Kinetic Characterization of Human DNA Polymerase ε

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of The Ohio State University

By

Walter John Zahurancik

The Ohio State University Biochemistry Graduate Program

The Ohio State University

2018

Dissertation Committee

Zucai Suo, Advisor

Dmitri Kudryashov

Karin Musier-Forsyth

Richard Swenson

1

Copyrighted by

Walter John Zahurancik

2018

2

Abstract

Cell survival and proliferation is dependent upon highly efficient and accurate

DNA replication. To ensure that the genetic information encoded in DNA is faithfully copied and passed on from generation to generation, cells have evolved DNA polymerases, which are specialized enzymes that proficiently replicate DNA through stringent selection of nucleotides based on their preference for forming geometrically and energetically favorable base pairs. In eukaryotes, DNA replication is a tightly regulated process involving the coordinated action of three replicative DNA polymerases. DNA polymerase ε (Polε) is responsible for catalyzing continuous synthesis of the leading strand. Human Polε (hPolε) is a heterotetrameric complex consisting of the catalytic p261 subunit and the non-catalytic p59, p17, and p12 subunits. Though very limited structural studies have been carried out for hPolε, structures of the Saccharomyces cerevisiae homolog have provided insight into the catalytic properties of Polε. Cryo-electron microscopy structures of the yeast Polε heterotetramer and subassemblies revealed that

Polε is characterized by a large globular head-like domain and an extended tail-like domain which is hypothesized to interact with long stretches of newly-synthesized double-stranded DNA (Asturias et al, 2006). Additionally, X-ray crystal structures of the catalytic domain of the yeast Polε catalytic subunit have shown that the catalytic domain possesses a novel P subdomain which allows it to fully encircle DNA, thereby enhancing

ii

DNA binding and polymerization processivity (Hogg et al, 2014). Though informative, these structures are only suggestive of the DNA polymerization activity of hPolε during leading strand DNA synthesis.

One of the long-term research goals in the Suo lab is to develop a more comprehensive mechanistic understanding of human DNA replication, which is a complex process that is executed by the carefully orchestrated activities of many proteins and protein complexes. To approach this daunting task, I set out to perform detailed mechanistic studies of hPolε. I initially carried out pre-steady-state kinetic studies to determine the polymerization mechanism and fidelity of the catalytic domain of the p261 subunit. I determined that the catalytic domain catalyzes nucleotide incorporation using the same kinetic mechanism utilized by nearly all other kinetically-characterized DNA polymerases and is highly accurate due to both stringent nucleotide selection and efficient removal of mismatched bases. Subsequently, I purified hPolε heterotetramer from baculovirus-infected insect cells to investigate the effects of the p261 C-terminal domain and the accessory subunits on the catalytic activities of hPolε. I observed that the C- terminal domain and subunits do not impact polymerization but do attenuate proofreading activity, perhaps to promote processive polymerization under normal DNA synthesis conditions.

The results of my work lay the foundation for future studies exploring the effects of other DNA replication enzymes on the polymerization and fidelity of hPolε.

Furthermore, my work provides the basis for investigating the mechanism of mutagenesis

iii in tumors containing hPolε exonuclease domain mutants. Overall, this study represents an important advance toward our comprehensive understanding of human DNA replication.

iv

Dedication

Dedicated to my mother and father

v

Acknowledgments

First, I would like to thank my advisor Dr. Zucai Suo for his dedicated and selfless mentorship throughout my graduate career. Over the last six years, Dr. Suo has been one of my greatest advocates, offering his unwavering support time and again.

Under his guidance, I have developed into an independent and confident researcher. I would also like to thank the members of my advisory committee, Drs. Dmitri

Kudryashov, Karin Musier-Forsyth, and Richard Swenson, who have continually expressed their confidence in my abilities as a student and researcher. It has been with their support that my achievements as a graduate student have been made possible. I extend my sincere gratitude to each of them.

I would like to thank the current members of the Suo lab. I am thankful for the friendship of Austin Raper, Andrew Reed, Anthony Stephenson, and Jack Tokarsky.

Over the last four years we have worked together, we have grown and learned as graduate students and as a family. Through this long and arduous process, their companionship has been invaluable and irreplaceable. They supported me, made me laugh, and challenged me every day, and I owe much of my development as an independent scientist to them. I would also like to thank past members of the Suo lab. In particular, I would like to thank

Drs. Jason Fowler, Brian Maxwell, David Taggart, and Rajan Vyas, who were among my first mentors in the lab and who taught me how to have fun with science, pay attention to

vi details, and trust my skills. Their contributions to my development early in my career cannot be understated. I would also like to thank Dr. Varun Gadkari, who has been my biggest supporter and steadfast friend since we started graduate school together. Through my highest and lowest moments, Varun stuck by me and always looked out for what was best for me. He still watches over me, even from ‘That School up North’! Finally, I would like to acknowledge Seth Klein, my first mentee and co-author. It was a pleasure working alongside Seth and reaching my first first-author publication milestone with him.

I would like to acknowledge my friends, colleagues, and faculty in the

Department of Chemistry and Biochemistry and OSBP, MCDB, and Biophysics graduate programs. The breadth of meaningful relationships built over the years is enormous, and I look fondly on the memories I have gathered. Moreover, I look forward to their future successes. I wish the very best for my colleagues and hope they achieve everything they strive for. I would like to extend a special thanks to Dr. Venkat Gopalan, whose selflessness and passion for discovery, integrity, and honor have set the standard for the researcher I would like to become. I would also like to thank Dr. Lien Lai and Stella Lai for treating me as a part of their family and making Columbus feel even more like home.

Finally, I would like to thank my family for their support and encouragement. I would like to thank my mother, Angela, and my stepfather, Dave, for their unconditional love and tireless thoughts and prayers. Special thanks to Uncle Ross for being my go-to guy for laughs at the dinner table. I would like to thank my father, Paul, and my stepmother, Deanna, for always making me feel welcome in their home. I would like to thank my older brother Paul for putting up with me when I was little and becoming one

vii of my closest friends as I have grown older. I would also like to thank my younger brother John for making me put up with him and making me feel important and special as his older brother. Lastly, I would like to thank the other members of my family, who are too numerous to name individually. I will never tire of answering the question, “What are you working on now?”, at family gatherings.

viii

Vita

2008 – 2012...... B.S. Biochemistry,

The Ohio State University, Columbus, OH

2012 – 2018...... Ph.D. Ohio State Biochemistry Program,

The Ohio State University, Columbus, OH

2012 – 2013...... Graduate Teaching Associate,

Department of Chemistry and Biochemistry,

The Ohio State University, Columbus, OH

2013 – 2014; 2015 – 2016 ...... Chemistry-Biology Interface Training

Program Fellow,

Department of Chemistry and Biochemistry,

The Ohio State University, Columbus, OH

2014 – 2015...... Graduate Research Associate,

Department of Chemistry and Biochemistry,

The Ohio State University, Columbus, OH

2016 – 2017...... Pelotonia Graduate Fellow,

Department of Chemistry and Biochemistry,

The Ohio State University, Columbus, OH

ix

2017 – 2018...... Presidential Research Fellow,

Department of Chemistry and Biochemistry,

The Ohio State University, Columbus, OH

Publications

*Denotes Co-First-Author Publications

1. Kimsey, I. J., Szymanski, E. S., Zahurancik, W. J., Shakya, A., Xue, Y., Chu, C.

–C., Sathyamoorthy, B., Suo, Z., and Al-Hashimi, H. M. (2018) Dynamic Basis

for dG-dT Misincorporation via Tautomerization and Ionization. Nature, 554,

195-201. doi:10.1038/nature25487.

2. Campbell, B. B., Light, N., Fabrizio, D., Zatzman, M., Fuligni, F., de Borja, R.,

Davidson, S., Edwards, M., Elvin, J. A., Hodel, K. P., Zahurancik, W. J., Suo,

Z., Lipman, T., Wimmer, K., Kratz, C. P., Bowers, D. C., Laetsch, T. W., Dunn,

G. P., Johanns, T. M., Grimmer, M. R., Smirnov, I. V., Larouche, V., Samuel, D.,

Bronsema, A., Osborn, M., Stearns, D., Raman, P., Cole, K. A., Storm, P. B.,

Yalon, M., Opocher, E., Mason, G., Thomas, G. A., Sabel, M., George, B.,

Ziegler, D. S., Lindhorst, S., Issai, V. M., Constantini, S., Toledano, H., Elhasid,

R., Farah, R., Dvir, R., Dirks, P., Huang, A., Galati, M. A., Chung, J.,

Ramaswamy, V., Irwin, M. S., Aronson, M., Durno, C., Taylor, M. D., Rechavi,

G., Maris, J. M., Bouffet, E., Hawkins, C., Costello, J. F., Meyn, M. S., Pursell, Z.

x

F., Malkin, D., Tabori, U., and Shlien, A. (2017) Comprehensive Analysis of

Hypermutation in Human Cancer. Cell, 171 (5), 1042-1056.

doi:10.1016/j.cell.2017.09.048.

3. Wu, J., Wang, P., Li, L., Williams, N. L., Zahurancik, W. J., You, C., Wang, J.,

Suo, Z., and Wang, Y. (2017) Replication Studies of Carboxymethylated DNA

Lesions in Human Cells. Nucleic Acids Research, 45 (12), 7276-7284.

doi:10.1093/nar/gkx442.

4. Vyas, R., Reed, A. J., Raper, A. T., Zahurancik, W. J., Wallenmeyer, P. C., and

Suo, Z. (2017) Structural Basis for the D-stereoselectivity of Human DNA

Polymerase β. Nucleic Acids Research, 45 (10), 6228-6237.

doi:10.1093/nar/gkx252.

5. Tokarsky, E. J., Gadkari, V. V., Zahurancik W. J., Malik, C. K., Basu, A. K.,

and Suo, Z. (2016) Pre-steady-state Kinetic Investigation of Bypass of a Bulky

Guanine Lesion by Human Y-family DNA Polymerases. DNA Repair, 46, 20-28.

doi:10.1016/j.dnarep.2016.08.002.

6. Dunn, M. R., Larsen, A. C., Zahurancik, W. J., Fahmi, N. E., Meyers, M., Suo,

Z., and Chaput, J. C. (2015) DNA Polymerase-mediated Synthesis of Unbiased

Threose Nucleic Acid (TNA) Polymers Requires 7-deazaguanine to Suppress G:G

Mispairing during TNA Transcription. J. Am. Chem. Soc, 137 (12), 4014-4017.

doi:10.1021/ja511481n.

7. Zahurancik, W. J., Baranovskiy, A. G., Tahirov, T. H., and Suo, Z. (2015)

Comparison of the Kinetic Parameters of the Truncated Catalytic Subunit and

xi

Holoenzyme of Human DNA Polymerase ε. DNA Repair, 29, 16-22.

doi:10.1016/j.dnarep.2015.01.008.

8. Zahurancik, W. J., Klein, S. J., and Suo, Z. (2014) Significant Contribution of

the 3′→5′ Exonuclease Activity to the High Fidelity of Nucleotide Incorporation

Catalyzed by Human DNA Polymerase ε. Nucleic Acids Research, 42(22), 13853-

13860. doi:10.1093/nar/gku1184.

9. Vyas, R.*, Zahurancik, W. J.*, and Suo, Z. (2014) Structural Basis for the

Binding and Incorporation of Nucleotide Analogs with L-stereochemistry by

Human DNA Polymerase λ. PNAS Plus, 111(30), E3033-3042.

doi:10.1073/pnas.1401286111.

10. Zahurancik, W. J., Klein, S. J., and Suo, Z. (2013) Kinetic Mechanism of DNA

Polymerization Catalyzed by Truncated Human DNA Polymerase ε.

Biochemistry, 52, 7041-7049. doi:10.1021/bi400803v.

11. Göksenin, A. Y., Zahurancik, W., LeCompte, K. G., Taggart, D. J., Suo, Z.,

Pursell, Z. F. (2012) Human DNA Polymerase Epsilon Is Able to Efficiently

Extend from Multiple Consecutive Ribonucleotides. J. Biol. Chem., 287, 42675-

42684. doi:10.1074/jbc.M112.422733.

12. Doddapaneni, K., Zahurancik, W., Haushalter, A., Yuan, C., Jackman, J., Wu, Z.

(2011) RCL Hydrolyzes 2′-deoxyribonucleoside 5′-monophosphate via Formation

of a Reaction Intermediate. Biochemistry, 50, 4712-4719. doi:10.1021/bi101742z.

xii

Field of Study

Major Field: The Ohio State University Biochemistry Graduate Program

xiii

Table of Contents

Abstract ...... ii

Dedication ...... v

Acknowledgments...... vi

Vita ...... ix

Table of Contents ...... xiv

List of Schemes ...... xix

List of Tables ...... xx

List of Figures ...... xxi

Chapter 1. Introduction ...... 1

1.1 DNA Polymerases ...... 1

1.2 Mechanism and Fidelity of DNA Polymerases ...... 2

1.3 DNA Polymerases at the Eukaryotic Replication Fork ...... 5

1.4 DNA Polymerase ε ...... 6

1.5 Human Polε Mutations in Cancer ...... 9

1.6 Focus of the Dissertation ...... 10

1.7 Schemes ...... 13

1.8 Figures ...... 14

Chapter 2. Kinetic Mechanism of DNA Polymerization Catalyzed by Truncated Human DNA Polymerase ε...... 17

2.1 Introduction ...... 18

xiv

2.2 Materials and Methods ...... 19

2.2.1 Materials ...... 19

2.2.2 Expression and Purification of Human Polε ...... 20

2.2.3 Pre-Steady-State Kinetic Assays ...... 20

2.2.4 Measurement of the Rate Constant of DNA Dissociation from the Binary Complex ...... 21

2.2.5 Active Site Titration Assay ...... 21

2.2.6 Processive Polymerization Assay ...... 22

2.2.7 Elemental Effect of Nucleotide Incorporation ...... 22

2.2.8 Pulse-Chase and Pulse-Quench Experiments ...... 22

2.2.9 Product Analysis ...... 23

2.2.10 Data Analysis ...... 23

2.3 Results ...... 25

2.3.1 Expression and Purification of an Exonuclease-Deficient Mutant of Human Polε ...... 25

2.3.2 Burst Kinetics ...... 26

2.3.3 Active Site Titration ...... 27

2.3.4 Pre-Steady-State Kinetics of Correct dTTP Incorporation by Polε exo- ...... 28

2.3.5 Processive Polymerization ...... 29

2.3.6 Elemental Effect of Nucleotide Incorporation ...... 30

2.3.7 Pulse-Chase and Pulse-Quench Experiments ...... 31

2.4 Discussion ...... 33

2.5 Schemes ...... 38

2.6 Tables ...... 39

2.7 Figures ...... 40

xv

Chapter 3. Significant Contribution of the 3′→5′ Exonuclease Activity to the High Fidelity of Nucleotide Incorporation Catalyzed by Human DNA Polymerase ε ...... 49

3.1 Introduction ...... 50

3.2 Materials and Methods ...... 52

3.2.1 Materials ...... 52

3.2.2 DNA Substrates ...... 53

3.2.3 Polymerase and Exonuclease Single-Turnover Assays ...... 53

3.2.4 Product Analysis ...... 54

3.2.5 Data Analysis ...... 54

3.3 Results ...... 55

3.3.1 Substrate Specificity of hPolε exo- ...... 55

3.3.2 Mismatch Extension Fidelity of hPolε exo- ...... 56

3.3.3 Excision of Matched and Mismatched DNA Substrates by hPolε exo+ ...... 57

3.4 Discussion ...... 58

3.5 Tables ...... 64

3.6 Figures ...... 68

Chapter 4. Comparison of the Kinetic Parameters of the Truncated Catalytic Subunit and Holoenzyme of Human DNA Polymerase ε ...... 71

4.1 Introduction ...... 72

4.2 Materials and Methods ...... 74

4.2.1 Materials ...... 74

4.2.2 DNA Substrates ...... 75

4.2.3 Purification of the Human DNA Polymerase ε Heterotetramer and the p261N Fragment ...... 75

4.2.4 Reaction Buffers ...... 76

xvi

4.2.5 Pre-Steady-State Kinetic Assays ...... 76

4.2.6 Active Site Titration Assay ...... 76

4.2.7 Processivity Assays ...... 77

4.2.8 Product Analysis ...... 77

4.2.9 Data Analysis ...... 77

4.3 Results and Discussion ...... 78

4.3.1 Burst Assays ...... 78

4.3.2 Active Site Titration Assay ...... 80

4.3.3 Processivity Assays ...... 81

4.3.4 Excision of Matched and Mismatch DNA Substrates by hPolε ...... 82

4.3.5 Concluding Remarks ...... 84

4.4 Tables ...... 85

4.5 Figures ...... 86

Chapter 5. The Accessory Subunits of Human DNA Polymerase ε Regulate the 3′→5′ Exonuclease Activity ...... 92

5.1 Introduction ...... 92

5.2 Materials and Methods ...... 95

5.2.1 Materials ...... 95

5.2.2 Preparation of the Human DNA Polymerase ε Heterotetramer ...... 95

5.2.3 DNA Substrates ...... 96

5.2.4 Pre-Steady-State Kinetic Assays ...... 96

5.2.5 Active Site Titration Assay ...... 96

5.2.6 Measurement of the E•DNA Complex Dissociation Rate Constant ...... 97

5.2.7 Measurement of the Steady-State Rate Constant of Correct Nucleotide Incorporation ...... 97

xvii

5.2.8 Measurement of the Elemental Effect on Nucleotide Incorporation ...... 97

5.2.9 Single-Turnover Exonuclease Assays ...... 97

5.2.10 Product Analysis ...... 98

5.2.11 Data Analysis ...... 98

5.3 Results ...... 100

5.3.1 Active Site Titration ...... 100

5.3.2 Measurement of the E•DNA Complex Dissociation Rate Constant ...... 101

5.3.3 Pre-Steady-State Kinetics of Correct Nucleotide Incorporation ...... 102

5.3.4 Elemental Effect on Nucleotide Incorporation ...... 103

5.3.5 Pre-Steady-State Kinetics of Incorrect Nucleotide Incorporation ...... 104

5.3.6 Pre-Steady-State Kinetics of Mismatch Extension ...... 104

5.3.7 Pre-Steady-State Kinetics of Matched and Mismatched Base Pair Excision 106

5.4 Discussion ...... 107

5.5 Tables ...... 114

5.6 Figures ...... 118

Chapter 6. Epilogue ...... 123

6.1 Future Studies on Human DNA Polymerase ε ...... 123

6.2 Future Studies on Cancer-Associated Mutations of Human DNA Polymerase ε . 124

6.3 Future Studies on the Leading Strand Synthesis ...... 125

6.4 Concluding Remarks ...... 126

References ...... 127

xviii

List of Schemes

Scheme 1.1: Minimal kinetic mechanism of DNA polymerization ...... 13

Scheme 2.1: Minimal kinetic mechanism of DNA polymerization catalyzed by human DNA polymerase ε ...... 38

xix

List of Tables

Table 2.1: Kinetic parameters for correct nucleotide incorporation catalyzed by Polε exo- at 20 °C ...... 39

Table 3.1: Sequences of DNA substrates ...... 64

Table 3.2: Kinetic parameters for correct and incorrect nucleotide incorporation catalyzed by hPolε exo- at 20 °C ...... 65

Table 3.3: Kinetic parameters for mismatch extension and excision catalyzed by hPolε exo- and hPolε exo+ at 20 °C ...... 66

Table 3.4: Comparison of the contribution of 3′→5′ exonuclease activity to the overall fidelity of replicative DNA polymerases when encountering a single base mismatch in the staggering end of a DNA substrate ...... 67

Table 4.1: DNA substrates ...... 85

Table 5.1: Sequences of DNA substrates ...... 114

Table 5.2: Kinetic parameters for correct and incorrect nucleotide incorporation onto the D-1 DNA substrate catalyzed by hPolε exo- at 20 °C ...... 115

Table 5.3: Kinetic parameters for mismatch extension and excision catalyzed by hPolε exo- and hPolε exo+ at 20 °C ...... 116

Table 5.4: Comparison of kinetic parameters determined for p261N and hPolε at 20 °C ...... 117

xx

List of Figures

Figure 1.1: DNA polymerase structure is highly conserved ...... 14

Figure 1.2: Cryo-EM model of Saccharomyces cerevisiae Polε heterotetramer ...... 15

Figure 1.3: Ternary structure of Saccharomyces cerevisiae Pol2 catalytic domain bound to DNA and an incoming dATP...... 16

Figure 2.1: Pre-steady-state and steady-state kinetics of correct dTTP incorporation into a DNA substrate D-1 by Polε exo- at 20 °C ...... 40

Figure 2.2: Conservation of Exo I and Exo II motifs across the B-family DNA polymerases...... 41

Figure 2.3: DNA dissociation from the Polε•D-1 binary complex ...... 42

Figure 2.4: Active site titration of Polε exo- at 20 °C ...... 43

Figure 2.5: Concentration dependence on the pre-steady-state rate constant of correct dTTP incorporation catalyzed by Polε exo- at 20 °C ...... 44

Figure 2.6: Processive polymerization catalyzed by Polε exo- at 20 °C ...... 45

Figure 2.7: Elemental effect on the rate constant of correct and incorrect nucleotide incorporation catalyzed by Polε exo- at 20 °C ...... 47

Figure 2.8: Pulse-chase and pulse-quench experiments at 20 °C ...... 48

Figure 3.1: Nucleotide concentration dependence on the pre-steady-state kinetic parameters of correct dCTP and incorrect dATP incorporation opposite dG catalyzed by hPolε exo- at 20 °C ...... 68

Figure 3.2: Extension of a mismatched base pair catalyzed by hPolε exo- at 20 °C ...... 69

Figure 3.3: Excision of primers with matched and mismatched 3′ termini catalyzed by hPolε exo+ at 20 °C ...... 70 xxi

Figure 4.1: Analysis of hPolε purity by SDS-PAGE ...... 86

Figure 4.2: Biphasic kinetics of correct dTTP incorporation by hPolε and p261N at 20 °C ...... 87

Figure 4.3: Active site titration of hPolε at 20 °C ...... 88

Figure 4.4: Processive DNA synthesis by hPolε and p261N on singly-primed M13mp2 ssDNA templates at 20 °C ...... 89

Figure 4.5: Comparison of processivities of hPolε and p261N on the 21-mer M13 and 45- mer M13 DNA substrates at 20 °C ...... 90

Figure 4.6: Excision of matched and mismatched DNA by hPolε at 20 °C ...... 91

DNA Figure 5.1: Active site titration assay to measure the Kd for hPolε exo- binding to DNA ...... 118

Figure 5.2: Measurement of E•DNA complex dissociation rate constant ...... 119

Figure 5.3: Pre-steady-state kinetics of correct nucleotide incorporation ...... 120

Figure 5.4: Elemental effect on correct and incorrect nucleotide incorporation ...... 121

Figure 5.5: Pre-steady-state kinetics of matched and mismatched base pair excision ... 122

xxii

Chapter 1. Introduction

1.1 DNA Polymerases

In all living organisms, DNA contains the coded instructions for synthesizing proteins which are involved in nearly all biological processes. Accordingly, conservation of genetic information is of the utmost importance for organism survival and proliferation. To ensure efficient and accurate passage of genetic information from generation to generation, cells have evolved specialized enzymes termed DNA polymerases (Pols) to replicate their DNA. During DNA replication, DNA polymerases add the four nucleoside triphosphates (dNTPs; N = A, T, C, or G) to the growing DNA primer strand while the complementary DNA strand serves as a template, guiding the selection of the next dNTP through recognition of geometrically and energetically favorable Watson-Crick base pairs (A-T or C-G). In this fashion, the genetic code is preserved and transmitted in a near-error-free manner.

DNA polymerases are phylogenetically grouped into seven families (A, B, C, D,

X, Y, and RT) based on sequence homology1. Though DNA polymerases across or even within the seven families differ in their primary amino acid sequences, number of domains, biological roles, DNA substrate preferences, and DNA polymerization fidelities, all structurally characterized DNA polymerases share a common polymerase core domain which consists of thumb, fingers, and palm subdomains arranged in a “right

1 hand” configuration1-3 (Figure 1.1). The thumb subdomain makes direct contacts with the minor groove of double-stranded DNA. Binding to DNA triggers a conformational change in the thumb subdomain, enabling the polymerase to more fully encircle DNA.

This motion is thought to enhance DNA polymerase processivity, allowing the DNA polymerase to synthesize longer stretches of DNA prior to dissociation. The fingers subdomain binds to and aids in the selection of dNTPs by stabilizing Watson-Crick base pairs at the enzyme active site while rejecting mismatched base pairs with distorted base pairing geometry. Correct dNTP binding promotes a conformational change in the fingers subdomain which closes around the active site and positions active site residues, essential divalent metal ions, the primer 3′-OH, and the dNTP 5′-triphosphate in close proximity for catalysis. In contrast, incorrect dNTP binding prevents proper closing of the fingers subdomain, allowing the mispaired dNTP to dissociate and the correct dNTP to bind.

Finally, the palm subdomain contains the conserved catalytic aspartic acid residues which coordinate two Mg2+ ions that are required for catalysis. Beyond the polymerase core domain, many DNA polymerases possess additional domains that pertain directly to their biological function. For example, most replicative DNA polymerases from the A- and B- families have exonuclease domains which catalyze removal of mismatched base pairs that are erroneously inserted during DNA replication.

1.2 Mechanism and Fidelity of DNA Polymerases

Pre-steady-state and steady-state kinetic methods have established a minimal kinetic mechanism of nucleotide incorporation (Scheme 1.1) that is shared by all DNA polymerases4. First, the DNA polymerase binds to its DNA substrate to form the E•DNA

2 binary complex (Step 1). Next, an incoming nucleotide binds to the E•DNA complex to form the E•DNA•dNTP ground-state ternary complex (Step 2). The DNA polymerase then undergoes a conformational change to form the active ternary complex (Step 3,

E*•DNA•dNTP), thereby repositioning active site residues, divalent metal ions, and reactive groups for catalysis. The 3′-OH of the DNA primer then makes an in-line nucleophilic attack on the α-phosphate of the incoming dNTP to form a new phosphodiester bond (Step 4). Following the chemistry step, the DNA polymerase undergoes a reverse conformational change to return to its initial state (Step 5). Finally, the pyrophosphate (PPi) product dissociates from the binary complex, allowing another dNTP to bind to restart the catalytic cycle.

The rate-limiting step of the nucleotide incorporation reaction in Scheme 1.1 has been extensively investigated for many DNA polymerases using a pre-steady-state kinetic approach. For nearly all kinetically characterized DNA polymerases5-17, it has been suggested that the pre-chemistry conformational change (Step 3) is rate-limiting for correct nucleotide incorporation on the basis of two pieces of evidence: i) there is a small elemental effect (<4-fold) on the rate of correct nucleotide incorporation in reactions using a 5′-[α-thio]triphosphate-substituted dNTP in place of a natural dNTP; and ii) a significantly higher product amplitude is observed when reactions with [α-32P]-labeled dNTP are chased with a large excess of unlabeled dNTP (pulse-chase) compared to reactions that are immediately quenched with acid (pulse-quench). In contrast, the chemistry step (Step 4) has been suggested to be rate-limiting for incorrect nucleotide incorporation due to measurement of very large elemental effects4. Although this model

3 has been accepted for years, recent stopped-flow fluorescence studies on the mechanism of DNA polymerization have revealed that conformational changes preceding chemistry are too rapid to be rate limiting for correct nucleotide incorporation18,19. Thus, the physical nature of the rate-limiting step remains to be unambiguously determined.

Pre-steady-state kinetic methods have also been used to measure and calculate the fidelity of a wide variety of DNA polymerases8,20-29. Fidelity is defined as the selectivity for correct over incorrect nucleotides and is typically expressed as the frequency of errors made by a polymerase per nucleotide incorporation event. Though all DNA polymerases share a common mechanism for nucleotide incorporation, they differ widely in their fidelity. Importantly, replicative DNA polymerases that are responsible for copying the vast majority of genomic DNA are among the most accurate DNA polymerases, with reported fidelities in the range of 10-6-10-8 (i.e., 1 error per 106-108 nucleotides incorporated)30,31.

Based on kinetic studies, the high fidelity of replicative DNA polymerases is achieved two-fold. First, replicative DNA polymerases incorporate correct nucleotides two to three orders of magnitude faster and bind correct nucleotides one to two orders of magnitude more tightly than incorrect nucleotides. Accordingly, replicative DNA polymerases are reported to have base substitution fidelities in the range of 10-4-10-6

8,21,24,27-29. Second, in addition to having high specificity for correct nucleotides, most A- family and B-family replicative DNA polymerases also possess 3′→5′ exonuclease activity which removes mismatched nucleotides inserted during DNA replication. While removal of correctly paired nucleotides is relatively slow, the rate of excision is

4 significantly enhanced in the presence of a mismatched nucleotide. Moreover, the rate of mismatch excision typically exceeds that of mismatch extension by as much as 100- fold5,29,32. Thus, the 3′→5′ exonuclease activity enhances the fidelity of replicative DNA polymerases by 100-fold, resulting in an overall DNA polymerization fidelity of 10-6-10-

8.

1.3 DNA Polymerases at the Eukaryotic Replication Fork

Accurate and efficient replication of the eukaryotic nuclear genome involves the coordinated action of three essential B-family replicative DNA polymerases (Pols): Polα,

Polδ, and Polε. At the replication fork, the Cdc45-Mcm2-7-GINS (CMG) helicase complex tracks along single-stranded DNA in the 3′→5′ direction and unwinds double stranded DNA, generating the single-stranded lagging and leading strand templates33.

DNA synthesis on both strands is initiated by the heterotetrameric Polα-primase complex34,35 which catalyzes de novo, template-directed oligonucleotide synthesis to generate short RNA/DNA primers36. The lagging strand is synthesized in a discontinuous fashion in the direction opposite fork unwinding. As single-stranded DNA is generated,

Polα-primase synthesizes primers that are extended by Polδ to form 100- to 200- nucleotide Okazaki fragments. The Okazaki fragments are then processed by the coordinated effort of flap endonuclease 1, Polδ, and DNA ligase to remove ribonucleotides and form a complete complement strand37,38. In contrast, the leading strand is synthesized in a continuous fashion from a single primer. Genetic studies have provided evidence that Polε is the main polymerase to synthesize the leading strand during bulk genomic DNA replication39-41. Near-atomic resolution structures of S.

5 cerevisiae Polε bound to the CMG helicase have been captured by single-particle cryo- electron microscopy (cryo-EM)42,43 and demonstrate that Polε interacts primarily with the

C-terminal face of the CMG helicase, receiving the single-stranded DNA template as it passes through the central channel of the helicase. Consequently, DNA unwinding and synthesis on the leading strand are tightly coupled33.

1.4 DNA Polymerase ε

In all eukaryotes, Polε is a heterotetramer consisting of a large, multi-domain catalytic subunit and three additional small, non-catalytic subunits. Like most high fidelity replicative DNA polymerases, Polε catalyzes both 5′→3′ template-directed polymerase activity as well as 3′→5′ exonuclease activity that removes mismatches generated during replication44. The vital role of Polε in cell survival is emphasized by its involvement in a variety of cellular processes beyond DNA replication including cell cycle regulation45-48, gene silencing49,50, sister chromatid cohesion51,52, and base excision repair53,54.

Human Polε (hPolε) is a heterotetramer that is comprised of the p261 catalytic subunit and three small subunits (p59, p17, and p12)55,56. The p261 catalytic subunit is divided into two distinct domains: an N-terminal domain (p261N) that is homologous to archaeal B-family DNA polymerases and possesses the conserved polymerase and exonuclease subdomains, and a C-terminal domain which is homologous to another smaller class of archaeal B-family DNA polymerases but contains inactivating mutations in the predicted catalytic motifs57. In addition to the non-functional polymerase- exonuclease module, the C-terminal domain also contains two Zn fingers which are

6 critical for forming a stable interaction between p261 and p5958. The p17 and p12 subunits also interact with the C-terminal domain of p261, but the Zn fingers do not appear to play a role in mediating these interactions58. Though hPolε exists as a heterotetramer in vivo, stable subassemblies of hPolε, such as p261-p59 and p17-p12 dimers and a p261-p17-p12 trimer, are capable of forming in vitro58.

Previously, structures of Saccharomyces cerevisiae Pol2 (p261 homolog), the

Pol2-Dpb2 dimer (p261-p59), and the full S. cerevisiae Polε heterotetramer were determined by cryo-EM to 20 Å resolution59 and revealed that the complex is characterized by a globular head-like structure comprised of the Pol2 catalytic subunit and a flexible tail-like structure corresponding to the three small subunits (Figure 1.2).

Accompanying biochemical assays supported the hypothesis that the extra density afforded by the small subunits engages longer stretches of double-stranded DNA and increases the processivity of Pol2. This hypothesis is consistent with other studies demonstrating that the small subunits interact with double-stranded DNA60 and are essential for maximal processivity61.

Recently, crystal structures of both the wild type62 and exonuclease-deficient63 N- terminal catalytic domain of yeast Pol2 bound to DNA and an incoming nucleotide were determined (Figure 1.3). The structures show that the catalytic domain adopts the canonical DNA polymerase right-hand structure consisting of fingers, palm, and thumb subdomains in addition to N-terminal and 3′→5′ exonuclease subdomains that are typically found in A-family and B-family replicative polymerases. Interestingly, Pol2 possesses a novel P subdomain that is inserted into the palm subdomain which allows the

7 polymerase to fully encircle double-stranded DNA and is indispensable for processive

DNA synthesis. The structure of the C-terminal domain of the catalytic subunit from any organism remains to be determined, and it is expected that a high-resolution structure of this domain will provide key insights into how the C-terminal domain mediates interactions between the catalytic subunit and the small subunits.

Like the catalytic subunit, there is only limited structural information for the small subunits. The N-terminal domain (residues 1-75) of the hPolε p59 subunit was previously characterized by NMR64. Interestingly, the p59 N-terminal domain shares a highly similar fold with the C-domains of AAA+ proteins, a group which includes helicases, processivity sliding clamp loaders, and other proteins involved in DNA metabolism.

However, distortion of this domain relative to functional clamp loaders suggests that the p59 N-terminal domain does not likely recognize and bind ATP. More recently, an X-ray crystal structure of full-length p59 bound to a short C-terminal fragment of p261 confirmed that Zn2+ is required for p59 binding to p26165. Additionally, the structure of the Drosophila melanogaster p17 homolog, CHRAC14, complexed with CHRAC16, a subunit of a nucleosome remodeling complex, was determined by X-ray crystallography and revealed that CHRAC14 and CHRAC16 adopt a histone fold as predicted by sequence analysis66. Consistently, the p17 and p12 subunits of hPolε were also predicted to contain histone fold motifs56, and the yeast Dpb4 (p17 homolog) subunit was shown to interact with Dls1, a chromatin accessibility complex protein49. Thus, the p17 subunit of hPolε likely partners with chromatin accessibility complex proteins in vivo, but the

8 importance of this subunit sharing between complexes is not well understood. Lastly, the structure of the p12 subunit, or any of its homologs, has not yet been described.

1.5 Human Polε Mutations in Cancer

Recent whole-exome and whole-genome sequencing of 452 colorectal cancer

(CRC) and endometrial cancer (EEC) tumors by The Cancer Genome Atlas Network

(TCGA) identified a subset of tumors that were microsatellite stable (MSS) and hypermutated with a mutation frequency exceeding 100 mutations per 106 bases67.

Interestingly, each of these tumors had an hPolε exonuclease-domain mutation (EDM) and the V411L and P286R EDMs were the most frequently identified. It is currently estimated that 3% of CRCs and 8% of EECs have hPolε EDMs68. Remarkably, these

EDMs are associated with the highest number of mutations found in cancer genome sequencing studies to date and give rise to a unique mutation signature69,70 consisting primarily of TCT→TAT transversions and TCG→TTG transitions. Furthermore, an exhaustive study in which over 81,000 tumors were examined by sequencing analysis discovered previously unidentified cancer-driving hPolε EDMs that were associated exclusively with hypermutated tumors71. Notably, several of the identified cancer- associated hPolε EDMs significantly affect the 3′→5′ exonuclease activity of p261N41,71 and likely decrease the overall fidelity of DNA replication by hPolε. Thus, the observed effect of these mutations on proofreading by hPolε may partly give rise to such a mutation signature and contribute to tumorigenesis. Importantly, a mechanistic basis for the mutation signature in hPolε EDM tumors has not yet been explored.

9

1.6 Focus of the Dissertation

One of the long-term research goals in the Suo lab is to define the kinetic mechanism of human DNA replication at the replication fork. This is no small task as

DNA replication is divided into two distinct and tightly coordinated processes, leading and lagging strand synthesis, each of which are uniquely performed and regulated by a staggering network of diverse proteins including three heterotetrameric DNA polymerase complexes, an 11-subunit helicase comprised of three separate protein complexes, a sliding clamp and its associated clamp loader complex, and numerous other proteins and protein complexes. Therefore, given the complexity of human DNA replication, it is critical to define the individual contributions of each of these components to the overall kinetics of DNA synthesis in order to develop a detailed kinetic mechanism.

At present, there are limited kinetic studies describing the catalytic activities of the human replicative DNA polymerases72,73. Thus, in order to begin to understand human DNA replication, we must carry out more detailed kinetic investigation of these polymerases. Among the three human replicative DNA polymerases, hPolε has remained the most elusive due to the difficulty of preparing sufficient quantities of active protein required for thorough kinetic characterization. The goal of my dissertation project has been to overcome this barrier and perform a full kinetic investigation of the catalytic activities of hPolε.

To approach this challenge, we initially worked with a truncation of the p261 catalytic subunit which comprises only the N-terminal catalytic domain and lacks the entire C-terminal domain. The rationale for this approach was that the N-terminal

10 catalytic domain possesses all of the conserved polymerase and exonuclease motifs58 and can be overexpressed in E. coli and purified to homogeneity in milligram quantities which would be suitable for pre-steady-state kinetic studies. In Chapter 2, I describe the preparation of an exonuclease-deficient mutant of the catalytic domain of hPolε p261. We determined that the catalytic domain rapidly catalyzes correct nucleotide incorporation at

20 °C and exhibits a pre-chemistry rate-limiting conformational change like other kinetically characterized replicative DNA polymerases. In Chapter 3, I describe a complete pre-steady-state kinetic investigation of the fidelity of nucleotide incorporation catalyzed by the catalytic domain. Notably, we observed that the 3′→5′ exonuclease activity of the catalytic domain significantly enhances its DNA polymerization fidelity, resulting in an overall fidelity of 10-6-10-11, which is sufficient to replicate the human haploid genome (3x109 bp) in an error-free manner.

Shortly after completion of these studies, we were provided with purified and fully-assembled hPolε heterotetramer by our collaborators in the Tahirov lab at the

University of Nebraska. In Chapter 4, I describe initial kinetic studies to compare the activity of the heterotetramer with that of the catalytic domain alone. While the polymerization activity appeared unaffected by the intact p261 catalytic subunit and the presence of the p59, p17, and p12 accessory subunits, we observed the surprising result that the 3′→5′ exonuclease activity of the heterotetramer was actually reduced against both matched and mismatch-containing DNA substrates compared to the catalytic domain alone. In an effort to carry out more detailed kinetic studies of the fully-assembled heterotetramer, we established a system in our lab for overexpressing wild type and

11 exonuclease-deficient hPolε heterotetramer using insect cells and baculoviruses with the assistance of Yoshihiro Matsumoto at the University of New Mexico. In Chapter 5, I describe preparation and pre-steady-state kinetic characterization of fully-assembled hPolε heterotetramer. Significantly, we confirmed the result that the 3′→5′ exonuclease activity of the heterotetramer was indeed attenuated compared to the catalytic domain alone. The significance of this result remains to be determined.

With the completion of these studies, we have now set the foundation for future work in our lab investigating the impact of replication fork proteins on the kinetics of

DNA polymerization catalyzed by hPolε. Additionally, our efforts to kinetically characterize the fidelity of DNA polymerization by hPolε have provided a baseline for assessing the impact of putative cancer-driving exonuclease domain mutations on the fidelity of hPolε. Overall, these studies serve as a significant stepping stone toward a complete understanding of the mechanism of human DNA replication.

12

1.7 Schemes

Scheme 1.1: Minimal kinetic mechanism of DNA polymerization. Figure adapted from Joyce and Benkovic, 20044.

13

1.8 Figures A B

C D

Figure 1.1: DNA polymerase structure is highly conserved (A) A-family polymerase Klentaq1. (B) B-family polymerase RB69 DNA polymerase.

(C) X-family polymerase human Polβ. (D) Y-family polymerase Sulfolobus solfataricus

Dpo4. The fingers subdomain is in yellow, the palm subdomain is in red, and the thumb subdomain is in green. Additional polymerase-specific subdomains are colored in white or grey. Adapted from Rothwell and Waksman, 20051. 14

Figure 1.2: Cryo-EM model of Saccharomyces cerevisiae Polε heterotetramer Adapted from Asturias et al, 200659.

15

Figure 1.3: Ternary structure of Saccharomyces cerevisiae Pol2 catalytic domain bound to DNA and an incoming dATP The fingers (blue), palm (green), thumb (red), N-terminal (white), exonuclease (orange), and P (purple) subdomains of Pol2 are shown as ribbons. The DNA primer-template substrate is shown as a cartoon and is colored orange. The incoming dATP is shown as sticks and is colored cyan. Prepared using PDB code 4M8O63.

16

Chapter 2. Kinetic Mechanism of DNA Polymerization Catalyzed by Truncated Human DNA Polymerase ε

Reproduced in part with permission from Zahurancik, W.J., Klein, S.J., Suo, Z.

(2013) Kinetic Mechanism of DNA Polymerization Catalyzed by Human DNA

Polymerase ε. Biochemistry. 52, 7041−7049. Copyright 2013 American Chemical

Society. The full article is available at https://pubs.acs.org/doi/abs/10.1021/bi400803v.

Walter J. Zahurancik overexpressed and purified Polε exo-. W.J.Z. planned and performed the kinetic experiments with assistance from Seth J. Klein. W.J.Z. and Zucai

Suo analyzed the results. W.J.Z. wrote the initial draft of the manuscript. Z.S. conceived the research and modified the manuscript. This work was supported by National Institutes of Health grant (GM079403) and National Science Foundation grant (MCB0960961) to

Z.S. and an REU training grant (DBI-1062144) to S.J.K.

17

2.1 Introduction

Three DNA polymerases (Pols), α, δ, and ε, efficiently and accurately replicate the majority of eukaryotic genomes36. Specifically, the Polα-primase complex primes the leading- and lagging-strand synthesis during DNA replication74. Studies involving mutagenic derivatives of Polε and Polδ have demonstrated that Polε is primarily responsible for leading-strand synthesis while Polδ plays a major role in lagging-strand replication40,74,75.

The Polε holoenzyme is a heterotetramer comprised of a single catalytic subunit

(p261) and three smaller subunits (p59, p12, and p17). The N-terminal half of p261 contains the catalytic domain, while mapping studies have shown that the C-terminal half is necessary for interaction with the three smaller subunits58. Furthermore, in vitro DNA replication assays have shown that the catalytic domain fragment of p261 is as active as the Polε holoenzyme58. As with most replicative DNA polymerases, Polε possesses both

DNA polymerase and 3′→5′ exonuclease proofreading activities, with the latter increasing the accuracy of DNA replication beyond the intrinsic fidelity of the former76.

Among six families (A, B, C, D, X, and Y) of DNA polymerases, Polε belongs to the B- family. Due to a combination of high intrinsic fidelity and 3′→5′ proofreading activity, most B-family replicative DNA polymerases have fidelities in the range of 10-6 to 10-8, or one error per 106 to 108 nucleotide incorporations31. In addition to synthesizing the leading-strand during genomic replication, Polε is also involved in long-patch base excision repair in vivo54.

18

The kinetic mechanism of DNA polymerization catalyzed by the B-family replicative DNA polymerases such as bacteriophage T4 DNA polymerase9 and

Sulfolobus solfataricus PolB112 have been characterized by extensive pre-steady-state kinetic analysis. These polymerases catalyze nucleotide incorporation via an induced-fit mechanism which is outlined in Scheme 2.1. Within this mechanism, the polymerase first binds DNA and then an incoming nucleotide to form the ground-state ternary complex

(E•DNA•dNTP). Next, the enzyme undergoes a conformational change (E′•DNA•dNTP) prior to phosphodiester bond formation and product releasing steps.

Recently, pre-steady-state kinetics has been applied to investigate correct and incorrect nucleotide incorporation catalyzed by yeast Polδ21. However, the mechanism of nucleotide incorporation catalyzed by either Polδ or Polε of humans has not yet been established. Here, we report a minimal kinetic mechanism for correct nucleotide incorporation catalyzed by human Polε.

2.2 Materials and Methods

2.2.1 Materials

These chemicals were purchased from the following companies: [α-32P]dTTP and

[γ-32P]ATP from Perkin-Elmer Life Sciences (Boston, MA), dNTPs from Bioline

(Taunton, MA), Sp-dTTPαS and Sp-dATPαS from Biolog-Life Science Institute (Bremen,

Germany). The DNA substrate (D-1) in Figure 2.1A was prepared as previously described23.

19

2.2.2 Expression and Purification of Human Polε

An orf encoding residues 1-1189 of the p261 subunit of human Polε was inserted into a modified pGEX4T3 vector C-terminal to a GST tag and a TEV cleavage site to generate vector pGEX4T3(TEV)/Polε(1-1189)-His6. To generate an exonuclease-deficient mutant, two rounds of site-directed mutagenesis were used to make three amino acid substitutions

(D275A/E277A/D368A). The TEV protease expression vector pRK603 was described

77 previously . Vectors pGEX4T3(TEV)/Polε exo-(1-1189)-His6 and pRK603 were co- transformed into Escherichia coli expression strain BL21(DE3) Rosetta cells. The LB cultures were grown in the presence of 25 µg/mL kanamycin and 50 µg/mL carbenecillin at 37 °C until the OD600 of the cells reached 0.6. The protein expression was then induced with 70 M IPTG. The cells were harvested and lysed by French press. The cell lysate was then clarified by ultracentrifugation (160,000 g at 4 °C for 40 minutes). Polε exo- was isolated from the cleared lysate by nickel ion affinity chromatography, followed by a

MonoQ anion-exchange column, and finally by a Sephacryl S-200 HR size-exclusion column. Polε exo- was determined to be ~95% homogeneous by SDS-PAGE analysis.

The concentration of purified Polε exo- was measured by UV spectrometry at 280 nm using a calculated extinction coefficient of 156760 M-1 cm-1.

2.2.3 Pre-Steady-State Kinetic Assays

All assays using Polε exo- were performed at 20 °C using reaction buffer E (50 mM Tris-OAc, pH 7.4 at 20 °C, 8 mM Mg(OAc)2, 1 mM DTT, 10% glycerol, 0.1 mg/mL

BSA, and 0.1 mM EDTA). Fast reactions were carried out using a rapid chemical quench flow apparatus (KinTek). Notably, to prevent any residual cleavage of DNA by Polε exo-

20

, Polε exo- was pre-incubated with DNA in the absence of Mg2+, the catalytic metal ion14.

All reported concentrations are final.

2.2.4 Measurement of the Rate Constant of DNA Dissociation from the Binary Complex

A pre-incubated solution of Polε exo- (50 nM) and 5′-radiolabeled D-1 (100 nM) was mixed with an unlabeled D-1 trap (2.5 µM) for 10 s to 20 min. After the mixing period, dTTP (133 µM) was added to the reaction mixture to initiate nucleotide incorporation. The reaction mixture was then allowed to incubate for an additional 15 s prior to be quenched by 0.37 M EDTA. Products were analyzed by sequencing gel analysis. The product concentrations were plotted against mixing times and the data were fit to the following equation: [product] = Aexp( - kt) + C, where A is the product concentration in the absence of the DNA trap and k is the DNA dissociation rate constant

32 (k-1), and C is the P-labeled product concentration in the presence of a trap for unlimited time.

2.2.5 Active Site Titration Assay

DNA The equilibrium dissociation constant (Kd ) of the Polε exo-•DNA binary complex was determined by an active site titration. A pre-incubated solution of Polε exo-

(64 nM, UV concentration) and increasing concentrations of 5′-radiolabeled D-1 DNA substrate (5 to 175 nM) was mixed rapidly with a solution containing dTTP (100 µM).

Each time point was quenched at 50 ms to ensure maximum product formation. All reactions were repeated in triplicate. The products were quantified via sequencing gel analysis.

21

2.2.6 Processive Polymerization Assay

The processivity of Polε exo-, calculated as the average rate constant of nucleotide incorporation divided by the average rate constant of dissociation of the binary complex (E•DNA), was measured via a processive polymerization assay. A pre-incubated solution of Polε exo- (60 nM, active site concentration) and 5′-radiolabeled D-1 (100 nM) was rapidly mixed with a solution containing dTTP, dGTP, and dCTP (100 µM each). At various time points, the reaction was quenched with EDTA to a final concentration of

0.37 M. The products were separated by sequencing gel electrophoresis and quantified by computer simulation.

2.2.7 Elemental Effect of Nucleotide Incorporation

To measure the elemental effect of correct nucleotide incorporation, a pre- incubated solution of Polε exo- (40 nM, active site concentration) and 5′-radiolabeled D-1

(20 nM) was rapidly mixed with dTTP or Sp-dTTPαS (1.25 µM) in buffer E and quenched with 0.37 M EDTA at various time points. To measure the elemental effect of incorrect nucleotide incorporation, the same pre-incubated Polε exo- and D-1 solution was rapidly mixed with dATP or Sp-dATPαS (600 µM). The Sp isomers were used due to previously observed stereoselectivity78. All products were quantified by sequencing gel analysis.

2.2.8 Pulse-Chase and Pulse-Quench Experiments

To observe an enzyme conformational change preceding the chemistry step, a pre- incubated solution of Polε exo- (50 nM, active site concentration) and unlabeled D-1 (50 nM) was rapidly mixed with [α-32P]dTTP (2.5 µM) for time points ranging from 4 to 650

22 ms. In the pulse-quench assay, reactions were quenched immediately with 1 M HCl. In the pulse-chase assay, reactions were chased with 1 mM unlabeled dTTP and were subsequently quenched with 1 M HCl after 30 s. For both the pulse-chase and the pulse- quench assays, acid-quenched mixtures were treated with chloroform and then neutralized in 1 M NaOH prior to analysis by PAGE.

2.2.9 Product Analysis

Reaction products were separated by denaturing PAGE (17% acrylamide, 8 M urea, and 1x TBE running buffer) and quantified by using a Typhoon TRIO (GE

Healthcare) and ImageQuant (Molecular Dynamics). For product analysis of the pulse- chase and pulse-quench assays, reaction products were separated using a 20% highly cross-linked polyacrylamide gel matrix as described previously79.

2.2.10 Data Analysis

Data were fit by nonlinear regression using Kaleidagraph (Synergy Software).

Data from the burst assay was fit to eq 2.1,

[product] = A[1 - exp( - k1t) + k2t] (eq 2.1)

where A is the amplitude of active enzyme, k1 is the observed burst rate constant, and k2 is the observed steady-state rate constant.

Data from the experiments performed under steady-state conditions were fit to eq

2.2

23

[product] = kssE0t + E0 (eq 2.2)

where kss is the steady-state rate constant of dNTP incorporation at the initial active enzyme concentration of E0

For the active site titration experiment, the resulting product concentration

(equivalent to burst amplitude) was plotted versus the concentration of D-1 and the data were fit to eq 2.3

DNA DNA 2 1/2 [E•DNA] = 0.5(Kd + E0 + D0) – 0.5[(Kd + E0 + D0) – 4E0D0] (eq 2.3)

DNA where Kd represents the equilibrium dissociation constant for the binary complex

(E•DNA), E0 is the active enzyme concentration, and D0 is the DNA concentration.

Data from the experiments performed under single-turnover conditions were fit to eq 2.4

[product] = A[1 – exp( - kobst)] (eq 2.4)

where A is the amplitude of product formation and kobs is the observed single-turnover rate constant. The kobs values were plotted against dTTP concentrations and the data were fit to eq 2.5

dTTP kobs = kp[dTTP]/(Kd + [dTTP]) (eq 2.5)

24

dTTP where kp is the maximum rate constant of dTTP incorporation and Kd is the equilibrium dissociation constant of dTTP.

Data from the processive elongation of 21/41-mer to 27/41-mer were modeled by using KinTek Explorer (KinTek)80.

2.3 Results

2.3.1 Expression and Purification of an Exonuclease-Deficient Mutant of Human Polε

Due to the difficulty of preparing the four subunit Polε holoenzyme for pre- steady-state kinetic analysis, we chose to express and purify a 140-kDa N-terminal fragment of p261, the catalytic subunit of Polε, which has been characterized previously77. The truncation fragment spans from residues 1-1189 and contains all conserved polymerase and exonuclease motifs81. Previously, the polymerase function of this N-terminal fragment has been shown to be as active as that of the full-length catalytic subunit58. It is also known that the 140-kDa fragment possesses a very active 3′→5′ exonuclease function76. This exonuclease activity is expected to compete with the polymerization activity of the 140-kDa fragment and complicate the interpretation of kinetic data. To circumvent this issue, we prepared an exonuclease-deficient mutant of the 140-kDa fragment by substituting three highly conserved carboxylates D275, E277, and D368 with alanine (Figure 2.2). These residues are located within the Exo I and II sequence motifs, which are highly conserved among replicative, B-family DNA polymerases82. Substitution of these particular residues for alanine has resulted in successful generation of exonuclease-deficient B-family DNA polymerases12,83. The

25 purified triple mutant of the 140-kDa fragment, denoted as Polε exo-, was then tested for its polymerase and exonuclease activities. Polε exo- possessed the wild-type level of polymerase function but lacked any noticeable exonuclease activity even after an hour of its incubation with DNA and Mg2+ at 20 °C (data not shown). Thus, this triple mutant is suitable for pre-steady-state kinetic investigation of polymerization mechanism of Polε.

2.3.2 Burst Kinetics

Replicative DNA polymerases typically exhibit biphasic kinetics during single nucleotide incorporation7,9,12. A burst assay was used to observe whether Polε exo- also followed a similar biphasic pattern. Specifically, a pre-incubated solution of Polε exo-

(200 nM, UV concentration) and 5′-radiolabeled DNA substrate D-1 (120 nM, Figure

2.1A) was rapidly mixed with correct dTTP (1.25 µM) in buffer E (see Experimental

Procedures) for various times at 20 °C. Notably, 37 °C was not used because the rate

-1 constant of product formation catalyzed by Polε exo- (kp > 500 s ) was too fast to be accurately measured by using a rapid chemical quench apparatus. The time course for product formation reveals biphasic kinetics characterized by a burst of product formation followed by a slower linear phase (Figure 2.1B). This suggests that Polε exo-, as other kinetically characterized DNA polymerases7,9,10,12,13, likely follows the same minimal kinetic mechanism (Scheme 2.1) when catalyzing correct nucleotide incorporation4. The burst rate constant of nucleotide incorporation was determined to be 12 ± 2 s-1 while the linear phase occurred with a rate constant of 0.023 ± 0.006 s-1 (Figure 2.1B). To evaluate if the linear phase rate constant was the steady-state rate constant of nucleotide incorporation, we measured the steady-state rate constant directly. Polε exo- (1 nM,

26 active site concentration) was pre-incubated with a large excess of D-1 (250 nM) and then mixed with dTTP (100 µM) for various time intervals. The observed steady-state rate constant was 0.026 s-1 (Figure 2.1C), which is nearly identical to the rate constant of the linear phase in Figure 2.1B, suggesting that the linear phase was indeed the steady- state phase of product formation. The steady-state rate of polymerase-catalyzed nucleotide incorporation is usually representative of the slow dissociation rate of the

E•DNA binary complex, which limits the rate of multiple turnovers12. To confirm that

Polε exo- exhibited the same pattern, we directly measured the rate of DNA dissociation.

A pre-incubated solution of Polε exo- (50 nM) and D-1 (100 nM) were mixed with unlabeled trap D-1 (2.5 µM) for various times. Then dTTP (133 µM) was added for 15 s to allow extension of any labeled D-1 bound by Polε exo- (E•DNA). The data were fit to a single-exponential equation (see Experimental Procedures) to yield an E•DNA

-1 dissociation rate constant (k-1) of 0.021 ± 0.007 s (Figure 2.3, Table 2.1), which is

-1 similar to the measured steady-state rate constant of 0.023 ± 0.006 s . Thus, we conclude that the aforementioned steady-state or linear phase (Figure 2.1) was limited by DNA dissociation from the E•DNA binary complex.

2.3.3 Active Site Titration

The burst phase in Figure 2.1B suggests the formation of a stable binary complex

(E•DNA) which bound and incorporated dTTP rapidly during the first turnover. Thus, the

DNA equilibrium dissociation constant (Kd ) of the binary complex can be measured by titrating the active site of Polε exo- with varying concentrations of DNA based on the fact that the amount of E•DNA complex formed is given by the amplitude of the first

27 turnover7. A pre-incubated solution of Polε exo- (64 nM, UV concentration) and increasing concentrations of 5′-radiolabeled D-1 (5 to 175 nM) was rapidly mixed with dTTP (100 µM) and Mg2+ for 50 ms, allowing ample time for full product formation before being quenched by EDTA. Each reaction was repeated in triplicate and the mean product concentration (E•DNA) was plotted versus the DNA concentration. The data

DNA were then fit to a quadratic equation (eq 2.3) to yield a Kd of 79 ± 14 nM (Table 2.1) and an active enzyme concentration (E0) of 10.1 ± 0.9 nM (Figure 2.4). Based on the UV concentration, only 15.8% of Polε exo- was active although the protein was purified to greater than 95% homogeneity based on the result of SDS-PAGE (data not shown).

Henceforth, all experiments described below were performed using the active site concentration of Polε exo-. From the above measured E•DNA dissociation rate constant

-1 DNA (k-1) of 0.021 s , the second-order rate constant of DNA binding (k1 = k-1/Kd ) was calculated to be 2.7 x 105 M-1s-1 (Table 2.1).

2.3.4 Pre-Steady-State Kinetics of Correct dTTP Incorporation by Polε exo-

Following the DNA binding event, Polε exo- binds an incoming nucleotide to form the ground-state E•DNA•dNTP ternary complex. To the measure both the maximum rate constant of correct nucleotide incorporation (kp) catalyzed by Polε exo- as

dTTP well as the equilibrium dissociation constant (Kd ) for a correct incoming nucleotide, we carried out a series of single-turnover experiments with increasing concentrations of correct dTTP. A pre-incubated solution of Polε exo- (80 nM) and 5′-radiolabeled D-1 (20 nM) was mixed rapidly with dTTP (0.625 to 200 μM) for various times. For each dTTP concentration, the product concentration was plotted against reaction time and the data

28 were fit to a single-exponential equation (eq 2.4) to yield kobs (Figure 2.5A). The kobs values were then plotted against the corresponding dTTP concentrations, and the data

-1 dTTP were fit to a hyperbolic equation (eq 2.5) to yield a kp of 248 ± 6 s (k3) and a Kd of

31 ± 2 µM (Table 2.1, Figure 2.5B). Furthermore, if the association rate constant (k2) of a small molecule dTTP approaches the diffusion limit of 1.0 x 108 M-1s-1, the upper-limit of

dTTP -1 the dTTP dissociation rate constant (k-2 = k2Kd ) was calculated to be 3,100 s (Table

2.1), which is too fast to be measured by current pre-steady-state techniques.

2.3.5 Processive Polymerization

The processivity of a DNA polymerase is defined as the number of nucleotides incorporated by the polymerase per DNA binding event, which can be quantitatively estimated as the ratio of the polymerization rate constant divided by the DNA dissociation rate constant. To kinetically measure this ratio, a pre-incubated solution of

Polε exo- (60 nM) and 5′-radiolabeled D-1 (100 nM) was mixed with three nucleotides

(100 µM each) for various times to allow the primer 21-mer to be elongated into a 27- mer product. The DNA substrate was used in molar excess over Polε exo-, which was to ensure that the measured kinetics of sequential elongation steps were a function of single enzyme binding events. The time courses of D-1 and products (Figure 2.6) were fit to a mechanism consisting of six consecutive nucleotide incorporation and dissociation events by the computer simulation program KinTek Explorer80. The rate constants of each intermediate product formation was: 169 ± 6 s-1 for 22-mer, 20 ± 1 s-1 for 23-mer, 14 ± 1 s-1 for 24-mer, 79 ± 20 s-1 for 25-mer, 30 ± 5 s-1 for 26-mer, and 36 ± 4 s-1 for 27-mer.

Thus, the rate constant of polymerization is in the range of 14 to 169 s-1 with an average

29 of 58 s-1. The following DNA dissociation rate constants were also generated: 12 ± 1 s-1 for 22/41-mer, 2.2 ± 0.9 s-1 for 23/41-mer, 6.2 ± 4.0 s-1 for 24/41-mer, 4.0 ± 1.9 s-1 for

25/41-mer, 2.2 ± 1.1 s-1 for 26/41-mer, and 0.9 ± 0.2 s-1 for 27/41-mer. These DNA dissociation rate constants (0.9 to 12 s-1) are significantly larger than the above measured

E•DNA dissociation rate constant (0.021 to 0.026 s-1) and reflected the rates for the dissociation of various E•DNA•dNTP complexes during processive polymerization12.

Similar fast DNA dissociation rates were also observed for other replicative polymerases such as S. solfataricus PolB1, Staphylococcus aureus PolC, T4 DNA polymerase, and

HIV-1 RT6,9,12,84. Based on the average rate constants of polymerization (58 s-1) and

-1 E•DNA•dNTP dissociation (5 s , Table 2.1, k7), the processivity of Polε exo- was calculated to be 11.

2.3.6 Elemental Effect of Nucleotide Incorporation

The observation of a distinct burst phase in Figure 2.1B suggests that a single nucleotide incorporation cycle is limited by either phosphodiester bond formation (step 4 in Scheme 2.1), a conformational change preceding phosphodiester bond formation (step

3 in Scheme 2.1), or both. To probe the rate-limiting step, single-turnover measurements were carried out with both dTTP and Sp-dTTPαS. Sp-dTTPαS contains a non-bridging phosphothioate substitution at the α-phosphate where bond breakage and formation take place. A pre-incubated solution of Polε exo- (40 nM) and 5′-radiolabeled D-1 (20 nM) was rapidly mixed with dTTP or Sp-dTTPαS (1.25 µM). Data were fit to a single-

-1 -1 exponential equation (eq 2.4) to give kobs values of 9 ± 1 s and 10 ± 1 s for the incorporation of dTTP and Sp-dTTPαS, respectively (Figure 2.7A). An elemental effect

30 of 0.9 for correct incorporation of dTTP was calculated from these kobs values.

Previously, an elemental effect of 4-11 has been used as a suggestive evidence for a rate- limiting chemistry step85. Thus, our result suggests that the rate of correct nucleotide incorporation catalyzed by Polε exo- is probably not limited by step 4 in Scheme 2.1.

To determine whether step 3 or 4 in Scheme 2.1 was rate-limiting for incorrect nucleotide incorporation, kobs was measured for the incorporation of incorrect dATP or its analog Sp-dATPαS (600 µM) into D-1. The kobs values were determined to be 0.25 ± 0.02

-1 -1 s and 0.0015 s for the incorporation of dATP and Sp-dATPαS, respectively, giving an elemental effect of 167 (Figure 2.7B). This large elemental effect suggests that the chemistry step during incorrect nucleotide incorporation catalyzed by Polε exo- is likely rate-limiting85. Similar elemental effect values for correct and incorrect nucleotide incorporations have previously been obtained for S. solfataricus PolB112. However, a large elemental effect may be due to steric clash between active site residues and the polymerase-bound nucleotide caused by the sulfur atom substitution4. Thus, more evidence is required to identify the rate-limiting step for correct or incorrect nucleotide incorporation catalyzed by Polε exo-.

2.3.7 Pulse-Chase and Pulse-Quench Experiments

To provide more conclusive evidence for diagnosing the rate-limiting step, we performed pulse-chase and pulse-quench assays which have been used to identify the rate-liming protein conformational change (step 3 in Scheme 2.1) for other DNA polymerases4,12,13,86. If the conformational change step also exists with Polε exo-, we would expect to see an increase in product formation for the pulse-chase assay relative to

31 the pulse-quench assay. A pre-incubated solution of Polε exo- (50 nM) and unlabeled D-1

(50 nM) was mixed rapidly with [α-32P]dTTP (2.5 µM) for various times. In the pulse- chase assay, each reaction was chased with a large excess of unlabeled dTTP (1 mM) for an additional 30 s before being quenched with 1 M HCl. In the pulse-quench assay, the reactions were quenched immediately with 1 M HCl without the addition of unlabeled dTTP.

During each pulse-chase assay, the E•DNA•dNTP complex that was in equilibrium with the forward and reverse intermediate species would presumably be pushed in the forward direction by the chase with a large molar excess of unlabeled dTTP, thus increasing product formation. This result would suggest that the chemistry step is preceded by a stable E•DNA•dNTP ternary complex that is distinct from the ground-state ternary complex. A product formation rate constant of 69 ± 8 s-1 and amplitude of 21.7 ± 0.7 nM were determined for the pulse-chase assay, whereas they were 66 ± 6 s-1 and 15.3 ± 0.5 nM for the pulse-quench assay (Figure 2.8). Since the reactions in the two sets of assays are the same, the incorporation rates were expected to be equal and this condition was met based on almost identical measured product formation rates. In contrast, the amplitude difference of 6.4 nM (42%) suggests the existence of a stable and distinct complex that accumulated prior to the chemistry step and was chased into the product by large molar excess of cold dTTP. The stable intermediate could be E•DNA, E•DNA•dNTP, or E′•DNA•dNTP in Scheme 2.1.

However, it was not the E•DNA binary complex since it would bind cold dTTP under the pulse-chase conditions, form undetectable cold product, and lead to no change in reaction

32 amplitude between the pulse-chase and pulse-quench assays. Similarly, the stable intermediate was not the E•DNA•dNTP complex either since the 42% amplitude difference would require that the complex partitioned between product formation (58%,

66 s-1), dTTP dissociation to form the E•DNA binary complex (calculated as [66/(58%) -

66 - 5] = 43 s-1 based on the kinetic partitioning from E•DNA•dNTP to E•product,

E•DNA, and E + DNA + dNTP), and the dissociation of E•DNA•dNTP to form E, DNA,

-1 and dNTP (average rate constant = 5 s , Scheme 2.1, k7). In comparison, the rate constant for the E•DNA•dNTP dissociation (k-2 in Scheme 2.1) to form the E•DNA complex has been estimated to be 3,100 s-1 (Table 2.1) which is much larger than 43 s-1. Thus, the stable complex had to be E′•DNA•dNTP, not E•DNA•dNTP. Alternatively, this conclusion can be reached based on the facts that the formation of E•DNA•dNTP from

-1 E•DNA and dNTP was a rapid equilibrium and k-2 (3,100 s ) in Scheme 2.1 was much faster than the product or the stable intermediate formation rate constant (66 to 69 s-1) as measured in Figure 2.8. Lastly, in order for E′•DNA•dNTP (21.7 - 15.3 = 6.4 nM) to accumulate, a slow conformational change step (step 5 in Scheme 2.1) after the rapid chemistry step must occur and serve as a kinetic “roadblock”10. This step resulted in 15.3 nM of E′•DNAn+1•PPi (Scheme 2.1) and a calculated internal equilibrium constant of 2.4

(= 15.3/6.4)10.

2.4 Discussion

In this paper, we carried out pre-steady-state kinetic studies in order to establish a minimal kinetic mechanism for correct nucleotide incorporation catalyzed by Polε exo-, an N-terminal truncation fragment of the catalytic subunit of human Polε. The burst assay

33 revealed that correct dTTP incorporation exhibited biphasic kinetics characterized by a

“burst” phase for the first turnover followed by a slow linear phase (Figure 2.1B). The later phase was demonstrated to represent the steady-state phase of product formation during subsequent turnovers (Figure 2.1C). A DNA trap assay further confirmed that the steady-state phase was limited by the E•DNA dissociation. These kinetic assays suggest that Polε exo- catalyzes correct nucleotide incorporation by following a similar mechanism (Scheme 2.1) which has been established for numerous DNA polymerases and reverse transcriptases4. With this mechanism, Polε exo- first binds a DNA substrate to form the E•DNA binary complex and then an incoming nucleotide to form the

E•DNA•dNTP ground-state ternary complex. Through the active site titration assay, the

DNA observed equilibrium dissociation constant Kd for the formation of E•DNA from Polε exo- and the DNA substrate D-1 (Figure 2.1A) was determined to be 79 nM (Figure 2.4).

dTTP The Kd for correct dTTP binding to E•DNA was then measured to be 31 µM (Figure

2.5B). To identify the rate-limiting step for single nucleotide incorporation during the first turnover, two lines of kinetic evidence were provided. First, a small elemental effect of 0.9 was determined for the incorporation of correct dTTP versus its thio-analog Sp- dTTPαS (Figure 2.7A), suggesting that the chemistry step was unlikely rate-limiting since the elemental effect fell significantly outside the elemental effect range of 4-11 for a rate-limiting chemical reaction85. In contrast, a much larger elemental effect (167) was determined for incorrect dNTP incorporation (Figure 2.7B). These elemental effect values suggest that correct and incorrect nucleotide incorporations did not share the same rate-limiting step and that the chemistry step (step 4 in Scheme 2.1) at least partially

34 controlled the rate of incorrect nucleotide incorporation. However, as noted previously, elemental effect measurements alone are not substantial pieces of evidence for the absence or presence of a rate-limiting conformational change preceding chemistry because an unusually large elemental effect or an elemental effect just below the expected range of 4-11 is difficult to interpret and the elemental effect is significantly affected by the exact active site structure of E′•DNA•dNTP4,78. To provide more definitive evidence for the rate-limiting step for correct dTTP incorporation, we performed pulse-chase and pulse-quench assays. There was a 42% reaction amplitude difference (Figure 2.8) between the pulse-chase and pulse-quench assays, suggested that a stable intermediate complex E′•DNA•dNTP accumulated prior to step 4 in Scheme 2.1.

The observed accumulation of E′•DNA•dNTP further indicates the presence of two slow steps (steps 3 and 5) that flank the fast chemistry step (step 4) in Scheme 2.110,13. Thus, step 3 limited dTTP incorporation in the forward direction while step 5 controlled the reverse polymerization direction or pyrophosphorolysis and served as a kinetic

“roadblock” (Scheme 2.1).

The processivity of Polε exo- was estimated to be 11 nucleotide incorporations per DNA binding event at 20 °C, given by the ratio of the average polymerization rate constant (58 s-1) divided by the average E•DNA•dNTP dissociation rate constant (5 s-1).

The calculated processivity of Polε exo- is remarkably lower than that of other replicative

DNA polymerases7,24 and is inadequate for catalyzing leading-strand synthesis in vivo.

However, the pre-steady-state rates used for calculating the processivity values for T7

35

DNA polymerase and human mitochondrial DNA polymerase were measured in the presence of their processivity cofactors. For example, E. coli thioredoxin, the cofactor of

T7 DNA polymerase, has been shown to increase the processivity of this replicative enzyme by more than two orders of magnitude87. The low processivity of Polε exo- is mainly contributed by the high dissociation rate constant measured in the absence of any cofactors. Previously, proliferating cell nuclear antigen (PCNA) has been shown to enhance DNA synthesis catalyzed by Polε58. Similarly, PCNA has also been demonstrated to significantly enhance the processivity of mammalian Polδ88,89. Taken together, PCNA is likely to be the processivity cofactor for Polε during leading-strand synthesis and we are currently investigating this possibility.

DNA Notably, the measured Kd (79 nM) of Polε exo- (Figure 2.4) is roughly 4- to

40-fold higher than those of DNA polymerases from the A-, B-, and Y-families6,7,10,12,13.

The low DNA binding affinity of Polε exo- could be caused by the absence of the smaller subunits of the Polε holoenzyme. Saccharomyces cerevisiae Dpb3p and Dpb4p, which are respectively homologous to the p12 and p17 subunits of the Polε holoenzyme, contain a double-stranded DNA binding site60. Thus, it is possible that the presence of p12 and p17 will enhance the DNA binding affinity of Polε. In addition, PCNA, which has been shown to stimulate Polε activity in vitro90, may also improve the DNA binding affinity of

Polε. Similar conclusions have been drawn for T4 DNA polymerase, which possesses a

DNA 9 comparable Kd of 70 nM in the absence of any replication factors , and yeast Polδ, which binds DNA with a 5-fold tighter affinity in the presence than in the absence of

PCNA91.

36

In summary, our kinetic analysis demonstrates that Polε exo-, like other replicative DNA polymerases, extends DNA via an induced-fit mechanism. The studies presented here will provide a foundation for future kinetic investigation of human Polε holoenzyme.

37

2.5 Schemes

Scheme 2.1: Minimal kinetic mechanism of DNA polymerization catalyzed by human DNA polymerase ε

38

2.6 Tables

Table 2.1: Kinetic parameters for correct nucleotide incorporation catalyzed by Polε exo- at 20 °C parameter value parameter value

-1 -1 -1 k1 0.27 µM s k-2 3,100 s

-1 dTTP k-1 0.021 s Kd 31 µM

DNA -1 Kd 79 nM k3 248 s

-1 -1 -1 k2 100 µM s k7 5 s

39

2.7 Figures A D-1 5′-CGCAGCCGTCCAACCAACTCA-3′ 3′-GCGTCGGCAGGTTGGTTGAGTAGCAGCTAGGTTACGGCAGG-5′

B C

Figure 2.1: Pre-steady-state and steady-state kinetics of correct dTTP incorporation into a DNA substrate D-1 by Polε exo- at 20 °C (A) The DNA substrate D-1 containing a 21-mer primer and a complementary 41-mer template. (B) A pre-incubated solution of Polε exo- (200 nM, UV concentration) and 5′- radiolabeled D-1 (120 nM) was rapidly mixed with dTTP (1.25 µM) and quenched at various time intervals with 0.37 M EDTA. The data were fit to eq 2.1 to yield a burst phase rate constant of 12 ± 2 s-1 and a linear phase rate constant of 0.023 ± 0.006 s-1. (C)

A pre-incubated solution of Polε exo- (1 nM, active site concentration) and 5′- radiolabeled D-1 (250 nM) was mixed with dTTP (100 µM) and quenched at various time intervals with 0.37 M EDTA. The data were fit to eq 2.2 to yield a steady-state rate constant of 0.026 s-1.

40

Figure 2.2: Conservation of Exo I and Exo II motifs across the B-family DNA polymerases Exonuclease motifs I and II, marked by the sequences DxE and NxxxF, respectively, where x is any amino acid residue, are highly conserved in the B-family. Polε is DNA polymerase ε from human, PolB1 is DNA polymerase I from Sulfolobus solfataricus, T4 is DNA polymerase from bacteriophage T4, and Rb69 is DNA polymerase from bacteriophage Rb69. Absolutely conserved residues are shown in purple, highly conserved residues are in red, and semi-conserved residues are in yellow.

41

Figure 2.3: DNA dissociation from the Polε•D-1 binary complex A pre-incubated solution of Polε exo- (50 nM) and 5′-radiolabeled D-1 (100 nM) was mixed with an unlabeled D-1 trap (2.5 µM) for various time before quenching with 0.37

M EDTA. The data were fit to the equation [product] = Aexp( - kt) + C to yield a dissociation rate constant (k) of 0.021 ± 0.007 s-1.

42

Figure 2.4: Active site titration of Polε exo- at 20 °C A pre-incubated solution of Polε exo- (64 nM, UV concentration) and increasing concentrations of 5′-radiolabeled D-1 (5 to 175 nM) was rapidly mixed with dTTP (100

µM). All reactions were quenched after 50 ms with 0.37 M EDTA. The product concentrations were plotted versus D-1 concentrations and the data were fit to eq 2.3,

DNA yielding a Kd of 79 ± 14 nM for the equilibrium dissociation constant of the Polε exo-

•D-1 complex and an enzyme amplitude of 10.1 ± 0.9 nM.

43

A B

Figure 2.5: Concentration dependence on the pre-steady-state rate constant of correct dTTP incorporation catalyzed by Polε exo- at 20 °C (A) A pre-incubated solution of Polε exo- (80 nM) and 5′-radiolabeled D-1 (20 nM) was mixed with increasing concentrations of dTTP [0.625 (●), 1.25 (●), 2.5 (●), 5 (●), 10 (●),

20 (●), 40 (●), 80 (●), and 200 (●) µM] for various times. The solid lines are the best fits to eq 2.4. (B) The observed rate constants of product formation (kobs) were plotted versus

-1 dTTP concentrations. The data were then fit to eq 2.5 to yield a kp of 248 ± 6 s and a

dTTP Kd of 31 ± 2 µM.

44

A

B

Figure 2.6: Processive polymerization catalyzed by Polε exo- at 20 °C A pre-incubated solution of Polε exo- (60 nM) and 5′-radiolabeled D-1 (100 nM) was rapidly mixed with dTTP, dCTP, and dGTP (100 µM) for various time intervals before being quenched by 0.37 M EDTA. Products were resolved by sequencing gel electrophoresis. (A) Gel image of the processivity assay. (B) The amount of remaining substrate [21-mer (●)] and each intermediate product [22-mer (●), 23-mer (●), 24-mer

(continued)

45

Figure 2.6 continued

(●), 25-mer (●), 26-mer (●), and 27-mer (●)] were plotted as a function of time. The solid lines represent the best fits generated by computer simulation using a model consisting of six single nucleotide incorporation events and six DNA dissociation events. The polymerization rate constants were 169 ± 6 s-1 for 22-mer, 20 ± 1 s-1 for 23-mer, 14 ± 1 s-

1 for 24-mer, 79 ± 20 s-1 for 25-mer, 30 ± 5 s-1 for 26-mer, and 36 ± 4 s-1 for 27-mer. The dissociation rate constants were 12 ± 1 s-1 for 22/41-mer, 2.2 ± 0.9 s-1 for 23/41-mer, 6.2

± 4.0 s-1 for 24/41-mer, 4.0 ± 1.9 s-1 for 25/41-mer, 2.2 ± 1.1 s-1 for 26/41-mer, and 0.9 ±

0.2 s-1 for 27/41-mer.

46

A B

Figure 2.7: Elemental effect on the rate constant of correct and incorrect nucleotide incorporation catalyzed by Polε exo- at 20 °C (A) A pre-incubated solution of Polε exo- (40 nM) and 5′-radiolabeled D-1 (20 nM) was rapidly mixed with 1.25 µM dTTP (●) or Sp-dTTPαS (■) in parallel time courses. The

-1 -1 data were fit to eq 2.4 to yield kobs values of 9 ± 1 s and 10 ± 1 s for dTTP and Sp- dTTPαS, respectively. The elemental effect was calculated to be 0.9. (B) The same Polε exo- and D-1 solution was rapidly mixed with 600 µM dATP (●) or Sp-dATP (■) in parallel time courses. The data were fit to eq 2.4 for dATP and a linear fit for Sp-dATPαS.

-1 -1 The kobs values were 0.25 ± 0.02 s and 0.0015 s for dATP and Sp-dATPαS, respectively. The elemental effect was calculated to be 167.

47

Figure 2.8: Pulse-chase and pulse-quench experiments at 20 °C A pre-incubated solution of Polε exo- (50 nM) and unlabeled D-1 (50 nM) was rapidly mixed [α-32P]dTTP (2.5 µM) for various time intervals. The pulse-quench (■) reaction mixtures were immediately quenched with 1 M HCl while the pulse-chase (●) reaction mixtures were chased with an excess of unlabeled dTTP (1 mM) for 30 s, followed by acid-quenching. The data were fit to eq 2.4 to yield a reaction rate constant of 69 ± 8 s-1 and a reaction amplitude of 21.7 ± 0.7 nM for the pulse-chase assay, and 66 ± 6 s-1 and

15.3 ± 0.5 nM for the pulse-quench assay.

48

Chapter 3. Significant Contribution of the 3′→5′ Exonuclease Activity to the High Fidelity of Nucleotide Incorporation Catalyzed by Human DNA Polymerase ε

Reproduced in part with permission from Zahurancik, W. J., Klein, S.J., and Suo,

Z. (2014) Significant Contribution of the 3’→5′ Exonuclease Activity to the High

Fidelity of Nucleotide Incorporation Catalyzed by Human DNA Polymerase ε. Nucleic

Acids Research. 42 (22), 13853-60. Copyright 2014 Oxford University Press. The full article is available at https://academic.oup.com/nar/article/42/22/13853/2411276.

Walter J. Zahurancik planned and performed the kinetic experiments with assistance from Seth J. Klein. W.J.Z. analyzed the results and wrote the initial draft of the manuscript. Zucai Suo conceived the research and modified the manuscript. This work was supported by National Institutes of Health [ES009127 to Z.S., T32 GM008512 to

W.J.Z.]; National Science Foundation [MCB0960961 to Z.S.]; REU [DBI-1062144 to

S.J.K.]. Funding for open access charge: the National Science Foundation [MCB-

0960961].

49

3.1 Introduction

DNA polymerases (Pols) perform a wide variety of biological functions that are critical to the proliferation and maintenance of genomic DNA including DNA replication,

DNA repair, and translesion DNA synthesis. DNA polymerases are organized into seven families (A, B, C, D, X, Y, and RT) and they share a structurally similar polymerase core consisting of finger, palm, and thumb domains that together form a right-hand geometry92-94. Besides the conserved polymerase core, DNA polymerases from different families possess additional domains and structural features that broaden their functional diversity in vivo. For instance, many replicative A- and B-family DNA polymerases possess a 3′→5′ exonuclease domain containing conserved carboxylate residues that are required for coordinating divalent metal ions to catalyze the excision of mismatched bases from the primer 3′ terminus94-98.

Highly accurate DNA synthesis is critical for eukaryotic genome replication and stability. To ensure that DNA is faithfully copied from generation to generation, cells employ high fidelity DNA polymerases that make only a single error per 106-108 nucleotide incorporation events9,28,29,31. Kinetically, the polymerase active site alone in a replicative DNA polymerase has been found to exhibit a nucleotide selectivity of 104-107

8,21,25,28,29. It was originally hypothesized that the amplification of free energy differences between correct and incorrect nucleotide incorporation by DNA polymerases was sufficient to account for the fidelity of DNA replication99. More recently, measured energetic difference between correct and incorrect nucleotide incorporation by three

DNA polymerases account for most of the high fidelity displayed by these enzymes100.

50

Overall, nucleotide selection by DNA polymerases is guided by a wide variety of factors, such as base stacking101, nucleotide desolvation102, induced-fit conformational changes8, and shape complementarity101. In addition to the contributions of these factors to DNA polymerase fidelity, the 3′→5′ proofreading activity found in most A- and B-family DNA polymerases further improves the fidelity of DNA replication by as much as 200- fold5,29,32.

In eukaryotes, three replicative DNA polymerases from the B-family, Polα, Polδ, and Polε, are responsible for the majority of DNA replication44. Human Polε (hPolε) is a heterotetramer, consisting of a catalytic subunit, p261, as well as three smaller subunits: p59, p12, and p1758. Though the structure of hPolε remains elusive, the crystal structure of the truncated catalytic subunit of yeast Polε (yPolε) was recently solved and shows the canonical right-hand configuration consisting of finger, thumb, and palm domains in addition to an N-terminal domain and a 3′→5′ exonuclease domain. Surprisingly, the palm domain of yPolε was found to contain additional structural elements, including a previously unidentified “P domain” which may play a role in aiding processive DNA synthesis catalyzed by Polε62,63.

Genetic studies have shown that Polε is primarily responsible for synthesizing the leading strand during DNA replication39-41,75. To serve this role effectively, Polε must be able to synthesize DNA efficiently and accurately. Recently, our lab utilized pre-steady- state kinetics to elucidate a minimal kinetic mechanism of correct nucleotide incorporation catalyzed by an exonuclease-deficient version of the N-terminal fragment

(residues 1-1189) of the catalytic subunit p261 of hPolε (hPol exo-)103. Our studies

51 reveal that hPolε inserts the correct nucleotide via an induced-fit mechanism and the rate- determining step is a protein conformational change step that occurs prior to phosphodiester bond formation. The proposed kinetic mechanism has been observed in most kinetically characterized DNA polymerases7,9,10,12-16. For hPol exo-, forward mutation assays estimated that it has a base substitution fidelity of 10-5, which is similar to the background of the assays and thus the error rate may even be overestimated77.

However, the overall fidelity of hPolε, as a function of its two enzymatic functions, has not yet been determined through pre-steady-state kinetic methods. In this paper, we determined the base substitution fidelity of hPol exo- using pre-steady-state kinetic methods. Moreover, we investigated the contributions of mismatch extension and exonuclease activity to the overall fidelity of the wild-type, exonuclease-proficient N- terminal fragment of p261 of hPolε (hPol exo+).

3.2 Materials and Methods

3.2.1 Materials

The chemicals used for experiments were purchased from the following sources:

[-32P]ATP from Perkin-Elmer Life Sciences (Boston, MA); Optikinase from USB

(Cleveland, OH); and dNTPs from Bioline (Taunton, MA). Both the wild-type (hPolε exo+) and the exonuclease-deficient triple mutant (D275A/E277A/D368A, hPolε exo-) forms of the truncated hPolε catalytic subunit were overexpressed and purified as described previously103.

52

3.2.2 DNA Substrates

The DNA substrates listed in Table 3.1 were purchased from Integrated DNA

Technologies, Inc. (Coralville, IA) and purified as described previously23. The 21- and

22-mer primer strands were 5′-radiolabeled by incubation with [-32P]ATP and

Optikinase for 3 hours at 37 °C, and then purified from free [-32P]ATP by passing through a Bio-Spin 6 column (Bio-Rad). The 5′-radiolabeled primers were then annealed to the 41-mer templates by incubating the primer with a 1.15-fold excess of template at

95 °C for 5 minutes before cooling slowly to room temperature over several hours.

3.2.3 Polymerase and Exonuclease Single-Turnover Assays

All assays using hPol exo- or hPolε exo+ were performed at 20 C in reaction buffer E (50 mM Tris-OAc, pH 7.4 at 20 C, 8 mM Mg(OAc)2, 1 mM DTT, 10% glycerol, 0.1 mg/mL BSA, and 0.1 mM EDTA). Fast reactions were carried out using a rapid chemical quench-flow apparatus (KinTek). Notably, all reactions were performed at

20 °C since the rate constant for correct nucleotide incorporation at 37 °C was too fast (kp

> 500 s-1) to be measured accurately by using the rapid chemical quench-flow apparatus.

For polymerization single-turnover assays, a pre-incubated solution of hPolε exo- (260 nM) and a 5′-radiolabeled DNA substrate (20 nM) in buffer E was rapidly mixed with

Mg2+ (8 mM) and varying concentrations of dNTP. For exonuclease assays, a pre- incubated solution of hPolε exo+ (200 nM) and a 5′-radiolabeled DNA substrate (20 nM) in buffer E was rapidly mixed with Mg2+ (8 mM) in the absence of nucleotide to initiate the excision reaction. All reactions were quenched with the addition of 0.37 M EDTA.

All reported concentrations are final. Most data, unless otherwise specified, were

53 collected from single trials due to insufficient amount of hPolε to repeat each measurement in triplicate.

3.2.4 Product Analysis

Reaction products were separated by denaturing PAGE (17% acrylamide, 8 M urea, and 1X TBE running buffer) and quantified using a Typhoon TRIO (GE

Healthcare) and ImageQuant (Molecular Dynamics).

3.2.5 Data Analysis

All kinetic data were fit by nonlinear regression using KaleidaGraph (Synergy

Software). Data from polymerization assays under single-turnover conditions were fit to eq 3.1

[product] = A[1 – exp(- kobst)] (eq 3.1)

where A is the amplitude of product formation and kobs is the observed single-turnover rate constant.

Data from the plot of kobs versus dNTP concentration were fit to eq 3.2

kobs = kp[dNTP]/(Kd + [dNTP]) (eq 3.2) where kp is the maximum rate constant of nucleotide incorporation and Kd is the equilibrium dissociation constant for dNTP binding. When Kd is very large, the plot of kobs versus dNTP concentration was fit to eq 3.3

54 kobs = (kp/Kd)[dNTP] (eq 3.3)

to yield the substrate specificity constant, kp/Kd.

Data from exonuclease assays under single-turnover conditions were fit to eq 3.4

[product] = A[exp(- kexot)] + C (eq 3.4)

where A is the reaction amplitude and kexo is the overall DNA excision rate constant.

All reported errors were generated by fitting the data to the above equations through Kaleidagraph.

3.3 Results

3.3.1 Substrate Specificity of hPolε exo-

In our recent publication we revealed through pre-steady-state kinetics that hPolε, like all other kinetically characterized polymerases, catalyzes correct nucleotide incorporation via an induced-fit mechanism103. At 20 C, hPolε exo- binds and

-1 incorporates correct dTTP opposite dA with a maximum rate constant, kp, of 248 s and

103 an equilibrium dissociation constant, Kd, of 31 µM . However, kp and Kd for an incorrect incoming nucleotide have not yet been determined. We expected that hPolε, like other replicative DNA polymerases, exhibits high selectivity for correct incoming nucleotides versus incorrect nucleotides through the combination of both a faster incorporation rate constant and a higher ground-state binding affinity (1/Kd). To confirm this hypothesis, we measured the substrate specificities (kp/Kd) for each of the 15

55 remaining possible incoming nucleotide and templating base combinations through four perfectly matched DNA substrates (D-1, D-6, D-7, and D-8) listed in Table 3.1. As examples, the plots of kobs versus dNTP concentration for the extension of the 21-mer primer in D-6 are shown for correct dCTP and incorrect dATP in Figures 3.1A and 3.1B, respectively. The plot in Figure 3.1A was fit to eq 3.2 (see Experimental Procedures) to

-1 -1 -1 obtain a kp of 268±14 s and a Kd of 19±4 µM as well as a calculated kp/Kd of 14 µM s for correct dCTP incorporation. Likewise, the plot in Figure 3.1B was fit to eq 3.2 to

-3 -1 2 -6 -1 -1 yield a kp of (8.8±0.4) x 10 s , a Kd of (9±1) x 10 µM and a kp/Kd of 9.8 x 10 µM s for incorrect dATP incorporation. Similarly, the kinetic parameters for all other combinations of nucleotides and templating bases were determined at 20 °C and are listed in Table 3.2. Notably, the kp and Kd values for dCTP misincorporation opposite dC could not be determined due to the extremely weak binding affinity (>2 mM) of the incorrect dCTP. In this case, the plot of kobs versus dCTP concentration (data not shown) was fit to

-5 -1 -1 a linear equation (eq 3.3) to give the corresponding kp/Kd value (1.5 x 10 µM s , Table

-4 3.2). Overall, the base substitution fidelity (Fpol) of hPolε exo- was determined to be 10 -

10-7 (Table 3.2).

3.3.2 Mismatch Extension Fidelity of hPolε exo-

After a misincorporation event, hPolε will excise the nascent mismatched base pair, dissociate from the DNA substrate, or further extend the mismatched base pair.

Following selective inhibition of its 3′→5′ exonuclease activity by mutating three highly conserved carboxylate residues (D275/E277/D368) at the exonuclease active site to

103 alanine , we were able to determine the kp/Kd values for the incorporation of both a

56 correct nucleotide and an incorrect nucleotide on DNA substrates containing a single mismatched base at the primer 3′ terminus (M-1, M-7 and M-8 in Table 3.1). As an example, the plot of kobs versus dCTP concentration for the extension of M-7 (Figure 3.2)

-2 -1 was fit to eq 3.2 (see Experimental Procedures) to yield a kp of (4.3±0.4) x 10 s and a

3 Kd of (1.6±0.2) x 10 µM. Notably, M-7 contains a C:T mismatch at the primer-template junction, but is otherwise identical to the four correctly matched DNA substrates (D-1, D-

6, D-7, and D-8 in Table 3.1). Interestingly, both correct dCTP and incorrect dGTP with

M-7 had very low substrate specificities which were comparable to the values measured for incorrect nucleotide incorporation into a correctly matched DNA substrate (Table

3.3). Similarly, the kinetic parameters for correct dCTP and incorrect dGTP incorporation into the other two mismatched DNA substrates, M-1 and M-8, in Table 3.1 at 20 °C were determined and are listed in Table 3.3.

3.3.3 Excision of Matched and Mismatched DNA Substrates by hPolε exo+

hPolε, like most A- and B-family replicative DNA polymerases, possesses a

3′→5′ exonuclease proofreading activity that is proficient in removing mismatched bases from the primer 3′ terminus. It is expected that the exonuclease activity of hPolε will be kinetically favored over its polymerase activity in the presence of a mismatched primer terminus due to a significantly higher rate of excision versus extension. On the other hand, excision of a matched base pair should be much slower than correct nucleotide incorporation to prevent futile competition with 5′→3′ primer extension during processive DNA synthesis. To verify this hypothesis, we measured the overall excision rate constants (kexo) of matched versus mismatched base pairs by hPolε exo+. The D-8

57 and M-8 substrates (Table 3.1) were used to measure the kexo values for a matched and mismatched primer-template pair, respectively. The concentration of remaining substrate was plotted versus time and the data were fit to eq 3.4 (see Experimental Procedures) to

-1 -1 yield kexo (Figure 3.3). The kexo values were determined to be 0.17±0.02 s and 3.0±0.7 s for matched (D-8) and mismatched (M-8) primer-template pairs at 20 °C, respectively.

These measurements were repeated at a lower enzyme concentration and kexo was found to be unaffected by the ratio of hPolε exo+ to DNA (data not shown). Notably, the measured kexo is not the true excision rate constant at the exonuclease active site (kx) since it is a function of kx, the forward and backward transfer rates of the primer 3-terminal nucleotides between the polymerase and exonuclease active sites, and DNA dissociation and rebinding rates from the exonuclease active site. Similarly, we measured kexo for the mismatched DNA substrates M-1 and M-8 (Table 3.1) and the kexo values are listed in

Table 3.3. Interestingly, Table 3.3 shows that the overall rate constant of excision was not significantly affected by the identity of the 3′ mismatched base pair in a DNA substrate.

3.4 Discussion

To determine if hPolε synthesizes DNA with high fidelity as observed with other replicative DNA polymerases, we used pre-steady-state kinetics to measure the kinetic parameters of nucleotide incorporation and excision on both matched and single-base mismatched DNA substrates. First, we calculated the base substitution fidelity of hPolε exo- by measuring the kp and Kd values at 20 °C for all 16 possible combinations of incoming nucleotides and templating bases. Correct nucleotides were incorporated with

-1 an average kp and Kd of 252 s and 23 µM, respectively. The kp values for incorrect

58 nucleotide incorporation varied widely from (8.8±0.4) x 10-3 s-1 to 5.2±0.9 s-1 while the

2 3 Kd values ranged between (2.4±0.3) x 10 to (2.0±0.6) x 10 µM. Strikingly, the kp difference between correct and incorrect nucleotide incorporation [(kp)correct/(kp)incorrect] contrasts broadly, varying by one to four orders of magnitude. A similar result was previously obtained from pre-steady-state kinetic analysis of hPolγ exo-25. Overall, hPolε exo- incorporated a correct nucleotide with a 48- to 3.0 x 104-fold faster rate constant than an incorrect nucleotide, and bound a correct nucleotide with a 10- to 100-fold higher affinity. Thus, the differences in both kp and Kd were major determinants of the base substitution fidelity of hPolε exo-, which was calculated to be 10-4-10-7 (Table 3.2).

Similar kinetic patterns of incorrect nucleotide discrimination were determined for other highly accurate replicative DNA polymerases, including hPolγ, T7 DNA polymerase, and

RB69 DNA polymerase8,25,104. Interestingly, all DNA polymerases including hPolε exo-

(Table 3.2) possess sequence-dependent base substitution fidelity.

The fidelity of DNA synthesis catalyzed by replicative DNA polymerases is further enhanced by an associated 3′→5′ exonuclease proofreading activity that selectively excises mismatched base pairs. We calculated the contribution of proofreading (Fexo) to the fidelity of DNA synthesis catalyzed by hPolε by taking the ratio of the overall rate constant of mismatch excision (kexo) versus the rate constant of mismatch extension at a typical intracellular nucleotide concentration of 100 μM (kobs).

For example, in the case of correct dCTP incorporation onto the mismatched M-1

-2 -1 substrate by hPolε exo-, the kp and Kd values were determined to be (4.0±0.4) x 10 s

2 and of (5.4±1.3) x 10 μM, respectively (Table 3.3). Using eq 3.2, kobs was calculated to

59

-1 - be 0.0062 s . For the same mismatched DNA substrate, the kexo was measured to be 2.2 s

1 with hPolε exo+ (Table 3.3). Thus, the contribution of proofreading to the overall fidelity of hPolε was calculated to be ~350-fold (Table 3.4). When factored together with the base substitution fidelity of hPolε exo- (10-4-10-7), the overall in vitro polymerization fidelity of hPolε was determined to be 10-6-10-9 with a C:A mismatch (M-1). It should be noted that incorrect incorporation over a mismatch is much slower and less efficient than correct incorporation and thus, misincorporations were not considered in the determination of Fexo (Table 3.3).

Interestingly, the substrate specificity for the next correct nucleotide with hPolε exo- varied widely depending on the identity of the single base mismatch (Table 3.3). A similar result was obtained for E. coli Klenow fragment which catalyzed mismatch extension with a rate constant that differed by as many as three orders of magnitude in a sequence dependent manner105. In contrast, the overall rate constant of mismatch excision by hPolε exo+ is not significantly affected. This is comparable to the observation that the rate constant of excision of a single base mismatch catalyzed by hPolγ is independent of mismatch identity32. As a consequence of both a highly variable extension rate constant and a similar excision rate constant, the 3′→5′ exonuclease activity of hPolε appears to enhance its overall fidelity by two to four orders of magnitude based on the mismatched bases (Table 3.3). For better comparison, the Fexo values were calculated for several other replicative DNA polymerases (Table 3.4). Notably, the Fexo values are much larger with hPolε exo+ than with S. solfataricus PolB1, hPolγ, and T7 DNA polymerase and this is beneficiary to faithful replication of the vast nuclear human genome. However, the rate

60 constants listed for extension of a primer containing a single base mismatch by S. solfataricus PolB1, hPolγ, and T7 DNA polymerase in Table 3.4 were determined only for one specific mismatched base pair. Therefore, it is possible that the Fexo for these replicative DNA polymerases, as observed with hPolε exo+, varies in a large range depending on the identity of the single base mismatch.

Though the 3′→5′ proofreading activity of hPolε is highly efficient at removing mismatched base pairs, the possibility that hPolε may partition toward removal of a correctly matched base pair must be considered. For example, the extension rate constant

(kp) on the D-8 substrate in the presence of the next correct nucleotide, dGTP, was

-1 measured to be 219±13 s (Table 3.2), while the overall excision rate constant (kexo) was

0.17±0.02 s-1 (Figure 3.3). Since typical cellular nucleotide concentrations (100 M) are significantly higher than the Kd value (9 M, Table 3.2) for dGTP with D-8, the dGTP incorporation rate constant should approach kp. Thus, the probability of matched base pair excision, given by kexo/(kexo + kp), was calculated to be only 0.08% while the probability of further extension kp/(kp + kexo) approached 100%. In contrast, for a single base mismatched terminus in a DNA substrate, the kinetic partitioning between excision kexo/(kexo + kobs) and extension kobs/(kobs + kexo) was calculated to be 99.719-99.991% and

0.009-0.281%, respectively (Table 3.4). Thus, the 3′→5′ proofreading activity of hPolε is very efficient at removing mismatched nucleotides without interfering with continuous faithful DNA synthesis.

From the combined contributions of both high polymerase selectivity (10-4-10-7,

Table 3.2) and efficient 3′→5′ proofreading activity (3.5 x 102 to 1.2 x 104, Table 3.3),

61 hPolε exhibits overall polymerization fidelity of 10-6-10-11 in vitro. Such high fidelity of

DNA synthesis qualifies hPolε as a main enzyme to catalyze accurate replication of large human nuclear genome (3 x 109 base pairs). As the key polymerase responsible for leading strand synthesis during nuclear genomic replication, hPolε must synthesize long stretches of DNA without making an error. Consistently, the fidelity of DNA replication in normal human cells was estimated to be 10-9-10-10 106-108. Strikingly, somatic mutations in the 3′→5′ exonuclease domain of hPolε impair the proofreading activity, cause a high frequency of errors (>10-4 mutations per base) in the leading strand, elevate recurrent nonsense mutation rates in key tumor suppressors such as TP53, ATM and PIK3R1, and ultimately lead to the formation of various cancers41. This error frequency is greater than the high end of the fidelity range of hPolε exo- (10-4-10-7) measured here. Such a discrepancy suggests other cellular factors also contribute to the high leading strand mutation rate in tumors carrying inactivating mutations of the proofreading domain of hPolε.

Notably, the lower limit (10-6) of the fidelity range of hPolε (10-6-10-11) is significantly higher than the error frequency of normal human genome replication (10-9-

10-10)106-108. It is likely that this difference is accounted for by post-replication mismatch repair in vivo, which enhances replication fidelity by one to three orders of magnitude in

E. coli and Saccharomyces cerevisiae61,108-111. Additionally, it is possible that interactions between the p261 catalytic subunit and the smaller subunits or other proteins in the replisome may further enhance the fidelity of DNA replication in vivo. To investigate this

62 hypothesis, we are currently studying the effect of the smaller subunits on the catalytic properties of p261 of hPolε.

63

3.5 Tables

Table 3.1: Sequences of DNA substrates 5′-CGCAGCCGTCCAACCAACTCA-3′ D-1 3′-GCGTCGGCAGGTTGGTTGAGTAGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCA-3′ D-6 3′-GCGTCGGCAGGTTGGTTGAGTGGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCA-3′ D-7 3′-GCGTCGGCAGGTTGGTTGAGTTGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCA-3′ D-8 3′-GCGTCGGCAGGTTGGTTGAGTCGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCAC-3′ M-1 3′-GCGTCGGCAGGTTGGTTGAGTAGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCAC-3′ M-7 3′-GCGTCGGCAGGTTGGTTGAGTTGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCAC-3′ M-8 3′-GCGTCGGCAGGTTGGTTGAGTCGCAGCTAGGTTACGGCAGG-5′

64

Table 3.2: Kinetic parameters for correct and incorrect nucleotide incorporation catalyzed by hPolε exo- at 20 °C

-1 -1 -1 a dNTP kp (s ) Kd (µM) kp/Kd (µM s ) Fpol

Template dA (D-1) dTTPb 248±6 31±2 8 dATP 0.61±0.04 (6±1) x 102 1.0 x 10-3 1.2 x 10-4 dCTP 5.2±0.9 (2.0±0.6) x 103 2.6 x 10-3 3.2 x 10-4 dGTP (1.13±0.04) x 10-2 (3.2±0.3) x 102 3.5 x 10-5 4.4 x 10-6 Template dG (D-6) dCTP 268±14 19±4 14 dTTP 0.63±0.06 (7±2) x 102 9.0 x 10-4 6.4 x 10-5 dATP (8.8±0.4) x 10-3 (9±1) x 102 9.8 x 10-6 7.0 x 10-7 dGTP (8.6±0.2) x 10-2 (2.4±0.3) x 102 3.6 x 10-4 2.6 x 10-5

Template dT (D-7) dATP 275±12 33±5 8 dTTP (4.7±0.4) x 10-2 (9±2) x 102 5.2 x 10-5 6.5 x 10-6 dCTP (7.4±0.6) x 10-2 (1.1±0.2) x 103 6.7 x 10-5 8.4 x 10-6 dGTP 0.58±0.06 (1.1±0.2) x 103 5.3 x 10-4 6.6 x 10-5 Template dC (D-8) dGTP 219±13 9±2 24 dTTP 3.1±0.3 (6±1) x 102 5.2 x 10-3 2.2 x 10-4 dATP 1.2±0.1 (9±2) x 102 1.3 x 10-3 5.4 x 10-5 dCTP - - 1.5 x 10-5 6.2 x 10-7 a Calculated as (kp/Kd)incorrect/[(kp/Kd)correct + (kp/Kd)incorrect]. bTable 2.1.

65

Table 3.3: Kinetic parameters for mismatch extension and excision catalyzed by hPolε exo- and hPolε exo+ at 20 °C k /K dNTP k (s-1) K (µM) p d F a k (s-1)b k (s-1) F c p d (µM-1s-1) ext obs exo exo C:A mismatch (M-1) dCTP (4.0±0.4) x 10-2 (5.4±1.3) x 102 7.4 x 10-5 6.2 x 10-3 - dGTP (3.6±0.3) x 10-4 (5.3±1.3) x 102 6.8 x 10-7 9.1 x 10-3 5.7 x 10-5 ------2.2±0.1 350 C:T mismatch (M-7) dCTP (4.3±0.4) x 10-2 (1.6±0.2) x 103 2.7 x 10-5 2.5 x 10-3 - dGTP (6.3±0.5) x 10-4 (6.4±1.0) x 102 9.8 x 10-7 3.5 x 10-2 8.5 x 10-5 ------2.9±0.3 1200 C:C mismatch (M-8) dCTP - - 2.6 x 10-6 2.6 x 10-4 - dGTP (6.1±0.3) x 10-4 (1.5±0.1) x 103 4.1 x 10-7 0.14 3.8 x 10-5 ------3.0±0.7 12000 a Calculated as (kp/Kd)incorrect/[(kp/Kd)correct + (kp/Kd)incorrect]. b Calculated as kp[dNTP]/(Kd + [dNTP]) during extension from a mismatched primer terminus at an intracellular nucleotide concentration of 100 μM. c dCTP Calculated as kexo/kobs .

66

Table 3.4: Comparison of the contribution of 3′→5′ exonuclease activity to the overall fidelity of replicative DNA polymerases when encountering a single base mismatch in the staggering end of a DNA substrate

-1 -1 a b c Polymerase Mismatch kexo (s ) kobs (s ) Fexo excision%

hPolεd C:A 2.2 6.2 x 10-3 350 99.719 C:T 2.9 2.5 x 10-3 1200 99.914 C:C 3.0 2.6 x 10-4 12000 99.991 S. solfataricus PolB1e A:A 1.86 0.012 160 99.359 hPolγf T:T 0.4 0.1 4 80.000 T7 DNA polymeraseg A:A 2.3 0.012 190 99.481 a Calculated as kp[dNTP]/(Kd + [dNTP]) during extension from a mismatched primer terminus at an intracellular nucleotide concentration of 100 μM. b Calculated as kexo/kobs. c Calculated as kexo/(kexo + kobs) for a single base mismatch. dThis work (performed at 20 °C). eReference 29 (performed at 37 °C). fReference 32 (performed at 37 °C). gReference 5 (performed at 20 °C).

67

3.6 Figures A B

Figure 3.1: Nucleotide concentration dependence on the pre-steady-state kinetic parameters of correct dCTP and incorrect dATP incorporation opposite dG catalyzed by hPolε exo- at 20 °C (A) A pre-incubated solution of hPolε exo- (260 nM) and 5′-radiolabeled D-6 (20 nM) was mixed with increasing concentrations of correct dCTP and Mg2+ for various times.

The plot of product concentration versus time was fit to eq 3.1 to yield kobs (data not shown). The resulting kobs values were plotted against dCTP concentration and fit to eq

-1 3.2 to yield a kp of 268±14 s and a Kd of 19±4 µM; (B) hPolε exo- and 5′-radiolabeled

D-6 were mixed with increasing concentrations of incorrect dATP and Mg2+ as described

-3 -1 above. The data were similarly processed to yield a kp of (8.8±0.4) x 10 s and a Kd of

(9±1) x 102 µM.

68

Figure 3.2: Extension of a mismatched base pair catalyzed by hPolε exo- at 20 °C A pre-incubated solution of hPolε exo- (260 nM) and 5′-radiolabeled M-7 (20 nM) was rapidly mixed with increasing concentrations of dCTP and Mg2+ for various times. The product concentration was plotted against time and fit to eq 3.1 to yield kobs (data not shown). The kobs values were plotted against dCTP concentration and fit to eq 3.2 to yield

-2 -1 3 a kp of (4.3±0.4) x 10 s and a Kd of (1.6±0.2) x 10 μM.

69

Figure 3.3: Excision of primers with matched and mismatched 3′ termini catalyzed by hPolε exo+ at 20 °C A pre-incubated solution of 200 nM of hPolε exo+ and 20 nM of 5′-radiolabeled D-8 (■) or M-8 (●) was rapidly mixed with Mg2+ for various times before been quenched with

0.37 M EDTA. The remaining substrate concentration was plotted versus time and fit to

-1 -1 eq 3.4 to yield a kexo of 0.17±0.02 s for the matched D-8 substrate and 3.0±0.7 s for the mismatched M-8 substrate.

70

Chapter 4. Comparison of the Kinetic Parameters of the Truncated Catalytic Subunit and Holoenzyme of Human DNA Polymerase ε

Reproduced in part with permission from Zahurancik, W. J., Baranovskiy, A.G.,

Tahirov, T.H., and Suo, Z.* (2015) Comparison of the Kinetic Parameters of the

Truncated Catalytic Subunit and Holoenzyme of Human DNA Polymerase ε. DNA

Repair. 29, 16-22. Copyright 2015 Elsevier. The full article is available at https://www.sciencedirect.com/science/article/pii/S1568786415000208.

Walter J. Zahurancik planned and performed the kinetic experiments and analyzed the results. Andrey G. Baranovskiy and Tahir H. Tahirov purified the hPolε holoenzyme. W.J.Z. wrote the initial draft of the manuscript with assistance from A.G.B.

Zucai Suo and T.H.T. conceived the research. Z.S. modified the manuscript. This work was supported by National Institutes of Health grants [ES024585 and ES009127 to

Z.S., GM101167 to T.H.T., and T32 GM008512 to W.J.Z.].

71

4.1 Introduction

Since its identification in budding yeast, DNA polymerase (Pol1) ε has been of major interest on account of its role in a wide variety of biological processes in eukaryotes. Polε, along with Polα and Polδ, is a key eukaryotic DNA replication enzyme44, and is believed to be the primary leading strand synthesis polymerase during nuclear genome replication based on findings in budding yeast40, fission yeast39, and humans41. In addition to its role in nuclear DNA replication, Polε has been implicated in cell cycle regulation45-48, gene silencing49,50, sister chromatid cohesion51,52, and base excision repair54.

Polε is a heterotetramer with an overall architecture that was shown to be conserved in humans56,58, budding yeast112, and African clawed frog113. The p261 catalytic subunit (Pol2 in yeast) consists of an N-terminal domain containing the conserved polymerase and 3′→5′ exonuclease subdomains114,115, as well as a C-terminal domain that is required for interaction with the three small subunits, p59, p12, and p17

(Dpb2, Dpb3, and Dpb4 in yeast)56,58,116,117. A low-resolution (20 Å) structure of the yeast

Polε heterotetramer obtained by cryo-electron microscopy has shown that Pol2 forms a globular head-like structure, while the three small subunits associate with Pol2 to form an extended tail-like structure that is suggested to interact with newly-synthesized double- stranded DNA (dsDNA)59. Recently, two ternary crystal structures of the N-terminal domain of Pol2 were solved and show that Pol2 possesses a novel P domain that makes additional contacts with the double-stranded region of the DNA substrate and contributes to processive DNA synthesis62,63.

72

Notably, only Pol2 and Dpb2 are essential in yeast, while deletions of Dpb3 and

Dpb4 are non-lethal55,118. Interestingly, only the C-terminal domain of Pol2 is essential as deletions of the entire N-terminal domain are viable, albeit with a prolonged S phase116,117. Bioinformatics tools have revealed that the C-terminal domain contains a distantly related copy of the exonuclease-polymerase module in which both enzymatic activities are nonfunctional57. Such inactive polymerase domains are likely to play a key structural role in assembly of replication complexes119,120. Taken together, these observations suggest that a critical role of Polε in yeast DNA replication involves protein-

DNA and protein-protein interactions at the replication fork mediated by the C-terminal domain of Pol2 and the Dpb2 subunit, and that the polymerase activity of Polε is important for timely replication fork progression.

Expression systems in yeast121,122 and insect cells123 have allowed for purification of the yeast Polε heterotetramer in sufficient quantities for studies of its catalytic properties. Furthermore, proteolysis of Polε in yeast generates a highly conserved124 and active N-terminal fragment of Pol2 that is readily isolated from full-length Polε76,122.

Purification of both forms has enabled investigation of the effects of the small subunits on the catalytic properties of yeast Polε. Biochemical assays have demonstrated that the

Dpb3-Dpb4 dimer is able to bind dsDNA and that the yeast Polε heterotetramer contains an additional DNA binding site that has similar affinity for single-stranded DNA

(ssDNA) and dsDNA60. Moreover, longer regions of dsDNA were previously shown to slightly increase the processivity of the yeast Polε heterotetramer, but not the Pol2

73 subunit alone59,61. These results suggest that the small subunits may contribute to the catalytic activity of yeast Polε in vivo.

In contrast, characterization of the catalytic activities of human Polε has been more limited. While the N-terminal domain of the human Polε catalytic subunit (p261N) can be readily overexpressed and purified from E. coli in quantities suitable for biochemical analysis77, thorough kinetic studies of the human heterotetramer of Polε

(hereafter referred to as hPolε) have been precluded by the difficulty of expressing comparable amounts of active protein. Recently, hPolε was obtained from a baculovirus expression system and was shown to be as active as and catalytically similar to the full- length p261 subunit and the p261N fragment under the reported conditions58. Using a similar approach, we have reconstituted and purified fully-assembled hPolε from a baculovirus-insect cell system. We then used pre-steady-state kinetics to measure kinetic parameters of hPolε for the first time. We find that the small subunits do not appear to affect DNA synthesis by hPolε but enhance DNA binding and decrease the 3′→5′ exonuclease activity, suggesting a potential role in regulating the proofreading activity of hPolε.

4.2 Materials and Methods

4.2.1 Materials

Materials used for experiments described below were purchased from the following sources: [γ-32P]ATP from Perkin-Elmer Life Sciences (Boston, MA);

Optikinase from USB (Cleveland, OH); and dNTPs from Bioline (Taunton, MA).

74

4.2.2 DNA Substrates

All DNA substrates listed in Table 4.1 were purchased from Integrated DNA

Technologies, Inc. (Coralville, IA) and purified as described previously23. The M13mp2 ssDNA template was generously provided by Dr. Zachary Pursell from the Tulane

University School of Medicine. All primers were 5′-radiolabeled by incubating with [γ-

32P]ATP and Optikinase for 3 hours at 37 °C. Excess [γ-32P]ATP was removed by passing the reaction mixture through a Bio-Spin 6 column (Bio-Rad). The 5′-radiolabeled primers were annealed to their respective templates (41-mer in Table 4.1 or M13mp2 ssDNA) by incubating the primer and template in a 1:1.15 ratio for 5 minutes at 95 °C and then cooling slowly to room temperature over several hours.

4.2.3 Purification of the Human DNA Polymerase ε Heterotetramer and the p261N Fragment

The cDNAs for the human Polε subunits p12 (Clone ID 5443810), p17 (Clone ID

2822216), and p59 (Clone ID 8991936) were obtained from Open Biosystems. The p261 gene was amplified from pCR-XL carrying the Polε gene125. The genes encoding for all full-length human Polε subunits were cloned into a pFastBac-1 transfer vector (Life

Technologies). During cloning, a His6 tag was added to the N-terminus of the p59 subunit. All cloned genes were verified by DNA sequencing. Preparation of high titer baculoviruses and protein expression in insect cells were performed using the Bac-to-Bac baculovirus expression system (Life Technologies) as described previously126. 1.8x109

Sf21 cells in 1 L of shaking culture were infected simultaneously with four recombinant baculoviruses encoding for each subunit and were cultivated at 25°C for 65 hours. The wild-type Polε heterotetramer was isolated from lysate by Ni-IDA affinity 75 chromatography (Bio-Rad), followed by a HiTrap Heparin HP affinity column, and finally by a Superose 12 size-exclusion column (GE Healthcare). The concentration of purified hPolε was determined by UV spectrometry at 280 nm. SDS-PAGE analysis revealed that hPolε was purified to near-homogeneity (Figure 4.1). The pure peak fractions were combined, aliquoted, and flash-frozen in liquid nitrogen for long-term storage at -80 °C. The wild-type p261N fragment was overexpressed and purified as described previously103.

4.2.4 Reaction Buffers

All assays were performed at 20 °C in reaction buffer E (50 mM Tris-OAc, pH

7.4 at 20 °C, 8 mM Mg(OAc)2, 1 mM DTT, 10% Glycerol, 0.1 mg/mL BSA, and 0.1 mM

EDTA).

4.2.5 Pre-Steady-State Kinetic Assays

In the burst assays, a pre-incubated solution of hPolε or p261N (10 nM) and 5′- radiolabeled D-1 DNA (40 nM) in buffer E was rapidly mixed with dTTP (5 μM) and

Mg2+ (8 mM). In the exonuclease assays, a pre-incubated solution of hPolε or p261N

(100 nM) and 5′-radiolabeled D-1 or M-1 DNA (20 nM) in buffer E was rapidly mixed with Mg2+ (8 mM) to initiate the excision reaction. All reactions were quenched with the addition of 0.37 M EDTA. All reactions were performed using a rapid chemical quench- flow apparatus (KinTek).

4.2.6 Active Site Titration Assay

A pre-incubated solution of hPolε (50 nM) and increasing concentrations of 5′- radiolabeled D-1 DNA (10-80 nM) was rapidly mixed with dTTP (5 μM) and Mg2+ (8

76 mM). Each time point was quenched at 100 ms to ensure maximum product formation from completion of the burst phase. All reactions were performed on a rapid chemical quench-flow and repeated in triplicate.

4.2.7 Processivity Assays

A pre-incubated solution of hPolε or p261N (250 nM) and 5′-radiolabeled 21- or

45-mer primer annealed to M13mp2 ssDNA (25 nM) in buffer E was mixed with all four dNTPs (100 μM each) and Mg2+ (8 mM). The reaction was quenched at various time points with the addition of 0.37 M EDTA.

4.2.8 Product Analysis

Reaction products from the pre-steady-state kinetic assays and the active site titration assay were separated by denaturing PAGE (17% polyacrylamide, 8 M urea, and 1X TBE running buffer) and quantified using a Typhoon TRIO (GE Healthcare) and ImageQuant

(Molecular Dynamics). Reaction products from the processivity assays were separated by denaturing PAGE using a 17% or an 8% gel for the 21-mer M13 and 45-mer M13 substrates, respectively.

4.2.9 Data Analysis

All kinetic data were fit by nonlinear regression using KaleidaGraph (Synergy

Software). Data for the burst assay were fit to eq 4.1

[product] = A[1 - exp( - k1t) + k2t] (eq 4.1)

77 where A is the amplitude of active enzyme, k1 is the observed burst rate constant, and k2 is the observed steady-state rate constant.

Data from the active site titration assay were fit to eq 4.2

DNA DNA 2 1/2 [E•DNA] = 0.5(Kd + E0 + D0) – 0.5[(Kd + E0 + D0) – 4E0D0] (eq 4.2)

DNA where Kd represents the equilibrium dissociation constant for the binary complex

(E•DNA), E0 is the active enzyme concentration, and D0 is the DNA concentration.

Data from exonuclease assays under single-turnover conditions were fit to eq 4.3

[product] = A[exp( - kexot)] + C (eq 4.3)

where A is the reaction amplitude and kexo is the observed DNA excision rate constant.

4.3 Results and Discussion

4.3.1 Burst Assays

Previously, we expressed and purified an exonuclease-deficient form of p261N from E. coli for pre-steady-state kinetic analysis103. We found that a single-nucleotide incorporation by p261N, like most kinetically-characterized DNA polymerases, is limited by a conformational change following nucleotide binding, while additional nucleotide incorporations are limited by dissociation of the enzyme from the E•DNA binary complex7,9,10,12-16. Both phases can be observed at once by performing a burst assay in which the enzyme is pre-incubated with an excess of the DNA substrate to allow

78 formation of a stable E•DNA complex. To see if hPolε behaves similarly to p261N, we performed burst assays with both hPolε and p261N under identical conditions. Briefly, a pre-incubated solution of hPolε or p261N (10 nM) and 5′-radiolabeled D-1 DNA (40 nM,

Table 4.1) was rapidly mixed with dTTP (5 µM) and Mg2+ (8 mM) at 20 °C for various durations of time. Notably, the D-1 DNA substrate used in this study is identical to the

D-1 DNA substrate previously used to kinetically characterize p261N27,103. Both hPolε and p261N exhibited a fast burst of product formation with rate constants of 90±28 s-1 and 101±14 s-1, respectively (Figure 4.2). Similarly, the rate constants for nucleotide incorporation by the yeast Polε catalytic subunit and holoenzyme were found to be nearly identical to each other127. Following the burst phase, hPolε and p261N catalyzed additional product formation characterized by slower linear phases with rate constants of

0.047±0.006 s-1 and 0.018±0.004 s-1, respectively. Thus, hPolε follows a similar kinetic pattern to p261N. The rate constant of the linear phase for p261N is similar to what was previously measured for the exonuclease-deficient mutant, and this rate constant was shown to be equivalent to the steady-state rate of product formation at 20 °C as well as

103 the E•DNA dissociation rate constant (koff) . Therefore, it is likely that the rate constant

-1 of the linear phase (0.047 s ) for hPolε is identical to koff. Unexpectedly, the koff measured for hPolε is 2.6-fold higher than that of p261N. A possible source of this difference may be due in part to the high abundance of acidic residues in the C-terminal portion of the p17 subunit. Notably, the p17 subunit is also a component of the human

CHRAC-15/17 histone fold complex and its negatively charged C-terminal region was previously shown to be necessary for full enhancement of nucleosome sliding by the

79

CHRAC-15/17 complex128. Similarly, this C-terminal region may be essential for facilitating sliding by hPolε along the DNA substrate during highly processive DNA synthesis.

4.3.2 Active Site Titration Assay

To determine if the increase in koff observed for hPolε resulted in overall weaker binding affinity for DNA, we measured the equilibrium dissociation constant of the

DNA E•DNA binary complex (Kd ). Observation of a burst phase is indicative of formation of a stable E•DNA complex that is able to rapidly form product upon nucleotide binding.

Therefore, the amplitude of product formation during the burst phase is directly related to the concentration of the E•DNA complex. By titrating the enzyme with increasing amounts of DNA, the dependency of the burst amplitude, or rather the concentration of the E•DNA binary complex, on free DNA concentration can be determined. From this

DNA DNA relationship, the Kd is derived. To determine the Kd for hPolε, a pre-incubated solution of hPolε (50 nM) and increasing concentrations of 5′-radiolabeled D-1 DNA (10-

80 nM) was rapidly mixed with dTTP (5 µM) for 100 ms before quenching with 0.37 M

EDTA. The reactions were performed in triplicate and the data were fit to eq 4.2 to yield

DNA a Kd of 33±5 nM and an active enzyme concentration (E0) of 9.0 ± 0.7 nM, corresponding to 18% enzyme activity (Figure 4.3). A similarly low activity was observed for the p261N exonuclease-deficient mutant103. Interestingly, the measured

DNA DNA Kd for hPolε (33 nM) is 2.4-fold lower than the Kd of 79 nM determined previously for the p261N exonuclease-deficient mutant103. This difference indicates that the extended structure of hPolε relative to p261N increases the DNA binding affinity of

80 hPolε, most likely by increasing the number of contacts that hPolε is able to make with

-1 the DNA substrate. Using the koff of 0.047 s estimated from the burst assay, the second-

DNA 6 -1 -1 order rate constant of DNA binding (kon = koff/Kd ) is calculated to be 1.4 x 10 M s ,

103 which is over 5-fold higher than the kon calculated for p261N . Thus, the 2.6-fold increase in koff of hPolε relative to p261N is compensated by an even larger increase in kon, suggesting that the small subunits assist hPolε in associating with the primer-template more efficiently.

4.3.3 Processivity Assays

Previously, it was shown that the small subunits of yeast Polε slightly enhanced the processivity of DNA synthesis relative to the Pol2 catalytic subunit alone, but only when the length of the dsDNA region of the singly-primed DNA substrate was 40 nucleotides or longer59. This observation correlated well with the extra length of yeast

Polε afforded by interaction between Pol2 and the small subunits. To see if the processivity of hPolε demonstrated a similar dependence on both the presence of its small subunits and the length of the primer, we compared the processivities of hPolε and p261N during DNA synthesis on an M13mp2 ssDNA template containing a 21- or 45-nucleotide primer. A pre-incubated solution of hPolε or p261N (250 nM) and 5′-radiolabeled 21- or

45-mer M13 primer (Table 4.1) annealed to M13 ssDNA (25 nM) was mixed with all four nucleotides (100 µM) for various times and the products were separated by denaturing PAGE. During extension from the 21-mer M13 primer (Figure 4.4A), both hPolε and p261N showed strong pauses after the addition of 22 or 23 nucleotides, indicating that the presence of the small subunits does not affect the processivity of hPolε

81 relative to p261N on the 21-mer M13 DNA substrate. Surprisingly, hPolε and p261N show similar pausing patterns during extension from the 45-mer M13 primer as well

(Figure 4.4B). Thus, under these conditions, the small subunits do not appear to enhance the processivity of DNA synthesis by hPolε. However, it should be noted that the nucleotide sequence of the 21-mer M13 primer is offset by 3 nucleotides at the 5′ end relative to the 45-mer M13 primer. When the samples from both extension reactions are separated on a 12% denaturing PAGE gel (Figure 4.5), the resulting pausing patterns show a slight offset that is accounted for by the aforementioned sequence offset in the two primers. Thus, the observed pausing pattern is likely a consequence of secondary structure forming at various positions in the M13mp2 ssDNA that is blocking continued

DNA synthesis by both hPolε and p261N. It is evident that highly processive DNA synthesis by hPolε in vivo requires additional processivity factors, including PCNA, RFC, and RPA. Consistently, synthesis of large products on an M13 ssDNA substrate by hPolε was previously shown to be dependent on PCNA and RFC in vitro58.

4.3.4 Excision of Matched and Mismatch DNA Substrates by hPolε

Like many replicative DNA polymerases94-98, hPolε possesses a 3′→5′ exonuclease domain which catalyzes proofreading activity62,63 that is responsible for excising mismatched base pairs formed during DNA synthesis. To determine whether the small subunits have any effect on the proofreading activity of hPolε, we compared the

3′→5′ exonuclease activities of hPolε and p261N on a matched DNA substrate (D-1) and a DNA substrate containing a single mismatched base pair (M-1, Table 4.1). As with the

D-1 DNA substrate, the M-1 DNA substrate was previously used to evaluate the

82 contribution of proofreading to the overall fidelity of p261N27. Briefly, a pre-incubated solution of hPolε or p261N (100 nM) and 5′-radiolabeled D-1 or M-1 DNA (20 nM) was rapidly mixed with Mg2+ (8 mM) for various durations of time. Each plot of remaining substrate versus reaction time was fit to eq 4.3 to yield the overall excision rate constant

(kexo) as a function of the rate constant of DNA transfer from the polymerase site to the exonuclease site, the rate constant of DNA dissociation and rebinding to the exonuclease site, and the true excision rate constant. For the matched D-1 substrate, hPolε and p261N

-1 -1 catalyzed excision with measured kexo values of 0.018±0.002 s and 0.041±0.004 s , respectively (Figure 4.6A). This 2.3-fold decrease in kexo for hPolε relative to p261N suggests that the small subunits may somehow regulate the 3′→5′ exonuclease activity of hPolε against excision of correctly matched DNA, either by limiting transfer of the matched DNA substrate from the polymerase active site to the exonuclease active site or by decreasing the true rate constant of ssDNA excision. The kexo values of both enzymes were then measured in the presence of a single mismatched base pair using the M-1 DNA substrate and were determined to be 0.19±0.03 s-1 and 1.4±0.2 s-1 for hPolε and p261N, respectively (Figure 4.6B). Interestingly, excision by p261N was enhanced by over 30- fold in the presence of a single mismatch, while hPolε only experienced a 10-fold stimulation. Notably, a similar pattern was observed for enhancement of the excision rate constant of the yeast Polε catalytic subunit and holoenzyme127. This result provides evidence that the excision rate constant is mostly limited by the transfer of the DNA substrate from the polymerase active site to the exonuclease active site, as the excision rate constant of ssDNA at the exonuclease site should be independent of DNA duplex

83 stability. It is possible that the extended structure of hPolε is limiting DNA substrate transfer to the exonuclease site relative to p261N. However, investigation of DNA substrate transfer between the exonuclease and polymerase active sites of both the yeast

Polε catalytic subunit and holoenzyme suggests that the additional subunits have no effect on this transfer129. Thus, further studies are required to completely characterize the mechanism of 3′→5′ exonuclease activity catalyzed by both hPolε and p261N and determine the cause of this difference.

4.3.5 Concluding Remarks

We have performed the first kinetic analysis of the four-subunit hPolε holoenzyme and compared its activity to that of the p261N catalytic fragment. We found that the small subunits increase DNA binding affinity to hPolε, but do not appear to affect the processive polymerization activity of hPolε. In contrast, the reduction of the overall excision rate constant of hPolε relative to p261N indicates that the small subunits may sway the enzyme activity toward the DNA synthesis direction. To further explore this hypothesis, we are currently performing a thorough kinetic study of the mechanisms of the 3′→5′ exonuclease activities of both hPolε and p261N. Finally, it is worth noting that p261N was overexpressed and purified from E. coli, while hPolε was prepared from insect cells. Therefore, it is possible that post-translational modification of the catalytic subunit during overexpression in insect cells may account for some of the differences determined in this study, and this hypothesis cannot be ruled out without performing a kinetic analysis of p261N isolated from insect cells.

84

4.4 Tables

Table 4.1: DNA substrates 5′-CGCAGCCGTCCAACCAACTCA-3′ D-1 3′-GCGTCGGCAGGTTGGTTGAGTAGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCAC-3′ M-1 3′-GCGTCGGCAGGTTGGTTGAGTAGCAGCTAGGTTACGGCAGG-5′ 21-mer M13 5′-ACGGCTACAGAGGCTTTGAGG-3′ 45-mer M13 5′-GCAACGGCTACAGAGGCTTTGAGGACTAAAGACTTTTTCATGAGG-3′

85

4.5 Figures

Figure 4.1: Analysis of hPolε purity by SDS-PAGE To evaluate the purity of hPolε, the final protein sample was run on a 12% SDS-PAGE gel: lane 1, protein marker; lane 2, eluate from Superose 12 size-exclusion column. The p12 and p17 subunits have identical mobilities and are indistinguishable.

86

Figure 4.2: Biphasic kinetics of correct dTTP incorporation by hPolε and p261N at 20 °C A pre-incubated solution of 10 nM hPolε (●) or p261N (■) and 40 nM 5′-radiolabeled D-

1 DNA was rapidly mixed with 5 μM dTTP and 8 mM Mg 2+ and quenched after various times with 0.37 M EDTA. The data were fit to eq 4.1 to yield burst phase rate constants of 90±28 s-1 and 101±14 s-1 and linear phase rate constants of 0.047±0.006 s-1 and

0.018±0.004 s-1 for hPolε and p261N, respectively.

87

Figure 4.3: Active site titration of hPolε at 20 °C A pre-incubated solution of hPolε (50 nM) and increasing concentrations of 5′- radiolabeled D-1 DNA (10-80 nM) was rapidly mixed with dTTP (5 µM). All reactions were quenched after 100 ms with the addition of 0.37 M EDTA. The data were fit to eq

DNA 4.2 to yield a Kd of 33±5 nM and an enzyme active concentration of 9.0 ± 0.7 nM.

88

A B

Figure 4.4: Processive DNA synthesis by hPolε and p261N on singly-primed M13mp2 ssDNA templates at 20 °C A pre-incubated solution of hPolε or p261N (250 nM) and 5′-radiolabeled (A) 21- or (B)

45-mer primer annealed to M13 ssDNA (25 nM) was mixed with all four dNTPs (100

µM) for various times before quenching with the addition of 0.37 M EDTA. Products extended from the 21-mer primer were separated by 17% denaturing PAGE, while products extended from the 45-mer primer were separated by 8% denaturing PAGE.

89

Figure 4.5: Comparison of processivities of hPolε and p261N on the 21-mer M13 and 45-mer M13 DNA substrates at 20 °C Samples from the processivity assays monitoring extension from both the 21-mer M13 primer and the 45-mer M13 primer were run together on a single 12% denaturing PAGE gel. A similar pausing pattern is observed on both DNA substrates, which is indicative of termination of DNA synthesis caused by secondary structure forming in the M13 ssDNA template. 90

A B

Figure 4.6: Excision of matched and mismatched DNA by hPolε at 20 °C A pre-incubated solution of 100 nM hPolε (●) or p261N (■) and 20 nM 5′-radiolabeled

(A) D-1 or (B) M-1 DNA was rapidly mixed with Mg2+ and quenched after various times with 0.37 M EDTA. The data were fit to eq 4.3 to yield kexo. For the matched D-1 DNA

-1 -1 (A), the measured kexo values were 0.018±0.002 s and 0.041±0.004 s for hPolε and p261N, respectively. For the mismatched M-1 DNA (B), the measured kexo values were

0.19±0.03 s-1 and 1.4±0.2 s-1 for hPolε and p261N, respectively.

91

Chapter 5. The Accessory Subunits of Human DNA Polymerase ε Regulate the 3′→5′ Exonuclease Activity

5.1 Introduction

Three B-family DNA polymerases (Pols) are responsible for replicating the majority of the nuclear genome: Polα, Polδ, and Polε130. DNA replication is initiated on both the leading and lagging strands by the Polα-primase complex. Briefly, primase synthesizes short primers of 7-12 ribonucleotides (rNTPs) that are extended by Polα with an additional 20-25 deoxyribonucleotides (dNTPs)131. Following primer synthesis, processive DNA synthesis is taken over by Polε and Polδ on the leading and lagging strands, respectively39-41,75.

The eukaryotic replicative DNA polymerases all exist as multi-subunit complexes in vivo130. Each of the replicative DNA polymerases shares a common core architecture consisting of a single large catalytic subunit and a B subunit (e.g. Polα p180-p70, Polδ p125-p50, and Polε p261-p59 in humans). In addition to this core complex, the replicative DNA polymerases associate with unique accessory subunits. For example, the human Polα (hPolα) p180 catalytic subunit and its p70 B subunit form a tetrameric complex with the heterodimeric primase consisting of the p49 catalytic subunit and the p58 regulatory subunit132, while human Polδ (hPolδ) associates with p68 and p12 accessory subunits26 and human Polε (hPolε) interacts with p17 and p12 (distinct from hPolδ p12) accessory subunits56. Low-resolution structures of Saccharomyces cerevisiae 92

Polα35, Polδ133, and Polε59 illustrate that all three replicative polymerases are structurally characterized by a globular catalytic core and an extended structure that is comprised of the accessory subunits. Interestingly, a recent crystal structure of the hPolα-primase complex reveals that the catalytic subunits of Polα and primase are linked to each other through the Polα p70 B subunit and the primase p58 regulatory subunit to allow for efficient coordination of the polymerase and primase activities during primer synthesis34.

Though high-resolution structures of Polδ and Polε from any organism remain to be determined, the extended structures observed for S. cerevisiae Polδ and Polε through low- resolution structural determination methods are hypothesized to increase affinity for

DNA and facilitate coordination at the replication fork during lagging and leading strand synthesis59,133.

In the absence of their respective B and accessory subunits, the catalytic subunits of each of the replicative DNA polymerases are capable of catalyzing template-dependent

DNA synthesis in vitro. However, the accessory subunits either regulate or enhance DNA polymerization activities of the full replicative polymerase complexes. For example, the hPolα p70 B subunit limits the processivity of RNA primer extension by the p180 catalytic subunit134. Furthermore, the polymerase activity of the trimeric hPolδ complex

(p125/p50/p68) is enhanced 4.6-fold in the presence of the p12 accessory subunit, while its 3ʹ→5ʹ exonuclease activity is reduced by as much as 5-fold on a DNA substrate containing a single-base mismatch at the primer 3ʹ terminus73. Moreover, the polymerase processivity by S. cerevisiae Polε is enhanced in the presence of all three accessory subunits59,127. Thus, in order to accurately characterize the activities of the eukaryotic

93 replicative polymerases in vitro, it is imperative that studies are carried out with the full, in-tact DNA polymerase complexes.

Detailed analysis of the full hPolε heterotetrameric complex has been hampered by the difficulty of preparing sufficient quantities of pure complex. Initial studies of hPolε heterotetramer overexpressed in insect cells revealed that accessory subunits do not enhance the activity of the catalytic subunit alone58. However, these studies were performed with excess enzyme on a large M13 DNA template58, which does not allow for a quantitative comparison of DNA binding and nucleotide binding and incorporation kinetics between the catalytic subunit and the heterotetramer. Furthermore, these studies did not investigate the impact of the accessory subunits on the 3ʹ→5ʹ exonuclease activity of hPolε. To determine if the accessory subunits affect the DNA binding and nucleotide incorporation of kinetics of hPolε, we overexpressed and purified fully-assembled hPolε heterotetramer in insect cells and performed a pre-steady-state kinetic analysis of the in- tact complex. We determined that the accessory subunits enhance the DNA binding affinity of hPolε by reducing the DNA dissociation rate constant. While the accessory subunits provide do not significantly affect base substitution fidelity, the contribution of

3ʹ→5ʹ exonuclease activity to fidelity is surprisingly reduced compared to that of the catalytic subunit alone, primarily due to a 10-fold attenuation of the excision rate constant on a DNA substrate containing a single-base mismatch. Together, these observations suggest that the accessory subunits drive hPolε toward processive DNA synthesis to ensure rapid synthesis of the leading strand in vivo.

94

5.2 Materials and Methods

5.2.1 Materials

Reagents used for the experiments described were purchased from the following sources: [γ-32P]ATP from Perkin-Elmer Life Sciences (Boston, MA); Optikinase from

USB (Cleveland, OH); and dNTPs from Bioline (Taunton, MA).

5.2.2 Preparation of the Human DNA Polymerase ε Heterotetramer

The vectors 11A/hPolε1 and 11A/hPolε1exo-, which encode either wild type or exonuclease-deficient (D275A/E277A/D368A) variants of the p261 catalytic subunit, and

11A/hPolε234, which encodes the p59, p17, and p12 accessory subunits, were kindly provided by Dr. Yoshihiro Matsumoto at the University of New Mexico (Albuquerque,

NM). The wild type and exo- p261 sequences contain C-terminal PKA, FLAG, and His16 tags. The p59 sequence contains an N-terminal StrepII tag, a thrombin cleavage site, a

PKA tag, and an HA tag. The p17 and p12 sequences are wild type and contain no tags.

Recombinant baculoviruses were generated using the Bac-to-Bac expression system (Life

Technologies) and amplified to a high titer according to the manufacturer’s instructions.

Wild type or exo- hPolε heterotetramer were overexpressed by simultaneously infecting

1.8x109 Sf9 cells in a 1 L shaking culture with either wild type or exo- p261 virus and virus encoding all three accessory subunits. The cells were incubated at 25 °C for 60 hours, and then harvested and lysed by Dounce homogenization. The lysate was clarified by ultracentrifugation at 20,000 g and hPolε heterotetramer was purified from the cleared lysate essentially as described135.

95

5.2.3 DNA Substrates

The DNA substrates listed in Table 5.1 were purchased from Integrated DNA

Technologies, Inc. (Coralville, IA) and purified by denaturing PAGE as described23. All primers were radiolabeled by incubation with [γ-32P]ATP and Optikinase at 37 °C for 3 hours. Following the labeling reaction, the kinase was inactivated by heating the reaction mixture at 95 °C for 5 minutes, and unreacted [γ-32P]ATP was removed by passing the reaction mixture through a Bio-Spin 6 column (Bio-Rad). The 5′-radiolabeled primers were annealed to the templates in Table 5.1 by incubating the primer with a 1.15-fold molar excess of template at 95 °C for 5 minutes and then slowly cooling the mixture to room temperature over several hours.

5.2.4 Pre-Steady-State Kinetic Assays

All assays were performed at 20 °C in reaction buffer containing 50 mM Tris-

OAc (pH 7.4), 8 mM Mg(OAc)2, 1 mM DTT, 10% glycerol, 0.1 mg/mL BSA, and 0.1 mM EDTA. Fast reactions were performed using a rapid chemical quench-flow apparatus

(KinTek). All reported concentrations are final after mixing.

5.2.5 Active Site Titration Assay

A pre-incubated solution of hPolε exo- (50 nM by UV absorbance) and 5′- radiolabeled D-1 DNA substrate (10-250 nM) was rapidly mixed with a solution containing dTTP (100 μM). The reaction was allowed to proceed for 50 ms and then was quenched with the addition of 0.37 M EDTA. All reactions were repeated in triplicate.

96

5.2.6 Measurement of the E•DNA Complex Dissociation Rate Constant

A pre-incubated solution of hPolε exo- (50 nM) and 5′-radiolabeled D-1 DNA substrate (100 nM) was mixed with an excess of unlabeled D-1 DNA substrate (2.5 μM).

After varying incubation times, the reaction mixture was supplemented with dTTP (100

μM). The nucleotide incorporation reaction was allowed to proceed for an additional 15 seconds and then was quenched with addition of 0.37 M EDTA.

5.2.7 Measurement of the Steady-State Rate Constant of Correct Nucleotide Incorporation

A pre-incubated solution of hPolε exo- (1 nM, active site concentration) and 5′- radiolabeled D-1 DNA (400 nM) was mixed with a solution containing dTTP (100 μM).

After varying incubation times, an aliquot of the reaction mixture was quenched in 0.37

M EDTA.

5.2.8 Measurement of the Elemental Effect on Nucleotide Incorporation

A pre-incubated solution of hPolε exo- (100 nM) and 5′-radiolabeled D-1 DNA substrate (20 nM) was rapidly mixed with a solution containing either dTTP or Sp- dTTPαS (5 μM) for varying reaction times and then was quenched with the addition of

0.37 M EDTA. The elemental effect on incorrect nucleotide incorporation was measured under identical assay conditions, but with the addition of incorrect dATP or Sp-dATPαS

(500 μM).

5.2.9 Single-Turnover Exonuclease Assays

A pre-incubated solution of wild type hPolε (100 nM) and 5′-radiolabeled DNA substrate (20 nM) was rapidly mixed with a solution containing 8 mM Mg(OAc)2 to

97 initiate the 3′→5′ exonuclease reaction. After varying reaction times, the reaction was quenched with the addition of 0.37 M EDTA.

5.2.10 Product Analysis

Reaction substrates and products were separated by denaturing polyacrylamide gel electrophoresis (17% acrylamide, 8 M urea, and 1X TBE running buffer) and quantified using a Typhoon TRIO (GE Healthcare) and ImageQuant (Molecular

Dynamics).

5.2.11 Data Analysis

Data were fit by nonlinear regression using Kaleidagraph (Synergy Software).

Data from the active site titration assay were fit to eq 5.1

DNA DNA 2 1/2 [E•DNA] = 0.5(Kd + E0 + D0) – 0.5[(Kd + E0 + D0) – 4E0D0] (eq 5.1)

DNA where Kd is the equilibrium dissociation constant for hPolε binding to DNA to form the E•DNA binary complex, E0 is the enzyme concentration, and D0 is the DNA concentration.

Data from the DNA dissociation assay were fit to eq 5.2

[Product] = Aexp(–kofft) + C (eq 5.2)

98 where A is the product concentration in the absence of the DNA trap, koff is the DNA dissociation rate constant, and C is the 5′-radiolabeled product concentration in the presence of a trap for unlimited time.

Data from the steady-state nucleotide incorporation assay were fit to eq 5.3

[Product] = kssE0t + E0 (eq 5.3)

where kss is the steady-state rate constant of nucleotide incorporation at the initial enzyme concentration of E0.

Data from single-turnover polymerization assays were fit to eq 5.4

[Product] = Afast[1 – exp(–kfastt)] + Aslow[1 – exp(–kslowt)] (eq 5.4)

where Afast and Aslow are the amplitudes of product formation of the fast and slow phases and kfast and kslow are the observed rate constants of the fast and slow phases. The kfast values were plotted against nucleotide concentration and the data were fit to eq 5.5

dNTP kfast = kp[dNTP]/(Kd + [dNTP]) (eq 5.5)

dNTP where kp is the maximum rate constant of nucleotide incorporation and Kd is the

dNTP equilibrium dissociation constant for dNTP binding. When Kd is very large, the data were fit to eq 5.6

99

dNTP kfast = (kp/Kd )[dNTP] (eq 5.6)

dNTP to yield the substrate specificity constant, kp/Kd .

Data from single-turnover exonuclease assays were fit to eq 5.7

[Remaining substrate] = Aexo[exp(–kexot)] + Aexo2[exp(–kexo2t)] + C (eq 5.7)

where Aexo and Aexo2 are the amplitudes of substrate excision of the fast and slow phases and kexo and kexo2 are the observed excision rate constants of the fast and slow phases.

5.3 Results

5.3.1 Active Site Titration

To determine if p261C and the accessory subunits affect the binding of hPolε to a

DNA substrate, we measured the equilibrium dissociation constant of hPolε exo-binding

DNA to DNA to form the E•DNA binary complex (Kd ) using an active site titration assay.

A fixed concentration of hPolε exo- (50 nM) was pre-incubated with varying concentrations of D-1 DNA substrate (10-250 nM) to allow the E•DNA complex to form prior to initiation of the reaction with the addition of dTTP and Mg2+. The concentration of the product formed during the single-turnover phase was measured by quenching each reaction after 50 ms. Importantly, the amplitude of product formation under these conditions is a direct measurement of the amount of E•DNA complex that forms during the pre-incubation period. The concentration of E•DNA complex was plotted against the concentration of total D-1 DNA substrate and the data were fit to a quadratic equation (eq

100

DNA DNA 5.1) to give a Kd of 22 ± 4 nM (Figure 5.1). Notably, the Kd measured here is

DNA 135 comparable to the Kd of 33 nM that we measured previously for wild type hPolε , indicating that the mutations to deactivate the 3′→5′ exonuclease activity of hPolε did not

DNA impact the ability of hPolε to associate with a DNA substrate. Previously, a Kd of 79 nM was measured using the same assay to monitor the binding of p261N to the same D-1

DNA substrate103. Thus, p261C and the accessory subunits together increase the binding affinity of hPolε by 3.6-fold (Table 5.4).

5.3.2 Measurement of the E•DNA Complex Dissociation Rate Constant

To determine if the tighter binding of hPolε than p261N to a primed DNA substrate resulted from the formation of a more stable complex, we directly measured the rate constant of dissociation of hPolε exo- from DNA (koff) using a DNA trap assay. A pre-formed complex of hPolε exo- (50 nM) and 5′-radiolabeled D-1 DNA substrate (100 nM) was mixed with a large excess of unlabeled D-1 DNA substrate (2.5 μM) for varying incubation times. During the incubation period, any hPolε exo- that dissociates from the labeled DNA substrate will rebind to the unlabeled DNA substrate which is present in a

25-fold excess over the labeled DNA substrate. Thus, when the reaction is initiated with the addition of dTTP and Mg2+, only hPolε exo- that is still bound to labeled DNA substrate will catalyze observable product formation. Product concentration was plotted against trap incubation time and the data were fit to eq 5.2, resulting in a koff of 0.0064 ±

-1 -1 0.0007 s (Figure 5.2A). Interestingly, this value is 3-fold lower than the koff of 0.021 s measured previously for p261N (Table 5.4), and essentially accounts for the increase in

DNA binding affinity observed for the fully-assembled complex. Consistently, the

101

DNA second-order rate constant of DNA binding (kon = koff/Kd ) was calculated to be 2.9 x

5 -1 -1 5 -1 -1 10 M s , which is nearly identical to the kon of 2.7 x 10 M s measured for p261N.

Typically, the dissociation of a DNA polymerase from the E•DNA complex is the slowest step in a multiple-turnover reaction and is identical to the steady-state rate of nucleotide incorporation. To verify this assumption, we directly measured the steady- state DNA polymerization rate constant by mixing hPolε exo- (1 nM) and a large excess of D-1 DNA substrate (400 nM) with dTTP and Mg2+. The time course of product

-1 formation was fit to eq 5.3 and the kss was 0.007 s (Figure 5.2B), which was in good agreement with the DNA dissociation rate constant and confirmed that multiple enzyme turnovers are limited by the rate constant of DNA dissociation as observed for p261N.

Thus, p261C and the accessory subunits do not affect the identity of the rate-limiting step of multiple turnovers by hPolε.

5.3.3 Pre-Steady-State Kinetics of Correct Nucleotide Incorporation

We then sought to determine whether p261C and the accessory subunits affect the kinetics of nucleotide binding and incorporation. To explore this possibility, we measured

dNTP the equilibrium dissociation constant for correct nucleotide binding (Kd ) and the maximum nucleotide incorporation rate constant (kp) by monitoring the dependence of the rate constant of nucleotide incorporation on the concentration of nucleotide under single-turnover conditions. Briefly, a pre-incubated solution of hPolε exo- (100 nM) and

D-1 DNA substrate (20 nM) was rapidly mixed with increasing concentrations of dTTP and Mg2+. After varying reaction times, the reaction was quenched with the addition of

EDTA. The individual time courses of product formation at each nucleotide

102 concentration were fit to eq 5.4 to yield fast and slow observed rate constants for nucleotide incorporation (Figure 5.3A). The fast rate constants were plotted against dTTP

dTTP concentration and fit to eq 5.5 (Figure 5.3B). The resulting kp and Kd values were 411

-1 dTTP ± 26 s and 11 ± 2 μM, respectively, and the substrate specificity (kp/Kd ) was calculated to be 37 μM-1 s-1. Notably, the observed slow rate constants of product formation, which ranged from 0.32-0.65 s-1, showed no clear dependency on nucleotide concentration, which is suggestive of an additional slow step prior to rapid nucleotide binding and incorporation that is exhibited by a significant population of hPolε bound to

DNA. A similar result was observed for single-turnover nucleotide incorporation assays performed with S. cerevisiae Polε and the slow phase was attributed either to switching of the DNA primer terminus from the 3′→5′ exonuclease active site to the polymerase active site or to slow binding of Polε to DNA127.

5.3.4 Elemental Effect on Nucleotide Incorporation

To determine if p261C and the accessory subunits affect the rate-limiting step of correct nucleotide incorporation by hPolε exo-, we compared the incorporation rate constants of dTTP and α-thiophosphate-substituted dTTP (Sp-dTTPαS). A pre-incubated solution of hPolε exo- (100 nM) and D-1 DNA substrate (20 nM) was rapidly mixed with dTTP or Sp-dTTPαS (5 μM) and after varying incubation times the reaction was quenched with the addition of EDTA. The time courses of product formation were fit to eq 5.4 to yield observed incorporation rate constants of 127 ± 15 s-1 and 118 ± 21 s-1 for

dTTP Sp- dTTP and Sp-dTTPαS, respectively (Figure 5.4A). The elemental effect (kobs /kobs dTTPαS) was calculated to be 1.1, which suggests that phosphodiester bond formation is not

103 likely rate-limiting for correct nucleotide incorporation. A similarly small elemental effect was observed for correct nucleotide incorporation by p261N103.

5.3.5 Pre-Steady-State Kinetics of Incorrect Nucleotide Incorporation

The p261N catalytic domain of hPolε exhibits high base substitution fidelity (10-4-

-7 dNTP 10 ) resulting from decreases in both kp and the ground-state binding affinity (1/Kd ) for incorrect relative to correct nucleotides27. To probe whether p261C and the accessory subunits affect the base substitution fidelity of hPolε, we measured the pre-steady-state kinetic parameters for the incorporation of all three incorrect nucleotides with the D-1

dNTP DNA substrate as described above. The kp and Kd values for all four nucleotides are listed in Table 5.2. As for correct dTTP incorporation, we calculated the substrate specificities for each of the incorrect nucleotides and subsequently determined the base

5 7 substitution fidelity (Fpol) hPolε exo- which was 10 -10 (Table 5.2). We also measured the elemental effect on incorrect nucleotide incorporation by comparing the incorporation rate constant of dATP and Sp-dATPαS (500 μM). The observed incorporation rate

-1 -1 constants were 0.073 ± 0.007 s for dATP and 0.0016 s for Sp-dATPαS (Figure 5.4B),

dATP Sp-dATPαS yielding an elemental effect (kobs /kobs ) of 46, suggesting that chemistry may be partially rate-limiting for incorrect nucleotide incorporation. This is similar to the large elemental effect on incorrect nucleotide incorporation observed with p261N103.

5.3.6 Pre-Steady-State Kinetics of Mismatch Extension

Previously we determined that p261N very poorly extends single-base

dNTP mismatches due to large decreases in both kp and 1/Kd for the next correct nucleotide27. To determine if p261C and the accessory subunits affect the ability of hPolε

104 to extend a single-base mismatch, we measured the kinetic parameters for incorporation of the next correct nucleotide, dCTP, onto two different mismatched substrates containing either a C:A mismatch (M-1, Table 5.1) or a C:C mismatch (M-8, Table 5.1) at the primer 3′ terminus. For extension of both the C:A and the C:C mismatched substrates, the kp of correct nucleotide incorporation was reduced by 100- and 63,000- fold (Table 5.3), respectively, relative to extension of a correctly-matched A:T base pair

(D-1, Table 5.1 and Table 5.2). Furthermore, the binding affinity for the correct nucleotide was reduced by 22-fold when extending from the C:A base pair and 64-fold when extending from the C:C base pair. Overall, the substrate specificity of hPolε for the correct nucleotide was reduced by 2.2 x 103- and 4.0 x 106-fold for the C:A and C:C mismatch substrates (Table 5.3), respectively, indicating that the hPolε heterotetramer very poorly extends mismatched DNA substrates compared to matched substrates as observed with p261N. However, both the heterotetramer and the p261N catalytic domain demonstrate a marked variability in their abilities to extend different mismatched base pairs. Based on the current study, hPolε exhibits an 1,800-fold preference for extension of a C:A mismatch versus a C:C mismatch (Table 5.3), while p261N only displayed a 28- fold preference for C:A versus C:C27. The reason for this difference in mismatch type discrimination remains to be determined. We further tested the ability of hPolε exo- to extend the C:A mismatch with an incorrect nucleotide, dGTP. From the measured kinetic parameters, we calculated a substrate specificity of 1.6 x 10-7 μM-1 s-1 for dGTP (Table

5.3), which is similar to the substrate specificity values determined for extension of a mismatched base pair with an incorrect nucleotide by p261N27. Thus, both hPolε and

105 p261N are similarly unlikely to bury a mismatched base pair with an additional mismatch. Based on both a 20,000-fold decrease in kp and a 5.4-fold decrease in binding affinity relative to extension of the same mismatch with correct dCTP, we determined the

-6 base substitution fidelity during mismatch extension (Fext) to be 9.4 x 10 (Table 5.3), demonstrating that hPolε remains highly selective for the correct nucleotide even when extending from a mismatch.

5.3.7 Pre-Steady-State Kinetics of Matched and Mismatched Base Pair Excision

Previous pre-steady-state kinetic analysis of mismatch extension and excision by p261N revealed that the 3′→5′ exonuclease activity of p261N increased its fidelity of

DNA synthesis by 3.5 x 102- to 1.2 x 104-fold due to a strong preference for excision of a mismatched substrate coupled with poor efficiency of extension27. Here, we used a similar approach to measure the excision rate constants for a matched A:T (D-1, Table

5.1) and a mismatched C:A (M-1, Table 5.1) base pair by hPolε. A pre-incubated solution of wild type hPolε (100 nM) and 5′-radiolabeled DNA substrate (20 nM) and was rapidly mixed with Mg2+ in the absence of nucleotide. The concentration of remaining substrate was plotted against time and the data were fit to eq 5.7 to yield rate constants for fast

(kexo) and slow (kexo2) phases of excision (Figure 5.5). The kexo values for the D-1 and M-

1 substrates were 0.013 ± 0.004 s-1 and 0.4 ± 0.1 s-1, respectively. Thus, the excision rate constant increased by 31-fold in the presence of a single-base mismatch. This is comparable to the 15-fold 3′→5′ exonuclease activity enhancement measured previously for p261N in the presence of a single-base mismatch (Table 5.4). Similarly, we measured an excision rate constant of 0.5 ± 0.1 s-1 for a C:C mismatch (M-8, Table 5.1), indicating

106 that the identity of the mismatched base pair does not significantly affect the excision rate constant of hPolε as observed with p261N. Moreover, the kexo2 values, which were 0.004

± 0.001 s-1, 0.0099 ± 0.0006 s-1, and 0.0041 ± 0.0002 s-1 for the D-1, M-1, and M-8 substrates, respectively, appear to be only slightly affected by the presence of a mismatch, and may represent a slow transition of the DNA substrate from the polymerase to the 3′→5′ exonuclease active site5. Notably, the rate constants for excision of both matched and mismatched DNA substrates by hPolε are reduced by 5.6- and 13-fold, respectively, compared to the excision rate constants measured for p261N (Table 5.4).

Thus, the presence of p261C and the accessory subunits appears to restrict the 3′→5′ exonuclease activity of hPolε.

5.4 Discussion

We initially carried out the kinetic characterization of hPolε using an N-terminal fragment (residues 1-1189) of the p261 catalytic subunit (p261N) due to the relative ease of overexpression and purification of sufficient quantities from E. coli that are required for pre-steady-state kinetic studies. Importantly, the p261N fragment contains all the conserved polymerase and 3′→5′ exonuclease motifs. From our pre-steady-state kinetic analysis, we defined a minimal kinetic mechanism of nucleotide incorporation and established a kinetic basis for the high fidelity of DNA polymerization catalyzed by p261N27,103. In brief, p261N catalyzes correct nucleotide incorporation via an induced-fit mechanism that is generally followed by all DNA polymerases and is characterized by a rate-limiting pre-chemistry conformational change following nucleotide binding18.

Moreover, p261N rapidly incorporates correct nucleotides at a fast rate constant of 252 s-1

107 at 20 °C and with a modest binding affinity of 23 μM, and exhibits large decreases in both the maximum incorporation rate constant (kp) and ground-state nucleotide binding

dNTP affinity (1/Kd ) for incorrect nucleotides thereby achieving the high base substitution fidelity (10-4-10-7) that is a hallmark of replicative DNA polymerases136. Furthermore, p261N demonstrates a 102- to 104-fold preference for excision than extension of a single- base mismatch, leading to a calculated overall DNA polymerization fidelity of 10-6-10-11.

Though the fidelity of p261N measured in vitro suggests that hPolε is suitably accurate for replication of the human nuclear genome (3 x 109 base pairs), it is unclear whether the

C-terminal domain of p261 (p261C) or any of the three accessory subunits alter the interactions between hPolε, DNA, and an incoming nucleotide, or affect the kinetics of polymerization or 3′→5′ exonuclease activity. To investigate this possibility, we overexpressed wild type or exonuclease-deficient hPolε heterotetramer in Sf9 cells using recombinant baculoviruses encoding each of the four subunits. We successfully purified milligrams of fully-complexed hPolε and then used pre-steady-state kinetics to assess the impact of p261C and the accessory subunits on the catalytic activity of hPolε.

Earlier structural and biochemical studies of S. cerevisiae Polε implicated a direct role for the accessory subunits in modulating the polymerase processivity of Polε. A low- resolution structure of the S. cerevisiae Polε heterotetramer determined by cryo-electron microscopy (cryo-EM) revealed an extended tail-like structure which is hypothesized to encircle long stretches of dsDNA that trail Polε as it synthesizes the leading strand, thereby tethering Polε to DNA and enhancing processivity59. This hypothesis is supported by the observation that the S. cerevisiae Polε heterotetramer is able to synthesize longer

108 products than the Pol2 catalytic subunit alone under single-hit conditions but only when the length of dsDNA is at least 40 nucleotides long. Consistently, the Dpb3/Dpb4

(p12/p17 in humans) heterodimer forms a stable complex with dsDNA in the absence of

Pol2/Dpb2 (p261/p59 in humans)60, and significantly enhances the polymerase and 3′→5′ exonuclease processivities of the Pol2/Dpb2 heterodimer61, suggesting that the increased processivity results from additional interactions between Polε and DNA afforded by the accessory subunits. To determine if the accessory subunits increase the binding affinity of hPolε for a primed DNA substrate, we performed an active site titration assay and

DNA DNA measured a Kd of 22 nM (Figure 5.1), which is 3.6-fold lower than the Kd previously measured for p261N binding to an identical DNA substrate (Table 5.4).

Additionally, a DNA dissociation rate constant of 0.0064 s-1 was determined using an unlabeled DNA trap (Figure 5.2A). Notably, this dissociation rate constant is 3.3-fold slower than that measured for p261N (Table 5.4) and essentially accounts for the 3.6-fold difference in binding affinity, indicating that the accessory subunits enable hPolε to form a tighter, more stable complex with its DNA substrate. Surprisingly, the processivities of p261N and the hPolε heterotetramer are indistinguishable when the polymerases are in excess over DNA substrate135. In addition, p261N and hPolε are comparably active on a primed M13 DNA template and their activities are similarly enhanced in the presence of

PCNA, RFC, and RPA58. Thus, the polymerization processivity of hPolε does not appear to be affected by the accessory subunits despite the increase in binding affinity for DNA, contrasting sharply with the effect of the accessory subunits on the activity of S. cerevisiae Polε.

109

We then asked whether p261C and the accessory subunits affect the nucleotide binding and incorporation kinetics of hPolε. We first measured the pre-steady-state

dNTP kinetic parameters kp and Kd for correct nucleotide incorporation, and obtained values of 411 s-1 and 11 μM, which are comparable to the average kinetic parameters measured previously for correct nucleotide incorporation catalyzed by p261N (Table 5.4).

Furthermore, we measured the kinetic parameters for each of three incorrect nucleotides using the same DNA substrate, and observed that the accessory subunits have only a small impact (≤20-fold) on the nucleotide selectivity of hPolε. Such a small discrepancy can likely be attributed to differences in experimental design and in the activities of the two enzyme preparations. Moreover, we measured a small elemental effect of 1.1 for correct nucleotide incorporation, and a large elemental effect of 46 for incorrect dATP incorporation, which is consistent with the magnitudes of the elemental effects determined for p261N103. Taken together, the accessory subunits do not affect the nucleotide binding and incorporation kinetics of Polε, and it is expected that hPolε follows an identical mechanism of nucleotide incorporation involving a rate-limiting conformational change triggered by correct nucleotide binding that precedes phosphodiester bond formation. Notably, the pre-steady-state kinetic parameters determined for correct nucleotide incorporation by S. cerevisiae Polε are nearly identical to those measured for the Pol2 catalytic subunit, indicating that the accessory subunits also do not affect the kinetic mechanism of nucleotide incorporation catalyzed by S. cerevisiae Polε127. This conclusion is supported by the positioning of the accessory subunits away from the globular Pol2 catalytic domain in the low-resolution cryo-EM

110 structure of S. cerevisiae59. At present, hPolε structural data is limited to a solution structure of an N-terminal fragment of the p59 accessory subunit64. Though there are currently no structures of the full hPolε heterotetramer, it is expected that the accessory subunits are also positioned distantly from the p261 active site based on in vitro transcription/translation and immunoprecipitation studies which assigned the binding sites for the accessory subunits within the p261C domain58. Interestingly, hPolε catalyzes correct dCTP incorporation during extension from a single-base mismatch with a kp of

-1 -1 4.1 s for a C:A mismatch and 0.0065 s for a C:C mismatch (Table 5.3). This wide kp range (10-3-100 s-1) contrasts significantly from the values determined for p261N, which

-2 -1 generally catalyzes single-base mismatch extension with a kp on the order of 10 s regardless of mismatch identity (Table 5.4). The reason for this difference is not well understood, and stresses the critical importance of high-resolution structural data in understanding the catalytic mechanism of DNA polymerization catalyzed by hPolε.

Lastly, we sought to determine whether p261C and the accessory subunits impact the 3′→5′ exonuclease activity of hPolε. The single-turnover excision rate constant kexo was measured with DNA substrates containing either a correctly paired primer-template pair (D-1, Table 5.1) or a single-base mismatch (M-1 and M-8, Table 5.1). The measured

-1 -1 -1 kexo values were 0.013 s for excision of a matched A:T base pair and 0.4 s and 0.5 s for excision of a C:A and C:T mismatched base pair, respectively. Thus, the excision rate constant is enhanced roughly 35-fold in the presence of a single-base mismatch.

Surprisingly, the kexo values measured here are significantly lower than those measured for p261N, with kexo reduced by 13-fold for excision of a matched DNA substrate and

111

5.6-fold for excision of a mismatched DNA substrate (Table 5.4). Thus, the presence of p261C and the accessory subunits appears to modulate the 3′→5′ exonuclease activity of hPolε. Interestingly, this result contrasts sharply with the pre-steady-state kinetic analysis of S. cerevisiae Polε which demonstrated that the kexo values for excision matched, mismatched, and single-stranded DNA substrates were largely unaffected by the presence or absence of the Pol2 C-terminal domain and the accessory subunits127.

Previously, pre-steady-state kinetic methods have been used to examine the

3′→5′ exonuclease activities of hPolδ, the major lagging strand replicative DNA polymerase, and hPolγ, the DNA polymerase to replicate the mitochondrial genome, as well as their oligomeric assemblies. Notably, kinetic comparison of a heterotrimeric hPolδ assembly (p125-p50-p68) and the fully-assembled heterotetramer (p125-p50-p68- p12), revealed that addition of p12 to the hPolδ heterotrimer results in a 4.6-fold increase in polymerase activity, but 8.3- and 4.8-fold decreases in the excision rates of matched and single-base mismatched DNA substrates, respectively73. Thus, processive polymerization by the hPolδ heterotetramer is less likely to be interrupted by the slow switching of the DNA primer from the polymerase to the exonuclease active site.

Consistent with the enhanced polymerase activity and attenuated exonuclease activity, the hPolδ heterotetramer exhibits increased tolerance for translesion synthesis and mismatch formation and extension in the presence of p12137. Interestingly, it has been shown that treatment of human cells with UV, methyl methanesulfonate, and other DNA damaging agents results in degradation of p12138. Taken together, damage-dependent p12 degradation resulting in a more proofreading active form of hPolδ is a potential

112 mechanism that human cells have evolved for minimizing mutagenic DNA synthesis while maintaining the high rate of DNA synthesis required for timely genome replication under normal cellular conditions. Similarly, hPolγ catalyzes nucleotide incorporation 5- fold faster and is 7.8-fold more processive in the presence of the p55 accessory subunit139, but exhibits a 2-fold decrease in exonuclease activity and greater selectivity against excision of correctly-matched dsDNA32. However, unlike hPolδ, a similar mechanism of proofreading activity control has not been identified for hPolγ in vivo.

Finally, while the polymerization rate we have measured for the hPolε heterotetramer differs only slightly from that of p261N (411 s-1 vs 252 s-1, 1.6-fold), the >5-fold decrease in exonuclease-activity is consistent with the regulatory effects exerted upon hPolδ and hPolγ by their accessory subunits. As with hPolγ, the significance of such control is unknown. Overall, the proofreading activities of these three replicative DNA polymerases are comparably modulated in the presence of their respective accessory subunits, although the physiological role for proofreading control may not be common between these DNA polymerases.

Author Contributions

Walter J. Zahurancik overexpressed and purified hPolε exo- and wild type heterotetramer. W.J.Z. planned and performed the kinetic experiments and analyzed the results. W.J.Z. wrote this chapter. Zucai Suo conceived the research. This work was supported by a Pelotonia Graduate Fellowship to W.J.Z.

113

5.5 Tables

Table 5.1: Sequences of DNA substrates 5′-CGCAGCCGTCCAACCAACTCA-3′ D-1 3′-GCGTCGGCAGGTTGGTTGAGTAGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCAC-3′ M-1 3′-GCGTCGGCAGGTTGGTTGAGTAGCAGCTAGGTTACGGCAGG-5′ 5′-CGCAGCCGTCCAACCAACTCAC-3′ M-8 3′-GCGTCGGCAGGTTGGTTGAGTCGCAGCTAGGTTACGGCAGG-5′

114

Table 5.2: Kinetic parameters for correct and incorrect nucleotide incorporation onto the D-1 DNA substrate catalyzed by hPolε exo- at 20 °C k /K dNTP dNTP k (s-1) K dNTP (μM) p d F a F /F b p d (μM-1 s-1) pol pol, p261N pol dTTP 411±26 11±2 37 - - dATP 0.26±0.01 (4.6±0.5) x 102 5.7 x 10-4 1.5 x 10-5 8 dCTP - - 1.4 x 10-3 3.8 x 10-5 8.4 dGTP (1.50±0.06) x 10-2 (1.8±0.1) x 103 8.3 x 10-6 2.2 x 10-7 20 a dNTP dNTP dNTP Calculated as (kp/Kd )incorrect/[(kp/Kd )correct + (kp/Kd )incorrect]. b Values for Fpol, p261N reported in Table 3.2.

115

Table 5.3: Kinetic parameters for mismatch extension and excision catalyzed by hPolε exo- and hPolε exo+ at 20 °C k /K dNTP dNTP k (s-1) K dNTP (μM) p d F a k dCTP (s-1)b k (s-1) F c p d (μM-1 s-1) ext obs exo exo C:A mismatch (M-1) dCTP 4.1±0.3 (2.4±0.5) x 102 1.7 x 10-2 - 1.2 - dGTP (2.1±0.2) x 10-4 (1.3±0.2) x 103 1.6 x 10-7 9.4 x 10-6 1.5 x 10-5 ------0.4±0.1 0.3 C:C mismatch (M-8) dCTP (6.5±0.5) x 10-3 (7±1) x 102 9.3 x 10-6 - 8.1 x 10-4 ------0.5±0.1 617 a dNTP dNTP dNTP Calculated as (kp/Kd )incorrect/[(kp/Kd )correct + (kp/Kd )incorrect]. b dCTP Calculated as kp[dNTP]/(Kd + [dNTP]) during extension from a mismatched primer terminus at an intracellular dCTP concentration of 100 μM. c dCTP Calculated as kexo/kobs .

116

Table 5.4: Comparison of kinetic parameters determined for p261N and hPolε at 20 °C

a b p261N hPolε p261N/hPolε

-1 -1 koff 0.021 s 0.0064 s 3.3 DNA Kd 79 nM 22 nM 3.6 -1 -1 kp 252 s 411 s 0.6 dNTP, correct Kd 23 μM 11 μM 2.1 -2 -1 - 3 0 -1 -2 1 kp, mismatched 10 s 10 -10 s 10 -10 -1 -1 kexo, matched 0.17 s 0.013 s 13 -1 -1c kexo, mismatch 2.5 s 0.45 s 5.6 a dNTP, correct Tables 2.1, 3.2, and 3.3. Values for kp, Kd , kp, mismatched, kexo, matched, and kexo, mismatched are the averages of the reported values for all tested sequences. bCalculated by dividing the kinetic parameter for p261N by the corresponding value measured for hPolε c -1 -1 Averaged from measured kexo values of 0.4 s and 0.5 s for the M-1 and M-8 substrates, respectively.

117

5.6 Figures

DNA Figure 5.1: Active site titration assay to measure the Kd for hPolε exo- binding to DNA A pre-incubated solution of hPolε exo- (50 nM) and increasing concentrations of 5′- radiolabeled D-1 DNA substrate (10-250 nM) was rapidly mixed with dTTP (100 μM) and Mg2+ for 50 ms and then was quenched with the addition of EDTA. All measurements were performed in triplicate and the average concentration of E•DNA complex that formed during the pre-incubation period, given by product concentration, was plotted against total D-1 DNA concentration and the data were fit to eq 5.1 to yield a

DNA Kd of 22 ± 4 nM. Error bars represent the standard deviation from the calculated average product concentration.

118

A B

Figure 5.2: Measurement of E•DNA complex dissociation rate constant (A) A pre-incubated solution of hPolε exo- (50 nM) and 5′-radiolabeled D-1 DNA substrate (100 nM) was mixed with unlabeled D-1 DNA substrate (2.5 μM) for varying incubation times before the reaction was initiated with the addition of dTTP (100 μM) and Mg2+. The reaction was allowed to proceed for 15 s and then was quenched with the addition of EDTA. Product concentration was plotted against time and the data were fit to

-1 eq 5.2 to yield a koff of 0.0064 ± 0.0007 s . (B) The steady-state DNA polymerization rate constant was measured by mixing a pre-incubated solution of hPolε exo- (1 nM, active site concentration) and 5′-radiolabeled D-1 DNA substrate (400 nM) with dTTP

(100 µM) and Mg2+ for varying reaction times and then quenching the reaction with the addition of EDTA. Product concentration was plotted against time and the data were fit to

-1 eq 5.3 to yield a kss of 0.007 s .

119

A B

Figure 5.3: Pre-steady-state kinetics of correct nucleotide incorporation (A) A pre-incubated solution of hPolε exo- (100 nM) and 5′-radiolabeled D-1 DNA substrate (20 nM) was rapidly mixed with 1.25 μM (●), 2.5 μM (●), 5 μM (●), 10 μM

(●), 80 μM (●), or 200 μM (●) dTTP and Mg2+ for varying incubation times before the reaction was quenched with the addition of EDTA. Product concentration was plotted against time and the data were fit to eq 5.4. (B) The kfast values for each dTTP concentration were plotted against their respective dTTP concentration and the data were

-1 dTTP fit to eq 5.5 to yield a kp of 411 ± 26 s and a Kd of 11 ± 2 μM.

120

A B

Figure 5.4: Elemental effect on correct and incorrect nucleotide incorporation (A) A pre-incubated solution of hPolε exo- (100 nM) and 5′-radiolabeled D-1 DNA

2+ substrate (20 nM) was rapidly mixed with 5 μM dTTP (●) or Sp-dTTPαS (●) and Mg for varying incubation times before quenching with the addition of EDTA. Product concentration was plotted against time and the data were fit to eq 5.4 to yield rate

-1 -1 constants of 127 ± 15 s and 118 ± 21 s for dTTP and Sp-dTTPαS, respectively, resulting in an elemental effect of 1.1. (B) The elemental effect on incorrect dATP incorporation was tested under identical conditions, except the reaction was initiated with

2+ the addition of 500 μM dATP (●) or Sp-dATPαS (●) and Mg . Product concentration was plotted against time and the data for dATP incorporation were fit to eq 5.4, while the data for Sp-dATPαS were fit to a line. The incorporation rate constants were 0.073 ±

-1 -1 0.007 s for dATP and 0.0016 s for Sp-dATPαS, giving an elemental effect of 46.

121

Figure 5.5: Pre-steady-state kinetics of matched and mismatched base pair excision A pre-incubated solution of 100 nM hPolε exo- and 20 nM 5′-radiolabeled D-1 (●) or M-

1 (●) DNA substrate was rapidly mixed with Mg2+ for varying incubation times before quenching with the addition of EDTA. Remaining substrate concentration was plotted against time and fit to eq 5.7 to yield fast excision rate constants of 0.013 ± 0.004 s-1 and

0.4 ± 0.1 s-1 for the D-1 and M-1 DNA substrates, respectively.

122

Chapter 6. Epilogue

6.1 Future Studies on Human DNA Polymerase ε

The goal of this dissertation project was to carry out the first detailed pre-steady- state kinetic studies of hPolε, an essential DNA polymerase that is responsible for performing leading strand DNA synthesis during genomic DNA replication. The preceding chapters describe significant progress toward this goal, beginning with kinetic characterization of the polymerization and exonuclease activities of the N-terminal catalytic domain of the p261 catalytic subunit and finishing with investigation of the effects of the p261 C-terminal domain and the p59, p17, and p12 subunits on these activities. From our studies, we have made several important observations about hPolε- catalyzed DNA polymerization: i) hPolε shares the same minimal kinetic mechanism for nucleotide incorporation as other replicative DNA polymerases (Schemes 1.1 and 2.1); ii) hPolε catalyzes high fidelity DNA polymerization via stringent nucleotide selection and highly efficient proofreading; iii) the p261 C-terminal domain and the accessory subunits do not appear to affect the polymerization activity of the catalytic domain; and iv) the p261 C-terminal domain and the accessory subunits decrease the proofreading activity of the catalytic domain.

The final result is perhaps the most striking, but it is not unprecedented.

Comparison of the 3′→5′ exonuclease activity of fully-assembled hPolδ heterotetramer

123 and heterotrimer lacking the p12 subunit revealed that the excision rate of hPolδ is actually faster in the absence of its p12 subunit73,137. Importantly, the p12 subunit of hPolδ has been shown to degrade in response to UV damage138, indicating that DNA damage triggers a switch in hPolδ composition toward a proofreading active mode in order to mitigate the mutagenic effects of DNA damage. On the other hand, under normal

DNA replication conditions, the heterotetrameric form of hPolδ dominates, prioritizing processive DNA synthesis over proofreading. It is feasible that hPolε may undergo a similar compositional shift as hPolδ in response to DNA damage, given that both DNA polymerases play a major role in genomic DNA replication. However, stable subassemblies of hPolε lacking one or more of its accessory subunits remain to be identified in vivo. Moreover, it is unclear how the small subunits of either polymerase regulate the exonuclease activity of their respective catalytic subunits. Near-atomic resolution structure determination methods, such as single-particle cryo-electron microscopy, may prove to be a powerful tool for uncovering the molecular basis for the regulation of hPolε proofreading by its accessory subunits.

6.2 Future Studies on Cancer-Associated Mutations of Human DNA Polymerase ε

Kinetic investigation of the DNA polymerization fidelity of hPolε provides the groundwork for elucidating a mechanistic basis for mutagenic DNA synthesis catalyzed by cancer-associated hPolε exonuclease domain mutants (EDMs). As discussed in

Chapter 1.5, variants of hPolε containing mutations in the exonuclease domain have been identified as drivers of tumorigenesis due to their exclusive association with hypermutated tumors71. Based on the localization of these mutations within the

124 exonuclease domain, it does not come as a surprise that many of these mutations result in attenuation of hPolε proofreading activity41,71. Interestingly, however, the mutation rate associated with these hypermutated tumors (>100 mutations per 106 bases) is greater than the base substitution frequency of the exonuclease-inactivated mutant of hPolε characterized in our pre-steady-state kinetic studies (1 mutation per 105-107 bases)27.

Thus, the mechanism of mutagenesis by hPolε EDMs is likely more complex than exonuclease inactivation alone. For example, hPolε EDMs may have a higher tolerance for bypass of common DNA lesions such as 8-oxodG, dU, or apurinic/apyrimidinic sites, all of which promote mutagenic DNA synthesis. Detailed kinetic studies of hPolε EDMs and subsequent comparison with wild type hPolε will undoubtedly provide key insight into our understanding of tumorigenesis by cancer-associated hPolε EDMs.

6.3 Future Studies on the Leading Strand Synthesis

Though our kinetic studies have demonstrated that DNA synthesis by hPolε is rapid and highly accurate, both of which are major prerequisites for efficient leading strand DNA synthesis, we also observed that hPolε exhibits poor processivity, dissociating after synthesizing short stretches of DNA. Therefore, hPolε requires additional protein factors in order to synthesize the leading strand in a timely manner.

The Cdc45-Mcm2-7-GINS (CMG) helicase is the most likely candidate as it translocates along the leading strand template while unwinding double-stranded DNA140. Several pieces of evidence support a direct interaction between hPolε and CMG during leading strand DNA synthesis, including i) stimulation of CMG-catalyzed DNA unwinding in the presence of hPolε141; ii) enhanced hPolε DNA synthesis activity in the presence of

125 purified four-subunit GINS complex58; iii) successful reconstitution of leading strand

DNA synthesis in vitro using purified Saccharomyces cerevisiae Polε and CMG142; and iv) determination of several cryo-EM structures of S. cerevisiae Polε bound to CMG42,43.

Measurement of the rate and equilibrium constants along the kinetic pathway in Scheme

2.1 and comparison with the parameters measured for hPolε alone will reveal the role of

CMG in modulating DNA binding and polymerization by hPolε. Furthermore, measurement of the polymerization fidelity in the presence of CMG will elucidate whether any enhancement of hPolε activity by CMG comes at the expense of base substitution fidelity and proofreading.

6.4 Concluding Remarks

In summary, the work completed in this dissertation opens the door to more detailed kinetic studies of leading strand DNA synthesis catalyzed by hPolε and its interacting partners and also provides the basis for investigating the mechanism of mutagenic DNA synthesis catalyzed by cancer-associated hPolε EDMs. Completion of these studies will significantly expand our understanding of leading stand synthesis while guiding similar studies of lagging strand synthesis to produce a more comprehensive mechanism of human DNA replication.

126

References

1. Rothwell, P.J. & Waksman, G. Structure and mechanism of DNA polymerases. Adv Protein Chem 71, 401-40 (2005). 2. Ollis, D.L., Brick, P., Hamlin, R., Xuong, N.G. & Steitz, T.A. Structure of large fragment of Escherichia coli DNA polymerase I complexed with dTMP. Nature 313, 762-6 (1985). 3. Federley, R.G. & Romano, L.J. DNA polymerase: structural homology, conformational dynamics, and the effects of carcinogenic DNA adducts. J Nucleic Acids 2010(2010). 4. Joyce, C.M. & Benkovic, S.J. DNA polymerase fidelity: kinetics, structure, and checkpoints. Biochemistry 43, 14317-24 (2004). 5. Donlin, M.J., Patel, S.S. & Johnson, K.A. Kinetic partitioning between the exonuclease and polymerase sites in DNA error correction. Biochemistry 30, 538- 46 (1991). 6. Kati, W.M., Johnson, K.A., Jerva, L.F. & Anderson, K.S. Mechanism and fidelity of HIV reverse transcriptase. J Biol Chem 267, 25988-97 (1992). 7. Patel, S.S., Wong, I. & Johnson, K.A. Pre-steady-state kinetic analysis of processive DNA replication including complete characterization of an exonuclease-deficient mutant. Biochemistry 30, 511-25 (1991). 8. Wong, I., Patel, S.S. & Johnson, K.A. An induced-fit kinetic mechanism for DNA replication fidelity: direct measurement by single-turnover kinetics. Biochemistry 30, 526-37 (1991). 9. Capson, T.L., Peliska, J.A., Kaboord, B.F., Frey, M.W., Lively, C., Dahlberg, M. & Benkovic, S.J. Kinetic characterization of the polymerase and exonuclease activities of the gene 43 protein of bacteriophage T4. Biochemistry 31, 10984-94 (1992). 10. Dahlberg, M.E. & Benkovic, S.J. Kinetic mechanism of DNA polymerase I (Klenow fragment): identification of a second conformational change and evaluation of the internal equilibrium constant. Biochemistry 30, 4835-43 (1991). 11. Eger, B.T. & Benkovic, S.J. Minimal kinetic mechanism for misincorporation by DNA polymerase I (Klenow fragment). Biochemistry 31, 9227-36 (1992). 12. Brown, J.A. & Suo, Z. Elucidating the kinetic mechanism of DNA polymerization catalyzed by Sulfolobus solfataricus P2 DNA polymerase B1. Biochemistry 48, 7502-11 (2009). 13. Fiala, K.A. & Suo, Z. Mechanism of DNA Polymerization Catalyzed by Sulfolobus solfataricus P2 DNA Polymerase IV. Biochemistry 43, 2116-25 (2004). 127

14. Hsieh, J.C., Zinnen, S. & Modrich, P. Kinetic mechanism of the DNA-dependent DNA polymerase activity of human immunodeficiency virus reverse transcriptase. J Biol Chem 268, 24607-13 (1993). 15. Washington, M.T., Prakash, L. & Prakash, S. Yeast DNA polymerase eta utilizes an induced-fit mechanism of nucleotide incorporation. Cell 107, 917-27 (2001). 16. Kuchta, R.D., Mizrahi, V., Benkovic, P.A., Johnson, K.A. & Benkovic, S.J. Kinetic mechanism of DNA polymerase I (Klenow). Biochemistry 26, 8410-7 (1987). 17. Lin, T.C., Karam, G. & Konigsberg, W.H. Isolation, characterization, and kinetic properties of truncated forms of T4 DNA polymerase that exhibit 3'-5' exonuclease activity. J Biol Chem 269, 19286-94 (1994). 18. Johnson, K.A. The kinetic and chemical mechanism of high-fidelity DNA polymerases. Biochim Biophys Acta 1804, 1041-8 (2010). 19. Tsai, Y.C. & Johnson, K.A. A new paradigm for DNA polymerase specificity. Biochemistry 45, 9675-87 (2006). 20. Brown, J.A., Zhang, L., Sherrer, S.M., Taylor, J.S., Burgers, P.M. & Suo, Z. Pre- Steady-State Kinetic Analysis of Truncated and Full-Length Saccharomyces cerevisiae DNA Polymerase Eta. J Nucleic Acids 2010(2010). 21. Dieckman, L.M., Johnson, R.E., Prakash, S. & Washington, M.T. Pre-steady state kinetic studies of the fidelity of nucleotide incorporation by yeast DNA polymerase delta. Biochemistry 49, 7344-50 (2010). 22. Fiala, K.A., Abdel-Gawad, W. & Suo, Z. Pre-Steady-State Kinetic Studies of the Fidelity and Mechanism of Polymerization Catalyzed by Truncated Human DNA Polymerase lambda. Biochemistry 43, 6751-62 (2004). 23. Fiala, K.A. & Suo, Z. Pre-Steady-State Kinetic Studies of the Fidelity of Sulfolobus solfataricus P2 DNA Polymerase IV. Biochemistry 43, 2106-15 (2004). 24. Johnson, A.A. & Johnson, K.A. Fidelity of nucleotide incorporation by human mitochondrial DNA polymerase. J Biol Chem 276, 38090-6 (2001). 25. Lee, H.R. & Johnson, K.A. Fidelity of the human mitochondrial DNA polymerase. J Biol Chem 281, 36236-40 (2006). 26. Xie, B., Mazloum, N., Liu, L., Rahmeh, A., Li, H. & Lee, M.Y. Reconstitution and characterization of the human DNA polymerase delta four-subunit holoenzyme. Biochemistry 41, 13133-42 (2002). 27. Zahurancik, W.J., Klein, S.J. & Suo, Z. Significant contribution of the 3'-->5' exonuclease activity to the high fidelity of nucleotide incorporation catalyzed by human DNA polymerase. Nucleic Acids Res 42, 13853-60 (2014). 28. Zhang, H., Rhee, C., Bebenek, A., Drake, J.W., Wang, J. & Konigsberg, W. The L561A substitution in the nascent base-pair binding pocket of RB69 DNA polymerase reduces base discrimination. Biochemistry 45, 2211-20 (2006). 29. Zhang, L., Brown, J.A., Newmister, S.A. & Suo, Z. Polymerization fidelity of a replicative DNA polymerase from the hyperthermophilic archaeon Sulfolobus solfataricus P2. Biochemistry 48, 7492-501 (2009).

128

30. Bebenek, A. & Ziuzia-Graczyk, I. Fidelity of DNA replication-a matter of proofreading. Curr Genet (2018). 31. Kunkel, T.A. DNA replication fidelity. J Biol Chem 279, 16895-8 (2004). 32. Johnson, A.A. & Johnson, K.A. Exonuclease proofreading by human mitochondrial DNA polymerase. J Biol Chem 276, 38097-107 (2001). 33. Li, H. & O'Donnell, M.E. The Eukaryotic CMG Helicase at the Replication Fork: Emerging Architecture Reveals an Unexpected Mechanism. Bioessays 40(2018). 34. Baranovskiy, A.G., Babayeva, N.D., Zhang, Y., Gu, J., Suwa, Y., Pavlov, Y.I. & Tahirov, T.H. Mechanism of Concerted RNA-DNA Primer Synthesis by the Human Primosome. J Biol Chem 291, 10006-20 (2016). 35. Nunez-Ramirez, R., Klinge, S., Sauguet, L., Melero, R., Recuero-Checa, M.A., Kilkenny, M., Perera, R.L., Garcia-Alvarez, B., Hall, R.J., Nogales, E., Pellegrini, L. & Llorca, O. Flexible tethering of primase and DNA Pol alpha in the eukaryotic primosome. Nucleic Acids Res 39, 8187-99 (2011). 36. Kunkel, T.A. & Burgers, P.M. Dividing the workload at a eukaryotic replication fork. Trends Cell Biol 18, 521-7 (2008). 37. Stodola, J.L. & Burgers, P.M. Mechanism of Lagging-Strand DNA Replication in Eukaryotes. Adv Exp Med Biol 1042, 117-133 (2017). 38. Burgers, P.M. Polymerase dynamics at the eukaryotic DNA replication fork. J Biol Chem 284, 4041-5 (2009). 39. Miyabe, I., Kunkel, T.A. & Carr, A.M. The major roles of DNA polymerases epsilon and delta at the eukaryotic replication fork are evolutionarily conserved. PLoS Genet 7, e1002407 (2011). 40. Pursell, Z.F., Isoz, I., Lundstrom, E.B., Johansson, E. & Kunkel, T.A. Yeast DNA polymerase epsilon participates in leading-strand DNA replication. Science 317, 127-30 (2007). 41. Shinbrot, E., Henninger, E.E., Weinhold, N., Covington, K.R., Goksenin, A.Y., Schultz, N., Chao, H., Doddapaneni, H., Muzny, D.M., Gibbs, R.A., Sander, C., Pursell, Z.F. & Wheeler, D.A. Exonuclease mutations In DNA Polymerase Epsilon reveal replication strand specific mutation patterns and human origins of replication. Genome Res (2014). 42. Sun, J., Shi, Y., Georgescu, R.E., Yuan, Z., Chait, B.T., Li, H. & O'Donnell, M.E. The architecture of a eukaryotic replisome. Nat Struct Mol Biol 22, 976-82 (2015). 43. Zhou, J.C., Janska, A., Goswami, P., Renault, L., Abid Ali, F., Kotecha, A., Diffley, J.F.X. & Costa, A. CMG-Pol epsilon dynamics suggests a mechanism for the establishment of leading-strand synthesis in the eukaryotic replisome. Proc Natl Acad Sci U S A 114, 4141-4146 (2017). 44. Brieba, L.G. Template dependent human DNA polymerases. Curr Top Med Chem 8, 1312-26 (2008). 45. Kesti, T., McDonald, W.H., Yates, J.R., 3rd & Wittenberg, C. Cell cycle- dependent phosphorylation of the DNA polymerase epsilon subunit, Dpb2, by the Cdc28 cyclin-dependent protein kinase. J Biol Chem 279, 14245-55 (2004).

129

46. Lou, H., Komata, M., Katou, Y., Guan, Z., Reis, C.C., Budd, M., Shirahige, K. & Campbell, J.L. Mrc1 and DNA polymerase epsilon function together in linking DNA replication and the S phase checkpoint. Mol Cell 32, 106-17 (2008). 47. Muramatsu, S., Hirai, K., Tak, Y.S., Kamimura, Y. & Araki, H. CDK-dependent complex formation between replication proteins Dpb11, Sld2, Pol (epsilon}, and GINS in budding yeast. Genes Dev 24, 602-12 (2010). 48. Navas, T.A., Zhou, Z. & Elledge, S.J. DNA polymerase epsilon links the DNA replication machinery to the S phase checkpoint. Cell 80, 29-39 (1995). 49. Iida, T. & Araki, H. Noncompetitive counteractions of DNA polymerase epsilon and ISW2/yCHRAC for epigenetic inheritance of telomere position effect in Saccharomyces cerevisiae. Mol Cell Biol 24, 217-27 (2004). 50. Smith, J.S., Caputo, E. & Boeke, J.D. A genetic screen for ribosomal DNA silencing defects identifies multiple DNA replication and chromatin-modulating factors. Mol Cell Biol 19, 3184-97 (1999). 51. Carson, D.R. & Christman, M.F. Evidence that replication fork components catalyze establishment of cohesion between sister chromatids. Proc Natl Acad Sci U S A 98, 8270-5 (2001). 52. Edwards, S., Li, C.M., Levy, D.L., Brown, J., Snow, P.M. & Campbell, J.L. Saccharomyces cerevisiae DNA polymerase epsilon and polymerase sigma interact physically and functionally, suggesting a role for polymerase epsilon in sister chromatid cohesion. Mol Cell Biol 23, 2733-48 (2003). 53. Ganai, R.A., Zhang, X.P., Heyer, W.D. & Johansson, E. Strand displacement synthesis by yeast DNA polymerase epsilon. Nucleic Acids Res (2016). 54. Parlanti, E., Locatelli, G., Maga, G. & Dogliotti, E. Human base excision repair complex is physically associated to DNA replication and cell cycle regulatory proteins. Nucleic Acids Res 35, 1569-77 (2007). 55. Hogg, M. & Johansson, E. DNA polymerase epsilon. Subcell Biochem 62, 237-57 (2012). 56. Li, Y., Pursell, Z.F. & Linn, S. Identification and cloning of two histone fold motif-containing subunits of HeLa DNA polymerase epsilon. J Biol Chem 275, 23247-52 (2000). 57. Tahirov, T.H., Makarova, K.S., Rogozin, I.B., Pavlov, Y.I. & Koonin, E.V. Evolution of DNA polymerases: an inactivated polymerase-exonuclease module in Pol epsilon and a chimeric origin of eukaryotic polymerases from two classes of archaeal ancestors. Biol Direct 4, 11 (2009). 58. Bermudez, V.P., Farina, A., Raghavan, V., Tappin, I. & Hurwitz, J. Studies on human DNA polymerase epsilon and GINS complex and their role in DNA replication. J Biol Chem 286, 28963-77 (2011). 59. Asturias, F.J., Cheung, I.K., Sabouri, N., Chilkova, O., Wepplo, D. & Johansson, E. Structure of Saccharomyces cerevisiae DNA polymerase epsilon by cryo- electron microscopy. Nat Struct Mol Biol 13, 35-43 (2006). 60. Tsubota, T., Maki, S., Kubota, H., Sugino, A. & Maki, H. Double-stranded DNA binding properties of Saccharomyces cerevisiae DNA polymerase epsilon and of the Dpb3p-Dpb4p subassembly. Genes Cells 8, 873-88 (2003). 130

61. Aksenova, A., Volkov, K., Maceluch, J., Pursell, Z.F., Rogozin, I.B., Kunkel, T.A., Pavlov, Y.I. & Johansson, E. Mismatch repair-independent increase in spontaneous mutagenesis in yeast lacking non-essential subunits of DNA polymerase epsilon. PLoS Genet 6, e1001209 (2010). 62. Jain, R., Rajashankar, K.R., Buku, A., Johnson, R.E., Prakash, L., Prakash, S. & Aggarwal, A.K. Crystal structure of yeast DNA polymerase epsilon catalytic domain. PLoS One 9, e94835 (2014). 63. Hogg, M., Osterman, P., Bylund, G.O., Ganai, R.A., Lundstrom, E.B., Sauer- Eriksson, A.E. & Johansson, E. Structural basis for processive DNA synthesis by yeast DNA polymerase epsilon. Nat Struct Mol Biol 21, 49-55 (2014). 64. Nuutinen, T., Tossavainen, H., Fredriksson, K., Pirila, P., Permi, P., Pospiech, H. & Syvaoja, J.E. The solution structure of the amino-terminal domain of human DNA polymerase epsilon subunit B is homologous to C-domains of AAA+ proteins. Nucleic Acids Res 36, 5102-10 (2008). 65. Baranovskiy, A.G., Gu, J., Babayeva, N.D., Kurinov, I., Pavlov, Y.I. & Tahirov, T.H. Crystal structure of the human Pol B-subunit in complex with the C-terminal domain of the catalytic subunit. J Biol Chem 292, 15717-15730 (2017). 66. Hartlepp, K.F., Fernandez-Tornero, C., Eberharter, A., Grune, T., Muller, C.W. & Becker, P.B. The histone fold subunits of Drosophila CHRAC facilitate nucleosome sliding through dynamic DNA interactions. Mol Cell Biol 25, 9886- 96 (2005). 67. Cancer Genome Atlas, N. Comprehensive molecular characterization of human colon and rectal cancer. Nature 487, 330-7 (2012). 68. Rayner, E., van Gool, I.C., Palles, C., Kearsey, S.E., Bosse, T., Tomlinson, I. & Church, D.N. A panoply of errors: polymerase proofreading domain mutations in cancer. Nat Rev Cancer 16, 71-81 (2016). 69. Heitzer, E. & Tomlinson, I. Replicative DNA polymerase mutations in cancer. Curr Opin Genet Dev 24C, 107-113 (2014). 70. Wheeler, D.A. & Wang, L. From human genome to cancer genome: the first decade. Genome Res 23, 1054-62 (2013). 71. Campbell, B.B., Light, N., Fabrizio, D., Zatzman, M., Fuligni, F., de Borja, R., Davidson, S., Edwards, M., Elvin, J.A., Hodel, K.P., Zahurancik, W.J., Suo, Z., Lipman, T., Wimmer, K., Kratz, C.P., Bowers, D.C., Laetsch, T.W., Dunn, G.P., Johanns, T.M., Grimmer, M.R., Smirnov, I.V., Larouche, V., Samuel, D., Bronsema, A., Osborn, M., Stearns, D., Raman, P., Cole, K.A., Storm, P.B., Yalon, M., Opocher, E., Mason, G., Thomas, G.A., Sabel, M., George, B., Ziegler, D.S., Lindhorst, S., Issai, V.M., Constantini, S., Toledano, H., Elhasid, R., Farah, R., Dvir, R., Dirks, P., Huang, A., Galati, M.A., Chung, J., Ramaswamy, V., Irwin, M.S., Aronson, M., Durno, C., Taylor, M.D., Rechavi, G., Maris, J.M., Bouffet, E., Hawkins, C., Costello, J.F., Meyn, M.S., Pursell, Z.F., Malkin, D., Tabori, U. & Shlien, A. Comprehensive Analysis of Hypermutation in Human Cancer. Cell 171, 1042-1056 e10 (2017).

131

72. Baranovskiy, A.G., Duong, V.N., Babayeva, N.D., Zhang, Y., Pavlov, Y.I., Anderson, K.S. & Tahirov, T.H. Activity and fidelity of human DNA polymerase alpha depend on primer structure. J Biol Chem 293, 6824-6843 (2018). 73. Meng, X., Zhou, Y., Lee, E.Y., Lee, M.Y. & Frick, D.N. The p12 subunit of human polymerase delta modulates the rate and fidelity of DNA synthesis. Biochemistry 49, 3545-54 (2010). 74. Waga, S. & Stillman, B. Anatomy of a DNA replication fork revealed by reconstitution of SV40 DNA replication in vitro. Nature 369, 207-12 (1994). 75. McElhinny, S.A.N., Gordenin, D.A., Stith, C.M., Burgers, P.M.J. & Kunkel, T.A. Division of labor at the eukaryotic replication fork. Mol. Cell 30, 137-144 (2008). 76. Hamatake, R.K., Hasegawa, H., Clark, A.B., Bebenek, K., Kunkel, T.A. & Sugino, A. Purification and characterization of DNA polymerase II from the yeast Saccharomyces cerevisiae. Identification of the catalytic core and a possible holoenzyme form of the enzyme. J Biol Chem 265, 4072-83 (1990). 77. Korona, D.A., Lecompte, K.G. & Pursell, Z.F. The high fidelity and unique error signature of human DNA polymerase epsilon. Nucleic Acids Res 39, 1763-73 (2010). 78. Liu, J. & Tsai, M.D. DNA polymerase beta: pre-steady-state kinetic analyses of dATP alpha S stereoselectivity and alteration of the stereoselectivity by various metal ions and by site-directed mutagenesis. Biochemistry 40, 9014-22 (2001). 79. Arnold, J.J., Ghosh, S.K. & Cameron, C.E. Poliovirus RNA-dependent RNA polymerase (3D(pol)). Divalent cation modulation of primer, template, and nucleotide selection. J Biol Chem 274, 37060-9 (1999). 80. Johnson, K.A. Fitting enzyme kinetic data with KinTek Global Kinetic Explorer. Methods Enzymol 467, 601-26 (2009). 81. Pursell, Z.F., Isoz, I., Lundstrom, E.B., Johansson, E. & Kunkel, T.A. Regulation of B family DNA polymerase fidelity by a conserved active site residue: characterization of M644W, M644L and M644F mutants of yeast DNA polymerase epsilon. Nucleic Acids Res 35, 3076-86 (2007). 82. Iwai, T., Kurosawa, N., Itoh, Y.H., Kimura, N. & Horiuchi, T. Sequence analysis of three family B DNA polymerases from the thermoacidophilic crenarchaeon Sulfurisphaera ohwakuensis. DNA Res 7, 243-51 (2000). 83. Frey, M.W., Nossal, N.G., Capson, T.L. & Benkovic, S.J. Construction and characterization of a bacteriophage T4 DNA polymerase deficient in 3'-->5' exonuclease activity. Proc Natl Acad Sci U S A 90, 2579-83 (1993). 84. Lahiri, I., Mukherjee, P. & Pata, J.D. Kinetic Characterization of Exonuclease- Deficient Staphylococcus aureus PolC, a C-family Replicative DNA Polymerase. PLoS One 8, e63489 (2013). 85. Herschlag, D., Piccirilli, J.A. & Cech, T.R. Ribozyme-catalyzed and nonenzymatic reactions of phosphate diesters: rate effects upon substitution of sulfur for a nonbridging phosphoryl oxygen atom. Biochemistry 30, 4844-54 (1991).

132

86. Fiala, K.A., Sherrer, S.M., Brown, J.A. & Suo, Z. Mechanistic consequences of temperature on DNA polymerization catalyzed by a Y-family DNA polymerase. Nucleic Acids Res 36, 1990-2001 (2008). 87. Tabor, S., Huber, H.E. & Richardson, C.C. Escherichia coli thioredoxin confers processivity on the DNA polymerase activity of the gene 5 protein of bacteriophage T7. J Biol Chem 262, 16212-23 (1987). 88. Prelich, G., Kostura, M., Marshak, D.R., Mathews, M.B. & Stillman, B. The cell- cycle regulated proliferating cell nuclear antigen is required for SV40 DNA replication in vitro. Nature 326, 471-5 (1987). 89. Tan, C.K., Castillo, C., So, A.G. & Downey, K.M. An auxiliary protein for DNA polymerase-delta from fetal calf thymus. J Biol Chem 261, 12310-6 (1986). 90. Chilkova, O., Stenlund, P., Isoz, I., Stith, C.M., Grabowski, P., Lundstrom, E.B., Burgers, P.M. & Johansson, E. The eukaryotic leading and lagging strand DNA polymerases are loaded onto primer-ends via separate mechanisms but have comparable processivity in the presence of PCNA. Nucleic Acids Res 35, 6588-97 (2007). 91. Einolf, H.J. & Guengerich, F.P. Kinetic analysis of nucleotide incorporation by mammalian DNA polymerase delta. J Biol Chem 275, 16316-22 (2000). 92. Davies, J.F., 2nd, Almassy, R.J., Hostomska, Z., Ferre, R.A. & Hostomsky, Z. 2.3 A crystal structure of the catalytic domain of DNA polymerase beta. Cell 76, 1123-33 (1994). 93. Ling, H., Boudsocq, F., Woodgate, R. & Yang, W. Crystal structure of a Y-family DNA polymerase in action: a mechanism for error-prone and lesion-bypass replication. Cell 107, 91-102 (2001). 94. Wang, J., Sattar, A.K., Wang, C.C., Karam, J.D., Konigsberg, W.H. & Steitz, T.A. Crystal structure of a pol alpha family replication DNA polymerase from bacteriophage RB69. Cell 89, 1087-99 (1997). 95. Doublie, S., Tabor, S., Long, A.M., Richardson, C.C. & Ellenberger, T. Crystal structure of a bacteriophage T7 DNA replication complex at 2.2 A resolution. Nature 391, 251-8 (1998). 96. Freemont, P.S., Friedman, J.M., Beese, L.S., Sanderson, M.R. & Steitz, T.A. Cocrystal structure of an editing complex of Klenow fragment with DNA. Proc Natl Acad Sci U S A 85, 8924-8 (1988). 97. Savino, C., Federici, L., Johnson, K.A., Vallone, B., Nastopoulos, V., Rossi, M., Pisani, F.M. & Tsernoglou, D. Insights into DNA replication: the crystal structure of DNA polymerase B1 from the archaeon Sulfolobus solfataricus. Structure 12, 2001-8 (2004). 98. Beese, L.S. & Steitz, T.A. Structural basis for the 3'-5' exonuclease activity of Escherichia coli DNA polymerase I: a two metal ion mechanism. EMBO J 10, 25- 33 (1991). 99. Loeb, L.A. & Kunkel, T.A. Fidelity of DNA synthesis. Annu Rev Biochem 51, 429-57 (1982).

133

100. Olson, A.C., Patro, J.N., Urban, M. & Kuchta, R.D. The energetic difference between synthesis of correct and incorrect base pairs accounts for highly accurate DNA replication. J Am Chem Soc 135, 1205-8 (2013). 101. Kool, E.T. Hydrogen bonding, base stacking, and steric effects in dna replication. Annu Rev Biophys Biomol Struct 30, 1-22 (2001). 102. Petruska, J., Sowers, L.C. & Goodman, M.F. Comparison of nucleotide interactions in water, proteins, and vacuum: model for DNA polymerase fidelity. Proc Natl Acad Sci U S A 83, 1559-62 (1986). 103. Zahurancik, W.J., Klein, S.J. & Suo, Z. Kinetic mechanism of DNA polymerization catalyzed by human DNA polymerase epsilon. Biochemistry 52, 7041-9 (2013). 104. Xia, S. & Konigsberg, W.H. RB69 DNA polymerase structure, kinetics, and fidelity. Biochemistry 53, 2752-67 (2014). 105. Kuchta, R.D., Benkovic, P. & Benkovic, S.J. Kinetic mechanism whereby DNA polymerase I (Klenow) replicates DNA with high fidelity. Biochemistry 27, 6716- 25 (1988). 106. Loeb, L.A. Mutator phenotype may be required for multistage carcinogenesis. Cancer Res 51, 3075-9 (1991). 107. McCulloch, S.D. & Kunkel, T.A. The fidelity of DNA synthesis by eukaryotic replicative and translesion synthesis polymerases. Cell Res 18, 148-61 (2008). 108. Preston, B.D., Albertson, T.M. & Herr, A.J. DNA replication fidelity and cancer. Semin Cancer Biol 20, 281-93 (2010). 109. Lujan, S.A., Williams, J.S., Pursell, Z.F., Abdulovic-Cui, A.A., Clark, A.B., Nick McElhinny, S.A. & Kunkel, T.A. Mismatch repair balances leading and lagging strand DNA replication fidelity. PLoS Genet 8, e1003016 (2012). 110. Schaaper, R.M. Base selection, proofreading, and mismatch repair during DNA replication in Escherichia coli. J Biol Chem 268, 23762-5 (1993). 111. Umar, A. & Kunkel, T.A. DNA-replication fidelity, mismatch repair and genome instability in cancer cells. Eur J Biochem 238, 297-307 (1996). 112. Ohya, T., Maki, S., Kawasaki, Y. & Sugino, A. Structure and function of the fourth subunit (Dpb4p) of DNA polymerase epsilon in Saccharomyces cerevisiae. Nucleic Acids Res 28, 3846-52 (2000). 113. Shikata, K., Sasa-Masuda, T., Okuno, Y., Waga, S. & Sugino, A. The DNA polymerase activity of Pol epsilon holoenzyme is required for rapid and efficient chromosomal DNA replication in Xenopus egg extracts. BMC Biochem 7, 21 (2006). 114. Kesti, T., Frantti, H. & Syvaoja, J.E. Molecular cloning of the cDNA for the catalytic subunit of human DNA polymerase epsilon. J Biol Chem 268, 10238-45 (1993). 115. Morrison, A., Araki, H., Clark, A.B., Hamatake, R.K. & Sugino, A. A third essential DNA polymerase in S. cerevisiae. Cell 62, 1143-51 (1990). 116. Dua, R., Levy, D.L. & Campbell, J.L. Analysis of the essential functions of the C- terminal protein/protein interaction domain of Saccharomyces cerevisiae pol

134

epsilon and its unexpected ability to support growth in the absence of the DNA polymerase domain. J Biol Chem 274, 22283-8 (1999). 117. Kesti, T., Flick, K., Keranen, S., Syvaoja, J.E. & Wittenberg, C. DNA polymerase epsilon catalytic domains are dispensable for DNA replication, DNA repair, and cell viability. Mol Cell 3, 679-85 (1999). 118. Kawasaki, Y. & Sugino, A. Yeast replicative DNA polymerases and their role at the replication fork. Mol Cells 12, 277-85 (2001). 119. Rogozin, I.B., Makarova, K.S., Pavlov, Y.I. & Koonin, E.V. A highly conserved family of inactivated archaeal B family DNA polymerases. Biol Direct 3, 32 (2008). 120. Baranovskiy, A.G., Babayeva, N.D., Liston, V.G., Rogozin, I.B., Koonin, E.V., Pavlov, Y.I., Vassylyev, D.G. & Tahirov, T.H. X-ray structure of the complex of regulatory subunits of human DNA polymerase delta. Cell Cycle 7, 3026-36 (2008). 121. Chilkova, O., Jonsson, B.H. & Johansson, E. The quaternary structure of DNA polymerase epsilon from Saccharomyces cerevisiae. J Biol Chem 278, 14082-6 (2003). 122. Maki, S., Hashimoto, K., Ohara, T. & Sugino, A. DNA polymerase II (epsilon) of Saccharomyces cerevisiae dissociates from the DNA template by sensing single- stranded DNA. J Biol Chem 273, 21332-41 (1998). 123. Dua, R., Edwards, S., Levy, D.L. & Campbell, J.L. Subunit interactions within the Saccharomyces cerevisiae DNA polymerase epsilon (pol epsilon ) complex. Demonstration of a dimeric pol epsilon. J Biol Chem 275, 28816-25 (2000). 124. Uitto, L., Halleen, J., Hentunen, T., Hoyhtya, M. & Syvaoja, J.E. Structural relationship between DNA polymerases epsilon and epsilon* and their occurrence in eukaryotic cells. Nucleic Acids Res 23, 244-7 (1995). 125. Li, Y., Asahara, H., Patel, V.S., Zhou, S. & Linn, S. Purification, cDNA cloning, and gene mapping of the small subunit of human DNA polymerase epsilon. J Biol Chem 272, 32337-44 (1997). 126. Gu, J., Babayeva, N.D., Suwa, Y., Baranovskiy, A.G., Price, D.H. & Tahirov, T.H. Crystal structure of HIV-1 Tat complexed with human P-TEFb and AFF4. Cell Cycle 13, 1788-97 (2014). 127. Ganai, R.A., Osterman, P. & Johansson, E. Yeast DNA polymerase epsilon catalytic core and holoenzyme have comparable catalytic rates. J Biol Chem (2014). 128. Kukimoto, I., Elderkin, S., Grimaldi, M., Oelgeschlager, T. & Varga-Weisz, P.D. The histone-fold protein complex CHRAC-15/17 enhances nucleosome sliding and assembly mediated by ACF. Mol Cell 13, 265-77 (2004). 129. Ganai, R.A., Bylund, G.O. & Johansson, E. Switching between polymerase and exonuclease sites in DNA polymerase epsilon. Nucleic Acids Res (2014). 130. Doublie, S. & Zahn, K.E. Structural insights into eukaryotic DNA replication. Front Microbiol 5, 444 (2014). 131. Pellegrini, L. The Pol alpha-primase complex. Subcell Biochem 62, 157-69 (2012). 135

132. Muzi-Falconi, M., Giannattasio, M., Foiani, M. & Plevani, P. The DNA polymerase alpha-primase complex: multiple functions and interactions. ScientificWorldJournal 3, 21-33 (2003). 133. Jain, R., Hammel, M., Johnson, R.E., Prakash, L., Prakash, S. & Aggarwal, A.K. Structural insights into yeast DNA polymerase delta by small angle X-ray scattering. J Mol Biol 394, 377-82 (2009). 134. Zhang, Y., Baranovskiy, A.G., Tahirov, T.H. & Pavlov, Y.I. The C-terminal domain of the DNA polymerase catalytic subunit regulates the primase and polymerase activities of the human DNA polymerase alpha-primase complex. J Biol Chem 289, 22021-34 (2014). 135. Zahurancik, W.J., Baranovskiy, A.G., Tahirov, T.H. & Suo, Z. Comparison of the kinetic parameters of the truncated catalytic subunit and holoenzyme of human DNA polymerase varepsilon. DNA Repair (Amst) 29, 16-22 (2015). 136. Ganai, R.A. & Johansson, E. DNA Replication-A Matter of Fidelity. Mol Cell 62, 745-55 (2016). 137. Meng, X., Zhou, Y., Zhang, S., Lee, E.Y., Frick, D.N. & Lee, M.Y. DNA damage alters DNA polymerase delta to a form that exhibits increased discrimination against modified template bases and mismatched primers. Nucleic Acids Res 37, 647-57 (2009). 138. Zhang, S., Zhou, Y., Trusa, S., Meng, X., Lee, E.Y. & Lee, M.Y. A novel DNA damage response: rapid degradation of the p12 subunit of dna polymerase delta. J Biol Chem 282, 15330-40 (2007). 139. Johnson, A.A., Tsai, Y., Graves, S.W. & Johnson, K.A. Human mitochondrial DNA polymerase holoenzyme: reconstitution and characterization. Biochemistry 39, 1702-8 (2000). 140. Abid Ali, F., Renault, L., Gannon, J., Gahlon, H.L., Kotecha, A., Zhou, J.C., Rueda, D. & Costa, A. Cryo-EM structures of the eukaryotic replicative helicase bound to a translocation substrate. Nat Commun 7, 10708 (2016). 141. Kang, Y.H., Galal, W.C., Farina, A., Tappin, I. & Hurwitz, J. Properties of the human Cdc45/Mcm2-7/GINS helicase complex and its action with DNA polymerase epsilon in rolling circle DNA synthesis. Proc Natl Acad Sci U S A 109, 6042-7 (2012). 142. Georgescu, R.E., Schauer, G.D., Yao, N.Y., Langston, L.D., Yurieva, O., Zhang, D., Finkelstein, J. & O'Donnell, M.E. Reconstitution of a eukaryotic replisome reveals suppression mechanisms that define leading/lagging strand operation. Elife 4, e04988 (2015).

136