Thermodynamic Characterization of Tetraplex Dna Structures

THERMODYNAMIC CHARACTERIZATION OF TETRAPLEX DNA STRUCTURES

BYUL KIM

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Department of Pharmaceutical Sciences Leslie Dan Faculty of Pharmacy University of Toronto

Thermodynamic Characterization of Tetraplex DNA Structures

Byul Kim

Doctor of Philosophy

Department of Pharmaceutical Sciences Leslie Dan Faculty of Pharmacy

University of Toronto

2018

+ + in TM of preformed G-quadruplexes following the addition of nonstabilizing ions Li , Cs , and

TMA+ (tetramethylammonium). Our data suggest that the stabilizing action of cations on the G- quadruplex conformation is, primarily, due to the central ions which act as specifically bound

II ligands. Nonspecifically bound (condensed) counterions may slightly stabilize the G-quadruplex conformation or exert no influence on its stability (human telomeric G-quadruplexes) or may strongly destabilize it (c-MYC and VEGF). After investigating the salt effect in G-quadruplexes, the latter part of the thesis project was devoted to understanding the folding/unfolding transitions of i-motif, a distinct tetraplex structure formed by c-rich complementary sequences of G- quadruplexes.We described the pH-induced folding/unfolding transitions of i-motifs formed by c-MYC sequence by a linkage thermodynamics-based formalism in terms of three pKa‟s of cytosines, namely, an apparent pKa in the unfolded conformation, pKau, and two apparent pKa‟s in the folded state, pKaf1 and pKaf2.

III

I will love thee, O LORD, my strength.

Psalm 18:1

To my parents and family.

Acknowledgments

I would like to thank my supervisor Dr. Tigran V. Chalikian. I truly respect and appreciate the effort he put in to raise me as a doctor. I am the outcome of your patience and generosity, Dr. Chalikian!

I also would like to express appreciation to my committee members Dr. Robert B. Macgregor, Dr. Jeffrey Henderson, and Dr. Giuseppe Melacini. Thank you all for your encouragement and advice. I could not have done it without you.

Thank you, all my friends and colleagues. To name a few, I would like to express my sincere gratitude to Dr. Yuen Shek, Dr. Ikbae Son, Dr. Yang Li, Dr. Helen Fu, Pei Wang, Tameshwar Ganesh, Ujala Feroze, Lutan Liu, Alah Amsdr, and Jay Lee. I will not forget the quality times I spent with you. Thank you for encouraging me when I was down at times.

Lastly, but not least, I am very much indebted in love to my parents, sisters, and my one and only brother-in-law. Thanks Mom and Dad for always being there for me. Your prayer and words of encouragement have been and will always mean so much to me. I am blessed to be your daughter. My sisters, Jeeyeon and Sohyoung! I respect you big sissys for always holding up with my whining and moaning. Thank you for being my life-time best friends. My big brother Sungjin! I will never forget your patience in everything, especially in listening to my endless stories. Thank you so very much. Love y‟all!

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ……………………………………………………………….…… VI

TABLE OF CONTENTS …………………………………………………………………...… VII

PUBLISHED WORKS ………………………………………………………………..……..… XI

LIST OF TABLES ……………………………………………………………………….…… XII

LIST OF FIGURES ………………………………………………………………………..… XIII

CHAPTER 1: General Introduction ………………………………………………………..… 1

1.1: History of Nucleic Acid Discovery …………………………………………………….…... 1

1.2: Emergence of G-Quadruplexes ………………………………………………………..….… 1

1.3: G-quadruplexes in vivo ……………………………………………………………….…….. 2

1.3.1 Telomeric DNA …………………………………………………………………… 3

1.3.2 Oncogene Promoters …………………………………………………………….… 4

1.3.3 Drugs that Stabilize G-quadruplexes …………………………………….……...... 5

1.4: Structural Properties of G-quadruplexes …………………………………………………..... 6

1.4.1 Building Blocks ……………………………………………………………....…… 6

1.4.2 Hydrogen Bonding ……………………………………………………………...…. 8

1.4.3 Stacking Interactions ……………………………………………………………..... 9

1.5: Topological Diversity of G-quadruplexes ………………………………………………… 10

1.5.1 Tetramolecular G-quadruplexes ………………………………………………….. 11

1.5.2 Bimolecular G-quadruplexes …………………………………………………….. 11

1.5.3 Intramolecular G-quadruplexes ……………………………………………..…… 13

1.6: Cations and Nucleic Acid Structure ……………………………………………………….. 15

1.6.1 DNA as a polyelectrolyte ………………………………………………….…….. 15

VII

1.6.2 Salt Effects in G-quadruplexes …………………………………………….….…. 18

1.7: Energetics/Thermodynamics of Quadruplex Formation …………………….….…………. 20

1.8: Solvent Effects on the Stability of G-quadruplexes …………………………………….…. 21

1.9: i-motif Nucleic Acid Structures ………………………………………………………….... 23

1.10: Methods to Study G-quadruplex Structures ……………………………………………… 24

1.11: Gap in Literature and the Experimental Layout …………………………………………. 26

1.12: Reference ………………………….…………………………………………………...… 28

CHAPTER 2: Polyelectrolyte effects in G-quadruplexes ……………………………….….. 45

2.1: Abstract …………………………………………………..………………………………... 46

2.2: Introduction ……………………………………………………………...………………… 47

2.3: Materials and Methods ……………………………………………………………….……. 49

2.4: Results ………………………………………………………………………………..…..... 51

2.5: Discussion ………………………………………………………………………….……… 57

2.6: Conclusion …………...………………………………………………………….……..….. 61

2.7: Acknowledgements ………………………………………………………………...……… 62

2.8: Reference…………………………..………………………………………………………. 63

CHAPTER 3: Effects of Salt on the Stability of a G-quadruplex from the Human c-MYC

Promoter ……………………………………………………………………...…………...... 67

3.1: Abstract ………………………………………………………………...……………...... 68

3.2: Introduction …………………………………………………………………………...... 69

3.3: Materials and Methods ………………………………………………………………...... 72

3.4: Results ………………………...……………………………………………………...... 74

3.5: Discussion ……………………………………………………………………….…...... 86

VIII

3.6: Conclusion ………………………………………………………………….……...... 93

3.7: Acknowledgements ……………………………………………………….…………...... 94

3.8: Reference …………………………………………………………………...…...…...... 95

CHAPTER 4: Ionic effects on VEGF G-quadruplex Stability ……………...………...... 113

4.1: Abstract ……………………………………………………………………...………...... 114

4.2: Introduction …………………………………………………………………………...... 115

4.3: Materials and Methods ………………………………………………………………...... 117

4.4: Results ………………………………………………………………………………...... 119

4.5: Discussion ………………………………………………………………………………... 130

4.6: Concluding Remarks ………………………………………………………………...... 134

4.7: Supporting Information ……………………………………………………………...... 135

4.8: Acknowledgements ……………………………………………………………..…...... 136

4.9: References …………………………………………………………………………...... 137

CHAPTER 5: Thermodynamic linkage analysis of pH-induced folding and unfolding transitions of i-motifs ………………………………………………………………...... 144

5.1: Abstract …………………………………………………………………………...... 145

5.2: Introduction …………………………………………………………………………...... 146

5.3: Experimental Section ………………………………………………………………...... 155

5.4: Acknowledgements …………………………………………………………………...... 156

5.5: Reference ………………….……………………………………...………………...... 157

CHAPTER 6: General Conclusion and Future Perspectives …………………...…...... 162

6.1: Conclusions ………………….……………………………………………………...... 162

6.2: Future Prospect ……………………………………………………….………………...... 164

6.3: Reference …………………..………………………………………………………...... 166

APPENDIX A: Effect of Urea on G-quadruplex Stability ……………………………...... 167

A.1: Abstract …………………………………………………………………………...... 168

A.2: Introduction …………………………………………………………………………...... 169

A.3: Materials and Methods ………………………………………………………….……...... 170

A.4: Results ………………………………………………………………………………...... 172

A.5: Discussion …………………………………………………………………………...... 181

A.6: Conclusions ………………………………………………………………………..…...... 189

A.7: Acknowledgements …………………………………………………………………...... 190

A.8: Reference ……………….…………………………………………...……………...... 191

PUBLISHED WORKS

1. Kim, B. G., Shek, Y. L., and Chalikian, T. V. (2013). Polyelectrolyte effects in G- quadruplex. Biophys. Chem. 184 95-100.

2. Kim, B. G., Evans, H. M., Dubins, D. N., and Chalikian, T. V. (2015). Effects of salt on the stability of a G-quadruplex from the human c-MYC promoter. Biochemistry, 54, 3420-3430.

3. Byul G. Kim, Ji Long, David N. Dubins, and Tigran V. Chalikian (2016). Ionic Effects on G-quadruplex Stability. J. Phys. Chem. B, 120 (22), 4963-4971.

4. Byul G. Kim and Tigran V. Chalikian (2016). Thermodynamic linkage analysis of pH- induced folding and unfolding transitions of i-motifs. Biophysical chemistry, 216, 19-22.

5. Lusine Aslanyan, Jordan Ko, Byul G. Kim, Ishkhan Vardanyan, Yeva B. Dalyan, and Tigran V. Chalikian (2017). Effect of Urea on G-Quadruplex Stability. J. Phys. Chem. B, 121, 6511- 6519.

6. Lutan Liu, Byul G. Kim, Ujala Feroze, Robert Macgregor, and Tigran V. Chalikian (2018). Probing the Ionic Atmosphere and Hydration of the c-MYC i-motif. Journal of the American Chemical Society.

LIST OF TABLES

Chapter 1

Table 1-1 Ionic radii of G-quadruplex stabilizing cations.

Chapter 3

Table 3-1 Molecular volumes, VM, and solvent accessible surface areas, SA, of the c-MYC

and human telomeric Tel22 and Tel26 G-quadruplexes. Values normalized per

nucleotide are shown in parentheses.

XII

LIST OF FIGURES

Chapter 1

Figure 1-1 Components of a Nucleotide.

Figure 1-2 Base Geometry (Syn and Anti glycosylic bonds).

Figure 1-3 Guanine Nucleotides.

Figure 1-4 G-quartet.

Figure 1-5 Stacking of G-tetrads.

Figure 1-6 Schematic diagrams of various topologies of bimolecular quadruplexes. (a) lateral

loops connecting antiparallel strands to form head-to-head hairpin quadruplex; (b)

lateral loops connecting antiparallel strands on different sides; (c) external and

lateral loops; (d) external (propeller) chain reversal loops linking parallel strands;

(e) diagonal linkers connecting opposite anti-parallel strands.

Figure 1-7 Basket-type Anti-parallel G-quadruplex.

Figure 1-8 Guanine Tetrad and Associated Waters.

Figure 1-9 Cytosine-Cytosine+ Base Pairs.

XIII

Chapter 2

Figure 2-1 CD spectra of Tel22 in the presence of 20 mM NaCl and 0 (●), 180 (○), 380 (■),

and 780 (□) mM LiCl (panel A), CsCl (panel B), and TMACl (panel C).

Figure 2-2 CD spectra of Tel26 in the presence of 20 mM KCl and 0 (●), 180 (○), 380 (■),

and 780 (□) mM LiCl (panel A), CsCl (panel B), and TMACl (panel C).

Figure 2-3 UV melting profiles at 295 nm of Tel22 at 100 mM NaCl (●) and Tel26 at 100

mM KCl (○).

Figure 2-4 (a) Dependences of the melting temperature, TM, of Tel22 on the concentration of

Na+ (●), Li+ (○), Cs+ (■), and TMA+ (□) cations; (b) Dependences of the melting

+ + + temperature, TM, of Tel26 on the concentration of K (●), Li (○), Cs (■), and

TMA+ (□) cations.

Chapter 3

Figure 3-1 (a) CD spectra of MYC22-G14T/G23T in a 10 mM TBA-phosphate buffer at

various KCl concentrations at 25 °C. (b) Dependence of the molar ellipticity of

MYC22-G14T/G23T at 264 nm on the KCl concentration. Experimental data

were approximated with Eq. (1) for n = 3 (solid line).

Figure 3-2 (a) Representative UV melting profile at 295 nm of the MYC22-G14T/G23T G-

quadruplex at 1 mM KCl. (b) Dependence of the melting temperature, TM, of the

XIV

MYC22-G14T/G23T G-quadruplex on the concentration of K+ ions.

Experimental data were approximated by an exponential function solely to guide

the eye (solid line).

Figure 3-3 (a) CD spectra of the MYC22-G14T/G23T G-quadruplex at 300 μM KCl at

various CsCl concentrations. (b) CD spectra the MYC22-G14T/G23T G-

quadruplex preformed in 300 μM KCl with subsequently added 800 mM CsCl at

temperatures ascending from 25 to 80 °C and descending to 25 °C.

Figure 3-4 (a) CD spectra of the MYC22-G14T/G23T G-quadruplex at 300 μM KCl at

various TMACl concentrations. (b) CD spectra the MYC22-G14T/G23T G-

quadruplex preformed in 300 μM KCl with subsequently added 800 mM TMACl

at temperatures ascending from 25 to 80 °C and descending to 25 °C.

Figure 3-5 Dependences of the melting temperature, TM, of the MYC22-G14T/G23T G-

quadruplex on the concentration of CsCl at 300 (a), 600 (b), and 1000 (c) μM KCl.

Experimental data were approximated by an exponential function solely to guide

the eye (solid lines).

Figure 3-6 Dependences of the melting temperature, TM, of the MYC22-G14T/G23T G-

quadruplex on the concentration of TMACl at 300 (a), 600 (b), and 1000 (c) μM

KCl. Experimental data were approximated by an exponential function solely to

guide the eye (solid lines).

Figure 3-7 CsCl-dependence of the molar ellipticity of MYC22-G14T/G23T at 264 nm in

600 μM KCl at 51 °C. Experimental data were approximated with Eq. (3) (solid

line).

Chapter 4

Figure 4-1 (a) CD spectra of Pu22-T12T13 in a 10 mM TBA-phosphate buffer at various

KCl concentrations at 25 °C. (b) Dependence of the molar ellipticity of Pu22-

T12T13 at 264 nm on the KCl concentration. Experimental data were

approximated with Eq. (1) for n = 2 (solid line).

Figure 4-2 (a) Dependence of the melting temperature, TM, of the Pu22-T12T13 G-

quadruplex on the concentration of K+ ions. Experimental data were

approximated by a linear function (solid line). (b) The G-quadruplex-to-coil

transition van't Hoff enthalpies of Pu22-T12T13 plotted as a function of the

transition temperature, TM, at 2 (■), 5 (□), and 10 (●) mM KCl and TMACl

concentrations ranging from 0 to 800 mM and 2 (○), 5 (♦), an 10 (◊) mM KCl and

CsCl concentrations ranging from 0 to 100 mM.

Figure 4-3 CD spectra of the Pu22-T12T13 G-quadruplex at 2 (a), 5 (b), and 10 (c) mM KCl

at 25 °C and various concentrations of TMACl.

Figure 4-4 CD spectra of the Pu22-T12T13 G-quadruplex at 2 (a), 5 (b), and 10 (c) mM KCl

at 25 °C and various concentrations of CsCl.

XVI

Figure 4-5 Dependences of the melting temperature, TM, of the Pu22-T12T13 G-quadruplex

on the concentration of TMACl at 2 (a), 5 (b), and 10 (c) mM KCl. Experimental

data were approximated by an exponential function solely to guide the eye (solid

lines).

Figure 4-6 Dependences of the melting temperature, TM, of the Pu22-T12T13 G-quadruplex

on the concentration of CsCl at 2 (a), 5 (b), and 10 (c) mM KCl. Experimental

data were approximated by an exponential function solely to guide the eye (solid

lines).

Chapter 5

Figure 5-1 The CD spectra of the ODN between pH 9.5 and 4.5 (panel A) and pH 4.8 and 2.3

(panel B).

Figure 5-2 The pH-dependence of the molar ellipticity at 284 nm. Experimental data were

approximated with Eq. (2) (solid line).

Figure 5-3 The pH-dependence of the differential number of protons bound to the folded and

unfolded states of the ODN calculated with Eq. (3).

Figure 5-4 Transition profiles simulated with Eq. (2) for pKau, pKaf1, and pKaf2 of 4.8, 6.0,

and 3.6, respectively, and K0 equal to 0.0001 (dark yellow), 0.001 (green), 0.005

(blue), 0.02 (black), and 0.1 (red).

XVII

Figure 5-5 Transition profiles simulated with Eq. (2) for i-motifs containing 6 (black) and 8 (red)

+ C·C base pairs with pKau, pKaf1, and pKaf2 of 4.8, 6.0, and 3.6, respectively, and K0 of

0.0001.

Appendix A

Figure A-1 The CD spectra of Tel22 at 25 °C in the absence of urea at various NaCl

concentrations. The NaCl concentration (in mM) for each spectrum is given in

the inset.

Figure A-2 The CD spectra of Tel22 at 25 °C at various urea concentrations at 20 (panel a),

50 (panel b), 100 (panel c); 200 (panel d); 400 (panel e); 600 (panel f); 800 (panel

g); and 1000 (panel h) mM NaCl. The urea concentrations (in M) for the spectra

are given in the insets.

Figure A-3 The CD spectra of the urea-induced unfolded state at 20 mM NaCl and 8 M urea

(○) and temperature-induced unfolded states at 20 mM NaCl and 95 °C in the

absence of urea (■) of Tel22. The inset presents the temperature dependence of

the molar ellipticity at 257 nm of the urea-induced unfolded state at 20 mM NaCl

and 8 M urea.

Figure A-4 Representative UV melting profile at 295 nm Tel22 at 200 mM NaCl and 1 M

urea. Experimental data were approximated with Eq. (1) (solid line).

XVIII

Figure A-5 The G-quadruplex-to-coil transition temperature, TM, transition temperatures of

Tel22 as a function of urea at various NaCl concentrations. The NaCl

concentration (in mM) for each set of TM is given in the inset.

Figure A-6 Compilation of transition enthalpies, ΔHM, of Tel22 measured at various

combinations of NaCl and urea concentrations plotted against corresponding

transition temperatures, TM.

Figure A-7 The dependence of the molar ellipticity of the Tel22 at 295 nm on the urea

concentration at 20 mM NaCl. Experimental data were approximated with Eq. (2)

(solid line).

Figure A-8 The three-dimensional dependence of the transition temperature, TM, on the

concentrations of NaCl and urea.

XIX

Chapter 1: General Introduction 1

CHAPTER 1: General Introduction

1.1: History of Nucleic Acid Discovery

In 1953, Franklin and Wilkins crystallized a DNA molecule and recorded its X-ray diffraction patterns3. Subsequently, Watson and Crick used the diffraction patterns to construct three-dimensional model of the DNA4. The proposed model envisages a double helical structure in which two separate 5‟-3‟ strands run in an anti-parallel manner. The two strands are interconnected via hydrogen bonds between complementary bases resembling steps of a ladder. In agreement with the chemical evidence of uniform composition offered by Chargaff and his colleagues, the Watson - Crick model of a DNA duplex offers a molecular rationalization of the complementarity of A and T and G and C bases5.

The Watson and Crick model has since been corroborated through numerous fiber diffraction experiments and single-crystal X-ray measurements6. Structural studies of nucleic acids not only produced a deeper understanding of the canonical nucleic acid structure but also unveiled the environment and sequence-dependent polymorphism of DNA7. The conformational flexibility of nucleic acids under various environmental conditions and in the presence of DNA- binding proteins hints at the diversity of DNA structures adopted at different genomic loci and during various phases of the cell cycle7.

1.2: Emergence of G-Quadruplexes

More than 40 years prior to the development of the double helical DNA model, the Norwegian chemist Ivar Christian Bang found that concentrated samples of guanylic acid from ox pancreas could form viscous gels8. Guanylic acid at a concentration of 25.0 mg/mL forms a thick aqueous solution which suggests that self-association may take place9. This simple yet novel proposal has caused many controversies as some researchers argued that guanylic acid, a

Chapter 1: General Introduction 2

simple nucleotide similar to inosinic acid, could not form higher order structures10. This argument persisted until 1962, when Gellert and collaborators employed X-ray crystallography to illustrate that guanine bases, indeed, arrange themselves into a large helical tetraplex structure now referred to as a G-quadruplex11.

1.3: G-quadruplexes in vivo

In 2005, Huppert and Balasubramanian scanned through the human genome using the algorithm Quadparser to assess the prevalence of the G-quadruplex-forming sequences12. These authors have devised an algorithm that can be used to identify the sequences d(G3+N1–7 G3+N1–7 12 G3+N1–7 G3+) that may fold into putative G-quadruplex structures . Theoretically, there are 375,000 candidate sequences capable of folding into G-quadruplexes12. Although the sequences identified through the bioinformatics analysis may or may not adopt a G-quadruplex structure, the specific dispersion pattern suggests that the tetraplex formation may play a vital role at certain loci of the human genome12.

The Balasubramanian group (2013) has also succeeded in visualizing G-quadruplexes in human cells using structure-specific antibodies13. They have observed discrete G-quadruplexes distributed throughout the genome13. Telomeric ends at chromosomes favour formation of G- quadruplexes given the abundance of guanine-rich tandem repeats and the presence of single- stranded 3‟-overhangs14. It is worth noting, however, that the absence of the complementary strand is not a requirement for formation of G-quadruplexes. The canonical double helical DNA can undergo transient unwinding in the middle of the transcription or replication events, thereby providing an ideal condition for tetraplex DNA formation at a number of functionally important genomic loci such as promoter regions of proto-oncogenes14. Proto-oncogenes are essential for normal functioning of cells but once deregulated, they become oncogenes whose expression can lead to cancer initiation. The evidence of G-quadruplexes existing at sites crucial for transcriptional control and senescence marks these structures as potential targets for therapeutics.

Chapter 1: General Introduction 3

1.3.1 Telomeric DNA

In the early 1990s, Blackburn and her colleagues identified G-rich tandem repeats positioned at the end of chromosomes15. This region, which measures about 5-15 kilobases in length, is now referred to as a telomere16. The G-rich tandem repeat sequences are highly conserved across different species. Telomeres are composed of a double-stranded section that is a few thousand nucleotides long followed by a short dangling terminal end consisting of 100-200 bases16. Understandably, the duplex region has a complementary C-rich strand running in the opposite direction, whereas the single-stranded overhang is unprotected and exposed to the environment. Some common examples of the repeat units are TTAGGG in mammals17, TTGGGG in the ciliate Tetrahymena18.19, and TTTTGGGG in the protozoa Oxytricha nova20. The biomedical relevance of telomeres stems from the fact that their length is an accurate timer of cellular aging21-25. Telomeres are destined to shorten by about 40 to 50 bases with each DNA replication. When the chromosomal terminus reaches a critical length, the cell is signaled to undergo apoptosis26. There is an enzyme that elongates telomeres in order to systematically decelerate this aging clock27. The enzyme, telomerase, is a reverse transcriptase that has an RNA template complementary to the guanine rich telomeric sequence28. Nonetheless, this molecule is rarely abundant in human tissues unless the cell requires a high degree of development156. Hence, telomerase is active in germ cells or stem cells where cell proliferation is key29, 30. It is essential to control the telomerase activity. There is much evidence indicating that the maintenance of telomeric ends is central to making cells overproliferative31, 32, 156. Telomerase activity is very high in about 90% of immortalized cells, including tumour cells156. G-quadruplexes may play an essential role in this respect, since the formation of a tetraplex structure in the telomeric region has been shown to inhibit the telomerase activity33, 34. Finding ways to control cellular immortalization and incessant replication is the ultimate pathway to targeting tumourigenesis. Thus, G-quadruplexes have been evolved as a potential target in cancer therapy.

Chapter 1: General Introduction 4

1.3.2 Oncogene Promoters

Transcriptional regulatory regions within the genome also have a high propensity to fold into G-quadruplexes35. Putative G-quadruplex-forming sequences rich in guanine bases are found in 50% of all identified human genes, in particular, in promoters and nuclease hypersensitive elements within 1000 bases upstream of the transcription initiation site36. The nuclease hypersensitive elements and promoter regions are 230 times more likely to adopt G- quadruplex structures relative to other genomic loci36. Nuclease hypersensitive site is an area proximal to the promoter that is easily detected by cleavage enzymes such as DNase35. This region is characterized by the absence of histones (which make chromosomes compact) thereby enhancing the flexibility of the DNA and its propensity to fold into a G-quadruplex structure35. Although there is a large number of oncogene promoters, the regulatory pathways of which still need to be investigated, there are quite a few of them that have been well scrutinized. One example is c-MYC that encodes a transcription factor that is essential for regulating cell proliferation37-39. Deregulation of c-MYC expression triggers uncontrolled multiplication of cells which is associated with tumorigenesis in a variety of tissues37, 39. c-MYC can either act as a primary oncogene that becomes amplified41 or translocated40 or it can activate downstream pathways of other oncogenes in a chain reaction41. G-quadruplex formation at the promoter of c- MYC is thought to provide steric hindrance and impair the binding of initiation proteins42. This notion suggests that small ligands that stabilize tetraplex DNA can act as antitumor drugs.

Another well investigated proto-oncogene is vascular endothelial growth factor (VEGF). VEGF plays an essential role in angiogenesis by producing growth factors that are vital for generating embryonic circulatory system or for growing blood vessels from pre-existing vascular structure43, 44. Solid tumours that are about 1mm long in diameter usually depend on other host blood vessels to acquire oxygen and nutrients45. However, when cancer cells cannot sustain their activities with the substances they obtain from existing blood cells, they become angiogenic themselves by gaining the capability to generate new blood vessels46. The angiogenic ability is activated in tumour cells with elevated expression of VEGF47. The over-expression of human VEGF observed in many tumor tissues can be regulated at the stage of transcription. One way of achieving such a regulatory effect is to induce a G-quadruplex at the promoter region of the

Chapter 1: General Introduction 5

oncogene to prevent the binding and suppress the activity of transcriptional proteins. The oligomeric sequences derived from the c-MYC and VEGF promoter regions are the main subject of the experiments described in this thesis.

1.3.3 Drugs that Stabilize G-quadruplexes

The ultimate aim of scrutinizing the distribution and biological relevance of G- quadruplex-forming sequence is to develop an ability to induce G-quadruplexes at selected genomic loci via intervening low molecular weight compounds (drugs). In 1991, Zahler and his colleagues reported the inability of Oxytricha telomerase to elongate the telomeric oligonucleotide coiled into a G-quadruplex form in vitro48. Given this observation, much work has been invested to find natural molecules and their derivatives that can trigger G-quadruplex folding and disrupt telomere elongation in vivo49. Numerous compounds representing various chemical classes (such as porphyrin50, acridine51, perylene52, and bisquinolinium51, 53) have been shown to impede telomerase activity.

Telomestatin is one of the better understood high-affinity G-quadruplex binding molecules. This natural compound isolated from Streptomyces anulatus is a powerful telomerase inhibitor54. It stabilizes the G-quadruplex conformation, particularly, the anti-parallel topology, thereby blocking the telomerase activity54. Telomestatin blocks telomere elongation within time periods that are too short to measure if used at a dose higher than 5μM55, 56. At concentrations lower than 2μM, telomestatin has been shown to promote a gradual telomere degradation or programmed cell death in numerous tumor cell lines in vivo57. Telomestatin also induces a quick dissociation of the telomere capping proteins TRF2 and Pot158. Other low molecular weight compounds that bind G-quadruplexes and are capable of uncapping the 3‟ ends of chromosomes are BRACO19 and the acridine RHPS459, 60. The selective uncapping of chromosome by these drugs in cancer relative to normal somatic cells was a matter of concern at the early stages of research 59, 60. It is presently known that the affinity of these drugs for tumour cells is significantly higher compared to that for noncancerous cells for not very well understood reasons56, 61.

Chapter 1: General Introduction 6

A number of agents that bind to G-rich promoter sequences and modulate the transcription of proto-oncogenes have been identified and investigated. A porphyrin derivative TMPyP4 down-regulates the expression of c-MYC proto-oncogene in HeLa cells62. The Hurley lab has shown that TMPyP4 binds specifically to the c-MYC nuclease hypersensitive element 63 (NHE) . When the NHEIII1 region of c-MYC in Burkitt‟s lymphoma cell lines was removed, the porphyrin had no effect on c-MYC expression studied63.

There are a few G-quadruplex-binding ligands that have gone through clinical investigations64. Quarfloxin, a fluoroquinolone-based antitumor agent, is the first G-quadruplex interactive small molecule to have progressed through Phase II clinical trials64. This molecule was discovered by Hurley and his colleagues65, 66. It is tolerated well in neuroendocrine cancer patients and its effect is mediated through disruption of the interaction between nucleolin and G- quadruplex in rDNA, an important interaction amplified in tumour cells67. Other prospective G- quadruplex stabilizing molecules currently undergoing clinical trials include CX5461 and AS141167. The quest for G-quadruplex-binding agents for specific therapeutic applications within the genome has evolved into an important component of anticancer strategies.

1.4: Structural Properties of G-quadruplexes

1.4.1 Building Blocks

The building block of a G-quadruplex is a G-quartet, a planar assembly of four guanines interlinked through Hoogsteen hydrogen bonds68. Similar to other DNA structures, G- quadruplexes are composed of nucleotides that involve pentose sugars attached to phosphate groups (Figure 1-1)7. Though there are four bases (adenine, cytosine, thymine, and guanine) that are indiscriminately used in other DNA structures, G-quadruplexes need guanine bases to form their primary building blocks35. The torsion angle created by the bond between the base and the 1‟ carbon of the sugar is referred to as angle χ. Depending on the angle, the nucleotide can be classified as anti (-120<χ<180˚) or syn (0<χ<90˚) (Figure 1-2). The anti configuration is sterically more favourable relative to the syn configuration; the latter can be stabilized through

Chapter 1: General Introduction 7

various external factors35. The degree of this angle is a significant variant among different G- quadruplex conformations35.

Figure 1-1. Components of a Nucleotide35.

Chapter 1: General Introduction 8

Figure 1-2. Base Geometry (Syn and Anti glycosylic bonds).

The pentose sugar configuration is also a distinctive feature of different nucleic acid conformations. The C2‟-endo sugar pucker observed in B-DNA is the stable form in G-tetrads35.

1.4.2 Hydrogen Bonding

Nucleic acid bases contain hydrogen donors (amino or imino hydrogen) and acceptors (carbonyl oxygen or aromatic nitrogen)7. Hydrogen bonding between bases dictates the specific conformation DNA can fold into at given environmental conditions. In a double-helical DNA, pyrimidines form Watson-Crick hydrogen bonds with complementary purine bases. In G- quadruplexes, guanine bases engage in two hydrogen bonds with alternation of Hoogsteen and Watson-Crick faces as shown in Figure 335. In a G-quartet, the nitrogen and the oxygen attached to the ring carbon, N7 and O6, serve as hydrogen bond acceptors with N2-H and N1-H serving donors (Figure 4)35.

Figure 1-3. Guanine Nucleotides35.

Chapter 1: General Introduction 9

Figure 1-4. G-quartet14.

1.4.3 Stacking Interactions

DNA bases are non-polar and their interactions with water, a polar solvent, are energetically unfavourable7, 72. In order to minimize the area exposed to the aqueous environment, nucleic acid bases with planar ringed structures form stacking interactions. In the case of G-quadruplexes, planar G-tetrads, each composed of four guanines, aggregate on top of each other to create a structure similar to a jenga tower71, 72. The stacking that contributes to DNA stability originates from a combination of hydrophobic, electrostatic, and van der Waals forces35, 70-73. When the planar bases gather in close proximity to each other, π orbitals cause repulsion and create an unfavourable energetic state. However, stacked aromatic bases enable formation of π-π bonding central to G-quadruplex stabilization7, 35. Concurrently, the hydrogen atoms around the ring become electron-poor. Base stacking in G-quadruplexes interfaces is

Chapter 1: General Introduction 10

important for their stability 35. Due to steric constraints, G-quartets stack with a twist of 30˚ and a rise of 3.3 Å with each elevation (Figure 5)35. The eight carbonyl oxygen atoms of stacked bases eventually arrange into a bipyramidal antiprismatic construct35, 74.

Figure 1-5. Stacking of G-tetrads35.

1.5: Topological Diversity of G-quadruplexes

G-quadruplexes can be widely classified based on the number of strands involved in the structure. Intramolecular conformations comprise a single strand, whereas intermolecular quadruplex structures can be composed or two or more (2 and 4, typically) sequences35. The orientation of the strand involved in G-quadruplex formation offers another level of classification. If the adjacent strands have opposite directionalities, the conformation is designated as antiparallel, while G-quadruplex is parallel if all strands have the same orientation49. G-quadruplexes can also consist at a mixture of parallel and anti-parallel strands. The conformation adopted by a

Chapter 1: General Introduction 11

G-quadruplex depends on the nucleotide sequence, number of stacked G-tetrads, and numerous environmental factors such as the cation type and concentration and the extent of hydration35.

1.5.1 Tetramolecular G-quadruplexes

A tetramolecular G-quadruplex is composed of four strands rich in guanine bases. The guanines of each strand interact with the guanines from adjacent strands. Theoretically, there are a number of possibilities in which the strands can be oriented, but, experimentally, only the all- parallel structure has been observed35. High resolution X-ray and NMR studies have reported 104 that tetramolecular tetraplexes are extremely stiff . They fold from four XnGmXn or GmXnGm sequences120. However, tetramolecular G-quadruplexes did not receive as much attention because it is not found in human genome.

1.5.2 Bi-molecular G-quadruplexes

When less than four strands come together to form a G-quadruplex structure, loops come into play. The linkers (loops) can be classified as lateral, diagonal, or external14, 35. As shown in Figure 6, the lateral loops of a bimolecular G-quadruplex can be placed either on the same or the opposite side. When the linkers are on the same face, a head-to-head hairpin G-quadruplex forms120. However, when the loops are on the opposite side, all strands become anti-parallel to each other, creating a head-to-tail lateral loop dimer120. When at least one strand is anti-parallel to the other strands, the G-quadruplex conformation is classified as anti-parallel120. The antiparallel configuration has guanines with alternating syn and anti glycosidic angles35.

Chapter 1: General Introduction 12

Figure 1-6. Schematic diagrams of various topologies of bimolecular quadruplexes; (a) lateral loops connecting antiparallel strands to form head-to-head hairpin quadruplex; (b) lateral loops connecting antiparallel strands on different sides; (c) external and lateral loops; (d) external (propeller) chain reversal loops linking parallel strands; (e) diagonal linkers connecting opposite anti-parallel strands35.

Chapter 1: General Introduction 13

1.5.3 Intramolecular G-quadruplexes

The G-quadruplex conformation can be adopted by a monomolecular sequence containing G-tracts, separated by regions that may consist of any combination of residues. The consensus sequence of an intramolecular G-quadruple forming motif is (GmLnGmLpGmLqGm)r where m is the number of guanine repeats, and Ln, Lp, and Lq denote any combination of nucleotides forming loops35.

The length and the sequence of the loops determine the specific G-quadruplex conformation of the DNA molecule. Monomolecular structures may contain propeller, lateral, and diagonal loops 35. Hazel and his colleagues have postulated a general loop-length dependent folding pattern of G-quadruplex structures as described below 117. Using a combination of circular dichroism measurements and molecular dynamics simulations on ten intramolecular G- quadruplexes, it has been reported that three single nucleotide linkers facilitate formation of a parallel structure. When single nucleotide loops are present along with longer loops, DNA may fold into both parallel and anti-parallel G-quadruplexes. Guedin et al. (2010), based on the analysis of more than 80 DNA sequences, have concluded that single nucleotide linkers displayed the highest stability118. The two studies are in agreement with each other as it has been reported that parallel structures are thermodynamically more stable than anti-parallel G- quadruplexes119.

Conformational preferences of intramolecular G-quadruplexes also depend on the identity of the stabilizing cation120. The same sequence may frequently adopt different conformations depending on the cation present in the solution. For instance, the human telomeric + + sequence d(TTAGGG)4 may fold into different conformations in the presence of Na or K 120 + cations . In a Na solution, the 22-base telomeric sequence 5′-AGGG(TTAGGG)3-3‟ forms a basket-type anti-parallel G-quadruplex as shown in Figure 735. However, in a buffer with K+ ions, the same sequence folds into multiple G-quadruplex conformations with comparable free energies121, 122. It has been reported that alternating between Na+ and K+ can trigger interconversion between anti-parallel and parallel G-qudruplexes123. This agrees with one study reporting a minimal energy barrier between parallel and anti-parallel G-quadruplexes adopted by intramolecular sequences with TTA loops124. Thus, the conformational diversity of G-

Chapter 1: General Introduction 14

quadruplexes results from and is govered by the balance between the energetic contributions of a variety of structural and environmental factors and interactions35.

Figure 1-7. Basket-type Anti-parallel G-quadruplex.

The presence of a mixture of conformations complicates thermodynamic and kinetic studies and interpretation of macroscopic results. For longer and more complex sequences, the polymorphic nature of G-quadruplexes becomes even more apparent. The 27 nucleotide long c- MYC parent sequence contains five tracts of three or more guanine bases which, depending on the combination of the guanines involved in G-quartet formation, may adopt structurally diverse G-quadruplex conformations125. Mutations involving shortening the five runs of guanine bases to four or adding bases to elongate the loops may reduce the number of conformations available for sampling125. Conformational conversions may also occur depending on molecules crowding conditions induced by PEG126or the identity and concentration of divalent ions127. Parallel conformations are favoured under molecular crowding conditions126 and when divalent cations are present at high concentrations127.

Chapter 1: General Introduction 15

1.6: Cations and Nucleic Acid Structure

DNA and RNA are highly charged molecules given the backbone rich in negatively charged phosphate groups7. Association of two or more nucleic acid strands accompanying the formation of higher order structures causes a significant increase in charge density. High charge density attracts oppositely charged ions, making the discussion of ionic behavior important when examining the physical properties and the biological role of nucleic acids7, 75-78. Cations are especially important for G-quadruplexes because the structure cannot form in the absence of counterions specifically bound inside the central cavity. The ab initio calculations showed that specific cation binding in the central cavity contributes more to the stabilization of G-quadruplex structures than the hydrogen bonding or stacking interactions79. Electrostatic repulsion caused by the gathering of O6 atoms of four guanine bases in a G-quartet and clustering of phosphate groups must be compensated by a neutralizing effect of counterions35. It is therefore intuitively clear that folding, stability, and conformational preferences of G-quadruplexes are strongly modulated by the type and concentration of the cations present in solution.

1.6.1 DNA as a polyelectrolyte

Polyelectrolyte is a linear polymer with ionizable units along the backbone80. Nucleic acids in solution are natural polyelectrolytes given the abundance of negatively charged phosphates with the resulting high charge density81. The electrostatic interactions between counterions and nucleic acids modulate the ion distribution within the solution, thereby controlling the salt-dependent equilibria of conformations available for sampling82, 83. Stabilization of duplex DNA structures that are crucial for storing and expressing the hereditary information requires counterion condensation neutralizing the repulsive forces originating from the clustering of charged phosphate groups84. The charge density of a duplex is higher compared to that of its isolated constituent single strands. Hence, the degree of counterion condensation around the double helix should be greater relative to the single-stranded conformation, causing counterion release upon the duplex-single strand transition 85. Since maintaining a cloud of counterions is entropically unfavourable with the entropic penalty increasing with a decrease in

Chapter 1: General Introduction 16

the concentration of counterions in the bulk, the thermodynamic stability of duplex DNA increases with salt despite the fact that the local concentration of counterions in the vicinity of a DNA is essentially independent of that in the bulk84, 86, 87. On a similar note, an incrase in salt, generally, results in an increase in the thermal stability of G-quadruplexes87. However, G- quadruplexes, in addition to the polyelectrolyte effect and the related cloud of counterions, bind cations in a specific manner inside their central cavity where they become coordinated to O6 oxygens of the guanines 88, 89.

One of the oldest theoretical tools that has been used to quantify the distribution of mobile ions around polyelectrolytes in solution is the Poisson-Boltzmann (PB) theory90, 91. The PB theory links the electrostatic potential of a system and the charges that contribute to this potential80. The basic Poisson-Boltzmann equation is the following:

²ѱ(r) =-(4π/ε) ρ(r) (1)

(where ѱ(r) is the electrostatic potential at a distance r in a medium with a dielectric constant ε and a charge density ρ)92. The charge density is given by the sum qₑƩzᵢnᵢ(r) with the local concentration (ions/mL) of ions of type i, nᵢ(r), and the ionic charge of qₑzᵢ. The Boltzmann equation relates the local concentration of a species to its bulk concentration nbi and energy Ei(r) at r, which is determined by its charge and potential at r92.

(2)

Substituting equation (2) into equation (1) yields the final Poisson-Boltzmann equation92,

Chapter 1: General Introduction 17

(3)

The PB equation is a mean-field theory with the mobile ions in the region far from the polyion suface affecting each other only through their average contribution to the mean-field potential. In other words, the PB theory does not take into consideration the ion-ion correlation80. The PB equation is mathematically complicated because it is non-linear with the exponential term containing the electrostatic potential92. In addition, the PB theory incorporates the Gouy- Chapman model where the polyelectrolyte is viewed as an infinitely long double-layered cylinder with condensed ions placed within the space between the inner and the outer layers93. The mean charge density is measured by the number of charges present93.

The counterion condensation theory described by Manning simplifies the mathematical complexity of the PB theory, although the two theories are based on the same basic principle94-96. The counterions cluster around the polyelectrolyte, while the ions of the opposite charge are excluded from the vicinity85. The number of accumulated counterions depends only on the valency of the counterion and the axial charge density of the polyelectrolyte, being independent of the concentration in the bulk solution84. In the case of DNA, the theory predicts that cations are accumulated and anions are excluded and that the bulk concentration of ions does not affect the local concentration of the counterions in the close vicinity of the DNA. The counterions are free to change positions along the iso-potential energy paths created by the phosphates85.

If a macromolecule can be modeled by a linear array of N univalent (negative) charges with an average of spacing b, the extent of association of monovalent counterions M+ with the polyanion can be described by a single dimensionless parameter ξ84,86.

ξ=q2/εkTb

Chapter 1: General Introduction 18

Here, q signifies the protonic charge, ε is bulk dielectric constant of solvent, k is Boltzmann‟s constant, T is the thermodynamic temperature, and b is average axial intercharge spacing along the helical axis for DNA or along the contour axis for more flexible polymers.

If only univalent ions are present, the ξ value greater than unity signifies that the molecule is highly charged. On the other hand, if the net charge density is less than one, counterions do not condense 7, 84. The counterion condensation theory, as proposed by Manning and Record, is useful in studying the ionic distribution in polyelectrolyte solutions. However, the basis of this model that presents the polyelectrolyte as a linear charge with no thickness and counterions as point charges has limitations in explaining the competitive binding of distinct counterion molecules92.

1.6.2 Salt Effects in G-quadruplexes

Among the major classes of nucleic acid structures, G-quadruplexes are unique in that they interact with cations in both specific and non-specific manners97. The specific binding occurs within the central cavity where mono- or divalent cations of an appropriate size form inner sphere coordination bonds with O6 oxygens of guanine bases98, 99.

Arnott and colleagues (1974) have pioneered in suggesting that the O6 carboxyl oxygens of guanines form coordination bonds with cations100. Sequestered cations serve to weaken unfavourable interactions between carbonyl oxygens and consequently promote base-base stacking interactions102, 103. In agreement with this notion, molecular dynamics simulations reveal that G-quadruplexes are destabilized in the absence of centrally-bound cations104, 105.

The effect of specific internal cations was clearly demonstrated through studies observing the folding of G-quadruplexes in the atmosphere of Na+ and K+ ions106. The folding of the d(TG4T) G-quadruplex increases by a factor of 200 following a 10-fold increase in the concentration of Na+ ions35. The physiological abundance and functionality of K+ and Na+ ions particularly make them important targets of characterization in G-quadruplex studies35.

Chapter 1: General Introduction 19

Depending on the size, centrally bound cations may be located within the G-quartet plane 107 + + or be sandwiched between the planes . Ions such as K and NH4 (with radii 1.33 and 1.48 Å , respectively) are too large to be coordinated in the plane of a G-quartet, whereas Na+ (with a radius 0.95Å ) fits within the plane of a G-quartet. In general, n-1 large cations are 107 accommodated in n G-quartets . In this respect, Hud and collaborators used the [d(G4T4G4)]2 + 108 sequence to demonstrate that three NH4 ions are located within the four quartets . In contrast, the smaller Na+ ion allows for in-plane coordination. Multiple Na+ ions are not as restricted to the localized space as larger ions; they can move further away from each other to reduce electrostatic repulsions109.

An early study aimed to characterize the role of cations in the stability of G-quadruplexes have produced the following order of the stabilizing action of cations K+ > Ca2+ > Na+ > Mg2+ > Li+ and K+ > Rb+ > Cs+110. A subsequent study has revealed the following ranking of the stabilizing role of cations Sr2+ > Ba2+> Ca2+ > Mg2+ and K+ > Rb+ > Na+ > Li+ = Cs+ 111. In agreement with this result, Sr2+ and Ba2+ induce a thermodynamically more stable thrombin aptamer G- quadruplex than that induced by Mg2+ and Ca2+112.

Cations do not always make G-quadruplexes stable. Certain counterions can also destabilize the structure. Even at low concentrations, divalent transition metal cations such as Mn2+, Co2+or Ni2+ have been observed to disrupt K+-induced G -quadruplex structures113. Some studies have also shown that the destabilizing effect of divalent cations depends on the concentration of monovalent ions114. Hardin and his colleagues reported that divalent cations of concentration 10-40 mM stabilize the G-quadruplex conformation when the concentration of monovalent counterions is higher than 50mM. However, when the monovalent cation concentration is lower than 50mM, divalent ions destabilize G-quadruplex structures114. Although Li+ is commonly suggested to destabilize G-quadruplexes115, there are some indications that it neither stabilizes nor destabilizes G-quadruplex structures 35.

Chapter 1: General Introduction 20

Table 1-1. Ionic radii of G-quadruplex stabilizing cations35.

1.7: Energetics/Thermodynamics of Quadruplex Formation

G-quadruplexes have two or more G-tetrads stacked on top of another. Some researchers argue that G-quadruplex formation is accompanied by a decrease in free energy equal to or even exceeding that associated with Watson-Crick base pair formation128-131. Other researchers, however, suggest that G-quadruplex stability depends stronger on the identity of cations, the length and sequence of loops, and the presence co-solvents132. Kogut et al. (2016) have measured the energetics of dissociation of a single G-quartet from the telomeric G-quadruplex133. The change in free energy results from an interplay between the favourable ΔH and unfavourable – TΔS terms133. A change in enthalpy, ΔH, is around -35 kcal/mol for the tetrad formed in the telomeric sequence, whereas the value of –TΔS is 29 kcal/mol133. Thus, G-quartet formation and the subsequent stacking is an enthalpically driven event. This notion is in agreement with the results of Boncina et al. (2012) and Sengar et al. (2014) which reveal that G-quadruplex folding is enthalpy driven with an enthalpy change per G-quartet of -14 to -19 kcal/mol129, 134.

The van‟t Hoff analysis of melting curves enables one to calculate changes in enthalpy of association or dissociation. The applicability of this analysis in intermolecular G-quadruplexes is based on the two-state assumption35. Nonetheless, it has been reported that the enthalpies of intermolecular G-quadruplexes are comparable to those of intramolecular structures35.

Calorimetric studies of the (TG4T)4 tetramolecular G-quaruplex reveal a change in enthalpy of - 20 kcal/mol per quartet106, 135.

Chapter 1: General Introduction 21

In contrast to consistent values of enthalpy reported in literature, the reported changes in entropy depend strongly on the molecularity of G-quadruplexes and the loop length and identity35, 132.

1.8: Solvent Effects on the Stability of G-quadruplexes

In aqueous solutions, macromolecules are hydrated157. Hydration refers to the binding of water molecules to solutes159. It has been reported that hydration of some biopolymers is not limited to a single layer of water molecules but can involve two or more layers of water molecules158. The presence of cosolvents modulates the solute-solvent interactions of biomolecules and changes the hydrogen-bond dynamics of water molecules136. Nucleic acids reside and function in cellular environment where biomolecules occupy about 1/3rd of the whole volume137 with the macromolecular concentration ranging from 50 to 400 g/L138. DNA molecules including G-quadruplexes are surrounded by water molecules forming a hydrogen bonding network as illustrated in Figure 8.

Chapter 1: General Introduction 22

Figure 1-8. Guanine Tetrad and Associated Waters35.

The stability of DNA and RNA duplexes and triplexes involving Watson-Crick hydrogen bonds decreases with an increase in the concentration of crowding agents, but the effect of these agents on the stability of G-quadruplexes is opposite139. Sugimoto and coworkers have reported that cosolvents can induce transitions between G-quadruplex conformations141. In particular, polyethylene glycol (PEG) triggers a transition of d(G4T4G4) from the antiparallel to a more stable parallel conformation141. A similar result has been reported by Li et al. (2005)142. These findings suggest that the energetics of G-quadruplex polymorphism is modulated not only by nucleotide sequence but also by environmental factors including solvent composition.

The antiparallel G-quadruplex formed by a 15-base thrombin aptamer is stabilized under the molecular crowding conditions at various concentrations of PEG 200143. The stability of the G-quadruplex decreases linearly as the activity of water increases143. The number of water molecules released per nucleotide upon formation of an anti-parallel G-quadruplex has been estimated to be 4.5143. The release of water molecules upon G-quadruplex formation suggests

Chapter 1: General Introduction 23

that the folded conformation is less hydrated than the unfolded form. In contrast, according to osmotic stress experimental results, the folded state of a DNA duplex is more hydrated than the unfolded state143. The number of water molecules taken up following the formation of a duplex is 3.4 per nucleotide143. Thus, under molecular crowding conditions, non-canonical DNA structures like G-quadruplexes may be favoured over duplexes.

1.9: i-motif Nucleic Acid Structures

In the genomic DNA, G-quadruplex-forming guanine-rich strands are complementary to strands rich in cytosine, with an exemption of the telomeric single- stranded 3‟-overhang. In 1993, Gehring and his colleagues published the first report of the unique structure adopted by these cytosine-rich sequences (Figure 9)145. These tetraplex structures consist of two parallel duplexes are intercalated in an antiparallel manner. They are, generally, formed at slightly acidic conditions facilitating hemi- protonation of cytosines and formation of cytosine-cytosine+ base pairs (Figure 9).

Figure 1-9. Cytosine·Cytosine+ Base Pair145.

Chapter 1: General Introduction 24

There are two topological forms of i-motifs. In one form (3‟E), the cytosine-cytosine base pair is located at the 3‟ end, while in the other form (5‟E), the base-pairing occurs at the 5‟ end160. They are named 3‟-E (R-form) and 5‟-E (S-form) 35. Some sequences show a tendency to interconvert between the two forms consistent with the picture in which the energy barrier between the two configurations is not large. The third less common topology is the T-form where the base pair at both ends are not intercalated35. Analogous to G-quadruplexes, i-motifs may be inter- or intramolecular35.

Although i-motifs generally exist under slightly acidic conditions, they also can form under neutral146 or even basic conditions147. In particular, experiments conducted under conditions mimicking the crowded environment of the cell reveal that i-motifs can form at neutral pH146. Negative superhelicity has also been shown to facilitate the i-motif formation148. In one study by Sun and Hurley (2009), the c-MYC oncogene promoter sequence has been placed into a supercoiled plasmid148. Importantly, the G-rich and C-rich strands of the sequence have been shown to fold into the G-quadruplex and i-motif conformations, respectively148. These authors have used enzymatic and chemical footprinting to show that negative superhelicity supports the formation of tetraplex structures even at neutral pH148. Paradoxically, a more recent paper has refuted this finding and argued that supercoiling itself is not a sufficient condition to induce G-quadruplex formation in other sequences149. Thus, further studies are required to clarify the influence of supercoiling on tetraplex structure formation in the genomic DNA.

1.10: Methods to Study G-quadruplex Structures

A basic structural method of studying G-quadruplex structures is X-ray crystallography161. This technique involves crystallization of purified DNA sample161. By analyzing the diffraction patterns of the X-ray shone on the crystal structure, one can determine the three-dimensional image of the DNA of interest162. One disadvantage of X-ray crystallography is that only the solid state of a DNA structure can be detected161. The technique is of limited applicability when the structure is polymorphic, which is a common case for G-quadruplexes161. X-ray crystallography will detect and characterize only the conformation most favoured by crystallization161.

Chapter 1: General Introduction 25

Another technique that provides detailed structural data about G-quadruplex structures is nuclear magnetic resonance (NMR) spectroscopy161. Though high purification is needed, this method demands less work for sample preparation compared to X-ray crystallography161. NMR is useful for G-quadruplexes because guanine imino protons show a unique shift when they form hydrogen bonding161. In addition, NMR provides information about the polymorphism because it can be used for dynamics and kinetics studies161.

Native gel electrophoresis probes the anomalous migratory behaviour of folded G- quadruplexes relative to the unfolded DNA150. Owing to their compact structure, intramolecular G-quadruplexes migrate faster through a cation-containing gel than their linear counterparts151. To characterize other possible structures present in the bands, one must pay a careful attention to the speed (voltage) of the run151. DNA can be visualized through labeling or EtBr staining35.

Circular dichroism (CD) spectroscopy has been traditionally used as a tool that discriminates between various G-quadruplex conformations35. This method involves a sample exposed to circularly polarized light. If there is a chiral center in the molecule of interest, it will generally interact asymmetrically with the light, with the asymmetry varying with the wavelength. Although it may be difficult to unequivocally link a CD spectrum to a specific DNA structure, characteristic spectra corresponding to different G-quadruplex conformations have been determined empirically35. If used over a range of temperatures, CD spectroscopy can be used to observe a melting transition of a G-quadruplex and hence determine thermodynamic parameters such as the Tm, ΔH, and ΔG used to quantify structural stabilities35.

UV spectroscopy is a technique that can monitor the dissociation/association and folding/unfolding reactions of DNA structures35. For a duplex DNA, a hyperchromic shift is indicative of the unstacking of bases and unfolding of the structure35. This method has been applied to G-quadruplexes with UV absorption changes monitored at 295nm35, 121. The Tm, the temperature at which 50% of the structures is unfolded, can be determined from the melting profile and additional thermodynamic data can be obtained based on the two-state approximation of the melting curve35, 154. The temperature dependence of the equilibrium constant of the reaction yields the van‟t Hoff enthalpy of the reaction.

Chapter 1: General Introduction 26

Differential scanning calorimetry (DSC) is a fundamental method for examining the thermal stability of G-quadruplexes163. DSC provides a thermogram in which the apparent heat capacity is measured as a function of temperature163. This technique allows a direct measurement of the enthalpy change resulting from heat-induced unfolding of DNA163. The model- independent enthalpy obtained from DSC can be compared to the model-dependent van‟t Hoff enthalpy to determine if the transition contains intermediates163. If the two enthalpies coincide, the transition proceeds as a two-state event163. However, if the values are different, the transition involves intermediates or aggregation takes place163.

A powerful method for detecting and characterizing the G-quadruplex configuration and dynamics is single-molecule Fӧrster resonance energy transfer (FRET) 155. In FRET, oligonucleotides are labeled with a donor (e. g., 6-carboxy fluorescein FAM at 5‟ end) and an acceptor (e. g., tetramethyl-6-carboxyrhodamine TMA at the 3‟ end) fluorophores35. Upon DNA folding, the donor fluorophore transfers its energy to the acceptor, with an efficiency that depends on their distance and relative orientation. When the donor and the acceptor are close, the fluorescence is quenched, but when apart, the molecule becomes detectable with fluorescence. G-quadruplex melting proceeds with the separation of the two fluorophores with an increase in temperature. This technique is most appropriate for elongated sequences that form intramolecular G-quadruplexes with the folded conformation bringing the two ends into close proximity35.

1.11: Gap in Literature and the Experimental Layout

The importance of G-quadruplexes as therapeutic targets continually increases with their in vivo existence and important functionalities being revealed on a virtually, daily basis. Hence, the controlled modulation of G-quadruplex structures and the search of drugs triggering G- quadruplex formation being of high interest to researchers. Enhancing or inhibiting the formation of G-quadruplexes at specific sites within the genome is a promising approach to control expression of proto-oncogenes and control and prevent the immortality of tumour cells. Though there are many factors that contribute to stabilization of G-quadruplexes, one universal component that exists only in G-quadruplexes is the specific ion binding within the central cavity.

Chapter 1: General Introduction 27

The thermal stability of G-quadruplexes have been shown to strongly depend on the concentration of salt35. However, prior to our work, the effects of specific and non-specific cations have not been separated. Ionic effects are more complex in G-quadruplexes, since there are two types of ion binding. This thesis aims at resolving the differential effect of specific and non-specific ionic binding on the stability of G-quadruplexes. Our laboratory has devised an experimental approach to separate the stabilizing effect of centrally bound cations from that of the cloud of counterions. The method involves the initial saturation of the G-quadruplex by the stabilizing ions (Na+ or K+) followed by the addition of either a stabilizing or a non-stabilizing cation (Li+, Cs+, TMA+). This methodology, which started with telomeric sequences, has been subsequently applied to sequences derived from the promoter regions of the c-Myc and VEGF oncogenes.

Chapter 1: General Introduction 28

1.12 Reference

1. Avery, O., MacLeod, C., and McCarty, M. (1944). Studies on the Chemical Nature of the Substance Inducing Transformation of Pneumococcal Types. J. Expt. Med., 79, 137-157.

2. Hershey, A. D. and Chase, M. (1952). Independent Functions of Viral Protein and Nucleic Acid in Growth of Bacteria. J. Gen. Physiol., 36, 39-56.

3. Pray, L. (2008). Discovery of DNA structure and function: Watson and Crick. Nature Education, 1(1), 100.

4. Watson, J. D. and Crick, F. H. C. (1953). A structure for deoxyribose nucleic acid. Nature, 171, 737–738.

5. Dahm, R. (2005). Friedrich Miescher and the discovery of DNA, Developmental Biology, 278, 274–288.

6. Arnott. S. and Hukins, D. W. (1973). Refinement of the structure of B-DNA and implications for the analysis of X-ray diffraction data from fibers of biopolymers. J Mol Biol., 81(2), 93- 105.

7. Bloomfield, V. A., Crothers, D. M., and Tinoco, I. (2000). Nucleic acids: structures, properties, and functions. University Science Books.

8. Bang, I. (1910). Untersuchungen uber die Guanylsare. Biochem. Z., 26, 293–311

9. Fritzsche, W. and Spindler, L. (2013). Guanine Quartets: Structure and Application. Cambridge: Royal Society of Chemistry.

10. Kossel, A. (1901). Erwiderung. Z. Physiol. Chem., 31, 410.

11. Gellert, M., Lipsett, M.N., and Davies, D.R. (1962). Helix Formation by Guanylic Acid. Proc. Nat. Acad. Sci. USA, 48, 213–218.

Chapter 1: General Introduction 29

12. Huppert, J.L. and Balasubramanian, S. (2005). Prevalence of quadruplexes in the human genome. Nucleic Acids Res., 33, 2908–2916.

13. Biffi, G., Tannahill, D., McCafferty, J., and Balasubramanian, S. (2013) Quantitative visualization of DNA G-quadruplex structures in human cells. Nat. Chem., 5, 182–186.

14. Bryan, T. M. and Baumann, P. (2011). G-quadruplexes: From Guanine Gels to Chemotherapeutics. Mol. Biotechnol., 49(2): 198–208.

15. Blackburn, E. H. (2001). Switching and signaling at the telomere. Cell, 106, 661 – 673.

16. Wright, W. E., Tesmer, V. M., Huffman, K. E., Levene, S. D., and Shay, J. W. (1997), Genes Dev., 11, 2801–2809.

17. Lim, K.W., Ng, V.C., Martin-Pintado, N., Heddi B., and Phan A.T. (2013). Structure of the human telomere in Na+ solution: an antiparallel (2+2) G-quadruplex scaffold reveals additional diversity. Nucleic Acids Res., 41, 10556 – 10562.

18. Blackburn, E.H. and Gall J. G. (1978). A tandemly repeated sequence at the termini of the extrachromosomal ribosomal RNA genes in Tetrahymena. J. Mol. Biol., 120(1), 33-53.

19. Collins, K., Kobayashi, R., and Greider, C.W. (1995). Purification of Tetrahymena telomerase and cloning of genes encoding the two protein components of the enzyme. Cell, 81(5), 677-86.

20. Zahler, A. M. and Prescott, D. M. (1988).Telomere terminal transferase activity in the hypotrichous ciliate Oxytricha nova and a model for replication of the ends of linear DNA molecules. Nucleic Acids Res., 16(14B), 6953-6972.

21. Harley, C.B., Futcher A.B., and Greider C.W. (1990). Telomeres shorten during ageing of human fibroblasts. Nature, 345(6274), 458-60.

22. Hastie, N.D., Dempster, M., Dunlop, M.G., Thompson, A.M., Green, D.K., and Allshire, R.C. (1990). Telomere reduction in human colorectal carcinoma and with ageing. Nature, 346(6287), 866-8.

Chapter 1: General Introduction 30

23. Lindsey, J., McGill, N.I., Lindsey, L.A., Green, D.K., and Cooke, H.J. (1991). In vivo loss of telomeric repeats with age in humans. Mutation Research , 256(1), 45-48.

24. Vaziri, H., Schächter, F., Uchida, I., Wei, L., Zhu, X., Effros, R., Cohen, D., and Harley, C.B. (1993). Loss of telomeric DNA during aging of normal and trisomy 21 human lymphocytes. Am J Hum Genet., 52(4), 661-667.

25. Slagboom, P. E., Droog, S., and Boomsma, D. (1994). Genetic Determination of Telomere Size in Humans: A Twin Study of Three Age Groups. Am. J. Hum. Genet., 55, 876-882.

26. Shin, J. S., Hong, A., Solomon, M. J., and Lee, C. S. (2006). The role of telomeres and telomerase in the pathology of human cancer and aging. Pathology, 38, 103–113.

27. Blackburn, E.H., (1990) Telomeres: structure and synthesis. J. Biol. Chem. 265, 5919.

28. Kim, N. W., Piatyszek, M. A., Prowse, K. R., Harley, C. B., West, M. D., Ho, P.L., Coviello, G. M., Wright, W. E., Weinrich, S. L., and Shay, J. W. (1994). Specific association of human telomerase activity with immortal cells and cancer. Science, 266, 2011–2015.

29. Shammas, M.A. (2011). Telomeres, lifestyle, cancer, and aging. Curr. Opin. Clin. Nutr. Metab. Care. 14(1), 28–34.

30. Lee, H.W., Blasco, M.A., Gottlieb, G.J., Horner, J.W., Greider, C.W., and DePinho, R.A. (1998). Essential role of mouse telomerase in highly proliferative organs. Nature, 392(6676), 569-574.

31. Hahn, W. C., Counter, C. M., Lundberg, A. S., Beijersbergen, R. L., Brooks, M. W., and Weinberg, R. A. (1999). Creation of human tumour cells with defined genetic elements. Nature, 400, 464–468.

32. Greider, C. W. and Blackburn, E. H. (1985). Identification of a specific telomere terminal transferase activity in Tetrahymena extracts. Cell, 43, 405-413.

33. Riou, J. F., Guittat, L., Mailliet, P., Laoui, A., Renou, E., Petitgenet, O. Megnin-Chanet, F. Helene, C., and Mergny, J. L. (2002). Proc. Natl. Acad. Sci. USA, 99, 2672.

Chapter 1: General Introduction 31

34. Tauchi, T., Shin-Ya, K., Sashida, G., Sumi, M., Nakajima, A., Shimamoto, T., Ohyashiki, J.H., and Ohyashiki, K. (2003). Oncogene, 22, 5338.

35. Neidle, S. and Balasubramanian, S. (2006). Quadruplex Nucleic Acids. Cambridge: RSC Biomolecular Sciences.

36. Huppert, J. L. and Balasubramanian, S. (2007). G-quadruplexes in promoters throughout the human genome. Nucleic Acids Res., 35, 406–413

37. Oster, S. K., Ho, C. S. W., Soucie, E. L., and Penn, L. Z. (2002). The myc oncogene: MarvelouslY Complex. Adv. Cancer Res., 84, 81−154.

38. Eisenman, R. N. (2001). Deconstructing Myc. Genes Dev., 15, 2023−2030.

39. Wierstra, I., and Alves, J. (2008). The c-myc Promoter: Still MysterY and Challenge. Adv. Cancer Res., 99, 113−333.

40. Fahrlander, P. D., Sumegi, J., Yang, J. Q., Wiener, F., Marcu, K. B., and Klein, G. (1985). Proc. Natl. Acad. Sci. USA, 82, 3746.

41. Alitalo, K., Schwab, M., Lin, C.C., Varmus, H.E., and Bishop, J.M. (1983). Proc, Natl. Acad. Sci. USA, 80, 1707.

42. Siddiqui-Jain, A., Grand, C. L., Bearss, D. J., and Hurley, L.H. (2002). Direct evidence for a G-quadruplex in a promoter region and its targeting with a small molecule to repress c-MYC transcription. Proc. Natl. Acad. Sci.USA, 99, 11593–11598.

43. Ferrara, N. (1996). Vascular endothelial growth factor. Eur. J. Cancer, 32A, 2413−2422.

44. Hoeben, A., Landuyt, B., Highley, M. S., Wildiers, H., Van Oosterom, A. T., and De Bruijn, E. A. (2004). Vascular endothelial growth factor and angiogenesis. Pharmacol. Rev., 56, 549−580.

45. Sun, D., Guo, K. Rusche, J.J. and Hurley, L. H. (2005). Facilitation of a structural transition in the polypurine/polypyrimidine tract within the proximal promoter region of the human

Chapter 1: General Introduction 32

VEGF gene by the presence of potassium and G-quadruplex-interactive agents. Nucleic Acids Res., 33(18), 6070-80.

46. Nishida, N., Yano, H., Nishida, T., Kamura, T., and Kojiro, M. (2006). Angiogenesis in cancer. Vascular Health and Risk Management, 2(3), 213-219.

47. Goodsell, D. S. (2003). The molecular perspective: VEGF and angiogenesis. Stem Cells, 21, 118–119.

48. Zahler A. M., Williamson JR, Cech T. R., and Prescott DM. (1991). Inhibition of telomerase by G-quartet DNA structures. Nature, 350(6320), 718-20.

49. Bryan, T. M. and Baumann, P. (2011). G-Quadruplexes: From Guanine Gels to Chemotherapeutics. Mol. Biotechnol., 49(2), 198–208.

50. Wheelhouse, R. T., Sun, D. K., Han, H. Y., Han, F. X., and Hurley, L. H. (1998). Cationic porphyrins as telomerase inhibitors: the interaction of tetra-(N-methyl-4-pyridyl)porphine with quadruplex DNA. J. Am. Chem. Soc., 120, 3261–3262.

51. De Cian, A., Cristofari, G., Reichenbach, P., De Lemos, E., Monchaud, D., Teulade-Fichou, M. P., Shin-Ya, K., Lacroix, L., Lingner, J., and Mergny, J. L. (2007). Re-evaluation of telomerase inhibition by quadruplex ligands and their mechanisms of action. Proc. Natl. Acad. Sci. USA, 104, 17347–17352.

52. Fedoroff, O. Y., Salazar, M., Han, H., Chemeris, V. V., Kerwin, S. M., and Hurley, L. H. (1998). NMR Based model of a telomerase-inhibiting compound bound to G-quadruplex DNA. Biochemistry, 37, 12367–12374.

53. De Cian, A., Delemos, E., Mergny, J. L., Teulade- Fichou, M. P., and Monchaud, D. (2007). Highly efficient G-quadruplex recognition by bisquinolinium compounds. J. Am. Chem. Soc., 129, 1856–1857.

Chapter 1: General Introduction 33

54. Rezler, E. M., Seenisamy, J., Bashyam, S., Kim, M. Y., White, E., Wilson, W. D., and Hurley, L. H. (2005). Telomestatin and diseleno sapphyrin bind selectively to two different forms of the human telomeric G-quadruplex structure. J. Am. Chem. Soc., 127, 9439–9447.

55. Gomez, D., Paterski, R., Lemarteleur, T., Shin-Ya, K., Mergny, J. L., and Riou, J. F. (2004). Interaction of telomestatin with the telomeric single-strand overhang. J. Biol. Chem., 279, 41487–41494.

56. Tahara, H., Shin-Ya, K., Seimiya, H., Yamada, H., Tsuruo, T., and Ide, T. (2006). G- Quadruplex stabilization by telomestatin induces TRF2 protein dissociation from telomeres and anaphase bridge formation accompanied by loss of the 3‟ telomeric overhang in cancer cells. Oncogene, 25, 1955–1966.

57. Tauchi, T., Shin-Ya, K., Sashida, G., Sumi, M., Nakajima, A., Shimamoto, T., Ohyashiki, J. H., and Ohyashiki, K. (2003). Activity of a novel G-quadruplex-interactive telomerase inhibitor, telomestatin (SOT-095), against human leukemia cells: involvement of ATM dependent DNA damage response pathways. Oncogene, 22, 5338–5347.

58. Gomez, D., Wenner, T., Brassart, B., Douarre, C., O‟Donohue, M. F., El Khoury, V., Shin- Ya, K., Morjani, H., Trentesaux, C., and Riou, J. F. (2006). Telomestatin-induced telomere uncapping is modulated by POT1 through G-overhang extension in HT1080 human tumor cells. J. Biol. Chem., 281, 38721–38729.

59. Burger, A. M., Dai, F., Schultes, C. M., Reszka, A. P., Moore, M. J., Double, J. A., and Neidle, S. (2005). The G-quadruplex-interactive molecule BRACO-19 inhibits tumor growth, consistent with telomere targeting and interference with telomerase function. Cancer Res., 65, 1489–1496.

60. Leonetti, C., Amodei, S., D‟Angelo, C., Rizzo, A., Benassi, B., Antonelli, A., Elli, R., Stevens, M.F., D‟Incalci, M., Zupi, G., and Biroccio, A. (2004). Biological activity of the G- quadruplex ligand RHPS4 (3, 11-difluoro-6, 8, 13-trimethyl-8H-quino[4, 3, 2-kl]acridinium methosulfate) is associated with telomere capping alteration. Mol. Pharmacol., 66, 1138– 1146.

Chapter 1: General Introduction 34

61. Salvati, E., Leonetti, C., Rizzo, A., Scarsella, M., Mottolese, M., Galati, R., Sperduti, I., Stevens, M.F., D‟Incalci, M., Blasco, M., Chiorino, G., Bauwens, S., Horard, B., Gilson, E., Stoppacciaro, A., Zupi, G., and Biroccio, A. (2007). Telomere damage induced by the G- quadruplex ligand RHPS4 has an antitumor effect. J. Clin. Invest., 117, 3236–3247.

62. Grand, C.L., Han, H., Munoz, R.M., Weitman, S., Von Hoff, D.D., Hurley, L.H., and Bearss D.J. (2002).The cationic porphyrin TMPyP4 down-regulates c-MYC and human telomerase reverse transcriptase expression and inhibits tumor growth in vivo. Mol. Cancer Ther., 1(8), 565–573.

63. Siddiqui-Jain, A., Grand, C.L., Bearss, D.J., and Hurley, L.H. (2002). Direct evidence for a G-quadruplex in a promoter region and its targeting with a small molecule to repress c-MYC transcription. Proc. Natl. Acad. Sci. USA, 99, 11593-8.

64. Drygin, D., Siddiqui-Jain, A., O'Brien, S., Schwaebe, M., Lin, A., Bliesath, J., Ho, C.B., Proffitt, C., Trent, K., Whitten, J.P., Lim, J.K.C., Von Hoff, D., Anderes, K., and Rice, W.G. (2009). Anticancer Activity of CX-3543: A Direct Inhibitor of rRNA Biogenesis. Cancer Research, 69(19), 7653–7661.

65. Kim, M.Y., Duan, W.H., Gleason-Guzman, M., and Hurley, L.H. (2003). Design synthesis and biological evaluation of a series of fluoroquinoanthroxazines with contrasting dual mechanisms of action against topoisomerase II and G-quadruplexes. J. Med. Chem., 46(4), 571–583.

66. Duan, W., Rangan, A., Vankayalapati, H., Kim, M.Y., Zeng, Q., Sun, D., Han, H., Fedoroff, O.Y., Nishioka, D., Rha, S.Y., Izbicka, E., Von Hoff, D.D., and Hurley, L.H. (2001). Design and synthesis of fluoroquinophenoxazines that interact with human telomeric G- quadruplexes and their biological effects. Mol. Cancer Ther., 1(2), 103-20.

67. Drygin, D., Siddiqui-Jain, A., O'Brien, S. Schwaebe, M. Lin, A., Bliesath, J., Ho, C.B. Proffitt, C., Trent, K., Whitten, J.P. Lim, J.K.C. Von Hoff, D. Anderes, K. and Rice, W.G. (2009). Anticancer activity of CX-3543: a direct inhibitor of rRNA biogenesis. Cancer Res., 69, 7653–7661.

Chapter 1: General Introduction 35

68. Bates, P. J., Laber, D. A., Miller, D. M., Thomas, S. D., and Trent, J. O. (2009). Discovery and development of the G-rich oligonucleotide AS1411 as a novel treatment for cancer. Exp. Mol. Pathol., 86, 151-164.

69. Bates, P. J., Laber, D. A., Miller, D. M., Thomas, S. D., and Trent, J. O. (2009). Discovery and development of the G-rich oligonucleotide AS1411 as a novel treatment for cancer. Exp. Mol. Pathol., 86, 151-164.

70. Lech, C. J., Heddi, B., and Phan, A. T. (2013). Guanine base stacking in G-quadruplex nucleic acids. Nucleic Acids Research, 41(3), 2034–2046.

71. Hobza, P. (2006). Nature of base stacking: reference quantum-chemical stacking energies in ten unique B-DNA base-pair steps. Chem. Eur. J., 12, 2854–2865.

72. Sponer, J., Riley, K. E., and Hobza, P. (2008).Nature and magnitude of aromatic stacking of nucleic acid bases. Phys. Chem. Chem. Phys., 10, 2595–2610.

73. Dabkowska, I., Gonzalez, H. V., Jurecka, P., and Hobza, P. (2005).Stabilization energies of the hydrogen-bonded and stacked structures of nucleic acid base pairs in the crystal geometries of CG, AT, and AC DNA steps and in the NMR geometry of the 50- d(GCGAAGC)-30 hairpin: complete basis set calculations at the MP2 and CCSD(T) levels. J. Phys. Chem. A, 109, 1131–1136.

74. Zhang, D. Huang, T., Lukeman, P. S., and Paukstelis, P. J. (2014). Crystal structure of a DNA/Ba2+ G-quadruplex containing a water-mediated C-tetrad. Nucleic Acids Res., 42(21), 13422–13429.

75. Saenger, W. (1984). Principles of Nucleic Acid Structure. Springer Verlag, New York.

76. Cantor, C. R. and Schimmel, P. R. (1980). The Conformation of Biological Macromolecules, W. H. Freeman, San Francisco.

77. Cantor, C. R. and Schimmel, P. R. (1980). Techniques for the Study of Biological Structure and Function, W. H. Freeman, San Francisco.

Chapter 1: General Introduction 36

78. Cantor, C. R. and Schimmel, P. R. (1980). The Behavior of Biological Macromolecules, W. H. Freeman, San Francisco.

79. Gu, J., Leszczynski, J., and Bansal, M. A. (1999). New insight into the structure and stability of Hoogsteen hydrogen-bonded G-tetrad: an ab initio SCF study. Chem. Phys. Lett., 311, 209–214.

80. Ciferri, A. and Perico, A. (2012). Ionic Interactions in Natural and Synthetic Macromolecules. John Wiley & Sons, Inc.

81. Lohman, T. M. and Mascotti, D. P. (1992). Thermodynamics of ligand-nucleic acid interactions. Methods in Enzymology, 212, 400-424.

82. Record, M. T., Anderson, C. F., and Lohman, T. M. (1978). Thermodynamic analysis of ion effects on the binding and conformational equilibria of proteins and nucleic acids: the roles of ion association or release, screening, and ion effects on water activity. Quart. Rev. Biophys., 11, 103-178.

83. Anderson, C.F. and Record, M.T. (1995). Salt-nucleic acid interactions. Annu. Rev. Phys. Chem., 46, 657-700.

84. Manning, G. S. (1978). The molecular theory of polyelectrolyte solutions with applications to the electrostatic properties of polynucleotides. Quart. Rev. Biophys., 11, 179-246.

85. Volker, J., Klump, H. K., Manning, G. S., and Breslauer, K. J. (2001). Counterion association with native and denatured nucleic acids: an experimental approach. Journal of Molecular Biology, 310 (5), 1011-1025.

86. Manning, G. S. (2007). Counterion condensation on charged spheres, cylinders, and planes, J. Phys. Chem. B, 111, 8554–8559.

87. Anderson, C.F. and Record, M.T. (1982). Polyelectrolyte theories and their applications to DNA, Annu. Rev. Phys. Chem., 33, 191–222.

Chapter 1: General Introduction 37

88. Tikhomirova, A., Beletskaya, I. V., and Chalikian, T. V. (2006). Stability of DNA Duplexes Containing GG, CC, AA, and TT Mismatches. Biochemistry, 45, 10563-10571.

89. Shek, Y. L., Noudeh, G. D., Nazari, M., Heerklotz, H., Abu-Ghazalah, R. M. Dubins, D. N., and Chalikian, T. V. (2013). Folding thermodynamics of the hybrid-1 type intramolecular human telomeric G-quadruplex, Biopolymers, 101(3), 216-227.

90. Stigter, D. (1978). A Comparison of Manning‟s Polyelectrolyte Theory with the Cylindrical Gouy Model. The Journal of Physical Chemistry, 82 (14), 1603-1606.

91. Stigter, D. (1995). Evaluation of the Counterion Condensation Theory of Polyelectrolytes. Biophysical Journal, 69, 380-388.

92. Bloomfield, V. A. and Rouzina, I. (1998). Use of Poisson-Boltzmann Equation to Analyze Ion Binding to DNA. Methods in Enzymology, 295, 364-378.

93. Anderson, C. F. and Record, M. T. (1990). Ion distributions around DNA and other cylindrical polyions – theoretical descriptions and physical implications. Annu. Rev. Biophys. Biophys. Chem., 19, 423-465.

94. Manning, G. S. (1969a). Limiting laws and counterion condensation in polyelectrolyte solutions. I. Colligative properties. J. Chem. Phys., 51, 924-933.

95. Manning, G. S. (1969b). Limiting laws and counterion condensation in polyelectrolyte solutions. III. An analysis based on the Mayer ionic solution theory. J. Chem. Phys., 51, 3249-3252.

96. Manning, G. S. (1977). Limiting laws and counterion condensation in polyelectrolyte solutions. IV. The approach to the limit and the extraordinary stability of the charge fraction. Biophys. Chem., 7, 95-102.

97. Smirnov, I.V. and Shafer, R.H. (2006). Electrostatics Dominate Quadruplex Stability. Biopolymers, 85, 1.

Chapter 1: General Introduction 38

98. Smirnov, I. V., Kotch, F. W., (2002). Pickering, I. J., Davis, J. T., and Shafer, R. H. Biochemistry, 41, 12133–12139.

99. Phillips, K. Dauter, Z. Murchie, A. I. H. Lilley, D. M. J. and Luisi, B. (1997). J. Mol. Biol., 273, 171–182.

100. Arnott, S., Chandrasekaran, R., and Marttila, C.M. (1974). Biochem., 141(2), 537.

101. Xu, Q., Deng, H., and Braunlin, W.H. (1993). Biochemistry, 32, 13130-13137

102. Guschlbauer, W., Chantot, J. F., and Thiele, D. (1990). Four-Stranded nucleic acid structures 25 years later: from guanosine gels to telomer DNA. J. Biomol. Struct. Dyn., 8, 491 – 511.

103. Gilbert, D. E. and Feigon, J. (1999). Multistranded DNA structures. Curr. Opinion Struct. Biol., 9, 305 – 314.

104. Špacková, N. A., Berger, I. and Šponer, J. (1999). Nanosecond molecular dynamics simulations of parallel and antiparallel guanine quadruplex DNA molecules. J. Am. Chem. Soc., 121, 5519–5534.

105. Chowdhurdy, S. and Bansal, M. (2001). G-Quadruplex structure can be stable with only some coordination sites being occupied by cations: a six nanosecond molecular dynamics simulation. J. Phys. Chem. B, 31, 7572–7578.

106. Mergny, J. L. Cian, A. D., Ghelab, A., Sacca, B., and Lacroix, L. (2005). Kinetics of tetramolecular quadruplexes. Nucleic Acids Res., 33(1), 81-94.

107. Bhattacharyya, D., Arachchilage, M., and Basu, S. (2016). Metal Cations in G-quadruplex Folding and Stability. Front Chem., 4 (38).

108. Hud, N. V., Schultze, P., Sklenar, V., and Feigon, J. (1999). Binding Sites and Dynamics of Ammonium Ions in a Telomere Repeat DNA Quadruplex. J. Mol. Biol., 285, 233–243.

Chapter 1: General Introduction 39

109. Laughlan, G., Murchie, A. I., Norman, D. G., Moore, M. H., Moody, P. C., Lilley, D. M., and Luisi, B. (1994). The high-resolution crystal structure of a parallel-stranded guanine tetraplex. Science, 265, 520–524.

110. Hardin, C. C., Watson, T., Corregan, M., and Bailey, C. (1992). Cation- dependent

transition between the quadruplex and Watson-Crick hairpin forms of d(CGCG3GCG). Biochemistry, 31,833–841.

111. Venczel, E. A. and Sen, D. (1993). Parallel and antiparallel G-DNA structures from a complex telomeric sequence. Biochemistry, 32, 6220–6228.

112. Kankia B. I. and Marky L. A. (2001). Folding of the thrombin aptamer into a G-quadruplex with Sr2+: stability, heat, and hydration. J. Am. Chem. Soc., 123, 10799–10804.

113. Blume S. W., Guarcello V., Zacharias W., and Miller D. M. (1997). Divalent transition metal cations counteract potassium-induced quadruplex assembly of oligo (dG) sequences. Nucleic Acids Res., 25, 617–625.

114. Hardin, C. C., Perry, A. G., and White, K. (2001). Thermodynamic and kinetic characterization of the dissociation and assembly of quadruplex nucleic acids. Biopolymers, 56, 147–194.

115. Simonsson, T. (2001). G-quadruplex DNA structures–variations on a theme. Biol. Chem. 382, 621–628.

116. Krishnan-Ghosh, Y., Liu, D., and Balasubramanian, S. (2004). Formation of an Interlocked Quadruplex Dimer by d(GGGT). J. Am. Chem. Soc., 126(35), 11009.

117. Hazel, P., Huppert, J., Balasubramanian, S., and Neidle, S. (2004). Loop-length-dependent folding of G-quadruplexes. J. Amer. Chem. Soc., 1286, 16405–16415.

118. Guedin, A., Gros, J., Alberti, P., and Mergny, J.L. (2010). How long is too long? Effects of loop size on G-quadruplex stability. Nucleic Acids Research, 38 (21), 7858-7868.

Chapter 1: General Introduction 40

119. Lu, M., Guo, Q., and Kallenbach, N. R. (1993). Thermodynamics of G-tetraplex formation by telomeric DNAs. Biochemistry, 32, 598–601.

120. Burge, S. Parkinson, G. N., Hazel, P., Todd, A. K., and Neidle, S. (2006). Quadruplex DNA: sequence, topology, and structure. Nucleic Acids Research, 34(19), 5402–5415.

121. Olsen, C. M., Gmeiner, W. H., and Marky, L. A. (2006). Unfolding of G-quadruplexes: energetic, and ion and water contributions of G-tetrad stacking. J. Phys. Chem. B, 110, 6962– 6969.

122. Ying, L., Green, J. J., Li, H., Klenerman, D., and Balasubramanian, S. (2003). Studies on the structure and dynamics of the human telomeric G quadruplex by single-molecule fluorescence resonance energy transfer. Proc. Natl Acad. Sci. USA, 100, 14629–14634.

123. Sen, D. and Gilbert, W. (1988). Formation of parallel four-stranded complexes by guanine- rich motifs in DNA and its implications for meiosis. Nature, 334 (6180), 364-346.

124. Hazel, P., Huppert, J., Balasubramanian, S., and Neidle, S. (2004). Loop-length-dependent folding of G-quadruplexes. J. Amer. Chem. Soc., 1286, 16405–16415.

125. Le, H. T., Miller, M. C., Buscaglia, R., Dean, W. L., Holt, P. A., Chaires, J. B., and Trent, J. O. (2012). Not all G-quadruplexes are created equally: an investigation of the structural polymorphism of the c-Myc G-quadruplex-forming sequence and its interaction with the porphyrin TMPyP4. Org. Biomol. Chem., 10, 9393–9404.

126. Xue, Y., Kan, Z. Y., Wang, Q., Yao, Y., Liu, J., Hao, Y.H., and Tan, Z. (2007). Human Telomeric DNA Forms Parallel-Stranded Intramolecular G-Quadruplex in K+ under Molecular Crowding Condition. J. Am. Chem. Soc., 129, 11185–11191.

127. Miyoshi, D., Nakao, A., Toda, T., and Sugimoto, N. (2001). FEBS Lett., 496, 128.

128. Chaires, J. B. (2010). Human telomeric G-quadruplex: thermodynamic and kinetic studies of telomeric quadruplex stability. FEBS J., 277, 1098–1106.

Chapter 1: General Introduction 41

129. Boncina, M., Lah, J., Prislan, I., and Vesnaver, G. (2012). Energetic basis of human telomeric DNA folding into G-quadruplex structures. J. Am. Chem. Soc., 134, 9657–9663.

130. Kumar, N., Sahoo, B., Varun, K., Maiti, S., and Maiti, S. (2008). Effect of loop length variation on quadruplex-Watson Crick duplex competition. Nucleic Acids Res., 36, 4433– 4442.

131. Pilch, D. S., Plum, G. E., and Breslauer, K. J. (1995). The thermodynamics of DNA structures that contain lesions of guanine tetrads. Curr. Opin. Struct. Biol., 5, 334-342.

132. Lane, A. N., Chaires, J. B., Gray, R. D., and Trent, J. O. (2008). Stability and kinetics of G- quadruplex structures. Nucleic Acids Res., 36, 5482–5515.

133. Kogut, M., Kleist, C., and Czub, J. (2016). Molecular dynamics simulations reveal the balance of forces governing the formation of a guanine tetrad––a common structural unit of G-quadruplex DNA. Nucleic Acids Research, 44(7), 3020–3030.

134. Sengar, A., Heddi, B., and Phan, A. T. (2014). Formation of G-quadruplexes in poly-G sequences: structure of a propeller-type parallel-stranded G-quadruplex formed by a G15 stretch. Biochemistry, 53, 7718–7723.

135. Petraccone, L., Erra, E., Duro, I., Esposito, V., Randazzo, A., Mayol, L., Matia, C.A., Barone, G. and Giancola, C. (2005). Nucleos. Nucleot. Nucleic Acids, 24, 757–760.

136. Kashid, S. M., Jin, G. Y., Bagchi, S., and Kim, Y. S. (2015). Cosolvent Effects on Solute- Solvent Hydrogen-Bond Dynamics: Ultrafast 2D IR Investigations. Phys. Chem. B, 119 (49), 15334-15343.

137. Minton, A. P. (2001). The Influence of Macromolecular Crowding and Macromolecular Confinement on Biochemical Reactions in Physiological Media. J. Biol. Chem. 276, 10577- 10580.

138. Ellis, R. J. (2001). Macromolecular crowding: an important but neglected aspect of the intracellular environment. Curr. Opin. Struct. Biol., 11 (1), 114-119.

Chapter 1: General Introduction 42

139. Nakano, S. Karimata, H., Ohmichi, T., Kawakami, J., and Sugimoto, N (2004).The effect of molecular crowding with nucleotide length and cosolute structure on DNA duplex stability. Journal of the American Chemical Society, 126, 44, 14330–14331.

140. Miyoshi, D. Nakamura, K. Tateishi-Karimata, H. Ohmichi, T., and Sugimoto, N. (2009). Hydration of watson crick base pairs and dehydration of hoogsteen base pairs inducing structural polymorphism under molecular crowding conditions, Journal of the American Chemical Society, 131(10), 3522–3531.

141. Miyoshi, D., Nakao, A., and Sugimoto, N. (2002). Molecular Crowding Regulates the Structural Switch of the DNA G-Quadruplex. Biochemistry, 41, 15017-15024.

142. Li, J., Correia, J. J., Wang, L., Trent, J. O., and Chaires, J. B. (2005). Not so crystal clear: the structure of the human telomere G-quadruplex in solution differs from that present in a crystal. Nucleic Acids Res., 33, 4649-4659.

143. Miyoshi, D., Karimata, H., and Sugimoto, N. (2006). Hydration regulates thermodynamics of G-quadruplex formation under molecular crowding conditions, Journal of the American Chemical Society, 128(24), 7957–7963.

144. Ahmed, S. and Henderson, E. (1992). Formation of novel hairpin structures by telomeric C- strand oligonucleotides. Nucleic Acids Res., 20, 507-511.

145. Gehring, K., Leroy, J. L., and Gueron, M. (1993). A tetrameric DNA structure with protonated cytosine.cytosine base pairs. Nature, 363, 561–565.

146. Escaja, N., Viladoms, J., Garavis, M., Villasante, A., Pedroso, E., and Gonzalez, C. (2012). A minimal i-motif stabilized by minor groove G:T:G:T tetrads. Nucleic Acids Res., 40, 11737‐11747.

147. Zager, R. A., Schimpf, B. A., and Gmur, D. J. (1993). Physiological pH. Effects on posthypoxic proximal tubular injury.Circ. Res., 72, 837–846.

Chapter 1: General Introduction 43

148. Sun, D. and Hurley, L.H. (2009). The Importance of Negative Superhelicity in Inducing the Formation of G-Quadruplex and i-Motif Structures in the c-Myc Promoter: Implications for Drug Targeting and Control of Gene Expression. J. Med. Chem., 52, 2863–2874.

149. Sekibo, D.A.T. and Fox, K. R. (2017). The effects of DNA supercoiling on G-quadruplex formation. Nucleic Acids Research, 45 (21), 12069–12079.

150. Henderson, E., Hardin, C. C., Walk, S. K., Tinoco, I., and Blackburn, E. H. (1987). Telomeric DNA oligonucleotides form novel intramolecular structures containing guanine- guanine base pairs. Cell, 51, 899-908.

151. Williamson, J.R., Raghuraman, M.K., and Cech, T.R. (1989). Monovalent cation-induced structure of telomeric DNA: the G-quartet model. Cell, 59, 871-880.

152. Balagurumoorthy, P., Brahmanchari, S.K., Mohanty, D., Bansal, M., and Sasisekharan, V. (1992). Hairpin and parallel quartet structures for telomeric sequences. Nucleic Acids Res., 20, 4061-4067.

153. Hardin, C.C., Henderson, E., Watson, T., and Prosser, J.K. (1991). Monovalent cation- induced structural transitions in telomeric DNAs: G-DNA folding intermediates. Biochemistry, 30, 4460-4472.

154. Privalov, P.L. (2012). Microcalorimetry of Macromolecules. John Wiley and Sons, Inc.

155. Lee, J. Y., Okumus, B., Kim, D. S., and Ha, T. (2005). Extreme conformational diversity in human telomeric DNA. Proc. Natl. Acad. Sci. USA, 102, 18938–18943.

156. Hanahan, D. and Weinberg, R. A. (2011). Hallmarks of Cancer: The Next Generation. Cell. 144, 646-674.

157. Chalikian, T. V.; Sarvazyan, A. P.; Funck, Th.; Breslauer, K. J. Biopolymers 1994, 34, 541- 553.

158. Egli, M., Tereshko, V., Teplova, M., Minasov, G., Joachimiak, A., Sanishvili, R., Weeks, C. M., Miller, R., Maier, M. A., An, H., Dan, C. P., and Manoharan, M. (1998). X-ray

Chapter 1: General Introduction 44

crystallographic analysis of the hydration of A- and B-form DNA at atomic resolution. Biopolymers, 48:234–52.

159. Chalikian, T. V. and Macgregor R B. (2007) Nucleic acid hydration: a volumetric perspective. Physics of Life Reviews, 4:91–115.

160. A. T. Phan and J. L. Leroy, J. Biomol. Struct. Dyn., 2000, 17, 245–251.

161. Huppert, J. L. Four-stranded nucleic acids: structure, function and targeting of G- quadruplexes. (2008) Chemical Society Reviews, 37, 1375-1384.

162. Smyth, M. S. and Martin, J. H. J. (2000). X Ray Crystallography. J. Clin. Pathol: Mol. Pathol. 53: 8-14.

163. Pagano, B., Randazzo, A., Fotticchiaa, I., Novellino, E., Petraccone, L., and Giancola, C. (2013). Differential scanning calorimetry to investigate G-quadruplexes structural stability. Methods, 64: 43-51.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 45

CHAPTER 2: Polyelectrolyte Effects in G-quadruplexes

“Reprinted from Biophysical Chemistry, 184, Byul G. Kim, Yuen Lai Shek, and Tigran V. Chalikian. Polyelectrolyte Effects in G-quadruplexes, 95-100. Copyright (2013), with permission from Elsevier B.V.”.

Author„s Contribution: Byul G. Kim performed all experiments, analyzed the data, and participated in the preparation of the manuscript.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 46

2.1 Abstract

The role of counterion condensation as a dominant force governing the stability of DNA duplexes and triplexes is well established. In contrast, the effect of counterion condensation on the stability of G-quadrupex conformations is poorly understood. Unlike other ordered nucleic acid structures, G-quadruplexes exhibit a specific binding of counterions (typically, Na+ or K+) which are buried inside the central cavity and coordinated to the O6 carbonyls of the guanines forming the G-quartets. While it has been known that the G-quadruplex-to-coil transition temperature, TM, increases with an increase in the concentration of the stabilizing ion, the contributions of the specific (coordination in the central cavity) and nonspecific (condensation) ion binding have not been resolved. In this work, we separate the two contributions by studying + the change in TM of preformed G-quadruplexes following the addition of nonstabilizing ions Li , Cs+, and TMA+ (tetramethylammonium). In our studies, we used two G-quadruplexes formed by the human telomeric sequences which are distinct with respect to the folding topology and the identity and the number of sequestered stabilizing ions. Our data suggest that the predominant ionic contribution to G-quadruplex stability comes from the specifically bound Na+ or K+ ions and not from counterion condensation. We offer molecular rationalizations to the observed insensitivity of G-quadruplex stability to counterion condensation and emphasize the need to expand such studies to assess the generality of our findings.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 47

2.2 Introduction

Although the genomic DNA overwhelmingly exists in the B-double helical conformation, many genomic sequences may give rise to stable non B-DNA conformations. For example, guanine-rich single-stranded stretches are susceptible to spontaneous dissociation from their complementary strands and folding into various G-quadruplex structures in which guanine bases associate with each other in G-quartets, stable hydrogen-bonded arrangements1-3. The biomedical relevance of G-quadruplexes stems from the fact that the consensus G-quadruplex sequence motifs are found in critical loci of the genome, including telomeres, centromeres, immunoglobulin switch regions, mutation-prone hot spots, and promoter regions of many oncogenes2,4-8. Telomeric sequences have become an important target for anti-cancer drug development, since drug-induced G-quadruplex structures are incompatible with telomerase function and induce rapid senescence in cancer cells9. Search for G-quadruplex-stabilizing molecules have quickly emerged as part of novel anti-cancer strategies6, 8. The ultimate success of such strategies depends, in part, on our ability to understand thermodynamic forces stabilizing G-quadruplex conformations relative to their respective double- and single-stranded conformations and modulation of these forces by environmental conditions and the presence of G-quadruplex-stabilizing agents (drugs).

Owing to their polyelectrolyte nature, DNA and RNA are surrounded by a cloud of mobile counterions which exert a powerful influence on the stability of nucleic acid structures10-15. Counterion condensation in the vicinity of DNA and the concomitant entropic penalty introduce a thermodynamic leverage that can produce shift in an equilibrium of virtually any helix-to-coil and ligand binding reaction simply by a change in salt15. Helical forms of DNA, due to their 2 higher axial charge density, ξ = e /(ɛkBTb) (where e is electron charge, ɛ is the dielectric constant of the solvent, kB is the Boltzmann constant, T is the thermodynamic temperature, and b is the average axial distance between the polyelectrolyte charges), condense more counterions per phosphate relative to the single-stranded conformation10,14-17. Consequently, helix-to-coil transitions of nucleic acids are accompanied by a release of counterions to the bulk, while an increase in salt results in an increase in the thermal and thermodynamic stability of the helical forms due to a diminution of the mixing entropy penalty10, 13-16. At low ionic strength, the

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 48

number of associated counterions per phosphate is given by θ = 1 - ξ-1. For example, the values of θ for a double-stranded B-DNA and a single-stranded DNA are 0.76 and 0.44, respectively10. Thus, the helix-to-coil transition of a B-DNA is accompanied by a release of 0.32 monovalent cations to the bulk.

The conventional way of experimental determination of the number of cations, Δn = θh - θc, released to the bulk upon a helix-to-coil transition of a DNA relies on measuring the salt dependence of the transition temperature, TM:

2 + ΔnM+ = (ΔHM/RTM )(δTM/Δln[M ]) (1)

where [M+] is the molar concentration of the counterion11,16,18,19.

In contrast to all other ordered nucleic acid structures, G-quadruplexes, in addition to nonspecific binding (condensation), exhibit a specific binding of counterions (typically, Na+ or K+) which are buried inside the central cavity and coordinated to the O6 carbonyls of the guanines forming the G-quartets3,5,6. Hence, application of Eq. (1) to G-quadruplexes melting transitions, yields the number of cations released to the bulk from both the central cavity (bound specifically) and the cloud of counterions (bound nonspecifically). The problem of discrimination between the two cation populations remains unsolved which prevents one from elucidating the role of the polyelectrolyte effect in stabilization of G-quadruplex structures3. In one study, it has been concluded that the specific cation binding plays a more prominent role in stabilizing a G- quadruplex than counterion condensation20. This conclusion is in agreement with our results presented below. However, it has been reached based on comparative G-quadruplex stability studies in binary solvents containing water and a variety of cosolvents. In particular, it has been assumed that the influence of a particular cosolvent on G-quadruplex stability is predominantly determined by the dielectric constant of the medium. However, such an oversimplification may not be warranted given the complexity of the effect of cosolvents on the stability of biopolymers21-23.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 49

In this work, we evaluate the influence of the polyelectrolyte effect on G-quadruplex stability by separating the effect of the stabilizing cations in the central cavity from the effect of the cations condensed in the vicinity of the DNA. More specifically, after saturating a G-quadruplex by the stabilizing cation (Na+ or K+), a further increase in the solution ionic strength is accomplished by the addition of either the stabilizing or a non-stabilizing cation [Li+, Cs+, TMA+ (tetramethylammonium)]. The latter cannot penetrate the central cavity being either too large or too small and, therefore, influence the G-quadruplex stability only via the effect of counterion condensation. We study the differential effect of salt on the G-quadruplex-to-single strand transition temperature, TM, when the ionic strength is modulated by the stabilizing or nonstabilizing cations. To increase the generality of our work and the ensuing conclusions, we have chosen two G-quadruplexes varying in topology and the identity and number of the sequestered stabilizing cation (Na+ and K+). Our results suggest that, in contrast to the duplex and T-rich triplex conformations, the polyelectrolyte effect plays a small to negligible role in maintaining G-quadruplex stability. The observed salt-dependent increase in G-quadruplex stability predominantly reflects the specific binding of the Na+ or K+ cations.

2.3 Materials and Methods

Materials

The oligodeoxyribonucleotides, d[A(G3T2A)3G3] (Tel22) and d[A3(G3T2A)3G3A2] (Tel26), containing four repeats of the human telomeric DNA sequence were synthesized and cartridge purified by ACGT, Inc. (Toronto, ON, Canada). Potassium chloride, sodium chloride, cesium chloride, lithium chloride, tetramethylammonium chloride, and phosphoric acid were purchased from Sigma-Aldrich Canada (Oakville, ON, Canada). EDTA (free acid) was purchased from Fisher Biotech (Fair Lawn, NJ, USA). These reagents were of the highest grade commercially available and used without further purification. All solutions were prepared using doubly distilled water.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 50

Prior to all experiments, the Tel22 oligonucleotide was dissolved in and exhaustively dialyzed against a pH 7.0 buffer consisting of 10 mM sodium phosphate, 0.1 mM EDTA, and 20 mM NaCl. The Tel26 oligonucleotide was dissolved in and dialyzed against a pH 7.0 buffer consisting of 10 mM potassium phosphate, 0.1 mM EDTA, and 20 mM KCl. In these conditions, both Tel22 and Tel26 are fully folded in their respective G-quadruplex conformations [24;25]. Dialysis was carried out in 1000 Da molecular weight cut-off Tube-O-Dialyzers from G Biosciences (St. Louis, MO, USA). For UV melting and CD experiments, the ionic strengths of the DNA solutions were adjusted to the desired level by addition of aliquots of NaCl, KCl, LiCl, CsCl, or TMACl on top of the existing 20 mM of NaCl (Tel22) or KCl (Tel26).

The CD and UV melting experiments were performed at the following ionic conditions. For the experiments involving Tel22, the ionic conditions were 20, 50, 100, 200, 400, and 800 mM NaCl or 20 mM NaCl plus 30, 80, 180, 380, and 780 mM LiCl, CsCl, or TMACl. For the experiments involving Tel26, the measurements were performed at 20, 50, 100, 200, 400, and 800 mM KCl or 20 mM KCl plus 30, 80, 180, 380, and 780 mM LiCl, CsCl, or TMACl.

The concentrations of the oligonucleotides were determined from the UV light absorbance at 260 nm measured at 25 °C with a Cary 300 Bio spectrophotometer (Varian Canada, Inc., Mississauga, Ontario, Canada) using molar extinction coefficients of 228,500 and 278,200 M-1 cm-1 for the unfolded conformation of Tel22 and tel26, respectively. These extinction coefficients were calculated using an additive nearest neighbor procedure as described by Dr. Richard Owczarzy (http://www.owczarzy.net/extinct.htm). For our CD measurements and temperature-dependent UV light absorption measurements, the DNA concentrations were ~30 and ~3 μM, respectively.

Circular Dichroism Spectroscopy

The CD spectra of the Tel22 and Tel26 were recorded in a 1 mm path-length cuvette at 25 °C using an Aviv model 62 DS spectropolarimeter (Aviv Associates, Lakewood, NJ). CD spectroscopy was used to probe the conformations adopted by the oligonucleotides at the various ionic conditions employed in the present study.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 51

UV Melting Experiments

UV light absorption at 295 nm was measured as a function of temperature in DNA samples contained in a 1 cm path-length cuvette. These measurements were performed by a Cary 300 Bio spectrophotometer (Varian Canada, Inc., Mississauga, Ontario, Canada). The temperature was changed at a rate of 1 °C per minute. The G-quadruplex-to-coil transition temperatures, TM, and van't Hoff enthalpies, ΔHvH, were evaluated from our experimental UV melting profiles using standard procedures26,27.

It should be noted that temperature scanning experiments performed on Tel22 in NaCl solutions at a rate of 0.2 °C per minute did not reveal any change in the shape of the melting profiles. By extension, we assume that the scanning rate of 1 °C per minute is optimal for all DNA melting experiments reported here.

2.4 Results

CD spectra. In the presence of Na+ ions, the Tel22 human telomeric sequence, + d[A(G3T2A)3G3], forms an antiparallel structure with three Na ions coordinated within the central cavity28. To monitor the G-quadruplex formation in the presence of NaCl and its maintenance throughout our experimental protocols (in the presence of LiCl, CsCl, and TMACl), we measured the CD spectrum of the DNA at each salt. Figure 2-1 presents the CD spectra of Tel22 in the presence of 20 mM NaCl and various concentrations of LiCl (panel A), CsCl (panel B), and TMACl (panel C). Previous investigations have shown that the CD spectrum of antiparallel telomeric G-quadruplexes display a negative peak at 260 nm and a positive peak at 295 nm3. The CD spectra of Tel22 in Figure 1a, b, and c with the characteristic bands at 260 and 295 nm are consistent with the formation of an antiparallel G-quadruplex structure.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 52

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 53

Figure 2-1. CD spectra of Tel22 in the presence of 20 mM NaCl and 0 (●), 180 (○), 380 (■), and 780 (□) mM LiCl (panel A), CsCl (panel B), and TMACl (panel C).

+ In the presence of K ions, the Tel26 sequence d[A3G3(T2AG3)3A2] overwhelmingly exists as a hybrid-type (hybrid-1) intramolecular G-quadruplex consisting of three G-tetrads linked with mixed parallel/antiparallel G-strands29-31. The structure has a sequential double-chain-reversal side loop and two lateral loops each consisting of three nucleotides (-TTA-) thereby making the third G-strand antiparallel. In contrast to Tel22 with three sequestered Na+ ions, the Tel26 G- quadruplex coordinates only two K+ ions within its central cavity. The CD signature of the hybrid-1 conformation of Tel26 is a double-headed positive domain with the maxima at 260 and 295 nm and a negative domain with the minimum at 240 nm29. The CD spectra of Tel26 in the presence of 20 mM KCl and various concentrations of LiCl (Figure 2-2a), CsCl (Figure 2-2b), and TMACl (Figure 2-2c) display the characteristic bands at 240, 260, and 295 nm. Thus, the CD spectra of Tel26 are consistent with the CD spectrum reported for the hybrid-1 conformation of Tel2625,29. The CD spectra are representative of more than three repeated trials of equivalent experiments.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 54

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 55

Figure 2-2. CD spectra of Tel26 in the presence of 20 mM KCl and 0 (●), 180 (○), 380 (■), and 780 (□) mM LiCl (panel A), CsCl (panel B), and TMACl (panel C).

UV Melting Profiles. Figure 2-3 shows representative UV melting profiles measured at 295 nm for Tel22 in 100 mM NaCl and Tel26 in 100 mM KCl. These and similar profiles determined at other salt conditions all exhibit a sigmoidal shape. The UV melting profiles were approximated by the analytical relationship for the two-state transition:

-1 -1 -1 α = [1 - exp(ΔHvH(T - TM )/R)] (2)

where ΔHvH is the van't Hoff enthalpy of the transition, TM is the transition temperature, α =

[A(T) - AN(T)] / [AD(T) - AN(T)] is the fraction of the DNA that is unfolded; A(T) is the UV light absorption at a temperature T; and AN(T) and AD(T) signify the native and denatured baselines, respectively.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 56

Figure 2-3. UV melting profiles at 295 nm of Tel22 at 100 mM NaCl (●) and Tel26 at 100 mM KCl (○).

Our measured UV melting profiles were used in conjunction with Eq. (2) to determine the transition temperatures, TM, and enthalpies, ΔHvH, for Tel22 and Tel26 at each experimental salt condition. Figure 4 plots the melting temperatures, TM, against the natural logarithm of cation concentration for Tel22 (panel A) and Tel26 (panel B).

It needs to be pointed out that G-quadruplexes, in general, may not melt in the two-state 32, 33 manner with no single temperature to describe the melting transition . Therefore, the TM, as determined from Eq. (2), should be qualified as an effective or apparent melting temperature. Nevertheless, it represents a reliable measure of the thermal stability of G-quadruplexes in comparative studies reported here.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 57

2.5 Discussion

Counterions can influence the stability of a G-quadruplex in two ways by binding specifically or binding nonspecifically by condensing around the DNA polyanion. While specific binding may cause conformational transitions of G-quadruplex-forming sequences, no conformational changes are expected to accompany nonspecific ion binding. Any release of condensed counterions upon a helix-to-coil transition reflects a difference in charge density between the folded and unfolded forms. In the present study, we are interested in this type of binding (counterion condensation).

The Na+ and K+ ions bind specifically by penetrating the central cavity and stabilizing the G- quadruplex conformation3. In contrast, the Li+, Cs+, and TMA+ ions do not fit inside the cavity being either too small or too large. To additionally verify that these ions do not support G- quadruplex formation, we have recorded the CD spectra of Tel22 and Tel26 in the absence of the Na+ and K+ ions but in the presence of Li+, Cs+, and TMA+ up to 800 mM (data not shown). No G-quadruplex formation was detected.

Inspection of Figures 1 and 2 reveals that, in the presence of Li+, Cs+, and TMA+, the CD spectra of the preformed Tel22 and Tel26 G-quadruplexes undergo only quantitative, although, sizeable changes. The observed changes do not involve appearance of new CD bands or disappearance of the existing ones. Instead, the Li+, Cs+, and TMA+ ions only bring about a diminution of the intensity of the existing CD bands. This observation is consistent with the picture in which the two G-quadruplexes retain their respective global conformations with no new species (including unfolded species) appearing in the presence of the three nonstabilizing cations. Thus, we conclude that the Li+, Cs+, and TMA+ ions affect the thermal stability of the Tel22 and Tel26 G-quadruplexes predominantly via the effect of counterion condensation.

A note of caution is in order here. TMA+ ions interact in a differential manner with the AT and GC base pairs in DNA duplexes34-38. Hence, the effect of TMA+ on G-quadruplex stability may not be limited to counterion condensation. While not ignoring this possibility, we, nevertheless, do not separate the effect TMA+ ions from that of the Li+ and Cs+ ions because of the qualitative similarity of the influence of the three ions on the spectral and thermal properties

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 58

of the G-quadruplexes. In fact, the CD spectral changes of Tel22 and Tel26 caused by TMA+ are less pronounced compared to those caused by Li+ and Cs+ (see Figures 1 and 2).

Inspection of Figures 4a and b reveals that steep increases in TM for both the Tel22 and Tel26 oligonucleotides accompany increases in the concentration of NaCl and KCl, respectively. In other words, the thermal stability of the two G-quadruplexes increases with the rise in the concentration of their respective stabilizing ions (Na+ for Tel22 and K+ for Tel26). We use the + slope (ΔTM/Δln[M ]) in conjunction with Eq. (1) to calculate the number of sodium or potassium 11,18 ions, ΔnM+, released to the bulk upon G-quadruplex unfolding . The numbers of counterions released to the bulk are 1.7±0.5 Na+ for Tel22 and 1.5±0.2 K+ for Tel 26. These numbers are qualitatively similar to 3 and 2, the structurally detected numbers of Na+ and K+ cations coordinated within the central cavities of the Tel22 and Tel26 G-quadruplexes, respectively28,29,39.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 59

Figure 2-4. (a) Dependences of the melting temperature, TM, of Tel22 on the concentration of Na+ (●), Li+ (○), Cs+ (■), and TMA+ (□) cations; (b) Dependences of the melting temperature, + + + + TM, of Tel26 on the concentration of K (●), Li (○), Cs (■), and TMA (□) cations.

Further inspection of Figures 4a and b reveals that an increase in the concentration of the nonstabilizing ions Li+, Cs+, and TMA+ brings about only modest (Tel26) or no (Tel22) changes in the melting temperature, TM. Specifically, for Tel22, an increase in the concentration of these ions from 0 to 800 mM does not result in any increase in TM. For Tel26, an increase in the + + + concentration of the Li , Cs , and TMA ions causes modest, although non-zero, changes in TM. The strongest change is observed for Li+ ions for which an increase in concentration from 0 to

800 mM causes TM to increase from 46 to 58 °C. This is only a fraction of the increase in TM from 46 to 77 °C that was detected to accompany an increase in K+ concentration from 20 to 800 mM.

In the absence of fortuitous compensations, the observed insensitivity of the thermal stability of Tel22 with respect to the presence of the nonstabilizing ions Li+, Cs+, and TMA+ is consistent with the picture in which the polyelectrolyte effects exert a negligible influence on the stability

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 60

of the Tel22 G-quadruplex. By extension, this inference implies that the Tel22 oligomeric sequence condenses the same amount of counterions in its folded and unfolded states. It also suggests that the observed increase in the stability of the Tel22 G-quadruplexes upon the addition of NaCl results overwhelmingly from the specific binding of Na+ ions in the central cavity with the former acting as a ligand. From this perspective, the observed increase in G-quadruplex stability due to Na+ ions is similar to an enhancement of protein or DNA stability facilitated by ligand binding.

In contrast to Tel22, the thermal stability of the Tel26 G-quadruplex moderately increases with an increase in the concentration of the nonstabilizing ions Li+, Cs+, and TMA+. As is seen + from Figure 4b, the increase in TM is only 15 to 30 % of that caused by the stabilizing ion K . Thus, although, the effect of counterion condensation in Tel26 is not zero, the predominant ionic contribution to its stability still comes from the specific K+ binding and not from counterion condensation.

The greatly reduced effect of counterion condensation in G-quadruplex stability is related, at least, in part to the presence of coordinated cations within the central cavity with the concomitant decrease in the charge density of the DNA. This notion is supported by the fact that Tel22 with + its three sequestered Na ions exhibits no change in TM with an increase in the concentration of nonstabilizing cations. On the other hand, Tel26 with only two sequestered K+ ions and, hence, a higher charge density, exhibits a non-zero slope of the dependence of TM on the concentration of nonstabilizing cations. This interpretation is also in line with the lack of counterion condensation contribution to the stability of C-rich triplexes which, due to the presence of protonated cytosines in the third strand, exhibit a reduced charge density40, 41. Another contributing factor may be the globular shape of a G-quadruplex that increases the distance between the G-strands thereby diminishing the net charge density to the level characteristic of the single-stranded conformation. This conjecture is supported by the results of recent theoretical studies by Manning that revealed reduced counterion condensation around a charged spherical construct relative to a cylinder12, 13.

The results presented in this paper suggest that, for the two G-quadruplexes differing in topology and the type of the stabilizing cations, the polyelectrolyte effects play a small to

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 61

negligible role in maintaining their stability. This is in stark contrast to the thermodynamics of duplex and T-rich triplex DNA where counterion condensation provides a major contribution to thermal and thermodynamic stability10, 11, 14, 15, 18, 41. Our data suggest that the duplex-G- quadruplex equilibrium in a system with a G-rich strand and its complementary C-rich strand will shift towards the duplex state upon an increase in the concentration of nonstabilizing counterions. This notion may be of practical importance, for example, for understanding the balance of forces governing the conformational preferences of G-rich strands in the promoter regions of oncogenes where G-quadruplex formation is preceded by duplex dissociation.

The generality of our finding is supported by the ionic, structural, and topological dissimilarities between the two G-quadruplex structures studied in this work. However, further studies involving a larger and more diversified group of G-quadruplexes are needed to ascertain the generality of the relative insignificance of counterion condensation in maintaining G- quadruplex stability. In particular, the influence of the G-quadruplex topology, molecularity, sequence, loop length, and composition must be studied.

2.6 Conclusion

G-quadruplexes are non-canonical nucleic acid constructs that exist in the genome and have been implicated in cancer research. G-quadruplexes are unusual in that, in contrast to other DNA structures, they specifically bind cations in their central cavity. There has been a long- standing question about the ionic stabilization of G-quadruplexes. It was not clear if the stabilization comes from the specific binding, nonspecifc binding (counterion condensation), or both. In this work, we resolve the specific and nonspecific contributions to the thermal stability of G-quadruplexes by separating the effect of the centrally bound ions from the effect of the cations condensed around the DNA. Our data suggest that the predominant ionic contribution to G-quadruplex stability comes from the specifically bound Na+ or K+ ions and not from counterion condensation around the DNA. Our results shed light on the balance of forces governing the conformational preferences of guanine-rich sequences while also being of practical importance for controlled induction of G-quadruplexes in the genome.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 62

2.7 Acknowledgements

This work is supported by a grant from NSERC to TVC.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 63

2.8 Reference

1. M.A.Keniry, Quadruplex structures in nucleic acids, Biopolymers 56 (2000) 123-146.

2. R.H.Shafer and I.Smirnov, Biological aspects of DNA/RNA quadruplexes, Biopolymers 56 (2000) 209-227.

3. A.N.Lane, J.B.Chaires, R.D.Gray and J.O.Trent, Stability and kinetics of G-quadruplex structures, Nucleic Acids Res. 36 (2008) 5482-5515.

4. L.H.Hurley, DNA and its associated processes as targets for cancer therapy, Nat.Rev.Cancer 2 (2002) 188-200.

5. J.L.Huppert, Four-stranded DNA: cancer, gene regulation and drug development, Philos.Trans.RoyalSoc. 365 (2007) 2969-2984.

6. J.L.Huppert, Four-stranded nucleic acids: structure, function and targeting of G-quadruplexes, Chem.Soc.Rev. 37 (2008) 1375-1384.

7. L.Oganesian and T.M.Bryan, Physiological relevance of telomeric G-quadruplex formation: a potential drug target, Bioessays 29 (2007) 155-165.

8. L.Oganesian and T.M.Bryan, Physiological relevance of telomeric G-quadruplex formation: a potential drug target, Bioessays 29 (2007) 155-165.

9. L.R.Kelland, European Overcoming the immortality of tumour cells by telomere and telomerase based cancer therapeutics - current status and future prospects, J.Cancer 41 (2005) 971-979.

10. G.S.Manning, Molecular theory of polyelectrolyte solutions with applications to electrostatic properties of polynucleotides, Q.Rev.Biophys. 11 (1978) 179-246.

11. M.T.Record, C.F.Anderson and T.M.Lohman, Thermodynamic analysis of ion effects on binding and conformational equilibria of proteins and nucleic acids. Roles of ion association or release, screening, and ion effects on water wctivity, Q.Rev.Biophys. 11 (1978) 103-178.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 64

12. G.S.Manning, Electrostatic free energies of spheres, cylinders, and planes in counterion condensation theory with some applications, Macromolecules 40 (2007) 8071-8081.

13. G.S.Manning, Counterion condensation on charged spheres, cylinders, and planes, J.Phys.Chem.B 111 (2007) 8554-8559.

14. C.F.Anderson and M.T.Record, Polyelectrolyte theories and their applications to DNA, Annu.Rev.Phys.Chem. 33 (1982) 191-222.

15. C.F.Anderson and M.T.Record, Salt-nucleic acid interactions, Annu.Rev.Phys.Chem. 46 (1995) 657-700.

16. J.P.Bond, C.F.Anderson and M.T.Record, Conformational transitions of duplex and triplex nucleic acid helices. Thermodynamic analysis of effects of salt concentration on stability using preferential interaction coefficients, Biophys.J. 67 (1994) 825-836.

17. J.Volker, H.H.Klump, G.S.Manning and K.J.Breslauer, Counterion association with native and denatured nucleic acids: an experimental approach, J.Mol.Biol. 310 (2001) 1011-1025.

18. Y.K.Cheng and B.M.Pettitt, Stabilities of double strand and triple strand helical nucleic acids, Prog.Biophys.Mol.Biol. 58 (1992) 225-257.

19. R.Owczarzy, I.Dunietz, M.A.Behlke, I.M.Klotz and J.A.Walder, Thermodynamic treatment of oligonucleotide duplex-simplex equilibria, Proc.Natl.Acad.Sci.U.S.A 100 (2003) 14840- 14845.

20. I.V.Smirnov and R.H.Shafer, Electrostatics dominate quadruplex stability, Biopolymers 85 (2007) 91-101.

21. S.N.Timasheff, Protein hydration, thermodynamic binding, and preferential hydration, Biochemistry 41 (2002) 13473-13482.

22. J.A.Schellman, Protein stability in mixed solvents: a balance of contact interaction and excluded volume, Biophys.J. 85 (2003) 108-125.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 65

23. D.B.Knowles, A.S.LaCroix, N.F.Deines, I.Shkel and M.T.Record, Jr., Separation of preferential interaction and excluded volume effects on DNA duplex and hairpin stability, Proc.Natl.Acad.Sci.U.S.A 108 (2011) 12699-12704.

24. H.Y.Fan, Y.L.Shek, A.Amiri, D.N.Dubins, H.Heerklotz, R.B.Macgregor and T.V.Chalikian, Volumetric characterization of sodium-induced G-quadruplex formation, J.Am.Chem.Soc. 133 (2011) 4518-4526.

25. Y.L.Shek, G.D.Noudeh, M.Nazari, H.Heerklotz, R.M.Abu-Ghazalah, D.N.Dubins and T.V.Chalikian, Folding thermodynamics of the hybrid-1 type intramolecular human telomeric G-quadruplex, Biopolymers (2013) in press.

26. L.A.Marky and K.J.Breslauer, Calculating thermodynamic data for transitions of any molecularity from equilibrium melting curves, Biopolymers 26 (1987) 1601-1620.

27. K.J.Breslauer, Extracting thermodynamic data from equilibrium melting curves for oligonucleotide order-disorder transitions, Methods Enzymol. 259 (1995) 221-242.

28. Y.Wang and D.J.Patel, Solution structure of the human telomeric repeat d[AG3(T2AG3)3] G- tetraplex, Structure. 1 (1993) 263-282.

29. A.Ambrus, D.Chen, J.X.Dai, T.Bialis, R.A.Jones and D.Z.Yang, Human telomeric sequence forms a hybrid-type intramolecular G-quadruplex structure with mixed parallel/antiparallel strands in potassium solution, Nucleic Acids Res. 34 (2006) 2723-2735.

30. J.X.Dai, C.Punchihewa, A.Ambrus, D.Chen, R.A.Jones and D.Z.Yang, Structure of the intramolecular human telomeric G-quadruplex in potassium solution: a novel adenine triple formation, Nucleic Acids Res. 35 (2007) 2440-2450.

31. J.X.Dai, M.Carver and D.Z.Yang, Polymorphism of human telomeric quadruplex structures, Biochimie 90 (2008) 1172-1183.

32. R.D.Gray, R.Buscaglia and J.B.Chaires, Populated intermediates in the thermal unfolding of the human telomeric quadruplex, J.Am.Chem.Soc. 134 (2012) 16834-16844.

Chapter 2: Polyelectrolyte Effects in G-quadruplexes 66

33. R.Buscaglia, R.D.Gray and J.B.Chaires, Thermodynamic characterization of human telomere quadruplex unfolding, Biopolymers 99 (2013) 1006-1018.

34. J.T.Shapiro, B.S.Stannard and G.Felsenfeld, The binding of small cations to deoxyribonucleic acid. Nucleotide specificity, Biochemistry 8 (1969) 3233-3241.

35. W.B.Melchior, Jr. and P.H.von Hippel, Alteration of the relative stability of dAdT and dGdC base pairs in DNA, Proc.Natl.Acad.Sci.U.S.A 70 (1973) 298-302.

36. J.J.Delrow, P.J.Heath, B.S.Fujimoto and J.M.Schurr, Effect of temperature on DNA secondary structure in the absence and presence of 0.5 M tetramethylammonium chloride, Biopolymers 45 (1998) 503-515.

37. W.I.Wood, J.Gitschier, L.A.Lasky and R.M.Lawn, Base composition-independent hybridization in tetramethylammonium chloride: a method for oligonucleotide screening of highly complex gene libraries, Proc.Natl.Acad.Sci.U.S.A 82 (1985) 1585-1588.

38. K.A.Jacobs, R.Rudersdorf, S.D.Neill, J.P.Dougherty, E.L.Brown and E.F.Fritsch, The thermal stability of oligonucleotide duplexes is sequence independent in tetraalkylammonium salt solutions: application to identifying recombinant DNA clones, Nucleic Acids Res. 16 (1988) 4637-4650.

39. R.D.Gray and J.B.Chaires, Kinetics and mechanism of K+- and Na+-induced folding of models of human telomeric DNA into G-quadruplex structures, Nucleic Acids Res. 36 (2008) 4191-4203.

40. J.Volker and H.H.Klump, Electrostatic effects in DNA triple helices, Biochemistry 33 (1994) 13502-13508.

41. G.E.Plum, D.S.Pilch, S.F.Singleton and K.J.Breslauer, Nucleic acid hybridization: triplex stability and energetics, Annu.Rev.Biophys.Biomol.Struct. 24 (1995) 319-350.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 67

CHAPTER 3: Effects of Salt on the Stability of a G- quadruplex from the Human c-MYC Promoter

“Reprinted from Biochemistry, 54, Byul G. Kim, Heather M. Evans, David N. Dubins, and Tigran V. Chalikian. Effects of Salt on the Stability of a G-quadruplex from the Human c-MYC Promoter, 3420−3430. Copyright (2015), with permission from American Chemical Society”.

Author„s Contribution: Byul G. Kim performed all experiments, analyzed the data, and participated in the preparation of the manuscript.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 68

3.1 Abstract

In the presence of potassium ions, a modified c-MYC NHE III1 sequence with two G-to- T mutations (MYC22-G14T/G23T) forms a highly thermostable parallel-stranded G-quadruplex.

The G-quadruplex exhibits a steady increase in its melting temperature, TM, upon an increase in the concentration of the stabilizing cation K+. On the other hand, an increase in the concentration of nonstabilizing Cs+ or TMA+ cations at a constant concentration of K+ causes a + + sharp decline in TM followed by leveling off at ~200 mM Cs or TMA . At 51 °C and 600 mM K+, an increase in Cs+ concentration from 0 to 800 mM leads to a complete unfolding of the G- quadruplex. These observations are consistent with the picture in which the unfolded state of MYC22-G14T/G23T exhibits a stronger polyelectrolyte effect than the G-quadruplex state. We estimate that the unfolded state condenses roughly one extra counterion relative to the G- quadruplex state. Taken together with results of our earlier study of human telomeric G- quadruplexes, our data suggest that the stabilizing action of cations on the G-quadruplex conformation is, primarily, due to the central ions which act as specifically bound ligands. Nonspecifically bound (condensed) counterions may slightly stabilize the G-quadruplex conformation or exert no influence on its stability (human telomeric G-quadruplexes) or may strongly destabilize it (MYC22-G14T/G23T). We offer a structural rationalization for the extraordinary thermal stability and reduced polyelectrolyte effect of the MYC22-G14T/G23T G- quadruplex.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 69

3.2 Introduction

Guanine-rich single-stranded DNA stretches exhibit a propensity to spontaneously dissociate from their complementary strands and fold into various G-quadruplex structures in which guanine bases associate into G-quartets - plain hydrogen-bonded arrangements 1-3. More generally, the GmL1GmL2GmL3Gm sequence motifs (m refers to the number of guanine repeats within the G-stretch, while L1, L2, and L3 signify a nucleotide combination forming the loops between the respective G-stretches) may, under favorable conditions, form G-quadruplexes even when the complementary DNA strand is present 3-5. There is growing evidence that DNA and RNA G-quadruplexes exist in vivo with ensuing biological implications 6-9. Consensus G-quadruplex sequence motifs are found in important loci of the genome including telomeres, centromeres, immunoglobulin switch regions, mutation-prone hot spots, and promoter regions of oncogenes (such as c-myc, c-myb, c-fos, and c-abl) 2,5,10-13.

The c-MYC oncogene encodes a transcription factor that upregulates expression of many genes thereby playing an important role in cell proliferation, growth, differentiation, angiogenesis, and apoptosis 14-16. The aberrant overexpression of c-MYC is linked to the progression of a variety of human cancers including colon, cervical, breast, prostate, and small- cell lung cancers, as well as osteosarcomas, leukemias, and lymphomas 14,16,17. Roughly 90 % of c-MYC transcriptional activation is controlled by the highly conserved nuclease-hypersensitive element III1 (NHE III1) which is located 142-115 bp upstream from the P1 promoter of the 17-20 oncogene . NHE III1 consists of the complementary coding pyrimidine-rich and non-coding purine-rich strands which, in vitro, can separately adopt the i-motif and G-quadruplex conformations, respectively 18,21. Under conditions of transcriptionally induced negative superhelicity, the NHE III1 domain exists in a slow equilibrium between the B-DNA double stranded conformation, the single stranded conformation, and the combination of the tetra- stranded i-motif (coding strand) and G-quadruplex (non-coding strand) conformations 17,22,23.

The G-quadruplex that forms within NHE III1 acts as a silencer element facilitating downregulation of c-MYC transcription 17,24.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 70

The purine-rich strand of the c-MYC NHE III1, 5'- TGGGGAGGGTGGGGAGGGTGGGGAAGG-3' (Pu27), is a 27-mer containing five consecutive runs of guanines. Both the I, II, III, and IV and II, III, IV, and V consecutive guanine runs have the ability to fold into G-quadruplexes. Nevertheless, in the single-stranded state of Pu27, the first run of guanines (G2 to G5) is not involved in the formation of the physiologically relevant G-quadruplex. The II, III, IV, and V runs of guanines of NHE III1 predominantly form the basis of its folding into the G-quadruplex-based silencer element 24,25. However, when incorporated into a supercoiled plasmid and, consequently, in the context of duplex DNA under negative superhelicity, the more stable G-quadruplex appears to be the one which involves the I, II, III, and IV runs of guanines 23.

Studies of the c-MYC NHE III1 sequence as well as its myc-1234 (5'- TGGGGAGGGTTTTTAGGGTGGGGA-3') and myc-2345 (5'- TGAGGGTGGGGAGGGTGGGGAA-3') fragments and their modifications have established that, in the presence of K+ ions, all of them adopt the parallel G-quadruplex conformation with three double-chain-reversal loops bridging G-quartet layers 25-28. The presence of the two four- guanine stretches in the myc-1234 and myc-2345 sequences gives rise to ambiguities with respect to the actual topology of the formed G-quadruplex structures involving three G-quartets. Introduction of two G-to-T mutations at residue positions of 14 and 23 of the original c-MYC (Pu 27) sequence (5'-TGAGGGTGGGTAGGGTGGGTAA-3') restricts the modified sequence (MYC22-G14T/G23T) to adopting a single, parallel-stranded G-quadruplex conformation 27. This singularity renders the MYC22-G14T/G23T sequence an attractive candidate for thermodynamic investigations affording a more straightforward interpretation of resulting data while also being of biologically relevance.

Identification and thermodynamic characterization of forces that stabilize/destabilize G- quadruplex conformations of guanine-rich strands relative to their double- and single-stranded conformers is of fundamental and practical importance 3,29. In particular, they are important for elucidating mechanisms of biological control of G-quadruplex-based genetic switches and devising ways for operational intervention in the induction/prevention of G-quadruplex formation at desired genomic loci.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 71

One ubiquitous force that modulates the conformational preferences of nucleic acids emerges from the differential counterion condensation (polyelectrolyte effect) of the interconverting (e. g., helix and coil) species 30-34. Given the entropic forces originating from the polyelectrolyte effect, an increase in the solution ionic strength shifts the equilibrium between nucleic acid species towards the conformer that condenses more counterions in its vicinity 30-34. This is the case with the helix-to-coil transitions of nucleic acid duplexes and T-rich triplexes when an increase in salt favors the helical states with a concomitant increase in the melting 35,36 temperature, TM .

The thermal stability (TM) of G-quadruplexes increases with an increase in the concentration of the stabilizing ion 3. However, the situation with G-quadruplexes is more complicated, since, unlike other nucleic acid structures, G-quadruplexes display specific ion binding inside the central cavity in addition to counterion condensation (nonspecific binding).

Consequently, the observed increase in TM may originate from both the specific and nonspecific ion binding. In a previous publication, we have developed an experimental approach that enables one to separate the specific and nonspecific contributions to the salt-induced increase in G-quadruplex stability 37.

In this approach, the thermal stability (melting temperature, TM) of a G-quadruplex preformed at saturating concentrations of the stabilizing ion is measured as a function of the concentration of stabilizing or nonstabilizing [such as Cs+ or tetramethylammonium (TMA+) ion] 37 + + monovalent cations . A change in TM upon addition of the stabilizing cation (Na or K ) is governed by a combination of specific (internal binding) and nonspecific (counterion condensation) salt effects. In contrast, a change in TM upon the addition of a nonstabilizing cation is solely determined by counterion condensation. Hence, comparison of the dependences of TM on the concentration of stabilizing and nonstabilizing cations enables one to separate the effect of specific and nonspecific binding on G-quadruplex stability.

With this approach, we have studied two G-quadruplexes formed by human telomeric sequences which are distinct with respect to their folding topology and the identity and the number of sequestered stabilizing ions (Na+-stabilized antiparallel and K+-stabilized hybrid structures) 37. Our data were consistent with the picture in which the overwhelming electrostatic

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 72

contribution to G-quadruplex stability is due to the specifically bound Na+ or K+ ions and not from the effect of counterion condensation 37. The latter effect has no or only a slightly stabilizing influence on the thermal stability of the two telomeric G-quadruplexes 37.

In the present work, we extend our studies to characterizing the K+-stabilized parallel G- quadruplex formed by the modified MYC22-G14T/G23T sequence. As mentioned above, MYC22-G14T/G23T is an attractive subject of thermodynamic investigations given its biological relevance and the singularity of its structure. Another peculiarity of the MYC22- G14T/G23T G-quadruplex is its unusually high thermal stability and the ability to form G- quadruplex at extremely low K+ concentrations 27,38. Our results indicate a strongly destabilizing influence of the polyelectrolyte effect on the G-quadruplex formed by the MYC22-G14T/G23T sequence. We also find that the MYC22-G14T/G23T G-quadruplex can be unfolded by an increase in the concentration of nonstabilizing counterions. This is a remarkable result that, to the best of our knowledge, has never been reported for nucleic acid structures.

3.3 Materials and Methods

Materials. The MYC22-G14T/G23T oligonucleotide (ODN) sequence 5'- TGAGGGTGGGTAGGGTGGGTAA-3' was purchased from two sources - ACGT, Inc. (Toronto, ON, Canada) and Integrated DNA Technologies (Coralville, IA, USA). We did not observe any statistically significant differences in our experimental results obtained on the ACGT or IDT DNA samples. Potassium chloride, cesium chloride, tetramethylammonium chloride, tetrabutylammonium hydroxide, and phosphoric acid were obtained from Sigma- Aldrich Canada (Oakville, ON, Canada). EDTA (free acid) was purchased from Fisher Biotech (Fair Lawn, NJ, USA). All reagents were of the highest commercially available grade and used without further purification. The solutions were prepared using doubly distilled water.

The ODN was initially dissolved in and dialyzed against a buffer consisting of 10 mM phosphoric acid titrated to pH 7.0 by tetrabutylammonium hydroxide, 0.1 mM EDTA, and the desired levels of KCl, CsCl, or TMACl. Dialysis was performed in 1000 Da molecular weight

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 73

cut-off Tube-O-Dialyzers from G Biosciences (St. Louis, MO, USA). In the absence of any added salt, the tetrabutylammonium phosphate buffer does not support the folded G-quadruplex conformation of the ODN. Given their large size, tetrabutylammonium (TBA+) ions are fully excluded from the central cavity of G-quadruplex and, therefore, cannot stabilize a G-quadruplex.

The concentrations of the oligonucleotide were determined from the UV light absorbance at 260 nm measured at 25 °C with a Cary 300 Bio spectrophotometer (Varian Canada, Inc., Mississauga, Ontario, Canada) using molar extinction coefficient of 228,700 M-1 cm-1 for the unfolded conformation. The latter was calculated using an additive nearest neighbor procedure as described by Dr. Richard Owczarzy (http://www.owczarzy.net/extinct.htm). For our CD measurements and temperature-dependent UV light absorption measurements, the ODN concentrations were adjusted to ~30 and ~3 μM, respectively.

Circular Dichroism Spectroscopy. The CD spectra of the ODN were recorded in a 1 mm path-length cuvette at 25 °C using an Aviv model 62 DS spectropolarimeter (Aviv Associates, Lakewood, NJ). CD spectroscopic measurements were used to probe the conformations adopted by the ODN at various ionic conditions employed in the current study.

UV Melting Experiments. UV light absorption at 295 nm was measured as a function of temperature in ODN samples contained in a 1 cm path-length cuvette. The measurements were performed with a Cary 300 Bio spectrophotometer (Varian Canada, Inc., Mississauga, Ontario, Canada). The temperature was changed at a rate of 1 °C per minute. The G-quadruplex-to-coil transition temperatures, TM, and van't Hoff enthalpies, ΔHvH, were evaluated from our experimental UV melting profiles using the standard procedures 39,40.

Computation of Solvent Accessible Surface Areas and Molecular Volumes. Our analysis was based on the NMR-determined potassium-induced G-quadruplex structure of MYC22-G14T/G23T (PDB entry 1XAV) obtained from the RCSB Protein Databank. We calculated the solvent-accessible surface area, SA, for the G-quadruplex as the sum of the accessible surface areas of all atoms in the structure. The intrinsic volume, VM, of the ODN was calculated as molecular volumes as described by Richards 41,42. The program MSP (Molecular Surface Package) Version 3.9.3 was obtained from Dr. Michael Connolly at

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 74

www.biohedron.com and used to calculate the solvent-accessible surface area and molecular volume, using a 1.4 Å probe radius on a Linux platform.

3.4 Results

K+-induced G-quadruplex Formation. Figure 1a presents the CD spectra of MYC22- G14T/G23T in a 10 mM TBA-phosphate buffer at various KCl concentrations. Inspection of Figure 1a reveals that MYC22-G14T/G23T undergoes a K+-induced transition to a G-quadruplex conformation. The CD spectrum of the G-quadruplex exhibits a positive maximum at 264 nm and a negative minimum at 244 nm. These features are characteristic of the parallel structure with three double-chain-reversal loops bridging the G-quartets 43. This notion is consistent with the NMR-derived structure of the K+-stabilized MYC22-G14T/G23T G-quadruplex 27. Note that the isoelliptic point observed at 254 nm is suggestive of the two-state nature of the K+-induced folding transition of the ODN.

Figure 1b shows the K+-dependence of the molar ellipticity of MYC22-G14T/G23T at 264 nm. Inspection of Figure 3-1b reveals that the G-quadruplex formation is complete at ~80 μM K+ ions. Given the oligonucleotide concentration of ~35 μM, this observation suggests an extremely tight binding of potassium ions to MYC22-G14T/G23T with its subsequent folding into G-quadruplex. For comparison, human telomeric sequences form G-quadruplexes at Na+ or K+ concentrations of 10 to 20 mM 44,45.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 75

Figure 3-1. (a) CD spectra of MYC22-G14T/G23T in a 10 mM TBA-phosphate buffer at various KCl concentrations at 25 °C. (b) Dependence of the molar ellipticity of MYC22- G14T/G23T at 264 nm on the KCl concentration. Experimental data were approximated with Eq. (1) for n = 3 (solid line).

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 76

We used UV absorption at 295 nm to monitor the heat-induced G-quadruplex unfolding transitions at 100, 300, 600, and 1000 μM KCl 46. Figure 3-2a presents a representative UV melting profile of MYC22-G14T/G23T at 1mM KCl. The melting profiles were approximated by the two-state transition model from which the melting temperatures, TM, and van't Hoff 39,40,47 enthalpies, ΔHvH, were determined . It should be noted, however, that G-quadruplex melting is, generally, not a two-state event and may involve one or more intermediates between the folded and unfolded states 48.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 77

Figure 3-2. (a) Representative UV melting profile at 295 nm of the MYC22-G14T/G23T G- quadruplex at 1 mM KCl. (b) Dependence of the melting temperature, TM, of the MYC22- G14T/G23T G-quadruplex on the concentration of K+ ions. Experimental data were approximated by an exponential function solely to guide the eye (solid line).

Strictly speaking, if intermediates are present, a single TM becomes physically meaningless. More rigorous definitions may be the temperatures at which 50 % of the DNA is in the folded, TF, or fully unfolded, TU, state. Determination of TF and TU is experimentally demanding and, generally, cannot be accomplished based on a single UV melting profile.

However, the TM operationally defined as the transition midpoint is between TF and TU being close to (TF + TU)/2. In this respect, TM is a useful and convenient qualitative characteristics of G-quadruplex thermostability. Moreover, if the width of the transition does not change, a change in TM accompanying an alteration in solution conditions (e. g., ionic strength or solvent composition), ΔTM, is roughly equal to ΔTF ≈ ΔTU. Should the width of the transition change, the deviation of ΔTM from ΔTF and ΔTU should be on the order of the change in the transition width. Hence, ΔTM can be treated as a robust measure of the G-quadruplex stability response to a change in environmental conditions.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 78

On the other hand, the model-dependent van't Hoff enthalpy, ΔHvH, determined from UV melting profiles may significantly differ from the true transition enthalpy. Given the uncertainty related to the ΔHvH values, we limit their use to estimating the differential number of counterions,

ΔnM+, condensed around the G-quadruplex and coil states of the ODN with Eqs. (2) and (3). We treat our determined values of ΔnM+ as qualitative rather than quantitative estimates.

Figure 2b plots the transition temperatures, TM, of the MYC22-G14T/G23T G- quadruplex against the KCl concentrations. In agreement with a previous report, the MYC22- G14T/G23T G-quadruplex exhibits an unusually high thermal stability 27. For example, at 1 mM + KCl, the TM is 79.9±0.2 C°. For comparison, the melting temperature of the K -stabilized hybrid-1 G-quadruplex formed by the 26-meric human telomeric sequence d[A3G3(T2AG3)3A2] is 70.6±0.3 C° at 200 mM KCl45.

Conformational Preferences of the K+-stabilized G-quadruplex at Large Excess of Cs+ and TMA+ Ions. Due to their size, Cs+ and TMA+ ions do not fit inside the central cavity of the G-quadruplex. If added to a preformed G-quadruplex, Cs+ and TMA+ ions should predominantly exert their influence on the G-quadruplex stability through the polyelectrolyte effect. We conducted CD spectroscopic measurements to see if the K+-stabilized G-quadruplex undergoes any Cs+- or TMA+-induced transitions.

Figure 3a shows the CD spectra of the MYC22-G14T/G23T G-quadruplex in 300 μM KCl to which CsCl was incrementally added at concentrations ranging from 0 to 800 mM. Inspection of Figure 3a reveals the absence of Cs+-induced conformational changes of the MYC22-G14T/G23T G-quadruplex. Similar observations were made at 600 and 1000 μM KCl (data not shown).

To ensure lack of kinetically trapped conformations of MYC22-G14T/G23T, the CD spectra of the G-quadruplex in 300 μM KCl to which 800 mM CsCl was subsequently added were recorded as the temperature was stepwise increased from 25 to 80 °C (above unfolding) and brought back to 25 °C. Figure 3-3b presents the CD spectra at ascending temperatures of 25, 40, 60, and 80 °C and descending temperatures of 60, 40, and 25 °C. At 25 °C, the CD spectra of the original and annealed G-quadruplex coincide, an observation incompatible with the existence

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 79

of kinetically trapped states. However, if CsCl is added to the ODN prior to KCl, the ODN adopts, at least, partially, a kinetically trapped state(s), We arrive at this conclusion base on ~20 % increase in the CD peak at 265 nm of the annealed relative to the original ODN at 25 °C (data not shown).

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 80

Figure 3-3. (a) CD spectra of the MYC22-G14T/G23T G-quadruplex at 300 μM KCl at various CsCl concentrations. (b) CD spectra the MYC22-G14T/G23T G-quadruplex preformed in 300 μM KCl with subsequently added 800 mM CsCl at temperatures ascending from 25 to 80 °C and descending to 25 °C.

Figure 3-4a shows the CD spectra of the G-quadruplex preformed in 300 μM KCl to which TMACl was incrementally added at concentrations of up to 800 mM. Analogous to CsCl, no TMA+-induced changes in the conformation of the K+-stabilized MYC22-G14T/G23T G- quadruplex could be observed. We also did not observe TMA+-induced CD spectral changes for the G-quadruplex preformed in 600 and 1000 μM KCl (data not shown).

Figure 3-4b presents the CD spectra of the G-quadruplex preformed in the presence of 300 μM KCl, to which 800 mM TMACl was subsequently added, at temperatures increased in a stepwise manner from 25 to 80 °C and then reduced to 25 °C. The CD spectra of the initial and annealed ODN at 25 °C are identical which suggests the absence of kinetically trapped states. However, if TMACl was added to the ODN prior to KCl, the ODN adopts a kinetically trapped state(s). This inference is drawn based on a ~20 % increase in the intensity of the CD band at 265 nm when the sample is heated to 80 °C and subsequently cooled to 25 °C (data not shown).

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 81

Figure 3-4. (a) CD spectra of the MYC22-G14T/G23T G-quadruplex at 300 μM KCl at various TMACl concentrations. (b) CD spectra the MYC22-G14T/G23T G-quadruplex preformed in 300 μM KCl with subsequently added 800 mM TMACl at temperatures ascending from 25 to 80 °C and descending to 25 °C.

Thermal Stability of the K+-stabilized G-quadruplex in the Presence of Cs+ and + TMA Ions. Figures 3-5a, b, and c show the CsCl-dependences of the melting temperatures, TM, for the G-quadruplex preformed at 300, 600, and 1000 μM KCl, respectively. Inspection of

Figure 3-5 reveals that, for all KCl concentrations, the TM steeply decreases as the CsCl concentration increases from 0 to ~200 mM and levels off at higher concentrations.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 82

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 83

Figure 3-5. Dependences of the melting temperature, TM, of the MYC22-G14T/G23T G- quadruplex on the concentration of CsCl at 300 (a), 600 (b), and 1000 (c) μM KCl. Experimental data were approximated by an exponential function solely to guide the eye (solid lines).

Figures 3-6a, b, and c present the TMACl-dependences of the melting temperatures, TM, for the MYC22-G14T/G23T G-quadruplex preformed in 300, 600, and 1000 μM KCl, respectively. Analogous to CsCl, the addition of TMACl causes a steep decrease in TM between 0 and ~200 mM TMACl followed by a plateau at higher concentrations.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 84

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 85

Figure 3-6. Dependences of the melting temperature, TM, of the MYC22-G14T/G23T G- quadruplex on the concentration of TMACl at 300 (a), 600 (b), and 1000 (c) μM KCl. Experimental data were approximated by an exponential function solely to guide the eye (solid lines).

Given the diminution in the stability of the MYC22-G14T/G23T G-quadruplex under the influence of Cs+ and TMA+ ions, it can be anticipated that, at some temperatures, the G- quadruplex-to-coil unfolding transition can be triggered by adding nonstabilizing cations, such as Cs+ or TMA+. To test this hypothesis, we performed CD spectroscopic characterization of the MYC22-G14T/G23T G-quadruplex in 600 μM KCl at 51 °C and increasing concentrations of CsCl. At 51 °C and no CsCl, MYC22-G14T/G23T overwhelmingly exists in its G-quadruplex conformation. However, addition of CsCl brings about an unfolding of the MYC22-G14T/G23T G-quadruplex as is seen from Figure 3-7. The latter presents the CD-detected CsCl-induced unfolding profile of the MYC22-G14T/G23T G-quadruplex at 51 °C. Inspection of Figure 3-7 reveals that the MYC22-G14T/G23T G-quadruplex undergoes a sigmoidal transition with a midpoint of ~75 mM CsCl.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 86

5000

4500

4000

-1 cm

-1 3500

deg M

3 3000 , 10 ,

264 2500

]

 [

2000

1500 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5

log[CsCl]

Figure 3-7. CsCl-dependence of the molar ellipticity of MYC22-G14T/G23T at 264 nm in 600 μM KCl at 51 °C. Experimental data were approximated with Eq. (3) (solid line).

3.5 Discussion

The Stability of the MYC22-G14T/G23T G-quadruplex. The affinity of the ODN for K+ ions, with the G-quadruplex formation to follow, is very high. As is seen from Figure 3-1b, G-quadruplex formation is complete at a stoichiometric ratio of ~2 K+ ions per ODN which corresponds to the structural number of K+ ions sequestered inside the central hole of the G- quadruplex 27. The K+-induced coil (C)-to-G-quadruplex (Q) transition can be formally presented as the following binding reaction:

C + nK+ ↔ Q (Reaction 1)

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 87

By assuming a two-state transition, the binding constant of Reaction 1 can be expressed as follows:

+ n + n Kb = [Q] / ([C][K ] ) = α/ [(1 - α)([K ]T - αn[DNA]) ] (1)

+ where [K ]T is the total concentration of potassium ions in the solution; [DNA] = [C] + [Q] is the total concentration of the ODN in solution; [K+] is the concentration of unbound potassium ions; α = [Q]/[DNA] is the fraction of the ODN in the G-quadruplex conformation; and, consequently, (1 - α) = [C]/[DNA] is the fraction of the ODN in the coil conformation.

The fraction α can be evaluated for each experimental point of the titration profile presented in Figure 1b as the ratio α = ΔX/ΔXmax, where ΔX is the change in an experimental observable (here, molar ellipticity at 264 nm) relative to its initial value at [KCl] = 0; and ΔXmax is the asymptotic maximum change in X corresponding to the fully folded state of the ODN. We fitted the titration profile presented in Figure 1b with Eq. (1) with n = 1, 2, or 3. The fitting was performed numerically based on a direct search optimization technique for fitting nonlinear 49 systems . The best fit of the data was achieved at n = 3 with a binding constant Kb of 1.19ⅹ1014 M-3.

With n equal to 3, one can hypothesize that, in addition to the two NMR-detected potassium ions sequestered inside the central cavity 27, there is an additional K+ ion bound to the G-structure albeit in a non-localized and/or unimobilzed (and, therefore, invisible to NMR) manner. A note of caution is warranted here. Although the isoelliptic point at 254 nm in Figure 3-1a is consistent with the two-state nature of the K+-induced MYC22-G14T/G23T folding, the folding/unfolding transitions of G-quadruplexes are, generally, non-two-state 48,50. Consequently, we do not exclude the possibility that the best fit obtained at n = 3 may be an artifact related to the two-state approximation used to derive Eq. (1). At present, we view the resulting value of n of 3 as a qualitative estimate which calls for further studies to determine the nature of the K+- induced transition of MYC22-G14T/G23T and the number of associated potassium ions.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 88

+ In principle, the K -dependence of the melting temperature, TM, presented in Figure 3-2b can also be used to evaluate the differential number of counterions (M+) bound to the G- quadruplex and coil states 31,51-53:

+ 2 ΔnM+ = (δTM/δln[M ])ΔHM/RTM (2)

The analysis yields a ΔnM+ of 1.8±0.2. This result is close to 2, the number of centrally bound potassium ions. The analysis is compromised by the use in Eq. (2) of the model- dependent van't Hoff enthalpy that may significantly differ from the true transition enthalpy. Notwithstanding, our aggregate results suggest that the folding of MYC22-G14T/G23T is accompanied by an uptake of 2 to 3 potassium ions in good agreement with the structural data.

Cs+ and TMA+ Ions Do Not Cause Structural Alterations in the Preformed G- quadruplex. Once formed in the presence of K+ ions, the MYC22-G14T/G23T G-quadruplex retains its conformation even at a large excess of Cs+ or TMA+ ions (see Figures 3-3a and 3-4a). We infer that the G-quadruplex conformation corresponds to the global minimum of free energy and not to a kinetically trapped local minimum. This notion follows from the observation that the annealing of the preformed G-quadruplex to which CsCl or TMACl was subsequently added does not lead to a change in the CD spectrum (see Figures 3-3b and 3-4b).

On the other hand, if the ODN is initially in 800 mM CsCl or TMACl and KCl is added later, a kinetically trapped state(s) is (are) formed. The kinetically trapped state(s) converts into the thermodynamically stable one once the solution is heated to 80 °C and cooled back to 25 °C. In other words, annealing the ODN eliminates kinetically trapped species converting them into the thermodynamically stable conformation.

In the UV melting experiments discussed below, the G-quadruplexes were preformed in KCl with the desired amounts of CsCl or TMACl added afterward. In addition, the ODN

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 89

samples were annealed prior to the experiments. Therefore, when interpreting our data, we assume a single, thermodynamically stable G-quadruplex conformation.

Modulation of the stability of the K+-stabilized G-quadruplex by Cs+ and TMA+ ions.

At every KCl concentration studied, the thermal stability, TM, of the G-quadruplex sharply decreases as the concentration of Cs+ or TMA+ ions is increased and levels off above ~200 mM + (see Figures 3-5 and 3-6). At 300, 600, and 1000 μM KCl, Cs -induced decreases in TM are 25±2, 21±2, and 23±2 °C (with an average of 23±1 °C), respectively, while TMA+ ions cause a depression of TM by 15±1, 16±1, and 16±1 °C (with an average of 16±1 °C), respectively.

We have previously observed that nonstabilizing ions do not influence or weakly increase the thermal stability of the human telomeric G-quadruplexes 37. In the case of the c-MYC promoter sequence, the situation is more dramatic; Cs+ or TMA+ ions strongly destabilize the G- quadruplex conformation. We rationalize these experimental observations by proposing that counterion condensation is reduced in the G-quadruplex state of MYC22-G14T/G23T relative to the coil state. In other words, there are more counterions (Cs+ or TMA+) associated with the coil state compared to the G-quadruplex state. The differential number, ΔnM+, of counterions bound + to the G-quadruplex and coil states can be estimated from the initial slopes δTM/δln[Cs ] or + δTM/δln[TMA ] in Figures 3-5 and 3-6 using Eq. (2). The estimates yield the values of ΔnM+ of - 1.1±0.1 and -0.9±0.1 for Cs+ and TMA+, respectively.

The effect of differential counterion condensation on G-quadruplex stability ceases when the bulk concentration of the counterions approaches the local concentration in the vicinity of the coil state. Thus, the local counterion concentration in the vicinity of the coil state is on the order of ~200 mM compared to ~1 M, the estimate for duplex polymeric ODN 30.

Cs+-induced Unfolding of the K+-stabilized G-quadruplex. To the best of our knowledge, the unfolding profile shown in Figure 3-7 is the first reported salt-induced unfolding

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 90

transition of a nucleic acid structure. We analyzed the transition profile in Figure 3-7 within the framework of a two-state model in which the fraction unfolded, α, is given by

α = K / (1 + K) (3)

+ Δn + where K = K0(1 + k[M ]) M is the salt-dependent equilibrium constant for the reaction of interconversion between the coil and G-quadruplex states; K0 is the equilibrium constant in the absence of Cs+ ions (M+).

When fitting the data in Figure 3-7 with Eq. (3), the value of K0 was set to 0.0032. The latter was estimated from our UV melting data, as K0 = exp(-ΔG°/RT), where ΔG° = ΔHM(1 -

T/TM). For the experimental conditions in Figure 3-7 (600 μM KCl and 51 °C), the values of TM, -1 -1 ΔHM, and ΔG° are 74.3±0.3 °C, 54.7±1.4 kcal mol , and 3.7±1.4 kcal mol , respectively. The fit produces ΔnM+ of -1.4±0.1. This value is in good agreement with -1.1, the estimate made + above based on the Cs -dependence of TM. The agreement lends credence to our analysis and the two-state approximation used in Eq. (3).

Enhanced Stability and Reduced Polyelectrolyte Effect of the MYC22-G14T/G23T G-quadruplex. The exceptionally high thermal stability of the MYC22-G14T/G23T G- quadruplex is consistent with the picture in which the intramolecular phosphate-phosphate repulsion, a major destabilizing force, is greatly reduced. The reduction may be caused by structural features and/or neutralization of the negative phosphate charges by the internally bound potassium ions. These factors contribute to the high thermal stability of a G-quadruplex and, concomitantly, to weakening the polyelectrolyte effect.

In G-quadruplexes, the polyelectrolye effect may be weakened, partially, due to the presence of cations within the central cavity that decrease the charge density of the DNA. In addition, it may be further weakened due to the structural factors as the near spherical shape of a

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 91

G-quadruplex facilitates an increase in the distance between the G-strands thereby decreasing the charge density in the system. This possibility is underscored by recent theoretical studies which suggest that a charged spherical construct displays reduced counterion condensation relative to a cylinder 54,55. While these arguments can be put forward with respect to any G-quadruplex, their effect should be stronger in the MYC22-G14T/G23T G-quadruplex compared to, for example, human telomeric G-quadruplexes. The latter exhibit a moderate stability, a lower affinity for the stabilizing ion, and a stronger polyelectrolyte effect compared to c-MYC 37,44,45.

As a first approximation, the neutralizing influence of the internally bound potassium ions should correlate to their average separation from the phosphates within the G-quadruplex, while the interphosphate repulsion should depend on the average distance between the phosphate oxygens. One way to analyze the structural disparities between the c-MYC and telomeric G- quadruplexes is to compare their solvent accessible surface areas, SA, and molecular volumes,

VM. The solvent-accessible surface area of a G-quadruplex normalized per phosphate offers a measure of the average interphosphate distance, while the molecular volume per nucleotide characterizes the compactness of a G-quadruplex and, hence, the average distance between the central ions and the phosphates. Table 3-1 lists the computed data on SA and VM for the MYC22-G14T/G23T G-quadruplex and the telomeric Tel22 and Tel26 G-quadruplexes from our previous works 44,45.

Table 3-1. Molecular volumes, VM, and solvent accessible surface areas, SA, of the c-MYC and human telomeric Tel22 and Tel26 G-quadruplexes. Values normalized per nucleotide are shown in parentheses.

------

3 2 VM, Å SA, Å

------c-MYC 6178±16 4442±33

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 92

(281±1) (202±2)

------

Tel22a 6426±26 3907±99

(292±1) (178±5)

------

Tel26b 7092±27 3963±69

(273±1) (152±3)

------a from ref. 44. b from ref. 45.

Inspection of data presented in Table 3-1 reveals that the molecular volume of the c- MYC G-quadruplex normalized per nucleotide is 281±1 cm3mol-1. This number is intermediate between those of the Na+-stabilized antiparallel Tel22 (292±1 cm3mol-1) and the K+-stabilized hybrid-1 Tel26 (273±1 cm3mol-1) human telomeric G-quadruplexes. Thus, the c-MYC G- quadruplex exhibits an average compactness being less compact than the Tel26 G-quadruplex but more compact than the Tel22 G-quadruplex. Thus, we conclude that the central ion-induced neutralization of the phosphates cannot be the main reason for the observed extraordinary stability of the c-MYC G-quadruplex.

The solvent-accessible surface area occupied by each phosphate on the surface of the MYC22-G14T/G23T G-quadruplex is 15 to 30 % larger than that in the Tel22 and Tel26 G- quadruplexes. If the area assigned to each phosphate is viewed as a square, a 15 to 30 % increase in area translates into a 5 to 10 % increase in the interphosphate distance in the MYC22- G14T/G23T relative to the Tel22 and Tel26 G-quadruplexes. The detected increase in

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 93

interphosphate separation, in turn, signifies a lower charge-charge repulsion and a smaller charge density of the MYC22-G14T/G23T G-quadruplex. This notion is in agreement with the observed greater stability and lower polyelectrolyte effect of the MYC22-G14T/G23T G- quadruplex compared to human telomeric G-quadruplexes.

In the aggregate, our simplified analysis is consistent with the picture in which the outstanding stability of the MYC22-G14T/G23T G-quadruplex reflects and, in fact, results from structural features. However, given the limited set of analyzed topologies, it is currently unclear if the large interphosphate separation is an individual feature of the MYC22-G14T/G23T G- quadruplex or a general characteristics of the parallel topology. Further studies are required to shed light on the generality of our findings.

3.6 Conclusion

We report a systematic investigation of the thermal stability of a G-quadrupex formed by the modified c-MYC NHE III1 sequence with two G-to-T mutations (5'- TGAGGGTGGGTAGGGTGGGTAA-3') in the presence of stabilizing and non-stabilizing cations. Introduction of two G-to-T mutations at residue positions of 14 and 23 of the original c-MYC sequence restricts the modified sequence (MYC22-G14T/G23T) to adopting a single, parallel-stranded K+-stabilized G-quadruplex conformation. The MYC22-G14T/G23T G- quadruplex exhibits a steady increase in the melting temperature, TM, upon an increase in the concentration of the stabilizing cation K+. On the other hand, an increase in the concentration of nonstabilizing Cs+ or TMA+ cations at a constant concentration of K+ causes a sharp decline in + + + TM followed by leveling off at ~200 mM Cs or TMA . Moreover, at 51 °C and 600 μM K , an increase in CsCl concentration from 0 to 800 mM leads to unfolding of the G-quadruplex. These observations are consistent with the picture in which the unfolded state of MYC22-G14T/G23T exhibits a stronger polyelectrolyte effect than the G-quadruplex state. In fact, we estimate that the unfolded state of MYC22-G14T/G23T condenses roughly one extra counterion relative to the G-quadruplex state.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 94

Taken together with our earlier studies of human telomeric G-quadruplexes 37, our data suggest that the stabilizing action of cations on the G-quadruplex conformation is due, primarily, to the internally bound metal ions which act as ligands. Nonspecifically bound (condensed) counterions in the vicinity of DNA may exert a very slightly stabilizing or no influence on a G- quadruplex (human telomeric G-quadruplexes) or may strongly destabilize it (MYC22- G14T/G23T). The destabilizing influence of the polyelectrolyte effect on the MYC22- G14T/G23T G-quadruplex may be related to its structural features. In the MYC22-G14T/G23T G-quadruplex, the average solvent accessible surface area per phosphate is 15 to 30 % larger than that in human telomeric G-quadruplexes which translates into a 5 to 10 % greater interphosphate distance in the former. An increase in interphosphate separation signifies a reduced charge-charge repulsion and a lower charge density of the MYC22-G14T/G23T G- quadruplex. This notion is in agreement with the observed greater stability and lower polyelectrolyte effect of the MYC22-G14T/G23T G-quadruplex.

3.7 Acknowledgements

This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada to TVC.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 95

3.8 Reference

1. Burge, S.; Parkinson, G. N.; Hazel, P.; Todd, A. K.; Neidle, S. Quadruplex DNA: sequence, topology and structure. Nucleic Acids Res. 2006, 34, 5402-5415.

2. Huppert, J. L. Four-stranded nucleic acids: structure, function and targeting of G-quadruplexes. Chem. Soc. Rev. 2008, 37, 1375-1384.

3. Lane, A. N.; Chaires, J. B.; Gray, R. D.; Trent, J. O. Stability and kinetics of G-quadruplex structures. Nucleic Acids Res. 2008, 36, 5482-5515.

4. Keniry, M. A. Quadruplex structures in nucleic acids. Biopolymers 2000, 56, 123-146.

5. Shafer, R. H.; Smirnov, I. Biological aspects of DNA/RNA quadruplexes. Biopolymers 2000, 56, 209-227.

6. Lam, E. Y.; Beraldi, D.; Tannahill, D.; Balasubramanian, S. G-quadruplex structures are stable and detectable in human genomic DNA. Nat. Commun. 2013, 4, 1796.

7. Biffi, G.; Tannahill, D.; McCafferty, J.; Balasubramanian, S. Quantitative visualization of DNA G-quadruplex structures in human cells. Nat. Chem. 2013, 5, 182-186.

8. Biffi, G.; Di, A. M.; Tannahill, D.; Balasubramanian, S. Visualization and selective chemical targeting of RNA G-quadruplex structures in the cytoplasm of human cells. Nat. Chem. 2014, 6, 75-80.

9. Rodriguez, R.; Miller, K. M.; Forment, J. V.; Bradshaw, C. R.; Nikan, M.; Britton, S.; Oelschlaegel, T.; Xhemalce, B.; Balasubramanian, S.; Jackson, S. P. Small-molecule-induced DNA damage identifies alternative DNA structures in human genes. Nat. Chem. Biol. 2012, 8, 301-310.

10. Hurley, L. H. DNA and its associated processes as targets for cancer therapy. Nat. Rev. Cancer 2002, 2, 188-200.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 96

11. Huppert, J. L. Four-stranded DNA: cancer, gene regulation and drug development. Phiosl. Trans. Royal Soc. 2007, 365, 2969-2984.

12. Oganesian, L.; Bryan, T. M. Physiological relevance of telomeric G-quadruplex formation: a potential drug target. Bioessays 2007, 29, 155-165.

13. De Cian, A.; Lacroix, L.; Douarre, C.; Temime-Smaali, N.; Trentesaux, C.; Riou, J. F.; Mergny, J. L. Targeting telomeres and telomerase. Biochimie 2008, 90, 131-155.

14. Oster, S. K.; Ho, C. S. W.; Soucie, E. L.; Penn, L. Z. The myc oncogene: (M)under-bar- arvelousl(Y)under-bar (C)under-bar-omplex. Adv. Cancer Res. 2002, 84, 81-154.

15. Eisenman, R. N. Deconstructing myc. Genes Develop. 2001, 15, 2023-2030.

16. Wierstra, I.; Alves, J. The c-myc promoter: still MysterY and challenge. Adv. Cancer Res. 2008, 99, 113-333.

17. Gonzalez, V.; Hurley, L. H. The c-MYC NHE III(1): function and regulation. Annu. Rev. Pharmacol. Toxicol. 2010, 50, 111-129.

18. Simonsson, T.; Pecinka, P.; Kubista, M. DNA tetraplex formation in the control region of c- myc. Nucleic Acids Res. 1998, 26, 1167-1172.

19. Berberich, S. J.; Postel, E. H. PuF/NM23-H2/NDPK-B transactivates a human c-myc promoter-CAT gene via a functional nuclease hypersensitive element. Oncogene 1995, 10, 2343-2347.

20. Postel, E. H.; Berberich, S. J.; Rooney, J. W.; Kaetzel, D. M. Human NM23/nucleoside diphosphate kinase regulates gene expression through DNA binding to nuclease- hypersensitive transcriptional elements. J. Bioenergetics Biomembranes 2000, 32, 277-284.

21. Simonsson, T.; Pribylova, M.; Vorlickova, M. A nuclease hypersensitive element in the human c-myc promoter adopts several distinct i-tetraplex structures. Biochem. Biophys. Res. Commun. 2000, 278, 158-166.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 97

22. Brooks, T. A.; Hurley, L. H. The role of supercoiling in transcriptional control of MYC and its importance in molecular therapeutics. Nat. Rev. Cancer 2009, 9, 849-861.

23. Sun, D.; Hurley, L. H. The importance of negative superhelicity in inducing the formation of G-quadruplex and i-motif structures in the c-myc promoter: implications for drug targeting and control of gene expression. J. Med. Chem. 2009, 52, 2863-2874.

24. Siddiqui-Jain, A.; Grand, C. L.; Bearss, D. J.; Hurley, L. H. Direct evidence for a G- quadruplex in a promoter region and its targeting with a small molecule to repress c-MYC transcription. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 11593-11598.

25. Seenisamy, J.; Rezler, E. M.; Powell, T. J.; Tye, D.; Gokhale, V.; Joshi, C. S.; Siddiqui-Jain, A.; Hurley, L. H. The dynamic character of the G-quadruplex element in the c-MYC promoter and modification by TMPyP4. J. Am. Chem. Soc. 2004, 126, 8702-8709.

26. Phan, A. T.; Modi, Y. S.; Patel, D. J. Propeller-type parallel-stranded G-quadruplexes in the human c-myc promoter. J. Am. Chem. Soc. 2004, 126, 8710-8716.

27. Ambrus, A.; Chen, D.; Dai, J. X.; Jones, R. A.; Yang, D. Z. Solution structure of the biologically relevant g-quadruplex element in the human c-MYC promoter. implications for g-quadruplex stabilization. Biochemistry 2005, 44, 2048-2058.

28. Mathad, R. I.; Hatzakis, E.; Dai, J.; Yang, D. c-MYC promoter G-quadruplex formed at the 5'-end of NHE III1 element: insights into biological relevance and parallel-stranded G- quadruplex stability. Nucleic Acids Res. 2011, 39, 9023-9033.

29. Chaires, J. B. Human telomeric G-quadruplex: thermodynamic and kinetic studies of telomeric quadruplex stability. FEBS J. 2010, 277, 1098-1106.

30. Manning, G. S. Molecular theory of polyelectrolyte solutions with applications to electrostatic properties of polynucleotides. Q. Rev. Biophys. 1978, 11, 179-246.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 98

31. Record, M. T.; Anderson, C. F.; Lohman, T. M. Thermodynamic analysis of ion effects on binding and conformational equilibria of proteins and nucleic-acids. Roles of ion association or release, screening, and ion effects on water activity. Q. Rev. Biophys. 1978, 11, 103-178.

32. Anderson, C. F.; Record, M. T. Salt nucleic-acid interactions. Annu. Rev. Phys. Chem. 1995, 46, 657-700.

33. Anderson, C. F.; Record, M. T. Polyelectrolyte theories and their applications to DNA. Annu. Rev. Phys. Chem. 1982, 33, 191-222.

34. Lipfert, J.; Doniach, S.; Das, R.; Herschlag, D. Understanding nucleic acid-ion interactions. Annu. Rev. Biochem. 2014, 83, 813-841.

35. Plum, G. E.; Pilch, D. S.; Singleton, S. F.; Breslauer, K. J. Nucleic acid hybridization: triplex stability and energetics. Annu. Rev. Biophys. Biomol. Struct. 1995, 24, 319-350.

36. SantaLucia, J., Jr.; Hicks, D. The thermodynamics of DNA structural motifs. Annu. Rev. Biophys. Biomol. Struct. 2004, 33, 415-440.

37. Kim, B. G.; Shek, Y. L.; Chalikian, T. V. Polyelectrolyte effects in G-quadruplexes. Biophys. Chem. 2013, 184, 95-100.

38. Yang, D. Z.; Hurley, L. H. Structure of the biologically relevant G-quadruplex in the c-MYC promoter. Nucleos. Nucleot. Nucl. Acids 2006, 25, 951-968.

39. Marky, L. A.; Breslauer, K. J. Calculating thermodynamic data for transitions of any molecularity from equilibrium melting curves. Biopolymers 1987, 26, 1601-1620.

40. Breslauer, K. J. Extracting thermodynamic data from equilibrium melting curves for oligonucleotide order-disorder transitions. Methods Enzymol. 1995, 259, 221-242.

41. Richards, F. M. Areas, volumes, packing, and protein structure. Annu. Rev. Biophys. Bioeng. 1977, 6, 151-176.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 99

42. Richards, F. M. Calculation of molecular volumes and areas for structures of mnown geometry. Methods Enzymol. 1985, 115, 440-464.

43. Dapic, V.; Abdomerovic, V.; Marrington, R.; Peberdy, J.; Rodger, A.; Trent, J. O.; Bates, P. J. Biophysical and biological properties of quadruplex oligodeoxyribonucleotides. Nucleic Acids Res. 2003, 31, 2097-2107.

44. Fan, H. Y.; Shek, Y. L.; Amiri, A.; Dubins, D. N.; Heerklotz, H.; Macgregor, R. B.; Chalikian, T. V. Volumetric characterization of sodium-induced g-quadruplex formation. J. Am. Chem. Soc. 2011, 133, 4518-4526.

45. Shek, Y. L.; Noudeh, G. D.; Nazari, M.; Heerklotz, H.; Abu-Ghazalah, R. M.; Dubins, D. N.; Chalikian, T. V. Folding thermodynamics of the hybrid-1 type intramolecular human telomeric G-quadruplex. Biopolymers 2014, 101, 216-227.

46. Rachwal, P. A.; Fox, K. R. Quadruplex melting. Methods 2007, 43, 291-301.

47. Breslauer, K. J. Extracting thermodynamic data from equilibrium melting curves for oligonucleotide order-disorder transitions. Methods Mol. Biol. 1994, 26, 347-372.

48. Buscaglia, R.; Gray, R. D.; Chaires, J. B. Thermodynamic characterization of human telomere quadruplex unfolding. Biopolymers 2013, 99, 1006-1018.

49. Luus, R.; Jaakola, T. H. I. Optimization by direct search and systematic reduction of size of search segion. AIChE J. 1973, 19, 760-766.

50. Gray, R. D.; Trent, J. O.; Chaires, J. B. Folding and unfolding pathways of the human telomeric G-quadruplex. J. Mol. Biol. 2014, 426, 1629-1650.

51. Bond, J. P.; Anderson, C. F.; Record, M. T. Conformational transitions of duplex and triplex nucleic acid helices. Thermodynamic analysis of effects of salt concentration on stability using preferential interaction coefficients. Biophys. J. 1994, 67, 825-836.

52. Cheng, Y. K.; Pettitt, B. M. Stabilities of double-strand and triple-strand helical nucleic acids. Prog. Biophys. Moleec. Biol. 1992, 58, 225-257.

Chapter 3: Effects of Salt on the Stability of a G-quadruplexes from the Human c-MYC Promoter 100

53. Owczarzy, R.; Dunietz, I.; Behlke, M. A.; Klotz, I. M.; Walder, J. A. Thermodynamic treatment of oligonucleotide duplex-simplex equilibria. Proc. Natl. Acad. Sci. U. S. A 2003, 100, 14840-14845.

54. Manning, G. S. Electrostatic free energies of spheres, cylinders, and planes in counterion condensation theory with some applications. Macromolecules 2007, 40, 8071-8081.

55. Manning, G. S. Counterion condensation on charged spheres, cylinders, and planes. J. Phys. Chem. B 2007, 111, 8554-8559.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 113

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability

“Reprinted from The Journal of Physical Chemistry B, 120, Byul G. Kim, Ji Long, David N. Dubins, and Tigran V. Chalikian. Ionic Effects on VEGF G-quadruplex Stability, 4963-4971. Copyright (2016), with permission from American Chemical Society”.

Author„s Contribution: Byul G. Kim performed the majority of the experiments, trained the undergraduate student Ji Long to run some remaining experiments, analyzed the data, and participated in the preparation of the manuscript.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 114

4.1 Abstract

In a potassium solution, a modified 22-meric DNA sequence Pu22-T12T13 from a region proximal to the transcription initiation site of the human VEGF gene adopts a single parallel- stranded G-quadruplex conformation with a 1:4:1 loop-size arrangement. We measured the + thermal stability, TM, of the K -stabilized Pu22-T12T13 G-quadruplex as a function of + + + stabilizing K ions and nonstabilizing Cs and TMA ions. The thermal stability, TM, of the Pu22-T12T13 G-quadruplex increases with the concentration of the stabilizing potassium ions, while sharply decreasing upon the addition of the nonstabilizing cations. We interpret these results as underscoring the opposing effects of internal binding and counterion condensation on the stability of the Pu22-T12T13 G-quadruplex. While centrally bound ions stabilize the G- quadruplex conformation, counterion condensation destabilizes it favoring the coil conformation. + + From the initial slopes of the dependences of TM on the concentration of Cs and TMA cations, we estimate that the deleterious effect of counterion condensation stems from roughly one extra counterion associated with the coil relative to the G-quadruplex state of Pu22-T12T13. The reduced accumulation of counterions around the G-quadruplex state of Pu22-T12T13 relative to its coil state is due to the low surface charge density of the G-quadruplex reflecting its structural characteristics. Based on the analysis of our data along with the results of a previous study, we propose that the differential effect of internally (stabilizing) and externally (destabilizing) bound cations may be a general feature of parallel intramolecular G-quadruplexes.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 115

4.2 Introduction

G-quadruplexes have emerged as a major class of non-canonical DNA structures of biological importance and a potential target of anticancer and antitumor strategies 1-6. Guanine-rich DNA sequences with a potential to fold into a G-quadruplex conformation in vivo have been found in important loci of the genome including telomeres, centromeres, immunoglobulin switch regions, mutation-prone hot spots, and promoter regions of oncogenes 4,6-12. A growing body of experimental evidence suggests that DNA and RNA G-quadruplexes, in fact, do exist in the cell and play a regulatory role in genomic processes 13-15. Drug-mediated stabilization of local G- quadruplexes in telomeric ends of chromosomes to inhibit telomerase activity and/or in the promoter regions of target oncogenes to prevent their expression is under development as prospective cancer therapy 3,16.

Controlled regulation of G-quadruplex formation at specific DNA loci requires an understanding of the underlying thermodynamic forces that contribute to making a G-quadruplex conformation favored over the competing double- or single-stranded conformations. The G- quadruplex conformation is maintained by a complex interplay of a host of inter- and intramolecular interactions that act to stabilize or destabilize it12. Elucidation of the nature of these interactions and their thermodynamic characterization is also needed for rational design of low molecular weight compounds that would shift the population equilibrium between DNA structures towards the G-quadruplex conformation.

The electrostatic component of the balance of forces stabilizing ordered nucleic acid structures originates from the polyanionic nature of the DNA and is modulated by the identity and concentration of counterions present in solution 12,17-20. In contrast to other ordered DNA structures, G-quadruplexes are subject to both specific and nonspecific ion binding 12,21,22. The specific binding is due to metal ions coordinated inside the central cavity to the O6 oxygens of guanine bases, while the nonspecific binding involves external condensation of counterions around the phosphate groups 12. In two recent publications, we have developed an experimental approach that permits one to discriminate between the contributions of specific and nonspecific 21,22 cation binding to G-quadruplex stability . In this approach, the thermal stability, TM, of a

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 116

preformed G-quadruplex is measured as a function of the concentration of stabilizing and nonstabilizing monovalent cations. The stabilizing ions are the cations that support G- quadruplex formation by penetrating into the central cavity and forming coordination bonds with the N6 oxygens of the guanines. Nonstabilizing cations are organic or inorganic cations that, given their size (being either too large or too small), do not fit inside the central cavity of the G- quadruplex and, therefore, per se do not support G-quadruplex formation. While the effect of stabilizing cations on TM includes both the internal and external binding, the effect of nonstabilizing cations is limited to external binding as counterions. Comparative analysis of the dependences of TM on the concentration of stabilizing and nonstabilizing cations can be used to discriminate between the influence exerted by the internal (specific) and external (nonspecific) binding on G-quadruplex stability.

Our combined results suggest that specific cation binding stabilizes the G-quadruplex conformation, while nonspecific binding may be neutral or may slightly stabilize or strongly destabilize it 21,22. Counterion condensation (the polyelectrolyte effect) has been found to dramatically decrease the thermal stability, TM, of the parallel G-quadruplex formed by a modified DNA sequence from the promoter region of the c-MYC oncogene 22. Moreover, at some conditions, an increase in the concentration of Cs+ ions in the solution of the K+-stabilized c-MYC G-quadruplex brings about its complete unfolding to the coil conformation 22. This observation is in stark contrast with the counterion-induced increase in thermal stability conventionally observed for other nucleic acid structures establishing the parallel c-MYC G- quadruplex as a new paradigm for DNA-salt interactions 19,23-27.

Despite the progress made by these studies, our understanding of the electrostatic effects on the stability of G-quadruplexes is far from complete. In the present work, we assess the generality of the observed counterion-induced reduction in the stability of parallel G- quadruplexes commonly found in human promoter sequences. Specifically, we study a G- quadruplex formed by a modified 22-meric DNA sequence (Pu22-T12T13) from a region proximal to the transcription initiation site of the human vascular endothelial growth factor (VEGF) gene 28-30. VEGF is a homodimeric signal protein involved in vasculogenesis and angiogenesis 31,32. Overexpression of VEGF is implicated in development and progression of

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 117

many types of human cancers, rheumatoid arthritis, diabetic retinopathy, and age-related macular degeneration 32,33.

The proximal 39-base pair polyG/polyC tract (-88 to -50 relative to the transcription initiation site) is functionally significant for inductive VEGF expression 34. The wild type VEGF-Pu22 sequence (5‟-CGGGGCGGGCCGGGGGCGGGGT-3‟) is a 22-meric oligonucleotide at the 5'- end of the G/C-rich tract (-86 to -65 relative to the transition site). The Pu22-T12T13 sequence (5‟-CGGGGCGGGCCTTGGGCGGGGT-3‟) differs from the wild-type sequence VEGF-Pu22 by the wild-type G12 and G13 substituted with T12 and T13. In potassium solution, Pu22- T12T13 has been shown to adopt a single parallel-stranded G-quadruplex conformation with a 1:4:1 loop-size arrangement 35. In the Pu22-T12T13 G-quadruplex, the four-nucleotide middle loop stretches over the terminal quartet forming a specific caping structure involving the two flanking segments 35. This structural feature of Pu22-T12T13 is in contrast to our previously studied c-MYC G-quadruplex in which the two-nucleotide middle loop remains in the groove 22,36.

4.3 Materials and Methods

Materials. The Pu22-T12T13 oligonucleotide sequence 5‟- CGGGGCGGGCCTTGGGCGGGGT-3‟ was purchased from two sources - ACGT, Inc. (Toronto, ON), and Integrated DNA Technologies (Coralville, IA). No statistically significant inconsistencies were observed in the experimental results obtained on the ACGT or IDT DNA samples. Potassium chloride, cesium chloride, tetramethylammonium chloride, tetrabutylammonium hydroxide, and phosphoric acid were obtained from Sigma-Aldrich Canada (Oakville, ON). EDTA (free acid) was purchased from Fisher Biotech (Fair Lawn, NJ).

The DNA was initially dissolved in and dialyzed against a denaturing buffer consisting of 10 mM phosphoric acid titrated to pH 7.0 with tetrabutylammonium hydroxide and 0.1 mM EDTA. Given their large size, tetrabutylammonium (TBA+) ions do not stabilize a G-quadruplex. Dialysis was carried out with 1000 Da molecular mass cut off Tube-O-Dialyzers from G

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 118

Biosciences (St. Louis, MO). Once the DNA was fully unfolded through dialysis, desired levels of KCl and CsCl or TMACl were added to the solution.

The concentrations of the oligonucleotide were determined spectrophotometrically at 25 °C −1 −1 using a molar extinction coefficient, ɛ260, of 200,400 M cm for the unfolded conformation. The latter was calculated using a nearest neighbor procedure as described by Dr. Richard Owczarzy (http://www.owczarzy.net/extinctionDNA.htm). In the CD measurements and temperature-dependent UV light absorption measurements described here, the DNA concentrations were adjusted to∼30 and∼3μM, respectively.

Optical Spectroscopy. The circular dichroism (CD) spectra of Pu22-T12T13 were recorded at 25 °C in a 1 mm path-length cuvette using an Aviv model 62 DS spectropolarimeter (Aviv Associates, Lakewood, NJ). CD spectroscopic measurements were used to probe the conformations adopted by the DNA under each experimental condition employed in this study.

UV light absorption at 295 nm was measured as a function of temperature in Pu22-T12T13 samples contained in a 10 mm path-length cuvette. The measurements were performed with a Cary 300 Bio spectrophotometer (Varian Canada, Inc., Mississauga, ON). The temperature was changed at a rate of 1 °C/min. The G-quadruplex-to-coil transition temperatures, TM, and van‟t

Hoff enthalpies, ΔHvH, were quantified from our experimental UV melting profiles using the standard procedures 37-40.

Computation of Solvent Accessible Surface Areas and Molecular Volumes. Our analysis was based on the NMR structure of the potassium-induced Pu22-T12T13 G-quadruplex (Protein Data Bank entry 2M27) obtained from the RCSB Protein Data Bank. As 2M27 contains ten models, we treated each model as a distinct structure. We calculated the solvent accessible surface area, SA, for each structure as the sum of the accessible surface areas of all atoms in the 41,42 structure . The intrinsic volume, VM, of the DNA was calculated as a molecular volume as described by Richards 41,43. MSP (Molecular Surface Package) version 3.9.3 was obtained from Dr. Michael Connolly at www.biohedron.com and used to calculate the solvent accessible surface area and molecular volume using a 1.4 Å probe radius on a Linux platform.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 119

4.4 Results

K+-induced Formation and Stability of the Pu22-T12T13 G-quadruplex. Figure 4-1a presents the CD spectra of Pu22-T12T13 in a TBA-phosphate buffer in the absence and presence of KCl. Inspection of the CD spectra presented in Figure 1a reveals the K+-induced formation of a parallel G-quadruplex as can be judged by a minimum at 244 nm and a maximum at 264 nm 44,45. The CD-detected parallel structure of the G-quadruplex adopted by Pu22-T12T13 in the presence of potassium ions is consistent with the NMR solution structure 35.

Figure 4-1b shows the folding profile of Pu22-T12T13 as monitored by the K+-dependence of the molar ellipticity at 264 nm. It can be seen from Figure 4-1b that the Pu22-T12T13 G- quadruplex formation is complete above ~2 mM KCl. This value is intermediate between 80 μM and 15 mM KCl observed for the K+-induced G-quadruplex formation by the MYC22- G14T/G23T sequence and the human telomeric Tel26 sequence, respectively 22,46.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 120

Figure 4-1. (a) CD spectra of Pu22-T12T13 in a 10 mM TBA-phosphate buffer at various KCl concentrations at 25 °C. (b) Dependence of the molar ellipticity of Pu22-T12T13 at 264 nm on the KCl concentration. Experimental data were approximated with Eq. (1) for n = 2 (solid line).

G-quadruplex melting experiments performed at various KCl concentrations were monitored by UV absorption at 295 nm 39. Figure S1 of Supporting Information presents a representative UV melting profile obtained at 2 mM KCl in the absence of nonstabilizing cations. The transition temperatures, TM, and van't Hoff enthalpies, ΔHvH, were determined as described 37-39 + previously . Figure 4-2a presents the K -dependence of the transition temperatures, TM. As is seen from Figure 4-2a, the thermal stability, TM, of Pu22-T12T13 increases in a nearly linear manner as the concentration of stabilizing K+ ions increases. This observation is in agreement with the behavior of all studied G-quadruplexes that display an increase in TM with an increase in the concentration of stabilizing cations 12,21,22,46-48.

Figure 4-2b plots the transition van't Hoff enthalpies, ΔHvH, against the melting temperature,

TM. The use of van't Hoff enthalpies, ΔHvH, in this study is limited to calculation of the differential numbers of associated cations in the G-quadruplex and coil states with Eq. (2) (see below). For the salt conditions studied in this work, the combined dependences of the van't Hoff enthalpies, ΔHvH, for the G-quadruplex-to-coil transitions on the transition temperature, TM, can

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 121

-1 be described by ΔHvH(kcal mol ) = (-14±9) + (0.9 ± 0.1)TM(°C). The slope of the dependence of -1 ΔHvH on TM of 0.9±0.1 kcal mol suggests a positive change in heat capacity, ΔCP, associated with the G-quadruplex-to-coil transition of Pu22-T12T13. This notion is consistent with the results of a study from the Shafer laboratory which reports that melting of G-quadruplexes are accompanied by significant changes in heat capacity 49. However, in a previous work, we have not observed any substantial variation of the transition enthalpy with TM for the human telomeric 46 Tel26 G-quadruplex . The differential heat capacity, ΔCP, of the G-quadruplex and coil states may depend on the specific DNA sequence. Clearly, further studies are required to clarify this important point.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 122

Figure 4-2. (a) Dependence of the melting temperature, TM, of the Pu22-T12T13 G-quadruplex on the concentration of K+ ions. Experimental data were approximated by a linear function (solid line). (b) The G-quadruplex-to-coil transition van't Hoff enthalpies of Pu22-T12T13 plotted as a function of the transition temperature, TM, at 2 (■), 5 (□), and 10 (●) mM KCl and TMACl concentrations ranging from 0 to 800 mM and 2 (○), 5 (♦), an 10 (◊) mM KCl and CsCl concentrations ranging from 0 to 100 mM.

Cs+- and TMA+-induced Conformational Changes of Pu22-T12T13. Figures 3a, b, and c present the CD spectra of the Pu22-T12T13 G-quadruplex preformed in the presence of 2, 5, and 10 mM KCl at different concentrations of TMACl. Inspection of Figures 3a, b, and c reveals that, at all KCl concentrations studied here, the Pu22-T12T13 G-quadruplex retains its conformation as the concentration of TMA+ ions increases up to 800 mM.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 123

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 124

Figure 4-3. CD spectra of the Pu22-T12T13 G-quadruplex at 2 (a), 5 (b), and 10 (c) mM KCl at 25 °C and various concentrations of TMACl.

Figures 4-4a, b, and c show the CD spectra of the Pu22-T12T13 G-quadruplex formed in the presence of 2, 5, and 10 mM KCl at different concentrations of CsCl. In contrast to TMA+ ions, Cs+ ions cause visible changes in the CD spectra of the preformed Pu22-T12T13 G-quadruplex as the concentrations of CsCl increases above 100 mM. However, below 100 mM CsCl, the CD spectra of the Pu22-T12T13 G-quadruplex are similar to the original spectrum in the absence of Cs+ ions. As a plausible explanation, these observations may reflect conformational alterations resulting from replacement of some of the central K+ ions by Cs+ ions at high cesium-to- potassium ratios.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 125

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 126

Figure 4-4. CD spectra of the Pu22-T12T13 G-quadruplex at 2 (a), 5 (b), and 10 (c) mM KCl at 25 °C and various concentrations of CsCl.

It should be also noted that folding of heat-induced unfolded Pu22-T12T13 triggered by a decrease in temperature from 90 to 25 °C in the presence of a large excess of Cs+ over K+ ions takes approximately two hours to complete (see Figure S2 of Supporting Information). The slow folding kinetics in the environment of Cs+ ions is distinct from the fast folding of Pu22-T12T13 in the presence of K+ ions or in K+/TMA+ mixtures which is on the order of the dead time of the experiment (~1 minute). In contrast to Cs+ ions, apparently, due to their larger size, the organic TMA+ ions do not compete with K+ ions even at very high concentrations.

This work aims at understanding the external effect of nonstabilizing cations as counterions. Consequently, we limit the analysis of our Cs+-dependent results to CsCl concentrations of up to 100 mM when their influence can be considered to arise only from external binding.

+ + Cs - and TMA -induced Modulation of Melting Temperature, TM. Panels A, B, and C of + Figure 4-5 present the TMA -dependences of the transition temperatures, TM, of Pu22-T12T13 at

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 127

2, 5, and 10 mM KCl, respectively. Inspection of Figure 4-5 reveals sharp decreases in TM with an increase in the concentration of TMA+ ions following by leveling off above ~200 mM. The net changes in TM are -14±1, -12±1, and -10±1 °C for 2, 5, and 10 mM KCl, respectively.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 128

Figure 4-5. Dependences of the melting temperature, TM, of the Pu22-T12T13 G-quadruplex on the concentration of TMACl at 2 (a), 5 (b), and 10 (c) mM KCl. Experimental data were approximated by an exponential function solely to guide the eye (solid lines).

Figures 4-6a, b, and c plot the changes in TM of the Pu22-T12T13 G-quadruplex as the concentration of CsCl increases from 0 to 100 mM at 2, 5, and 10 mM KCl, respectively. + Inspection of Figure 4-6 reveals a steady decline in TM with as the concentration of Cs ions increases from 0 to 100 mM. The changes in TM are -16±1, -15±1, and -11±1 °C for 2, 5, and 10 mM KCl, respectively.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 129

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 130

Figure 4-6. Dependences of the melting temperature, TM, of the Pu22-T12T13 G-quadruplex on the concentration of CsCl at 2 (a), 5 (b), and 10 (c) mM KCl. Experimental data were approximated by an exponential function solely to guide the eye (solid lines).

4.5 Discussion

Stability of the Pu22-T12T13 G-quadruplex. The folding profile presented in Figure 1b can be formally analyzed as an isothermal G-quadruplex folding reaction induced by cation binding 50,51. Within the framework of an association reaction in which the binding of n potassium ions to the unfolded coil-like state (C) results in the folded G-quadruplex state (Q), one can write:

C + nK+ ↔ Q (Reaction 1)

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 131

Under the assumption of a two-state coil-to-G-quadruplex transition, the apparent binding constant, Kb, of Reaction 1 is given by the relationship:

+ n + n Kb = [Q] / ([C][K ] ) = α/[(1 – α)([K ]T – αn[DNA]) ] (1)

+ where [DNA] = [Q] + [C] is the total concentration of DNA; [K ]T is the total concentration of potassium ions; α = [Q]/[DNA] is the fraction of the DNA in the G-quadruplex conformation.

The best fit of experimental data in Figure 1b with Eq. (1) is produced by n equal to 2 and 6 -2 yields an apparent binding constant, Kb, of (2.89±0.01) ⅹ10 M . The n of 2 is consistent with the structural number of potassium ions sequestered inside the central cavity of the G-quadruplex 35.

The number of counterions that become released to the bulk upon the G-quadruplex unfolding + can also be evaluated from the K -dependence of the melting temperature, TM, presented in Figure 2a 18,52-54:

+ 2 ΔnM+ = (δTM/δln[M ]) ΔHM/RTM (2)

+ where ΔnM+ is the differential number of counterions (M ) thermodynamically associated with the G-quadruplex and coil conformations.

Using Eq. (2) in conjunction with the data presented in Figure 2a, we calculate a ΔnM+ of 1.7±0.2, consistent with our estimate based on Eq.(1) and Figure 1b.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 132

TMA+ Ions Do not Influence the Conformation of the K+-stabilized Pu22-T12T13 G- quadruplex. The CD spectroscopic spectra shown in Figures 4-3 and 4-4 suggest that, while TMA+ ions do not influence the structure of the preformed Pu22-T12T13 G-quadruplex, Cs+ ions cause its structural alterations at concentrations higher than ~100 mM. The presence of Cs+ cations leads to a slow folding kinetics of the thermally unfolded G-quadruplex which additionally complicates the temperature-dependent measurements conducted in this work. We, therefore, limit discussions of Cs+-dependent ionic effects to the concentrations of up to 100 mM.

The influence of TMA+ (between 0 and 800 mM) and Cs+ ions (between 0 and 100 mM) on the stability of the Pu22-T12T13 G-quadruplex preformed in the environment of potassium ions can be viewed as originating from and reflecting the polyelectrolyte nature of DNA (external nonspecific binding). At higher concentrations, Cs+ is likely to compete with K+ for the sites inside the central cavity of the G-quadruplex which would, consequently, bring about conformational alterations and slow the folding kinetics.

TMA+- and Cs+-induced Modulation of the Stability of K+-stabilized Pu22-T12T13 G- quadruplex. Results of our previous studies suggest that counterion condensation may slightly increase the stability of a G-quadruplex, may not have any sizeable impact on it, or may significantly destabilize it 21,22. The K+-stabilized parallel G-quadruplex formed by the modified MYC22-G14T/G23T sequence from the promoter region of the human c-MYC oncogene is structurally and functionally similar to the Pu22-T12T13 G-quadruplex studied here. The MYC22-G14T/G23T G-quadruplex preformed in the atmosphere of K+ ions is strongly destabilized in the presence of Cs+ and TMA+ cations 22.

Inspection of Figures 4-5a, b, and c reveals that the influence of TMA+ ions on the stability of the Pu22-T12T13 G-quadruplex is qualitatively similar to their effect on the stability of the c- + MYC G-quadruplex. At each K concentration studied in this investigation, the TM decreases sharply as the concentration of TMA+ ions increase from zero to ~200 mM and levels off at higher concentrations. On a similar note, an increase in Cs+ concentration from 0 to 100 mM also results in a sharp decline in TM (see Figures 6a, b, and c).

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 133

+ The observed decreases in TM of the preformed K -stabilized Pu22-T12T13 G-quadruplex with an increase in the concentration of nonstabilizing Cs+ and TMA+ cations unequivocally suggest that counterion condensation (the polyelectrolyte effect) destabilizes the G-quadruplex conformation. This notion implies a smaller number of counterions associated with the G- quadruplex relative to the coil state. The differential number of counterions condensed around + the G-quadruplex and coil states can be determined from the initial slopes (δTM/δln[M ]) in + Figures 4-5 and 4-6 using Eq. (2). For TMA ions, our calculated values of ΔnM+ are all equal to + -0.6±0.2 for 2, 5, and 10 mM KCl. For Cs ions, the values of ΔnM+ are equal to -0.6±0.2, - 0.8±0.2, and -0.9±0.2 for 2, 5, and 10 mM KCl (on average, -0.8±0.2). These numbers are qualitatively similar to -0.9±0.1 and -1.1±0.1, the differential number of TMA+ and Cs+ ions bound to the G-quadruplex and coil states of MYC22-G14T/G23T 22. Thus, similar to MYC22- G14T/G23T, the destabilizing effect of counterion condensation originates from roughly one extra counterion condensed around the coil relative to the G-quadruplex state of Pu22-T12T13.

There is one more important quantitative feature that is common for the Pu22-T12T13 and + MYC22-G14T/G23T G-quadruplexes. The TM-versus-[TMA ] dependences of the former and + + 22 the TM-versus-[TMA ] and TM-versus-[Cs ] dependences of the latter all level off at ~200 mM . We have interpreted this experimental result as suggesting that the local counterion concentration in the vicinity of the coil state of MYC22-G14T/G23T is on the order of ~200 mM 22. We now extend this estimate to the Pu22-T12T13 sequence and propose that a local counterion concentration of 200 mM may be a common property of single stranded DNA conformations.

Structural Considerations. A smaller number of counterions associated with the G- quadruplex state is consistent with a lower charge density of the folded relative to the unfolded conformations of Pu22-T12T13. A similar notion supported by subsequent structural calculations has been put forward with respect to the folded and unfolded states of MYC22- G14T/G23T 22. The Pu22-T12T13 and MYC22-G14T/G23T sequences are both 22-meric olgonucleotides that fold in the presence of potassium ions into a parallel G-quadruplex with two + K ions sequestered in the central cavity. The volume, ρq, and surface, ζq, charge densities of a G-quadruplex can be obtained by dividing the number of formal charges, N, by the molecular,

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 134

volume, VM, and solvent accessible surface area, SA, respectively. For the Pu22-T12T13 G- quadruplex, N equals 20 (22 negative phosphate charged minus two positive charges of the sequestered potassium ions). Our calculated values of VM and SA of the Pu22-T12T13 G- quadruplex are 6180±38 Å 3 and 4476±33 Å 2, respectively. These values are close to those 3 2 calculated for the parallel MYC22-G14T/G23T G-quadruplex (VM = 6178 Å and SA = 4442 Å ) but differ significantly from those calculated for the Na+-stabilized Tel22 antiparallel G- 3 2 22 quadruplex (VM = 6426 Å and SA = 3907 Å ) . The latter is bulkier being, at the same time, less solvent-exposed compared to the parallel G-quadruplexes.

The volume, ρq, and surface, ζq, charge densities of the Pu22-T12T13 G-quadruplex are equal to (3.24±0.02)ⅹ10-3 e/Å 3 (20 /6180) and (4.47±0.03)ⅹ10-3 e/Å 2 (20 / 4476), respectively.

These charge densities are similar to those of the MYC22-G14T/G23T G-quadruplex (ρq = 3.24 -3 3 -3 2 x 10 e/Å and ζq = 4.50 x 10 e/Å ) but differ significantly from those of the Tel22 G- -3 3 -3 2 22 quadruplex (ρq = 2.96 x 10 e/Å and ζq = 4.9 x 10 e/Å ) . In fact, the parallel Pu22-T12T13 and MYC22-G14T/G23T G-quadruplexes exhibit the lowest charge density amongst the G- duplexes and polymeric duplexes analyzed in our previous study 22. This result is consistent with and accounts for the strongly destabilizing effect of counterion condensation observed for the parallel Pu22-T12T13 and MYC22-G14T/G23T G-quadruplexes.

4.6 Concluding Remarks

We characterize the structural and thermodynamic properties of a G-quadruplex formed by a modified 22-meric DNA sequence from a region proximal to the transcription initiation site of the human VEGF gene in the presence and absence of stabilizing and nostabilizing cations. In a potassium solution, the Pu22-T12T13 sequence adopts a single parallel-stranded G-quadruplex conformation with a 1:4:1 loop-size arrangement. The thermal stability, TM, of the Pu22-T12T13 G-quadruplex increases with the concentration of the stabilizing K+ ions. In contrast, the G- + + quadruplex-to-coil transition temperatures, TM, sharply decrease as nonstabilizing Cs and TMA ions are added to the solution of the preformed K+-stabilized G-quadruplex. Stabilizing ions modulate the conformational preferences of guanine-rich DNA sequences by acting both as

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 135

internally bound ligands and as counterions forming the external ionic atmosphere of the folded (G-quadruplex) and unfolded (coil) states. In contrast, nonstabilizing ions may only act as counterions. With this notion, our results emphasize the opposing effects of internal binding and counterion condensation on the stability of the Pu22-T12T13 G-quadruplex. While centrally bound ions stabilize the G-quadruplex conformation, counterion condensation destabilizes it favoring the coil conformation. From the slope of the dependence of TM on the concentration of Cs+ and TMA+ cations, we estimate that the destabilizing effect of counterion condensation originates from roughly one extra counterion condensed around the coil relative to the G- quadruplex state of Pu22-T12T13. The reduced counterion binding by the G-quadruplex state of Pu22-T12T13 compared to the coil state is due to the low surface charge density of the G- quadruplex reflecting its structural characteristics. Taken together with the results of a previous study, our present results are consistent with the picture in which the differential effect of internally (stabilizing) and externally (destabilizing) bound cations is a common feature of parallel intramolecular G-quadruplexes. Additional studies are, however, needed to prove or refute the generality of this notion.

4.7 Supporting Information

Figure S1 presents a representative UV melting profile of Pu22-T12T13 at 295 nm obtained in the presence of 2 mM KCl. Figure S2 presents the time dependence of the molar ellipticity of Pu22-T12T13 at 264 nm following a rapid decrease in temperature from 95 to 25 °C in the presence of 5 mM KCl and 200 mM CsCl. This material is available free of charge via the internet at http://pubs.acs.org.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 136

4.8 Acknowledgements

The authors are grateful to Prof. Robert B. Macgregor, Jr., for useful discussions and constructive comments. This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada to TVC.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 137

4.9 Reference

1. Neidle, S.; Read, M. A. G-quadruplexes as therapeutic targets. Biopolymers 2000, 56, 195-208.

2. Neidle, S. Human telomeric G-quadruplex: the current status of telomeric G-quadruplexes as therapeutic targets in human cancer. FEBS J. 2010, 277, 1118-1125.

3. Balasubramanian, S.; Hurley, L. H.; Neidle, S. Targeting G-quadruplexes in gene promoters: a novel anticancer strategy? Nature Rev. Drug Disc. 2011, 10, 261-275.

4. Bochman, M. L.; Paeschke, K.; Zakian, V. A. DNA secondary structures: stability and function of G-quadruplex structures. Nature Rev. Genet. 2012, 13, 770-780.

5. Hurley, L. H. DNA and its associated processes as targets for cancer therapy. Nat. Rev. Cancer 2002, 2, 188-200.

6. Huppert, J. L. Four-stranded nucleic acids: structure, function and targeting of G-quadruplexes. Chem. Soc. Rev. 2008, 37, 1375-1384.

7. Huppert, J. L. Four-stranded DNA: cancer, gene regulation and drug development. Phil. Trans. Royal Soc. A 2007, 365, 2969-2984.

8. Shafer, R. H.; Smirnov, I. Biological aspects of DNA/RNA quadruplexes. Biopolymers 2000, 56, 209-227.

9. Oganesian, L.; Bryan, T. M. Physiological relevance of telomeric G-quadruplex formation: a potential drug target. Bioessays 2007, 29, 155-165.

10. De Cian, A.; Lacroix, L.; Douarre, C.; Temime-Smaali, N.; Trentesaux, C.; Riou, J. F.; Mergny, J. L. Targeting telomeres and telomerase. Biochimie 2008, 90, 131-155.

11. Burge, S.; Parkinson, G. N.; Hazel, P.; Todd, A. K.; Neidle, S. Quadruplex DNA: sequence, topology and structure. Nucleic Acids Res. 2006, 34, 5402-5415.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 138

12. Lane, A. N.; Chaires, J. B.; Gray, R. D.; Trent, J. O. Stability and kinetics of G-quadruplex structures. Nucleic Acids Res. 2008, 36, 5482-5515.

13. Lam, E. Y.; Beraldi, D.; Tannahill, D.; Balasubramanian, S. G-quadruplex structures are stable and detectable in human genomic DNA. Nat. Commun. 2013, 4, 1796.

14. Biffi, G.; Tannahill, D.; McCafferty, J.; Balasubramanian, S. Quantitative visualization of DNA G-quadruplex structures in human cells. Nat. Chem. 2013, 5, 182-186.

15. Biffi, G.; Di, A. M.; Tannahill, D.; Balasubramanian, S. Visualization and selective chemical targeting of RNA G-quadruplex structures in the cytoplasm of human cells. Nat. Chem. 2014, 6, 75-80.

16. Rodriguez, R.; Miller, K. M.; Forment, J. V.; Bradshaw, C. R.; Nikan, M.; Britton, S.; Oelschlaegel, T.; Xhemalce, B.; Balasubramanian, S.; Jackson, S. P. Small-molecule-induced DNA damage identifies alternative DNA structures in human genes. Nat. Chem. Biol. 2012, 8, 301-310.

17. Manning, G. S. Molecular theory of polyelectrolyte solutions with applications to electrostatic properties of polynucleotides. Q. Rev. Biophys. 1978, 11, 179-246.

18. Record, M. T.; Anderson, C. F.; Lohman, T. M. Thermodynamic analysis of ion effects on binding and conformational equilibria of proteins and nucleic acids: the roles of ion association or release, screening, and ion effects on water activity. Q. Rev. Biophys. 1978, 11, 103-178.

19. Anderson, C. F.; Record, M. T. Salt - nucleic acid interactions. Annu. Rev, Phys. Chem. 1995, 46, 657-700.

20. Anderson, C. F.; Record, M. T. Polyelectrolyte theories and their applications to DNA. Annu. Rev. Phys. Chem. 1982, 33, 191-222.

21. Kim, B. G.; Shek, Y. L.; Chalikian, T. V. Polyelectrolyte effects in G-quadruplexes. Biophys. Chem. 2013, 184, 95-100.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 139

22. Kim, B. G.; Evans, H. M.; Dubins, D. N.; Chalikian, T. V. Effects of salt on the stability of a G-quadruplex from the human c-MYC promoter. Biochemistry 2015, 54, 3420-3430.

23. Plum, G. E.; Pilch, D. S.; Singleton, S. F.; Breslauer, K. J. Nucleic acid hybridization: triplex stability and energetics. Annu. Rev. Biophys. Biomol. Struct. 1995, 24, 319-350.

24. Volker, J.; Klump, H. H.; Manning, G. S.; Breslauer, K. J. Counterion association with native and denatured nucleic acids: an experimental approach. J. Mol. Biol. 2001, 310, 1011-1025.

25. Volker, J.; Klump, H. H.; Breslauer, K. J. Communication between noncontacting macromolecules. Proc. Natl. Acad. Sci. U. S. A 2001, 98, 7694-7699.

26. Volker, J.; Klump, H. H. Electrostatic effects in DNA triple helices. Biochemistry 1994, 33, 13502-13508.

27. Santa Lucia, J., Jr.; Hicks, D. The thermodynamics of DNA structural motifs. Annu. Rev. Biophys. Biomol. Struct. 2004, 33, 415-440.

28. Guo, K.; Gokhale, V.; Hurley, L. H.; Sun, D. Intramolecularly folded G-quadruplex and i- motif structures in the proximal promoter of the vascular endothelial growth factor gene. Nucleic Acids Res. 2008, 36, 4598-4608.

29. Sun, D.; Liu, W. J.; Guo, K. X.; Rusche, J. J.; Ebbinghaus, S.; Gokhale, V.; Hurley, L. H. The proximal promoter region of the human vascular endothelial growth factor gene has a G- quadruplex structure that can be targeted by G-quadruplex-interactive agents. Mol. Cancer Therapeut. 2008, 7, 880-889.

30. Sun, D.; Guo, K.; Shin, Y. J. Evidence of the formation of G-quadruplex structures in the promoter region of the human vascular endothelial growth factor gene. Nucleic Acids Res. 2011, 39, 1256-1265.

31. Ferrara, N. Vascular endothelial growth factor. Eur. J. Cancer 1996, 32A, 2413-2422.

32. Hoeben, A.; Landuyt, B.; Highley, M. S.; Wildiers, H.; Van Oosterom, A. T.; De Bruijn, E. A. Vascular endothelial growth factor and angiogenesis. Pharmacol. Rev. 2004, 56, 549-580.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 140

33. Goel, H. L.; Mercurio, A. M. VEGF targets the tumour cell. Nat. Rev. Cancer 2013, 13, 871- 882.

34. Loureiro, R. M. B.; D'Amore, P. A. Transcriptional regulation of vascular endothelial growth factor in cancer. Cytokine Growth Factor Rev. 2005, 16, 77-89.

35. Agrawal, P.; Hatzakis, E.; Guo, K. X.; Carver, M.; Yang, D. Z. Solution structure of the major G-quadruplex formed in the human VEGF promoter in K+: insights into loop interactions of the parallel G-quadruplexes. Nucleic Acids Res. 2013, 41, 10584-10592.

36. Ambrus, A.; Chen, D.; Dai, J. X.; Jones, R. A.; Yang, D. Z. Solution structure of the biologically relevant G-quadruplex element in the human c-MYC promoter. Implications for G-quadruplex stabilization. Biochemistry 2005, 44, 2048-2058.

37. Marky, L. A.; Breslauer, K. J. Calculating thermodynamic data for transitions of any molecularity from equilibrium melting curves. Biopolymers 1987, 26, 1601-1620.

38. Mergny, J. L.; Lacroix, L. Analysis of thermal melting curves. Oligonucleotides 2003, 13, 515-537.

39. Rachwal, P. A.; Fox, K. R. Quadruplex melting. Methods 2007, 43, 291-301.

40. Breslauer, K. J. Extracting thermodynamic data from equilibrium melting curves for oligonucleotide order-disorder transitions. Methods Enzymol. 1995, 259, 221-242.

41. Richards, F. M. Areas, Volumes, packing, and protein structure. Annu. Rev. Biophys. Bioeng. 1977, 6, 151-176.

42. Connolly, M. L. Solvent accessible surfaces of proteins and nucleic acids. Science 1983, 221, 709-713.

43. Richards, F. M. Calculation of molecular volumes and areas for structures of known geometry. Methods Enzymol. 1985, 115, 440-464.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 141

44. Dapic, V.; Abdomerovic, V.; Marrington, R.; Peberdy, J.; Rodger, A.; Trent, J. O.; Bates, P. J. Biophysical and biological properties of quadruplex oligodeoxyribonucleotides. Nucleic Acids Res. 2003, 31, 2097-2107.

45. Vorlickova, M.; Kejnovska, I.; Sagi, J.; Renciuk, D.; Bednarova, K.; Motlova, J.; Kypr, J. Circular dichroism and guanine quadruplexes. Methods 2012, 57, 64-75.

46. Shek, Y. L.; Noudeh, G. D.; Nazari, M.; Heerklotz, H.; Abu-Ghazalah, R. M.; Dubins, D. N.; Chalikian, T. V. Folding thermodynamics of the hybrid-1 type intramolecular human telomeric G-quadruplex. Biopolymers 2014, 101, 216-227.

47. Olsen, C. M.; Gmeiner, W. H.; Marky, L. A. Unfolding of G-quadruplexes: Energetic, and ion and water contributions of G-quartet stacking. J. Phys. Chem. B 2006, 110, 6962-6969.

48. Fan, H. Y.; Shek, Y. L.; Amiri, A.; Dubins, D. N.; Heerklotz, H.; Macgregor, R. B.; Chalikian, T. V. Volumetric characterization of sodium-induced G-quadruplex formation. J. Am. Chem. Soc. 2011, 133 (12), 4518-4526.

49. Majhi, P. R.; Qi, J. Y.; Tang, C. F.; Shafer, R. H. Heat capacity changes associated with guanine quadruplex formation: an isothermal titration calorimetry study. Biopolymers 2008, 89, 302-309.

50. Gray, R. D.; Chaires, J. B. Linkage of cation binding and folding in human telomeric quadruplex DNA. Biophys. Chem. 2011, 159, 205-209.

51. Gray, R. D.; Chaires, J. B. Isothermal folding of G-quadruplexes. Methods 2012, 57, 47-55.

52. Bond, J. P.; Anderson, C. F.; Record, M. T. Conformational transitions of duplex and triplex nucleic acid helices: thermodynamic analysis of effects of salt concentration on stability using preferential interaction coefficients. Biophys. J. 1994, 67, 825-836.

53. Cheng, Y. K.; Pettitt, B. M. Stabilities of double- and triple-strand helical nucleic acids. Prog. Biophys. Molec. Biol. 1992, 58, 225-257.

CHAPTER 4: Ionic Effects on VEGF G-quadruplex Stability 142

54. Owczarzy, R.; Dunietz, I.; Behlke, M. A.; Klotz, I. M.; Walder, J. A. Thermodynamic treatment of oligonucleotide duplex-simplex equilibria. Proc. Natl. Acad. Sci. U. S. A 2003, 100, 14840-14845.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 144

CHAPTER 5: Thermodynamic Linkage Analysis of pH- induced Folding and Unfolding Transitions of i-motifs

“Reprinted from Biophysical Chemistry, 216, Byul G. Kim and Tigran V. Chalikian. Thermodynamic Linkage Analysis of pH-induced Folding and Unfolding Transitions of i-motifs, 19-22. Copyright (2016), with permission from Elsevier B.V.”.

Author„s Contribution: Byul G. Kim performed all experiments, analyzed the data, and participated in the preparation of the manuscript.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 145

5.1 Abstract

We describe the pH-induced folding/unfolding transitions of i-motifs by a linkage thermodynamics-based formalism in terms of three pKa‟s of cytosines, namely, an apparent pKa in the unfolded conformation, pKau, and two apparent pKa‟s in the folded state, pKaf1 and pKaf2. For the 5′-TTACCCACCCTACCCACCCTCA-3' sequence from the human c-MYC oncogene promoter region, the values of pKau, pKaf1, and pKaf2 are 4.8, 6.0, and 3.6, respectively. With these pKa‟s, we calculate the differential number of protons bound to the folded and unfolded states as a function of pH. Analysis along these lines offers an alternative interpretation to the experimentally observed shift in the pH-induced unfolded-to-i-motif transitions to neutral pH in the presence of cosolvents and crowders.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 146

5.2 Introduction

At slightly acidic to near-neutral solution pH, cytosine-rich DNA sequences exhibit a propensity to fold into an i-motif conformation. The latter represents a noncanonical DNA conformation with a tetrameric arrangement of two parallel-stranded duplexes which intercalate into each other in an antiparallel orientation1-4. Significantly, i-motifs are believed to be of biological importance, since cytosine-rich sequences capable of folding into i-motifs have been found in important genomic loci including but not limited to centromeric and telomeric regions of human chromosomes and the promoter regions of numerous genes2.

The building blocks of an i-motif are made of hemiprotonated cytosine·cytosine+ (C·C+) base pairs which are stabilized by an unusual arrangement of three hydrogen bonds and hold together the pairs of the parallel strands forming the i-motif structure2, 5. As cytosine protonation is involved in i-motif formation, the stability of i-motif structures is strongly pH-dependent with i- motifs being favored by slightly acidic pH. The maximum of i-motif stability is generally observed between pH 4 and 5 which corresponds to the pKa of the N3 nitrogens of cytosines in a free mononucleotide (and, presumably, in an unfolded DNA) of ~4.6 2,6-9. A further decrease in pH leads to protonation of the remaining cytosines with a subsequent disruption of the i-motif around ~pH 3 2,8,10.

The pH-dependence of i-motif stability, while being well documented, has never been analyzed in a rigorous way afforded by linkage thermodynamics. In general, the dissociation constant, pKa, of a titrable group is modulated by its microenvironment and the nature and extent of interactions with other groups. Clearly, the pKa of cytosines in an unfolded oligonucleotide and a folded i-motif should be significantly different. Moreover, there should be two distinct dissociation constants, pKa, for the two sets of protonated and deprotonated cytosines forming the hemiprotonated C·C+ base pairs in a folded i-motif. The favorable change in free energy, ΔG°, accompanying i-motif formation will drive the protonation of the initial set of cytosines in + the hemiprotonated C·C base pairs to pH values higher than ~4.6, the pKa of unfolded cytosines. Following the same logic, the protonation of the second set of cytosines (initially deprotonated in the i-motif conformation) will be suppressed and take place below pH 4.6.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 147

With these comments, we analyze below the pH-induced folding/unfolding transition profile of a representative i-motif formed by a 22-meric oligonucleotide (5′- TTACCCACCCTACCCACCCTCA-3') from the human c-MYC oncogene promoter region (ODN). The sequence is complementary to the guanine-rich G-quadruplex-forming sequence from the c-MYC promoter site that has been characterized in our recent work11. Cytosine-rich sequences from the promoter region of c-MYC have been shown to fold into i-motif conformations at slightly acidic pH1, 7, 10, 12-15.

Figure 5-1 presents the CD spectra of the ODN recorded at pH ranging from 9.5 to 4.5 (panel A) and from 4.8 to 2.3 (panel B). Inspection of Figures 5-1a and b reveals several important observations. Firstly, a decrease in pH to ~5 results in formation of an i-motif structure as can be judged by the CD spectrum with a characteristic maximum at 285 nm and a minimum at 258 nm (see Figure 1a) 2, 9, 10, 16. Secondly, the isoelliptic points observed at 246 and 276 nm in Figure 1a suggest that the pH-induced i-motif folding is likely a two-state transition and does not involve populated intermediates. Thirdly, a decrease in pH to below ~3 results in an unfolding of the i- motif (see Figure 5-1b). Finally, isoelliptic points at 233 and 271 nm in Figure 5-1b are consistent with the two-state nature of the second pH-induced transition (i-motif-to-unfolded transition). Given the isoelliptic points in Figures 5-1a and b, we analyze the pH-induced folding/unfolding transitions of the ODN within the framework of a two-state model with equilibrium constant, K = [F]/[U] (where [F] and [U] are concentrations of the folded and unfolded DNA states, respectively). A note of caution is warranted here. The observed isoelliptic points are consistent with but cannot be viewed as an absolute proof of the two-state nature of the folding and unfolding transitions. While, in this work, we treat our data in a two- state manner, further studies are required to clarify the nature of the pH-induced folding/unfolding transitions of i-motifs, in general, and the ODN, in particular.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 148

Figure 5-1. The CD spectra of the ODN between pH 9.5 and 4.5 (panel A) and pH 4.8 and 2.3 (panel B).

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 149

According to the thermodynamic linkage theory, the pH-dependence of the equilibrium constant, K, is given by the relationship17-19:

∏ K = K0 (1) ∏

where K0 is the equilibrium constant in the absence of protons (at alkaline pH); pKafi is the pKa of the i-th cytosine in the folded (i-motif) conformation; and pKaui is the pKa of the same cytosine in the unfolded conformation. The specific values of pKafi and pKaui may depend on the internal and external factors. The former reflect the sequence and structure of the i-motif- forming DNA and the position of the specific cytosine in it, while the latter reflect temperature, pressure, and solution composition, including the presence of cosovents and salts.

Figure 5-2 presents the pH-induced folding/unfolding profile of the ODN as traced by the molar ellipticity at 284 nm. Inspection of Figure 5-2 reveals two near-sigmoidal transitions centered at pH 6.1 and 3.6 representing the unfolded-to-i-motif and i-motif-to-unfolded transitions, respectively. We simplify Eq. (1) to fit the experimental data plotted in Figure 5-2. To this end, we assume that the cytosines of the ODN are uniformly characterized by a set of three distinct pKa's (in reality, the individual pKa‟s of cytosines may be distinct depending on their position). In the unfolded conformation, all 12 cytosines are assumed to exhibit the same dissociation constant of pKau. Further, we assume that, in the i-motif conformation, the six initially protonating cytosines exhibit a higher dissociation constant of pKaf1, while the remaining six cytosines are characterized by a lower pKa of pKaf2. With these approximations, Eq. (1) simplifies to the form:

K = K0 (2)

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 150

We used Eq. (2) to fit the experimental data points shown in Figure 5-2. The fit produced pKau, pKaf1, and pKaf2 of 4.8±0.1, 6.0±0.1, and 3.6±0.1, respectively, with K0 equal to 0.07±0.05. The equilibrium constant of 0.07±0.05 is rather large and consistent with a picture in which an i- motif population of a few percent persists even at alkaline pH with virtually no charged (protonated) cytosines. This notion is suggestive of the existence of highly unstable i-motifs containing unprotonated C·C base pairs. Alternatively, there is a possibility that this quantitative result is an artifact reflecting the approximations involved in deriving Eq. (2). Clearly, further studies are required to clarify this intriguing and important point.

8 -1

cm 6

-1

deg M

3 , 10 ,

284

]

 [

2 2 4 6 8 10 pH

Figure 5-2. The pH-dependence of the molar ellipticity at 284 nm. Experimental data were approximated with Eq. (2) (solid line).

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 151

The values of pKau, pKaf1, and pKaf2 we obtained can be used to calculate the differential number of protons bound to the folded (i-motif) state relative to the unfolded state:

Δn (pH) = ( ) (3) H+

where Ncyt is the total number of cytosines participating in i-motif formation.

Figure 5-3 plots the pH-dependence of the differential number of bound protons, ΔnH+, computed with Eq. (3). Inspection of Figure 3 reveals that, as the pH decreases from neutral to acidic, ΔnH+ initially increases from 0 to 2.7 (at ~pH 5.6), then decreases to -2.7 (at ~pH 4.0) crossing 0 at ~pH 4.8, and, finally, increases to zero at ~pH 2. Results of this simulation are in qualitative agreement with the results of previous works in which ΔnH+ are treated as constants 7,9 and determined from the slope (δTM/δpH) .

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 152

H+ 0

 -1

-2

-3

-4 2 4 6 8 pH

Figure 5-3. The pH-dependence of the differential number of protons bound to the folded and unfolded states of the ODN calculated with Eq. (3).

Eq. (2) yields valuable insights into an increase in the midpoint of the pH-induced i-motif formation to ~pH 7 or even higher in the presence of cosolvents and crowders12, 20, 21. As an explanation, it has been proposed that the pKa of cytosines increases in the presence of cosolvents20 which, indeed, may contribute to the observed increase in the transition midpoint.

In fact, all three pKa‟s in Eq. (2) may be affected. As an alternative or supplemental explanation, we propose that, given the excluded volume effect, cosolvents and crowders may stabilize the compact i-motif state relative to the extended unfolded state with an ensuing increae in the sequilibrium constant K0. To illustrate this point, we used Eq. (2) to simulate the transition profiles of an i-motif for different values of K0. Specifically, we calculate the fraction folded, α

= K/(1 + K), as a function of pH. Figure 5-4 presents these profiles calculated for pKau, pKaf1, and pKaf2 of 4.8, 6.0, and 3.6, respectively, and K0 equal to 0.0001, 0.001, 0.005, 0.02, and 0.1.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 153

Inspection of Figure 4 reveals that a decrease in K0 below 0.005 leads to incomplete i-motif folding. At K0 equal to 0.005, the folding is nearly complete with α ~ 98 % at pH 4.8 and a transition midpoint at pH 5.7. A 20-fold increase in K0 from 0.005 to 0.1 (ΔΔG° = ~1.8 kcal mol-1) leads to a shift in the transition midpoint from pH 5.7 to 6.3. This observation readily offers an alternative rationalization to the effect of cosolvents and crowders on i-motif folding transitions.

1.0

K = 0.1 0 0.8 0.02 0.005 0.001

0.6 0.0001

0.4 FractionFolded 0.2

0.0

0 2 4 6 8 10 pH

Figure 5-4. Transition profiles simulated with Eq. (2) for pKau, pKaf1, and pKaf2 of 4.8, 6.0, and

3.6, respectively, and K0 equal to 0.0001 (dark yellow), 0.001 (green), 0.005 (blue), 0.02 (black), and 0.1 (red).

It should be noted that an increase in K0 may be caused not only by environmental factors but also by modifications in structure and sequence, e. g. the number of C·C+ base pairs or loop sequence and length2, 22. Clearly, such modulations of i-motif stability will be reflected in the

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 154

corresponding shift in the midpoint of the pH-induced folding/unfolding transition. As a final note, an increase in the number of C·C+ base pairs alone will cause an increase the degree of folding and shift the transition midpoint to a higher pH even in the absence of any change in K0. This notion is demonstrated in Figure 5-5 that plots the transition profiles of two i-motifs with 6 + (black) and 8 (red) C·C base pairs with the same pKa's of cytosines and the same equilibrium + constant, K0, of 0.0001. Inspection of Figure 5-5 reveals that the presence of two extra C·C base pairs leads to an increases in the maximum of the fraction folded, α, from 44 to 94 % while also shifting the folding transition midpoint to a higher pH.

1.0

N = 12 cyt 0.8 N = 16 cyt

0.6

0.4 FractionFolded 0.2

0.0

0 2 4 6 8 10 pH

Figure 5-5. Transition profiles simulated with Eq. (2) for i-motifs containing 6 (black) and 8 (red) + C·C base pairs with pKau, pKaf1, and pKaf2 of 4.8, 6.0, and 3.6, respectively, and K0 of 0.0001.

In summary, the pH-induced folding/unfolding transitions of i-motifs can be effectively described by a linkage thermodynamics-based formalism in terms of three cytosine pKa‟s - the

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 155

apparent pKa of cytosines in the unfolded conformation, pKau, and two apparent pKa‟s of cytosines in the folded conformation, pKaf1 and pKaf2. For a specific case of the 5′- TTACCCACCCTACCCACCCTCA-3' sequence from the human c-MYC oncogene promoter region, the values of pKau, pKaf1, and pKaf2 are 4.8, 6.0, and 3.6, respectively. These pKa‟s enable one to calculate the differential number of protons associated with the folded and unfolded states as a function of pH. Analysis along these lines offers an alternative interpretation to the experimentally observed shift in the pH-induced unfolded-to-i-motif transitions to neutral pH in the presence of cosolvents and crowders. Our simulations reveal that a significant increase in the transition midpoint pH can be achieved by an increase in the equilibrium constant between the folded and unfolded DNA conformations due to the excluded volume effect.

5.3 Experimental Section

The oligomeric sequence 5′-TTACCCACCCTACCCACCCTCA-3' was purchased from Integrated DNA Technologies (Coralville, IA). Sodium chloride was obtained from Sigma- Aldrich Canada (Oakville, ON). EDTA (free acid) was purchased from Fisher Biotech (Fair Lawn, NJ). The DNA was initially dissolved in and dialyzed against an unbuffered solution containing 10 mM NaCl and 0.1 mM EDTA rather than a buffer to avoid complications related to the ionization-neutralization equilibria of the buffer in pH-titration experiments. Dialysis was carried out with 1000 Da molecular mass cut off Tube-O-Dialyzers from G Biosciences (St. Louis, MO). The concentrations of the oligonucleotide were determined spectrophotometrically −1 −1 at 25 °C with a molar extinction coefficient, ɛ260, of 191,500 M cm for the unfolded conformation. The latter was evaluated using a nearest neighbor procedure as described by Dr. Richard Owczarzy (http://www.owczarzy.net/extinctionDNA.htm). The UV light absorption spectral measurements were carried out with a Cary 300 Bio spectrophotometer (Varian Canada, Inc., Mississauga, ON).

The circular dichroism (CD) spectra of the DNA were recorded at 25 °C in a 1 mm path- length cuvette using an Aviv model 62 DS spectropolarimeter (Aviv Associates, Lakewood, NJ). CD spectroscopic measurements were used for conformational characterization of the ODN

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 156

under each solution condition employed in this study. CD pH titration experiments were performed by adding to the optical cell containing 0.25 mL of the ODN solution aliquots of HCl solution. The initial volume was delivered by a 1 mL Hamilton Syringe, while titrant solutions (50, 100, and 500 mM HCl) were added with 10 L Hamilton syringes (Hamilton Co. Reno, NV). The syringes were equipped with Chaney adapter that allowed a relative delivery accuracy of  0.1%. The pH values were measured separately with four-fold larger volumes of the ODN solutions and the added HCl aliquots using a VWR brand Benchtop model 8015 pH-meter equipped with an Accumet Ag/AgCl combination microelectrode. The absolute error in these pH measurements was ±0.01 pH unit.

5.4 Acknowledgements

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 157

5.5 Reference

1. H. A. Day, P. Pavlou and Z. A. Waller, i-motif DNA: Structure, stability and targeting with ligands, Bioorg. Med. Chem. 22 (2014) 4407-4418.

2. S. Benabou, A. Avino, R. Eritja, C. Gonzalez and R. Gargallo, Fundamental aspects of the nucleic acid i-motif structures, RSC Adv. 4 (2014) 26956-26980.

3. K. Gehring, J. L. Leroy and M. Gueron, A tetrameric DNA structure with protonated cytosine cytosine base pairs, Nature 363 (1993) 561-565.

4. J. L. Mergny, L. Lacroix, X. G. Han, J. L. Leroy and C. Helene, Intramolecuar folding of pyrimidine oligodeoxynucleotides into an i-DNA motif, J. Am. Chem. Soc. 117 (1995) 8887- 8898.

5. I. Berger, M. Egli and A. Rich, Inter-strand C-HO hydrogen bonds stabilizing four-stranded intercalated molecules: Stereoelectronic effects of O4' in cytosine-rich DNA, Proc. Natl. Acad. Sci. U. S. A 93 (1996) 12116-12121.

6. J. J. Christensen, J. H. Rytting and R. M. Izatt, Thermodynamics of proton dissociation in dilute aqueous solution. 8. pK, change in heat content, and change in entropy values for proton ionization from several pyrimidines and their nucleosides at 25 °C, J. Phys. Chem. 71 (1967) 2700-2705.

7. V. Mathur, A. Verma, S. Maiti and S. Chowdhury, Thermodynamics of i-tetraplex formation in the nuclease hypersensitive element of human c-myc promoter, Biochem. Biophys. Res. Commun. 320 (2004) 1220-1227.

8. N. Khan, A. Avino, R. Tauler, C. Gonzalez, R. Eritja and R. Gargallo, Solution equilibria of the i-motif-forming region upstream of the B-cell lymphoma-2 P1 promoter, Biochimie 89 (2007) 1562-1572.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 158

9. M. Kaushik, N. Suehl and L. A. Marky, Calorimetric unfolding of the bimolecular and i-

motif complexes of the human telomere complementary strand d(C3TA2)4, Biophys. Chem. 126 (2007) 154-164.

10. J. M. Dettler, R. Buscaglia, J. Cui, D. Cashman, M. Blynn and E. A. Lewis, Biophysical characterization of an ensemble of intramolecular i-motifs formed by the human c-MYC

NHE III1 P1 promoter mutant sequence, Biophys. J. 99 (2010) 561-567.

11. B. G. Kim, H. M. Evans, D. N. Dubins and T. V. Chalikian, Salt effects on the stability of a G-quadruplex from the human c-MYC promoter, Biochemistry 54 (2015) 3420-3430.

12. J. J. Cui, P. Waltman, V. H. Le and E. A. Lewis, The effect of molecular crowding on the stability of humans c-MYC promoter sequence i-motif at neutral pH, Molecules 18 (2013) 12751-12767.

13. J. Dai, A. Ambrus, L. H. Hurley and D. Yang, A direct and nondestructive approach to determine the folding structure of the i-motif DNA secondary structure by NMR, J. Am. Chem. Soc. 131 (2009) 6102-6104.

14. S. M. Reilly, D. F. Lyons, S. E. Wingate, R. T. Wright, J. J. Correia, D. M. Jameson and R.

M. Wadkins, Effect of interior loop length on the thermal stability and pKa of i-motif DNA, Biophys. J. 107 (2014) 1703-1711.

15. J. Choi and T. Majima, Conformational changes of non-B DNA, Chem. Soc. Rev. 40 (2011) 5893-5909.

16. H. Kanehara, M. Mizuguchi, K. Tajima, K. Kanaori and K. Makino, Spectroscopic evidence for the formation of four-stranded solution structure of oligodeoxycytidine phosphorothioate, Biochemistry 36 (1997) 1790-1797.

17. J. Wyman and S. J. Gill, Binding and Linkage. Functional Chemistry of Biological macromolecules. (University Science Books, Mill Valley, California, 1990).

18. C. Tanford, Protein denaturation, Adv. Protein Chem. 23 (1968) 121-282.

CHAPTER 5: Thermodynamic Linkage Analysis of Folding and Unfolding of i-motifs 159

19. C. Tanford, Protein denaturation. C. Theoretical models for the mechanism of denaturation, Adv. Protein Chem. 24 (1970) 1-95.

20. A .Rajendran, S. Nakano and N. Sugimoto, Molecular crowding of the cosolutes induces an intramolecular i-motif structure of triplet repeat DNA oligomers at neutral pH, Chem. Commun. 46 (2010) 1299-1301.

21. Y. P. Bhaysar-Jog, E. Van Dornshuld, T. A. Brooks, G. S. Tschumper and R. M. Wadkins, Epigenetic modification, dehydration, and molecular crowding effects on the thermodynamics of i-motif structure formation from c-rich DNA, Biochemistry 53 (2014) 1586-1594.

22. T. Fujii and N. Sugimoto, Loop nucleotides impact the stability of intrastrand i-motif structures at neutral pH, Phys. Chem. Chem. Phys. 17 (2015) 16719-16722.

CHAPTER 6: General Conclusion and Future Perspectives 162

CHAPTER 6: General Conclusion and Future Perspectives

6.1 Conclusions

As non-canonical DNA structures arise as potential therapeutic targets, understanding the factors that govern their stability and folding/unfolding transitions becomes increasingly important. G-quadruplex often occurs in important regulatory regions of the genome and acts to inhibit uncontrolled telomere elongation and oncogene expression. A unique characteristic that distinguishes a G-quadruplex from other DNA structures is that it requires specifically bound cations to be formed. Counterions of specific size become coordinated to negatively charged O6 atoms of the guanines within the central cavity of a G-quadruplex. Another tetraplex structure that is often studied in conjunction with G-quadruplexes is i-motif. In the genome, G- quadruplex-forming guanine-rich strands are complementary to strands rich in cytosine. This C- rich strand may form i-motif, a distinct structure that consists of hemi-protonated cytosine·cytosine+ base pairs that are, generally, formed under slightly acidic conditions. My doctoral thesis was devoted to understanding the thermodynamics of the stability of these tetraplex structures. Three projects of my thesis examined the salt effect in various G-quadruplex structures and one paper investigated the effect of pH on i-motif formation using the thermodynamic linkage analysis.

Chapters 2, 3, and 4 explore the influence of the polyelectrolyte effect on the stabilization of G-quadruplexes formed in telomeric and oncogene promoter regions of the genome. I have conducted a series of spectroscopic experiments aimed at characterizing the stability of various G-quadruplexes in solutions with varying concentrations of stabilizing (Na+ or K+) and non-stabilizing (Li+, TMA+, Cs+) counterions. The specific and non-specific ionic contributions to the thermal stability of G-quadruplexes were evaluated by separating the effect of the centrally bound ions from that of cations condensed around the DNA. The thermal stability of G-quadruplexes is known to increase with the concentration of stabilizing cations; however, the effects of specific and non-specific cation binding have not been separated1. We devised a method to discriminate between the effects of the stabilizing cations in the central cavity are counterions condensed in the vicinity of the DNA. The method used involved an initial

CHAPTER 6: General Conclusion and Future Perspectives 163

saturation of the G-quadruplex by the stabilizing ions (Na+ or K+) followed by addition of either stabilizing or non-stabilizing cations (Li+, Cs+, TMA+). Our data suggest that the predominant ionic contribution to G-quadruplex stability comes from the specifically bound Na+ or K+ ions and not from counterion condensation around the DNA. This is in contrast to the thermodynamics of duplex and T-rich triplex DNA where counterion condensation provides a major contribution to the conformational stability2, 3. Our results clarified the balance of forces governing the conformational preferences of guanine-rich sequences while also being of practical importance for controlled induction of G-quadruplexes in the genome.

Following the study of the salt effect in G-quadruplex structures, we focused on understanding the molecular basis of i-motif stability. pH plays an important role in modulating the i-motif stability. Hence we conducted pH-dependent titration studies with the c-MYC i-motif and performed a thermodynamic linkage analysis of the pH-induced folding and unfolding transitions.

CHAPTER 6: General Conclusion and Future Perspectives 164

6.2 Future Prospect

Though the work for my dissertation provided new insight about the environment- dependent stabilities of tetraplex structures, it had several limitations. First, our studies hinted at the generality of the differential effect of internally (stabilizing) and externally (destabilizing) bound cations in parallel G-quadruplexes. However, experiments must be conducted with more oligomers derived from other parts of the genome to test if the results are broadly applicable. The second limitation is that the projects examining G-quadruplexes and i-motif separately do not perfectly reflect the physiological environment. In the genome, G-quadruplex structures are formed in the presence of complementary strands (with the exception of telomeric 3'-overhangs of eukaryotes). G-quadruplexes in gene promoter regions are constrained by the duplex nature of genomic DNA. Thus, an important question is how secondary structures such as G-quadruplexes energetically compete with duplex DNA, in particular, since putative GC-rich sequences that form these structures are thermally and thermodynamically more stable than AT-rich regions. Importantly, limiting the environmental factors simply to changes in salt and pH does not provide a big picture because there are many other molecules in the cell that can modulate the DNA-water interactions.

Salt Effects in G-quadruplexes derived from other regions of DNA

We have investigated the polyelectrolyte effect in four different G-quadruplex oligomers. The results showed that non-stabilizing cations can have no effect (tel-22 sequence), slightly stabilizing (tel-26), or destabilizing (VEGF and c-MYC) effect on G-quadruplex structures. It is suspected that the differing effects stem from the fact that the conformations of the G- quadruplexes used are different. However, more studies on G-quadruplex with varying topology must be conducted to draw concrete conclusions about the generalizability of our experimental results obtained.

Duplex-Single strand-Tetraplex Equilibria

Sugimoto and collaborators have shown that G-quadruplexes compete with duplexes depending on the cations and pH values4. We are presently investigating the duplex-quadruplex

CHAPTER 6: General Conclusion and Future Perspectives 165

equilibria transitions using the complementary strands for a subset of our original G-rich sequences. We are modulating the environmental conditions including temperature, pH, and salt in order to observe how the equilibria between various conformations change as a function of solution conditions.

Effect of osmolytes and cosolvents

Characterizing the interactions of osmolytes and other cosolvents with proteins and nucleic acids represents an area of significant interest in our laboratory. This study is essential both for understanding the modulation of biological function and developing methods to stabilize specific conformational states of biopolymers. We have recently begun a cycle of experimental and theoretical works in which we extend volumetric measurements to studying the solvation properties of solutes in binary solvents. Our lab has applied this approach to characterizing the interactions of protein groups with urea and glycine betaine. We are extending this knowledge to nucleic acid structures, specifically to G-quadruplexes in both telomeric and non-telomeric sequences.

CHAPTER 6: General Conclusion and Future Perspectives 166

6.3 Reference

1. Lane, A. N., Chaires, J. B., Gray, R. D., Trent, J. O. (2008), Stability and kinetics of G- quadruplex structures, Nucleic Acids Res. 36, 5482–5515.

2. Manning, G. S. (1978), Molecular theory of polyelectrolyte solutions with applications to electrostatic properties of polynucleotides, Q. Rev. Biophys. 11, 179–246.

3. Record, M. T., Anderson, C. F., Lohman, T. M. (1978), Thermodynamic analysis of ion effects on binding and conformational equilibria of proteins and nucleic acids. Roles of ion association or release, screening, and ion effects on water activity, Q. Rev. Biophys.11, 103–178.

4. Li, W., Wu, P., Ohmichi, T., and Sugimoto, N. (2002) Characterization and thermodynamic properties of quadruplex/duplex competition. FEBS Letters, 526, 77-81.

APPENDIX A: Effect of Urea on G-quadruplex Stability 167

APPENDIX A: Effect of Urea on G-quadruplex Stability

“Reprinted from The Journal of Physical Chemistry B, 121, Lusine Aslanyan, Jordan Ko, Byul G. Kim, Ishkhan Vardanyan, Yeva B. Dalyan, and Tigran V. Chalikian. Effect of Urea on G- quadruplex Stability, Copyright (2017), with permission from American Chemical Society”.

Author„s Contribution: Byul G. Kim organized and analyzed the data, participated in the preparation of the manuscript, and trained the undergraduate student Jordan Ko to run the experiments.

APPENDIX A: Effect of Urea on G-quadruplex Stability 168

A.1 Abstract

G-quadruplexes represent a class of noncanonical nucleic acid structures implicated in transcriptional regulation, cellular function, and disease. An understanding of the forces involved in stabilization/destabilization of the G-quadruplex conformation relative to the duplex or single-stranded conformation is a key to elucidating the biological role of G-quadruplex-based genomic switches and the quest for therapeutic means for controlled induction or suppression of a G-quadruplex at selected genomic loci. Solute-solvent interactions provide a ubiquitous and, in many cases, the determining thermodynamic force in maintaining and modulating the stability of nucleic acids. These interactions involve water as well as water-soluble cosolvents that may be present in the solution or in the crowded environment in the cell. We present here the first quantitative investigation of the effect of urea, a destabilizing cosolvent, on the conformational preferences of a G-quadruplex formed by the telomeric d[A(G3T2A)3G3] sequence (Tel22). At 20 mM NaCl and room temperature, Tel22 undergoes a two-state urea-induced unfolding transition. An increase in salt mitigates the deleterious effect of urea on Tel22. The urea m- value of Tel22 normalized per change in solvent-accessible surface area, ΔSA, is similar to those for other DNA and RNA structures while being several-fold larger than that of proteins. Our results suggest that urea can be employed as an analytical tool in thermodynamic characterizations of G-quadruplexes in a manner similar to the use of urea in protein studies. We emphasize the need for further studies involving a larger selection of G-quadruplexes varying in sequence, topology (parallel, antiparallel, hybrid), and molecularity (monomolecular, bimolecular, tetramolecular) to outline the advantages and the limits of the use of urea in G- quadruplex studies. A deeper understanding of the effect of solvent and cosolvents on the differential stability of the G-quadruplex and duplex conformations is a step towards elucidation of the modulating influence of different types of cosolvents on duplex-G-quadruplex molecular switches triggering genomic events.

APPENDIX A: Effect of Urea on G-quadruplex Stability 169

A.2 Introduction

In a suitable environment, guanine-rich DNA and RNA sequences fold into one of the various G-quadruplex conformations in which guanines are collected into stacks of G-quartets 1-7. A G- quartet is a planar structure formed by guanine bases pairwise-interlinked by Hoogsteen hydrogen bonds and additionally stabilized by a mono- or divalent cation of a proper size coordinated to the O6 carbonyls of the guanines within the central cavity of the core of a G- quadruplex 1, 3-4, 8. Guanine-rich sequences potentially capable of folding into G-quadruplex have been identified in hundreds of thousand loci in the human genome including telomeres, centromeres, and promoter regions of oncogenes with clear connections between G-quadruplexes and cellular function and disease 2, 8-14. Human telomeric DNA consists of tandem repeats of the 5'-TTAGGG-3' sequence with a 3'-end single-stranded overhang of 100-200 nucleotides 7, 11, 15. The structural, kinetic, and thermodynamic properties of monomolecular G-quadruplexes containing four repeats of 5'-TTAGGG-3' have been extensively studied 5, 16-27. In the + atmosphere of Na ions, the 22-meric telomeric sequence d[A(G3T2A)3G3] (Tel22) forms a basket structure with antiparallel strands and a diagonal and two lateral loops 16. In an aqueous solution, the Tel22 G-quadruplex is fully formed at ~20 mM NaCl and becomes further stabilized with an increase in the concentration of Na+ ions 25.

Among all interactions governing the stability and conformational preferences of G- quadruplex-forming DNA sequences, solute-solvent interactions (solvation) stand out as a ubiquitous thermodynamic force prominently present in all major structural transitions involving nucleic acid structures (e. g., duplex-to-single strand) while also participating in fine tuning of more subtle transitions between conformational substates (e. g., A-to-B duplex transition or transitions between G-quadruplex topoisomers) 28-32. Solvation should be viewed to include not only solute-water interactions but also interactions between DNA and water-miscible cosolvents that may be present in solution. The conformational preferences of guanine-rich DNA sequences have been shown to be strongly modulated by water-soluble organic compounds 30-31, 33-35.

Urea is a quintessential, widely used cosolvent with a strongly denaturing action on both proteins and nucleic acids 36-37. It has been employed for over a century in protein studies as a

APPENDIX A: Effect of Urea on G-quadruplex Stability 170

means to disturb and direct folded/unfolded equilibria 38-39. Remarkable progress has been made in developing an understanding of the action of urea on protein stability at the molecular level 36, 40-41. Studies of the effect of urea on the stability of nucleic acid structures have been relatively scarce compared to proteins. While the influence of urea on the stability of DNA and RNA structures have been studied to some extent, quantitative investigations of the effect of urea on conformational equilibria involving G-quadruplexes are lacking 37, 42-51. The deficiency is unfortunate for a number of reasons. Firstly, G-quadruplexes are unusual nucleic acid structures sharing a number of common features with globular proteins; they are globular in shape, less charged compared to other nucleic acids, and compressible owing to the presence of the central cavity. Thus, urea may interact with G-quadruplexes in a manner that is distinct compared to canonical DNA and RNA structures. Secondly, given the polymorphism of G-quadruplexes, urea may prove to be an effective and subtle tool in modulating the equilibria between various G- quadruplex conformations. Thirdly, an understanding of urea-G-quadruplex interactions may form the basis for subsequent studies of molecular forces underlying the duplex-to-G-quadruplex and single strand-to-G-quadruplex conformational switches within the genome in the crowded environment of the cell. Finally, urea is a natural cosolvent present in large concentrations in various living organisms which makes its effect on G-quadruplexes of biological significance.

In this work, we explore the conformational states of Tel22 as a function of temperature and the concentrations of urea and sodium ions. Our results represent the first systematic investigation of the stability of a G-quadruplex within the temperature-urea-salt conformational space, while also producing the necessary experimental basis for further investigations of the preferential interactions of this important cosolvent with G-quadruplex-forming DNA sequences in their folded and unfolded states.

A.3 Materials and Methods

Materials. The 22-mer, d[A(G3T2A)3G3] (Tel22), containing four repeats of the human telomeric DNA sequence was synthesized and purified by Integrated DNA Technologies (Coralville, IA). Sodium chloride, phosphoric acid, and urea were purchased from Sigma-Aldrich Canada

APPENDIX A: Effect of Urea on G-quadruplex Stability 171

(Oakville, ON, Canada). EDTA (free acid) was purchased from Fisher Biotech (Fair Lawn, NJ, USA). These reagents were used without further purification. All solutions were prepared using doubly distilled water.

All measurements were performed in a pH 7.0 buffer containing 10 mM monosodium phosphate/disodium phosphate, 0.1 mM EDTA, NaCl between 20 and 1000 mM, and urea at concentrations between 0 and 8 M. Solutions of urea with concentrations of 1, 2, 4, 6, and 8 M were prepared by weighing urea and adding pre-estimated amounts of the buffer to achieve the desired molalities, m. The molar concentration, C, of urea was computed from its molal -1 concentration, m, using C = [1/(mρb) + ϕV/1000] , where ρb is the density of urea-less phosphate buffer and ϕV = 44.1 cm3mol-1 is the apparent molar volume of urea 52-53. Urea solutions were subsequently used as solvents for the DNA solutions.

The DNA was dissolved in and exhaustively dialyzed against a pH 7.0 phosphate buffer with NaCl and urea at desired concentrations. The dialysis was carried out in 1000 Da molecular weight cut-off Tube-O-Dialyzers from G Biosciences (St. Louis, MO, USA). To form the equilibrium structure, the DNA-containing solution was placed in a boiling water bath and allowed to cool to room temperature.

The concentration of the DNA was determined from the absorbance at 260 nm measured at 25 °C using a molar extinction coefficient of 228,500 M-1 cm-1 for the unfolded conformation. The latter was calculated using an additive nearest neighbor procedure described by Owczarzy and colleagues 54. For circular dichroism (CD) measurements and temperature-dependent UV light absorption measurements, the DNA concentrations were ~30 and 3 μM, respectively.

Circular Dichroism Spectroscopy. CD spectroscopy was used to probe the DNA conformation at each of the experimental conditions of this study. The CD spectra of Tel22 were recorded in a 1 mm path-length cuvette at 25 °C using an Aviv model 62 DS spectropolarimeter (Aviv Associates, Lakewood, NJ) or an Olis DSM 20 CD spectrophotometer (Olis, Inc., Bogart, GA).

APPENDIX A: Effect of Urea on G-quadruplex Stability 172

UV Melting Experiments. UV light absorption at 295 nm was measured as a function of temperature in a DNA sample contained in a 1 cm path-length cuvette. These measurements were performed using a Cary 300 Bio spectrophotometer (Varian Canada, Inc., Mississauga, Ontario, Canada) or a Lambda 800 spectrophotometer (Perkin Elmer, Inc., Waltham, MA). The temperature was changed at a rate of 1 °C per minute. The G-quadruplex-to-coil transition temperatures, TM, and the van't Hoff enthalpies, ΔHvH, were evaluated from the experimental UV melting profiles using standard procedures 55-59.

A.4 Results

Figure A-1 presents the CD spectra of Tel22 at 25 °C in the absence of urea at NaCl concentrations between 20 and 1000 mM NaCl. In the absence of urea, Tel22 exhibits a CD spectrum, characteristic of the antiparallel G-quadruplex conformation with a positive peak at 295 nm and a negative peak at 260 nm 5, 60-61. The CD spectrum does not change when the concentration of NaCl increases from 20 to 1000 mM.

4000

2000

-1

cm -1 0

20 mM , deg , M

] 50 mM 

[ 100 mM -2000 200 mM 400 mM 600 mM 800 mM 1000 mM -4000 200 220 240 260 280 300 320 , nm

APPENDIX A: Effect of Urea on G-quadruplex Stability 173

Figure A-1. The CD spectra of Tel22 at 25 °C in the absence of urea at various NaCl concentrations. The NaCl concentration (in mM) for each spectrum is given in the inset.

Figures A-2a, b, c, d, e, f, g, and h show the CD spectra of Tel22 at urea concentrations between 0 and 8 M at 20, 50, 100, 200, 400, 600, 800, and 1000 mM NaCl, respectively. As is seen from Figure A-2a, at 20 mM NaCl, Tel22 undergoes a urea-induced conformational transition as suggested by changes in the positions and the amplitudes of the original CD spectral peaks. However, an increase in salt leads to complete (at urea concentrations of up to 4 M) or partial (at 6 and 8 M urea) refolding of unfolded Tel22 (see Figures 2b to h).

APPENDIX A: Effect of Urea on G-quadruplex Stability 174

APPENDIX A: Effect of Urea on G-quadruplex Stability 175

APPENDIX A: Effect of Urea on G-quadruplex Stability 176

APPENDIX A: Effect of Urea on G-quadruplex Stability 177

Figure A-2. The CD spectra of Tel22 at 25 °C at various urea concentrations at 20 (panel a), 50 (panel b), 100 (panel c); 200 (panel d); 400 (panel e); 600 (panel f); 800 (panel g); and 1000 (panel h) mM NaCl. The urea concentrations (in M) for the spectra are given in the insets.

Figure A-3 compares the CD spectra of the urea-induced unfolded state (at 20 mM NaCl and 8 M urea) and the temperature-induced (at 20 mM NaCl and 95 °C) unfolded states of Tel22. The two states are clearly distinct with the urea-induced unfolded state being more structured than the heat-induced unfolded state.

APPENDIX A: Effect of Urea on G-quadruplex Stability 178

4000

1400

1200

1000 -1

cm 800 -1

600 , deg , M

257 400 ]

2000  [ 200

0 -1

-200 10 20 30 40 50 60 70 80 90 100

T, °C

cm -1

, deg , M

]

 [ -2000

-4000 200 220 240 260 280 300 320 , nm

Figure A-3. The CD spectra of the urea-induced unfolded state at 20 mM NaCl and 8 M urea (○) and temperature-induced unfolded states at 20 mM NaCl and 95 °C in the absence of urea (■) of Tel22. The inset presents the temperature dependence of the molar ellipticity at 257 nm of the urea-induced unfolded state at 20 mM NaCl and 8 M urea.

Figure A-4 shows a representative UV melting profile of Tel22 recorded at 295 nm in a solution containing 200 mM NaCl and 1 M urea. All UV melting profiles have been approximated by the two-state model of thermal denaturation:

-1 -1 A = AN(T) + [AD(T) - AN(T)]/[1 + exp(ΔHM(T - TM )/R)] (1)

where TM and ΔHM are the transition temperature and van‟t Hoff enthalpy, respectively; and

AN(T) and AD(T) are linear functions that approximate the pre- and post-denaturation baselines, respectively.

APPENDIX A: Effect of Urea on G-quadruplex Stability 179

0.28

0.26

0.24

0.22 295 A 0.20

0.18

0.16

0.14 0 20 40 60 80 100 T, °C

Figure A-4. Representative UV melting profile at 295 nm Tel22 at 200 mM NaCl and 1 M urea. Experimental data were approximated with Eq. (1) (solid line).

Figure A-5 plots TM as a function of urea at various NaCl concentrations. Inspection of

Figures A-5 reveals a steady decrease in TM as the concentration of urea increases, while TM increases with an increase in salt concentration.

APPENDIX A: Effect of Urea on G-quadruplex Stability 180

20 mM 50 mM 70 100 mM 200 mM 400 mM 60 600 mM 800 mM 1000 mM

, °C , M

T 40

10 0 2 4 6 8 10 [urea], M

Figure A-5. The G-quadruplex-to-coil transition temperature, TM, transition temperatures of Tel22 as a function of urea at various NaCl concentrations. The NaCl concentration (in mM) for each set of TM is given in the inset.

Figure 6 presents a compilation of van‟t Hoff transition enthalpies, ΔHM, plotted against respective transition temperatures, TM. Linear regression analysis of the data plotted in Figure 6 -1 -1 yields a slope, ΔΔHM/ΔTM, of 0.14±0.04 kcal mol K which may represent a change in heat capacity, ΔCP, associated with the heat-induced denaturation of Tel22. In fact, it has been reported that thermal unfolding of some G-quadruplexes is accompanied by an increase in heat capacity 62. On the other hand, our calorimetric measurements have not produced any significant changes in heat capacity accompanying thermal unfolding of the K+-stabilized Tel26 telomeric G-quadruplex in the hybrid-1 conformation 26.

APPENDIX A: Effect of Urea on G-quadruplex Stability 181

A.5 Discussion

Urea-induced unfolding of Tel22 is a two-state process. Inspection of Figures 2a to h reveals that, at each NaCl concentration employed in this study, an increase in urea leads to a conformational alteration of Tel22 as manifested in a decrease in the amplitude of the positive peak at 295 nm, a redshift and a significant decrease in the amplitude of the negative peak at 264 nm, and a redshift and a modest decrease in the amplitude of the positive peak at 246 nm. Comparative analysis of the CD spectra presented in Figures 2a to h reveals that the denaturing effect of urea weakens as the concentration of NaCl in solution increases. This observation clearly reflects an increase in G-quadruplex stability accompanying an increase in the concentration of stabilizing ions (in this case, Na+ ions) 5, 63-65. Thus, at increasing NaCl concentrations, increasingly higher concentrations of urea are required to unfold the Tel22 G- quadruplex.

Further inspection of Figures 2a to h reveals the presence of nearly isoelliptic points at 252 and 275 nm. This observation is suggestive of a two-state nature of the urea-induced unfolding transition of the Tel22 G-quadruplex. In contrast to urea-induced unfolding of the Tel22 G- quadruplex, its heat-induced unfolding is not a two-state event but involves intermediates 21-24. Given the two-state nature of the urea-induced transition, it may be treated within the framework of the linear extrapolation model (LEM) in a manner similar to that applied to urea-induced denaturation of proteins 66-68. Figure 7 presents the urea dependence of the molar ellipticity at 295 nm of Tel22 at 20 mM NaCl. The experimental points were approximated by the LEM- based two-state analytical model in which the differential free energy, ΔG°, of the folded and unfolded states is given by the relationship:

ΔG° = -RTlnK = ΔG0 – m[urea] (2)

APPENDIX A: Effect of Urea on G-quadruplex Stability 182

where K = α/(1 - α) is the folded/unfolded equilibrium constant; α is the fraction unfolded; ΔG0 is the differential free energy of the folded and unfolded states in the absence of urea; and m is the so-called “m-value”, a proportionality coefficient.

Fitting the data in Figure A-7 with Eq. (2) yields the values of ΔG0 and m of 2.5±0.4 kcal mol-1 and 0.57±0.09 kcal mol-1M-1, respectively. A more conventional way of determining the differential free energy, ΔG0, is based on the thermodynamic parameters (the melting temperature, TM, and enthalpy, ΔHM) of the UV melting profile; ΔG0 = ΔHM(1 – T/TM). At 20 mM NaCl and in the absence of urea, TM and ΔHM are equal to 49.6±0.1 °C and 38.1±0.6 kcal -1 -1 mol , respectively. With these numbers, we calculate at 25 °C a ΔG0 of 2.9±0.6 kcal mol in good agreement with the urea-unfolding-based value of 2.5±0.4 kcal mol-1. The agreement is significant and suggests that the LEM analysis of urea-induced unfolding profiles of G- quadruplexes provides an alternative way of characterizing G-quadruplex stability.

3000

2000

-1

cm -1

1000

, deg , M

295

]

 [ 0

-1000 0 2 4 6 8 10 [urea], M

Figure A-7. The dependence of the molar ellipticity of the Tel22 at 295 nm on the urea concentration at 20 mM NaCl. Experimental data were approximated with Eq. (2) (solid line).

APPENDIX A: Effect of Urea on G-quadruplex Stability 183

Our determined m-value of 0.57±0.09 kcal mol-1M-1 is within the range of similar values determined for other nucleic acid structures 37, 44-45. For example, the urea m-values for the formation of a number of 12-meric DNA and RNA duplexes range between 0.58 and 0.98 kcal mol-1M-1 37.

The nature of the urea-induced unfolded state. The CD spectrum of the urea-induced unfolded state of Tel22 at 20 mM NaCl and 8 M urea with a positive peak at 258 nm and a negative peak at 278 nm (see Figure A-2a) is similar to that the unfolded state Tel22 adopts in the atmosphere of tetrabutylammonium ions in the absence of stabilizing Na+ ions 25. Figure 3 compares the CD spectra of the urea-induced unfolded state of Tel22 at 20 mM NaCl and 8 M urea and the heat-induced unfolded state at 20 mM NaCl and 95 °C in the absence of urea. Inspection of Figure 3 reveals that the urea-induced unfolded state is more structured compared to the heat-induced state at 95 °C. The latter does not display the characteristic peaks at 258 and 278 nm. Significantly, an increase in temperature causes a gradual (non-cooperative) conversion of the CD spectrum of the urea-induced unfolded state of Tel22 into that of the heat-induced unfolded state. The inset in Figure 3 presents the temperature dependence of the molar ellipticity at 257 nm. Inspection of the inset reveals that the urea- and heat-induced unfolded states are not separated by a cooperative transition.

Given these results, we conclude that the denaturing effect of urea on Tel22 is milder compared to that of temperature. At room temperature, the urea-induced unfolded state of Tel22 displays a significant amount of residual base stacking. This result is in agreement with the data from the Record lab who, based on a comparative analysis of the experimental and predicted m- values for duplex DNA and RNA, have concluded that the separated single strands are 60 to 90 % stacked 37. A qualitatively similar inference has been drawn from comparing the UV absorbance spectra of single stranded DNA and the respective mononucleotides 69. In agreement with our data, it has been suggested that the degree of single strand unstacking and the related exposure of previously buried solvent-accessible surface area is essentially independent of urea concentration 37. However, the stacking decreases as the temperature increases 37. Incomplete denaturation of DNA in the presence of urea contrasts the denaturing action of urea on proteins. As a general

APPENDIX A: Effect of Urea on G-quadruplex Stability 184

rule, urea-induced unfolded states of proteins are more fully disordered compared to the heat- induced denatured states which may retain significant amounts of secondary structural elements 38-39.

Urea-salt phase diagram. Figure 8 presents the three-dimensional plot of the combined effect of NaCl and urea on the transition temperature, TM. The TM-surface shown in Figure A-8 separates the folded and unfolded conformations of Tel22. Inspection of Figure 8 along with Figure A-5 reveals two interesting features. Firstly, at all NaCl concentrations studied in this work, an increase in the concentration of urea brings about thermodynamic destabilization of the

G-quadruplex as reflected in a nearly linear decrease in TM (see Figure A-5). This observation is consistent with the reports of nonspecific detrimental effect of urea on the thermal stability of nucleic acid structures including an RNA pseudoknot and hairpin and polymeric DNA and RNA duplexes 43, 46, 48-49. Secondly, at each experimental urea concentration, an increase in the concentration of sodium ions leads to an increase in TM fully or partially offsetting the destabilizing influence of urea (see Figure A-5). While a similar behavior has been reported for RNA structures 43, the molecular nature of the stabilizing action of cations in a G-quadruplex and other (canonical) nucleic acid structures is very different 63-65. In G-quadruplexes, the stabilizing action of cations originates predominantly from their specific binding inside the central cavity, while the nonspecific external binding (counterion condensation) may be very slightly stabilizing, insignificant, or even strongly destabilizing 63-65. In contrast to G-quadruplexes, canonical nucleic acid structures, e. g. duplexes, are stabilized by counterion condensation 70-73.

APPENDIX A: Effect of Urea on G-quadruplex Stability 185

Figure A-8. The three-dimensional dependence of the transition temperature, TM, on the concentrations of NaCl and urea.

As mentioned above, transition enthalpy data shown in Figure A-6 exhibit a tendency to -1 -1 increase with temperature. The apparent slope, ΔΔHM/ΔTM, of 0.14±0.04 kcal mol K may represent a change in heat capacity, ΔCP, associated with the heat-induced denaturation of Tel22.

An alternative explanation may reflect the fact that ΔCP is markedly influenced by a change in DNA solvation following the unfolding transition 74-78. An increase in urea leads to increasing replacement of water by urea within the hydration shell of DNA with the resulting alteration of solute-solvent interactions. An alteration of the nature of solute-solvent interactions in the vicinity of the folded and unfolded states of Tel22 with an increase in urea concentration may result in the observed temperature dependence of the transition enthalpy, ΔHM, with an apparently positive ΔCP.

APPENDIX A: Effect of Urea on G-quadruplex Stability 186

45 -1

40 , kcal , mol

vH 35

H 

25 30 40 50 60 70 80 90

TM,°C

Figure A-6. Compilation of transition enthalpies, ΔHM, of Tel22 measured at various combinations of NaCl and urea concentrations plotted against corresponding transition temperatures, TM.

Interactions of urea with the folded and unfolded states of Tel22. The observed deleterious effect of urea on the stability of Tel22 is in line with the previous reports of urea destabilizing the secondary and tertiary structures of a wide range of DNA and RNA constructs as manifested, for 37, 43-45, 47-49 example, in a decrease in their melting temperatures, TM . The destabilizing effect of urea on nucleic acid structures is due to its favorable interactions with the rings and functional groups of nucleic acid bases which become exposed to solvent in the unfolded state 37. Significantly, urea interacts favorably with all components of nucleic acids with a local/bulk 37 microscopic partition coefficient, KP, ranging between 1.09 to 1.37 . If normalized per solvent- accessible surface area, urea interacts most favorably with the heterocyclic aromatic rings of all nucleic acid bases and with the methyl group of thymine 37. While this assessment is based on studies of small analogs of nucleic acids and has been applied to DNA and RNA duplex-to-

APPENDIX A: Effect of Urea on G-quadruplex Stability 187

single strand transitions 37, it is plausible to expect a similar molecular basis for the effect of urea on G-quadruplex stability.

A thermodynamically rigorous way of quantifying accumulation or exclusion of a cosolvent around a macromolecule in question is given by preferential interaction parameter Γ23 =

(δm3/δm2)T,P,μ3, where m2 and m3 are the molalities of the cosolvent and the macromolecule, 40, 79 respectively; while μ3 is the chemical potential of the cosolvent . The difference in preferential interaction parameter between the folded and unfolded macromolecular (protein or

DNA) states, ΔΓ23, is related to cosolvent-induced modulation of the folded/unfolded equilibrium constant, K, according to ΔΓ 23 = (δlnK/δlna3)T,P,m2, where a3 is the activity of the 79-82 cosolvent . The differential preferential interaction parameter, ΔΓ 23, is related to the transition-induced changes in excess numbers of solvent, Δn1, and cosolvent, Δn3, around the solute as follows:

ΔΓ 23 = Δn3 - (N3/N1)Δn1 (3)

where N1 and N3 are the numbers of mole of the principal solvent (water) and cosolvent in solution 40, 80, 83-86.

Urea is very slightly excluded from the surface of double-stranded DNA but significantly accumulated at the surface exposed in DNA and RNA unfolding, in particular, the amide-like surfaces of guanine, cytosine, and, especially, thymine and uracil bases 47. There are no data on the preferential interactions of urea with G-quadruplexes. Nonetheless, it might be expected that, by analogy, urea is also very slightly excluded (or very slightly accumulated) from the surface of the G-quadruplex conformation while being strongly accumulated around the unfolded state.

The m-value and differential interaction parameter, ΔΓ 23, are related via the following relationship 36:

APPENDIX A: Effect of Urea on G-quadruplex Stability 188

m ≈ RT ΔΓ 23/[urea] (4)

The urea m-value of a protein is proportional to a change in solvent-accessible surface area, 87 ΔSA, accompanying the unfolding transition . Comparison of Eqs. (3) and (4) reveals that the 2 ratio m/ΔSA correlates with and reflects the net cosolvent excess or depletion around 1 Å of the newly exposed solute surface. Hence, m/ΔSA reflects the complex of differential solute-water and solute-cosolvent interactions that determine the specific effect of a given cosolvent on the stability of the macromolecule in question. It is instructive to compare the ratio m/ΔSA for the unfolding of the Tel22 G-quadruplex with that determined for other DNA and RNA denaturation events. A change in solvent accessible surface area, ΔSA, accompanying the unfolding of the Na+-stabilized Tel22 G-quadruplex has been estimated to be 1230 Å 2 25. With this estimate and -1 -1 our determined m-value of 0.57±0.09 kcal mol M , we calculate the ratio m/ΔSA of 0.46 cal mol-1M-1Å -2.

Survey of the literature reveals a large variation of the reported values of m/ΔSA for DNA and 44 RNA unfolding reactions. Shelton et al. have estimated an average value of urea m/ΔSA of 0.099 cal mol-1M-1Å -2 for five self-complementary RNA duplexes of varying lengths and tRNAPhe. While this estimate is close to that observed for protein unfolding 87, it is much smaller than 0.46 cal mol-1M-1Å -2, our estimate for the unfolding of the Tel22 G-quadruplex. On the other hand, Lambert and Draper 45 have determined, for a number of RNA structures including a hairpin, the tar-tar* RNA kissing loop complex, the tetraloop-receptor (TLR) structural motif, and the aptamer domain of the adenine riboswitch (A-riboswich), higher values of urea m/ΔSA ranging between 0.48 (A-riboswitch) and 0.65 (hairpin) cal mol-1M-1Å -2 (close to our estimate for Tel22). Using the urea m-values determined for 12-meric DNA and RNA duplexes reported by Guinn et al. 37 and their calculated changes in solvent-accessible surface areas assuming 50 % stacking of the single stranded conformation, we evaluate the values of m/ΔSA ranging between -1 -1 -2 0.24 and 0.40 cal mol M Å (again, close to our estimate). Hence, the value of m/ΔSA we determined for the Tel22 G-quadruplex is within the range of similar values determined by Lambert and Draper 45 and Guinn et al. 37, while being larger than those reported by Shelton et al. 44 45 . As noted by Lambert and Draper , the smallness of the value of m/ΔSA reported by Shelton

APPENDIX A: Effect of Urea on G-quadruplex Stability 189

et al. 44 may be due to an unrealistically large estimate of the change in solvent-accessible surface area, ΔSA, used by these authors in their analysis (the unfolded state has been modeled by the fully unstacked conformation). The use of more realistic (smaller) values of ΔSA would bring m/ΔSA to higher values (similar to our estimate for Tel22). Given the similarity of urea 37, 45 m/ΔSA ratios determined for the Tel22 G-quadruplex and other DNA and RNA structures , we propose that the molecular basis of urea-induced destabilization of Tel22 is similar to that of other nucleic acid species.

A.6 Conclusions

We presented the first systematic investigation of the combined effect of urea, salt, and temperature on the stability of a G-quadruplex. We derived the urea-salt-temperature phase- diagram of a telomeric DNA G-quadruplex which, in the presence of Na+ ions, adopts the antiparallel conformation. At 20 mM NaCl and room temperature, urea fully denatures the Tel22 G-quadruplex. An increase in salt mitigates the deleterious effect of urea on Tel22. The determined urea m-value of Tel22 normalized per change in solvent-accessible surface area, ΔSA, is similar to those for other DNA and RNA structures while being several-fold larger than that of proteins. Taken together, our results suggest that urea can be used as an analytical tool in thermodynamic characterizations of G-quadruplexes in a manner similar to the use of urea in protein studies. Studies involving a larger selection of G-quadruplexes varying in sequence, topology (parallel, antiparallel, hybrid), and molecularity (monomolecular, bimolecular, tetramolecular) must be conducted to outline the advantages and the limits of the use of urea in G-quadruplex studies. In particular, a deeper understanding of the effect of solvent and cosolvents on the differential stability of the G-quadruplex and duplex conformations is a step towards elucidation of the modulating influence of different types of cosolvents on duplex-G- quadruplex molecular switches triggering genomic events.

APPENDIX A: Effect of Urea on G-quadruplex Stability 190

A.7 Acknowledgements

This work was supported by grant 203816-2012 from the Natural Sciences and Engineering Research Council of Canada (to T. V. C.) and grant 15T-1F048 from the State Committee of Sciences of the Republic of Armenia (to Y. B. D.).

APPENDIX A: Effect of Urea on G-quadruplex Stability 191

A.8 Reference

1. Keniry, M. A. Quadruplex structures in nucleic acids. Biopolymers 2000, 56, 123-146.

2. Shafer, R. H.; Smirnov, I. Biological aspects of DNA/RNA quadruplexes. Biopolymers 2000, 56, 209-227.

3. Davis, J. T. G-quartets 40 years later: from 5'-GMP to molecular biology and supramolecular chemistry. Angew. Chem. Int. Ed. Engl. 2004, 43, 668-98.

4. Burge, S.; Parkinson, G. N.; Hazel, P.; Todd, A. K.; Neidle, S. Quadruplex DNA: sequence, topology and structure. Nucleic Acids Res. 2006, 34, 5402-5415.

5. Lane, A. N.; Chaires, J. B.; Gray, R. D.; Trent, J. O. Stability and kinetics of G- quadruplex structures. Nucleic Acids Res. 2008, 36, 5482-5515.

6. Huppert, J. L. Four-stranded nucleic acids: structure, function and targeting of G- quadruplexes. Chem. Soc. Rev. 2008, 37, 1375-1384.

7. Phan, A. T. Human telomeric G-quadruplex: structures of DNA and RNA sequences. FEBS J. 2010, 277, 1107-17.

8. Huppert, J. L. Four-stranded DNA: cancer, gene regulation and drug development. Philos. Trans. Royal Soc. A 2007, 365, 2969-2984.

9. Hurley, L. H. DNA and its associated processes as targets for cancer therapy. Nat. Rev. Cancer 2002, 2, 188-200.

10. Oganesian, L.; Bryan, T. M. Physiological relevance of telomeric G-quadruplex formation: a potential drug target. Bioessays 2007, 29, 155-165.

11. De Cian, A.; Lacroix, L.; Douarre, C.; Temime-Smaali, N.; Trentesaux, C.; Riou, J. F.; Mergny, J. L. Targeting telomeres and telomerase. Biochimie 2008, 90, 131-155.

APPENDIX A: Effect of Urea on G-quadruplex Stability 192

12. Balasubramanian, S.; Hurley, L. H.; Neidle, S. Targeting G-quadruplexes in gene promoters: a novel anticancer strategy? Nature Rev. Drug Discov. 2011, 10, 261-275.

13. Maizels, N. G4-associated human diseases. EMBO Rep 2015, 16, 910-22.

14. Huppert, J. L.; Balasubramanian, S. Prevalence of quadruplexes in the human genome. Nucleic Acids Res. 2005, 33, 2908-2916.

15. Neidle, S.; Parkinson, G. N. The structure of telomeric DNA. Curr. Opin. Struct. Biol. 2003, 13, 275-283.

16. Wang, Y.; Patel, D. J. Solution structure of the human telomeric repeat d[AG3(T2AG3)3] G-tetraplex. Structure. 1993, 1, 263-282.

17. Parkinson, G. N.; Lee, M. P.; Neidle, S. Crystal structure of parallel quadruplexes from human telomeric DNA. Nature 2002, 417, 876-880.

18. Phan, A. T.; Patel, D. J. Two-repeat human telomeric d(TAGGGTTAGGGT) sequence forms interconverting parallel and antiparallel G-quadruplexes in solution: Distinct topologies, thermodynamic properties, and folding/unfolding kinetics. J. Am. Chem. Soc. 2003, 125, 15021-15027.

19. Li, J.; Correia, J. J.; Wang, L.; Trent, J. O.; Chaires, J. B. Not so crystal clear: the structure of the human telomere G-quadruplex in solution differs from that present in a crystal. Nucleic Acids Res. 2005, 33, 4649-4659.

20. Antonacci, C.; Chaires, J. B.; Sheardy, R. D. Biophysical characterization of the human

telomeric (TTAGGG)4 repeat in a potassium solution. Biochemistry 2007, 46, 4654-4660.

21. Buscaglia, R.; Gray, R. D.; Chaires, J. B. Thermodynamic characterization of human telomere quadruplex unfolding. Biopolymers 2013, 99, 1006-1018.

22. Gray, R. D.; Buscaglia, R.; Chaires, J. B. Populated intermediates in the thermal unfolding of the human telomeric quadruplex. J. Am. Chem. Soc. 2012, 134, 16834- 16844.

APPENDIX A: Effect of Urea on G-quadruplex Stability 193

23. Gray, R. D.; Trent, J. O.; Chaires, J. B. Folding and unfolding pathways of the human telomeric G-quadruplex. J. Mol. Biol. 2014, 426, 1629-1650.

24. Boncina, M.; Lah, J.; Prislan, I.; Vesnaver, G. Energetic basis of human telomeric DNA folding into G-quadruplex structures. J. Am. Chem. Soc. 2012, 134, 9657-9663.

25. Fan, H. Y.; Shek, Y. L.; Amiri, A.; Dubins, D. N.; Heerklotz, H.; Macgregor, R. B.; Chalikian, T. V. Volumetric characterization of sodium-induced G-quadruplex formation. J. Am. Chem. Soc. 2011, 133, 4518-4526.

26. Shek, Y. L.; Noudeh, G. D.; Nazari, M.; Heerklotz, H.; Abu-Ghazalah, R. M.; Dubins, D. N.; Chalikian, T. V. Folding thermodynamics of the hybrid-1 type intramolecular human telomeric G-quadruplex. Biopolymers 2014, 101, 216-227.

27. Boncina, M.; Vesnaver, G.; Chaires, J. B.; Lah, J. Unraveling the thermodynamics of the folding and interconversion of human telomere G-quadruplexes. Angew. Chem. Int. Ed. Engl. 2016, 55, 10340-4.

28. Chalikian, T. V.; Volker, J. Hydration of nucleic acids. In Wiley Encyclopedia of Chemical Biology, Begley, T. P., Ed. Wiley and Sons, Inc.: 2008.

29. Miyoshi, D.; Karimata, H.; Sugimoto, N. Hydration regulates thermodynamics of G- quadruplex formation under molecular crowding conditions. J. Am. Chem. Soc. 2006, 128, 7957-7963.

30. Nakano, S.; Yamaguchi, D.; Tateishi-Karimata, H.; Miyoshi, D.; Sugimoto, N. Hydration changes upon DNA folding studied by osmotic stress experiments. Biophys. J. 2012, 102, 2808-2817.

31. Nakano, S.; Miyoshi, D.; Sugimoto, N. Effects of molecular crowding on the structures, interactions, and functions of nucleic acids. Chem. Rev. 2014, 114, 2733-2758.

32. Saenger, W.; Hunter, W. N.; Kennard, O. DNA conformation is determined by economics in the hydration of phosphate groups. Nature 1986, 324, 385-388.

APPENDIX A: Effect of Urea on G-quadruplex Stability 194

33. Smirnov, I. V.; Shafer, R. H. Electrostatics dominate quadruplex stability. Biopolymers 2007, 85, 91-101.

34. Buscaglia, R.; Miller, M. C.; Dean, W. L.; Gray, R. D.; Lane, A. N.; Trent, J. O.; Chaires, J. B. Polyethylene glycol binding alters human telomere G-quadruplex structure by conformational selection. Nucleic Acids Res. 2013, 41, 7934-7946.

35. Marchand, A.; Ferreira, R.; Tateishi-Karimata, H.; Miyoshi, D.; Sugimoto, N.; Gabelica, V. Sequence and solvent effects on telomeric DNA bimolecular G-quadruplex folding kinetics. J. Phys. Chem. B 2013, 117, 12391-401.

36. Canchi, D. R.; Garcia, A. E. Cosolvent effects on protein stability. Annu. Rev. Phys. Chem. 2013, 64, 273-293.

37. Guinn, E. J.; Schwinefus, J. J.; Cha, H. K.; McDevitt, J. L.; Merker, W. E.; Ritzer, R.; Muth, G. W.; Engelsgjerd, S. W.; Mangold, K. E.; Thompson, P. J.; Kerins, M. J.; Record, M. T., Jr. Quantifying functional group interactions that determine urea effects on nucleic acid helix formation. J. Am. Chem. Soc. 2013, 135, 5828-5838.

38. Tanford, C. Protein denaturation. Adv. Protein Chem. 1968, 23, 121-282.

39. Tanford, C. Protein denaturation. C. Theoretical models for the mechanism of denaturation. Adv. Protein Chem. 1970, 24, 1-95.

40. Timasheff, S. N. Control of protein stability and reactions by weakly interacting cosolvents: the simplicity of the complicated. Adv. Protein Chem. 1998, 51, 355-432.

41. Guinn, E. J.; Pegram, L. M.; Capp, M. W.; Pollock, M. N.; Record, M. T., Jr. Quantifying why urea is a protein denaturant, whereas glycine betaine is a protein stabilizer. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 16932-16937.

42. Klump, H.; Burkart, W. Calorimetric measurements of the transition enthalpy of DNA in aqueous urea solutions. Biochim. Biophys. Acta 1977, 475, 601-604.

APPENDIX A: Effect of Urea on G-quadruplex Stability 195

43. Schwinefus, J. J.; Kuprian, M. J.; Lamppa, J. W.; Merker, W. E.; Dorn, K. N.; Muth, G. W. Human telomerase RNA pseudoknot and hairpin thermal stability with glycine betaine and urea: preferential interactions with RNA secondary and tertiary structures. Biochemistry 2007, 46, 9068-9079.

44. Shelton, V. M.; Sosnick, T. R.; Pan, T. Applicability of urea in the thermodynamic analysis of secondary and tertiary RNA folding. Biochemistry 1999, 38, 16831-16839.

45. Lambert, D.; Draper, D. E. Denaturation of RNA secondary and tertiary structure by urea: simple unfolded state models and free energy parameters account for measured m- values. Biochemistry 2012, 51, 9014-9026.

46. Nordstrom, L. J.; Clark, C. A.; Andersen, B.; Champlin, S. M.; Schwinefus, J. J. Effect of ethylene glycol, urea, and N-methylated glycines on DNA thermal stability: the role of DNA base pair composition and hydration. Biochemistry 2006, 45, 9604-9614.

47. Hong, J.; Capp, M. W.; Anderson, C. F.; Saecker, R. M.; Felitsky, D. J.; Anderson, M. W.; Record, M. T., Jr. Preferential interactions of glycine betaine and of urea with DNA: implications for DNA hydration and for effects of these solutes on DNA stability. Biochemistry 2004, 43, 14744-14758.

48. Aslanyan, V. M., Babayan, Y. S., Arutyunyan, S. G. Conformation and thermal stability of DNA in aqueous urea solution. Biophysics (USSR) 1984, 29, 410-414.

49. Babayan, Y. S. Conformation and thermostability of double-helical nucleic acids in aqueous solutions of urea. Molec. Biol. (USSR) 1988, 22, 1204-1210.

50. Lambert, D.; Draper, D. E. Effects of osmolytes on RNA secondary and tertiary structure stabilities and RNA-Mg2+ interactions. J. Mol. Biol. 2007, 370, 993-1005.

51. Holmstrom, E. D.; Dupuis, N. F.; Nesbitt, D. J. Kinetic and thermodynamic origins of osmolyte-influenced nucleic acid folding. J. Phys. Chem. B 2015, 119, 3687-96.

APPENDIX A: Effect of Urea on G-quadruplex Stability 196

52. Gucker, F. T.; Gage, F. W.; Moser, C. E. The densities of aqueous solutions of urea at 25 and 30 °C and the apparent molal volume of urea. J. Am. Chem. Soc. 1938, 60, 2582- 2588.

53. Lee, S. Y.; Chalikian, T. V. Volumetric properties of solvation in binary solvents. J. Phys. Chem. B 2009, 113, 2443-2450.

54. Tataurov, A. V.; You, Y.; Owczarzy, R. Predicting ultraviolet spectrum of single stranded and double stranded deoxyribonucleic acids. Biophys. Chem. 2008, 133, 66-70.

55. Marky, L. A.; Breslauer, K. J. Calculating thermodynamic data for transitions of any molecularity from equilibrium melting curves. Biopolymers 1987, 26, 1601-1620.

56. Breslauer, K. J. Extracting thermodynamic data from equilibrium melting curves for oligonucleotide order-disorder transitions. Methods Enzymol. 1995, 259, 221-242.

57. Rachwal, P. A.; Fox, K. R. Quadruplex melting. Methods 2007, 43, 291-301.

58. Mergny, J. L.; Phan, A. T.; Lacroix, L. Following G-quartet formation by UV- spectroscopy. FEBS Lett. 1998, 435, 74-78.

59. Mergny, J. L.; Lacroix, L. Analysis of thermal melting curves. Oligonucleotides 2003, 13, 515-537.

60. Vorlickova, M.; Kejnovska, I.; Sagi, J.; Renciuk, D.; Bednarova, K.; Motlova, J.; Kypr, J. Circular dichroism and guanine quadruplexes. Methods 2012, 57, 64-75.

61. Dapic, V.; Abdomerovic, V.; Marrington, R.; Peberdy, J.; Rodger, A.; Trent, J. O.; Bates, P. J. Biophysical and biological properties of quadruplex oligodeoxyribonucleotides. Nucleic Acids Res. 2003, 31, 2097-2107.

62. Majhi, P. R.; Qi, J. Y.; Tang, C. F.; Shafer, R. H. Heat capacity changes associated with guanine quadruplex formation: An isothermal titration calorimetry study. Biopolymers 2008, 89, 302-309.

APPENDIX A: Effect of Urea on G-quadruplex Stability 197

63. Kim, B. G.; Shek, Y. L.; Chalikian, T. V. Polyelectrolyte effects in G-quadruplexes. Biophys. Chem. 2013, 184, 95-100.

64. Kim, B. G.; Evans, H. M.; Dubins, D. N.; Chalikian, T. V. Effects of salt on the stability of a G-quadruplex from the human c-MYC promoter. Biochemistry 2015, 54, 3420-3430.

65. Kim, B. G.; Long, J.; Dubins, D. N.; Chalikian, T. V. Ionic effects on VEGF G- quadruplex stability. J. Phys. Chem. B 2016, 120, 4963-4971.

66. Pace, C. N.; Shaw, K. L. Linear extrapolation method of analyzing solvent denaturation curves. Proteins 2000, Suppl. 4, 1-7.

67. Makhatadze, G. I. Thermodynamics of protein interactions with urea and guanidinium hydrochloride. J. Phys. Chem. B 1999, 103, 4781-4785.

68. Greene, R. F., Jr.; Pace, C. N. Urea and guanidine hydrochloride denaturation of ribonuclease, lysozyme, a-chymotrypsin, and b-lactoglobulin. J. Biol. Chem. 1974, 249, 5388-5393.

69. Holbrook, J. A.; Capp, M. W.; Saecker, R. M.; Record, M. T., Jr. Enthalpy and heat capacity changes for formation of an oligomeric DNA duplex: interpretation in terms of coupled processes of formation and association of single-stranded helices. Biochemistry 1999, 38, 8409-8422.

70. Manning, G. S. Molecular theory of polyelectrolyte solutions with applications to electrostatic properties of polynucleotides. Q. Rev. Biophys. 1978, 11, 179-246.

71. Record, M. T.; Anderson, C. F.; Lohman, T. M. Thermodynamic analysis of ion effects on binding and conformational equilibria of proteins and nucleic acids. Roles of ion association or release, screening, and ion effects on water activity. Q. Rev. Biophys. 1978, 11, 103-178.

72. Anderson, C. F.; Record, M. T. Polyelectrolyte theories and their applications to DNA. Annu. Rev. Phys. Chem. 1982, 33, 191-222.

APPENDIX A: Effect of Urea on G-quadruplex Stability 198

73. Anderson, C. F.; Record, M. T. Salt-nucleic acid interactions. Annu. Rev. Phys. Chem. 1995, 46, 657-700.

74. Chalikian, T. V.; Volker, J.; Plum, G. E.; Breslauer, K. J. A more unified picture for the thermodynamics of nucleic acid duplex melting: a characterization by calorimetric and volumetric techniques. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 7853-7858.

75. Madan, B.; Sharp, K. A. Hydration heat capacity of nucleic acid constituents determined from the random network model. Biophys J. 2001, 81, 1881-1887.

76. Tikhomirova, A.; Taulier, N.; Chalikian, T. V. Energetics of nucleic acid stability: the

effect of DCP. J. Am. Chem. Soc. 2004, 126, 16387-16394.

77. Mikulecky, P. J.; Feig, A. L. Heat capacity changes associated with nucleic acid folding. Biopolymers 2006, 82, 38-58.

78. Vaitiekunas, P.; Crane-Robinson, C.; Privalov, P. L. The energetic basis of the DNA double helix: a combined microcalorimetric approach. Nucleic Acids Res. 2015, 43, 8577-89.

79. Timasheff, S. N. The control of protein stability and association by weak interactions with water: how do solvents affect these processes? Annu. Rev. Biophys. Biomol. Struct. 1993, 22, 67-97.

80. Timasheff, S. N. Water as ligand: preferential binding and exclusion of denaturants in protein unfolding. Biochemistry 1992, 31, 9857-9864.

81. Record, M. T., Jr.; Zhang, W.; Anderson, C. F. Analysis of effects of salts and uncharged solutes on protein and nucleic acid equilibria and processes: a practical guide to recognizing and interpreting polyelectrolyte effects, Hofmeister effects, and osmotic effects of salts. Adv. Protein Chem. 1998, 51, 281-353.

82. Courtenay, E. S.; Capp, M. W.; Saecker, R. M.; Record, M. T., Jr. Thermodynamic analysis of interactions between denaturants and protein surface exposed on unfolding:

APPENDIX A: Effect of Urea on G-quadruplex Stability 199

interpretation of urea and guanidinium chloride m-values and their correlation with changes in accessible surface area (ASA) using preferential interaction coefficients and the local-bulk domain model. Proteins 2000, Suppl 4, 72-85.

83. Record, M. T., Jr.; Anderson, C. F. Interpretation of preferential interaction coefficients of nonelectrolytes and of electrolyte ions in terms of a two-domain model. Biophys. J. 1995, 68, 786-794.

84. Parsegian, V. A.; Rand, R. P.; Rau, D. C. Osmotic stress, crowding, preferential hydration, and binding: A comparison of perspectives. Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 3987-3992.

85. Parsegian, V. A.; Rand, R. P.; Rau, D. C. Macromolecules and water: probing with osmotic stress. Methods Enzymol. 1995, 259, 43-94.

86. Tanford, C. Extension of the theory of linked functions to incorporate the effects of protein hydration. J. Mol. Biol. 1969, 39, 539-544.

87. Myers, J. K.; Pace, C. N.; Scholtz, J. M. Denaturant m values and heat capacity changes: relation to changes in accessible surface areas of protein unfolding. Protein Sci. 1995, 4, 2138-2148.