STRUCTURAL STUDIES OF KLEBSIELLA CAPSULAR POLYSACCHARIDES by MO-TAI YANG B.Sc, National Taiwan University, 1963 M.Sc, National Taiwan University, 1965 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department °* Chemistry •< We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1974 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r equ i r emen t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h C o l u m b i a , I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s tudy . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y pu rposes may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d that c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l owed w i thout my w r i t t e n p e r m i s s i o n . Department o r The U n i v e r s i t y o f B r i t i s h Co lumbia Vancouver 8. Canada D a t e ^ j r , ABSTRACT The genus K l e b s i e l l a b e l o n g s t o t h e f a m i l y E n t e r o b a c t e r i a c e a e . E i g h t y t y p e s o f K l e b s i e l l a h a v e b e e n s e r o l o g i c a l l y c l a s s i f i e d u s i n g t h e c a p s u l a r p o l y s a c c h a r i d e s as a n t i g e n s . I n o r d e r t o u n d e r s t a n d t h e c h e m i c a l b a s i s o f s e r o l o g i c a l r e a c t -i o n s , s t r u c t u r a l s t u d i e s o f t h e c a p s u l a r p o l y s a c c h a r i d e s f r o m t h r e e s t r a i n s ° f K l e b s i e l l a ^ n a m e l y . K 5 , K 6 2 and K 1 8 were c a r r i e d o u t by means o f c o n v e n t -i o n a l m e t h o d s ( s u c h as h y d r o l y s i s , m e t h y l a t i o n and p e r i o d a t e o x i d a t i o n ) as w e l l a s m o d e r n m e t h o d s ( s u c h as c i r c u l a r d i c h r o i s m ( c . d . ) f o r t h e a s s i g n m e n t o f t h e £ o r L c o n f i g u r a t i o n o f m o n o s a c c h a r i d e s , g a s - l i q u i d c h r o m a t o g r a p h y ( g . l . c . ) and mass s p e c t r o m e t r y f o r i s o l a t i o n and i d e n t i f i -c a t i o n o f s u g a r m o i e t i e s ) . I n t h e c o u r s e o f t h e p r e s e n t i n v e s t i g a t i o n , p . m . r . s p e c t r o s c o p y a t 9 5 ° f o r t h e a s s i g n m e n t o f a n o m e r i c c o n f i g u r a t i o n s o f the c o n s t i t u e n t s u g a r s , and S e p h a d e x g e l f i l t r a t i o n f o r t h e s e p a r a t i o n o f a s e r i e s o f a c i d i c o l i g o s a c c h a r i d e s o b t a i n e d f r o m t h e p a r t i a l h y d r o l y s i s o f p o l y s a c c h a r i d e s f o r s e q u e n t i a l a n a l y s i s h a v e b e e n d e v e l o p e d . The r e p e a t i n g u n i t s o f K 5 , K 6 2 and K18 a r e as f o l l o w s . CH3 COOH V 4 A 6 K5 K 6 2 / \ D-GlcAp D-Glcp ~ D-Manp ~ — - - y6 ~ i ~ s$ - ~ s& 2 - 0 A c ~ D-GlcAp ~ D-Manp ~ ^ D-Galp ~~~ D-Glcp ~ -D-Manp K 1 8 ~ L-Rhap ~ D-Galp ~ ~ D-Glcp ~ D-GlcAp L-Rhap i ^ -cx 11 D - G l c p The structure of K5 capsular polysaccharide lacks any side chain and is also unusual in affording the f i r s t example of a 4,6-0-(l-carboxyethylidene)-D-raannose unit In a natural product. Capsular polysaccharide of K.62 belongs to the prevailing structural pattern so far published in this genus which is composed of a repeating unit involving a single sugar in the side chain with three to four mono-saccharides in the backbone. Capsular polysaccharide from K18 is unusua complex in that i t consists of two L-rhamnoses in the hexasaccharlde repeat unit which renders the polysaccharide less viscous and acid l a b i l e . Finally, an attempted immunological study on K5 is briefly discussed. i i i TABLE OF CONTENTS Page I. INTRODUCTION 1 II. METHODOLOGY OF STRUCTURAL STUDY ON POLYSACCHARIDES 7 1. General Structural Remarks of Polysaccharides .... 7 2. Isolation and Purification 8 3. Total Hydrolysis of Polysaccharides 9 4. Methanolysis 16 5. Methylation 17 6. Carboxy1 Reduction of Acidic Carbohydrates ....... 21 7. Partial Hydrolysis 22 8. Periodate Oxidation '.. 24 9. Gas-liquid Chromatography 28 10. Mass Spectrometry of Partially Methylated A l d i t o l Acetates 31 11. Determination of Anomeric Configuration 36 12. Determination of D- and.L-Configuration of a Sugar Residue 39 13. Immunochemical Methods 39 III. THE STRUCTURE OF THE CAPSULAR POLYSACCHARIDE FROM KLEBSIELLA K-TYPE 5 43 SUMMARY 44 DISCUSSION 44 EXPERIMENTAL 51 IV. THE STRUCTURE OF THE CAPSULAR POLYSACCHARIDE FROM KLEBSIELLA K-TYPE 62 77 i v SUMMARY 78 DISCUSSION 7.9 EXPERIMENTAL 84 V. THE STRUCTURE OF THE CAPSULAR POLYSACCHARIDE FROM KLEBSIELLA K-TYPE 18 101 SUMMARY 102 DISCUSSION 103 EXPERIMENTAL 108 APPENDIX A: IMMUNOCHEMICAL STUDIES ON KLEBSIELLA K-TYPE 5 POLYSACCHARIDE 126 APPENDIX B: P.M.R.SPECTRA 129 K5 CAPSULAR POLYSACCHARIDE 130 SODIUM HYDROXIDE TREATED K5 131 TWICE PER10DATE OXIDIZED AND SODIUM BOROHYDRIDE REDUCED K5. . 132 METHYLATED K5 133 ; METHYLATED AND REDUCED K5 134 CELLOBIOURONIC ACID ( A l ) FROM K5 135 ALDOTRIOURONIC ACID (A2) FROM K5 136 NEUTRAL DISACCHARIDE (N3 ) FROM K5 137 PYRUVIC ACID-MANNOSE-ERYTHRONIC ACID COMPLEX FROM K5 138 K62 CAPSULAR POLYSACCHARIDE 139 MILDLY DEGRADED (DILUTE ACID) K6.2 140 ALDOBIOURONIC ACID (A2) FROM K62 141 . ALDOTRIOURONIC ACID (A3;) FROM K62 1 142 PSEUDOTETRAOURONIC ACID (A4) FROM K62 143 V P a £ e DISACGHARIDE GLYCOSIDE FROM SMITH DEGRADATION OF K62 144 K18 CAPSULAR POLYSACCHARIDE 145 ALDOBIOURONIC ACID (A2) FROM K18...' 146 NEUTRAL DISACCHARIDE ( N l ) FROM K18 147 PSEUDOTRIOURONIC ACID (A3) FROM K18 148 TETRASACCHARIDE.(A4) FROM K18 149 A5a FROM K18 150 A6a FROM K18 151 MILD ACID TREATED POLY ALCOHOL FROM K18 ..' 152 APPENDIX C: EXPRESSIONS OF RAW SUGAR RATIO DATA FROM G.L.C.-ANALYSIS .153 EXAMPLES FROM K5 154 EXAMPLES FROM K5 156 EXAMPLES FROM K18 158 BIBLIOGRAPHY 161 v i LIST OF TABLES Table No. Page TABLE I RATE CONSTANTS FOR HYDROLYSIS OF THE GLYCOSIDIC BONDS IN THE AEROBACTER AEROGENES A3 (Sl) POLYSACCHARIDE 14 TABLE II RATE CONSTANTS FOR HYDROLYSIS OF THE GLYCOSIDIC BONDS IN SEROTYPE 2 CAPSULAR POLYSACCHARIDE OF AEROBACTER AEROGENES 15 TABLE III PRIMARY FRAGMENTS IN THE MASS SPECTRA OF PARTIALLY METHYLATED SUGARS IN THE FORM OF THEIR ALDITOL ACETATES 37 TABLE IV OPTICAL ROTATIONS AND P.M.R. DATA ON KLEBSIELLA K 62 CAPSULAR POLYSACCHARIDE AND DERIVED OLIGOSACCHARIDES . 81 v i i LIST OF FIGURES Figure No. Page Figure 1 The repeating unit of the Aerobacter aerogenes A3 (Sl) polysaccharide 14 Figure 2 The structure of the serotype 2 capsular polysaccharide of Aerobacter aerogenes 16 Figure 3 Periodate degradation of Klebsiella K-type 62 polysaccharide 25 Figure 4 Periodate degradation of Klebsiella K-type 5 polysaccharide 27 v i i i ACKNOWLEDGEMENTS I am s i n c e r e l y g r a t e f u l to Professor G.G.S. Dutton f o r h i s patience, encouragement and advice throughout the e n t i r e research and during the pre p a r a t i o n of t h i s t h e s i s . I wish to thank Dr. K.B. Gibney, Dr. Y'.M. Choy, Dr. G'.M. Bebault, Dr. J.M. Berry and Roger Walker f o r t h e i r v a l u a b l e d i s c u s s i o n s and s t i m u l a t i o n s . I a l s o g r a t e f u l l y acknowledge the in v a l u a b l e a s s i s t a n c e extented by Dr. P.C. Simon i n the immunological study on K5, Dr. P.J. S a l i s b u r y f o r growing the b a c t e r i a , and Dr. P'.E. Reid f o r the e l e c t r o p h o r e t i c a n a l y s i s . F i n a l l y , I would l i k e to express my si n c e r e g r a t i t u d e to Miss O l i v e Cheung f o r her a s s i s t a n c e i n the pre p a r a t i o n of t h i s manuscript. I. INTRODUCTION As a l l diseases are fundamentally chemical phenomena of livi n g organisms, a chemical approach to understanding the disease at the molecular level is essential. The defensive mechanisms of li v i n g organisms consist, in part, of a mechanical barrier, such as the skin. After this barrier i s penetrated by an invasive foreign agent (antigen), e.g., bacterium, the hosts' immune system is stimulated to produce humoral antibodies and cellular immunity. Besides the phagocytosis by macrophages, the specific antibody fa c i l i t a t e s the mechanisms responsible for the destruction and removal of the foreign agents. If the immune system f a i l s to remove the antigen(s), the disease persists and the host requires medical treatment. The indispensable role of the immune system in the body's defence is thus evident, though many of i t s basic processes are s t i l l , unknown. However, since the discovery of vaccination by Edward Jenner in 1798, some major breakthroughs have been achieved; i.e., discovery of cellular events in the mechanism of immunity (in addition to humoral immunity), molecular basis of antigenicity, and structural elucidation of antibodies (1). The determination of the structure of the slime polysaccharides of Klebsiella-- also examining the molecular basis of their antigenicity--|is the principle concern of this thesis. The structures of three strains of Klebsiella capsular antigens (K-antigens), namely K5, K62 and K18, were determined. The "inoculation of rabbits, guinea pigs and mice with K5 has been carried out, and no demonstrable antibodies have been obtained. Nevertheless, a study of antigenic determinants of these K-antigens has 2 been conducted by Eriksen (2) and by Heidelberger (3). Their results are consistent with the structures proposed in this research, and w i l l be mentioned briefly later. Generally, polysaccharides are considered to be weakly or non-antigenic though they may have high molecular weights. Unlike proteins polysaccharides lack a definite three-dimensional structure to activate the immune system. Nevertheless the repeating nature of the capsular polysaccharides is somehow "impressive", and possesses antigenicity. The genus Klebsiella belongs to the family Enterobacteriac A l l c e l l s of Enterobacteriaceae have lipopolysaccharide (LPS) bound and external to the c e l l wall. LPS can be extracted by reagents such as phenol or ethylenediaminetetraacetic acid (EDTA), suggesting that the linkage between lipopolysaccharide and the c e l l wall may be ionic rather than covalent. The lipopolysaccharide layer may be further enveloped by an extracellular capsule in the case of encapsulated strains. Most of the Klebsiella strains possess heavy capsules (K+) whereas the noncapsulated variants are separable from K+ cultures. Most of the capsular antigens are negatively charged. The surface charge confers resistance to phagocytosis, thus imparting virulence to the bacteria. Generally, slime can be considered as excreted capsular polysaccharide. Wilkinson (4) has shown that capsular and slime poly-saccharides are identical in chemical composition. However; since M-antigen (5,6) and DNA released as a result of autolysis of the c e l l may also be present in the slime (7), their presence must be viewed with some caution. 3 In the course of the structural study of Klebsiella capsular antigen type 5, a problem of irreversible mutation from K5+ to K5- (8) (prskov's notation, less encapsulated substrain) was encountered. This resulted in greatly reduced slime production and greater contamination in the slime of the neutral polysaccharides which were cleaved from 0-antigen. The heavy contamination was very hard to remove completely. S e r o l o g i c a l l y K5+ and K5- are the same strain, but K5- gives better immunological response, possibly due to a much thinner capsule which results in less incidence of immunoparalysis in the immunized animal (8). That K5- is the better antigen and that i t s antiserum reacts solely against the capsular polysaccharide is of great importance. This property i s also similar to the blood group antigens whose end group monosaccharides are the antigenic determinants (9). These two examples are evidence for the principle that the exposed components are the determinants of the whole antigen. Additionally, allergenic constituents of pollen which induce hypersensitivity are shown to be located in the outer c e l l wall (exine) of the pollen grain (10). In the present study of Klebsiella capsular polysaccharides classic and modern methods and sometimes, new approaches have been employed. The chemistry of Klebsiella has been reviewed by Luderitz, Jann and Wheat (11). Now more than eighty strains have been serologically identified. Nimmich (12,13) has reported the qualitative analysis of the constituent sugars of K-types 1 to 80. Most of them contain D-glucuronic acid in combination with D-mannose, D-glucose, D-galactose and, to a lesser extent, with L-rhamnose and L-fucose. About half of the polysaccharides are found to incorporate pyruvic acid. Uronic acid and/or 4 pyruvic acid confer the acidity on a l l of the Klebsiella capsular polysaccharides. Wheat, Dorsch and Godoy were the f i r s t to report the presence of pyruvic acid in K. rhinoscleromatis polysaccharide (K-type 3) (14), and the f i r s t complete structural study was carried out on K54 by Sanford and Conrad (15,16) using improved methylation procedures devised by Hakomori (17). The occurrence of pyruvic acid as a covalently linked ketal unit in polysaccharides was f i r s t reported by Hirase (18) in a study on agar which contains 4,6-0_-(l-carboxyethylidene)-D-galactose units. The presence of 4,6-0-(l-carboxyethylidene)-D-glucose units was subsequently demonstrated by Sloneker and Orentas (19) in the exocellular polysaccharide from Xanthomonas campestrls and the occurrence of 3,4-0-(l-carboxyethylidene)-L-rhamnose was found in the capsular polysaccharide of Klebsiella K-type 72 (20). The presence of the D-galactose ketal has also been shown in the polysaccharides from Corynebacterium insidiosum (21) and Klebsiella K-type 21 (22). Pyruvic acid has been found as well in the M-antigens °f Salmonella and in E.coli linked as 3,4-0-(l-carboxyethylidene)-D-galact-ose (23,24). In addition to these ketals of known structure the existence of pyruvic acid has also been demonstrated in other bacteria (14,25,26) and in certain cases this acid may be an immunodominant group (26). When our structural study on Klebsiella capsular polysaccharides was started, the only known structures were K-types 2 (27), 8 (28), and 54 (15,16). Now the structures of K-types 9 (29), 20 (31), 21 (22), 24 (32), 38 (33), 47 (30), 52 (34), 56 (35) and 72 (20) have also been reported. A detailed structural study of K-types 5 (36,37), 62 and 18 w i l l be the contribution of this thesis. The structures of K5, 62 and 18 can be represented as follows. CH3 COOH K5 V 1 4 1 3 x 1 D - GlcAp ~g D - Glcp D - Manp ~g 2-0Ac K62 - 2 — | - GlcAp ^ D - Manp ^ D - Galp ^ D - Glcp ~ ^ 1 D - Manp K18 — L - Rhap ~ ~ D - Galp — - O - Glcp ~ 1 D - GlcAp—-L-Rhap D -Glcp K5 slime polysaccharide comprises a repeating unit of only three monosaccharides. This polysaccharide lacks any carbohydrate side chain and is also unusual in affording the f i r s t example of a 4,6-0-(l-carboxy-ethylidene)-D-mannose unit in a natural product. K62 slime polysaccharide belongs to the prevailing structural pattern of this genus, i.e., a repeating unit composed of one side unit to a backbone of t r i - or tetra-saccharide. K18 capsular polysaccharide is composed of hexasaccharide repeating units with two L-rhamnoses in each repeating unit and one D -glucose as the side chain. This polysaccharide is comparatively complicated due to the length of the repeating block and the presence of L-rhamnoses which are acid l a b i l e . These structures are proposed on the basis of methylation analysis, periodate oxidation and partial hydrolysis studies. The general methods of investigation are outlined in section II, and the specific evidence on which the structures of K5, 62 and 18 are based is discussed in sections III, IV and V, respectively. Finally, an attempted immunological study on K5 is briefly discussed. II. METHODOLOGY OF STRUCTURAL STUDY ON POLYSACCHARIDES 1. General Structural Remarks on Polysaccharides Polysaccharides, with or without obvious repeating units, can be cl a s s i f i e d into one of the following groups: A. Linear horaoglycans. Such as amylose, cellulose and pectin. B. Branched horaoglycans. Such as amylopectin, glycogen and dextran. C. Linear heteroglycans. Such as Pneumococcus type III and Klebsiella K-type 5 polysaccharides. D. Branched heteroglycans. Such as wood hemicelluloses and plant gums. In the case of a polysaccharide consisting of repeating units, a structural study should characterize the nature of the sugar residues, the position of linkage, the sequence, the anomeric configurations, and i f possible, the molecular weight. In other words, the exact structure of the repeating unit and an approximate degree of polymerization have to be demonstrated. On the other hand, the exact overall sequence can not be determined in the case of polysaccharides lacking repeating units. The investigations can only provide information on the "average structure" including the different monosaccharide units present, the type of end groups,th p o s i t i o n of the-branch points and the nature of linkages. Some polysaccharides may have a main chain (such as hemicelluloses) while others may consist of a multiple-branched system in which no main chain can be defined (e.g., amylopectin and gum exudates). The following discussions w i l l only be concerned with poly-saccharides with a repeating unit. Of course, most of the techniques are also applicable to the polysaccharides of irregular structure. 8 2. Isolation and Purification Each strain of Klebsiella was inoculated in a sucrose-yeast-extract broth culture in an Erlenmyer flask which was shaken for 24h unti l a definite turbidity due to the growth of bacteria was observed. The culture was then spread on trays on sucrose-yeast-extract agar for 3 or 4 days to reach a maximum growth of slime. Slime was harvested, diluted with aqueous phenol and centrifuged. The acidic polysaccharide in the supernatant was precipitated with 5% hexa-decyltrimethyl ammonium bromide (Cetavlon). The precipitate was dissolved in brine again precipitated with ethanol, then dialyzed against tap water. The purified, deionized acidic polysaccharide was subjected to micro-analysis for nitrogen and ash, as well as ultraviolet spectrophoto-metry analysis for protein and nucle i c acids content. The constituent sugars were qualitatively examined by paper chromatography following hydrolysis of the polysaccharide, while quantitative analysis was performed by gas-liquid chromatography (g.l.c.) on the alditol acetates derived from the sugar residues. Normally, i t is not d i f f i c u l t to obtain a purified acidic polysaccharide. Most of the contamination is by neutral polysaccharides which are not precipitated by Cetavlon and as a result remain in the supernatant. In the case of heavy contamination by neutral polysaccharides, Cetavlon precipitation has to be repeated until a constant sugar ratio of the monosaccharides i n the a c i d i c polysaccharide i s obtained. However, i f the contamination is overwhelming, there is d i f f i c u l t y of singling out the desired fraction and an examination of the slime-producing bacteria is necessary, as the Klebsiella bacterial culture may be contaminated or 9 undergoing mutation to a substrain which synthesizes only a minor fraction of the acidic polysaccharide, as in the case of Klebsiella K-type 5. However, the minute yet persistant contamination (e.g., a galactan in K5 acidic polysaccharide) can be ignored, for i t s presence is too l i t t l e to be considered part of the repeating structure. Cetavlon precipitation has been proven sufficient for the purpose of purification of Klebsiella capsular polysaccharides, provided that the i n i t i a l separation of the cells by centrifugation is complete and gives a clear yellowish supernatant. Separation, by anion exchange column using diethylaminoethyl (DEAE) Sephadex is only suitable for analytical purpose on a milligram scale due to the extraordinarily high viscosity of the acidic polysaccharide. Purification by precipitation with the specific antiserum to the acidic polysaccharide is not feasible, since the isolated acidic polysaccharide is a very weak immunogen and e l i c i t s antibody rather poorly. 3. Total Hydrolysis of Polysaccharides The f i r s t step in the investigation of a polysaccharide structure is total hydrolysis followed by analysis of the hydrolyzed products qualitatively and quantitatively. Hydrolysis using hydrochloric acid (IM) is known to be more destructive on sugars than sulfuric acid (0.5M) (38), and 6 to 8 h of heating at 100° in either acid is sufficient to break down completely a neutral polysaccharide into i t s monomers. Hydrochloric acid can be removed at diminished pressures at 30-40° while sulfuric acid has to be neutralized with barium carbonate, and after removal of the precipitate (barium sulfate plus excess barium carbonate) the supernatant is deionized by cation then anion exchange resins. 10 Two molar trifluoroacetic acid (TFA) (39) has approximately the same hydrolytic strength, as compared to IM hydrochloric acid and 0.5M sulfuric acid, and i t does not significantly degrade sugars in the conditions normally used for hydrolysis ( i . e . 100°, 6-8h). Moreover, trifluoroacetic acid is easily removed from the reaction mixture by evaporation, and has been frequently used in this study. As to acidic polysaccharides, the conditions employed in the hydrolysis of neutral polysaccharides are not strong enough to ensure a complete cleavage of the glycuronosyl-bonds, and a considerable amount of aldobiouronic acid remains intact. The incomplete release of the uronic acid and the neutral sugar attached to i t give rise to a discrepancy of the sugar ratio in the hydrolyzate and the sugar ratio in the original polymer. Nevertheless, incomplete hydrolysis i t s e l f is a useful indication of the possible combination of the aldobiouronic acid. In order to achieve complete hydrolysis, an acidic polysaccharide can be converted to the neutral alditois by methanolysis to form methyl glycosides and the methyl ester methyl glycoside which is reduced in methanol with sodium borohydride to the methyl glycoside. The mixture of methyl glycosides is then hydrolyzed with 2M trifluoroacetic acid and reduced to alditois. The sugar ratio is obtained by g.l.c* analysis of the alditoi acetates. Alternatively, a direct total hydrolysis without transforming the uronic acid to a neutral polyol can be achieved by i n i t i a l hydrolysis and reduction in the usual manner, followed by more stringent conditions to further hydrolyze the remaining aldobiouronic acid in the alditoi form which is more resistant to acidic degradation than reducing sugars. By removing the uronic acid in i t s acidic form from the hydrolyzate with anion-exchange resin, the ratio of the constituent 11 neutral monosaccharides can be obtained by g.l.c. analysis of the alditol acetates. By comparing the ratio of only neutral sugars with the ratio of neutral and the transformed uronic acid, the type of the uronic acid as well as the molar ratio of individual sugars can be obtained. The preceding design was applied to K5. acidic polysaccharide because of i t s particular relevancy and partially applied to K62 and K18. F i r s t l y , K5 was methanolyzed (4% methanolic hydrogen c h l o r i d e , 100°C 10 h) to methyl glycosides which were then reduced in methanol by sodium borohydride, then hydrolyzed, reduced, and the polyol mixture was analyzed by g.l.c. as alditol acetates. The ratio of D-glucitol hexaacetate (mp. 95°) to D-mannitol hexaacetate (mp. 123°) was 2:1. Secondly, in order to achieve a total hydrolysis followed by the removal of the acidic sugars, K5 was i n i t i a l l y hydrolyzed in the usual way (2M trifluoroacetic acid, 10 h, 97°). The hydrolysate was evaporated to dryness then made alkaline with 0.5M sodium hydroxide and reduced with sodium borohydride at room temperature for lh to a mixture of mono- and disaccharide alditols. Hydrochloric acid was added to a concentration of 2M and the solution was heated at 100° for 4h to achieve further hydrolysis. After sodium borohydride reduction, the hydrolyzate was acidified and passed through anion-exchange resin to completely remove L-gulonic acid. The neutral effuent was analyzed and found to contain D-glucose and D-mannose in a ratio of 1:1. Thus one can conclude that the uronic acid in K5 is D-glucuronic acid and the repeating unit of the polysaccharide consists of D-glucuronic acid, D-glucose and D-mannose in a 1:1:1 ratio. On the other hand, i f knowledge of the type of the acidic sugar 12 rather than the exact ratio of sugars is desired, a modified procedure can be used. The acidic polysaccharide is hydrolyzed with 2M hydrochloric acid at 100° for 7h. A portion of the hydrolyzate is treated with anion-exchange resin to remove a l l the acidic components and the neutral sugars are analyzed. A separate portion of the hydrolyzate i s evaporated to dryness. The residue is reduced with sodium borohydride in methanol to convert neutral sugars as well as glucuronolactone into alditois, which are then analyzed. These two analyses of sugars can provide the information as to the type of the uronic acid and an approximate sugar ratio. A milder reagent, 2M trifluoroacetic acid, can be used for hydrolysis instead of 2M hydrochloric acid with lesser depolymerization and degradation of sugars. After examination of various hydrolysis conditions applied to the depolymerization of polysaccharides, i t i s desirable to have an overall survey of the acid l a b i l i t y of glycosidic linkages to provide a basis for the choice of a certain condition of hydrolysis. Capon (40) has reviewed comprehensively the first-order rate constants for the acid-catalyzed hydrolysis of the glycosides of mono-saccharides. Those data can be extrapolated to provide information as to the l a b i l i t y of the glycosidic linkages in a polysaccharide. To demonstrate the difference in rate of hydrolysis with respect to different types of glycosidic linkage, examples from poly-saccharides w i l l be given for the purpose of easy comparison. For simplification, only the major factors governing the rate of hydrolysis are discussed, i.e. the ring size, the type of the sugars, the position of sugar units in the polymer, and the anomeric configuration of the glycosidic bonds. Several generalizations (41) can be made: 13 a. Furanosides are more labile than pyranosldes. b. Deoxysugars are more readily hydrolyzed than hexoses (27). c. Uronic acids are more resistant to hydrolysis and as a result aldobiouronic acids always persist at the end of hydrolysis. d. Amino sugars are more acid resistant than common hexoses. e. Pentopyranosides are more acid labile than hexopyranosides. f. Alpha glycosidic bonds are generally more labile than beta. g. Residues present as side chains are often more easily hydrolyzed than when present in the main chains. Examples demonstrating these main points are discussed below. A pentosan isolated from the water-insoluble portion of Durum wheat endosperm (42) was found to have a general structure consisting of D-xylopyranosyl units linked 1—»4 with single-unit L-arabinofuranosyl side chains attached to the 2 and 3 positions of the xylose units. On mild hydrolysis of the pentosan (0.025M sulfuric acid, 75-77° for 4.5h) and followed by precipitation with ethanol, at a fina l concentration of 85%, the supernatant was shown to contain 60% of the L-arabinose and trace of D-xylose; while the precipitate contained over 90% of the D-xylose. The observation of the extreme acid l a b i l i t y of the L-arabinose indicated that i t existed in the furanose form. The water-soluble wheat flour pentosan with a similar structure possessed the same acid-labile property (43)* Klebsiella 0 group 9 lipopolysaccharide (44) is composed of D-galactofuranose and D-galactopyranose units. When the polysaccharide was methylated and treated with 90% formic acid at 70° for 45 min., a methylated disaccharide was released as a result of preferential cleavage of positions corresponding to the furanosyl linkages. 14 From the study of the hydrolysis of Aerobacter aerogenes A3 (Sl) polysaccharide (Fig. 1), Conrad et a l . (16) have reported some useful rate constants for hydrolysis of glycosidic bonds by measuring the release and: subsequent/cleavage of oligosaccharides as well as the progressive release of monosaccharides during the hydrolysis. Figure 1: The repeating unit of the Aerobacter aerogenes A3 (Sl) polysaccharide. J L D-Glcp ^ ~ D-GlcAp ~ L-Fucp - — — — / O — _ o < _ — D-Glcp TABLE I RATE CONSTANTS FOR HYDROLYSIS OF THE GLYCOSIDIC BONDS IN THE AEROBACTER AEROGENES A3 (Sl) POLYSACCHARIDE. Rate Constant Bond cleaved (sec _ 1) X 10 3 L-fucose D-glucose 1.3 D-glucose D-glucose 0.13 D-glucose i ~ D-glucuronic acid 0.05 D-glucuronic acid i - 3 - L-fucose 0.004 In the Aerobacter aerogenes A3 (Sl) polysaccharide the l a b i l i t y of the L-fucosidic linkage i s about 300 times greater than that of D-glucuronosidic linkage and the st a b i l i t y of other glycosidic bonds l i e s in between. Although identical in type of linkage, D-glucose as a side 15 chain is 2.6 times as labile as compared with i t s l a b i l i t y when in the backbone. The marked l a b i l i t y of L-fucosyl bonds explains the observation that oligosaccharides werefridt found containing _L-fucosyl linkages. Thus the anomeric configuration of L-fucose in the polysaccharide could not be determined by p.m.r. spectroscopy of oligosaccharides. This kind of d i f f i c u l t y was circumvented in this laboratory by using intact poly-saccharide for p.m.r. measurements. A study of rate constants for hydrolysis of the glycosidic bonds in the serotype 2 capsular polysaccharide of Aerobacter aerogenes (Fig. 2) (27) gave the data lis t e d in Table II. Figure 2: The structure of the serotype 2 capsular polysaccharide of Aerobacter aerogenes. D-Glcp i - ^ D-Manp Kr D-Glcp - — — — /? = I —- = — Oi D-Glcp TABLE II RATE CONSTANTS FOR HYDROLYSIS OF THE GLYCOSIDIC BONDS IN SEROTYPE 2 CAPSULAR POLYSACCHARIDE OF AEROBACTER AEROGENES. Rate constant Bond cleaved (sec" 1) X 10 D-glucose D-mannose D-glucose D-mannose 0.119 0.285 0.011 0.362 16 One can notice the increased rate for hydrolysis of the D-glucuronic acid as the branch unit in the serotype 2 capsular polysaccharide of Aerobacter aero genes (0.011 sec "*•), as compared to the D-glucuronic acid (in main chain) in Aerobacter aerogenes A3 (Sl) polysaccharide (0.004 sec~\ Table I ) . Fi f t y percent of D-mannose in Klebsiella K62, f i f t y percent of D-glucose in Klebsiella K18, and aldobiouronic acid in Klebsiella K20 (31) are a l l branching units and preferentially released in the course of hydrolysis. 4. Methanolysis Bishop et a l . (45-47) have conducted systematic studies of the methanolysis of pentoses and hexoses, and reported in detail on the formation, isomerization, distribution, and stability of various anomeric and ring isomers in terms of conformational analysis. Entlicher and BeMiller (48!) reported that methanolysis of the glycosidic bonds of amylose was more facile than acid hydrolysis and that methanolysis gave predominantly, or perhaps complete, inversion. Sandford et a l . (49) resorted to methanolysis for the sugar analysis of black yeast NRRL Y-6272 polysaccharide which was composed of N-acetylglucosamine and N-acetylglucosaminuronic acid, because the acidic polysaccharide underwent extensive degradation to form humin upon acid hydrolysis. Methanolysis was employed in the complete depolymerization of acidic polysaccharides as mentioned in section II 3. F.m.r. spectroscopy is a useful physical method for the detection of acidic components such as formyl, acetyl and pyruvic acid ketal units (50,51) in a polysaccharide. Alternatively, a chemical method 17 consisting of methanolysis followed by alkaline hydroxylamine treatment (52) on the polysaccharide has been devised to convert the acidic components in the polysaccharide into their -hydroxamateswhich are subsequently analyzed by thin layer chromatography ( t . l . c ) . Polysaccharide from K5 was methanolyzed and treated with alkaline hydroxylamine. Hydroxamates of acetic, pyruvic and D-glucuronic acid were detected on t . l . c . by comparison with authentic standards. It was found in the course of methanolysis of K5 that the pyruvic acid ketal, in contrast to glycosidic bonds and ester linkages, was quite resistant to methanolysis. This discovery was applied to the partial hydrolysis of periodate-oxidized and sodium borohydride-reduced K5 to obtain an oligosaccharide containing pyruvic acid. An attempt was made to break down K5 by methanolysis to isolate oligomers carrying pyruvic acid. The product of methanolysis was separated by paper chromatography and fafractions with an p.m.r. signal-of the- pyruvate methyl group were obtained. But no further purification and identification was done. This method merits further investigation for i t s general applicability for the isolation of the complex of sugar-pyruvate ketal. 5. Methylation In the technique of methylation analysis,the free hydroxyl groups in a polysaccharide are f i r s t exhaustively etherified and then the glycosidic bonds are hydrolyzed to obtain partially methylated monosaccharides which are subsequently separated and identified. Structural information such as branchings and types of sugar can thus be derived. The methods of methylation generally used individually or in combination are: a. Hakomori methylation (17), b. Kuhn methylation (53), 18 and c. Purdie methylation (54). a. Hakomori methylation: The methylsulfinyl anion prepared by heating dimethyl sulfoxide (DMSO) with sodium hydride at 60-70° (15) was used to generate alkoxide ions from free hydroxyl groups in a polysaccharide. Methyl iodide was then added to the polysaccharide alkoxide to produce a methylated polysaccharide which may be purified by dialysis against tap water. A polysaccharide containing uronic acids is not suitable for more than one Hakomori methylation, because the methyl ester of the uronic acid can react with methylsulfinyl anion to form methylsulfinylmethy1 ketone; and the methyl ester can undergo ^-elimination in a strongly basic medium. The jB -elimination is particularly pronounced when the uronic acid is substituted at C4. These side reactions can be minimized i f the incompletely methylated polysaccharide is treated with dilute aqueous alkaline acetone solution prior to the second methylation. b. Kuhn methylation: When the i n i t i a l degree of methylation is low following the Hakomori method, the Kuhn methylation may be used next. The partially methylated polysaccharide is dissolved in a mixture of N, N-dimethyl-formamide (DMF) and methyl iodide. Silver oxide is added in small portions at various time intervals with st i r r i n g . The reaction is allowed to proceed for 16 h, then the mixture is centrifuged and the solid substance is washed with chloroform. The combined centrifugate and chloroform washings are washed with water then evaporated to dryness. c. Purdie methylation: . This method is generally used to ensure maximum methylation of the polysaccharide after Hakomori, or Hakomori and Kuhn methylations. 19 The partially methylated polysaccharide is dissolved in refiluxing methyl iodide and silver oxide is added in small portions at several time intervals over 16 h. The reaction mixture i s centrifuged and the solid material is extracted with boiling chloroform three times. The combined centrifugate and washings are evaporated to dryness and the product is examined by IR spectrophotometry to determine the extent of raethylation. Occasionally colloidal silver salts persist as a suspension in the chloroform layer. Evaporation to dryness followed by dissolving in carbon tetrachloride can sometimes break down the colloidal silver salt. If this treatment does not work, one can resort to complexing the silver ion with cyanide and removing Ag(CN>2 i o n b y extraction with water. The chloroform solution is washed several times with water, and the cyanide solution is discarded. Oligo- and monosaccharides are easily methylated by the Hakomori and Kuhn methods. The-procedures-used are s i m i l a r to those described in the preceding paragraph except for the dialysis step due to the fact that oligo- and monosaccharides are dialyzable. Therefore, various procedures for isolation and purification of the products of Hakomori or Kuhn methylation of low molecular weight carbohydrates are now discussed. For the isolation of methyl glycosides of tetramethyl monosaccharides, the methylation mixture is poured into ice water and the glycosides are extracted with petroleum ether (65-110°) which is then washed with small volumes of ice water. The petroleum ether layer is withdrawn, dried over anhydrous sodium sulfate and evaporated to dryness to collect the product. This method can be extended to 20 trimethyl glycosides but not to dimethyl glycosides, because of the higher solubility of the latter in water, and chloroform has to be used instead. Methylation mixtures of d i - and higher oligosaccharides can be poured into ice water then extracted with chloroform which is in turn washed with cold water. Generally i t is very d i f f i c u l t to remove completely dimethyl sulfoxide from the chloroform extract. For less volatile disaccharide (e.g. aldobiouronic acids) and higher oligomers, i t is more convenient to d i s t i l dimethyl sulfoxide under vacuum directly from the methylation mixture, after excess chloroform is removed. The solid residue is dissolved in water which is in turn extracted with chloroform. The organic layer is dehydrated over anhydrous sodium sulfate and evaporated to dryness. Traces of dimethyl sulfoxide remaining can be easily removed by d i s t i l l a t i o n under vacuum for a short period. If the disaccharides are very vo l a t i l e (e.g. sucrose and neutral disaccharides containing deoxysugars) the duration of vacuum d i s t i l l a t i o n should be as short as possible to avoid the loss of methylated sugars. When an oligosaccharide is methylated by the Kuhn method, the reaction mixture is centrifuged and the solid is washed with chloroform. The combined centrifugate and washings are evaporated to remove chloroform and further high vacuum d i s t i l l a t i o n is used to remove N, N-dimethyIformamide. In the course of methylation the sugar unit at the reducing end may undergo equilibration between furanoside and pyranoside. The methylation of the furanoside form may be substantial when the sugar has a stable furanoside isomer (e.g. galactose), and reaction at a low temperature is required to avoid the side reaction. 21 6. Carboxyl reduction of acidic carbohydrates. Reduction of carboxyl groups in acidic carbohydrates can be carried out at four different stages during a structural investigation: a. Direct reduction on native acidic polysaccharides (55). The polysaccharide is f i r s t propionylated with propionic anhydride in pyridine (or pyridine and formamide), followed by treatment with lithium borohydride in anhydrous tetrahydrofuran (THF). The reduction mixture is dialyzed against tap water for 24h and the carboxyl-reduced polysaccharide remaining in the dialysis tubing is freeze-dried. For the reduction of the carboxylic groups in K5 this method was used. b. Lithium aluminum hydride reduction of the methylated acidic polysaccharide. The reduction is carried out in anhydrous tetrahydrofuran or ether. This is one of the approaches to obtaining the exact sugar ratio as well as the nature of the uronic acid, since the neutral polysaccharide is readily hydrolyzable to i t s constituent units, and the neutral sugar derived from the acidic unit is easier to identify both qualitatively and quantitatively than the original uronic acid. c. Carboxyl reduction of acidic oligosaccharides.' The oligosaccharides are esterified either by diazomethane (CH^2) o r b v methylation. The methyl ester of the original oligosaccharide is reduced with sodium borohydride in methanol while the permethylated acidic oligosaccharide is reduced under the same conditions or in anhydrous tetrahydrofuran by lithium aluminum hydride. d. Conversion of uronic acid to the corresponding al d i t o l through the uronolactone (56). D-Glucurono- and D-mannuronolactone can be formed by d i s t i l l i n g concentrated hydrochloric acid or trifluoroacetic acid from the corresponding acids. The lactone is then reduced to the alditol by sodium borohydride in methanol. 22 7. Partial hydrolysis. A series of oligosaccharides from the partial hydrolysis of a polysaccharide is the key to the determination of the sequential arrangement of the constituent monosaccharides in that polymer, for the oligosaccharides reveal the relative position of one building block to another in an additive manner. To achieve the structural elucidation of oligosaccharides methods such as total hydrolysis, reducing end group analysis, progressive partial hydrolysis, methylation, p.m.r. spectroscopy and enzymatic hydrolysis are employed. From the discussion on rates of hydrolysis of glycosidic linkages in section II 3, i t is clear that appropriate conditions (concentration of acid, temperature and length of hydrolysis) can be chosen to effect preferential cleavage of certain weaker linkage(s) and thereby rel e a s i n g preferentially certain subunits of the polymer. For example, K5 was autohydrolyzed at 97° and pH 2.7 (1.2% aqueous solution) to produce a series of acidic oligosaccharides (see page 63, A^.g) and a neutral disaccharide. Hydrolysis of K62 in IM t r i f luoroacetic acid at 97° for Ih yielded-an a-ldotriouronicaacid as the major oligosaccharide plus tetra- and penta-saccharides with very l i t t l e aldobiouronic acid; while hydrolysis of K62 in IM trifluoroacetic acid at 97° for 5h gave an aldobiouronic acid as .the..major, product. When K18 o was heated at 97 in 0.5M trifluoroacetic acid for 0.5h aldotetrao-, aldopentaouronic acid and higher oligomers were produced, while heating K18 under the same conditions for 4h afforded aldobiouronic acid as the major product plus a trace of higher oligosaccharides. At the end of partial hydrolysis the solution was evaporated to dryness, and the residue was twice dissolved in water and again 23 evaporated to dryness to remove most of the trifluoroacetic acid. The partially hydrolyzed mixture can be separated directly by preparative paper chromatography, or applied to a charcoal-Celite (1:1) column using f i r s t water to elute the monosaccharides, which are.the major components of the mixture and interfere, with'the^sep^a'ration of oligosaccharides by 'gel- ' chromatography (57),and then eluting with aqueous ethanol and aqueous isopropanol to recover the oligo-saccharides. The fractions containing oligosaccharides are concentrated and fractionated on a column of Sephadex G 15 using water as eluent. Partially hydrolyzed K5 was separated on f i l t e r paper to obtain two disaccharides, one trisaccharide and higher oligomers. Partially hydrolyzed K62 was separated f i r s t on a charcoal-Gelite column to remove monosaccharides then successfully fractionated on Sephadex G 15 column to afford an aldobiouronic acid, an aldotriouronic acid, an aldotetraouronic acid which was further purified by paper chromatography, and aldopentaouronic acids. The partially hydrolyzed K18 was processed on a charcoal-Celite column to remove the monosaccharides. The oligo-saccharides of K18 were not separable into pure fractions by Sephadex G 15 but were successfully fractionated by preparative paper chromatography to obtain a series of oligosaccharides. To obtain an oligosacchride sufficiently pure for p.m.r. spectroscopic analysis of the anomeric protons, Sephadex (G 15. or G 10) gel f i l t r a t i o n was frequently employed to purify the oligosaccharide which had already been separated by paper chromatography. 24 8. Periodate oxidation (58). Vicinal diols are oxidatively cleaved by periodate resulting in the formation of dialdehydes . In the case of contiguous t r i o l s , formic acid w i l l be derived from the central carbon. The product of periodate oxidation, polyaldehyde, is reduced to a polyol which can be totally hydrolyzed or partially hydrolyzed with acids. Total hydrolysis of the polyol can provide considerable information of the structure of the polymer. Partial hydrolysis of the polyol from the oxidized and reduced polysaccharide was developed by F. Smith and coworkers. The Smith degradation is a powerful tool which can be used to cleave the poly-saccharide specifically at the sugar unit containing vicinal diols to provide fragments which aid in the structural analysis. The cause of the specific cleavage can be explained as follows. When the sugar unit in a polysaccharide is cleaved by periodate ions and subsequently reduced, the resultant polyalcohol is a true acetal and is more susceptible to acid than the glycosidic linkages of the intact sugar unit which is immune to periodate oxidation. (The relative rate (59) for hydrolysis is 10 3 to 10 4 :1). The usefulness of Smith degradation of K62 and total hydrolysis of the polyol is demonstrated in Figure 3. (See page 25) Periodate oxidation is complicated both by over and under oxidation (60-63). Over oxidation can be minimized by using dilute solutions of periodate buffered at a proper pH (e.g. 4), and by keeping the reaction mixture in . the dark at low temperature (4°). Incomplete oxidation can occur due to the formation of formyl esters of the reducing sugars, the formation of intramolecular hemiacetal through the 25 Figure 3. Periodate degradation of Klebsiella K-type 62 polysaccharide (annular protons are not shown). CH2OH CH2OH HCOOH + C HO OH C C°2H HOHoC 9 9 „ , CH20H 2 | 0 / HO J-LQ CH2OH HOH2C COOH H0H2p HOH2C CH20H C H20H HOH2&) total hydrolysis CH2OH partial hydrolysis CH9OH I z CHo0H HO-C-H * 1 CHO COOH CHO °\j-0 - C-H +•• H-C-QH +. H-C - OH t CH2OH 0 y 1 1 CH20H CH9OH CH20H OH Z CH2OH HO CH20H H , O H -+ H O CH 2 0 H •f H-C-OH I .CH2OH C H 2 O H H-fe-OH- C H 2 O H C H O " COOH' j H , O H + H - C - O H H C - O W + H C - Q H . + H C ^ - O H + ( ! ; H 2 O H C H 2 O H C H ^ H C H 2 0 H ' 2o dialdehyde generated by i n i t i a l periodate cleavage, as well as steric hindrance that limits the accessibility of the vicinal diols to periodate ions. As demonstrated in the periodate oxidation of K.62, the side chain unit, as a rule i s oxidized more rapidly than the backbone unit due to the steric factor. While carrying out the periodate oxidation of K5 (Figure 4), a persistent underoxidation was observed, even after the 0-acetyl group on C 2 of the D-glucose unit was removed by saponification. Hydrolysis of the oxidized and reduced K5 gave rise to D-glucose as well as O-mannose, demonstrating that D-glucose was protected through inter-residual hemiacetal by the condensation of the C 3 hydroxyl on D-glucose to the C 2 aldehyde of the degraded D-glucuronic acid. On the other hand, the carboxyl reduced K5 underwent complete reaction with periodate, both the D-glucose and the D-glucuronic acid being oxidized, and thus hydrolysis of the polyol gave rise only to D-mannose. The anomalous underoxidation of the native K5 is similar to the incomplete periodate oxidation of alginate, as well as the extremely slow uptake of periodate by polymer having contiguous I—>4 linked pyranosyl units, such as 1—>k linked galactan of lupin seeds, maize-cob xylan, and amylose. Painter and colleagues have reported the study of the under-oxidation of alginate (61). Three repetitions of periodate oxidation followed by reduction were required to reach the theoretical oxidation, li m i t . The cleavage of the C2-C3 diol to dialdehyde in the 1—>4 linked polymer 27 Figure 4. Periodate degradation of Klebsiella K-type 5 polysaccharide (annular protons not shown). 2 8 followed by the formation of a 6-membered dioxane-type stable inter-residue hemiacetal ring (see Figure 4) provide complete protection against the oxidation of the immediate neighboring sugar units. No intra-residue 6-membered hemiacetal involving the C6 group i s possible at C6, since this is a carboxyl group and not a primary alcohol. It should be noted that an acidic group may, through i t s inductive effect, possibly contribute to the pronounced st a b i l i t y of the protective inter-residue hemiacetal- a property less remarkable in a neutral polysaccharide. In the cases of 1—*4 linked D-glucan and D-galactan, circumstantial evidence points to an equilibrium favouring inter-residue instead of intra-residue hemiacetal, judging by the sluggish periodate uptake of those polysaccharides. The intra-residue hemiacetal leaves the neighboring sugar units free and susceptible to periodate oxidation. In the case of 1—»4 linked neutral polysaccharides, the intr i n s i c instability of the inter-residue hemiacetal and the existance of the competitive intra-residue hemiacetal in the equilibrium result in the ultimate theoretical periodate uptake. 9. Gas-liquid chromatography The application of gas-liquid chromatography (g.l.c.) to the analysis of hydrolysis products of a polysaccharide has provided a qualitative as well as a rapid, accurate and quantitative analysis of the constituent sugars in the polymer. In order to make g.l.c. applicable to carbohydrates, efforts have long been made to render the sugars volatile by suitable derivatization. Sweeley and colleagues (64) introduced O-trimethylsilyi groups to the sugar molecule using a very convenient reaction to provide derivatives of suitable v o l a t i l i t y . Each 29 monosaccharide may give rise to a maximum of four peaks through anomeric and ring isomerizations thus yielding a complex chromatogram. An elegant method was devised by Gunner et a l . (65) by converting a sugar into i t s alditoi acetate which has suitable v o l a t i l i t y and gives only a single peak on g . l . c , corresponding to the parent sugar. The isolation of the alditoi acetates and part i a l l y methylated alditoi acetates has been improved by Sawardeker and co-workers (67) using a copolymer of ethylene glycol succinate polyester and a n i t r i l e silicone polymer (ECNSS-M) as the liquid phase. Lindberg and colleagues have published a very informative review (68) on gas-liquid chromatography and mass spectrometry in methylation analysis of polysaccharides. An extensive review of g.l.c. analysis of methylated methyl glycosides was reported by Aspinall (69) using butane -1,4-diol succinate polyester (BDS) and the less polar polyphenyl ether (m-bis- (m-phenoxyphenoxy)-benzene) as liquid phases. In summary, g.l.c. analysis of sugar derivatives can be stated as follows: various derivatizations of the reducing sugars and their alditois by methylation, trimethylsilylation, or acetylation, or, i f necessary, a combination of these methods and subsequent separation of the derivatives using various liquid phases or a combination of liquid phases to achieve the best separation. Applications of these principles w i l l be discussed. In the analysis of the partially methylated a l d i t o i acetates of K18, the separation of the acetates of 3,4 -di-O-methyl-L-rhamnitol, 2,4-di-0-methy1-L-rhamnitol, 2,3,4,6-1e tra-0-me thy1-D-glucitol and 2,4,6-tri-O-methyl-D-galactitol (mixed with 2,6-di-O-methyl-D-glucitol) was achieved by an Apiezon L column. Further separation of the mixture 30 of 2,4,6-tri-CMnethyl-D-galactitol triacetate and 2,6-di-0-methyl-D-glucitol tetraacetate was effected on a 3% ECNSS-M column. In the case of K24 (32), the acetates of 2,4,6-tri-0-methyl- and 3,4,6-tri-0-roethyl-D-hexitols were not resolvable on columns of ECNSS-M and BDS, so that the methylated sugars were pertrimethylsilylated. Separation was then effectively achieved on ECNSS-M and BDS columns. These tri-O-raethyl-D-hexoses can also be readily separated as their acetates. The effluent^of each component separated by g.l.c. can be collected from the outlet into a capillary tube. As a rule, alditol acetates can be identified by the melting points, while the identification of part i a l l y methylated alditol acetates can be accomplished by mass spectrometry, or melting points: j f they are crystallizable. Generally, sufficient quantities of partially methylated alditol acetate can be collected from g.l.c. for mass spectrometric analysis by raethylating ca. 2mg of an oligosaccharide. While trimethyl-silylated disaccharides (70) are readily separable by g . l . c , a direct application of g.l.c. - mass spectrometry on a permethylated sodium borodeuteride reduced aldotriouronic acid has been demonstrated (29). G.l.c. and g.l.c.-mass spectrometry which are capable of analyzing a sample on a microgram scale have a great potential for the structural analysis of minute amounts of biological material containing carbohydrate, such as glycoproteins (71,72)and glycolipids. Glycoproteins are widely believed to be the receptor site on the c e l l membrane for various biological functions of the c e l l . Understanding of the structure of c e l l membrane and i t s biological function are of primary importance in the f i e l d of l i f e science. 31 10. Mass spectrometry of partially methylated alditoi acetates Of the three variations of mass spectrometer available today, i.e., electron impact (E.I.) (73), chemical ionization (G.I.) (74) and f i e l d ionization (F.I.) (75) mass spectrometers, only electron impact spectrometry which gives rise to desirable fragmentations of the sugar derivatives was employed in the present study. C.I. and F.I. mass spectrometry can produce large fragments and molecular peaks, thus fa c i l i t a t i n g the molecular weight determination. F.I. is particularly suitable for nonvolatile compounds such as free sugars. Heyns et a l . (76) reported the results of mass spectrometry of fu l l y methylated glycosides, while mass spectrometry of trimethylsilylated ethers of partially methylated glycosides was investigated by Kochetkov et a l . (77), and by Samuelson et a l . (78). Chizhov and colleagues (79) have reported their study of mass spectra of alditoi acetates and determined that primary fragments from al d i t o i acetates arise by fission between two adjacent carbon atoms in the molecule, and either fragment can carry the p o s i t i v e charge. Secondary fragments are derived from the primary fragments by elimination of acetic acid (60) and/or ketene (42). Intensity of primary fragments increased with the decreased size of the fragment. These investigators have also shown that the intensities of the primary fragmentations from the aldi t o i acetates are much lower than the intensities of the corresponding peaks from the partially methylated alditoi acetates. Therefore, the fission of the carbon-carbon bond in structures I and II takes place preferentially as compared to that in structure III. 32 I I l H—C — OCH, H-C-OCH, H-G-OCOCH, I 3 I 3 I 3 H—C—OCH, H-C-OCOCH. H-C—OCOCH. I 3 j 3 , 3 I II III Moreover, no fragments containing vicinal methoxy are observed unless there are three pairs of vicinal methoxy groups in the same molecule and two of them are cleaved. This indicates the preferred fission of structure I to II. Through systematic study of the fission patterns of partially methylated ald i t o l acetates, Lindberg (68) has made the following generalizations: 1. Derivatives with the same substitution pattern (e.g. 2,3, 4,6-tetra-O-methyl derivatives of hexitols) give very similar mass spectra, typical of that substitution pattern. 2. The base peak in almost a l l determined spectra is m/e 43 + \ (CH 3C""0). 3. Primary fragments are formed by fission between carbon atoms in the chain. Fission between a methoxylated and an acetoxylated carbon is preferred over fission between two acetoxylated carbons. The fragments with the methoxyl groups carry the positive charge because the positive ion is stabilized by the methoxyl group. The alditol acetate derived from 2-0-methyl-D-mannose therefore only gives two main primary fragments. ™ 2 _ _ _ 0 A C „ 33 333 117 CH3o — G • I-- H A c0 -1 C — H H — C 1 - 0 A c H — 1 G | - 0 A C 1 C H2GAC 4. When the molecule contains two adjacent methoxylated carbons, the fission between them is the preferred reaction and both fragments are found as positive ions. CH 0— OA-I 2 HC 0A_ 1 HC OCH3 189 131 CH3O— CH A.O-CH *» 1 CH3 5. The secondary fragments are derived from the primary fragments by single or consecutive elimination of acetic acid (60), ketene (42), methanol (32), or formaldehyde (30). The important fragments according to Lindberg's designation are li s t e d below: A. m/e 45: The lowest molecular weight primary fragment which is formed by a molecule with a methoxyl group at C I. B. m/e 59: Formed from 6-deoxyhexitols when position 5 is methylated. C. m/e 89: Obtained from the ald i t o i with positions 1 and 2 methylated. This is the exception mentioned above. This species w i l l 34 give rise to (B) after elimination of formaldehyde. D. m/e 117: Derived from an alditol when C 1 is acetylated and C 2 methylated. E. m/e 131: Obtained from a 6-deoxyhexitol when C 4 is methylated and C 5 acetylated. F l . m/e 161: Obtained in low intensities from alditols methylated at positions 2 and 3. It may become prominent when C 4 is also methylated. F2. m/e 161: Always prominent when C 1 and C 3 are methylated. The secondary fragments having m/e 129 and m/e 101 are obtained from Fl and F2 by loss of methanol (32) and acetic acid (60), respectively. Further loss of ketene (42) from m/e 129 gives m/e 87, and loss of formaldehyde (30) from m/e 101 might give m/e 71. G. m/e 175: Obtained from a 6-deoxyhexitol methylated at positions 2,3 and 4. H. m/e 189: From alditols methylated at C 3 but not C 1 and C 2. The secondary fragment m/e 129 may be formed by elimination of acetic acid, and the fragments m/e 99 and m/e 87 by further loss of formaldehyde and ketene, respectively. I. m/e 203: From a 6-deoxyhexitol, methylated at C 3 but not at positions 4 or 5. KI. m/e 205: Observed in low intensities when positions 2, 3, and 4 are methylated. When C 5 is also methylated i t w i l l be obtained in high intensities. K2. m/e 205: From alditols methylated at positions 1,2,4 and 5. K3. m/e 205: From alditols methylated at positions 1,3,4 and 5. An important peak, m/e 145, is obtained from K3 by loss of acetic acid. 35 L. m/e 233: Obtained from alditois methylated at positions 1 and 4. A prominent secondary peak, m/e 113, is obtained from (H) by loss of two molecules of acetic acid. M. m/e 261: Derived from alditois methylated at C 3 and acetylated at C 4, 5 and 6. CH 2 = 0 CH, HC = 0 - CH, + HC = 0 CH„ + HC = 0 CH, CH, H2C 0 - CH, H2C - OCOCH3 A 45 B 59 C 89 D 117 HC + 0 CH, HC = 0 CH, HC i - CH. HC - OCOCH- HC - OCH„ HC - OCOCH, CH, H2C - OCOCH3 HC - OCOCH. CH, E 131 Fl 161 I 203 + HC = 0 - CH. HC - OCH, HC - OCH, I 3 H2C - 0C0CH3 Kl 205 + 0 HC I HC - 0C0CH I HC - OCH. I 3 H2C CH, OCH, K2 205 HC = 0 - CH, I HC - OCH„ HC - OCOCH I 3 H2C - 0 - CH3 K3 205 + HC = 0 - CH, HC - OCOCH, + HC = 0 - CH, I HC - OCOCH. HC - OCOCH, I ' H2C - 0CH3 L 233 HC - OCOCH, H0C - OCOCH, M 261 On reduction some pairs of methylated sugars, e.g., a 3-0-methyl- and a 4-0-methyl-hexose, give rise to aldit o i acetates with the same substitution pattern. This can be overcome i f the reduction is carried out with sodium deuterioborate. CHDOA I HC -. OA. CHDOA HC - OA. 261 CH30 - CH 190 A„0 - CH HC - OA HC - OA. 262 HC - OCH, 189 CH 2 - OAc HC - 0A„ I H2C - 0AC This same labelling technique can be used in the reduction of carboxylic group of the uronic. acid moiety. The origin of the reduced neutral sugar can readily be distinguished by mass spectrometry. The prominent primary peaks of mass spectrometry of various methylated aldi t o i acetates are compiled in Table III. 11. Determination of anomeric configurations Optical rotations have long been employed to provide an indication of the overall anomeric configurations of a polysaccharide, although i t provides no information concerning the individual sugar residues in the polymer. However, the optical rotation of a series of oligosaccharides obtained by partial hydrolysis of a polysaccharide together with the result of specific enzymatic hydrolysis, may reveal the anomeric configuration relating to each sugar unit. Recently, use has been made of the fact that oligosaccharide aldi t o i acetates containing j8 - linkages can be oxidized by chromium trioxide in acetic anhydride 37 TABLE III. PRIMARY FRAGMENTS (a) IN THE MASS SPECTRA OF PARTIALLY METHYLATED SUGARS IN THE FORM OF THEIR ALDITOL ACETATES. i m/e Position of CH3 45 59 89 117' 131 161 Pentoses 2 (4) X 3 2,3 (3,4) X 2,4 X 2,5 X X 3,5 X X 2,3,4 X X 2,3,5 X X X Hexoses 2 (5) X 3 (4) 6 X 2,3 X 2,4 (3,5) X 2,5 X 2,6 X X 3,4 3,6 X 4,6 X X 5,6 X X 2,3,4 X X 2,3,5 X 2,3,6 X X 2,4,6 X X X 2,5,6 X X X 3,4,6 X X 2,3,4,6 X X X 2,3,5,6 X X X 6-Deoxyhexoses 2 X 3 4 X 2,3 X 2,4 X X 3,4 X 2,3,4 X X X 2,3,5 X X 3,6-Dideoxyhexo ses 2 X 2,4 X X X X X X X X X X X X X X X X (a) Primary fragments having mass numbers higher than m/e 261, which are formed by fission between a methoxyl and an acetoxyl group, are of low intensity and are not included in the Table. 38 much faster than those containing cv - linkages (80), providing a means for differentiatingoC - from jS - configurations. The non-destructive technique of p.m.r. spectroscopy (27,50) has been applied to oligo- as well as polysaccharides to obtain definitive information on the anomeric configuration of each sugar residue, and i t was extensively used in this study. Generally, the proton at the axial position of C. 1 in an aldppyranoside^ (equivalent toj8 -D-1inkage) resonates at higher f i e l d c than the proton at equatorial position (50,51) (equivalent totx -D-linkage). Both types of anomeric proton have larger chemical shifts (S*) than the rest of annular protons. Furthermore, spin-spin coupling between the protons at C 1 and C 2 can provide useful information. Thus, in a trans-diaxial interaction as in the case of ft - D-glucosyl and >5-D-galactosyl linkages, ^ i s about 7-9 Hz, while J± 2 i s about 1-3 Hz for CX -D-glucosyl and d acetate from D-glucose pentaacetate (Rf 0.56). Faint yet distinct spots from pyruvic acid as compared to intensive acetic acid spot indicated the pyruvic acid in the poly-saccharide was mostly intact under the methanolysis conditions. Proton Magnetic Resonance Spectra of Original arid De-O-acetylated Polysaccharides. " Capsular polysaccharide (ca. 20mg) was dissolved in D20 and freeze-dried. After two further treatments, the 100 MHz spectrum was run at 90° and showed sharp singlets at T^8.50 (pyruvate) and 7.93 (acetate) together with signals at7T5.50 (2H) and 5.15 (IH). The spectrum of deacetylated material was similar with the absence of the peak a t t 7.93. The 60 MHz spectra were run on deacetylated polysaccharide at pH 1 and 11, showing the chemical shifts of pyruvate CH^ at?" 8.50 and 8.55, respectively. Methylation of Original Capsular Polysaccharide To vacuum dried polysaccharide (400mg) in anhydrous dimethyl 57 sulfoxide (20ml) was added an excess of methylsulfinyl carbanion prepared from l g sodium hydride (55% o i l dispersion) according to the method of Conrad and Sandford (15). The flask was sealed with a serum cap and shaken for 12h. An excess of methyl iodide (4ml) was added slowly to the flask while sti r r i n g and keeping the temperature below 20°. The reaction mixture was dialyzed against tap water for 24h and extracted with chloroform (2 X 30ml). The chloroform layer was dried over anhydrous sodium sulfate and evaporated to dryness, giving 100 mg of methylated polysaccharide which showed no hydroxyl absorption in the IR spectrum. The aqueous layer was freeze-dried (300mg) and showed no hydroxyl absorption in IR after three Purdie methylations. The p.m.r. spectrum (chloroform-d) of the methylated polysaccharide showed ^8.5 (3 H, pyruvate CH 3 ) 6-7(40H, calcd. 43H for ring protons and 0CH 3»s), and T~5.55 and 5.44 (3 H, anomeric protons). The methylated polysaccharide (300mg) was digested in sulfuric acid (72%, 3ml) with ice cooling for 2h then diluted with water (24ml) and refluxed for 4h. The cooled hydrolyzate was neutralized with barium carbonate and the supernatant was collected after centrifugation. The precipitate was washed twice with a small amount of water. The combined supernatant and washings were concentrated to a small volume, filter e d , passed through successive columns of Amberlite 1R-120 (IT1") resin and Duolite A-4(0H") resin. The solution was evaporated to dryness, giving 178mg of neutral sugars. Acidic sugars were recovered by eluting the Duolite A-4(0H") resin column with 10% formic acid which was then evaporated to dryness, giving 60mg of an acidic mixture. 58 The neutral sugars were chromatographed using solvent D. The chromatogram showed two spots: the fast moving spot (R f 0.55) was identical to 2,3,6-tri-O-methyl-D-glucose in mobility and characteristic color response, and the slower running spot (R^ 0.07) was in the region of monomethylated sugars. To the water solution of neutral sugars (lOmg) sodium borohydride (15mg) was added. The reduction was allowed to proceed for 24h, then worked up as usual. The methylated alditois were acetylated and the methylated alditoi acetates were dissolved in a small volume of chloroform and injected onto column a at 180° (helium flow rate 88 ml/min) giving peaks corresponding to the alditoi acetates of 2,3,6-tri-0-methyl-D-glucose (6.4 min) and a monomethylated sugar (18 min) in 1:1.2 ratio. Samples were collected and mass spectra were taken. The component at R^ 6.4 min gave a fragmentation pattern identical to that from the alditoi acetate of 2,3,6-tri-O-methyl-D-glucose while the second component (Rj, 18 min) gave a fragmentation pattern similar to the a l d i t o i acetate of 2-0-methylaldoses. The neutral sugars (170mg) was separated on f i l t e r paper in solvent D. Guide strips cut from the chromatogram were developed with p-anisidine trichloroacetate and the corresponding sections of the two components were cut out from the paper chromatogram, eluted with water, and the water solutions were concentrated to dryness. The component at R^ 0.55 (58rag) was crystallized twice from ethyl ether, giving m.p. 119-121°, mixed m.p. 117-119° with authentic 2,3,6-tri-O-methyl-D-glucose. The second component (R^ 0.07) was a syrup (40mg) which was seeded with authentic 2-0-methyl-D-mannose. The crystals were collected 59 after removal of the surrounding syrup by cold ethanol. The ethanol solution was concentrated, seeded and cooled to give clusters of o platelets. The crystals were recrystallized from ethanol, m.p. 139-140 , mixed m.p. 138-140° with the authentic 2-0-methyl-D-mannose. The acidic mixture (5mg), eluted from Duolite A-4(0H~) resin, was refluxed .for ,6h with methanolic hydrogen chl o r i d e ( 3%, 3ml ). neutralization (silver carbonate) and centrifugation, the methanol solution was concentrated to dryness. Injection in chloroform, onto column b at 170° and using a helium flow rate of 75 ml/min, gave characteristic peaks corresponding to methyl 2,3,6-tri-0-methyl-D-gluco-pyranoside (4.4 min (m), 6.0 min (s) ) and methyl 2,3-di-0-methyl-D-glucuronic acid methyl ester (11.5 min (m), and 13.5 min (s) ). A. portion of the acidic mixture (5mg) was refluxed with methanolic hydrochloride (3%, 5ml) for 4h. The cooled reaction mixture was neutralized with silver carbonate and centrifuged. The combined centrifugate and washings were evaporated to dryness and treated with sodium borohydride (lOmg in 10 ml water). The reduction was allowed to proceed for 24h. The excess sodium borohydride was destroyed by adding dilute acetic acid and the solution was passed through a column of Amberlite IR-120 (H +) resin, then evaporated to dryness. The residue was dissolved in trifluoroacetic acid (2M, 1ml) and the solution was heated on a steam bath for 4h. The reaction mixture was evaporated to dryness. The hydrolyzate was examined by paper chromatography in solvent D, giving, two spots corresponding to 2,3,6-tri-O-methyl-D-glucose (R f 0.56) and 2,3-di-0-methyl-D-glucose (Rf 0.28). The same hydrolyzate was further treated with sodium borohydride (lOmg in 0.5ml water) and worked up as usual. Injection in chloroform, onto column a 60 at 200° and with a helium flow rate of 86 ml/min, gave two peaks i d e n t i c a l i n retention time to the a l d i t o l acetates of 2,3,6-tri-O-methyl-D-glucose (6.1 min) and 2,3-di-0-methyl-D-glucose (11.9 min). Samples' were c o l l e c t e d and the mass spectra were run and the fragmentation patterns were i d e n t i c a l , to that of a l d i t o l acetates of 2,3,6-tri^-methyl-D-glucose and 2,3-di-^-methyl-D-glucose, r e s p e c t i v e l y . A solution of methylated polysaccharide (0.17 g) in anhydrous tetrahydrofuran (20 ml) was added dropwise with s t i r r i n g to a suspension of lithium aluminum hydride (0.3 g) in tetrahydrofuran (30 ml) at room temperature. The reaction mixture was s t i r r e d for 18h more, then the excess lithium aluminum hydride was destroyed by cautious addition of a c e t i c a c i d . The whole reaction mixture was evaporated to dryness and the residue was extracted with b o i l i n g chloroform (3 X 30ml). The combined extracts were washed with water, dried over anhydrous sodium su l f a t e , then evaporated to dryness, giving 0.13g of reduced methylated polysaccharide. The p.m.r. spectrum (chloroformrd) showed the signals atX8.6 (pyruvate CH3,3H),6-7 (rin g protons and OCH^ protons, 40 H), 5.3-5.6 (anomeric protons, 3H). Two methyl esters at"c"6.2-6.4 were absent. The methylated carboxyl-reduced polysaccharide (56mg) was digested i n formic a c i d (907«) at 97° for l h , concentrated to dryness, and then hydrolyzed i n 2M t r i f l u o r o a c e t i c acid on a steam bath for 4h to give, by paper chromatography in solvent D, three spots of s i m i l a r i n t e n s i t y and m o b i l i t i e s i d e n t i c a l to 2-C|-methyl-D-mannose (Rf 0.10), 2,3-di-0-methy 1-D-glucose (R f 0.30), and 2,3,6-tri-0-methyl-D-glucose (Rf 0.58). A portion of the hydrolyzate (36mg) was chromatographicaHy 61 separated in solvent D. The guide strips were cut out from the chromatogram and developed by p-anisidine spray. The areas corresponding to the three components were cut out and eluted with water. Each fraction was concentrated to a syrup. The fraction from Rf 0.10 (6mg) was crystallized from ethanol, m.p. 137-139°, undepressed/ with authentic 2-0-methyl-D-mannose. The fraction from Rf 0.58 (9.7mg) was crystallized from ethyl ether, m.p. 113-115°, undepressed with authentic 2,3,6-tri-0-methyl-D-glucose. The fraction from Rf 0.30 (lOmg) had the same mobility as 2,3-di-O-methyl-D-glucose. The ald i t o l acetate was prepared and i t s retention time in g.l.c. analysis and the mass spectrum were both identical to the alditol acetate of authentic 2,3-di-O-methyl-D-glucose. Another portion of the hydrolyzate (lOmg) was reduced with sodium borohydride. After the usual work up the methylated alditols were acetylated and injected in chloroform onto column a programmed from 160 to 190° at 2°/min. Using a helium flow rate of 86 ml/min three peaks corresponding to the alditol acetates of 2,3,6-tri-0-methyl-D-glucose (16.0 min), 2,3-di-0-methyl-D-glucose (22.4 min), and 2-0-methyl-D-mannose (26.1 min) were obtained in a ratio of 1:1:1. Methylation of degraded polysaccharide A degraded polysaccharide with the pyruvic acid group removed was prepared by autohydrolysis of the original polysaccharide (0.5g) in 60ml aqueous solution-(pH 2.5) in a *'steam bath for 20h. The hydrolyzate was dialyzed against d i s t i l l e d water (2 X 300ml);T The:material in the dialyzing tubing was freeze-dried, giving degraded polysaccharide (0.25g) which did not show the pyruvate CH- in i t s p.miri^spectrum. 62 To this polysaccharide (200mg in 10 ml dimethyl sulfoxide) methylsulfinyl carbanion prepared from 0.5g sodium hydride was added at a temperature below 20°. The mixture was shaken overnight, and an excess of methyl iodide (3ml) was added while the reaction mixture was kept below 20°. The mixture was dialyzed against tap water for 24h, then the material in the dialyzing bag was freeze-dried. The freeze-dried methylated polysaccharide (170mg) was made anhydrous by evaporation to dryness with a small volume of benzene, and showed a small hydroxyl absorption band in the IR spectrum (in carbon tetrachloride). It was thus methylated by the method of Purdie to a permethylated polysaccharide without absorption at 3500 era"*. A portion of the methylated polysaccharide (30mg) was hydrolyzed with trifluoroacetic acid (2M, 8ml) on a steam bath for 8h. The solution was evaporated to dryness and the residue was dissolved in a small volume of water and examined chromatographically in solvent D. The chromatogram showed a major spot corresponding to trimethyl hexoses (Rf 0.55) and a faint spot corresponding to tetramethyl hexoses (Rf 0.78). The remaining hydrolyzate was treated with sodium borohydride (40mg) and the reaction mixture was allowed to stand at room temperature for 24h and worked up as usual. The partially methylated alditois were acetylated and injected in chloroform onto column a, programmed from 170 to 200°at 2°/min with a helium flow rate of 67 ml/min, giving two major peaks with retention times identical to those of the alditoi acetates of 2,4,6-tri-0-methyl-D-mannose (16.1 min) and 2,3,6-tri-O-methyl-D-glucose (18.8 min) along with a small peak corresponding to the alditoi acetate of 2,3,4,6-tetra-O-raethyl-D-glucose (9.3 min). 63 Samples were collected and the fraction at 16.1min was crystallized from petroleum ether (30-60°), giving crystals of m.p. 64-66°, undepressed with the authentic alditol acetate of 2,4,6-tri-0-methyl-D-mannose. Partial Hydrolysis of Capsular Polysaccharide Acidic polysaccharide (3.5g) was dissolved in water (300ml, i n i t i a l pH2.7), heated on a steam bath for 12h, and the solution was dialyzed against d i s t i l l e d water (2 X 600 ml). The non-dialyzable material was rehydrolyzed and the dialyzates were, combined and concentrated. Only a small amount of product was obtained and chromatography in solvent A revealed pyruvic acid (Rf 0.7), D-glucose, D-mannose, and two components having R^an 1.2 and 1.7 with traces of other compounds. The solution containing the non-dialyzable material was concentrated to 150ml, heated on a steam bath for 24h, and dialyzed against d i s t i l l e d water (2 X 500 ml). The process was repeated for a total of 148 h heating with periodic concentration to keep the volume below 300ml. The dialyzates were combined, concentrated, and passed through successive columns of Amberlite IR-120 (H +) and Duolite A-4 (OH") resins. The neutral components were fractionated by preparative paper chromatography in solvent B to give (30mg), ^ (20mg) and (15mg, R_, 0.36), and the components eluted from Duolite A-4 (OH") column —klc with 107. formic acid were separated on paper in solvent C to give Aj-A^. The neutral compounds and N 2 were chromatographically identical to D-mannose and D-glucose and were not further examined. The oligosaccharides were analyzed as follows. 64 Structural study of compound N3 Compound crystallized from water and had m.p. 164-166°, (°9D .7.3° (c 2.18, water) which agrees with the value (m.p. 165°) given for 3-0-^-D-glucopyranosyl-D-mannose (89). Compound N-j was hydrolyzed in t r i f luoroacetic acid (2M) at 100° for 6h. The hydrolyzate was evaporated to dryness and the residue was dissolved in a few drops of water and chromatographed in solvent B, giving two spots identical to the mobilities of D-mannose and D-glucose. Another portion of was reduced with sodium borohydride in water, worked up as usual then hydrolyzed in trifluoroacetic acid (2M) for 4h at 100°. The hydrolyzate was evaporated to dryness then dissolved in a few drops of water and examined on a duplicate f i l t e r paper using solvent E. One of the chromatograms was developed with silver nitrate-sodium hydroxide, revealing two spots with the mobilities identical to D-glucose and D-mannitol (R^^ 1.07). The other chromatogram was sprayed with p_-anisidine trichloroacetate, revealing only the spot equivalent to D-glucose. Compound N3 (5mg) was methylated according to the method of Hakomori , then hydrolyzed (2M trifluoroacetic acid, 100°, 4h) and concentrated to dryness. The residue was dissolved in a few drops of water and chromatogrammed in solvent D, revealing two spots of similar intensity and mobility identical to 2,3,4,6-tetra-O-methyi-D-glucose (R f 0.78) and 2,4,6-tri-O-methyl-D-mannose (Rf 0.56). The rest of the hydrolyzate was reduced with sodium borohydride (8mg) and acetylated with acetic anhydride-pyridine (1:1) in the usual way. Injection of the partially methylated a l d i t o i acetates in chloroform onto column a at 185° and using a helium flow rate of 60 ml/min, gave two peaks, in 1:1 ratio identical in retention times to the alditol acetates of 2,3,4,6-tetra-O-methyl-D-glucose (6.5 min) and 2,4,6-tri-0-methy1-D -mannose (13.1 min). The p.m.r. spectrum of the original N-j in Dr>0 was run after two deuterium-exchanges, showing signals at "Z'5.46 (J^ 2 6.0 Hz, IH), 5.l6(0.3H)and 4.81 ( J t 2 1Hz, 0.7H). The structure of N.j is therefore 3-0-^ -D-glucopyranosyl-D-mannose. Structural study of A^ Compound A^ (58mg, R^ c^ 0.23 solvent A) had an equivalent weight of 394 and£ 5.36 (7) 0.6 J$ - Gal-OH 5.20 (7) 1 £ - GlcA U L Man-OH 4.71 I 1 3 Cx - Man Gal-OH 4.58 (3) 1 1 2 CX - Glc GlcA-OH 4 1 2 1 2 1 3 I G l c - ^ - GlcA Man Gal ~g~ 4.58 1 cx 3 1 4.81 1 Three CX-linkages Man 4.85 1 5.35 (5) 1 e i-5.49 (7) 1 Two jQ-linkages, * Spectra r u n in DoO with external tetramethvlsilane ( T = 10) at 100 M H_. 82 of the glycosidic linkages to be assigned. The structures given were determined by 1) hydrolysis with acid and enzyme, 2) progressive partial hydrolysis and 3) methylation as described in the experimental. Capsular polysaccharide was methylated (15,17), hydrolyzed and the partially methylated sugars were examined. 2,3,4,6-Tetra-O-methyl-D-mannose, 2,3,6-tri-0-methyl-D-glucose, 2,4,6-tri-0-methyl-D-galactose, 4,6-di-0-methyl-D-mannose were found in a molar ratio of 1:1:1:1 which was determined by g.l.c. of the al d i t o i acetates (92). The acidic component was found to be 3,4-di-0-methyl-D-glucuronic acid as identified by the derived 3,4-di-0-methyl-D-glucose. The analysis of methylated, carboxyl reduced polysaccharide showed that 3,4-di-O-methyl-D-glucose was present in equimoiar ratio to the other sugar moieties, A simple chromatographic differentiation on paper of 2,3,4,6-tetra-Onmethyl-D-mannose and 2,3,4,6-tetra-0-methyl-D-glucose was devised by making the chromatographic tank unsaturated with solvent at the beginning of chromatography. The presence of tetramethylated D-mannose indicates that D-mannose is in the side chain and the branch point i s another D-mannose since a dimethyl-D-mannose was the only dimethylated sugar obtained. The exact sequential arrangement is established by the study of the oligosaccharides. The identities of the individual partially methylated sugars were determined by comparison with standards using a) paper chromatography, 2) g.l.c. of the methyl glycosides, c) g.l.c. - m.s. of the alditoi acetates, d) m.p. of crystalline compounds, and e) demethylation to the parent sugar aldi t o i acetate in the case of tetramethyl-D-mannose. The methylation data in conjunction with the results of analysis of the acidic oligosaccharides enable the repeat unit of Klebsiella K 62 capsular polysaccharide to be written as: 4 1 2 1 2 1 3 1 D-Glcp D-GlcAp D-Manp D-Galp * 1 .D-Manp Confirmation of this structure was sought by the Smith degradation procedure. The polysaccharide was oxidized with periodate and 0.82 mole per hexose unit was consumed. The derived polyalcohol was totally hydrolyzed to give D-mannose, D-galactose, erythritol and glycerol in a molar ratio of 1:1:1:1. Erythritol i s derived from D-glucose and glycerol from D-mannose in the side chain (Figure 3). Mild acid hydrolysis by the acidification of the polyol with Amberlite IR-120 (H +) followed by evaporation to dryness at room temperature yielded, by gel f i l t r a t i o n , glycerol and a disaccharide glycoside composed of D-mannose, D-galactose and erythritol. The evidence presented shows clearly that the structure of the capsular polysaccharide of Klebsiella K-type 62 is as given above. 84 EXPERIMENTAL Isolation and Properties of K62 Capsular Polysaccharide Klebsiella K62 (5711-52) was grown in the medium as for K5 and harvested to give 400ml of slime and c e l l s from 41 of agar medium. Phenol (5g) in water (100ml) was added to the slime and the mixture was diluted to 1.5 1 with water. Centrifugation of the diluted slime at 27,000 r.p.m. (68,000g) for 1 h yielded a clear light yellow supernatant which was treated with Cetavlon (107., 25ml) to precipitate the acidic polysaccharide. The neutral polysaccharide remaining in the supernatant was dialyzed and freeze-dried, yielding 0.7g. The precipitated acidic polysaccharide was dissolved in sodium chloride (2M), then precipitated in five volumes of ethanol. The precipitate, collected by centrifugation, was dissolved in water, decationized, dialyzed against d i s t i l l e d water, then freeze-dried to give 3.7g of pure acidic polysaccharide. The electrophoretic mobility on cellulose acetate was 2.4cm for 30 min at pH 8.8 and 300 V. Ash content n i l : nitrogen n i l . This product, after methylation, had (<*)D +59.7° (c 2.3, carbon tetrachloride). The equivalent weight of this polysaccharide was found to be 806 by titration with 0.01M sodium hydroxide (phenolphthalein). The p.m.r. spectrum of a 27. solution of the original polysaccharide in D20 at 95° showed signals of anomeric protons at .T5.49 (IH), 5.35 (IH), 4.85 (IH), 4.81 (IH) and 4.58 (IH). A mildly degraded K62 polysaccharide was obtained by heating the original polysaccharide with trifluoroacetic acid (0.2M) at 97° for 30 min, followed by dialyzing against tap water for 24 h. The material in the dialysis tubing was freeze-dried, and i t s p.m.r. spectrum in D^ O gave signals of anomeric protons similar to that 85 of original polysaccharide with an improved resolution: f 5.49 (J 7Hz, IH), 5.35 (J, 5Hz, IH), 4.85 (IH), 4.81 (IH) and 4.58 (IH) (see Table IV). Analysis of Constituent Sugars The capsular polysaccharide (4mg) was hydrolyzed with trifluoroacetic acid (2M) at 100° for 9h. After evaporation, the hydrolyzate was examined by paper chromatography in solvent B revealing three major spots for D-mannose, D-glucose, D-galactose and an acidic component at the origin. Paper chromatography in solvent A showed the presence of D-glucuronic acid, and aldobiouronic acid (RQ^ 0.45) and a trace of aldotriouronic acid ( R ^ j ^ 0.16), in addition to the neutral hexoses. The polysaccharide (lOmg) was heated in hydrochloric acid (2M) on a steam bath for 8h. The solution was passed through a Duolite A-4 (OH") column to neutralize hydrochloric acid and to remove the acidic sugars. After evaporation, the neutral fraction was analyzed and found to contain D-mannose, D-galactose and D-glucose in an approximate 2:1:1 ratio as determined by the g.l.c. of the alditoi acetates using column a. The individual components were collected and identified as D-mannitol hexaacetate, m.p. 119-121°, galactitol hexaacetate, m.p. 161-163° and D-glucitol hexaacetate, m.p. 93-95°. The configuration of mannose and glucose were found by dissolving the mannitol hexaacetate and glucitol hexaacetate collected from g.l.c. in acetonitrile. The positive circular dichroism curves at 213 my , identical to those given by authentic samples, confirmed the D-configurations of mannose and glucose. A sample of polysaccharide (lOmg) was refluxed in methanolic 86 hydrochloride (3%) for lOh on a steam bath. After neutralization (s i l v e r carbonate), reduction (sodium borohydride) in methanol, the reaction mixture was neutralized with Amberlite 1R-120 (H+) cation exchange resin and d i s t i l l e d three times with methanol. The residue was further hydrolyzed (2M trifluoroacetic acid), reduced and acetylated. The acetylated mixture was found by g.l.c. analysis on column a to contain D-mannitol hexaacetate (m.p. and mixed m.p. 121-123°), galactitol hexaacetate (m.p. and mixed m.p. 162-164°), and p_-glucitol hexaacetate (m.p. and mixed m.p. 95-97°) in an approximate ratio of 2:1:2. Methylation of the Capsular Polysaccharide Dry polysaccharide (Ig) was dissolved in anhydrous dimethyl sulfoxide (40ml). To this solution methyisulfinyl anion (25ml, 3M) was added and the mixture was stirred for 12h. Methyl iodide (8ml) was added slowly while s t i r r i n g and keeping the temperature below 20°. After 0.5h the solution was dialyzed against running water for 24h then extracted with chloroform (3 X 100ml). The chloroform extract was dried over anhydrous sodium sulfate followed by evaporating the chloroform solution to dryness to give 0.9g of methylated polysaccharide which showed absorption of hydroxyl groups at 3600 cm"^. The incompletely methylated polysaccharide was dissolved in 8ral of N, N-dimethyl formamide and 20ml of methyl iodide. While stirr i n g , silver oxide (5g) was added. Another portion of silver oxide (5g) was added lOh later and the sti r r i n g was continued for 12h more. The reaction mixture was centrifuged and the solid was washed with chloroform. The combined centrifugate and washings were washed with water (2 X 10ml) 87 and the organic layer was dried and evaporated to dryness. The residue was methylated with Purdie's reagents to give a product showing no absorption at 3600 cm"l,(c*)D +59.7° (c 2.3, carbon tetrachloride). Methylated polysaccharide (O.lg) was heated at 100° in formic acid (90%) for lh, and the solution was evaporated to dryness. The residue was subsequently dissolved in hydrochloric acid (2M) and heated for 6h.AAf.ter evaporation to dryness,, analysis of the hydrolyzate by-paper chromatography (solvent D and F), showed the presence of 2,3,4,6-tetra-0-methyl-D-mannose (R f 0.80, solvent D; 0.82, solvent F), 2,3,6-tri-O-methyl-D-glucose (R f 0.55, D; 0.72, F), 2,4,6-tri-O-methyl-D-galactose (R^ 0.40, D; 0.63, F), 4,6-di-0-methyl-D-mannose (R f 0.28, D; 0.56, F), and an acidic component (origin, D; R^ 0.29, F). Separation on anion-exchange resin (Duolite A-4) gave a neutral fraction which showed on paper chromatography components corresponding to 2,3,4,6-tetra-0-methyl-D-mannose, 2,3,6-tri-O-methyl-D-glucose, 2,4,6-tri-0-methyl-D-galactose and 4,6-di-0-methyl-D-mannose. The neutral fraction was reduced with sodium borohydride, and after removal of the borate, the material was acetylated in a sealed tube by using acetic anhydride and pyridine (1:1, 1 ml, 97°, 20min). After removal of the reagents, the residue was dissolved in a small volume of methanol and analyzed by g.l.c. on column a (195°, helium flow rate 71 ml/min) and column d (212°, helium flow rate 97 ml/min). Four peaks in the ratio of 1:1:1:1 were observed. They were identified by their retention times and mass spectra as acetates of 2,3,4,6-tetra-0-methyl-D-mannitol (8.6 min column 25.5 min, column d), 2,3,6-tri-0-methyl-D-glucitol (20.5 min, a; 32.7 min 2,4,6-tri-O-methyl-D-galactitol (18.4 min, a; 36.6 min, d), and 4,6-di-0-methyl-D-mannitol (26.1 min, a; 39.7 min, d). 88 A separate portion of the methylated polysaccharide (0.7g) was heated in formic acid (90%) on a steam bath for 40 min. After evaporation to dryness, the mixture was further hydrolyzed in trifluoro-acetic acid (2M) on a steam bath for 3h. The hydrolyzate, after removal of trifluoroacetic acid by evaporation, was separated into a neutral fraction (0.42g) and an acidic fraction (0.21g) using Duolite A-4 (OH-) ion exchange resin. The neutral fraction was separated into four components by preparative paper chromatography using solvent D. Fraction I (51mg) had chromatographic mobility in solvent D identical to 2,3,4,6-tetra-O-methyt-D-mannose, and could be differentiated from 2,3,4,6-tetra-O-methyl-D-glucose in solvent D i f the chromatographic tank were i n i t i a l l y not saturated with respect to the solvent system and with several drops of ammonium hydroxide on the bottom of the tank instead of using a tank saturated with the solvent. Under these special conditions i t took longer (ca. lOh) to develop the chromatogram, however, i t provided a ready separation of 2,3,4,6-tetra-OHnethyl-D-mannose and 2,3,4,6-tetra-0-methyl-D-glucose into two distinct spots with a ratio of mobility Me4 Man/Me^ Glc = ca. 0.94. A portion of this fraction (5mg) was methanolyzed with 3% methanolic hydrochloride, and after neutralization (silver carbonate) and concentration, the product was examined on column a, at 150° and a helium flow rate of 60 ml/min. Two peaks were obtained in approximately 10:1 ratio, identical in ratio and retention times to methyl 2,3,4,6-tetra-O-methyl-tf-D-mannoside (20.9 min) and methyl 2,3,4,6-tetra-O-methyl-^-D-mannoside (27.5 min) obtained by methanolysis of an authentic sample of the sugar. 89 A separate portion of fraction I (5mg) was dissolved in methylene chloride (lml) and chilled at -75°. Boron trichloride (2ml) withdrawn from a cylinder while chilled at -75° was added to the methylene chloride solution and the reaction mixture was kept at that temperature for 30 min, then allowed to stand at room temperature for I6h. After evaporation and d i s t i l l a t i o n s with methanol, the de-0-methylated product was reduced o o and acetylated and separated on column a programmed from 185-120 at 1 /min, with a helium flow rate of 100 ml/rain, to give a peak identical to the authentic D-mannitol hexaacetate (18 min). A sample was collected and the m.p. and mixed m.p. with D-mannitol hexaacetate was 120-122°. Fraction II (30mg) had the same chromatographic mobility as 2,3,6-tri-O-methyl-D-glucose (R f 0.55, solvent D). The m.p. was 115-117° (from ethyl ether) and undepressed with an authentic sample of 2,3,6-tri-0-methyl-D-glucose. Fraction III (32mg) had the same chromatographic mobilitiy as 2,4,6-tri-0-methyl-D-galactose (R^ 0.40, solvent D), and was crystallized from chloroform and recrystallized from ethyl ether to give crystals of m.p. 103-105°, undepressed with an authentic sample of 2,4,6-tri-0-methyl-D-galactose. The c.d. spectrum of 2,4,6-tri-0-methyl-galactitol acetate was identical to that of a standard sample of 2 , 4 , 6 - t r i -O-methyl-D-galactitol confirming the D-configuration of galactose. Fraction IV had the same chromatographic mobility as 4,6-di-0-methy1-D-mannose (R^ 0.28, solvent D). Attempted crystallization of fraction IV was not successful and i t was subsequently reduced, acetylated, and examined on column a (195°, helium flow rate 71 ml/min) to give a component identical in retention time (26.1 min) and mass spectrum to that of the alditol acetate of authentic 4,6-di-O-methyl-D-mannose. 90 Analysis of the acidic mixture A portion of the acidic fraction (20mg) was hydrolyzed in hydrochloric acid (2M) on a steam bath for 8h. After evaporation, the hydrolyzate was shown by paper chromatography to contain 4,6-di-O-methyl-D-mannose (Rf 0.28, solvent D) as the major component, together with small amounts of 2,3,6-tri-O-methyl-D-glucose (R f 0.55, solvent D) and 2,4,6-tri-0-methyl-D-galactose (Rf 0.40, solvent D) and having acidic components at the origin. A separate portion of the acidic fraction (lOOmg) was heated under reflux overnight with 3% methanolic hydrochloride (25ml) to give the ester glycosides which were reduced by refluxing with lithium aluminum hydride (0.5g) in tetrahydrofuran (15ml) for 4h. The excess lithium aluminum hydride was destroyed by adding ethyl acetate, evaporated to dryness, and extracted with chloroform. The chloroform extract was concentrated and the residue was hydrolyzed with triftuoro-acetic acid (2M). The hydrolyzate was examined by paper chromatography to show 4,6-di-O-methyl-D-mannose (Rf 0.28, solvent D) overlapped with another component, 2,3,6-tri-O-methyl-D-glucose (R^ 0.55, solvent D), and 2,4,6-tri-O-methyl-D-galactose (R f 0.40, solvent D). No acidic components were observed at the origin. The hydrolyzate was subsequently o reduced, acetylated, and examined by g.l.c. on column a (195 , helium flow rate 71 ml/min) to show peaks identical in retention times to 2,3, 4,6-tetra-0-methyl-D-mannose (8.6 min, trace), 2,3,6-tri-O-methyl-D-glucose (20.5 min), 2,4,6-tri-0-methyl-D-galactose (18.4 min), 4,6-di-O-methyl-D-mannose (26.1 min), as well as a peak at 42.1 min which was derived from the D-glucuronic acid moiety in the polysaccharide. The mass spectrum of the last peak showed the fragmentation pattern 91 characteristic of an ald i t o l acetate of a 3,4-di-0-methylhexose. Therefore the last component was 3,4-di-O-methyl-D-glucitoi acetate which reflected the fact that D-glucuronic acid in the polysaccharide was connected at position 2. Reduction of the Methylated Polysaccharide A portion of the methylated K62 (O.lg) was dissolved in tetrahydrofuran (30ml). To this solution lithium aluminum hydride (0.l6g) in tetrahydrofuran (8ml) was added,and the reaction was stirred overnight. The excess lithium aluminum hydride was destroyed by careful addition of ethyl acetate. After removal of the solvents by evaporation the residue was extracted with chloroform and the organic solution was evaporated to dryness, yielding 0.09g. The reduced product was heated with formic acid (90%) on a % steam bath for lh.- After concentration, the residue was further hydrolyzed with tfifluoroacetic acid (2M) for 4h. Quantitative analysis of the derived alditol acetates by g.l.c. on column a (195°, helium flow rate 71 ml/min) and column d (212°, helium flow rate 97 ml/mirt)showedthe p r e s e n c e o f i t h e a l d i t o l a c e t a t e s o f 2,3 , 4 , 6 - t e t r a - 0 - m a t h y1 - D - m a n n o s e ( 8 . 6 m i n , a ; 25.5 min, d), 2,3,6-tri-O-methyl-D-glucose (20.5 min, a; 32.7 min, d), 2,4,6-tri-0-methyl-D-galactose (18.4 min, a; 36.6 min, d), 4,6-di-0-methyI-D-mannose (26.1 min, a; 39.7 min, d), and 3,4-di-O-methyl-D-glucose (42.1 rain, a; 48.8 min, d) in a molar ratio of 1:1:1:1:1. The identity of the methylated sugars was confirmed by mass spectrometry of the derived alditol acetates. 9 2 Partial Hydrolysis of Capsular Polysaccharide Autohydroiysis of the capsular polysaccharide was carried out at pH 3.5 and 100° up to 37h. D-Mannose as the only neutral monosaccharide and traces of oligosaccharides were found by paper chromatography in solvents A and; C. Polysaccharide (0.4g) was hydrolyzed with trifluoroacetic acid (IM) on a steam bath for 5h. After evaporation, the hydrolyzate was found, by paper chromatography in solvent A, to contain D-mannose, D-glucose, D-galactose, D-glucuronic acid, and a component with the mobility of an aldobiouronic acid (15^ 0.37). The hydrolyzate was separated into neutral and acidic fractions using ion-exchange resins (Amberlite IR-120 (H +) and Duolite A-4 (0H~) ). The acidic fraction (iOOmg) was applied to a Sephadex G 15 column (110 X 2 cm) which was irrigated with water at a flow rate of ca. 5 ml/h. The eluent was collected ca. 2ml each tube and the content in each tube was examined on paper in solvent C. The component corresponding to an aldobiouronic acid, A 2, (60mg) was collected from fractions about the elution volume of 140ml, followed by fractions containing the mixture of an aldobiouronic acid and D-glucuronic acid, and fi n a l l y followed by fractions containing only D-glucuronic acid (lOmg). The glucuronic acid was evaporated with the addition of a few drops of concentrated hydrochloric acid. The c.d. curves of the isolated D-glucuronic acid and that of a standard were both positive (^£ H2° +2.82 and +3.32, respectively). 219 In order to obtain oligosaccharides higher than the aldobiouronic acid (A 2), a milder hydrolysis condition was used by heating the polysaccharide (0.8g) in t r i f luoroacetic acid CW for lh. The solution was concentrated 93 and evaporated three times with water and fractionated using a charcoal-Celite column (a mixture of active carbon (30g) and Gelite (30g) was poured into a Buchner funnel (10cm i.d.) to make a column of 1.4 cm thickness). Monosaccharides were eluted by d i s t i l l e d water (2.8 1) and a series of oligosaccharides was eluted with 207. aqueous ethanol (2 1) and 107. aqueous isopropanol (500ml); yield of oligosaccharides about 0.34g. The mixture of oligosaccharides was separated (lOOmg per run) on a Sephadex G 15 column (110 X 2cm) into four fractions (J^j 0.46, 0.19, 0.12 and 0.06, solvent C), namely di-(A 2, 4mg), t r i - ( A 3 , 52mg), tetra- and pentasaccharides (50mg), and Ag (31mg). A successful separation of the third fraction into tetrasaccharide (A^ 28mg) and A 5 (13rag) was achieved by a prolonged paper chromatography (7 days) in solvent C. The oligosaccharides were analyzed as follows. Structural study of Aldobiouronic Acid (A 2) The aldobiouronic acid was found to have |e*)D-32.0° (c 6.37, H 20). (a) Constituent sugars ""' A portion (15mg) of A was hydrolyzed with hydrochloric acid (2M) at.97° for 8h, then evaporated to dryness. Paper chromatography in solvent B showed D-mannose as the only neutral sugar present, whereas the presence of D-mannose, D-glucuronic acid, and D-giucuronolactone was demonstrated by using solvent A. The rest of the hydrolyzate was dissolved in 37. methanolic hydrochloride (lOmg) and the solution was refluxed for 6h. The solution was neutralized (silver carbonate) centrifuged, and evaporated. The residue was dissolved in methanol and reduced with sodium borohydride. The solution was l e f t overnight at 94 room temperature and was then carefully neutralized and decationized by passage through an Amberlite IR-120 (H+) column. The residue, after removal of borate, was hydrolyzed with trifluoroacetic acid (IM) at 97° for 4h and shown in solvent A to give D-mannose and D-glucose. The mixture of neutral sugars was reduced, acetylated and found by g.l.c. on column a to contain D-raannitol hexaacetate (m.p. and mixed m.p. 120-122°;) and D-glucitol hexaacetate (m.p. and mixed m.p. 95-97°) in a ratio of 1:1. (b) Methylation. A^ (20mg) was methylated according to the method of Hakomori. One third of the methylated compound was hydrolyzed (hydro-chloric acid, 2M) to give a trimethyl sugar with the identical chromatographic mobility and characteristic color response of 3,4,6-tri-0-methyl-D-mannose (R^ 0.51, solvent D, a green color response to the p-anisidine spray), and an acidic component (R^ 0.2, solvent F). The remaining methylated aldobiouronic acid was treated with Purdie's reagent and the methyl ester methyl glycoside was then reduced in methanol with sodium borohydride and hydrolyzed. Paper chromatography showed two unresolved spots (R^ 0.56 and 0.58, solvent D). G.l.c. , analysis of the alditol acetates on column a at 180° and a helium flow rate of 43 ml/min gave two peaks identical in retention times to those of the alditol acetates of 3,4,6-tri-0-methyl-D-mannose (15min) and 2,3,4-tri-0-methyl-D-glucose (I8.4min) in a ratio of 1:0.8, The identity of these two compounds was confirmed by comparison of their mass spectra with those of authentic compounds. (c) Anomeric configuration A portion of A^ (20mg) was dissolved in Drfi (1ml) and freeze-dried. The deuterium exchange was repeated once more. P.m.r. of 95 this product (in Do0) showed peaks at T 5.40 (J 7.0 Hz, IH), 5.0 the degraded L-rhamnose. T A f t e r dialysis against running water and deionization, the product was freeze-dried to give a degraded polyol (121mg). The degraded polyol was hydrolyzed (2M hydrochloric acid, 97°, 7h) and examined on paper in solvent B to contain D-galactose, D-glucoee, L-rhamnose and D-erythronolactone (Kglc 5.26). A separate portion of the degraded polyol (40mg) was methylated according to the method of Hakomori, dialyzed against running water for 24h, then hydrolyzed and examined on paper in solvent D and F to give 125 2,4-di-0-methyl-L-rhamnose (R^ 0.62 solvent D; 0.73, solvent F), 2,3,6-tri-O-methyl-D-glucose (R^ 0.54, D; 0.70, F) and 2,4,6-tri-0-methyl-D-galactose (R f 0.35, D; 0.53, F). Authentic standards were used for comparison. The remaining hydrolyzate was reduced (sodium borohydride), acetylated and analyzed by g.l.c. on column a (180°, helium flow rate 55 ml/min) and column d (180°, helium flow rate 60 ml/min) to give the alditoi acetates of 2,4-di-O-methyl-L-rhamnose (6.1 min, column a; 15.6 min, column d), 2,3,6-tri-O-methyl-D-glucose (16.3 min, aj 26.5 min, d) and 2,4,6-tri-O-methyl-D-galactose (14.7 min, a; 30,9 min, d) in a ratio of 1:1:1. Samples were collected and the structures of the sugars were confirmed by mass spectrometry. The acid degraded polyol (30mg) was twice dissolved in DO and freeze-dried. The p.m.r. spectrum of the solution (in D 20) was run at 95° and showed a sharp signal at 1^8.65 ( J 5 g 6 Hz, 3H, CH3 of L-rhamnose) and signals of anomeric protons a t T 5.43 (lH,^*-linkage) and 4.83 (2H, -1inkages). Same mild Smith degradation of the polyol was carried out in trifluoroacetic acid (2M) at room temperature for 4h and 12h in addition to 8h. Similar degree of hydrolysis of the polyol was observed by the analysis of the products. 126 APPENDIX A: Immunochemical Studies on Klebsiella K-Type 5 Polysaccharide Many serological studies and cross-reactions have been reported in the Klebsiella group (2,3,8,12,13,94,95). However, quantitative inhibition studies of the homologous antisera, using oligosaccharides derived from the parent capsular polysaccharides have not been reported. Data were desired on the relative antigenic potency of the subunits of the polysaccharides as well as the size of the antibody reacting site in terms of the size of the oligosaccharide inducing 100% inhibition of the antiserum for the subsequent reaction with the homologous antigen. In the course of the structural study of K5, i t was found that this capsular polysaccharide was composed of such potentially antigenic deterrainents as D-glucuronic acid, pyruvic acid and acetyl group, and that a series of oligosaccharides from the capsular poly-saccharide were available (37). These properties make K5 an excellent material for the immunochemical study i f a potent homologous antiserum could be e l i c i t e d from the experimental animals. It has been found that purified capsular polysaccharides from Klebsiella are immunogenic only to mice and men but not immunogenic to rabbits (96). 0rskov (8) had also reported immunological paralysis induced in rabbits by a heavilycapsulated Klebsiella K5 strain. In this study, attempts have been made to immunize rabbits and mice with Klebsiella K-type 5 bacteria using either whole c e l l s or the purified capsular polysaccharide. The schedule for immunization and bleeding was as follows. New Zealand albino rabbits (three month old) No. 1 and 2 were 127 immunized intravenously with 0.1ml of bacteria suspended in saline (1.2XK?ce&sjmO which was k i l l e d by heating the bacterial culture and washed by saline. The subsequent immunizations were carried out using bacteria k i l l e d by formalin (0.5%). The animals were immunized every 4 days unt i l 4 injections were given. The doses of the 2nd, 3rd and 4th injections were 0.2, 0.4 and 0.8ml of a c e l l suspension of the same concentration. Both rabbits were bled 3 days after the 3rd immunization. Eight days after the 4th injection, the animals were challenged with 1 ml of r l i v i n g bacteria which had been cultured for 48h. Both rabbits died of the challenge. Rabbits No. 3 and 4 were immunized with formalin-killed 8 bacterial culture (1.8 X 10 cells/ml) and challenged by li v i n g bacteria according to the schedule stated in the preceding paragraph. The rabbits were bled by cardiac puncture 4 days after the challenge. A group of three Swiss albino mice (8 weeks old, ca. 20g in weight) was given 4 subcutaneous injections with 0.2ml of K5 polysaccharide (17. in saline) at 4 day intervals. Fourteen days after the last injection the mice were challenged with 0.1ml of a liv e culture (6 X 10 cells/ml). Bleeding was performed 4 days after the challenge using the technique of cutting the t a i l vein. Another group of three Swiss albino mice were immunized with I Q 0.2ml of a formalinized, washed broth culture (3 X 10* cells/ml) by giving 4 injections at 4 day intervals. Fourteen days after the last immunization the mice were challenged with 0.1ml of a l i v i n g broth culture of the bacteria. One mouse died on the 9th day of challenge. The rest were further challenged 10 days after the f i r s t challenge with 128 o 0.2ml of•living bacteria (6 X 10 cells/ml). Mice were bled 4 days after the second challenge by cutting the t a i l vein. For the preliminary test of the antisera, the ring precipitin test was employed by layering 0.15ml of 0.5, 0.1, 0.02 and 0.004% of the capsular polysaccharide in saline (pH adjusted to 8 by IM sodium hydroxide) over 0.15ml of serum in narrow tubes at room temperature. No precipitation was observed over a period of 24h for a l l the sera from rabbits and mice. Ouchterlony gel diffusion tests were also performed on each serum by putting the serum in the central well of an Ionagar plate which was surrounded by five perlferal wells containing 10-fold serial dilutions of the antigen beginning from 0.5% of the polysaccharide in saline (pH adjusted to 8 by IM sodium hydroxide). No precipitin bands were observed up to 7 days. Plate agglutination tests were conducted on the heat-killed K 5 bacteria by gently mixing the bacteria with each of the serial dilutions of a serum which had been heated at 56° for 30 min to inactivate the complement. No flocculation was observed for any of the tested sera. APPENDIX B: P.M.R. SPECTRA 130 K5 CAPSULAR POLYSACCHARIDE 8.0 SWEEP OFISEl (Hi): SPECTRUM AMPLITUDE INTEGRAL AMPLITUDE SPINNING UAH HPS) 2«f CQROiaQ C M A i i t a J 7.0 6.0 MANUAL 5.0 4.0 I AUTO [3] 1 SAMPIE; I UM 3.0 1 • 2.0 REMARKS: 1.0 ^ SWEEP IIME (SEC): i M M u » i , . SWEEP WIDTH (Hi): jfV 1» j«y">»l,jy 1 IK»I FILTER: 1^1' I'J'J'.I' i.'ii.l ! I > l | JJi^ RF POWER LEVEL: . O.O\Q | I o> I , SOLVE II:')-0 K 5 PotroL DATE: . . . \ T \ \ I 1 •* OPERATOR: . P.O. TWICE PERIODATE OXIDIZED AND SODIUM BOROHYDRODE REDUCED K5 60 MHl NMR SPECTRUM NO. 5 ro 133 METHYLATED K5 134 METHYLATED AND REDUCED K5 135 ! I I I ! ! I ! 1 U U I I 1 ! I I ! 1 I I I ! I '~~T~ I 1 I ' ' I PPM : ' 1 I • • • i | ! ' ; i | ! f ! ! | ! I i i | i i ! i | i i i ! | i i i : | i i ! i | i i i ; | i ; i ! | : i ; ' | -CELLOBIOURONIC ACID (Al) FROM K5 136 I I J I i t i I L U — L _ J i i i l i i J—!— I i~~T- l i i i i I I I | ; i i i | i i i i i i I I i i 1 TT~1 i . i . I T ~ n — | i i i i | i I T i l l M i l -- . .-U... . ! . i :•<> .Gfi'-fGlcjinan. • 5.U- • 500 10° • ! i i • •• I t ' l l ' 1 1 1 I I I I I- I ' t I I I I I I I I I I I I I I I I 1 I I I I I I I I I I I 1 I ALDOTRIOURONIC ACID (A2) FROM K5 137 NEUTRAL DISACCHARIDE (N3) FROM K5 PYRUVIC ACID-MANNOSE-ERYTHRONIC ACID COMPLEX FROM K5 139 K62 CAPSULAR POLYSACCHARIDE 140 i i i i ' i i i i i 11 i i i i i i i i i i i i i i i r~ i i i i i i P P M I I I I I I I I ! I I I I ! | I I I I | I I I I | ! I I I | I j I I | M I I | I I I I | I I I I | I I i i i i | i i i i 111 i i i | ; i i i | I I I I | I I I I | I I I I | I I I MILDLY DEGRADED (DILUTE ACID) K62 141 i i i i I i i II i I | | 1 I 1—!—I—I—L J ' 1 1 1 1 ! 1 1 1 1 1 1 1 ,1 \ri 1 1 1 • i._J ...L J _J—1—1—1—LJ—1 l_l t i i : i i i i i T - r l 111 i i i i i T T T T r n - i i i i i i i i " i ALDOBIOURONIC ACID ( A 2 ) FROM K 6 2 142 ALDOTRIOURONIC ACID (A3) FROM K6 2 143 ' ' ' » " ' ' H i "I ' i i " i i i ' i ' i - t i "'I . i—r J—I I I I I I I I _L "' M i i ; i | ; i i i | | i | | i l l ! ' I ' I I I I I j I I I | ] | | | | | "| 1 ,' ' ' ' I ' ', ' ' , ' 1 ' ', ,' ' I ' I ' I I I | I I I I | | I | , | | | | | | | ; - r - T 1500-I r I l I i i i i„ i.. i i i i | i i 7: ....)., 100'.; I' I I I / 3 ~i r r r T PSEUD0TETRA0UR0NIC ACID (A4) FROM K62 144 DISACCHARIDE GLYCOSIDE FROM SMITH DEGRADATION OF K62 1 4 5 J I I I LU i i i i I I I I L I I l ~ T I I ' 1 ' ' ' ' ' ' i • i ' i I M I I | I I I I | i I I ! I. | I I I I I I I I I I J . I I I : | I I I I | ~\—i— i—i— i—rh— i—i—i—i—i—i—r-„ i , I,.! I I I H I I I • I I II I f, I I I I I JL_J—l—L—L I I i ' l . l I I I I | I I I I I I I • ! I I I I I I I I 1 I I I I ! I ! I I I I I I I I I III I I I I I I I I I I I I I I I I ! I t T T I I I I I I i I I I I I I I I I I I I I I I I I I I I I I III K18 CAPSULAR POLYSACCHARIDE 146 n i i i i i i i i i i r ~ r i i L * t M I ' ' i 1 i i i i I i ; i i | i i i i I i i i i 1 i ! i i I , i i ! i i i i i i i u i i i ALDOBIOURONIC ACID (A2) FROM K18 147 "B—J 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I I I I t I I I I I I I I I I II I NEUTRAL DISACCHARIDE (Nl) FROM K18 148 J 1 ! l I I I I I " I I I I | PSEUDOTRIOURONIC ACID (A3) FROM K18 149 i i 11 i I i l I i l i i i i i i i r r i i i i i i i i i , i i i i , , , • , I ' I 11 ' ' I I i I I* I I I' I I I" I'"I i T r w r i Tl I I T T 1 T"l t I | I • I I TETRASACCHARIDE (A4) FROM K18 150 A5a FROM K18 151 I i it I l I l l I l i i i i i i i i i i i i i i i i i i i i i i i 't i- n I • P P M . ' ! 1 1 1 ' 1 1 ' ' ' I ' I I I I "I I I 1 I i I I I I I I I I I V I I I I -J A6a FROM K18 152 i i II i i J L J I I L - l _ J ! I I L_J I i i I ' i j m_ MILD ACID TREATED POLYALCOHOL FROM K18 153 APPENDIX C: Expressions of Raw Sugar Ratio Data from G.L.C••Ana l y s i s Raw sugar r a t i o s g . l . c . a n alysis have been rounded o f f to the nearest i n t e g r a l numbersi to represent the simple i n t e g r a l molar r a t i o of the repeating unit i n the polysaccharides as well as the r a t i o s of the monosaccharides in the oligosaccharides derived from the polysaccharides. The relati'onshlpbbet'Weentt'heaaetiual experimental values represented by the p r i n t outs of peak areas by the d i g i t a l integrator to the corresponding rounded o f f sugar r a t i o s i s exemplified by the following seven examples. i :-l H 1 I I I i ! i i I i i ! I IliTil ANALYSIS OF K5 CAPSULAR POLYSACCHARIDE AFTER GLUCURONIC ACID HAS BEEN TRANSFORMED INTO GLUCOSE 01 -o i t ! h r r ' J IT! •"?" j n i l n-HJ! ri I I I ! I i ANALYSIS OF AUTO HYDROLYZED K5 1 2- !-I ! -l! I I i i i i i i .1 : j i ! i I-: -I J;! j J! | i : ! ! H I M ! .11 i i • i U i i I I ANALYSIS OF METHYLATED K62 CAPSULAR POLYSACCHARIDE ANALYSIS OF THE METHYLATED, REDUCED K18 OO -----| ••• T I - | I - . T : I ; i• i - : !-{'-!"! |.;. .1....).,.. ;.i ••hlTr'rf-TI Lj:!:; •I •!--1 H7 7 i 5 7 5 3 , 1 7 9 1 * 8 2 5 -i ]r':':-r .1! b j t . '^ " • I il. • r:].:::/ II: :!: j.FH'fi •it[-:.f:n. t-H mmkmm A *.<£..b:±L it-lr-fl [./ip^:J{:.:.|./.a :::; j [;; ii Liu;: H:; • i ; i RESOLUTION OF PEAK NO.4 IN PAGE 158 TREATED POLYALCOHOL FROM K18 o l'6-I BIBLIOGRAPHY 1. R.R. Porter, Essays Biochem., 3> 1-20 (1967). 2. S.D. Henriksen and Jorunn Eriksen, Acta path. et. microbiol. scandinav., 51^ 259 (1961). 3. M. Heidelberger and G.G.S. Dutton, J . Immunol., 111. 857 (1973). 4. W.F. Dudman and J.F. Wilkinson, Biochemical J., 62, 289 (1956). 5. P. J . Garegg, B. Lindberg, T. Onn, and T. Holme, Acta Chem. Scand., 25, 1185 (1971). 6. P. J. Garegg, B. Lindberg, T. Onn, and I.W, Sutherland, Acta Chem. Scand., 25, 2103 (1971). 7. K.J. Carson and R.G. Eagon, Can. J . Microbiol., 10, 467 (1964). 8. I. 0rskov, Acta path, et microbiol. scandinav., 38., 375 (1956) 9. E.A. Kabat, Blood Group Substances, Their Chemistry and Immuno-chemistry. New York, Academic Press, Inc., 1956. 10. T. Hubscher and A.H. Eisen, Internat. Arch. Allergy and Applied Immunology, 42, 466 (1972). 11. 0. Luderitz, K. Jann and R. Wheat, Comprehensive Biochem., 26A. 105 (1968). 12. W. Nimmich, Z. Med. Mikrobiol. Immunol., 154. 117 (1968). 13. W. Nimmich, Acta b i o l . med. germ., 26, 397 (1971). 14. R.W. Wheat, C. Dorsch and G. Godoy, J. Bacterid., 89, 539 (1965). 15. P.A. Sandford and H.E. Conrad, Biochemistry, 5, 1508 (1966). 16. H.E. Conrad, J.R. Bamburg, J.D. Epley, and T.J. Kindt, Biochemistry, 5, 2802 (1966). 17. S.I. Hakomori, J. Biochem. (Tokoyo), 55, 205 (1964). 18. S. Hirase, Bull, Chem. Soc. Japan, 30, 75 (1957). 19. ^ J.H. Sloneker and O.G. Orentas, Can. J . Chem., 40, 2188 (1962). 20. Y.M. Choy and G.G.S. Dutton, Can. J. Chem., 52, 684 (1974). 21. P.A.J. Gorin and J.F.T. Spencer, Can. J. Chem., 42, 1230 (1964). 22. Y.M. Choy and G.G.S. Dutton, Can. J. Chem., M, 198 (1973). 162 23. P.J. Garegg, B. Lindberg, T. Onn, and T. Holme, Acta Chem. Scand., 25, 1185 (1971). 24. P.J. Garegg, B. Lindberg, T. Onn and I.W. Sutherland, Acta Chem. Scand., 25, 2103 (1971). 25. B.J. Gormus, R.W. Wheat, and J.F. Porter, J . Bacteriol., 107. 150 (1971). 26. M. Heidelberger, W.F. Dudman, and W. Nimmich, J. Immunol., 104. 1321 (1970). 27. L.C. Gahen, P.A. Sandford, and H.E. Conrad, Biochemistry, 6, 2755 (1967). 28. I.W. Sutherland, Biochemistry, 9, 2180 (1970). 29. B. Lindberg, J. Lonngren, and J.L. Thompson, Carbohydr. Res., 25, 49 (1972). 30. H. Bjorndal, B, Lindberg, J . Lonngren, K. Rosell, and W. Nimmich, Carbohydr. Res., 27, 373 (1973). 31. Y.M. Choy and G.G.S. Dutton, Can. J. Chem., 51, 3015 (1973). 32. Y.M. Choy, G.G.S. Dutton, and A.M. Zanlungo, 51, 1819 (1973). 33. B. Lindberg, K. Samuelsson, Carbohyd. Res., 30_, 63 (1973). 34. H. Bjorndal, B. Lindberg, J . Lonngren, and M. Meszaros, Carbohyd. Res., 31, 93 (1973). 35. Y.M.Choy and G.G.S. Dutton, Can. J. Chem., 51, 3021 (1973). 36. G.G.S. Dutton and M.T. Yang, Can. J. Chem., 50, 2382 (1972). 37. G.G.S. Dutton and M.T. Yang, Can. J. Chem., 51, 1826 (1973). 38. CK. De-Bruyne and J . Wouters-Leysen, Carbohyd. Res., 17_, 45 (1971). 39. P. Albersheim, D.J. Nevis, P.D. English, and A. Karr, Carbohydr. Res., 5, 340 (1967). 40. B. Capon, Chemical Review, 69, 407 (1969). 41. G.A. Adams, Methods Carbohyd. Chem., 5, 269 (1965). 42. D.G. Medcalf and K.A. G i l l e s , Cereal Chem., 45, 550 (1968). 43. A.S. Perlin, Cereal Chem., 28, 382 (1951). 44. B. Lindberg and J; Lonngren, Carbohydr. Res., 23, 47 (1972). 45. C.T. Bishop and F.P. Cooper, Can. J. Chem., 41, 2742 (1963). 163 46. V. Smirnyagin and C.T. Bishop, Can. J . Chem., 46, 3085 (1968). 47. V. Smirnyagin, C.T. Bishop, and F.P. Cooper, Can. J. Chem., 43, 3109 (1965). 48. G. Entlicher and J.N. BeMiller, Carbohydr. Res., 16, 363 (1971). 49. P.A. Sandford, P.R. Watson and A, Jeans, VI International Symposium on Carbohydrate Chemistry. Madison, U.S.A. August 1972. Abstracts of paper presented. P. 35. 50. Y.M. Choy, G.G.S. Dutton, A.M. Stephen and M.T. Yang, Anal. Lett., 5, 675 (1972). 51. G.M. Bebault, Y.M. Choy, G.G.S. Dutton, N. Funneli, A.M. Stephen, and M.T. Yang, J. Bacteriol., 113, 1345 (1973). 52. M. Abdel-Aker and F. Smith, J . Amer. Chem. Soc, 73, 5859 (1951). 53. R. Kuhn, H. Trischmann, and I. Low, Angew. Chem., 67, 32 (1965). 54. T. Purdie and J.C. Irvine, J . Chem. Soc, 83, 1021 (1903). 55. G.G.S. Dutton and S. Kabir, Anal. Lett., 4, 95 (1971). 56. P.D. Bragg and L. Hough, J. Chem. Soc, 4347 (1957). 57. S.C. Churms, Advances Carbohydrate Chem., 25, 13 (1970). 58. I.J. Goldstein, G.W. Hay, B.A. Lewis and F. Smith, Methods Carbohydrate Chem., 5, 361 (1965). 59. B. Lindberg, Private Communication. 60. R.J. Yu and C.T. Bishop, Can. J. Chem., 45, 2195 (1967). 61. T. Painter and B. Larsen, Acta Chem. Scand., 24, 813 (1970). 62. M.F. Ishak and T. Painter, Ibid., 25, 3875 (1971). 63. A.M. Stephen, J . Chem. Soc, 2030 (1962). 64. CC. Sweeley, R. Bentley, M. Makita and W.W. Wells, J . Amer. Chem. Soc, 85, 2497 (1968). 65. S.W. Gunner, J.K.N. Jones, and M.B. Perry, Can. J . Chem., 39, 1892 (1965). 66. CO. Aspinall, J. Chem. Soc, 1676 (1963). 67. J.S. Sawardeker, J.H. Sioneker and A. Jeanes, Anal. Chem., 37, 1602 (1965). . 164 68. H. Bjorndal, C.G. Hellerqvist, B. Lindberg and S. Svenssen, Angew. Chem. internat. Edit., 9, 610 (1970). 69. G.O. Aspinall, J . Chem. Soc, 1676 (1963). 70. E. Percival, Carbohydr. Res., 4, M l (1967). 71. A.G. Cooper, J.F. Codington, and M.C. Brown, Proc, Nat. Acad. Sci. U.S.A., 71, 1224(1974). 72. D. Allan, J . Auger and M.J. Crumpton, Nature New Biology, 236. 23 (1972). 73. K. Biemann, "Mass Spectrometry, Application to Organic Chemistry" McGraw-Hill, New York, 1962. 74. M.S.B. Munson and F.H. Field, J . Amer. Chem. Soc, 88, 2621 (1966). 75. H.R. Schulten and H.D. Beckey, Org. Mass Spectrometry, J_, 861(1973). 76. W. Heyns and H. Scharmann, Liebigs Ann. Chem., 667. 183 (1963). 77. O.S. Chizkov, N.N. Moldostov and N.K. Kochetkov, Carbohydr. Res., 4, 273 (1967). 78. G. Pettersson, 0. Samuelson, K. Anjou and E. von Sydow, Acta Chem. Scand., 21, 1251 (1967). 79. O.S. Chizhov, L.S. Golovkina and N.S. Wulfson, Izv. Akad. Nauk. SSSR, Ser. Khim., 1915 (1966). 80. J . Hoffman, B. Lindberg and S. Svensson, Acta Chem. Scand., 26, 661 (1972). 81. S.L. Patt and B.D. Sykes, J . Chem. Phys. 56, 3182 (1972). 82. G.M. Bebault, J.M. Berry, Y.M. Choy, G.G.S. Dutton, N. Funnell, L. D. Hayward, and A.M. Stephen, Can. J. Chem., 51, 324 (1973). 83. M. Duckworth and W. Yaphe, Chem. Ind. (London), 747 (1970). 84. J.H. Sloneker and Jeanes, Can. J. Chem., 40, 2066 (1962). 85. J.O. Deferrari, E.G. Gros, and 1.0. Mastronardi, Carbohydr. Res., 4, 432 (1967). 86. B. Lindberg and L. Selleby, Acta Chem. Scand., 14, 1051 (1960). 87. J.G. Slonecker, D.G. Orentas, C.A. Knutson, P.R. Watson, and A. Jeanes, Can. J. Chem., 46, 3353 (1968). 88. A.N. de Belder and B. Norrman, Carbohydr. Res., 8, 1 (1968). 165 89. R.W. Bailey, Oligosaccharides. Pergamon, Oxford, 1965, p. 99. 90. J.K.N. Jones and M.B. Perry, J. Aner. Chem. Soc, 79, 2787 (1957). 91. R.L. Whistler and D.F. Durso, J . Amer. Chem. Soc, 72, 677 (1950). 92. H. BjSrndal, B. Lindberg and S. Svensson, Acta Chem. Scand., 2JL, 1801 (1967). 93. Z. Dische, ,: Methods Carbohyd. Chem., 1, 478 (1962). 94. S.D. Henriksen, Acta path, et microbiol. scandinav., 34, 249 (1954). 95. M. Heidelberger, W.F. Dudman and W. Nimmich, J. Immunol., 104. 1321 (1970). 96. J . Erikseri; ; ^, Bakteriologiske Institutt, Rikshosptalet, Oslo, Norway; Personal communication. 97. M. Stacey and S.A. Barker, Polysaccharides of Micro-organisms. Oxford; Clarendon Press, 1967. 98. M.J. How, J.S. Brimacombe, and M. Stacey, Adv. Carbohydr. Chem., 19, 313 (1964). 99. E.J. Bourne and S. Peat, Adv. Carbohydr. Chem., 5, 145 (1950). 100. M. Heidelberger, Bacteriol. Rev., 3, 49 (1939). PUBLICATIONS Lung Ching L i n and Mo-tai Yang, Anomalous N i t r a t i o n Reactions of Halotropones and Tribromotropolone, J. Chinese Chem. S o c , I I , L3, 178 (1966.) J.J.M.-Rowe, K.B. Gibney, M.T. Yang, G.G.S. Dutton, . Periodic Acid-Dimethyl Sulfoxide Mixtures, a P o t e n t i a l Hazard, J.A.C.S., 90, 1924 (1968) G.G.S. Dutton and M.T. Yang, 4,6-0-(1-Carbox-yethylidene)-D_-mannose as a Str u c t u r a l Unit • i n Capsular Polysaccharide of K l e b s i e l l a K-• type 5, Can. J . Chem., 50, 2382 (1972) G.G.S. Dutton 'and M.T. Yang, 'The Structure of the '< Capsular' Polysaccharide of K l e b s i e l l a • i K-type 5, Can. J.' Chem., 51_, 1826 (1973)' Y.M. Choy, G.G.S. Dutton, A.M. Stephen, and M.T. Yang, P.M.R. Spectroscopic Analysis of B a c t e r i a l Polysaccharides Containing Pyruvic Acid, A n a l y t i c a l L e t t e r s , 5(10), 675 (1972) G.M. Bebault, Y.M. Choy, G.G.S. Dutton, N. Funnell, A.M. Stephen, and M.T. Yang, Proton Magnetic Resonance Spectroscopy of K I e b s i e l l a Capsular Polysaccharides, J . B a c t e r i d . , 113, 1345 (1973) i