Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A pursuit of sequence specific RNase A mimicking DNAzymes Wang, Yajun 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_may_wang_yajun.pdf [ 52.89MB ]
Metadata
JSON: 24-1.0229566.json
JSON-LD: 24-1.0229566-ld.json
RDF/XML (Pretty): 24-1.0229566-rdf.xml
RDF/JSON: 24-1.0229566-rdf.json
Turtle: 24-1.0229566-turtle.txt
N-Triples: 24-1.0229566-rdf-ntriples.txt
Original Record: 24-1.0229566-source.json
Full Text
24-1.0229566-fulltext.txt
Citation
24-1.0229566.ris

Full Text

A PURSUIT OF SEQUENCE SPECIFIC RNASE A MIMICKING DNAZYMES by  Yajun Wang  M.Sc., Peking Union Medical College & Medical School of Tsinghua University, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   April 2016  © Yajun Wang, 2016 ii  Abstract  DNAzymes are single-stranded DNA molecules capable of catalysis, and they are the DNA counterparts of ribozymes and protein enzymes. M2+-independent RNA cleaving DNAzymes are intriguing due to their robust activity that is not compromised by low Mg2+ that is found in cells. Hence, they hold great promise for RNA regulation in vivo. To mimic the metal independent protein endonuclease RNase A, three chemically modified nucleotides dAimTP, dCaaTP, and dUgaTP, that are adorned with respective side-chain functionalities of histidine, lysine, and arginine, have been simultaneously introduced in in vitro selections by our group, and led to the development of families of M2+-independent DNAzymes targeting chimeric DNA/RNA substrates, attaining kobs as high as ~0.6 min-1 at pH 7.4, 37°C. In order to further select such DNAzymes capable of highly efficient all-RNA cleavage and multiple-turnover, a novel unimolecular selection scheme containing an all-RNA substrate derived from the HIV-1 LTR-promoter allows a direct selection of all-RNA cleavers meanwhile fostering the subsequent conversion of the cis-cleaving species into a trans-acting catalyst was constructed. An optimized in vitro selection cycle combining selection, re-selection, and evolution that permitted greater sequence space sampling and pursuit of catalytically improved sequences through generation-specific mutagenesis was designed. The application of the novel construct in the optimized in vitro selection cycle gave rise to two families of desired DNAzyme candidates. Under simulated physiological conditions (pH 7.45, 150 mM K+, 0.5 mM Mg2+, 37°C), the best representative, Dz7-38-32t, attained kcat and KM values of ~0.24 min-1 and 2.72 µM, respectively, corresponding to a catalytic efficiency of ~105 M-iii  1min-1. Dz7-38-32 can be spontaneously taken up by HeLa cells after 45 h incubation at 0.9 µM due to its similarity to cell penetrating peptides (CPPs) regarding the appended functionalities. To understand fundamental aspects by which these three modified bases function in in vitro selection, investigations of biophysical and enzymatic properties of them in the context of discretely modified oligonucleotides were performed. These studies identified certain shortcomings in the use of modified nucleosides while providing clear evidence of negligible mutagenicity in terms of both primer extension and enzymatic read-through.   iv  Preface  This thesis is submitted as part of the requirements for the degree of Doctor of Philosophy in Chemistry at the University of British Columbia. All the experiments in this thesis were designed primarily by Dr. David M. Perrin and myself unless otherwise indicated.  The modified deoxynucleotides: 8-histaminyl-adenosine triphosphate and 5-guanidinoallyl-uridine triphosphate used in this thesis were previously synthesized by Dr. Curtis H Lam.   Chapter 2. All the oligonucleotide sequences and constructs for in vitro selection were designed by myself. All the experiments were performed by myself. The DH5α competent cells for cloning were provided by Dr. Jessie Chen and Dr. Elena Polishchuk at the Department of Chemistry’s BioServices Laboratory. This chapter was written by myself.   Chapter 3. This chapter was written by myself. All the characterization of enzymatic parameters, elucidation of catalytic properties, and probing of RNase A-mimicking mechanisms were performed by myself. The work described in this chapter is included in a manuscript in preparation: Densely modified, Highly-active, Turnover-capable DNAzymes Operating under Simulated Physiological Conditions.   Chapter 4. This chapter was written by myself. All the cell culture and imaging work were performed by myself. HeLa cells used by this chapter were provided by Dr. Jessie Chen and Dr. Elena Polishchuk at the Department of Chemistry’s BioServices Laboratory. The bright field imaging was performed in the Department of Chemistry’s v  BioServices Laboratory. The confocal imaging was performed at BioImaging Facility at the University of British Columbia.  Chapter 5. This chapter was written by myself. This work has been submitted for publication: Yajun Wang, Nicole Ng, Erkai Liu, Curtis H. Lam, and David M. Perrin. A Systematic Study of Constraints Imposed by Modified Nucleosides with Protein-like Side Chains: Imidazole-dA, Guanidino-allyl-dU, and Amino-allyl-dC. This manuscript was written by me and David M. Perrin. The phosphoramidites were synthesized by Dr. Erkai Liu and Dr. Curtis H Lam; oligonucleotide synthesis was performed by Dr. David Sabatino; biophysical and enzymatc investigations were performed by me and Nicole Ng. Only the experiments from the manuscript that were performed by myself have been included in this chapter.     vi  Table of Contents  Abstract ........................................................................................................................... ii Preface ........................................................................................................................... iv Table of Contents .......................................................................................................... vi List of Tables ................................................................................................................ xii List of Figures .............................................................................................................. xiii List of Abbreviations .................................................................................................. xvii Acknowledgements ..................................................................................................... xxi Chapter 1: General Introduction to the Chemistry and Catalysis of RNA Cleavage Reaction .......................................................................................................................... 1 1.1 Nucleic Acids and In Vitro Selection .................................................................. 1 1.1.1 Nucleic acids and naturally-occurring ribozymes ........................................... 1 1.1.2 In vitro combinatorial selection (SELEX) and artificial ribozymes .................. 2 1.1.3 Artificial DNA enzymes/ deoxyribozymes / DNAzymes .................................. 4 1.2 Comparison of DNAzymes with Ribozymes and Proteins: Motivation for Developing DNAzymes ................................................................................................. 5 1.2.1 Comparison of DNAzymes to ribozymes ........................................................ 5 1.2.2 Comparison of DNAzymes to protein enzymes .............................................. 6 1.2.3 Advantages of synthetic DNAzymes over synthetic ribozymes ...................... 7 1.3 RNA Cleavage Reaction: Knowledge about RNA Targets ................................ 8 1.3.1 RNA cleavage by transesterification reaction ................................................. 8 1.3.2 Transesterification reaction catalyzed by small self-cleaving ribozymes ..... 10 vii  1.3.3 In vitro selection methodology of RNA-cleaving DNAzymes ........................ 12 1.3.4 Development of RNA-cleaving DNAzymes for therapeutic applications: lessons learned ....................................................................................................... 13 1.3.5 M2+-independent RNA-cleaving DNAzymes ................................................. 18 1.4 Transesterification RNA Cleavage Reaction Catalyzed by RNase A .............. 19 1.5 Towards the Development of Sequence Specific and M2+-independent RNase A Mimics ..................................................................................................................... 20 1.5.1 RNase A mimics based on rational design ................................................... 20 1.5.2 In vitro combinatorial selection of modified DNAzymes with functionalities derived from RNase A ............................................................................................. 22 1.5.3 Further expanding the functionalities of DNAzymes .................................... 28 1.6 Potential Advantages of Modified DNAzymes as Therapeutic Candidates for RNA Manipulation In Vivo ........................................................................................... 32 1.7 Specific Research Aims ................................................................................... 33 Chapter 2: Towards the Design of a Generalizable In Vitro Selection Strategy for Modified All-RNA Cleaving DNAzymes ...................................................................... 36 2.1 Introduction ...................................................................................................... 36 2.2 Optimizing Conditions for More Efficient Primer Extension Reactions to Address the Production of Modified DNA ................................................................... 38 2.3 Unimolecular Construct Optimization for In Vitro Selection ............................. 42 2.4 In Vitro Selection .............................................................................................. 47 2.4.1 In vitro selection of DNAzymes cleaving a chimeric substrate containing three embedded RNA linkages ............................................................................... 50 viii  2.4.2 In vitro selection of all-RNA cleaving DNAzymes ......................................... 54 2.5 Cloning, Sequencing, and Single-clone Activity Screening ............................. 56 2.5.1 Cloning and sequencing of the selection targeting the chimeric substrate containing three embedded RNAs .......................................................................... 56 2.5.2 Cloning and sequencing of the 7th Generation (G7) capable of cleaving an all-RNA substrate .................................................................................................... 59 2.6 Re-selection and Evolution of All-RNA Cleaving DNAzymes .......................... 63 2.6.1 Initiation of two lineages of re-selections ...................................................... 63 2.6.2 Mutagenic PCR ............................................................................................ 64 2.6.3 Re-selection progress .................................................................................. 68 2.6.4 Analysis of resulting sequences form re-selection Family I .......................... 71 2.6.5 Analysis of resulting sequences form re-selection Family II ......................... 73 2.7 Re-selection and Evolution of Degenerated Dz12-91 ...................................... 74 2.7.1 Re-selection process of the degenerated Dz12-91 ...................................... 76 2.7.2 Alignment and analysis of sequences resulting from re-selection of degenerated Dz12-91 ............................................................................................. 78 2.8 Discussions and Conclusion ............................................................................ 81 2.9 Experimental .................................................................................................... 85 2.9.1 General reagents, materials, and enzymes .................................................. 85 2.9.2 General protocols ......................................................................................... 85 2.9.3 Optimizing conditions for more efficient primer extension reactions to address the production of modified DNA ................................................................ 86 2.9.4 In vitro selection ........................................................................................... 88 ix  Chapter 3: Characterization of Dz7-38-32, A Representative RNase A-mimicking DNAzyme ..................................................................................................................... 101 3.1 Introduction .................................................................................................... 101 3.2 Characterization of Cleavage Activity In Cis .................................................. 102 3.2.1 Cis-cleavage kinetics .................................................................................. 102 3.2.2 The simultaneous need of all three modifications ...................................... 107 3.2.3 Effect of temperature on cis-cleavage activity of Dz7-38-32 ...................... 110 3.3 Characterizations In Trans ............................................................................. 112 3.3.1 Conversion of cis-acting DNAzymes to trans-acting species ..................... 112 3.3.2 Trans-acting characterization of Dz7-38-32t .............................................. 116 3.3.3 Attempts to increase catalytic efficiency at 37°C by lengthening the substrate-recognition arms .................................................................................... 121 3.3.4 Probing the function of 8-histaminyl-deoxyadenosine in the 3’ substrate-recognition arm of Dz7-38-32t ............................................................................... 122 3.3.5 Activity of Dz7-38-32t on a chimeric DNA/RNA substrate containing exclusive DNA bases in substrate-recognition arms ............................................. 125 3.4 Effect of Varying the RNA Bases at the Cleavage Site .................................. 127 3.5 Analysis of Cleavage Product ........................................................................ 130 3.6 Analysis of pH Dependence ........................................................................... 132 3.7 Divalent Metal Ion Effects on Cleavage Activity ............................................. 136 3.8 Discussions and Conclusion .......................................................................... 139 3.9 Experimental .................................................................................................. 141 3.9.1 General reagents, materials, and enzymes ................................................ 141 x  3.9.2 General protocols ....................................................................................... 142 3.9.3 Buffers/solutions ......................................................................................... 142 3.9.4 Oligonucleotides (5’ to 3’, ON) ................................................................... 143 3.9.5 Cis-cleavage characterization .................................................................... 145 3.9.6 Trans-cleavage characterization ................................................................ 148 Chapter 4: Image Based Analysis of Spontaneous Cellular Uptake of  Dz7-38-32t ................................................................................................................... 155 4.1 Introduction .................................................................................................... 155 4.2 Preparation of Rhodamine-labeled Dz7-38-32t ............................................. 158 4.3 Cellular Uptake Assay .................................................................................... 160 4.4 Discussions and Conclusion .......................................................................... 165 4.5 Experimental .................................................................................................. 167 4.5.1 General reagents and materials for cell culture, enzymes, and oligonucleotides .................................................................................................... 167 4.5.2 Preparation of Dz7-38-32t-m and Dz7-38-32t-u ......................................... 167 4.5.3 Cellular uptake ........................................................................................... 170 Chapter 5: A Systematic Study of Constraints on In Vitro Selection Imposed by Modified Nucleosides Histaminyl-dA, Guanidinoallyl-dU, and Aminoallyl-dC ..... 172 5.1 Introduction .................................................................................................... 172 5.2 Impact of Modified Nucleosides on Duplex Thermal Stability ........................ 175 5.3 Base Pairing Properties of Modified Bases .................................................... 178 5.3.1 Base-pairing properties of dAim .................................................................. 178 5.3.2 Base-pairing properties of dUga .................................................................. 180 xi  5.3.3 Base-pairing properties of dCaa .................................................................. 181 5.4 Full-Strand Synthesis Continuing Beyond Multiple Modified Pairs ................ 184 5.5 Overall Fidelity of a Complete Replication Cycle ........................................... 185 5.6 Discussions and Conclusion .......................................................................... 191 5.7 Experimental .................................................................................................. 193 5.7.1 General ....................................................................................................... 193 5.7.2 Oligonucleotides (ON, 5’ to 3’) ................................................................... 194 5.7.3 Buffers/solutions ......................................................................................... 195 5.7.4 De-protection of oligonucleotides ............................................................... 196 5.7.5 Thermal melting temperature measurement .............................................. 196 5.7.6 Standing-start single-nucleotide insertion survey ....................................... 197 5.7.7 Strand synthesis continuing beyond multiple modified base pairs ............. 198 5.7.8 Test of fidelity of a complete replication cycle ............................................ 199 Chapter 6: Summary and Future Work ..................................................................... 202 6.1 Summary of Research ................................................................................... 202 Bibliography ............................................................................................................... 207 Appendices ................................................................................................................. 225 Appendix A Chapter 4 ............................................................................................... 225 Appendix B Chapter 5 ............................................................................................... 227  xii  List of Tables  Table 2.1 RNase A-mimicking modified DNAzymes developed in the Perrin lab .......... 37 Table 2.2 Clone sequences isolated from selection against the chimeric substrate containing three embedded RNA linkages ..................................................................... 58 Table 2.3 Active sequences isolated from selection of all-RNA cleaving DNAzymes .... 61 Table 2.4 Conditions for standard PCR, mutagenic PCR devised by Cadwell and Joyce, and mutagenic PCR we adapted from Cadwell and Joyce ............................................ 68 Table 2.5 Sequence alignment of the initial 10 clones with improved catalytic rates (kobs) compared with wild-type Dz7-38 .................................................................................... 72 Table 2.6 Top 10 active sequence alignment referring to wild-type Dz7-45 (W.T) ........ 74 Table 2.7 Active sequences from the re-selection of degenerated Dz12-91 aligned in reference to the wild-type Dz12-91 (W.T) ...................................................................... 79 Table 3.1 Two families of highly conserved sequences ............................................... 103 Table 3.2 Biphasic enzymatic parameters of Dz7-38-32 and Dz7-45-28 ..................... 107 Table 3.3 Effect of temperature on cis-cleavage activity of Dz7-38-32 ........................ 111 Table 3.4 Trans-acting enzymatic parameters of Dz7-38-32t ...................................... 119 Table 3.5 Effects of varying the RNA bases at the cleavage site ................................. 130 Table 3.6 Effects of divalent metal ions on trans-cleavage reaction of Dz7-38-32t ..... 139 Table 5.1 Thermal melting temperatures (Tm) of duplexes formed by ON 1, ON 2, and ON3 with the complementary ON 4 in pH variant buffers, respectively ....................... 177  xiii  List of Figures  Figure 1.1 Schematic representation of in vitro selection. ................................................ 3 Figure 1.2 Mechanism of RNA cleavage reaction through internal transesterification reaction. ............................................................................................................................ 9 Figure 1.3 Scheme for in vitro selection of RNA-cleaving DNAzymes. .......................... 13 Figure 1.4 Composition of the Dz10-23 and Dz8-17 catalytic motifs. ............................ 15 Figure 1.5 RNase A active site structure and catalytic mechanism. .............................. 20 Figure 1.6 Sequence specific RNase A mimics generated by rational design. .............. 22 Figure 1.7 Dz16.2-11 containing C5-imidazole-functionalized deoxyuridine. ................. 24 Figure 1.8 RNase A-mimicking Dz925-11. ...................................................................... 27 Figure 1.9 Sidorov's modified DNAzyme. ....................................................................... 28 Figure 1.10 RNase A-mimicking DNAzymes containing three modified nucleosides. ... 31 Figure 2.1 Chemical structures of modified nucleotides for RNase A mimicry. .............. 39 Figure 2.2 Template-directed primer extension reaction using Sequenase v2.0 in the presence of dAimTP, dTTP, dCTP, and dGTP at different temperatures. ...................... 41 Figure 2.3 5’-biotinylated unimolecular constructs for in vitro selections. ...................... 44 Figure 2.4 Sequence schemes used for selections. ....................................................... 46 Figure 2.5 In vitro selection cycle for RNA-cleaving modified DNAzymes. .................... 49 Figure 2.6 Representative gel pictures of different rounds of selection against the chimeric substrate containing three embedded RNA linkages. ...................................... 52 Figure 2.7 In vitro selection progress of DNAzymes cleaving the chimeric substrate containing three embedded RNA linkages. .................................................................... 53 xiv  Figure 2.8 In vitro selection progress of all-RNA cleaving DNAzymes. .......................... 55 Figure 2.9 Comparison of G10 to G12 on their activity of cleaving the chimeric substrate containing three RNA linkages in standard cleavage buffers. ........................................ 56 Figure 2.10 Activity comparison of G6 to G9 in cleavage buffer of simulated physiological conditions. ................................................................................................. 60 Figure 2.11 Kinetic self-cleavage reactions of representative clones. ........................... 62 Figure 2.12 Partially degenerated libraries for re-selection. ........................................... 64 Figure 2.13 Functional components of a nucleic acid-based evolution system. ............ 65 Figure 2.14 Re-selection progress of all-RNA cleaving DNAzymes. .............................. 70 Figure 2.15 Construct of Dz12-91 for re-selection and evolution. .................................. 76 Figure 2.16 Re-selection progress of degenerated Dz12-91. ........................................ 77 Figure 2.17 Cis-cleavage reaction of Dz12-91-5. ........................................................... 81 Figure 3.1 Hypothetical cleavage mechanism of chemically modified RNase A-mimicking DNAzymes. ................................................................................................. 102 Figure 3.2 Cis-cleavage activity of Dz7-38-32 and Dz7-45-28. .................................... 106 Figure 3.3 The necessity of individual modifications for catalytic competence of Dz7-38-32. ................................................................................................................................ 109 Figure 3.4 Temperature profile of cis-cleavage activity of Dz7-32-38. ......................... 112 Figure 3.5 Trans-acting DNAzymes. ............................................................................ 114 Figure 3.6 Qualitative trans-cleavage activity assay of Dz7-38-32t and Dz7-45-28t. ... 116 Figure 3.7 Multiple-turnover catalytic activity of Dz7-38-32t (5 nM) under simulated physiological ionic strength: 0.5 mM Mg2+, 150 mM K+, pH 7.45. ................................ 118 xv  Figure 3.8 Hypothetical 2D structures of Dz7-38-32t, Dz7-38-39t, Dz7-38-90t, and Dz7-38-83t in complex with the 19 nt all-RNA substrate, respectively. ............................... 120 Figure 3.9 Effect of substrate-recognition arm length on trans-cleavage rate at 37oC. 122 Figure 3.10 The function of dAim in the modified 3’ substrate-recognition arm. ........... 124 Figure 3.11 Activity of Dz7-38-32t on the chimeric DNA/RNA substrate. ..................... 127 Figure 3.12 PAGE gel (20%) pictures for substrates containing different RNA bases at the cleavage site. ......................................................................................................... 129 Figure 3.13 Analysis of the cleavage product. ............................................................. 132 Figure 3.14 The pH dependence study of Dz7-38-32t. ................................................ 135 Figure 3.15 Effects of divalent metal ions on the trans-cleavage activity of Dz7-38-32t. ..................................................................................................................................... 138 Figure 3.16 Concentration determination of Dz7-38-32t using autoradiographic calibration curve. .......................................................................................................... 151 Figure 4.1 The structures of CPP and Dz7-38-32. ....................................................... 157 Figure 4.2 Preparation of fluorescent Dz7-38-32t. ....................................................... 159 Figure 4.3 Bright field images of cells treated with 1.35 µM of Dz7-38-32t-m at increasing time points up to 45 h of incubation time. ................................................... 161 Figure 4.4 Confocal microscopy images of HeLa cells treated with 0.9 µM of rhodamine-labeled Dz7-38-32t of various constructs.. ................................................................... 164 Figure 5.1 Discretely modified oligonucleotides containing modified nucleoside(s). .... 175 Figure 5.2 Impact of 8-histaminyl-dA on duplex thermal stability. ................................ 176 Figure 5.3 Base-pairing properties of dAim. .................................................................. 179 xvi  Figure 5.4 Standing-start single-nucleotide insertion assays opposite dAim through 180 min of reaction time. ..................................................................................................... 180 Figure 5.5 Single-nucleotide insertion assay opposite dUga in template ON 2. ............ 181 Figure 5.6 Single-nucleotide insertion assay opposite dCaa in template ON 2. ............ 183 Figure 5.7 Strand synthesis continuing beyond modified base pairs using Vent (exo-) DNA polymerase. ......................................................................................................... 185 Figure 5.8 Assessment of the overall fidelity throughout a complete replication/selection cycle. ............................................................................................................................ 187 Figure 5.9 Self-cleavage and purification of Dz7-38-32 and “dummy” control. ............ 188 Figure 5.10 PCR assay for unmodified template DNA contaminants. .......................... 189 Figure 5.11 cDNA sequence alignment used to evaluate modified DNA replication fidelity. .......................................................................................................................... 191 Figure A1 Bright field images of cells treated with 0.9 µM of Dz7-38-32t-m at increasing time points up to 45 h of incubation time…………………………………..…………….. 225 Figure A2 Stability of Dz7-38-32t-m and Dz7-38-32t-u in Opti-MEM at 37°C………..  226 Figure A3 Denaturing purification PAGE (10%, 7M urea) for ON 1, ON 2 and ON 3...227 Figure A4 Thermal melting study at pH 6. ………………………………………………..228 Figure A5 Thermal melting study at pH 8.5……………………………………………….229  xvii  List of Abbreviations  A    adenosine  Arg    L-arginine  ATP    adenosine triphosphate   C    cytidine   cDNA   complementary DNA  cis   from the same molecule  CPP    cell penetrating peptide  dA    2′-deoxyadenosine  dAim   8-histaminyl-2′-deoxyadenosine dATP    2′-deoxyadenosine triphosphate  dAimTP  8-histaminyl-2′-deoxyadenosine triphosphate dC    2′-deoxycytidine  dCaa   5-aminoallyl-2′-deoxycytidine dCTP    2′-deoxycytidine triphosphate  dCaaTP  5-aminoallyl-2′-deoxycytidine triphosphate DEPC   diethylpyrocarbonate  DPBS   Dulbecco’s phosphate-buffered saline DMEM  Dulbecco’s modified eagle’s medium dG    2′-deoxyguanosine  dGTP    2′-deoxyguanosine triphosphate  DNA    deoxyribonucleic acid  xviii  DNAzyme   DNA enzyme  dNTPs   2′-deoxynucleoside triphosphates  dT    2′-deoxythymidine  DTT    (2S), (3S)-dithiothreitol  dTTP    thymidine triphosphate  dU    2′-deoxyuridine  dUga   5-guanidinoallyl-2′-deoxyuridine  dUTP    2′-deoxyuridine triphosphate   dUgaTP  5-guanidinoallyl-2′-deoxyuridine  triphosphate  E.coli   Escherichia coli EDTA   ethylenediaminetetraacetic acid   EGR-1  early growth response-1  FBS   fetal bovine serum HDV   hepatitis delta virus His    L-histidine  HIV   human immunodeficiency virus kcat    multiple-turnover catalytic rate constant  KM   Michaelis-Menten constant of a substrate  kobs   observed rate constant  LB   lysogeny broth  LNA    locked nucleic acid  LTR   long terminal repeat xix  Lys    L-lysine  mRNA   messenger ribonucleic acid  N20    random 20 nucleobase DNA  N40    random 40 nucleobase DNA nt    nucleotides  PAGE   polyacrylamide gel electrophoresis  PCR    polymerase chain reaction  pKa    negative log of acid dissociation constant   PNK    polynucleotide kinase  rA   riboadenosine (as opposed to the 2′-deoxyadenosine) rC    ribocytidine (as opposed to the 2′-deoxycytidine) rG   riboguanosine (as opposed to the 2′-deoxyguanosine) rU   ribouridine (as opposed to the 2′-deoxyuridine) RNA    ribonucleic acid  RNase A   ribonuclease A  SELEX   systematic evolution of ligands by exponential enrichment  Sequenase  engineered Family A DNA polymerase lacking 3'-5' exonuclease activity  siRNA   small interfering ribonucleic acid  T    thymidine  Taq    Family A DNA polymerase from Thermus aquaticus  TEN    tris, EDTA, NaCl buffer  Tm   thermal melting temperature  xx  Trans    from a different molecule  Tris    tris(hydroxymethyl) aminomethane  U    uridine  UV    ultraviolet  VEGFR  receptor for vascular endothelial growth factor Vent  Family B DNA polymerase derived from Thermococcus  litoralis  Vis visible light  λ exonuclease  5'-3' exodeoxyribonuclease derived from Escherichia coli    xxi  Acknowledgements  I would first like to thank my supervisor Dr. Perrin for his support throughout my degree. His guidance, training and mentorship were critical in developing my knowledge, skills and persistence as a researcher, and making my academic experience as graduate student a success.  I would also like to thank all past and present members of the Perrin laboratory for providing the foundation for my research projects and for their support and company throughout my time there. In particular, I would like to thank Dr. Curtis H Lam for his invaluable training in my early years of graduate study. I would also like to thank Emily Miller and Dr. Erkai Liu for their insightful advice and for all the exciting and frustrating moments we shared with each other.  This thesis was proofread by Dr. David Perrin, Dr. David Chen, Ms. Emily Miller, Mr. Abid Hasan, Mr. Omar Sadek, and Mr. Edward Mount. Thank you for all your suggestions, corrections and feedback. I am grateful to Nicole Ng and Edward Mount, the two undergraduate students who worked with me, for their help of carrying out some fundamental experiments, and also for the freshness they always brought me.  I must also extend my thanks to all the staff of the offices, shops and services of the Chemistry Department. Whenever I needed something ordered, fixed or analyzed, I could always find immediate and kind help. I would particularly like to thank Jessie Chen xxii  and Dr. Elena Polishchuk at the BioServices Laboratory for their work in helping me prepare the biological materials I required.  I am deeply grateful to my amazing friends, for the company, support, encouragement, and entertainment they have given me throughout my degree.  I am greatly appreciated that my sister and brother have taken so much of my share of responsibility to take care of our parents.   Most importantly, I profoundly thank my parents for their endless support and for their efforts to motivate me to explore my interests and finish such a long journey of pursuing education and science. The values and work ethic that they strongly hold have been crucial in developing my personality, as well as my beliefs in life and science.    1  Chapter 1: General Introduction to the Chemistry and Catalysis of RNA Cleavage Reaction  1.1 Nucleic Acids and In Vitro Selection 1.1.1 Nucleic acids and naturally-occurring ribozymes Nucleic acids, including ribonucleic acid (RNA) and deoxyribonucleic acid (DNA), are one of the four major biological macromolecules essential for all forms of life. “DNA makes RNA makes protein” is the shorthand version of Francis Crick’s ‘‘central dogma’’ of molecular biology1,2, which more specifically states the flowing of genetic information from DNA to RNA to protein in a cell.  Whereas DNA is considered the primary transcript of a cell responsible for genetic information storage, RNA is the intermediate used to translate the genetic code from DNA into proteins. For a long time, RNA was thought merely to fill the role of an intermediate until the discovery of natural catalytic RNA molecules (ribozymes) in early 1980’s. The fact that naturally-occurring ribozymes facilitate a number of processes including RNA self-cleavage3-7, RNA splicing8-10, RNA cleavage11, and peptide bond formation12,13 revised the traditional conception of RNA and led to the hypothesis of “the RNA world”14. This hypothesis postulates that RNA molecules were the primitive polymers that had both the hereditary and catalytic functions, and thus were capable of propagating life at its earliest stages. 2  1.1.2 In vitro selection (SELEX) and artificial ribozymes  Starting from 1990s, the discovery of natural ribozymes spurred the identification of artificial ribozymes using in vitro selection; one of these techniques is commonly referred to as SELEX (Systematic Evolution of Ligands by EXponential enrichment) (Figure 1.1).15-18 Briefly, starting from a pool of RNA sequences containing up to 1014 random sequences19, molecules that conform to the desired phenotype are separated from those that do not in a process known as “selection”. The isolated desired sequences are then scaled up to the size of the original pool by use of the polymerase chain reaction (PCR) that provides exponential amplification. The direct result of a single round of in vitro selection is a pool of molecules that is enriched (relative to the originally random pool) in sequences with the desired phenotype. The overall selection process is then repeated until the desired phenotype is retained at the same ratio in the selection pool and no further improvement is brought about by more selection cycles. More than one round of selection is required because inactive sequences have a small but nonzero probability of surviving through any particular round of selection; therefore, many selection rounds are needed to ensure that the surviving sequences are reproducibly competent at the desired catalytic activity.20 As selection proceeds, reaction parameters such as reaction time, pH, temperature, salt concentration, and sometimes substrate concentration may be altered in order to favor the development of molecules that are adapted to the adjusted conditions.    3   Figure 1.1 Schematic representation of in vitro selection.  (1) The sequence pool preparation via a template-directed primer incorporation step. (2) An isolation step to select the desired active sequences. (3) An amplification step to exponentially scale up the active sequences.   Since the advent of in vitro selection, in addition to ribozymes that are found in nature, artificial ribozymes catalyzing a wide range of novel reactions have been isolated. Based on an RNA ligation reaction catalyzed by the first artificial ribozyme isolated by in vitro selection in Szostak group21, these reactions catalyzed by artificial ribozymes also include (1) a Michael addition between a fumaramide-derivatized RNA pool (the Michael-acceptor) and a biotinylated cysteine (the Michael-donor)22, (2) an alcohol oxidation reaction to convert a benzyl alcohol that was covalently attached to the 5’-end of the RNA pool to a benzyl aldehyde and the reverse reduction reaction to reduce a 2. Isolation 3. Amplification 1. Incorporation 4  corresponding benzyl aldehyde to an alcohol23,24, (3) a Diels-Alder reaction of a biotinylated maleimide with an RNA-tethered anthracene25, (4) a decarboxylative Claisen condensation reaction between a pool of malonyl-PEG-RNA conjugates and the biotinylated acyl-CoA26, and (5) a urea synthesis reaction between an NHS-carbamate activated N-terminus of a small noncharged hydrophobic peptide and a primary amine that was tethered to the 5’-end of the RNA pool27, among others.  1.1.3 Artificial DNA enzymes/ deoxyribozymes / DNAzymes Analogous catalysis by artificial DNA enzymes (also referred to as deoxyribozymes or  DNAzymes) was first demonstrated experimentally in 1994 by Breaker and Joyce under the assumption that DNA can have catalytic activity as well, considering that it contains most of the same functional groups as RNA, except for the lack for 2’-hydroxyl group. Unsurprisingly, they successfully applied in vitro selection methodology to a random pool of DNA molecules to identify the first Pb2+-dependent DNAzyme capable of catalyzing the cleavage of an RNA phosphodiester linkage.28 Since this first report, hundreds of DNAzymes have been isolated using the in vitro selection methodology to catalyze as broad range of reactions as their ribozyme counterparts do. In addition to RNA cleavage29-31 and ligation32 reactions, which are well within the catalytic capability of DNAzymes, these DNAzyme-catalyzed reactions also include (1) DNA cleavage reactions through an oxidation mechanism with the facilitation of Cu2+ as a cofactor33,34 and an alternative hydrolysis mechanism with Zn2+ 35 or Zn2+ and Mn2+ as cofactors 36, (2) DNA ligation reactions by the condensation mechanism of the 5’-hydroxyl of one oligonucleotide and the 3’-phosphorimidazolide of another oligonucleotide37 or by a 5  mechanism that is analogous to the last step of the joining reaction of T4 DNA ligase38 39, (3) porphyrin metalation reactions by the insertion of Cu2+ and Zn2+ 40,41, and (4) phosphorylation of tyrosine residues of peptides using an ATP bound to a discrete ATP-aptamer42 or using a phosphoryl group of a 5’-triphosphorylated RNA oligonucleotide43,44, among others45-47. However, no naturally-occurring catalytic DNAzyme has been reported so far, presumably because (barring some exceptions for certain viral genomes and replication intermediates) nearly all the DNA in biological organisms exits as a complete duplex, which precludes it from adopting complex secondary and tertiary structures required for catalytic function.23,24,28 1.2 Comparison of DNAzymes with Ribozymes and Proteins: Motivation for Developing DNAzymes Among the broad spectrum of catalytic activities catalyzed by DNAzymes, the RNA cleavage reaction is still the most commonly studied reaction. This is in part due to its long laboratory history, and more importantly, because RNA represents an important target for ultimately regulating protein production. Consequently, RNA-cleaving DNAzymes hold more promise for practical applications compared to DNAzymes of other catalytic activities, in particular for use as therapeutic agents for in vivo RNA manipulation. The focus of this thesis is the development of divalent metal ion independent RNA-cleaving DNAzymes that is undertaken largely due to this promising application potential.  1.2.1 Comparison of DNAzymes to ribozymes When comparing DNA and RNA in regards to their catalytic capabilities, the primary 6  concern is whether the lack of the 2’-hydroxyl group in DNA causes it to be catalytically inferior. Cech25,48 and Rich49 pointed out that the 2’-hydroxyl group as a hydrogen bond donor and acceptor gives RNA more versatility than DNA to form complex secondary and tertiary structure, which might make RNA more competent to perform catalysis. In reality, the catalytic rate enhancements displayed by DNAzymes fall into the same range as those associated to their respective ribozyme analogues, as summarized in two separate reviews on ribozymes50 and DNAzymes51 by Silverman. When a direct comparison was made based on specific experimental data, DNAzymes were demonstrated to catalyze carbon-carbon formation in Diels-Alder reactions with equal catalytic efficiency as that observed for ribozymes.52 Moreover, Silverman et al.53 isolated RNA-cleaving DNAzymes functioning through alternative hydrolysis mechanism instead of the readily available transesterification mechanism through in vitro selection including a stringent selection pressure for hydrolysis. The isolation of such DNAzymes suggests that DNAzymes also have the catalytic ability to direct a chemical reaction down an otherwise disfavored pathway even when a more favored mechanism is readily available. Collectively, evidence appears to suggest that DNAzymes are not intrinsically inferior to ribozymes in catalytic capability.  1.2.2 Comparison of DNAzymes to protein enzymes Natural selection and evolution has chosen protein as the dominant catalyst of biological world, so it is not surprising that most proteins have significantly higher rate enhancements compared to DNAzymes and ribozymes. The catalytic inferiority of nucleic acids over proteins is attributed to both the lack of functional groups analogous 7  to those on protein side chains which are most frequently identified as catalytic residues, and the high charge and flexibility of their backbones, which limits precise positioning.54 Breaker and coworkers carried out a detailed investigation and conceptual analysis about the rate limits of ribozymes and DNAzymes based on their RNA-cleavage activities. They concluded that if nucleic acid enzymes could employ multiple catalytic strategies simultaneously as their protein counterparts do, enormous rate enhancements could likely be achieved.55,56 Bevilacqua et al. also proposed that the HDV ribozyme could attain a rate enhancement as high as RNase A if optimal general acid-base catalysis was used.57 In addition, DNAzymes as well as all other nucleic acid enzymes possess inherent high substrate specificity relative to protein enzymes, because they process nucleic acid substrate recognition through Watson-Crick base pairing.58 Thus, an overall catalytic performance comparable to that of protein enzymes is worth expecting for DNAzymes as well as other nucleic acid enzymes.  1.2.3 Advantages of synthetic DNAzymes over synthetic ribozymes  DNA lacks the 2’-hydroxyl group found in RNA, and consequently is almost completely inert to cleavage through hydrolysis or transesterification by small molecule catalysts, with a few notable exceptions.59-63 Consequently, the phosphodiester linkage making up the backbone of DNA has an estimated half-life of 200 billion years at pH 7 and 25°C64, which is approximately 100,000 times of that of RNA65,66. In other words, DNAzymes would be more likely to survive extreme conditions, making them more suitable for a broader range of applications. Regarding the in vitro selection process, DNAzymes also require fewer selection steps than ribozymes, because ribozymes almost always require 8  reverse transcription of RNA into cDNA67. This remains the case although several techniques including isothermal amplification strategies68-70 and automated RNA selection71 have been developed to reduce the manipulations of ribozyme selections. Additionally, DNA phosphoramidite reagents are less costly than their RNA counterparts, making the large-scale synthesis of DNA more cost-efficient. Overall, DNA solid phase synthesis is of higher-yield and less tedious.  1.3 RNA Cleavage Reaction: Knowledge about RNA Targets 1.3.1 RNA cleavage by transesterification reaction Owing to the presence and participation of the 2’-hydroxyl group, RNA undergoes a spontaneous self-cleavage reaction through an intramolecular transesterification reaction in the absence of any added catalyst. This transesterification reaction proceeds through a concerted SN2-like mechanism, wherein the 2’ ribose oxygen acts as the nucleophile to attack the adjacent phosphodiester backbone, generating a phosphorane-like transition state.72-74 Cleavage products have a 2’, 3’-cyclic phosphate (5’-cleavage product), and a 5’-OH (3’-cleavage product), respectively (Figure 1.2). Under near physiological conditions (pH 7.0, 250 mM K+, 5 mM Mg2+, 23°C), the spontaneous cleavage of RNA typically occurs with a rate constant of about 10-8 min-1.65  Different classes of enzymes employ distinct catalytic strategies to accelerate this otherwise slow transesterification reaction that leads to RNA cleavage. These strategies were systematically summarized by Breaker and coworkers56 and are shown in Figure 1.2: in-line nucleophilic attack of the adjacent phosphodiester backbone (α catalysis)75; 9  electrostatic stabilization of the negative charge on the non-bridging phosphate oxygen (β catalysis)76; deprotonation of the 2’-hydroxyl group for nucleophile activation (γ catalysis)65; and neutralization of the negative charge on the 5’-oxygen of the downstream (3’) cleavage product (δ catalysis)76. At neutral pH, the total rate enhancement that is theoretically available for enzymes can be at least as high as 1019, given an exploitation of all catalytic strategies α through δ.77  Figure 1.2 Mechanism of RNA cleavage reaction through internal transesterification reaction.  The four catalytic strategies that accelerate the reaction are depicted as: α, in-line attack of the phosphodiester backbone; β, electrostatic stabilization of the negative charge on the non-bridging phosphate oxygen; γ, deprotonation of the 2’-hydroxyl group for nucleophile activation; and δ, neutralization of the negative charge on the 5’-oxygen of the downstream (3’) cleavage product. Reproduced from Reference 55 and 56.    OOOP OOOB5' RNAHO BO OHOOOPOB5' RNAOO BO OHOOOOB5' RNAHOO BO OHPOO3' RNA3' RNA 3' RNAαβγδ10  1.3.2 Transesterification reaction catalyzed by small self-cleaving ribozymes  Natural, small, self-cleaving ribozymes catalyze this transesterification reaction either by making use of their own nucleobases with perturbed pKas (as in the case of hammerhead78-80, hairpin81-83, and VS ribozymes84), or by a combination of nucleobases and associated divalent metal cations (as in the case of the HDV ribozyme85,86). These moieties participate in a general acid/base reaction to achieve rate enhancements of approximately 109-fold under the most frequently used reaction conditions for studying these natural, small, self-cleaving ribozymes (10 mM Mg2+, pH 7.5 and 25°C).87-89 Although divalent metal cations (M2+s) can play either direct roles by participating in catalysis, or indirect roles by involving in folding into catalytically competent structures in the catalytic processes of different classes of natural ribozymes, almost all ribozymes require millimolar Mg2+ to achieve a moderate or an optimal catalytic activity.90 According to Uhlenbeck’s investigation, the hammerhead ribozyme can attain a very diminished catalytic rate constant of 2.4×10-4 min-1 in the absence of any divalent metal ions, which is approximately 1300-fold slower than the same rate constant in the presence of 10 mM Mg2+.91 Similarly, the self-cleavage rate of HDV ribozyme decreased to 1×10-3 min-1 in the absence of Mg2+ at pH 7.92,93 Regarding to the catalytic strategies depicted in Figure 1.2, small self-cleaving ribozymes are proposed to have implemented α and γ catalysis, but not make use of either the β or δ strategies.55,56,94 In other words, ribozymes, as well as other nucleic acid enzymes, are poor at negative charge neutralization, despite the presence of M2+ cofactors.95 Ribozymes can be used as sequence-specific RNA cleavage agents in vitro was first 11  demonstrated by Cech et al. using the Tetrahymena ribozyme as an RNA restriction endonuclease to cut a variety of tetranucleotide sequences.96,97 Evoked by this first demonstration, the therapeutic value of ribozymes as mRNA disruption reagents was explored and successfully demonstrated mostly by using the catalytic motifs of the hammerhead and hairpin ribozymes. Mostly, at the cellular level and in a few cases in vivo, the hammerhead and hairpin ribozymes were reported to inhibit target mRNAs derived from HIV-1, HBV, HCV and so on.98,99 Nevertheless, these preliminary inhibition activities of ribozymes against viral mRNAs were undermined by the low stability of RNA both in vitro and in vivo. Attempts to improve ribozyme stability through chemical modification of the 2’-OH group led to limited success, because such modifications always resulted in undesired effects that compromised the activity.100-102  Hence, at the time, DNAzymes became more attractive as therapeutic agents due to their noted stability compared to RNA. Kurrreck et al. did a comparative study of DNAzymes and ribozymes against the same full-length mRNA of the Vanilloid receptor subtype I, in which they demonstrated that the most efficient DNAzyme had a ~15-fold higher reaction rate and a ~100-fold higher catalytic efficiency than the best ribozyme, and concluded that DNAzymes were an inexpensive, very stable and active alternative to ribozymes for specific cleavage of RNA.103 Therefore, enthusiasm towards mRNA disruption using ribozymes has been shifted to DNAzymes, which displayed comparable or higher catalytic activities under similar conditions and did not suffer from the same stability issues. 12  1.3.3 In vitro selection methodology of RNA-cleaving DNAzymes The in vitro selection methodology for RNA-cleaving DNAzymes was pioneered by Breaker and Joyce28,104 and used widely by different groups with appropriate adaptions. Generally, a biotin moiety attached to the 5’-end of the sequences that make up the pool serves for immobilization through binding to a streptavidin-coated solid support, normally magnetic beads or a column. Active sequences become detached from the solid support as a consequence of the self-cleavage reaction, allowing the separation of active sequences (partitioned into the cleavage buffer) from inactive ones (retained on solid support). Once the active sequences have been separated, they are amplified by PCR, a process that is facilitated by the inclusion of constant primer-binding regions that flank the random region. Then the amplified active population serves as the template for the next round of selection (Figure 1.3).    13    Figure 1.3 Scheme for in vitro selection of RNA-cleaving DNAzymes.  Biotin-streptavidin interaction serves for the separation methodology for the active and inactive sequences.   1.3.4 Development of RNA-cleaving DNAzymes for therapeutic applications: lessons learned In analogy to ribozymes, Mg2+ in most cases is critical to the structure and/or catalytic activity of DNAzymes.94,105,106 DNAzymes normally require ~10 mM Mg2+ for optimal activity.107-109 In a few cases DNAzymes can make use of ~0.1 mM Ca2+/Zn2+ to achieve optimal activity as well. Notwithstanding the low cellular concentration of free Mg2+ (~0.5 mM)110,111, it is still the most abundant one relative to Ca2+ (<1 µM112) and Zn2+ (<10 nM113). The selection of RNA-cleaving DNAzymes for therapeutic applications included the presence of Mg2+ as the cofactor.  Primer extensionBind to streptavidinB0.1 N NaOHBIncubation incleavage bufferxxxxActive population collectionPCRPrimer-binding regionxxxxRandomized library region5' multi-T regionRNA target regionLibrary regionCleavage region5'-BiotinBBBBBStreptavidin coated support14  One year after the selection of the first ever DNAzyme with Pb2+-dependent RNA cleavage activity28, Breaker and Joyce set out to select a Mg2+-dependent RNA-cleaving DNAzyme, due to the high potential for biomedical applications and relevance to the cellular environment104. The outcome of this selection led to the discovery of a DNAzyme operating in the manner of multiple-turnover under intracellular-compatible conditions with a rate enhancement of ~105-fold, which raised the possibility that DNAzymes might be made to operate in vivo for the first time. Although this DNAzyme, which was selected with a chimeric DNA/RNA substrate containing only one embedded ribonucleoside at the cleavage site, was unable to cleave an all-RNA substrate, its properties suggested that selecting a DNAzyme with general-purpose all-RNA cleavage activity under physiological conditions might be attainable.  With this inspiration, Joyce and Santoro obtained DNAzymes 10-23 (Dz10-23) and 8-17 (Dz8-17) after a total of 52 rounds of intensive in vitro selection in the pursuit of physiologically compatible and catalytically efficient DNAzymes that cleave all-RNA targets. Both Dz10-23 and Dz8-17 are capable of multiple-turnover catalysis with a catalytic rate constant of ~0.1 min-1 in the presence of 2 mM Mg2+, at pH 7.5, 37°C (Figure 1.4).29,114,115 Notably, they can be engineered to cleave virtually any all-RNA target that contains a purine-pyrimidine junction.116  15   Figure 1.4 Composition of the Dz10-23 and Dz8-17 catalytic motifs. The DNAzyme (bottom strand in black) binds the RNA substrate (top strand in red) through Watson-Crick pairing. Arrows indicate cleavage sites. R = A or G; Y = U or C. Reproduced from Reference 29.   Consequently, Dz10-23 has been intensively studied for reduction of a target mRNA in cells and in vivo owing to its superior catalytic efficiency compared to Dz8-17.117-123 It is encouraging that Dz10-23 targeting VEGFR121, c-Jun119, and EGR-1122 significantly reduced tumor sizes in relevant mouse models. More strikingly, Khachigian and colleagues reported the first phase I clinical trial with Dz13, an engineered version of Dz10-23 targeting c-Jun mRNA, in which they established the safety and tolerability of this DNAzyme on human beings in 2013.124  The mRNA inhibition activities displayed by Dz10-23 in cells and in vivo demonstrate that RNA-cleaving DNAzymes are viable therapeutic agents, yet raise the question whether the working mechanism is catalytic cleavage or simply mRNA silencing through more standard mechanisms such as RNase H or simple antisense effects. In some 3'--3'5'--5'R YRGGCTAGCTAAGCAACCatalytic coreSubstrate-recognition armSubstrate-recognition armCleavage siteRNA substrateDNAzymeDz10-23A3'--3'5'--5'A GTAGC AGCCG GCC GABDz8-1716  cases it is certain that the observed protein reductions were caused by mRNA disruptions through the cleavage mechanism of DNAzymes. This conclusion was supported by the fact that antisense analogues of DNAzymes displayed significant smaller effects, as exemplified by utilizing Dz10-23 for targeting HIV-1 leader region RNA125 and insulin-like growth factor I mRNA123. However, in some cases DNAzymes and the antisense analogues induced similar or slightly different levels of protein reduction, which is suggestive that the DNAzyme attained substrate inhibition through an antisense effect or both antisense and cleavage effects. This trend was observed when Dz10-23 was used to target RNAs of hepatitis B126, ornithine decarboxylase127, and VEGF receptor128.  To address the possible mechanism associated with the observed intracellular activity of Dz10-23, Cieslak et al.109 carried out a study on folding transition of Dz10-23 as a function of Mg2+ concentrations, in which they found out that the folding of the DNAzyme induced by binding of Mg2+ may occur in several distinct stages. At 0.5 mM Mg2+, which simulates the physiological conditions, the DNAzyme was only partially folded, did not bind to its substrate, and showed only residual catalytic activity. It was hypothesized at this Mg2+ concentration the DNAzyme may be catalytically inactive in the transfected cells and instead behave like an antisense oligonucleotide, as certain studies concluded121,126,127. Conversely, at 15 mM Mg2+, Dz10-23 adopted a secondary structure consisting of a compact catalytic domain with high substrate binding affinity and efficient catalytic activity. Based on the relationship between kcat and Mg2+ concentration, the estimated kcat value of Dz10-23 under low-Mg2+ cellular conditions 17  would fall into a low to moderate range of 0.001-0.02 min-1. Despite of the claims of generality against a broad spectrum of substrate mRNA sequences, Cieslak’s study showed significant variation in catalytic efficiency (kcat/KM) when Dz10-23 was used to target mRNAs of different sequences. In Cieslak’s study, Dz10-23 was used to cleave the β-integrin subunit mRNA, however it attained a maximum catalytic efficiency of merely 1.09×106 M-1min-1 in 50 mM Tris (pH 8.0) buffer containing 15 mM Mg2+ at 37°C. In contrast, a catalytic efficiency of 1.4×108 M-1min-1 was attained by Dz10-23 against the HIV-1 gag/pol mRNA under simulated physiological conditions containing only 2 mM Mg2+.29 The fact that Dz10-23 relies on 15 mM Mg2+ to achieve its optimal enzymatic activity raises two questions with respect to the full employment of its catalytic potential, one of which is the low concentration of Mg2+ (~0.5 mM)111 under cellular conditions as mentioned at the beginning of 1.3.4, while the other one is the dichotomy between the optimal DNAzyme activity (favored by high concentrations of Mg2+) and the high substrate accessibility (favored by low concentrations of Mg2+). Long mRNA strands are prone to adopting compact folds at high concentrations of Mg2+, which greatly decreases their accessibility for catalysts.129  Additionally, in comparison to other approaches for sequence-specific knockdown of mRNA, the catalytic activity of DNAzymes endows them with the capacity for independent substrate cleavage without the reliance on any cellular machinery.130-132 If DNAzymes can be shown to function catalytically with intracellular multiple-turnover activity, the very high stoichiometry requirements of DNAzyme administration would be 18  alleviated, which would be likely to lead to increased in vivo safety and tolerability.133,134 Therefore, it is conceivable that DNAzymes can be evolved and then engineered so that the cleavage mechanism is preferred over the antisense mechanism. This would foster the evolution of a more efficient and potent DNAzyme under physiological conditions.103 Thus, a M2+-independent RNA-cleaving DNAzyme with high catalytic efficiency and sequence specificity is likely to provide an alternative solution to the paradoxes described above by providing a more efficient route to in vivo gene regulation through a catalytic RNA cleavage reaction. The catalytic activity of such a DNAzyme would not be compromised by low concentration of Mg2+ that is found in cells. 1.3.5 M2+-independent RNA-cleaving DNAzymes Attempts made by Sen and Geyer30, Faulhammer and Famulok135, as well as Benner136 towards the development of M2+-independent DNAzymes led to the discovery of three such DNAzymes, all of which attained self-cleavage rate constants of 10-4 to 10-3 min-1 in the absence of any M2+. Their catalytic performances are comparable to that of the hammerhead ribozyme in the absence of any M2+.91,137 Only the species reported by Benner showed a very slow turnover rate (~2 in 100 h). These early findings suggest the difficulty for pursuing M2+-independent DNAzymes that attain catalytic rate constants of greater than 10-3 min-1 and efficient multiple-turnover activity. In contrast to these previously reported M2+-independent DNAzymes, a Na+-dependent DNAzyme, denoted NaA43, reported by Yi Lu et al. in 2015 attained an observed cis-cleavage rate constant of 0.11 min-1 for cleaving a single embedded adenosine ribonucleotide (rA) in the presence of 400 mM Na+.138 This represents the so far best catalytic rate attained by 19  unmodified DNAzymes under M2+-free and relatively low concentrations of Na+. Notably, NaA43 was converted into a trans-acting true enzyme with catalytic turnover. Since NaA43 was selected for Na+ sensing under cellular conditions, the authors did not demonstrate its propensity to cleave an all-RNA substrate. 1.4 Transesterification RNA Cleavage Reaction Catalyzed by RNase A In contrast to both RNA-cleaving ribozymes and DNAzymes, the protein enzyme RNase A uses essentially only three critical amino acids at the active site simultaneously exploit all four catalytic strategies shown in Figure  1.2, with an extraordinary rate constant about 80,000 min-1 in optimal conditions, corresponding to an rate enhancement ~1013.139,140 Specifically, the imidazoles of His12 and His119 afford general base and general acid catalysis, respectively, while the cationic ammonium ion of Lys41 electrostatically stabilizes a pentacoordinate anionic phosphate transition state (Figure 1.5).140-142 It is particularly noteworthy that RNase A catalyzes RNA cleavage with high efficiency without the involvement of any M2+ cofactor.141,143 Additionally, RNase A is one of the best characterized enzymes with the precise catalytic mechanism assigned to specific amino acid residues.144,145 This level of characterization is quite different from the current status of self-cleaving ribozymes, the catalytic mechanisms of which are still ambiguous and of some dispute146,147. Therefore, the active site mechanism of RNase A represents a venerable model and benchmark for designing and evaluating M2+-independent artificial ribonucleases.     20    Figure 1.5 RNase A active site structure and catalytic mechanism.  Reproduced from Reference 139 and 140.  1.5 Towards the Development of Sequence Specific and M2+-independent RNase A Mimics   Efforts in the development of artificially M2+-independent RNase A mimics which integrate properties of high sequence specificity (characteristic of ribozymes, DNAzymes, as well as other antisense oligonucleotide reagents) and high catalytic efficiency (characteristic of RNase A), have been directed in two alternative paths: rational design and in vitro selection.  1.5.1 RNase A mimics based on rational design  Numerous sequence-specific RNase A mimics have been rationally designed. In general, a non-specific RNA-cleaving moiety is conjugated to a DNA oligonucleotide that confers sequence specificity (Figure 1.6). The RNA cleaving moieties contain OOOPO N NHB5' RNAHis12BOH O3' RNALys41NH2NNHHHis119HO OO21  properly  positioned functionalities derived from those found at the active site of RNase A, most commonly imidazoles and amines. Analysis of successful examples reveals that these catalysts suffer from very low catalytic rate constants (≤ 10-3 min-1) and turnover incapability.63,148-154 For example, in the presence of 10-fold excess of conjugate depicted in Figure 1.6C, catalytic rate constants of 0.042-0.056 h-1 were observed depending on different substrate sequences with no detectable turnover.153 In fact, their catalytic rate constants are actually in the same order of magnitude with those described for M2+-independent DNAzymes mentioned beforehand in 1.3.5, which currently excludes them from further development towards therapeutic applications. These studies  underscore the difficulty in rational design to produce sufficiently efficient RNA-cleaving catalysts yet highlight the significant interest in the alternative way of in vitro selection methodology, which has generated more promising results towards the development of more efficient and M2+-independent RNA cleavers.   22   Figure 1.6 Sequence specific RNase A mimics generated by rational design. (A) Oligonucleotide conjugate containing an ethylenediamine attached at the 3’-end (reproduced from Reference 154). (B) Oligonucleotide conjugate containing a diamidazole construct attached at the 3'-end (reproduced from Reference 151). (C) Oligonucleotide conjugate containing trisaminobenzimidazole and cationic amine attached at the 5'-end (reproduced from Reference 153).   1.5.2 In vitro selection of modified DNAzymes with functionalities derived from RNase A  In contrast to the methodology of rational design, in in vitro selections of RNA cleaving O NHONH NH2A5'-5'-NHNO NHONHONHNNHNBHNNNHNNHNHNNHNHNH NH SS OPO OO -3'3C23  catalysts, no assumptions are made regarding the position of RNA-cleaving moieties appended on nucleosides in the oligonucleotide sequences that are produced. Instead, an unbiased (or minimally biased) exploration of “sequence space” is carried through the whole in vitro selection process, which eventually leads to the enrichment of sequences that catalyze some reaction be it RNA cleavage or other, by making use of the readily available appendages, for example, imidazoles and cationic amines.    Soon after the development of artificial nucleic acid enzymes through in vitro selection, many modified nucleotide analogues were synthesized and used in in vitro selections, first for RNA155-160 and quickly extending to DNA161-168. The goal was to supplement the meager functionalities of nucleic acids, thereby enriching their structural and functional diversity. Functionalized ribozymes bearing a single appended pyrimidine functional group for Diels-Alder cyclization catalysis169 and a single imidazole functional group for amide synthesis catalysis170 were first selected by Eaton and coworkers in 1997, with unprecedented catalytic performances compared to the unmodified counterparts. Similarly, Barbas et al. synthesized a series of deoxyuridine derivatives that were tolerated by some thermostable DNA polymerases for in vitro selection conditions.171 By using one of these derivatives, a C5-imidzole-functionalized dUTP, in place of natural dTTP in an in vitro selection, Barbas and Santoro selected the first ever modified RNA-cleaving DNAzyme, denoted Dz16.2-11 (Figure 1.7).172 Trans-acting Dz16.2-11 was subsequently synthesized by coupling the phosphoramidite of the modified deoxyuridine, and it operated with multiple-turnover with an absolute dependency on micromolar Zn2+ as a cofactor for coordinating the appended imidazoles. In the 24  presence of 30 µM Zn2+ at pH 7.5, Dz16.2-11 attained a maximum catalytic rate constant of 3.1 min-1. Compared with a DNAzyme that used L-histidine as cofactor31, Dz16.2-11 represented a step closer to RNase A mimics with respect to the immediate availability of intramolecular imidazole group, but was still limited by the Zn2+-dependence.     Figure 1.7 Dz16.2-11 containing C5-imidazole-functionalized deoxyuridine. (A) Minimal composition of Dz16.2-11 containing catalytically essential functionalized deoxyuridines (U). RNA substrate sequence is in red. Arrow indicates the cleavage site. (B) Chemical structure of the C5-imidazole-functionalized deoxyuridine. Reproduced from Reference 172.   A year later, the first DNAzyme containing two modified nucleosides was identified. Denoted Dz925-11, it was obtained by simultaneous use of two different modified nucleotides in an in vitro selection designed to pursue totally M2+-independent RNase A mimics.173 The two modified nucleotides were 8-histaminyl-deoxyadenosine (dAimTP), a dATP analogue for mimicking the activity of His119 and His12 of RNase A, and 5-5'--5'3'--3'G U AG_ UT GGACC CCUUHNNOOOONHOU =NNHODz16.2-11A BZn2+25  aminoallyl-deoxyuridine (dUaaTP), a dUTP analogue for mimicking the activity of Lys41 of RNase A. They were first validated as suitable substrates for simultaneous enzymatic polymerization into modified DNA chains, as well as the subsequent reverse transcription of modified DNA back to unmodified cDNA before the initiation of the in vitro selection.174 In addition to catalytic strategy α (in-line nucleophilic attack of the adjacent phosphodiester backbone), that is within the capability of RNA-cleaving unmodified DNAzymes, the appended primary amine (potentially cationic) was designed to employ the catalytic strategy β to neutralize the negative charges on the non-bridging oxygen of the RNA backbone; the appended imidazole was designed to employ the catalytic strategy γ and δ to deprotonate the 2’-hydroxyl group of the nucleoside at the cleavage site and protonate the 5’-oxygen of the downstream (3’) cleavage product as general acid and base, respectively. These catalytic strategies are depicted in Figure 1.2. Adorned by an imidazole and a primary amine, Dz925-11 catalyzed a self-cleavage reaction of a chimeric DNA/RNA substrate at the site of a single embedded ribocytidine with an observed rate constant of 0.044 min-1 at 37°C, pH 7.5 in the absence of any divalent M2+. Dz925-11 was subsequently engineered into a multiple-turnover catalyst, with a moderate catalytic rate constant (kcat) of ~0.037 min-1 and a catalytic efficiency (kcat/KM) of ~5.3 ×105 M-1min-1 at 24°C, pH 7.5.175,176 No Mg2+ is needed for the folding or catalysis, yet the presence of Mg2+ up to 5 mM does not interfere with catalysis. The RNase A-mimicking mechanism of Dz925-11 was experimentally elucidated. As expected, the appended imidazoles were especially crucial for both proper folding and catalysis, with dAim19 and dAim23 acting as the respective general acid and base to 26  enhance transesterification reaction, as His119 and His12 do at the active site of RNase A, respectively; meanwhile, the cationic amine side chain of dUaa21 was responsible for the stabilization of the non-bridging scissile phosphate oxygen (Figure 1.8).177 These enzymatic properties distinguished it from all other catalysts generated by both rational design and combinatorial selection using unmodified libraries. The success of Dz925-11 also demonstrates that a combinatorial DNAzyme selection experiment can result in biomimetic catalysts that function with desired properties of both proteins and nucleic acids to catalyze similar reactions, thereby improving catalytic performances. Nevertheless, Dz925-11 is unable to cleave an all-RNA substrate, and its optimum temperature is 13°C, suggesting insufficient stabilization of the transition state or transient intermediate.178 Thus, further improvements to increase the temperature optimum and catalytic rate on an all-RNA substrate is greatly needed in order to achieve the ultimate goal of in vivo application.    27   Figure 1.8 RNase A-mimicking Dz925-11.  (A) Trans-acting Dz925-11 in complex with the chimeric DNA/RNA substrate containing one embedded ribocytidine (rC) at the cleavage site. The nucleosides in the catalytic core are in bold, with modified ones in blue. A19 and A23 act as the respective general acid and base, and U21 stabilizes the non-bridging scissile phosphate oxygen. (B) The structures of modified nucleosides: 8-histaminyl-deoxyadenosine (A), and 5-aminoallyl-deoxyuridine (U). Reproduced from Reference 177.   In 2004, Sidorov et al. used 5-imidazolyl-modified-dUTP and 3-(aminopropynyl)-7-deaza-dATP in place of respective dTTP and dATP in an in vitro selection with the explicit aim of generating a DNAzyme that acted as an RNase A mimic as well.179 This work led to the isolation of another M2+-independent modified DNAzyme capable of cleaving a 12 nt all-RNA substrate at two cleavage sites, with observed self-cleavage rate constants of 0.06 ± 0.01 and 0.07 ± 0.01 min-1 at pH 7.5, 37°C, respectively (Figure 1.9). The rate constants of this DNAzyme are slightly higher than that of Dz925-11, and its activity can tolerate Mg2+ concentrations as high as 5 mM as well, yet it is not capable of multiple-turnover. When either or both modified nucleotides was/were 3'-TTGTCTGrCCCGTGCG-5'5'-CCAACAG GGCACGC-3'U8U9CU11CA13U14C C GU18A19GU21GA24A23 NN NNNH2OOONHNHNHNNOOOONH2OA = U =A BDz925-118-histaminyl-deoxyadenosine 5-aminoallyl-deoxyuridine28  replaced by their respective natural counterpart(s), this DNAzyme still displayed a certain albeit reduced degree of activity. In particular, the DNAzyme containing only 5-imidazolyl-modified-deoxyuridine still retained 30% of the optimal activity. The observed modification-activity relationship of this DNAzyme suggests that the modifications may play more of a structural role than a catalytic role.   Figure 1.9 Sidorov's modified DNAzyme. (A) Sidorov’s self-cleaving DNAzyme. The 12 nt all-RNA substrate is in red. Arrows indicate the sites of RNA cleavage. The nucleosides in the catalytic region are in bold, with modified ones in blue. (B) The structures of modified nucleosides: 3-(aminopropynyl)-7-deaza-deoxyadenosine (A), and 5-imidazolyl-modified-deoxyuridine (U). Reproduced from Reference 179.   1.5.3 Further expanding the functionalities of DNAzymes  The modest catalytic properties of Dz925-11 along with those of Sidorov’s self-cleaving NN NNH2OOOHNNOOOONHOA =U =NHNNH2OGG AAAA A G   U   A  A C   U   A G A G AUUU C U C UU U C   G   U U G  GCAUGUGUCGGUAAGGCAC5'-3'-A B29  DNAzyme indicate that two modified bases might be not sufficient to improve enzymatic parameters to a higher degree. In addition, the low optimum temperature of Dz925-11 (13°C) also suggested the use of more than two modified nucleotides, especially those adorned with cationic functionalities which can provide further electrostatic stabilization to active folds. In order to improve the temperature optimum, many studies reported the highly stabilizing effect afforded by a gunidinium group primarily due to its cationic character over a wide pH range (pKa 12-13).180-182 As a landmark example, a glutamate-recognizing DNA aptamer containing guanidinium-functionalized deoxyuridine displayed enhanced binding affinity and specificity, further highlighting its potential for anion recognition.182 In addition, appended guanidinium group, which would enhance catalytic action, might also endow duly modified DNAzymes with cell-penetrating ability183 and increased nuclease resistance184, both of which would benefit the subsequent in vivo application of DNAzymes. Therefore, guanidinium was introduced as the third appended functionality in in vitro selections by the Perrin lab. The simultaneous use of 8-histaminyl-dATP (dAimTP), 5-aminoallyl-dCTP (dCaaTP), 5-guanidinoallyl-dUTP (dUgaTP) in in vitro selections led to the isolation of Dz9-96185, Dz10-66186, and Dz12-91187, the three most densely modified DNAzymes bearing three appended functional groups: an imidazole, a potentially basic cationic amine, and a fully cationic guanidinium (Figure 1.10). In contrast to Dz925-11, both Dz9-86 and Dz10-66 achieved temperature optima of above 37°C on the same chimeric DNA/RNA substrate containing one embedded ribocytidine, with improved self-cleavage rate constants of ~0.15 min-1 and ~0.63 min-1 at pH 7.5 and 37°C, respectively, suggesting the enhanced 30  stability of active folds possibly provided by the added guanidinium group. Dz10-66 was successfully transformed into trans-acting DNAzyme, with similar catalytic rate (kcat) constant of ~0.2 min-1 and catalytic efficiency (kcat/KM) of ~ 6×105 M-1min-1 at 24°C and 37°C, pH 7.5. However, neither Dz9-86 nor Dz10-66 was active on the corresponding all-RNA substrate. Dz12-91 was selected with a 12 nt all-RNA substrate, and it underwent self-cleavage reaction on this all-RNA substrate with a maximum rate constant of 0.088 min-1 at 37°C, pH 7.5, yet it was incapable of carrying out an intermolecular cleavage reaction, and thus was also unable to perform multiple-turnover catalysis.    31   Figure 1.10 RNase A-mimicking DNAzymes containing three modified nucleosides. (A) Cis-acting Dz9-86 containing the chimeric DNA/RNA substrate (top panel), trans-acting Dz10-66 in complex with the chimeric DNA/RNA substrate (middle panel), and cis-acting Dz12-91 containing the 12 nt all-RNA substrate (bottom panel). Nucleosides 5'-3'-G C G T G C C rC G T C T G T T G GC G C C C C G C A C G G  A C A  A C CCCCTAGAC1U2A3C5 A6C8C1024U23C21C19CA11 A12U14A1625AU26A28U30U32U36C3438C40UU42U1C2A3U4C6A7C911C13UA1416UU18C19GGGGGG GGGGGGGGGGGGGGGGG3'-G C G T G C C rC G T C T G T T G G-3' C C C C G C A C G G A C A  A C CCCCTAG43AU44CCCTA A   C A  A C C5'-3'- G C G C C CGr(G C G U G C C C G U C U) G T T G G C G C A C G GU2G3A4U5C7A8C1012C14U16UU18C19GGGA13GGA1 Dz12-91   Dz9-86  Dz10-66A   =NNNNNH2HNNHNOOOU =NHN OONHH2NNH2OOONN ONH2H2NOOOC  =5'- 5'-B8-histaminyl-deoxyadenosine5-guanidinoallyl-deoxyuridine5-aminoallyl-deoxycytidine32  in the catalytic cores are numbered with the modified ones in blue. Substrate sequences are in red. Arrows indicate the sites of RNA cleavage. (B) The structures of modified nucleosides: 8-histaminyl-deoxyadenosine (A), 5-guanidinoallyl-deoxyuridine (U), and 5-aminoallyl-deoxycytidine (C).   These findings demonstrate the practicality of using three modified nucleotides in in vitro selection of RNase A mimicking DNAzymes, and the potential utility of such DNAzymes under physiological conditions where modifications play catalytic and/or structural roles to afford absolute Mg2+-independence. 1.6 Potential Advantages of Modified DNAzymes as Therapeutic Candidates for RNA Manipulation In Vivo  Factors that are likely to influence the eventual therapeutic applications of DNAzymes include their stability, local divalent cation concentration, efficient cellular uptake, as well as the subsequent subcellular localization.188-191 Several modifications, such as 2’-OMe ribonucleotides123,134,192 or a 3’-3’ inverted thymidine at the 3’-end193,194, and LNA194,195 have been successfully used to protect DNAzymes from nucleolytic degradations. A number of nucleobase-modified nucleotides have also been demonstrated their resistance to nucleases.196-200 Oligonucleotides containing 8-histaminyl-deoxyadenosine (dAim) and 5-aminoallyl-deoxyuridine (dUaa) showed resistance to several endonuclease including HindIII, XbaI and XmaI.174 Although still no such assessment on oligonucleotides bearing all dAim, dCaa and dUga has been performed thus far, it would be worth expecting that they also display increased resistance to variable cellular nucleases.201 Meanwhile, oligonucleotides containing grafted cationic functionalities, 33  particularly amine202,203 and guanidinium183, have been demonstrated to have mammalian cell membrane penetration ability owing to their property of cell penetrating peptide (CPP) mimicry conferred by their functional groups. CPPs containing multiple “pH-responsive” histidine residues exhibited high transfection efficiencies of their oligonucleotide cargos because of the improved endosomal escape mediated by imidazole functionalities through acidification of endocytic vesicles.204-209 Taking into consideration of all of these properties, RNase A-mimicking DNAzymes represent a highly promising route towards the eventual development of RNA-manipulating therapeutics.  1.7 Specific Research Aims As described in section 1.5.2 and 1.5.3, many efforts have been directed towards the selection, characterization and mechanism elucidation of Dz925-11, Dz9-86, Dz10-66 and Dz12-91. With gradual improvements in catalytic performance including increased temperature optima, catalytic rate, multiple-turnover, propensity to cleave an all-RNA substrate and so forth, these DNAzymes are becoming increasingly close to true in vivo applicability. Nevertheless, none of these species is versatile enough to merit further development. Dz12-91, which has the greatest propensity to cleave an all-RNA substrate, requires the presence of four DNA bases in the substrate sequence to facilitate substrate positioning (bottom panel in Figure 1.10A). Therefore, strictly speaking, Dz12-91 is still not a fully competent all-RNA cleaver. This thesis is focused on further development of RNase A-mimicking DNAzymes capable of M2+-independent all-RNA cleavage reaction with comprehensive properties of high sequence specificity, 34  moderate to high catalytic efficiency, and multiple-turnover, which would provide potent candidates for the development of in vivo RNA-manipulating therapeutics. With the goal of developing such versatile DNAzymes, Chapter 2 discusses (1) the design and optimization of a new selection construct to allow a direct selection of all-RNA cleavers that simultaneously fosters the subsequent cis- to trans-acting DNAzyme transformation. In order to achieve this dual objective, an in vitro selection system integrating selection, re-selection and evolution to enable exploration of an enlarged sequence space and the pursuit of higher-performing sequences through directed and generation-specific mutations was developed. Also detailed in Chapter 2, is (2) the application of a novel construct and the optimized in vitro selection to two parallel selections containing a DNA/RNA chimeric substrate with three RNA bases at the directed cleavage site and a 17 nt all-RNA substrate, respectively. The first selection aimed to test the working efficiency of the new in vitro selection construct and the selection cycle, while the second selection led to the development of all-RNA cleavers. Lastly, Chapter 2 discusses (3) the initial screening of the all-RNA cleavers from the second selection process to find the most efficient candidate for systematic characterization. Chapter 3 demonstrates the systematic characterization of all enzymatic parameters, the validation of sequence specificity, and the probing of an RNase A-mimicking mechanism, using the best DNAzyme obtained (Dz7-38-32). Chapter 4 tests the spontaneous cellular uptake of Dz7-38-32 by HeLa cells due to its similarity to cell penetrating peptides in regard to the functionalities it contains. To understand fundamental aspects by which these three modified bases utilized 35  (histaminyl-dA, guanidinoallyl-dU, and aminoallyl-dC) function in combinatorial selection steps, Chapter 5 investigates the biophysical and enzymatic properties of these three modified nucleosides in the context of discretely modified oligonucleotides. Chapter 6 provides an overall summary of the whole thesis, as well as some insight for the future experiments that may be carried out.       36  Chapter 2: Towards the Design of a Generalizable In Vitro Selection Strategy for Modified All-RNA Cleaving DNAzymes  2.1 Introduction  As introduced in 1.5.2 and 1.5.3 in Chapter 1, for more than a decade since the discovery of the first characterized RNase A-mimicking Dz925-11 containing two modified nucleosides (8-histaminyl-deoxyadenosine and 5-aminoallyl-deoxyuridine), the Perrin lab has been working on developing M2+-independent RNase A-mimicking DNAzymes through in vitro selection using chemically modified nucleotide analogues. For additional stabilizing effects in terms of electrostatic interactions, a highly basic and positively charged alkylguanidinium ion mimicking the side chain of arginine grafted on the 5-position of dUTP was used. Thereafter, Dz9-86, Dz10-66, and Dz12-91 (as depicted in Figure 1.10) containing three modified nucleosides were selected and characterized, with improved catalytic properties beyond those found for their unmodified counterparts30,135,136 and the elegant RNase A mimics resulting from rational design.150,153,154 Nevertheless, they share one common drawback that currently prevents their consideration for therapeutic use in vivo: to date, such catalysts are incapable of cleaving an all-RNA substrate of sufficient length to define a specific mRNA target19, not to mention with multiple-turnover (Table 2.1). The lack of activity on an all-RNA substrate excludes them from any possible in vivo therapeutic applications.   37  Table 2.1 RNase A-mimicking modified DNAzymes developed in the Perrin lab DNAzyme a kobs in cis (min-1) kcat /KM in trans M-1min-1 Catalytic core (nt ) Substrate Turnover Dz925-11 0.044 a 5x105 17 1 embedded RNA base Y Dz9-86 0.134 NA 19 1 embedded RNA base NA Dz10-66 0.57 b 6x105 44 1 embedded RNA base Y Dz12-91 0.06 NA 19 12 nt all-RNA NA a Enzymatic parameters were characterized in M2+-free buffer conditions: 50 mM cacodylate (pH 7.45), 200 mM NaCl, 1 mM EDTA, at room temperature (21-24°C). b Catalytic efficiency was characterized in the presence of physiological concentration of Mg2+: 50 mM cacodylate (pH 7.45), 200 mM NaCl, 0.2 mM Mg2+, at 24°C.  “Y”: capable of turnover, and with characterized catalytic efficiency; “NA”: without detectable turnover, or without characterized catalytic efficiency.  In general, therapeutically-efficient DNAzymes should probably have, at the very least, the following characteristics: (1) validated sequence specificity for cleaving an all-RNA substrate, (2) robust activity at low or no Mg2+ characteristic of cells, and (3) moderate to high catalytic efficiency >105 M-1min-1, with a kcat value > 0.05 min-1 to ensure reasonable mRNA disruption before cell division causes DNAzyme dilution.210,211 With the goal of developing such versatile DNAzymes, this chapter focuses on revising the previously used unimolecular selection construct and the selection cycle in regards to several critical steps affecting sequence space explored, as well the as the subsequent 38  practice of in vitro selections making use of the revised selection construct and selection cycle.  2.2 Optimizing Conditions for More Efficient Primer Extension Reactions to Address the Production of Modified DNA  The simultaneous use of three modified nucleotides has successfully resulted in the discovery of DNAzyme families summarized in Table 2.1, and leads to the hypothesis that enzymatic incorporation rather than highly efficient incorporation is the prerequisite of in vitro selection using modified nucleotides. It is a known fact that the performance of natural polymerases is limited to the acceptance of just a few, or even single unnatural nucleotides; even though many polymerases can accept some components of unnatural genetic systems. Whereas 5-guanidinoallyl-dUTP (dUgaTP) and 5-aminoallyl-dCTP (dCaaTP) (2 and 3, Figure 2.1) are normally tolerated by polymerases, due to the appendages that fall in the major groove of the duplex163,168,212,213, 8-histaminyl-dATP (dAimTP) (1, Figure 2.1) is a confirmed poor substrate that cannot be readily incorporated by most DNA polymerases more than three times in a row214. The failure to isolate DNAzymes capable of cleaving an all-RNA substrate may be related to the compromised sequence space imposed by limited substrate ability of modified nucleotides, in particular dAimTP. To enlarge the sequence space that can be created in the process of transcribing modified library through a template-directed primer extension reaction, I attempted to optimize conditions towards favoring dAimTP incorporation.  39   Figure 2.1 Chemical structures of modified nucleotides for RNase A mimic. dAimTP 1, dUgaTP 2, and dCaaTP 3.  Of various DNA polymerases, Sequenase v2.0 displayed the greatest activity for incorporating dAimTP214, as well as efficient incorporation of the three modified nucleotides at the same time on certain test sequences as well as those that survive selection. Therefore, Sequenase v2.0 has been consistently used for the production of modified libraries at 37°C, its optimum temperature, as described in previous reports.185-187,215 However, the incorporation of dAimTP in the growing strand would likely introduce instability to the newly synthesized double-stranded DNA owing to the bulkiness of the histaminyl appendage on the incorporated dAim. Especially when more than two consecutive dAimTPs were incorporated, presumed collisions between the histaminyl group and the polymerase caused polymerase pausing or stalling, which limits the production of A-rich sequences.  This consequently led to a reduction in sequence space along with the accumulation of a truncation product. A lower temperature of 32°C instead of 37°C and a longer reaction time were found to solve this problem in part. To NNNNNH2HNNHN8-histaminyl-dATP (dAimTP)NHN OONHH2NNH5-aminoallyl-dCTP (dCaaTP)OOH4-O9P3OO4-O9P3ONN ONH2H2NOOH4-O9P3O5-guanidinoallyl-dUTP (dUgaTP)OH1 2 340  monitor the temperature impact on the kinetic reaction progress, template-directed primer extension reactions were carried out using the template for transcribing Dz10-66 at 32°C, 42°C, and temperatures cycled between 32°C and 42°C, respectively, in the presence of dAimTP and unmodified dTTP, dCTP, and dGTP (see Figure 2.2A for primer and template).  Not surprisingly, in early time points, Sequenase v2.0 paused as evidenced by the presence of truncation bands caused by the incorporation of two consecutive dAimTPs. Yet, Sequenase v2.0 was able to extend the aborted bands to fully extended product after long reaction times at 32°C. In contrast, most of the primer just ended up with the aborted bands at the higher temperature of 42°C or temperatures that cycled between 32°C and 42°C (Figure 2.2).     41   Figure 2.2 Template-directed primer extension reaction using Sequenase v2.0 in the presence of dAimTP, dTTP, dCTP, and dGTP at different temperatures. (A) Template (T, ON 1) and primer (P, ON 2) used for primer extension reaction. The aborted sites in the template are indicated by dashed arrows. (B) The autoradiograph of the primer extension reaction. Lane 1, 5’-32P labeled template as size marker for full-length product; lane 2 to lane 5, reaction progress for 0.5 h, 1 h, 2 h, and 3 h at 32°C, respectively; lane 6 to lane 9, reaction progress for 0.5 h, 1 h, 2 h, and 3 h at 42°C, respectively; lane 6 to lane 9, reaction progress for 0.5 h, 1 h, 2 h, and 3 h at temperatures cycled between 32°C and 42°C, respectively; lane 14, reaction mixture omitting template; lane 15, 5’-32P labeled primer.   Therefore, for a template-directed primer extension reaction, a longer reaction time of 4 h at a relatively lower temperature of 32°C was employed to polymerize modified libraries for in vitro selections simultaneously using dAimTP, dCaaTP, and dUgaTP.  1 2 3 4 5 6 7 8 9 10 11 12 13 14 1542°C32°C 32°C & 42°C cycleExtension for 0.5 h, 1 h, 2 h, 3 hP: 5'-32P-CGTCTGTTGGGCCCTACCA-3'T: 3'-TGTGCTCGCACGGGCAGACAACCCGGGATGGTTGTGATCGTCGCGTTCACTCCGCGCGATACTCACACGCACGCACATACGTGCGTTTTT-5'Aborted (indicated in red in the template)Full-length productSize markerABON 1ON 2dAimTP, dCTP, dTTP, and dGTP42  2.3 Unimolecular Construct Optimization for In Vitro Selection  DNAzymes selected with DNA/RNA chimeric substrates have generally proven to be catalytically inept in terms of cleaving all-RNA substrates, a feature seen in all modified DNAzymes in Table 2.1, as well as the first Mg2+-dependent DNAzyme developed by Breaker and Joyce104. Even though Dz12-91 was originally selected with a 12 nt all-RNA substrate, replacing the original substrate with a corresponding 17 nt all-RNA substrate by substituting the five DNA bases involved in substrate positioning still resulted in 2-fold decrease of cis-cleavage rate.187 Moreover, no turnover was possible in constructs designed for such. In contrast, most DNAzymes that have been selected with all-RNA substrates usually show an equal propensity for cleaving corresponding DNA/RNA chimeric substrates, as nicely exemplified by Dz8-17.  No explicit conclusion among different interpretations has been drawn to explain these observations. Cedergren and coworkers216 found that the hammerhead ribozyme showed lower optimal temperatures along with much higher KM values and lower kcat values for cleaving chimeric DNA/RNA substrates wherein RNA bases in the substrate-recognition arms were replaced by DNA bases. In comparison to an all-RNA substrate, which formed a RNA/RNA duplex with the hammerhead ribozyme, a chimeric DNA/RNA substrate formed a DNA/RNA heteroduplex with the hammerhead ribozyme, which had a lower thermal stability relative to RNA/RNA duplex, as reflected by a higher KM value. In addition, the presence of a DNA/RNA heteroduplex instead of the original RNA/RNA duplex also caused a conformational change of the catalytic core within the ribozyme-substrate complex. This catalytic core conformation change resulted in a decreased kcat 43  value. Collectively, similar trends in terms of decreased catalytic performance are noted when DNAzymes and ribozymes were used to target chimeric substrates of varied DNA or RNA compositions. These findings would suggest it might be more practical and straightforward to just design a new unimolecular construct containing an intact all-RNA substrate region towards the selection of real all-RNA cleavers capable of multiple-turnover under physiological conditions.  One practical reason that a 12 nt all-RNA substrate was built in the unimolecular construct for the selection of Dz12-91 (Figure 2.3A) was the requirement of enough constant region length for primer binding in the PCR amplification process, as depicted by P2. Consequently, five DNA bases (5’-GTTGG-3’) at the 5’-end of P2 rather than real RNA bases had to participate in substrate positioning as part of the substrate region. To convert the cis-acting (intramolecular) to a trans-acting (intermolecular) DNAzyme by removing the dispensable Loop I, extra RNA bases had to be added to the 3’-end of the 12 nt substrate to provide a longer substrate that would in turn provide sufficient binding affinity as the 5’-end bases did in P2. However, after transformation in this manner, Dz12-91 showed no activity in trans, suggesting that such sequence engineering is highly likely to introduce catalytic conformational change in the DNAzyme-substrate complex and which consequently caused a decrease or loss of catalytic rate and catalytic efficiency. Therefore, a revised construct in Figure 2.3B was designed by directly building a 17 nt all-RNA substrate as well as a larger loop (Loop II) which provides sufficient length for highly specific primer binding during the PCR amplification step. As an extra benefit, the new construct in Figure 2.3B is not expected to form an 44  intramolecular hairpin structure formed in P2.  The lack of such secondary structure further increases the specificity of PCR amplification. Additionally, it is known that a larger loop in a hairpin structure leads to decreased rate of hairpin formation and stability.217,218 By the same token, the relatively large Loop II in the new construct would hypothetically make the intramolecular interaction between the substrate domain and the DNAzyme domain increasingly similar to the intermolecular reaction. In other words, the larger Loop II on one hand increases selection stringency for intramolecular cleavage reaction, on the other hand eases the conversion of a cis-acting DNAzyme to a trans-acting DNAzyme by removing the dispensable loop.  Figure 2.3 5’-biotinylated unimolecular constructs for in vitro selections.  Letters in red in the parenthesis denote all-RNA substrate, and they are positioned right opposite the bulge (drawn in blue) of a stem-bulge structure formed between the RNA substrate and the substrate-recognition arms (Arm I and Arm II) of the DNAzyme CCCTA A   C A  A C CB5'-3'-C T C G A  G C G C C CGr(G C G U G C C C G U C U) G T T G G C G C A C G GCatalytic coreP1P2 A   C A  A C CB5'-3'-C G C G A  G C G C G Cr(G C G U G C C C G U C U  G U U G G) C G C A C G GCatalytic coreP1'P2'TT TTCTCCCTCGGGGGGArm IArm IArm IIArm IILoop ILoop IIAB45  sequence. Primer binding regions are labeled with arrows named P1 and P2 in construct (A), and P1’ and P2’ in construct (B). (A) The old selection construct leading to the isolation of Dz12-91 containing the 12 nt all-RNA substrate and the small dispensable loop (Loop I). (B) The revised selection construct containing a 17 nt all-RNA substrate and the enlarged dispensable loop (Loop II).   With this new construct, two new selections were initiated, one of which still used a DNA/RNA chimeric substrate containing only three unpaired RNA bases at the directed cleavage sites for testing the new construct (Figure 2.4A), while the second selection contained a 17 nt all-RNA substrate was towards the development of new families of real all-RNA cleaving DNAzymes (Figure 2.4B).  The substrates here were the same sequence derived from HIV-1 LTR-promoter mRNA that was previously used for DNAzymes listed in Table 2.1. Different functional regions in the constructs are color coded as illustrated in Figure 2.4C. 46    Figure 2.4 Sequence schemes used for selections. 5’-poly-T tail in cyan serves for a large difference in migration on PAGE gel between the intact full-length product and that undergoes self-cleavage reaction. Substrate region is depicted in red, with potential cleavage reactions directed to happen at the unpaired bases indicated by arrows. Modified sequence region in blue starts from the beginning of the randomized library region (N40) till the 3’-end. The primer binding regions for PCR amplification are highlighted by the magenta dashed lines. Sequence region in black hybridizes to template for template-directed primer extension reaction to make the modified library. (A) Sequence scheme containing the chimeric substrate containing 5'-T40-C C C G G G T T T T T (G C G U G C C C G U C U G U U G G)rG A C A A C CC G C A C GN40TTTTCTCCCTC-C G C G A G C G C G C3'5' multi-T region Substrate sequence region derived from HIV-LTR promoterModified regionPCR primer binding regionGGGGGG5'-T40-C C C G G G T T T T T  G C G T G C C (C G U) C T G T T G G)rG A C A A C CC G C A C GN40TTTTCTCCCTC-C G C G A G C G C G C3'GGGGGGDirected cleavage siteABCBinding site for primer extension reaction47  three embedded RNA linkages. (B) Sequence scheme containing the all-RNA substrate. (C) Color codes of different functional regions in the unimolecular constructs.   2.4 In Vitro Selection  In vitro selections were performed following the protocol described by Perrin173, which was invented by Breaker and Joyce28 using the streptavidin-biotin methodology. After some adaptive revisions, the selection cycle is detailed in Figure 2.5. In detail, the modified library was prepared by a standard template-directed primer extension reaction using Sequenase v2.0. The primer in question was a 5’-biotinylated DNA/RNA chimeric sequence containing the RNA substrate sequence and a poly-T (T45) tail at the 5’-end that provided a large electrophoretic migration difference on PAGE gel between the intact full-length modified DNA molecule and the so-called active one that underwent a self-cleavage reaction. Based on the different substrate properties of dAimTP, dCaaTP and dUgaTP, unequal concentrations of them were used along with the natural dGTP (50 µM of dAimTP, 25 µM of dCaaTP, 10 µM of dUgaTP and 10 µM dGTP) in the primer extension reaction to decrease, at least partially, the sequence bias, that is caused by using modified nucleotides in in vitro selection. The enzymatically synthesized modified DNA, associated in heteroduplex form containing the template strand was immobilized on streptavidin coated magnetic beads through the streptavidin-biotin interaction. The template strand was then removed by five consecutive washes using diluted NaOH (0.1 M, 1 mM EDTA). The wash process here must be rapid in order to minimize the background cleavage of substrate RNA by NaOH.  The resulting single-stranded modified DNA was subjected to self-cleavage in absolutely M2+-free buffers (50 mM 48  sodium cacodylate, pH 7.45, various concentrations of NaCl, 1 mM EDTA) for different periods of time. The NaCl concentrations in cleavage buffers as well as the reaction times were the two general paremeters for controling different selection stringencies throughout the selection processe. Sometimes, other reagents such as denaturants of nucleic acids can be added to different levels for the isolation of very stable secondary structures. The active population that underwent self-cleavage reaction leading to removal from the beads was collected, precipitated, resolved and purified using denaturing PAGE gel (7 M urea). Following elution from the gel, the material was PCR amplified twice. The first PCR served to copy the modified strands back into unmodified DNA strands. Due to the intrinsic difficulty for DNA polymerase to use modified DNA stands as templates to synthesize complementary unmodified DNAs, the first PCR amplification can be artifact-prone, therefore 20 to 25 cycles were preferred instead of >30 cycles to decrease the accumulation of artifacts. The product of the first PCR amplification was purified by a denaturing PAGE gel according to previous reports185,186 and this product was then subjected to reamplification in a second PCR. In the second PCR, the primer for the first strand (template strand) synthesis was substituted by the same sequence with a 5’-T20 tail to give a product of greater length that also migrated differently from the self-cleaving species. The non-template stand which was amplified from a 5’-phosphorylated PCR primer was digested using lambda exonuclease, and the resulting template strand containing the 5’-T20 tail served as the template for the following round of selection.   49   Figure 2.5 In vitro selection cycle for RNA-cleaving modified DNAzymes. 1) Modified DNA library preparation by template-directed primer extension reaction using Sequenase v2.0 in the presence of unequal concentrations of nucleotides: 50 µM dAimTP, 25 µM dCaaTP, 10 µM dUgaTP, and 10 µM dGTP. Template contains a randomized region (N40) flanked by constant primer binding sequences at the 5’- and 3’-termini. The extension primer is a 5’-biotinylated DNA/RNA chimeric sequence which serves as substrate for RNA cleavage in cis.  2) Modified library is captured by streptavidin beads, allowing removal of template by quick wash with 0.1 M NaOH. 3) Single-stranded modified library is incubated in reaction buffer. Active DNAzymes adopting acitve conformations cleave the RNA substrate in cis at the preprogrammed B -3'- -5'5' Biotin3'N40B- -3'B5'-3'-N40xxxx-5'-B - -3'5'-intact uncleaved population-3'B -5'- -3'5'-cleaved population-3'5'-5'- -3'2)3)4)unequal molar of modified dNTPs5)6)new library template1)PAGE separationPCR amplificaion(1st- & 2nd-PCR)50  unpaired bulge region (indicated by xxxx). 4) Size separation of cleaved DNAzymes from uncleaved library using denaturing PAGE (7%). 5) Self-cleaved DNAzymes are gel extracted using gel elution buffer (10 mM Tris, pH 8.0, 1% LiClO4). 6) PCR amplification of self-cleaved DNAzymes using Vent (exo-) DNA polymerase. The 1st-PCR amplification reverse transcribes self-cleaved DNAzymes into unmodified cDNA molecules (modified DNAzymes→unmodified cDNA molecules). To minimize artifacts accumulation, 20 to 25 cycles of nested PCR are carried out. Non-template strand of the PCR product generated by the 5’-phosphorylated primer is digested using lambda exonuclease to leave only the template strand. The 2nd-PCR (prep-PCR, 30 cycles) and subsequent lambda exonuclease digestion of 5’-phosphorylated non-template strand to produce a new library template for initiating a new round of selection.  2.4.1 In vitro selection of DNAzymes cleaving a chimeric substrate containing three embedded RNA linkages  The selection targeting a chimeric substrate containing three RNA linkages started from a totally randomized library containing 40 randomized regions, which had a theoretical library size of 440 (=1024). In reality however, 15 pmole of modified DNA were used for each round of in vitro selection, which covered about 7.5x10-10 % of the theoretical sequence space, corresponding to a population of ~1013 of variants at most. Following sequestration on streptavidin and removal of the template strand, the modified library was incubated in standard cleavage buffer (50 mM cacodylate, pH 7.45, 200 mM NaCl, 1 mM EDTA) for 1 h, the first visible band resulting from self-cleavage reaction was observed in selection round 3. The activity rapidly evolved to provide robust cleavage starting in round 4 as evidenced by the kinetic increase of self-cleavage activity. Over the course of a 1 h cleavage reaction, the cleaved population was collected at different 51  time points from round 4 onwards, but only the population collected from the earliest time point representing the fast-to-cleave species was isolated for PCR amplification and propagation to the following round of selection. As an increase of selection stringency, the earliest time point was decreased from 60 min to 1 min from round 4, and to 0.5 min from round 5 (round 1 to 3: 60 min; round 4: 1 min; round 5 to 10: 0.5 min). Additionally, the cleavage reactions were carried out in sodium phosphate buffers (25 mM sodium phosphate, pH 7.4, 1 mM EDTA) with gradually decreasing NaCl concentrations from 100 mM (round 7) to 50 mM (round 8 and 9), and then to 0 mM (round 10). Afterward, the self-cleavage reactions were carried out in urea-denatured buffers from round 11-12 as a further increase of selection stringency. Representative selection gel pictures are shown in Figure 2.6. The most noticeable aspect of gel pictures G9 and G11 was that lane 3 to 6 all contained substantial amounts of uncleaved (full-length) product. This observation reflected an incomplete separation of active population from inactive one, which was probably caused by the imperfect retention by poor streptavidin magnetic particles of that specific batch. Additionally, the presence of 1 M urea in the cleavage buffer of G11 might have disrupted the streptavidin-biotin interaction to a certain level, which also led to the increased amount of full-length material release by the magnetic beads. The gel purification step helped to remove these uncleaved sequences, which would otherwise contaminate the selection and consequently affect the progress of selective enrichment of active sequences.  52   Figure 2.6 Representative gel pictures of different rounds of selection against the chimeric substrate containing three embedded RNA linkages. All 1st lanes show the intact full-length DNAs staying on the streptavidin beads after self-cleavage reaction, and all 2nd lanes are 0.1 M NaOH treated streptavidin beads after self-cleavage reaction. (G3) Selection round 3 in standard cleavage buffer. (G4) Selection round 4 in standard cleavage buffer. (G9) Selection round 9 in cleavage buffer containing 25 mM sodium phosphate (pH 7.4), 0 mM NaCl, 1 mM EDTA. (G11) Selection round 11 in cleavage buffer containing 10 mM sodium phosphate (pH 7.4), 1 M urea, 1 mM EDTA. Standard cleavage buffer: 50 mM sodium cacodylate (pH 7.45), 200 mM NaCl, 1 mM EDTA.  Full-length productSelf-cleavage productBeads+_Beads+_Beads+_Beads+_Full-length productSelf-cleavage productSelf-cleavage in 1, 10, 60 minSelf-cleavage in 0.5, 1, 10, 60 minSelf-cleavage in 60 min1 2 3 1 2 3 4 51 2 3 4 5 6 1 2 3 4 5 6  G3 G4G9  G1153  Selection progress is summarized in Figure 2.7. Notably, the population passed through all the 12 rounds of selections can still function in 2 M urea denatured conditions with about 5% self-cleavage yield within 1 h. This observation suggests that some active sequences adopt active folds that are tolerant to the presence of denaturants, probably due to the additional interactions afforded by the grafted functionalities. To the best of my knowledge, no DNAzymes have been reported to function under such denaturing conditions containing no supplemental divalent metal cations as has been the case for the last two rounds of selections.    Figure 2.7 In vitro selection progress of DNAzymes cleaving the chimeric substrate containing three embedded RNA linkages. Plot of percentage of population cleaved versus selection round (G1 through G12), [NaCl] (mM) 200 100 50 0[Urea] (M) 0 1 20	2	4	6	8	10	12	G1	 G2	 G3	 G4	 G5	 G6	 G7	 G8	 G9	 G10	 G11	 G12	Percent Cleavage  Generation  Progress of selection against the chimeric substrate containing three embedded RNAs   0.5	min	1	min	10	min	60	min	54  where G1 designates the population resulting from amplification of self-cleaved species from the first round of selection, and G12 designates the DNA population resulting from 12 rounds of selective amplification. The NaCl and urea concentrations contained in cleavage buffers are labeled underneath the graph.  2.4.2 In vitro selection of all-RNA cleaving DNAzymes  Different from the selection against the chimeric substrate containing three RNA linkages, the 15 pmole of starting pool for the selection against the all-RNA substrate contained 50% of N40 random sequence, and 50% of the G3 (third generation) obtained from the selection targeting the chimeric substrate containing three RNA linkages. By introducing an amount equaling to approximately half of the partially converged library capable of cleaving the chimeric substrate (7% of crude cleavage in 1h), some of the newly introduced species were expected to behave as “breeders” for the selective enrichment of all-RNA cleavers with an assumption that these added species might be active on both the chimeric and the all-RNA substrates. As expected, all-RNA cleavage activity was observed from the very first round of selection. However, the manner of starting an in vitro selection with a partially converged library might raise the risk of losing genotypic diversity in favor of a quick discovery of desired phenotypes.  Following the similar selection pathway that I performed with the chimeric substrate containing three embedded RNA linkages (described in 2.4.1), nine rounds of selective amplification of fast-to-cleave populations accompanied with gradually increased selection stringency by shortening reaction time and decreasing the ionic strength were carried out until no further improvement in activity was observed. Stringency control with 55  denaturants was not applied to this selection considering the ultimate function arena of cellular conditions for these selected all-RNA cleavers. G9 displayed crude cleavage yields of  ~4% and ~6% in 10 min and 60 min of cleavage reaction time, respectively. The selection progress is shown in Figure 2.8.    Figure 2.8 In vitro selection progress of all-RNA cleaving DNAzymes.  Plot of percentage of populations cleaved versus selection round (G1 through G9), where G1 designates the population resulting from amplification of self-cleaved species from the first round of selection, and G9 designates the DNA population resulting from 9 rounds of selection amplification. NaCl concentrations contained in cleavage buffers are labeled underneath the graph.   0"2"4"6"8"10"12"G1" G2" G3" G4" G5" G6" G7" G8" G9"Percent cleavage Generation  Progress of selection against all-RNA substrate    0.5 min  1 min  10 min  60 min  [NaCl] (mM) 200 100 50 056  2.5 Cloning, Sequencing, and Single-clone Activity Screening 2.5.1 Cloning and sequencing of the selection targeting the chimeric substrate containing three embedded RNAs  Before proceeding to cloning, the last three generations of selection for cleavage of the chimeric substrate with three embedded RNAs (G10 to G12) were compared for crude cleavage activity in the standard cleavage buffer conditions of the same stringency (50 mM cacodylate, pH 7.45, 200 mM NaCl, 1 mM EDTA). G11 displayed the best activity in terms of both the fast-to-cleave population in 1 min reaction time and the total fraction cleaved in 60 min (Figure 2.9).   Figure 2.9 Comparison of G10 to G12 on their activity of cleaving the chimeric substrate containing three RNA linkages in standard cleavage buffers. The amplicon resulting from amplification of fraction cleaved in 1 min reaction time from G11 was cloned and sequenced (as indicated by the arrow).  The population collected after 1 min cleavage reaction from G11 was amplified using Taq DNA polymerase with thermal cycles containing a final extension step of 30 min to 0	5	10	15	20	25	30	G10	 G11	 G12	Percent cleavage  Generation  Chimeric substrate cleavage activity in the presence of 200 mM NaCl  1	min		10	min		60	min		57  generate PCR products with 3’-A overhangs. The amplicons were purified on a 2% agarose gel using GeneJet Gel Extraction Kit before they were TOPO cloned into the pCR2.1-TOPO vector according to the manufacture’s instruction. The vector was then used to transform E.coli DH5α following a normal chemical transformation protocol. White transformed colonies were picked by Blue-White screening on LB Agar plate containing 100 mg/L ampicillin for inoculation. Plasmids were prepared by using PureLink Quick Plasmid Miniprep Kit (Invitrogen) followed by fragment insertion check by endonuclease EcoR I digestion. Plasmids containing a single insertion of the amplicon were sequenced using M13R primer by Nucleic Acid Protein Service Unit of UBC (NAPS-UBC). 29 individually different sequences were finally obtained and are described in Table 2.2. These sequences had randomized regions ranging from 39 to 42 nucleosides, most of which were G-rich, with an average number of 18 Gs, a value that is significantly higher than the statistically random number of 10. This G-rich phenomenon is more likely to be attributed to the selection criterion of cleaving the chimeric substrate containing three RNA linkages and the accumulative selection pressure imposed by multiple rounds of selection. In particular, the presence of urea in selection round 11 was prone to G-rich sequences that are inclined to fold into more denaturant-resistant structures.219,220 Conversely, the G-rich phenomenon here is very unlikely to be caused by the commonly observed “unidirectional losses of unnatural components” accompanied with amplification of them,221,222 since the three modifications (dAim, dCaa, and dUga) have been experimentally proven to be not very “deleterious”, which is demonstrated in detail in Chapter 5 of this thesis.   58  Table 2.2 Clone sequences isolated from selection against the chimeric substrate containing three embedded RNA linkages Clone (Dz) Sequence (5’-3’, catalytic) 11-1 TAAGCAGCGCATGTGAGGCACGCCATGAGTGTGCGTGTGTG 11-2 TAAGCAGCGCATGTGAGGCATGCTATGAGTGTGCGTGAGTGT 11-3 TAAGCAGCGCATGTGAGGTGCGCTATGACTGTGAGTGCGTGT 11-4 TATGCGTGCGGTCCCAGATGAGGTGGTATGCAGTCCGCAT 11-5 TAGTGGTATGTGTCCGCGTGGTCGGAGGTGGACCGCGTGT 11-6 TGGCGGTGTCGAGGGGTTGCGTATGCAGTGTGGGTCGTAT 11-7 TGTTCGCGTTGGGGTCGGTGCTGGTTGCTCGGTGTGTCAC 11-8 TGAGTAGGGGGCTCGGTAGGGCGTGGGGGGGCCTGTGTGT 11-9 TGTGCAGGTGTGGGTCGGGGGTTGCCTGGAGGGTGTGCGG 11-10 TTTCAGCGCATGTGTCGCACGCATGTGAGTGTGTGTGTTT 11-11 GCGGTGTTGGGTGGTGCACAGAGTACGGTGGTCGAGCTAT 11-12 GCGGTGTTGGGTGGTGCACAGAGTACGGTGGTCGAGCTAT 11-13 GCCATGGGGCGATTGTCGTGCGCTTGGGGGATCAGTGCGT 11-14 GTGGGAGCTGTGCGGCGGGGGTGTCAGTGTGTTGCCCTT 11-15 GTGGGGCTGGCTTGGTATTGGGTCGTGCCGGTGTGGGTGT 11-16 GTGTGTAGCGGTGGAGTTGCGCTGGGAGGGTTTTGGGGTT 11-17 CTGGGGGCTATGCGTCGGGGGGTGTCAGTGTGTTGCAGTT 11-18 CTTGCAGCGGGGGCCGGGCTGTAGTGTAGCGGGTGTTGTA 11-19 CCCAGTGCTCTGGCCGCACGGTCTAGTGGCGAGAGGTGTT 11-20 CGCCGTAGTGAGTGTGTGCGGGTTGTGGTTGGGCGGTGTT 11-21 CGGCCCATGGTCGGGCCTTTGTGTAAGGTCGGGGCCTGTT 11-22 CGGTGGCTGGGTGGGTAGCGGGTTGGCATGCGGTGTCCGT 11-23 ATGCAGCGCACGTGTCGCACGCATGCGAGTGTGTGTGTAT 11-24 ATGCAGCGCATGTGAGGTGCGCTATGAGTGAGAGGGCGTGTA 11-25 ATGCGGGTCGCAGGTGGGGTGGTATGTGCCGGCTTCGTAT 11-26 ATCCGCCGGTGCCCGCCTGACCCCACAAACACGGCATCCG 59  Clone (Dz) Sequence (5’-3’, catalytic) 11-27 ATGAAATCCCGTCTATCGGATTTGCTATGTACCGCTGAT 11-28 ACAAAAACATCACAGCGATGAACCACACATCGCGCTTATT 11-29 AGCCGGTGAGGATGGTGTGCGCTTGGTTGGGGTGGCCTTA Only the sequence in the randomized region is shown from the 5’ end to the 3’ end.   The selection for cleavage of the chimeric substrate containing three embedded ribose linkages was performed primarily for testing the new unimolecular construct, which has been demonstrated to work in the context of selections with different versions of RNA substrates, respectively. It was then discontinued after obtaining individual sequence information. Efforts were focused on the re-selection, single-clone activity screening, and systematic characterization of DNAzymes isolated for cleaving the all-RNA substrate due to the long-term goal of seeking efficient all-RNA cleavers for mRNA disruption in vivo. 2.5.2 Cloning and sequencing of the 7th Generation (G7) capable of cleaving an all-RNA substrate  To decide which generation to clone for the selection targeting the all-RNA substrate, G6 to G9 were compared for their self-cleavage activity under simulated physiological buffer conditions (10 mM sodium phosphate, pH 7.45, 150 mM KCl, 0.5 mM MgCl2), in line with the long-term goal of seeking efficient all-RNA cleavers for mRNA disruption in vivo. G7 was finally chosen for cloning and sequencing in regards to its highest percent of fast-to-cleave population (~ 2.3% in 1 min) and active population in total (fraction cleaved in 60 min, ~10%). The activity comparison chart is shown in Figure 2.10.  60    Figure 2.10 Activity comparison of G6 to G9 in cleavage buffer of simulated physiological conditions. The amplicons of cleaved population corresponding to 1 min cleavage reaction from G7 (as indicated by the arrow) were subjected to cloning and sequencing.   As described for the cloning and sequencing work done for the selection with the chimeric substrate containing three embedded RNA linkages in 2.5.1, the collected population of sequences capable of cleavage within 1 min was re-amplified using Taq DNA polymerase, cloned, sequenced, and screened for self-cleavage activity. Initial kinetic analysis of the acquired clones by fitting the percent of self-cleavage collected at 5 different time points (0, 1, 10, 60 and 180 min) to a single exponential first-order 0"2"4"6"8"10"12"G6" G7" G8" G9"Percent cleavage  Generation All-RNA cleavage under simulated physiological ionic strength 1"min""5"min""10"min""60"min""61  reaction (Equation 2.1) revealed that 11 clones were active among 58 screened ones. These active clones were categorized to two families according to the respective consensus of 5’-sequences (Table 2.3).  Equation 2.1: P! = P! 1− 𝑒!!!"#! , where Pt  is the total fraction cleaved at time point t, P! is the fraction underwent cleavage reaction at the end time point, and kobs is the first order rate constant. Table 2.3 Active sequences isolated from selection of all-RNA cleaving DNAzymes Family Clone (Dz) Sequence (5’ to 3’, catalytic)   7-10 7-38 7-44 7-50 7-55 7-57 7-60 TTACAGTGGTAGCGGTTGGCACGTGTGCAGCGTAAGTGGGCG  TTACAGTGGTAGCGGTTGGCATGTGTGCAGCGTAGGTGGGCG  TTACAGTGGTAGCGGTGGCATGTGTACAGCGTTAGTGAGTG I TTATAGTGGTAGCGGTTGGCATGTGTGTGGCGTGAGTTAGTG  TTACAGTGGTAGCGGTTGGCATGTGTGGAGCGTAAGTGAGCG  TTACAGTGGTAGCGGTTGGCATGTGTGCAGCGTAAGTGGGCG  TTACAGTCGTAGCGGTGGCATGCGTGCAGAGTAAGTGGGCG II 7-45 AAGCAGCGCATGTGATGGCACGCGTAGTGGCTGGGTAGCT Only the sequence in the randomized region is shown from the 5’ end to the 3’ end. Letters in blue denote modified bases.   62  The activities they displayed did not seem to be satisfying, with kobs values of ~0.02 min-1 and ~0.01 min-1 for the best representatives from each family, Dz7-38 and Dz7-45, respectively (Figure 2.11). The most probable reason for which faster self-cleavers could not be isolated was attributed to the limited sequence diversity given by the starting pool. Both the use of modified nucleotides and the introduction of a partially converged library from the selection against the chimeric substrate containing three embedded RNA linkages might have caused considerable reduction in sequence space.  Figure 2.11 Kinetic self-cleavage reactions of representative clones.  In each panel, lane 1 represents full-length product, lane 2 is NaOH treated DNAzyme as size marker for the self-cleavage product, and lane 3 to lane 7 show the kinetic self-cleavage progress from time point 0 min to 180 min. (A) Initial kinetic self-cleavage gel of Dz7-38. (B) Initial kinetic self-cleavage gel of Dz7-45. Beads+_Beads_ +Full-length productSelf-cleavage product 0     1    10    60  180   min1 2 3 4 5 6 7 1 2 3  4   5   6    7ADz7-38, kobs~0.02 min-1 0     1    10    60   180   minBDz7-45,  kobs~0.01 min-163  2.6 Re-selection and Evolution of All-RNA Cleaving DNAzymes  2.6.1 Initiation of two lineages of re-selections  To improve the catalytic performance of all-RNA cleaving DNAzymes, whilst gaining an insight into the conserved catalytic motif, two re-selections were initiated by partially randomizing the catalytic regions of Dz7-38 and Dz7-45. In detail, random mutations at a frequency of 15% per nucleoside position over all the positions that encompassed the catalytic center were introduced to the parental sequences. Thereby, two separate libraries containing ~0.1% and ~0.15% of their own parental sequences were built, respectively (Figure 2.12). Members of the two libraries were then challenged to perform all-RNA cleavage reaction by being subjected to iterative rounds of selection and amplification as described in 2.4.2.  64   Figure 2.12 Partially degenerated libraries for re-selection.  Each nucleoside in the underlined sequence regions was mutagenized at a degeneracy of 15%. N1, N2, N3, and N4 represent partially degenerated A, C, G, and T, respectively. “85” represents the percent of the original nucleoside; “05” represents the percent of introduced mutagenicity of each of the other three nucleosides. (A) Re-selection Family I using library degenerated from Dz7-38. (B) Re-selection Family II using library degenerated from Dz7-45.   2.6.2 Mutagenic PCR  In addition to re-selection, “directed evolution” through mutagenic PCR aiming to introduce additional sequence diversity was integrated into the re-selection process. In one of the early Joyce papers, the authors pointed out that evolution of nucleic acids requires the integration of three chemical processes: selection, amplification, and A Family I, Dz7-38 degenerated library:  5’-GCGCTCGCGCGGCGTGC C(N2:05850505)G(N3:05058505)(N2)(N2)(N2)A(N1:85050505)(N2)(N2)T(N4:05050585)(N1)(N2)(N3)(N2)(N4)(N3)(N2)(N1)(N2)(N1)(N2)(N1)(N4)(N3)(N2) (N2)(N1)(N1)(N2)(N2)(N3)(N2)(N4)(N1)(N2)(N2)(N1)(N2)(N4)(N3)(N4)(N1)(N1) CTGTTGGCGCAGGCCGACGC-3’. B Family II, Dz7-45 degenerated library:  5’-GCGCTCGCGCGGCGTGC A(N1:85050505)G(N3:05058505)C(N2:05850505)T(N4:05050585)(N1)(N2) (N2)(N2)(N1)(N3)(N2)(N2)(N1)(N2)(N4)(N1)(N2)(N3)(N2)(N3)(N4)(N3)(N2)(N2) (N1)(N4)(N2)(N1)(N2)(N1)(N4)(N3)(N2)(N3)(N2)(N4)(N3)(N2)(N4)(N4) CTGTTGGCGCAGGCCGACGC-3’ 65  mutation. Ideally, selection operates to reduce variability by excluding those individuals not conforming to the fitness criterion imposed by the selection at the level of phenotype; amplification results in additional copies of the genetic information stored in the selected phenotypes; mutation operates at the level of genotype to introduce variability, which in turn codes for a range of behavioral phenotypes. These three processes must be linked so that molecules with selected phenotypes are enriched by amplification, and subjected to mutations to produce new distributions of mutant individuals (Figure 2.13).223,224    Figure 2.13 Functional components of a nucleic acid-based evolution system.  Arrows represent individual nucleic acid species; thickness of arrows corresponds to copy number. Reproduced from Reference 224.   AMPLIFICATIONSELECTIONEVOLUTIONMUTATION66  In practice, Ying-Fu Li et al. demonstrated the strategy of evolving a degenerate DNAzyme library produced by chemical mutagenesis through further application of mutagenic PCR, which facilitated the isolation of rare DNAzyme motifs that were not isolated in the original selection. They showed that when provided with appropriate mutations, DNAzymes can adopt different structural folds from pre-existing structures to give rise to better catalytic performance.225 In addition, even though comprehensive strategies especially a better template-directed primer extension conditions have been employed to alleviate the decreased sequence space imposed by using three modified nucleotides in in vitro selection, the requirement of selecting better all-RNA cleavers with improved catalytic rates encourages the introduction of mutagenic PCR, which operates to introduce a certain level of new sequence variance to allow an exploration of an enlarged sequence space. Meanwhile, iterative amplification unavoidably leads to accumulations of artifacts, which may be further magnified when the populations being amplified are densely modified DNA sequences. Mutagenic PCR, by producing sequence variability through introducing errors, also might hypothetically help to circumvent the “overwhelmingness” of in vitro selections by PCR artifacts.19 There are several standardized protocols to promote the incorporation and extension of mismatched nucleosides, and they either utilize altered reaction conditions or a mutant version of a thermostable DNA polymerase. The protocol used by Li et al. just mentioned above is a “hyper-mutagenic” PCR procedure that achieves an error rate of about 10% per nucleotide position with only modest sequence bias. Compared to a standard PCR, this protocol employs added MnCl2, increased concentrations of MgCl2, 67  and heavily unbalanced concentrations of the four dNTPs, as well as increased number of thermal cycles (50 instead of normally used 30).225 Another protocol devised by Cadwell and Joyce226 achieves a moderate error rate of 0.66% per nucleotide position with roughly equal probability of all possible transitions and transversions. In this protocol, in addition to the utilization of added MnCl2, increased concentration of MgCl2, and lightly unbalanced concentrations of the four dNTPs, increased concentration of Taq DNA polymerase is used as well. We adapted the Cadwell and Joyce procedure to the second amplification step, and the reaction conditions are: 10 mM Tris (pH 8.3), 50 mM KCl, 7 mM MgCl2, 0.5 mM MnCl2, 0.01% (wt/vol) gelatin, 0.2 mM of dGTP and dATP, 1 mM of dCTP and dTTP, 8.7 µM of first primer, 6.5 µM of second primer, and 50 units of Taq polymerase in a 100 µL of reaction volume (Table 2.4). Specifically, (1) the concentration of MgCl2 is increased to 7 mM to stabilize mispairs;227,228 (2) the addition of 0.5 mM of MnCl2 diminishes the template specificity of the polymerase;229 (3) increased concentrations of dCTP and dTTP promote misincorporation;230 (4) and the significantly increased amount of Taq polymerase promotes chain extension beyond positions of mispairs. A comparison of conditions between standard PCR, Cadwell and Joyce’s mutagenic PCR, and the mutagenic PCR after adaption by this chapter is tabled in Table 2.4.     68  Table 2.4 Conditions for standard PCR, mutagenic PCR devised by Cadwell and Joyce, and mutagenic PCR we adapted from Cadwell and Joyce Reaction condition Standard PCR for selection Mutagenic PCR Cadwell & Joyce Our adaption  Reaction buffer: 10 mM Tris (pH 8.3), 50 mM KCl MgCl2 (mM) 3 7 7 MnCl2 (mM) 0 0.5 0.5 Primer concentration (µM) 1st primer 8.7 0.3 8.7 2nd primer 6.52 0.3 6.52   Nucleotide concentrations (mM) dATP 0.3 0.2 0.2 dGTP 0.3 0.2 0.2 dTTP 0.3 1 1 dCTP 0.3 1 1 DNA polymerase (units/100 µL) Vent (exo-), 10 Taq, 5 Taq, 50 All the concentrations are under 1X reaction conditions.   2.6.3 Re-selection progress For both of the degenerated libraries, self-cleavage activities were restored from round 2, with very obvious cleavage bands on 7% denaturing PAGE (7 M urea). From round 3 69  to 10, active species that had self-cleaved within 10 min were amplified for the following round of selection. In contrast to the selection process of all-RNA cleavers, in which active species that had self-cleaved within 0.5 min were enriched in the later stages of selection, reaction times for self-cleavage of later generations were shortened to and kept at 5 min without further time reduction from round 11 until the last round of the re-selection process. The relatively lax selection stringency in terms of reaction time was chosen to balance the desire for a fast catalytic rate and the need to maintain sequence diversity. In addition to templates generated from amplifying the self-cleaved species using regular PCR protocol, templates from rounds 5, 8, 10 and 13 were supplemented with 10% of the DNA that had been subjected to mutagenic PCR for amplifying the self-cleaved populations resulting from previous rounds. Interestingly, the observed self-cleavage activity dropped in rounds containing introduced mutations yet increased immediately in the following round. This phenomenon indicates that some sequences introduced by mutagenic PCR bore valuable mutations for improving on pre-existing active species. Along with shortening reaction time, the reaction stringency was also increased by decreasing concentrations of NaCl in cleavage buffers. A total of 14 rounds of selection were carried out for each lineage of re-selection before subsequent cloning. Cleavage yields of ~7.3% and ~4% within 5 min reaction time for the two end pools were obtained, respectively. Re-selection progress is summarized in Figure 2.14.    70   Figure 2.14 Re-selection progress of all-RNA cleaving DNAzymes.  Plot of percentage of population cleaved versus selection round (G1 through G14), where G1 designates the population resulting from amplification of cleaved species from the first round of re-selection, and G14 designates the DNA population resulting from 14 rounds of selective amplification. NaCl concentrations contained in selection buffers for [NaCl] (mM) 200 100 50 100[NaCl] (mM) 200 100 50 100AB71  different rounds of selection were labeled underneath the graph. Mutagenic PCR generated library was used in round 5, 8, 10 and 13. The amplicons of cleaved population within 5 min cleavage reaction from selection round 14 (family I) and round 12 and 14 (family II) were subjected to cloning, as indicated by arrows.  2.6.4 Analysis of resulting sequences form re-selection Family I Alignment of the cloned sequences from Family I revealed that not a single one conforming to the 5’-conserved region of Dz7-45, the wild-type sequence of re-selection lineage II (Family II). Initial kinetic analysis of 60 sequences was performed by fitting the kinetic cleavage data to a single exponential first-order reaction (Equation 2.1) revealed 10 cis-cleavers with improved kobs, ranging from ~0.08 to ~0.53 min-1, corresponding to an improvement of 4- to 25-fold compared to the wild-type Dz7-38. Generally, all sequences were three or four bases longer than wild-type Dz7-38 caused by insertions located at the 3’-end of the catalytic region, which were probably due to polymerase errors that were induced by the mutagenic PCR conditions. Additionally, these sequences contained 14 to 17 mutations compared with the wild-type Dz7-38, most of which were also scattered close to the 3’-end of the catalytic region. Among these sequences, Dz7-38-32, Dz7-38-39, Dz7-38-83 and Dz7-38-90 shared high sequence homology, with less than two bases in difference from one another. However, apart from these, no obvious sequence consensus was observed for the other 6 clones that showed good activity. Not surprisingly, Dz7-38-32, Dz7-38-39, Dz7-38-83 and Dz7-38-90 displayed higher catalytic rates relative to others as well (Table 2.5).   72  Table 2.5 Sequence alignment of the initial 10 clones with improved catalytic rates (kobs) compared with wild-type Dz7-38   Sequence of the wild-type Dz7-38 is shown on the top. Clone names are shown at the left, with 7-38 denoting the wild-type sequence, and the last figures representing respective clone numbers. The kobs values shown in the last column were calculated by fitting the data obtained from initial kinetic analysis to a single exponential first-order equation (Equation 2.1). The initial kinetic analysis was carried out in 50 mM sodium cacodylate buffer (pH 7.45) containing 200 mM NaCl, 1 mM EDTA.  Dz7-38-32 was chosen for further systematic characterization of catalytic parameters in cis and in trans, as well as for RNase A-mimicking mechanism probing. All these contents are covered in the following Chapter 3.  Clone (Dz)kobs (min-1)W.T T T A C A G T G G T A G C G G T T G G C A T G T G T G C A G C G T A G G T G G G C G - - - -0.027-38-43 . . . . . . . . . . . . . . . - . . . T . . . C A . . . . . . . A G A . G T C . T . G T T T0.087-38-9 . . . G . . . . . . . . . . . C . C . T . . . . . . . . G . G . . T . . G T C C T . C T T G0.097-38-21 . . . . . . . . . . . . . . . . A . . G . . . . . . . T C C . . G T . . G C T T T . C C T G0.137-38-38 . . . T . . . C . . T . . . . . A . . . . . . . . G . T . . . . . C A . G T T . T T C C C G0.187-38-80 . . . T . . . . . . . . . . . A . . . A . G . G . G . . . . . A . G A . G - T . T . C C C G0.197-38-39 . . . . . . . . . . . T . . A . . . . G . C . . . . . G . . . . . C A . G A T . G T A T A -0.277-38-89 . . . . . . . . . . . . . . . . . T T . . . . C C . . . C . T . . . A . G . T . T . C C T G0.297-38-32 . . . . . . . . . . . T . . A . . . . G . C . . . . . G . . . . . C A . G A T . G T A C A -0.397-38-90 . . . . . . . . . . . T . . A . . . . G . C . . . . . G . . . . . C A . G A T . G . A T C A0.407-38-83 . . . . . . . . . . . T . . A . . . . G . C . . . . . T . . . . . C A . G A T . G . A T C A0.53Sequence (5'-3', catalytic)73  2.6.5 Analysis of resulting sequences form re-selection Family II As aforementioned in 2.6.4, sequences isolated from the re-selection Family I (derived from Dz7-38) only conform to the 5’ consensus of the wild-type Dz7-38. In contrast, re-selection Family II (derived from Dz7-45) gave rise to two families of sequences conforming to the 5’ consensus of either Family I or Family II. In the initial kinetic analysis of 70 different clones, sequences conforming to Family I displayed no activity improvement compared to the ancestral Dz7-38 and Dz7-45, whereas, ten sequences conforming to Family II showed kobs values in cis ranging from ~0.04 to ~0.40 min-1, representing a rate improvement of 4- to 40-fold. These ten active clones were aligned in reference to the wild-type Dz7-45 in Table 2.6 in the order of slow to fast cleavers. Compared to the sequence of wild-type Dz7-45, all of the improved sequences contained more than 10 mutations, an observation that also happened in the reselection for improved cis-cleavers derived from Dz7-38. Among these mutations, one “T” to “G” mutation (highlighted in yellow) was observed for all the 10 active clones obtained by reselection, indicative of its critical role to an improved self-cleavage rate. Besides this common mutation, no explicit conclusion in terms of other mutations that resulted in improved catalytic activity can be drawn without systematic substitution study.187,231  Due to the improved catalytic performance in regards to both catalytic rate and extent of reaction completion among the ten active clones in Table 2.6, Dz7-45-28 was subjected to more detailed kinetic studies in cis together with the best representative (Dz7-38-32) from re-selection Family I, which is demonstrated in Chapter 3.   74  Table 2.6 Top 10 active sequence alignment referring to wild-type Dz7-45 (W.T)   Sequence of the wild-type Dz7-45 is shown on the top. Clone names are shown at the left, with 7-45 denoting the wild-type sequence, and the last figures representing clone numbers. The one common “T” to “G” mutation is highlighted in yellow. The kobs values shown in the last column were calculated by fitting the data obtained from initial kinetic analysis to a single exponential first-order equation (Equation 2.1). The initial kinetic analysis was carried out in 50 mM sodium cacodylate buffer (pH 7.45) containing 200 mM NaCl, 1 mM EDTA.   2.7 Re-selection and Evolution of Degenerated Dz12-91  The success in isolating catalytic sequences with up to a 40-fold improvement in catalytic rates by performing re-selections and mutagenic evolution prompted a question Clone (Dz)kobs (min-1)W.T A A G C A G C G C A T G T G A T G G C A C G C G T A G T G G C T G G G - - T A G C T - - 0.017-45-42 . . . . . . . . G . . . - . . G . . - . . . . . G . . . . C . A . . . T C . G . T C T T 0.047-45-30 . . . . . . T . G . . C G . T G . . - . . . . . G . . . . . G A . . . A T . G C . . T - 0.077-45-19 . . . . . . . . G . C . . . . . . . - . T . . . G . . . . C . A . . . T C . G . . C C T 0.107-45-44 . . . . . . T . G . . . G . . G . . - . . . . . G . . . . C . A . . . A G . G . . C C T 0.137-45-29 . . . . . . T . G . C A G . . G . . - . . . . . G . . . . C . A . . . T C . G . . C T T 0.157-45-6 . . . . . . . . - - . . . . T . A . . G . . . . G . . . T T . C C . C T T A G . T C C C 0.217-45-10 . . . . . . . . G . . . . . G . . T G . . . . . G . . . - - . C C . A G T A G . G C T T 0.227-45-28 . . . . . . . . G . C . - . . . . . - . T . . . G . . . . C . . . A . T C A G . . C T T 0.257-45-22 . . . . . . . . G . . . . . G . . T G . . . . . G . . . - - . C C . . G T A G . G C T T 0.347-45-12 . . . . . . . . - - . . . . . . A . . T . . . . G . . . T T . C C . C T T A G . T C C C 0.40Sequence (5' to 3', catalytic)75  of whether the same methodology can be used to evolve Dz12-91. Dz12-91, which had been selected from a library consisting of 20 randomized positions, attained a cis-cleavage rate of 0.06 min-1 on a 12 nt all-RNA substrate. If DNAzymes with self-cleavage rates as high as those from N40 libraries (up to 0.53 min-1) can be obtained through a re-selection process using a library degenerated from Dz12-91, these relatively small DNAzymes would be more promising for in vivo study. Because of the composition of fewer nucleotides, they would be easier to synthesize on solid phase and also less restricted by issues of administration and in vivo stability.29,99 Meanwhile, by sequence comparison with N40 DNAzymes, insight in terms of the essential catalytic residues may be provided by examining the extent to which highly conserved catalytic motifs are found in newly evolved catalysts targeting the same 17 nt all-RNA substrate.  To generate a partially degenerated library based on the catalytic sequence of Dz12-91, random mutations at a frequency of 15% per nucleoside position over all the 19 positions (A1 to C19 in the top panel of Figure 2.15) that encompassed the catalytic core were made. Meanwhile, the original 12 nt all-RNA substrate was replaced by the 17 nt one based on the same rational aforementioned in 2.3 (Figure 2.15).  76   Figure 2.15 Construct of Dz12-91 for re-selection and evolution.  Bases in red denote the RNA substrates: the one in the construct is the original 12 nt substrate, and the other one on top is the new 17 nt all-RNA substrate. Potential cleavage sites are indicated by arrows. Bases in blue are modified with structures depicted in the bottom panel. Bases in the catalytic core numbered from 1 to 19 were subjected to partial randomization.   2.7.1 Re-selection process of the degenerated Dz12-91  The enrichment process of the re-selection based on the degenerated Dz12-91 sequence was relatively slow as reflected by the low percentage of cleavage in 1 h reaction prior to round 7. In round 7, the selection stringency was increased by shortening reaction time from 1 h to 10 min, yielding 4% for the self-cleavage reaction. In the following round (round 8), the total activity increased to 12% in 10 min reaction  G A   C A  A C CB5'-3'-C U C G A  G C G C C Cr(G C G U G C C C G U C U) G T T G G C G C A  C GU2G3A4U5C7A8C1012C14U16UU18C19GGGA13GGA1r(G C G U G C C C G U C U  G U U G G)A   =NNNNNH2HNNHNOOOU =NHN OONHH2NNH2OOONN ONH2H2NOOOC  =8-histaminyl-deoxyadenosine 5-guanidinoallyl-deoxyuridine5-aminoallyl-deoxycytidine77  time. Then the reaction time was shortened to 5 min starting in round 11 until the final round 13. In addition, NaCl concentration in the cleavage buffer was decreased to 100 mM for round 12 and the final round 13 as well. The fast-to-cleave population that was isolated following a 5 min reaction time collected from round 11 was amplified using Taq DNA polymerase, cloned, sequenced, and screened for activity. The re-selection progress is summarized in Figure 2.16.     Figure 2.16 Re-selection progress of degenerated Dz12-91. Plot of percentage of population self-cleaved versus selection round (G1 through G13). NaCl concentrations contained in selection buffers were labeled underneath the graph. Mutagenic PCR generated library was used in round 6, 9, and 12, denoted G6_m, G9_m, and G12_m. The amplicons of self-cleaved population corresponding to 5 min cleavage reaction from round G11 was subjected to cloning, as indicated by arrow.  [NaCl] 200 mM 100 mM0	2	4	6	8	10	12	14	16	18	G1	 G2	 G3	 G4	 G5	 G6_m	 G7	 G8	 G9_m	 G10	 G11	 G12_m	 G13	Percent cleavage  Generation  Progress of re-selection of degenerated Dz12-91 5 min  10 min  30 min  60 min  78  2.7.2 Alignment and analysis of sequences resulting from re-selection of degenerated Dz12-91  Sequence alignment of the 42 clones resulting from re-selection of degenerated Dz12-91 revealed 34 different sequences, among which clone 11 (Dz12-91-11) was repeated twice, both clone 3 (Dz12-91-3) and clone 5 (Dz12-91-5) were repeated three times, and clone 2 (Dz12-91-2) was repeated four times. Notably, most of the sequences that emerged contained 5’ conserved regions conforming to either Family I (5’-TTAT/CA-3’) or Family II (5’-AAGCA-3’) of the N40 libraries rather than the wild-type Dz12-91, which probably suggests that these sequences are the dominant 5’ motifs required by the cleavage of the 17 nt all-RNA substrate, at least in the context of the current unimolecular construct. The two exceptions, Dz12-91-16 and Dz12-91-22, retained 16 consecutive positions that were found in the sequence of wild-type Dz12-91. Initial kinetic analysis of the 34 clones by fitting the percent of self-cleavage collected at 5 different time points (0, 3, 10, 60, and 180 min) to a single exponential first-order reaction (Equation 2.1) revealed 11 active sequences. Not surprisingly, the highly abundant Dz12-91-3, Dz12-91-5 and Dz12-91-2 sequences were all active, with Dz12-91-5 showing the best activity among all the screened sequences, which reflects the normally strong correlation between the abundance of a particular sequence and its catalytic activity (Table 2.7).      79  Table 2.7 Active sequences from the re-selection of degenerated Dz12-91 aligned in reference to the wild-type Dz12-91 (W.T)   Sequence of the wild-type Dz12-91 is shown on the top. Clone names were shown at the left, with 12-91 denoting the wild-type sequence, and the last figures representing clone numbers after re-selection. The kobs values shown in the last column were calculated by fitting the data obtained from initial kinetic analysis to a single exponential first-order function (Equation 2.1). The initial kinetic analysis was carried out in 50 mM sodium cacodylate buffer (pH 7.45) containing 200 mM NaCl, 1 mM EDTA. a Dz12-91-16 and Dz12-91-22 conform to 16 consecutive positions of wild-type Dz12-91. b Repeating numbers were counted out of 42 obtained sequences.   Clone (Dz) bRepeats kobs (min-1)W.T A T G A T G C A G C G C A T G T G T C - - - - - - - - - - N/A 0.0312-91-11 - - - C A C A . . T . G T A T C . G T A G G C A T G T - - 2 0.004a12-91-16 - - - . . . . . . . . . . . . . . . . - - G C A C G C T T 1 0.00612-91-25 - - - T . A . . . T . G T A . C . G T - G G C A C G T G T 1 0.00612-91-1 - - - T . A T . . T . G T A . C . G A - G G C A T G T A T 1 0.00712-91-26 - - - T . A T . . T . G T A . C . G T - G G C A C G C T - 1 0.0112-91-13 - - - T . A . . . T . G T A . C . G A - G G C A T G T A T 1 0.014a12-91-22 - - - . . . . . . . . . . . . . . . . - - G C A C G C A T 1 0.01612-91-3 - - - T . A T . . T . G T A . C . G A - G G C A T G T G T 3 0.01612-91-2 - - - . A . . . . . . . . . . . . . . - - G C A C G C A T 4 0.01712-91-4 - - - T . A T . . T C G T . . C . G T - G G C A C G C T - 1 0.0212-91-5 - - - . C A G C T . . A G . C C . . T T G C G G C A T - - 3 0.034Sequence (5' to 3', catalytic )80  With the intramolecular 17 nt all-RNA substrate, Dz12-91-5 displayed the highest self-cleaving rate constant of ~0.034 min-1 (Figure 2.17), a rate constant just comparable to that displayed by the wild-type Dz12-91, ~0.03 min-1. The re-selection used library derived from degenerated Dz12-91 toward isolating DNAzymes with self-cleavage rates comparable to the N40 clones was not achieved. Although the possibility of missing clones with better catalytic activity in the step of single-clone sequencing cannot be readily ruled out, it is more likely that the N20 library is intrinsically deficient in providing modified DNAzymes with as high catalytic rates as those from N40 libraries. A similar difference in catalytic rate has been observed for Dz9-86 and Dz10-66 previously, the best representatives targeting the same DNA/RNA chimeric substrate containing one embedded RNA base, but isolated from libraries of N20 and N40, respectively (Table 2.1). Dz9-86 only attained a kobs of ~0.13 min-1 in cis, and it was turnover-incapable,185 whereas, Dz10-66 attained a kobs value of ~0.57 min-1 in cis under the same cleavage conditions (50 mM sodium cacodylate, pH 7.45, 200 mM NaCl, 1 mM EDTA), and it was capable of multiple-turnover.186      81   Figure 2.17 Cis-cleavage reaction of Dz12-91-5.  (A) Gel image (7% PAGE) showing cis-cleavage time course. The uncleaved products are indicated by a filled arrowhead, while the cleavage products are indicated by an open arrowhead. (B) Plot of cis-cleavage percentage versus incubation time after fitting to a single exponential first-order equation (Equation 2.1); kobs = 0.034 min-1 (R2=0.994). Values for curve fitting were derived from the PAGE band intensities in part (A). The data fitting and curve generation were carried out using Prism 6.  2.8 Discussions and Conclusion  The practical difficulty in transforming RNase A-mimicking DNAzymes targeting a DNA/RNA chimeric substrate into equally efficient all-RNA cleaving DNAzymes prompted the optimization of a previous construct containing a partially all-RNA substrate (12 nt all-RNA) (Figure 2.3A) into the one containing a 17 nt all-RNA substrate (Figure 2.3B). Meanwhile, constraints imposed by the in vitro selection process and by the simultaneous use of three modified nucleotides emphasized the design of a more comprehensive in vitro selection cycle. Firstly, the simultaneous use of -30 0 30 60 90 120 150 180 210153045607590Time (min)Percent cleavage Dz12-91-5Time (min)0 10 603 180uncleavedcleavedA B82  three chemically modified nucleotides unavoidably diminishes the sequence space available for enzymatic synthesis of the modified library. Secondly, PCR amplification of multiple natural oligonucleotide templates can favor certain sequences over others.232,233 Consequently, good catalysts that fold into more stable secondary structures for enhanced function would suppress their ability to act competently as templates in PCR, which eventually leads to a significant accumulation of undesired artifacts during a 25-30 cycle PCR process. Thirdly, modified sequences afford templates that are not readily read through by polymerases, which always causes “unidirectional loss” of modifications.221,222 Strategies including optimization of primer extension conditions for transcribing modified library at a lower temperature (32ºC) for a longer period of time (4 h), use of a re-selection process by partially degenerating the best acting clones out of selection, and implementation of mutagenic PCR to introduce round-specific mutations seemed to amend these drawbacks at least in terms of identifying catalysts with demonstrably improved activity.  DNAzymes displaying kobs values as high as 0.53 min-1 (Dz7-38-83) and 0.40 min-1 (Dz7-45-12) in initial kinetic assays of self-cleavage were selected from N40 libraries. To the best of my knowledge, when assayed under absolutely M2+-free conditions, they attained the best-ever all-RNA cleavage rates among all reported RNA-cleaving DNAzymes, including both unmodified29,30,135,136 and modified ones179,187. Their discovery demonstrated the efficiency of the new in vitro selection cycle which integrates re-selection and evolution to provide sufficient sequence space for the exploration of efficiently all-RNA cleaving DNAzymes that had not been isolated before. 83  The same selection cycle has been generalized to another selection targeting an all-RNA substrate derived from c-MYC mRNA in the Perrin lab, and led to the isolation of a family of highly efficient DNAzymes. More importantly, the isolation of these highly efficient DNAzymes provides candidates for the long-term goal of in vivo mRNA disruption using DNAzymes. Systematic characterization of these new DNAzymes is covered in Chapter 3. It is hard to clarify the reasons for failing to evolve DNAzymes with improved activity over that of Dz12-91, by re-selecting new DNAzymes using a library generated by partially degenerating the catalytic core of Dz12-91. In vitro selection is a long and laborious process in which many different factors can affect the final outcome. In this case, the presumed catalytic deficiency of DNAzymes re-selected starting with Dz12-91 is probably related to the random region length. A shorter random region allows more coverage of sequence space, as reflected by the appearance of repeating sequences after re-selection of degenerated Dz12-91. However, it may not permit the structural complexity necessary for catalysis. This might be especially true in the pursuit of M2+-independent DNAzymes, which have to make use of a sufficient number of appended side-chains to supplement the function of M2+-cofactors. In contrast, a longer random region cannot be sampled completely in in vitro selection but allows greater possibility to adopt larger and thus more complex structural features that enable more efficient catalytic function. Mutagenic PCR, intended to explore more sequence diversity, helped to some extent. Such has been observed in unmodified selections as well; for example Silverman et al. carried out a systematic examination of effects of random region length 84  on DNA cleavage and tyrosine-RNA nucleopeptide linkage formation reactions.234 In this study, they found that robust DNA cleavers were isolated from N20 and N30 libraries, but not from longer N50 or N60 ones, whereas, an N20 library was unsuccessful in providing sequences capable of catalyzing Tyr-RNA linkage formation.234 Analogous interplays between optimal random region length and particular types of binding or catalysis have been reported by a number of other groups in selections for RNA aptamers235 and ribozymes236,237 as well. Collectively, random region length should be an important experimental variable in designing in vitro selections of functional nucleic acids. In conclusion, a new in vitro selection construct enabled improved catalytic activity against a longer all-RNA substrate by virtue of appending a large dispensable loop that effectively separated the substrate region and catalytic domain to provide a construct enabled efficient read-out during the PCR step. In addition, a more efficient selection cycle with a comprehensive consideration of efficient transcription and enlarged sequence space for selection using chemically modified nucleotides was designed and verified. The simultaneous use of a transcription step at 32ºC for 4 h, a re-selection process, and a moderate mutagenic PCR protocol for introducing “real-time” sequence variability has led to the development of two families of efficient all-RNA cleavers, as represented by Dz7-38-32 and Dz7-45-28.   85  2.9 Experimental 2.9.1 General reagents, materials, and enzymes  dAimTP  and dUgaTP were synthesized according to literatures185,214 by Dr. Curtis Lam. dCaaTP was obtained from TriLink Bio Technologies. Oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA), and purified by denaturing PAGE (7 M). Vent (exo-) DNA polymerase, Taq DNA polymerase, lambda exonuclease, T4 DNA ligase, and T4 polynucleotide kinase (T4 PNK) were obtained from New England Biolabs. Sequenase v2.0 and pyrophosphatase were purchased from Affymetrix. Steptavidin magnetic particles were purchased from Roche. α-32P-dGTP and γ-32P-ATP were obtained from Perkin Elmer. pCR2.1-TOPO vector system kit and PureLink Quick Plasmid Miniprep Kit were purchased from Invitrogen. Sephadex G25 resin was obtained from GE.  2.9.2 General protocols  The protocols described here were generally followed by all related experiments throughout this thesis, unless otherwise mentioned.   Denaturing PAGE gels. Polyacrylamide gels were made from 40% acrylamide/bis-acrylamide (29:1) stock solutions. The gels were denatured by the presence of 7 M urea in final 1X TBE solutions. TBE solutions of 1X or 0.5X were also used as electrolytic solutions. Gels used in the Perrin lab are always made for polymerization overnight. Gels of two different sizes were used for analyzing or purification, denoted small gel (170 x 165 x 1 mm, L x W x D) and big gel (42 x 33 x 0.04 cm, L x W x D), respectively. 86  All oligonucleotides were heated to 90ºC for 5 min followed by snap-cool down to 0ºC prior to loading. Gels were either visualized by autoradiography on a Typhoon 9200 phosphorimager from GE healthcare or by UV shadowing.  Gel elution. Desired gel band was sliced off with surgical blade, and smashed into fine particles using a flame-sealed pipette tip. Then to the gel particles, gel elution buffer was added, agitated on a vortexer, and incubated at 65ºC for 15 min. The mixture was centrifuged, and the supernatant was recovered. This elution process was repeated two more times to collect the supernatant. The combined supernatant was evaporated by speed-vac, and followed by ethanol precipitation. The resulting pellet was re-dissolved in H2O and run through G25 spin column to desalt. The final concentration of eluted oligonucleotides was measured on a Beckman Coulter DU800 spectrophotometer.   G-25 spin column. Columns were prepared freshly right before purification usage. Sterile G-25 Sephadex was added into 1 ml glass-wool plugged pipette tips, and washed three times with 400 µL of DEPC-treated sterile water to wash away preservative NaN3. Normally, columns were packed to ~3 to 5 cm in length, depending on the quantity of oligonucleotides purified. 2.9.3 Optimizing conditions for more efficient primer extension reactions to address the production of modified DNA Buffers/solutions. 1X T4 polynucleotide kinase buffer: 70 mM Tris-HCl (pH 7.6 @ 25 °C), 10 mM MgCl2, 5 mM DTT. 87  1X Sequenase buffer: 40 mM Tris-HCl (pH 8.0), 50 mM NaCl, 20 mM MgCl2.  Gel loading buffer: Formamide, 25 mM EDTA, 0.01% bromophenol blue and 0.01% xylene cyanole. Gel elution buffer: 10 mM Tris-HCl, (pH 8.0), 1% LiClO4. 1X TBE solution: 89 mM Tris base, 89 mM Boric acid, 2 mM EDTA.  Oligonucleotides (ON, 5’ to 3’). Primer for template-directed primer extension, ON1: CGTCTGTTGGGCCCTACCA.  Dz10-66 derived template for template-directed primer extension, ON 2:  TTTTTGCGTGCATACACGCACGCACACTCATAGCGCGCCTCACTTGCGCCGCGTTGCTAGTGTTGGTAGGGCCCAACAGACGGGCACGCTCGTGT3. Labeling of ON 1 at the 5’ end by T4 polynucleotide kinase. 5 µL of ON 1 stock solution (1 mM) was mixed with 5 µL of 10X T4 polynucleotide kinase buffer, 3 µL of γ-32P-ATP (~ 30 µCi), and 5 µL of T4 polynucleotide kinase (units), then 32 µL of H2O was added to make a final reaction volume of 50 µL. The reaction was incubated at 37ºC for 3 h and then terminated by heating at 65ºC for 20 min. After adding into 50 µL of gel loading buffer, the resulting 100 µL of solution was equally loaded to 2 wells of a 20% small denaturing PAGE. Product was visualized by UV-shadowing after electrophoresis, and gel slice containing the labeled ON 1 was cut out, eluted, and subjected to G-25 column desalting before measuring concentration.   Template-directed primer extension. The template-directed primer extension 88  reactions were run in a final volume of 40 µL. Equal amount (pmole) of gel purified template (ON 2) and primer (5’-32P-ON 1) were heated to above 95ºC for 5 min in the presence of 8 µL of 5X Sequenase buffer, and then cooled down to room temperature slowly. By doing so the two pieces of oligonucleotides were annealed together to form template-primer duplex. Then 8 µL of 5X dNTP mixture containing 250 µM dAimTP, and 50 µM of each dTTP, dCTP, and dGTP, 2 µL of 100 mM DTT, 1 µL of inorganic pyrophosphatase, and 1 µL of Sequenase v2.0 were added to the annealed reaction sequentially. H2O was added to bring the reaction volume up to 40 µL. Reactions were initiated by incubating in thermocycler at different temperatures. For the reaction cycled between 32ºC and 42ºC, each reaction cycle included a 15 min at 32ºC and a 15 min at 42ºC. For all the tested reactions, 10 µL of reactions were removed into 10 µL of gel loading buffer at 0.5 h, 1 h, 2 h, and 3 h of reaction times, respectively. The 10 µL control reaction contained every reagent at the same concentrations, omitting the template, was subjected to the same treatments, and incubated at 32ºC for 3 h before loading to the gel. A 24-well big gel (10%) was used for analyzing all the samples.  2.9.4 In vitro selection Buffers/solutions. Wash buffer (TEN): 50 mM Tris-HCl (pH 7.5), 200 mM NaCl, 1 mM EDTA.  Neutralization buffer: 25 mM sodium cacodylate (pH 6.0), 1 mM EDTA. Cleavage buffer: 50 mM sodium cacodylate (pH 7.4), 200 mM NaCl, 1 mM EDTA. Gel elution buffer: 10 mM Tris-HCl (pH 8.0), 1% LiClO4. 89  Gel loading buffer: Formamide, 25 mM EDTA, 0.01% bromophenol blue and 0.01% xylene cyanole. Gel loading buffer containing biotin: gel loading buffer/100 mM biotin in DMF (99:1). 5X modified nucleotides cocktail:  250 µM dAimTP, 125 µM dCaaTP, 50 µM dUgaTP, 50 µM dGTP. 5X first amplification cocktail (each 230 µL): P2 (9 nmol), P3 (7.5 nmol), dNTPs (345 nmol each), MgSO4 (1.15 µmol), 10X Thermopol buffer (115 µL), and H2O for a final volume of 230 µL.  5X second amplification cocktail (each 230 µL):  P4 (10 nmol), P3 (7.5 nmol), dNTPs (345 nmol each), MgSO4 (1.15 µmol), 10X Thermopol buffer (115 µL), and H2O for a final volume of 230 µL.  10X mutagenic PCR buffer: 100 mM Tris-HCl (pH 8.3), 70 mM MgCl2, 500 mM KCl, 0.1% (wt/vol) gelatin.  10X dNTP mixture for mutagenic PCR:  2 mM dGTP, 2 mM dATP, 10 mM dCTP, and 10 mM dTTP. 5 mM MnCl2. 5X TOPO cloning PCR cocktail:  P2 (9 nmol), TOPO-P3 (7.5 nmol), dNTPs (345 nmol each), MgSO4 (1.15 µmol), 10X 90  Thermopol buffer (115 µL), and H2O for a final volume of 230 µL.  6X agarose gel loading buffer:  20 mM Tris-HCl (pH 8.0), 15% Ficoll®-400, 66 mM EDTA, 0.1% SDS, 0.09% bromophenol blue.  Oligonucleotides (5’ to 3’, ON).  Selection primer fragment 1, ON 3: Biotin-T40CCC.  Selection primer fragment 2.1, ON 4:  Phosphate-GGGTTTTTGCGTGCCr(CGU)CTGTTGGTTTTGCGTCGGCCTGC GCCAACAG. Selection primer fragment 2.2, ON 5:  Phosphate-GGGTTTTTr(GCGUGCCCGUCUGUUGG)TTTTGCGTCGGCCTGC GCCAACAG. Splint for ligation, ON 6: AAAAACCCGGGAAAAA. N40 library template, ON 7:  GCGCTCGCGCGGCGTGCN40CTGTTGGCGCAGGCCGACGC.  PCR primers:  91  P2 (ON 8), GCGCTCGCGCGGCGTGC. P3 (ON 9), phosphate-GCGTCGGCCTGCGCCAACAG.  P4 (ON 10), TTTTTTTTTTTTTTTTTTTTGCGCTCGCGCGGCGTGC.  P5 (ON 11), phosphate-TTTTGC GTCGGCCTGCGCCAACAG.  TOPO-cloning primer:  TOPO-P3 (ON 12), GCGTCGGCCTGCGCCAACAG.  7-38 degenerated library (ON 13): GCGCTCGCGCGGCGTGC C(N2:05850505)G(N3:05058505)(N2)(N2) (N2)A(N1:85050505)(N2)(N2)T(N4:05050585)(N1)(N2)(N3)(N2)(N4)(N3) (N2) (N1)(N2)(N1)(N2)(N1)(N4)(N3)(N2) (N2)(N1)(N1) (N2)(N2)(N3)(N2)(N4)(N1) (N2)(N2)(N1)(N2)(N4)(N3)(N4)(N1)(N1) CTGTTGGCGCAGGCC GACGC. 7-45 degenerated library (ON 14):  GCGCTCGCGCGGCGTGC A(N1:85050505)G(N3:05058505)C(N2:05850505)T (N4:05050585)(N1)(N2)(N2)(N2)(N1)(N3)(N2)(N2)(N1)(N2)(N4)(N1)(N2)(N3)(N2) (N3)(N4)(N3)(N2)(N2)(N1)(N4)(N2)(N1)(N2)(N1)(N4)(N3)(N2) (N3)(N2)(N4) (N3)(N2)(N4) (N4) CTGTTGGCGCAGGCCGACGC. 12-91 degenerated library (ON 15):  GCGCTCGCGCGGCGTGC G(N3:05058505)A(N1:85050505)C(N2:05850505) 92  (N1)(N2)(N1)T(N4:05050585)(N3)(N2)(N3)(N2)(N4)(N3)(N2)(N1)(N4)(N2)(N1) (N4) CTGTTGGCGCAGGCCGACGC. “r” designates a stretch of RNA bases in oligonucleotide 4 and 5. “N40” in oligonucleotide 7 represents 40 randomized positions. In ON 13, 14, and 15, N1, N2, N3, and N4 represent partially degenerated A, C, G, and T, respectively. They were defined with randomization ratios once following the first appearances in the randomized regions (underlined).  Ligation of selection primer fragment 1 and selection primer fragment 2.1/selection primer fragment 2.2. The ligation reaction was carried out in a final 60 µL of 1X DNA ligase buffer conditions (50 mM Tris-HCl (pH 7.5), 10 mM MgCl2, 1 mM ATP, 10 mM DTT). 3 nmol of each ON 3 and ON 4/ON 5, and 6 nmol (2 eq.) of splint DNA (ON 6) were annealed together in the presence of 6 µL of 10X DNA ligase buffer by heating up to 95ºC for 5 min and then gradually cooling down to room temperature. Subsequently, 2 µL (400,000 units/ml) of T4 DNA ligase was added to the annealing mixture and incubated at 14ºC overnight. The ligation reaction was then extracted by equal volume of phenol-chloroform-isoamyl alcohol, and followed by ethanol precipitation before dissolving on a 10% small purification gel. Gel slice containing the ligation product was sliced out for gel elution, precipitation, and G-25 desalting before applying in in vitro selection.  Template-directed primer extension for transcribing modified library for selection. To make 15 pmole of modified library used for each round of selection and re-selection, 15 pmole of each template and ligated selection primer, 8 µL of 5X Sequenase buffer, 93  and H2O were combined to a volume of 26.5 µL, and heated to above 95°C for 5 min. The solution was then left at room temperature to slowly cool down to room temperature. Following that, reagents including 8 µL of 5X modified dNTP cocktail (250 µM dAimTP 125 µM dCaaTP, 50 µM dUgaTP, and 50 µM dGTP), 2 µL of DTT (100 mM), 1 µL of pyrophosphatase, 1.5 µL of α-32P-dGTP (~15 µCi), and 1 µL of Sequenase v2.0 were added sequentially to the annealing solution. The reaction was then mixed well and covered with 20 µL of mineral oil. The polymerization reaction was run at 32°C for 4 h and then quenched by the addition of 2 µL of EDTA (0.5 M, pH 8.0) to a final concentration of ~ 25 mM.  Selection. 50 µL of streptavidin beads was magnetized, decanted, and subjected to three washes with TEN buffer (100 µL per wash). The primer extension product was incubated with the washed beads for 15 min at room temperature to immobilize the DNA duplex on beads. Following two more washes with 100 µL of TEN after immobilization, the template strand was stripped away by five quick washes (no longer than 30 seconds per wash) with 100 µL of 0.1 M NaOH containing 1 mM EDTA. The resulting modified strand sticking on streptavidin beads was immediately neutralized using 200 µL of neutralization buffer, followed by a final 100 µL of DEPC-treated sterile water wash. The modified DNA library on beads was then allowed to fold and cleave in 100 µL of cleavage buffers of different ionic strength for varying time spans with the progress of in vitro selection.  Selection workup and gel purification. After controlled reaction time, the 100 µL of cleavage reaction was magnetized, of which only 90 µL of the supernatant was 94  recovered to minimize the contamination caused by the agitation of magnetic beads. To the recovered supernatant, 2 µL of 5X 1st amplification cocktail and 800 µL of 1% LiClO4 in acetone were added. The resulting solution was thoroughly mixed with a vortexer, centrifuged for 15 min and decanted. Then the transparent DNA pellet was washed with 1 ml of ethanol by agitated with a vortexer and centrifuged for 15 min and decanted as well. The resulting DNA pellet was fully dried on 65°C heat-block for 5 min. 20 µL of gel loading solution was used to dissolve the pellet for purification by denaturing PAGE (7%).  The remaining magnetic beads after selective self-cleavage reactions were suspended in 20 µL of water. This sample was divided into two aliquots as positive and negative controls by removing 10 µL of beads in suspension into another 1.5 ml eppendorf tube. To make the negative control, 10 µL of gel loading buffer containing 1 mM biotin was directly added into one tube for subsequent gel loading. The positive control was made from treating the other tube of beads (10 µL) with 2 µL of 1 M NaOH at 65 °C for 20 min prior to the addition of gel loading solution containing 1 mM biotin.  The controls and selection samples were heated at 95°C for 5 min and then snap-cooled to 0°C on ice-block. The selection samples were loaded onto denaturing PAGE (7%) directly. The controls containing magnetic beads were magnetized and only the supernatants were loaded onto the purification gel. For the selections with the three-RNA substrate, gel pieces/bands that lined up with the positive control were cut out. For the selections with the all-RNA substrate, gel pieces/bands migrated several-base higher than the positive control were cut out. After gel elution, the resulting solutions 95  were evaporated to ~100 µL by speed-vac. To each of the mixtures, 2 µL of 1st amplification cocktail and 1 ml of ethanol were added, agitated using a vortexer, and centrifuged for 15 min to precipitate. The DNA pellet left behind after decanting of solvent was further washed with another 1 ml of ethanol by repeating the centrifuge and decanting step. The dried DNA pellet was re-suspended in 60 µL of water.  First PCR amplification. The first PCR amplification was carried out in a 40 µL of reaction containing 25 µL of selection sample, 8 µL of 1st amplification cocktail, 1 µL of α-32P-dGTP (~10 µCi), 2 µL of Vent (exo-) DNA polymerase, and supplemental volume of H2O. The positive control was made from replacing the selection sample with 2 µL (2 pmol) of corresponding original library and H2O. Each sample was equally split into two PCR tubes and thermocycled for 25 cycles:  3 min at 95°C, 25X (15 seconds at 95°C, 15 seconds at 58°C and 40 seconds at 75°C), and 5 min at 72°C.  Each pair of samples was recombined and extracted with equal volume of phenol-chloroform-isoamyl alcohol (40 µL). To the recovered top aqueous layer containing DNA amplicons, 400 µL of ethanol was added, and the resulting mixture was agitated on a vortexer, centrifuged for 15 min, decanted, washed again with 400 µL of ethanol and dried on a heat block at 65°C for 5 min. The resulting DNA pellet was dissolved in 35 µL of water. This DNA solution was subjected to λ-exonuclease digestion by adding into 4 µL of 10X λ-exonuclease buffer and 1µL (5 units) of λ-exonuclease and incubated at 37°C for 3 h. The digested sample was subjected to the same workup procedures as the products obtained from the 1st amplification step. To each dried DNA sample, 20 µL 96  of H2O and 2 µL of NaOH (1 M) were added and agitated to mix prior to the addition of 20 µL of gel loading buffer. All samples were resolved on a big denaturing PAGE (10%). Once again, radioactive bands that lined up with the positive control were cut out, and run through steps of gel elution, and ethanol precipitation. The resulting DNA pellet was dissolved in 20 µL of H2O. Second PCR amplification. The 2nd PCR amplification was performed in a total of 200 µL reaction. Following a quick spin-down, 10 µL of the 20 µL 1st-PCR amplification product was added into 40 µL of 5X 2nd amplification cocktail, followed by the addition of 140 µL of H2O, and 10 µL (20 units) of vent (exo-) DNA polymerase. The mixture was split into five PCR vials (40 µL for each vial). A control reaction was performed in 40 µL of reaction containing 1 µL (2 pmol) of the original template in replacement of the 1st amplification product, 8 µL of 5X 2nd- PCR amplification cocktail, 2 µL (4 units) of vent (exo-) DNA polymerase and 29 µL of H2O. All the sample tubes were thermocycled for 30 cycles:  3 min at 95°C, 30X (15 seconds at 95°C, 15 seconds at 58°C and 40 seconds at 75°C), and 5 min at 72°C.  The samples were then recombined and subjected to phenol-chloroform-isoamyl alcohol extraction and ethanol precipitation steps as described in the “First PCR amplification”. The resulting sample DNA and control DNA pellets were dissolved in 70 µL and 35 µL of H2O, respectively. To the sample DNA tubes, were 8 µL of 10X λ-exonuclease buffer, and 2 µL (10 units) of λ-exonuclease added. To the control DNA tube, 4 µL of 10X λ-exonuclease buffer, and 1 µL (5 units) of λ-exonuclease were 97  added. The mixture of sample DNA was equally split into 2 vials. All DNAs (samples and control) were incubated at 37°C for 3 h. Following that, sample DNAs were recombined, extracted with equal volume of phenol-chloroform-isoamyl alcohol, precipitated with ethanol, and re-suspended in 18 µL of H2O. To each tube, 2 µL of 1M NaOH was added prior to the addition of 20 µL of gel loading buffer. All the samples were then purified on a small PAGE gel  (10%). Bands were visualized by UV-shadowing, and those lined up with control band were cut out, subjected to gel elution, and ethanol precipitation. The resulting DNA samples were re-suspended in 50 µL of H2O, and run through G-25 spin columns to desalt. 5 µL of the each column eluent was then diluted into 300 µL with H2O for UV-Vis quantification on a DU800 Spectrometer.  Mutagenic PCR. Mutagenic PCR was carried out in 100 µL of reaction. 10 µL of 10X mutagenic PCR buffer, 10 µL of 10X dNTP cocktail, 0.87 nmole of primer P4, 0.65 nmole of primer P5, 10 µL of the 1st PCR amplification product, and an amount of H2O were combined to make a volume of 80 µL. Then 10 µL of 5 mM MnCl2 was added. Lastly, 10 µL (50 units) of Taq DNA polymerase was added to make the final volume of 100 µL. This reaction was split into 50 µL aliquots to run 30 cycles of PCR:  3 min at 95°C, 30X (15 seconds at 95°C, 15 seconds at 58°C and 40 seconds at 75°C), and 5 min at 72°C.  The samples were then recombined and subjected to phenol-chloroform-isoamyl alcohol extraction, ethanol precipitation. The resulting DNA pellet was re-suspended in 35 µL of H2O. The subsequent λ-exonuclease digestion reaction was carried out in 40 µL of reaction supplemented by 4 µL of 10X λ-exonuclease buffer and 1 µL (5 units) of 98  λ-exonuclease at 37°C for 3 h. Following the same workup procedures described in “Second PCR amplification”, the mutagenic PCR product was purified on a small PAGE (10%) as well. The resulting DNA pellet was dissolved in 30 µL of H2O and desalted using G-25 column. 5 µL of each eluent was diluted to 300 µL and quantified for concentration on a DU800 UV-Vis Spectrometer.  TOPO Cloning. 40 µL of 5X TOPO cloning PCR cocktail was combined with 10 µL of the 1st-PCR amplification product resulting from the generation proceeding to cloning, 140 µL of H2O, and 10 µL of Taq DNA polymerase to make a final reaction volume of 200 µL. This mixture was then split into 5 vials of 40 µL aliquots and run through thermocycles containing a final 30 min extension step at 72°C to produce amplicons bearing 3’-A overhangs required for TOPO cloning:  3 min at 95°C, 30X (15 seconds at 95°C, 15 seconds at 58°C and 40 seconds at 75°C), and 30 min at 72°C. The PCR products were recombined, extracted by equal volume of phenol-chloroform-isoamyl alcohol, precipitated with ethanol, and dissolved in 25 µL of H2O. The re-suspended sample was loaded to a 2% purification agarose gel following the addition of 5 µL of 6X agarose gel loading buffer. Gel slice containing the amplicons was purified using GeneJet Gel Extraction Kit according to the manufacture’s instruction. The purified amplicons were TOPO cloned into the pCR2.1-TOPO vectors. These vectors were then used to transform E.coli DH5α competent cells following a normal chemical transformation protocol. White transformant colonies were picked by Blue-White screening on a LB Agar plate containing 100 mg/L ampicillin for 3 ml of inoculation in 99  liquid LB medium. Plasmids were prepared by using PureLink Quick Plasmid Miniprep Kit (Invitrogen), and sequenced by the Nucleic Acid Protein Service Unit of UBC (NAPS-UBC) using the M13R sequencing primer.  Initial kinetic analysis. The catalyst was prepared following the same procedures for transcribing the modified library in in vitro selection as described in “Template-directed primer extension for transcribing modified library for selection”. Instead of running 15 pmole of template directed primer extension reaction, 6 pmole of each synthetic template, 6 pmole ligated primer (containing the all-RNA substrate sequence), and 3.2 µL of 5X Sequenase buffer, and an amount of H2O were mixed to a volume of 10.6 µL. This mixture was heated to 95°C for 5 min followed by slowly cooling down to room temperature for annealing. To the annealing mixture, 3.2 µL of 5X modified nucleotides, 0.8 µL of DTT (100 mM), 0.4 µL of pyrophosphatase, 0.5 µL of α-32P-dGTP (~5 µCi), and 0.4 µL of Sequenase v2.0 were added sequentially to make a final reaction volume of 16 µL. The reaction was run for 4 h at 32°C with mineral oil on the top to minimize the effect of condensation on modified strand synthesis.  The extension product was immobilized on 40 µL of pre-washed (100 µL of TEN buffer per wash) streptavidin coated magnetic beads by incubating at room temperature for 15 min. Following two more washes with 100 µL of TEN, the template strand was stripped away by 5 quick washes with 100 µL of 0.1 M NaOH containing 1 mM EDTA. The resulting modified strand sticking to streptavidin beads was immediately neutralized using 200 µL of neutralization buffer followed by a final 100 µL of water wash. Then the modified DNA on beads was incubated in 100 µL of standard cleavage buffer (50 mM 100  cacodylate, pH 7.45, 200 mM NaCl, 1 mM EDTA), which formed a mixture of bead-bound uncleaved materials and cleaved products. 5 µL of the mixture was removed and quenched in 15 µL of gel loading buffer containing 1 mM biotin at time points 0 min, 3 min, 10 min, 60 min, and 180 min, respectively. All samples were denatured by incubating at 95°C for 5 min, followed by a snap-cooling step on ice before resolving the supernatant after magnetization on a denaturing PAGE (7%). Visualizations and quantifications were carried out by using Typhoon 9200 PhosphorImager (Amersham). The autoradiographic data of the cleavage reactions (represented by pixel volumes on the Imagequant software) were obtained by drawing polygons around the bands corresponding to the cleaved and uncleaved species. These data were fitted to single exponential first-order reaction with Prism 6 using Equation 2.1: P! = P! 1− 𝑒!!!"#! , where Pt is the total fraction cleaved at time point t, P∞ is the fraction underwent cleavage reaction at the end time point, and kobs is the first order rate constant.  101  Chapter 3: Characterization of Dz7-38-32, A Representative RNase A-mimicking DNAzyme  3.1 Introduction  The previous work in Chapter 2 led to the isolation of two DNAzyme families capable of efficient all-RNA cleavage, with kobs values as high as 0.53 min-1 (Dz7-38-83) and 0.40 min-1 (Dz7-45-12) based on initial kinetic assays of self-cleavage. In order to evaluate the potential capacity of these DNAzymes as therapeutic agents, this chapter focuses on: (1) characterization of enzymatic parameters of representative DNAzymes in terms of cis-cleavage (i.e. self-cleavage) and in trans cleavage (i.e. intermolecular cleavage under either single-turnover or multiple-turnover), (2) elucidation of substrate sequence specificity-generality238 and cleavage site preference, and (3) examination of the effects of temperature, pH, and divalent metal ions, factors that have a significant impact on the catalytic efficiency of DNAzymes29,116.  Fully characterizing and understanding these parameters will 1) bring a greater understanding of the potential therapeutic applications of these DNAzymes, 2) lead to a better understanding of the general properties of modified catalytic DNA, and 3) contribute more insight into this particular class of DNAzymes bearing RNase A-mimicking functionalities. Additionally, the possible catalytic strategies employed by Dz7-38-32 will be discussed, thereby providing an overall evaluation of these unique species that bridge the gap between proteins and nucleic acids (Figure 3.1). Lastly, the 102  catalytic performance of the best cleaver will validate efficiency of the optimized in vitro selection system demonstrated in Chapter 2, and provide insight into possible shortcomings in these catalysts which can be considered in the future for designing appropriate strategies in in vitro selection towards the development of a broader range of novel catalysts or aptamers using modified nucleotides.   Figure 3.1 Hypothetical cleavage mechanism of chemically modified RNase A-mimicking DNAzymes.   3.2 Characterization of Cleavage Activity In Cis  3.2.1 Cis-cleavage kinetics  Based on the sequence alignment of active sequences and the activities they displayed in initial kinetic analysis, four individual sequences from family I, and three sequences OOOP OOONNHOBaseBaseOH O3' RNANNHHHH3NNH2NHH2NO5' RNA3' DNA5' DNA103  from family II were found to be highly conserved within their own families, with sequence variance of only one or two nucleosides (Table 3.1). Hence Dz7-38-32 and Dz7-45-28 were chosen for detailed cis-cleavage characterization due to their better catalytic performance in terms of both rate and extent of self-cleavage reaction.   Table 3.1 Two families of highly conserved sequences  Family Clone Sequence in the catalytic region (5’-3’)   I Dz7-38-32 TTACAGTGGTATCGATTGGGACGTGTGGAGCGTCAGGATGGTACA Dz7-38-39 TTACAGTGGTATCGATTGGGACGTGTGGAGCGTCAGGATGGTATA Dz7-38-83 TTACAGTGGTATCGATTGGGACGTGTGTAGCGTCAGGATGGGATCA Dz7-38-90 TTACAGTGGTATCGATTGGGACGTGTGGAGCGTCAGGATGGGATCA  II Dz7-45-10 AAGCAGCGGATGTGGTGTGACGCGGAGTCCCGAGTAGGGCTT Dz7-45-22 AAGCAGCGGATGTGGTGTGACGCGGAGTCCCGGGTAGGGCTT Dz7-45-28 AAGCAGCGGACGGATGGATGCGGAGTGCCTGAGTCAGGCCTT Only sequences in the catalytic region are shown. All the A, T, and C in blue are modified, corresponding to dAim, dUga, and dCaa, respectively.    The kinetic cis-cleavage data of both Dz7-38-32 and Dz7-45-28 under standard in vitro selection conditions (50 mM cacodylate, pH 7.45, 200 mM NaCl, 1 mM EDTA, at 25°C) were found to fit a double-exponential equation (Equation 3.1) better than a single-exponential equation.  104  Equation 3.1: P! = P!"#$ ! 1− 𝑒!!!"# !"#$! + P!"#$ ! 1− 𝑒!!!"# !"#$!       The apparent biphasic kinetic properties of these self-cleaving DNAzymes suggested that they must exist in more than one conformation, leading to the apparent partition of a fast phase and a slow phase. Both phases had distinct catalytic rates (kobs fast and kobs slow) and fractions that cleaved (defined by the phase amplitudes) where kobs fast and kobs slow are the observed cis-cleavage rates of the fast and slow phases, and P!"#$ !and P!"#$ ! are the amplitudes of the two phases, respectively. In a 150 min reaction time, Dz7-38-32 reached an overall cleavage yield (reaction plateau, P∞) of 62.50%: the amplitude of the fast phase (P!"#$ !) was calculated to be 36.23% with an extraordinary apparent catalytic rate constant (kobs fast) of 4.86 min-1, and for the slow phase it reached a reaction amplitude (P!"#$ !) of 26.27% with an apparent catalytic rate constant of 0.06 min-1. Dz7-45-28 reached a higher reaction plateau (P∞) of 99.83%, reflecting a higher degree of reaction completion. It attained a fast phase amplitude P!"#$ !  of 51.87%, with a substantially lower apparent catalytic rate constant (kobs fast) of 0.21 min-1 compared to Dz7-38-32, and a slow phase amplitude P!"#$ !) of 47.96%, with a slow phase apparent catalytic rate constant (kobs slow) of 0.007 min-1. These biphasic enzymatic parameters are summarized in Table 3.2. The cis-cleavage gel pictures and fitting curves are shown in Figure 3.2.  While the observation of biphasic kinetic cleavage data is not novel in this field, it is difficult to interpret such data and even more difficult to evaluate the dependence on temperature and pH, particularly because both temperature and pH can alter the relatively amplitudes of each phase. Biphasic kinetics have been observed for other 105  DNAzymes containing no expanded functionalities225,239,240 as well as ribozymes, where this effect has been studied extensively.241,242 In addition, Dz925-11 was characterized in detail and displayed biphasic trans-cleavage kinetics under single-turnover cleavage conditions at its optimum temperature of 13°C. Systematic investigation of the origin of the observed biphasic kinetics of Dz925-11 suggested the existence of two conformations of the DNAzyme-substrate complex, both of which led to substrate cleavage.243 However, for the biphasic cleavage kinetics of Dz7-38-32 and Dz7-45-28, an unambiguous conclusion as to the nature of the biphasic kinetics in trans cannot be made without pulse-chase experiments followed by stopped-quench kinetics and PAGE gel analysis, a line of experimentation that is beyond the scope of the characterization undertaken herein. Other explanations including the presence of synthetic impurities arising from enzymatic errors, the existence of kinetically inactive folds,241 and the existence of reverse reaction (ligation) rate might contribute as well.244    106   Figure 3.2 Cis-cleavage activity of Dz7-38-32 and Dz7-45-28.  (A) Representative PAGE gel images (7%) showing cis-cleavage time courses of Dz7-38-32 (top) and Dz7-45-28 (bottom), respectively. The uncleaved DNAzymes of full-length are indicated by filled arrowheads, and the cis-cleavage products are indicated by open arrowheads. (B) Plots of cis-cleavage percentage versus incubation time for Min 0 0.17 0.33 0.50 1 35 10 159020 30 60 754521.50.67105 120 150Min 0.5 31 1511874305150uncleavedcleaveduncleavedcleavedDz7-38-32Dz7-45-28AB-30 0 30 60 90 120 15015304560Time (min)Percent cleavageDz7-38-32Fast phase (kobs fast= 4.86 min-1, Pfast ∞ = 36.23%)Slow phase (kobs slow= 0.06 min-1), Pslow ∞ = 26.27%) -30 0 30 60 90 120 150153045607590Time (min)Percent cleavage Dz7-45-28Fast phase (kobs fast= 0.21 min-1, Pfast ∞ = 51.87%)Slow phase (kobs slow= 0.007 min-1), Pfast ∞ = 47.96%) 107  Dz7-38-32 (left panel) and Dz7-45-28 (right panel), respectively. The time-dependent fraction of cis-cleavage as a percentage of the total cleavage was derived from corresponding band intensity in part (A) and the data were fit to a double-exponential Equation 3.1 (curves in black). The blue curves and red curves were generated by plotting reaction rates and amplitudes of respective fast phase and slow phase to a single-exponential equation (Equation 2.1). All the data fitting and curve generation were carried out using Prism 6.  Table 3.2 Biphasic enzymatic parameters of Dz7-38-32 and Dz7-45-28               DNAzyme   Plateau (%)  Fast phase  Slow phase kobs fast (min-1) Amplitude (%) kobs slow  (min-1) Amplitude (%) Dz7-38-32 62.50 4.86 36.23 0.06 26.27 Dz7-45-28 99.83 0.21 51.87 0.007 47.96 Cis-cleavage conditions: 50 mM cacodylate (pH 7.45), 200 mM NaCl, 1 mM EDTA, at 25°C.  3.2.2 The simultaneous need of all three modifications Once the basic kinetic parameters had been determined, the necessity of individual modifications for catalytic competence was investigated. Using Dz7-38-32, I demonstrated that the observed high catalytic rate was due to the simultaneous presence of all the three modifications. This was shown by replacing one, two or all three modified nucleosides with their respective natural unmodified counterparts in the 108  enzymatic synthesis of the catalytic strand. Not surprisingly, replacement of any one of the modified nucleosides with an unmodified congener during the enzymatic synthesis gave self-cleaving strands that displayed near-total loss of cleavage activity, with less than ~5% cleavage observed over 840 minutes, suggesting that the modified nucleotides in the DNAzyme play crucial roles in terms of either structure or catalysis or both. The observed suppression of activity observed in the absence of one or more modified nucleosides was similar to that observed for Dz9-86. Interestingly, in contrast to Dz9-86, for which the replacement of modification(s) introduced significant amount of truncated products during the enzymatic synthesis of the self-cleaving strand, the synthesis of full-length Dz7-38-32 was not affected by replacing any of the modifications, despite its relatively large size compared to Dz9-86.      109   Figure 3.3 The necessity of individual modifications for catalytic competence of Dz7-38-32. (A) Gel images (PAGE 7%) demonstrating the necessity of all three modifications for self-cleavage activity of Dz7-38-32. As a control, the elongated strands were also cleaved with 0.1 M NaOH at 65°C for 20 min to identify the migration of the cleavage products, as denoted with OH-. Time points were 1, 5, 60, 120, and 840 min. (B) Chemical structures of dAimTP (1), dCaaTP (2), and dUgaTP (3).   dAimTPdCaaTPdUgaTPdAimTPdCaaTPdUgaTP+---+---+---++-+-+-++Time (1, 5, 15, 60, 120,840 min)1   8-histaminyl-dATP (dAimTP)2   5-aminoallyl-dCTP (dCaaTP)3   5-guanidinoallyl-dUTP (dUgaTP)NNNNNH2HNNHNO4-O9P3OOHNN ONH2H2NOOH4-O9P3ONHN OONHH2NNHOOH4-O9P3OOH-OH- OH- OH-110  3.2.3 Effect of temperature on cis-cleavage activity of Dz7-38-32 Changes in temperature generally have a significant effect on the catalytic rates of M2+-independent RNase A-mimicking DNAzymes. As mentioned in Chapter 1, Dz925-11 had a kobs (in cis) value of 0.3 min-1 at its optimum temperature of 13°C, but the observed value for the self-cleavage rate constant dropped to 0.044 min-1 at 37°C.173,177,243 The subsequent introduction of dUgaTP in the in vitro selection process was a major step forward in enhancing the thermal stability of active conformations.185 Consequently, the kobs of Dz10-66 for self-cleavage increased from 0.50 min-1 at 24°C to 0.6 min-1 at 37°C.186 Herein, the kobs values of Dz7-38-32 as a function of temperature, ranging from 15°C to 50°C, were investigated. All cis-cleavage reactions were performed at a range of temperatures in standard cleavage buffer conditions (50 mM cacodylate, pH 7.45, 200 mM NaCl, 1 mM EDTA), and the kinetic cleavage progression was sampled at 0.5, 1, 3, 5, 15, 30, 60, 90, and 120 min, respectively. Values of kobs were obtained by fitting the cleavage data using Prism 6 to the double-exponential equation (Equation 3.1).  Not unexpectedly, the obtained values of kobs and amplitude revealed that both the kobs and the partition of the two phases (slow and fast) were temperature-related. Thus, in addition to the temperature dependence of kobs fast (line in blue, Figure 3.4) and kobs slow (line in red, Figure 3.4), kapparent was calculated by integrating phase amplitude and corresponding kobs into Equation 3.2 to provide an overall reflection of the effect of temperature on catalytic performance.   111  Equation 3.2: 𝑘!""!#$%& = 𝑘!"# !"#$×P!"#$ ! + 𝑘!"# !"#$×P!"#$ !   A bell-shaped temperature profile (line in black, Figure 3.4) was generated based on the calculated kapparent values (Table 3.3), with a maximum kapparent found at 37°C among all the tested temperatures. This temperature profile was similar to what had been previously reported for Dz9-86185 and Dz12-91187, suggesting a temperature-related rearrangement of various secondary and/or tertiary conformations within the catalytic core or possibly the entire DNAzyme-substrate complex.  Table 3.3 Effect of temperature on cis-cleavage activity of Dz7-38-32   Temperature  (°C)  Fast phase Slow phase  kapparent (min1) kobs (min-1) Amplitude kobs (min-1) Amplitude 15 2.31 23.92 0.033 57.44 0.57 25 1.56 47.27 0.020 28.06 0.74 30 4.02 35.89 0.076 26.22 1.46 37 6.60 24.44 0.043 36.62 1.63 42 3.31 33.76 0.021 30.01 1.12 50 2.45 22.80 0.006 34.90 0.56 All the values of kobs and amplitude were calculated using Prism 6 based on cleavage data obtained under standard cleavage buffer conditions: 50 mM cacodylate (pH 7.45), 200 mM NaCl, 1 mM EDTA. kapparent was obtained by fitting kobs and amplitude to Equation 3.2.  112   Figure 3.4 Temperature profile of cis-cleavage activity of Dz7-32-38.  All cleavage reactions were measured in 50 mM cacodylate (pH 7.45), 200 mM NaCl, 1 mM EDTA at temperatures of 15°C, 25°C, 30°C, 37°C, 42°C and 50°C, respectively.   3.3 Characterizations In Trans  3.3.1 Conversion of cis-acting DNAzymes to trans-acting species  Following cis-cleavage characterization, efforts were directed to transform the cis-acting species to trans-acting ones for intermolecular and multiple-turnover substrate cleavage, thereby allowing the DNAzymes to function as true enzymes. Theoretically, the conversion can be readily achieved by simply removing the dispensable loop and separating the catalytic region and the substrate region (Figure 3.5A). However practically, the resulting DNAzymes are not necessarily active in trans because of the way the selection is configured such that selection for active species demands only cis-0 10 20 30 40 50 600.000.050.102468Temperature (°C) k (min-1)kapparentkobs fastkobs slow113  cleavage and does not provide a means of selecting directly for in trans cleavage. As mentioned previously in Chapter 2, a relatively large dispensable loop was assembled to increase the chance of success of this transformation (refer to Figure 2.3B in Chapter 2). To engineer a trans-acting activity of Dz7-38-32 and Dz7-45-28, trans-acting DNAzymes were enzymatically synthesized following the scheme shown in Figure 3.5B.  A synthetic DNA template containing a 5’-biotin along with a DNA primer corresponding to the 3’ sequence of the stem region in the hairpin loop structure (depicted as a black stick in the unimolecular construct in Figure 3.5A) were used in a template-directed polymerase-catalyzed primer extension reaction that provided the desired modified strand following the same protocol used for in vitro selection. The DNAzyme was separated from the template which remained immobilized on streptavidin beads following a brief 0.1 M NaOH wash for 2 min at room temperature, followed by an immediate neutralization with 0.1 M HCl to a pH ~7.45. The resulting DNAzyme was further desalted using G-25 spin column which removed the salt and provided pH equilibration in 10 mM Tris-HCl. The quantity of DNAzyme produced was determined by relating autoradiographic density to signal volume and ultimately to the number of pmole of α-32P-dGTP as previously reported.186 More details are described in the following Experimental part: Enzymatic synthesis of Dz7-38-32t.    114   Figure 3.5 Trans-acting DNAzymes. (A) Schematic representation of converting cis-acting DNAzymes to trans-acting ones. (B) Scheme for enzymatic synthesis of trans-acting DNAzymes. (1) Transcription of the C C G UNn-3'5'-.......TTTTTTTTTTTTTTT- Dispensable loop3' dispensable sequenceUnimolecular chimeric construct containing RNA substrate (in red) and catalytic DNA (in blue)RNA substrateCatalytic DNAC C G UNn5'- -3'-5'3'-Cis-acting to trans-actingBimolecular DNAzyme-substrate complexA  NeutralizationBTranscription(3)(1) dAimTP, dCaaTP, dUgaTP, dGTP+ α-32P-dGTPB3'-3'--3'-3'5'-Sequenase V2.05'-Strand separation 0.1 M NaOH0.1 M HCl5'-DNAzyme in transB3'-5'-(2)StreptavidinImmobilization-3'B115  modified strand in the presence of the three modified nucleotides, natural dGTP, and trace amount of α-32P-dGTP by template-directed primer extension reaction using Sequenase v2.0. (2) Immobilization of the resulting duplex on streptavidin coated magnetic beads through biotin-streptavidin binding. (3) Removal of the newly synthesized Dz7-38-32t using 0.1 M NaOH followed by neutralization with 0.1 M HCl.  As a direct consequence of the adoption of the larger dispensable loop, which represented a novel addition to the selection process, the cis- to trans- transformation was successfully achieved. In the initial qualitative assays for trans-acting activity, both Dz7-38-32 and Dz7-45-28 (denoting Dz7-38-32t and Dz7-45-28t), representative sequences from the two DNAzyme families were found to cleave all-RNA substrates in trans. They were both found to be optimally active on a 19 nt all-RNA substrate (resulting from a single base addition to each terminus of the original 17 nt substrate in the unimolecular construct), but not on the original 17 nt all-RNA substrate. This might suggest that the DNAzyme-substrate association step would be rate-limiting in the trans-cleavage reaction, which is consistent with what has been observed for other RNA-cleaving unmodified DNA enzymes.116,172  The cleavage site of each DNAzyme was assigned by comparing the cleaved substrate to an electrophoretic ladder that was generated by treating the substrate with carbonate leading to partial hydrolysis. It was revealed that Dz7-38-32t cleaved after the rG, whereas Dz7-45-28t cleaved after rU in the four-base preprogrammed bulge (5’-r(CCGU)-3’) opposite the catalytic core (Figure 3.6).  116   Figure 3.6 Qualitative trans-cleavage activity assay of Dz7-38-32t and Dz7-45-28t.  (A) PAGE gel (20%) picture of the trans-cleavage reaction progression. Dz: DNAzyme; S: the intact substrate; P: cleavage product; CO32-: partially alkaline-hydrolyzed substrate as ladder. The reactions were carried out in 50 mM cacodylate buffer (pH 7.5) containing 0.5 mM MgCl2, 150 mM KCl, at 30°C. [Dz]=not measured, [S]=500 nM. (B) DNAzyme in complex with 5’-32P-labeld 19 nt all-RNA substrate. The cleavage sites are indicated by arrows.   Based on the outstanding catalytic performance both in cis and in trans, Dz7-38-32t displayed more promise towards fulfilling all the criteria required to obtain a therapeutically competent DNAzyme. Therefore, it was chosen for further study focused on detailed characterization of the species’ enzymatic parameters in trans.   3.3.2 Trans-acting characterization of Dz7-38-32t The characterization of trans-cleavage was carried out under multiple-turnover 0 180 300 1320Min 0 180 300 1320MinDz7-38-32t Dz7-45-28tCO32-DzSP5'-32P-G G C G U G C  C  C  G  U  C U G U U G G C-3'3'-C C G C A C G G A C A  A  C  C G-5'ACatalitic coreDz7-38-32t Dz7-45-3-28tCleavage siteB117  conditions to determine the enzymatic parameters of Dz7-38-32t; DNAzyme ([Dz]) at 5 nM was used along with varying concentrations of substrate ([S]) in at least 5-fold excess, ranging from 25 nM to 4 µM or 8 µM, depending on the reaction temperature. The multiple-turnover reactions proceeded under simulated physiological ionic conditions: 50 mM sodium cacodylate (pH 7.45), 0.5 mM MgCl2, 150 mM KCl (Figure 3.7A). The kobs value of Dz7-38-32t at each substrate concentration was determined from a minimum of five time points obtained from the first 10-20% of the cleavage reaction. Each kobs reflects an overall cleavage rate and does not give a nuanced delineation of multiple cleavage phases, should they exist. Multiple-turnover profiles (Figure 3.7B) at both 30°C and 37°C were then generated using Prism 6 by fitting the kobs values and substrate concentrations into the Michaelis-Menten equation (Equation 3.3), where vobs is the initial cleavage rate (the first 10-20% of total reaction progression) observed for different substrate concentrations, and [E] and [S] represent the respective DNAzyme and substrate concentrations.  Equation 3.3: !!"#[!] = 𝑘!"# !!!! !          At 30°C and 37°C, Dz7-38-32t attained kcat values of 1.06±0.08 min-1 and 0.24±0.01 min-1, and KM values of 1.37±0.24 µM and 2.72±0.34 µM, corresponding to catalytic efficiencies of 7.7x105 M-1min-1 and 9x104 M-1min-1, respectively. These values are summarized in Table 3.4. Dz7-38-32 represents the most efficient all-RNA cleaving DNAzyme capable of multiple-turnover activity discovered to date in terms of the catalytic rates and catalytic efficiencies it attains in the presence of only 0.5 mM Mg2+, which underscores the potential of this DNAzyme as a therapeutically efficient agent. 118   Figure 3.7 Multiple-turnover catalytic activity of Dz7-38-32t (5 nM) under simulated physiological ionic strength: 0.5 mM Mg2+, 150 mM K+, pH 7.45.  (A) Representative autoradiographic images following PAGE for reactions containing 500 nM 5’-32P labeled RNA substrate. Left panel: reaction at 30°C. Right panel: reaction at 37°C. Time points and corresponding reaction turnovers are labeled on top of the PAGE images. Dz: Dz7-38-32t; S: the intact 5’-labeled substrate; P: 5’ cleavage product; CO32-: partially alkaline-hydrolyzed substrate as ladder. (B) Multiple-turnover profiles. Left panel: multiple-turnover profile based on three independent experiments at 30°C; substrate concentrations are 25, 50, 100, 300, 500, 1000, 2000, and 4000 nM; Time (min)Turnovers0 3010 60 110 220 12000 116 17 22 30 37   30°CDzSPTime (min)Turnovers0 3010 60 120 330 13200 32.2 4.7 7.5 15 35CO32-37°CADzSPBCO32-0 2000 4000 6000 8000 100000.000.050.100.150.2037°CSubstrate concentration (nM)v obs/[Dz] (min-1)kcat = 0.24±0.01 min-1 KM = 2716±342.7nM0 1000 2000 3000 4000 50000.00.20.40.60.81.0Substrate concentration (nM)v obs /[Dz] (min-1) 30°Ckcat = 1.06 ±0.08 min-1KM = 1372±245.2 nM119  kcat=1.06±0.08 min-1, KM=1.37±0.24 µM (R2=0.96). Right panel: multiple-turnover profile based on one set of experiment at 37°C; substrate concentrations are 100, 300, 500, 1000, 2000, 4000, and 8000 nM; kcat=0.24±0.01 min-1, KM=2.72±0.34 µM (R2=0.99).  Table 3.4 Trans-acting enzymatic parameters of Dz7-38-32t  Enzymatic parameters  30°C 37°C kcat ( min-1) 1.06±0.08 0.24±0.01 KM  (µM) 1.37±0.24 2.72±0.34 kcat/KM (M-1 min-1) 7.7x105 9 x104 All the cleavage reactions were carried out under simulated physiological ionic conditions: 50 mM sodium cacodylate (pH 7.45), 0.5 mM Mg2+, 150 mM K+.  The free software program m-fold was used to predict the hypothetical 2D structures of the DNAzyme-substrate complex of Dz7-38-32t and the other three close sequences from Family I (as listed in Table 3.1) under simulated physiological conditions (Figure 3.8). In contrast to the two way stem-loop structure adopted by the catalytic core of Dz10-66186 (Figure 1.10A, Chapter 1), three-way helical junctions formed by three stem-loop structures constituted the catalytic cores of these four DNAzymes, with the four unpaired RNA bases in substrate preprogrammed for cleavage directly opposite the junction. Nonetheless, m-fold cannot predict the structures brought about by the additional appendages which are highly possible to participate in folds.  120   Figure 3.8 Hypothetical 2D structures of Dz7-38-32t, Dz7-38-39t, Dz7-38-90t, and Dz7-38-83t in complex with the 19 nt all-RNA substrate, respectively.  A, C, and U in blue represent the modified nucleosides dAim, dCaa and dUga, respectively. The bases boxed represent the only one variant base between Dz7-38-32t and Dz7-38-39t (top panel), and between Dz7-38-90t and Dz7-38-83t (bottom panel).  5'-r(G   G   C   G   U   G  CG   A   C   A   A   C   C   G-5'U1UACA5GU G U G GU10AUCG15 AUGGC3'-C   C   G   C   A   C   GA35GGAU39A45A21CGUUGC30GG25UGGAGG42UA CC   U   G   U   U   G   G   C)-3'CCGUcleavage siteG   A   C   A   A   C   C   G-5'U1UACA5GU G U G GU10AUCG15 AUGGC3'-C   C   G   C   A   C   GA35GGAU39A45A21CGUUGC30GG25UGGAGG42UA UDz7-38-39tG   A   C   A   A   C   C   G-5'U1UACA5GU G U G GU10AUCG15 AUGGC3'-C   C   G   C   A   C   GA35G36GAU39A46A21CGUUGC30GG25UGUAGGUAG C45G   A   C   A   A   C   C   G-5'U1UACA5GU G U G GU10AUCG15 AUGGC3'-C   C   G   C   A   C   GA35G36GAU39A46A21CGUUGC30GG25UGGAGGUAG C455'-r(G   G   C   G   U   G  C C   U   G   U   U   G   G   C)-3'CCGUcleavage siteDz7-38-32t5'-r(G   G   C   G   U   G  C C   U   G   U   U   G   G   C)-3'CCGUcleavage site5'-r(G   G   C   G   U   G  C C   U   G   U   U   G   G   C)-3'CCGUcleavage siteDz7-38-90t Dz7-38-83t121  3.3.3 Attempts to increase catalytic efficiency at 37°C by lengthening the substrate-recognition arms  Dz7-38-32 showed a higher cis-cleavage rate at 37°C than at 30°C (Figure 3.4). However, when operating in trans, both its kcat and catalytic efficiency decreased several folds at 37°C relative to 30°C as summarized in Table 3.4. The catalytic efficiency of DNAzymes is normally limited by the rate of DNAzyme-substrate association, which is in turn highly sensitive to the length of the two substrate-recognition arms. For example, when targeting the start codon of HIV-1 gag-pol mRNA, the catalytic efficiency of Dz10-23 rose sharply as the length of the substrate-recognition arms increased from 4+4 to 7+7.116 Therefore, to address the decreased catalytic efficiency at 37°C, two longer substrates (21 nt and 25 nt) with additional bases at both termini were tested for multiple-turnover trans-acting activity under simulated physiological conditions at 37°C (Figure 3.9), based on the fact that distal base pairs have been found to have negligible or no effect on cleavage rate245. Values of kobs at 1 µM substrate concentration were compared with those obtained for the 19 nt substrate. Instead of improved trans-cleavage rate on lengthened substrates, Dz7-38-32t displayed better activity against the 19 nt substrate (kobs=0.05 min-1) compared to either the 21 nt substrate (kobs=0.016 min-1) or the 25 nt substrate (kobs=0.03 min-1), suggesting slightly decreased catalytic activity at 37°C that can not be readily restored by simply lengthening the substrate-recognition arms to facilitate DNAzyme-substrate association and enhance the stability of DNA-RNA heteroduplex.   122   Figure 3.9 Effect of substrate-recognition arm length on trans-cleavage rate at 37oC.  Sequences involved in DNAzyme-substrate binding are underlined. RNA substrates are in red. In Dz7-38-32t, modified nucleosides are in blue, with A=dAim, C=dCaa, and U=dUga, and unmodified nucleosides are in black.  The kobs values were measured under reaction conditions: 50 mM cacodylate (pH 7.45), 0.5 mM Mg2+, 150 mM KCl, at 37°C. [Dz]=5 nM, [S]=1 µM.   3.3.4 Probing the function of 8-histaminyl-deoxyadenosine in the 3’ substrate-recognition arm of Dz7-38-32t The modified 3’ substrate-recognition arm of Dz7-38-32 is inherently fixed by the selection method where this region must be synthesized enzymatically that results in the incorporation of modified nucleosides in the guide arm as well. The histaminyl appendage in the substrate-recognition arm (Figure 3.10A) would likely introduce some duplex instability to the DNAzyme-substrate complex, and thus may contribute to the G   A   C   A    A    C   C   G   C   G   T-5'U1UACA5GU G U G GU10AUCG15 AUGGC3'-G   C   G   C   C   G   C   A   C   GA35GGAU39A45A21CGUUGC30GG25UGGAGG42UA C5'-C   G   G   C   G   U   G  C C   U   G   U   U   G   G   C   G-3'C C G Ucleavage site5'-C   G   C   G   G   C   G   U   G  C C   U   G   U   U   G   G   C   G   C   A-3'C C G U5'- G   G   C   G   U   G  C C   U   G   U   U   G   G   C-3'C C G U 19 nt, kobs=0.05 min-121 nt, kobs=0.016 min-125 nt, kobs=0.03 min-1SubstrateDz7-38-32t123  observed decreased catalytic efficiency at 37°C. Indeed in Chapter 5, I show that the 8-histaminyl-dA mildly destabilizes duplexes that contain the 8-histaminyl-dA.  Moreover, this specific histaminyl-dA within the guide arm would be expected to be catalytically indispensable with the imidazole group acting potentially as the general acid or base, or structurally important in promoting an active conformation of the DNAzyme-substrate complex. To address this question, along with a more general question of substrate specificity, a mutated Dz7-38-32t bearing a natural “dG” in replacement of the dAim in the substrate-recognition arm, was transcribed and used to cleave the original 19 nt substrate (Figure 3.10B). In parallel, and a mutated substrate with a “rU to rC” change, which fully matched the mutated “dAim to dG” Dz7-38-32t, was tested for cleavage by this mutated Dz7-38-32t as well (Figure 3.10C). Not surprisingly, in either case where the dAim was replaced in the substrate-recognition arm, considerable reduction in catalytic activity was observed, as evidenced by the gel pictures in Figure 3.10. This finding suggests that the modified dAim plays a critical role in promoting a catalytically competent conformation of Dz7-38-32, reflective of unintended selection constraints imposed by the selection of using modified nucleotides. Nevertheless the exact role of this dAim is unclear insofar as it may play roles in folding, catalysis or both. This finding suggests that some constant sequences initially intended to serve solely for primer binding or substrate positioning were recruited for folding or catalysis in the process of selection, reflecting the complicated nature of in vitro selection.20 By the same token, my effort of using mutated Dz7-38-32 to cleave all-RNA substrates that have different sequences flanking the four unpaired RNA bases at the cleavage site was not successful. Dz7-38-32, unlike Dz10-23, is not insofar as a general-purpose RNA-124  cleaving DNAzyme.   Figure 3.10 The function of dAim in the modified 3’ substrate-recognition arm.  S: the intact 19 nt all-RNA substrate; P: 5’ cleavage product. Arrow indicates the site of cleavage. (A) Dz7-38-32t targeting the original 19 nt all-RNA substrate. (B) “dAim to dG” mutated Dz7-38-32t targeting the original 19 nt all-RNA substrate. (C) “dAim to dG” mutated Dz7-38-32t targeting the “U to C” mutated all-RNA substrate. Reaction conditions: 50 mM cacodylate (pH 7.45), 0.5 mM Mg2+, 150 mM KCl, at 30°C. [Dz]=5 nM, [S]=1 µM.   Analogous observations from other groups also revealed that in some cases pre-designed sequences that were expected to be dispensable and/or mutable to other sequences turned out to be indispensable to the optimal catalytic activity of the selected 5'-r( G G C G U G C  C C G U     C U G U U G G C)-3' C  C G C A C G G A C A  A C C G3'- -5'Catalytic coreNHNNH5'-r( G G C G U G C  C C G U     C U G U U G G C)-3' C  C G C G C G G A C A  A C C G3'- -5'Catalytic core5'-r( G G C G C G C  C C G U     C U G U U G G C)-3' C  C G C G C G G A C A  A C C G3'- -5'Catalytic coreTime (min):    0        60     120     12600 60 120 180 1260Time (min):    5   10  30  60 120 360 1260SPSPSPTime (min):AB C125  DNAzymes246 or the highest binding affinity of selected aptamers247. For example, Silverman et al. reported a phosphorserine lyase DNAzyme, in which the constant sequences included for primer binding in both the DNAzyme and the substrate were absolutely needed for optimal activity.246 Therefore, it is not so surprising that the attempts described here to generalize Dz7-38-32t for targeting substrates other than the original HIV-1 LTR-promoter target by changing the substrate-recognition sequences were not successful. 3.3.5 Activity of Dz7-38-32t on a chimeric DNA/RNA substrate containing exclusive DNA bases in substrate-recognition arms  As was previously discussed, efforts toward engineering DNAzymes selected to cleave DNA/RNA chimeric substrates into all-RNA cleavers were unsuccessful. This includes Dz10-66, which displayed exceptionally impressive trans-acting catalytic efficiency on a DNA/RNA chimeric substrate containing one embedded RNA base at the cleavage site, yet had no activity on the corresponding all-RNA substrate. On the other hand, all-RNA cleaving DNAzymes normally show an equal propensity for cleaving DNA/RNA chimeric substrates containing a single embedded ribonucleoside. This is nicely exemplified by Joyce’s Dz8-17, the activity of which is always greater in the presence of a chimeric substrate.248 To gain further insight in this trend and in regards to the question of substrate selectivity, Dz7-38-32t was studied for its capability to cleave a DNA/RNA chimeric substrate containing four RNA bases that do not form apparent base pairs with the guide arms, and where DNA bases were present for pairing in the substrate-recognition arms (Figure 3.11A). As observed for other all-RNA cleaving DNAzymes, 126  Dz7-38-32t was capable of cleaving this chimeric DNA/RNA substrate in the manner of multiple-turnover at 30°C (Figure 3.11B). However, a greatly improved catalytic efficiency was not achieved as expected, with calculated kcat and KM values of 0.46±0.04 min-1 and 3.5±0.728 µM, respectively (Figure 3.11C). The weaker binding reflected by the greater KM value is likely in part due to a generally lower thermal stability of a DNA-DNA duplex compared to an RNA-DNA duplex.249,250 Meanwhile, the increased value of KM value as well as the decreased value of kcat might also suggest that the active conformation of the Dz7-38-32t-substrate complex is impaired by the change of duplex shape, from A-form (formed between DNAzyme and the all-RNA substrate) to B-form (formed between the DNAzyme and the nearly all-DNA found in the chimeric substrate). Moreover, Dz7-38-32t also showed compromised cleavage specificity at the cleavage site, evidenced by the appearance of two cleavage bands at higher substrate concentrations, again reflecting a certain level of catalytic impairment that is reflected by increased catalytic promiscuity and a decrease in substrate specificity. It is not so surprising that Dz7-38-32t operated more effectively and specifically on the original all-RNA substrate, since the RNA substrate had been used throughout the selection process. Interestingly, the hammerhead ribozyme also displayed a higher catalytic efficiency on all-RNA substrates compared to chimeric DNA/RNA substrates.216  127    Figure 3.11 Activity of Dz7-38-32t on the chimeric DNA/RNA substrate.  (A) Dz7-38-32t in complex with the chimeric DNA/RNA substrate containing DNA bases in the substrate-recognition arms (underlined) and unpaired RNA bases (in red) at the cleavage region. (B) Representative PAGE gel (20%) of trans-cleavage reaction in the presence of 2 µM substrate. S: the intact substrate; P1: the major cleavage product; P2: non-specific cleavage product. Reaction conditions: 50 mM sodium cacodylate (pH 7.45), 0.5 mM MgCl2, 150 mM KCl, at 30°C. [Dz]=5 nM, [S]=2 µM. (C) Multiple-turnover profile. Substrate concentrations are 50, 100, 300, 500, 1000, 2000, 4000, 6000, and 8000 nM. kcat=0.81±0.08 min-1, KM=7.80±1.15 µM (R2=0.99).  3.4 Effect of Varying the RNA Bases at the Cleavage Site  To gain an insight into the base preference of Dz7-38-32t at the cleavage site, three additional substrates were constructed. Each substrate contained one of the other three SP1P2rCrCrGrU5'--3'5 10 30 60 120 300 1260 MinCO32-5'-G G C G T G C  r(C C G U )   C T G T T G G C-3'3'-C C G C A C G G A C A  A  C  C G-5'NHNHNDz7-38-32tP1 P2ABC0 2000 4000 6000 8000 100000.00.10.20.30.4Substrate concentration (nM)v obs /[Dz] (min-1)kcat = 0.46 ±0.04 min-1KM = 3503±728.6 nM128  standard ribonucleosides at the cleavage site, with the remainder of the sequence unchanged from the original substrate (Table 3.5). Multiple-turnover cleavage assays ([Dz]=5 nM, [S]=1 µM) revealed that Dz7-38-32t displayed optimal activity on the original 5’-rGrU-3’ junction, while exhibiting ~20-fold less activity on a 5’-rArU-3’ junction, and no activity on either of the two pyrimidine-rU junctions (Figure 3.12).  In addition to the low tolerance to various modifications in the substrate-recognition arm, Dz7-38-32t displayed high specificity to the base at the cleavage site as well. The overall high sequence specificity is critical to exclusive HIV-1 LTR-promoter mRNA disruption, especially considering applications in vivo, wherein high level of target discrimination from other cellular RNAs is absolutely necessary.    129   Figure 3.12 PAGE gel (20%) pictures for substrates containing different RNA bases at the cleavage site.  (A) The original substrate containing rG at the cleavage site. (B) The substrate containing rA at the cleavage site. (C) The substrate containing rC at the cleavage site. (D) The substrate containing rU at the cleavage site. All the cleavage reactions were carried out under simulated physiological ionic conditions: 50 mM cacodylate (pH 7.45), 0.5 mM Mg2+, 150 mM KCl, at 30°C. [Dz]=5 nM, [S]=1 µM.    5'-C C A U-3'0 105 621260360120300 2010 801409184400 2010 80140918440Min5 3010 120126036060MinAC DB5'-C C G U-3'5'-C C C U-3' 5'-C C U U-3'SubstrateCleavage product130  Table 3.5 Effects of varying the RNA bases at the cleavage site                   RNA substrate kobs (min-1) 5’-GGCGUGCCCGUCUGUUGGC-3’  0.43 5’-GGCGUGCCCAUCUGUUGGC-3’  0.024 5’-GGCGUGCCCCUCUGUUGGC-3’  n.ma 5’-GGCGUGCCCUUCUGUUGGC-3’  n.m The ribonucleosides at the cleavage site are underlined. All the cleavage reactions were carried out under simulated physiological ionic conditions: 50 mM cacodylate (pH 7.45), 0.5 mM Mg2+, 150 mM KCl, at 30°C. [Dz]=5 nM, [S]=1 µM. a Not measurable.  3.5 Analysis of Cleavage Product  As introduced in detail in 1.3.1, most self-cleaving ribozymes251,252 and RNA-cleaving DNAzymes253, as well as RNase A140 cleave the RNA phosphodiester bond through a transesterification-like mechanism, whereby the 2’-ribose oxygen attacks the adjacent 3’-phosphate. This generates a 5’-cleavage product having a 2’, 3’-cyclic phosphate, and a 3’-cleavage product having a 5’-OH (Figure 3.13A). Qualitative analysis methods that enable analysis of the cleavage pathway include checking the extension disability of the 5’-cleavage product in the presence of α-32P-dATP and terminal deoxynucleotidyl transferase; the phosphorylation ability of the 3’-cleavage product by γ-32P-ATP and T4 polynucleotide kinase allow for identification of cleavage products. Phosphorylation of the trans-cleavage reaction mixture of Dz7-38-32t by γ-32P-ATP and T4 polynucleotide 131  kinase resulted in a radiolabeled 3’-cleavage product of the substrate, which is in consistent with the presence of a free 5’-OH, suggesting the transesterification reaction pathway is indeed operative in this case (Figure 3.13B). However, the 5’-cleavage product remained uncharacterized, which means there is a possibility that the cleavage occurred through hydrolysis (H2O attacks the adjacent 3’-phosphate instead of the 2’-ribose oxygen). While this possibility can not be totally ruled out, it is more likely that Dz7-38-32 generated a 2’, 3’-cyclic phosphate given the fact that the direct addition of water to phosphate (hydrolysis reaction) is substantially disfavored relative to the transesterification reaction (the uncatalyzed half-life of the former is ~30 million years254 while the latter is ~10 years65 in the absence of divalent metal ions53). Additionally, Dz7-38-32 showed resistance to substrates containing four unaired DNA bases at the cleavage region (data not shown), which provides further evidence against the H2O hydrolysis cleavage pathway.   132   Figure 3.13 Analysis of the cleavage product.  (A) Schematic representation of trans-cleavage reaction and products with characteristic functionalities of transesterification reaction. (B) Gel image (20%) for the 5’-phophorylation of the trans-cleavage mixture. Lane 1, trans-cleavage reaction mixture at time point 0 min; lane 2, trans-cleavage reaction mixture after overnight incubation; lane 3, polynucleotide kinase phosphorylated trans-cleavage reaction after overnight incubation. Trans-cleavage reaction conditions: 50 mM cacodylate (pH 7.45), 0.5 mM Mg2+, 150 mM KCl, at 30°C. [Dz]=5 nM, [S]=1 µM. The polynucleotide kinase phosphorylation reaction was carried out under 1X polynucleotide kinase buffer conditions at 37°C for 3 h.   3.6 Analysis of pH Dependence  The general acid-base catalysis mediated by the appended imidazole groups was C C G UCatalytic core5'-32P- -T15-3'-5'3'-SubstrateDz7-38-32Cleavage site5'-32P- C COGOPOOOOOPOOOOUHOOHPOOO -T15-3'5'-cleavage product3'-cleavage productOOUOOHPOOOPhosphorylated 3'-cleavage productPOOO 32γ-32P-ATPT4 PNK-T15-3'AS, 34 nt5'-cleavage product, 10 ntPhosphorylated 3'-cleavage productDz1 2 3B+133  experimentally demonstrated for Dz925-11.177 Moreover, the bell-shaped pH profile displayed by both Dz10-66 and Dz9-86 also suggests that the imidazole groups play a catalytic role.185,186 Here, the kobs dependence of Dz7-38-32t on pH was investigated under multiple-turnover trans-cleavage conditions (Figure 3.14A). The logarithmic value of kobs increased linearly within pH range of 6.0 to 7.0 with a slope value of 1.139, whereas the logarithmic value of kobs decreased linearly within pH range of 7.0 to 8.5 with a negative slope value of -1.143 (Figure 3.14B). The close to unity slope value suggests that either the deprotonation step at pH values lower than the apparent pKa or protonation step at pH values higher than the apparent pKa is rate-limiting in the concerted SN2-like transesterification reaction depicted in Figure 1.2, Chapter 1. Fitting the obtained kobs values to a double ionization model reflected by Equation 3.4 generated a bell-shaped rate pH-dependence profile over the tested pH range (Figure 3.14C), suggesting the participation of two catalytically relevant groups in a two-step protonation-deprotonation mechanism.83,255  Equation 3.4: 𝑘!"# = !!"#!!!" !"!!!!! !!" !!!!!!" !!" !!!!!!!!!   The calculated 𝑝𝐾!! and 𝑝𝐾!! were exactly the same, with a value of 7.0. While this cannot decisively demonstrate two ionizable groups, the appearance of a bell-shaped pH-rate profile is consistent with a mechanism involving two appended imidazole groups engaged in general acid and base catalysis. However, other possibilities that might lead to a bell-shaped pH-profile cannot be excluded, particularly nucleobase-mediated catalysis. Mechanisms involving a “histidine-like” cytosine acting as either general acid57 or general base,256  and an adenine257 acting as a general acid have been found in self-134  cleaving ribozymes. Moreover, pH-dependent conformational change may also contribute to the bell-shaped pH-profile. With the current data, it is premature to conclude that the bell-shaped pH-rate profile necessarily reflects general acid and general base catalysis, or that it is due to two specific imidazoles. Future studies including affinity labeling177 and structural elucidation by NMR258 or crystallography259 will help to identify the catalytic mechanism and critical groups involved in catalysis.    135   Figure 3.14 The pH dependence study of Dz7-38-32t.  (A) Gel (20%) pictures of trans-cleavage reactions under 1X pH variant buffers containing 0.5 mM MgCl2, 150 mM KCl, at 30°C. S: the intact substrate, P: cleavage product. pH 6.0-7.0 buffers: 50 mM sodium cacodylate. pH 7.5-8.5 buffers: 50 mM sodium phosphate. [Dz]=7.5 nM, [S]=1.5 µM. (B) The logkobs-pH linear regressions in pH 6-7 (blue line, y=1.139X-8.284), and pH 7-8.5 (red line, y=-1.143X+7.745), respectively. (C) pH-rate profile generated by fitting values of kobs and pH to Equation 3.4, R2=0.85. 5 6 7 8 9-2.5-2.0-1.5-1.0-0.50.0pHlogkobs5 6 7 8 90.00.10.20.30.40.5pHk obsASPSP10 4020 90126012060Min240pH=6.0 pH=6.5 pH=7.0pH=7.5 pH=8.0 pH=8.5B C136  3.7 Divalent Metal Ion Effects on Cleavage Activity  Beyond the very fundamental concern of exploring the advantages and limitations of modified DNAzymes, the ultimately practical initiative for isolating and characterizing M2+-independent RNase A-mimicking DNAzymes is to find DNAzymes that could be used as therapeutic reagents under physiological conditions. In other words, M2+-independent DNAzymes that are also M2+-tolerant should be such that absence or presence of at least physiological concentrations of divalent metal ions would not cause significant catalytic efficiency loss as that found with Dz10-23.109  Dz7-38-32t displayed tolerance to physiological concentrations of Mg2+ (0.5 mM)110 in the previous trans-cleavage assays. Additionally, Mg2+ concentrations of up to 10 mM caused only negligible difference in trans-acting kobs. Various other divalent metal cations were also investigated to determine their effect on the trans-acting activity Dz7-38-32t; Zn2+, Cu2+, Hg2+ were examined because of their high affinity to the appended functionalities, particularly imidazoles; Ca2+ was tested due to its high physiological relevance; and Mn2+ and Pb2+ were investigated because of the high relevance of these ions to the activity of other reported DNAzymes. The results are shown in Figure 3.15. The experimentally obtained values for kobs are summarized in Table 3.6. Generally, Dz7-38-32t was more tolerant to the presence of divalent metal ions to certain levels relative to other RNase A mimics reported by the Perrin lab previously. In contrast to Dz925-11260 and Dz9-86185, Zn2+ and Cu2+ did not cause significant change in kobs of Dz7-38-32t with concentrations up to 0.5 mM, but in the presence of 100 µM Pb2+, Dz7-38-32t lost almost all activity. Conversely, Dz7-38-32t still displayed a kobs of ~ 0.10 min-137  1 in the presence of 50 µM of Hg2+, whereas the activities of Dz925-11 and Dz9-86 were totally depleted by the presence of less than 10 µM Hg2+, probably owing to its “poisoning” effect on the appended imidazoles.215 These findings suggest that the intramolecular secondary or tertiary structure adopted by Dz7-38-32 is more robust to external interference compared to these other DNAzymes, which again reflects its superiority as a representative of a new generation of RNase A-mimicking DNAzymes.    138   Figure 3.15 Effects of divalent metal ions on the trans-cleavage activity of Dz7-38-32t.  S: the intact substrate; P: the cleavage product. All the reactions were carried out in 50 mM cacodylate buffer (pH 7.45) containing 150 mM KCl, and the tested concentration of M2+, at 30°C. P: cleavage product. [Dz]=5 nM, [S]= 1 µM.   0.5 mM Zn2+M2+-free0.5 mM Ca2+0.5 mM Mn2+0.5 mM Cu2+Min: 5 10 20 30 60 12036013805 10 20 60 120 3601380Min: Min: 0 5 10 30 60 12036012605 10 30 60 120 300 1260Min:0.05 mM Hg2+SPSPSPSPSPSP0.05 mM Pb2+Min: 5 10 20 30 60 1203601380Min: 5 10 20 30 60 12036013805 10 30 60 120 300 1260Min:SP139  Table 3.6 Effects of divalent metal ions on trans-cleavage reaction of Dz7-38-32t M2+ Concentration (mM) kobs  (min-1) M2+-free 0 0.43 Mg2+ 0.5 0.51 Cu2+ 0.5 0.36 Zn2+ 0.5 0.41 Ca2+ 0.5 0.46 Mn2+ 0.5 0.48 Pb2+  0.05 0.26 Hg2+ 0.05 0.10 All the reactions were carried out in 50 mM cacodylate buffer (pH 7.45) containing 150 mM KCl, and the tested concentration of M2+, at 30°C. [Dz]=5 nM, [S]= 1 µM.  3.8 Discussions and Conclusion Using enzymatically transcribed materials, this chapter focused on the characterization of enzymatic parameters and elucidation of catalytic properties and sequence specificity of Dz7-38-32. The experiments were inspired by analogous experiments used by other RNA-cleaving DNAzymes and RNase A-mimicking DNAzymes. The concentrations of DNAzymes produced by enzymatic synthesis were measured by relating autoradiographic densities to signal volumes and ultimately the number of pmole of α-32P-dGTP incorporated. It is assumed that these calculated concentrations should 140  roughly reflect the true DNAzyme concentrations. However, no PAGE gel purification was applied to the very little material made by enzymatic polymerization, which means the signal resulting from some truncated products caused by incomplete synthesis would be likely to lead to a higher apparent concentration of the fully-intact DNAzyme. In other words, the true catalytic efficiency of Dz7-38-32t should be higher than that obtained here. To perform accurate enzymatic parameter characterization, DNAzyme structure elucidation, and cell based or in vivo activity and toxicity assays, pure DNAzymes made on solid phase through the coupling of respective phosphoramidites of dAimTP, dCaaTP, dUgaTP are required.  The obtained kobs fast for self-cleavage (25°C) and kcat in intermolecular cleavage (30°C) under no-M2+ or low-M2+ conditions correspond to a greater than 108-fold enhancement of the uncatalyzed rate constant for RNA transesterification (~10-8 min).65 As previously discussed in Chapter 1 (Figure 1.2), there are two catalytic strategies commonly employed by RNA-cleaving DNAzymes: in-line nucleophilic attack (strategy α) and deprotonation of the 2’-OH group (strategy γ). Theoretically, the accumulative rate enhancement derived from perfecting utilization of these catalytic strategies is 108.55,56 Therefore, Dz7-38-32 almost certainly employed at least one additional catalytic strategy to achieve such a rate enhancement. This hypothesis is further supported by the bell-shaped pH-profile, which is highly indicative of catalytic strategy δ (neutralization of 5’-oxygen of 3’-cleavage product by protonation, presumably mediated by one of the imidazoles in histaminyl-dA that would play the role of acid catalysis). The existence of multiple strategies in the catalytic process of Dz7-38-32t provides more 141  evidence for the notion that nucleic acids have the ability to employ combinations of catalytic strategies, although in this case some of these strategies are possible because they are supported by the grafted functionalities.  In conclusion, Dz7-38-32t, the best representative from two families of RNase A- mimicking DNAzymes isolated by in vitro selection, attained kcat and KM values of ~0.24 min-1 and 2.72 µM under physiological conditions (pH 7.45, 150 mM K+, 0.5 mM Mg2+, 37°C).  This corresponds to a catalytic efficiency of ~105 M-1min-1, the best catalytic efficiency achieved by chemically modified DNAzymes under low-M2+ characteristic of cellular conditions up to date. Additionally, mutations in the guide arms and in terms of sequence specificity of cleavage confirmed its high specificity for the original substrate derived from the HIV-1 LTR-promoter mRNA, which makes it a superior candidate for mRNA disruption in vivo, wherein a very high degree of substrate differentiation is an absolute requirement.261,262 These properties endow Dz7-38-32t with all the characteristics of therapeutically efficient DNAzymes described in 2.1, Chapter 2, including validated sequence specificity on an all-RNA substrate, robust activity at low or no Mg2+ characteristic of cells, and moderate to high catalytic efficiency (>105 M-1min-1, with a kcat value > 0.05 min-1) to ensure a reasonably high rate of mRNA disruption before the DNAzyme is diluted due to cell division. 3.9 Experimental 3.9.1 General reagents, materials, and enzymes  dAimTP and dUgaTP were synthesized according to literatures by Dr. Curtis Lam. 142  dCaaTP was obtained from TriLink Bio Technologies. Oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA), and purified by denaturing PAGE (7 M). Vent (exo-) DNA polymerase, Taq DNA polymerase, lambda exonuclease, T4 DNA ligase, and T4 polynucleotide kinase were obtained from New England Biolabs. Sequenase v2.0 and pyrophosphatase were purchased from Affymetrix. Steptavidin magnetic particles were purchased from Roche. α-32P-dGTP and γ-32P-ATP were obtained from Perkin Elmer. Sephadex G25 resin was obtained from GE.   3.9.2 General protocols  The general protocols including PAGE gel electrophoresis, PAGE gel purification of oligonucleotides, and all other basic procedures for manipulating oligonucleotides generally followed the protocols described in 2.9.2, Chapter 2 without otherwise mentioning.  3.9.3 Buffers/solutions 1X T4 polynucleotide kinase buffer: 70 mM Tris-HCl (pH 7.6 @ 25°C), 10 mM MgCl2, 5 mM DTT. 1X Sequenase buffer: 40 mM Tris-HCl (pH 8.0), 50 mM NaCl, 20 mM MgCl2.  Gel loading buffer: formamide, 25 mM EDTA, 0.01% bromophenol blue and 0.01% xylene cyanole. Gel loading buffer containing biotin: gel loading buffer/100 mM biotin in DMF (99:1). Wash buffer (TEN): 50 mM Tris-HCl (pH 7.5), 200 mM NaCl, 1 mM EDTA.  143  Neutralization buffer: 25 mM sodium cacodylate (pH 6.0), 1 mM EDTA. 1X cis-cleavage buffer: 50 mM sodium cacodylate (pH 7.4), 200 mM NaCl, 1 mM EDTA.  10X trans-cleavage buffer (containing Mg2+): 0.5 M sodium cacodylate (pH 7.4), 1.5 M KCl, 5 mM MgCl2. 10X M2+-free trans-cleavage buffer: 0.5 M sodium cacodylate (pH 7.4), 1.5 M KCl. 5X pH variant trans-cleavage buffer: pH 6.0-7.0, 250 mM sodium cacodylate, 750 mM KCl, 2.5 mM MgCl2; pH 7.5-8.5, 250 mM sodium phosphate, 750 mM KCl, 2.5 mM MgCl2. M2+ stock solutions in H2O: 5 mM CuCl2, ZnCl2, CaCl2, and MnCl2; 0.5 mM Pb(CH3CO2)2 and Hg(CH3CO2)2.  Gel elution buffer: 10 mM Tris-HCl, (pH 8.0), 1% LiClO4. 1X TBE solution: 89 mM Tris base, 89 mM boric acid, 2 mM EDTA.  5X modified nucleotides cocktail: 250 µM dAimTP, 125 µM dCaaTP, 50 µM dUgaTP, 50 µM dGTP. 3.9.4 Oligonucleotides (5’ to 3’, ON) Primer for cis-cleavage characterization, ON 1: Biotin-T40CCCGGGTTTTTr(GCGUGCCCGUCUGUUGG)TTTTGCGTCGGCC TGCGCCAACAG.  Template for transcription of cis-acting Dz7-38-32, ON 2: 144  GCGCTCGCGCGGCGTGCCGCCCACCTACGCTGCACACATGCCAACCGCTACCACTGTAACTGTTGGCGCAGGCCGACGC. Template for transcription of cis-acting Dz7-45-28, ON 3: GCGCTCGCGCGGCGTGCAGCTACCCAGCCACTACGCGTGCCATCACATGCGCTGCTTCTGTTGGCGCAGGCCGACGC. Primer for transcription of Dz7-38-32t, ON 4: GCCTGCGCCAACAG. Template for transcription of trans-acting Dz7-38-32t, ON 5:  Biotin-GCGCGGCGTGCTGTACCATCCTGACGCTCCACACGTCCCAATCGAT ACCACTGTAACTGTTGGCGCAGGC. Template for transcription of trans-acting Dz7-45-28t, ON 6: Biotin-GCGCGGCGTGCCGCCCACCTACGCTGCACACATGCCAACCGCTA CCACTGTAACTGTTGGCGCAGGC. Original 19 nt all-RNA substrate, ON 7:  r(GGCGUGCCCGUCUGUUGGC).  21 nt all-RNA substrate, ON 8:  r(CGGCGUGCCCGUCUGUUGGCG). 25 nt all-RNA substrate, ON 9:  145  r(CGCGGCGUGCCCGUCUGUUGGCGCA). 19 nt all-RNA substrate containing rU to rC mutation, ON 10:  r(GGCGCGCCCGUCUGUUGGC).  19 nt DNA/RNA chimeric substrate, ON 11:  GGCGTGC r(CCGU) CTGTTGGC.  19 nt all-RNA substrate containing mutated RNA base at the cleavage site, ON 12, ON13, and ON 14:  r(GGCGUGCCCAUCUGUUGGC), r(GGCGUGCCCCUCUGUUGGC), and r(GGCGUGCCCUUCUGUUGGC). 31 nt all-RNA substrate with 3’-T12 tail for cleavage product characterization, ON 15:  r(GGCGUGCCCAUCUGUUGGC)-TTTTTTTTTTTT.  “r” designates a stretch of RNA bases. Underlined bases in all different versions of substrates represent the base at the cleavage site. “C” in red in ON 9 highlighted the U to C mutation in the substrate.   3.9.5 Cis-cleavage characterization  Kinetic analysis in cis. Both Dz7-38-32 and Dz7-45-28 were prepared in the same way as the modified library in the selection as described in “Template-directed primer extension for transcribing modified library for selection, 2.9.4, Chapter 2”. 15 pmole of synthetic template (ON 2 or ON 3), 15 pmole of primer (ON 1), and 8 µL of 5X 146  Sequenase buffer, and an amount of H2O were mixed to a volume of 26.5 µL. This mixture was heated to 95°C for 5 min followed by slowly cooling down to room temperature for annealing. To the annealing mixture, 8 µL of 5X modified nucleotides cocktail, 2 µL of DTT (100 mM), 1 µL of pyrophosphatase, 1.5 µL of α-32P-dGTP (~15 µCi), and 1 µL of Sequenase v2.0 were added sequentially to make a final reaction volume of 40 µL. The reaction was gently mixed well and incubated at 32°C for 4 h with mineral oil on top to minimize the effect of condensation. The reaction was quenched by the addition of 2 µL of EDTA (0.5 M, pH 8.0) to a final concentration of ~25 mM. The extension product was immobilized on 100 µL of pre-washed (by TEN buffer, 100 µL per wash) streptavidin coated magnetic beads by incubating at room temperature for 15 min. Following two more washes with 100 µL of TEN, the template strand was removed by 5 quick washes with 100 µL of 0.1 M NaOH containing 1 mM EDTA. The resulting modified strand sticking to streptavidin beads was immediately neutralized using 200 µL of neutralization buffer followed by a final 100 µL of water wash. Then the modified DNA on beads was incubated in 100 µL of standard cleavage buffer (50 mM cacodylate pH 7.45, 200 mM NaCl, 1 mM EDTA), which formed a mixture of uncleaved (sticking to streptavidin beads) and cleaved (falling off streptavidin beads into solution) materials. For Dz7-38-32, 5 µL of the mixture was removed and quenched in 15 µL of gel loading buffer containing 1 mM biotin at time points 0, 0.17, 0.33, 0.50, 0.67, 1, 1.5, 2, 3, 5, 10, 15, 20, 30, 45, 60, 75, 90, 105, 120, and 150 min, respectively. For Dz7-45-28, 5 µL of the mixture was removed and quenched in 15 µL of gel loading buffer containing 1 mM biotin at time points 0.50, 1, 3, 5, 10, 15, 30, 74, 118, and 150 min, 147  respectively.  All samples were denatured by incubating at 95°C for 5 min, followed by snap-cooling on ice before resolving the supernatant after magnetization on denaturing PAGE (7%). Visualizations and quantifications were carried out using Typhoon 9200 PhosphorImager (Amersham). The autoradiographic data of the cleavage reactions (represented by pixel volumes on the Imagequant software) were obtained by drawing polygons around the bands corresponding to the cleaved and uncleaved species. These data were fitted to a double exponential first-order reaction using Prism 6. Equation 3.1: P! = P!"#$ ! 1− 𝑒!!!"# !"#$! + P!"#$ ! 1− 𝑒!!!"# !"#$! , where kobs fast and kobs slow are the observed cis-cleavage rates of the fast and slow phases, and the Pfast ∞and Pslow ∞are the amplitudes of the two phases, respectively.  Testing the need for modifications. 5X nucleotides cocktails containing natural counterpart(s) in replacement of any one or two, or all the three modifications (dAimTP, dUgaTP, dCaaTP) contained in the 5X modified nucleotides cocktails were used into template-directed primer extension reactions for transcribing 15 pmole of partially modified or totally unmodified Dz7-38-32. Then kinetic cis-cleavage reactions were carried out following the same procedures described above in “Kinetic analysis in cis”. Time points were taken at 1, 5, 15, 60, 120, and 840 min for all different versions of Dz7-38-32, respectively. All samples were denatured by incubating at 95°C for 5 min, followed by snap-cooling on ice before resolving the supernatant after magnetization on denaturing PAGE (7%). Gel pictures were taken using Typhoon 9200 PhosphorImager.  Addressing temperature effect on cis-cleavage reaction. 60 pmole of primer (ON 1), 148  60 pmole of synthetic template of Dz7-38-32 (ON 2), and 32 µL of 5X Sequenase buffer, and an amount of H2O were mixed to a volume of 106 µL. This mixture was heated to 95°C for 5 min followed by slowly cooling down to room temperature for annealing. To the annealing mixture, 32 µL of 5X modified nucleotides cocktail, 8 µL of DTT (100 mM), 4 µL of pyrophosphatase, 6 µL of α-32P-dGTP (~15 µCi), and 4 µL of Sequenase v2.0 were added sequentially to make a final reaction volume of 160 µL. The reaction was mixed well and split into 4 tubes of 40 µL aliquots. Then the reaction was run for 4 h at 32°C with mineral oil on top of solution to minimize the effect of condensation. The reactions in the four tubes were recombined, and quenched by adding 8 µL of EDTA (0.5 M, pH 8.0) to a final concentration of 25 mM.  Kinetic cis-cleavage reactions at different temperatures (15°C, 25°C, 30°C, 37°C, 42°C and 50°C) were performed in 1X standard cleavage buffer following the same procedures described above in “Kinetic analysis in cis”. 28 µL of the recombined reaction above was used for assay at each temperature. Kinetic cleavage progression was sampled at 0.5, 1, 3, 5, 15, 30, 60, 90, and 120 min reaction times, respectively. kobs values were obtained by fitting the cleavage data to the double exponential equation (Equation 3.1) using Prism 6. 3.9.6 Trans-cleavage characterization  5’-32P labeling of substrates using T4 polynucleotide kinase. 5 µL of substrate stock solution (1 mM) was mixed with 5 µL of 10X T4 polynucleotide kinase buffer, 3 µL of γ-32P-ATP (~30 µCi), and 5 µL of T4 polynucleotide kinase (50 units), then 32 µL of H2O was added to make a final reaction volume of 50 µL. The reaction was incubated at 149  37ºC for 3 h and then terminated by incubating at 65ºC for 20 min. After adding 50 µL of gel loading solution, the resulting 100 µL of solution was loaded to 2 wells of a 20% small denaturing PAGE (50 µL per well). The reaction product was visualized by UV-shadowing after electrophoresis and gel slice containing the labeled substrate was cut out, eluted, and subjected to G-25 column desalting before concentration measurement on the DU800 Spectrophotometer.  Enzymatic synthesis of Dz7-38-32t. According to the methodology used for producing Dz10-66t1, 60 pmole of primer for transcribing of Dz7-38-32t (ON 4) was annealed to 60 pmol of the synthetic template of Dz7-38-32t containing a 5’-biotin moiety (ON 5) in a mixture of 55 µL solution containing 16 µL of 5X Sequenase buffer following the same protocol as described for the transcription of cis-cleaving Dz7-38-32. Following that, the catalytic strand was enzymatically synthesized at 32°C for 4 h using 26 units of Sequenase v2.0 (2 µl) in a final volume of 80 µL reaction supplemented with 16 µL of 5X modified nucleotides cocktail, 4 µL of DTT (100 mM), 2 µL of pyrophosphatase, and 1 µL of α-32P-dGTP (~10 µCi). 4 µL of 0.5 M EDTA (pH 8.0) was added to a final concentration of 25 mM to quench the reaction at the end. 1 µL aliquot of this solution was diluted sequentially into 3X, 10X, 100X, 1000X, and 2000X solutions before the rest of the crude reaction was subjected to DNAzyme “workup”.   The catalytic stand in duplex with its template was immobilized on prewashed streptavidin magnetic beads through the 5’-biotin moiety on the template. After two washes with 100 µL of wash buffer, the catalytic strand was stripped off with 30 µL of 0.1 M NaOH followed by immediate neutralization with 0.1 M HCl to obtain a final pH of 150  ~7.5. The resulting DNAzyme solution was spun down through G25 column to desalt. 1 µL of the desalted DNAzyme was diluted for 10 times in H2O, 1 µL of which along with 1 µL of each of the solutions resulting from sequential dilutions of the crude reaction were spotted on a TLC-plate (Figure 3.16A). An autoradiographic calibration curve (Figure 3.16B) represented by Equation 3.5 which relates autoradiographic density to signal volume and ultimately to the number of pmole α-32P-dGTP (specificity) was generated by plotting logarithm of the signal intensities obtained from sequential dilutions against the logarithm of the dilution factors using linear regression. Then the concentration of DNAzyme produced can be calculated using Equations 3.5 and 3.6.   Equation 3.5:  𝑦 =  𝑎𝑥+ 𝑏 Equation 3.6: [𝐷𝑁𝐴𝑧𝑦𝑚𝑒] = !! ! !" !!"#$ ×!"!"∙!!!  In Equation 3.5, 𝑎 is the slope and 𝑏 is the intercept of the linear calibration curve; in Equation 3.6, 𝑥  is the calculated value after applying the logarithm of the signal intensity of the pure DNAzyme (Figure 3.16A, the spot in the blue rectangular) into Equation 3.5; the factors of 10 and 21 arise from the 10-fold dilution of the pure DNAzyme spotted on the TLC plate and the presence of 21 dGs in the polymerized sequence region of Dz7-38-32t, respectively.  Dz7-38-32t was found to be in the range of 135 to 270 nM corresponding to an estimated yield of ~8.1-16.2 pmole out of a 60 pmole of primer extension reaction.  151   Figure 3.16 Concentration determination of Dz7-38-32t using autoradiographic calibration curve. (A) Autoradiogram of the standard curve representing the quantity of the total radioisotope at various dilutions. (B) Calibration curve of the logarithm of the signal intensity versus the logarithm of the dilution factor and the corresponding linear least square fit (𝑦 = -1.093· 𝑥 +17.97, R2=0.9978). Dz7-38-32t, indicated by the blue square on the generated curve, has a concentration of 270 nM for this particular assay.  3X 10X 100X 1000X 2000X 10XA0 2 4 6 8 1069121518ln (dilution factor)ln (volume)Dilution factor:DNAzymeDNAzymeB152  General protocol for trans-cleavage reactions. For trans-reaction assays carried out under multiple-turnover conditions, normally 5 nM to 10 nM of DNAzyme ([Dz]) was used, and the concentrations of 5’-32P-labeled substrate ([S]) used were always at least 10-fold in excess over that of the enzyme. DNAzyme and substrate were firstly incubated in 18 µL of solution at 95°C for 2 min followed by cooling on ice for 5 min before 2 µL of 10X trans-reaction buffer was added to initiate the cleavage reaction. All reactions were carried out in vessels coated with PAM oil to prevent material aggregation. Mineral oil was used to cover the reaction to minimize the effect of evaporation and condensation. At certain time points, 1.5 µL of sample was removed from reaction and added into 10 µL of gel loading buffer. Reaction products were separated by denaturing 20% PAGE, then visualized and quantified using PhosphorImager. Values of vobs were calculated by fitting a minimum of five data points obtained over the first 10%-20% of cleavage reaction to linear least square regression.  Characterization of kcat and KM with the optimal 19 nt all-RNA (ON 7). These assays were carried out under multiple-turnover conditions in 1X trans-cleavage buffer containing 0.5 mM MgCl2, and 150 mM KCl at both 30°C and 37°C. For assays at 30°C, the DNAzyme concentration was 5 nM, and substrate concentrations were 25, 50, 100, 300, 500, 1000, 2000, and 4000 nM, respectively. For reactions at 37°C, the DNAzyme was used at 10 nM, and substrate concentrations were 100, 300, 500, 1000, 2000, 4000, and 8000 nM, respectively. Values of kcat and KM were determined by fitting obtained values of vobs and corresponding substrate concentrations to Michaelis-Menten equation (Equation 3.3).   153  Characterization of kcat and KM with the chimeric DNA/RNA substrate (ON 11). 5 nM of Dz7-38-32t, and the chimeric DNA/RNA substrate with concentrations of 50, 100, 300, 500, 1000, 2000, and 4000 nM were used into multiple-turnover assays in 1X trans-cleavage buffer containing 0.5 mM MgCl2, and 150 mM KCl at 30°C, respectively. Values of kcat and KM were calculated by fitting the obtained values of vobs and corresponding substrate concentrations to Michaelis-Menten equation (Equation 3.3).   Analysis of cleavage product. 20 µL of trans-cleavage reaction containing 5 nM Dz7-38-32t, 1 µM all-RNA substrate with 3’-T12 tail (ON 15), and 2 µL of 10X trans-cleavage buffer was set up following “General protocol for trans-cleavage reactions” described above. 2 µL of reaction mixture was removed into 10 µL of gel loading buffer before incubation at 30°C. To produce enough amount of cleaved substrate, the reaction was incubated at 30°C overnight. Again, 2 µL of the reaction mixture was removed into 10 µL of gel loading buffer. Then, another 6 µL of the trans-cleavage mixture was mixed with 3 µL of 10X polynucleotide kinase buffer, 1 µL of γ-32P-ATP (~10 µCi), 2 µL of polynucleotide kinase, and 18 µL of dH2O. The resulting 30 µL reaction mixture was incubated at 37°C for 3 h followed by deactivation at 65°C for 20 min. G-25 spin column was used to remove salts and unreacted γ-32P-ATP. 10 µL of the G-25 column eluent was added into 10 µL of gel loading buffer. All the three samples were then resolved on a 20% small PAGE gel and visualized by PhosphorImager.   Analysis of pH dependence. The pH dependence of Dz7-38-32t was studied under multiple-turnover conditions in the presence 7.5 nM Dz7-38-32t and 1500 nM substrate (ON 7) in 1X pH variant buffers (50 mM sodium cacodylate: pH 6, 6.5, and 7; 50 mM 154  sodium phosphate: pH 7.5, 8 and 8.5) containing 0.5 mM MgCl2 and 150 mM KCl at 30°C, respectively. The reaction progression was sampled at 8 different time points (10, 20, 40, 60, 90, 120, 240 and 1260 min) through an overnight reaction period, and quantified by PhosphorImager after small denaturing PAGE (20%) electrophoresis. Values of kobs for each pH was calculated by fitting at least five data collected for the fist 10-20% cleavage reaction into linear least square regression as aforementioned.    M2+ effects on cleavage activity. To evaluate the effect of different M2+s on cleavage reaction, 2 µL of 10X M2+-free trans-cleavage buffer and 2 µL of the tested M2+ stock solution were added to 16 µL of the snap-cooled reaction mixture containing 5 nM Dz7-38-32t and 1 µM substrate (ON 7), and reactions were initiated by incubation at 30°C. Each reaction progression was sampled at 7 or 8 different time points through an overnight reaction period. All samples were analyzed on small denaturing PAGE gels (20%) and quantified by PhosphorImager.  Other assays for substrate specificity and so on. Other assays including the attempt to increase the catalytic efficiency at 37°C by lengthening substrate-recognition arms in 3.3.3, probing the function of 8-histaminyl-deoxyadenosine in the 3’ substrate-recognition arm in 3.3.4, and study the effect of varying the RNA bases at the cleavage site in 3.4 were all carried out under multiple-turnover conditions in the presence of 5 nM DNAzyme and 1 µM tested substrate in 1X trans-cleavage buffer containing 0.5 mM Mg2+ following the procedures described in “General protocol for trans-cleavage reactions” without otherwise mentioning.     155  Chapter 4: Image Based Analysis of Spontaneous Cellular Uptake of Dz7-38-32t  4.1 Introduction  Natural oligonucleotides do not efficiently enter most types of cultured cells when added alone.202 Commercially available cell penetrating peptides (CPPs),263 cationic lipids,264,265 polycationic dendrimers,266 or polyethyleneimine (PEI)267,268 are usually necessary to achieve a useful level of uptake of oligonucleotides. Chemically modified oligonucleotides with cationic amines203 and guanidinium ions183 exhibit both increased affinity for target complementary strands and improved mammalian cell membrane penetration with no report of toxicity. Moreover, only four terminal guanidinium functionalized dTs appended at the 3’-end of a 24 nt oligonucleotide facilitated its spontaneous cellular uptake by HeLa cells.183 The guanidinium functionality has been implicated for its central role in the “mysterious” cell membrane penetration activity of CPPs, and an effect that is informally referred to as “arginine magic”. The very representative CPP, HIV-1 TAT peptide, is composed of nine amino acids, of which six are arginines, two are lysines, and one is glutamine without positive charge269 (Figure 4.1A). The RNase A-mimicking DNAzymes containing a number of guaninidium and amine groups isolated and characterized in Chapter 2 and more specifically in Chapter 3, represented by Dz7-38-32t (Figure 4.1B), mimic TAT in terms of the guanidinium functionality. Therefore, it would be interesting to see how these DNAzymes behave 156  with respect to spontaneous cellular uptake. To probe this, spontaneous cellular uptake research of rhodamine-labeled Dz7-38-32t on HeLa cells using confocal fluorescence microscopy was assessed. At the current stage, full-length Dz7-38-32 made on solid-phase by coupling of modified phosphoramidites is still unavailable, and only very limited amounts of DNAzyme are available by enzymatic polymerization for this research. Therefore, all the assays were performed on 384-well glass bottomed cell culture plates.       157   Figure 4.1 The structures of CPP and Dz7-38-32.  (A) The HIV-1 TAT peptide. (B) Chemically modified Dz7-38-32 (top panel), and the constituting modified nucleosides (bottom panel).  All the modified nucleosides are in blue.  H3N NHHNNHHNNH OHNONH OHNOONHNH2H2NONH3OOHNHN NH2ONHNH2H2NONH3H2N ONHN2HH2NNHNH2H2NNH3AG   A   C   A    A    C   C   G-5'U1UACA5GU G U G GU10AUCG15 AUGGC3'-C   C   G   C   A   C   GA35GGAU39A45A21CGUUGC30GG25UGGAGG42UA CA  =NNNNNH2HNNHNOOOU  =NHN OONH2NNH2OOONN ONH2H2NOOOC   =B158  4.2 Preparation of Rhodamine-labeled Dz7-38-32t To prepare the fluorescent Dz7-38-32t, a 5’-rhodamine labeled primer containing three consecutive 2’-O-methyl nucleosides at the 5’-end conferring nuclease resistance (Figure 4.2A) was used in the template-directed primer extension reaction.123 Following the same scheme and procedures for enzymatically synthesizing the trans-acting Dz7-38-32t via primer extension, a 5’-biotinylated synthetic DNA template was hybridized to the 5’-rhodamine labeled primer, and Sequenase v2.0 was subsequently used to incorporate dAimTP, dCaaTP and dUgaTP along with dGTP or a set of all the four natural dNTPs to make modified and unmodified counterparts of Dz7-38-32t, heretofore referred to as Dz7-38-32t-m and Dz7-38-32t-u, respectively. The enzymatically polymerized strand was recovered by treating the heteroduplex with 0.1 M NaOH followed by immediate neutralization with 0.1 M HCl to a pH of ~7.5  (Figure 4.2B). To produce suitable amounts of material, a template-directed primer extension reaction of large scale containing 300 pmole of synthetic template and rhodamine labeled primer along with all other proportionally scaled reagents was carried out in a 400 µl of reaction. Based on previously observed yields ranging from 13.5%-27% in small-scale primer extension reactions used to produce modified trans-acting Dz7-38-32t-m (Enzymatic synthesis of Dz7-38-32t, 3.8, Chapter 3), the resulting fluorescent DNA was estimated to be in the range of ~40 to 80 pmole, which corresponded to a concentration of 5 µM to 10 µM in 8 µl of H2O. In addition to Dz7-38-32t-m and Dz7-38-32t-u, a 5’-rhodamine labeled unmodified Dz7-38-32t-u also containing three consecutive 2’-O-methyl nucleotides at the 5’-end, ordered commercially from IDT 159  (referred to as synthetic-Dz7-38-32t-u), and the 5’-rhodamine labeled primer for transcribing Dz7-38-32t-m and Dz7-38-32t-u (denoting Rhodamine-Pri) were also employed as controls.   Figure 4.2 Preparation of fluorescent Dz7-38-32t.  (A) 5’-rhodamine labeled primer for template-directed transcription of Dz7-38-32t-m and Dz7-38-32t-u. (B) Schematic representation of making 5’-rhodamine labeled Dz7-38-32t-m and Dz7-38-32t-u.    Strand separation (0.1 M NaOH)+ Neutralization (0.1 M HCl)BTranscription (Sequenase V2.0)(2)(1)(dAimTP, dCaaTP, dUgaTP,dGTP)or (dATP, dCTP, dTTP,dGTP)B3'-RR3'-R -3'-3'O NSO3SONH ONNNN3OOPOOOmGmCmCTGCGCCAACAG-3'AB160  4.3 Cellular Uptake Assay HeLa cells were seeded on to 384-well plates 48 h before treatments in order to get tight enough cell adherence at the plate bottom. A confluency of ~50% to 60% at the moment of treatment was preferable, which was determined by the subsequent long-time (~48 h) cell culture in the presence of tested oligonucleotides to allow enough time for uptake. After removing the original cell culture medium, two concentrations (0.9 µM and 1.35 µM) of each Dz7-38-32t-m, Dz7-38-32t-u, synthetic-Dz7-38-32t-u, and Rhodamine-Pri dissolved in Opti-MEM were applied to cells in different wells, respectively. Cell morphology changes, along with what appeared to be extensive cell death was observed for cells cultured in Opti-MEM containing 1.3 µM Dz7-38-32t-m after 23 h, but not for cells cultured in the presence of the same concentrations of other control oligonucleotides, suggesting some cytotoxicity of the modified Dz7-38-32t-m. Bright-field cell pictures taken using Olympus IX70 equipped with a 40X/0.60 objective over a period of 45 h culture time suggested cytotoxicity caused by cell membrane perturbation (indicated by arrows in Figure 4.2). The observed cytotoxicity here was unusual compared to the low toxicities displayed by CPPs,270 some of which displayed no significant cytotoxicity even at concentrations up to 100 µM.271 Moreover, in a parallel assay for addressing the effect of 5-guanidinoallyl-dU on the cellular uptake of Dz10-23, Dz10-23 constructs (39 nt) containing different numbers (2 to 6) of terminal 5-guanidinoallyl-dUs did not display obvious toxicity on HeLa cells with concentrations up to 8 µM through a 48 h of incubation time (data not shown). Therefore, it is most likely that the high density of modifications in Dz7-32-38t-m, and its relatively large size (70 161  nt) led to the observed high toxicity. Subsequent confocal imaging process was discontinued for all the cells treated with 1.35 µM of oligonucleotides.   Figure 4.3 Bright field images of cells treated with 1.35 µM of Dz7-38-32t-m at increasing time points up to 45 h of incubation time.  Cell morphology change and cell death are indicated by arrows. Scale bar: 25 µm. Dz7-38-32t-m 0 min Dz7-38-32t-m 130 minDz7-38-32t-m 300 min Dz7-38-32t-m 23 hDz7-38-32t-m 27 h Dz7-38-32t-m 45 h162  A slight change in cell morphology with less than 10% cell death was also observed for cells treated in Opti-MEM with 0.9 µM of Dz7-38-32t-m over a period of 45 h. All the treated cells were then washed with Dulbecco’s phosphate buffered saline (DPBS) after removing oligonucleotide containing Opti-MEM, and fixed with 4% formaldehyde in DPBS before proceeding to image with Perkin Elmer Spinning disc confocal microscopy in BioImaging, University of British Columbia.    Confocal images revealed remarkable differences in the distribution of fluorescent signals between cells treated with the various Dz7-38-32 constructs  (Figure 4.4). It seemed that substantial amount of Dz7-38-32t-m was taken up into cytoplasm by HeLa cells, the pattern of which showed a predominantly punctate appearance. This observation indicates that most of the oligonucleotides that entered the cell might be still trapped in endosomes as is the case for most internalized nucleic acids,272 CPPs,209 and other transfection reagents in complex with cargos.273,274 No visible nuclear entry was observed based upon the spatial localization of the fluorescent signals on the merged cell picture (Figure 4.4, panel A). Although nuclear uptake was not observed, apparent cytosolic uptake was evident and in contrast to the very weak fluorescent signal that was obtained from the cells treated with Dz7-38-32t-u, with no obvious signal distribution in the cytoplasm (Figure 4.4, panel B). This control essentially helped to rule out the possibility that the spontaneous internalization of Dz7-38-32t-m was caused by factors introduced in the process of its enzymatic preparation and subsequent purification, since these two samples were made from the same transcription and purification steps. The Rhodamine-Pri (Figure 4.4, panel C) and synthetic-Dz7-38-32t-u 163  (Figure 4.4, panel D) made by IDT showed strong accumulation on cell membranes, presumably due to those oligonucleotides that still aggregated on the outer cell membranes after all the wash steps. This is a consequence of the very gentle washes applied to cells using needle controlled by vacuum to reduce cell loss, which is a constraint imposed by using 384-well plate for cell culture. Yet as expected, no apparent uptake into cells was observed.   164   Figure 4.4 Confocal microscopy images of HeLa cells treated with 0.9 µM of rhodamine-labeled Dz7-38-32t of various constructs: fluorescent channel (left), transmitted light (middle), and overlay (right).  (A) Cells treated with Dz7-38-32t-m. (B) Cells treated with Dz7-38-32t-u. (C) Cells treated with Rhodamine-Pri. (D) Cells treated with synthetic-Dz7-38-32t-u. The figures show Z-projection of five central planes in the Z-axis. Scale bar: 20 µm. 165  4.4 Discussions and Conclusion  As hypothesized, Dz7-38-32t-m at a concentration of ~0.9 µM spontaneously enters the cytoplasm of HeLa cells after ~45 h of incubation, most probably because of its “TAT-mimicking” functionalities presented by dUga and dCaa in the sequence. Other reports addressing spontaneous cellular uptake of oligonucleotides bearing grafted cationic amine203 and guanidinium183 functionalities observed uptakes with oligonucleotide concentrations of as high as 8 µM and 4 µM, respectively. This suggests a very strong membrane penetrating ability of these RNase A-mimicking modified DNAzymes. The noted toxicity of Dz7-38-32t-m at concentrations above 1.3 µM would limit the concentration of Dz7-38-32t-m that can be administrated to cells, which would in turn affect the amount of Dz7-38-32t-m that is eventually taken up into cytoplasm by cells. Nonetheless, low nanomolar concentrations of antisense single stranded oligonucleotides (ASOs) taken up by endocytosis were shown to reduce target RNAs within hours in hepatocyte cell lines.275 In terms of the stability of Dz7-38-32t-m under physiological context, albeit no enhanced plasma and serum stability of Dz7-38-32 has been experimentally confirmed so far, it is reasonable to expect that the modifications will block it from degradation by at least some nucleases, since other chemically modified oligonucleotides have been demonstrated to have improved nuclease protection properties.276,277 Once intracellular, even natural ASOs exhibit long half-lives (2-4 weeks) and prolonged activity in suppressing or altering expression of their target RNAs.278 Taking all these facts together, Dz7-38-32t-m holds high promise for mRNA 166  disruption inside cells if given appropriate administration with optimal dosage to minimize its cytotoxicity.  Limited by the amount of chemically modified Dz7-38-32t-m that can be provided by enzymatic polymerization of chemically modified dAimTP, dCaaTP, dUgaTP, together with dGTP, systematic studies including concentration effects on both cytotoxicity and cellular uptake efficiency, and a profile of internalization as a function of incubation time cannot be carried out at the moment. To identify the dependence of the observed cellular uptake on the modifications, different versions of Dz7-38-32 containing one or two kind(s) of modified nucleotide(s) replaced by their natural counterparts were still needed. Additionally, to rule out the possibility that the spontaneous internalization observed was not just attributed to the specific sequence of Dz7-38-32t-m, scrambled variants of Dz7-38-32t-m or similar RNase A-mimicking DNAzymes from other families must be tested as controls. Apparently, questions including the spontaneous internalization activity on cell lines other than HeLa and the target RNA disruption activity inside cells after internalization still need to be addressed in the future experiments.  In summary, I demonstrated that Dz7-38-32t-m can readily penetrate the cell membrane of HeLa cells yet at higher concentrations, it manifests noted cytotoxicity. The preliminary data here provide a possible solution to the problem of cellular uptake, one of the important factors that must be solved for eventual therapeutic applications of DNAzymes as mentioned in 1.6, Chapter 1. It also gives rise to a possibility of using 167  these modified oligonucleotides as transfection reagents to transport other cargos across cell membranes.  4.5 Experimental  4.5.1 General reagents and material for cell culture, enzymes, and oligonucleotides  Dulbecco’s Modified Eagle’s medium (DMEM, 4.5 g/L D-glucose, 4.0 mM L-glutamine), Opti-MEM, trypsin (0.25%, 1X), Fetal Bovine Serum (FBS), Dulbecco’s Phosphate buffered Saline (DPBS, 0.5 mM Mg2+, 0.9 mM Ca2+), and 100X Antibiotic-Antimycotic solution were obtained from ThermoFisher Scientific.  Corning® 384-well high content imaging glass bottom microplates were obtained from Corning. Sequenase v2.0 and pyrophosphatase were purchased from Affymetrix. Streptavidin magnetic particles were purchased from Roche. Sephadex G25 resin was obtained from GE. Oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA), and purified by denaturing (7 M urea) PAGE gels. 4.5.2 Preparation of Dz7-38-32t-m and Dz7-38-32t-u Buffers and solutions.  1X Sequenase buffer: 40 mM Tris-HCl (pH 8.0), 50 mM NaCl, 20 mM MgCl2. 5X modified nucleotides cocktail: 250 µM dAimTP, 125 µM dCaaTP, 50 µM dUgaTP, 50 µM dGTP. 5X unmodified nucleotides cocktail: 50 µM of each dATP, dCTP, dTTP, and dGTP.  168  Wash buffer (TEN): 50 mM Tris-HCl (pH 7.5), 200 mM NaCl, 1 mM EDTA.  Oligonucleotides (5’ to 3’, ON).  Template for transcribing Dz7-38-32t-m and Dz7-38-32t-u, ON 1:  Biotin-GCGCGGCGTGCTGTACCATCCTGACGCTCCACACGTCCCAATCGATA CCACTGTAACTGTTGGCGCAGGC. 5’-rhodamine labeled elongation primer (Rhodamine-Pri), ON 2:  Rhodamine-mGmCmCTGCGCCAACAG.  Unmodified synthetic-Dz7-38-32t-u made by IDT, ON 3:  Rhodamine-mGmCmCTGCGCCAACAGTTACAGTGGTATCGATTGGGACGTG TGGAGCGTCAGGATGGTACAGCACGCCGCGC.  “m” designates 2’-O-methyl modification in ON 2 and ON 3. Transcription of Dz7-38-32t-m. The annealing reaction for template-directed primer extension reaction was carried out in two 1.5 ml eppendorf tubes. In each tube, 150 pmole of template (ON 1) and equal amount of rhodamine labeled primer (ON 2), 80 µL of 5X Sequenase buffer, and H2O were combined to a volume of 280 µL, and heated to above 95°C for 5 min. The solution was then left at room temperature to slowly cool down to room temperature to anneal the template and primer. Following the annealing, reagents including 80 µL of 5X modified dNTP cocktail (250 µM dAimTP 125 µM dCaaTP, 50 µM dUgaTP, and 50 µM dGTP), 20 µL of DTT (100 mM), 10 µL of pyrophosphatase, and 10 µL of Sequenase v2.0 were added sequentially to the annealing solution to 169  make a final reaction volume of 400 µL for each reaction. The reaction mixture was gently mixed well, and then split into 80 µL aliquots into five PCR tubes. The solution in each tube was covered with 20 µL of mineral oil on top. The polymerization reaction was carried out at 32°C for 4 h.  Transcription of unmodified Dz7-38-32t-u. The same procedures and reaction scale were followed for transcribing the unmodified Dz7-38-32t-u. After annealing, 80 µL of 5X unmodified dNTP cocktail (50 µM of each dATP, dCTP, dTTP, and dGTP) instead of the same amount of 5X modified dNTP cocktail along with all other reagents and enzymes were added into the 280 µL annealing reaction. Then the reaction mixture was also split into PCR tubes (80 µL per tube), covered with 20 µL mineral oil, and incubated at 32°C for 4 h. Recovery of Dz7-38-32t-m and Dz7-38-32t-u from the 5’-biotinylated synthetic template. 200 µL of streptavidin coated magnetic beads was used for each 400 µL of transcription reaction. The beads were washed with wash buffer for three times (200 µL per wash). The transcribed DNAzyme strand in duplex with the synthetic template was immobilized on streptavidin coated magnetic beads through the 5’-biotin moiety on the template after 30 min of incubation at room temperature. Following the magnetization of the beads, the supernatant solution was decanted, and the beads were washed for two times with wash buffer (200 µL per wash). The DNAzyme strand was then recovered with 60 µL of 0.1 M NaOH followed by immediate neutralization with 0.1 M HCl to obtain a final pH of ~7.5. The resulting DNAzyme solution was spun down through G-25 column to desalt. Pinkish DNA pellet was obtained by drying all the solution down on a 170  speed-vac, which was subsequently re-suspended in 8 µL of autoclaved H2O.  4.5.3 Cellular uptake Cell culture. HeLa cells were maintained and cultivated in Dulbecco’s Modified Eagle’s medium (supplemented with 10% FBS and 1X Antibiotic-Antimycotic) in 37ºC cell culture incubator (5% CO2 and 95% humidity). The cells were seeded at a density of 2x103 cells per well in a 200 µm glass bottom 384-well plate 48 h before treatments in order to reach a confluency of 50-60% and a tight adherence to the plate bottom at the moment of treatment with oligonucleotides.  Cellular uptake. The rhodamine-labeled oligonucleotides were diluted to desired concentrations (0.9 µM and 1.3 µM) in warm Opti-MEM to a final volume of 20 µL. For Dz7-38-32t-m and Dz7-38-32t-u, which were made by enzymatic polymerization and separated from the template binding to streptavidin-coated magnetic beads, residual amount of magnetic beads contamination might exit in the re-suspended solution. To minimize possible magnetic beads contamination of Opti-MEM, which might cause cell aggregation279, an extra magnetization step was applied to the re-suspended DNAzyme solutions before carefully pipetting the desired amount of supernatant into warm Opti-MEM. Then the original cell culture medium was carefully removed using needle connected with vacuum. The warm Opti-MEM containing different oligonucleotides was gently added to cells using micropipette equipped with a sharp gel-loading tip to reduce cell detachment. After 48 h of incubation, the cells were processed for the microscopy as following. Firstly, the oligonucleotides containing Opti-MEM was removed, and cells were washed twice with 30 µL of warm DPBS (containing 0.5 mM Mg2+ and 0.9 mM 171  Ca2+). Then cells were fixed by incubating in 30 µL of 4% formaldehyde in DPBS at room temperature for 20 min. After removing fixation solution, cells were washed with 30 µL of warm DPBS for two more times. The cells were left in DPBS and imaged immediately. Bright field imaging. Bright field cell pictures were taken on an inverted Olympus IX70 microscope equipped with a 40X/0.60 objective and an Olympus U-BMAD CCD camera in BioService Center, Chemistry Department, University of British Columbia. Confocal imaging. Confocal imaging was performed on an inverted Perkin Elmer Ultraview VoX Spinning Disk Confocal Microscope with a 63X glycerol objective in BioImaging, University of British Columbia. Bottom-to-top Z stack of 80-100 steps with 0.2 µm per step were scanned and captured by the equipped Hamamatsu 9100-02 electron multiplier CCD camera. The cell pictures in Figure 4.4 were produced by Z-projection of 5 central planes on the ImageJ.   	172  Chapter 5: A Systematic Study of Constraints on In Vitro Selection Imposed by Modified Nucleosides Histaminyl-dA, Guanidinoallyl-dU, and Aminoallyl-dC  5.1 Introduction  The importance of the three chemically modified nucleotides 8-histaminyl-dATP (dAimTP), 5-guanidinoallyl-dUTP (dUgaTP), and 5-aminoallyl-dCTP (dCaaTP) has been demonstrated in the previous reports of the Perrin lab185,186 and most recently in Chapter 2 and Chapter 3 of this thesis with the selection of several M2+-independent RNase A-mimicking DNAzymes with rate constants up to 3 orders of magnitude over unmodified catalysts. The catalytic roles of the appended imidazole groups and the cationic amines have been experimentally identified for Dz925-11.177 Additionally, oligonucleotides containing these modified nucleosides displayed spontaneous cell membrane penetration activity, as demonstrated using Dz7-38-32t in Chapter 4. Parallel studies using two different modified nucleotides, one with imidazole and the other one with cationic amine, in in vitro selection also led to the isolation of a M2+-independent RNA-cleaving DNAzyme.179 Similarly, it was shown that for aptamers, modifications afford both qualitative improvements in terms of an increased range of targets capable of recognition, and quantitative improvements in terms of higher binding affinity.280 These recent reports featured selections with a single modified dUTP that was appended with a phenyl or indole group, each of which provides a “hydrophobic” 173  effect for folding as well as specific hydrophobic interactions against several targets.281 Collectively, these studies demonstrate the practicability of using modified nucleotide analogues in in vitro selections, and more importantly highlight their critical contributions to improved catalytic abilities of modified DNAzymes and high binding affinities of modified aptamers by making use of the protein-like functionalities that play either catalytic or structural roles, or both.   Four prerequisites must be met for using synthetic nucleotides with appended functionalities in in vitro selection, including (1) substrate ability for at least one DNA polymerase, (2) capability for replacing unmodified counterparts at every position specified by the template sequence, (3) retention of base-pairing capacity during the iterative cycling of information transfer, and (4) capability of transferring genetic information to unmodified cDNA complements during each round of PCR amplification.173,174 In other words, efficient transcription and reverse transcription steps, which provide faithful inter-conversions between unmodified genotypic DNAs and modified phenotypic DNAs over iterative rounds of selections, are critical factors to the success of an in vitro selection.  Natural polymerases, even though many of them have been shown to accept some components of unnatural genetic systems, have limited capacity to accept just a few, or even single, or nonadjacent unnatural nucleotides214,221, the consequence of which is biased library production with compromised “sequence space” in in vitro selection. In addition, all synthetic systems examined to date suffer unidirectional losses when amplifying unnatural components.222 The successful selections simultaneously using 174  dAimTP, dCaaTP and dUgaTP support the notion that the genetic information stored in the modified DNAs can be reversely transcribed back to unmodified cDNAs with high fidelity such that sequences with the desired phenotype are enriched, yet no experimental data have been collected in regards to the base pairing properties of these three modified bases to directly answer the extent to which they change the physiochemical and biological properties of DNA as genetic informational molecules when appearing in DNA sequences. To address these critical yet unclear questions and provide insight for people wishing to perform similar selections with the three modified dNTPs, this chapter studied their impact on duplex stability and their respective base pairing properties, and the template properties of modified oligonucleotides along with an assessment of the mutagenicity of these modifications. To perform these studies, oligonucleotides based on the sequence of a RNA-cleaving DNAzyme previously identified from a library consisting of 30 degenerate positions, were synthesized on the solid-phase with either a single histaminyl-dA (ON 1), or all three modified nucleosides (ON 2) using the corresponding phosphoramidites that previously reported by the Perrin lab166,175,282, or no modified nucleoside (ON 3). These are shown in Figure 5.1.       175   Figure 5.1 Discretely modified oligonucleotides containing modified nucleoside(s). (A) Phosphoramidite structures of 8-histaminyl-deoxyadenosine (dAim, 1), 5-aminoallyl-deoxycytidine (dCaa, 2), and 5-guanidinoallyl-deoxyuridine (dUga, 3), respectively. (B) Sequences and constructs of ON 1, ON 2, and ON 3. ON 1 contains only dAim; ON 2 contains a set of dAim, dCaa, and dUga; ON 3 contains no modified nucleosides. “Cm” designates 2’-OMe ribocytidine, and “TInv” represents an inverted deoxythymidine at the 3’-end of the oligonucleotide.  5.2 Impact of Modified Nucleosides on Duplex Thermal Stability  The oligonucleotides were thermally denatured in the presence of a complementary sequence (ON 4) at buffered pH 6 and 8.5 at which the imidazole (pKa 7.4) is either fully protonated or nearly entirely deprotonated (89%). Measurement at pH 7.5, which is very near the pKa of the imidazolinium ion, was deliberately avoided as such data would be difficult to interpret (Figure 5.2 A).   NNNNHNOODMTOHNNHNHNNODMTOOONHNHNNNODMTOOHNNHBzPNO CNOPNO CNO ONCOONCF3COOPNO CNBz1 2 35'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3'5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGT   C   TTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 1ON 2AB5'-CmCmCmTGCGGAGGGGCTGCCAGTA    GT   C   TTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3176   Figure 5.2 Impact of 8-histaminyl-dA on duplex thermal stability. (A) Oligonucleotide sequences used for thermal denaturation study. ON 1, ON 2, and ON 3 in duplex with a complementary strand ON 4 for thermal melting studies, with the modified pair(s) positioned close to the center. (B) Watson-Crick base pairing (1) and Hoogsteen base pairing (2) formed between dAim and dT.    The presence of a single dAim residue lowered duplex stability by 0.9 ˚C and 1.7 ˚C at pH 6 and 8.5, respectively, when compared to the all-natural duplex (melting was assessed with ON 4). The greater destabilization at pH 8.5 likely reflects the loss of a positive charge on the imidazole group.  The presence of dCaa and dUga in ON 2 increased the stability of duplexes by 2˚C compared with that observed for ON 1 when measured under the same conditions. This is also consistent with several reports on grafted cationic amine species in 5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGT   C   TTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 15'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 25'-CmCmCmTGCGGAGGGGCTGCCAGTA    GT    C  TTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 33'-GACGGTCAT   CA   G   AACACC-5' ON 4ANNNNNNNHNHHHNNOOHBNNNNNHNNHNNNOOHHH1 Watson-Crick 2 Hoogsteenanti anti syn anti177  oligonucleotides enhance their duplex stability.183,203,282,283 All the acquired thermal melting temperatures (Tm) are summarized in Table 5.1. Table 5.1 Thermal melting temperatures (Tm) of duplexes a formed by ON 1, ON 2, and ON3 with the complementary ON 4 in pH variant buffers b, respectively  Entry ON 1 (dAim) ON 2 (dAim, dCaa, dUga) ON 3 (no modification) pH 6.0 8.5 6.0 8.5 6.0 8.5 Tm (°C) 60.5 59.6 62.5 61.6 61.4 61.3 ∆Tm (°C) -0.9 -0.9 -0.1 a Duplex concentrations were 1 uM, with 1:1 ratio of each of the two complementary strands. b 1X pH variance buffers: 10 mM sodium phosphate (pH 6 and pH 8.5), 150 mM KCl.  Absorbance was measured at 260 nm. Numerical data were based on the average of three separate experiments of forward and reverse melting temperatures.  These preliminary melting experiments demonstrate that the 8-histaminyl-dA is not particularly destabilizing, suggesting that canonical Watson-Crick base pair depicted in Figure 5.2B (1) still formed between dAim and dT, with trivial local distortion introduced by the bulky alkylimidazole appendage which led to destabilization of duplex by less than 2˚C under tested conditions above. This observation likely excludes the formation of dAim-dT Hoogsteen pair depicted in Figure 5.2B (2) due to the presence of C8-NH proton, which would cause significant loss of duplex stability. When an aminoallyl-dC and a guanidinoallyl-dU are added, the melting temperature is increased as might be expected. 178  5.3 Base Pairing Properties of Modified Bases  To address the respective base-pairing properties of dAim, dCaa, and dUga, single-nucleotide insertion assays were carried out using Vent (exo-) DNA polymerase, the polymerase that previously afforded robust PCR amplification of modified DNA into unmodified cDNA based on the experience from the previous success in in vitro selections. Briefly, pre-hybridized primer-template duplexes were prepared to test the effects of the modified nucleoside contained in the template.  Several primers (ON 5, 6 and 7) were chosen to address incorporation opposite template dAim, dUga, and dCaa, respectively. Vent (exo-) DNA polymerase was added to primer extension reactions pre-heated to 72˚C, a temperature below the melting temperature of the primer-template heteroduplex. Standing-start single-nucleotide insertion was initiated immediately after the addition of different dNTPs. All the primer extension assays were carried out under “non-forcing” conditions (1 µM primer-template duplex, 10 µM dNTP, 1 unit of Vent (exo-) DNA polymerase, at 72˚C) as adapted from previously reported protocols284, so that inefficiently processed base pairs such as mispairs and non-canonical Watson-Crick pairs, would not form or were less likely to form. 5.3.1 Base-pairing properties of dAim In the case ON 1 where dAim is the template base, a 3 min reaction was performed with each nucleotide considering the intrinsic pairing difficulty imposed by the presence of the 8-histaminyl group. In addition to the four natural nucleotides, dUgaTP was also tested for its pairing ability with template dAim. Vent (exo-) DNA polymerase successfully 179  incorporated a dT and dUga vis-à-vis the dAim, however none of the other three dNTPs (dATP, dCTP, or dGTP) functioned as a substrate. Notably, the extent of insertion reaction for dTTP and dUgaTP were about the same with more than 90% completion in 3 min reaction time. Due to the cationic nature of the 5-guanidinium cation on dUgaTP, its corresponding insertion band showed a retarded electrophoretic mobility compared to that obtained with dTTP (Figure 5.3).    Figure 5.3 Base-pairing properties of dAim.  (A) Primer-template duplex used for single-nucleotide insertion assay with the dAim in template (ON 1) just downstream of the 3’-end of a 32 nt primer (ON 5). Unmodified ON 3 is for reference.  (B) Single-nucleotide insertion results on PAGE gel (20%). The band corresponding to 33 nt after one dUgaTP insertion (lane 6) is higher than that observed for dTTP (lane 3) due to the retardation caused by the 5-guanidinoallyl group. The duplex was used at 1 µM, and all the dNTPs were used at 10 µM. The reactions were stopped after 3 min reaction time.  Lane       1         2         3          4        5         6       7         8        9       10       11      12A         T         C         G        Uga_______________________________     dAim____33 nt5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3'3'- CAGAACACCGCAAGCAAACAAGCGCCGCGCGA-5'____32 ntdNTPON 1ON 5dTTP/dUgaTP over dATP, dCTP, and dGTPA         T         C         G        Uga_______________________________     dA____33 nt____33 nt5'-CmCmCmTGCGGAGGGGCTGCCAGTA   GTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3ATemplate baseB180  Furthermore, single-nucleotide insertion assays with reaction time points up to 180 min revealed that even after long incubation times, only less than 5% of dAim-dA mispair happened (lane 14, Figure 5.4), which was highly likely attributable to the “A-rule”285. No dAim-dC, or dAim-dG mispairs formed at all (Figure 5.4). Therefore, dAim did not seem to compromise its base pairing specificity when appearing as a template base in an otherwise unmodified DNA sequence, which is one of the indispensable prerequisites for achieving highly faithful and efficient in vitro selections.   Figure 5.4 Standing-start single-nucleotide insertion assays opposite dAim through 180 min of reaction time.   Lane 1 and lane 10 are 5’-32P labeled standing-start primer ON 5, 32 nt. Lane 14 shows the 33 nt product representing a dAim-dA mispair (less than 5%). The duplex was used at 1 µM, and all the dNTPs were used at 10 µM.  5.3.2 Base-pairing properties of dUga In the case of ON 2, where dUga is the template base, the natural dATP was able to be inserted to form a dUga-dA pair in 1 min reaction time exclusively. Interestingly, no insertion was observed when dAimTP was used as the substrate, irrespective of whether dCTPTime (min) 2 5 30 180 2 5 30 180Lane 1 2 3 4 5 6 7 8 932 nt10 11 12 13 1432 nt33 nt___________________________________________ dGTP_______________________________ dATP_______________________________2 5 30 180dNTP181  the template base is the unmodified dT or the modified dUga, an observation that is consistent with the previous finding that dAimTP is not an as efficient substrate for Vent (exo-) DNA polymerase214 (Figure 5.5).   Figure 5.5 Single-nucleotide insertion assay opposite dUga in template ON 2.  (A) Template-primer duplexes used for single-nucleotide insertion assay. dUga in template (ON 2) is just downstream of the 3’-end of a 30 nt primer (ON 6). Template ON 3 containing no modification was used as a reference. (B) Autoradiograph showing the single-nucleotide insertion results after 1 min reaction time. Lane 1 and lane 7 are 5’-32P labeled ON 6.  5.3.3 Base-pairing properties of dCaa In contrast to the high-fidelity base pairing displayed by template strands containing dAim and dUga, dCaa formed mismatched pairs with both dA and dT under single-nucleotide insertion conditions with unmodified substrates dATP and dTTP. 3'-G  AACACCGCAAGCAAACAAGCGCCGCGCGA-5' ON 6dATP/dAimTP over dGTP, dCTP, and dTTP5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 2Lane       1         2         3          4        5         6        7           8          9        10        11      12A        Aim       T         C       G_______________________________     dUgadNTPTemplate baseA        Aim       T         C       G__________________________     dT____31 nt____30 nt5'-CmCmCmTGCGGAGGGGCTGCCAGTA   GT    C   TTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3AB182  Surprisingly, the mispair with dTTP was also observed with the unmodified template (ON 3) containing dC although at somewhat lower amounts. More strikingly, the formation of the dCaa-dA mispair did not cause polymerase stalling and following incorporation, a downstream dUga-dA pair formed immediately, as evidenced by the predominant appearance of the 31 nt product corresponding to the consecutive insertion of two dAs (lane 2, Figure 5.6B). We cannot exclude the possibility that in the absence of dGTP, the enzyme recognized the n+1 base (in this case the dT or dUga), and inserted a dA. Subsequently, with a dA facing the dC (or dCaa), the polymerase extended the 3’ mismatch with another dA. Indeed, it is nevertheless puzzling that dTTP was well incorporated once against the dCaa, and yet even against the unmodified control, the same effect was seen, albeit at lower amounts. These results are shown in Figure 5.6. Reducing the time of incorporation or reducing the concentration of either dATP or dTTP lowered the overall yield of incorporation but did little to change the relative amounts of incorporated products (data not shown).   183   Figure 5.6 Single-nucleotide insertion assay against dCaa in template ON 2.  (A) Template-primer duplexes used for single-nucleotide insertion assay. dCaa in template (ON 2) is just downstream of the 3’-end of a 29 nt primer (ON 7). Template ON 3 containing no modification was used as a reference. (B) Autoradiograph showing the single-nucleotide insertion results after 1 min reaction time. Lane 1, 5’-32P labeled ON 7 (29 nt). Lane 10, 5’-32P labeled ON 6 (30 nt) as size marker for one nucleotide insertion.    Whereas the propensity to form mispairs with the template dCaa was notable, this was also observed with the unmodified template albeit to a lower extent.  Such misinsertion is likely to reflect an artifact of incorporation assays that are run under single-nucleotide incorporation conditions, which are thermodynamically favored by the release of pyrophosphates.286 When given a full complement of substrate dNTPs, the insertion of the correct dNTP would be likely to prevail over any misinsertions, a point that was addressed subsequently (vide infra).  Lane       1         2         3          4        5          6         7          8         9        10___________________     dCaadNTPTemplate baseA         T         C         G___________________A         T         C         G     dC____31nt____29 nt____30 nt____30 nt3'-AACACCGCAAGCAAACAAGCGCCGCGCGA-5' ON 7dGTP over dATP, dCTP, and dTTP5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 25'-CmCmCmTGCGGAGGGGCTGCCAGTA   GT    C   TTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3AB184  5.4 Full-Strand Synthesis Continuing Beyond Multiple Modified Pairs Efficient nucleotide insertion does not guarantee successful replication since the newly formed modified pairs may, to some extent, be similar to mismatches and lesions to DNA polymerases that normally cause polymerase stalling and consequent pause or termination of strand elongation. Therefore, the ability of Vent (exo-) DNA polymerase to synthesize a strand continuing beyond modified pairs was further tested. To address this issue, Vent (exo_) DNA polymerase was used to elongate ON 7 (29 nt) along ON 2 in the presence of all four dNTPs (Figure 5.7A). As shown in Figure 5.7B, ON 7 elongated itself along ON 2 with comparable efficiency as that observed for unmodified control (ON 3), and no significant truncation bands were observed.   185   Figure 5.7 Strand synthesis continuing beyond modified base pairs using Vent (exo-) DNA polymerase.  (A) Template-primer duplexes used to evaluate strand synthesis progress. (B) Strand synthesis results through 180 min of reaction time. Lane 1 and lane 7, 5’-32P labeled primer, 29 nt (ON 7). Lane 2 to lane 6, strand synthesis progress on template ON 2. Lane 8 to lane 12, strand synthesis progress on template ON 3. Lane 13, 5’-32P labeled ON 2 as size-marker for full-length product. The observed multiple bands on the top part of this gel picture representing full-length products were attributed to the three consecutive 2’-OMe-rC at the 5’-end of the templates (ON 2 and ON 3).  5.5 Overall Fidelity of a Complete Replication Cycle  To further address the overall fidelity throughout a complete replication/selection cycle, which includes the transcription of an unmodified template into an modified DNAzyme using Sequenase v2.0, followed by the amplification of self-cleaved DNAzyme back into 3'- AACACCGCAAGCAAACAAGCGCCGCGCGA-5'5'-CmCmCmTGCGGAGGGGCTGCCAGTA   GT   C    TTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3ON 2ON 729 nt51 nt5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3'Lane          1             2                3             4            5              6              7               8             9             10                11           12           135                10          30           120           180_______________________________   ON 2Time (min) 5            10            30              120          180________________________   ON 3____29 nt____54 nt54 ntdATP, dTTP, dCTP, dGTPAB186  unmodified cDNA (Figure 5.8), the all-RNA cleaving Dz7-38-32 characterized in Chapter 3 of this thesis was chosen to assess the overall mutagenicity of modified bases. To do this, the synthetic, unmodified DNA template for transcribing Dz7-38-32 was hybridized to a 5’-biotinylated primer containing the all-RNA target sequence. Sequenase v2.0 was used to incorporate dAimTP, dCaaTP and dUgaTP along with dGTP. Following standard template strand removal with NaOH, the self-cleaving strand composed of modified DNA was allowed to self-cleave. The self-cleaved DNAzyme was subsequently amplified into unmodified cDNA which was then subjected to cloning and sequencing.   187   Figure 5.8 Assessment of the overall fidelity throughout a complete replication/selection cycle. (A) Modified catalytic core of Dz7-38-32 made by Sequenase v2.0. (B) Schematic representation of a complete replication cycle constituted by (1) synthetic template directed transcription of modified DNAzyme, (2) catalytic strand separation from unmodified template, (3) and (4) catalytic folding and self-cleavage, and (5) amplification of self-cleavage product using Taq or Vent (exo_) DNA polymerase.  The thick line in red denotes RNA substrate region; the thick line in blue denotes modified DNA region; the thick line in black denotes unmodified DNA region.    A2. Template Strand RemovalB1. Primer Extension (Sequenase V2.0)    dAimTP, dCaaTP, dUgaTP, dGTP4. Self-cleavage5. Amplification (Taq  or Vent (exo-) DNA polymerase)3. FoldingxB5'-TTACAGTGGTATCGATTGGGACGTGTGGAGCGTCAGGATGGTACA-3'B6. Cloning and sequencingOverall fidelity????188  To eliminate the small but nonzero possibility of contamination by the unmodified template, the self-cleaved modified sequences were subjected to 7% denaturing PAGE (7 M urea) purification before PCR amplification (Figure 5.9). Meanwhile, to ensure that the cDNA sequenced was not resulting from amplifying adventitious amounts of unmodified template DNA, which was presumed to be removed with 5 brief NaOH (0.1 M) washes, a “dummy” reaction that included Sequenase v2.0, primer and template but no dNTPs was run. The “dummy” sample was worked up accordingly and a band was eluted from the gel in the place where the modified self-cleavage product normally ran (Figure 5.9).   Figure 5.9 Self-cleavage and purification of Dz7-38-32 and “dummy” control.  (A) Schematic representations of full-length modified DNA (Dz7-38-32), product after self-cleavage reaction, and the size marker for the self-cleavage product generated by treatment of full-length DNAzyme with 0.1 M NaOH. (B) PAGE gel image showing the BSelf-cleavagex______________Dz7-38-32 __________"Dummy" control____Full-length modified            DNA (154 nt)____Self-cleaved modified                 DNA (94 nt)          NaOH  cleaved  modified DNA (86 nt)____Lane 1        2        3              4       50.1 M NaOHLane 1Lane 2Lane 3A B189  DNA bands represented by the schemes in (A). Self-cleavage product in red rectangular in lane 2 and the same position from “dummy” control (lane 5) were sliced off for gel elution and subsequent PCR reaction using either Vent (exo-) or Taq DNA polymerase.    Whereas PCR of the modified DNA gave an intense band by agarose after 25 cycles, the same reaction on the “dummy” sample gave no product in the case of Vent (exo-) (Figure 5.10A), and only a small amount of product that was of lesser size in the case of Taq (Figure 5.10B).    Figure 5.10 PCR assay for unmodified template DNA contaminants.  PCR assay for unmodified template DNA contaminants. PCR analysis confirmed that 0              5              10             15             25_______________________________Dz7-38-32Cycle________________________"Dummy" control0               5             10               15            25____150 bp____100 bp ____75 bp ____82 bp0              5              10             15             25_______________________________Dz7-38-32Cycle________________________"Dummy" control0               5             10               15            25____150 bp____100 bp ____75 bp ____82 bpAB190  correct DNA amplicons (82 bp) were only observed for amplification of self-cleaved Dz7-38-32, but not for the “dummy” control, which ruled out the small but nonzero possibility that the amplicons were resulting from amplification of co-migrating unmodified DNAs. (A) Amplification of self-cleaved Dz7-38-32 using Vent (exo-) DNA polymerase. (B) Amplification of self-cleaved Dz7-38-32 using Taq DNA polymerase.  Confident that the obtained PCR amplicons derived only from the modified DNA, they were ligated into pCR2.1-TOPO vectors following the manufacture’s protocol. The ligated product was transformed into E. coli DH5α, following a normal chemical transformation protocol, cloned, and sequenced. Of 10 individual clones derived from amplicons amplified by Vent (exo-), there were 12 mutations in a total of 450 positions. Notably 8 of these mutations occurred at an unmodified G while two others occurred at a modified A and the other 2 occurred at modified C. Since two-thirds of the mutations occurred at an unmodified G, we attributed this relatively high mutation rate to be due to the lack of a proof-reading (exo) activity or perhaps to nearest-neighbor effects. Gratifyingly, all 14 individual sequences amplified by Taq DNA polymerase revealed an overall fidelity of 100% with no mutations observed through a total of 630 modified positions (Figure 5.11). The fidelity we evidence now provides concrete evidence that these three modified nucleotides maintain hereditary information transfer activity in the in vitro selection cycle, which is one of the characteristic signatures of genetic polymers.  Moreover, it provides an exception to the previous claim that “all synthetic genetic systems examined to date suffer unidirectional losses when amplifying unnatural components”.221,222  191   Figure 5.11 cDNA sequence alignment used to evaluate modified DNA replication fidelity.  cDNA sequence alignment used to evaluate modified DNA replication fidelity of a complete replication cycle. All individual sequences were aligned in reference to the original catalytic sequence. 10 individual sequences derived from amplicons amplified by Vent (exo-), denoted Vent-1 to Vent-10, revealed 12 mutations (highlighted in blue and yellow) through a total of 450 positions in the modified sequence region, only 4 of these mutations were caused by modified nucleosides (highlighted in blue). All 14 individual sequences derived from amplicons amplified by Taq DNA polymerase, denoted Taq-1 to Taq-14, revealed an overall fidelity of 100% with no mutations observed through a total of 630 modified positions.   5.6 Discussions and Conclusion Compared with a reference duplex containing no modified base pairs, the presence of only one dAim-dT pair introduced a maximum of 1.7°C decrease in duplex stability when measured in pH 8.5 phosphate buffer conditions. Marked decrease in duplexes thermal 10 20 30 40Reference T T A C A G T G G T A T C G A T T G G G A C G T G T G G A G C G T C A G G A T G G T A C ATaq-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Taq-14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Vent-1 . . . . . . . . . . . . G . . . . . . . . . . . . . A . . T . . . . . . . T . . . . . . .Vent-2 . . . . . . . . . . . . G . . . . . . . . . . . . . A . . T . . . . . . . T . . . . . . .Vent-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Vent-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Vent-5 . . . . . . . . A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Vent-6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Vent-7 . . . . . . . . A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Vent-8 . . . . . . . . . . . . . . . . . A . . . . . . A . . . . . . . . . . . . . . . . . . . .Vent-9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Vent-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192  stabilities ranging from 13-16°C and 7-9°C have been reported for duplexes containing 8-(acetylarylamino)-dG or 8-(arylamino)-dG adducts under similar conditions with confirmed syn- and anti- conformations, respectively.287 Therefore, it is highly likely that destabilization of this level can be attributed to the trivial distortion caused by the appended bulky alkylimidazole group rather than any significant anti- to syn- conformation change that can lead to the formation of a Hoogsteen base pair. Consistent with many reports on oligonucleotides bearing cationic functionalities, dCaa and dUga introduced thermal stability to duplexes, which would compensate the destabilization effect caused by dAim when they were simultaneously built into an oligonucleotide sequence. Notably, both dAim and dUga maintained perfect base pairing properties with just their complementary nucleotides, respectively. dCaa displayed propensity to form mispairs under single-nucleotide insertion conditions by Vent (exo-) DNA polymerase for unknown reasons. Similar mispairs were also observed for Klenow (exo-) DNA polymerase and human polymerase β (Pol β) mediated single-nucleotide insertion opposite natural dC assays reported by Meier, which has not been fully discussed.287,288 The overall high fidelity for a complete cycle of replication supports the previous hypothesis that when given a full complement of substrate dNTPs, insertion of the correct dNTP would prevail over misinsertions which are thermodynamically favored by the release of pyrophosphate under single-nucleoside incorporation conditions. While we cannot claim that all sequences (with all nearest-neighbor constraints) will be amplified with equal fidelity, the fact that several families of DNAzymes with 20-40 193  degenerate positions have survived selection suggests these sequences represent viable sequence combinations that can, in the presence of a suitable polymerase (Sequenase), be converted from unmodified templates into modified strands which can then be viably amplified by Vent (exo-) or Taq with good or excellent retention of genetic information, respectively.  In summary, the studies here have experimentally demonstrated the physicochemical and biological impact imposed on DNA sequences by dAim, dCaa, and dUga, including duplex thermal stability, base-pairing properties, and the overall transcription and amplification fidelity. Despite the high density of modifications in DNA molecules, exemplified by Dz7-38-32, the assembled modified nucleotides retain the characteristic signature of coding and decoding as genetic polymers. With the grafted functionalities from the side chains of histidine, lysine, and arginine, respectively, there is no doubt that dAimTP, dCaaTP, and dUgaTP, as well as modified DNA molecules containing them can have broad applications in exploring nucleic acids with enlarged functional repertoire and improved performance. The studies here identify certain shortcomings in the use of modified nucleotides while providing insights for those wishing to perform researches with DNA containing the same or similar functionalities. 5.7 Experimental 5.7.1 General  dAimTP  and dUgaTP were synthesized according to literatures.185,214 dCaaTP was obtained from TriLink Bio Technologies. Non-modified oligonucleotides were purchased from Integrated DNA Technologies. Modified oligonucleotides (with protection groups) 194  were synthesized by Dr. Sabatino, at the department of Biochemistry, Seton Hall University. All oligonucleotides were purified by 10-20% denaturing PAGE (7 M urea). Ultrapure dNTPs were obtained from Fermentas. Vent exo (-) DNA polymerase and T4 polynucleotide kinase were obtained from New England Biolabs. Sequenase v2.0 and pyrophosphatase were purchased from Affymetrix. γ-[32P]-ATP and α-[32P]-dGTP were purchased from Perkin Elmer. Sephadex G25 resin was obtained from GE. UV spectrometry was performed on a Beckman Coulter DU800 spectrophotometer. 5.7.2 Oligonucleotides (ON, 5’ to 3’) CmCmCmTGCGGAGGGGCTGCCAGTAimGTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv ON 1, CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCInv ON 2,  CmCmCmTGCGGAGGGGCTGCCAGTAGTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv ON 3,  CCACAAGACTACTGGCAG ON 4,  AGCGCGCCGCGAACAAACGAACGCCACAAGAC ON 5, AGCGCGCCGCGAACAAACGAACGCCACAAG ON 6,  AGCGCGCCGCGAACAAACGAACGCCACAA ON 7, GCGCTCGCGCGGCGTGCTGTACCATCCTGACGCTCCACACGTCCCAATCGATACCACTGTAACTGTTGGCGCAGGCCGACGC ON 8,  195  Biotin-T40CCCGGGTTTTT r(GCGUGCCCGUCUGUUGG)TTTTGCGTCGGCC TGCGCCAACAG ON 9, GCGCTCGCGCGGCGTGC ON 10, GCGTCGGCCTGCGCCAACAG ON 11. In oligonucleotides 1, 2 and 3, “Cm” designates 2’-OMe ribocytidine, and “TInv” represents an inverted deoxythymidine at the 3’-end of the oligonucleotide. The three consecutive 2’-OMe ribocytidine at the 5’-end and the inverted deoxythymidine at the 3’-end were constructed for enhanced stability for cell based studies in other research project. “r” designates a stretch of RNA bases in ON 9. 5.7.3 Buffers/solutions 1 pH variant phosphate buffer for thermal melting study: 10 mM sodium phosphate (pH 6, and 8.5), 150 mM KCl.  2 1X thermopol buffer: 20 mM Tris-HCl (pH 8.8), 10 mM (NH4)2SO4, 10 mM KCl, 2 mM MgSO4, 0.1% Triton® X-100. 3 1X reaction buffer for Sequenase v2.0: 40 mM Tris-HCl (pH 8.0), 50 mM NaCl, 20 mM MgCl2.  4 Wash buffer: 50 mM Tris-HCl (pH 7.5), 200 mM NaCl, 1 mM EDTA.  5 Neutralization buffer: 25 mM sodium cacodylate (pH 6.0), 1 mM EDTA. 6 Cleavage buffer: 50 mM sodium cacodylate (pH 7.4), 200 mM NaCl, 1 mM EDTA. 196  7 Gel elution buffer: 10 mM Tris-HCl, (pH 8.0), 1% LiClO4. 8 Gel loading buffer: formamide, 25 mM EDTA, 0.01% bromophenol blue and 0.01% xylene cyanole. 9 5X modified nucleotides cocktail: 250 µM dAimTP, 125 µM dCaaTP, 50 µM dUgaTP, 50 µM dGTP. 5.7.4 De-protection of oligonucleotides ON 1 and ON 3 were cleaved off the resin and fully de-protected by direct incubation in concentrated ammonium hydroxide (28-30%, wt) at 55°C overnight.166 ON 2 was first treated with piperidine/H2O (1:1, v/v) at room temperature overnight to remove the β-cyanoethoxycarbonyl groups, meanwhile cleave the oligonucleotides off the resin. Then the supernatant was recovered and dried by speed-vac prior to the concentrated ammonium hydroxide (28-30%, wt) treatment as ON 1 and ON 3.282,289 Following lyophilization, all the crude oligonucleotides were dissolved in 10 mM Tris-HCl (pH 8.0) buffer, and then subjected to 10% denaturing PAGE (7 M urea) purification.  5.7.5 Thermal melting temperature measurement  Thermal melting temperatures were measured on a Beckman Coulter DU800 spectrophotometer using its “Experimental Tm Analysis” program. Oligonucleotides 1, 2, and 3 at a final concentration of 1 µM were heated up to 95°C with ON 4 at one-to-one ratio in 1X different pH buffer conditions, respectively, and then allowed to cool down to room temperature slowly to form duplexes, prior to the thermal melting temperature measurement. Samples were monitored for absorbance at 260 nm from 20-85°C at a 197  rate of 1°C/min with data points at every 1°C. To minimize the buffer concentration change caused by evaporation during the process of thermal melting, mineral oil was applied on top of the solution inside the Teflon-stopped 1 cm path length quartz microcell. Three separate experiments of forward and reverse melting temperatures were obtained for each buffer pH. The final reported Tm for each pH was the average of three separate experiments of forward and reverse melting temperatures.  5.7.6 Standing-start single-nucleotide insertion survey  Primers (ON 5, ON 6, and ON 7) for standing-start single-nucleotide insertions were 5’-labeled with γ-32P-ATP by T4 polynucleotide kinase and purified with 15% denaturing PAGE gel (7 M urea) before use. The insertion reaction was carried out in 20 µL of 1X thermopol buffer (Buffer 2) under non-forcing conditions: the duplex concentration was 1 µM, the tested nucleotide concentration was 10 µM, and the Vent (exo-) DNA polymerase used for each 20 µL reaction was 1 unit, which was a sufficient amount for incorporating 10 nmole of dNTP in 30 min at 75°C. Firstly, 9.5 µL of annealing mixture containing 1 µL of each 20 µM labeled primer and tested template, 2 µL of 10X thermopol buffer, 0.2 µL of 100 mM MgSO4 (supplemented Mg2+ for further stabilization of enzyme•duplex complex), as well as 5.3 µL of H2O was heated to above 95°C for 5 min, then allowed to cool down to room temperature slowly to form template-primer duplex. At this point, 0.5 µL (1 unit) of Vent (exo-) DNA polymerase was added to the annealed duplex, and incubated at 72°C for 5 min to allow the formation of enzyme•duplex complex. Then, 10 µL of 20 µM of tested dNTP was added into the enzyme•duplex complex, and the insertion assay was readily initiated by incubating the 198  reaction at 72°C. The reaction was stopped by adding into equal volume of gel loading buffer 8 (20 µL of formamide stop solution containing 25 mM EDTA, 0.01% bromophenol blue and 0.01% xylene cyanole) after designed reaction time. Reaction products were separated by 20% denaturing PAGE, and visualized by PhosphorImager. For the test of exploring whether dAim forms mispairs with dATP, dCTP, and dGTP after long-time incubation, 20 µL of insertion reaction with each dNTP was set up as described above, respectively. 2 µL of reaction was sampled into 10 µL of gel loading buffer at different time points including 2 min, 5 min, 30 min and 180 min, through 180 min reaction time at 72°C, followed by visualization using PhosphorImager.  5.7.7 Strand synthesis continuing beyond multiple modified base pairs   The assay for strand synthesis continuing beyond multiple modified base pairs was carried out under the same non-forcing conditions as single-nucleotide insertion survey. 1 µL of 20 µM 5’-32P-labeled 29 nt primer (ON 7) was annealed to an equal amount of modified template ON 2 and unmodified control template ON 3, respectively. Then an incubation step with 0.5 µL (1 unit) of Vent (exo-) DNA polymerase at 72°C for 5 min was done to form enzyme•duplex complex following the same procedure as described for single-nucleotide insertion asaay. Instead of using 10 µM of just one nucleotide, 10 µL of nucleotide mixture containing 10 µM of each dATP, dCTP, dGTP, and dTTP was added into 10 µL of pre-formed enzyme•duplex complex. The primer extension assay was immediately initiated by incubation at 72°C. 2 µL of reaction was taken into 10 µL of gel loading buffer at different time points, including 5 min, 10 min, 30 min, 120 min, and 199  180 min. Samples were separated by denaturing 20% PAGE gels (7 M urea) and visualized by PhosphorImager.  5.7.8 Test of fidelity of a complete replication cycle Transcription of Dz7-38-32 and “Dummy” control. Transcription of self-cleaving Dz7-38-32 using synthetic template ON 8 and primer ON 9 containing the all-RNA substrate was carried out in a 40 µL of reaction (15 pmole) following the “Template-directed primer extension for transcribing modified library for selection, 2.9.4, Chapter 2”. In parallel with the preparation of Dz7-38-32, the “dummy” control was prepared following the same procedures. In the dummy reaction, instead of including dNTP mixtures (dAimTP, dCaaTP, dUgaTP, and dGTP), equal volume of dH2O was used to supplement the 40 µL of transcription reaction. Self-cleavage and purification of modified Dz7-38-32 and dummy control.  The resulting reaction mixture from transcription was incubated with 50 µL of pre-washed streptavidin beads for 15 min at room temperature to immobilize the DNA duplex on beads. Following two washes with 100 µL of wash buffer 4 after beads incubation, the template strand was stripped away by 5 quick washes (no longer than 30 seconds per wash) with 100 µL of 0.1 M NaOH containing 1 mM EDTA. The resulting modified strand sticking to streptavidin beads was immediately neutralized using 200 µL of neutralization buffer 5 followed by a final wash with 100 µL of water. The modified DNA on beads was then allowed to fold and cleave in 100 µL of cleavage buffer 6 for 30 min. The cleavage reaction was then magnetized, and the reaction supernatant containing self-cleaved Dz7-38-32 was recovered from beads. This was then precipitated using 1% 200  LiClO4 in acetone and washed with ethanol. The modified DNA was re-suspended in formamide loading buffer containing 25 mM EDTA, 0.01% bromophenol blue and 0.01% xylene cyanole, and resolved by a 7% denaturing PAGE (7 M urea). Once again, the same treatments were applied to the “dummy” control. Gel pieces corresponding to the desired self-cleavage products were sliced off, eluted using gel elution buffer 7.  Amplification of self-cleaved modified DNA. The gel eluted DNAs (self-cleaved DNAzyme and “dummy” control), which were re-suspended into 20 µL of dH2O, were used as templates for PCR amplification using Vent (exo-) or Taq DNA polymerase, respectively. The amplification reactions were performed in 200 µL of 1X thermopol buffer conditions, supplemented with 1 mM of MgSO4, 7 µM of primers (ON 10 and ON 11), 0.3 mM of each of the four natural dNTPs, and 0.1 unit/µL of Vent (exo-) or 0.25 unit/µL of Taq DNA polymerase. The reactions were cycled for 25 cycles: 15 seconds at 95°C, 15 seconds at 58°C and 40 seconds at 75°C. To generate PCR products with 3’-A overhangs required by subsequent TOPO cloning, a final extension step of 30 min at 75°C was added to the Taq amplification reaction. To monitor the progress of DNA amplification, which was used to assay the self-cleaved modified DNA for possible unmodified template DNA contamination, aliquots (1 µL) were removed at cycle 0, 5, 10, 15 and 25 and stored in fridge. Once all the cycles were taken, the samples were analyzed on a 2% agarose gel stained with ethidium bromide.   Fidelity analysis through cDNA sequencing. PCR amplified DNA was purified using 2% agarose gel. The amplicons contained in the gel slice was purified using GeneJet Gel Extraction Kit first according to the manufacture’s instruction before they were 201  TOPO cloned into the pCR2.1-TOPO vectors. The vectors were then used to transform E.coli DH5α following a normal chemical transformation protocol. White transformant colonies were picked by Blue-White screening on LB Agar plates containing 100 mg/L ampicillin for inoculation. Plasmids were prepared using PureLink Quick Plasmid Miniprep Kit (Invitrogen), and sequenced by the Nucleic Acid Protein Service Unit of UBC (NAPS-UBC) using the M13R sequencing primer. The alignment of individual sequences was done on the MEGA 6.06 platform.    202  Chapter 6: Summary and Future Work  6.1 Summary of Research  This thesis has addressed several aspects regarding the pursuit of RNase A-mimicking DNAzymes through the use of three chemically modified nucleotides: 8-histaminyl-deoxyadenosine triphosphate (dAimTP), 5-guanidinoallyl-deoxyuridine triphosphate (dUgaTP), and 5-aminoallyl-deoxycytidine triphosphate (dCaaTP) in in vitro selection. Chapter 2 discussed the design and optimization of a novel construct for use in selections and a comprehensive selection strategy to allow a direct selection of all-RNA cleavers that simultaneously foster the subsequent conversion of cis-cleaving to a trans-acting DNAzyme. Also detailed in Chapter 2, was the successful application of the novel construct and the optimized in vitro selection strategy to selections containing a chimeric DNA/RNA substrate and a 17 nt all-RNA substrate, respectively. The latter selection led to the isolation of two families of DNAzymes capable of rapid all-RNA cleavage. Chapter 3 demonstrated the systematic characterization of all enzymatic parameters, the validation of sequence specificity, and the probing of an RNase A-mimicking mechanism, using the best DNAzyme obtained (Dz7-38-32). Dz7-38-32 attained kcat and KM values of ~0.24 min-1 and 2.72 µM under simulated physiological conditions (pH 7.45, 150 mM K+, 0.5 mM Mg2+, 37°C). This corresponds to a catalytic efficiency of ~105 M-1min-1, the best catalytic efficiency achieved by chemically modified DNAzymes under a low-M2+ regime characteristic of cellular conditions up to date. Mutations in the guide arms confirmed its high specificity for the original substrate 203  derived from the HIV-1 LTR-promoter mRNA. The bell-shaped pH-rate profile along with the high M2+-independence and M2+-tolerance reflect the RNase A-mimicking properties of Dz7-38-32. Chapter 4 probed the spontaneous cellular uptake of Dz7-38-32 due to its similarity with cell penetrating peptides (CPPs) regarding functional groups contained. After 45 h of incubation with HeLa cells, only modified Dz7-38-32 displayed significant punctate distribution in cytoplasm relative to cells treated with the same concentration of various control oligonucleotides. The cell membrane permeability displayed by Dz7-38-32 further highlights its potential as therapeutic reagents. Chapter 5 investigated the biophysical and enzymatic properties of these three modified nucleosides (dAim, dUga and dCaa) in the context of discretely modified oligonucleotides. The presence of a single dAim residue resulted in decreased stability in duplexes by 0.9 ˚C and 1.7 ˚C at pH 6 and 8.5, respectively, when compared to the all-natural duplex, whereas, the presence of dCaa and dUga increased the stability of duplexes by 2˚C. Therefore, the simultaneous presence of all these three modified nucleosides increases the melting temperature of duplexes to within the range of the unmodified sequence. Single-nucleotide insertion assays and assessment of the overall fidelity throughout a complete replication/selection cycle revealed a negligible mutagenicity in terms of both primer extension as well as enzymatic read-through. The characteristic signature of coding and decoding as genetic polymers exemplified by Dz7-38-32 along with the biophysical property evidenced here strongly support the continued use of these three modified nucleotides in combinatorial selection.  204  6.2 Future Directions  It has been more than a decade since the report of Dz925-11173, the experimentally elucidated RNase A-mimicking DNAzyme with two extra chemical functionalities175,177. Ever since that discovery, much has been achieved185-187, much has yet to be accomplished. From the perspective of development of therapeutically efficient DNAzymes that are capable of highly efficient and sequence-specific mRNA disruption in vivo, Dz7-38-32 represents the best candidate to date. Apart from Dz10-23 and Dz8-17, no other DNAzyme has ever been studied for activity in cells or in vivo. Dz7-38-32, with its characteristic distinction of M2+-independence from Dz10-23, is worth studying for its HIV-1 LTR-promoter mRNA inhibition activity in cells and in vivo. Dz10-23 will serve for a good benchmark for the evaluation of both the activity of Dz7-38-32 and the significance of M2+-independence to the application of modified DNAzymes for catalytic cleavage of target mRNAs. Towards this end, modified phosphoramidites allowing high yield coupling and sufficient amount of Dz7-38-32 synthesized on solid phase by coupling the modified phosphoramidites will be required.  Dz7-38-32 has a catalytic core of 45 nucleosides. Parallel selections using libraries consisting of 20 (N20) or 30 (N30) degenerate positions did not give rise to as efficient cleavers as those using N40 library. Hence it might be worth trying to carry out “cutdown” or “boundary” experiments290 to define a more precise catalytic core. To do so, one nucleoside, or a stretch of nucleosides in the catalytic core is truncated from its 5’ or 3’ end and the cleavage activity of the resulting smaller DNAzyme is evaluated. Dz16.2-11 containing only chemically essential residues in the catalytic core (as 205  depicted in Figure 1.7, Chapter 1) was obtained using this methodology.291 Strategies including a re-selection step by partial sequence degeneration and an evolution step by mutagenic PCR that introduce more sequence variation to subsequently increase the sequence space during selection afforded DNAzymes with much improved catalytic efficiency. This achievement emphasizes the importance of sufficient sequence space for the isolation of rapid self-cleavers. The fact that several families of DNAzymes from N20 to N40 libraries have survived selection suggests that these sequences represent viable sequence combinations that can, in the presence of a suitable polymerase (Sequenase), be converted from unmodified templates into modified strands which can then be viably amplified by a high temperature polymerase with good retention of genetic information. The study in Chapter 5 further confirmed these perspectives experimentally and supported the use of these dNTPs in in vitro selection with transcription and reverse-transcription mediated by Sequenase v2.0 and Vent (exo-) or Taq DNA polymerase, respectively. As an alternative methodology, several reports address the engineering of polymerases that can synthesize modified nucleic acid polymers from DNA templates or can reverse transcribe modified nucleic acid polymers into DNA.292-296 These studies lead to the discovery of a number of engineered polymerases capable of replicating unnatural genetic polymers with unbiased sequence space. An adapted methodology for engineering DNA polymerases that can replicate these three modified nucleotides along with natural dGTP to provide library with unbiased sequence would eventually allow a better use of these modified nucleotides in in vitro selection.  206  In conclusion, the body of this work now provides a set of DNAzymes worthy of synthetic scale-up for full toxicity assays and ultimately in vivo investigation.  At the same time, fundamental studies generally demonstrate that the three modified nucleotides used herein can continue to provide a rich source of new M2+-independent catalysts for RNA cleavage as well as perhaps other reactions of fundamental and/or medical interest.    207  Bibliography  (1) Crick, F. H. On protein synthesis.; Symposia of the Society for Experimental Biology, 1958.  (2) Crick, F. Nature 1970, 227 (5258), 561–563.  (3) Prody, G. A.; Bakos, J. T.; Buzayan, J. M.; Schneider, I. R.; Bruening, G. Science 1986, 231 (4745), 1577–1580.  (4) Buzayan, J. M.; Gerlach, W. L.; Bruening, G. Nature 1986, 323 (6086), 349–353.  (5) Sharmeen, L.; Kuo, M. Y.; Dinter-Gottlieb, G.; Taylor, J. J. Virol. 1988, 62 (8), 2674–2679.  (6) Saville, B.; Collins, R. Cell 1990, 61 (4), 685–696.  (7) Winkler, W. C.; Nahvi, A.; Roth, A.; Collins, J. A.; Breaker, R. R. Nature 2004, 428 (6980), 281–286.  (8) Cech, T.; Zaug, A.; Grabowski, P. Cell 1981, 27 (3), 487–496.  (9) Kruger, K.; Grabowski, P.; Zaug, A.; Sands, J. Cell 1982, 31 (1), 147–157.  (10) Cech, T. Annu. Rev. Biochem. 1990, 59, 543–568.  (11) Guerrier-Takada, C.; Gardiner, K.; Marsh, T.; Pace, N.; Altman, S. Cell 1983, 35, 849–857.  (12) Noller, H.; Hoffarth, V.; Zimniak, L. Science 1992, 256 (5062), 1416–1419.  (13) Nissen, P.; Hansen, J.; Ban, N.; Moore, P.; Steitz, T. Science 2000, 289 (5481), 920–930.  (14) Gilbert, W. Nature 1986, 319 (6055), 618.  (15) Ellington, A. D.; Szostak, J. W. Nature 1990, 346 (6287), 818–822.  (16) Robertson, M. P.; Ellington, A. Nat. Biotechnol. 1999, 17 (1), 62–66.  (17) Robertson, D. L.; Joyce, G. F. Nature 1990, 344 (6265), 467–468.  208  (18) Tuerk, C.; Gold, L. Science 1990, 249 (4968), 505–510.  (19) Joyce, G. F. Annu. Rev. Biochem. 2004, 73 (1), 791–836.  (20) Baum, D. A.; Silverman, S. K. Cell. Mol. Life Sci. 2008, 65 (14), 2156–2174.  (21) Bartel, D. P.; Szostak, J. W. Science 1993, 261 (5127), 1411–1418.  (22) Sengle, G.; Eisenführ, A.; Arora, P.; Nowick, J. Chem. Biol. 2001, 8 (5), 459–473.  (23) Tsukiji, S.; Pattnaik, S.; Suga, H. Nat. Struct. Mol. Biol. 2003, 10 (9), 713–717.  (24) Tsukiji, S.; Pattnaik, S.; Suga, H. J. Am. Chem. Soc. 2004, 126 (16), 5044–5045.  (25) Seelig, B.; Jäschke, A. Chem. Biol. 1999, 6 (3), 167–176.  (26) Ryu, Y.; Kim, K.-J.; Roessner, C. A.; Scott, A. I. Chem. Commun. 2006, No. 13, 1439–1441.  (27) Nieuwlandt, D.; West, M.; Cheng, X. Chembiochem 2003, 4 (7), 649–662.  (28) Breaker, R. R.; Joyce, G. F. Chem. Biol. 1994, 1 (4), 223–229.  (29) Santoro, S. W.; Joyce, G. F. Proc. Natl. Acad. Sci. U.S.A. 1997, 94 (9), 4262–4266.  (30) Geyer, C. R.; Sen, D. Chem. Biol. 1997, 4 (8), 579–593.  (31) Roth, A.; Breaker, R. R. Proc. Natl. Acad. Sci. U.S.A. 1998, 95 (11), 6027–6031.  (32) Purtha, W. E.; Coppins, R. L.; Smalley, M. K.; Silverman, S. K. J. Am. Chem. Soc. 2005, 127 (38), 13124–13125.  (33) Carmi, N.; Shultz, L. A.; Breaker, R. R. Chem. Biol. 1996, 3 (12), 1039–1046.  (34) Carmi, N.; Balkhi, S. R.; Breaker, R. R. Proc. Natl. Acad. Sci. U.S.A. 1998, 95 (5), 2233–2237.  (35) Gu, H.; Furukawa, K.; Weinberg, Z.; Berenson, D. F.; Breaker, R. R. J. Am. Chem. Soc. 2013, 135 (24), 9121–9129.  (36) Chandra, M.; Sachdeva, A.; Silverman, S. K. Nat. Chem. Biol. 2009, 5 (10), 209  718–720.  (37) Cuenoud, B.; Szostak, J. W. Nature 1995, 375 (6532), 611–614.  (38) Lehman, I. R. Science 1974, 186 (4166), 790–797.  (39) Sreedhara, A.; Li, Y.; Breaker, R. J. Am. Chem. Soc. 2004, No. 126, 3454–3460.  (40) Li, Y.; Sen, D. Nat. Struct. Biol. 1996, 3 (9), 743–747.  (41) Li, Y.; Sen, D. Biochemistry 1997, 36 (18), 5589–5599.  (42) Dokukin, V.; Silverman, S. K. Chem. Commun. 2014, 50 (66), 9317–9320.  (43) Walsh, S. M.; Sachdeva, A.; Silverman, S. K. J. Am. Chem. Soc. 2013, 135 (40), 14928–14931.  (44) Pradeepkumar, P. I.; Höbartner, C.; Baum, D. A.; Silverman, S. K. Angew. Chem. Int. Ed. 2008, 47 (9), 1753–1757.  (45) Wang, Y.; Silverman, S. K. J. Am. Chem. Soc. 2003, 125 (23), 6880–6881.  (46) Wang, Y.; Silverman, S. K. Angew. Chem. Int. Ed. 2005, 44 (36), 5863–5866.  (47) Brandsen, B. M.; Velez, T. E.; Sachdeva, A.; Ibrahim, N. A.; Silverman, S. K. Angew. Chem. Int. Ed. 2014, 126 (34), 9191–9196.  (48) Cech, T. Science 1987, 236 (4808), 1532–1539.  (49) Quigley, G.; Rich, A. Science 1976, 194 (4267), 796–806.  (50) Silverman, S. K. Wiley Encyclopedia of Chemical Biology 2008.  (51) Baum, D. A.; Silverman, S. K. Cell. Mol. Life Sci. 2008, 65 (14), 2156–2174.  (52) Chandra, M.; Silverman, S. K. J. Am. Chem. Soc. 2008, 130 (10), 2936–2937.  (53) Parker, D. J.; Xiao, Y.; Aguilar, J. M.; Silverman, S. K. J. Am. Chem. Soc. 2013, 135 (23), 8472–8475.  (54) Narlikar, G. J.; Herschlag, D. Annu. Rev. Biochem. 1997, 66, 19–59.  (55) Emilsson, G. M.; Nakamura, S.; Roth, A.; Breaker, R. R. RNA 2003, 9 (8), 907–918. 210  (56) Breaker, R. R. RNA 2003, 9 (8), 949–957.  (57) Nakano, S.-I.; Chadalavada, D. M.; Bevilacqua, P. C. Science 2000, 287 (5457), 1493–1497.  (58) Trawick, B.; Daniher, A.; Bashkin, J. Chem. Rev. 1998, 98 (3), 939–960.  (59) Takasaki, B.; Chin, J. J. Am. Chem. Soc. 1994, 116 (3), 1121–1122.  (60) Wahnon, D.; Hynes, R. C.; Chin, J. Journal of the Chemical Society, Chem. Commun. 1994, 12, 1441.  (61) Basile, L.; Raphael, A.; Barton, J. J. Am. Chem. Soc. 1987, 109 (24), 7550–7551.  (62) Schnaith, L. M.; Hanson, R. S.; Que, L. Proc. Natl. Acad. Sci. U.S.A. 1994, 91 (2), 569–573.  (63) Komiyama, M.; Kodama, T.; Takeda, N. J. Biochem. 1994, 115 (5), 809–810.  (64) Chin, J.; Banaszczyk, M.; Jubian, V. J. Am. Chem. Soc. 1989, 111 (1), 186–190.  (65) Li, Y.; Breaker, R. R. J. Am. Chem. Soc. 1999, 121 (23), 5364–5372.  (66) Radzicka, A.; Wolfenden, R. Science 1995, 267 (5194), 90–93.  (67) Chapman, K. B.; Szostak, J. W. Curr. Opin. Struct. Biol. 1994, 4, 618–622.  (68) Breaker, R. R. Chem. Rev. 1997, 97 (2), 371–390.  (69) Boese, B. J.; Corbino, K.; Breaker, R. R. Nucleosides Nucleotides & Nucleic Acids 2008, 27 (8), 949–966.  (70) Compton, J. Nature 1991, 350 (6313), 91–92.  (71) Cox, J. C.; Rudolph, P.; Ellington, A. D. Biotechnol. Prog. 1998, 14 (6), 845–850.  (72) Westheimer, F. Acc. Chem. Res. 1968, 1 (12), 381.  (73) Westheimer, F. Acc. Chem. Res. 1968, 1 (3), 70–78.  (74) Dennis, E. A.; Westheimer, F. H. J. Am. Chem. Soc. 1966, 88 (14), 3432–3433. 211  (75) Soukup, G. A.; Breaker, R. R. RNA 1999, 5 (10), 1308–1325.  (76) Emilsson, G. M. RNA 2003, 9 (8), 907–918.  (77) Perreault, D. M.; Anslyn, E. V. Angew. Chem. Int. Ed. 1997, 36 (5), 432–450.  (78) Martick, M.; Scott, W. G. Cell 2006, 126 (2), 309–320.  (79) Martick, M.; Lee, T.-S.; York, D. M.; Scott, W. G. Chem. Biol. 2008, 15 (4), 332–342.  (80) Chi, Y.-I.; Martick, M.; Lares, M.; Kim, R.; Scott, W. G.; Kim, S.-H. PLoS Biol. 2008, 6 (9), e234.  (81) Rupert, P.; Ferré-D'Amaré, A. Nature 2001, 410 (6830), 780–786.  (82) Rupert, P.; Massey, A.; Sigurdsson, S. Science 2002, 298 (5597), 1421–1424.  (83) Bevilacqua, P. C. Biochemistry 2003, 42 (8), 2259–2265.  (84) Murray, J. B.; Seyhan, A. A.; Walter, N. G.; Burke, J. M.; Scott, W. G. Chem. Biol. 1998, 5 (10), 587–595.  (85) Nakano, S.; Chadalavada, D. M.; Bevilacqua, P. C. Science 2000, 287 (5457), 1493–1497.  (86) Das, S. R.; Piccirilli, J. A. Nat. Chem. Biol. 2005, 1 (1), 45–52.  (87) Fedor, M. J. J. Mol. Biol. 2000, 297 (2), 269–291.  (88) Clouet-d'Orval, B.; Uhlenbeck, O. C. Biochemistry 1997, 36 (30), 9087–9092.  (89) Lafontaine, D. A.; Wilson, T. J.; Norman, D. G.; Lilley, D. M. J. Mol. Biol. 2001, 312 (4), 663–674.  (90) Pyle, A. M. Science 1993, 261 (5122), 709–714.  (91) Dahm, S. C.; Uhlenbeck, O. C. Biochemistry 1991, 30 (39), 9464–9469.  (92) Wadkins, T. S.; Shih, I.; Perrotta, A. T.; Been, M. D. J. Mol. Biol. 2001, 305 (5), 1045–1055.  (93) Perrotta, A. T.; Shih, I.; Been, M. D. Science 1999, 286 (5437), 123–126.  (94) Cochrane, J. C.; Strobel, S. A. Acc. Chem. Res. 2008, 41 (8), 1027–1035. 212  (95) Doudna, J. A.; Lorsch, J. R. Nat. Struct. Mol. Biol. 2005, 12 (5), 395–402.  (96) Zaug, A. J.; Been, M. D.; Cech, T. R. Nature 1986, 324 (6096), 429–433.  (97) Cech, T. R. JAMA 1988, 260 (20), 3030–3034.  (98) Christoffersen, R.; Marr, J. J. Med. Chem. 1995, 38 (12), 2023–2037.  (99) Benitez-Hess, M.; Alvarez-Salas, L. Letters in Drug Design & Discovery 2006, 3 (6), 390–404.  (100) Paolella, G.; Sproat, B. S.; Lamond, A. I. EMBO J. 1992, 11 (5), 1913–1919.  (101) Perreault, J.; Wu, T.; Cousineau, B.; Ogilvie, K. Nature 1990, 344 (6266), 565–567.  (102) Hendry, P.; McCall, M. J.; Santiago, F. S.; Jennings, P. A. Nucleic Acids Res. 1992, 20 (21), 5737–5741.  (103) Kurreck, J.; Bieber, B.; Jahnel, R.; Erdmann, V. A. J. Biol. Chem. 2002, 277 (9), 7099–7107.  (104) Breaker, R. R.; Joyce, G. F. Chem. Biol. 1995, 2 (10), 655–660.  (105) Scott, W. G.; Klug, A. Trends Biochem. Sci. 1996, 21 (6), 220–224.  (106) Bratty, J.; Chartrand, P.; Ferbeyre, G.; Cedergren, R. Biochim. Biophys. Acta 1993, 1216 (3), 345–359.  (107) Silverman, S. K. Nucleic Acids Res. 2005, 33 (19), 6151–6163.  (108) Kim, H.-K.; Rasnik, I.; Liu, J.; Ha, T.; Lu, Y. Nat. Chem. Biol. 2007, 3 (12), 763–768.  (109) Cieslak, M.; Szymanski, J.; Adamiak, R. W.; Cierniewski, C. S. J. Biol. Chem. 2003, 278 (48), 47987–47996.  (110) Suzuki, Y.; Komatsu, H.; Ikeda, T.; Saito, N.; Araki, S.; Citterio, D.; Hisamoto, H.; Kitamura, Y.; Kubota, T.; Nakagawa, J.; Oka, K.; Suzuki, K. Anal. Chem. 2002, 74 (6), 1423–1428.  (111) Andrea; Romani, M. P. Frontiers in Bioscience 2000, 5 (1), d720.  (112) Marban, E.; Kitakaze, M.; Kusuoka, H.; Porterfield, J. K.; Yue, D. T.; Chacko, V. P. Proc. Natl. Acad. Sci. U.S.A. 1987, 84 (16), 6005–6009. 213  (113) Bozym, R. A.; Thompson, R. B.; Stoddard, A. K.; Fierke, C. A. ACS Chem. Biol. 2006, 1 (2), 103–111.  (114) Peracchi, A. J. Biol. Chem. 2000, 275 (16), 11693–11697.  (115) Brown, A. K.; Li, J.; Pavot, C. M. B.; Lu, Y. Biochemistry 2003, 42 (23), 7152–7161.  (116) Santoro, S. W.; Joyce, G. F. Biochemistry 1998, 37 (38), 13330–13342.  (117) Santiago, F. S.; Lowe, H. C.; Kavurma, M. M.; Chesterman, C. N.; Baker, A.; Atkins, D. G.; Khachigian, L. M. Nat. Med. 1999, 5 (11), 1264–1269.  (118) Takahashi, H.; Hamazaki, H.; Habu, Y.; Hayashi, M.; Abe, T.; Miyano-Kurosaki, N.; Takaku, H. FEBS Letters 2004, 560 (1-3), 69–74.  (119) Zhang, G.; Dass, C. R.; Sumithran, E.; Di Girolamo, N.; Sun, L.-Q.; Khachigian, L. M. J. Natl. Cancer Inst. 2004, 96 (9), 683–696.  (120) Yuan, B.-F.; Xue, Y.; Luo, M.; Hao, Y.-H.; Tan, Z. Int. J. Biochem. Cell Biol. 2007, 39 (6), 1119–1129.  (121) Zhang, L.; Gasper, W. J.; Stass, S. A.; Ioffe, O. B.; Davis, M. A.; Mixson, A. J. Cancer Res. 2002, 62 (19), 5463–5469.  (122) Mitchell, A.; Dass, C. R.; Sun, L.-Q.; Khachigian, L. M. Nucleic Acids Res. 2004, 32 (10), 3065–3069.  (123) Fokina, A. A.; Meschaninova, M. I.; Durfort, T.; Venyaminova, A. G.; François, J.-C. Biochemistry 2012, 51 (11), 2181–2191.  (124) Cho, E.-A.; Moloney, F. J.; Cai, H.; Au-Yeung, A.; China, C.; Scolyer, P. R. A.; Yosufi, B.; Raftery, M. J.; Deng, J. Z.; Morton, S. W.; Hammond, P. T.; Arkenau, H.-T.; Damian, D. L.; Francis, D. J.; Chesterman, C. N.; Barnetson, R. S. C.; Halliday, G. M.; Khachigian, L. M. The Lancet 2013, 381 (9880), 1835–1843.  (125) Jakobsen, M. R.; Haasnoot, J.; Wengel, J.; Berkhout, B.; Kjems, J. Retrovirology 2007, 4, 29.  (126) Wo, J.-E.; Wu, X.-L.; Zhou, L.-F.; Yao, H.-P.; Chen, L.-W.; Dennin, R.-H. World J. Gastroenterol. 2005, 11 (23), 3504–3507.  (127) Ackermann, J. M.; Kanugula, S.; Pegg, A. E. Biochemistry 2005, 44 (6), 2143–2152. 214  (128) Zhang, L.; Gasper, W. J.; Stass, S. A.; Ioffe, O. B.; Davis, M. A.; Mixson, A. J. Cancer Res. 2002, 62 (19), 5463–5469.  (129) Russell, R.; Zhuang, X.; Babcock, H. P.; Millett, I. S.; Doniach, S.; Chu, S.; Herschlag, D. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (1), 155–160.  (130) Corey, D. Nat. Chem. Biol. 2007, 3 (1), 8–11.  (131) Baker, B. F.; Monia, B. P. Biochim. Biophys. Acta 1999, 1489 (1), 3–18.  (132) Scherer, L. J.; Rossi, J. J. Nat. Biotechnol. 2003, 21 (12), 1457–1465.  (133) Trepanier, J.; Tanner, J.; Momparler, R. Journal of Viral Hepatitis 2006, 13 (2), 131–138.  (134) Trepanier, J. B.; Tanner, J. E.; Alfieri, C. Virology 2008, 377 (2), 339–344.  (135) Faulhammer, D.; Famulok, M. J. Mol. Biol. 1997, 269 (2), 188–202.  (136) Carrigan, M. A.; Ricardo, A.; Ang, D. N.; Benner, S. A. Biochemistry 2004, 43 (36), 11446–11459.  (137) Kuimelis, R.; McLaughlin, L. J. Am. Chem. Soc. 1995, 117 (44), 11019–11020.  (138) Torabi, S.-F.; Wu, P.; McGhee, C. E.; Chen, L.; Hwang, K.; Zheng, N.; Cheng, J.; Lu, Y. Proc. Natl. Acad. Sci. U.S.A. 2015, 112 (19), 5903–5908.  (139) Thompson, J. E.; Raines, R. T. J. Am. Chem. Soc. 1994, 116 (12), 5467–5468.  (140) Raines, R. T. Chem. Rev. 1998, 98 (3), 1045–1066.  (141) Kartha, G.; Bello, J.; Harker, D. Nature 1967, 214 (5085), 862–865.  (142) Findlay, D.; Herries, D. G.; Mathias, A. P.; Rabin, B. R.; Ross, C. A. J. Biochem. 1962, 85 (1), 152–153.  (143) Wlodawer, A.; Bott, R.; Sjölin, L. J. Biol. Chem. 1982, 257 (3), 1325–1332.  (144) Cuchillo, C. M.; Nogués, M. V.; Raines, R. T. Biochemistry 2011, 50 (37), 7835–7841.  (145) Thompson, J. E.; Kutateladze, T. G.; Schuster, M. C.; Venegas, F. D.; Messmore, J. M.; Raines, R. T. Bioorganic chemistry 1995, 23 (4), 471–481.  (146) Takagi, Y.; Warashina, M.; Stec, W. Nucleic Acids Res. 2001, 29 (9), 1815–215  1834.  (147) Westhof, E. Science 1999, 286 (5437), 61–62.  (148) Bashkin, J.; Gard, J.; Modak, A. J. Org. Chem. 1990, 55 (17), 5125–5132.  (149) Bashkin, J.; McBeath, R.; Modak, A. J. Org. Chem. 1991, 56 (9), 3168–3176.  (150) Reynolds, M.; Beck, T.; Say, P. Nucleic Acids Res. 1996, 24 (4), 760–765.  (151) Vlassov, V.; Abramova, T.; Godovikova, T.; Giege, R.; Silnikov, V. Antisense Nucleic Acid Drug Devel. 1997, 7 (1), 39–42.  (152) Lönnberg, H. Org. Biomol. Chem. 2011, 9 (6), 1687–1703.  (153) Gnaccarini, C.; Peter, S.; Scheffer, U.; Vonhoff, S.; Klussmann, S.; Göbel, M. W. J. Am. Chem. Soc. 2006, 128 (24), 8063–8067.  (154) Komiyama, M.; Inokawa, T. J. Biochem. 1994, 116 (4), 719–720.  (155) Aurup, H.; Williams, D.; Eckstein, F. Biochemistry 1992, 31 (40), 9636–9641.  (156) Eaton, B.; Pieken, W. Annu. Rev. Biochem. 1995, 64, 837–863.  (157) Eaton, B. Curr. Opin. Chem. Biol. 1997, 1 (1), 10–16.  (158) Dewey, T.; Mundt, A.; Crouch, G. J. Am. Chem. Soc. 1995, 117 (32), 8474–8475.  (159) Matulic-Adamic, J.; Daniher, A.; Karpeisky, A. Bioorg. Med. Chem. Lett. 2000, 10, 1299-1302.  (160) Vaish, N. K.; Fraley, A. W.; Szostak, J. W.; McLaughlin, L. W. Nucleic Acids Res. 2000, 28 (17), 3316–3322.  (161) Latham, J.; Johnson, R.; Toole, J. Nucleic Acids Res. 1994, 22 (14), 2817–2822.  (162) Battersby, T. R.; Ang, D. N.; Burgstaller, P.; Jurczyk, S. C.; Bowser, M. T.; Buchanan, D. D.; Kennedy, R. T.; Benner, S. A. J. Am. Chem. Soc. 1999, 121 (42), 9781–9789.  (163) Roychowdhury, A.; Illangkoon, H.; Hendrickson, C. L.; Benner, S. A. Org. Lett. 2004, 6 (4), 489–492.  216  (164) Lee, S. E.; Sidorov, A.; Gourlain, T.; Mignet, N.; Thorpe, S. J.; Brazier, J. A.; Dickman, M. J.; Hornby, D. P.; Grasby, J. A.; Williams, D. M. Nucleic Acids Res. 2001, 29 (7), 1565–1573.  (165) Gourlain, T.; Sidorov, A.; Mignet, N.; Thorpe, S. J.; Lee, S. E.; Grasby, J. A.; Williams, D. M. Nucleic Acids Res. 2001, 29 (9), 1898–1905.  (166) Lermer, L.; Hobbs, J.; Perrin, D. M. Nucleosides Nucleotides & Nucleic Acids 2002, 21 (10), 651–664.  (167) Held, H.; Benner, S. Nucleic Acids Res. 2002, 30 (17), 3857–3869.  (168) Kuwahara, M.; Takahata, Y.; Shoji, A.; Ozaki, A. Bioorg. Med. Chem. Lett. 2003, 13, 3735-3738.  (169) Tarasow, T. M.; Tarasow, S. L.; Eaton, B. E. Nature 1997, 389 (6646), 54–57.  (170) Wiegand, T. W.; Janssen, R. C.; Eaton, B. E. Chem. Biol. 1997, 4 (9), 675–683.  (171) Sakthivel, K.; Barbas, C. Angew. Chem. Int. Ed. 1998, 37 (20), 2872–2875.  (172) Santoro, S. W.; Joyce, G. F.; Sakthivel, K.; Gramatikova, S.; Barbas, C. F. J. Am. Chem. Soc. 2000, 122 (11), 2433–2439.  (173) Perrin, D. M.; Garestier, T.; Hélène, C. J. Am. Chem. Soc. 2001, 123 (8), 1556–1563.  (174) Perrin, D. M.; Garestier, T.; Hélène, C. Nucleosides Nucleotides 1999, 18 (3), 377–391.  (175) Lermer, L.; Roupioz, Y.; Ting, R.; Perrin, D. M. J. Am. Chem. Soc. 2002, 124 (34), 9960–9961.  (176) Ting, R.; Thomas, J. M.; Lermer, L.; Perrin, D. M. Nucleic Acids Res. 2004, 32 (22), 6660–6672.  (177) Thomas, J. M.; Yoon, J.-K.; Perrin, D. M. J. Am. Chem. Soc. 2009, 131 (15), 5648–5658.  (178) Herschlag, D. Proc. Natl. Acad. Sci. U.S.A. 1991, 88 (16), 6921–6925.  (179) Sidorov, A. V.; Grasby, J. A.; Williams, D. M. Nucleic Acids Res. 2004, 32 (4), 1591–1601.  217  (180) Roig, V.; Asseline, U. J. Am. Chem. Soc. 2003, 125 (15), 4416–4417.  (181) Calnan, B. J.; Tidor, B.; Biancalana, S.; Hudson, D.; Frankel, A. D. Science 1991, 252 (5009), 1167–1171.  (182) Ohsawa, K.; Kasamatsu, T.; Nagashima, J.; Hanawa, K. Analytical Sciences 2008, 24 (1), 167–172.  (183) Ohmichi, T.; Kuwahara, M.; Sasaki, N.; Hasegawa, M.; Nishikata, T.; Sawai, H.; Sugimoto, N. Angew. Chem. Int. Ed. 2005, 44 (41), 6682–6685.  (184) Barawkar, D. A.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A. 1998, 95 (19), 11047–11052.  (185) Hollenstein, M.; Hipolito, C. J.; Lam, C. H.; Perrin, D. M. Nucleic Acids Res. 2009, 37 (5), 1638–1649.  (186) Hollenstein, M.; Hipolito, C. J.; Lam, C. H.; Perrin, D. M. Chembiochem 2009, 10 (12), 1988–1992.  (187) Hollenstein, M.; Hipolito, C. J.; Lam, C. H.; Perrin, D. M. ACS Comb. Sci. 2013, 15 (4), 174–182.  (188) Khachigian, L. M. J. Clin. Invest. 2000, 106 (10), 1189–1195.  (189) Chan, C. W. S.; Khachigian, L. M. Intern. Med. J. 2009, 39 (4), 249–251.  (190) Burgess, D. J. Nat. Rev. Cancer 2012, 12 (8), 508–509.  (191) Deleavey, G. F.; Damha, M. J. Chem. Biol. 2012, 19 (8), 937–954.  (192) Warashina, M.; Kuwabara, T.; Nakamatsu, Y.; Taira, K. Chem. Biol. 1999, 6 (4), 237–250.  (193) Dass, C. R.; Saravolac, E. G.; Li, Y.; Sun, L.-Q. Antisense Nucleic Acid Drug Dev. 2002, 12 (5), 289–299.  (194) Schubert, S.; Gül, D. C.; Grunert, H. P.; Zeichhardt, H.; Erdmann, V. A.; Kurreck, J. Nucleic Acids Res. 2003, 31 (20), 5982–5992.  (195) Edwards, S. L.; Poongavanam, V.; Kanwar, J. R.; Roy, K.; Hillman, K. M.; Prasad, N.; Leth-Larsen, R.; Petersen, M.; Marušič, M.; Plavec, J.; Wengel, J.; Veedu, R. N. Chem. Commun. 2015, 51 (46), 9499–9502.  (196) Mačková, M.; Boháčová, S.; Perlíková, P.; Poštová Slavětínská, L.; Hocek, M. 218  Chembiochem 2015, 16 (15), 2225–2236.  (197) Macíčková-Cahová, H.; Pohl, R.; Hocek, M. Chembiochem 2011, 12 (3), 431–438.  (198) Fujita, H.; Nakajima, K.; Kasahara, Y.; Ozaki, H.; Kuwahara, M. Bioorg. Med. Chem. Lett 2015, 25 (2), 333–336.  (199) Vaníková, Z.; Hocek, M. Angew. Chem. Int. Ed. 2014, 53 (26), 6734–6737.  (200) Valinluck, V.; Wu, W.; Liu, P.; Neidigh, J. W.; Sowers, L. C. Chem. Res. Toxicol. 2006, 19 (4), 556–562.  (201) Strauss, J. K.; Prakash, T. P.; Roberts, C.; Switzer, C.; Maher, L. J. Chem. Biol. 1996, 3 (8), 671–678.  (202) Braasch, D. A.; Corey, D. R. Biochemistry 2002, 41 (14), 4503–4510.  (203) Milton, S.; Honcharenko, D.; Rocha, C. S. J.; Moreno, P. M. D.; Smith, C. I. E.; Strömberg, R. Chem. Commun. 2015, 51 (19), 4044–4047.  (204) Kichler, A.; Mason, A. J.; Bechinger, B. Biochim. Biophys. Acta 2006, 1758 (3), 301–307.  (205) Lan, Y.; Langlet-Bertin, B.; Abbate, V.; Vermeer, L. S.; Kong, X.; Sullivan, K. E.; Leborgne, C.; Scherman, D.; Hider, R. C.; Drake, A. F.; Bansal, S. S.; Kichler, A.; Mason, A. J. Chembiochem 2010, 11 (9), 1266–1272.  (206) Midoux, P.; Pichon, C.; Yaouanc, J.-J.; Jaffrès, P.-A. Br. J. Pharmacol. 2009, 157 (2), 166–178.  (207) Pichon, C.; Roufaï, M. B.; Monsigny, M.; Midoux, P. Nucleic Acids Res. 2000, 28 (2), 504–512.  (208) Midoux, P.; Kichler, A.; Boutin, V.; Maurizot, J. C.; Monsigny, M. Bioconjug. Chem. 1998, 9 (2), 260–267.  (209) Douat, C.; Aisenbrey, C.; Antunes, S.; Decossas, M.; Lambert, O.; Bechinger, B.; Kichler, A.; Guichard, G. Angew. Chem. Int. Ed. 2015, 54 (38), 11133–11137.  (210) Tafech, A.; Bassett, T.; Sparanese, D.; Lee, C. Current Medicinal Chemistry 2006, 13 (8), 863–881.  (211) Isaka, Y. Curr. Opin. Mol. Ther. 2007, 9 (2), 132–136. 219  (212) Hashimoto, H.; Nelson, M.; Switzer, C. J. Am. Chem. Soc. 1993, 115 (16), 7128–7134.  (213) Fischhaber, P. L.; Reese, A. W.; Nguyen, T. Nucleosides Nucleotides 1997, 16 (4), 365–377.  (214) Lam, C.; Hipolito, C.; Perrin, D. M. Eur. J. Org. Chem. 2008, 2008 (29), 4915–4923.  (215) Hollenstein, M.; Hipolito, C.; Lam, C.; Dietrich, D.; Perrin, D. M. Angew. Chem. Int. Ed. 2008, 47 (23), 4346–4350.  (216) Yang, J. H.; Perreault, J.-P.; Labuda, D.; Usman, N.; Cedergren, R. Biochemistry 1990, 29 (51), 11156–11160.  (217) Bonnet, G.; Krichevsky, O.; Libchaber, A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95 (15), 8602–8606.  (218) Kuznetsov, S. V.; Ren, C.-C.; Woodson, S. A.; Ansari, A. Nucleic Acids Res. 2008, 36 (4), 1098–1112.  (219) Smith, F. W.; Feigon, J. Nature 1992, 356 (6365), 164–168.  (220) Blackburn, E. Nature 1991, 350 (6319), 569–673.  (221) Johnson, S. C.; Sherrill, C. B.; Marshall, D. J.; Moser, M. J.; Prudent, J. R. Nucleic Acids Res. 2004, 32 (6), 1937–1941.  (222) Yang, Z.; Chen, F.; Alvarado, J. B.; Benner, S. A. J. Am. Chem. Soc. 2011, 133 (38), 15105–15112.  (223) Joyce, G. F. Pure Appl. Chem. 1993, 65 (6), 1205–1212.  (224) Joyce, G. F. Gene 1989, 82 (1), 83–87.  (225) Chiuman, W.; Li, Y. Chem. Biol. 2006, 13 (10), 1061–1069.  (226) Cadwell, R. C.; Joyce, G. F. PCR Methods Appl. 1994, 3 (6), 136–140.  (227) Eckert, K.; Kunkel, T. PCR Methods Appl. 1991, 1 (1), 17–24.  (228) Ling, L. L.; Keohavong, P.; Dias, C.; Thilly, W. G. PCR Methods Appl. 1991, 1 (1), 63–69.  (229) Beckman, R.; Mildvan, A.; Loeb, L. Biochemistry 1985, 24 (21), 5810–5817. 220  (230) Cadwell, R. C.; Joyce, G. F. Genome Research 1992, 2 (1), 28–33.  (231) Zaborowska, Z.; Fürste, J. P.; Erdmann, V. A.; Kurreck, J. J. Biol. Chem. 2002, 277 (43), 40617–40622.  (232) Kanagawa, T. J. Biosci. Bioeng. 2003, 96 (4), 317–323.  (233) Polz, M. F.; Cavanaugh, C. M. Appl. Environ. Microbiol. 1998, 64 (10), 3724–3730.  (234) Velez, T. E.; Singh, J.; Xiao, Y.; Allen, E. C.; Wong, O. Y.; Chandra, M.; Kwon, S. C.; Silverman, S. K. ACS Comb. Sci. 2012, 14 (12), 680–687.  (235) Legiewicz, M.; Lozupone, C.; Knight, R.; Yarus, M. RNA 2005, 11 (11), 1701–1709.  (236) Huang, F.; Bugg, C.; Yarus, M. Biochemistry 2000, 39 (50), 15548–15555.  (237) Coleman, T. M.; Huang, F. Chem. Biol. 2002, 9 (11), 1227–1236.  (238) Xiao, Y.; Wehrmann, R. J.; Ibrahim, N. A.; Silverman, S. K. Nucleic Acids Res. 2012, 40 (4), 1778–1786.  (239) Schlosser, K.; Lam, J. C. F.; Li, Y. Nucleic Acids Res. 2009, 37 (11), 3545–3557.  (240) Vazin, M.; Huang, P.; Matuszek, Ż.; Liu, J. Biochemistry 2015, 54 (39), 6132–6138.  (241) Esteban, J. A.; Banerjee, A. R.; Burke, J. M. J. Biol. Chem. 1997, 272 (21), 13629–13639.  (242) Hoffmann, B.; Mitchell, G. T.; Gendron, P.; Major, F.; Andersen, A. A.; Collins, R. A.; Legault, P. Proc. Natl. Acad. Sci. U.S.A. 2003, 100 (12), 7003–7008.  (243) Ting, R.; Thomas, J. M.; Perrin, D. M. Canadian Journal of Chemistry-Revue Canadienne De Chimie 2007, 85 (4), 313–329.  (244) Zamel, R.; Poon, A.; Jaikaran, D.; Andersen, A.; Olive, J.; De Abreu, D.; Collins, R. A. Proc. Natl. Acad. Sci. U.S.A. 2004, 101 (6), 1467–1472.  (245) Werner, M.; Uhlenbeck, O. C. Nucleic Acids Res. 1995, 23 (12), 2092–2096.  (246) Chandrasekar, J.; Wylder, A. J. Am. Chem. Soc. 2015, 137 (30), 9575–9578.  221  (247) Xu, J.; Carrocci, T. J.; Hoskins, A. A. Chem. Commun. 2016, 52 (3), 549–552.  (248) Schlosser, K.; Li, Y. Chembiochem 2010, 11 (7), 866–879.  (249) Ota, N.; Warashina, M.; Hirano, K.; Hatanaka, K.; Taira, K. Nucleic Acids Res. 1998, 26 (14), 3385–3391.  (250) Casey, J.; Davidson, N. Nucleic Acids Res. 1977, 4 (5), 1539–1552.  (251) Doudna, J. A.; Cech, T. R. Nature 2002, 418 (6894), 222–228.  (252) Ferré-D'Amaré, A. R.; Scott, W. G. Cold Spring Harb Perspect Biol 2010, 2 (10), a003574–a003574.  (253) Silverman, S. K. Angew. Chem. Int. Ed. 2010, 49 (40), 7180–7201.  (254) Schroeder, G.; Lad, C.; Wyman, P. Natl. Acad. Sci. U.S.A. 2006, 103 (6), 4052–4055.  (255) Kath-Schorr, S.; Wilson, T. J.; Li, N.-S.; Lu, J.; Piccirilli, J. A.; Lilley, D. M. J. J. Am. Chem. Soc. 2012, 134 (40), 16717–16724.  (256) Perrotta, A. T.; Shih, I.; Been, M. D. Science 1999, 286 (5437), 123–126.  (257) Wilson, T. J.; Li, N.-S.; Lu, J.; Frederiksen, J. K.; Piccirilli, J. A.; Lilley, D. M. J. Proc. Natl. Acad. Sci. U.S.A. 2010, 107 (26), 11751–11756.  (258) Zhang, N.; Zhang, S.; Szostak, J. W. J. Am. Chem. Soc. 2012, 134 (8), 3691–3694.  (259) Nowakowski, J.; Shim, P. J.; Prasad, G. S.; Stout, C. D.; Joyce, G. F. Nat. Struct. Biol. 1999, 6 (2), 151–156.  (260) Thomas, J. M.; Ting, R.; Perrin, D. M. Org. Biomol. Chem. 2004, 2 (3), 307.  (261) Chu, C.-C.; Wong, O. Y.; Silverman, S. K. Chembiochem 2014, 15 (13), 1905–1910.  (262) Jakobsen, M. R.; Haasnoot, J.; Wengel, J.; Berkhout, B.; Kjems, J. Retrovirology 2007, 4, 29.  (263) Prater, C. E.; Miller, P. S. Bioconjug. Chem. 2004, 15 (3), 498–507.  (264) Lewis, J. G.; Lin, K. Y.; Kothavale, A.; Flanagan, W. M.; Matteucci, M. D.; DePrince, R. B.; Mook, R. A.; Hendren, R. W.; Wagner, R. W. Proc. Natl. 222  Acad. Sci. U.S.A. 1996, 93 (8), 3176–3181.  (265) Helbling-Leclerc, A.; Scherman, D.; Wils, P. Biochim. Biophys. Acta 1999, 1418 (1), 165–175.  (266) Yoo, H.; Juliano, R. L. Nucleic Acids Res. 2000, 28 (21), 4225–4231.  (267) Robaczewska, M.; Guerret, S.; Remy, J. S.; Chemin, I.; Offensperger, W. B.; Chevallier, M.; Behr, J. P.; Podhajska, A. J.; Blum, H. E.; Trepo, C.; Cova, L. Gene Ther. 2001, 8 (11), 874–881.  (268) Boussif, O.; Lezoualc'h, F.; Zanta, M. A.; Mergny, M. D.; Scherman, D.; Demeneix, B.; Behr, J. P. Proc. Natl. Acad. Sci. U.S.A. 1995, 92 (16), 7297–7301.  (269) Gasparini, G.; Bang, E.-K.; Montenegro, J.; Matile, S. Chem. Commun. 2015, 51 (52), 10389–10402.  (270) Sebbage, V. Bioscience Horizons 2009, 2 (1), 64–72.  (271) Jones, S. W.; Christison, R.; Bundell, K.; Voyce, C. J.; Brockbank, S. M. V.; Newham, P.; Lindsay, M. A. Br. J. Pharmacol. 2005, 145 (8), 1093–1102.  (272) Gilleron, J.; Querbes, W.; Zeigerer, A.; Borodovsky, A.; Marsico, G.; Schubert, U.; Manygoats, K.; Seifert, S.; Andree, C.; Stöter, M.; Epstein-Barash, H.; Zhang, L.; Koteliansky, V.; Fitzgerald, K.; Fava, E.; Bickle, M.; Kalaidzidis, Y.; Akinc, A.; Maier, M.; Zerial, M. Nat. Biotechnol. 2013, 31 (7), 638–646.  (273) Yuan, A.; Laing, B.; Hu, Y.; Ming, X. Chem. Commun. 2015, 51 (30), 6678–6680.  (274) Thompson, D. B.; Villaseñor, R.; Dorr, B. M.; Zerial, M.; Liu, D. R. Chem. Biol. 2012, 19 (7), 831–843.  (275) Koller, E.; Vincent, T. M.; Chappell, A.; De, S.; Manoharan, M.; Bennett, C. F. Nucleic Acids Res. 2011, 39 (11), 4795–4807.  (276) Macíčková-Cahová, H.; Hocek, M. Nucleic Acids Res. 2009, 37 (22), 7612–7622.  (277) Kielkowski, P.; Brock, N. L.; Dickschat, J. S.; Hocek, M. Chembiochem 2013, 14 (7), 801–804.  (278) Geary, R. S.; Norris, D.; Yu, R.; Bennett, C. F. Advanced Drug Delivery Reviews 2015, 87, 46–51. 223  (279) Sefah, K.; Shangguan, D.; Xiong, X.; O'Donoghue, M. B.; Tan, W. Nat. Protoc. 2010, 5 (6), 1169–1185.  (280) Rohloff, J. C.; Gelinas, A. D.; Jarvis, T. C.; Ochsner, U. A.; Schneider, D. J.; Gold, L.; Janjic, N. Mol. Ther. Nucleic Acids 2014, 3, e201.  (281) Davies, D. R.; Gelinas, A. D.; Zhang, C.; Rohloff, J. C.; Carter, J. D.; O'Connell, D.; Waugh, S. M.; Wolk, S. K.; Mayfield, W. S.; Burgin, A. B.; Edwards, T. E.; Stewart, L. J.; Gold, L.; Janjic, N.; Jarvis, T. C. Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (49), 19971–19976.  (282) Lam, C. H.; Perrin, D. M. Bioorg. Med. Chem. Lett. 2010, 20 (17), 5119–5122.  (283) Deglane, G.; Abes, S.; Michel, T.; Prévot, P.; Vives, E. Chembiochem 2006, 7 (4), 684–692.  (284) Moran, S.; Ren, R. X.; Kool, E. T. Proc. Natl. Acad. Sci. U.S.A. 1997, 94 (20), 10506–10511.  (285) Goodman, M. F.; Creighton, S.; Bloom, L. B.; Petruska, J. Crit. Rev. Biochem. Mol. Biol. 1993, 28 (2), 83–126.  (286) Bryant, F. R.; Johnson, K. A.; Benkovic, S. J. Biochemistry 1983, 22 (15), 3537–3546.  (287) Krüger, S.; Meier, C. Eur. J. Org. Chem. 2013, 2013 (6), 1158–1169.  (288) Höfler, K.; Sarac, I.; Meier, C. Chembiochem 2015, 16 (14), 2046–2053.  (289) Prakash, T. P.; Püschl, A.; Manoharan, M. Nucleosides Nucleotides & Nucleic Acids 2007, 26 (2), 149–159.  (290) Uhlenbeck, O. C. RNA 2003, 9 (12), 1415–1417.  (291) Santoro, S.; Joyce, G. F.; Sakthivel, K.; Gramatikova, S.; Barbas, C. J. Am. Chem. Soc. 2000, 122 (11), 2433–2439.  (292) Loakes, D.; Holliger, P. Chem. Commun. 2009, No. 31, 4619–4631.  (293) Pinheiro, V. B.; Taylor, A. I.; Cozens, C.; Abramov, M.; Renders, M.; Zhang, S.; Chaput, J. C.; Wengel, J.; Peak-Chew, S.-Y.; McLaughlin, S. H.; Herdewijn, P.; Holliger, P. Science 2012, 336 (6079), 341–344.  (294) Yu, H.; Zhang, S.; Dunn, M. R.; Chaput, J. C. J. Am. Chem. Soc. 2013, 135 (9), 3583–3591. 224  (295) Yu, H.; Zhang, S.; Chaput, J. C. Nat. Chem. 2012, 4 (3), 183–187.  (296) Taylor, A. I.; Pinheiro, V. B.; Smola, M. J.; Morgunov, A. S.; Peak-Chew, S.; Cozens, C.; Weeks, K. M.; Herdewijn, P.; Holliger, P. Nature 2014, 518 (7539), 427–430.            225  Appendices Appendix A  Chapter 4  Figure A1 Bright field images of cells treated with 0.9 µM of Dz7-38-32t-m at increasing time points up to 45 h of incubation time. Scale bar: 25 µm. Dz7-38-32t-m 0 min Dz7-38-32t-m 180 minDz7-38-32t-m 26 hDz7-38-32t-m 45 hDz7-38-32t-m 19 h226    Figure A2 Stability of Dz7-38-32t-m and Dz7-38-32t-u in Opti-MEM at 37°C. Time points were taken at 0 min, 30 min, 330 min, 22 h, 26 h, 30h, and 48 h, respectively.               Time (0 min, 30 min, 330 min, 22 h, 26 h, 30 h, 48h)Dz7-38-32t-m Dz7-39-32t-u227  Appendix B  Chapter 5   Figure A3 Denaturing purification PAGE (10%, 7M urea) for ON 1, ON 2 and ON 3. (A) Sequences of ON 1, ON 2 and ON 3, respectively. (B) Gel picture visualized by UV-shadowing. Lane 1, ON 3 containing no modified nucleoside; lane 2, ON 1 containing modified dAim; lane 3, ON 2 containing modified dAim, dCaa and dUga. Bands representing full-length products are cut out, as indicated by the arrow.     ON 1 ON 2ON 31 2 35'-CmCmCmTGCGGAGGGGCTGCCAGTAimGTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 15'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 25'-CmCmCmTGCGGAGGGGCTGCCAGTA    GTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3AB228    Figure A4 Thermal melting study at pH 6. (A) Oligonucleotide sequences used for thermal denaturation study. ON 1, ON 2, and ON 3 in duplex with a complementary strand ON 4 for thermal melting studies. (B) Thermal melting curves of three separate experiments of forward and reverse melting processes. 5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3'3'-GACGGTCAT   CAGAACACC-5'ON 1ON 45'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 25'-CmCmCmTGCGGAGGGGCTGCCAGTA    GTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3A20 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 6-1st-forward ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 6-1st-reverse ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 6-2nd-forward ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 6-2nd-reverse ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 6-3rd-forward ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 6-3rd-reverse ON 3ON 1ON 2B229    Figure A5 Thermal melting study at pH 8.5. (A) Oligonucleotide sequences used for thermal denaturation study. ON 1, ON 2, and ON 3 in duplex with a complementary strand ON 4 for thermal melting studies. (B) Thermal melting curves of three separate experiments of forward and reverse melting processes. 5'-CmCmCmTGCGGAGGGGCTGCCAGTAimGTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3'3'-GACGGTCAT   CAGAACACC-5'ON 1ON 45'-CmCmCmTGCGGAGGGGCTGCCAGTAimGUgaCaaTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 25'-CmCmCmTGCGGAGGGGCTGCCAGTA    GTCTTGTGGCGTTCGTTTGTTCGCGGCGCGCTInv-3' ON 3A20 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 8.5-1st-forward ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 8.5-1st-reverse ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 8.5-2nd-forward ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 8.5-2nd-reverse ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 8.5-3rd-reverse ON 3ON 1ON 220 30 40 50 60 70 80 900.350.400.450.500.550.600.65TemperatureA260pH 8.5-1st-forward ON 3ON 1ON 2B

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0229566/manifest

Comment

Related Items