UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Novel catalysis, photocaging, and positron emission tomography through bioconjugate chemistry Ting, Richard 2007

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2008_spring_ting_richard.pdf [ 9.67MB ]
Metadata
JSON: 24-1.0061555.json
JSON-LD: 24-1.0061555-ld.json
RDF/XML (Pretty): 24-1.0061555-rdf.xml
RDF/JSON: 24-1.0061555-rdf.json
Turtle: 24-1.0061555-turtle.txt
N-Triples: 24-1.0061555-rdf-ntriples.txt
Original Record: 24-1.0061555-source.json
Full Text
24-1.0061555-fulltext.txt
Citation
24-1.0061555.ris

Full Text

NOVEL CATALYSIS, PHOTOCAGING, AND POSITRON EMISSION TOMOGRAPHY THROUGH BIOCONJUGATE CHEMISTRY by RICHARD TING B.Sc. (Honors), University of British Columbia, 2000 A THESIS SUBMITTED iN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA December 2007 © Richard Ting, 2007 ABSTRACT This thesis is comprised of three projects that serve to address current topics in the field of bioorganic chemistry. Chapter 1 describes the emulation of enzyme catalysis through the kinetic analysis of the DNAzyme 925-1 it, a combinatorially selected RNase A mimic utilizing imiclazole and amine groups hybridized to DNA. The rate constants measured for this system are the largest to date forM2-independent self cleavage (0.020 mhi1), trans cleavage (0.28 ± 0.02 min’), and multiple turnover (0.030 ± 0.002 min’) by a biomimetic system at physiological ionic strength and pH. These constants rival most combinatorially selected metal dependent DNAzymes and naturally occurring ribozymes even at physiological concentrations ofM2. Chapters 2 and 3 discuss a novel photochemical motif, its application to biologically relevant molecules, characterization of the photochemical mechanism, and its utility in the generation of alkenes. Light is considered superior to other chemical reagents as its spatial and temporal properties can be precisely controlled and its penetrative ability makes it a perfect reagent for the non- invasive perturbation of cellular processes. Chapter 2 details the discovery of a novel photochemical reaction and its use in photochemically regulating gene function and nucleic acid chemistry. The chemistry described in Chapter 2 holds potential for the photocaging of all adenine substrates, cofactors, and products in biological systems. Chapter 3 identifies the photolytic mechanism and highlights its application to photolytic alkene synthesis. It is predicted that the photolytic thioether mechanism identified in this chapter can be extrapolated to the photolysis of a wide range of other aromatic thioethers. Chapter 4 discusses the application of the ‘8F acceptor, boron, to most sensitive of in vivo molecular imaging techniques: positron emission tomography. This aqueous approach simplifies the state of the art by multiple chemical steps and multiplies the final specific radioactivity of the final radiotracer by a factor of 3. This tool is expected to widen the currently limited scope of biomarkers available for in vivo imaging and will enhance our ability to image biochemical targets and pathways such that insight may be gained in the progression, diagnosis, and treatment of disease. 11 TABLE OF CONTENTS ‘1’L]3IJE OF 1N[’IiN’I’S .....................................................................iii I_dISrf OF nI4i3iFs............................................................................... xv I41srf OF AJND SY1’1BOI..S ..... ... .... ... . . . . . .. . ...... . . . . . . . . . . . . .xxxiii xl xli (2 ()_4{Jrj( I{Sf{IP ... .. ... ... ... . ... ........ ...... .... ....... .. .... xiii CHAPTER 1: AN IMIDAZOLE AND CATIONIC AMINE 1’1ODIF’IED E]NA.SE 4. 1’1II’1IC ......................................................... 1 1.1 INTRODUCTION 1 1.1.1 RNase A 1 1.1.1.1 EarlyRNaseAmimics 3 1.1.1.2 Sequence specific RNase A mimi 3 1.1.1.3 Drawbacks to rational approaches to generating RNase A mimics 4 1.1.2 Nucleic acid catalysts 5 1.1.3 In vitro selected ribozymes 6 1.1.4 In vitro selected deoxyribozymes 8 1.1.5 Drawbacks of ribozymes and deoxyribozymes 9 1.1.5.1 Nucleic acid catalysts require divalent metals for fast activity 10 1.1.6 Chemically modified deoxyribozymes 11 1.1.6.1 The successful selection of modified nucleic acid systems 12 1.1.7 Proposed research 13 111 1.2 SYNTHESIZING925-llt AS A CATALYST .14 1.2.1 MATERIALS AND METHODS 17 1.2.1.1 Synthetic methods 17 1.2.1.1.1 Enzymes and Chemicals 17 1.2.1.1.2 The cis-cleaving sequence 925-11 c 17 1.2.1.1.3 Oligonucleotides 18 1.2.1.1.4 2P-oligonucleotide labeling 18 1.2.1.2 General kinetic procedures 18 1.2.1.2.1 Erroranalysis 19 1.2.1.3 Cis cleavage by 925-11 c 20 1.2.1.3.1 Determination of the observed single turnover rate constant, keat STR. . .20 1.2,1.4 Trans cleavage by925-i it under multiple turnover (MTR) conditions 20 1.2.1.5 Estimation of the maximum multiple turnover rate, kCatMTR 20 1.2.2 RESULTS 22 1.2.2.1 The kinetic profile of the cis-cleaving substrate, 925-i ic 22 1.2.2.2 The synthesis of a catalytic sequence, 925-i it 24 1.2.2.3 The temperature dependence of the catalyst, 925-1 it 26 1.2.3 DISCUSSION 28 1.3 THE KINETICS ANALYSIS OF THE CATALYST925-llt at 24°C 30 1.3.1 MATERIALS AND METHODS 32 1.3.1.1 General kinetic procedures 32 1.3.1.1.1 Error analysis 32 1.3.1.2 Trans cleavage by925-lit under single turnover (sTR) conditions 33 1.3.1.2.1 Determination of the observed single turnover rate constant, kobs STR 33 1.3.1.2.2 Determination of the maximum first—order rate constant (koat STR) and the catalyst concentration at which the rate is half-maximal (KM 5TR) under single turnover conditions 34 1.3.1.2.3 Determination of the substrate dissociation rate constant (k..i, k..1 fast, and k..1 Denaturing methods on the substrate, Si 34 1.3.1.2.4 Determination of the substrate dissociation rate constant (k..1 S1-OMe, k..1 fast 51-OMe, and k..1 Si-OMe SiOW). Non-denaturing methods on a non-hydrolysable substrate analogue 35 1.3.1.2.5 Determination of equilibrium substrate dissociation rate constants 36 1.3.1.2.6 Determination of product dissociation rate constants (k..3, k..4,k5, and k. 6) 37 1.3.1.3 Trans cleavage under multiple turnover (MTR) conditions 37 1.3.1.3.1 Determination of the maximum multiple turnover rate, kcat MTR, under conditions where no burst phase is observed 37 i .3.1.4 Attempts at observing a reverse reaction (ligation) 38 1.3.1.4.1 Single Turnover Product Ligation Experiments 38 1.3.1.4.2 External Equilibrium Shift Experiments 38 iv 1.3.2 RESULTS .39 1.3.2.1 Determination of the maximum first—order rate constant at saturating catalyst concentrations, (koat STR), and the concentration of catalyst at which the reaction rate is half-maximal (KM sTR) 39 1.3.2.2 Determination of the substrate Si dissociation rate constant (k-1) through denaturing polyacrylamide gel methods 40 1.3.2.3 Determination of the substrate dissociation rate constant (k..1 Si-OMe). Non- denaturing methods on a non-hydrolysable substrate analogue 42 1.3.2.4 24 °C Multiple turnover kinetic profiles 43 1.3.2.5 Gel-shift assays for equilibrium dissociation constants of catalyst-substrate analogue complexes (Kd) 44 1.3.2.6 Gel-shift assays for dissociation rate constants of product oligonucleotides (lc ,k4lcs,andko) 45 1.3.2.7 24 °C Substrate specificity studies 46 1.3.2.8 Effects of 2 mMMgon9s-11tat24 °C 47 1.3.3 DISCUSSION 49 1.3.3.1 The kinetic model for 925-1 it at 24°C 49 1.3.3.2 The importance of catalyst excess conditions (s-ni) vs. substrate excess conditions (MTR) 50 1.3.3.3 Regarding ligation 51 1.3.3.4 925-1 it Substrate specificity 52 1.3.3.5 Comparison to other nucleotide catalysts 53 1.4 THE KINETIC ANALYSIS OF THE CATALYST925-llt at 13°C 56 1.4 THE KINETIC ANALYSIS OF THE CATALYST925-llt at 13°C 56 1.4.1 MATERIALS AND METHODS 57 1.4.1.1 Trans cleavage by925-i it under single turnover (STR) conditions 57 1.4.1.2.1 Determination of the observed single turnover rate constant, kobs STR. .57 1.4.1.2.2 Determination of the maximum first-order rate constant (kcat s’ri) and - the catalyst concentration at which the rate is half-maximal (KM sTR) under single turnover conditions 58 1.4.1.2.3 Determination of the substrate dissociation rate constant (k-1,k-1 ft, and k..1 siow). Denaturing methods on the substrate, Si 58 1.4.1.2.4 Determination of the substrate dissociation rate constant (k-1 Si-OMe, k..1 fast Si-OMe, and k-1 Si-OMe Non-denaturing methods on a non-hydrolysable substrate analogue 58 1.4.1.2.5 Determination of equilibrium substrate dissociation rate constants 59 1.4.1.2.6 Determination of product dissociation rate constants (k-3, k..4,k5, and k.. 6) 59 1.4.1.2 Trans-cleavage under multiple turnover (MTR) conditions 60 1.4.1.2.1 Determination of the maximum multiple turnover rate, kcatMTR, under conditions where burst phase kinetic profiles are observed 60 1.4.1.3 Attempts at observing a reverse reaction (ligation) 61 v 1.4.2 RESULTS .62 1.4.2.1 Determination of the maximum first—order rate constant at saturating catalyst concentrations, (kcat sm), and the concentration of catalyst at which the reaction rate is half-maximal (K sm.) 62 1.4.2.2 Attempts to restore monophasic kinetic profiles 65 1.4.2.3 Investigating the nature of the biphase 66 1.4.2.4 Explaining the biphasic nature of925-ilt at 13°C 68 1.4.2.5 Explaining the biphasic nature of 925-1 it at 13 °C. Determination of the substrate dissociation rate constant. (k-1 fast, and k-i slow). Denaturing methods on the substrate, Si 70 1.4.2.6 Explaining the biphasic nature of 925-1 it at 13 °C. Determination of the substrate dissociation rate constant (k..i Si-OMe fast and k-1 Si-OMe slow). Non-denaturing methods on a non-hydrolysable substrate analogue 71 1.4.2.7 Multiple turnover cleavage of SI by 925-i it at 13 °C 72 1.4.2.8 Slow product release does not account for the burst phase 74 1.4.2.9 Native PAGE multiple turnover analysis provides evidence against slow product release 75 1.4.2.10 Attempts at measuring a ligation rate for925-i it 77 1.4.2.11 Temperature cycling enhances the rate of multiple turnover and demonstrates catalyst stability 78 1.4.3 DISCUSSION 80 1.4.3.1 The kinetic model for925-iit at 13°C 80 1.4.3.1.1 The kinetic model for Si cleavage by 925-i it at 13 °C 82 1.4.3.1.2 Attempts at Linear Multiple Turnover. Kinetics with Different substrates 85 1.4.3.2 Exploiting the different temperature optima. Temperature cycling 85 1.5 CONCLUSIONS 87 1.6 REFERENCES 89 CHAPTER 2: ARYL THIOETHERS AS NOVEL PHOTOCAGING GROUPS FOR GENE REGULATION............94 2.1 INTRODUCTION 94 2.1.1 Photocaging 96 2.1.2 Irreversible photocaging 97 2.1.3 Drawbacks of current photocaging strategies 99 2.1.4 Photo regulation of gene expression 100 2.1.5 Proposed Research 103 vi 2.2 THE DISCOVERY OF A NOVEL PHOTOCHEMICAL REACTION 104 2.2.1 MATERIALS AND METHODS 106 2.2.1.1 General synthetic methods 106 2.2.1.2 Specific synthetic methods 107 2.2.1.2.1 Triphosphate preparation 109 2.2.1.2.2 Imidazole ethyl thioether adenosine preparation 114 2.2.1.2.3 Deoxyribophosphoramidite preparation 116 2.2.1.3 Enzymatic, Kinetic, and DNA preparation procedures 118 2.2.1.3.1 Enzymatic incorporation assays 118 2.2.1.3.2 Synthesis and Purification of Substrates and Modified Catalysts 119 2.2.1.3.3 DNA MALDI-TOF mass spectrometry 119 2.2.1.3.4 Photochemical Procedures 120 2.2.2 RESULTS 121 2.2.2.1 The photolysis of C-8 thioethers of adenosine gives adenine 121 2.2.2.1.1 Photolysis of the triphosphate 2.5 121 2.2.2.1.2 Photolysis of the ribonucleoside 2.6 122 2.2.2.2 Solvent isotope dependence for the photolysis of 2.6 125 2.2.2.3 A CONTROL REACTION. Photo-excitation of unmodified adenine does not promote H/Deuterium exchange on C-8 of adenosine 127 2.2.2.4 The photoactivation of adenine in DNA synthesized with the thioether phosphoramidite 2.8 128 2.2.3 DISCUSSION 130 2.2.3.1 The mechanism of photoactivation 130 2.2.3.2 Contributions to photocaging functionality 131 2.3 THE APPLICATION OF IMIDAZOLE THIOETHER ADENINE AS A PHOTOCAGING STRATEGY FOR THE DNAzyme, 17E 132 2.3.1 MATERIALS AND METHODS 134 2.3.1.1 Synthesis and purification of substrates and modified catalysts 134 2.3.1.2 DNA MALDI-TOF mass spectrometry 135 2.3.1.3 Kinetic assays 136 2.3.1.4 Photochemical procedures 137 2.3.2 RESULTS 138 2.3.2.1 Investigating the background activity of 17E and its derivatives ml through m4 138 2.3.2.2 Mass verification of 17E and 17E-ml synthesis 139 2.3.2.3 The photolytic activation of the DNAzyme, 17E-ml 140 2.3.2.3.1 Gel and kinetic proof of 17E photoactivation 140 2.3.2.3.2 MALDI-TOF verification of the irradiative conversion of 17E-ml into 17E 143 2.3.3 DISCUSSION 144 vii 2.3.3.1 Research implications 144 2.4 CONCLUSIONS 145 2.4.1 Future work 145 2.5 REFERENCES 148 CHAPTER 3: CLEAN ALKENE GENERATION FROM ARYL THIOETEIERS ............................................................................ 150 3.1 INTRODUCTION 150 3.1.1 Alkene generation in organic chemistry 150 3.1.2 Radical generation of alkenes 151 3.1.3 Photolytic generation of aikenes 151 3.1.4 Photolysis of sulfoxides and thioethers 152 3.1.5 Proposed Research 156 3.2 THE INTERMEDIATES AND PRODUCTS OF THIOETHER PHOTOLYSIS 160 3.2.1 MATERIALS AND METHODS 160 3.2.1.1 General synthetic methods 160 3.2.1.2 Specific synthetic procedures 161 3.2.1.3 Photochemical procedures 164 3.2.2 RESULTS 166 3.2.2.1 The intermediate of photolysis. Photolysis of 3.1 proceeds through an 8- mercapto adenine intermediate 166 3.2.2.2 The intermediate of photolysis. Photolysis of 3.1 proceeds through an 8- mercapto adenine intermediate 3.1-I 168 3.2.2.3 The product of photolysis. Ethene is evolved through the photolysis of 3.1. 170 3.2.3 DISCUSSION 171 3.2.3.1 On the mechanism of photolysis 172 3.2.3.1.1 Regarding N-9 adenine, or N-Benzimidazole substitution 172 3.2.3.1.2 On the generation of an alkene 172 3.2.3.1.3 Onthe fate ofthe sulfur 173 3.3 GENERALIZATION OF PHOTOLYSIS TO OTHER CHROMOPHORES 174 3.3.1 MATERIALS AND METHODS 174 3.3.1.1 General synthetic methods 174 viii 3.3.1.2 Specific synthetic procedures 175 3.3.1.3 Photochemical procedures 176 3.3.2 RESULTS 176 3.3.2.1 UV-Vis absorbance characterization of the photolysis of 3.2 176 3.3.2.1.1 Ethene is evolved through the photolysis of 3.2 178 3.3.3 DISCUSSION 179 3.3.3.1 On the generalizability of thioether photolysis 179 3.4 THE MECHANISM OF ARYL THIOETHER PHOTOLYSIS 181 3.4.1 MATERIALS AND METHODS 183 3.4.1.1 Specific synthetic procedures 183 3.4.1.2 Photochemical procedures 186 3.4.2 RESULTS 186 3.4.2.1 UV-Vis absorbance characterization of the photolysis of 3.3 186 3.4.2.1.1 3.3-P, N-Ethenyl-benzimidazole, and 3.3-I, 2-Mercapto-N-ethenyl- benzimidazole, are generated in the photolysis of 3.3 187 3.4.2.2 Verification of the products resulting from the photolysis of 3.3 188 3.4.3 DISCUSSION 190 3.4.3.1 Evidence against a Norrish-like mechanism 190 3.4.3.2 On the mechanism of thioether photolysis 191 3.5 PIIOTOCHEMICAL GENERALIZATION TOWARDS THE PRODUCTION OF OTHER ALKENES 192 3.5.1 MATERIALS AND METHODS 193 3.5.1.1 Specific synthetic procedures 193 3.5.1.2 Photochemical procedures 195 3.5.2 RESULTS 196 3.5.2.1 UV-Vis absorbance characterization of the photolysis of 3.4 196 3.5.2.1.1 1-Butene is evolved through the photolysis of 3.4 197 3.5.2.2 UV-Vis absorbance characterization of the photolysis of 3.5 198 3.5.2.2.1 ‘H-NMR Characterization of the photolysis of 3.5 199 3.5.2.3 UV-Vis absorbance characterization of the photolysis of 3.6 200 3.5.2.3.1 2-Cyclohexenone is generated by the photolysis of 3.6 201 3.5.3 DISCUSSION 202 3.5.3.1 On alkene generation 202 3.5.3.2 On the synthesis of a-13 unsaturated ketones and esters 203 3.6 THE QUANTUM YIELDS OF PHOTOLYSIS 204 3.6.1 MATERIALS AND METHODS 205 ix 3.6.1.1 Determination of the incident photon intensity, Po, that is transferred through a quartz cell using the handheld 254 nm irradiation source 205 3.6.1.2 Determination of the quantum efficiency of the photolysis reaction, 4 , at 254 nm 205 3.6.1.3 Determining the constant, kobs and the choice of wavelength for measuring quantum yields 206 3.6.2 RESULTS 209 3.6.2.1 Quantum yield for the photolysis of 3.1 209 3.6.2.2 Quantum yield for the photolysis of 3.2 210 3.6.2.3 Quantum yield for the photolysis of 3.3 211 3.6.2.4 Quantum yield for the photolysis of 3.4 212 3.6.2.5 Quantum yield for the photolysis of 3.5 213 3.6.2.6 Quantum yield for the photolysis of 3.6 214 3.6.3 DISCUSSION 215 3.7 CONCLUSIONS 217 3.7.1 FUTURE STUDIES 218 3.8 REFERENCES 220 CHAPTER 4: BORON BASED STRATEGIES FOR 1’IOI4FCULAR IfvIi-.C-INC-............................................................... 221 4.1 INTRODUCTION 221 4.1.1 Current medical imaging techniques 222 4.1.2 Positron emission tomography (PET) 224 4.1.2.1 Choice of PET isotope 225 4.1.2.2 2-[18F] fluoro-2-deoxy-D-glucose 227 4.1.3 Current technologies for PET labeling biomolecules 229 4.1.3.1 ‘8F labeling strategies 229 4.1.3.2 The state of the art in PET labeling biomolecules 231 4.1.3.3 Limitations of current technology 233 4.1.4 A new PET labeling strategy 234 4.1.4.1 Radiochemical requirements of a PET strategy 236 4.1.4.2 Biological requirements 237 4.2 THE AQUEOUS STABILITY OF 18F ARYL TRIFLUOROBORATES DEPENDS ON THE ARYL SUBSTITUENTS 238 4.2.1 MATERIALS AND METHODS 242 x 4.2.1.1 General synthetic methods 242 4.2.1.2 Radiochemical procedures 242 4.2.1.2.1 18F preparation 242 4.2.1.2.2 Trifluoroborate formation 242 4.2.1.2.3 ‘8F TLC Trifluoroborate formation rate experiments 243 4.2.1.2.4 ‘8F TLC Trifluoroborate isotopic exchange rate experiments 243 4.2.1.3 Kinetic procedures 246 4.2.1.3.1 Buffered aqueous fluoride ‘9F NMR spectroscopy dissociation rate experiments 246 4.2.2 RESULTS 246 4.2.2.1 The syntheses of trifluoroborates in Table 4.2 are rapid with respect to the half-life of‘8F 247 4.2.2.2 The stability of‘8F trifluoroborates in 200 mM pH 7.5 buffered phosphate depends on the aryl substituents 247 4.2.2.3 ‘9F NMR spectroscopic verification of TLC trifluoroborate isotopic fluoride exchange kinetics — 3 representative examples 254 4.2.2.3.1 ‘9F NMR spectroscopy of fluoride dissociation from 4- carboxyphenyltrifluoroborate 254 4.2.2.3.2 ‘9F NMR spectroscopy of fluoride dissociation from 4- sulphonamidylphenyl trifluoroborate 257 4.2.2.3.3 ‘9F NMR spectroscopy of fluoride dissociation from 2,4,6- trifluorophenyl trifluoroborate 260 4.2.2.4 Hammet plots of fluoride loss from aryl trifluoroborates 264 4.2.3 DISCUSSION 270 4.2.3.1 Labeling is rapid with respect to the half-life of ‘8F 270 4.2.3.2 The mechanism of aryl B-’8F trifluoroborate dissociation and association .270 4.2.3.3 Some aryl B-’8Ftrifluoroborates are highly stable 272 4.2.3.4 Towards stable trifluoroborates for use in PET 273 4.3 THE SYNTHESIS OF A ‘8F TRIFLUOROBORATE WHICH MEETS THE RADIOCHEMICAL REQUIREMENTS OF A PET LABELING STRATEGY 274 4.3.1 MATERIALS AND METHODS 278 4.3.1.1 General synthetic methods 278 4.3.1.2 Specific synthetic methods 279 4.3.1.2.1 Boronic Esters 279 4.3.1.2.2 Trifluoroborates 286 4.3.1.3 Radiochemical procedures 288 4.3.1.3.1‘8Fpreparation 288 4.3.1.3.2 Trifluoroborate formation 288 4.3.2 RESULTS 289 4.3.2.1 The boron containing electron-withdrawing carboxylate 4.1 meets the fluoridation, purification, and stability requirements for a PET label 289 4.3.2.1.1 4.1 can be rapidly converted to the trifluoroborate 4.1 TFB 289 xi 4.3.2.1.2 4.1 TFB is easily pufied .290 4.3.2.1.3 4.1 TFB exchanges ‘8F with its environment at a rate of 0.00093 ± 0.00009 min1 (ti,2 =744±70 mm) 292 4.3.2.2 The amide “conjugate” 4.2 meets purification and stability requirements for a PET label 293 4.3.2.2.1 4.2 can be transformed to the trifluoroborate 4.2 TFB 294 4.3 .2.2.2 4.2 TFB is easily purified 295 4.3 .2.2.3 4.2 TFB exchanges ‘8F with its environment at a rate of 0.00012 ± 0.00004 mm4 (t112 = 5550 ± 1745 mm) 298 4.3.3 DISCUSSION 300 4.3.3.1 Trifluoroborate formation 300 4.3.3.2 Trifluoroborates can be easily separated from free fluoride 301 4.3.3.3 Some aryl B-’8Ftrifluoroborates are highly stable 301 4.4 IN VIVO IMAGING OF‘8F-BIOTIN AND18F-BIOTIN LABELED NEUTRAVIDI1’4 304 4.4.1 MATERIALS AND METHODS 305 4.4.1.1 General synthetic methods 305 4.4.1.2 Specific synthetic methods 305 4.3.1.2.1 Trifluoroborates 307 4.4.1.3 Radiochemical procedures 308 4.4.1.3.1‘8Fpreparation 308 4.4.1.3.2 Low-activity radiochemical synthesis of the trifluoroborate of[‘8”9F]- 4.7 TFB for gel electrophoresis and isotopic exchange experiments 309 4.4.1.3.3‘8F-Isotope exchange experiments- using the low-activity radiochemical synthesis of[‘8F]-4.7 TFB 310 4.4.1.3.4 Polyacrylamide Gel Electro,horesis Experiments- using the low- activity radiochemical synthesis of [1 F] -4.7 TFB 310 4.4.1.3.5 High-activity radiochemical synthesis of the trifluoroborate of [18F] -4.7 TFB for in vivo imaging 311 4.4.1.3.6 In vivo and post-mortem biodistribution studies 312 4.4.2 RESULTS 313 4.4.2.1 The biotin “conjugate” 4.7 meets fluoridation, purification, and stability requirements for a PET label 313 4.4.2.1 .1 Biotin 4.7 can be converted to the trifluoroborate 4.7 TFB 313 4.4.2.1.2 4.7 TFB is easily purified 314 4.3.2.1.3 Biotin [‘8F]-4.7 TFB is slow to exchange ‘8F. Biotin [‘8F]-4.7 TFB exchanges ‘8F with its environment at a rate of 0.00012 ± 0.00006 min1 (ti,2 = 5800 ± 2000 mm) 315 4.4.2.2 Conjugation to the PET label 4.1 does not alter the desired biological properties of biotin 4.7 TFB 317 4.4.2.3 PET monitored [‘8F]-4.7 TFB biodistribution 318 4.4.2.4 Biodistribution data for[18F]-4.7 TFB 321 4.4.3 DISCUSSION 322 xii 4.4.3.1 Biotin trifluoroborate [‘8Fj-4.7 TFB meets the biological requirements of the trifluoroborate PET labeling strategy 322 4.4.3.2 PET images generated with the biotin trifluoroborate 4.7 TFB validates the use of conjugates of the boron species 4.1 for in vivo PET imaging 323 4.5 IN VIVO IMAGING OF 18F LABELED MARIMASTAT 4.8 TFB 324 4.5.1 MATERIALS AND METHODS 326 4.5.1.1 General synthetic methods 326 4.5.1.2 Specific synthetic methods 326 4.5.1.3 Radiochemical procedures 328 4.5.1.3.1 Radiochemical synthesis of the trifluoroborate of 4.8 for in vivo imaging 328 4.5.1.3.2 Radiochemical synthesis of the trifluoroborate of 4.1 for in vivo imaging 329 4.5.1.3.3 In vivo and post-mortem biodistribution studies 330 4.5.2 RESULTS 331 4.5.2.1 The marimastat “conjugate” 4.8 meets the fluoridation, purification, and stability requirements of a PET label 331 4.5.2.1.1 4.8 can be rapidly converted to the trifluoroborate 4.8 TFB 331 4.5.2.1.2 4.8 TFB is easily purified 333 4.5.2.1.3 4.8 TFB is slow to exchange 18F 334 4.5.2.2 In vivoPET imaging using boron trifluoroborates 335 4.5.2.3 Conjugation to the PET label 4.1 does not alter the desired biological properties of marimastat 335 4.5.2.4 Biodistribution data for[18F]-4.1 TFB or[‘8F]-4.8 TFB 337 4.5.2.5 PET data suggests [‘8F]-4.8 TFB is specific for the tumor 340 4.5.2.6 Time dependent marimastat biodistribution 342 4.5.2.7 On the choice of control 344 4.5.3 DISCUSSION 346 4.5.3.1 Trifluoroborates meet the biological requirements of a PET labeling strategy. 346 4.5.3.2 ‘8F does not dissociate from the probe under in vivoconditions 346 4.5.3.3 The trifluoroborate PET label does not alter the desired biological properties of the biomolecule 347 4.5.3.4 The trifluoroborate PET label clears rapidly from areas in which the tracer does not display its intended activity 348 4.6 CONCLUSIONS 349 4.7 REFERENCES 351 354 xiii Chapter3.354 Determination of the incident photon intensity, Po, that is transferred through a quartz cell using the handheld 254 nm irradiation source 354 Determination of the quantum efficiency of the photolysis reaction, 4 , at 254 nm.357 Determining the constant, and the choice of wavelength for measuring quantum yields 357 ‘H NMR spectrum of 2-mercaptobenzimidazole (MBI) 361 ‘H NMR spectrum ofbenzimidazole (BI) 361 Crystal Data for8-mercapto-9-ethyl adenine, 3.1-I 362 REFERENCE 365 Chapter 4 366 UBC Research Ethics Board’s Certificate of Approval. Research Involving Animals. 366 xiv LIST OF TABLES Table 1.1. Comparison Of kCa(STR s and KMSTR s for substrate analogues of Si. Underscored bold type indicates specific base deviations from unmodified Si. kreiatjve values relative to substrate Si were obtained by dividing the value of k’ obtained for other substrate analogues by the value Of kcat in entry 1 (data courtesy of Jason Thomas) 46 Table 1.2. Comparison of kinetic rate constants for different trans-cleaving ribozymes systems60’62, 63, 67, 71-73 Entries 1 to 3 are DNAzymes, while entries 4 to 8 are ribozymes. Where multiple literature reports exist, rate constants are reported at lowered Mg2 concentrations. (table reproduced with the help of Jason Thomas) 53 Table 1.3. Structures and kinetic constants of sequences shown to restore monophasic kinetic profiles to the 925-1 it system at 13 °C 66 Table 1.4. Rate constants measured at 13 °C for925-llt. Where no “slow” rate is reported, the reaction is monophasic 81 Table 3.1. Thioether containing heterocycles photolyzed in this chapter and their products. Errors are calculated from the sum of the percentage errors of the two largest sources of error; the calculation ofk0b and Po 159 Table 4.1. List of common nuclides used in imaging and their nuclear properties. Mean and Maximum Positron energy reported for the most abundant positron produced. Table compiled from data obtained from Table of Nuclides © Nuclear Data Evaluation Lab. 2000 Korea Atomic Energy Research Institute (http://atom.kaeri.re.kr/ton!nuc9.html). 226 Table 4,2. Aryl borate library assayed for trifluoroborate formation and trifluoroborate 18F-’9 exchange. Compound names are those provided by the borate supplier. R equals benzopinacol 249 Table 4.3. Values for the ‘8”9F TLC Exchange Rate Constants, ‘9F NMR Defluoridation Rate Constants and the substituent constant, a. Values for a are found in Hansch.’62 265 Table 4.4. Biotin [‘8F]-4.7 TFB and [‘8F]-4.7 TFB Biotin-neutravidin biodistribution data in mice collected following 90 mm of PET scanning. Biodistribution data for [‘8F]-4.7 TFB Biotin is shown in Entries 1-6 and biodistribution data for the {‘8F]-4.7 TFB Biotin-Neutravidin complex is shown in Entries 7-12 321 Table 4.5. CONTROL Biodistribution data on a 66c14-luc tumor bearing mouse imaged with [‘8F]-4.1 TFB at 180 mm post injection. 37.5 gCi (counted at 2 pm) was injected at 2:00 pm, PET scanning was done from 2:15 - 3:55 pm and biodistribution data were collected at 4:48 pm (3h post injection.). The number of background counts was 48. 337 xv Table 4.6. Biodistribution data on a 66c14-luc tumor bearing mouse imaged with ‘8F- marimastat,[’8F]-4.8 TFB, collected at 195 mm. 67.5 iCi (counted at 10:55 am) was injected at 11:15 am, PET scanning was done from 11:30 - 2:00 pm and biodistribution data were collected at 2:38 pm (—‘3.25 h post injection.). The number of background counts was 59. The final column “Ratio vs. Control” was calculated by directly dividing marimastat counts by the control counts. This column is relevant because the total scintillation counts in the uterus in both the control and the marimastat samples are serendipitously equivalent 338 xvi LIST OF FIGURES Figure 1.1. Mechanism of RNase A catalyzed RNA Cleavage with His 12, His 119 and Lys 41.10 2 Figure 1.2. The sequence, 925-lie, prepared by templated enzymatic polymerization of synthetic monomer triphosphates and the synthetically prepared sequence, 925-i it, shown hybridized to its substrate Si 15 Figure 1.3. A, left, Time courses for the cleavage of the cis-cleaving sequence 925-lie at different temperatures in CB200 buffer. Radiochemical amounts of 925-iic were incubated in CB200 at (o) 8 °C, (V) 13 °C, (z) 22 °C, and (D) 30 °C (4 °C and 45 °C not shown). Lines shown are mathematical fits of data points to equation 1. R is greater than 0.983; the mean R2 value is 0.989. B, right, Observed cleavage rates, kseij cleave, obtained from single exponential fits were plotted against temperature demonstrating an optimal rate constant, kseificieave, of 0.28 mind at 13 °C. All self- cleavage reactions were performed on the same day using 6 different temperature baths to minimize error and ensure internal consistency (Figure reproduced courtesy of Jay Thomas). Data from one replicate at each temperature was obtained using the same batch of catalyst and substrate 23 Figure 1.4. Trans-cleaving sequences of 925-il prepared by solid phase for a structure-activity assay. All sequences are shown hybridized to their substrate, Si. Emboldened A and Us represent the modified nucleotides shown at right. The base, rC, indicates the target ribophosphodiester bond. Predicted Watson-Crick pairs are indicated by a dash between blue-coloured complementary bases 24 Figure 1.5. Rough optimal temperature determination for the trans-cleaving catalyst 925-i it. A, left, Time course for the cleavage of 925-1 it at different temperatures. Cleavage reactions were performed with 100 nM Catalyst, 15 jiM unlabeled substrate Si, and <[ 5’ 32 labeled substrate in buffer CB200 at (.) 6°C, (ti) 15°C, (LI) 24°C, or (o) 30°C. B, right, slopes obtained from regression analyses were divided by the catalyst concentration (100 nM) and plotted against temperature to give a keat MTR temperature optimum near 24°C. Data from one replicate at each temperature were obtained 27 Figure 1.6. The minimal kinetic scheme for cleavage of substrate Si by 925-1 it at 24 °C in CB200 (50 mM Tris HC1 pH 7.5, 1 mM EDTA, and 200 mM NaC1). E, Catalyst, 5, Substrate Si, P5, 5’ Oligonucleotide product, and P3, 3’ Oligonucleotide product. Kinetic constants are indicated beside their respective steps. The underlined value was calculated from measured constants 30 Figure 1.7. A, left, Plots of the fraction of substrate Si cleaved vs. time at 24°C. Cleavage reactions were performed with varying concentrations of catalyst and < 1 nM 5‘-32P Si in buffer CB200 at 24°C. The catalyst concentrations at which kinetic profiles were xvii observed are: (•) 1200 nM (D) 150 nM, () 80 nM, (v) 40 nM, (0) 20 nM, or (o) 10 nM Catalyst. Data from one replicate at each concentration of catalyst were obtained. All catalyst concentrations assayed not shown. B, right, Plot of kØb vs. catalyst concentration. Rate constants obtained from the single exponential fits in A were fit to a hyperbolic plot against catalyst concentration equation 4. A non-linear least squares fit gave a maximum first—order rate constant (kcat STR) of 0.037 ± 0.001 min1 and a catalyst concentration at which the reaction rate is half-maximal (Km STR) of 69 ± 7 nM. R2 is 0.978 39 Figure 1.8. Pulse-chase experiment for k.4. Catalyst is “pulsed” with 32P-labeled substrate and given enough time to form a quantifiable quantity of catalyst-label complex. After a time, t1, the chase is added which competes for the 5’-32P labeled Si. The chase is applied before a significant amount of E*S is cleaved (3 mm). Due to the excess of the unlabeled chase over both the labeled substrate and total catalyst content, all free catalyst is complexed in a form that cannot associate with 5’-32P labeled Sl. Further formation of catalyst-S‘-32P labeled Si complex is essentially halted, allowing one to quantitate kinetic constants of 32P-[ES] decomposition without having to deal with 32P-[ES] formation. Bold arrow indicates removal of free E following the chase 40 Figure 1.9. Determination of the Substrate Dissociation Rate Constant for Substrate S 1 from Pulse-chase Experiments at 24°C. Saturating amounts of Catalyst (Control reactions, (•) 1800 nM and (•) 1200 nM) were incubated with <1 nM S’-32P Si and run as a control for pulse-chase experiments. Three replicates of pulse-chase reactions (, :J, and 0) were performed with 1000 nM Catalyst, <1 nM 5’ 3P labeled Sl and 75 iM of excess unlabeled Si as the chase. Fraction cleaved values for chased reactions have been corrected for substrate cleavage that occurred prior to the chase. Zero time for the chased reactions were referenced to the time at which the chase was added. Results of the fit for the chased reaction: kehase = 0.41 ± 0.05 min1,Amplitude = 4.0 ± 0.1 %, R is 0.849 41 Figure 1.10. Pulse-chase analysis by native gel electrophoresis to determine the first order rate constant for substrate analogue dissociation from925-lit at 24°C. A trace of 5’-32P Si-OMe substrate analogue was incubated with catalyst (1 tiM) in CB200. Following the addition of an equal volume of the chase (100 jiM substrate in CB200), aliquots were taken at various times and loaded onto native PAGE gels. The fraction of S’-32P Si -OMe dissociated was plotted versus time for three experimental replicates (o, cJ, and 0), A non-linear least squares fit gave a first order dissociation rate constant, k1 2. SI-OMe, for the substrate analogue Si-OMe of 0.44 ± 0.03 mm R is 0.979 (Figure reproduced courtesy of Jay Thomas) 42 Figure 1.11. A, left, Multiple turnover analyses on the cleavage of Si at 24 °C. Cleavage reactions were performed with varying concentrations of catalyst, 15 tM Si, and <1 nM 5’ labeled Si in buffer CB200 at 24°C. The concentrations of catalyst at which kinetic experiments were observed are: (•) 1000 nM (D) 500 nM, (Li) 200 nM, (V) xviii 100 nM, or (0) 50 nM Catalyst. Data from one replicate at each concentration of catalyst were obtained. Data were fit to linear equations. The lowest R2 value was 0.996. Cleavage was allowed to proceed until a maximum of 20 % of the substrate was cleaved. Reactions were quenched in formamide at the times indicated on the graph. B, right, Rate constant determination for the maximum steady state rate constant. Values ofk0b5 MTR was plot against catalyst concentration to obtain a steady state rate constant at saturating substrate concentrations. The intercept for this value is 0.000 ± 0.001 jiM!min. The slope, kcatjR, is 0.030 ± 0.002 min’ and R2 is 0.981. 43 Figure 1.12. Non-denaturing PAGE analysis of the binding affinities of non hydrolysable substrate analogues of Si to 925-1 it. Data were fit to equation 7. A, left, 32P labeled Si -OMe bound to catalyst vs. catalyst concentration. Data from two replicates were obtained at each concentration of catalyst. B, right, Plot of the fraction 32P labeled Si - DNA bound to catalyst vs. catalyst concentration. The value ofKd for Si-OMe was 82 ± 10 nM and had an R2 values of 0.967. The value of Kd for Si-DNA was 43 ± 9 nM and had an R2 value of 0.970. Data from one replicate at each concentration of catalyst were obtained (Figure reproduced courtesy of Jason Thomas) 45 Figure 1.13. Multiple turnover analysis on the cleavage of Si at 24 °C in the presence and absence of 2 mM MgSO4. Cleavage reactions were performed with 150 nM catalyst, 15 tM Si, and <1 nM 5’ labeled Si in 50 mM Tris HC1 pH 7.5 buffer, 200 mlvi NaCl, and either (.) 1 m]V1 EDTA or (o) 2 mM MgSO4 at 24°C. Cleavage was allowed to proceed until a maximum of 20 % of the substrate was cleaved. Reactions were quenched in formamide at the times indicated on the graph. Linear regression analyses gave Si cleavage rates (k0b5M) of (.) 0.0048 ± 0.0001 pM’ mu1 and (0) 0.00554 ± 0.00003 tM miii’. Applying the observed rates to equation 3, the steady state cleavage rate (kc0tjI,J’R) of 925-1 it is 0.032 ± 0.007 miii’ in the presence of 2 mM EDTA (.) and 0.03 7 ± 0.002 miii’ in the presence of 2 mlvi MgSO4(0) 48 Figure 1.14. The minimal kinetic scheme for cleavage of substrate Si by 925-lit at 24 °C in CB200 (50 mM Tris HC1 pH 7.5, 1 mM EDTA, and 200 mM NaCl). E, Catalyst, 5, Substrate Si, P5, 5’ Oligonucleotide Product, and P3’, 3’ Oligonucleotide product. Corroborative data are shown for various experiments with the non cleavable 2’ ribo methyl ester of Si and the DNA equivalent of Si 49 Figure 1.15. Comparison of identical data, fit to single and double exponential equations. Both cleavage reactions were performed with 900 nM catalyst and < i nM substrate S 1 in buffer CB200 under single turnover cleavage reactions at 13 °C. A, top, Fitting data to a single exponential 1 gives a value of 0.020 ± 0.005 for k0b5 and 0.704 for R. B, bottom, Identical data fit to double exponential equation 2 gave an observed rate constant for the fast cleavage phase, k0b5 fast STR, of 0.21 ± 0.02 miii1 (fast phase amplitude is 25 ± 0.7 %) and an observed rate constant for the slow cleavage phase, k0b5 slow STR, of 0.0026 ± 0.0002 mind (slow phase amplitude is 53 ± 1 %). R2 was 0.996. (graph at right details the early time points enclosed in the coloured box) 63 xix Figure 1.16. Effect of 925-1 it concentration on the fast and slow phase rate constants and phase amplitudes in single turnover experiments at 13 °C in CB200. Observed rate constants at different catalyst concentrations, k0b5 ft STR’ were obtained by fitting substrate cleavage data to double exponential equation 8. The values k0b5fast sm and k0b5 slow s at various 925-lit concentrations were fit separately to equation 4. Data from one replicate at each catalyst concentration were obtained. A, top left: Effect of 925-lit concentration on k0b5 ft STR: the maximum first-order fast cleavage rate constant, kcat fast STR is 0.20 ± 0.01 min’ and the catalyst concentration at which the reaction rate is half-maximum for the fast phase, KM fast STR, is 170 ± 30 nM (R2 is 0.952). B top right: Effect of 925-1 it concentration on the fast phase amplitude. The fraction of substrate cleaved through the large cleavage rate constant (not normalized for the extent of cleavage), [Pfast]a,, obtained from fits to equation 2 are shown as a function of catalyst concentration. The average value of [Pfast]oo is 28 %. C bottom left: Effect of925-i it concentration onk0b55lOWSTh: kcat slow STR is 0.0036 ± 0.0001 min1 and KM STR is 21 ± 4 nM (R2 is 0.892). D bottom right: Effect of 925-1 it concentration on the slow phase amplitude. The average value of [Pi0], is 51 %. The relative invariance of [Pft]c, and [P0]as a function of catalyst concentration suggests that these fractions are independent of the catalyst concentration. Catalyst concentrations ranged from 5 nM to 1500 nM 64 Figure 1.17. Pulse-chase dissociation constant experiments that can give rise to monophasic or biphasic kinetic profiles depending on the nature of the slow phase. After the substrate is added to the catalyst in the initial pulse, a time, ti, is given for complex formation before chase is added. A, right, The kinetic system that would give monophasic pulse-chase kinetic profiles but biphasic single turnover kinetic profiles. When this system is chased with Si, turnover of 32P-S 1 is observed by the active complex but a biphase is not observed because the inactive complex cannot cleave 32P-S 1. B, left, The kinetic chase system observed with 925-lit. Biphasic single turnover kinetic profiles are a result of two active complexes both capable of performing turnover but at different rates. When this system is chased, a biphasic plot is obtained because there are two distinct catalyst-substrate complexes which turnover substrate at different rates. Bold arrow indicates removal of free E following the chase 69 Figure 1.18. Determination of substrate dissociation rate constants for substrate Si by pulse- chase cleavage assay with 925-i it in CB200 13°C. A left: Control reaction, (.) 900 nM Catalyst was incubated with <1 nM Si, or Chase reaction, (0) 112 nM catalyst and <1 nM Si were challenged with 50 .tM of excess unlabeled Si at 3 mm. Data for the unchased reaction were fit to equation 8 giving 0.21 min1 for kOb5 fast STR and 0.0026 min1 for kOb5 slow STR (R2 is 0.996). Data for chased reaction were fit to equation 8 giving 0.233 min1 for kchase fast and 0.0041 min’ for keijase slow (R2 is 0.981). These values were treated according to equation 5 (see section 1.3) to give the substrate dissociation rate constants, k1 ft and 1(j slow B right: rescaling the fraction cleaved axis for the chased reaction clearly shows biphasic behavior. Dotted lines indicate the constituent curves for the fast and slow phases 70 xx Figure 1019. Pulse-chase analysis by native gel electrophoresis to determine the first order rate constant for substrate analogue dissociation from 925-i it at 13°C. A trace of 5’-32P labeled Si-OMe was incubated with 925-i it (i.tM) in CB200. Following the addition of an equal volume of the chase (100jiM substrate in CB200), aliquots were taken at various times and loaded onto native PAGE gels. The fraction of substrate analogue dissociated was plotted versus time for multiple experimental trials. The fraction of Si -OMe dissociated was plotted versus time. Data from at least two experimental replicates were obtained. The data were fit to a double exponential equation 5, which returned the following dissociation rate constants and phase amplitudes: k.qft Si-OMe = 0.014 ± 0.004 min1,k110 S1-OMe= 0.0023 ± 0.0005 min1, amplitude fast = 27 ± 7 %, amplitude slow = 46 ± 6 %. Dotted lines indicate the constituent curves for the fast and slow phases. (Figure reproduced courtesy of Jay Thomas) 71 Figure 1.20. A, left, Multiple Turnover Analyses on the cleavage of substrate Si at 13 °C. Cleavage reactions were performed with varying concentrations of catalyst, 7.5 1iM unlabeled substrate Si, and <1 nM 5’ labeled substrate 51 in buffer CB200 at 13 °C. (•) 1000 nM (D) 500 nM, (z) 200 nM, (V) 100 nM, (0) 50 nM Catalyst. Data were fit to equation 12 (f=C( 1 expSt t)+ksteady state t). The average R2 is greater than 0.996 except for data obtained at 50 nM catalyst which is 0.968. Cleavage was allowed to proceed until a maximum of 20 % of the substrate was cleaved. Reactions were quenched in formamide at the times indicated on the graph. B, right, Rate constant determination for the steady state rate constant. For all catalyst concentrations, linear regression analyses were performed on multiple turnover substrate cleavage for time points after 255 mi This rate was plot against catalyst concentration to obtain a steady state rate constant. Data from one replicate at each catalyst concentration were obtained. The intercept for this value is 0.0000 ± 0.0001 jiM min1.The steady state rate constant is 0.003 8 ± 0.0002 min1.R2 is 0.994 72 Figure 1.21. Multiple turnover cleavage of substrate Si by925-lit at 13 °C under native PAGE conditions. A top: native PAGE analysis of the time course for multiple turnover at 1 jiM catalyst and 15 jiM Sl in CB200 at 13°C (lanes 1-10). Control lanes (11-14) reveal electrophoretic motilities of: uncleaved Si (lane ii), E•S 1 complex (lane 12), P5’ (lane 13), and E•P5 complex (lane 14). The reaction lanes show a small, steady state amount of E•Si, and no observable E•P5. B middle: The same time course (lanes 1-10) for the same reaction as in A run on a denaturing PAGE gel. C bottom: Plots of substrate cleaved vs. time for both the native (blue trace) and denaturing gels (green trace). Production of unbound product as observed in the native gel matches that observed by denaturing gel, indicating that the burst phase is not due to slow release of the 5’ product. (Figure reproduced courtesy of Jay Thomas) 76 Figure 1.22. The effect of temperature cycling on the multiple turnover rate of Si substrate cleavage by 925-1 it at 13 °C. 100 nM catalyst, 15 jiM unlabeled Si, and <1 nM 5’ labeled Si was incubated in buffer CB200 at 13 °C. The reaction was placed into a PCR temperature cycler programmed as follows: 13 °C for 15 s, 22 °C for 7 mm, and 95 °C for 15 sec. With temperature ramping time included, the time for one cycle was ‘-9 mi Aliquots were removed for analysis after every five cycles. Data were fit to a xxi linear equation: k0b the,,,,oc!eMTR is 0.58 ± 0.02 cycl&’ or 0.064 ± 0.002 min’ for the 9 mm cycle (R2 is 0.996). Data from one replicate at every five cycles were obtained. The linearity of the graph indicates that the catalyst is not deactivated upon heating. A control reaction done with only labeled substrate showed a negligible amount of cleavage during the duration of temperature program 79 Figure 1.23. The minimal kinetic scheme for cleavage of substrate Si by925-lit at 13 °C in CB200 (50 mM Tris HC1 pH 7.5, 1 mM EDTA, and 200 mlvi NaC1). E, Catalyst 925- lit, 5, Substrate Si 83 Figure 2.1. Early photocaged molecules. A, top, z-chymotrypsin deactivated in its cis cinnamate ester form is activated upon irradiation through isomerization of the cinnamate to its trans$orm which is 1 0 times more susceptible to hydrolysis.89 Subsequent hydrolysis releases a-chymotrypsin which catalyzes the hydrolysis of tyrosine ester to tyrosine which is further converted to the detectable pigment melanine by tyrosinase. B, bottom, the light activated ATP from which the term photocaging was coined.90 Caged nitrobenzyl ATP is irradiated with light to release nitrosoaldehyde and ATP 95 Figure 2.2. Examples of reversibly photocaged molecules applied to biologically relevant systems. A, top, an example of azobenzene caged DNAyzme activity.97 B, Bottom, spiropyran caged a-chymotrypsin activity.96 Different frequencies of light favor different isomers of the double bond 96 Figure 2.3. Common reagents used in biochemical photochemistry and their photoproducts.88’ 99-101 98 Figure 2.4. The photolysis mechanism of a caged nitrobenzyl complex shown with aci-nitro (A), isoxazole (B), and nitrosobenzyl hemiacetal (C) intermediates 99 Figure 2.5. Photocaged ribose and nucleotide base photocagin strategies as described in MacMillan, Silverman Heckel, and Dmochowski.91’1081 102 Figure 2.6. A., Structure of the histaminyl modified deoxyriboadenosine NH-dA” required for the activity of DNAzyme 9-25-il c. B., Structure of the imidazole ethyl thioether modified deoxyriboadenosine isostere S-dA 104 Figure 2.7. C8-linked 2-(4-imidazolyl)ethylthio)adenosine photolyzes to unmodified adenosine and an imidazole product 105 Figure 2.8. The photolysis of triphosphate 2.5 as monitored by absorbance spectroscopy. Blue line represents the absorbance spectrum of a 6 jiM solution of 2.5 in water prior to irradiation. Red line represents the absorbance spectrum of a 6 jiM solution of 2.5 in water following ih of irradiation at 254 nm 122 Figure 2.9. The photolysis of 40 jiM of ribonucleoside 2.6 with a 280 nm irradiation source as monitored by absorbance spectroscopy 123 xxii Figure 2.10. The photolysis of nucleoside 2.6 as monitored by ‘H-NMR spectroscopy. A. top, Reference spectrum of adenosine in DMSO-d6 obtained from the spectral database for organic compounds (SDBS) (http://www.aist.go.jp/RIODB/SDBS/cgi binlcre_index.cgi). Peaks present at 7.4, 5.5 and 5.3 ppm correspond to exchangeable hydroxyl protons signals that are not be observed in MeOH-d4. B, middle,1H-NMR spectrum of 2.6 at -5 mM acquired in MeOH-d4 prior to irradiation. Protons corresponding to the 2-(4-imidazolyl) ethyl moiety are shown in red. C, bottom, ‘H NMR of spectrum B following 14 hours of 254 nm irradiation in undeuterated methanol. ‘H-NMR spectrum acquired in MeOH-d4. The newly incorporated proton corresponding to the C-8 position of adenosine is indicated by the blue arrow 124 Figure 2.11. The photolysis of nucleoside 2.6 monitored by ‘H-NMR spectroscopy in methanol. A, top, ‘H-NMR spectrum following the photolysis of 5 mM of 2.6 in undeuterated methanol. B, bottom, ‘H-NMR spectrum following the photolysis of 5 mM of 2.6 in deuterated methanol. The blue arrow indicates the missing ‘H-NMR shift that is indicative of the incorporation of deuterium at the C-8 of ademine 126 Figure 2.12. UV mediated destruction of the DNA sequence 5’-TCCCCCCTTTTCTTTTAAG- 3’. A, left, The unirradiated DNA sequence 5 ‘-TCCCCCCTTTTCTITTAAG-3’, where the two A’s indicate positions where 2.8 has been incorporated. DNA B, middle, The result of 1 hour of irradiation from a 1 mW, 254 nm shortwave mercury handheld UV lamp on A. C, right, the sequence, 5 ‘-TCCCCCCTTTTCTTTT-3’, that was used as a sizing control to gauge the photolysis of lane B 129 Figure 2.13. The catalytic motif of Joyce’s Mg2 dependent DNAzyme 8-17 (left) and Lu’s Zinc dependent 17E motif (right). We chose to photocage four derivatives of 17E with single substitutions of 2,8 at the adenines shown by the coloured arrows. Two of these adenines are conserved between the Joyce and Lu DNAzymes (ml and m2). Predicted Watson-Crick pairs are indicated by a dash between complementary bases. 1 7E is shown paired to its all RNA substrate, 1 7RS 133 Figure 2.14. 17RS dark cleavage by 17E variants in 100 j.tM ZnBr2, 50 mM pH 7.0 HepesHCl buffer, and <1 nM 5’ 3P of labeled 17RS RNA by variants of 17E at 5 jiM. A, top, Denaturing PAGE gel depicting cleavage efficiencies of modified catalysts in HepesHCl buffered kinetic experiments. B, bottom left, Time dependence of RNA cleavage for 17E and its derivatives: 17E-ml (D), 17E-m2 (A), 17E-m3 (V), 17E-m4 (o). C. bottom right, Table summarizing the first order rate constants, errors, and activity relative to 1 7E for 1 7E and its derivatives 138 Figure 2.15. RNA 17RS cleavage by 17E, caged 17E-mland photoactivated l7E-ml. Gel data for single turnover RNA cleavage. Cleavage reactions were performed with 100 mM ZnBr2, 50 mM pH 7.0 TrisHC1 buffer, and <1 nM 5’ 32P labeled RNA. Catalysts were resuspended with 1.33 mL of water in quartz cuvettes at 300 nM, flushed with argon for 30 mm, irradiated, dried and incubated with the substrate cocktail to give a final catalyst concentration of 5 jiM. A, top, Irradiation at ‘-.1 mW with a 254 nm shortwave mercury handheld UV lamp Left to right: Time course of cleavage for; I) 1 7E irradiated for 10 mm, II) Unirradiated 1 7E-ml, and III) 1 7E-m 1 irradiated for 10 xxiii mm. B, bottom, frradiation at 10 mW with a 310 nm Band Pass Filtered Arc Lamp. From left to right: Time course of cleavage for; IV) 1 7E irradiated for 15 mm, V) Unirradiated 17E-ml, and VI) 17E-ml irradiated for 15 mm 141 Figure 2.16. Plots of fraction cleaved vs. time for irradiated samples of 1 7E and 1 7E-ml. The fraction of 1 7RS cleaved was plot against time to give data that were fit to equation 1. The constants and errors returned from fitting these data to the equation [Pit = [Pj(1- el0l)S t) using Sigma Plot 2001 v7. 101 are detailed in the inset within each graph. Condition IV and VI refer to conditions described in Figure 2.15 142 Figure 3.1. Reagents and products of the photolysis of thiol containing compounds as investigated by A, top, Jenks’25 and B, bottom, Grimrne’26 (yields in brackets) 153 Figure 3.2. Reagents and products generated in high yield from the photolysis of thiol containing compounds as investigated by A, top, Kutateladze 23 and B, bottom, Kropp127 (yields in brackets) 154 Figure 3.3. Kropp’s’27 characterization of the radical/ionic nature of photolytic C-S bond cleavage. A, top, Proposed homolytic (top) or heterolytic (bottom) mechanisms of C S bond photolyses. B, bottom, Predicted mechanistic products as applied to the photolysis of 2-S-phenylthioether tetrahydropyran. Homolytic photolysis would give rise to tetrahydropyran while heterolytic photolysis would give rise to 2-methoxy tetrahydropyran 155 Figure 3.4. Spectrophotometric absorbance profile for photolysis at 254 nm of a 26.5 1iM solution of 3.1 in H20. Because the photolysis for the conversion of 3.1 to 3.1-I is much faster than the photolysis of 3.1-Ito 3.1 -P, isosbestic points could be identified within the spectra. A. left, data to 7 mm detailing the conversion of 3.1 (green) to 3.1-I (blue). Plots are shown at 0.5 to 1 mm intervals to highlight isosbestic points indicated by the blue arrows. B, right, data from 7 mm onwards detailing the conversion of 3.1-I (blue) to 3.1-P (red). Red arrows indicate isosbestic points corresponding to the transformation of 3.1-I to 3.1-P. Plots are shown at 1 to 5 mill intervals to highlight isosbestic points indicated by the red arrows 167 Figure 3.5. Intermediate crystallized from the irradiation of 3.1 in MeOH. A) top. ORTEP structure obtained from x-ray diffraction studies. B) bottom. UV/Vis absorption spectra of the crystallized intermediate 3.1-I (Data courtesy of Jen Steele) 169 Figure 3.6. NMR spectral visualization of the photolysis of 0.04 M 3.1 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 4.2 mg of 3.1 in 500 iL MeOH-d4. Arrows in the t 1 h spectrum indicate NMR signals corresponding to the formation of 3.1-I, arrows in the t= 24 h spectrum indicate NMR signals corresponding to the formation of 3.1-P, and the green arrow in the top spectrum indicates the shift of an ethene standard in MeOH-d4. Times at which spectra were acquired are indicated to the left of spectra. Ethene had a shift of 5.39 ppm, this value matched the respective peak at 1 hour and 24 hours. No ethene was observed in the 192 hour spectrum. The lack of signal was attributed to volatilization of ethene that had occurred due to the lack of an air-tight seal on the NMR tube 170 xxiv Figure 3,7. Proposed general pathway for the photolysis of adenine 3.1. The first step in the photolysis of 3.1 involves photolytic cleavage of the thioether resulting in ethene and 8-mercapto-9-ethyl adenine, while the second step involves the replacement of a heterocyclic C-S bond with a C-H bond to give 9-ethyl adenine 172 Figure 3.8. Spectrophotometric absorbance profile for the photolysis at 254 nm of a 20 jiM solution of 3.2 in MeOH. Because the rate of photolysis for the conversion of 3.2 to thiobenzimizole is much larger than the transition from 2- mercaptobenzimizole to benzimidazole, isosbestic points could be identified within the spectra. A. right, data to 2.5 mm detailing the conversion of 3.2 (green) to the intermediate 2- mercaptobenzimidazole (blue). Plots are shown at < 1 mm intervals to highlight isosbestic points indicated by the blue arrows. B, left, data from 2.5 mm onwards detailing the conversion of 2-mercaptobenzimidazole (MBI, blue) to benzimidazole (BI, red). Red arrows indicate isosbestic points corresponding to the transformation of 2-mercaptobenzimidazole to benzimidazole. Plots are shown at 1 to 5 mm intervals to highlight isosbestic points indicated by the red arrows 177 Figure 3.9. NMR spectral visualization of the photolysis of 3.2 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 3.2. Times at which spectra were acquired are indicated to the left of spectra. A predicted spectra for that of ethene is shown in the uppermost spectrum. Arrows in the t= 90 h spectrum indicate NMR signals corresponding to the formation of 2-mercaptobenzimidazole, arrows in the t= 19. h spectrum indicate NMR signals corresponding to the formation of benzimidazole, and the green arrow in the top spectrum indicates the shift of an ethene standard in MeOH-d4. See appendix for 2-mercaptobenzimidazole and benzimidazole reference spectra. Ethene is observed at 90h of irradiation of 3.2 due to the use of an air tight seal on the NMR tube 178 Figure 3.10. Mechanism adapted from Kropp’27 as applied to the photolysis of 3.1 181 Figure 3.11. Norrish type II fragmentation mechanism for the photochemical hydrogen abstraction of 4-alkylpyrimidines’32 182 Figure 3.12. A. Top. Norrish type II like mechanism proposed for the photolysis of 1-ethyl-2- thioethyl benzimidazole thioether. B. Bottom 3.3 with a y-ethyl thioether proton tied up in a 5- membered ring such that the abstracing nitrogen has limited access to its abstraction 183 Figure 3.13. Spectrophotometric absorbance profile for photolysis at 254 nm of a 34.1 jiM solution of 3.3 in MeOH. A. right, data to 3.5 mm detailing the conversion of 3.3 (green) to 3.3-I (blue). Plots are shown at < 1 mm intervals to highlight isosbestic points indicated by the blue arrows. B, left, data from 3.5 mm onwards detailing the conversion of 3.3-I (blue) to 3.3-P (red). Red arrows indicate isosbestic points corresponding to the transformation of 3.3-I to 3.3-P. Plots are shown at 1 to 5 mm intervals to highlight isosbestic points indicated by the red arrows 187 Figure 3.14. NMR spectroscopy on the photolysis of 3.3 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 8 mg of 3.3. Arrows in t = 2 h xxv spectrum indicate NMR signals corresponding to the formation of the thiol alkene 3.3- I and arrows in t = 8 h spectrum indicate NMR signals corresponding to the formation of the sulfur deficient species, 3.3-P. Times at which spectra were acquired are indicated to the left of spectra. The rate of photolysis for the conversion of 3.3 to 3.3- I is much larger than the rate of photolysis from 3.3-I to 3.3-P 188 Figure 3.15. Spectrophotometric absorbance profile for photolysis at 254 nm of a 20 jiM solution of 3.4 in MeOH. A. right, data to 2 mm detailing the conversion of 3.4 (green) to 2-mercaptobenzimidazole (blue). Black lines are drawn at < 1 mill intervals to highlight isosbestic points indicated by the blue arrows. B, left, data from 2 mm onwards detailing the conversion of 2-mercaptobenzimidazole (MBI, blue) to benzimidazole (BI, red). Red arrows indicate isosbestic points corresponding to the transformation of 2-mercaptobenzimidazole to benzimidazole. Black lines are drawn at 1 to 5 mm intervals to highlight isosbestic points indicated by the red arrows 196 Figure 3.16. ‘H NMR spectroscopy visualization of the photolysis of 3.4 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 3.4. Times at which spectra were acquired are indicated to the left of spectra. A predicted spectra for that of 1-butene by Chemdraw Ultra 9.0 ‘H-NMR prediction software is shown in the uppermost spectrum (green arrows) 197 Figure 3.17. Spectrophotometric absorbance profile for photolysis at 254 nm of a 20 jiM solution of 3.5 in MeOH. A. right, data to 2 mm detailing the conversion of 3.5 (green) to 2-mercaptobenzimidazole (blue). Black lines are drawn at < 1 mm intervals to highlight isosbestic points indicated by the blue arrows. B, left, data from 2 mm onwards detailing the conversion of 2-mercaptobenzimidazole (MBI, blue) to benzimidazole (BI, Red). Red arrows indicate isosbestic points corresponding to the transformation of 2-mercaptobenzimidazole to benzimidazole. Black lines are drawn at 1 to 5 mm intervals to highlight isosbestic points indicated by the red arrows 198 Figure 3.18. ‘H NMR spectroscopy of the photolysis of 3.5 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 3.5. Times at which spectra were acquired are indicated to the left of spectra. A predicted spectra for that of 1 -butene, as well as spectra for NEAT samples of E, and Z butane obtained from spectral database for organic compounds are shown in the uppermost spectra 199 Figure 3.19. Spectrophotometric absorbance profile for the photolysis at 254 nm of a 20 jiM solution of 3.6 in MeOH. A. right, data to 6 mm detailing the conversion of 3.6 (green) to 2-mercaptobenzimidazole (blue). Plots are shown at < 1 mm intervals to highlight isosbestic points indicated by the blue arrows. B, left, data from 6 mm onwards detailing the conversion of 2-mercaptobenzimidazole (MBI, blue) to benzimidazole (BI, Red). Plots are shown at 1 to 5 mm intervals 200 Figure 3.20. 1H NMR spectroscopy on the photolysis of 3.6 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 3.6. Times at which spectra were acquired are indicated to the left of spectra. A spectrum of cyclohexenone obtained in MeOH-d4 is shown in the uppermost spectrum (green arrows) 201 xxvi Figure 3.21. Plots of Absorbance at 310 nm vs. time for the photolysis of a 26.5 tM solution of 3.1 in water. Data to 7 mm (red trace) were fit to equation (15), AbSt0tai =yO+a*(1 - exp(-k0bt)). The constant, yO is 0.002 ± 0.004, a is 0.3 18 ± 0.004, k0b is 0.54 ± 0.02 and R2 is 0.998. Data from 7 mm onwards (blue trace) were fit to equation (15) in the form, AbStotai =yO+a*exp(k t). The constant, yO is 0.026 ± 0.002, a is 0.31 ± 0.002, k0b is 0.0 147 ± 0.0003 and R2 is 0.999 209 Figure 3.22. Plots of absorbance at 307 nm vs. time for the photolysis of a 20 jiM solution of 3.2 in MeOH. Data to 2.5 mm (red trace) were fit to equation (15), AbStotai =yO+a*(1 exp(-k0bt)). The constant, yO is 0.003 ± 0.002, a is 0.121 ± 0.003, k0b is 2.4 ± 0.1 and R2 is 0.998. Data from 2.5 mm onwards (blue trace) were fit to equation (15), AbStotai =y0+a*exp(kObS t). The constant, yO is 0.0 10 ± 0.0045, a is 0.23 ± 0.02, k0b is 0.31±0.03,andRis0.996 211 Figure 3.23. Plots of absorbance vs. time for the photolysis of a 34.1 jiM solution of 3.3 in MeOH. A. Left. Absorbance at 310 nm vs. time. Data to 3.5 mm (red trace) were fit to equation (15). The constant, yO is 0.003 ± 0.007, a is 0.44 ± 0.01, k0b is 0.72 ± 0.05 and R2 is 0.998. Data from 3.5 mm onwards (blue trace) were fit to equation (14). The constant, yO is 0.03 ± 0.01, a is 0.61 ± 0.02, kob8 is 0.11 ± 0.01, and R2 is 0.990 212 Figure 3.24. Plots of absorbance at 310 nm vs. time for the photolysis of a 20 jiM solution of 3.4 in MeOH. Data to 2 mm (red trace) were fit to equation (15), AbStotal =y0+a*(1 exp(-k0bt)). The constant, yO is 0.0035 ± 0.0009, a is 0.082 ± 0.001, k0b3 is 2.6 ± 0.1 and R2 is 0.999. Data from 2 miii onwards (blue trace) were fit to equation (15), AbStotai =yO+a*exp(k t). The constant, yO is 0.026 ± 0.003, a is 0.14 ± 0.02, k0b3 S 0.43 ± 0.07 and R2 is 0.970 213 Figure 3.25. Plots of absorbance vs. time for the photolysis of a 20 jiM solution of 3.5 in MeOH. A. Left. Absorbance at 310 nm vs. time. Data to 2 mm (red trace) were fit to equation (15), AbStotal =y0+a*(1exp(kObS t)). The constant, yO is 0.009 ± 0.001, a is 0.209 ± 0.001, k0b is 2.94 ± 0.04 and R2 is 0.999. Data from 2 mm onwards (blue trace) were fit to equation (15), AbStøtai =yO+a*exp(k03t). The constant, yO is 0.012 ± 0.004, a is 0.32 1 ± 0.005, and k0b is 0.257 ± 0.01 and R2 is 0.994 214 Figure 3.26. Plots of absorbance vs. time for the photolysis of a 20 jiM solution of 3.6 in MeOH. A. Left. Absorbance at 307 nm vs. time. Data to 6 mm (red trace) were fit to equation (15), AbStatai =y0+a*(1exp(kObS t)). The constant, yO is -0.001 ± 0.002, a is 0.0480 ± 0.002, k0b is 0.75 ± 0.09 and R2 is 0.995. Data from 6 mm to 20 mm (blue trace) were fit to equation (15), AbStotal =yO+a*exp(k0t). The constant, yO is 0.027 ± 0.001, a is 0.17 ± 0.1, k0 is 0.32 ± 0.1, and R2 is 0.960 215 Figure 3.27. Proposed general mechanism for photolytic adenine thioether photolysis 217 Figure 4.1. Recent schemes used for generating PET labeled tracers. A. top, method for oligonucleotide labeling,’50 B. middle, Scheme for antibody labeling,’5’C. bottom, scheme for peptide labeling.’52 232 xxvii Figure 4.2. Our proposed one-step scheme for generating PET labeled biomolecules 235 Figure 4.3. The chemical structure of a stable boron containing precursor 4.1, that meets the radiochemical requirements of a PET strategy when converted to its ‘8F labeled trifluoroborate, 4.1 TFB 236 Figure 4.4. Proposed mechanism for aqueous aryl trifluoroborate formation and dissociation. R, R’ equals different aryl substituents 239 Figure 4.5. Mechanism of cumyl chloride solvolysis.’6° R, R’ equals different aryl substituents 240 Figure 4.6. ‘H NMR and ‘9F NMR spectra of 4-carboxyphenyl trifluoroborate in DMSO-d6.255 Figure 4.7. Kinetic profile of 4-carboxyphenyltrifluoroborate dissociation in aqueous solutions. A, top left, 4-carboxyphenyltrifluoroborate dissociation as monitored by ‘9F NMR sDectroscopy. B, bottom left, plot of trifluoroborate fraction vs. time as measured by ‘F NMR. Data were fit to the function y a(e0bt), k0b = 0.082 ± 0.005, a = 0.164 ± 0.006, R2 = 0.994, and t112= 8.4 ± 0.6 mm. C, top right, 4- carboxyphenyltrifluoroborate ‘8F-’9 exchange as monitored by TLC. D, bottom right, plot of‘8F-trifluoroborate fraction vs. time as measured by TLC. Data were fit to the function y= a(exp0bst) k0b = 0.054 ± 0.004, a = 0.51 ± 0.03, R2 = 0.991, and t112=12±lmin 256 Figure 4.8. ‘H NMR and ‘9F NMR spectra of 4-sulphonamidylphenyl trifluoroborate in DMSO d6 258 Figure 4.9. Kinetic profile of 4-sulphonamidylphenyl trifluoroborate dissociation in aqueous solutions. A, top left, 4-sulphonamidylphenyl trifluoroborate dissociation as monitored by ‘9F NIVIR spectroscopy. B, bottom left, plot of trifluoroborate fraction vs. time as measured by ‘9F NMR spectroscopy. Data were fit to the function y= a(exp0t) k0b = 0.03 1 ± 0.002, a = 0.97 ± 0.01, R2 = 0.998, and t,= 22 ± 2 mm. C, top right, 4-sulphonamidylphenyl trifluoroborate ‘8F-’9 exchange as monitored by TLC. D, bottom right, plot of‘8F-trifluoroborate fraction vs. time as measured by TLC. Data were fit to the function y= a(exp0t) k0b = 0.0170 ± 0.0015, a = 0.67 ± 0.05, R2 = 0.994, and t,,2= 40±4 mm 259 Figure 4.10. ‘H NMR and ‘9F NIvIR spectra of 2,4,6-trifluorophenyl trifluoroborate in DMSO d6 261 Figure 4.11. Kinetic profile of 2,4,6-trifluorophenyl trifluoroborate dissociation in aqueous solutions. A, top left, 2,4,6-trifluorophenyl trifluoroborate dissociation as monitored by ‘9F NMR spectroscopy. B, Plot of trifluoroborate fraction vs. time as measured by ‘9F NMR spectroscopy. Data were fit to the function y= a(exp0t), k0b = 0.0026 ± 0.0003, a 0.93 ± 0.03, R2 = 0.985, and ty2= 266 ± 9 mm. C, 2,4,6-Trifluorophenyl trifluoroborate‘8F-’9 exchange as monitored by TLC. D, Plot of‘8F-trifluoroborate fraction vs. time as measured by TLC. Data were fit to the function y a(exp0t), k0b = 0.0029 ± 0.0005, a = 0.73 ± 0.03, R2 = 0.941, and 234 ± 40 mm 262 xxviii Figure 4.12. Hammet Plot in the form log(k) ap + log(ko) for data in Table 4.3. Data for the ‘8”9F Exchange TLC experiment (.) and the ‘9F NMR fluoride dissociation experiment (o) were plot against a. Linear regression analysis of the ‘8119F Exchange TLC experiment (.) (black line) gave the value -0.91 ± 0.12 for the reaction constant for trifluoroborate isotopic exchange, p, -1.20 ± 0.06 for log(ko), and 0.8 16 for R2. Linear regression analysis of the 19F NMR fluoride dissociation experiment (o) (blue line) gave the value -0.99 ± 0.13 for the reaction constant for trifluoroborate fluoride loss, p, -0.92 ± 0.07 for log(ko), and 0.802 for R2 268 Figure 4.13. Mechanism for aqueous aryl trifluoroborate formation and dissociation 271 Figure 4.14. Requirements detailing the ideal ‘8F captor — for ‘8F trifluoroborate based PET imaging 274 Figure 4.15. Strategies for the syntheses of electron-withdrawing/ortho sterically hindered boronic acids with a suitable tether for bioconjugation 276 Figure 4.16. The hypothesized mechanism for base catalyzed protio-deboronation of pinacol boronic acids and our proposed strategy to prevent it 277 Figure 4.17. ‘9F NMR kinetic data for the fluoridation of 4.1 in MeOH. Kinetic experiments were carried out at 5.3 mM Boron and 200 mM Fluoride in 80 % MeOH 289 Figure 4.18. NMR analysis of the TLC purification of trifluoroborate of 4.1, 4.1 TFB. A, top, ‘9F NIVIR spectra before silica chromatography. B, middle and C, bottom, ‘9F NMR (5 % NH4OH/EtOH) spectra of the silica prepared material and ‘H NMR (D20) after silica chromatography and extraction 291 Figure 4.19. A. Kinetic assay of ‘8F -4.1 TFB loss as monitored by TLC in an ‘8F -‘9F exchange experiment. B. Plot of trifluoroborate fraction vs. time as measured by TLC. Data were fit to the function y= a(exp0t) k0b3 = 0.00093 ± 0.00009, a = 0.991 ± 0.007, R2 = 0.95549. t112= 740 ± 70 mm. C. Plot B rescaled relative to the half-life of‘8F 292 Figure 4.20. ‘9F NMR kinetic data for the fluoridation of 4,2 in MeOH. Experiments were carried out at 2 mM Boron and 400 mM Fluoride in 90 % MeOH 294 Figure 4.21. TLC analysis at 366 tim with a 20 % MeOH/CHC13developing solution on borate 4.2 before fluoridation (left), during fluoridation (middle), and after TLC purification 4.2 TFB (right). Photo taken with a common digital camera and a 366 nm UV lamp. 296 Figure 4.22. NMR spectroscopic analysis of the TLC purification of trifluoroborate of 4.2, 4.2 TFB. A, top, ‘9F NMR spectra before TLC chromatography. B, middle and C, bottom,’9NMR and 1H NMR spectra (MeOH-d4) after TLC chromatography 297 Figure 4.23. Kinetic assay of ‘8F -4.2 TFB loss as monitored by TLC in an ‘8F -‘9F exchange experiment. A. 366 tim fluorescence image of the kinetic assay used to monitor 1 F loss from ‘8F -4.2 TFB by TLC in an ‘8F -‘9F exchange experiment. B. xxix Autoradiograph of the kinetic assay used to monitor ‘8F loss from ‘8F -4.2 TFB by TLC in an ‘8F exchange experiment. C. Plot of trifluoroborate fraction vs. time as measured by TLC. Data were fit to the function y’= a(exp0l)t) k0b = 0.000 12 ± 0.00004, a = 0.984 ± 0.005, R2 = 0.7341. t,,2= 5550 ± 1745 mm. D. Plot C expanded relative to 3 half-lives of‘8F (330 mm) 299 Figure 4.24. Structure of biotin, the biotin conjugate to 4.1, 4.7, and the trifluoroborate 4.7 TFB 304 Figure 4.25. ‘9F NMR kinetic data for the fluoridation of a mixture of 4.7 in MeOH. Experiments were carried out at 19 mM (4.7) in 200 mM Fluoride in 95 % MeOH. No ‘8F was used in the NMR experiment 313 Figure 4.26. NMR analysis of the TLC purification of trifluoroborate 4.7 TFB. A, top, 19F NMR spectroscopy before silica chromatography. B, middle, ‘9F NMR (5 % NH4OH/EtOH) spectroscopy of the silica prepared material and C, bottom, ‘H NMR (MeOH-d4) spectroscopy of 4.7 TFB of the silica prepared material 314 Figure 4.27. Time-dependent solvolytic exchange of the biotin ‘8F-4.7 TFB with excess ‘9F- fluoride. A) Autoradiograph of the kinetic assay of[‘F]-fluoride loss from 18F-4.7 TFB as monitored by TLC in an isotopic exchange experiment, 5:95 NH4OH:ethanol. The purified trifluoroborate ‘8F-4.7 TFB was spotted twice for reproducibility. B) Plot of relative autoradiographic density corresponding to the‘8F-4.7 TFB fraction vs. time. Data were fit to the function y a(e0t) k0b3 = 0.00012 ± 0.00006, a = 0.965 ± 0.007, R2 = 0.532. t,,2= 5800 ± 2000 mm. C. Plot B rescaled relative to a period of 300 minutes 316 Figure 4.28. Polyacrylamide gel electrophoresis of neutravidin. A. left, Visualization by 680 nm excitation fluorescence of different amounts of neutravidin bound to fluorescent biotin (left four lanes) and without biotin (right two lanes). B. Right. Visualization of different amounts of neutravidin bound to [‘8F]-4.7 TFB by autoradiography 318 Figure 4.29. PET monitored distribution of the biotin {‘8F]-4.7 TFB as free biotin and the biotin-avidin complex. PET scans showing the coronal view of biotin and biotin neutravidin distribution at 0, 15, 30, and 90 mm following the start of the PET experiment. Injection of the biotin-Neutravidin bound composition occurred 13 minutes prior to the free [‘8F]-4.7 TFB (right mouse) 320 Figure 4.30. Marimastat: Structure, substitution site, and the synthesis of its suitable PET probe 4.8 325 Figure 4.31. ‘9F NMR kinetic data for the fluoridation of a mixture of 4.8 and 4.1 in MeOH. Experiments were carried out at 2.7 mM 4.8 and 1.6 mM 4.1 in 200 mM Fluoride in 80 % MeOH. No ‘8F was used in the NMR experiment 332 Figure 4.32. NMR analysis of the TLC purification of trifluoroborates 4.8 TFB. A, top, ‘9F NMR spectroscopy before silica chromatography. B, middle, ‘9F NMR (5 % NH4OH/EtOH) spectroscopy of the silica prepared material and C, bottom, ‘9F NMR xxx (5 % NH4OH/EtOH) of 4.1 TFB from Figure 4.17 suggesting that the contamination in 4.8 TFB is the benzoic acid 4.1 TFB 333 Figure 4.33. Kinetic assay of{‘8F]-4.1 TFB loss as monitored by TLC in an ‘8F -‘9F exchange experiment 334 Figure 4.34. IC50 assays with MT1 -MMP expressing MDA-MB-23 I cells on 4 different chromatographic preparations of{‘9F]-4.8 TFB eluted with 5 %NH4OH/EtOH. 30 tL of void volume collected off a 400 mg column. Blue, IC50 =6 nM, Red, 10 nM, Green, 5.5 nM, Yellow 4 nIVI. (Image courtesy of Dr. Ulrich auf dem Keller, Overall lab UBC) 336 Figure 4.35. CONTROL [‘8F]-4.1 TFB data. A, left. coronal near infrared spectroscopy mouse image showing signal localized to the tumor. Data were collected following activation of an JR luciferin substrate by luciferase expressed by the 66c 1 4-luc tumor. B, middle. Coronal and sagittal PET data collected 25 to 35 mm post injection (10- 20 mm into scanning). (Images courtesy of Ulrich auf dem Keller). C. Right. Structure of the control: [‘8F]-4.1 TFB. Images are shown approximately to scale 340 Figure 4.36. Marimastat ‘8F-4.8 TFB data. A, left. Coronal near infrared spectroscopy mouse image showing signal localized to the tumor. Data were collected following activation of an JR luciferin substrate by luciferase expressed by the 66c1 4-luc tumor B, middle. Coronal and sagittal PET data collected 25 to 35 mm post injection (10- 20 mm into scanning). (Images courtesy of Ulrich auf dem Keller). C. Right. Structure of the marimastat:‘8F-4.8 TFB. Images are shown approximately to scale 341 Figure 4.37. PET monitored distribution of control [‘8F]-4. 1 TFB in 66c14-luc mice. 5 mm scans of the coronal and sagittal views at 0, 10, 25, and 85 mm following the start of the PET experiment. Injection occurred 15 mm prior to the start of the experiment342 Figure 4.38. PET monitored distribution of marimastat [‘8F]-4.8 TFB in 66c14-luc mice. 5 mm scans of the coronal and sagittal views at 0, 10, 25, and 85 mm following the start of the PET experiment. Injection occurred 15 mm prior to the start of the experiment. 343 Figure 4.39. ‘8F-free fluoride control figures that represent coronal (right) and sagittal (left) PET imaged mice imaged with an aqueous Fluoride-18 (‘8F) control buffered to pH 7.5 with sodium bicarbonate and injected as a 100 mM saline (NaC1) solution. Note the typical distribution of Fluoride-l 8 (‘8F) to the bladder and skeletal system of the mouse. The Fluoride-18 solution was introduced through an interperitoneal injection. Positron emission tomography images were reconstructed from images obtained 60 minutes post-injection 345 Figure 5.1. Rate constant determination for the irradiation of potassium iron(IIJ) oxalate with a hand-held UVP UVGL-55 254 nm shortwave mercury lamp A, top left. Raw kinetic data for the absorption of 700 nm light at different irradiation times. B, top right. Scatter plot and least square fit of kinetic data. C, top left. Results of least square fit to xxxi equation 1. The rate constant k0b is 0.57 miii1 ± 0.07 miii’ for a 1.48 mM solution of potassium iron(III) oxalate 356 xxxii ABBREVIATIONS AND SYMBOLS Phosphate. 1.709 MeV beta emitter ‘8F Fluoride. Positron emitter (two 511 keV emissions produced upon positron annihilation) A Angstrom (1010 metres) f3 Positron 6 Chemical shift (ppm) Molar extinction coefficient (M’cm’) Quantum yield. Measure of the efficiency by which absorbed light produces an effect, Microliter (1 06 litre) Micromolar (106 mole litre ‘) A Adenosine A260 Absorbance at 260 nm Arrhenius law Equation used to predict the temperature dependence of a chemical reaction rate. k = Where A is the preexponential, Ea, the Arrhenius activation energy (can be regarded as a constant), R is the gas constant 8.31 J K’ mor1,T is temperature, and k is a measured rate constant. ATP Adenosine 5’-triphosphate AMP Adenosine 5’-monophosphate xxxiii C Cytidine CB200 Buffer (50 mM TrisHCl pH 7.5, 1 mM EDTA, and 200 mlvi NaC1) CTP Cytidine 5’-triphosphate CD21 Dichioro dideutero methane CDCl3 Trichioro detero methane CH21 Dichioromethane (methylene chloride) CHC13 Trichioromethane (chloroform) CTP Cytidine 5’-triphosphate d Doublet dA 2’-Deoxyriboadenosine dATP 2’-Deoxyriboadenosine 5’-triphosphate dAMP 2’-Deoxyriboadenosine 5’-monophosphate dC 2’-Deoxyribocytidine dCTP 2’-Deoxyribocytidine 5 ‘-triphosphate dG 2’-Deoxyriboguanosine dGTP 2’-Deoxyriboguanosine 5’-triphosphate dT 2’-Deoxyribothymidine dU 2’-Deoxyribouridine D20 Deuterium oxide D-PAGE Denaturing polyacrylamide gel electrophoresis DMF Dimethylformamide (CH3)2N OH DMSO Dimethylsulfoxide ((CH3)2S0) xxxiv DMT Dimethoxytrityl (OH and NH Protecting group) DNA 2’-Deoxyribonucleic acid EDC N-(3 -dimethylaminopropyl)-N’-ethyl carbodiimide hydrochloride EDTA Ethene diamine tetraacetate ESI MS Electrospray ionization mass spectrometry [ES] Enzyme-substrate complex. Also used in this thesis to describe the complex formed between the DNAzyme925-llt and its substrates. [EP] Enzyme-product complex. Also used in this thesis to describe the complex formed between the DNAzyme 925-ilt and its products. [EJT The total concentration of enzyme. The sum of all forms of an enzyme that is free and in substrate-complexes. Also used in this thesis to describe the total concentration of the catalyst925-i it. Et Ethyl EtOH Ethanol f A fraction fo The fraction at the start of the experiment, t = 0 mm fC(. The fraction at the end of an experiment. G Guanosine g Gram GTP Guanosine 5’-triphosphate h Hour HOBt N-Hydroxybenzotriazole (peptide coupling reagent) xxxv HPLC High pressure/performance liquid chromatography HR(MS) High resolution (mass spectrometry) Hz Hertz (s’) J Coupling constant (Hz) k1 The first order rate constant describing the dissociation of substrate from an [ES] complex. Also used in this thesis to describe the dissociation of substrate from DNAzyme925-llt complex kcat MTR The maximum first order rate constant of a catalyst as measured under substrate excess/ multiple turnover conditions. Also the turnover number of a catalyst (minj. kcat STR The maximum first order rate constant of a species measured under substrate limiting/ single turnover conditions (min’). This constant does necessarily reflect species ability for catalysis kcat steady state A kinetic rate constant measured under multiple turnover conditions. An observed first order rate constant that describes a system after it has gone through a burst phase. kchase A first order rate constant that describes the stability of an [ES] complex in pulse-chase kinetic assays. A term that equals the sum of the rate constants which describe [ES] cleavage and [ES] substrate dissociation Kd The equilibrium dissociation rate constant xxxvi k0b5 An observed rate constant. In biphasic systems, the rate constants kOb5fast and k0b5fast are used to describe their respective phase. KMMTR The concentration of substrate required for a catalyst to display a half-maximal turnover rate as measured under substrate excess! multiple turnover conditions. Also the Michaelis constant. KMSTR The concentration of catalyst required for a species to display a half-maximal turnover rate as measured under substrate limiting! single turnover conditions. The pseudo Michaelis constant. LET Linear energy transfer LSI MS Liquid secondary ion-mass spectrometry M Molar (mole litre’) M2 Divälent metal cation. m Multiplet MALDI-TOF Matrix bsorption laser deionization time Qflight (mass spectrometry) MBI 2-Mercaptobenzimidazole (photolysis intermediate) Me Methyl (CH3) MeOH Methanol MeOH-d4 Deuterated methanol. CD3O . MHz Megahertz (106 s’) mm Minute mL Millilitre (10 litre) xxxvii mM Millimolar (lO- mole litre’) MW Molecular weight NWdAjm N-(Histaminyl)-8-deoxyriboadenosine NIRS Near-infrared pectroscopy nm Nanometre (1 O metres) nM Nanomolar (lOg mole litre’) NMR Nuclear magnetic resonance nt Nucleotides MTR Multiple turnover reaction conditions. This notation represents conditions where the substrate is in excess of the catalyst. Under these conditions, a catalyst has the potential to turnover multiple substrates. PAGE olycrylamide gel electrophoresis PCR Polymerase chain reaction PEG Polyethene glycol (water soluble bioconjugates linker) PET Positron çmission tomography pM Picomolar (1012 mole litre’) ppm Parts per million rA Adenosine (ribose sugar) rC Cytidine (ribose sugar) rG Guanosine (ribose sugar) rU Uridine (ribose sugar) xxxviii s Singlet Si The RNA containing substrate that the catalyst925-lit hydrolyzes the fastest. Contains one embedded ribose, rC. 5’- GCGTGCCrCGTCTGTT-3’ STR Single turnover reaction conditions. This notation represents conditions where the catalyst is in excess of the substrate. Under these conditions a maximum of one substrate is processed per catalytic unit. SPECT ing1e,.photon mission computed tomography T Thymidine t Time TBA Tetrabutylammonium (salt) TLC Thin layer chromatography TTP Guanosine 5’-triphosphate THF Tetrahydrofuran (C4H80) UTP Thymidine 5’-triphosphate UV Ultraviolet Vis Visible xxxix ACKNOWLEDGMENTS I would like to acknowledge the assistance of my supervisor Prof. David M. Perrin. Our scientific discussions have led to the body of work that I proudly present in this thesis and his unique guidance has been catalytic in my realization of my potential as a scientist. Most importantly, I would like to thank my family and close friends for their undying faith in my scientific potential during my tenure as a Ph.D student. I always have, and will continue to, attribute my scientific success to their support. These people will always be my mentors in my struggle to balance the difficulties of life with my pursuit of scientific excellence. These people are my mother, Mrs. Liza Jang; my father, Mr. Houng Seng Ting; my brothers, Mr. Robert Ting and Mr. Eric Ting; and my close ffiends, Mr. Raymond Yip, Mr. Allen Yang, Ms. Lily Trinh, Ms. Angela Ho, Ms. Adelphie Cheung, Mr. Richard Chang, Ms. Vivian Shum, Mr. Lu Wang, Ms. Kat Tse, Mr. Minh Ta, Dr. Wayne Chou, Dr. Bobby Lee, Mr. Mickey Wong, and Prof. David M. Perrin. For the DNAzyme work presented in Chapter 1, I would like to acknowledge the scientific mentorship, kinetic data, and scientific advice provided by the Pert-in lab, especially Dr. Leonard Lermer, Dr. Jonathan P. May, Dr. Yoann Roupioz, Dr. Vivian Yip, Mr. Jason M. Thomas, and Mr. Curtis Lam. For the photocaging strategy presented in Chapter 2, I would like to acknowledge the assistance provided by Prof. John Hepburn and his lab, especially Mr. Qichi Hu in assisting with dye laser experiments. I acknowledge Perrin lab members Mr. Chris Hippolito for providing invaluable synthetic precursors pertaining to this project and Dr. Leonard Lermer for providing assistance with MALDI and solid phase DNA synthesis experiments. For the mechanistic strategies for photolytic olefin generation presented in Chapter 3, I would like to acknowledge the assistance of Ms. Julie Lau, Ms. Jennifer S. Steele and Mr. Jeff Chan in generating data elucidating the mechanism of arylthioether photolysis. I would also like to acknowledge Mr. David Dietrich for helpful scientific discussions regarding photocaging. For the PET data presented in Chapter 4, I would like to acknowledge Ms. Julie Lau, Ms. Gillian Lai, Mr. Eric Lamsa, and Mr. Justin Lo regarding work done on identifying candidate trifluoroborates for positron emission tomography (PET). I would like to acknowledge our collaborators at TRIUMF, Dr. Michael J. Adam, and Dr. Thomas J. Ruth for the provision of ‘8F and invaluable scientific discussions. Regarding the in vivo application of the boron technology, I would like to acknowledge Dr. Curtis Harwig for the synthesis of chemical precursors leading to the synthesis of marimastat. I would like to thank Prof. Chris Overall and Dr. Ulrich auf dem Keller who carried out much of the in vivo injection and biodistribution murine work. Other members of the PET team to acknowledge include Ms. Siobhan McCormick, Ms. Pamela Austin, Dr. Yuanmei Lou Prof. Cal Roskelley, and Prof. Shoukat Dedhar. xl This thesis is dedicated to my parents, Mrs.Liza Jang and Mr. Houng Seng Ting xli CO-AUTHORSHIP STATEMENT Prof. David M. Perrin is acknowledged for his help in co-authoring the design of the research program regarding the DNAzyme work presented in Chapter 1. Dr. Leonard Lermer, Dr. Jonathan P. May, Dr. Yoann Roupioz, Dr. Vivian Yip, Mr. Jason M. Thomas, and Mr. Curtis Lam are acknowledged for their help in performing this research. Prof. David M. Perrin is acknowledged for his help in co-authoring the design of the research program regarding the photocaging strategy presented in Chapter 2. Prof. John Hepburn, Mr. Qichi Hu, Mr. Chris Hippolito, and Dr. Leonard Lermer are acknowledged for their help in performing this research. Prof. David M. Perrin is acknowledged for his help in co-authoring the design of the research program regarding mechanistic strategies for photolytic olefin generation presented in Chapter 3. Ms. Julie Lau, Ms. Jennifer S. Steele, Mr. Jeff Chan, and Mr. David Dietrich are acknowledged for their help in performing this research. Prof. David M. Perrin is acknowledged for his help in co-authoring the design of the research program regarding boron based PET tracers presented in Chapter 4. Ms. Julie Lau, Ms. Gillian Lai, Mr. Eric Lamsa, and Mr. Justin Lo are acknowledged for their help in research regarding the identification of candidate trifluoroborates for positron emission tomography (PET). TRIUMF, Dr. Michael J. Adam, and Dr. Thomas J. Ruth are acknowledged for the provision of ‘8F Prof. David M. Perrin, Prof. Chris Overall, Prof. Cal Roskelley, and Prof. Shoukat Dedhar are acknowledged for the design of the research program regarding the in vivo application of this boron technology. Dr. Curtis Harwig is acknowledged for the synthesis of marimastat and Dr. Ulrich aufdem Keller, Ms. Pamela Austin, Ms. Siobhan McCormick, and Dr. Yuanmei Lou are a.cknowledged for carrying out the in vivo injection and biodistribution work. xlii CHAPTER 1: AN IMIDAZOLE AND CATIONIC AMINE-MODIFIED RNASE A MIMIC 11 INTRODUCTION 1.1.1 RNaseA The ability to catalyze reactions under aqueous biological conditions without the need for heightened temperatures or elevated reagent concentrations has unsurprisingly inspired the imitation of the enzyme. The field of biomimetic chemistry is devoted to this goal,’ and few enzymes have received more attention in the past century than RNase A.24 Since its discovery, RNase A has been a cornerstone in the study of enzymes. It is the first enzyme to have its amino acid sequence determined,5the third enzyme to be resolved by X-ray diffraction,6and in 1972, and 1984, this enzyme has been the topic of Nobel prizes regarding the correlation of its chemical structure with its catalytic activity7’ 8 and its solid phase synthesis.9 Publication statement: Versions of this of this thesis chapter have been published. 1) Lermer, L.; Roupioz, Y.; Ting, R.; Perrin, D. M., Toward an RNaseA mimic: A DNAzyme with imidazoles and cationic amines. J. Am. Chem. Soc. 2002, 124, 9960. 2) Ting, R.; Thomas, 3. M.; Lermer, L.; Perrin, D. M., Substrate specificity and kinetic framework of a DNAzyme with an expanded chemical repertoire: a putative RNaseA mimic that catalyzes RNA hydrolysis independent of a divalent metal cation. Nuci. Acids Res. 2004, 32, 6660. 3) Ting, R.; Thomas, 3. M.; Pen-in, D. M., Kinetic characterization of a cis- and trans-acting M2+-independent DNAzyme that depends on synthetic RNaseA-like functionality - Burst-phase kinetics from the coalescence of two active DNAzyme folds. Can. I Chem. 2007, 85, 313. 1 Perhaps the primary reason that bioorganic chemists choose to emulate RNase A is because of its deceptively apparent simplicity. RNase A is a small enzyme (124 residues, 13.7 kDa), its activity is not complicated with cofactor requirements (i.e. divalent metals), and RNase A is capable of diffusion controlled transphosphorylative cleavage of RNA with a first order rate constant, kcat, of 83 000 min’. The enzymatic mechanism of RNase A is relatively simple: RNA cleavage involves two catalytic histidines, Hisl2 and Hisi 19, which cleave RNA through a general acid-base mechanism to produce a 2 ‘-3’ cyclic phosphorane intermediate stabilized by a nearby lysine 41 (Figure 1.1). The cyclic intermediate can then be later hydrolyzed by water with the aid of RNase A and His 12.10 Hisl2 Hisl2His 12 e x”\\ e9 ______ 0—P-old 8 H’ \\ 8 Wtransfer QO\ /08 _______ e OO-H H, 0010 ) NH3 e0\/ O HO ) N 0 OH / N 0 OHLys 41 / Lys 41\(\> /N G i /Lys4l GH 0 OH / \ I> O-P=O / His 119H His 1 His 119”N/” Hisl2 O 12 N _____ I8 His 119 L- o--- Lys41 H, 0 OH NH ,,-2O-H /G \ N H // \ iI> /\ .-N / /\ II) / His 119 Figure 1.1, Mechanism of RNase A catalyzed RNA Cleavage with His 12, His 119 and Lys 41.10 2 1.1.1.1 Early RNase A mimics The seemingly simplistic composition of RNase A’s active site has inspired the mimicry of RNase A’s activity through the localization of imidazole and amine chemical functionalities near RNA phosphodiester bonds. Early work by Breslow involved the functionalization of cyclodextrin supports with imidazoles.’1 Three different cyclodextrins were tested, and a first order cleavage rate, kcat, of 0.084 mind (KM was 0.18 mM, kcat/KM is 466 M’ min’) was reported for the hydrolysis of a catechol cyclic phosphate. Since Breslow’s report, many instances of RNase A mimics have been described involving imidazole functionalized supports such as peptides, acridines and polyalkylamines decorated with imidazoles, amines, bis-imidazoles, di- and triamines, and guanidines.’2’ 13 Such functionalized supports are capable of RNA cleavage but do not approach the high cleavage rate constants seen by RNase A. 1.1.1.2 Sequence specific RNase A mimics To make up for lowered RNA cleavage rate constants of their RNase A mimics, researchers have opted to convey sequence specificity onto RNse A mimics. Sequence specific RNA hydrolysis is not an intrinsic property of RNase A, where only a limited sequence specificity has been shown for tetranucleotide sequences through x-ray crystallographic data.’4 The gene specific targeting of RNase A mimics is usually achieved by attaching Watson-Crick pairing oligonucleotide elements to RNase A mimics to achieve sequence specificity.’5By conveying sequence specificity onto their RNase A mimics, researchers can hope to match sequence- specific enzymatic rate constants• which are decreased in relation to RNase A’s rate constant. The relationship between substrate specificity and large rate constants has been described by 3 Jencks’6 as the “Circe effect” where the activation energy of a reaction is lowered (and a reaction rate accelerated) through the destabilization of an enzyme-substrate complex (rather than the selective stabilization of a transition state). When elements of sequence-specificity are introduced through Watson-Crick oligonucleotide pairing elements, the converse happens, the activation energy of a reaction is increased through the stabilization of the enzyme-substrate complex and hence the rate of reaction is decreased. Thus, high sequence specificfty can lead to a reduction in reaction rates. Despite the negative impact of the Circe effect, the ability to target RNAse activity to specific genes relates to the application of RNAse mimics to RNA manipulation, diagnostic use, the pharmaceutical development of clinical antiviral agents, and the silencing of genes associated with disease.’7 Unfortunately, turnover in sequence specific RNase A mimics has not been described or is very low. This reduced turnover is often ascribed to the fact that sequence-specific guide sequences form Watson-Crick double helix hybrids that are sufficiently stable such that one of the two cleavage products does not dissociate from RNase A mimics following cleavage. 1.1.1.3 Drawbacks to rational approaches to generating RNase A mimics Data on early RNase A mimics suggest that the simple tethering of a scaffold with imidazoles and amines is not sufficient to convey RNase A-like activity onto a mimic, nor is the simple attachment of oligonucleotide pairing elements on a catalytic support sufficient to simultaneously convey sequence specificity and turnover. 4 To achieve the cleavage rates of RNase A, a mimic’s scaffold must contribute to RNA hydrolysis, either by arranging imidazoles at a precise distance from its substrate, through direct contacts with the substrate, or by setting up a hydrophobic environment to perturb pKa’s. To convey sequence specificity onto a mimic; it is more effective to situate the catalytic motif between two recognition domains to reduce individual Watson-Crick pairing interactions in product-catalyst complexes. The syntheses of such species are synthetically challenging and do not guarantee that the mimic will retain activity.’8 Two drawbacks associated with Breslow’s rational design approach hamper the syntheses of catalytically applicable sequence specific RNase A mimics. The first is that the functionalization of different supports with imidazoles requires a significant amount of work on the part of the chemist and the second is that no sufficiently accurate theoretical method exists for predicting the rate enhancement that is contributed by other factors (e.g. hydrophobic forces, solvent interactions, and hydrogen bonding). At the current state of knowledge in enzyme mimicry, sequence specific mimics of RNase A must be identified through a combinatorial approach. Luckily, at about the same time the first cyclodextrin RNase A mimics were being prepared, advancements occurring in the fields of ribozymes would provide the needed combinatorial approach for the selection of RNase A 1.1.2 Nucleic acid catalysts Through nucleic acid hydrogen bonding, n-stacking, and metal ion coordination, single stranded RNA and DNA take on defined secondary structure and irregular tertiary structure similar to the structures responsible for exercising the catalytic diversity seen in proteins. This comparison was first pointed out by Francis H. Crick in 1968 when he proposed that the shared 5 structural similarities between single stranded nucleic acids and proteins would also mean shared chemical properties.22 For RNA, Crick’s hypothesis was shown to be true; In 1989 Tom Cech and Sidney Altman were awarded the 1989 Nobel prize in chemistry for their discovery of RNA-mediated cleavage activity that Cech termed ribozymes.23 These specific RNA sequences, like enzymes, were shown to mediate RNA phosphodiester hydrolysis.2325 1.1.3 In vitro selected ribozymes Unfortunately, naturally occurring ribozyme chemistry is generally limited to M2 dependent phosphodiester cleavage, ligation, or in the case of the ribosome, peptide bond synthesis. In order to expand the role of ribozymes in chemistry, two advances in nucleic acid chemistry make it possible for us to focus scientific efforts on the in vitro generation of synthetic ribozymes in the laboratory. The first advance was discovered when Cech and Altman determined the primary structure of the tetrahymena ribozyme.23 They reported that the tetrahymena ribozyme’s phenotype (its catalytic activity) is directly related to its genotype (its genetic sequence). Like proteins, it is the secondary and tertiary structure of single stranded RNA that gives it its catalytic properties. Unlike proteins, whose sequence information cannot be enzymatically decoded, the primary sequence of RNA (and DNA) can be enzymatically replicated. In vitro selection techniques make use of this direct link between phenotype and genotype in order to directly select ribozymes with new chemical activities.1921 The second advance for the in vitro selection of new ribozyme activity was the development of a set of in vitro RNA replication procedures from which desired sequences could be selected and amplified. In 1990, both the Szostak and Joyce groups reported the first such procedures for 6 the in vitro selection of ribozyme activity.’9’26Since these reports, many different derivatives of these procedures have led to successful ribozyme selection.2734 To discover a ribozyme in vitro, a selective pressure is placed on a large pooi of up to 1015 different RNA sequences that are made under solid phase using heterogeneous mixtures of phosphoramidites. The selection pressure consists of a test which evaluates a sequence’s ability to catalyze a bond formation or catalyze a bond cleavage event. Sequences which show the desired response to the selective pressure are physically separated from undesired sequences based on resulting charge and size differences typically by gel electrophoresis, affinity chromatography, or high performance liquid chromatography (HPLC). Selected sequences are then transcribed and amplified by a high temperature DNA polymerase in the polymerase chain reaction (PCR). The greater the magnitude of the amplification afforded by PCR, the lower the required quantity of starting material, and thus, the larger the pool of different RNA sequences that can be sampled.35’6The resulting amplified products are then reverse transcribed back into active RNA and the resulting pool is passed through an increased selection pressure. This process is repeated until the desired ribozyme activity has been selected. Szostak’s ribozyme was selected to accelerate RNA oligonucleotide hydrolysis.’9 More significantly, Joyce’s ribozyme was capable of cleaving DNA making it the first in vitro selected ribozyme capable of catalyzing a reaction that naturally occurring ribozymes do not.26 Proof that new ribozymes could be generated under in vitro conditions is significant in terms of the noted therapeutic value that oligonucleotides offer to anti-sense technologies and RNAse mimic! bioconjugate chemistry. Since these initial reports, many other ribozymes have been discovered using these in vitro techniques. Many remarkable RNA mediated activities have been described including lead dependent RNA phosphodiester cleavage activity,27 7 polynucleotide kinase activity,28 aminoacyl transfer activity,29’ 30 C-N bond formation,3’ phosphothiolate alkylation,32bridged biphenyl isomerization,33and Diels-Alder chemistry.34 1.1.4 In vitro selected deoxyribozymes Having shown that ribozymes could be selected in vitro, the selection of catalytic DNA became the next goal for bioorganic chemists in the early 90’s. The selection of DNA based catalysts proved to be significant for several reasons: firstly, DNA based catalysis is not a naturally occurring phenomenon, thus the selection of a DNA based catalyst would convey to the scientific community a sense of advancement in the field of biomimetic chemistry. Secondly, the selection and amplification process for in vitro catalytic DNA is simpler than that for ribozyme systems, as transcription steps could be omitted in the selection to reduce the difficulty of carrying out an in vitro combinatorial selection. Thirdly, the chemical stability of DNA is much greater than that of RNA due to the lack of a 2’ hydroxyl group on DNA’s nucleoside sugar. RNA is much more susceptible to hydrolysis as its 2’ hydroxyl is capable of anchimerically assisting in the hydrolysis of its phosphodiester linkages. Finally the process for chemically synthesizing DNA is much more efficient than that of RNA. Again, this is due to the chemical yields involved with the protection and deprotection of RNA’s nucleoside 2’ hydroxyl groups. Thus, longer in vitro-selected DNA catalysts could be mass produced in greater quantities, at lower prices, and stored under aqueous conditions for longer times than a ribozyme. The first selection of a DNA catalyst was carried out by Breaker and Joyce.37 This sequence facilitated the sequence specific cleavage of RNA. Such DNA catalysts were coined deoxyribozymes, or DNAzymes, as a play on the words “ribozyme”, which pertained to RNA, 8 and “enzyme”, which was reserved for protein based catalysts. Before this discovery, no DNA mediated reactions have been known to exist. Following Breaker and Joyce’s selection,37many DNAzyme activities have been selected including light mediated thymine dimer photo reversion,38 DNA deglycosylation,39porphyrin metalation,4°DNA adenylation,4’and DNA- metal mediated oxidative cleavage.42 1.1.5 Drawbacks of ribozymes and deoxyribozymes In terms of mimicking existing enzymatic activities, the chemical rates displayed by in vitro selected DNA and RNA catalysts generally fall short of those displayed by their protein enzyme counterparts. Nowhere is this more apparent than in DNA!RNA based RNA hydrolysis. The most rapid of catalytic RNA hydrolyzing DNAzymes is the 10-23 DNAzyme. The 10-23 DNAzyme does not exactly mimic RNase A as it is Mg2 dependent. Nevertheless, it is capable of catalytically hydrolyzing RNA sequence specifically.43 This DNAzyme acts with a second order cleavage rate constant, kCaS/KM, of i0 M1 miii’, a value comparable to the diffusion controlled limit for a second order reaction (2 x 108 M’ miii’). At 50 mlvi MgCl2,10- 23 acts with a first order cleavage rate constant, kcat, of 3.4 mu1. RNase A, on the other hand, is capable of diffusion controlled second order RNA hydrolysis in a M2 independent setting. RNase A acts with a first order cleavage rate constant, kcat, of 84 000 miii’, 24 700 times that of the 10-23 DNAzyme. Examples of M2 independent DNAzymes that cleave RNA are uncommon.4446 Given that enzymes have a more diverse pooi of chemical functionality from which their activity can be derived from (20 different amino acids as opposed to 4 nucleobases), nucleic acid catalysts sequester divalent metal cofactors to 9 supplement the limited functionality and enhance their chemical activity under physiologically relevant conditions, The requirement of high divalent metal concentrations for fast activity represents a limitation of nucleic acid systems that is not always well defined in novel reports of nucleic acid systems. 1.1.5.1 Nucleic acid catalysts require divalent metals for fast activity At concentrations of Mg2 measured closer to the physiologically relevant estimate of 0.6 mM Mg2,47’48 the first order rate constant for the 10-23 DNAzyme RNase A mimic drops 34 fold from 3.4 mu1 (measured at 50 mM MgC12) to 0.1 mu1 (measured at 2 mM MgC12).43 DNAzyme activity for 10-23 in the absolute absence of divalent metals has not been observed. Even naturally occurring ribozymes have significant M2 dependences. The hammerhead ribozyme can cleave RNA with a first order cleavage rate constant of 870 mu’.49 Unfortunately, when hammerhead derivatives are reconstructed for second order turnover at physiological conditions (pH 7.0 100 1iM MgCl2), this first order cleavage rate drops almost 3000 fold to 0.3 miii’. It is unlikely that faster nucleic acid basedM2-dependent RNase A mimics will come about from further in vitro selection using standard RNA or DNA based in vitro selection strategies. Evidence justifying this statement can be found in the comparison of the DNAzyme selections for catalytic RNA cleavage carried out in the labs of Famulok,5°Joyce,43 and Lu.5’ The three labs carried out independent selections that differed in the degenerate length of their combinatorial library. Famulok’s selection was dependent on Ca2 activity, Joyce’s selection focused on Mg2 activity, and Lu’s selection focused on Zn2 dependent activity. Interestingly enough, the 3 independent selections gave DNA sequences that had nearly identical conserved 10 regions (the 8-17 region43), a region hypothesized to tightly bind these metals interchangeably. This library convergence suggests that the size of a combinatorial nucleic acid library is not as diverse as the 1015 different nucleic acid sequences that are hypothesized, but instead, is limited to the number of DNA folds capable ofbinding metals near RNA targets. The actual rate of RNA cleavage by a nucleic acid system in the absence of M2 has been estimated in three separate accounts where values that vary between 10 to min’ have been reported.446 Only in one case was limited multiple turnover reported (2 turnovers in 100 h).46 At physiological pH and ionic strength, all the aforementioned DNAzymes and ribozymes, both natural and selected, require divalent metal cofactors in order to display first order cleavage rate constants that are larger than these benchmark values. 1.1.6 Chemically modified deoxyribozymes Selection strategies utilizing chemically enhanced nucleic acids have been proposed as a way of broadening the catalytic potential of ribozymes and DNAzymes.36’52, 53 It is hypothesized that by synthetically appending nucleic acids with protein like functionality, nucleic acid catalysts bearing these nucleic acids would not only display cleavage rates more similar to those of enzymes, but modified DNAzyme systems may be relieved of their metal dependence. Unfortunately, the substitution of natural nucleic acids with chemically modified ones also introduces new problems from the perspective of being able to combinatorially select modified nucleic acid catalysts. Before a combinatorial nucleic acid selection can be carried out with modified nucleotide triphosphates, it must be demonstrated that modified nucleic acid triphosphate monomers pair 11 with oligomers in a Watson-Crick fashion such that RNA or DNA polymerases can 1) replicate modified nucleic acids into their complementary Watson-Crick sequences, and 2) generate template DNAs that are amplifiable in the polymerase chain reaction that are used to accurately template the polymerization of chemically modified nucleic acid species from chemically modified RNA or DNA sequences.52 1.1 .61 The successful selection of modified nucleic acid systems The ability to polymerize and then select a system capable of accelerating a reaction using chemically modified nucleic acids was first reported in 1997 by Eaton and coworkers on a copper-dependent modified RNAzyme capable of Diels-Alderase activity using pyridyl modified uridine-C5 RNA oligonucleotides.36Barbas later reported the discovery of a zinc- dependent modified DNAzyme capable of cleaving RNA derived from imidazole functionalized deoxyuridines at C5.53 These initial examples are novel for their display of modified nucleic acid utility in selection; however, they did not show the rate improvement that modified nucleotides were expected to convey onto a nucleic acid system. Eaton’s system could not perform more than one chemical turnover,36 Barbas’ system did not convey a significant rate improvement over an unmodified in vitro selected DNAzyme counterpart,5’and metal- dependent activity is still observed in both systems. A significant rate improvement in the absence of divalent metals was not realized until the selection of the self-cleaving DNAzyme, 925-11. This selection was carried out using both imidazole-modified deoxyadenosines and amino-modified deoxyuridines in 2001. The DNAzyme 925-11, was selected with the expectation that it would mimic the metal-independent nuclease activity of RNase A and could eventually be engineered to hydrolyze RNA under 12 multiple turnover conditions.55A similar discovery was made three years later by the Williams group with a metal independent deoxyribozyme modified at C7 of deazadeoxyadenosine.56 These discoveries are both significant because they verify that by synthetically appending nucleic acids with protein like functionality, nucleic acid systems are made more enzyme-like: they can be relieved of metal-dependence and they display large cleavage rates. 1.1.7 Proposed research The discovery of 925-11 highlights technology that can lead to the selection of catalytic activity never before observed by unmodified systems. The ability to catalytically cleave RNA in the absence of divalent metals is particularly significant for in vivo applications57 where the availability of such cations at millimolar concentrations may limit cleavage activity. 925-11 demonstrates the largest rate constant forM2-independent activity at physiological ionic strength and pH by any nucleic acid catalyst reported to date. The discovery of 925-11 provides the first opportunity to evaluate the catalytic enhancement that can be synthetically conveyed onto DNA by incorporating protein-like chemical functionality onto an easily amplifiable nucleic acid system. This chapter builds on work carried out by Perrin54 following the selection of the cis-cleaving species 925-11 c. This chapter is divided into 3 sections which serve to further characterize the 925-11 system: Section 1.2 discusses the conversion of the self-cleaving925-llc species into the fully catalytic trans-cleaving species, 925-11 t, through solid phase methodology. This section details the use 13 of orthogonal chemical methodology (solid phase) to verify that 925-1 l’s enzymatically incorporated modified nucleotides are required for the hydrolysis of RNA containing substrates. Kinetic studies are reported in this section that allow us to draw initial comparisons between the cis and trans-cleaving modified nucleic acid systems. Section 1.3 investigates the kinetic profile of the 925-1 it catalyst at 24 °C. It is at this temperature where the most rapid turnover of an RNA containing substrate occurs. The rate constant measured at 24 °C represents the largest catalytic rate constant for M2 independent RNA cleavage reported to date for a RNase A mimic under physiological pH and ionic strength. Finally, Section 1.4 forecasts biphasic kinetics as a potential problem for future selected modified DNAzymes and details the kinetic profile of925-lit at 13 °C, the temperature at which the largest first order rate constant, kcat, for M2 independent RNA cleavage by a catalytic RNase A mimic is reported. 1.2 SYNTHESIZING925-lltAS A CATALYST The species 925-1 ic (Figure 1.2) was selected as a cis-cleaving structure, i.e. it is a species capable of accelerating the cleavage of an internal RNA phosphodiester bond, but is incapable of multiple substrate turnover.54 In order to convert- the selected sequence into a catalytic species, it had to be shown that imidazole-containing adenosines and amine-containing uridine phosphoramidites could be polymerized using solid phase methodology.58 Having shown this, 925-11 t (where the letter “t” denotes trans-cleaving) was synthesized by standard solid phase methods.55 14 925-llc GC S c 3’ - CUCGAGCGCCCCGCACCACG GACAACC C A U AT AU G C U U G C A A U U GCC HN j1-S_ -f =A(b1d)5’ -GCGTGCCrCGTCTGTT- 3’ 3’-éàAáà AàAAcc-s’925-llt A U I A U G C CSNJ U U G C A A U U GCC Figure 1.2. The sequence, 925-i ic, prepared by templated enzymatic polymerization of synthetic monomer triphosphates and the synthetically prepared sequence,925-ilt, shown hybridized to its substrate Si. Emboldened A and U represent the modified nucleotides shown at right. The rC indicates the target ribophosphodiester bond. Predicted Watson-Crick pairs are indicated by a dash between blue-coloured complementary bases. Figure reproduced from Canadian Journal of Chemistry.59 The trans-cleaving species, 925-1 it (Figure 1.2), consists of 31 bases, 17 of which form a catalytic domain containing 4 imidazole and 6 allylamino modified nucleotides. This sequence recognizes the 15 nucleotide long substrate, Si, and effects transphosphorylative cleavage at a single embedded ribocytosine at the eighth position to give a 5’-RNA-product terminating in a 2’,3’-phosphodiester and a 3’-DNA-product terminating in a 5’-hydroxyl. Like its parent self- cleaving sequence, 925-1 it catalyzes sequence-specific ribophosphodiester hydrolysis in the total absence of a divalent metal cation, in low ionic strength at pH 7.5, and in the presence of EDTA. The solid phase synthesis of 925-1 it not only represents an important breakthrough in terms of being able to synthesize relatively small biomimetic catalysts but it is the closest chemical 15 synthesis of a sequence specific RNase A mimic in terms of general acid/base catalyst properties,54 rate, physiological operating conditions (pH and ionic strength), and M2 independence. This section details the strategy involved in the reengineering of 925-il as the catalytic species 925-1 it and will present preliminary data that serve to 1) evaluate 925-1 it’s potential as a catalyst, and 2) allow us to account for activity that is lost in the reengineering of 925-il as the catalyst925-i it. i6 1.2.1 MATERIALS AND METHODS 1.2.1.1 Synthetic methods 1.2.1.1.1 Enzymes and Chemicals Enzymes, 32P-labeled nucleotides, reagents, and buffers were purchased from commercial sources. 5-aminoallyl-deoxyuridine and 8-histaminyl-deoxyadenosine phosphoramidites were synthesized as previously described by Perrin eta?.58 1.2.1.1.2 The cis-cleaving sequence 925-11C Oligonucleotides used in the preparation if 925-lie were gifts from the Laboratoire de Biophysique Museum National d’Histoire Natürelle, in Paris, France. The sequence,925-lie (where the letter “c” denotes cis-cleaving), was synthesized by annealing a biotinylated primer; 5’-biotin-GCGTGCCrCGTCTG TTGGGCCCTA CCAACA-3’, to the template DNA with the sequence 5 ‘-T20 GAGCTCGCGG GGCGTGCGCG TGCCTTCACT ACGGATGAGA ACTGTTGGTA GGGCCCAACA GAGGGCACGC TCGTGTCGT-3’ (where rC cytidine). Modified DNA was enzymatically polymerized at 37 °C in the presence of 5-aminoallyl- deoxyuridine triphosphate, 8-histaminyl-deoxyadenosine triphosphate,52[a-32PJdGTP, and the unmodified triphosphates dCTP and dGTP using the enzyme Sequenase 2.0 as previously described.54 The cis cleaving substrate, 925-lic, could be stored at 4°C as the polymerized double helix for up to a week. 17 1.2.1.1.3 Oligonucleotides All oligonucleotides used in the analysis of the catalyst 925-1 it were synthesized using standard automated solid-phase methods on applied Biosystems DNA synthesizers by the Nucleic Acids and Protein Synthesis (NAPS) unit at UBC and the University Core DNA and Peptide Services Unit at the University of Calgary. These oligonucleotides and their sequences were: Substrate Si, 5 ‘-GCGTGCCrCGT CTGH-3’ (where rC = cytidine); the trans-cleaving catalyst925-iit, 5’-CCAACAGUUC UCAUCCGUAG UGAAGGCACG C- 3’, and its inactive derivatives: 925-11t,5’-CCAACAGUUC CUUCGGGUGA AGGCACGC-3’, 925-lit3, 5 ‘-CCAACAGUUC UCAUCCGUAG AGGCACGC-3’ (where U = 5-aminoallyl- deoxyuridine and A = 8-histaminyl-deoxyadenosine),58and And 925-1 it4, the unmodified sequence 5’-CCAACAGUUC UCAUCCGUAG UGAAGGCACG C-3’. 1.2.1.1.432P-oligonucleotide labeling Substrate 51, was 5’ end labeled with T4 polynucleotide kinase lacking 3 ‘-phosphatase activity (Fermentas) and [‘y-32P]ATP. This sequence was purified by electrophoresis using denaturing 20 % (29:1, monomer: his) polyacrylamide/ 7M urea (D-PAGE) gels, ethanol precipitated, desalted on a G-25 spin column, and stored in water prior to use. 1.2.1.2 General kinetic procedures The procedures used in the kinetic characterization of 925-1 ic and 925-1 it were adapted from the procedures used in previous studies of catalytic RNAs.6063 All experiments were performed in mineral oil suspensions to prevent non-specific binding to microcentrifuge tubes and water 18 evaporation during the course of the experiment. Temperatures were kept constant by carrying out reactions in a Julabo F- 10 temperature controlled water bath. Reaction solutions were buffered in CB200 (50 mM Tris-HC1 pH 7.5, 1 mM EDTA, and 200 mM NaCI). Visualization of 32p labeled complexes, substrates, and products were accomplished by exposing gels on Molecular Dynamics phosphor screens at —20 °C overnight and scanning the screens on a Molecular Dynamics Typhoon 9200 phosphoimager. Polygons were drawn around distinct bands with linageQuant v5.2. Quantified volumes were corrected for background phosphorescence by ImageQuant v5.2 histogram peak correction prior to any mathematical treatment. The kinetic constants reported in this text are the result of non-linear least square fits of the cumulative plot of all data sets for identical experiments using the Sigma Plot 2001 v7. 101 data analysis program. Errors associated with the constants reported in this thesis are the standard errors generated from fits. 1.2.1.2.1 Error analysis Regarding rate constants, the standard error values we report were generated by the software used to calculate the value in question. Variance was found to be as high as 60 % when the resynthesis of catalyst925-lit was considered. Others have reported the same for unmodified ribozymes.60’62-64 For the self-cleaving sequence, 925-11 c, errors from day to day were generally lower (approximately 10-20 %) and variance seemed to depend on the freshness of both the Sequenase used to synthesize925-lie and the [u—32P]dGTP used for internal labeling. To control for such error, all self-cleavage measurements were conducted on the same day. 19 1.2.1.3 Cis cleavage by925-llc 1.2.1.3.1 Determination of the observed single turnover rate constant, kcat STR The double helix form of925-llc was bound to magnetic avidin particles (Roche), washed five times with 100 i.tL 0.2 M NaOH, once with 100 p.L of 10 mM cacodylate buffered at pH 6.5, then two more times with 100 p.L of deionized water. The final slurry was suspended in 80 tL of deionized water. The avidin-bound modified DNA strands were divided into 5 p.L time points and incubated at the desired temperatures. Reactions were started by the addition of an equal volume of 2x solution of CB200. Time points were quenched with 20 j.tL of 95 % Formarnide containing 1 mM EDTA and 1 mM biotin. Samples were heated for 5 mm at 95 °C and resolved by 7M urea denaturing 20 % 29:1 monomer:bis polyacrylamide gel electrophoresis (20 % D-PAGE). Data were fit to the single-exponential equation 1: [Pit {P](1_eTRt) 1 where [Pit, and [P] are the fractions of substrate cleaved at time t and the end point respectively and kob is the observed first order rate constant. 1.2.1.4 Trans cleavage by925-llt under multiple turnover (MTR) conditions. 1.2.1.5 Estimation of the maximum multiple turnover rate, kcat MTR Multiple turnover experiments were performed at a ratio of [S]/[E]? 15 in CB200 at different temperatures. 32P labeled substrate was prepared by mixing trace amounts of 5 ‘-32P labeled Si with unlabeled Si. Catalyst preparation, the preincubation of catalysts and substrates, reaction initiation, aliquot quenching and resolution by D-PAGE is as described previously. Multiple turnover kinetic data were fit to the equation: 20 P=k0bMTRt 2 where P is the concentration of Si (in M) cleaved at time t (mm) and k0b MTR is the observed steady state rate of product formation (jtM min’). Reaction carried out at saturating concentrations of Si, values of k0b MTh obtained at different enzyme concentrations were divided by the catalyst concentration, [E]T. Linear regression yielded a slope that represents the maximum first order rate at saturating substrate concentrations, kcat MTR, as described by the following equation. kobs MTR = [EJ keat MTR 3 21 1.2.2 RESULTS The self-cleaving nature of 925-iiC allows us to measure a chemical cleavage rate constant without having to entertain the effects of bimolecular enzyme-substrate association and dissociation. This rate constant will provide a reference from which we can evaluate the kinetic potential of different trans-cleaving derivatives of925-lit under single and multiple turnover conditions, Before a meaningful kinetic study can be carried out, variables such as temperature, salinity, pH, and metal concentrations must be defined in order to standardize kinetic conditions at which kinetic experiments are carried out. These conditions can be chosen to accent any number of reaction aspects, for example, the utility of a system in physiological conditions (i.e. pH, temperature, and ionic strength). In the field of nucleic acid chemistry, these conditions are usually chosen to highlight a particular kinetic property of a system, e.g. the cleavage rate constant of the reaction (kcat) or the conditions that will give the highest number of chemical turnovers. The study of 925-i ic begins with a survey for optimal standard kinetic conditions. 1.2.2.1 The kinetic profile of the cis-cleaving substrate,925-llc Conditions hypothesized to cause an increase in the value of the observed self-cleavage rate constant, kselfcleave, were M2, buffer, ionic strength, and temperature. Between 20-200 mM NaC1 there was a linear increase in kseijcieave. No increase was observed at higher ionic strength (e.g. 0.2-i M NaCl). Mg2 up to 5 mM also had little effect on the value Of kseijciee. Based on these data, the standard kinetic buffer chosen to monitor RNA cleavage consisted of 50 mM TrisHC1 pH 7.5 in order to approximate physiological pH, 200 mM NaC1 in order to approximate physiological ionic strength, and 1 mM EDTA in order to highlight 925-i i’s ability 22 0 10 20 30 40 50 lime (mm) Temperature ( C) Figure 1.3. A, left, Time courses for the cleavage of the cis-cleaving sequence 925-1 ic at different temperatures in CB200 buffer. Radiochemical amounts of925-llc were incubated in CB200 at (o) 8 °C, (V) 13 °C, (Li) 22 °C, and () 30 °C (4 °C and 45 °C not shown). Lines shown are mathematical fits of data points to equation 1. R2 is greater than 0.983; the mean R2 value is 0.989. B, right, Observed cleavage rates, kseieaye, obtained from single exponential fits were plotted against temperature demonstrating an optimal rate constant, ksecieave, of 0.28 min1 at 13 °C. All self-cleavage reactions were performed on the same day using 6 different temperature baths to minimize error and ensure internal consistency (Figure reproduced courtesy of Jay Thomas). Data from one replicate at each temperature was obtained using the same batch of catalyst and substrate. to cleave RNA in the absence of divalent metal cations. This buffer is referred to as CB200 and is used throughout this chapter in order to characterize both the cis and trans-cleaving derivatives of 925-11. The variable associated with the largest change in the value of kselfcleave was temperature. This was determined by plotting the fraction of 925-11 c that is self-cleaved against time. Cleavage approached 70 % at all temperatures, are monophasic, and data fit to equation 1 (Figure 1.3A). In order to check whether a folding step was rate limiting,63’65 pre-folding was attempted by preincubating 925-11 c at various temperatures under various conditions known to suppress cleavage. This was followed by the sudden restoration of permissive conditions.65 No sigmoid nor lag-phase kinetic profiles were observed. Of the temperatures surveyed, the largest cleavage rate constant was observed at 13 °C with a cleave of 0.28 ± 0.02 min1 (Figure 1 .3B). A B > C.) C 0 LI. . . . . 025 —. 020 C 015 0.10 0.05 0 10 20 30 40 50 60 . . 23 1.2.2.2 The synthesis of a catalytic sequence,925-lit The discovery of the cis cleaving species 925-11 c,54 was significant enough to warrant its resynthesis as a trans-acting catalytic version. Had 925-11 c been an unmodified nucleic acid system, the ability to probe its structure-activity relationship could be immediately carried out through solid phase syntheses. As 925-1 ic contained modified nucleotides, the ability to probe its structure-activity relationship required that the modifications associated with this system also be compatible with solid phase techniques. This compatibility was shown with the successful incorporation of imidazole-modified adenosine and amine-modified uracil as modified phosphoramidites in solid phase syntheses (Figure 1.2, Figure 1 4)58 This discovery allowed us to perform a structure-activity relationship assay on trans-cleaving derivatives of 925-11. The four initial sequences synthesized for the determination of the minimal catalytic motif required for trans cleavage are shown in Figure 1.4. 5’ -CCGTGCrCGTCTGTT- 3’ 5’ -GCGTGCCrCGTCTGTT- 3’ 3’ AéAAec-s’ 3’ cACà4 AáAcc-s’ AU AU AU AU CC C C 9-i1t u 92-iit2 G C - CC C U CU CCC 5’ ‘ 5’ -GCGTGCCrCGTCTGTT-3’ 3’ CCACGG GACAACC-5’ 3’ -A AAiec-s’AU AT 925-1 it3 911t T T CcuA A T T T CCC Figure 1.4. Trans-cleaving sequences of 925-11 prepared by solid phase for a structure-activity assay. All sequences are shown hybridized to their substrate, Si. Emboldened A and Us represent the modified nucleotides shown at right. The base, , indicates the target ribophosphodiester bond. Predicted Watson-Crick pairs are indicated by a dash between blue-coloured complementary bases. 24 The rationale behind the syntheses of the sequences seen in Figure 1.4 is as follows: The sequence 925-lit, was expected to be fully active as it contained all 10 of the modified nucleotides that were observed in the active site of the cis cleaving sequence 925-11 c. The sequence 925-i it2 contained only the four modified nucleotides located closest to the cleaved substrate ribose. These modifications include two imidazole-containing adenosines and two amine-containing uridines, chemical functionality reflective of the imidazoles as well as the amines comprising the active site of RNase A. If this sequence is found to be active, it would represent the minimal motif required for RNA substrate cleavage. The sequence 925-1 it3, contained 8 modifications and a three base deletion within the active site. The sequence 925- lit4,was a control sequence which contained only unmodified nucleotides. When substrate Si cleavage was assayed under metal-independent single turnover conditions, only 925-1 it was found to support activity.55 No activity was observed with the three other constructs. These results lead to the following conclusions: 1) that imidazole and amine modified DNA are absolutely required for divalent metal-independent cleavage and 2) that the putative active site of925-lit is structured such that the requirements for RNA hydrolysis activity are much more complex than the localization of the four modified nucleotides located closest to the cleaved substrate ribose. The modified nucleotides in925-ilt are likely to be vital in determining both the secondary and tertiary structure of 925-i it and are likely to be responsible for the actual chemistry involved in phosphodiester hydrolysis. In order to verify these rationales, future work should envision generating structural or computational models of both active and active substrate-catalyst complexes. For the purpose of identifying faster and smaller catalysts, a more comprehensive look into the structure-activity relationship of925-lit needs to be carried out. Prior to this, we were content 25 with analyzing the kinetic profile of 925-i it to elaborate on preliminary results which suggest that: 1) 925-lit exhibits preferential activity towards ribo cytidine when different nucleosides were substituted for the targeted ribose, and 2)13 turnovers were observed in buffered EDTA, the first such observation reported for a nucleic acid system.55 1.2.2.3 The temperature dependence of the catalyst,925-llt A temperature survey was carried out on 925-i lt with substrate Si to determine how the multiple turnover cleavage rate constant varied with temperature. Multiple turnover assays were performed with 100 nM 925-lit catalyst and 15 iM substrate Si at four different temperatures (Figure 1.5). For this initial survey, the occurrence of product inhibition or folding complications was initially ignored. Data fit to equation 2 gave observed multiple turnover rates, k0bMTh, that were divided by the catalyst concentration (100 nM) to give the rate constant keat MTR (equation 3). These data were collected again at 10 tM Si (not shown). As the value of k0b MTR did not vary from data carried out at 15 pM Si, we concluded that the rate constants measured at 15 iM Si were done at conditions in which the catalyst was saturated ([Si] >> Km). The plots Of kcatM shown in Figure i.5B represent a first order rate constant at saturating amounts of Si. This constant can be compared to ksejjccjeave for self cleavage by 925- ii c as a direct measure of the cleavage ability of the 925-il system. 26 - 0.6 I 0.4 8, a z(0 0.2 0.8 A B 0.025 0.020 0.015 0.010 0.005 0.0 . 100 200 300 0 10 20 30400 Temperature (‘C)Time (mm) Figure 1.5. Rough optimal temperature determination for the trans-cleaving catalyst 925-i it. A, left, Time course for the cleavage of925-lit at different temperatures. Cleavage reactions were performed with 100 riM Catalyst, 15 iM unlabeled substrate Si, and <1 nM 5’ 3P labeled substrate in buffer CB200 at (.) 6°C, (/x) 15°C, (U) 24°C, or (o) 30°C. B, right, slopes obtained from regression analyses were divided by the catalyst concentration (100 riM) and plotted against temperature to give a kMTR temperature optimum near 24°C. Data from one replicate at each temperature were obtained. Of the temperatures surveyed, the largest multiple turnover cleavage rate constant was observed at 24 °C with a kCatMTR of approximately 0.026 min1 (Figure 1.5B). 27 1.2.3 DISCUSSION 925-lie is the first nucleic acid catalyst to be combinatorially selected from two nucleotide triphosphates modified with synthetic functionalities that are characteristic of protein enzymes. 925-lic displays RNA self cleavage rates, kseijccieav, of 0.20 min’ and 0.28 min’ at 24 °C and 13 °C respectively. These rates represent a 102 to 1 0 fold improvement over the largest reported rate constants forM2-independent nucleic acid systems4446 and are the largest rate constants forM2tindependent activity at physiological pH and ionic strength by any nucleic acid catalyst reported to date. 925-lie was selected based on its ability to rapidly self cleave at one internal ribophosphodiester linkage. It was not known whether this RNA cleavage activity could be reconfigured to operate in a multiple turnover context until the extraneous DNA which facilitated 925-iiC’S selection was removed following resynthesis of925-iiC as the catalyst 925- lit on the solid phase. The solid phase synthesis of 925-i it was an important feat for two reasons. Not only did this synthesis verify through chemical correlation that the modified nucleotides that are enzymatically incorporated into925-lie are also required for the hydrolysis of RNA containing substrate, but it also demonstrates that chemically modified in vitro selected DNAzyme species can be resynthesized as catalysts on a large scale through solid phase means. Initial kinetic studies on the solid phase constructs of 925-li, suggest that the 925-11 catalytic motif is highly dependent on multiple chemically modified nucleic acids. Of the constructs tested, 925-i it is the only kinetically active trans-cleaving construct of 925-i 1. The products of 925-i it catalyzed cleavage are a 5 ‘-cleavage product containing a 3 ‘-(2 ‘.3 ‘cyclic phosphodiester) and a 3’- 28 cleavage product containing a 5 ‘-hydroxy terminus as determined by MALDI-TOF analysis (carried out by Jason Thomas). 925-1 it’s kinetic profile was investigated in order to draw comparisons between the cis and trans- cleaving 925-11 modified nucleic acid systems. Activity demonstrated by the construct 925-lit suggests that its M2 independent RNAse activity is indeed catalytic. Unfortunately,925-lit, did not facilitate optimal multiple turnover RNA cleavage at the rate or the temperature that was expected. The optimal temperature for multiple turnover catalysis by the trans-cleaving species925-lit was found to be ‘- 24 °C, not 13 °C, where an apparent first order rate constant, kcat Mfl?, was initially calculated to be 0.026 ± 0.010 miii’ Figure 1.5). To complicate matters, the optimal temperature for both single turnover cleavage by the originally selected cis cleaving species is at 13 °C (kselfic!eave, Figure 1.3), yet multiple turnover by 925-1 it was much slower at 13 °C than at 24 °C. The choice of which standard temperature conditions were to be chosen to assay the trans cleaving 925-lit system was complicated by the observation of different temperature optima and rate constants between the cis and trans-cleaving constructs of 925-11. At 13 °C, self cleavage by 925-il c occurs with the largest observed metal independent single turnover rate constant for any RNase A mimic previously reported. However, it is at 24 °C and not 13 °C where925-lit displays the largest multiple turnover rate constant. As the multiple turnover rate constant of 925-1 it at 24 °C is reflective of optimal conditions for the application of925-llt as a catalyst, the kinetic profile of925-lit were first pursued at 24 °C. 29 1.3 THE KINETICS ANALYSIS OF THE CATALYST 925- 11tat24°C. The temperature at which 925-i it could perform the fastest multiple turnover, 24 °C, was initially chosen for the characterization of925-lit. These kinetics were investigated despite the 7.5-fold reduction in keat observed upon the conversion of 925-1 ic into 925-1 it. The multiple turnover rate constant displayed by925-lit at 24 °C represents the activity of the fastest nucleic acid RNase A mimic capable of catalytic M2-independent RNA cleavage activity at physiological pH and ionic strength. The simplest minimal kinetic scheme for nucleic acid substrate cleavage by a nucleic acid catalyst is shown in Figure 1.6. The kinetic behavior of 925-1 it at 24 °C fits this scheme. E P5. + P. = 5.9 0.7 x 106 M1:in ± 0.001 E+S - ES EP5.3 E+P5+P3. k=0.41±::5min EP3.+P5 Figure 1.6. The minimal kinetic scheme for cleavage of substrate Si by925-lit at 24 °C in CB200 (50 mlvi Tris IIC1 pH 7.5, 1 mM EDTA, and 200 mM NaCl). E, Catalyst, S, Substrate Si, P5., 5’ Oligonucleotide product, and P3., 3’ Oligonucleotide product. Kinetic constants are indicated beside their respective steps. The underlined value was calculated from measured constants. This section details the necessary kinetic experiments required to describe 925-i it activity in the minimal kinetic scheme shown in Figure 1.6. The kinetic dissection herein includes the discussion of experimental values for the Michaelis constant (Km), the rate constant for Si dissociation from925-lit (k1), the rate constant for Si cleavage (kcat), and product release rate 30 constants (k3, 1c4, k5, k.6) measured undór single or multiple turnover conditions. From these values, the substrate association rate constant, k1, can be calculated. Single turnover kinetic profiles on different substrates and different Mg2 conditions will be investigated to show that at physiological pH, ionic strength, and in both Mg2 andMg2-independent conditions, 925-i it acts as a sequence specific catalyst that is limited only by keat at 24 °C. 31 1.3.1 MATERIALS AND METHODS 1.3.1.1 General kinetic procedures The procedures used in the kinetic characterization of 925-1 it were adapted from the procedures used in previous studies of catalytic RNAs.6063 All experiments were performed under mineral oil to prevent non-specific binding and evaporation. Temperatures were kept constant by carrying out reactions in a Julabo F-b temperature controlled water bath. Reaction solutions and chases were buffered in CB200 (50 mM Tris’HC1 pH 7.5, 1 mM EDTA, and 200 mM NaC1) except when 1 mM Mg2was present in which case EDTA was absent. Solutions were preincubated at 24 °C for a minimum of lh in order to ensure the reaction of only thermodynamically favored oligonucleotide folds. At least three independent data sets were collected for all experiments unless specified otherwise. Visualization of 32P labeled complexes, substrates, and products were accomplished by exposing gels on Molecular Dynamics phosphor screens at —20 °C overnight and scanning the screens on a Molecular Dynamics Typhoon 9200 phosphoimager. Polygons were drawn around distinct bands with ImageQuant v5.2. Quantified volumes were corrected for background phosphorescence by ImageQuant v5.2 histogram peak correction prior to any mathematical treatment. The kinetic constants reported in this text are the result of non-linear least square fits using the Sigma Plot 2001 v7.101 data analysis program. Errors associated with the constants reported in this thesis are the standard errors generated from fits. 1.3.1.1.1 Error analysis Regarding rate constants, the standard errors values we report are generated by the sofiware used to calculate the value in question. Variance from week to week was found to be as high as 32 30 %, 60 % when the resynthesis of catalyst925-lit was considered. Others have reported the same for unmodified ribozymes.60’62, 64 For the self-cleaving sequence, 925-lic, errors from day to day were generally lower (approximately 10-20 %) and variance seemed to depend on the freshness of both the Sequenase used to synthesize 925-i ic and the dGTPcL32Pused for internal labeling. To control for such error, all self-cleavage measurements were conducted on the same day. 1.3.1.2 Trans cleavage by925-llt under single turnover (STR) conditions 1.3.1.2.1 Determination of the observed single turnover rate constant, kobs STR Cleavage reactions in trans were initiated by mixing equal volumes of925-lit at different concentrations (5 nM to 1500 nM) with trace radiochemical quantities of labeled substrate Si (— 1 nM) in CB200 (buffer) at 24 °C. Aliquots were removed at various time points, quenched with two volumes of a 9:1 formamide: water, 50 mM EDTA, 0.01 % bromophenol blue, 0.01 % xylene cyanol solution, and resolved by 7M urea denaturing 20 % 29:1 monomer:bis polyacrylamide gel electrophoresis (20 % D-PAGE). First order rate constants, k0b STR, were obtained by fitting data to the single-exponential equation 1. [Pjtt= [P](l_eTRt) 1 where [Pit, and [P] are the fractions of substrate cleaved at time t and the end point respectively and kobs is the observed first order rate constant. 33 1.3.1.2.2 Determination of the maximum first—order rate constant (kcat sTR) and the catalyst concentration at which the rate is half-maximal (KM sTR) under single turnover conditions For each of the fast and slow cleaving phases, kcat STR and KM STR were determined by plotting k0bSTR against total925-lit concentration, [Er], using the hyperbolic equation: kcat STR [Er] k0bsTR= KMSTR+[EJ 1.3.1.2.3 Determination of the substrate dissociation rate constant (k..1, k..1 fast, and k..1 slow). Denaturing methods on the substrate, Si Pulse-chase cleavage experiments followed from literature precedent.62’63 The basis for determining the Si dissociation rate constant through a pulse-chase experiment under single turnover conditions lies in the isolation of the catalyst-32P-S1 complex. This is achieved in the pulse phase of this experiment. The addition of the chase molecule (excess unlabeled Si) prevents any rebinding of dissociated 32P-labeled Si to the catalyst. Post-chase, the 32P-S 1- catalyst complex decays through two parallel pathways: substrate cleavage (described by the rate constant keag) and substrate dissociation (described by the rate constant k1). Only the results of the cleavage step can be detected by autoradiography. Experimentally, the “pulse” phase is initiated by combining 1000 nM of catalyst (a saturating concentration of catalyst) with trace amounts (< 1 nM) of 32P labeled substrate at 24 °C. The “chase” is initiated after quantitative [ES] formation, but before Si cleavage has occurred. At saturating concentrations of925-lit, a time of 3 minutes following the pulse was found to be 34 sufficient. The “chase” is initiated by the addition of unlabeled Si in CB200 in 50-fold excess over catalyst. Aliquots were quenched and resolved by D-PAGE as previously described. To account for substrate dissociation, a single turnover experiment was run in parallel at a saturating concentration of catalyst. Data from both the pulse-chase and control experiments were fit to equation 1 to obtain values for kchase and kcat STR respectively. The Si dissociation rate constant, k1, was determined from these constants, according to equation 4:63 k1 = kchase - keat s’R 5 1.3.1.2.4 Determination of the substrate dissociation rate constant (k..1 Si-OMe, k1 fast Si-OMe, and k..1 Si-OMe slow). Non-denaturing methods on a non-hydrolysable substrate analogue Aliquots of 925-1 it (1000 nM) and trace radiochemical quantities of 5’-32P labeled Si-OMe were equilibrated in CB200 at 24 °C overnight. The experiment was initiated by the addition of an equal volume of a chase mixture, consisting of 100 iM unlabeled substrate at 100-fold excess over the 5 ‘-32P labeled Si -OMe in CB200. Aliquots were removed at different times and added to 1110th volume of 70 % sucrose, 0.01 % bromophenol blue and 0.01 % xylene cyanol in CB200 held in an ice bath. Time points were run one to two centimetres into a native 20 % 29:1 monomer:bis polyacrylaniide gel poured in CB200, with CB200 as the electrophoretic running buffer. The electrophoresis apparatus, gel, and buffers were pre-cooled to 4 °C in order to slow substrate dissociation that may occur as the complex runs through the gel. Due to the high salt concentration, the electrophoresis buffer had to be recirculated in order to prevent detrimental pH changes. 35 The first order dissociation rate constant, k1 S1-OMe, was obtained by fitting data to the single- exponential equation: f=f0+f(i -e’ Si-OMe t) 6 wheref is the fraction of Si -OMe dissociated from the catalyst at time t, f° is the fraction that was not complexed prior to addition of the chase,f, is the maximum fraction dissociated, and k.. I SI-OMe is the Si -OMe dissociation rate constant. 1.3.1.2.5 Determination of equffibrium substrate dissociation rate constants 5’-32P labeled substrate analogues Si-OMe and Si-DNA were incubated with varying concentrations of catalyst at 24 °C in CB200 and 7 % sucrose for 1.5 hours. Aliquots were run on native gels as described above. The electrophoresis apparatus was maintained at 24 °C. The fractions of bound substrate analogue were determined and fit against the total catalyst concentration according to the hyperbolic relationship given by the equation: [EP*] — [BiT [P*i+[EP*] (Kd + [BiT) where [EP*]/([P*J+[EP*]) is the fraction of 5 ‘-32P labeled oligonucleotide bound to catalyst at the different total catalyst concentrations (FElT), and Kd is the equilibrium dissociation constant for the catalyst-substrate analogue complex. 36 1.3.1.2.6 Determination of product dissociation rate constants (k..3,k, k..5, and k) Product dissociation constants for 925-1 it were measured on native gels for k1 SI-OMe using 5’- 32p labeled P5 or 3 ‘-32P labeled P3’. These experiments were carried out with each product in the absence or presence of the other at 50 M. Rate constants were determined using equation 6. 1.3.1.3 Trans cleavage under multiple turnover (MTR) conditions 1.3.1.3.1 Determination of the maximum multiple turnover rate, kcat Mm, under conditions where no burst phase is observed Multiple turnover experiments were performed at a ratio of [S]/[E]? 15 in CB200 24 °C. 32P labeled substrate was prepared by mixing trace amounts of 5 ‘-32P labeled Si with unlabeled Si. Catalyst preparation, the preincubation of catalysts and substrates, reaction initiation, aliquot quenching and resolution by D-PAGE is as described for single turnover experiments. No burst phase kinetic profiles were observed. Multiple turnover kinetic data were fit to the equation: P=k0b5Mmt 2 where P was the concentration of Si (in 1.iM) cleaved at time t (mm) andk0bM is the observed steady state rate of product formation (tiM min’). At saturating concentrations of Si, values of k0b MTR obtained at different enzyme concentrations were plotted versus catalyst concentration, [E]T. Linear regression yielded a slope that represents the maximum first order rate at saturating substrate concentrations, kCaIMTR, as described by the following equation. kobs MTR [E1j keat MTR 3 37 ‘1.3.1.4 Attempts at observing a reverse reaction (ligation) 1.3.1.4.1 Single Turnover Product Ligation Experiments Ligation experiments were performed by mixing equal volumes of solutions containing 2000 nM catalyst, 30 iiM unlabeled P3’, and a solution containing trace 5 ‘-32P labeled P5’ (< 1 nM) at 24 °C, resulting in a final solution of 1000 nM catalyst, 15 j.tM 3 ‘-product, and trace quantities of the 5’-32P labeled P5 (— 1 nM) in CB200. Aliquots were removed, quenched in the formamide solution described above, and resolved on a 20 % D-PAGE gel. This experiment was repeated with 15 i.tM unlabeled P5’ and trace amounts of 3’- 32P-labeled P3. Data for all experiments were collected over a 5 to 2100 mm period. 1.3.1.4.2 External Equffibrium Shift Experiments Catalyst was incubated with trace quantities of 5‘-32P labeled Si in CB200 at 24 °C to give a reaction that was 200 nM in catalyst. This reaction was allowed to proceed until the substrate had cleaved to near completion (19.5 hours or 50.1 half-lives). A chase solution of unlabeled P3 in CB200 was then added so that the final concentration of 3 ‘-unlabeled product was 2.7 tiM. A control reaction was set up where an equivalent volume of CB200 buffer absent P3’ was added as the chase. Aliquots from both sets of reactions were removed, quenched in the described formamide solution, and resolved on a 20 % D-PAGE gel. Quantities of 5 ‘-32P labeled Si and 5 ‘-32P labeled P5 were monitored at different times in attempts to measure a ligation rate. Data were collected over a 5 to 2100 mm period following the chase. 38 1.3.2 RESULTS 1.3.2.1 Determination of the maximum first—order rate constant at saturating catalyst concentrations, (kcat sTR), and the concentration of catalyst at which the reaction rate is half-maximal (KM sTR) The pre-steady state kinetic constants kcat STR uid Km STR were determined under single turnover conditions. Data from single turnover experiments obtained at catalyst concentrations ranging from 10 nM to 2500 nM were fit to equation I (Figure 1 .7A). The observed pseudo-first-order constants, k0b3, obtained from these fits were plot against catalyst concentration in the hyperbolic equation 4, to give values for JCCatSTR and Km STR (Figure 1 .7B). A B CD CD C) C 0 C) CD LL Figure 1.7. A, left, Plots of the fraction of substrate Si cleaved vs. time at 24°C. Cleavage reactions were performed with varying concentrations of catalyst and < mM 5‘-32P Si in buffer CB200 at 24°C. The catalyst concentrations at which kinetic profiles were observed are: (•) 1200 nM (D) 150 nM, () 80 nM, (V) 40 nM, (0) 20 nM, or () 10 nM Catalyst. Data from one replicate at each concentration of catalyst were obtained. All catalyst concentrations assayed not shown. B, right, Plot of kObS vs. catalyst concentration. Rate constants obtained from the single exponential fits in A were fit to a hyperbolic plot against catalyst concentration equation 4. A non linear least squares fit gave a maximum first—order rate constant (kcat sTR) of 0.037 ± 0.001 min’ and a catalyst concentration at which the reaction rate is half-maximal (Km STR) of 69 ± 7 nM. R2 is 0.97 8 The maximum 24 °C first-order rate constant at saturating catalyst concentration, kcat STR, is 0.037 ± 0.001 min1, and the concentration of catalyst at which the reaction rate is half- 0 500 1000 1500 2000 1000 1500 Time (mm) Catalyst (nM) 39 maximal, Km STR, is 69 ± 7 nM. From these pararnetres, a second-order-rate constant (kcatSTR/Km sri?) of 5.3 ± 0.5 x iO M’ miii’ was calculated (equation 4). 1.3.2.2 Determination of the substrate SI dissociation rate constant (k1) through denaturing polyacrylamide gel methods The dissociation rate constant of Si was first determined by pulse-chase experiments under single turnover conditions. The basis for the pulse-chase experiment lies in the isolation of a 32P-labeled catalyst-substrate complex, 32P-[ES] (Figure 1.8). The formation of this complex occurs in the pulse step, whereby 32P labeled Si is incubated with catalyst. The isolation of this complex is achieved upon addition of a large quantity of the chase species, in this case unlabeled Si. This addition halts the formation of any further 32P-labeled complex [ES]. The isolated post-chase 32P-[ES] complex can decay through either of two pathways: substrate cleavage (kcaj, or dissociation of the substrate from the substrate-catalyst complex (k1) (Figure 1.8). Because only the results of substrate cleavage (kcat) can be detected in denaturing gels, the observed rate of 32P-[ES] decay through Si cleavage will appear to proceed with a rate constant, kchase, that is the sum of kcat and k1 (Figure 1.8). k1 keat E ÷ . [E’S] - E + k1 Quench with Excess S (cold) Prevents k1 from occuring [E’S] B + ‘P k1 Figure 1.8. Pulse-chase experiment for k1. Catalyst is “pulsed” with32P-labeled substrate aid given enough time to form a quantifiable quantity of catalyst-label complex. After a time, t1, the chase is added which competes for the 5’-32P labeled Si. The chase is applied before a significant amount of E*S is cleaved (3 mm). Due to the excess of the unlabeled chase over both the labeled substrate and total catalyst content, all free catalyst is complexed in a form that cannot associate with 5’-32P labeled Si. Further formation of catalyst-5’-32Plabeled Si complex is essentially halted, allowing one to quantitate kinetic constants of 32P-[ES] decomposition without having to deal with32P-[ES] formation. Bold arrow indicates removal of free E following the chase. 40 Substrate dissociation rate constants were determined by incubating sub-saturating amounts of catalyst with trace amounts of substrate for 0.5 mm to 3 mm before the chase was added. The resulting data were fit to equation 5. kcat was subtracted from the chased decay constant, kche, to give values for k..1. The time between the pulse and the chase was chosen with these two considerations in mind: enough time was allocated between the pulse and the chase so that a quantifiable amount of ES complex could be formed, and the chase was added at a time where minimal turnover has occurred. With a koat of 0.037 min’1, or a half-life of 19 mm, a 3 mm interval between pulse and chase is sufficient (i.e. < 10 % of the pulse is cleaved). Experimental pulse-chase data were fit to equation 1 to obtain values for the observed decay constant kchase shown in Figure 1.9. 8 200 400 600 800 1000 1200 0 20 40 - 60 80 100 120 140 160 Time (mm) Time (mm) Figure 1.9. Determination of the Substrate Dissociation Rate Constant for Substrate Si from Pulse-chase Experiments at 24°C. Saturating amounts of Catalyst (Control reactions, (•) 1800 nM and (•) 1200 nM) were incubated with <1 nM 5‘-32P Si and run as a control for pulse-chase experiments. Three replicates of pulse-chase reactions (G, C, and 0) were performed with 1000 nM Catalyst, <1 nM 5’ 32P labeled Si and 75 i.tM of excess unlabeled Si as the chase. Fraction cleaved values for chased reactions have been corrected for substrate cleavage that occurred prior to the chase. Zero time for the chased reactions were referenced to the time at which the chase was added. Results of the fit for the chased reaction: kche = 0.41 ± 0.05 mm4,Amplitude = 4.0 ± 0.1 %, R2 is 0.849, The Si dissociation constant, k..1, was determined by subtracting the observed first order rate constant from the chased reaction from the rate constant of the saturated control reaction according to equation 5 (kchase - kcaj.63 The observed rate constant of the chased reaction, kcat 41 0.6 0 5 0.4 ci STR is 0.037 ± 0.001 min’. This value was subtracted from the observed rate constant of the chased reaction, kche, 0.41 ± 0.05 min1. The resulting value for the dissociation rate constant for Si, k.j, was found to be 0.37 ± 0.05 min’. 1.3.2.3 Determination of the substrate dissociation rate constant (k..1 SI-OMe). Non- denaturing methods on a non-hydrolysable substrate analogue Because the error associated with k1 is high, it was necessary to corroborate the dissociation rate constant for Si through pulse-chase native gel experiments on the non-cleavable 2’cytidine methyl ester analogue of Si, Si -OMe. In these experiments, the pulse was initiated with the addition of5’32P- labeled Si-OMe. The chase was accomplished through the addition of a large excess of unlabeled Si -OMe. Native gel shift data and fits to equation 6 are shown in Figure 1.10. 1.0 n o 8 0.8 0a) 0.6 0 0 0.4 U a) LI 0.2 0.0 0 20 40 60 80 Time (mm) Figure 1.10. Pulse-chase analysis by native gel electrophoresis to determine the first order rate constant for substrate analogue dissociation from925-lit at 24°C. A trace of 5‘-32P Si -OMe substrate analogue was incubated with catalyst (1 i.tM) in CB200. Following the addition of an equal volume of the chase (100 jiM substrate in CB200), aliquots were taken at various times and loaded onto native PAGE gels. The fraction of 5‘-32P Si -OMe dissociated was plotted versus time for three experimental replicates ci, and 0). A non-linear least squares fit gave a first order dissociation rate constant, Icj SJ-OMe, for the substrate analogue S 1-OMe of 0.44 ± 0.03 min’ R2 is 0.979, (Figure reproduced courtesy of Jay Thomas) 42 The value 0.44 ± 0.03 min’ was measured for the dissociation rate constant of the 2’OMe analogue of Si from925-llt, k1 SI-OMe. This value is consistent with the measured value of k4 for Si. 1.3.2.4 24°C Multiple turnover kinetic profiles Multiple turnover kinetic experiments were performed at varying catalyst concentrations (50 — 1000 nM) and saturating Si concentrations (15 p.M). Burst and lag phase kinetic profiles were not apparent in plots of Si cleaved as a function of time (Figure 1.11 A). B 0.04 0.03 C S .S 0.02 a 0.01 0.00 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Time (miii) Catalyst (M) Figure 1.11. A, left, Multiple turnover analyses on the cleavage of Si at 24 °C. Cleavage reactions were performed with varying concentrations of catalyst, 15 iM Si, and <1 nM 5’ labeled Si in buffer CB200 at 24°C. The concentrations of catalyst at which kinetic experiments were observed are: (•) 1000 nM (D) 500 nM, (z) 200 nM, (V) 100 nM, or (0) 50 nM Catalyst. Data from one replicate at each concentration of catalyst were obtained. Data were fit to linear equations. The lowest R2 value was 0.996. Cleavage was allowed to proceed until a maximum of 20 % of the substrate was cleaved. Reactions were quenched in formamide at the times indicated on the graph. B, right, Rate constant determination for the maximum steady state rate constant. Values ofk0b3M71? was plot against catalyst concentration to obtain a steady state rate constant at saturating substrate concentrations. The intercept for this value is 0.000 ± 0.001 p.M/mm. The slope, ktMm, is 0.030 ± 0.002 min’ and R2 is 0.981. Linear regression analyses (equation 2) of these data gave observed rates (k0b MTR) that varied directly with catalyst concentration (equation 3) (Figure 1.11 B). The first order rate constant at saturating substrate conditions for multiple turnover substrate Si cleavage, kcat MTR, is 0.030 ± 0 100 200 300 400 500 600 43 0.002 min1. This value is very close to the maximum rate constant for Si cleavage obtained under single turnover conditions (kcatSTR = 0.037 ± 0.001 min’). 1.3.2.5 Gel-shift assays for equilibrium dissociation constants of catalyst- substrate analogue complexes (Kd) Under single turnover conditions, the concentration of catalyst at which the reaction rate is half- maximal, Km STR, was measured to be 69 ± 7 nM (Figure 1.7). Assuming that the interaction between925-lit and its substrate 51 were a simple bimolecular process, this value would also represent the Michealis constant, KMMTR, measured under substrate excess conditions. The value of KM is also an appropriate estimate of the equilibrium dissociation constant, Kd, for the catalyst-substrate complex when the dissociation rate constant, k1, for the substrate is much greater than the rate constant of cleavage, kcat. Under single turnover conditions, this is the case as k1 is 16 times larger than kcat. If the association of 925-i it with its substrate Si were a simple biomolecular process, Kd values of the noncleavable DNA and 2 ‘OMe substrate analogues of Si could be measured to confirm the value of Km STR determined for S 1(69 ± 7 nM). This was carried out by incubating different amounts of DNA and 2’OMe Si substrate analogues with 925-lit at different catalyst concentrations (Figure 1.12). Data were fit to equation 7. 44 A 120 100 60 •60 0 40 20 0 B 100 80 Eeo z 0 i40 20 0 0 200 400 600 800 1800 1200 0 200 400 600 800 1000 1200 Catalyst Concentration (nM) Catalyst Concentration (nM) Figure 1.12. Non-denaturing PAGE analysis of the binding affinities of non hydrolysable substrate analogues of Si to 925-i it. Data were fit to equation 7. A, left, 32P labeled S 1-OMe bound to catalyst vs. catalyst concentration. Data from two replicates were obtained at each concentration of catalyst. B, right, Plot of the fraction 32P labeled Si-DNA bound to catalyst vs. catalyst concentration. The value of Kd for Si -OMe was 82 ± iO nM and had an R2 values of 0.967, The value of Kd for Si-DNA was 43 ± 9 nM and had an R2 value of 0.970. Data from one replicate at each concentration of catalyst were obtained (Figure reproduced courtesy of Jason Thomas) Data measured by native gel shift electrophoresis were fit to equation 7 to give the values 43 ± 9 nM for Si-DNA, and 82 ± 10 nM for Si -OMe. These values of Kd derived from noncleavable substrate analogs agree reasonably well with the value of Km STR (69 ± 7 nM) measured for Si under single turnover conditions. . •• . I I I • 1.3.2.6 Gel-shift assays for dissociation rate constants of product oligonucleotides (k..3, k..4, k5, and k..6) The dissociation rates for both the 5’ and 3’ oligonucleotide products were too large to measure using non-denaturing native gel-shift methods. In the case of the 3’ product, no product— catalyst complex could be observed in a native gel at 1 mM 925-1 it. In the case of the 5’ product, product-catalyst complex formation was observed, but complete dissociation prior to even the earliest time point (30 sec.) made the accurate measurement of a rate constant impossible. These results force us to conclude that the dissociation of the cleavage products 45 from catalyst-product complexes are much faster than the cleavage step, kcat STR, at 24 °C and that product inhibition of multiple turnover is unlikely. 1.3.2.7 24 °C Substrate specificity studies To test the sequence specificity of925-lit, unpaired nucleotide bases at or downstream from the site of phosphodiester hydrolysis were synthesized and tested in a single turnover context. The six sequence variants of Si are shown in Table 1.1. Table 1.1. Comparison of kcat STR s and KM STR s for substrate analogues of Si. Underscored bold type indicates specific base deviations from unmodified Si. kreiatjve values relative to substrate Si were obtained by dividing the value of kcat’ obtained for other substrate analogues by the value of kcat STR in entry 1 (data courtesy of Jason Thomas). Entry Substrate KMSTR (nM) kcat STR (min’) kreiative 1 5’-GCGTGCCrCGTCTGTT-3’ (Si) 69± 7 0.0370 ± 0.001 1 2 5’-GCGTGCCrAGTCTGTT-3’ 50±20 0.0101 ± 0.0004 0.67 3 5’-GCGTGCCrUGTCTGTT-3’ Not measured 0.001390 ± 0.00003 0.09 4 5’-GCGTGCCrGGTCTGTT-3’ Not measured 0.000659 ± 0.00003 0.04 5 5’-GCGTGCCrCATCTGTT-3’ (+i) Not measured 0.00013 ± 0.00003 0.009 6 5’-GCGTGCCrCACCTGTT-3’ Not measured 0.00015 ± 0.00003 0.01 7 5’-GCGTGCCrCGCCTGTT-3’ (+2) Not measured 0.00 1230 ± 0.000005 0.08 As was done for Si, first order rate constants (kcat sTR) were determined for each substrate at saturating catalyst concentrations. Substrate 51 was cleaved the fastest by925-lit. For the substrates that displayed kcat STR values similar to those of S 1, values of Km STR were calculated from plots of pseudo first order rate constants (k0b) at different catalyst concentrations (equation 4). For substrates where the first order rate constants, kcat STR, were ten fold lower than that of Si, catalyst saturation was confirmed by comparing the observed first order rate constants, k0b, at 46 two concentrations of catalyst both seen to be saturating in the case of Si. At 5 M or 10 i.tM catalyst, values ofk0b STR differed by less than 5 % for all substrates assayed, therefore it was assumed that at 5 jiM catalyst, substrates were saturated and the measured values of k0b sr were accurate approximations of kcat STR• The substitution of cytidine for any other base at the site of cleavage, effected a drop in kcat sji values by up to a factor of 50 (Table 1.1, Entries 1-4). Purine to purine and pyrimidine to pyrimidine substitutions were investigated at sites +1 and +2 from the ribose cleavage site. Of the measured substitutions, a purine to purine G—*A substitution next to the cleavage site (Table 1.1, Entries 5-6) produced the most significant drop in rate constant relative to Si, while substitution of T for C at the +2 position only reduced keat by an order of magnitude. This result is suggestive of a non-Watson-Crick pair with guanidine at the + 1 site of Si that is detrimental in Si substrate cleavage by 925-i it. To identify the reasons for nucleobase substitution affects on activity, future work should envision generating structural or computational models of nucleobase substituted complexes. 1.3.2.8 Effects of 2 mM Mg2’ on 925-I It at 24 °C In order to evaluate the hypothetical utility of modified, M2tindependent DNAzymes for targeting mRNA in cells, the multiple turnover activity of 925-lit was investigated at a physiologically relevant concentration ofMg2. We chose to study the activity of925-lit in the presence of 2 mM Mg2, a concentration which reflects the intracellular physiological concentration ofMg2more accurately estimated at ‘- 0.6 mM (Figure 1.13). 47,48 47 2.5 2.0 - 1.5 ci) > ci) 0 1.0 U, 0.5 0.0 I I 0 100 200 300 400 500 Time (mm) Figure 1.13. Multiple turnover analysis on the cleavage of Si at 24 °C in the presence and absence of 2 mlvi MgSO4. Cleavage reactions were performed with 150 nM catalyst, 15 iM SI, and <1 nM 5’ labeled SI in 50 mM Ti-is HC1 pH 7.5 buffer, 200 mM NaC1, and either (.) 1 mM EDTA or (0) 2 mM MgSO4 at 24°C. Cleavage was allowed to proceed until a maximum of 20 % of the substrate was cleaved. Reactions were quenched in formamide at the times indicated on the graph. Linear regression analyses gave Si cleavage rates (k0bSM7’.1) of (.) 0.0048 ± 0.000 1 p.M’ ntin’ and (o) 0.00554 ± 0.00003 j.tM1 min’. Applying the observed rates to equation 3, the steady state cleavage rate (kcatR) of925-llt is 0.032 ± 0.007 min1 in the presence of 2 mM EDTA (.) and 0.037 ± 0.002 mm1 in the presence of 2 mM MgSO4 (0). The value for kcatMTh is 0.037 ± 0.002 min1 in the presence of 2 mM Mg2,while the value of kcat MTR is 0.032 ± 0.007 min’ in the presence of 1 mM EDTA. These two constants agree within error. Not only can 925-i it act independently of Mg2,but its rate is unaffected by the presence of Mg2. From this study it is predicted that physiological concentrations of Mg2 will not affect catalysis by 925-1 it making these results important for physiological applications of 925-i it. This data implies that future selected imidazole and amine modified nucleic acid catalysts will not be affected by Mg2 at 2 mM. 48 1.3.3 DISCUSSION 1.3.3.1 The kinetic model for925-llt at 24°C The minimal kinetic scheme for Si cleavage by 925-i it at 24 °C and the kinetic rate constants measured in this chapter are shown in Figure 1.14. Some values that were directly measured with Si were corroborated with rate constants measured with the non-cleavable substrate analogues, Si -OMe and Si-DNA. These values are also shown in Figure 1.14. ‘cat STR = 0.037 ± 0.001 mm4 E P5’ + P3. 1cat MTR 0.030 ± 0.002 mm1 k1=5.9±0.7x_106 M4min-1 E+S - ES - EP5.3 E+P5.+3 ±0.05 min1 E P3*p5. KmSTR=69±7flM Kd Si DNA =43±9 nIVI KdSIOMe=82±lOnM Figure 1.14. The minimal kinetic scheme for cleavage of substrate Si by 925-1 it at 24 °C in CB200 (50 mM Tris HC1 pH 7.5, 1 mM EDTA, and 200 mM NaC1). E, Catalyst, S, Substrate Si, P5., 5’ Oligonucleotide Product, and P3., 3’ Oligonucleotide product. Corroborative data are shown for various experiments with the non cleavable 2’ ribo methyl ester of Si and the DNA equivalent of Si. The substrate dissociation rate constant, k1, was confirmed through pulse-chase denaturing assays on Si as well as native gel assays for the dissociation of Si -OMe (k4 SI-OMe) . Both studies show that substrate dissociation is much faster than Si cleavage allowing us to confirm values of Km STR measured in single turnover assays through measurements of substrate equilibrium dissociation rate constants (Kd) on the non cleavable analogues of Si, Si-DNA, and Si -OMe. Rapid Si association with925-lit, relative to Si cleavage and Si dissociation, made it difficult to measure an association rate constant through experimental precedent.6063 The Si 49 association rate constant, k1, as calculated from values of Km S77?, kcat STR, and k1 is (5.9 ± 0.7) x 106 M’ min’. This value was verified using the calculated value of the association constant for Si-OMe (4.4 ± 0.4) x 106 M’ min’ from values ofKdSJQMe and k1 Si-OMe. Regarding events that occur prior to Si cleavage, the simple single-step binding mechanism for 925-1 it in Figure 1.14 was proposed based on agreement between constants measured under kinetic pulse-chase experiments (k1) and dissociation constants measured with substrate Si - OMe in native gel shift assays (lci SI-OMe). The constant k1 is used to measure substrate dissociation indirectly from an experiment where product formation was monitored, while k.1 si OMe was measured by directly monitoring substrate analogue dissociation. If a stable [ES] intermediate were formed prior to the complex capable of cleavage66 these two rates would not be in agreement. Product dissociation constants could not be accurately measured by pulse-chase/native gel experiments because they were too large. These data suggest that the product dissociation constants are much larger than kcat. This observation is confirmed by the absence of a burst phase in 24 °C multiple turnover assays (Figure 1.11) and by the agreement between the value Of kcat in catalyst excess conditions (kcat sTR) and substrate excess conditions (kcatMTR). 1.3.3.2 The importance of catalyst excess conditions (STR) vs. substrate excess conditions (MTR). single iurnover reaction kinetic experiments (denoted with the STR notation) are useful in monitoring rates up to and including the substrate cleavage event without having to deal with complicating factors that follow cleavage. The use of STR conditions is common in the field of 50 nucleic acid catalysts and allows one to avoid complicating effects such as slow product release. Conversely, for catalytic applications, the steady state rate constant (kcat MTR), as observed in multiple turnover kinetic profiles, is a more relevant constant to measure as it more accurately describes the behavior of a system that is used as a catalyst. In the case of 925-i it at 24 °C, kcat, measured under multiple turnover conditions (kcat MTR = 0.030 ± 0.002 min’) agrees with the rate of Si cleavage measured under single turnover conditions (kcatSTR = 0.037 ± 0.001 min’). This agreement suggests that the Si hydrolysis step is the common rate limiting step under both STR and MTR conditions and confirms the model shown in Figure i.i4. Finally, this agreement is used to conclude that slow product release does not limit cleavage. 1.3.3.3 Regarding ligation It was important to investigate the possibility of ligation by925-iit. The reaction generating the two product oligonucleotides involves transphosphorylative cleavage of the RNA phosphodiester bond. This reaction is independent of water, thus the ligation of the 3’ and 5’ product oligonucleotides generated from the cleavage of Si could represent a significant reverse rate that is not discouraged by the large concentration of 55 M water. Ligation is not seen for925-lit. By the principal of microscopic reversibility, all nucleic acid systems should be able to ligate their products. This is true for some systems such as the hammerhead6°and hairpin62 ribozymes as well as the i 0-23 DNAzyme.67 Thermodynamic issues can make it difficult to identify a ligation product i.e. when fast substrate cleavage is followed by rapid product release63 or catalyst conformational changes occur following substrate cleavage.68 These phenomena have been used to account for the lack of ligation observed in the naturally 51 occurring HDV ribozyme system. Ligation products are not observed with 925-i it, so these phenomena may account for the lack of ligation in 925-1 it. I .33.4925-lit Substrate specificfty The substrate specificity of 925-11 was investigated with different substrates in Table 1.1. 925- 11 is quite sensitive to the nucleotide base nature of the RNA nucleotide that is to be cleaved. C—U substitution at this site gives a 25-fold reduction, C—÷G a 50-fold reduction, and C—>A 4- fold reduction in rate (kcat STR). These results suggest that the hydrogen donating exocyclic amine on A and C and/or the hydrogen acceptor (Ni-of adenine and N3- of cytosine) are critical for recognition. G—*A substitution at the base +1 to the scissile ribose (Table 1.1, Entries 5 and 6) produced the largest drop in the value of kcat (150- fold) while substitution at the at the base +2 to the scissile ribose resulted in a less pronounced twelve-fold drop in the value of kcat. (Table 1.1, Entry 7). Interestingly enough, Watson-Crick pairs cannot be proposed for the bases +1 and +2 to the scissile ribose. This is common to many other DNAzymes, even those that act in M2 dependent settings.43’69, 70 The activity displayed by 925-i it in an M2 independent setting suggests that the added synthetic functionality on 925-i it provides a recognition motif with the +1 and +2 bases that may be used to reduce transition state energies during the cleavage of 925- lit. The identification of such motifs may prove useful in designing new RNase A mimics. The x-ray crystallographic resolution of this motif may explain why 925-i it can work without clivalent metals while all the other nucleic acid systems require them. 52 1.3.3.5 Comparison to other nucleotide catalysts Rate constants for different DNAzymes and ribozymes capable of multiple turnover cleavage have been compiled in Table 1.2 such that a comparison can be made between 925-1 it and other well-known Mj-dependent systems that display turnover.60’62, 63, 67, 71-73 It is important to point out that the RNA cleavage rate constant, kcat, for many of these entries have been emphasized through their measurement in conditions where divalent metal concentrations have been elevated past physiologically relevant values (i.e. >2 mM) and that values of kcat reported in Table 1.2 are not relative in conditions where M2 is absent. Nevertheless, this table is presented to defme the rate constants that future M2 independent modified nucleic acid selections must surpass in order to demonstrate superior catalytic potential over M2 aided unmodified nucleotide ribozymes and DNAzymes. Table 1.2. Comparison of kinetic rate constants for different trans-cleaving ribozymes systems6°’ 62, 63, 67, 71-73 Entries 1 to 3 are DNA.zymes, while entries 4 to 8 are ribozymes. Where multiple literature reports exist, rate constants are reported at lowered Mg2 concentrations. (table reproduced with the help of Jason Thomas). Substrate length Substrate Substrate Multiple Catalytic (Number of Association Dissociation Turnover Efficiency Proposed Watson- Rate Constant, Rate Cleavage Rate kcat/Km Crick Base Pairs Ribozyme Jr1 Constant, k Constant, k (10 M1 (5’ Froduct/3’ (Construct) Conditions (106 M4 minj (min’) (min1) (mind) min’) Product)) pH 7.5, 24°C 1 925-llt 1 mM EDTA, 5.9 0.37 0.037 0.030 5 15 (7/5) 2mM M - -*.- — 2 10-23 2mMMgC1,pH 470 <0.018 0.18 0.18 3200 17(8/8) ., . 17 o2 ‘.‘ 20400 17(8/7) Hammerhead 10 M, pH 18 0.003 1 0.01 18 (8/8) H1I1fl 12mMMgC1,pH 270 <0.01 0.16 0.13 200 14(4/6) HDV 10 mM MgCI2, pH 21 1:4 91 0006L . 10(0/6) Group II 100 mM MgCI2, 7 Intron pH7.5,500mM NA NA 0.041 0.11 5.5 58 (D5.exDl23) KCI 45°C 5 mM MgC12,pH 8 VS 8.0, 2 mM NA NA 0.7 0.6 54* 32(G1 1/Ava 5) spermidine 30 C 53 The choice of which constants in Table 1.2 are to be used in comparing nucleic acid catalysts can be a difficult choice to make. From an applied standpoint, the catalytic efficiency, kcat/K,, is an important value from which to draw comparisons, as it defines the ability of a system to carry out turnover at low concentrations of catalyst and substrate. Factors unrelated to the motif capable of chemistry can be used to emphasize kca/Km values. For example, sequence specificity is often conveyed onto nucleic acid catalysts through the introduction of guide sequences which form sequence specific Watson-Crick pairs. The conditions which favor Watson-Crick pair formation and hence substrate and product association and dissociation can be controlled by altering the number of Watson-Crick pairs formed by the guide sequence. The values ofk1, k1, and K,, can be influenced by increasing the length of the Watson-Crick double helix forming guide sequence between substrate and catalyst. Unfortunately, long guide sequences have a detrimental effect on catalysis as slow product release of highly paired Watson-Crick products becomes a problem. For nucleic acid systems which suffer from product inhibition, large values of kcat/Km, is not reflective of efficient catalysis. From a biomimetic standpoint, kcat/K,, when measured under STR conditions is not the best constant for which RNA and DNA catalysts can be compared. For those interested in the degree of chemical enhancement given by imidazole and amine enhanced DNA (such as those in the field of biomimetic chemistry) it is more appropriate to compare the rate constant of cleavage, kcat. Watson-Crick guide sequences used to convey sequence specificity onto a nucleic acid system have a lessened effect on this value. 54 In terms of the practical application of nucleic acid catalysts, the M2 independence of a system is not necessarily relevant. It can be said that the metal independent rate of RNA hydrolysis by 925-i it (kcat MTR 0.03 min’) compares reasonably well with its M2dependent counterparts (925- lit is 5” on the list in terms of multiple turnover). The addition of Mg2 to 925-i it leaves its multiple turnover rate unaffected: a result that is not reciprocated when M2 is removed from unmodified nucleic acid systems. The ability to cleave RNA in the absence of M2 represents a significant advance in terms of being able to generate sequence specific enzyme mimics. The large rate constant displayed by 925-i it has never been observed by any other nucleic acid catalyst in the absence of M2. Furthermore, the Mg2 independent multiple turnover rate seen with 925-lit suggests that future catalysts that are selected in the absence ofMg2using the modified nucleotides in 925-il will remain active even at physiological Mg2 concentrations. 55 1.4 THE KINETIC ANALYSIS OF THE CATALYST 925- lit at 13 °C. This section will focus on the rate discrepancies between the self-cleaving species 925-11 c, and the trans-cleaving species 925-i it observed in section 1.2. At 13 °C, cif cleavage by 925-i ic occurs with the largest observed M2 independent first order rate constant for any RNase A mimic previously reported. For trans-cleavage, it was not at 13 °C, but at 24 °C where the largest multiple turnover cleavage rate constant for the trans-cleaving construct 925-lit was observed (Figure 1.5). To complicate matters further, self cleavage by 925-11 c at 13 °C occurred at a rate 7.5 times faster than multiple turnover cleavage by925-lit at 24 °C (Figure 1.3). We decided to pursue the kinetic characterization of925-lit at 13 °C in order to 1) identifi the cause of the zi 1 °C temperature optima discrepancy and the origin of the 7.5 rate reduction between maximal cis cleaving activity at 13 °C and maximal trans-cleaving activity at 24 °C, and to 2) report the largest observed metal independent rate constant for an RNase A mimic capable of catalytic activity. This section will address the kinetic experiments that were carried out at 13 °C in order to rationalize the temperature and rate discrepancies observed between 925-i ic and 925-1 it. As was done at 24 °C, this kinetic dissection includes experimental values for the Michaelis constant (Km), the rate constant for Si dissociation (k1), and the rate constant for Si cleavage (kcat) as measured under single or multiple turnover conditions. 56 1.4.1 MATERIALS AND METHODS General kinetic procedures and error analyses were carried out at 13 °C as previously discussed in the materials and methods for the kinetic characterization of925-llt at 24 °C (Section 1.3). 1.4.1.1 Trans cleavage by925-llt under single turnover (STR) conditions 1.4.1.2.1 Determination of the observed single turnover rate constant, kobs STR. Cleavage reactions in trans were initiated by mixing equal volumes of925-lit at different concentrations (5 nM to 1500 nM) with trace32P-radiochemical quantities of labeled substrate Si ( 1 nM) in CB200 (50 mM TrisHCl pH 7.5, 1 mM EDTA, and 200 mM NaC1) at 13 °C and 24 °C. Aliquots were removed at various time points, quenched with two volumes of a 9:1 formamide: water, 50 mM EDTA, 0.01 % bromophenol blue, 0.01 % xylene cyanol solution, and resolved by 7 M urea denaturing 20 % 29:1 monomer:bis polyacrylamide gel electrophoresis (20 % D-PAGE). First order rate constants, k0b STR, were obtained by fitting data to the single-exponential equation 1. At 13 °C, where a biphasic rate equation described kinetic data more accurately, rate constants for biphasic curves were obtained by fitting data to the following double-exponential equation: [Pit [Pfastico( 1 t) + [Pi0j(1 _elC0 slowSTR t) 8 where the subscripts “fast” and “slow” identify the constants that describe the fast and slow phases respectively. 57 1.4.1.2.2 Determination of the maximum first—order rate constant (‘(cat sTR) and the catalyst concentration at which the rate is half-maximal (KM sTR) under single turnover conditions For each of the fast and slow cleaving phases, kcat STR and KM STR were determined by plotting k0bSTR against total925-lit concentration, [Er], using the hyperbolic equation: kcat STR [Es] k0bSTR = 4 KM STR + {E] 1.4.1.2.3 Determination of the substrate dissociation rate constant (k-i, k-i fast, and k.1 slow). Denaturing methods on the substrate, Si Pulse-chase cleavage experiments were carried out at 13 °C as described in section 1.3. 1.4.1.2.4 Determination of the substrate dissociation rate constant (k-i Si-OMe, ki fast Si-OMe, and k-i Si-OMe Non-denaturing methods on a non-hydrolysable substrate analogue Aliquots of925-lit (1000 riM) and trace radiochemical quantities of 5 ‘-32P labeled Si -OMe were equilibrated in CB200 at 13 °C overnight. The experiment was initiated by the addition of an equal volume of a chase mixture, consisting of 100 tM unlabeled substrate at 100-fold excess over the 5 ‘-32P labeled Si -OMe in CB200. Aliquots were removed at different times and added to 1110th volume of 70 % sucrose, 0.01 % bromophenol blue, and 0.01 % xylene cyanol in CB200 held in an ice bath. Time points were run one to two centimetres into a native 20 % 29:1 monomer:bis polyacrylamide gel poured in 58 CB200, with CB200 as the electrophoretic running buffer. The electrophoresis apparatus, gel, and buffers were pre-cooled to 4 °C in order to slow substrate dissociation that may occur as the complex runs through the gel. Due to the high salt concentration, the electrophoresis buffer had to be recirculated in order to prevent detrimental pH changes. In cases at 13 °C, where a biphasic rate equation described kinetic data more accurately, the first order dissociation rate constants, k1 fast SI-OMe and ki slow SI-OMe, were obtained by fitting biphasic data to the double-exponential equation: f=f° +fco!ast( 1 _efa3t SI-OMe t) +j slow( 1 —e’ slow SI-Olvie t) 9 whereffast and fco slow are the respective maximum fractions of bound substrate analogue that dissociates in the fast or slow dissociating phases andf0 is the fraction of Si -OMe that was not bound to925-lltatt=O. 1.4.1.2.5 Determination of equffibrium substrate dissociation rate constants Substrate equilibrium dissociation rate constants were carried out at 13 °C as described in section 1.3. 1.4.1.2.6 Determination of product dissociatiQn rate constants (k..3,k..4,k5, and k..6) Product dissociation constants for925-lit were measured on native gels for k1 SI-OMe using 5’- 32P-labeled P5’ or 3‘-2P-labeled P3’. These experiments were carried out with each product in the absence or presence of the other at 50 tiM. Rate constants were determined using equation 5. 59 1.4.1.2 Trans-cleavage under multiple turnover (MTR) conditions. 1.4.1.2.1 Determination of the maximum multiple turnover rate, kcat MTR, under conditions where burst phase kinetic profiles are observed For the purpose of obtaining a steady state rate constant, k0bMfl?, multiple turnover kinetic data following the burst phase were fit to the equation: P=k0bMTRt+Po 10 where P0 is the amplitude of the burst phase. Values of k0b5 MTR obtained at different catalyst concentrations were plotted versus total catalyst concentration, {E], according to the following equation: * k0b3MTR = [Ed kcat steady state MTR 11 To obtain values for the burst phase constants, multiple turnover data were fit to the equation: P = C(1 -e”’)+ k0b5MTR t 12 The initial rate of product formation was determined by linear regression of cleavage data obtained at early time points (up to 12 minutes). A first order rate constant was obtained by dividing the initial rate by the total catalyst concentration adjusted to the fraction 0.35, which 60 reflects the fraction of catalyst initially in the fast cleaving conformation at 13 °C (see results and discussion for the rationale of attributing this value). 1.4.1.3 Attempts at observing a reverse reaction (ligation) Ligation experiments were carried out at 13 °C as was discussed in section 1.3. Ligation was not observed. 61 1.4.2 RESULTS 1.4.2.1 Determination of the maximum first—order rate constant at saturating catalyst concentrations, (kcat STR), and the concentration of catalyst at which the reaction rate is half-maximal (Km STR). At 13 °C, the initial characterization of 925-1 it was carried out by attempting to determine the values of kcat STR and Kmfast STR. Unlike the kinetic profile at 24 °C, no amount of data collection would give plots which fit to the single exponential function given by equation 1 (see section 1.2). A typical attempt to fit data to single exponential equation 1 is shown in Figure 1.1 5A. Instead, data fit much better to double exponential equation 8 as shown in Figure 1.1 5B. 62 A0 C., Ll a, a, C-) C 0 C, a, I1 0.8 0.6 0.4 0.2 0.0 B 0 200 400 600 800 1000 1200 1400 1600 Time (mm) f= a’(1 exp(b*x)) 0.4 0.3 g 0.2 0 a, LI.. 0.1 0 200 400 600 800 1000 0.0 0 20 40 60 Early Time Points (mm) 1200 1400 1600 Time (mm) Figure 1.15. Comparison of identical data, fit to single and double exponential equations. Both cleavage reactions were performed with 900 nM catalyst and < mM substrate Si in buffer CB200 under single turnover cleavage reactions at 13 °C. A, top, Fitting data to a single exponential 1 gives a value of 0.020 ± 0.005 for k0bS and 0.704 for R2. B, bottom, Identical data fit to double exponential equation 2 gave an observed rate constant for the fast cleavage phase, kobsfastST of 0.21 ± 0.02 min’ (fast phase amplitude is 25 ± 0.7 %) and an observed rate constant for the slow cleavage phase, k0b5 slow STR, of 0.0026 ± 0.0002 min’ (slow phase amplitude is 53 ± 1 %). R2 was 0.996. (graph at right details the early time points enclosed in the coloured box). The biphasic fit of Si cleavage data to equation 8 implied that there were two different forms of 925-lit capable of trans-cleavage. In order to investigate the dependence of these two forms on substrate concentration, observed rate constants, k0b5 fast sm or k0b5 slow STR, were obtained at varying concentrations of catalyst for both the fast and slow phases of substrate Si cleavage. For each phase, a value of kcat STR and a value of the pseudo-Michaelis constant, KM STR, were 63 obtained by fitting k0b ft sri? or k0b3 sm to a hyperbolic plot against total catalyst 0.8 0. a a 2 0.6 0. Co ‘ 0.4 a •0 a 0.2 0.0 Figure 1.16. Effect of 925-i it concentration on the fast and slow phase rate constants and phase amplitudes in single turnover experiments at 13 °C in CB200. Observed rate constants at different catalyst concentrations, k0b5fast STR’ were obtained by fitting substrate cleavage data to double exponential equation 8. The values k0b, fast STR and k0bs slow STR at various 925-i it concentrations were fit separately to equation 4. Data from one replicate at each catalyst concentration were obtained. A, top left: Effect of925-i it concentration on kobs fast STR: the maximum first- order fast cleavage rate constant, kcatfastSTR, is 0.20 ± 0.01 min1 and the catalyst concentration at which the reaction rate is half-maximum for the fast phase, KMft STR, is i70 ± 30 nM (R2 is 0.952). B top right: Effect of 925-1 it concentration on the fast phase amplitude. The fraction of substrate cleaved through the large cleavage rate constant (not normalized for the extent of cleavage), [Pfastlw, obtained from fits to equation 2 are shown as a function of catalyst concentration. The average value of [Pfastjw is 28 %. C bottom left: Effect of 925-1 it concentration on kObSSlQ,STR: k0 slow STR is 0.0036 ± 0.0001 min’ and KM slow STR is 2i ± 4 nM (R2 is 0.892). D bottom right: Effect of 925-1 it concentration on the slow phase amplitude. The average value of [PsiøwJw is 51 %. The relative invariance of [Pfastjw and [P010w]w as a function of catalyst concentration suggests that these fractions are independent of the catalyst concentration. Catalyst concentrations ranged from 5 nM to i 500 nM. 0.8 0.6 concentration (equation 4, Figure 1.16). A B a a a a C ‘C a a 0. 1) a 0. a Ca LI 0.4 a a 0.2 0 200 400 0.005 . 1200600 800 1000 1400 Catalyst (nM) D 0 200 400 600 800 1000 1200 1400 Catalyst (nM) 0.004 0.003 0.002 0.001 0.000 . . . 0 200 400 600 800 1000 1200 Catalyst (nM) 1400 0 200 400 600 800 1000 Catalyst (nM) 1200 1400 64 For all catalyst concentrations, —35 % of the substrate is cleaved in the fast phase (Figure 1.1 6B) while —65 % of the substrate is cleaved in the slow phase (Figure 1.1 6D), this distribution was static at saturating concentrations of catalysts. The values so determined for KMfast STR and kcat fast STR are 170 ± 30 nM and 0.20 ± 0.01 min1 and for KM SlOW STR and kcatsiow are 21 ± 4 nM and 0.0036 ± 0.0001 Based on these values, the ratio of keat sn /KM STR for the more active conformation is calculated to be 1.2 .106 M’ min1. The key to addressing both the temperature and rate discrepancies seen between the cis and trans-cleaving derivatives of925-lilies in the discovery of this fast phase. Its presence suggests that a significant fraction of925-lit displays a kcat S value comparable to kseificjeave for 925-i ic. 1.4.2.2 Attempts to restore monophasic kinetic profiles In an attempt to simplify the kinetic profile of 925-i it at 13 °C , we tried to find conditions that would restore a monophasic kinetic profile.65 925-i it was subjected to various treatments74 in the hope of attaining simplified kinetic profiles which include, a) adding MgSO4, b) heat treating the catalyst followed by slow or fast cooling to 13 °C preceding kinetics, c) heating and cooling ES complexes in the presence of a Hg2 inhibitor followed by the addition of EDTA to initiate cleavage,75 d) pre-incubating the catalyst at the inactive pH of 6 followed by the addition of a large excess of pH 7.5 buffer to initiate cleavage, e) prebinding substrate to the catalyst in the absence of NaC1, followed by the addition of NaC1 to initiate the reaction (we assumed the substrate had sufficient time to bind to the guide sequences), and 1) running the 65 reaction in 10 % formamide or 1-2 M urea.63 Under all conditions, biphasic kinetic profiles persisted under single turnover conditions. Conditions were eventually found in which monophasic kinetic profiles were restored. These conditions involved the substitution of Si substrates containing a lesser number of bases at either the 5’ or the 3’ terminus (Table 1.3): Table 1.3. Structures and kinetic constants of sequences shown to restore monophasic kinetic profiles to the 925- lit system at 13 °C Substrate KMSTR (mM) kCatSTR (miii’) Monophasic Si 5’-GCGTGCCrCGTCTGTT-3’ 170±30 iiJvI 0.20 ± 0.01 min’ No — 21±4nM 0.0036±0.0001mm 3D2 5’-GCGTGCCrCGTCTG -3’ 121 ± 16 0.0069± 0.0003 Yes 5D2 5’- GTGCCrCGTCTGTT-3’ 590±70 0.038 ± 0.002 Yes With respect to Si, substrate 5D2 was truncated by two bases at the 5’ terminus whereas substrate 3D2 was truncated by two bases at the 3’ terminus. Both truncates gave monophasic single turnover kinetic profiles at 13 °C, however, the values for kcat STR obtained with these substrates were both much lower than the value of kcatjast STR observed with Si. These results suggest that 925-1 it folding appears to be especially sensitive to substrate length. Despite the persistent biphasic kinetic profiles, substrate Si was retained for continued investigation because of its heightened value of kcatfast STR. 1.4.2.3 Investigating the nature of the biphase Unable to simplify the biphasic nature of 925-i it at 13 °C without incurring a diminishment in kcat, we took steps to identify the reason for its occurrence. A large number of studies regarding 66 biphasic kinetic profiles and catalytic heterogeneity in catalytic RNA are available,62’64, 65, 76-78 however, there are very few examples reported with DNAzymes. From ribozyme precedent, reasonable explanations that could account for the observed biphasic single turnover kinetic profiles include: 1) the presence of synthetic impurities arising from the solid phase synthesis of 925-i it; 2) catalyst dimerization!aggregation;79’80 3) the existence of a significant reverse (ligation) reaction rate;78 or 4) the presence of kinetically inactive catalyst folds (conformational heterogeneity).62’8 The presence of synthetic impurities is unlikely given that; i) two independent syntheses of 925- lit and three different purifications (6 purifications in total) all resulted in nearly identical biphasic activity at 13 °C, ii) a poor catalyst synthesis could not have given rise to biphasic kinetic profiles, as biphasic profiles at i3 °C would also be observed at 24 OC,82 suggesting that the observed biphase is dependent on low temperature and not synthesis, and iii) single turnover kinetic profiles with alternate substrates 3D2 and 5D2 at 13 °C were monophasic (see supporting information page S2 to S3). Dimerization dependent activity has been observed in hammerhead ribozyme variants.79’80 It may be possible to rationalize dimer structures of 925-1 it which would give rise to hairpin dimers capable of Si cleavage activity. However, such activity must be excluded based on the observations that i) aggregated925-lit dimers in native gel analyses of 925- ii t were not present (vide infra, Figure 1.20), ii) single turnover amplitudes for both phases are invariant with catalyst concentration (vide infra, Figure i .1 6B and D), and iii) the multiple turnover steady- state Si cleavage rate increases linearly with catalyst concentration over a range of 50 nM to i iM catalyst (vide infra, Figure 1 .20B). 67 Finally, the existence of a backward reaction rate (ligation rate) could be used to explain biphasic kinetic profiles, as it has been used to explain the biphasic kinetic profiles observed with VS ribozyme derivatives.78 Because no ligation rate could be measured by 925-i it at both 13 °C and 24 °C, ligation is unlikely to be responsible for the biphasic kinetic profiles of925-lit at 13 °C. Having excluded the first three hypotheses for biphasic kinetic profiles, conformational heterogeneity remained the only candidate that could be used to explain the biphasic nature of 925-1 it. 1.4.2.4 Explaining the biphasic nature of925-llt at 13°C Having excluded synthetic impurities, dimerization, and ligation as the source of our biphasic kinetic profiles, we turn to pulse-chase substrate dissociation rate constant measurements to explain the behavior of925-lit.62 It is clear from single turnover experiments on Si that there are two 925-i lt-Sl complexes present at 13 °C (Figure i.i5, Figure i.i6). Pulse-chase experiments were carried out in order to determine if both these [ES] complexes were capable of cleaving Si. The basis for the pulse-chase experiment lies in the isolation of two labeled catalyst-substrate complexes 32P-[ES] (Figure 1.17). The formation of this complex is allowed to occur in the pulse step, whereby 32P labeled Si is incubated with catalyst. Th isolation of this complex is achieved upon addition of a large quantity of the chase species, in this case unlabeled Si. This addition halts the formation of any further 32P-labeled complex 32P-[ES]. The isolated post 68 chase 32P-[ESj complex can decay through either of two pathways: substrate cleavage (kcat) or dissociation of the substrate from the substrate-catalyst complex (k1). Because only the results of substrate cleavage (kcat) can be detected in denaturing gels, the observed rate of 32P-[ESj decay through Si cleavage will appear to proceed with a rate constant, kche, that is the sum of kcat and k.1 (Figure 1.17). If both isolated 32P-[ES] complexes are capable of cleaving substrate Si, the results of the pulse-chase experiments would be biphasic. If only one of the two isolated 32P-[ES] complexes are capable of cleaving Si, pulse-chase experiments would be monophasic.62This description is illustrated in Figure 1.17. k1 kcatst B k311EfaI + S [E S] E + P PULSE _j!_— [E ‘S] - E + PULSE 1ç E + k.1 k1 Eve + S [E *Slthac1ve Ic. [E k.1 1 CHASE CHASE k [E S] E + [E S]f E + k.1 k.1 [ *5]. [E SJ, E + Ic1’ k.1 Figure 1.17. Pulse-chase dissociation constant experiments that can give rise to monophasic or biphasic kinetic profiles depending on the nature of the slow phase. After the substrate is added to the catalyst in the initial pulse, a time, t1, is given for complex formation before chase is added. A, right, The kinetic system that would give monophasic pulse-chase kinetic profiles but biphasic single turnover kinetic profiles. When this system is chased with Si, turnover of 32P-S 1 is observed by the active complex but a biphase is not observed because the inactive complex cannot cleave 32P-Sl. B, left, The kinetic chase system observed with 925-1 it. Biphasic single turnover kinetic profiles are a result of two active complexes both capable of performing turnover but at different rates. When this system is chased, a biphasic plot is obtained because there are two distinct catalyst-substrate complexes which turnover substrate at different rates. Bold arrow indicates removal of free E following the chase. 69 1.4.2.5 Explaining the biphasic nature of 925-I It at 13 °C. Determination of the substrate dissociation rate constant. (k..1 fast, and k..1 slow). Denaturing methods on the substrate, SI The dissociation rate constant of Si was determined by pulse-chase experiments under single turnover conditions. Experimental pulse-chase data were fit to equation 8 (Figure 1.18) to obtain values for the observed decay constant kche. A B _____________________________________________________________________ 0.07 __ / 0.O ______________ ______________ 0 200 400 600 800 1000 1200 1400 1600 0 4l3 600 800 1OX) 1200 1400 lime (mm) •lime (non) Figure 1.18. Determination of substrate dissociation rate constants for substrate Si by pulse-chase cleavage assay with 925-i it in CB200 13°C. A left: Control reaction, (.) 900 nM Catalyst was incubated with <1 nM Si, or Chase reaction, (0) 1 i2 nM catalyst and <1 nM Si were challenged with 50 jiM of excess unlabeled Si at 3 mm. Data for the unchased reaction were fit to equation 8 giving 0.21 mm4 for kObS fast STR and 0.0026 min’ for STR (R is 0,996). Data for chased reaction were fit to equation 8 giving 0.233 nun for kchase fast and 0.0041 nun for kcho.se slow (R2 is 0.981). These values were treated according to equation 5 (see section 1.3) to give the substrate dissociation rate constants, 11 and k,1 slowS B right: rescaling the fraction cleaved axis for the chased reaction clearly shows biphasic behavior. Dotted lines indicate the constituent curves for the fast and slow phases. Unlike the Hairpin ribozyme system, the substrate dissociation rate of Si from 925-lit was biphasic. Two observed first order rate constants were measured from the biphasic kinetic profiles observed in the pulse-chase experiment: kchasejtst = 0.23 mm4 and kciiase slow = 0.004 1 mind. Using equation 5 (see section 1.3) and the respective values of k01 STR obtained in the control reactions, we calculated two values for substrate SI dissociation constants: kJf1 is 0.03 mm’1 and k1 slow is 0.0005 mm4. The subtraction of kcat fast from kobs chase slow gave irrational values for Jci making it easy to determine which phase in the biphasic pulse-chase experiment 70 belonged to the respective phase in single turnover kinetic profiles. The biphasic nature of these pulse-chase experiments, suggests that both 925-i it ES complexes can lead to substrate cleavage. 1.4.2.6 Explaining the biphasic nature of925-llt at 13 °C. Determination of the substrate dissociation rate constant (k..j SI-OMe fast and k.1 SI-OMe Non denaturing methods on a non-hydrolysable substrate analogue To corroborate the dissociation constants and to verify that only two major ES complexes exist between substrate Si and 925-llt, pulse-chase native gel-shift assays were performed with a non-hydrolyzable substrate analog, Si -OMe. These data were also biphasic and rate constants could be calculated from equation 9 (Figure 1.19): 1.0 0.8 0.6 0 0 a C 0.4 0 0.2 0.0 Time (mm) 1000 1200 1400 Figure 1.19. Pulse-chase analysis by native gel electrophoresis to determine the first order rate constant for substrate analogue dissociation from 925- lit at 13°C, A trace of 5‘2P labeled S 1 -OMe was incubated with 925-1 it (ljiM) in CB200. Following the addition of an equal volume of the chase (lOOi.tM substrate in CB200), aliquots were taken at various times and loaded onto native PAGE gels. The fraction of substrate analogue dissociated was plotted versus time for multiple experimental trials. The fraction of Sl -OMe dissociated was plotted versus time. Data from at least two experimental replicates were obtained. The data were fit to a double exponential equation 5, which returned the following dissociation rate constants and phase amplitudes: kJf S1-OM€ = 0.0 14 ± 0.004 min1, slow SJ-OMe 0.0023 ± 0.0005 mm’1,amplitude fast = 27 ± 7 %, amplitude slow = 46 ± 6 %. Dotted lines indicate the constituent curves for the fast and slow phases. (Figure reproduced courtesy of Jay Thomas) 0 200 400 600 800 71 The two dissociation constants were 1(j fact Si-OMe , which was 0.0 14 ± 0.004 min1 and had an 0.003 0.002 amplitude of 26 ± 7 %, and k4 SI-OMe, which was 0.0023 ± 0.0005 min’ and had a slow phase amplitude of 46 ± 6 %. Despite the structural difference imparted by the 2’OMe substitution, these dissociation constants are consistent with those observed for substrate Si in the kinetic pulse-chase experiments (Figure 1.18). Likewise, the amplitudes are in good agreement with those observed in single turnover cleavage experiments (Figure 1.1 6B). 1.4.2.7 Multiple turnover cleavage of SI by925-Ilt at 13°C Multiple turnover experiments were conducted at 13 °C with a saturating concentration of substrate (Si) and various concentrations of catalyst. The non-linearity of substrate cleavage at early times is clear evidence that a burst phase is present (Figure 1 .20A). A B 3.0 2.5 2.0 1.5 I.e Cl, 0.5 0.004 zz’zz 0.000 500 1000 1500 Time (mm) 0.0 0 2000 0 200 400 600 800 1000 1200 Catalyst (nM) Figure 1.20. A, left, Multiple Turnover Analyses on the cleavage of substrate Si at 13 °C. Cleavage reactions were performed with varying concentrations of catalyst, 7.5 pM unlabeled substrate Si, and <1 nM 5’ labeled substrate Si in buffer CB200 at 13 °C. (•) 1000 nM (0) 500 nM, (Lx) 200 nM, (v) 100 nM, (0) 50 nM Catalyst. Data were fit to equation 12 (C(1expt )+kteady state t). The average R2 is greater than 0.996 except for data obtained at 50 nM catalyst which is 0.968. Cleavage was allowed to proceed until a maximum of 20 % of the substrate was cleaved. Reactions were quenched in formamide at the times indicated on the graph. B, right, Rate constant determination for the steady state rate constant. For all catalyst concentrations, linear regression analyses were performed on multiple turnover substrate cleavage for time points after 255 mm. This rate was plot against catalyst concentration to obtain a steady state rate constant. Data from one replicate at each catalyst concentration were obtained. The intercept for this value is 0.0000 ± 0.0001 pM mm4.The steady state rate constant is 0.0038 ± 0.0002 min1.R2 is 0.994. 72 When the steady state multiple turnover rate constant is plotted against enzyme concentration (see equations ii and 12), the value of keat steaa’y state ivrr was found to be 0.0077 ± 0.0007 miii’ (Figure 1 .20B). This value is within a factor of 2 of the value for keat slow STR (0.003 8 ± 0.0002 miii’) observed in the single turnover experiments with Si (Figure 1.16), suggesting that the steady state rate represents turnover by a catalyst population that exists largely as the slow cleaving conformation. Addressing the burst phase was much more difficult. The constant, “C”, in equation 12 represents the fraction of catalyst associated with the burst phase, while the burst rate constant, kburst, represents the rate constant that describes the decay of the burst phase. The rate constant kburst was estimated by the method of initial rates.83 This estimation required that we assume: 1) that the slow cleaving conformation of the catalyst contributes negligibly to Si cleavage at early time points (equation 12 , P = C( 1 ett)+ k0bs MTR t, becomes P = C( 1 ett) only at early time points) and 2) that the initial rate of product formation in the burst phase reflects the entire amount of 925-1 it that is present as the fast cleaving conformation in single turnover experiments (0=0.35). In Figure 1.16 this value was estimated to be 0.35 % of the catalyst59 and is independent of 925-1 it concentration (Figure 1.16B), From the 1 mM plot (Figure 1.20, (•)) the initial first order rate constant, kburst, is estimated to be 0.10 ± 0.03 miii’, which is within a factor of 2 to the value of the first order rate constant (kcat fast STR 0.2 ± 0.01 miii’) extracted for the fast phase under the single turnover conditions. The contention that conformational heterogeneity gives rise to the kinetic profiles defined by equation 12 is contingent upon data which suggest that the observed burst phase kinetic profile in Figure 1.20 does not arise from product inhibition. Since slow product release is commonly 73 invoked to explain burst-phase kinetic profiles and slow product release would be consistent with a reduction in temperature, slow product release resulting in burst phase kinetic profiles was investigated and ultimately excluded. 1.4.2.8 Slow product release does not account for the burst phase Three orthogonal experiments were carried out in order to rule out product inhibition as the source of the burst phase in multiple turnover kinetic profiles at 13 °C. Initially, we attempted to measure dissociation rate constants for both P5 and P3’ by attempting to resolve 925-1 it-product complexes on non-denaturing polyacrylamide gels. As was previously attempted at 24 °C, the rate of product dissociation at 13 °C was too fast to be effectively measured by native PAGE assays. These observations suggest that the product dissociation rate constants for P5’ and P3 are much larger than either STR or kcat slow STR• The possibility of product inhibition was also evaluated under conditions designed to inhibit the burst phase observed in multiple turnover kinetic profiles. By incubating 1 iM catalyst with one or both cleavage products at 50 iiM prior to the addition of 32P labeled substrate Si at 15 jiM, we attempted to form a 925-1 it-product complex in situ prior to the initiation of kinetic experiment. Incubating this solution prior to substrate addition produced no effect on the burst phase. Again, the lack of a result suggests that product inhibition is unlikely to affect multiple turnover at 13 °C. 74 1.4.2.9 Native PAGE multiple turnover analysis provides evidence against slow product release Additional evidence showing that slow product release does not contribute to the burst phase in multiple turnover experiments has been observed in kinetic experiments carried out in native PAGE gels. Were the steady state rate constant, kcat steady state Ml’]?,, to reflect slow product release rather than chemistry, then kcat steady state MTR would closely represent a product dissociation constant. As kcat steady state MTR would represent a half-life of 0.0077 ± 0.0007 miii , the half-life for either the EP5 or the EP3 complex, would be -90 minutes and would be observable by native PAGE gel retardation experiments. In an attempt to observe product inhibition in native PAGE gels, multiple turnover reactions were initiated at different times at 13 °C, stopped by placing on ice, and immediately analyzed by native PAGE at 4 °C. The experiment was performed for both 3’ and 5 ‘-labeled Si. Multiple turnover cleavage of 5’ labeled Si was monitored by both native and denaturing PAGE as shown in Figure i.21A and B respectively. The presence of 3’ or 5’ product-925-lit complexes were not observed. 75 j\ 10 25 35 50 65 75 90 120 240 300 Time (mm) - 4 E•Si complex 4 L_ E•P5 complex 1i.t Si * P5 1 2 3 4 5 6 7 8 9 10 ii 12 13 14 13 10 25 35 50 65 75 90 120 240 300 4 Time (mm) ——————— I— Si ø—P5,. —bd = 1 2 3 4 56 7 8 910 C : Substrate — cleaved 2.5 (M) 2 1.5 0.5- —______ 0 —.-— denaturing 0 100 200 300 400 —4-— native Time (mm) Figure 1.21. Multiple turnover cleavage of substrate Si by 925-i it at 13 °C under native PAGE conditions. A top.. native PAGE analysis of the time course for multiple turnover at 1 1iM catalyst and 15 jtM Si in CB200 at 13°C (lanes 1-10). Control lanes (11-14) reveal electrophoretic motilities of: uncleaved Si (lane ii), E•S 1 complex (lane 12) , P5 (lane 13), and EP5 complex (lane 14). The reaction lanes show a small, steady state amount of ES 1, and no observable E•P5. B middle: The same time course (lanes 1-10) for the same reaction as in A run on a denaturing PAGE gel. C bottom: Plots of substrate cleaved vs. time for both the native (blue trace) and denaturing gels (green trace). Production of unbound product as observed in the native gel matches that observed by denaturing gel, indicating that the burst phase is not due to slow release of the 5’ product. (Figure reproduced courtesy of Jay Thomas) Plots of the appearance of product over time as observed by native and denaturing PAGE both show a burst phase and overlap closely (see Figure 1.21C). The good correlation between 76 denaturing and non denaturing kinetic data (Figure 1.21 C) suggests that the 3’ or 5’ Product- 925-lit complexes dissociate from the catalyst too quickly to be observed. This same result was obtained with the 3 ‘-labeled substrate (data not shown). In order to confirm that this native PAGE technique can be used to detect slow product release at a rate comparable to the steady state rate of925-i it, Jason Thomas applied this technique to a hammerhead ribozyme variant known to be inhibited by slow product release, HH16,84 where burst phase kinetic profiles were observed in denaturing gels under multiple turnover conditions (data not shown). 1.4.2.10 Attempts at measuring a ligation rate for925-llt It was important to investigate the possibility of ligation by925-lit. The reaction generating the two product oligonucleotides involves transphosphorylative cleavage of the RNA phosphodiester bond. This reaction is independent of water, thus the ligation of the 3’ and 5’ product oligonucleotides generated from the cleavage of Si could represent a significant reverse rate that is not discouraged by the large concentration of 55 M water. Had a ligation rate been observed by 925-i it on the products of Si substrate cleavage, the kinetic scheme describing 925-i it at 13 °C would have been considerably more complicated. No such ligation was observed. Attempts were made at both 13 °C and 24 °C to observe a reverse reaction for 925-lit in single turnover ligation experiments. Catalyst was incubated with either; trace amounts of 5’- 32P labeled P5 and an excess of unlabeled P3’, or trace amounts of 3’- 32P labeled P3’ and an excess of unlabeled P5’. In all cases ligation in the form of substrate Si formation was not observed. Attempts were also made to observe ligation through a shift in the external equilibrium of the Si cleavage reaction. The effect of adding excess 3 ‘-product to single turnover cleavage 77 reactions was investigated in an effort to shift the external equilibrium between Si and its cleavage products. 5 ‘-32P labeled Si was incubated with saturating amounts of 925-i it for 50 half-lives. The reaction was chased with unlabeled P3 in CB200 in order to favor ligation. This reaction was compared to a control where unlabeled P3’ was not added. Data for all experiments were collected over a 5 to 2100 mm period (post chase). We were unsuccessful in measuring a rate of ligation for925-i it in all ligation experiments attempted. 1.4.2.11 Temperature cycling enhances the rate of multiple turnover and demonstrates catalyst stability The largest reported rate for Si cleavage by 925-i it occurs under single turnover conditions at i 3 °C (Figure 1.3), while the largest reported value for the steady state multiple turnover cleavage rate occurs at 24 °C (Figure i .5). It makes sense that a thermally cycled regime be used to exploit constants at both temperatures to obtain a higher multiple turnover rate. By cycling the temperature (i 5 sec i 3 °C, 7 mm at 24 °C, and i 5 sec at 95 °C), we obtained an apparent first order rate constant k0b theocyc1e MTR, of 0.064 ± 0.002 min1 (Figure 1.22). 78 2.0 1.5 a, a’ 0 a, 1.0 a, C’) C’) 0.5 0.0 Number of Cycles (-9 mm / cycle) Figure 1.22. The effect of temperature cycling on the multiple turnover rate of Si substrate cleavage by 925-i it at 13 °C. 100 nM catalyst, 15 .tM unlabeled Si, and <1 nM 5’ labeled Si was incubated in buffer CB200 at i3 °C. The reaction was placed into a PCR temperature cycler programmed as follows: 13 °C for 15 s, 22 °C for 7 mm, and 95 °C for 15 sec. With temperature ramping time included, the time for one cycle was —9 mm. Aliquots were removed for analysis after every five cycles. Data were fit to a linear equation: k0b5 thermoc1e MTR is 0.58 ± 0.02 cycle1 or 0.064 ± 0.002 mm4 for the 9 mm cycle (R2 is 0.996). Data from one replicate at every five cycles were obtained. The linearity of the graph indicates that the catalyst is not deactivated upon heating. A control reaction done with only labeled substrate showed a negligible amount of cleavage during the duration of temperature program. The value ofk0b thermocycle MTR represents a multiple turnover rate constant that represents a 20 fold improvement over the steady state multiple turnover rate at 13 °C, and a modest improvement over the previously reported steady state rate at 24 °C (0.030 ± 0.002). Presumably the mechanism for this increase is through the restoration of the larger burst phase rate at 13 °C with each thermal cycle. In terms of precise characterization, k0b thermoc1e MTR, is a less meaningful rate constant that comprises all the reaction steps during the temperature cycling. However, the ability to extract a higher rate constant using thermocycling allows 925-i it to overcome slow multiple turnover seen at both 13 °C and 24 °C. Finally, the linearity maintained in the plot of Si cleavage vs. time by 925-i it following 17 turnovers, demonstrates that the synthetic modifications are 0 10 20 30 40 79 thermally stable over a wide temperature range and underscored the resilience of the modified nucleic acid catalyst and its chemical modifications. 1.4.3 DISCUSSION 1.4.3.1 The kinetic model for925-llt at 13°C We carried out kinetic experiments on 925-i it at 13 °C in order to ask whether a value similar to the single turnover rate constant for self cleavage, kseijcieave, could be reciprocated on a species capable of multiple turnover. The kinetic rate constants measured for 925-iit at 13 °C have been compiled in Table i .4. 80 Table 1.4. Rate constants measured at 13 °C for 925-i it. ‘Where no “slow” rate is reported, the reaction is monophasic. Observed Rate Method of AtSubstrateConstant for Determination 13 °C 925-llc kselfcleave - Single turnover 0.28 ± 0.02 min’ 925-llt STR kcatfastS Si Single turnover 0.20 ± 0.01 min’ kcatsiow Si Single turnover 0.0036 ± 0.000 1 min’ KMfOStSTR Si Single turnover 170 ± 30 nM KM slow STR Si Single turnover 21 ± 4 nM Pulse-chase single .0.233 mmkchase fast Si turnover Pulse-chase single . kchase slow Si 0.0041 mmturnover Pulse-chase single .kijast Si turnover 0.03 mm Pulse-chase singlek1 slow Si 0.0005 min1turnover k-1fast SI-OMe Si-OMe Pulse-chase/native PAGE 0.0 14 ± 0.004 min’ kislowSiOMe Si-OMe Pulse-chase/native PAGE 0.0023 ± 0.0005 min’ kcat5D2STR 5D2 Single turnover 0.038 ± 0.002 min’ KM 5D2STR 5D2 Single turnover 590 ± 70 nM Pulse-chase single kchasejD2 5D2 0.066 ± 0.002 mimi’turnover Pulse-chase single “-15D2 5D2 0.O36min’turnover kcat3D2STR 3D2 Single turnover 0.0069 ± 0.0003 mimi’ KM 3D2STR 3D2 Single turnover 121 ± 16 nM 9-11tMTR kcat steady state MTR Si Multiple turnover 0.0077 ± 00007i3 kcatMTR 5D2 5D2 Multiple turnover 0.0 193 ± 0.0006 mimi’ kcatMTR 3D2 3D2 Multiple turnover 0.0041 ± 0.0003 mimi’ k0b5 thermocycleMTR Si Multiple turnover 0.064 ± 0.002 min’ Despite complicated biphasic kinetic profiles, we identified a large cleavage rate constant for 925-i it at 13 °C (kcat fast STR = 0.2 miii’, Figure 1.16). This value compares favorably with the 81 rate constant for self-cleavage by its cis-cleaving counterpart,925-lie, at 13 °C (kseijcieave, of 0.28 ± 0.02 min’) (Figure 1.3). These rate constants represent the highest values reported for Mtindependent RNA hydrolysis for both RNA trans-cleavage and self-cleavage to date by a biomimetic system at pH 7.4 and at physiological ionic strength. These rate constants rival those displayed by most combinatorially selected metal dependent DNAzymes and naturally occurring ribozymes (Table 1.2). The cleavage rate constants of the fast phases observed in 13 °C biphasic kinetic profiles are similar in value to 925-lic at 13 °C suggesting that kcat STR is optimal at 13 °C for both cis cleaving and trans-cleaving systems. In terms of multiple turnover, this elevated cleavage rate constant is observed in a burst phase. The fact that burst phase kinetics complicate the kinetic profiles of925-lit at 13 °C also explain why the temperature optimum for steady state multiple turnover is at 24 °C and not 13 °C. 1.4.3.1.1 The kinetic model for Si cleavage by925-iit at 13 °C To transform the kinetic constants reported in Table 1.4 into a 13 °C kinetic model for 925-i it the following key observations were required: 1) Single turnover kinetic profiles were biphasic (Figure 1.16) and two dissociation rate constants were observed in the dissociation of Si and Si-OMe from 925-i it at 13 °C (Figure 1.18 and Figure 1.19 respectively). These data indicate that two conformations of the ES complex are formed and that both can lead to substrate cleavage.62 82 2) Burst phase kinetics observed under multiple turnover conditions could not be attributed to irreversible catalyst inactivation (ruled out through multiple turnover temperature cycling experiments Figure 1.22) and could not be attributed to slow product release (Figure 1.21). 3) The distribution of the amount of substrate cleaved by the slow cleaving phase and the fast cleaving phase remain constant despite different catalyst concentrations (Figure 1.1 6B and D). One scheme that fits these applications has been proposed by Esteban et a!. but has never been observed.62 This scheme, shown in Figure 1.23, accounts for both biphasic single turnover kinetics and burst phase multiple turnover kinetics that does not occur as a result of slow product release or catalyst inactivation. This scheme requires that the catalyst is pre-partitioned into species that are capable of binding and cleaving substrates in separate fast cleaving and slow cleaving complexes. In this scheme, burst phases observed in multiple turnover experiments were observed with successive turnover due to the accumulation of catalyst in the less active, substrate-bound catalyst form after multiple turnovers. Furthermore, the slow cleaving, substrate-bound catalyst must possess a substrate dissociation rate constant that is less than or comparable to the value Of kcatsiow. k1 fast heat fast Efast + S ESfast Efast k_1 fast fast P slow keat slow E510 + S - ES510 - E1 k1 slow P Figure 1.23. The minimal kinetic scheme for cleavage of substrate Si by925-lit at 13 °C in CB200 (50 mM Tris HC1 pH 7.5, 1 mM EDTA, and 200 mlvi NaC1). E, Catalyst 925-1 it, S, Substrate Si. 83 The scheme in Figure 1.23 is consistent with all the experimental data observed with 925-i it at 13 °C. The biphasic nature of direct Si cleavage by 925-1 it at 13 °C (Figure 1.15) and substrate dissociation experiments measured on two different substrates from 925-lit (Si: Figure 1.18 and Si OMe Figure 1.19) verify that there are two species present that are capable of cleaving Si (ESi0 and ESfast). Figure 1.1 6B and D suggests that the fraction of925-lit catalyzed cleavage of Si is distributed in a ratio of 35 % to 65 % of ESft and ES510 respectively. This distribution ratio is independent of the substrate-bound catalyst concentration, which suggests that free catalyst in the Ei0 and Efast forms are in rapid equilibrium. The slow cleavage (kcat STR slow .4\ . ..0.003 6 ± 0.0001 mm fi and dissociation (k1 0.0005 mm rate constants pertammng to the decomposition of the ES510 complex are both less than the rate limiting steady state multiple turnover rate constant (kcatsteadysteMTh 0.0077 ± 0.0007 min1)in multiple turnover assays. This suggests that a thermally reversible accumulation of ES5i0under multiple turnover conditions may account for the burst phase. There are alternate kinetic schemes that can be proposed to rationalize the experimental data seen in the kinetic profiles of 925-i it at 13 °C. The description of these data transcend the scope of this thesis and can be found in published literature.59 This literature can be summarized by stating that925-lit kinetic data at 13 °C fits best to the scheme shown in Figure 1.23. At 13 °C,925-lit displays the largest reported rate constant for a trans-cleaving RNase A mimic in an M2 independent RNA hydrolysis context (kcat fast 0.2 min1). The agreement between this rate and the rate of the cis-cleaving species925-ilc at 13 °C (kseijcieave 0.28 min1)suggests that the replication of the active site of925-lie is accurately reflected in the solid phase 84 synthesis of 925-1 it. Furthermore, the complicated kinetic profiles observed with 925-1 it at 13 °C suggests that complicated biphasic kinetic profiles may foreshadow future selections for fast acting modified DNAzymes. 1.4.3.1.2 Attempts at Linear Multiple Turnover. Kinetics with Different substrates In an attempt to restore monophasic kinetic profiles to 925-i it such that simplified kinetic measurements, high values Of kcat, and simplified multiple turnover activity could be reported at 13 °C, 925-i it’s activity on different substrates were tested (Table 1.3). The restoration of monophasic kinetic profiles with 5D2 and 3D2 came at the expense of a high kcatfast which lead us to hypothesize that the two bases at the 3’ and 5’ ends of Si are required for the large value of kcat seen with 925-i lt on Si at 13 °C. 1.4.3.2 Exploiting the different temperature optima. Temperature cycling Temperature cycling has long been utilized in PCR to overcome burst phase kinetics attributed to slow product release. The temperature cycling of925-lit gives both a linear response in plots of Si cleavage vs. time and a higher turnover rate than previously achieved at any other single temperature. These data are reconciled with the scheme shown in Figure 1.23 through the hypothesis that quick multiple turnover is obtained through the thermal denaturing of the partition of catalyst that accumulates in the slow cleaving ESi0 complex allowing for the free 925-lit to rebind substrate into a fast cleaving ESfast form. This thermal denaturation process is repeated before 925-i it has an opportunity to accumulate in the ES10 form. Through cycling conditions, the multiple turnover rate for Si cleavage by925-lit is increased to 0.06 min’, 200 % larger than the multiple turnover rate constant reported at 24°C (Figure 1.21). 85 This experiment is used to forward the concept that thermal cycling can be used to extract larger rates from DNA based catalysts, furthermore, the linearity of this experiment suggests that 925-i it is robust under thermal conditions even after being modified to become chemically more active. This degree of stability displayed by 925-lit is not necessarily observed in enzymes and can be used to conclude that future selected modified nucleic acid catalyst displaying low keatS values may make up for their lowered rates with a heightened catalyst stability. 86 15 CONCLUSIONS In this chapter, we demonstrate how synthetic organic chemistry can be used to complement combinatorial selections for the production of biomimetic catalysts. The DNAzyme 925-lit, uses protein-like modifications to catalyze sequence-specific RNA hydrolysis under multiple turnover conditions in the absence of M2. The techniques demonstrated in both the selection and the solid phase production of this RNase A mimic represent an advance in terms of the synthesis of high molecular weight biomimetic catalysts of defined structure. The modified nucleic acid requirement and the ability to act in the absence ofM2 distinguishes 925-1 it from other nucleic acid systems in terms of a novel structure (and possibly a novel mechanism) required for nucleic acid mediated RNA cleavage. Multiple turnover RNA cleavage in the absence of M2 underscores the potential for RNA cleavage at physiological concentrations where the availability of divalent metal cations is limited. These two observations warrant the kinetic study of925-lit carried out in sections 1.3 and 1.4. The study of925-lit at 24 °C reveals that the multiple turnover rate constant displayed by 925- lit at 24 °C (kcat MTR = 0.030 ± 0.002 mm4, kcat/KM = 5.3 x 10) is reflective of the largest multiple turnover rate constant for a nucleic acid RNase A mimic capable of catalytic M2- independent RNA cleavage at physiological pH and ionic strength. The study of925-lit at 13 °C, identifies a large cleavage rate constant for 925 lit (keat fast STR = 0.2 min’) that represents the largest first order rate constant reported forM2-independent RNA hydrolysis by a trans-cleaving system at pH 7.4 and at physiological ionic strength. Most importantly the agreement between this rate and the rate of the cis-cleaving species 925-il c at 87 13 °C (kseijcieave 0.3 min’) implies that the active site of 925-11 c may be represented in the solid phase synthesis of925-1 it. This kinetic study can be used as precedent for the characterization of newly selected cis- and trans-RNA cleaving motifs, it serves as a benchmark for examining the limitations of unmodified nucleic acid catalysts, and it highlights the possible advantages and disadvantages (complicated kinetic profiles) imparted by the use of modified nucleotides in selections. This study may prove pertinent to such DNAzyme systems such as that selected by Sidorov et a?., who have demonstrated self-cleavage of an all-RNA target with modified DNA from similarly modified imidazole containing triphosphates, but have not yet reproduced RNA cleavage in a multiple turnover context.56 In another instance, these kinetic analyses may prove valuable to a recent report that has described biphasic kinetic profiles under single-turnover conditions at 10 mIVI Mg2 for a 10-23 sequence that targeted mRNA relating to the vanilloid receptor subtype.85 Regarding new selections, the identification of larger rates at lower temperatures in the case of 925-lie and 925-llt would suggest that higher rate constants at physiologically relevant temperatures may be realized in future selections that are carried out at higher temperatures. The original selection of925-ilc was carried out on the bench-top where temperatures were speculated to be as low as 13 °C. By maintaining physiological temperatures throughout new combinatorial selections, it is hypothesized that the limitations of biphasic kinetics and a low temperature optimum may be overcome. This study can serve as a guide for new selections carried out with the triphosphates shown in Figure 1.2. 88 1.6 REFERENCES 1. Breslow, R., Biomimetic chemistry and artificial enzymes - catalysis by design. Ace. Chem. Res. 1995, 28, 146. 2. Raines, R. T., Ribonuclease A. Chem. Rev. 1998, 98, 1045. 3. Rowan, S. J.; Sanders, J. K., Enzyme models: design and selection. Curr. Opin. Chem. Biol. 1997, 1, 483. 4. Kirby, A. 3., Enzyme mechanisms, models, and mimics. Angew. Chem. mt. Edit. 1996, 35, 707. 5. Hirs, C. H.; Moore, S.; Stein, W. H., The sequence of the amino acid residues in performic acid-oxidized ribonuclease. J. Biol. Chem. 1960, 235, 633. 6. Kartha, G.; Bello, J.; Harker, D., Tertiary structure of ribonuclease. Nature. 1967, 213, 862. 7. Anfinsen, C. B., Principles that govern folding of protein chains. Science. 1973, 181, 223. 8. Moore, S.; Stein, W. H., Chemical Structures of Pancreatic Ribonuclease and Deoxyribonuclease. Science. 1973, 180, 458. 9. Gutte, B.; Merrifield, R. B., The total synthesis of an enzyme with ribonuclease A activity. J. Am. Chem. Soc. 1969, 91, 501. 10. Voet, D.; Voet, J. G., Biochemistiy. 2nd ed. ed.; John Wiley & Sons, Inc.: New York, 1995. 11. Anslyn, E.; Breslow, R., Geometric evidence on the ribonuclease model mechanism. J. Am. Chem. Soc. 1989, 111, 5972. 12. Lorente, A.; Espinosa, J. F.; Fernandez Saiz, M.; Lehn, J. M.; Wilson, W. D.; Zhong, Y. Y., Syntheses of imidazole-acridine conjugates as ribonuclease A mimics. Tet. Lett. 1996, 37, 4417. 13. Fouace, S.; Gaudin, C.; Picard, S.; Corvaisier, S.; Renault, J.; Carboni, B.; Felden, B., Polyamine derivatives as selective RNaseA mimics. Nuci. Acids Res. 2004, 32, 151. 14. Birdsall, D. L.; McPherson, A., Crystal structure disposition of thymidylic acid tetramer in complex with ribonuclease A. J. Biol. Chem. 1992, 267, 22230. 15. Beloglazova, N. G.; Fabani, M. M.; Zenkova, M. A.; Bichenkova, E. V.; Polushin, N. N.; SiPnikov, V. V.; Douglas, K. T.; Vlassov, V. V., Sequence-specific artificial ribonucleases. I. Bis-imidazole-containing oligonucleotide conjugates prepared using precursor-based strategy. Nuc. Acids Res. 2004, 32, 3887. 16. Jencks, W. P., Binding energy, and enzyrnic catalysis: the circe effect. Adv. Enzymol. Relat. Areas Mo!. Biol. 1975, 43, 219—410 17. Dass, C. R., Deoxyribozymes: cleaving a path to clinical trials. Trends Pharmacol. Sci. 2004, 25, 395. 18. Kaukinen, U.; Lonnberg, H.; Perakyla, M., Stabilisation of the transition state of phosphodiester bond cleavage within linear single-stranded oligoribonucleotides. Org. Biomol. Chem. 2004, 2, 66. 19. Green, R.; Ellington, A. D.; Szostak, J. W., In vitro genetic analysis of the Tetrahyrnena self-splicing intron. Nature. 1990, 347, 406. 20. Breaker, R. R.; Joyce, G. F., A DNA enzyme that cleaves RNA. Chem. & Biol. 1994, 1, 223. 89 21. Tuerk, C.; Gold, L., Systematic evolution of ligands by exponential enrichment - RNA ligands to bacteriophage-T4 DNA-polymerase. Science. 1990, 249, 505. 22. Crick, F. H., The origin of the genetic code. J. Mol. Biol. 1968, 38, 367. 23. Kruger, K.; Grabowski, P. J.; Zaug, A. 3.; Sands, 3.; Gottschling, D. E.; Cech, T. R., Self-splicing RNA: autoexcision and autocyclization of the ribosomal RNA intervening sequence of Tetrahymena. Cell. 1982, 31, 147. 24, Guerrier-Takada, C.; Gardiner, K.; Marsh, T.; Pace, N.; Altman, S., The RNA moiety of ribonuclease P is the catalytic subunit of the enzyme. Cell. 1983, 35, 849. 25. Collins, C. A.; Guthrie, C., The question remains: is the spliceosome a ribozyme? Nat. Struct. Biol. 2000, 7, 850. 26. Robertson, D. L.; Joyce, G. F., Selection in vitro of an RNA enzyme that specifically cleaves single-stranded DNA. Nature. 1990, 344, 467. 27. Pan, T.; Uhlenbeck, 0. C., In vitro selection of RNAs that undergo autolytic cleavage with Pb2+. Biochemistry. 1992, 31, 3887. 28. Lorsch, J. R.; Szostak, 3. W., In vitro evolution of new ribozymes with polynucleotide kinase activity. Nature. 1994, 371, 31. 29. Lohse, P. A.; Szostak, J. W., Ribozyme-catalysed amino-acid transfer reactions. Nature. 1996, 381, 442. 30. Illangasekare, M.; Sanchez, G.; Nickles, T.; Yams, M., Aminoacyl-RNA synthesis catalyzed by an RNA. Science. 1995, 267, 643. 31. Wilson, C.; Szostak, J. W., In vitro evolution of a self-alkylating ribozyme. Nature. 1995, 374, 777. 32. Wecker, M.; Smith, D.; Gold, L., In vitro selection of a novel catalytic RNA: characterization of a sulfur alkylation reaction and interaction with a small peptide. RNA. 1996, 2, 982. 33. Prudent, J. R.; Uno, T.; Schultz, P. G., Expanding the scope of RNA catalysis. Science. 1994, 264, 1924. 34. Seelig, B.; Jaschke, A., A small catalytic RNA motif with Diels-Alderase activity. Chem. &Biol. 1999,6, 167. 35. Jensen, K. B.; Atkinson, B. L.; Willis, M. C.; Koch, T. H.; Gold, L., Using in vitro selection to direct the covalent attachment of human immunodeficiency virus type 1 Rev protein to high-affinity RNA ligands. Proc. Nati. Acad. Sci. USA. 1995, 92, 12220. 36. Tarasow, T. M.; Tarasow, S. L.; Eaton, B. E., RNA-catalysed carbon-carbon bond formation. Nature. 1997, 389, 54. 37. Breaker, R. R.; Joyce, G. F., A DNA Enzyme with Mg2+-Dependent RNA Phosphoesterase Activity. Chem. & Biol. 1995, 2, 655. 38. Chinnapen, D. 3. F.; Sen, D., A deoxyribozyme that harnesses light to repair thymine dimers in DNA. Proc. Nati. Acad. Sci. USA. 2004, 101, 65. 39. Sheppard, T. L.; Ordoukhanian, P.; Joyce, G. F., A DNA enzyme with N-glycosylase activity. Proc. Nati. Acad. Sci. USA. 2000, 97, 7802. 40. Li, Y. F.; Sen, D., A catalytic DNA for porphyrin metallation. Nat. Struct. Biol. 1996, 3, 743. 41. Li, Y. F.; Liu, Y.; Breaker, R. R., Capping DNA with DNA. Biochemistry. 2000, 39, 3106. 42. Carmi, N.; Shultz, L. A.; Breaker, R. R., In vitro selection of self-cleaving DNAs. Chem. &Biol. 1996,3, 1039. 43. Santoro, S. W.; Joyce, G. F., A general purpose RNA-cleaving DNA enzyme. Proc. Natl. Acad. Sci. USA. 1997, 94, 4262. 90 44. Geyer, C. R.; Sen, D., Evidence for the metal-cofactor independence of an RNA phosphodiester-cleaving DNA enzyme. Chem. & Biol. 1997, 4, 579. 45. Faulhammer, D.; Famulok, M., Characterization and divalent metal-ion dependence of in vitro selected deoxyribozymes which cleave DNA!RNA chimeric oligonucleotides. J. Mo!. Biol. 1997, 269, 188. 46. Carrigan, M. A.; Ricardo, A.; Ang, D. N.; Benner, S. A., Quantitative analysis of a RNA-cleaving DNA catalyst obtained via in vitro selection. Biochemistiy. 2004, 43, 11446. 47. Murphy, E.; Steenbergen, C.; Levy, L. A.; Raju, B.; London, R. E., Cytosolic free magnesium levels in ischemic rat heart. J. Biol. Chem. 1989, 264, 5622. 48. Mulquiney, P. J.; Kuchel, P. W., Free magnesium-ion concentration in erythrocytes by 31P NMR: the effect of metabolite-haemoglobin interactions. NMR Biomed. 1997, 10, 129. 49. Canny, M. D.; Jucker, F. M.; Kellogg, E.; Khvorova, A.; Jayasena, S. D.; Pardi, A., Fast cleavage kinetics of a natural hammerhead ribozyme. I Am. Chem. Soc. 2004, 126, 10848. 50. Faulhammer, D.; Famulok, M., The Ca2+ ion as a cofactor for a novel RNA-cleaving deoxyribozyme. Angew. Chem. mt. Edit. 1996, 35, 2837. 51. Li, J.; Zheng, W.; Kwon, A. H.; Lu, Y., In vitro selection and characterization of a highly efficient Zn(II)-dependent RNA-cleaving deoxyribozyme. Nuci. Acids Res. 2000, 28,481. 52. Perrin, D. M.; Garestier, T.; Helene, C., Expanding the catalytic repertoire of nucleic acid catalysts: simultaneous incorporation of two modified deoxyribonucleoside triphosphates bearing ammonium and imidazolyl functionalities. Nucleosides & Nucleotides. 1999, 18, 377. 53. Santoro, S. W.; Joyce, G. F.; Sakthivel, K.; Gramatikova, S.; Barbas, C. F., RNA cleavage by a DNA enzyme with extended chemical functionality. I Am. Chem. Soc. 2000, 122, 2433. 54. Perrin, D. M.; Garestier, T.; Helene, C., Bridging the gap between proteins and nucleic acids: a metal-independent RNAseA mimic with two protein-like functionalities. J. Am. Chem. Soc. 2001, 123, 1556. 55. Lermer, L.; Roupioz, Y.; Ting, R.; Perrin, D. M., Toward an RNaseA mimic: A DNAzyme with imidazoles and cationic amines. I Am. Chem. Soc. 2002, 124, 9960. 56. Sidorov, A. V.; Grasby, J. A.; Williams, D. M., Sequence-specific cleavage of RNA in the absence of divalent metal ions by a DNAzyme incorporating imidazolyl and amino functionalities. Nuci. Acids Res. 2004, 32, 1591. 57. Christoffersen, R. E.; Marr, J. J., Ribozymes as human therapeutic agents. I Med. Chem. 1995, 38, 2023. 58. Lermer, L.; Hobbs, J.; Perrin, D. M., Incorporation of 8-histaminyldeoxyadenosine [8- (2-(4-imidazolyl)ethylamino)-2 ‘-deoxyriboadenosine] into oligodeoxyribonucleotides by solid phase phosphoramidite coupling. Nucleos. Nucleot. Nucl. 2002, 21, 651. 59. Ting, R.; Thomas, J. M.; Perrin, D. M., Kinetic characterization of a cis- and trans acting M2+-independent DNAzyme that depends on synthetic RNaseA-like functionality - Burst-phase kinetics from the coalescence of two active DNAzyme folds. Can. I Chem. 2007, 85, 313. 60. Hertel, K. J.; Herschlag, D.; Uhlenbeck, 0. C., A kinetic and thermodynamic framework for the hammerhead ribozyme reaction. Biochemistry 1994, 33, 3374. 61. Hegg, L. A.; Fedor, F. J., Kinetics and thermodynamics of intermolecular catalysis by hairpin ribozymes. Biochemistry. 1995, 34, 15813. 91 62. Esteban, J. A.; Banderjee, A. R.; Burke, J. M., Kinetic mechanism of the hairpin ribozyme. I Biol. Chem. 1997, 272, 13629. 63. Shih, I.; Been, M. D., Kinetic scheme for intermolecular RNA cleavage by a ribozyme derived from hepatitis delta virus RNA. Biochemistiy. 2000, 39, 9055. 64. Beebe, J. A.; Fierke, C. A., A kinetic mechanism for cleavage of precursor tRNA catalyzed by the RNA component of Bacillus subtilis Ribonuclease P. Biochemistry. 1994, 33, 10294. 65. Russell, R.; Herschlag, D., New pathways in folding of the tetrahymena group I RNA enzyrne.IMol.Biol. 1999,291, 1155. 66. Bevilacqua, P. C.; Kierzek, R.; Johnson, K. A.; Turner, D. H., Dynamics of ribozyme binding of substrate revealed by fluorescence-detected stop-flow methods. Science. 1992, 258, 1355. 67. Santoro, S. W.; Joyce, G. F., Mechanism and utility of an RNA-cleaving DNA enzyme. Biochemistry. 1998, 37, 13330. 68. Ke, A.; Zhou, K.; Ding, F.; Cate, 3. H. D.; Doudna, 3. A., A conformational switch controls hepatitis delta virus ribozyme catalysis. Nature. 2004, 429, 201. 69. Soukup, G. A.; Breaker, R. R., Relationship between internucleotide linkage geometry and the stability of RNA. RNA. 1999, 5, 1308. 70. Scott, W. G., Ribozyme catalysis via orbital steering. I Mo?. BioL 2001, 311, 989. 71. Bonaccio, M.; Credali, A.; Peracchi, A., Kinetic and thermodynamic characterization of the RNA-cleaving 8-17 deoxyribozyme. Nuc?. Acids Res. 2004, 32, 916. 72. Pyle, A. M.; Green, J. B., Building a kinetic framework for group II intron ribozyme activity: Quantitation of interdomain binding and reaction rate. Biochemistry. 1994, 33, 2716. 73. Guo, H. C. T.; Collins, R. A., Efficient trans-cleavage of a stem-loop RNA substrate by a ribozyme derived from Neurospora VS RNA. EMBO. 1995, 14, 368. 74. Rosenstein, S. P.; Been, M. D., Self-cleavage of hepatitis delta virus genomic strand RNA is enhanced under partially denaturing conditions. Biochemistry. 1990, 29, 8011. 75. Thomas, J. M.; Ting, R.; Perrin, D. M., High affinity DNAzyme-based ligands for transition metal cations- a prototype sensor for Hg2. Org. Biomol. Chem. 2004, 2, 307. 76. Esteban, J. A.; Walter, N. G.; Kotzorek, G.; Heckman, J. E.; Burke, J. M., Structural basis for heterogeneous kinetics: Reengineering the hairpin ribozyme. Proc. Nat?. Acad. Sci. USA. 1998, 95, 6091. 77. Brown, T. S.; Chadalavada, D. M.; Bevilacqua, P. C., Design of a highly reactive HDV ribozyme sequence uncovers facilitation of RNA folding by alternative pairing and physiological ionic strength. I Mo?. Biol. 2004, 341, 695. 78. Zamel, R.; Poon, A.; Jaikaran, D.; Andersen, A.; Olive, J.; Dc Abreu, D.; Collins, R. A., Exceptionally fast self-cleavage by a neurospora varkud satellite ribozyme. Proc. Nat?. Acad. Sd. USA. 2004, 101, 1467. 79. Amontov, S. V.; Taira, K., Hammerhead minizymes with high cleavage activity: A dimeric structure as the active conformation of minizymes. I Am. Chem. Soc. 1996, 118, 1624. 80. Kuwabara, T.; Amontov, S. V.; Warashina, M.; Ohkawa, 3.; Taira, K., Characterization / of several kinds of dimer minizyme: simultaneous cleavage at two sites in HIV-l tat mRNA by dimer minizymes. Nuc?. Acids Res. 1996, 24, 2302. 81. Stage-Zimmermann, T. K.; Uhlenbeck, 0. C., Hammerhead ribozyme kinetics. RNA. 1998, 4, 875. 82. Ting, R.; Thomas, J. M.; Lermer, L.; Pen-in, D. M., Substrate specificity and kinetic framework of a DNAzyme with an expanded chemical repertoire: a putative RNaseA 92 mimic that catalyzes RNA hydrolysis independent of a divalent metal cation. Nuci. Acids Res. 2004, 32, 6660. 83. Fersht, A., Enzyme Structure and Mechanism. p.195-196. 2nd ed.; W.H. Freeman & Co.: New York, 1985. 84. Hertel, K. J.; Herschlag, D.; Uhienbeck, 0. C., A kinetic and thermodynamic framework for the hammerhead ribozyme reaction. Biochemistiy. 1994, 33, 3374. 85. Kurreck, 3.; Bieber, B.; Jahnel, R.; Erdmann, V. A., Comparative study of DNA enzymes and ribozymes against the same full-length messenger RNA of the vanilloid receptor subtype I. J. Biol. Chem. 2002, 277, 7099. 93 CHAPTER 2: ARYL THIOETHERS AS NOVEL PHOTOCAGING GROUPS FOR GENE REGULATION 2.1 INTRODUCTION The precise control that we have over light makes it an attractive means for probing both biological processes and directing photosensitized reagents. As light does not influence many naturally occurring biochemical stimuli such as concentration gradients, variations in temperature, protein signals, and the physical presence of membranes, the use of light in combination with light-sensitive photoactivatable bioconjugates is useful for precicely probing specific biological stimuli without also altering other variables of a biological system.86’7 Originally, photolabile chemical functionalities were applied as protecting groups in synthetic chemistry.88 The first biochemical application of a photolabile protecting strategy was not reported until 1971 with caged chymotrypsin (Figure 2.A).89 It was Hoffhian in 1978 who coined the term photocaging with the synthesis of a caged adenosine triphosphate derivative (Figure 2.B).9° The term photocaging is often misleading, as it implies that the photolabile group masks the solvent exposed surface of a biologically relevant molecule. In this field, the term photocaging is more loosely used to refer to the masking of a biochemical activity. Publication statement: A version of this of this thesis chapter has been published. 1) Ting, R.; Lermer, L.; Perrin, D. M., Triggering DNAzymes with light: a photoactive C8 thioether-linked adenosine. J. Am. Chem. Soc. 2004, 126, 12720. 94 The masking of a species’ activity alone does not guarantee that the caged state also masks undesired biochemical activity. ____ + H0 chymotrypsin NH2 N-L <-juO 0 N N HO--O--O-I-O---1 OH OH OH + OH A B hv H2O hvO 0II II Figure 2.1. Early photocaged molecules. A, top, ct-chymotrypsin deactivated in its cis-cinnamate ester form is activated upon irradiation through isomerization of the cinnamate to its trans-form which is 1 O times more susceptible to hydrolysis.89 Subsequent hydrolysis releases cL-chymotrypsin which catalyzes the hydrolysis of tyrosine ester to tyrosine which is further converted to the detectable pigment melanine by tyrosinase. B, bottom, the light activated ATP from which the term photocaging was coined.90 Caged nitrobenzyl ATP is irradiated with light to release nitrosoaldehyde and ATP. Photocaging works well when applied to biological systems exhibiting short kinetic processes in the minute to sub-microsecond range.91 Many biochemical processes that fall in this range have relevantly photocaged molecules, the most famous of which is the cellular unit of energy and cellular signaling molecule, ATP,9° whose photocaged derivatives are commercially available (see molecular probes). Other photocaged small molecules include amino acids, steroids, sugars, and lipids.92 Proteins and peptides that have been photocaged include proteases, kinases, nuclease, toxins, cell-matrix proteins, receptors, serum proteins, galactosidases, and antibodies.87’9395 95 2.11 Photocaging There are two different strategies to the photoactivation of a biologically relevant molecule: reversible and irreversible photocaging. In the former strategy, isomerizable photo-switches such as azobenzenes and spiropyrans have been used to modify enzymes96 and DNAzymes.97 These chemical groups reversibly isomerize in the presence of light such that activity can be switched on and off (Figure 2.2). The obvious advantage to this photocaging approach is that light altered activity can be reversed. Unfortunately, the reversible switches are usually quite large and hydrophobic, limiting their use to large molecules and sterically sensitive applications. The major criticism of reversible photocaging is that the chemical differences between active and inactive states of a reversibly caged molecule are minor, therefore, some background activity is almost always observed by the inactive state. Activities that are exhibited by molecules in their inactive form often complicate the interpretation of experimental data. hv1 - hv2 cL-amylase Figure 2.2. Examples of reversibly photocaged molecules applied to biologically relevant systems. A, top, an example of azobenzene caged DNAyzme activity.97 B, Bottom, spiropyran caged t-chymotrypsin activity.96 Different frequencies of light favor different isomers of the double bond. hv1 ooff hv2 ‘Ion” 96 2.1.2 Irreversible photocaging The irreversible photocaging strategy is more successful than reversible strategies in obtaining caged molecules with no activity. Irreversible caging strategies function by masking vital chemical functionalities within a system. Active biomolecules or substrates are generated from photocaged species through the photolytic removal of the unnatural caging group. This strategy is not limited by a molecule’s size and has been successfully applied to systems both large and small. An irreversible photocaging strategy usually involves the synthetic incorporation of a photoreactive group onto a biologically active molecule. This incorporation masks the activity of the molecule resulting in a species that is inactive in the absence of light. The orthonitrobenzyl group is photoreactive group that is currently the most utilized photocaging functionality in the literature. Orthonitrobenzyl caged reagents photolyze to nitrosobenzaldehydes (Figure 2.3A) and are often used to protect phosphates, alcohols, carbonyls and amines. Since its first report in 1970 by Woodward88 this photolabile group has been chemically modified for photolytic efficiency with the introduction of alkyl groups at the benzyl center and can be tuned for deprotection at different wavelengths of light by the addition of electron donating substituents.98 Other common systems used for photocaging are the coumarin and the p-hydroxy phenacyl systems. Both systems can be used to cage the same functionalities as orthonitrobenzyls but vary in their physical properties. p-hydroxyphenacyl derivatives photolyze to p hydroxyphenylacetic acids (Figure 2.3B)99 which are less toxic than nitrosoaldehydes, the byproduct of orthonitrobenzyl decaging. Coumarin caged molecules photolyze very efficiently 97 in aqueous conditions to give their hydroxyl coumarin derivatives (Figure 2.3 C).’°° Coumarins have the added advantage that they can be employed to mask aldehydes, and ketones.’°’ Biomolecule N Aldehydes and Ketones: Biomolecule OH HO Br 0 A B Biomolecule hv H20 1O N.. + Biomolecule x,. hv H20 C Nucleophiles: OOH BiomoecuIe HO XH OH Biomolecule + I N X.-H hv H20 hv H20 Biomolecule + I O) X=O, N,orS Figure2.3. Common reagents used in biochemical photochemistry and their photoproducts.88’99-101 Other less studied photocaging groups exist, many of which have been applied in biological systems. These newer photocaging groups have not received nearly as much attention as the examples shown in Figure 2.3, but are applied with similar strategy in mind: a conjugated or aromatic core is utilized to collect light. The acquired energy is focused into breaking an 98 oxygen or nitrogen based bond that is used to tether the biological molecule to the caging group. 2.1.3 Drawbacks of current photocaging strategies Alternatives to the orthonitrobenzyl protecting strategy, such as the coumarin and p methoxyphenacyl systems have been developed in response to disadvantages associated with the nitrobenzyl chromophore. The use of the nitrobenzyl protecting strategy is limited to monitoring reactions that occur on a moderate timescale. This is due to the slow decomposition of intermediates that follow aci-nitro intermediate formation (Figure 2.4).b02 These rate limiting steps occurs on the order of 1 sec to 10 mm and poses a problem for monitoring faster processes.103 Biomolecule I Biomolecule hv XH X ,—.CkCN e H N,.0 + Cij) siowf x (Th Figure 2.4. The photolysis mechanism of a caged nitrobenzyl complex shown with aci-nitro (A), isoxazole (B), and nitrosobenzyl hemiacetal (C) intennediates. Other problems99 with nitrobenzyl caging include: (1) high UV absorbtivities of intermediates and products that serve to reduce the observed quantum yield of a system, (2) side reactions that occur between the highly reactive and often toxic nitroso products with primary amines and other nucleophiles,99 (3) premature hydrolysis,99 and subsequent biochemical problems B 99 associated with nitrobenzene caged derivatives prior to photorelease, and (4) unwanted side reactions that occur between photolysis and nitrosobenzyl hemiacetal decomposition. The problems associated with nitrobenzyl protection are being dealt with through research into the development of new photolabile groups. An ideal photocaging group must embody high molar extinction coefficients, quantum yields that are near unity, and must not possess the problems associated with nitrobenzyl protection. The search for new photocaging strategies is still relevant, as no single photoprotecting strategy which meets all these criteria has been described. 2.1.4 Photo regulation of gene expression Interest in the application of photocaging groups to the control of gene expression grows with each new report of oligonucleotide signaling, recognition, and catalysis. Being able to photocage nucleic acids would have widespread implications on a large number of applications including cellular signaling (cyclic’04 and linear’05 nucleotide triphosphates), gene regulation (RNAi, mRNA, antisense approaches), molecular diagnosis (aptamers9’and microarrays), nano materials, and catalysis. The strategy of “backbone caging” was first reported by Monroe106 who nonspecifically photocaged, transfected, and expressed a gene coding for green fluorescent protein (GFP) through a strategy which involved nonspecifically protecting the phosphate backbone of the entire plasmid coding for green fluorescent protein with approximately 270 caging groups. This approach is simple but is still being applied.’07 Unfortunately, it suffers from drawbacks associated with non-specific chemistry and low compounded yields from the required deprotection of multiple caging groups. 100 Alternative chemistries for nucleotide caging have targeted the nucleotide ribose sugar, or the nucleobase itself. MacMillan108 employed a nitrobenzyl strategy for protecting the 2’ hydroxyl functionality of RNA. These modified RNA bases were incorporated into the hammerhead ribozyme system where activity was completely quenched in the absence of the light, only following irradiation was activity restored. This protection strategy could also be used for protecting RNA during its solid phase synthesis. The 2’ hydroxyl of RNA is typically protected on the solid phase resins with silyl protecting groups that are removed with fluoride. Macmillan’s photocaging strategy utilizes light to deprotect the 2’ hydroxyl of RNA, reducing the need for solution phase reagents for 2’ hydroxyl RNA deprotection. In terms of protecting the nucleobase, reports exist for the nitrobenzyl protection of all four nucleobases on an exocyclic nitrogen or oxygen (Figure 2.5).91 109, 110 while these nucleobases pertain to different research in the fields of transcription,” aptamer activity,9’structure-activity relationships, and DNA replication, these photolabile protection strategies function similarly by preventing the photocaged species from forming nucleobase interactions such that the processes that they are designed to study cannot occur until the irradiation of the system with light. 101 RNA 2 Hydroxy Protection MacMillan RNAIDNA Nucleobase Protection UIT Ni-I2 N HO N Os O 0% ,Ne Figure 2.5. Photocaged ribose and nucleotide base photocaging strategies as described in MacMillan, Silvennan Heckel, and Dmochowski.91’108410 Although the biological activity involved with nucleotide phosphate, ribose, and nucleobase caging strategies in Figure 2.5 are novel, the photochemistry is not. With the exception of Ando’s coumarin based system,107 all described strategies suffer from the general problems associated with nitrobenzyl caging groups discussed in the previous section. G A C 102 2.1.5 Proposed Research We have serendipitously discovered a novel aryl thioether based photoactivatable group that is efficient in photocaging adenine. To demonstrate that this photoactivatable group can be applied to gene expression, we have applied this photocaging strategy to the Zn2 dependent DNAzyme, 8 1 7E, a nucleic acid catalyst capable of catalytic site specific cleavage of an all RNA sequence. This chapter will be divided into 2 sections. Section 2.2 will discuss the experiments that were required for the discovery of the photoactivity of the aryl thioether photoactivatable group on DNA as well as adenosine. The source of hydrogen during the photoconversion of thioether modified adenosine tounmodified adenosine will also be investigated in this section. Section 2.3 will describe aspects of the application of this thioether moiety to the DNAzyme 8- 1 7E including the incorporation of this moiety into solid phase syntheses and the activation of photocaged 8-1 7E with light. 103 2.2 THE DISCOVERY OF A NO VEL PHOTOCHEMICAL REACTION Imidazole ethyl thioether photoactivity was initially discovered through experiments designed to study the structure-activity relationship between the chemical elements of the histaminyl modified deoxyriboadenosine NH-dA’m(Figure 2.6A) and the M2 independent RNA cleavage activity that this modification conveys onto the DNAzyme 925-1 ic (Chapter 1). It was hypothesized that by modifying adenosine at the C-8 position with histamine, two chemical functionalities novel to DNA are incorporated onto adenosine; an imidazole moiety and a hypothetical guanidinium moiety (Figure 2.6A). There is no reported value for the pKa of 8- aminoadenine and verification of the hypothetical guanidinium pKa is difficult. In order to investigate the importance of this potential guanidinium moiety to the activity of DNAzyme 925-11 c, a sulfur containing isostere of NH-dA1m,s-dA’m (Figure 2.6B), was synthesized as a triphosphate in anticipation that it could be polymerized into a sulfur containing derivative of 925-1 lc. A Imidazoe B Imidazole GU I5 S NH-dA’m SdAIm Figure 2.6. A., Structure of the histaminyl modified deoxyriboadenosine NHdAim required for the activity of DNAzyme 9-25-1 ic. B., Structure of the imidazole ethyl thioether modified deoxyriboadenosine isostere S-dA1m. 104 The triphosphate of S-dA’m,2.5, was synthesized and assayed as a polymerase substrate for the DNA polymerase Klenow (DNA polymerase I). Unfortunately, the incorporation of modified triphosphate 2.5 into short DNA sequences gave anomalous results during their analyses in denaturing polyacrylamide (D-PAGE) gels. Mobility analyses of DNA sequences with presumably incorporated S_dAim gave products that were more similar to that of unmodified adenosine containing DNA than that of NH-dA’m containing DNA. We initially hypothesized that preparations of 2.5 were contaminated with an unmodified dATP impurity that had been carried through 5 reaction steps. Confusingly, multiple sequential purifications of 2.5 on preparative TLC and HPLC gave products that still contained the unmodified dATP contamination. We began to suspect that 2.5 was reacting with UV-vis radiation output by a handheld UV lamp, a stimulus that was present in both preparative TLC and the HPLC chromatography used to isolate 2.5. This section will discuss how the photoconversion of C8-linked 2-(4- imidazolyl)ethylthio)adenosine to unmodified adenosine (Figure 2.7) was discovered through S-dA’m triphosphate incorporation experiments. This novel photolysis will be applied to the imidazole thioether containing triphosphate 2.5, the imidazole thioether adenosine ribonucleoside 2.6, and DNA synthesized with the S-dA’m containing phosphoramidite 2.8. The source of the C-8 proton that is incorporated into the final adenosine photoproduct will also be investigated. HN—/ NH2 hv N NH2 280 nm,254nm HN N MeOHorH2 + HJ3 SH R= alkyl, ribose, or deoxyribose Organic Product Nucease Product Figure 2.7. C8-linked 2-(4-imidazolyl)ethylthio)adenosine photolyzes to unmodified adenosine aid an imidazole product. 105 2.2.1 MATERIALS AND METHODS 2.2.1.1 General synthetic methods All chemicals were purchased from Sigma, Aldrich, or Lancaster. Deuterated solvents were purchased from Cambridge Isotope Laboratories. Thin layer chromatography retention factor (Rf’s) are reported on Silica Gel 60 F254 Glass TLC plates from EMD Chemicals. Oligonucleotide syntheses were performed on an Applied Biosystems automated DNA synthesizer. Polynucleotide kinase was purchased from New England Biolabs. ATPy32Pwas purchased from Perkin-Elmer. All ‘H and 32P nuclear magnetic resonance (NMR) spectra were recorded at room temperature on a Bruker Avance 300 or 400 MHz instrument. Chemical shifts are reported using the scale in ppm and all coupling constants (J) are reported in hertz (Hz). Unless specified, ‘H NMR spectra are referenced to the tetramethylsilane peak (6 = 0.00) and 32P NMR spectra are referenced to the phosphoric acid (H3P04)peak (6 0.00). Mass spectrometry was performed at the Mass Spectrometry lab of the University of British Columbia (U.B.C.) Chemistry Department and the Scripps Research Center for Mass Spectrometry. DNA and RNA oligomers were synthesized by the University Core DNA and Protein Oligonucleotide synthesis laboratory at the University of Calgary. 106 2.2.1.2 Specific synthetic methods 2.1. 4(5)-(2-Haloethyl)imidazole hydrochloride. The title compound was synthesized according to the procedures of Stensiö and Wahlberg.”2 A 250 mL round bottom flask containing 42.4 mL of 1.5 M sulphuric acid, histamine dihydrochloride (5.34 g, 28.9 mmol) and potassium bromide (11.52 g, 96.0 mmol) was fitted with a magnetic stirrer and cooled to -15 °C using a ethene glycol/ dry ice bath. A saturated solution of sodium nitrite (2.58 g, 37.2 mmol) was made up in 3.8 mL of water. The saturated sodium nitrite solution was added all at once to the magnetically stirred sulphuric acid! histamine solution kept at -15 °C. A colour change from colourless to a deep orange-brown was observed and a gas had evolved in the reaction flask immediately upon addition. 30 mm into the reaction, the reaction mixture was warmed to room temperature. After a total of 3 hours, no more bubbling was observed and the reaction mixture had changed from a deep orange-brown to an almost colourless light yellow. The reaction mixture was cooled to -15 °C and made to a pH of 10 with the drop wise addition of 27.6 mL of 5 M sodium hydroxide. This basic solution was transferred to a 100 mL separatory funnel where the desired product was quickly extracted with chloroform (5 x 10 mL). The colourless chloroform layer was added directly from the separatory funnel to 64 mL of 0.5 N HC1/ ethyl acetate (32 mmol HC1) in a 250 mL round bottom flask, This acidification step was necessary in order to prevent the formation of intramolecular substitution products during storage. Concentration of this organic layer under vacuum gave a white crystalline hydrochloride salt. NMR spectroscopy and mass spectrometry analyses showed that the salt was a mixture of 4(5)-(2-bromoethyl)imidazole hydrochloride and 4(5)-(2- chloroethyl)imidazole hydrochloride in roughly a 3:1 ratio. LSI MS (D20)= 177 MH (Br) (175 (100 %) 177 (97.3 %) calculated forC5H8BrN2),133 MH(Cl) (133 (100 %), 135 (32 %) calculated forC5H8lN2). 4(5)-(2-bromoethyl)imidazole hydrochloride ‘H NMR (200 MHz, 107 D20, pH = 2): 6 8.60 (1H, s) (H-2 imidazole), 6 7.35 (1H, s) (H-4 imidazole), 6 3.84 (2H, t, J = 6.5 Hz) (Br-C-CH-imidazo1e), 6 3.21 (2H, t, J 6.5 Hz) (Br-CH2-C-imidazole). 4(5)-(2- chloroethyl)imidazole hydrochloride ‘H NMR (200 MHz, D20, pH = 2): 8.60 (1H, s) (H-2 imidazole), 7.35 (1H, s) (H-4 imidazole), 3.69 (2H, t, J = 6.3 Hz), 3.31 (2H, t, J = 6.5 Hz). 2.2. 4(5)-(2-(S-Thiouronium)ethyl)imidazole dihydrohalide. A 250 mL round bottom flask was charged with a magnetic stirrer, 12.2 mL of water, and 25 mmol of 4(5)-(2- haloethyl)imidazole hydrochloride 2.1. A solution of thiourea was prepared by dissolving thiourea (2.1 g, 27.6 nimol) in 49 mL of water in a 125 mL Erlenmeyer flask. The thiourea solution was added to the magnetically stirring light yellow solution of 2.1 which was then heated to 65 °C in an oil bath and left for 8 hours. The contents of the entire reaction pot were lyophilized to give a hygroscopic, fluffy white solid. This solid was washed twice with 5 mL of acetonitrile to remove excess thiourea. The acetonitrile was decanted off and the white precipitate that was retained was placed under vacuum. Resulting NMR analyses showed that both the chloro- and the bromo- species of 2.1 were converted entirely to the thiouronium salt 2.2. ‘H NMR (200 MHz, D20, pH = 2): 8.71 (1H, s) (H-2 imidazole), 7.43 (1H, s) (H-4 imidazole), 3.52 (2H, t, J = 6.6 Hz), 3.25 (2H, t, 3 = 6.6 Hz). 108 2.2.1.2.1 Triphosphate preparation NH2 Br—) ___ NaO—P—O I 1. NaOH ONa pH13, OH 2. Doex-2 2.3 NH4OAc H NL N s—j O o o N N O_ O_ O_ OH 2.5 2.3. Sodium 8-bromo-2’-deoxyriboadenosine-5’-monophosphate (8-Br-dAMP). The synthesis of 8-Br-dAMP was adapted from a protocol reported by Ikehara and Uesugi.113 The disoclium salt of 2’-deoxyriboadenosine monophosphate dihydrate (dAMP) (0.55 g, 1.3 mmol) was dissolved in a 250 mL round bottom flask charged with a magnetic stirrer and 52 mL of a 0.25M barium acetate buffer Q,H 4.5, 52 mmols acetate! acetic acid and 13 mrnol Ba2). The colourless solution was then degassed with nitrogen and a solution of saturated bromine-water, prepared by stirring bromine (0.13 mL, 2.54 mmol) into 13 mL of degassed water, was added to the dAMP solution via a syringe. The reaction was stoppered and allowed to proceed at room temperature for 12 hours. Completion of the reaction was observed by TLC with 5:2:3 1- butanol : acetic acid : water as the running solvent (Rf = 0.38 for sodium 8-Br-dAMP and Rf = 0.23 for sodium dAMP). Excess bromine was removed by flushing the open round bottom flask with nitrogen. The resulting solution was then lyophilized overnight, resuspended in 5 The following scheme details the synthetic route for the synthesis of triphosphate 2.5. S N N H H H2N°NH N N2 _ ____ NaNO2, H2S04 _______ _ N°°-NH3+ XS KBr, -15°C H2O, 65°C f.°°°—°°s NH2 IHCI C HCIX Br:CI 3:1 X= Br, CI I2.1 2.2 I NH2 <ijN II NaO-P-O O4N ONa OH 1. Br2, Ba(OAc)2/HOAc pH4.5 r.t. 2. Na2SO4 2. ç’N°N DMF r.t. MeOH, HO—P—O-P—OH . 1.5 TBA OH OH DMF r.t. H N NH2 N N N.-S_</ N OH 2.4 109 mL water, and 5 mL of ethanol was added to promote precipitation. The precipitate was collected by centrifugation, and the ethanol/water precipitation was repeated two more times before washing the precipitate with pure ethanol (2 x 10 mL). The fmal precipitate was dissolved in 10 mL of water resulting in an orange suspension. To this suspension, a solution of sodium sulphate (1.85g, 0.013 mol) in 6 mL water was added. The now blood-red mixture was centrifuged to remove a barium sulphate pellet that was washed with water (2 x 5 mL). The washes from the barium sulphate concentrates were collected and lyophilized in a 250 mL round bottom flask. A wavelength spectrum scan on a Unicam UV4 UV-Vis spectrophotometre showed that the resulting lyophilate had a 2max at 264 nm. 11 900 OD units (—‘1 .1 mmol) were obtained. NH NH2 x:frBr Br-jJ NH4°O1,NH4°°iO1, NH4 OH NH4 OH Syn Anti 2.3 MALDI MS (3-NBA matrix): = 410 (M-H) (408 (100 %), 410 (99.3 %) calculated for C10H2BrN5O6P). 1H NMR mixture of syn and anti (see appendix) (300 MHz, D20): 7.92, 7.95 (1H, s) (CH2), 6.29-6.23 (1H,m) (CH”), 4.53-4.57 (1H, m) (CH3’), 3.97-3.92, 3.87-3.84, 3.80- 3.73 (1H, m, 1H, m, 1H, m) (CH4’, CH5’, CH5’), 3.17-3.07 (1H, m) (CH2’), 2.26-2.16 (1H, m,) (CH2’). 110 2.4. Ammonium 8-(2-(4-imidazolyl)ethylthio)-2 ‘-deoxyriboadenosine-5’-monophosphate (S-dA’mMP). A 50 mL round bottom flask was charged with a magnetic stirrer, 11 mL water and 8-Br-dAMP 2.3 (1.1 mmol). The thiouronium salt 2.2 (2.0 mmol) was dissolved in 3.3 mL of water and added to the magnetically stirred round bottom flask containing the 8-Br-dAMP 2.3. Following this addition, 0.66 mL of 5 M sodium hydroxide (3.3 mmol) was added and the reaction was heated to 65 °C in an oil bath. The progress of the reaction was monitored on a Unicam UV4 UV-Vis spectrophotometre. After 4 hours, the complete disappearance of the reactant, as detected by observing the reagent absorption maximum at 264 nm, and the appearance of an absorption maximum at 282 nm indicated that the reaction was complete. Near the end of the 4 hours, the formation of a black precipitate was also observed in the reaction flask. This precipitate was removed by centrifugation and washed with water (2 x 1 mL). 9020 OD units with a max at 282 nm were recovered in the supematant which was lyophilized to give a white powder. In order to separate the desired nucleotide from free 4(5)- (2-thiolethyl)imidazole, the white lyophilate was resuspended in 2 000 mL of water in a round bottom flask for absorption on an ion exchange chromatography matrix. The pH of the 2 000 mL solution was adjusted to a pH of 10.5 with 1.24 mL 5 M sodium hydroxide and 12 g of Dowex 1X8-400 (chloride form) was added to the 2 000 mL vessel. UV-Vis analysis of the supernatant showed that 7 805 OD units with a 2max at 282 nm were absorbed onto the Dowex. The nucleotide-bound Dowex was placed into a 1.5 cm x 7 cm glass column. A linear gradient of 500 mL water to 500 mL 2 M ainmonium acetate pH 4.0 was used to elute the ammonium salt of the desired product. 15 mL fractions were collected and 8260 OD units (0.59 mmols, 2rnax at 282 nm) of the desired product 2.4 was obtained between fractions 7 and 13. ill N H N 12 I NH211( 7 los’y NH4 1’ OH I 4 Fl4 2.4 ‘H NMR (300 MHz, D20, water suppression): 8.46 (1H, s) (CH’3), 8.20 (1H, s) (CH2), 7.22 (1H, s) (CH’2), 6.50 (1H, t, 3 = 7.1 Hz) (CH”), 4.26-4.35 (2H, m) (CH’°), 4.20-4.10 (2H, m) (CH3”4’), 3.23-3.29 (2H, m) (CH5’), 3.23-3.29 (2H, m) (CH”), 3.10-3.20 (1H) (CH2’), 2.38-2.45 (1H, m) (CH2’). 2.5. Triethyl ammonium 8-(2-(4-imidazolyl)ethylthio)- 2’-deoxyriboadenosine- 5’- triphosphate. The ammonium salt of 2.4 was converted to the triethyl ammonium salt of 2.4 (TEA 2.4) by gel filtration chromatography on a Sephadex Gi 5 column. This was accomplished by suspending 2550 OD units (2’.m at 282 nm) of the lyophilized ammonium salt of 2.4 in 1.5 mL of 0.5 M triethyl ammonium acetate pH 9. This volume was loaded onto a 1 cmx 20 cm column of Sephadex G15. 0.4 mL fractions were collected and 2270 OD units (2.max at 282 nm) of TEA 2.4 eluted between fractions 14 - 23. These fractions were pooled, lyophilized overnight and resuspended in 5 mL ethanol. The TEA salt of 2.4 was converted to the tributylammonium salt (TBA 2.4) by the addition of tributylamine (1.26 mL 5.29 mmol) to the ethanol solution. Concentration of this solution by rotary evaporation removed the ethanol and remaining free triethylamine. The remaining oil was washed with 25 mL of hexanes to remove excess free tributyl ammonium acetate from the TBA 2.4 salt. To ensure complete 112 conversion of TEA 2.4 into TBA 2.4, the precipitate was resuspended in 2 mL more ethanol and more tributylamine (0.3 mL, 1 .26mmol) was added. This suspension was allowed to sit for 30 mm before it was washed with 25 mL more hexane. The hexane layer was decanted off, and the precipitate was washed with three additional 30 mL hexane washes. The resulting white, fluffy solid, TBA 2.4, was evaporated to dryness under a vacuum. The conversion of TBA 2.4 to the triphosphate 2.5 was performed according to procedures documented by Hoard and Ott.’14 TBA 2.4 was placed into a sealed, flame-dried, 100 mL round bottom flask that was pre-charged with 1.31 mL of distilled anhydrous dimethyl formamide (DMF), a magnetic stir bar, and a positive nitrogen pressure. 1,1 ‘-carbonyldiimidazole (0.131 g, 0.807 mmol) was added despite the fact that TBA 2.4 displayed only partial solubility in DMF. The reaction mixture was allowed to sit overnight before it was quenched with methanol (65 pi, 1.6 mmol). The quenched reaction was allowed to stir for 30 mm before tributylammonium pyrophosphate (H4P207.1.5 C,2H7N(TBA) 0.2 17 g, 0.475 mrnol) and 15 mL more anhydrous DMF were added. The coupling of pyrophosphate to the activated phosphoimidazolide took place over 4 hours. The progress of the reaction was checked every hour by adding 50 jil aliquots of the reaction to 3 % lithium perchiorate in acetone to obtain 2.5 as a lithium salt. This salt was resolved by TLC, with 6:1:4 p-dioxane: ammonium hydroxide : water as the mobile phase (Rf = 0.14 for lithium triphosphate and Rf = 0.39 for lithium monophosphate). Upon completion of the reaction, the reaction flask was placed under vacuum for two days to remove DMF. To separate the desired triphosphate from monophosphate, diphosphate, and imidazole impurities, this oil (3 mL) was transferred to a 2 L round bottom flask and dissolved in 1 L of water for purification on an ion exchange column. The pH of the 1 L solution was adjusted to 9 with 0.4 mL of triethylamine before 7 g of Dowex 1X8-400 (chloride form) was added to the vessel. UV-Vis analysis of the supernatant showed that 805 OD units with a 2max at 282 nm 113 were absorbed onto the Dowex. The nucleotide-bound Dowex was placed into a 1.5 cm x 7 cm glass column and a linear gradient of 500 mL 0.1 M triethylammoriium acetate pH 9 - 500 mL 2 M ammonium acetate pH 4.0 was used to elute the animonium salt of the desired product. 13 mL fractions were collected and 345 OD units at 282 nm) of pure ammonium 2.5 was isolated. The very pure ammonium 2.5 sample was lyophilized to dryness in a 500 mL flask before conversion to the lithium 2.5 salt. 50 mL of 3 % lithium perchiorate in acetone was added to the lyophilized sample to form a biphasic mixture. A 200 mL solution of acetone, then 250 mL of diethyl ether was added to the round bottom to precipitate out a white solid. The supematant was decanted and the precipitation was repeated. The isolated white solid was washed with acetone (3 x 150 mL) and then washed with ethanol (4 x 5 mL) to give a yellow coloured solid, the lithium salt of 2.5. This salt was purified by HPLC on a 4.6 mm x 250 mm Phenomenex Jupiter lOj.im C4 300 A reverse phase column. The desired product eluted at 4.1 ± 0.4 mm at a 1 mL/ mm flow rate using a linear gradient of 20 mM NH4OAc (2 % MeCN) to 20 mM NH4OAc (16.4 % MeCN) run over 35 mm at 1 mL/min. The peak eluting at 4.1 ± 0.4 mm 2.5 was concentrated in a speed vac for further use in enzymatic study. MALDI MS (HPA matrix): 616.0 (M-H) (616.02 calculated forC15H21N702P3S). 2m = 280 nm, E280= 18 000 M’cm’. 2.2.1.2.2 Imidazole ethyl thioether adenosine preparation 2.6. 8-(2-(4-Imidazolyl)ethylthio)-adenosine. A 50 mL round bottom flask was charged with a magnetic stirrer, 30 mL DMSO, and 1 g (2.89 mmol) of 8-Bromoadenosine. A 1 g mass of the thiouronium salt 2.2 (2.0 mmol) was dissolved in 10 mL of DMSO. A 2 mL solution of an aqueous 5 M sodium hydroxide (10 mmol) solution was added to the thiouronium salt and this 114 basic solution was transferred to the magnetically stirring round bottom flask containing 8- bromoadenosine. The progress of the reaction was monitored through wavelength spectrum scans on a Unicam UV4 UV-Vis spectrophotometre. The reaction was monitored by observing the disappearance of an absorption maximum at 264 nrn corresponding to the reactant and the appearance of an absorption maximum at 280 nm indicating the formation of product (UV-Vis in methanol). After 5 hours, the product was precipitated upon addition of 200 mL of chloroform to the DMSO solution. This precipitate was centrifuged and collected by vacuum filtration before being triturated and filtered with 25 mL of water. The resulting solid was further triturated in 25 mL of a 10 % solution of methanol in chloroform. The supematant was collected by filtration and concentrated to oil. A precipitate was formed upon addition of 50 mL chloroform to the resulting oil. This precipitate was collected, dissolved in 1 mL of methanol and concentrated to oil. This oil was precipitated again with an additional 50 mL of chloroform. The resulting solid was collected by vacuum filtration and concentrated to give 0.14 g (0.36 mmol, 13 % yield) of a pink powder. 12 NH2 13 HO OHOH 2.6 ESI MS (1 % TFA in MeOH): = 394.1290 MH (394.12920 calculated forC,5H2oN7O4S). max = 280 nm, 6280 18 400 M’cm’. ‘H NMR (300 MHz, MeOH-d4): 8.08 (1H, s) (CH2), 7.76 (1H, s) (CH’3), 7.00 (1H, s) (CH’2), 5.95 (1H, t, J 7.24 Hz) (CH”), 4.99 (1H, m) (CH2’), 4.33 (1H, dd, J = 5.25 Hz, 1.35 Hz) (CH3’), 4.17 (1H, m) (CH4’), 3.86 (1H, dd, J = 12.71 Hz, 115 2.34 Hz) (CH5’), 3.76 (1H, dd, J = 12.74 Hz, 2.51 Hz) (CH5”), 3.59 (2H, m) (CH’°), 3.11 (2H, t, J = 6.61 Hz) (CH11). 2.2.1.2.3 Deoxyribophosphoramidite preparation The following scheme details the synthetic route for the synthesis of triphosphate 2.8. 0 /N :: ci 0N O HNI I III S. Su I I CH2I,EtiPr2N rsil ‘_/ L) ONzON 2.8 2.7. Synthesis of 8(-2-(4-imidazolyl)ethylthio)-5’-DMT-2‘deoxyriboadenosine. A 50 mL round bottom flask was charged with a magnetic stirrer, 7.5 mL DMSO, and 0.3 g of the thiouronium salt 2.2 (2.31 mmol). A 1 g mass (1.36 mmol) of N-(9-(5-(Bis-(4-methoxy- phenyl)-phenyl-methoxymethyl)-4-hydroxy-tetrahydro-furan-2-yl)-8-bromo-9H-purin-6-yl)- benzamide (synthesis previously reported)58 was dissolved in 7.5 mL of DMSO and placed in an addition funnel. The pH of the stirring thiouronium salt solution was adjusted to 10 with the addition of 950 jL of a 5 M sodium hydroxide solution (4.75 mmol). The protected bromoadenosine was immediately added to the basic magnetically stirred round bottom flask. The formation of product was observed by TLC where 2.7 has an Rf of 0.32 in a 10 % MeOH: CHC13 developing solution. This reaction was deemed to be complete after 20 mm. The entire DMSO solution was added to 100 mL of water and extracted against 100 mL of CHC13. The 116 CHC13 layer was washed with an additional 100 mL of water, dried over anhydrous magnesium sulphate, filtered and concentrated to give orange oil. The product was isolated by flash chromatography with 5 % MeOH: CHC13 as the eluant. The recovered product was concentrated in the presence of hexanes to give 0.9 g (1.14 mmol, 83 % yield) of white foam. 13 12 ii 1615 17 1ost018 28f)—!-Q--—i5 \zzJ’ 120 4’ ‘J 1’ 27 26 -21 ‘—.‘ 22 OH 24 2.7 ESI MS (MeOH): = 784.2900 MH (784.29118 calculated forC43N27O6Sj. ‘H NMR (300 MHz, MeOH-d4): 8.41 (1H, s) (CH2), 8.06 (2H, d, J = 7.8 Hz) (CH’6), 7.67 (1H, s) (CH’3), 7.65 (1H, t, J 7.5 Hz) (CH’8), 7.56 (2H, t, J = 7.8 Hz) (CH’7), 7.22 to 7.08 (9H, m) (CH 21, 26, 2728) 6.93 (1H, s) (CH’2), 6.75 (1H, d, J = 9.0 Hz) (CH28), 6.71 (4H, d, J = 9.0 Hz) (CH22), 6.39 (1H, t, J = 6.7 Hz) (CH”), 4.79 (1H, dt, 3 = 6.6 Hz, 4.4 Hz) (CH3’), 4.13 (1H, dt, J = 4.3 Hz, 5.2 Hz) (CH4’), 3.73 (3H, s) (CH24), 3.72 (3H, s) (CH24’), 3.64 (2H, dt, J = 7.5 Hz, 3.5 Hz) (CH5’), 3.53 (1H, q4, J = 6.5 Hz) (CH2’), 3.13 (2H, t, J = 7.2 Hz) (CH”), 2.31 (1H, ddd, J = 11.9 Hz, 7.2 Hz, 4.6 Hz) (CH2’). 2.8. Synthesis of 8(-2-(4-Imidazolyl)ethylthio)-5 ‘-DMT-2 ‘deoxyriboadenosine 3 ‘dilsopropyl-f3-cyano-ethoxy-phosphoramidite. The reagent 2.7 (0.9 g, 1.14 mmol) was added to a nitrogen flushed 50 mL round bottom flask charged with a stir bar and 14 mL of dry CH21. At room temperature, 0.5 mL (2.78 mmol) of 117 dry diisopropylethylamine and 0.5 mL (2.24 mmol) 2-cyanoethyl diisopropyl chioro phosphoramidite was added to the stirring flask over a nitrogen atmosphere. The reaction was monitored by TLC. The diasteromeric products of 2.8 had RfS of 0.31 and 0.20 in a 10 % MeOH: CHC13 developing solution. The reaction was stopped after 30 mm and washed 3 times with 15 mL of a 5 % bicarbonate solution. The organic layer was dried over anhydrous magnesium sulphate, filtered and concentrated. The product was isolated by flash chromatography using 230-400 mesh silica with 3 % MeOH: CHC13 as the eluant. The recovered product was concentrated in the presence of hexanes to give 0.72 g (0.74 mmol, 65 % yield) of white foam. The product contained trace amounts of oxidized phosphoramidite which complicated ‘H-NMR, but was not expected to couple during solid phase DNA synthesis. ESI MS (CHC13): = 984.7 MH (984.40 calculated forC52H9N9O7PS). 31P NMR diastereomers (300 MHz, CDC13): 8 150.16, 8 149.88. 2.2.1.3 Enzymatic, Kinetic, and DNA preparation procedures 2.2.1.3.1 Enzymatic incorporation assays Attempts at synthesizing imidazole ethyl thioether-containing DNA using the triphosphate 2.5 were made by annealing 50 pmol of 5’ 32P labeled primer with 62.5 pmol of template in 20 tL of lOx Kienow Eco poll buffer (NEB). In the following order, 180 tL of a 4 !.LM dGTP, 20 jiM 2.5 solution, 3.4 j.iL of 100 mM DTT, and 1.6 Units (1.6 p.L) of Kienow exo polymerase was added and the reaction was allowed to proceed for 3 hours at 37 °C. After polymerization, the sample was heated at 95 °C for 5 mm and loaded in a 20 % Denaturing PAGE gel. Elution of the sequence from the gel was performed using the crush soak method with 1 % LiC1O4 and 0.7 mM triethylamine as the eluant. The eluant was concentrated, washed twice with ethanol, 118 resuspended in 40 iiL H20 and passed over a G-25 spin column. This PAGE gel purification was repeated to ensure purity. 2.2.1.3.2 Synthesis and Purification of Substrates and Modified Catalysts 8(-2-(4-Imidazolyl)ethylthio)-5 ‘-DMT-2 ‘deoxyriboadenosine-3 ‘diisopropyl-3- cyanoethoxyphosphoramidite 2.8 was incorporated into oligonucleotides as described in Chapter i’ 58 Oligonucleotides were gel-purified by 20 % D-PAGE and identified by UV shadowing. Only the perimetres of UV absorbing bands corresponding to DNA were visualized by UV shadowing. The interior of these bands were protected by cardboard such that these interior regions were not irradiated. Only this interior region was extracted from gels. Elution of catalysts from gels were performed using the crush soak method using 1 % LiC1O4 and 0.7 mM triethylamine as the eluant. The eluant was concentrated, washed twice with ethanol, resuspended in 40 p.L H2O and passed over a G-25 spin column. This purification resulted in a sacrifice of approximately half of each band by UV shadowing so that there would be no premature activation. The concentration of catalyst was determined by summing standard values at 260 nm for each unmodified base and using 6260 mn=’4 400 cm1 M’ for modified adenosine. Oligonucleotides were stored at a final concentration of 20 iiM in preparation for kinetic assays. 2.2.1.3.3 DNA MALDI-TOF mass spectrometry MALDI-TOF procedures were performed as detailed. A 10 LL solution of oligonucleotide at a concentration of 100 pmol/ tL (100 nM) was added to -20 Bio-Rad AG5OW-X8 cation exchange beads (NH4 form) and allowed to sit for 15 mm at room temperature. A 1 1iL solution of a saturated solution of 3-HPA (Aldrich, 98 %) prepared fresh in 50/50 119 Acetonitrile/H20and was spotted on a MALDI-TOF plate. Immediately, 1 i.tL of the NH4 exchanged oligo solution was mixed with the saturated 3-HPA spot. This spot was allowed to dry under vacuum at room temperature prior to MALDI-TOF analysis. 2.2.1.3.4 Photochemical Procedures For all photochemical activations, catalysts or nucleosides were diluted to 600 .tL and placed in quartz cuvettes. These cuvettes were degassed with argon for 20 mm before being sealed. Irradiation took place with a handheld 1 mW 254 nm shortwave mercury handheld IJV lamp, a 10 mW Xe arc lamp, or a 10 mW tunable dye laser at 280 nm. All sources photolyzed to their respective unmodified adenines. Photolyzed ribosides were transferred to NMR tubes for ‘H NMR analysis while photoactivated 1 7E derivatives were transferred out of quartz irradiation cuvettes into eppendorf tubes and dried in a speedvac for kinetic analyses. 120 2.2.2 RESULTS The application of 4-imidazolyl)ethylthio as a photocaging group began with the serendipitous discovery that C8-linked 2-(4-imidazolyl)ethyltbio)adenosine in DNA photolyzed to unmodified adenosine in DNA. The triphosphate 2.5 was initially synthesized in order to probe the structure-activity relationship of the adenine-histamine linkage in modified adenosines in the DNAzyme 925-1 lc on RNA cleavage. Attempts to incorporate 2.5 opposite from thymidine in primed DNA by the DNA polymerase Klenow (DNA polymerase I), revealed the presence of a species suggestive of a dATP impurity. This impurity could not be removed from 2.5 by TLC and HPLC. It was hypothesized that the exposure of 2.5 to UV sources during purification had converted 2.5 into dATP. 2.2.2.1 The photolysis of C-8 thioethers of adenosine gives adenine 2.2.2.1.1 Photolysis of the triphosphate 2.5 In order to verify that UV light facilitates the conversion of 2.5 into dATP, the photolysis of the thioether-containing triphosphate 2.5 was investigated spectrophotometricaly. A 5 1iM, unbuffered pH 7.5 solution of 2.5 was placed in a quartz cuvette which was irradiated for 1 hour with a 1 mW, 254 nm shortwave mercury handheld UV lamp. Absorbance measurements were taken before and after irradiation as shown in Figure 2.8. 121 0.16 -- I I i I 0.14 — t0, pM Triphosphte 25 —tlh Irradiation 0.08 0.06 0.04 0.02 - 0 —————— — ______ — 220 240 260 280 300 320 340 360 380 400 Wavelength (nm) Figure 2.8. The photolysis of triphosphate 2.5 as monitored by absorbance spectroscopy. Blue line represents the absorbance spectrum of a 6 i.tM solution of 2.5 in water prior to irradiation. Red line represents the absorbance spectrum of a 6 tM solution of 2.5 in water following lh of irradiation at 254 nm. Following one 1 hour of irradiation, the triphosphate 2.5 photolyzed to a new product with an absorbance maximum at 260 nm. Knowing this result, we suspected that the triphosphate 2.5 had photolyzed to unmodified adenine triphosphate, which possesses an absorbance maximum at 260 nm. 2.2.2.1.2 Photolysis of the ribonucleoside 2.6 Further analysis of this novel photolysis mechanism was carried out with the thioether adenosine ribonucleoside 2.6 instead of the triphosphate 2.5 for the following two reasons: 1) the triphosphate 2.5 is difficult to prepare in yields that are high enough for ‘H-NMR spectroscopy and mass spectrometry analyses and 2) that the photolytic conversion of the 122 ribonucleoside 2.6 to the adenosine ribonucleoside would suggest that this novel photolysis is independent of the 5’ ribose phosphate and the 2’ C-H ribose moieties of the triphosphate 2.5. In order to confirm that unmodified adenosine is the product of the photolysis of 2.6, a 40 jiM solution of 2.6 was prepared and irradiated with a 15 mW, 280 nm tunable dye laser. The progress of this irradiation was followed by absorbance spectroscopy (Figure 2.9). 220 240 260 280 300 320 340 360 380 400 Figure 2.9. The photolysis of 40 jiM of ribonucleoside 2.6 with a 280 nm irradiation source as monitored by absorbance spectroscopy. As was seen with 2.5, the irradiation of 2.6 produced a shift in the absorbance maximum, max, of the starting material at 280 nm to a new product with a of 260 nm. Mass spectrometry analyses that were carried out following irradiation was used to verify that unmodified adenosine is the product of this photolysis. A shift in the primary (M+H) signal from that of the starting material, 394.1(394.13 calculated forC15H2oN7O4S)to the (M+H) signal of the natural adenosine nucleoside, 268.1 (268.10 calculated for C,oH14N5O4)was observed. This conclusion was verified by ‘H NMR. 1 mg of 2.6 was dissolved in 600 jiL of MeOH-d4 to give a 5 mM solution that was analyzed by 1H NMR. MeOH-d4 was removed under vacuum 123 and replaced with undeuterated methanol. This sample was irradiated in a quartz cuvette with a 1 mW, 254 nm shortwave mercury handheld UV lamp for 14 hours, enough time to ensure the full photolysis of 2.6. Following irradiation, the final sample was concentrated and redissolved in MeOH-d4 for a fmal analysis by 1H NMR spectroscopy (Figure 2.10). A Reference spectrum for the photolysis product adenosine NMR in DMSO-d6. 254 nm Irradiation of 2.6 NMR in MeOH-d4 Figure 2.10. The photolysis of nucleoside 2.6 as monitored by ‘H-NMR spectroscopy. A. top, Reference spectrum of adenosine in DMSO-d6 obtained from the spectral database for organic compounds (SDBS) (http://www.aist.go.jp/RIODB/SDBS/cgi-binlcre_index.cgi). Peaks present at 7.4, 5.5 and 5.3 ppm correspond to exchangeable hydroxyl protons signals that are not be observed in MeOH-d4. B, middle,1H-NMR spectrum of 2.6 at —5 mM acquired in MeOH-d4 prior to irradiation. Protons corresponding to the 2-(4-imidazolyl) ethyl moiety are shown in red. C, bottom,1HNMR of spectrum B following 14 hours of 254 nm irradiation in undeuterated methanol. ‘H-NMR spectrum acquired in MeOH-d4. The newly incorporated proton corresponding to the C-8 position of adenosine is indicated by the blue arrow. ‘H-NMR analysis of the photolysis product gives a spectrum that is consistent with that of adenosine (Figure 2.1OA). In the irradiated spectrum (Figure 2.1OC), evidence for the adenosine B 8-(2-(4-imi olyl)ethylthio)-riboadenosine 2.6 No frradiatio NMR in Me ___ u_i J. ILZL ppm (fl) 124 product is observed at 8.21 ppm, (blue arrow (Figure 2.1OC)) a signal that corresponds to the C- 2 proton of the nucleobase. The protons corresponding to the 2-(4-imidazolyl) ethyl thiol half of 2.6 are absent (the two aromatic imidazole protons at 6 7.67 ppm and 6 6.93 ppm, as well as the ethyl protons at 6 3.11 and 6 3.59, red arrows in Figure 2.1OB. 2.2.2.2 Solvent isotope dependence for the photolysis of 2.6 The 1H NMR spectroscopy and mass spectrometry data presented for the photolysis of 2.6 suggest that unmodified adenine is produced cleanly upon the photolysis of 2.6. Two major outstanding issues remain regarding this photolysis: The first issue, which will be discussed in the next chapter, was the fate of the 2-(4-imidazolyl)ethyl thiol half of the molecule. The second issue, discussed here is the origin of the C-8 hydrogen that is acquired by unmodified adenosine in the photolysis. It was not suspected that the source for abstracted hydrogen was the ribose sugar, as irradiated and unirradiated ‘H-NIvIR spectra did not differ in the region of Figure 2.10 contained by 6 6.00 ppm and 6 3.60 ppm. We also do not believe that the abstracted hydrogen originates from a disproportionation reaction between two molecules of 2.6 because its photolysis gave a product with a single “max of 260 nm. The remaining candidates for the source of the C-8 proton were the solvent or the departing alkyl group. To investigate the role of the solvent, the photolysis of 2.6 was carried out in both deuterated and undeuterated methanol. A 5 mM sample (‘ 1 mg in 0.5 mL) of 2.6 was prepared in either deuterated or non-deuterated methanol. Both samples were irradiated with a 1 mW, 254 nm shortwave mercury handheld UV lamp for a time long enough to ensure the quantitative photolysis of 2.6. 125 Two aromatic singlets corresponding to the C-2 and C-8 protons of unmodified adenine (at 8.21 ppm, and ö 8.39 ppm respectively) are observed in the photolysis of 2.6 in undeuterated methanol (Figure 2.1 1A), while photolysis in deuterated methanol shows only a single peak corresponding to only the C-2 proton of unmodified adenosine ( 8.21 ppm). The C-8 hydrogen is not observed due to the incorporation of deuterium at this position (Figure 2.1 1B). A coa ciioH Irradiation of 2.6 in Undeuterated Methanol 1MR in MeOH-d4 LdiJ___ Irradiation of 2.6 in MeOH-d4 LU I I I I I I I I I I I I I I I I I I I I I I I I I ppm (fi) 8.0 7. 6.0 50 40 Figure 2.11. The photolysis of nucleoside 2.6 monitored by ‘H-NMR spectroscopy in methanol. A, top, ‘H-NMR spectrum following the photolysis of 5 mM of 2.6 in undeuterated methanol. B, bottom, 1H-NMR spectrum following the photolysis of 5 mM of 2.6 in deuterated methanol. The blue arrow indicates the missing1H-NMR shift that is indicative of the incorporation of deuterium at the C-8 of adenine. High-resolution electrospray mass spectrometry carried out on the irradiated samples in Figure 2.11 was used to corroborate ‘H-NMR suggesting that the abstracted hydrogen originates from the solvent, methanol. The primary (M+H) signal observed for the sample irradiated in undeuterated methanol is 268.1(394.10 calculated forC10H4N504),while the primary (M+H) signal observed for the sample irradiated in MeOH-d4 is 269.1084 (269.11021 calculated for C1oH3DN5O4). 126 2.2.2.3 A CONTROL REACTION. Photo-excitation of unmodified adenine does not promote H/Deuterium exchange on C-8 of adenosine The photolysis data presented in Figure 2.11 suggests that deuterium substitution can be made on C-8 of adenine through the photolysis of imidazole ethyl thioether derivatives of adenine in deuterated water. This reaction is significant if deuterium incorporation at C-8 does not occur on unmodified adenine upon irradiation in deuterated solvents. It is known that proton exchange on C-8 does not occur quickly on adenosine under ambient conditions.”5 The following control was carried out to show that proton exchange does not naturally occur at C-8 of adenine under irradiative conditions: A sample of adenosine was resuspended in MeOH-d4. The C-8 proton peak was stably observed at 6 8.39 despite irradiation for days. These data were corroborated by mass spectrometry. Deuterium substitution can be made on C-8 of adenine through the photolysis of imidazole ethyl thioether derivatives in deuterated solvents. This phenomenon does not occur when unmodified adenosine is irradiated in deuterated solvents. From the lack of irradiative proton exchange at C-8 of adenine, the fact that the proton regions contained by 6 6.00 and 6 3.60 in Figure 2.1 1A and B are similar, and the observation that an exchangeable solvent hydrogen is incorporated into the final adenine photolysis product (Figure 2.11), we contend that the photolysis of 2.6 somehow involves hydrogen abstraction from the solvent and that adenosine is the product of the photolysis of 2.6 in protic solvents. This mechanistic investigation is the subject of the next chapter. 127 2.2.2.4 The photoactivation of adenine in DNA synthesized with the thioether phosphoramidite 2.8 In order to investigate the impact of UV irradiation on thioether-containing oligonucleotides, the phosphoramidite 2.8 was incorporated into the DNA sequence, RT-4-37C; 5’-. TCCCCCCTTTTCTTTT 2.8 2.8 G-3’ through solid phase means. This sequence was purified by denaturing polyacrylamide gel electrophoresis in the dark. The synthesis of RT-4-37C; was verified by MALDI-TOF analysis. The observed mass in the positive mode is 5905.0 (M) while the predicted mass of the DNA sequence (M) 5’- TCCCCCCTTTTCTfTT 2.8 2.8 G — 3 ‘is 5903.2. This value was compared to control sequences 5 ‘-TCCCCCCTTTTCTT’TTG — 3’ (MALDI-TOF 5028.8 (M) (5026.8 calculated)) and 5 ‘-TCCCCCCTTTTCTTTT 2.8 G — 3’ (MALDI-TOF + 5467.0 (M) (5465.9 calculated)). The mass difference between these sequences and the control were 437.8 and 878.5 corresponding to a difference of 1 or 2 incorporations of 2.8 respectively. Following MALDI verification of its synthesis and purification, RT-4-37C was 5 ‘-32P labeled and irradiated for 1 hour with a 254 nm, 1 mW shortwave mercury handheld UV lamp (Figure 2.12). 128 TOP- 20% PAGE Figure 2.12. UV mediated destruction of the DNA sequence 5 ‘-TCCCCCCTTI7CTflTAAG-3’. A, left, The unirradiated DNA sequence 5 ‘-TCCCCCCTTTrCTTTTAAG-3’, where the two A’s indicate positions where 2.8 has been incorporated. DNA B, middle , The result of 1 hour of irradiation from a 1 mW, 254 nm shortwave mercury handheld UV lamp on A. C, right, the sequence, 5 ‘-TCCCCCC1TITCTflT-3’, that was used as a sizing control to gauge the photolysis of lane B. The photolysis of A to B in Figure 2.12 clearly demonstrates that the sequence 5’- TCCCCCCTTTTCTTTTAAG-3’ is photolytically converted to a new product of a given molecular weight. A control experiment was carried out with a DNA sequence containing 8- histaminyl-deoxyadenosine. Following similar irradiation for lh, no photolysis was observed in these control experiments allowing us to conclude that the photolytic nature of RT-4-37C is dependent on the thioether bond incorporated into DNA along with 2.8. <5‘32PTCCCCCCTTTTCHTTG3’ <-- 1 hour irradiation product <5‘32PTCCCCCCTCTTTT3’ A B C 129 2.2.3 DISCUSSION The 4-(imidazolyl-2-ethyl)thio moiety was synthesized with the intention of developing a probe for the structure/activity relationship of modified adenine within the DNAzyme 925-1 it. Knowing that polymerase enzymes can incorporate very small amounts of dATP in repurifled triphosphate 2.5 samples, and diligence in the characterization of anomalous artifacts during DNA preparations led to the application of 4-(imidazolyl-2-ethyl)thio as a photocaging strategy for RNA and DNA. The applicability of this moiety as a caging group has been demonstrated with the imidazole thioether-containing triphosphate 2.5, the imidazole thioether adenosine ribonucleoside 2.6, and DNA synthesized with the S-dA”t’containing phosphoramidite 2.8. 2.2.3.1 The mechanism of photoactivation The mechanism by which a C-H bond is unveiled in adenine involves the rapid and high- yielding photolysis of a weak S-C bond resulting in the conversion of C-8 thioether conjugated adenine into adenine. Because all thioether linked adenine derivatives 2.4 (data not shown), 2.5, and 2.6 photolyze quickly to their respective unmodified adenine components, it is expected that this photolysis mechanism is generalizable to any adenine containing system. The exact mechanism by which photolysis occurs is discussed in Chapter 3 where it is revealed that the photolytic mechanism at 254 nm occurs through a process that involves two photochemical steps: quick thioether dealkylation followed by a slower light-dependent desulfurization. While this two step mechanism is likely to be true at the wavelengths of light used to carry out photolyses discussed in this chapter, the relative rates of these photolyses (dealkylation being faster than desulfurization) may not hold true at all wavelengths of light. 130 We have shown at 280 nm, desulfurization must occur at a rate on par with desulfurization (Figure 2.9) and as such, we exploited this to unmask catalytic activity in a DNAzyme. The source of the proton at C-8 that is incorporated into the adenine product originates from the solvent. This was identified by the quantitative incorporation of deuterium into C-8 of adenosine in photolysis experiments carried out in deuterated methanol. The ability to incorporate isotopes of hydrogen into adenine is perhaps one of the most directly applicable results of this work.”6 The ability to photochemically deuterate or tritiate adenosine at a position that does not rapidly exchange this isotope with a protic solvent leads to what is undoubtedly the easiest known method for generating isotopically labeled adenosine. To isotopically C-8 label the triphosphate, the monophosphate, or the ribonucleoside of adenine, one need simply dissolve 2.4, 2.5, or 2.6 in isotope labeled water and irradiate. Isotopically labeled DNA at specific adenine bases can be made similarly through solid phase syntheses involving 2.8. 2.2.3.2 Contributions to photocaging functiona!ity As an alterative to the orthonitrobenzyl photocage, the 4-(imidazolyl-2-ethyl)thio moiety may address some of the disadvantages associated with the orthonitrobenzyl functionality. The products of imidazole thioether photolysis (imidazoles, adenine, and elemental sulfur (see Chapter 3)) are not expected to display toxicity at the level of the nitroso product that is produced upon orthonitrobenzyl photolysis.99 131 2.3 THE APPLICATION OF IMIDAZOLE THIOETHER ADENINE AS A PHOTOCAGING STRATEGY FOR THE DNAzyme, 17E Knowing that the photolysis of 8-(2-(4-imidazolyl)ethylthioether adenosine gives unmodified adenosine as a product, we proceeded to exploit this novel fmding to the photocaging of DNA. As DNAzymes offer a means for controlling gene expression by catalyzing sequence specific RNA cleavage, we contemplated the possibility of resynthesizing the DNA catalyst,925-lit, as an imidazolyl ethyl thioether containing sequence and photolytically deactivating its RNA cleavage activity with light. However, we were concerned that this application would be complicated by competing non-specific photobleaching phenomena. The fact that photobleaching is well known in the literature deterred us from investigating a process that could be deactivated with light. Instead, we decided that it was more interesting to the scientific community to be able to activate a catalytic system with light. To achieve the light-activation of an oligonucleotide, we chose the unmodified DNAzyme system 8-17 (Figure 2.13). This DNAzyme system was originally selected to be Mg2 dependent by Joyce.43 The variant 17E, a zinc dependent variant containing Joyce’s 8-17 motif was later reselected in 2000 by Lu.51 As it is important to completely mask the background RNA cleavage activity of a photocaged species, we chose to photocage Lu’s system over Joyce’s based on the fact that imidazoles show affinity toward zinc. Any interactions between our imidazole containing thioether functionality and the zinc atoms required for 1 7E’ s activity are expected to further reduce any background activity of the photocaged species. In order to generate a photocaged version of 1 7E with no background activity, we synthesized the 8-(2-(4- 132 imidazolyl)ethylthioether adenosine phosphoramidite 2.8, which was incorporated by standard solid phase methods into four variants of 17E. We chose to synthesize four derivatives of 17E with single substitutions of 2.8 at the 4 separate adenines contained in the active site of 1 7E. Two of these adenines are conserved between Joyce and Lu’s DNAzyme systems (Figure 2.13). m4 17 RS Substrate 3’- GrA -5’ 3’-GTAGGAAGrAkTC.CTCA-5’ I It ill iii —lIt tit iii gt * t —Wts tt I 5’— T —3’ 5’—CATCTCTTCT ATAGTGAGT—3’ C4 A CG A—.-.-...8-17 (Joyce) cAcG 17E(Lu) m m2 Figure 2.13. The catalytic motif of Joyce’s Mg2 dependent DNAzyme 8-17 (left) and Lu’s Zinc dependent 17E motif (right). We chose to photocage four derivatives of 1 7E with single substitutions of 2.8 at the adenines shown by the coloured arrows. Two of these adenines are conserved between the Joyce and Lu DNAzymes (ml and m2). Predicted Watson-Crick pairs are indicated by a dash between complementary bases. 1 7E is shown paired to its all RNA substrate, 1 7RS. We chose to preserve the nomenclature and substrates defined by Lu51 in naming these singly substituted adenine variants so the photocaged catalysts, 17E-m(1 through 4), were named according to the site of substitution in Figure 2.13. The active, unmodified variant, 17E, was also synthesized as a control. This section will describe the photocaging of DNAzyme 8-17E with the imidazole ethyl thioether moiety discussed in section 2.2. 133 2.3.1 MATERIALS AND METHODS 2.3.1.1 Synthesis and purification of substrates and modified catalysts 8(-2-(4-Imidazolyl)ethylthio)-5 ‘-DMT-2 ‘deoxyriboadenosine-3 ‘diisopropyl- cyanoethoxyphosphoramidite 2.8 was incorporated into oligonucleotides as described in Chapter 58 Catalysts 1 7E-A4 were synthesized by the trityl-on method and deprotected in concentrated NH4O at 65 °C for 2 hours. The catalysts 17E-A4 were gel-purified by 20 % D-PAGE and identified by UV shadowing. Only the perimetres of UV absorbing bands corresponding to DNA were visualized by UV shadowing. The interiors of these bands were protected by cardboard such that these interior regions were not irradiated. Only this interior region was extracted from gels. Elutions of catalysts from gels were performed using the crush soak method using 1 % LiC1O4 and 0.7 mM triethylamine as the eluant. The eluant was concentrated, washed twice with ethanol, resuspended in 40 j.iL H20 and passed over a G-25 spin column. This purification ensured that there would be no premature photoactivation of photocaged DNA but resulted in a sacrifice of approximately half of all material. The concentration of catalyst was determined by summing standard values at 260 nm for each unmodified base and using s =14 400 cm’ M’ for modified adenosine. Oligonucleotides were stored at a final concentration of 20 iiM in preparation for kinetic assays. Substrate 17RS was labeled with 10-20 !ICi of {‘y-32P]ATP with 1-10 units of polynucleotide kinase in PNK buffer for 30 minutes at 37 °C, run over a GlO spin column, and purified by 20 % D-PAGE. The gel was briefly exposed and the band corresponding to the radiolabeled material was isolated. Elution of each catalyst from gel was performed using the crush soak method using 1 % LiC1O4 and 0.7 mM triethylamine as the eluant. The eluant was 134 concentrated, washed twice with ethanol, resuspended in 40 iL H20 and passed over a G-25 spin column. The sequences of the catalysts and substrates are as follows: DNA catalysts. 1 7E: 5 ‘-CATCTCTTCTCCGAGCCGGTCGAAATAGTGAGT-3’. 1 7E-m: 5 ‘-CATCTCTfCTCCGAGCCGGTCGAAATAGTGAGT-3’. 1 7E-m2: 5 ‘-CATCTCTTCTCCGAGCCGGTCGAAATAGTGAGT-3’. 1 7E-m3: 5 ‘-CATCTCTrCTCCGAGCCGGTCGAAATAGTGAGT-3’. 17E-m4: 5’-CATCTCTI’CTCCGAGCCGGTCGAAATAGTGAGT-3’. Where A = Imidazolylethyithiol-adenosine. RNA substrate. 1 7RS: 5’ -rArCrTrCrArCrTrArTrArGrGrArArGrArGrArTrG-3’. 2.3.1.2 DNA MALDI-TOF mass spectrometry MALDI-TOF procedures were performed as detailed. A 10 tL solution of oligonucleotide at a concentration of 100 pmol/ tL (100 nM) was added to --20 Bio-Rad AG5OW-X8 cation exchange beads (NH4 form) and allowed to sit for 15 mm at room temperature. 1 p.L of a saturated solution of 3-HPA (Aldrich, 98 %) prepared fresh in 50/50 Acetonitrile and H20 was spotted on a MALDI-TOF plate. Immediately, 1 iL of the NH4 exchanged oligo solution was mixed with the saturated 3-HPA spot. This spot was allowed to dry under vacuum at room temperature prior to MALDI-TOF analysis. 135 2.3.1.3 Kinetic assays Single turnover assays were performed at 24 °C with 5 tM catalyst, 100 ZnBr2 jiM, and trace 5’ 32p labeled RNA substrate 1 7RS. Kinetic experiments were buffered in either 50 mM Hepes HC1 pH 7.0’ or 50 mM of Tris HC1 pH 7.0. The use of Hepes gave poor PAGE resolution of the cleavage product, however the use of Hepes HC1 was mandatory such that results consistant with literature precedent could be shown.5’ Subsequently, it was found that replacement of Hepes with Tris gave clearer visual results in denaturing PAGE analysis and did not affect kinetic results; thus Tris HC1 was preferred as the buffer. In both buffers, the catalyst-substrate interaction was taken to be saturated based on literature precedent.5’Single turnover assays were initiated by adding the catalyst at four times the target concentration (20 jiM) to the substrate mixture containing 66.7 mM of TrisHCl pH 7.0, 133 jiM ZnBr2 and 2 to 20 nM of 5’ 32p labeled RNA substrate 17RS at 24 °C. Aliquots were taken out at desired times and quenched with 95 % formamide containing 1 mM EDTA, bromophenol blue and xylene cyanol to give a final concentration of 75 % formamide and 0.75 mM EDTA. The aliquots were resolved on a 20 % D-PAGE gel. Gels were exposed overnight on phosphor screens and quantitated on a Molecular Dynamics typhoon 9200 phosphorimager. Polygons were drawn around distinct bands. The volume in each polygon corresponding to the product was divided by the sum of the volumes in both the cleaved and the uncleaved polygons to give the fraction of substrate cleavage. First order rate constants (k0bs) for monophasic systems were obtained by fitting data to the single-exponential equation 1 using the Sigma Plot 2001 v7.101 data analysis program. {P]1= [P](1_e0TRt) 1 136 where [Pit, and [P] are the fractions of substrate cleaved at time t and the end point, respectively and k0b is the observed first order rate constant. 2.3.1.4 Photochemical procedures For all photochemical activations, catalysts or nucleosides were diluted to 600 jiL and placed in quartz cuvettes. These cuvettes were degassed with argon for 20 mm before being sealed. Irradiation took place with a handheld 1 mW 254 nm shortwave mercury handheld UV lamp, a 10 mW Xe arc lamp, or a 10 mW tunable dye laser at 280 nm. All sources photolyzed to their respective unmodified adenines. Photolyzed ribosides were transferred to NMR tubes for ‘H NMR spectroscopy analysis while photoactivated 1 7E derivatives were transferred out of quartz irradiation cuvettes and into eppendorf tubes and concentrated for kinetic analyses. 137 2.3.2 RESULTS 2.3.2.1 Investigating the background activity of 17E and its derivatives ml through m4 Newly synthesized 17E variants ml through m4 were assayed for their ability to cleave the all RNA sequence, 1 7RS, in the dark. Unmodified 1 7E was simultaneously assayed as a control. The cleavage gels, kinetic curves, and rate constants are shown in Figure 2.14. A Figure 2.14. 17RS dark cleavage by 17E variants in 100 j.tM ZnBr2, 50 mM pH 7.0 HepesHCl buffer, and <1 nM 5’ 32P of labeled 17RS RNA by variants of 17E at 5 jiM. A, top, Denaturing PAGE gel depicting cleavage efficiencies of modified catalysts in HepesHCl buffered kinetic experiments. B, bottom left, Time dependence of• RNA cleavage for 17E and its derivatives: 17E-ml (s), 17E-m2 (A), 17E-m3 (V), 17E-m4 (o). C. bottom right, Table summarizing the first order rate constants, errors, and activity relative to 1 7E for 1 7E and its derivatives. Substrate Only Cleaved Substrate 17E 17E-ml 17E-m2 17E-m3 17E-m4 .-. u t’J - C ‘J t’3 . CN i—’ ‘.3 4 C k) -. C C C C C C C C C C C C C C C C C C C B C a, C,, a, C) C C C) CC, 20 4J Time (mm) 138 The observed rate constant for the control, 1 7E, is reported as a maximum first order rate constant, kcat STR, based on evidence presented by Lu that at 5 j.tM 17E, 100 jiM ZnBr2 and 50 mM pH 7.0 HepesHCl buffer, the observed RNA cleavage rate by 17E does not increase with an increased DNAzyme concentration.51 The rate constant, kcat sm, measured for the 1 7E control was 0.099 min’, 50 % larger than previously reported by Lu at 0.064 mm4. The observed rate constants for the modified 17E variants are reported in Figure 2.14C. For all photocaged 17E variants ml through m4, RNA cleavage activity is either reduced or nonexistent relative to the 1 7E control. The caged species, 1 7E-ml, displays the most reduced activity of the caged systems. Interestingly enough, this species is modified at an adenosine that is conserved in both Joyce and Lu’s metal dependent systems. Because the species 17E-ml is the only caged species that displays no competing background RNA cleavage in the absence of light, it was chosen for further photoactivation analysis. 2.3.2.2 Mass verification of 17E and 17E-ml synthesis Before the photoactivation of 17E-ml was attempted, the proper syntheses of 17E and 17E-ml was verified by MALDI-TOF mass spectral analysis. The value obtained upon MALDI-TOF analysis of 17E-ml was 10314.0 (M). The predicted mass for the sequence 5’- CATCTCTTCTCCGAGCCGGTCGAAATAGTGAGT-3’,C326H413N1 20201P3S,(where A = Imidazolylethyithiol-adenosine) is 10305.69 for the exact mass and 10310.66 for the molecular 139 + . . + weight (M) . The value obtained upon MALDI-TOF analysis of 17E was 10188.7 (M) . The predicted mass for the sequence 5’ -CATCTCTTCTCCGAGCCGGTCGAAATAGTGAGT-3’, C321H407N1 02P3 is 10179.67 for the exact mass and 10184.48 for the molecular weight of the (M) ion. Both syntheses of 17E and 17E-ml were deemed to be correct as only a 4-9 Dalton differences between measured and calculated masses were observed. 2.3.2.3 The photolytic activation of the DNAzyme, 17E-ml 2.3.2.3.1 Gel and kinetic proof of 17E photoactivation Three light sources were used to analyze the photoactivation of 1 7E-ml. frradiation was carried out with a 10 mW Xe-Hg arc lamp with a band pass filter to cut off radiation greater than 310 nm, a 1 mW 254 nm UV lamp, or a 8 mW dye laser tuned to 280 nm. Although all irradiation sources gave similar results, only the photoactivation results for the handheld UV lamp and the arc lamp are shown in Figure 2.15. 140 A. Ixradiatiou with a 25$ mu Hamfheld Lamp I II lu 17E 10 minliradiation 17E-ml No hradiatian 17E-ml 10 ranliradiation Full Length Substrate I7RS- —,- ‘•— 3— C1eaed Stibstrite 17RS - — — ! — — -u . — — 4_. — B. liradiation with a 310 bandpass firer VI 17E-ml 15mm liradiation Full Length Substrate 1 7RS - Cleaved Substrate 1IRS - 3 3 I I — - - = — — 3:3 - a.C C :3 :3 Figure 2.15. RNA 1 7RS cleavage by 1 7E, caged 1 7E-ml and photoactivated 1 7E-ml. Gel data for single turnover RNA cleavage. Cleavage reactions were performed with 100 mM ZnBr2, 50 mM pH 7.0 TrisHCl buffer, and <1 4 5 32 labeled RNA. Catalysts were resuspended with 1.33 mL of water in quartz cuvettes at 300 riM, flushed with argon for 30 mm, irradiated, dried and incubated with the substrate cocktail to give a final catalyst concentration of 5 jiM. A, top, Irradiation at —1 mW with a 254 rim shortwave mercury handheld UV lamp Left to right: Time course of cleavage for; I) 17E irradiated for 10 mm, II) Unirradiated 17E-ml, and III) 17E-ml irradiated for 10 mm. B, bottom, Irradiation at 10 mW with a 310 nm Band Pass Filtered Arc Lamp. From left to right: Time course of cleavage for; IV) 17E irradiated for 15 mm, V) Unirradiated 17E-ml, and VI) 17E-ml irradiated for 15 mm. To control for photobleaching, 17E was irradiated for the same length of time as photoactivated 1 7E-ml. There is a noticeable photobleaching activity associated with the light sources being used so short irradiation times were preferred. Activity is restored upon irradiation of 17E-ml. Iv 17E 15 rain Irradiation V 1 7E-ml No Irradiation 141 In order to kinetically verify that the 17E-ml is converted to 17E upon photoactivation, data for conditions IV and VI were fit to equation 1 (Figure 2.16). The use of this equation assumes that any irradiative artifacts of 17E-ml and 17E are either extremely slow cleaving or do not cleave substrate 17RS. Condition IV Condition VI Unmodified 17E 15 mm Irradiation 17E-ml 15 mm Irradiation 0.30 0.30 . 0.25 0.25 0.20 020 0.15 7 0.15 / 0.10 • Li.. 0.10 / kcatSi? = 0.096 ± 0.005 mn kcatSTR = 0.11 ± 0.02 mm4 0.05 / [P]00 26.2 % 0.05 [P]00 = 27.7 % 0.00 0.00 0 20 40 60 0 20 40 60 lime (mm) Time (mm) Figure 2.16. Plots of fraction cleaved vs. time for irradiated samples of 1 7E and 1 7E-m 1. The fraction of 1 7RS cleaved was plot against time to give data that were fit to equation 1. The constants and errors returned from fitting these data to the equation [P]1 = [P](1 e0I)0t) using Sigma Plot 2001 v7. 101 are detailed in the inset within each graph. Condition IV and VT refer to conditions described in Figure 2.15. The value of kcat STR obtained in both plots in Figure 2.16 correlates with the value of kcat STR obtained for unirradiated 17E. The value of kcat STR for the control sequence 17E is 0.099 ± 0.007 min1, while the value of kcat STR returned for the 15 mm irradiation of unmodified 17E was 0.096 ± 0.005 min1, and the value of kcat siz returned for the irradiation of 17E-ml was 0.11 ± 0.02 min’. These values agree within error and suggest that 17E is generated from 17E- ml under irradiative conditions. It was necessary to monitor the kinetic profiles on an irradiated version of unmodified 1 7E in order to address the concern that the reduced extent of cleavage ([P]00. equation 1) by photoactivated 17E-ml is due to a photobleaching effect and not due to the thiol imidazole modification on 1 7E-ml. 142 The amount of [ES] complex available for 1 7RS cleavage can be gauged by the value of [P]. Kinetic profiles of unirradiated 17E gave a value of 0.66 ± 0.01 for [P] (see Figure 2.14B) while irradiated 17E returned only 50 % of this value ([P] = 0.26 ± 0.01) indicating that about half of 1 7E had been photobleached. This inactivation is attributed to standard irradiative DNA damage to 17E and 17E-A1 (T-T dimerization etc.). The diminution of [P] on 17E does not remove from the significance of the photoactivation of 17E-A1. The value Of kcat obtained for the irradiated product of 17E-ml suggests that 17E can be generated cleanly. This observation was verified by mass spectrometry analysis. 2.3.2.3.2 MALDI-TOF verification of the irradiative conversion of 17E-ml into 17E A MALDI-TOF analysis was carried out on 17E-ml prior to and following a 30 mm. irradiation with a —1 mW 254 nm shortwave mercury handheld UV lamp. The MALDI-TOF value for 17E-ml prior to irradiation is 10313.6 (M). The value obtained following irradiation is 10185.6 (M+H). These values agree with the predicted masses for 17E-ml, which is 10305.69 (M+H) for the exact mass and 10310.66 (M) for the molecular weight, and the predicted mass of 17E, which is 10179.67 (M) for the exact mass and 10184.48 for the molecular weight. As there is less than an 8 Dalton difference between measured and calculated masses for 1 7E and 17E-ml, this MALDI data confirms the kinetic data which suggests that 17E is produced upon the photo irradiation of 1 7E-ml. 143 2.3.3 DISCUSSION The application of the aryl thioether moiety as a photocaging group has been demonstrated with the DNAzyme 17E. This DNAzyme was synthesized on solid phase with the phosphoramidite 2.8 to give a photocaged variant, 17E-ml, which possesses very little background activity. The DNAzyme 17E-ml can be photoactivated over a period of minutes with a low wattage irradiation source relative to most other photoactivation procedures. Not only is this strategy unique in that thioether-nucleotide caging groups have not been previously reported in the literature, but it is novel because it offers an alternative to photocaging nucleic acids at the nitrogens and oxygens on the Watson-Crick pairing face of the nucleobases. Two reviews of the photocaged 1 7E DNAzyme work described in this chapter”7emphasize the novelty of this protection strategy for the C-8 C-H bond on adenine.87’110 2.3.3.1 Research implications As many ribozymes and DNAzymes can be made through phosphoramidite methodology, the photocaged phosphoramidite 2.8 can be applied to ribozymes and deoxyribozymes in order to photochemically investigate transcription,’ aptamer activity,9’ and secondary and tertiary nucleic acid structure forniation)’8Furthermore, 2.8 can be incorporated into other catalytic nucleic acid systems to photolytically regulate gene expression in RNAi, mRNA’7 and antisense approaches. 144 2.4 CONCLUSIONS Multiple strategies have been developed for caging ATP at the phosphate oxygen9°or at the Watson-Crick pairing face of adenine.’°9The imidazole ethyl thioether strategy presented in this chapter is the only reported strategy for caging adenine at C-8. The photoactivation of imidazole ethyl thioether photocaged adenine nucleosicles is achieved through the localization of an adenine chromophore (max = 280 nm, e28o 14 000 Mcm’), next to a relatively weak sp2 carbon -sp3 sulfur bond. The weak aromatic C-S bond acts as an energy sink for the photoexcited state of adenine as it relaxes back to its ground state. This photocaging strategy is unique in that thioether based nucleotide caging groups have not been previously reported in the literature. Since our publication in 2004,117 two reviews on the topic of photocaging have described this methodology as a rare example of a “traceless” cage, in that our thioether protecting group can be used to cage and ultimately reveal a C-H bond. 86,87 The application of this photocaging group has been demonstrated for the decaging of adenine nucleosides as an imidazole thioether containing triphosphate 2.5, a ribonucleoside 2.6, and a DNA oligonucleotide. Furthermore, the potential for photo controlled gene regulation has been demonstrated in with the DNAzyme 1 7E, where the imidazole ethyl thioether photocaging group can be used to successfully inhibit RNA hydrolysis until exposed to light. 2.4.1 Future work Sometimes a photocaged background activity can be desired. The C-8 adenine protection strategy in this chapter may be advantageous because it introduces a degree of steric bulk at the 145 C-8 position of adenine, yet it does not mask the Watson-Crick pairing face of adenine. It may be of interest to investigate the application of 2.5 as substrates for ATP requiring enzymes. The steric bulk introduced by the C-8 thioether modification of 2.5 may discourage many ATP requiring enzymes from its utilization. However, if enzymes were found that could tolerate the modifications of 2.5, the product of this enzymatic incorporation provides a method for generating a product that may not necessarily act as a substrate for the next reaction in the enzymatic pathway. Thus, enzymes that are capable of utilizing 2.5 provide a method for generating photocaged substrates for the next reaction. New substrates incorporated with any of 2.4, 2.5, or 2.6, may also be photoactivated into a deuterium or tritium labeled aclenine species. One drawback to this thioether photocaging strategy is the observation of a photo bleaching effect on 1 7E associated with the light source used for photoactivation. In this chapter, we have shown that the photoactivation of adenine can proceed with different irradiation sources with different wavelengths of light including a 254 nm shortwave mercury handheld UV lamp, a 10 mW Xe band pass arc lamp, or a tunable dye laser at 280 nm. These data suggest that the activation of aryl thioether caged species can be achieved in a range of light wavelengths. It may be useful to investigate the quantum yields of photoactivation at different wavelengths of light in order to reduce the DNA photo bleaching effect. This may come in useful when applying this thioether caging group in the presence of proteins which are more sensitive to photobleaching. 146 Because of the wide ranging applicability of the ethyl thioether photocaged strategy, the next chapter will detail the mechanism of this photolysis as well as its generalizability to other chromophores and thioether substituents. 147 2.5 REFERENCES 86. Tang, X. J.; Dmochowski, I. J., Regulating gene expression with light-activated oligonucleotides. Mol. Biosyst. 2007, 3, 100. 87. Mayer, G.; Heckel, A., Biologically active molecules with a “light switch”. Angew. Chem. mt. Edit. 2006, 45, 4900. 88. Patchom, A.; Amit, B.; Woodward, R. B., Photosensitive Protecting Groups. J Am. Chem. Soc. 1970, 92, 6333. 89. Martinek, K.; Varfolomeyev, S. D.; Berezin, I. V., Interaction of aipha-chymotrypsin with N-cinnamoylimidazole. Substrate sensitive to light. Eur. J. Biochem. 1971, 19, 242. 90. Kaplan, J. H.; Forbush, B.; Hoffman, J. F., Rapid photolytic release of adenosine 5’- triphosphate from a protected analog - utilization by Na-K pump of human red blood- cell ghosts. Biochemistry. 1978, 17, 1929. 91. Heckel, A.; Mayer, G., Light regulation of aptamer activity: an anti-thrombin aptamer with caged thymidine nucleobases. J. Am. Chem. Soc. 2005, 127, 822. 92. Dorman, G.; Prestwich, G. D., Using photolabile ligands in drug discovery and development. Trends Biotechnol. 2000, 18, 64. 93. Rothman, D. M.; Shults, M. D.; Imperiali, B., Chemical approaches for investigating phosphorylation in signal transduction networks. Trends Cell Biol. 2005, 15, 502. 94. Young, D. D.; Deiters, A., Photochemical control of biological processes. Org. & Biomol. Chem. 2007, 5, 999. 95. Lawrence, D. S., The preparation and in vivo applications of caged peptides and proteins. Curr. Opin. Chem. Biol. 2005, 9, 570. 96. Aizawa, M.; Namba, K.; Suzuki, S., Photo control of enzyme activity of alpha-amylase. Arch. Biochem. Biophys. 1977, 180, 41. 97. Liu, Y.; Sen, D., Light-regulated catalysis by an RNA-cleaving deoxyribozyme. I Mol. Biol. 2004, 341, 887. 98. Adams, S. R.; Kao, 3. P. Y.; Tsien, R. Y., Biologically useful chelators that take up Ca- 2+ upon illumination. I Am. Chem. Soc. 1989, 111, 7957. 99. Park, C. H.; Givens, R. S., New photoactivated protecting groups. 6. p hydroxyphenacyl: A phototrigger for chemical and biochemical probes. I Am. Chem. Soc. 1997, 119, 2453. 100. Hagen, V.; Frings, S.; Wiesner, B.; Helm, S.; Kaupp, U. B.; Bendig, J., [7- (dialkylamino)coumarin-4-yl]methyl-caged compounds as ultrafast and effective long- wavelength phototriggers of 8-bromo-substituted cyclic nucleotides. Chembiochem. 2003, 4, 434. 101. Lu, M.; Fedoryak, 0. D.; Moister, B. R.; Dore, T. M., Bhc-diol as a photolabile protecting group for aldehydes and ketones. Org. Lett. 2003, 5, 2119. 102. Il’ichev, Y. V.; Schworer, M. A.; Wirz, J., Photochemical reaction mechanisms of 2- nitrobenzyl compounds: methyl ethers and caged ATP. J. Am. Chem. Soc. 2004, 126, 4581. 103. Niggli, E.; Lederer, W. J., Restoring forces in cardiac myocytes. Insight from relaxations induced by photolysis of caged ATP. Biophys. J. 1991, 59, 1123. 148 104. Engels, J.; Schlaeger, E. J., Synthesis, structure, and reactivity of adenosine cyclic 3’,5’- phosphate benzyl triesters. J. Med. Chem. 1977, 20, 907. 105. Mccray, J. A.; Herbette, L.; Kihara, T.; Trentham, D. R., A new approach to time- resolved studies of ATP-requiring biological-systems - laser flash-photolysis of caged ATP. Proc. Nati. Acad. Sci. USA. 1980, 77, 7237. 106. Monroe, W. T.; McQuain, M. M.; Chang, M. S.; Alexander, J. S.; Haselton, F. R., Targeting expression with light using caged DNA. J. Biol. Chem. 1999, 274, 20895. 107. Ando, H.; Furuta, T.; Tsien, R. Y.; Okamoto, H., Photo-mediated gene activation using caged RNA/DNA in zebrafish embryos. Nat. Genet. 2001, 28, 317. 108. Chaulk, S. G.; MacMillan, A. M., Caged RNA: photo-control of a ribozyme reaction. Nuc. Acids Res. 1998, 26, 3173. 109. Hobartner, C.; Silvemian, S. K., Modulation of RNA tertiary folding by incorporation of caged nucleotides. Angew. Chem. mt. Edit. 2005, 44, 7305. 110. Tang, X.; Dmochowski, I. J., Phototriggering of caged fluorescent oligodeoxynucleotides. Org. Lett. 2005, 7, 279. 111. Krock, L.; Heckel, A., Photoinduced transcription by using temporarily mismatched caged oligonucleotides. Angew. Chem. mt. Edit, 2005, 44, 471. 112. Stensiö, K.-E.; Wahlberg, K.; Wabren, R., Synthesis of brominated imidazoles. Acta. Chemica Scand. 1973 27, 2179. 113. ilcehara, M.; Uesugi, S., Studies of Nucleosides and Nucleotides .53. Purine Cyclonucleosides .18. Selective Tosylation of Adenine-Nucleotides - Synthesis of 8,2’- Anhydro-8-Mercapto-9-Beta-Arabinofuranosyladenine 5’- and 3 ‘,5-Cyclic Phosphate. Tetrahedron. 1972, 28, 3687. 114. Hoard, D. E.; Ott, D. G., Conversion of mono- and oligodeoxyribonucleotides to 5’- triphosphates. I Am. Chem. Soc. 1965, 87, 1785. 115. Shelton, K. R.; Clark, J. M., Jr., A proton exchange between purines and water and its application to biochemistry. Biochemistry. 1967, 6, 2735. 116. Chirakul, P.; Litzer, J. R.; Sigurdsson, S. T., Preparation of base-deuterated 2 ‘- deoxyadenosine nucleosides and their site-specific incorporation into DNA. Nucleos. Nucleot. Nuci. 2001, 20, 1903. 117. Ting, R.; Lermer, L.; Pen-in, D. M., Triggering DNAzymes with light: a photoactive C8 thioether-linked adenosine. J. Am. Chem. Soc. 2004, 126, 12720. 118. Wenter, P.; Furtig, B.; Hainard, A.; Schwalbe, H.; Pitsch, S., Kinetics of photoinduced RNA refolding by real-time NMR spectroscopy. Angew. Chem. mt. Edit. 2005, 44, 2600. 149 CHAPTER 3: CLEAN ALKENE GENERATION FROM ARYL-THIOETHERS 3.1 INTRODUCTION 3.1.1 Alkene generation in organic chemistry The alkene functional group finds many important applications in synthetic and industrial chemistry. Alkenes can be easily and diversely functionalized for use in total syntheses, or alkenes can be polymerized to generate polymers of high practical value in industrial applications. Numerous methods for alkene syntheses exist,119 for example, they can be synthesized from; alcohols, halides, and amines through elimination reactions; from aldehydes and ketones through phosphonium ylides and suiphones; from alkynes through their selective reduction; and through olefin metathesis reactions which offer a means of constructing new alkenes from existing ones. With the numerous methods available for the generation of alkenes, factors such as yield, reactivity under atmospheric conditions, and reactivity in mild protic solvents would justify the use of a certain olefin preparation over another. Phosphonium ylide’2° and metathesis reactions121 for example, display tolerance to mild aqueous conditions and are often preferentially employed over halide elimination or radical based selenium methods for alkene generation,’22 where reagents or products can be intolerant of oxidants, bases, and high temperatures. There are three other factors that would cause one to favor a new method of 150 alkene generation over existing methodologies. These are: 1) that the new method has advantages in inducing reactivity in states of matter where existing methods cannot, for example, under intracellular, solid, or gel conditions, 2) that the method gives high yields with environmentally safe byproducts, and 3) that the method is specific enough to allow for multiple subsequent reactions to be carried out in one pot. 3.1.2 Radical generation of alkenes Radical means for the generation of alkenes include Barton olefination, Norrish reactions, and oxidative selenium based Grieco chemistry.’22Unfortunately, the production of alkenes through radical methods is usually low yielding, mainly due to the non-specific reactivity between radical reaction intermediates and new chemical functionalities. Furthermore, the pursuit of new radical methods for the generation of alkene functionality is deterred by the hazardous nature and environmental costs of working with typical peroxides and tin radical initiators/reagents’23in organic solvents. 3.1.3 Photolytic generation of alkenes The development of new radical methods for alkene generation is still attractive in cases where light is used to initiate chemistry. Light can be superior to other reactants for reasons that include; its relatively low cost, the ease with which we can control its flux, and the lack of a need for the clean-up of light as a reaction byproduct. The relative transparency of light towards matter allows it to be employed in many instances where the introduction of a chemical reagent could not be used. For example, chemistries carried out in intracellular, solid, or gel states require a method of non-destructively introducing a reagent, a requirement met by light. 124 These unique properties of light continue to draw interest in developing novel 151 photochemical transformations that are often plagued by both low quantum yields and low chemical yields. 3.1.4 Photolysis of sulfoxides and thioethers Considering that the photolyisis of the C-S bond can be exploited due to its weak bond energy (—65 kcal/mol), it is interesting that the utility of C-S containing molecules in photochemistry is not more pronounced. The following two examples in Figure 3. offer some insight into why radical C-S photolyses may be under-utilized in organic chemistry. Photochemical strategies have been described by for the generation of new products from alkyl aryl sulfoxides’25in an attempt to draw analogy between photochemical a-cleavage in carbonyls and sulfoxides. At 267 nm, phenylsulfoxides were irradiated in different solvents to give complex mixtures of products (See Figure 3 .A). Jenks was able to conclude that sulfoxides are more susceptible to photochemical ct-cleavage than carbonyl analogues. Despite this conclusion, the application of Jenks’ photolabile sulfoxides to synthetic methodolgy is deterred by low chemical yields and multiple side products. Grimme et al.’26 also saw similar complex product distributions while investigating the photolysis of naphthylthioethers in studies designed to determine whether the photochemical mechanism of S-C bond cleavage occurs as the primary photochemical step through thermally active crossing of singlet and triplet tn states into a dissociative itcy* ones (See Figure 3.B).’26 152 A 9 hv =267 nm (18%) HO t-BuOH (25%) (8%) ÷ 9 + I I (22%) (5%) HO- II I + (22%) + 0 (4%) (34%) (12%) hv =267 nm t-BuOH --s hv =405 nm MeOH B (15%) QSSJDJ (3%) SH cc + + coy Figure 3.1. Reagents and products of the photolysis of thiol containing compounds as investigated by A, top, Jenks125 and B, bottom, Grimme’26 (yields in brackets). There are examples in the field of photolytic sulfur chemistry where high yielding photolyses are reported. In 1997, Kutateladze reported that a-(methylthio)ethylacetate could photochemically transfer ethylacetate to alkenes to generate alkylesters (Figure 3 .2A).’3 Kutateladze reported that this transfer gave high yields, was insensitive to oxygen, did not require a special reducing agent, and was compatible with polar/protic functional groups and solvents. 153 A+ (64%) 0 hv =300-578 nm 0 + H0 MeOH (73%) + (48%) B - hv=254nm ( o) MeOH ço (67%) Figure 3.2. Reagents and products generated in high yield from the photolysis of thiol containing compounds as investigated by A, top, Kutateladze’23and B, bottom, Kropp’27 (yields in brackets). In another high-yielding application, photolytic olefin generation has been described using alkyl phenyl thioethers. Kropp et al. were able to photochemically synthesize 2-norbomene from 2-norbornyl tbioether in 58 % yield, and dihydropyran from 2-tetrahydropyranyl thioether in 67 % yield (Figure 3.2B).’27 In an effort to classify photolytic C-S bond cleavage as a caged radical or a caged ionic photoprocess (Figure 3.3A), Kropp’s goal was to identify either tetrahydropyran or 2-methoxy tetrahydropyran to verify the respective radical or ionic nature of the photolytic C-S bond cleavage (Figure 3.3B). 154 A. R Y Radical Photoprocess e e R Y Ionic Photoprocess (67%)sQ] MeOH LS Ionic Photoprocess Figure 3.3. Kropp’s127 characterization of the radical/ionic nature of photolytic C-S bond cleavage. A, top, Proposed homolytic (top) or heterolytic (bottom) mechanisms of C-S bond photolyses. B, bottom, Predicted mechanistic products as applied to the photolysis of 2-S-phenylthioether tetrahydropyran. Homolytic photolysis would give rise to tetrahydropyran while heterolytic photolysis would give rise to 2-methoxy tetrahydropyran. Because dihydropyran was identified as the major product, neither the speculated radical or ionic C-S bond photolysis mechanism could be verified as the major route by which product is •• 1 _____ C-S Electron Transfer reel - LR yj B . H Radical Photoprocess a e 0 155 generated (Figure 3.3). Despite reporting the yield in the materials and methods, Kropp gave no mention of the high yielding alkene product in the discussion or the conclusion.’27 A more thorough report detailing the synthetic potential of Kropp’s result has not yet been published. Based on the high photochemical yield for alkene generation reported by Kropp, it is likely that alkene generation can be achieved efficiently from photolytic thioether methods. The high yielding nature of Kropp’s chemistry has influenced our desire to fully characterize the photolysis of imidazole ethyl thioethers of adenine and its related thioethers. 3.1.5 Proposed Research In the previous chapter, a novel aryl thioether photolysis reaction was identified. The moiety, 8- (2-(4-imidazolyl)ethyl- 1 -thio)-adenine, was investigated as a means of photocaging different ribosides of adenine.”7Although not exhaustive in nature, this initial characterization provides a starting point from which we can study the mechanism and application of this photolysis. The following points regarding aryl thioether photolysis remain unanswered by the work presented in the preceding chapter and form an experimental basis for the continued investigation of this photolysis: 1) The nature of the photolysis product(s), any possible intermediates, and the fate of sulfur had not been accounted for. 2) The specificity of this photolysis with respect to an adenine chromophore and whether it could be extrapolated to other chromophores 3) The mechanism of photolysis 4) The application of this photolysis towards the syntheses of other alkene products 156 5) The quantum yield of photolyses. This chapter will be divided into five sections which serve to delineate this novel photolysis and probe the generalizability of this photolysis to other heterocycles for the generation of alkenes. Section 3.2 will investigate: 1) the mechanism of photolysis in terms of intermediates that are produced and the identity of the non-adenine product that is generated. 8-S-Mercaptoethyl-9- ethyl-adenine, 3.1 (vide infra Table 3. for structures), was synthesized as a minimal photolyzable structure and photolyzed in order to identify intermediates and products of thioether adenine photolysis. Section 3.3 will investigate: 2) the generalizability of this photolysis to non-adenine chromophores (i.e. benzimidazole containing systems), through the photolysis of 2-ethylthio benzimidazole, 3.2. Section 3.4 will attempt to determine the mechanism of photolysis through the photolysis of the strained compound 2, N-cyclothioethyl-benzimidazole 3.3. Section 3.5 will investigate the application of this novel photolysis mechanism to the generation of other alkene products through the photolysis of 3.4, 3.5, and 3.6. Section 3.6 details how the quantum yields of the photolyses in this chapter were calculated. 157 The following table, Table 3., provides all the data regarding the compounds that are photolyzed in this chapter. It is recommended that this table be detached and kept aside for reference, as the structures contained within this table are referred to often in this chapter. 158 Qu an tum M ol ar Ef f Qu an tum Qu an tum Qu an tum e th e r In te rm ed ia te R el at iv e to In te rm ed ia te A lk en e Pr od uc ed to Pr od uc t 2 54 (cm 1M 1) 3 1 25 4 nm (4 ) 3 1 25 4 nm V t2 54 ) 60 10 0. 00 7 ± 0. 00 1 1. 0 3.1 72 30 0. 02 4 * 0. 00 4 3.6 44 47 0. 01 2± 0. 00 2 1. 8 I N H 2 + 3.1 -I + 2- M er ca pt ob en zi m id az ol e 3.3 -I 0, 00 01 8 ± 0. 00 00 2 1 0. 00 7 ± 0. 00 1 39 0, 00 18 ±0 .0 00 4 10 — 73 05 0. 02 6± 0. 00 4 3.9 Et he ne Et he ne + 1- B ut en e Ie 1- B ut en e + an d E- B ut en e + 6- C yc lo he xe no ne — o E 0 C D - I C D 0 C D — C D CD 0 0 . + ) C) C D 0 I’; CD ç. C) C)CD 0 og _+)Cl) o. ‘tip.) •CD C) p.) C) N H 2 3 .1 -P B en zi m id az ol e 3. 3- P B en zi m id az ol e B en zi m id az ol e B en zl m id az ol e 0. 02 6 ± 0. 00 3 39 N H . 2- M er ca pt ob en zi m id az ol e 10 88 0 0. 00 6 ± 0. 00 1 0. 8 0. 00 4 ± 0. 00 1 24 . 0. 00 43 ± 0. 00 07 24 - 0. 00 2 ± 0. 00 1 13 3.2 THE INTERMEDIATES AND PRODUCTS OF THIOETHER PHOTOL YSIS In Chapter 2, it was identified that the aryl thioether moiety could be used to photocage different ribosides of adenine. One product of this photolysis, adenine, was clearly identified, but the other product of photolysis was not. In order to identify other products and intermediates that arise from aryl thioether photolysis, compound 3.1 was synthesized on a large scale and photolyzed as a minimal photolyzable aryl thioether structure. Compound 3.1 has ethane as its N-9 substituent, thus its photolysis serves to identify the importance of the ribose substitution at N-9 of adenine between this photolysis and the photolyses studied in Chapter 2. The analysis of the photolyses of a thioether containing adenosine, 2.6, in the previous chapter saw the production of a minor intermediate with a 2max at 310 rim (Chapter 2, Figure 2.9). 3.1 was synthesized and photolyzed on a large scale in order to identify both the non-adenine product of this photolysis and an intermediate of photolysis through ‘H-NMR spectroscopy and x-ray crystallography. 3.2.1 MATERIALS AND METHODS 3.2.1.1 General synthetic methods All chemicals were purchased from Sigma, Aldrich, or Acros Organic. Deuterated solvents were purchased from Cambridge Isotope Laboratories. Thin layer chromatography Rf’s are reported on Silica Gel 60 F254 Glass TLC plates from EMD Chemicals. 160 All ‘H Nuclear magnetic resonance (NMR) spectra were recorded at room temperature on a Bruker Avance 300 or 400 MHz instrument. Chemical shifts are reported using the 6 scale in ppm and all coupling constants (J) are reported in hertz (Hz). ‘H NMR spectra are referenced to the tetramethylsilane peak (6 = 0.00). Mass spectrometry was performed at the mass spectrometry lab of the University of British Columbia (U.B.C.) chemistry department. 3.2.1.2 Specific synthetic procedures 3.1. 8-Mercaptoethyl-9-ethyl-adenine was synthesized through the following synthetic scheme. NH2 NH2 NH 12 NN AcOcONa NaSCH N N 8-Bromo-9-ethyl-adenine 8-Bromo-9-ethyl-adenine. N-9-Ethyladenine 3.1-P (0.53g, 3.248 mmol), 17.5 M glacial acetic acid (11 mL), and 1.25 g (15.26 mmol, 4.70 equiv.) of CH3O2Na were added to a 50 mL round bottom flask equipped with a stir bar. The pH of the resulting solution was verified to be 4. 0.54 mL (10.43 mmol, 3.21 equiv.) of Br2 was added to the mixture and TLC with a running solvent of 5 % MeOH/CHC13v/v was used to follow the course of the reaction. Multiple additions of Br2 were required in order to drive the reaction to completion. After 1 hr, Br2 (0.54 mL, 3.21 equiv., total 6.42 equiv.) more was added via syringe. After 1.5 hrs, an additional 0.28 mL (1.6 equiv., total 8.04 equiv.) was added. After 3 hrs, unwanted starting material was extracted with EtOAc (30 mL), washed with 10% NaHSO3 (3 x 20 mL), then was washed with sat. NaHCO3 (2 x 20 mL). The aqueous washes were pooled and 8-Bromo-9-ethyl-adenine 161 was extracted with CHC13 (3 x 50 mL). The combined CHC13 extracts were dried over NaSO4, filtered, and concentrated. The cOncentrate was dissolved in a minimal amount of CHC13 and loaded onto a silica column. Elution of 8-bromo-9-ethyl-adenine was achieved with an increasingly polar MeOH/CHC13gradient (500 mL CHC13,200 mL x 2 % MeOH/CHC13v/v, 400 mL 5 % MeOH/CHC13). Concentration of the appropriate pooled fractions resulted in 0.29 g (37 % yield) of the desired product 8-bromo-9-ethyl-adenine. ‘H NMR (CDC13,300 MHz): 8.30 (s, 1H, CH Ar3), 5.92 (s, 2H, CNH2), 4.27-4.24 (q, J = 7.2 Hz, 2H, NCH2), 1.44-1.39 (t, J 7.2 Hz, 3H, CH3) LRMS (ES): (M+H) m!z found: 242.0, 244.0, calculated forC7H9BrN5:242.0041 (100.0 %), 244.0021 (97 %) 3.1. 8-S-Mercaptoethyl-9-ethyl-adenine. Sodium ethane thiolate (0.69 g, 8.20 mmol) was added to a stirring solution of 8-Bromo-9-ethyl-adenine (0.10 g, 0.413 mmol) in 100 rnL of MeOH. After 24 hrs, the reaction was concentrated, suspended in 50 mL ofH20, and 3.1 was extracted with CHC13 (3 x 30 mL). The combined CHC13 extracts were dried over anhydrous Na2SO4,filtered, and concentrated. The resulting residue was dissolved in a minimal amount of CHC13 and purified by flash chromatography with an eluant gradient ranging from 2% MeOH: CHC13 to 10 % MeOH: CHC13.A mass of 0.06 g (65 % yield) of 3.1 was obtained. 1H NMR (MeOD, 300 MHz): 8.09 (s, 1H, CHAr2),4.18-4.11 (q, J= 7.2 Hz, 2H, NCH2), 3.34- 3.29 (q, J= 7.3 Hz, 2H, SCH2), 1.46-1.41 (t, J= 7.3 Hz, 3H, CR3), 1.36-1.32 (t, J= 7.2 Hz, 3H, CR3). ‘3C NMR (MeOH-d4, 400 MHz): 122.9, 101.0, 97.1, 27.1, 15.1. HRMS calculated for C9H14N5S(M+H4): 224.0970. Found: 224.0967. UV/Vis in MeOH. 2.max: 279.0 nm, 8279: 1.8 x i0 M’ cm1. 8254: 6.0 x l0 M’ cm’. 162 3.1-I. 8-Mercapto-9-ethyl-adenine. A solution of 3.1 was irradiated, concentrated, and then loaded onto a flash chromatography column. The product 3.1-I was isocratically eluted with 97:3 CHCI:MeOH. The appropriate fractions were collected and crystals of 3.1-I were grown from boiling water. The value calculated for the HRMS analysis ofC7H1ON5S(M+H) is 196.0651. The value found was 196.0657. 3.1-P. 9-Ethyl-adenine. The title compound was synthesized twice, both photolytically and chemically. In the chemical synthesis, a 2 L flame dried round bottom flask was charged with a stir bar, 2 g (14.8 mmol) of adenine, and 500 mL of DMF that had been dried over 4A molecular sieves. The stirred yellow mixture was flushed with N2 gas for 5 mm before 30 mL (60 mmol, 4.05 equiv.) of a 2.0 M tBuLi solution in cyclohexane was added. The reaction was allowed to stir at room temperature for 1.5 Irs before 2.4 mL (30.0 mmol, 2.02 equiv.) of ethyl iodide was added via syringe. Product formation was observed by TLC with a 20 % MeOH/CHC13 developing solution. The product 3.1-P has a Rf value of 0.40, two minor products are present with Rf values of 0.50 and 0.60 respectively. After 16 hours, DMF was removed by rotary evaporation. The residue was resuspended in MeOH and concentrated under vacuum twice in order to facilitate the quantitative removal of DMF. The resulting residue was suspended in a saturated solution of NaHCO3 (100 mL) and extracted with CHC13 (4 x 100 mL). The combined CHC13 extracts were dried over anhydrous Na2SO4, filtered, and concentrated. This concentrate was dissolved in a minimal amount of CHC13 and loaded onto a silica column. The product 3.1-P was eluted with a MeOH/CHC13gradient (400 mL x 2 % MeOH/CHC13,400 mL 5 % MeOH/CHC13,800 mL 10 % MeOH/CHC13v/v). Concentration of the appropriate pooled fractions resulted in 1.10 g (46 % yield) of the desired product 3.1-P. 163 To synthesize 3.1-P photolytically, a solution of 3.1 was irradiated at 254 nm to give the product 3.1-P. The resulting residue was loaded onto a flash chromatography column. The product 3.1-P was eluted with a 10 % MeOH/CHC13gradient. Both syntheses gave identical spectral analyses data. ‘H NMR (CDC13,300 MHz): 8.35 (s, 1H, CH Ar2), 7.81 (s, 1H, CH Ar8) 5.80 (s, 2H, CNH2), 4.28-4.21 (q, J= 7.3 Hz, 2H, NCH2), 1.55-1.39 (t, J 7.3 Hz, 3H, CH3). HRMS calculated for C7H,0N5(M+H): 164.0931. Found: 164.0938. UV/Vis in MeOH: 2max: 261.0 fl, 626,: 1.12 x l0 M’ ciii’. 3.2.1.3 Photochemical procedures The photolyses of the thioethers reported in this chapter proceed efficiently at sub-molar concentrations. For photochemical reactions that were monitored by absorbance spectroscopy, argon degassed solutions were prepared at concentrations of no more than 0.2 absorbance units at 254 nm in a Heilma Quartz Suprasil cuvette. This cuvette was degassed with argon for 20 mm before being sealed and fastened onto the glass filter of hand-held UVP UVGL-55 254 nm shortwave mercury lamp. Irradiation was allowed to take place for a given amount of time. Time points were monitored by removing the cuvette from the UV lamp and placing it in the spectrophotometre. The spectrophotometre possessed insufficient photon intensity to affect the photochemical reaction. Furthermore, a photolysis intermediate with a of 310 nm was produced upon photolysis. This intermediate is stable in the absence of 254 nm light and could be analyzed 164 spectrophotometricaly, transferred back to the UV lamp, and irradiated until the next time point without intennediate decomposition. For photochemical reactions that were monitored by ‘H-NMR, 0.5 to 1 mL solutions in deuterated solvent at -5 mg/mL concentrations were degassed with argon then transferred to a Norell S-5-500-QTZ quartz NMR tube. This tube was fastened onto the glass filter of a hand- held UVP UVGL-55 254 nm shortwave mercury lamp. Irradiation was allowed to take place for a given time before the quartz NMR tube was transferred to the NMR instrument for analysis. 165 3.2.2 RESULTS In the previous chapter, we discussed the photolysis of a thioether modified heterocycle, 8-(2- (4-imidazolyl)ethyl-l-thio)-2’deoxyriboadenosine.”7Photolysis of this species was originally achieved with a 10 mW dye laser at 280 nm. The appearance of an intermediate with an absorbance at 310 nrn was observed but not discussed due to its transient nature (Chapter 2, Figure 2.9). Two complications were associated with the light source used in the initial report: 1) the generation of a monochromatic, 280 nrn beam of light required a frequency doubling crystal in conjunction with a tunable dye laser, equipment to which access was limited and 2) in order to isolate and characterize intermediates generated by photolysis, a less intense light source was needed. These concerns regarding the light source were resolved through the photolysis of targets in a quartz cuvette using a hand-held 254 nm UVP UVGL-55 shortwave mercury lamp. This light source will be used to identify the intermediate and products of thioether photolyses in this chapter. 3.2.2.1 The intermediate of photolysis. Photolysis of 3.1 proceeds through an 8- mercapto adenine intermediate. The prototype adenine derivative, 8-mercaptoethyl-9-ethyl-adenine 3.1 was synthesized as a surrogate for the thioether adenosine species initially reported in Chapter 2 for in-depth photolysis experiments. Compound 3.1 was irradiated as a minimal photolyzable structure in order to verify the importance of the ribose substitution on N-9 of adenine as well as the need for the imidazole functionality on the thioether component. Figure 3.4 shows the absorbance changes of a solution of 3.1 irradiated in MeOH at 254 nm over time. 166 A_____________ B _____________ 0.6 3.1 3.1-1 0.6 3.1-I 220 240 260 280 300 320 340 360 380 400 220 240 260 280 300 320 340 360 380 400 Wavelength (nm) Wavelength (nm) Figure 3.4. Spectrophotometric absorbance profile for photolysis at 254 am of a 26.5 i.tM solution of 3.1 in 1120. Because the photolysis for the conversion of 3.1 to 3.1-I is much faster than the photolysis of 3.1-I to 3.1-P, isosbestic points could be identified within the spectra. A. left, data to 7 mm detailing the conversion of 3.1 (green) to 3.1-I (blue). Plots are shown at 0.5 to 1 min intervals to highlight isosbestic points indicated by the blue arrows. B, right, data from 7 mm onwards detailing the conversion of 3.1-I (blue) to 3.1-P (red). Red arrows indicate isosbestic points corresponding to the transformation of 3.1-I to 3.1-P. Plots are shown at 1 to 5 mm intervals to highlight isosbestic points indicated by the red arrows. The kinetic profiles of the photolysis of 3.1, as monitored by absorbance spectrophotometry, produced a quantitative shift in the absorbance maximum from 280 nm to 310 nm after 6 mm of irradiation (isosbestic points for this conversion are shown by blue arrows in Figure 3.4A). Following irradiation for 51 mm more, an absorbance shift to 260 nm, the characteristic absorbance of adenine, was observed (isosbestic points for this conversion are shown in red arrows in Figure 3.4B). The UV-vis absorbance profile of the product 3.1-P suggests that the final photolysis product is 9-ethyladenine. ‘H-NMR evidence is later used to confirm this. The identity of the intermediate was more mysterious. Data thus far suggest that the intermediate, with an absorption maximum at 310 nm, is stable in the absence of UV light. A large scale irradiation of 3.1, was undertaken in order to identify this intermediate. 167 3.2.2.2 The intermediate of photolysis. Photolysis of 3.1 proceeds through an 8- mercapto adenine intermediate 3.1-I. In order to identify the intermediate 3.1-I corresponding to the 310 nm absorbance maximum in Figure 3.4, a large scale photolysis was carried out on 3.1. This intermediate is stable in the absence of 254 nm light, could be isolated by flash chromatography, and was crystallized. Crystallographic and absorbance spectroscopy data of the isolated material show a structure and UV/Vis absorbance maximum, max, at 310 nm that suggest that this intermediate is 8- mercaptoadenine (Figure 3.5). 168 A. B. 0.10 0.08 0.06 6, 0.04 0.02 0.00 220 Figure 3.5. Intermediate crystallized from the irradiation of 3.1 in MeOH. A) top. ORTEP structure obtained from x-ray diffraction studies. B) bottom. TJV/Vis absorption spectra of the crystallized intermediate 3.1-I (Data courtesy of Jen Steele). H IIJ 117 NI cr C2 240 260 280 300 320 340 360 380 400 Wavelength (nm) 169 3.2.2.3 The product of photolysis. Ethene isevolved through the photolysis of 3.1. ‘H-NMR studies on the photolysis of 3.1 suggest that the fate of the ethyl component of the thioether in 3.1 is ethene. CH,OH CH,OH Spectrum of ethene in MeOH-d4 t =192 — t=24h ,I.JJ I t = 1 h —_I L_______ I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 80 70 60 50 40 30 2 10 ppm 01) Figure 3.6. NMR spectral visualization of the photolysis of 0.04 M 3.1 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 4.2 mg of 3.1 in 500 tL MeOH-d4, Arrows in the t = 1 h spectrum indicate NMR signals corresponding to the fonnation of 3.1-I, arrows in the t= 24 h spectrum indicate NMR signals corresponding to the formation of 3.1-P, and the green arrow in the top spectrum indicates the shift of an ethene standard in MeOH-d4. Times at which spectra were acquired are indicated to the left of spectra. Ethene had a shift of 5.39 ppm, this value matched the respective peak at 1 hour and 24 hours. No ethene was observed in the 192 hour spectrum. The lack of signal was attributed to volatilization of ethene that had occurred due to the lack of an air-tight seal on the Nivik tube. ‘H-NMR data of the products and intermediates of photolyses compare well with authentic standards confirming that the photolysis intermediate, 3.1-I, is 8-mercaptoadenine, and that the photolysis product 3.1-P, is 9-ethyladenine (see appendix for NMR spectra of authentic 170 references). An authentic sample of ethene dissolved in MeOH-d4 was prepared through the sonicating of MeOH-d4 in an NMR tube that was capped with an ethene filled balloon. The presence of ethene in this ‘H NMR spectrum is shown by the green arrow in Figure 3.6. The appearance of a peak at 6 5.4 following irradiation of 3.1 suggests that ethene is generated from the photolysis of 3.1. Based on isotope incorporation experiments in chapter 2, it is hypothesized that the proton incorporated into 3.1-P derives from solvent. 3.2.3 DISCUSSION Chapter 2 detailed the use of an alkylthioether as a photocaging group for adenine. In this section, strong evidence is presented towards this photolysis being a two step process. X-ray crystallographic, ‘H-NMR and UV-Vis absorbance spectroscopy data presented in this section suggest that the products of the first photolysis step is 8-Mercapto-9-ethyl adenine, 3.1-I, and ethene. Further photolysis of the intermediate, 3.1-I, involves the replacement of a heterocyclic C-S bond with a C-H bond to give 9-ethyl adenine 3.1-P. It is assumed that the proton incorporated into C-8 of 9-ethyl adenine derives from the protic solvent, as was identified in Chapter 2. The product, 3.1-P, was verified by chemical correlation, where 9-ethyl-adenine was synthesized orthogonally through the reaction of the N-9 lithium salt of adenine with ethyl iodide. The scheme in Figure 3.7 for the photolysis of 3.1 summarizes these data. 171 hv 2:\sH ) ) + Sulfur ) Figure 3.7. Proposed general pathway for the photolysis of adenine 3.1. The first step in the photolysis of 3.1 involves photolytic cleavage of the thioether resulting in ethene and 8-mercapto-9-ethyl adenine, while the second step involves the replacement of a heterocyclic C-S bond with a C-H bond to give 9-ethyl adenine. Both processes proceed quickly at 280 nm (see Chapter 2, Figure 2.9), so irradiation with a low wattage source at wavelengths removed from the absorbance maxima had to be employed to properly visualize the two step process. From the quantum yields of photolysis presented for 3.1 (Table 3. (vide infra)), it is clear that desulfurization occurs slower than deallcylation at 254 nm. This may not necessarily hold true when other wavelengths of light are used as the irradiation source. 3.2.3.1 On the mechanism of photolysis 3.2.3.1.1 Regarding N-9 adenine, or N-Benzimidazole substitution The rapid photolysis of 3.1 suggests that this photolysis is independent of the phosphate or sugar moieties seen on thioether adenosines in Chapter 2. The ethyl moiety on 3.1 can be substituted with ribose, deoxyribose, and 2’-deoxyribo-5’-monophosphate without a major effect on the rate of photolysis. In simplifring the minimal photolyzable unit to 3.1, the ‘H NMR spectrum was also simplified such that the fate of the non-adenine fragment could be easily identified. 3.2.3.1.2 On the generation of an alkene 172 The ability to photolytically generate an alkene has potential for use in synthetic and biological applications. As an alternative to other alkene synthesis chemistries, it is envisioned that the thioether chemistry described herein could provide a photochemical means to alkene generation in cases where other methods are not practical. In this study, the shortwave emission wavelength of mercury vapor, 254 nm, from a hand-held UVP UVGL-55 mercury lamp was used to photolyze our compounds. We chose this irradiation source over an arc lamp or a tunable dye laser source as it is inexpensive and accessible to most scientific labs. 3.2.3.1.3 On the fate of the sulfur Attempts to isolate the final state of sulfur following desulfurization have not proven successful. Based on previous literature, it is hypothesized that the ultimate fate of sulfur is elemental sulfur. The photodesulfurization of episulfides has been proposed to occur by extrusion of elemental sulfur.’28 Further evidence is observed in the photodesulferization of orthomercapto-heterocycles which has been thought to proceed through 3-membered azaepisulfides.’29 Photodesulfurization of triazole-3-thiones has also been shown, where irradiation in the presence of quenchers and sensitizers have suggested that decomposition proceeds via a triplet state episulfide formed through a thiol tautomer.’3° These examples would imply that the mechanism for benzimidazole desulfurization may also occur through an episulfide intermediate. Thus, it is speculated that following initial dealkylation of the thioether alkyl component, a mercaptan is left at the C-8 position that can photoisomerize through the episulfide. Sulfur photoextrusion of the episulfide occurs followed by abstraction of hydrogen from solvent. 173 3.3 GENERALIZATION OF PHOTOLYSIS TO OTHER CHROMOPHORES. Having identified the intermediate and products of the photolysis of 3.1, we decided to investigate the application of this photolysis to a chromophore other than adenine., Entry 3.2 was studied in order to show that the photolysis described for N-alkylated adenines can be extrapolated to unalkylated C-2 substituted benzimidazoles. Benzimidazole thioethers can be synthesized using the readily available and inexpensive synthon; 2-mercaptobenzimidazole (< lOt CDN/ gram Acros Organic). In this section it will be shown that the photolysis of 3.2 gives 2-mercaptobenzimidazole and ethene gas in similar yields as observed for the photolysis of the adenine derivative 3.1, indicating that the described thioether photolysis may be generalizable to other heterocycles. 3.3.1 MATERIALS AND METHODS 3.3.1.1 General synthetic methods All chemicals were purchased from Sigma, Aldrich, or Acros Organic. Deuterated solvents were purchased from Cambridge Isotope Laboratories. Thin layer chromatography Rf’s are reported on Silica Gel 60 F254 Glass TLC plates from EMD Chemicals. All ‘H Nuclear magnetic resonance (NMR) spectra were recorded at room temperature on a Bruker Avance 300 or 400 MHz instrument. Chemical shifts are reported using the 6 scale in ppm and all coupling constants (J) are reported in hertz (Hz). ‘H NMR spectra are referenced to 174 the tetramethylsilane peak (6 = 0.00). Mass spectrometry was performed at the mass spectrometry lab in the University of British Columbia (U.B.C.) chemistry department. 3.3.1.2 Specific synthetic procedures Authentic standards of 2-mercaptobenzimidazole and benzimidazole were purchased from Acros Organics and Aldrich, respectively. 3.2. 2-Ethylthio benzimidazole. 2-Mercaptobenzimidazole (0.5 g, 3.3 mmol) was added to neat ethyl iodide (8.0 mL, 100 mmol). Vigorous bubbling followed by precipitate formation was observed upon mixing. Solid sodium carbonate (1.3g 15.4 mmols) and 5 mL of DMSO were added such that full solubility was observed. The reaction was allowed to sit for 5 hours after which two products are observed with TLC RfS of 0.450 (2-ethylthio benzimidazole 3.2) and 0.70 (9-ethyl-2-ethylthio benzimidazole) with a 40 % EtOAc/ Hexanes v/v developing solution. After 5 hr, the reaction was added to 50 mL of SM NaOH and extracted into 50 mL CH21. The organic layer was washed twice with with 50 mL H20, dried over Na2SO4, filtered, and concentrated. The concentrate was sonicated in 50 mL of EtOAc and centrifuged. The EtOAc layer was decanted from remaining solids, concentrated, and isolated by flash chromatography with 33 % EtOAc: Hexanes as an eluant in a 20 cm x 1.5 cm column. Two products were collected in the following elution order: 60 mg of 9-ethyl-2-ethylthio benzimidazole followed by 80 mg of 3.2. 175 ‘H NMR (MeOH-d4, 400 MHz): 7.46-7.44 (m, 2H, CH Ar), 7.17-7.15 (m, 2H, CR Ar), 3.25 (q, J=7.3 Hz, 2H, CR2), 1.37 (t, J=7.3 Hz, 3H, CR3). HRMS (ES) calculated for C9H,,N2S (M+H): 179.06374 m!z, found: 179.0636. UV/Vis: Xrnax: 249 rim, 6249: 0.8 x M’ cuf’; 2max 282.0 nm, 6284: 1.55 x iü M’ cm’; Xmax: 291.0 rim, 629,: 1.50 x M’ cm’. 8254: 7.2 x M’ cm’, 3.3.1.3 Photochemical procedures Photochemical procedures were carried out as described in section 3.2.1 3.3.2 RESULTS 3.3.2.1 UV-Vis absorbance characterization of the photolysis of 3.2. Compound 3.2 was irradiated to verify whether 1) the thioether photolysis is limited to adenine or could be extrapolated to other heterocyclic thioethers and 2) the necessity of the N-9 adenine substituent (N-i on benzimidazole) to the photolysis reaction. The quantum yields reported in Table 3. for the conversion of 3.2 were calculated from the absorbance data reported in this section. The UV profile of the photolysis of 3.2 is shown in (Figure 3,8). 176 A B 0.4 0.3 0 a, .0 0.2 0.1 0.0 Wavelength (nm) Wavelength (nm) Figure 3.8. Spectrophotometric absorbance profile for the photolysis at 254 nm of a 20 jiM solution of 3.2 in MeOH. Because the rate of photolysis for the conversion of 3.2 to thiobenzimizole is much larger than the transition from 2- mercaptobenzimizole to benziniidazole, isosbestic points could be identified within the spectra. A. right, data to 2.5 mm detailing the conversion of 3.2 (green) to the intermediate 2-mercaptobenzimidazole (blue). Plots are shown at < 1 mm intervals to highlight isosbestic points indicated by the blue arrows. B, left, data from 2.5 mm onwards detailing the conversion of 2-mercaptobenzimidazole (MBI, blue) to benzimidazole (BI, red). Red arrows indicate isosbestic points corresponding to the transformation of 2-mercaptobenzitnidazole to benziniidazole. Plots are shown at 1 to S mm intervals to highlight isosbestic points indicated by the red arrows. UV-Vis absorbance data and isosbestic points indicated in Figure 3.15 suggest that photolysis of 3.2 gives 2-mercaptobenzimidazole as an intermediate which photolyzes further to benzimidazole. This is confirmed through ‘H-NMR analyses. 220 240 260 280 300 320 340 360 380 400 220 240 260 280 300 320 340 360 380 400 177 3.3.2.1.1 Ethene is evolved through the photolysis of 3.2. The ‘H-NMR spectroscopic study in Figure 3.9 details the photolysis of 3.2 and confirms that ethene is produced. -__ 1.I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 6.0 5.0 4.0 3.0 2.0 Figure 3.9. NMR spectral visualization of the photolysis of 3.2 in MeOH-d4. Relative irradiation times are indicated for the low power photolysis of 3.2. Times at which spectra were acquired are indicated to the left of spectra. A predicted spectra for that of ethene is shown in the uppermost spectrum. Arrows in the t= 90 h spectrum indicate NMR signals corresponding to the formation of 2-mercaptobenzimidazole, arrows in the t= 19 h spectrum indicate NMR signals corresponding to the formation of benzimidazole, and the green arrow in the top spectrum indicates the shift of an ethene standard in MeOH-d4. See appendix for 2-mercaptobenzimidazole and benzimidazole reference spectra. Ethene is observed at 90h of irradiation of 3.2 due to the use of an air tight seal on the NMR tube. Spectrum of ethene in MeOWd4 .0. CH3OkI I A. .0- t= 19h t=2h L A = Q hours I. ppm (ti) 8.0 7.0 1.0 178 ‘H-NMR spectra of the products and intermediates of photolyses compare well with authentic standards confirming that the photolysis intermediate is 2-mercatobenzimidazole, and that the photolysis product is benzimidazole (see appendix for NMR spectra of authentic references). The appearance of a peak at 6 5.4 following the irradiation of 3.2 suggests that ethene is generated as a photolysis product. These data in this section confirm that benzimidazoles can be used in lieu of adenine for the clean generation of ethene. The photolysis of 3.2 proceeds through a 2-mercatobenzimidazole intermediate before being converted to benzimidazole. This was confirmed through comparisons to authentic samples of 2-mercatobenzimidazole and benzimidazole (see appendix). This 2-mercatobenzimidazole intermediate produced in the photolysis 3.2 is analogous to the 8-mercapto adenine intermediate seen in the photolysis of 3.1 suggesting that the mechanism of thioether adenine photolysis is extendable to thioether beuzimidazoles. Benzimidazoles are cheaper synthons, are easier to crystallize, and are easier to chromatograph than adenines due to their less polar, aromatic nature. 3.3.3 DISCUSSION 3.3.3.1 On the generalizabillty of thioether photolysis Compound 3.2 was synthesized to confirm the following two hypotheses; 1) that the thioether photolysis mechanism proposed in Figure 3.5 is generalizable to other heterocyclic thioethers and 2) that the alkyl substitution at N-9 of adenine 3.1 (N-i on benzimidazole 3.2) does not affect the photolysis mechanism. Analysis of the photolysis of 3.2 by 1H NMR spectroscopy 179 shows that ethene is generated as a product (Figure 3.9). In the photolysis of 3.2, a mercaptan intermediate, 2-mercaptobenzimidazole, is produced that is analogous to the adenine intermediate 3.1-I that is produced in the photolysis of 3.1. This intermediate, 2-mercapto benzimidazole, not only has an absorbance maximum at 310 nni like 3.1-I, but it is also the starting material for the synthesis of 3.2, so its identity was easily verified by comparing absorbance profiles and ‘H NMR shifts to authentic standards. Irradiating 2-mercapto benzimidazole further gave the product benzimidazole. From these data it is concluded that the general photolysis mechanism proposed in Figure 3.7 extends indeed to the photolysis of benzimidazole thioether 3.2. As was observed with thioether-protected adenines in Chapter 2, it is proposed that the source of the C-2 proton in the benzimidazole product is the solvent; evidence for this has been obtained with mass spectrometry and ‘H-NMR spectroscopy data acquired from the irradiation of 3.3 in MeOH-d4. 180 3.4 THE MECHANISM OF ARYL THIOETHER PHOTOL YSIS Having shown that our thioether photolysis is generalizable to adenine and benzimidazole (3.1 and 3.2), we can begin to propose mechanisms which allow us to further describe this photolysis. Mechanistic analogy for thioether photolysis can be drawn from work by Kropp’27 (Figure 3.3) where the photolysis of 3.1 and 3.2 proceed through a radical or ionic mechanism during photolytic C-S bond cleavage. The following mechanism is proposed for the photolysis of 3.1 (Figure 3.10). — H—OMe cfl shNH2 [ sH2] j ,, Electron TransferN Hfl3.1 [ esN:H ] QeSNNH ‘ N ‘ N ‘LOM Figure 3.10. Mechanism adapted from Kropp’27 as applied to the photolysis of 3.1. 181 Alternatively, a Norrish type II — like fragmentation reaction mechanism can be proposed to account for the products produced in the photolysis of 3.1 section 3.2. This mechanism invokes the abstraction of a y-hydrogen in systems traditionally containing carbonyl oxygens undergoing radical cleavage. Heterocyclic versions of this fragmentation mechanism have been observed in the photolysis of 2-substituted quinolines,’3’pyrimidyl alkyl ketones, and 4- alkylpyrimidines.’32The latter example, being the closest relative to our thioether system, is shown in Figure 3.11.132 In the alkylpyrimidine system, an alkene is generated through intramolecular y-hydrogen abstraction by nitrogen through a 6-membered ring, a mechanism that is analogous to y-hydrbgen abstraction by carbonyl oxygen in Norrish type II radical cleavage.’32 1 . H3C H3C CH3 H3C H2 H3C Figure 3.11. Norrish type II fragmentation mechanism for the photochemical hydrogen abstraction of 4- allcylpyrimidines’32 Knowing that the Norrish type II mechanism requires a y-hydrogen to abstract, 3.3 was synthesized with the intention of producing a thioether with a y-hydrogen that could not be abstracted by the sp2 hybridized nitrogen (Figure 3.12). 182 * absiaction d HN \ / 2c + a / H2 ) diradical (3.3) Figure 3.12. A. Top. Norrish type II like mechanism proposed for the photolysis of 1-ethyl-2-thioethyl benzimidazole thioether. B. Bottom 3.3 with a y-ethyl thioether proton tied up in a 5- membered ring such that the abstracing nitrogen has limited access to its abstraction. If general thioether photolysis required a y-hydrogen for abstraction, the photolysis of 3.3 was expected to occur with a reduced quantum yield or result in different intermediates or products of photolysis. 3.4.1 MATERIALS AND METHODS 3.4.1.1 Specific synthetic procedures In cases where intermediates and products of photolyses could not be purchased, they were synthesized. The notation —I or —P following the numerical abbreviation of a compound indicates that the compound is a photolysis intermediate or product respectively. 183 3.3. 2,N-Cyclothioethyl-benzimidazole. 3.3 was synthesized through the following synthetic scheme. 1.LiAH4 ‘;‘: thiázolo[2,3-b]benzimidazole-3-(2H)-one 2-(1H-benzo[djimidazot-2-ylthio)ethanol 2-(1H-benzo[d]imidazol-2-ylthio)ethanol. A 50 mL RBF was charged with a stir bar, 25 mL of THF that had been distilled over sodium metal, and 1 g (5.25 mmol) of thiazolo{2,3- b]benzimidazole-3-(2H)-one. The mixture was cooled to —78 °C in an acetone/CO2()bath. 1 g of LiA1H4 (26.3 mmol, 5.0 equiv.) was added and the mixture was allowed to warm to room temperature. Product formation was observed after 5 mm by TLC with a 10 % MeOH!CHC13 developing solution where 2-(1H-benzo[d]imidazol-2-ylthio)ethanol. displays a Rf value of 0.15. The crude mixture was treated with MeOH saturated with 1 g of NH4C1 then filtered under vacuum with a buchner funnel. The filtrate was dried by rotary evaporation, then isolated by flash chromatography with 9:1 MeOH:CHC13as the eluant. ‘H NMR (MeOD, 300 MHz): 7.41-7.39 (m, 2H, CR Ar), 7.10-7.07 (m, 2H, CR Ar), 3.85-3.81 (t, J= 8.0 Hz, 2H, CH2), 3.38-3.33 (t, J= 10 Hz, 2H, CH2). LRMS (ES): (M+Li) m!z found: 201.1, calculated: 201.0668; (M+K) m!z found: 233.1, calculated: 233.0145; (M-H) mlz found 193.4, calculated: 193.044. 3.3. 2,N-Cyclothioethyl-benzimidazole. A 100 mL RBF was charged with a stir bar, 40 mg (0.21 mmol) of 2-(2-hydroxy)-thioethane-benzimidazole, 5 mL of 13.7 M (68.5 mmol) thionyl chloride and THF (50 mL). Product formation was observed by TLC with a 9:1 184 CHC13:MeOH v/v developing solution after 5 mm. The product 3.3 displayed a Rf value of 0.50. The crude mixture was dried by rotary evaporation. Et20 (30 mL) was added and the sample was concentrated to facilitate removal of THF. The crude product was resuspended in MeOH (25 mL). 5 M NaOH solution (15 mL) was added, causing