Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Enzymatic synthesis and NMR investigation of cellulose I and II Ng, Emily Siu Pan 2007

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

Item Metadata

Download

Media
831-ubc_2007-0534.pdf [ 6.73MB ]
Metadata
JSON: 831-1.0060167.json
JSON-LD: 831-1.0060167-ld.json
RDF/XML (Pretty): 831-1.0060167-rdf.xml
RDF/JSON: 831-1.0060167-rdf.json
Turtle: 831-1.0060167-turtle.txt
N-Triples: 831-1.0060167-rdf-ntriples.txt
Original Record: 831-1.0060167-source.json
Full Text
831-1.0060167-fulltext.txt
Citation
831-1.0060167.ris

Full Text

ENZYMATIC SYNTHESIS AND NMR INVESTIGATION OF CELLULOSE I AND II by EMILY SIU PAN NG B.Sc, The University of British Columbia, 2005 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA June 2007 © Emily Siu Pan Ng , 2007 Abstrac t Cellulose is one of the most abundant naturally occurring polymers on earth. It has also long been a polymer of high interest. Surprisingly, the complete structure of cellulose still remains a mystery to this date. Therefore, this thesis will utilize the recent advances of fluoro-sugar chemistry and enzymology to produce celluloses and modified cellulose whose structures could be studied by high-resolution solid-state nuclear magnetic resonance spectroscopy. The in-vitro syntheses of cellulose I and II using wild-type enzymes (Trichoderma viride Onozuka R-10, Trichoderma viride, Trichoderma reesei, Aspergillus niger, Cellulomonasfimi) and mutant enzymes (El97A, E197G and E197S of endoglucanase I from Fusarium oxysporum) were investigated. In-vitro syntheses of cellulose I by varying the buffer systems in which the enzymatic polymerization reactions were carried out in was studied. Furthermore, the possibilities of investigating cellulose structure by incorporating fluorine atoms into its polymer structures were also explored. The substrates necessary for the enzymatic syntheses of cellulose, namely o> cellobiosyl fluoride (a-CBF), P-cellobiosyl fluoride (P-CBF), 6'-fluoro-6'-deoxy-a-cellobiosyl fluoride (6'F-a-CBF) and 6'-fluoro-6'-deoxy-p-cellobiosyl fluoride (6'F-P-CBF) were all synthesized. Then, various in-vitro enzymatic polymerizations of the above substrates using different enzymes were carried out in an array of different buffer systems. The buffer systems used in this study include an aqueous buffer system and 2:1,5:1 and 7:1 ratios of acetonitrile to aqueous buffer systems. Interestingly, different buffer systems have dramatic effects on the amount of solid cellulose precipitates produced in each in-vitro polymerization regardless of the enzymes used. The resulting solid cellulose precipitates collected were then analysed using CP/MAS solid state NMR spectrometry. Unlike reports by Kobayashi and his group r e t '"3, the in-vitro synthesis of cellulose I using 2:1 acetonitrile and aqueous buffer cannot be achieved on any of the enzymes tested. All cellulose samples produced were identified as cellulose II. u Both 6'F-p-CBF and 6'F-a-CBF were successfully utilized by wild-type and mutant cellulases respectively as substrates and produced fluorinated cellulose polymers. Resulting fluorinated cellulose polymer were also analysed using CP/MAS solid state NMR spectrometry, proving that it is possible to use 1 9F CP/MAS NMR in combination with 1 3C and 'H CP/MAS NMR to further the study of cellulose structures. in Table of Contents Abstract ii Table of Contents iv List of Tables ix List of Figures x List of Illustrations xvi List of Abbreviations xviii Acknowledgments xx Dedication xxii Chapter 1 Cellulose 1 1.1 Cellulose Structure and Introduction 1 1.2 Structure of Cellulose I & II 6 1.2.1 Structural Information from the 1 3C CP/MAS Spectra of the Cellulose I and II Polymorphs 11 Chapter 2 Enzymatic Synthesis of Cellulose 15 2.1 The Chemistry of Cellulose Polymerization 15 2.2 Glycosylations and Polysaccharides Syntheses 16 2.3 Enzymatic Polymerization 17 2.4 Glycosidases 18 iv 2.5 Transglycosylation 22 2.6 Artificial Glycosynthases (Mutated Glycosidases) 24 2.7 Glycosyl Fluorides 26 Chapter 3 Nuclear Magnetic Resonance 27 3.1 Introduction 27 3.2 Relaxation Times 34 3.2.1 Spin-Lattice Relaxation Time (T [) 34 3.2.2 Spin-Spin Relaxation Time (T2) 39 3.2.3 Spin-Lattice Relaxation Time in the Rotating Frame (Ti p ) 43 3.3 High Resolution Solid State NMR 45 3.3.1 Nuclear Spin Interactions 46 3.3.2 Zeeman Nuclear Spin Interaction H z 47 3.3.3 Dipolar Interaction H D 48 3.3.4 Magnetic Shielding (chemical shift) HCs 49 3.3.5 Indirect Spin-Spin Coupling (J coupling) H s c 50 3.4 High Resolution CP/MAS 51 3.4.1 Magic Angle Spinning 51 3.4.2 Cross Polarization 53 3.4.3 Hartmann-Hahn Experiment 54 3.4.4 Cross Polarization Experiment 56 Chapter 4 Results and Discussion 60 4.1 Syntheses of Cellobiosyl Fluorides 60 4.1.1 a-Cellobiosyl Fluoride 60 4.1.2 p-Cellobiosyl Fluoride 61 4.1.3 Stability 62 4.2 Synthesis of Fluorine-Labelled Cellobiosyl Fluorides 63 4.2.1 Fluorine Labelling 64 Protection of the Hydroxyl Groups 65 Reactions at the 6'-Position 65 4.2.2 6'-Deoxy-6'-Fluoro-a-Cellobiosyl Fluoride 66 4.2.3 6'-Deoxy-6'-Fluoro-p-Cellobiosyl Fluoride 66 v 4.3 Enzymatic Synthesis 68 4.3.1 Enzymatic Synthesis of Cellulose using Wild Type Enzymes 68 4.4 Enzymatic Synthesis using P-Cellobiosyl Fluoride 70 4.4.1 Common Cellulose II Polymorph Misconceptions .....74 4.4.2 Cellulose Production in Different Acetonitrile/Water Buffers 76 Cex-Cellulomonas fimi 83 Endoglucanase 86 Modified Cellulose Production using 6'-Deoxy-6'-Fluoro-Cellobiosyl Fluoride Substrates 90 4.5 Solubility Factor 92 4.6 Cellulose Production by Wild Type Enzyme vs Mutants 93 4.7 Further Studies of the Monomeric Substrates 94 4.7.1 Solid State NMR Studies of the a-Cellobiosyl Fluoride Substrates 96 l 9F Solid State NMR Studies 101 Chapter 5 Conclusions and Suggestions for Further Work 103 5.1 Summary and Conclusions 103 5.2 Suggestions for Further Work 104 Chapter 6 Materials and Methods 105 6.1 Chemical Synthesis of Cellobiose Derivatives 105 6.1.1 a-Cellobiosyl Fluoride 106 2, 3, 6, 2', 3', 4', 6' Hepta-O-acetyl-a-cellobiosyl fluoride 106 a-Cellobiosyl fluoride 107 6.1.2 p-Cellobiosyl Fluoride 108 2, 3, 6, 2', 3', 4', 6' Hepta-O-acetyl-a-cellobiosyl bromide 108 2, 3, 6, 2', 3', 4', 6' Hepta-O-acetyl-p-cellobiosyl fluoride 109 p-Cellobiosyl fluoride 110 6.1.3 6'-Deoxy-6'-Fluoro-a-Cellobiosyl Fluoride Ill 1, 2, 3, 6, 2\ 3' Hexa-O-acetyl-4', 6'-benzylidene cellobiose Ill 1, 2, 3, 6, 2', 3' Hexa-O-acetyl cellobiose 112 1, 2, 3, 6, 2', 3' Hexa-0-acetyl-6'-deoxy-6'-fluoro cellobiose 113 2, 3, 6, 2', 3' Penta-0-acetyl-6'-deoxy-6'-fluoro-a-cellobiosyl fluoride 114 vi 6'-Deoxy-6'-fluoro-a-cellobiosyl fluoride 115 6.1.4 6'-Deoxy-6'-Fluoro-p-Cellobiosyl Fluoride 116 2, 3, 6, 2', 3', 4' Hexa-0-acetyl-6'-deoxy-6'-fluoro-a-cellobiosyl bromide 116 2, 3, 6, 2', 3', 4' Hexa-0-acetyl-6'-deoxy-6'-fluoro-P-cellobiosyl fluoride 117 6'-Deoxy-6'-fluoro-P-cellobiosyl fluoride 118 6.2 Enzymatic Syntheses 119 6.2.1 Trichoderma viride Onozuka R-10 120 Acetonitrile/aqeous sodium phosphate buffer ratio 7:1 120 Acetonitrile/aqeous sodium phosphate buffer ratio 5:1 120 Acetonitrile/aqeous sodium phosphate buffer ratio 2:1 120 6.2.2 Cellulose Production in 2:1 Acetonitrile/Aqueous Buffer 121 Aspergillus niger 121 Trichoderma reesei 121 Trichoderma viride 121 Cex 121 6.3 Cellulose Production with Glycosynthase Mutants 122 6.3.1 E197G 122 6.3.2 E197A 122 6.3.3 E197S : 122 6.4 Syntheses of 6'-Fluorinated Cellulose using 6'F-CBF 123 6.4.1 E197A 123 6.4.2 Cex 123 6.5 NMR Spectroscopy 124 6.5.1 NMR Spectrometer 124 6.5.2 Probes 124 6.5.3 Magic Angle Spinning 125 6.5.4 Small Sample NMR 126 6.5.5 Low Temperature MAS 127 6.5.6 Reference Samples 129 6.5.7 13C{'H} CP MAS NMR 129 6.5.8 Pulse Sequences 130 6.6 Single Crystal X-ray Diffraction 133 6.6.1 Diffractometers, Data Collection and Processing 133 a-Cellobiosyl Fluoride 133 6'-Deoxy-6'-Fluoro-a-Cellbiosyl Fluoride 134 References 136 vii Appendix I. X-Ray Crystal Structure Data for a-Cellobiosyl Fluoride 146 A. Crystal Data 146 B. Intensity Measurements 147 C. Structure Solution and Refinement 148 Appendix II. X-Ray Crystal Structure Data for 6'-Deoxy-6'-Fluoro-a-Cellobiosyl Fluoride 157 A. Crystal Data 157 B. Intensity Measurements 158 C. Structure Solution and Refinement 159 viii I List of Tables Table 1-1 Characteristic 1 3C chemical shifts of cellulose II and I allomorphs 11 Table 2-1 Bond-lengths and var der waals radii of some elements 26 Table 4-1 Names, sources and abbreviations of wild-type cellulose polymerizing enzymes used in this thesis 70 Table 4-2 Comparison of the 1 3C chemical shifts of insoluble cellulose precipitates produced by Ono R-10 in 7:1 acetonitrile/water buffer with the diagnostic 1 3C chemical shifts of the cellulose II and I polymorphs 76 Table 4-3 Comparison of the 1 3C chemical shifts of insoluble cellulose precipitates produced by Ono R-10 in 5:1 and 2:1 acetonitrile/water buffer with the diagnostic l 3C chemical shifts of cellulose II and I polymorphs 78 Table 4-4 Comparison of the 1 3C chemical shifts of insoluble cellulose precipitates produced by T. viride and T. reesei in 2:1 acetonitrile/water buffer with the diagnostic l 3C chemical shifts of cellulose II and I allomorphs 79 Table 4-5 Comparison of the 1 3C chemical shifts of insoluble cellulose precipitates produced by T. viride and T. reesei in 2:1 acetonitrile/water buffer with diagnostic 1 3C chemical shifts of cellulose II and I polymorphs 86 ix List of Figures Figure 1 -1 Basic polymer chain of cellulose 2 Figure 1-2 Illustration of (a) 4Ci chair conformation and (b) P~(l-4) linkage 4 Figure 1 -3 Polymorphy of cellulose 5 Figure 1-4 Solid-state 1 3C NMR spectra of various cellulose I samples 6 Figure 1 -5 Solid-state 1 3C CP-MAS NMR spectrum of cellulose I and II 7 Figure 1-6 Alignment of cellulose chains 8 Figure 1-7 Model of a rosette TC 9 Figure 1 -8 Illustration of a cellulose synthase rosette in the plant cell plasma membrane 9 Figure 1-9 Illustration of in vivo cellulose I formation 10 Figure 1-10 CP/MAS 1 3C NMR spectra of Cladophora, tunicate and cellulose Ia 13 Figure 1-11 CP/MAS 1 3C NMR spectra of Cladophora, tunicate and cellulose Ip 13 Figure 1-12 Subspectrum expansion of C2, C3, and C5 regions of the CP/MAS 1 3C NMR spectra of pure Ia phases of Cladophora cellulose 14 Figure 1-13 Subspectrum expansion of C2, C3, and C5 regions of the CP/MAS l 3C NMR spectra of pure Ip phases of Cladophora cellulose 14 Figure 1-14 Expansion of spectrum of the mercerized cellulose II obtained from 1 3C -enriched bacterial cellulose 14 Figure 2-1 Illustration of the stereo-chemical outcome of (a) retaining and (b) inverting glycosidases 18 Figure 2-2 General illustration of the distance between the two catalytic carboxylic acid of (a) retaining glycosidase and (b) inverting glycosidase 21 Figure 3 -1 Description of a proton 27 Figure 3-2 Energy difference between spin states of spin lA nuclei in the presence of external magnetic field BQ 29 Figure 3-3 Magnetic behavior of protons in a magnetic field B0 30 Figure 3-4 Precession of net magnetization 31 Figure 3-5 Effect of Bi pulse along x-axis on net magnetization 32 Figure 3-6 The transformation of time domain FID signal to the frequency domain NMR spectrum through a Fourier transform 33 Figure 3-7 The process of spin-lattice relaxation in a rotating frame 35 Figure 3-8 Recovery of net magnetization along z-axis in terms of 7/ 36 Figure 3-9 Measuring Tj spin-lattice relaxation times using inversion recovery method 37 Figure 3-10 Measuring Ti spin-lattice relaxation times using saturation recovery method 38 Figure 3-11 The process of spin-spin relaxation in a rotating frame 39 Figure 3-12 Method of measuring T2 spin-spin relaxation times 41 Figure 3-13 Multi-echo pulse sequence for measuring T2 relaxation time 42 Figure 3-14 Method of measuring Ti p relaxation times 44 Figure 3-15 Pictorial representation of the internuclear vector in the presence of external magnetic field B0 48 xi Figure 3-16 Pictorial representation of a spinning rotor in the process of magic angle spinning (MAS) in a magnetic field B0 52 Figure 3-17 Energy level difference diagram of the cross polarization spin-locking 'thermal contact' 55 Figure 3-18 Vector representation of the cross polarization process 58 Figure 3-19 The cross polarization pulse sequence 59 Figure 4-1 Illustration of a-cellobiosyl fluoride and P-cellobiosyl fluoride 60 Figure 4-2 Fluorine labeled cellulose polymer 63 Figure 4-3 General approach for direct fluorination 64 Figure 4-4 l 3C solid state CP/MAS NMR spectrum of bacterial cellulose 1 72 Figure 4-5 1 3C solid state CP/MAS NMR spectrum of cellulose II 72 Figure 4-6 1 3C solid state CP/MAS NMR spectra of the mercerized cellulose II 73 Figure 4-7 2D MAS-J-HMQC spectrum of the mercerized cellulose II obtained from l3C-enriched bacterial cellulose : 73 Figure 4-8 1 3C solid state CP/MAS NMR spectrum of rayon obtained from the cuprammonium process of filter paper 74 Figure 4-9 ' 3C solid state CP/MAS NMR spectrum of filter paper 75 Figure 4-10 1 3C solid state CP/MAS NMR spectrum of cellulose from cellulase Ono R-10 in 7:1 acetonitrile / water buffer 77 Figure 4-11 1 3C solid state CP/MAS NMR spectrum of cellulose from Ono R-10 in 5:1 acetonitrile / water buffer 77 Figure 4-12 l 3 C solid state CP/MAS NMR spectrum of cellulose from Ono R-10 in 2:1 acetonitrile / water buffer 80 xii Figure 4-13 1 3C solid state CP/MAS NMR spectrum of cellulose from T. viride in 2:1 acetonitrile / water buffer 80 Figure 4-14 MALDI-TOF MS spectrum of cellulose from T. viride. produced in 2:1 acetonitrile / water buffer 81 Figure 4-15 1 3C solid state CP/MAS NMR spectrum of cellulose from T. reesei in 2:1 acetonitrile / water buffer 81 Figure 4-16 MALDI-TOF MS spectrum of cellulose from T. reesei in 2:1 acetonitrile / water buffer 82 Figure 4-17 MALDI-TOF MS spectrum of cellulose from Cex in 2:1 acetonitrile / water buffer 84 Figure 4-18 MALDI-TOF MS spectrum of cellulose from Cex in 2:1 acetonitrile / water buffer 84 Figure 4-19 l 3C solid state CP/MAS NMR spectrum of cellulose produced from glycine mutant E197G of Endoglucanase I from F. oxysporum in aqueous buffer 88 Figure 4-20 MALDI-TOF MS spectrum of cellulose from Endoglucanase IF. Oxysporum in 5:1 acetonitrile / water 88 Figure 4-21 MALDI-TOF MS spectrum of cellulose from Endoglucanase IF. Oxysporum in 2:1 acetonitrile / water 89 Figure 4-22 1 3C solid state CP/MAS NMR spectrum of cellulose II from endoglucanase F. oxysporum mutant E197A in 2:1 acetonitrile / water buffer 89 Figure 4-23 l 3C solid state CP/MAS NMR spectrum of solid material produced from T. viride using 6'F (3-CFB in 2:1 acetonitrile / water buffer 91 X l l l Figure 4-24 Mass separation differences between cellulose polymer chains produced by wild type enzymes and mutant enzymes 94 Figure 4-25 Stereoview of the X-ray crystal structure of a-cellobiosyl fluoride 95 Figure 4-26 Stereoview of the X-ray crystal structure of 6'F a-cellobiosyl fluoride ...95 Figure 4-27 1 3C solid state CP/MAS NMR spectrum of per-acetylated a-CBF 97 Figure 4-28 Comparison of the 1 3C solid state CP/MAS NMR spectrum expansion and the high-resolution 1 3C solution state NMR spectrum of per-acetylated a-CBF 97 Figure 4-29 1 3C solid state CP/MAS NMR spectrum of cellulose of per-acetylated a -CBF 99 Figure 4-30 1 3C solid state CP/MAS NMR spectrum of a-CBF 98 Figure 4-31 1 3C High resolution solution state NMR spectrum of a-CBF 99 Figure 4-32 Comparison of the 1 3C solid state CP/MAS NMR spectrum and the high -resolution 1 3C solution state NMR spectrum of a-CBF 99 Figure 4-33 1 3C solid state CP/MAS NMR spectrum of a-CBF 100 Figure 4-34 1 3C solid state CP/MAS NMR spectrum of cellulose of 6'F-a-CBF 100 Figure 4-35 1 9F solid state single pulse NMR spectrum of a-CBF 102 Figure 4-36 l 9F solid state single pulse NMR spectrum of 6'F-a-CBF 102 Figure 6-1 Small sample setup for the 4mm rotor 126 Figure 6-2 Small sample setup for the 7mm rotor 126 Figure 6-3 Schematic setup of low temperature MAS NMR experiments 128 xiv Figure 6-4 Pulse sequence for 19F solid-state NMR experiments with a 90° pulse 130 Figure 6-5 Pulse sequence for 1 9F solid-state NMR with a 30° pulse 131 Figure 6-6 Pulse sequence for 1 3C -> ]H CP/MAS solid-state NMR without *H decoupling 131 Figure 6-7 Pulse sequence for 1 3C -» *H CP/MAS solid-state NMR with *H decoupling 132 Figure 6-8 Pulse sequence for 1 3C -> [H CP/MAS solid-state NMR with simultaneous *H and l 9F decoupling 132 xv List of Illustrations Scheme 2-1 Elongation of cellulose chain using UDP-Glc with inversion of anomeric center 15 Scheme 2-2 Generalized double displacement mechanism of a retaining P-glycosidase 19 Scheme 2-3 Generalized direct displacement mechanism of an inverting P-glycosidases 20 Scheme 2-4 The propsed retaining mechanism of the Thr26His nucleophile mutant of T4 phage lysozyme 21 Scheme 2-5 Formation of glycosyl enzyme intermediate with glycosyl fluoride donor. 22 Scheme 2-6 Hydrolysis and transglycosylation reactions proceeding from the glycosyl enzyme intermediate 23 Scheme 2-7 Generalized inverting mechanism of a retaining P-glycosidase nucleophile mutant 24 Scheme 4-1 Outline of the chemical synthesis of ct-cellobiosyl fluoride 61 Scheme 4-2 Outline of the chemical synthesis of P-ellobiosyl fluoride 62 Scheme 4-3 The outline of selective protection and DAST fluorine labeling 64 Scheme 4-4 Outline of the chemical synthesis of the 6'-deoxy-6'-fluoro-a-cellobiosyl fluoride 66 Scheme 4-5 Outline of the chemical synthesis of 6'-deoxy-6'-fluoro-p-cellobiosyl fluoride 67 xvi Scheme 4-6 Enzymatic synthesis of cellulose polymer by cellulase using P-cellobiosyl fluoride 69 Scheme 4-7 Enzymatic synthesis of cellulose polymer using a-cellobiosyl fluoride in mutant enzyme Endoglucanase I, E197G 85 xvii List of Abbrev ia t ions AgF silver fluoride CP cross polarization CSA chemical shift anisotropy CBD cellulose binding domain CBF cellobiosyl fluoride DAST diethylaminosulfur trifluoride DCM methylene chloride DMF dimethyl formamide DP degree of polymerization FID free induction decay FT Fourier transformation H Hamiltonian HRLSIM high-resolution liquid secondary ion mass spectrometry Hz Hertz LCMS liquid chromatography mass spectrometry MALDI-TOF matrix-assisted laser desorption/ionization - time of flight MAS magic angle spinning MSL spin-locked magnetization NMR nuclear magnetic resonance ppm parts per million rf radio frequency S/N signal to noise TC terminal complexes xviii TLC thin layer chromatography UDP-Glc uridine 5'-diphospho-a-D-glucose TsOH toluene-4-suphonic acid xix A c k n o w l e d g m e n t s This thesis will not be possible without the help and the excellent guidances provided by both of my supervisors, Professor Colin A. Fyfe and Professor Stephen G. Withers. Professor Stephen G. Withers' professionalism and insights are the greatest inspiration to anyone who aspires to be a scientist. Professor Colin A. Fyfe's ingenuity and persistence in finding and proving the truth in his field of study is something all should learn from. I am grateful for the opportunity to study and learn under the guidance of two of the greatest people I have ever met. On top of my professors, many members from the department of Chemistry at UBC also lend numerous help in making this thesis possible. First I would like to thank the group members in both the Withers' lab and Fyfe lab, whose advices and friendships I should take with me for the rest of my life. I would also like to thank Dr. Hamzah Mohd-Salleh for providing the precious mutant enzymes of endoglucanase I from Fusarium oxysporum and David Poon for providing the enzyme of Cellomonas fimi. Dr. Celine Schneider for the help in solid states NMR and the much-needed company for the days we spend in the dark basement of the E-wing. Miss Emily Kwan in the Withers' cellulase lab for providing the bacterial cellulose I sample. Dr. Johannes Mullegger for all the help in the attempt synthesis of endoglucanase I, F. oxysporum mutants. Dr Brian Patrick for help in obtaining the x-ray crystallography of the substrates and Marshall Lapawa in the UBC Mass Spectrometry Centre for those greatly needed help in operating the MALDI-TOF mass spectrometry machine and sample preparations. I would also like to thank Dr. Sophia Nussbaum for providing precious teaching experiences and a chance to giving back to the UBC community as lecturer. Ann Thomas and Angelo for making my teaching experience easy and fun. Also I would like to thank Vivian Yip, my good friend since high school, for her support and companionship throughout the years making my experience at UBC complete. To my mom Lee Lee Yu and my dad Kwok Kwong Ng, I would like to thank you for your love and your continuous support throughout my study despite you disapproval of my pursue in science. To Dr. Hoa D. Ly, I would like to thank you for introducing me to the wonderful world of research. Last but not least, I would like to thank Joesph Edward xx Peschisolido for being there through good times and bad times. Also to Papaya, Peanutty and Coconutty who always give me the warmest welcome everytime I come home. Emily Ng xxi Dedication xxii Chapter 1 Cellulose 1.1 Cellulose Structure and Introduction The structure of cellulose, a seemingly simple polysaccharide of linear P(l->4) linkages, has mystified scientists for centuries. Ever since the 1920s, cellulose had been an important target molecule with respect to its structure, molecular weight, and derivatization reactions for scientists, especially in the field of polymer science 4. The first attempt at a chemical synthesis of cellulose was in 1941 and then no successful synthesis had been reported for half a century. Also, despite the vast interest and tremendous effort applied in attempting to solve its structure, the complete structure of cellulose remains somewhat a mystery to this date and is often an area of debate. 1 As one of the most abundant organic polymers on earth, cellulose is found in the vast majority of plant forms and is produced by a number of other organisms, including marine life forms and some pathogenic bacteria5. There is even evidence which suggests that cyanobacteria, one of the most ancient life forms on earth also synthesize cellulose 6. Biologically, cellulose holds great significance. It is involved in many of the structural and metabolic pathways of numerous organisms. It is estimated that some 1015 kg of cellulose is biologically synthesized and degraded every year 1. The manipulation and utilization of cellulose is as old as civilization itself, dating back many millennia. Historically, the most well known uses of cellulose are in the cotton-based textile industry and the pulp and paper industry. Being a key structural component of wood, cellulose has also influenced the daily life of most cultures through its utilization. Hence, over the past 100 years, much of the research on cellulose has been focused on its utilization in industrial and construction processes. However, as many of the vital biological functions of cellulose are recognized, cellulose is now also considered as a biological molecule of major importance with many distinctive properties. At the same time, due to its unique biocompatibility, chirality, structure-forming capacity, and environmentally benign properties 4'8'9, much research is underway to use cellulose as an advanced material. ^ ^ O H \ OH ^ ^ • O H -OH I \ \ OH OH Figure 1-1 Basic polymer chain of cellulose (linear P-(l-4) linked homo-polymer of anhydroglucose) 2 The term cellulose was first introduced by the French agriculturalist Anselm Payen in 1842. He recognized that the key constituent of plant cell walls is a neutral polysaccharide with its monomeric constituent being a carbohydrate. Later on, as the establishment of the polymer hypothesis consolidated in the 1920-30s, cellulose was recognized as the linear (3-(l-4) linked homo-polymer of anhydroglucose 1 0 (Figure l-2b). Recently, the polymer was also redefined as the homo-polymer of anhydrocellobiose 1 0 (Figure 1-2). As a polymer of pyranose rings in the 4 C i chair conformation (Figure l-2a) with hydroxyl groups in the equatorial positions, one would expect its solubility in aqueous solvents to be high. However, the number of glucose monomeric units required to produce an insoluble product is only about eight. In fact, cellulose has poor solubility in most simple solvents; almost all of the solvent systems for cellulose are multi-component solvents that form strong associative bonds with the molecule. Because the methodologies of modern polymer science analysis often rely on the measurement of properties of the polymers in solution, its low solubility has become one of the major hindrances in identifying its structures in detail. Even though the primary structure of cellulose has been well established for some time, its secondary and tertiary structure has baffled scientists for many years. Since many native celluloses found in plant forms are highly ordered structures that are elaborately integrated into biological structural tissues, the final structure of cellulose is very much a function of its primary structure as well as its state of aggregation. Therefore, it was difficult to isolate a uniform "cellulose" that is common to all plants. It was later discovered that all of the celluloses from plant sources are blends or composites of two forms of cellulose, Ia and Ip, with the particular blend specific to the species and tissue from which the cellulose is derived. In high plants the Ip forms always appear to be dominant whereas in bacterial and algal celluloses the Ia form dominates n ' 1 2 . In general, there are six polymorphs of cellulose categorized (I, II, IIIi, IIIn, IVi and IVn). Cellulose I is only found in unprocessed cellulose in nature, it can be further sub-divided into Ia and Ip depending on the source from which it is derived 1 U 2 . Cellulose II is the most extensively studied and most prevalent form seen in industrial applications. It is 3 generally derived through manipulation of cellulose I by one of two possible methods: (a) regeneration of cellulose I by solubilization and re-precipitation by dilution in water, or (b) mercerization, the swelling of native fibres in concentrated sodium hydroxide 1 3 - 1 5 . Celluloses III/, and III;/ 1 6' 1 7 are formed from a reversible process in which cellulose I or II are treated with liquid ammonia or amines respectively 18~20. Lastly, cellulose IV/ and IV// 2 I ' 2 2 are formed by heating celluloses III/ and III// respectively to 206°C in glycerol. All of the above polymorphs of cellulose can interconvert between each other with the exception of cellulose I (Figure 1-3). Even though cellulose I is the most abundance form of cellulose found in nature, structurally, it is not the most stable form of cellulose, it cannot be synthesized by crystallization of pre-formed cellulose chains. Crystallization of pre-formed cellulose chains in vitro will only give rise to the thermodynamically more favourable cellulose II. Since all the other forms of cellulose are basically derived from either cellulose I and II, the focus of this thesis will be on the structures of cellulose I and II. Figure 1-2 Illustration of (a) 4C[ chair conformation and (b) P-(l-4) linkage. 4 boiling in acid mereertzatum or regeneration boiling wi acid Figure 1-3 Polymorphy of cellulose. Adapted from ref. 23. 5 1.2 Structure of Cellulose I & II Early Raman spectral studies first revealed that the chains in solid cellulose exists in a twofold helical conformation. However, it was found later on that the adjacent anhydroglucose units in a cellulose chain are actually non-equivalent and that the non-equivalences takes different forms in celluloses I and II24. Later on, early solid state l 3C NMR studies revealed that there exists a wider structural diversity among native cellulose than other sensitive measurements had previously detected n ' 2 4 . Figure 1-4 illustrates this variety in samples of native celluloses from different sources and a sample of cellulose I that has been regenerated from phosphoric acid solution at elevated temperatures 1 1 '24. C-2,3,5 • » ' < ' i i — i — i t i i _ C O HX) Kl> 60 ppm Figure 1-4 Solid-state 1 3 C NMR spectra of various cellulose I samples (a) ramie; (b) cotton; (c) hydrocellulose from cotton linters; (d) a low DP regenerated cellulose I; (e) Acetobacter xylinum cellulose; (0 Valonia Ventricosa cellulose. The (x) marks the first spinning side band of linear polyethylene, which was used as a chemical shift reference. Adapted from ref. 11. 6 Despite the differences, all of the spectra showed a number of similar complex multiplet resonances and the assignments of these resonances have been well established for some time 2 5. Beginning from downfield: d between 104 and 108 ppm; sharp resonances of C 4 between 88 and 92 ppm, and broad components ranging down to about 83 ppm; a cluster assigned to C2, C 3 and C 5 in the range between 70 and 80 ppm; resonances for Ce below 70 ppm25. All of the spectra for native cellulose I can be resolved into linear combinations of two spectra of Ia and Ip as shown in Figure 1-5. Along with the figure a spectrum of cellulose II is also shown as a comparison between the two forms. Even though similar in structure, the spectra of cellulose II show some distinct differences including different chemical shifts especially in C-4 and C-6, and different resonance patterns in C-2,3,5. C-2,3,5 100 XO 60 ppm Figure 1-5 Solid-state l 3 C CP-MAS NMR spectrum of cellulose I and II (a) low-DP cellulose II; (b & c) low-DP cellulose I sample from Acetobacter. (b) cellulose Ia; (c) cellulose Ip. Adapted from ref. 11. 7 Figure 1-6 Alignment of cellulose chains. (a) Parallel structure of cellulose I chains, (b) Anti-parallel structure of cellulose II chains. Adapted from ref. 26. Among the two high-order crystalline structures of cellulose allomorphs, it is proposed that the glucan chains in cellulose I exhibit a parallel27 arrangement whereas cellulose II has an anti-parallel arrangement ' (Figure 1-5). Even though cellulose II is the thermodynamically more stable form with an additional hydrogen bond per glucose residue, normally living cells only produce cellulose I with the exceptions of several algae and bacteria 3 0. In organisms where cellulose production is studied in detail, it is found that cellulose I is synthesized with a distinct number of parallel glucan chains arranged to form the nano-structure known as a microfibril31. Since 1976, scientists have identified through freeze-fracturing studies, a series of enzyme complexes associated with cellulose production from organisms ranging from algae to tunicates to bacteria and plants 3 2. It is proposed that cellulose is formed at the outside of the plasma membrane via rosettes of cellulose synthase complexes known as terminal complexes (TCs) (Figure 1-7) 3 3 - 3 5 . These complexes are involved in the elongation of cellulose microfibrils and all the TCs in a rosette elongate their microfibril at the same rate producing parallel cellulose chains. Then the final crystallization process of individual cellulose chains is thought to be brought about by a top protein accompanying the TC 3 6 producing crystalline parallel cellulose structures (Figure 1 -7 & Figure 1-8). 8 Figure 1-7 Model of a rosette TC. It shows the possible origin of glucan chains. Each unit contains six subunits and each subunit synthesizes a 6-glucan-chain sheet held together by van der Waals forces. Hence, this TC produces a crystalline microfibril with approximately 36 glucan chains. The sheets are then "stacked" by hydrogen bonding to form a three-dimensional microfibril. Adapted from ref. 37-39. o Kerrigan CesAe CesAi CesAi/e Sitosterol Glucose Figure 1-8 Illustration of a cellulose synthase rosette in the plant cell plasma membrane. (A) Longitudinal view of the rosette composed of six elementary particles during elongation of cellulose microfibrils (MF). Sucrose synthase (SuSy) on the cytoplasmic face of the plasma membrane (PM) may channel UDP-glucose to the 36 growing glucan chains, which are then extruded into the plant cell wall where they coalesce to form microfibrils. (B) Possible substructures within one elementary particle of the rosette, (i) The particle contains six elongating polypeptides (CesAe), one initiating polypeptide (CesAi), and one copy of Korrigan cellulase. (ii) The particle contains three copies each of the two types of elongating polypeptide (CesAe), one initiating polypeptide (CesAi), and one copy of Korrigan cellulase. (iii) The particle contains three copies each of two types of polypeptide that both initiate synthesis and promote chain elongation (CesAi/e), and one copy of Korrigan cellulase. (C) Initiation of cellulose synthase. UDP-glucosyl transferase (UGT) transfers a glucose residue onto a sitosterol molecule on the cytoplasmic face of the plasma membrane, forming sitosterol-P-glycoside (SG). The short glucose chain is extended with UDP-glucose by an initiating CesAi subunit which "flips" to the outer face of the plasma membrane. The cellodextrin chain is then cleaved by Korrigan cellulase, binds to the elongating CesAe, and is extended into a glucan chain by addition of UDP-glucose provided by sucrose synthase. Adapted principally from ref. 40-42. 9 Formation of Cellulose f JL Micro-fibrils i : I Lipid Cellulose \ . * Bilayers Synthase Figure 1 -9 Illustration of in vivo cellulose I formation. Adapted from ref. 26. 1 0 1.2.1 Structural Information from the 1 3 C C P / M A S Spectra of the Cellulose I and II Polymorphs Although the exact arrangements of individual chains in cellulose I and II polymorphs are still under debate, l 3C CP/MAS solid state NMR spectra are completely diagnostic in distinguishing the two. In general, almost all cellulose I materials consist of a combination of two forms, I„ and Ip, whose relative proportions depend on the source of the sample. The presence of two different chain forms within a single polymorph leads to multiple signals, which are most evident in the CI and C4 signals. On the other hand, all of the cellulose II spectra are identical regardless of the source, synthetic or regenerated. The shifted amorphous peaks normally found in cellulose I spectra are usually absent. The CI, C4 and C6 signals in cellulose II samples each consist of two resonances of equal intensities indicating two different rings in the asymmetric unit consistent with the structure containing two non-equivalent orientations of the cellulose chains. Also, detailed comparisons of the 1 3C chemical shifts of the two polymorphs show small but reproducible differences, particularly in the resolved signals for CI, C4 and C6 (Table 1-1). 1 3C CP/MAS NMR is thus a very simple, direct and diagnostic method for assigning polymorphic form of cellulose samples. CI C4 C2,3,5 C6 (ppm) (ppm) (ppm) (ppm) I 104-106 88-92 70-78 65-68 II 105-107 87-89 70-80 62-63 Table 1-1 Characteristic l 3 C chemical shifts of cellulose II and I allomorphs. 11 The assignments of the 1 3C CP/MAS NMR resonances have been well established for both cellulose I and II. Typically, in cellulose I, both forms (Ia and Ip) have spectra with CI between 104 and 106 ppm; C4 with sharp resonances between 88 and 92 ppm and broad components ranging down to about 83 ppm; a cluster of resonances assigned to C2, C3, and C5 is in the range between 70 and 80 ppm; and the resonances below 70 ppm are associated with C625 (Figure 1-1043 & Figure 1-1143). For cellulose II, CI is two signals at 105 and 107 ppm; C4 is two signals at 87 and 89 ppm; a cluster between 70 and 78 ppm is assigned to C2, C3, and C5; the resonances between 62 and 63 ppm are attributed to C623. Traditionally, the precise assignment of the C2, C3, and C5 has been difficult due to the overlapping nature of the chemical shifts of these particular carbons. However, due to recent advances in 2D solid state NMR techniques, the cluster assigned to C2, C3 and C5 has now been assigned through various 2D techniques for both cellulose I (Figure 1-1244, Figure 1-1344) and II (Figure 1-1423). On top of the chemical shift differences, the two polymorphs can also be identified using the splitting patterns of the signals. Due to the fact that cellulose I is generally comprised of two forms (Ia & Ip), most cellulose I spectra contain two overlapping spectra. Hence, multiplet peaks are observed for each of the cellulose I signals. Generally, signals for CI, C4 and C6 appear to be comprised of two sets of overlapping peaks, indicating the presence of two cellulose I forms. While in cellulose II spectra, signals for the same carbons generally appear to be comprised of only one set of peaks, indicating the presence of one cellulose form. 12 3 OH OH k B ci C4 C4 C2 CH C2;5 C3 ca! f ""AS ' C6 35 2 6 3 25 110 100; 90; 80 l 3C Chemical shift 70 ppm Figure 1-10 (A) CP/MAS 1 3 C NMR spectra of Cladophora ( l a - r i c h , / ^ = 64/36, solid line) and tunicate (Ip-rich,y /^/p = 8/92, dotted line) celluloses. (B) The 1 3 C subspectrum of the pure cellulose Ia. The l 3 C subspectrum of I a phase was derived by subtracting the l 3 C spectrum of Cladophora cellulose (solid line in figure A) from that of tunicate cellulose (doted line in figure A) cellulose. (C) l 3 C Line spectra of the two magnetically nonequivalent anhydroglucose residues composing cellulose I a. The lines indicate the C chemical shifts of each carbon of the anhydroglucose residues. The line spectra were determined by the 2D refocused CP-INADEQUATE spectrum of Cladophora cellulose. Adapted from ref. 43. CHzOH B C1C1 A . C4 iC4 C2.5 C5 ( C 3 \ C ? \ C6 Ii II i 3 52 ;6 3 25 6 110 100 90 80 70 ppm C Chemical shift Figure 1-11 (A) CP/MAS 1 3 C NMR spectra of Tunicate (Ip-rich,/a//p = 8/92, solid line) and Cladophora ( la -r ich, / , / /^ = 64/36, dotted line) celluloses. (B) The l 3 C subspectrum of the pure cellulose Ip. The 1 3 C subspectrum of Ip phase was derived by subtracting the l 3 C spectrum of tunicate cellulose (solid line in figure A) from that of Cladophora cellulose (doted line in figure A) cellulose. (C) 1 3 C Line spectra of the two magnetically nonequivalent anhydroglucose residues composing cellulose Ip. The lines indicate the 1 3 C chemical shifts of each carbon of the anhydroglucose residues. The line spectra were determined by the 2D refocused CP-INADEQUATE spectrum of tunicate cellulose. Adapted from ref. 43. 13 C5; | . ppmlruiTiTMS| Figure 1-12 Subspectrum expansion of C2, C3, and C5 regions of the CP/MAS 1 3 C NMR spectra of pure I a phases of Cladophora cellulose. Doted lines indicated the result of a line-fitting analysis of the spectrum. Adapted from ref. 44. I ppn* troro TMS| Figure 1-13 Subspectrum expansion of C2, C3, and C5 regions of the CP/MAS l 3 C NMR spectra of pure Ip phases of Cladophora cellulose. Doted lines indicated the result of a line-fitting analysis of the spectrum. Adapted from ref. 44. 1 1 0 1 0 0 9 0 8 0 7 0 ppm Figure 1-14 Expansion of spectrum of the mercerized cellulose II obtained from l3C-enriched bacterial cellulose. Adapted from ref. 23. 14 Chapter 2 Enzymatic Synthesis of Cellulose 2.1 The Chemistry of Cellulose Polymerization In nature, cellulose is synthesized by the cellulose synthase enzymes through biosynthetic pathways using uridine 5'-diphospho-a-D-glucose (UDP-Glc) as monomer building blocks. It has been proposed that the poly-condensation reaction takes place through a single displacement mechanism where the nucleophilic C4-OH at the non-reducing chain end attacks the a-Gj position of UDP-Glc 4 5' 4 6. Therefore, the chain propagation process occurs with the inversion of the anomeric configuration with the UDP-Glc C] carbon inverting from a to P resulting in a P-(l-4)-linkage (Scheme 2-1). The scission of the phosphoester bond between uridine diphosphate (UDP) and the monsaccharide will supply the free energy necessary for the formation of the glycosidic linkage. As mentioned in the previous chapter, in cellulose producing plants, cellulose synthases are found in TC (terminal complexes) embedded in cell membranes and the production of cellulose requires a series of enzyme complexes. Synthetically, it is possible to synthesize cellulose with isolated synthases, UDP-Glc; the propagation of the cellulose chain can also be achieved in-vitro with the above mechanism 4 6 . However, the high cost associated with preparing nucleotide diphosphate sugar substrates often makes the large-scale in-vitro production of cellulose uneconomical. Therefore, there is a need to develop alternative methods for synthesizing cellulose in-vitro. OH OH Scheme 2-1 Elongation of cellulose chain using UDP-Glc with inversion of anomeric center. 15 2.2 Glycosylations and Polysaccharides Syntheses Recently, it has been shown that complex carbohydrates such as oligo- and polysaccharides are very important in cellular recognition and in the immune responses of living systems 4 7 ~ 4 9 . Many practical applications involving oligosaccharides such as liquid crystalline polymers 5 0 , selective membranes 5 1 , sensor matrices 5 2 and bioactive and biocompatible materials 5 3 are also recognized. In a biological context, due to their extensive involvement in many biological processes, complex carbohydrate compounds are also regarded as having highly promising therapeutic potential. Thus there is an increasing demand for reliable methods of the synthesis of polysaccharides which guarantee the production of carbohydrates on a large scale with precise structural control 5 4. Generally, there are two major areas of concern in the synthesis of polysaccharides: (1) controlling the stereochemistry of the anomeric Ci carbon, (2) regioselectively forming the bond with the correct hydroxyl. Therefore, numerous efforts have been devoted to the development of new glycosylating processes which proceed in a regio- and stereoselective manner5 5"5 7. Over the years, methods have been developed for the formation of glycosidic linkages in a controlled manner such as: (i) the Koenigs-Knorr reaction, using bromine or chlorine at Ci of glycosyl donors with AgC104 or Hg(CN)2 as promoter5 8, (ii) glycosyl fluoride with AgC104-SnCi2as promoter 5 9' 6 0, (iii) pentenyl glycoside with C F 3 S O 3 O H - N -iodosuccinimide as promoter6 1, (iv) thioglycoside with C H 3 C O S O 2 C F 3 as promoter 6 2' 6 3, (v) the imidate method and the seteroselective glycosylation method which has been refined recently with glycosyl fluoride to achieve high yields and selectivity using the Lewis-acid catalyst triphenylmethyl tetrakis(pentafluorophenyl)borate 6 4 . However, most of the coupling reactions between saccharide units mentioned above involve laborious protecting-group manipulations and the reactions often result in a range of unwanted side products. Thus, a versatile and general method that can be applied to the preparation of a variety of carbohydrate polymers or oligomers has not been well established. The application of enzymes in polysaccharide synthesis as mentioned in the previous section can be a possible solution to the above problems. 16 2.3 Enzymatic Polymerization Enzymatic polymerization is defined as in-vitro polymerization via non-biosynthetic (non-metabolic) pathways catalyzed by an isolated enzyme. It produces polysaccharides in a free form from a one-step reaction, thereby contrasting with the laborious, multi-step chemical synthesis of polysaccharides involving many protecting groups. Compared to standard chemical synthesis, enzymatic polymerizations also offer much faster reaction rates and the molecular recognition of substrates at the binding site of the enzymes offers regio-and stereo-selective control of the reaction 6 5. At the same time, enzymatic polymerizations are more environmentally 'friendly' and are often referred to as "green polymer chemistry" because of their environmentally benign processes 6 6. The use of heavy metals (as catalysts), organic solvents and production of organic waste can be reduced and in some cases avoided. However, enzymatic synthesis is not without its limitations. Traditionally, the limited availability of appropriate enzymes, specificity of substrates and high production cost are major drawbacks in enzymatic synthesis. However, through modern recombinant DNA technology and site-directed mutagenesis, the availability of enzymes has been significantly increased. Together with added improvements such as substrate selectivity, resistance to high temperatures or co-solvents, the improved enzymes are also capable of carrying out reactions that differ from their original process 6 7. Depending on the type of polymer desired, there is a wide range of 'designer' enzymes suited for different purposes. Generally, enzymes capable of polymerizing carbohydrates can be classified into 4 groups 6 7: (i) phosphorylases (which are involved in the non-Leloir pathway and require sugar-1-phosphates as donors), (ii) glycosyltransferases (which are involved in the Leloir pathway and require sugar nucleotides as donors), (iii) glycosidases (which normally breaks glycosidic linkages but can be manipulated to form glycosidic linkages in a kinetic and/or thermodynamic approach) and (iv) artificial glycosynthases (mutated glycosidases to abolish the hydrolytic activity)68. For the purpose of this thesis, the focus will mainly be on glycosidases and artificial glycosynthases. 17 2.4 Glycosidases As mentioned above, glycosidases are enzymes that break or form glycosidic linkages between saccharide units. They are amongst most studied enzymes to date due to their glycoside hydrolysis ability and wide availability. They are also particularly useful enzymes in this thesis due to their tolerance to organic solvents such as acetonitrile. In general, there are two distinct mechanisms for the process of glycoside hydrolysis 69"72. One is the direct displacement mechanism that results in the 'inversion' of anomeric configuration whereas in the other, a double displacement mechanism results in the 'retention' of the anomeric configuration (Figure 2-1). Despite their differences, both reactions take place through two catalytic carboxylic acids residues (Scheme 2-2, Scheme 2-3). In the direct displacement inverting mechanism, one of the carboxylic acid residues acts as a general acid when the other acts as a general base (Scheme 2-3). However, in the double displacement retaining mechanism, one of the residues acts as general acid/base while the other acts as a nucleophile/leaving group O R ' Figure 2-1 Illustration of the stereo-chemical outcome of (a) retaining and (b) inverting glycosidases. 18 a) Acid/Base Catalyst A / H ^ ° H 0 -•OH [ ^ 0 H Q V . -r-L^Q ^ Nucleophile d) ^ ° H O . Acid/Base Catalyst t e) Nucleophile k Acid/Base Catalyst .A / H O H 0 ^ ° H 0 -.OH b) Acid/Base Catalyst A I ^ H O -c) H ft -o-A,-- HO-O H ; 0 Nuc _ L Nucleophile i O H •0 Acid/Base Catalyst H c 0 H -o— O H Nucleophile O H Nucleophile Scheme 2-2 Generalized double displacement mechanism of a retaining P-glycosidase. 19 a) Acid Catalyst fJ N f 0HO-OH f ° V OH O J 1 & Base Catalyst Acid Catalyst Base Catalyst Scheme 2-3 Generalized direct displacement mechanism of an inverting p-glycosidases. The mechanisms are distinguished by identifying first-formed products - typically by 'H-NMR analysis73. Interestingly, in order to accommodate the nucleophilic attack of a water molecule, the two residues are generally farther apart in the inverting than in the retaining glycosidases. In the retaining glycosidases, the distance between the two catalytic carboxylic acid side chains are generally between 5-6 A. apart while in the inverting glycosidases it is anywhere between 7-10 A (Figure 2-2)74"76. Therefore theoretically, it is 20 possible to interconvert the two mechanisms from one to the other by changing the separation in the active site via mutations of catalytic residue(s)74. Kuroki and co-workers have illustrated that the inverting glycosidase T4 phage lysozyme can be converted into a retaining glycosidase by the mutation Thr26His 7 4 , 7 7 where the imidazole side-chain of Thr26His in the mutant occupies the position normally assumed by the water molecule, hence blocking the nucleophilic attack of water (Scheme 2-4). a) ^ ° H O Acid/Base Catalyst OH ( ° H i o<C^o 1? 5 -6 A Acid/Base Catalyst b) ^ ° H 0 Acid/Base Catalyst o<°^o OH ( OH ) H H J 1 - 10 A Acid/Base Catalyst Figure 2-2 General illustration of the distance between the two catalytic carboxylic acid of (a) retaining glycosidase and (b) inverting glycosidase. Scheme 2-4 The proposed retaining mechanism of the Thr26His nucleophile mutant of T4 phage lysozyme. 21 2.5 Transg lycosy la t ion For most wild-type glycosidases under normal conditions, the hydrolysis of glycosidic bonds is favoured. However, retaining glycosidases have been extensively used in chemo-enzymatic oligosaccharide synthesis under controlled conditions which favor reversal of the hydrolytic reaction of wild-type retaining glycosidases: (1) equilibrium controlled synthesis 7 8 or (2) kinetically controlled transglycosylation 6 7 (Scheme 2-6). The first approach utilizes the Le Chatelier's principle where a large excess of mono- or oligosaccharides (glycosyl donors)79'80, elevated reaction temperatures and the addition of various organic co-solvents ' and salts (to lower water activities) can be used alone or in combination to reverse the direction of the reaction from hydrolysis to glycosylation. The second approach focuses on promoting the formation of the glycosyl enzyme intermediate by using activated glycosyl donors such as glycosyl fluorides or aryl glycosides (Scheme 2-5). The rationale is that the reaction between the glycosyl enzyme intermediate and glycosyl acceptor is expected to occur more easily than that with the water molecule, leading to mainly glycosylation. However, due to the inherent hydrolase activity of the enzyme, the newly-formed glycoside bond can also be readily hydrolysed leading to rather low yields of condensation products at times. a) b) Acid/Base Catalyst Acid/Base Catalyst Nucleophile glycosyl enzyme intermediate Scheme 2-5 Formation of glycosyl enzyme intermediate with glycosyl fluoride donor. 22 c) d) Acid/Base Catalyst Acid/Base Catalyst glycosyl enzyme Nucleophile intermediate Scheme 2-6 Hydrolysis and transglycosylation reactions proceeding from the glycosyl enzyme intermediate. R = H gives hydrolysis reaction; R = sugar gives transglycosylation reaction. 23 2.6 Artificial G l y c o s y n t h a s e s (Mutated Glycos idases ) Glycosynthases are defined as mutated glycosidases that form glycosidic bonds without hydrolysing the products 8 4. The Withers group first introduced the concept of glycosynthases in 1998 with the exo-enzyme p-glucosidase from Agrobacterium faecalis. Through site-directed mutagenesis, the catalytic nucleophile of a retaining p-glycosidase is replaced by a smaller non-nucleophilic residue (e.g. Ala or Gly) in order to produce an enzyme which is hydrolytically inactive towards the unactivated O-glycosidic linkages 85~87. a) Acid/Base Catalyst b) Acid/Base Catalyst CH, CH, non-nucleophilic residue non-nucleophilic residue C) Acid/Base Catalyst OH OH CH, non-nucleophilic residue Scheme 2-7 Generalized inverting mechanism of a retaining P-glycosidase nucleophile mutant. 24 However, the mutated enzyme will still be able to catalyze the transglycosylation reactions if glycosyl fluoride donors with opposite anomeric configuration are used instead of the normal substrate of the wild-type glycosidase (Scheme 2-7). In the active site of a nucleophile-less mutant of a P-glycosidase, an a-glycosyl fluoride donor can be used to mimic the glycosyl-enzyme intermediate of the wild-type enzyme and transfer its glycone unit to an acceptor in the acceptor sub-site. Because the anomeric carbon of an a-glycosyl fluoride is directly attacked by the glycosyl acceptor from the p-side resulting in products with inversion of the anomeric configurations, the transglycosylation of the mutant enzyme proceeds via an inverting mechanism instead of the retaining mechamism of the parent wild-type enzyme. Also, due to the lack of a nucleophilic residue, the mutated enzymes will be unable to hydrolyse the product(s) formed, therefore the yields of transglycosylation will be very high compared to those of the wild-type glycosidases under equilibrium or kinetic control85'88. Other than exo-enzyme P-glucosidase, many other glycosynthases have been constructed since then from various endo- and exo- glucanase, glucosidase and cellulase89. 25 2.7 Glycosyl Fluorides In both the wild-type and mutant glycosynthases, the use of glycosyl fluorides is crucial to the formation of glycosidic bonds. More specifically, for the enzymatic synthesis of cellulose in this thesis, cellobiosyl fluorides and derivatives of them are used as the monomeric starting materials for a number of reasons. Firstly, the monomer design using a two-glucose-unit structure is very important because the disaccharide unit (cellobiose) is the smallest unit readily recognized by cellulases in producing cellulose. When glucosyl fluoride with a monosaccharide structure was used, only cello-oligomers are produced 4 5 making it an unsuitable substrate for true polymerization. Secondly, the small atomic radius of fluorine is very similar to that of oxygen or even the hydroxyl group (Table 2-1), which provides structural similarity and little steric demand for molecular/substrate recognition in the active centre 9 0 - 9 2 . Next, the fluoride anion is also a good leaving group, facilitating the nucleophilic attack of the acceptor. Of all the glycosyl halides, the glycosyl fluoride is the only one that is sufficiently chemically stable in the unprotected form in aqueous buffers to avoid appreciable rates of spontaneous hydrolysis 90"93. Additionally, the fluorine atom in a C-F bond is believed to be capable of forming hydrogen bonds with hydrogen bond donors but not hydrogen bond acceptors 9 I' 9 4. Lastly, as the naturally occurring isotope of fluorine, 1 9F has a spin number of Vi and can be easily detected by NMR spectroscopy making the direct detection and monitoring of fluorine possible even when the molecule of interest is interacting with an enzyme. Element Bond-length (C-X) van der waals radius (A) H 1.09 1.20 0 1.43 1.40 F 1.37-1.42 1.35 Table 2-1 Bond-lengths and var der waals radii of some elements. 26 Chapter 3 Nuclear Magnetic Resonance 3.1 Introduction Nuclear Magnetic Resonance (NMR) spectroscopy 9 5 - 1 0 1 is one of the most powerful and useful techniques available to chemists for the investigation and elucidation of molecular structures and dynamics and can be applied to a diverse range of complex chemical systems. NMR occurs when certain nuclei in a static magnetic field are exposed to an oscillating radiofrequency source. These interactions provide a great deal of information about the local magnetic environments of nuclei, from which information about local structures and dynamics can be obtained. The theory of NMR is discussed in great detail in a number of (a) (b) Precessional orbit U l l i ( -•JNuclear magnetic dipole, p. r Spinning proton Figure 3-1 Description of a proton. (a) Rotating on its nuclear axis generating a magnetic dipole. (b) In a magnetic field B 0 precessing about B 0 . 27 Nuclear Magnetic Resonance takes advantage of the fact that all atoms contain protons. These protons have an innate fundamental property called spin or angular momentum. In a nucleus, each of these individual nucleons can couple their individual spins or angular momenta to yield a total nuclear spin that can have an integral, half-integral or zero value. Any nucleus with a spin / not equal to zero will show NMR properties. When such a nucleus rotates (Figure 3-1 a) it generates a magnetic moment ju (Figure 3-1 b) which is related to the total angular momentum, P and the nuclear gyromagnetic constant (^an innate property of the nucleus) as expressed by Equation 3-1. Equation 3-1 fl= y P In the presence of an external magnetic field Bc, the magnetic moment will precess about the field at its Larmor frequency (Figure 3-1 b) which has a characteristic value for a given type of nucleus. In this case, each spin can be subdivided into m/ = (21+1) spin states of different energies as expressed by Equation 3-2. In the absence of an external magnetic field, the spin states still exist but they are degenerate and have identical energies. Equation 3-2 E = -y h miB„ The nuclei of interest in this study, *H, 1 3C and 1 9F are all 'spin !4' nuclei; that is, they have (27+1) = (2 (!4) + 1) = 2 spin states: (Figure 3-2) a spin "up" (mt = + !4) and a spin "down" (mi = -14) with an energy separation of AE as shown by Equation 3-3, where v is the Larmor precession frequency in Hz. Equation 3-3 AE = h v= I yft B01 28 m, = +1/2 magnetic field Bo=0 m, = +l/2 m, = -l/2 A E = yftB0 incraesing magnetic _ field strength B 0 Figure 3-2 field B„. Energy difference between spin states of spin Vi nuclei in the presence of external magnetic According to the Boltzmann distribution (Equation 3-4), there will be an excess population of spin "up" (lower energy states) at equilibrium when an external magnetic field B0 is applied. This uneven distribution of spin states will result in a net magnetization M (Equation 3-5), along B0 or the positive z-axis as shown by Figure 3-3. Equation 3-4 A r/7V+ = e' Equation 3-5 M 2kT where AE is the energy difference between the spin states; k is the Boltzmann constant, 1.3805xl0"23 J/Kelvin; and T is the temperature in degrees Kelvin. 29 z Figure 3-3 Magnetic behavior of protons in a magnetic field B 0 . (a) Precession of the magnetic moment about the external field B„. (b) Excess of spins in spin "up" state in a 'precessional cone' (mi = + V-i) in the presence of field BQ. (c) Depiction of the excess of spins in the spin "up" state as represented by a net magnetization vector ( M ) in the field BQ. When a radio frequency (rf) field B i « B 0 that matches the resonance or Larmor frequency of the system of interest is applied perpendicular to B 0 , the more abundant lower spin states can absorb the energy and be promoted to a higher energy level spin state. In most spectrometers the radiofrequency is applied through a coil surrounding the sample that is connected to an appropriately matched and tuned rf circuit. Therefore in order to properly visualize the process of magnetization and relaxation, it is necessary to change the viewing frame from a traditional static Cartesian frame of reference to the rotating frame of reference (Figure 3-4). The rotating frame rotates around the z-axis ( B 0 ) at a frequency equal to the angular frequency of the applied rf field. Unless otherwise mentioned, in the rest of this thesis, all of the vector diagrams will be described using a rotating frame of reference. 30 • precessing about B 0 with frequency <o0 after time t e) B , Cartesian i j rotating at oo i rotating frame o M Figure 3-4 Precession of net magnetization M . (a) 90° pulse (b-c) precession of net magnetization M around B„ in a lab frame (d) a rotating Cartesian frame (e-f) net magnetization M in a rotating frame as time passes by. In a rotating frame, the applied field B i causes the spins to precess about the B i field at a frequency a>^=yBx. It is possible to rotate the net magnetization M by any angle by applying B i field for a set period of time. The changes in the direction of M are given by the equation Equation 3-6 0 = O)J 31 In general, the rf field is applied along the x-axis for a period of time t such that the net magnetization is rotated to an angle 9= 90° onto the xy-plane (Figure 3-5). Once the magnetization is rotated 90°, the applied field is turned off and the system is allowed to undergo 'free induction decay' (FID) and relax back to its Boltzmann distribution. Figure 3-5 (a) Effect of B, pulse along x-axis on net magnetization M . (b) Net magnetization vector M after a 90° pulse. The NMR spectrum is obtained by monitoring the free induction decay (FID) of the magnetization M to zero as the spins dephase. The precession of the net magnetization in the xy-plane about B0 induces an electrical current in the coil, which is then amplified and detected. A mathematical function, the Fourier transformation (FT), is applied to translate the FID signals in the time domain to yield a NMR spectrum in the frequency domain (Figure 3-6). Depending on the environment each nucleus is in and its interactions with other surrounding nuclei, it will experience a slightly different magnetic field causing a shift in the resonance frequency of the particular nucleus, thereby shifting the rf absorbance in the NMR spectrum giving us diagnostic information about its chemical environment. This is formalized as the 'chemical shift' of the resonance. 32 FID NMR Spectrum Figure 3-6 The transformation of time domain FID signal to the frequency domain NMR spectrum through a Fourier transform. 33 3.2 Relaxation T i m e s The decay of the magnetization after a pulse (FID) and the re-establishment of the Boltzmann distribution are described by 'time constants' related to the first order rate constants of the processes. They often define the limits of NMR experiments involving multiple pulses, spin-echoes (T?) or spin-locking (T/p). Relaxation arises from motion-induced changes in local magnetic fields such as shielding anisotropy, dipole-dipole interactions, quadrupolar interactions and interactions with unpaired electrons. These changes can be measured and quantified by NMR techniques, providing additional information about the systems under study. 3.2.1 Sp in -La t t i ce Re laxat ion T i m e (7\) When performing FT-NMR experiments there are two very important relaxation times to consider, Ti and T2. Ti is the 'spin-lattice' relaxation time and describes the rate of recovery of the longitudinal magnetization to its equilibrium population level (Figure 3-7). To be exact, Tj is equal to 1/k for the first order recovery process. It takes five times Ti for a perturbed magnetized system to effectively recover its spin populations to the Boltzmann equilibrium distribution. In other words, Ti determines the minimum delay time required between collections of successive FIDs if quantitative information is to be obtained. It limits the S/N that can be achieved during a given experimental time. In terms of the vector model, T; is the time required for the net magnetization vector to return to its original position and magnitude along the z-axis after excitation. This relaxation process is described as a first order kinetic process by the differential equation Equation 3-7 — Mz = -(Mz - M0)/ T, dt with Mo being the unperturbed equilibrium magnetization. 34 Figure 3-7 The process of spin-lattice relaxation in a rotating frame. (a) Net magnetization vector in M the presence of external field BD. (b) Net magnetization vector M after a 90° pulse along the x-axis. (c-e) Vector representations of the spin-lattice relaxations. Ti is usually measured by the inversion-recovery sequence shown in Figure 3-9106. As mentioned above we can "tip" or change the direction of the net magnetization vector by applying the rf pulse for the appropriate period of time. In the inversion-recovery sequence, a 180° pulse is applied to change the net magnetization from the positive z-axis to the negative z-axis. Then after a time period x during which the system relaxes towards its original Boltzmann equilibrium, a 90° pulse is applied to tip the magnetization to the xy-plane where the detector measures the sign and the magnitude of the resulting magnetization. A series of spectra is obtained for different x periods, and the rvalue where maximum signal is obtained is equal to ~ 5 x Ti (Figure 3-8). It is important to obtain accurate Ti values because in order to obtain quantitative data in NMR studies, one should have a repetition time TR of at least five times Ti between successive repetitions of pulse sequences to allow for the system to fully revert back to its equilibrium state. The change in magnetization can be shown by the following equation. r Equation 3-8 Mz = M„ (1 - 2 e T' ) Ti is therefore defined as the time required to change the Z component of magnetization by a factor of (1 -r 2 e 71). 35 Intensity Figure 3-8 Recovery of net magnetization along z-axis in terms of Tt. Maximum intensity is recovered after 5T,. 2T, 3T] 4T, 5T, 6T, 7T, 8T, Relaxation time In general, Ti relaxation time can be obtain through either inversion recovery (Figure 3-9) or saturation recovery (Figure 3-10). To reduce experiment time, T/ in our experiments will be obtained by saturation recovery. 36 I n v e r s i o n R e c o v e r y a) 1 8 0 ° ( . r ) 9 0 ° ( r ) Figure 3-9 Measuring Tt spin-lattice relaxation times using inversion recovery method, (a) Inversion recovery pulse sequence (f) Effect of x on signal intensity 37 Figure 3-10 Measuring T/ spin-lattice relaxation times using saturation recovery method, (a) Saturation recovery pulse sequence. repeats (b-e) vector description of the net magnetization as a consequence of the saturation recovery pulse sequence t (f) Effect of x on signal intensity 38 3.2.2 Spin-Spin Relaxation Time (T2) Once the T\ of a system is obtained one can measure the spin-spin relaxation time T2, which is always less than or equal to Ti. T2 is the time required for the transverse magnetization to return to its equilibrium value (zero) after the application of the B) field. It is the rate at which net magnetization is lost in the xy-plane after a pulse. The T2 relaxation time reflects the environment and behaviour of the system under study due to the dephasing of the individual magnetic moments as they precess around BD (Figure 3-11) and is also a very important parameter in NMR experiments that depend on conserving transverse magnetization such as INADEQUATE, TEDOR and REDOR. The relaxation process can again be described by first order kinetics: Equation 3-9 —Mx= ~ M X IT2 dt In a 'homogeneous' environment, after the application of a 90° pulse, the FID will decay only through the spin-spin relaxation mechanism with a time constant T2. However due to magnetic field inhomogeneities, following a 90° pulse the net magnetization of most system will be dephasing at a rate of T2* instead of the true spin-spin relaxation time T2. The Fourier transformation of an exponentially decaying FID with the time constant T2 or T2* provides a Lorentzian peak with a line width of (x T^)'1. Therefore a decay time of T2* causes a line broadening of (^TV-T!?))"1 m m e resulting resonance. In order to remove inhomogeneous broadening, the spin-echo experiment is used to 'refocus' these inhomogeneities. The decay of the magnetization at the end of the spin echo is the result of only spin-spin relaxation. Figure 3-11 The process of spin-spin relaxation in a rotating frame. (a) Net magnetization vector after a 90° pulse along the x-axis. (b-d) The process of dephasing of the net magnetization vectors in the spin-spin relaxation process. 39 c) (e) observed magnetization as a function of the delay time x. T2 is usually determined using the spin-echo sequence 1 0 7 shown in Figure 3-12. First, a 90° pulse is applied along the y-axis in the rotating frame tipping the magnetization onto the xy-plane. Then, during a delay period of time x, the individual spins in the ensemble will experience slightly different magnetic field strengths and rotate at slightly different Larmor frequencies because of the field inhomogeneities. Therefore, the spins will start to dephase, leading to a decrease in signal intensity. After a delay time x, a 180° pulse is applied to the system in order to rotate the magnetization about the x-axis. Now the slower moving spins will be moving ahead of the faster moving spins. After another x period of delay, the spins refocus themselves as the effects of field inhomogeneity are exactly removed and the signal intensity is recorded. There is, however decay due to T? and the echo signal will decrease in intensity. The two x periods of delay are equal to the echo time, TE-As in the T\ experiment, in order to get the time of echo TE, a series of spectra are obtained at different x values and the T? value obtained using Equation 3-10 MZ = M„ e An alternative method to obtain T2 values is by using a multi-echo sequence (Figure 3-13). The advantage of a multi-echo experiment is that it allows us to obtain T2 values in a much shorter time. After the first 90° (spin) and 180° (echo) pulse pair, another echo 180° pulse is applied, and the sequence repeated by adding 180° pulses from there on. 40 Figure 3-12 Method of measuring T2 spin-spin relaxation times, (a) Spin-echo pulse sequence y 41 Figure 3-13 number of echoes (a) Multi-echo pulse sequence for measuring T 2 relaxation time c) 90° (x) pulse / / ' / t = 0 T delay spins dephase in ^ v-plane spin refocus to form echo t = 2T 180°(y) pulse fast slow multi-echo (b-e) Vector description of the multi-echo pulse sequence. 42 3.2.3 Spin-Lattice Relaxation Time in the Rotating Frame {T1p) Besides Ti and T2, there is another important relaxation time in solid-state NMR studies. As described above, applying a rf pulse along they-axis can tip the net magnetization onto the xy-plane along the x-axis which is static in the rotating frame of reference. Normally spin-spin relaxation (T2) follows. However, if at this time, the r.f. field Bi is "phase shifted" and applied along the x-axis, the net magnetization will be 'spin-locked' in the xy-plane along x-axis and the system will experience a different type of relaxation, the spin-locked magnetization (MSL)- In this 'spin-locked' state, the net magnetization spins precess about Bi as long as Bi is applied rather than decaying due to 71/ or 7]?. In the rotating frame, this spin-locking Bi field acts as a static magnetic field, and decay of the magnetization occurs in an analogous manner to conventional spin-lattice relaxation. This decay of the spin-locked magnetization (MSL) can be described by the spin-locked relaxation time Tip. Equation 3-11 ^ M S L = ~ M S L I T ^ P The measurement of T!p relaxation times by the spin-locking experiment is illustrated in Figure 3-14. 43 a) 9 0 " (x) spin-lock ing B, (y) 4 X Figure 3-14 Method of measuring T ] p relaxation times, (a) Spin-locking pulse sequence. t = 0 (b-e) Vector representation of the magnetization and spins in spin-locking experiment ^ M ( r ) = M0e""n" K M(r) t = T 1 I (f) Observed magnetization as a function of spin lock time x. 44 3.3 High Resolution Solid State NMR Traditionally, NMR studies have mainly focused on liquid and solution samples. In solution, the molecules are free to rotate and tumble isotropically and therefore many of the orientation-dependent nuclear spin interactions are averaged out, resulting in narrow and highly resolved signals. However, in solid samples, the molecules experience much less motion and the orientationally dependent nuclear spin interactions are not averaged out, resulting in a series of broad and featureless signals (with the exception of single crystal NMR). Early proton NMR studies of solids were mainly based on exploration of the dipolar interactions between groups of protons as probes of molecular motions in the solid state. There was no possibility of acquiring spectral information in a detailed manner as can be done for liquid state NMR spectra. However, for materials that are insoluble or amorphous, structural studies are difficult. Polymers, cellulose, zeolites, heterogeneous catalysis, semiconductors, resins, glasses, coal, wood, soils and minerals are such examples 1 0 8 - 1 1 0 . 45 3.3.1 Nuclear Spin Interactions In solution, the nuclear spin interactions contributing to the final NMR spectrum can be described using the following Hamiltonian Equation 3-12 H = Hz + H'^ + H'^ where in a solution the Zeeman nuclear spin interaction Hz , the magnetic shielding H'cs and the scalar spin-spin interactions J (H™s) dominate. However, in solids the interactions are "anisotropic" or orientation-dependent and several additional nuclear spin interactions must be included as described by the general Hamiltonian Equation 3-13 H = Hz+ Hcs + HD + Hsc + HQ where HD is the dipolar through-space interaction between nuclei and HQ is the quadrupolar interaction with an electric field gradient for spins > Vi. Since all of the nuclei involved in this thesis are of spin lA, HQ does not affect the resulting spectra and will not be discussed here. In addition to the dipole-dipole interactions (HD) and quadrupolar interaction (HQ) there are also the anisotropic forms of the nuclear shielding Hcs and scalar coupling Hsc that will affect the resulting spectra. Therefore the observed NMR spectra will be dependent on the orientation of each nucleus and, in a powder, the final spectrum will be the sum of many overlapping NMR resonance frequencies of different 'orientations' leading to severe line broadening of the resonances 10°. 46 3.3.2 Zeeman Nuclear Spin Interaction Hz One of the most important spin interactions of both solution and solid-state NMR spectra is the Zeeman interaction. As mentioned previously, in solution and the solid state, when the appropriate rf frequency is applied, transitions between energy levels are possible: Equation 3-14 a>0 = = y B0 h where co0 is the 'Larmor frequency' in rad s"1 and ra0 = G>0/2K is the Larmor frequency in Hz. Compared to the Zeeman interaction, all other spin interactions are small perturbations in comparison. However, as in solution, these small variations in the Zeeman energy levels often can yield structural information about the system under study. 47 3.3.3 Dipolar Interaction HD The dipolar interaction is the most important isotropic mechanism causing line broadening in the solid state. It is a through-space interaction that describes the local magnetic field a nucleus experiences from the magnetic moments of the nuclear spins of neighbouring nuclei (Figure 3-15). It can either be between nuclei of the same type (homonuclear interaction e.g. 'rl-'H) or between nuclei of different types (heteronuclear interaction e.g. 'H-13C). These interactions influence the Zeeman energy levels causing a small perturbation in the energies. The magnitude of the dipolar coupling (bis) is described using the following equation where b/s is a function of the gyromagnetic ratios yi and ys of the nuclei, the internuclear distance r and the orientation of the internuclear vector with respect to the applied static magnetic field defined by the angle 9. d/^ and Dis are the 'dipolar coupling constants': Equation 3-15 Equation 3-16 = M0r,rsh 4nr3 (in rad s") Equation 3-17 = dis = MJJsn In %n2r3 (in Hz) / where jua = An x 10 7 kg m s2 A 2 is the 'permitivity of free space' and h = hl2n where h = 6.6262 x 10~34 J s is the 'Planck constant'. The effect of dipolar interactions on solid-state spectra can be reduced using MAS and removed by high power decoupling techniques as will be described later. Figure 3-15 Pictorial representation of the internuclear vector in the presence of external magnetic field B 0 . 48 3.3.4 Magnetic Shielding (chemical shift) H c s Nuclei are not isolated entities; they themselves are surrounded by spinning charges such as electrons and other nuclei. These spinning charges can influence the magnetic field or local electronic environment each nuclei experiences through a process called 'shielding'. The variation in local magnetic field experienced can be defined using Equation 3-18 AB = ±-^(3 cos2 9- l)r~3 — 2 4/r where AB is the local magnetic field variation caused by the mutual interaction of two nuclear magnetic moments separated by a distrance r, u0 is the permeability in free space and 9 is the angle by which the orientation of the vector between the nuclear magnetic moments deviates from B0. Depending on the orientations of surrounding electrons, each nucleus will experience a local magnetic field slightly different from the applied static magnetic field. This slight variation in experienced magnetic field gives rise to transitions at slightly different frequencies than the Larmor frequency, which in rum, translates into different 'chemical shifts' in the resulting spectra. Hence, the local electronic shielding provides information about the local environments of the nuclei within the system. Since all molecules are freely rotating and tumbling in a solution, all of the orientations are averaged out and the shielding of a nucleus by the surrounding electrons is usually 'isotropic', that is, the average value over all-orientations. However, in a solid powder sample where local molecular motion is limited, the chemical shift will be dependent on the orientation of the nuclei in the different crystallites with respect to the static magnetic field leading to 'chemical shift anisotropy' (CSA). As a result, the resulting spectrum will contain broad lines due to the overlapping of many chemical shifts. Fortunately, this anisotropy can be removed by the magic angle spinning (MAS) technique described below giving isotropic chemical shifts analogous to those seen in solution spectra. 49 3.3.5 Indirect Spin-Spin Coupling (J coupling) Hsc Unlike the dipolar coupling, the indirect spin-spin coupling is a through-bond ' J-coupling' between two nuclei and is independent of the applied magnetic field strength. It is much smaller than the dipolar coupling, the ./-coupling and can also be both homonuclear and heteronuclear in nature. However, due to its small size, its anisotropy is almost always ignored. However it still provides perturbations to the Zeeman energy levels, and can give rise to additional resonance frequency separations resulting in multiplet structures as in solution. These multiplet structures could provide very useful information about the conformations and structures of molecules. However since its effect drops rapidly as the number of bonds between the nuclei increases, it is most useful for describing geometric arrangements between closely bonded nuclei. 50 3.4 High Resolut ion C P / M A S 3.4.1 Magic Angle Spinning As mentioned before, the fixed orientation of molecules in powdered solids causes an anisotropic broadening of the chemical shift (CSA) making it difficult to obtain sharp peaks. Recall that the variation of local magnetic field experienced by a nucleus due to chemical shift anisotropy can be described by the general equation: Equation3-19 AB = ±-/y(3cos2 6?-l)r~3^°-2 4n and 6 is the angle the magnetic moment vector makes with the external magnetic field B0. In a solution, due to the random thermal translation and rotational of molecules, the orientation dependent factor of 3cos2 0-\ can be readily replaced by the time average sum^ (3 cos2 6X y . -l)/3 which equals zero for freely rotating molecules. In a solid, the sum of the orientation-dependent factors does not equal to zero therefore resulting in low resolution lines of peaks that are several kHz wide 10°. However if we take the orientation dependent factor of 3cos2r7-l and set it to zero, one finds that at an angle of 54.74°, the orientation dependence term will be equal to zero and the CSA will be averaged to its isotropic value as in solution. 3cos2r7-l = 0 r? = cos-'(l/V3) = 54.74° 51 Experimentally, if one rapidly rotates a solid sample packed in a rotor around an axis orientated at 54.74° with respect to B0 (Figure 3-16), the CSA can be removed, producing a high-resolution NMR spectrum with isotropic chemical shifts. This technique is called magic angle spinning (MAS) 1 1 1 - 1 1 3 with 54.74° being the 'magic' angle where all CSA disappear 1 0 0 It is also interesting to note that at an angle of 54.74°, the dipolar coupling also gives a time-average value of zero. Recall in dipolar coupling, the magnitude of the dipolar coupling (bis) is described by Equation 3-20 bis = ±^-(3 COS 2 5-1) where the orientation dependent factor is also 3 cos2 & -1 Therefore, all the dipolar interactions will cancel at the angle of 54.74° and theoretically the effects of dipolar couplings can be removed by magic angle spinning. However, in order to remove all the heteronuclear and homonuclear dipolar couplings, the sample rotation rate must be greater than that of the strength of the interactions in a completely static sample. In the case where MAS cannot completely remove large dipolar interactions ('H being the most common situation), it is also possible to employ high-power decoupling techniques to remove such interactions if they are heteronuclear (e.g. 'H/13C). spinning rotor Figure 3-16 Pictorial representation of a spinning rotor in the process of magic angle spinning (MAS) in a magnetic field B 0 with 6m 54.74°. 52 3.4.2 Cross Polarization Other than CSA and dipolar couplings, depending on the nature of the nuclei under study, there are also other complicating factors in attempting to efficiently obtain high-resolution spectra. In general, nuclei can be divided into four groups: Group 1: high natural abundance spin lA nuclei (e.g. lH, 19F) Group 2: low natural abundance spin Vi nuclei (e.g. 1 3C, 15N, 29Si, 31P) Group 3: full integer spin I> 1/2 nuclei (e.g. 2H, 6Li) Group 4: Vi integer spin I>l/2 nuclei (e.g. 1 'B, 170) Of the four groups, the most problematic are the group I nuclei where there are strong homonuclear dipolar interactions due to the high natural abundance. Groups 3 and 4 are also problematic due to the presence of quadrupolar interactions. It is in group 2 where the nuclei are 'dilute' either due to low natural abundance (e.g. 1 3C, 29Si) or their presence being rare in most systems (e.g. 31P) that most of today's solid state techniques are applied 10°. There are no homonuclear dipolar interactions because they are 'dilute' but this also means low sensitivity. Central to them is one of the most widely used NMR techniques, the cross polarization (CP) experiment. 53 3.4.3 H a r t m a n n - H a h n E x p e r i m e n t Cross polarization (CP) initially started out as the Hartmann-Hahn experiment114. It is found that if both the abundant I spin and the S spin are 'spin-locked' in the xy-plane by two 'matched' spin locking rf frequencies, polarization can be transfered from the I spin to the S spin if the following Hartmann-Hahn conditions are satisfied. Equation 3-21 7/B i / = ysB2s where yj and ys are the gyromagnetic ratios of the I and S spins, By « Bo and Bis « Bo are the spin locking rf fields of I and S respectively. In order for polarization to occur the two spin locking fields must be applied in such a fashion that the magnetizations are precessing about the applied fields at matching angular frequencies C O l / = C O i s where coi/ = y/Bi/ and coi,s = ysBis When the above conditions are satisfied, the resonance exchange of energy between the two spin systems can take place through a mutual spin flip mechanism mediated by the I-S dipolar couplings (Figure 3-17). The polarization of the dilute S spin can be enhanced to a maximum factor of yi/js-54 r I spins r S spins Y /B 1 / =CO 1 / = 03 2 S =Y 5 B 1 5 <Oo/= Y/Bo m 0 5 = Y s B o spin-locked states 'thermal contact' Figure 3-17 Energy level difference diagram of the cross polarization spin-locking 'thermal contact'. 55 3.4.4 Cross Polarization Experiment In general, the dynamics of cross polarization can be understood thermodynamically as the flow of heat ('thermal contact') between the / and S spin reservoirs. Due to its large abundance, the spin / system is generally described as a large thermal reservoir with high heat capacity whereas the dilute S spin system is described as a small reservoir with a small heat capacity. Because the / nuclei are generally strongly interacting with a short spin-spin relaxation, time and a long spin-lattice relaxation time, it is permissible to describe it as having a 'spin temperature' T\ which is different from the lattice temperature To. Equation 3-22 T, = (Bu/ B0) Tn In thermodynamic studies, temperature is inversely proportional to the polarization/magnetization. Therefore, in the spin locking rf field B i / « Bo, the abundant spin / system can be described as having a low 'spin temperature' whereas the rare S spin system has a high 'spin temperature'. When the two spin systems are brought into 'thermal contact' with each other by satisfying the Hartmann-Hahn conditions, the S spin system can be 'cooled' towards the spin temperature of the / spin system. Experimentally, the / spin system is first excited by a 90° pulse along the x-axis, and then a 'spin-locking' rf field B i / with nutation frequency of © i / is applied along the v-axis. In this spin-locking state, the relative populations of the two energy levels are Equation 3-23 N~ IN+ = e(-yhB" l k T ' ] Recall the population differences arising from the Zeeman interaction of the spins with the strong magnetic field BQ. 56 Equation 3-4 N~ /N = e 1 where AE = h v= | y ^  B01 is the energy difference between the spin states giving Equation 3-24 N- / N + =e(-r*BJkTL) Comparing the above two population equations one gets Equation 3-25 T,/TL = B,/B, For a spin locking rf field Bi«B 0 , one would get the spin temperature Ti«TL. As mentioned above, spin temperatures are inversely proportional to the magnetizations. Therefore, in an rf field Bi«B 0 , there will be a large magnetization precessing about they-axis. Simultaneously, another spin locking rf field Bis with matching angular frequency o>is = oil/ about Bis is also applied along the y-axis. The common angular frequencies between the two systems provide a pathway where the energy-conserving spin-flip (transfer of polarization) can occur through mutual dipole-dipole interaction. Depending on the duration of the 'contact' time T, the magnetization of the S spin can be increased. The polarization process will continue until the spin temperatures of the two spin systems are equalized. Since the heat capacity of the abundant I spin is much larger than that of the dilute S spin, the S spin system can be cooled to the initial spin temperature of the I spin system without significantly affecting the I spin temperature. Equation 3-26 Ts = T, = (B,/B0) TL = (ys/yd (B2/B„) TL 57 According to the above equation the spin temperature of the S spin system can be decreased by a factor of ys/yi- Since polarization is inversely proportional to the spin temperature, the polarization or magnetization of the dilute S spin can be enhanced up to a maximum factor of ji/js- For example, in a 'H - 1 3C cross polarization it is possible to enhance the polarization of the 1 3C by a factor of approximately four (Figure 3-18). z z apply spin-lock z z cross polariation of rBgncuzauon from abundant I spin to rare S spin 4 z apply notching z net magnetization z Figure 3-18 Vector representation of the cross polarization process. Other than the magnetization enhancement provided by the cross polarization process, the signal to noise (S/N) ratio is also improved during a CP experiment because typically the spin-spin relaxation time T\ of the abundant spins is much shorter than that of the dilute S spins. Since in a CP experiment the delay time between collecting successive FIDs only depends on the 7/ relaxation time of the abundant / nuclei, in a given amount of time more FIDs can be collected thus increasing the S/N of the spectra (Figure 3-19). 58 9 0 ° (.,) spin- locking B ] (v) polarization transfer spin- locking B 2 0') d e c o u p l i n g Figure 3-19 The cross polarization pulse sequence. 59 Chapter 4 Results and Discussion 4.1 Syntheses of Cellobiosyl Fluorides In order to be utilized as a substrate by cellulases to produce cellulose, cellobiose has to be converted into a cellobiosyl fluoride for the reasons mentioned in previous chapters. Synthetically, cellobiosyl fluorides are also good choices due to their ease of preparation and the possibility of long-term storage without significant decomposition. Because both wild type and mutant enzymes will be employed for the polymerizations in this thesis, it is necessary to synthesize both a- and P- cellobiosyl fluorides due to the fact that the different mechanisms involved require different starting materials in order for the polymerization reaction to proceed. The details of the mechanisms and reasons for the requirement of the different starting materials will be described later in this chapter. In general, wild type enzymes require P-cellobiosyl fluoride as substrate while the nucleophile mutant enzymes require the activated a-cellobiosyl fluoride as substrate. 4.1.1 a-Cellobiosyl Fluoride Cellobiosyl fluoride was synthesized using the commercially available 1,2,2',3,3',4',6,6'-per-0-acetylated cellobiose (1) as starting material. First, a-cellobiosyl a) b) F Figure 4-1 Illustration of (a) a-cellobiosyl fluoride and (b) p-cellobiosyl fluoride 60 fluoride per-O-acetate (2) was synthesized according to a published procedure using hydrogen fluoride (HF/pyridine) " 5. Due to the anomeric effect, the nucleophilic displacement of the anomeric acetate group with fluoride ion results in the preferred alpha-orientation of the fluorine atom. The 2,2',3,3',4',6,6'-per-0-acetylated a-fluoride compound (2) is then deprotected using a catalytic amount of sodium methoxide providing a good yield of the end product, a-cellobiosyl fluoride (3) (Scheme 4-1). OH OH (3) Scheme 4-1 Outline of the chemical synthesis of a-cellobiosyl fluoride. 4.1.2 p-Cellobiosyl Fluoride The synthesis of P-cellobiosyl fluoride is similar to that of the a-equivalent. However, due to the anomeric effect, the direct displacement using fluoride will result in the 'wrong' anomer. Hence, the fluorination of (1) is carried out through the displacement of the anomeric acetate with a bromide ion to give an a-cellobiosyl bromide (4) and then the a-bromide is displaced with a fluoride source (silver fluoride, AgF) 1 1 6 , 1 1 7 resulting in the per-61 O-acetylated P-cellobiosyl fluoride (5). As for the case of a-cellobiosyl fluoride, compound (5) is deprotected using a catalytic amount of NaOMe/MeOH, yielding P-cellobiosyl fluoride (6) (Scheme 4-2). 4.1.3 Stabil i ty While cellobiosyl fluorides are reasonably stable, for prolonged storage purposes they should be kept in the per-O-acetylated form at low temperature and deacetylated (with NaOMe/MeOH) prior to use. Compared to their P-equivalents, the a-anomers are much more stable. In aqueous solution, a-cellobiosyl fluorides can be kept intact for periods of weeks or longer if kept below 0°C and buffered to minimize auto-decomposition arising from acid-catalyzed solvolysis. For P-cellobiosyl fluorides, solutions should be made up immediately prior to use. 62 4.2 Synthes is of F luor ine-Label led Ce l lob iosy l F luor ides In order to study the structure of the cellulose polymer further, the syntheses of fluorine-labelled cellobiosyl fluorides were carried out. Due to the abundance of carbon and hydrogen atoms in cellulose polymers, signals of the carbons and hydrogens will be largely overlapped; hence detailed structural information on the cellulose polymer cannot be easily obtained by using only traditional carbon and proton solution NMR. Therefore, one of the hydroxyl groups on the cellobiosyl fluoride starting material was replaced with a fluorine atom in order to produce fluorine labelled cellulose polymer (Figure 4-2). It was anticipated that the resulting cellulose polymer structure could then be studied in detail using 19F NMR. Figure 4-2 Fluorine labeled cellulose polymer. In this thesis the hydroxyl group at the C-6' position was replaced by a fluorine atom by direct fluorination followed by the fluorination of the anomeric C-l position to yield both the 6-deoxy-6'-fluoro a-and P-cellobiosyl fluorides. All supporting characterization data can be found in Chapter 5-Materials and Methods section. 63 4.2.1 F luor ine Labe l l ing As mentioned before, in order to perform 1 9F NMR studies, a portion of the cellobiosyl fluorides were synthesized in which an additional fluorine atom had been imported at a non-anomeric position. The fluorine labelling was achieved by direct fluorination, replacing the 6'-hydroxyl group on the sugar with a fluorine substituent. Out of the many hydroxyl groups on the cellobiose molecules, the 6'-hydroxyl group was chosen for several reasons: Firstly, in terms of accessibility, this hydroxyl can be conveniently isolated without the need for extensive protection schemes. In addition, the 6'-position is a primary position with two attached hydrogen atoms; hence there are no concerns regarding altering the stereochemistry of the compound during reaction at this centre (Figure 4-3). From the standpoint of fluorine introduction through displacement reactions, the 6'-position is also ideal because this position offers less steric hindrance compared to many of the other possible positions. OH OAc Figure 4-3 General approach for direct fluorination. Oh OAc (10) MeO-<\ />—HOMe OH ] O DMF, TsOH MeO' OH OH 2. Ac 2 0, Pyridine OAc (g) DAST ^ A c (Diethylaminosulfur trifluoride) O MeOH -40°C to RT. OAc OAc OAc OAc OAc OAc H 2 , PdOH/C EtOH, MeOH Glacial Acetic Acid OAc OAc (9) OAc Scheme 4-3 The outline of selective protection and DAST fluorine labeling. 64 Protection of the Hydroxyl Groups In order to isolate the 6'-OH, the p-methoxybenzylidene protecting group was employed to selectively protect the 4'- and 6'- hydroxyls of the cellobiose (7). The remaining hydroxyl groups were then protected as acetate esters as shown in Scheme 4-3. Reactions at the 6'-Position Starting from (8), attempts were made to remove the p-methoxybenzylidene protecting group by hydrogenolysis using palladium catalyst on carbon (Pd/C). However after two days at room temperature, no observable product was dectected by thin layer chromatography (TLC). Therefore, a more active catalyst, palladium hydroxide on carbon (PdOH/C) was used, yielding the desired product, 1,2,2',3,3',6-hexa-O-acetyl-cellobiose (9) (Scheme 4-3). With the 6'-OH now isolated (9), the next step was to carry out fluorine substitution at the primary alcohol. A common fluorinating method involves the use of diethylaminosulfur trifluoride (DAST) 1 1 8. DAST is both an activating agent and a fluoride source. The 6'-position of (8) was fluorinated by treating (8) with 1.1 equivalent of DAST at -40°C and allowing the reaction mixture to warm to room temperature as the reaction proceed. After three hours of reaction, the 6'-position was selectively fluorinated. Since the 6'-position is a primary site whereas the 4'-position is secondary, the initial adduct forms preferentially at the primary centre. Furthermore, fluoride sources generally favour attacking the 6'- over the 4'-position (Scheme 4-3). 65 4.2.2 6'-Deoxy-6'-Fluoro-a-Cellobiosyl Fluoride N a O M e / M e O H M e O H Scheme 4-4 Outline of the chemical synthesis of the 6'-deoxy-6'-fluoro-a-cellobiosyl fluoride. The approach for the displacement of the anomeric acetate with fluorine in the l,2,2',3,3',6'-hexa-0-acetate cellobiose (10) was similar to that of the per-O-acetylated cellobiose described above. Starting from (10), fluorination of the anomeric centre was carried out using HF/pyridine, yielding (11). Deprotection was then performed using NaOMe/MeOH, yielding the final product (12) (Scheme 4-4). 4.2.3 6'-Deoxy-6'-Fluoro-p-Cellobiosyl Fluoride The displacement reaction of the anomeric acetate group to give the P-fluoride (10) was similar to that performed on the per-O-acetylated cellobiose. However, because of the reactive reagents involved, it was necessary to protect the free 4'-OH group before proceeding. Hence (10) was first acetylated with pyridine and acetic anhydride yielding (13). Then the per-0-acetyl-6'-deoxy-6'-fluoro-P-cellobiosyl fluoride was obtained by treating (13) with hydrobromic acid (HBr) in acetic acid followed by silver fluoride (AgF) yielding (15). Lastly, the acetate protecting groups were removed using NaOMe yielding the final product 6'-deoxy-6'-fluoro-P-cellobiosyl fluoride (16) (Scheme 4-5). 66 MeONa/MeOH MeOH (16) Scheme 4-5 Outline of the chemical synthesis of 6'-deoxy-6'-fluoro-p-cellobiosyl fluoride. 67 4.3 Enzymat ic Synthes is 4.3.1 Enzymatic Synthesis of Cellulose using Wild Type Enzymes As mentioned previously, many wild-type cellulases are retaining glycosidases, able to form glycosidic bonds repeatedly when the proper conditions are provided. The polycondensation of P-D-cellobiosyl fluoride (P-CBF) using cellulase was the first successful in vitro synthesis of artificial cellulose \ When P-cellobiosyl fluoride was used as substrate it was found that the polymerization reaction proceeded smoothly to form synthetic cellulose via a polycondensation that liberates one molecule of HF per condensation (Scheme 4-6). Subsequently, the polycondensation product can be characterized by solid state NMR and mass spectrometry to confirm its molecular structure and determine its polymorphic form. However, in most aqueous environments, the newly formed polysaccharides are cleaved by the inherent hydrolase action of the cellulase, resulting in a low yield of polycondensation products. In natural cellulases, the polymerization process generally involves two carboxylic acid residues; one acting as a catalytic acid/base, the other acting as a nucleophile. Normally, the two carboxylic acid residues in the active site assist the cleavage of the glycoside bond with one protonating the glycosidic oxygen atom and the other acting as a general base forming a reactive glycosyl enzyme intermediate on the nucleophilic carboxylic acid. When an activated substrate such as P-CBF is used to form the intermediate, the 4-hydroxyl group of another P-CBF, or the growing chain end, then attacks the CI carbon of the intermediate. This gives a cellobiose-unit-elongated product, which can be further elongated to produce a polymer of synthetic cellulose. 68 a) Acid/Base Catalyst b) Acid/Base Catalyst i d) Acid/Base Catalyst Cellulose Polymer OH OH OH OH Nucleophile Scheme 4-6 Enzymatic synthesis of cellulose polymer by cellulase using p-cellobiosyl fluoride. 69 4.4 Enzymatic Synthesis using p-Cellobiosyl Fluoride In the study of enzymatic synthesis of cellulose utilizing wild-type enzymes, Cellulomonas fimi was purified by Dave Poon in the Withers lab and a variety of commercially available cellulases were used, including: cellulase from Trichoderma viride "Onozuka R-10", cellulase from Trichoderma viride, cellulase from Aspergillus niger and cellulase from Trichoderma reesei. For simplicity's sake, abbreviations of the above mentioned cellulases will be used instead of their full names in the rest of the thesis (Table 4-1). In normal aqueous buffer systems, using a-cellobiosyl fluoride as starting material, the polymerizations of cellulose by the above mentioned cellulases all produce crystalline cellulose. As will be seen later, the cellulose polymers obtained in this thesis using enzymatic systems regardless of the enzymes, substrates or buffer systems have all been identified as perfect crystalline cellulose II using solid-state 1 3C CPA/IAS NMR. Enzyme Name Source , Abbreviation Cellulomonas fimi Withers lab (Dave Poon) Cex Trichoderma viride "Onozuka R-10" Serva Ono R-10 Trichoderma viride Sigma Aldrich T. viride Trichoderma reesei Sigma Aldrich T. reesei Aspergillus niger Sigma Aldrich A. niger Table 4-1 Names, sources and abbreviations of wild-type cellulose polymerizing enzymes used in this thesis. Abbreviations instead of full names will be used for the rest of the thesis. There has been a report in the literature by Japanese researchers Kobayashi and coworkers that the production of the cellulose I polymorph is possible using commercial enzymes in a mixture of organic solvent with aqueous buffers 1 3 . Kobayashi and coworkers investigated the enzymatic formation of cellulose by using transmission electron micrographs (TEM) and electron diffraction (ED). According to their report, the in-vitro synthesis of cellulose I was observed in 2:1 acetonitrile/water buffer system whereas in the 5:1 and 7:1 acetonitrile/water buffer systems, cellulose II was obtained. In their TEM experiments, 70 cellulose I and II were differentiated by observing the microfibril formations of cellulose, where elongated microfibrils were taken to be indicative of cellulose I and shorter irregular rodlet microfibrils indicative of cellulose II. Electron diffraction experiments were also used to characterize the cellulose formation. In this case, the presence of a 6.0-A reflection was considered to be evidence for the cellulose I polymorph. As mentioned in previous chapters, the synthesis of cellulose I in nature is believed to require an organized array of catalytic subunits to assemble parallel glucan chains uni-directionally 1 , 9 . It was postulated by Kobayashi and coworkers that the addition of acetonitrile to the buffer system created a situation where microscopic phase separation could occur, creating micellar formations which simulated, in some way, the organized structure of the native cellulose I-synthesizing terminal complexes 2. Due to the surprising nature of the above mentioned claim and its potential importance, a decision was made to attempt to reproduce the above work as part of this thesis research to either confirm or refute the "in-vitro" formation of cellulose I. Since the cellulose polymers are relatively short chains and it is known that changing recrystallization solvents can produce different polymorphic forms of organic molecules, the report was considered seriously. In this thesis, enzymatic syntheses of celluloses were carried out using various commercially available wild type enzymes in various ratios of acetonitrile/water buffer solutions. The insoluble cellulose precipitates were then analyzed using 1 3C solid state CP/MAS NMR. For the sake of comparison, the 1 3C solid state CP/MAS NMR spectrum of cellulose I produced by bacteria in the Withers lab was used as a cellulose I reference (Figure 4-4) and cellulose II spectrum obtained by Atalla et al120 was used as cellulose II reference (Figure 4-5). More recent work23 has confirmed the assignment of the cellulose II resonances in detail as shown in Figure 4-6 and Figure 4-7. 71 C*e ! l u l o . s e I f r o m E J a e t e r i u l S o u r c e C 2,3,5 a 1 x:xs> i x o a o s x o o s>s s » o s o v s 7 0 6 3 S O S S S O I S Figure 4-4 3 C solid state CP/MAS NMR spectrum of bacterial cellulose I. (400 MHz) Obtained at a spinning rate of 8.5 kHz, with a contact time of 1 sec, repeat time of 1 msec and 26001 scans. Data were transformed with 0 Hz linebroadening. The peaks Ci to C 6 are labeled, with C 4 & C6 clearly consisting of several peaks. The peaks marked 'a' are amorphous peaks for Q and C 4 . The amorphous peaks for other carbons are presumably hidden by the crystalline peaks. Figure 4-5 , 3 C solid state CP/MAS NMR spectrum of cellulose II. (200 MHz) (where no amorphous peaks for C 4 and C 6 are observed) Obtained from ref. 1 2 0 . 72 C2,3.5 1 4 325 6 110 100 90 80 70 60 ppm 1 3 C Chemical shift Figure 4-6 ' 3 C solid state CP/MAS NMR spectra of the mercerized cellulose II. Obtained from Valonia (A) and l3C-enriched bacterial cellulose (B). 1 3 C line spectra of the two magnetically nonequivalent anhydroglucose residues composing cellulose II (C). The lines indicate the 1 C chemical shifts of each carbon of the anhydroglucose residues. The line spectra were determined by the INADEQUATE spectrum of the ,3C-enriched bacterial cellulose. Adapted from ref2 3. 5.0-i 5.5-} 110 100 90 80 70 ppm 1 3 C Chemical shift Figure 4-7 2D MAS-J-HMQC spectrum of the mercerized cellulose II obtained from 1 C-enriched bacterial cellulose. The solid and dotted lines indicate the 'H and I 3 C chemical shifts of the 1 3 C - ' H through-bond correlation of the cellulose II. Adapted from ref.2 3. 73 4.4.1 Common Cellulose II Polymorph Misconceptions In the search for good reference spectra for cellulose I and cellulose II it was observed that while the literature often categorized materials such as filter papers as examples of cellulose I and rayon and cellophane as examples of cellulose II, a lot of these materials actually exist in their amorphous forms. When solid state NMR was performed on store bought filter paper and rayon synthesized for this thesis using the cuprammonium process121'122, it was observed that, instead of the expected high quality cellulose I and II spectrum, NMR spectra such as that shown in Figure 4-8 and Figure 4-9 were obtained. Both specta displayed broad signals that lack the fine features necessary to properly identify the material's polymorph, indicating that, even though the rayon and filter paper used in the experiments may in fact be cellulose I and II respectively, they exist in a relatively non-crystalline, more amorphous form. Rayon from Cuprammonium process i i i i i i i (PPm) Figure 4-8 1 3 C solid state CP/MAS NMR spectrum of rayon obtained from the cuprammonium process of filter paper. Obtained at 8.5 kHz, with a contact time of 1 msec, repeat time of 5 sec and 26001 scans. Data were transformed with 20 Hz of linebroadening. 74 F i l t e r P a p e r r • • — — f • • • 1 I • • • • - — • i " - " 1 ' 1 • • — - - « — — . . . . r . ^ - . . . . . • ^ ^ ^ . ^ ^ ^ r ^ ~ , ~ , . ^ ^ ~ ^ . ^ r . b o a . 0 . 0 a _ o o s> o s o - ? o s o s o «a Figure 4-9 1 3 C solid state CP/MAS NMR spectrum of filter paper. Obtained at 8.5 kHz with a contact time of 1 msec, repeat time of 5 sec and 26001 scans. Data were transformed with 50 Hz of linebroadening. 75 4.4.2 Cellulose Production in Different Acetonitrile/Water Buffers Amongst many commercially available enzymes capable of producing cellulose with P-CBF, one of the most commonly used in similar studies is the cellulase Ono R-10. In this thesis various cellulose-producing experiments were also carried out using this particular enzyme. Ono R-10 was used to examine the validity of producing different polymorphs of cellulose by manipulating the acetonitrile to water ratio in the buffer systems. In general, in vitro cellulose productions in this thesis were performed by first dissolving the substrates (e.g. a/p-CBF) in the appropriate acetonitrile/water buffer mixtures followed by the cellulase enzyme addition. After the completion of polymerization reaction, solid participates were collected by centrifuging, washing and drying of the insoluble products. When Ono R-10 was used in the polymerization of p-cellobiosyl fluoride in a 7:1 acetonitrile/water buffer, a white precipitate was observed whose 1 3C solid state CP/MAS NMR spectrum is shown in Figure 4-10. The spectrum displayed a CI doublet at 106.5 ppm, a C4 doublet at 88 ppm, a cluster of C2, C3 and C5 between 79 - 70 ppm and the C6 doublet at 62.5 ppm. Therefore, not only did the chemical shifts indicate that the resulting cellulose precipitates were indeed of the cellulose II polymorph, but the "doublet" splitting pattern also confirmed this conclusion (Table 4-2). Cellulose polymorph CI (ppm) C4 (ppm) C2,3,5 (ppm) C6 (ppm) I 104-106 88-92 70-78 65-68 II 105-107 87-89 70-80 62-63 II (7:1 CH 3 CN/H 2 0 Ono R-10) 106.5 88 70-79 62.5 Table 4-2 Comparison of the l 3 C chemical shifts of insoluble cellulose precipitates produced by Ono R-10 in 7:1 acetonitrile/water buffer with the diagnostic l 3 C chemical shifts of the cellulose II and I polymorphs. 76 ^Cellulose f r o m O n o z u k a R - 1 O in 7; 1 i-icettoriittrilc/t>tifFesr C 2 , 3 , 5 I . ^ - ^ — ^ — — » ' i • • - • i • • • • • i — ™ - ~ - r ^ . . . , . . - ^ — ^ , . . , . . . . . . . _ r , - r . . . . . . -I I S U O J . O S l O O , 9 5 S O S S S O T S T O 6 5 6 0 5 5 S O 4 5 4 p p m + Figure 4-10 1 3 C solid state CP/MAS NMR spectrum of cellulose from cellulase Ono R-10 in 7:1 acetonitrile / water buffer obtained at 8.5 kHz with a contact time of 1 msec, repeat time of 2.0 sec and 2000 scans. Data were transformed with 10.00 Hz of linebroadening. Ceniil0se':fir6m Ono'zuKa R - 1 0 ,ihC5:;l • acetoniiri:ie/buff er ^2,3,5 I Figure 4-11 1 3 C solid state CP/MAS NMR spectrum of cellulose from Ono R-10 in 5:1 acetonitrile / water buffer obtained at 8.5 kHz with a contact time of 1 msec, repeat time of 5 sec and 30,000 scans. Data were transformed with 10.00 Hz of linebroadening. 77 When similar polymerizing reactions are carried out in 5:1 and 2:1 acetonitrile/water buffer using cellulase Ono R-10 and a-cellobiosyl fluoride, cellulose polymer precipitates were also obtained which yielded similar 1 3C solid state CP/MAS NMR spectra. Both the 5:1 and 2:1 spectra contained the characteristic cellulose II doublet splitting pattern of CI, C4 and C6 signals (Figure 4-11 & Figure 4-13). The chemical shifts of both spectra were also typical of that of cellulose II samples (Table 4-3). Therefore, all cellulose polymer precipitates obtained were identified as cellulose II. Cellulose polymorph CI (ppm) C4 (ppm) C2,3,5 (ppm) C6 (ppm) I 104-106 88-92 70-78 65-68 II 105-107 87-89 70-80 62-63 II (5:1 CH 3 CN/H 2 0 Ono R-10) 103 85 68-78 60 II (2:1 CH 3 CN/H 2 0 Ono R-10) 106 88 68-78 62 Table 4-3 Comparison of the l 3 C chemical shifts of insoluble cellulose precipitates produced by Ono R-10 in 5:1 and 2:1 acetonitrile/water buffer with the diagnostic 1 3 C chemical shifts of cellulose II and I polymorphs. Besides confirming that the cellulose II polymorph was produced in all the different acetonitrile/water ratio buffers, it was also observed that the yield of insoluble cellulose precipitate generally increased as the ratio of acetonitrile to aqueous buffer increased (with minimal yield in pure aqueous systems) using Ono R-10. Similar trends were also observed when using other cellulases. Because the Japanese report indicated that the production of cellulose I can be achieved in 2:1 acetonitrile/water buffers2, subsequent experiments placed more emphasis on the study of cellulose production in 2:1 buffer systems. A series of experiments using different commercially available cellulase enzymes were carried out in 2:1 acetonitrile/water buffer using T. viride, T. reesei and A. niger, and all three enzymes produced insoluble cellulose precipitates in the 2:1 acetonitrile/water buffer system. Cellulase T. reesei gave the highest yield of insoluble cellulose products, followed by A. niger and then T. viride. In terms of morphology, T. viride and T. reesei both produced cellulose samples with cellulose 78 II13C solid state CP/MAS NMR spectra with characteristic chemical shift and splitting patterns (Table 4-4, Figure 4-13 & Figure 4-15). In order to further characterize the insoluble solids obtained in each enzymatic polymerization, MALDI-TOF Mass Spectra were also obtained for selected samples. As can be seen in Figure 4-14 & Figure 4-16, both spectra display the characteristic mass separation (A162 g/mol) of typical cellulose polymers. A series of peaks can be observed indicating the presence of cellulose polymers of different chain lengths. The spectra also show degrees of polymerization (DP)„ of at least 18 or 19*. However, a word of caution is that MALDI-TOF MS is a qualitative but not a quantitative method, meaning the actual degree of polymerization cannot be determined solely by MALDI-TOF MS. The inability to observe peak(s) of higher degree(s) of polymerization in a MALDI-TOF MS spectrum does not indicate the lack of polymers of higher degree(s) of polymerization. Hence, all of the MALDI-TOF MS spectra obtained can only provide supplemental information regarding the presence of cellulose polymer chains, the mass differences between various cellulose polymer chains and the minimum "observable" (DP)n. Cellulose polymorph CI (ppm) C4 (ppm) C2,3,5 (ppm) C6 (ppm) I 104-106 88-92 70-78 65-68 II 105-107 87-89 70-80 62-63 II (2:1 CH 3 CN/H 2 0 T. viride) 106.5 88.5 68-79 62.5 II (2:1 CH 3 CN/H 2 0 T. reesei) 106.5 88 68-79 62 Table 4-4 Comparison of the l 3 C chemical shifts of insoluble cellulose precipitates produced by T. viride and T. reesei in 2:1 acetonitrile/water buffer with the diagnostic 1 3 C chemical shifts of cellulose II and I allomorphs. * "Degree of polymerization (DP)n" in this thesis is defined as the mean individual values which is not the same as the average degree of polymerization. 79 Cellulose from Onozuka R - 1 0 in 2:1 acetonitrile/buffer '2,3,5 I IA 120 1:1.0: 100 90 go 70 60. SO Figure 4-12 1 3 C solid state CP/MAS NMR spectrum of cellulose from Ono R-10 in 2:1 acetonitrile / water buffer. Obtained at 8.5 kHz with a contact time of 1 msec, repeat time of 5 sec and 30,000 scans. Data were transformed with 10.00 Hz of linebroadening. C e l l u l o s e f r o m T. viirlcle i n 2 -. X a c e o t o n i t r i i e / b u f f e r ft A c 4 1A \ C 2,3,5 * 0 - 8 & 3:<S. Figure 4-13 1 3 C solid state CP/MAS NMR spectrum of cellulose from T. viride in 2:1 acetonitrile / water buffer. Obtained at 8.5 kHz with a contact time of 1 msec, repeat time of 5 sec and 10,000 scans. Data were transformed with 1.00 Hz of linebroadening. 80 C t - l I u l o s e P o l y m e r p r o d u c e d I>> " I ' r i c h o d e r m a - v i r i d e A162 g/mol j DP(n) 19 A t\ ul z o o o 3 0 0 0 I ' l / -Figure 4-14 MALDI-TOF MS spectrum of cellulose from T. viride. produced in 2:1 acetonitrile / water buffer with mass separation of A162 g/mol between peaks. Maximum observable degree of polymerization (DP)„ of 19. G 2 , 3 , 5 •! *1 •;C'.1v •S •i! :H A Figure 4-15 l 3 C solid state CP/MAS NMR spectrum of cellulose from T. reesei in 2:1 acetonitrile / water buffer obtained at 8.5 kHz with contact time of 1 msec, repeat time of 5 sec and 30,000 scans. Data were transformed with 1.00 Hz of linebroadening. 81 <.t o o *«-ooo H E* £*> o o H s'>,*=soo H 3?* o o o C Q I I U I O S G polymer from T. rr?&&&i in^:1 aaetonitriio/fcauffor A162 g/mol -I o o o S I O O O DP ( n ) 17 • r "5 O O O Figure 4-16 MALDI-TOF MS spectrum of cellulose from T. reesei in 2:1 acetonitrile / water buffer with mass separation of A162 g/mol between peaks. Maximum observable degree of polymerization (DP) of 17. 82 Cex-Cellulomonas fimi Another enzyme Cellulomonas fimi (Cex) was prepared in the Withers lab by David Poon. The enzyme was prepared in two forms, one having all domains and the other having only the catalytic domain, lacking the CBD and the linker domain. Interestingly, both enzymes show high polymerization activity for monomer (3-CF, producing synthetic cellulose. As time progresses, the cellulose produced gradually disappears when the intact full-length enzyme is used because of hydrolysis; however the product cellulose hardly changes when catalytic domain is employed. These results suggest that the CBD plays an important role in the hydrolysis of the product but not in the polymerization of P-CF while The polymerization needs only the catalytic domain. Polymeric celluloses produced using Cex were also investigated to identify the structure of the cellulose produced. Insoluble cellulose was produced from a 2:1 acetonitrile/aqueous buffer mixture. Both P-CBF and 6'F P-CBF produced solid precipitate using 2:1 acetonitrile/aqueous buffer mixture. Both the solid precipitates obtained were identified using MALDI-TOF MS as cellulose polymer with mass separation of A324 g/mol for cellulose polymer produced from P-CBF (Figure 4-17) and A328 g/mol for cellulose polymer produced from 6'F P-CBF (Figure 4-18). It was also observed in the MALDI-TOF MS spectrum of cellulose produced by Cex using P-CBF (Figure 4-17) that other than the major species of polymer with the A324 g/mol mass separation, a minor polymer species with A162 g/mol was also observed due to the hydrolysis and re-extension of the major polymer species. 83 Cellulose polymer produced by Cex using P-CBF A324 g/mol (Major species) A162 gfmol (M inor species) DP(n) 21 U i l i i J i i i L u k ,i Figure 4-17 MALDI-TOF MS spectrum of cellulose from Cex in 2:1 acetonitrile / water buffer using P-CBF with mass separation of A324 g/mol between peaks. Maximum observable degree of polymerization (DP) > 21. Reflector :'1.000 II 6TS 90 Cellulose polymer producd by Cex using 6'F-P-CBF A328 g/mol DP(n) 16 • Figure 4-18 MALDI-TOF MS spectrum of cellulose from Cex in 2:1 acetonitrile / water buffer using 6'F P-CBF with mass separation of A328 g/mol between peaks. Maximum observable degree of polymerization (DP) of 16. 84 Enzymatic synthesis of cellulose using mutant cellulases In the enzymatic synthesis of cellulose using mutant glycosidases, a-glycosyl fluorides are used as starting materials instead of the usual p-glycosyl fluorides. Even though the a-glycosyl fluorides are of the 'wrong' anomeric form and cannot be accommodated in the active site of the enzyme in normal circumstances, since the catalytic nucleophile has been replaced by a much smaller non-nucleophilic residue such as Ala or Gly through site-directed mutagenesis, the mutated P-glycosidases now have enough free space in the active centre to accept a-glycosyl fluorides. In the active centre, the a-glycosyl fluoride will mimic and occupy the same position as the a-glycosyl enzyme intermediate (Scheme 4-7). The anomeric C1 carbon of the a-glycosyl fluoride can then be directly attacked by the 4-hydroxy group of another a-CBF or end of the growing chain attacks the C1 carbon from the p-side, resulting in glycosylation with the inversion of anomeric configuration (Scheme 4-7). In this thesis, a-cellobiosyl fluoride was used as substrate for the synthesis of cellulose polymer using mutant cellulases according to the following scheme (Scheme 4-7). Because the mutant enzyme is used, the reactions are not reversible. A c i d / B a s e C a t a l y s t O H 0 H(0 O H O H M u t a t e d n o n - n u c l c o p h i h c r e s i d u e E 1 9 7 G A c i d / B a s e C a l a l y s t — O H — - O H j O H — O H M u t a t e d n o n - n u c l e o p h i l i c res idue E197G t A c i d / B a s e C a l a l y s t Celulose Polymer — O H , — O H — O H . - - O H M u t a t e d n o n - n u c l e o p h i l i c res idue E 1 9 7 G Scheme 4-7 Enzymatic synthesis of cellulose polymer using a-cellobiosyl fluoride in mutant enzyme Endoglucanase I, E197G. 85 Endoglucanase The mutant enzyme used for cellulose synthesis belongs to the endoglucanase family. Most endoglucanases are cellulases that hydrolyze interior bonds within cellulose contain 3 domains: (1) a cellulose binding domain (CBD), (2) a catalytic domain and (3) a linker domain 1 2 3. In the hydrolysis process, the CBD would first bind to the crystalline part of cellulose, and then the catalytic domain catalyzes the hydrolysis. Recently, mutants of endoglucanase I from Fusarium oxysporum have been generated in the Withers lab by Dr. Hamzah Mohd-Salleh. These are nucleophile mutants where the catalytic nucleophile is replaced by a small non-nucleophilic residue. Hence these mutants are only capable of synthesizing cellulose but lack the ability to hydrolyse it. Three types of mutant enzymes were employed in the study, the glycine mutant E197G, the alanine mutant E197A and the serine mutant E197S. In initial studies using E197G mutants with only buffered aqueous solution, white precipitates of cellulose were observed after one day of reaction time. Then 2:1 and 5:1 acetonitrile/aqueous buffer systems were used with the E197G mutants, producing similar cellulose precipitates. 1 3C CP/MAS NMR analysis of the solid produced by the E197G mutant in pure aqueous solvent was shown to be cellulose II (Figure 4-19) from its characteristic chemical shifts and doublet splitting patterns. Cellulose polymorph CI (ppm) C4 (ppm) C2,3,5 (ppm) C6 (ppm) I 104-106 88-92 70-78 65-68 II 105 - 107 87-89 70-80 62-63 II (H20 endoglucanase I F. Oxysporum E197G) 106.5 88.5 68-78 62.5 Table 4-5 Comparison of the l 3 C chemical shifts of insoluble cellulose precipitates produced by 71 viride and T. reesei in 2:1 acetonitrile/water buffer with diagnostic l 3 C chemical shifts of cellulose II and I polymorphs. 86 E197G mutants also successfully produced cellulose polymer in 7:1, 5:1 and 2:1 ratios of acetonitrile/aqueous buffer systems. The 13C{'H} CP/MAS NMR spectrum clearly identified the cellulose polymer produced using E197G in aqueous buffer as cellulose II (Figure 4-19). The presence of cellulose polymer in the insoluble precipitate produced in acetonitrile/aqueous buffer system can be confirmed by MALDI-TOF MS spectra showing a series of peaks with a cellulose polymer mass separation of A324 g/mol. However due to low yield and limited amount of the E197G mutant available, only MALDI-TOF MS spectrum was obtained for the cellulose produced using E197G in 5:1 (Figure 4-20) and 2:1 (Figure 4-21) acetonitrile/aqueous buffers, further NMR studies of these cellulose products were not possible. Attempts have been made to produce more of the same mutant, unfortunately without success. As mentioned before, most of the enzymatic cellulose polymerization studies focused on the production of cellulose in acetonitrile/water buffer systems, therefore further studies with acetonitrile/water buffer system were performed on the E197A and E197S mutants. Polymerizations were carried out in 2:1 acetonitrile/water buffer using both El97A and E197S mutants and both produced insoluble crystalline products. However, both strains produced low yields of crystalline product with El 97A giving a slightly better yield than E197S. MALDI-TOF MS studies confirmed the presence of cellulose polymers with the expected A324 g/mol mass differences between adjacent peaks. More importantly, the insoluble celluloses collected from El97A are clearly identified as cellulose II using 1 3C solid state CP/MAS NMR (Figure 4-22). 87 Figure 4-19 1 3 C solid state CP/MAS NMR spectrum of cellulose produced from glycine mutant E197G of Endoglucanase I fromF. oxysporum in aqueous buffer obtained at 8.5 kHz with a contact time of 1.0 msec, repeat time of 2.0 sec and 37,340 scans. Data were transformed with 10.00 Hz of linebroadening. -1 o o o s o o o —I C e l l u l o s e p o l y m e r p r o d u c e d b y E n d o g l u c a n s e m u t a n t 5 : 1 a c e t o n i t r i l e / w a t e r SI SI V V V A3 2 4 g / m o l 3 0 0 0 n i / . . Figure 4-20 MALDI-TOF MS spectrum of cellulose from Endoglucanase IF. Oxysporum in 5:1 acetonitrile / water with mass separation of A324 g/mol between peaks. 88 C e l l u l o s e p o l y m e r p r o d u c e d t>y E n d o g l u c a n a s e m u t a n t E 1 9 7 G 2 :1 a c e t o n i t r i l e / w a t e r s o o —j i 3 7 . y O 1 A- - O 1 6 6 1 - 7 U L 324 g/mol 1 9 8 5 . 7 2 3 O 9 . 6 2 6 3 5 - 5 ^1 «<_><_> Figure 4-21 MALDI-TOF MS spectrum of cellulose from Endoglucanase I F. Oxysporum in 2:1 acetonitrile / water with mass separation of A324 g/mol between peaks. C e l l u l o s e p r o d u c e d b y e n d o g l u c a n a s e F. oxysporum m u t a n t E 1 9 7 A in 2:1 a c e t o n i t r i l e / a q e o u s buffer ft. ^ o a. a~ o (ppm) Figure 4-22 1 3 C solid state CP/MAS NMR spectrum of cellulose II from endoglucanase F. oxysporum mutant E197A in 2:1 acetonitrile / water buffer obtained at 8.5 kHz with contact time of 1 msec, repeat time of 5 sec and 32768 scans. Data were transformed with 50.00 Hz of linebroadening. 89 Modified Cellulose Production using 6'-Deoxy-6'-Fluoro-Cellobiosyl Fluoride Substrates Generally 6' fluoro cellobiosyl fluoride substrates are less well tolerated than regular a or (3 - cellobiosyl fluoride substrates resulting in low yields in our cellulases and glycosynthases enzymatic reactions. Successful cellulose synthesis was obtained with T. viride using P-cellobiosyl fluoride, obtaining close to 8 mg of final products. Normally, in order to obtain a reasonable solid-state NMR spectrum with a 4 mm diameter spinner, a minimum of 50 mg of solid is needed. However, constrained by the small sample size, a custom-made insert was made with boron nitride to contain the sample to the center of the spinner where the signal is most easily obtained. After two nights of experiment or 40960 scans, a spectrum was obtained showing a typical cellulose type pattern. However, due to the small sample size, there is substantial noise and it is necessary to use substantial line broadening precluding the identification of the exact chemical shifts and multiplet patterns of the resonances, and the characterization of which polymorph is present. However it was proven that with the suitable enzymes under suitable conditions, fluorinated cellulose polymers could be produced using 6'F cellobiosyl fluorides. Also, with adequate sample sizes, further studies regarding the cellulose structure could be carried out using fluorinated cellulose polymers. 90 BC ! II i j CP MAS of Cellulose produced by T, wide using 6T {J-CBF A 200 150 (ppm) * m SO Figure 4-23 l 3 C solid state CP/MAS NMR spectrum of solid material produced from T. viride using 6'F p-CFB in 2:1 acetonitrile / water buffer obtained at 8.5 kHz with contact time of 1 msec, repeat time of 4 sec and 40960 scans. Data were transformed with 50.00 Hz of linebroadening. 91 4.5 Solubil i ty Factor As mentioned in previous sections, the yields of insoluble cellulose products generally increase as the ratio of acetonitrile/water in buffer increases. There were concerns that the increase in crystalline products observed may simply be caused by the insolubility of the cellulose polymer in an organic solvent (acetonitrile), much like inducing crystallization by adding an additional solvent with low solubility index for the compound of interest. In order to eliminate the possibility that the differences were simply due to solubility effects, the following experiments were performed: In a pure aqueous buffer system, a reaction was carried out using mutant endoglucanase E197A and a-CBF. After one day of reaction, no observable insoluble products were observed. At this time a seven times excess of acetonitrile was added to the reaction mixture slowly. If reaction had proceeded and produced cellulose polymers which were not observed because of solubility, the addition of acetonitrile should have induced cloudiness and/or their precipitation. After the addition of acetonitrile, no cloudiness nor precipitates were observed, proving the lack of insoluble products in low acetonitrile/water ratio buffers was not caused by solubility factors and the low yield observed in low acetonitrile/water ratio buffers may have been due to low reaction rates of the enzymes involved. To further prove that solubility has not caused the low yield, the whole reaction mixture was freeze-dried, re-hydrated with water and freeze-dried three more times in order to completely remove the ammonium bicarbonate buffer and solvents. Then, 1 3C solid state CP/MAS NMR was performed on the freeze-dried solid to detect the presence of any cellulose polymer. The resulting spectrum showed the presence of neither cellulose I nor II, suggesting that no cellulose polymers had been produced. Similar experiments were also performed on A. niger using P-CBF as starting material with identical results. 92 4.6 Cellulose Production by Wild Type Enzyme vs Mutants One interesting feature of the various MALDI-TOF MS spectra is that the difference in mass between adjacent peaks directly reflects the type of enzyme used. In this thesis all of the starting materials used were disaccharides. Since disaccharides are used as polymer building blocks, the MALDI-TOF MS spectra of the cellulose polymers should have shown mass differences of one disaccharide unit between adjacent peaks, for in each step of the polymerization process, one additional disaccharide unit would have been added to the growing chain end. In the MALDI-TOF MS spectra of the cellulose polymers produced by the mutant enzymes, the expected results were observed. Between the series of adjacent peaks, mass differences corresponding to the weight of one disaccharide were consistently observed. However, in the MALDI-TOF MS spectra of the cellulose polymers produced by the wild type enzymes, mass differences between the series of adjacent peaks correspond to the mass of a monosaccharide. This is due the fact that wild type cellulase enzymes not only possesses the ability to form cellulose chains, they also possesses the ability to hydrolyse them. In the process of polymerization, due to the nature of our cellobiose substrate, growing polymer chains were elongated two sugar units at a time producing even units of polymerized chains. However, in the process of hydrolysis, polymer chains were hydrolyzed randomly, producing even and odd units of polymerized chains. Hence for enzymes only capable of polymerization, only even units of polymerization are possible. For enzymes capable of both polymerization and hydrolysis both even and odd numbers of polymerization can be observed. Therefore, in the MALDI-TOF MS spectra of cellulose polymer produced by wild type cellulases, mass differences between adjacent peaks corresponding to one monosaccharide can be observed. 93 Wild Type Cellulase Cellulose polymer produced by T r i c h o d e r m a rewl I : M I I I I i I I I Iffigitd Mutant Cellulase C e l l u l u . r polymer produced hy Rnilugluri in, 2 : l acetonitrile/water 324g/nd Figure 4-24 Mass separation differences between cellulose polymer chains produced by wild type enzymes and mutant enzymes. Typical mass separation between peaks of cellulose polymer using cellobiosyl fluoride as starting material is A324 g/mol (as can be seen in samples formed using the mutant enzyme). However, with the hydrolysis ability of wild type cellulases, mass separations of A162 g/mol can also be observed as the growing chain is also being hydrolyzed as the same time (as seen in samples formed using the wild type enzyme). 4.7 Further Studies of the Monomer ic Substrates In order to facilitate future structural studies by solid state NMR on the structures of the cellulose polymorphs by incorporating fluorine atoms, preliminary investigations of the monomeric fluorinated building blocks were carried out using various analytical techniques to serve as reference data. In addition to high-resolution solution state NMR studies, X-ray structure determinations of a-cellobiosyl fluoride (Figure 4-25) and 6'F-a-cellobiosyl fluoride (Figure 4-26) were also obtained and revealed the three-dimensional conformations of both cellobiosyl fluorides. (For full data of the X-ray crystal structural data of both a-CBF and 6'F-a-CBF please refer to Appendix I and Appendix II respectively.) Attempts to crystallize p-cellobiosyl fluoride and 6'F-P-cellobiosyl fluoride were without success and hence X-ray crystallography could not be performed on those compounds. 94 g-Cellobiosyl Fluoride Figure 4-25 Stereoview of the X-ray crystal structure of a-cellobiosyl fluoride. 6'F a-Cellobiosyl Fluoride Figure 4-26 Stereoview of the X-ray crystal structure of 6'F a-cellobiosyl fluoride. 95 4.7.1 Solid State NMR Studies of the a-Cellobiosyl Fluoride Substrates In the solid state NMR studies on the alpha enzyme substrates (a-CBF and 6'F a-CBF) (Figure 4-30,Figure 4-33 and Figure 4-34) and its precursor (per-acetylated a-CBF) (Figure 4-27), two different NMR probes with different size spinners, a 7 mm and a 4 mm were used. The 7 mm probe is a three-channel probe that utilizes a 7 mm spinner containing the sample with a cooling channel allowing the adjustment of temperature. It is known that friction between the spinning rotor and the bearing gas under magic angle spinning conditions can lead to heating of the sample. Therefore due to concerns of over-heating the more unstable samples, the 7 mm probe was used in some experiments involving fully deprotected monomeric substrates in order to avoid possible decomposition during the experiment. Other than the disadvantage of requiring a large sample size, the 7 mm probe also has the disadvantage of a lower spin rate. As explained in earlier chapters, in CP/MAS solid-state experiments, the sample rotation rate is crucial to remove large chemical shift anisotropics and some dipolar couplings. Generally, as the spinner size increases, the maximum possible spinning rate it can achieve decreases, resulting, in some cases, in unwanted spinning side bands. However, depending on the size of the interactions, the effect of spinning rate on the final spectrum will be different as seen in Figure 4-27 and Figure 4-29 versus Figure 4-30 and Figure 4-33. Therefore, in the case where large interactions existed, the 4 mm probe was used. The resulting l 3C -» 'H CP/MAS spectra (Figure 4-27,Figure 4-28, Figure 4-30 to Figure 4-32) all show similar chemical shifts and patterns as their respective high-resolution NMR with the exception of the spectra of a-CBF at 3.5 kHz where large spinning side bands exists. The spectrum of 6'F a-CBF also shown similar chemical shifts and patterns as its high-resolution NMR spectrum with carbon signals with fluorine coupling at chemical shifts 106.8 ppm and 74.3 ppm. 96 13CJ1 MS'CR. M A S on ot-cel1obios>il: fluoride "oclaacccatc. with .1-H • decoupling at 8.5 kHz C H 3 C=0 — ~ — — ~ ~ — ™ — ™ j — r - — — r — r > — ~ — — — - — r ; — r ~ . — r - ~ - ~ — . — ^ _ _ ^ r , _ r . _ — , - — . , — , , — . . r r - , , . . , „ , , , — „ » _ — . — _ p.- ast> 20:0' jsd: (pp,„j> -'-lop'' ' s'o.',": Figure 4-27 1 3 C solid state CP/MAS NMR spectrum of per-acetylated a-CBF obtained at 8.5 kHz with contact time of 1 msec, repeat time of 4 sec and 3934 scans. Data were transformed with 0 Hz of linebroadening. Figure 4-28 Comparison between the 1 3 C solid state CP/MAS NMR spectrum expansion (top) and the high-resolution 1 3 C solution state NMR spectrum (bottom spectrum) of per-acetylated a-CBF. 97 c = o J\i\j V-C H 3 Figure 4-29 l 3 C solid state CP/MAS NMR spectrum of cellulose of per-acetylated a-CBF obtained at 3.5 kHz with contact time of 1 msec, repeat time of 4 sec and 2048 scans. Data were transformed with 20 Hz of linebroadening and (*) indicates spinning side bands. Figure 4-30 1 3 C solid state CP/MAS NMR spectrum of a-CBF obtained at 8.5 kHz with contact time of 1 msec, repeat time of 4 sec and 20480 scans. Data were transformed with 10.00 Hz of linebroadening. 98 A * p t i # -c«vi l . o t o i o K i y J , F l u o r i d e H i<3ft K e c o l u c i o n s o l u t i o n s t a t e i:ic N M R Figure 4-31 1 3 C High resolution solution state NMR spectrum of a-CBF obtained with 3447scans. Data were transformed with 1.00 Hz of linebroadening. Figure 4-32 Comparison between the C solid state CP/MAS NMR spectrum expansion (top) and the high-resolution 1 3 C solution state NMR spectrum expansion (bottom) of a-CBF. 99 Figure 4-33 1 3 C solid state CP/MAS NMR spectrum of a-CBF obtained at 3.5 kHz with contact time of 1 msec, repeat time of 4 sec and 1024 scans. Data were transformed with 0 Hz of linebroadening. I 3 C J 1H} C P / M A S o f 6 'F a - C B F @ 2 7 5 k (pp«i> Figure 4-34 l 3 C solid state CP/MAS NMR spectrum of cellulose of 6'F-a-CBF obtained at 3.5 kHz at a temperature of 275 K with contact time of 1 msec, repeat time of 4 sec and 1024 scans. Data were transformed with 0 Hz of linebroadening. 100 F Solid State NMR Studies 19F solid-state NMR spectra were successfully obtained on both alpha-cellobiosyl fluoride (Figure 4-35) and 6'F a-cellobiosyl fluoride (Figure 4-36) using the 4 mm probe. Due to the long 1 9F relaxation time Ti of 165 seconds, a regular 90° single pulse program with delay time of five times Ti means one would have to wait 15 minutes in between successive scans. Therefore a time saving approach of using a 30° pulse was employed in order to reduce the delay time between successive scans down to 2 minutes. Even though the 30° pulse would produce a lower signal intensity, the loss in signal intensity was made up by the gain in the number of scans in the same time period. Also, due to the compound's large dipolar coupling, a high spin rate is necessary in order to produce reasonable peaks in the NMR spectrum. Even at a spinning rate of 15 kHz, minor spinning side bands can still be observed. Attempts were also made to obtain the 19F ->'H CP/MAS solid-state NMR spectra with no success. 19F ->'H CP/MAS has been performed previously with a fast spinning rate of 35 kHz performed on a highly fluorinated highly ordered rigid non-organic species of poly(vinyl fluoride) (PVF)124'. We feel that the low spinning rate, low fluorine concentration and the low degree of crystallinity of the sample all contributed to the inability to perform l 9F -VH CP/MAS on our sample. 101 19F SP cm «-t;ellobiosvAl fluoride.@ 15fcHx -149.7 ., , p- —• ~- , — ^ ^™-™r™.,™.,. . J——.- ^ . ~^  , ^ . ™ p J. . -46' -.60 ..~86.- -100. -iZ'O .-14.0. t.1.60 -ifec -200 -220 -2.4 0 ~.2'6Q. (ppm) Figure 4-35 1 9 F solid state single pulse NMR spectrum of a-CBF obtained at 15 kHz with repeat time of 300 sec and 200 scans. Data were transformed with 0 Hz of linebroadening. Signal corresponding to the a-fluoride of a-CBF was found at -149.7 ppm and (*) indicates spinning side bands. 19F SP-'6n:'6!R^^ll6hK»^;WTu'oriae;@ii:5kH?:-.143 -241 .., , . ,. —. ^ , 1 ra -r — -1 —*—7- • f ~" — - V" ~— '— —SO : - 6 Q --80', lOa; -lip -'-X4'0. ~.-160 -1,30 -.200' -22.0 -240 -260 : Figure 4-36 l 9 F solid state single pulse NMR spectrum of 6'F-a-CBF obtained at 15 kHz with repeat time of 300 sec and 200 scans. Data were transformed with 0 Hz of linebroadening. Signals corresponding to the a-fluoride and 6' fluoride of 6'F-a-CBF were found at -143 ppm and -241 ppm respectively. (*) Indicates spinning side bands. 102 Chapter 5 C o n c l u s i o n s and S u g g e s t i o n s for Further Work 5.1 S u m m a r y and C o n c l u s i o n s Mutant Endoglucanase fromF. oxysporum (E197A, E197G, E197S: prepared in the Withers lab) and selected commercial cellulase enzymes (T. viride, T. reesei, A. niger, Ono R-10, Cex) all produced solid cellulose in various ratios of acetonitrile and aqueous buffer mixtures (0:1, 2:1, 5:1, 7:1 acetonitrile/aqueous buffer). Though all of the enzymes used in this thesis produced solid cellulose when cellobiosyl fluoride was used as substrates, it was found that not all enzymes produced solid cellulose precipitates when 6'F cellobiosyl fluoride was used as a substrate. In the case where the substrate 6'F a-cellobiosyl fluoride successfully produced solid cellulose, such as mutant Endoglucanase enzyme El97A, the yield was very low in comparison with that obtained with regular a-cellobiosyl fluoride. From 1 3C solid state CP/MAS NMR studies, all cellulose synthesized from a-CBF or P-CBF starting material produced high quality cellulose II NMR spectra indicating high crystallinity of the solids obtained. Successful 13C{'H} CP/MAS solid state NMR spectra were also carried out on polymeric cellulose synthesized using 6'F a-CBF with mutant enzyme E197A. In contrast to the results of Kobayashi's group 1 - 3 who reported producing cellulose I using pure enzymes and a-CBF in 2:1 acetonitrile/aqueous buffer, cellulose I production could not be achieved using similar conditions with a variety of enzymes in any of our experiments. This result indicates that the cellulose I production observed in Kobayashi's group might be a unique situation and cannot be generalized and applied to all enzymes or the observations are in error. In a preliminary study, successful single pulse 19F solid state MAS NMR and 1 3C solid state CPMAS NMR studies were performed on the a-CBF and 6'F a-CBF. Solid state ^Fl'H} CP/MAS NMR were also attempted on cellobiosyl fluorides and a fluorinated polymeric cellulose compound successfully produced in enzymatic synthesis. However, this 103 was not successful due to hardware limitations. With the proper instrumentation, the characterization of fluorinate cellulose derivatives using 6'F a-CBF and cellulases using solid state 19F MAS in combination with 19F{'H} CP/MAS and 13C{!H} CP/MAS NMR should be possible in the future. 5.2 S u g g e s t i o n s for Further Work The enzymatic synthesis of cellulose using cellobiosyl fluoride in 2:1 acetonitrile/aqueous buffers should be extended to other cellulase enzymes to further investigate the cellulose I producing phenomenon reported by Kobayashi and group '~3. Possible 1 3C CPMAS spectra should be run on this sample itself if it is made available. Larger scale enzymatic synthesis of fluorinated polymeric cellulose should be performed using the 6'F cellobiosyl fluoride in order to better characterize its structures and polymorphs. Also high crystallinity cellobiosyl fluoride substrates and fluorinated cellulose should be prepared to increase the quality of the solid state NMR spectra obtained. Last of all, with the increase in crystallinity and quantity of fluorinated cellulose, solid state 19F{'H} CP/MAS NMR should be performed and structural investigations carried out on the cellulose II analogs produced by enzymatic polymerization. 104 Chapter 6 Materials and Methods 6.1 C h e m i c a l Syn thes is of Ce l lob iose Derivatives General methods All buffer chemicals and other reagents were obtained from Sigma or Aldrich Chemical Cos. unless otherwise noted. 'H solution NMR spectra for all products and intermediates were obtained using Bruker AC-200, AV-300 or WH-400 spectrometers. l 9F solution NMR spectra were recorded on the Bruker AC-200 or AV-300 spectrometers with C F 3 C O 2 H as reference. High-resolution liquid secondary ion mass spectrometry (HRLSIM) was performed by the staff of the mass spectrometry facility in the Department of Chemistry at U.B.C. (LCMS). Matrix-Assisted Laser Desportion/Ionization (MALDI) mass spectrometry was performed using a Bruker Biflex IV system. All single crystal x-ray diffraction studies were carried out using a Bruker X8 APEX CCD diffractometer. All thin layer chromatography was carried out using E. Merck pre-coated aluminium-backed sheets of Silica Gel 60 F254 with detection effected by heating with 10% ammonium molybdate in 2M H2SO4. Column chromatography was performed using Silica Gel 60 (230-400 mesh) from Silicycle Inc. under positive pressure. 105 6.1.1 a - C e l l o b i o s y l F luor ide (3) 2, 3, 6, 2', 3', 4', 6' Hepta-O-acetyl-a-cellobiosyl fluoride (2) a-Cellobiosyl fluoride per-O-acetate was synthesized according to a published procedure l l 5 . Cellobiose per-O-acetate (1) (5.0 g, 7.4 mmol) was placed in a 125 mL polypropylene Nalgene™ bottle in an ice bath and flushed with argon. HF/pyridine (70% hydrogen fluoride/pyridine, 5 g) was added and the mixture stirred until homogeneous. The reaction vessel was then sealed and allowed to warm up to room temperature. After one hour, the reaction was terminated by the addition of saturated sodium bicarbonate (NaHCOs) solution drop-wise until the pH of the mixture was neutral. The reaction was then diluted with dichloromethane (DCM, 100 mL) and the organic phase washed with 100 mL saturated NaHC03 followed by 100 mL water. The organic phase was then dried over magnesium sulphate (MgS04) and filtered. After the solvent was evaporated, a-cellobiosyl fluoride per-O-acetate (2) (4.6 g, 7.2 mmol, 97%) was obtained as white crystals. *H NMR data were in agreement with that reported previously " 5. Rf (3:1 EtOAc/PE) = 0.50; 'H NMR (400 MHz CDC13): 5 5.65 (dd, 1 H, JlF= 53.1 Hz, J u = 2.7 Hz, H-l), 5.44 (dd, 1 H, J3,2= 10.0 Hz, J3.4 = 9.8 Hz, H-3), 5.13 (dd, 1 H, JIA =9.5 Hz, J y x = 8.9 Hz, H-3'), 5.05 (dd, 1 H, J4j=9J Hz, J4J =9.5 Hz, H-4'), 4.91 (dd, 1 H, Jr,r = 8.9 Hz, J2;r= 8.1 Hz, H-2'), 4.82 (ddd, 1 H, J2,s = 10.0Hz,J2.F = 24.3Hz,yZ/ = 2.7Hz,H-2),4.51 (m, 1 H, H-6a), 4.50 (d, 1 W,J,;2 = 8.1 Hz, H-l'), 4.35 (dd, 1 H,J6H,5= 12.5 Hz, J6B,A = 4.5 Hz, H-6b), 4.15-4.05 (m, 2 H, H-6'ab), 4.02 (dd, J5,6b = 12.5 Hz, J5.6A = 2.0 Hz, H-5), 3.80 (dd, 1 H, J4J = 9.8 Hz, J4J = 9.7 Hz, H-4), 3.65 (ddd, 1 H, Jy,4- = 9.7 Hz, Js-_6-ab = 4.4, 2.2 Hz, H-5'), 2.3-1.9 (m, 21 H, 7 x OAc); 1 3C NMR (100 MHz, CDC13): 5 -170.47, 170.23, 170.17, 169.46, 169.27, 168.92 (6 x C H 3 C H O ) , 103.61 (d,JC.F = 229.52 Hz, C-l), 100.52 (C-l'), 75.35 (C-4), 72.88 (C-3'), 71.99 (C-5'), 71.55 (C-2'), 70.65 (d, JC.H = 4.24 Hz, C-5), 70.32 (d, JC.F = 24.3 Hz, C-2), 68.72 (C-3), 67.76 (C-4'), 61.57 (C-6), 61.09 (C-6'), 20.81, 20.62, 20.55, 20.52 (4 x C H 3 C H O ) ; l 9F NMR (188 MHz, CDCI3, ref. To CF3C02H): 8-73.1 (dd, 1 F, JFM., = 53.1 Hz, JF.H.2 = 24.3 Hz); LCMS: m/z: 661.6, calcd. For C26H35F017 [M+Na] 661.54 106 a-Cellobiosyl fluoride (3) OH OH^ a-Cellobiosyl fluoride was obtained by treatment of a-cellobiosyl fluoride per-O-acetate (2) (4.0 g, 6.3 mmol) with catalytic sodium methoxide (1.0 M, 0.94 mL) in dry methanol (MeOH, 55 mL) for 30 minutes. The product crystallized spontaneously upon completion of the reaction, after which the reaction mixture was neutralized with Amberlite IR-120 (H+) resin and the white crystals filtered off (3) (1.6 g, 4.8 mmol, 76%). Rf (5:2:1 EtOAc/MeOH/H20) = 0.46; 'H NMR (400 MHz, D20): 8 5.68 (dd, 1 H, J,,F = 53.4 Hz, Ju = 2.7 Hz, H-l),4.51 (d, 1 H, J,-.r = 8.0 Hz, H-l'), 3.95 (m, 1 H, H-5), 3.94-3.87 (m, 3 H, H-6a,b and H-6'a), 3.85 (dd, 1 H, Jl4 = Ju = 9.5 Hz, H-3), 3.75 (dd, 1 H, J4J = J4,s = 9.5 Hz, H-4), 3.71 (dd, 1 H, J6%5 = 12.6 Hz, J6b.6-a = 5.9 Hz, H-6'b), 3.65 (ddd, 1 H, J2j = 9.5, J2.F = 26.2 Hz, J2,i = 2.7 Hz, H-2), 3.50 (dd, 1 H, Jy,r =Ji:r = 9.1 Hz, H-3'), 3.47 (m, 1 H, H-5'), 3.4 (dd, 1 H,J4J=9.2 Hz,J4 r = 9.1 Hz, H-4'), 3.31 (dd, 1 H,J 2 ,r = 9.1 H z , 8 . 0 Hz, H-2'); 1 9F NMR (188 MHz, CDC13, ref. To CF3C02H): 5 -74.0 (d, 1 F, JF,H., = 53.4 Hz, JF,H.2 = 26.2 Hz, F-l); 107 6.1.2 B-Cellobiosyl Fluoride (6) •OAc OAc AcO-2, 3, 6, 2', 3', 4', 6' Hepta-O-acetyl-a-cellobiosyl bromide (4) Br Cellobiose octa-acetate (3.1 g, 4.6 mmol) was dissolved in anhydrous DCM (10 mL) and the system flushed with argon gas and placed in an ice bath. Hydrobromic acid (> 33% HBr/AcOH, 4.1 mL) was added slowly. The reaction mixture was stirred and allowed to warm up to room temperature. After four hours of reaction time, the mixture was added to 40 mL of ice and solid NaHCC*3 added until neutral. The mixture was then extracted with three 25 mL portions of DCM and the combined organic phase washed with water (30 mL). The organic phase was then dried over MgSCv, filtered and the solvents evaporated off. a-Cellobiosyl bromide per-O-acetate was obtained as off-white solids (3.0 g, 4.4 mmol, 95%). Rf (1:1 EtOAc/PE) = 0.24; 'H NMR (400 MHz, D20): 5 6.50 (d, 1 H, Ju = 4.0 Hz, H-l), 5.53 (dd, 1 H, J3,2 = 9.8 Hz, J3A = 9.8 Hz, H-3), 5.33 (dd, 1 H, JrA- = 3.4 Hz, J r x = 0.9 Hz, H-4'), 5.10 (dd, 1 H , J 2 j = 10.4 Hz, J2-j • = 7.8 Hz, H-2'), 4.94 (dd, 1 H,J3-,2 • = 10.4 Hz, J3;4 =3AUz, H-3'), 4.74 (dd, 1 H, J2,3 = 9.8, J2,i = 4.0 Hz, H-2), 4.50 (d, 1 H, J,-i2- = 7.8 Hz, H-l'), 4.5-4.0 (m, 1 H, H-6a), 4.24-4.15 (m, 1 H, H-5), 4.15-4.00 (m, 3 H, H6b, H6'a, H6'b), 3.91-3.82 (m, 1 H, H5'), 3.83 (dd, 1 H, J4,3 = 9.8 Hz, J4J = 9.8 Hz, H-4), 2.40 - 2.00 (m, 24 H, 8 x OAc); 108 2, 3, 6, 2 ' , 3 ' , 4', 6 ' Hepta-O-acetyl-p-cellobiosyl fluoride (5) OAc OAc a-Cellobiosyl bromide-per-O-acetate (3.0 g, 4.4 mmol) was dissolved in anhydrous acetonitrile ( C H 3 C N , 12 mL) under argon gas and then two equivalents of silver fluoride (AgF, 1.1 g) were added. The reaction vessel was then covered with aluminum foil and the reaction was allowed to proceed for two hours. Upon completion, the reaction mixture was mixed with Celite® and filtered to remove silver and unreacted silver fluorides. The solvents were evaporated off and DCM (25 mL) plus water (30 mL) added. The aqueous phase was extracted with two more portions of 25 mL of DCM and then the combined organic phases were washed with water (30 mL). The organic phase was dried over MgSC>4, filtered and the solvents evaporated off. The resulting solid was then purified using column chromatography with 3:1 EtOAc:PE and P-cellobiosyl fluoride per-O-acetate was obtained as a white solid (2.3 g, 3.6 mmol, 82%). Rf (3:1 EtOAc:PE) = 0.61; *H NMR (400 MHz CDCI3): 5 5.26 (dd, 1 H,J/./r=52.5 Hz,7/2 = 5.6 Hz, H-l), 5.10 (m, 1 H, H-4'), 5.07 (dd, 1 H, J R A - = 9.2 Hz, J 3 ; r = 9.0 Hz, H-3'), 4.97 (dd, 1 H, J3.2 = J3.4 = 9.6 Hz, H-3), 4.91 (m, 1H, H-2), 4.84 (dd, 1 H , J r , r = 8.3 Hz, J r , r = 9.0 Hz, H-2'), 4.49 (d, 1 H,Jr,2=8.3 Hz, H-l'), 4.46 (m, 1 H, H-6a), 4.28 (dd, 1 H, J 5 M = 12.5 Hz, J S A = 4.3 Hz, H-5), 4.00 (m, 1 H, H-6'a), 3.95 (dd, 1 H, J6B.5 = 12.5, J6b.6a = 1.9 Hz, H-6b), 3.87 (dd, 1 H, JY.6-A = 9.1 Hz, J5A = 9.0 Hz, H-5'), 3.75 (m, 1 H, H-6'b), 3.60 (m, 1 H, H-4), 2.07 - 1.86 (m, 21 H, 7 x OAc); 1 3C NMR (100 MHz, CDCI3); 8 170.3, 170.1, 170.0 (3 x C H 3 C H O , G) 169.4, 169.1, 169.1, 168.9 (4 x CH3CHO, G') 105.8 (d, JC,F = 218.43 Hz, C-l), 100.7 (C-l'), 75.2 (C-5'), 72.6 (C-6'), 72.4 (d, J c.// = 2.61 Hz, C-3'), 71.8 (C-4), 71.5 (d, JC,H = 7.2 Hz, C-2'), 71.3 (C-4'), 71.1 (d, JC,F= 29.8 Hz, C-2), 67.6 (C-3), 61.5, 61.4 (C-5,6), 20.60, 20.57, 20.45, 20.39, 20.33, 20.31 (6 x C H 3 C H O ) ; 19F NMR (188 MHz, CDCI3, ref. To CF3C02H): 8 - 58.2 (dd, 1 F, JFH., = 52.5 Hz, JFH.2 = 8.1 Hz, F-l); 109 P-Cellobiosyl fluoride (6) OH OH P-Cellobiosyl fluoride-per-O-acetate (4.2 g, 6.6 mmol) was dissolved in anhydrous MeOH (115 mL) under argon gas and then saturated sodium methoxide in methanol (NaOMe/MeOH, 0.7 mL) was added drop-wise. The reaction was allowed to proceed for one hour and the reaction terminated by the addition of Amberlite IR-120 (H+) resin until almost neutral. The solvents were evaporated off, and the solid material was purified by column chromatography with 15:4:1 EtOAc:MeOH:H20 yielding white crystals (2.5 g, 6.2 mmol, 95%). Rf (15:4:1 EtOAc:MeOH:H20) = 0.20; *H NMR (400 MHz, D20): 8 5.29 (dd, 1 H, Ju = 53.0 Hz, H-l), 4.53 (d, 1 H, J, , r = 7.9 Hz, H-l'), 4.02-3.98 (m, 2 H, H-5, H-2'), 3.98-3.82 (m, 3 H, H-6a,b and H-6'a), 3.75 (dd, 1 H, J6%5 = 12.5 Hz, H-6'b), 3.73 (dd, 1 H, Ji.2=JiA = 9.3 Hz, H-3), 3.68 (dd, 1 H, J4J = 9.3 Hz, J4,s = 8.2 Hz, H-4), 3.52-3.48 (m, 2 H, H-5' andH-6'b), 3.48-3.40 (dd, 1 U,J4J=J4J = 9.3 Hz, H-4'), 3.38-3.20 (m, 1 H, H-3'), 3.25-3.18 (m, 1 H, H-2); 8 - 67.1 (dd, 1 F, JFH-i = 53.0 Hz, JFH.2 = 12.9 Hz, F-l); 19F NMR (188 MHz, CDC13, ref. To CF3C02H): 8-67.1 (dd, 1 F, JFH., = 52.7 Hz, JFH.2 = 12.9 Hz, F-1); LCMS: m/z: 367.0, 711.3 calcd. For Ci2H21FOi0 [M+Na] 367.3, [2M+Na] 711.6. 110 6.1.3 6'-Deoxy-6'-Fluoro-a-Cellobiosyl Fluoride (12) 1, 2, 3, 6, 2', 3' Hexa-O-acetyl-4', 6'-benzylidene cellobiose (8) Cellobiose (20.0 g, 58.4 mmol) was dissolved in anhydrous dimethyl formamide (DMF, 800 mL). After the addition of 1.5 equivalent of p-anisaldehyde dimethyl acetal (13 mL), a spatula of toluene 4-sulfonic acid was added. The reaction was then placed under light vacuum and 60°C water bath. After six hours, undissolved and unreacted cellobiose was filtered off and the mixture neutralized with triethylamine. After the solvents were evaporated off, yellow syrup was obtained. Acetylation was then performed by the addition of pyridine (200 mL) and acetic anhydride (100 mL) and mixture allowed to react overnight. In the workup, the solvents were evaporated off and 10% v/v HC1 (300 mL) added and then the mixture extracted with two 100 mL portions of DCM. The combined organic phase was then washed with saturated NaHCC>3 (300 mL), brine (300 mL) and water (300 mL). The organic phase was then dried over MgSC^ t and the solvents evaporated off. The product was then purified using column chromatography with 3:1 EtOAc: PE was obtained as off white crystals. Rf (3:1 EtOAc:PE) = 0.60; *H NMR (300 MHz CDC13): 5 7.32 (d, 2 H, JiPhAPh = 8.8 Hz, H-3 Ph), 6.90 (d, 2 H, JiPhAPh = 8.8 Hz, H-2 Ph), 6.23 (d, 41% H, Jla2 = 3.7 Hz, H-l a-anomer), 5.64 (d, 59% H, J,p,2 = 9.3 Hz, H-l p-anomer), 5.43 (1H, CH), 5.29-5.18 (m, 2 H, H-3, H-3'), 5.02 (dd, 1 U,J2.ia = 9.3 Hz, J2J = 9.2 Hz, H-2), 4.90 (dd, 1 H,J2j- = 8.6 Hz, J2j/=7.7Hz,H-2'),4.57(dd, 1 H,J , , r = 7.7 Hz, J / j = 2.9 Hz, H-l'), 4.46 (dd, 1 H, J6aj = 12.5 Hz, J6aM =2.0 Hz, H-6a), 4.4-4.25 (m, 1 H, H-6'), 4.08 (dd, 1 H, J6b,5 = 12.5 Hz, J6h,6a = 2.0 Hz, H-6b), 3.6 (m, H-5), 3.82 (dd, 1 H, J4J, J4,5 = 8.9, 3.8 Hz, H-4), 3.64 (dd, 1 H, J4;5 J4J • = 9.6, 2.1 Hz, H-4'), 3.46 (m, 1 H, H-5'), 2.1 - 1.9 (m, 18 H, 6 x OAc). I l l 1, 2, 3, 6, 2', 3' Hexa-O-acetyl-4', 6'-benzylidene cellobiose (8) (3.678 g, 5.16 mmol) was dissolved in 50 mL EtOAc followed by the addition of MeOH (50 mL) and then palladium hydroxide over carbon catalyst (PdOH/C, 0.82 g) was added along with glacial acetic acid in catalytic amounts (20.53 uL). The reaction vessel was then placed on the hydrogenator where the system was evacuated and hydrogen then supplied to the system. The reaction was completed after two days and the catalyst filtered off with the addition of Celite®. The solvents were evaporated off and the product purified by column chromatography with 3:1 EtOAc/PE and then 4:1 EtOAc/PE, obtaining a pure white solid (2.79 g, 4.69 mmol, 91%). Rf (3:1 EtOAc/PE) = 0.20; !H NMR (400 MHz CDC13): 8 6.18 (d, 0.28 M H, Jia.2a = 3.7 Hz, H-la), 5.61 (d, 0.72 M H, Jm2p = 8.2 Hz, H-l(3), 5.36 (dd, 0.28 M H, .7=9.6, 9.4, H-3'a), 5.17 (dd, 1 H,J34 0 = 9.1 Hz, J3,2 = 8.9 Hz, H-3p),4.98 (dd, 1 H,J23 = 8.9 Hz, J2,!a = 8.2 Hz, H-2), 4.92 (dd, 1 H, J3:2- = 9.2 Hz, Jr,4- = 4.1 Hz, H-3'), 4.77 (dd, 1 H,J2.i=9.2 Hz, J2-,r = 1.5 Hz, H-2'), 4.51 (d, 1 H,Jr,2 =7.5 Hz, H-l'), 4.41 (m, 1 H, H-6b), 4.04 (m, 1 H, H-5b), 3.95 (m, 1 H, H-6a), 3.82 (dd, 1 H, J4J = 9.3, J4,3 = 9.1 Hz, H-4), 3.78-3.70 (m, 3 H, H-5 and H-6'a,b), 3.67-3.60 (m, 1 H, H-4'), 3.38 (m, 1 H, H-5'); 112 1, 2, 3, 6, 2', 3' Hexa-0-acetyl-6'-deoxy-6'-fluoro cellobiose (10) OAc 1, 2, 3, 6, 2', 3' Hexa-O-acetyl cellobiose (9) (73 mg, 0.123 mmol) was placed in a flamed-dried round bottom flask and dissolved in DCM (1.5 mL). The reaction flask was flushed with argon and cooled to -40°C in a dry ice bath. 1.1 equivalents of diethylaminosulfur trifluoride (DAST, 0.018 mL) was added dropwise. The reaction was then allowed to warm up to room temperature and proceed for four hours. Upon completion, the reaction was worked up by adding saturated NaHCCh dropwise until neutral. The mixture was then extracted with three portions of 5 mL DCM. The combined organic phase was washed with brine, water and then dried over MgSC>4, filtered and the solvents evaporated off. After purification by column chromatography using 1:1 EtOAc:PE, a pure white solid was obtained (12.9 mg, 0.022 mmol, 18%). Rf (3:1 EtOAc: PE) = 0.41; 'H NMR (200 MHz CDC13): 8 6.22 (d, 0.37 M H, J,a2a = 3.7 Hz, H-l a-anomer), 5.63 (d, 0.63 M H, Jifi2p= 8.3 Hz, H-l P-anomer, a/p = 3.7/8.3), 5.41 (dd, 0.37 M H, J 3 a 2 a = 10.1 Hz, JiaA = 9.5 Hz, H-3 a-anomer), 5.19 (dd, 0.63 M H, J3p2p = 9.3 Hz, J3pA = 9.1 Hz, H-3 p-anomer), 5.25 (dd, 0. 63 M H, J2p3p = 9.4, J2pjp= 8.4 Hz, H-2 p-anomer), 4.99 (dd, 0.37 M H, J2a.3a = 10.1 Hz, J 2 a J a = 3.7 Hz, H-2 a-anomer),4.90 (dd, 1 H, Jr,r = 9.6 Hz, J3A- = 5.3 Hz, H-3'), 4.86 (dd, 1 H, J2-,y = 9.6 Hz, Jrx = 7.6 Hz, H-2'), 4.75- 4.50 (m, 2 H, H-6'a and H-6'b), 4.46 (d, 1 H, Ji -j' ~ 7.6 Hz, H-l'), 4.44 (ddd, 1 H, J6a.5 = 12.0 Hz, J6a,6b = 4.0 Hz, J6aA = 2.0 Hz, H-6a), 4.09 (ddd, 1 H, J6bJ = 12.0 Hz, J6b.6a = 4.0 Hz, J6bA = 1.7 Hz, H-6b), 3.85 - 3.65 (m, 3 H, H-4, H-4' and H-5), 3.8 - 3.2 (m, 1 H, H-5'), 2.82 (s, 1 x OH), 2.3-1.8 (m, 18 H, 6 x OAc) 19F NMR (188 MHz, CDC13, ref. To CF3C02H): - 158.41 (td, 1 F, Jr,H.y= 72.1 Hz, JFM.6a = Jr.H-6'b = 24.9 Hz, F-6') LCMS: m/z: 619.5, calcd. For C24H33FO16 [M+Na] 619.5 113 2, 3, 6, 2', 3' Penta-0-acetyl-6'-deoxy-6'-fluoro-a-cellobiosyl fluoride (11) HO' 1, 2, 2', 3, 3', 6-hexa-O-acetate 6'-fluoro-cellobiosyl fluoride was synthesized according to a published procedure 1 1 5 . 1, 2, 2', 3, 3', 6-hexa-0-acetyl-6'-deoxy-6'-fluoro-cellobiose (10) (290 mg, 0.50 mmol) was placed in a 50 mL polypropylene Nalgene™ bottle flushed with argon which was placed in an ice bath. HF/pyridine (70% hydrogen fluoride/pyridine, 1.44 mL) was added and mixture stirred until it was homogeneous. The reaction vessel was then sealed and allowed to warm up to room temperature. After one hour, the reaction was terminated by the addition of saturated sodium bicarbonate solution drop-wise until the pH of mixture was neutral. The reaction was then diluted with dichloromethane (DCM, 20 mL) and the organic phase washed with 100 mL saturated NaHC03 followed by 20 mL water. The organic phase was then dried over magnesium sulphate (MgS04) and filtered. After the solvent was evaporated off, 1, 2, 2', 3, 3', 6-hexa-0-acetyl-6'-deoxy-6'-fluoro-cellobiosyl fluoride (217 mg, 0.40 mmol, 80%) was obtained as white crystals. Rf (3:1 EtOAc:PE) = 0.62; 'HNMR (200 MHz CDC13): 8 5.65 (dd, 1 H,J,,F = 53.1 Hz, J,,F = 2.5 Hz, H-l), 5.42 (dd, 1 H, JiA = 9.8 Hz, J3J = 9.5 Hz, H-3), 5.00-4.85 (m, 1 H, H-2), 4.85 (dd, Jrx = 9.5 Hz, Jrj• = 7.6 Hz, H-2'), 4.80-4.60 (m, 2 H, H-5, H-4'), 4.59-4.55 (dd, 1 H,J6a,6b= 12.2 Hz, J6a,5= 1.7 Hz, H-6a), 4.45 (d, 1 H, J, -2 = 7.6 Hz, H-l'), 4.19-4.10 (dd, 1 H, J6b.6a = 12.2 Hz, J6b,5 = 4.1 Hz, H-6b), 4.08-3.95 (m, 1 H, H-5'), 3.78 (dd, 1 H, JiA- = 9.8 Hz, Jy.r = 9.5 Hz, H-3'), 3.68 (dd, 1 H, J4j=J4,5 = 9.8 Hz, H-4), 3.55-3.45 (m, 2 H, H-6'ab), 2.75 (m, 1 H, OH), 2.2-1.9 (m, 15 H, 5 x OAc);19F NMR (200 MHz, CDC13, ref. To CF3CO2H): 8 -73.27 (dd, IF, JF,H., = 53.1 Hz, JFH.2 = 24.1 Hz, F-l) - 158.385 (ddd, 1 F, JF:H-S- = 70.2 Hz, JF-.H-6 « = Jrwb = 24.5 Hz, F-6'); LCMS: m/z: 579.4, calcd. For C22H30F2O14 [M+Na] 579.4 114 6'-Deoxy-6'-fluoro-a-cellobiosyl fluoride (12) 1, 2, 2', 3, 3', 6-Hexa-0-acetyl-6'-deoxy-6'-fluoro-cellobiosyl fluoride (11) (217 mg, 0.39 mmol) was dissolved in 2.99 mL of anhydrous MeOH in a flame-dried round bottom flask under argon gas. A catalytic amount of NaOMe/MeOH (0.15 equivalent, 58.49 uL) was added and the reaction allowed to proceed for five hours. Upon completion, it was neutralized with Amberlite IR-120 (Trt) resin, filtered and the solvents evaporated off. The products were then purified by column chromatography using 7:2:1 EtOAc:MeOH:H20 and a-6'-fluoro-cellobiosyl fluoride was obtained as a white solid (40 mg, 0.12 mmol, 30%). Rf (7:2:1 EtOAc:MeOH:H20) = 0.50; 'H NMR (400 MHz CDC13): 5 5.53 (dd, 1 H, J,.F = 53.4 Hz, Ju = 2.7 Hz, H-l), 4.61-4.55 (m, H-2'), 4.6-4.4 (m, 1 H, H-3'), 4.39 (d, 1 H,J,-,2 =8.0 Hz, H-l'), 3.80 (dd, Jy,4-= 8.8, Jy,6- = 2.28 Hz, H-5'), 3.77-3.71 (m, 2 H, H-6'a,b), 3.70 (dd, 1 H, J4,} = 9.5 Hz, J4.5 = 9.4 Hz, H-4), 3.59 (dd, 1 H, J32 = 9.6 Hz, J34 = 9.5 Hz, H-3), 3.55-3.42 (m, 2 H, H-6a,b), 3.50 (ddd, 1 H, J2,F = 25.9 Hz, J2,3 = 9.6 Hz, J2J = 2.7 Hz, H-2), 3.39-3.32 (m, 1 H, H-5), 3.18 (dd, 1 H, J4\y = 8.8 Hz, J4-,y = 8.3 Hz, H-4'); 19F NMR (200 MHz, CDC13, ref. To C F 3 C O 2 H ) : 5 -73.3 (dd, 1 F, JF,H-i = 53.8 Hz, JFH.2 = 25.9 Hz, F-l) - 159.4, (ddd, 1 F, JFH.y = 73.4 Hz, JFH.6-a = JFH.6b = 25.4 Hz, F-6'); LCMS: m/z: 369.2, 715.3; calcd. For C12H20F2O9 [M+Na], 369.3, [2M+Na] 715.5 115 6.1.4 6'-Deoxy-6'-Fluoro-p-Cellobiosyl Fluoride 2, 3, 6, 2', 3', 4' Hexa-0-acetyl-6'-deoxy-6'-fluoro-a-cellobiosyl bromide (14) 1, 2, 3, 6, 2', 3' Hexa-0-acetyl-6'-deoxy-6'-fiuoro cellobiose (10) (6.5 g, 11.1 mmol) was dissolved in pyridine (150 mL) and acetic anhydride (75 mL). The reaction was allowed to proceed for one hour and twenty minutes and then terminated with neutralization using solid sodium bicarbonate. The mixture was extracted with three 50 mL portions of DCM and the combined organic phase washed with water (50 mL), brine (50 mL) and then water (50 mL) again. The organic phase was dried over MgSC<4, filtered and the solvents evaporated off. Solid was then dissolved in anhydrous DCM (17.6 mL) in a flame-dried round bottom flask under argon and placed in an ice bath. HBr/AcOH (> 33%, 7.25 mL) was added dropwise and the reaction allowed to proceed for one hour and thirty-five minutes. The reaction was terminated by the addition of water (100 mL) followed by the addition of solid sodium bicarbonate until neutral. The mixture was then extracted with two 150 mL portions DCM. The combined organic phase is then washed with saturated NaHC03 (200 mL), brine (100 mL) and water (200 mL) was dried over MgSC>4, filtered off and the solvents evaporated. Purification by column chromatography using 1:1 EtOAc:PE yielded an off-white solid (0.481 g, 0.731 mmol, 6.6%). Rf (1:1 EtOAc:PE) = 0.45; l 9F NMR (200 MHz, CDC13, ref. To CF3CO2H): - 155.297 (ddd, 1 F, J= 68.73, 21.54, 21.5 Hz, F-6'); LCMS: m/z: 681.3, 1341.6; calcd. For C24H32BrFO,5 [M+Na], 682.4; [2M+Na], 1341.8. 116 2, 3, 6, 2', 3', 4' Hexa-0-acetyl-6'-deoxy-6'-fluoro-P-cellobiosyl fluoride (15) OAc 6'-Deoxy-6'-fluoro-a-cellobiosyl bromide per-O-acetate (14) (2.17 g, 3.29 mmol) was dissolved in anhydrous C H 3 C N (8.5 mL) in a flame-dried round bottom flask under argon. AgF (1.1 equivalent, 0.62 g) was added and the reaction flask wrapped in foil and stirred for one hour and ten minutes. Upon completion, the reaction mixture was mixed with Celite® and filtered off to remove silver and unreacted silver fluorides. The solvents were evaporated off and DCM (10 mL) plus water (50 mL) added. The aqueous phase was extracted with two 25 mL portions of DCM and then combined organic phase washed with water (20 mL). The organic phase was dried over MgS04, filtered and the solvents evaporated off. The resulting solid was then purified using column chromatography with 3:1 EtOAc:PE and a white solid of 6'-deoxy-6'-fluoro-p-cellobiosyl fluoride per-O-acetate was obtained as a white solid (1.75 g, 2.93 mmol, 89%). Rf (1:1 EtOAc: PE) = 0.43; *H NMR (400 MHz CDCl3):5 5.15(dd, 1 H,y,,F = 53.1 Hz, .//,, = 7.2 Hz, H-l), 4.42 (d, 1 U,Ji;2 = 7.9 Hz, H-l'), 3.90-3.83 (d, 1 H, JX6 = 11.8 Hz, H-5), 3.85-3.70 (m, 2 H, H-4, H-4'), 3.64-3.53 (m, 4 H, H-3, H-5', H-6'ab), 3.43-3.35 (m, 4 H, H-2, H-6ab, H3'), 3.22 (dd, 1 H, Jr.y = 9.13 Hz, J2;r = 7-9 Hz, H-2'); 19F NMR (200 MHz, CDC13, ref. To CF3C02H): - 155.3 (ddd, 1 F, J= 71.3, 21.1, 21.1 Hz, F-6'), 5 - 58.5 (dd, 1 F, JF,H-I ~ 52.5 Hz, JFM.2 = 9.1 Hz, F-l); 117 6'-Deoxy-6'-fluoro-P-cellobiosyl fluoride (16) 6'-Deoxy-6'-fluoro-P-cellobiosyl fluoride per-O-acetate (15) (1.75 g, 2.93 rnrnol) was dissolved in anhydrous MeOH (24.2 mL) in a flame-dried round bottom flask under argon. A catalytic amount of NaOMe/MeOH (411.7 uL) was added and the reaction allowed to proceed for one hour fifteen minutes at room temperature. Upon completion, the reaction was neutralized with Amberlite IR-120 (H+) resin, filtered and the solvents evaporated off. The product was then purified with column chromatography using 7:2:1 EtOAc:MeOH:H20 yielding a white solid (0.35 g, 1.01 mmol, 35%). Rf (7:2:1 MeoH:EtOAc:H20) = 0.52; 'H NMR (400 MHz CDC13): 5 5.15 (dd, 1 H,//./=• =53.1 Hz,J />2 = 7.2 Hz, H-l), 4.44 (d, 1 H, Jr.r = 7-9 Hz H-l'), 3.88 (d, 1 H, J5,6a,b =11.8 Hz, H-5), 3.82 - 3.70 (m, 2 H, H-4, H-4'), 3.65 - 3.55 (m, 3 H, H-3, H-5', H-6'a,b), 3.45-3.38 (m, 3 H, H-2, H-3', H-6a,b), 3.23 (dd, Jr.y = 8.5 Hz, J2;y = 7.9 Hz, H-2'); 1 3C NMR (100 MHz, CDCI3); 5 109.4 (d, JC.F = 195.7 Hz, C-l), 102.8 (C-l'), 58.2 (C-5), 81.2 (C-2'), 77.6 (C-3'), 75.1 (C-2'), 74.3 (d, JCF= 17.7 Hz, C-4), 72.9 (C-5'), 72.7 (C-6'), 70.8 (C-5), 68.2 (d, JC,H = 6.8 Hz, C-5), 59.2 (C-6'); 19F NMR (300 MHz, CDCI3, ref. To CF3C02H): - 67.393 (dd, IF, J = 52.39, 13.6 Hz, F-l) -153.73 (ddd, 1 F,J = 53.1, 21.3, 21.3 Hz, F-6'); 19F NMR 'H decoupled (300 MHz, CDC13, ref. To CF3C02H): 5 -67.45 (F-l), -153.73 (F-6'); 118 6.2 Enzymat ic S y n t h e s e s General methods All buffer chemicals and other reagents were obtained from Sigma or Aldrich Chemical Cos. unless otherwise noted. All enzymes were obtained from Sigma or Serva Chemical Cos. unless otherwise noted. All NMR spectra for all products synthesized in this part of the experiment were obtained using Bruker Advance 400 9.6 Telsa magnet. Matrix-Assisted Laser Desorption/Ionization (MALDI) mass spectrometry was performed using a Bruker Biflex IV system. All single crystal x-ray diffraction studies were carried out using a Bruker X8 APEX CCD diffractometer. All thin layer chromatography was carried out using E. Merck pre-coated aluminium-backed sheets of Silica Gel 60 F254 with detection being effected by heating with 10% ammonium molybdate in 2M H2SO4. General enzymatic synthesis procedures All of the cellulose polymer synthesizing reactions are started by first dissolving the desired starting materials (a-CBF, P-CBF, 6'F a-CBF, 6'F P-CBF) in aqueous buffers followed by the addition of acetonitrile if necessary and then the addition of enzymes. Reaction is then allowed to proceed until no more precipitate is observed or until the mixture is acidic. At this point the insoluble cellulose is centrifuged and the solution is separated from the solid. The solid will then be washed with 5:1 methanol/water mixture, centrifuged and solution removed. The above "wash" procedure will be repeated three times and then the solid obtained will be dried in the freeze drier. General MALDI-TOF procedures All of the cellulose polymer samples for MALDI-TOF are placed onto the target in a "Sandwiched" fashion: 30 mg DHB matrix material is dissolved in 1 mL 2:1 H 2 O / C H 3 C N with 0.1% TFA. 1 uL of the matrix mixture is applied onto the target and allowed to dry. Meanwhile samples are dissolved in 4:1 H2O/DMSO, then 1 uL of the sample solution is loaded onto the dried matrix bed and let dry. Another 1 uL of the matrix mixture is applied on top of the dried sample and let dry again. Spectra are then collected using positive reflectron mode with detection limit optimized for 1 to 3 kDa. 119 6.2.1 Trichoderma viride O n o z u k a R-10 55 mg of P-CBF is dissolved in various ratios of acetonitrile and 50 mM sodium phosphate buffer at pH 6.7 along with 1 mg of enzyme and allowed reaction to proceed for one day. Solid precipitate then collected as mentioned in the General Methods. Acetonitrile/aqeous sodium phosphate buffer ratio 7:1 Reaction carried out in 4.4 mL C H 3 C N and 0.63 mL sodium phosphate buffer. 38 mg of solid precipitate obtained (69% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 1 3C CP/MAS NMR identified as cellulose II. Acetonitrile/aqeous sodium phosphate buffer ratio 5:1 Reaction carried out in 4.2 mL C H 3 C N and 0.63 mL sodium phosphate buffer. 35 mg of solid precipitate obtained (63% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 1 3C CP/MAS NMR identified as cellulose II. Acetonitrile/aqeous sodium phosphate buffer ratio 2:1 Reaction carried out in 3.3 mL C H 3 C N and 1.6 mL sodium phosphate buffer. 14 mg of solid precipitate obtained (25% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 1 3C CP/MAS NMR identified as cellulose II. 120 6.2.2 Cellulose Production in 2:1 Acetonitrile/Aqueous Buffer 120 mg of P-CBF is dissolved in 3.3 mL of acetonitrile and 1.7 mL of 50 mM sodium phosphate buffer at pH 6.7 along with 5 mg of enzyme and allowed to proceed for one day. Solid precipitate then collected as mentioned in the General Methods. Aspergillus niger 39 mg of solid precipitate obtained (33% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 1 3C CP/MAS NMR identified as cellulose II. Trichoderma reesei 44 mg of solid precipitate obtained (37% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 1 3C CP/MAS NMR identified as cellulose II. Trichoderma viride 29 mg of solid precipitate obtained (24% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 1 3C CP/MAS NMR identified as cellulose II. Cex 15 mg of solid precipitate obtained (13% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 121 6.3 Cellulose Production with Glycosynthase Mutants 6.3.1 E197G 150 mg of a-CBF is dissolved in 1.5 mL of 50 mM sodium phosphate buffer at pH 6.7 along with 500 uL of 1.6 mg/mL of enzyme and allowed to proceed for two days. Solid precipitate then collected as mentioned in the General Methods. 43.00 mg of solid precipitate obtained (29% yield). MALDI-TOF identified the solid collected as polymer of cellulose. 1 3C CP/MAS NMR identified as cellulose II. 6.3.2 E197A 200 mg of a-CBF is dissolved in 4.0 mL of acetonitrile and 1.0 mL of 50 mM ammonium bicarbonate buffer at pH 8.0 along with 1000 uL of 1.6 mg/mL of enzyme and allowed to proceed for two days. Solid precipitate then collected as mentioned in the General Methods. MALDI-TOF identified the solid collected as polymer of cellulose. 31.8 mg of solid precipitate obtained (16% yield). 6.3.3 E197S 50 mg of a-CBF is dissolved in 1.0 mL of acetonitrile and 0.25 mL of 50 mM ammonium bicarbonate buffer at pH 8.0 along with 250 pL of 1.6 mg/mL of enzyme and allowed to proceed for two days. Solid precipitate then collected as mentioned in the General Methods. MALDI-TOF identified the solid collected as polymer of cellulose. 6 mg of solid precipitate obtained (12% yield). 122 6.4 Syntheses of 6'-Fluorinated Cel lu lose us ing 6 ' F - C B F 6.4.1 E197A 100 mg of 6'F a-CBF is dissolved in 2.0 mL of acetonitrile and 0.5 mL of 50 mM ammonium bicarbonate buffer at pH 8.0 along with 500 uL of enzyme and reaction allowed to proceed for two days. Solid precipitate then collected as mentioned in the General Methods. MALDI-TOF identified the solid collected as polymer of cellulose. 8 mg of solid precipitate obtained (8% yield). 1 3C CP/MAS NMR identified as cellulose II 6.4.2 Cex 100 mg of 6'F P-CBF is dissolved in 2.0 mL of acetonitrile and 0.5 mL of 50 mM ammonium bicarbonate buffer at pH 8.0 along with 500 uL of enzyme and reaction allowed to proceed for one day. Solid precipitate then collected as mentioned in the General Methods. MALDI-TOF identified the solid collected as polymer of cellulose. 10 mg of solid precipitate obtained (10% yield). 123 6.5 N M R S p e c t r o s c o p y 6.5.1 NMR Spectrometer Solid state NMR experiments were performed on a Bruker ADVANCE DSX-400 NMR spectrometer operating at frequencies of 400.13 MHz for 'H, 376.343 MHz for l 9F and 100.64 MHz for 1 3C. The spectrometer was equipped with one 1000 W amplifier for 'H or 1 9F and two 300 W amplifiers for the lower frequency nuclei such as 1 3C. For experiments that involve the generation of *H and 19F frequencies simultaneously, an additional 1000 W amplifier (from a Bruker MSL-400 spectrometer) was added. A Silicon Graphics workstation running version 2.6 of Bruker'sXWIN-NMR software controls the spectrometer with FIDs collected with digital filtering. Pulse sequence programs for the NMR experiments performed throughout this thesis will be provided in later sub-sections. 1 3C NMR spectra were referenced to adamantane and 1 9F NMR spectra were referenced to octadecasil. 6.5.2 Probes Solid state NMR experiments were performed using two different triple-resonance magic angle spinning probes, a Bruker H/X/Y triple-resonance CP MAS probe equipped with a 4 mm MAS stator and a custom-built H/F/X triple-resonance probe equipped with a 7 mm stator. The 4 mm triple-resonance probe was capable of spinning samples at rates up to 15 kHz with a precision and stability of ± 5 Hz. The spinning rate was detected with an internal optical spin rate detector in a feed-back loop with the MAS unit. The MAS unit also controlled the spin rate by moderating drive and bearing gas pressures. The custom-built H/F/X triple-resonance probe equipped with a 7 mm stator was originally a Bruker H/X 7 mm CP MAS probe with a single solenoid coil which could be 124 double-tuned to either the 'H or 19F frequency and a lower frequency such as 1 3C, 29Si, or 15N. The probe was later modified by replacing the original boron-nitride stator with a cylindrical standard-speed 7 mm Kel-F MAS stator from Doty Scientific. The 7 mm probe was also equipped with a heater and thermocouple (in a controlled feed-back loop) making it possible to perform variable temperature experiments ranging between 150 K and 400 K. 6.5.3 Magic Angle Spinning In magic angle spinning experiments the spinning rate was automatically measured by the optical spinning rate detector in the 4 mm probe. However, since the 7 mm probe was not equipped with the rate detector, a small amount (-10-20 mg) of KI must be incorporated into the sample to allow the measurement of the spinning rate. KI can either be mixed directly into samples or place on top at the bottom of the rotor if samples were to be recovered after experiment. To prevent cross contamination of the recovered sample, a small amount of Teflon tape was also used to separate KI from the sample. The 127I resonance (80.055 MHz) of KI in the rotor will produce spinning sidebands spaced by the spinning frequency and the spinning rate can then be deduced by measuring the frequency separation between the spinning sidebands. To spin the rotor at a high speed, dried compressed air controlled by a Bruker pneumatic MAS control unit (Part no. H2620) was used to create the bearing and drive pressure. In the 4 mm probe an average of 2500 mbar of bearing and 1500 mbar of drive is necessary to achieve a spinning rate of 15 kHz. For the 7 mm probe, 1500 mbar of bearing and 800 mbar were required to achieve a spinning rate of 8 kHz. 125 6.5.4 Small Sample NMR The 4 mm rotors were designed to hold about 100 mg of sample while the 7mm rotors held around 250 mg of sample. For small samples of less than half of the available spinner volumes, specialized boron nitride inserts containing small sample wells were inserted inside the rotor placing the sample at the center of the coil where signals can most efficiently be detected. Figure 6-1 shows the setup for the 4mm rotor. Due to only one single open end, the bottom insert is placed in the rotor permanently while the top insert can be removed using the small hole on the top of the top insert. Figure 6-2 shows the setup for the open-ended 7mm rotor where both inserts can be removed. All samples were packed into 4 mm, 7 mm rotors with Kel-F or Vespel caps respectively. boron nitride insert Figure 6-1 Small sample setup for the 4mm rotor. Setup includes two boron nitride inserts and one Kel-F cap. The two boron nitride inserts each contains small sample well to reduce the sample size. Figure 6-2 Small sample setup for the 7mm rotor. Setup includes two boron nitride inserts and two vespel caps. The two boron nitride inserts each contain a small sample wells to reduce the sample size. 126 6.5.5 Low Temperature MAS For samples sensitive to higher temperatures, low temperature MAS experiments were performed using the 7 mm H/F/X probe with the experimental setup as seen in Figure 6-3. Low temperature bearing gas was supplied by the boil-off of a large 200L self-pressurizing liquid N 2 "cooling dewar", the bearing gas then passes through another dewar filled with liquid N2 before reaching the MAS probe. Once inside the probe, the flowing bearing gas was adjusted to the desired temperature by an internal heating coil and thermocouple controlled by a BVT3000 temperature control unit. The driving gas was provided by a compressed N2 gas cylinder with its pressure controlled by the Bruker pneumatic MAS control unit. For prolonged experiments at low temperatures, a reservoir system supplying cooling liquid N2 was necessary. The liquid N2 "cooling dewar" was periodically refilled by attaching another dewar the ("refill dewar"). A programmable timer (consumer product obtained form a local hardware store) with a three-way solenoid valve connects pressurized N 2 gas and the "refill dewar" allowed the refilling of the "cooling dewar" from the "refill dewar" every three to five hours by pressurizing the "refill dewar" and "pushing" liquid N2 over to the "cooling dewar". For every refill, the solenoid valve connecting the "refill dewar" and the "cooling dewar" was set to opened for about three to five minutes allowing the transfer of N 2 into the "cooling dewar" ensuring the "cooling dewar" would never go empty during experiments. 127 pressure Pressure liquid building , regulator S ( c l o s e d ) m m ) s (closed) c n ~ > < ^ ^ |l&35j 68 0 Programmable timer Pressure regulator pressure vent Automated liquid N,fill Insulated transfer , tube cooled bearing gas (12 psi) Cooling dewar liquid N2(40L) From dried compressed air or N,gas cylinder (> 4 bar) MAS control unit Drive gas (400 mbar) l-way solenoid valve gas Refill dewar liquid N,(40L) MAS probe Frame flush (-100 mbar) Heater Thermocouple Temperature Control Unit Figure 6-3 Schematic setup of low temperature MAS NMR experiments. The programmable timer allowed automated filling of the cooling liquid N 2 dewar providing about 12 hours of unsupervised operation. The refill liquid N 2 dewar was filled once or twice per day. Adapted from ref. I 2 5. 128 6.5.6 R e f e r e n c e S a m p l e s All experimental parameters such as pulse lengths, CP matching conditions, and chemical shift references were usually found and optimized on reference samples prior to performing experiments on the particular sample under investigation. A simple 'PI or 19F MAS NMR spectrum was first collected to obtain the 'H of 19F r.f. offset value. With the 'H or 1 9F channel on resonance, the CP matching condition for a given 'FI or 19F power level and spinning rate was then identified using the 'paropt' feature in XWIN- NMR. Finally the 1 3C chemical shifts were referenced to liquid tetramethylsilane (TMS). The l 9F chemical shifts were referenced by setting the isotropic peak in the 19F MAS NMR spectrum of octadecasil to -37.8 ppm (with respect to liquid CFCI3). 6.5.7 1 3 C { 1 H } C P M A S N M R 19F {'H} & 1 3C {'H} CP MAS NMR spectra were collected after shimming and setting up the CP conditions, and 19F chemical shift reference on octadecasil reference sample. Before CP experiments, a *H spectrum was first collected to optimizes the 'H offset frequency for cross polarization and/or decoupling. In general, quantitatively reliable l 3C MAS NMR spectra were collected with a 90° pulse with recycle delays more than five times the Ti relaxation time. For compounds with long relaxation times (e.g. 19F compounds), a 30° pulse was used instead. The optimal recycled delay time (Trecycie) can be calculated by the "Ernst angle" relationship99 (Equation 6-1) using the pulse angle a and relaxation time Ti. Equation 6-1 Cos a = exp (-TreCy|e/Ti) 129 6.5.8 Pulse Sequences Simple 90° single pulse NMR experiment (Figure 6-4) can be used to obtain the 1 9F NMR spectra for fluorinated compounds. However for fluorinated cellobiosyl fluoride compounds, the relaxation times were too long to obtain good spectra within reasonable experiment times, and therefore a 30° single pulse NMR experiment (Figure 6-5) was used. Figure 6-6 shows the pulse sequence for l 3C {'H} cross polarization experiments with the 30° pulse length for 1 9F being 1.0 ps, spinning rate of 15 kHz and 6.00 dB of power were used for each spectrum. For l 3C {*H} cross polarization experiments (Figure 6-6), spectra were collected with 4.50 ps of 90° pulses, contact time of 1.0 second, recycle delays between 4 to 5 seconds depending on the compound being experimented on and a spinning rate of 8.5 kHz. During the contact time the power level used were 12.00 dB on 'H and -2.00 dB on 1 3C For 1 3C {'H} cross polarization experiments with *H dipolar decoupling (Figure 6-7) all the other parameters stays the same with the addition 10.00 dB of decoupling on the 'H during acquisition time. For 1 3C {'H} cross polarization experiments with 'H and 1 9F dipolar decoupling, another additional 12.00 dB of 19F decoupling was added to the pulse sequence during acquisition (Figure 6-8). 90°i '(x) 19 F (n) Number of repeats Figure 6-4 Pulse sequence for l 9 F solid-state NMR experiments with a 90° pulse. 130 30° (x ) •(5xTj)/3' Number of repeats Figure 6-5 Pulse sequence for l 9 F solid-state N M R with a 30° pulse instead of the 90° pulse for fluorine nuclei with long relaxation times. a) H 90° contact (CP) Figure 6-6 Pulse sequence for 1 3 C -» ' H C P / M A S solid-state N M R without ' H decoupling X being the 1 3 C channel. 131 ' H 9 0 ° dipolar C P decoupling (DD) V Figure 6-7 Pulse sequence for l 3 C -> 'H CP/MAS solid-state NMR with 'H decoupling. X being the 1 3 C channel. D D 19T Figure 6-8 decoupling. X being the l 3 C channel Pulse sequence for , 3 C 'H CP/MAS solid-state NMR with simultaneous 'H and l 9 F 132 6.6 S ingle Crysta l X-ray Diffraction 6.6.1 Diffractometers, Data Collection and Processing The single crystal XRD data for a-CBF and 6'F a-CBF were collected (by Dr. B. Patrick) on a Bruker X8 APEX diffractometer with graphite monochromated Mo-Ka radiation. Data were collected and integrated using the Bruker SAINT126 software package, corrected for absorption effects using the multi-scan technique (SADABS127) and were corrected for Lorentz and polarization effects. The structure was solved by direct methods128. All non-hydrogen atoms were refined anisotropically. All O-H hydrogen atoms were located in difference maps and refined isotropically, while all other hydrogen atoms were included in calculated positions but not refined. Neutral atom scattering factors were taken from Cromer and Waber129. Anomalous dispersion effects were included in Fcalc130; the values for Af and Af' were those of Creagh and McAuley131. The values for the mass attenuation coefficients are those of Creagh and Hubbell132. All refinements were performed using the SHELXTL133 crystallographic software package of Bruker-AXS. a-Cellobiosyl Fluoride A colourless irregular crystal of C21H21O10F having approximate dimensions of 0.15 x 0.07 x 0.03 mm was mounted on a glass fiber and the data were collected at a temperature of -100.0 + 0.1 °C to a maximum 20 value of 50.1°. Data were collected in a series of (f) and co scans in 0.50° oscillations with 20.0 second exposures. The crystal-to-detector distance was 38.01 mm. 133 Of the 7929 reflections that were collected, 2441 were unique (Rint = 0.040; Friedels not merged); equivalent reflections were merged. The linear absorption coefficient, u, for Mo-Ka radiation is 1.42 cm" 1. Data were corrected for absorption effects with minimum and maximum transmission coefficients of 0.840 and 0.996, respectively. The final cycle of full-matrix least-squares refinement134 on F 2 was based on 2441 reflections and 236 variable parameters and converged (largest parameter shift was 0.00 times its esd) with unweighted and weighted agreement factors of: Rl =E||Fo|-|Fc||/Z|Fo| = 0.069 wR2 = [ I (w (Fo2 - Fc2)2 )/ Z w(Fo2)2]1/2 = 0.097 The standard deviation of an observation of unit weight135 was 1.03. The weighting scheme was based on counting statistics. The maximum and minimum peaks on the final difference Fourier map corresponded to 0.18 and -0.22 e"/A^, respectively. 6'-Deoxy-6'-Fluoro-a-Cellbiosyl Fluoride A colourless needle crystal of C|2H2oF209.MeOH having approximate dimensions of 0.50 x 0.10 x 0.03 mm was mounted on a glass fiber and the data were collected at a temperature of -100.0 + 0.1 °C to a maximum 28 value of 50.3°. Data were collected in a series of <|) and co scans in 0.50° oscillations with 40.0 second exposures. The crystal-to-detector distance was 38.04 mm. Of the 19246 reflections that were collected, 2862 were unique (Rint = 0.084); equivalent reflections were merged. The linear absorption coefficient, p, for Mo-Ka radiation is 1.46 cm~l. Data were corrected for absorption effects with minimum and maximum transmission coefficients of 0.796 and 0.996, respectively. 134 The material crystallizes with one molecule of MeOH in the asymmetric unit. The final cycle of full-matrix least-squares refinement134 on F 2 was based on 2862 reflections and 255 variable parameters and converged (largest parameter shift was 0.00 times its esd) with unweighted and weighted agreement factors of: Rl =S||Fo|- |Fc||/S|Fo| = 0.087 wR2 = [ S ( w (Fo2 - Fc2)2 )/ S w(Fo2)2]1/2 = 0.109 The standard deviation of an observation of unit weight135 was 1.02. The weighting scheme was based on counting statistics. The maximum and minimum peaks on the final difference Fourier map corresponded to 0.20 and -0.22 e'/A ,^ respectively. 135 Kobayashi, S., Kashiwa, K., Kawasaki, T. & Shoda, S. Novel method for polysaccharide synthesis using an enzyme: the first in vitro synthesis of cellulose via a nonbiosynthetic path utilizing cellulase as catalyst. Journal of American Chemical Society 113, 3079-84(1991). Lee, J. H., Brown, M., Kuga, S., Shoda, S. I. & S, K. Assembly of synthetic cellulose I. Proceedings of the National Academy of Sciences of the United States of America 91,7425-7429(1994). Kobayashi, S. et al. Formation and structure of artificial cellulose spherulites via enzymatic polymerization. Biomacromolecides 1, 168-173 (2000). Mark, H. Fifty years of cellulose research. Cellulose Chemical Technology 14, 569-581 (1980). Zogai, X., Nimtz, N., Rodhe, M., Bokranzl, W. & Romling, U. The multicellular morphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as the second component of the extracellular matrix. Molecular Microbiology 39, 1452 (2001). Nobles, D. R., Romanovicz, D. K. & Brown, R. M. Cellulose in cyanobacteria. Origin of vascular plant cellulose synthase? Plant Physiology 127, 529-542 (2001). Stryer, 1. Biochemistry. (W. H. Freeman, New York) (1988). Heinze, T. New ionic polymers by cellulose functionalization. Macromolecular Chemistry Physics 199, 2341-64 (1998). Klemm, D., Heinze, T., Philipp, B. & Wagenknecht, W. New approaches to advanced polymers by selective cellulose functionalization. Acta Polymerica 48, 277-297 (1997). Purves, C. B. Cellulose and Cellulose Derivatives. Pt. I (eds. Spurlin, E. O. H. M. & Graffline, M. W.) (Wiley, New York, 1954). VanderHart, D. L. & Attalla, R. H. Studies of microstructure in native celluloses using solid state C-13 NMR. Macromolecules 17, 1465-1472 (1984). Sugiyama, J., Persson, J. & Chanzy, H. Combined IR and electron diffraction study of the polymorphism of native cellulose. Macromolecules 24, 2461-2466 (1991). Marchessault, R. H. & Sarko, A. in Advanced Carbohydrate Chemistry (ed. Wolfrom, M. L.) (Academic Press, New York, 1967). Marchessault, R. H. & Sundararajan, P. R. 11 (Academic Press, New York, 1983). 136 15. Walton, A. G. & Blackwell, J. Biopolymers (Academic Press, New York, 1973). 16. Mann, J. & Marrinan, H. J. Crystalline modifications of cellulose. Part II. A study with plane-polarized infrared radiation. Journal of Polymer Science 32, 357-370 (1958). 17. Hayashi, J., Sufoka, A., Ohkita, J. & Watanabe, S. The confirmation of existence of cellulose IIIi, IIIn, IVi and IVn by X-ray method. J. Polymer Sci.: Polymer Letters Edition 13,23-27(1975). 18. Davis, W. E., Barry, A. J., Peterson, F. C. & King, A. J. X-ray studies of reactions of cellulose in non-aqueous systems. II. Interaction of cellulose and primary amines. Journal of American Chemical Society 65, 1294-1300 (1943). 19. Sarka, A., Southwick, J. & Hayashi, J. Packing analysis of carbohydrates and polysaccharides .7. Crystal structure of cellulose IIIi and its relationship to other cellulose polymorphs. Macromolecules 9, 857-863 (1976). 20. Sarko, A. Cellulose - How much do we know about its structure (Ellis Horwood, Chichester, 1987). 21. Hess, K. & Kissig, H. Zur Kenntnis der Hochtemperatur-Modifikation der Cellulose (Cellulose IV). Zietschrift fur Physikalische Chemie B - Chemie der Element Arprozesse Auf Bau der Materie 49, 235-239 (1941). 22. Gardiner, E. S. & Sarko, A. Packing analysis of carbohydrates and polysaccharides. The crystal structures of celluloses IVi and IVn. Can. J. Chemistry 63, 173-180 (1985). 23. Kono, H., Numata, Y., Erata, T., Takai, M. 1 3C and !H resonance assignment of mercerized cellulose II by two-dimensional MAS NMR spectroscopies. Macromolecules, 5310-5316 (2004). 24. Atalla, R. H. & Vanderhart, D. L. Native cellulose - a composite of 2 distinct crystalline forms. Science 223, 283 (1984). 25. Gast, J. C, Atalla, R. H. & McKelvey, R. D. The C-13-NMR spectra of the xylo-oligosaccharides and cello-oligosaccharides. Carbohydrate Research 84, 137-146 (1980). 26. Kobayashi, S. Challenge of Synthetic Cellulose. Journal of Polymer Science 43, 693-710(2005). 27. Kuga, S. & Brown Jr., R. M. Silver labelling of the reducing ends of bacterial cellulose. Journal of Carbohydrate Research 180, 345-350 (1988). 28. Gardener, K. H. & Blackwell, J. Structure of native cellulose. Biopolymers 613, 1975-2001 (1974). 137 29. Gardner, K. H. & Blackwell, J. Structure of native cellulose. Biopolymers 13, 1975-2001 (1974). 30. Hespell, R. B. & Canale-Parola, E. Carbohydrate metabolism in Spirochaeta stenostrepta. Journal of Bacteriology 103, 216-226 (1970). 31. Roelofsen, P. A. Cell wall structure as related to surface growth. Acta Bot Neerl 7, 77-89(1958). 32. Tsekos, I. J. JPhycol 35 , 635 (2000). 33. Lamport, A. D. T. Cell wall metabolism. Ann. Rev. Plant Physiol, 235-270 (1970). 34. Brett, C. & Waldron, K. Physiology and Biochemistry of Plant Cell Walls (Unwin Hyman, London, 1990). 35. Okuda, K., Tsekos, L. & Brown, R. M. Cellulose microfibril assembly in erythrocladia subintegra Rosenv.: an ideal system for understanding the relationship between synthesising complexes (TCs) and microfibril crystallization. Protoplasma 180, 49-58 (1994). 36. Sugiyama, J., Chanzy, H. & Revol, J. F. On the polarity of cellulose in the cell wall of Valonia. Planta, 260-265 (1994). 37. Cousins, S. K. & Brown, R. M. X-ray diffraction and ultrastructural analyses of dye-altered celluloses support van der Waals forces as the initial step in cellulose crystallization. Polymer 38 , 897-902 (1997). 38. Cousins, S. K. & Brown, R. M. Cellulose-I microfibril assembly - computational molecular mechanics energy analysis favours bonding by vanderwaals forces as the initial step in crystallization. Polymer 36, 3885-3888 (1995). 39. Brown, R. M. Cellulose Structure and Biosynthesis: What is in Store for the 21st Century? Journal of Polymer Science 42, 487-495 (2004). 40. Scheible, W. R., Eshed, R., Richmond, T., Delmer, D. & Somerville, C. Modifications of cellulose synthase confer resistance to isoxaben and thiazolidinone herbicides in Arabidopsis Ixrl mutants. Proceedings of the National Academy of Sciences of the United States of America 98, 10079-84 (2001). 41. Brown, R. M. & Saxena, I. M. Cellulose biosynthesis: A model for understanding the assembly of biopolymers. Plant Physiol. Biochem. 38 , 57-67 (2000). 42. Read, S. M. & Bacic, R. Plant biology - Prime Time for Cellulose. Science 295 , 59-60 (2002). 138 43. Kono, H., Erata, T. & Takai, M. Determination of the through-bond carbon-carbon and carbon-proton connectivities of the native celluloses in the solid state. Macromolecules 36, 5131-5138 (2003). 44. Kono, H. et al. CP/MAS 1 3C NMR study of cellulose and cellulose derivatives. 1. Complete assignment of the CP/MAS 1 3C NMR spectrum of the native cellulose. Journal of the American Chemical Society 124, 7506-7511 (2002). 45. Saxena, I. M., Brown, R. M., Fevre, M., Germia, R. A. & Henrissat, B. J. Bacteriol, 1419-1424(1995). 46. Salmon, S. & Hudson, S. M. Crystal morphology, biosynthesis, and physical assembly of cellulose, chitin, and chitosan. Journal of Macromolecular Science Review Macromolecular Chemistry Physics C 3 7 , 199-276 (1997). 47. Kennedy, J. F. Bioactive Carbohydrates (eds. Kennedy, J. F. & White, C. W.) (Ellis Horwood, Chichester, 1983). 48. Paulson, J. C. Glycoproteins - What are the sugar chains for? Trends in Biochemical Sciences 14, 272-276 (1989). 49. Hakomori, S. Tumor-associated carbohydrate antigens. Annual Review of Immunology 2,103-126(1984). 50. Kamide, K., Miyamoto, I. & Okajima, K. Formation and properties of the lyotropic mesophase of the cellulose mixed inorganic acid system. Polymer Journal 25, 453-61 (1993). 51. Schwarz, H. H., Richau, K. & Paul, D. Membranes from poyelectrolyte complexes. Polymer Bulletin 25, 95-100 (1991). 52. Benkert, A., Scheller, F. W., Schoessler, W., Micheel, B. & Warwinke, A. Size exclusion redox-labeled immunoassay (SERI): a new forma for homogeneous amperometric creatinine determination. Electroanalysis 12, 1381-21 (2000). 53. Risbud, M. V. & Bhonde, R. R. Suitability of cellulose molecular dialysis membrane for bioartificial pancreas: in vitro biocompatibility studies. Journal of Biomedical Materials Research 54, 436-44 (2001). 54. Kobayashi S, S. J., Kimura S. In vitro synthesis of cellulose and related polysaccharides. Progress in Polymer Science 26, 1525-60 (2001). 55. Paulsen, H. Advances in selective chemical syntheses of complex oligosaccharides. Angewandte Chemie - International Edition in English 21, 155-173 (1982). 56. Schmidt, R. R. New methods for the synthesis of glycosides and oligosaccharides -are there alternatives to the Koenings-Knorr method? Angewandte Chemie -International Edition in English 25, 212-235 (1986). 139 57. Kunz, H. Synthesis of glycopeptides, partial structures of biological recognition components. Angewandte Chemie - International Edition in English 26, 294-308 (1987). 58. Igarashi, K. The Koenings-Knorr reaction. Advances in Carbohydrate Chemistry and Biochemistry 34, 243-83 (1977). 59. Nicolaou, K. C, Dolle, R. E. & Papahatjis, D. P. Practical synthesis of oligosaccharides - partial synthesis of avermectin-BlA. Journal of the American Chemical Society 106,4189-92 (1984). 60. Mukaiyama, T., Murai, Y. & Shoda, S. An efficient method for glucosylation of hydroxy compounds using glucopyranosyl fluoride. Chemistry Letters, 431-2 (1981). 61. Lopez, J. C. & Fraserreid, B. Normal-pentenyl esters versus normal-pentenyl glycosides - synthesis and reactivity in glycosidation reactions. Journal of the Chemical Society - Chemistry Communications, 159-161 (1991). 62. Loenn, H. Glycosylation using a thioglycoside and methyl trifluoromethanesulfonate - a new and efficient method for cis and trans glycoside formation. Journal of Carbohydrate Chemistry 6, 301-6 (1987). 63. Pozsgay, V. & Jennings, H. J. A new stereoselective synthesis of methyl 1,2-trans-l-thioglycosides. Tetrahedron Letters 28, 1375-6 (1987). 64. Takeuohi, K. & Mukaiyama, T. Trityl tetrakis(pentafluorophenyl)borate catalyzed stereoselective glycosylation using glycopyranosyl fluoride as a glycosyl donor. Chemical Letter, 555-6 (1998). 65. Whitesides, G. M. & Wong, C. H. Enzymes as catalysts in synthetic organic chemistry. Angewandte Chemie - International Edition in English, 617-718 (1985). 66. Kobayashi, S. Enzymatic polymerization - synthesis of artificial polymers catalyzed by natural polymers. High. Polym. Jpn. 48, 124-7 (1999). 67. Gijsen, H. J. M., Qiao, L., Fitz, W. & Wong, C. H. Recent advances in the chemoenzymatic synthesis of carbohydrates and carbohydrate mimetics. Chemical Review 96, 443-73 (1996). 68. Ichikawa, Y., Look, G. C. & Wong, C. H. Enzyme-catalyzed oligosaccharide synthesis. Analytical Biochemistry 202, 215-38 (1992). 69. Sinnott, M. L. Enzyme Mechanisms (eds. Page, M. L. & Williams, A.) (Royal Society of chemistry, London, 1987). 70. Sinnott, M. L. Catalytic mechanisms of enzymatic glycosyl transfer. Chemical Reviews 90, 1171-1202 (1990). 140 71. Koshland, D. E. Stereochemistry and the mechanism of enzymatic reactions. Biological Reviews of the Cambridge Philosophical Society 28, 416-436 (1953). 72. Legler, G. Glycoside hydrolases - mechanistic information from studies with reversible and irreversible inhibitors. Advances in Carbohydrate Chemistry and Biochemistry 48, 319-384 (1990). 73. Gebler, J. C, Aebersold, R. & Withers, S. G. Glu-537, Not Glu-461, Is the Nucleophile in the Active-Site of (Lac-Z) Beta-Galactosidase from Escherichia-Coli. Journal of Biological Chemistry 267, 11126-11130 (1992). 74. Wang, Q., Graham, R. W., Trimbur, D., Warren, R. A. J. & Withers, S. G. Changing enzymatic reaction mechanisms by mutagenesis: conversion of a retaining glucosidase to an inverting enzyme. Journal of the American Chemical Society 116, 11594-11595 (1994). 75. Davies, G. & Henrissat, B. Structures and mechanisms of glycosyl hydrolases. Structure 3, 853-9 (1995). 76. McCarter, J. & Withers, S. G. Mechamisms of enzymic glycoside hydrolysis. Current Opinion in Structural Biology 4, 885-92 (1994). 77. Kuroki, R., Weaver, L. H. & Matthews, B. W. A covalent enzyme-substrate intermediate with saccharide distortion in a mutant T4 losozyme. Science 262, 2030-2033 (1993). 78. Fernandez-Mayoralas, A. Synthesis and modification of carbohydrates using glycosidases and lipases. Glycoscience Synthesis of Oligosaccharides and Glycoconjugates Topics in Current Chemistry 186, 1-20 (1997). 79. Hehre, E. J., Okada, G. & Genghof, D. S. Configurational specificity: unappreciated key to understanding enzymic reversions and de novo glycosidic bond synthesis. Archives of Biochemistry and Biophysics 135, 75-80 (1969). 80. Millqvist-Fureby, A., MacManus, D. A., Davies, S. & Vulfson, E. N. Enzymatic transformations in supersaturated substrate solutions: I, A general study with glycosidases. Biotechnology and Bioengineering 60, 197-203 (1998). 81. Ismail, A., Soultani, S. & Ghoul, M. Optimization of the enzymatic synthesis of butyl glucoside using response surface. Biotechnological Progress 14, 874-8 (1998). 82. Akkara, J. A., Ayyagari, M. S. R. & Bruno, F. F. Enzymatic synthesis and modification of polymers in nonaqueous solvents. Trends in Biotechnology 17, 67-73 (1999). 83. Rajnochova, E., Dvorakova, J., Hunkova, A. Z. & Kren, V. Reverse hydrolysis catalysed by P-A^ -acetylhexosaminidase from Aspergillus oryzae. Biotechnology Letters 19, 869-72(1997). 141 84. Ly, H. D. & Withers, S. G. Mutagenesis of glycosidases. Annual Review of Biochemistry 68, 487-522 (1999). 85. Mackenzie, L. F., Wang, Q., Warren, R. A. J. & Withers, S. G. Glycosynthases: mutant glycosidases for oligosaccharide synthesis. Journal of the American Chemical Society 120, 5583-4(1998). 86. Withers, S. G. Understanding and exploiting glycosidases. Canadian Journal of Chemistry 11, 1-11 (1999). 87. Wang, Q., Graham, R. W., Trimbur, d., Warren, R. A. J. & Withers, S. G. Changing enzymatic-reaction mechanisms by mutagensis - conversion of a retaining glucosidase to an inverting enzyme. Journal of American Chemical Society 116, 11594-5 (1994). 88. Mayer, C, Zechel, D. L., Reid, S. P., Warren, R. A. J. & Withers, S. G. The #358S mutant of agrobacterium sp. P-glucosidase is a greatly improved glycosynthase. FEBS Letter 466, 208-12 (2000). 89. Williams, S. J. & Withers, S. G. Glycosynthases: Mutant glycosidases for Glycoside Synthesis. Australian Journal of Chemistry 55, 3-12 (2002). 90. Walsh, C. Fluorinated substrate-analogs - routes of metabolism and selective toxicity. Advances in Enzymology and Related Areas of Molecular Biology 55, 197-289 (1983). 91. O'Hagan, D. & Rzepa, H. S. Some influences of fluorine in bioorganic chemistry. Chemical Communications, 645-652 (1997). 92. Harper, D. B. & O'Hagan, D. The fluorinated natural-products. Natural Product Reports 11, 123-133 (1994). 93. Shoda, S., Fujita, M. & Kobayashi, S. Glycanase-catalyzed synthesis of non-natural oligosaccharides. Trends in Glycoscience Glycotechnology 10, 279-89 (1998). 94. Howard, J. A. K., Hoy, V. J., O'Hagan, D. & Smith, G. T. How good is fluorine as a hydrogen bond acceptor? Tetrahedron 52, 12613-12622 (1996). 95. Slichter, C. P. Principles of Magnetic resonance (Springer-Verlag, New York, 1983). 96. Ernst, R. R., Bodenhausen, G. & Wakaun, A. Principles of Nuclear Magnetic Resonance in One and Two Dimensions (Clarendon Press, Oxford, 1987). 97. Harris, R. K. Nuclear Magnetic Resonance Spectroscopy: A Physicochemical View (Longmann Scientific and Technical, United Kingdom, 1987). 98. Sanders, J. Modern NMR Spectroscopy, A Guide for Chemists (Oxford university Press, 1987). 142 99. Derome, A. E. Modern NMR Techniques for Chemistry Research (Pergamon Press, New York, 1987). 100. Gunther, H. NMR Spectroscopy- Basic Principles, Concepts, and Applications in Chemistry (John Wiley and Sons, New York, 1995). 101. Levitt, M. H. Spin Dynamics. Basics of Nuclear Magnetic Resonance (Wiley, Chichester, 2001). 102. Harris, R. K. Nuclear Magnetic Resonance Spectroscopy (Longman Scientific & Technical, Essex, 1987). 103. Dybowsky, C. & Licter, R. L. NMR Spectroscopy Techniques (Marcel Dekker Inc., New York, 1987). 104. Freeman, F. A Handbook of Nuclear Magnetic Resonance (Longman Scientific & Technical, Essex, 1983). 105. Abragam, A. Principles of Nuclear Magnetism (Clarendon Press, Oxford, 1983). 106. Wong, G. Imaging Investigations of Swelling HPMC Tablets. MSc Thesis., Department of Chemistry., The University of British Columbia. Vancouver, B.C., Canada, 100 (2001). 107. Blazek, A. NMR Imaging Investigations of Swelling-Controlled Drug Delivery. PhD Thesis., Department of Chemistry., The University of British Columbia., Vancouver, B.C., Canada, 212 (1998). 108. Melhring, M. Pinciples of High Resolutiion NMR in Solids (Springer- Verlag, Berlin, 1983). 109. Fyfe, C. A. Solid State NMR for Chemists (C.F.C. Press, Guelph, 1983). 110. Stejskal, E. O. & Memory, J. D. High Resolution NMR in the Solid State, Fundamentals of CP MAS NMR (Oxford University Press, 1994). 111. Herzfeld, J. & Berger, A. E. Sideband intensities in NMR-spectra of samples spinning at the magic angle. Journal of Chemical Physics 73, 6021-6030 (1980). 112. Yannoni, C. S. High-resolution NMR in solids - the CPMAS experiment. Accounts of Chemical Research 15, 201-208 (1982). 113. Farrar, T. C. Concepts in Magnetic Resonance 2, 1 (1990). 114. Hartmann, S. R. & Hahn, E. L. Nuclear double resonance in rotating frame. Phys. Rev. 128, 2042(1962). 143 115. Junnemann, J., Thiem, J. & Pedersen, C. Facile synthesis of acetylated glycosyl fluorides derived from disaccharides and trisaccharides. Carbohydrate Research 249, 91-94(1993). 116. Filippo, J. S. & Romano, L. J. Mechanism of reaction of alkyl bromides and iodides with mercury(II) and silver(I) fluorides. Journal of Organic Chemistry 40, 782-787 (1975). 117. Nash, S. A. & Gammill, R. B. The synthesis of 2-fluorokhellin and 3-fluorokhellin. Tetrahedron Letters 28, 4003-4006 (1987). 118. Card, P. J. & Reddy, G. S. Fluorinated carbohydrates .2. Selective fluorination of glucopyranosides and mannopyranosides - use of 2-D NMR for structural assignments. J. Org. Chem. 48, 4734-43 (1983). 119. Mueller, S. C. & Brown, R. M. Evidence for an intermembrane component associated with a cellulose microfibril-synthesizing complex in higher-plants. Journal of Cell Biology 84, 315-326 (1980). 120. Atalla, R. FL, Gast, J. C, Sindorf, D. W., Bartuska, V. J. & Maciel, G. E. C-l3 Nmr-Spectra of Cellulose Polymorphs. Journal of the American Chemical Society 102, 3249-3251 (1980). 121. Shakashiri, B. Z. Chemical Demonstrations: A Handbook for Teachers of Chemistry, 247 (University of Wisconsin: Madison, 1983). 122. Knopp, M. A. Rayon from dryer lint: A demonstration.. Journal of Chemical Education 74,401 (1997). 123. Nakamura, I., Sugiyama, J., Ohmae, M., Kobayashi, S. & Kimura, S. Polym. Prepr. Jpn. 53, 5280 (2004). 124. Ando, S., Harris, R. K., Holstein, P., Reinsberg, S. A. & Yamauchi, K. Solid-state lH-static, 1H-MAS, and 1H->19F/19F->1H CP/MAS NMR study of polyvinyl fluoride). Polymer 42, 8137-8151 (2001). 125. Brouwer, D. H. Location, Disorder, and Dynamics of Guest Species in Zeolite Frameworks Studied by Solid State NMR and X-ray Diffraction. PhD Thesis., Department of Chemistry., The University of British Columbia., Vancouver, B.C., Canada (2003). 126. SAINT. (Madison, Wisconsin, USA, 1999). 127. SADABS. (Madison, Wisconsin, USA.). 128. Altomare, A., Cascarano, M., Giacovazzo, C. & Guagliardi, A. SIR92:. J. Appl. Cryst. 26, 343 (1994). 144 129. Cromer, D. T. & Waber, J. T. (The Kynoch Press, Birmingham, England, 1974). 130. Ibers, J. A. & Hamilton, W. C. Acta Crystallogr. 17, 781 (1964). 131. Creagh, D. C. & McAuley, W. J. in International Tables for Crystallography (ed. Wilson, A. J. C.) 219-222 (Kluwer Academic Publishers, Table 4.2.6.8, Boston, 1992). 132. Creagh, D. C. & Hubbell, J. H. in International Tables for Crystallography (ed. Wilson, A. J. C.) 200-206 (Kluwer Academic Publishers, Table 4.2.4.3, Boston, 1992). 133. SHELXTL. (Madision, Wisconsin, USA, 1997). 134. . 135. Standard deviation of an observation of unit weight: [Sw(Fo2-Fc2)2/(No-Nv)] 1/2 where: No = number of observations Nv = number of variables. 145 A p p e n d i x I. X - R a y Crysta l Structure Data for a -Ce l lob iosy l F luor ide A . Crysta l Data Empirical Formula Formula Weight Crystal Color, Habit Crystal Dimensions Crystal System Lattice Type Lattice Parameters Space Group Z value Dcalc FfJOO p(MoKa) C12H21O10F 344.29 colourless, irregular 0.15X0.07X0.03 mm triclinic primitive a = 4.878(2) A b = 7.575(3) A c= 10.435(5) A a = 92.14(2)° P = 95.62(2) 0 y= 104.72(2)° V = 370.4(3) A 3 P 1 (#1) 1 1.544 g/cm3 182.00 1.42 cm"1 146 B. Intensity Measurements Diffractometer Radiation Data Images Detector Position 26max No. of Reflections Measured Corrections Bruker X8 APEX MoKcx (A. = 0.71073 A) graphite monochromated 2639 exposures @ 20.0 seconds 38.01 mm 50.1« Total: 7929 Unique: 2441 (Rmt = 0.040;Friedels not merged) Absorption (Tmin = 0.840, Tmax= 0.996) Lorentz-polarization 147 C. Structure Solution and Refinement Structure Solution Refinement Function Minimized Least Squares Weights Anomalous Dispersion No. Observations (I>0.00a(I)) No. Variables Reflection/Parameter Ratio Residuals (refined on F 2, all data): Goodness of Fit Indicator No. Observations (I>2.00a(I)) Residuals (refined on F): Rl; wR2 Max Shift/Error in Final Cycle Maximum peak in Final Diff. Map Minimum peak in Final Diff. Map Direct Methods (SIR97) Full-matrix least-squares on F 2 £ w (Fo2 - Fc2)2 w=l/(a2(Fo2)+(0.0515P) 2+0.0360P) All non-hydrogen atoms 2441 236 10.34 Rl;wR2 0.069; 0.097 1.03 1867 0.043; 0.089 0.00 0.18 e7A3 -0.22 e7A3 . I Table 1 Atomic coordinates (xlO4) and equivalent isotropic displacement parameters (A 2 x 103) for a-CBF. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor. x y z U(eq) C ( l ) - 2 1 5 6 (8) - 6 9 6 9 ( 5 ) 5496 (4) 27 (1) C ( l ' ) 2505 (8) - 2 2 9 5 (5) 9465 (3) 20 (1) C ( 2 ' ) 2906 (8) - 3 2 7 (5) 9959 (3) 2 1 ( 1 ) C ( 2 ) - 9 8 2 (8) - 7 5 0 9 (5) 6789 (3) 2 3 ( 1 ) C ( 3 ) - 6 5 6 (8) - 6 0 4 7 (5) 7884 (3) 21 (1) C ( 3 ' ) 4953 (8) 68 (5) 1 1 2 0 8 ( 3 ) 21 (1) C ( 4 ) 816 (8) - 4 1 6 6 (5) 7464 (3) 2 1 ( 1 ) C ( 4 ' ) 3967 (8) - 1 3 1 1 (4) 12205 (3) 19 (1 ) C ( 5 ) - 5 8 4 (8) - 3 7 7 9 (5) 6163 (3) 21 (1) C ( 5 ' ) 3277 (8) - 3 2 7 4 (5) 11602 (3) 20 (1) C ( 6 ' ) 1836 (9) - 4 6 6 2 (5) 12484 (3) 2 5 ( 1 ) C ( 6 ) 950 (8) - 2 0 1 6 ( 5 ) 5634 (3) 2 3 ( 1 ) 0 ( 1 ) - 6 7 8 ( 5 ) - 5 2 3 5 (3) 5206 (2) 24 (1) O ( l ' ) 1297 (5) - 3 4 4 4 (3) 1 0 4 4 6 ( 2 ) 21 (1) 0 ( 2 ' ) 4023 (6) 8 8 6 ( 4 ) 9002 (2) 2 6 ( 1 ) 0 ( 2 ) - 2 6 4 5 ( 6 ) - 9 2 4 5 (4) 7101 (3) 30 (1) 0 ( 3 ) 8 9 6 ( 6 ) - 6 6 0 1 (3) 8954 (3) 2 9 ( 1 ) 0 ( 3 ' ) 5254 (6) 1896 (3) 11730 (2) 27 (1) 0 ( 4 ) 556 (5) - 2 7 0 3 (3) 8351 (2) 2 1 ( 1 ) 0 ( 4 ' ) 6218 (5) - 1 0 5 6 (3) 13235 (2) 2 7 ( 1 ) 0 ( 6 ) 3884 (5) - 1 9 6 9 (3) 5499 (2) 25 (1) 0 ( 6 ' ) 901 (6 ) - 6 4 6 9 (3) 11850 (2) 27 (1) F ( l ) - 5 0 4 8 (4) - 6 9 9 3 (3) 5559 (2) 3 9 ( 1 ) 149 App. I Table 2 Bond lengths [A] and angles [deg] for a-CBF. C(l) -0(1) 1. 389(5) c ( l ) - F ( l ) 1. 414 (4) C(l) -C(2) 1. 524(5) C(l) -H(1A) 1. 0000 C ( l ' )-0(4) 1 . 399 (4) C ( l ' )-0(l') 1. 441(4) C ( l ' )-C(2') 1. 517(5) C ( l ' )-H(lB) 1. 0000 C (2 ' )-0(2') 1 . 435(4) C (2 ' )-C(3') 1. 532(5) C (2 ' )-H(2B) 1. 0000 C (2) -0(2) 1. 427(5) C (2) -C(3) 1 529(5) C (2) -H(2A) 1 0000 C (3) -0(3) 1 424 (4) C (3) -C(4) 1 525(5) C(3) -H(3A) 1 0000 C (3 ' )-0(3') 1 436(4) C(3' )-C(4') 1 526 (5) C (3' )-H(3B) 1 0000 C (4) -0(4) 1 456(4) C (4) -C(5) 1 531(5) C (4) -H(4A) 1 0000 C (4 )-0(4') 1 430 (4) C (4 )-C(5') 1 534(5) C (4 )-H(4B) 1 0000 C (5) -0(1) 1 449(4) C (5) -C(6) 1 509(5) C (5) -H(5A) 1 0000 C (5 )-0(l') 1 450 (4) C (5 )-C(6') 1 507 (5) C (5 )-H(5B) 1 0000 C (6 )-0(6') 1 441(4) C (6 )-H(6C) 0 9900 C (6 )-H(6D) 0 9900 C (6 -0(6) 1 443 (5) C (6 -H(6A) 0 9900 C (6 -H(6B) 0 9900 0 (2 )-H(2') 0 .83 (5) 0(2 -H(2) 0 .88 (5) 0(3 -H(3) 0 .71 (4) 0 (3 )-H(3') 1 .01 (6) 0(4 )-H(4') 0 .90(4) 0(6 -H(6) 0 .93(5) 0(6 )-H(6') 0 .88(4) 0(1 - C d ) - F ( l ) 108 .9(3) 0(1 -C(l)-C(2) 112 .1 (3) F ( l -C(l)-C(2) 108 .2 (3) 0(1 -C(1)-H(1A) 109 .2 F ( l -C(1)-H(1A) 109 .2 C (2 -C(1)-H(1A) 109 .2 0(4 -Cd')-O(l') 107 .8(3) 150 O(4) -C(1')-C (2 ' ) 0 (1 1 ) -C(1')-C (2 ' ) 0 (4) -C (1' ) -H (IB) 0 (1 ' ) -C (1 ' ) -H(1B) C (2 ' ) -C (1 ' ) -H (IB) 0(2')-C(2 ,)-C(l'-) O (2 ' ) -C (2 ' ) -C (3 ' ) C(l')-C(2')-C(3') 0(2')-C(2')-H(2B) C(1')-C(2')-H(2B) C (3')-C(2')-H(2B) 0 (2) -C (2) -C (1) 0 (2) -C (2) -C (3) C (1) -C (2) -C (3) 0(2)-C(2)-H(2A) C(1)-C(2)-H(2A) C (3) -C (2) -H (2A) O (3) -C (3) -C (4) 0 (3) -C (3) -C (2) C (4) -C (3) -C (2) 0(3)-C(3)-H(3A) C (4) -C (3) -H (3A) C (2) -C (3) -H (3A) 0(3')-C(3')-C(4') 0 (3 ' ) -C (3 ' ) -C (2 ' ) C (4 ' ) -C (3 ' ) -C (2 ' ) 0(3 ' ) -C(3')-H(3B) C(4')-C(3•)-H(3B) C(2')-C(3')-H(3B) O (4) -C (4) -C (3) 0 (4) -C (4) -C (5) C (3) -C (4) -C (5) 0 (4) -C (4) -H (4A) C (3) -C (4) -H (4A) C(5)-C(4)-H(4A) 0 (4 ' ) -C (4 ' ) -C (3 ' ) 0(4')-C(4')-C(5') C (3' ) -C (4 ' ) -C (5 ' ) 0(4 ' ) -C(4')-H(4B) C(3 ' )-C(4')-H(4B) C(5')-C(4')-H(4B) 0(1) -C(5)-C(6) 0(1) -C(5)-C (4) C (6) -C (5) -C (4) 0 (1) -C (5) -H (5A) C (6) -C (5) -H (5A) C (4) -C (5) -H (5A) 0(1')-C(5')-C(6') 0 (1 ' ) -C (5 ' ) -C (4 ' ) C(6')-C(5')-C(4') 0(1' )-C(5*)-H(5B) C (6 ' ) -C (5 ' ) -H (5B) C (4 ' ) -C (5 ' ) -H (5B) 0(6' )-C(6')-C (5 ' ) 0 (6 1 ) -C (6 1 ) -H (6C) C(5')-C(6')-H(6C) 0(6' )-C(6')-H(6D) 110.1 (3) 107.2 (2) 110 . 6 110.6 110 . 6 109.8(3) 110.3 (3) 109.3(3) 109.2 109.2 109.2 111.0 (3) 111.6(3) 113.0 (3) 106.9 106.9 106.9 113.2 (3) 106.1 (3) 110.4 (3) 109.0 109.0 109.0 110.4(3) 110.3 (3) 111.7 (3) 108 .1 108.1 108 .1 111.8(3) 104.3 (2) 111.6(3) 109.7 109.7 109.7 108.1(3) 109.5 (3) 110.7 (3) 10.9.5 109.5 109.5 106.8 (3) 109.9 (2) 114.1(3) 108.6 108.6 108 . 6 106.4 (3) 109.2(3) 112.3 (3) 109.6 109.6 109.6 111.7(3) 109.3 109.3 109.3 151 C(5') -C(6')-H(6D) 109.3 H (6C) -C(6')-H(6D) 108 .0 0(6) -C(6)-C(5) 111.6(3) 0(6)- C (6) -H(6A) 109.3 C(5)- C(6) -H(6A) 109.3 0(6) -C(6) -H(6B) 109.3 C(5)- C(6)-H(6B) 109.3 H (6A) -C (6) -H(6B) 108.0 C(D- 0(1)-C(5) 115.2 (2) C ( l ' ) -0(1')-C(5') 112.3(3) C(2') -0(2')-H(2') 101 (4) C(2)- 0(2)-H(2) 111(3) C(3) -0(3)-H(3) 116 (4) C(3') -0(3')-H(3') 110(3) C(l') -0(4)-C(4) 116.5(3) C(4') -0(4')-H(4') 108(3) C(6)- 0(6) -H(6) 105 (3) C(6') -0(6')-H(6') 110(3) Symmetry transformations used to generate equivalent atoms: 1 5 2 I Table 3 Anisotropic displacement parameters (A 2 x 103) for a-CBF. The anisotropic displacement factor exponent takes the form: -2 piA2 [ hA2 a*A2 U l 1 + ... + 2 h k a* b* U12 ] U l l U22 U33 U23 U13 U12 C(l) 25(2) 27 (2) 29 (2) -2 (2) 5(2) 4 (2) C(l') 25 (2) 19(2) 15 (2) 0 (2) 3 (2) 7 (2) C(2') 26 (2) 16 (2) 21 (2) 0 (2) 11 (2) 4 (2) C(2) 24 (2) 17(2) 25 (2) -6(2) 7 (2) 1 (2) C(3) 26(2) 19(2) 18 (2) -5(2) 6(2) 5 (2) C(3') 23 (2) 20 (2) 20 (2) -3(2) 5 (2) 3(2) C(4) 28 (2) 19 (2) 17 (2) -4 (2) 7 (2) 9(2) C(4') 24 (2) 14 (2) 19 (2) -2 (2) 4 (2) 2 (2) C(5) 24 (2) 23 (2) 17 (2) -5(2) 2 (2) 10 (2) C(5') 24 (2) 22 (2) 15 (2) 0 (2) 1 (2) 6(2) C(6') 34 (2) 17 (2) 20 (2) -6 (2) 2 (2) 4 (2) C(6) 33 (2) 22 (2) 17 (2) 2 (2) 4 (2) 12 (2) 0(1) 34 (2) 22 (2) 15 (1) -5(1) 5(1) 3(1) 0(1' ) 29(1) 19 (1) 14 (1) 0(1) 4 (1) 2(1) 0(2') 39 (2) 19(2) 24 (1) 7 (1) 14 (1) 9(1) 0(2) 41 (2) 18 (2) 28 (2) -6(1) 12 (1) 0(1) 0(3) 52 (2) 21 (2) 16(1) -2 (1) K D 15 (1) 0(3' ) 31 (2) 12 (1) 35 (2) -10 (1) 4 (1) 3(1) 0(4) 30 (2) 17 (1) 15 (1) -7 (1) 0(1) 7 (1) 0(4') 28 (2) 31 (2) 18 (1) -1 (1) -4 (1) 0(1) 0(6) 30 (2) 23 (2) 19(1) -4 (1) 3(1) 6(1) 0(6' ) 29 (2) 13 (2) 35 (2) -3(1) K D 2 (1) F(l) 25 (1) 45 (2) 41 (1) 0 (1) K D 1 (1) 153 App. I Table 4 Hydrogen coordinates ( x 104) and isotropic displacement parameters (A 2 x 103) for a-CBF. x y z U(eq) H(1A) - 2 0 3 5 - 7 8 8 1 4799 33 H U B ) 4372 - 2 5 1 6 9287 23 H (2B) 1021 - 1 4 3 10143 25 H(2A) 972 - 7 6 4 7 6687 27 H (3A) - 2 5 9 0 - 6 0 3 0 8117 25 H(3B) 6866 - 3 1 10995 26 H (4A) 2877 - 4 0 8 5 7404 25 H(4B) 2231 - 1 0 8 8 12547 23 H (5A) - 2 5 8 2 - 3 7 3 1 6265 25 H (5B) 5070 - 3 5 5 9 11376 24 H(6C) 3182 - 4 6 7 6 13256 30 H (6D) 172 - 4 3 0 2 12776 30 H (6A) - 3 7 - 1 8 8 5 4781 28 H(6B) 886 - 9 7 2 6219 28 H(2) - 3 4 8 0 (110) - 9 1 4 0 ( 7 0 ) 7800 (50) 66 (17 ) H ( 2 ' ) 3110 (110) 1 6 7 0 ( 8 0 ) 9060 (50) 61 (16) H(3) 7 0 0 ( 1 0 0 ) - 6 2 5 0 (70) 9570 (40) 48 (17) H ( 3 ' ) 7350 (130) 2540 (80) 1 1 9 6 0 ( 5 0 ) 8 1 ( 1 8 ) H ( 4 ' ) 5440 (90) - 1 3 0 0 (60) 13970 (40) 4 2 ( 1 3 ) H ( 6 ' ) 2320 (90) - 6 9 9 0 (60) 11900 (40) 2 8 ( 1 2 ) H(6) 4950 (110) - 9 3 0 (70) 6 0 0 0 ( 5 0 ) 6 5 ( 1 6 ) 154 App. I Table 5 Torsion angles [deg] for a-CBF. 0(4) -C (1 ' ) -C (2 ' ) -0 (2 ' ) -61 . 2 (4) 0(1' )-C(l')-C(2')-0(2') -178 . 1(3) 0(4) -C (1' ) -C (2 ' ) -C (3 ' ) 177 . 7 (3) 0(1' )-C (1 ' )-C (2 ' )-C (3 ' ) 60 . 8 (4) 0(1) -C (1) -C (2) -0 (2) 176. 4 (3) F(l) -C (1) -C (2) -0 (2) 56. 3(4) 0(1) -C (1) -C (2) -C (3) 50. 1(4) F(l) -C (1) -C (2) -C (3) -70 . 0(4) 0 (2) -C (2) -C (3)-0(3) 63 . 8(4) C(l) -C (2)-C(3)-0(3) -170 . 2 (3) 0(2) -C (2) -C (3) -C (4) -173. 2 (3) C(l) -C (2) -C (3) -C (4) -47 2 (4) 0(2 ' ) -C (2 ' ) -C (3' ) -0 (3 ' ) 62 1(4) C ( l ' ) -C (2 ' ) -C (3 ' ) -0 (3 ' ) -177 1(3) 0(2' ) -C(2')-C(3')-C (4') -174 8 (3) C ( l )-C(2•)-C(3')-C (4') -54 0 (4) 0(3) -C (3) -C (4) -0 (4) -74 9 (4) C (2) -C(3)-C(4)-0(4) 166 3(3) 0(3) -C (3) -C (4) -C (5) 168 7 (3) C (2) -C (3) -C (4) -C (5) 49 9 (4) 0(3 ) -C (3 ' ) -C (4 ' ) -0 (4 ' ) -67 2 (3) C (2 ) -C (3' ) -C (4 ' ) -0 (4 ' ) 169 7(3) 0(3 )-C(3')-C(4')-C(5') 172 9 (3) C (2 ) -C (3 ' ) -C (4 • ) -C (5' ) 49 8(4) 0(4) -C (4) -C (5) -0 (1) -175 3(3) C (3) -C(4)-C(5)-0(1) -54 5 (4) 0(4) -C (4) -C (5) -C (6) 64 7 (3) C (3) -C (4) -C (5) -C (6) -174 4 (3) 0(4 ) -C (4 ' ) -C (5 ' ) -0 (1' ) -171 8(3) C (3 ) -C (4 ' ) -C (5' ) -0 (1 ' ) -52 7 (4) 0(4 ) -C (4 ' ) -C (5' ) -C (6' ) 70 5(4) C (3 )-C(4')-C(5')-C(6') -170 4 (3) 0(1 ) -C(5')-C(6')-0 (6' ) 53 8 (4) C (4 )-C(5')-C(6')-0(6') 173 2(3) 0(1 -C(5)-C(6)-0(6) -64 8(3) C (4 -C (5)-C (6)-0(6) 56 9 (4) F ( l -C (1) -0 (1) -C (5) 63 1(3) C (2 -C (1) -0 (1) -C (5) -56 6 (4) C (6 -C (5) -0 (1) -C (1) -176 8(3) C (4 -C (5) -0 (1) -C (1) 59 0 (4) 0(4 -C(l')-0(1')-C(5') 173 .8(2) C (2 ) -C (1' ) -0 (1' ) -C (5 ' ) -67 .7(3) C (6 ')-C(5')-0(l')-C(l') -175 . 0 (3) C (4 ')-C(5')-0(l')-C(l') 63 .5(3) 0(1 ' ) -C(1')-0(4)-C (4) -88 • 9(3) C (2 ' ) -C (1' ) -0 (4) -C (4) 154 .5(3) C (3 -C(4)-0 (4)-C (1') 82 .3(4) C (5 -C(4)-0 (4)-C (1') -157 . 0 (3) 155 App. I Table 6 Hydrogen Bonds for a-CBF Donor H....Acceptor [ ARU ] D - H H...A D...A D -H. . .A 0(2) -- H(2) . • 0(2') [ 1445 01] 0 88 (5) 1 83 (5) 2 699(4) 167(5) 0(2') -- H(2') . . 0(3) [ 1565 01] 0 83(6) 1 90 (6) 2 724 (4) 172(6) 0(3) - - H(3) . • O(l') [ ] 0 71 (4) 2 22 (5) 2 758(4) 134 (5) 0(3) - - H(3) . . 0(6') [ ] 0 71 (4) 2 39(4) 3 020 (4) 150(5) 0(3') - - H(3') . . 0(6') [ 1665 01] 1 02 (6) 1 72 (6) 2 706 (4) 162(5) 0(4' ) -- H(4') . • 0(6) [ 1556 01] 0 89 (4) 1 87 (4) 2 758 (3) 176(4) 0(6) - - H(6) . . 0(2) [ 1665 01] 0 93 (5) 1 79(5) 2 708 (4) 169 (4) 0(6') -•- H(6" ) . • 0(3') [ 1545 01] 0 88 (5) 1 86(5) 2 .727 (4) 171 (4) T r a n s l a t i o n of ARU-code to E q u i v a l e n t P o s i t i o n Code [ 1445. ] = -1+x,-1+y,z [ 1545. ] = x,-1+y,z [ 1665. ] = 1+x,1+y,z [ 1565. ] = x,1+y,z [ 1556. ] = x,y,1+z [ 1655. ] = 1+x,y,z 156 A p p e n d i x II. X -Ray Crysta l Structure Data for 6'-Deoxy-6'-Fluoro - a -Ce l lob iosy l F luor ide A . Crysta l Data Empirical Formula Formula Weight Crystal Color, Habit Crystal Dimensions Crystal System Lattice Type Lattice Parameters Space Group Z value Dcalc F000 u(MoKa) C13H24F2O10 378.32 colourless, needle 0.50X0.10X0.03 mm orthorhombic primitive a = 5.110(l) A b= 16.172(4) A c= 19.614(5) A a = 90.0 0 (3 = 90.0 0 y = 90.0 0 V= 1620.9(7) A 3 P 2,2,2) (#19) 4 1.550 g/cm3 800.00 1.46 cm-1 157 B. Intensity Measurements Diffractometer Radiation Data Images Detector Position 26max No. of Reflections Measured Corrections Bruker X8 APEX MoKa (X = 0.71073 A) graphite monochromated 1284 exposures @ 40.0 seconds 38.04 mm 50.3° Total: 19246 Unique: 2862 (Rmt = 0.084) Absorption (Tmin = 0.796, Tmax= 0. Lorentz-polarization 158 C. Structure Solution and Refinement Structure Solution Refinement Function Minimized Least Squares Weights Anomalous Dispersion No. Observations (I>0.00a(I)) No. Variables Reflection/Parameter Ratio Residuals (refined on F 2 , all data): Goodness of Fit Indicator No. Observations (I>2.00a(I)) Residuals (refined on F): Rl; wR2 Max Shift/Error in Final Cycle Maximum peak in Final Diff. Map Minimum peak in Final Diff. Map Direct Methods (SIR92) Full-matrix least-squares on F 2 I w (Fo2 - Fc2)2 W=1/(CT2(FO2)+(0.0583P) 2+0.1080P) All non-hydrogen atoms 2862 255 11.22 Rl;wR2 0.087; 0.109 1.02 1977 0.043; 0.093 0.00 0.20 e"/A3 -0.22 e"/A3 159 App. II Table 1 Atomic coordinates ( x 104) and equivalent isotropic displacement parameters (A 2 x 103) for 6'F-a-CBF. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor. X y z U(eq) C(l') 804(6) 9652 (2) 2061 (2) 20 (1) C(l) 5250 (7) 11507 (2) 172 (2) 23(1) C(2') 1375 (7) 8862 (2) 2460 (2) 20 (1) C (2) 4068 (7) 11900 (2) 804 (2) 21 (1) C(3') -796(6) 8688 (2) 2980 (2) 19(1) C (3) 4093 (6) 11290 (2) 1397 (2) 19(1) C(4') -1451(6) 9445 (2) 3406 (2) 21(1) C (4) 2752 (6) 10489 (2) 1171 (2) 19(1) C (5) 4036 (7) 10140 (2) 525 (2) 20 (1) C(5' ) -1997 (6) 10182 (2) 2942 (2) 21 (1) C (6) 2538 (6) 9420 (2) 219(2) 26(1) C(6') -2637 (8) 10981 (2) 3300 (2) 31 (1) C (7) 6554 (9) 7683 (3) 751(2) 51 (1) 0(1) 4135 (4) 10761 (1) -5(1) 24 (1) O(l') 278 (4) 10328 (1) 2516(1) 23(1) 0(2') 1509 (5) 8170(2) 2009(1) 26(1) 0(2) 5245 (5) 12681 (1) 947 (2) 26(1) 0(3' ) -6(5) 8021 (2) 3414 (1) 23 (1) 0(3) 2812 (5) 11671 (1) 1966 (1) 26(1) 0(4') -3779(4) 9325 (2) 3805 (1) 26(1) 0(4) 3066 (4) 9865 (1) 1693 (1) 21 (1) 0(6) -115 (5) 9613 (2) 48 (1) 29(1) 0(7) 6387(6) 7882 (2) 1452 (2) 49(1) F(l) 7963 (4) 11396 (1) 308 (1) 30 (1) F(6' ) -560 (4) 11243 (1) 3720(1) 42 (1) 160 II Table 2 Bond lengths [A] and angles [deg] for 6'F-a-CBF. C (1 ' )-0(4) 1. 406(4) C (1' )-0(l') 1. 437 (4) C (1 ' )-C(2') 1. 527 (4) C (1 ' )-H(lB) 1. 0000 C(l) -0(1} 1. 378 (4) C(l) -F(l) 1. 424 (4) C(l) -C(2) 1 . 519 (5) C(l) -H(1A) 1. 0000 C (2 ' )-0(2') 1. 429(4) C (2 ' )-C(3') 1 . 533 (5) C (2 ' )-H(2B) 1 . 0000 C (2) -0(2) 1. 428(4) C (2) -C(3) 1. 525 (4) C (2) -H(2A) 1 0000 C (3 ' )-0(3') 1 431 (4) C (3 ' )-C(4') 1 519 (5) C (3 ' )-H(3B) 1 0000 C (3) -0(3) 1 433 (4) C (3) -C(4) 1 531 (4) C (3) -H(3A) 1 0000 C (4 ' )-0(4') 1 437 (4) C (4 ' )-C(5') 1 526(5) C (4 )-H(4B) 1 0000 C (4) -0(4) 1 447 (4) C (4) -C(5) 1 535(5) C (4) -H(4A) 1 0000 C (5) -0(1) 1 446(4) C (5) -C(6) 1 516(5) C (5) -H(5A) 1 0000 C (5 )-0(l') 1 451 (4) C (5 )-C(6') 1 508(5) C (5 )-H(5B) 1 0000 C (6) -0(6) 1 430 (4) C (6) -H(6B) 0 .9900 C (6 -H(6A) 0 .9900 C (6 )-F(6') 1 .408 (4) C (6 )-H(6C) 0 . 9900 C (6 )-H(6D) 0 .9900 C(7 -0(7) 1 .414 (5) C(7 -H(7A) 0 . 9800 C(7 -H(7B) 0 . 9800 C(7 -H(7C) 0 .9800 0(2 )-H(2') 0 .77 (6) 0 (2 -H(2) 0 .78 (4) 0(3 •)-H(3') 0 .81 (3) 0(3 -H(3) 0 .87 (4) 0(4 ')-H(4') 0 .88(3) 0(6 -H(6) 0 .86 (5) 0(7 -H(7) 0 • 71 (6) 0(4 -Cd')-O(l') 106 .6(2) 0(4 )-C(1 1)-C (2 1 ) 108 • 1 (3) 0(1 ')-C(l')-C(2') 110 .7 (3) 161 0 (4) -C (1 ' ) -H (IB) 0 (1 ' ) -C(1')-H(IB) C(2')-C(1')-H (IB) 0(1)-C(1)-F(1) 0 (1) -C (1) -C (2) F (1) -C (1) -C (2) 0 (1) -C (1) -H (IA) F (1) -C (1) -H (IA) C (2) -C (1) -H (IA) O (2 ' ) -C (2 ' ) -C (1' ) 0 (2 * ) -C (2 ' ) -C (3 1 ) C (1 ' ) -C(2')-C (3 ' ) 0(2')-C(2')-H(2B) C(1')-C(2')-H(2B) C(3')-C(2')-H(2B) 0 (2) -C (2) -C (1) 0 (2) -C (2) -C (3) C (1) -C (2) -C (3) 0(2)-C(2)-H(2A) C (1) -C (2) -H (2A) C (3) -C (2) -H (2A) 0(3')-C(3')-C(4') 0 (3 ' ) -C (3 ' ) -C (2 1 ) C (4 ' ) -C (3 ' ) -C (2 ' ) 0(3')-C(3')-H(3B) C(4')-C(3')-H(3B) C(2*)-C(3')-H(3B) 0(3)-C (3)-C (2) 0 (3) -C (3) -C (4) C (2)-C (3)-C (4) 0(3)-C(3)-H(3A) C (2) -C (3) -H (3A) C (4) -C (3) -H (3A) 0 (4 ' ) -C (4 ' ) -C (3 ' ) 0 (4 ' ) -C (4 ' ) -C (5 ' ) C (3 ' ) -C (4 ' ) -C (5 ' ) 0(4 ' ) -C(4')-H(4B) C(3')-C(4')-H(4B) C(5')-C(4')-H(4B) 0 (4) -C (4) -C (3) 0 (4) -C (4) -C (5) C (3) -C (4) -C (5) 0(4)-C(4)-H(4A) C(3)-C(4)-H(4A) C(5)-C(4)-H(4A) 0(1)-C(5)-C(6) 0 (1) -C (5) -C (4) C (6) -C (5) -C (4) 0 (1) -C (5) -H (5A) C (6) -C (5) -H (5A) C (4) -C (5) -H (5A) 0(1')-C(5')-C(6') 0(1')-C(5')-C (4 ' ) C(6')-C(5')-C(4') 0 (1' ) -C (5 ' ) -H (5B) C(6')-C(5')-H(5B) C(4 ' ) -C(5')-H(5B) 110.4 110.4 110.4 109.9 (2) 114.0(3) 106.6 (3) 108 . 7 108.7 108.7 110.3 (3) 107.7 (3) 110.9 (3) 109.3 109.3 109.3 111.4(3) 114.7(3) 110.4(3) 106.6 106.6 106 . 6 110.0(3) 109.2 (2) 112.1(3) 108.5 108 . 5 108 . 5 108.2 (2) 112.6(3) 108.8(3) 109.1 109.1 109.1 111.9 (3) 106.2 (2) 110.0(3) 109.5 109.5 109.5 109.6(3) 106.2 (2) 111.0 (3) 110.0 110.0 110.0 105.4(3) 110.7 (2) 113.2 (3) 109.2 109.2 109.2 107.6 (3) 108.9 (2) 115.6(3) 108.2 108.2 108.2 162 0 (6) -C (6) -C (5) 113. 8 (3) 0 (6) -C (6) -H (6B) 108. 8 C (5) -C (6) -H (6B) 108. 8 0 (6) -C (6) -H (6A) 108 . 8 C (5) -C (6) -H (6A) 108 8 H (6B) -C (6) -H (6A) 107 7 F(6')-C(6')-C(5') 111 5(3) F(6' )-C(6')-H(6C) 109 3 C(5')-C(6')-H(6C) 109 3 F(6')-C(6')-H(6D) 109 3 C(5 ' ) -C(6')-H(6D) 109 3 H(6C)-C(6')-H(6D) 108 0 0(7)-C(7)-H(7A) 109 5 0(7) -C(7) -H(7B) 109 5 H(7A)-C(7)-H(7B) 109 5 0(7)-C(7)-H(7C) 109 5 H(7A)-C(7)-H(7C) 109 5 H(7B)-C(7)-H(7C) 109 5 C(l)-0(1)-C(5) 116 2 (3) C(l')-0(1')-C(5') 112 5 (2) C(2')-0(2')-H (2 ' ) 118 (4) C (2) -0 (2) -H (2) 107(3) C (3 ' ) -0(3' ) -H (3 ' ) 108 (2) C(3)-0(3)-H(3) 107 (3) C(4')-0(4')-H(4 ' ) 111 (2) C(l')-0(4)-C(4) 116 .3 (2) C (6) -0 (6) -H (6) 105 (4) C(7)-0(7)-H(7) 114 (5) Symmetry transformations used to generate equivalent atoms: 163 App. II Table 3 Anisotropic displacement parameters (A 2 x 103) for 6'F-a-CBF. The anisotropic displacement factor exponent takes the form: -2 piA2 [ hA2 a*A2 U l 1 + ... + 2 h k a* b* U12 ] U l l U22 U33 U23 U13 U12 C ( l ' ) 11(2) 22 (2) 25(2) 1 (2) 3(2) -1(1) C(l) 7 (2) 24 (2) 37 (2) 3 (2) 0 (2) 0(1) C(2') 11 (2) 19(2) 31 (2) 0(2) 1 (2) 0 (1) C (2) 10 (2) 17 (2) 37 (2) 0(2) -1 (2) 2(1) C(3' ) 9(2) 20 (2) 28 (2) 2 (2) -1 (2) -5(1) C (3) 10 (2) 20 (2) 27 (2) 0 (2) 2 (2) 0(1) C(4') 7 (2) 27 (2) 28 (2) 3 (2) 7 (2) -2(1) C (4) 7 (2) 24 (2) 27 (2) 5(2) 1 (2) 2 (1) C (5) 12 (2) 25 (2) 22 (2) 4 (2) 2 (2) 0(1) C(5' ) 8 (2) 30 (2) 26(2) 2 (2) 4 (2) -2(1) C (6) 16(2) 28 (2) 32 (2) 0 (2) 2 (2) K D C(6') 32 (2) 24 (2) 36(3) -2 (2) 5(2) 0 (2) C(7) 49(3) 57 (3) 46(3) -13(2) -5(2) 20(2) 0(1) 18 (1) 26(1) 29(2) 2(1) 2(1) -4 (1) 0(1' ) 16(1) 24 (1) 28 (1) 0(1) 5(1) 0(1) 0(2') 16(2) 23(1) 38 (2) -5(1) 3(1) K D 0(2) 17 (2) 21 (1) 39(2) 2(1) -6(1) -1 (1) 0(3' ) 14 (1) 21 (1) 34 (2) 6(1) 0(1) - K D 0(3) 25 (2) 21 (1) 31 (2) -2(1) 5(1) - K D 0(4' ) 16(1) 25 (1) 36 (2) 5(1) 10 (1) -2(1) 0(4) 10 (1) 24 (1) 29(2) 7 (1) 5(1) 0(1) 0(6) 14 (1) 33 (2) 39(2) 3(1) -2(1) -4 (1) 0(7) 17 (2) 81 (2) 49(2) -25 (2) -1 (2) 4(1) F(l) 7 (1) 39(1) 45(1) 3(1) 2 (1) - K D F(6' ) 45 (2) 34 (1) 46(2) -10 (1) -16(1) -6(1) 164 App. II Table 4 Hydrogen coordinates ( x 104) and isotropic displacement parameters (A 2 x 103) for 6'F-a-CBF. U(eq) H(1B) -705 9565 1744 23 H(1A) 5055 11899 -219 27 H (2B) 3083 8922 2704 24 H(2A) 2188 12012 697 26 H (3B) -2405 8515 2727 22 H (3A) 5948 11166 1523 23 H(4B) 51 9576 3714 25 H (4A) 850 10593 1087 23 H (5A) 5857 9957 635 24 H (5B) -3501 10037 2638 26 H(6B) 3456 9234 -198 31 H (6A) 2549 8955 547 31 H (6C) -3011 11414 2956 37 H(6D) -4228 10905 3581 37 H(7A) 7294 8152 501 76 H(7B) 4801 7560 575 76 H(7C) 7682 7198 691 76 H(2) 6290 (80) 12600(20) 1228(18) 21 (12) H(3) 2110 (90) 11270(20) 2210 (20) 42(13) H(6) -20 (110) 9980 (30) -270 (30) 80(18) H(2' ) 2820 (120) 8080 (30) 1820 (30) 100 (20) H(3') -660 (70) 7600 (20) 3271 (17) 19(11) H(4 ' ) -3930 (70) 8810 (20) 3934 (17) 24(10) H(7) 7600 (120) 7890 (30) 1620 (30) 70 (20) 165 App. II Table 5 Torsion angles [deg] for 6'F-a-CBF. 0(4) -C (1 ' ) -C (2 ' ) -0 (2 ' ) -72 . 1 (3) 0(1' ) -C (1' ) -C (2 ' ) -0 (2 ' ) 171. 4 (2) 0(4) -C(l')-C(2')-C(3') 168 . 8(3) 0(1' ) -C (1' ) -C (2 ' ) -C (3 ' ) 52 . 3(3) 0(1) -C (1)-C (2)-0 (2) -178 . 1 (3) F(l) -C (1)-C(2)-0(2) 60 . 4 (3) 0(1) -C (1) -C (2) -C (3) 53. 2(4) F(l) -C (1) -C (2) -C (3) -68. 2(3) 0(2 ' ) -C (2 ' ) -C (3' ) -0 (3 ' ) 68 . 2 (3) C ( l ' ) -C (2 ' ) -C (3' ) -0 (3' ) -171. 1 (3) 0(2 ' )-C(2')-C(3')-C(4') -169. 6(3) C ( l ' )-C(2')-C(3')-C(4') -48 9(4) 0(2) -C (2) -C (3) -0 (3) 56 8(3) C(l) -C (2) -C (3) -0 (3) -176 4 (3) 0(2) -C (2) -C (3) -C (4) 179 5(3) C(l) -C(2)-C(3)-C (4) -53 7(3) 0(3' )-C(3')-C(4')-0(4') -68 5 (3) C (2 )-C(3')-C(4')-0 (4') 169 7 (3) 0(3 ) -C (3 ' ) -C (4 ' ) -C (5 ' ) 173 7 (2) C (2 )-C(3')-C(4')-C(5') 51 9(3) 0(3) -C (3) -C (4) -0 (4) -67 8(3) C (2) -C (3) -C (4) -0 (4) 172 3 (2) 0(3) -C (3) -C (4) -C (5) 175 2 (3) C (2) -C (3) -C (4) -C (5) 55 2(3) 0(4) -C (4)-C (5)-0(1) -172 5 (2) C (3) -C (4) -C (5) -0 (1) -53 3(3) 0(4) -C (4) -C (5) -C (6) 69 5(3) C (3) -C(4)-C(5)-C(6) -171 4 (3) 0(4 ) -C (4 ' ) -C (5 ' ) -0 (1 ' ) -179 2(3) C (3 ) -C (4 ' ) -C (5 ' ) -0 (1 ' ) -57 9(3) 0(4 ) -C(4')-C(5')-C (6') 59 6(4) C (3 ) -C (4 ' ) -C (5' ) -C (6' ) -179 1 (3) 0(1 -C (5)-C(6)-0(6) -63 6(4) C (4 -C (5)-C (6)-0(6) 57 5(4) 0(1 ) -C(5')-C(6')-F (6') -61 2 (4) C (4 )-C(5')-C(6')-F (6') 60 .8 (4) F ( l -C (1) -0 (1) -C (5) 66 .4(3) C (2 -C (1) -0 (1) -C (5) -53 .3(4) C (6 -C (5)-0 (1)-C (1) 175 • 4(3) C (4 -C (5) -0 (1) -C (1) 52 .7 (3) 0(4 -C(l')-0(1')-C(5') -178 .8 (2) C (2 ) -C (1' ) -0 (1' ) -C (5 ' ) -61 .4(3) C (6 ' ) -C(5 ' ) -0 (1' ) -C(1' ) -170 .0(3) C (4 ' ) -C(5 ' ) -0 (1' ) -C(1' ) 64 .0(3) 0(1 ' ) -C (1' ) -0 (4) -C (4) -73 .8(3) C (2 ' ) -C (1' ) -0 (4) -C (4) 167 .1(3) C (3 -C (4) -0 (4) -C (1' ) 109 .2(3) C (5 -C(4)-0(4)-C(l') -130 .7(3) 1 6 6 App. II Table 6 Hydrogen Bonds Donor — H...Acceptor [ ARU ] D - H H...A D...A D-H...A 0(2) — H(2) . . 0(3') [ 4655 01] 0 78 (4) 2 14 (4) 2 791 (4) 142 (3) 0(2') — H(2') . . 0(7) [ ] 0 78 (6) 1 99 (6) 2 761 (4) 173(6) 0(3) — H(3) . . 0(1' ) [ ] 0 88 (4) 1 89 (4) 2 749(3) 166 (4) 0(3) — H(3) . . 0(4) [ ] 0 88 (4) 2 54 (3) 2 973 (3) 111 (3) 0(3' ) ~ H(3'). . 0(3) [ 4545 01] 0 81 (3) 1 92 (3) 2 716 (3) 169 (3) 0(4') — H(4 ' ) . • 0(3') [ ] 0 87 (3) 2 59(3) 2 959 (3) 107 (3) 0(4') — H(4'). • 0(2) [ 4545 01] 0 87 (3) 1 96 (3) 2 805 (3) 162 (3) 0(6) — H(6) . . 0(1) [ ] 0 86(5) 2 52 (5) 2 859 (3) 104 (4) 0(6) -- H(6) . . 0(4') [ 2474 01] 0 86(5) 2 22 (6) 3 036 (4) 157(5) 0(7) — H(7) . . 0(2') [ 1655 01] 0 70 (6) 2 19(6) 2 874 (4) 167(6) T r a n s l a t i o n of ARU-code to E q u i v a l e n t P o s i t i o n Code [ 1655. ] = l+x,y,z [ 4655. ] = 1-x,1/2+y,1/2-z [ 4545. ] = -x,-1/2+y,1/2-z [ 2474. ] = -1/2-x,2-y,-1/2+z 167 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0060167/manifest

Comment

Related Items