NUCLEOSOME REMODELING BY hMSH2- hMSH6

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By

Sarah Javaid

Graduate Program in Biophysics

The Ohio State University

2010

Dissertation Committee:

Professor Richard Fishel, Advisor

Professor Charles Brooks

Professor Mark Foster

Professor Joanna Groden

Copyright by

Sarah Javaid

2010

Abstract

The human MutS homologues (MSH), hMSH2 and hMSH6, forms a heterodimer

(hMSH2-hMSH6) that plays a central role in mismatch repair (MMR). hMSH2-hMSH6 is required for the recognition of mismatched nucleotides and insertion/deletion loops

(IDLs) generated by misincorporation during DNA replication. Mutations in either of the hMSH2 or hMSH6 genes result in elevated spontaneous mutation rate and susceptibility to the common cancer predisposition syndrome, Lynch Syndrome or hereditary non- polyposis colorectal cancer (LS/HNPCC).

Mismatches that are recognized by hMSH2-hMSH6 arise in vivo within chromosomes that are a complex mixture of DNA and protein (chromatin). A fundamental unit of chromatin is the nucleosome which consists of ~147 bp of DNA wrapped twice around a histone octamer containing two H2A-H2B dimers and an H3-H4 tetramer. The biophysical/biochemical effect of chromatin on MMR is unknown.

Moreover, little is known about the effect of more than a 100 post-translational modifications (PTMs) that may decorate the human histones during MMR processes.

This dissertation discusses chromatin and MMR. Chapter 1 serves as an introduction to

MMR and chromatin. Chapter 2 provides the thesis rationale.

Chapter 3 involves analysis of a mismatched DNA substrate containing a single well-defined nucleosome. We demonstrate that hMSH2-hMSH6 can catalyze the disassembly of a nucleosome adjacent to a mismatch. In addition, we have constructed

ii nucleosomes containing acetylations of the histone H3 dyad residues K115 and K122 by a semi-synthetic intein-based strategy. We find that hMSH2-hMSH6 nucleosome disassembly is considerably enhanced when nucleosomes contain H3(K115, K122) acetylation modifications. Moreover, lysineglutamine substitution mutation of histone

H3(K56), used to mimic the lysine acetylation, also enhances nucleosome disassembly.

Disassembly of the nucleosome requires ATP binding by hMSH2-hMSH6. In addition, nucleosome disassembly is blocked by LacI/LacO placed between the mismatch and the nucleosome arguing in favor of a “cis” or “moving” mechanism.

Chapter 4 is devoted to analyzing single acetylation and mimics of the acetylation of histone H3(K115) and/or H3(K122). We observe that hMSH2-hMSH6 nucleosome disassembly is enhanced with acetylation and mimicked acetylation modifications of

H3(K115) and/or H3(K122). Moreover, hMSH2-hMSH6 nucleosome disassembly is dependent on the nucleosome positioning sequence (NPS). hMSH2-hMSH6 nucleosome disassembly is considerably enhanced with the physiological relevant Xenopus 5S rDNA

NPS. Disassembly of the nucleosome by hMSH2-hMSH6 is masked by the high affinity nonphysiological 601 and pMP2 NPS’s. Replacement of the DNA sequence at the nucleosomal-dyad axis in the Xenopus 5S rDNA NPS with the 601 NPS reduces nucleosome disassembly by hMSH2-hMSH6 ~2-fold. We also find that hMSH2-hMSH6 can recognize and bind to a mismatch within the nucleosome. hMSH2-hMSH6 binding affinity and nucleosome disassembly is increased when the mismatch is located at the entry-exit region of the nucleosome compared to a mismatch located at the LRS (loss of rDNA silencing) or the nucleosomal-dyad axis region. Furthermore, the ability of

iii hMSH2-hMSH6 to disassemble nucleosomes containing mismatches is enhanced when nucleosomes contain H3(K115, K122) acetylation modifications. These results highlight that nucleosome disassembly by hMSH2-hMSH6 is dependent on NPS and histone octamer modifications. Moreover, acetylation/mimic modifications enhance hMSH2- hMSH6 nucleosome disassembly.

Chapter 5 focuses on histone H3(K56) acetylation which occurs during DNA replication and repair and is located in the entry-exit region of the nucleosome. We find that a single nucleosome containing a H3(K56) acetylation with an adjacent mismatch is disassembled by hMSH2-hMSH6.

Chapter 6 is centered on histone H3(T118) phosphorylation, which is located at the nucleosomal-dyad axis region of the histone-DNA interface, directly adjacent to the

DNA backbone. We analyzed a single nucleosome reconstituted on the high affinity pMP2 artificial sequence containing an adjacent mismatch and observed enhanced hMSH2-hMSH6 nucleosome disassembly compared to unmodified nucleosomes reconstituted on the pMP2 sequence. The results detailed in this thesis improve our understanding of MMR in the context of chromatin. Together, these results suggest a novel passive mechanism for nucleosome disassembly by hMSH2-hMSH6.

iv

Dedication

This document is dedicated to my family,

Abdul R. Javaid, Tasnim Javaid, Shazia Azam, Zainab Javaid, Mustafa S. Javaid, Ally R.

Javaid, and Haadi I. Javaid

v

Acknowledgments

I wish to thank my advisor Richard Fishel for his support; none of this would have been possible without his guidance. I would also like to thank Joanna Groden, Mark Foster, and Charles Brooks for making the time and effort to serve on my thesis committee.

Special thanks to all the members of the lab who have been so great to work with. The time spent there would not have been so enjoyable without them. This project would not have been possible without the help and advice from Samir Acharya, Michael Mcilhatton,

Kang-Sup Shim, Michael Poirier, and Jennifer Ottesen.

vi

Vita

1999-2003 ...... B.S. Biology, James Madison University

2004 to present ...... PhD. Biophysics, The Ohio State University

Publication

Javaid, S. Manohar, M. Punja, N. Mooney, A. Ottesen, JJ. Poirier, MG. Fishel R. 2009. Nucleosome remodeling by hMSH2-hMSH6. Molecular Cell. 36:1086-1094

Field of Study

Major Field: Biophysics

vii

Table of Contents

Abstract ...... ii Dedication ...... v Acknowledgments...... vi Vita ...... vii List of Tables ...... xi List of Figures ...... xii List of Abbreviations ...... xv Preface...... xix

Chapter 1: Introduction ...... 1-52 I. DNA Repair Pathways ...... 1 I.a. Nucleotide Excision Repair...... 2 I.b. Base Excision Repair ...... 3 I.c. Mismatch Repair ...... 3 II. Hereditary Non-Polyposis Colorectal Cancer ...... 4 III. Mismatch Repair ...... 8 III.a. The Mutator Phenotype ...... 8 III.b. Prokaryotic Mismatch Repair...... 9 III.c. Eukaryotic Mismatch Repair ...... 14 III.d Signaling Models for Strand Discrimination in MMR...... 26 IV. Mismatch Repair and DNA damage signaling ...... 30 V. Chromatin and Mismatch Repair ...... 32 V.a. Nucleosome Structure...... 33 V.b. Nucleosome Positioning ...... 36 V.c. Histone Variants ...... 37 V.d. Post-translational Modifications of Histones ...... 38 V.e. Chromatin Remodelers ...... 42 viii

V.f. Chromatin Remodeling ...... 47 V.g. Chromatin and Mismatch Repair ...... 51

Chapter 2: Thesis Rationale ...... 53-54

Chapter 3: Nucleosome Remodeling by hMSH2-hMSH6...... 55-88 I. Summary ...... 55 II. Introduction ...... 56 III. Results ...... 56 IV. Discussion ...... 70 V. Experimental Procedures ...... 77 VI. Acknowledgements ...... 80 VII. Supplemental Figures ...... 81

Chapter 4: hMSH2-hMSH6 nucleosome disassembly is dependent on post-translational modifications and nucleosome localization sequences ...... 89-138 I. Summary ...... 89 II. Introduction ...... 90 III. Results ...... 94 IV. Discussion ...... 115 V. Experimental Procedures ...... 119 VI. Supplemental Figures ...... 126

Chapter 5: Histone H3(K56) acetylation enhances hMSH2-hMSH6 nucleosome disassembly ...... 139-147 I. Summary ...... 139 II. Introduction ...... 139 III. Results and Discussion ...... 141 IV. Experimental Procedures ...... 143 V. Supplemental Figures ...... 145

Chapter 6: Histone H3(T118) phosphorylation enhances hMSH2-hMSH6 nucleosome disassembly ...... 148-156 I. Summary ...... 148 ix

II. Introduction ...... 148 III. Results and Discussion ...... 150 IV. Experimental Procedures ...... 154 V. Supplemental Figures ...... 155

Conclusions and Future Directions ...... 157-170

References ...... 171-188

x

List of Tables

Table 1. Amsterdam Criteria 1 and II and Bethesda Guidelines ...... 6

Table 2. Identity and functions of E. coli and human proteins involved in MMR ..... 15-16

xi

List of Figures

Figure 1. X-ray crystal structure of hMSH2-hMSH6/ADP/G•T heteroduplex complex . 18

Figure 2. Cis and trans models for mismatch repair ...... 27

Figure 3. Crystal Structure of nucleosome core particle ...... 34

Figure 4. Binding of hMSH2-hMSH6 to nucleosome-DNA ...... 58

Figure 5. Nucleosome disassembly by hMSH2-hMSH6 ...... 61-62

Figure 6. Analysis of the ATP requirement for hMSH2-hMSH6 nucleosome disassembly ...... 64-65

Figure 7. The effect of intervening Lac I on hMSH2-hMSH6 nucleosome disassembly . 71

Figure 8. Two passive models for chromatin remodeling by hMSH2-hMSH6 ...... 75

Figure 9. Purification of nucleosome-DNA ...... 81

Figure 10. hMSH2-hMSH6 catalyzed nucleosome disassembly using nucleosome-DNA containing a G/C duplex ...... 82

Figure 11. Representative experimental and controls for Figure 6 ...... 83-84

Figure 12. ATP/ATPγS hydrolysis and the ability to provoke the formation of hMSH2- hMSH6 sliding clamps by ATP analogs ...... 85-86

Figure 13. Representative experimental and controls for Figure 7 ...... 87-88

Figure 14. hMSH2-hMSH6 nucleosome disassembly is enhanced by acetylation modifications and/or acetylation mimics ...... 96-97

Figure 15. Nucleosome disassembly by hMSH2-hMSH6 is dependent on the nucleosome localization sequence ...... 102

xii

Figure 16. Effect of nucleosome localization sequence variation at the nucleosomal-dyad axis on hMSH2-hMSH6 nucleosome disassembly ...... 105-106

Figure 17. Binding of hMSH2-hMSH6 to nucleosome-Mismatch substrates ...... 109-110

Figure 18. Disassembly of nucleosome-Mismatch substrates by hMSH2-hMSH6. 113-114

Figure 19. Nucleosome-DNA substrates and gel analysis of purified nucleosome-DNA substrates ...... 126

Figure 20. Gel analysis of purified nucleosome-DNA substrates reconstituted with high affinity nucleosome localization sequences ...... 127

Figure 21. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-DNA substrates reconstituted with high affinity 601 nucleosome localization sequence ...... 128

Figure 22. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-DNA substrates reconstituted with high affinity pMP2 nucleosome localization sequence ...... 129

Figure 23. Gel Analysis of purified nucleosome-DNA substrates reconstituted with variations in the nucleosome localization sequences ...... 130

Figure 24. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-DNA substrates reconstituted with variations in nucleosome localization sequences ...... 131-132

Figure 25. Nucleosome-Mismatch substrates and gel analysis of purified nucleosome- Mismatch substrates ...... 133-134

Figure 26. Nucleosome binding by hMSH2-hMSH6 to nucleosome-Mismatch substrates ...... 135-136

Figure 27. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-Mismatch substrates ...... 137-138

Figure 28. H3(K56Ac) enhances hMSH2-hMSH6 nucleosome disassembly ...... 142

Figure 29. Nucleosome-DNA substrates and gel analysis of purified nucleosome-DNA substrates ...... 145-146

Figure 30. Representative gels showing the nucleosome disassembly reaction catalyzed by hMSH2-hMSH6 ...... 147

Figure 31. H3(T118ph) crystal structure ...... 152

xiii

Figure 32. H3(T118ph) enhances hMSH2-hMSH6 nucleosome disassembly ...... 153

Figure 33. Preparation of nucleosomes ...... 155

Figure 34. hMSH2-hMSH6 disassembly of unmodified nucleosomes with a DNA mismatch and of H3(T118ph) containing nucleosomes without a DNA mismatch ...... 156

xiv

List of Abbreviations

AP Apurinic/Apyrimidinic sites

IDL(s) Insertion/Deletion Loops

DSB(s) Double-Strand Breaks

HRR Homologous Recombination Repair

NHEJ Non-Homologous End Joining

NER Nucleotide Excision Repair

BER Base Excision Repair

MMR Mismatch Repair

HNPCC Hereditary Non-Polyposis Colorectal Cancer

XP Xeroderma Pigmentosum

TTD Trichothiodystrophy

CS Cockayne’s Syndrome

6-4PPs 6-4 Photoproducts

CPD Cyclobutane Pyrimidine Dimers

TCR Transcription Coupled Repair

GGR Global Genome Repair

CRC Colorectal Cancer

LS Lynch Syndrome

MSI Microsatellite Instability

xv

EpCAM Epithelial Cell Adhesion Molecule

FAP Familial Adenomatous Polyposis

PCR Polymerase Chain Reaction

GI Gastrointestinal bp Basepair

E. coli Escherichia coli

MSH MutS Homologue

MLH MutL Homologue

PMS Post Meiotic Segregation

SSB Single-Strand Binding Protein

RFC Replication Factor C

RPA Replication Protein A

PCNA Proliferating Cellular Nuclear Antigen

HMG1 High Mobility Group Box 1 ssDNA Single-strand DNA

ExoI Exonuclease I

ExoVII Exonuclease VII

ExoX Exonuclease X

MNNG N-Methyl-N-Nitro-N-Nitrosoguanidine

MNU N-Methyl-N-Nitrosourea

N7-MeG 7-Methylguanine

N3-MeA 3-Methyladenine

xvi

06-meG 06 Methylguanine

NCP Nucleosome Core Particle

ARR Access-Repair-Restore

PTM Post-Translational Modifications

DDR DNA damage Response

FRET Fluorescence Resonance Energy Transfer

XPC-hHR23B Xeroderma Pigmentosum Group C-Human Rad23 Homologue B

CSA Cockayne Syndrome group A

CSB Cockayne Syndrome group B

UV-DDB UV-Damaged DNA-Binding Protein

TACSTD1 Tumor-Associated Calcium Signal Transducer 1

ADP Adenosine Diphosphate

ATP Adenosine Triphosphate

DAM DNA Adenine Methylase

PAGE Polyacrylamide Gel Electrophoresis

PMSF Phenylmethylsulfoynl Fluoride

CENPA Centromere Protein A

LRS Loss of rDNA Silencing

S. cerevisiae Saccharomyces cerevisiae

NPS Nucleosome Positioning Sequence

MSI-H MSI-High

ABC ATP Binding Cassette

xvii

PWWP Proline-Tryptophan-Tryptophan-Proline

X. laevis Xenopus Laevis

SWI/SNF Switch/Sucrose Nonfermentable

RSC Remodels the Structure of Chromatin

ISWI Imitation Switch

NURF Nucleosome Remodeling Factor

CHRAC Chromatin Accessibility Factor

ACF1 ATP-utilizing Chromatin Assembly and Remodeling Factor

CHD Chromodomain, Helicase, DNA Binding

INO80 Inositol-requiring 80

SIN SWI/SNF-independent

MMS Methyl Methanesulfonate

SWR1 SWI/SNF-related Protein

BZA Benzamidine

EPL Expressed Protein Ligation

HO Histone Octamer pMP2 Modified High Affinity 601 Positioning Sequence

MESNA Mercaptoethanesulfonic Acid

RP-HPLC Reverse Phase HPLC kDa Kilodalton

Da Dalton kb Kilobase

xviii

Preface

Some of the contents presented here have been published or are in preparation for publishing.

Chapter 1: Introduction

Chapter 2: Thesis Rationale

Chapter 3: Javaid, S. Manohar, M. Punja, N. Mooney, A. Ottesen, JJ. Poirier, MG. Fishel R. 2009. Nucleosome remodeling by hMSH2-hMSH6. Molecular Cell. 36:1086- 1094

Chapter 4: Javaid, S. Manohar, M. Punja, N. Mooney, A. Ottesen, JJ. Poirier, MG. Fishel R. 2010. hMSH2-hMSH6 nucleosome remodeling is highly dependent on post- translational modifications and nucleosome localization sequences. Manuscript in preparation

Chapter 5: North, JA. Javaid, S. Ottesen, JJ. Fishel, R. Poirier, MG. 2010. Contribution of Author: Histone H3(K56) acetylation enhances nucleosome remodeling by hMSH2-hMSH6. Manuscript in preparation

Chapter 6: North, JA. Javaid, S. Ferdinand, M. Chatterjee, N. Picking, J. Schoffner, M. Nakkula, R. Bartholomew, B. Ottesen, JJ. Fishel, R. Poirier, MG. 2010. Histone H3(T118) phosphorylation switches nucleosome remodeling. Submitted. Contribution of Author: Reconstitution and analysis of histone H3(T118) phosphorylation effect on nucleosome disassembly by hMSH2-hMSH6

Forties, RA. North, JA. Javaid, S. Tabbaa, OP. Fishel, R. Poirier, MG. Bundschuh, R. 2010. A quantitative model of nucleosome dynamics and applications to hMSH2- hMSH6 nucleosome disassembly. Submitted

Heinen, CD. Punja, N. Cyr, JL. Sakato, M. Javaid, S. Martin-Lopez, J. Hignorani, M. Fishel, R. 2010. hMSH2 controls the molecular switch activity of the hMSH2-hMSH6 mismatch repair recognition heterodimer. Manuscript in preparation

xix

Chapter 1

Introduction

I. DNA Repair Pathways

In mammalian cells, it has been estimated that 10,000 DNA lesions occur every day [1]. DNA damage can occur by endogenous mechanisms (i.e. polymerase errors, allele recombination) or by exogenous exposure to chemical and physical damage [2]. If these lesions are not repaired by the cell, random mutations may occur in the genome.

Some of the lesions caused by DNA damage can be classified as modified/damaged bases, apurinic/apyrimidinic (AP) sites, mismatched nucleotides, insertion-deletion loops

(IDLs), single-strand breaks in the DNA backbone, and double-strand breaks (DSBs) [3].

To combat challenges posed by DNA damage, cells have damage responses: i) removal of damaged DNA and restoration of the DNA duplex; ii) activation of DNA damage signaling; and iii) apoptosis [2]. Mammalian cells utilize five major DNA repair pathways to restore damaged DNA: homologous recombination (HRR), non-homologous end joining (NHEJ), nucleotide excision repair (NER), base excision repair (BER), and mismatch repair (MMR). MMR, BER, and NER are excision repair processes that depend upon the complementary DNA strand to faithfully direct repair of the lesion [4].

Mutations in MMR genes are associated with Lynch Syndrome or hereditary non- polyposis colorectal cancer (LS/HNPCC), while mutations in NER genes cause

1 xeroderma pigmentosum (XP), trichothiodystrophy (TTD), and Cockayne’s syndrome

(CS) [5].

I.a. Nucleotide Excision Repair

NER is largely responsible for excision of DNA lesions that distort the DNA backbone (i.e. 6-4 photoproducts (6-4PPs), cyclobutane pyrimidine dimers (CPDs), toxins, or cancer chemotherapeutics like cisplatin) [6]. However, only patients with XP are predisposed to UV sunlight-induced skin carcinomas [7]. NER and BER pathways overlap in their specificity for abasic sites and oxidative lesions [5]. The repair of damaged DNA involves at least 30 proteins within two different pathways of NER: transcription-coupled repair (TCR-NER) and global-genome repair (GGR-NER) [8-10].

TCR-NER refers to repair of lesions located in actively transcribed strand of genes by

RNA polymerase II [11]. GGR-NER repairs lesions in non-transcribed areas of the genome as well as non-transcribed strands of expressed genes [11]. Repair of DNA damage by NER occurs more frequently on the transcribed strand of DNA [12]. The main step that differs between GGR-NER and TCR-NER is recognition of the DNA lesion. In GGR-NER, XPC-hHR23B (XP group C-human Rad23 homologue B) and UV-

DDB (UV-damaged DNA-binding protein) complexes are responsible for the process while in TCR-NER, it is the stalling of RNA polymerase II that triggers a cellular response with the help of CSA (CS group A) and CSB (CS group B) [11]. In eukaryotes,

NER is carried out by excision of ~30 nucleotides [13].

2

I.b. Base Excision Repair

Lesions within the DNA can result from exogenous and/or endogenous events

(oxidative and alkylating damage, AP-sites, and single-strand breaks), which are repaired by BER [14]. These types of lesions do not significantly induce a distortion of the DNA to stall replication forks or transcription machinery [15]. The BER pathway is initiated by removal of the damaged base that is mediated by DNA glycosylases [15]. In bacteria, mammalian cells, as well as yeast, several DNA glycosylases have been identified, each specific for a limited number of distinct types of damaged bases [15]. Their activity induces formation of an AP-site that is processed by an AP-endonuclease or glycosylase- associated AP-lyase [5]. BER can be classified as “short-patch” BER and “long-patch”

BER. In short-patch BER, excision involves one nucleotide whereas in long-patch BER, excision involves 2-10 nucleotides [16].

I.c. Mismatch Repair

MMR is responsible for repairing mismatches and IDLs that occur during replication. Deficiency in this pathway is associated with increased genomic instability and LS/HNPCC [17]. MMR is conserved from bacteria to humans. In mammals, it involves two heterodimers that are homologous to the prototypical bacterial MutS and

MutL. The MutS homologues (MSH) contain MSH2 complexed with MSH6 (MSH2-

MSH6) or MSH3 (MSH2-MSH3) [18]. The MutL homologues (MLH) contain MLH1 complexed with PMS2 (Post Meiotic Segregation) (MLH1-PMS2) or MLH3 (MLH1-

MLH3) [18]. Both MSH2 and MLH1 have partial redundancy [17]. MSH recognizes and binds to mismatches or IDLs in the newly synthesized strand [19]. Upon addition of

3

ATP, MSH undergoes an ATP-driven conformation to form a hydrolysis-independent sliding clamp [20]. In one model, the MSH sliding clamp diffuses from the mismatch site and recruits MLH [21]. The MSH-MLH sliding clamp diffuses along the DNA backbone until MSH-MLH encounters downstream proteins (i.e. PCNA) [17]. ExoI is proposed to degrade the strand containing the mismatch [17, 22]. In addition to correcting replication errors, MMR also participates in recombination [17, 22].

Moreover, MMR proteins are responsible for cellular responses in reaction to DNA damage and participate in triggering cell cycle arrest upon exposure to damaging agents

[23]. As MMR is the focus of this thesis, the MMR pathway will be discussed in detail in subsequent sections.

II. Hereditary Non-Polyposis Colorectal Cancer

In the 2009, approximately one million patients were diagnosed with colorectal cancer (CRC). Approximately 3% (30,700 cases) will have Lynch Syndrome or hereditary non-polyposis colorectal cancer (LS/HNPCC), the most common hereditary

CRC [24]. LS/HNPCC was identified as a cancer susceptibility locus on chromosome

2p22-21 (MSH2) and 3p21 (MLH1) [25]. LS/HNPCC is defined as having a germline mutation in MMR genes: MSH2 (~38%), MLH1 (~54%), MSH6 (<10%), and PMS2

(<5%) [26]. Two of major MMR proteins, MSH2 and MLH1, are stabilized by interactions with other DNA MMR proteins (i.e. MSH6 and PMS2, respectively). Due to partial redundancy, MSH6 and PMS2 alterations are less common in LS/HNPCC. For

LS/HNPCC to be clinically verified, a defect inherited in MMR genes needs to be demonstrated [27]. LS/HNPCC is characterized by autosomal dominant inheritance, high 4 gene penetrance (~80-90%), early age of cancer onset (~45 years), and early onset of tumors in the proximal colon as well as extracolonic regions [28]. Germline mutations in at least one of the MMR genes can be found in 80% of all cases in LS/HNPCC [29].

Most of the MSH2 and MLH1 mutations are truncations, missense, nonsense, frameshift, and splice-junction mutations that cause loss of MMR function [30]. LS/HNPCC may be divided into two syndromes, depending on the presence or absence of extracolonic tumors, respectively [25].

LS/HNPCC has a spectrum of extracolonic tumors originating from the endometrium, ovary, stomach, bile duct, kidney, bladder, ureter, and skin [25]. The clinical hallmarks of LS/HNPCC resulted in a classification scheme designated

Amsterdam I and later modified to Amsterdam II to incorporate extracolonic cancers [31]

(Table 1).

In a defective MMR system, mutations occur frequently in small repetitive DNA sequences known as microsatellites [32]. The length of microsatellites varies within different regions of the genome [27]. Defects in MMR cause widespread changes in the length of microsatellite repeat sequences in tumors compared to normal samples [30].

This variation in length or size of microsatellites is called microsatellite instability (MSI)

[33]. Thus, MSI is considered a hallmark of LS/HNPCC since 95% of all LS/HNPCC- associated cancers show MSI [32]. However, some families with LS/HNPCC show

MSH6 mutations without the presence of widespread MSI in tumors [34]. MSI can be revealed by PCR (polymerase chain reaction) of microsatellite sequences that can evaluate for changes in microsatellite repeat lengths. Moreover, MSI is present in ~15%

5

Table 1. Amsterdam Criteria 1 and II and Bethesda Guidelines

Amsterdam I At least three relatives with verified colorectal cancer 1) One is a first-degree relative of the other two 2) At least two successive generations are affected 3) At least one of the relatives with colorectal cancer is diagnosed at <50 years of age 4) Familial adenomatous polyposis (FAP) is excluded Amsterdam II At least three relatives with LS/HNPCC-associated cancer (colorectal cancer, endometrial, stomach, ovary, ureter/renal pelvis, brain, small bowel, and skin) 1) One is a first-degree relative of the other two 2) At least two successive generations are affected 3) At least one of the relatives with colorectal cancer is diagnosed at <50 years of age 4) Familial adenomatous polyposis (FAP) is excluded 5) Tumors are verified Bethesda Guidelines for Testing Using Microsatellite Instability (MSI) 1) Colorectal cancer diagnosed in a patient <50 years of age 2) Presence of synchronous or metachronous colorectal cancer 3) Colorectal cancer with MSI-H (MSH-High) diagnosed in a patient <60 years of age 4) Colorectal cancer or syndrome associated tumor diagnosed in patient <50 years in at least one degree relative 5) Colorectal cancer of syndrome associated tumor diagnosed at any age in two first- or second-degree relatives

Adapted from: Clin Genet. 2009 July; 76(1): 1–18. [25]

6 of sporadic CRCs [35]. Sporadic CRCs is caused by somatic hypermethylation of the

MLH1-gene promoter or somatic mutation of both alleles at a MMR locus [32]. DNA methylation is an epigenetic modification that targets the cytosine at CpG dinucleotides

[32]. Regions in the genome that contain high frequencies of CpG dinucleotides are called CpG islands. CpG islands are present in approximately 40% of all human genes, including the MLH1 gene [36]. Hypermethylation of cytosine in CpG islands in the

MLH1 promoter inhibits gene transcription thereby mimicking an inactivating gene mutation [36-38]. If both copies of the genes are inactivated (bi-allelic hypermethylation), DNA MMR function of MLH1 is lost [32]. In addition to hypermethylation of MLH1, a new mechanism for germline MSH2 hypermethylation has been discovered [39]. MSH2 hypermethylation occurs by germline deletion of the last two exons of TACSTD1 (tumor-associated calcium signal transducer 1), a gene just upstream of the MSH2 encoding epithelial cell adhesion molecule (EpCAM), leading to inactivation of the MSH2 gene by promoter hypermethylation [39].

The development of mouse models that knock out MMR genes has helped to characterize MMR [40]. Msh2-/- mice have a reduced lifespan, high incidence of lymphomas (T-cell), small intestinal adenomas and adenocarcinomas, and exhibit high

MSI (MSI-H) tumors [40]. Msh6-/- mice have low incidence of late-onset GI

(gastrointestinal) tumors with moderate-high MSI [40]. Msh3-/- mice tumors have little or no MSI. Msh3-/- Msh6-/- mice have phenotypes similar to Msh2-/- mice and exhibit high

MSI. Mlh1-/- mice are similar to Msh2-/- mice [40]. These mouse models underline the significance of Msh2, Mlh1, and Msh6 in MMR and LS/HNPCC.

7

III. Mismatch Repair

Mutations normally arise at a frequency of 1 in 109 to 1010 bp per cell division

[41, 42]. Mutations can arise from misincorporation of nucleotides during DNA replication, chemically damaged nucleotides (i.e. 8-oxoguanine, carcinogen adducts, UV photoproduct), or strand slippage (IDLs) [43]. DNA polymerase base incorporation and proofreading results in an error rate of 10-7 bp per genome [17]. Mistakes that escape are corrected by MMR and increase fidelity 50-1000-fold [17]. The MMR pathway is conserved from bacteria to humans. In the mammalian MMR pathway, a mismatched nucleotide is recognized by MSH and that in combination with MLH, replaces the mismatched nucleotide on the newly synthesized strand via an excision repair reaction

[17]. MMR may eliminate severely damaged cells and prevent mutagenesis and tumorigenesis.

III.a. The Mutator Phenotype

The mutator hypothesis suggested that mutations in genes that maintain the genome result in genomic instability [44]. It is manifested by increases in mutation rates and in genetic evolution of cancer cells that drive cancer progression [44]. The loss of

MMR was proposed to contribute to tumorigenesis by creating cells that accumulated mutations at an increased rate [45]. However, loss of MMR did not lead to a growth advantage, as with classical tumor suppressors. Rather, it increased the likelihood that other proto-oncogenes and tumor suppressors would be mutated [46].

8

III.b. Prokaryotic Mismatch Repair

The bacterial MMR system is best characterized in Escherichia coli (E. coli) where it has been extensively studied genetically and biochemically. In E. coli, MMR is methyl-directed and occurs in an ATP/MutS/MutL/MutH-dependent manner [17].

Meselson and colleagues observed that mismatches can provoke their own repair with E. coli transfected with phage λ DNA containing mismatches [47, 48]. Mismatches on the λ

DNA were repaired with different efficiencies indicating that recognition is dependent on the mismatch [47, 48]. In E. coli, MMR recognizes and repairs G/T, A/C, G/A, T/C,

A/A, G/G, and T/T mismatches efficiently [49-51]. G/A and C/T mismatches can be weak substrates depending on sequence context [49]. C/C mismatches were subject to little or no repair [49]. IDLs up to five unpaired bases were also efficiently processed by the MMR pathway [52, 53].

The components of E. coli MMR are MutS, MutL, MutH, DNA helicase II

(UvrD), four exonucleases (ExoI, ExoVII, ExoX, and RecJ), single-stranded DNA binding protein (SSB), DNA polymerase III holoenzyme, and DNA ligase [54, 55].

MutS, MutL, and MutH initiate and play special roles in MMR. E. coli strains that are deficient in MutS, MutL, MutH, or UvrD are deficient in MMR [50, 56, 57]. To date, all of the MMR proteins required for successful MMR have been purified and the entire

MMR reaction reconstituted in vitro [54, 58-62]. Analysis of MMR in vitro and with E. coli extracts under conditions where DNA synthesis is blocked, indicates that the entire

MMR reaction can be divided into five steps: i) recognition of the mismatch; ii) mismatch-dependent incision of the unmethylated strand at a hemimethylated d(GATC)

9 site; iii) excision of the portion spanning the single-strand incision and the mismatch; iv)

DNA synthesis; and v) ligation [63].

In E. coli, mismatch-dependent incision and excision are controlled by the status of adenine methylation (N6 position of adenine is methylated) at d(GATC) sequences

[63]. In E. coli, newly synthesized DNA is subject to modification at d(GATC) sequences by Dam methylase after a transient delay (~1 minute) [64]. The presence of hemimethylated d(GATC) sequences distinguishes the daughter strand from the parental strand [57, 64-66]. MMR of hemimethylated DNA occurs on the unmodified strand (i.e. daughter strand). DNA, which lacks the Dam modification on either strand has little or no strand bias, whereas DNA that is methylated on both strands is not repaired [17]. E. coli, when deficient in Dam methylase, are mutators [56, 57]. In prokaryotic MMR, it is the hemimethylated d(GATC) sites that determines strand specificity for repair. The d(GATC) site may reside on either side of the mismatch indicating that MMR functions in a bidirectional manner.

MutS. MutS is responsible for the initiation and recognition of E. coli MMR [67].

MutS recognizes base-base mismatches as well as IDLs [67]. The footprint of MutS bound to a mismatch is 24-28 bp centered around the mismatch [68]. MutS is a 95kDa protein that possesses intrinsic ATPase activity [67]. The structure of MutS bound to a mismatch was determined by X-ray crystallography [69, 70]. Crystallography revealed that MutS binds to a mismatch as a homodimer [69, 70]. The subunits form a Θ where the mismatched DNA is located in the bottom channel [69, 70]. Only one subunit of the bacterial MutS protein makes contact with the mismatch. There is a conserved Phe-X-

10

Glu motif in bacterial MutS where the Phe residue intercalates with the mismatch [69,

70]. The carboxyl-group of the Glu residue hydrogen bonds with the mismatch [69, 70].

Furthermore, DNA is sharply kinked at 60° towards the major groove of the mismatch

DNA [69, 70]. Thus, MutS acts in an asymmetric fashion when bound to the mismatch.

This characteristic is mimicked by eukaryotic MutS homologues that function as heterodimers instead of homodimers. MutS possesses weak ATPase activity and mutations within the ATP binding domain result in a dominant mutator phenotype [71].

The presence of ATP in a binding reaction between MutS and mismatched DNA abolishes mismatch recognition specificity [21]. This demonstrated a close association between mismatch DNA and ATP-binding by MutS [21, 72].

MutL. MutL, a 68 kDa protein, is recruited to the mismatched DNA in a MutS- and ATP-dependent manner [73-75]. MutL is believed to act downstream of MutS in the initiation of MMR. The footprint of MutS-MutL is approximately 100 bp and protects

DNA on both sides of the mismatch [76]. The increase in the MutS-MutL footprint may reflect limited migration of MutS in the presence of ATP before MutL binds to MutS.

Alternatively, it may reflect multiple MutS-MutL sliding clamps clustered around the mismatch [76]. MutL physically interacts with MutS and actively recruits and activates

MutH [53, 77]. In E. coli, MutL does not possess endonuclease activity (MLH possesses endonuclease activity) [78]. Like MutS, MutL functions as a homodimer and contains intrinsic ATPase activity [79]. Mutations in the ATP-binding domains lead to a dominant negative mutator phenotype [80]. Recently, it was observed that MutL physically interacts with the clamp loader subunits of DNA polymerase III [81, 82], indicating that

11

MutL may promote binding of DNA polymerase III to MMR intermediates.

Additionally, MMR may be coupled to DNA replication [22].

MutH. Assembly of the MutS-MutL complex allows for activation of downstream activators. MutH, a downstream activator, is a 25-kDa latent endonuclease

[83, 84]. MutH binds to unmodified d(GATC) sequences. MutH is activated by the

MutS-MutL complex in a mismatch-specific and ATP-dependent manner [83, 84]. MutH incises the unmethylated strand of the hemimethylated d(GATC) 5’ to the G [85]. If both strands (daughter and parental) are unmethylated at d(GATC) sequences, MutH can cleave both strands resulting in a DSB [17]. The incision by MutH can occur 3’ or 5’ to the mismatch on the unmethylated, daughter strand. This results in a strand break that serves as an initiation signal for strand excision on the unmethylated strand towards the mismatch [56]. The excision tract from the strand break to the mismatch was confirmed by electron microscopy and in vitro reconstituted systems comprised of purified proteins in MMR [86, 87]. In vitro, it was observed that mismatch-provoked-excision was localized to the unmethylated strand where it extended from the d(GATC) site and terminated 100-250 nucleotides past the mismatch [86]. The excision tract can be as great as a 1000 bp but the efficiency of the reaction decreases as the excision tract increases to 2000 bp [85, 88, 89].

Other components. DNA helicase II (UvrD), exonuclease I (ExoI), exonuclease

VII (ExoVII), recombination protein J (RecJ), exonuclease X (ExoX), SSB, DNA polymerase III, and DNA ligase are also vital in MMR. The presence of a nick either 5’ or 3’ to the mismatch on the DNA revealed an orientation-specific requirement of

12 exonucleases in MMR reactions (independent of MutH) [90]. A nick located at the 5’- end requires ExoVIII or RecJ whereas a nick located at the 3’-end, requires ExoI or ExoX

[90]. DNA helicase II unwinds the nicked DNA strand toward the mismatch in a reaction stimulated by MutS, MutL, SSB, and ATP.

E. coli MMR Model. In the MMR pathway, mismatch recognition and binding occurs by MutS. A hemimethylated d(GATC) site 5’ or 3’ to the mismatch is located and cleaved by the concerted action of MutS, MutL, MutH, and ATP [17]. There are two classes of models proposed to address how mismatch binding of MutS leads to cleavage of the hemimethylated d(GATC) site. These models will be discussed in the eukaryotic

MMR section. The strand-specific nick at hemimethylated d(GATC) by MutH is the initial starting point for excision of the mismatched base. Via MutL, helicase II (UvrD), is loaded at the nick (on the unmethylated strand) and unwinds the DNA from the nick towards the mismatch [91]. This generates single-stranded DNA rapidly bound by SSB.

SSB bound to single-stranded DNA prevents nuclease attack [92]. Depending on the position of the strand break relative to the mismatch, ExoI (3’5’ exonuclease), ExoX

(3’5’ exonuclease), ExoVII (5’3’ exonuclease), or RecJ (5’3’ exonuclease), excises the unmethylated strand from the strand break to slightly past the mismatch. The resulting single-strand gap undergoes repair and ligation via DNA synthesis by DNA polymerase III, SSB, and DNA ligase [67].

Inactivation of RecJ and ExoVIII abolishes 5’-directed MMR in E. coli. Genetic inactivation of ExoI, ExoVIII, and ExoX eliminates 3’-directed MMR [93, 94]. Similar reactions were observed in vitro using purified MMR proteins [93, 94]. ExoVIII and

13

RecJ provide redundancy in 5’-directed MMR whereas ExoI, ExoVIII, and ExoX provide redundancy in 3’-directed MMR. Inactivation of SSB results in reduction of MMR [55].

DNA polymerase III supports the repair synthesis step and DNA integrity is restored by

DNA ligase [55].

III.c. Eukaryotic Mismatch Repair

As mentioned, the MMR pathway is conserved from bacteria to humans.

Similarities between the bacterial and eukaryotic MMR pathway include mismatch specificity, bidirectionality, and nick-directed strand specificity [67]. The role of MutH,

DNA helicase II, and the hemimethylated d(GATC) site for strand discrimination in E. coli are not conserved in eukaryotic MMR. In eukaryotes, a strand-specific gap or nick is sufficient to direct MMR in extracts of mammalian, Drosophila cells, as well as Xenopus egg extracts [17]. These findings, coupled with the observation that MMR is more efficient on the lagging strand of DNA at replication forks [95], suggests that DNA termini occurring during replication (3’ terminus on the leading strand, 3’ and 5’ termini on the lagging strand) may suffice as strand signals to direct correction of DNA errors

[95].

Multiple MMR proteins have been identified based on their homology to E. coli

MMR proteins (Table 2). The eukaryotic MMR components include MSH [96-99], MLH

[100, 101], ExoI [102-104], replication factor C (RFC), single-strand binding protein

RPA (replication protein A) [105], proliferating cellular nuclear antigen (PCNA) [106-

108], DNA polymerase δ [109], and DNA ligase I [62].

14

Table 2. Identity and functions of E. coli and human proteins involved in MMR E. coli Protein Function Human Homologues Function MutS Binds mismatches hMSH2-hMSH6 Repairs single base- base and 1-2 base IDL mismatches

hMSH2-hMSH3 Repairs some single base IDLs and large IDLs. Partially redundant with hMSH2-hMSH6

hMSH4-hMSH5 Participates in meiosis MutL Downstream affector hMLH1-hPMS2 Downstream affector that coordinates that coordinates multiple steps in events by binding to MMR MutS homologues

hMLH1-hMLH2 Function unknown

hMLH1-hMLH3 Participates in meiosis MutH Nicks nascent None unmethylated DNA at hemimethylated d(GATC) sites γ-δ complex Loads β-clamp onto RFC complex Loads PCNA, DNA modulates excision polarity Β-clamp Interacts with MutS PCNA Interacts with MutS and may recruit it to and MutL mismatches and/or homologues. Recruits replication fork. MMR proteins to Enhances processivity mismatches, increases of DNA polymerase mismatch binding III affinity of MutS homologues, participates in excision and maybe signaling, participates in DNA repair synthesis, participates in DNA re-synthesis Helicase II Loaded onto DNA at None nick by MutS and MutL. Unwinds DNA to allow excision of ssDNA 15

ExoI, ExoX Perform 3’5’ ExoI Excision of dsDNA excision of ssDNA RecJ, ExoVII Perform 5’3’ 3’ exo of polymerase Excision of ssDNA excision of ssDNA. δ, 3’ exo of ExoVII can also polymerase ε perform 3’5’ excision DNA polymerase III Accurate re-synthesis DNA polymerase δ Accurate repair of DNA synthesis SSB Participates in excision RPA Participates in and DNA synthesis excision and DNA synthesis DNA ligase Seals nick after DNA ligase Seals nick after completion of DNA completion of DNA synthesis synthesis

Adapted from: Mech Ageing Dev. 2008; 129(7-8): 391–407. [19]

16

Eukaryotic MutS and MutL homologues. Eukaryotic homologues of E. coli MutS and

MutL function as heterodimers [96, 101]. However, no yeast and human homologue of

MutH has been identified [17]. MutH has only been identified in gram-negative enteric bacteria (E. coli). Eukaryotic homologues of MutS are designated MSH1-MSH6. MSH1 is required for mitochondrial DNA stability in S. cerevisiae [110]. MSH2 heterodimerizes with MSH6 or MSH3 and is implicated in mitotic genetic stability [63].

Both MSH2-MSH6 and MSH2-MSH3 are ATPases that recognize and bind to a mismatch and/or IDLs to initiate repair. MSH2-MSH6 recognizes single mismatches

(including C-C), DNA damage by chemotherapeutic agents, and IDLs of one to two nucleotides [63]. Genetic studies in fission yeast with MSH2 and MSH6 mutants were associated with elevated levels of insertion mutations of G/T dinucleotide repeats [63].

MSH2-MSH3 recognizes larger IDLs (approximately 10 nucleotides) [63]. In eukaryotic cells, MSH2 dimerizes with MSH3 and MSH6. Studies have revealed that 85% of MSH2 in a cell dimerizes with MSH6 [111]. MSH4 and MSH5 function in meiosis as a heterodimer that plays an important role in crossing-over in yeast and mammals [112-

115].

The structure of human MSH2-MSH6 (hMSH2-hMSH6), determined by crystallography [116], is remarkably similar to the crystal structure of bacterial MutS [69,

70]. The structure of full-length hMSH2 with a protease-resistant fragment of hMSH6 lacking the first 340 amino acids (hMSH6Δ340) was crystallized [116] (Figure 1). hMSH2-hMSH6(Δ340) was analyzed in a complex with a mismatched G/T DNA [116]. hMSH2-hMSH6 is a 260 kDa complex [22] and a member of the ATP-binding cassette

17

Figure 1

Figure 1. X-ray crystal structure of hMSH2-hMSH6/ADP/G•T heteroduplex complex. Light Gray, hMSH6; Dark Grey, hMSH2; Orange Ribbon, DNA; Blue, ADP and Mg2+ ions. The ABC ATPase domain and the two channels in hMSH2-hMSH6 are indicated. Long α helices connect the ATPase and DNA binding domain.

Adapted from: Mol. Cell. 2007 May 25; 26(4):579-92. [116]

18

(ABC)-transporter superfamily [117]. Both hMSH2 and hMSH6 share common domain structure but vary in sequence and length: hMSH2 is 104 kDa whereas hMSH6 is 156 kDa [116]. Bacterial MutS share limited sequence identity with hMSH2-hMSH6. The

MutS homodimer has 21% and 24% identical to conserved regions with hMSH2 and hMSH6, respectively [116]. Moreover, MutS is approximately 600 amino acids smaller than hMSH2-hMSH6. The hMSH2-hMSH6(Δ340) heterodimer forms an asymmetric dimer (approximately 125 Å tall, 110 Å wide, and 65 Å thick) with two channels with hMSH2 and hMSH6 lining the sides [116]. The two ATPase domains, contributed by hMSH2 and hMSH6 are located at the C-terminus. The mismatch binding domain

(contributed by hMSH6) is located at the N-terminus. The mismatched G/T DNA bound by hMSH2-hMSH6 is located in the larger of the two channels, which is the farthest away from the ATPase domain and closest to the mismatch binding domain [116]. The hMSH6 dimer makes specific contacts with the G/T mismatch whereas hMSH2 makes one contact with the DNA backbone. hMSH2 and hMSH6 are divided into five domains: mismatch binding, connector, lever, clamp, and the ATPase domains [116].

The mismatch binding domain contains amino acids 1-124 and 362-518 of hMSH2 and hMSH6, respectively [116]. The domains are a mixture of α/β structure

[116]. The mismatch binding domain differs from that of E. coli MutS. The non- mismatch binding domain in MutS (equivalent to hMSH2) makes extensive contact [69,

70] with the DNA backbone, whereas hMSH2 makes one contact [116]. The mismatch binding domain of hMSH2 is rotated away from the DNA and packs against the mismatch binding domain of hMSH6. The hMSH6 mismatch binding domain interacts

19 with the G/T mismatch in the DNA. In the crystal structure of hMSH2-hMSH6(Δ340), the mismatched base remains intrahelical and the Phe of the Phe-X-Glu mismatch recognition motif of hMSH6 intercalates into the DNA via the minor groove to stack with the mismatched G/T base (Phe36 and Phe432 in E. coli MutS and hMSH6, respectively)

[116]. The Phe and Glu residues are conserved from bacterial MutS to eukaryotic hMSH6 [116]. The substitution of PheAla of hMSH6 results in a drastic decrease in the mismatch recognition ability of hMSH2-hMSH6 [116]. Glu434 residue of the conserved Phe-X-Glu motif hydrogen bonds with the mismatched thymine, sandwiched between Phe432 and Met459 [116]. The backbone carbonyl of Val429 accepts a hydrogen bond from the mismatched G [116]. These interactions along with nonspecific protein-DNA interactions widens the minor DNA groove at the G/T mismatch [116].

This tilts the thymine of the mismatch so that its O4 carbonyl interacts with the mismatched G [116]. These interactions with the G/T mismatch result in a kink of about

45˚ in the DNA backbone [116].

The ATPase domains of hMSH2 and hMSH6 share 48% sequence homology with

E. coli MutS [116]. The domains are a mixture of α/β structures [116]. Genetic and biochemical characterization of mutations of the conserved ATPase domain have demonstrated a central role of these ATPase domains. The ATPase domains require a mismatch to form stable dimer interfaces and disengage ADP [116]. Loss of hMSH2- hMSH6 dimerization results in loss of mismatch specific ATPase activity. The ATPase sites in hMSH2-hMSH6 have different nucleotide affinities [118]. The ATPase binding

20 domain and mismatch binding domain are connected by long α helices that act as levers

[116].

The N-terminal region of hMSH6 contains an extended 340 amino acids compared to the N-terminus of E. coli MutS [116]. The N-terminus of hMSH6 and/or hMSH3 but not hMSH2, contains a conserved PCNA binding motif (PIP box). The PIP box will be discussed in the PCNA section. In addition to the PIP box, the N-terminus of hMSH6 has another conserved motif called the PWWP (Proline-Tryptophan-Tryptophan-

Proline) domain [119]. The PWWP domain is thought to be a potential protein-protein or protein-chromatin interaction domain [116].

There are four human MutL homologues: hMLH1, hMLH3, hPMS1, and hPMS2. hMLH1 heterodimerizes with hPSM2, hPMS1, and hMLH3 [100, 120-122]. hMLH1- hPMS2 is required for MMR and accounts for ~90% of hMLH1 in human cells [17]. hMLH1-hMLH3 plays a role in meiosis whereas no significant biological activity of hMLH1-hPMS1 has been discovered [100, 120-122]. hMLH1-hPMS2 binds to hMSH2- hMSH6 and forms a ternary complex required for MMR. hMLH1-hPMS2 has intrinsic

ATPase activity [63]. In the human MMR system, hMLH1-hPMS2 harbors latent endonuclease activity that nicks the discontinuous strand of the mismatched DNA in a hMSH2-hMSH6-, PCNA-, RFC-, and ATP-dependent manner [78]. In eukaryotic 3’- repair, the pre-existing strand discontinuity does not serve as an entry point for ExoI to support the following excision, the entry point for ExoI is introduced by the endonuclease activity of hMLH1-hPMS2 [78]. The DQHA(X)2E(X)4 motif located in the C-terminus of the hPMS2 subunit of hMLH1-hPMS2 comprises the metal-binding site that is

21 essential for endonuclease activity of hMLH1-hPMS2 in MMR [78]. Amino acid substitution of this motif abolishes endonuclease activity of hMLH1-hPMS2 [78]. This motif is present in eukaryotes, archael, and eubacterial MutL proteins but is absent from gram-negative bacteria like E. coli that rely on d(GATC) methylation [78].

PCNA. PCNA plays a role in the initiation of MMR, DNA excision, and resynthesis steps of MMR [22]. Moreover, PCNA physically interacts with hMSH6 and hMSH3 via a conserved PIP box [22]. Mutations in the PIP box of hMSH6 abolish interactions between PCNA and the MSH2-MSH6 heterodimer resulting in a partial mutator phenotype in yeast [123]. In gel-shift experiments, it was observed that PCNA increases specific binding of yeast MSH2-MSH6 to mismatches [124]. Another study observed that PCNA transferred yeast MSH2-MSH6 onto a mismatch [125]. Cytological studies with MSH3 and MSH6 demonstrate that these proteins co-localize with the replication fork [126]. This demonstrates that PCNA may help load MutS homologues onto newly replicated daughter strands at mismatch sites. PCNA is essential in 3’ nick- directed MMR but is not essential in 5’ nick-directed MMR [127]. Depletion of PCNA by p21, a cell cycle-regulated protein that tightly binds and sequesters PCNA, abolishes

3’-directed MMR but inhibits 5’-directed MMR approximately 50% [60]. PCNA, with the help of the clamp loader, RFC, loads onto the 3’-end of Okazaki fragments or the 3’- end of the leading strand [58, 62].

Exonuclease I. In E. coli MMR, multiple exonucleases are implicated in the excision step. In human cells, only one exonuclease is implicated in MMR. The ExoI protein is a member of the Rad2 family and has 5’3’ polarity and 5’ flap endonuclease

22 activity [17]. ExoI is involved in both 5’- and 3’- directed MMR and interacts with hMSH2 and hMLH1. ExoI carries out 5’ mismatch excision in the presence of hMSH2- hMSH6 and RPA. In 3’ mismatch excision, ExoI requires hMSH2-hMSH6 and hMLH1- hPMS2 that is activated by PCNA and RFC [22]. ExoI-deficient mice display modest cancer predisposition and ExoI-deficient cells exhibit MSI and a mutator phenotype

[128]. Exo1 deletion strains of S. cerevisiae exhibit a mild mutator phenotype.

Other Components. Other components in human MMR include RPA, RFC

(PCNA clamp loader), HMGB1 (high mobility group box I protein), and DNA polymerase δ [22]. RPA is involved in all aspects of MMR: binding to nicked DNA, stimulating MMR-provoked excision, protecting ssDNA gaped region during excision, and facilitating DNA synthesis [22]. DNA polymerase δ binds to DNA gapped substrate for DNA resynthesis. Upon addition of DNA polymerase δ, RPA is phosphorylated, reducing RPA’s affinity for ssDNA [129]. Unphosphorylated RPA possesses high DNA binding affinity, stimulates the excision step in MMR more efficiently, and protects ssDNA from nuclease attack [129]. Unphosphorylated RPA may also displace the hMSH2-hMSH6/hMLH1-hPMS2 tetramer from the DNA. Phosphorylated RPA facilitates DNA synthesis by DNA polymerase δ [22]. Additionally, HMGB1, a non- histone chromatin protein, may play a role in the initiation of mismatch-provoked excision in nuclear extracts [130].

Human MMR Model. The human MMR reaction has been reconstituted in vitro using purified human proteins [58, 62]. The simplest MMR pathway is comprised of hMSH2-hMSH6, hMLH1-hPMS2, ExoI, and RPA that supports excision with 5’3’

23 directionality when the nick is located 5’ to the mismatch [131]. hMLH1-hPMS2 is not necessary for 5’-directed MMR reaction but it enhances mismatch dependence. The hMSH2-hMSH6 protein activates ExoI hydrolysis on a mismatched DNA containing a nick 5’ to the mismatch in a mismatch- and ATP-dependent manner [60, 131]. In the absence of RPA, hMSH2-hMSH6 stimulates ExoI hydrolysis of nicked DNA, making

ExoI highly processive [60]. In the presence of RPA, ExoI processivity is reduced from approximately 2000 nucleotides to 250 nucleotides [60]. This reduction leads to termination of the MMR reaction upon excision of the mismatch. Additionally, RPA- bound DNA is a poor substrate for ExoI reloading. hMSH2-hMSH6 promotes reloading of ExoI provided that the DNA contains a mismatched nucleotide [60]. Upon mismatch removal, ExoI excision is highly attenuated because hMSH2-hMSH6 is unavailable to assist. This effect of hMSH2-hMSH6 suppresses ExoI hydrolysis on DNA lacking a mismatch leading to termination of excision by ExoI just past the mismatch site [60].

Thus, excision on 5’-mismatched DNA (nick located 5’ to the mismatch) proceeds via a set of intermediates that differ in size of about 250 nucleotides, an effect attributed to multiple loading of ExoI by hMSH2-hMSH6 [60]. The four protein system (hMSH2- hMSH6, hMLH1-hPMS2, ExoI, and RPA) has 5’3’ directionality because ExoI hydrolysis proceeds 5’3’ from the strand break regardless of the nick location (5’ or 3’ to the mismatch).

A purified human MMR system that supports bidirectional excision has six components: hMSH2-hMSH6, hMLH1-hPMS2, ExoI, RPA, PCNA, and RFC. When the nick is located 5’ to the mismatch, excision proceeds in the 5’3’ direction. However on

24 a 3’-mismatch substrate (nick is located 3’ to the mismatch), hydrolysis proceeds 5’3’ from the strand break that is in the wrong polarity for mismatch removal [90]. The default polarity of ExoI in MMR is 5’3’. PCNA does not have a significant effect on the restricted directionality of MMR. Upon addition of PCNA and RFC (RFC loads

PCNA onto the DNA), MMR supports mismatch removal in both 5’ and 3’ mismatched-

DNA (nick located either 5’ or 3’ to the mismatch) [90]. When the nick is located 3’ to the mismatch, ExoI 5’3’ hydrolysis initiating at the nick is repressed by RFC and excision occurs with apparent 3’5’ polarity resulting in mismatch removal [78]. In previous studies, it was observed that an ExoI active site mutant failed to support 5’- and

3’- directed MMR. Thus, mismatch removal was attributed to an unknown exonuclease with cryptic ExoI 3’5’ hydrolytic function responsible for the 3’5’ excision of the mismatch [90]. However, the idea of an unknown exonuclease containing cryptic 3’5’ hydrolytic activity was rendered moot when it was discovered that hMLH1-hPMS2 possesses latent endonuclease activity activated by hMSH2-hMSH6, RFC, and PCNA in an ATP- and mismatch-dependent manner [90]. The incision by hMLH1-hPMS2 occurs on both 3’ and 5’ mismatched DNA (nick located 3’ or 5’ to the mismatch) and is biased to the nicked DNA strand [78]. In the case of a 3’-mismatched DNA (nick located 3’ to the mismatch), incision is distal to the mismatch providing an initiation site for mismatch removal by the 5’3’ action of hMSH2-hMSH6 activated ExoI [78]. This implies that the nick directing repair serves as a strand signal but not as a site for excision initiation by ExoI. Thus, excision initiation by ExoI occurs at a strand break produced by hMLH1-

25 hPMS2 [78]. This mode of excision is different from E. coli methyl-directed pathway, where hydrolysis initiates at a nick located either 3’ or 5’ from the mismatch [17].

III.d. Signaling Models for Strand Discrimination in MMR

There are two classes of models that explain how information is propagated from the mismatch site to the excision site located some distance away: in cis (moving) or in trans (stationary) [132] (Figure 2). In a trans activation model (Figure 2), communication occurs between the MutS-MutL complex at a mismatch site and proteins that operate downstream (i.e. MutH) by a protein-protein interaction facilitated by DNA bending [132]. In this model, the MutS-MutL complex remains bound to the mismatch.

This was based on experiments showing MutH activation and DNA cleavage with the mismatch and the d(GATC) site on two different DNA strands [76]. Further evidence was provided when HeLa cell extracts were not substantially inhibited by the presence of a barrier (i.e. DNA hairpin or biotin-streptavidin) between the mismatch and the initiating nick [133, 134]. However, these extracts were subsequently shown to remove the biotin block [77].

In the cis activation model, MutS-MutL complexes utilize ATP and the DNA helix to travel down the DNA until encountering a strand break or downstream proteins

(i.e. PCNA). MutS-MutL then activate the excision step (Figure 2) [132]. There are two cis models: “ATP-dependent translocation” and the “molecular switch” model [90]. In the ATP-dependent translocation model [135], ATP reduces the mismatch binding affinity of MutS. ATP hydrolysis drives unidirectional translocation of MutS proteins along the DNA [135]. The mismatched DNA is threaded through the protein complex

26

Figure 2

Figure 2. Cis and trans models for mismatch repair. In a cis mechanism, an intact DNA helix is required so that a protein molecule can communicate (i.e. slide) along the DNA helix with the second site. In a trans mechanism, contacts occur between two sites and the intervening DNA is looped out.

Adapted from: Proc Natl Acad Sci U S A. 2007 August 7; 104(32): 12953–12954. [132]

27 until the strand discrimination signal, forming a DNA loop [135]. In the molecular switch model [136], mismatch binding by MutS triggers a conformational change that allows an ADPATP exchange; this exchange allows a second conformational change of MutS forming a hydrolysis-independent sliding clamp [136]. The sliding clamp can diffuse bidirectionally leaving the mismatch site open for iterative loading of multiple

MutS [136]. In this model, it is the binding of ATP, not ATP hydrolysis that signals downstream events, including formation of a ternary complex with MutL and sliding of multiple MutS-MutL clamps from the mismatch site to the strand break [136]. Moreover,

Zhang et al. [62] show that multiple MSH-MLH complexes are essential for processing a single mismatch providing evidence for the molecular switch model.

Recent studies by Paul Modrich have argued in favor of the “moving” rather than the “stationary” mechanism [77]. The Modrich lab demonstrated that a protein

“roadblock” (i.e. EcoRI (E111Q) – a hydrolytically defective form of EcoRI endonuclease) between the mismatch and nick inhibits in vitro MMR with recombinant

E. coli proteins [77]. Nucleotide binding site occupancy regulates MutS interaction with

DNA. MutS undergoes a conformational change upon binding to ATP [17]. However, the effects of adenine nucleotides on MutS and DNA are not well understood. Initial attempts to evaluate the effect of ATP on MutS-DNA interaction used visualization of complexes with MutS, mismatched and homoduplex DNA [135]. The experiment demonstrated mismatch- and ATP-dependent formation of α-shaped DNA loop structure up to several kilobases in size, stabilized by MutS at the base [135]. Loop size increased with time and the mismatch was present within the loop [135]. MutL enhanced the rate

28 of MutS-mediated DNA loop growth; both MutS and MutL are bound at the base of the

α-loop structures [63]. Non-hydrolyzable ATP analogues failed to support loop formation. Furthermore, loop growth stops after addition of excess non-hydrolyzable

ATP analogs [135]. These effects are contributed to the ATP-dependent translocation model in which MutS leaves the mismatch in a unidirectional manner along the DNA

[135]. However, these experiments are complicated by the use of low NaCl concentration (at which heteroduplex DNA does not stimulate hMSH2-hMSH6 activity above homoduplex DNA) [135, 137]. This suggests that the interpretation of these results must be regarded as uncertain. Another shortcoming of this model is that it is difficult to reconcile with the modest rate of ATP hydrolytic turnover by MutS homologues.

Another mechanism for nucleotide binding was demonstrated using a 41-mer mismatched DNA by placing physical barriers at ends of DNA [20]. MSH dissociates rapidly from mismatched DNA with biotin tags at both termini upon challenge with ATP and magnesium [20]. Dissociation of ATP-bound MSH is blocked if the terminal biotins are bound by streptavidin [20]. Also, DNA structures such as four-way junctions, hairpins, and lac repressor placed near the end of DNA inhibit ATP-induced release of

MSH [20, 138]. Furthermore, blocking the ends of heteroduplex DNA inhibits 90% of

MSH ATPase activation [20, 21]. This effect was interpreted as movement of MSH from the mismatch along the DNA and led to the initial suggestion that MSH may form a clamp-like structure around the DNA [20]. In the molecular switch model, MSH initially binds to mismatched DNA in an ADP-bound state. ATP binding by MSH results in the

29 formation of a stable hydrolysis-independent sliding clamp that is capable of diffusing for at least 1 kb along the DNA [21]. Moreover, under certain conditions, ATPγS (a non- hydrolyzable analog of ATP) challenge of end-blocked (DNA is blocked at both end)

MSH results in long-lived mobile complexes [20, 138].

IV. Mismatch Repair and DNA damage signaling

Cell cycle arrest is important for preventing DNA-damaged induced instability. A large number of studies have characterized G2- or S-phase checkpoints and the proteins involved (i.e. ATM, ATR, p53, p73, Chk1, and Chk2) [22]. In addition to repair of mismatches, the MMR system also signals and/or processes different types of DNA damage. MMR-deficient cells are more resistant to apoptosis induced by several different chemicals than MMR-proficient cells [139]. Studies show that functional MMR is required for cytotoxicity of specific alkylation chemicals (i.e. N-methyl-N-nitro-N- nitrosoguanidine (MNNG) and N-methyl-N-nitrosourea (MNU)) [140]. SN1 alkylating agents produce DNA damage such as 7-methylguanine (N7-MeG), 3-methyladenine (N3-

MeA), and O6-methylguanine (O6-meG) [140]. Cells lacking MMR are 100-fold more resistant to apoptosis induced by methylating agents [139]. Tolerance to cytotoxic effects of SN1 alkylators occur in E. coli strains deficient in MMR and MMR-deficient mammalian cells [141]. In cells lacking MMR, p53 and p73 are not phosphorylated in response to DNA damage [142, 143]. p53 and p73 are phosphorylated by ATM, ATR, or c-Abl; each interact with hMSH2-hMSH6 and hMLH1-hPMS2 [142, 143]. This implicates hMSH2-hMSH6 and hMLH1-hPMS2 in a signaling cascade that leads from

DNA damage to cell cycle arrest and apoptosis. 30

There are two models that suggest a role for MMR in DNA damage signaling

[19]. The first model is the “futile DNA repair cycle” model. In this model, MMR recognizes damage (i.e. O6-meG) and engages in a futile cycle that activates the DNA damage pathway and apoptosis. MSH interacts with MLH to initiate excision of the damaged strand. As the damage is on the parental strand, which is not excised by MMR, abortive repair cycles would lead to generation of intermediate structures, nicks, and single-strand gaps resulting in DSBs and force checkpoint activation [144]. This model is supported by exposure of cells to MNNG that induce DNA breaks/gaps, cell cycle arrest, and nuclear foci at sites of DNA damage [144, 145]. Furthermore, studies in vitro of MMR using mammalian proteins are consistent with iterative rounds of excision repair

[146].

The second model, “direct signaling”, argues that hMSH2-hMSH6/hMLH1- hPMS2 directly triggers DNA damage signaling by recruiting ATM or ATR/ATRIP to the lesion activating the checkpoint response. This model is supported by direct interaction of ATR and ATRIP with hMSH2-hMSH6/hMLH1-hPMS2 in the presence of

O6-meG opposite a T [147]. Moreover, there are two different separation-of-function mutant mouse models, Msh2G674A and Msh6T1217D that support the “direct signaling” model. Both mouse models lose MMR function but retain apoptotic responses implying that proper repair function is unnecessary for triggering the DNA damage signaling pathway [19].

31

V. Chromatin and Mismatch Repair

DNA damage detection, signaling for checkpoint responses, and DNA damage repair processes all take place in the context of chromatin. The basic repeating unit of the chromatin is the nucleosome core particle (NCP) that consists of ~147 bp of DNA wrapped around an octamer of four proteins (2 copies of H2A, H2B, H3, and H4). The core histones have an unsaturated N-terminal tail containing residues subject to modifications (i.e. acetylation, phosphorylation, methylation, and ubiquitination) [148].

The compact structure of chromatin hinders nuclear processes such as transcription, replication, and repair.

All cells have evolved mechanisms to interact and remodel chromatin to alter access to DNA. The first evidence of chromatin structure alteration due to DNA damage was observed in NER processes following UV damage [149]. This report led to a model called “Access-Repair-Restore” (ARR) describing how DNA repair occurs in a chromatin environment [150].

ARR hypothesizes that chromatin organization is locally destabilized by DNA damage to accommodate repair machinery. After completion of DNA damage repair, the chromatin configuration is restored [151]. Due to the highly condensed nature of chromatin, it was considered that chromatin posed a constraint on cellular responses, causing DNA lesions to be difficult to detect and repair. Studies in the past decade show that chromatin is an active participant in DNA repair through covalent modification of histone tails and chromatin structure rearrangements such as deposition or evictions of histones [148, 151]. Chromatin remodelers are characterized into two different groups: 32 histone modifiers and ATP-dependent remodelers. Histone modifiers and ATP- dependent remodelers participate in DNA repair [152, 153].

Histone modifiers catalyze the attachment or removal of post-translational modifications (i.e. acetylation, methylation, and phosphorylation) [154]. The result is control of the condensation state of chromatin via alteration of DNA-histone contacts and/or recruitment of non-histone proteins to chromatin [155, 156]. ATP-dependent remodelers use the energy of ATP-hydrolysis to alter chromatin structure by disrupting

DNA-histone contacts. Disruption of DNA-histone contacts can reposition or slide the nucleosomes, changing the accessibility of DNA to other proteins [157]. For the scope of this thesis, my focus will be the structure of the nucleosome, histone post-translational modifications, and chromatin remodeling involved in DNA replication and/or repair.

V.a. Nucleosome Structure

The nucleosome consists of ~147 bp of DNA wrapped ~1.7 times around eight histone proteins. The nucleosome is the basic repeating unit of chromatin. Nucleosomes are separated by a linker DNA of variable length that may be associated with linker histone H1. The core histones are characterized by the presence of a histone-fold domain

(composed of three α helices connected by two loops) and N-terminal tails of variable length subject to post-translational modifications (PTMs) [158]. Core histones are small

(11-16 kDa) basic proteins that are highly conserved in length and sequence [159]. A prominent feature of the histone octamer is the positively charged surface [159]. There are >120 direct atomic interactions between histones, DNA, and water [159]. Direct protein-DNA interactions are not spread evenly about the octamer surface, but are located

33

Figure 3

Figure 3. Crystal structure of nucleosome core particle. Ribbon traces for the 146-bp DNA phosphodiester backbones (orange) and eight histone protein main chains (green).

Adapted from: 1AOI - pdb

34 at discrete sites. There are 14 sites where the nucleosome makes direct contact with the

DNA [159]. DNA binding occurs at the sugar phosphate backbone over short stretches of each helical turn where the minor groove faces inward towards the histone octamer

[159]. The structure of the nucleosome was crystallized using Xenopus laevis (X. laevis) recombinant histones and an asymmetric 147 bp DNA fragment derived from human α- satellite DNA [160, 161] (Figure 3). The histone-fold domains organize the central 121 bp of DNA with an additional 13 bp at each end organized by the N-terminal extension of

H3 [162]. The twist of free B-form DNA in solution is 10.5 bp per run; the overall twist of nucleosomal-DNA is 10.2 bp per turn [159].

Histone H3-H4 forms a tetramer, whereas H2A-H2B forms two dimers. Two

H2A-H2B dimers interact with the H3-H4 tetramer via two H2B-H4 interactions to complete the histone octamer that is assembled in the presence of DNA or high salt conditions [161]. H3 also makes contacts with DNA at points of entry and exit from the nucleosome. The core histones make contact with the DNA primarily through the phosphodiester backbone [159]. The lack of specific contacts between the core histones and the DNA backbone explains how nucleosomes can pack DNA in a sequence- independent manner. In addition to the histone-fold domains, the four core histones have disordered tails that protrude from the nucleosomal core. These tails are rich in basic residues and are subject to multiple PTMs. Structural similarities between the core histones reveal similarities between various organisms [159, 163]. H2A and H2B are the most variable pair, whereas H3 and H4 are the most conserved.

35

A single molecule technique (DNA unzipping) directly measures histone-DNA energetics [164, 165]. DNA unzipping is a force-feedback apparatus that pulls apart two strands of a single duplex DNA (unzipping the Watson-Crick bp) [166]. DNA unzipping revealed distinct histone contacts that occur approximately every 5 bp intervals around the nucleosome; the two strands of DNA at each minor groove site contribute independently to histone binding rather than as a collective unit [164, 165]. Unzipping also revealed three strong interactions of histone-DNA: a strong central contact around the nucleosomal-dyad (the center of the nucleosome around which there is an overall pseudo two-fold symmetry) and two lesser contacts about 50-60 bp away on either side

[166].

V.b. Nucleosome Positioning

Naturally occurring DNA sequences can be packaged into nucleosomes whereas dsRNA, RNA-DNA hybrids, and Z-form DNA cannot. In most cases, DNA sequences wrapped around the nucleosome adopt a preferred rotational and translational positioning relative to the histone octamer [162]. Translational positioning refers to the 146-147 bp sequence occupied by the histone octamer whereas rotational positioning is referred to as the face of the DNA that contacts histone octamer [162].

Early experiments using 5S rDNA sequence demonstrated that DNA was similarly positioned on nucleosomes formed from histone octamer of chicken, yeast, and frog [167, 168]. Mutated 5S rDNA sequence indicates that the central 40-60 bp is central for positioning DNA sequence on the nucleosome [169]. This central region must accommodate sharp bends and distortion seen at the nucleosome-dyad. The 5S rDNA

36 sequence consists of short runs (4-6 bp) of oligo d(A):oligo d(T) per helical turn, imparting a natural curvature to the DNA, making it a favored sequence to wrap around the octamer [170]. Previous studies demonstrate that AA, TT, and TA dinucleotides occur at 10 bp intervals on nucleosome positioning sequences [171]. GC dinucleotides were observed every 10 bp but were offset by 5 bp with the AA, TT, and TA dinucleotides [171]. This pattern of 10 bp periodical provides a rotational setting of the

DNA on the histone surface because AA or TT dinucleotides tend to expand the major groove of the DNA, whereas GC dinucleotides tend to contract the major groove [171].

This might facilitate DNA wrapping around the histone octamer when the dinucleotides are in phase with the helical twist of DNA.

V.c. Histone Variants

Nucleosomes can be modified in their composition, structure, and location by chromatin remodeling complexes that introduce PTMs to core histones [159].

Additionally, chromatin can also be modified by incorporation of histone variants [172].

Histone variants change the local structure of chromatin by promoting nucleosome subunit exchange to facilitate cellular processes such as transcription, replication, and development.

Histone H2A. Histone H2A is the core histone with the largest number of variants. The variants of H2A are H2ABbd, MacroH2A, H2AZ, and H2AX. These variants are characterized by divergent C-terminal sequences and genome localization [158]. H2AZ is implicated in transcriptional activation in yeast [173]. H2ABbd localizes with transcriptionally active chromatin and is excluded from inactive X chromosomes [174].

37

MacroH2A is found primarily in inactive X chromosomes [175]. H2AX is involved in

DNA repair and recombination [176].

Histone H2B. Histone H2B variants are few in number and those that occur have specialized roles in chromatin compaction during gametogenesis. However, their specific roles remain unknown.

Histone H3. Histone H3 variants include H3.1, H3.2, H3.3, CENPA (Centromere protein

A), and H3.4. H3.3 is a histone variant that is not S-phase regulated and is found in transcriptionally active chromatin [177]. H3.4 is testis-specific, whereas CENPA is localized on centromeric chromatin [158]. H3.1 is coupled to DNA synthesis during

DNA replication [172]. H3.2 might be involved in transcriptional silencing [172].

Histone H4. Histone H4 is the most highly conserved histone. H4 makes extensive contacts with the other three core histones in the nucleosome. H4 has no known sequence variants.

V.d. Post-translational Modifications of Histones

There are at least eight different types of modifications that modify histones: acetylation, methylation, phosphorylation, ubiquitylation, sumoylation, ADP- ribosylation, deimination, and proline isomerization [178]. Histones are modified at multiple sites. In histones, over 60 different amino acids with modifications have been detected either by antibodies or by mass spectrometry. PTMs can be complicated by the fact that methylation at lysines or arginines can come in three different forms: mono-, di-, trimethyl for lysines or mono-, or dimethyl for arginines [178]. Complexity also arises because not all the modifications will be on the same histone at the same time. The

38 timing of PTMs will be dependent on signaling conditions of the cell. There are two mechanisms for the function of PTMs: i) disrupt contacts between histone and DNA in order to loosen the chromatin; and ii) recruitment of non-histone proteins [178].

During DNA repair, a number of histone modifications occur. The functional significance of these modifications is not fully understood. Histone modifications influence chromatin structure by changing the contacts of chromatin through structural histone changes, influencing electrostatic interactions, and recruiting non-histone proteins to chromatin [178]. For the scope of this thesis, I will focus on PTMs coupled with

DNA replication and/or DNA repair.

Acetylation. Different lysines in both histone H3 and H4 are targets for acetylation that neutralize the basic charge of lysine. Acetylations potentially alter interactions between adjacent histones and/or between histones and DNA. In response to UV irradiation, histones become hyperacetylated and DNA repair more efficient [179, 180]. This suggests that changes in chromatin structure induced by acetylation make the nucleosomal-DNA more accessible for DNA repair activities [180].

Histone H3K56 acetylation is a PTM tightly coupled with DNA replication and

DNA repair. Studies have observed a connection between H3K56 acetylation and chromatin assembly following DNA replication and DSB repair [181]. In budding yeast,

H3K56 acetylation is deposited on newly synthesized histones during S phase. It was observed that in the absence of damage, H3K56 acetylation disappears in G2. However, when DNA damage is present, deacetylases for H3K56, Hst3 and Hst4 (paralogs of Sir2), are downregulated and the modification persists [182]. In yeast, it appears that H3K56

39 acetylation drives AsfI-dependent assembly of chromatin after DNA repair [183, 184].

H3K56 acetylation may signal for DNA repair completion, as lack of H3K56 acetylation by histone acetyltranferase Rtt109 mutants leads to persistent activation of checkpoint protein Rad53 [185]. CAF-1 is also involved in acetylated H3K56-driven chromatin restoration [184]. Acetylation of H3K56 promotes the association of histone H3 with

CAF-1 and Rtt109, and promotes assembly of nucleosomes by CAF-1.

In mammals, histone H3K56 acetylation by DNA damage is a matter of debate

[186, 187]. IR, UV, hydroxyurea (HU), and MMS induce histone H3K56 acetylation in human cells [186]. HU is a ribonuclease reductase inhibitor (inhibits the formation of deoxyribonucleotides thereby inhibiting DNA synthesis) which activates MMR.

Acetylated H3K56 colocalizes with γH2AX immediately following IR. However,

Tjeertes et al. observed that H3K56 acetylation is rapidly and reversibly reduced in response to DNA damage [187]. Future work will clarify the importance of H3K56 acetylation in DNA repair in mammalian cells.

Phosphorylation. Amongst the known histone modifications, H2A phosphorylation is one of the most well known. Phosphorylation of H2A in mammals occurs on the histone variant H2AX. Phosphorylation of H2AX in response to DDR (DNA damage response) is observed by induction of DSBs by ionizing radiation [188]. Histone H2AX is a histone variant that occurs ~10% of the time in the H2A population of mammals. Histone H2AX contains a conserved SQE motif and is phosphorylated on serine 139 in mammals. In yeast, serine 129 is phosphorylated in H2A. The serine phosphorylation in mammals and yeast is commonly referred to as γH2AX [188]. Histone H2AX is phosphorylated by

40

ATM and DNA-protein kinases and the phosphorylation occurs in a 1 to 2 megabase range surrounding the DSB region in mammals [189]. In yeast, phosphorylated H2A is detected up to 50 kilobases on either side of the DSB break [190]. ATM/ATR phosphorylates H2AX in response to replication stress and UV irradiation [191].

Phosphorylation of H2AX serves as a platform to stabilize the association of proteins involved in DNA damage and DNA repair.

Ubiquitination. Ubiquitination occurs in conjugation with ubiquitin or ubiquitin-like moieties to lysine residues [192]. All four histones are target for ubiquitination, although its precise function remains unknown. H2A ubiquitination is generally associated with gene silencing whereas H2B ubiquitination is related to both gene activation and silencing [192]. Ubiquitination of H3 and H4 is less abundant. UV irradiation increases

H2A ubiquitination but also induces H3 and H4 ubiquitination transiently [193].

Moreover, H2AX is also ubiquitinated in response to DNA damage [194]. H3 and H4 ubiquitination occurs early in DDR in contrast to H2A ubiquitination.

Methylation. Histones can be mono-, di-, or trimethylated and can be a marker of transcriptionally active or inactive chromatin. This depends on the type of methylation and residue involved. Methylation of lysines H3K4, H3K36, and H3K79 are associated with transcriptional activation whereas methylation of H3K9, H3K27, and H4K20 are connected with transcriptional repression [192]. UV irradiation causes histone methylation on H3 (H3K79) and H4 (H4K20) [195, 196]. Histone methylation of H3K79 and H4K20 occurs during checkpoint signaling after DSB induction.

41

V.e. Chromatin Remodelers

Chromatin remodelers are large multi-subunit complexes that uses ATP as energy to alter the position and composition of nucleosomes [197]. Remodelers change the state of chromatin by moving, ejecting, or restructuring the nucleosome. Chromatin remodelers work with other chromatin factors to control packaging and unpackaging of

DNA needed for various processes (i.e. gene transcription, DNA replication, DNA repair, and DNA recombination) [197]. Most ATP-dependent chromatin remodelers are conserved from yeast to humans. Chromatin remodelers utilize ATP hydrolysis to alter histone-DNA contacts and share similar ATPase domains with the superfamily II helicase-related proteins [198]. The superfamily II helicase family has an ATPase domain containing DExx and HELICc regions separated by a linker. The ATPase domain has 3’5’ translocase activity on naked DNA generating a torsional strain in the presence of DNA and nucleosomes [199-201]. There are four known families of chromatin remodelers (SWI/SNF, ISWI, CHD, and INO80) grouped by their homology of ATPase subunits. Five properties shared by these chromatin remodelers are: i) affinity for nucleosomes over DNA; ii) domains that recognize histone modifications; iii) similar

ATPase domains required for remodeling; iv) domains that regulate the ATPase domain; and v) domains that interact with other proteins (chromatin or transcription factors) [197].

SWI/SNF family. The SWI/SNF genes were originally discovered by genetic studies in S. cerevisiae, by Stern and Neigeborn, as activities required for expression of the HO endonuclease gene required for mating type switching (initiates mating type switch

(SWI)) and SUC1 (encodes invertase required for sucrose fermentation; for sucrose-

42 nonfermenting) gene [202, 203]. In S. cerevisiae, Drosophila, and humans, there are two versions of SWI/SNF remodelers (SWI/SNF and RSC). Both are involved in transcriptional activation and DNA repair. RSC (remodels the structure of chromatin) is ten times more abundant in the cell (yeast whole-cell extract) than SWI/SNF and is essential for growth whereas SWI/SNF is not [204]. The SWI/SNF families of chromatin remodelers are composed of 8 to 14 subunits whereas RSC consists of 15 to 17 subunits

[202, 203]. SWI/SNF and RSC have distinct roles that do not overlap. The catalytic subunit of SWI/SNF is the Swi2 or the Snf2 protein. For RSC, it is the Sth1 subunit

[205]. SWI/SNF complex regulates gene expression whereas RSC is linked to a variety of roles including DSB repair, kinetochore function, and sister chromatid cohesion [206].

In yeast, RSC catalyzes the position and eviction of nucleosomes in vitro [204].

RSC translocates along the dsDNA in an ATP-dependent manner to catalyze eviction of histone octamers [207]. RSC may play a role in DSB repair [208]. In yeast, RSC promotes homologous recombination at the stage of strand invasion [206]. RSC is also implicated in NHEJ [208]. RSC physically interacts with core NHEJ components, Ku and Mre11 [208]. Their interaction is crucial for cell survival after DNA damage by

DSB-causing agents [208, 200].

SWI/SNF uses the energy of ATP hydrolysis to slide and eject nucleosomes

[197]. SWI/SNF changes the translational position of nucleosomes [209]. Based on single molecule experiments, the translocase domain binds to a location on the nucleosomal-DNA situated two turns away from the nucleosomal-dyad [209]. ATP hydrolysis carries a directional (3’5’) DNA translocation destroying histone-DNA

43 contacts [210, 211]. This creates a transient DNA loop that propagates around the nucleosome resulting in nucleosome repositioning [212]. SWI/SNF has been suggested to play a role in DSB repair [213]. In yeast, SWI/SNF is involved in the early steps of homologous recombination and is involved in NER [143, 214]. As RSC occurs at the step of gene conversion [215], this suggests that RSC and SWI/SNF may cooperate in distinct steps of homologous recombination.

ISWI. The ISWI (imitation switch) family consists of 2 to 4 subunits. Many ISWI family complexes encourage nucleosome spacing to promote chromatin assembly and transcription repression [197]. In S. cerevisiae, there are two ISWI genes, ISW1 and

ISW2 [206]. Other members include the Drosophila NURF (nucleosome remodeling factor), CHRAC (chromatin accessibility factor), and ACF1 (ATP-utilizing chromatin assembly and remodeling factor). In budding yeast, ISW1 forms two distinct complexes,

ISW1a and ISWIb [216]. ISWIa has strong nucleosome spacing activity whereas ISW1b does not [216]. ISW2 also has nucleosome activity but is not as tightly regulated as

ISW1a. Furthermore, ISW2 has no detectable nucleosome disruption activity [217].

ISWI changes the translational position of the nucleosome [209]. ISWI moves the entire nucleosome far enough to place the nucleosome localization site into the extranucleosomal DNA region [206]. Moreover, ISWI has a strong directional preference for nucleosome mobilization.

ACF1 is a 2 subunit complex comprised of the ISWI ATPase domain and ACF1.

The ACF complex can assemble periodically spaced nucleosomes in the presence of histone chaperones such as NAP1 or CAF-1, in an ATP-dependent manner [218].

44

Human ACF1 recruits SNF2h (ISWI homologue) to replication foci, as indicated by co- immunoprecipitation and immunofluorescence co-localization [219]. Depletion of human ACF1 increased heterochromatinization of newly replicated DNA as seen from increased levels of HP1 and heterochromatin-specific histone modifications [220].

CHD. The CHD (chromodomain, helicase, DNA binding) family remodelers have 1 to

10 subunits. Some CHD remodelers slide or eject nucleosomes to promote transcription.

Others have repressive roles in transcription. The CHD family is highly conserved from yeast to humans but the function of these proteins remains unknown. The chromodomain and helicase domains of CHD are required for association with nucleosomes [221]. The role of CHD includes chromatin assembly, disruption, and positioning. Biochemical analysis of S. cerevisiae Chd1 reveal ATPase activity that effects DNA-histone interactions within the nucleosome [222]. Yeast Chd1 preferentially relocates the nucleosome to the center of DNA fragments [223].

INO80. The INO80 (inositol-requiring 80) family contains more than 10 subunits, including SWR1-related complexes [224]. INO80 mobilizes nucleosomes in an ATP- dependent manner. INO80 has diverse functions including transcriptional activation and

DNA repair [224]. SWR1 is highly related to INO80. However, SWR1 has the unique ability to restructure the nucleosome by removing the canonical H2A-H2B dimers and replacing them with H2A.Z-H2B dimers [225]. INO80 is involved in transcriptional activation and DNA repair [226, 227]. INO80 also exhibits ATP dependent 3’5’ helicase activity [206]. Yeast strains that lack INO80 have misregulated transcription

45 and are hypersensitive to DNA damaging agents (i.e. MMS) suggesting that INO80 regulates transcription as well as facilitates DNA repair (NHEJ and HR) [226, 227].

Genetic studies suggest that INO80 functions in both HR and NHEJ [226, 227].

INO80 may function in error-prone NHEJ by stimulating the binding of Ku80 and

Mre11, two core proteins involved in NHEJ [228]. INO80 is also involved in HR [228].

One of the earliest steps in the DNA damage response to DSB is the rapid phosphorylation of H2AX adjacent to the DNA break site [188]. Recent studies have shown a strong interaction between INO80 and γH2AX [227]. This interaction is stable under harsh conditions (i.e. sonication) and provides a mechanism for INO80 recruitment to DSBs [227]. The composition of INO80 suggests additional roles in homologous recombination. As INO80 possesses helicase activity, this could disrupt nucleosomes proximal to the break. INO80 mutants are hypersensitive to DNA damaging agents (i.e.

MMS) [229]. Studies also show that transcriptional and checkpoint responses to DNA damage were normal in INO80 mutants by comparing specific gene and global expression patterns [229]. This suggests that INO80 participates in the DNA damage response independent of transcription [229]. Furthermore, it was recently shown that histone eviction near DSBs is mediated by INO80 remodeling activity dependent on the

MRX complex, a DNA damage sensor [230].

SWR1 has been implicated in error-free NHEJ by its recruitment of Ku80 to the

DSB site [228]. SWR1 deposits H2AZ in exchange for γH2AX in a rapid and transient fashion near the break site [231]. This may alter the local chromatin structure and

46 facilitate DNA repair. Also, SWR1 mutants are hypersensitive to DNA damage-induced agents (MMS and hydroxyurea) implicating SWR1 in DDR [225, 232].

V.f. Chromatin Remodeling

A major question in the chromatin remodeling field is how chromatin remodelers engage and manipulate the nucleosome to deposit, remove, and slide nucleosomes along the DNA.

Twist diffusion model. An early idea for how the histone octamer is shifted on DNA was known as the twist diffusion model [233-235]. This model suggests that a single bp can be transferred between linker DNA and the DNA wrapped around the histone core [234].

This implies that segments of nucleosomal-DNA twist or untwist to accommodate the loss or gain of one or more bps [166]. The advantage of this model is that it allows movement of the nucleosome without causing any large changes in the core nucleosomal region. A loss or a gain would generate a twist defect; only a small twist could be tolerated without disrupting the histone-DNA contacts [233]. As the DNA is helical in nature, shifting the DNA past the histone core by 1 bp would result in a ~35˚ rotation of the DNA with each bp [166]. Studies have suggested that nucleosome movement occurs in increments larger than one bp [206]; chromatin remodelers that slide nucleosomes in increments of 10 bp maintain the rotational phasing of nucleosomal-DNA [211]. This indicates that DNA does not shift in 1 bp increments around the histone core.

Loop/bulge propagation model. A second model is the loop/bulge propagation model

[162, 236, 237]. It is similar to the twist diffusion model because the loop/bulge propagation model suggests that DNA from one linker transiently shifts onto the

47 nucleosome, creating a loop or a bulge of DNA that is rapidly diffused around the histone core emerging on the other side [162, 236, 237]. An important difference between the twist diffusion and loop/bulge propagation model is that the DNA being pushed onto the nucleosome is in larger segments, thus disrupting one or more contacts within the histone core. The loop/bulge model is more energetically expensive than the twist diffusion model as more histone-DNA contacts have to be broken [238]. Within the loop/bulge propagation model, there are two different variations.

The first variation of the loop/bulge propagation model is that the ATPase motor utilizes its DNA translocase ability to directly pull DNA in from the nearest entry/exit

(where the DNA enters and exits the nucleosome) site in a continuous fashion. By pumping DNA directly toward the nucleosomal-dyad, the remodeler can create a loop/bulge that diffuses to the distal entry/exit site [239]. In this model, the remodeler maintains its position (at the SHL2 (superhelical turn 2) region – two helical turns from the dyad axis) and attempts to translocate away from the dyad [166]. Continuous translocation on DNA would also explain the processivity of the remodeler as the

ATPase motor would not be dissociated from the nucleosome [239]. However, continuous translocation along the DNA would alter the twist and supercoil the DNA; this is in conflict with experimental evidence showing that DNA does not appreciably twist during remodeling [211, 219, 240]. A more recent version of this model suggests that each single bp step is accompanied by an additional 10 bp step made by a combination of remodeler domains. This would reduce the rotation of the DNA duplex but still pump the DNA toward the dyad [239].

48

The second variation of the loop/bulge propagation model suggests that the

ATPase motor remains fixed on the DNA (at the SHL2 region) as the loop is formed

[211] and shifts the loop towards the dyad. The energy stored in the loop would be sufficient to promote loop propagation around the histone octamer [166, 211]. Recent evidence suggesting that the strongest histone-DNA contacts are at the dyad region challenged the idea that loop propagation started at the entry/exit region where the loop would have to travel a long distance to cross the dyad [166]. This question was resolved by the finding that ATPase motors engage at an internal site on the nucleosome (the

SHL2 region) close to the dyad, thus propagating the loop [166, 211].

Nucleosome sliding. The third model proposes nucleosome sliding without wave-like propagations along the DNA [166]. This suggests that the strong histone-DNA contacts around the dyad would require an alternative strategy for nucleosome remodeling [166].

This strategy suggests that chromatin remodelers engaged within the SHL2 region of the nucleosomal-DNA would alter histone-DNA contacts at the dyad region [239]. Instead of creating a loop/bulge, the entire segment of DNA wrapped around the histone core would shift in concert [166]. This is supported by electrostatic interactions between charged residues and the DNA binding surface. If the charged residues were neutralized

(i.e. acetylated), the disruption of the histone-DNA contacts would allow for sliding. An example of this is the SIN mutations (SWI/SNF-independent) localized to where histones

H3 and H4 contact the DNA [241]. In nucleosomes, the SIN mutant histone octamer shifts more easier than wildtype [241].

49

Breathing of Nucleosomes. Several lines of evidence suggest that nucleosomes are dynamic, periodically releasing stretches of nucleosomal-DNA from their edges [242].

Such partial unwrapping or “breathing” is transient and reversible. Proteins may access nucleosomal-DNA by capturing partially unwrapped nucleosomal-DNA during rounds of unwrapping [242]. Nucleosomes reconstituted with 5S rDNA NPS have a progressive decrease in restriction site accessibility, as a function of distance from nucleosomal-DNA edge [242]. This was consistent with site exposure of progressive unwrapping and wrapping [242]. The site exposure was dependent on the DNA sequence as a subsequent study with nucleosomes positioned on 601 (~100-fold stronger NPS compared to 5S rDNA) were characterized by a 10-100 fold reduction of site exposure compared to nucleosomes reconstituted on 5S rDNA [243]. Fluorescence resonance energy transfer

(FRET) was also used to study nucleosomal-DNA unwrapping rate [244]. Nucleosomes were observed to remain completely wrapped for 250 msec before unwrapping for 10-15 msec and rewrapping again [244]. FRET studies indicate that the unwrapping rate may be fast enough on a biological time scale for proteins to gain access.

Partial unwrapping on nucleosomal-DNA may constitute an intrinsic mechanism of nucleosomal-DNA site exposure. DNase I footprint studies in vitro observed that GR

(glucocorticoid receptor) could bind to GRE (glucocorticoid receptor response elements) elements packed into nucleosomes [245]. Thus, intrinsic nucleosome dynamics may facilitate access of passive or actively-driven proteins to nucleosomal-DNA resulting in chromatin changes.

50

V.g. Chromatin and Mismatch Repair

During DNA replication, nucleosomes undergo disassembly/assembly. The nucleosomes ahead of the replication fork are disrupted and then rapidly reassembled on nascent DNA behind the replication fork [246]. There is increasing evidence that MMR is coupled with DNA replication and that active MMR components must access long distances of DNA (i.e. ~1000 bp) during the MMR reaction [17, 81]. The human MMR system has been reconstituted in vitro using naked mismatched DNA, but the effect of chromatin on human MMR is relatively unknown [58, 62]. Several reports have suggested the involvement of chromatin structure during MMR [247]. In yeast, MMR efficiency varies in the different regions of the genome and suggests that this could be due to chromosome structure [247].

Recent data suggest that nucleosomes may inhibit MMR in vitro [248]. Human hMSH2-hMSH6 was shown to have: i) reduced ATPase and ADP binding activities when interacting with the nucleosomal heteroduplexes; ii) hMSH2-hMSH6 bound to the mismatch within a nucleosome with lower affinity than naked heteroduplex; and iii) nucleosomes flanking a mismatch prevented hMSH2-hMSH6 from sliding along the

DNA helix [248]. The study suggests that reduced binding of hMSH2-hMSH6 to the mismatch within the nucleosome may lead to reduced MMR activity. Moreover, nucleosomes flanking a mismatch may completely or partially block MMR [248].

However, these experiments were conducted using unmodified histone octamers positioned on a high affinity NPS (601).

51

One dimensional diffusion of yeast MSH2-MSH6 (searching for a mismatch) on a chromatin lattice (λ DNA reconstituted with unmodified histones) suggests that nucleosomes act as semipenetrable barriers and MSH2-MSH6 becomes trapped between nucleosomes [249]. On high density nucleosome arrays, most of the molecules of

MSH2-MSH6 were found to be immobile or oscillated between tightly confined regions

[249]. This suggests that MSH2-MSH6 experiences a barrier upon encountering nucleosomes. Therefore, MSH2-MSH6 must disengage the DNA to continue searching for targets or the DNA must be cleared of nucleosomes to allow unhindered access to the

DNA. However, these experiments used λ DNA that did not contain a mismatch [249].

Therefore, MMR complexes may possess chromatin-remodeling activities on natural chromatin which contains mismatches. However, at the moment, it remains unclear whether any specific chromatin remodeling events, including histone modifications, or chromatin remodelers, are required for MMR.

52

Chapter 2

Thesis Rationale

Mutations of the human DNA MMR genes, including hMSH2, hMSH6, and hMLH1, are associated with LS/HNPCC [250]. The loss of MMR in a cell gives rise to a mutator phenotype [251, 252] that is associated with carcinogenesis through accumulation of secondary mutations. The DNA MMR pathway is primarily recognized as a system responsible for post-replicative polymerase error correction. A detailed understanding of MMR depends on identifying the components involved in the process and defining the activity of each component in the process.

In the last few years, MMR proteins involved in MMR have been purified and the

MMR reaction has been reconstituted in vitro using naked mismatched DNA substrates

[59, 62]. However, mismatches and lesions that are recognized by hMSH2-hMSH6 arise in vivo within chromosomes. The effect of chromatin structure on human MMR is unknown. Moreover, no chromatin remodeling activities have been linked to MMR and little is known about the molecular roles PTMs (involved in DNA replication and/or repair) play in MMR. The functions of hMSH2-hMSH6 are: i) recognition and binding of a mismatch; ii) mismatch provoked ADPATP exchange; iii) ATP conformational transitions resulting in a hydrolysis-independent sliding clamp; and iv) hydrolysis of ATP upon dissociation from the DNA. However, the effect of nucleosomes on hMSH2-

53 hMSH6 function(s) is a significant unknown. Moreover, the ability of nucleosomes to block diffusion of hMSH2-hMSH6 sliding clamps and potentially impede MMR is unknown.

To determine the effect nucleosomes have on hMSH2-hMSH6 functions(s), we developed a simple mismatch DNA substrate containing a single well-defined nucleosome. These studies will provide the first physiological relevant characterization of how hMSH2-hMSH6 recognizes and interacts with mismatches in chromatin.

54

Chapter 3

Nucleosome Remodeling by hMSH2-hMSH6

I. Summary

DNA nucleotide mismatches and lesion arise on chromosomes that are a complex assortment of protein and DNA (chromatin). The fundamental unit of chromatin is a nucleosome that contains ~147 bp DNA wrapped around an H2A, H2B, H3, and H4 histone octamer. We demonstrate that the mismatch recognition heterodimer hMSH2- hMSH6 disassembles a nucleosome. Disassembly requires a mismatch that provokes the formation of hMSH2-hMSH6 hydrolysis-independent sliding clamps, which translocate along the DNA to the nucleosome. The rate of disassembly is enhanced by actual or mimicked acetylation of histone H3 within the nucleosome entry-exit and dyad axis that occurs during replication and repair in vivo and reduces DNA-octamer affinity in vitro.

Our results support a passive mechanism for chromatin remodeling where hMSH2- hMSH6 sliding clamps trap localized fluctuations in nucleosome positioning and/or wrapping that ultimately leads to disassembly, and highlights unanticipated strengths of the Molecular Switch Model for mismatch repair (MMR).

55

II. Introduction

Mismatched nucleotides arise in DNA as a result of polymerase misincorporation errors, recombination between heteroallelic parental chromosomes or as a result of chemical and physical damage [2]. MutS homologues (MSH) and MutL homologues

(MLH/PMS) are highly conserved proteins, and are essential for the MMR excision reaction that removes mismatches/lesions from DNA [132]. Mutation of hMSH2, hMSH6, hMLH1, and hPMS2 are the causes of a common human cancer predisposition syndrome, Lynch Syndrome or hereditary non-polyposis colorectal cancer (LS/HNPCC)

[253]. The hMSH2-hMSH6 heterodimer is required for the initial recognition of mismatches during MMR as well as lesion recognition for specific damage-induced signaling pathway(s) [96, 147]. Although MMR occurs in the context of chromatin in vivo, previous biochemical studies have relied exclusively on naked DNA substrates [58,

62]. The effect of chromatin on MMR is unknown. Moreover, no chromatin remodeling activities have been linked to MMR in spite of numerous cellular and genetic surveys [4,

148].

III. Results

Constructing a defined nucleosome DNA containing a mismatch

To determine the effect of nucleosomes on hMSH2-hMSH6 function(s), we have constructed a model DNA substrate containing the Xenopus 5S rDNA nucleosome localization sequence linked to a lac O sequence, mismatch, and terminal biotin on a 3’- tail (Fig. 4A). A single nucleosome was reconstituted on this DNA substrate by salt

56 dialysis, using purified H2A, H2B, H3, and H4 histones that were refolded into histone octamers as previously described [254, 255]. Nucleosome substrates were formed with three types of histone octamers: those containing no modifications (UN), those containing an acetylation mimic where the H3 lysine-56 is substituted with glutamine [H3(K56Q)], and those containing site-specific acetylation of the histone H3 K115 and K122 residues

[H3(K115Ac,K122Ac)]. H3(K56) is located in the nucleosome entry-exit region while

H3(K115, K122) are located in the nucleosome dyad beneath the wrapped DNA. All three residues appear important for normal replication, transcription and DNA repair

[256-258]. Site-specific acetylation of histone H3(K115, K122) was accomplished by intein-mediated protein ligation that links a recombinant H3 thioester truncated at L109 with a synthetic peptide containing acetylated K115 and K122; this method generates a native peptide bond and H3 protein sequence [259]. The mono-nucleosome DNA substrates were then purified on a 5-30% sucrose gradient (Suppl. Fig. 9A and 9B;

[260]). The nucleosome positions were mapped using an ExoIII protection assay and found to largely occupy the 5S rDNA sequence shielding ~145 bp of DNA, as well as a number of lower frequency positioning sites (Suppl. Fig. 9C). The protection footprint suggests that the nucleosomes are composed of histone octamers and the additional positioning sites appear consistent with the gel migration pattern (Suppl. Fig. 9B). hMSH2-hMSH6 binds to nucleosome DNA containing a mismatch

To determine the effect of nucleosomes on the initiation of MMR, we examined hMSH2-hMSH6 binding to the nucleosome-DNA substrates (Fig. 4B and 4C). We found little difference in hMSH2-hMSH6 mismatch binding between the free-DNA substrate

57

Figure 4

Figure 4. Binding of hMSH2-hMSH6 to nucleosome-DNA. (A) The Nucleosome- DNA substrate contains 17 bp 3’ of the 147 bp 5S rDNA nucleosome localization sequence (red) followed by a 28 bp linker, 24 bp lac O sequence (yellow), and 47 bp containing a mismatch site 20 bp from the 3’-end that contains a terminal biotin (light blue). (B) Representative gel showing specific binding of hMSH2-hMSH6 to the G/T mismatch nucleosome-DNA substrate containing an unmodified nucleosome. Boxes above indicate added reaction components (+), the concentration of hMSH2-hMSH6 (nM), and the inclusion of nucleosome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left. The DNA substrate is colored as shown in (A) with a nucleosome (blue oval); hMSH2-hMSH6 (purple clamp); and streptavidin (green circle). (C) Quantitative analysis of hMSH2- hMSH6 binding to free-DNA containing a G/T mismatch (G/T) or G/C duplex (G/C) without or with biotin-streptavidin (-b*) on the 3’-end; and nucleosome-DNA with an unmodified (UN) or H3(K115Ac, K122Ac) modified (2Ac) nucleosome without or with (-b*). Standard deviations were determined from at least three independent experiments and error bars shown (some within the symbol).

58 and the UN and H3(K115Ac,K122Ac) nucleosome-DNA substrates (Fig. 4B and 4C; KD

(G/T) = 24nM; KD (G/T•b•UN) = 27 nM; KD (G/T•b•K115Ac/K122Ac) = 22 nM), and in the presence of streptavidin that induces a physical block to one end of the DNA substrate (Fig. 4C; KD

(G/T•b-SA•UN) = 26 nM; KD (G/T•b-SA•K115Ac/K122Ac) = 10 nM). The binding of hMSH2-hMSH6 to identical DNA substrates without the mismatch (G/C) was over 50-fold less efficient

(Fig. 4B and 4C; KD (G/C) = 1808 nM; KD (G/C•b•K115Ac/K122Ac) = 1342 nM; KD (G/C•b-

SA•K115Ac/K122Ac) = 1198 nM). Similar mismatch specific binding was observed for the

H3(K56Q) nucleosome-DNA substrate. These results demonstrate that the nucleosome-

DNA substrates containing a mismatch outside of the predominant nucleosome localization region are efficiently recognized by hMSH2-hMSH6.

The addition of ATP to hMSH2-hMSH6 bound to a mismatch provokes the formation of a hydrolysis-independent sliding clamp that may diffuse off an open DNA end [20, 138, 261]. ATP-dependent release of the sliding clamp from the mismatch allows iterative cycles of hMSH2-hMSH6 loading and clamp formation [20, 21, 262].

These iterative cycles can result in multiple ATP-bound hMSH2-hMSH6 clamps that may be trapped on the DNA by blocking the ends with biotin-streptavidin or by using a circular DNA substrate [20, 21, 138, 262].

Nucleosomes are highly stable protein-DNA complexes that are known to sterically occlude DNA binding proteins from their target sites [242, 244, 263]. The ability of a nucleosome to block the diffusion of hMSH2-hMSH6 sliding clamps and potentially impede MMR is a significant unknown. Consistent with previous work, we found that addition of streptavidin to a free-DNA (F) substrate containing a single biotin

59 on the 3’-tail resulted in a mobility shift (Fig. 5A-C, compare lane 1 and 2; Suppl. Fig.

10A and 10B, compare lane 1 and 2; [20]). An additional mismatch-specific shift on this single end blocked DNA was observed with hMSH2-hMSH6 (Fig. 5A-C, lane 3; compared to Suppl. Fig. 10A and 10B, lane 3) that was released from the remaining open-end of the DNA substrate with the addition of ATP (Fig. 5A-C, lane 4). These results are consistent with previous studies that demonstrated a single biotin-streptavidin blocked-end is not sufficient to retain ATP-bound hMSH2-hMSH6 sliding clamps on mismatched DNA [20]. We found that the unmodified, H3(K56Q) modification mimic, or H3(K115Ac,K122Ac) modified nucleosome-DNA substrate (N) with an open 3’-tail behaved similarly to the free-DNA (F) substrate containing a single biotin-streptavidin blocked-end (Fig. 5A-C, compare lanes 2-4 with lanes 5-7; Suppl. Fig. 10A and B, lanes

5-7). In this case, the nucleosome-DNA substrate (Fig. 5A-C, lane 5) was bound specifically with hMSH2-hMSH6 (see*, Fig. 5A-C, lane 6; compare Suppl. Fig. 10A and

10B, lane 6), which was then released upon addition of ATP (Fig. 5A-C, lane 7).

Nucleosome disassembly is catalyzed by hMSH2-hMSH6

To determine whether nucleosomes blocked the sliding of hMSH2-hMSH6 clamps we examined the nucleosome-DNA substrates containing biotin-streptavidin blocked 3’-tails (Fig. 5A-C, lanes 8-15; Suppl. Fig. 10A and 10B, lanes 8-15).

Nucleosome stability may be calculated from data with the nucleosome-DNAs containing a G/C duplex, where hMSH2-hMSH6 displays insignificant binding activity (Suppl. Fig.

10A and 10B, lane 8-15; t1/2 (G/C•UN) = 578 min, t1/2 (G/C•K115Ac/K122Ac) = 347 min; Fig. 5D).

60

Figure 5

Figure 5. Nucleosome disassembly by hMSH2-hMSH6. Representative gels showing the nucleosome disassembly reaction catalyzed by hMSH2-hMSH6 with (A) G/T mismatch nucleosome-DNA containing an unmodified nucleosome, (B) G/T mismatch nucleosome-DNA containing an H3(K56Q) acetylation mimic nucleosome, and (C) G/T mismatch nucleosome-DNA containing an H3(K115Ac, K122Ac) modified nucleosome. Black bars indicate image splicing from a single gel where spliced out lanes were redundant with Fig. 6B. Boxes above indicate added reaction components (+) and the inclusion of free-DNA (F) or nucleosome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left and right of the gel panels. The DNA substrate is colored as shown in Fig. 4A with a nucleosome (blue oval); hMSH2-hMSH6 (purple clamp); and streptavidin (green circle). Asterisks indicate the mobility of nucleosome-DNA substrate with bound hMSH2-hMSH6 and without a biotin-streptavidin bound 3’-tail. Red arrow indicates the gel mobility of the nucleosome disassembly product. Asterisk (*) indicates the position of the nucleosome substrate bound by hMSH2-hMSH6; multiple bands are consistent with multiple nucleosome positions surrounding the 5S rDNA localization site (see Suppl. Fig. 9). (D) Quantitative analysis of the nucleosome disassembly reactions. Data analysis includes representative 61

Continuing Figure 5 gels shown in panel A and B as well as Supplemental Fig. 10. Each data set was fit to a single exponential decay to calculate and t1/2. Key: Unmodified nucleosome substrate containing duplex DNA (G/C) and biotin-streptavidin blocked (b*) 3’-tail (UN Nuc-G/C- b*); Unmodified nucleosome substrate containing a G/T mismatch and biotin- streptavidin blocked 3’-tail (UN Nuc-G/T-b*); H3(K56Q) acetylation mimic nucleosome substrate containing a G/T mismatch and biotin-streptavidin blocked 3’-tail (K56Q Nuc- G/T-b*); H3(K115Ac, K122Ac) nucleosome substrate containing duplex DNA (G/C) and biotin-streptavidin blocked 3’-tail (2Ac Nuc-G/C-b*); H3(K115Ac, K122Ac) nucleosome substrate containing a G/T mismatch and biotin-streptavidin blocked 3’-tail (2Ac Nuc-G/T-b*). Standard deviations were determined from at least three independent experiments and error bars shown (some within the symbol).

62

These results demonstrate that unmodified and H3(K115Ac,K122Ac) modified nucleosomes are stable for 10-20 hrs under our experimental conditions. A similar stability is observed with H3(K56Q) nucleosomes (data not shown). In contrast, we found that incubation of the biotin-streptavidin blocked G/T mismatch nucleosome DNA substrates with hMSH2-hMSH6 and ATP resulted in the eviction of the histone octamer

(Fig. 5A-C, lanes 8-15, red arrow; quantified in Fig. 5D). These results suggest that a nucleosome does not block ATP-bound hMSH2-hMSH6 sliding clamps and that the nucleosome appeared to be disassembled by hMSH2-hMSH6. Moreover, there was a significant difference in the ability of hMSH2-hMSH6 to disassemble unmodified versus the H3(K56Q) mimic or H3(K115Ac,K122Ac) modified nucleosomes (Fig. 5D; t1/2

(G/T•UN) = 117 min, t1/2 (G/T•K56Q) = 53 min, t1/2 (G/T•K115Ac/K122Ac) = 23 min, respectively). Our previous work has demonstrated that H3(K115Ac, K122Ac) increases the rate of thermal repositioning and reduces the DNA-histone binding free-energy compared to unmodified nucleosomes [259]. These observations are consistent with the conclusion that nucleosomes containing H3 acetylation mimics and/or modifications that reduce their intrinsic DNA affinity may be disassembled more efficiently by hMSH2-hMSH6.

ATP binding by hMSH2-hMSH6 is required for nucleosome disassembly

ATP-dependent chromatin remodeling is required for numerous cellular DNA transactions including transcription, replication, and repair [246]. Disassembly of a nucleosome from a localized region on DNA suggests that hMSH2-hMSH6 performs a chromatin remodeling reaction. To explore the mechanism behind this new hMSH2- hMSH6 function, we examined the ATP requirement for chromatin remodeling (Fig. 6;

63

Figure 6

Figure 6. Analysis of the ATP requirement for hMSH2-hMSH6 nucleosome disassembly. Boxes above indicate added reaction components (+), the inclusion of free- DNA (F) or Nucleosome-DNA (N), and the time of incubation (min). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left and right of the gel panels. The DNA substrate is colored as shown in Fig. 4A with a nucleosome (blue oval); hMSH2-hMSH6 (purple clamp); and streptavidin (green circle). Asterisks indicate the mobility of nucleosome-DNA substrate with bound hMSH2- hMSH6 and without a biotin-streptavidin bound 3’-tail. Red arrow indicates the gel mobility of the nucleosome disassembly product. (A) Nucleosome disassembly by hMSH2(K675A)-hMSH5(K1140A). Black bar indicates image splicing from a single gel where spliced lanes contained redundant controls shown in Fig. 5A, 5B, and 6B (lanes 6 and 7). (B) Nucleosome disassembly by hMSH2-hMSH6 in the presence of ATPγS. (C) and (D) Quantitative analysis of (A) plus Suppl. Fig. 12 A-C and (B) plus Suppl. Fig. 12D-F, respectively. Each data set was fit to a single exponential decay to calculate and t1/2. See Fig. 5 for Key. Standard deviations were determined from at least three 64

Continuing Figure 6 independent experiments and error bars shown (some within the symbol). (E) hMSH2- hMSH6 steady-state ATPase activity. hMSH2-hMSH6 ATPase activity was determined in the absence of DNA (no DNA), with free-DNA containing a G/T mismatch (G/T) or G/C duplex (G/C) with one (-b*) or two (*b-X-b*) biotin-streptavidin blocked ends, or with nucleosome-DNA containing an unmodified (UN-Nuc) or H3(K115Ac. K122Ac) modified (2Ac-Nuc) nucleosome and a G/T mismatch (G/T) or G/C duplex (G/C) without or with (-b*) a biotin-streptavidin blocked 3’-end. Standard deviations were determined from at least three independent experiments and error bars shown. A diagram of two ATPase cycles is shown on the right. Cycle A illustrates an ATPase cycle for free-DNA containing a single biotin-streptavidin blocked 3’-end [20]. Cycle B illustrates a hypothetical requirement for disassembly of a nucleosome from nucleosome-DNA containing a biotin-streptavidin blocked 3’-end to complete an ATPase cycle consistent with the data. The dashed blue arrow shows that the two cycles are connected by the product of nucleosome disassembly, which is identical to free-DNA containing a single biotin streptavidin blocked 3’-end that initiates cycle A.

65

Suppl. Fig. 11). The hMSH2(K675A)-hMSH6(K1140A) mutant heterodimer binds mismatched DNA similar to the wild type heterodimer, but is incapable of ATP binding and/or hydrolysis (N.P, S.J. and R.F. in preparation; [71]). We found that in spite of a normal mismatch binding activity, hMSH2(K675A)-hMSH6(K1140A) was incapable of catalyzing the disassembly of unmodified or H3(K115Ac, K122Ac) modified nucleosomes (Fig. 6A and 6C; Suppl. Fig. 11A-C). These results suggest that ATP binding and/or hydrolysis are required for hMSH2-hMSH6 catalyzed chromatin remodeling. Since the hMSH2(K675A)-hMSH6(K1140A) protein was purified by an identical method to the wild type protein, these results also imply that preparation contaminants are unlikely to be responsible for the chromatin remodeling activity.

To examine the role of ATP hydrolysis on hMSH2-hMSH6 chromatin remodeling activity we performed nucleosome disassembly studies with the ATP analog adenosine

5′-[γ-thio]-triphosphate (ATPγS). We determined that the rate of ATPγS hydrolysis (kcat) by hMSH2-hMSH6 in the absence of DNA (0.04 ± 0.02 min-1) or in the presence of mismatched DNA (0.06 ± 0.05 min-1; Suppl. Fig. 12A), and compared it to the well- known rate of ATP hydrolysis in the absence of DNA (1 ± 0.5 min-1) or in the presence of mismatched DNA (22 ± 1.2 min-1; [264]). These results clearly demonstrate that hMSH2-hMSH6 is more than 350-fold less capable of hydrolyzing ATPγS compared to

ATP when a mismatch is present, and that repeated rounds of mismatch-dependent hydrolysis are dramatically suppressed by ATPγS. Perhaps more importantly, ATPγS is the only analog of ATP that appears to bind hMSH2-hMSH6 and provoke the formation

66 of a sliding clamp similar to ATP; although the kinetics of sliding clamp formation appear slower than ATP (Suppl. Fig. 12B; [72]).

Control reactions with free-DNA demonstrated streptavidin binding (Fig. 6B, compare lane 1 and 2), specific mismatch binding by hMSH2-hMSH6 (Fig. 6B, lane 3), and the release of hMSH2-hMSH6 upon addition of ATPγS (Fig. 6B, lane 4). These results are similar to previous studies and are consistent with the conclusion that ATP- binding by hMSH2-hMSH6 results in the formation of a hydrolysis-independent sliding clamp [20, 138, 261]. In addition, the single nucleosome substrate DNA containing a

G/T mismatch (Fig. 6B, lane 5) specifically binds hMSH2-hMSH6 (see *, Fig. 6B, lane

6) that is largely released upon the addition of ATPγS (Fig. 6B, lane 7). We note that for both the free-DNA and the nucleosome-DNA substrates the efficiency of ATPγS-induced release appears reduced compared to ATP. These observations are consistent with kinetic analysis (Suppl. Fig. 12B; [72])and suggest that the nucleosome-DNA substrates provoke hMSH2-hMSH6 to form a sliding clamp in the presence of ATPγS, which although modestly slower appears nearly identical to single biotin-streptavidin blocked- end free DNA (Fig. 6B, compare lanes 1-4 with lane 5-7). The addition of ATPγS to the pre-bound hMSH2-hMSH6 in a chromatin remodeling reaction suggests reduced but significant nucleosome disassembly (Fig. 6B, lanes 8-15; t1/2 (G/T•K115Ac/K122Ac) = 108 min;

Fig. 6D; Suppl. Fig. 11D-F). Contrasting the ~4-fold slower rate for nucleosome disassembly in the presence of ATPγS, to the ~350-fold slower rate of ATPγS hydrolysis compared to ATP (Suppl. Fig. 12C), and assuming that the rate-limiting step(s) of the disassembly reaction remain similar, these observations support the notion that γ-

67 phosphate hydrolysis is unlikely to be a significant contributor to the disassembly process. It is important to note that these studies are complicated by a competitive

ATPγS pre-binding reaction that inactivates hMSH2-hMSH6 mismatch binding and freezes iterative mismatch-dependent loading of sliding clamps, which may ultimately contribute to the reduced rate of ATPγS-induced nucleosome disassembly [20, 21].

Taken as a whole these observations are consistent with the conclusion that ATP binding and not hydrolysis is the most significant contributor to hMSH2-hMSH6 chromatin remodeling, and that iterative ATP binding likely sustains an efficient reaction.

In the absence of DNA, hMSH2-hMSH6 displays an intrinsic low-level ATP hydrolysis (ATPase) activity (Fig. 6E, bar 1) that is stimulated by mismatched DNA (Fig.

6E, lane 2). This mismatch-dependent hMSH2-hMSH6 ATPase activity (Fig. 6E, compare bar 2 with bar 5) may be progressively reduced to the background level in the absence of DNA (red line) when one and then both of the DNA ends are blocked with biotin-streptavidin (Fig. 6A, compare bar 2 with bars 3 and 4 or bar 5 with bars 6 and 7).

These results are consistent with previous studies that have demonstrated the hMSH2- hMSH6 ATPase is accelerated by mismatch provoked ADPATP exchange and hydrolysis only occurs when hMSH2-hMSH6 translocates off a DNA end (Fig. 6E, cycle

A; [20]). We examined the hMSH2-hMSH6 ATPase activity with the unmodified and

H3(K115Ac,K122Ac) modified single nucleosome substrates containing a biotin- streptavidin blocked 3’-tail (Fig. 6E, bar 8-11). Unlike traditional chromatin remodelers that display an increased ATPase activity with nucleosome substrates [206], we found that the ATPase activity of hMSH2-hMSH6 with the biotin-streptavidin blocked G/T

68 mismatch nucleosome substrates was reduced compared to G/T mismatch free-DNA containing a single biotin-streptavidin blocked-end (Fig. 6E; see blue bars, compare bar 3 with bar 8 and 10). As expected, the hMSH2-hMSH6 ATPase activity with the biotin- streptavidin blocked G/T mismatch nucleosome substrates was greater than the corresponding biotin-streptavidin blocked G/C duplex nucleosome substrates (Fig. 6E, compare grey bars with blue bars or bar 8 and 10 with bar 9 and 11). Moreover, we found that the ATPase activity was greater with the H3(K115Ac, K122Ac) modified nucleosome substrate compared to the unmodified nucleosome substrate (Fig. 6E, compare bar 8 with 10). These results mirror the hMSH2-hMSH6 catalyzed chromatin remodeling studies and suggest an intimate connection between ATPase activity, a mismatch, and the ability to disassemble a nucleosome. Taken together with our previous studies [20, 262], we consider it likely that the ATPase activity with nucleosome substrates results from a combination of two ATPase cycles, since the product of nucleosome disassembly is a single end-blocked free-DNA substrate (Fig. 6E, cycle B to cycle A via dashed blue arrow). This would explain the reduced ATPase activity since efficient hydrolysis with nucleosome-free DNA (cycle A) would be delayed until the nucleosome was disassembled (cycle B). Alternatively, nucleosomes might enhance hMSH2-hMSH6 ATPase cycling on the DNA. However, it is hard to reconcile the catalytic enhancement of ATPase cycling by histone modifications like H3(K115Ac,

K122Ac) that are buried in the nucleosome dyad.

69 hMSH2-hMSH6 must translocate along the DNA to disassemble a nucleosome

Chromatin remodeling proteins typically interact directly with nucleosomes [206].

To determine whether hMSH2-hMSH6 catalyzed chromatin remodeling requires a mismatch to load hMSH2-hMSH6 sliding clamps that must translocate along the DNA

(cis) or interacts directly with nucleosomes (trans), we placed a lac O sequence between the mismatch and the nucleosome (Fig. 4A). The addition of Lac I protein to a lac O sequence has been previously shown to provide a high-affinity block to the diffusion of

MSH2-MSH6 sliding clamps [138]. We found that the addition of Lac I to the biotin- streptavidin blocked G/T mismatch nucleosome substrate induces a near complete inhibition of hMSH2-hMSH6 catalyzed nucleosome disassembly (Fig. 7; Suppl. Fig. 13; t1/2 (G/T•UN) = 791 min, t1/2 (G/T•K115Ac/K122Ac) = 198 min). These results strongly suggest that the hMSH2-hMSH6 chromatin remodeling activity requires a mismatch in cis with the nucleosome and that hMSH2-hMSH6 must translocate from the mismatch to the nucleosome for disassembly.

IV. Discussion

Nucleosomes are disassembled in front of and reassembled behind a replication fork [246]. The first fully formed nucleosome may be found approximately 250 bp behind the replication fork with intermediates in the assembly process occurring in the intervening region [265, 266]. Post-replication MMR is likely to be initiated in vivo shortly after a mismatch escapes the replication machinery and has been shown to form excision tracts that encompass 100-1000 bp in vitro [267]. These observations suggest

70

Figure 7

Figure 7. The effect of intervening Lac I on hMSH2-hMSH6 nucleosome disassembly. (A) Lac I blocks hMSH2-hMSH6 nucleosome disassembly. Black bar indicates image splicing from a single gel where spliced lanes contained redundant controls shown in Fig. 5A, 5B, and 6B (lanes 6 and 7). Boxes above indicate added reaction components (+), the inclusion of free-DNA (F) or Nucleosome-DNA (N), and the time of incubation (min). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left and right of the gel panels. The DNA substrate is colored as shown in Fig. 4A with a nucleosome (blue oval), hMSH2-hMSH6 (purple clamp); streptavidin (green circle); and Lac I (orange). Asterisks indicate the mobility of nucleosome-DNA substrate with bound hMSH2-hMSH6 and without a biotin-streptavidin bound 3’-tail. Red arrow indicates the gel mobility of the nucleosome disassembly product. Green arrows are a redundant control with Fig. 5B and indicate gel mobility of the nucleosome-DNA and the disassembly product following 60 min incubation without Lac I. (B) Quantitative analysis of (A) plus Suppl. Fig. 13A-C. Each data set was fit to a single exponential decay to calculate and t1/2. See Fig. 5 for Key. Standard deviations were determined from at least three independent experiments and error bars shown (some within the symbol).

71 that the human MMR machinery may encounter both fully formed nucleosomes as well as nucleosome assembly intermediates.

Here we have demonstrated a new chromatin remodeling function for the MMR initiation heterodimer hMSH2-hMSH6. Chromatin remodeling by hMSH2-hMSH6 requires a cis-mismatch and translocation of the heterodimer along the DNA, ATP binding but not ATP hydrolysis, and it is enhanced by histone post-translational modifications that increase thermal repositioning and/or reduce histone-DNA affinity.

We used the 5S rDNA positioning sequence, which strongly localizes nucleosomes compared to native DNA [268]. These observations suggest that genome wide nucleosome disassembly by hMSH2-hMSH6 may be significantly more efficient.

Moreover, artificially high affinity nucleosome positioning sequences, such as the non- physiological 601 positioning sequence, may mask the hMSH2-hMSH6 nucleosome disassembly process [268].

While we have demonstrated that the H3(K56Q) mimic of the replication- associated acetylation modification H3(K56Ac) clearly enhances nucleosome disassembly by hMSH2-hMSH6, there is growing evidence that bona fide histone acetylations additionally accelerate nucleosome thermal repositioning, which may substantially enhance hMSH2-hMSH6 dependent chromatin remodeling [259].

Moreover, SWI/SNF-independent (SIN) histone mutations that are located in the nucleosome dyad near H3(K115) and H3(K122) appear to increase the rate of nucleosome repositioning following thermal heating [241, 269] and reduce DNA-histone

72 interactions [270, 271]; thus reducing or eliminating the requirement for these chromatin remodeling factors in several DNA transactions [272].

The rate of nucleosome disassembly (t1/2 = 23 min) appears well within the window of MMR in vitro [58, 62], although the rate of MMR may be somewhat reduced in the presence of nucleosomes compared to naked DNA. Our results are consistent with the conclusion that hMSH2-hMSH6 performs two important functions for MMR: 1) it specifically recognizes mismatched nucleotides to initiate repair, and 2) it creates a nucleosome-free and perhaps protein-free environment surrounding the mismatch for the excision reaction. A requirement for translocation and the lack of any detectable interaction(s) with histone components or nucleosomes strongly suggests that hMSH2- hMSH6 chromatin remodeling functions are uniquely linked to its ability to form sliding clamps. A related reaction has been considered for RAD51 polymerization-dependent chromatin remodeling [273]. Because chromosomes throughout phylogeny contain complex mixtures of protein-DNA, our observations might be generalized to suggest that all MutS homologues that form sliding clamps function similarly. Several mechanisms for MMR have been proposed and remain controversial (for review see ref. [132]). The

Molecular Switch Model posits the mismatch-dependent loading of multiple MSH hydrolysis-independent sliding clamps that recruit MLH/PMS proteins, and connect mismatch recognition to an iterative dynamic and redundant strand excision process [20,

21]. Our observations appear to highlight an unanticipated strength of the Molecular

Switch Model by suggesting that the iterative MSH hydrolysis-independent sliding clamps also perform chromatin remodeling.

73

Examining the role of hMSH2-hMSH6 in chromatin remodeling in vivo is complicated by the overlapping requirement for sliding clamps in both MMR and nucleosome disassembly. Thus, dissociating the hMSH2-hMSH6 chromatin remodeling activity from MMR activity has been impracticable. One prediction of our studies is that there may be a synergistic phenotype when partially defective alterations of the MMR machinery and chromatin modifying machinery are combined. While these studies are underway, they are technically challenging and may be subtle as a result of the significant redundancies associated with histone modification enzymes.

The absence of an energetic component associated with the translocation of hMSH2-hMSH6 sliding clamps suggests a unique passive mechanism for chromatin remodeling. We consider two models in which hMSH2-hMSH6 sliding clamps might trap inherent structural fluctuations in nucleosomes leading to disassembly (Fig. 8). One model proposes that the formation of iterative sliding clamps may capture thermally induced position-shifts of the nucleosome away from the mismatch; ultimately “nudging” the nucleosome off the open-end of our model DNA substrates (Fig. 8). Since free DNA ends are rare in vivo, such a Nudging Model would be envisioned to detain nucleosomes away from the mismatch along the DNA. A second model considers thermal fluctuations

(breathing) by the nucleosome DNA [242, 244], which might be irreversibly captured in the open-state by hMSH2-hMSH6 sliding clamps (Fig. 8). In this Unwrapping Model, hMSH2-hMSH6 sliding clamps would iteratively occupy the DNA of a breathing nucleosome, beginning at the entry-exit region, until a critical DNA length is engaged and the nucleosome spontaneously disassembles. Both models do not appear to be

74

Figure 8

Figure 8. Two passive models for chromatin remodeling by hMSH2-hMSH6. Both models use the translocation of hMSH2-hMSH6 hydrolysis-independent sliding clamps to trap thermal fluctuations in the nucleosome structure. See text.

75 mutually exclusive and may occur in concert. Passive chromatin remodeling has been considered for transcription factors where binding sites are occluded by nucleosomes

[242]. However, this would be the first case in which stable translocating DNA clamps, with no DNA sequence specificity, provoke the disassembly of nucleosomes. Regardless of the detailed mechanics, it appears that hMSH2-hMSH6 typifies an entirely new class of passive DNA lesion-dependent chromatin remodeling factors.

76

V. Experimental Procedures

Protein and DNA substrates – hMSH2–hMSH6 and the hMSH2(K675A)- hMSH6(K1140A) were purified as previously described [72]. Lac I protein was a generous gift from Dr. Kathleen Matthews (Rice University). The G/C and G/T oligonucleotides (5’-GCT TAG GAT CAT CGA GGA TCG AGC TCG GTG CAA TTC

AGC GGG-3’with the complementary strand 5’-T CGA CCC GCT GAA TTG CAC

CGA GCT (T/C)GA TCC TCG ATG ATC CTA AGC-3’ containing a 3’-biotin moiety) were synthesized (Midland Certified Research Company), annealed and purified by

HPLC using a Waters Gen-Pak column [72]. The site of the mismatch is indicated in bold. The Xenopus 5S rDNA nucleosome localization sequence containing the lac O sequence (5’-TGG AAT TGT GAG CGG ATA ACA ATT-3’) on the 3’-end was amplified by PCR from a pBluescript (SK-) plasmid containing the Xenopus 5S rDNA sequence using tailed primers [5’- GCC CGG GGG ATC CAC TAG TTC - 3’; 5’- ACC

GCC TGG GCC TGG TAC AAT TGT TAT CCG CTC ACA ATT CCA CTC GAG

CGA -3’]. The PCR product (5S rDNA plus the lac O sequence) was digested with XhoI on the 3’-end and SmaI on the 5’-end. The annealed synthetic oligonucleotide containing a G/C duplex or G/T mismatch was ligated to the PCR product, purified by native PAGE and verified by restriction analysis.

Preparation of site-specific acetylated Histone H3 – Histone H3 acetylated at K115 and K122 was prepared by expressed protein ligation [259]. A peptide containing amino acids 110-135 was synthesized manually on Boc-Ala-PAM resin (Novabiochem) using standard Boc-N protection strategies and HBTU activation protocols. K115 and K122 77 were acetylated prior to HF cleavage from the resin and purified by RP-HPLC. Truncated histone H3 (residues 1-109) was cloned as a fusion protein with the GyrA intein into the pTXB1 vector (New England Biolabs). The H3-intein fusion protein was expressed in E. coli BL21 (DE3) cells and purified from inclusion bodies by ion exchange and gel filtration chromatography. The purified protein was refolded by dialysis into a high-salt buffer. Thiolysis was then initiated by addition of 100 mM MESNA

(mercaptoethanesulfonic acid) and allowed to continue for 24 hours at 4°C. The buffer components were then adjusted to generate protein-ligation buffer I: 50 mM HEPES (pH

7.5), 6 M urea, 1 M NaCl, 1 mM EDTA, 50 mM MESNA and the protein concentrated to

> 1 mg/mL of the thioester and stored at –80°C. Expressed protein ligation was done with ten molar equivalents of the acetylated H3(110-135) peptide to the H3(1-109) thioester in protein ligation buffer II: 50 mM HEPES pH 7.5, 6 M urea, 1 M NaCl, 1 mM

EDTA, 20 mM TCEP, which proceeded overnight at room temperature with gentle agitation. Full-length semisynthetic H3 was then purified by ion exchange chromatography over a TSKgel SP-5PW column (TOSOH Bioscience).

Histone octamer preparation – Recombinant unmodified histones: H2A, H2B, H3 and

H4 were expressed and purified as previously described [254, 255]. The unmodified,

H3(K56Q) and H3(K115Ac,K122Ac) histones were unfolded separately in: 7 M guanidine, 20 mM Tris (pH 7.5) and 10 mM DTT for 1 to 3 hours and then spun to remove aggregates. The four core histones were combined at equal molar ratio with total histone concentration adjusted to 5 mg/ml in 200 ul. The octamer was refolded by double dialysis in: 2 M NaCl, 10 mM Tris-HCl (pH 7.5), 1 mM EDTA and 5 mM BME. The

78 recovered refolded octamer was centrifuged to remove large aggregates and then purified over a Superdex 200 (GE healthcare) column. The purity of each octamer was confirmed by SDS-PAGE and mass spectrometry.

Nucleosome Reconstitution – Nucleosomes were reconstituted with [32P]-labeled nucleosome-DNA substrate (Fig. 4A) and with octamer containing unmodified,

H3(K56Q), or H3(K115Ac,K122Ac) histones by salt double dialysis as previously described [274]. The reconstituted nucleosomes were purified by ultracentrifugation on a

5-30% sucrose gradient. Fractions corresponding to the peak of reconstituted nucleosomes were pooled and concentrated in a Centricon 30 concentrator (Amicon) and washed twice with 0.5 X TE. The nucleosome purity was verified with a 5% native polyacrylamide gel containing 1/3 X TBE.

Binding Studies and ATPase – Reactions were performed in: 25 mM HEPES (pH 7.8),

15% glycerol, 100 mM NaCl, 1 mM DTT, 2 mM MgCl2, containing 20 ng/ L poly dI-dC,

200 g/mL acetylated BSA (Promega), and approximately 5 fmols of [32P]-labeled mono- nucleosome or the 265 bp free-DNA substrate in a final volume of 20 ul. hMSH2- hMSH6 (at the indicated concentration) was preincubated with the nucleosome-DNA on ice for 10 min. Reactions were separated on a 5% native polyacrylamide/5% glycerol in

1/3 X TBE at 4°C for 3 hours. Gels were dried, quantified by phosphorimager

(Molecular Dynamics), and represented as percent substrate shifted. Standard deviation was calculated from at least three separate experiments. The ATPase activity was determined in: 25 mM HEPES (pH 7.8), 100 mM NaCl, 10 mM MgCl2, 1 mM DTT,

0.01 mM EDTA, 15% glycerol, 200 g/mL acetylated BSA (Promega), 500 μM unlabeled

79

ATP and 16.5 nM [γ-32P]-ATP in a final volume of 20 μl. Steady-state reactions were performed using 25 nM hMSH2-hMSH6 and 25 nM free-DNA, nucleosome-DNA or without DNA as indicated. We determined that ATP hydrolysis was linear under these conditions for at least 2 hr. Reactions were incubated at 37°C for 60 min and processed as described previously [72].

VI. Acknowledgements

The authors wish to thank Michael Smerdon and Ravindra Amunugama for 5S rDNA plasmids and constructs; Kathleen Matthews for Lac I protein; Justin North and

Robin Nakkula for help in octamer preparation; Thomas Haver for technical assistance; and Kristine Yoder and Jessica Tyler for helpful discussions. This work was funded by

NIH/NCI grants CA067007 and GM062556 (R.F.); GM083055 (M.G.P. and J.J.O.) and a

Career Award in Basic Biomedical Sciences from the Burroughs Welcome (M.G.P.).

80

VII. Supplemental Figures

Figure 9

Figure 9. Purification of nucleosome-DNA. (A) Representative 5-30% sucrose gradient fractionation analysis. Embedded box shows key to fractionation curves. (B) 5% acrylamide, 1/3 X TBE gel analysis of peak fractions from Panel A. (C) Representative Exonuclease III mapping of reconstituted H3(K115Ac, K122Ac) nucleosomes. Similar digestion patterns were observed with unmodified and H3(K56Q) reconstituted nucleosomes. The major stop sites from both the Cy5 and Cy3 ends indicate that the majority of nucleosomes occupy the 5S rDNA localization site. Minor stop sites indicate numerous additional positions that include the Cy5 arm. A schematic of the nucleosome position is shown below panel with darker blue indicating the major position and lighter blue the region(s) of minor position. The multiple closely spaced banding pattern observed in gel shift analysis is indicative of these major and minor affinity sites within the 5S rDNA substrate [275].

81

Figure 10

Figure 10. hMSH2-hMSH6 catalyzed nucleosome disassembly using nucleosome- DNA containing a G/C duplex. Representative gels showing the nucleosome disassembly reaction catalyzed by hMSH2-hMSH6 with (A) G/C duplex nucleosome- DNA containing an unmodified nucleosome and (B) G/C duplex nucleosome-DNA containing an H3(K115Ac, K122Ac) modified nucleosome. Black bars indicate image splicing from a single gel where spliced out lanes are redundant with Fig. 6B. Boxes above indicate added reaction components (+) and the inclusion of free-DNA (F) or nucleosome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left and right of the gel panels. The DNA substrate is colored as shown in Fig. 4A with a nucleosome (blue oval); hMSH2-hMSH6 (purple clamp); and streptavidin (green circle). Asterisks indicate the mobility of nucleosome- DNA substrate with bound hMSH2-hMSH6 and without a biotin-streptavidin bound 3’- tail. Grey arrow and brackets indicates the putative gel mobility of substrates when bound by hMSH2-hMSH6.

82

Figure 11

Figure 11. Representative experimental and controls for Figure 6. Representative gels showing the nucleosome disassembly reaction. (A) hMSH2(K675A)- hMSH6(K1140A) with G/T mismatch and unmodified nucleosomes, (B) hMSH2(K675A)-hMSH6(K1140A) with G/C duplex and unmodified nucleosomes, (C) hMSH2(K675A)-hMSH6(K1140A) with G/C duplex and H3(K115Ac, K122Ac) modified nucleosomes (D) ATPγS with G/T mismatch and unmodified nucleosomes, (E) ATPγS with G/C duplex and unmodified nucleosomes, and (F) ATPγS with G/C duplex and H3(K115Ac, K122Ac) modified nucleosomes. Black bars indicate image splicing from a single gel where spliced out lanes are redundant with Fig. 6A. Boxes above indicate added reaction components (+) and the inclusion of free-DNA (F) or nucleosome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left and right of the gel panels. The DNA substrate

83

Continuing Figure 11 is colored as shown in Fig. 4A with a nucleosome (blue oval); hMSH2-hMSH6 (purple clamp); and streptavidin (green circle). Asterisks indicate the mobility of nucleosome- DNA substrate with bound hMSH2-hMSH6 and without a biotin-streptavidin bound 3’- tail. Grey arrow and brackets indicates the putative gel mobility of schematic substrates when bound by hMSH2-hMSH6.

84

Figure 12

Figure 12. ATP/ATPγS hydrolysis and the ability to provoke the formation of hMSH2-hMSH6 sliding clamps by ATP analogs. (A) ATPγS hydrolysis. Reactions were performed as in the Experimental Methods with the indicated amounts of ATPγS and 200 nM hMSH2-hMSH6 in a 20 ul reaction at 37oC for 30 min without DNA or in the presence of 600 nM homoduplex or mismatched DNA. Total hydrolysis was maintained below 10% and protein independent hydrolysis subtracted. Concentrations were chosen to fully cover the KD of ATPγS (KD•ATP S = 1 M; N.P., S.J. and R.F., unpublished). Data was converted into the rate of hydrolysis, fit to Michaelis-Menten, and the kcat was determined (see text). (B) ATP analog dependent dissociation from a mismatched DNA. hMSH2-hMSH6 was bound to mismatched DNA attached to a surface plasmon resonance (Biacore) surface via a biotin-stretavidin linkage. The addition of ATP has been shown to result in the formation of a sliding clamp that dissociates from the remaining free-end of the attached DNA [138] . The t1/2 was calculated from the dissociation rate constants (koff): t1/2•ATP = 1 sec; t1/2•ATPγS = 34 sec; t1/2•AMP-PNP = 500 sec; t1/2•AMP-PCP = 5000 sec. (C) Normalized comparison of mismatch-dependent ATP hydrolysis rate with ATP S hydrolysis rate and the rate of

85

Continuing Figure 12

ATP-dependent disassembly of the H3(K115Ac,K122Ac) nucleosome with the rate of ATPγS disassembly of the H3(K115Ac,K122Ac) nucleosome.

86

Figure 13

Figure 13. Representative experimental and controls for Figure 7. Representative gels showing the nucleosome disassembly reaction catalyzed by hMSH2-hMSH6 in the 87

Continuing Figure 13 presence of Lac I with (A) G/T mismatch and unmodified nucleosomes, (B) G/C duplex and unmodified nucleosomes, and (C) G/C duplex and H3(K115Ac, K122Ac) modified nucleosomes. Controls for free-DNA and nucleosome DNA are shown in Fig. 5C; black bar indicates image splicing from a single gel where spliced out lanes are redundant with Fig. 6B. Boxes above indicate added reaction components (+) and the inclusion of nucleosome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left and right of the gel panels (see Fig. 7 Legend and text). The DNA substrate is colored as shown in Fig. 4A with a nucleosome (blue oval); hMSH2-hMSH6 (purple clamp); and streptavidin (green circle); and Lac I (orange). Asterisks indicate the mobility of nucleosome-DNA substrate with bound hMSH2-hMSH6 and without a biotin-streptavidin bound 3’-tail. Grey arrow and brackets indicates the putative gel mobility of substrates when bound by hMSH2- hMSH6.

88

Chapter 4

hMSH2-hMSH6 nucleosome disassembly is dependent on post- translational modifications and nucleosome localization sequences

I. Summary

The influence of chromatin on mismatch repair (MMR) has been a matter of debate in recent years. Chromatin is composed of nucleosomes which contain ~147 bp of

DNA wrapped around a tetramer of H3-H4 flanked on either side by H2A-H2B dimers.

Human MMR has been reconstituted in vitro on naked mismatched DNA using purified human proteins. However, the influence of chromatin on MMR is unknown. We previously demonstrated that the hMSH2-hMSH6 mismatch recognition complex disassembles nucleosomes. This work demonstrates that the rate of nucleosome disassembly by hMSH2-hMSH6 is enhanced by actual or mimicked acetylation of histone H3 that occur during DNA replication and repair. Moreover, high affinity nucleosome localization sequences mask hMSH2-hMSH6 nucleosome disassembly. We observe that hMSH2-hMSH6 can recognize and disassemble nucleosomes containing mismatches within the nucleosome. Disassembly of nucleosomes by hMSH2-hMSH6 is dependent on the position of the mismatch within the nucleosome. These results confirm that hMSH2-hMSH6 can disassemble nucleosomes in eukaryotic cells and that this

89 process may be dependent on post-translational modifications (PTMs) on histones of nucleosomes.

II. Introduction

Mismatched nucleotides are generated by misincorporation during DNA replication as well as exposure to exogenous chemicals and endogenous reactive metabolites [2]. Mismatch repair (MMR) recognizes and corrects mismatched nucleotides and insertion/deletion loops (IDLs) [132]. MMR is conserved from bacteria to humans and the MutS homologues (MSH) and MutL homologues (MLH/PMS) are essential for MMR [132]. In human cells, there are three MutS homologues (hMSH2, hMSH3, and hMSH6) which assemble to form two functional heterodimers and initiate the MMR process (hMSH2-hMSH3 and hMSH2-hMSH6) [17]. Mutations in hMSH2, hMSH6, hMLH1, and hPMS2 genes results in susceptibility to the common cancer predisposition syndrome, Lynch Syndrome or hereditary non-polyposis colorectal cancer

(LS/HNPCC) [253]. The hMSH2-hMSH6 heterodimer recognizes single base mismatches and small IDLs [96]. MMR occurs in three distinct steps in E. coli: i) recognition of the mismatch nucleotides or IDLs by MutS; ii) removal of the mismatch nucleotides by nucleases; and iii) DNA synthesis by DNA polymerase [17]. In comparison with E. coli MMR, eukaryotic MMR is more complicated and involves more proteins and cofactors [22]. In eukaryotes, MMR occurs in chromatin and it is highly likely that chromatin structure influences MMR in human and other eukaryotic cells.

The fundamental unit of chromatin is the nucleosome that consists of ~147 bp of

DNA wrapped twice around the histone octamer containing a central H3-H4 tetramer 90 flanked by two H2A-H2B dimers. Previous studies have demonstrated that MMR is coupled to DNA replication [22, 82, 276, 277]. During DNA replication, nucleosomes undergo disassembly/assembly. The DNA in front of the replication fork is removed and is reassembled on the DNA behind the replication fork [246]. As mismatches can arise during replication, MMR complexes must slide over long distances (~100-1000 bp) which requires disassembly of nucleosomes formed behind the replication fork [72, 77,

278]. Although MMR has been reconstituted in vitro on naked mismatched DNA, surprisingly little is known about MMR in the context of chromatin [4, 148].

Here, we examine nucleosome disassembly by hMSH2-hMSH6 with nucleosomes containing PTMs on histone H3 that may alter chromatin structure, stability, dynamics, positioning, and can play a role in regulating biological processes such as

DNA replication and repair [154, 279]. Acetylations can alter chromatin structure and function directly or may act to recruit other factors to the genome (i.e bromodomain containing proteins) [154]. We focused on histone PTMs located at the nucleosomal- dyad axis (the center of the nucleosome around which there is an overall pseudo two-fold symmetry), LRS (loss of rDNA silencing – located in between the nucleosomal-dyad axis and entry-exit region) and the entry-exit region (where the DNA enters and exits the nucleosome). The histone H3(K115) and H3(K122) residues are located in the nucleosome-dyad beneath the wrapped DNA. The dyad is essential for several interactions including essential protein-protein contacts at the H3/H4 tetramer interface and histone-DNA contacts at the center of the positioned DNA sequence [257, 259]. In yeast, replacing lysine to a glutamine, an acetylation mimic, on H3(K115) and H3(K122)

91 reduces transcriptional silencing at ribosomal DNA and telomeres [257]. Previous studies suggested that DNA repair might be influenced by H3(K115) and/or H3(K122) acetylation [244]. H3(K115Q) increases sensitivity to hydroxyurea, a ribonucleotide reductase inhibitor, whereas H3(K115A), H3(K122A), and H3(K122Q) were shown to be highly sensitive to Zeocin, a DNA DSB mimic [256, 257].

Histone H3(K56) acetylation is conserved from yeast to humans and has been shown to play an essential role in DNA repair and replication, regulation of transcription, and chromatin assembly [182, 183, 186, 257]. Histone H3(K56) is located at the entry- exit region at the first α helix of H3 and makes a water-mediated contact close to the

DNA [280]. This led to the idea that H3(K56) acetylation may modulate binding of the

DNA to the nucleosome [281]. In vitro studies report increased nucleosome mobility when a glutamine was substituted for a lysine for H3(K56), mimicking the acetylation

[246]. Moreover, H3(K56Q) increases sensitivity to microccocal nuclease digestion in yeast [282]. For the scope of this study, we used the H3(K56Q) mimic as the H3(K56) acetylation is difficult to synthesize in vitro.

Histone H4(K77, K79) acetylation modifications, commonly referred to as LRS mutants, are located on the lateral surface of the nuclesome within the DNA-protein interface (Suppl. Fig. 25B). Histone H4(K77Ac, K79Ac) are within close proximity of each other at the opposite end of the crescent shaped H3-H4 heterodimer [159]. The

H4(K77Ac, K79Ac) modifications are involved in telomere silencing in yeast [283]. A recent study suggests a role for H4(K77Ac) in replication and repair [284]. However, very little is known about the biochemistry of H4(K77, K79) acetylations.

92

The goal of the study was to examine the interaction of hMSH2-hMSH6 and a single mismatch with different nucleosome substrates. Previous results have shown that

ATP-bound hMSH2-hMSH6 can disassemble nucleosomes [285]. We report that the rate of disassembly of nucleosomes by ATP-bound hMSH2-hMSH6 is dependent on histone

PTMs (i.e. acetylations) and/or mimics (substitution of lysineglutamine) on histone

H3. We also find that hMSH2-hMSH6 nucleosome disassembly is dependent on the nucleosome positioning sequence (NPS). The rate of nucleosome disassembly by ATP- bound hMSH2-hMSH6 is ~5-20-fold reduced when higher affinity nucleosome localization sequences are used for reconstitution of the histone octamer compared to nucleosomes reconstituted with Xenopus 5S rDNA NPS. Moreover, hMSH2-hMSH6 is able to recognize a mismatch within the nucleosome. The recognition affinity by hMSH2-hMSH6 is increased when the mismatch is located within the entry-exit region of the nucleosome compared to a mismatch located at the LRS or the nucleosomal-dyad axis. Moreover, hMSH2-hMSH6 binds to a mismatch within the nucleosome with lower affinity compared to a mismatch adjacent to the nucleosome or free mismatched DNA.

ATP-bound hMSH2-hMSH6 is able to disassemble nucleosomes which contain a mismatch. The nucleosome disassembly rate by ATP-bound hMSH2-hMSH6 is greatly enhanced when the mismatch is located at the entry-exit region within the nucleosome compared to a mismatch at the LRS or the nucleosomal-dyad axis. These results highlight that hMSH2-hMSH6 can disassemble nucleosomes to promote or modulate

MMR in eukaryotic cells.

93

III. Results

Reconstitution of nucleosome-DNA containing a mismatch adjacent to the nucleosome

Nucleosome substrates were constructed from two different substrates: substrate I was assembled from Xenopus 5S rDNA NPS, linked to a mismatch and a terminal biotin on the 3’-end, and substrate II was constructed from Xenopus 5S rDNA NPS linked to a terminal Cy5 on the 5’-end (Suppl. Fig. 19A). Substrate I and II assembled into single nucleosomes when reconstituted by salt dialysis, using purified H2A, H2B, H3, and H4 histones that were first assembled into histone octamer [254, 255]. Nucleosome substrates I and II were formed with three types of histone octamer: those containing no modifications (UN), those containing an acetylation mimic where the H3(K56),

H3(K115), and/or H3(K122) lysine has been substituted with a glutamine (H3[K56Q],

H3[K115Q], H3[K122Q], and H3[K115Q, K122Q]), and those containing a site-specific acetylation of histone H3(K115) and/or H3(K122) residues (H3[K115Ac], H3[122Ac], and H3[115Ac, K122Ac]). H3(K56) is located at the entry-exit region whereas

H3(K115) and H3(K122) are located at the nucleosomal-dyad axis beneath the wrapped

DNA (Suppl. Fig. 19B). All three residues (H3[K56], H3[K115], and H3[K122]) are important for transcription, replication, and repair [256-258]. Site-specific acetylation of histone H3(K115) and/or H3(K122) was accomplished by intein-mediated protein ligation that links a recombinant H3 thioester with a synthetic peptide containing acetylated K115 and/or K122 [259]. The mononucleosomes were purified on a 5%-30% sucrose gradient and analyzed on 5% native PAGE (Suppl. Fig. 19C) [260].

94

Histone H3 modifications on nucleosomes enhance hMSH2-hMSH6 nucleosome disassembly

To determine whether histone H3 modifications enhance nucleosome disassembly by hMSH2-hMSH6, EMSAs (Electrophoretic Mobility Shift Assays) were used to examine interactions between hMSH2-hMSH6 and nucleosome-DNA mismatch substrates with biotin-streptavidin blocked 3’-tails (substrate I in Suppl. Fig. 19A).

Nucleosome stability, as previously reported [285], was calculated from data with the nucleosome-DNA containing a G/C duplex, where hMSH2-hMSH6 displays insignificant binding (Fig. 14A,B t1/2 (G/C•UN) = 578±7 min, t1/2 (G/C•K115Ac/K122Ac) = 347±29 min) [285].

Furthermore, previous work demonstrated that H3(K56Q) mimic and H3(K115Ac,

K122Ac) modification on histone H3 enhance nucleosome disassembly by hMSH2- hMSH6 compared to unmodified nucleosome-DNA mismatch substrate (Fig. 14A,B t1/2

(G/T•UN) = 117±14 min, t1/2 (G/T•K56Q) = 53±3 min, t1/2 (G/T•K115Ac/K122Ac) = 23±0.4 min) [285].

To determine whether single histone H3 acetylations (H3[K115Ac] and

H3[K122Ac]) or acetylation mimics, (H3[K115Q], H3[K122Q], and H3[K115Q,

K122Q]), enhance nucleosome disassembly by hMSH2-hMSH6 compared to unmodified nucleosome-DNA, we examined nucleosome-DNA substrates containing modifications on histone H3. We observed that there is a significant difference in the ability of hMSH2-hMSH6 to disassemble nucleosomes between unmodified versus acetylation mimic nucleosome-DNA substrates (Fig. 14A,B t1/2 (G/T•K122Q) = 80±6 min, t1/2 (G/T•K115Q) =

79±4 min, t1/2 (G/T•K115Q/K122Q) = 63±6 min, t1/2 (G/T•K56Q) = 53±3 min). Moreover, we

95

Figure 14

Figure 14. hMSH2-hMSH6 nucleosome disassembly is enhanced by acetylation modifications and/or acetylation mimics. (A-B) Quantitative analysis of the nucleosome disassembly reactions. Data analysis included representative gels (data not shown). (A) Datum was fit to a single exponential decay to calculate τ and t1/2. (B) Table of rate disassembly values (t1/2) and ΔΔG values for DNA-histone binding affinity. Key: unmodified nucleosome substrate containing a duplex DNA (G/C) and biotin- streptavidin blocked (b*) 3’-tail (UN Nuc-G/C-b*), unmodified nucleosome substrate containing a G/T mismatch and a biotin-streptavidin blocked 3’-tail (UN Nuc-G/T-b*), H3(K115Ac, K122Ac) nucleosome substrate containing duplex DNA (G/C) and biotin- streptavidin 3’-tail (2Ac Nuc-G/C-b*), H3(K122Q) acetylation mimic nucleosome substrate containing a G/T mismatch and a biotin-streptavidin 3’-tail (K122Q Nuc-G/T- b*), H3(K115Q) acetylation mimic nucleosome substrate containing a G/T mismatch and a biotin-streptavidin 3’-tail (K115Q Nuc-G/T-b*), H3(K115Ac) nucleosome substrate containing a G/T mismatch and a biotin-streptavidin 3’-tail (K115Ac Nuc-G/T-b*), H3(K115Q, K122Q) acetylation mimic nucleosome substrate containing G/T mismatch and a biotin-streptavidin 3’-tail (2KQ Nuc-G/T-b*), H3(K122Ac) nucleosome substrate containing G/T mismatch and a biotin-streptavidin 3’-tail (K122Ac Nuc-G/T-b*), H3(K56Q) acetylation mimic nucleosome substrate containing a G/T mismatch and a biotin-streptavidin 3’-tail (K56Q Nuc-G/T-b*) and H3(K115Ac, K122Ac) nucleosome

96

Continuing Figure 14 substrate containing a G/T mismatch and a biotin-streptavidin 3’-tail (2Ac Nuc-G/T-b*). Standard deviations were determined from at least three independent experiments and error bars are shown (some within the symbol). The values for UN Nuc-G/C-b*, 2Ac Nuc-G/C-b*, UN Nuc-G/T-b*, K56Q Nuc-G/T-b*, and 2Ac Nuc-G/T-b* are from our previous publication [285].

97 observed hMSH2-hMSH6 nucleosome disassembly was enhanced with H3(K115Ac),

H3(K122Ac), and H3(K115Ac, K122Ac) acetylation modifications compared to the

H3(K115Q), H3(K122Q), and H3(K115Q, K122Q) acetylation mimics (Fig. 14A,B t1/2

(G/T•K115Ac) = 70±8 min; t1/2 (G/T•K122Ac) = 53±8 min, t1/2 (G/T•K115Ac/K112Ac) = 23±0.4 min), respectively [285]. These observations demonstrate that nucleosomes containing histone

H3 acetylation modifications or acetylation mimics are disassembled more efficiently by hMSH2-hMSH6 compared to unmodified nucleosome-DNA mismatch substrate.

Moreover, there is a ~2-fold difference in the ability of hMSH2-hMSH6 to disassemble

H3(K115Ac, K122Ac) acetylations compared to the H3(K115Q, K122Q) acetylation mimic nucleosomes-DNA mismatch substrates (Fig. 14A,B; t1/2 (G/T•K115Ac/K122Ac) = 23±0.4 min, t1/2 (G/T•K115Q/K122Q) = 63±6 min, respectively). These observations suggest that lysineglutamine mutations, which are generally considered to mimic actual acetylations, do not entirely capture the entire features of lysine acetylation.

Acetylation modifications or acetylation mimics of Lysine-115 and Lysine-122 on histone H3 reduce the free energy of histone octamer binding to Xenopus 5S rDNA nucleosome localization sequence

We performed nucleosome competitive reconstitutions to examine the effect of

H3(K115) and/or H3(K122) acetylation modification or acetylation mimic on the binding affinity of histone octamer to DNA [259]. Competitive reconstitutions were carried out as previously described with the Xenopus 5S rDNA NPS (substrate II in Suppl. Fig. 19A) in the presence of excess low affinity competitor DNA [259]. Under these conditions, the free DNA and histone octamer establish a dynamic equilibrium that can be determined by gel shift analysis with native PAGE [274]. We compared the binding equilibrium of

98 specific histone octamer modifications to the binding equilibrium of unmodified octamer to generate a ΔΔG. Competitive reconstitutions with octamer containing H3(K115Ac),

H3(K122Ac), and H3(K115Ac, K122Ac) acetylation modifications, or H3(K115Q),

H3(K122Q), and H3(K115Q, K122Q) acetylation mimics, were performed at least three times and calculated as reported previously [259]. We find that H3(K115Ac) and

H3(K115Q) do not reduce the free energy of the histone octamer binding to DNA (Fig.

14B ΔΔG K115Ac = 0.021±0.17 kcal/mol, ΔΔG K115Q = 0.055±0.12 kcal/mol). However,

H3(K122Ac), H3(K115Ac, K122Ac), H3(K122Q), and H3(K115Q, K122Q) do reduce the free energy of histone octamer binding to Xenopus 5S rDNA (Fig. 14B ΔΔG K122Ac =

-1.12±0.19 kcal/mol, ΔΔG K115Ac/K122Ac = -1.44±0.55 kcal/mol, ΔΔG K122Q = -0.98±0.16 kcal/mol, ΔΔG K115Q/K122Q = -0.78±.14 kcal/mol). These results demonstrate that acetylation modifications and acetylation mimics of histone H3(K115) may not contribute to the overall binding affinity of histone octamer to DNA. The major contribution to reducing the free energy of histone octamer to DNA is H3(K122) and the combination of H3(K115/K122). However, in regards to hMSH2-hMSH6 nucleosome disassembly, ΔΔG does not predict hMSH2-hMSH6 nucleosome disassembly kinetics.

Formation of nucleosome-DNA mismatch substrates with high affinity nucleosome localization sequences

Nucleosomes can typically form on any DNA sequence. However, certain DNA sequences have ~100-fold increased affinities for histone octamer compared to the affinity of arbitrary sequence DNA [162]. We reconstituted nucleosome-DNA mismatch substrate with high affinity NPS’s (601 and pMP2) using unmodified (UN) and

H3(K115, K122) acetylation modification histone octamer. H3(K115, K122) acetylation 99 modification has the highest rate of nucleosome disassembly by hMSH2-hMSH6 compared to unmodified nucleosome-DNA substrates when reconstituted with Xenopus

5S rDNA NPS (Fig. 14A,B). Single nucleosomes were reconstituted with nonphysiological 601 (a synthetic sequence that was derived in vitro), and a sequence- altered derivative of the 601 sequence called pMP2, with an adjacent mismatch and a terminal biotin at the 3’-tail [260, 286]. The Xenopus 5S rDNA and 601 NPS’s have the lowest and the highest affinity for nucleosome formation, respectively [162]. pMP2 and

601 have a 100-300-fold higher affinity for nucleosome formation compared to Xenopus

5S rDNA NPS, respectively [260, 286]. Unmodified and H3(K115, K122) acetylation modification histone octamer were reconstituted with NPS of 601 and pMP2, and were purified via a 5%-30% sucrose gradient and analyzed by PAGE (Suppl. Fig. 20). hMSH2-hMSH6 nucleosome disassembly is dependent on the nucleosome localization sequence

We observe that hMSH2-hMSH6 can recognize a mismatch adjacent to a single nucleosome (unmodified or H3[K115Ac, K122Ac])-DNA substrates positioned on 601 and pMP2 consistent with previous published studies (unpublished data) [248, 285].

Upon addition of ATP, hMSH2-hMSH6 disassembles unmodified nucleosomes positioned on Xenopus 5S rDNA whereas with the stronger affinity sequences, 601 and pMP2, we were unable to detect disassembly of unmodified nucleosome-DNA mismatch substrates (Fig. 15A,B t1/2 (5S G/T•UN) = 117±14 min, t1/2 (601 G/T•UN) = 693±47 min, t1/2 (pMP2

G/T•UN) = 533±53 min; Suppl. Fig. 21-22A, compare lanes 4-9) [285]. This may occur if the product of nucleosome disassembly by hMSH2-hMSH6 was below the threshold of detection. Moreover, there is not a significant difference in the ability of hMSH2- 100 hMSH6 to disassemble unmodified versus H3(K115Ac, K122Ac) modified nucleosomes-

DNA positioned on 601 or pMP2 NPS (Fig. 15A,B t1/2 (601 G/T•K115Ac/K122Ac) = 408±21 min, t1/2 (pMP2 G/T•K115Ac/K122Ac) = 289±29 min; Suppl. Fig. 21-22B, compare lanes 4-9). These observations indicate that Xenopus 5S rDNA has the highest rate of disassembly for unmodified nucleosome-DNA mismatch substrates whereas the nonphysiological 601 or pMP2 sequence mask nucleosome disassembly by hMSH2-hMSH6. Moreover, the rate of disassembly of nucleosome-DNA substrates by hMSH2-hMSH6 reconstituted with

601 and pMP2 was slightly enhanced (~1.5-fold) by H3(K115Ac, K122Ac) modified compared to unmodified nucleosomes. However, the rate of hMSH2-hMSH6 nucleosome disassembly was ~10-20-fold reduced compared to H3(K115Ac, K122Ac) modified nucleosome-DNA substrate reconstituted on Xenopus 5S rDNA.

These results clarify a previous study which suggests that ATP-bound hMSH2- hMSH6 cannot disassemble nucleosomes [248]. The study was based on nucleosome-

DNA mismatch substrate with a biotin-streptavidin blocked 3’-tail that was reconstituted with high affinity 601 sequence using unmodified human histone octamer [248]. Their results are consistent with our results suggesting that hMSH2-hMSH6 nucleosome disassembly is dependent on NPS. Moreover, higher affinity NPS’s may mask hMSH2- hMSH6 nucleosome disassembly. These observations indicate that hMSH2-hMSH6 can disassemble nucleosomes. This activity is observed on modified or more natural histone octamer which are reconstituted on physiological relevant sequences (i.e. Xenopus 5S

101

Figure 15

Figure 15. Nucleosome disassembly by hMSH2-hMSH6 is dependent on the nucleosome localization sequence. (A-B) Quantitative analysis of the nucleosome disassembly reactions. Data analysis includes representative gels (Suppl. Fig. 21-22). (A) Datum was fit to a single exponential decay to calculate τ and t1/2. (B) Table of rate disassembly values (t1/2). Key: unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (5S rDNA UN-Nuc-G/T-b*), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/T and biotin- streptavidin blocked (b*) 3’-tail (5S rDNA 2Ac-Nuc-G/T-b*), unmodified nucleosome substrate with 601 nucleosome positioning sequence containing a G/T and biotin- streptavidin blocked (b*) 3’-tail (601 UN-Nuc-G/T-b*), H3(K115Ac, K122Ac) nucleosome substrate with 601 nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (601 2Ac-Nuc-G/T-b*), unmodified nucleosome substrate with pMP2 nucleosome positioning sequence containing a G/T and biotin- streptavidin blocked (b*) 3’-tail (pMP2 UN-Nuc-G/T-b*), H3(K115Ac, K122Ac) nucleosome substrate with pMP2 nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (pMP2 2Ac-Nuc-G/T-b*). Standard deviations were determined from at least three independent experiments and error bars are shown (some within the symbol). The 5S rDNA UN Nuc-G/T-b* and 5S rDNA 2Ac Nuc-G/T- b* values are from our previous publication [285].

102 rDNA). These results are consistent with the conclusion that NPS’s may play an important role in the ability of hMSH2-hMSH6 to disassemble nucleosomes.

Variations in the nucleosome localization sequence at the nucleosomal-dyad axis effect hMSH2-hMSH6 nucleosome disassembly

We have shown that nucleosome localization sequences play a role in nucleosome disassembly by hMSH2-hMSH6. To determine whether variations within the NPS at the nucleosomal-dyad axis impacts nucleosome disassembly by hMSH2-hMSH6, we reciprocally replaced six nucleotides from Xenopus 5S rDNA (5S rDNA) nucleosomal- dyad axis with the 601 nucleosomal-dyad axis (5S rDNA-601 and 601-5S rDNA). The pMP2 nucleosomal-dyad axis was also replaced with the 5S rDNA nucleosomal-dyad axis (pMP2-5S rDNA). The pMP2 and the 601 nucleosomal-dyad axis were not altered as they share the exact six nucleotides. The nucleosome-DNA substrates were reconstituted with these reciprocally replaced nucleosome localization sequences containing an adjacent mismatch and a terminal biotin at the 3’-tail using unmodified or

H3(K115Ac, K122Ac) modified histone octamer. The nucleosomes were purified on a

5%-30% sucrose gradient and examined on a PAGE gel (Suppl. Fig. 23).

We determined whether nucleosome disassembly by hMSH2-hMSH6 is affected by nucleosomal-dyad axis replacements in the nucleosome localization sequence.

Disassembly of unmodified or H3(K115Ac, K122Ac) modified 5S rDNA-601 nucleosome-DNA mismatch substrate by hMSH2-hMSH6 was reduced ~2-fold compared to unmodified or H3(K115Ac, K122Ac) nucleosome-DNA mismatch substrate that were reconstituted using the original 5S rDNA nucleosome localization sequence

(Fig. 16A t1/2 (G/T•UN) = 117±14 min, t1/2 (5S-601 G/T•UN) = 277±38 min, t1/2 (G/T•K115Ac/K122Ac) = 103

23±0.4 min, t1/2 (5S-601 G/T•K115Ac/K122Ac) = 46±1.2 min; Suppl. Fig. 24A lanes 4-9). These results suggest that nucleosome disassembly by hMSH2-hMSH6 is ~2-fold dependent on nucleosome localization sequences.

Disassembly of unmodified or H3(K115Ac, K122Ac) modified pMP2-5S rDNA nucleosome-DNA mismatch substrate by hMSH2-hMSH6 was increased ~2-fold compared with unmodified or H3(K115Ac, K122Ac) modified nucleosome-DNA mismatch substrates reconstituted with the original pMP2 sequence. However, there was almost no difference between the unmodified or H3(K115Ac, K122Ac) modified 601-5S rDNA nucleosome-DNA mismatch substrate by hMSH2-hMSH6 compared to the original 601 sequence (Fig. 16B-C t1/2 (601 G/T•UN) = 693±47 min, t1/2 (601 G/T•K115Ac/K122Ac) =

408±21 min, t1/2 (601-5S G/T•UN) = 630±43 min, t1/2 (601-5S G/T•K115Ac/K122Ac) = 365±35 min, t1/2

(pMP2 G/T•UN) = 533±53 min, t1/2 (pMP2 G/T•K115Ac/K122Ac) = 289±29 min, t1/2 (pMP2-5S G/T•UN) =

330±18 min, t1/2 (pMP2-5S G/T•K115Ac/K122Ac) = 204±8 min; Suppl. Fig. 21-22A,B; Suppl. Fig.

24B,C, compare lanes 4-9). These observations indicate that variations in NPS at the nucleosomal-dyad axis have an impact on nucleosome disassembly by hMSH2-hMSH6.

Nucleosome structure does not inhibit binding of hMSH2-hMSH6 to a mismatch

EMSAs were used to examine the interactions between hMSH2-hMSH6 and a

241-bp DNA NPS substrate containing a mismatch located at the entry-exit (where the

DNA enters and exits the nucleosome), LRS (located between entry-exit region and nucleosomal-dyad axis), or nucleosomal-dyad axis within the Xenopus 5S rDNA NPS

104

Figure 16

Figure 16. Effect of nucleosome localization sequence variation at the nucleosomal- dyad axis on hMSH2-hMSH6 nucleosome disassembly. (A-D) Quantitative analysis of the nucleosome disassembly reactions. Data analysis includes representative gels (Suppl. Fig. 24A-C). (A-C) Datum was fit to a single exponential decay to calculate τ and t1/2. (B) Table of rate disassembly values (t1/2). Key: unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (UN-Nuc-G/T-b*), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (2Ac-Nuc-G/T-b*), unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning 105

Continuing Figure 16 sequence exchanged with 601 nucleosome positioning sequence at the nucleosomal-dyad axis containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (UN-Nuc-G/T-b* 5S- 601), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence exchanged with 601 nucleosome positioning sequence at the nucleosomal-dyad axis containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (2Ac-Nuc-G/T-b* 5S-601), unmodified nucleosome substrate with 601 nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (601 UN-Nuc-G/T-b*), H3(K115Ac, K122Ac) nucleosome substrate with 601 nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (601 2Ac-Nuc-G/T-b*), unmodified nucleosome substrate with 601 nucleosome positioning sequence exchanged with Xenopus 5S rDNA nucleosome positioning sequence at the nucleosomal-dyad axis containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (601 UN-Nuc-G/T-b* 601-5S), H3(K115Ac, K122Ac) nucleosome substrate with 601 nucleosome positioning sequence exchanged with Xenopus 5S rDNA nucleosome positioning sequence at the nucleosomal-dyad axis containing a G/T and biotin- streptavidin blocked (b*) 3’-tail (601 2Ac-Nuc-G/T-b* 601-5S), unmodified nucleosome substrate with pMP2 nucleosome positioning sequence containing a G/T and biotin- streptavidin blocked (b*) 3’-tail (pMP2 UN-Nuc-G/T-b*), H3(K115Ac, K122Ac) nucleosome substrate with pMP2 nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (pMP2 2Ac-Nuc-G/T-b*), unmodified nucleosome substrate with pMP2 nucleosome positioning sequence exchanged with Xenopus 5S rDNA nucleosome positioning sequence at the nucleosomal-dyad axis containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (pMP2 UN-Nuc-G/T-b* pMP2-5S), H3(K115Ac, K122Ac) nucleosome substrate with pMP2 nucleosome positioning sequence exchanged with Xenopus 5S rDNA nucleosome positioning sequence at the nucleosomal-dyad axis containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (pMP2 2Ac-Nuc-G/T-b* 5S-601 ). Standard deviations were determined from at least three independent experiments and error bars are shown (some within the symbol).

106

(Suppl. Fig. 25A). The 241-bp DNA contains a 17 bp 5’-tail followed by 147 NPS with a

75 bp 3’-tail. The 241-bp DNA containing a mismatch in the entry-exit region (referred to as nucleosome-A) was reconstituted using unmodified, H3(K115Ac, K122Ac) modified, or the H3(K56Q) mimic histone octamer. The 241-bp DNA containing a mismatch in the LRS region (referred to as nucleosome-B) was reconstituted using unmodified, H3(K115Ac, K122Ac) modified, or H4(K77Ac, K79Ac) modified histone octamer. A 241-bp DNA containing a mismatch in the nucleosomal-dyad axis was reconstituted (nucleosome-C) using unmodified or H3(K115Ac, K122Ac) modified histone octamer. Nucleosome-Mismatch (A, B, or C) substrates were reconstituted with unmodified, H3(K115Ac, K122Ac) modified, H4(K77Ac, K79Ac) modified, or

H3(K56Q) mimic histone octamer (Suppl. Fig. 25B) depending on nucleosome-

Mismatch (A, B, or C) substrate with a terminal biotin on the 3’-end (Suppl. Fig. 25A).

The nucleosome-Mismatch substrates were purified on a 5%-30% sucrose gradient and analyzed by a PAGE gel (Suppl. Fig. 25C).

To determine the effect of mismatches within the NPS on the initiation of MMR, we examined hMSH2-hMSH6 binding to nucleosome-Mismatch (A, B, or C) substrates.

There was a significance difference between hMSH2-hMSH6 mismatch binding affinity to free mismatch DNA substrate, a mismatch adjacent to a nucleosome, and nucleosome-

Mismatch (A, B, or C) substrates. As shown in Figure 17A, hMSH2-hMSH6 binds to free mismatched DNA containing a biotin-streptavidin blocked 3’-end [285]. hMSH2- hMSH6 binding affinity for nucleosome-Mismatch-A, -B, and –C substrates (regardless of mimic or modified octamer) is reduced ~5-22-fold compared to naked mismatched

107

DNA (Fig. 17B,C KD (G/T) = 5.4±0.7 nM, KD (Nuc(G/T)-A•UN) = 44±7 nM, KD (Nuc(G/T)-

A•K115Ac/K122Ac) = 49±3 nM, KD (Nuc(G/T)-A•K56Q) = 91±37.5 nM, KD (Nuc(G/T)-B•UN) = 37±4.2 nM, KD (Nuc(G/T)-B•K115Ac/K122Ac) = 119±47 nM, KD (Nuc(G/T)-B•H4(K77Ac/K79Ac)) = 51±11.5 nM,

KD (Nuc(G/T)-C•UN) = 56±6.5 nM, KD (Nuc(G/T)-C•K115Ac/K122Ac) = 54±3.5 nM; Suppl. Fig. 26A-

D) [285]. Moreover, the binding affinity of hMSH2-hMSH6 to a nucleosome or free-

DNA that does not contain a mismatch is ~10-60-fold less compared to the hMSH2- hMSH6 binding affinity for a mismatch located within the NPS of a nucleosome (Fig.

17B,C KD (Nuc(G/C)•K115Ac/K122Ac) = 1198±43.5 nM; KD (G/C) = 2261±85 nM) [285]. This indicates that hMSH2-hMSH6 binding to a mismatch is a specific interaction with the mismatch within the nucleosome and not the nucleosome itself. These results demonstrate that a mismatch located within the nucleosome is recognized by hMSH2- hMSH6. However, hMSH2-hMSH6 binding affinity for a mismatch is reduced by the presence of the histone octamer compared to naked mismatched DNA.

The percentage of maximum binding (Bmax) by hMSH2-hMSH6 is approximately equivalent for H3(K115Ac, K122Ac) modified or H3(K56Q) acetylation mimic compared to unmodified nucleosome-Mismatch-A substrate (Fig. 17A-C Bmax (Nuc(G/T)-

A•UN) = 0.47±.05, Bmax (Nuc(G/T)-A•K115Ac/K122Ac) = 0.54±0.03, Bmax (Nuc(G/T)-A•K56Q) =

0.79±.09). hMSH2-hMSH6 maximum binding (%) for a mismatch located in the entry- exit region of a nucleosome is ~1.2-4-fold enhanced compared to a mismatch at the LRS

(nucleosome-Mismatch-B) or the nucleosomal-dyad axis (nucleosome-Mismatch-C) within the nucleosome regardless of modification (Fig. 17B,C Bmax (Nuc(G/T)-B•UN) =

108

Figure 17

Figure 17. Binding of hMSH2-hMSH6 to nucleosome-Mismatch substrates. (A) Representative gel showing specific binding of hMSH2-hMSH6 to unmodified and H3(K115Ac, K122Ac) acetylation modification nucleosome-Mismatch-A substrate containing an entry-exit mismatch. Boxes above indicate added reaction components (+), the concentration of hMSH2-hMSH6 (nM), and the inclusion of nucleosome-Mismatch- A substrate (N). A schematic DNA species with arrows or brackets indicating gel mobility position is shown to the left. The DNA substrate is colored with a nucleosome (blue oval), hMSH2-hMSH6 (purple clamp), and streptavidin (green circle). (B-C) Quantitative Analysis of hMSH2-hMSH6 binding to free-DNA containing a G/T mismatch (G/T) or G/C duplex (G/C) with biotin-streptavidin (-b*) on the 3’-end, and nucleosome-Mismatch substrate (A, B, or C) with an unmodified (UN), H3(K115Ac, K122Ac) (2Ac), H3(K56Q) (H3(K56Q)), H4(K77Ac, K79Ac) (H4(K77Ac, K79Ac)) with (-b*). Key: unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing an entry-exit mismatch (G/T) and biotin-streptavidin 109

Continuing Figure 17 blocked (b*) 3’-tail (A UN Nuc-(G/T)-b*), H3(115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing an entry-exit mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (A 2Ac Nuc-(G/T)-b*), H3(K56Q) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing an entry-exit mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (A H3 K56Q-Nuc-(G/T)-b*), unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a LRS mismatch (G/T) and biotin- streptavidin blocked (b*) 3’-tail (B UN Nuc-(G/T)-b*), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a LRS mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (B 2Ac Nuc-(G/T)-b*), H4(K77Ac, K79Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a LRS mismatch (G/T) and biotin- streptavidin blocked (b*) 3’-tail (B H4(K77Ac, K79Ac) Nuc-(G/T)-b*), unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a DNA-dyad axis mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (C UN Nuc-(G/T)-b*), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a DNA-dyad axis mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (C 2Ac Nuc-(G/T)-b*), DNA substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a mismatch and biotin- streptavidin blocked (-b*) 3’-tail (G/T-b*), DNA substrate with Xenopus 5S rDNA nucleosome positioning sequence with biotin-streptavidin blocked (-b*) 3’-tail (G/C-b*), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence and biotin-streptavidin blocked (b*) 3’-tail (2Ac Nuc-G/C-b*). Standard deviations were determined from at least three independent experiments and error bars are shown (some within the symbol). G/T-b*, G/C-b* and 2Ac Nuc-G/C-b* values were published previously [285].

110

0.28±.03, Bmax (Nuc(G/T)-B•K115Ac/K122Ac) = 0.39±0.07, Bmax (Nuc(G/T)-B•H4(K77Ac/K79Ac)) =

0.20±0.02, Bmax (Nuc(G/T)-C•UN) = 0.27±0.02, Bmax (Nuc(G/T)-C•K115Ac/K122Ac) = 0.31±0.03).

This increased recognition by hMSH2-hMSH6 for a mismatch located in the entry-exit region may occur if the histone octamer is not positioned fully over the NPS. Some of the mismatches located in the entry-exit region of the nucleosome may be exposed (not occluded by the histone octamer) leading to increased mismatch binding affinity by hMSH2-hMSH6.

Nucleosome structure reduces hMSH2-hMSH6 nucleosome disassembly of nucleosomes containing mismatches located within the nucleosome localization sequence

To determine whether hMSH2-hMSH6 can disassemble nucleosomes containing mismatches, we examined nucleosome-Mismatch substrates (A, B, or C). Consistent with previous work, we found that addition of streptavidin to unmodified or H3(K115Ac,

K122Ac) modified nucleosome-Mismatch-A substrate containing a single biotin on the

3’-tail resulted in mobility shift (Fig. 18A; compare lanes 1 and 2) [285]. An additional mismatch specific shift on the unmodified or H3(K115Ac, K122Ac) modified nucleosome-Mismatch-A substrate was observed upon addition of hMSH2-hMSH6 (Fig.

18A; lane 3). In addition, we found that incubation of unmodified or H3(K115Ac,

K122Ac) modified nucleosome-Mismatch-A substrate with hMSH2-hMSH6 and ATP resulted in eviction of the histone octamer (Fig. 18A; compare lanes 4-9). These results suggest that nucleosomes containing a mismatch can be disassembled by ATP-bound hMSH2-hMSH6. Moreover, there is a significant difference in the ability of hMSH2- hMSH6 to disassemble unmodified versus H3(K56Q) mimic or H3(K115Ac, K122Ac)

111 modified nucleosomes (Fig. 18B, C t1/2 (Nuc(G/T)-A•UN) = 315±18 min; t1/2 (Nuc(G/T)-A•K56Q) =

210±12 min; t1/2 (Nuc(G/T)-A•K115Ac/K122Ac) = 69±1 min; Suppl. Fig. 27C). However, the rate of disassembly of nucleosome-Mismatch substrates (B or C) by hMSH2-hMSH6 was ~3-

9-fold reduced when the mismatch was placed either at the LRS or the nucleosomal-dyad axis compared to a mismatch placed at the entry-exit region of nucleosome, regardless of modification (Fig. 18B,C t1/2 (Nuc(G/T)-B•UN) = 990±48 min, t1/2 (Nuc(G/T)-B•K115Ac/K122Ac) =

630±21 min, t1/2 (Nuc(G/T)-B•H4(K77Ac/K79Ac)) = 745±24 min, t1/2 (Nuc(G/T)-C•UN) = 788±127 min, t1/2 (Nuc(G/T)-C•K115Ac/K122Ac) = 204±7 min; Suppl. Fig. 27A-B,D). Moreover, the rate of disassembly of H4(K79Ac, K77Ac) modified histone octamer reconstituted with Xenopus

5S rDNA with an adjacent G/T and a biotin-streptavidin blocked 3’-tail was not enhanced compared to unmodified nucleosomes with an adjacent mismatch (Fig. 18B,C t1/2

(G/T•H4(K77Ac/K79Ac)) = 122±6 min; Suppl. Fig. 27E), indicating that the H4(K77Ac, K79Ac) modified nucleosome is not efficiently disassembled by hMSH2-hMSH6. This might occur if H4(K77Ac, K79Ac) modifications do not increase the rate of thermal repositioning and/or do not reduce the histone-DNA free energy. Entry-exit mismatches within the nucleosome are recognized and nucleosomes are disassembled by hMSH2- hMSH6. Disassembly of a nucleosome containing a entry-exit mismatch is considerably enhanced by H3(K115Ac, K122Ac) modification compared to H3(K56Q) mimic or the unmodified nucleosome-Mismatch-A substrate. A nucleosome containing a mismatch located within the LRS region or the nucleosomal-dyad axis is not efficiently disassembled by hMSH2-hMSH6. However, hMSH2-hMSH6 nucleosome disassembly

112

Figure 18

Figure 18. Disassembly of nucleosome-Mismatch substrates by hSMH2-hMSH6. (A) Representative gel showing the nucleosome disassembly reaction catalyzed by hMSH2-hMSH6 with a G/T mismatch in the entry-exit region of the nucleosome containing unmodified or H3(K115Ac, K122Ac) modified nucleosome. Boxes above indicate added reaction components (+) and inclusion of nucleosome-Mismatch-A substrate. A schematic of DNA species with arrows or brackets indicating gel mobility position is shown to the left. The nucleosome-Mismatch-A substrate is colored with a nucleosome (blue oval), hMSH2-hMSh6 (purple clamp), and streptavidin (green circle). Red arrow indicates the gel mobility of the nucleosome disassembly product. Multiple bands are consistent with multiple nucleosome positions surrounding the 5S rDNA localization site. (B-C) Quantitative analysis of the nucleosome disassembly reactions. Data analysis includes representative gels shown in (A) and Suppl. Fig. 27A-E. Each

113

Continuing Figure 18 data was fit to a single exponential decay to calculate τ and t1/2. Key: unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing an entry-exit mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (A UN Nuc-(G/T)-b*), H3(115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing an entry-exit mismatch (G/T) and biotin- streptavidin blocked (b*) 3’-tail (A 2Ac Nuc-(G/T)-b*), H3(K56Q) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing an entry-exit mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (A H3 K56Q Nuc-(G/T)-b*), unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a LRS mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (B UN Nuc-(G/T)-b*), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a LRS mismatch (G/T) and biotin- streptavidin blocked (b*) 3’-tail (B 2Ac Nuc-(G/T)-b*), H4(K77Ac, K79Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a LRS mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (B H4(K77Ac, K79Ac) Nuc- (G/T)-b*), unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a DNA-dyad axis mismatch (G/T) and biotin- streptavidin blocked (b*) 3’-tail (C UN Nuc-(G/T)-b*), H3(K115Ac, K122Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a DNA-dyad axis mismatch (G/T) and biotin-streptavidin blocked (b*) 3’-tail (C 2Ac Nuc-(G/T)-b*), H4(K77Ac, K79Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence with an adjacent mismatch and biotin- streptavidin blocked (b*) 3’-tail (H4(K77Ac, K79Ac) Nuc-G/T-b*). Standard deviations were determined from at least three independent experiments and error bars are shown (some within the symbol).

114 on nucleosome-Mismatch-A substrate is ~4-5-fold enhanced by the H3(K115Ac,

K122Ac) modification compared to unmodified nucleosome-Mismatch-A and –C substrates. Moreover, nucleosome disassembly by hMSH2-hMSH6 is ~1.6-fold enhanced by H3(K115Ac, K122Ac) modification compared to unmodified nucleosome-

Mismatch-B substrate. This suggests that some of the mismatches in the entry-exit and nucleosomal-dyad axis region may be exposed and not occluded by the histone octamer whereas a mismatch located at the LRS region is mainly occluded by histone octamer.

Exposure of the mismatch site will increase the mismatch binding affinity of hMSH2- hMSH6 and upon addition of ATP, increase the rate of hMSH2-hMSH6 nucleosome disassembly. Additionally, these observations indicate that additional mechanisms may recognize and disassemble nucleosomes containing mismatches located within NPS that are not easily recognized by hMSH2-hMSH6.

IV. Discussion

For MMR to occur, nucleosomes must be moved or disassembled. There is growing evidence that supports an effect of chromatin in MMR [247]. Yeast studies have suggested that local chromatin structure influences MMR efficiency because mutation rate of an identical polyGT tract varied significantly in different locations in the yeast genome [247]. Moreover, recent data has suggested that MMR is coupled to replication

[22]. During replication, nucleosomes have to be disassembled and reassembled behind a replication fork [246]. The first fully formed nucleosome occurs ~250 bp behind the replication fork with nucleosome intermediates in the assembly process occurring behind the replication fork and the first fully formed nucleosome [265, 266]. Post-replicative 115

MMR is likely to occur shortly after a mismatch escapes the replication machinery.

MMR has been shown to form excision tracts that range ~100-1000 bp in vitro [267].

These observations indicate that MMR machinery might encounter chromatin in vivo.

Here, we demonstrate that nucleosome disassembly by hMSH2-hMSH6 is dependent on histone modifications and the nucleosome localization sequence. The data presented here highlights: (i) hMSH2-hMSH6 nucleosome disassembly reconstituted on physiological relevant Xenopus 5S rDNA is enhanced by acetylation modifications and/or acetylation mimics; (ii) chromatin disassembly by hMSH2-hMSH6 is highly dependent on NPS; nucleosomes reconstituted with Xenopus 5S rDNA have the highest rate of disassembly whereas nucleosomes reconstituted with the artificial 601 and pMP2 sequences have the least rate of disassembly by hMSH2-hMSH6; iii) hMSH2-hMSH6 binds to a mismatch within the nucleosome with lower affinity than free mismatched

DNA or a mismatch adjacent to the nucleosome; and iv) hMSH2-hMSH6 can disassemble nucleosomes containing mismatches; however the rate of disassembly by hMSH2-hMSH6 is reduced compared to a mismatch adjacent to a nucleosome.

We demonstrate that H3(K56Q), H3(K115Q), H3(K122Q), and H3(K115Q,

K122Q) acetylation mimic enhance nucleosome disassembly by hMSH2-hMSH6.

Nucleosome disassembly by hMSH2-hMSH6 is also enhanced with H3(K115Ac),

H3(K122Ac), and H3(K115Ac, K122Ac) acetylation modifications. This indicates that nucleosome disassembly by hMSH2-hMSH6 may be enhanced by PTMs which reduce the free energy of histone octamer binding to DNA or increase thermal repositioning.

Acetylations of H3(K115) and H3(K122) have been shown to reduce histone-DNA

116 affinity [259]. In this study, we also observed reduction in histone-DNA affinity with

H3(K115, K122) acetylations. Moreover, there is a significant difference in the rate of nucleosome disassembly by hMSH2-hMSH6 between actual acetylation modifications and acetylation mimics. Previous studies have demonstrated significant differences between acetylation modifications and mimics [259]. LysineGlutamine substitutions mimics the change in charge of a acetylation but is a very poor mimic of steric effects of acetylations [259]. This may occur if steric effects play a more significant role within the histone octamer. These observations indicate that acetylation mimics do not accurately reproduce actual acetylations.

We also demonstrate that nucleosome localization sequences play a significant role in hMSH2-hMSH6 nucleosome disassembly. We observe that hMSH2-hMSH6 disassembly of nucleosomes reconstituted on high affinity artificial NPS’s (i.e. 601 and pMP2) is masked regardless of acetylation modification. These experiments explain a previous study by Li et al. that demonstrated that nucleosome disassembly is inhibited by hMSH2-hMSH6 [248]. Their study used unmodified histone octamer reconstituted on the high affinity 601 NPS [248, 268]. Their data collaborates with our results suggesting that the use of nonphysiological artificial sequences to reconstitute nucleosomes may mask hMSH2-hMSH6 nucleosome disassembly.

hMSH2-hMSH6 is able to recognize mismatches within the nucleosome.

However, the binding affinity of hMSH2-hMSH6 to mismatches within the NPS is reduced ~5-22-fold compared to a mismatch adjacent to a nucleosome or naked mismatched DNA [72, 248, 285]. Moreover, hMSH2-hMSH6 nucleosome disassembly

117 is dependent on mismatch placement within the NPS. An entry-exit mismatch in nucleosome-Mismatch substrates has the highest rate of nucleosome disassembly by hMSH2-hMSH6 compared to a mismatch located at the LRS or the nucleosomal-dyad axis region. Moreover, the rate of nucleosome disassembly by hMSH2-hMSH6 of a mismatch within the NPS is reduced ~2-10-fold compared to a mismatch adjacent to the nucleosome. However, a caveat in the study is that the binding affinity of hMSH2- hMSH6 detected in gel-shift assays may be hMSH2-hMSH6 background binding of free mismatches. Thus, hMSH2-hMSH6 may not recognize and bind to mismatches within the nucleosome; hMSH2-hMSH6 recognizes and binds to mismatches which are not occluded by the histone octamer. If the mismatch is not repaired within the NPS of a nucleosome, the mismatch may persist. These results suggest that additional chromatin remodeling and/or histone modification activities may be required to repair mismatches within the NPS of nucleosomes in addition to the hMSH2-hMSH6 complex.

These findings have significant implications in our understanding of the mechanism of MMR in eukaryotic cells. First, we demonstrate that acetylation modifications or acetylation mimics that may occur during DNA repair and replication, enhance nucleosome disassembly by hMSH2-hMSH6. There is growing evidence suggesting that histone acetylations accelerate nucleosome thermal repositioning and/or reduce free energy of histone octamer to DNA [259] that may enhance hMSH2-hMSH6 nucleosome disassembly. This may reduce or eliminate the requirement for ATP- dependent chromatin remodeling factors [272]. Our results propose that MMR is not inhibited by nucleosomes. Several mechanisms for MMR have been proposed and

118 remain controversial [17, 132]. The molecular switch model proposes that mismatch dependent loading of multiple MSH hydrolysis-independent sliding clamps recruits

MLH/PMS proteins [20, 21]. Moreover, multiple hMSH2-hMSH6 hydrolysis- independent sliding clamps may enhance nucleosome disassembly [285]. The results presented here further demonstrate interactions between MMR and chromatin.

V. Experimental Procedures

Protein and DNA substrates – hMSH2-hMSH6 was purified as previously described

[72]. Substrates containing Xenopus 5S rDNA, 601 (a gift from Dr. Jonathan Widom), and pMP2, a variant of the 601 sequence (a gift from Dr. Michael Poirier), were synthesized and purified. The G/C and G/T oligonucleotides (5’-GCT TAG GAT CAT

CGA GGA TCG AGC TCG GTG CAA TTC AGC GGG-3’ with the complementary strand 5’-TCG ACC CGC TGA ATT GCA CCG AGC T(T/C)G ATC CTC GAT GAT

CCT AAG C-3’ containing a 3’-biotin moiety) were synthesized (Midland Certified

Research Company), annealed, and purified by HPLC using a Waters Gen-Pak column.

The site of the mismatch is indicated in bold. The Xenopus 5S rDNA nucleosome localization sequence was amplified by PCR from a pBluescript (SK-) plasmid containing the Xenopus 5S rDNA sequence using tailed primers with an XhoI digestion site on the primer annealing on the 3’-end and a SmaI digestion site on the 5’-end primer.

The PCR product was digested with XhoI on the 3’-end and SmaI digested on the 5’end.

The annealed synthetic oligonucleotide containing a G/C duplex or a G/T mismatch was ligated to the PCR product, purified by native PAGE, verified by restriction digest, and will be referred to as substrate I. The 601 and the pMP2 nucleosome localization 119 sequence were amplified by PCR from a pUC19 plasmid containing the 601 or the pMP2 sequence using tailed primers with an XhoI digestion site on the primer annealing on the

3’-end and the SmaI digestion site on the 5’-end. The PCR product of 601 or the pMP2 was digested with XhoI on the 3’-end and SmaI digested on the 5’-end. The annealed synthetic oligonucleotide containing the G/C or a G/T mismatch was ligated to the PCR product of 601 or pMP2, purified by native PAGE, and verified by restriction digest. The

DNA substrate used for competitive reconstitutions was a 192 bp sequence of the

Xenopus 5S rDNA nucleosome localization sequence and will be referred to as substrate

II. Substrate II was amplified by PCR from a pBluescript (SK-) plasmid containing the

Xenopus 5S rDNA sequence.

DNA substrates with sequence change at the nucleosomal-dyad axis – The 601 and pMP2 NPS’s have the exact same nucleosomal-DNA axis sequence (GCGCTG). The

Xenopus 5S rDNA NPS at the dyad region is CGTAGG, which is differs from that of 601 and pMP2. A two-primer PCR mutagenesis [287] was used reciprocally to exchange the nucleosomal-dyad axis of 5S rDNA sequence (from pBluescript (SK-)) with the nucleosomal-dyad axis of 601 (from pUC19); the 601 and/or pMP2 DNA dyad region were exchanged with Xenopus 5S rDNA using two-primer PCR mutagenesis [287]. The

Xenopus 5S rDNA with the exchanged sequence from 601 and/or pMP2 will be called

5S-601. The 601 and pMP2 with the exchanged sequence from Xenopus 5S rDNA will be called 601-5S and pMP2-5S, respectively. The exchanged regions were verified by

DNA sequencing. The 5S-601, 601-5S, and pMP2-5S oligonucleotides containing a G/T

120 mismatch adjacent to the NPS were constructed as described in the previous section

[285].

Preparation of DNA substrates with mismatches within the NPS – Xenopus 5S rDNA

NPS was used to construct mismatches within the nucleosome localization sequence.

The G/T mismatch located at the entry-exit, LRS, or nucleosomal-dyad axis within the

NPS were 241-nucleotide PCR derived substrates with a 3’-terminal biotin moiety and were prepared in the following manner [138]. For simplicity sake, the methodology about the G/T mismatch in the nucleosomal-dyad region of the NPS will be provided.

The G/T mismatch in the entry-exit or the LRS region was prepared similarly. The biotinylated “G” strand was prepared by PCR amplification of a 241-bp product, using the reverse 5’-biotinylated oligonucleotide (5’-biotin- GGG GCC TAG GTG ATC AAG

ATC) and the forward 5’-phosphorylated oligonucleotide (5’-phos-CGG GCC CCC TAG

GTG ATC AAG) with plasmid SJ851 as template DNA. The “T” strand for the G/T mismatch substrate was amplified with plasmid template SJ852 with the oligonucleotide having the same sequence, except the reverse primer was 5’-phosphorylated and forward primer was not phosphorylated. The plasmid SJ851 has the same sequence as SJ852 except for a single nucleotide change that places an AT base pair instead of a GC base pair 92 bases from the 5’-end of the forward primer. The two PCR products (GbCp and

ApT) were digested with λ exonuclease (New England Biolabs), which is specific for the

5’-phosphorylated strand of the double-stranded DNA; thus digestion of these PCR products with λ exonuclease produces a biotinylated G top strand, as well as the T bottom strand. The G and T strands were mixed together and annealed by heating to 95 ˚C and

121 cooling slowly to create a G/T mismatch. The double-stranded G/T mismatch DNA was then purified using the Qiagen PCR purification kit (Qiagen) and analyzed on a PAGE gel.

Preparation of site-specific acetylations and/or mimics on histone H3 and H4 – Site specific acetylations and or mimics of histone H3 (H3[K115] and/or H3[K122]) and

H3[K56Q]) mimic were constructed as previously described [259, 285]. The K77 and

K79 acetylation on histone H4 were generated by employing a native chemical ligation to generate full length histone H4 with K77 and K79 acetylation very similar to the

H3(K115) and H3(K122) acetylation (manuscript in preparation). Site specific acetylation on H4(K77Ac, K79Ac) were generated by expressed protein ligation. The alanine on H4-A76 was mutated to a cysteine (H4-A76C). H4-A76C-102 was synthesized manually. Truncated histone H4-(1-75) was cloned as a fusion protein with the GyrA intein and a chitin binding domain into the pTXB1 vector (New England

Biolabs). Fusion protein was overexpressed in E. coli BL21 (DE3) cells and purified from inclusion bodies. Histone H4 (1-75) thioester was generated by thiolysis of the fusion protein with MESNA. The resultant thioester was ligated with the synthetic peptide overnight. Free radical initiated desulfurization of the crude ligation mixture resulted in the native H4(K77Ac, K79Ac) which was purified by RP-HPLC. (Personal communication with John C. Shimko and Dr. Jennifer Ottesen).

Histone octamer preparation – Recombinant unmodified histones H2A, H2B, H3, and

H4 were expressed and purified as previously described [239]. The unmodified,

H3(K115Q), H3(K122Q), H3(K115Q, K122Q), H3(K115Ac), H3(K122Ac),

122

H3(K115Ac, K122Ac), H3(K56Q), and H4(K77Ac, K79Ac) were purified as previously described [285]. The purity of each octamer was confirmed by SDS-PAGE and mass spectrometry.

Nucleosome Reconstitution – Nucleosomes were reconstituted with 32P-labeled DNA substrate and with histone octamer containing unmodified, H3(K115Q), H3(K122Q),

H3(K115Q, K122Q), H3(K115Ac), H3(K122Ac), H3(K115Ac, K122Ac), H3(K56Q), and H4(K77Ac, K79Ac) by salt double-dialysis as previously described [285]. The nucleosome purity was verified by 5% native polyacrylamide gel containing 1/3x TBE.

Competitive Reconstitutions – Competitive reconstitutions were modified from published protocols to measure the change in DNA-histone binding induced by

H3(K115) and/or H3(K122) modifications [274]. Reconstitutions were prepared in 2 M

NaCl, 0.5x TE, 1 mM BZA with 1 ng/μl labeled substrate II, 1.5 ng/μl unlabeled substrate II, 80 ng/μl buffer DNA, and 20 ng/μl of HO (either unmodified, modified or mimicked) in a volume of 50 μl. To minimize variation in DNA and HO concentrations, we first prepared a master mix containing substrate II, buffer DNA, TE, and BZA. For each type of HO, we prepared a dilution of HO that was added to three aliquots of the

DNA master mix. Each HO sample was then dialyzed separately. Each sample was dialyzed against the same reservoir containing 200 mL of 2 M NaCl, 0.5x TE, and 1 mM

BZA. The concentration of salt in the dialysis reservoir was slowly reduced to <50 mM over 24 h; the samples were then dialyzed overnight against 0.5x TE and 1 mM BZA to reduce the final NaCl concentration to <1 mM NaCl. The reconstitution products were

123 examined by PAGE, scanned with a Typhoon 8600 variable mode imager (GE

Healthcare), and analyzed with ImageJ.

Binding Studies – Reactions were performed in 25 mM Hepes (pH 7.8), 15% glycerol,

100 mM NaCl, 1 mM DTT, and 2 mM MgCl2 containing 20ng/ul poly dI-dC, 200 ug/ml acetylated BSA (Promega) and approximately 5 fmol of 32P-labeled nucleosome-DNA or the 241-bp free DNA substrate in a final volume of 20 ul. Where indicated, 900 nM of streptavidin was included for 5 minutes prior to the addition of hMSH2-hMSH6. hMSH2-hMSH6 (at the indicated concentration) was preincubated with nucleosome-

DNA on ice for 10 minutes. The reactions were separated on a 5% native polyacrylamide/5% glycerol in 1/3x TBE at 4 ˚C for three hours. Gels were dried, quantified by phosphoimager (Molecular Dynamics), and represented as percent substrate shifted. Standard deviation was calculated from at least three separate experiments.

Chromatin Remodeling – Chromatin remodeling was analyzed as previously described

[285]. Briefly, reactions were performed in 25 mM Hepes (pH 7.8), 15% glycerol, 100 mM NaCl, 1 mM DTT, 5 mM MgCl2, 20 ng/ul poly dI-dC, 200 ug/ml acetylated BSA, and approximately 5 fmol of 32P-labeled nucleosome-DNA in a 20 ul reaction. Where indicated, 900 nM of streptavidin was included for 5 min on ice prior to the addition of hMSH2-hMSH6. Reactions were incubated with hMSH2-hMSH6 (250 nM) on ice for

10 minutes. Dissociation with 1 mM ATP was performed where indicated by addition of nucleotide and a further incubation from 10 to 60 minutes at 37 ˚C. The reactions were separated on a 5% native polyacrylamide gel/5% glycerol in 1/3x TBE at 4 °C for 3

124 hours. Gels were dried and quantified by phosphoimager (Molecular Dynamics).

Standard deviations were calculated from at least three different experiments.

125

VI. Supplemental Figures

Figure 19

Figure 19. Nucleosome-DNA substrates and gel analysis of purified nucleosome- DNA substrates. (A) The nucleosome-DNA substrate I contains 17 bp 5’ of the 147 bp Xenopus 5S rDNA nucleosome localization sequence (red) followed by 77 bp containing a mismatch 20 bp from the 3’ end that contains a terminal biotin (light blue). The nucleosome-DNA substrate II contains 10 bp 5’ of the 147 bp Xenopus 5S rDNA nucleosome localization sequence (red) followed by 35 bp and has a Cy5 (black) attached to the 5’-end. (B) H3-K115 and H3-K122 are buried below the DNA at the nucleosome dyad. The H3-K56 is buried below the DNA at the entry-exit region of the nucleosome. Face view of the nucleosome (Protein Data Bank code 1AO1). H3 residues 1-109 are shown in yellow. H3-K115 is shown as blue spheres, H3-K122 is shown as red spheres, and H3-K56 is shown as green spheres. (C) 5% acrylamide, 1/3x TBE gel analysis of purified nucleosome-DNA substrates reconstituted with substrate I in (A).

126

Figure 20

Figure 20. Gel Analysis of purified nucleosome-DNA substrates reconstituted with high affinity nucleosome localization sequences. 5% acrylamide, 1/3x TBE gel analysis of purified nucleosome-DNA substrates reconstituted with substrate I in Suppl. Fig. 19A except that the nucleosome localization sequence was either that of Xenopus 5S rDNA, 601, or the pMP2 nucleosome positioning sequence. The unmodified and the H3(K115Ac, K122Ac) modified nucleosomes reconstituted on Xenopus 5S rDNA are from Suppl. Fig. 19C.

127

Figure 21

Figure 21. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-DNA substrates reconstituted with high affinity 601 nucleosome localization sequence. (A-B) Representative gels showing nucleosome disassembly reaction catalyzed by the hMSH2-hMSH6 with (A) G/T mismatch nucleosome-DNA containing an unmodified nucleosome reconstituted on a 601 nucleosome localization sequence (B) G/T mismatch nucleosome-DNA containing a H3(K115Ac, K122Ac) modified nucleosome reconstituted on a 601 nucleosome localization sequence. Boxes above indicate added reaction components (+) and inclusion of nuclesome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left of the gel panels. The DNA substrate is colored with a nucleosome (blue oval), hMSH2- hMSH6 (purple clamp), and streptavidin (green circle). Red arrow indicates the gel mobility of the nucleosome disassembly product.

128

Figure 22

Figure 22. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-DNA substrates reconstituted with high affinity pMP2 nucleosome localization sequence. (A-B) Representative gels showing nucleosome disassembly reaction catalyzed by the hMSH2-hMSH6 with (A) G/T mismatch nucleosome-DNA containing an unmodified nucleosome reconstituted on a pMP2 nucleosome localization sequence (B) G/T mismatch nucleosome-DNA containing a H3(K115Ac, K122Ac) modified nucleosome reconstituted on a pMP2 nucleosome localization sequence. Boxes above indicate added reaction components (+) and inclusion of nuclesome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left of the gel panels. The DNA substrate is colored with a nucleosome (blue oval), hMSH2- hMSH6 (purple clamp), and streptavidin (green circle). Red arrow indicates the gel mobility of the nucleosome disassembly product.

129

Figure 23

Figure 23. Gel analysis of purified nucleosome-DNA substrates reconstituted with variations in the nucleosome localization sequences. 5% acrylamide, 1/3x TBE gel analysis of purified nucleosome-DNA substrates reconstituted with substrate I in Suppl. Fig. 19A except that the nucleosome localization sequence was composed of Xenopus 5S rDNA-601, 601-5S rDNA, or the pMP2-5S rDNA nucleosomal-dyad axis.

130

Figure 24

131

Continuing Figure 24

Figure 24. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-DNA substrates reconstituted with variations in nucleosome localization sequences. (A-C) Representative gels showing nucleosome disassembly reaction catalyzed by the hMSH2- hMSH6 with (A) G/T mismatch nucleosome-DNA containing an unmodified and H3(K115Ac, K122Ac) modified nucleosome reconstituted on a 5S-601 nucleosome localization sequence (B) G/T mismatch nucleosome-DNA containing an unmodified and H3(K115Ac, K122Ac) modified nucleosome reconstituted on pMP2-5S nucleosome localization sequence. (C) G/T mismatch nucleosome-DNA containing an unmodified and H3(K115Ac, K122Ac) modified nucleosome reconstituted on 601-5S nucleosome localization sequence. Boxes above indicate added reaction components (+) and inclusion of nuclesome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left of the gel panels. The DNA substrate is colored with a nucleosome (blue oval), hMSH2-hMSH6 (purple clamp), and streptavidin (green circle). Red arrow indicates the gel mobility of the nucleosome disassembly product.

132

Figure 25

Figure 25. Nucleosome-Mismatch substrates and gel analysis of purified nucleosome-Mismatch substrates. (A) The nucleosome-DNA substrate A (entry-exit region mismatch) contains 17 bp 5’ of the 147 bp Xenopus rDNA nucleosome localization sequence (red) followed by 77 bp containing a mismatch (G/T) 84 bp from the 3’ end that contains a terminal biotin (light blue). The nucleosome-DNA substrate B (LRS region mismatch) contains 17 bp 5’ of the 147 bp Xenopus rDNA nucleosome localization sequence (red) followed by 77 bp containing a mismatch (G/T) 103 bp from the 3’ end that contains a terminal biotin (light blue). The nucleosome-DNA substrate C (nucleosomal-dyad axis mismatch) contains 17 bp 5’ of the 147 bp Xenopus rDNA

133

Continuing Figure 25 nucleosome localization sequence (red) followed by 77 bp containing a mismatch (G/T) 148 bp from the 3’ end that contains a terminal biotin (light blue). (B) H3-K115 and H3- K122 are buried below the DNA at the nucleosomal-dyad. The H3-K56 is buried below the DNA at the entry-exit region of the nucleosome. H4-K77 and H4-K79 are located at the DNA between the entry-exit and the DNA dyad axis (LRS region). Face view of the nucleosome (Protein Data Bank code 1AO1). H3 residues are shown in yellow. H4 residues are shown in orange. H3-K115 is shown as blue spheres, H3-K122 is shown as red spheres, H3-K56 is shown as green spheres, H4-K79 and H4-K77 are shown in purple spheres (C) 5% acrylamide, 1/3x TBE gel analysis of purified nucleosome- Mismatch substrates reconstituted with substrate A, B, or C (A) with unmodified, H3(K115Ac, K122Ac), H3(56Q), or H4(K77Ac, K79Ac) modified octamer.

134

Figure 26

Figure 26. Nucleosome binding by hMSH2-hMSH6 to nucleosome-Mismatch substrates. (A-D) Representative gels showing specific binding of hMSH2-hMSH6 to a G/T mismatch within the nucleosome (A) nucleosome-Mismatch-B substrate (LRS region mismatch) containing an unmodified and H3(K115Ac, K122Ac) nucleosome (B) nucleosome-Mismatch-C substrate (nucleosomal-dyad axis region mismatch) containing an unmodified and H3(K115Ac, K122Ac) nucleosome (C) nucleosome-Mismatch-A substrate (entry-exit region mismatch) containing a H3(K56Q) nucleosome. (D) nucleosome-Mismatch-B substrate (LRS region mismatch) containing a H4(K77Ac, K79Ac) nucleosome Boxes above indicate added reaction components (+) and inclusion of nuclesome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left of the gel panels. The DNA 135

Continuing Figure 26 substrate is colored with a nucleosome (blue oval), hMSH2-hMSH6 (purple clamp), and streptavidin (green circle).

136

Figure 27

Figure 27. Nucleosome disassembly by hMSH2-hMSH6 of nucleosome-Mismatch substrates. (A-E) Representative gels showing nucleosome disassembly reaction catalyzed by of hMSH2-hMSH6 (A) nucleosome-Mismatch-B substrate (LRS region mismatch) containing an unmodified and H3(K115Ac, K122Ac) modified nucleosome (B) nucleosome-Mismatch-C substrate (nucleosomal-dyad axis region mismatch) containing an unmodified and H3(K115Ac, K122Ac) modified nucleosome (C) nucleosome-Mismatch-A substrate (entry-exit region mismatch) containing an H3(K56Q) nucleosome (D) nucleosome-Mismatch-B substrate (LRS region mismatch) containing an H4(K77Ac, K79Ac) nucleosome (E) nucleosome-DNA substrate with an adjacent mismatch containing an H4(K77Ac, K79Ac) nucleosome. Boxes above indicate added reaction components (+) and inclusion of nuclesome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left of the gel panels. The DNA substrate is colored with a nucleosome (blue oval), hMSH2- hMSH6 (purple clamp), and streptavidin (green circle). Red arrow indicates the gel 137

Continuing Figure 27 mobility of the nucleosome disassembly product.

138

Chapter 5

Histone H3(K56) acetylation enhances hMSH2-hMSH6 nucleosome disassembly

I. Summary

DNA repair in vivo occurs in the context of mismatches and lesions which arise within chromosomes. Chromosomes consist of DNA and protein (chromatin) where

~147 bp of DNA is wrapped twice around a histone octamer. The octamer consists of two H2A-H2B dimers and a H3-H4 tetramer. The biochemical effect of mismatch repair

(MMR) on chromatin is unknown. Post-translational modifications (PTMs) of histones play a central role in genome expression, replication, and repair. Histone H3(K56) acetylation is essential for transcription, replication, and repair. We demonstrate that

H3(K56Ac) enhances nucleosome disassembly by hMSH2-hMSH6. This indicates that certain PTMs (involved in DNA repair and DNA replication) enhance nucleosome disassembly by hMSH2-hMSH6.

II. Introduction

Mismatches and lesions can arise in DNA as a result of misincorportion or chemical and physical damage [2]. The MMR pathway corrects mismatched bases and insertion/deletion loops (IDLs) [132]. The hMSH2-hMSH6 heterodimer is responsible 139 for the initial recognition of mismatched nucleotides and small IDLs. Mutations in MMR genes result in Lynch Syndrome or hereditary non-polyposis colorectal cancer

(LS/HNPCC). Although MMR occurs in the context of chromatin, the majority of biochemical studies have extensively studied MMR in the context of naked mismatched

DNA substrates.

The eukaryotic genome is organized of units of nucleosomes which contain ~147 bp of DNA wrapped twice around H2A, H2B, H3, and H4 [288]. DNA wrapped around octamer is occluded from DNA interacting proteins [242]. However, the DNA must be accessed for replication and repair. Structural alterations within the nucleosomes are controlled by PTMs [289]. PTMs on histone tails can function as binding modules for chromatin associated proteins and/or influence chromatin structure [290, 291].

Moreover, PTMs that occur within the nucleosome are often inaccessible and may alter chromatin structure and histone-DNA dynamics [259, 283].

Acetylation of histone H3 at lysine-56 [H3(K56)] is required for numerous DNA processes [181, 292, 293]. The enzymes responsible for the acetylation and deacetylation have been identified in yeast and humans [186, 294-296]. Histone H3(K56) is acetylated prior to deposition onto newly replicated DNA [183] and may be important in DNA repair, genomic stability, and transcriptional activation [292, 297]. The crystal structure of the nucleosome demonstrates that lysine-56 contacts DNA (Suppl. Fig. 29B) and suggests that it may increase the accessibility of DNA binding proteins to access nucleosomal-DNA.

140

Here we report the influence of histone H3(K56Ac) on hMSH2-hMSH6 nucleosome disassembly. These results provide further evidence that PTMs can considerably increase chromatin disassembly by hMSH2-hMSH6.

III. Results and Discussion hMSH2-hMSH6 nucleosome disassembly

We have recently demonstrated that hMSH2-hMSH6 can disassemble nucleosomes and this activity is enhanced 5-fold by the acetylaton of histone H3 at lysine

115 and 122 [285]. Moreover, nucleosomes were disassembled 2-fold faster by the acetylation mimic of histone H3 lysine 56 [H3(K56Q)], where the lysine was substituted for a glutamine [285]. The disassembly of nucleosomes containing H3(K56Ac) was studied using the Xenopus 5S rDNA nucleosome positioning sequence (NPS) [285]

(Suppl. Fig. 29A). The mononucleosome substrates were purified on a 5%-30% sucrose gradient and analyzed on a PAGE gel (Suppl. Fig. 29C). We found that H3(K56Ac) nucleosomes disassembled ~2 fold faster (Fig. 28A,B t1/2•H3(K56Ac) = 69±3 min) than unmodified nucleosomes (Fig. 28A,B t1/2•UN = 117±14 min; Suppl. Fig. 30A,B) by hMSH2-hMSH6 [285]. Moreover, mononucleosomes reconstituted with H3(K56Q) mimic histone octamer disassembled similar to that of H3(K56Ac) acetylated nucleosomes (Fig. 28A,B t1/2•H3(K56Q) = 53±3 min) by hMSH2-hMSH6 [285]. This indicates that H3(K56Ac) enhances nucleosome disassembly by hMSH2-hMSH6.

During DNA replication, nucleosomes are assembled with H3(K56Ac) [298].

Deletion of Rtt109 (which acetylates H3-K56) or mutation of H3(K56) to Arg

[H3(K56R)] in yeast, causes defects in post-replication DNA repair as unacetylated 141

Figure 28

Figure 28. H3(K56Ac) enhances hMSH2-hMSH6 nucleosome disassembly. (A-B) Quantitative analysis of the nucleosome disassembly reactions. Data analysis includes representative gels (Suppl. Fig. 29A,B) [285]. (A) Datum was fit to a single exponential decay to calculate τ and t1/2. (B) Table of rate disassembly values (t1/2). Key: unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/C and biotin-streptavidin blocked (b*) 3’-tail (UN Nuc-G/C-b*), H3(K56Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/C and biotin-streptavidin blocked (b*) 3’-tail (K56Ac Nuc-G/C- b*), unmodified nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (UN Nuc-G/T-b*), H3(K56Ac) nucleosome substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (K56Ac Nuc-G/T- b*), H3(K56Q) substrate with Xenopus 5S rDNA nucleosome positioning sequence containing a G/T and biotin-streptavidin blocked (b*) 3’-tail (K56Q Nuc-G/T-b*). The UN Nuc-G/T-b*, UN Nuc-G/C-b*, and K56Q Nuc-G/T-b* values are from our previous publication [285].

142

H3(K56) leads to genomic instability [184]. We recently showed that hMSH2-hMSH6 can disassemble nucleosomes near a mismatch and that this activity is 2-fold enhanced when nucleosomes contain H3(K56Q) [285]. Moreover, this study finds that histone

H3(K56Ac) increases nucleosome disassembly by hMSH2-hMSH6 2-fold compared to unmodified nucleosomes. Together, these observations indicate that H3(K56Ac) facilitates nucleosome disassembly by hMSH2-hMSH6.

IV. Experimental Procedures

Synthesis of H3-K56Ac (A47C, A91Ac, C110A) – All peptides were synthesized by in situ neutralization with Boc SPPs utilizing HBTU activation. The N-terminal and middle peptide segments were synthesized on propionamide thioester resin, cleaved with HF, and purified by RP-HPLC. The C-terminal peptide was synthesized on Wang resin and purified by RP-HPLC. Sequential native chemical ligation was employed to generate synthetic H3-K56Ac (A47Ac, A91Ac, C110A). Following the determination of the ligation product, unpurified ligation mixture containing H3(K56Ac) (A47C, A91C,

C110A) was directly desulfurized prior to purification. Reaction progress was monitored by MALDI and upon completion, the final desulferized product H3-K56Ac (C110A) was purified by RP-HPLC. (Personal Communication with John C. Shimko and Dr. Jennifer

Ottesen).

Nucleosome Preparation and DNA constructs – The DNA construct and nucleosome octamer were prepared and purified as previously described [285]. Nucleosomes were prepared by salt dialysis with purified histone octamer and DNA, and purified by sucrose gradient centrifugation [259]. 143 hMSH2-hMSH6 nucleosome disassembly assay – Nucleosome disassembly reactions were carried out as previously described [285] with 0.25 nM of H3(K56Ac) nucleosomes reconstituted with the Xenopus 5S rDNA NPS with an adjacent mismatch and a terminal

3’-biotin. Kinetic analysis was performed by staggering the time after ATP was added.

The fraction of disassembled nucleosome were analyzed by gel shifts on polyacrylamide gels as previously described [285].

144

V. Supplemental Figures

Figure 29

Figure 29. Nucleosome-DNA substrates and gel analysis of purified nucleosome- DNA substrates containing H3(K56Ac). (A) The nucleosome-DNA contains 17 bp 5’ of the 147 bp Xenopus 5S rDNA nucleosome localization sequence (red) followed by 77 bp containing a mismatch 20 bp from the 3’ end that contains a terminal biotin (light

145

Continuing Figure 29 blue). (B) Histone H3 is shown in yellow. H3-K56 contacts the DNA at the entry-exit region of the nucleosome. Face view of the nucleosome (Protein Data Bank code 1AO1). H3-K56 is shown as green spheres. (C) 5% acrylamide, 1/3x TBE gel analysis of purified nucleosome-DNA substrates reconstituted with (A) with and without a mismatch.

146

Figure 30

Figure 30. Representative gels showing the nucleosome disassembly reaction catalyzed by hMSH2-hMSH6 with (A) G/T mismatch with H3(K56Ac) nucleosomes, (B) G/C duplex with H3(K56Ac) nucleosomes. Boxes above indicate added reaction components (+) and the inclusion of nucleosome-DNA (N). A schematic of DNA species with arrows or brackets indicating gel mobility position is shown on the left of the gel panels. The DNA substrate is colored with a nucleosome (blue oval), hMSH2-hMSH6 (purple clamp), and streptavidin (green circle). Brackets indicate the putative gel mobility of substrates when bound by hMSH2-hMSH6.

147

Chapter 6

Histone H3(T118) phosphorylation enhances hMSH2-hMSH6 nucleosome disassembly

I. Summary

DNA repair in vivo is complicated by the fact that mismatches/lesions arise within chromosomes. Chromosomes are a mixture of DNA and protein (chromatin). Chromatin consists of ~147 bp of DNA wrapped twice around a histone octamer consisting of two

H2A-H2B dimers and a H3-H4 tetramer. The biochemical effect of MMR on chromatin is relatively unknown. Moreover, little is known about the effect of the more than 100 post-translational modifications (PTMs) that may decorate the human histones on MMR processes. The histone H3(T118) residue is phosphorylated H3(T118ph) and is important for transcription and repair. We demonstrate that H3(T118ph) enhances nucleosome disassembly by the hMSH2-hMSH6 mismatch recognition complex.

II. Introduction

Mismatches in DNA can arise as a result of misincorporation errors or as a result of chemical and physical damage [2]. MutS homologues (MSH) and MutL homologues

(MLH/PMS) play central roles in human mismatch repair (MMR) [132]. hMSH2- hMSH6 is required for the initial recognition of single nucleotide and small

148 insertion/deletion loop (IDL) type DNA mismatches as well as damage-induced nucleotide lesions. Mutations in hMSH2, hMSH6, hMLH1, and hPMS2 genes result in an elevated spontaneous mutation rate and susceptibility to a cancer predisposition syndrome, Lynch Syndrome or hereditary non-polyposis colorectal cancer (LS/HNPCC).

Although MMR occurs in the context of chromatin, relatively little is known.

Histone PTMs have been implicated in the regulation of chromatin remodeling

[178]. Histone H3(T118I) was identified in S. cerevisiae through identification of

SWI/SNF-independent (SIN) allele [272]. SIN mutations are thought to suppress mutations in the ATP-dependent nucleosome remodeling complex (SWI/SNF) by causing increased mobility of the nucleosome [257]. Replacing H3(T118) with glutamic acid (or alanine) is lethal to yeast and exhibits a dominant lethal effect [257]. Histone

H3(T118A) shows a slight increase in sensitivity to hydroxyurea (HU) [257]. The crystal structure of H3(T118I) has been crystallized [269] and exhibits weakened DNA contacts with the histone (Fig. 31A,B). Histone H3(T118) is located near the nucleosomal-DNA dyad symmetry axis at the DNA-histone interface (Fig. 31A). Phosphorylation of

H3(T118) is predicted to weaken histone-DNA interactions and may influence chromatin structure and/or histone-DNA dynamics.

Here we report the influence of H3(T118ph) on hMSH2-hMSH6 nucleosome disassembly. These results provide further evidence that PTMs may increase hMSH2- hMSH6 nucleosome disassembly.

149

III. Results and Discussion hMSH2-hMSH6 nucleosome disassembly

Recently, we demonstrated that the mismatch recognition complex hMSH2- hMSH6 is capable of disassembling nucleosomes and that this activity is enhanced 5-fold by acetylation of histone H3 at lysine 115 and 122 [H3(K115Ac, K122Ac)] [285]. The disassembly of nucleosomes containing H3(T118ph) was initially studied using the

Xenopus 5S rDNA nucleosome positioning sequence (NPS) similar to our previous studies [285]. The nucleosome-DNA substrates (unmodified and H3(T118ph)) were reconstituted and purified on a 5-30% sucrose gradient [260]. We found that nucleosomes reconstituted with 5S rDNA NPS containing H3(T118ph) reconstituted poorly and disassembled too rapidly for quantification of the data. We used a significantly more stable pMP2 (a variant of the 601 sequence) (Fig. 31B, Suppl. Fig.

33), and found that H3(T118ph) nucleosomes were disassembled 25±7 times faster (Fig.

32A,B t1/2•H3(T118ph) = 55±4 min) than unmodified nucleosomes (Fig. 32A,B t1/2•H3 =

1400±400 min; Suppl. Fig. 34A,B) by hMSH2-hMSH6. This indicates that H3(T118ph) dramatically increases nucleosome disassembly by hMSH2-hMSH6. Moreover, this modification may significantly accelerate nucleosome disassembly near DNA mismatches.

The H3(T118ph) places a negative charge close to the DNA phosphate backbone.

This suggests an electrostatic repulsion between the threonine and the DNA backbone may reduce nucleosome stability. The H3(T118I) SIN mutant displays similar but less pronounced effects on nuclosome mobility [241, 269]. The crystal structure of 150 nucleosomes containing H3(T118I) and H3(T118H) resulted in a distortion of the nucleosomal-DNA [269]. These observations indicate that the properties and locations of

H3(T118ph) may be responsible for reduction in nucleosome stability and remodeling.

Our studies demonstrate that histone phosphorylation within the nucleosome DNA- histone interface enhances hMSH2-hMSH6 nucleosome disassembly.

We have demonstrated that unmodified and H3(K115, K122) acetylated nucleosomes reconstituted on high affinity artificial NPS’s (i.e. pMP2 and 601) mask nucleosome disassembly by hMSH2-hMSH6. However, H3(T118ph) nucleosomes reconstituted on pMP2 are efficiently disassembled by hMSH2-hMSH6. These studies indicate that different PTMs (i.e acetylation and phosphorylation) may have different impacts on nucleosome stability and/or disassembly by hMSH2-hMSH6.

151

Figure 31

Figure 31. H3(T118ph) crystal structure. (a) The side view of the nucleosome structure with histone H3 in yellow and H3(T118) in cyan [299]. (b) The DNA constructs: mp2-GT and mp2-GC contain 15 bp 5’ of the mp2 NPS, 70 bp 3’ of the mp2 NPS containing a biotin attached to the 5’-end. mp2-GT contains a GT mismatch 35 bp to the right of the mp2 NPS.

152

Figure 32

Figure 32. H3(T118ph) enhances hMSH2-hMSH6 nucleosome disassembly. (a) Electrophoretic mobility shift analysis of H3(T118ph) nucleosomes disassembled by hMSH2-hMSH6. Lane 1) sucrose gradient purified H3(T118ph)-containing nucleosomes, Lane 2) H3(T118ph) nucleosomes bound by streptavidin, Lane 3) H3(T118ph) nucleosomes bound by streptavidin and hMSH2-hMSH6, Lanes 4 – 9) kinetic analysis of streptavidin-bound H3(T118ph) nucleosome disassembly by hMSH2- hMSH6 in the presence of 1 mM ATP. (b) The fraction of unmodified mp2-GC NPS nucleosomes (triangle), H3(T118ph) mp2-GT NPS nucleosomes (circles) and H3(T118ph) mp2-GC NPS nucleosomes (squares) verses time in the presence of hMSH2-hMSH6 (250 nM) and ATP (1 mM). The error bars were determined from the standard deviation of at least 3 separate experiments. The fraction of nucleosomes vs. time were fit excluding the zero time point to a single exponential decay, A*e-t/τ with τmp2-GT pT118 = 55±4 min and τmp2-GT unmod = 1400±400 min. We interpret the negative decay time of H3(T118ph) mp2-GC NPS nucleosomes to imply that no nucleosomes were disassembled within the uncertainty of the measurement.

153

IV. Experimental Procedures

Nucleosome Preparation and DNA constructs – Nucleosomes were prepared by salt dialysis with purified histone octamer and DNA and purified by sucrose gradient centrifugation [259]. The additional of 0.5 mM MgCl2 was required to maintain stability during the purification of H3(T118ph) nucleosomes. H3(T118ph) was prepared by expressed protein ligation as previously described [259]. The peptide synthesized with the phosphothreonine was prepared using standard HBTU protocols. Wild-type histones were prepared by recombinant expression in E. coli and purified as previously described

[254, 255]. The DNA construct, mp2-MM, was prepared and purified as previously described [285]. hMSH2-hMSH6 nucleosome disassembly assay – Nucleosome disassembly reactions were carried out as previously described [285] with 0.25 nM of H3(T118ph) nucleosomes reconstituted with the mp2-GT NPS and 5 nM of unlabeled unmodified nucleosomes reconstituted with just the MP2 NPS. Kinetic analysis was performed by staggering the time after ATP was added. The fraction of disassembled nucleosome were analyzed by gel shifts on polyacrylamide gels as previously described [285].

154

V. Supplemental Figures

Figure 33

Figure 33. Preparation of nucleosomes. Phosphorimage analysis of purified H3(T118ph) containing nucleosomes with: Lane 1) mp2-GT mismatch NPS and Lane 2) mp2-GC duplex NPS.

155

Figure 34

Figure 34. hMSH2-hMSH6 disassembly of unmodified nucleosomes with a DNA mismatch and of H3(T118ph) containing nucleosomes without a DNA mismatch. (a) and (b) are native PAGE analysis of disassembly by hMSH2-hMSH6 of H3(T118ph) containing nucleosomes with a GC duplex and of unmodified nucleosomes with a GT mismatch, respectively. For each gel: Lane 1) the sucrose gradient purified nucleosome construct, Lane 2) the nucleosome construct bound by streptavidin, Lane 3) the nucleosome bound by streptavidin and hMSH2-hMSH6 (250 nM), and Lanes 4 – 9) nucleosomes with streptavidin and hMSH2-hMSH6 (250 nM) for increasing time in the presence of ATP (1 mM).

156

Chapter 7

Conclusions and Future Directions

Prior to these studies, assays in vitro using purified human proteins in MMR reactions have been performed on naked mismatched DNA [58, 62]. The influence of chromatin structure on human MMR is largely unknown. Moreover, we also do not fully understand how human MMR proteins recognize and interact with mismatches in chromatin. By reconstituting a single nucleosome containing a mismatch adjacent or within the nucleosome, we were able to investigate the biochemical properties of the major eukaryotic mismatch recognition factor, hMSH2-hMSH6, in chromatin. These studies reveal a function of the hMSH2-hMSH6 mismatch initiation heterodimer.

Nucleosome disassembly by hMSH2-hMSH6 requires a cis mismatch and translocation of the heterodimer along the DNA, ATP binding but not hydrolysis, is enhanced by histone post-translational modifications that increase thermal repositioning and/or reduce histone affinity, and is dependent on the nucleosome localization sequence. Taken together, our observations appear to suggest a novel framework for a passive mechanism for nucleosome disassembly by hMSH2-hMSH6.

157 hMSH2-hMSH6 recognition of a mismatch located adjacent to or within the nucleosome localization sequence of a nucleosome

hMSH2-hMSH6 recognizes and binds to a mismatch located adjacent to a single nucleosome with similar affinity as mismatches on naked heteroduplex DNA. However, recognition and binding of mismatches within the nucleosome by hMSH2-hMSH6 occurs with decreased affinity compared to naked mismatched DNA or a mismatch adjacent to a nucleosome. We find that hMSH2-hMSH6 recognizes a mismatch located at the entry- exit region with higher affinity than a mismatch located at the LRS or the nucleosomal- dyad axis region. One could speculate that efficient processing of mismatches within the nucleosome requires additional help of chromatin modifying and/or remodeling proteins

[300]. Alternatively, novel PTMs or combinations of PTMs on histones of nucleosomes may allow mismatches within the nucleosome to become accessible to hMSH2-hMSH6

[259]. Another mechanism for accessing mismatches within the nucleosome is nucleosome sliding and/or translocation that may occur via recapture of free nucleosomal-DNA at a different site on the histone octamer leading to a translational shift of the nucleosome. However, the role of processing this translational shift in facilitating DNA processes in unknown. A different mechanism to access nucleosomal-

DNA is direct binding to DNA on the surface of the histone octamer. This was demonstrated by transcription factors (i.e. glucocorticoid receptor and heat shock protein)

[301, 302]. Moreover, intrinsic nucleosomal-dynamics involving transient, spontaneous release of nucleosomal-DNA have been shown to facilitate access by transcription factors and other DNA binding proteins to nucleosomal-DNA buried within the nucleosome

[242, 244, 303]. Unwrapping of nucleosomal-DNA may account for the ability of DNA 158 ligase I to access nucleosomal-DNA [304, 305]. Moreover, partial unwrapping may account for certain exonucleases and RNA polymerase to access nucleosomal-DNA [306-

308]. Consistent with this, we find that hMSH2-hMSH6 maximum binding increases 2- fold when the mismatch is located in the entry-exit region compared to a mismatch in the

LRS or the nucleosomal-dyad region. We speculate that a mismatch in the entry-exit region within the nucleosome is captured by hMSH2-hMSH6 during episodes of partial unwrapping of the nucleosome. Mismatches within the LRS and the nucleosomal-dyad axis on the NPS may be occluded by the histone octamer leading to decreased capture of the mismatch by hMSH2-hMSH6. The rate of dynamic partial unwrapping is influenced by nucleosomal-DNA sequence, with stronger nucleosomal positioning sequence shown to be associated with reduced site exposure [243]. Processing of the mismatch will primarily rely on mismatch capture by hMSH2-hMSH6 during the exposure of long stretches of nucleosomal-DNA. Moreover, the frequency of DNA unwrapping may also be influenced by a mismatch-induced distortion; distorting mismatches may destablize histone-DNA interactions, leading to increased unwrapping frequency [309]. Moreover, previous reports have sugggested that histone PTMs increase nucleosome thermal repositioning and/or reduce the histone-DNA affinity [259]. Thus, mismatches in the entry-exit region of nucleosomes in combination with histone PTMs on nucleosomes may further destabilize histone-DNA interactions leading to increased unwrapping of nucleosomal-DNA allowing hMSH2-hMSH6 easier access to a mismatch.

hMSH2-hMSH6 nucleosome disassembly and nucleosome localization sequence

A recent study showed that the yeast genome is nine times more likely to be

159 occupied by nucleosomes in vitro compared to that of E. coli [310]. Our results suggest that NPS’s have a significant impact on hMSH2-hMSH6 nucleosome disassembly. Li et al. suggested that nucleosomes inhibit MMR [248]. This study was based on reconstituted nucleosomes from histone octamers formed from unmodified histone proteins which were positioned on a high affinity NPS [248]. We analyzed a physiological relevant Xenopous 5S rDNA, artifical high affinity 601, and the artifical high affinity pMP2, NPS’s, and find that hMSH2-hMSH6 can disassemble nucleosomes on Xenopus 5S rDNA but nucleosome disassembly is masked on the artifical 601 and the pMP2 NPS’s. However, previous studies have demonstrated that varying the strength of the binding interface between histones and nucleosomal-DNA does not affect the rate of nucleosome remodeling [311]. Partensky el al. suggested that chromatin remodeling enzymes remodel nucleosomes reconstituted on various sequences at similar rates [311].

However, Partensky et al. study was based on ATP-dependent chromatin remodelers which require ATP to remodel chromatin. Moreover, hMSH2-hMSH6 does not require

ATP to disassemble chromatin; thus the affinity of a histone octamer to a particular sequence may play a much greater role in the ability of hMSH2-hMSH6 to disassemble chromatin.

Post-translational Modifications and hMSH2-hMSH6 nucleosome disassembly

Histone PTMs may directly alter the interaction between DNA and histones to modify the DNA sequence preference of nucleosomes which could effect nucleosome repositioning and histone-DNA interactions [259]. This may determine the ease in which the nucleosome can be translocated along the DNA [283]. Histone PTMs may change

160 the affinity of histone octamer for DNA resulting in equilibrium changes between unmodified and modified nucleosomes. We observe that nucleosome disassembly by hMSH2-hMSH6 is greatly enhanced by histone PTMs compared to unmodified nucleosomes. Moreover, we also observe that phosphorylation and acetylation modifications differ in their ability to alter nucleosome structure; however, both modifications enhance nucleosome disassembly by hMSH2-hMSH6. We observe that nucleosomes containing a phosphorylation (H3[T118]) compared to a dual acetylation

(H3[K115Ac, K122Ac]) on histone H3 reconstituted on the high affinity pMP2 NPS, has increased disassembly by hMSH2-hMSH6. This may suggest that H3(T118) phosphorylation on histone H3 reduces nucleosome stability to a greater degree than

H3(K115, K122) acetylations (by impacting the histone-DNA interface or increasing thermal mobility), thus increasing the rate of hMSH2-hMSH6 nucleosome disassembly. hMSH2-hMSH6 nucleosome disassembly occurs via a passive mechanism

Several mechanisms for MMR have been proposed [17, 132]. The molecular switch model proposes that addition of ATP to hMSH2-hMSH6 bound to a mismatch results in the formation of a diffusable clamp that remains stably associated with the

DNA backbone [20, 21, 138]. hMSH2-hMSH6 sliding clamps diffuse bidirectionally and stochastic loading of multiple hMSH2-hMSH6 sliding clamps onto the DNA backbone signals the remaining repair machinery. Diffusion of ATP-bound hMSH2- hMSH6 does not appear to require ATP hydrolysis. Instead, ATP hydrolysis appears to occur after the dissociation of ATP-bound hMSH2-hMSH6 from a free end (in vitro), resulting in recycling of hMSH2-hMSH6 mismatch binding activity [20, 21, 138]. The

161 translocation model suggests that mismatch DNA activates an ATP hydrolysis-dependent translocation of hMSH2-hMSH6 along the DNA backbone [137].

Recognition of a mismatch by hMSH2-hMSH6 must occur within the chromatin.

Our results indicate that hMSH2-hMSH6 recognizes mismatches and also facilitates the

MMR reaction by ensuring that the DNA surrounding the mismatch is nucleosome and perhaps protein-free. We demonstrated that hMSH2-hMSH6 disassembles nucleosomes adjacent to/containing a mismatch and that histone PTMs enhance the rate of nucleosome disassembly by hMSH2-hMSH6 [285]. We propose that hMSH2-hMSH6 disassembles nucleosomes via a novel passive mechanism by trapping DNA unwrapping fluctuations through stochastic loading of multiple hMSH2-hMSH6 at the DNA mismatch [285].

This may occur if hMSH2-hMSH6 exerts pressure on the nucleosome. The pressure builds as additional hMSH2-hMSH6 sliding clamps are loaded onto the DNA at the mismatch. This may shift the equilibrium amount of DNA wrapped around the nucleosome to a more unwrapped state. The nucleosome eventually reach an unwrapped state that allows for histone octamer disassembly. PTMs that increase nucleosome thermal mobility and/or reduce histone-DNA interactions may increase the probability for nucleosome unwrapping, allowing the nucleosome to be disassembled at a faster rate by hMSH2-hMSH6. Consistent with this, we find that dyad modifications, H3(T118ph) and

H3(K115Ac, K122Ac), and entry-exit modification, H3(K56Ac), enhance hMSH2- hMSH6 nucleosome disassembly.

Our model explains the contradictory results that report nucleosomes are barriers to hMSH2-hMSH6 [248, 249]. Li et al. [248] observed no nucleosome disassembly by

162 hMSH2-hMSH6. However, Li et al. used the 601 NPS and unmodified histone octamer.

As discussed earlier, we also find that the 601 NPS largely masks hMSH2-hMSH6 nucleosome disassembly with unmodified histone octamer. Gorman et al. [249] reported that nucleosomes create a barrier to yeast MSH2-MSH6 diffusion. They used λ DNA that contains nucleosomes but no mismatch. Gorman et al. tracked MSH2-MSH6 using a quantum dot along the DNA after washing out free MSH2-MSH6. As there is no mismatch and no additional MSH2-MSH6 can load onto the DNA, multiple MSH2-

MSH6 cannot be loaded onto the λ DNA between nucleosomes. Thus, a single MSH2-

MSH6 heterodimeric protein may not be able to disassemble nucleosomes as the amount of DNA wrapped around the nucleosome cannot be shifted into the unwrapped state, without the help of additional MSH2-MSH6 complexes. This would indicate why

Gorman et al. reported that nucleosomes become barriers for a single MSH2-MSH6 sliding clamp. It is important to note that Gorman et al. assume that MSH2-MSH6 can form a sliding clamp on λ DNA that does not contain a mismatch.

Downstream events for mismatch repair

In E. coli, recognition of a mismatched nucleotide by MutS results in a formation of an ATP-bound MutS that recruits MutL. The MutS/MutL tetramer diffuses along the

DNA backbone and signals the MutH endonuclease either upstream or downstream of the mismatch [17]. MutH nicks the unmethylated strand of a hemi-methylated d(GATC) site

[83] and stimulates unwinding of the nicked strand towards the mismatch via DNA helicase II [17]. In yeast and humans, similar association between MSH2-MSH6 and

163

MLH1-PMS2 has been observed in the presence of ATP. Moreover, MLH1-PMS2 has latent endonuclease activity.

As in E. coli, human MMR occurs via a mechanism that involves bidirectional excision and synthesis of long single-stranded DNA tracts [267]. However, in contrast to

E. coli, DNA in yeast and human cells occur in chromatin. Thus, for MMR to occur, the area surrounding the mismatch may have to be nucleosome- and perhaps protein-free. In the context of the sliding clamp model, it is possible that multiple hMSH2-hMSH6 sliding clamps disassemble nucleosomes creating a nucleosome and/or protein-free environment around the mismatch. It is also possible that the rate of nucleosome disassembly may be enhanced by tetramers of hMSH2-hMSH6/hMLH1-hPMS2 sliding clamps.

Integrating hMSH2-hMSH6 nucleosome disassembly and mismatch repair phenotypes

Mutator Phenotype Previous studies with the human hMSH2-hMSH6 [20, 72] mismatch heterodimeric complex are consistent with yeast and bacterial studies which demonstrate that mutations within the adenosine nucleotide domain (binding and hydrolysis) result in a dominant mutator phenotype [71, 312]. There are two different alterations of hMSH2-hMSH6 that can result in a dominant mutator phenotype: i) hMSH2-hMSH6 cannot bind and/or exchange ADP for ATP; and ii) hMSH2-hMSH6 cannot hydrolyze ATP. An inability to bind and/or exchange ADP for ATP would result in a permanent mismatch bound hMSH2-hMSH6. This would block additional MMR machinery from the mismatch site. An inability to hydrolyze ATP would result in hMSH2-hMSH6 that is permanently inactive for mismatch binding. We demonstrate that 164 a hMSH2-hMSH6 mutant heterodimer (incapable of ATP binding and/or hydrolysis) is incapable of nucleosome disassembly [285]. Moreover, studies performed with an ATP non-hydrolyzable analog adenosine 5’-[γ-thio]-triphosphate (ATPγS), resulted in a slower rate for nucleosome disassembly [285]. The rate of hMSH2-hMSH6 nucleosome disassembly was decreased as the studies were complicated by a ATPγS prebinding reaction that inactivates hMSH2-hMSH6 mismatch binding and freezes the stochastic loading of hMSH2-hMSH6 sliding clamps. Both of these conditions may result in increased mutation rates as a consequence of unrepaired mismatched nucleotides.

Hereditary cancer-associated mutations in hMSH2-hMSH6 and nucleosome disassembly Cells which lack functional MMR have increased mutation rate [313].

Germline mutations in hMSH2 or hMLH1 account for a majority of LS/HNPCC cases; however, a subset of MMR mutations have been reported for hMSH6 [314]. The majority of reported mutations are truncations, deletions, or missense mutations whose functional significance is unclear. These mutations may disrupt MMR function, hMSH2- hMSH6 nucleosome disassembly, and may contribute to disease. Whether known

LS/HNPCC mutations in hMSH2 or hMSH6 affect hMSH2-hMSH6 nucleosome disassembly has yet to be determined. Further investigation should reveal the significance of hMSH2-hMSH6 nucleosome disassembly and its contribution to

LS/HNPCC.

Mismatch repair, histone post-translational modifications, and LS/HNPCC

Recently it has been suggested that misregulation of histone PTMs (i.e. phosphorylation, methylation, and acetylation) may contribute to cancers [315].

165

Modulation of chromatin through covalent histone modifications is a way of regulating

DNA accessibility during DNA repair processes (i.e. gene transcription, DNA damage repair, and DNA replication). Histone modifications may directly effect nucleosome structure or provide a signaling platform to recruit proteins. Recent evidence has suggested that genetic alterations and epigenetic aberrations may contribute to cancers.

Changes in global histone modification pattern have been observed in multiple types of cancers (i.e. prostrate cancer) [316]. New evidence also suggests that histone marks and chromatin assembly processes are altered in cells experiencing replication problems

[284]. Colorectal cancers show epigenetic abnormalities (i.e. DNA methylation and histone modifications) [317]. However, whether MMR genes in combination with histone modifications confer special roles in cancer initiation and progression remains unclear.

Our work suggests that histone PTMs (involved in DNA replication and/or DNA repair) on nucleosomes play a role in the initiation step of MMR (i.e. histone PTMs enhance hMSH2-hMSH6 nucleosome disassembly). However, very little is known of

LS/HNPCC in combination with chromatin. DNA MMR genes function genetically as tumor supressors, yet it is not always clear how alteration identified in the gene effects gene function. Moreover, there does not appear to be partial function of truncated MMR proteins [46]. The ability to determine whether a variant of MMR genes that disrupt

MMR function may provide the best evidence of its pathogenicity. To that end, it would be interesting to examine the functional consequence of variants (disease-associated variants) in MMR genes in regards to hMSH2-hMSH6 nucleosome disassembly.

166

Examining the ability of a variant-containing hMSH2-hMSH6 to disassemble nucleosomes would allow us to observe the functionality of hMSH2-hMSH6 in chromatin containing PTMs. We may observe whether histone PTMs alleviates the requirement for hMSH2-hMSH6 complex. This may allow histone PTMs on nucleosomes to substitute for hMSH2-hMSH6 nucleosome disassembly function.

Consistent with this, previous studies have identified SIN mutations in yeast that alleviate the requirement for the yeast SWI/SNF remodeling complex [241]. It would also be interesting to determine whether inhibition of histone PTMs (by inhibiting histone modifying proteins) on nucleosomes and disease-associated variants of hMSH2-hMSH6

(partially defective alterations in MMR machinery) would create a synergistic mutator phenotype. Previous results have demonstrated that mutations in subunits of SWI/SNF remodeling complex exhibit synthetic lethality with mutations in components of the

GCN5-dependent histone acetyltransferese (HAT) complexes [318, 319]. Thus, histone

PTMs may highly impact MMR function and contribute to LS/HNPCC.

Further defining the mechanism of chromatin and mismatch repair

There are several unanswered questions in this area. It will be interesting to know if nucleosomes retain stoichiometric amounts of core histones upon hMSH2-hMSH6 binding to a mismatch within the NPS on the nucleosome. Certain transcription factors

(i.e. Gal4) have been found to destablize histone octamers [320] and may facilitate nucleosome disassembly. Moreover, it will be interesting to know if mismatch orientation (a mismatch facing inward toward the histone octamer or towards the surface) will effect hMSH2-hMSH6 mismatch binding. Consistent with this, Prasad et al.

167 reported that efficiency of BER of lesions in nucleosomes varies with the lesion's helical and translational position relative to the histone octamer [309]. This would suggest that mismatch substrates occluded by histone octamer may be processed slowly whereas mismatch substrates which are accessible for hMSH2-hMSH6 binding may be processed with efficiency closer to naked mismatched DNA.

It has been shown that hMSH2-hMSH6 binds to mismatches followed by recruitment of hMLH1-hPMS2. hMSH2-hMSH6/hMLH2-hPMS2 may signal subsequent repair steps. By blocking the stochastic cycles of hMSH2-hMSH6 loading and clamp formation, it would be interesting to test if the subsequent steps of MMR initiation alter hMSH2-hMSH6 nucleosome disassembly. It would also be interesting to know if nucleosome disassembly is observed on dinucleosome substrates or nucleosome arrays which better mimic cellular conditions.

No reports have yet identified the mechanism by which hMSH2-hMSH6 disassembles chromatin. Fluorescence resonance energy transfer (FRET) using labeled donor fluorophore on histone octamer and labeled acceptor fluorophore on DNA may identify whether the “nudging” or the “unwrapping” model accounts for the hMSH2- hMSH6 heterodimeric ability to disassemble chromatin. Changes in FRET between the fluorescence donor and fluorescence acceptor will allow us identify the mechanism of hMSH2-hMSH6 nucleosome disassembly. Addition of biotin-streptavidin on the 5’-end of the nucleosome-DNA in addition to biotin-streptavidin on the 3’-end may also clarify the hMSH2-hMSH6 nucleosome disassembly mechanism This would let us observe

168 whether the nucleosome is spontaneously disassembling or whether an open end is needed for nucleosome displacement.

Another idea would be to delete the PWWP (ΔPWWP) domain in hMSH6. The

PWWP domain is thought to be a potential protein-chromatin binding domain. It would be interesting to see whether hMSH2-hMSH6(ΔPWWP) can disassemble nucleosomes and whether the PWWP domain of hMSH6 is required for hMSH2-hMSH6 nucleosome disassembly.

A MMR system has been reconstituted using purified MMR proteins and naked mismatched DNA [58, 62]. The nature of how proteins downstream of hMSH2-hMSH6 interact with chromatin remains a mystery. It would be interesting to observe whether

MMR proteins downstream of hMSH2-hMSH6 have chromatin remodeling activity. By fully reconstituting an in vitro MMR system using chromatin, we may observe the interaction of purified MMR proteins in chromatin.

Future work may assay the activities of MMR proteins in a nucleosomal-array context that may provide more insight into MMR proteins activities on higher order structures. Nucleosomes are the basic repeating units of chromatin; DNA in eukaryotic cells are complexed with linker histone and other nonhistone proteins into 30-nm chromatin fibers and other higher-order structures. DNA damage that occurs in these structures would need to be repaired. It would be interesting to observe the activity of

MMR proteins on such substrates and what factors, such as histone tail PTMs may affect this activity.

169

Interaction studies between hMSH2-hMSH6 and chromatin may reveal chromatin remodelers which participate in MMR that have yet to be identified. At the moment, no chromatin remodeling activity and/or modifications have been observed in MMR. Future studies may reveal whether hMSH2-hMSH6 is the chromatin remodeler in MMR or whether other, yet undetermined, chromatin remodelers also exist..

Conclusion

This thesis provides a initial analysis of DNA binding, adenosine nucleotide binding, and ATPase properties of hMSH2-hMSH6 in chromatin. Our observations support the molecular switch model in that hMSH2-hMSH6 mismatch recognition is regulated by ADP and ATP. The idea that multiple hMSH2-hMSH6 sliding clamps are needed to disassemble nucleosomes provides a passive mechanism for hMSH2-hMSH6 nucleosome disassembly. Moreover, these findings suggest several testable predictions regarding the mechanism by how hMSH2-hMSH6 nucleosome disassembly is taking place. Future studies of MMR on chromatin will be interesting because it appears to be independent of the ATP-dependent remodelers that DSB and NER utilize to access their target sites in chromatin.

170

References

1. Lindahl, T., Instability and decay of the primary structure of DNA. Nature, 1993. 362(6422): p. 709-15. 2. Friedberg, E.C., et al., DNA repair: from molecular mechanism to human disease. DNA Repair (Amst), 2006. 5(8): p. 986-96. 3. Sharova, N.P., How does a cell repair damaged DNA? Biochemistry (Mosc), 2005. 70(3): p. 275-91. 4. Ataian, Y. and J.E. Krebs, Five repair pathways in one context: chromatin modification during DNA repair. Biochem Cell Biol, 2006. 84(4): p. 490-504. 5. Sancar, A., et al., Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem, 2004. 73: p. 39-85. 6. Thoma, F., Light and dark in chromatin repair: repair of UV-induced DNA lesions by photolyase and nucleotide excision repair. Embo J, 1999. 18(23): p. 6585-98. 7. Daya-Grosjean, L. and A. Sarasin, The role of UV induced lesions in skin carcinogenesis: an overview of oncogene and tumor suppressor gene modifications in xeroderma pigmentosum skin tumors. Mutat Res, 2005. 571(1-2): p. 43-56. 8. de Boer, J. and J.H. Hoeijmakers, Nucleotide excision repair and human syndromes. Carcinogenesis, 2000. 21(3): p. 453-60. 9. Friedberg, E.C., How nucleotide excision repair protects against cancer. Nat Rev Cancer, 2001. 1(1): p. 22-33. 10. van Steeg, H., L.H. Mullenders, and J. Vijg, Mutagenesis and carcinogenesis in nucleotide excision repair-deficient XPA knock out mice. Mutat Res, 2000. 450(1- 2): p. 167-80. 11. Hanawalt, P.C., Subpathways of nucleotide excision repair and their regulation. Oncogene, 2002. 21(58): p. 8949-56. 12. Hoeijmakers, J.H., Genome maintenance mechanisms for preventing cancer. Nature, 2001. 411(6835): p. 366-74. 13. Hanawalt, P.C., et al., DNA repair in bacteria and mammalian cells. Annu Rev Biochem, 1979. 48: p. 783-836. 14. Fortini, P., et al., The base excision repair: mechanisms and its relevance for cancer susceptibility. Biochimie, 2003. 85(11): p. 1053-71. 15. Hitomi, K., S. Iwai, and J.A. Tainer, The intricate structural chemistry of base excision repair machinery: implications for DNA damage recognition, removal, and repair. DNA Repair (Amst), 2007. 6(4): p. 410-28.

171

16. Lindahl, T., P. Karran, and R.D. Wood, DNA excision repair pathways. Curr Opin Genet Dev, 1997. 7(2): p. 158-69. 17. Iyer, R.R., et al., DNA mismatch repair: functions and mechanisms. Chem Rev, 2006. 106(2): p. 302-23. 18. Kolodner, R.D. and G.T. Marsischky, Eukaryotic DNA mismatch repair. Curr Opin Genet Dev, 1999. 9(1): p. 89-96. 19. Hsieh, P. and K. Yamane, DNA mismatch repair: molecular mechanism, cancer, and ageing. Mech Ageing Dev, 2008. 129(7-8): p. 391-407. 20. Gradia, S., et al., hMSH2-hMSH6 forms a hydrolysis-independent sliding clamp on mismatched DNA. Mol Cell, 1999. 3(2): p. 255-61. 21. Acharya, S., et al., The coordinated functions of the E. coli MutS and MutL proteins in mismatch repair. Mol Cell, 2003. 12(1): p. 233-46. 22. Li, G.M., Mechanisms and functions of DNA mismatch repair. Cell Res, 2008. 18(1): p. 85-98. 23. Fink, D., S. Aebi, and S.B. Howell, The role of DNA mismatch repair in drug resistance. Clin Cancer Res, 1998. 4(1): p. 1-6. 24. Hampel, H., et al., Feasibility of screening for Lynch syndrome among patients with colorectal cancer. J Clin Oncol, 2008. 26(35): p. 5783-8. 25. Lynch, H.T., et al., Review of the Lynch syndrome: history, molecular genetics, screening, differential diagnosis, and medicolegal ramifications. Clin Genet, 2009. 76(1): p. 1-18. 26. Boland, C.R., Evolution of the nomenclature for the hereditary colorectal cancer syndromes. Fam Cancer, 2005. 4(3): p. 211-8. 27. Silva, F.C., et al., Mismatch repair genes in Lynch syndrome: a review. Sao Paulo Med J, 2009. 127(1): p. 46-51. 28. Lynch, H.T., et al., Phenotypic and genotypic heterogeneity in the Lynch syndrome: diagnostic, surveillance and management implications. Eur J Hum Genet, 2006. 14(4): p. 390-402. 29. De Jong, A.E., et al., The role of mismatch repair gene defects in the development of adenomas in patients with HNPCC. Gastroenterology, 2004. 126(1): p. 42-8. 30. de la Chapelle, A., Genetic predisposition to colorectal cancer. Nat Rev Cancer, 2004. 4(10): p. 769-80. 31. Vasen, H.F. and J. Wijnen, Clinical implications of genetic testing of hereditary nonpolyposis colorectal cancer. Cytogenet Cell Genet, 1999. 86(2): p. 136-9. 32. van Lier, M.G., et al., A review on the molecular diagnostics of Lynch syndrome: a central role for the pathology laboratory. J Cell Mol Med. 14(1-2): p. 181-97. 33. de la Chapelle, A., Microsatellite instability. N Engl J Med, 2003. 349(3): p. 209- 10. 34. Miyaki, M., et al., Germline mutation of MSH6 as the cause of hereditary nonpolyposis colorectal cancer. Nat Genet, 1997. 17(3): p. 271-2. 35. Rustgi, A.K., The genetics of hereditary colon cancer. Genes Dev, 2007. 21(20): p. 2525-38.

172

36. Kane, M.F., et al., Methylation of the hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res, 1997. 57(5): p. 808-11. 37. Herman, J.G. and S.B. Baylin, Gene silencing in cancer in association with promoter hypermethylation. N Engl J Med, 2003. 349(21): p. 2042-54. 38. Herman, J.G., et al., Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci U S A, 1998. 95(12): p. 6870-5. 39. Ligtenberg, M.J., et al., Heritable somatic methylation and inactivation of MSH2 in families with Lynch syndrome due to deletion of the 3' exons of TACSTD1. Nat Genet, 2009. 41(1): p. 112-7. 40. Edelmann, L. and W. Edelmann, Loss of DNA mismatch repair function and cancer predisposition in the mouse: animal models for human hereditary nonpolyposis colorectal cancer. Am J Med Genet C Semin Med Genet, 2004. 129C(1): p. 91-9. 41. Drake, J.W., A constant rate of spontaneous mutation in DNA-based microbes. Proc Natl Acad Sci U S A, 1991. 88(16): p. 7160-4. 42. Drake, J.W., The distribution of rates of spontaneous mutation over viruses, prokaryotes, and eukaryotes. Ann N Y Acad Sci, 1999. 870: p. 100-7. 43. Barnes, D.E. and T. Lindahl, Repair and genetic consequences of endogenous DNA base damage in mammalian cells. Annu Rev Genet, 2004. 38: p. 445-76. 44. Loeb, L.A., A mutator phenotype in cancer. Cancer Res, 2001. 61(8): p. 3230-9. 45. Fishel, R. and R.D. Kolodner, Identification of mismatch repair genes and their role in the development of cancer. Curr Opin Genet Dev, 1995. 5(3): p. 382-95. 46. Heinen, C.D., Genotype to phenotype: analyzing the effects of inherited mutations in colorectal cancer families. Mutat Res. 693(1-2): p. 32-45. 47. Wagner, R., Jr. and M. Meselson, Repair tracts in mismatched DNA heteroduplexes. 1976. DNA Repair (Amst), 2005. 4(1): p. 126-30; discussion 103-4, 131. 48. Wildenberg, J. and M. Meselson, Mismatch repair in heteroduplex DNA. Proc Natl Acad Sci U S A, 1975. 72(6): p. 2202-6. 49. Dohet, C., R. Wagner, and M. Radman, Repair of defined single base-pair mismatches in Escherichia coli. Proc Natl Acad Sci U S A, 1985. 82(2): p. 503-5. 50. Kramer, B., W. Kramer, and H.J. Fritz, Different base/base mismatches are corrected with different efficiencies by the methyl-directed DNA mismatch-repair system of E. coli. Cell, 1984. 38(3): p. 879-87. 51. Su, S.S., et al., Mispair specificity of methyl-directed DNA mismatch correction in vitro. J Biol Chem, 1988. 263(14): p. 6829-35. 52. Dohet, C., et al., Large non-homology in heteroduplex DNA is processed differently than single base pair mismatches. Mol Gen Genet, 1987. 206(1): p. 181-4. 53. Parker, B.O. and M.G. Marinus, Repair of DNA heteroduplexes containing small heterologous sequences in Escherichia coli. Proc Natl Acad Sci U S A, 1992. 89(5): p. 1730-4. 173

54. Burdett, V., et al., In vivo requirement for RecJ, ExoVII, ExoI, and ExoX in methyl-directed mismatch repair. Proc Natl Acad Sci U S A, 2001. 98(12): p. 6765-70. 55. Lahue, R.S., K.G. Au, and P. Modrich, DNA mismatch correction in a defined system. Science, 1989. 245(4914): p. 160-4. 56. Lu, A.L., S. Clark, and P. Modrich, Methyl-directed repair of DNA base-pair mismatches in vitro. Proc Natl Acad Sci U S A, 1983. 80(15): p. 4639-43. 57. Pukkila, P.J., et al., Effects of high levels of DNA adenine methylation on methyl- directed mismatch repair in Escherichia coli. Genetics, 1983. 104(4): p. 571-82. 58. Constantin, N., et al., Human mismatch repair: reconstitution of a nick-directed bidirectional reaction. J Biol Chem, 2005. 280(48): p. 39752-61. 59. Dzantiev, L., et al., A defined human system that supports bidirectional mismatch- provoked excision. Mol Cell, 2004. 15(1): p. 31-41. 60. Genschel, J. and P. Modrich, Mechanism of 5'-directed excision in human mismatch repair. Mol Cell, 2003. 12(5): p. 1077-86. 61. Genschel, J. and P. Modrich, Functions of MutLalpha, replication protein A (RPA), and HMGB1 in 5'-directed mismatch repair. J Biol Chem, 2009. 284(32): p. 21536-44. 62. Zhang, Y., et al., Reconstitution of 5'-directed human mismatch repair in a purified system. Cell, 2005. 122(5): p. 693-705. 63. Kunkel, T.A. and D.A. Erie, DNA mismatch repair. Annu Rev Biochem, 2005. 74: p. 681-710. 64. Geier, G.E. and P. Modrich, Recognition sequence of the dam methylase of Escherichia coli K12 and mode of cleavage of Dpn I endonuclease. J Biol Chem, 1979. 254(4): p. 1408-13. 65. Lyons, S.M. and P.F. Schendel, Kinetics of methylation in Escherichia coli K-12. J Bacteriol, 1984. 159(1): p. 421-3. 66. Marinus, M.G., Adenine methylation of Okazaki fragments in Escherichia coli. J Bacteriol, 1976. 128(3): p. 853-4. 67. Modrich, P. and R. Lahue, Mismatch repair in replication fidelity, genetic recombination, and cancer biology. Annu Rev Biochem, 1996. 65: p. 101-33. 68. Biswas, I. and P. Hsieh, Interaction of MutS protein with the major and minor grooves of a heteroduplex DNA. J Biol Chem, 1997. 272(20): p. 13355-64. 69. Lamers, M.H., et al., The crystal structure of DNA mismatch repair protein MutS binding to a G x T mismatch. Nature, 2000. 407(6805): p. 711-7. 70. Obmolova, G., et al., Crystal structures of mismatch repair protein MutS and its complex with a substrate DNA. Nature, 2000. 407(6805): p. 703-10. 71. Haber, L.T. and G.C. Walker, Altering the conserved nucleotide binding motif in the Salmonella typhimurium MutS mismatch repair protein affects both its ATPase and mismatch binding activities. Embo J, 1991. 10(9): p. 2707-15. 72. Gradia, S., S. Acharya, and R. Fishel, The human mismatch recognition complex hMSH2-hMSH6 functions as a novel molecular switch. Cell, 1997. 91(7): p. 995- 1005.

174

73. Galio, L., C. Bouquet, and P. Brooks, ATP hydrolysis-dependent formation of a dynamic ternary nucleoprotein complex with MutS and MutL. Nucleic Acids Res, 1999. 27(11): p. 2325-31. 74. Grilley, M., et al., Isolation and characterization of the Escherichia coli mutL gene product. J Biol Chem, 1989. 264(2): p. 1000-4. 75. Spampinato, C. and P. Modrich, The MutL ATPase is required for mismatch repair. J Biol Chem, 2000. 275(13): p. 9863-9. 76. Schofield, M.J., et al., Interaction of Escherichia coli MutS and MutL at a DNA mismatch. J Biol Chem, 2001. 276(30): p. 28291-9. 77. Pluciennik, A. and P. Modrich, Protein roadblocks and helix discontinuities are barriers to the initiation of mismatch repair. Proc Natl Acad Sci U S A, 2007. 104(31): p. 12709-13. 78. Kadyrov, F.A., et al., Endonucleolytic function of MutLalpha in human mismatch repair. Cell, 2006. 126(2): p. 297-308. 79. Ban, C. and W. Yang, Crystal structure and ATPase activity of MutL: implications for DNA repair and mutagenesis. Cell, 1998. 95(4): p. 541-52. 80. Aronshtam, A. and M.G. Marinus, Dominant negative mutator mutations in the mutL gene of Escherichia coli. Nucleic Acids Res, 1996. 24(13): p. 2498-504. 81. Li, F., et al., Escherichia coli mismatch repair protein MutL interacts with the clamp loader subunits of DNA polymerase III. Mutat Res, 2008. 637(1-2): p. 101- 10. 82. Lopez de Saro, F.J., et al., The beta sliding clamp binds to multiple sites within MutL and MutS. J Biol Chem, 2006. 281(20): p. 14340-9. 83. Au, K.G., K. Welsh, and P. Modrich, Initiation of methyl-directed mismatch repair. J Biol Chem, 1992. 267(17): p. 12142-8. 84. Welsh, K.M., et al., Isolation and characterization of the Escherichia coli mutH gene product. J Biol Chem, 1987. 262(32): p. 15624-9. 85. Lahue, R.S., S.S. Su, and P. Modrich, Requirement for d(GATC) sequences in Escherichia coli mutHLS mismatch correction. Proc Natl Acad Sci U S A, 1987. 84(6): p. 1482-6. 86. Grilley, M., J. Griffith, and P. Modrich, Bidirectional excision in methyl-directed mismatch repair. J Biol Chem, 1993. 268(16): p. 11830-7. 87. Su, S.S., et al., Gap formation is associated with methyl-directed mismatch correction under conditions of restricted DNA synthesis. Genome, 1989. 31(1): p. 104-11. 88. Bruni, R., D. Martin, and J. Jiricny, d(GATC) sequences influence Escherichia coli mismatch repair in a distance-dependent manner from positions both upstream and downstream of the mismatch. Nucleic Acids Res, 1988. 16(11): p. 4875-90. 89. Lu, A.L., Influence of GATC sequences on Escherichia coli DNA mismatch repair in vitro. J Bacteriol, 1987. 169(3): p. 1254-9. 90. Modrich, P., Mechanisms in eukaryotic mismatch repair. J Biol Chem, 2006. 281(41): p. 30305-9.

175

91. Dao, V. and P. Modrich, Mismatch-, MutS-, MutL-, and helicase II-dependent unwinding from the single-strand break of an incised heteroduplex. J Biol Chem, 1998. 273(15): p. 9202-7. 92. Ramilo, C., et al., Partial reconstitution of human DNA mismatch repair in vitro: characterization of the role of human replication protein A. Mol Cell Biol, 2002. 22(7): p. 2037-46. 93. Cooper, D.L., R.S. Lahue, and P. Modrich, Methyl-directed mismatch repair is bidirectional. J Biol Chem, 1993. 268(16): p. 11823-9. 94. Viswanathan, M., et al., Redundant exonuclease involvement in Escherichia coli methyl-directed mismatch repair. J Biol Chem, 2001. 276(33): p. 31053-8. 95. Pavlov, Y.I., I.M. Mian, and T.A. Kunkel, Evidence for preferential mismatch repair of lagging strand DNA replication errors in yeast. Curr Biol, 2003. 13(9): p. 744-8. 96. Drummond, J.T., et al., Isolation of an hMSH2-p160 heterodimer that restores DNA mismatch repair to tumor cells. Science, 1995. 268(5219): p. 1909-12. 97. Fishel, R., et al., The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell, 1994. 77(1): p. 1 p following 166. 98. Leach, F.S., et al., Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell, 1993. 75(6): p. 1215-25. 99. Palombo, F., et al., GTBP, a 160-kilodalton protein essential for mismatch- binding activity in human cells. Science, 1995. 268(5219): p. 1912-4. 100. Bronner, C.E., et al., Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature, 1994. 368(6468): p. 258-61. 101. Li, G.M. and P. Modrich, Restoration of mismatch repair to nuclear extracts of H6 colorectal tumor cells by a heterodimer of human MutL homologs. Proc Natl Acad Sci U S A, 1995. 92(6): p. 1950-4. 102. Schmutte, C., et al., Human exonuclease I interacts with the mismatch repair protein hMSH2. Cancer Res, 1998. 58(20): p. 4537-42. 103. Tishkoff, D.X., et al., Identification of a human gene encoding a homologue of Saccharomyces cerevisiae EXO1, an exonuclease implicated in mismatch repair and recombination. Cancer Res, 1998. 58(22): p. 5027-31. 104. Tishkoff, D.X., et al., Identification and characterization of Saccharomyces cerevisiae EXO1, a gene encoding an exonuclease that interacts with MSH2. Proc Natl Acad Sci U S A, 1997. 94(14): p. 7487-92. 105. Lin, Y.L., et al., The evolutionarily conserved zinc finger motif in the largest subunit of human replication protein A is required for DNA replication and mismatch repair but not for nucleotide excision repair. J Biol Chem, 1998. 273(3): p. 1453-61. 106. Gu, L., et al., ATP-dependent interaction of human mismatch repair proteins and dual role of PCNA in mismatch repair. Nucleic Acids Res, 1998. 26(5): p. 1173- 8.

176

107. Johnson, R.E., et al., Evidence for involvement of yeast proliferating cell nuclear antigen in DNA mismatch repair. J Biol Chem, 1996. 271(45): p. 27987-90. 108. Umar, A., et al., Requirement for PCNA in DNA mismatch repair at a step preceding DNA resynthesis. Cell, 1996. 87(1): p. 65-73. 109. Longley, M.J., A.J. Pierce, and P. Modrich, DNA polymerase delta is required for human mismatch repair in vitro. J Biol Chem, 1997. 272(16): p. 10917-21. 110. Reenan, R.A. and R.D. Kolodner, Characterization of insertion mutations in the Saccharomyces cerevisiae MSH1 and MSH2 genes: evidence for separate mitochondrial and nuclear functions. Genetics, 1992. 132(4): p. 975-85. 111. Drummond, J.T., et al., DHFR/MSH3 amplification in methotrexate-resistant cells alters the hMutSalpha/hMutSbeta ratio and reduces the efficiency of base-base mismatch repair. Proc Natl Acad Sci U S A, 1997. 94(19): p. 10144-9. 112. Bocker, T., et al., hMSH5: a human MutS homologue that forms a novel heterodimer with hMSH4 and is expressed during spermatogenesis. Cancer Res, 1999. 59(4): p. 816-22. 113. Kneitz, B., et al., MutS homolog 4 localization to meiotic chromosomes is required for chromosome pairing during meiosis in male and female mice. Genes Dev, 2000. 14(9): p. 1085-97. 114. Snowden, T., et al., hMSH4-hMSH5 recognizes Holliday Junctions and forms a meiosis-specific sliding clamp that embraces homologous chromosomes. Mol Cell, 2004. 15(3): p. 437-51. 115. Snowden, T., et al., hMSH4-hMSH5 adenosine nucleotide processing and interactions with homologous recombination machinery. J Biol Chem, 2008. 283(1): p. 145-54. 116. Warren, J.J., et al., Structure of the human MutSalpha DNA lesion recognition complex. Mol Cell, 2007. 26(4): p. 579-92. 117. Locher, K.P., Structure and mechanism of ABC transporters. Curr Opin Struct Biol, 2004. 14(4): p. 426-31. 118. Mazur, D.J., M.L. Mendillo, and R.D. Kolodner, Inhibition of Msh6 ATPase activity by mispaired DNA induces a Msh2(ATP)-Msh6(ATP) state capable of hydrolysis-independent movement along DNA. Mol Cell, 2006. 22(1): p. 39-49. 119. Stec, I., et al., The PWWP domain: a potential protein-protein interaction domain in nuclear proteins influencing differentiation? FEBS Lett, 2000. 473(1): p. 1-5. 120. Lipkin, S.M., et al., MLH3: a DNA mismatch repair gene associated with mammalian microsatellite instability. Nat Genet, 2000. 24(1): p. 27-35. 121. Nicolaides, N.C., et al., Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature, 1994. 371(6492): p. 75-80. 122. Papadopoulos, N., et al., Mutation of a mutL homolog in hereditary colon cancer. Science, 1994. 263(5153): p. 1625-9. 123. Shell, S.S., C.D. Putnam, and R.D. Kolodner, The N terminus of Saccharomyces cerevisiae Msh6 is an unstructured tether to PCNA. Mol Cell, 2007. 26(4): p. 565-78.

177

124. Flores-Rozas, H., D. Clark, and R.D. Kolodner, Proliferating cell nuclear antigen and Msh2p-Msh6p interact to form an active mispair recognition complex. Nat Genet, 2000. 26(3): p. 375-8. 125. Lau, P.J. and R.D. Kolodner, Transfer of the MSH2.MSH6 complex from proliferating cell nuclear antigen to mispaired bases in DNA. J Biol Chem, 2003. 278(1): p. 14-7. 126. Kleczkowska, H.E., et al., hMSH3 and hMSH6 interact with PCNA and colocalize with it to replication foci. Genes Dev, 2001. 15(6): p. 724-36. 127. Guo, S., et al., Differential requirement for proliferating cell nuclear antigen in 5' and 3' nick-directed excision in human mismatch repair. J Biol Chem, 2004. 279(17): p. 16912-7. 128. Wei, K., et al., Inactivation of Exonuclease 1 in mice results in DNA mismatch repair defects, increased cancer susceptibility, and male and female sterility. Genes Dev, 2003. 17(5): p. 603-14. 129. Guo, S., et al., Regulation of replication protein A functions in DNA mismatch repair by phosphorylation. J Biol Chem, 2006. 281(31): p. 21607-16. 130. Yuan, F., et al., Evidence for involvement of HMGB1 protein in human DNA mismatch repair. J Biol Chem, 2004. 279(20): p. 20935-40. 131. Genschel, J., L.R. Bazemore, and P. Modrich, Human exonuclease I is required for 5' and 3' mismatch repair. J Biol Chem, 2002. 277(15): p. 13302-11. 132. Kolodner, R.D., M.L. Mendillo, and C.D. Putnam, Coupling distant sites in DNA during DNA mismatch repair. Proc Natl Acad Sci U S A, 2007. 104(32): p. 12953-4. 133. Wang, H. and J.B. Hays, Mismatch repair in human nuclear extracts: effects of internal DNA-hairpin structures between mismatches and excision-initiation nicks on mismatch correction and mismatch-provoked excision. J Biol Chem, 2003. 278(31): p. 28686-93. 134. Wang, H. and J.B. Hays, Signaling from DNA mispairs to mismatch-repair excision sites despite intervening blockades. Embo J, 2004. 23(10): p. 2126-33. 135. Allen, D.J., et al., MutS mediates heteroduplex loop formation by a translocation mechanism. Embo J, 1997. 16(14): p. 4467-76. 136. Fishel, R., Mismatch repair, molecular switches, and signal transduction. Genes Dev, 1998. 12(14): p. 2096-101. 137. Blackwell, L.J., et al., Nucleotide-promoted release of hMutSalpha from heteroduplex DNA is consistent with an ATP-dependent translocation mechanism. J Biol Chem, 1998. 273(48): p. 32055-62. 138. Mendillo, M.L., D.J. Mazur, and R.D. Kolodner, Analysis of the interaction between the Saccharomyces cerevisiae MSH2-MSH6 and MLH1-PMS1 complexes with DNA using a reversible DNA end-blocking system. J Biol Chem, 2005. 280(23): p. 22245-57. 139. Jiricny, J., MutLalpha: at the cutting edge of mismatch repair. Cell, 2006. 126(2): p. 239-41. 140. Jiricny, J., The multifaceted mismatch-repair system. Nat Rev Mol Cell Biol, 2006. 7(5): p. 335-46. 178

141. Karran, P., Mechanisms of tolerance to DNA damaging therapeutic drugs. Carcinogenesis, 2001. 22(12): p. 1931-7. 142. Duckett, D.R., et al., Human MutSalpha recognizes damaged DNA base pairs containing O6-methylguanine, O4-methylthymine, or the cisplatin-d(GpG) adduct. Proc Natl Acad Sci U S A, 1996. 93(13): p. 6443-7. 143. Gong, J.G., et al., The tyrosine kinase c-Abl regulates p73 in apoptotic response to cisplatin-induced DNA damage. Nature, 1999. 399(6738): p. 806-9. 144. Stojic, L., R. Brun, and J. Jiricny, Mismatch repair and DNA damage signalling. DNA Repair (Amst), 2004. 3(8-9): p. 1091-101. 145. Yan, T., et al., DNA mismatch repair (MMR) mediates 6-thioguanine genotoxicity by introducing single-strand breaks to signal a G2-M arrest in MMR-proficient RKO cells. Clin Cancer Res, 2003. 9(6): p. 2327-34. 146. York, S.J. and P. Modrich, Mismatch repair-dependent iterative excision at irreparable O6-methylguanine lesions in human nuclear extracts. J Biol Chem, 2006. 281(32): p. 22674-83. 147. Yoshioka, K., Y. Yoshioka, and P. Hsieh, ATR kinase activation mediated by MutSalpha and MutLalpha in response to cytotoxic O6-methylguanine adducts. Mol Cell, 2006. 22(4): p. 501-10. 148. Escargueil, A.E., et al., What histone code for DNA repair? Mutat Res, 2008. 658(3): p. 259-70. 149. Smerdon, M.J. and M.W. Lieberman, Nucleosome rearrangement in human chromatin during UV-induced DNA- reapir synthesis. Proc Natl Acad Sci U S A, 1978. 75(9): p. 4238-41. 150. Polo, S.E. and G. Almouzni, [Chromatin dynamics during the repair of DNA lesions]. Med Sci (Paris), 2007. 23(1): p. 29-31. 151. Pandita, T.K. and C. Richardson, Chromatin remodeling finds its place in the DNA double-strand break response. Nucleic Acids Res, 2009. 37(5): p. 1363-77. 152. Fillingham, J., M.C. Keogh, and N.J. Krogan, GammaH2AX and its role in DNA double-strand break repair. Biochem Cell Biol, 2006. 84(4): p. 568-77. 153. Keogh, M.C., et al., A phosphatase complex that dephosphorylates gammaH2AX regulates DNA damage checkpoint recovery. Nature, 2006. 439(7075): p. 497- 501. 154. Wang, G.G., C.D. Allis, and P. Chi, Chromatin remodeling and cancer, Part I: Covalent histone modifications. Trends Mol Med, 2007. 13(9): p. 363-72. 155. Martin, C. and Y. Zhang, The diverse functions of histone lysine methylation. Nat Rev Mol Cell Biol, 2005. 6(11): p. 838-49. 156. Peterson, C.L. and M.A. Laniel, Histones and histone modifications. Curr Biol, 2004. 14(14): p. R546-51. 157. Johnson, C.N., N.L. Adkins, and P. Georgel, Chromatin remodeling complexes: ATP-dependent machines in action. Biochem Cell Biol, 2005. 83(4): p. 405-17. 158. Marino-Ramirez, L., et al., Histone structure and nucleosome stability. Expert Rev Proteomics, 2005. 2(5): p. 719-29. 159. Luger, K., Structure and dynamic behavior of nucleosomes. Curr Opin Genet Dev, 2003. 13(2): p. 127-35. 179

160. Luger, K., et al., Characterization of nucleosome core particles containing histone proteins made in bacteria. J Mol Biol, 1997. 272(3): p. 301-11. 161. Luger, K., et al., Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature, 1997. 389(6648): p. 251-60. 162. Widom, J., Role of DNA sequence in nucleosome stability and dynamics. Q Rev Biophys, 2001. 34(3): p. 269-324. 163. Chakravarthy, S., et al., Structure and dynamic properties of nucleosome core particles. FEBS Lett, 2005. 579(4): p. 895-8. 164. Hall, M.A., et al., High-resolution dynamic mapping of histone-DNA interactions in a nucleosome. Nat Struct Mol Biol, 2009. 16(2): p. 124-9. 165. Shundrovsky, A., et al., Probing SWI/SNF remodeling of the nucleosome by unzipping single DNA molecules. Nat Struct Mol Biol, 2006. 13(6): p. 549-54. 166. Bowman, G.D., Mechanisms of ATP-dependent nucleosome sliding. Curr Opin Struct Biol, 2010. 20(1): p. 73-81. 167. Pham, T.A., et al., Modulation of progesterone receptor binding to progesterone response elements by positioned nucleosomes. Biochemistry, 1992. 31(5): p. 1570-8. 168. Gottesfeld, J.M., DNA sequence-directed nucleosome reconstitution on 5S RNA genes of Xenopus laevis. Mol Cell Biol, 1987. 7(5): p. 1612-22. 169. Hayes, J.J. and A.P. Wolffe, Preferential and asymmetric interaction of linker histones with 5S DNA in the nucleosome. Proc Natl Acad Sci U S A, 1993. 90(14): p. 6415-9. 170. Hayes, J.J., T.D. Tullius, and A.P. Wolffe, The structure of DNA in a nucleosome. Proc Natl Acad Sci U S A, 1990. 87(19): p. 7405-9. 171. Jiang, C. and B.F. Pugh, Nucleosome positioning and gene regulation: advances through genomics. Nat Rev Genet, 2009. 10(3): p. 161-72. 172. Talbert, P.B. and S. Henikoff, Histone variants--ancient wrap artists of the epigenome. Nat Rev Mol Cell Biol. 11(4): p. 264-75. 173. Meneghini, M.D., M. Wu, and H.D. Madhani, Conserved histone variant H2A.Z protects euchromatin from the ectopic spread of silent heterochromatin. Cell, 2003. 112(5): p. 725-36. 174. Chadwick, B.P. and H.F. Willard, A novel chromatin protein, distantly related to histone H2A, is largely excluded from the inactive X chromosome. J Cell Biol, 2001. 152(2): p. 375-84. 175. Costanzi, C., et al., Histone macroH2A1 is concentrated in the inactive X chromosome of female preimplantation mouse embryos. Development, 2000. 127(11): p. 2283-9. 176. Sarma, K. and D. Reinberg, Histone variants meet their match. Nat Rev Mol Cell Biol, 2005. 6(2): p. 139-49. 177. Ahmad, K. and S. Henikoff, The histone variant H3.3 marks active chromatin by replication-independent nucleosome assembly. Mol Cell, 2002. 9(6): p. 1191-200. 178. Kouzarides, T., Chromatin modifications and their function. Cell, 2007. 128(4): p. 693-705.

180

179. Ramanathan, B. and M.J. Smerdon, Changes in nuclear protein acetylation in u.v.-damaged human cells. Carcinogenesis, 1986. 7(7): p. 1087-94. 180. Ramanathan, B. and M.J. Smerdon, Enhanced DNA repair synthesis in hyperacetylated nucleosomes. J Biol Chem, 1989. 264(19): p. 11026-34. 181. Ozdemir, A., et al., Histone H3 lysine 56 acetylation: a new twist in the chromosome cycle. Cell Cycle, 2006. 5(22): p. 2602-8. 182. Celic, I., et al., The sirtuins hst3 and Hst4p preserve genome integrity by controlling histone h3 lysine 56 deacetylation. Curr Biol, 2006. 16(13): p. 1280-9. 183. Chen, C.C., et al., Acetylated lysine 56 on histone H3 drives chromatin assembly after repair and signals for the completion of repair. Cell, 2008. 134(2): p. 231- 43. 184. Li, Q., et al., Acetylation of histone H3 lysine 56 regulates replication-coupled nucleosome assembly. Cell, 2008. 134(2): p. 244-55. 185. Zhu, Q. and A.A. Wani, Histone modifications: crucial elements for damage response and chromatin restoration. J Cell Physiol, 2010. 223(2): p. 283-8. 186. Das, C., et al., CBP/p300-mediated acetylation of histone H3 on lysine 56. Nature, 2009. 459(7243): p. 113-7. 187. Tjeertes, J.V., K.M. Miller, and S.P. Jackson, Screen for DNA-damage-responsive histone modifications identifies H3K9Ac and H3K56Ac in human cells. Embo J, 2009. 28(13): p. 1878-89. 188. Rogakou, E.P., et al., DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J Biol Chem, 1998. 273(10): p. 5858-68. 189. Rogakou, E.P., et al., Megabase chromatin domains involved in DNA double- strand breaks in vivo. J Cell Biol, 1999. 146(5): p. 905-16. 190. Shroff, R., et al., Distribution and dynamics of chromatin modification induced by a defined DNA double-strand break. Curr Biol, 2004. 14(19): p. 1703-11. 191. Ward, I.M. and J. Chen, Histone H2AX is phosphorylated in an ATR-dependent manner in response to replicational stress. J Biol Chem, 2001. 276(51): p. 47759- 62. 192. Dinant, C., A.B. Houtsmuller, and W. Vermeulen, Chromatin structure and DNA damage repair. Epigenetics Chromatin, 2008. 1(1): p. 9. 193. Wang, H., et al., Histone H3 and H4 ubiquitylation by the CUL4-DDB-ROC1 ubiquitin ligase facilitates cellular response to DNA damage. Mol Cell, 2006. 22(3): p. 383-94. 194. Ikura, T., et al., DNA damage-dependent acetylation and ubiquitination of H2AX enhances chromatin dynamics. Mol Cell Biol, 2007. 27(20): p. 7028-40. 195. Bostelman, L.J., et al., Methylation of histone H3 lysine-79 by Dot1p plays multiple roles in the response to UV damage in Saccharomyces cerevisiae. DNA Repair (Amst), 2007. 6(3): p. 383-95. 196. Sanders, S.L., et al., Methylation of histone H4 lysine 20 controls recruitment of Crb2 to sites of DNA damage. Cell, 2004. 119(5): p. 603-14. 197. Clapier, C.R. and B.R. Cairns, The biology of chromatin remodeling complexes. Annu Rev Biochem, 2009. 78: p. 273-304.

181

198. Singleton, M.R. and D.B. Wigley, Modularity and specialization in superfamily 1 and 2 helicases. J Bacteriol, 2002. 184(7): p. 1819-26. 199. Gavin, I., P.J. Horn, and C.L. Peterson, SWI/SNF chromatin remodeling requires changes in DNA topology. Mol Cell, 2001. 7(1): p. 97-104. 200. Havas, K., et al., Generation of superhelical torsion by ATP-dependent chromatin remodeling activities. Cell, 2000. 103(7): p. 1133-42. 201. Saha, A., J. Wittmeyer, and B.R. Cairns, Chromatin remodeling by RSC involves ATP-dependent DNA translocation. Genes Dev, 2002. 16(16): p. 2120-34. 202. Neigeborn, L. and M. Carlson, Genes affecting the regulation of SUC2 gene expression by glucose repression in Saccharomyces cerevisiae. Genetics, 1984. 108(4): p. 845-58. 203. Stern, M., R. Jensen, and I. Herskowitz, Five SWI genes are required for expression of the HO gene in yeast. J Mol Biol, 1984. 178(4): p. 853-68. 204. Cairns, B.R., et al., RSC, an essential, abundant chromatin-remodeling complex. Cell, 1996. 87(7): p. 1249-60. 205. Du, J., et al., Sth1p, a Saccharomyces cerevisiae Snf2p/Swi2p homolog, is an essential ATPase in RSC and differs from Snf/Swi in its interactions with histones and chromatin-associated proteins. Genetics, 1998. 150(3): p. 987-1005. 206. Gangaraju, V.K. and B. Bartholomew, Mechanisms of ATP dependent chromatin remodeling. Mutat Res, 2007. 618(1-2): p. 3-17. 207. Fischer, C.J., A. Saha, and B.R. Cairns, Kinetic model for the ATP-dependent translocation of Saccharomyces cerevisiae RSC along double-stranded DNA. Biochemistry, 2007. 46(43): p. 12416-26. 208. Shim, E.Y., et al., The yeast chromatin remodeler RSC complex facilitates end joining repair of DNA double-strand breaks. Mol Cell Biol, 2005. 25(10): p. 3934-44. 209. Kassabov, S.R., et al., SWI/SNF unwraps, slides, and rewraps the nucleosome. Mol Cell, 2003. 11(2): p. 391-403. 210. Saha, A., J. Wittmeyer, and B.R. Cairns, Chromatin remodeling through directional DNA translocation from an internal nucleosomal site. Nat Struct Mol Biol, 2005. 12(9): p. 747-55. 211. Zofall, M., et al., Chromatin remodeling by ISW2 and SWI/SNF requires DNA translocation inside the nucleosome. Nat Struct Mol Biol, 2006. 13(4): p. 339-46. 212. Zhang, Y., et al., DNA translocation and loop formation mechanism of chromatin remodeling by SWI/SNF and RSC. Mol Cell, 2006. 24(4): p. 559-68. 213. Park, J.H., et al., Mammalian SWI/SNF complexes facilitate DNA double-strand break repair by promoting gamma-H2AX induction. Embo J, 2006. 25(17): p. 3986-97. 214. Chai, B., et al., Distinct roles for the RSC and Swi/Snf ATP-dependent chromatin remodelers in DNA double-strand break repair. Genes Dev, 2005. 19(14): p. 1656-61. 215. Shim, E.Y., et al., RSC mobilizes nucleosomes to improve accessibility of repair machinery to the damaged chromatin. Mol Cell Biol, 2007. 27(5): p. 1602-13.

182

216. Vary, J.C., Jr., et al., Yeast Isw1p forms two separable complexes in vivo. Mol Cell Biol, 2003. 23(1): p. 80-91. 217. Gelbart, M.E., et al., Interactions of Isw2 chromatin remodeling complex with nucleosomal arrays: analyses using recombinant yeast histones and immobilized templates. Mol Cell Biol, 2001. 21(6): p. 2098-106. 218. Ito, T., et al., ACF, an ISWI-containing and ATP-utilizing chromatin assembly and remodeling factor. Cell, 1997. 90(1): p. 145-55. 219. Strohner, R., et al., A 'loop recapture' mechanism for ACF-dependent nucleosome remodeling. Nat Struct Mol Biol, 2005. 12(8): p. 683-90. 220. Poot, R.A., et al., The Williams syndrome transcription factor interacts with PCNA to target chromatin remodelling by ISWI to replication foci. Nat Cell Biol, 2004. 6(12): p. 1236-44. 221. Flanagan, J.F., et al., Double chromodomains cooperate to recognize the methylated histone H3 tail. Nature, 2005. 438(7071): p. 1181-5. 222. Tran, H.G., et al., The chromo domain protein chd1p from budding yeast is an ATP-dependent chromatin-modifying factor. Embo J, 2000. 19(10): p. 2323-31. 223. Stockdale, C., et al., Analysis of nucleosome repositioning by yeast ISWI and Chd1 chromatin remodeling complexes. J Biol Chem, 2006. 281(24): p. 16279- 88. 224. Shen, X., et al., A chromatin remodelling complex involved in transcription and DNA processing. Nature, 2000. 406(6795): p. 541-4. 225. Mizuguchi, G., et al., ATP-driven exchange of histone H2AZ variant catalyzed by SWR1 chromatin remodeling complex. Science, 2004. 303(5656): p. 343-8. 226. Morrison, A.J., et al., INO80 and gamma-H2AX interaction links ATP-dependent chromatin remodeling to DNA damage repair. Cell, 2004. 119(6): p. 767-75. 227. van Attikum, H., et al., Recruitment of the INO80 complex by H2A phosphorylation links ATP-dependent chromatin remodeling with DNA double- strand break repair. Cell, 2004. 119(6): p. 777-88. 228. van Attikum, H., O. Fritsch, and S.M. Gasser, Distinct roles for SWR1 and INO80 chromatin remodeling complexes at chromosomal double-strand breaks. Embo J, 2007. 26(18): p. 4113-25. 229. Bao, Y. and X. Shen, INO80 subfamily of chromatin remodeling complexes. Mutat Res, 2007. 618(1-2): p. 18-29. 230. Tsukuda, T., et al., Chromatin remodelling at a DNA double-strand break site in Saccharomyces cerevisiae. Nature, 2005. 438(7066): p. 379-83. 231. Kalocsay, M., N.J. Hiller, and S. Jentsch, Chromosome-wide Rad51 spreading and SUMO-H2A.Z-dependent chromosome fixation in response to a persistent DNA double-strand break. Mol Cell, 2009. 33(3): p. 335-43. 232. Kobor, M.S., et al., A protein complex containing the conserved Swi2/Snf2-related ATPase Swr1p deposits histone variant H2A.Z into euchromatin. PLoS Biol, 2004. 2(5): p. E131. 233. Kulic, I.M. and H. Schiessel, Chromatin dynamics: nucleosomes go mobile through twist defects. Phys Rev Lett, 2003. 91(14): p. 148103.

183

234. van Holde, K. and T. Yager, Models for chromatin remodeling: a critical comparison. Biochem Cell Biol, 2003. 81(3): p. 169-72. 235. Yager, T.D. and K.E. van Holde, Dynamics and equilibria of nucleosomes at elevated ionic strength. J Biol Chem, 1984. 259(7): p. 4212-22. 236. Langst, G. and P.B. Becker, Nucleosome remodeling: one mechanism, many phenomena? Biochim Biophys Acta, 2004. 1677(1-3): p. 58-63. 237. Schiessel, H., et al., Polymer reptation and nucleosome repositioning. Phys Rev Lett, 2001. 86(19): p. 4414-7. 238. Schiessel, H., The nucleosome: a transparent, slippery, sticky and yet stable DNA-protein complex. Eur Phys J E Soft Matter, 2006. 19(3): p. 251-62. 239. Cairns, B.R., Chromatin remodeling: insights and intrigue from single-molecule studies. Nat Struct Mol Biol, 2007. 14(11): p. 989-96. 240. Aoyagi, S. and J.J. Hayes, hSWI/SNF-catalyzed nucleosome sliding does not occur solely via a twist-diffusion mechanism. Mol Cell Biol, 2002. 22(21): p. 7484-90. 241. Flaus, A., et al., Sin mutations alter inherent nucleosome mobility. Embo J, 2004. 23(2): p. 343-53. 242. Polach, K.J. and J. Widom, Mechanism of protein access to specific DNA sequences in chromatin: a dynamic equilibrium model for gene regulation. J Mol Biol, 1995. 254(2): p. 130-49. 243. Anderson, J.D. and J. Widom, Sequence and position-dependence of the equilibrium accessibility of nucleosomal DNA target sites. J Mol Biol, 2000. 296(4): p. 979-87. 244. Li, G. and J. Widom, Nucleosomes facilitate their own invasion. Nat Struct Mol Biol, 2004. 11(8): p. 763-9. 245. Perlmann, T., Glucocorticoid receptor DNA-binding specificity is increased by the organization of DNA in nucleosomes. Proc Natl Acad Sci U S A, 1992. 89(9): p. 3884-8. 246. Groth, A., et al., Chromatin challenges during DNA replication and repair. Cell, 2007. 128(4): p. 721-33. 247. Hawk, J.D., et al., Variation in efficiency of DNA mismatch repair at different sites in the yeast genome. Proc Natl Acad Sci U S A, 2005. 102(24): p. 8639-43. 248. Li, F., et al., Evidence that nucleosomes inhibit mismatch repair in eukaryotic cells. J Biol Chem, 2009. 284(48): p. 33056-61. 249. Gorman, J., et al., Visualizing one-dimensional diffusion of eukaryotic DNA repair factors along a chromatin lattice. Nat Struct Mol Biol, 2010. 17(8): p. 932- 8. 250. Fishel, R., et al., The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell, 1993. 75(5): p. 1027-38. 251. Prindle, M.J., E.J. Fox, and L.A. Loeb, The mutator phenotype in cancer: molecular mechanisms and targeting strategies. Curr Drug Targets, 2010. 11(10): p. 1296-303. 252. Prindle, M.J., E.J. Fox, and L.A. Loeb, The Mutator Phenotype in Cancer: Molecular Mechanisms and Targeting Strategies. Curr Drug Targets, 2010. 184

253. Boland, C.R. and R. Fishel, Lynch syndrome: form, function, proteins, and basketball. Gastroenterology, 2005. 129(2): p. 751-5. 254. Luger, K., T.J. Rechsteiner, and T.J. Richmond, Expression and purification of recombinant histones and nucleosome reconstitution. Methods Mol Biol, 1999. 119: p. 1-16. 255. Luger, K., T.J. Rechsteiner, and T.J. Richmond, Preparation of nucleosome core particle from recombinant histones. Methods Enzymol, 1999. 304: p. 3-19. 256. English, C.M., et al., Structural basis for the histone chaperone activity of Asf1. Cell, 2006. 127(3): p. 495-508. 257. Hyland, E.M., et al., Insights into the role of histone H3 and histone H4 core modifiable residues in Saccharomyces cerevisiae. Mol Cell Biol, 2005. 25(22): p. 10060-70. 258. Zhang, L., et al., Identification of novel histone post-translational modifications by peptide mass fingerprinting. Chromosoma, 2003. 112(2): p. 77-86. 259. Manohar, M., et al., Acetylation of histone H3 at the nucleosome dyad alters DNA-histone binding. J Biol Chem, 2009. 284(35): p. 23312-21. 260. Lowary, P.T. and J. Widom, New DNA sequence rules for high affinity binding to histone octamer and sequence-directed nucleosome positioning. J Mol Biol, 1998. 276(1): p. 19-42. 261. Selmane, T., et al., Formation of a DNA mismatch repair complex mediated by ATP. J Mol Biol, 2003. 334(5): p. 949-65. 262. Gradia, S., S. Acharya, and R. Fishel, The role of mismatched nucleotides in activating the hMSH2-hMSH6 molecular switch. J Biol Chem, 2000. 275(6): p. 3922-30. 263. Utley, R.T., et al., In vitro analysis of transcription factor binding to nucleosomes and nucleosome disruption/displacement. Methods Enzymol, 1996. 274: p. 276- 91. 264. Mazurek, A., et al., Sequence context effect for hMSH2-hMSH6 mismatch- dependent activation. Proc Natl Acad Sci U S A, 2009. 106(11): p. 4177-82. 265. Jackson, V., Deposition of newly synthesized histones: hybrid nucleosomes are not tandemly arranged on daughter DNA strands. Biochemistry, 1988. 27(6): p. 2109-20. 266. Sogo, J.M., et al., Structure of replicating simian virus 40 minichromosomes. The replication fork, core histone segregation and terminal structures. J Mol Biol, 1986. 189(1): p. 189-204. 267. Fang, W.H. and P. Modrich, Human strand-specific mismatch repair occurs by a bidirectional mechanism similar to that of the bacterial reaction. J Biol Chem, 1993. 268(16): p. 11838-44. 268. Thastrom, A., et al., Sequence motifs and free energies of selected natural and non-natural nucleosome positioning DNA sequences. J Mol Biol, 1999. 288(2): p. 213-29. 269. Muthurajan, U.M., et al., Crystal structures of histone Sin mutant nucleosomes reveal altered protein-DNA interactions. Embo J, 2004. 23(2): p. 260-71.

185

270. Kurumizaka, H. and A.P. Wolffe, Sin mutations of histone H3: influence on nucleosome core structure and function. Mol Cell Biol, 1997. 17(12): p. 6953-69. 271. Ura, K., et al., Histone acetylation: influence on transcription, nucleosome mobility and positioning, and linker histone-dependent transcriptional repression. Embo J, 1997. 16(8): p. 2096-107. 272. Kruger, W., et al., Amino acid substitutions in the structured domains of histones H3 and H4 partially relieve the requirement of the yeast SWI/SNF complex for transcription. Genes Dev, 1995. 9(22): p. 2770-9. 273. Dupaigne, P., et al., Rad51 polymerization reveals a new chromatin remodeling mechanism. PLoS One, 2008. 3(11): p. e3643. 274. Thastrom, A., P.T. Lowary, and J. Widom, Measurement of histone-DNA interaction free energy in nucleosomes. Methods, 2004. 33(1): p. 33-44. 275. Flaus, A., et al., Mapping nucleosome position at single base-pair resolution by using site-directed hydroxyl radicals. Proc Natl Acad Sci U S A, 1996. 93(4): p. 1370-5. 276. Simmons, L.A., et al., Beta clamp directs localization of mismatch repair in Bacillus subtilis. Mol Cell, 2008. 29(3): p. 291-301. 277. Smith, B.T., A.D. Grossman, and G.C. Walker, Visualization of mismatch repair in bacterial cells. Mol Cell, 2001. 8(6): p. 1197-206. 278. Gorman, J., et al., Dynamic basis for one-dimensional DNA scanning by the mismatch repair complex Msh2-Msh6. Mol Cell, 2007. 28(3): p. 359-70. 279. Wang, G.G., C.D. Allis, and P. Chi, Chromatin remodeling and cancer, Part II: ATP-dependent chromatin remodeling. Trends Mol Med, 2007. 13(9): p. 373-80. 280. Davey, C.A., et al., Solvent mediated interactions in the structure of the nucleosome core particle at 1.9 a resolution. J Mol Biol, 2002. 319(5): p. 1097- 113. 281. Neumann, H., et al., A method for genetically installing site-specific acetylation in recombinant histones defines the effects of H3 K56 acetylation. Mol Cell, 2009. 36(1): p. 153-63. 282. Masumoto, H., et al., A role for cell-cycle-regulated histone H3 lysine 56 acetylation in the DNA damage response. Nature, 2005. 436(7048): p. 294-8. 283. Cosgrove, M.S., J.D. Boeke, and C. Wolberger, Regulated nucleosome mobility and the histone code. Nat Struct Mol Biol, 2004. 11(11): p. 1037-43. 284. Jasencakova, Z., et al., Replication stress interferes with histone recycling and predeposition marking of new histones. Mol Cell. 37(5): p. 736-43. 285. Javaid, S., et al., Nucleosome remodeling by hMSH2-hMSH6. Mol Cell, 2009. 36(6): p. 1086-94. 286. Poirier, M.G., et al., Spontaneous access to DNA target sites in folded chromatin fibers. J Mol Biol, 2008. 379(4): p. 772-86. 287. Guerrette, S., et al., Interactions of human hMSH2 with hMSH3 and hMSH2 with hMSH6: examination of mutations found in hereditary nonpolyposis colorectal cancer. Mol Cell Biol, 1998. 18(11): p. 6616-23.

186

288. Schwarz, P.M. and J.C. Hansen, Formation and stability of higher order chromatin structures. Contributions of the histone octamer. J Biol Chem, 1994. 269(23): p. 16284-9. 289. Fischle, W., Y. Wang, and C.D. Allis, Histone and chromatin cross-talk. Curr Opin Cell Biol, 2003. 15(2): p. 172-83. 290. Shogren-Knaak, M., et al., Histone H4-K16 acetylation controls chromatin structure and protein interactions. Science, 2006. 311(5762): p. 844-7. 291. Taverna, S.D., et al., How chromatin-binding modules interpret histone modifications: lessons from professional pocket pickers. Nat Struct Mol Biol, 2007. 14(11): p. 1025-40. 292. Downs, J.A., Histone H3 K56 acetylation, chromatin assembly, and the DNA damage checkpoint. DNA Repair (Amst), 2008. 7(12): p. 2020-4. 293. Rufiange, A., et al., Genome-wide replication-independent histone H3 exchange occurs predominantly at promoters and implicates H3 K56 acetylation and Asf1. Mol Cell, 2007. 27(3): p. 393-405. 294. Celic, I., A. Verreault, and J.D. Boeke, Histone H3 K56 hyperacetylation perturbs replisomes and causes DNA damage. Genetics, 2008. 179(4): p. 1769-84. 295. Han, J., et al., Acetylation of lysine 56 of histone H3 catalyzed by RTT109 and regulated by ASF1 is required for replisome integrity. J Biol Chem, 2007. 282(39): p. 28587-96. 296. Xu, F., et al., Sir2 deacetylates histone H3 lysine 56 to regulate telomeric heterochromatin structure in yeast. Mol Cell, 2007. 27(6): p. 890-900. 297. Adkins, M.W., S.R. Howar, and J.K. Tyler, Chromatin disassembly mediated by the histone chaperone Asf1 is essential for transcriptional activation of the yeast PHO5 and PHO8 genes. Mol Cell, 2004. 14(5): p. 657-66. 298. Xu, F., K. Zhang, and M. Grunstein, Acetylation in histone H3 globular domain regulates gene expression in yeast. Cell, 2005. 121(3): p. 375-85. 299. Richmond, T.J. and C.A. Davey, The structure of DNA in the nucleosome core. Nature, 2003. 423(6936): p. 145-50. 300. Gontijo, A.M., C.M. Green, and G. Almouzni, Repairing DNA damage in chromatin. Biochimie, 2003. 85(11): p. 1133-47. 301. Pederson, D.S. and T. Fidrych, Heat shock factor can activate transcription while bound to nucleosomal DNA in Saccharomyces cerevisiae. Mol Cell Biol, 1994. 14(1): p. 189-99. 302. Perlmann, T. and O. Wrange, Specific glucocorticoid receptor binding to DNA reconstituted in a nucleosome. Embo J, 1988. 7(10): p. 3073-9. 303. Tomschik, M., et al., Fast, long-range, reversible conformational fluctuations in nucleosomes revealed by single-pair fluorescence resonance energy transfer. Proc Natl Acad Sci U S A, 2005. 102(9): p. 3278-83. 304. Chafin, D.R., et al., Human DNA ligase I efficiently seals nicks in nucleosomes. Embo J, 2000. 19(20): p. 5492-501. 305. Pascal, J.M., et al., Human DNA ligase I completely encircles and partially unwinds nicked DNA. Nature, 2004. 432(7016): p. 473-8.

187

306. Kirov, N., et al., In vitro transcription through nucleosomes by T7 RNA polymerase. Embo J, 1992. 11(5): p. 1941-7. 307. Lorch, Y., J.W. LaPointe, and R.D. Kornberg, Nucleosomes inhibit the initiation of transcription but allow chain elongation with the displacement of histones. Cell, 1987. 49(2): p. 203-10. 308. Studitsky, V.M., et al., Mechanism of transcription through the nucleosome by eukaryotic RNA polymerase. Science, 1997. 278(5345): p. 1960-3. 309. Prasad, A., S.S. Wallace, and D.S. Pederson, Initiation of base excision repair of oxidative lesions in nucleosomes by the human, bifunctional DNA glycosylase NTH1. Mol Cell Biol, 2007. 27(24): p. 8442-53. 310. Zhang, Y., et al., Intrinsic histone-DNA interactions are not the major determinant of nucleosome positions in vivo. Nat Struct Mol Biol, 2009. 16(8): p. 847-52. 311. Partensky, P.D. and G.J. Narlikar, Chromatin remodelers act globally, sequence positions nucleosomes locally. J Mol Biol, 2009. 391(1): p. 12-25. 312. Alani, E., et al., Genetic and biochemical analysis of Msh2p-Msh6p: role of ATP hydrolysis and Msh2p-Msh6p subunit interactions in mismatch base pair recognition. Mol Cell Biol, 1997. 17(5): p. 2436-47. 313. Duval, A. and R. Hamelin, Mutations at coding repeat sequences in mismatch repair-deficient human cancers: toward a new concept of target genes for instability. Cancer Res, 2002. 62(9): p. 2447-54. 314. Woods, M.O., et al., A new variant database for mismatch repair genes associated with Lynch syndrome. Hum Mutat, 2007. 28(7): p. 669-73. 315. Chi, P., C.D. Allis, and G.G. Wang, Covalent histone modifications--miswritten, misinterpreted and mis-erased in human cancers. Nat Rev Cancer. 10(7): p. 457- 69. 316. Seligson, D.B., et al., Global histone modification patterns predict risk of prostate cancer recurrence. Nature, 2005. 435(7046): p. 1262-6. 317. Issa, J.P., The epigenetics of colorectal cancer. Ann N Y Acad Sci, 2000. 910: p. 140-53; discussion 153-5. 318. Pollard, K.J. and C.L. Peterson, Role for ADA/GCN5 products in antagonizing chromatin-mediated transcriptional repression. Mol Cell Biol, 1997. 17(11): p. 6212-22. 319. Roberts, S.M. and F. Winston, Essential functional interactions of SAGA, a Saccharomyces cerevisiae complex of Spt, Ada, and Gcn5 proteins, with the Snf/Swi and Srb/mediator complexes. Genetics, 1997. 147(2): p. 451-65. 320. Gutierrez, J.L., et al., Activation domains drive nucleosome eviction by SWI/SNF. Embo J, 2007. 26(3): p. 730-40.

188