Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Laser desorption ionization Fourier transform ion cyclotron resonance (L.D.I.-F.T.-I.C.R.) mass spectrometric… Lam, Zamas 1989

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

Item Metadata

Download

Media
831-UBC_1990_A1 L35.pdf [ 12.95MB ]
Metadata
JSON: 831-1.0059719.json
JSON-LD: 831-1.0059719-ld.json
RDF/XML (Pretty): 831-1.0059719-rdf.xml
RDF/JSON: 831-1.0059719-rdf.json
Turtle: 831-1.0059719-turtle.txt
N-Triples: 831-1.0059719-rdf-ntriples.txt
Original Record: 831-1.0059719-source.json
Full Text
831-1.0059719-fulltext.txt
Citation
831-1.0059719.ris

Full Text

LASER DESORPTION IONIZATION FOURIER TRANSFORM ION CYCLOTRON RESONANCE (L.D.I.-F.T.-I.C.R.) MASS SPECTROMETRIC STUDIES OF SOME UNDERIVATIZED BACTERIAL OLIGOSACCHARIDES by ZAMAS LAM B. Sc. (Hons.) Thames Polytechnic, London, England, 1984 M . Sc. University of British Columbia, Canada, 1987 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (DEPARTMENT of CHEMISTRY) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Nov 1989 © ZAMAS LAM, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada Date niK Aiov, /w DE-6 (2/88) - i i -ABSTRACT This thesis describes the novel application of laser desorption ionization Fourier transform ion cyclotron resonance (l.d.i.-F.t.-i.c.r.) mass spectrometry for determining the sequence of underivatized oligosaccharides. As a starting point, four model oligosaccharides obtained from bacteriophage degradation of the corresponding bacterial capsular polysaccharides were analyzed. The experiments progressed from a linear pentasaccharide (Klebsiella K44), to a linear pentasaccharide containing an acid-labile pyruvic acid (Klebsiella K3), to a linear tetrasaccharide containing an acetamido hexose and an acid-labile neuraminic acid (E. coli K9), and finally to a "3+1" branched tetrasaccharide containing an lactic acid ether and a base-labile 0-acetate (Klebsiella ¥22). The glycosidic bond and significant ring cleavage fragmentations observed in the mass spectra could be used for the sequencing of all three linear oligosaccharides. Furthermore, these unusual ring-cleaved fragments also provided some tentative indication on the positions of linkage on several monosaccharide residues. The fragments derived from these model oligosaccharides were used to propose a logical interpretation of the l.d.i.-F.t.-i.c.r. mass spectra of commercially available polysaccharide reported by others. Polysaccharides that have similar primary and tertiary structures gave similar mass spectra. This may be rationalized by the fact that fragmentation occurred at the sample layer or the selvedge. Consequently, the mass spectra reflected the primary and tertiary structures of the polysaccharides in the solid state. Finally, this l.d.i.-F.t.-i.c.r. technique was employed to sequence an unknown oligosaccharide obtained by anhydrous hydrogen fluoride degradation of the bacterial capsular polysaccharide. It was concluded that the unknown sample is composed of two tetrasaccharides with the same carbohydrate components, but with one structure substituted with serine and the other with threonine. Furthermore, indirect evidence suggests that the amino acids are linked to the carboxylic acid group of the hexuronic acid. However, two possible structures could be proposed for these oligosaccharides. - i i i -The experimental results showed that l.d.i.-F.t.-i.c.r. mass spectrometry is well suited for the analysis of underivatized oligosaccharides, where the abundant ring fragments observed, can be used for a tentative indication of positions of linkage information. Furthermore, the high resolution data can be used to confirm the presence of unusual sugars and substituents. - iv -TABLE OF CONTENTS Contents Pages ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vii LIST OF FIGURES viii LIST OF MASS SPECTRA x LIST OF ABBREVIATIONS xii ACKNOWLEDGMENTS xiii DELICATION xiv I INTRODUCTION 1 I. 1 IMMUNOLOGICAL IMPORTANCE OF BACTERIAL EXOPOLYSACCHARIDES 3 II CHEMICAL METHODS FOR STRUCTURAL ANALYSIS OF BACTERIAL CAPSULAR POLYSACCHARIDE 8 II. 1 ISOLATION AND PURIFICATION 9 11.2 TOTAL SUGAR ANALYSIS 10 11.2.1 HYDROLYSIS IN AQUEOUS MEDIUM 10 11.2.2 HYDROLYSIS IN NON-AQUEOUS MEDIUM 11 11.2.3 CHARACTERIZATION OF THE MONOSACCHARIDES 11 11.2.4 REDUCTION OF URONIC ACID FOR G.C. ANALYSIS 12 11.2.5 DETERMINATION OF ABSOLUTE CONFIGURATION 12 11.2.6 DETERMINATION OF ANOMERIC CONFIGURATION 13 11.3 METHYLATION ANALYSIS 13 11.4 ANALYSIS OF NON-CARBOHYDRATE SUBSTITUENTS 15 11.5 DEGRADATION OF POLYSACCHARIDES INTO SMALLER OLIGOSACCHARIDES 15 II.5.1 PARTIAL ACID HYDROLYSIS 15 - V -11.5.2 P-ELIMINATION 16 11.5.3 PERIODATE OXIDATION/SMITH DEGRADATION 16 11.5.4 BACTERIOPHAGE 17 III INSTRUMENTAL METHODS USED IN CARBOHYDRATE ANALYSIS 19 111.1 NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY 20 III. 1.1 CHEMICAL SHIFTS 20 III. 1.2 RELATIVE INTEGRAL INTENSITIES 21 III.l.3 COUPLING CONSTANTS 21 111.2 GAS CHROMATOGRAPHY 22 111.3 MASS SPECTROMETRY 23 IH.3.1 ELECTRON IONIZATION 26 III.3.1.1 E.I.-M.S. OF PERMETHYLATED ALDITOL ACETATES 29 111.3.2 CHEMICAL IONIZATION & DESORPTION CHEMICAL IONIZATION 30 111.3.3 FAST ATOM BOMBARDMENT & SECONDARY ION MASS SPECTROMETRY 31 111.3.4 LASER DESORPTION IONIZATION FOURIER TRANSFORM ION CYCLOTRON RESONANCE 35 111.3.4.1 LASER DESORPTION IONIZATION 36 111.3.4.2 FOURIER TRANSFORM ION CYCLOTRON RESONANCE 38 IV LASER DESORPTION IONIZATION FOURIER TRANSFORM ION CYCLOTRON RESONANCE MASS SPECTROMETRY OF SOME UNDERIVATIZED OLIGOSACCHARIDES DERIVED FROM SPECIFIC DEGRADATION OF BACTERIAL CAPSULAR POLYSACCHARIDES 42 IV.l INTRODUCTION 43 IV.2 EXPERIMENTAL PROCEDURES 44 IV.2a DATA ANALYSIS 45 IV.3 RESULTS AND DISCUSSIONS 47 IV.3a NOMENCLATURE 48 - vi -IV.3b DESORPTION MECHANISM 49 IV.3c FRAGMENTATION MECHANISM 49 IV.3d RING FRAGMENTS 52 IV.3.1 NEGATIVE-ION L.D.I.-F.T.-I.C.R. OF KLEBSIELLA K44 OLIGOSACCHARIDE 57 IV.3.2 POSITIVE- AND NEGATIVE-ION L.D.I.-F.T.-I.C.R. OF MALTOSE AND CELLOBIOSE 63 IV.3.3 POSITIVE-ION L.D.I.-F.T.-I.C.R. OF KLEBSIELLA K3 OLIGOSACCHARIDE 67 IV.3.4 NEGATIVE-ION L.D.I.-F.T.-I.C.R. OF KLEBSIELLA K3 OLIGOSACCHARIDE 73 IV.3.5 POSITIVE-ION L.D.I.-F.T.-I.C.R. OF E. coli K9 OLIGOSACCHARIDE 76 IV.3.6 NEGATIVE-ION L.D.I.-F.T.-I.C.R. OF E. coli K9 OLIGOSACCHARIDE 78 IV.3.7 POSITIVE-ION L.D.I.-F.T.-I.C.R. OF KLEBSIELLA K22 OLIGOSACCHARIDE 83 IV.3.8 NEGATIVE-ION L.D.I.-F.T.-I.C.R. OF KLEBSIELLA K22 OLIGOSACCHARIDE 87 IV.3.9 STRUCTURAL AND REACTION ASSIGNMENTS FOR SOME COMMERICAL POLYSACCHARIDES 94 IV.3.10 NEGATIVE-ION L.D.I.-F.T.-I.C.R. OF E. coli K49 OLIGOSACCHARIDE 106 IV.3.11 POSITIVE-ION L.D.I.-F.T.-I.C.R. OF E. coli K49 OLIGOSACCHARIDE 118 IV.4 GENERAL DISCUSSION 123 IV.5 CONCLUSION 127 IV.6 FUTURE WORK 128 V BIBLIOGRAPHY 130 - vii -LIST OF TABLES Tables Page IV. 1 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF KLEBSIELLA K44 DE-0-ACETYLATED OLIGOSACCHARIDE 62 IV.2a POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF MALTOSE 65 IV.2b NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF MALTOSE 65 IV.2c POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF CELLOBIOSE 66 IV.2d NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF CELLOBIOSE 66 IV.3 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF KLEBSIELLA K3 OLIGOSACCHARIDE 72 IV.4 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF KLEBSIELLA K3 OLIGOSACCHARIDE 75 IV.5 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF £. coli K9 OLIGOSACCHARIDE 77 IV. 6 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF E. coli K9 OLIGOSACCHARIDE 82 IV.7 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF KLEBSIELLA K22 OLIGOSACCHARIDE 86 IV.8 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF KLEBSIELLA K22 OLIGOSACCHARIDE 88 IV.9 SUMMARY OF WILKINS' ION SERIES A-R OF THE FORM [(162)N+X+K]+ FOR THE L.D.I.-F.T.-I.C.R. MASS SPECTRA OF SIX POLYHEXOSES 95 IV. 10 PROPOSED STRUCTURES AND LABELS OF THE WILKINS' ION SERIES A-R. FOLLOWING KOCHETKOV'S CONVENTION 105 IV. 11 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF E. coli K49 OLIGOSACCHARIDE 117 IV.12 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF £. coli K49 OLIGOSACCHARIDE 119 - viii -LIST OF FIGURES Figure Page I. 1 SCHEMATIC REPRESENTATIONS OF GRAM-POSITIVE AND GRAM-NEGATIVE CELL WALL 5 II. 1 SCHEMATIC DIAGRAM TO ILLUSTRATE BACTERIOPHAGE DEGRADATION OF A CAPSULAR POLYSACCHARIDE 18 III. 1 SOME FRAGMENTATION PATTERNS OF A PERMETHYLATED HEXOSE PROPOSED BY CHIZHOV AND KOCHETKOV 27 111.2 MORE FRAGMENTATION PATTERNS OF A PERMETHYLATED HEXOSE PROPOSED BY CHIZHOV AND KOCHETKOV 28 111.3 PREFERRED ORDER OF FRAGMENTATION FOR PARTIALLY METHYLATED ALDITOL ACETATES 29 111.4 FRAGMENTATION PATTERN OF l,5,6-TRI-0-ACETYL-2,3,4-TRI-0-METHYLGLUCITOL, AS PROPOSED BY LINDBERG ET AL 30 IV. 1 PROPOSED GLYCOSIDIC BOND CLEAVAGE VIA PROTONATION 51 IV.2 SOME RING CLEAVAGES GIVING -C 2H 30 2 FRAGMENTS 53 IV.3 SOME RING CLEAVAGES GIVING - C 3 H 5 O 3 FRAGMENTS 54 IV.4 RING CLEAVAGE GIVING - C 4 H 7 O 3 FRAGMENT 55 IV.5 STRUCTURE OF KLEBSIELLA SEROTYPE K44 DE-O-ACETYLATED OLIGOSACCHARIDE OBTAINED FROM BACTERIOPHAGE DEGRADATION OF THE CAPSULAR POLYSACCHARIDE 57 IV. 6 STRUCTURE OF THE NON-REDUCTING END RING FRAGMENTS OBTAINED FROM NEGATIVE-ION L.D.I.-F.T.-I.C.R. OF KLEBSIELLA K44 DE-O-ACETYLATED OLIGOSACCHARIDE 60 IV.7 STRUCTURE OF THE REDUCTTNG END RING FRAGMENTS OBTAINED FROM NEGATIVE-ION L.D.L-F.T.-I.C.R. OF KLEBSIELLA K44 DE-O-ACETYLATED OLIGOSACCHARIDE 61 IV.8 STRUCTURES OF THE DISACCHARIDES: MALTOSE AND CELLOBIOSE 63 IV.9 SOME RING CLEAVAGES GIVING - C 4 H 7 O 4 FRAGMENTS 64 IV.10 STRUCTURE OF KLEBSIELLA SEROTYPE K3 OLIGOSACCHARIDE OBTAINED FROM BACTERIOPHAGE DEGRADATION OF THE CAPSULAR POLYSACCAHARIDE 68 IV. 11 STRUCTURE OF THE FRAGMENT ION C4H704-(2Hex)-C4H703 70 IV.12 STRUCTURE OF THE FRAGMENT ION C3H403-(3Hex) 73 IV.13 DEPROTONATION OF C-3 HYDROXY HYDROGEN OF THE GALACTURONIC ACID AND SUBSEQUENT CLEAVAGES OF THE C-3.C-4 AND C-5.0-HEMIACETAL BOND 74 - ix -fV.14 STRUCTURE OF E. coli SEROTYPE K9 OLIGOSACCAHRIDE OBTAINED FROM BACTERIOPHAGE DEGRADATION OF THE CAPSULAR POLYSACCHARIDE 76 IV. 15 BOTH HexNAc AND HexN-C2H302 HAVE THE SAME ELEMENTAL FORMULA OF C 8 H i 5 0 6 N 78 IV.16 STRUCTURE OF THE FRAGMENT ION C3H4O3N-C4H7O3 79 IV. 17 FRAGMENTATION OF THE R-(l-»4)NeuNAc 80 IV. 18 STRUCTURE OF KLEBSIELLA SEROTYPE K22 OLIGOSACCHARIDE OBTAINED FROM BACTERIOPHAGE DEGRADATION OF THE CAPSULAR POLYSACCAHARIDE 83 IV. 19 DEPROTONATION OF A -C3H3O4 FRAGMENT 85 IV.20 A NON-REDUCING END PERMETHYLATED HEXOSE RESIDUE WILL CLEAVE TO GIVE AN OXONIUM ION FRAGMENT M/Z 219 90 IV.21 CLEAVAGE OF A 2-SUBSTTTUTED NON-REDUCING RESIDUE TO GIVE -C 2H 30 FRAGMENT CONTAINING A C-l.C-2 CENTRE 99 IV.22 CLEAVAGE OF A 3- AND 4-SUBSTITUTED REDUCING RESIDUE TO GIVE -C 2H 30 FRAGMENT CONTAINING A C-3.C-4 CENTRE 99 IV.23 PROPOSED REACTION PATHWAY FOR CELLULOSE AND CHTTTN TO GIVE -C3H5O2 FRAGMENT CONTAINING A C-4. C-5.C-6 CENTRE 100 IV.24 PROPOSED REACTION PATHWAY FOR DEXTRAN TO GIVE - C 3 H 5 O 3 FRAGMENT CONTAINING EITHER A C-l C-2.C-3 OR A C-4, C-5.C-6 CENTRE 101 IV.25 PROPOSED DOUBLE RING CLEAVAGES OF STARCH TO GIVE A -C 2H 30 AND A -C2H3O2 RING FRAGMENTS 102 IV.26 PROPOSED DOUBLE RING CLEAVAGES OF CELLULOSE TO GIVE TWO -C 2H 30 2 RING FRAGMENTS 103 IV.27 ANHYDROUS HYDROGEN FLUORIDE HYDROLYSIS (METHANOL QUENCHED) OF A CAPSULAR POLYSACCHARIDE 107 IV.28 POSSIBLE LINKAGES OF SERINE RESIDUE TO THE TETRASACCHARTDE 112 IV.29 STRUCTURE OF THE FRAGMENT ION (HexN,HexA)~C3H503 114 IV.30 STRUCTURE OF THE FRAGMENT ION (Hex.HexA,OMe)~C2H30NAc 118 IV.31 STRUCTURE OF THE FRAGMENT ION (HexfHexA>OMe)~CH02 118 - X -LIST OF MASS SPECTRA Spectra Page IV. 1 LOW RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF KLEBSIELLA K44 DE-O-ACETYLATED OLIGOSACCHARIDE 156 IV.2 HIGH RESOLUTION NEGATTVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF KLEBSIELLA K44 DE-O-ACETYLATED OLIGOSACCHARIDE 157 IV.3 LOW RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF MALTOSE 162 IV.4 HIGH RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF MALTOSE 163 IV.5 LOW RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF MALTOSE 164 IV.6 HIGH RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF MALTOSE 165 IV.7 LOW RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF CELLOBIOSE 166 IV.8 HIGH RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF CELLOBIOSE 167 IV.9 LOW RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF CELLOBIOSE 168 IV. 10 HIGH RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF CELLOBIOSE 169 IV. 11 LOW RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF KLEBSIELLA K3 OLIGOSACCHARIDE 170 IV. 12 HIGH RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF KLEBSIELLA K3 OLIGOSACCHARIDE 171 IV. 13 LOW RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF KLEBSIELLA K3 OLIGOSACCHARIDE 173 IV. 14 HIGH RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF KLEBSIELLA K3 OLIGOSACCHARIDE 175 IV. 15 LOW RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF E. coli K9 OLIGOSACCHARIDE 176 IV. 16 HIGH RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF E. coli K9 OLIGOSACCHARIDE 177 IV.17 LOW RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF E. coli K9 OLIGOSACCHARIDE 178 IV. 18 HIGH RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF E. coli K9 OLIGOSACCHARIDE 179 - xi -TVJ9 LOW RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF KLEBSIELLA K22 OLIGOSACCHARIDE 182 tV.20 HIGH RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF KLEBSIELLA K22 OLIGOSACCHARIDE 183 IV.21 LOW RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF KLEBSIELLA OLIGOSACCHARIDE 185 IV.22 HIGH RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF KLEBSIELLA K22 OLIGOSACCHARIDE 186 IV.23 LOW RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF E. coli K49 OLIGOSACCHARIDE 188 IV .24 HIGH RESOLUTION NEGATIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF E. coli K49 OLIGOSACCHARIDE 189 IV.25 LOW RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.CJI. SPECTRUM OF E. coli K49 OLIGOSACCHARIDE 190 IV.26 HIGH RESOLUTION POSITIVE-ION L.D.I.-F.T.-I.C.R. TABULATED SPECTRUM OF E. coli K49 OLIGOSACCHARIDE 191 - xii -LIST OF ABBREVIATIONS GPS capsular polysaccharide LPS lipopolysaccharide P M A A partially methylated alditol acetate Gal galactose Glc glucose Man mannose Rha rhamnose GalA galacturonic acid GalNAc 2-acetamido-2-deoxygalactose GlcA glucuronic acid GlcNAc 2-acetamido-2-deoxyglucose deoxyHex deoxyhexose Hex hexose HexA hexose uronic acid HexNAc acetamido hexose Kdo 2-keto-3-deoxy-manno-octulonsonic acid NeuNAc N-acetyl neuraminic acid Ac acetyl pyr pyruvic acid acetal g.c. gas chromatography m.s. mass spectrometry 252Cf-p.d. Californium-252 plasma desorption c i . chemical ionization d.c.i. desorption chemical ionization e.i. electron ionization f.a.b. fast atom bombardment f.d. field desorption F.t.-i.c.r. Fourier transform ion cyclotron resonance Ldi . laser desorption ionisation l.s.i.m.s. liquid secondary ion mass spectrometry t.o.f. time-of-flight amu atomic mass unit eV electron volt fx fragment ion m/z mass to charge ratio n.m.r. nuclear magnetic resonance - xiii -ACKNOWLEDGMENTS I would like to express my sincere gratitude to the following: Professors Guy G.S. Dutton and Melvin B. Comisarow under whose guidance this research was conducted, for their patient, inspiration and very enthusiastic discussion on various academic and non-academic topics during my course of studies. Professor Dutton in particular for his financial support for the various ICSs, IMSC and CIC conferences that he so actively encourages me to attend. Professor Comisarow for his financial support for the ASMS conference. Dr. Gtinter K. Eigendorf for his encouragment and the opportunity to learn some of the mass spectrometric techniques in his laboratory. And his staffs for the weekly Friday nutrients. To the numerous graduate students (Drs. Andrew V.S. Lim, Linda M . Beynon, Ms. Sandra Taylor), post-doc. (Dr. Neil Ravenscroft), visiting scientists (Drs. Haralambos Parolis, Lesley A.S. Parolis, Stephen K. Ng and Ms. Paola Cescutti) in the Department of Chemistry for their friendship and general assistance. In particular, to Linda, Neil, Paola and Sandra for their helpful discussions and in the perparation of this thesis. Drs. Asgeir Bjaranson (Science Institute, University of Iceland) and David A. Weil (Nicolet Analytical Instruments) for running the l.d.i.-F.t.-i.c.r. spectra. Nicolet Analytical Instruments for the loan of the FTMS-2000. The Department of Chemistry (U.B.C.) for their financial support in the form of a Research Assistantship and the Faculty of Graduate Studies (U.B.C.) for a Travel Award. To the friends in Vancouver (in particular B & B & B) for their kindness, comfort and encouragement that have made the past five years so extremely enjoyable. - xiv -Delicated with love and gratitude to: my parents LAM CHE HIM & CfflU KAI YIN my sister and niece SYLVIA & LAURA TO C H A P T E R I I N T R O D U C T I O N - 2 -INTRODUCTION Carbohydrates are a group of naturally occurring biological materials which have been known to man since prehistoric times and are still being used as both anabolic materials and as a readily available food source. Carbohydrates are composed of polyhydroxy aldehydes, ketones, alcohols, acids, or simple derivatives thereof. They can be linked into chains by acetal linkages to form oligo- or poly-saccharides. No sharp distinction is drawn between oligo- and poly-saccharides, for their structures are similar and only the molecular weights are different. The term "oligosaccharides" usually implies carbohydrates with two to nine monosaccharide residues in each molecule. Polysaccharides (PS) are defined as those carbohydrate polymers that contain periodically repeating structures in which the dominant inter-unit linkages are of the <9-glycosidic types. Polysaccharides containing only one type of monosaccharide unit are called homopolysaccharides or homoglycans, examples of which are cellulose and chitin. Polysaccharides containing different monosaccharide units are termed heteroglycans. Plants are composed of 50 to 80% (dry weight) of carbohydrates, found exclusively as polysaccharide polymers. Although the principal building material of higher animals is proteins, carbohydrates are often found associated with proteins (e.g., glycoproteins, proteoglycans and peptidoglycans), as well as with other materials to form glycolipids, lipopolysaccharides (LPS), teichoic and nucleic acids1. In some lower animals the major constituent of the exoskeleton is chitin, which is a polymer of N-acetylated glucosamine. Carbohydrates in nature can serve either a structural role or as a source of energy (e.g., glycogen and starch). Structural polysaccharides can be divided into two main classes: i) the fibrous polysaccharides, such as cellulose and chitin; or ii) the gel-forming matrix polysaccharides, such as pectin. Furthermore, complex carbohydrates are essential in biological recognition processes and protection. As early as 300 A.D., man exploited naturally occurring carbohydrates in the food industry. Nowadays, the thickening and gelling properties of carbohydrates are used to control the texture, flavor, appearance and color of food. Carbohydrates also play a major role - 3 -in the cosmetic, textile, paper and paint industries; they are recendy being utilized as drilling fluids in the petroleum industry2. Readily accessible mono- or di-saccharides can also be used as starting materials in the synthesis of complex chiral compounds3. Polysaccharides can usually be represented by an average or regularly repetitive structure. Plant polysaccharides are generally comprised of block polymers with no true sense of a repeating structure. One of the few plant homoglycans is cellulose, which is the polycondensation product of 1,4-linked p-D-glucopyranoses. Plant polysaccharides are mostly composed of heteroglycans of average block repeating units, therefore, are very difficult to characterize. On the other hand, a majority of microbial exopolysaccharides are composed of regular oligosaccharide repeating units. Thus, structural elucidation of these microbial oligosaccharides should lead to the complete structural characterization of the polysaccharides. The recent resurgence of interest in carbohydrates is largely due to a greater understanding of their role in biological cell-cell recognition processes. Furthermore, the discovery of the highly glycosylated protein coat of the HIV virus has encouraged a large number of "non-carbohydrate" chemists to try their hand in this oldest branch of biological chemistry4. 1.1 IMMUNOLOGICAL IMPORTANCE OF BACTERIAL EXOPOLYSACCHARIDES The bacterial cell wall encloses the plasma membrane of the bacterium. Lipopolysaccharide (LPS) which exhibits both immunogenicity and full endotoxicity is a main component of the cell wall (10 to 15% dry weight). This species-specific LPS contains the somatic polysaccharide antigen and is called the bacterial O antigen. In general, the outermost cell component of pathogenic microbes plays a critical role in the immune response. Many bacteria, both Gram-positive and Gram-negative, produce extracellular (exo-)polysaccharides either in the form of a discrete capsule surrounding the bacterial cell or in the form of loose slime unattached to the cell surface5. (Fig. 1.1 is a schematic representation of the Gram-positive and Gram-negative cell walls of bacteria.) - 4 -This capsular polysaccharide (CPS) is called the K antigen and constitutes the principal antigen of most pathogenic microorganisms. For many years these microbial capsular polysaccharides were considered to function simply as energy reserves, structural polymers and to have merely a limited role in protection against phagocytosis. It is now well established that they are also essential in biological recognition where they: i) act as highly specific receptors for bacterial viruses (bacteriophages) and bacteriocins; ii) are immunogenic; and iii) are specific surface antigens6. These capsular polysaccharides also mask the cell wall O antigens and thus interfere with their serological detection7. Furthermore they also protect the bacteria against desiccation and the action of complement. Nevertheless, because these capsular polysaccharides are immunogenic, they play an important part in the immune response to bacterial infection. Extensive reviews of the immunology of bacterial capsular polysaccharides have been published8,9. An animal's immunological defence system is based on its antibody molecules recognizing a portion of the structure of the bacterial polysaccharide antigens. These recognized partial structures are termed the immunodominant or immunodeterminant structures, and they may be monosaccharide units linked in a specific way or an oligosaccharide in a particular conformation or a non-carbohydrate substituent, e.g. pyruvic acid acetal, or N- and O- acetyls10. In acidic polysaccharides the immunodominant sugars are often the uronic acid residues. In the case of branched polysaccharides, the immunodeterminant is usually the side chain. Since capsular polysaccharides consist of regular oligosaccharide repeating units, the immunodominant structures are therefore expressed repetitively. Several bacterial polysaccharides may share the same antigenic determinant, and can be recognized by the same antibody regardless of their origin. These immunologically related polysaccharides are said to cross-react serologically. Cross-reactions can be used to provide information about the structure of many polysaccharides11"14. Capsular polysaccharide t&t: ZipopolysacehariAc (somatic 0 antigen) -Lipoprotein \ S\ S\ /\ /\ S\ S\ y\ y S 5 f » ^ A A A A / * ^ '^ S-* Pariphamie enzymes ; Peptiioglycan "Phospholipid, bilayer with various membrane protein*,, enzymes anil permease* Interior of c e n ' ^ C ^ ^ ^ . J i j ^ ' . - . ^ " . . • ? : . > ' ^ / : " ^ - . V , ' V l , ; < : • •• *=>•. • » : Cell envelope of the Gram-negative cell vail ure 1.1 Schematic representations of Gram-positive and Gram-negative cell wall. Gram staining, perhaps the most widely used reaction in microbiology, was discovered by H.C.J. Gram (1884). It consists of staining heat-fixed smears with an aqueous solution of a basic gentian violet dye, mordanting with aqueous iodine, and then dehydration with ethanol. Those organisms that retain the blue-black dye are termed Gram-positives, and those which decolorized are termed Gram-negatives. - 6 -Both dead and live "attenuated" bacteria and toxins are commonly used as vaccines for disease prevention. Many exopolysaccharides coupled to carrier proteins can also be used as vaccines8. The first example dates from the 1940s, when bacterial polysaccharides were used to protect against pneumococcal infections1 5. Since then, many microbial polysaccharides have been shown to have non-cytotoxic antitumor properties, and hence have been extensively studied in cancer research • . Due to the immunological and serological importance of microbial exopolysaccharides, our laboratory has been part of an international collaborative program aimed at determining the molecular structures of these polysaccharides. This includes the characterization of their primary chemical structures as well as their tertiary structures in the solid and solution states. This is mainly achieved by using classical carbohydrate chemistry, combined with nuclear magnetic resonance spectroscopy (n.m.r.) and mass spectrometry (m.s.). With the advances of scientific instruments, a number of researchers are attempting to elucidate carbohydrate structures solely with the use of these spectroscopic techniques. The obvious advantages of this instrumental approach are the reduction in analysis time and sample quantity. N.m.r. being a non-destructive technique is often favored for structural elucidation. Furthermore, there exists a large library of n.m.r. data which can be used as cross-references for unknown structures. The relatively large sample quantity needed (~ 10 to 20 mg) for n.m.r. analysis is sometimes prohibitory. On the other hand, mass spectrometry, although it is a destructive technique, often only requires sub-ng quantities. Therefore m.s. is becoming a more prominent tool in structural analysis, in addition to its more routine usage in molecular weight determination. Although the electron ionization (e.i.) mass spectra of alditol acetates and oligosaccharide alditols are well documented18,19, there is a paucity of data relating to underivatized oligosaccharides. The mass spectrometric investigations described in this and a previous dissertation20 are specifically related to the structural characterization of Klebsiella and E. coli capsular polysaccharide antigens. In the previous studies20, the preliminary investigation of some - 7 -desorption/ionization (desorption chemical ionization (d.c.i.), fast atom bombardment (f.a.b.) and laser desorption ionization Fourier transform ion cyclotron resonance (l.d.i.-F.t.-i.c.r.)) mass spectrometric techniques for the sequencing of both underivatized and derivatized oligosaccharides was presented. In this thesis, the application of l.d.i.-F.t.-i.c.r. mass spectrometry is extended to the sequencing and characterization of underivatized bacterial oligosaccharides obtained by specific chemical or enzymic degradation of bacterial capsular polysaccharides. As a starting point for the mass spectrometric studies, the application of l.d.i.-F.t.-i.c.r. to four underivatized model oligosaccharides is described in Section IV.3.1 to IV.3.8. These investigations served to illustrate the fragment ions formed using this technique and thus assisted in the analysis of the l.d.i.-F.t.-i.c.r. mass spectra of some commercial polysaccharides (Section IV.3.9). Finally the usefulness of this l.d.i.-F.t.-i.c.r. technique was evaluated by application to an unknown oligosaccharide which is described in Section rv.3.10 and IV.3.11. A l l the model oligosaccharides investigated were generated by specific degradation of the corresponding bacterial capsular polysaccharide. As capsular polysaccharides are comprised of a regular oligosaccharide repeating unit, the structural elucidation of the oligosaccharides may represent complete characterization of the native capsular polysaccharide. Although this thesis is concerned with the analysis of pure carbohydrate compounds, it must be emphasized that any m.s. methods elaborated on bacterial carbohydrate antigens are directly transferable and relevant to other biological molecules containing sugars, such as glycoproteins and glycolipids. - 8 -CHAPTER n CHEMICAL METHODS FOR STRUCTURAL ANALYSIS OF BACTERIAL CAPSULAR POLYSACCHARIDES - 9 -C H E M I C A L M E T H O D S F O R S T R U C T U R A L A N A L Y S I S O F P O L Y S A C C H A R I D E S The complete structural characterization of any carbohydrate material can be quite complex, due to the different permutations of sugar residues, glycosidic linkages, anomeric configurations and possible non-carbohydrate substituents that may be present. The normal analytical procedure involves: i) identification and quantitation of monosaccharide components; ii) analysis of non-carbohydrate substituents if present; and the determination of iii) absolute configuration; iv) anomeric configuration; v) ring size; vi) linkage position; vii) sugar sequence; and viii) tertiary structure. Although the present study did not involve traditional structural carbohydrate chemistry, this chapter will describe some of these more common, albeit complex procedures used for the structural elucidation of these polysaccharides in order to emphasize the usefulness and relative simplicity of the instrumental approaches. II. 1 I S O L A T I O N A N D P U R I F I C A T I O N The heterogeneity of capsular polysaccharide preparations arises from the other cell components. As capsular polysaccharides are soluble in water, the first step in its purification involves removal of non water-soluble components by high speed centrifugation. Extraction using dilute alkali is not normally used as the possibility exists for base-catalyzed degradation. Similarly, utilization of dilute acid increases the risk of hydrolysis of glycosidic linkages during isolation. LPS and nucleic acids can be separated from the capsular polysaccharides by liquid phenol precipitation from aqueous solvent21. In addition, LPS exists as micelles in aqueous solution and, therefore, can be removed during the centrifugation step. The polysaccharide in the supernatant can be separated from any proteins and nucleic acids present by addition of the supernatant to ethanol22. Because capsular polysaccharides are usually acidic, polysaccharides can be further purified by precipitation with potassium chloride23, cupric acetate24 or cetyltrimethylammonium bromide (Cetavlon)25. - 10 -Purification by chromatographic techniques, such as gel filtration26 or molecular sieve oo chromatography on cross-linked dextrans are usually employed when dealing with OH. milligram quantities of polysaccharide. Ion-exchange chromatography is used for acidic 00 polysaccharides, while affinity chromatography is employed extensively for the purification of N-glycosidic glycopeptides . The absence of heterogeneity, rather than the presence of homogeneity is used to established the purity of the polysaccharide. Purity is obtained where the isolate attains consistent chemical composition (such as sugar residues, analysis of particular functional groups and spectroscopic data) and the physical properties (optical rotation, gel chromatography31 and electrophoresis32,33) of the isolate. II.2 TOTAL SUGAR ANALYSIS The initial step of structural characterization involves the total hydrolysis of polysaccharide into its component monosaccharides. As all sugars are to some extent degraded by acid, the experimental conditions must be chosen with care. The choice of reagent depends very much on the nature of the sugars, the type of glycosidic bonds present and the general stability of the individual monosaccharide units to acid . Therefore a series of experiments with different hydrolytic conditions have to be used to establish the true composition of the polysaccharide. II.2.1 HYDROLYSIS IN AQUEOUS MEDIUM Hydrolysis may be carried out successfully using hydrochloric and sulphuric acids. Although hydrochloric acid is more easily removed than sulphuric acid, it usually causes greater degradation35. Therefore the volatile and easily removed trifluoroacetic acid has largely replaced mineral acids for aqueous hydrolysis . For 2-acetamido-2-deoxy sugars, preliminary hydrolysis with acetic acid is usually carried out to avoid N-deacetylation37. Acetamido sugars can then be hydrolyzed with anhydrous hydrogen fluoride 3 8 , 3 9. -11 -Although neutral sugars can be completely hydrolyzed with minimum degradation with 2 M trifluoroacetic acid at 100°C for 6-8 hours, other more acid labile saccharides, such as deoxysugars, ketoses and sialic acids are largely decomposed under these conditions, while 2-amino-2-deoxy sugars and uronic acids are incompletely hydrolyzed. Consequently, different acid strengths and exposure times are required in order to measure the exact number of monosaccharide residues in the oligosaccharide repeating unit. 11.2.2 HYDROLYSIS IN NON-AQUEOUS MEDIUM Methanolysis with methanolic hydrogen chloride leads to the formation of methyl glycoside methyl esters and methyl glycosides, which stabilize the normally acid labile sugars, and thus is less destructive to deoxysugars and sialic acids 4 0 , 4 1. In the case of polysaccharides containing uronic acid residues, methanolysis of the glycosidic bonds induces esterification of the uronic acid residue, which can then be reduced to its corresponding primary alcohol with sodium borodeuteride42. The mixture of methyl glycosides is hydrolyzed and converted into alditol acetate derivatives to be analyzed by gas chromatography (g.c). Although some polysaccharides are insoluble in aqueous solution (e.g. cellulose), the acetylated derivatives are readily soluble in acetic anhydride-acetic acid mixture. Acetolysis of the peracetylated material with a catalytic amount of sulphuric acid has the advantage of cleaving l->6 linkages, which are more resistant than the other linkages to acid hydrolysis. Sialic acid linkages are usually very stable to the above process, leading to the isolation of sialic acid containing oligomers . 11.2.3 CHARACTERIZATION OF T H E MONOSACCHARIDES Conventional qualitative analytical techniques involve paper chromatography44,45, thin-layer chromatography46 and paper electrophoresis47. Spectrometric analysis permits classification into broad groups 4 8 , 4 9. Gas chromatography34,50, high performance liquid chromatography51 (h.p.l.c.) and m.s. are increasingly popular for the quantification of - 12 -monosaccharides. G.c. methods require the sugar to be derivatized to impart volatility, whereas h.p.l.c. can be performed on the underivatized, hydrolyzed monosaccharides or oligosaccharides. 11.2.4 REDUCTION OF URONIC ACID FOR G.C. ANALYSIS Polysaccharides containing uronic acid residues are difficult to analyze for the following reasons: i) the carboxylic function cannot be easily analyzed by g.c. or h.p.l.c; ii) the glycosidic bond of the uronic acid is resistant to acid hydrolysis; and iii) incomplete periodate oxidation results from electrostatic repulsion between the periodate and the carboxylate ion (discussed in Section II.5.3). The best approach to use in the investigation of acidic polysaccharides involves reduction of the carboxylic function to the primary alcohol, though the pH of the reaction medium has to be rigidly controlled. Normally an aqueous solution of the acidic polysaccharide is treated with a water-soluble carbodhrnide to give 0-acylisourea, which is then reduced with sodium borohydride52. This procedure, unlike methanolysis, leaves all glycosidic linkages intact. The uronic acid residues can also be reduced, after methanolysis, to the methyl glycoside methyl ester with sodium borodeuteride42. Another approach involves reduction of the permethylated polysaccharide, usually with lithium aluminum deuteride in tetrahydrofuran . The reduction of the uronic acid using the deuteride approach, has the added advantage of labelling the acidic sugar which can be identified by the different g.c. retention time, as well as the increase in mass observable in the mass spectrum. 11.2.5 DETERMINATION OF ABSOLUTE CONFIGURATION Routine use of chromatographic and spectrometric methods cannot distinguish between enantiomers, therefore, polarimetric measurements are employed: i) optical rotation [a]o'» and ii) circular dichroism measurement of the alditol acetates or the partially methylated alditol acetates54. The use of optically active alcohols (e.g., (-)-2-butanol55, (+)-2-octanol56) to convert the enantiomers to diastereomers, followed by g.c. separation of the volatile - 13 -derivatives, such as acetates or trimethylsilyl ethers can be achieved in milligram quantities. Enzymatic assay can presently only be used for a few sugars (e.g., D-glucose oxidase57 and D-galactose oxidase58.) II.2.6 DETERMINATION OF ANOMERIC CONFIGURATION The monosaccharides can be either a- or B- linked in the polysaccharide. The assignment of the anomeric configurations to specific linkages can be achieved by: i) the use of specific exoglycanase enzymes (e.g., a- or B- glucosidase)59; ii) measurement of the coupling constants and chemical shifts in the n.m.r. spectra (discussed in Section HI. 1.1); iii) the use of chromium dioxide oxidation on the peracetylated material, where the B-linked saccharides are preferentially degraded60. II.3 METHYLATION ANALYSIS The second step in the characterization of polysaccharides involves the determination of the positions of substitution of each individual monosaccharide residue. This is accomplished by methylation analysis, where the free hydroxyl groups of the polysaccharide are protected by etherification (methylation), followed by hydrolysis, reduction and acetylation. The partially methylated alditol acetate (PMAA) derivatives obtained are then separated by g.c. and identified by g.c.-m.s. The information obtained can also be used to plan appropriate degradation methods for the polysaccharide into small oligosaccharide fragments. Nevertheless, methylation analysis does not provide information on the sequence or the anomeric linkages. The most commonly used methylation procedure was developed by Hakomori 6 1. Methylation can be achieved by treating the polysaccharides with the strong base dimethyl methylsulphinyl methanide (sodium dimsyl, formed by the reaction of sodium hydride with (limethyl sulphoxide) and followed by addition of methyl iodide. Hakomori methylation leads to esterification of uronic acids, pyruvate residues and N-methylation of acetamide groups in amino sugars. - 14 -Dimsyl can also be generated using potassium rerf-butoxide62, potassium hydride63 and butyllithium64. These reagents generally produce dimsyl as efficiendy as sodium hydride, and are more rapid, convenient and safer. These methods also result in less interfering peaks in g.c.-m.s. analyses of methylated polysaccharides. The latest methylation procedure involves the use of methyl iodide, a solid base (sodium hydroxide) and DMSO 6 5 . Excellent yields of permefhylated products can be obtained in a very short reaction time, and non-carbohydrate peaks are not observed in the chromatogram. The use of a strong base during methylation causes removal of base-labile substituents (e.g., acetates), resulting in loss of substitution information. Nevertheless, O-acyl groups can be detected using the Prehm6 6 methylation procedure, where the polysaccharide is dissolved in trimethyl phosphate and then methylated with methyl trifluoromethane sulphonate using 2,6-di-(rerf-butyl) pyridine as the proton scavenger. If undermethylation is suspected, then complete methylation can be obtained by treating the partially methylated polysaccharide with silver oxide and methyl iodide 6 7. A second treatment by the Hakomori procedure will cause p-elimination of any uronic acid residue present. Polysaccharides containing uronic acid can be carboxyl-reduced before (carbodiimide reduction52) or after (lithium aluminium deuteride53) the permethylation steps. The usual cause of undermethylation is the incomplete dissolution of the sample. Some polysaccharides like cellulose, are insoluble in DMSO; however, these can be methylated with a N-methylmorpholine-N-oxide (MMNO)-DMSO mixture. Solubility can also be improved by using the free acid form of the polysaccharide, by ultrasonication and by heating to ~ 50°C before methylation. The methylated product is recovered by dialysis or partition between water and chloroform. Infrared spectroscopy is used to check for completeness of methylation (absence of hydroxyl absorption at 3600 cm"1). - 15 -11.4 ANALYSIS OF NON-CARBOHYDRATE SUBSTITUENTS The most common non-carbohydrate substituents in bacterial polysaccharides are the base-labile acetates and acid-labile acetal-linked pyruvate groups. If present, these non-carbohydrate substituents usually function as the antigenic determinants10. Therefore, determination of the location of these groups is of paramount importance. The presence of pyruvate is best identified and quantified by n.m.r. spectroscopy (discussed in Section III.l). The positions of linkage can be obtained by comparing the g.c.-m.s. data from the methylation analysis of the pyruvylated and the de-pyruvylated polysaccharide. The iV-acetyl group is usually associated with 2-acetamido-2-deoxyhexoses and is considered to be part of the sugar. Identification and quantification of the number of O-acetyl groups per repeating unit of a capsular polysaccharide can be measured from its ^-n.m.r. spectrum (discussed in Section DXl). Fast atom bombardment (f.a.b.)68 mass spectrometry can be used to determine to which monosaccharide residues the O -acetyl groups are attached 6 9, however, the exact positions of linkage can only be attained by Prehm methylation66 or the de Belder and Normann procedure70. Methylation using a strong base will cause de-O-acetylation, therefore, pertrifluoroacetyl derivatives are normally used in f.a.b.-m.s.69 for the location of <9-acetyl groups. 11.5 DEGRADATION OF POLYSACCHARIDES INTO SMALL OLIGOSACCHARIDES Specific cleavages of polysaccharides into small oligosaccharide fragments, followed by isolation and characterization of these units are the established methods for polysaccharide sequencing. A number of specific degradation methods have been developed over the years. These include: partial acid hydrolysis, p-elimination7 1 , 7 2, periodate oxidation73 and enzymic degradation74. II.5.1 PARTIAL ACID HYDROLYSIS Partial acid hydrolysis is the most common procedure for cleaving polysaccharide into smaller fragments. The mode of cleavage is dependent on the different rates of hydrolysis of - 16 -different glycosidic linkages75. The factors affecting the rates of hydrolysis depend on: ring size, configuration, conformation, positions of linkage, position of saccharide within the primary structure and any non-carbohydrate substituents present. Uronic acids and amino sugars are generally more resistant to acid hydrolysis than neutral sugars, and therefore, oligosaccharides containing these two residues are frequently isolated. The disadvantages of acid hydrolysis are: i) the possible loss of acid-labile substituents; ii) that partial hydrolysis usually gives non-specific cleavages, and therefore, a mixture of oligosaccharides is generated. Nevertheless, the mixture of oligosaccharides generated by this method, after permethylation, can be analyzed by g.c.-c.i.-m.s. to give sequence information on the polysaccharide itself76. 11.5.2 p-ELIMINATION Any methylated polysaccharides containing 4-O-substituted uronic acids will undergo p-elimination when treated with a strong base, such as dimsyl anion 7 1 , 7 2. If the uronic acid is in the side chain, the resultant new polysaccharide can give an indication of where and how the acid is linked. On the other hand, if the uronic acid residue is in the main chain, this will result in fission of the polysaccharide and the oligosaccharides generated can provide information on the saccharide residues adjacent to the uronic acid residue. The 90 77 oligosaccharides can also be used for m.s. sequencing of the PS • . 11.5.3 PERIODATE OXIDATION/SMITH DEGRADATION Water-soluble polysaccharides containing 1,2-diol and 1,2,3-triol systems can be degraded by periodate oxidation 7 8 , 7 9. However, polysaccharides containing uronic acid residues will be incompletely oxidized due to the electrostatic repulsion between the periodate and the carboxylate ion. Therefore, the polysaccharides should be reduced using the carbodiimide procedure52. The oxidation can be monitored by spectrophotometry80 or by titration of the liberated formic acid and formaldehyde81 from a 1,2,3-triol. The resultant - 17 -polyaldehyde is usually reduced to the polyalcohol to prevent hemiacetal formation of the highly reactive aldehydes. This procedure is often rnis-identifled as the Smith degradation82. The Smith degradation involves periodate oxidation of the polysaccharide, reduction of the polyaldehyde, and a mild acid hydrolysis of the resultant polyalcohol at room temperature to cleave acyclic acetal linkages and leave the remaining glycosidic linkages intact 8 2. Information on the oxidized saccharide can be obtained from the resultant aglycones (e.g., a 4-O-substituted hexopyranose will produce a 2-0-tetritol and a 2-0-substituted hexopyranose will produce a 2-0-glycerol). The hydrolysis step can be monitored by g.c. analysis of the trimethylsilyl derivatives of the product83. II.5.4 BACTERIOPHAGE DEGRADATION Bacteriophages are serotype-specific viruses that first infect and then propagate within the bacterium thus destroying the host ce l l 8 4 " 8 6 . They are often designated by the Greek letter 0, followed by the number of the serotype of the host strain. The absorption of the phage to its receptor is highly specific, where the exopolysaccharide acts as a receptor for phage b i n d i n g 8 7 . The phage-associated endoglycanases depolymerize this exopolysaccharide, and subsequent penetration of the host cell by the phage is followed by the release of viral D N A 8 8 . The use of these phage-borne enzymes for depolymerization of bacterial capsular polysaccharides allows the isolation of selectively cleaved oligosaccharides corresponding to one or more oligosaccharide repeating units (termed PI, P2, etc.) . Furthermore, acid- and base-labile non-carbohydrate substituents remain intact on the oligosaccharide. This is usually difficult to achieve using chemical means of degradation. Fig. U.1 shows a schematic diagram of selective bacteriophage (<]>) cleavages of a capsular polysaccharide to give PI and P2 units. The characterization of the repeating oligosaccharides represents the structural elucidation of the polysaccharide itself; thus oligosaccharides generated by bacteriophage degradation are excellent substrates for mass spectrometric studies89. - 18 -4> <t> pyr pyr -D- •B- •D- -B- -D-OAc OAc 37 °C for 3 days pyr A B C D PI unit OAc pyr pyr -B- •B- •D P2 unit OAc OAc Fig.n.l Schematic diagram to illustrate bacteriophage (<)>) degradation of a capsular polysaccharide (made up of tetrasaccharide repeating unit) to give PI (tetrasaccharide) and P2 (octasaccharide) units. Note that base- or acid-hydrolysis will remove either the O-acetyl or the pyruvic acid acetal groups respectively, with loss of substituent information. The P2 unit contains all the information required to elucidate the structure of the CPS, while PI unit contains only partial (3/4) information. - 19 -CHAPTER HI INSTRUMENTAL METHODS USED IN CARBOHYDRATE ANALYSIS - 20 -INSTRUMENTAL METHODS USED IN CARBOHYDRATE ANALYSIS The advance in analytical instrumentation, has generated a phenomenal improvement in instrumental techniques. Moreover, with the introduction of pulsed Fourier transform spectroscopy and superconducting magnets, complete characterization of polysaccharides may be possible in the near future by both n.m.r. and m.s. III . l NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY N.m.r. being a non-destructive physical technique is used extensively in carbohydrate chemistry for structural, configurational and conformational analyses90"93. The traditional n.m.r. experiments have been conducted in solution, but recent solid state experiments using cross polarization-magic angle spinning are developing as the technique improves94. The most useful one-dimensional n.m.r. parameters in carbohydrate chemistry are: i) chemical shifts; ii) relative integral intensities; iii) coupling constants; and iv) proton spin-lattice relaxation times (T i ) 9 5 , 9 6 . In recent years, two dimensional n.m.r. techniques have became more prominent in structural elucidation9 7 , 9 8. Complete characterization can be achieved using a combination of classical methods and high field n.m.r9 9. The n.m.r. approach suffers from the main disadvantage of requiring a relatively large amount of material (~ 10 to 20 mg). III.1.1 C H E M I C A L SHIFTS The chemical and magnetic environment of a nucleus governs its chemical shifts (8). The 8 value is usually referenced to acetone and expressed relative to tetramethylsilane. The 8 scale can be divided into three main regions for a *H spectrum and four for a 1 3 C spectrum: i) The anomeric region - ! H : H - l a-anomers 8 5.5 to 5.0 and p-anomers 8 5.0 to 4.5; 1 3 C : a-pyranoses 8 98 to 103, p-pyranoses 8 103 to 106 and furanoses 8 106 to 109. ii) The ring proton region - the *H spectrum is not well resolved (8 3.0 to 4.5). In contrast, the 1 3 C spectrum can be used to assign linkage position. A free primary alcohol gives a - 21 -signal at 8 60 to 62. If it is substituted then the signal shifts downfield by 7 to 10 ppm. The rest of the ring carbons resonate between 8 65 to 80. iii) The high field region where the methyl groups of pyruvate (depending on the R or S stereochemistry of the acetal100, - 8 1.5; 1 3 C : 8 17 to 25, ); acetate (*& 8 2.0 to 2.2; 1 3 C : 8 20 to 25) and 6-deoxyhexoses ^ H : - 8 1.3; 1 3 C : 8 16 to 18) can be detected. iv) The carboxy and carbonyl region which is only observed for 1 3 C spectra: ~ 8 170. 111.1.2 R E L A T I V E INTEGRAL INTENSITIES A quantitative estimation of the ratio of a- to 8-anomers, the number of 6-deoxyhexoses, pyruvate and acetate in the repeating oligosaccharide unit can be accomplished by measuring the relative integrals of the appropriate signals in the *H polysaccharide spectrum. Due to saturation and the nuclear Overhauser enhancement (n.O.e.) effect, quantitation with a proton decoupled 1 3 C spectrum is not reliable. Nevertheless, by counting the number of 1 3 C signals with the same number of hydrogen atoms in the spectrum the number of sugar residues per repeating unit can be determined. Oligosaccharides mutarotate, and exhibit two signals (a- and B-) for the "mixture" of reducing sugar. Therefore, the degree of polymerization of an oligosaccharide can be determined by comparing the number of anomeric signals to those of the reducing sugar. 111.1.3 COUPLING CONSTANTS Coupling constants (J values) arise from spin-spin coupling between two or more nuclei possessing magnetic moments. The coupling constant is mediated by the bonding electrons, and its magnitude is dependent on the type of chemical bond, the bond angle, the nature of the two nuclei and their distance. Therefore, coupling constants are used extensively for configurational and conformational analyses. The Karplus equation describes the relationship of J to the torsional angle (0) of the coupled nuclei 1 0 1. The value is maximum when 0 is 0° or 180° and minimum at 0 of 90°. For a pyranosyl system, the splitting of transdiaxial protons - 22 -is ~ 8 to 12 Hz, while equatorial-equatorial and equatorial-axial protons have coupling constants of ~ 1 to 3 Hz and ~ 1 to 6 Hz respectively. III.2 GAS CHROMATOGRAPHY Although the separation and identification of saccharides is now dominated by g.c.-m.s., paper chromatography is still used for initial identification of the sugar residues from the color formed by reaction with p-anisidine HC1 together with their R f values102. The principle of gas liquid chromatographic separation depends on the partition of volatile components between a mobile gas phase and a stationary liquid phase. In modern capillary columns, the liquid phase is directly bonded onto the fused silica column wall. These wall coated open tubular columns provide excellent separation of compounds previously very difficult or impossible to separate on packed columns. Therefore, capillary g.c. is an excellent quantification and quantitation method for sugars. The flame ionization detector provides a large linear dynamic range for measuring both high and low concentration samples. As the availability of gas chromatographs fitted with mass selective detectors increases, g.c.-m.s. may play an even more important role in structural analysis. Because carbohydrates are non-volatile, and thus are not directly amenable to g.c. analysis, they must therefore be chemically derivatized. The two most common derivatives used in g.c. analysis are the trimethylsilyl (TMS) and the alditol acetate (AA) derivatives 1 0 3 , 1 0 4. Medium polarity columns, such as DB-17™ (100% methylphenyl-TM polysiloxane, J&W Scientific) or DB-225 (50% cyanopropylmethyl-(50%)-methylphenyl-polysiloxane, J&W Scientific) are normally used for g.c. separation. Although trimethylsilyl derivatives are readily formed, the presence of a multiplicity of peaks due to the many isomeric forms (a-, B-, furanose and pyranose forms) complicates the chromatogram. Therefore, alditol acetate derivatization is commonly used because it generates a simple chromatogram and it facilitates easier quantification without using response factors105. Several comprehensive reviews on g.c. and g.c.-m.s. of alditol acetates have been published18 ,34. For co-eluting partially methylated alditol acetates (PMAAs), their partially - 23 -ethylated alditol acetates can be used1 0 6. There is little published data on methylated amino sugars, except for the common 2-acetamido-2-deoxyhexoses107,108. Amino sugar derivatives have longer retention times than their corresponding neutral sugars. It should be noted that when working with the simple regular repeating units of microbial capsular polysaccharides, exact quantification is not critical, rather the ratio of the sugar species to each other is more important. Oligosaccharides generated by the different chemical or enzymic methods are chiefly separated by paper chromatography, followed by g.c.-m.s analysis of the hydrolyzate as the PMAA derivatives. However, for oligosaccharides with similar Rf value or molecular weights, the separation can be incomplete. Therefore, g.c.-c.i.-m.s. using non polar columns, TM such as a DB-1 (100% dimethyl-polysiloxane, J&W Scientific) is becoming more prominent for the simultaneous separation and characterization of permethylated oligosaccharides7 6 , 1 0 9 , 1 1 0. Nevertheless, both g.c. and c.i.-m.s. data on oligosaccharides are not readily available. Furthermore, the volatility and thus molecular weight limit imposed by g.c. makes separation of high molecular weight oligosaccharides difficult, although, the new generation of high temperature non-polar columns has demonstrated the potential of g.c. separation with a number of malto-oligosaccharides110. I l l .3 MASS SPECTROMETRY Due to their high molecular weights (~ 106 daltons), direct use of mass spectrometric methods for the analyses of capsular polysaccharides is at present not realistic. Thus, mass spectrometry in the past has played a small role in polysaccharide analysis, largely being limited to the use of g.c.-m.s. for detection and identification of derivatized monosaccharides obtained by chemical degradation of the biopolymers. Nevertheless, the molecular weight of the oligosaccharides generated by specific degradation of capsular polysaccharides is well within the mass detection range of modern mass spectrometers. Furthermore, in the last two decades a number of novel desorption/ionization techniques have been developed which enable very polar non-volatile compounds to be analyzed without prior derivatization. The - 24 -different desorption and/or ionization techniques used include: field desorption (f.d.) 1 1 1, desorption chemical ionization (d.c.i.)1 1 2, Califomium-252 plasma desorption ( 2 5 2 Cf p.d.) 1 1 3, laser desorption ionization Fourier transform ion cyclotron resonance (l.d.i.-F.t.-i.c.r.)114, thermospray 1 1 5 , fast atom bombardment (f.a.b.)67 and liquid secondary ion mass spectrometry (l.s.i.m.s.) 1 1 6. Although mass spectrometry is a destructive chemical technique, the quantity required for analysis is often in the sub-ng level. Therefore, a number of researchers in the bacterial capsular polysaccharide field, including ourselves, are exploiting this powerful combination of desorption mass spectrometry and specific degradation to sequence the repeating oligosaccharide. It is important to note that historically, m.s. was and still is used in carbohydrate 1 ft chemistry primarily for characterizing monosaccharide derivatives separated by g.c. . Mass spectral identification is achieved by cross-referencing with reference spectra of all the common PMAA derivatives (Section UI.3.1.1)18. Although previous workers have attempted to characterize oligosaccharides by m.s., these experiments were not successful due to the low volatility of the samples even with derivatization1 9 , 1 1 7. Earlier, it was hoped that m.s. could be used to distinguish the different positions of linkage for each individual monosaccharide residue. The only ionization technique available at that time was electron ionization (e.i.), but this method, unfortunately, causes a lot of fragmentation and was found to be useful only for the analysis of low mass permethylated oligosaccharides. Furthermore, the analysis of these very complicated mass spectra was too superficial in comparison to the well-documented methylation analysis. With the development of novel desorption/ionization techniques, it was possible to desorb and ionize large oligosaccharides. Nevertheless, in f.a.b., l.s.i.m.s. and 2 5 2 C f p.d., fragmentations occur predominantly at the glycosidic bonds, due to the low energy transferred to the carbohydrate molecules, and the only information obtained from these ionization techniques is the molecular weight and the sequence information. The lack of ring cleavage fragmentations implies that no information can be obtained on the different stereoisomers, substitution patterns or anomeric configurations. Tandem mass spectrometry, such as f.a.b.-m.s.-.m.s., was carried out by a number of - 25 -workers on mono- or di-saccharides to obtain extra structural information relating to substitution patterns or anomeric configurations118"121. However, most correlations were based on the relative intensities of different peaks, which can sometimes be misleading. Although f.a.b. analysis can locate a monosaccharide residue which carries a substituent, it is incapable of defining the exact position of the linkage. The positions of linkage for these non-carbohydrate substituents are normally determined by e.i.-m.s. of the derivatives of the hydrolysate obtained from chemical degradation. Therefore, for any single m.s. technique to totally characterize oligosaccharides, it is important that it provides the following: i) glycosidic bond cleavages for sequencing; and ii) carbon backbone cleavages to provide information on the substitution patterns. The importance of carbon backbone cleavage data for providing linkage information is illustrated in Section IV.3.6. So far, no single mass spectrometric technique can be used to distinguish stereoisomers of different monosaccharide units (e.g., glucose vs galactose). One possible method is by reaction mass spectrometry, where simple mono- and di-saccharides react with borate additives in the f.a.b. matrix to give specific complexes122. The different stereoisomeric complexes gives different spectra and the hexose identity can then be obtained by correlating the relative intensities of different peaks. However, a simpler classical carbohydrate method to characterize monosaccharide residues is methylation analysis (Section n.3 and III.2). Methylation analysis provides the following information: i) the identity of the monosaccharide residues; ii) the positions of linkage of each monosaccharide residue; iii) the structural pattern of the oligosaccharide, e.g., linear or branched; and iv) the identity of the branch point monosaccharide residue. Therefore, the general instrumental approach to analyze an unknown oligosaccharide requires the characterization of the monosaccharide residues with methylation analysis and then the sequencing of the oligosaccharide by n.m.r. or m.s. Mass spectrometers in general consist of four components: a) the ion source, where the sample is transferred to the gas phase and ionized; b) the analyzer, where the ions are mass analyzed according to their mass to charge ratio; c) the detector, where the mass-resolved ion beams are detected, amplified and their intensity recorded; d) the vacuum system, which - 26 -provides a stable environment for the above processes. A detailed review of mass spectrometry will not be discussed here, although, a few of the most frequently used techniques are described in the following sections. III.3.1 ELECTRON IONIZATION The basic process whereby a molecule is ionized under electron impact is: M(g) + e- - » M ( g ) + * + 2e" This is the most common procedure used for generating ions in the gas phase. When electrons collide with neutral particles, the collisions can be elastic or inelastic. For elastic collisions the impact velocity is relatively low, and no appreciable loss of energy of the electrons is detected. At higher impact velocities the neutral particles are excited and the collisions become inelastic, but these neutral particles are not charged and hence not detected. Above the ionization potential (~ 12 eV for most molecules), many configurational changes are possible, with ionization and dissociation being most prominent. Above 25 eV, doubly charged ions and fragmentation can occur. The maximum of the ion intensity vs electron energy curve is located approximately at 70 eV. Thus, e.i.-m.s. utilizes mainly 70 eV energy electrons. The excess energy acquired by the molecule causes it to undergo single or multi-stage fragmentation; consequently, molecular ions are rarely observed for non-volatile materials. E.i.-m.s. is, therefore, performed primarily on volatile, low mass samples, such as monosaccharide alditol acetate derivatives (discussed in Section 0X3.1.1). Higher mass samples will frequently decompose thermally or fragment such that structural information is lost. Although molecular ions of derivatized oligosaccharides can be generated with low eV electron ionization, the positive radical molecular ions (i.e. an odd-electron species) are normally unstable and therefore, not observed. Chizhov and Kochetkov1 1 7 have reported a series of e.i. fragmentation patterns for cyclic permethylated sugars and proposed nomenclature for these ionization pathways. Cleavage of the glycosidic bond gives rise to fragments of the A series, the C-l.C-2 bond to the C, D, F, H and K series, the C-4,C-5 - 27 -bond to the J, H, B and D series (Fig. DXl and UI.2). A number of workers have attempted to use these fragmentation patterns for structural characterization of permethylated oligosaccharides with varying degrees of success • . Fig.m.l Some fragmentation patterns of a permethylated hexose proposed by Chizhov and Kochetkov117. - 28 -MeO '+ MeO 4 \. ) OMe .m.2 Some fragmentation patterns of a permethylated hexose proposed by Chizhov and Kochetkov 1 1 7. - 29 -in.3.1.1 EJ.-M.S. OF PERMETHYLATED ALDITOL ACETATES Of all the carbohydrates, it is the alditol derivatives that exhibit the simplest mass spectrometric fragmentation patterns18. The following generalizations can be applied to the mass spectra of PMAAs. i) Derivatives with the same substitution pattern generate very similar mass spectra. Therefore, the mass spectrum of D-mannitol hexaacetate is representative of all the peracetylated hexitols. ii) No molecular ion is observed for e.i., and the base peak is customarily m/z 43 (C2H3O). iii) Primary fragments result from a-cleavage of the carbon backbone, with the relative intensities of the ions decreasing with increasing molecular weight. iv) The fission between carbon atoms is determined by the stability of the resulting radical, with the methoxylated radical being more stable than the acetoxylated radical. The primary fragmentation of amino alditol acetates largely depends on the substitution of the acetamido group. Preferred bond cleavages are in the following order. 1 Me I I HC—N—Ac HC—OMe > 1 Me I I H C — N - A c HC—OAc > H C - O M e H C - O M e > 1 1 7 I HC—OMe ^ H C — N - A c ^ HC—OAc H C - O A c HC—OAc H C - O A c I I I Fig. III.3 Preferred order of fragmentation for partially methylated alditol acetates 18 v) Secondary fragments are obtained by single or consecutive loss of acetic acid (60 amu), acetamide (59), ketene (42), methanol (32) or formaldehyde (30). - 30 -Deuterium labelling studies using sodium borodeuteride or lithium aluminum deuteride can be used to identify the reducing end of an oligosaccharide and to distinguish an acidic residue from the neutral sugars by reduction, prior to sugar analysis by g.c.-m.s. For g.c.-m.s. analysis of partially methylated alditol acetates, both the g.c. retention time and the fragmentation pattern are important. The retention time is especially crucial when different epimers are under investigation (e.g., glucose vs mannose). Fig. III.4 illustrates the fragmentation pattern of a typical PMAA (e.g., l,5,6-tri-0-acetyl-2,3,4-tri-0-methylglucitol, mw = 350 amu)18. CH 2OAc 57,75- 117 Kfr^C/J^ "  22 3q 73 101,119,131—161 M e O - C H uX" "n^V""" "1"8"9" ^ 169,147,129 175- 205 ?5r7 O M e HC—OAc I CH 2OAc Fig.m.4 Fragmentation pattern of l,5,6-tri-0-acetyl-2,3,4-tri-0-methylglucitol, as proposed by Lindberg et a/ 1 8. III.3.2 CHEMICAL IONIZATION/DESORPTION CHEMICAL IONIZATION Chemical ionization was first introduced in 1966124. It was developed directly from fundamental studies of gas phase ion/molecule reactions. In essence, the reagent gas which is present in large excess (1000 x) is ionized by electron impact. This is followed by ion/molecule reactions between the primary ions and the neutral reagent gas which produces the c.i. reagent ions. As a result of collisional relaxation, the c.i. reagent ions have a much lower energy than the ionizing electrons, consequently c.i. is a milder form of ionization. The mode of ionization is a type of Lewis acid-Lewis base reaction, where the molecule is converted to a relatively low energy even-electron species. Therefore, an abundance of pseudo-molecular ions can be observed. - 31 -The most widely used positive-ion reagent gas systems are those which yield Br0nsted acids 1 2 5 . The major reaction mode of the Br0nsted acids, BH*. is proton transfer to give initially the protonated molecule M H + . B H + + M -> M H + + B In principle, the B H + ions can also react by hydride ion abstraction or by charge exchange. B H + + M [M - H ] + + B H 2 (or B + H 2 ) B H + + M -» M + - + BH- (or B + H ) By selecting the appropriate reagent gas, pseudo-molecular ions may be observed in the presence of fragment ions, with similar fragmentation patterns obtained from e.i.-m.s. For carbohydrates, the most common c i . reagent gas used is ammonia and the pseudo-molecular ion observed in the positive-ion mode is generally M+NH4 + . Positive-ion chemical ionization using methane as the reagent gas usually gives the fragment ion (M+H-AcOH)+, i.e., (M-59)+. Although c.i.-m.s. can directly analyze underivatized oligosaccharides, the molecular weight Umit of such samples is fairly low, hence derivatization is normally carried out for m.s. analysis. G.c-c.i.-m.s. analysis of permethylated oligosaccharides, obtained by partial acid hydrolysis, has been used successfully for the sequencing and characterization of capsular polysaccharide . Another method of increasing the abundance of the pseudo-molecular ions is to desorb materials from sample probes which are positioned directly within the reagent plasma (d.c.i.) 1 1 2. Since desorption chemical ionization still involves heating and sample-surface interaction, pyrolysis can still be a problem. Nevertheless, d c i . is frequently used for the sequencing of permethylated oligosaccharides126. Furthermore, structural information may be obtained if Chizhov's fragment ions are observed. III.3.3 FAST ATOM BOMBARDMENT/LIQUID SECONDARY ION MASS SPECTROMETRY Mass spectrometric analysis of ions produced by bombarding surfaces with ions in the keV energy range is well established as a surface technique127. The early applications of - 32 -secondary ion mass spectrometry (s.i.m.s.) were mostly in fundamental studies which utilized the fact that s.i.m.s. provides high absolute sensitivity for many surface components. Furthermore, s.i.m.s. unlike some other available techniques, could detect hydrogen and could determine the isotopic composition of elements on the surface128. S.i.m.s. was first shown to be a very sensitive technique for detection and identification of organic compounds in 1976129. The main disadvantage of s.i.m.s. is the radiation damage caused by the exciting primary ion beam. This damage can be mirumized by operating in the "static mode", in which the surface is bombarded at an extremely low primary ion current density. Hence, during the analysis only a small probability exists for a surface area damaged by the impact of an ion to be struck by another primary ion during measurement. The secondary ions formed can be either atomic or molecular. Formation of molecular ions can be accomplished by three different processes: i) canonization and anionization (these result from chemical reaction during or following the concurrent production of metal ions and the liberation of organic molecules by sputtering or by thermal vaporization); ii) electron transfer (the resulting anion and cation radicals M"" and M + " , can be observed in s.i.m.s. spectra, sometimes associated with fragment ions which correspond to known products of their uriimolecular dissociation); and iii) the direct sputtering of organic molecular ions from the solid to the gas phase130. These ions arise by momentum transfer from the primary ions. A l l s.i.m.s. spectra show high abundances of pseudo-molecular ions that allow determination of molecular weight. Secondary ion yields are strongly dependent on the substrate and the properties of the primary beam, with the highest parent-like emissions obtained from compounds deposited on noble metal substrates and utilising a C s + ion beam. With few exceptions, low-energy bombardment of solids has used keV ion beams rather than neutrals, largely as a matter of experimental convenience since ion beams in this energy range are relatively easy to produce131. It is generally understood that the approaching ion is neutralized by long-range electron transfer as it approaches the surface so that the major effects of the impact would be the same for neutrals as for singly charged ions 1 3 2. - 33 -F.a.b. was first developed in 1981 6 8 and is presently the most common desorption/ionization technique used for the sequencing of biological molecules 1 3 3 , 1 3 4 . Secondary ions are produced by bombarding the sample in a liquid matrix with a primary neutral beam (Ar° or Xe°) in the 8 kV energy range produced by charge exchange neutralization of an ion beam. It is now well established that the presence or absence of charge on the impacting primary particle has little effect on the desorption process, but the use of the neutral beam is somewhat more convenient with magnetic instruments in which the ion source is at high potential . Nevertheless, the increase in secondary ion intensities resulting from the use of a 20 to 30 kV ion beam (Cs+) has convinced a lot of the original f.a.b. spectrometrists to work with liquid secondary ion mass spectrometry (l.s.Lm.s.) . With the use of the liquid matrix the requirement for working in the "static mode" to avoid radiation damage on the surface is removed. The beam current density used is several orders of magnitude larger than those of s.i.m.s. and the spectra are stable for up to 30 minutes 1 3 7. The reason for this is that the matrix presents a mobile, constantly renewed surface to the bombarding beam, which continously replenishes undamaged sample molecules to be ionized. However, substantial radiation damage to the matrix does occur 1 3 8. One disadvantage of the liquid matrix is that a very high background is produced by the ionization of the matrix and its radiation products. Also clustering of the matrix with sample ions is observed with formation of mixed cluster ions when impure samples or mixtures are analysed. Mass spectral data showed that the same secondary ions are produced whether the sample is in a liquid or solid matrix. The distribution of the secondary ions is representative of the composition of the liquid phase. If odd-electron species are just as stable in solution as even-electron species they can be detected just as easily. Although the ion production mechanism can be considered as identical in f.a.b. and s.i.m.s., a solid matrix usually leads to more fragmentation due to the process of primary energy dissipation. Solutes, molecules and ions are less tightly bound in liquids and the energy dissipation occurs mainly as translation instead of as vibration of the matrix 1 3 9. This may lead to the ejection of larger solvated - 34 -clusters where energy dissipation can easily be accomplished by excessive loss of solvent molecules without breaking intramolecular bonds140. Bombarding with lower energy primary particles has been found to increase the extent of fragmentation probably via the ejection of smaller clusters where the energy dissipation by solvent loss is less efficient. The size distribution of the ejected clusters, depending on the primary beam conditions, could be reflected in the degree of fragmentation. The second aspect in which solid and liquid matrix differ is related to the dynamic nature of the liquid sample. The liquid matrix must be considered to be mobile in the spectra-recording time scale. Hence, f.a.b. can reflect the equilibria of the solution; it has been used to measure acidity constants, stability constants, and to monitor enzyme reactions 1 4 1 , 1 4 2. A situation of particular importance occurs when the solvent-solute interactions are strong enough to induce heterogeneities in the distribution of species within the liquid matrix. If one solute behaves like a surfactant, a concentration gradient will develop, resulting in the enrichment of the surface with this compound. This behaviour is not confined to certified surfactants such as the long chain quarternary ammonium salts but also occurs with amino acids and dipeptides 1 4 3 , 1 4 4. A crucial step in the f.a.b. technique is the choice of a suitable matrix composition145"147. The same solute can produce ions that vary in abundances by orders of magnitude as the matrix composition is altered. Some general requirements concerning the solvent properties of the matrix are summarised as follows: i) The sample must be soluble in the matrix, ii) Only low-vapour pressure solvents can be used in the vacuum of the mass spectrometer. Volatile solvents could in principle be employed with high pumping rates, provided that stable surfaces could be obtained on the time scale of recording a spectrum, iii) The viscosity of the solvent must be low enough to ensure the diffusion of the solutes to the surface, iv) Ions from the matrix itself must be as unobstrusive as possible in the secondary ion mass spectrum, v) The matrix must be chemically inert; if specific ion formation reactions are used to promote secondary ion yield the reaction must be reproducible. The usual matrices used for carbohydrates are glycerol, thioglycerol or a 1:1 mixture of both. Frequently additives - 35 -such as 1% volatile acid (acetic and hydrochloric), NaCl and KBr are added to improve molecular ion intensity. However, cationization of the molecular ion will decrease fragmentation, with possible loss of sequence information. Although underivatized samples can be used, derivatization is frequently carried out to improve ion intensities. F.a.b. produces simple fragmentation patterns, with dehydration and glycosidic bond cleavage for underivatized samples, and primarily oxonium ion formation (Chizhov's A series) from derivatized materials. III.3.4 LASER DESORPTION IONIZATION FOURIER TRANSFORM ION CYCLOTRON RESONANCE D.c.i. and f.a.b. usually give abundant sequence information on derivatized oligosaccharides, however, they provide little or no information on either the positions of linkage between the saccharide residues or the anomeric configuration of the linkages1 4 8. Tandem mass spectrometric experiments, such as collisional activation dissociation (c.a.d.) experiments119 and mass ion kinetic energy (m.i.k.e.s.) experiments118 have been carried out on either permethylated disaccharides or methyl glycosides in an attempt to distinguish the position of linkage and the anomeric configuration. The commonly used derivatization methods are esterification of the reducing end with ethyl p-aminobenzoate136 or reduction followed by either permethylation or peracetylation148. A l l these derivatization techniques can remove any acid/base labile substituents that may be present and structural information consequently will be lost6 9. It would be preferable to sequence native carbohydrates without having to use derivatization. The application of laser desorption for mass analysis of non-volatile and thermally labile biomolecules, such as oligosaccharides, glycosides, lipid A compounds and oligopeptides has been shown to be useful 1 4 9" 1 6 0 . Most of the pseudo-molecular ions observed in positive-ion spectra were either cationized or protonated but the exact mechanism of desorption/ionization is still under investigation 1 6 1" 1 6 9. The potential of l.d.i.-F.t.-i.c.r. for the analysis of oligo- and poly-saccharides was demonstrated by Coates - 36 -and W i l k i n s 1 7 0 , 1 7 1 . From the l.d.i.-F.t.-i.c.r. mass spectra of nine commercially available polysaccharides they observed "extensive fragmentation of the saccharide chains, from both within the sugar rings and between them ... . Although similarities are seen in the spectra of some compounds, each displays a characteristic fragmentation pattern."151 Using the high resolution data obtained from the positive-ion l.d.i.-F.t.-i.c.r. mass spectra of an underivatized de-O-acetylated pentasaccharide derived from bacteriophage hydrolysis of Klebsiella K44 capsular polysaccharide, we have shown that l.d.i.-F.t.-i.c.r. can be used to 90 179 sequence underivatized oligosaccharides (Appendix i y 4 U - 1 , A , Encouraged by these results, a systematic study involving application of positive- and negative-ion l.d.i.-F.t.-i.c.r. to a number of well characterized underivatized model oligosaccharides was initiated. These studies will be discussed in Section IV.3.1 to IV.3.8. Moreover, the fragmentation pattern observed for these underivatized model oligosaccharides can also be used to explain the polysaccharide mass spectra observed by Coates and Wi lk ins 1 7 1 , 1 7 3 . This will be discussed in Section IV.3.9. In Section IV.3.10 and IV.3.11, this l.d.i.-F.t.-i.c.r. technique was applied to an unknown oligosaccharide, and results obtained compared with those obtained by classic carbohydrate chemistry. The two different aspects of l.d.i.-F.t.-i.c.r. are discussed below. IH.3.4.1 LASER DESORPTION IONIZATION Soon after the advent of the laser several mass spectrometrists realized the potential of a laser mass spectrometer combination174. The nature of the radiation produced by the laser has some specific properties, which makes the laser a highly attractive source for sample desorption. Laser radiation is coherent, monochromatic, directional and intense. The high intensity of a laser beam enables sample vaporation to be on a very short time scale. The direction of the beam is defined by the geometry of the lasing system. Generally, a beam with a diameter of a few rnillimeters and a very low divergence is generated. This low divergence facilitates the operation of the laser at a relatively long working distance from the sample, without severe radiation loss. Furthermore, the coherence and directional properties together enable focusing of the beam to a very small spot, thereby increasing the power density. The - 37 -monochromatic nature of the radiation leads to a narrow wavelength band. Although monochromaticity is an important factor in many other laser applications, no extentive use of this property can be made in laser desorption ionization (l.d.i.). In principle, with a tunable laser very accurate scanning of wavelength over the absorption bands of the sample can be performed and the effects of selective excitations can be studied. Nevertheless, using multiphoton mass spectrometry Schlag et. al. demonstrated that the isotopic species mono-C-benzene can be preferentially ionized in a natural isotopic mixture by shifting the wavelength by 1.6 cm"1 from the absorption band of light benzene175. This demonstrated that trace components in a mixture can be ionized without ionizing the major components if the intermediate-state spectrum shows sharp features at a resolution of 1 cm"1. They also proposed that multiphoton ionization occurs only subsequent to multiphoton absorption into the final energtic state of the neutral molecule, further multiphoton absorption will lead to dissociation. Laser radiation is produced either in pulses or continously174. The normal pulse range is from 1 us to 1 ms. However, each pulse exhibits a fine structure and is effectively built up of a train of separate pulses 0.1 to 1 us wide. By the so called Q-switching technique part of the total energy can be released in one short pulse of ~10 ns duration. Even shorter pulses in the pico second range can be obtained with an additional mode-locking technique. Although the total energy in a pulse is kept low, a 1 uJ pulse of 10"8 s focused on 10"6 cm 2 from a Q-switched Nd-YAG laser leads to an instantaneous high power density of 108 Wcm"2. As a result of the energy transferred to the sample material, heating, melting, vaporization and ionization processes take place. Under severe conditions large amounts of substrate can be vaporized leaving a crater in the solid. Two classes of power density can be distinguished. With low power density (< 108 W cm"2) evaporation of surface layers can be obtained, in the form of intact neutrals and ionized molecules. This evaporation process is reproducible, and can be looked upon as a thin vapour layer moving into the solid at a steady state. In the high power domain (> 108 W cm"2) a high degree of irreproducibility is observed and mainly atomic and small molecular fragments in an ionized state are desorbed. After - 38 -vaporization, collision between neutral and primary ions in the plasma can lead to formation of secondary ions. The dominant process is canonization by alkali ions. M(S) + hy —> Mi( g ) + M2(g) + M 3 ( g ) + etc... Due to the pulse nature of the ion current generated, special mass analyzers are required with emphasis on the scan speed and signal registration. Spectrometers equipped with time-of-flight detectors have been most frequently used because of their ability to record a complete spectrum for each laser generated ion pulse and high transmission. The major drawback of this system is the relative low dynamic range of the recording system due to the inherently short sampling time176. Quadrupole mass filter and scanning magnetic sector type instruments have also been used, but are too slow to obtain a complete spectrum for each pulse 1 5 0 , 1 7 7 . Therefore high, repetitive rate lasers have been used recording only one mass peak per laser pulse. Magnetic analyzers, both single and double focusing instruments, can be efficiently applied employing simultaneous ion recording techniques such as photoplates and channelplate-camera systems 1 7 8 , 1 7 9. Since Fourier transform ion cyclotron resonance (F.t.-i.c.r.) 1 8 0 , 1 8 1 mass spectrometry is an inherent pulse experiment which gives the whole mass spectrum at once 1 8 0" 1 8 5 , it is well suited for l.d.i. Gross et al first demonstrated the feasibility of l.d.i.-F.t-i.c.r.1 8 2. IH.3.4.2 FOURIER TRANSFORM ION CYCLOTRON RESONANCE In a uniform magnetic field of strength "B", a moving ion of charge "z", and of mass "m", will be subjected to the Lorentz force, which is perpendicular to the direction of the ion, hence, constraining it to a circular orbit with its radius proportional to its velocity. This orbit will have a characteristic cyclotron frequency, co, and the orbital motion can be described as co = zB/m The ion cyclotron resonance (i.cr.) spectrometer measures the cyclotron frequencies of an ensemble of ions of differing masses (illustrated in Fig. BI.5). The i.cr. experiment is divided into two parts: i) excitation of cyclotron motion by an alternating electric field between the plates of a capacitor (illustrated in Fig. m.6); ii) detection of the excited cyclotron motion of - 39 -tons (illustrated in Fig. HI.6). The rotating ions induce an alternating current in an external circuit; the voltage induced across this circuit is the i.cr. signal 1 8 3. In a F.t.-i.c.r. tpectrometer, all the ions are excited together by using a fast frequency sweep from a radio frequency oscillator. The i.cr. signal from all the ions is then digitally sampled and Fourier transformed to give a mass spectrum 1 8 4 , 1 8 5. This F.t.-i.c.r. concept was first demonstrated by Comisarow and Marshall in 1974 1 8 0 , 1 8 1 . Ion Cyclotron Resonance + + + + B + + + + B = 3 T e s l a I r——1 1 1 j 1 1 1 1 p- m (Da) 0 500 1000 - i 1 1 1 f ( H z ) 1 MHz 200 kHz 100 kHz 50 kHz - 4 1 -FT -1 C R excitation detection Kt) V(t) V(t) = K< cos (f«) + K 2 cos (f2) + . . . Fourier V(t) » whole mass spectrum Transform Fig.m.6 Schematic to illustrate the principle of Fourier transform ion cyclotron resonance mass spectrometry. - 42 -CHAPTER IV LASER DESORPTION IONIZATION FOURIER TRANSFORM ION CYCLOTRON RESONANCE MASS SPECTROMETRY OF SOME UNDERIVATIZED OLIGOSACCHARIDES DERIVED FROM SPECIFIC DEGRADATION OF BACTERIAL CAPSULAR POLYSACCHARIDES. - 43 -LASER DESORPTION IONIZATION FOURIER TRANSFORM ION CYCLOTRON RESONANCE MASS SPECTROMETRY OF SOME UNDERIVATIZED OLIGOSACCHARIDES DERIVED FROM SPECIFIC DEGRADATION OF BACTERIAL CAPSULAR POLYSACCHARIDES. IV. 1 INTRODUCTION The mass spectrometric investigations described in this and the previous M . Sc. thesis20 are specifically related to the structural characterization of Klebsiella and E. coli capsular polysaccharide antigens. In the previous studies20, a preliminary investigation was carried out on three different types of desorption/ionization mass spectrometry (desorption chemical ionization (d.c.i.), fast atom bombardment (f.a.b.) and laser desorption ionization Fourier transform ion cyclotron resonance (l.d.i.-F.t.-i.c.r.)) for the sequencing of oligosaccharides. It was demonstrated that positive-ion l.d.i.-F.t.-i.c.r. mass spectrometry could be used to sequence an underivatized pentasaccharide obtained by bacteriophage degradation of the bacterial capsular polysaccharide172 (Appendix 1). Furthermore, the ring fragments observed could be used to provide some tentative indication on the positions of linkage. Encouraged by these results, the application of both positive- and negative-ion l.d.i.-F.t.-i.c.r. to this and other underivatized oligosaccharides generated by specific degradation of the native bacterial capsular polysaccharide was further investigated. This thesis seeks to characterize the abundant ring fragmentation patterns observed in the l.d.i.-F.t.-i.c.r. spectra of model underivatized oligosaccharides (Section IV.3.1 to Section IV.3.8). The experiments progressed from a linear pentasaccharide {Klebsiella K44), to a linear pentasaccharide containing an acid-labile pyruvic acid acetal (Klebsiella K3), to a linear tetrasaccharide containing an acetamido hexose and a sialic acid (E. coli K9), and finally to a "3+1 branched" tetrasaccharide containing a base-labile 0-acetyl group and a lactic acid ether (Klebsiella K22). Using the fragmentation patterns observed from these model underivatized oligosaccharides, a logical interpretation could be proposed for the mass spectra of commercial polysaccharides published by Coates and Wilkins 1 7 1 (Section IV.3.9). In Section IV.3.10 and IV.3.11, the results obtained from l.d.i.-F.t.-i.c.r. of an unknown - 44 -oligosaccharide are presented, and compared to the structure obtained by classic carbohydrate chemistry. In Section IV.4, all the mass spectral data presented in this thesis was correlated to project possible uses of l.d.i.-F.t.-i.c.r. in carbohydrate chemistry. It must be emphasized that the focus of this thesis is not to illustrate that l.d.i.-F.t.-i.c.r. mass spectrometry can be used as a stand-alone method for the complete characterization of carbohydrate materials. But rather it serves to illustrate that l.d.i.-F.t.-i.c.r. may be used to minimize the amount of classic carbohydrate chemistry required to elucidate the structure of an oligosaccharide sample. IV.2 EXPERIMENTAL PROCEDURES A l l the oligosaccharides discussed in this thesis are obtained by bacteriophage or anhydrous hydrogen fluoride degradation of the native capsular polysaccharide. Due to the experimental approach in the isolation of capsular polysaccharide, the acidic component, such as hexuronic acid and pyruvic acid acetal have alkali halide as the counter ion. Therefore, the mass spectra reflect the amount of sodium and/or potassium ion associated with the oligosaccahride samples. Mass spectra were obtained by Drs. Asgeir Bjarnason and David A. Weil at Nicolet Analytical Instruments, with a Nicolet FTMS-2000 mass spectrometer equipped with a dual cell, a 3.0 T superconducting magnet and a Nicolet laser desorption interface. The laser was a Tachisto 216 pulsed C O 2 laser operating as a stable resonator with aperture-controlled beam characteristics. Equipped with a total reflector, the laser delivers up to 2 J in a 40 T|S wide pulse emitting at X = 10.59 micron. Output energy can be controlled by adjusting the aperture and was estimated to be on the order of 0.05 J per pulse for these experiments. The spot size of the focused laser beam on the probe tip was of the order of 100 microns. Approximately 100 n mol. of the sample, dissolved in 50 uL of methanol/water mixture, was used for mass analyses. Typically, each sample was examined three times: once with no dopant, once with KBr or NrLtBr dopant and once with KBr/NaCl or NHtBr/NaCl dopant. This solution was transferred to and allowed to evaporate on the stainless steel tip of the - 45 -direct insertion probe of the mass spectrometer. The vacuum system was pumped down to ~ 1 x 10'8 torr after insertion of the sample probe, prior to mass analysis. After each laser pulse, the probe was rotated so that the laser always struck a fresh spot on the sample surface. After the laser pulse, a delay of 2 to 3 s was employed to reduce the pressure in the cell, before ion detection. Mass spectra were collected in the direct (broadband) mode 1 8 5 with a high frequency cutoff at 500 KHz (corresponding to a lower mass limit of ~ 90 amu). Normally, 64K-datapoint transients were collected and when the signals were weak, the spectra from several laser shots were averaged to increase the signal to noise ratio. Positive- and negative-ion spectra were obtained in each case. For exact mass measurements, the series of ions corresponding to K n B r n _ i + or K n B r n + i " were used as internal mass calibrants. When these ions were absent or weak, a few peaks in the oligosaccharide spectra that could be assigned with certainty were used as internal calibrants. After data collection, the sample was 'recovered', doped with successive solutions of methanolic KBr and NaCl and rerun as necessary. Actual sample consumption was thus limited to several picomoles for each laser shot. IV.2a DATA ANALYSES The data analyses of the l.d.i.-F.t.-i.c.r. spectra were carried out as follows: i) The baseline noise level was first determined and all peaks above twice the baseline noise were considered to be real signals. A l l the spectra in this thesis have an average baseline noise of less than 5% of the base peak. ii) The highest-mass peaks in each spectrum were taken to represent the pseudo-molecular ions, which indicate the number of residues present in the oligosaccharide. A l l masses resulting from glycosidic bond cleavages gave "whole saccharide" structures which could then be used to derive a number of possible sequences for the oligosaccharide. In the positive-ion spectra, the ions are typically canonized (either from the added alkali salt or by salt present in the sample). Deprotonated structures are the most common species in the negative-ion spectra. Nevertheless, some chloridation of high intensity peaks are sometimes observed. - 46 -In both the positive- and negative-ion spectra, many of the observed masses are assigned to dehydrated formulae, formed by loss of one or more water molecules. iii) Due to the ease of dehydration of oligosaccharides, it is important to distinguish whether the sample contains any anhydro-saccharide residues, or if such fragments are dehydration products formed during the desorption/ionization process. If the pseudo-molecular ions are dehydrated, then all the hydroxy groups on all the fragment ions have to be carefully accounted for. iv) It should be noted that for a homoglycan, such as a mannan, the molecular formula for the molecule is [(A)n+18], where A is the molecular weight of the anhydro-residue, n is the degree of polymerization, and 18 represents the molecular weight of a water molecule. For example, the molecular formula of chitin is [(203)n+18], and the individual residues are separated by 203 amu in the mass spectrum. For each different class of saccharide residue, there are different mass separations. Consequently, the sequence of a heteroglycan can be obtained by adding the different mass separations. For a hexose, the mass separation is 162; for a deoxyhexose it is 146; for a hexuronic acid it is 176 and for an acetamido hexose it is 203. v) After obtaining a number of possible sequences for the oligosaccharide, the mass errors for the peaks derived from glycosidic bond cleavage were then calculated. If the mass errors were too high (i.e. £ ±50 ppm), a few peaks that can be assigned with certainty were used to recalibrate the spectra. vi) The recalibrated masses were again correlated with proposed sequences. Any masses that were not assigned to glycosidic bond cleavage were then considered to be derived from ring cleavages. Thus, elemental formulae could be assigned and structure(s) proposed in the context of the most probable sequence to delineate the oligosaccharide. vii) With the assumption that ring fragments were formed by simple ring cleavages, rather than extensive rearrangement of the monosaccharide residue, the ring fragments could then be used to elucidate the structure of the oligosaccharide in greater detail (i.e., positions of linkage). If possible, these fragments were then correlated to the known structure with - 47 -respect to anomeric configuration, epimeric stereochemistry, and ring size. The use of ring fragments in e.i. spectra for the determination of positions of linkage for low molecular weight permethylated oligosaccharides is well documented19. Nomenclature has long been proposed 117 for these ring fragments by Chizhov and Kochetkov . However, due to pyrolysis in both e.i. and c i . , underivatized oligosaccharides have not been studied extensively, viii) The error limits of high resolution data according to some mass spectrometrists have to be below ± 10 ppm 1 8 6 , therefore, the mass errors in some of the spectra presented in this thesis are too high to be considered "proper" high resolution results. Nevertheless, due to the nature of carbohydrate materials (which are predominantly made up of carbon, hydrogen, oxygen and nitrogen, all of a certain ratio), ultra-high mass resolution is not required, as the low resolution data limits the number of possible monosaccharide residues in the sample. IV.3 RESULTS AND DISCUSSIONS In the interpretation of l.d.i.-F.t.-i.c.r. spectra of oligosaccharides, it is not always essential for all the ions to be assigned exact structures. The low resolution spectra provide a framework of limited possibilities for the oligosaccharide sequences. Therefore, only selected ions of high intensity, in the high resolution spectra need be assigned to structures for complete sequencing. The structures of these ions are "whole oligosaccharides" attached to one or more sub-monosaccharide units (ring fragments). In some cases, two or more possible structures could be rationalized for a particular ion with the same elemental composition. The structures could be derived from different oligosaccharides attached to different ring fragments, or the same oligosaccharide attached to one or more ring fragments. Generally, only one of these structures exists in the framework designated by the low resolution spectra. If however, both of these structures were possible, then either both structures or only the most "weighted" probable structure contribute to the intensity of the ion. This "weighting" process was predominantly determined by whether: i) the ring fragment could be derived from a particular part of a monosaccharide residue; ii) the ring fragments in - 48 -the structures have been observed previously; and iii) the same ring fragment exists in the spectrum derived from different monosaccharide residues. The ring fragments observed are usually unsaturated and can be categorized into four different types: one-carbon, two-carbon, three-carbon and four-carbon fragments. Each of these can be further divided into sub-groups, with differing compositions of oxygen or hydrogen atoms. Any unsaturated structure can be written in a cyclic or a chain form. In this study most of these unsaturated fragments are arbitrarily drawn as the cyclic form as no mechanistic studies have been carried out to identify their exact structure. Furthermore, no charges are placed on any part of the fragment structures, but the figure captions will indicate what types of alkaline ion attachment was involved in providing the ion charge. For any four-carbon ring fragment structures, it is possible to propose alternative structures with two-carbon fragments attached to each terminal of the oligosaccharide. However, the presence of multiple ring fragment structures imply that the oligosaccharides were energetic enough to undergo multiple fragmentation at both termini. Therefore, single ring fragmentation was usually assumed to have occurred, and the presence of multiple ring fragment structures were considered only if no simple single ring fragments could be assigned. IV.3a NOMENCLATURE Although a systematic nomenclature has been proposed recently for higher oligosaccharides (which is similar to the nomenclature for protein fragmentation)187, it can not be used to designate fragmentations of a homoglycan. Therefore, in this thesis, the fragmentation products were designated with a different "cleavage label". The two-carbon fragments were considered to be derived from M-, three-carbon from P- and four-carbon fragments from N-series cleavages. Any subscript designated to a particular series merely serves to denote the sub-group with differing compositions for oxygen and hydrogen atoms. The superscripts °, 2 or 3 denotes none, one, two or three unit of dehydration. - 49 -IV.3b DESORPTION MECHANISM Laser-induced desorption of organic solids has been shown to be essentially a rapid heating effect resulting in flash evaporation of neutrals and ions which subsequently can undergo gas phase ion-molecule r eac t ions 1 5 5 , 1 5 7 , 1 6 7 , 1 8 8 , 1 8 9 . A number of workers have measured the kinetic energy spread of the laser desorbed ions, which supports this "thermal desorption" m o d e l 1 5 5 , 1 8 8 , 1 9 0 , 1 9 1 . The long neutral1 9 2 and i o n 1 5 2 , 1 5 5 , 1 9 2 , 1 9 3 emission time after a short laser pulse also strongly strengthen this thermal model for the desorption mechanism. Cotter et. al. using a 40 ns laser pulse observed the emission life-time of ions to be 1 u.s, while neutrals were still observed after 100 us, which suggested that canonization or deprotonation of the neutrals will be the dominant species observed in the mass spectra192. IV .3C FRAGMENTATION MECHANISM After the initial desorption of the oligosaccharides from the sample layer, fragmentation of the oligosaccharides can occur either in the gas phase or in the desorption plume (also termed as the selvedge) by ion-molecule reactions, or even in the sample layer if excess thermal energy is deposited by the laser pulse. Ionization processes, such as cation ion attachment in the positive-ion mode and deprotonation in the negative-ion mode can occur either before or after the primary and secondary fragmentations. Using repetitive laser desorption mass spectrometry on the disaccharide, sucrose Heresh observed a series of oligosaccharide (2 to 7 hexose units) fragment ions and with both glycosidic and ring cleavages194. The author suggested that the thermal energy deposited by the repetitive laser pulses, caused oligosaccharide polymerization and fragmentation in the sample layer, followed by desorption and ionization in the gas phase. In contrast to Heresch's experiments, all the oligosaccharide mass spectra contained in this thesis were obtained by a single laser pulse on the sample. Therefore, due to the low molecular weight of the oligosaccharides and the relatively small energy transfer (as compared to Heresch's repetively laser experiments), all the samples investigated in this - 50 -thesis could be desorbed into the selvedge as the intact oligosaccharides, where primary and secondary ion-molecule reactions can occur . In this thesis, all the ions observed in mass spectra are "even-electron" species, and are likely to undergo chemistry which is similar in the gaseous and aqueous phases. Consequendy, for some of these frequently observed ring fragments, structures are proposed by analogy to known "even-electron" gas phase fragmentations117 and "even-electron" solution chemistry. An example of this "even-electron" species approach for fragmentation is the glycosidic bond cleavage via protonation of the glycosidic oxygen, to generate the reducing end oligosaccharides, and the non-reducing end oligosaccharide oxonium ions as shown in Fig. rv.l. The proposed reaction scheme shows proton donation by water to a glycosidic oxygen. This is a bimolecular reaction, which could occur in the desorption plume following the laser pulse. Carbohydrates are highly hygroscopic and the sample handling procedure, involving dissolution in water/methanol and introduction via the solid probe followed by evaporation of the solvent, probably leaves sufficient water for the protonation reaction to occur. HO HO OH 2 3 Fig.IV.l . Proposed glycosidic bond cleavage via protonation to give an oxonium ion and a reducing end fragment. If fragmentation is a result of ionization followed by gas phase unimolecular or ion-molecule reactions, then the formation of the ring fragments would be strongly influenced by the first generation ions, and different ring fragments would be observed in the positive- and negative-ion modes. This assumption is supported by the positive- and negative-ion f.a.b.-m.s.-m.s. experiment on a series of underivatized monosaccharides, where the authors observed different ring fragments in the different modes, and they attributed the results to the. different first generation ions obtained under the different modes120. On the other hand, if fragmentation occurs before ionization, then all the ring fragments observed in both the positive- and negative-ion spectra would be the same. In this thesis, the same ring fragments (with the exception of two) were observed in both positive- and negative-ion experiments. This suggests that ionization followed by gas phase ion-molecular reactions may not be the dominant fragmentation process, but rather fragmentation occurs before - 52 -ionization. However, once the desorbed fragments are in the gas phase further fragmentation could occur by ion-molecule reactions. The only two ring fragments that were observed exclusively in the negative-ion mode are: -C3H3O4 and -C3H4O3N. These two ring fragments both contain an a,a-unsaturated system, therefore, deprotonation between the two double bond is very facile, and the deprotonated species would be much more stable than the corresponding cationized species. Hence, they were observed in relatively high abundance in the negative-ion spectra (Sections IV.3.4 & IV.3.6). IV.3d RING FRAGMENTS Due to the lack of literature data on ring-cleavage of underivatized oligosaccharide, the mass spectral data of all the oligosaccharides in this thesis has to be reviewed collectively in order to analyze the ring fragments. A general discussion of the different ring fragments is outlined in the following paragraphs in order to better understand the mass spectral interpretation presented in this thesis. More detailed analysis, and the way in which these ring fragments were used for structural characterization, will be discussed in subsequent sections. The most common ring fragment in both positive- and negative-ion modes is the -C2H3O2 fragment. This two-carbon ring fragment has been observed occasionally in both f.a.b and l.s.i.m.s. spectra 1 3 6 , 1 4 8. The authors attributed this fragment to a C-l,C-2 centre or a C-3.C-4 centre (Fig. rv.2). Therefore these fragments are associated with both the reducing and the non-reducing end oligosaccharides. It should be noted that fragments containing a C-5.C-6 centre will also give a -C2H3O2 ring fragment. Hence, the existence of a -C2H3O2 fragment attached to a non-reducing end oligosaccharide implies that the ring cleaved hexose could be linked via position 3, 4, 5 or 6. Therefore the analytical aspect of the -C2H3O2 ring fragment for deterrnining positions of linkage is not very informative. However, if one of the H's were to be replaced by a substituent, then the two-carbon ring fragment might be useful for a tentative position-of-linkage assignment. In our previous study on - 53 -Klebsiella K44 oligosaccharide (Appendix 1), it was noted that a 2-linked rhamnose residue cleaved to give a -C2H3O ring fragment, which may be interpreted to indicate 2-linked residues would be cleaved to give such a ring fragment. 1 2 ROvrLrE====ayvrOR' 3 4 "RQv/v /^VOQR"' 6 5 "ROx/v- ^ A ^ O H Fig.IV.2. Some ring cleavages giving two-carbon -C2H3O2 fragments common to both positive- and negative-ion modes. Two of the products are of the structural type C2H2O2R2. If one of the R groups is replaced with a H atom, then the fragment formula will be C2H3C>2~R. In this thesis, the three-carbon fragment, -C3H5O3, was also abundant in both the positive- and negative-ion spectra. These -C3H5O3 fragments could contain either a C-l,C-2,C-3 or C-2,C-3,C-4 or C-4,C-5,C-6,0-hemiacetal centre (Fig. IV.3). Fragments derived from the C-1.C-2.C-3 centre imply that the hexose is linked via position 1, 2 or 3 (1/2/3), while fragments derived from the C-2,C-3,C-4 centre imply that the hexose is linked via position 2, 3 or 4, similarly fragments derived from the C-4,C-5,C-6,0-hemiacetal centre imply that the hexose is linked via position 4, 5 or 6. Fragments derived from the C-l,C-2,C-3 and the C-4,C-5,C-6,0-hemiacetal centres imply that the hexose cleaves by breakage of the C-l.O-hemiacetal bond and the C-3.C-4 bond. However, if the first stage of the fragmentation is via glycosidic bond cleavage in the desorption plume, then the oxonium ion formed would have a double bond between C- l and the ring oxygen (Fig. IV. 1, structure 2). In this case, cleavage between the C-l.O-hemiacetal double bond would be unlikely. Therefore, we propose that if gas phase ion-molecule reactions play a prominant role in fragmentation of oligosaccharides, any -C3H5O3 fragments attached to non-reducing end oligosaccharides are derived from the C-2,C-3,C-4 centre, which implies that the cleaved residue is linked via position 2, 3 or 4. If the non-reducing end oligosaccharide were to - 54 -undergo fragmentation to give a -C3H5O3 fragment, then the three-carbon fragment would contain the C-l,C-2,C-3 centre. If fragmentation occurs as a result of thermal pyrolysis in the desorption plume, then the cleavage would be influenced by the conformation of the oligomers in the solid state. Thus any of the three centres could contribute to a -C3H5O3 fragment. A l l the oligosaccharide samples investigated in this thesis have degrees of polymerization lower than five, therefore as a result of the thermal energy deposited by the laser pulse, desorption of the intact oligosaccharides into the gas phase followed by fragmentation is possible. Therefore, we propose that if gas phase ion-molecule reaction plays a prominant role in fragmentation of lower oligosaccharides, any - C 3 H 5 O 3 fragments attached to non-reducing end oligosaccharides are derived from the C-2,C-3,C-4 centre, which implies that the cleaved residue is linked via position 2, 3 or 4. Fig.rv.3. Some ring cleavages giving three-carbon -C3H5O3 fragments common to both positive- and negative ion modes. Two of the products are of the structural type C3H3O3R3. If two of the R groups are replaced with H atoms, then the fragment formula will be C3H5C>3~R. The structural formula of the other product is C3H4O3R2. If one of the R groups is replaced with a H atom, then the fragment formula will be CsHsC^-R. - 55 -In a recent f.a.b.-m.s.-m.s. study of some monosaccharides120, the authors suggested three different fragmentation reactions in the negative-ion mode for the formation of this -C3H5O3 fragment containing the three centres discussed above. These fragmentation pathways all involve the initial deprotonation of the C-2 hydroxyl group. However, this -C3H5C>3 fragment was not observed in their positive-ion f.a.b.-m.s.-m.s. monosaccharide spectra. In contrast the ions containing this -C3H5O3 fragment were relatively abundant in both the positive- and negative-ion l.d.i.-F.t.-i.c.r. spectra obtained in this study, suggesting that fragmentation occurred before ionization. Another fragment seen in both positive- and negative-ion l.d.i.-F.t.-i.c.r. spectra is the -C4H7O3 fragment, and this fragment could only be derived from the C-3,C-4,C-5,C-6 centre of a hexose (Fig. IV.4). ""RO '"RO Fig.lV.4. Ring cleavage giving a four-carbon -C4H7O3 fragment common to both positive-and negative-ion modes. If two of the R groups are replaced with H atoms, the fragment formula will be C4H703~R. Other ring fragments that have been observed in this thesis include: -CH0 2 ; - C 2 H 5 0 2 ; - C 3 H 3 0 4 ; - C 3 H 4 0 3 N ; - C 4 H 7 0 4 ; and - C 4 H 9 0 3 . The possible analytical implications of these ring fragments will be discussed individually in the following sections. The low resolution spectra and the "raw" high resolution tabulated spectra of the various oligosaccharides are listed in Appendix 2. The positive- and negative-ion spectra of each oligosaccharide are discussed in separate sections and accompanied by the "processed" high resolution data tables. These tabulated data include the observed mass (amu), calculated mass, relative intensity (%), mass error (ppm) and proposed carbohydrate structures, but will not include any alkali-halide clusters ions observed in the mass spectra. - 56 -The "whole oligosaccharide" fragments will be briefly mentioned in the text, however, the ring fragments will be discussed in more detail. Some masses could not be assigned to carbohydrate structures. Those with negative mass defect could be assigned to alkali-halide clusters from the KBr/NaCl matrix. Generally, all peaks considered to be signal ions could be assigned structures. In some spectra, a few ions are still to be structurally identified. - 57 -IV.3.1 NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRUM OF Klebsiella K44 DE-0-ACETYLATED OLIGOSACCHARIDE OBTAINED BY BACTERIOPHAGE DEGRADATION OF THE NATIVE CAPSULAR POLYSACCHARIDE This linear pentasaccharide was chosen as the first example to demonstrate the application of l.d.i.-F.t.-i.c.r. to a linear oligosaccharide contain common sugars (Fig. IV.5). It was generated by bacteriophage degradation of the corresponding Klebsiella K44 capsular polysaccharide. The structure of the capsular polysaccharide is made up of a linear pentasaccharide repeating unit, consisting of one p-D-glucosyluronic acid, two a-L-rhamnose, one p-D-glucose and one 6-0-acetylated-a-D-glucose195,196. COOH HO— i HO—. O p-D-GlcA(l->2)-a-L-Rha(l-^3)-a-L-Rha(l-^3)-p-D-Glc(l->4)-D-Glc Fig.lV.5. Structure of Klebsiella K44 de-0-acetylated oligosaccharide obtained by bacteriophage degradation of the capsular polysaccharide. The negative-ion spectrum (Spec. IV. 1 & IV.2) was dominated by KBr cluster-ions, however these inorganic ions were easily identifiable by their negative mass defect. Other than these inorganic ions and two other peaks (m/z 623.170 - 7% and 273.012- 8%), all the peaks above 5% could be assigned to carbohydrate structures (Table IV. 1). Although the spectrum was recalibrated, the mass error of two fragments was still above 50 ppm (m/z 791.319, 5%, 92.0 ppm and 453.134,10%, -60.7 ppm). The low resolution mass spectral data indicated that the sample is a pentasaccharide, consisting of two hexoses (2 x Hex), two deoxyhexose (2 x deoxyHex) and one hexuronic acid (HexA). This pentasaccharide after successive glycosidic bond cleavages gave nine - 58 -different fragment structures: (2deoxyHex)\ (2Hex); (Hex.deoxyHex); (deoxyHex.HexA); (Hex,2deoxyHex); (2deoxyHex,HexA)\ (2Hex,deoxyHex)\ (2Hex,2deoxyHex) and (Hex,2deoxyHex,HexA). These fragments suggested that: i) the two hexose residues are contiguous; ii) the two deoxyhexose residues are adjacent; and iii) the hexuronic acid is a terminal residue and is linked to a deoxyhexose. Thus, the low resolution glycosidic bond cleavage data suggested two possible sequences for this linear pentasaccharide: HexA-»deoxyHex-»deoxyHex—»Hex-»Hex Hex-»Hex-»deoxyHex-»deoxyHex-»HexA Four different types of ring fragment were observed, viz: -C2H3O, -C2H3O2, -G2H5O2 and -C3H5O3. None of these fragments could be used conclusively to locate the reducing end of the oligosaccharide. However, from our previous positive-ion mass spectral data (Appendix 1), it was shown that the reducing residue is a hexose and the sequence is: HexA-»deoxyHex-»deoxyHex-»Hex-»Hex The fragment ion m/z 521.172 (7%) could be written as a deprotonated, di-dehydrated structure of a (2deoxyHex.HexA) unit linked to a -C3Hs0 3 ring fragment. If the first step of fragmentation for this species is via oxonium ion formation, then the -C3H5O3 ring fragment suggests that the cleaved hexose is linked to the deoxyhexose via position 2, 3 or 4. However, none of the other two-carbon ring fragments could be used to determine the positions of linkage for other monosaccharide residues. Therefore, the linkage information obtained from the negative-ion spectral data is: Hex A( 1 ->?)deoxyHex( 1 ->?)deoxyHex( 1 -»2/3/4)Hex( 1 ->?)Hex Combining this information and the information obtained from the positive-ion spectral data detailed in Appendix 1: i) HexA(l-»2)deoxyHex; ii) deoxyHex(l-»3/4/5/6)Hex; and iii) Hex(l-»3/4/5/6)Hex, gives: HexA( 1 -»2)deoxyHex(l ->?)deoxyHex(l -*3/4)Hex( 1 ->3/4/5/6)Hex The most interesting observation for this negative-ion spectrum is that the two-carbon ring fragments seem to reflect the stereochemistry of the anomeric centre of the monosaccharide residues. For the non-reducing oligosaccharide structures, the ring - 59 -fragments correlated with the anomeric configuration of the monosaccharide residue linked to the ring cleaved residue (Fig. IV.6). In this case, a-anomers gave a -C2H5O2 fragment (cleavage from the 3-0-substituted-p-D-glucose residue and the 3-0-substituted-a-L-rhamnose residue), while p-anomers gave a -C2H3O2 fragment (cleavage from the reducing end 4-0-substituted-D-glucose residue and the 2-0-substituted-a-L-rhamnose residue). For the reducing oligosaccharide structures, the ring fragments correlated with the anomeric configuration of the ring cleaved residue itself (Fig. rv.7). In this case, a-anomers gave a - C 2 H 3 O fragment (cleavage from the 2-0-substituted-a-L-rhamnose residue and the 3-0-substituted-a-L-rhamnose residue), while p-anomers gave a -C2H3O2 fragment (cleavage from the non-reducing end p-D-glucose residue and the 3-O-substituted-p-D-glucose residue). It could be argued that the two-carbon fragments shown in Fig. IV.6 and rv.7 may indicate the configuration of the highest-numbered chiral carbon, i.e., a D- or L-series sugar, rather than the the configuration of the anomeric carbon. The fragmentation of a monosaccharide ring, however, is more likely to manifest itself as either a neighbouring-group effect (as observed in Fig. IV.6) or as a representation of the stereochemistry of the fragment centre (as observed in Fig. IV.7), rather than a reflection of the stereochemistry of a relatively distant group, i.e., C-5 in this case. Although the pseudo-molecular ion showed one unit of dehydration, this oligosaccharide does not contain any anhydro-residues. The fragment ions m/z 689.2144 (deprotonated (Hex,2deoxyHex,HexA) unit attached to a -C2H3O2 fragment) and 675.2143 (deprotonated (2Hex,2deoxyHex) unit attached to a -C2H3O2 fragment) indicated that none of the monosaccharide residues are anhydro-derivatives. Therefore, the dehydration occurred during the desorption/ionization process. - 60 -COOH °* O - C 2 H 3 O C 2 H 3 0 2 fragment attached to p-D-glucuronic acid COOH 0 HO/1 ° v O - C 2 H 5 0 C 2 H 5 0 2 fragment attached to a-L-rhamnose OH C 2 H 5 0 2 fragment attached to a-L-rhamnose HO HO-0 x O - C 2 H 3 0 C 2 H 3 0 2 fragment attached to p-D-glucose Fig.IV.6. Structures of the non-reducing end ring cleavage fragments obtained from negative-ion l.d.i.-F.t.-i.c.r. of Klebsiella K44 de-O-acetylated oligosaccharide. Deprotonation of these structures will give the correspon<ling ions. - 61 -p-D-glucose cleaved to give C2H3O2 fragment O H 3 C 2 - O OH a-L-rhamnose cleaved to give C 2 H 3 0 fragment a-L-rhamnose cleaved to give C 2 H 3 0 fragment H 3 C 2 - O OH OH p-D-glucuronic acid cleaved to give C 2 H 3 0 2 fragment Fig.IV.7. Structures of the reducing end ring cleavage fragments obtained from negative-ion l.d.i.-F.t.-i.c.r. of Klebsiella K44 de-O-acetylated oligosaccharide. Deprotonation of these structures will give the corresponding ions. TABLE IV. 1 - 62 -NEGATIVE ION LDI-FT-ICR FRAGMENT STRUCTURES OF KLEBSIELLA K44 DE-0-ACETYLATED OLIGOSACCHARIDE Observed Calculated Relative Mass Proposed mass mass intensity error structure (amu) (amu) (%) (ppm) 791.319 791.246 5 92.0 [(2Hex,2deoxyHex,HexA)-H20-H]~ 689.214 689.214 5 0.6 [(Hex,2deoxyHex,HexA)+C2H302-H]" 675.214 675.235 11 -31.1 [(2Hex,2deoxyHex)+C2H302-H]" 629.170 629.194 7 -37.0 [(Hex,2deoxyHex,HexA)-H20-H]" 623.170 7 ? 615.224 615.214 5 15.9 [(2Hex,2deoxyHex)-H20-H]-555.196 555.193 7 5.4 [(Hex,2deoxyHex)+2C2H302-H]' 529.171 529.177 6 -10.6 [(2deoxyHex,HexA)+C 2H 50 2-H] _ 521.172 521.151 7 39.1 [(2deoxyHex,HexA)+C 3H 50 3-2H 20-H]-513.160 513.183 14 -44.8 [(2Hex ,deoxyHex)+C2H30-H]" 485.154 485.151 15 5.6 [(2deoxyHex,HexA)-H]~ 469.151 469.156 7 -12.2 [(2Hex,deoxyHex)-H 20-H]_ 467.134 467.141 19 -14.8 [(2deoxyHex,HexA)-H20-H]" 453.134 453.162 10 -60.7 [(Hex,2deoxyHex)-H20-H]" 449.113 449.130 6 -38.4 [(2deoxyHex,HexA)-2H20-H]" 383.120 383.119 7 2.4 [(deoxyHex,HexA)+C2H502-H]" 367.124 367.125 16 -2.7 [(2Hex)+C 2H 30-H]' 323.106 323.098 5 24.5 [(2Hex)-H 20-H]~ 321.086 321.083 16 11.5 [(deoxyHex,HexA)-H20-H]~ 309.126 309.119 6 22.0 [(2deoxyHex)-H]-307.108 307.104 17 13.7 [(Hex,deoxyHex)-H20-H]~ 303.078 303.072 8 20.1 [(deoxyHex,HexA)-2H20-H]~ 291.110 291.109 18 4.5 [(2deoxyHex)-H20-H]" 273.012 8 ? 235.042 235.045 9 -14.5 [(HexA)+C 2 H 3 0 2 -H] ' 221.069 221.067 16 11.8 [(Hex)+C 2H 30 2-H]-203.057 203.056 6 5.9 [(Hex)+C 2 H 3 0 2 -H 2 0-H]" 193.036 193.035 12 4.7 [(HexA)-H]-185.045 185.046 12 -2.7 [(Hex)+C 2 H 3 0 2 -2H 2 0-H]-179.059 179.056 11 14.5 [(Hex)-H]-175.028 175.025 9 16.6 [(HexA)-H 2 0-H]" 173.048 173.045 9 16.8 [ (HexA)+C 2 H 3 0 2 -C0 2 -H 2 0-H]-161.046 161.046 15 0.6 [(Hex)-H 2 0-H]-145.052 145.051 21 6.9 [(deoxyHex)-H2Q-H]" - 63 -IV.3.2 POSITIVE- AND NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRA OF MALTOSE AND CELLOBIOSE To confirm the empirical observation of anomeric configuration correlating to mass fragments, the mass spectra of two underivatized disaccharides were obtained. These disaccharides were: maltose [ct-D-glucose (l-»4) D-glucose] and cellobiose [p-D-glucose (l->4) D-glucose] (Fig. IV.8). HO HO Maltose [a-D-Glc (l->4) D-Glc] Fig.IV.8. Structure of the disaccharides: maltose [<x-D-Glc(l-»4)-D-Glc] and cellobiose [p-D-Glc(l-*4)-D-Glc] There were no obvious differences between the positive- and negative-ion spectra of these disaccharides (Spec. IV.3-10; Table rv.2a-d). Although five different ring fragments were observed (-C 2 H 3 0, - C 2 H 3 0 2 , -C3H5O3, - C 4 H 7 0 3 , - C 4 H 7 0 4 ) , none of these fragments was characteristic for either disaccharide. The fragment -C4H7O4 could have three possible structures, and could be linked either to the non-reducing or the reducing end (Fig. IV.9). A possible explanation for the very similar spectra observed, may be due to the thermal model - 64 -of laser desorption. As both disaccharides are of low molecular weight, they can be easily desorbed into the gas phase by the thermal energy deposited by the laser pulse. Due to the similar structures of the disaccharides, they could have similar gas phase unimolecular and ion-molecule reactions, thereby giving identical spectra. OH Fig.IV.9. Some ring cleavages giving R - C 4 H 7 0 4 fragments common to both positive- and negative-ion modes, containing either a C-l ,C-2,C-3,C-4, or a C-3,C-4,C-5,C-6,0-hemiacetal centre. - 65 -TABLE IV.2a POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF MALTOSE Observed Calculated Relative Mass Proposed mass mass intensity error structure (amu) (amu) (%) (ppm) 381.081 381.079 24 5.3 [(2Hex)+K]+ 365.103 365.105 72 -6.8 [(2Hex)-H20+K]+ 325.114 325.113 24 3.3 [(2Hex)-2H20+K]+ 321.055 321.058 16 -11.5 [(2Hex)+C4H704+K]+ 305.088 305.084 33 12.2 [(2Hex)+C4H704+Na]+ 271.085 271.079 12 20.9 [(2Hex)+C 4H 70 3-H 20+Na]+ 265.096 265.092 14 14.1 [(2Hex)+C 4H 70 4-H 20+H]+ 247.084 247.081 13 10.8 [(2Hex)+C 4H 70 4-2H 20+H]+ 229.071 229.071 24 1.0 [(2Hex)+C 4H 70 4-3H 20+H] + 187.061 187.060 17 5.0 [(Hex)+C 2 H 3 0 2 -2H 2 0+H] + 169.051 169.050 11 7.4 [(Hex)+C 2H 30 2-3H 20+H]+ 163.061 163.060 56 4.3 [(Hex)-H20+H]+ TABLE IV .2b NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF MALTOSE Observed Calculated Relative Mass Proposed mass mass intensity error structure (amu) (amu) (%) (ppm) 341.124 341.109 21 43.8 [(2Hex)-H]~ 281.094 281.088 19 22.9 [(Hex)+C 4H 70 4-H]" 263.083 263.077 23 23.2 [(Hex)+C 4 H 7 0 4 -H 2 0-H]-251.083 251.077 25 24.2 [(Hex)+C 3H 50 3-H]" 221.071 221.067 58 21.4 [(Hex)+C 2H 30-H]" 179.060 179.056 100 20.6 [(Hex)-H]" 161.049 161.046 70 20.4 [(Hex)-H 20-H]~ 143.039 143.035 18 24.7 [(Hex)-2H 20-H]~ 119.038 119.035 11 25.3 [ (C 4 H 8 0 4 ) -H] -101.026 101.024 17 14.5 [ ( C 4 H 8 0 4 ) - H 2 0 - H ] -- 66 -TABLE IV.2c POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF CELLOBIOSE Observed Calculated Relative Mass Proposed mass mass intensity error structure (amu) (amu) (%) (ppm) 381.080 381.079 13 0.8 [(2Hex)+K]+ 365.110 365.105 50 13.1 [(2Hex)-H20+K]+ 325.127 325.113 15 44.6 [(2Hex)-2H 20+K] + 305.093 305.084 17 27.9 [(2Hex)+C4H704+Na]+ 271.089 271.079 10 38.4 [(2Hex)+C 4H 70 3-H 20+Na] + 229.074 229.071 9 15.3 [(2Hex)+C4H 70 4-3H 20+H] + 187.062 187.060 9 8.0 [(Hex)+C 2H 30 2-2H 20+H]+ 163.061 163.060 57 4.3 [(Hex)-H20+H]+ TABLE IV.2d NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF CELLOBIOSE Observed Calculated Relative Mass Proposed mass mass intensity error structure (amu) (amu) (%) (ppm) 377.129 377.086 15 115 [(2Hex)+Cl]~ 341.142 341.109 32 95.6 [(2Hex)-H]" 263.091 263.077 36 53.6 [(Hex)+C 4H 704-H 20-H]-221.077 221.067 18 46.6 [(Hex)+C 2H 30-H]~ 179.060 179.056 60 20.7 [(Hex)-HJ-161.049 161.046 100 20.5 [(Hex)-H 20-H]" 143.039 143.035 15 24.5 [(Hex)-2H 20-H]" 101.026 101.024 12 14.8 [ ( C 4 H 8 0 4 ) - H 2 0 - H ] -- 67 -IV.3.3 POSITIVE-ION L.DJ.-F.T.-I.C.R. SPECTRA OF Klebsiella K3 OLIGOSACCHARIDE OBTAINED BY BACTERIOPHAGE DEGRADATION OF THE NATIVE CAPSULAR POLYSACCHARIDE This linear oligosaccharide was chosen as it contains five common sugars and an acid-labile pyruvic acid acetal group (Fig. IV.IO). The structure of the Klebsiella K3 capsular polysaccharide is made up of a "4+1" branched pentasaccharide repeating unit, consisting of one pyruvic acid acetal 4,6-linked to a a-D-mannose, one a-D-galactosyluronic acid, two a-D-mannose and one oc-D-galactose. The branching pyruvylated mannose residue is linked to the branch point galactosyluronic acid residue 1 9 7. The bacteriophage endoglycanase hydrolyze the galactose-galactosyluronic acid glycosidic linkages, therefore, generating a linear pentasaccharide repeating unit. A l l peaks above 10% intensity of the base peak have been assigned to carbohydrate structures (Spec. IV. 11 & IV. 12). Listed in Table IV.3 are the tabulated high resolution mass spectral data for the oligosaccharide. These ions can be divided into two categories: i) fragment ions arising from glycosidic bond cleavages; and ii) fragment ions derived from ring cleavages. As the oligosaccharides obtained from the enzymic procedure were not desalted, multiple peaks were often observed for the same carbohydrate structures, such as m/z 907.202 (potassiated species) and 891.250 (the corresponding sodiated species). The low resolution mass spectral data indicated that the capsular polysaccharide is made up of a pentasaccharide repeating unit, consisting of four hexoses f4 xHex), one hexuronic acid (HexA) and one pyruvic acid acetal (pyr). From the high resolution data, the pseudo-molecular ion m/z 917.257 (11%) can be assigned to a sodiated, dehydrated (4Hex,HexA,pyr) structure. Although the other pseudo-molecular ions (m/z 913.199, 907.202, 891.250) were decarboxylated, they were not dehydrated, thus implying that the sample does not contain any anhydro-saccharides. - 68 -g Fig.IV.lO. Structure of Klebsiella serotype K3 oligosaccharide obtained by bacteriophage degradation of the capsular polysaccharide. - 69 -The pentasaccharide after successive glycosidic bond cleavages, gave seven different fragment structures: (2Hex); (3Hex); (Hex.HexA); (2Hex,HexA); (3Hex,HexA)\ and (3Hex,HexA,pyr). These fragments suggested that: i) the three hexose residues are contiguous; ii) the fourth hexose is a terminal residue and is linked to the other (3Hex) via the hexuronic acid residue; iii) the absence of (2Hex,pyr) & (3Hex,pyr) fragments suggested that the pyruvic acid acetal is located either on the fourth hexose unit or on the hexuronic acid. Thus, the low resolution glycosidic bond cleavage data suggested four possible sequences for this linear pentasaccharide: Hex(pyr)-»HexA-»Hex-*Hex-»Hex Hex-»HexA (pyr) -»Hex-»Hex-»Hex Hex-»Hex-»Hex-»HexA(pyr)-»Hex Hex-»Hex-»Hex-»HexA-»Hex(pyr) Four different types of ring cleavage fragments were observed: -C2H3O2, -C3H5O3, -C4H7O3, and -C4H7O4. These ring fragments can be used to ascertain the location of labile substituents and alternative reducing end structures. The site of pyruvylation was deduced from the ring fragment, m/z 293.090 (C11H17O9), which could be assigned to [(//ex,pyrj+C2H 30 2+H] + . Thus, the pyruvic acid acetal is attached to the terminal hexose residue and not to the hexuronic acid. This further narrowed down the choice of possible sequences to: Hex(pyr)-»HexA-»Hex-»Hex-»Hex Hex-»Hex-*Hex -»Hex A -»Hex(pyr) Based upon the low resolution data, the high resolution spectrum contained two fragment ions, each of which could be assigned to more than one structure. The fragment ion m/z 609.155 (C22H34O17K) could be assigned to: [(3//^)+C 4 H 7 0 4 -2H20+K] + or [(2//ex,//e^)+C4H903-2H20+K]+ or [(2//ex,//exA)+C2H302+C2H30-H20+K]+ or [(2Hex,HexA)+2C2K30+K]+ - 70 -The fragment ion m/z 553.176 (C2oH340i6Na) could be assigned to: [ ( 3 i / « ) + C 2 H 3 0 + N a ] + or [(2Hex)+C4H704+C4H703+Na]+ The -C4H9O3 ring fragment is fully saturated. Since all the other observed ring fragments have one unit of unsaturation, it would seems that structures containing the saturated -C4H9O3 ring fragment are less likely to exist than structures containing unsaturated -C4H7CV As discussed in Section IV.3.3, the approach in this study is to consider that four-carbon ring fragments are more likely than two two-carbon ring fragments in a particular structure. Therefore, it is assumed that structures with the -C 4 H70 4 fragment are more likely, thus the fragment ions m/z 609 and 553 have been assigned to structures containing this fragment. It should be noted though that the structures containing -C2H3O might contribute to the intensity of the fragment m/z 553. Therefore, the proposed structure of the fragment ion m/z 553 is a sodiated (2Hex) unit sandwiched between - C 4 H 7 0 4 and -C4H7C>3 ring fragments. The -C 4 H 7 03 fragment can only be derived from the C-3,C-4,C-5,C-6 centre of a hexose, therefore, the sequence of this m/z 553 fragment is C4H704-f2//cjc>)-C4H703 (Fig. rv . l l ) . The reverse of this sequence would imply a (4Hex) unit, which was not observed in the mass spectrum. HO Fig.IV.ll. Structure of fragment ion C4H704-(2Hex)-C4H703. Sodiation of this structure will give m/z 553.176. - 71 -Therefore, the sequence of the linear repeating pentasaccharide is: Hex(pyr)-»HexA-»Hex-»Hex-»Hex Further analyses of the ring structure also provide some tentative information on the positions of linkage. The fragment ion m/z 801.215 (21%) could be assigned to a sodiated, decarboxylated (3Hex,HexA,pyr) unit linked to a -C3H5O3 ring fragment, implying that the cleaved residue is located at the reducing end, and probably linked via positions 2, 3 or 4. Subsequent, sequential loss of the non-reducing end hexose from this m/z 801 fragment would give fragment ion m/z 613.157 (21%) {[(2Hex,HexA)+C3B.50s+Nsi]+}, and with further loss of the hexuronic acid would give fragment ion m/z 437.130 (25%) {[(2//ex)+C3H503+Na]+}. It could be argued that the fragments m/z 613 and 437 could have been derived from the secondary and tertiary cleavage of a (Hex—>HexA-*Hex->Hex) tetrasaccharide unit. However, the mass spectral data did not show any other fragment ions that would favour one of these possibilities. The reducing end hexose also fragmented to give a -C4H7O3 unit which is derived from the C3-C4-C5-C6 centre (m/z 553.176, 13%). Therefore the position of linkage for the reducing end hexose is either at C-3 or C-4. Another observed four-carbon fragment, -C4H7O4, derived from fragmentation of the galacturonic acid and attached to the reducing (3Hex) unit. However, immediate structural implication could not be drawn from this fragment. Unfortunately, the pentasaccharide did not undergo enough ring fragmentation for all the positions of linkage to be determined. Therefore the only information obtained from the positive-ion l.d.i.-F.t.-i.c.r. spectra for this oligosaccharide is: Hex(pyr)( 1 ->?)HexA( 1 -»?)Hex( 1 -*?)Hex( 1 -»3/4)Hex - 72 -TABLE IV.3 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF Klebsiella K3 OLIGOSACCHARIDE OBTAINED BY PHAGE DEGRADATION Observed Calculated Relative Mass Proposed mass mass intensity error structure amu amu % ppm 917.257 917.238 10 20.2 [(4Hex,HexA,pyr)-H20+Na]+ 913.199 913.241 11 -45.7 [(4Hex,HexA,pyr)-C02-H+2Na]+ 907.202 907.233 29 -33.9 [(4Hex,HexA,pyr)-C02+K]+ 891.250 891.259 57 -9.5 [(4Hex,HexA,pyr)-C02+Na]+ 801.215 801.227 21 -15.6 [(3Hex,HexA,pyr)+C3H503-C02+Na]+ 745.184 745.201 13 -22.7 [(3Hex,HexA)+C2H302+Na]+ 729.204 729.206 24 -2.6 [(3Hex,HexA,pyr)-C02+Na]+ 719.155 719.164 19 -12.5 [(3Hex,HexA)+K]+ 711.188 711.196 20 -10.8 [(3Hex,HexA,pyr)-C02-H20+Na]+ 703.186 703.190 50 -6.1 [(3Hex,HexA)+Na]+ 685.174 685.180 17 -8.3 [(3Hex,HexA)-H20+Na]+ 641.182 641.190 17 -13.3 [(3Hex,HexA)-C02-H20+Na]+ 613.157 613.159 21 -2.5 [(2Hex,HexA)+C3H503+Na]+ 609.155 609.143 27 20.4 [(3Hex)+C 4H 70 4-2H 20+K] + 569.160 569.169 23 -16.3 [(3Hex)+C2H302+Na]+ 553.176 553.174 13 3.7 [(2Hex)+C 4H 70 4+C 4H 70 3+Na] + 549.142 549.140 18 3.2 [(3Hex)-H+2Na]+ 543.143 543.132 15 20.6 [(3Hex)+K]+ 541.137 541.138 22 -0.7 [(2Hex,HexA)+Na]+ 527.158 527.158 100 -0.8 [(3Hex)+Na]+ 523.127 523.127 21 0.0 [(2Hex,HexA)-H20+Na]+ 509.150 509.148 16 4.1 [(3Hex)-H 20+Na] + 479.142 479.137 11 9.9 [(2Hex)+C 2H 30 2+C 3H 50 3+Na]+ 437.130 437.127 25 6.9 [(2Hex)+C 3H 50 3+Na] + 421.095 421.095 26 -1.7 [(Hex,HexA)+C 2H 30 2+Na] + 365.106 365.106 35 2.2 [(2Hex)+Na]+ 363.099 363.092 17 17.6 [(Hex,HexA)+C 2H 30 2-2H 20+H]+ 361.079 361.074 12 13.9 [(Hex,HexA)-H20+Na]+ 347.099 347.095 11 10.9 [(2Hex)-H20+Na]+ 293.090 293.087 11 11.6 [(Hex,pyr)+C 2H 30 2+H]+ 187.063 187.060 15 12.8 [(Hex)+C 2H 30 2-2H 20+H]+ - 73 -IV.3.4 NEGATIVE-ION L.DJ.-F.T.-I.C.R. SPECTRA OF Klebsiella K3 OLIGOSACCHARIDE The negative-ion spectrum of Klebsiella K3 oligosaccharide (Spec. IV. 13 & IV. 14, Table IV.4) confirmed the sequence obtained from the positive-ion spectrum, and in addition, established one other position of linkage. The fragment ion m/z 497.110 (10%) could be assigned to a deprotonated (Hex,HexA,pyr) unit linked to a -C3H5O3 ring fragment, implying that the cleaved residue was attached to the hexuronic acid, and is either linked via position 2,3 or 4. Another three-carbon ring fragment, -C3H3O4, observed in the spectrum could be assigned to a species with two units of unsaturation. The fragment ions m/z 589.166 (6%), 571.142 (16%), and 553.142 (39%) were the deprotonated and the "successive" dehydration products of a (3Hex) unit linked to a -C3H3O4 ring fragment (Fig. TV. 12). This ring fragment was derived from the C-3,C-2,C-l,0-hemiacetal centre of the hexuronic acid. A simple mechanism that can be proposed for this fragmentation is illustrated in Fig. IV. 13. This mechanism involves the formation of an a,a-unsaturated hydroxyl-keto-aldehyde structure at the C-3,C-2,C-l,0-hemiacetal centre. O. \ 3 Q HO OH HO HO Fig.IV.12. Structure of fragment ion C3H304-(3Hex). Deprotonation of this structure will give m/z 589.166. Successive dehydration of this ion gave fragment ions m/z 571.142, and 553.142. - 74 -6 HOOC Fig.IV.13. Deprotonation of C-3 hydroxy hydrogen of the galacturonic acid and subsequent cleavages of the C-3,C-4 and C-5,0 bonds. Other than these two unique fragment ions and a greater abundance of dehydrated ions observed in the negative-ion spectrum, both the positive- and negative-ion spectra were very similar. In conclusion, the positive- and negative-ion l.d.i.-F.t.-i.c.r. spectra provided the following sequence and linkage information: Hex(pyr) (1 ->?)Hex A( 1 ->2/3/4)Hex( 1 -»?)Hex( 1 ->3/4)Hex - 75 -TABLE IV.4 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF Klebsiella K3 OLIGOSACCHARIDE OBTAINED BY PHAGE DEGRADATION Observed Calculated Relative Mass Proposed mass mass intensity error structure amu amu % ppm 867.281 867.262 34 21.1 [(4Hex,HexA,pyr)-C0 2-H]-759.230 759.220 15 13.3 [(3Hex,HexA,pyr)+C3H503-C02-3H20-H]-705.202 705.210 65 -10.4 [(3Hex,HexA,pyr)-C0 2-H]-703.192 703.194 14 -2.1 [(3Hex,HexA)+C 2H 30 2-H 20-H]-687.202 687.199 10 4.9 [(3Hex,HexA,pyr)-C0 2-H 20-H]-679.196 679.194 36 3.3 [(3Hex,HexA)-H]" 661.182 661.183 39 -2.4 [(3Hex,HexA)-H 20-H]-643.179 643.173 6 10.1 [(3Hex,HexA)-2H 20-H]-589.166 589.162 6 5.8 [(3Hex)+C 3H 30 4-H]-585.164 585.167 11 -5.5 [(2Hex,HexA,pyr)+C2H302-C02-H]-571.142 571.152 16 -16.6 [(3Hex)+C 3 H 3 0 4 -H 2 0-H]-559.155 559.152 5 5.6 [(2Hex,HexA)+C 2H 30 2-H]-553.142 553.141 39 0.8 [(3Hex)+C 3 H 3 0 4 -2H 2 0-H]-543.158 543.157 7 1.5 [(2Hex,HexA,pyr)-C0 2-H]-525.148 525.146 13 4.0 [(2Hex,HexA,pyr)-C0 2-H 20-H]-517.138 517.141 61 -5.2 [(2Hex,HexA)-H]-515.123 515.125 5 -5.6 [(Hex,HexA,pyr)+C3H503+H20-H]-499.130 499.131 100 -1.7 [(2Hex,HexA)-H 20-H]-497.110 497.115 10 -9.8 [(Hex,HexA,pyr)+C 3H 50 3-H]-481.122 481.120 51 4.4 [(2Hex,HexA)-2H 20-H]-397.098 397.099 31 -2.7 [(Hex,HexA)+C 2H 30 2-H]-381.105 381.104 48 3.1 [(Hex,HexA,pyr)-C0 2-H]-379.087 379.088 55 -3.4 [(Hex,HexA)+C 2 H 3 0 2 -H 2 0-H]-355.088 355.088 7 -0.8 [(Hex.HexA)-H]-337.077 337.078 29 -2.1 [(Hex,HexA)-H 20-H]-335.099 335.098 8 2.8 [ ( H e x , H e x A ) + C 2 H 3 0 2 - C 0 2 - H 2 0 - H ] -323.100 323.098 6 5.6 [(2Hex)-H 20-H]-321.084 321.083 7 1.6 [(Hex,pyr)+C2H302-H]-319.067 319.067 39 -1.8 [(Hex,HexA)-2H 20-H]-303.073 303.072 5 3.8 [(Hex,pyr)+C 2H 302-H 20-H]-193.036 193.035 8 5.2 [(HexA)-H]-179.057 179.056 6 3.9 [(Hex)-H]" 175.026 175.025 18 4.6 [(HexA)-H 2 0-H]-- 7 6 -IV.3.5 POSITIVE-ION L.D.I .-F .T.-I.C.R. SPECTRA OF E. coli K 9 OLIGOSACCHARIDE OBTAINED BY BACTERIOPHAGE DEGRADATION OF T H E NATIVE CAPSULAR POLYSACCHARIDE This linear tetrasaccharide was chosen as it contains two common sugars, one amino sugar and an acid-labile sialic acid (Fig. IV. 14). It was generated by bacteriophage degradation of the corresponding E. coli K9 capsular polysaccharide. The structure of the capsular polysaccharide is made up of a linear tetrasaccharide repeating unit, consisting of one p-D-galactose, one a-D-galactose, one iV-acetyl p-D-galactosamine and one iV-acetyl neuraminic acid 1 9 8 . C H 2 O H I CHOH COOH HO NAc OH ^ p-D-Gal( 1 -»3)-p-D-GalNAc( 1 ->4)-cc-D-Gal( 1 -»4 ) NeuNAc Fig.lV.14. Structure of Escherichia coli serotype K9 oligosaccharide obtained by bacteriophage degradation of the capsular polysaccharide No pseudo-molecular ions were observed in any of the positive-ion spectra (Spec. IV.15 & rv.16, Table IV.5). They showed only ions corresponding to (2Hex,HexNAc), (Hex,HexNAc) and (HexNAc). Although only limited structural information could be obtained from these spectra, they were consistent with the more detailed analysis obtained from the negative-ion spectrum. - 77 -TABLE IV.5 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF E. coli K9 OLIGOSACCHARIDE OBTAINED BY PHAGE DEGRADATION Observed Calculated Relative Mass Proposed mass mass intensity error structure amu amu % ppm 584.171 584.159 12 20.5 [(2Hex,HexNAc)+K]+ 568.188 568.185 34 5.8 [(2Hex,HexNAc)+Na]+ 550.176 550.174 11 3.6 [(2Hex,HexNAc)-H20+Na]+ 388.120 388.122 100 -5.2 [(Hex,HexNAc)-H20+Na]+ or *[(Hex,HexNAc)+C2H3O2-H20+Na]+ 366.140 366.140 24 0.5 [(Hex,HexNAc)-H20+H]+ or * [ (Hex,HexN Ac)+C 2H 3O 2-H 20+H]+ 204.087 204.087 29 -0.5 [(HexNAc)-H20+H]+ or * [(HexN Ac)+C 2H 3O 2-H 20+Na]+ 186.076 186.076 15 0.5 [(HexNAc)-2H 20+H]+ or *[(HexNAc)+C2H3O2-2H20+H]+ Loss of a kctene from a N-acetylated amino sugar, -C 2 H 2 Q  - 78 -IV.3.6 NEGATIVE-ION L.DJ.-F.T.-I.C.R. SPECTRA OF E. coli K9 OLIGOSACCHARIDE All peaks above 10% intensity of the base peak, with the exception of ten peaks, eight of which are alkali-halide clusters, have been assigned to the corresponding carbohydrate structures (Spec. IV. 17 & IV. 18). The high resolution mass spectral data for the oligosaccharide are listed in Table IV.6. The negative-ion mass spectrum was dominated by deprotonated structures. A number of even mass ions were observed suggesting that the analyte contains an odd number of nitrogen atoms 1 8 6. Although fragments containing one nitrogen atom are easily identified by their even masses, the presence of iV-acetyl sugars in oligosaccharide structures complicates the mass spectral analysis of the sugar. For example, a fragment at m/z 220.083 in the negative-ion spectrum reflects the elemental formula of CsHuOeN, which is either a N-acetamido hexose or a hexosamine attached to a -C2H3O2 ring fragment (Fig. IV. 15). Therefore, extra care has to be taken when analyzing this type of compound. Acetamido hexose Amino hexose attached to a C2H3O2 ring fragment Fig.IV.15. Both HexNAc and HexN-C2Hs02 have the same elemental formula of C8H15O6N The low resolution mass spectral data imply that the capsular polysaccharide is made up of a tetrasaccharide repeating unit, consisting of two hexoses (2 x Hex), one iV-acetamido hexose (HexNAc) and one N-acetyl neuraminic acid (NeuNAc). The three main fragments observed after successive glycosidic bond cleavages were: (2Hex,HexNAc); (Hex,NeuNAc); (Hex,HexNAc). These fragments suggest four possible sequences for this linear tetrasaccharide: - 79 -HexNAc->Hex-»Hex-»NeuNAc NeuNAc->Hex-»Hex-»HexNAc Hex-»HexNAc—»Hex-»NeuNAc NeuNAc-»Hex-»HexNAc-»Hex The fragment ion m/z 170.045 (C7H.8O4N, 38%) in the high resolution data could be a deprotonated, dehydrated C3H4O3N+C4H7O3 structure (Fig. rv.16), implying that a hexose residue is 3-, 4-, 5- or 6-linked to the amino hexose unit. The -C3H4O3N fragment probably has a similar structure to the -C3H3O4 fragment (derived from cleavage of the hexuronic acid residue, Fig. IV. 12) observed in the negative-ion spectra of Klebsiella K3 oligosaccharide. 5 N H 2 Fig.IV.16. Structure of fragment ion C3H4O3N-C4H7O3. Deprotonation and dehydration of this structure will give m/z 170.045. This structure suggested that a hexose is 3/4/5/6 linked to a hexosamine. Fragment ions at m/z 626.182 (13%) indicated that the NeuNAc residue can be located either at the reducing or the non-reducing end of the sequence and is linked via positions 1, 2 or 4. Fragment ions at m/z 497.192 (21%), 467.172 (32%) and 407.153 (11%) are all derived from the same series with successive loss of C-9, C-8 and C-7 from the (NeuNAc) residue. This series demonstrated that the NeuNAc is linked via position 4 to the (Hex,HexNAc) unit. The fragment ion at m/z 395.159 (13%) confirms this inference (Fig. IV. 17). - 80 NH. AcN H 2 N 5 C3H7O3 626.1818 amu 497.1920 amu H 2N. AH502 467.1721 amu RO H 2N. \ £ 4 407.150 amu RO Fig.IV.17. Fragmentation of the R (l->4) NeuNAc unit. These structures implied that the neuraminic acid is located at the reducing end and is 4-linked to a (R=Hex,HexNAc) unit. Therefore, the negative-ion l.d.i.-F.t.-i.c.r. spectrum provided the following sequence and linkage information: Hex( 1 ->?)HexN Ac(l -»3/4/5/6)Hex(l -»4)NeuN Ac The pseudo-molecular ion m/z 817.274 could be assigned to the deprotonated, dehydrated tetrasaccharide. Therefore it is important to distinguish whether this tetrasaccharide contains any anhydro-saccharide. As the fragment ion m/z 264.094 could be - 81 -assigned to the deprotonated, decarboxylated NeuNAc residue, and the fragment ions m/z 497.199, 467.188, 407.167 and 395.167 could be assigned to the series of deprotonated (Hex.HexNAc) unit linked to ring fragments of NeuNAc , it is safe to assume that the anhydro-saccharide in the pseudo-molecular ion is the non-reducing end hexose residue. However, there was no peak that could confirm that the oligosaccharide contains an intrinsic anhydro-hexose or whether dehydration occurred during the desorption/ionization process. - 82 -TABLE IV.6 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF E. coli K9 OLIGOSACCHARIDE OBTAINED BY PHAGE DEGRADATION Observed Calculated Relative Mass Proposed mass mass intensity error structure amu amu % ppm 817.274 817.273 24 0.7 [(2Hex,HexNAc,NeuNAc)-H20-H]" 626.182 626.194 12 -19.2 [(2Hex,HexNAc)+C4H5O4-H20-H]" or *[(2Hex,HexNAc)+C2H3O2+C4H5O4-H20-H]_ 580.163 11 ? 526.178 526.178 20 -0.3 [(2Hex,HexNAc)-H 20-H]" or * [(2Hex,HexN Ac)+C 2 H 3 O 2 -H 2 0-H]" 497.192 497.199 21 -13.8 *[(Hex,HexNAc)+C 7H 1 204N-H]" 496.197 10 ? 484.166 484.167 14 -2.3 * [(2Hex ,HexN Ac)-H 20-H]" 467.172 467.188 32 -34.7 *[(Hex,HexNAc)+C 6Hi 0O 3N-H]" 466.156 466.157 27 -0.5 *[(2Hex,HexNAc)-2H20-H]" 462.111 462.120 12 -19.9 [(Hex,HexNAc)+C 2H302+3 7Cir 460.122 460.123 30 -0.9 [(Hex,HexNAc)+C 2H302+3 5Cir 452.134 452.141 16 -16.6 [(Hex,NeuNAc)-H20-H]~ or * [(Hex,NeuNAc)+C 2H 3O 2-H 20-H]" 424.130 424.146 100 -37.4 [(Hex,HexNAc)+C 2H 30 2-H]' 410.152 410.130 12 51.7 *[(Hex,NeuNAc)-H20-H]" 407.153 407.167 11 -34.8 *[(Hex,HexNAc)+C 4H 6ON-H]" 395.159 395.167 13 -19.9 * [(Hex,HexN Ac)+C 3H 6ON-H]" 364.114 364.125 53 -30.3 [(Hex,HexNAc)-H20-H]~ or *[(Hex,HexNAc)+C 2H 3O 2-H 20-H]" 290.088 290.088 34 0.0 [(NeuNAc)-H 20-H] _ or *[(NeuNAc)+C 2H 3O 2-H 20-H]" 272.075 272.078 15 -10.7 [(NeuNAc)-2H20-H]~ or *[(NeuNAc)+C 2H 3O 2-2H 20-H]" 264.094 264.109 10 -56.1 [(NeuNAc)-C0 2-H]-or *[(NeuNAc)+C 2H 30 2-C02-H]" 262.088 262.093 18 -20.4 [(HexNAc)+C2H 30 2-H] _ 202.071 202.072 15 -6.6 [(HexNAc)-H 20-H]" or * [(HexN Ac)+C 2 H 3 O 2 -H 2 0-H] _ 170.045 170.046 38 -6.5 [(C3H 4O 3N+C4H7O 3)-H 20-H]" * Loss of a ketene from a JV-acetylated sugar, - C 2 H 2 Q - 83 -IV.3.7 POSITIVE-ION L.D.I.-F.T.-I.C.R. SPECTRA OF Klebsiella K22 OLIGOSACCHARIDE OBTAINED BY BACTERIOPHAGE DEGRADATION OF THE NATIVE CAPSULAR POLYSACCHARIDES The structure of Klebsiella K22 capsular polysaccharide is made up of a 2+2 branched tetrasaccharide repeating unit, consisting of one B-D-glucose, one 6-O-acetylated p-D-galactose, one ot-D-glucose and one lactic acid ether-linked to an a-D-glucosyluronic acid 1 9 9, with the p-D-glucose and the 6-0-acetylated p-D-galactose in the main chain. The serotype specific bacteriophage K22 cleaves the main chain galactose-glucose glycosidic linkages, to generate a "3+1" branched structure (Fig. IV. 18). Consequently, this tetrasaccharide was chosen for its interesting "3+1" structure, the base-labile O-acetyl group and the ether-linked lactic acid group. HO—j \ i—OAc OH OH OAc i 6 lac(2-»4)-p-D-GlcA(l->6)-<x-D-Glc(l -»4)-D-Gal 3 t 1 p-D-Glc Fig.IV.18. Structure of Klebsiella serotype K22 oligosaccharide obtained by bacteriophage degradation of the capsular polysaccharide - 84 -The tabulated high resolution mass spectral data for the oligosaccharide are listed in Table IV.7. Al l peaks above 4% of the base peak can be assigned to carbohydrate structures (Spec. IV. 19 & IV.20). The low resolution mass spectral data indicated that the oligosaccharide consists of three hexoses (3 x Hex), one hexuronic acid {HexA), one lactyl group (lac), and one acetyl group {OAc). The tetrasaccharide, after successive glycosidic bond cleavages, gave three different fragment structures: {Hex); {2Hex) and {3Hex). These fragments suggested that: i) the three hexoses are linked together, and ii) the hexuronic acid is a terminal residue. The oligosaccharide generated by bacteriophage degradation of the native capsular polysaccharide have a 3+1 branched structure. Therefore, the low resolution glycosidic bond cleavage data suggested four possible sequences for this tetrasaccharide: Hex —> Hex —> Hex t HexA Hex —» Hex —» Hex t HexA HexA —> Hex —»Hex t Hex Hex -> Hex -» HexA T Hex From the high resolution data, the fragment ion m/z 271.043 (CpHnOsNa) could be assigned to two different sodiated, dehydrated structures: {HexAJac) and {Hex^CiHiO*. The (//ejc)+C3H3C«4 structure implies the formation of an <x,a-unsaturated hydroxyl-keto-aldehyde structure at the C-3,C-2,C-l,0-hemiacetal centre (Fig. IV. 19). This type of ring fragment has only been previously observed in negative-ion spectra but not in the positive-ion spectra (Sections rv\3.4 & rv.3.6). This can be rationized by the fact that the the C-2 hydrogen is acidic; and deprotonation of this a,a-unsaturated three-carbon fragment is facile under the desorption/ionization condition. Therefore, this a,a-unsaturated - 85 -three-carbon fragment will be unstable in the positive-ion mode. With this assumption, the fragment ion m/z 271 was assigned to the sodiated, dehydrated (HexAJac) structure, implying that the lactyl group is attached to the hexuronic acid. O OR - H + O v % ©;V— OR OH Fig.IV.19 Deprotonation of the C-3 hydroxy hydrogen of the glucuronic acid and subsequent cleavage of the C-3.C-4 and the C-5,0-hemiacetal bonds to give a keto-enol three-carbon -C3H3O4 fragment, containing a C-l,C-2,C-3,0-hemiacetal centre. Due to the acidity of the C-2 hydrogen, this three-carbon fragment is unstable in the positive-ion mode. The three structures (Hex.OAc); (//ex)+C2H302; and (HexA,lac)-C02 all share the same elemental composition, C8H14O7. These structures can be sub-units of larger structural fragments, such as fragment ions m/z 569.161 and 407.115. Due to this ambiguity in assignment, there is no direct evidence to locate the acetyl group or to obtain further sequence information on this 3+1 branched oligosaccharide. No fragments, however, could be assigned to a (HexA.OAc) or a (HexA,lac,OAc) structure, suggesting that the O-acetyl group is attached to one of the hexose residues. - 86 -TABLE IV.7 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF Klebsiella K22 OLIGOSACCHARIDE Observed Calculated Relative Mass Proposed mass mass intensity error structure amu amu % ppm 817.224 817.222 4 2.0 [(3Hex,HexA,lac,OAc)+Na]+ 585.143 585.143 29 0.2 [(3Hex,OAc)+K]+ or [(3Hex)+C2H302+K]+ or [(2Hex,HexA,lac)-C02+K]+ 569.161 569.169 100 -14.1 [(3Hex,OAc)+Na]+ or [(3Hex)+C2H302+Na]+ or [(2Hex,HexA,lac)-C02+Na]+ 543.130 543.132 15 -4.6 [(3Hex)+K]+ 527.155 527.158 51 -6.6 [(3Hex)+Na]+ 407.115 407.116 42 -2.0 [(2Hex,OAc)+Na]+ or [(2Hex)+C2H302+Na]+ or [(Hex,HexA,lac)-C02+Na]+ 389.102 389.105 8 -8.1 [(2Hex,OAc)-H20+Na]+ or [(2Hex)+C 2H 30 2-H 20+Na] + or [(Hex,HexA,lac)-C02-H20+Na]+ 365.105 365.106 24 -0.3 [(2Hex)+Na]+ 347.094 347.095 9 -2.0 [(2Hex)-H20+Na]+ 271.043 271.043 13 0.4 [(HexA,lac)-H 20+Na]+ 205.070 205.071 26 2.0 [(Hex,OAc)-H20+H]+ or [(Hex)+C 2H 30 2-H 20+H]+ or [(HexA,lac)-C0 2-H 20+H]+ 203.053 203.053 9 1.0 [(Hex)+Na]+ 187.060 187.060 23 0.0 [(Hex,OAc)-2H20+H]+ or [(Hex)+C 2H 30 2-2H 20+H]+ or [(HexA,lac)-C02-2H20+H]+ 145.049 145.050 4 -2.1 [(Hex)-2H20+H]+ - 87 -IV.3.8 NEGATIVE ION L.DJ.-F.T.-I.C.R. SPECTRA OF Klebsiella K22 OLIGOSACCHARIDE Listed in Table IV.8 are the tabulated high resolution mass spectral data for the oligosaccharide. With the exception of three peaks (m/z 581.138 - 13%, 405.105 - 15%, and 215.032 - 12%) all peaks above 10% can be assigned to carbohydrate structures (Spec. IV.21 & IV.22). The pseudo-molecular ions confirmed the positive-ion data, that the oligosaccharide is made up of three hexoses, one hexuronic acid, one lactyl and one acetyl group. As with the positive-ion mass spectra, the negative-ion mass spectral data of this oligosaccharide are very complicated due to the possible double and triple assignment of a number of fragment ions. This is due to the fact that the structures (HexAJac) and (Hex)+C3rl3C>4 share the same elemental composition (C9H14O9), and the fact that deprotonation of both structures is equally facile in the negative-ion mode. Deprotonation can occur at the C-2 of the -C3H3O4 ring fragment or at the carboxylic group of the hexuronic acid. The exact assignment of any ion containing a -C3H3O4 ring fragment is of importance, as it will prove that the hexuronic acid is not located at the reducing end, thus reducing the number of possible sequences. No extra sequence information could be obtained from the negative-ion spectra. - 88 -TABLE IV.8 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF KLEBSIELLA K22 OLIGOSACCHARIDE Observed Calculated Relative Mass Proposed mass mass intensity error structure (amu) (amu) (%) (ppm) 793.225 793.226 31 -1.0 [(3Hex,HexA,lac,OAc)-H]" 775.208 775.215 10 -9.2 [(3Hex,HexA,lac,OAc)-H20-H]" 751.213 751.215 10 -2.5 [(3Hex,HexA,lac)-H]" 721.190 721.205 16 -20.5 [(3Hex,HexA,OAc)-H]" 615.169 615.178 12 -14.3 [(2Hex,HexA,lac,OAc)+CH0 2-C0 2-H]-or [(2Hex,HexA,lac)+C 2H 30 2+CH0 2-C0 2-H]" 581.138 13 427.110 427.109 100 1.2 [(Hex,HexA,lac)-H]~ or [(2Hex)+C 3H 30 4-H]-409.099 409.099 52 0.7 [(Hex,HexA,lac)-H 20-H]" or [(2Hex)+C 3 H 3 0 4 -H 2 0-H] _ 405.105 15 355.087 355.088 29 -4.8 [(Hex,HexA)-H]~ 349.075 349.078 13 -8.4 [(HexA,lac)+C 4 H 7 0 4 -H 2 0-H]" or [ (Hex)+C 3 H 3 0 4 +C 4 H 7 0 4 -H 2 0-H]" 337.076 337.078 50 -5.9 [(HexA,lac)+C 3H 50 3-H]~ or [(Hex)+C 3 H 3 0 4 +C 3 H 5 0 3 -H]-or [(Hex,HexA)-H 20-H]~ 319.066 319.067 15 -4.4 [(HexA,lac)+C 3 H 5 0 3 -H 2 0-H]' or [ (Hex)+C 3 H 3 0 4 +C 3 H 5 0 3 -H 2 0-H]" or [(Hex,HexA)-2H 20-H]~ 277.056 277.057 11 -3.6 [(HexA)+C 4 H 7 0 4 -H 2 0-H]" 275.077 275.077 22 -1.6 [(HexA,lac)+C 3 H 5 0 3 -C0 2 -H 2 0-H]" or [(Hex,HexA)-C0 2 -2H 2 0-H]" 265.057 265.057 13 0.0 [(HexA.lac)-H]" or [(Hex)+C 3H 30 4-H]" 247.044 247.046 30 -7.7 [(HexA,lac)-H 20-H]' or [ (Hex)+C 3 H 3 0 4 -H 2 0-H] ' 215.032 12 179.055 179.056 10 -6.7 [(Hex)-H]" 175.023 175.025 12 -8.6 [(HexA)-H 2 0-H]" - 89 -In summary, the l.d.i.-F.t-i.c.r. spectra of this 3+1 branched tetrasaccharide suggested four possible sequences A, B, C or D: Hex —> Hex —> Hex t A HexA(lac) Hex —> Hex —> Hex t B HexA(lac) HexA(lac) -> Hex -> Hex T C Hex Hex -» Hex -» HexA(lac) T D Hex Because the structures (HexA,lac) and (Hex) + C$ll304, {Hex,OAc) and (//ex)+C2H302 and (HexA,lac)-C02 share the same elemental formulae, the location of the 0-acetyl group and the exact sequence of this "3+1" branched tetrasaccharide could not be determined. However, indirect evidence suggests that the O-acetyl group is attached to one of the hexose residues. This ambiguous sequence information should not be looked upon as a drawback of the l.d.i.-F.t.-i.c.r. technique, as all single-stage mass spectrometric methods are unable to distinguish between alternative underivatized branched oligosaccharides. The most effective mass spectrometric method for the sequencing of this branched structure is to analyze the permethylated and pertrifluoroacetylated derivatives69. This derivatization approach will encourage oxonium ion formation, which effectively "labels" the non-reducing end residues of the branched oligosaccharide (Fig. rv.20). Permethylation will remove all base-labile substituents, therefore, an ether-linked lactyl group will be stable to permethylation, while an ester-linked lactyl group will be removed. For this oligosaccharide, the lactyl group is ether-linked, therefore the non-reducing end permethylated iHexA,lac) will give an oxonium ion fragment m/z 305. Pertrifluoroacetylation has the added advantage - 90 -of not removing any 0-acyl groups. Therefore, an acetylated residue could be distinguished from a non-acetylated residue due to the differences in the mass increment. 219 amu F i g . I V . 2 0 A non-reducing end permethylated hexose residue will cleave to give an oxonium ion fragment of 219 mass unit. In theory, each of the possible branched sequences has a characteristic permethylated oxonium ion fragment, therefore, the appearance of one of these fragments in the spectra will define the sequence. For example, B will give a (2Hex) fragment m/z 423, D will give a (3Hex) fragment m/z 627, C will give a (Hex,HexA,lac) fragment m/z 509, and A will give a (2Hex,HexA,lac) fragment m/z 713. - 91 -713 amu Hex —» Hex - » Hex t ' HexA(lac) 423 amu Hex -> Hex -]» Hex B t HexA(lac) 509 amu HexA(lac) -> Hex -[» Hex T Hex 627 amu Hex —> Hex -f> HexA(lac) T D Hex Similarly, the pertrifluoroacetylated 3+1 branched tetrasaccharide will also give characteristic pertrifluoroacetylated oxonium ion fragments. A non-reducing end (HexAJac) will give a fragment m/z 428. A non-reducing end (Hex) will give a fragment m/z 547, while a non-reducing end (Hex) substituted with one O-acetyl group will give fragment m/z 493. Therefore, in the case of r^rtrifluoroacetylated sample, B will give a (2Hex) fragment m/z 997, D will give a (3Hex) fragment m/z 1147, C will give a (HexJtiexAJac) fragment m/z 878, and A will give a (2HexJJexAJac) fragment m/z 1328. - 92 -1328 amu Hex -> Hex 4> Hex t HexA (lac) 997 amu Hex -» Hex 4» Hex B t HexA (lac) 878 amu HexA(lac) -» Hex -[» Hex T Hex 1147 amu Hex -> Hex 4» HexA(lac) T D Hex Furthermore, as pertrifluoroacetylation does not remove 0-acetyl groups, any characteristic oxonium ion fragments that are 97 amu lower than expected, implies that the acetylated hexose is part of the oxonium ion. For example, in A , if the O-acetyl group is located on one of the non-reducing end hexoses, then the characteristic oxonium ion fragment will be m/z 1231. Furthermore, a fragment peak at m/z 547 indicates that the 0-acetyl group is on the branching hexose, while a fragment peak at m/z 450 indicates that the <9-acetyl group is on the non-reducing end hexose. As structure C is the real sequence for this 3+1 branched oligosaccharide, the pertrifluoroacetylated oligosaccharide will give the following oxonium fragment ions : m/z 428, 547, 878, indicating that the 0-acetyl group is not attached on any of these residues and, therefore, must be linked to the reducing end hexose. 93 428 878 HexA(lac) -f Hex -|> Hex <- OAc Hex"] 547 Although this Section illustrates that l.d.i.-F.t.-i.c.r. could not sequence this 3+1 underivatized branched tetrasaccharide, sequencing could, in principle, be accomplished with the derivatization methods as described above. - 94 -IV.3.9 STRUCTURAL ASSIGNMENTS FOR SOME COMMERCIAL POLYSACCHARIDES USING L.DJ.-F.T.-I.C.R. Coates and Wilkins have demonstrated the potential of l.d.i.-F.t.-i.c.r. for the analysis of some commercial oligo- and poly-saccharides 1 7 0 , 1 7 1 . They categorized their polysaccharide mass spectral data by the differences in the "masses" of the ring fragments observed. Their results and interpretations are described in Section IV.3.9.A. Based on the l.d.i.-F.t.-i.c.r. studies of the four model underivatized oligosaccharides investigated in this thesis, mechanisms could be proposed for the formation of ring fragments observed by Coates and Wilkins in their polysaccharide mass spectra. These proposed mechanisms are detailed in Section IV.3.9.B. IV.3.9.A SUMMARY OF L.DJ.-F.T.-I.C.R. FRAGMENTATION DATA ON POLYSACCHARIDES OBTAINED BY COATES AND WILKINS. From the low resolution positive-ion l.d.i.-F.t.-i.c.r. mass spectra of nine commercially available polysaccharides171 (locust bean gum, white dextrin, dextran, cellulose, starch, xanthan gum, agarose, agar and chitin) they observed "extensive fragmentation of the saccharide chains, from both within the sugar rings and between them ... . Although similarities are seen in the spectra of some compounds, each displays a characteristic fragmentation pattern." The observed ion masses in their fingerprint mass spectra of polyhexoses can be described by the empirical formula [(162)n + X + K ] + or [(Hex)n + X + K ] + n = l ,2,3, . . . (1) where n corresponds to the number of hexose rings in the fragment ion, X corresponds to the mass of a ring fragment and K corresponds to a potassium ion attachment. For the polyhexoses analyzed, they observed sixteen series of fragment ions, with differing X , each of which was assigned an arbitrary label from A to R (Table IV.9). Fewer ion series were observed in the chitin spectrum than in the polyhexose spectra, primarily only those for which X = 0 (A), 60 (F), 74 (G), and 185. They noted that TABLE IV.9 SUMMARY OF WILKIN'S ION SERIES A-R OF THE FORM [(162)n+X+K]+ FOR THE L.D.I.-F.T.-I.C.R. MASS SPECTRA OF SIX POLYHEXOSES % of relative abundance of series peaks (number of series peaks present) X" 5 C D E F G H J IC L M N P <5 R polyhexose [0)h [18] [24] [42] [44] [60] [74] [84] [88] [90] [102] [104] [120] [126] [144] [148] locust bean guma 48 (4) 10(2) 4(2) 6(2) 15(3) 18(4) white dextran 27(5) 5 (3) 4 (4) 5 (4) 46 (7) 5 (4) 4 (3) 5 (4) dextran 15(5) 4(2) 65 07) 8(4) 9(5) cellulose 26(6) 10 (6) 47 (6) 3 (3) 5 (4) 6 (4) starch 19 (6) 6 (6) 31 (7) 8 (4) 24 (6) 11 (5) xanthan gum 6 (2) 10 (4) 18 (5) 31 (5) 7 (3) 14 (4) 5 (3) 10 (4) "The entries in this table can be explained by using locust bean as an example. There are 4 peaks in its mass spectrum whose masses fit the A series pattern [(162)„ +X + K]+. Thus, the number in parentheses in the A column entry is 4. The first number in this entry, 48, is obtained by summing the intensities of these peaks and dividing that sum by the sum of the intensities of all the ion peaks in the spectrum. The other entries in this table are obatined in the same way. bNumbers in brackets correspond to Xin[(162)n+X + K] +. - 96 -"The series for which X = 185 is equivalent to the polyhexose series for which X = 144, shifted in mass reflect the acetylated amine substituent on carbon 2. The fact that the series for which X = 60 and X = 74 do not shift establishes that carbon 2 is not present in those fragments after cleavage across the sugar ring." IV.3.9.B PROPOSED STRUCTURES AND REACTIONS PATHWAYS OF L.D.I.-F.T.-I.C.R. DATA ON POLYSACCHARIDES OBTAINED BY COATES AND WILKINS. The most common ion series were A (X = 0), D (X = 42), F (X = 60), and Q (X = 144), with the A and F series ions being the most intense. These four series were observed in each of the mass spectra of the polysaccharides reported and thus are likely to be fragmentation patterns characteristic of hexose rings, rather than unique fingerprints of the individual polysaccharides. A l l the other ion series are probably derived from specific ring cleavages of each individual polysaccharide. When analyzing these data, comparison should be made between the spectra of similar polysaccharides, e.g., cellulose and chitin. Cellulose being the polymer of 1,4-linked p-D-glucose residues and chitin is poly- p-l,4-linked iV-acetyl D-glucosamine. Some insight into the chemical origin of the ions in the Coates-Wilkins A-R series is provided by writing their empirical formula (1) in a different form. For any polyhexose, their molecular formula can be written as [(162)„ + 18] (2) where n is the degree of polymerization, 162 being the molecular weight of a C6H10O5 anhydro-hexose residue and 18 is the molecular weight of water. Successive loss of hexose residues in the mass spectrum of this polysaccharide would give rise to fragment ions whose formula is given by [(162)n + 18], n = n-1, n-2, n-3, ... (3) Loss of water from these ions would give rise to ions of the formula [(162)n + 18 - (18)m], n = n, n-1, n-2 , . . . , m = 1, 2, 3,... (4) Ring cleaved oligohexoses in the mass spectrum would give rise to ions of the formula - 97 -[(162)n + 18 - (18)m + Y] n = n-1, n-2, n-3,..., m = 0, 1, 2,... (5) where Y is the mass of the ring fragment. Similarly, for any non-polyhexose homoglycan, the mass number, 162, would be replaced by the molecular weight of the anhydro-residue which makes up the homoglycan. Chitin, which is a poly- N-acetylated glucosamine would have a mass number of 203. Similarly, for a linear heteroglycan composed of two different saccharide residues alternately linked, the mass spectrum should reflect this alternating sequence. Agarose which is composed of galactose ( C 6 H 1 0 O 5 , 162) and 3,6-anhydrogalactose (C6H8O4, 144) alternating in the chain, would reflect the following formula in the mass spectrum [(162)n + (144)p + 18 - (18)m + Y + Z] p = n-1, n, n+1 (6) where n and p are the number of galactose and 3,6-anhydrogalactose residues in the chain respectively, m is the degree of dehydration, Y is the mass of the ring fragment derived from galactose and Z is the mass of the ring fragment derived from 3,6-anhydrogalactose. Applying the ring fragment data obtained from the high resolution positive- and negative-ion l.d.i.-F.t.-i.c.r. experiments performed on model underivatized oligosaccharides and the known primary and tertiary molecular structure of the commercial polysaccharides, plausible mechanisms and corresponding. structures are proposed for the reactants and products of fourteen of the sixteen ion series observed by Coates and Wilkins. These fourteen proposed structures/reactions are for the most frequently occurring high intensity 117 peaks. In order to rnmimize confusion between the classical Kochetkov and Chizhov A to K ion series on permethylated glycosides and the Coates-Wilkins series, in this thesis these latter new ion series are labelled from L onwards. Following Kochetkov's convention, we label glycosidic bond cleavage of the saccharide chain the L-type fragmentation. Each ion in the Coates-Wilkins B (X = 18) series fits formula (3) and corresponds to the L° fragmentation route shown in Fig. IV. 1. The superscript 0 denotes no loss of a water molecule from the oligosaccharide (superscript \ 2 or 3 implies one, two or three unit of dehydration respectively). It should be noted that there are two ways in which the L° cleavage can take place. One way places the ion charge on the - 98 -oligosaccharide at the non-reducing end, and the other places the charge on the reducing end oligomer. For any polyhexose, these two fragmentation paths give products which cannot be distinguished from each other. Therefore, in order to deterrnine the reducing end of the sequence of any oligosaccharide either i) the oligosaccharide would have to be "labelled" by chemical means, or ii) the oligomers would have to contain at least two different classes of monosaccharides at different locations. Furthermore, there must also be an abundance of ring cleavages to distinguish the reducing and non-reducing end residues. Carbohydrates are easily dehydrated in the mass spectrometer, and the L 1 , L 2 and L 3 fragmentations are derived from the L° fragmentations by successive loss of water molecules. L 1 , L 2 and L 3 fragmentations give rise to Coates-Wilkins A (X = 0), Q (X = 144) and P (X = 126) series ions respectively, which fit formula (4). Results reported in this thesis indicate that unsaturated two-carbon fragments are the most abundant ring fragments. The reaction pathways giving two-carbon ring fragments are designated as the M-type fragmentation. The positive-ion spectrum of Klebsiella K44 oligosaccharide showed two different fragment structures derived from this M fragmentation. Namely, a 2-substituted rhamnose attached to the reducing oligosaccharide gave an unsaturated - C 2 H 3 0 ring fragment, while non 2-substituted monosaccharides gave rise to an unsaturated - C 2 H 3 0 2 ring fragment. The fragmentation forming - C 2 H 3 0 is labeled the Mi reaction. This could be visualized by the loss of the glycosidic oxygen linked at position-2 of the saccharide (Fig. rv.21). Similarly, it stands to reason that for a - C 2 H 3 0 ring fragment attached to a non-reducing oligosaccharide, either position-3 or position-4 would be substituted (Fig. IV.22). The Coates-Wilkins E (X = 44) series ions could be formed via an Mi fragmentation. These E series ions were of low intensity and were only observed in white dextrin. White dextrin is derived from starch, a poly- 1,4- and 1,6-linked a-D-glucose, by acid/heat treatment. It is known that this treatment causes rearrangements, and the possibility of some 2-, 3- or 4-linked units in white dextrin is conceivable200. HO OR" HO OH Fig.IV.20 Cleavage of a 2-substituted non-reducing residue to give a - C 2 H 3 O fragment containing a C-l.C-2 centre. HO OH "RO OH CH 2 OH CH 2 OH OR' HO OH O OH Fig.IV.22 Cleavage of a 3- and 4-substituted reducing residue to give -C2H3O fragments containing the C-3.C-4 centre. For glycosides that are not 2-linked, the formula for an unsaturated fragment from ring cleavage is - C 2 H 3 0 2 (Fig. rv.2). This ring cleavage is termed the M 2 reaction. The M 2 0 fragmentation gives rise to the Coates-Wilkins F (X = 60) series ions. Successive loss of water molecules from the M 2 0 fragmentation products gives the M 2 1 and the M 2 2 products which correspond to Coates-Wilkins D series (X= 42) and C series (X = 24) ions respectively. The Coates-Wilkins G series (X = 74) was observed only in cellulose (poly- 1,4-linked P-D-glucose) and chitin (poly- 1,4-linked p-D-iV-acetyl glucosamine) spectra. This 74 mass unit represents a three-carbon fragment with one unit of unsaturation. As Coates pointed - 100 -out, the fact that this mass unit did not shift in the chitin spectrum suggested that C-2 was not present in the fragment. A probable structure of this three-carbon fragment is -C3H5O2, containing the C-4,C-5,C-6 centre. This ring cleavage reaction is termed the Pi-reaction. A possible reaction pathway is illustrated in Fig. IV.23. Fig.IV.23 Proposed reaction pathway for cellulose and chitin to give -C3H5O2 ring fragment containing a C-4,C-5,C-6 centre. The substituent X on C-2 can be a hydroxy (cellulose) or an acetamido (chitin) group. The Coates-Wilkins K series (X = 90, average intensity 8%) was observed only in dextran (poly- 1,6-linked a-D-glucose) spectra. This 90 mass unit could represent a -C3H5O3 ring fragment containing either a C-l,C-2,C-3 or a C-4,C-5,C-6,0-hemiacetal centre. As suggested by Heresch 1 9 4 in his repetitive laser experiments on sucrose, the fragmentation could have occurred in the solid. Dextran being a highly polymerized carbohydrate material, would have fragmentation occurring more readily in the solid rather than in the gas phase, thus giving a - C 3 H 5 O 3 ring fragment containing a C-4,C-5,C-6,0-hemiacetal centre, linked to the non-reducing oligosaccharides. This ring cleavage is termed the P2-reaction. A possible pathway is illustrated in Fig. IV.24. Fig.IV.24 Proposed reaction pathway for dextran (poly- a- 1,6-linked glucose) to give - C 3 H 5 O 3 ring fragments containing either a C-l,C-2,C-3 or a C-4,C-5,C-6,0-hemiacetal centre. The Coates-Wilkins J series (X = 88, average intensity 8%) was observed only in the starch (poly- 1,4-linked a-D-glucose) spectrum. This 88 mass unit could represent two two-carbon ring fragments attached to the oligosaccharides. With the following assumptions that: i) ring cleavages of a-anomers give -C2H3O fragments and residues linked to a-anomers give -C2H5O2 fragments (negative-ion data on Klebsiella K44 oligosaccharide); and ii) fragmentation occurs before ionization, this 88 mass unit could be interpreted as "whole oligosaccharide" structures associated with a - C 2 H 3 0 (Mi reactions) fragment attached to the non-reducing end residue (derived from cleavages of a a-glucose unit) and a -C2Hs0 2 ( M 3 reactions) fragment attached to the reducing end residue (derived from cleavages of a glucose residue linked to a-glucose residue) ring fragments. A possible pathway giving two ring fragments is illustrated in Fig. rv.25. Fig.IV.25 Double ring cleavages of starch (poly- 1,4-linked a-D-glucose) to give a -C2H3O (derived from cleavages of a-glucose residue) and a -C2H5O2 (derived from cleavages of glucose unit linked to a-glucose residue) ring fragments. The Coates-Wilkins L series (X = 102 ) was observed in the spectra of white dextrin (average intensity 5%), cellulose (average intensity 5%) and xanthan gum (average intensity 14%). Xanthan gum could be considered as cellulose, substituted on alternating glucose units by a trisaccharide composed of two mannose residues, one glucuronic acid residue, one O-acetate and one pyruvic acid acetal201. This 102 mass unit could represent a four-carbon ring fragment with two units of unsaturation, or two two-carbon ring fragments at both ends. Assuming that: i) ring cleavages of both p-anomers and residues linked to p-anomers give - C 2 H 3 O 2 fragments (negative-ion data on Klebsiella K44 oligosaccharide); ii) fragmentation occurs before ionization; and iii) xanthan gum gives a cellulose backbone after the initially loss of the side chain, this 102 mass unit could be interpreted to give two -C2H3C>2 ring fragments derived from cleavages of p-glucose and glucose residues linked to p-glucose ( M 2 reactions). A possible pathway is illustrated in Fig. IV.26. This simplistic "p-anomers" argument could not be applied to white dextrin, whose exact structure is not known, as extensive rearrangement could have occurred during its formation by acid treatment of starch. - 103 HO OH HO Fig.IV.26 Double ring cleavages of cellulose (poly- 1,4-linked p-D-glucose) to give two -C2H3O2 ring fragments derived from cleavages of p-glucose and glucose linked to p-glucose. The Coates-Wilkins M series (X = 104, average intensity 24%) was observed only in the starch (poly- 1,4-linked a-D-glucose) spectrum. This 104 mass unit could be interpreted as a -C4H7O3 ring fragments containing a C-3,C-4,C-5,C-6 centre (Fig. IV.4). This ring cleavage reaction is termed the N2 fragmentation. The Coates-Wilkins N series (X = 120, average intensity 4%) was observed only in the white dextrin spectrum. This 120 mass unit could be interpreted as a -C4H7O4 ring fragment (as illustrated in Fig. IV.9). This ring cleavage reaction is termed the N3 fragmentation. No structures could be proposed for the other two Coates-Wilkins series: H (X = 84) and R (X = 148). A summary of the proposed interpretation of Coates-Wilkins l.d.i.-F.t.-i.c.r. mass spectral data is listed in Table. IV. 10. It should be noted that while a particular ion formula in the Coates-Wilkins scheme allows its formation to be assigned to a specific reaction type, it is understood that the last reaction forming the ion does not necessarily have to be of that type. For example, consider an ion in the Coates-Wilkins C series. As argued above, the formation of this ion can be assigned to a reaction of the M 2 2 type in which the ion is formed by a type M 2 cleavage with additional loss of two water molecules. The last reaction forming this ion could have been a type M 2 2 fragmentation, or it could have been a type L 1 fragmentation preceded by a type M 2 1 fragmentation. Similarly, an ion in the Coates-Wilkins P series (X = 126) could have been formed by a L 3 reaction or three successive L 1 reactions. Usually, several combinations of reactions can be written to account for the formation of any particular fragment ion. - 104 -M.s.-m.s. or other experiments, which delineate fragmentation pathways, will be needed to ascertain the precise sequence of reactions leading to each of the observed fragment ions. TABLE IV. 10 PROPOSED STRUCTURES AND LABELS, FOLLOWING KOCHETKOVS CONVENTION Proposed label Coates- Mass of the Structure Polysaccharide For proposed following Wilkins sub-monosaccharide of the in which fragmentation Kochetkov's series fragment, X ring fragments the ring fragment reaction and convention is prominant structure see Fig. L 1 A 0 Glycosidic bond cleavage All polysaccharides rv.i L° B 18 Glycosidic bond cleavage All polysaccharides rv.i M 2 2 C 24 R~C2H302 Xanthan gum (10%) rv.2 M 2 * D 42 R~C 2 H 3 0 2 All polysaccharides rv.2 M i 0 E 44 R~C 2H 30 White dextrin ( 5 % ) rv.22 M 2 ° F 60 R~C 2 H 3 0 2 All polysaccharides rv.2 Pl° G 74 R - C 3 H 5 O 2 Cellulose (3%) F/23 Chitin (10%) H 84 7 Xanthan gum (7%) Mi°,M 3 ° J 88 OH3C2-R-C2H502 Starch (8%) rv.25 P2° K 90 R~C 3 H 5 0 3 Dextran (8%) rV.3 JV.24 M 2 ° ,M 2 ° L 102 0 2H 3C2-R-C2H 30 2 White dextrin ( 5 % ) IV.26 Cellulose ( 5 % ) Xanthan gum (14%) N 2 ° M 104 R-C4H 70 3 Starch (24%) rv.4 N 3 ° N 120 R~C4H704 White dextrin (4%) rv.9 L 3 P 126 Glycosidic bond cleavage All polysaccharides rv.i L 2 Q 144 Glycosidic bond cleavage All polysaccharides IV. 1 R 148 7 Starch (11%) - 106 -IV.3.10 NEGATIVE-ION L.D.I.-F.T.-I.C.R. SPECTRA OF E. coli K49 OLIGOSACCHARIDE OBTAINED BY ANHYDROUS HYDROGEN FLUORIDE HYDROLYSIS OF THE NATIVE CAPSULAR POLYSACCHARIDE Anhydrous hydrogen fluoride degradation of capsular polysaccharides under sub-ambient temperatures (-40 to 0°C) is a selective chemical degradation method that cleaves predominantly only one specific type of linkage 3 8 ' 2 0 2" 2 0 5. Therefore this method is analogous to bacteriophage degradation, which leads to the isolation of oligosaccharide repeating unit. Due to differences in the behavior of different linkages to different specific reagents, the CPS will cleave at different sites, thus generating different repeating oligosaccharides. A number of specific degradation techniques were employed by Dr. L . M . Beynon in the structural characterization of the capsular polysaccharide from E. coli K49 bacteria. The particular oligosaccharide investigated in this Section was obtained by hydrolyzing E. coli K49 capsular polysaccharide with anhydrous hydrogen fluoride. The reaction was quenched by addition of methanol and the volatiles removed under nitrogen. The oligosaccharide thus obtained would have methyl esters of any free acidic groups (e.g., a hexuronic acid will give a methyl hexuronate) and a methyl glycoside at the reducing end of the newly generated oligosaccharide derived from the non-reducing end (Fig. 1Y.27). - 107 -CH 2 OH COOH Fig.lV.27 Anhydrous hydrogen fluoride hydrolysis (methanol quenched) of a capsular polysaccharide to give a methyl glycoside and an oligosaccharide with a methyl hexuronate ester. Different hydrolytic conditions will give different oligosaccharides. In this diagram, the hydrolysis occurs between the galactose and glucuronic acid residues. As discussed in Section ELI, all capsular polysaccharides are acidic, since they contain acidic sugar residues and/or acidic non-carbohydrate substituents. Although the most commonly found acidic sugars in CPS are hexuronic acid residues (HexA), some structures have been shown to contain either 2-keto-3-deoxy-manno-octulosonic acid (Kdo, CgHnOg) 2 0 6 " 2 1 4 oriV-acetylated neuraminic acid (NeuNAc, C 1 1 H 1 9 O 9 N ) 2 1 5 ' 2 1 8 . A few CPS lack these acidic sugar residues, and their acidity is conferred by non-carbohydrate acidic substituents, such as pyruvic acid 2 1 9 or phosphates220"225. As most CPS are made up of simple sugars, the first approach in analyzing the mass spectra of E. coli K49 oligosaccharide - 108 -is to presume that the oligosaccharide repeating unit contains an acidic sugar residue, rather than an acidic non-carbohydrate substituent. The second approach is to assume that the oligosaccharide lacks any acidic monosaccharide residues but the acidity is conferred by phosphate groups. If however, both of these approaches do not give reasonable oligosaccharide structures, then, a pyruvic acid residue will be considered to be the sole acidic component. Al l these alternatives assume that the acidic component is present as the methyl ester. Therefore, the true molecular weight of the non esterified oligosaccharide would be at least 28 mass units lower than the pseudo-molecular ion (loss of CH2 from the C- l of the methyl glycoside and the other CH2 from the methyl ester of the acidic component). The negative-ion spectrum of E. coli K49 oligosaccharide (Spec. IV.23 and IV.24) showed a number of both even and odd mass peaks in the low mass region and a number of odd masses in the high mass region, suggesting that the oligosaccharide contains two nitrogen atoms183. One of the nitrogen atoms could be accounted for by an acetamido hexose adjacent to a hexose residue (m/z 364.123). However, none of the other peaks were easily identified. The high mass ions at m/z 835.280, 821.268 and 791.246 could be deprotonated pseudo-molecular ions. The ion at m/z 835 was assumed to be the deprotonated molecular ion, suggesting that the true molecular weight of the non-esterified oligosaccharide is 808 (836-28) amu. With these assumptions, the fragment ion m/z 821 was derived from m/z 835 by loss of a CH2 group from either the C- l of the reducing methyl glycoside or the methyl ester of the acidic component Similarly, the fragment ion m/z 791 was derived from m/z 835 by loss of a molecule of carbon dioxide, suggesting that the oligosaccharide contains a carboxylic acid group. This carboxylic group is likely to be derived from C-6 of an acidic monosaccharide residue, implying that these groups are present rather than an acidic non-carbohydrate substituent, such as phosphate. However, methanol treatment would not leave any free carboxylic acid group in the oligosaccharide. Therefore, loss of a molecule of carbon dioxide from the oligosaccharide implies that either i) the ion m/z 835 is not the true deprotonated molecular ion, or ii) the analyte was not pure and contained more than one oligosaccharide. - 109 -Loss of a hexose residue from the ion m/z 835 will give a fragment ion at m/z 673.247, while loss of a hexose and a CH2 unit will give a fragment ion at m/z 659.200. The existence of these two fragments showed that the non-reducing sugar is a hexose residue and that it does not carry any non-carbohydrate substituents. Similarly, loss of a hexose residue from the fragment ion m/z 791 will give m/z 629.214. It is interesting to note that all these six high mass ions (m/z 835, 821, 791, 673, 659 and 629) could be closely correlated with each other. This suggests that if the analyte were not pure, the different oligosaccharides must have similar structures. As a result, the following assumptions were made in order to analyze the spectra: i) the oligosaccharide contains at least one hexose linked to an acetamido hexose; ii) the non-reducing end sugar is a hexose residue; iii) any non-carbohydrate substituents present are not linked to the non-reducing end hexose; iv) the oligosaccharide contains two nitrogen atoms; v) the oligosaccharide consists of mostly common 'neutral' monosaccharide residues (hexose, deoxyhexose and pentose); vi) the oligosaccharide contains one acidic monosaccharide residue (HexA, Kdo, NeuNAc); vii) the oligosaccharide contains two -OMe groups attached to the C - l of the reducing residue and to the carboxylate group of the acidic sugar; viii) the possible true molecular weight of one of the non-esterified oligosaccharides is 808 amu; ix) in addition to the one hexose and one acetamido hexose, the other component in the oligosaccharide must have one nitrogen atom. However, no assumptions were made as to whether the oligosaccharide was linear or branched. An oligosaccharide with a molecular weight of 808 amu is likely to be a tetra- or penta-saccharide structure, consisting of one acidic sugar, one hexose, one acetamido hexose and one or two unknown components. The molecular weight of the anhydro disaccharide (Hex,HexNAc) is 365, and therefore, the molecular weight of the unknown components is 443 (808-365) amu. The oligosaccharide could be represented by the following structure: Hex + HexNAc + X + Y + Z where X is the acidic monosaccharide residue, and Y and Z are the other components - 110 -This 443 amu unit must contain an acidic monosaccharide residue, e.g., HexA, Kdo or NeuNAc. The molecular weight of an anhydro NeuNAc residue is 291 amu, which means that the fourth component has a molecular weight of 152 (443-291) amu. As NeuNAc contains one nitrogen atom, this 152 amu component would not contain any nitrogen. No common monosaccharide has a molecular weight of 152. Therefore, it seems unlikely that the analyte contains a NeuNAc residue. Similarly, the molecular weight of an anhydro Kdo residue is 220 amu, which implies that the other component has a molecular weight of 223 (443-220) amu. As Kdo contains no nitrogen atom, this 223 amu component must contain one nitrogen atom. No known acetamido sugars have a molecular weight of 223 amu. The three most common 'normal' sugars are hexose, deoxyhexose and pentose. The molecular weights of their anhydro residues are 162 amu, 146 amu and 132 amu respectively. Therefore, the possible molecular weights of the nitrogen-containing components are 61 (223-162) amu, 77 (223-146) amu or 91 (223-132) amu respectively. If the bacteria incorporate any non-carbohydrate substituents into their capsular polysaccharide, then the substituent should be readily available from the food source. Amino acids, being the most readily available nitrogen compounds are the most likely non-carbohydrate substituents containing nitrogen. Furthermore, E. coli capsular polysaccharides have been shown to contain amino acids . No naturally occurring amino acids have molecular weight of 61, 77, or 91 amu. Therefore, it is unlikely that the analyte contains a Kdo residue. The molecular weight of an anhydro HexA residue is 176 amu, which means that the molecular weight of the nitrogen-containing component is 267 (443-176) amu. No known acetamido sugars have this mass. The three most common 'normal' sugars are hexose (162 amu), deoxyhexose (146 amu) and pentose (132 amu). Therefore, the possible molecular weights of the nitrogen-containing components are 105 (267-162) amu, 121 (267-146) amu or 135 (267-132) amu respectively. No naturally occurring amino acids have a molecular weight of 135 amu. However, 105 amu is characteristic of the hydroxyamino acid serine, - I l l -H2NCH(CH20H)COOH; while 121 amu suggests the presence of the thioamino acid cysteine, H 2 NCH(CH 2 SH)COOH. The above analysis gives two possible structures for the oligosaccharide. One structure is (2Hex,HexA.HexNAc,2OMe) substituted with serine (C3iH5 2024N2, 836.2911 amu) and the other is (Hex,deoxyHex,HexA,HexNAc,20Me) substituted with cysteine (C31H52O22N2S, 836.2733 amu). None of the known E. coli capsular polysaccharides have been shown to contain cysteine residues However, some have been found to contain serine, and some of these CPS containing serine also showed partial incorporation of threonine, H 2 N C H ( C H 3 C H O H ) C O O H 2 0 7 " 2 0 8 . This partial incorporation of both serine and threonine residues would explain the inconsistency observed in the higher mass peaks (m/z 791 and 629), as the fragment ions containing the different amino acid residues would be fourteen mass units apart. However, the tetrasaccharide substituted with cysteine cannot explain the high mass inconsistency, and is unlikely to be the true structure of the analyte, and thus is discounted. There are five linear, nine 3+1 branched and two 2+i+l branched carbohydrate structures possible for this 'serinated' tetrasaccharide: Hex —» HexNAc —> HexA —> Hex ~ OMe Hex -» HexNAc - » Hex -+ HexA ~ OMe Hex —» HexA -» HexNAc - » Hex ~ OMe Hex —» HexA - » Hex —> HexNAc ~ OMe Hex —» Hex - » HexNAc —> HexA ~ OMe HexNAc - » Hex (Hex) HexA ~ OMe HexNAc - » Hex - » HexA (Hex) ~ OMe Hex -> HexNAc (Hex) - » HexA ~ OMe Hex -> HexNAc -> HexA (Hex) ~ OMe Hex - » HexA - » HexNAc (Hex) ~ OMe HexA -» HexNAc (Hex) Hex ~ OMe HexA -> HexNAc Hex (Hex) ~ OMe - 1 1 2 -HexA -» Hex (Hex) -> HexNAc ~ OMe HexA -» Hex -» HexNAc (Hex) ~ OMe Hex -» HexNAc (HexA) (Hex) ~ OMe HexNAc -> Hex (HexA) (Hex) ~ OMe Al l structures would have a 'free' hexose residue located at the non-reducing end. The serine (Ser) or threonine (Thr) residue could be located on any of the HexNAc or HexA or the non-reducing end Hex residues. However, there were no masses that could be assigned to serine or threonine linked to either HexNAc or Hex or HexA. Furthermore, there is no indication of how the amino acid residues are linked to the carbohydrate backbone. The amino acid could be linked via an amide bond through the N-terminal to the hexuronic acid, or an ester bond through the hydroxyl group to the hexuronic acid or through the C-terminal to any hexose residues (Fig. rv.28). Amide bond through Ester bond through Ester bond through the N-terminal to the the hydroxyl group the C-terminal to C-2 hexuronic acid to the hexuronic acid of an hexose Fig.IV.28 The serine residue could be linked to the carbohydrate backbone through any of the above linkages. The ester bond through the C-terminal was arbitrarily drawn at the C-2 of an hexose. None of the hydroxyl groups on the hexose ring are shown, due to the unknown positions of linkage. The carbomethoxy group in the first two structure arises from the methanolic work-up. If one considers that the major amino acid substituent is serine, then the oligosaccharide, after glycosidic bond cleavages at all possible locations, would give the - 113 -following disaccharide fragment structures: (HexNAc,Hex); (HexNAc,Hex,Ser); (HexNAc,Hex,OMe); (HexNAc,Hex,OMe,Ser); (HexNAc,HexA); (HexNAc,HexA,Ser); (HexNAc,Hex A,OMe); (HexNAc,HexA,OMe,Ser); (HexNAc ,HexA,20Me,Ser); (Hex,HexA); (Hex,HexA,Ser); (Hex,HexA,OMe); (Hex,HexA,OMe,Ser); (Hex,HexA,20Me,Ser); (2Hex); (2Hex,Ser); and (2Hex,Ser,OMe). The fragment ions m/z 470.1435 and 456.1299 could be assigned to (Hex,HexA,2OMe,Ser) and (Hex,HexA,OMe,Ser) respectively. Dehydration of these ions gives fragments m/z 452.131 and 438.116. Moreover, the fragment ions m/z 369.107 could be assigned to (Hex,HexA,OMe), while simultaneous dehydration at two sites could give fragment ion m/z 333.078. On the other hand, the fragment ion m/z 424.134 could be assigned to two different structures: [(2Hex,Ser,OMe)-rl20-H.]~ or [(Hex,HexNAc)+C2H302-rI]~. The ion series m/z 470, 456, 452, 438 and 369 strongly indicates that the serine residue is linked to the (Hex,HexA) disaccharide. Therefore, fragment ion m/z 424 is more likely to be [(Hex,HexNAc)+C2rl302-m'-The ion series m/z 470 ,456, 452, 438, 369 and 333 suggests that the oligosaccharide has a serine residue linked to either a hexuronic acid and or a hexose at the reducing end. This effectively limits the oligosaccharide to the following six possible carbohydrate backbones: Hex - HexNAc - HexA - Hex ~ OMe Hex - HexNAc - Hex- HexA ~ OMe HexNAc - Hex (Hex) - HexA ~ OMe HexNAc - Hex - HexA (Hex) ~ OMe Hex - HexNAc - HexA (Hex) ~ OMe HexNAc - Hex (HexA) (Hex) ~ OMe The three most prominent fragment ions (470.144 - 100%; 426.115 - 100%; 408.110 -90%) in the spectra were all even masses, implying that they all contain one nitrogen atom. Most of the fragment ions observed in the spectrum could be explained by invoking either a - 114 -glycosidic bond or simple two-carbon ring cleavage argument. However, as the two very intense ions, m/z 426 and 408, could not be assigned by these simple fragmentations, theV were assumed to be derived from more complex ring cleavages. Their nitrogens could be\ derived either from the amino sugar or the amino acid residue. The ion m/z 426.115 could, be attributed to a deprotonated, deacetylated disaccharide (HexA,HexNAc) linked to a -C3H5O3 ring fragment (Fig. rV.29). Dehydration of this ion would give the fragment m/z 408.110. Fig.IV.29 Structure of fragment ion (HexN.HexA^CiHsOi. The hydroxyl groups of the monosaccharide residues are not shown. The -C3H5O3 ring fragment could be attached to either the hexuronic acid or the amino hexose residues. Deprotonation of ion m/z 426.1146, and subsequent dehydration will give fragment m/z 408.1098. From the mass spectra of all the model oligosaccharides in this thesis, no fragment ions were observed that have a ring-cleaved fragment linking two "whole" monosaccharide residues. Therefore, it is assumed that all ring fragments observed are associated with either the non-reducing or the reducing end. Hence, the two fragment ions m/z 425 and 408 suggest that the acetamido hexose is adjacent to the hexuronic acid residue. Therefore, the oligosaccharide backbone could only be the following linear and 3+1 branched structures: Hex (l->?) HexNAc (l->?) HexA (l->?) Hex ~ OMe Hex (l->?) HexNAc (1-*?) HexA ~ OMe ? Hex - 115 -A l l the fragment ions mentioned above can be fitted to a tetrasaccharide containing either serine or threonine residues, which suggests that there are a mixture of two oligosaccharides in the sample, one with a serine substituent and the other substituted with a threonine. The very slight differences in the oligosaccharide structures will make their complete separation and isolation of both components difficult. However, from the mass spectra there is no direct evidence to locate the exact position of the amino acid. Nevertheless, there is some indirect evidence that points to the location and the linkage of the amino acid, i) It is postulated above that the amino acids are linked to either the hexuronic acid or the hexose residue. As discussed before, there are only three different ways that the amino acid could be linked (Fig. IV.26). If the amino acid is linked via the C-terminal to any hydroxyl group, then the 'free' C-6 of the uronic acid would be esterified during the work-up procedure, and the disaccharide fragment (Hex,Hex A,2 OMe) should be observed. No ions could be assigned to that structure. This may imply that the amino acid is linked to C-6 of the uronic acid, and loss of the esterifed amino acid would give fragment ion m/z 369 and 333, (Hex,HexA,OMe). ii) The fragment ion m/z 791 was assigned to the deprotonated, decarboxylated structure (2Hex,HexA,HexNAc,Thr,OMe). This fragment could in theory be assigned to the deprotonated, decarboxylated structure (2Hex,HexA,HexNAc,Ser,20Me). However, due to the work-up procedure, this assignment would imply that there were three sites of esterification (C-l of the reducing monosacharide, C-6 of the hexuronic acid and the C-terminal of the amino acid) and the amino acids were linked via the N-terminal to a hydroxyl group of the oligosaccharide, which is unlikely. Therefore, the structure of the fragment ion m/z 791 could only be a tetrasaccharide substituted with a threonine residue. This suggests that the amino acid is linked to C-6 of the uronic acid, and loss of a C 2 H 2 O 2 unit from the esterified C-terminal of the 'threoninated' pseudo-molecular ion would give fragment ion m/z 791. Further loss of a hexose residue would give fragment ion m/z 629. Therefore, the indirect evidence suggests that the amino acids are linked (either via the N-terminal or the hydroxyl group) to C-6 of the hexuronic acid. -116 -The negative-ion l.d.i.-F.t.-i.c.r. fragment ions of the proposed tetrasaccharide structures substituted with serine and threonine residue are listed in Table IV. 11. There are two ions (m/z 528.095 and 484.060) that could not be assigned exact structures, however, their masses suggest that both are chloridated fragments containing one nitrogen atom. / - 117 -Table IV. 11 NEGATIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF E. coli K49 OLIGOSACCHARIDE OBTAINED BY ANHYDROUS HYDROGEN FLUORIDE HYDROLYSIS OF THE NATIVE CAPSULAR POLYSACCHARIDE (METHANOL QUENCHED)  Observed Calculated Relative Mass Proposed mass mass intensity error structure amu amu % ppm 835.280 835.284 34 -0.7 [(2Hex,HexA,HexNAc,Ser,20Me)-H]~ or [(2Hex ,Hex A,HexNAc ,Thr,OMe) -H]" 821.268 821.268 10 0.0 [ (2Hex ,Hex A ,HexN Ac, S er ,OMe )-H]" or [(2Hex,HexA,HexNAc,Thr)-H]~ 791.246 791.294 40 60.8 [(2Hex,HexA)HexNAc,Thr,OMe)-C02-H]" 673.247 673.231 30 24.4 [(Hex,HexA,HexNAc,Ser,20Me)-H]~ or [(Hex,HexA,HexNAc,Thr,OMe)-H]~ 659.200 659.215 16 -23.1 [(Hex,HexA,HexNAc,Ser,OMe)-H]~ or [(Hex,Hex A.HexN Ac ,Thr)-H j" 629.214 629.241 27 -42.9 [(Hex,HexA,HexNAc,Thr,OMe)-C02-H]~ 528.095 13 484.060 11 470.144 470.152 100 -17.2 [(Hex,HexA,Ser,20Me)-H]~ or [(Hex,HexA,Thr,OMe)-H]~ 460.107 460.123 15 -35.4 [(Hex,HexNAc)+C2H302+Cl]~ 456.130 456.136 22 -13.2 [(Hex,HexA,Ser,OMe)-H]~ or [(Hex,HexA,Thr)-H]~ 452.131 452.141 19 -21.5 [(Hex,HexA,Ser,2OMe)-H20-H]~ or [(Hex,HexA,Thr,OMe)-H20-H]~ 438.116 438.125 10 -21.2 [(Hex,HexA,Ser,OMe)-H20-H]" or [(Hex,HexA,Thr)-H20-H]~ 426.115 426.125 100 -25.2 *[(HexA,HexNAc)+C3H503-H]" 424.134 424.146 13 -28.5 [(Hex,HexNAc)+C2H302-H]_ 408.110 408.115 90 -12.2 *[(HexA,HexNAc)+C3H5O3-H20-H]" 400.086 400.102 12 -39.9 [(Hex,HexNAc)-H20+Cl]' 369.107 369.104 16 8.5 [(Hex,HexA,OMe)-H]~ 364.123 364.125 15 -4.9 [(Hex,HexNAc)-H20-H]" 333.078 333.083 15 -14.8 [(Hex,HexA,OMe)-2H20-H]" Loss of a ketene from an amino sugar, -C 2 H 2 Q - 118 -IV.3.11 POSITIVE-ION L.DJ.-F.T.-I.C.R. SPECTRA OF E. coli K49 OLIGOSACCHARIDE The positive-ion l.d.i.-F.L-i.c.r. spectral data (Spec, rv.25 and IV.26, Table IV. 12) were in accordance with the proposed sequences obtained from the negative-ion spectra. The fragment ions m/z 510.115, 494.154 and 466.116 confirmed that the amino acid residue was located on either the hexuronic acid or the hexose. The fragment ions m/z 432.160 (Fig. IV.30) and 464.128 (Fig. IV.31) are consistent with an acetamido hexose linked to the reducing end (HexA,Hex) residues. However, there was no indication of the exact location of the amino acids. COOH •O ^ ^ * ( O H ) x NHAc Fig.IV.30 Structure of fragment ion (Hex,HexA,OMe)+C2H30NAc. Decarboxylation, and sodiation of this structure will give m/z 432.160. None of the hydroxyl groups on the hexose ring were drawn in, due to the unknown positions of linkage. COOH >~0\ ( v< W \ A / * 0 , / v w w * ^ / \ f \*Jti / N > - ( O H ) , xl Fig.lV.31 Structure of fragment ion (Hex,HexA,OMe)+CH02- Addition of the threonine or serine+OMe residue, loss of two molecules of water, and subsequent sodiation will give m/z 464.128. None of the hydroxyl groups on the hexose ring were drawn in, due to the unknown positions of linkage. - 119 -Table IV.12 POSITIVE-ION LDI-FT-ICR FRAGMENT STRUCTURES OF E. coli K49 OLIGOSACCHARIDE OBTAINED BY ANHYDROUS HYDROGEN FLUORIDE HYDROLYSIS OF THE NATIVE CAPSULAR POLYSACCHARIDE (METHANOL QUENCHED) Observed Calculated Relative Mass Proposed mass mass intensity error structure amu amu % ppm 845.024 845.265 18 -285.0 [(2Hex,HexA,HexNAc,Ser,OMe)+Na]+ or [(2Hex,Hex A,HexN Ac ,Thr)+Na]+ 829.267 829.212 11 65.8 [(2Hex,HexA,HexNAc,Ser)-H20+K]+ 813.224 813.239 16 -18.3 [(2Hex,HexA,HexNAc,Ser)-H20+Na]+ 510.115 510.122 15 -13.3 [(Hex,HexA,Ser,20Me)+K]+ or [(Hex,HexA,Thr,OMe)+K]+ 494.154 494.148 14 12.6 [(Hex,HexA,Ser,20Me)+Na]+ or [(Hex,HexA,Thr,OMe)+Na]+ 466.116 466.117 11 -1.7 [(Hex,HexA,Ser)+Na]+ 464.128 464.140 58 -25.0 [(Hex,HexA,Ser,20Me)+CH02-2H20+Na]+ or [(Hex,HexA,Thr,OMe)+CH02-2H20+Na]+ 448.150 448.143 43 15.5 [(Hex,HexNAc)+C2H302+Na]+ 432.160 432.148 12 29.4 [(Hex,HexA,OMe)+C2H3ONAc-C02+Na]+ 422.113 422.106 17 16.3 [(Hex,HexNAc)+K]+ 406.130 406.132 15 -4.7 [(Hex,HexNAc)+Na]+ 404.082 404.095 40 -32.7 [(Hex,HexNAc)-H20+K]+ 388.121 388.122 100 -1.5 [(Hex,HexNAc)-H20+Na]+ 226.071 226.069 12 11.1 [(HexNAc)-H20+Na]+ 204.087 204.087 19 0.5 [(HexNAc)-H20+H]+ - 120 -In summary, the positive- and negative-ion spectra of E. coli K49 oligosaccharide suggests that the oligosaccharide is substituted with either a serine or a threonine residue. Some of the known E. coli capsular polysaccharides that are substituted with serine also partially incorporate threonine . However, there was no indication of the relative abundance of the two substituted oligosaccharides. Although there were no fragments that could be used to locate the exact position of the amino acids, indirect evidence suggested that the linkage is via C-6 of the hexuronic acid. No positions of linkage could be identified from the spectra. Future m.s-m.s. experiments are planned to locate the exact positions of linkage for the oligosaccharide backbone and the positions of linkage of the amino acid substituent. In conclusion, the negative- and positive-ion l.d.i.-F.t.-i.c.r. spectra of the sample suggest that there are two closely related oligosaccharides present. Both structures have the same tetrasaccharide backbone, and both are substituted at C-6 of the hexuronic acid with either a serine or a threonine residue. However, from the spectra it was not possible to distinguish whether the tetrasaccharide backbone is a linear or a 3+1 branched structures. The proposed structures of the oligosaccharides are therefore: - 121 -a) Ser - OMe Hex (1-*?) HexNAc (l->?) HexA (l-»?) Hex ~ OMe Thr-OMe Hex (l->?) HexNAc (l-»?) HexA (l->?) Hex - OMe or b) Ser - OMe Hex (l->?) HexNAc (l->?) HexA - OMe ? Hex Thr-OMe Hex (l->?) HexNAc (l->?) HexA - OMe Hex Although these negative- and positive-ion spectra would not distinguish whether this underivatized oligosaccharide was a linear or a 3+1 branched structure, a straight forward methylation analysis (Section n.3) would reveal the exact sequence of this tetrasaccharide backbone. A 3+1 branched structure would give two tetra-0-methylated hexitol acetates, while a linear structure would yield one tetra-0-methylated and one tri-O-methylated hexitol acetates. - 122 -Another approach to sequence this oligosaccharide is by permethylation. The series of permethylated oxonium ions for the proposed linear tetrasaccharide is m/z 219, 476, 694. Although the fragment ions m/z 219 and 476 would also be observed in the proposed 3+1 branched structure, m/z 694 would not be observed . Subsequent to the development of the mass spectral analysis, the primary structure of this oligosaccharide was established independently by classical carbohydrate chemistry208 and 2-D n.m.r. of the native capsular polysaccharide2 0 9, and found to be a linear tetrasaccharide substituted with either threonine (75%) and serine (25%) through an amide bond to the C-6 of the glucuronic acid: Glc (l-»3) GalNAc (l->4) GlcA (NHCH(CH2OH)COOMe) (l->6) Gal Glc (l-»3) GalNAc (l->4) GlcA (NHCH(CHOHCH3)COOMe) (l->6) Gal - 123 -IV.4 GENERAL DISCUSSION In the earlier Sections, it was proposed that for high molecular weight polysaccharides, fragmentation occurs principally in the solid or the solid-gas interface (selvedge), rather than in the gas phase, with ionization only arising before i.c.r. excitation and detection. This postulate implied that positive- and negative-ion mode laser desorption ionization of carbohydrates would give similar ring fragments. The ring fragments observed would therefore reflect the conformational integrity of the polysaccharide in the solid state. If this "macro-molecular" effect plays a major role in laser mass spectrometry, then the spectra of similar polysaccharides would have similar features. Furthermore, the ring fragments observed in negative-ion mass spectra can be used to interpret positive-ion mass spectra and vice versa. As illustrated in this thesis, most of the different types of ring-cleaved fragments are observed in both the positive- and negative-ion mass spectra, which support the proposed "fragmentation before ionization" model. When a large number of oligosaccharides with different degrees of polymerization were observed in the mass spectra, they could have resulted from either sequential loss of one or more residues in the gas phase or selvedge, or the simultaneous desorption of a whole range of oligosaccharide fragments from the solid sample. If an oligosaccharide were to be "frozen" in a particular conformation in the solid, and fragmentation occurs in the selvedge, then the oligosaccharide spectrum would have similar features to that of the polysaccharide spectrum. However, once the oligosaccharides were desorbed into the gas phase, they would be energetic enough to take up different conformations. Consequently, some stereospecificity would be lost, and the spectra observed would reflect the most probable unimolecular or ion-molecule reactions in the gas phase. This fragmentation model can explain the very similar spectra of the disaccharides, maltose and cellobiose. As both disaccharides are of low molecular weight, they can be easily be desorbed into the gas phase rather than fragmenting in the solid. Due to the similar structures of the disaccharides, they would have similar gas phase unimolecular and ion-molecule reactions, thereby giving identical mass spectra. - 124 -One possible way of encouraging low molecular weight oligosaccharides to fragment in the solid, would be to dope the sample with a large amount of alkali halide salt The thermal energy absorbed by the oligosaccharide "trapped" in a alkali halide matrix could be dissipated either by vaporization followed by fragmentation in the gas phase or fragmentation in the solid or selvedge. Due to the biological nature of the sample, all the oligosaccharides investigated in this thesis contain a high abundance of attached alkali halide. Therefore, it is not possible to compare the spectra of a real "undoped" vs a heavily doped sample. The Coates-Wilkins' mass spectral data on different polysaccharides (Table V . l l ) can be interpreted to strengthen the above "macro-molecular" model for laser mass spectrometry of polysaccharides. A l l the polysaccharides analyzed were of very high molecular weight. Therefore, if fragmentation occurred in the solid or selvedge, then the fragments observed from polysaccharides with different gross molecular structures would be different. This postulation also suggested that if two polysaccharides shared a similar tertiary structure, then their mass spectra would have a number of common features. Any differences observed in their spectra could have resulted from the differences at the "micro-molecular" level. Of the nine polysaccharide mass spectra obtained by Coates-Wilkins, several of them have a number of common gross molecular features. Their mass spectra are discussed below. Cellulose is a polymer composed of 1,4-linked B-D-glucose in a straight chain. The gross molecular structure of cellulose is considered to be a number of long strand fibers running antiparallel to each other. Similarly, chitin is a fibrous polymer composed of 1,4-linked B-D-N-acetyl glucosamine. Therefore, the only difference in the "micro-molecular" structure is the difference in substitution at C-2 of the glucose residue. Both polysaccharides gave similar spectra, with the ion series A (X = 0), F (X = 60), G (X = 74), and Q (X = 144 or 185). The only ion series observed for cellulose but not for chitin, is the L-series (X = 102). This may reflect the subtle differences at the micro-molecular level. According to our interpretation, this L-series was derived from two ring fragments located at each end of the oligosaccharides. It may be that the acetamido substituent in chitin does not favor M2-ring cleavage between the C-2, C-3 bond of the acetamido glucose. - 125 -Similarity, xanthan gum has a cellulose backbone, and should show fragmentation similar to cellulose. Both polysaccharides gave the ion series A (X = 0), F (X = 60), L (X = 102), and Q (X = 144). Although both spectra showed a number of similar features, there are some differences in their spectra, which may reflect the substitution pattern of xanthan gum. The differences in their spectra are: i) ion series C (X = 24) and P (X = 126) were observed for xanthan gum but not for cellulose; these two ion series correspond to two units of dehydration for glycosidic bond (P, X = 126) and ring cleavage between C-2 and C-3 (C, X = 24) reactions; ii) G (X = 74) was not observed for xanthan gum. Starch is a poly- 1,4-linked a-D-glucose (with varying degree of branching by 1,6-linkage), while white dextrin is the acid/heat-treated lower molecular weight polymer of starch with a lower degree of branching. Except for glycosidic bond cleavages and M 2 -reactions, the spectra obtained for starch and white dextrin are very different. Starch gave ion series of J (X = 88), M (X = 104) and R (X = 148); while white dextrin gave E (X = 44), L (X = 102) and N (X = 120). Both polysaccharides possess the same monosaccharide residues, position of linkage, anomeric configuration and cc-helix structure. The differences in their mass spectra could be a reflection of the differences in the "macro-molecular" structure of the polysaccharide in the solid state; starch being more branched than white dextrin would have a "less ordered packing". The same argument can also be used to account for the spectral differences observed for starch and cellulose. Starch gave ring fragments of ion series J (X = 88), M (X = 104) and R (X = 148); while cellulose gave G (X = 74) and L (X = 102). The differences in the mass spectra could have arisen from differences in the tertiary structure of the polysaccharides. The proposed "macro-molecular" model requires that some glycosidic and ring cleavages occur in the solid state or at the solid-gas interface. The two main desorption techniques for depth profiling of solid materials are laser desorption and s.i.m.s. It is likely that the laser pulse removes the first layer of sample, causing pyrolysis, fragmentation and spontaneous desorption of the oligosaccharide. The energetic oligosaccharide could then - 126 -undergo further fragmentation in the gaseous phase, corresponding to losses of one or more monosaccharide residues, dehydration and further ring cleavages. In the l.d.i.-F.t.-i.c.r. spectra, the same ring fragment structures are observed in both positive- and negative-ion mode. The only difference is that the carbohydrate structures are deprotonated in the negative-ion mode, and cationized by protonation or by alkali-ion attachment in the positive-ion spectra. If fragmentation occurs predominantly in the selvedge, then subsequent deprotonation or protonation will give the same structure in both the positive- and negative-ion spectra. Although the application of l.d.i.-F.t.-i.c.r. to carbohydrates is relatively new compared to f.a.b or d.c.L, it has already shown a lot of potential. Although routine f.a.b. sequencing of both peptides and oligosaccharides is being carried out in most laboratories, it is still seen as backup or complementary techniques. This can be explained as follows: i) High resolution f.a.b.-m.s.-m.s. are not routine experiments, and high resolution m.s.-m.s. is required to analyze a completely unknown sample. Often, by performing a limited amount of simple non-instrumental chemistry, some tentative composition data can be obtained. Therefore, most biological samples require only medium resolution m.s. analysis. ii) Soft ionization techniques, such as f.a.b. and l.s.i.m.s., result in mainly glycosidic bond cleavages, thereby permitting only the sequencing of the sample. Consequently, some workers have attempted to use tandem m.s. to correlate these m.s.-m.s. data to additional structural features 1 1 8" 1 2 1. L.d.i.-F.t.-i.c.r. spectra are very rich, having both fragments derived from glycosidic bond and ring cleavages. We have shown that these ring fragments can be used for a tentative indication of positions of linkage and possibly a correlation to other stereochemical features. Further studies would have to be performed on a large number of model oligo- and poly-saccharides to ascertain the exact structures of the ring fragments proposed in this thesis. From the five l.d.i.-F.t.-i.c.r. model oligosaccharide experiments described in this thesis. A few suggestions can be made with respect to the analysis of unknown oligosaccharides: - 127 -a) The first set of experiments (positive- and negative ion mode) should be performed on the underivatized material to obtain information on the component, as derivatization may cause loss of acid- or base-labile substituents. These experiments should also provide some tentative sequences and linkage information on most linear and some branched oligosaccharides. b) The second set of experiments (positive- and negative ion mode) should be performed on the permethylated material to obtain complementary sequence information. This will provide: i) components information including base-stable substituents, and ii) sequencing from the non-reducing end. c) The third set of experiments (positive- and negative ion mode) should be performed on the peracetylated or the pertrifluoroacetylated material to obtain complementary sequence information. This will provide: i) component information including acid-stable substituents, and ii) sequencing from the non-reducing end. IV.5 CONCLUSION In conclusion, the advantages of l.d.i.-F.t.-i.c.r. are as follows: i) underivatized oligosaccharides can be analyzed; ii) sequence information can be deduced from both the positive- and negative-ion spectra; iii) intense pseudo-molecular ions are obtained; iv) intense fragment ions from both glycosidic and ring cleavages are obtained; v) high resolution spectra can confirm the presence of unusual monosaccharides, such as deoxysugars, acetamido sugars, sialic acids and of labile substituents, such as O- & iV-acetyls, pyruvic acid acetals and lactyls; vi) ring cleavage fragments may be used to distinguish alternative reducing-end structures; vii) ring cleavage fragments may also provide tentative information on the positions of linkage; viii) the characteristic fragmentation patterns seen in oligosaccharides can be used to analyze the characteristic fingerprint mass spectra of polysaccharides. - 128 -IV.6 FUTURE STUDIES i) M.s.-m.s. studies of oligosaccharide fragments. Although l.d.i.-F.t.-i.c.r. provides abundant ring fragmentation, sometimes extensive ring cleavages can complicate the delineation of the oligosaccharide structures. Therefore, m.s.-m.s. studies may be useful to characterize some ambiguous oligosaccharide framents. ii) Dopant vs fragment ion structures. For all our model oligosaccharides, addition of dopants (NaCl, NH^Br and KBr) did not alter the structural features of the mass spectra in the positive-ion mode. With our sample preparation technique, the undoped oligosaccharides were analyzed first in both positive- and negative-ion modes, followed by addition of either NH4Br or KBr and finally NaCl for re-examination if necessary. Although only pico-mole levels of samples were used for each laser shot, at the end of the experiments a substantial amount of the oligosaccharides had been used and a decrease in the intensity of the spectra was normally observed. Nevertheless, with the addition of halide ions, a slight increase in fragmentation was observed in the negative-ion spectra. Due to the complexity of these "halidated" spectra, they were not analyzed in detail. Exact mass measurement using the heterodyne mode should be performed on desalted as well as doped samples to determine the exact nature of the fragmentation, and the relationship between the degree of doping with increase in fragmentation. iii) Laser intensity vs fragment ion structures of polysaccharides. It was noted by Coates during the experiments on polysaccharides that, except for overall intensity, the l.d.i.-F.t.-i.cr. mass spectra did not change with laser power . If fragmentation of the polysaccharides occurred in the selvedge, and the mass spectra reflect the macro-molecular structure in the solid state, then laser power should not substantially alter the fragmentation. Therefore, one aspect of further studies should concentrate on analyzing a series of homoglycans and simple linear heteroglycans to evaluate the different possible ring fragments observed under different laser power. The observed different ring fragments should be correlated with the primary and tertiary structure of the glycans to evaluate the proposed "macro-molecular" model for l.d.i.-F.t.-i.c.r. mass spectra. - 129 -iv) Laser intensity vs fragment ion structures of oligosaccharides. The mass spectra of oligosaccharides fragmenting in the gas-phase should be different from that of oligosaccharides undergoing pyrolysis and fragmentation in the solid state. If the oligosaccharides were desorbed into the gas-phase before fragmentation, then ion-molecule reactions would govern the fragmentation pathways. F.a.b.-m.s.-m.s. positive- and negative-ion spectra of monosaccharides have already shown different fragmentation patterns, and this is attributable to the different unimolecular and ion-molecule reactions occurring under the two different conditions120. In order to ascertain the site of fragmentation, a series of simple oligosaccharide repeating units derived from bacterial capsular polysaccharide (PI, P2, P3, P4, etc..) should be examined using different laser powers to determine the relationship between the degree of polymerization and the tendency for fragmentation to occur predominantly at the selvedge. Furthermore, m.s.-m.s. studies could be performed on the molecular and glycosidic bond cleaved ions to ascertain the exact nature of the fragmentation patterns of oligosaccharides in the gas phase. v) The ring fragment structures of underivatized and permethylated oligosaccharides obtained by l.d.i.-F.t.-i.c.r. should be compared to the fragments obtained from f.a.b.-m.s.-m.s. studies. - 130 -CHAPTER V BIBLIOGRAPHY - 131 -BIBLIOGRAPHY 1 G.O. Aspinall, in "The Polysaccharides" Vol. 2, 36-131, Ed. G.O. Aspinall, Academic Press, New York, 1983. 2 P.A. Sandford and J. Baird, in "The Polysaccharides". Vol. 2, 411-490, Ed. G.O. Aspinall, Academic Press, New York, 1983. 3 S. Hanessian, "Total Synthesis of Natural Products: The 'Chiron' Approach", Pergamon Press, New York, 1983. 4 "A Single Issue on AIDS" Scientific America, 259, 8 (1988). 5 C.T. Bishop and H J . Jennings, in "The Polysaccharides" Vol. 2, 291-330, Ed. G.O. Aspinall, Academic Press, New York, 1982. 6 W.F. Dudman, in "Surface Carbohydrates of the Prokaryotic Cell" 357-414, Ed. I. Sutherland, Academic Press, London, 1977. 7 K. Jann and B Jann, Prog. Allergy, 33 (1983) 53-79. 8 H J . Jenning, Adv. Carbohydr. Chem. Biochem., 44 (1983) 155-208. 9 K. Jann and O. Westphal, in "The Antigens", Vol. 3., 1-125, Ed. M . Sela, Academic Press, New York, 1975. 10 M . Heidelberger and G.G.S. Dutton, / . Immunol., 111 (1973) 857-859. 11 M . Heidelberger and W. Nimmich, Immunochem., 13 (1976) 67-80. 12 Y J . Lew and M . Heidelberger, Carbohydr. Res., 52 (1976) 255-258. 13 M . Heidelberger, Res. Immunochem. Immunobiol., 3 (1973) 1-40. 14 M . Heidelberger, Trends Biochem. Sc., 7 (1982) 261-263. 15 R.L. Whistler, A .A . Bushway, P.P. Singh, I. Nakahara, and R. Tokuzen, Adv. Carbohydr. Chem. Biochem., 32 (1976) 235-275. 16 J.R. Crook, W.K. Otto, and R.S. Jones, Proc. Soc. Exp. Biol. Med., 109 (1962) 552-556. 17 K. Miyamoto, R. Koshiura, T. Hasegawe, and N. Kato, Jpn. J. Pharmacol., 36 (1984) 51-57. - 132 -18 P.-E. Jannson, L. Kenne, H. Liedgren, B. Lindberg, and J. Lonngren, Chem. Commun., Univ., Stockholm, 8 (1976) 1-75. 19 J. Lonngren and S. Svensson, Adv. Carbohydr. Chem. Biochem., 29 (1974) 84-106. 20 Z. Lam, M . Sc. Thesis, University of British Columbia, (1986). 21 O. Westphal and K. Jann, Methods Carbohydr. Chem., 5 (1965) 83-91. 22 R.L. Whistler and J.L. Sannella, Methods. Carbohydr. Chem., 5 (1965) 34-36. 23 T.J. Painter, Methods Carbohydr. Chem., 5 (1965) 98-100. 24 G.O. Aspinal, J.A. Molloy, and J.W.T. Craig, Can. J. Biochem., 47 (1969) 1063-1070. 25 J.E. Scott, Methods Carbohydr. Chem., 5 (1965) 38-44. 26 K.A. Granath, Methods Carbohydr. Chem., 5 (1965) 20-28. 27 Y.C. Yee and R. Montgomery, Methods Carbohydr. Chem., 5 (1965) 28-34. 28 H. Neukon and W. Kuendig, Methods Carbohydr. Chem., 5 (1965) 14-17. 29 I.J. Goldstein and C.E. Hayes, Adv. Carbohydr. Chem. Biochem., 35 (1978) 127-340. 30 S. Narasimhan, J.R. Wilson, E. Martin, and H. Schachter, Can. J. Biochem., 57 (1979) 83-96. 31 S.C. Churms, Adv. Carbohydr. Chem. Biochem., 25 (1970) 13-51. 32 D.H. Northcote, Methods Carbohydr. Chem., 5 (1965) 49-53. 33 M . Breen, H.G. Weinstein, L.J. Black, M.S. Borcherding, and R.A. Sittig, Methods Carbohydr. Chem., 7 (1966) 101-115. 34 G.G.S. Dutton, Adv. Carbohydr. Chem. Biochem., 28 (1973) 11-160. 35 PJ . Garegg and B. Lindberg, Acta Chem. Scand., 14 (1960) 871-876. 36 P. Albersheim, D.J. Nevins, P.D. English, and A. Karr, Carbohydr. Res., 5 (1967) 340-345. 37 C.G. Hellerqvist and A.A. Lindberg, Carbohydr. Res., 16 (1971) 39-48. 38 A J . Mort, and D.T.A. Lamport, Anal. Biochem., 82 (1977) 289-309. 39 H J . Jennings and C. Lugowski, Can. J. Chem., 58 (1980) 2610-2612. 40 J.F. Codington, K.B. Linsley, and C. Silber, Methods Carbohydr. Chem., 7 (1976) 226-232. - 133 -41 S. Inoue, G. Matsumura, and Y. Inoue, Anal. Biochem., 125 (1982) 118-124. 42 G.G.S. Dutton and M.T. Yang, Can. J. Chem., 51 (1973) 1826-1832. 43 Y.C. Lee and C.E. BaUou, Biochem., 4 (1965) 257-264. 44 K. Macek, in "Paper Chromatography", Eds. I.M. Hais and K. Macek, Academic Press, New York, 1963. 45 K. Macek, in "Chromatography", 2nd Edit., Ed. E. Heftman, Reinhold Publishing Corporation, New York, 1967. 46 R.E. Wing and J.N. BeMiller, Methods Carbohydr. Chem., 6 (1972) 42-64. 47 H. Weigel, Adv. Carbohydr. Chem., 18 (1963) 61-97. 48 Z. Dische, Methods Carbohydr. Chem., 1 (1962) 477-514. 49 D. Aminoff, W.W. Binkley, R. Schaffer, and R.W. Mowry, in "The Carbohydrates", 2nd Edit., Vol. 2B, 734-807, Eds. W. Pigman and D. Horton, Academic Press, New York, 1970. 50 G.G.S. Dutton, Adv. Carbohydr. Chem. Biochem., 30 (1974) 9-110. 51 G.D. McGinnis and D. Fang, Methods Carbohydr. Chem., 8 (1980) 33-43. 52 R.L. Taylor and H.E. Conrad, Biochem., 11 (1972) 1383-1388. 53 J.K.N. Jones and M.B. Perry, / . Am. Chem. Soc, 79 (1957) 2787-2793. 54 G.M. Bebault, J.M. Berry, Y . M . Choy, G.G.S. Dutton, N . Funnel, L.D. Hayward, and A M . Stephen, Can. J. Chem., 51 (1973) 324-326. 55 K. Leontin, B. Lindberg, and J. Lonngren, Carbohydr. Res., 62 (1978) 359-362. 56 G.J. Gerwig, J.P. Kamerling, and J.F.G. Vliegenthart, Carbohydr. Res., 77 (1979) 1-7. 57 J.H. Pazur, Methods Enzymol., 9 (1966) 82-87. 58 D. Amaral, F. Kelly-Falcoz, and B.L. Horecker, Methods Enzymol., 9 (1966) 87-92. 59 Y.T. L i and S.C. L i , Methods Carbohydr. Chem., 7 (1976) 221-225. 60 S.J. Angyal and K. James, Aust. J. Chem., 23 (1970) 1209-1215. 61 S.I. Hakomori, J. Biochem. (Tokyo), 55 (1964) 205-208. 62 J.I. Brauman, J.A. Bryson, D.C. Kahl, and N.J. Nelson, / . Am. Chem. Soc, 92 (1970) 6679-6680. - 134 -63 L.R. Phillips and B.A. Fraser, Carbohydr. Res., 90 (1981) 149-152. 64 A.B. Blakeney and B.A. Stone, Carbohydr. Res., 140 (1985) 319-324. 65 I. Ciucanu and F. Kerek, Carbohydr. Res., 131 (1984) 209-217. 66 P. Prehm, Carbohydr. Res., 78 (1980) 372-374. 67 T. Purdie and J.C. Irvin, / . Chem. Soc., 83 (1903) 1021-1037. 68 M . Barber, R.S. Bordoli, G J . Elliott, R.D. Sedgwick, and A.N. Tyler, Biomed.Mass Spectrom., 8 (1981) 492-495. 69 A. Dell and P.R. Tiller, Biochem. Biophys. Res. Comm., 135 (1986) 1126-1134. 70 A.N. de Belder and B. Normann, Carbohydr. Res., 8 (1968) 1-6. 71 B. Lindberg, J. Lonngren, and J.L. Thompson, Carbohydr. Res., 28 (1973) 351-357. 72 G O . Aspinall and K.-G. Rosell, Carbohydr. Res., 57 (1977) c23-c26. 73 L. Malaprade, Bull. Soc. Chim. Fr., 43 (1928) 683-696. 74 H. Geyer, K. Himmelspach, N . Kwiatkowski, S. Schlect, and S. Stirm, Pure Appl. Chem., 55 (1983) 637-653. 75 B. Capon, Chem. Rev., 69 (1969) 407-498. 76 L .M. Beynon and G.G.S. Dutton, Carbohydr. Res., 179 (1988) 419-423. 77 A.V.S. Lim, Ph. D. Thesis, University of British Columbia, (1987). 78 J.M. Bobbitt, Adv. Carbohydr. Chem., 11 (1956) 1-40. 79 R.D. Guthrie, Adv. Carbohydr. Chem., 16 (1961) 105-157. 80 G O . Aspinall and R J . Ferrier, Chem. Ind. (London) (1957) 1216. 81 P. Fleury and J. Lange, / . Pharm. Chim., 17 (1933) 107-113. 82 I.J. Goldstein, G.W. Hay, B.A. Lewis, and F. Smith, Methods Carbohydr. Chem., 5 (1965) 361-370. 83 G.G.S. Dutton and K.B. Gibney, Carbohydr. Res., 25 (1972) 99-105. 84 M . Adams, "Bacteriophages", Interscience Publishers Inc., New York, 1957. 85 J. Douglas, "Bacteriophages", Chapman and Hall, London, 1975. 86 C.K. Mattews, "Bacteriophages Biochemistry", Van Nostrand Reinhold Co., N.Y. 1971. - 135 -87 A.A. Lindberg, in "Surface Carbohydates of the Prokaryotic Cell.", Ed. I. Sutherland, Academic Press, London, 1977. 88 D. Rieger-Hug and S.Stirm, Virology, 113 (1981) 363-378. 89 G.G.S. Dutton, G.K. Eigendorf, Z. Lam, and A.V.S. Lim, Biomed. Environ. Mass Spectrom., 15 (1988) 459-460. 90 B. Coxon, Adv. Carbohydr. Chem., 19 (1964) 51-93. 91 L.D. Hall, Adv. Carbohydr. Chem. Biochem., 27 (1972) 7-83. 92 K. Bock and C. Pedersen, Adv. Carbohydr. Chem. Biochem., 41 (1983) 27-66. 93 K. Dill , E. Bereman, and A.A. Pavia, Adv. Carbohydr. Chem. Biochem., 43 (1985) 2-49. 94 P.E. Pferrer, / . Carbohydr. Chem., 3 (1984) 613-639. 95 C M . Preston and LX>. Hall, Carbohydr. Res., 37 (1974) 267-282. 96 K. Bock, Pure and Appl. Chem., 55 (1983) 605-622. 97 F. Miction, J.B. Brisson, R.Roy, F.E. Ashton, and H.J. Jennings, Biochem., 24 (1985) 5592-5598. 98 J. Dabrowski, H. Egge, and U. Dabrowski, Carbohydr. Res., 114 (1983) 1-9. 99 D.R. Bundle, M . Gerken, and M.B. Perry, Can. J. Chem., 64 (1986) 225-264. 100 P.J. Garegg, P.E. Jansson, B. Lindberg, F. Lindh, J. Lonngren, I. KvarnstrQm, and W. Nimmich, Carbohydr. Res., 78 (1980) 127-132. 101 M . Karplus, J. Chem. Phys., 30 (1959) 11-15. 102 L. Hough, J.K.N. Jones, and W. Wadman, / . Chem. Soc, (1950) 1702-1706. 103 W.F. Lehnherdt and R.J. Winder, / . Chrmatogr., 34 (1968) 471-479. 104 C C . Sweeley, R. Bentley, M . Makita, and W.W. Wells, / . Am. Chem. Soc, 85 (1963) 2497-2507. 105 D.P. Sweet, R.H. Shapiro, and P. Albershiem, Carbohydr. Res., 40 (1975) 217-225. 106 D.P. Sweet, P. Albershiem, and R.H. Shapiro, Carbohydr. Res., 40 (1975) 199-216. 107 B. Fournet, G. Strecker, Y. Leroy, and J. Montreuil, Anal. Biochem., 161 (1981) 489-502. - 136 -108 R. Geyer, H. Geyer, S. Ktthrihardt, W. Mink, and S. Strim, Anal. Biochem., 121 (1982) 263-274. 109 H. Yamada, H. Yiyohara, and Y. Otsuka, Carbohydr. Res., 170 (1987) 181-191. 110 H. Karlson and G.C. Hansson, Proc. XTVth Int. Carbohydr. Symp., (1988) 107. 111 H.D. Beckey, Int. J. Mass Spectrom. Ion Phys., 2 (1969) 500-503. 112 R.J. Cotter, Anal. Chem., 52 (1980) 1589A-1606A. 113 RX). Macfarlane and D.F. Torgerson, Science, 191 (1976) 920-925. 114 D.A. McCrery, E.B. Ledford, Jr., and M.L. Gross, Anal. Chem., 54 (1982) 1435-1437. 115 C R . Blakely, J.J. Carmody, and M.L. Vestal, / . Am. Chem. Soc, 102 (1980) 5931-5933. 116 W. Aberth, K . M . Straub, and A.L. Burlingame, Anal. Chem., 54 (1982) 2029-2034. 117 O.S. Chizhov and N.K. Kochetkov, Adv. Carbohydr. Chem. Biochem., 21 (1966) 29-93. 118 G. Puzo, J. C. Prom6 ,and J. Fournier, Carbohydr. Res., 140 (1985) 131-134. 119 E.G. de Jong, W. Heerma, and G. Dijksrra, Biomed. Mass Spectrom., 7 (1980) 127-31. 120 J. W. Dallinga and W. Heerma, Biomed. Environ. Mass Spectrom., 18 (1989) 363-372. 121 V. Kovacik, E. Petrakova, J. Hirsch, V. Mihalov, W. Heerma, and C. Versluis, Biomed. Environ. Mass Spectrom., 17 (1988) 455-458. 122 Y.-Z. Chen, S. Hua, H. L i , and H. Yang, Proc. 36th Con. Mass Spectrom. Allied Topics, 36 (1988) 163-164. 123 N. Ravenscroft, Ph. D. Thesis, University of Cape Town, (1988). 124 B. Munson, Anal. Chem., 43 (1971) 28A-43A. 125 A.G. Harrison, "Chemical Ionization Mass Spectrometry", CRC Press, Florida 1983. 126 V.N. Reinhold, Methods Enzymol., 138 (Complex Carbohydr. Pt. E) (1988) 59-84. 127 H. Liebl, Anal. Chem., 46 (1974) 22A-30A. 128 A. Benninghoven, Surf. Sci., 35 (1973) 427-429. 129 A. Benninghoven and W.K. Sichtermann, Anal. Chem., 50 (1978) 1180-1184. 130 R.J. Day, S.E. Unger, and R.G. Cook, J. Am. Chem. Soc, 101 (1979) 1180-1182. - 137 -131 F.M. Devienne and J.C. Roustan, Org. Mass Spectrom., 17 (1982) 173-181. 132 M.L. Vestal, Mass Specrom. Rev., 2 (1983) 447-480. 133 R.S. Johnson and K. Biemann, Biochemistry, 26, (1987) 1209-1214. 134 H. Egge and J. Peter-Katalinic, in "Mass Spectrometry in the Health and Life Sciences" 401-450, Eds. A.L . Burlingame and N . Castagonli, Jr., Elsevier Science Publishers, Amsterdam, 1985. 135 V.N. Reinhold and S.A. Carr, Mass Spec. Rev., 2 (1983) 153-221. 136 L. Pointer, R. Karrer, K. Jiang, K.; B.L. Gillece-Castro, and A.L . Burlingame, Proc. 36th Con. Mass Spectrom. Allied Topics, 36 (1988) 921-922. 137 M . Barber, R.S. Bordoli, G J . Elliott, R.D. Sedgewick, and A.N. Tyler, Anal. Chem., 54 (1982) 645A-657A. 138 R H . Field, J. Phys. Chem., 26 (1982) 5115-5123. 139 R.A.W. Johnson, I.A.S. Lewis, and M.E. Rose, Tetrahedron, 39 (1983) 1597-1603. 140 G. Puzo, and J.-C. Prome, Org. Mass Spectrom., 19 (1984) 448-451. 141 R.M. Caprioli, Anal. Chem., 55 (1983) 2387-2389. 142 R.M. Caprioli and L.A. Smith, Biomed. Mass Spectrom., 11 (1984) 392-395. 143 M . Barber, R.S. Bordoli, G J . Elliot, R.D. Sedwick, N . Tyler, J. Chem. Soc, Faraday Trans. 1, 79 (1983) 1249-1255. 144 E. de Pauw, G. Pelzer, V.D. Dao, and J. Marien, Biochem. Biophys. Res. Comm., 123 (1984) 27-32. 145 J.L. Gower, Biomed. Mass Spectrom., 12 (1985) 191-196. 146 M . Arita, M . Iwamori, T. Higuchi, Y . Nagai, / . Biochem., 93 (1983) 319-322. 147 S.A. Martin, C.E. Costello, and K. Biemann, Anal. Chem., 54 (1982) 2362-2368. 148 A. Dell, Adv. Carbohydr. Chem. Biochem., 45 (1987) 19-72. 149 M.A. Postumus, P.G. Kistemaker, H.L.C. Meuzelaar, and M.C. Ten Noever de Brauw, Anal. Chem., 50 (1978) 985-991. 150 F. Heresch, E.R. Schmid, and J.F.K. Huber, Anal. Chem., 52 (1980) 1803-1807. 151 R. J. Cotter, Anal. Chem., 52 (1980) 1767-1770. - 138 -152 H J . Heinen, S. Meier, H. Vogt, and R. Wechsung, Adv. Mass Spectrom., 8 (1980) 942-153 D. Zakett, A.E. Svhoen, R.G. Cooks, and P.H. Hemberger, / . Am. Chem. Soc, 103 (1981) 1295-1296. 154 K. Balasnmugan, R.A. Dang, R J . Day, D.M. Hercules, Anal. Chem., 53 (1981) 2296-2298. 155 G.J.Q. van der Peyl, K. Isa, J. Haverkamp, and P.G. Kistemaker, Org. Mass Spectrom., 16 (1981) 416-420. 156 S.W. Gramham, P. Dowd and D.H. Hercules, Anal. Chem., 54 (1982) 649-654. 157 R. Stoll and F.W. Rollgen, Z. Naturforsch., 31A (1982) 9-14. 158 U . Seydel and B. Lindner, in Ion Formation from Organic Solids, Springer Series in Chemical Physics 25, 240-244, Ed. A. Benninghoven, Springer-Verlag, 1983. 159 K. Takayama, N . Qureshi, K. Hyver, J. Honovich, R J . Cotter, P. Mascagni, and H. Schneider, / . Biol. Chem., 261 (1986) 10624-10631. 160 R J . Cotter, J. Honovich, N . Qureshi, and K. Takayama, Biomed. Environ. Mass Spectrom., 14 (1987) 591-598. 161 F. Hillenkamp, in Ion Formation from Organic Solids, Springer Series in Chemical Physics 25, 190-205, Ed. A. Benninghoven, Springer-Verlag, 1983. 162 R J . Cotter, J. Honovich, J. Olthoff, P. Demirev, and M . Alaim, in Ion Formation from Organic Solids IF OS III, Springer Proceedings in Physics 9, Springer-Verlag, 1986, 142-146, Ed. A. Benninghoven, Springer-Verlag, 1986. 163 D.M. Hercules, in Ion Formation from Organic Solids IFOS III, Springer Proceedings in Physics 9, Springer-Verlag, 1986, 147-152, Ed. A. Benninghoven, Springer-Verlag, 1986. 164 F. Hillenkamp, D. Holtkamp, M . Karas, and P. Kliisener, in Ion Formation from Organic Solids IFOS III, Springer Proceedings in Physics 9, Springer-Verlag, 1986, 153-167, Ed. A. Benninghoven, Springer-Verlag, 1986. - 139 -165 B. Lindner and U . Seydel, in Ion Formation from Organic Solids IFOS III, Springer Proceedings in Physics 9, Springer-Verlag, 1986, 158-162, Ed. A. Benninghoven, Springer-Verlag, 1986. 166 M.P. Chiarelli, D.A. McCrery, and M.L. Gross, in Ion Formation from Organic Solids IFOS III, Springer Proceedings in Physics 9, Springer-Verlag, 1986, 204-208, Ed. A. Benninghoven, Springer-Verlag, 1986. 167 M.P. Chiarelli and M.L. Gross, Int. Mass Spectrom. Ion Proc, 78 (1987) 37-52. 168 M . Karas, D. Bachmann, U. Bahr, and F. Hillenkamp, Int. Mass Spectrom. Ion Proc, 78 (1987) 53-68. 169 J. Grotemeyer, K. Walter, U . Boesl, and E.W. Schlag, Int. Mass Spectrom. Ion Proc, 78 (1987) 69-84. 170 M.L. Coates and CL.Wilkins, Biomed. Mass Spectrom., 12 (1985)424-428. 171 M.L. Coates and CL.Wilkins, Anal. Chem., 159 (1987) 197-200. 172 Z. Lam, M.B. Comisarow, G.G.S. Dutton, D. Weil, and A. Bjarnason, Rap. Commun. Mass Spectrom., 1 (1987) 83-87. 173 Z. Lam, M.B. Comisarow, and G.G.S. Dutton, Anal. Chem., 60 (1988) 2304-2306. 174 E.W. Schlag and H.J. Neusser, / . Am. Chem. Soc, 103 (1981) 5058-5061 and Acc. Chem. Res., 16 (1983) 355-360. 175 J.F. Ready, "Effect of High-Power Laser Radiation", Academic Press, New York, 1971. 176 F. Hillenkamp, E. Unsold, R. Konfmann, and R. Nitsche, Appl. Phys., 8 (1975) 341-348. 177 E.D. Hardin and M.L. Vestal, Proc 28th Con. Mass Spectrom. Allied Topics, 28 (1980) 616-617. 178 G.D. Davis, Adv. Mass Spectrom., 8 (1980) 1012-1013. 179 M.A. Postumus, Anal. Chem., 50 (1978) 985-991. 180 M.B. Comisarow and A.G. Marshall, Chem. Phys. Lett., 25 (1974) 282-283. 181 M.B. Comisarow and A.G. Marshall, Chem. Phys. Lett., 26 (1974) 489-490. - 140 -182 D.A. McCrery and M.L. Gross, Anal. Chim. Acta, 178 (1985) 91-103. 183 M.B. Comisarow, / . Chem. Phys., 69 (1978) 4097-4104. 184 M B . Comisarow, Adv. Mass Spectrom., 7 (1978) 1042-1046. 185 M.B. Comisarow, Adv. Mass Spectrom., 8 (1980) 1698-1706. 186 F.W. McLafferty, Interpretation of Mass Spectra, 3rd Edit, University Science Book, New York, 1980. 187 B. Domon and C.E. Costello, Gly conjugates, 5 (1988) 397-409. 188 G.J.Q. van der Peyl, J. Haverkamp, P.G. Kistemaker, Int. J. Mass Spectrom. Ion Phys., 42 (1982) 125-141. 189 R J . Cotter and A.L. Yergey, Anal. Chem., 53 (1981) 1306-1307. 190 G.J.Q. van der Peyl, W.J. van der Zonde, K. Bederski, A.J.H. Boerboom, and P.G. Kistemaker, / . Int. Mass Spectrom. Ion Phys., 47 (1983) 7-10. 191 G.J.Q. van der Peyl, K. Isa, J. Haverkamp, and P.G. Kistemaker, / . Int. Mass Spectrom. Ion Phys., 47 (1983) 11-14. 192 R.B. van Breemen, M . Snow, and R.J. Cotter, Int. J. Mass Spectrom. Ion Phys., 49 (1983) 35-50. 193 R.J. Cotter, Anal. Chem., 53 (1981) 719-720. 194 F. Heresch, in Ion Formation from Organic Solids, Springer Series in Chemical Physics 25, Springer-Verlag, 1983, 217-221, Ed. A. Benninghoven, Springer-Verlag, 1983. 195 G.G.S. Dutton and T.E. Folkman, Carbohydr. Res., 78 (1980) 305-15. 196 D.N. Karunaratne, Ph. D. Thesis, University of British Columbia, (1985). 197 G.G.S. Dutton, H. Parolis, J.-P. Joseleau, and M.-F. Marais, Carbohydr. Res., 149 (1986) 411-424. 198 G.G.S. Dutton, H. Parolis, and L.A.S. Parolis, Carbohydr. Res., 170 (1987) 193-206. 199 L.A.S. Parolis, H. Parolis, H . Niemann, and S. Strim, Carbohydr. Res., 179 (1988) 301-314. 200 C T . Greenwood, in "The Carbohydrates" Vol. 2B, 504, Eds., W. Pigman and D. Horton, Academic Press, New York, 1970. - 141 -201 P.A. Sandford and J. Baird, in "The Polysaccharides" Vol. 2, 471, Ed. G.O. Aspinall, Academic Press, New York, 1983. 202 A J . Mort and W.D. Bauer, / . Biol. Chem., 257 (1982) 1870-1875. 203 P.-E. Jannson, B. Lindberg, and P.A. Sandford, Carbohydr. Res., 124 (1983) 135-139. 204 M.-S. Kuo and A.J. Mort, Carbohydr. Res., 145 (1986) 247-265. 205 P. Mesner and F.M. Unger, Biochem. Biophys. Res. Comm., 96 (1980) 1003-1010. 206 H.J. Jennings, K.-G. Rosell, and K J . Johnson, Carbohydr. Res., 105 (1982) 45-56. 207 M.A. Schmidt and K. Jann, FEMS Microbiol. Lett., 14 (1982) 69-74. 208 W.F. Vann and K. Jann, Infect. Immun., 25 (1979) 85-92. 209 W.F. Vann, T. Soderstrom, W. Egan, F.-P. Tsui, R. Scheerson, I. 0rskov, and F. 0rskov, Infect. Immun., 39 (1983) 623-629. 210 B. Jann, P. Hofmann, and K. Jann, Carbohydr. Res., 120 (1983) 131-141. 211 B. Jann, R. Ahrens, T. Dengler, and K. Jann, Carbohydr. Res., 177 (1988) 273-277. 212 R. Ahrens, B. Jann, K. Jann, and H. Blades, Carbohydr. Res., 179 (1988) 223-231. 213 T. Dengler, B. Jann, and K. Jann, Carbohydr. Res., 142 (1985) 269-276. 214 E J . McGurie and S.B. Binkley, Biochem., 3 (1964) 247-251. 215 G.G.S. Dutton, H. Parolis, and L.A.S. Parolis, Carbohydr. Res., 170 (1987) 193-206. 216 M.R. Lifely, J.C. Lindon, J.M. Williams, and C. Moreno, Carbohydr. Res., 143 (1985) 191-205. 217 W. Egan, T.-Y. Liu, D. Dorow, J.S. Cohen. J.D. Robbins. E.C. Gotshich, and J.B. Robbins, Biochem., 16 (1977) 3687-3692. 218 A.N. Anderson, H. Parolis, L.A.S. Parolis, Carbohydr. Res., 163 (1987) 81-90. 219 K. Jann and M.A. Schmidt, FEMS Microbiol. Lett., 7 (1980) 79-81. 220 W. Fischer, M.A. Schmidt, B. Jann, and K. Jann, Biochem., 21 (1982) 1279-1284. 221 M.L. Rodriguez, B. Jann, and K. Jann, Carbohydr. Res., 173 (1988) 243-253. 222 B. Jann, T. Dengler, K. Jann, FEMS Microbiol. Lett., 29 (1985) 257-261. 223 P. Hofmann, B. Jan, and K. Jann, Eur. J. Biochem., 147 (1985) 601-609. - 142 -224 F.-P. Tsui, W. Egan, M.F. Summers, R.A. Byrd, R. Schneerson, and J.B. Robbins, Carbohydr. Res., 173 (1988) 65-74. 225 T. Dengler, B. Jann, and K. Jann, Carbohydr. Res., 150 (1986) 233-240. 226 P. Hofmann, B. Jann, and K. Jann, Carbohydr. Res., 139 (1985) 261-271. 227 L .M. Beynon, Ph. D. Thesis, University of British Columbia, (1988). 228 L . M . Beynon and G.G.S. Dutton, paper in preparation for Carbohydr. Res. - 143 -APPENDIX I The structural elucidation of bacterial capsular polysaccharides using laser desorption ionization Fourier transform ion cyclotron resonance spectroscopy (l.d.i.-F.t.-i.c.r.). - 144 -The structural elucidation of bacterial capsular polysaccharides using laser desorption ionization F o u r i e r transform ion cyclotron resonance spectroscopy (l.d.i.-F.t.-i.c.r.). ZAMAS LAM, MELVTN B. COMISAROW, GUY G. S. DUTTON Department of Chemistry, The University of British Columbia, Vancouver, B.C., V6T1Y6 (Canada) DAVID A. WEIL Nicolet Analytical Instruments, 5225-1 Verona Road, Madison, WI 53711-0508 (USA) AND ASGEIR BJARNASON Science Institute, University of Iceland, Dunhaga 3,107 Reykjavik, (Iceland) SPONSOR REFEREE: Alma L. Burlingame, University of California, USA Rapid Communication in Mass Spectrometry, 1 (1987) 83-86 (Received 1 Sept 1987; accepted 1 Sept 1987) INTRODUCTION Bacterial polysaccharides are important in immunology as potential vaccines1 and it is desirable that the chemical structure of these biopolymers be established. An important aspect of such structural studies involves the determination of the sequence of the monosaccharide residues in the biopolymer. The molecular weights of these biopolymers are in the range of 106 daltons1, obviating direct use of mass spectrometric methods for this sequencing. It is now well documented1 that the structures of most bacterial polysaccharides can be considered in terms of repeating units, which may contain two to seven monosaccharide residues, depending upon the species and the strain. Furthermore, there exist endoglycanases, associated with bacterial viruses (bacteriophages), capable of depolymerising the native polysaccharides into oligosaccharides corresponding to such repeating units2"4. The enzymatic depolymerisation reactions take place under neutral conditions and the oligosaccharides formed have the same sequence of monosaccharide residues as the native polymers. The molecular weight of, for example, a hexasaccharide repeating unit is approximately 10 daltons and within the mass range of modern mass spectrometers. The combination of bacteriophage degradation to form oligosaccharides and - 145 -mass spectrometric sequencing of these oligosaccharides thus has considerable potential for bacterial polysaccharide analysis. Oligosaccharides are involatile compounds whose mass spectrometric analysis has been greatly facilitated by the development of desorption ionisation techniques. The two most frequently used desorption methods are desorption chemical ionisation (d.c.i.)5 , 6 and 7 o fast atom bombardment (f.a.b.) ' . These ionisation techniques usually give abundant sequence information on derivatised samples and we have recently sequenced a capsular antigen from Escherichia coli K44 using d.c.i.9. Unfortunately, these techniques provide little or no information on either the positions of linkage between the saccharide residues or the anomeric configuration of the linkage. Collision activation dissociation (c.a.d.) experiments10 and mass ion kinetic energy (m.i.k.e.s.) experiments11 have been carried out on either permethylated disaccharides or methyl glycosides to distinguish the position of linkage10 and the anomeric configuration11 with varying degrees of success. It is obviously preferable, whenever possible, to sequence an oligosaccharide without derivatisation. Laser desorption ionisation (l.d.i.) has been shown to produce molecular or pseudomolecular ions from underivatised organic molecules, usually with time-of flight (t.o.f.) mass analysers. T.o.f. mass spectrometers give the entire mass spectrum from a single laser pulse, although the mass resolution is typically l o w 1 2 , 1 3 . Fourier transform ion cyclotron resonance (F.t.-i.c.r.) mass spectrometry14 ,15 can be used to overcome this resolution limitation. F.t.-i.c.r. provides ultrahigh mass resolution16"17, high mass capability16"17 and is inherently a pulse experiment16"17, and thus is well suited for l .di . experiments19,20. Unit-mass resolution l.d.i.-F.t.-i.c.r. mass spectrometry, using a Nicolet FTMS-1000 instrument, has shown the potential of the technique for carbohydrate analysis2 1 , 2 2. We report here a high resolution, accurate mass l.d.i.-F.t.-i.c.r. mass spectrum of an underivatised bacterial capsular oligosaccharide. This spectrum can be interpreted to give complete sequence information and also, in some cases, the linkage positions of the individual sugar residues. - 146 -The structure of the capsular polysaccharide of Klebsiella K44 (Figure 1) has been investigated previously 2 3 , 2 4. The linear pentasaccharide repeating unit, generated by bacteriophage enzyme action on the polysaccharide, consists of a glucuronic acid residue, two rhamnose residues, two glucose residues and an 0-acetyl group. F.a.b.-m.s. only provides total sequence information if this particular oligosaccharide, obtained from bacteriophage degradation, is reduced and peracetylated25. Glucuronic acid - Rhamnose - Rhamnose - Glucose - Glucose Fig. 1. Structure of O-de-acetylated oligosaccharide obtained from Klebsiella K44 by bacteriophage degradation. E X P E R I M E N T A L Mass spectra were obtained with a Nicolet FTMS-2000 dual cell Fourier transform mass spectrometer equipped with a 3.0 T superconducting magnet and a Nicolet laser desorption interface. The laser was a Tachisto 216 pulse CO2 laser operating as a stable resonator with aperture controlled beam characteristics. Equipped with a total reflector, the laser delivers up to 2 J in a 40 ns wide pulse emitting at 10.59 micron. Output energy can be controlled by adjusting the aperture and was estimated to be on the order of 0.05 J per pulse for this experiment. The spot size for the focus laser beam on the probe tip was on the order of 100 micron. Approximately 1 mg of the sample was dissolved in 0.2 ml of methanol solution saturated with NaCl / KBr. A drop or two of water was added to completely dissolve the sample. This solution was transferred to and allowed to evaporate on the stainless steel - 147 -probe tip of the direct insertion probe of the mass spectrometer. The vacuum system was pumped down to ~lxlO" 8 torr after insertion of the sample probe, prior to mass analysis. Mass spectra were collected in the broadband mode16 with a high frequency cutoff at 500 KHz (corresponding to a lower mass limit of ~90 amu). Normally, 64K data point transients were collected and when needed the spectra from several laser shots were averaged to increase the S/N ratio. After each laser pulse, the probe was rotated so that the laser always struck a fresh spot on the sample surface. After the laser pulse, a delay of 5-10 s was employed to reduce the pressure of neutrals in the cell before ion detection. For exact mass measurements, the series of ions corresponding to K n B r n _ i + was used as an internal calibranL Results and Discussion Table I contains a list of all the l.d.i.-F.t.-i.c.r. positive structural ions, above 5% of the base peak, of the O-de-acetylated oligosaccharide obtained from Klebsiella K44 by bacteriophage degradation. In the spectrum, pseudomolecular and fragment ions were observed in abundance, typically as the sodiated and potassiated ions. The calculated and observed masses in Table I agree within 30 ppm. Three distinct types of fragmentation pattern were observed as illustrated in Figures 2 and 3. These fragmentation routes do not belong to the classical A - K fragmentation series for permethylated methyl glycosides described by Kochetkov and Chizhov 2 6 , 2 7 . For convenience, the observed fragmentation pathways are labeled from L onwards to distinguish them from Kochetkov's series. The glycosidic bond cleavages (L-cleavage) provided some sequence information (Figure 2) while the ring cleavages (M- and N-cleavages) confirmed the sequence of the oligosaccharide and gave some positions of linkage between the individual sugar residues (Figure 3). - 148 -657.2223 511.1629 365.0984 333.1220 495.1692 815.2431 Fig. 2. Some fragment ions arising from glycosidic bond cleavages (L-cleavage) in the l.d.i.-F.t.-i.c.r. spectrum of the 0-de-acetylated oligosaccharide obtained from Klebsiella K44 by bacteriophage degradation. Na + and K + have been omitted from the structures in this figure. The formulae of the cationised ions are assigned in Table I to specific masses . 699.2308 537.1790 597.2062 741.2286 Fig. 3. Some fragment ions arising from ring cleavages (M- and N-cleavages) in the l.d.i.-F.t.-i.c.r. spectrum of the O-de-acetylated oligosaccharide obtained from Klebsiella K44 by bacteriophage degradation. Na + and K + have been omitted from the structures in this figure, as in Fig. 1. - 149 -TABLE I . POSITIVE ION L.D.I.-F.T.-I.C.R. POSITIVE ION SPECTRUM OF THE O-DE-ACETYLATED OLIGOSACCHARIDE OBTAINED FROM Klebsiella K44 BY BACTERIOPHAGE DEGRADATION Observed Calculated Relative Mass Proposed mass mass intensity error structure (a) (amu) (amu) (%) (ppm) 831.2297 831.2167 8.6 15.6 [(2Hex,2deoxyHex,HexA)-H20+K]+ 815.2431 815.2428 24.9 0.4 [(2Hex,2deoxyHex,HexA)-H20+Na]+ 771.2657 771.2530 9.7 16.5 [(2Hex,2deoxyHex,HexA)-H20-C02+Na]+ 741.2286 741.2430 6.0 -18.6 [(2deoxyHex,Hex,HexA)+C4H702+Na]+ 715.2131 715.2058 11.1 10.2 [(2Hex,2deoxyHex)+C2H302+K]+ 699.2308 699.2318 37.4 -1.4 [(2Hex,2deoxyHex)+C2H302+Na]+ 697.2088 697.1952 5.9 19.4 [(2Hex,2deoxyHex)+C2H302-H20+K]+ 681.2240 681.2213 25.0 4.0 [(2Hex,2deoxyHex)+C2H 30 2-H 20+Na]+ 673.2067 673.1952 18.5 17.1 [(2Hex,2deoxyHex)+K]+ 657.2223 657.2213 47.9 1.6 [(2Hex,2deoxyHex)+Na]+ 655.1926 655.1846 26.0 12.2 [(2Hex,2deoxyHex)-H20+K]+ 639.2097 639.2107 100.0 -1.5 [(2Hex,2deoxyHex)-H20+Na]+ 613.1672 613.1741 13.4 11.2 [(2deoxyHex,HexA)+C4H903+K]+ 597.2062 597.2001 32.8 10.1 [(2deoxyHex,HexA)+C4H903+Na]+ 579.1959 579.1896 9.1 10.9 [(2deoxyHex,HexA)+C4H903-H20+Na]+ 553.1681 553.1530 31.2 27.3 [(2Hex,deoxyHex)+C2H30+K]+ 537.1790 537.1791 80.5 -0.2 [(2Hex,deoxyHex)+C2H302+Na]+ 519.1643 519.1685 5.6 -8.0 [(2Hex,deoxyHex)+C 2H30-H20+Na]+ 511.1629 511.1634 17.6 1.0 [(2Hex,deoxyHex)+Na]+ 509.1300 509.1267 6.9 6.4 [(2Hex,deoxyHex)-H20+K]+ 495.1692 495.1684 27.4 1.6 [(2deoyHex,Hex)+Na]+ 493.1526 493.1526 27.2 -1.0 [(2Hex,deoxyHex)-H20+Na]+ 477.1580 477.1579 13.7 0.3 [(2deoxyHex,Hex)-H20+Na]+ 365.0984 365.1055 6.5 -19.3 [(2Hex)+Na]+ 349.1127 349.1105 6.9 6.2 [(Hex,deoxyHex)+Na]+ 347.0981 347.0949 18.4 9.3 [(2Hex)-H20+Na]+ 333.1220 333.1156 9.9 19.3 [(2deoxyHex)+Na]+ 315.1083 315.1050 6.3 10.4 [(2deoxyHex)-H20+Na]+ (a) Hex = hexose, deoxyHex = deoxyhexose, HexA = hexuronic acid and the formation of each glycosidic bond involves the loss of a molecule of water.  - 150 -For ring cleavage, two fragmentation routes were observed (Figure 4). The M-series fragments, with the positive charge residing on the reducing end, contain C - l and C-2, while N-series fragments possibly hold C-3, C-4, C-5 and C-6. M-series N-series Fig. 4. Two different types of ring cleavages. As both types of ring cleavage fragments were observed, the reducing end could be determined without reducing the oligosaccharide prior to mass spectrometric analysis (Figure 3). From the Klebsiella K44 oligosaccharide, two M-series fragments, m/z = 699.2308 and m/z = 537.1790 were observed. The first fragment corresponds to two hexoses plus two deoxyhexoses plus a C2H302ring fragment. The second corresponds to two hexoses plus one deoxyhexose plus a C2H3O ring fragment. The M-series fragment derived from the terminal glucuronic acid was C2H3O2, while the 2-linked rhamnose gave a C 2 H 3 O fragment, presumably by a glycosidic oxygen loss (Figure 5). Two N-series fragments were also observed from this oligosaccharide, m/z = 741.2286 and m/z = 597.2062. The first N-series fragment corresponds to an uronic acid, plus two deoxyhexoses plus a hexose, plus a C4H7O2 ring fragment. The second N-series fragment corresponds to an uronic acid plus two deoxyhexoses plus a C4H9O3 ring fragment. The 3-linked terminal glucose gave a C4H7O2 fragment, while the adjacent 4-linked glucose gave a fragment of C4H9O3 (Figure 6). This may indicate that 4-linked residues give a C4H9O3 fragment, and 3-linked residues give a C 4 H 7 O 2 fragment. Dehydration of high molecular - 151 -weight oligosaccharides, however, is very facile, and further work is required to confirm these observations. - 152 -HO HO—i O i ^ . J O OH3C2' OH OH m/z=699.2308 H 3 C 2 - 0 OH OH m/z=537.1790 OH Fig. 5. M-series fragment ions observed for l.d.i.-F.t.-i.c.r. of O-de-acetylated oligosaccharide obtained from Klebsiella K44 by bacteriophage degradation. COOH HO •O k OH \ 1 \ HO O J N i m/z=741.2286 COOH • \f 7 OH \HO, 0^ m/z=597.2062 OH OH r O-C 4 Ho0 2 OH Fig. 6. N-series fragment ions observed for l.d.i.-F.t.-i.c.r. of O-de-acetylated oligosaccharide obtained from Klebsiella K44 by bacteriophage degradation. - 153 -High resolution l.d.i.-F.t.-i.c.r. shows promise for complete sequencing, and for determination of some linkage positions for the underivatised, linear oligosaccharide shown in Figure 1. Further work is in progress to corifirm the generality of our proposed ring cleavage fragmentation patterns in other oligosaccharides. ACKNOWLEDGMENT This research was supported by the Natural Sciences and Engineering Research Council of Canada. REFERENCES 1 L. Kenne and B. Lindberg, in G.O. Aspinall (Ed.), "The Polysaccharides," Vol. 1 and 2, Academic Press, New York (1983) 2 G.G.S. Dutton, J.L. di Fabio, D.M. Leek, E.H. Merrifield, J.R. Nunn and A . M . Stephen, Carbohydr. Res. 97, 127-38 (1981) 3 H. Geyer, K. Himmelspach, B. Kwiatkowski, S. Schlecht and S. Strim, Pure Appl. Chem. 55, 637-53 (1983) 4 D. Rieger-Hug and S. Stirm, Virology 113, 363-78 (1981) 5 A.K. Ganguly, N.F. Cappuccino, H . Fujiwara and A.K. Bose, / . Chem. Soc, Chem. Comm. 148 (1979) 6 K.-I. Harada, S. Ito, N . Takeda, M . Suzuki and A. Tatematsu, Biomed. Mass Spectrom. 10, 5-12,(1983) 7 H.R. Morris, A. Dell, M . Panico and R.A. McDowell, in A.L . Burlingame and N . Castagnoli, Jr. (Eds.), "Mass Spectrometry in the Health and Life Sciences", p. 363-78. Elsevier, Amsterdam (1985) 8 H. Egge and J. Peter-Kiitalinic, in A.L. Burlingame and N. Castagnoli, Jr. (Eds.), "Mass Spectrometry in the Health and Life Sciences", p. 401-24, Elsevier, Amsterdam (1985) 9 G.G.S. Dutton and A.V.S. Lim, submitted for publication 10 E.G. de Jong, W. Heerma and G. Dijkstra, Biomed. Mass Spectrom. 7, 127-31 (1980) 11 G. Puzo, J.-C. Prome, and J.-J. Fournie, Carbohydr. Res. 140, 131-34 (1985) - 154 -12 M.A. Posthumus, P.G. Kistemaker, H.L.C. Meuzelaar and M.C. Ten Noever de Brauw, Anal Chem. 5 0 , 985-91 (1978) 13 R J. Cotter and J.-C. Tabet Anal., Chem. 56, 1662-67 (1984) 14 M.B. Comisarow and A.G. Marshall, Chem. Phys. Lett. 2 5 , 282-84 (1974) 15 M.B. Comisarow and A.G. Marshall, Chem. Phys. Lett. 26, 489-90 (1974) 16 M.B. Comisarow, Adv. Mass Spectrom. 8 ,1698-706 (1980) 17 R.B. Cody, Jr., J.A. Kinsinga, S.G. Hadai, I.J. Austin and F.W. McLafferty, Anal. Chim. Acta 1 7 8 , 43-66 (1985) 18 D.A. Laude, Jr., C L. Johlman, R S . Brown, D.A. Weil and C L . Wilkins, Mass Spectrom. Rev. 5 , 107-66 (1986) 19 D.A. McCrery, E.B. Ledford, Jr. and M.L. Gross, Anal. Chem. 5 4 , 1435-37 (1982) 20 C L . Wilkins, D.A. Weil, C.L.C. Yang and G.F. Ijames, Anal. Chem. 5 7 , 520-24 (1985) 21 M.L. Coates and C L . Wilkins, Biomed. Mass Spectrom. 1 2 , 424-28 (1985) 22 M.L. Coates and C L. Wilkins, Anal. Chem. 5 9 , 197-200 (1987) 23 G.G.S. Dutton and T.E. Folkman, Carbohydr. Res. 7 8 , 305-15 (1980) 24 D.N. Karunaratne, Ph. D. Thesis, UBC, (1985) 25 A. Dell and P.R. Tiller, Biochem. Biophys. Res. Comm. 1 3 5 , 1126-34 (1986) 26 O.S. Chizhov and N.K. Kochetkov, Advan. Carbohydr. Chem. 2 1 , 39-93 (1966) 27 J. Lonngren and S. Svensson, Advan. Carbohydr. Chem. 29, 41-106 (1974) - 155 -APPENDIX II Laser desorption ionization Fourier transform ion cyclotron resonance mass spectra of underivatized oligosaccharides. -156 -s o u D p u n q D s } n j o s q v OB 09 0* 03 I » • » ! • • • • » • — i i i I i i—i—• i i • • . I • i o h O O • o Cn O O OO 001 Spec, rv.i Low resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K44 de-0-acetylated oligosaccharide. - 1 5 7 -Nominal mass Measured mass Rel . i n t e n s i t y Abs . In tens i ty 100 amu 100.39799 amu 3.1432 X 2585.546875 m 101 amu 101.02504 amu • 5.8857 X 4841.406250 m 10B amu 108.23066 amu 1.4945 X 1229.296875 m 111 amu 111.00949 amu 1.9B83 X 1635.546875 m 112 amu 112.017B5 amu 1.9012 X 1629.687500 m 113 amu 112.91869 amu 1.3363 X 1099.218750 m 113 amu 113.02414 amu 51 .4519 X 42323.046875 m 114 amu 114.03139 amu 3.9292 X 3232.031250 m 115 amu 115.00537 amu 1.6203 X 1332.812500 m 115 amu H5.041B2 amu 1.6493 X 1356.640625 m 11? amu 119.03660 amu 3.4168 X 2810.546875 m 125 amu 125.02448 amu 7.3421 X 6039.453125 m 126 amu 126.03230 amu 3.3674 X 2769.921875 m 127 amu 127.04004 amu 16.8654 X 13B73.046875 m 128 amu 128.04697 amu 1.9608 X 1612.890625 m 129 amu 129.02072 amu 3.7601 X 3092.968750 m 131 amu 131.03692 amu 6.5524 X 5389.843750 m 131 amu 131.31233 amu 2.6418 X 2173.046875 m 133 amu 133.04733 amu 4.5641 X 3754.296875 m 137 amu 137.02520 amu 1.4384 X 1183.203125 m 139 amu 139.00274 amu 3.3075 X 2720.703125 m 139 amu 139.04021 amu 1.9841 X 1632.031250 m 141 amu 141 .01996 amu 4.9122 X 4040.625008 m 142 amu 142.02819 amu 4.0954 X * 3368.750000 m 143 amu 142.88925 amu 3.7919 X 3119.140625 m 143 amu 143.03526 amu 53.0271 X 43618.750000 m 144 amu 144.04105 amu 6.6018 X 5430.468750 m 145 amu 145.05173 amu 20.9147 X 17203.906250 m 146 amu 146.05591 amu 1.5614 X 1284.375000 m 149 amu 149.04416 amu 1 .5681 X 1289.843750 m 153 amu 153.01888 amu 1.52B6 X 1257.421875 m 154 amu 153.54873 amu 1.5571 X 1280.859375 m 155 amu 155.03429 amu 3.0084 X 2474.609375 m 157 amu 157.01365 amu 3.9738 X 3268.750000 m 159 amu 159.02930 amu 2.3820 X 1959.375000 m 161 amu 160.85783 amu 1.4099 X 1159.765625 m 161 amu 161.04565 amu 15.2845 X 12572.656250 m 162 amu 162.05137 amu 1.6711 X 1374.609375 m 163 amu 162.87B79 amu 1.2979 X 1067.578125 m 163 amu 163.06320 amu 3.3759 X 2776.953125 m 165 amu 164.87663 amu 1.3335 X 1096.875000 m 166 amu 165.97945 amu 2.4841 X 2043.359375 m 167 amu 167.03378 amu 1.9850 X 1632.812500 m 173 amu 173.04790 amu 9.2673 X 7623.046875 m 175 amu 175.02766 amu 8.5859 X 7062.500000 m 177 amu 176.89859 amu 2.3274 X 1914.453125 m 179 amu 178.89672 amu 2.2538 X 1853.906250 m 179 amu 179.05863 amu 10.5481 X 8676.562500 m 181 amu 180.82961 amu 6.4299 X 5289.062500 m 163 amu 182.8252? amu 15.6867 X 12903.515625 m 185 amu 184.82379 amu 6.8112 X 5602.734375 ft 185 amu 185.04501 amu 11 .8915 X 9781.640625 m 186 amu 186.04913 amu 2.1018 X 1728.906250 m 187 amu 187.03633 amu 2.4604 X 2023.828125 m 189 amu 188.89590 amu 3.3560 X 2760.546875 m 189 amu 189.06228 amu 1.9940 X 1640.234375 m 191 amu 190.89556 amu 2.6004 X 2139.062500 m 193 amu 193.03628 amu 11.9836 X 9857.421875 m 197 amu 196.52167 amu 3.0616 X 2518.359375 m 197 amu 196.80179 amu 45.7220 X 37609.765625 m Spec. IV.2 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K44 de-O-acetylated oligosaccharide. - 158 -199 amu 198.57323 amu 1.4B31 X 1219 m 199 amu 198.79789 amu 100.0000 X 82257.421875 m 199 amu 199.17332 amu 2.3241 X 1911.718750 m 201 amu 200.50344 amu 3.2999 X 2714.453125 m 201 amu 200.79654 amu •54.2162 X 44596.875000 m 203 amu 202.80016 amu 2.7258 X 2242.187500 m 203 amu 203.05728 amu •5.7005 X 4689.062500 m 205 amu 205 .05097 amu 2.0472 X 1683.984375 m 212 amu 212.07516 amu 2.0724 X 1704.687500 m 215 amu 215.05690 amu 2.3910 X 1966.796875 m 217 amu 217.03003 amu 1.6445 X 1352.734375 m 219 amu 219.17677 amu 1.4636 X 1203.906250 m 221 amu 221.06929 amu 15.5704 X 12807.812500 m 222 amu 222.07255 amu 1.7281 X 1421.484375 m 225 amu 225.19204 amu 1.5334 X 1261.328125 m 227 amu 227.20813 amu 2.5696 X 2113.671875 m 231 amu 230.90875 amu 2.6361 X 2168.359375 m 233 amu 232.90348 amu 3.1651 X 2603.515625 m 233 amu 233.07082 amu 1.8549 X 1525.781250 m 235 amu 235.04196 amu 9.4083 X 7739.062500 m 243 amu 243.19780 amu 3.3427 X 2749.609375 m 245 amu 245.06697 amu 1.9807 X 1629.296875 m 247 amu 247.06507 amu 2.4414 X 2008.203125 m 249 amu 249.06055 amu 1.5714 X 1292.578125 m 251 amu 251.07501 amu 2.5449 X 2093.359375 m 253 amu 253.07754 amu 1 .5339 X 1261.718750 m 255 amu 255.01468 amu 1.6464 X 1354.296875 m 256 amu 256.24759 amu 2.1384 X 1758.984375 m 259 amu 259.08538 amu 2.0130 X 1655.859375 m 263 amu 263.08267 amu 4.7227 X 3884.765625 m 266 amu 266.21642 amu 1.4227 X 1170.312500 m 267 amu 267.07543 amu 1.4831 X 1219.921875 m 273 amu 273.01149 amu 7.7054 X 6338.281250 m 275 amu 275.07167 amu 2.2866 X 18B0.859375 m 277 amu 277.09481 amu 1.5714 X 1292.578125 m 287 amu 287.08251 amu 2.2110 X 1818.750000 m 289 amu 289.09055 amu 2.7334 X 2248.437500 «n 291 amu 290.54479 amu 1.6744 X 1377.343750 m 291 amu 291.10976 amu 17.9747 X 14785.546875 m 292 amu 292.12026 amu 2,3583 X 1939.843750 m 295 amu 294.69020 amu 1.3126 X 1079.687500 m 297 amu 296.69039 amu 1.4460 X 1189.453125 m 303 amu 303.07819 amu 5.7019 X 4690.234375 m 305 amu 305.08928 amu 4.3190 X 3552.734375 m 307 amu 307.10772 amu 16.5653 X 13626.171875 m 308 amu 308.13267 amu 2.6579 X 2186.328125 m 309 amu 309.12592 amu 6.1866 X 5089.062500 m 313 amu 313.10020 amu 2.3112 X 1901.171875 m 317 amu 316.67037 amu 2.8920 X 2378.906250 m 319 amu 318.67847 amu 3.5963 X 2958.203125 m 319 amu 319.11274 amu 3.6191 X 2976.953125 m 320 amu 320.22441 amu 2.8084 X 2310.156250 m 321 amu 321.08635 amu 15.5266 X 12773.437500 m 322 amu 322.10114 amu 2.1346 X 1755.859375 m 323 amu 323.10629 amu 4.9649 X 4083.984375 m 325 amu 325.11370 amu 2.9761 X 2448.046875 m 331 amu 331.99844 amu 1.7005 X 1398.828125 m 333 amu 333.13448 amu 1.3672 X 1124.609375 m 335 amu 335.12329 amu 1.4636 X 1203.906250 m 337 amu 337.10070 amu 1.3206 X 1066.328125 m 338 amu 338.23648 amu 2.8664 X 2357.812500 m 339 amu 339.10481 amu 4.5869 X 3773.046875 m 347 amu 347.09926 amu 1.2898 X 1060.937500 m Spec, rv.2 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K44 de-<9-acetylated oligosaccharide, con't. - 159 -350 amu 350.21382 amu 1 .39B5 X 11W.390625 A 351 amu 351.15329 amu 2.0406 X 1678.515625 m 353 amu 353.10851 amu 2.0016 X 1646.484375 m 357 amu 357.01166 amu 1 .6417 X 1350.390625 m 361 amu 361.12282 amu • 1.8126 X 1491.015625 m 365 amu 365.11105 amu 1.5324 X 1260.546675 m 367 amu 367.12362 amu 16.3179 X 13422.656250 m 368 amu 368.12341 amu 2.5891 X 2129.687500 m 369 amu 369.13584 amu 1.5543 X 1278.515625 m 375 amu 375.09450 amu 1.3197 X 1084.765625 m 379 amu 379.11441 amu 2.4604 X 2023.828125 m 381 amu 381.13461 amu 3.0910 X 2542.578125 m 383 amu 383.11986 amu 7.2989 X 6003.906250 m 391 amu 391.06797 amu 2.0016 X 1646.484375 m 393 amu 393.10957 amu 2.2790 X 1874.609375 m 395 amu 395.12418 amu 1.8506 X 1522.265625 m 397 amu 397.10614 amu 1.7390 X 1430.468750 m 403 amu 403.12981 amu 3.1485 X 2589.843750 m 405 amu 405.15696 amu 2.9956 X 2464.062500 m 409 amu 409.12720 amu 2.5805 X 2122.656250 m 421 amu 421.11636 amu 1.3401 X 1102.343750 m 423 amu 423.15313 amu 1.6502 X 1357.421B75 m 433 amu 433.14681 amu 1 .5495 X 1274.609375 m 435 amu 435.12693 amu 2.4380 X 2005.468750 m 437 amu 437.14533 amu 1.7442 X 1434.765625 m 445 amu 445.14367 amu 1.4284 X 1175.000000 m 449 amu 449.11259 amu 5.5651 X 4577.734375 m 450 amu 450.11620 amu 1.3563 X 1115.625000 m 451 amu 451.12473 amu 2.4019 X 1975.781250 m 453 amu 453.13386 amu 10.1021 X 8309.765625 m 454 amu 454.14077 amu 2.1754 X 1789.453125 m 461 amu 461.14057 amu 3.3023 X 2716.406250 m 465 amu 465.12886 amu 2.2068 X 1815.234375 m 467 amu 467.13366 amu 18.9397 X 15579.296675 m 468 amu 468.13519 amu 4.2801 X 3520.703125 m 469 amu 469.15059 amu 6.8934 X 5670.312500 m 471 amu 471.17449 amu 4.0769 X 3353.515625 m 477 amu 477.13700 amu 2.2433 X 1845.312500 m 478 amu 478.16007 amu 1.3658 X 1123.437500 m 479 amu 479.13361 amu 4.1809 X 3439.062500 m 483 amu 483.12755 amu 1.6083 X 1487.500000 m 485 amu 485.15393 amu 14.6548 X 12054.687500 m 486 amu 486.15599 amu 3.1784 X 2614.453125 m 490 amu 490.22520 amu 2.4594 X 2023.046875 m 493 amu 493.15579 amu 2.1883 X 1800.000000 m 495 amu 495.14310 amu 3.0312 X 2493.359375 m 499 amu 499.18398 amu 1.6939 X 1393.359375 m 508 amu 508.19807 amu 1.8102 X 1489.062500 m 509 amu 509.19470 amu 1.44B4 X 1191.406250 m 513 amu 513.15951 amu 13.5916 X 11180.078125 m 514 amu 514.16399 amu 3.2434 X 2667.968750 m 515 amu 515.17125 amu 1.6459 X 1353.906250 m 520 amu 520.14306 amu 2.1773 X 1791.015625 m 521 amu 521.17163 amu 7.1579 X 5887.890625 m 522 amu 522.17129 amu 2.0463 X 1683.203125 m 523 amu 523.10335 amu 1.4156 X 1164.453125 tn 525 amu 525.14567 amu 1.5063 X 1239.062500 m 527 amu 527.16297 amu 1.7005 X 1398.628125 m 529 amu 529.17132 amu 5.6107 X 4615.234375 m 535 amu 535.08291 amu 1.5533 X 1277.734375 m 537 amu 537.14524 amu 2.8279 X 2326.171875 m 539 amu 539.16367 amu 1.8259 X 1501.953125 m 541 amu 541.16333 amu 2.1588 X 1775.781250 m Spec. rv\2 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K44* de-O-acetylated oligosaccharide, con't. - 160 -549 amu 549.07773 amu 1.2703 X 1044.921375 % 551 amu 551.08894 amu 2.4352 X 2003.124996 m 553 amu 553.08434 amu 2.3796 X 1957.421875 m 555 amu 555.19607 amu 6.9736 X 5736.328125 m 556 amu 556.16368 amu • 2.0420 X 1679.687500 m 557 amu 557.16467 amu 1.7181 X 1413.281250 m 559 amu 559.17612 amu 1.287? X 1059.37500© m 567 amu 567.18662 amu 2.1826 X 1795.312500 m 573 amu 573.21254 amu 1 .6031 X 1483.203125 m 581 amu 581.17570 amu 1.6573 X 1363.281250 m 593 amu 593.03071 amu 2.1830 X 1795.703125 m 595 amu 595.04914 amu 2.3146 X 1903.906250 m 597 amu 597.12860 amu 1.9869 X 1634.375000 m 599 amu 599.22210 amu 1.6113 X 1325.390625 m 607 amu 607.12346 amu 1.4845 X 1221.093750 m 609 amu 609.14681 amu 1.7499 X 1439.453125 m 611 amu 611 .12476 amu 2.9680 X 2441.406250 m 613 amu 613.21547 amu 2.0192 X 1660.937500 m 615 amu 615.22409 amu 4.9725 X 4090.234367 m 616 amu 616.23230 amu 1.5115 X 1243.359375 m 622 amu 622.17071 amu 1.7575 X 1445.703125 m 623 amu 623.17006 amu 7.0278 X 57B0.B59375 m 624 amu 624.17683 amu 2.4414 X 2008.203125 m 625 amu 625.19818 amu 1.8264 X 1502.343750 m 627 amu 627.20173 amu 1.7922 X 1474.218750 m 629 amu 629.17021 amu 6.5838 X 5415.625000 m 630 amu 630.11762 amu 2.1341 X 1755.468750 m 631 amu 631.16373 amu 1 .4156 X 1164.453125 m 633 amu 633.27863 amu 1.8216 X 1498.437500 m 635 amu 635.17409 amu 1.4926 X 1227.734375 m 641 amu 641.15691 amu 2.8783 X 2367.57B125 m 645 amu 645.12459 amu 1.7542 X 1442.968750 m 647 amu 647.11616 amu 3.7060 X 3048.437500 m 653 amu 653.06394 amu 1 .4964 X 1230.859375 m 655 amu 655.10251 amu 1 .4370 X 1182.031250 m 667 amu 667.20917 amu 1.5106 X 1242.578125 m 668 amu 668.1332? amu 1.9765 X 1625.781250 m 669 amu 669.23758 amu 1.5814 X 1300.781250 m 671 amu 671.21441 amu 3.8005 X 3126.171875 m 672 amu 672.16005 amu 1.9589 X 1611.328125 m 675 amu 675.21431 amu 11.3895 X 9368.750000 m 676 amu 676.23102 amu 3.8803 X 3191.796875 m 677 amu 677.1?8?3 amu 1.3810 X 1135.937500 m 683 amu 683.21717 amu 4.2345 X 3483.203125 m 687 amu 687.25618 amu 1.8525 X 1523.828125 m 689 amu 689.21444 amu 5.1430 X 4230.468750 m 690 amu 690.32308 amu 1.7613 X 1448.828125 m 695 amu 695.13758 amu 3.6495 X 3001.953125 m 697 amu 697.12196 amu 3.9914 X 3283.203125 m 698 amu 698.12776 amu 1.3810 X 1135.937500 m 699 amu 699.21494 amu 1.4251 X 1172.265625 m 703 amu 703.23110 amu 1.9617 X 1613.671875 m 708 amu 708.43248 amu 1.4574 X 1198.828125 m 713 amu 713.13644 amu 2.3606 X 1941.796B75 m 715 amu 715.20980 amu 1.9769 X 1626.171875 m 717 amu 717.22120 amu 1.6635 X 1368.359375 m 731 amu 731.18433 amu 1.4954 X 1230.078125 m 743 amu 743.08624 amu 1.5624 X 1285.156250 m 755 amu 755.08885 amu 1.5628 X 1285.546875 m 757 amu 757.15537 amu 1.2898 X 1060.937500 m 773 amu 773.23565 amu 2.8222 X 2321.484375 m 791 amu 791 .31906 amu 5.1040 X 419B.437500 m B09 amu 809.34701 amu 2.3046 X 1895.703125 m Spec. IV.2 High resolution negative-ion l.d.i.-F.t.-i.cr. spectrum of Klebsiella K44 de-0-acetylated oligosaccharide, con't. - 161 -833 amu 833.4(9003 amu 1.7499 X } 4 2 2 - £ 5 2 J 2 5 m cm* amu 851.32165 amu 2.0154 X 1657.812500 * ?38 Z 937.58386 amu 1.7742 X 1459.375000 m Spec. IV.2 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K44 de-O-acetylated oligosaccharide, con't. 0 o—[ o J (.1—1 o O—l > cn-asoireui jo uirujoads I D I - T j - T P i uoi-3Apisod uopnjpsai M o q £AI aads 1 00 RELATIVE INTENSITY 50 - L 1 L — L L L I 1 I • ' • » . . . . c. c o c - 391 -- 163 -Nominal mass Measured mass R e l . intensity Abs. intensity 97. amu 97.02798 amu 31.0 X 10.9138 99. amu 99.04321 amu 7.1 X 2.5178 193. amu 103.03867 amu 11.6 X 4.0837 109. amu 109.02938 amu 4.1 X 1.4302 115. amu 115.03917 amu 5.1 X 1.7886 127. amu 127.03953 amu 41 .9 X 14.7449 131 . amu 131.03459 amu 6.1 X 2.1602 145. amu 145.04910 amu 100.0 X 35.2197 146. amu 146.05352 amu 7.5 X 2.6262 163. amu 163.06083 amu 56.0 X 19.7100 169. amu 169.05079 amu 10.6 X 3.7173 173. amu 173.04505 amu 2.9 X 1.0374 1B7. amu 187.06103 amu 17.3 X 6.1038 191 . amu 191.05566 amu 4.5 X 1.5852 199. amu 199.06069 amu 3.9 X 1.3691 205. amu 205.07119 amu 9.7 X 3.4275 229. amu 229.07090 amu 23.8 X 8.3918 245. amu 245.06231 amu 8.4 X 2.9712 247. amu 247.08389 amu 12.7 X 4.4607 259. amu 259.08421 amu 4.2 X 1.4658 265. amu 265.09554 amu 13.6 X 4.8013 271 . amu 271.08449 amu 12.3 X 4.3235 289. amu 289.09974 amu 5.5 X 1.9504 305. amu 305.08804 amu 33.2 X 11-7104 306. amu 306.08679 amu 5.3 X 1.8762 321 . amu 321.05457 amu 16.1 X 5.6655 325. amu 325.11401 amu 24.0 X 8.4424 326. amu 326.11003 amu 4.8 X 1.7002 365. amu 365.10297 amu 71 .7 X 25.2476 3&£> • amu 366.11907 amu 9.6 X 3.3799 367. amu 367.11674 amu 6.0 X 2.1262 381 . amu 381.08139 amu 23.8 X 8.3906 Spec. IV.4 High resolution positive-ion l.d.i.-F.L-i.c.r. spectrum of maltose. 0 asoirem jo amuoads T O I - T H - ! T ) T . uoi-9\vvs2w uopnpsai Moq C/AI ",0 RELATIVE INTENSITY 5 0 I I 1 l _ o . o cn-> n c. - W T -- 165 -Nominal M B S Measured B I B S 9 6 . amu 9 6 . 3 9 5 2 3 amu 9 7 . •mu 9 7 . 0 3 0 7 4 amu 9 9 . amu 9 9 .00978 amu iee . •mu ' 1 0 0 . 0 1 7 9 5 amu 141 . •mu • 101 . 0 2 5 8 5 amu n o . •mu 1 0 9 . 5 8 9 8 2 amu i n . •mu 1 1 0 . 5 3 6 7 8 amu 1 1 3 . •mu 1 1 3 . 0 2 6 9 7 amu 1 1 4 . •mu 1 1 4 . 0 3 4 1 2 amu 1 1 9 . •mu 1 1 9 . 0 3 7 9 9 amu 1 2 6 . •mu 1 2 6 . 0 3 5 0 9 amu 1 2 7 . •mu 1 2 7 . 0 4 3 0 7 amu 1 3 0 . •mu 1 3 0 . 0 2 9 8 7 amu 131 . •mu 1 3 1 . 0 3 7 7 7 amu 1 3 2 . •mu 1 3 2 . 0 4 7 1 4 amu 1 4 2 . •mu 1 4 2 . 0 3 1 5 7 amu 1 4 3 . •mu 1 4 3 . 0 3 8 5 1 amu 1 4 4 . •mu 1 4 4 . 0 4 6 1 3 amu 1 4 5 . •mu 1 4 5 . 0 5 3 8 6 amu 1 4 9 . amu 1 4 9 . 0 4 9 0 0 amu 1 5 9 . amu 1 5 9 . 0 3 3 8 7 amu 1 6 0 . amu 1 6 0 . 0 4 1 4 5 amu 161 . amu 1 6 0 . 9 4 2 1 5 amu 161 . amu 1 6 1 . 0 4 6 8 3 amu 1 6 2 . amu 1 6 2 . 0 5 6 4 4 amu 1 7 3 . amu 1 7 3 . 0 4 9 9 9 amu 1 7 8 . amu 1 7 8 . 0 5 2 6 1 amu 1 7 9 . amu 1 7 6 . 9 2 8 0 6 amu 1 7 9 . amu 1 7 9 . 0 5 9 8 0 amu 1 8 0 . amu 1 8 0 . 0 6 4 5 6 amu 181 . amu 1 8 1 . 0 6 6 0 3 amu 1 8 5 . amu 1 8 5 . 0 5 4 4 4 amu 2 0 9 . amu 2 0 8 . 6 4 8 7 1 amu 221 . •mu 2 2 0 . 6 6 7 3 6 amu 221 . amu 2 2 1 . 0 7 1 3 5 amu 2 2 2 . •mu 2 2 2 . 0 7 7 1 6 amu 2 2 3 . amu 2 2 3 . 0 6 1 7 8 amu 2 3 4 . •mu 2 3 4 . 0 6 0 4 5 amu 251 . •mu 2 5 1 . 0 6 3 3 1 amu 2 5 2 . •mu 2 5 2 . 0 6 9 8 7 amu 2 6 3 . •mu 2 6 3 . 0 6 3 3 4 amu 2 6 4 . •mu 2 6 4 . 0 9 1 6 3 amu 2 6 5 . •mu 2 6 5 . 0 9 6 1 6 amu 2 8 1 . amu 2 8 1 . 0 9 4 2 4 amu 2 8 2 . •mu 2 6 2 . 0 9 6 4 0 amu 2 9 3 . • M U 2 9 3 . 1 0 1 0 4 amu 3 1 7 . M U 3 1 7 . 0 6 3 7 0 amu 3 1 9 . •mu 3 1 9 . 0 7 5 0 2 amu 3 2 3 . •mu 3 2 3 . 1 2 9 3 6 amu 3 4 0 . •mu 3 4 0 . 1 2 0 0 9 amu 3 4 1 . amu 3 4 1 . 1 2 3 8 8 •mu 3 4 2 . •mu 3 4 2 . 1 2 9 0 1 amu Rel . intensity Abs. intensity 2 . 7 % 1 . 0 6 0 8 6 . 1 X 2 . 3 7 9 2 2 . 1 X 8 0 5 . 1 7 5 8 m 2 . 1 X 6 1 7 . 3 8 2 9 m 1 6 . 9 X 6 . 5 5 6 8 2 . 0 X 7 B 4 . 7 9 0 1 m 2 . 4 X 931 . 2 7 4 5 m 6 . 6 X 3 . 4 3 0 4 2 . 5 X 964 . 2 3 3 5 m 1 1 . 4 X 4 . 4 4 9 2 5 . 5 X 2 . 1 3 2 9 1 . 6 X 6 3 5 . 6 2 0 2 m 1 . 9 X 7 5 7 . 2 0 2 2 m 6 . 9 X 2 . 6 7 4 8 3 . 6 X 1 . 4 1 4 8 1 . 3 X 5 0 4 . 6 3 6 7 m 1 8 . 4 X 7 . 1 4 4 0 1 9 . 1 X 7 . 4 0 8 3 2 . 1 X 6 1 7 . 5 0 4 9 m 4 . 3 X 1 . 6 6 9 4 5 . 8 X 2 . 2 5 8 8 1 . 5 X 5 8 5 . 9 3 7 5 m 3 . 2 X 1 . 2 4 0 0 6 9 . 6 X 2 7 . 0 6 7 4 1 5 . 8 X 6 . 1 3 8 2 1 . 3 X 5 1 9 . 1 6 5 1 m 2.4 X 9 4 9 . 9 5 1 2 m 3 . 4 X 1 . 3 0 7 0 1 0 0 . 0 X 3 8 . 6 7 1 0 6 . 1 X 2 . 3 6 6 0 1 .4 X 5 5 9 . 5 7 0 4 m 1 . 5 X 6 0 0 . 4 6 3 9 m 1 . 2 X 4 8 3 r S 2 0 5 m 2 . 0 X 7 9 1 . 6 2 6 0 m 5 7 . 4 X 2 2 . 3 2 1 9 4 . 2 X 1 . 6 2 7 3 1 . 3 X 5 2 3 . 6 6 1 7 m 4 . 7 X 1 . 6 3 9 7 2 4 . 6 X 9 . 6 3 0 2 6 . 3 X 2 . 4 5 5 7 2 3 . 0 X 6 . 9 4 2 1 15.4 X 5 . 9 8 0 2 1 . 6 X 6 0 4 . 6 5 6 4 m 1 9 . 3 X 7 . 5 0 9 3 2 . 2 X 6 4 5 . 2 1 4 9 m 2 . 3 X 9 0 4 . 0 5 2 6 m 4 . 2 X 1 . 6 1 8 2 2 . 3 X 9 0 7 . 6 3 7 0 m 1.1 X 4 4 0 . 5 5 1 8 m 1 . 9 X 7 3 6 . 8 1 6 5 m 2 0 . 5 X 7 . 9 6 3 4 2 . 8 X 1 . 0 7 9 0 Spec, rv.6 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of maltose. •asoiqoipo jo rxrruwads TO-I-TJ-TPT uoi-3Apisod uopmosai Moq L'M ' ^ S RELATIVE INTENSITY 50 100 _i - 991 -- 1 6 7 -Nominal mass Measured mass R e l . intensity Abs. intensity 97. amu 97.02796 amu 47.3 X 14.9760 99. amu 99.04297 amu 10.3 X 3.2780 103. amu 103.03896 amu 10.7 X 3.3915 109. amu 109.02935 amu 7.9 X 2.5123 115. amu 115 .03939 amu 4.6 X 1.4703 127. amu 127.03985 amu 53.0 X 16.7947 128. amu 128.04235 amu 4.3 X 1.3630 131 . amu 131.03520 amu 3.9 X 1.2507 145. amu 145.04899 amu 100.0 X 31.6772 146. amu 146.05420 amu 8.7 X 2.7607 163. amu 163.06076 amu 56.9 X 18.0388 169. amu 169.05131 amu 5.8 X 1.8502 187. amu 187.06159 amu 8.9 X 2.8054 205. amu 205.06935 amu 3.2 X 1.0127 223. amu 223.09723 amu 8.1 X 2.5504 229. amu 229.07417 amu 8.9 X 2.8126 245. amu 245.06191 amu 5.6 X 1.7745 265. amu 265.09583 amu 4.3 X 1.3635 271 . amu 271 .08924 amu 9.7 X 3.0707 305. amu 305.09277 amu 16.9 X 5.3479 321 . amu 321.05221 amu 3.9 X 1.2480 325. amu 325.12740 amu 15.4 X 4.8666 365. amu 365.11009 amu 50.0 X 15.8491 . 366. amu 366.12288 amu 7.1 X 2.2537 381 . amu 381.07965 amu 12.9 X 4.0824 Spec. IV.8 High resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of cellobiose. •asoiqorpo jo rarujosds joi-Tj-n-pT. uoi-aAUBSau uopmosai A\oq 6AI '^S RELATIVE INTENSITY 50 ICO _L I I I I L I 1 1 1 1 I - 891 -- 169 -l inal mass Measured mmt >s R e l . in tern i i t y A b s . i n t e n s i t y 97. amu 97.03105 amu 4.8 X 2.6567 101. amu 101.02667 amu 12.0 X 6.6138 110. amu 109.59081 amu 2.9 X 1.5747 ' 113. amu 113.02776 amu 5.6 X 3.0701 119. amu 119.03863 amu 5.8 X 3.1899 131 . amu 131.03919 amu 4.3 X 2.3838 143. amu 143.04009 amu 15.0 X 8.2671 144. amu 144.04772 amu 5.7 X 3.1074 161 . amu 161.05019 amu 100.0 X 54.9341 162. amu 162.05629 amu 9.0 X 4.9714 179. amu 179.06297 amu 60.0 X 32.9849 180. amu 180.06919 amu 3.8 X 2.1050 m . amu 221 amu 1 / .V X 245. amu 245.08070 amu 2.8 X 1.5549 263. amu 263.«?9127 amu 36.2 X 19.8672 264. amu 264.09470 amu 5.9 X 3.2192 281 . amu 281.10914 amu 4.3 X 2.3877 341 . amu 341 .14146 amu 31 .8 X 17.4441 342. amu 342.15577 amu 4.2 X 2.2832 353. amu 353.13494 amu 4.8 X 2.6516 377. amu 377.12893 amu 15.5 X 8.4883 379. amu 379.11234 amu 7.5 X 4.1084 387. amu 387.16607 amu 2.7 X 1.4951 Spec, rv.10 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of cellobiose. - 170 -a o u D p u n q o s ^ n j o s q v • • 1 • • • • 1 • • • S t E Z I * • * • • • • * • • • • * • • • • I • •' • * • • • • I •»• • * • • • • I • • • • * • • • • I • * • • * • • • • Oo o V o i i I I 001 T OS o CD - O — O OO (=> " CO N o r ~ o LO o -o CO o •o CNJ i i i (%) souDpunqD S A T ) D j a y Spec, rv.ll Low resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K3 oligosaccharide.. - 171 -Non'in a* ma v.*. 1 73 amu 1B7 aim 205 ami.; 264 amu 293 a:nu 303 amu 'J 17 amu 331 A mi. A 34 7 amu 361 amu 363 amu 365 amu 3Q7 ant. 3B9 amu 39 2 amu 407 amu 421 amu 431 amu 435 amu 437 479 amu 6*9 afT.lt 523 524 amu 525 amu 527 amu 5?f.' amu 529 amu 53V amu 541 amu 543 amu 549 amu •5,-i amu ht>a«ii.irt>0 fii&ss> 173 .04753 aim 1B7.06254 amu 205.073'.-.5 amu 263.58277 amu 293 .(4)901 0 amu 303.07414 amu 317 .08650 amu 331 .11122 tmu 34 7.09U74 amu 361 .079If. «mu 363.09B6<5 amu 365.10627 amu 307.08069 amu 389.10813 amu 391 .13046 amu 407.1J124 amu 421. .09456 amu 431 .07B20 amu 435.10978 amu 437.12962 amu 479.14189 amu 509.14977 amu 523.12701 amu 52".12893 amu 525.1363B amu 527.15790 amu 520.16095 amu 529.17321 amu 539.09447 amu 541.13721 amu 543.14342 amu 549.14199 amu 551 .15*9^ aim. Rel . intensity B.1151 '/ 15.3512 '/ 7 .9363 '/. 6.2022 \ 10.9215 'A 5.3B19 'A 7.124'! •A 5.3B99 "/. 11.4327 •A 12.4472 'A 16.5860 'A 35.3161 'A 9.7790 'A 5.5325 */. 5.132V 'A 5.5625 'A 25.7292 'A 6.7540 5.2822 y. 24 .7677 X 10.7016 v. 16.4707 7. 21.2536 •/. 5.5960 'A 5.6711 X 100.0000 X 21 .2730 X 5.9364 X 5.4972 'A 22.2010 'A 15.4253 X IB.0062 X 9.R1B7 'A Abb. intensity 560.974121 m 1062.1D774 4 m 546.614502 m 426.741455 m 754.974365 m 372.039795 m 492.492676 m 372. 5 8 9 1 J l m 790.313721 m 860.443115 m 1146.606445 m 2441 .314697 m 675.994373 in 3B2.4462B9 m 354.8278131 ;., 3B4.S214B4 m 177a. S94 971 m 466.8BB42B m 365.142022 m 1712.127686 m 739.776611 m 1138.5B0322 m 1469.207764 m 386.84OB20 m 392.028809 m 6912.750244 m 1470.550537 m 410.369B73 m 3B0 .0V4RQ3 n, 1534.69B4B6 m 1066.314697 m 1244.720459 m 678.741455 ir Spec. IV.12 High resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K 3 oligosaccharide. - 172 -amu &53 . 1 7603 amu 13.1037 'A 905.822754 m 557 amu 557.12094 amu 6.6339 'A 45B.587646 m 565 565.12821 amu 5.5585 'A 384 .246826 m 567 amu 567.15411 amu 9.11213 A 629.913330 m 569 569.15959 amu 22.952FJ '/. 1586.669922 m 570 amu 576.16926 amu 5.9541 •/. 411 .590576 m 5 B 3 amu 5C3.13305 amu 6 .0990 V. 421.661377 m 5 9 ' • amu 593.13605 amu 6.8233 V. 471.679688 m a m a 595.15354 *mu 7.3075 'A 510.68115? m amu 597.1613 7 amu 6.1360 •A 424.163810 in f.e9 609.15524 amu 26.7053 'A 1B46.069336 m 610 ar" . 610.16089 amu 6.9770 'A 4fl?. .\'99U;»5 ni 613 613.1573.2 amu 20.959! 'A 1448 .B32539 m 625 aim • 675.13763- amu 7.5169 •A SI9.622803 m 641 a r r . u 641 . lG15iVamu 17.0729 •A 1180.206299 m 657 situ 657.16526 amu 5.1775 'A 357.910156 n> 6B5 amu 6H5.17414 amu 16.739 3 'A 1157.135010 in 686 amu 606.16901 amu 5.1:356 'A 355.010906 m 701 «mu 70J .14579 amu 8.004 7 y. 553.344727 m 703 MTV..'. 7*3 .1UM 4 amu 4 9 .<,855 7. 34 34.63134b m 704 704 . 19624 amu 14 .2749 'A 966 ,765(>09 m Vii 71! .18780 amu 2i" .4082 •A 1410 .766602 m • / » t i cvrnu 712.17067 amu 5.OBI 7 'A 406.585693 m VI 9 719 .15531 amu 16.9359 'A J30b.990479 m 720 7:-0.14 605 amu 5.0107 */. 346 .374512 m 729 uim.i 729 .20410 amu 24.1717 X 167© . 9 2 H 9 ' ' " 5 in CdiU • 73V .20658 amu 7 .4339 '/. 513.8B549t: m 7 3 • 731 . 19078 amu 5.4910 "/. 3 79.577637 ri 7 4-.. 745. i. 1:395 a mi A i:;.034:-.: 'A 90 3 .031494 m 755 755.1QQ2B ©il'.U 7.2599 'A 501 .t'foi572 i> RCii amlt 801 .21470 amu 20.7146 '/. 143 3 .945801 m 802 002 .''.'1342 amu 5.5669 •1 1. 384.826660 in U17 c-t!*:U 817.18921 amu 5.4226 •A 374 .847412 m 87:? iimu 073.23701 amu J? . 1552 'A 563.751221 m B8f: BBE' .35460 t l f l l U 6 .157h •A 425 .6591U0 Hi BttV •arti  i 839.20085 5.5307 -/ Ja 302.324219 ft. R91 an.u 891 .25044 afu 56.7251 '/. 3921 .264648 n, «i<TlU 89?. .26040 amu 21.7869 V. 1506.07299H m 893 amu 093 .25237 amu 7.6873 'A 531 .4025BQ rn 9ti7 amu 907 .20198 amu 26.9908 'A 2004.0SBB3B n> 90B amu 9015.22112 amu 11.2393 'A 776.947021 m 909 amu 909 .18495 amu 6.5320 'A 451.699121 ni 913 amu • 913.19913 amu 10.9109 •A 754.241943 m 917 5./TU.I 917.25661 amu 10.6177 'A 733.97B2 71 (Tl 933 amu 933.25635 amu 5.6278 •A 3B9.0380B6 m 935 aniu 935.26465 amu 5.2945 'A 365.997314 ni Spec. IV.12 High resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K 3 oligosaccharide, con't. - 173 -8 9 t Z I • • • • 1 • • • • I * • • * 1 * • * * I • • * •—i—i—•—•—•—I —•—>— i a . 001 ( % ) s p U D p u n q D &-AnD I 3 ci Spec, rv.13 Low resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K3 oligosaccharide. - 174 -Nominal mass Measured mass R e l . intensity Abs. intensity 157 amu 157.01535 amu 2.4756 X 202.099609 m 159 amu 159.03072 amu 2.4122 X 196.923828 m 161 amu 161 .04631 amu 4.9613 X 405.029297 m 175 amu 175.02564 amu 17.5113 X 1429.589844 m 179 amu 179.05682 amu 5.8650 X 478.808594 m 193 amu 193.03643 amu 7.7760 X 634.814453 m 221 amu 221.06751 amu 4.7830 X 390.478516 m 111 amu 226.73973 amu 2.7142 X 221 .582031 m 249 amu 249.06278 amu 2.7022 X 220.605469 m 263 amu 263.07834 amu 2.4965 X 203.808594 m 275 amu 275.07775 amu 2.3834 X 194.580078 m 303 amu 303.07325 amu 5.2496 X 428.564453 m 319 amu 319.06651 amu 38.6788 X 3157.666016 m 320 amu 320.06976 amu 4.7149 X 384.912109 m 321 amu 321 .08369 amu 6.9859 X 570.312500 m 323 amu 323.10024 amu 5.9362 X 484.619141 m 335 amu 335.09930 amu 7.8782 X 643.164062 m 337 amu 337 .07685 amu 28.5457 X 2330.419922 m 338 amu 338.08278 amu 4.7974 X 391.650391 m 341 amu 341 .11144 amu 2.8518 X 232.812500 m 355 amu 355.08791 amu 6.5917 X 538.134766 m 361 amu 361 .07876 amu 3.3536 X 273.779297 m 363 amu 363.09293 amu 3.2776 X 267.578125 m 379 amu 378.65965 amu 2.7794 X 226.904297 m 379 amu 379.08685 amu 54.5399 X 4452.539063 m 380 amu 380.09387 amu 10.3998 X 849.023437 m 381 amu 381.10510 amu 48.5158 X 3960.742187 m 382 amu 382.10787 amu 7.8232 X 638.671875 m 383 amu 383.11490 amu 2.5964 X 211.962891 m 395 amu 395.12164 amu 2.2273 X 181.835938 m 397 amu 397.09773 amu 31.3909 X 2562.695312 m 398 amu 398.10344 amu 4.8590 X 396.679688 m 425 amu 425.08926 amu 3.7992 X 310.156250 m 48 i amu 4"BT. 12201 amu 56.9564 X 4155.908203 m 482 amu 482.12587 amu 10.2204 X 834.375000 m 483 amu 483.13049 amu 3.9499 X 322.460938 m 497 amu 497.10992 amu 9.8813 X 806.689453 m 498 amu 498.12282 amu 2.2309 X 182.128906 m 498 amu 498.38577 amu 3.7854 X 309.033203 m 499 amu 498.97976 amu 2.0599 X 168.164063 m 499 amu 499.12956 amu 100.0000 X 8163.818359 m 500 amu 500.13698 amu 20.4127 X 1666.455078 m 501 amu 501.14006 amu 5.4918 X 448.339844 m 507 amu 507.13586 amu 2.6263 X 214.404297 m 509 amu 509.15291 amu 4.0922 X 334.082031 m 515 amu 515.12249 amu 5.1485 X 420.312500 m 516 amu 516.36078 amu 4.1622 X 339.794922 01 517 amu 517.13834 amu 61 .4151 X 5013.818359 m 518 amu 518.14706 amu 12.8730 X 1050.927734 m 519 amu 519.14561 amu 4.1544 X 339.160156 m 525 amu 525.14819 amu 13.1661 X 1074.853516 m 526 amu 526.15408 amu 3.0031 X 245.166016 m 541 amu 541 .14005 amu 3.8105 X 311.083984 m 543 amu 543.15747 amu 6.5863 X 537.695313 m 545 amu 545.13705 amu 4.2459 X 346.630859 m 553 amu 553.14154 amu 38.6232 X 3153.125000 m Spec. IV.14 H i g h resolution negative-ion l .d . i . -F . t . - i . c .r . spectrum of Klebsiella K3 oligosaccharide. - 175 -554 amu 554.14677 amu 8.9596 X 731.445312 m 555 amu 555.15359 amu 2.8314 X 231.152344 m 559 amu 559.15472 amu 5.2705 X 430.273438 m 567 amu 567.15365 amu 2.1053 X 171.875000 m 571 amu 571.14206 amu 15.9975 X 1306.005859 m 572 amu 572.14603 amu 3.3661 X 274.804688 m 573 amu 573.15420 amu 2.0120 X 164.257813 m 585 amu 585.16400 amu 10.7234 X 875.439453 m 586 amu 586.17451 amu 2.7698 X 226.123047 m 589 amu 589.16560 amu 6.3106 X 515.185547 m 643 amu 643.17917 amu 5.9380 X 484.765625 m 645 amu 645.19567 amu 2.1580 X 176.171875 m 660 amu 659.90122 amu 3.3667 X 274.853516 m 661 amu 661 .18173 amu 38.9043 X 3176.074219 m 662 amu 662.18508 amu 10.8622 X 886.767578 m 663 amu 663.18934 amu 4.7968 X 391.601563 m 678 amu 677.84485 amu 2.4648 X 201.220703 m 679 amu 678.90380 amu 2.1861 X 178.466797 m 679 amu 679.19605 amu 36.1757 X 2953.320312 m 680 amu 680.20202 amu 10.5960 X 865.039063 m 681 amu 681.19835 amu 2.9995 X 244.873047 m 687 amu 687.20232 amu 9.7766 X 798.144531 m 688 amu 688.20417 amu 3.1484 X 257.031250 m 703 amu 703.19243 amu 13.3712 X 1091.601563 m 704 amu 703.75996 amu 5.7687 X 470.947266 m 704 amu 704.20106 amu 4.3034 X 351.318359 m 705 amu 705.20224 amu 65.1036 X 5314.941406 m 706 amu 706.21032 amu 20.1250 X 1642.968750 m 707 amu 707.20407 amu 10.5165 X 858.544922 m 70S amu 708.20853 amu 2.2435 X 183.154297 m 721 amu 721.19356 amu 2.2118 X 180.566406 m 731 amu 731.22004 amu 4.4194 X 360.791016 m 733 amu 733.18738 amu 2.1693 X 177.099609 m 747 amu 747.22866 amu 2.6957 X 220.068359 m 749 amu 749.20120 amu 4.2992 X 350.976563 m 759 amu 759.23024 amu 15.3234 X 1250.976563 m 760 amu 760.23156 amu 5.2484 X 428.466797 m 777 amu 777.21992 amu 4.0910 X 333.984375 m 803 amu 803.20610 amu 2.1215 X 173.193359 m 841 amu 841 .25590 amu 2.9624 X 241.845703 m 849 amu 849.24959 amu 2.2112 X 180.517578 m oo i amu amu 34.1362 X 2786.816406 m 868 amu 868.28306 amu 11.8167 X 964.697266 m 869 amu 869.28911 amu 4.4481 X 363.134766 m 893 amu 893.26589 amu 3.1777 X 259.423828 m 911 amu 911.22090 amu 2.5653 X 209.423828 m Spec, rv.14 H i g h resolution negative-ion l .d . i . -F . t . - i . c .r . spectrum of Klebsiella K3 oligosaccharide, con't. - 176 -SI d O u o p u n q D s}n|Osqv i i i p • • • • 1 •—i—i—i—l—i i i i • • i i 001 s in 5 I i i i i j i i i i I i i i i (%) d O U O p u n l ^ D 8 A T } D f 9 f c j O .o o -o CD O •o GO O o CO N E • o -o to o . o —o © r -O Spec. IV.15 Low resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of E. coli K9 oligosaccharide.. - 177 -Nominal mass M e a s u r e d mats R e l . i n t e n s i t y Abs. i n t e n s i t y 158 amu 15B.30B13 amu 7.5522 X 1219.921375 m 166 amu 166.26356 amu 5.B383 X 943.066406 m 160 amu 160.06632 amu 5.670B X 916.015625 m 166 amu 1B6.07621 amu 15.3009 X 2471.582031 m 10B amu 187.79260 amu 6.3920 X 1032.519531 m 204 amu 204 .08645 amu 29.0625 X 4694.531250 m 208 amu 20B.05762 amu 5.1073 X 825.000000 nt 226 amu 226 .06826 amu 8.7559 X 1414.355469 •Tl 274 amu 274.09083 amu 9.29^9 X 1500.781250 m 366 amu 366.13965 amu 24 .3106 X 3926.953125 m 367 amu 367.14015 amu 5.2313 X 845.019531 tn 366 amu 387.68734 amu 7.1737 X 1156 .789063 m amu 388.11952 amu 100.0000 X 16153.222687 m 387 amu 389.12455 amu 17.3908 X 2809.17968 7 m 406 amu 406.13025 amu 7.B055 X 1260.839844 m ".708 amu 50D.16437 amu 6.5854 X 13B6.816406 m 550 amu 550.17634 amu 11.1014 •/. 1B06.152344 m 56B amu 56B.10C14 amu 34 .0973 X 5507.812500 m 569 amu 569.19233 amu 8.092 7 X 1307.226563 m 584 amu 584.17066 amu 11 .6776 X 1918.652344 m Spec, rv.16 High resolution positive-ion l.d.i.-F.t.-i.c.r. oligosaccharide. spectrum of E. coli K9 - 178 -9 s o u D p u n q D s i n f o s q v 9 t } I . . . . i . . . • i i i -I I I I I L. o o "O CD o - o CO o l - o CM 0 0 1 1 ( 1 1 T 1 1 1 1 " "I " 1 1 1 ' 1 1 I OS o (%) a D U D p u n q D a A T ^ D j a y Spec. IV.17 L o w resolution negative-ion l .d . i . -F . t . - i . c . r . spectrum of E. coli K9 oligosaccharide. - 179 -Nominal mass Measured mass Re^. intensity Abs, intensity 159 amu 159.49366 amu 6.2559 X 530.371094 m 161 amu 160.73590 amu 11.7061 X 992.431641 m 170 amu 170.04480 amu 3B.3633 X 3252.392578 m 202 amu 202.07084 amu 14.6694 X 1243.652344 m 212 amu 212.0786© amu 5.5112 X 467.236328 m 262 amu 262.08791 amu 17.6395 % 1495.458984 m 264 amu 264.09414 amu 10.4223 X 883.593750 m 272 amu 272.07474 amu 15.2926 X 1296.484375 m 290 amu 290.08815 amu 33.0496 X 2869.726562 m 291 amu 291.08508 amu 5.8113 X 492.675781 m 308 amu 3O8.09403 amu 8.5856 X 727.880859 m 332 amu 332.08634 amu 8.6265 X 731.347656 m 333 amu 353 .13939 amu 6.1580 X 522.070313 m 355 amu 355.15804 amu 9.8810 X 837.695313 m 364 amu 364.11388 amu 53.1939 4509.716797 m 365 amu 365.11853 amu 12.5614 X 1064.941406 m 369 amu 369.14954 amu 7.9112 X 670.703125 m 370 amu 370.14022 amu 6.8325 X 579.248047 m 372 amu 372.09508 amu 6.5307 X 553 .662109 m 3G1 amu 381.17351 amu 7.0847 X 600.634766 m 382 amu 382.12471 amu 7.354B X 623.535156 m 390 amu 390.10935 amu B.5033 X 720.898437 m 394 amu 394.12061 amu 5.3926 X 457.177734 m 395 amu 395.15934 amu 12.6570 X 1073.046875 m 396 amu 396.16276 amu 6.3245 X 536.181641 m 397 amu 397.17559 amu 5.9363 X 503.271484 m 398 amu 398.16979 amu 5.6564 X 479.541016 m 400 amu 400.08613 amu 9.5187 X 806.982422 m 406 amu 406.16413 amu 6.7950 X 576.074219 m 407 amu 407.15304 amu 10.6366 X 901.757812 m 409 amu 409.16689 amu 8.6288 X 731.542969 m 7--88 .lSi.3,4 ID: .r#Sw56 m 411 amu 411.17614 amu 9.5561 X 810.156250 m 412 amu 412.17435 amu 9.1409 X 774.951172 m 413 amu 413.16830 amu 5.3724 X 455.468750 tn 418 amu 41B. 11402 amu 5.1392 X 435.693359 m 424 amu 424.13016 amu 100.0000 X 8477.880859 m 425 amu 425.14126 amu 17.2859 X 1465.478516 m 426 amu 426.13883 amu 5 .5579 X 471.191406 m 428 amu 428.17097 amu 5 .6374 X 477.929688 m 432 amu 432.12512 amu 7.1884 X 609.423828 m 436 amu 436.15662 amu 7.0387 X 596.728516 m 437 amu 437.19004 amu 5.0591 X 428.906250 m 43Q amu 438.17940 amu 7.6762 X 650.781250 m 440 amu 440.1801B amu 7.1158 X 603.271484 m 450 amu 450.14200 amu 7.3848 X 626.074219 m 452 amu 452.13349 amu 16.0655 X 1362.011719 m 453 amu 453.12806 amu 5.0557 X 428.613281 m 454 amu 454.16195 amu 9.8556 X 835.546875 m 455 amu 455.17506 amu 6.9822 X 591.943359 m 456 amu 456.17697 amu 6.4662 X 548.193359 m 460 amu 460.12225 amu 29.9412 X 2538.378906 m 461 amu 461.10968 amu 5.5723 X 472.412109 m 462 amu 462.11058 amu 12.2959 X 1042.431641 m 466 amu 466.15643 amu 27.2561 X 2310.742187 m 467 amu 467.17213 amu 32.0924 X 2720.751953 m Spec. IV.18 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of E. coli K9 oligosaccharide. - 180 -4 6 3 amu 4 6 8 . 1 8 1 8 3 amu 1 1 . 2 8 4 5 X 9 5 6 . 6 8 9 4 5 3 m 4 6 9 amu 4 6 9 . 1 6 8 6 0 amu 6 . 3 8 3 5 X 5 4 1 . 3 5 7 4 2 2 m 4 7 0 amu 4 7 0 . 1 4 4 0 3 amu 7 . 7 4 4 2 X 6 5 6 . 5 4 2 9 6 9 m 4 7 7 amu 4 7 7 . 1 7 3 5 2 amu 5 . 9 2 3 6 X 5 0 2 . 1 9 7 2 6 6 m 4 7 0 amu 4 7 8 . 2 0 7 7 7 amu 5 . 3 1 8 3 X 4 5 0 . 8 7 8 9 0 6 m 4 7 9 amu 4 7 9 . 1 9 3 2 6 amu 5 . 0 4 6 5 X 4 2 7 . 8 3 2 0 3 1 m 481 amu 4 8 1 . 1 8 5 3 9 amu 6.2 7 9 0 X 5 3 2 . 3 2 4 2 1 9 m 4 8 2 amu 4 8 2 . 1 9 3 2 0 amu 9.4461 X 8 0 0 . 8 3 0 0 7 8 m 4 8 3 amu 4 8 3 . 2 0 3 4 2 amu 9 . 0 9 4 2 X 7 7 0 . 9 9 6 0 9 4 m 4 8 4 amu 4 8 4 . 1 6 6 1 4 amu 1 3 . 9 5 4 1 X 1 1 8 3 . 0 0 7 8 1 2 m 4 8 5 amu 4 3 5 . 1 9 5 0 1 amu 7.69 8 1 X 6 5 2 . 6 3 6 7 1 9 m 4 9 1 amu 4 9 1 . 1 6 1 6 5 amu 5 . 2 4 1 7 X 4 4 4 . 3 8 4 7 6 6 m 4 9 2 amu 4 9 2 . 1 5 0 0 0 amu 9 . 3 0 6 2 X 7 8 8 . 9 6 4 8 4 4 m 4 9 3 amu 4 9 3 . 1 8 6 9 2 amu 5 . 6 7 4 8 X 4 8 1 . 1 0 3 5 1 6 m 4 9 4 amu 4 9 4 . 2 1 8 7 5 amu 7 . 8 9 9 7 X 6 6 9 . 7 2 6 5 6 2 m 4 9 5 amu 4 9 3 . 2 1 2 1 0 amu 7 . 9 5 7 9 X 6 7 4 . 6 5 8 2 0 3 m 4 9 6 amu 4 9 6 . 1 9 6 4 5 amu 1 0 . 2 0 8 7 X 8 6 5 . 4 7 8 5 1 6 tn 4 9 7 amu 4 9 7 . 1 9 2 0 3 amu 2 1 . 2 4 6 1 X 1 8 0 1 . 2 2 0 7 0 3 m 4 9 8 amu 4 9 8 . 1 9 9 5 6 amu 9 . 4 1 8 5 X 7 9 8 . 4 8 6 3 2 8 m 4 9 9 amu 4 9 9 . 2 0 6 2 4 amu 7 . 7 4 1 3 X 6 5 6 . 2 9 8 8 2 8 m 5 0 2 amu 5 0 2 . 1 2 1 1 7 amu 6.7017 X 5 6 8 . 1 6 4 0 6 3 m S 0 8 amu 5 0 8 . 1 9 4 5 0 amu 6.7011 X 5 6 8 . 1 1 5 2 3 4 m 5 0 9 amu 5 0 9 . 2 0 6 3 0 amu 1 2 . 7 3 5 3 X 1 0 7 9 . 6 8 7 5 0 0 tn 5 1 0 amu 5 1 0 . 2 1 3 7 1 amu 9 . 3 0 7 9 X 7 8 9 . 1 1 1 3 2 8 m 511 amu 5 1 1 . 2 2 7 2 2 amu 1 0 . 3 1 5 8 X 8 7 4 . 5 6 0 5 4 7 m 5 1 2 amu 5 1 2 . 2 0 9 6 2 amu 8 . 9 2 3 2 X 7 5 6 . 4 9 4 1 4 1 m 5 1 3 amu 5 1 3 . 2 2 6 7 7 amu 6 . 6 4 3 6 X 5 6 3 . 2 3 2 4 2 2 m 5 1 4 amu 5 1 4 . 2 1 2 3 2 amu 6 . 6 8 5 0 X 5 6 6 . 7 4 8 0 4 7 m 5 1 5 amu 5 1 5 . 2 0 2 9 8 amu 5 . 6 4 7 7 X 4 7 8 . 8 0 8 5 9 4 m 5 2 0 amu 5 2 0 . 1 2 5 5 1 amu 6 . 9 3 3 8 X 5 8 7 . 8 4 1 7 9 7 m 5 2 3 amu 5 2 3 . 2 2 1 7 0 amu 7. 8 8 8 8 X 6 6 8 . 7 9 8 8 2 8 m 5 2 4 amu 5 2 4 . 2 1 3 2 0 amu 6 . 5 3 9 9 X 5 5 4 . 4 4 3 3 5 9 m 5 2 5 amu 5 2 5 . 2 0 9 5 8 amu 8 . 9 7 1 5 X 7 6 0 . 5 9 5 7 0 3 m 5 2 6 amu 5 2 6 . 1 7 7 5 7 amu 1 9 . 5 0 4 5 X 1 6 5 3 . 5 6 4 4 5 3 m 5 2 7 amu 5 2 7 . 1 9 4 5 4 amu 1 0 . 8 6 7 0 X 9 2 1 . 2 8 9 0 6 2 m 5 2 8 amu 5 2 8 . 2 0 0 7 3 amu 6 . 3 8 1 5 X 5 4 1 . 0 1 5 6 2 5 m 5 2 9 amu 5 2 9 . 2 1 3 9 0 amu 5. 1 8 8 1 X 4 3 9 . 0 4 3 7 5 0 m 534 amu 5 3 4 . 1 9 3 7 0 amu 5 . 9 9 7 3 X 5 0 8 . 4 4 7 2 6 6 m 5 3 5 amu 5 3 5 . 1 8 4 0 B amu 5 . 2 7 6 3 X 4 4 7 . 3 1 4 4 5 3 m 5 3 8 •mu 5 3 8 . 2 2 0 6 4 amu 6 . 1 0 1 6 X 5 1 7 . 2 8 5 1 5 6 m 5 3 9 amu 5 3 9 . 2 4 7 6 0 amu 7 . 7 7 0 1 X 6 5 8 . 7 4 0 2 3 4 m 5 4 0 amu 5 4 0 . 2 3 8 9 5 amu 7 . 2 6 2 1 X 6 1 5 . 6 7 3 8 2 8 m 541 amu 54 1 . 2 1 1 2 5 amu 8 . 4 3 1 9 X 7 1 4 . 8 4 3 7 5 0 m 5 4 2 amu 5 4 2 . 1 9 9 0 5 amu 7 . 1 6 3 1 X 6 0 7 . 2 7 5 3 9 1 in 5 4 3 amu 5 4 3 . 1 7 6 6 4 amu 5 . 5 1 6 4 X 4 6 7 . 6 7 5 7 8 1 m 5 4 4 amu 54 4 .19368 amu 5 . 2 0 4 9 X 4 4 8 . 0 4 6 8 7 5 m ' 5 4 5 amu 5 4 5 .18600 amu 5 . 8 3 2 0 X 4 9 4 . 4 3 3 5 9 4 m 5 4 9 amu 5 4 9 . 2 1 6 2 6 amu 6.5 0 2 4 X 5 5 1 . 2 6 9 5 3 1 m 5 5 0 amu 5 5 0 . 2 5 0 2 7 amu 5 . 1 2 0 7 X 4 3 4 . 1 3 0 8 5 9 m 5 5 1 amu 5 5 1 . 2 4 1 5 8 amu 8 . 5 9 9 5 X 7 2 9 . 0 5 2 7 3 4 m 5 5 2 amu 5 5 2 . 1 8 3 4 7 amu 9 . 6 8 6 9 X 8 2 1 .240234 m 5 5 3 amu 5 5 3 .2.3551 amu 9.5924 X 8 1 3 . 2 3 2 4 2 2 m '354 amu 5 5 4 . 2 3 8 4 5 amu 6.2277 X 5 2 7 . 9 7 8 5 1 6 m 5 5 5 amu 5 5 5 . 1 9 6 6 6 amu 7 . 3 6 5 8 X 6 2 4 , 4 6 2 0 9 1 m 5 5 6 amu 5 5 6 . 2 0 7 5 0 amu 5 . 3 9 3 6 X 4 7 4 . 2 1 8 7 5 0 m 5 6 2 amu 5 6 2 . 1 5 2 7 7 amu 7 . 7 3 5 5 X 6 5 5 . 8 1 0 5 4 7 m 5 6 3 amu 5 6 3 . 2 3 9 6 8 amu 5 . 8 5 4 5 X 4 9 6 . 3 3 7 8 9 1 m 5 6 5 amu 5 6 5 . 2 7 3 6 4 amu 7 . 1 5 5 6 X 6 0 6 . 6 4 0 6 2 5 m 5 6 6 amu 5 6 6 . 2 8 5 4 9 amu 7 . 3 8 4 8 X 6 2 6 .074219 m .18 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of E. Cc?// oligosaccharide, con't. 181 -567 amu 567.2S8B6 amu 15.9416 X 1351.5136 72 m amu 56B.24310 amu 11.0012 X 932.666016 m 56V amu 567.24199 amu 8.065© X 751.5625W0 HI 570 amu 570.22537 amu 5.54 IB X 469.B24219 m 572 amu 5 72.22374 amu 6.3141 '/. 535.302734 ni 579 579.26379 amu 7.6636 X 649.707031 m amu 530.16251 amu 11.1 69'? •A 946.9 726^6 581 amu 581.21563 &mu 7 , 47«>0 X 633 .3«i«'>7Bl (tl amu '-Hi . 1 7991 ami i 7.9654 X 675.292969 m r:.83 amu 583 .22092 amu V.. 9829 X 507.226562 m a.nu 584 .22680 amu B.7400 7. 740.966777 in r.86 amu 5B6.1751B amu 5.9426 X 503 .808594 m 592 amu 392.25132 a.nu 5 .6259 't im 476.933125 m •f.93 amu 593.24039 amu 6.5B25 V. 558.056641 m 594 amu 594 .27166 amu 6.0423 X 512.255859 ri. S96 amu 596.24197 amu 6 .8762 V. 582.958984 m 597 amu 597.26007 amu 6.0065 X 577.050781 rn 59B amu 598.24926 amu 7.0974 /• 601.708984 m 610 amu 610.25410 amu 10. /33V X 910.009766 m 61 1 amu 611 .26193 amu 9.2872 X 787.353516 m 612 amu 612.22240 amu 9.9501 'A 844 .238201 ni 613 amu 613.22044 amu 6.4944 X 550.5S5938 m 617 A . T . U 619 .22210 amu 6.06B2 X 514.453125 M 620 amu 620.27371 amu 6.7484 X 572.119141 m 621 amu <>:i .24261 amu 5.2123 X 441.894531 :I\ 623 amu 623.23474 amu 6.4408 X 546.044922 .Tl 624 amu 624.0937 7 amu 9.1662 X 777.099609 rr; 625 amu 625.24137 amu 5.90B6 X 500.927735 m 626 amu 626.10101 amu 12 .4911 X 1&58.984375 m 627 amu 627.21042 amu 6.2162 X 527.001953 (Ti 636 amu 636.31231 amu 17.4702 X 1481.103516 m 637 amu 637.29440 amu 10.5001 X B90.1B5547 m 630 amu 630.26519 airu.i 7 .' 1 786 X 600.593750 m 639 amu 639.30133 Amu 5.2659 •A 446 .435547 m 632 amu 652 .32108 «mu 8.1704 X 692.675701 m 653 amu 653.28205 amu 5.7295 X 485.742188 m 65'.:. amu 65'... 23636 amu 9.6886 X 821.386719 m 664 amu 664.34 IBB amu 7.B755 X 667.675781 m 667 amu 667.2 7066 amu 6.6004 X 559.570313 m 66b amu 668.25B21 amu 5.0856 X 431.152344 m 677 amu 677.27166 amu 5.6195 X 476.416016 m 67V amu 679.25608 amu 5.4900 X 466.113281 m 680 amu 680.27322 amu 6.4034 X 542.B71094 m 721 amu 721.39801 amu 5.1599 X 437.451172 m 724 amu 724.29827 amu 5.4790 X 464.501953 m 817 amu 817.27382 amu 24.24B0 X 2055.712B91 m 818 amu , BIB. 28671 amu 10.2271 X 867.041016 m 835 amu 835.28946 amu 8.0909 X 685.937500 m Spec. IV.18 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of E. coli K9 oligosaccharide, con't. - 182 -s o u D p u n q D a j n f o s q v I i i i i i i i—i—• I ' • • • i • i • . I o 2 < O CM CM CM CM CD < 00L (%) S O U D p U n q D 3 A n D l 8 e l Spec. IV.19 Low resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K 2 2 oligosaccharide.. - 183 -Nominal m * S B Measured m a s s 94 •nu 93.53036 •mu 97 •mu 97.02904 •mu 99 amu 99.04397 •mu 102 •mu 102.21946 •mu 103 •mu 102.53582 •mu 103 •mu 103.03929 •mu 110 •mu 110.48207 •mu 115 •mu 115.03864 •mu 127 •mu 127.03945 •mu 129 •mu 129.05383 •mu 133 •mu 132.90525 •mu 133 •mu 133.04978 •mu 145 •mu 145.04929 •mu 150 •mu 149.71473 •mu 153 •mu 153.05544 •mu 155 •mu 155.03154 •mu 157 •mu 157.05030 •mu 159 •mu 159.02939 •mu 169 •mu 169.04912 amu 173 •mu 173.04386 •mu 1B5 •mu 185.04306 amu 187 •mu 187.06016 amu 186 •mu 168.06247 amu 203 •mu 203.05276 amu 205 •mu 204.92232 amu 205 •mu 205.07030 amu 206 •mu 206.07386 amu 209 •mu 209.03993 amu 215 •mu 215.05613 amu 227 •mu 227.05096 amu 229 •mu 229.07024 amu 231 •mu 231.05149 amu 233 •mu '233.06574 amu 245 •mu 245.06185 •mu 253 •mu 253.03845 •mu 257 •mu 257.06274 •mu 264 •mu 263.57396 •mu 271 •mu 271.04263 •mu 275 •mu 275.07516 •mu 285 •mu 284.58033 •mu 265 •mu 265.08217 •mu 293 •mu 292.57330 •mu 329 •mu 329.09107 •mu 347 •mu 347.09416 •mu 349 •mu 349.11490 •mu 365 •mu 364.62644 •mu 365 •mu 365.10536 •mu 366 •mu 366.10966 •mu 375 •mu 375.02169 •mu 369 •mu 369.10226 •mu 390 •mu 390.10495 •mu 407 •mu 406.52575 •mu 407 •mu 407.11523 •mu 406 •mu 408.11672 •mu 409 •mu 409.11C58 •mu 435 •mu 435.10266 •mu Rel . Intensity Ab*. Intensity 1.4387 X 253.784180 ID 2.5279 X 445.922852 m 1.6366 X 323.974609 m 1.4242 X 251.220703 m 1.6601 X 326.125000 m 3.6026 X 635 .496047 m 2.1522 X 379.636672 m 2.0609 X 367.065430 m 6.0800 X 1425.292969 m 1.4685 X 262.573242 m 3.7161 X 655.517578 m 1.6276 X 287.109375 m 4.0227 X 709.594727 m 3.6788 X 664.204102 m 1.5390 X 271.484375 m 1.4366 X 253.417969 m 3.4456 X 607.788086 m 2.3729 X 418.579102 m 3.7826 X 667.236328 m 3.3404 X 589.233398 m 3.3051 X 583.007613 m 23.4116 X 4129.760742 m 1.6311 X 322.996047 m 9.4191 X 1661.499023 m 2,2705 X 400.512695 m 26.3306 X 4644.653320 m 1.9668 X 347.290039 m 1.7591 X 310.302734 m 2.9556 X 521.362305 m 2.0336 X 358.764646 m 2.1252 X 374.677930 m 1.6442 X 290.039063 m 2.2968 X 405.151367 m 3.1736 X 559.814453 m 1.6034 X 316.115234 m 1.4214 X 250.732422 m 3.3680 X 594.116211 m 12.7061 X 2241.333008 m 2.6248 X 498.291016 m 11.2833 X 1990.356445 m 3.1148 X 549.436477 m 1.7051 X 300.781250 n 1.5785 X 278.442363 m 9.0371 X 1594.116211 ID 1.9051 X 336.059570 m 2.6317 X 499.511719 m 23.8664 X 4213.500977 m 3.0476 X 537.597656 m 1.5550 X 274.291992 HI 8.1457 X 1436.869648 m 1.4000 X 246.948242 m 2.6539 X 503.417969 m 41.9556 X 7400.878906 <n 6.5942 X 1163.208006 m 1.6761 X 296.020508 m 1.4442 X 254.760742 m Spec. IV.20 High resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K22 oligosaccharide. - 184 -509 amu 509.14772 amu 3.4961 X 616.699219 m 526 amu 526.17843 amu 2.6110 X 460.571289 m 527 amu 526.64842 amu 1.7446 X 307.739258 m 527 amu 527.15477 amu 50.5720 X 8920.776367 m 528 amu 528.15965 amu 11.2314 X 1981.201172 m 529 amu 529.14901 amu 3.0553 X 538.940430 m 543 amu 543.12971 amu 15.0977 X 2663.208008 m 544 amu 544.127BB amu 2.7750 X 489.501953 <n 545 amu 545.13283 amu 1.7231 X 303.955078 m 551 amu 551.14761 amu 1.6228 X 321.533203 tn 568 amu 567.9700B amu 1.9356 X 341.430664 m 569 amu 568.606B5 amu • 3.2386 X 571.289063 m 569 amu 569.16087 amu 100.0000 X 17639.770508 m 570 amu 570.16937 amu 21 .4602 X 3785.522461 tn 571 amu 571.17809 amu 5.6088 X 9B9.379BB3 m 584 amu 583.97258 amu 2.2110 X 390.014646 m 585 amu 585.14293 amu 29.1935 X 5149.658203 m SB6 amu 586.14278 amu 6.9742 X 1230.224609 m 587 amu 587.15141 amu 4.2504 X 749.755859 tn 611 amu 611.17594 amu 3.8933 X 666.767578 m 641 amu 641.17731 amu 3.60B2 X 636.474609 ai 683 amu 683.23459 amu 2.4470 X 431.640625 m 685 amu 685.18407 amu 1 .7411 X 307.128906 m 727 amu 727.20140 amu 1.5605 X 275.268555 m 731 amu 731.24246 amu 1.5245 X 268.920898 m 757 amu 757.18150 amu 1.7563 X 309.614453 m 799 amu 799.25831 amu 3.4324 X 605.466750 m 815 amu 815.23083 amu 1.9729 X 348.022461 m 817 amu 817.22371 amu 4.4504 X 785.034160 tn 833 amu B33.16937 amu 2.0186 X 356.079102 m Spec. IV20 High resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K22 oligosaccharide, con't. •3puBqooBSo3r[o ZZyL VU3*SQ9M J ° nmijoads \ro-i-TH-i'P'I UOI-3AUB33U uoiiniosaj Moq IZAI M&S R e l a t i v e a b u n d a n c e (%) 50 1 • 1 ' 1 I L 1 1 I Q ~ ~ 1 1 1 1 I 1 1 1 ' | 1 1 1 1 I 1 1 1 1 j 1 ' 1 1 I 1 1 ' 1 | 1 1 1 1 I 1 1 ' ' I I I I ' | I I I I j ' 0 1 2 3 A b s o l u t e a b u n d a n c e - £81 -- 186 -Nominal mass Measured mats 113 amu 113.02327 •mu 143 amu 143.03333 •mu 150 amu 149.72364 •mu 157 amu 157.01422 •mu 139 amu 159.03146 •mu 161 amu 161.04603 amu 175 amu 175.02329 •mu 17B amu 178.04629 amu 179 •mu 179.05487 amu 165 amu 185.04541 amu 187 •mu 167.06000 •mu 203 amu 203.05633 •mu 215 amu 215.03153 •mu 217 amu 217.03336 •mu 235 •mu 235.04647 •mu 247 •mu 247.04399 •mu 24B •mu 246.03043 •mu 263 •mu 263.07355 •mu 265 amu 265.03646 •mu 269 •mu 269.08194 •mu 275 amu 275.07676 •mu 276 •mu 276.06304 •mu 277 amu 277.05552 •mu - 287 •mu 267.03629 •mu 266 amu 288.06268 •mu 289 •mu 269.03273 •mu 293 amu 293.08495 •mu 295 •mu 295.06693 •mu 299 •mu 299.04692 •mu 305 •mu 305.09000 amu 306 •mu 306.08941 •mu 307 •mu 307.05462 amu 319 •mu 319.06566 •mu 320 •mu 320.074B5 •mu 323 •mu 323.09707 •mu 329 •mu 329.04186 •mu 335 •mu 335.03663 •mu 337 •mu 336.70920 •mu 337 •mu 337.07562 •mu 338 amu 336.07679 •mu 349 •mu 349.07469 •mu 355 •mu 355.06649 •mu 356 •mu 356.0B625 •mu 359 •mu 359.05648 •mu 359 •mu 359.12067 amu 365 •mu 365.10350 •mu 366 •mu 366.11443 •mu 367 •mu 367.08368 •mu 379 •mu 379.08467 •mu 391 •mu 391.06665 •mu 393 •mu 393.09877 •mu 401 •mu 401.07614 •mu 405 •mu 405.10474 •mu 406 •mu 406.10491 •mu 407 •mu 407.07869 •mu Rel . intensity Abs. intensity 7.6821 X 367.939433 m 5.6140 X 293.599446 m 6.7877 X 342.773438 m 3.0579 X 154.418943 m 4.2729 X 213.779623 m 4.4643 X 225.443522 m 12.1641 X 614.276156 m 4.9449 X 249.715169 m 10.1570 X 512.919107 m 3.2659 X 165.934245 m 7.2265 X 364.929199 m 5.0465 X 254.943848 m 11 .5445 X 582.967467 m 7.7409 X 390 .909631 m 3.6960 X 166.643508 m 29.6678 X 1499.206543 m 4.4657 X 226.521810 m 3.9736 X 200.663249 m 12.9171 X 652.303061 m 3.7871 X 191 .243490 m 21 .5766 X 1089.599609 m 3.8572 X 194.783529 m 10.56S9 X 533.569336 m 7.0202 X 354.512532 m 4.1154 X 207.624707 m 6.5697 X 432.759603 m 3.1392 X 156.526646 m 7.3604 X 372.701009 m 5.0146 X 253.234663 m 3.2390 X 264.567058 m 4.7606 X 240.417460 m 4.3132 X 217.614127 m 13.5247 X 682.963398 m 7.0923 X 358.154297 m 6.7663 X 442.667988 m 3.9300 X 299.458621 m 6.6979 X 336.236491 m 2.9491 X 148.925781 m 90.4166 X 2546.000160 m 7.0963 X 336.357747 m 12.3461 X 633.565266 m 29.2241 X 1475.7B93B7 m 4.1029 X 207.194010 m 4.0191 X 202.962240 m 4.1960 X 211.993717 m 7.7286 X 390.299479 m 3.3977 X 161.681315 m 9.3247 X 470.666230 m 3.4225 X 172.631217 m 7.5733 X 382.446269 m 3.3044 X 166.870117 m 3.6900 X 166.340332 m 14.9444 X 754.679361 <n 3.4376 X 173.604330 m 4.2246 % 213.336216 m Spec, rv.22 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K22 oligosaccharide. - 187 -409 amu 409.09695 amu 51 .5352 X 2602.478027 m 410 amu 410.10311 amu 11.4466 X 578.043619 m 411 amu 411.11267 amu 3.3608 X 169.718424 m 421 amu 421.10048 amu 6.6341 X 436.014812 m 423 amu 423.12063 amu 3.2359 X 163.411456 m 425 amu 425.08669 amu 7.5540 X 381.469727 m 427 amu 426.52266 amu 6.1919 X 312.663105 m 427 amu 427.10978 amu 100.0000 X 5049.906418 m 426 amu 428.11567 amu 16.0192 X 909.952799 m 429 amu 429.11678 amu 4.4011 X 222.249349 m 449 amu 449.08971 amu 3.9369 X 196.611649 m 463 amu 463.09710 amu 4.4152 X 222.961426 m 465 amu 465.13098 amu 3.4356 X 173.502604 m 477 amu 477.13950 amu 3.3669 X 170.023600 m 461 amu 481.11229 amu 2.9531 X 149.129232 m 493 amu 493.12016 amu 6.0621 X 306.131999 m 495 amu 495.13158 amu 2.9366 X 148.396810 m 507 amu 507.14363 amu 5.6161 X 283.610026 m 511 amu 511.12571 amu 8.7804 X 443.400065 m 525 amu 525.14990 amu 6.6281 X 435.709636 m 539 amu 539.11301 amu 6.6225 X 334.431966 m 541 amu 541 .12177 amu 3.1582 X 159.484863 m 555 amu 555.15096 amu 5.2419 X 264.709473 m 573 amu 573.17500 amu 3.9909 X 201.538086 m 561 amu 581.13754 amu 12.7694 X 645.853676 m 563 amu 583.12633 amu 4.5948 X 232.035319 m 597 amu 597.16319 amu 7.8747 X 397.664368 m 615 amu 615.16897 amu 11 .9406 X 602.9B6654 m 679 amu 679.18124 amu 5.1029 X 257.690430 m 691 amu 691.36671 amu 4.5356 X 229.044596 m 703 amu 703.19201 amu 5.6488 X 285.257976 m 721 amu 721.18972 amu 15.6964 X 602.754721 m 722 amu 722.16838 amu 4.9514 X 250.040690 <n 733 amu 733.20234 amu 4.1643 X 211.303711 m 751 amu 751.21309 amu 10.4116 X 525.777162 m 752 amu 752.19484 amu 3.3016 X 166.727702 m 775 amu 775.20791 amu 10.1417 X 512.145996 m 776 amu 776.22393 amu 4.0872 X 206.400553 m 793 amu 793.22478 amu 30.9210 X 1561.482746 m 794 amu 794.23952 amu 10.3887 X 524.617514 m 795 amu 795.16052 amu 3.2226 X 162.740072 m Spec. IV.22 High resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of Klebsiella K22 oligosaccharide, con't. - 188 -o I I I i 0^1 OS o • o DC o * o • § = i CD ' o - o o - o CO o - o CM A11SN31NI 3AIlV"l3cJ Spec, r v . 2 3 Low resolution negative-ion l.d.i.-F.t.-i.c.r. spectrum of E. coli K49 oligosaccharide.. - 189 Nominal mats Measured mass 188 amu 188.05702 amu 202 amu 202.07237 amu 333 amu 333.07777 amu 351 amu 351.08551 amu 364 amu 364.12315 amu 366 amu 366.10456 amu 368 amu 368.12358 amu 369 amu 369.01619 amu 369 amu 369.10702 amu 400 amu 400.08572 amu ' 402 amu 402.06656 amu 408 amu 408.10979 amu 409 amu 409.10914 amu 410 amu 410.10927 amu 420 amu 420.10804 amu 424 amu 424.13403 amu 426 amu 426.11463 amu 427 amu 427.12305 amu 434 amu 434.12144 amu 438 amu 438.11596 amu 440 amu 440.12333 amu 452 amu 452.13133 amu 456 amu 456.12985 amu 460 amu 460.10650 amu 462 amu 462.09651 amu 466 amu 466.01652 amu 469 amu 468.81784 amu 470 amu 470.14345 amu 471 amu 471.14614 amu 484 amu 484.06001 amu 486 amu 486.05056 amu 528 amu 528.09480 amu 572 amu 572.20427 amu 588 amu 588.17213 amu 629 amu 629.21419 amu 630 amu 630.22390 amu 657 amu 657.19213 amu 659 amu 659.20014 amu 660 amu 660.23585 amu 673 amu 673.24735 amu 674 amu 674 .24813 amu 716 amu 716.20160 amu 734 amu 734.25780 amu 791 amu 791.24591 amu 792 amu 792 .22926 amu B21 amu 821 .26608 amu 835 amu 835.28005 amu 836 amu 636.23190 amu 837 amu 837.30852 amu R e l . intensity Abs . Intensity 6.7795 X 621.044922 m 8.4905 X 777.783203 m 15.3319 X 1404.492187 m 6.7630 X 619.531250 m 14.9163 X 1366.601563 m 5.2434 X 460.322266 m 5.7806 X 529.541016 m 5.7449 X 526.269531 m 15.6789 X 1436.279297 m 11.9696 X 1096.4B4375 m 5.8185 X 533.007813 m 90.1780 X 6260.639644 m 15.5765 X 1426.904297 m 6.7401 X 617.431641 m 5.5088 X 504.638672 m 13.2749 X 1216.064453 m 99.8342 X 9145.410156 m 18.2694 X 1673.583984 m 6.4954 X 595.019531 m 10.3972 X 952.441406 m 5.8713 X 537.641797 m 19.2262 X 1761 .230469 m 21.5427 X 1973.437500 m 15.0451 X 1376.222656 m 7.6105 X 697.167969 m 5.1101 X 468.115234 m 5.3537 X 490.429688 m 100.0000 X 9160.595703 m 21 .8641 X 2002.880836 m 11.4787 X 1051.313672 m 5.1069 X 467.822266 m 12.5639 X 1150.927734 m 5.1554 X 472.265625 m 5.1767 X 474.218750 m 27.3713 X 2507.373047 m 6.5646 X 784.570312 m 7.1437 X 654.589844 m 15.9219 X 1458.544922 m 5.0248 X 460.302734 m 30.1873 X 2765.332031 m 8.8B44 X 613.867187 m 6.2220 X 369.970703 m 5.7428 X 326.074219 m 39.9267 X 3657.319531 m 13.4311 X 1230.371094 m 10.0803 X 923.437500 m 33 .9205 X 3107.324219 m 11.8223 X 1063.007613 * 5.3046 X 485.937500 m Spec. TV2A High resolution negative-ion l.d.i.-F.t.-i.c.r. oligosaccharide. spectrum of E. coli K49 - 190 -i i | — i i i AlISN31NI J 3AIlV13cJ Spec. IV.25 Low resolution positive-ion l.d.i.-F.t.-i.c.r. spectrum of E. coli K49 oligosaccharide. - 191 -Nominal mats Measured mass Rel. intensity Abs. intensity 204 amu 204.0B675 amu 19.0477 X 1578.320312 m 226 amu 226.07116 amu 11.6337 X 963.984375 m 366 amu 366.13983 amu 7.4895 X 620.585938 m 375 amu 375.10968 amu 5.9672 X 494.453125 m 388 amu 388.12091 amu 100.0000 X 6286.132812 at 389 amu 389.13688 amu 22.2138 X 1840.664062 m 390 amu 390.13884 amu 5.5571 X 460.468750 m 404 amu 404.08224 amu 39.8713 X 3303.789062 m 405 amu 405.09224 amu 7.9920 X 662.226562 m 406 amu 406.13010 amu 15.4230 X 1277.968750 m 420 amu 420.16247 amu 5.2464 X 434.726563 m 422 amu 422.11281 amu 17.0319 X 1411.289062 m 432 amu 432.16040 amu 11.9166 X 987.421875 m 436 amu 436.12365 amu 6.1384 X 50B.632B13 m 448 amu 448.14951 amu 42.8563 X 3551.132812 m 449 amu 449.14350 amu 9.5566 X 791.875000 m 450 amu 450.13256 amu 8.5992 X 712.539062 m 464 amu 464 .12825 amu 57.9093 X 4798.437500 m 465 amu 465.12927 amu 12.7776 X 1058.789062 m 466 amu 466.11601 amu 11 .4381 X 947.773437 m 488 amu 488.11823 amu 5.8140 X 481.757812 m 494 amu 494.15428 amu 13.6981 X 1135.039063 m 508 amu 508.16300 amu 7.7549 X 642.578125 m 510 amu 510.11520 amu 14.7116 X 1219.023437 m 524 amu 524.17479 amu 7.6317 X 648.945312 m 635 amu 635.23124 amu 5.1720 X 426.554688 m 651 amu 651.15376 amu 7.8308 X 648.867168 m 683 amu 683.21279 amu 7.7256 X 640.156250 m 711 amu 711 .30716 amu 5.0895 X 421.718750 m 756 amu 756.10590 amu 7.8015 X 646.445312 m 813 amu 813.22359 amu 15.6342 X 1295.468750 m 814 amu 814.19814 amu 5.4487 X 451.484375 in 829 amu B29.26689 amu 11.1387 X 922.968750 m 845 amu 845.02400 amu 18.0219 X 1493.320312 m 846 amu 846.04426 amu 6.4047 X 530.703125 m 875 amu 875.06413 amu 5.8343 X 483.437500 m 889 amu 669.10161 amu 7.8025 X 646.523437 <n 889 amu 869.46572 amu 9.4642 X 784 .216750 m Spec. IV.26 High resolution positive-ion l.d.i.-F.t.-i.c.r. oligosaccharide. spectrum of E. coli K49 

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.831.1-0059719/manifest

Comment

Related Items