Investigation Into Catalytic Metallodrugs That Target Hepatitis C IRES RNA

Investigation into Catalytic Metallodrugs that Target Hepatitis C IRES RNA:

Development, Characterization, and Mechanism

DISSERTATION

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

Martin James Ross

B.A. Chemistry

Graduate Program in Chemistry

The Ohio State University

2015

Dissertation Committee:

James A. Cowan (Advisor)

A. Douglas Kinghorn

Hannah Shafaat

Claudia Turro

Martin James Ross

2015

Abstract

Metals have been used for therapeutic purposes since the dawn of civilization including the ancient Egyptians using copper jars to sterilize their water. This use of metal and others were commonplace until the discovery of penicillin. With the discovery of penicillin, small molecules with a rational and target approach became the standard of the drug industry. Traditional drugs to this today mostly consist of organic compounds, composed primarily of carbon, hydrogen, nitrogen, oxygen, chlorine, and fluorine.

The discovery of cis-platin and the advancement of our understanding of how the body works, including how our bodies handle metals, and advance techniques have created the environment for the renaissance in interest and development of bioinorganic compounds for therapeutic use. Metal complexes offer unique opportunities and properties that traditional small molecules lack.

One such approach is through the development of catalytic metallodrugs which by design are able to recognize and target multiple of the same therapeutic target. This ability enable these compounds to be dose at lower dosage, sub-stoichiometric equivalents, which will lead to fewer off-targeting and side-effects. Limited preliminary studies have applied this approach towards ribonucleic acids, such as Hepatitis C IRES RNA.

Hepatitis C Virus (HCV) affects over 200 million people globally which unchecked can lead to cirrhosis or liver cancer. Unfortunately, there is not a vaccine available for HCV like hepatitis A or hepatitis B. The current approach towards treatment involve cocktails, mixtures of several compounds each with a different therapeutic target. Initial reports have demonstrated the activity of the Cu-GGHYrFK, copper peptide complex, in recognition of stem-loop IIb of the HCV IRES RNA.

This research starts by understanding the products and binding of Cu-GGHYrFK to stem-loop IIb (SLIIb). With the lead compound, pathways and mechanism for oxidative degradation of RNA were developed. Derivatives including the all D-configuration and all

L-configuration of this peptide were synthesized to examine the importance of stereochemistry on reactivity.

After this, a structure activity relationship study based upon the all L configuration was preformed to evaluate the role of each of the targeting domain amino acids on binding, reactivity, and cellular uptake. This was then continued to the first position after the Cu-

GGH domain to examine catalytic properties.

A series of different metal ions, Ni2+, Co3+, Pd2+, Pt2+, Au3+, were incorporated into

GGHYrFK, the lead compound, to investigate the importance of the metal ion with reactivity. A further more in-depth mechanistic study was carried out with Cu-GGHYrFK

18 with the use of heavy water, H2 O to determine the source of oxygen into products of RNA degradation. The novel 5’-product, 5’-geminal diol was reported.

iii

Finally, several catalytic drugs based on reported peptides that bind stem-loop IV of the HCV IRES RNA were developed and characterized. These compounds represent a paradigm shift in therapeutic approach for the treatment of hepatitis c virus.

Dedication

This dissertation is dedicated to Marty and Sue Ross, my parents, who always supported

every goal, taught me to dream, and instilled the passion of learning and to Keisha

Neidrich for her patience, support, and love.

Acknowledgments

This endeavor would not have been possible without many talented and wonderful individuals. At this time I would like to recognize a few.

I would initially would like to thank Dr. James Cowan, my advisor, for recognizing my potential, allowing the freedom to explore concepts and ideas while fostering the chemist in me.

To the Department of Chemistry and Biochemistry at the Ohio State for their continued financial support and access to some of the most talented researchers, educators, and instruments in the world.

Thank you is such an understatement for all the assistance from the former and current Cowan lab members that aided in this research; between the lively discussions to the social gatherings. Dr. Seth Bradford, thank you for being the one person that listened to all the ideas I had; whether they were crazy or far-fetch. Further, thank you for the opportunity to continue the work on the project you conceived and started. Dr. Jeff Joyner, thank you for the development of the MassDaddy and our heated debates in interpretation of data. Also thank you for your entertaining group meetings which have come to be

measured in Joyners. To Insiya Fidai for the countless hours of trouble-shooting, sorting, and eventually producing informative computation data. In addition to all the CD and ITC data in assistance in completing this.

vii

Vita

2009 to Current ...... The Ohio State University, Department of

Chemistry and Biochemistry

2009...... B.A. Chemistry, Hiram College

2005...... Graduated from Meadville Area Senior High

1986...... Born in Meadville, PA

Publications

Bradford S; Ross, M. J.; Fidai, I.; Cowan, J.A. “Insight into the recognition, binding, and reactivity of catalytic metallodrugs targeting stem loop IIb of hepatitis C IRES RNA.” ChemMedChem. 2014, 9, 1275-1285.

Fields of Study

Major Field: Chemistry viii

Table of Contents

Abstract ...... ii

Dedication ...... v

Acknowledgments...... vi

Vita ...... viii

List of Tables ...... xvi

List of Figures ...... xx

– Introduction ...... 1

– History of Metals in Medicine from Serendipity to Design ...... 3

1.3 – Cisplatin and derivatives ...... 7

1.4 – US FDA ...... 12

1.5 – Clinical and FDA-approved metallodrugs ...... 14

1.5.1 First Row Transition Metals ...... 14

1.5.2 Second and Third Row Transition Metals ...... 16

1.5.3 Other notable elements ...... 20

1.6 – Novel Approaches – Catalytic Metallodrugs ...... 22

1.6.1 Therapeutic Approach ...... 22

1.6.2 Design and Double Filter Effect ...... 25

1.6.3 Amino Terminal Copper Nickel Motif (ATCUN) ...... 28

1.6.4 Previous applications of the Cu-GGH motif ...... 29

Chapter 2: Insights into the Recognition, Binding and Reactivity of Catalytic

Metallodrugs Targeting Stem Loop IIb of Hepatitis C IRES RNA ...... 31

2.1 – Introduction ...... 31

2.2 – Materials ...... 33

2.3 – Methods ...... 34

2.3.1 Binding Constant Determination ...... 34

2.3.2 Melting Temperatures ...... 35

2.3.3 Molecular Dynamics ...... 35

2.3.4 Reaction Kinetics via Fluorescence...... 36

2.3.5 Mass Spectrometry...... 37

2.3.6 HCV Cellular Replicon Assay...... 38

2.3.7 RT-PCR...... 39

2.4 – Results ...... 39

2.4.1 Synthesis and Characterization...... 39

2.4.2 Metallopeptide Binding and Thermal Melts...... 40

2.4.3 RNA Cleavage Reactivity ...... 42

2.4.4 MALDI-TOF Mass Spectrometry...... 43

2.4.5 Molecular Modeling – Dynamic and Docking ...... 44

2.4.6 HCV Cellular Replicon Assay ...... 47

2.4.7 RT-PCR...... 49

2.5 – Discussion ...... 50

2.5.1 Comparison of Cu-GGHYrFK-amide and Cu-GGhyrfk-amide ...... 50

2.5.2 Binding and Thermal Melts ...... 51

2.5.3 Reactivity ...... 53

2.5.4 Sites of Reactivity ...... 56

2.5.5 Binding Model ...... 64

2.5.6 HCV Replicon Assays ...... 69

2.6 – Conclusion ...... 71

2.7 – Supplemental Material ...... 72

3.1 – Introduction ...... 82

3.2 – Materials and Methods ...... 83

3.2.1 Binding Constant Determination ...... 83

3.2.2 Complex Optimization ...... 84

3.2.3 In silico Complex docking ...... 84

3.2.4 Reaction Kinetics via Fluorescence...... 86

3.2.5 Reaction Rates Monitored by Gel Electrophoresis ...... 86

3.2.6 MALDI-TOF Mass Spectrometry...... 87

3.2.7 Cellular Uptake...... 88

3.2.8 Peptide Synthesis...... 89

3.3 – Results and Discussion ...... 89

3.3.1 Binding Constant Determination...... 89

3.3.2 In silico docking and inhibition constants ...... 93

3.3.3 Reaction Kinetics via Fluorescence ...... 96

3.3.4 Reactivity measured by fluorescence in gel assays...... 99

3.3.5 Reactivity and Mechanistic Insights...... 102

xii

3.3.6 Cellular uptake ...... 109

3.4 – Conclusions ...... 115

3.5 – Supplemental Figures and Tables ...... 116

Chapter 4: Binding and Initial Reactivity of Various Metal Ions Complexed to

GGHYrFK with an In-depth Characterization of Mechanism with GGHYrFK-Cu with

HCV SLIIb IRES RNA ...... 138

4.1 - Introduction ...... 138

4.2 – Materials ...... 139

4.3 – Synthesis...... 139

4.3.1 Complexation of Au (III), Pd (II), and Pd (II) with GGHYrFK ...... 139

4.3.2 Synthesis of Co (III) with GGHYrFK (1-Co) ...... 140

4.4 – Methods ...... 141

4.4.1 Quantification of stock complexes ...... 141

4.4.2 Circular Dichroism Titrations ...... 141

4.4.3 Isothermal Calorimetry ...... 141

4.4.4 Reaction monitored by agarose gels ...... 142

4.4.5 LC-MS of SLIIb with 1-Cu...... 142

4.5 – Results and Discussion ...... 143 xiii

4.5.1 Binding Assays ...... 143

4.5.2 Crystal Structure Trends ...... 146

4.5.3 Entropic and enthalpic contributions and binding comparision ...... 149

4.5.4 Reactivity...... 151

4.5.5 Mechanistic hypothesis of 1-Pt and 1-Au ...... 154

4.5.6 LC-MS of 1-Cu with SLIIb...... 156

4.5.7 Overhang analysis and mechanism pathway ...... 164

4.5.8 18O Incorporation ...... 166

4.6 – Conclusion ...... 169

4.7 – Supplemental Material ...... 171

Chapter 5: Catalytic Metallodrugs Based on the LaR2C Peptide Target HCV SLIV IRES

RNA ...... 195

5.1 – Introduction ...... 195

5.2 – Materials ...... 198

5.3 – Methods ...... 199

5.3.1 pUC19 Isolation ...... 199

5.3.2 Copper Peptides ...... 199

xiv

5.3.3 Binding via Fluorescence ...... 199

5.3.4 Reaction Kinetics via Fluorescence ...... 200

5.3.5 Kinetic Reactions via PAGE Sequencing Gel...... 200

5.3.6 pUC19 Kinetic Reactions via Gel ...... 201

5.3.7 HIV-RRE (Stem loop IIb) Kinetic Reactions by Gel...... 201

5.3.8 Mass Spectrometry...... 202

5.3.9 HCV Cellular Replicon Assay ...... 203

5.4 – Results and Discussion ...... 204

5.4.1 Binding and reactivity toward IRES SLIV ...... 204

5.4.2 Kinetics of 2-Cu and 3-Cu via PAGE ...... 208

5.4.3 Selectivity of 2-Cu and 3-Cu ...... 208

5.4.4 Mass Spectrometric Analysis of RNA Cleavage ...... 211

5.4.5 Cellular Replicon Activity ...... 220

5.5 – Conclusions ...... 222

5.6 – Supplemental Material ...... 223

References ...... 244

List of Tables

Table 1.1. Bulk, trace, and essential metals in biology ...... 2

Table 1.2. Excerpt from Rosenberg’s 1965 Nature Letters showing the impact of other transition metals complexes on cell viability...... 8

Table 2.1. HCV cellular replicon data for all D-amino acid analog of 1-Cu...... 48

Table 2.2. Michaelis-Menten parameters for degradation of SLIIb...... 55

Table 2.3. Subcluster assignments of 1-Cu based on proximity of the copper atom to the site of reactivity as determined by mass spectrometry...... 72

Table 2.4. Subcluster assignments of 2-Cu based on proximity of the copper atom to the site of reactivity as determined by mass spectrometry...... 74

Table 3.1. Summary of binding constant for peptides binding SLIIb...... 91

Table 3.2. Summary of in silico screening of complexes...... 95

Table 3.3. Pseudo Michaelis-Menten perimeters determined by fluorescence...... 97

Table 3.4. Initial Rates as determine by fluorescein-labelled SLIIb in agarose gel...... 101

Table 3.5. Summary of initial rates based on MALDI-TOF mass spectrometry for complexes 3-Cu through 7-Cu...... 104

Table 3.6. Description of overhangs and possible mechanisms...... 108

xvi

Table 3.7. Cellular Uptake Values ...... 110

Table 3.8. Mass list of expected products for RNA cleavage...... 116

Table 3.9. Expanded docking values showing the contribution of each cluster towards the average value reported in Table 3.2 ...... 122

Table 3.10. Matches at the 120 min time point for 3-Cu with SLIIb...... 126

Table 3.11. Matches at the 120 min time point for 4-Cu with SLIIb...... 127

Table 3.12. Matches at the 120 min time point for 5-Cu with SLIIb...... 128

Table 3.13. Matches at the 120 min time point for 6-Cu with SLIIb...... 129

Table 3.14. Matches at the 120 min time point for 7-Cu with SLIIb...... 130

Table 4.1. Dissociation constants as determined by circular dichroism and isothermal titration calorimetry...... 146

Table 4.2. Selected bond lengths and angles for crystal structures of Cu (II)-GGH, Pd

(II)-GGH, and Au (III)-GGH...... 148

Table 4.3. ITC derived values: binding affinities, enthalpic and entropic terms...... 152

Table 4.4. Initial Rate and consumption parameters from agarose gels ...... 154

Table 4.5. Example of a LC-MS product table ...... 160

Table 4.6. Example of data obtained for all the identified products for a particular time point by LC-MS analysis ...... 161

Table 4.7. Global analysis of products with DI water and 18O water...... 162

xvii

Table 4.8. Incorporation of 18O into determined products by overhang analysis through

LC-MS...... 162

Table 4.9. Oligo standards used for LC-MS...... 173

Table 4.10. Mass list of expected products of SLIIb...... 179

Table 4.11. Summary of products at each peak from the reaction of 1-Cu with SLIIb

18 under anaerobic conditions without O, H2O...... 184

Table 4.12. Summary of products at each peak from the reaction of 1-Cu with SLIIb

18 under anaerobic conditions with O, H2O...... 190

Table 5.1. Dissociation constants to HCV IRES SLIV for peptides based on LaR2C. 204

Table 5.2. Michaelis-Menten parameters for degradation of SLIV RNA...... 205

Table 5.3. Comparison of initial rate measurements obtained from fluorescence and

PAGE profiles ...... 209

Table 5.4. Selectivity of 2-Cu and 3-Cu as represented by initial rates as monitored from gels...... 209

Table 5.5. Relative initial rate for SLIV cleavage reactions mediated by 2-Cu and 3-Cu determined by MALDI-TOF experiments...... 218

Table 5.6. Initial rates for select overhang products observed following reactions promoted by 2-Cu and 3-Cu as monitored by time-dependent MALDI-TOF mass spectrometry ...... 219

Table 5.7. HCV cellular replicon activity for 2-Cu and 3-Cu...... 220

xviii

Table 5.8. MALDI-TOF peak global assignments for each time point from the 2-Cu promoted reaction with SLIV in the presences of co-reagents...... 231

Table 5.9. MALDI-TOF peak global assignments for each time point from the 3-Cu promoted reaction with SLIV in the presences of co-reagents...... 235

xix

List of Figures

Figure 1.1. Medicinal arsenic compounds at the turn of the 20th century...... 6

Figure 1.2. Cisplatin accumulation in the cells...... 9

Figure 1.3. Cisplatin interacting with d(GpGp) to form a 1,2-intrastrand crosslink...... 10

Figure 1.4. Currently approved platinate drugs in the US and around the world...... 11

Figure 1.5. First row transition metal currently approved by FDA or in clinical trials. .. 15

Figure 1.6. Second and third row transition metals currently FDA approved or in clinical trials...... 19

Figure 1.7. Additional metallotheraputics FDA approved or in clinical trials. A

Pentostam B. Glucantime C. Bismuth Subsalicylate D. Magnevist E. Dotarem F. Lithium

Carbonate ...... 22

Figure 1.8. Mechanism of traditional versus catalytic metallodrugs...... 24

Figure 1.9. Overview of a catalytic metallodrugs with three components: targeting domain, linker, and catalytic metal domain...... 26

Figure 1.10. Illustration of the double filter effect...... 27

Figure 1.11. Structure of the Cu-ATCUN motif...... 29

Figure 2.1. Structure of previously reported compound, 1-Cu ...... 32

Figure 2.2. Melting profiles with or without 1-Cu or 2-Cu...... 41 xx

Figure 2.3. Variation of shifts in IRES SLIIb melting temperature in the presence of peptides, relative to the SLIIb RNA alone ...... 41

Figure 2.4. The determination of KD for 2 via changes in tyrosine fluorescence ...... 42

Figure 2.5. Mass spectrometric analysis of RNA cleavage by copper-peptide complexes

...... 43

Figure 2.6. Summary of the rates of appearance of cleavage products at sites on SLIIb after reaction with 1-Cu (orange) and 2-Cu (blue)...... 44

Figure 2.7. The metal ion, shown as a sphere, for each bound metallopeptide conformer from docking to the top conformation of the NMR structure of SLIIb in Autodock ...... 47

Figure 2.8. RT-PCR results showing the preferential cleavage of HCV RNA (left) over ribosomal RNA (right) at increasing dosages of 1-Cu...... 49

Figure 2.9. Sites of reactivity based on MALDI assignments of fragments that show a time dependence in the presence of 1-Cu with SLIIb (left) and 2-Cu with SLIIb (right). 56

Figure 2.11. Time dependent formation of 2’, 3’-cyclic phosphates and 3’-phosphates as determined by MALDI-MS ...... 61

Figure 2.12. The computed solution structures of Cu-GGHYrFK-amide (left) and Cu-

GGhyrfk-amide (right) ...... 63

xxi

Figure 2.13. Comparison of the reactive sites based upon the MALDI-TOF MS (the colored residues) versus the simulated binding proposed by Autodock (orange spheres represent copper atoms) for 1-Cu (left), and 2-Cu (right)...... 66

Figure 2.14. Sites of interaction (yellow) based on modeling of 1-Cu with SLIIb (orange) and 2-Cu with SLIIb (blue)...... 68

Figure 2.15. Summary of cleavage sites on SLIIb RNA promoted by 1-Cu as determined by mass spectrometric measurements ...... 76

Figure 2.16. Summary of cleavage sites on SLIIb RNA promoted by 2-Cu as determined by mass spectrometric measurements ...... 77

Figure 2.17. Overlap of the 20 NMR structures available in the PDB file 1P5N ...... 78

Figure 2.18. Summary of cleavage sites on SLIIb RNA by 1-Cu as determined by mass spectrometry, showing the formation of products at each position as a function of time following reaction mediated by catalyst in the absence of co-reagents ...... 79

Figure 2.19. Summary of cleavage sites on SLIIb RNA by 2-Cu as determined by mass spectrometry, showing the formation of products at each position as a function of time following reaction mediated by catalyst in the absence of co-reagents...... 80

Figure 2.20. Relative amounts of each class of product observed by mass spectrometry displayed as a function of time following reaction with 1-Cu (top) and 2-Cu (bottom) .. 81

Figure 3.1. Binding affinity of 3-Cu to 2.5 M SLIIb...... 90

xxii

Figure 3.2. Positions of the top three copper atoms in the representative clusters for complexes 3-Cu through 7-Cu (spheres)...... 94

Figure 3.3. Concentration of 3-Cu versus initial rate of cleavage with 333 M SLIIb and

1 mM ascorbic acid and 1 mM hydrogen peroxide at pH 7.4...... 96

Figure 3.4. Turnover determination for 3-Cu, 4-Cu, 6-Cu, and 7-Cu...... 98

Figure 3.5. Time-dependence of 5-Cu with fluorescein-labelled SLIIb...... 100

Figure 3.7. Percent overhangs for predicted 3'-overhangs ...... 106

Figure 3.8. Percent overhangs for predicted 5'-overhangs ...... 107

Figure 3.9. Cellular Uptake for 3-Cu with time...... 112

Figure 3.10. Comparison of different conditions from cellular uptake assay ...... 113

Figure 3.11. Suggested model of cellular uptake and accumulation in the cell assuming an energy independent pathway ...... 114

Figure 3.12. CD titration profiles for 3-Cu through 13-Cu at selected wavelengths. .... 121

Figure 3.13. Pseudo Michaelis-Menten plots for 3-Cu through 13-Cu...... 123

Figure 3.14. Time dependent reactions monitoring the dissapearence of full length fluorescein labelled SLIIb with time...... 124

Figure 3.15. Plots of the concentration of Fl-SLIIb versus time (in minutes)...... 125

xxiii

Figure 3.16. Normalized intensity for 3-Cu with SLIIb at each position with time...... 131

Figure 3.17. Normalized intensity for 4-Cu with SLIIb at each position with time...... 132

Figure 3.18. Normalized intensity for 5-Cu with SLIIb at each position with time...... 133

Figure 3.19. Normalized intensity for 6-Cu with SLIIb at each position with time...... 134

Figure 3.20. Normalized intensity for 7-Cu with SLIIb at each position with time...... 135

Figure 3.21. Determination of relative initial rates (normalized intensity/min) from time- dependent MALDI-TOF MS reactions of nucleic acids with increasing amount of product detected...... 136

Figure 3.22. Cellular uptake of 3-Cu through 12-Cu with time...... 137

Figure 4.1. Binding curve with corresponding Mn+-GGHYrFK complex and SLIIb .... 145

Figure 4.2. Crystal structures of GGH-Mn+ ...... 147

Figure 4.3. A model of Mn+-GGH with highlighted atoms that are discussed in Table 4.2.

...... 149

Figure 4.4. ITC profile of 1-Cu into SLIIb ...... 153

Figure 4.5. Reactivity of 1-Cu towards Fl-SLIIb in the presence of ascorbate and hydrogen peroxide monitored up to 24 hours...... 153

Figure 4.6. Traces of LC-MS chromatogram of various reaction conditions of 1-Cu with

SLIIb...... 159

Figure 4.7. Comparison of output from LC-MS versus MALDI-TOF MS...... 163

Figure 4.8. Proposed formation of the 5'-aldehyde and 5'-gem diol...... 165

xxiv

Figure 4.9. Proposed mechanism of 4’-H abstraction to yield the 3’-phosphoglycolate 168

Figure 4.10. Proposed mechanism of 3'-H abstraction to yield the 3'- phosphoglycaldehyde ...... 169

Figure 4.11. ESI-MS of Mn+-GGHYrFK ...... 171

Figure 4.12. Quantification of 1-Mn+ complexes by UV/Vis monitoring the tyrosine absorbance...... 172

Figure 4.13. Control LC-MS chromatograms...... 174

Figure 4.14. CD spectra over the range of 350 nm to 200 nm, with increasing amount of

1-Mn+ added to 2.5 M SLIIb...... 175

Figure 4.15. ITC traces for 1-Mn+ complexes with SLIIb...... 176

Figure 4.16. Gel Assays of 1-Mn+ with fluorescein labelled SLIIb...... 177

Figure 4.17. Plots of concentration of fluorescein labelled SLIIb versus time...... 178

Figure 5.1. The highly structured Internal Ribosome Entry Site (IRES) in the 5’ untranslated region of hepatitis C viral mRNA ...... 197

Figure 5.2. Structure of the metal binding ATCUN motif with targeting R groups that define the domains that recognize the GCAC sequences of SLIV ...... 197

Figure 5.3. NMR solution structure of the human La protein showing all 20 structures including the secondary elements superimposed ...... 205

Figure 5.4. Sample kinetic trace for the reaction of 7.5 µM 2-Cu with HCV IRES SLIV

...... 206

xxv

Figure 5.5. Pseudo Michaelis-Menten plots for reaction of 2-Cu (left) and 3-Cu (right) with HCV IRES SLIV...... 207

Figure 5.6. Variation of initial rate with [NaCl] for reaction of 3-Cu with HCV IRES

SLIV ...... 207

Figure 5.7. Time-dependence of MALDI-TOF chromatograms following reactions catalyzed by 2-Cu (top) and 3-Cu (bottom) with SLIV RNA in the presence of coreagents...... 210

Figure 5.8. Three and Two dimensional representations of SLIV RNA ...... 212

Figure 5.9. (Below) Mass spectrometric analysis of cleavage products for reactions of

HCV IRES SLIV mediated by 2-Cu (top) and 3-Cu (bottom) ...... 213

Figure 5.10. Relative amount of each class of product, following catalytic reaction with

2-Cu (top) and 3-Cu (bottom) ...... 217

Figure 5.11. Sample curve for binding of 3 to HCV IRES SLIV...... 223

Figure 5.12. Disappearance of full-length Fl-HCV-SLIV ...... 224

Figure 5.13. Reactivity of co-reagents with Fl-HIV-RRE with or without the presence of either 2-Cu or 3-Cu ...... 225

Figure 5.14. Reactivity of co-reagents with 10 M bp supercoiled pUC19 with or without the presence of either 2-Cu or 3-Cu ...... 226

Figure 5.15. The principle of the double-filter effect ...... 227

xxvi

Figure 5.16. Time dependence of MALDI-TOF chromatograms for reactions promoted by 2-Cu with Fl-SLIV in the presence of coreagents ...... 228

Figure 5.17. Time dependence of MALDI-TOF chromatograms for reactions promoted by 3-Cu with Fl-SLIV in the presence of coreagents ...... 229

Figure 5.18. Time dependence MALDI-TOF chromatogram of Fl-SLIV in the presence of coreagents ...... 230

Figure 5.19. Variation of overhang with time as determined for 2-Cu and HCV SLIV from time-dependent MALDI-MS analysis ...... 239

Figure 5.20. Variation in overhang with time as determined for 3-Cu and HCV SLIV from time-dependent MALDI-MS analysis ...... 240

Figure 5.21. Change in relative normalized intensity at each overhang with time as determined for 2-Cu and HCV SLIV ...... 241

Figure 5.22. Change in relative normalized intensity at each overhang with time as determined for 3-Cu and HCV SLIV ...... 242

Figure 5.23. Proposed mechanisms for nuclease activity ...... 243

xxvii

Chapter 1: Metals in Biology

– Introduction

Since the days of antiquity metals have been used for therapeutic reasons. Over

5,000 years ago Egyptians were using copper to sterilize drinking water.1 The Chinese and

Indian in 2500-2600 BC believed that gold had vitalizing power; power that could be used to cure impotency, epilepsy, and syphilis.2 Today most prescribed drugs do not contain metals. There are notable exceptions such as cisplatin and its derivatives, a platinum compound used to treat cancer tumors, and bismuth subsalicylate better known by the trade name Pepto-Bismol®

This does not imply the roles of metals are unimportant. Our bodies uses various metal ions in a variety of roles. Many enzymes incorporate a metal ion, Cu, Fe, Co, and/or

Zn to name a few to help facilitate their function which can generally be characterized as structural, transport, and/or catalysis.3 RNA and DNA incorporate alkaline and alkaline earth metals to stabilize the negatively charged phosphate backbone structure.3-4 Iron in hemoglobin is used to transport oxygen from the lungs through the blood and eventually to the cells. In ACE (angiotensin converting enzyme) and ACE II, two of the enzymes

associated with blood pressure regulation, zinc will be found at the active site which selectively and catalytically hydrolyzes the peptide bond of their substrates.5 A further listing of the relative concentrations and roles some of these metal ions have can be found in Table 1.1.

Table 1.1. Bulk, trace, and essential metals in biology Adapted from Wohrle and Kaneko in Metal Complexes and Metals in Macromolecules (2003).6

Concentrations Daily Element In Human Role in biology Required (mg) (mg) Calcium 106 800 Bones, teeth, muscle activity Nerve axon potential, osmotic Potassium 1.5 x 105 cell pressure Sodium 1x105 Nerve axon potential Magnesium 3 x 104 500 Nucleic acids Many Enzymes, respiratory Iron 4.5 x 103 15 proteins Hydrolytic enzymes, nucleic Zinc 2 x 103 12 acid synthesis Many enzymes, dioxygen Copper 100 3 transport Manganese 20 3 Enzyme activation Molybdenum 5 0.2 Many redox enzymes

Cobalt 1.5 0.3 Vitamin B12 Chromium 1 0.06 Glucose tolerance factor Vanadium <0.1 <0.1 Cholesterol metabolism Nickel <0.1 0.4 Ureases, Reduced hemopoiesis

– History of Metals in Medicine from Serendipity to Design

Since the dawn of civilization, society has used metals in medicine. The Chinese in

2500 BC were using gold for medical applications. The Egyptians utilized copper for sterilizing water. In 400 BC, mercury became medically used by the Greeks and Persians.

It was use in the treatment of trachoma, syphilis, and as a laxative.1, 7 It was applied either as a cream in the form of cinnabar or as metallic mercury being ingested with milk or wine.

In the 1600’s minerals of antimony, arsenic, and mercury were administered as a diuretic; work pioneered by Paracelsus.1 This was followed by Nicholas Culpepper an advocate of the gold elixir aurum potabile.1 Dr. William Robinson in the 1907 Critic and Guide discussed the usage of the antimony “bullet” with the distinct property of causing purging as often as swallowed and could be administered as many times after it was passed. Quoting the article the “Everlasting Pill:”

“This, as Dr. J. A. Paris says, was economy in right earnest, for a single pill

would serve a whole family during their lives and might be transmitted as

an heirloom to their posterity. We have heard of a lady, says the Doctor,

who, having swallowed one of these pills became seriously alarmed at its

not passing. "Madam," said her physician, "fear not. It has already passed

thru a hundred patients without any difficulty."”8

An intriguing thought is to realize there was little to no chemical or biochemical understanding as to how or why these metals were used. Their applications arose from observations. Further, little to no attention was given to the side effects of administration.

One of the more well-known side effects known today is the psychological effects of mercury.9 Hat makers were commonly using mercury in their designs and suffered from erythrism also known as “Mad hatter disease,” as is Lewis Carroll’s character the Mad

Hatter from the book Alice’s Adventures in Wonderland.10

One of the earliest drugs that entailed a rational approach was Paul Ehrlich’s

Salvarsan® (3-amino-4-hydroxyphenyl arsenic (III), Figure 1.1).11 Ehrlich’s goal was to create a “magic bullet” that would be selective towards bacterium and not the host organism. As a medical student, he was amazed that anilines and other recently developed synthetic dyes could be used to stain bacterium. It was from there he set to tackle the epidemic of the day, syphilis. Starting with atoxyl (Figure 1.1), a previously abandoned organoarsenic compound because of optic atrophy, his team of scientists developed hundreds of molecules with each one tested for the biological activity, toxicity, and distribution in infected rabbits.12 This was the forerunner of what is typically referred to as a structure activity relationship assay, SAR, a common practice in the pharmaceutical world today. Compound 606 ended up becoming Salvarsan® (derived from Latin for

“arsenic that saves”).13 Salvarsan® would become a global sensation in the medical world, being the most prescribed drug in its hay-day. It went on to replace the use of mercury salts

in a remarkably short time span following its discovery in 1909, to clinical use in 1910.11

Unfortunately, Salvarsan® was not without its side-effects. Some of which resulted from the disconnect between the lab and the doctors administering it. One of the issues was dissolving the compound. Ehlirch continued making derivatives of Salvarsan® until compound 914 became Neo-salvarsan®, a more water soluble form. Eventually

Salvarsan® and Neo-salvarsan®, would be replaced by the discovery of penicillin.

Figure 1.1. Medicinal arsenic compounds at the turn of the 20th century. Atoxyl was the compound that Ehrlich used as his starting compound. Compound 418 was the parent compound that went on to inspire compound 606, Salvarsan®. There was much debate over the actual structure of Salvarsan® and it was not until 2005 that it was determined that the structure in solution comprised mostly of three and five member rings of arsenic.14 In 1919, Jacobson and Heidelberger developed Tryparsamide™, which was marketed by May & Baker for the treatment of trypanosomiasis and was in general dual administered with Suramin which saw use well into the 1960’s.1

Ehrlich’s systematic approach created the pillars of medicinal chemistry and pharmaceutics, but the world of bioinorganic medicinal chemistry was forever changed with the serendipitous compound Rosenberg formed while studying the impact of electric fields on cell growth and division published in 1965.

1.3 – Cisplatin and derivatives

Cisplatin would later be identified as the active compound that prevented cell division and ultimately cellular death. Interestingly in Rosenberg’s 1965 publication, he had not yet determine the active species.15 He noticed that inoculation of the solution with

(NH3)2PtCl4 at concentrations of 8 ppm produced the fibrils, but was unable to determine at the time what the active species was. Rosenberg then went on to experiment with other metal complexes to test the impact that the complexes had (Table 1.2). Of importance were the questions that were then proposed at the end of the of his letter,

“What is the mechanism of action of these metal ions? Where is the locus

of action in the bacterial cell? How does the effect of these metal ions relate

to the actions of the other causative agents of filamentous growth – is there

a weak link that all operate on? Can these metal ions inhibit cell division in

other bacteria, or cells?

These questions guided the discovery of mechanism of what is arguably the most renowned inorganic drug, cisplatin, which gained FDA approval in 19781 for the treatment of cancer. The simple but robust molecule works through an ingenious mechanism (Figure

1.2). When a patient is dosed with cisplatin intravenously, the compound enters the patient’s blood which has a sodium chloride concentration high enough, ~100 mM, to keep the chlorides attached to the platinum. Following cellular uptake, the lower concentration of chloride, ~2-30 mM, shifts the equilibrium of the complex to dissociate the bound chlorides which are replaced by water molecules. As a result of the preference of the cytoplasm for charged species, the active drug is essentially trapped within the cell.16,17

This species then interacts with nuclear DNA. In particular, it crosslinks with guanosine at the N7 position, with 1, 2-intrastrand d(GpGp) crosslink as the major product (Figure 1.3).

This interaction causes the DNA phosphate backbone to kink approximately 40o and the helix to unwind ~13o, which inhibits DNA replication and transcription.

Table 1.2. Excerpt from Rosenberg’s 1965 Nature Letters showing the impact of other transition metals complexes on cell viability. Concentrations were held at 8 ppm for 2 hours.15

Caused bacterial death Caused no change Caused elongation

+ + + 2- CoCl2 [Co(NH3)6]Cl3 K , NH4 , H -- [PtCl4]

(NH4)IrCl6 K2Ir(NO3)6 (NH4)2PtBr6

NiCl2 [Ni(NH3)6]Cl2 (NH4)2PtI6

(NH4)2OsCl6 [Pt(en)3]Cl4 Cis and trans (NH4)2PdCl4 RhCl3 [Rh(en)2Cl2]NO3

[Rh(NH3)5Cl]Cl2 (NH4)3RhCl6

PdCl2 [Ru(NH3)4ClOH]Cl

Figure 1.2. Cisplatin accumulation in the cells. In the extracellular environment, the chloride concentration is sufficient enough to shift the equilibrium of the complex towards the dichloro species. Upon crossing the cell membrane, the concentration of chloride is lowered and the equilibrium now shifts towards the diaqua species. Once the diaqua species forms, it is no longer able to readily cross the cell membrane. In addition, the prodrug, cisplatin, is now in the activated form and can react with DNA.

Figure 1.3. Cisplatin interacting with d(GpGp) to form a 1, 2-intrastrand crosslink. In both cases, the Pt is coordinated to the N7 of both guanine bases. The crosslinking of the bases causes a distortion in the helix of the DNA of about 40o and an unwinding of approximately 13o which prevents DNA replication and transcription. Additional stability of the structure may arise through hydrogen bonding of the amines with the carbonyl oxygen. Structure is adapted from S. E. Sherman et al. Science18 and J. A. Cowan19.

Cisplatin is responsible for approval of over 700 additional platinum compounds to be approved by the United States Food and Drug Administration (FDA) for clinical trials since its inception.16 For all the advances that cisplatin offered, there is room for improvement. Derivatives were developed to reduce side effects, enhance the therapeutic index, as well as to combat cisplatin-resistant tumors. From these compounds arose carboplatin and oxaliplatin which have had great clinical success and are registered worldwide. Available in Japan is nedaplatin, heptaplatin in South Korea and in China lobaplatin (Figure 1.4). However, these have been the exceptions as the vast majority have been abandoned as result of efficacy, toxicity and/or solubility.

Figure 1.4. Currently approved platinate drugs in the US and around the world. Carboplatin and oxaliplatin are FDA approved. Nedaplatin is approved for use in Japan and is in FDA phase II trials. Heptaplatin is approved for use in South Korea and Lobaplat is approved in China and is in phase II trials in the US.

1.4 – US FDA

The FDA can trace its foundation back to President Lincoln in 1862 with the appointment of chemist Charles M. Wetherill to head the Chemical Division in the new

U.S. Department of Agriculture (USDA). Peter Collier became the successor of Wetherill as chief chemist in the USDA and began addressing the issue of food adulteration.

However, it was Collier’s successor, Harvey Wiley in 1883 that developed the division into the Bureau of Chemistry in 1901.20 Wiley began his career in the division concerned with the admixture of glucose and sugar syrup to commercial maple syrup. This was merely the start and in his 1890 report A Popular Treatise on the Extent and Character of Food

Adulteration, which stated that nearly every food was impure in some manner and went further to state that the extent of impurities or adulterations caused impairment to the consumer’s health, including death.21 However, it was not enough to cause legislative changes.

As therapeutics became available on the market, tragedies occurred as a result of side-effects and impurities in the production thus causing the need for action. In 1902, pharmaceutical companies producing the diphtheria vaccine distributed vaccines tainted with tetanus. This episode led to the death of children and teens and created a distrust not only in the vaccine but in all biological agents. This led to the passage in 1906 of the Pure

Food and Drug Act signed by President Theodore Roosevelt which had the nickname of the “Wiley Act.”21 This was the first time the Bureau of Chemistry was charged with

enforcement of certifying the purity of the nation’s food and drugs. In 1927, Congress authorized the formation of Food, Drug, and Insecticide Administration under the Bureau of Chemistry and in 1930 it became known as the Food and Drug Administration (FDA).20

However, this was not enough to prevent the next crisis. In the 1930’s, sulfanilamide was identified and became one of the first synthetically successful drugs for the treatment of streptococcal infections.

In the 1937, the S. E. Massengill Company of Bristol in attempts to produce an easier administered medication discovered sulfanilamide was easily dissolved in diethylene glycol.21 The mixture was welcomed for its sweeten taste and fragrance compared to the standard. However, Watkins, the chief scientist at Massengill, did not perform any toxicity assays in humans or animals for the new formulation, Elixir

Sulfanilamide. The new formulation caused deaths of over 100 children before it was finally pulled from the market.20 This event led to the passage of the 1938 Federal Food,

Drug, and Cosmetic Act. This enacted premarket reviews (clinical trials) of the safety of compounds, assigned tolerance levels for poisonous substances, authorized factory inspections and made all new drugs available to consumers only through prescription, instilling many of the foundations of the current FDA regulations today.21

1.5 – Clinical and FDA-approved metallodrugs

While much attention thus far has focused on the development of Pt approved metallodrugs, this section will span the periodic table and showcase the versatility of metal complexes compared to traditional organic drugs.

1.5.1 First Row Transition Metals

1.5.1.1 Titanium

While best known for its use in implants, Ti (IV) were the first metal compounds after platinum to enter clinical trials for cancer treatment.22 The issue that led to the abandonment of these trials were a result of instability in water.23 Additional research is on-going, trying to improve the stability of Ti(IV) complexes. Two such leads are titanocene (Figure 1.5 A)24 derivatives and titanium salan (Figure 1.5 B)25 complexes.

1.5.1.2 Vanadium

The compound BEOV, bis(2-ethyl-3-hydroxy-4-pyrone) oxovanadium(IV) (Figure

1.5 C) is in phase IIa trials for the treatment of diabetes. The active form is the dissociated

V(IV) form, which forms an irreversible bond with the insulin receptor.26

1.5.1.3 Iron

Iron is one of the most abundant essential transition metal elements at approximately 4.8 g per person. Currently in phase II for the treatment of malaria is a Fe(II) compound, Ferroquine™ (Figure 1.5 E), which is based on quinine. Available through

Hospira is the drug Nitropress®, sodium nitroprusside (Figure 1.5 D), which slowly releases nitric oxide which acts as a hypotensive agent (reducing blood pressure).27

1.5.1.4 Cobalt

Doxovir®, CTC-96 (Figure 1.5 F), is a Co(III) compound which has completed phase I trials for ophthalmic herpetic keratitis, which is the leading cause for corneal blindness28, and adenoviral conjunctivitis. It has also completed phase II trials for treatment of herpes labials (cold sores).22

Figure 1.5. First row transition metal currently approved by FDA or in clinical trials. A. Titanocene dichloride B. Titanium salan C. BEOV D. Nitropress® E. Ferroquine™ F. Doxovir®

1.5.2 Second and Third Row Transition Metals

1.5.2.1 Yttrium

Radionuclide 90Y compounds, usually bound by a DOTA chelating group coupled to some targeting domain (as seen with octreotide Figure 1.6 A), are used clinically for cancer treatment as a pure -emitter by conversion of a neutron to a proton by an ejection of an electron from the nucleus.22

1.5.2.2 Technetium

99mTc is a metastable man-made radioisotope generated by relative stable 99Mo that is a -emitter with a half-life of 6 h. Gamma-emission consists of high energy electron emission and as such does not change the atomic number. It is used in tens of millions of single photon emission computed tomography (SPECT) diagnostics each year. As with radioisotopes, their imaging effectiveness is directed by the targeting groups attached.

Currently approved diagnostics are Cardiolite® (Figure 1.6 B) and Neurolite® (Figure 1.6

C) for folate-receptor positive tumors, while 99mTc-MIP-1404 (Figure 1.6 D) is in clinical phase II trials for prostate cancer imaging.22

1.5.2.3 Ruthenium

NAMI-A (Figure 1.6 E) and KP1019 (Figure 1.6 F) are two complexes involving

Ru (III) that are in clinical trials and have cleared phase I trials for cancer treatment.

1.5.2.4 Silver

Silver has a long history as an antimicrobial. There are several different mechanisms that this activity is attributed to. First, silver (I) interaction of thiol groups and inactivation of enzymatic functions of proteins. Second, Ag+ causes the release of potassium.29 Third, it can bind to nucleic acids30 and fourth, generates intracellularly superoxide.31 Although silver compounds are not as common use as they were before the advent of penicillin, they are still available, including silver sulfadiazine (SSD, Figure 1.6

F) cream which is used a topical agent for burns.4 In fact, silver may be part of the solution for antibiotic resistance as it was recently shown to increase the antimicrobial efficiency by a thousand fold.32 A form of silver, Acticoat® absorbent, is currently in phase IV trials as a sterile coating for wound dressings.22

1.5.2.5 Gold

Gold in medicine appeared on the scene as the result of Robert Koch’s gold cyanide solutions to treat tuberculosis.20 Starting in the 1930’s sanocyrin, Myocrisin®, and

Solganol® were developed for the treatment of rheumatoid arthritis, which were replaced by auranofin (Ridaura®).33 Although, these compounds are not part of standard treatment for rheumatoid arthritis, there has been interest in using auranofin (Figure 1.6 G) for anti-

HIV and lung cancer.22 Another promising lead is the use of gold nanoparticles with conjugated targeting biomolecules. These nanoparticles have potential for the use in the field of photo thermal and imaging agents.34

1.5.2.6 Mercury

Much concern has been expressed over mercury in vaccines in recent years; especially over the possibility it leading to autism; none of which have been substantiated.

Organometallic mercurials have been used in topical disinfectants as well as preservatives for vaccines, such as thiomersal (Figure 1.6 H) and merbromin (Figure 1.6 I).

GlaxoSmithKline’s Pandemrix™ contains 5 g/0.5 mL35 which is far below the recommended maximum limit for the far more toxic methyl mercury which is 1.6 g/kg body weight per week.4

Figure 1.6. Second and third row transition metals currently FDA approved or in clinical trials. A. 90DOTA-Phe-Tyr-octreotide B. Cardiolite® C. Neurolite® D. 99mTc-MIP-1404 E. NMIA-A F. KP1019 G. Silver Sulfadiazine H. auranofin I. thiomersal J. merbromin

1.5.3 Other notable elements

1.5.3.1 Antimony

The drugs Pentostam® (Figure 1.7A) and Glucantime™ (Figure 1.7B) have been used for 60 years and are available for the treatment of leishmaniosis. Leishmaniosis is a parasitic infection that is passed through sand flies that damage the spleen and liver. The mechanism of action of these compounds are still unclear but it is believed to involve the reduction of Sb(V) to Sb(III). In the reduced form, Sb(III) acts as the drug and interacts with trypanothione reductase and inhibits its function, which is vital for parasite surivial.36

1.5.3.2 Bismuth

As a result of bismuth typically existing in vivo as Bismuth (III), it can occupy Fe

(III) and Zn (II) binding spots in proteins.37 Bismuth compounds have been used for 200 years as anti-microbial agents4. The most common known form of bismuth to the consumer is Pepto-Bismol™ (bismuth subsalicylate, Figure 1.7C) which is used for gastrointestinal disorders. Bi (III) strongly binds to proteins in the acidic environment of the stomach to create a protective layer of the stomach lining. Further, bismuth binding to proteins in acidic conditions causes an increase in local prostaglandin levels which increases bicarbonate production, which helps neutralize acid and allows ulcers to heal.38

1.5.3.3 Gadolinium

Gadolinium (III) with seven unpaired electrons and a slow electronic relaxation time makes this ion ideal for magnetic resonance imaging (MRI) as a contrast agent.

Gadolinium compounds were first approved for clinical use in 1988 as Magnevist® (Figure

1.7D) and Dotarem® (Figure 1.7E). These compounds are kinetically and thermodynamically stable, which is part of the reason why these can be safely injected in up to gram quantities.

1.5.3.4 Lithium

Lithium salts have been used in the treatment of metal disorders, in particular bipolar disorder, which has effected roughly 2 million Americans since the 1950’s. The most common form is administered as lithium carbonate (Figure 1.7F).22 Li+ is required to be in millimolar concentrations to compete with the concentration of sodium ions to make cells less electrically activated.4

Figure 1.7. Additional metallotheraputics FDA approved or in clinical trials. A Pentostam™ B. Glucantime® C. Bismuth subsalicylate D. Magnevist® E. Dotarem® F. Lithium Carbonate

1.6 – Novel Approaches – Catalytic Metallodrugs

The last section highlighted compounds that have already started clinical trials or are already approved for usage. In this section, catalytic metallodrugs and their properties will be discussed.

1.6.1 Therapeutic Approach

Catalytic metallodrugs have a different mode of action compared to traditional drugs. Traditional drugs act through reversible inhibition (Figure 1.8A), here the drug, usually a small molecule, binds to a target site, usually an enzyme active site, and blocks

the function of the enzyme. Eventually, this drug dissociates either as a result of a cellular response or degradation of the target or drug itself. The first step of catalytic drugs is consistent with that of the traditional drug; binding to a target. However, when a catalytic drug finds the appropriate target, the metal center associated with the catalytic drug induces an irreversible modification of the target. This modification may be anything from oxidizing an amino acid side chain in either the active site, which attenuates the activity or knocks it out completely, a surface amino acid which would cause an allosteric change in the enzyme keeping it locked in a certain position, or creating cross-linking between amino acids (cysteine’s and tyrosine’s predominately) to actually cleaving bonds of proteins or nucleic acids. The beauty of the system is that once the modification/cleavage occurs the binding affinity towards the target changes, ideally decreases, and the drug is then able to find another target and repeat the cycle (Figure 1.8B).

Figure 1.8. Mechanism of traditional versus catalytic metallodrugs. (A) Traditional drugs may either act as competitive inhibitors by binding to the enzyme or as uncompetitive inhibitors which bind to the enzyme-substrate complex. (B) Catalytic metallodrugs act initially like traditional drugs in the first step. This changes in the second step where modification of the target occurs. This is then followed by release of the catalyst and ultimately turnover of the drug as it finds a new target.

1.6.2 Design and Double Filter Effect

Figure 1.9 demonstrates the basic design of a catalytic metallodrugs. The drug consists of three parts: targeting domain, metal binding domain, and linker (optional). The targeting domain is what incorporates specificity into the drug. It is as important as the metal binding domain. If the targeting domain is not specific enough, it will have find additional binding sites of therapeutic and non-therapeutic targets. Further, the design of the drug has a way to deal with off-target binding. This is referred to as a double-filter effect. In short, for irreversible cleavage to occur two conditions must be met. First, the drug (via the targeting domain) must bind to a target. Second, the metal binding domain must be oriented with the target to perform efficient chemistry. Figure 1.10, illustrates this point. In the case, where only binding is occurring, the drug acts as a reversible inhibitor like that of a traditional drug design. This then brings up an additional advantage of catalytic drugs. Traditional drugs are given in rather high dosages to ensure binding of the drug to the target; recall that binding is driven by equilibrium. Catalytic drugs on the other hand do not need to be given in saturating dosages. By design, they can be given at substoichiometric equivalents as the binding is only half the criteria for action. This substoichiometric use also reduces off-target binding as well.

As important as the targeting domain is the catalytic metal domain. The metal binding domain must efficiently perform chemistry to the target. If the chemistry is too robust, it will have a greater level of off-target modifications. Similarly too slow, and the

drug will not be efficient enough for therapeutic relevance. This also speaks to the type of chemistry the catalyst performs. Ideally, the catalyst should not produce diffusible reactive oxygen or nitrogen species (ROS and RNS). As aforementioned, this method of chemistry would have a higher level of off-target chemistry occurring. The ideal mechanism, would be one that would have a metal-associated species responsible for cleavage. An example of this would be a metal oxo species in which the oxygen, when near a target, could perform hydrogen abstraction or act as Lewis base. This non-diffusible, but localized chemistry combines with the targeting domain to construct the ideal catalytic metallodrug. The linker group is used to tune the binding and chemistry to allow the metal binding domain flexibility/constraint in interaction with the target. Care must be implemented as not to disrupt the targeting domain by placing the metal binding domain in a position that impedes binding.

Figure 1.9. Overview of a catalytic metallodrugs with three components: targeting domain, linker, and catalytic metal domain.

Figure 1.10. Illustration of the double filter effect. (Left) When off-target binding occurs, the metal binding domain is out of position to perform irreversible chemistry and therefore acts as reversible inhibitor. (Right) The drug is now with the correct target and the catalytic metal domain is oriented properly to preform irreversible chemistry.

1.6.3 Amino Terminal Copper Nickel Motif (ATCUN)

The metal binding domain used in this research is the amino terminal copper nickel motif. This motif consists of a tripeptide of the form X-X-H, where X can be any amino acid except for proline. The most basic of ATCUN motifs is the Gly-Gly-His. Figure 1.11 depicts the crystal structure of this peptide bound to Cu(II).39 This binding domain is a naturally occurring motif found in several proteins, including human serum albumin40-42,

HSA, neuromedins C and K,43 human sperm protamine P2a, and histatins44.41, 45 Binding to metal ions, primarily Cu2+ and Ni2+, consists of the amino terminal nitrogen, two deprotonated amide backbones, and an imidazole nitrogen from the histidine. The

3+ 3+ relatively hard ligands and planar chelating group induces the stability of the Cu and Ni oxidation state. The Cu3+/2+ redox couple is defined as quasi-reversible transfer with a potential for Cu3+ reduction at 800 mV versus silver/silver chloride electrode.46 Further, there is no reduction of the Cu2+ to Cu1+ in this system. This is advantageous as Cu(I) would dissociate form the peptide and would cause additional problems. Outside of Cu (II) and

Ni (II), additional metal complexes of the ATCUN have been made with Co (II),47 Au

(III)48-49, Pd (II)48-49, Pt (II)49, and V(IV)(oxo)50. The primary focus of this research will focus on the Cu (II) derivative.

Figure 1.11. Structure of the Cu-ATCUN motif. . (Left) Crystal structure of Cu-Gly-Gly- His. (Right) Generic representation of the ATCUN motif. The R groups show that the peptide can be customized to contain additional side group other than hydrogens, as is the case for glycine.

1.6.4 Previous applications of the Cu-GGH motif

Previous work has utilized Cu-GGH derivatives for a myriad of targets including proteins, deoxynucleic acids, DNAs, and ribonucleic acids, RNAs. Most of the preliminary work with Cu-GGH has been towards DNA.51 Dervan and co-workers synthesized Cu-

GGH attached to the amino terminus of Hin recombinase, residue 139-190 (binds to DNA) and showed moderate cleavage.52 Long and et al. used Tyr-Ala-()-Orn-Gly-His. Orn is ornithine (2,5-Diaminopentanoic acid) that allows the “N-terminus” to be available by coupling through the sidechain amine. This compound also showed comparable activity.53

Additional work was shown by Jin et al with Lys-Gly-His.46 Additional work recently has been done by Joyner to explore the nuclease activity towards DNA.54 Targeting of proteins

with Cu-GGH complexes have also been studied by Gokhale55-57, Hocharoen58, Joyner59, and Fidai60.

The focus of the research presented here will focus on the cleavage of RNA through the use of Cu-ATCUN derivate. There is little literature on these compounds for work towards ribonucleic acid. Most of the work has come from the Cowan group headed by former graduate students of Jin, Chen, Bradford, and Joyner.47, 61-68 It is on this foundation that this work is built on.

Chapter 2: Insights into the Recognition, Binding and Reactivity of Catalytic Metallodrugs Targeting Stem Loop IIb of Hepatitis C IRES RNA

2.1 – Introduction

Previous work has shown that the tetrapeptide sequence YrFK-amide (where the arginine is in the D-conformation) binds to stem loop IIb (SLIIb) of the Internal Ribosomal

Entry Site (IRES) of Hepatitis C Virus (HCV) RNA69 and that incorporation of a metal- binding ATCUN40 (amino terminal copper and nickel binding, XXH, where X ≠ P) motif results in catalytic degradation of the target RNA under oxidative conditions.69 The SLIIb domain of IRES RNA has been shown to be important for the initiation of translation for

HCV derived proteins and represents a promising target for therapeutic intervention.70,71 In fact, the activity of the copper-peptide complex CuGGHYrFK-amide (1-Cu, Figure 2.1) was demonstrated in a cellular replicon assay and exhibited additive to slightly synergic activity when given in combination with recombinant interferon alpha-2b69, one of the current treatments for HCV infection.

Figure 2.1. Structure of previously reported compound, 1-Cu Note the arginine, r, is lowercase indicating that it is in the D-configuration.

The incorporation of D-amino acids into a peptide sequence is a common approach to improve the stability of a peptide toward digestion by proteases, a recurrent problem for the use of peptides in vivo.72,73 It also addresses a fundamentally interesting question regarding the impact of stereochemical inversion on binding, as well as the reactivity for the all D-form stereoisomer (Gly excluded), relative to the parent peptide, and provides additional insights into the mechanism of action and binding. Stereochemical change at the α-carbon alters the conformation of the peptide and therefore potentially affects its ability to interact with its target. A complex set of factors, including binding affinity, off- rate for metallopeptide release from the inactivated target, orientation of the metal relative to the site of reaction, reduction potential of the metal cofactor, and cellular uptake and stability, will all determine the efficacy of the metallopeptide complex. If conversion of the amino acids to the D-form negatively impacts any one of these factors it could be detrimental to its effectiveness as a therapeutic. However, there is extensive precedent for

retention of bioactivity for proteins and peptides following conversion to the D-amino acid form. Examples include HIV-1 protease where the D-form of the enzyme retained activity but with inverted specificity74, an peptide derived from the N-terminus of HIV-1 gp41 retained activity in the opposite configuration75, while the all-D form of the Rev peptide showed enhanced binding to its RRE RNA target76, in addition to the D-configuration form of channel-forming peptides that are used as antibiotics.77 In the case of an amyloid beta assembly inhibitor peptide, the observed activity also increased.78 The binding of the

YrFK-amide targeting peptide to the SLIIb of the HCV IRES was only recently reported and there is no structural information available on the mode of binding.69 This current report investigates the impact, following conversion of the peptide sequence GGHYrFK- amide to the corresponding sequence containing all D-amino acids, GGhyrfk-amide (2), on binding as well as solution and cellular reactivity. Computational analysis suggests a model for how this class of peptides interacts with and promotes cleavage of its target

RNA, and is supported by mass spectrometric analysis of product fragments and RNA melting profiles. Mass spectrometric experiments also support specific primary mechanisms for RNA cleavage by this class of metallodrugs.

2.2 – Materials

RNA was purchased from Dharmacon, part of Thermo Fisher Scientific (Lafayette,

CO). Peptides were purchased from Genemed Synthesis, Inc. (South San Francisco, CA).

The sequence for the IRES SLIIb RNA incorporated 5’-fluorescein-

GGCAGAAAGCGUCUAGCCAUGGCGUUAGUA UGCC-3’, for the IRES SLIV RNA was 5’-fluorescein-GGACCGUGCACCAUGAGCACGAAUCC-3’, and for HIV Rev response element (RRE) RNA was 5’-fluorescein-UUGGUCUGGGCGCAGCGC

AAGCUGACGGUAC AGGCC-3’. The sequence of the 5-mer RNA used for calibration in the MALDI-TOF experiments was 5’-fluorescein(T)-UGUG-3’. All RNA was annealed by heating to 95 ºC and then cooling slowly to room temperature before use. Sodium chloride, sodium hydroxide, and acetonitrile were purchased from Fisher and HEPES, ammonium citrate, and 3-hydroxypicolinic acid were purchased from Sigma. C18 ZipTips were obtained from Millipore. All experiments were performed using diethyl pyrocarbonate (DEPC) treated water and autoclaved pipette tips and Eppendorf’s™.

Copper complexes form quickly and were made by mixing in a 1.1:1 peptide:CuCl2 ratio and waiting at least 5 min for complex to form. Structures were rendered in the

Visualization Applet for RNA (VARNA)79 for RNA secondary structures and the PyMOL

Molecular Graphics System, Version 1.5.0.4 (Schrödinger, LLC) or AutoDockTools, for three dimensional structures.

2.3 – Methods

2.3.1 Binding Constant Determination

For GGhyrfk-amide, RNA binding experiments were performed in the presence of

84 nM GGhyrfk-amide (2) in 20 mM HEPES (pH 7.4), 100 mM NaCl. Aliquots of unlabeled IRES SLIIb RNA were added and tyrosine emission was monitored (ex = 280

nm, em = 313 nm). Data were then fit to a one-site binding model using Origin 7.0 software.

2.3.2 Melting Temperatures

This experiment was carried out by Seth Bradford. Melting temperatures were measured by monitoring fluorescein emission (ex = 485 nm, em = 518 nm) with increasing temperature (15 to 95 degrees Celsius heated at a rate of 3 degrees Celsius per minute) in the presence or absence of metal-free peptide and/or varying [NaCl]. Melting temperatures were determined under conditions of 20 mM HEPES, pH 7.4, and 1 μM fluorescein-labeled

SLIIb IRES. Melting profiles were fit to a three-state consecutive melting model using

Origin 7.0 software to obtain TM1 and TM2. The values shown are an average of at least 3 trials.

2.3.3 Molecular Dynamics

Molecular Dynamics were performed in conjunction with Inisya Fidai. The peptide structures were initially optimized with Gaussian 09 vA01* using an Amber MM/MD force field in an aqueous solvent followed by a more thorough refinement with a DFT/B3LYP/3-

21G force field.80 Solvent interactions, water in this case, were considered by using the recent universal solvation model, SMD, by Truhlar and co-workers.81 The charge for the peptide structures was set to +3 with a multiplicity of 1. The solution state NMR structure of stem loop IIb (SLIIb) of the HCV IRES RNA is readily available from the protein data bank (PDB: 1P5N). The top solution state NMR structure (out of the 20 available) was

used for docking simulations using AutoDock 4.2. The entire SLIIb domain was considered for docking. In all cases, the complex was made flexible except for aromatic carbons, peptide bonds, and the metal binding domain. Docking analyses were performed on the UNIX system at the Ohio Supercomputer Center using the OAKLEY cluster platform with 12 CPU processors running on 1 computing node and a total wall time of

120 hours.

2.3.4 Reaction Kinetics via Fluorescence.

Reactions were performed by Seth Bradford. HCV IRES RNA cleavage was monitored in vitro by fluorescence using 5’ fluorescein end-labeled RNA with excitation and emission wavelengths of 485 nm and 518 nm, respectively. Reactions were carried out at 25 ºC in reaction volumes of 100 µL in the presence of 1 mM ascorbic acid and 1 mM H2O2 in HEPES buffer (pH = 7.4, 100 mM NaCl) with 1 μM fluorescein labeled IRES

SLIIb and analyzed according to the change in fluorescence observed as the reaction occurred. Both a time-dependence and a concentration-dependence of catalyst activity were observed. The initial velocity of the time dependence plot was used to generate the pseudo Michaelis-Menten plots which were then fit to the Michaelis-Menten equation. All fits were performed using Origin 7.0 software. The values shown are an average of at least three trials. A turnover number was estimated based on the limiting amount of peptide catalyst consumed by a specific amount of RNA.

2.3.5 Mass Spectrometry.

Reactions for MALDI-TOF analysis were run as described above, but using 10 µM fluorescein labeled IRES SLIIb and 10 µM copper-peptide incubated for up to two hours.

Reactions were then quenched by being placed on ice and ZipTipped. Zip Tip was performed using C18 Zip Tips from Millipore Co. in order to desalt the reaction mixtures prior to mass spectrometric analysis. Zip Tips were wetted with a 50:50 mixture of acetonitrile:water and equilibrated with 2 M triethylammonium acetate (TEAA), pH 7.0.

The reaction mixture was then bound to the Zip Tip, washed with nanopure water, and eluted with 50:50 acetonitrile:water. These samples were spotted onto a Bruker ground steel 96 target microScout plate by first spotting with 1 µL of a matrix solution containing

0.3 M 4-hydroxypicolinic acid (HPA) and 30 mM ammonium citrate in 30% acetonitrile, drying, spotting with 1 µL of a 2:1 RNA:matrix mixture, and allowed to dry. A calibration mixture containing 3 RNAs covering a range of molecular weights (Fl-5mer, Fl-IRES

SLIV, and Fl-RRE, with molecular weights of 2057.5, 8861.5, and 12172.5 amu, respectively) were used to calibrate the instrument. All MALDI-TOF MS analysis was performed on a Bruker MicroFlex LRF instrument equipped with a gridless reflectron, using negative ion mode and reflectron mode. The pulsed ion extraction time was 1200 ns. Typically, at least 1000 shots were summed per spectrum to acquire an accurate representation of the reaction. Data analysis was performed using Bruker flexAnalysis software. Assignment of peaks was performed by comparison of each peak list with the

expected masses for possible cleavage products.68 Only m/z values > 1500 amu and those with a signal to noise ratio greater than five were considered, since excessive spectral crowding occurred at lower m/z ranges. Peaks which showed time dependence were then compared to controls containing SLIIb RNA in the presence of copper-peptide. Reactions for the time dependent assay were collected at: 2 min, 10 min, 20 min, 30 min, 45 min, 60 min, 90, and 120 min. A comparison to controls of two hour reactions containing SLIIb

RNA with 1 mM ascorbic acid and/or 1 mM H2O2 as well as SLIIb RNA alone was used to identify new products of cleavage by the metallopeptide. Heat maps and initial rates were generated by first summing the total change at a position within the RNA sequence and determining the fraction of associated change at each time point. An apparent initial rate was then determined from the linear region of the data.

2.3.6 HCV Cellular Replicon Assay.

Samples were submitted to Dr. Zhuhui Huang and his staff at Southern Research

Institute. A stable cell line ET (luc-ubi-neo/ET) was employed in the assay. The ET is a

Huh7 human hepatoma cell line that contains an HCV RNA replicon with a stable luciferase (Luc) reporter and three cell culture-adaptive mutations. The HCV RNA replicon antiviral evaluation assay examined the effects of compounds at six half-log concentrations each. Human interferon alpha-2b was included in each run as a positive control compound. Sub-confluent cultures of the ET line were plated out into 96-well plates that were dedicated for the analysis of cell numbers (cytotoxicity) or antiviral

activity, and various concentrations of metallodrugs and controls were added to the appropriate wells the following day. Cells were processed 72 hours later when the cells were still sub-confluent. Six half-log serial dilutions of the compound were performed, and values derived for IC50 (the concentration that inhibited virus replication by 50%), TC50

(the concentration that lowered cell viability by 50%) and TI (the selectivity index:

TC50/IC50). HCV RNA replicon levels were assessed as the replicon-derived Luc activity.

The toxic concentration of drug that reduced cell numbers (cytotoxicity) was assessed by the CytoTox-1 cell proliferation colorimetric assay (Promega).

2.3.7 RT-PCR.

From the above replicon assays, medium was replenished and compound was added every 3 days for a total of 9 days. The cells were collected at 0, 3, 6 and 9 days for

RNA extraction and measurement of HCV RNA copies by qRT-PCR (TaqMan).82

Ribosomal RNA determined simultaneously using qRT-PCR was used for calibration of

HCV RNA reduction.

2.4 – Results

2.4.1 Synthesis and Characterization.

Peptides were monitored by tyrosine absorbance at 274 nm (ε = 1400 M-1cm-1).

Stock copper (II) chloride solutions were calibrated by titrating into a fixed amount of peptide at 37 oC every 5 min and fitting the absorbance at 240 nm and 525 nm.

2.4.2 Metallopeptide Binding and Thermal Melts.

It has been shown that the isolated stem loop IIb domain of the HCV IRES maintains the structure adopted in the full length RNA and is a useful probe for in vitro assays.83 Binding and cleavage assays were performed using 5'-fluorescein labeled SLIIb.

A dissociation constant (KD) of 76 nM for GGhyrfk-amide binding to SLIIb RNA was determined by monitoring the shift in the tyrosine emission of the peptide as a function of

RNA concentration under the conditions shown in Figure 2.4 (left). Next, the melting temperature of SLIIb was determined in the presence and absence of peptides and the results are shown in Figure 2.2. Melting profiles were fit to a three state consecutive model

(with an initial melt, TM1, of the lower stem; and a subsequent melt of the upper stem, TM2 as illustrated in Figure 2.2 (right)). Figure 2.2 (left) also shows fits for conditions of RNA alone, and in the presence of 50 μM GGHYrFK-amide or 50 μM GGhyrfk-amide. To further compare the binding profile for the two forms of the peptide, the salt dependence of thermal melts with and without the addition of 50 μM peptide was studied (Figure 2).

At high to intermediate salt concentrations, no noticeable shift in the melting temperature was observed. At lower salt concentrations, however, both of the peptides show an increase in TM1 on the order of 2 to 4 K, but only the D-amino acid form experienced a significant shift in TM2 of up to 18 K.

Figure 2.2. Melting profiles with or without 1-Cu or 2-Cu. (Left) Variation of fluorescein emission as a function of temperature. SLIIb alone (black), SLIIb plus 50 µM GGHYrFK- amide (1) (orange), SLIIb plus 50 µM GGhyrfk-amide (2) (blue). Fits are shown are for a three state consecutive melt with the proposed melts corresponding to TM1 and then TM2 (model shown on the right).

Figure 2.3. Variation of shifts in IRES SLIIb melting temperature in the presence of peptides, relative to the SLIIb RNA alone (indicated by the black line), SLIIb RNA and 50 µM GGHYrFK-amide (1) (orange), and SLIIb RNA and 50 µM GGhyrfk-amide (2) (blue). [RNA] = 1 µM; [peptide] = 50 µM, [HEPES] = 20 mM, pH = 7.4.

2.4.3 RNA Cleavage Reactivity

The reactivity of the metal complexes was determined by following the degradation of 5’-fluorescein labeled SLIIb by time-dependent fluorescence spectrophotometry. Initial velocities were plotted as a function of catalyst concentration to generate a pseudo-

Michaelis-Menten plot to generate KM and kcat values (Figure 2.4). A turnover number was estimated based on the limiting amount of peptide catalyst consumed by a specific amount of RNA. The turnover number for the all D-amino acid metallopeptide, 2-Cu, could not be directly determined because it exceeded the ratio of catalyst:RNA that was practical to use, but the turnover number was greater than 40 compared to the previously determined value of ~ 32 determined for 1-Cu.

Figure 2.4. The determination of KD for 2 via changes in tyrosine fluorescence; KD = 76 nM ± 3 nM. [2-Cu] = 84 nM, pseudo Michaelis-Menten profile for reactivity of Cu- GGhyrfk-amide with fluorescein labeled SLIIb RNA; KM = 7.9 µM, Vmax = 0.14 µM -1 RNA/min, kcat = 0.14 min .

2.4.4 MALDI-TOF Mass Spectrometry

MALDI-TOF analysis of RNA cleavage products is complicated by the presence of a large number of fragmentation peaks that are present even in the absence of catalyst

(Figure 2.5, left), and has been reported for other systems.84 The identification of cleavage products was determined by considering only those peaks that exhibited time dependence and were not observed in controls of either RNA alone, RNA in the presence of the coreagents (ascorbate and H2O2), or in the absence of catalyst. By use of these criteria it can be seen in Figure 2.5 (right) that the major product peaks are centered on m/z = 5100 and m/z = 6400. The new peaks and their assignments are shown in Tables 2.3 and 2.4.

Figure 2.6 shows the sites of cleavage observed based on the time dependence of a position in the RNA for 1-Cu and 2-Cu and how they map to the structure (based upon the data in

Figures 2.15 and 2.16 in the supplemental).

Figure 2.5. Mass spectrometric analysis of RNA cleavage by copper-peptide complexes RNA alone (black), and after reaction with 1-Cu (orange) or 2-Cu (blue). New product peaks are centered on m/z = 5100 and m/z = 6400 (right).

Figure 2.6. Summary of the rates of appearance of cleavage products at sites on SLIIb after reaction with 1-Cu (orange) and 2-Cu (blue).

2.4.5 Molecular Modeling – Dynamic and Docking

The peptide structures were initially optimized with Gaussian 09 vA01* using an

Amber MM/MD force field in an aqueous solvent, followed by a more thorough refinement with a DFT/B3LYP/3-21G force field.80 Solvent interactions, water in this case, were considered by using the recent universal solvation model, SMD, by Truhlar and co- workers.81 The charge for the peptide structures was set to +2 with a multiplicity of 1. The structures of the metal complexes were optimized by attaching the crystal structure for Cu-

GGH (inverting the stereocenter for the D-His in 2-Cu) to the optimized peptide and

resubmitting the complexes using the same force field and solvation model. The charge for the complex structures was set to +3 with a multiplicity of 1. The Cu-GGH atoms in the complexes were held fixed during the geometry optimization of the complexes by freezing the atoms in Gaussian 09. The optimized peptides and complexes were used for docking simulations.

The solution state NMR structure of stem loop IIb (SLIIb) of the HCV IRES RNA is readily available from the protein data bank (PDB: 1P5N). The top solution state NMR structure (out of the 20 available) was used for docking simulations using AutoDock 4.2.85

The entire SLIIb domain was considered for docking. The Cu-ATCUN sequence containing the targeting domain was simulated to investigate the site of localization of the peptide/complex on the RNA. In all cases, the complex was made flexible except for aromatic carbons, peptide bonds, and the metal binding domain. As a result of the high number of flexible bonds already in the complex, the RNA was restricted to its initial state.

Autodock does not have parameters for copper so an iron atom was substituted to mimic the geometry of the copper with a charge of +0.8. Iron and copper should have similar ionic radii and the iron simulates the charge on the copper ion. A value of +0.8 is used to simulate the +2 charge in order to compensate for the tendency of Autodock to overestimate electrostatic interactions.86 The geometry around the metal was based on the

X-ray crystal structure of Cu-GGH and was restricted to a square planar configuration.39 a

Lamarckian Genetic Algorithm and the following parameters were used for docking:

population size of 150, a random starting position and conformation, a maximal mutation of 2Å in translation and 50o in rotations (elitism of 5), iterations of Solis and Wets local search of 300, torsional degrees of freedom of 23 for the peptide complex, an external grid energy of 1000 kcal/mol, a mutation rate of 0.02 and a crossover rate of 0.8, and local search rate of 0.06. Simulations were performed with a maximum of 107 energy evaluations and a maximum of 27000 generations. The total number of hybrid GA-LS runs was set to

200, yielding an output of 200 clusters.

Figure 2.7 provides a summary of the results for docking of each of the complexes to the RNA. 1-Cu and 2-Cu do not bind at the same location. Cluster analysis was performed with an RMSD = 8 Å for the complex (Figure 2.7). Further analysis with a particular cluster was guided by positioning of the metal binding domain based on the experimental reaction data obtained from mass spectrometric (Table 2.4, 1-Cu; Table 2.5,

2-Cu, both available in the supplemental).

Figure 2.7. The metal ion, shown as a sphere, for each bound metallopeptide conformer from docking to the top conformation of the NMR structure of SLIIb in Autodock (Left box, left image) Positions of the docked metal ion for each conformation of the RNA- complexed metallopeptide 1-Cu. (Left box, right image) The subclusters of metal ions distributed around bases (individually color-coded) which correspond to experimentally observed sites of reaction for 1-Cu. (Right box, left image) Positions of the docked metal ion for each conformation of the RNA-complexed metallopeptide 2-Cu. (Right box, right image) The subclusters of metal ions distributed around bases which correspond to experimentally observed sites of reaction for 2-Cu.

2.4.6 HCV Cellular Replicon Assay

The activity of metallodrug 2-Cu was evaluated in cell culture by use of a cellular

HCV replicon assay that mimics the native HCV replication system as described previously. The results are shown in Table 2.1 and are compared to the data for the parent metallodrug 1-Cu. The diastereomer of 1-Cu showed activity similar to the parent metallopeptide (IC50=1.92 µM for 2-Cu versus an IC50 = 0.58 µM for 1-Cu). As in the case of 1-Cu, no activity was observed for controls of the RNA binding domain alone, Cu-GGH 47

alone, free copper, or the peptide without copper. Recombinant IFNα-2b was used as a positive control. The Cu-GGh complex was not tested, but is not expected to differ from the Cu-GGH response.

Table 2.1. HCV cellular replicon data for all D-amino acid analog of 1-Cu.

Antiviral Cytotoxicity Selectivity Compound IC50 (μM) TC50 (μM) Index (TI)

[GGHYrFK-amide]-Cu2+, 1-Cu 0.58 > 100 > 172

[GGhyrfk-amide]-Cu2+, 2-Cu 1.92 > 100 >52

YrFK-amide > 100 > 100 na

yrfk-amide > 100 > 100 na

[GGH]-Cu2+ > 100 > 100 na

Cu2+(aq) > 100 > 100 na na = not applicable

2.4.7 RT-PCR

To confirm the proposed mode of action (RNA degradation) for the metallopeptide in a cellular environment the disappearance of target RNA in cellular replicon assays by

RT-PCR experiments was demonstrated. These were performed to look at RNA levels for both HCV RNA and rRNA (as a control) in the replicon assays. It can be seen in Figure

2.8 that there is a preference for HCV RNA as the normalized copies of HCV RNA approach zero while still maintaining activity even up to 9 days. A small reduction in levels of rRNA is also observed.

Figure 2.8. RT-PCR results showing the preferential cleavage of HCV RNA (left) over ribosomal RNA (right) at increasing dosages of 1-Cu 0 µM (black), 2.5 µM (blue), 5 µM (green), 10 µM (orange), 20 µM (red). The graph on the left is normalized to the RNA levels in the absence of metallopeptide.

2.5 – Discussion

2.5.1 Comparison of Cu-GGHYrFK-amide and Cu-GGhyrfk-amide

The catalytic metallodrug candidates described herein are based on a metallopeptide design that incorporates well-characterized metal-binding domains with high stability combined with a targeting domain to provide specificity. The stability and mechanism of action of these compounds is consistent with the replicon assay data which displayed antiviral IC50 = 1.92 µM for 2-Cu compared to 0.58 µM for 1-Cu. This activity was only observed in the presence of all of the components of the metallodrug (copper ion, the metal-binding domain, and the RNA targeting tetrapeptide). Cellular toxicity was not observed up to the highest concentration tested (100 μM) and neither the individual components nor the metal-free peptide demonstrated activity. Previous work has also shown that incorporation of uptake sequences into the parent peptide does not significantly enhance activity suggesting that these peptides are already readily taken up by cells, and in fact are based on sequences known to facilitate cellular uptake.87,88

A common approach to improving the stability of a peptide in vivo is the incorporation of D-amino acids which are not recognized by proteases. Depending on the mechanism of action of the peptide, however, this inversion of chirality can be detrimental to its efficacy. This diastereomer can also be used to probe structure activity relationships and determine how these changes impact chemistry. For example, while an enhancement in binding might be expected to improve activity, it is also possible that a reduction in

binding could enhance release from the cleaved RNA and, if release of the catalyst is rate determining, actually improve activity. The in vitro and in vivo activity of the all D-amino acid version, CuGGhyrfk-amide of the previously described is reported and compared to

CuGGHYrFK-amide in terms of both binding and reactivity and RT-PCR data confirms a significant reduction in HCV RNA levels. The results provide further insights into the mechanism of this new class of compounds and suggest future routes for their optimization.

2.5.2 Binding and Thermal Melts

Sample melting profiles are shown in Figure 2.2 and melting temperatures were determined by measuring the change in fluorescence emission of fluorescein labeled RNA as a function of temperature. The resulting melting profiles were fit to a three state consecutive model for melting of the SLIIb secondary structure, consistent with initial melting of the double stranded region at the 3’- and 5’-ends prior to melting of the double- stranded region near the upper loop (Figure 2.2, left). This model also provided the best and most consistent fit to the data. Other fits that were tried were a two state model and a three state nonconsecutive model. Figure 2.3 shows that addition of peptide to the RNA caused the melting temperature to increase, consistent with the peptide-RNA complex being more stable than free RNA, and is consistent with the observation that these assays had to be performed at lower salt concentrations in order to see a noticeable shift in melting temperature where higher salt concentrations compete with the peptide for binding to the

negatively charged RNA. This suggests a mode of binding for 2, in particular, that is dominated by electrostatic interactions.

The original peptide GGHYrFK-amide exhibits surprisingly tight binding (KD = 44 nM) for such a small peptide. Therefore, it might be expected that all of the amino acid residues are important for binding. While there is no current structure of the peptide bound to the IRES SLIIb, it would be expected that the lysine and the arginine make electrostatic and hydrogen bond contacts with the negatively-charged RNA backbone. The tyrosine, phenylalanine, and/or arginine could also be involved in stacking interactions with the bases. The peptide backbone would be flexible but tighter binding could also be achieved by preordering the side chains, for example by π-stacking between the tyrosine and phenylalanine of YrFK-amide (and/or arginine) or by hydrogen bonding between functional groups.

Following inversion of stereochemistry at the α-carbon, the relative positioning of the amino acid side chains changes and intramolecular interactions would still be possible but with differences in orientation, as noted in the case of D-Rev peptide binding to stem-

76 loop IIb RRE RNA. Only a modest change in KD is observed following conversion to the all D-amino acid peptide (44 nM for 1 versus 76 nM for 2), consistent with flexibility in the side chains of lysine and arginine, which are more tolerant of changes than perturbations to hydrophobic/base interactions that would be more readily lost. This is also consistent with the effect of increasing [NaCl] on shifts in melting temperature for the IRES

SLIIb in the presence and absence of the two peptides as shown in Figure 2.3. As expected, both TM1 and TM2 for RNA alone increase with increasing [NaCl] due to stabilization of the negatively charged phosphate backbone by sodium binding. The original peptide

GGHYrFK-amide does not show a significant effect on TM2, and ΔTM2 ~1 K, presumably due to a balance of stabilizing electrostatic interactions and destabilizing interactions with the bases. If the binding of GGhyrfk-amide is more dominated by electrostatics, little or no shift would be observed at high [NaCl] due to the loss of electrostatic interactions with the target RNA and at lower [NaCl] there would be a stabilizing effect, consistent with the data shown in Figure 2.3. For both 1 and 2, there is an increase in TM1 at low salt concentrations (for 1 ΔTM1 ~ 2 to 4 K and for 2 ΔTM1 ~ 18 K). These differences are a reflection of subtle and ill-defined differences in binding of 1 versus 2, but significantly greater in the case of 2, and consistent with the dominance of electrostatic contacts.

2.5.3 Reactivity

Reaction kinetics were monitored by following the change in fluorescein emission over the course of the reaction as previously described for 1-Cu. Initial velocities were plotted as a function of CuGGhyrfk-amide concentration to generate a pseudo Michaelis-

-1 Menten plot (Figure 2.4). A kcat value of 0.14 min and a KM of 7.9 µM were determined

(Table 2.2). The parent peptide 1 demonstrated a KD (44 nM), much lower than its KM

(850 nM), which is consistent with a high affinity binding peptide where release of the

RNA from the peptide catalyst is most likely rate determining. The previously reported

data for 1-Cu was obtained under conditions of fixed catalyst concentration with varying

[RNA], whereas 2-Cu showed inconsistent behavior at high [RNA] and so was studied under pseudo Michaelis-Menten conditions although the values are directly comparable.

-1 The kcat value for 1-Cu is slightly greater than the value obtained for 2-Cu (0.53 min

-1 versus 0.14 min ) and the KM is about an order of magnitude higher (0.85 + 0.09 μM versus

7.9 + 1.2 μM), consistent with different binding modes that affect the chemistry without significantly changing the overall affinity. For 1-Cu, it was estimated by comparing the

KM and KD that k2 ~ 18 x k-1. In the case of 2-Cu, the k2 ~ 9 x k-1 and therefore k2 is also significant relative to k-1 and chemistry is fast relative to dissociation from the RNA. This is consistent with a system where catalyst release, not chemistry or initial binding, is determining the catalytic efficiency, kcat/KM, which is still significantly better for 1-Cu.

The kcat/KM isn’t the only parameter that is important for determining the activity of a catalytic metallodrug. A comparison of turnover numbers is also important. An exact turnover number could not be determined for 2-Cu except that it is greater than 40. This is higher than the turnover number of 32 previously reported for 1-Cu and provides an example of the complexity of factors that ultimately determine the efficacy of a catalytic metallodrug. A classical reversible inhibitor would show reduced efficacy with reduced binding, such as in the case of a mutation of the drug target that promotes resistance by the virus. The catalytic metallodrug approach, however, provides the possibility for the retention or enhancement of activity upon mutation of the target binding site. This emphasizes the fact that standard predictors for the efficacy of a drug, such as binding

affinity, are not necessarily the best predictors for the efficacy of a catalytic metallodrug. As long as specificity can be achieved, the catalyst can still be effective. Along with the targeting domain, specificity is also achieved by a double filter mechanism where both binding and positioning of the metal binding domain must be optimized for a given target.

Table 2.2. Michaelis-Menten parameters for degradation of SLIIb.

+ kcat KM kcat/KM KD Turnover Compound ref (min-1) (μM) (μM-1 min-1) (nM) Number 1-Cu 0.53 0.85 0.62 44 32 69 This 2-Cu 0.14 7.9 0.018 76 > 40 work

+ KD values listed are for the free peptide; the complex without the metal present

Figure 2.9. Sites of reactivity based on MALDI assignments of fragments that show a time dependence in the presence of 1-Cu with SLIIb (left) and 2-Cu with SLIIb (right). The scale reflects the relative initial velocity for reaction at each position with time (per minute) as described in the experimental section.

2.5.4 Sites of Reactivity

The mechanism and site of cleavage for 1-Cu was probed by MALDI-TOF mass spectrometry to map the sites of cleavage (Figures 2.6, Figure 2.15, and Figure 2.16 the latter two are in the supplementary). RNA alone, in the absence of reaction with catalytic metallodrug, showed fragmentation during the MALDI process (Figure 2.5). This has been shown previously for other RNA systems. Cleavage products from reaction with 1-Cu

were analyzed by considering those peaks that met strict requirements of having a signal to noise ratio greater than five and time dependence. These peaks were then compared to the predicted masses for expected products based on hydrolysis and known pathways for the oxidation of DNA. It should be noted that pathways for the oxidative cleavage of DNA are well characterized, but those for RNA oxidation are not.89 Correlations to the pathways for DNA can be made, however, even if hydrogen abstraction is expected to be more difficult for RNA than for DNA. Oxidation of RNA is most likely facilitated by the high oxidation state of copper (3+) and formation of intermediate reactive oxygen species, a characteristic of Cu-ATCUN complexes.68,54,90,91,92

Reaction of 1-Cu with isolated SLIIb showed reaction products primarily centered on m/z = 5100 and m/z = 6400 (Figure 2.5, right). The assignments for major new peaks, along with errors in ppm, are shown in Table 2.3. Overall trends show common residues for reaction (U14, A15, C18, A19, and G21) and these sites are mapped to the three dimensional structure of SLIIb and shown in Figure 2.9 (left). A recent paper performed selective 2’ hydroxyl acylation analyzed by primer extension (SHAPE) on the full length

HCV IRES and showed high accessibility at U14 and A15, consistent with the reactivity observed.93 Therefore, the observed preference for reaction in this area reflects both the placement and orientation of the metal ion, as well as the intrinsic reactivity and accessibility of these residues as shown in Figure 2.18 (supplemental material). Previous work by Kalliampakou et al. examined mutations of the apical region of SLIIb and showed

that single substitutions at the proposed residues (U14, A15, C18, and A19)94 cause dramatic reductions in IRES mediated translation, with a ~ 30-60% reduction for a single residue. This demonstrates the importance of these residues, but also suggests that activity for 1-Cu would be expected even if dissociation of the RNA strands does not occur after cleavage, and particularly in the context of the full length IRES and under physiological conditions where most of the hydrogen bonding interactions would remain intact. Clearly structural perturbations at the positions of those bases have a pronounced effect on translation, and cleavage at those positions would be expected to have a similar impact.

The sites where cleavage occurs for 2-Cu were further probed by MALDI-TOF mass spectrometry as described for 1-Cu (Table 2.4). Reactions of 2-Cu with SLIIb showed several similar reactive residues to those of 1-Cu in addition to a wider distribution of sites of reactivity (Figure 2.9). As mentioned for 1-Cu, the proposed sites of cleavage at the top loop have been shown to be important for translation. This is consistent with a system that occasionally binds to one of these apical residues similar to 1-Cu, at A15, A19, and G21, but as a result of the different stereochemistry around the α-carbons 2-Cu is less selective and reacts over a wider range of sites than 1-Cu, preferring residues between A6 and G9.

Figure 2.10. Under oxidative conditions, catalysts 1-Cu and 2-Cu can promote RNA cleavage chemistry either through “hydrolytic” (A) or “oxidative” (B) mechanisms promoted through the transient formation of higher valent Cu3+ The illustration shows initial intermediates proposed for RNA cleavage mechanisms. The requirement for redox coreagents to effect cleavage supports the requirement for Cu3+ formation and transient intermediates are shown that invoke either a role as Lewis acid in a classical hydrolysis mechanism involving a cyclic phosphate ester (A), or by an oxidative pathway involving a copper-associated reactive oxygen species (B). For the analogous reactions for oxidative cleavage of DNA, both C4’-H or C5’-H hydrogen abstraction are most common although the oligonucleotide 3’-phosphate product can be observed for oxidative C-H scission at any ribose carbon.37 For RNA, the oxidative reaction (B) is suggested to most likely occur by C4’-H or C5’-H hydrogen abstraction because these hydrogens are typically the most accessible and the pathways are less compromised by the presence of the C2’-OH.36

Reactions were also performed both in the presence and absence of coreagents. It can be seen from Figures 2.19 and 2.20 (supplemental material) that the overall reactivity in the absence of coreagents is negligible compared to reactions with coreagents (Figure

2.15 and Figure 2.16, supplemental material), consistent with a requirement for the coreagents in promoting the reactivity of the catalyst. Previous work has suggested that

the ATCUN motif mediates cleavage through a metal-associated reactive oxygen species,54 and the ability of the ATCUN motif to produce this reactive intermediate is enhanced in the presence of coreagents. An unexpected product from the 3’-overhang that is observed for the reaction of both 1-Cu and 2-Cu is the 2’,3’-cyclic phosphate, in addition to the 3’- phosphate, and for the 5’-overhang the major product is the 5’OH. Given the absence of such a reaction product with 1-Cu alone, the former cyclic phosphate product is consistent with a hydrolytic mechanism (Figure 2.10) that proceeds through a high oxidation state copper center serving as a Lewis acid to promote deprotonation of the 2’-OH, as shown in

Figure 2.10A. The rate of cyclic phosphate formation (Figure 2.11) is greater for 1-Cu

-1 -1 versus 2-Cu (kobs = 0.063 ± 0.017 min for 1-Cu and kobs = 0.021 ± 0.003 min for 2-Cu).

- These rates directly correlate with the observed differences in kcat values (0.53 ± 0.02 min

1 -1 for 1-Cu vs. 0.14 ± 0.01 min for 2-Cu) for each metallopeptide. Variability between kobs and kcat most likely reflect solution and methodology differences between the solution fluorescence assay and mass spectrometric detection.

Figure 2.11. Time dependent formation of 2’, 3’-cyclic phosphates and 3’-phosphates as determined by MALDI-MS (Left) The time dependent formation of 2’, 3’-cyclic phosphates in the presence of 1-Cu (orange) and 2-Cu (blue). (Right) The time dependent formation of 3’-phosphates in the presence of 1-Cu (orange) and 2-Cu (blue).

For 1-Cu the primary cleavage reaction appears to proceed via formation of the 2’,

3’-cyclic phosphate product and the accompanying 5’-OH fragment (Figure 2.21 supplemental material). Minor production of 3’-phosphate products can arise either from hydrolysis of the 2’, 3’-cyclic phosphates,95 or by an alternative hydrogen abstraction pathway. This latter mechanism appears to be more prevalent for cleavage reactions catalyzed by 2-Cu, with 3’-phosphates arising due to hydrogen abstraction pathways analogous to those for DNA. While 3’- phosphate formation can in principal occur as a result of hydrogen abstraction at any position on the ribose, for RNA hydrogen abstraction is expected to be most likely at C4’ or C5’ 96 and, when the observed product types are

taken into account, the most likely hydrogen abstraction pathway involves C5’-H abstraction (Figure 2.10B) which generates a 3’-phosphate and 5’-aldehyde (the mass difference of the 5’aldehyde product compared to the 5’-OH product is only 2 amu and, at the high molecular weights under consideration, are essentially indistinguishable by

MALDI-MS; consequently these products are not differentiated in Figure 2.21). Because the C5’ hydrogens are located next to the phosphate ester, these hydrogens are more accessible to solvent and would also be more accessible to the copper ion of the complex.

It should again be noted, however, that this hydrogen abstraction mechanism does not account for the production of 2’, 3’-cyclic phosphates, and so the “hydrolysis”-type mechanism is required and is a significant, and most likely dominant component of the overall RNA cleavage reaction.

Figure 2.12. The computed solution structures of Cu-GGHYrFK-amide (left) and Cu- GGhyrfk-amide (right) Distances are shown in Angstroms. Structures were optimized in Gaussian and images were produced in PyMol.

As an alternative to a direct hydrogen abstraction pathway, one can invoke 3’- phosphate formation entirely through hydrolysis of the 2’,3’-cyclic phosphate, especially in the presence of a powerful Lewis acid.95 However, the absence of significant 3’- phosphate formation for reactions catalyzed by 1-Cu, relative to one third of the products derived from reaction mediated by 2-Cu, suggest independent reaction routes and varying rates of reaction for “hydrolysis” versus “oxidative” paths (Figure 2.10). Based upon the kobs values noted earlier, it would be expected that 1-Cu would generate 3’-phosphates at a higher rate than 2-Cu if this product were formed by rapid hydrolysis cyclic phosphates - but this is not observed (Figure 2.11, right). Such variability in reactivity is expected because each catalyst will likely have a distinct orientation for the catalytic metal relative

to scissile bonds on the RNA, and the intrinsic reactivity toward each available pathway

(in this case hydrolysis versus hydrogen abstraction) will be locally determined by the stereoelectronic constraints imposed by each reaction site.

2.5.5 Binding Model

In order to further evaluate the proposed binding model shown in Figure 2.9, and derived from experimental RNA cleavage profiles, computational studies were performed to generate energy optimized structures for 1-Cu and 2-Cu (Figure 2.12) using a B3LYP

DFT calculation in Gaussian 09. Involvement of the tyrosine and phenylalanine aromatic rings in contributions to RNA binding would be expected to involve interactions with nucleic acid bases, whereas contributions from arginine should involve either electrostatic,

H-bonding and/or pi-stacking interactions, while lysine would be expected to engage primarily in electrostatic and H-bonding interactions with the phosphate backbone. The optimized structure for CuGGHYrFK-amide (1-Cu) shows the tyrosine and phenylalanine arranged on the same side of the peptide backbone potentially allowing their sidechains to pi-stack. By contrast, it can be seen from the optimized structure for CuGGhyrfk-amide

(2-Cu) that the phenylalanine is flipped out as the tyrosine forms an intramolecular hydrogen bond with the guanidine group of arginine. This reduces the availability of both of the aromatic residues to pi-stack in the groove of the RNA and illustrates the impact of changing the configuration of the amino acids. The alternating spacing of aromatic

residues and positive charges in the YrFK-amide of CuGGHYrFK-amide could provide the appropriate spacing to optimize both base interactions and electrostatic interactions.

To further test this model, these peptides were docked to the NMR structure of

SLIIb. Initial docking was performed using the optimized peptide structures from Gaussian and the binding was simulated using Autodock 4.2. Two hundred structures were simulated and the localization of the second lowest energy cluster is shown in the left box of Figure 2.7 for 1-Cu, and the lowest energy cluster is shown in the right box of Figure

2.7 for 2-Cu. These clusters, and the clusters of metals that they contain, are chosen as the working model for the proposed sites of reactivity because they are in the best agreement with the experimentally-determined cleavage sites determined by mass spectrometry

(Figure 2.13), where reactivity is centered on the copper ion.

The common docking sites based on clustering are different for each complex. This once again emphasizes the effect that conversion to D-amino acids can have not only on the conformation of the isolated peptide, but also on how it interacts with the target RNA.

The lowest energy structure for 1-Cu shows both aromatic residues oriented in the same direction, pointing towards the bases of the RNA, with the tyrosine hydroxyl group hydrogen bonding with the base on G16 and the phenylalanine positioned near the base of

G22. Further, the copper atom is placed near the phosphate group of A19, close to one of the major sites of reactivity (Figure 2.14). The top structure for 2-Cu, on the other hand, maintains the intramolecular salt bridge between the tyrosine and the arginine as shown in the optimized peptide structure and prevents the tyrosine from interacting with the RNA.

The phenylalanine is positioned between the backbone phosphate groups of C23 and G22 and the copper atom interacts with the phosphate group of A8, one of the major sites of reactivity as determined by MALDI-TOF MS (Figure 2.14). This mode of interaction with the RNA is consistent with the binding model described in the previous section where 2-

Cu relies more on electrostatic interactions from the arginine and lysine side chains.

Figure 2.14. Sites of interaction (yellow) based on modeling of 1-Cu with SLIIb (orange) and 2-Cu with SLIIb (blue).

An analysis of the binding energies obtained from docking experiments is shown in Figure 2.22 (supplemental material). The largest differences in contributions to overall energy are reflected in electrostatic contributions as opposed to other non-electrostatic factors that include hydrogen bonding (HB), van der Waals interactions (vdW), and desolvation energy. The calculations provide the overall combination of the last three factors, however, and so it is possible that there is a compensatory change in one or more of these properties without changing their overall sum. The differences between 1-Cu and

2-Cu reflect the placement of the metal binding domain, where the metal binding domain in 1-Cu is better positioned to directly interact with the RNA and mediate more effective 68

cleavage chemistry, and is consistent with the relative k2 values predicted from the experimental KM and KD values discussed earlier. However, the smaller electrostatic contribution exhibited by 1 and the greater activation barrier expected for release of 1, relative to 2 (because EA is greater when breaking H-bonds rather than electrostatic

97 contacts) results in a smaller koff for 1-Cu, relative to 2-Cu. Also, this trend in koff would be expected to impact the observed turnover numbers, as it in fact does, with 2-Cu (turnover number > 40) showing better turnover than 1-Cu (turnover number ~ 32, Table 2.2).

2.5.6 HCV Replicon Assays

Complex 2-Cu was also tested in the HCV cellular replicon assay previously described to assess activity in vivo. This cellular replicon assay is a widely accepted measure of drug efficacy for HCV treatment. The copper complex of 2 was found to have an IC50 of 1.92 μM, which is comparable to the previously reported data for 1-Cu (IC50 =

0.58 μM), and it exhibited no cytotoxicity up to 100 μM (Table 2.1). Despite a ~10 fold higher KM and ~34 fold worse kcat/KM, the observed cellular activity was approximately the same. This is a reflection of the complex combination of factors that determine the activity of these metallodrugs (Figure 2.17, supplemental material), and the possible limiting step of cellular uptake, which is not addressed in this work. The all D-amino acid analog also seems to maintain the ability to be taken up by cells and the lack of toxicity shows the specificity for the virus.

To further confirm that the mechanism of action in the cellular replicon assays is consistent with that proposed for catalytic metallodrugs, real-time polymerase chain reaction (RT-PCR) was performed to measure the RNA levels for both HCV RNA and ribosomal RNA (rRNA). The results can be seen in Figure 2.8 (left) and show a clear preference for HCV RNA over rRNA. Significantly, 1-Cu still showed activity up to final day tested, 9 days later, and reduced RNA levels to close to zero (Figure 2.8, left, data is normalized). In contrast, the rRNA levels decreased much less although there was still a small decrease relative to the control. This data is complicated by the fact that it will reflect both the amount of RNA being made as well as the amount of RNA being consumed. Also, the rate of signal amplification can be different for the two RNAs which further complicates any quantitative comparison. The data for copies of rRNA (Figure 2.8, right, not normalized) is shown and it is seen that the amount of rRNA being made over time increases both with and without 1-Cu but that the rRNA levels are slightly higher in the absence of catalyst. This could be due to small levels of background cleavage of the rRNA but is consistent with the idea that higher levels of rRNA are being made to compensate for the presence of the IRES, thereby reducing the effective concentration of rRNA available for use by the cell. In the presence of catalytic metallodrug, however, the reduction in the concentration of IRES means that more rRNA is available for the cell to use and, therefore, it does not need to make as much. Regardless, there is a clear preference for the HCV RNA and no significant overall cellular toxicity was observed as evidenced

by the TC50 values (greater than 100 µM). All of these are consistent with the metallodrug behaving in the cellular replicon assays in the similar way that is proposed for catalytic metallodrugs and shown for the in vitro experiments.

2.6 – Conclusion

The diasteromeric analog of 1-Cu, which incorporates D-configuration amino acids, shows potential for use in vivo. The assays described show that the chiral analog maintains activity, but the expected enhancement of efficacy due to increased stability to proteases is not reflected in these cellular experiments. Therefore, it is expected that the all D-amino acid version would perform better in an animal model. It binds to the target

RNA with a KD of 76 nM, and, despite differences in catalytic efficiency (kcat/Km = 0.0018

M-1min-1 for 2-Cu versus 0.62 M-1min-1, for 1-Cu), both show good activity in an FDA approved HCV replicon assay. This similar efficacy, combined with potentially higher in vivo stability, should lead to a more effective catalytic metallodrug in more complex systems such as animal models or humans.

2.7 – Supplemental Material

Table 2.3. Subcluster assignments of 1-Cu based on proximity of the copper atom to the site of reactivity as determined by mass spectrometry.

Pose Atom RNA Residue RNA Atom Distance Å Assigned Cluster 2 Cu U12 OP1 1.4 U14 6 Cu C13 OP1 1.5 U14 7 Cu U12 OP2 2.2 U14 11 Cu U12 OP1 1.6 U14 15 Cu U12 OP1 1.6 U14 16 Cu U12 OP2 4 U14 19 Cu C13 OP2 1.6 U14 26 Cu U14 O4 2.3 U14 27 Cu C13 OP2 3.7 U14 32 Cu C13 OP1 2.9 U14 33 Cu U14 OP2 1.7 U14 35 Cu C13 OP2 1.6 U14 36 Cu U12 OP2 4.2 U14 37 Cu C13 OP2 1.7 U14 39 Cu U14 O4 2.8 U14 44 Cu U14 OP2 4.3 U14 10 Cu C18 OP1 1.4 C18 25 Cu C18 OP1 4.2 C18 31 Cu C18 OP1 1.7 C18 38 Cu C18 OP1 1.6 C18 45 Cu G16 OP2 6.4 C18 1 Cu A19 OP2 1.7 A19 17 Cu A19 OP2 4.4 A19 22 Cu A19 OP2 3.2 A19 23 Cu A19 OP2 4.1 A19 28 Cu A19 OP2 4.2 A19 29 Cu A19 OP2 8 A19 40 Cu A19 OP2 5.5 A19 42 Cu A19 OP2 3.1 A19 43 Cu A19 OP2 3.5 A19 continued

Table 2.3 continued

3 Cu C10 OP2 1.5 G21 4 Cu U20 OP2 3.4 G21 5 Cu U20 OP2 1.7 G21 8 Cu C10 OP2 1.9 G21 9 Cu U20 OP2 1.6 G21 12 Cu G9 OP2 2.3 G21 13 Cu C10 OP2 1.5 G21 18 Cu C10 OP1 2.9 G21 20 Cu C10 OP1 2 G21 21 Cu G9 OP1 1.6 G21 24 Cu G11 OP1 1.7 G21 30 Cu U20 OP2 1.8 G21 34 Cu C10 OP2 1.7 G21 36 Cu U12 OP2 4.2 G21 41 Cu U20 OP2 1.6 G21

Table 2.4. Subcluster assignments of 2-Cu based on proximity of the copper atom to the site of reactivity as determined by mass spectrometry.

Pose Atom RNA Residue RNA Atom Distance Å Assigned Cluster 25 Cu A6 OP1 3.8 A6 27 Cu A6 OP1 2.8 A6 29 Cu A6 OP1 4.8 A6 33 Cu A6 HO2' 2.7 A6 76 Cu A6 OP1 3.6 A6 87 Cu A6 OP2 4.2 A6 14 Cu A7 OP1 3.5 A7 17 Cu A7 OP1 1.7 A7 22 Cu A7 OP1 1.4 A7 30 Cu A7 OP1 1.7 A7 34 Cu A7 OP1 2.6 A7 38 Cu A7 OP1 2.1 A7 39 Cu A7 OP1 1.3 A7 40 Cu A7 OP1 1.6 A7 42 Cu A7 OP1 4.5 A7 43 Cu A7 OP1 1.6 A7 50 Cu A7 OP1 1.7 A7 59 Cu A7 OP1 1.6 A7 74 Cu A7 OP2 4 A7 78 Cu A7 OP1 2.8 A7 1 Cu A8 OP1 1.5 A8 3 Cu A8 OP1 2.4 A8 5 Cu A8 OP1 1.7 A8 6 Cu A8 OP1 1.6 A8 7 Cu A8 OP1 4.7 A8 8 Cu A8 OP1 3.4 A8 10 Cu A8 OP1 1.6 A8 16 Cu A8 OP1 1.5 A8 19 Cu A8 OP1 1.7 A8 20 Cu A8 OP1 1.5 A8 24 Cu A8 OP1 2.2 A8 32 Cu A8 OP1 2 A8 35 Cu A8 OP1 1.5 A8 continued

Table 2.4 continued 36 Cu A8 OP1 1.6 A8 41 Cu A8 OP1 3.1 A8 51 Cu A8 OP1 1.5 A8 53 Cu A8 OP1 1.8 A8 55 Cu A8 OP1 1.5 A8 57 Cu A8 OP2 4.4 A8 60 Cu A8 OP1 3.1 A8 62 Cu A8 OP1 1.9 A8 66 Cu A8 OP1 4.9 A8 69 Cu A8 OP1 5 A8 72 Cu A8 OP1 1.6 A8 77 Cu A8 OP2 3.7 A8 82 Cu A8 OP2 3.2 A8 83 Cu A8 OP1 1.5 A8 9 Cu C23 OP2 3.7 C23 15 Cu C23 OP2 2.6 C23 61 Cu C23 OP2 4.1 C23 64 Cu C23 OP1 6.6 C23 68 Cu C23 OP2 3.6 C23 73 Cu C23 OP2 4.1 C23 85 Cu C23 OP2 2.5 C23 86 Cu C23 OP2 4 C23 2 Cu G22 OP1 1.6 G22 21 Cu G22 OP2 3.3 G22 31 Cu G22 OP2 3.8 G22 37 Cu G22 OP2 1.7 G22 44 Cu G22 OP1 3.6 G22 46 Cu G22 OP1 3.8 G22 47 Cu G22 OP2 1.7 G22 48 Cu G22 OP1 1.5 G22 52 Cu G22 OP2 1.7 G22 54 Cu G22 OP1 3.6 G22 67 Cu G22 OP2 2.3 G22 71 Cu G22 OP2 4.2 G22 81 Cu G22 OP1 4.4 G22 84 Cu G22 OP1 1.4 G22 88 Cu G22 OP1 2.6 G22

Figure 2.15. Summary of cleavage sites on SLIIb RNA promoted by 1-Cu as determined by mass spectrometric measurements The plot shows the formation of products at each position as a function of time following reaction with 1-Cu in the presence of coreagents. Reaction sites are defined as those with products at six or more time points.

Figure 2.16. Summary of cleavage sites on SLIIb RNA promoted by 2-Cu as determined by mass spectrometric measurements The plot shows the formation of products at each position as a function of time following reaction with 2-Cu in the presence of coreagents. Reaction sites are defined as those with products at six or more time points.

Figure 2.17. Overlap of the 20 NMR structures available in the PDB file 1P5N The top of the RNA is the most dynamic portion of stem loop IIb, while the rest is more rigid.

are those with products at six or more time points.

Figure 2.20. Relative amounts of each class of product observed by mass spectrometry displayed as a function of time following reaction with 1-Cu (top) and 2-Cu (bottom) 1- Cu produced almost exclusively 2’, 3’-cyclic phosphates for the 3’-overhang whereas for 2-Cu the relative amounts were about two-thirds 2’, 3’-cyclic phosphates and one-third 3’- phosphates

Chapter 3: Miniature Structure Activity Relationship Based on the GGHYRFK-Cu Complex Targeting HCV SLIIb IRES RNA

3.1 – Introduction

Hepatitis C Virus, HCV, effects approximately 2 million people. Recent advances in drug advances such as Sovaldi®, sofosbuvir, and Harvoni®, a mixture of sofosbuvir and ledipasvir, (Gilead) have taken the market by storm with $10.2 billion in total revenue for

2014 Sovaldi® and $2.1 billion total revenue for Harvoni®, which launched in October of

2014.98 These current treatments are far superior to the prior treatments that involved ribavirin and interferon, both of which were not specific and had numerous side effects leading to problems with patient compliance. However, these new drugs along with newer protease drugs are facing issues associated with the cost of treatment. One complication has arisen through Medicaid expansion. States that have not expanded coverage, are in general, withholding coverage of payment until the patients have exhausted alternatives, or worse, experience irreversible liver damage.99 Thus there exists a need for additional therapeutic targets. The approach to targeting HCV is moving to a multi-drug regimen with each drug targeting a different portion of the virus. This cocktail approach overwhelms the virus and prevents it from advantageously mutating to develop resistance. Our goal has been to add to the arsenal of this approach, in particular through catalytic metallodrugs that target HCV IRES RNA.

We have recently reported two complexes, GGHYrFK-Cu (1-Cu) and GGhyrfk-Cu

(2-Cu) that target SLIIb of the HCV IRES domain. These compounds have demonstrated activity in vitro as well as a in a FDA approved replicon assay. Further, 1-Cu has been shown when given with interferon- to have an additive to synergistic effect towards

HCV.62

Previous studies have shown the importance of the stereochemistry of amino acids in these compounds that target the Hepatitis C Virus RNA.63 This study aims to build upon that knowledge by looking at the all L-configuration of the lead compound, Cu-GGHYrFK,

1-Cu, which contains arginine in the D-orientation and compare it the other previously reported compound Cu-GGhyrfk, 2-Cu, which has all of the amino acids in the D- configuration. In addition, we aimed to systematically investigate the role of each amino acid in the targeting domain, YRFK, to determine the impact on binding, reactivity/chemistry, as well as cellular uptake.

3.2 – Materials and Methods

3.2.1 Binding Constant Determination

Insiya Fidai performed these experiments. For the complexes, RNA binding experiments were performed in the presence of 2.5 M SLIIb in 20 mM phosphate buffer

(pH 7.4) containing 100 mM NaCl. Serial aliquots of the peptides were added and the CD signal was monitored at selected wavelengths of 263 nm, another in the range 230-240 nm, and at 283 nm. Data was fit to a one-site binding model using Origin software. A quadratic one site binding equation, Eq 1,was used.

F = F0 + (KD + R0 + P0 - (((KD + R0 + P0)2) - (4 x R0 x P0))0.5) ÷ 2 ÷ R0 x (F1 - F0) + m x P0

Eq1. Quadratic one site binding equation. This equation is based upon two lines intersecting. F0 is the y-intercept of the first line and F1 is the y-intercept of the second line. KD and R0 correspond to the dissociation constant and inflection point. To account for any additional phases (a non-zero slope of the second line) an m term, or slope, is used to define this. P0 is the independent variable in concentration.

3.2.2 Complex Optimization

The peptide structures were optimized with Gaussian 09 vA01* using an Amber

MM/MD force field in an aqueous solvent, followed by a more thorough refinement with a DFT/B3LYP/3-21G force field.100 Solvent interactions, water in this case, were considered by using the universal solvation model, SMD, by Truhlar and co-workers.101

The structures of the metal complexes were optimized by attaching the crystallographically determined structure for CuGGH39 to the peptide and submitting the complexes using the above-mentioned force field and solvation model.

3.2.3 In silico Complex docking

This was carried out in conjunction with Insiya Fidai. The Cu-GGH atoms in the complexes were held fixed during the geometry optimization of the complexes by freezing the atoms in Gaussian 09. The optimized peptides and complexes were used for docking simulations. The solution state NMR structure of stem loop IIb (SLIIb) of the HCV IRES

RNA is readily available from the protein data bank (PDB: 1P5N). The top solution state

NMR structure (out of the 20 available) was used for docking simulations using AutoDock

4.2. The entire SLIIb domain was considered for docking. In all cases, the complex was made flexible except for aromatic carbons, peptide bonds, and the metal binding domain.

The Lamarckian Genetic Algorithm allows a large degree of flexibility in the selection of

the initial starting position: essentially searching a local area, identifying the thermodynamically most favorable orientation, and then “mutating” to a different spot on the RNA. Comparison of the docking in this manner provides a measure of how readily the complex is able to be position in a manner to perform effective chemistry.

Autodock does not have parameters for copper so an iron atom was substituted to mimic the geometry of the copper with a charge of +0.8. Iron and copper should have similar ionic radii and iron simulates the charge on the copper ion. A value of +0.8 is used to simulate the +2 charge in order to compensate for the tendency of Autodock to overestimate electrostatic interactions.86 We have previously demonstrated this approach.63 The geometry around the metal was based on the X-ray crystal structure of Cu-

GGH and was restricted to a square planar configuration.39 a Lamarckian Genetic

Algorithm and the following parameters were used for docking: population size of 150, a random starting position and conformation, a maximal mutation of 2 Å in translation and

50 o in rotations (elitism of 5), iterations of Solis and Wets local search of 300, torsional degrees of freedom of 23 for the peptide complex, an external grid energy of 1000 kcal/mol, a mutation rate of 0.02 and a crossover rate of 0.8, and local search rate of 0.06. Simulations were performed with a maximum of 107 energy evaluations and a maximum of 27000 generations. The total number of hybrid GA-LS runs was set to 200. Docking analyses were performed on the UNIX system at the Ohio Supercomputer Center using the

OAKLEY cluster platform with 12 CPU processors running on 1 computing node and a total wall time of 120 hours.

3.2.4 Reaction Kinetics via Fluorescence.

HCV IRES RNA cleavage was monitored in vitro by fluorescence using 5’ fluorescein end-labeled RNA with excitation and emission wavelengths of 491 nm and 518 nm, respectively. Reactions were carried out at 25 ºC in reaction volumes of 150 µL in the presence of 1 mM ascorbic acid and 1 mM H2O2 in HEPES buffer (pH = 7.4, 100 mM

NaCl) with 333 nM fluorescein labeled HCV SLIIb and analyzed according to the change in fluorescence observed as the reaction occurred. A calibration curve of concentration of

SLIIb versus fluorescence intensity was constructed to convert the RFI to intensity. Both a time-dependence and a concentration-dependence of catalyst activity were observed. The initial velocity of the time dependence plot was used to generate the pseudo Michaelis-

Menten plots which were then fit to the Michaelis-Menten equation. All fits were performed using Origin software. The values shown are an average of at least three trials.

A turnover number was determined based on the limiting amount of complex and determining the amount consumed by a specific amount of RNA by measuring the change in intensity at 518 nm.

3.2.5 Reaction Rates Monitored by Gel Electrophoresis

Reactions were carried out in 10 L containing 750 nM Fl-SLIIb and 750 nM of the appropriate copper peptide complex in addition to the co-reagents of 1 mM ascorbate and 1 mM hydrogen peroxide. Reactions were staggered to have them end at the same time.

Time points were 180, 120, 90, 60, 45, 30, 15, 5 minutes. A control lane of 750 nM Fl-

SLIIb was also included. Five microliters of 20% sterile glycerol was added to each reaction before 12 L of the reaction mixture was loaded onto a 4% agarose gel. The gel

was ran at 70 mV for approximately 20 minutes. Gels were quantified in Quantity One and then fitted in Origin software using an exponential decay.

3.2.6 MALDI-TOF Mass Spectrometry.

Reactions for MALDI-TOF analysis were run as described above, but using 10 µM fluorescein labeled IRES SLIIb and 10 µM copper-peptide incubated for up to 90 minutes.

Reactions were then quenched by being placed on ice and Zip Tipped. Zip Tip was performed using C18 Zip Tips from Millipore Co. in order to desalt the reaction mixtures prior to mass spectrometric analysis. Zip Tips were wetted with a 50:50 mixture of acetonitrile:water and equilibrated with 2 M triethylammonium acetate (TEAA), pH 7.0.

0.3 M 4-hydroxypicolinic acid (HPA) and 30 mM ammonium citrate in 30% acetonitrile, drying, spotting with 1 µL of a 2:1 RNA:matrix mixture, and allowed to dry. A calibration mixture containing 4 RNAs covering a range of molecular weights, namely (GU)3, (GU)9,

(GU)14, (GU)20, with molecular weights of 1,892.1, 5,800.4, 9,057.3, 12,965.6 amu, respectively, was used to calibrate the instrument. All MALDI-TOF MS analysis was performed on a Bruker MicroFlex LRF instrument equipped with a gridless reflectron, using negative ion mode and reflectron mode. Typically, at least 1000 shots were summed per spectrum to acquire an accurate representation of the reaction. Data analysis was performed using Bruker flexAnalysis software. Assignment of peaks was performed by comparison of each peak list with the expected masses for possible cleavage products

(Table 3.8). Only m/z values > 1500 amu and those with a signal to noise ratio greater than 87

five were considered, since excessive spectral crowding occurred at lower m/z ranges.

Time dependence was used as a determination of peak validity.

Reactions for the time dependent assay were collected at: 2 min, 10 min, 20 min,

30 min, 45 min, 60 min, and 90 min. Heat maps and initial rates were generated by first summing the total change in intensity at a position within the RNA sequence and determining the fraction of associated change at each time point. An apparent initial rate was then determined from the linear region of the data (first 30 minutes).

3.2.7 Cellular Uptake.

This assay was carried out by Zhen Yu. Human liver cancer cells HuH-7 was cultured in DMEM (Dulbecco’s modified Eagle's medium) supplemented with 10% FBS

(fetal bovine serum) in an atmosphere containing 5% CO2 at 37°C. Cells were seeded to 6- well plates and incubated for 24 hr to allow cells to attach. Once approximately 1 million cells were in each well, 10 µM of the corresponding copper complex was added. After incubation for a set time, 0, 1, 2, 4, 8, or 16 hr, at 37°C with the indicated complex, the cells were rinsed with PBS and lysed in a urea buffer containing protease inhibitor cocktails. Total protein concentration of cell lysate was measured through a Bradford assay. A calibration curve for the determination of concentration of the peptide along with the principle first fragmentation was determined through the use of a hydrophilic interaction liquid chromatography (HILIC) column (Atlantis, 4.6 x 5.0 mm, 5 µM). A two solution system was employed with the first solution, water with 0.1% formic acid (solution

A), and the second solution, acetonitrile (solution B). Initially the system was set to 20% solution A and 80% solution B. Over the course of 5 minutes, solution B was ramped up to 30% (2%/min). The mass spectrometer, Agilent 6460 Triple Quadrupole System, was 88

operated under the multiple reactions monitoring (MRM) mode with a collision energy 25-

60 eV depending on peptides. The dwell time was 200 ms for each transition. The m/z transitions (precursor to product) monitored were 432.2 to 110.1 (GGHYRFK), 771.4 to

252.1 (Y4A), 778.3 to 252.1 (R5A), 787.4 to 252.1 (F6A), 806.4 to 252.1 (K7A), 432.2 to

110.1 (I1), 842.4 to 252.1 (Y4F), 815.4 to 449.3 (Y4D), 814.4 to 252.1 (Y4N), 828.4 to

234.1 (Y4K), 270.1 to 110.1 (GGH). Lysate samples were run in triplicate. The concentration of the peptide was determined in g/g of lysate and plotted versus time point.

An exponential fit was used to determine the half-life of uptake as well as level of saturation of the peptide when possible. Linear fits were used to determine the initial uptake rate. All fits were determine in Origin software.

3.2.8 Peptide Synthesis.

Peptides were synthesized by standard solid phase peptide synthesis utilizing Fmoc chemistry on a PS3 synthesizer (Protein Technologies). All peptides were purified by

HPLC using reverse-phase Axia packed C18 Gemini 5 m 100 x 21.20 mm column

(Phenomenex). A gradient of 2%/min from water with 0.1% TFA (100%) to acetonitrile with 0.1% TFA (100%) was employed. Product was determined by ESI-MS and stock of

- - peptides were determined by either tyrosine determination (270nm = 1400 M cm ) and/or serial titration with a standardized CuCl2 solution monitoring at 250 nm and 520 nm.

3.3 – Results and Discussion

3.3.1 Binding Constant Determination.

It has been shown that the isolated stem loop IIb domain of the HCV IRES maintains the structure adopted in the full length RNA and is a useful probe for in vitro assays. Peptide binding studies were performed using circular dichroism. A dissociation 89

constant, KD, for peptides 1 and complexes 1-Cu and 3-Cu through 12-Cu binding to SLIIb

RNA were observed in the range from 10 to 3000 nM (Table 3.1) determined by monitoring the change in the CD signal at least two wavelengths. A representative binding curve is shown in Figure 3.1 (the remainder are available as Figure 3.12 in the supplementary).

CD signal at 263 nm CD signal at 15

14 0 1 2 3 4 5 6 7 8 9 10 11 12 [3-Cu] M

Figure 3.1. Binding affinity of 3-Cu to 2.5 M SLIIb. Additional binding curves are available in the supplementary material as Figure 3.12.

Table 3.1. Summary of binding constant for peptides binding SLIIb.

K (nM) K (nM) Complex D I Determined by CD Docking Simulation

1 GGHYrFK 66.3 ± 1.4 a

1-Cu GGHYrFK-Cu 18.7 ± 0.8 7.7

2 GGhyrfk 76 ± 3 b

3-Cu GGHYRFK-Cu 72.8 ± 7.5 69

4-Cu GGHARFK-Cu 93.6 ± 16.8 108

5-Cu GGHYAFK-Cu 189.2 ± 54.2 159

6-Cu GGHYRAK-Cu <10 c 9.4

7-Cu GGHYRFA-Cu 56.6 ± 4.6 61.7

8-Cu GGHKFRY-Cu 38.0 ± 2.1 69.6

9-Cu GGHDRFK-Cu 258.2 ± 41.4

10-Cu GGHNRFK-Cu 129.7 ± 35.3

11-Cu GGHKRFK-Cu 69.0 ± 12.3

12-Cu GGHFRFK-Cu 47.1 ± 35.5

13-Cu GGH-Cu 3840 ± 1228 a Previously reported value of 44 nM determine by fluorescence based tyrosine emission assay for free peptide alone.62 This shows values of this assay are comparable in this study to that of the previously reported values. b Previously reported value as determined by fluorescence based tyrosine emission assy.63 c A lower limit of 10 nM was used in the regression analysis fitting in Origin.

The parent compound in this report, 3-Cu, has a binding affinity of the same order of magnitude as the previously reported compound, 1-Cu (72.8 ± 7.5 nM versus 18.7 ± 0.8 nM). The decrease in binding is a result of the change in the stereochemistry in the arginine residue. This shows that the stereochemistry of this residue is important and may be result of steric hindrance. Complexes 3-Cu and 4-Cu are similar in binding affinity whereas 9-

Cu and 10-Cu show a decrease in affinity of 258.2 nM and 129.7 nM respectively. The role of charge at the fourth position is revealed, in complex 9-Cu through 12-Cu as may be expected, a negatively charged residue at this position decreases the affinity (258.2 nM) whereas a positively charged residue increases the affinity (69.0 nM). A neutral residue modification has an affinity between these two (72.8 nM, 93.6 nM, 129.7 nM, and 47.1 nM, for 3-Cu, 4-Cu, 10-Cu, 12-Cu, respectfully). Interestingly, removal of the hydroxyl group of tyrosine, 3-Cu, to a phenylalanine residue, 12-Cu, increases the binding affinity, suggesting a relaxation in steric interaction or allowing a greater  - interaction. The impact this position has on reactivity will be explored below in the next section.

Another observed trend may be seen between 3-Cu, 4-Cu, 5-Cu, 6-Cu, and 7-Cu.

Complexes 4-Cu and 6-Cu represent changes in the aromatic residues of the parent compound 3-Cu. These changes to alanine show a negligible to a slightly favorable increase of 93.6 nM to 56.6 nM, respectfully, compared to 72.8 nM for 3-Cu. Complexes

5-Cu and 7-Cu, which represents a change from a positively charged residue (arginine or lysine) have a larger influence on the binding affinity; 189 nM and 56.6 nM respectfully.

Of interest is the observation that the modification at the phenylalanine and lysine position to alanine increase the overall affinity of the complexes while changes at the tyrosine or arginine positions decrease the affinity. 92

3.3.2 In silico docking and inhibition constants

Previously we have reported the geometric optimization and simulated docking of

1-Cu and 2-Cu to SLIIb.63 This process was carried out for the copper complexes 3-8. The inhibition constant, was calculated by taking the weighted average of the KI’s of the first structure of the lowest energy clusters to cover the majority of the 200 docked structures

(Table 3.12). This KI, is compared to the determined KD from circular dichroism (Table

3.1). As the KD represents the total sum of microstates of bound versus unbound form it is not surprising that the weighted KI’s obtained are similar to the experimentally determined values. Further, weighted averages were used to calculate binding energies, non- electrostatic and electrostatic contributions (Table 3.2). Available in Figure 3.2 is the docking of the top clusters with respect to the structure of SLIIb.

With confidence in the binding, these structures also provide usefully models to predict localization of the complex and in particular where the metal binding domain is located as well as the mode of binding (Figure 3.2). For all of the complexes simulated, the interaction of the complexes with RNA has a greater contribution from non-electrostatic contributions (the summation of van derr Waal’s, , and hydrogen bonding) than electrostatic.

Figure 3.2. Positions of the top three copper atoms in the representative clusters for complexes 3-Cu through 7-Cu (spheres). 3-Cu (red) cluster 1, 46 structures; (green) cluster 2, 65 structures; (blue) cluster 3, 55 structures. 4-Cu (red) cluster 1, 29 structures; (green) cluster 2, 63 structures; (blue) cluster 3, 15 structures; (orange) cluster 4, 57 structures. 5- Cu (red) cluster 1, 69 structures; (green) cluster 2, 67 structures. 6-Cu (red) cluster 1, 80 structures; (green) cluster 2, 62 structures. 7-Cu (red) cluster 1, 35 structures; (green) cluster 2, 90 structures; (blue) cluster 3, 25 structures.

Table 3.2. Summary of in silico screening of complexes. Total # of Binding Intermolecular Non-electrostatic Electrostatic Torsional K Internal Complex Docked I Energy Energy Energy Energy Energy (nM) Energy Structures (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol)

3-Cu 166 69.0 -10.36 -17.22 -10.76 -6.45 -4.09 6.86

4-Cu 164 108.5 -9.53 -15.93 -8.60 -7.37 -2.64 6.56

95 5-Cu 136 159.1 -9.38 -15.05 -9.01 -6.04 -3.20 5.67

6-Cu 142 9.4 -11.11 -17.38 -10.01 -7.37 -1.92 6.26

7-Cu 150 61.7 -10.12 -15.49 -9.94 -5.19 -4.11 5.37

8-Cu 161 69.6 -9.83 -16.69 -9.03 -7.67 -3.80 6.86

a This is the total number of docked states in the clusters used out of a maximum total of 200. b This represents the sum of van der Waal's, hydrogen bonding, and desolvation energy

3.3.3 Reaction Kinetics via Fluorescence

Pseudo Michaelis-Menten parameters were determined for all the copper complexes. An example of complex concentration versus initial rate for 3-Cu is shown in

Figure 3.3 (additional figures for the remaining complexes is available in the supplemental material as Figure 3.13). Table 3.3 summarizes the results, including a turnover number for a select few of the compounds. The traces for the turnover values is available in Figure

3.4.

Inital Rate (nM/min) Inital Rate

0 5 10 15 20 25 30 35 40 45 50 55 [3-Cu] M

Figure 3.3. Concentration of 3-Cu versus initial rate of cleavage with 333 M SLIIb and 1 mM ascorbic acid and 1 mM hydrogen peroxide at pH 7.4. Error bars represent the standard deviation of at least three trials.

Table 3.3. Pseudo Michaelis-Menten perimeters determined by fluorescence.

V k KM kcat/KM Cu-Peptide max cat TO # (nM/min) (min-1) (M) (M-1 min-1)

1-Cu 62 53 0.53 ± 0.02 0.85 ± 0.1 620 x 10-3 32 GGHYrFK

2-Cu 63 140 0.14 ± 0.01 7.9 ± 1.2 18 x 10-3 >40 GGhyrfk

3-Cu 26.3 ± 0.8 0.08 ± 0.01 55 ± 3 1.4 x 10-3 ± 0.1 x 10-3 32 GGHYRFK

4-Cu 169 ± 11.6 0.51 ± 0.03 13 ± 2 38 x 10-3 ± 9 x 10-3 60 GGHARFK

5-Cu 457 ± 29 1.37 ± 0.09 78 ± 7 18 x 10-3 ± 3 x 10-3 N.D. GGHYAFK

6-Cu 38 ± 1 0.11 ± 0.01 117 ± 3 1 x 10-3 ± 0.1 x 10-3 48 GGHYRAK

7-Cu 305 ± 10 0.92 ± 0.10 36 ± 2 26 x 10-3 ± 2 x 10-3 26 GGHYRFA

8-Cu 759 ± 34 2.28 ± 0.02 335 ± 17 7 x 10-3 ± 1 x 10-3 N.D. GGHKFRY

9-Cu 484 ± 23 1.45 ± 0.07 45 ± 4 32 x 10-3 ± 4 x 10-3 N.D. GGHDRFK

10-Cu 19 ± 2 0.06 ± 0.01 14 ± 2 4 x 10-3 ± 1 x 10-3 N.D. GGHNRFK

11-Cu 3.6 ± 0.4 0.01 ± 0.01 0.5 ± 0.4 22 x 10-3 ± 55 x 10-3 N.D. GGHKRFK

12-Cu 174 ± 5 0.52 ± 0.28 49 ± 2 11 x 10-3 ± 1 x 10-3 N.D GGHFRFK N.D. – not determined TO # - Turnover determined by amount of RNA consumed divided by the catalyst concentration

Figure 3.4. Turnover determination for 3-Cu, 4-Cu, 6-Cu, and 7-Cu. Determination of turnover was carried out in the presence of 1 mM ascorbate and 1 mM hydrogen peroxide and usually 10 nM of complex except in the case of 3-Cu when 1 M of the complex was used.

It is truly interesting that the conversion of the D-arginine in 1-Cu to the L-arginine for 3-Cu results in a ~400 fold decrease in catalytic efficiency (kcat/KM). This is a combination of a decrease in the kcat by ~6 fold as well as an increase in KM by ~64 fold.

This is remarkable in the dissociation constants determined for these compounds were 19 nM and 72 nM for 1-Cu and 3-Cu respectfully. The stereochemistry of the arginine position is once again highlighted for the importance to facilitate the correct orientation for effective chemistry. This is reflected by substituting the tyrosine position with a smaller amino acid 98

to relax the steric constraint, as can been seen in the trend of the kcat’s of 0.08, 0.51, 1.45, and 0.52 min-1 for 3-Cu, 4-Cu, 9-Cu , and 12-Cu respectfully. With changing the tyrosine to alanine, 4-Cu, or to phenylalanine, 12-Cu, the kcat’s are now on par with 1-Cu, while

-1 changing to aspartic acid yield a greater kcat (1.45 min ). The aspartic acid makes sense if the residue kinks to orient away from the RNA and in return forces the metal binding domain closer to the RNA. Complexes 10-Cu and 11-Cu were ignored as a result of the decrease in kcat values that most likely results from additional contacts, either hydrogen- bonding (10-Cu) or electrostatic (11-Cu), and complex stabilization is evident by the two having the lower KM’s in the series; 0.5 and 49 M respectfully.

Comparing 3-Cu through 7-Cu, an increase in kcat is observed following substitution of alanine for the positively charged residues of arginine, 5-Cu, and lysine, 7-

Cu. Although the kcat’s for these are relative close, the KM for the 7-Cu is approximately two-fold better than 5-Cu; 36 M and 78 M respectfully. This is consistent with the binding affinities of 56 nM and 189 nM for 5-Cu and 7-Cu, respectfully. This would imply that although 7-Cu has a tighter binding, it lacks the orientation of the copper domain to perform as effect as 5-Cu.

The change on the phenylalanine in position 6 has drastic effects on the catalytic properties for 6-Cu. The KM for the complexes is now greater than 100 M (greater than 2 fold from for 3-Cu) and the catalytic efficiency is the lowest of all the complexes screened.

This would indicate that the phenylalanine is an important residue for recognition of SLIIb.

3.3.4 Reactivity measured by fluorescence in gel assays.

Reactivity was measured by using fluorescein-labelled SLIIb and loading reaction mixtures of varying time length onto a 4% agarose gel. This was performed to help 99

collaborate the initial rates through a secondary method. A figure containing all of the gels is available in the supplemental material as Figure 3.14. An example of a time versus concentration plot is shown in Figure 3.5 (all the fits are available in the supplementary as

Figure 3.15). Table 3.4 contains a summary of the initial rates determined from quantifying and fitting the gel.

Figure 3.5. Time-dependence of 5-Cu with fluorescein-labelled SLIIb. Additional plots of the remaining complexes are available in the supplemental figures.

Comparison of the initial rates in Table 3.4 reveals that complex 7-Cu has the fastest initial rate followed by 8-Cu, 5-Cu, and 11-Cu respectively. However, there does not appear to be a correlation between the KD, kcat, or KM or amount of SLIIb consumed when it comes to predicting the initial rate. The majority of the complexes approached full consumption of the RNA present; exceptions include 9-Cu, 10-Cu, 11-Cu and 12-Cu. Both

10-Cu and 11-Cu have the lowest kcat of the complexes tested and it make sense that they would not consume as much of the SLIIb given the possibility of catalyst degradation over

100

the longer time frame. Furthermore, 9-Cu displayed the highest dissociation constant, outside of 13-Cu, in addition to having the negative charge next to the catalytic domain.

This most like results in electrostatic repulsion and may explain why the amount of SLIIb consumed is reduced. The most likely scenario is that the tyrosine is involved in the catalytic process. The only complex that still showed complete reactivity without the tyrosine present is 4-Cu.

Table 3.4. Initial Rates as determine by fluorescein-labelled SLIIb in agarose gel.

Complex Initial Rate (nM/min) % SLIIb consumed

3-Cu 1 ± 1 102 ± 2

4-Cu 8 ± 2 93 ± 7

5-Cu 26 ± 3 92 ± 9

6-Cu 9 ± 1 82 ± 8

7-Cu 53 ± 11 96 ± 14

8-Cu 32 ± 6 81 ± 9

9-Cu 2 ± 1 25 ± 7

10-Cu 6 ± 4 8 ± 3

11-Cu 8 ± 1 69 ± 5

12-Cu 20 ± 9 20 ± 6

101

3.3.5 Reactivity and Mechanistic Insights.

As previously reported, time-dependent reactivity was also performed by MALDI-

TOF characterization. The mass spectra for each time point was screened using the script

Mass Daddy68 for possible products based upon previously reported work with DNA (the mass list is available in the supplemental material as Table 3.8). The output of this data is available in the supplemental material (Table 3.10 through 3.14 and Figure 3.16 through

3.20). This data was filtered to ensure the product appeared across the time range and most importantly showed a time-dependence in the intensity of the product peak. A summary of reactivity with overlap of predicted Cu positions from Figure 3.2 is available in Figure 3.6.

The individual initial rates for each position is available in Table 3.5.

Figure 3.6 allows an additional level of scrutiny of the proposed binding model as predicted through the docking simulation. In general the reactivity of the complexes, 3-Cu through 7-Cu, corresponds well with the proposed binding sites. Furthermore, insights into the binding motif (cluster) that is most responsible for reactivity are gained. For 3-Cu, the majority of the reactivity is centered towards the top of SLIIb. Both the second cluster and third cluster have the copper atoms position near this region. This suggests that even though the first cluster may be the most thermodynamically favorable for binding, it is not the primary site for reactivity. A similar finding has been reported in the past for 1-Cu. It is worth noting, that the second and third cluster also have a larger population than the first cluster as well.

For 4-Cu the predicted binding and reactivity has shifted towards the middle bulge of SLIIb RNA. With this shift, there is a slight decrease in overall reactivity compared to that of 3-Cu. In general, when the predicted binding is in this middle bulge the overall 102

reactivity is decreased. This would suggest that either the copper metal center is not positioned in a fashion to facilitate efficient chemistry, or the binding affinity is tighter which results in a decrease in the release of the complex and ultimately the turnover, or a combination of both.

Figure 3.6. Heat maps of complexes 3-Cu through 7-Cu which depicts the relative initial rate per minute at each position in relation to the simulated docking of Cu atoms from the copper complexes (white to black spheres). Scale: red > 3.0, orange 2.5 – 3.0, yellow 2.0 – 2.5, green 1.5 – 2.0, blue 1.0 – 1.5, purple 0.5 – 1.0. A numerical representation is available in Table 3.5. The fit of the initial rates is available in Figure 3.21. 3-Cu (white) cluster 1, 46 structures; (grey) cluster 2, 65 structures; (black) cluster 3, 55 structures. 4- Cu (white) cluster 1, 29 structures; (light grey) cluster 2, 63 structures; (dark grey) cluster 3, 15 structures; (black) cluster 4, 57 structures. 5-Cu (white) cluster 1, 69 structures; (black) cluster 2, 67 structures. 6-Cu (white) cluster 1, 80 structures; (black) cluster 2, 62 structures. 7-Cu (white) cluster 1, 35 structures; (grey) cluster 2, 90 structures; (black) cluster 3, 25 structures.

103

Table 3.5. Summary of initial rates based on MALDI-TOF mass spectrometry for complexes 3-Cu through 7-Cu.

Initial Rate (Normalized Intensity/min multiplied by 1000)

Position 3-Cu 4-Cu 5-Cu 6-Cu 7-Cu

A7 1.0 ± 0.3 0.8 ± 0.1 0.5 ± 0.1 0.6 ± 0.1

A8 0.7 ± 0.1 0.5 ± 0.1

C10 0.6 ± 0.1

G11 1.2 ± 0.5 0.4 ± 0.1

U12 0.2 ± 0.1 0.2 ± 0.1

G16 1.2 ± 0.3 2.1 ± 0.2 0.7 ± 0.1

C17 0.3 ± 0.1 1.0 ± 0.2

C18 1.56 ± 0.3 0.7 ± 0.2 0.9 ± 0.1 0.7 ± 0.2 1.0 ± 0.3

A19 2.3 ± 0.4 1.7 ± 0.3 0.6 ± 0.2

U20 2.5 ± 0.4 1.3 ± 0.2 0.2 ± 0.3

G21 0.2 ± 0.1 1.3 ± 0.1

G22 0.5 ± 0.1

G24 1.6 ± .3 0.6 ± 0.1 0.2 ± 0.1

U25

U26 2.7 ± 0.5 1.2 ± 0.1 0.20 ± 0.1

A27 1.84 ± 0.2 2.2 ± 0.5 0.3 ± 0.2 0.7 ± 0.2

G28 0.5 ± 0.1 3.4 ± 0.7 0.4 ± 0.1

U29 0.1 ± 0.1

A30 0.4 ± 0.1

104

With the high throughput accessibility of using an expected product list in combination with Mass Daddy, the relative abundance of each overhang may be determined. A summation of the 3’ predicted products (2’, 3’-cyclic phosphates, 3’- phosphates, 3’-hydroxyl, 3’-phosphglycaldehyde, 3’-phosphoglycolate, 3’-a-b) as well as the 5’ predicted products (5’-hydroxyl, 5’-aldehyde, 5’-diol, 5’-phosphate, and 5’-z) was performed at each time point and was divided into each overhang to determine a percent of that overhang for each time point. Figure 3.7 shows the percent of each 3’ product overhang at a particular time point where as Figure 3.8 shows the respective percent of 5’ products. The primary products observed are the 2’, 3’ - cyclic phosphates and the 5’ - hydroxyl. This would suggest that major mode of action for these complexes would be through an “oxidative hydrolytic” mechanism. This has been discussed earlier, where it is the formation of a transient “Cu3+” species under the condition of ascorbate and hydrogen peroxide that acts as a Lewis acid that promotes the hydrolysis. However, little additional insights can be gleamed from those products.

Of more interest are the products of lower accumulation. Table 3.6 describes the different predicted products along with the expected mechanism. Complexes 3-Cu, 5-Cu,

6-Cu, and 7-Cu show formation of 5’-phosphates, that may arise either as the minor product of hydrolysis or is a sign of H’-abstraction. It has also been reported earlier that both 1-Cu and 2-Cu can promote hydrogen abstraction.63 The H’-abstraction would be expected to arise primarily from 4’ or 5’ – hydrogen abstraction as these are the most solvent accessible.

5’-H abstraction is the prevalent mechanism as the signature product of 4’-H abstraction, the 3’-phosphoglycolate, is not present. Further, the relative amounts of the 5’-aldehyde (a unique product of 5’-H abstraction) may be underrepresented as a result of peak resolution 105

from MALDI; the mass difference between the 5’-OH and 5’-aldehyde is 2 amu. The 5’- aldehyde is observed in 4-Cu and 6-Cu.

Figure 3.7. Percent overhangs for predicted 3'-overhangs (left to right: 2’, 3’-cyclic phosphates, 3’-hydroxyl, 3’-phosphates, 3’-phosphglycaldehyde, 3’-phosphoglycolate, 3’- a-b) with time. (A) 3-Cu, (B) 4-Cu, (C) 5-Cu, (D) 6-Cu, (E) 7-Cu. Relatively little 3’- product was observed for 7-Cu, which explains the differences between the other graphs.

106

Figure 3.8. Percent overhangs for predicted 5'-overhangs (left to right: 5'-hydroxyl, 5'- aldehyde, 5'-diol, 5'-phosphate, 5'-z). A) 3-Cu, (B) 4-Cu, (C) 5-Cu, (D) 6-Cu, (E) 7-Cu.

107

Table 3.6. Description of overhangs and possible mechanisms.

mass Overhang Possible Mechanism a (amu)

3’-OH 0.00 Hydrolysis b

2',3'-Cyclic Phosphates +61.96 Hydrolysis

Hydrolysis, 3'-Phosphates +79.98 1’, 2’, 3’, 4’, or 5’-H abstraction c

3'-Phosphoglycolates +138.02 4’H abstraction

5’-Hydroxyl 0.00 Hydrolysis

5’-Aldehyde -2.016 5’-H

5’-H, followed by hydration of 5’-Diol +16.00 5’-aldehyde

Hydrolysis, 5'-phosphates +79.98 1’, 2’, 3’, 4’, or 5’-H abstraction c

a. Proposed mechanisms are based upon the equivalent DNA mechanism. b. There is a closely related oxidative product resulting for 3’-H abstraction with mass difference of 2 amu which may inflate this value. c. Although 1’ through 5’ H-abstraction is possible, 4’ and 5’-H abstraction are expected to be the most prevalent as these positions are the more solvent exposed.

108

3.3.6 Cellular uptake

Complexes 3-Cu through 12-Cu were incubated with cells and at various time points were lysed to determine the amount of complex that was within the cell. Table 3.7 contains the information of the initial rate uptake, the saturation t1/2, and the apparent saturation levels. The concentration is presented in grams of complex per grams of total protein of the lysate as was determined by a Bradford assay. An example time dependence of concentration versus time plot is available for 3-Cu in Figure 3.9 (additional figures are available in Figure 3.22).

109

Table 3.7. Cellular Uptake Values

Initial Rate of Uptake Saturation t Saturation Complex 1/2 (ng/g/hr) a (hr) (ng/g) a

3-Cu 2.2 ± 0.3 16.1 ± 0.8 60 ± 2

4-Cu 9.3 ± 0.6 >50 268 ± 35 b

5-Cu 1.8 ± 0.2 21.3 ± 0.7 60 ± 2

6-Cu 4.0 ± 0.2 >50 112 ± 12 b

7-Cu 12.3 ± 0.9 67.9 ± 8.3 405 ± 70 b

8-Cu 9.9 ± 0.4 10.7 ± 1.7 150 ± 9

9-Cu 11.4 ± 1.6 4.4 ± 0.7 70 ± 2

10-Cu 0.3 ± 0.1 0 <5 c

11-Cu 26.2 ± 0.2 2.5 ± 0.1 130 ± 3

12-Cu 0.9 ± 0.1 7.3 ± 0.8 100 ± 2

13-Cu 0.3 ± 0.1 0 N.D. d a ng of complex per gram of cell lysate b Values are lower estimates using a t1/2 of 24 hrs. c Value is overestimated based upon a the single 18 hr time point. All earlier points had shown little uptake (< 1 ng of complex/g of cell lysate) d The data fitting utilizing an exponential equation was not able to provide a reliable value and therefore no value is reported.

110

Cationic aromatic peptides, a peptide with a balance of aromatic residues and positive charge residues, have been described before as having the ability to enhanced cellular uptake.88, 102-103 Previous work in the field has shown the targeting sequence of 1-

Cu, Tyr-D-Arg-Phe-Lys-amide, to be able to cross the cell membrane through an energy- independent method. Additional work has suggested it to have poor blood-brain barrier permeability.87 This is remarkable as the endogenous opioid peptide, Tyr-Gly-Gly-Phe-

Met, is uptake by receptor-mediated.104 A simple modification of changing the tyrosine to

2, 6-dimethyltyrosine allowed the compound to cross the blood-brain barrier. This was surprising as a result of its relative large size (>500), the highly polar backbone and a 3+ net charge under physiological conditions. A hall mark of both of these compounds is the alternating aromatic-cationic structural motif; Tyr, Dmt, or Phe as the aromatic residues and Arg and Lys as the basic residues. This motif allows for the intramolecular cation-pi interaction between the conjugated pi system of the aromatic and the adjacent cation.105-106

The energy for this interaction is on the same order of magnitude as a hydrogen-bond and it may be this interaction that allows for shielding of the positive charge to enable cellular uptake.104

The series of complexes (3-Cu through 12-Cu) were compared across a variety of variables, including cellular up-take rate, saturation time, saturation amount, approximate volume of the complex, change in charge, as well as the hydropathy index (HI). The HI was determined using the values of Kyte and Doolittle.107 Out of all of these, there does seem to be a correlation when considering the HI value. As the HI value becomes less negative the longer the time the complex takes to saturate the cell (Figure 3.10E).

Consistently, as the HI increases, the saturation value increases (Figure 3.10A). As 111

suggested by the previous two trends, there is also a correlation between saturation of complex with increasing saturation time (Figure 3.10C).

0.04

0.03

0.02

g/g of lysate) of g/g 0.01



0.00

3-Cu, ( 3-Cu, -0.01

0 3 6 9 12 15 18 21 24 Time (hr)

Figure 3.9. Cellular Uptake for 3-Cu with time. The concentration is expressed in gram of complex per grams of lysate. The concentration of complex was determined by LC- MS/MS based upon a calibration curve and the lysate concentration was determine by Bradford assay. Time points collected were 0, 1, 2, 4, 8, and 16 hr.

It makes sense that the HI would a large impact on cellular uptake. The more negative the HI, the more hydrophilic in nature it becomes. These complexes have a larger charge density associated. This causes issues in crossing the bilayer. A proposed model

(Figure 3.10 top) have these complexes binding to the surface with their hydrophobic portions sticking into the bilayer. These complexes may eventually be able to diffuse across the membrane by formation of aggregates that help disguise the charge. These complexes saturating the surface leads to the lower saturation time and lower accumulation in the cell observed.

112

Figure 3.10. Comparison of different conditions from cellular uptake assay (A) A direct relationship of the saturation of complex in the cell compared to the hydropathy score. (B) The inversion relationship of the initial rate of cellular uptake compared to the hydropathy score. (C) The direct correlation of the saturation within the cell with the half saturation time. (D) The comparison of the half saturation time compared to the initial rate; no correlation is apparent. (E) The comparison of the saturation time compared to the hydropathy score. Each complex is labelled by color as well by the identity; black 3-Cu, red 4-Cu, lime green 5-Cu, blue 6-Cu, cyan 7-Cu, magenta 8-Cu, yellow 9-Cu, light brown 10-Cu, navy blue 11-Cu, purple 12-Cu, dark brown 13-Cu. Hydropathy index was calculated using the value of Kyte and Doolittle.107

113

Complexes with a HI value more positive than -8 are more hydrophobic and are able to pass through the bilayer with greater ease (Figure 3.10 bottom). This is seen in the saturation values for 4-Cu, 6-Cu, and 7-Cu which are lower estimates as result of the complexes not leveling after the 16 hrs.

Figure 3.11. Suggested model of cellular uptake and accumulation in the cell assuming an energy independent pathway. (Top) Complexes that are more hydrophilic, a larger negative hydropathy index less than -10, (green shapes) accumulate on the outside of the cell membrane as they are unable to cross the bilayer as result of their charge causing quicker saturation of the complex. These complexes may eventually be able to cross once aggregation occurs. (Bottom) Complexes that are more hydrophobic, with less negative hydropathy index, greater than -6 (orange shapes) are able to pass through the bilayer into the cytosol allowing for additional accumulation of the complex within the cell.

114

3.4 – Conclusions

A series of complexes based upon the reported complexes 1-Cu and 2-Cu were synthesized and characterized for reactivity towards the HCV IRES RNA SLIIb.

Complexes 3-Cu through 12-Cu were shown to have decreased catalytic efficiency compared to the previously reported complexes, which stemmed mostly from a decrease in the KM. Furthermore, complexes 3-Cu through 7-Cu cleavage products from time- dependent MALDI-TOF MS were determined. From this, 3-Cu and 4-Cu displayed the highest reactivity with the sites of reactivity of the tested complexes agreeing comparably with the docking sites from in silico simulation. Additionally, complexes 3-Cu through 13-

Cu were screened for cellular uptake in a liver cancer line. The initial uptake rate was comparable across the complexes, ~10 ng of complex per gram of cell lysate per hour. The saturation levels of the complexes varied among the complex and showed a correlation to the hydropathy index value. Complexes with a hydropathy index of -10 or more negative reach saturation of ~100 ng of complex per gram of cell lysate with the saturation half point of less than 10 hrs. This is compared to the complexes with a hydropathy index more positive than -6 reaching a saturation of at least 200 ng of complex per gram of cell lysate and a saturation half point of greater than 20 hrs. This once again demonstrates the importance of not only considering in vitro results but also the in vivo properties.

115

3.5 – Supplemental Figures and Tables

Table 3.8. Mass list of expected products for RNA cleavage.

Mass Position Overhang Mass Position Overhang 344.211 1 2',3'-cyclic phosphate 240.184 34 5'aldehyde 689.421 2 2',3'-cyclic phosphate 545.364 33 5'aldehyde 994.601 3 2',3'-cyclic phosphate 890.574 32 5'aldehyde 1323.811 4 2',3'-cyclic phosphate 1196.744 31 5'aldehyde 1669.021 5 2',3'-cyclic phosphate 1525.954 30 5'aldehyde 1998.231 6 2',3'-cyclic phosphate 1832.124 29 5'aldehyde 2327.441 7 2',3'-cyclic phosphate 2177.334 28 5'aldehyde 2656.651 8 2',3'-cyclic phosphate 2506.544 27 5'aldehyde 3001.861 9 2',3'-cyclic phosphate 2812.714 26 5'aldehyde 3307.041 10 2',3'-cyclic phosphate 3118.884 25 5'aldehyde 3652.251 11 2',3'-cyclic phosphate 3464.094 24 5'aldehyde 3958.421 12 2',3'-cyclic phosphate 3769.274 23 5'aldehyde 4263.601 13 2',3'-cyclic phosphate 4114.484 22 5'aldehyde 4569.771 14 2',3'-cyclic phosphate 4459.694 21 5'aldehyde 4898.981 15 2',3'-cyclic phosphate 4765.864 20 5'aldehyde 5244.191 16 2',3'-cyclic phosphate 5095.074 19 5'aldehyde 5549.371 17 2',3'-cyclic phosphate 5400.254 18 5'aldehyde 5854.551 18 2',3'-cyclic phosphate 5705.434 17 5'aldehyde 6183.761 19 2',3'-cyclic phosphate 6050.644 16 5'aldehyde 6489.931 20 2',3'-cyclic phosphate 6379.854 15 5'aldehyde 6835.141 21 2',3'-cyclic phosphate 6686.024 14 5'aldehyde 7180.351 22 2',3'-cyclic phosphate 6991.204 13 5'aldehyde 7485.531 23 2',3'-cyclic phosphate 7297.374 12 5'aldehyde 7830.741 24 2',3'-cyclic phosphate 7642.584 11 5'aldehyde 8136.911 25 2',3'-cyclic phosphate 7947.764 10 5'aldehyde 8443.081 26 2',3'-cyclic phosphate 8292.974 9 5'aldehyde 8772.291 27 2',3'-cyclic phosphate 8622.184 8 5'aldehyde 9117.501 28 2',3'-cyclic phosphate 8951.394 7 5'aldehyde 9423.671 29 2',3'-cyclic phosphate 9280.604 6 5'aldehyde 9752.881 30 2',3'-cyclic phosphate 9625.814 5 5'aldehyde 10059.051 31 2',3'-cyclic phosphate 9955.024 4 5'aldehyde 10404.261 32 2',3'-cyclic phosphate 10260.204 3 5'aldehyde 10709.441 33 2',3'-cyclic phosphate 10605.414 2 5'aldehyde NA 34 2',3'-cyclic phosphate NA 1 5'aldehyde continued

116

Table 3.8. Continued

Mass Position Overhang Mass Position Overhang 404.261 1 3'-enol/aldehyde 258.2 34 5'diol 749.471 2 3'-enol/aldehyde 563.38 33 5'diol 1054.651 3 3'-enol/aldehyde 908.59 32 5'diol 1383.861 4 3'-enol/aldehyde 1214.76 31 5'diol 1729.071 5 3'-enol/aldehyde 1543.97 30 5'diol 2058.281 6 3'-enol/aldehyde 1850.14 29 5'diol 2387.491 7 3'-enol/aldehyde 2195.35 28 5'diol 2716.701 8 3'-enol/aldehyde 2524.56 27 5'diol 3061.911 9 3'-enol/aldehyde 2830.73 26 5'diol 3367.091 10 3'-enol/aldehyde 3136.9 25 5'diol 3712.301 11 3'-enol/aldehyde 3482.11 24 5'diol 4018.471 12 3'-enol/aldehyde 3787.29 23 5'diol 4323.651 13 3'-enol/aldehyde 4132.5 22 5'diol 4629.821 14 3'-enol/aldehyde 4477.71 21 5'diol 4959.031 15 3'-enol/aldehyde 4783.88 20 5'diol 5304.241 16 3'-enol/aldehyde 5113.09 19 5'diol 5609.421 17 3'-enol/aldehyde 5418.27 18 5'diol 5914.601 18 3'-enol/aldehyde 5723.45 17 5'diol 6243.811 19 3'-enol/aldehyde 6068.66 16 5'diol 6549.981 20 3'-enol/aldehyde 6397.87 15 5'diol 6895.191 21 3'-enol/aldehyde 6704.04 14 5'diol 7240.401 22 3'-enol/aldehyde 7009.22 13 5'diol 7545.581 23 3'-enol/aldehyde 7315.39 12 5'diol 7890.791 24 3'-enol/aldehyde 7660.6 11 5'diol 8196.961 25 3'-enol/aldehyde 7965.78 10 5'diol 8503.131 26 3'-enol/aldehyde 8310.99 9 5'diol 8832.341 27 3'-enol/aldehyde 8640.2 8 5'diol 9177.551 28 3'-enol/aldehyde 8969.41 7 5'diol 9483.721 29 3'-enol/aldehyde 9298.62 6 5'diol 9812.931 30 3'-enol/aldehyde 9643.83 5 5'diol 10119.101 31 3'-enol/aldehyde 9973.04 4 5'diol 10464.311 32 3'-enol/aldehyde 10278.22 3 5'diol 10769.491 33 3'-enol/aldehyde 10623.43 2 5'diol NA 34 3'-enol/aldehyde NA 1 5'diol continued

117

Table 3.8. Continued

Mass Position Overhang Mass Position Overhang 282.241 1 3'-OH 242.2 34 5'OH 627.451 2 3'-OH 547.38 33 5'OH 932.631 3 3'-OH 892.59 32 5'OH 1261.841 4 3'-OH 1198.76 31 5'OH 1607.051 5 3'-OH 1527.97 30 5'OH 1936.261 6 3'-OH 1834.14 29 5'OH 2265.471 7 3'-OH 2179.35 28 5'OH 2594.681 8 3'-OH 2508.56 27 5'OH 2939.891 9 3'-OH 2814.73 26 5'OH 3245.071 10 3'-OH 3120.9 25 5'OH 3590.281 11 3'-OH 3466.11 24 5'OH 3896.451 12 3'-OH 3771.29 23 5'OH 4201.631 13 3'-OH 4116.5 22 5'OH 4507.801 14 3'-OH 4461.71 21 5'OH 4837.011 15 3'-OH 4767.88 20 5'OH 5182.221 16 3'-OH 5097.09 19 5'OH 5487.401 17 3'-OH 5402.27 18 5'OH 5792.581 18 3'-OH 5707.45 17 5'OH 6121.791 19 3'-OH 6052.66 16 5'OH 6427.961 20 3'-OH 6381.87 15 5'OH 6773.171 21 3'-OH 6688.04 14 5'OH 7118.381 22 3'-OH 6993.22 13 5'OH 7423.561 23 3'-OH 7299.39 12 5'OH 7768.771 24 3'-OH 7644.6 11 5'OH 8074.941 25 3'-OH 7949.78 10 5'OH 8381.111 26 3'-OH 8294.99 9 5'OH 8710.321 27 3'-OH 8624.2 8 5'OH 9055.531 28 3'-OH 8953.41 7 5'OH 9361.701 29 3'-OH 9282.62 6 5'OH 9690.911 30 3'-OH 9627.83 5 5'OH 9997.081 31 3'-OH 9957.04 4 5'OH 10342.291 32 3'-OH 10262.22 3 5'OH 10647.471 33 3'-OH 10607.43 2 5'OH 10952.651 34 3'-OH 10952.64 1 5'OH continued

118

Table 3.8. Continued

Mass Position Overhang Mass Position Overhang 362.221 1 3'-phosphate 322.18 34 5'Phosphates 707.431 2 3'-phosphate 627.36 33 5'Phosphates 1012.611 3 3'-phosphate 972.57 32 5'Phosphates 1341.821 4 3'-phosphate 1278.74 31 5'Phosphates 1687.031 5 3'-phosphate 1607.95 30 5'Phosphates 2016.241 6 3'-phosphate 1914.12 29 5'Phosphates 2345.451 7 3'-phosphate 2259.33 28 5'Phosphates 2674.661 8 3'-phosphate 2588.54 27 5'Phosphates 3019.871 9 3'-phosphate 2894.71 26 5'Phosphates 3325.051 10 3'-phosphate 3200.88 25 5'Phosphates 3670.261 11 3'-phosphate 3546.09 24 5'Phosphates 3976.431 12 3'-phosphate 3851.27 23 5'Phosphates 4281.611 13 3'-phosphate 4196.48 22 5'Phosphates 4587.781 14 3'-phosphate 4541.69 21 5'Phosphates 4916.991 15 3'-phosphate 4847.86 20 5'Phosphates 5262.201 16 3'-phosphate 5177.07 19 5'Phosphates 5567.381 17 3'-phosphate 5482.25 18 5'Phosphates 5872.561 18 3'-phosphate 5787.43 17 5'Phosphates 6201.771 19 3'-phosphate 6132.64 16 5'Phosphates 6507.941 20 3'-phosphate 6461.85 15 5'Phosphates 6853.151 21 3'-phosphate 6768.02 14 5'Phosphates 7198.361 22 3'-phosphate 7073.2 13 5'Phosphates 7503.541 23 3'-phosphate 7379.37 12 5'Phosphates 7848.751 24 3'-phosphate 7724.58 11 5'Phosphates 8154.921 25 3'-phosphate 8029.76 10 5'Phosphates 8461.091 26 3'-phosphate 8374.97 9 5'Phosphates 8790.301 27 3'-phosphate 8704.18 8 5'Phosphates 9135.511 28 3'-phosphate 9033.39 7 5'Phosphates 9441.681 29 3'-phosphate 9362.6 6 5'Phosphates 9770.891 30 3'-phosphate 9707.81 5 5'Phosphates 10077.061 31 3'-phosphate 10037.02 4 5'Phosphates 10422.271 32 3'-phosphate 10342.2 3 5'Phosphates 10727.451 33 3'-phosphate 10687.41 2 5'Phosphates NA 34 3'-phosphate NA 1 5'Phosphates continued

119

Table 3.8. Continued

Mass Position Overhang 420.261 1 3'-phosphoglycolate 765.471 2 3'-phosphoglycolate 1070.651 3 3'-phosphoglycolate 1399.861 4 3'-phosphoglycolate 1745.071 5 3'-phosphoglycolate 2074.281 6 3'-phosphoglycolate 2403.491 7 3'-phosphoglycolate 2732.701 8 3'-phosphoglycolate 3077.911 9 3'-phosphoglycolate 3383.091 10 3'-phosphoglycolate 3728.301 11 3'-phosphoglycolate 4034.471 12 3'-phosphoglycolate 4339.651 13 3'-phosphoglycolate 4645.821 14 3'-phosphoglycolate 4975.031 15 3'-phosphoglycolate 5320.241 16 3'-phosphoglycolate 5625.421 17 3'-phosphoglycolate 5930.601 18 3'-phosphoglycolate 6259.811 19 3'-phosphoglycolate 6565.981 20 3'-phosphoglycolate 6911.191 21 3'-phosphoglycolate 7256.401 22 3'-phosphoglycolate 7561.581 23 3'-phosphoglycolate 7906.791 24 3'-phosphoglycolate 8212.961 25 3'-phosphoglycolate 8519.131 26 3'-phosphoglycolate 8848.341 27 3'-phosphoglycolate 9193.551 28 3'-phosphoglycolate 9499.721 29 3'-phosphoglycolate 9828.931 30 3'-phosphoglycolate 10135.101 31 3'-phosphoglycolate 10480.311 32 3'-phosphoglycolate 1078.491 33 3'-phosphoglycolate NA 34 3'-phosphoglycolate

120

Figure 3.12. CD titration profiles for 3-Cu through 13-Cu at selected wavelengths.

121

Table 3.9. Expanded docking values showing the contribution of each cluster towards the average value reported in Table 3.2 # in K Binding IM vdw/hb/desolv Electrostatic Total Internal Torsional Complex I cluster (nM) Energy Energy energy Energy Energy Energy 46 5.6 -11.26 -18.12 -12.93 -5.19 -4.3 6.86 65 13.89 -10.72 -17.58 -10.87 -6.71 -4.36 6.86 3-Cu 55 187.23 -9.18 -16.04 -8.83 -7.21 -3.6 6.86 Total: 166 69.03 -10.36 -17.22 -10.76 -6.45 -4.09 6.86 29 75.32 -9.72 -16.28 -10.8 -5.48 -2.47 6.56 63 81.22 -9.67 -16.24 -8.11 -8.13 -2.61 6.56 4-Cu 15 124.6 -9.42 -15.98 -7.74 -8.24 -3.17 6.56 57 151.41 -9.3 -15.41 -8.26 -7.26 -2.62 6.56

Total: 164 108.54 -9.53 -15.93 -8.60 -7.37 -2.64 6.56 122 69 72.38 -9.74 -15.41 -8.19 -7.22 -2.83 5.67 5-Cu 67 248.31 -9.01 -14.68 -9.86 -4.82 -3.58 5.67 Total: 136 159.05 -9.38 -15.05 -9.01 -6.04 -3.20 5.67 80 3.76 -11.49 -17.76 -9.9 -7.86 -1.65 6.26 6-Cu 62 16.63 -10.61 -16.88 -10.15 -6.73 -2.27 6.26 Total: 142 9.38 -11.11 -17.38 -10.01 -7.37 -1.92 6.26 35 9.59 -10.94 -16.31 -12.46 -3.85 -2.9 5.37 90 40.87 -10.08 -15.45 -9.45 -5.4 -4.79 5.37 7-Cu 25 209.75 -9.11 -14.48 -8.2 -6.28 -3.36 5.37 Total: 150 61.72 -10.12 -15.49 -9.94 -5.19 -4.11 5.37 104 43.97 -10.04 -16.9 -8.91 -7.99 -3.32 6.86 8-Cu 57 116.42 -9.46 -16.32 -9.24 -7.08 -4.67 6.86 Total: 161 69.62 -9.835 -16.69 -9.03 -7.67 -3.80 6.86 122

Figure 3.13. Pseudo Michaelis-Menten plots for 3-Cu through 13-Cu.

123

Figure 3.14. Time dependent reactions monitoring the disappearance of full length fluorescein labelled SLIIb with time. Reaction conditions: 750 nM Fl-SLIIb, 750 nM of complex, 1 mM ascorbic acid, 1 mM hydrogen peroxide in HEPES buffer pH 7.4. Lanes from left to right; control, 5 min, 15 min, 30 min, 45 min, 60 min, 90 min, 120 min, 180 min. Gels are comprised of 4 % agarose.

124

Figure 3.15. Plots of the concentration of Fl-SLIIb versus time (in minutes). The concentration is from the quantification of the gels in Figure 3.14.

125

Table 3.10. Matches at the 120 min time point for 3-Cu with SLIIb.

120 min Mass Theor Obsd Peak Error Position Overhang Norm Mass Mass Area (ppm) 2179.35 2179.338 -5.7 28 5'OH 26572.8 2.11E-01 2508.56 2508.636 30.3 27 5'OH 26161.2 2.08E-01 1527.97 1527.783 -122.3 30 5'OH 12438.3 9.90E-02 1834.14 1834.006 -72.8 29 5'OH 11885.9 9.46E-02 5097.09 5097.128 7.5 19 5'OH 6049.8 4.81E-02 2814.73 2814.918 66.8 26 5'OH 5629.5 4.48E-02 4767.88 4768.125 51.4 20 5'OH 4870.3 3.87E-02 3466.11 3466.359 72.0 24 5'OH 4059.7 3.23E-02 5402.27 5402.313 7.9 18 5'OH 2832.6 2.25E-02 3120.9 3121.000 32.2 25 5'OH 2255.2 1.79E-02 6993.22 6993.562 49.0 13 5'OH 767.1 6.10E-03 1669.021 1669.131 65.8 5 2',3'-cPO4 5918.4 4.71E-02 3001.861 3001.809 -17.4 9 2',3'-cPO4 5207.6 4.14E-02 5244.191 5243.858 -63.6 16 2',3'-cPO4 2636.9 2.10E-02 2656.651 2656.459 -72.1 8 2',3'-cPO4 2503.3 1.99E-02 1998.231 1998.486 127.5 6 2',3'-cPO4 2492.3 1.98E-02 3652.251 3652.156 -26.0 11 2',3'-cPO4 1650.0 1.31E-02 2327.441 2327.554 48.4 7 2',3'-cPO4 1470.9 1.17E-02 5854.551 5853.609 -160.9 18 2',3'-cPO4 294.0 2.34E-03

126

Table 3.11. Matches at the 120 min time point for 4-Cu with SLIIb.

120 min Mass Theor Obsd Peak Error Position Overhang Norm Mass Mass Area (ppm) 2508.56 2508.833 109.0 27 5'OH 29952.6 2.46E-01 5097.09 5097.099 1.7 19 5'OH 21622.5 1.77E-01 2179.35 2179.639 132.5 28 5'OH 4102.8 3.37E-02 5707.45 5708.497 183.5 17 5'OH 627.8 5.15E-03 10952.64 10952.518 -11.1 1 5'OH 150.0 1.23E-03 4645.821 4645.926 22.5 14 3'-PG 3550.7 2.91E-02 6507.941 6506.921 -156.7 20 3'-phosphate 317.7 2.61E-03 1607.051 1607.194 89.1 5 3'-OH 3506.2 2.88E-02 1936.261 1936.089 -89.1 6 3'-OH 2706.0 2.22E-02 10952.651 10952.518 -12.1 34 3'-OH 150.0 1.23E-03 4569.771 4569.749 -4.8 14 2',3'-cPO4 26909.0 2.21E-01 5854.551 5854.653 17.4 18 2',3'-cPO4 8391.5 6.89E-02 3307.041 3307.114 22.0 10 2',3'-cPO4 4421.2 3.63E-02 3001.861 3001.781 -26.5 9 2',3'-cPO4 3981.3 3.27E-02 5549.371 5549.308 -11.4 17 2',3'-cPO4 3691.6 3.03E-02 2327.441 2327.473 13.8 7 2',3'-cPO4 2626.1 2.15E-02 3958.421 3958.087 -84.3 12 2',3'-cPO4 2367.3 1.94E-02 4263.601 4263.807 48.4 13 2',3'-cPO4 2304.9 1.89E-02 3652.251 3651.975 -75.5 11 2',3'-cPO4 2144.6 1.76E-02

127

Table 3.12. Matches at the 120 min time point for 5-Cu with SLIIb.

120 min Mass Theor Obsd Peak Error Position Overhang Norm Mass Mass Area (ppm) 2179.350 2179.263 -40.1 28 5'OH 11792.7 1.98E-01 2508.560 2508.531 -11.7 27 5'OH 9331.7 1.57E-01 4767.880 4767.868 -2.6 20 5'OH 7568.7 1.27E-01 5097.090 5096.979 -21.8 19 5'OH 5034.8 8.47E-02 1527.970 1527.840 -85.3 30 5'OH 4427.0 7.45E-02 5402.270 5402.099 -31.6 18 5'OH 2806.2 4.72E-02 1834.140 1834.002 -75.1 29 5'OH 2369.1 3.99E-02 3466.110 3465.935 -50.5 24 5'OH 2301.7 3.87E-02 2814.730 2814.672 -20.5 26 5'OH 2032.0 3.42E-02 5707.450 5707.548 17.1 17 5'OH 1074.1 1.81E-02 4461.710 4461.736 5.8 21 5'OH 857.2 1.44E-02 3120.900 3120.955 17.6 25 5'OH 682.1 1.15E-02 3771.290 3771.342 13.7 23 5'OH 623.4 1.05E-02 8624.200 8623.988 -24.6 8 5'OH 272.8 4.59E-03 8953.410 8952.813 -66.7 7 5'OH 82.3 1.38E-03 7299.390 7298.199 -163.2 12 5'OH 26.3 4.43E-04 8951.394 8952.813 158.5 7 5'aldehyde 82.3 1.38E-03 7297.374 7298.199 113.1 12 5'aldehyde 26.3 4.43E-04 7642.584 7642.108 -62.3 11 5'aldehyde 12.6 2.12E-04 7944.831 7944.491 -42.7 24 3'-a-b 21.8 3.66E-04 1669.021 1669.001 -12.1 5 2',3'-cPO4 3725.5 6.27E-02 3001.861 3001.687 -58.1 9 2',3'-cPO4 1967.2 3.31E-02 1998.231 1998.361 65.2 6 2',3'-cPO4 848.4 1.43E-02 2656.651 2656.539 -42.0 8 2',3'-cPO4 846.8 1.42E-02 2327.441 2327.293 -63.5 7 2',3'-cPO4 620.5 1.04E-02 8136.911 8136.223 -84.5 25 2',3'-cPO4 3.2 5.33E-05

128

Table 3.13. Matches at the 120 min time point for 6-Cu with SLIIb.

120 min Mass Theor Obsd Peak Error Position Overhang Norm Mass Mass Area (ppm) 2508.560 2508.464 -38.4 27 5'OH 64599.7 2.83E-01 2179.350 2179.023 -150.0 28 5'OH 63111.8 2.76E-01 10952.640 10954.075 131.0 1 5'OH 8729.6 3.82E-02 5707.450 5707.682 40.7 17 5'OH 7630.3 3.34E-02 6381.870 6381.904 5.4 15 5'OH 7509.1 3.29E-02 6688.040 6687.894 -21.8 14 5'OH 6539.3 2.86E-02 8624.200 8623.048 -133.6 8 5'OH 4970.5 2.18E-02 4461.710 4462.196 109.0 21 5'OH 4838.4 2.12E-02 3120.900 3120.970 22.6 25 5'OH 4571.5 2.00E-02 8953.410 8952.826 -65.2 7 5'OH 3943.1 1.73E-02 4116.500 4116.862 88.0 22 5'OH 3820.0 1.67E-02 6993.220 6992.941 -39.8 13 5'OH 3326.8 1.46E-02 3771.290 3771.634 91.3 23 5'OH 3182.2 1.39E-02 6052.660 6052.273 -64.0 16 5'OH 2206.5 9.67E-03 7644.600 7644.775 22.9 11 5'OH 1333.5 5.84E-03 9282.620 9281.252 -147.3 6 5'OH 1146.8 5.02E-03 7299.390 7298.978 -56.4 12 5'OH 802.9 3.52E-03 8294.990 8293.868 -135.3 9 5'OH 696.7 3.05E-03 8622.184 8623.048 100.2 8 5'aldehyde 4970.5 2.18E-02 8951.394 8952.826 160.0 7 5'aldehyde 3943.1 1.73E-02 9280.604 9281.252 69.9 6 5'aldehyde 1146.8 5.02E-03 8292.974 8293.868 107.8 9 5'aldehyde 696.7 3.05E-03 10727.451 10727.639 17.5 33 3'-phosphate 30.0 1.31E-04 10952.651 10954.075 130.0 34 3'-OH 8729.6 3.82E-02 3652.251 3652.493 66.2 11 2',3'-cPO4 5625.7 2.46E-02 3307.041 3307.160 35.8 10 2',3'-cPO4 3697.2 1.62E-02 3958.421 3958.525 26.3 12 2',3'-cPO4 1651.8 7.24E-03 4898.981 4898.645 -68.5 15 2',3'-cPO4 1490.0 6.53E-03 4569.771 4569.827 12.4 14 2',3'-cPO4 1195.3 5.24E-03 4263.601 4263.677 17.7 13 2',3'-cPO4 1143.1 5.01E-03 5854.551 5853.639 -155.8 18 2',3'-cPO4 1079.6 4.73E-03

129

Table 3.14. Matches at the 120 min time point for 7-Cu with SLIIb.

120 min Mass Theor Obsd Peak Error Position Overhang Norm Mass Mass Area (ppm) 2588.520 2588.288 -89.6 27 5'-phosphate 987.9 1.07E-02 2259.310 2258.973 -149.1 28 5'-phosphate 413.6 4.48E-03 3546.070 3546.287 61.2 24 5'-phosphate 231.7 2.51E-03 3200.860 3200.678 -56.8 25 5'-phosphate 222.6 2.41E-03 2894.690 2895.090 138.2 26 5'-phosphate 177.1 1.92E-03 3851.250 3851.253 0.9 23 5'-phosphate 128.9 1.40E-03 2179.330 2179.319 -5.1 28 5'-OH 27524.5 2.98E-01 2508.540 2508.575 14.0 27 5'-OH 20721.4 2.25E-01 1527.950 1527.776 -113.6 30 5'-OH 13447.8 1.46E-01 1834.120 1833.998 -66.6 29 5'-OH 7956.7 8.63E-02 2814.710 2814.757 16.6 26 5'-OH 4358.9 4.73E-02 4767.860 4767.793 -14.1 20 5'-OH 3462.2 3.75E-02 5097.070 5096.922 -29.1 19 5'-OH 2825.9 3.06E-02 3466.090 3466.215 36.1 24 5'-OH 1988.9 2.16E-02 3120.880 3120.864 -5.2 25 5'-OH 1245.7 1.35E-02 5402.250 5402.265 2.8 18 5'-OH 954.7 1.04E-02 4461.690 4461.772 18.3 21 5'-OH 390.5 4.23E-03 5707.430 5707.302 -22.4 17 5'-OH 299.1 3.24E-03 3771.270 3771.357 23.1 23 5'-OH 295.7 3.21E-03 4116.480 4116.492 2.9 22 5'-OH 285.7 3.10E-03

130

131

Figure 3.16. Normalized intensity for 3-Cu with SLIIb at each position with time. Shown are the peaks that consistently show up at each time point. 131

132

Figure 3.17. Normalized intensity for 4-Cu with SLIIb at each position with time. Shown are the peaks that consistently show up at each time point. 132

133

Figure 3.18. Normalized intensity for 5-Cu with SLIIb at each position with time. Shown are the peaks that consistently show up at each time point. 133

134

Figure 3.19. Normalized intensity for 6-Cu with SLIIb at each position with time. Shown are the peaks that consistently show up at each time point. 134

135

Figure 3.20. Normalized intensity for 7-Cu with SLIIb at each position with time. Shown are the peaks that consistently show up at each time point. 135

Figure 3.21. Determination of relative initial rates (normalized intensity/min) from time- dependent MALDI-TOF MS reactions of nucleic acids with increasing amount of product detected. This data is represented in the 3-dimensional figure of Figure 3.6 and Table 3.5.

136

Figure 3.22. Cellular uptake of 3-Cu through 12-Cu with time. The y-axis is in g of complex detected by LC-MS/MS per gram of cell lysate as determined by Bradford assay.

137

Chapter 4: Binding and Initial Reactivity of Various Metal Ions Complexed to GGHYrFK with an In-depth Characterization of Mechanism with GGHYrFK-Cu with HCV SLIIb IRES RNA

4.1 - Introduction

The hepatitis C virus (HCV) affects approximately 200 million people worldwide and 2% here in the United States108. Advances to treat HCV have been made with the inclusion of sofosbuvir (Gilead). However, treatments for antivirals are now moving towards cocktails as opposed to single drug treatment. Even the treatment of sofosbuvir is used in conjugation with ledipasvi (trade name Harvoni®). The advantage of cocktails are to be able to attack the target on multiple fronts and as a result prevent the disease sufficient time to form resistance. While effective, concerns have been made about the cost for treatment. An article in June 2015, discusses this issue with the concern of Medicare dropping the coverage until a patient has already suffered liver damage before being approved.99 Previously we have reported the compound GGHYrFK-Cu (1-Cu) and its ability to target SLIIb of the HCV IRES RNA. Also reported in literature are compounds with the ATCUN (amino terminal copper and nickel) motif bound to nickel39, cobalt109, gold48-49, 110, palladium48-49, and platinum48-49. Of past interest in the bioinorganic medicinal field are compounds with coordinated gold and platinum, as can be seen through the clinical and FDA approved drugs cisplatin1, 18 and auranofin111.112-113 Auranofin is currently under investigation for its antimicrobial properties.114-116 These metal complexes

138 are of interest as they allow insight into the role of the oxidation state, inertness of the metal ion, mass to charge ratio, and redox potential in relation to reactivity.

4.2 – Materials

Cobalt (II) Chloride was acquired from Baker and Adamson. Cobalt (III) hexamine chloride and 3-hydroxypicolinic acid were obtained from ACROS. Potassium tetrachloroaurate was from Alfa Aesar. The peptide GGHYrFK was purchased from

Neopeptide and used without further purification. Palladium (II) chloride, platinum (II) chloride, methanol, 30% hydrogen peroxide, sodium chloride, trizma, acrylamide, N, N’- methylenebisacrylamide, and diethyl pyrocarbonate (DEPC) were from Sigma Aldrich.

Ascorbic acid, urea, sodium hydroxide, formamide, ammonium hydroxide were from

Fisher. The IRES SLIIb RNA: 5’-fluorescein-GGC AGA AAG CGU CUA GCC AUG

GCG UUA GUA UGC C-3 was purchased from Dharmacon.

All reactions were carried out in nano-pure water (18.2 ) with autoclaved pipette tips and Eppendorf’s™. Reactions involving RNA samples were in buffered solutions that were treated with 0.1% DEPC water.

Hepes buffer refers to 20 mM Hepes, 100 mM sodium chloride in 0.1% DEPC treated water pH 7.4. Phosphate buffer refers to 20 mM sodium phosphate, 100 mM sodium chloride in 0.1% DEPC treated water at pH 7.4.

4.3 – Synthesis

4.3.1 Complexation of Au (III), Pd (II), and Pd (II) with GGHYrFK

The complexation of the respective metal ion to the peptide was carried out in similar fashion as described elsewhere.48-49 The peptide concentration was determined by

139 measuring the absorbance of the peptide at 275 nm ( = 1400 M-cm-) in water. Equivalent molar ratios of the peptide and metal salt (5.53 x 10-6 mol) were mixed and stirred at 40 oC for 24 hours. In all cases, the solution of the peptide alone was adjusted to pH of 5.5 with the use of 1 M ammonium hydroxide before addition of the metal salt. Upon addition of the palladium (II) chloride and platinum (II) chloride the pH did not require additional adjustment. With the addition of the potassium tetrachloroaurate (III), additional 600 L of 1 M ammonium hydroxide was required. Solutions were then dried under vacuum.

Crude products were resuspended in water before purification by HPLC. A reverse phase

XTERRA C18 column was used. A gradient of 0% acetonitrile to 60% acetonitrile over 35 minutes used to separate the complex from the free peptide. ESI traces are available in the supplementary material as Figure 4.11. Yield: 1-Au: 15%, 1-Pd: 59%, 1-Pt: 61%.

4.3.2 Synthesis of Co (III) with GGHYrFK (1-Co)

Cobalt (II) chloride hexahydrate (13.8 mg, 5.79 x 10-5 mol) and the peptide

GGHYrFK-amide (55.2 mg, 6.39 x 10-5 mol) were dissolved in 1 mL of methanol. The solution rapidly changed a blue to a pink solution upon mixing the solution. The solution was stirred for an additional hour. The methanol was dried by vacuum leaving behind a blue solid, which was then redissolved in 1 mL of aqueous ammonium hydroxide. The solution turned to an orange/red solution almost instantaneously but was stirred for additional two hours. The product was separated from cobalt hexamine with four equivalents of acetonitrile. The solution was spun down to produce a yellowish solution and a purplish solid. The yellowish solution was kept and dried under vacuum. The yellowish product was kept dried and wrapped in aluminum foil in the -20 oC until used where a few milligrams were removed. Upon dissolving the product in water the pH of the

140 solution was slightly basic (pH ~8). ESI m/z 307, 3+, 410, 2+, and 920, 1+ peaks were observed. (ESI trace available in Figure 4.11)

4.4 – Methods

4.4.1 Quantification of stock complexes

Ten L of a stock solution of the complex was added to 140 L of buffer. The absorbance intensity at 274 nm, tyrosine, for all the complexes was monitored by serial

- - dilution to determine the concentration of the stock solution (274= 1400 M cm ). Nickel and copper solutions were calibrated by titration with a fixed amount of peptide whose concentration was predetermined by the tyrosine absorbance at 274 nm. Plots of the quantification is available in the supplementary material (Figure 4.12).

4.4.2 Circular Dichroism Titrations

A JASCO J-815 circular dichroism spectrophotometer was used. A starting solution of 2.5 M solution of SLIIb in phosphate buffer in a cuvette was initially screened and to this solution, serial addition of ~1-2 L of 40 M of the metal-peptide complex was titrated every 5-7 min, and the wavelength was scanned from 400 to 200 nM.

4.4.3 Isothermal Calorimetry

A 5 M solution of SLIIb in a volume of 1.8 mL in Hepes buffer was loaded into the Microcal ITC cell. To this aliquots of either 4 L, for the first 10 points, or 6 L, for the last 20 points, of 100 M of the respective metal-peptide was added. The data was fit to a two site binding model for all complexes, except for 1-Pt which was fit to a sequential site binding. All fits were performed within Origin software.

141

4.4.4 Reaction monitored by agarose gels

Reactions were set up with 10 L per time point. Time points were staggered at 10,

30, 60, 120, 180, 240, 300, 360, 1500 min. Reaction contained 10 M of Fl-SLIIb, 10 M of the corresponding complex with or without 1 mM ascorbic acid and hydrogen peroxide.

Reaction mixtures were loaded on to a 4% agarose gel and ran for 15-20 min before imaging by use of a Bio-Rad Gel Doc camera. Gels were then quantified in QuantityOne.

Concentration at individual time points were determine by comparing the intensity of the control lane. This fraction was multiplied by 10 M to determine the amount of RNA at the time point. Plots of concentration of Fl-SLIIb versus time were then constructed in

Origin and fitted to an exponential decay from which initial rates and consumption of RNA was determined.

4.4.5 LC-MS of SLIIb with 1-Cu

Four different conditions relating to the reactivity of 1-Cu with SLIIb were performed each with a reaction volume of 50 l. In all cases, the final concentration of 1-

Cu and SLIIb was 50 M. Aliquots of 100 M SLIIb were dried and resuspended in the appropriate water solution, either DI water or 18O water. The first condition was carried out aerobically in the presence of 1 mM ascorbic acid and hydrogen peroxide. The second condition was carried out the same, but anaerobically under an argon atmosphere. The third condition was similar to the second condition, but the reaction was carried out in 18O labelled water. The final condition was with 1-Cu and SLIIb under an argon atmosphere with 18O labelled water. Reactions were allowed to proceed for 2 hours before being flash frozen in advance of separation by LC-MS.

142

LC-MS was carried out in a collaboration with the Campus Chemical Instrument

Center, CCIC, at Ohio State University. The LC-MS was calibrated with a mixture of eight oligonucleotides from Bruker (Table 4.9 in the supplementary). Reaction mixtures were zip-tipped as follows. First the zip tip was washed 3x in 50/50 (v/v) mixture of water and acetonitrile. This was followed by rinsing the tip 3x with 2 M TEAA. The sample was then cycled through the tip 10x while ensuring no air is introduced into the sample. The tip was rinsed 3x with TEAA and 3x with water. Finally, the sample was eluted in 5 L by cycling the zip tip 10x in a 50/50 mixture of water/acetonitrile. The reaction mixture was separated using a 1 x 100 mm Waters X Bridge BEH C18 reverse phase column. A two solution system was used as described, 5 mM hexylammonium acetate at pH 7 (solution A) and

100% acetonitrile (solution B). The following gradient profile was used for the separation.

Initially a 5 minute loading at 20% solution B. This was followed by increasing solution B to 60% over 20 min (3%/min gradient). Finally to ensure all the reaction mixture was removed, solution B was increased to 90% B over 4 minutes (7.5%/min gradient).

4.5 – Results and Discussion

4.5.1 Binding Assays

Circular dichroism was performed to assay the binding of the various metal complexes to IRES SLIIB of HCV RNA. Dissociation constants (KD) for the complexes were determine in the range from 18 to 600 nM (Table 4.1). Binding to SLIIb RNA was determined by monitoring the change in the CD signal at least two wavelengths (263 nm and usually 240 nm). The binding curves are shown in Figure 4.1 along with a summary of the values in Table 4.1. The free peptide, 1, had an average KD of 66 nM which is comparable to the previously reported value of 44 nM;62 determined by a fluorescence

143 assay following tyrosine emission. This shows that the sensitivity of this assay is comparable to the tyrosine emission assay.

In general, complexes with first row transition metals enhanced the binding, second row transition metals were approximately the same as the free peptide, and finally the third row transition metals have negatively impacted the binding affinity. This holds true for the

Ni and Cu group metals used. The binding affinity for 1-Ni2+, Pd2+, Pt2+, increases from

24.1 nM, 89.6 nM to 380.6 nM respectfully, whereas the Cu2+ and Au3+ complexes, increase from 18.6 nM to 186 nM. The outlier in this situation is the 1-Co3+ complex which has the worst affinity of all the compounds tested. The reason for this may be the full coordination sphere that it contains. This in conjunction with the cobalt being a low-spin d6 ion in this complex, which makes it extremely stable, and as such the axial water ligands are not favorable to exchange and as a result produce additional steric constraints when it binds. For 1-Pd, 1-Au, 1-Pt, these compounds contain d8 metals ions which adopt a square planar geometry.

144

Figure 4.1. Binding curve with corresponding Mn+-GGHYrFK complex and SLIIb at a given wavelength as measured by Circular Dichroism. The individual additions of the 1- Mn+ to SLIIb is available in the supplementary as Figure 4.14.

145

Table 4.1. Dissociation constants as determined by circular dichroism and isothermal titration calorimetry.

b KD CD KD1 ITC Summary (nM) (nM)

1 66.3 ± 1.4 a 20.7 a

1-Au 186.6 ± 44.3 130

1-Co 568.6 ± 27.2 592

1-Cu 18.7 ± 0.8 29.9

1-Ni 24.1 ± 3.7 125

1-Pd 89.6 ± 20.6 245

1-Pt 380.6 ± 9.8 490

a A previous value of 44 nM had been reported based upon a fluorescence based assay. b Values are average from at least 2 different wavelengths.

4.5.2 Crystal Structure Trends

Previously, crystal structures have been solved for GlyGlyHis-Cu39, GlyGlyHis-

Pd48, and GlyGlyHis-Au48. Figure 4.2 presents the solved crystal structures. A summary of selected bond angles and bond distances are available in Table 4.2. Comparing the three, the ionic radius of the metal ion increases from Cu(II), to Au(III), to Pd(II). This is a result of the charge to radius ratio decreasing across the series and the extension of the d-orbitals.

The radius of Au and Pd atoms are comparable, but the higher oxidation state of Au(III) makes the ionic radius smaller than Pd (II). With increased radius, the peptide chelator

146 adjusts to accommodate the larger metal ion. This primarily occurs through an increase in

n+ o o o angle 1 (Figure 4.3), N1-M -N2; 79.3 , 82.8 , 83.4 respectfully for Cu (II)-GGH, Au (III)-

n+ o GGH, and Pd (II)-GGH. There is a corresponding decrease in angle 4, N4-M -N1; 102.6 ,

100.4o, 99.6o respectfully. To comment briefly on the Ni2+, Pt2+, and Co3+ complexes

2+ (complexes without a solved crystal structure), Ni would be expected to be similar to

Cu2+. Similarly, Pt2+ would expected to be similar to Pd2+. The ionic radii of the second and third row transition metals are relative similar in size and as such should have very similar structure. Co(III) would be expected to be smaller in ionic radius than Cu2+ as a

n+ result of a higher oxidation state and would result in smaller N1-M -N2 and a larger N4-

n+ M -N1 angle.

Figure 4.2. Crystal structures of GGH-Mn+ where Mn+ is Cu2+ (brown sphere, left), Au3+ 2+ 2+ (yellow sphere, middle), Pd (orange sphere, right). Ni would be expected to be similar to Cu2+. Similarly, Pt2+ would expected to be similar to Pd2+. The Co3+ would expect to be 2+ n+ smaller in ionic radius than Cu and would result in smaller N1-M -N2 and a larger N4- n+ M -N1 angle.

147

Table 4.2. Selected bond lengths and angles for crystal structures of Cu (II)-GGH, Pd (II)- GGH, and Au (III)-GGH. Figure 4.3 illustrates the atoms of selected in the table.

Cu2+-GGH Pd2+-GGH Au3+-GGH

Bond lengths (Å)

n+ N1-M 2.0 2.1 2.0

n+ N2-M 2.0 1.9 1.9

n+ N3-M 1.9 2.0 2.0

n+ N4-M 2.0 2.0 2.0

Angles (degree)

n+ N1-M -N2 79.3 83.4 82.8

n+ N2-M -N3 82.5 82.1 83.5

n+ N3-M -N4 95.2 94.9 93.3

n+ N4-M -N1 102.6 99.6 100.4

148

Figure 4.3. A model of Mn+-GGH with highlighted atoms that are discussed in Table 4.2.

4.5.3 Entropic and enthalpic contributions and binding comparison

Isothermal titration calorimetry was performed with the 1-Mn+ series and SLIIb by serial addition of the metal-complex was added to the SLIIb solution in the cell. Insights into the binding parameters were obtained and the parameters are listed in Table 4.3. The individual traces are available in the supplementary as Figures 4.15. Figure 4.4 contains an example ITC trace for 1-Cu.

Comparison of the dissociation constant, KD1, from ITC (the inverse of the association constant, KA1) with the calculated KD from CD, show relative consistency between the two values. Some variation is noticed in the 1-Ni and 1-Pd value; 24.1 nM versus 125 nM and 89.6 nM versus 245 nM (CD compared to ITC) respectively. For all the complexes, except for 1-Pt, a two-site binding was the best model obtained. A single- site as well as sequential binding were consider. This is not surprising, previous work with

1-Cu suggests there are two distinct sites for binding (Chapter 2). In the case of 1-Pt, a

149 sequential binding model was the preferred, suggesting that the first binding site causes a reorganization of the RNA to accommodate the second binding site.

Trends emerge in the enthalpic contributions to binding. As result of the inertness of 1-Co it will not be included in the general trends discussed. For all the complexes, the enthalpy of binding is negative and on the range of -6 to 0 kcal/mol. The enthalpy of binding decreases, less exothermic, from the first row transition metals, with an average of

-4.8 kcal/mol, to second row, -3.4 kcal/mol, and to the third row transition metals, -1.2 kcal/mol. The enthalpy would be expected to decrease, become more positive, as the ionic radii increases. Similarly as the charge increases, the enthalpy would also be expected to be more exothermic. This is highlighted between 1-Au and 1-Pt. Au has a charge of 3+ compared to Pt’s charge of 2+ which translates to a -1.6 kcal/mol compared to -0.8 kcal/mol enthalpy of binding. Complex 1-Co had the lowest enthalpy of binding. This is understandable. With a full coordination sphere, the metal ion in the complex does not interact with SLIIb during the binding. Further, due to its bulk as a result of the axial ligands, it shows decreased binding affinity compared to the free peptide.

An inverse trend is observed for the entropic contribution. The first row transition metals, TM, now have the lowest entropic contribution, 16.8 cal/mol/oC. The second row

TM, is 18.7 cal/mol/ oC, followed by the third row TM and 1-Co, 26.3 cal/mol/ oC and 27.6 cal/mol/ oC respectfully. It is expected that the larger the ionic radii, the greater the entropy contribution would be. As the metal center increases, it is able to occupy a larger space and displace more water molecules and in exchange increase the entropy for the system.

150

4.5.4 Reactivity

Reactivity profile for the different metal complexes were obtained by agarose gel experiment using a fluorescein-labelled SLIIb. Reactions were allowed to proceed up to 24 hrs. An example plot of reactivity is available in Figure 4.5 (1-Cu) with the remaining in the supplementary as Figure 4.17. Table 4.4 summarizes the initial rate as well as the percent of SLIIb that was consumed during the reaction.

151

Table 4.3. ITC derived values: binding affinities, enthalpic and entropic terms.

KA1 KD1 H1 S1 KA2 KD2 H2 S2 Complex (M-1) (nM) (kcal/mol) (cal/mol/ oC) (M-1) (M) (kcal/mol) (cal/mol/ oC)

1 4.82 x 107 20.7 -2.9 25.6 7.77 x 105 1.3 -1.3 22.7

1-Au 7.86 x 106 130 -1.6 26.2 1.78 x 105 5.6 1.9 17.7

1-Co 1.69 x 106 592 -0.3 27.6 1.39 x 105 14 -2.1 27.6 152

1-Cu 3.35 x 107 29.9 -5.6 15.6 1.76 x 106 0.6 -1.3 24.3

1-Ni 1.25 x 107 125 -4.1 17.9 1.14 x 105 8.8 -1.7 17.4

1-Pd 4.08 x 106 245 -3.4 18.7 4.81 x 105 208 -0.8 23.3

1-Pt a 2.04 x 106 490 -0.8 26.3 1.00 x 106 1.0 -0.1 27.2

a All complexes were fit to a two site model except 1-Pt. 1-Pt preferred a two-site sequential.

152

0.0

-1.0

-2.0

kcal/mole of injectant -3.0

-4.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 Molar Ratio

Figure 4.4. ITC profile of 1-Cu into SLIIb . The data is fit to a two site binding model.



[HCV Fl-SLIIb] [HCV Fl-SLIIb] 2

0 0 5 10 15 20 25 Time (hr)

Figure 4.5. Reactivity of 1-Cu towards Fl-SLIIb in the presence of ascorbate and hydrogen peroxide monitored up to 24 hours. Reaction conditions: 10 M Fl-SLIIb, 10 M 1-Cu, 1 mM ascorbic acid and hydrogen peroxide in Hepes buffer, pH 7.4.

153

Table 4.4. Initial Rate and consumption parameters from agarose gels Initial Rate Mn+-GGHYrFK % Fl-SLIIb consumed nM/min Au (III) 3.3 ± 1.1 23.5 ± 2.4 Pd (II) 3.9 ± 0.9 30.2 ± 2.0 Pt (II) 2.2 ± 0.3 100 ± 1.5 Co (III) 4.8 ± 1.6 19.4 ± 3.6 None a 6.0 ± 1.0 10.0 ± 2.4 Co (III) + coreagents b 4.0 ± 2.3 20.0 ± 3.9 Cu (II) + coreagents b 1.8 ± 0.28 41.8 ± 1.6 Ni (II) + coreagents b ND c ND c a. No metal present, reaction preformed with the metal free peptide b. Coreagents imply in the presence of 1 mM ascorbic acid and hydrogen peroxide c. Not determined

Comparing the results of Table 4.4, several trends between the initial rate and percent consumed emerge. Conditions with the largest initial rate, 1 and 1-Co with or without coreagents, consumed the least amount of fluorescein-labelled SLIIb, max of 20% of the RNA. Inversely, those with the slowest rate, 1-Cu and 1-Pt, degraded the most SLIIb,

42% to 100% respectively. This could also arise as result of stability of the metal-complex.

4.5.5 Mechanistic hypothesis of 1-Pt and 1-Au

1-Cu has been reported before to degrade SLIIb (Chapter 2), however, it is surprising it was the slowest in reactivity compared to the other complexes tested. The tested complexes ranged in initial rate of 1.8 nM/min to 6.0 nM/min. Intriguing is the 1-Pt complex, which appears to consume all SLIIb if allowed to continue to completion. This complex merits further investigation into the mechanism of action.

154

The Pt(II) metal center would be expected to be unreactive with the d8 configuration and the square planar conformation. It has been shown by Giandomenico et al. for the formation of Pt(IV) prodrugs through the use of hydrogen peroxide.117 For this case, no hydrogen peroxide was present, therefore, the Pt (II) derivate would be expected. A possible mechanism is one where 1-Pt binds the RNA and one of the Pt(II) axial positions is oriented towards a phosphate group or 2’-hydroxyl; activing as a Lewis acid pulling electron density away and weakening the respective bond. This weakened bond is then attacked by a nucleophile, most likely water in this scenario, and cleaves SLIIb. Additional evidence would be necessary to support this model.

Au(III) complexes are of particular interest since they are isoelectronic to Pt(II) which has been one of the most successful ions used in therapy; cisplatin for example. The derivative of Au(III)-GGH has been reported before and is stable for at least 48 hours with no degradation or change in oxidation state.110 This has been one issue of Au(III) complexes where the reduction to Au (0) can occur.111 Further Au(III)-GGH complexes have shown select toxicity towards different lines of cells. One in vitro growth inhibition assay has shown Au (III)-GGH to be more effective than cisplatin in HeLa (human cervix carcinoma) and HL-60 (human promyelocytic leukemia) cell lines, IC50 value of 0.0045

M versus 2.02 M and 2.98 M versus 10.31 M respectfully for Au (III)-GGH and cisplatin.110

The mechanism of Au (III) complexes towards nucleic acids degradation is limited.

One of the few reports, describes the ability of Au (III) in the presence of Hepes buffer to form a nitrogen based radical which is the active species against plasmid DNA. Au(III) is reduced to Au(I) and eventually precipitates to form nanoparticles.118 In cases, where the

155

Au(III) is complexed and stabilized, no reduction is observed and another mechanism must be proposed.119 Additionally, studies with these complexes also suggest the binding to

DNA to be weak, reversible and mostly electrostatic.119-120

The differences between the previously reported compounds and the one reported is here is the incorporation of a targeting domain which will seek out the HCV SLIIb RNA.

From the ITC and CD assay, there is clear binding, although the binding affinity is not as great as for 1-Cu or 1. Furthermore from the gel assays, reactivity is seen in the decrease intensity of the fluorescein-labelled SLIIb RNA (Table 4.4 and Figure 4.16 and 4.17; the latter two are in the supplementary). This complex is most likely acting through a Lewis acid interaction, promoting hydrolysis of SLIIb. Further research however is needed to determine the mechanism.

4.5.6 LC-MS of 1-Cu with SLIIb

Reactions under oxidative conditions (in the presence of ascorbic acid and hydrogen peroxide) and under aerobic conditions were initially performed to obtain a baseline profile for the reactivity of 1-Cu with SLIIb. Next, the same reaction was performed under anaerobic conditions, argon atmosphere, to determine if dioxygen was a species responsible for RNA cleavage. Comparison of Figure 4.6, aerobic oxidative (Figure

4.6C) and anaerobic oxidative (Figure 4.6B), show the chromatograms to have similar profiles. As result of this result, the reaction was carried out under anaerobic conditions with heavy water present (Figure 4.6A). This was used to determine whether or not water was being incorporated into the reaction products, or if hydrogen peroxide was the only source of oxygen in the products. Finally, the last reaction of SLIIb with 1-Cu under

156 anaerobic conditions was performed to ensure that the use of co-reagents was indeed required for reactivity to occur.

For the anaerobic oxidative reactions, with and without 18O, each peak was investigated to determine the products present. An expected mass list of products, (Table

4.10 in the supplementary), was used to scan for products. Another product list with the incorporation of 18O (mass = +2 amu) was also used to scan for incorporation of heavy water into the product. The charge of the relative product was determine based upon the separation of the masses as well as the elution time. The elution time was compared with that of the standard oligonucleotides. From there, the amount of hexylammonium acetate,

HAA, adducts were determined, by dividing the difference of the observed peak minus the product peak by the mass of HAA (mass = 101.1 amu). At this point, the mass was compared to the aforementioned mass lists and the part per million (ppm) deviation was calculated. Masses within a ppm deviation of under 400 were assigned as belonging to the product. The sum of intensities for all masses corresponding to the product were summed and the average ppm for the individual masses corresponding to the product was determined (example in Table 4.5). A summary table for each peak with all the products was constructed (example in Table 4.6). In this, the normalized intensity for a product was determined by taking the sum of the intensity for a particular product and dividing it by the total intensities of the products. In the case of the anaerobic reaction with heavy water, a ratio of 18O product to 16O was calculated. Elution times were determine to be consistent between the anaerobic with and without the 18O water; 2.0-2.2 min, 2.6-2.8min, 3.5-3.6 min, 4.7-4.9 min, 6.8-7.3 min, 12.0-12.3 min, 12.6-12.8, 13.1-13.4, 13.6-13.7, 13.9-14.0, and 15.6-16.0 min.

157

One goal of this experiment was to determine where the oxygen that is incorporated into the cleavage product is coming from. This search started with three possible sources, dioxygen from the atmosphere/dissolved in the water, water, or a derivative of hydrogen peroxide. Comparing the data in Table 4.5, the dioxygen can be ruled out; the reactivity profile between aerobic oxidative (B) and anaerobic oxidative (C), are similar. The solution is purged of any oxygen by bubbling with argon. This then leaves the water and hydrogen peroxide as the potential source of oxygen in the products. To ensure that the reactivity does not change with additional 18O water, reaction (C) and the reaction under anaerobic oxidative conditions with heavy water (D) were analyzed and compared. The elution times for respective products are were found to be the same between the two reactions and the relative amounts of each overhang observed is approximately the same Table 4.7.

158

Figure 4.6. Traces of LC-MS chromatogram of various reaction conditions of 1-Cu with 18 SLIIb. (A) Reaction under Argon with coreagents and O H2O (B) Reaction under Argon 18 with coreagents and without O H2O. (C) Aerobic reaction with coreagents and without 18 18 O H2O. (D) Reaction under Argon with no coreagents and O H2O.

159

Table 4.5. Example of a LC-MS product table . Example is for the anaerobic reaction with no 18O water at elution time of 4.7-4.9 min. The top row represents the product that was screen for. Below are the different masses detected that correspond to the product in the top row. Bottom row shows the sum of the intensity of the peaks and the average ppm or deviation for the product. 2',3'-cyclic Product 7 2327.441 phosphate I Obs # of HAA # m/z Res. S/N I FWHM Adj Mass ppm % Mass adducts 43 809.4399 66790 53.7 127 2 0.0121 2428.32 0.99 2327.22 95.1 80 843.1117 33292 93.9 221 3.6 0.0253 2529.335 1.99 2327.135 131.4 81 843.4418 53222 61.6 145 2.4 0.0158 2530.325 2.00 2328.125 -294.1

169 1214.6560 87544 45.2 106 1.7 0.0139 2429.312 1.01 2328.212 -331.3 160

204 1264.6641 32169 221.7 523 8.5 0.0393 2529.328 2.00 2327.128 134.4 205 1265.1701 39358 143 337 5.5 0.0321 2530.34 2.01 2328.14 -300.4 266 1315.2289 36290 121.5 287 4.6 0.0362 2630.458 3.00 2327.158 121.7 267 1315.7303 46628 92.2 217 3.5 0.0282 2631.461 3.01 2328.161 -309.2 304 1365.7796 104818 48.5 114 1.9 0.013 2731.559 4.00 2327.159 121.1 305 1366.2916 82832 43.4 102 1.7 0.0165 2732.583 4.01 2328.183 -318.9 Sum 2179 Average -95.0

160

Table 4.6. Example of data obtained for all the identified products for a particular time point by LC-MS analysis The data is for the anaerobic reaction with no 18O water at elution time of 4.7-4.9 min. Highlighted in red is the example from Table 4.5. All product tables obtained under anaerobic conditions are available in the supplemental as Table 4.11 and Table 4.12.

Elution Fraction Product Average Normalized Product (min) (mass) (ppm) Intensity 4.7-4.9 1914.12 29 5'Phosphates 21.07 0.06

4.7-4.9 1936.261 6 3'-OH 74.05 0.05

4.7-4.9 2177.334 28 5'aldehyde -262.39 0.01

4.7-4.9 2179.35 28 5'OH 332.94 0.01

4.7-4.9 2195.35 28 5'diol 298.03 0.01

4.7-4.9 2259.33 28 5'Phosphates 134.31 0.01 2',3'-cyclic 4.7-4.9 2327.441 7 -95.01 0.03 phosphate 4.7-4.9 2345.451 7 3'-phosphate 347.80 0.01 3'- 4.7-4.9 2387.491 7 -56.96 0.02 enol/aldehyde 4.7-4.9 2506.544 27 5'aldehyde 15.32 0.26

4.7-4.9 2508.56 27 5'OH 106.24 0.17

4.7-4.9 2524.56 27 5'diol 75.85 0.12

4.7-4.9 Unassigned 0.25

161

Table 4.7. Global analysis of products with DI water and 18O water. Determined accumulation from all overhangs from each peak and product. % Overhangs % Overhangs Overhang Detected Deviation w/o 18O water w/ 18O water 2',3'-cyclic phosphate 23.3 29.4 -6.1 3'-phosphoglycaldehyde 8.9 12.6 -3.7 3'-OH 27.2 21.3 +5.9 3'-phosphate 23.6 15.7 +7.9 3'-phosphoglycolate 17.1 21.1 -4.0 5'-aldehyde 33.4 34.8 -1.4 5'-diol 25.1 34.1 -9.0 5'-OH 19.2 10.4 +8.8 5'-Phosphates 22.3 20.7 +1.6

Table 4.8. Incorporation of 18O into determined products by overhang analysis through LC-MS.

Ratio of 18O/16O Overhang % Present Average a Weighted Average b 2',3'-cyclic phosphate 29.376 0.637 0.616 3'-phosphoglycaldehyde 12.552 2.210 1.640 3'-OH 21.291 0.614 0.543 3'-phosphate 15.710 1.726 1.461 3'-phosphoglycolate 21.072 1.213 2.534 5'aldehyde 34.794 0.913 0.702 5'diol 34.050 1.439 1.389 5'OH 10.436 0.694 0.352 5'Phosphates 20.720 0.984 0.822 a. This is the straight average of the ratio of 18O/16O without taking the intensity into consideration b. This considers the relative amount of each individual product by considering the normalized intensity as a weight factor.

162

Figure 4.7. Comparison of output from LC-MS versus MALDI-TOF MS. (Top) Chromatogram showing the elution of products at various time points. This particular chromatogram is for 1-Cu with SLIIb under anaerobic oxidative conditions with 18O labelled water. (Second from top). The actual mass spectrum for the peak that eluted between 4.7-4.8 min. This is then characterized for the products present. (Third from top) Mass spectrum for a reaction between SLIIb and 1-Cu under aerobic and oxidative condition acquired by MALDI-TOF MS as discusses in Chapter 2. (Bottom) Zoomed in region of mass spectrum acquired by MALDI-TOF MS.

163

4.5.7 Overhang analysis and mechanism pathway

Previously, the cleavage profile for 1-Cu with SLIIb was discussed in Chapter 2.

This earlier method used MALDI-TOF mass spectrometry. The peaks present in that method are considerably broader and are lower in resolution (Figure 4.7 bottom). This is in stark contrast to the output of LC-MS. The chromatogram shows separation of peaks allowing for higher level of accuracy and less overlap of products (Figure 4.7 top). In addition, the peaks are of higher resolution with precision to the fourth decimal position

(Figure 4.7 second from top). With this higher resolution additional overhangs can be distinguished such as 5’-aldehyde from the 5’-hydroxyl (mass = 2 amu). The 5’-aldehyde has been reported to arise from 5’-H abstraction.121 A previously unreported overhang is the formation of the 5’-gem diol.

This product may arise through two different mechanisms; an intramolecular and an intermolecular pathways. The intramolecular pathway is seen in Figure 4.8A. After 5’-

H abstraction, a hydroxyl radical attacks the 5’C. The neighboring phosphate extracts a proton from the newly added 5’-hydroxyl group. This yields a 3’-phosphate and 5’- aldehyde. The 5’-aldehyde may undergo hydration by heavy water to produce the 18O labelled 5’-gem diol.

Alternatively after the initial 5’-H abstraction, the 5’-gem diol may arise from an intermolecular attack of a hydroxide, producing the 3’-phosphate and yielding the 5’-gem diol directly without the 5’-aldehyde intermediate (Figure 4.8B). This intermolecular mechanism is not as likely to occur as the intramolecular reaction would expected to be faster.

164

Figure 4.8. Proposed formation of the 5'-aldehyde and 5'-gem diol. The first step is the 5’- H abstraction, a hydroxyl radical then attacks this position. From here two paths are possible. (A) Intramolecular attack of an oxygen on the phosphate group leads to the formation of the 5’-aldehyde. This 5’-aldehyde can then undergo hydration to form the 5’- geminal diol. (B) An intermolecular attack by a hydroxide to the phosphate group, produces a 3’-phosphate and a 5’-geminal diol directly without the formation of the aldehyde. (In box) 18O incorporation arising from tautomerization of the aldehyde to the enol.

From the discussion in Chapter 2, we postulated two different mechanisms in which

1-Cu acts. The first of which is through an oxidative hydrolytic mechanism and the other through H’-abstraction method. Taking into account the different overhangs and the relative abundance of each (Table 4.7), it appears that both mechanisms are plausible again.

The 2’, 3’-cyclic phosphate and the 5’-OH generally arise through a hydrolysis mechanism.

This reaction seems to occur about a third of the time based upon the relative abundance

165 of the overhangs observed. The minor products of hydrolysis would be the 5’-phosphate and 3’-hydroxyl. However, the 5’-phosphate may also arise from hydrogen abstractions at the 3’, 4’ and 5’ positions. In the case of the 3’-H abstraction, the expected 3’-product would be the 3’phosphoglycaldehyde, which accounts for ~10% of the reactivity. For the

4’-H abstraction the expected 3’-product is the 3’-phosphoglycolate accounting for 20%.

Finally, the 5’-H abstraction product would be expected to produce 5’-aldehyde/5’-gem diol and the formation of the 3’-phosphate; which for the 5’-products would account for

~50% of the reactivity. This is consistent with our expectations that the 4’-H and the 5’-H would be the most accessible hydrogens for abstraction.

4.5.8 18O Incorporation

Table 4.8 shows the ratio of 18O incorporation into the various overhangs. Overall, the highest ratio of 18O to 16O in products is seen in 3’-phosphoglycolates, 2.5, followed by 3’-phosphoglycaldehyde, 1.6, 3’-phosphates, 1.5, and 5’-geminal diols, 1.4. All other products had a ratio of less than one suggesting that water is not the primary source of oxygen incorporation in those complexes. Looking at the 3’-phosphates and the 5’- hydroxyl it makes sense that the 3’-phosphates have a larger ratio of incorporation of 18O than the 5’-hydroxyl. There is a far greater chance in the product to form an 18O derivative in the phosphate. Keeping with the oxidative hydrolysis mechanism, the 2’, 3’-cyclic phosphates also have a low incorporation rate. This once again makes sense as the 2’- hydroxyl is normally associated with the formation of the cyclic phosphates. Once again the 5’-product, the 5’-hydroxyl, should not have an incorporated 18O. In the minor product of hydrolysis, the 5’-phosphate and 3’-hydroxyl, the 5’-phosphate would have the greater

166 chance of 18O incorporation as is the case with a value of 0.8 compared to 0.5 for 3’- hydroxyl.

Considering now the mechanisms involving H-abstraction. Starting with the 5’-H abstraction, the unique products are the 5’-aldehyde and 5’-geminal diol (as shown in

Figure 4.8). The aldehyde and the diol had an incorporation of 0.7 and 1.4 respectively. It makes sense that the geminal diol had a larger incorporation rate (2x) compared to that of the aldehyde, as the 5’-gem diol is a hydration product of the 5’-aldehyde. With the 5’- aldehyde, in the proposed mechanism there is no opportunity for heavy water to be incorporated. However, aldehydes can undergo tautomerization to the enol. At this point, heavy water can exchange with the unlabeled hydroxyl group (Figure 4.8). This formation will be impeded by the energy barrier to the enol form.

Continuing around the ribose ring to consider 4’-H abstraction, the unique product would be identified by the 3’-phosphoglycolate as well as the 5’-phosphate. Figure 4.9 shows a predicted mechanism for the formation of these products. Following hydrogen abstraction, a hydroxyl radical attacks. This is followed by proton abstraction by a base in the area (could be water), which eventually leads to the removal of the 2’-hydroxyl group.

Heavy water may then attack the carbon (C4 position) of the newly formed ester, displacing the remainder of what was the original ribose ring and yielding the 3’-phosphoglycolate.

Further degradation of the 5’-product will yield a 5’-phosphate. Recalling the 3’- phosphoglycolate had the highest incorporation of 18O, this mechanism also accounts for this by having water attack the C4 position.

The 3’-H abstraction leads to the unique product formation of the 3’- phosphoglyaldehyde as well as a 5’-phosphate. A proposed mechanism of formation is

167 available in Figure 4.10. Following 3’-H abstraction, a hydroxyl radical attacks the site. A base removes the hydrogen of this hydroxyl group, opening the ribose ring and displacing the 2’-hydroxyl group. This yields in the formation of the 3’-phosphoglycaldehyde (PGA) and a 5’-phosphodiester with a propene-base adduct. This 5’-product can be attacked by water to produce a 5’-phosphate and basepropenol. In this mechanism, there is no clear method of 18O incorporation into the 3’-phosphoglcaldehyde. However, like the 5’- aldehyde, the 3’-aldehyde can undergo tautomerization to the enol form. The enol in this situation is less constrained then the 5’-aldehyde. This accounts for the data collected which had a ratio of 1.6.

Figure 4.9. Proposed mechanism of 4’-H abstraction to yield the 3’-phosphoglycolate (3’- PG), a unique product for 4’H abstraction. Further degradation of the 5’-product may result in the formation of the 5’-phosphate.

168

Figure 4.10. Proposed mechanism of 3'-H abstraction to yield the 3'-phosphoglycaldehyde (3'-PGA), a unique product for 3'-H abstraction. Further degradation of the 5’-product will result in the formation of the 5’-phosphate. (In box) 18O incorporation arising from tautomerization of the aldehyde to the enol.

4.6 – Conclusion A series of metal derivates based upon 1-Cu have been synthesized and characterized. These include Co (III), Ni (II), Pd (II), Pt (II), and Au (III). All of these compounds showed binding in the 10-600 nM range towards SLIIb of the HCV IRES RNA through two independent assays, CD and ITC. From ITC, most of the complexes showed two site binding which is consistent with what has been reported before for 1-Cu.

Reactivity towards SLIIb was monitored by agarose gels. Most of the complexes had limited reactivity towards cleavage of the SLIIb. This is the first time a Au(III) complex has been used with targeted design towards nucleic acid. Two notable compounds did stand out in reactivity assays, 1-Cu and 1-Pt. Further investigation into the products of 1-Pt and

1-Au would be beneficial to support mechanism of action.

169

Previously 1-Cu reactivity towards SLIIb had been described (Chapter 2).

However, this analysis was limited as result of the method of MALDI-TOF MS. This previous research was expanded through the use of LC-MS and heavy water (18O).

Analysis of the 18O incorporation allowed further insights into the source of the oxygen that is incorporated into the products of 1-Cu with SLIIb. Results showed that 18O was added to products of 3’-phosphates, 3’-phosphoglycolate and 5’-gem diol. The latter two products require the water for a nucleophilic attack after initial chemistry has occurred.

One overhang, the 3’-aldehyde, did not correlate with the ratio obtained. The ratio indicated that 18O should be able to be incorporated. This is achieved through 18O exchange when the aldehyde tautomerizes to the enol form. Expanding upon the previous work, additional support to the mechanism of action for 1-Cu has been developed. 1-Cu primarily acts through H-abstraction at the 5’ (40-50%), 4’ (20%), and 3’ (10%) positions. For the remaining 20-30%, the catalyst was found to promote oxidative hydrolytic cleavage of

SLIIb. Also reported for the first the formation of a 5’-gem diol which is a unique product of 5’-H abstraction.

170

4.7 – Supplemental Material

Figure 4.11. ESI-MS of Mn+-GGHYrFK. 1-Au (A), 1-Pd (B), 1-Pt (C), 1-Co (D).

171

Figure 4.12. Quantification of 1-Mn+ complexes by UV/Vis monitoring the tyrosine absorbance . Mn+ = Au (III), Pd (II) and Pt (II)

172

Table 4.9. Oligo standards used for LC-MS. Included are the elution times for the method used. Figure 4.13 contains the LC-MS chromatogram for these standards as well as the control of SLIIb alone.

Oligonucleotide Mass Elution Time (min) Oligo 4 1173.8 2.25 min Oligo 5 1487.0 2.75 min Oligo 7 2105.4 3.50 min Oligo 9 2722.8 7.50 min Oligo 11 3341.2 12.2 min Oligo 12 3645.4 13.0 min Oligo 20 6177.0 15 min Oligo 30 9191.0 16 min

173

Figure 4.13. Control LC-MS chromatograms. (Top) Injection contains the 8 standard oligonucleotides. (Bottom) Injection of only the SLIIb RNA.

174

Figure 4.14. CD spectra over the range of 350 nm to 200 nm, with increasing amount of 1-Mn+ added to 2.5 M SLIIb.

175

Figure 4.15. ITC traces for 1-Mn+ complexes with SLIIb.

176

Figure 4.16. Gel Assays of 1-Mn+ with fluorescein labelled SLIIb. Time increases from left to right with the complex in the assay on the far left.

177

Figure 4.17. Plots of concentration of fluorescein labelled SLIIb versus time. Quantification of gels from Figure 4.16.

178

Table 4.10. Mass list of expected products of SLIIb. 2’, 3’- cPO4, 2’, 3’-cyclic phosphates; 3’-PO4, 3’-phosphates; 3’-PGA, 3’-phosphoglycaldehyde; 3’-PG, 3’-phosphoglycolate; 5’-PO4, 5’-phosphate

Mass (amu) Position Overhang Mass (amu) Position Overhang 344.211 1 2',3'-cPO4 NA 1 5'-aldehyde 689.421 2 2',3'-cPO4 10605.414 2 5'-aldehyde 994.601 3 2',3'-cPO4 10260.204 3 5'-aldehyde 1323.811 4 2',3'-cPO4 9955.024 4 5'-aldehyde 1669.021 5 2',3'-cPO4 9625.814 5 5'-aldehyde 1998.231 6 2',3'-cPO4 9280.604 6 5'-aldehyde 2327.441 7 2',3'-cPO4 8951.394 7 5'-aldehyde 2656.651 8 2',3'-cPO4 8622.184 8 5'-aldehyde 3001.861 9 2',3'-cPO4 8292.974 9 5'-aldehyde 3307.041 10 2',3'-cPO4 7947.764 10 5'-aldehyde 3652.251 11 2',3'-cPO4 7642.584 11 5'-aldehyde 3958.421 12 2',3'-cPO4 7297.374 12 5'-aldehyde 4263.601 13 2',3'-cPO4 6991.204 13 5'-aldehyde 4569.771 14 2',3'-cPO4 6686.024 14 5'-aldehyde 4898.981 15 2',3'-cPO4 6379.854 15 5'-aldehyde 5244.191 16 2',3'-cPO4 6050.644 16 5'-aldehyde 5549.371 17 2',3'-cPO4 5705.434 17 5'-aldehyde 5854.551 18 2',3'-cPO4 5400.254 18 5'-aldehyde 6183.761 19 2',3'-cPO4 5095.074 19 5'-aldehyde 6489.931 20 2',3'-cPO4 4765.864 20 5'-aldehyde 6835.141 21 2',3'-cPO4 4459.694 21 5'-aldehyde 7180.351 22 2',3'-cPO4 4114.484 22 5'-aldehyde 7485.531 23 2',3'-cPO4 3769.274 23 5'-aldehyde 7830.741 24 2',3'-cPO4 3464.094 24 5'-aldehyde 8136.911 25 2',3'-cPO4 3118.884 25 5'-aldehyde 8443.081 26 2',3'-cPO4 2812.714 26 5'-aldehyde 8772.291 27 2',3'-cPO4 2506.544 27 5'-aldehyde 9117.501 28 2',3'-cPO4 2177.334 28 5'-aldehyde 9423.671 29 2',3'-cPO4 1832.124 29 5'-aldehyde 9752.881 30 2',3'-cPO4 1525.954 30 5'-aldehyde 10059.051 31 2',3'-cPO4 1196.744 31 5'-aldehyde 10404.261 32 2',3'-cPO4 890.574 32 5'-aldehyde 10709.441 33 2',3'-cPO4 545.364 33 5'-aldehyde NA 34 2',3'-cPO4 240.184 34 5'-aldehyde continued

179

Table 4.10 Continued

282.241 1 3'-OH NA 1 5'-diol 627.451 2 3'-OH 10623.43 2 5'-diol 932.631 3 3'-OH 10278.22 3 5'-diol 1261.841 4 3'-OH 9973.04 4 5'-diol 1607.051 5 3'-OH 9643.83 5 5'-diol 1936.261 6 3'-OH 9298.62 6 5'-diol 2265.471 7 3'-OH 8969.41 7 5'-diol 2594.681 8 3'-OH 8640.2 8 5'-diol 2939.891 9 3'-OH 8310.99 9 5'-diol 3245.071 10 3'-OH 7965.78 10 5'-diol 3590.281 11 3'-OH 7660.6 11 5'-diol 3896.451 12 3'-OH 7315.39 12 5'-diol 4201.631 13 3'-OH 7009.22 13 5'-diol 4507.801 14 3'-OH 6704.04 14 5'-diol 4837.011 15 3'-OH 6397.87 15 5'-diol 5182.221 16 3'-OH 6068.66 16 5'-diol 5487.401 17 3'-OH 5723.45 17 5'-diol 5792.581 18 3'-OH 5418.27 18 5'-diol 6121.791 19 3'-OH 5113.09 19 5'-diol 6427.961 20 3'-OH 4783.88 20 5'-diol 6773.171 21 3'-OH 4477.71 21 5'-diol 7118.381 22 3'-OH 4132.5 22 5'-diol 7423.561 23 3'-OH 3787.29 23 5'-diol 7768.771 24 3'-OH 3482.11 24 5'-diol 8074.941 25 3'-OH 3136.9 25 5'-diol 8381.111 26 3'-OH 2830.73 26 5'-diol 8710.321 27 3'-OH 2524.56 27 5'-diol 9055.531 28 3'-OH 2195.35 28 5'-diol 9361.701 29 3'-OH 1850.14 29 5'-diol 9690.911 30 3'-OH 1543.97 30 5'-diol 9997.081 31 3'-OH 1214.76 31 5'-diol 10342.291 32 3'-OH 908.59 32 5'-diol 10647.471 33 3'-OH 563.38 33 5'-diol 10952.651 34 3'-OH 258.2 34 5'-diol continued

180

Table 4.10 Continued

362.221 1 3'-PO4 10952.64 1 5'-OH 707.431 2 3'-PO4 10607.43 2 5'-OH 1012.611 3 3'-PO4 10262.22 3 5'-OH 1341.821 4 3'-PO4 9957.04 4 5'-OH 1687.031 5 3'-PO4 9627.83 5 5'-OH 2016.241 6 3'-PO4 9282.62 6 5'-OH 2345.451 7 3'-PO4 8953.41 7 5'-OH 2674.661 8 3'-PO4 8624.2 8 5'-OH 3019.871 9 3'-PO4 8294.99 9 5'-OH 3325.051 10 3'-PO4 7949.78 10 5'-OH 3670.261 11 3'-PO4 7644.6 11 5'-OH 3976.431 12 3'-PO4 7299.39 12 5'-OH 4281.611 13 3'-PO4 6993.22 13 5'-OH 4587.781 14 3'-PO4 6688.04 14 5'-OH 4916.991 15 3'-PO4 6381.87 15 5'-OH 5262.201 16 3'-PO4 6052.66 16 5'-OH 5567.381 17 3'-PO4 5707.45 17 5'-OH 5872.561 18 3'-PO4 5402.27 18 5'-OH 6201.771 19 3'-PO4 5097.09 19 5'-OH 6507.941 20 3'-PO4 4767.88 20 5'-OH 6853.151 21 3'-PO4 4461.71 21 5'-OH 7198.361 22 3'-PO4 4116.5 22 5'-OH 7503.541 23 3'-PO4 3771.29 23 5'-OH 7848.751 24 3'-PO4 3466.11 24 5'-OH 8154.921 25 3'-PO4 3120.9 25 5'-OH 8461.091 26 3'-PO4 2814.73 26 5'-OH 8790.301 27 3'-PO4 2508.56 27 5'-OH 9135.511 28 3'-PO4 2179.35 28 5'-OH 9441.681 29 3'-PO4 1834.14 29 5'-OH 9770.891 30 3'-PO4 1527.97 30 5'-OH 10077.061 31 3'-PO4 1198.76 31 5'-OH 10422.271 32 3'-PO4 892.59 32 5'-OH 10727.451 33 3'-PO4 547.38 33 5'-OH NA 34 3'-PO4 242.2 34 5'-OH continued

181

Table 4.10 Continued.

404.261 1 3'-PGA NA 1 5'-PO4 749.471 2 3'-PGA 10687.41 2 5'-PO4 1054.651 3 3'-PGA 10342.2 3 5'-PO4 1383.861 4 3'-PGA 10037.02 4 5'-PO4 1729.071 5 3'-PGA 9707.81 5 5'-PO4 2058.281 6 3'-PGA 9362.6 6 5'-PO4 2387.491 7 3'-PGA 9033.39 7 5'-PO4 2716.701 8 3'-PGA 8704.18 8 5'-PO4 3061.911 9 3'-PGA 8374.97 9 5'-PO4 3367.091 10 3'-PGA 8029.76 10 5'-PO4 3712.301 11 3'-PGA 7724.58 11 5'-PO4 4018.471 12 3'-PGA 7379.37 12 5'-PO4 4323.651 13 3'-PGA 7073.2 13 5'-PO4 4629.821 14 3'-PGA 6768.02 14 5'-PO4 4959.031 15 3'-PGA 6461.85 15 5'-PO4 5304.241 16 3'-PGA 6132.64 16 5'-PO4 5609.421 17 3'-PGA 5787.43 17 5'-PO4 5914.601 18 3'-PGA 5482.25 18 5'-PO4 6243.811 19 3'-PGA 5177.07 19 5'-PO4 6549.981 20 3'-PGA 4847.86 20 5'-PO4 6895.191 21 3'-PGA 4541.69 21 5'-PO4 7240.401 22 3'-PGA 4196.48 22 5'-PO4 7545.581 23 3'-PGA 3851.27 23 5'-PO4 7890.791 24 3'-PGA 3546.09 24 5'-PO4 8196.961 25 3'-PGA 3200.88 25 5'-PO4 8503.131 26 3'-PGA 2894.71 26 5'-PO4 8832.341 27 3'-PGA 2588.54 27 5'-PO4 9177.551 28 3'-PGA 2259.33 28 5'-PO4 9483.721 29 3'-PGA 1914.12 29 5'-PO4 9812.931 30 3'-PGA 1607.95 30 5'-PO4 10119.101 31 3'-PGA 1278.74 31 5'-PO4 10464.311 32 3'-PGA 972.57 32 5'-PO4 10769.491 33 3'-PGA 627.36 33 5'-PO4 NA 34 3'-PGA 322.18 34 5'-PO4 continued

182

Table 4.8. Continued

420.261 1 3'-PG 765.471 2 3'-PG 1070.651 3 3'-PG 1399.861 4 3'-PG 1745.071 5 3'-PG 2074.281 6 3'-PG 2403.491 7 3'-PG 2732.701 8 3'-PG 3077.911 9 3'-PG 3383.091 10 3'-PG 3728.301 11 3'-PG 4034.471 12 3'-PG 4339.651 13 3'-PG 4645.821 14 3'-PG 4975.031 15 3'-PG 5320.241 16 3'-PG 5625.421 17 3'-PG 5930.601 18 3'-PG 6259.811 19 3'-PG 6565.981 20 3'-PG 6911.191 21 3'-PG 7256.401 22 3'-PG 7561.581 23 3'-PG 7906.791 24 3'-PG 8212.961 25 3'-PG 8519.131 26 3'-PG 8848.341 27 3'-PG 9193.551 28 3'-PG 9499.721 29 3'-PG 9828.931 30 3'-PG 10135.101 31 3'-PG 10480.311 32 3'-PG 10785.491 33 3'-PG NA 34 3'-PG

183

Table 4.11. Summary of products at each peak from the reaction of 1-Cu with SLIIb under 18 anaerobic conditions without O, H2O. Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 15.7-16.0 9279.596 6 5'aldehyde -19.2 0.07 15.7-16.0 9298.620 6 5'diol -21.7 0.07 15.7-16.0 9625.814 5 5'aldehyde 88.9 0.03 15.7-16.0 9627.830 5 5'OH -69.2 0.05 15.7-16.0 9643.830 5 5'diol -51.8 0.06 15.7-16.0 9690.911 30 3'-OH -9.6 0.06 15.7-16.0 9707.810 5 5'Phosphates 93.5 0.05 15.7-16.0 9812.931 30 3'-enol/aldehyde -3.9 0.04 15.7-16.0 9828.931 30 3'-phosphoglycolate -38.4 0.05 15.7-16.0 9955.024 4 5'aldehyde 28.6 0.04 15.7-16.0 9973.040 30 3'-phosphate -9.2 0.06 15.7-16.0 9999.081 31 3'-OH -11.1 0.05 15.7-16.0 10037.020 4 5'Phosphates 11.3 0.05 2',3'-cyclic 15.7-16.0 10059.051 31 -5.3 0.03 phosphate 15.7-16.0 10077.061 31 3'-phosphate 0.7 0.06 15.7-16.0 10952.651 Full length SLIIb -1.7 0.06 15.7-16.0 Unknown 0.16

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 13.9-14.0 4114.484 22 5'aldehyde -30.2 0.05 13.9-14.0 4116.500 22 5'OH -36.3 0.04 13.9-14.0 4132.500 22 5'diol -28.6 0.08 13.9-14.0 4196.480 22 5'Phosphates 232.0 0.06 2',3'-cyclic 13.9-14.0 4263.601 13 phosphate 84.1 0.05 13.9-14.0 4459.694 21 5'aldehyde 256.4 0.08 13.9-14.0 4477.710 21 5'diol -25.9 0.04 13.9-14.0 4509.801 14 3'-OH -26.2 0.07 2',3'-cyclic 13.9-14.0 4569.771 14 phosphate -79.3 0.06 13.9-14.0 4587.781 14 3'-phosphate -55.5 0.02 13.9-14.0 4765.864 20 5'aldehyde -5.5 0.02 continued

184

Table. 4.11 Continued

13.9-14.0 4783.880 20 5'diol -0.9 0.02 13.9-14.0 4839.011 15 3'-OH 88.1 0.06 13.9-14.0 4959.031 15 3'-aldehyde -185.1 0.02 13.9-14.0 Unassigned 0.32

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 13.6-13.7 3787.290 23 5'diol -12.6 0.09 13.6-13.7 3851.270 23 5'Phosphates 172.7 0.03 13.6-13.7 3958.421 12 2',3'-cyclic phosphate 177.5 0.05 13.6-13.7 3976.431 12 3'-phosphate 297.4 0.03 13.6-13.7 4018.471 12 3'-enol/aldehyde -213.6 0.01 13.6-13.7 4034.471 12 3'-phosphoglycolate -19.9 0.04 13.6-13.7 4114.484 22 5'aldehyde 416.3 0.06 13.6-13.7 4132.500 22 5'diol -9.9 0.03 13.6-13.7 4196.480 22 5'Phosphates 227.5 0.07 13.6-13.7 Unassigned 0.58

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 13.1-13.4 3546.090 24 5'Phosphates 39.8 0.07 13.1-13.4 3590.281 11 3'-OH -1.9 0.05 13.1-13.4 3652.251 11 2',3'-cyclic phosphate -35.7 0.06 13.1-13.4 3670.261 11 3'-phosphate -58.4 0.06 13.1-13.4 3712.301 11 3'-enol/aldehyde 73.2 0.08 13.1-13.4 3728.301 11 3'-phosphoglycolate -54.9 0.04 13.1-13.4 3769.274 23 5'aldehyde -2.7 0.04 13.1-13.4 3771.290 23 5'OH -89.9 0.05 13.1-13.4 3787.290 23 5'diol 28.6 0.04 13.1-13.4 3851.270 23 5'Phosphates -23.8 0.06 13.1-13.4 3896.451 12 3'-OH -70.3 0.02 13.1-13.4 3958.421 12 2',3'-cyclic phosphate -77.6 0.03 13.1-13.4 Unassigned 0.40 continued

185

Table 4.11 Continued

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 12.6-12.8 3077.911 9 3'-phosphoglycolate -98.3 0.05 12.6-12.8 3118.884 25 5'aldehyde 23.0 0.11 12.6-12.8 3120.900 25 5'OH -55.8 0.06 12.6-12.8 3136.900 25 5'diol -104.6 0.08 12.6-12.8 3200.880 25 5'phosphate 120.4 0.05 12.6-12.8 3245.071 10 3'-OH 24.5 0.02 12.6-12.8 3307.041 10 2',3'-cyclic phosphate 194.4 0.03 12.6-12.8 3367.091 10 3'-enol/aldehyde 159.8 0.03 12.6-12.8 3383.091 10 3'-phosphoglycolate -66.9 0.09 12.6-12.8 3464.094 24 5'aldehyde 412.3 0.03 12.6-12.8 3546.090 24 5'Phosphates 98.5 0.05 12.6-12.8 3590.281 11 3'-OH -78.3 0.04 12.6-12.8 Unassigned 0.0 0.35

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 12.0-12.3 2812.714 26 5'aldehyde 185.8 0.05 12.0-12.3 2814.730 26 5'OH -152.0 0.05 12.0-12.3 2830.730 26 5'diol -60.4 0.08 12.0-12.3 2894.710 26 5'Phosphates -162.7 0.09 12.0-12.3 2939.891 9 3'-OH -29.7 0.03 12.0-12.3 3077.911 9 3'-phosphoglycolate -146.2 0.03 12.0-12.3 3118.884 25 5'aldehyde -56.4 0.05 12.0-12.3 3136.900 25 5'diol -59.2 0.04 12.0-12.3 3200.880 25 5'Phosphates -78.6 0.04 12.0-12.3 3245.071 10 3'-OH -168.4 0.03 12.0-12.3 3383.091 10 3'-phosphoglycolate -34.3 0.01 12.0-12.3 Unassigned 0.50 continued

186

Table 4.11 Continued

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 6.8-7.3 1936.261 6 3'-OH -84.2 0.10 6.8-7.3 1998.231 6 2',3'-cyclic phosphate -0.3 0.02 6.8-7.3 2016.241 6 3'-phosphate 110.1 0.12 6.8-7.3 2179.334 28 5'OH -234.4 0.03 6.8-7.3 2195.350 28 5'diol -129.7 0.04 6.8-7.3 2259.330 28 5'Phosphates 362.0 0.01 6.8-7.3 2265.471 7 3'-OH 58.6 0.05 6.8-7.3 2327.441 7 2',3'-cyclic phosphate 219.4 0.04 6.8-7.3 2403.491 7 3'-phosphoglycolate 224.9 0.05 6.8-7.3 2506.544 27 5'aldehyde -55.6 0.15 6.8-7.3 2588.540 27 5'Phosphates 312.3 0.04 6.8-7.3 Unassigned 0.35

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 4.7-4.9 1914.120 29 5'Phosphates 21.1 0.06 4.7-4.9 1936.261 6 3'-OH 74.1 0.05 4.7-4.9 2177.334 28 5'aldehyde -262.4 0.00 4.7-4.9 2179.350 28 5'OH 332.9 0.00 4.7-4.9 2195.350 28 5'diol 298.0 0.01 4.7-4.9 2259.330 28 5'Phosphates 134.3 0.01 4.7-4.9 2327.441 7 2',3'-cyclic phosphate 90.4 0.03 4.7-4.9 2345.451 7 3'-phosphate 347.8 0.01 4.7-4.9 2387.491 7 3'-enol/aldehyde -57.0 0.02 4.7-4.9 2506.544 27 5'aldehyde 15.3 0.26 4.7-4.9 2508.560 27 5'OH 106.2 0.17 4.7-4.9 2524.560 27 5'diol 75.9 0.12 4.7-4.9 Unassigned 0.25

187

Continued

Table 4.11 Continued

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 3.5-3.6 1525.954 30 5'aldehyde 442.9 0.02 3.5-3.6 1527.970 30 5'OH 14.9 0.02 3.5-3.6 1543.970 30 5'diol 1292.5 0.02 3.5-3.6 1607.950 30 5'Phosphates -2.9 0.04 3.5-3.6 1687.031 5 3'-phosphate -104.5 0.07 3.5-3.6 1745.071 5 3'-phosphoglycolate -178.2 0.02 3.5-3.6 1832.124 29 5'aldehyde -55.4 0.04 3.5-3.6 1834.140 29 5'OH 165.0 0.02 3.5-3.6 1850.140 20 5'diol 378.2 0.02 2',3'-cyclic 3.5-3.6 1998.231 6 -93.3 0.11 phosphate 3.5-3.6 2016.241 6 3'-phosphate 294.0 0.05 3.5-3.6 2074.281 6 3'-phosphoglycolate -435.6 0.04 3.5-3.6 2179.350 28 5'OH -73.3 0.09 3.5-3.6 Unassigned 0.43

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 2.6-2.8 1012.611 3 3'-phosphate 135.6 0.03 2.6-2.8 1196.744 31 5'aldehyde 188.3 0.08 2.6-2.8 1261.841 4 3'-OH -196.6 0.04 2.6-2.8 1278.740 31 5'Phosphates -25.5 0.05 2',3'-cyclic 2.6-2.8 1323.811 4 -350.6 0.04 phosphate 2.6-2.8 1341.821 4 3'-phosphate 59.1 0.12 2.6-2.8 1383.861 4 3'-enol/aldehyde 366.9 0.05 2.6-2.8 1399.861 4 3'-phosphoglycolate 286.0 0.04 2.6-2.8 1525.954 30 5'aldehyde -377.2 0.02 2.6-2.8 1527.970 30 5'OH 100.3 0.11 188

2.6-2.8 1543.970 30 5'diol -85.5 0.08 2.6-2.8 1607.051 5 3'-OH -68.1 0.03 2.6-2.8 1607.950 30 5'Phosphates -90.5 0.03 2.6-2.8 Unassigned 0.29 continued

Table 4.11 Continued

Elution Average Normalized Fraction Product (mass) Product (ppm) Intensity (min) 2.0-2.2 863.000 GGHYRFK -157.9 0.02 2.0-2.2 890.574 32 5'aldehyde -237.5 0.02 2.0-2.2 892.590 32 5'OH 225.7 0.01 2.0-2.2 908.590 32 5'diol 26.0 0.03 2.0-2.2 1261.841 4 3'-OH -31.1 0.06 2.0-2.2 1323.811 4 2',3'-cyclic phosphate 47.4 0.08 2.0-2.2 1341.821 4 3'-phosphate 56.1 0.01 2.0-2.2 1525.954 30 5'aldehyde -79.4 0.06 2.0-2.2 1527.970 30 5'OH 47.1 0.04 2.0-2.2 Unassigned 0.69

189

Table 4.12. Summary of products at each peak from the reaction of 1-Cu with SLIIb under 18 anaerobic conditions with O, H2O. Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 15.7-16.0 9279.596 6 5'aldehyde 56.6 0.06 0.56 15.7-16.0 9298.620 6 5'diol 128.5 0.03 0.22 15.7-16.0 9624.806 5 5'aldehyde 295.0 0.01 0.00 15.7-16.0 9643.830 5 5'diol 37.8 0.08 0.77 15.7-16.0 9690.911 30 3'-OH 260.2 0.02 0.00 15.7-16.0 9707.810 5 5'Phosphates 209.8 0.03 0.00 15.7-16.0 9812.931 30 3'-enol/aldehyde 7.8 0.11 0.94 15.7-16.0 9828.931 30 3'-phosphoglycolate 113.8 0.06 0.27 15.7-16.0 9955.024 4 5'aldehyde 257.9 0.03 0.00 15.7-16.0 9973.040 30 3'-phosphate 225.4 0.03 100 15.7-16.0 9999.081 31 3'-OH 121.2 0.05 0.00 15.7-16.0 10037.020 4 5'Phosphates 34.1 0.11 0.78 15.7-16.0 10059.051 31 2',3'-cyclic phosphate -40.2 0.15 0.66 15.7-16.0 10077.061 31 3'-phosphate 13.6 0.11 1.09 15.7-16.0 10952.651 Full length SLIIb 87.6 0.04 15.7-16.0 Unassigned 0.08

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 13.9-14.0 4114.484 22 5'aldehyde -80.1 0.05 1.04 13.9-14.0 4132.500 22 5'diol -128.2 0.21 1.47 13.9-14.0 4196.480 22 5'Phosphates 33.5 0.05 0.54 13.9-14.0 4263.601 13 2',3'-cyclic phosphate 161.3 0.09 0.54 13.9-14.0 4339.651 13 3'-phosphoglycolate 203.2 0.04 0.00 13.9-14.0 4459.694 21 5'aldehyde 27.3 0.05 0.73 13.9-14.0 4477.710 21 5'diol 163.5 0.06 0.00 13.9-14.0 4509.801 14 3'-OH -180.2 0.02 0.00 13.9-14.0 4569.771 14 2',3'-cyclic phosphate -116.1 0.07 0.67 13.9-14.0 4587.781 14 3'-phosphate 144.4 0.04 0.16 13.9-14.0 4765.864 20 5'aldehyde 11.8 0.10 6.98 13.9-14.0 4783.880 20 5'diol 55.5 0.07 0.77 13.9-14.0 4839.011 15 3'-OH -254.8 0.03 100 13.9-14.0 4959.031 15 3'-enol/aldehyde -14.7 0.03 0.82 13.9-14.0 Unassigned 0.08 Continued

190

Table 4.12 Continued

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 13.6-13.7 3787.290 23 5'diol -92.9 0.36 1.74 13.6-13.7 3851.270 23 5'Phosphates -196.6 0.03 0.97 13.6-13.7 3958.421 12 2',3'-cyclic phosphate -54.4 0.03 0.87 13.6-13.7 3976.431 12 3'-phosphate 28.5 0.03 1.07 13.6-13.7 4018.471 12 3'-enol/aldehyde -86.7 0.04 2.06 13.6-13.7 4034.471 12 3'-phosphoglycolate 26.4 0.05 0.23 13.6-13.7 4114.484 22 5'aldehyde 110.7 0.04 1.14 13.6-13.7 4132.500 22 5'diol -116.3 0.13 1.22 13.6-13.7 4196.480 22 5'Phosphates -47.4 0.13 0.56 13.6-13.7 Unassigned 0.16

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 13.1-13.4 3546.090 24 5'Phosphates 43.8 0.05 0.05 13.1-13.4 3590.281 11 3'-OH 122.5 0.04 0.36 13.1-13.4 3652.251 11 2',3'-cyclic phosphate 1.0 0.04 1.00 13.1-13.4 3670.261 11 3'-phosphate -98.0 0.06 2.08 13.1-13.4 3712.301 11 3'-enol/aldehyde 39.5 0.08 0.77 13.1-13.4 3728.301 11 3'-phosphoglycolate 17.5 0.07 0.86 13.1-13.4 3768.266 23 5'aldehyde -38.9 0.09 1.18 13.1-13.4 3787.290 23 5'diol 24.9 0.16 1.86 13.1-13.4 3851.270 23 5'Phosphates -43.8 0.09 0.89 13.1-13.4 3896.451 12 3'-OH -14.7 0.07 1.19 13.1-13.4 3958.421 12 2',3'-cyclic phosphate -35.8 0.06 1.01 13.1-13.4 Unassigned 0.19

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 12.6-12.8 3077.911 9 3'-phosphoglycolate 95.0 0.04 0.00 12.6-12.8 3118.884 25 5'aldehyde 53.7 0.07 1.02 12.6-12.8 3120.900 25 5'OH -22.0 0.03 0.47 12.6-12.8 3136.900 25 5'diol -113.1 0.13 2.19 12.6-12.8 3200.880 25 5'phosphate -81.1 0.03 0.71 continued

191

Table 4.12 Continued

12.6-12.8 3367.091 10 3'-enol/aldehyde -153.1 0.09 3.08 12.6-12.8 3383.091 10 3'-phosphoglycolate -31.9 0.08 0.93 12.6-12.8 3464.094 24 5'aldehyde 46.3 0.07 0.61 12.6-12.8 3590.281 11 3'-OH 4.1 0.04 1.10 12.6-12.8 3546.090 24 5'Phosphates -32.9 0.08 0.91 12.6-12.8 3245.071 10 3'-OH 188.1 0.04 3.54 12.6-12.8 3307.041 10 2',3'-cyclic phosphate 76.8 0.03 0.43

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 11.7-12.4 2830.730 26 5'diol -14.7 0.15 0.44 11.7-12.4 2894.710 26 5'Phosphates 8.7 0.08 0.99 11.7-12.4 2939.891 9 3'-OH 134.8 0.05 1.74 11.7-12.4 3077.911 9 3'-phosphoglycolate 303.6 0.01 0.00 11.7-12.4 3118.884 25 5'aldehyde -84.5 0.08 0.77 11.7-12.4 3136.900 25 5'diol -160.9 0.07 1.62 11.7-12.4 3200.880 25 5'Phosphates 11.6 0.08 0.67 11.7-12.4 3245.071 10 3'-OH 0.9 0.06 0.67 11.7-12.4 3383.091 10 3'-phosphoglycolate 3.9 0.08 0.91 11.7-12.4 Unassigned 0.33

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 6.8-7.3 1936.261 6 3'-OH 126.7 0.04 0.81 6.8-7.3 1998.231 6 2',3'-cyclic phosphate 147.1 0.02 0.00 6.8-7.3 2016.241 6 3'-phosphate -55.8 0.04 0.71 6.8-7.3 2179.334 28 5'OH 138.0 -0.05 0.12 6.8-7.3 2195.350 28 5'diol -111.5 0.05 0.00 6.8-7.3 2259.330 28 5'Phosphates -62.9 0.08 0.33 6.8-7.3 2265.471 7 3'-OH 49.0 0.06 0.53 6.8-7.3 2327.441 7 2',3'-cyclic phosphate -18.6 0.07 0.75 6.8-7.3 2403.491 7 3'-phosphoglycolate 83.3 0.02 0.00 6.8-7.3 2506.544 27 5'aldehyde -64.8 0.41 0.49 6.8-7.3 2588.540 27 5'Phosphates 239.2 0.05 0.39 6.8-7.3 Unassigned 0.21

continued

192

Table 4.12 Continued

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 4.7-4.9 1914.120 29 5'Phosphates -79.4 0.06 1.55 4.7-4.9 1936.261 6 3'-OH -4.5 0.03 0.74 4.7-4.9 2177.334 28 5'aldehyde -63.7 0.01 0.82 4.7-4.9 2179.350 28 5'OH -44.2 0.01 1.16 4.7-4.9 2195.350 28 5'diol -60.7 0.02 0.76 4.7-4.9 2259.330 28 5'Phosphates -91.9 0.01 1.00 4.7-4.9 2327.441 7 2',3'-cyclic phosphate -91.3 0.03 0.28 4.7-4.9 2345.451 7 3'-phosphate -74.5 0.01 0.98 4.7-4.9 2387.491 7 3'-enol/aldehyde -169.7 0.01 6.64 4.7-4.9 2506.544 27 5'aldehyde -31.1 0.51 0.35 4.7-4.9 2508.560 27 5'OH 3.3 0.16 0.20 4.7-4.9 2524.560 27 5'diol -78.9 0.03 0.50 4.7-4.9 Unassigned 0.13

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 3.5-3.6 1525.954 30 5'aldehyde 171.2 0.01 0.75 3.5-3.6 1543.970 30 5'diol 130.3 0.02 1.04 3.5-3.6 1607.950 30 5'Phosphates 55.8 0.03 2.04 3.5-3.6 1687.031 5 3'-phosphate 75.6 0.03 0.90 3.5-3.6 1745.071 5 3'-phosphoglycolate -6.1 0.03 1.00 3.5-3.6 1832.124 29 5'aldehyde -5.3 0.01 1.14 3.5-3.6 1834.140 29 5'OH -26.1 0.02 1.96 3.5-3.6 1850.140 20 5'diol -122.3 0.01 0.32 3.5-3.6 1998.231 6 2',3'-cyclic phosphate 28.5 0.09 1.30 3.5-3.6 2016.241 6 3'-phosphate 58.3 0.02 0.32 3.5-3.6 2058.281 6 3'-enol/aldehyde -74.1 0.02 1.98 3.5-3.6 2074.281 6 3'-phosphoglycolate 2.9 0.14 125.09 3.5-3.6 2179.350 28 5'OH 5.8 0.25 0.03 3.5-3.6 Unassigned 0.31

continued

193

Table 4.12 Continued

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 2.6-2.8 1261.841 4 3'-OH -435.5 0.09 2.18 2.6-2.8 1278.740 31 5'Phosphates -96.3 0.02 4.83 2.6-2.8 1323.811 4 2',3'-cyclic phosphate -94.7 0.04 2.31 2.6-2.8 1341.821 4 3'-phosphate 151.8 0.09 0.88 2.6-2.8 1383.861 4 3'-enol/aldehyde 99.2 0.02 1.38 2.6-2.8 1399.861 4 3'-phosphoglycolate 90.6 0.04 0.36 2.6-2.8 1525.954 30 5'aldehyde -90.7 0.06 4.97 2.6-2.8 1527.970 30 5'OH 44.0 0.10 0.92 2.6-2.8 1543.970 30 5'diol 35.5 0.09 0.97 2.6-2.8 1607.051 5 3'-OH 372.8 0.02 0.39 2.6-2.8 1607.950 30 5'Phosphates 344.3 0.02 0.52 2.6-2.8 1196.744 31 5'aldehyde 207.8 0.05 0.85 2.6-2.8 Unassigned 0.35

Elution Product Average Normalize Ratio of Fraction Product (mass) (ppm) Intensity 18O/16O (min) 2.0-2.2 863.000 GGHYRFK 220.8 0.03 0.00 2.0-2.2 890.574 32 5'aldehyde 707.0 0.01 0.00 2.0-2.2 908.590 32 5'diol 329.0 0.02 >10 2.0-2.2 1261.841 4 3'-OH -205.1 0.02 0.44 2.0-2.2 1323.811 4 2',3'-cyclic phosphate -72.7 0.23 0.80 2.0-2.2 1341.821 4 3'-phosphate 108.0 0.04 0.81 2.0-2.2 0 Unassigned 1.00 0.00

194

Chapter 5: Catalytic Metallodrugs Based on the LaR2C Peptide Target HCV SLIV IRES RNA 5.1 – Introduction

Treatment of hepatitis C virus (HCV) infections is limited by the lack of an effective vaccine and toxicity as well as the development of resistance. To circumvent resistance problems, HCV therapy is expected to evolve towards multidrug therapy consisting of drugs with complementary mechanisms of action.122 The catalytic metallodrugs described herein act on a target distinct from current HCV designed drugs

(protease inhibitors, polymerase inhibitors, and the general antivirals interferon and ribavirin), as well as functioning by a novel mechanism, and so they have the potential to complement other therapeutic treatments under development.

Previous reports of catalytic metallodrugs has demonstrated the potential of this drug design strategy.55, 59, 69, 123-127 The HCV internal ribosomal entry site (IRES) is a structured RNA present at the beginning of the viral mRNA and is important for the translation of viral proteins.128 The IRES initiates translation by recruiting proteins to form a complex that binds to the ribosome. Translation results in the synthesis of a single polyprotein precursor that is processed by the host and its viral proteases to form mature viral proteins for virus assembly. Structurally, the HCV IRES consists of 4 domains with distinct functions that lead to the initiation of translation (Figure 1 with stem-loop 1 omitted). HCV IRES RNA has been previously demonstrated to be important for the life cycle of the hepatitis C virus. The metallopeptide Cu-GGHYrFK-amide (1-Cu, Figure 2),

195 consisting of a YrFK-amide motif for RNA binding and a Cu-GGH ATCUN (amino terminal copper and nickel-binding) motif to perform chemistry on the target, was shown to have high activity, both in vitro against HCV IRES stem loop IIb (SLIIb), and in an

HCV cellular replicon assay.69 In fact, it showed additive to slightly synergic activity when given in combination with recombinant interferon α-2b (rIFNα-2b). Herein is described a similar approach that has been applied to develop a new set of peptides targeting stem loop

IV (SLIV) of the HCV IRES, which contains the AUG start codon. Prior studies have demonstrated the importance of SLIV and its potential as a target for therapeutic intervention, including targeting the GCAC sequences near the AUG start codon of the short hairpin RNA motif, 129 and the use of the LaR2C peptide to block binding of the full length protein.130 Catalytic metallodrugs targeting SLIV should help to improve selectivity and reduce toxicity by taking advantage of a double filter mechanism wherein both binding and proper positioning of the metal center are necessary for irreversible RNA cleavage activity.131 Therefore, nonspecific binding should not lead to side effects and toxicity. This unique mechanism of action complements current therapies and will help to combat the development of resistance.

196

Figure 5.1. The highly structured Internal Ribosome Entry Site (IRES) in the 5’ untranslated region of hepatitis C viral mRNA (left). The secondary structure (center), and three dimensional structure based on SAXS, small angle x-ray scattering, data132 (right) of stem-loop IV (SLIV) which contains the internal start codon (AUG), and residues 5’-G8 to C11-3’ and 5’-G17 to C20-3’ that correspond to the known binding site on the human La protein.

Figure 5.2. Structure of the metal binding ATCUN motif with targeting R groups that define the domains that recognize the GCAC sequences of SLIV. Complex 1-Cu (containing a D-configuration arginine)69 binds selectively to SLIIb IRES RNA, while 2- Cu and 3-Cu are based on the LaR2C peptide that recognizes HCV IRES SLIV.133,134

197

The human La protein belongs to a class of proteins that contain RNA recognition motifs (RRMs) that interact with a variety of RNA’s, including viral RNA’s as well as RNA polymerase III transcripts.135 It is also necessary for HCV IRES mediated translation.136 LaR2C is a peptide derived from the RRM2 of La protein and is responsible for most of the binding to SLIV.133 Further truncation of the LaR2C provides a 7-mer that retains most of the binding affinity.134 Cryoelectron microscopy (cryo-EM) revealed the structure of the IRES bound to the 40S ribosomal subunit, but at low resolution.137 Recent crystallographic studies of the HCV IRES pseudoknot domain has provided a superimposition of the structures of the individual domains providing further structural insight into the cryo-EM data.138 The design strategy described herein consists of the incorporation of a metal-binding ATCUN motif into the LaR2C peptide (2) as well as the truncated 7-mer (3), as shown in Figure 2.

5.2 – Materials

RNA was purchased from Dharmacon, part of Thermo Fisher Scientific (Lafayette,

CO). Peptides were purchased from Genemed Synthesis Inc. (South San Francisco, CA).

The sequence used for the IRES SLIIb RNA was 5’-fluorescein-

GGCAGAAAGCGUCUAGCCAUGGCGUUAGUAUGC C-3’, for the IRES SLIV RNA was 5’-fluorescein-GGACCGUGCACCAUGAGCACGAAUCC-3’, and for the HIV RRE was 5’-fluorescein-UUGGUCUGGGCGCAGCGCAAGCUGACGGUACAGGCC-3’.

The sequences used for calibration of the MALDI-MS instruments included: (GU)3, (GU)9,

(GU)14, and (GU)20. All RNA for binding and cleavage reactions was annealed by heating to 95 ºC and then cooled slowly to room temperature before use. Sodium chloride, sodium hydroxide, and acetonitrile were purchased from Fisher, while HEPES, ammonium citrate, and 3-hydroxypicolinic acid were purchased from Sigma. C18 Zip Tips were obtained 198 from Millipore. All experiments were performed using diethyl pyrocarbonate (DEPC) treated water and autoclaved pipette tips and tubes.

5.3 – Methods

5.3.1 pUC19 Isolation

Plasmid pUC19 DNA was isolated from a transformed DH5α Escherichia coli cell line and purified by use of a Qiagen maxiprep kit. The plasmid was eluted with 20 mM

HEPES and 100 mM NaCl (pH 7.4) instead of TE-Tris-EDTA buffer. The isolated DNA was quantified for purity and concentration by UV-Vis spectrophotometry. This was performed by Jessica Alexander.

5.3.2 Copper Peptides

Peptide concentrations were calibrated by titrating a stock solution of copper (II) chloride to a fixed amount of peptide at 37 oC at 5 min intervals, and then fitting the absorbance profiles at 240 nm and 525 nm. High affinity binding yielded a sharp transition at the concentration corresponding to peptide in solution. Copper complexes of the peptides were observed to form quickly and were prepared by mixing peptide and Cu (II) in a 1.1: 1 ratio, respectively, and left to equilibrate for at least 5 min prior to use.

5.3.3 Binding via Fluorescence

Seth Bradford performed these experiments. Binding data for

GGHKYKETDLLILFKDDYFAKKNEERK-amide (2) and GGHKYKETDL-amide (3) were obtained by adding serial aliquots of IRES SLIV RNA to 500 nM of the peptide and monitoring tyrosine emission (ex = 280 nm, em = 313 nm). Data was then fit to a one-site binding model using Origin 7.0 software.

199

5.3.4 Reaction Kinetics via Fluorescence

HCV IRES RNA cleavage was assessed in vitro by monitoring the change in fluorescence that accompanies reaction chemistry, using 5’ fluorescein end-labeled RNA with excitation and emission wavelengths of 485 nm and 518 nm, respectively. Reactions were carried out at 25 ºC in reaction volumes of 100 µL, and in the presence of 1 mM ascorbate and 1 mM H2O2 in HEPES buffer (pH = 7.4, 100 mM NaCl) with 1 μM fluorescein-labeled IRES SLIV. The resulting time-dependent data was analyzed according to the change in fluorescence emission observed as the reaction occurred. Both a time-dependence and a concentration-dependence of catalyst were observed. Reactions involving one phasic were fit to a first order exponential. All fits were performed using

Origin software. The initial velocity values recorded in the text and tables are an average of at least three trials.

5.3.5 Kinetic Reactions via PAGE Sequencing Gel.

Reactions were carried out in a solution volume of 10 L that contained 10 M fluorescein-labeled IRES SLIV, 1 mM ascorbate and 1 mM H2O2 in Hepes buffer with or without 10 M 1-Cu or 2-Cu present. The reactions were staggered with time points of 10,

20, 30, 40, 50, 60, 90, 120 min. Samples were then incubated with 5 L of 47.5% formamide, 0.5 mM EDTA and 20 % glycerol solution and heated at 65 oC to denature the samples before loading 8 L onto a 20% PAGE gel. The gel was imaged on a GE Typhoon scanner. The gel was quantified by use of Quantity One (Biorad) to determine the relative

200 concentration of the full-length. The values were then plotted to determine the initial rates which were performed in Origin software using either an exponential decay or linear fit.

5.3.6 pUC19 Kinetic Reactions via Gel

Reactions were carried out in a solution volume of 10 L that contained 10 M base pair concentration of pUC19 and 10 M of complex in the presence of 1 mM ascorbate and 1 mM hydrogen peroxide. Reactions were staggered with time points of 10, 20, 30, 40,

50, 60, 75, 90 and 120 min. A 5 L volume of a solution of 47.5% formamide, 0.5 mM

EDTA and 20 % glycerol solution was then added, and from this mixture, 7 L of the samples were loaded onto 1% agarose gels made with gel red (2 L/50 mL of gel solution) with Tris-EDTA-Borate buffer and run at 100 V, and imaged by use of a BioRad Gel Doc before quantification by use of Quantity One software. The intensity of the supercoiled band was multiplied by a correction factor of 1.4 to adjust for difference in Gel Red binding.139 All fits were performed using Origin software.

5.3.7 HIV-RRE (Stem loop IIb) Kinetic Reactions by Gel.

Reactions were carried out as described above but with 10 M Fl-RRE. Samples were then incubated with 5 L of 47.5% formamide, 0.5 mM EDTA and 20 % glycerol solution and heated at 65 oC to denature the samples before loading onto a 4% agarose gel.

The gels were run in parallel at 100 V for 20 min and imaged on a Biorad Gel Doc before quantifying band intensities by use of Quantity One. All fits were performed using Origin software.

201

5.3.8 Mass Spectrometry

Reactions for MALDI-TOF analysis were performed as described earlier123 by use of 10 µM fluorescein-labeled IRES SLIIb with 10 µM copper-peptide, and incubated for up to 90 min. Reactions were then quenched by placing on ice and desalted by use of C18

Zip Tips from Millipore Co. prior to mass spectrometric analysis. Zip Tips were wetted with a 50:50 mixture of acetonitrile:water and equilibrated with 2 M triethylammonium acetate (TEAA), pH 7.0. The reaction mixture was then bound to the Zip Tip, washed with nanopure water, and eluted with 50:50 acetonitrile:water. Samples were spotted onto a

Bruker ground steel 96 target microScout plate by first spotting with 1 µL of the 0.3 M 3- hydroxypicolinic acid (HPA), 30 mM ammonium citrate matrix solution in 30% acetonitrile, drying, then spotting with 1 µL of a 2:1 RNA:matrix mixture, and allowed to dry. A calibration mixture containing 4 RNAs covering a range of molecular weights, namely (GU)3, (GU)9, (GU)14, (GU)20, with molecular weights of 1,892.1, 5,800.4, 9,057.3,

12,965.6 amu, respectively, was used to calibrate the instrument. All MALDI-TOF MS analysis was performed on a Bruker MicroFlex LRF instrument, equipped with a gridless reflectron, using negative ion mode and reflectron mode. The pulsed ion extraction time was 1200 ns. Data analysis was performed using Bruker flexAnalysis software. Only m/z values > 1500 amu were considered, since excessive spectral crowding occurred at lower m/z ranges. Reaction peaks were compared to controls containing RNA alone and RNA in the presence of 1 mM ascorbate, 1 mM H2O2 and new peaks were identified as products of cleavage by the metallopeptide which showed time-dependence.

202

5.3.9 HCV Cellular Replicon Assay

Samples were submitted and run by the Southern Research Institute (Dr. Zhuhui

Huang). A stable cell line ET (luc-ubi-neo/ET), a Huh7 human hepatoma cell line that contains an HCV RNA replicon with a stable luciferase (Luc) reporter and three cell culture-adaptive mutations was employed in the assay. The HCV RNA replicon antiviral evaluation assay examined the effects of compounds at six half-log concentrations each.

Human interferon alpha-2b was included in each run as a positive control compound. Sub- confluent cultures of the ET line were plated out into 96-well plates that were dedicated for the analysis of cell numbers (cytotoxicity) or antiviral activity, and various concentrations of metallodrugs and controls were added to the appropriate wells the following day. Cells were processed 72 h later when the cells were still sub-confluent. Six half-log serial dilutions of the compound were performed, and values derived for IC50 (the concentration that inhibited virus replication by 50%), TC50 (the concentration that lowered cell viability by 50%) and TI (the selectivity index: TC50/IC50). HCV RNA replicon levels were assessed as the replicon-derived Luc activity. The toxic concentration of drug that reduced cell numbers (cytotoxicity) was assessed by use of the CytoTox-1 cell proliferation colorimetric assay (Promega).

203

5.4 – Results and Discussion

5.4.1 Binding and reactivity toward IRES SLIV

Dissociation constants for binding of 2 and 3 to SLIV RNA were obtained by serial addition of RNA to peptide and following tyrosine emission as previously described for binding of 1 to SLIIb RNA.123 A sample plot is shown in Figure 5.11 (supplemental) and the data is summarized in Table 5.1. Both 2 and 3 displayed similar KD values for binding to SLIV RNA.

Table 5.1. Dissociation constants to HCV IRES SLIV for peptides based on LaR2C..

Peptide SLIV KD (μM)

[GGHKYKETDLLILFKDDYFAKKNEERK-amide] (2) 9.9 ± 0.5

[GGHKYKETDL-amide] (3) 9.7 ± 0.1

Binding of LaR2C to SLIV is already well established, 133 and is confirmed by the in vitro fluorescence assays described in Table 1 for both peptides 2 and 3. However, the similarity in affinity further suggests a binding model focused on the KYKETDL motif at the N-terminus of the LaR2C peptide. The NMR structure of the human La protein (PDB

1S79) is available136 and is illustrated in Figure 5.3, with the region corresponding to the

LaR2C peptide highlighted in green and purple. The region corresponding to the 7-mer

KYKETDL-amide that appears to dominate binding to the SLIV RNA is shown in green.

204

Figure 5.3. NMR solution structure of the human La protein showing all 20 structures including the secondary elements superimposed (based on PDB 1S79) with the portion corresponding to LaR2C, 2, highlighted (green and purple). The green domain corresponds to the truncated 7-mer, 3. The left side shows the portion of the La protein from which the compounds were derived, as well as the highly conserved structure (secondary structure, top and atomic resolution, bottom).

Table 5.2. Michaelis-Menten parameters for degradation of SLIV RNA.

-1 kcat/KM Compound kcat (min ) KM (μM) (μM-1 min-1)

2-Cu 0.98 ± 0.07 4.6 ± 1.1 0.21

3-Cu 1.13 ± 0.06 17.2 ± 2.2 0.07

To determine the ability of these compounds to catalytically degrade RNA, cleavage activity was determined by use of an in vitro fluorescence assay, as reported for

Cu-GGHYrFK-amide,69, 123 by use of 5’-fluorescein-labeled SLIV RNA. Time- dependence plots were found to display first order exponential behavior. These initial rates were then plotted as a function of catalyst concentration (Figure 5.5) and the pseudo 205

Michaelis-Menten parameters are summarized in Table 5.2. It is noteworthy that the kcat’s are similar for 2-Cu and 3-Cu, suggesting similar positioning relative to scissile bonds on the RNA. However as result of a smaller targeting domain of 3-Cu compared to 2-Cu, there is a decrease in the binding affinity as reflected in the KM which is most notable in the catalytic efficiency term (kcat/KM).

To further investigate the binding and mechanism of reaction, the salt dependence on the reaction rate constant was determined for 3-Cu with SLIV RNA (Figure 5.6) and showed a significant dependence on [NaCl], most likely reflecting interference with electrostatic interactions to the negatively-charged RNA substrate at higher salt concentration.

225 200 175 150 125 100 75

RFI 518 nm RFI 518 50 25 0 0 2 4 6 8 10 12 14 16 18 20 Time (min)

Figure 5.4. Sample kinetic trace for the reaction of 7.5 µM 2-Cu with HCV IRES SLIV. showing monophasic behavior with the fit shown in red.

206

Figure 5.5. Pseudo Michaelis-Menten plots for reaction of 2-Cu (left) and 3-Cu (right) with HCV IRES SLIV.

1.2

1.0

0.8

)

-1

0.6

k (min

0.4

0.2

0 100 200 300 400 500 [NaCl] (mM)

Figure 5.6. Variation of initial rate with [NaCl] for reaction of 3-Cu with HCV IRES SLIV. Conditions for assay were 20 M 3-Cu, 1 M fluorescein labelled SLIV, 1 mM ascorbate and 1 mM hydrogen peroxide in HEPES buffer pH 7.4.

207

5.4.2 Kinetics of 2-Cu and 3-Cu via PAGE

SLIV RNA cleavage was also examined for both 2-Cu and 3-Cu through sequence

PAGE gels (Table 5.3, Figure 5.12 in the supplementary). Inspection shows that both complexes accelerate cleavage of the SLIV IRES RNA with similar rates of 349 ± 93 nM/min and 314 ± 76 nM/min, respectively, relative to 7 ± 6 nM/min observed for the control with co-reagents only. These results are consistent with the initial rates exhibited in the fluorescence assays (Table 5.3). Closer examination of the gel reveals product bands that are faintly observed, but are not sufficiently resolved for accurate quantification.

5.4.3 Selectivity of 2-Cu and 3-Cu

To probe the selectivity of 2-Cu and 3-Cu for SLIV RNA, additional oligonucleotides, including the rev-response element (RRE) RNA of HIV stem-loop II, and pUC19 (supercoiled DNA) were each tested (Table 5.4 and Figure 5.13 and Figure

5.14, supplementary). No significant cleavage was observed with Fl-HIV RRE and very limited disappearance of supercoiled pUC19 with time. In both cases the rate of cleavage was found to be less than 20 nM/min, relative to the enhanced rate of 300-500 nM/min for

HCV SLIV. Additional data that addresses potential problems with non-selective cleavage are addressed in the section on cellular replicon assays.

These observations reflect an important and unique feature of the approach, namely, the double-filter mechanism. While either of these compounds could bind to other nonspecific targets and act as reversible inhibitors, especially at higher concentrations, nevertheless, it is only in the scenario when these complexes cleave the RNA or oligonucleotide target that they would become the irreversible inhibitors that is the unique property of our design. For this scenario to occur, not only do the metallodrug complexes, 208 such as 2-Cu and 3-Cu, need to bind, but they must also place the catalytic metal in an orientation to produce/provide effective cleavage chemistry on the target (Figure 5.15, supplementary).

Table 5.3. Comparison of initial rate measurements obtained from fluorescence and PAGE profiles

Initial Rate (nM/min) Condition Fluorescence a PAGE b (nM/min) (nM/min)

2-Cu 509 ± 122 349 ± 93

3-Cu 423 ± 105 314 ± 77

coreagents c 8 ± 5 7 ± 6 The graphs are available in supplementary material (Figure 5.12). a Reaction conditions: 1 M Fl-HCV SLIV in the presence of 10 M complex and 1 mM co-reagents. b Reaction conditions: 10 M Fl-HCV SLIV in the presence of 10 M complex and 1 mM co-reagents. c Coreagents added in the absence of 2-Cu and 3-Cu.

Table 5.4. Selectivity of 2-Cu and 3-Cu as represented by initial rates as monitored from gels.

Initial Rate (nM/min) Condition Fl-HCV pUC19 b Fl-HIV RRE a SLIV a

2-Cu 349 ± 93 19 ± 4 7 ± 4

3-Cu 314 ± 77 10 ± 2 14 ± 16

coreagents c 7 ± 6 2 ± 1 0 ± 12 The gels and the graphs are available in supplementary material (Figures 5.12, 5.13, 5.14). a Reaction conditions: 10 M of RNA or 10 M bp of pUC19 in in the presence of 10 M complex and 1 mM co-reagents. b Reaction conditions: 10 M bp of pUC19 in in the presence of 10 M complex and 1 mM co-reagents. c Coreagents added in the absence of 2-Cu and 3-Cu.

209

Figure 5.7. Time-dependence of MALDI-TOF chromatograms following reactions catalyzed by 2-Cu (top) and 3-Cu (bottom) with SLIV RNA in the presence of coreagents.. Time points from front to back are 2, 10, 20, 30, 45, 60, and 90 min. The y-axis is the normalized fraction of the total intensity at each time point and does not account for mass bias. The initial rate was determined at each position, and those that showed a measurable time-dependence are summarized in Table 5.5. The contributions of 3’ and 5’-overhangs are available in Figures 5.16 and SM 5.17 in addition to the control reaction in Figure 5.18.

210

5.4.4 Mass Spectrometric Analysis of RNA Cleavage

To further probe the mechanism of reaction and map potential cleavage sites,

MALDI-TOF mass spectrometry was performed for the reaction of 2-Cu and 3-Cu with

SLIV. MALDI-TOF analysis of RNA cleavage products is complicated by the presence of a large number of fragmentation peaks that are present even in the absence of catalyst and occur as a result of ionization from the instrument as we have previously described.140

To circumvent the problem of identifying actual metallopeptide-mediated cleavage products, only those peaks that were either new and not contained in controls of RNA alone or RNA in the presence of the coreagents ascorbate and H2O2 in the absence of catalyst) or displayed a time-dependent change in intensity were considered. Illustrative time- dependent mass spectra are shown in Figure 5.9 and peak assignments for each time point are detailed in Tables 5.11 and 5.12 (supplementary).

Most of the observed products occur within a common locus (Figure 5.9). For reactions with 2-Cu, the most significant sites of reactions are at G15 and A16, which is the last and the nucleotide immediately following the internal AUG start codon sequence

(from site 13 to 15). For 3-Cu the main cleavage occurs at G8 which is part of one of the

GCAC motifs. For both of these complexes, reactivity is centered on the GCAC domains

(Figure 5.8 and 5.9) that are known to be important for ribosome assembly. A visual representation of the initial rates is available in Figure 5.9 in both a 2D and 3D representation with the actual values summarized in Table 5.5. It is important to keep in mind that the proposed 3D structure is not based on crystallographic or NMR information,

211 but rather was derived from linear-least fitting of SAXS data information, and is an average of the relative populations of various structures.

Figure 5.8. Three and Two dimensional representations of SLIV RNA. The three dimensional representation is based on a SAXS structure predicted by Perard et al.132 The two dimensional representation is based on VARNA structure prediction. The 5’-GCAC- 3’ sequences highlighted correspond to potential target recognition sites for 2-Cu and 3-Cu (blue and orange), while, the AUG start codon (green) at the top of the hair-pin is also highlighted.

212

Figure 5.9. (Next page) Mass spectrometric analysis of cleavage products for reactions of HCV IRES SLIV mediated by 2-Cu (top) and 3-Cu (bottom) including heat maps of reactivity obtained by assessing the relative initial rate values at each nucleotide position. The black bases or gray highlighted region represents the two known binding sequences of 5’-GCAC-3’ which are labelled. It can be seen that reactivity occurs on either side of this region. This suggests that the copper complexes induce a structural change which brings the two domains close in proximity which facilitates the range of cleavage. Relative Scale: Red x > 10, Orange 8 < x < 10, Yellow 6 < x < 8, Green 4 < x < 6, Blue 2 < 0 < 4, Purple x < 2. Values are from the initial rates observed from MALDI-MS multiplied by 1000. Include reference to Figure 5.7 and 5.8.

213

Figure 5.9

214

The data appears to suggest that each complex preferentially reacts at different sites; primarily at the top of the stem-loop for 2-Cu (with modest activity in the lower domain), and in the lower domain for 3-Cu. This in turn could reflect the presence of two potential GCAC binding motifs within the SLIV sequence (Figure 5.9), with two distinct binding orientations for the peptide depending on the preferred GCAC that is targeted by each complex. However, such a model is not reflected by binding or kinetic data (Table

5.1 and Figure 5.4) that evidence only one site. An alternative and more likely scenario reflects the conformational flexibility of the RNA. We propose that with binding of 2-Cu, the extra C-terminal amino acids provide additional contacts that facilitate the further

“bending” of the RNA to permit more efficient chemistry, as reflected by comparison of both the kcat/KM ratios (Table 5.2) and relative initial rates from MALDI experiments

(Figure 5.9 and Table 5.5). Evidence in support of this idea comes from the 3D structural model based on SAXS studies (Figure 5.9), which shows the SLIV domain to be already bent. In the absence of base pairing in the lower bulge and the top of the hair pin, these regions are expected to be flexible and accessible by the catalytic metal, and the kinetic data support a model whereby the RNA folds in a manner that brings the bulge and hairpin in closer proximity to the reactive metal center. For 3-Cu, the reactivity is more focused around one GCAC motif (G17-C20).

Additional mechanistic insight was obtained by examining the types of overhang produced by each reaction (Figure 5.10). Because little work has been reported for direct characterization of the kind of overhang expected from RNA cleavage reactions, product patterns were based on those reported for DNA fragmentation.141 Of interest is the rate at

215 which different overhangs are produced. In general, the relative rates are similar to both complexes across the different overhangs observed (Table 5.6). This is not surprising as their reactivity has been approximately the same as determined by other techniques. This also includes the general sphere of reactivity towards SLIV. 2-Cu appears to be slightly faster in the production of 2’, 3’-cyclic phosphates and 3’-phosphates, which suggests that additional contacts from the longer peptide stabilizes the metallopeptide-RNA complex to preform slightly more efficient chemistry. The proposed mechanism of interaction of action of these compounds have been reported and is available in Figure 2.10 (included for convenience as Figure 5.23).123

216

Figure 5.10. Relative amount of each class of product, following catalytic reaction with 2- Cu (top) and 3-Cu (bottom). as observed by mass spectrometry and displayed as a function of time. Kinetic data fits are available in the supplementary material (Figure 5.19 and 5.20).

217

Table 5.5. Relative initial rate for SLIV cleavage reactions mediated by 2-Cu and 3-Cu determined by MALDI-TOF experiments.

Initial Rate (Normalized Intensity/min) x 1000

2-Cu 3-Cu

G2 0.3 ± 0.1

A3 9.3 ± 1.0

C4 4.5 ± 0.1 3.3 ± 0.4

G6 3.1 ± 0.5

G8 37.1 ± 8.7

A10 3.5 ± 0.4

C12 6.1 ± 0.3

G15 9.1 ± 0.8 0.6 ± 0.4

A16 13.8 ± 4.2 1.1 ± 0.3

G17 1.4 ± 0.3

A19 0.8 ± 0.1

C20 1.0 ± 0.3

G21 3.1 ± 0.1 0.9 ± 0.3

U24 0.5 ± 0.1

Kinetic fits for the initial rates are available in the supplemental material (Figure 5.21 for 2-Cu and Figure 5.22 for 3-Cu).

218

Table 5.6. Initial rates for select overhang products observed following reactions promoted by 2-Cu and 3-Cu as monitored by time-dependent MALDI-TOF mass spectrometry. (see also Tables 5.11 and 5.12 and Figures 5.19 and 5.20 available in the supplementary).

mass 2-Cu 3-Cu Possible Overhang (amu) (%/min) (%/min) Mechanism

2',3'-Cyclic 61.96 10.5 ± 4.3 4.3 ± 0.5 Hydrolysis Phosphates

Hydrolysis, 1’, 2’, 3’, 4’, or 3'-Phosphates 79.98 6.4 ± 1.3 2.8 ± 0.2 5’-H abstraction a

3'-Phosphoglycolates 138.02 4.5 ± 1.4 4.0 ± 1.8 4’H abstraction

5-Hydroxyl 0.00 2.3 ± 0.1 1.1 ± 0.1 Hydrolysis

Hydrolysis, 1’, 2’, 3’, 4’, or 5'-phosphates 79.98 2.8 ± 2.5 6.1 ± 1.4 5’-H abstraction a a Although 1’, 2’, 3’, 4’, or 5’-H-abstraction are possible 4’, 5’-H abstraction are expected to be more common as a result of greater solvent exposure.

219

5.4.5 Cellular Replicon Activity

After binding and reactivity were established in vitro, the activity of 2-Cu and 3- Cu were evaluated in cell culture by use of a cellular HCV replicon assay that mimics the native HCV replication process and is accepted by the FDA as a measure of the efficacy of potential HCV treatments. Results are shown in Table 5.7 and are compared to the data previously reported for 1-Cu.69 Both metallopeptides showed activity similar to 1-Cu and, as for 1-Cu, no activity was observed in controls of the RNA binding domain alone, Cu-

GGH alone, free copper, or the peptide without copper. Recombinant IFNα-2b was used as a positive control.

Table 5.7. HCV cellular replicon activity for 2-Cu and 3-Cu.

Antiviral Cytotoxicity Selectivity Compound IC50 (μM) TC50 (μM) Index TI

[GGHKYKETDLLILFKDDYFAKKNEERK]-Cu, 0.75 > 50 > 67 2-Cu [GGHKYKETDL]-Cu, 2.17 > 50 > 23 3-Cu

[GGHYrFK]-Cu, 0.58 > 100 > 172 1-Cu

[KYKETDLLILFKDDYFAKKNEERK] > 100 > 100 na a (2)

[GGH]-Cu2+ > 100 > 100 na a

Cu2+ (aq) > 100 > 100 na a

a na- not applicable

220

The absence of replicon activity for the metal-free peptides, the GGH metal binding domain, and RNA binding domains alone are each consistent with 2-Cu and 3-Cu reacting in the cellular assays by targeted catalytic cleavage, as observed in vitro, as opposed to nonspecific reaction or competitive inhibition. The absence of replicon activity by 2

(without copper) contrasts with a previous report of cellular translational inhibition by the

LaR2C peptide.130 Given that the HCV cellular replicon assays used herein are the standard assays accepted by the FDA, while the earlier studies were performed using a simpler, less robust assay, these differences can reasonably be attributed to the use of distinct cellular assays. This cellular replicon activity, along with the lack of toxicity, further supports a model where these metallopeptides exhibit specificity for HCV RNA.

The metallopeptides described herein expand the available pool of potential catalytic metallodrugs for the treatment of HCV infection. The results help to validate the targeting of IRES RNA by 2-Cu and 3-Cu as a possible therapeutic approach to the treatment of HCV infection and further demonstrate the general applicability of the catalytic metallodrug approach to drug design. Building on the prior report of SLIIb RNA cleavage by Cu-GGHYrFK-amide (1-Cu), and the demonstration of its effectiveness as an antiviral agent when used in combination with rIFNα-2b, the current results suggest that 2-

Cu and 3-Cu could also find application in combination with current therapeutics, which now include both protease and polymerase inhibitors. In particular, their novel mechanism of action renders them ideal for use in combination treatments. Alternatively, such compounds could be used in tandem with compounds such as 1-Cu as a mixture that targets various IRES stem loop structures.

221

5.5 – Conclusions

Several currently approved FDA drugs have developed resistance to HCV and, therefore, the need for a multiple target approach has only grown. To circumvent resistance problems, HCV treatment will require multidrug therapy. Drugs that target SLIV of the

HCV IRES RNA represent a novel approach and mode of action, as well as a distinct therapeutic target for the treatment of HCV infection and have the potential to complement both current treatments and those still in development.

222

5.6 – Supplemental Material

1.2

1.0

0.8

0.6

0.4

Fraction Bound 0.2

0.0 0 5 10 15 20 25 30 [IRES SLIV] (M)

Figure 5.11. Sample curve for binding of 3 to HCV IRES SLIV.

223

Figure 5.12. Disappearance of full-length Fl-HCV-SLIV with 2-Cu in the presence of ascorbic acid and hydrogen peroxide. (top left, purple arrow), co-reagents only (top middle, red arrow), or 3-Cu with co-reagents (top right, green arrow). The control lane (C) is shown on the on the far right. The time points for each data set were (left to right) 10, 20, 30, 40, 50, 60, 90, and 120 min. Reaction conditions included 10 M RNA and 10 M complex with 1 mM ascorbic acid and 1 mM hydrogen peroxide. (Bottom) Graphical representation of the disappearance of the full-length Fl-HCV SLIV, 2-Cu (purple circles), 3-Cu (green triangles), and co-reagents only (red squares).

224

Figure 5.13. Reactivity of co-reagents with Fl-HIV-RRE with or without the presence of either 2-Cu or 3-Cu. Lanes from left to right, Control, 10, 20, 30, 40, 50, 60, 75, 90, 120 minutes. (Top) RNA with co-reagents and 2-Cu; (middle) RNA with co-reagents and 3- Cu; (bottom) Fl-HIV-RRE with co-reagents. (Bottom) Graphical representation of the disappearance of the full-length Fl-HIV RRE, 2-Cu (purple circles), 3-Cu (green triangles), co-reagents only (red squares).

225

Figure 5.14. Reactivity of co-reagents with 10 M bp supercoiled pUC19 with or without the presence of either 2-Cu or 3-Cu. Lanes from left to right, Control, 10, 20, 30, 40, 50, 60, 75, 90, 120 minutes. (Top) pUC19 with co-reagents and 2-Cu; (middle) pUC19 with co-reagents and 3-Cu; (bottom) pUC19 with co-reagents. (Bottom) Graphical representation of the disappearance of pUC19, 2-Cu (purple circles), 3-Cu (green triangles), co-reagents only (red squares).

226

Figure 5.15. The principle of the double-filter effect. (Left) graphical representation of a metallopeptide, depicting the catalytic metal domain (Cu-GGH moiety), and a generic targeting domain. (Middle and right) Binding of the targeting domain to two different RNAs. Even though both show binding, it is only the target on the right that is subject to irreversible cleavage and inactivation. This results from the positioning of the catalytic metal domain to perform chemistry on the target but not on the alternative RNA. Use of lower concentrations will favor destruction of the therapeutic target and minimize reversible inhibition of the alternative RNA motif. In both of these cases the compound acts a reversible inhibitor until cleavage occurs.

227

Figure 5.16. Time dependence of MALDI-TOF chromatograms for reactions promoted by 2-Cu with Fl-SLIV in the presence of coreagents. (Top). Time points from front to back include 2, 10, 20, 30, 45, 60, 90 min. The y-axis is the normalized fraction of the total intensity at each time point and does not account for mass bias. (Bottom left) Shows the contribution of the 3’-overhangs (2’/3’-phosphates, 2’, 3’-cyclic phosphates, 3’-OH, 3’- phosphoglycolates, 3’-enol/aldehydes, and 3’-a-B). (Bottom right) Shows the contribution of the 5’-overhangs (5’-OH/aldehydes, 5’-phosphates, and 5’-z).

228

Figure 5.17. Time dependence of MALDI-TOF chromatograms for reactions promoted by 3-Cu with Fl-SLIV in the presence of coreagents.(Top). Time points from front to back include 2, 10, 20, 30, 45, 60, and 90 min. The y-axis is the normalized fraction of the total intensity at each time point and does not account for mass bias. (Bottom left) The contribution of the 3’-overhangs (2’/3’-phosphates, 2’, 3’-cyclic phosphates, 3’-OH, 3’- phosphoglycolates, 3’-enol/aldehydes, and 3’-a-B). (Bottom right) The contribution of the 5’-overhangs (5’-OH/aldehydes, 5’-phosphates, and 5’-z).

229

Figure 5.18. Time dependence MALDI-TOF chromatogram of Fl-SLIV in the presence of coreagents. (Top). Time points from front to back include 2, 10, 20, 30, 45, 60, and 90 min. The y-axis is the normalized fraction of the total intensity at each time point and does not account for mass bias. (Bottom left) The contribution of the 3’-overhangs (2’, 3’- phosphates, 2’,3’-cyclic phosphates, 3’-OH, 3’-phosphoglycolates, 3’-enol/aldehydes, and 3’-a-B). (Bottom right) The contribution of the 5’-overhangs (5’-OH/aldehydes, 5’- phosphates, and 5’-z).

230

Table 5.8. MALDI-TOF peak global assignments for each time point from the 2-Cu promoted reaction with SLIV in the presences of co-reagents. 2 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7977.89 7977.99 12.90 2 5'-OH 140.21 0.0039 1632.15 1632.07 -49.04 3 3'-phosphoglycolate 1125.06 0.0313 7632.68 7632.52 -21.40 3 5'-OH 595.55 0.0166 1975.37 1975.39 8.19 4 3'-a-B 3734.01 0.1038 2511.67 2511.55 -48.61 6 2',3'-cyclic phosphate 1577.77 0.0438 3486.24 3486.73 141.75 9 3'-phosphate 12239.36 0.3401 4178.67 4177.58 -261.49 11 3'-phosphoglycolate 3959.29 0.1100 3471.18 3470.55 -181.25 16 5'-OH 9862.97 0.2741 1937.17 1936.70 -245.01 21 5'-phosphate 2750.41 0.0764

10 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 8057.87 8059.70 226.69 2 5'-phosphate 108.50 0.0032 7632.68 7633.42 96.55 3 5'-OH 2558.92 0.0750 2511.67 2511.92 101.17 6 2',3'-cyclic phosphate 696.12 0.0204 6693.11 6694.94 273.32 6 5'-OH 26.40 0.0008 3163.05 3162.25 -252.23 8 2',3'-cyclic phosphate 2848.47 0.0835 3181.06 3180.47 -185.89 8 3'-phosphate 1889.24 0.0554 3815.45 3815.06 -102.18 10 3'-phosphate 1797.83 0.0527 5062.13 5060.80 -262.24 11 5'-OH 245.54 0.0072 5061.19 5060.80 -76.56 14 3'-phosphate 245.54 0.0072 5388.39 5387.54 -157.26 15 2',3'-cyclic phosphate 1714.60 0.0503 3471.18 3470.57 -175.79 16 5'-OH 15345.22 0.4498 3141.97 3142.78 258.18 17 5'-OH 585.85 0.0172 2796.76 2796.76 1.31 18 5'-OH 1026.97 0.0301 1937.17 1936.73 -227.27 21 5'-phosphate 1281.15 0.0376 7347.59 7347.30 -39.48 21 2',3'-cyclic phosphate 116.36 0.0034 8312.18 8313.68 181.00 24 2',3'-cyclic phosphate 168.98 0.0050 8860.57 8861.30 81.90 26 3'-OH 3460.01 0.1014

continued

231

Table 5.8 continued

20 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7977.89 7975.79 -263.53 2 5'-OH 170.83 0.0079 1937.33 1936.80 -274.70 4 3'-phosphoglycolate 1530.67 0.0709 1975.37 1975.83 231.98 4 3'-a-B 1164.63 0.0539 2511.67 2511.44 -89.78 6 2',3'-cyclic phosphate 1285.11 0.0595 6693.11 6692.58 -79.81 6 5'-OH 39.13 0.0018 3486.24 3486.31 21.40 9 3'-phosphate 9239.77 0.4278 4178.67 4177.79 -210.91 11 3'-phosphoglycolate 3008.07 0.1393 3453.17 3453.48 89.35 16 5'-z 2261.79 0.1047 3141.97 3141.94 -8.41 17 5'-OH 972.75 0.0450 1937.17 1936.80 -190.58 21 5'-phosphate 1530.67 0.0709 7347.59 7347.15 -60.06 21 2',3'-cyclic phosphate 153.88 0.0071 8312.18 8311.82 -43.72 24 2',3'-cyclic phosphate 238.61 0.0110

30 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7977.89 7978.53 80.63 2 5'-OH 353.36 0.0070 7303.47 7302.64 -113.84 4 5'-OH 73.01 0.0014 2511.67 2511.50 -67.54 6 2',3'-cyclic phosphate 4504.17 0.0887 6693.11 6693.24 20.00 6 5'-OH 265.06 0.0052 3486.24 3486.63 111.01 9 3'-phosphate 22575.42 0.4447 3815.45 3814.63 -215.04 10 3'-phosphate 4971.40 0.0979 5388.39 5387.08 -243.77 15 2',3'-cyclic phosphate 9483.08 0.1868 5406.40 5405.03 -254.23 15 3'-phosphate 5067.30 0.0998 3453.17 3453.98 234.25 16 5'-z 3251.93 0.0641 7002.38 7002.22 -23.49 20 2',3'-cyclic phosphate 90.17 0.0018 8250.21 8250.30 11.21 24 3'-OH 129.45 0.0025

Continued

232

Table 5.8 Continued

45 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 1632.15 1632.19 20.19 3 3'-phosphoglycolate 1332.04 0.0407 1670.19 1669.74 -270.47 3 3'-a-B 845.60 0.0258 1975.37 1975.21 -83.41 4 3'-a-B 3375.83 0.1032 7303.47 7302.32 -157.22 4 5'-OH 11.61 0.0004 2242.51 2242.32 -88.22 5 3'-phosphoglycolate 1316.72 0.0402 6693.11 6692.69 -62.89 6 5'-OH 92.58 0.0028 3486.24 3486.94 201.20 9 3'-phosphate 8455.90 0.2585 3544.28 3543.45 -233.94 9 3'-phosphoglycolate 3058.81 0.0935 3471.18 3470.37 -232.85 16 5'-OH 6458.36 0.1974 3453.17 3453.59 122.40 16 5'-z 1811.80 0.0554 3141.97 3142.72 239.85 17 5'-OH 964.61 0.0295 2242.35 2242.32 -15.53 20 5'-phosphate 1316.72 0.0402 1937.17 1936.70 -240.91 21 5'-phosphate 2426.28 0.0742 1591.96 1592.23 171.24 22 5'-phosphate 1250.02 0.0382

60 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7632.68 7631.88 -104.32 3 5'-OH 4603.02 0.0577 2511.67 2511.80 49.83 6 2',3'-cyclic phosphate 2201.31 0.0276 6693.11 6693.21 15.57 6 5'OH 99.49 0.0012 3163.05 3162.45 -189.21 8 2',3'-cyclic phosphate 4503.11 0.0564 3815.45 3815.17 -72.96 10 3'-phosphate 3355.86 0.0421 4178.67 4179.35 162.37 11 3'-phosphoglycolate 3015.34 0.0378 4467.85 4467.09 -170.88 12 3'-enol/aldehyde 4856.58 0.0609 4755.02 4755.47 95.30 13 3'-phosphate 2489.12 0.0312 5388.39 5387.85 -100.48 15 2',3'-cyclic phosphate 4786.66 0.0600 5406.40 5406.55 27.71 15 3'-phosphate 1861.22 0.0233 3471.18 3470.48 -201.59 16 5'-OH 38085.95 0.4774 3141.97 3141.66 -99.17 17 5'-OH 1451.17 0.0182 2796.76 2796.54 -80.15 18 5'-OH 3562.47 0.0447 6367.99 6368.51 82.30 18 2',3'-cyclic phosphate 24.28 0.0003 7347.59 7348.36 104.73 21 2',3'-cyclic phosphate 98.34 0.0012 continued 233

Table 5.8 Continued

1591.96 1592.14 111.97 22 5'-phosphate 954.51 0.0120 8250.21 8249.58 -76.27 24 3'-OH 49.97 0.0006 8860.57 8861.67 123.73 26 3'-OH 3780.34 0.0474

90 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7977.89 7977.80 -10.93 2 5'-OH 114.20 0.0016 1937.33 1936.89 -229.50 4 3'-phosphoglycolate 1531.99 0.0218 1975.37 1975.36 -5.40 4 3'-a-B 1326.77 0.0189 1861.28 1861.28 -1.64 4 2',3'-cyclic phosphate 587.69 0.0083 1879.29 1879.32 16.14 4 3'-phosphate 510.07 0.0072 2242.51 2242.21 -134.79 5 3'-phosphoglycolate 972.87 0.0138 2511.67 2511.56 -45.26 6 2',3'-cyclic phosphate 2326.01 0.0330 6693.11 6691.69 -212.16 6 5'-OH 152.38 0.0022 3163.05 3162.11 -295.64 8 2',3'-cyclic phosphate 2625.14 0.0373 3486.24 3487.02 224.06 9 3'-phosphate 13143.42 0.1867 3815.45 3815.02 -113.38 10 3'-phosphate 2531.31 0.0360 4178.67 4178.21 -110.32 11 3'-phosphoglycolate 2673.34 0.0380 4120.63 4120.67 10.58 11 3'-phosphate 1973.55 0.0280 4102.62 4102.38 -57.86 11 2',3'-cyclic phosphate 1431.46 0.0203 4483.85 4484.92 238.94 12 3'-phosphoglycolate 3238.32 0.0460 4425.81 4426.31 113.35 12 3'-phosphate 2507.36 0.0356 4407.80 4407.28 -117.07 12 2',3'-cyclic phosphate 1509.98 0.0215 4755.02 4755.02 -0.86 13 3'-phosphate 1650.74 0.0235 5388.39 5387.63 -140.48 15 2',3'-cyclic phosphate 2622.08 0.0373 5406.40 5405.71 -128.34 15 3'-phosphate 2429.07 0.0345 3471.18 3470.39 -227.59 16 5'-OH 20089.61 0.2854 3141.97 3141.64 -103.54 17 5'-OH 493.55 0.0070 2796.76 2796.41 -126.11 18 5'-OH 698.09 0.0099 2242.35 2242.21 -62.11 20 5'-phosphate 972.87 0.0138 1937.17 1936.89 -145.38 21 5'-phosphate 1531.99 0.0218 1511.98 1511.99 3.34 22 5'-OH 740.56 0.0105

234

Table 5.9. MALDI-TOF peak global assignments for each time point from the 3-Cu promoted reaction with SLIV in the presences of co-reagents.

2 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 1632.15 1632.10 -35.31 3 3'-phosphoglycolate 520.36 0.0593 7632.68 7631.31 -179.58 3 5'-OH 118.81 0.0021 1975.37 1975.07 -152.69 4 3'-a-B 524.76 0.0564 2511.67 2511.23 -176.92 6 2',3'-cyclic phosphate 350.47 0.0546 6693.11 6692.87 -36.16 6 5'-OH 14.14 0.0003 3486.24 3485.91 -94.18 9 3'-phosphate 3196.97 0.3624 3453.17 3453.51 99.10 16 5'-z 1328.08 0.0194 3141.97 3142.39 135.22 17 5'-OH 367.11 0.0064

10 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7632.68 7634.41 226.50 3 5'-OH 185.77 0.0020 1937.33 1937.12 -107.78 4 3'-phosphoglycolate 1880.90 0.0893 1975.37 1975.67 150.64 4 3'-a-B 1719.87 0.0862 2242.51 2242.12 -175.72 5 3'-phosphoglycolate 977.80 0.0509 2511.67 2511.71 14.22 6 2',3'-cyclic phosphate 601.53 0.0249 6693.11 6693.76 96.78 6 5'-OH 33.53 0.0004 3486.24 3485.38 -247.77 9 3'-phosphate 6557.81 0.3133 3468.23 3469.03 231.81 9 2',3'-cyclic phosphate 4843.92 0.2060 3453.17 3453.09 -21.87 16 5'-z 1465.26 0.0164 3123.96 3124.79 265.15 17 5'-z 1524.76 0.0173 3141.97 3141.83 -43.31 17 5'-OH 1471.02 0.0159 2242.35 2242.12 -103.04 20 5'-phosphate 977.80 0.0106 1937.17 1937.12 -23.65 21 5'-phosphate 1880.90 0.0206

continued

235

Table 5.9 Continued

20 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7632.68 7632.88 26.39 3 5'-OH 202.60 0.0017 1937.33 1937.30 -18.65 4 3'-phosphoglycolate 4773.50 0.1112 1975.37 1975.35 -11.27 4 3'-a-B 4393.78 0.1017 2242.51 2241.85 -297.73 5 3'-phosphoglycolate 3223.34 0.0845 3468.23 3468.62 112.48 9 2',3'-cyclic phosphate 17997.33 0.3371 3486.24 3485.22 -291.90 9 3'-phosphate 12981.06 0.3036 4407.80 4407.93 29.74 12 2',3'-cyclic phosphate 472.44 0.0134 4738.94 4738.30 -135.98 12 5'-z 37.14 0.0003 4737.01 4738.30 271.39 13 2',3'-cyclic phosphate 37.14 0.0011 3123.96 3124.55 189.18 17 5'-z 1268.99 0.0114 2242.35 2241.85 -225.06 20 5'-phosphate 3223.34 0.0269 1937.17 1937.30 65.50 21 5'-phosphate 4773.50 0.0410

30 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 8323.10 8323.09 -1.25 1 5'-OH 7.29 0.0001 1937.33 1937.54 108.48 4 3'-phosphoglycolate 1563.45 0.0793 2242.51 2242.25 -118.53 5 3'-phosphoglycolate 890.95 0.0491 2280.55 2280.21 -150.43 5 3'-a-B 922.51 0.0321 6693.11 6693.12 1.99 6 5'-OH 22.62 0.0002 3468.23 3468.28 13.70 9 2',3'-cyclic phosphate 4041.27 0.2637 4737.01 4737.20 40.55 13 2',3'-cyclic phosphate 42.51 0.0038 3453.17 3452.17 -288.85 16 5'-z 1452.24 0.0139 3141.97 3141.23 -234.92 17 5'-OH 1978.76 0.0181 3123.96 3124.30 108.85 17 5'-z 1344.91 0.0130 2796.76 2796.26 -178.05 18 5'-OH 603.06 0.0056 2242.35 2242.25 -45.85 20 5'-phosphate 890.95 0.0083 1937.17 1937.54 192.64 21 5'-phosphate 1563.45 0.0147 8555.39 8555.19 -23.87 25 3'-OH 21.95 0.0014

continued

236

Table 5.9 Continued

45 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 1937.33 1937.78 228.78 4 3'-phosphoglycolate 2177.68 0.1270 1975.37 1975.37 -1.77 4 3'-a-B 1757.59 0.1042 2242.51 2241.99 -234.21 5 3'-phosphoglycolate 1267.00 0.0838 3468.23 3467.59 -185.72 9 2',3'-cyclic phosphate 3280.43 0.2776 4102.62 4103.11 120.12 11 2',3'-cyclic phosphate 221.01 0.0259 4738.94 4737.92 -216.03 12 5'-z 110.85 0.0017 4737.01 4737.92 191.31 13 2',3'-cyclic phosphate 110.85 0.0133 3123.96 3123.78 -58.31 17 5'-z 269.47 0.0042 2242.35 2241.99 -161.54 20 5'-phosphate 1267.00 0.0193 7347.59 7345.53 -280.74 21 2',3'-cyclic phosphate 72.67 0.0089 1634.00 1633.99 -6.48 22 5'-enol/aldehyde 1167.88 0.0183 8617.36 8616.23 -131.10 25 2',3'-cyclic phosphate 114.83 0.0141

60 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7977.89 7977.38 -63.76 2 5'-OH 32.03 0.0003 1632.15 1632.35 121.09 3 3'-phosphoglycolate 2225.07 0.1404 1879.29 1879.12 -92.69 4 3'-phosphate 1335.56 0.1984 3468.23 3467.21 -293.43 9 2',3'-cyclic phosphate 4800.17 0.4161 4104.55 4104.79 59.63 14 5'-z 189.43 0.0020 3123.96 3123.29 -214.84 17 5'-z 392.07 0.0042 7347.59 7348.09 67.83 21 2',3'-cyclic phosphate 4.16 0.0006 1591.96 1592.36 252.28 22 5'-phosphate 2556.38 0.0268 8250.21 8249.41 -97.16 24 3'-OH 35.40 0.0029 8860.57 8862.18 181.59 26 3'-OH 494.06 0.0411

Continued

237

Table 5.9 continued

90 min Mass Norm Theor Obsd Peak Error Position Overhang Peak Mass Mass Area (ppm) Area 7977.89 7975.88 -251.96 2 5'-OH 19.59 0.0004 1574.11 1573.96 -93.43 3 3'-phosphate 69.55 0.0059 7632.68 7631.27 -184.23 3 5'-OH 70.69 0.0016 1975.37 1974.88 -246.81 4 3'-a-B 1611.27 0.0756 1879.29 1879.12 -89.18 4 3'-phosphate 246.52 0.0212 1861.28 1860.89 -209.92 4 2',3'-cyclic phosphate 15.03 0.0013 3486.24 3486.32 24.01 9 3'-phosphate 2375.82 0.2089 4737.01 4737.26 52.10 13 2',3'-cyclic phosphate 6.18 0.0005 3471.18 3470.35 -238.77 16 5'-OH 920.37 0.0210 3141.97 3142.23 81.19 17 5'-OH 600.52 0.0140 2796.76 2796.30 -166.20 18 5'-OH 392.75 0.0093 2242.35 2241.68 -297.58 20 5'-phosphate 977.70 0.0251 1511.98 1512.18 135.29 22 5'-OH 2908.68 0.0694 1591.96 1592.30 214.05 22 5'-phosphate 1497.62 0.0394 8555.39 8554.41 -113.98 25 3'-OH 28.33 0.0024 8860.57 8862.00 161.60 26 3'-OH 134.50 0.0114

238

Figure 5.19. Variation of overhang with time as determined for 2-Cu and HCV SLIV from time-dependent MALDI-MS analysis; (A) 2’,3’-cyclic phosphates (B) 3’-phosphates (C) 3’-phosphoglycolates, (D) 5’-hydroxyls (E) 5’- phosphates.

239

Figure 5.20. Variation in overhang with time as determined for 3-Cu and HCV SLIV from time-dependent MALDI-MS analysis; (A) 2’,3’-cyclic phosphates (B) 3’-phosphates (C) 3’-phosphoglycolates, (D) 5’-hydroxyls (E) 5’- phosphates.

240

0.5

0.4 G2 C4 G6 0.3 A10 C11 G15 0.2 A16 G17 0.1 G21 U24

Normalized Relative Intensity 0.0 0 10 20 30 Time (min)

Figure 5.21. Change in relative normalized intensity at each overhang with time as determined for 2-Cu and HCV SLIV from time-dependence MALDI-MS analysis linear fits to determine relative initial rates are included.

241

0.7

0.6 A3 C4 0.5 G8 G15 0.4 A16 A19 0.3 C20 G21 0.2

0.1

Normalized Relative Intensity 0.0 0 10 20 30 Time (min)

Figure 5.22. Change in relative normalized intensity at each overhang with time as determined for 3-Cu and HCV SLIV from time-dependent MALDI-MS analysis linear fits to determine relative initial rates are included.

242

Figure 5.23. Proposed mechanisms for nuclease activity. Under oxidative conditions, catalysts can promote RNA cleavage chemistry either through “hydrolytic” (A) or “oxidative” (B) mechanisms promoted through the transient formation of higher valent Cu3+. For full discussion see Figure 2.10.

243

References

1. Fricker, S. P., Metal based drugs: from serendipity to design. Dalton transactions 2007, (43), 4903-17. 2. Whitehouse, M. W., Therapeutic gold. Is it due for a come-back? Inflammopharmacology 2008, 16 (3), 107-9. 3. Reedijk, J., Bioinorganic chemistry. Naturwissenschaften 1987, 74 (2), 71-77. 4. Farrer, N. J. S., Peter, J., Medicinal Inorganic Chemistry: State of the Art, New Trends, and a Vision of the Future. In Bioinorganic Medicinal Chemistry, Alessio, E., Ed. Wiley-VCH: 2011. 5. Joyner, J. C.; Hocharoen, L.; Cowan, J. A., Targeted Catalytic Inactivation of Angiotensin Converting Enzyme by Lisinopril-Coupled Transition Metal Chelates. Journal of the American Chemical Society 2012, XXX (XX), XX. 6. Wöhrle, D.; Pomogailo, A. D. 7. Norn, S.; Permin, H.; Kruse, E.; Kruse, P. R., [Mercury--a major agent in the history of medicine and alchemy]. Dansk medicinhistorisk arbog 2008, 36, 21-40. 8. Robinson, W. J., Everlasting Pill. The Medical Critic and Guide 1907, 8-10 (2). 9. Risher, J. F.; Amler, S. N., Mercury exposure: evaluation and intervention the inappropriate use of chelating agents in the diagnosis and treatment of putative mercury poisoning. Neurotoxicology 2005, 26 (4), 691-9. 10. O'Carroll, R. E.; Masterton, G.; Dougall, N.; Ebmeier, K. P.; Goodwin, G. M., The neuropsychiatric sequelae of mercury poisoning. The Mad Hatter's disease revisited. The British Journal of Psychiatry 1995, 167 (1), 95-98. 11. Yarnell, A., Salvarsan. Chemical & Engineering News 2005/06/20, 2005. 12. Frith, J., Arsenic – the “Poison of Kings” and the “Saviour of Syphilis”. Journal of Military and Veterans' Health 2013, 21 (4). 13. Thorburn, A. L., Paul Ehrlich: pioneer of chemotherapy and cure by arsenic (1854-1915). Sexually Transmitted Infections 1983, 59 (6), 404-405. 14. Lloyd, N. C.; Morgan, H. W.; Nicholson, B. K.; Ronimus, R. S., The composition of Ehrlich's salvarsan: resolution of a century-old debate. Angewandte Chemie 2005, 44 (6), 941-4. 15. Rosenberg, B.; Van Camp, L.; Krigas, T., Inhibition of Cell Division in Escherichia coli by Electrolysis Products from a Platinum Electrode. Nature 1965, 205 (4972), 698-699. 16. Dhar, S. L., Stephen J., Current Status and Mechanism of Action of Platinum- Based Anticancer Drugs. In Bioinorganic Medicinal Chemistry, Alessio, E., Ed. Wiley- VCH: 2011. 244

17. Cowan, J. A., Catalytic metallodrugs. Pure Appl. Chem. 2008, 80 (8), 1799-1810. 18. Sherman, S.; Gibson, D.; Wang, A.; Lippard, S., X-ray structure of the major adduct of the anticancer drug cisplatin with DNA: cis-[Pt(NH3)2(d(pGpG))]. Science 1985, 230 (4724), 412-417. 19. Cowan, J. A., Inorganic Biochemistry: An introduction. 2 ed.; Wiley-VCH: 1997. 20. Panyala, N. R.; Pena-Mendez, E. M.; Havel, J., Gold and nano-gold in medicine: overview, toxicology and perspectives. J Appl Biomed 2009, 7 (2), 75-91. 21. Hamowy, R. Medical Disasters and the Growth of the FDA; The Independent Institue: 2010. 22. Chellan, P.; Sadler, P. J., The elements of life and medicines. Philos Trans A Math Phys Eng Sci 2015, 373 (2037). 23. Tinoco, A. D.; Thomas, H. R.; Incarvito, C. D.; Saghatelian, A.; Valentine, A. M., Cytotoxicity of a Ti(IV) compound is independent of serum proteins. Proc Natl Acad Sci U S A 2012, 109 (13), 5016-21. 24. Sweeney, N. J.; Mendoza, O.; Müller-Bunz, H.; Pampillón, C.; Rehmann, F.-J. K.; Strohfeldt, K.; Tacke, M., Novel benzyl substituted titanocene anti-cancer drugs. Journal of Organometallic Chemistry 2005, 690 (21-22), 4537-4544. 25. Peri, D.; Meker, S.; Shavit, M.; Tshuva, E. Y., Synthesis, characterization, cytotoxicity, and hydrolytic behavior of C2- and C1-symmetrical Ti(IV) complexes of tetradentate diamine bis(phenolato) ligands: a new class of antitumor agents. Chemistry 2009, 15 (10), 2403-15. 26. Lichawska, M. E.; Bodek, K. H.; Jezierska, J.; Kufelnicki, A., Coordinative interaction of microcrystalline chitosan with oxovanadium (IV) ions in aqueous solution. Chem Cent J 2014, 8 (1), 50. 27. Salvemini, D.; Currie, M. G.; Mollace, V., Nitric oxide-mediated cyclooxygenase activation. A key event in the antiplatelet effects of nitrovasodilators. J Clin Invest 1996, 97 (11), 2562-8. 28. Mathers, W. D.; Sutphin, J. E.; Folberg, R.; Meier, P. A.; Wenzel, R. P.; Elgin, R. G., Outbreak of Keratitis Presumed to Be Caused by Acanthamoeba. American Journal of Ophthalmology 1996, 121 (2), 129-142. 29. Russell, A. D.; Hugo, W. B., 7 Antimicrobial Activity and Action of Silver. 1994, 31, 351-370. 30. Arakawa, H.; Neault, J. F.; Tajmir-Riahi, H. A., Silver(I) Complexes with DNA and RNA Studied by Fourier Transform Infrared Spectroscopy and Capillary Electrophoresis. Biophysical Journal 2001, 81 (3), 1580-1587. 31. Park, H. J.; Kim, J. Y.; Kim, J.; Lee, J. H.; Hahn, J. S.; Gu, M. B.; Yoon, J., Silver-ion-mediated reactive oxygen species generation affecting bactericidal activity. Water Res 2009, 43 (4), 1027-32. 32. Morones-Ramirez, J. R.; Winkler, J. A.; Spina, C. S.; Collins, J. J., Silver enhances antibiotic activity against gram-negative bacteria. Sci Transl Med 2013, 5 (190), 190ra81. 33. Mjos, K. D.; Orvig, C., Metallodrugs in medicinal inorganic chemistry. Chemical reviews 2014, 114 (8), 4540-63.

245

34. Dreaden, E. C.; Alkilany, A. M.; Huang, X.; Murphy, C. J.; El-Sayed, M. A., The golden age: gold nanoparticles for biomedicine. Chemical Society reviews 2012, 41 (7), 2740-79. 35. Jones, T., GSK's novel split-virus adjuvanted vaccines for the prevention of the H5N1 strain of avian influenza infection. Curr Opin Mol Ther 2009, 11 (3), 337-45. 36. Frezard, F.; Demicheli, C.; Ribeiro, R. R., Pentavalent antimonials: new perspectives for old drugs. Molecules 2009, 14 (7), 2317-36. 37. Jin, L.; Szeto, K. Y.; Zhang, L.; Du, W.; Sun, H., Inhibition of alcohol dehydrogenase by bismuth. J Inorg Biochem 2004, 98 (8), 1331-7. 38. Lambert, J. R.; Midolo, P., The actions of bismuth in the treatment of Helicobacter pylori infection. Alimentary Pharmacology and Therapeutics 1997, 11 (S1), 27-33. 39. Camerman, N.; Camerman, A.; Sarkar, B., Molecular design to mimic the copper(II) transport site of human albumin. The crystal and molecular structure of copper(II)-glycylglycyl-L-histidine-N-methyl amide monoaquo complex. Can. J. Chem. 1976, 54, 1309-16. 40. Lau, S.-J.; Kruck, T. P. A.; Sarkar, B., Peptide molecule mimicking the copper(II) transport site of human serum albumin. Comparative study between the synthetic site and albumin. J. Biol. Chem. 1974, 249 (18), 5878-84. 41. Sarkar, B.; Wigfield, Y., Evidence for albumin-Cu(II)-amino acid ternary complex. Can. J. Biochem. 1968, 46, 601-7. 42. Sivertsen, A.; Isaksson, J.; Leiros, H. K.; Svenson, J.; Svendsen, J. S.; Brandsdal, B. O., Synthetic cationic antimicrobial peptides bind with their hydrophobic parts to drug site II of human serum albumin. BMC Struct Biol 2014, 14, 4. 43. Harford, C.; Sarkar, B., Neuromedin C binds Cu(II) and NI(II) via the ATCUN motif: implications for the CNS and cancer growth. Biochem. Biophys. Res. Commun. 1995, 209, 877-82. 44. Grogan, J.; McKnight, C. J.; Troxler, R. F.; Oppenheim, F. G., Zinc and copper bind to unique sites of histatin 5. FEBS Letters 2001, 491, 76-80. 45. Harford, C.; Sarkar, B., Amino Terminal Cu(II)- and Ni(II)-Binding Motif of Proteins and Peptides: Metal Binding, DNA Cleavage, and Other Properties. Acc. Chem. Res. 1997, 30, 123-130. 46. Jin, Y.; Lewis, M. A.; Gokhale, N. H.; Long, E. C.; Cowan, J. A., Influence of stereochemistry and redox potentials on the single- and double-strand DNA cleavage efficiency of Cu(II) and Ni(II) Lys-Gly-His-derived ATCUN metallopeptides. J Am Chem Soc 2007, 129 (26), 8353-61. 47. Joyner, J. C.; Keuper, K. D.; Cowan, J. A., Kinetics and Mechanisms of Oxidative Cleavage of HIV RRE RNA by Rev-Coupled Transition Metal Chelates. Chemical science 2013, 4 (4), 1707-1718. 48. Best, S. L.; Chattopadhyay, T. K.; Djuran, M. I.; Palmer, R. A.; Sadler, P. J.; Sóvágó, I.; Varnagy, K., Gold(III) and palladium(II) complexes of glycylglycyl-L- histidine: crystal structures of [AuIII(Gly-Gly-L-His-H−2)]Cl ·H2O and [PdII(Gly-Gly- L-His-H−2)] ·1.5H2O and HisɛNH deprotonation. Journal of the Chemical Society, Dalton Transactions 1997, (15), 2587-2596. 246

49. Koleva, B. B.; Zareva, S.; Kolev, T.; Spiteller, M., New Au(III), Pt(II) and Pd(II) complexes with glycyl-containing homopeptides. Journal of Coordination Chemistry 2008, 61 (22), 3534-3548. 50. Garribba, E.; Micera, G.; Lodyga-Chruscinska, E.; Sanna, D.; Sanna, G., Binding of Oxovanadium(IV) to Tripeptides Containing Histidine and Cysteine Residues and Its Biological Implication in the Transport of Vanadium and Insulin-Mimetic Compounds. European Journal of Inorganic Chemistry 2005, 2005 (24), 4953-4963. 51. Harford, C.; Sarkar, B., Amino Terminal Cu(II)- and Ni(II)-Binding (ATCUN) Motif of Proteins and Peptides: Metal Binding, DNA Cleavage, and Other Properties†. Accounts of chemical research 1997, 30 (3), 123-130. 52. Mack, D. P.; Dervan, P. B., Sequence-specific oxidative cleavage of DNA by a designed metalloprotein, nickel(II).cntdot.GGH(Hin139-190). Biochemistry 1992, 31 (39), 9399-9405. 53. Shullenberger, D. F.; Eason, P. D.; Long, E. C., Design and synthesis of a versatile DNA-cleaving metallopeptide structural domain. Journal of the American Chemical Society 1993, 115 (23), 11038-11039. 54. Joyner, J. C.; Reichfield, J.; Cowan, J. A., Factors Influencing the DNA Nuclease Activity of Iron, Cobalt, Nickel, and Copper Chelates. Journal of the American Chemical Society 2011, 133 (39), 15613–15626. 55. Gokhale, N., Cowan, J., Inactivation of Human Angiotensin Converting Enzyme by Copper Peptide Complexes Containing ATCUN Motifs. Chemical Communications (Cambridge, United Kingdom) 2005, 5916-5918. 56. Gokhale, N. H.; Bradford, S.; Cowan, J. A., Catalytic Inactivation of Human Carbonic Anhydrase I by a Metallopeptide-Sulfonamide Conjugate is Mediated by Oxidation of Active Site Residues. J. Am. Chem. Soc. 2008, 130, 2388-2389. 57. Gokhale, N. H.; Cowan, J. A., Metallopeptide-promoted inactivation of angiotensin-converting enzyme and endothelin-converting enzyme 1: Toward dual-action therapeutics. J. Biol. Inorg. Chem. 2006, 11, 937-947. 58. Hocharoen, L.; Cowan, J. A., Metallotherapeutics: Novel Strategies in Drug Design. Chem.--Eur. J. 2009, 15 (35), 8670-8676. 59. Joyner, J. C.; Hocharoen, L.; Cowan, J. A., Targeted Catalytic Inactivation of Angiotensin Converting Enzyme by Lisinopril-Coupled Transition Metal Chelates. J. Am. Chem. Soc. 2011. 60. Fidai, I.; Hocharoen, L.; Bradford, S.; Wachnowsky, C.; Cowan, J. A., Inactivation of sortase A mediated by metal ATCUN complexes. J Biol Inorg Chem 2014, 19 (8), 1327-39. 61. Bradford, S. The Design and Evaluation of Catalytic MetalloDrugs Targeting HCV IRES RNA: Demonstration of a New Therapeutic Approach. The Ohio State University, OhioLINK Electronic Theses and Dissertations Center, 2012. 62. Bradford, S.; Cowan, J. A., Catalytic metallodrugs targeting HCV IRES RNA. Chemical Communications (Cambridge, United Kingdom) 2012, 48 (25), 3118-3120. 63. Bradford, S. S.; Ross, M. J.; Fidai, I.; Cowan, J. A., Insight into the recognition, binding, and reactivity of catalytic metallodrugs targeting stem loop IIb of hepatitis C IRES RNA. ChemMedChem 2014, 9 (6), 1275-85. 247

64. Chen, C.-A.; Cowan, J. A., In vivo cleavage of a target RNA by copper kanamycin A. Direct observation by a fluorescence assay. Chemical Communications (Cambridge, United Kingdom) 2002, (3), 196-197. 65. Jin, Y.; Cowan, J. A., Cellular activity of Rev response element RNA targeting metallopeptides. J. Biol. Inorg. Chem. 2007, 12, 637-644. 66. Jin, Y.; Cowan, J. A., Targeted Cleavage of HIV Rev Response Element RNA by Metallopeptide Complexes. J. Am. Chem. Soc. 2006, 128, 410-411. 67. Joyner, J. C.; Cowan, J. A., Targeted Cleavage of HIV RRE RNA by Rev- Coupled Transition Metal Chelates. Journal of the American Chemical Society 2011, 133 (25), 9912–9922. 68. Joyner, J. C.; Keuper, K.; Cowan, J. A., Analysis of RNA Cleavage by MALDI- TOF Mass Spectrometry Nucl. Acids Res. 2012, 41 (1). 69. Bradford, S.; Cowan, J. A., Catalytic MetalloDrugs targeting HCV IRES RNA. Chem. Commun. 2012, 48, 3118-3120. 70. Honda, M.; Beard, M. R.; Ping, L.-H.; Lemon, S. M., A phylogenetically conserved stem-loop structure at the 5' border of the internal ribosome entry site of hepatitis C virus is required for cap-independent viral translation. J. Virol. 1999, 73, 1165-1174. 71. Lukavsky, P. J.; Kim, I.; Otto, G. A.; Puglisi, J. D., Structure of HCV IRES domain II determined by NMR. Nat. Struct. Biol. 2003, 10, 1033-1038. 72. Erspamer, V.; Melchiorri, P.; Broccardo, M.; Erspamer, G. F.; Falaschi, P.; Improota, G.; Negri, L.; Renda, T., The brain-gut-skin triangle: new peptides. Peptides 1981, 2 Suppl 2, 7-16. 73. Montecucchi, P. C.; De Castiglione, R.; Piani, S.; Gozzini, L.; Erspamer, V., Amino acid composition and sequence of dermorphin, a novel opiate-like peptide from the skin of Phyllomedusa sauvagei. Int. J. Pept. Protein Res. 1981, 17, 275-83. 74. Milton, R. C. L.; Milton, S. C. F.; Kent, S. B. H., Total chemical synthesis of a D- enzyme: the enantiomers of HIV-1 protease show demonstration of reciprocal chiral substrate specificity. Science 1992, 256 (5062), 1445-8. 75. Pritsker, M.; Jones, P.; Blumenthal, R.; Shai, Y., A synthetic all D-amino acid peptide corresponding to the N-terminal sequence of HIV-1 gp41 recognizes the wild- type fusion peptide in the membrane and inhibits HIV-1 envelope glycoprotein-mediated cell fusion. Proc. Natl. Acad. Sci. U. S. A. 1998, 95 (13), 7287-7292. 76. Litovchick, A.; Rando, R. R., Stereospecificity of short Rev-derived peptide interactions with RREIIB RNA. RNA 2003, 9, 937-948. 77. Wade, D.; Boman, A.; Waahlin, B.; Drain, C. M.; Andreu, D.; Boman, H. G.; Merrifield, R. B., All-D-amino acid-containing channel-forming antibiotic peptides. Proc. Natl. Acad. Sci. U. S. A. 1990, 87 (12), 4761-5. 78. Chalifour, R. J.; McLaughlin, R. W.; Lavoie, L.; Morissette, C.; Tremblay, N.; Boule, M.; Sarazin, P.; Stea, D.; Lacombe, D.; Tremblay, P.; Gervais, F., Stereoselective Interactions of Peptide Inhibitors with the Î²-Amyloid Peptide. J. Biol. Chem. 2003, 278 (37), 34874-34881. 79. Darty, K.; Denise, A.; Ponty, Y., VARNA: Interactive drawing and editing of the RNA secondary structure. Bioinformatics 2009, 25 (15), 1974-1975. 248

80. M. J. Frisch, G. W. T., H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian 09, Revision A.1. Gaussian, Inc., Wallingford CT 2009. 81. Saielli, G., Differential Solvation Free Energies of Oxonium and Ammonium Ions: Insights from Quantum Chemical Calculations. The Journal of Physical Chemistry A 2010, 114 (26), 7261–7265. 82. Chen, C.; Ridzon, D. A.; Broomer, A. J.; Zhou, Z.; Lee, D. H.; Nguyen, J. T.; Barbisin, M.; Xu, N. L.; Mahuvakar, V. R.; Andersen, M. R.; Lao, K. Q.; Livak, K. J.; Guegler, K. J., Real-time quantification of microRNAs by stem-loop RT-PCR. Nucleic Acids Research 2005, 33 (20), e179/1-e179/9. 83. Kim, I.; Lukavsky, P. J.; Puglisi, J. D., NMR Study of 100 kDa HCV IRES RNA Using Segmental Isotope Labeling. J. Am. Chem. Soc. 2002, 124 (32), 9338-9339. 84. Andersen, T. E.; Kirpekar, F.; Haselmann, K. F., RNA Fragmentation in MALDI Mass Spectrometry Studied by H/D-Exchange: Mechanisms of General Applicability to Nucleic Acids. Journal of the American Society for Mass Spectrometry 2006, 17 (10), 1353-1368. 85. Morris, G. M., Huey, R., Lindstrom, W., Sanner, M. F., Belew, R. K., Goodsell, D.S., Olson, A. J. , Journal of Computational Chemistry 2009, 30 (16), 2485-2791. 86. Park, H.; Kim, S.; Kim, Y. E.; Lim, S.-J., A Structure-Based Virtual Screening Approach toward the Discovery of Histone Deacetylase Inhibitors: Identification of Promising Zinc-Chelating Groups. ChemMedChem 2010, 5 (4), 591-597. 87. Samii, A.; Bickel, U.; Stroth, U.; Pardridge, W. M., Blood-brain barrier transport of neuropeptides: analysis with a metabolically stable dermorphin analog. Am. J. Physiol. 1994, 267, E124-E131. 88. Szeto, H.; Zhao, K.; Birk, A. V.; Robertson, H. D. Method and carrier complexes for delivering molecules to cells. 2005001023, 2005. 89. Thorp, H. H., The importance of being r: greater oxidative stability of RNA compared with DNA. Chemistry & Biology 2000, 7 (2), R33-R36. 90. Bossu, F. P.; Chellappa, K. L.; Margerum, D. W., Ligand effects on the thermodynamic stabilization of copper(III)-peptide complexes. J Am Chem Soc 1977, 99 (7), 2195-203. 91. Margerum, D. W.; Owens, G. D., Copper(III) complexes and their reactions. Metal Ions Biol Syst 1981, 12, 75-132.

249

92. McDonald, M. R.; Fredericks, F. C.; Margerum, D. W., Characterization of Copper(III)-Tetrapeptide Complexes with Histidine as the Third Residue. Inorganic Chemistry 1997, 36 (14), 3119-3124. 93. Berry, K. E.; Waghray, S.; Mortimer, S. A.; Bai, Y.; Doudna, J. A., Crystal Structure of the HCV IRES Central Domain Reveals Strategy for Start-Codon Positioning. Structure (Cambridge, MA, United States) 2011, 19 (10), 1456-1466. 94. Kalliampakou, K. I.; Psaridi-Linardaki, L.; Mavromara, P., Mutational analysis of the apical region of domain II of the HCV IRES. FEBS Letters 2002, 511 (1-3), 79-84. 95. Markham, R. S., J. D., The Structure of Ribonucleic Acids. Journal of Biochemistry 1952, 52 (4), 552-557. 96. Balasubramanian, B. P., W; Tullius, T. D., DNA strand breaking by the hydroxyl radical is governed by the accessible surface areas of the hydrogen atoms of the DNA backbone. Proc. Natl. Acad. Sci. 1998, 95 (17), 9738-9743. 97. Black, C. B.; Cowan, J. A., Quantitative Evaluation of Electrostatic and Hydrogen-Bonding Contributions to Metal Cofactor Binding to Nucleic Acids. J. Am. Chem. Soc. 1994, 116, 1174-1178. 98. Gilead Sciences Announces Fourth Quarter and Full Year 2014 Financial Results. Gilead Sciences, Inc.: 2015. 99. Barua, S.; Greenwald, R.; Grebely, J.; Dore, G. J.; Swan, T.; Taylor, L. E., Restrictions for Medicaid Reimbursement of Sofosbuvir for the Treatment of Hepatitis C Virus Infection in the United States. Ann Intern Med 2015. 100. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery Jr., J. A.; J. E. Peralta; F. Ogliaro; M. Bearpark; J. J. Heyd; E. Brothers; K. N. Kudin; V. N. Staroverov; R. Kobayashi; J. Normand; K. Raghavachari; A. Rendell; J. C. Burant; S. S. Iyengar; J. Tomasi; M. Cossi; N. Rega; J. M. Millam; M. Klene; J. E. Knox; J. B. Cross; V. Bakken; C. Adamo; J. Jaramillo; R. Gomperts; R. E. Stratmann; O. Yazyev; A. J. Austin; R. Cammi; C. Pomelli; J. W. Ochterski; R. L. Martin; K. Morokuma; V. G. Zakrzewski; G. A. Voth; P. Salvador; J. J. Dannenberg; S. Dapprich; A. D. Daniels; Ö. Farkas; J. B. Foresman; J. V. Ortiz; J. Cioslowski; Fox, D. J., Gaussian 09, Revision A.1. Gaussian, Inc., Wallingford CT 2009. 101. Saielli, G., Differential Solvation Free Energies of Oxonium and Ammonium Ions: Insights from Quantum Chemical Calculations. J. Phys. Chem. A 2010, 114 (26), 7261–7265. 102. Wilson, D. T. Combination therapies using cyclosporine and aromatic cationic peptides. 2014. 103. Wilson, D. T.; Lerman, L. O.; Textor, S. C. Aromatic-cationic peptides and methods for using same. PCT/US2012/058091, Sep. 28, 2012, 2014. 104. Szeto, H. H.; Schiller, P. W., Novel therapies targeting inner mitochondrial membrane--from discovery to clinical development. Pharm Res 2011, 28 (11), 2669-79.

250

105. Gopal, R.; Seo, C. H.; Song, P. I.; Park, Y., Effect of repetitive lysine-tryptophan motifs on the bactericidal activity of antimicrobial peptides. Amino Acids 2013, 44 (2), 645-60. 106. Pletneva, E. V.; Laederach, A. T.; Fulton, D. B.; Kostić, N. M., The Role of Cation−π Interactions in Biomolecular Association. Design of Peptides Favoring Interactions between Cationic and Aromatic Amino Acid Side Chains. Journal of the American Chemical Society 2001, 123 (26), 6232-6245. 107. Kyte, J.; Doolittle, R. F., A simple method for displaying the hydropathic character of a protein. Journal of Molecular Biology 1982, 157 (1), 105-132. 108. Friedman, S. L., "There are decades where nothing happens; and there are weeks where decades happen"--Vladimir Ilyich Lenin. J Hepatol 2014, 60 (3), 471-2. 109. Hawkins, C. J.; Martin, J., Cobalt(III) complex of glycylglycyl-L-histidine: preparation, characterization, and conformation. Inorganic Chemistry 1983, 22 (26), 3879-3883. 110. Glisic, B.; Stanic, Z.; Rajkovic, S.; Kojic, V.; Bogdanovic, G.; Djuran, M., Solution study under physiological conditions and cytotoxic activity of the gold(III) complexes with L-histidine-containing peptides. Journal of the Serbian Chemical Society 2013, 78 (12), 1911-1924. 111. Lammer, A. D.; Cook, M. E.; Sessler, J. L., Synthesis and anti-cancer activities of a water soluble gold(III) porphyrin. J Porphyr Phthalocyanines 2015, 19 (1-03), 398-403. 112. Johnson, A.; Shiraishi, T., Biocompatibility of precious metals for medical applications. 2014, 37-55. 113. Arvizo, R. R.; Bhattacharyya, S.; Kudgus, R. A.; Giri, K.; Bhattacharya, R.; Mukherjee, P., Intrinsic therapeutic applications of noble metal nanoparticles: past, present and future. Chemical Society reviews 2012, 41 (7), 2943-70. 114. Jackson-Rosario, S.; Cowart, D.; Myers, A.; Tarrien, R.; Levine, R. L.; Scott, R. A.; Self, W. T., Auranofin disrupts selenium metabolism in Clostridium difficile by forming a stable Au-Se adduct. J Biol Inorg Chem 2009, 14 (4), 507-19. 115. Tejman-Yarden, N.; Miyamoto, Y.; Leitsch, D.; Santini, J.; Debnath, A.; Gut, J.; McKerrow, J. H.; Reed, S. L.; Eckmann, L., A reprofiled drug, auranofin, is effective against metronidazole-resistant Giardia lamblia. Antimicrob Agents Chemother 2013, 57 (5), 2029-35. 116. Cassetta, M. I.; Marzo, T.; Fallani, S.; Novelli, A.; Messori, L., Drug repositioning: auranofin as a prospective antimicrobial agent for the treatment of severe staphylococcal infections. Biometals 2014, 27 (4), 787-91. 117. Giandomenico, C. M.; Abrams, M. J.; Murrer, B. A.; Vollano, J. F.; Rheinheimer, M. I.; Wyer, S. B.; Bossard, G. E.; Higgins, J. D., Carboxylation of Kinetically Inert Platinum(IV) Hydroxy Complexes. An Entr.acte.ee into Orally Active Platinum(IV) Antitumor Agents. Inorg Chem 1995, 34 (5), 1015-21. 118. Habib, A.; Tabata, M., Oxidative DNA damage induced by HEPES (2-[4-(2- hydroxyethyl)-1-piperazinyl]ethanesulfonic acid) buffer in the presence of Au(III). J Inorg Biochem 2004, 98 (11), 1696-702. 119. Messori, L.; Orioli, P.; Tempi, C.; Marcon, G., Interactions of selected gold(III) complexes with calf thymus DNA. Biochem Biophys Res Commun 2001, 281 (2), 352-60. 251

120. Marcon, G.; Carotti, S.; Coronnello, M.; Messori, L.; Mini, E.; Orioli, P.; Mazzei, T.; Cinellu, M. A.; Minghetti, G., Gold(III) Complexes with Bipyridyl Ligands: Solution Chemistry, Cytotoxicity, and DNA Binding Properties. Journal of Medicinal Chemistry 2002, 45 (8), 1672-1677. 121. Pogozelski, W. K.; Tullius, T. D., Oxidative Strand Scission of Nucleic Acids: Routes Initiated by Hydrogen Abstraction from the Sugar Moiety. Chemical Reviews (Washington, D. C.) 1998, 98 (3), 1089-1107. 122. Shimakami, T.; Lanford, R. E.; Lemon, S. M., Hepatitis C: recent successes and continuing challenges in the development of improved treatment modalities. Curr. Opin. Pharmacol. 2009, 9, 537-544. 123. Bradford, S. S.; Ross, M. J.; Fidai, I.; Cowan, J. A., Insight into the Recognition, Binding, and Reactivity of Catalytic Metallodrugs Targeting Stem Loop IIb of Hepatitis C IRES RNA. ChemMedChem 2014, 9, 1275–1285. 124. Joyner, J. C.; Cowan, J. A., Targeted Cleavage of HIV RRE RNA by Rev- Coupled Transition Metal Chelates. J. Am. Chem. Soc. 2011, 133 (25), 9912–9922. 125. Suh, J.; Chei, W. S., Metal complexes as artificial proteases: toward catalytic drugs. Curr. Op. Chem. Biol. 2008, 12, 207-213. 126. Suh, J.; Chei, W. S.; Lee, T. Y.; Kim, M. G.; Yoo, S. H.; Jeong, K.; Ahn, J. Y., Cleavage agents for soluble oligomers of human islet amyloid polypeptide. J. Biol. Inorg. Chem. 2008, 13, 693-701. 127. Suh, J.; Yoo, S. H.; Kim, G.; Chae, P. S.; Lee, T. Y.; Lee, J.; Lee, J.; Jang, Y. A.; Ko, E. H., Cleavage Agents for Soluble Oligomers of Amyloid Beta Peptides. Angew. Chem. 2007, 119 (37), 7194-7197. 128. Sarnow, P., Viral internal ribosome entry site elements: novel ribosome-RNA complexes and roles in viral pathogenesis. J. Virol. 2003, 77, 2801-2806. 129. Subramanian, N.; Mani, P.; Roy, S.; Gnanasundram, S. V.; Sarkar, D. P.; Das, S., Targeted delivery of hepatitis C virus-specific short hairpin RNA in mouse liver using Sendai virosomes. Journal of General Virology 2009, 90 (8), 1812-1819. 130. Pudi, R.; Ramamurthy, S. S.; Das, S., A peptide derived from RNA recognition motif 2 of Human La protein binds to hepatitis C virus internal ribosome entry site, prevents ribosomal assembly, and inhibits internal initiation of translation. J. Virol. 2005, 79 (15), 9842-9853. 131. Hocharoen, L.; Cowan, J. A., Metallotherapeutics - Novel Strategies in Drug Design. Chem. Eur. J. 2009, 15, 8670-76. 132. Perard, J.; Leyrat, C.; Baudin, F.; Drouet, E.; Jamin, M., Structure of the full- length HCV IRES in solution. . Nature Commun. 2013, 4, 1612. 133. Pudi, R.; Abhiman, S.; Srinivasan, N.; Das, S., Hepatitis C Virus Internal Ribosome Entry Site-mediated Translation Is Stimulated by Specific Interaction of Independent Regions of Human La Autoantigen. J. Biol. Chem. 2003, 278 (14), 12231- 12240. 134. Mondal, T.; Ray, U.; Manna, A. K.; Gupta, R.; Roy, S.; Das, S., Structural determinant of human La protein critical for internal initiation of translation of hepatitis C virus RNA. J. Virol. 2008, 82 (23), 11927-11938.

252

135. Qin, N.; Chen, Z.; Yao, H.-P.; Yang, Z.-G.; Liu, K.-Z.; Wu, L.-L., Inhibition of human La protein by RNA interference downregulates hepatitis B virus mRNA in 2.2.15 cells. World Journal of Gastroenterology 2004, 10 (14), 2050-2054. 136. Alfano, C.; Sanfelice, D.; Babon, J.; Kelly, G.; Jacks, A.; Curry, S.; Conte, M. R., Structural analysis of cooperative RNA binding by the La motif and central RRM domain of human La protein. Nature Structural & Molecular Biology 2004, 11 (4), 323-329. 137. Siridechadilok, B.; Fraser, C. S.; Hall, R. J.; Doudna, J. A.; Nogales, E., Structural roles for human translation factor eIF3 in initiation of protein synthesis. Science 2005, 310, 1513-1515. 138. Berry, K. E.; Waghray, S.; Mortimer, S. A.; Bai, Y.; Doudna, J. A., Crystal Structure of the HCV IRES Central Domain Reveals Strategy for Start-Codon Positioning. Structure (Cambridge, MA, United States) 2011, 19 (10), 1456-1466. 139. Lloyd, R. S.; Haidle, C. W.; Robberson, D. L., Noncovalent intermolecular crosslinks are produced by bleomycin reaction with duplex DNA. Proc Natl Acad Sci U S A 1979, 76 (6), 2674-8. 140. Andersen, T. E.; Kirpekar, F.; Haselmann, K. F., RNA Fragmentation in MALDI Mass Spectrometry Studied by H/D-Exchange: Mechanisms of General Applicability to Nucleic Acids. J. Am. Soc. Mass Spectr. 2006, 17 (10), 1353–1368. 141. Pogozelski, W.; Tullius, T., Oxidative Strand Scission of Nucleic Acids: Routes Initiated by Hydrogen Abstraction from the Sugar Moiety. Chemical Reviews (Washington, D. C.) 1998, 98, 1089-1107.

253