nature Rev Gen. 10(8):551-64. Hastings PJ, Ira G, Lupski JR. 2009. A microhomology-mediated break-induced replication model for the origin of human copy number variation. PLoS Genetics. 5(1):e1000327.

Jinks-Robertson, S., Michelitch, M., and Ramcharan, S. 1993. Substrate length requirements for efficient mitotic recombination in Saccharomyces cerevisiae. Mol Cell Biol 13, 3937-3950.

Kendall, M. G., & Gibbons, J. D. 1990. Rank Correlation Methods 5th ed.. London: Edward Arnold.

Iadonato, S.P., and Gnirke, A. 1996. RARE-cleavage analysis of YACs. Methods Mol Biol 54, 75-85.

Kim, J.M., Vanguri, S., Boeke, J.D., Gabriel, A., and Voytas, D.F. 1998. Transposable elements and genome organization: a comprehensive survey of retrotransposons revealed by the complete Saccharomyces cerevisiae genome sequence. Genome research 8, 464-478. 99

Klein, H. L. 1988. Different types of recombination events are controlled by the RAD1 and RAD52 genes of Saccharomyces cerevisiae. Genetics 120, 367-377.

Kolodner, R.D., Putnam, C.D., and Myung, K. 2002. Maintenance of genome stability in Saccharomyces cerevisiae. Science 297, 552-557.

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, 305-309.

Krogh, B. O., and Symington, L. S. 2004. Recombination proteins in yeast. Annu Rev Genet 38, 233-271.

Kruskal, W.H., and Wallis, W.A. 1952. Use of ranks in one-criterion variance analysis. Journal of American Statistical Association 47, 583-621.

Lee, J.A., Carvalho, C.M., and Lupski, J.R. 2007. A DNA replication mechanism for generating nonrecurrent rearrangements associated with genomic disorders. Cell 131, 1235-1247.

Lemoine, F.J., Degtyareva, N.P., Lobachev, K., and Petes, T.D. 2005. Chromosomal translocations in yeast induced by low levels of DNA polymerase a model for chromosome fragile sites. Cell 120, 587-598.

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

Lo, A.W., Sabatier, L., Fouladi, B., Pottier, G., Ricoul, M., and Murnane, J.P. 2002. DNA amplification by breakage/fusion/bridge cycles initiated by spontaneous telomere loss in a human cancer cell line. Neoplasia New York, NY 4, 531-538.

Lobachev, K. S., Stenger, J. E., Kozyreva, O. G., Jurka, J., Gordenin, D. A., and Resnick, M. A. 2000. Inverted Alu repeats unstable in yeast are excluded from the human genome. Embo J 19, 3822- 3830.

Lobachev, K.S., Gordenin, D.A., and Resnick, M.A. 2002. The Mre11 complex is required for repair of hairpin-capped double-strand breaks and prevention of chromosome rearrangements. Cell 108, 183-193.

Lobachev, K.S., Rattray, A., and Narayanan, V. 2007. Hairpin- and cruciform-mediated chromosome breakage: causes and consequences in eukaryotic cells. Front Biosci 12, 4208- 4220.

Long, D. T., and Kreuzer, K. N. 2008. Regression supports two mechanisms of fork processing in phage T4. Proc Natl Acad Sci U S A 105, 6852-6857.

Lydeard JR, Jain S, Yamaguchi M, Haber JE. 2007. Break-induced replication and telomerase- independent telomere maintenance require Pol32. Nature. 448(7155):820-3. 100

McClintock, B. 1941. The Stability of Broken Ends of Chromosomes in Zea Mays. Genetics 26, 234-282.

Maser, RS , and RA DePinho. 2002. "Connecting Chromosome, Crisis, and Cancer. " 297:565-670.

Murnane, J.P., and Sabatier, L. 2004. Chromosome rearrangements resulting from telomere dysfunction and their role in cancer. Bioessays 26, 1164-1174.

Myung, K., Chen, C., and Kolodner, R.D. 2001. Multiple pathways cooperate in the suppression of genome instability in Saccharomyces cerevisiae. Nature 411, 1073-1076.

Narayanan, V., Mieczkowski, P.A., Kim, H.M., Petes, T.D., and Lobachev, K.S. 2006. The pattern of gene amplification is determined by the chromosomal location of hairpin-capped breaks. Cell 125, 1283-1296.

Pannunzio, N.R., Manthey, G.M., and Bailis, A.M. 2008. RAD59 is required for efficient repair of simultaneous double-strand breaks resulting in translocations in Saccharomyces cerevisiae. DNA Repair Amst 7, 788-800.

Paques, 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.

Paulovich, A. G., Armour, C. D., and Hartwell, L. H. 1999. The Saccharomyces cerevisiae RAD9, RAD17, RAD24 and MEC3 genes are required for tolerating irreparable, ultraviolet-induced DNA damage. Genetics 150, 75-93.

Payen, C., Koszul, R., Dujon, B., and Fischer, G. 2008. Segmental duplications arise from Pol32- dependent repair of broken forks through two alternative replication-based mechanisms. PLoS Genet 4, e1000175.

Pennaneach, V. and Kolodner, R.D. 2009. Stabilization of dicentric translocations through secondary rearrangements mediated by multiple mechanisms in S. cerevisiae. PLOS one 4, 6389.

Putnam, C.D., Pennaneach, V., and Kolodner, R.D. 2004. Chromosome healing through terminal deletions generated by de novo telomere additions in Saccharomyces cerevisiae. Proceedings of the National Academy of Sciences of the United States of America 101, 13262-13267.

Putnam C.D., Hayes T.K. and Kolodner R.D. 2009 Specific pathways prevent duplication- mediated genome rearrangements. Nature, 460:984-989.

Rattray, A.J., Shafer, B.K., Neelam, B., and Strathern, J.N. 2005. A mechanism of palindromic gene amplification in Saccharomyces cerevisiae. Genes Dev 19, 1390-1399.

Richard, G.F., Kerrest, A., and Dujon, B. 2008. Comparative genomics and molecular dynamics of DNA repeats in eukaryotes. Microbiol Mol Biol Rev 72, 686-727. 101

Sabatier, L., Ricoul, M., Pottier, G., and Murnane, J.P. 2005. The loss of a single telomere can result in instability of multiple chromosomes in a human tumor cell line. Mol Cancer Res 3, 139- 150.

Schmidt, K.H., Wu, J., and Kolodner, R.D. 2006. Control of translocations between highly diverged genes by Sgs1, the Saccharomyces cerevisiae homolog of the Bloom's syndrome protein. Mol Cell Biol 26, 5406-5420.

Segurado, M., and Diffley, J. F. 2008. Separate roles for the DNA damage checkpoint protein kinases in stabilizing DNA replication forks. Genes Dev 22, 1816-1827.

Sherman, F., Fink, G.R., and Hicks, J.B. 1986. Methods in yeast genetics Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press.

Smith, CE., Llorente, B., and Symington, LS. 2007. Template switching during break-induced replication. Nature. 447(7140):102-5.

Smith, CE., Lam, AF., and Symington, LS. 2009. Aberrant double-strand break repair resulting in half crossovers in mutants defective for Rad51 or the DNA polymerase delta complex. Mol Cell Biol. 29(6):1432-41.

Surosky R.T. and Tye B.K. 1985. Resolution of dicentric chromosomes by Ty-mediated recombination in yeast. Genetics 110, 397-419.

Thomas, B. J., and Rothstein, R. 1989. The genetic control of direct-repeat recombination in Saccharomyces: the effect of rad52 and rad1 on mitotic recombination at GAL10, a transcriptionally regulated gene. Genetics 123, 725-738.

Tillier, E.R.M., and Collins, R.A. 2000. Genome rearrangement by replication-directed translocation. Nat Genet 26, 195-197.

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, 309-312.

Voineagu, I., Narayanan, V., Lobachev, K.S., and Mirkin, S.M. 2008. Replication stalling at unstable inverted repeats: interplay between DNA hairpins and fork stabilizing proteins. Proceedings of the National Academy of Sciences of the United States of America 105, 9936- 9941.

Wang, Y., and Leung, F. C. 2009. A study on genomic distribution and sequence features of human long inverted repeats reveals species-specific intronic inverted repeats. Febs J 276, 1986- 1998.

Warburton, P. E., Giordano, J., Cheung, F., Gelfand, Y., and Benson, G. 2004. Inverted repeat structure of the human genome: the X-chromosome contains a preponderance of large, highly homologous inverted repeats that contain testes genes. Genome Res 14, 1861-1869. 102

Weinert, T.A., and Hartwell, L.H. 1990. Characterization of RAD9 of Saccharomyces cerevisiae and evidence that its function acts posttranslationally in cell cycle arrest after DNA damage. Mol Cell Biol 10, 6554-6564.

Weinstock, D.M., Elliott, B., and Jasin, M. 2006. A model of oncogenic rearrangements: differences between chromosomal translocation mechanisms and simple double-strand break repair. Blood 107, 777-780.

Wong, K.K., et al. 2000. Telomere dysfunction impairs DNA repair and enhances sensitivity to ionizing radiation. Nat Genet 26, 85-88.

Wood, J.S., and Hartwell, L.H. 1982. A dependent pathway of gene functions leading to chromosome segregation in Saccharomyces cerevisiae. J Cell Biol 94, 718-726.

Zhao, X., Muller, E. G., and Rothstein, R. 1998. A suppressor of two essential checkpoint genes identifies a novel protein that negatively affects dNTP pools. Mol Cell 2, 329-340.

Zhou, Z.H., Akgun, E., and Jasin, M. 2001. Repeat expansion by homologous recombination in the mouse germ line at palindromic sequences. Proc Natl Acad Sci U S A 98, 8326-8333.

103

Figure Legends

Figure 1. Chromosome system to detect instability. A. Two homologs of ChrVII and mutant alleles on each allow for genetic detection of chromosome changes. The CAN1 gene has been removed from ChrV and inserted into one copy of ChrVII. Selection for the loss of the CAN1 gene allows growth of cells with any of three types of chromosome changes, including simple loss, allelic recombinants, and mixed colonies. Mixed colonies contain cells of multiple genotypes, including a specific translocation. See Experimental Procedures for details. B. Configuration of elements in the ChrVII403 site and the geometry and order of how fusion may occur. Two tRNA genes (pentagons) transcribe towards the oncoming fork and slow replication. Fusion between the two LTR sigma repeats (S2, S3), shown diagrammatically, forms a dicentric, followed by recombination between the two LTR delta sequences (D7, D11) to form the specific translocation.

Figure 2. Inverted repeat fusion generates dicentrics that cause further chromosome instability. A. Shown are the structures of normal ChrVII and the putative dicentric with position of PCR primers 1 and 2 used to detect the presence of the dicentric. B. Genomic DNA from unselected cultures of Rad+, rad9 and rad51 were subjected to PCR amplification using primers 1 and 2. Gels show qualitative PCR results. The rad9 cultures were also analyzed for their frequencies of mixed colonies and quantitative PCR to determine the amount of dicentric. Spearman correlation test was performed to correlate instability and dicentrics. Spearman rank correlation coefficient (rho) is 0.973, P-value <0.0001. Three of the Rad+ cultures are faintly positive for the dicentric fragment; lanes 4, 8 and 9. Primers to RSP5 were used as PCR controls.

104

Figure 3. Instability is affected by the Ctf13 protein, suggesting that dicentrics are intermediates in instability. A. Model of how Ctf13 function may affect the fate of a dicentric chromosome. See Text for discussion. B. rad9 and rad9 ctf13-30 strains were grown at permissive (23˚C) or restrictive temperatures (36˚C), then analyzed for instability events, translocations and frequency of dicentrics by quantitative PCR as described in Experimental Procedures.

Figure 4. Inverted repeats at 5 sites in the yeast genome fuse to form dicentrics and/or acentrics. A. The sites screened for fusion contain tRNA genes (pentagons), repeat sequences (grey and brown arrows are involved in the fusion reaction, whereas clear ones are not), and in one case an origin (square). Inverted repeat fusion forms either dicentric or acentric fragments using primers in schemes shown in Figure 4C. B. DNA fragments are detected at all five sites tested, using appropriate primers specific for each site. Primer sequences are in Figure S9. Analysis of three sites yielded both acentric and dicentric fragments. Asterisks indicate where DNA fragments should have appeared, were they formed. The DNA sequence of all the DNA fragments has been confirmed (Figure S5). C. If the fusion junctions for acentric and dicentric chromosomes are identical, then they could have arisen by a symmetrical mechanism. This mechanism can conceptually be viewed as a crossover depicted here; one event generates both acentrics and dicentrics. (Not shown here are stalled forks as intermediates; this diagram is meant to convey the idea of a symmetrical event only.) D. On the other hand, if the fusion junctions for acentric and dicentric chromosomes are not identical, then we infer that acentrics and dicentrics must be formed by separate events. In this case, the DNA polymerase is “jumping’ between different sequences to form acentrics than it does to form dicentrics. Note that symmetric joints can also be formed by an asymmetrical mechanism, if the DNA polymerases for the acentric and dicentric “jump” between identical sequences. 105

Figure 5. Synthetic inverted repeats fuse to form dicentrics, and then recombine to form translocations. A. Three segments of the URA3 gene were cloned into two modules (see Figure S6 for details of module construction and chromosome insertion). The two modules were then inserted into the specific sites in ChrVII as indicated. Inverted repeat recombination joins UR to RA modules via recombination of shared “R” sequences, to form a dicentric. A second recombination event joins “A” to “A3” to form an intact URA3 gene, rendering the cells Ura+. B. Each rad9 strain contains a UR or RA module in one of three sites on the left arm of ChrVII, and all strains contain the A3 module at one site on the right arm. The UR modules differ in amount of sequence homology shared with the RA module (size of "R" is 200, 60, 20 or 0 base pairs of homology, as indicated). Each modified rad9 strain was analyzed for the presence of the specific PCR fragment diagnostic of a dicentric chromosome, in four independent cultures, using dicentric URA primers (Figure S9). CanR mixed colonies were generated from each rad9 strain and the frequency of Ura+ cells in cells from mixed colonies was determined (see Experimental Procedures). **The frequency of Ura+ colonies in Cans cells is shown in Figure S8.

Figure 6. A model of the pathways that may act on stalled replication forks. DNA polymerase stalls at a lesion (triangle). The stalled fork may undergo a DSB (i) or sister- strand annealing, template switch events forming a regressed fork (ii) or hemicatenane (iii). DSBs may undergo DNA rejoining by non-homologous or microhomology mediated end joining (NHEJ, MMEJ), homologous recombination (HR) or single strand annealing (SSA), or telomere addition (not shown). Some genes tested in this study that may regulate each pathway are shown.

106

Figure 7. DNA replication-based mechanism of inverted repeat fusion. Top shows a chromosome region that contains inverted repeat sequences (red box, S2 and blue box S3), separated by several kilobases of DNA. The rightward fork stalls (not shown), then regresses to form the “chicken foot” structure with some S3 sequence at the 3’ end of the regressed strand. On the right, the regressed fork normally reinvades at the appropriate S3 sequence, forms a Holliday structure that then is cleaved and resolved to restore a fork (after Atkinson and McGlynn, 2009). On the left, the regressed fork reinvades at the wrong S2 sequence. Then, the unpaired regressed strand (dark green) is degraded (pac man), the invaded strand begins to replicate using the blue strand as template (displacing the blue strand from the red strand and allowing annealing of light green to red, forming the S2-S3 joint; dotted line). Upon completion of replication, single-strand DNA breaks must separate leftward red and blue strands, gap is filled (thin red arc, details not shown), forming the dicentric with hybrid S2-S3 repeat and one intact S2 repeat. The dicentric may then undergo further instability.

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

Figure S7: Ura+ translocations formed from the URA3 modules

A EtBr-stained pulse-field gel B Probed with YGL250 (to ChrVII right arm) C Frequencies of Ura+ 461x535 (URA) 403x535 (URA) 461x535 (URA) 403x535 (URA) 1 2 3 Cont 1 2 3 1 2 3 Cont 1 2 3

% Ura+ CanS unselected: <0.000001 R 1,200- Can allelic recomb.: <0.000001 1,200- 1,100 1,100 CanR mixed colonies: 20.9%

Molecular and genetic analysis of Ura+ translocations from rad9 URA module strains. Cells (TYXXX; rad9 403- RA:NAT1:RU, 535-A3:KanMx, TYXXX; rad9 461-RA:NAT1:RU, 535-A3:KanMx, ) analyzed for instability as discussed in figure 4. In A and B, Ura+ cells were isolated from mixed colonies and samples analyzed by pulsefield gels. The arrows indicate the position of novel chromosome bands that are of the expected sizes for Ura+ translocations. The identity of the bands were confirmed by Southerns (B) using a probe to the right arm of ChrVII. We used PCR to further confirm the structure of the translocation junction (not shown). In C, we report the frequencies of Ura+ cells from colonies of the CanS starting strain (N= 5), the Can R allelic recombinants (N=5), and from mixed colonies (N=10), all from the TYXXX; rad9 403-RA:NAT1:RU, 535-A3:KanMx.

131

132

133

TABLE S1. Saccharomyces cerevisiae strains used in this study Strain Genotypea Source TY200 RAD+Wild-type Admire et al., 2006 TY206 rad9∆::ura3 Admire et al., 2006 TY388 rad51∆::URA3 This study TY436 rad52∆::URA3 This study TY438 rad59∆::KanMX4 This study TY440 rad18∆::KanMX4 This study TY442 rad1∆::HPH This study TY444 yku80∆::KanMX4 This study TY447 lig4∆::KanMX4 This study TY359 sgs1∆::URA3 This study TY354 srs2∆::URA3 This study TY340 rad9∆::ura3 rad51∆::URA3 This study TY448 rad9∆::ura3 rad52∆::URA3 This study TY451 rad9∆::ura3 rad59∆::KanMX4 This study TY452 rad9∆::ura3 rad18∆::KanMX4 This study TY453 rad9∆::ura3 rad1∆::HPH This study TY454 rad9∆::ura3 yku80∆::KanMX4 This study TY456 rad9∆::ura3 lig4∆::KanMX4 This study TY458 rad9∆::ura3 sgs1∆::URA3 This study TY356 rad9∆::ura3 srs2∆::URA3 This study TY364 rad51∆::KanMX4 srs2∆::URA3 This study TY459 rad9∆::ura3 rad51∆::KanMX4 srs2∆::URA3 This study TY461 pol32∆::KanMX4 This study TY471 rad9∆::ura3 pol32∆::KanMX4 This study TY416 rad9∆::ura3 403::RA::nat1MX::RU 535::A3::KanMX4 (200bp)b This study TY422 rad9∆::ura3 403::RA::nat1MX::RU 1067::A3::KanMX4 (200bp) This study TY424 rad9∆::ura3 461::RA::nat1MX::RU 1067::A3::KanMX4 (200bp) This study TY426 rad9∆::ura3 110::RA::nat1MX::RU 1067::A3::KanMX4 (200bp) This study TY428 rad9∆::ura3 461::RA::nat1MX::RU 535::A3::KanMX4 (200bp) This study TY430 rad9∆::ura3 110::RA::nat1MX::RU 535::A3::KanMX4 (200bp) This study TY481 rad9∆::ura3 403::RA::nat1MX::RU 535::A3::KanMX4 (0bp) This study TY482 rad9∆::ura3 403::RA::nat1MX::RU 535::A3::KanMX4 (5bp) This study TY483 rad9∆::ura3 403::RA::nat1MX::RU 535::A3::KanMX4 (10bp) This study TY484 rad9∆::ura3 403::RA::nat1MX::RU 535::A3::KanMX4 (20bp) This study TY485 rad9∆::ura3 403::RA::nat1MX::RU 535::A3::KanMX4 (60bp) This study TY486 rad9∆::ura3 403::RA::nat1MX::RU 535::A3::KanMX4 (80bp) This study TY487 rad9∆::ura3 ctf13-30 This study TY488 rad1∆::HPH rad52∆::URA3 This study TY491 mus81::URA3 This study TY492 rad9::ura3 mus81::URA3 This study TY493 rad9::ura3 rad50::KanMX4 This study 134

aAll strains are disomic for Chr VII and derivatives of TY200 MATa +/hxk2::CAN1 lys5/+ cyhr/CYHs trp5/+ leu1/+ Centromere ade6/+ +/ade3, ura3-2 (A364a genetic background) except for mutations listed (Admire et al., 2006). bURA3 module was integrated into the CAN1 Chr VII homolog for strains TY416-TY486 and number in parenthesis represents number of base pair homology between UR and RA fragments.

135

APPENDIX B

THE ROLE OF REPLICATION BYPASS PATHWAYS IN DICENTRIC CHROMOSOME

FORMATION IN BUDDING YEAST.

Andrew L. Paek*, Hope Jones*, Salma Kaochar*, and Ted Weinert*

* Department of Molecular and Cellular Biology, University of Arizona, Tucson, AZ

85721, USA

Running Title: Genome instability, post replication repair, template switch

Corresponding Author: Ted Weinert

E-mail: [email protected]

Tel (520) 621-8511

Fax (520) 621-3709

136

ABSTRACT

This may reflect separate roles for Srs2 in different phases of the cell cycle. Third, we provide evidence for a faulty template-switch model by studying mutants of DNA polymerases; defects in DNA pol delta (lagging strand polymerase) and Mgs1 (a pol delta interacting protein) lead to a defect in fusion events as well as allelic recombination. Pol delta and Mgs1 may collaborate either in strand annealing and/or DNA replication involved in fusion and allelic recombination events. Fourth, by studying genes implicated in suppression of GCRs in other studies, we found that inverted repeat fusion has a profile of genetic regulation distinct from these other major forms of GCR formation.

137

INTRODUCTION

All organisms are prone to large scale changes (gross chromosomal rearrangements; GCRs) to their genomes that include deletions, inversions and translocations. These large scale changes are thought to drive evolutionary events such as speciation, and contribute to human pathology such as Pelziaeus-Merzbacher syndrome and other genetic disorders (LEE et al. 2007; STANKIEWICZ and LUPSKI 2010).

Thus, a firm understanding of how cells normally prevent such rearrangements, and how they accumulate, is critical to our understanding of both evolution and pathology.

GCRs arise by many different mechanisms, and there is growing evidence that errors during DNA replication are a major source (ADMIRE et al. 2006b; MIZUNO et al.

2009; MYUNG et al. 2001a). Errors are thought to arise when replication forks encounter

“lesions" on the template strand. Lesions can consist of protein complexes bound to

DNA, or lesions in the DNA itself. Replication forks bypass lesions by several different mechanisms that are still poorly understood (ATKINSON and MCGLYNN 2009; WEINERT et al.

2009). We believe that understanding lesion bypass mechanisms is central to understanding both how GCRs are prevented, and how they form when lesion bypass mechanisms fail.

All lesion-bypass pathways utilize sequence homology to restart replication

(ATKINSON and MCGLYNN 2009; WEINERT et al. 2009). Use of sequence homology during restart may limit the frequency of GCRs, as it lowers the probability of annealing to non- 138

allelic sequences. Repetitive sequences present a problem because lesion bypass at sites near repetitive sequences may lead to annealing of non-allelic sequences and thus to GCR formation (ARGUESO et al. 2008; LEMOINE et al. 2005; NARAYANAN et al. 2006).

Indeed in yeast and in other organisms, GCRs occur frequently in repeat sequences

(ARGUESO et al. 2008; DI RIENZI et al. 2009; DUNHAM et al. 2002). Some rearrangements do occur between so-called “single-copy sequences” with either no homology or limited homology (micro-homologies of 5-9 base pairs; (KOLODNER et al. 2002; MYUNG et al.

2001a; PUTNAM et al. 2005) though evidence suggests these rearrangements occur less frequently than rearrangements between repetitive sequences (PUTNAM et al. 2009).

Interestingly, it has been shown that some genes are required to prevent the fusion of repetitive elements yet have no affect on rearrangements between single-copy sequences (PUTNAM et al. 2009). Currently it is not clear how these pathways act to suppress repeat-mediated events and why they are not required to prevent rearrangements between single copy sequences.

Our current understanding of the mechanisms underlying GCR formation is mostly derived from assays designed to detect specific changes to yeast chromosomes

(ADMIRE et al. 2006b; CHEN and KOLODNER 1999; HUANG and KOSHLAND 2003; LAMBERT et al.

2005; MIZUNO et al. 2009; MYUNG et al. 2001a; NARAYANAN et al. 2006; PAEK et al. 2009;

PANNUNZIO et al. 2008; PAYEN et al. 2009; RATTRAY et al. 2005; SCHMIDT et al. 2006a; SMITH et al. 2007). Previously we reported on GCR formation in the budding yeast

Saccharomyces cerevisiae using an assay we developed. We found that a major source 139

of genome instability involves the fusion of nearby inverted repeats (with approximately

20 to 200bp of sequence homology, separated by 1 to 5 kb) to form either dicentric or acentric chromosomes (Figure 1D; (PAEK et al. 2009). We also found that fusion of inverted repeats is general: fusion occurred between inverted repeats at all five different locations tested on four different yeast chromosomes, as well as between synthetic inverted repeats (PAEK et al. 2009). Genetic data suggests that these events most likely occur during replication of DNA (ADMIRE et al. 2006b). Further genetic analysis suggested that the mechanism of inverted repeat fusion differed from that of direct repeat recombination, in that inverted repeat fusion did not require genes involved in homologous recombination (HR) or single strand annealing (SSA) pathways

(PAEK et al. 2009). In addition, fusion events are unlikely to involve double strand breaks, as genes in the non-homologous end joining (NHEJ) and micro-homology mediated end joining (MMEJ) are not required for fusion events (PAEK et al. 2009).

Indeed gene knockouts in the HR (RAD52, RAD51, RAD59), SSA (RAD52, RAD1) and post replication repair (PRR, RAD18) pathways actually increased the frequency of fusion of an inverted repeat on chromosome VII (Chr VII; (PAEK et al. 2009); these pathways normally suppress inverted repeat fusion.

To further our previous studies, we analyzed three groups of genes implicated in the maintenance of genome stability. We tested how these genes affect the overall stability of chromosome VII (Chr VII), focusing on the fusion of nearby inverted repeats to form a specific dicentric Chr VII and the resolution of the dicentric into a monocentric 140

translocation (which we term the 403-535 translocation; Figure 1 D-F). First, we analyzed several genes in the PRR pathway, and found that error free bypass, but not translesion synthesis, is required for the prevention of inverted repeat fusion.

Surprisingly, we found that siz1 mutants, which are defective for Srs2 recruitment to replication forks, and srs2 mutants had opposite effects on instability. This may reflect separate roles for Srs2 in different phases of the cell cycle. Second we analyzed several mutations in genes that are associated with replication forks. We found that mutants in

POL3 (polymerase delta) and MGS1 (encoding a single-strand annealing protein which binds polymerase delta) significantly reduced the frequency of dicentric formation and allelic recombinants that arise in the checkpoint mutant rad9 (GIOT et al. 1997; HISHIDA et al. 2001; PAEK et al. 2009). Finally we studied genes associated with rearrangements involving repeats or single copy sequences, as well as a subset of mutants involved in recombination. Generally, we find that the mechanisms of nearby inverted repeat fusion are distinct from mechanisms fusing longer repeats or single copy sequences.

MATERIALS AND METHODS

Yeast Strains: All strains are derived from TY200 and TY206 that are derivatives of the

A364a yeast strain (ADMIRE et al. 2006b; WEINERT and HARTWELL 1990). TY200 is a Chr VII disome that is MATα HXK2 lys5 cyh2 trp5 leu1 cenVII ade6 ADE3 / hxk2:CAN1 LYS5

CYH2 TRP5 LEU1 cenVII ADE6 ade3. TY206 contains a rad9 null mutation, rad9::ura3, which was made as described in Weinert and Hartwell, 1990. It is a RAD9 deletion marked with URA3, which was then subjected to FOA selection to isolate an ura3 141

mutant cell to allow use of the URA3 gene for further genetic changes. Additional null mutants were made by transformation of PCR products derived from Euroscarf strains with appropriate PCR primers and selection for resistance to geneticin (KRUSKAL and

WALLIS 1952; WACH et al. 1994). Mutations were verified by PCR and phenotypic assays

(i.e. damage sensitivity) where appropriate. The pol3-13 mutation was introduced using the two step gene replacement strategy (GIOT et al. 1997). First a pY9 plasmid (Gift of

Dmitry Gordenin) bearing the pol3-13 mutation and URA3 was digested with Hpa1 and

Ura+ colonies were selected. The native POL3 gene was removed by selecting for loss of

URA3 by plating to 5-FOA, and candidates were verified by sequencing. A similar strategy using the Ycplac22-POL30 (Gift of Anja-Katrin Bielinsky) plasmid bearing various

POL30 mutations was used to generate the PCNA mutations (HOEGE et al. 2002). For all mutants reported, two separate mutants were made and analyzed.

Chromosome instability assays: All assays were performed as described in Admire et al.

(2006). Cells were grown on YEP plates supplemented with 2% dextrose (YEPD) for 2-3 days at 30°C (unless otherwise noted). Single colonies were then resuspended in water, counted and plated onto selective medium with 60 ug/ml canavanine and all essential amino acids except arginine and serine (Can plates), and in addition, the same selective medium yet lacking adenine (Can –Ade plates). Colonies that grew on Can plates and were Ade- were scored as loss events. CanR Ade+ colonies that were round in appearance were scored as allelic recombinants, while those with a serrated or sectored appearance (see Figure 1) were scored as "mixed colonies". We identify colonies as 142

"sectored" visually, and verify their genetic content as "mixed" by lineage assays. We verified by a lineage assay (ADMIRE et al. 2006b), that each round colony consisted of cells with the same genotype (>95%), while each mixed colony contained cells of multiple genotypes. In rare cases where this was not the case, frequencies were corrected accordingly. For example if one out of ten sectored colonies proved to have a stable genotype (i.e. all cells were of one genotype), then the mixed colony frequency would be adjusted to 90% of the sectored colony frequency observed. The lineage assay also provides a means of determining where events in the mixed colonies had occurred.

We can determine where events occurred by assessing the markers on Chr VII (see genotype of Chr VII above). In all strains tested in this study, the distribution of recombination events in mixed colonies are similar to results found in Rad+ and rad9 strains. That is, the majority of events (roughly 60%) occurred in the E2 genetic region linked to the 403 fragile site lies (data not shown). Frequencies were determined from analysis of at least 12 colonies, 6 from each of two independently-derived mutants.

Statistical tests were performed using the Kruskal-Wallis method (Kruskal and Wallis

1952).

Detection of the Chr VII403-535 translocation: For each strain we screened for the presence of the Chr VII403-535 translocation in both round and mixed CanR Ade+ colonies

(See Figure 1F). Cells from CanR Ade+ colonies were grown in 3 mls YEPD, and genomic

DNA was prepared and subjected to PCR amplification using primers which detect the

Chr VII403-535 translocation (See Figure 1F). The coordinates for these primers in SGD are 143

VII 406041-406022 and VII 535582-535562. The 1.1 kb translocation band was sequenced in a previous publication (ADMIRE et al. 2006b). For each strain at least six independent CanR Ade+ colonies were analyzed by PCR to judge if the translocation was generated.

Quantification of a specific dicentric chromosome by real-time PCR: In order to determine the frequency of inverted repeat fusion to form a dicentric chromosome at the site on Chr VII that is 403 kb from the left telomere ("the 403 fragile site"), we employed a quantitative PCR assay explained previously (PAEK et al. 2009). Briefly, cells that initially contained an intact Chr VII were grown on YEPD for two days at 30°C to allow formation of dicentrics (the pol3-13 mutants were grown at different temperatures, as noted). Single colonies were then picked and grown overnight in YEPD liquid cultures to generate enough cells to isolate genomic DNA. Two separate qPCR assays were carried out on each genomic DNA preparation, one to a “normal” unrearranged DNA region of Chr V (SGD coordinates Chr V 410165-410189, Chr V 412617-

412639) to determine the relative amount of genomic DNA in the sample, and the second qPCR detected the specific dicentric (SGD coordinates VII 405579-405606 VII

402296-402319, see Figure 1E). Each qPCR reaction was done in triplicate, and for each strain six independent colonies were analyzed. Average cycle number was calculated and compared to standards which consisted of purified PCR product of known concentration. By comparing standard concentrations to samples we were able to determine the number of copies of the specific dicentric in each sample. 144

RESULTS

The system: The genetic system that we use is a haploid strain of yeast containing two copies of Chromosome VII (a Chr VII disome, Figure 1A (CARSON and HARTWELL 1985a).

Having an extra copy of Chr VII allows for changes to occur on one copy without affecting cell viability. To identify cells with an altered copy of Chr VII, we used a strain that has a copy of CAN1 on one left arm (Figure 1A; (CARSON and HARTWELL 1985a). We identified changes to this chromosome by selecting for loss of CAN1 by plating on medium containing canavanine. Canavanine is a toxic analog of arginine that is imported into the cell by the Can1 protein (arginine permease), causing cell death

(Walker 1955). Therefore, in order to survive on canavanine medium, cells must lose the CAN1 gene. This is predominantly done in one of three ways; allelic recombination

(Figure 1B), chromosome loss (Figure 1C), or non-allelic recombination that results in formation of “mixed” colonies (see mixed colony in Figure 1D; (PAEK et al. 2009).

Chromosome loss frequency is measured by the proportion of CanR Ade- cells in a colony. Allelic recombinants are those that form CanR Ade+ colonies and are round in appearance. Mixed colonies are those that are CanR Ade+ and have a sectored appearance (see mixed colony in Figure 1D; Materials and Methods).

Each “mixed” colony consists of cells with multiple genotypes (Figure 1D). We showed previously, using a genetic test that mixed colonies arise from cells with dicentrics (PAEK et al. 2009). At least one dicentric that is associated with mixed colonies 145

is formed is by the fusion of one set of nearby inverted repeats on Chr VII (figure 1E; S2 and S3 repeats in the 403 fragile site; (PAEK et al. 2009). S2 and S3 are fragments of the ends of retrotransposons (LTR fragments). The S2-S3 fusion event leads to the formation of an unstable dicentric chromosome. The dicentric can be resolved by a chromosome loss event, allelic recombination, or a non-reciprocal translocation termed the 403-535 translocation (ADMIRE et al. 2006b). We believe that any of several inverted repeats may fuse to form dicentrics, though only the S2-S3 dicentric has been positively identified, and the frequency of S2-S3 dicentrics correlates with the frequency of mixed colonies

(PAEK et al. 2009).

For each mutant strain we measured the frequency of chromosome loss, allelic recombination and dicentric chromosome formation. We test every gene in two separate mutant isolates, and the results from pairs of strains were in all cases concordant. In general we can infer the frequency of dicentric formation from the frequency of mixed colonies, because in most mutants the frequency of mixed colonies does correlate with the formation of a specific dicentric chromosome (PAEK et al. 2009).

Formation of a mixed colony from a cell containing a dicentric chromosome requires the resolution of the dicentric (Figure 1D-F), and it is therefore possible that mutants which are defective for mixed colony formation still form dicentrics but cannot resolve them.

Therefore, for some mutants we also quantified directly the frequency of the specific dicentric chromosome, present in cells before selection, using a quantitative PCR technique (Figure 1E; Material and Methods). In addition we determined if the 403-535 146

translocation was present, in mixed colonies, by a qualitative PCR technique (Figure 1F); the translocation most likely arises from recombination of the specific dicentric chromosome. In all cases thus far, mutants that form the translocation in fact form the dicentric. In all mutants reported here, strains that formed the specific dicentric also formed the specific translocation. We therefore focus here on mechanisms involved in inverted repeat fusion to form dicentrics. (We previously reported a single mutant, rad59, that formed dicentrics but failed to form the translocation; (PAEK et al. 2009).

In our analyses, we first determine how fusion events are normally prevented by measuring, in single mutants, the frequency of mixed colonies and dicentrics (of GENEX, say). If a geneX mutant has an increased frequency of instability (of specific dicentric or mixed colonies), then we infer that GENEX normally prevents fusion events. If the single mutants are stable, we conclude that GENEX is not needed to prevent instability (though

GENEX may prevent instability by a mechanism redundant with another gene). Finally, to test if GENEX has a role in forming rearrangements, we tested GENEX function in a rad9 checkpoint mutant background which has a high level of instability, and determined if events still occurred in the rad9 geneX double mutant (ADMIRE et al.

2006b). If events still occur in the double mutant, we conclude that GENEX has no detectable role in forming the rearrangements, while if events do not occur (or occur to a lesser extent), we conclude GENEX has a role in forming the rearrangement. 147

The error free branch of post-replication repair, but not translesion synthesis, prevents the fusion of nearby inverted repeats: The post replication repair pathway

(PRR) is one of the major mechanisms the cell uses to bypass lesions during DNA replication (Lee and Myung 2008). There are at least two ways in which the PRR pathway is thought to bypass barriers to replication forks (Figure 2A). These separate functions are commonly referred to as translesion synthesis and error-free bypass, and the choice between these two pathways is tightly regulated by the ubiquitylation of

PCNA by Rad6, Rad18, Ubc13 and Mms2 (HOEGE et al. 2002; LEE and MYUNG 2008; STELTER and ULRICH 2003). When PCNA is mono-ubiquitylated by a Rad6-Rad18 heterodimer, translesion DNA polymerases such as Rev3 and Rad30 are recruited to the lesion sites, and are able to replicate past these lesions often in an error-prone manner, leading to frequent point mutations (Figure 2A; (Stelter and Ulrich 2003). In contrast, the poly- ubiquitylation of PCNA by Rad5, Ubc13 and Mms2 is thought to be required for error- free bypass, a poorly understood pathway that may involve template switching, and does not lead to an increase in point mutations (BRANZEI et al. 2008; STELTER and ULRICH

2003).

We previously showed that PRR-deficient rad18 mutants have an increased frequency of mixed colony formation (PAEK et al. 2009). In order to determine whether the affect we see in rad18 mutants was due to translesion synthesis and/or error-free bypass, we analyzed mutants that are reported to be involved in either branch but not both (Stelter and Ulrich 2003). We found that mutants in the translesion synthesis 148

pathway, such as rad30 and rev7, had no effect on the frequency of mixed colony formation, in either cells with a RAD9 defect or in cells with functional RAD9 (Table 1).

Translesion polymerases therefore do not cause instability when defective, and play no detectable role in rearrangement events themselves. In contrast, we found that rad5 and ubc13 single mutants both showed a dramatic increase in mixed colonies (Table 1).

(The phenotype of these mutants with a rad9 mutation is addressed below). Taken together these data imply that the error-free bypass branch of PRR (regulated by Rad5 and Ubc13), but not the translesion synthesis branch, acts to prevent the fusion of nearby inverted repeats. These data also suggests that neither the error-prone polymerases nor the error-free proteins are involved in the proposed faulty template switch events that form dicentrics.

Mutants in the error free branch of post-replication repair show different levels of instability: Both Rad5 and Ubc13 are required to poly-ubiquitylate PCNA at lysine 164 in order to promote error-free bypass (Figure 2A; (HOEGE et al. 2002). Thus it was puzzling as to why rad5 mutants formed mixed colonies much more frequently than did ubc13 mutants (61 x 105 and 13 x 105 respectively, Table 1). The different phenotypes seen in these single mutants might be explained by either of two separate models. First Rad5 may act in both PRR and in a pathway separate from PRR, and both pathways act independently to prevent inverted repeat fusion. There is evidence for this, as Rad5 is required for ubiquitylation of lysine 107 of PCNA, while Ubc13 is not (DAS-BRADOO et al.).

Alternatively, the suppression of mixed colonies by Rad5 and Ubc13 might be exclusively 149

through error-free bypass, yet Rad5 is essential for PRR and Ubc13 plays some lesser role.

In order to formally distinguish between these two models, we made use of an allele of PCNA (pol30-K164R) which is unable to be ubiquitylated on this residue and therefore is proposed to be defective for all branches of PRR (Figure 2A; (HOEGE et al.

2002; STELTER and ULRICH 2003). If the first model is correct, where Rad5 acts in PRR and in a pathway separate from PRR, then a pol30-K164R rad5 double mutant should form mixed colonies more frequently than a pol30-K164R single mutant. In contrast, if the second model is correct and both RAD5 and UBC13 are suppressing mixed colonies through PRR, then pol30-K164R, rad5 pol30-K164R and ubc13 pol30-K164R should show the same frequencies of mixed colony formation. This was indeed the case; POI30 is epistatic to both RAD5 and UBC13 (Table 1). Together these data suggest that Pol30,

Rad5 and Rad18 all act through one pathway, that Rad5 and Rad18 have greater roles in

PRR than does Ubc13, and that all these activities suppress inverted repeat fusion. We suggest that Ubc13 may simply have a lesser role in ubiquitylation than Rad5 and Rad18

(see Discussion).

We also found that rad9 rad18, rad9 ubc13 and rad9 pol30-K164R double mutants showed a synergistic (more than additive) increase in both mixed colony formation and chromosome loss when compared to corresponding single mutants

(Table 1). In contrast rad9 rad5 mutants have approximately additive levels of mixed 150

colony formation (Table 1). It is unclear why the first three genes have different genetic interactions with RAD9 than does RAD5.

The role of Srs2, Siz1, and PCNA SUMOylation in allelic recombination and inverted repeat fusion (mixed colony formation): In a previous study, we provided evidence for a genetic interaction between SRS2 and RAD51 in mediating instability of Chr VII (PAEK et al. 2009). Srs2 is a DNA helicase which disrupts the Rad51 presynaptic filament (KLEIN

2001; KREJCI et al. 2003; VEAUTE et al. 2003). Previously, we showed that a srs2 mutant suppressed the high frequency of mixed colony formation (shown in a rad9 background;

(PAEK et al. 2009). Mutants in SRS2 also increased the frequency of allelic recombinants, as expected from previous studies (Table 1; (PAEK et al. 2009). We further showed that rad51 mutants have a high frequency of instability, suggesting that srs2 mutants decrease instability by increasing Rad51 activity. In accordance with this model rad51 mutants are epistatic with srs2 mutants, that is, both srs2 rad51 double mutants and rad9 srs2 rad51 triple mutants have the same high frequency of mixed colony formation as rad51 and rad9 rad51 mutants, respectively (Figure 2B; (PAEK et al. 2009).

We therefore decided to examine regulators that are known to permit Srs2 association with the replication fork. Srs2 is recruited to the replication fork in part via

Siz1-dependent SUMOylation (small ubiquitin-related modifier) of PCNA (Figure 2A;

(PAPOULI et al. 2005; PFANDER et al. 2005). During S-phase PCNA (Pol30) is SUMOylated by Siz1 at lysines 127 and 164 (HOEGE et al. 2002). SUMOylated PCNA then recruits Srs2 151

to replication forks, which prevents Rad51 from acting at these forks (PAPOULI et al.

2005). The absence of Rad51 at the fork may prevent ssDNA from engaging in other reactions (i.e. allelic and non-allelic recombination), and may give PRR proteins unfettered access to regulate stalled forks. We therefore tested the instability of a siz1 mutant, expecting a siz1 mutant would fail to recruit Srs2 to forks, and thus siz1 and srs2 would have similar instability phenotypes. Unexpectedly, we found that siz1 mutants showed more instability than wild-type cells, not less as in srs2 mutants, and there was a dramatic synergistic increase in rad9 siz1 double mutants, not less as in rad9 srs2 double mutants (Table 1). We also found that siz1 mutants had only a modest increase in allelic recombinants, not 20-fold more as in srs2 mutants (Table 1). Similar results (that siz1 and srs2 have different rates of homologous recombination) have been reported previously (PAPOULI et al. 2005).

In addition to analyzing siz1 mutants, we also assessed the role of PCNA

SUMOylation directly. In order to do this we utilized pol30-K127R and pol30-K164R mutants (discussed above), which are defective for PCNA SUMOylation, and thus are defective for Srs2 recruitment (Figure 2A; (HOEGE et al. 2002; PFANDER et al. 2005). Since both residues may be SUMOylated, neither allele is completely deficient for

SUMOylation of PCNA. It was previously shown that the pol30-K164R allele is severely defective for Srs2 recruitment while the pol30-K127R allele has a milder phenotype

(PFANDER et al. 2005). The pol30-K127R allele did not show any instability, while the pol30-K164R allele showed a high frequency of instability (Table 1). Furthermore, pol30- 152

K127R mutants did not alter the phenotype of rad9 single mutants, while pol30-K164R synergized with instability of rad9 single mutants (Table 1). This suggests that Pol30

K164 modification, by SUMOylation or ubiquitylation, is critical for preventing instability.

We favor the idea that Pol30 K164 ubiquitylation is more critical for preventing instability than SUMOylation, as rad5 mutants (that can SUMOylate PCNA, yet cannot poly-ubiquitylate PCNA) have a high-instability phenotype similar to pol30-K164R mutants (deficient for SUMOylation and ubiquitylation of PCNA, Table1).

The fact that siz1 and pol30-K164R (high instability, deficient for Srs2 recruitment to forks) mutants have such different phenotypes than srs2 mutants (low instability, deficient for all Srs2 activity) may suggest separate roles of Srs2 in S-phase versus other cell cycle stages. In the discussion we speculate on what these separate roles might be.

Analysis of DNA polymerase-associated mutants for roles in faulty template switching:

From our previous studies, we proposed that nearby inverted repeat fusion occurs by a faulty template switch reaction (Figure 1; (PAEK et al. 2009). To date we know of few mutants defective for this reaction (PAEK et al. 2009); most mutants we have tested either have no phenotype or have an increase in instability (ADMIRE et al. 2006b; PAEK et al. 2009). In an attempt to identify additional mutants defective in the faulty template switch reaction, we turned our attention to genes that are involved in DNA replication itself. 153

We first examined the role of non-essential subunits of DNA polymerase epsilon

(DPB3, DPB4 leading strand polymerase; (ARAKI et al. 1991; OHYA et al. 2000). Deletion of either of these subunits lead to an approximately 6-fold increase in the formation of mixed colonies in a Rad+ background (compare Rad+ to dpb3, dpb4) and a further 2-fold increase in a rad9 background (compare rad9 to rad9 dpb3, rad9 dpb4; Table 2). A similar effect was seen previously in pol32 mutants, a non-essential subunit of DNA polymerase delta (PAEK et al. 2009). Thus, deleting non-essential subunits of DNA polymerase epsilon (dpb3, dpb4) and delta (pol32) increased instability. So disrupting

DNA replication increases inverted repeat fusion, yet these mutants do not shed light on how events occur.

We then turned to analyzing DNA polymerase delta (Pol3) itself, using a temperature-sensitive allele (pol3-13), and Mgs1, a single-strand binding protein that has a strong genetic interaction with Pol3 (GIOT et al. 1997; HISHIDA et al. 2001; HISHIDA et al. 2002; VIJEH MOTLAGH et al. 2006). We studied each single mutant, pol3-13 mgs1 double mutants, as well as each single and the double mutant in combination with a highly unstable rad9 mutant (Table 2). The strongest phenotype is observed in the rad9 pol3-13 mgs1 triple mutant. We found that, relative to unstable rad9 single mutants, rad9 pol3-13 mgs1 mutants grown at the semi-restrictive temperature (36°C) had a 20- fold decrease in mixed colonies, a 5-fold decrease in dicentric chromosome formation, and a 200-fold decrease in allelic recombinants (Table 2; compare rad9 36°C to rad9 pol3-13 mgs1 36°C). 154

The strong phenotype of the triple mutant suggests Pol3 and Mgs1 have roles in forming rearrangements. Analyses of single and double mutants are largely consistent with this view (Table 2). Both pol3-13 and mgs1 single mutants had an increase in the formation of dicentric chromosomes (compared to Rad+), yet rad9 pol3-13 and rad9 mgs1 strains had a modest 2-3 fold decrease in both dicentric chromosome and mixed colony formation (Table 2; compared to rad9). In the discussion, we speculate as to why the single mutants show an increase in dicentric chromosomes, while the double and triple mutants show a decrease in these events.

The triple mutant is not only defective in the formation of dicentrics, but also in resolving them, as the decrease in mixed colony formation (20-fold) is greater than the decrease in dicentrics (5-fold). We propose, therefore, that the triple mutant also has a defect in resolving dicentric chromosomes that is required to form a mixed colony

(Table 2). We suggest that, in particular, the triple mutant has a profound defect in completing normal allelic recombination (Table 2, Figure 3). In sum, the triple mutant fails to form allelic recombinants either from a dicentric intermediate (which occurs in mixed colonies) or through a DSB formed independent of dicentrics (200-fold decrease in allelic recombinants, Table 2). The defect in allelic recombination in the triple mutant is addressed in the discussion.

The model that the triple mutant fails to complete allelic recombination makes a simple and testable prediction. We suggest that rad9 pol3-13 mgs1 mutants may start 155

to make allelic recombinants at 36°C, but can't resolve them at 36°C, yet can resolve them at 23°C. In support of this hypothesis, we found that cells grown in rich medium at

36°C but plated to selective medium at 23°C readily form allelic recombinants, while cells grown in rich medium at 23°C and plated to selective medium at 36°C could not

(Table 2, rad9 pol3-13 mgs1 36°-23°, rad9 pol3-13 mgs1 23°-36°). Formation of mixed colonies followed the same trend; when grown at 36°C in rich medium and plated to selective medium at 23°C, mixed colonies formed, but cells grown at 23°C in rich medium and plated to selective medium at 36°C did not form mixed colonies (Table 2, rad9 pol3-13 mgs1 36°-23°, rad9 pol3-13 mgs1 23°-36°). Therefore early events in formation of either allelic recombinants or mixed colonies (dicentrics) can form at 36°C, but their resolution requires full Pol3 function (see discussion and Supplementary Figure

1).

Additional genes involved in chromosomal instability: In separate genome instability assays developed by Kolodner and colleagues, fusion events occur between sequences with either limited homology (5-9 base pairs; (CHEN and KOLODNER 1999; KOLODNER et al.

2002; MYUNG et al. 2001a; PUTNAM et al. 2005) or with larger regions of homology (~4.2 kb; (PUTNAM et al. 2009; PUTNAM et al. 2010). They refer to these rearrangements as single copy and duplication-mediated rearrangements respectively, and interestingly, they found that certain genes were required to prevent duplication-mediated rearrangements and not single copy rearrangements, and vice versa (PUTNAM et al.

2009; PUTNAM et al. 2010). In the reactions that we study, inverted repeat fusions seem 156

in general to require more homology than single copy reactions (though see discussion), and less homology than duplication-mediated events (PAEK et al. 2009; PUTNAM et al.

2009). In order to better understand the mechanisms underlying inverted repeat fusion, we next analyzed a subset of genes that affect single copy and/or duplication-mediated rearrangements for their roles in inverted repeat fusion.

First we analyzed RAD27, the ortholog of human FEN-1, a 5’ flap endonuclease

(Kolodner and Marsischky 1999). Rad27 is required for Okazaki fragment processing and maturation during replication. Rad27 also has roles in base-excision repair and NHEJ

(LIU et al. 2004; WU et al. 1999). Previously it was found that rad27 mutants have about a 1000-fold increase in single-copy GCRs and a 140-fold increase in duplication- mediated GCRs (CHEN and KOLODNER 1999; PUTNAM et al. 2009). In our assay, although rad27 mutants have an increased frequency of mixed colony formation, the increase was far less dramatic (approximately 10-fold; Table 3); thus Rad27 has a more dramatic role in preventing both single-copy and duplication-mediated events than in preventing inverted repeat fusion. (rad27 mutants also had a significant increase in allelic recombinants and chromosome loss, similar to what others have seen; (Vallen and Cross

1995).

The Slx5-Slx8 complex is a SUMO-dependent ubiquitin ligase complex thought to bind and ubiquitylate SUMOylated proteins, and has been shown to have a role in the

DNA damage response (II et al. 2007; ROUSE 2009). Interestingly, slx5 mutants were 157

shown to have a high rate of duplication-mediated GCRs (25-fold increase) but not single copy GCRs (PUTNAM et al. 2009). Consistent with the role of Slx5 in preventing repeat mediated rearrangements, we saw an increase in mixed colony formation in slx5 single mutants compared to wild-type (Table 3), yet again, as with rad27 mutants, the increase in mixed colony formation in slx5 mutants in our assay was less dramatic (9- fold increase) than in the duplication-mediated assay (25-fold increase, Table 3;(PUTNAM et al. 2009). (slx5 strains also showed a significant increase in chromosome loss and allelic recombinants in our assay. The rad9 slx5 double mutant displayed a synergistic increase in chromosome loss compared to either single; rad9 slx5, 2350 x 10^5; rad9,

273 x 10^5; slx5, 313 x 10^5, Table 3).

Next, we examined the Cullin Rtt101 protein that is involved in replication through natural and MMS-induced pause sites, and has been shown to increase the frequency of single-copy GCRs (to our knowledge it has not yet been tested in the duplication- mediated GCR assay; (LUKE et al. 2006). Rtt101 is a multi-subunit E3 ubiquitin ligase that, together with Mms1 and Mms22, aids in the replication of damaged DNA (ZAIDI et al.

2008). At least one way Rtt101 is thought to do this is through the recruitment to stalled forks of Rtt107, a BRCT domain protein implicated in replication restart following

DNA damage (CHIN et al. 2006; ROUSE 2004). Rtt101 and Rtt107 likely act independently as well, because rtt101 rtt107 double mutants have a higher degree of MMS sensitivity than either single (ROBERTS et al. 2008; ZAIDI et al. 2008). Interestingly we see a very high frequency of mixed colony formation in both rtt101 and rtt107 single mutants, as well 158

as in rad9 rtt101 and rad9 rtt107 doubles (Table 3). rtt107 single mutants had approximately three times more mixed colonies and five times more chromosome loss than rtt101 single mutants. Taken together these data imply that both Rtt101 and

Rtt107 act to prevent nearby inverted repeat fusion and chromosome instability, and that Rtt107 has a greater role than does Rtt101.

Finally we analyzed yen1 mutants. Yen1 acts as a Holliday junction resolvase, though there are several other proteins that also have resolvase activity (IP et al. 2008;

SVENDSEN and HARPER). To our knowledge yen1 mutants have not been analyzed by either the single-copy or duplication-mediated GCR assays, however we feel this is an important mutant as one version of inverted repeat fusion occurs by a potential Yen1- dependent resolution of a regressed fork (chicken foot; (PAEK et al. 2009). yen1 mutants have a very modest to negligible phenotype in our instability assay, arguing against a role for regressed forks (Table 3).

DISCUSSION

Post-Replication Repair: The Post-replication repair pathway is thought to be one of the predominant mechanisms that cells utilize in order to bypass lesions on the template strand during DNA replication (LEE and MYUNG 2008; ZHANG and LAWRENCE

2005). Using mutants that are deficient for translesion synthesis (rad30 and rev7), we have shown that this sub-pathway of PRR does not prevent the fusion of inverted repeats, nor do the error-prone polymerases appear to have a role in rearrangements 159

(Table 1). In contrast, mutants in the error-free bypass branch (rad5 and to a lesser extent ubc13) all increased the frequency of mixed colonies. Indeed, rad5 mutants had a phenotype strikingly similar to that of pol30-K164R mutants that are deficient for both branches of PRR (Table 1). Thus it is likely that the chromosome instability phenotypes that we see in these mutants are solely due to a defect in the error-free bypass branch of post-replication repair. We propose that PRR prevents fusion of inverted repeats by promoting a template switch of the stalled nascent strand to the correct sequence on its sister chromatid, forming a hemicatenane as proposed by others (Figure 4, IIA-IIB,

(BRANZEI et al. 2008). PRR may thus prevent a “faulty” template switch event to the non- allelic inverted repeat (Figure 4).

Individual mutants in the error-free branch of PRR had different levels of inverted repeat fusion, implying that the wild type proteins play slightly different roles

(Table 1). That is, rad5 mutants had a much higher level of mixed colony formation than ubc13 mutants, though both mutants increased mixed colony formation to some degree. A similar trend was seen when these mutants were analyzed for single-copy

GCR formation, and also for MMS sensitivity (MOTEGI et al. 2006; PUTNAM et al. 2010;

XIAO et al. 2000). Since pol30-K164R mutants are epistatic to rad5 and ubc13 for mixed colony formation, this implies that mixed colonies in these mutants are due to aberrant

PRR, and not some other (PRR-independent) pathway that these genes might be involved in (Table 1, Figure 2). How can we explain the observation that instability is higher in rad5 mutants than in ubc13 mutants? Both Rad5 and Ubc13 have been shown 160

to be required for poly-ubiquitylating PCNA, however the poly-ubiquitylation assays were not quantitative; it may be that Rad5 is required for full poly-ubiquitylation, while

Ubc13 plays a lesser role (HOEGE et al. 2002). Thus it could be that in ubc13 mutants

PCNA is poly-ubiquitylated to some extent, allowing greater amount of PRR function than present in rad5 or rad18 mutants (Hoege et al 2002).

Another interesting facet in PRR regulation is the SUMOylation of PCNA at lysine

127 and lysine 164 by the Siz1 sumo-ligase (HOEGE et al. 2002; PAPOULI et al. 2005;

PFANDER et al. 2005; STELTER and ULRICH 2003). PCNA SUMOylation recruits the Srs2 helicase that removes Rad51 from DNA, and thus may provide access to proteins in the

PRR pathway (PAPOULI et al. 2005; PFANDER et al. 2005). We found that mutants which are deficient for Srs2 recruitment to replication forks (siz1, pol30-K164R) had a high level of instability whereas srs2 mutants had a low level of instability (Table 1; (PAEK et al. 2009). How can we explain that srs2 and siz1 mutants have opposite phenotypes, if

Siz1-dependent PCNA SUMOylation is required for Srs2 recruitment to a fork? We suggest that when Srs2 is recruited to stalled replication forks via Siz1-dependent PCNA

SUMOylation, Srs2 prevents improper recombination events between sisters (by blocking Rad51 assembly). We suggest, however, that Srs2 may also prevent recombination between sisters in G2 (also by blocking Rad51 assembly), independent of the replication fork. And, Siz1 may promote PRR independent of Srs2 function (by a mechanism that is not yet known). Therefore, in a siz1 mutant PRR fails to some extent, and some inverted repeat fusion ensues in G2 (Figure 4). In srs2 mutants, replication 161

forks may fail, yet when cells reach G2 the lesions are repaired by Rad51-dependent homologous recombination, and not faulty template switching (Figure 4). In support of this model, it has been shown that PCNA is SUMOylated mainly in S-phase, while ubiquitylation mainly occurs in G2 (HOEGE et al. 2002). In addition, it has been shown recently that PRR can operate efficiently in the G2/M phase of the cell cycle, and PRR’s predominant role might be at single stranded gaps left after replication fork restart

(DAIGAKU et al.; KARRAS and JENTSCH).

Polymerase Delta and Mgs1 promote fusion events in rad9 mutants, and Rad9, Pol

Delta and Mgs1 are required for the propagation of allelic recombinants and cells with dicentric chromosomes: Surprisingly we found that rad9 pol3-13 mgs1 mutants grown at the restrictive temperature showed a dramatic 20-fold decrease in the formation of mixed colonies and a 200-fold decrease in allelic recombinants (Table 2). We propose that the decrease in mixed colony formation in rad9 pol3-13 mgs1 cells is due to a defect both in the formation of dicentric chromosomes (dicentrics down 5-fold relative to rad9 single mutants), and in an inability to properly resolve dicentrics by allelic recombination once dicentrics are formed (Figure 3-4).

Both mgs1 and pol3-13 mutants decreased the frequency of dicentric chromosome formation in a rad9 mutant background, yet these mutations caused an increase in a wild-type background (Table 2). We suggest that as single mutants (pol3-

13 36°C, mgs1), defects in replication lead to an increased number of defective forks, 162

which then lead to dicentrics through the use of the other intact gene products. In double mutants (rad9 pol3-13 36°C, rad9 mgs1), the increase in fork defects (caused by a rad9 mutation) is offset by defects in inverted repeat fusion (by Mgs1 and Pol3 defects) needed to complete events. Finally, in the triple mutant (rad9 pol3-13 mgs1

36°C), most lesions are not converted into dicentrics, as we see a 5-fold decrease in dicentric chromosomes.

There seems to be a synergistic interaction between all three genes, an interaction that occurs to a lesser extent in any double mutant (Table 2). Therefore a plausible molecular explanation requires three interactions, and all three may have some role in replication fork behavior. Mgs1 is a DNA dependent ATPase with single strand annealing activities, interacts with pol delta (BRANZEI et al. 2002; HISHIDA et al.

2001), and has been proposed to promote replication restart (VIJEH MOTLAGH et al. 2006).

Thus we propose that Mgs1 is partly responsible for annealing of the nascent strand to the non-allelic template strand (in inverted repeat fusion, Figure 4, IIIA). After annealing, pol delta is likely required for replication to the end of the neighboring

Okazaki fragment (Figure 4, IIIB) thus fusing the two sisters in inverted repeat fusion or completing structures in allelic recombination (Figure 3, step 4; (MALOISEL et al. 2008).

Mgs1 could act at these polymerization steps as well, as it has been shown that Mgs1 binds to Pol delta, and is required to stabilize temperature sensitive Pol delta strains

(Figure 4, IIIB; (BRANZEI et al. 2002; VIJEH MOTLAGH et al. 2006). Rad9's role is either in a 163

cell cycle delay (Figure 3 Step 5a) or in promoting fork stability (via Chk1;(Segurado and

Diffley 2008). Thus proper annealing and synthesis, be it to non-allelic or allelic sites, requires the combined action of Mgs1, Pol3 and Rad9.

Although speculative, this model does have additional support. First, it has been suggested that Mgs1 stabilizes replication forks in strains with pol delta defects (VIJEH

MOTLAGH et al. 2006). Second, genetic data supports the idea that pol delta is the major polymerase during mitotic recombination (GIOT et al. 1997; MALOISEL et al. 2008). Third, temperature shift experiments showed that when the triple mutant is grown in rich medium at the restrictive temperature, then plated to selection medium at the permissive temperature, mixed colony and allelic recombinants occur. We suggest that resolution of dicentrics requires events similar to those required for allelic recombination. A surprising prediction of this model is that most allelic recombinants in the triple mutant are not resolved before cells are subjected to selection, but are rather are resolved on selective medium (Table 2). We feel that wild type cells probably do resolve recombination events before selection, and provide a model in the supplement that explains this phenomenon in rad9 pol3-13 mgs1 cells (Supplementary Figure 1).

Finally we note that genes which have defects at other stages of recombination do not show a decrease in mixed colony formation; mutants with defects in the resection

(mre11), pairing (rad51, rad52) and Holliday junction resolution (sgs1, mus81, yen1) stages of homologous recombination do not show decreases in mixed colony formation 164

(KROGH and SYMINGTON 2004; PAEK et al. 2009); unpublished data; this study). Thus we infer a specific defect in resolution of allelic recombinants in rad9 pol3-13 mgs1 triple mutants that these other single mutants are proficient in.

Relationship of single copy and repeat copy rearrangements to inverted repeat fusion:

Finally, in order to compare inverted repeat fusion and dicentric chromosome formation to rearrangements seen in other GCR assays, we analyzed genes with roles in single copy or repeat mediated rearrangements (Table 3). We saw little correlation between the increase in dicentric chromosome formation and either single copy or duplication- mediated GCRs. For example, rad27 mutants have a 1000-fold increase in single copy

GCR formation, and a 140-fold increase in duplication-mediated GCRs (CHEN and

KOLODNER 1999; PUTNAM et al. 2009). In contrast, slx5 mutants have only a modest (~10- fold) increase in inverted repeat fusion events (Table 3). Analysis of PRR mutants also distinguish both single copy GCRs and duplication-mediated GCRs from inverted repeat fusion events (MOTEGI et al. 2006; PUTNAM et al. 2010) This implies that inverted repeat fusions, single copy and repeat sequence rearrangements are likely formed by separate mechanisms. Indeed, duplication mediated GCRs occur between separate chromosomes, between sequences with 4.2 kb of homology, and appear to be Rad52 dependent (PUTNAM et al. 2009). In contrast, the GCRs in this study arise from inverted repeat fusion, occur on the same chromosome (likely by sister chromatid fusion), between 300 bp repeats, and occur independent of Rad52 (ADMIRE et al. 2006b; PAEK et al. 2009). 165

ACKNOWLEDGEMENTS

We thank Dmitry Gordenin and Anja-Katrin Bielinsky for strains. Also we thank Dr.

Steven Smith for statistical assistance. T.W was funded by an NACRP/NCI grant and NIH

1R01 GM076186, A.P. was funded by NSF-IGERT grant DGE-0114420, H.J. is funded by

National Institute of General Medical Sciences Award Number F31GM087120, and S.K. was supported by GM008659-11A1. We also thank the JMST student forum for useful discussions.

166

LITERATURE CITED

ARAKI, H., R. K. HAMATAKE, A. MORRISON, A. L. JOHNSON, L. H. JOHNSTON et al., 1991 Cloning DPB3, the gene encoding the third subunit of DNA polymerase II of Saccharomyces cerevisiae. Nucleic Acids Res 19: 4867-4872.

ARGUESO, J. L., J. WESTMORELAND, P. A. MIECZKOWSKI, M. GAWEL, T. D. PETES et al., 2008 Double-strand breaks associated with repetitive DNA can reshape the genome. Proc Natl Acad Sci U S A 105: 11845-11850.

ATKINSON, J., and P. MCGLYNN, 2009 Replication fork reversal and the maintenance of genome stability. Nucleic Acids Res 37: 3475-3492.

BRANZEI, D., M. SEKI, F. ONODA and T. ENOMOTO, 2002 The product of Saccharomyces cerevisiae WHIP/MGS1, a gene related to replication factor C genes, interacts functionally with DNA polymerase delta. Mol Genet Genomics 268: 371-386.

BRANZEI, D., F. VANOLI and M. FOIANI, 2008 SUMOylation regulates Rad18-mediated template switch. Nature 456: 915-920.

CARSON, M. J., and L. HARTWELL, 1985 CDC17: an essential gene that prevents telomere elongation in yeast. Cell 42: 249-257.

CHEN, C., and R. D. KOLODNER, 1999 Gross chromosomal rearrangements in Saccharomyces cerevisiae replication and recombination defective mutants. Nat Genet 23: 81-85.

CHIN, J. K., V. I. BASHKIROV, W. D. HEYER and F. E. ROMESBERG, 2006 Esc4/Rtt107 and the control of recombination during replication. DNA Repair (Amst) 5: 618-628.

DAIGAKU, Y., A. A. DAVIES and H. D. ULRICH, 2010 Ubiquitin-dependent DNA damage bypass is separable from genome replication. Nature 465: 951-955.

DAS-BRADOO, S., H. D. NGUYEN, J. L. WOOD, R. M. RICKE, J. C. HAWORTH et al., 2010 Defects in DNA ligase I trigger PCNA ubiquitylation at Lys 107. Nat Cell Biol 12: 74-79.

167

DI RIENZI, S. C., D. COLLINGWOOD, M. K. RAGHURAMAN and B. J. BREWER, 2009 Fragile genomic sites are associated with origins of replication. Genome Biol Evol 2009: 350-363.

DUNHAM, M. J., H. BADRANE, T. FEREA, J. ADAMS, P. O. BROWN et al., 2002 Characteristic genome rearrangements in experimental evolution of Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 99: 16144-16149.

GIOT, L., R. CHANET, M. SIMON, C. FACCA and G. FAYE, 1997 Involvement of the yeast DNA polymerase delta in DNA repair in vivo. Genetics 146: 1239-1251.

HISHIDA, T., H. IWASAKI, T. OHNO, T. MORISHITA and H. SHINAGAWA, 2001 A yeast gene, MGS1, encoding a DNA-dependent AAA(+) ATPase is required to maintain genome stability. Proc Natl Acad Sci U S A 98: 8283-8289.

HISHIDA, T., T. OHNO, H. IWASAKI and H. SHINAGAWA, 2002 Saccharomyces cerevisiae MGS1 is essential in strains deficient in the RAD6-dependent DNA damage tolerance pathway. Embo J 21: 2019-2029.

HOEGE, C., B. PFANDER, G. L. MOLDOVAN, G. PYROWOLAKIS and S. JENTSCH, 2002 RAD6- dependent DNA repair is linked to modification of PCNA by ubiquitin and SUMO. Nature 419: 135-141.

HUANG, D., and D. KOSHLAND, 2003 Chromosome integrity in Saccharomyces cerevisiae: the interplay of DNA replication initiation factors, elongation factors, and origins. Genes Dev 17: 1741-1754.

II, T., J. FUNG, J. R. MULLEN and S. J. BRILL, 2007 The yeast Slx5-Slx8 DNA integrity complex displays ubiquitin ligase activity. Cell Cycle 6: 2800-2809.

IP, S. C., U. RASS, M. G. BLANCO, H. R. FLYNN, J. M. SKEHEL et al., 2008 Identification of Holliday junction resolvases from humans and yeast. Nature 456: 357-361.

KARRAS, G. I., and S. JENTSCH, 2010 The RAD6 DNA damage tolerance pathway operates uncoupled from the replication fork and is functional beyond S phase. Cell 141: 255-267.

KLEIN, H. L., 2001 Mutations in recombinational repair and in checkpoint control genes suppress the lethal combination of srs2Delta with other DNA repair genes in Saccharomyces cerevisiae. Genetics 157: 557-565.

168

KOLODNER, R. D., and G. T. MARSISCHKY, 1999 Eukaryotic DNA mismatch repair. Curr Opin Genet Dev 9: 89-96.

KOLODNER, R. D., C. D. PUTNAM and K. MYUNG, 2002 Maintenance of genome stability in Saccharomyces cerevisiae. Science 297: 552-557.

KREJCI, L., S. VAN KOMEN, Y. LI, J. VILLEMAIN, M. S. REDDY et al., 2003 DNA helicase Srs2 disrupts the Rad51 presynaptic filament. Nature 423: 305-309.

KROGH, B. O., and L. S. SYMINGTON, 2004 Recombination proteins in yeast. Annu Rev Genet 38: 233-271.

KRUSKAL, W. H., and W. A. WALLIS, 1952 Use of ranks in one-criterion variance analysis. Journal of American Statistical Association 47: 583-621.

LEE, J. A., C. M. CARVALHO and J. R. LUPSKI, 2007 A DNA replication mechanism for generating nonrecurrent rearrangements associated with genomic disorders. Cell 131: 1235-1247.

LEE, K. Y., and K. MYUNG, 2008 PCNA modifications for regulation of post-replication repair pathways. Mol Cells 26: 5-11.

LEMOINE, F. J., N. P. DEGTYAREVA, K. LOBACHEV and T. D. PETES, 2005 Chromosomal translocations in yeast induced by low levels of DNA polymerase a model for chromosome fragile sites. Cell 120: 587-598.

LIU, Y., H. I. KAO and R. A. BAMBARA, 2004 Flap endonuclease 1: a central component of DNA metabolism. Annu Rev Biochem 73: 589-615.

LUKE, B., G. VERSINI, M. JAQUENOUD, I. W. ZAIDI, T. KURZ et al., 2006 The cullin Rtt101p promotes replication fork progression through damaged DNA and natural pause sites. Curr Biol 16: 786-792.

MALOISEL, L., F. FABRE and S. GANGLOFF, 2008 DNA polymerase delta is preferentially recruited during homologous recombination to promote heteroduplex DNA extension. Mol Cell Biol 28: 1373-1382.

169

MOTEGI, A., K. KUNTZ, A. MAJEED, S. SMITH and K. MYUNG, 2006 Regulation of gross chromosomal rearrangements by ubiquitin and SUMO ligases in Saccharomyces cerevisiae. Mol Cell Biol 26: 1424-1433.

MYUNG, K., C. CHEN and R. D. KOLODNER, 2001 Multiple pathways cooperate in the suppression of genome instability in Saccharomyces cerevisiae. Nature 411: 1073-1076.

OHYA, T., S. MAKI, Y. KAWASAKI and A. SUGINO, 2000 Structure and function of the fourth subunit (Dpb4p) of DNA polymerase epsilon in Saccharomyces cerevisiae. Nucleic Acids Res 28: 3846-3852.

PANNUNZIO, N. R., G. M. MANTHEY and A. M. BAILIS, 2008 RAD59 is required for efficient repair of simultaneous double-strand breaks resulting in translocations in Saccharomyces cerevisiae. DNA Repair (Amst) 7: 788-800.

PAPOULI, E., S. CHEN, A. A. DAVIES, D. HUTTNER, L. KREJCI et al., 2005 Crosstalk between SUMO and ubiquitin on PCNA is mediated by recruitment of the helicase Srs2p. Mol Cell 19: 123-133.

PFANDER, B., G. L. MOLDOVAN, M. SACHER, C. HOEGE and S. JENTSCH, 2005 SUMO-modified PCNA recruits Srs2 to prevent recombination during S phase. Nature 436: 428- 433.

170

PUTNAM, C. D., T. K. HAYES and R. D. KOLODNER, 2009 Specific pathways prevent duplication-mediated genome rearrangements. Nature 460: 984-989.

PUTNAM, C. D., T. K. HAYES and R. D. KOLODNER, 2010 Post-replication repair suppresses duplication-mediated genome instability. PLoS Genet 6: e1000933.

PUTNAM, C. D., V. PENNANEACH and R. D. KOLODNER, 2005 Saccharomyces cerevisiae as a model system to define the chromosomal instability phenotype. Mol Cell Biol 25: 7226-7238.

RATTRAY, A. J., B. K. SHAFER, B. NEELAM and J. N. STRATHERN, 2005 A mechanism of palindromic gene amplification in Saccharomyces cerevisiae. Genes Dev 19: 1390-1399.

ROBERTS, T. M., I. W. ZAIDI, J. A. VAISICA, M. PETER and G. W. BROWN, 2008 Regulation of Rtt107 recruitment to stalled DNA replication forks by the cullin Rtt101 and the Rtt109 acetyltransferase. Mol Biol Cell 19: 171-180.

ROUSE, J., 2004 Esc4p, a new target of Mec1p (ATR), promotes resumption of DNA synthesis after DNA damage. Embo J 23: 1188-1197.

ROUSE, J., 2009 Control of genome stability by Slx protein complexes. Biochem Soc Trans 37: 495-510.

SCHMIDT, K. H., V. PENNANEACH, C. D. PUTNAM and R. D. KOLODNER, 2006 Analysis of gross- chromosomal rearrangements in Saccharomyces cerevisiae. Methods Enzymol 409: 462-476.

SEGURADO, M., and J. F. DIFFLEY, 2008 Separate roles for the DNA damage checkpoint protein kinases in stabilizing DNA replication forks. Genes Dev 22: 1816-1827.

SMITH, C. E., B. LLORENTE and L. S. SYMINGTON, 2007 Template switching during break- induced replication. Nature 447: 102-105.

STANKIEWICZ, P., and J. R. LUPSKI, 2010 Structural variation in the human genome and its role in disease. Annu Rev Med 61: 437-455.

STELTER, P., and H. D. ULRICH, 2003 Control of spontaneous and damage-induced mutagenesis by SUMO and ubiquitin conjugation. Nature 425: 188-191.

171

SVENDSEN, J. M., and J. W. HARPER, 2010 GEN1/Yen1 and the SLX4 complex: Solutions to the problem of Holliday junction resolution. Genes Dev 24: 521-536.

VALLEN, E. A., and F. R. CROSS, 1995 Mutations in RAD27 define a potential link between G1 cyclins and DNA replication. Mol Cell Biol 15: 4291-4302.

VEAUTE, X., J. JEUSSET, C. SOUSTELLE, S. C. KOWALCZYKOWSKI, E. LE CAM et al., 2003 The Srs2 helicase prevents recombination by disrupting Rad51 nucleoprotein filaments. Nature 423: 309-312.

WACH, A., A. BRACHAT, R. POHLMANN and P. PHILIPPSEN, 1994 New heterologous modules for classical or PCR-based gene disruptions in Saccharomyces cerevisiae. Yeast 10: 1793-1808.

WALKER, J. B., 1955 Canavanine and homoarginine as antimetabolites of arginine and lysine in yeast and algae. J Biol Chem 212: 207-215.

WEINERT, T., S. KAOCHAR, H. JONES, A. PAEK and A. J. CLARK, 2009 The replication fork's five degrees of freedom, their failure and genome rearrangements. Curr Opin Cell Biol 21: 778-784.

WEINERT, T. A., and L. H. HARTWELL, 1990 Characterization of RAD9 of Saccharomyces cerevisiae and evidence that its function acts posttranslationally in cell cycle arrest after DNA damage. Mol Cell Biol 10: 6554-6564.

WU, X., T. E. WILSON and M. R. LIEBER, 1999 A role for FEN-1 in nonhomologous DNA end joining: the order of strand annealing and nucleolytic processing events. Proc Natl Acad Sci U S A 96: 1303-1308.

XIAO, W., B. L. CHOW, S. BROOMFIELD and M. HANNA, 2000 The Saccharomyces cerevisiae RAD6 group is composed of an error-prone and two error-free postreplication repair pathways. Genetics 155: 1633-1641.

ZAIDI, I. W., G. RABUT, A. POVEDA, H. SCHEEL, J. MALMSTROM et al., 2008 Rtt101 and Mms1 in budding yeast form a CUL4(DDB1)-like ubiquitin ligase that promotes replication through damaged DNA. EMBO Rep 9: 1034-1040. 172

173

174

175

176

FIGURE LEGENDS

FIGURE 1. Experimental setup for the detection of inverted repeat fusion and chromosome instability. Objects are not drawn to scale. A) The starting strain has two copies of Chr VII. One copy contains the CAN1 gene is ADE6, ade3, while the other copy is ade6, ADE3. Cells are plated to canavanine, and three types of colonies are formed: B) Allelic recombinants are round in appearance and are Ade+ C) Colonies that form by loss of Chr VII are round in appearance and Ade- D) Cells which contain unstable dicentric chromosomes form by the fusion of inverted repeats. One specific case of this fusion (the S2/S3 dicentric) is shown in brackets. Cells with dicentrics form mixed colonies which contain allelic recombinants, chromosome loss events, as well as a translocation between D7 and D11. The bar in the S2/S3 repeat represents a fusion junction.

E) The specific dicentric is detected by dicentric primers DP1 and DP2 and F) a monocentric translocation that is detected with translocation primers TP1 and

TP2.

FIGURE 2. Model of the post replication repair pathway and regulation of Srs2.

A) Ubc9 and Siz1 sumoylate PCNA at lysines 164 and 127. SUMOylation recruits

Srs2 to PCNA and prevents Rad51 from accessing forks. In addition, PCNA is mono-ubiquitylated by a Rad6-Rad18 heterodimer at lysine 164 which recruits translesion polymerases (Rev3, Rad30) to sites of damage. Rad5, Ubc13 and

Mms2 poly-ubiquitylate PCNA in order to allow for error-free bypass. B) Srs2 177

inhibits Rad51, Rad51 promotes allelic recombination and prevents mixed colonies from forming. Thus srs2 mutants show an increase in allelic recombinants and a decrease in mixed colony formation.

FIGURE 3. Model for defects in rad9 pol3-13 mgs1 strains at the restrictive temperature. 1) When a double strand break occurs on a dicentric or monocentric chromosome, 2) the ends are resected and 3) strand invasion and second end capture occurs. 4) This is followed by Pol Delta and Mgs1 dependent polymerization, which by this model proceeds slowly in rad9 pol3-13 mgs1 mutants 5a) In a RAD9 cell this signals a checkpoint response, which prevents chromosome segregation. However in a rad9 cell mitosis proceeds, which leads to chromosome missegregation. 5b) A normal cell is able to form and 6) resolve a Holliday junction, leading to allelic recombinants.

FIGURE 4. Model of faulty template switching events, red arrows indicate inverted repeats, the blue asterisk indicates a "lesion" in the DNA. The model is drawn for a leading strand stall, but events could initiate on lagging strand instead. IA) When a replication fork stalls at a lesion, IB) Srs2 promotes replication restart downstream of the stall site, by preventing Rad51 from acting.

A faulty template switch event is also possible at this step (see Figure 1D). IIA)

In G2 Rad18 and other PRR proteins can promote a “normal” template switch to the nascent strand of the sister chromatid IIB) which forms a hemicatenane.

Alternatively, after resection of DNA by a nuclease, IIIA) a faulty template switch 178

event initiates when Mgs1 anneals the stalled nascent strand to the non-allelic repeat sequence on the template strand. IIIB) Pol Delta and Mgs1 dependent

DNA synthesis leads to the formation of a dicentric chromosome. IVA) In srs2 cells Rad51 gains access to ssDNA and promotes strand invasion and second end capture. IVB) Resolution leads to formation of an allelic recombinant.

179

180

181

182

183

TABLE S1. SACCHAROMYCES CEREVISIAE STRAINS USED IN THIS STUDY

Strain Genotypea Source

TY200 RAD+Wild-type Admire et al., 2006

TY206 rad9∆::ura3 Admire et al., 2006

TY440 rad18∆::KanMX4 Paek et al., 2009

TY354 srs2∆::URA3 Paek et al., 2009

TY452 rad9∆::ura3 rad18∆::KanMX4 Paek et al., 2009

TY356 rad9∆::ura3 srs2∆::URA3 Paek et al., 2009

TY545 pol30 K164R::TRP5 This study

TY546 rad5∆::KanMX4 This study

TY547 ubc13∆::KanMX4 This study

TY548 rev7∆::KanMX4 This study

TY549 rad30∆::KanMX4 This study

TY550 pol30 K127R::TRP5 This study

TY551 siz1∆::KanMX4 This study

TY552 rad9∆::ura3 pol30 K164R::TRP5 This study

TY553 rad5∆::KanMX4 pol30 K164R::TRP5 This study

TY554 ubc13∆::KanMX4 pol30 K164R::TRP5 This study

TY555 rad9∆::ura3 rad5∆::KanMX4 This study

TY556 rad9∆::ura3 ubc13∆::KanMX4 This study

TY557 rad9∆::ura3 rev7∆::KanMX4 This study

TY558 rad9∆::ura3 rad30∆::KanMX4 This study

TY559 rad9∆::ura3 pol30 K127R::TRP5 This study

TY560 rad9∆::ura3 siz1∆::KanMX4 This study

TY561 mgs1∆::KanMX4 This study

TY562 dpb3∆::KanMX4 This study

TY563 dpb4∆::KanMX4 This study 184

TY564 pol3-13 This study

TY565 mgs1∆::KanMX4 pol3-13 This study

TY566 rad9∆::ura3 mgs1∆::KanMX4 This study

TY567 rad9∆::ura3 dpb3∆::KanMX4 This study

TY568 rad9∆::ura3 dpb4∆::KanMX4 This study

TY569 rad9∆::ura3 pol3-13 This study

TY570 rad9∆::ura3 pol3-13 mgs1∆::KanMX4 This study

TY571 rad27∆::URA3 This study

TY572 rtt101∆::KanMX4 This study

TY573 rtt107∆::KanMX4 This study

TY574 slx5∆::KanMX4 This study

TY575 yen1∆::KanMX4 This study

TY576 rad9∆::ura3 rtt101∆::KanMX4 This study

TY577 rad9∆::ura3 rtt107∆::KanMX4 This study

TY578 rad9∆::ura3 slx5∆::KanMX4 This study

TY579 rad9∆::ura3 yen1∆::KanMX4 This study aAll strains are disomic for Chr VII and derivatives of TY200 MATa +/hxk2::CAN1 lys5/+ cyhr/CYHs trp5/+ leu1/+ Centromere ade6/+ +/ade3, ura3-52 (A364a genetic background) except for mutations listed (Admire et al., 2006).

185

SUPPLEMENTARY FIGURE 1. The fate of double strand breaks in rad9 pol3-13 mgs1 mutants grown at the restrictive temperature (36°C). 1) After a double strand break occurs on a dicentric or monocentric chromosome, the ends are resected and 2) strand invasion and second end capture occurs. 3) This is followed by Pol Delta and Mgs1 dependant polymerization, which proceeds slowly in rad9 pol3-13 mgs1 mutants 4) Chromosome segregation leads to either a chromosome loss event or to a one-ended DSB. 6) This leads to a break induced replication event, which again proceeds slowly in rad9 pol3-13 mgs1 mutants. Chromosome Segregation either leads to a one-ended DSB or a chromosome loss event.