Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structural studies of Klebsiella capsular polysaccharides Choy, Matthew Yuen-Min 1973

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

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

Item Metadata

Download

Media
831-UBC_1973_A1 C46.pdf [ 5.48MB ]
Metadata
JSON: 831-1.0060204.json
JSON-LD: 831-1.0060204-ld.json
RDF/XML (Pretty): 831-1.0060204-rdf.xml
RDF/JSON: 831-1.0060204-rdf.json
Turtle: 831-1.0060204-turtle.txt
N-Triples: 831-1.0060204-rdf-ntriples.txt
Original Record: 831-1.0060204-source.json
Full Text
831-1.0060204-fulltext.txt
Citation
831-1.0060204.ris

Full Text

15103 STRUCTURAL STUDIES OF KLEBSIELLA CAPSULAR POLYSACCHARIDES BY MATTHEW YUEN-MIN CHOY B.Sc, University of Hong Kong, 1964 M.Sc, Simon Fraser University, 1970 A THESIS SUBMITTED T.N PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF Doctor of Philosophy in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA A p r i l , 1973 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, 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 available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by h i s representatives.' It i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The University of B r i t i s h Columbia Vancouver 8, Canada STRUCTURAL STUDIES OF KLEBSIELLA CAPSULAR POLYSACCHARIDES ABSTRACT Eighty types of Klebsiella are recognised on the basis of immunochemical tests. Qualitative analyses of the capsular polysaccharides produced by these bacteria have been provided by Nimmich. Nevertheless very few detailed structures are known in contrast to the qualitative data published. The present investigation deals with the structural studies of capsular polysaccharides from four strains of these bacteria, for example K20, K21, K24 and K56. Besides the classical methods used, new techniques have been used or developed in order to obtain detailed information on the structures. Modern methods used include the circular dichroism (c.d.) for the assignment of the D or L configuration of sugar constituents. Gas-liquid chromatography (g.l.c.) and mass spectro-metry (m.s.) were used to analyse the partially methylated sugars which can be characterized further by demethylation and reacetylation to give crystalline derivatives. In the course of our studies, p.m.r. spectroscopy at 95° has been developed and becomes a very powerful tool for the assignment of the anomeric linkages of oligosaccharides and polysaccharides. It has also been found to be an excellent method for the detection and quantitative assay of other functional groups such as O-acetyl or pyruvic acid without degrading the polymer. The application of the above modern techniques together with use of the classical methods affords the detailed structures of K20, K21, K24 and K56, the repeating units of which are as follow: i i i K20 2 1 3 1 — D-Manp ^-^ D-Galp ~ 2~ a ~ 3 1 D-Galp 3~ 3 D-GlcAp + -OAc group K21 3 1 3 3 2 1 3 1 — D-GlcAp D-Manp — - D-Manp — - D-Galp — . - a = - ct = - a = - 3 4 a 1 D-Galp 4 V6 CH3-C-COOH K24 2 D-GlcAp D-Manp -^-2- D-Manp -^-^ D-Glcp . - a = - a = - a = 4 3 1 D-Manp + -OAc group K56 1 3 1 3 1 3 D-Glcp —z— D-Galp — — D-Galp D-Galp = . - 6 = L 8 = „ - a = -4 V 6 CH3-C-COOH 1 L-Rhap The structure of K20 is the f i r s t of these Klebsiella capsular polysaccharides to be encountered with an aldobiouronic acid side chain. The capsular polysaccharides K21 and K24 ?ach contains the same-type " °f aldobiouronic,.:ac-id_with..branching on- the Djsglucuronic acid. The .capsular polysaccharide K56 has a L-rhamnose side chain. The acidity of the iv polysaccharide is due only to the pyruvic acid present and this constitutes the first investigation of those Klebsiella polysaccharides lacking uronic acid. V TABLE OF CONTENTS Page INTRODUCTION i 1 METHODS USED FOR STRUCTURAL ANALYSIS ' 4 A. Isolation and Purification 4 B. Hydrolysis and Characterization of Sugar Constituents. 4 C. Configuration of Constituents 6 D. Methylation Analysis 7 (i) Gas-liquid chromatography of methylated sugars.. 7 ( i i ) Mass spectrometry of methylated a l d i t o l acetates 8 ( i i i ) Demethylation and reacetylation of partially methylated a l d i t o l acetates 12 E. Smith Periodate Degradation ^ F. Partial Hydrolysis 15 G. Anomeric Linkages 16 H. Detection of Other Functional Groups 18 STRUCTURES 'INVESTIGATED 20 BIBLIOGRAPHY 22 PART I. THE STRUCTURE OF THE CAPSULAR POLYSACCHARIDE FROM KLEBSIELLA K-TYPE 20 25 SUMMARY 26 DISCUSSION 27 v i Page EXPERIMENTAL . 34 General Methods 34 Isolation and Properties of K20 Capsular Polysaccharide ... 35 Hydrolysis of Polysaccharide 36 Partial Hydrolysis of Polysaccharide 38 Methylation Analysis of the Polysaccharide 39 Methylation of Degraded Polysaccharide 41 Reduction of the Capsular Polysaccharide 41 Smith Periodate Degradations and Methylation Studies of the Degraded Polysaccharides 42 BIBLIOGRAPHY 48 PART I I . THE STRUCTURE OF THE CAPSULAR POLYSACCHARIDE FROM KLEBSIELLA K-TYPE 56 49 SUMMARY 50 DISCUSSION 51 EXPERIMENTAL 57 Isolation and Properties of K56 Capsular Polysaccharide ... 57 Hydrolysis of the Polysaccharide 58 Methylation 58 Smith Degradation 59 Partial Hydrolysis of the Polysaccharide 60 Partial Hydrolysis of Methylated Capsular Polysaccharide... 62 BIBLIOGRAPHY 69 v i i Page APPENDIX I. The Structure of the Capsular Polysaccharide from Klebsiella K-Type 21. Can. J. Chem., 51, 198 (1973) 71 APPENDIX II. The Structure of the Capsular Polysaccharide of Klebsiella K-Type 24. Can. J. Chem., in press... 82 v i i i LIST OF TABLES Table - Page 1 Primary fragments in the mass spectra of partially methylated sugars in the form of their a l d i t o l acetates.. 11 PART I 1 P.m.r. data on Klebsiella K20 capsular polysaccharide and derived polysaccharides and oligosaccharide 45 2 Methyl ethers from the hydrolysis of methylated Klebsiella K20 polysaccharides 46 3 Diagnostic prominent peaks (m/e) in the mass spectra of acetates of methylated alditols 47 PART II 1 P.m.r. data on Klebsiella K56 capsular polysaccharide and derived oligosaccharides 65 2 Methyl ethers from the hydrolysis of methylated Klebsiella K56 polysaccharides 67 3 Diagnostic prominent peaks (m/e) in the mass spectra of acetates of methylated alditols 68 ix LIST OF FIGURES Figure Page 1 P.m.r. spectra of K21 Klebsiella capsular polysaccharides in D20 at 95° 17 2 P.m.r. spectra of K24 Klebsiella capsular polysaccharides in DO at 95° 17 ACKNOWLEDGEMENTS The author wishes to express his sincere gratitude to Dr. G.G.S. Dutton for his guidance, inspiration and assistance. The financial assistance from the Chemistry Department, University of British Columbia and National Research Council of Canada are also gratefully acknowledged. 1 INTRODUCTION The polysaccharides of micro-organisms can be divided into three major groups according to their morphological localization: (i) extra-cellular surface polysaccharides located outside the c e l l wall and frequently termed capsular polysaccharides, ( i i ) c e l l wall polysaccharides, and ( i i i ) somatic or intracellular polysaccharides located inside the cytoplasmic membrane. The presence or absence of these and the amounts and nature of them w i l l differ widely from organism to organism. There is a distinction^ between the c e l l wall of a bacterium and. the capsule, for the latter may be described as a covering layer outside the c e l l wall. The presence of a capsule is one of the conditions for the virulence of a c e l l . This layer is usually composed of a single polysaccharide. The capsular polysaccharides give specific agglutination or precipitation reactions with the antisera obtained by the action of the bacteria concerned on animals. These polysaccharides may act as antigens including the production of the appropriate antibody in the animals, the antibodies then conferring immunity against the particular 2 bacterial infection concerned. The whole cellular structure of some organisms such as Escherichia or Klebsiella may be embedded in a hydrophilic, acidic polysaccharide (often termed capsular or K-antigen) which is loosely bound to the c e l l surface. Such polysaccharides may sometimes also be found dissolved in the liquid culture f i l t r a t e in which the bacteria are grown, hence the term extracellular polysaccharides. 2 Bacteria of the genus Klebsiella belong to the family Enterobacteriaceae, .. ' 3 the chemistry of which has been reviewed by Luderitz, Wheat and Jann. Approximately 80 types of Klebsiella are known on the basis of immunochemical tests. One of the characteristics of Klebsiella bacteria is the formation of a capsular polysaccharide which is antigenic and the composition of which has been stated to be the same as that of the 2 slime polysaccharide excreted into the medium. Nimmichf'^ has reported the qualitative composition of K-types 1 to 80 and has shown that the great majority contain D-glucuronic acid in combination with hexoses such as D-glucose, D-galactose and D-mannose. Many strains contain, in addition, L-rhamnose and a few contain L-fucose. The presence of bound pyruvic acid in Klebsiella was f i r s t reported by Wheat, Dorsch and Godoy in the case of K. rhinoscleromatis (type 3) and pyruvic acid has subsequently been found by Wheat and his colleagues to be present in several different species,^ in fact nearly half of the Klebsiella polysaccharides have been found to contain pyruvic acid. It is interesting to note that a l l the capsular Klebsiella polysaccharides are acidic in nature, containing either uronic acid or pyruvic acid or both; there are a few containing an unknown ketoacid. In contrast to the data on the qualitative composition of most Klebsiella capsular polysaccharides there have been few detailed structures published of these carbohydrate polymers. Structures have 8 9 been given for the capsular polysaccharides of K-types 2, 8, and ,.^10,11 ^ each of the capsular polysaccharides so far examined the structures may be expressed as multiples of a repeating unit containing three to five sugar units. 3 To analyze the structures of these Klebsiella polysaccharides f u l l use of the classical and modern techniques has been made. 4 METHODS USED FOR STRUCTURAL ANALYSIS The determination of the structure of a polysaccharide may be achieved easily i f f u l l use is made of the present knowledge of these substances and the available modern analytical techniques. The following is an outline of the methods' used in the present investigation of the structures of the capsular polysaccharides from four strains of Klebsiella: A. Isolation and Purification Since i t is believed that the slime polysaccharide excreted into the 2 medium of the Klebsiella culture has the same composition as the capsular polysaccharide, the capsular polysaccharide is usually obtained by water extraction. Detergent cations such as cetyltrimethylammonium react with polyanions to form salts which are very insoluble in water. Neutral polysaccharides do not react except as borate complexes. Cetyltrimethylammonium 12 bromide has therefore been widely employed for purifying Klebsiella polysaccharides most of which are quite acidic. B. Hydrolysis and Characterization of Sugar Constituents Before hydrolysis, the purified polysaccharide is characterized by the determination of (i) the optical rotation in water and ( i i ) the equivalent weight by titration with sodium hydroxide. The individual sugar constituents are obtained by hydrolysis which is usually done with 13 0.5 M or 1 M sulfuric acid. In recent years, trifluoroacetic acid has been widely used in our laboratory to hydrolyze polysaccharides. In the case of trifluoroacetic acid, advantage has been taken of i t s easy 5 removal by evaporation at reduced pressure. Paper partition chromatography introduced i n 1944 for the separation and determination of amino acids, has been highly successful for the 14 15 separation of sugars. It was demonstrated ' that closely related sugars could be separated from each other provided a suitable solvent or combination of solvents could be selected. Quantitative determination 16 may for instance be accomplished by the phenol-sulfuric acid method. Nowadays, gas liquid chromatography (g.l.c.) has been well developed to the point where i t may be generally employed for the qualitative and quantitative analysis of sugars. One advantage of the method i s that only small amounts of material are required. The great impact of g.l.c. separation of carbohydrates started in 1963 when Sweeley and coworkers'^ published a paper on the application of trimethylsilyl derivatives of carbohydrates in g.l.c. The discovery that these derivatives were easily formed and that they were volatile revolutionized carbohydrate analysis. A problem i s encountered in quantitative analysis of mixtures of sugars, as their trimethylsilyl ethers, due to the overlapping of peaks which complicates the interpretation of the results. Early attempts to analyze sugars as their f u l l y acetylated a l d i t o l s , where each sugar gives a single peak, were not entirely successful, as the resolution of isomeric substances 18 was often unsatisfactory. However, Sawardeker and coworkers found that improved separation of acetylated alditols could be obtained using a copolymer of ethylene glycol succinate polyester and a n i t r i l e silicone polymer (ECNSS-M) as the stationary phase. One excellent advantage is that g.l.c. of the a l d i t o l acetates permits the isolation and characteriza-tion of crystalline derivatives of the common hexoses even with semi-6 microquantities. The other merit of the method is that each sugar gives a single peak and thus, quantitative evaluation of the chromatogram is considerably facilitated. C. Configuration of Constituents The D or L configuration of monosaccharides is usually assigned from the results obtained by measurement of the optical rotation which involves large quantities or alternatively by the action of specific enzymes, i f they are available. With the modern development of circular dichroism (cd.), one can now assign the configuration of 19 monosaccharides using only a semimicroquantity. In the classical method of configuration assignment, the monosaccharides are isolated by paper chromatography and their optical rotations measured. Large quantities are required in order to get an accurate measurement. However, such amounts are not always available in some cases of bacterial polysaccharides. To overcome this difficulty, a semimicroanalytical technique has been developed in our laboratory by measuring the circular dichroism curves of the alditol acetates. The acetoxy group has a very strong absorption at >^ 213 mp. Thus the c d . curve may be obtained with less than 1 mg of sample, which is usually isolated via the g.l.c. Identical samples will have identical c d . curves (i.e., same extinction coefficient and sign) so by comparing c d . curves of standard samples and unknowns the configuration (g or L) of a monosaccharide may be assigned unambiguously. 7 D. Methylation Analysis After purification of the polysaccharide, determination of the composition and characterization of the component sugars, one is faced with the problem of determining (a) the mode of union of various component sugars, (b) the sequential order of the components, and (c) the anomeric nature of the glycosidic linkages uniting the components. The f i r s t problem is solved by subjecting the polysaccharides themselves to methylation analysis. Generally the methylation proceeds normally by treatment with sodium hydride in methyl sulfoxide and methyl i o d i d e ,a process which gives good methylation within a reasonably short time. The analysis of the partially methylated polysaccharide has been greatly simplified in recent years by three major improvements in experimental techniques. These are (i) the separation of the methylated 21 sugars or their derivatives by g . l . c , ( i i ) the determination of the 22 position of methoxyl groups by mass spectrometry (m.s.), and ( i i i ) demethylation and reacetylation of the partially methylated a l d i t o l 23 acetates to their parent a l d i t o l acetates; discussions of these follow: (i) G.l.c. of methylated sugars G.l.c. has become an important method in methylation analysis of polysaccharides, both as an aid in identifying individual methylated sugars and for their quantitative measurement. The methylated sugars usually cannot be separated directly by g.l.c. but have to be transformed into more volatile derivatives such as methyl glycosides. In most cases a methylated sugar gives a mixture of glycosides in f a i r l y constant propor-tions which sometimes fa c i l i t a t e s identification. In more complex 8 mixtures the multiplicity of peaks becomes a disadvantage, as the quantitative evaluation becomes more d i f f i c u l t . The same problem i s encountered in quantitative analysis of mixtures of sugars, as their 16 trimethylsilyl ethers. By reduction of the methylated sugars to alditols a single derivative i s obtained from each sugar. The separation of isomeric al d i t o l derivatives (e.g., acetates) was not, however, better than that obtained from isomeric methyl glycoside derivatives u n t i l after the introduction of a new phase for g . l . c , ECNSS-M.^ Columns packed with butanediol succinate (BDS) are also found to be highly effective in separating methylated a l d i t o l acetates and in some cases give better resolution than ECNSS-M. For example, 2,4,6- and 3,4,6-tri-O-methyl-D-mannitol acetates are not separated on an ECNSS-M column but are well 2 A resolved on a 5% butanediol succinate column. In situations where the partially methylated a l d i t o l acetates are not separated by the above columns(see analysis of capsular polysaccharide of K24'in appendix), satisfactory resolutions have been obtained by 25 using the partially methylated aldose acetates. Although this method causes each compound to give up—to r f our- peaks-, i t - i s a-supplement-1 to- the above"procedures. ( i i ) Mass spectrometry of methylated a l d i t o l acetates Recently, mass spectrometry has become an important tool in carbo-hydrate chemistry and several groups have studied the mass spectra of 2 6 various carbohydrates. Heyns and colleagues investigated the mass 27 spectra of f u l l y methylated glycosides, and Kochetkov and coworkers 28 and Samuelson and coworkers examined the mass spectra of the trimethylsilyl 9 derivatives of parti a l l y methylated glycosides. Mass spectrometry of p a r t i a l l y methylated a l d i t o l acetates has been studied by Lindberg 22 29 and coworkers ' who made a systematic investigation and made the following generalizations:-(a) 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. (b) 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 a l d i t o l acetate derived from 2-0_-methyl-D-glucose therefore only gives two main primary fragments. CH„OAc 2 HC-OCEL 333 .117. | J AcO-CH HC-OAc I HC-OAc I CH2OAc (c) When the molecule contains two adjacent methoxylated carbons, e.g. the a l d i t o l acetates derived from 2,3-di-O-methyl-D-glucose, fission between them is preferred over fission between one of these and a neighboring acetoxylated carbon. CH2OAc CHo0-CH 261 HC-OAc I HC-OAc I CH OAc 10 (d) Secondary fragments are formed from the primary fragments by single or consecutive loss of acetic acid (m/e 60), ketene (m/e 42), methanol(m/e 32), or formaldehyde (m/e 30). For example, two primary fragments having m/e 161 are expected: F^ and F^ (Lindberg's terminology) + + HC=0-CHo HC=0-0CH. I 3 | 3 HC-O-CHL HC-OCOCH,, | 3 | 3 HoC-0C0CH, HoC-0-CH, 1 3 2. 3 F F 1 2 Loss of methanol from F^ gives a secondary fragment having m/e 129, while F^ gives m/e 101 by loss of acetic acid. Further loss of ketene from m/e 129 gives m/e 87, and loss of formaldehyde from m/e 101 gives m/e 71. On reduction some pairs of methylated sugars, e.g. a 3-0-methyl and a 4-0-methyl hexose, give an a l d i t o l with the same substitution pattern thus providing the same mass spectra. The loss of information brought about by reduction of the sugars to alditols can be avoided i f the reduction is performed with sodium deuterioborate; the two alditols then give different mass spectra. CHDOAc CHDOAc I I HC-OAc HC-OAc 261 CHo0-CH ' . AcO-CH 3_.| 190___ _ _ j HC-OAc HC-OCH. 189 I .-__2p2 | 3 HC-OAc "" . ~" HC-OAc I CH20Ac CH20Ac 11 The following table i s a summary of the prominent primary peaks of different partially methylated a l d i t o l acetates. Table 1. Primary fragments in the mass spectra of partially methylated sugars in the form of their a l d i t o l acetates. Position m/e of CH3 45 59 89 117 131 161 175 189 203 205 217 233 261 305 333 Pentoses 2(4) x x 3 x 2,3(3,4) x x 2.4 x x 2.5 x x x 3,5 x x x 2.3.4 x x 2.3.5 x x x Hexoses 2(5) x x 3(4) x x 6 x 2.3 x x 2,4(3,5) x x 2.5 x 2.6 x x 3.4 x 3,6 x x x 4,6 x x x 5,6 x x x 2.3.4 x x x x 2.3.5 x x 2.3.6 x x x 2,4,6 x x x x 2,5,6 x x x x 3,4,6 x x x 2,3,4,6 x x x x 2,3,5,6 x x x x 6-Deoxyhexoses 2 x 3 x x 4 x x 2.3 x x 2.4 x x x 3,4 x x 2.3.4 x x x x 2.3.5 x x x 12 Table 1 (continued) Position m/e 45 59 89 117 131 161 175 189 203 205 217 233 261 305 333 3,6-Dideoxyhexoses 2 x x 2,4 x x X X ( i i i ) Demethylation and reacetylation of partially methylated  a l d i t o l acetates Mass spectrometry has become an important method for the tentative identification of the methylated sugars obtained on methylation analysis of polysaccharides. One advantage of this method is that only small amounts of materials are required. However, isomeric a l d i t o l acetates (e.g., 2,4,6~tri-0-methyl-D-galactitol and mannitol acetates) give very similar mass spectra, typical of that substitution pattern. Mass spectrometry cannot distinguish between galactitol and mannitol. Attempts to characterize the partially methylated sugar by demethylation 23 and reacetylation have been made. The par t i a l l y methylated a l d i t o l acetate 30 is demethylated by boron trichloride and reacetylated with pyridine and acetic anhydride. In this case the partially methylated common hexitol acetates (e.g., gal a c t i t o l , mannitol and glucitol acetates) are converted to hexaacetates which can be isolated via g.l.c. as crystalline derivatives. In the study of K21 capsular polysaccharide, 3,4,6-tri-0_-methyl-D-mannose and 2,4,6-tri-0_-methyl-D-mannose were individually converted to D-mannitol hexaacetate, m.p. and mixed m.p. 1 1 8 - 1 2 1°. 23 13 Together with the result obtained by g.l.c.-m.s., the identity of a methylated sugar can be characterized without ambiguity. The results obtained w i l l show the composition of the polysaccharide, the nature of the building units, how they are joined together, and the number of residues in the average repeating unit. The findings also enable the units at which branching occurs to be designated. However, by themselves, methylation studies on the polysaccharides provide l i t t l e knowledge concerning the exact sequence of the building units. Methylation results become more diagnostic when considered in conjunction with Smith periodate degradation and partial hydrolysis studies. E. Smith Periodate Degradation Non-reducing terminal units in a polysaccharide or (1-4-6)-linked non-terminal units having three adjacent hydroxyl groups w i l l be cleaved by two molecular proportions of periodate to give one molecular proportion of formic acid. Non-terminal units joined by (l->-2) or (1-^ -4) bonds undergo cleavage by one molecular proportion of periodate, but no formic acid i s generated. Units which do not possess adjacent hydroxyl groups such as non-terminal units joined by (1-K3) bonds or units involved in branching at or are not affected by periodate. Thus oxidation of a polysaccharide and quantitative determination of the proportion of the surviving sugar units w i l l give information concerning the nature and proportion of the glycosidic linkages present 31 in the polysaccharide. Smith and coworkers reported that when a sugar residue of a polysaccharide i s cleaved by periodate and reduced, 14 the resulting alcoholic derivatives, being a true acetal, i s sensitive to mild acid hydrolysis, whereas when a sugar unit which survives cleavage i s joined to a unit which i s cleaved, the surviving unit appears as a glycoside which is relatively stable to mild acid hydrolysis. Because of the marked difference in st a b i l i t y between true acetals and glycosides, i t is now possible to obtain glycosides of mono-, di- and oligosaccharides from a wide variety of polysaccharides after the Smith periodate degradation. The analysis of the structures of these glycosides w i l l throw light on the fine structure of the parent polysaccharides. In our study of the capsular polysaccharide of K20, successive Smith degradations have been used to give a series of degraded polysaccharides. For example, in the original carboxyl reduced polysaccharide, only p-glucose can be removed by the Smith degradation. 1) 10 ~ 2 1 3 1 D-Manp D-Galp = 3 a = f 1 D-Galp 2) BHZ 3) H30 2 1 3 1 — D-Manp ^ --f2- D-Galp + a = 1 D-Galp 1 D-Glcp K20 1) i o 4 2) BH, 3) H 0' + 2 1 3 1 — D-Manp D-Manp — S „ In the degraded polysaccharide S^, only D-galactose in the side chain is susceptible to periodate oxidation. Removal of the D-galactose 15 side chain w i l l give a new polysaccharide which i n turn gives a galactosyl-glyceraldehyde on further degradation. For K56 capsular polysaccharide a l l sugars are resistant to periodate oxidation because of l->3 linkages, except the L-rhamnose side chain which i s linked l->2 to the main chain. 3 1 3 1 3 1 3 1 - D-Glcp D-Galp—2- D-Galp D-Galp P - - p - | 2 - a -6 1 CH -C-COOH L-R n aP K56 Instead of using partial hydrolysis to remove the L-rhamnose side chain the Smith degradation has been found to be excellent for the above purpose. F. Partial Hydrolysis If the hydrolysis of a polysaccharide i s stopped before i t s completion, oligosaccharides may be isolated. Analysis of the structures of these d i - , t r i - and higher oligosaccharides w i l l provide evidence for the mode of linkage between them. Partial hydrolysis i s usually done by using lower acid concentrations than for complete hydrolysis (e.g. 0.125 M sulfuric acid or 0,5 M TFA) under the ordinary conditions or by using the dialysis apparatus described by Galanos, Luderitz and 32 Himmelspach. The monosaccharides can be separated by charcoal column chromatography while acidic oligosaccharides can be separated from neutral oligosaccharides by an ion exchange resin (Duolite A4). 16 Oligosaccharides up to a D.P. of 4 are separated by paper chromatography , 33 Alternatively, they can be separated by using gel chromatography (e.g. K56 using Sephadex G-15). G. Anomeric linkages Proton magnetic resonance (p.m.r.) spectroscopy which is widely used in carbohydrate chemistry has now been extended to polysaccharide 8 3' and oligosaccharide chemistry for the assignment of anomeric linkages. ' In general the protons at C-l of aldopyranoses in deuterium oxide resonate at lower field than the remaining annular protons. Moreover, the anomeric protons in equatorial positions (ct-D-conf iguration) resonate to lower field than those occupying corresponding axial positions (B-D-configuration). In glucose and galactose a trans diaxial interaction between the protons on C-l and C-2 (g-anomer) results in a spin-spin coupling constant (JO ^) which is larger than that given by a gauche interaction (a-anomer). In the case of Klebsiella polysaccharides, which have very high molecular weights, some difficulty was experienced in making qualitative and quantitative determinations of the anomeric protons on account of the large HOD peak present. This peak appears about T 5-6 and partly covers those regions of the spectrum associated with anomeric and ring protons. The magnitude of the HOD peak was largely due to the necessity of working with solutions of less than 2% concentration because of their viscosity. This same viscosity prevented the solutions being cooled in order to move the HOD peak downfield. It has now been found that excellent p.m.r. spectra may be obtained by dissolving the sodium 17 salt of the polysaccharide, after D20 exchange, i n D20 and running 35 36 the spectra at 9 5° . » In this way the HOD peak is shifted upfield to T 5.5 leaving the region of the anomeric protons clear. C H , C C O O H GlcA^r^Mcn Man—Gal 1 K 2 ! m GlcA——Man1—^ Man-—-Glc1-^-a a a j3 4 + -OAc group K 2 4 Z r 4 ~T 6 7 3 9 T Figures 1-2 P.m.r. spectra (100 MHz, x scale) of Klebsiella capsular polysaccharides in D20 at 95' 1. K21, 2. K24 18 The above figures are p.m.r. spectra of K21 and K24 Klebsiella capsular polysaccharides. In K21, the anomeric signals appear at T 4.65, (4.9 + 4.95), 5.05 and 5.4 amounting to 1:2:1:1 indicating a penta-saccharide repeating unit. The signal at T 5.4 shows a distinct doublet with coupling constant of 7 Hz. This i s evidence of a B-linkage of the D-galactose moiety, while a l l the other anomeric signals are at lower f i e l d and have small coupling constants ( ct-linkages) . These results are i n agreement with those obtained from optical rotation. In the case of K24, signals at T 4.6, 4.95 (5.27 + 5.31) i n a ratio of 2:1:2 also show a pentasaccharide repeating unit. The signals at T (5.27 + 5.31) show the presence of a B-linked mannose unit whose coupling constant i s very small, together with a B-linked glucose, having a coupling constant of 6 Hz. A l l other signals are at low f i e l d and have small coupling constants indicating they are a l l a-linkages. P.m.r. : . is currently used as a routine screening process from which the polysaccharide samples may be recovered unchanged. Together with the results obtained from spectra of the derived oligosaccharides, the anomeric linkages can be assigned without d i f f i c u l t y . H. Detection of Other Functional Groups It is interesting to note that pyruvic acid i s present as a 4,6-ketal i n more than one third of Klebsiella polysaccharides. H K 4,6-0-(l-carboxyethylidene)-1 9 Also present i s the less common O^-acetyl group, since a few of the sugar hydroxyl groups may be esterified. Pyruvic acid and acetate can be identified by hydrolysing the polysaccharide. Pyruvic acid i s then 35 characterized by making the 2,4-dinitrophenylhydrazone derivative while 38 the acetate forms a hydroxamate derivative. The above technique is a destructive method which involves total hydrolysis of the polysaccharide. A non-destructive method was designed in our laboratory 35 3 6 again by using the above p.m.r. techniques. ' Pyruvate ketal shows a sharp singlet at x 8.5 while the O-acetyl is at x 7.8. Other •functional groups such as CH^ of 6-deoxyhexose can also be ascertained (doublet at x 8.7). P.m.r. spectroscopy also provides quantitative information on the functional groups present in the polysaccharide. From the figure of K21, the presence of one pyruvate ketal per five sugar units i s indicated. That of K24 shows the presence of an O-acetyl group and the absence of pyruvate. The acetate content corresponds to one O-acetyl group per 7 or 8 sugar units. 20 STRUCTURES INVESTIGATED The object of the present investigation i s to isolate and study the structures of several Klebsiella capsular polysaccharides. The capsular polysaccharides studied in this thesis are from Klebsiella K-type 20, K-type 21, K-type 24 and K-type 56 whose structures are given as follows: K20 2 1 3 1 — D-Manp ^ — ^ D-Galp —r = . j- a = - t a 1. D-Galp D-GlcAp + -OAc group K21 3 1 3 1 2 1 3 1 ~ "IcAp ±-2- D-Manp ±-=- D-Manp D-Galp — , - a = - a . = - a = - t D-G; a 1 D-Galp CH3-C-C00H K24 2 1 3 12 1 3 1 — D-GlcAp D-Manp ^ - ^ D-Manp ^ - ^ D-Glcp , - a = - a = L a = - f 4 6 1 D-Manp + -OAc group K56 3 1 3 1 3 1 3 1 — D-Glcp D-Galp — - D-Galp ^ D-Galp CH3-C-C00H 1 L-Rhap >*- a = 21 The structures found for K21, K24 and K56 are similar in the sense that they each have a repeating unit of five sugars including a single unit side chain. The structure of K20 is different from the above in that i t has a repeat of four sugars including a two unit side chain and is the f i r s t investigation of a Klebsiella polysaccharide which contains an aldobiouronic acid side chain. K56 represents the f i r s t investigation of these capsular polysaccharides lacking a uronic acid. Structures of the capsular polysaccharides from K-type 20 and 56 are reported in Part I and II of the thesis. The work on the capsular polysaccharides K21 and K24 has already been reported in the Canadian Journal of Chemistry and Xerox copies are included as appendices. 22 BIBLIOGRAPHY 1. J.F. Wilkinson, Bact. Rev., 22, 46 (1956). 2. M. Stacey and S.A. Barker, Polysaccharides of Micro-organisms, Oxford, 1960. 3. 0. Liideritz, K. Jann and R. Wheat, Comprehensive Biochem., 26A, 105 (1968). 4. W. Nimmich, Z. Med. Mikrobiol. Immunol., 154, 117 (1968). 5. W. Nimmich, Acta b i o l . med. germ., 2h_, 397 (1971). 6. R.W. Wheat, C. Dorsch and G. Godoy, J . Bacteriol., 89, 539 (1965). 7. B.J. Gormus, R.W. Wheat and J.F. Porter, J . Bacterid., 107, 150 (1971), 8. L.C. Gahen, P.A. Sandford and H.E. Conrad, Biochemistry, t5, 2755 (1967) 9. I.W. Sutherland, Biochemistry, 9, 2180 (1970). 10. P.A. Sandford and H.E. Conrad, Biochemistry, _5, 1508 (1966). 11. I.W. Sutherland, Biochem. J . , 104, 278 (1967). 12. J.E. Scott, Chem. & Ind. (London), 1568 (1955). 13. P. Albersheim, D.J. Nevins, P.D. English and A. Karr, Carbohydr. Res., 5_, 340 (1967). 14. . S.M. Partridge, Nature, 158, 270 (1946). 15. S.M. Partridge and R.G. Westall, Biochem. J. , 42_, 238 (1968). 16. N. Dubois, J.K. Hamilton, K.A. G i l l e s , P.A. Rebers and F. Smith, Anal. Chem., 28, 350 (1956). 17. C C . Sweeley, R. Bentley, M. Makita and W.W. Wells, J . Amer. Chem. Soc, 85, 2497 (1965). . 18. J.S. Sawardeker, J.H. Sloneker and A. Jeanes, Anal. Chem., 37, 1602 (1965). 23 19. G.M. Bebault, J . Berry, Y.M. Choy, G.G.S. Dutton, N. Funnell, L.D. Hayward, and A.M. Stephen, Can. J . Chem., 5_1, 324 (1973). 20. S.I. Hakomori, J . Biochem. (Tokyo), 55_, 205 (1964). 21. H. Bjorndal, B. Lindberg and S. Svensson, Acta Chem. Scand., 21, 1801 (1967). 22. H. Bjorndal, C.G. Hellerqvist, B. Lindberg and S. Svensson, Angew. Chem. internat. Edit., 9, ' 610 (1970). 23. G.G.S. Dutton and Y.M. Choy, Carbohydr. Res., 21, 169 (1972). 24. Y.M. Choy, G.G.S. Dutton, K.B. Gibney, S. Kabir and J.N.C. Whyte, J . Chromatog., 72, 13 (1972). 25. G.M. Bebault, G.G.S. Dutton and R.H. Walker, Carbohydr. Res., J23, 430 (1972). 26. W. Heyns and H. Scharmann, Liebigs Ann. Chem., 667, 183 (1963). 27. O.S. Chizhov, N.V. Moldostov and N.K. Kochetkov, Carbohydr. Res., 4, 273 (1967). 28. G. Pettersson, 0. Samuelson, K. Anjou and E. von Sydow, Acta Chem. Scand., 21, 1251 (1967). 29. H. Bjorndal, B. Lindberg and S. Svensson, Carbohydr. Res., 5_, 433 (1967). 30. S. Allen, T.J. Bonner, E.J. Bourne and N.M. Saville, Chem. & Ind. (London), 630 (1958). 31. I.J. Goldstein, G.W. Hay, B.A. Lewis and F. Smith, Methods Carbohydr. Chem., 5_, 361 (1965). 32. C. Galanos, 0. Liideritz and K. Himmelspach, Eur. J . Biochem., >3, 332 (1969). 33. S.C. Churms, Advances Carbohydrate Chem., 25, 13 (1970). 34. L.D. Ha l l , Advances Carbohydr. Chem., 19, 51 (1964). 24 35. Y.M. Choy, G.G.S. Dutton, A.M. Stephen and M.T. Yang, Anal. Lett., 5_, 675 (1972). 36. G.M. Bebault, Y.M. C hoy, G.G.S. Dutton, N. Funnell, A.M. Stephen, and M.T. Yang, J . Bacteriol., March (1973). 37. H.T. Openshaw, Qualitative organic analysis, Cambridge, 1946. 38. M. Abdel-Aker and F. Smith, J . Amer. Chem. S o c , 73., 5859 (1951). PART I THE STRUCTURE OF THE CAPSULAR POLYSACCHARIDE FROM KLEBSIELLA K-TYPE 20 26 SUMMARY Methylation, periodate oxidation and partial hydrolysis studies on the capsular polysaccharide, and on the carboxyl reduced polymer, of Klebsiella K20 show the structure to consist of a repeating unit. 2 1 3 1 — D-Manp D-Galp — a 1 D-Galp 3~ 1 D-GlcAp The anomeric linkages were determined by p.m.r. spectroscopy of the carboxyl reduced polysaccharide and periodate degraded poly-saccharides. The p.m.r. spectroscopy of the original polysaccharide also showed the presence in the polysaccharide of one O-acetyl group per 8 sugar residues. 27 DISCUSSION For a l l the Klebsiella capsular polysaccharides so far published, the structures are composed of a repeating unit involving a single sugar side chain. The structure of K~type 20 is the f i r s t of these capsular polysaccharides to be encountered with an aldobiouronic acid side chain. A preliminary report"'" has appeared which gives other background references. A culture of Klebsiella K20 (889/50) was obtained from Dr. I. 0rskov, Copenhagen as an agar slant and grown on sucrose yeast-extract agar. Cells were harvested after 3 days, diluted with water containing 1% phenol, and centrifuged at 60,000 x g for 30 minutes. The clear supernatant f l u i d was poured into ethanol and the product was purified 2 by Cetavlon precipitation. The precipitated acidic polysaccharide was recovered and the small amount of neutral material in the supernatant was discarded. The polysaccharide purified by Cetavlon, deionization and freeze 20 drying had [ct]^ +94° (c 0.21, water) and a neutralization equivalent (by titration) of 701. One hexuronic acid to three hexose units requires a neutralization equivalent of 661. The p.m.r. spectrum of a 2% solution of the sodium salt of the polysaccharide in B^O run at room temperature showed a singlet at T 7.8 3 4 characteristic of 0-acetyl and the absence of pyruvate. ' Determination of the acetyl:sugar ratio was hampered by the presence of the large HOD signal due to the high viscosity. However integration of the ring protons and the acetate signal indicated one acetyl group to approximately eight sugar units. The presence of 0~acetyl groups was also confirmed by 28 formation of the hydroxamic ester. The attempt to locate the O-acetyl group using the procedure of de Belder and Norrman^ was unsuccessful due to the very low yi e l d , probably because of the high molecular weight (5-9 x IO"') of the Klebsiella capsular polysaccharide. Acid hydrolysis of the polysaccharide showed the rapid liberation of an aldobiouronic acid and after 2 hours at 100° with 2 M trifluoro-acetic acid (TFA) D-mannose, D-galactose and an aldobiouronic acid were shown by paper chromatography. Separation of the hydrolysate into neutral and acidic fractions, either by ion exchange resin or via the 20 barium salt of the acid, gave an acidic component having [ctl^ +17° (c 0.75, water) and a neutralization equivalent of 349 ('-'12^ 20^ 12 r e <^u^r e s 356). Analysis of the neutral fraction showed D-mannose and D-galactose in a molar ratio of 1:1 as determined by the gas liquid chromatography (g.l.c.) of the a l d i t o l acetates. Collection of the a l d i t o l acetates permitted characterization of D-mannitol hexaacetate, m.p. 118-120° and galactitol hexaacetate, m.p. 1 6 2 - 1 6 4°. A portion of the aldobiouronic acid was dissolved in 0.3 M acetate buffer (pH 4.8) and 8-D-glucuronidase was added, and the solution was incubated at 38° for 6 hours. Paper chromatography showed the formation of D-galactose as the only neutral sugar, together with D-glucuronic acid, and D-glucuronolactone. Similar results were obtained by hydrolysis of the aldobiouronic acid by using 3.5 M hydrochloric acid at 100° for 2 1/2 hours. The configuration of D-galactose was established by D-galactose oxidase, D-glucuronic acid as D-glucose, by D-glucose oxidase and D-mannose by the circular dichroism curve (cd.) of D-mannitol hexaacetate.^ Hydrolysis of the polysaccharide with 0.4 M TFA at 95° for 2 hours 29 gave the aldobiouronic acid as the only component mobile on paper chromatography, thus suggesting the presence of this unit as a side chain. Methylation of the aldobiouronic acid gave on hydrolysis 2,3,4-tri-O-methyl-D-glucuronic acid and 2>4,6-tri-0-methyl-g-galactose demonstrating a 1*3 linkage. A portion of the original capsular polysaccharide was par t i a l l y hydrolyzed at 95° with 0.4 M TFA for 30 hours using the apparatus g described by Galanos, Liideritz and Himmelspach. The dialyzate was evaporated and separated into neutral and acidic fractions. Paper chromatography of the acidic fraction showed the presence of aldobiouronic acid as the major component, with related aldotetrauronic acid together with traces of aldotriouronic acid. The aldobiouronic and aldotetra-uronic acid were separated by preparative paper chromatography. Partial hydrolysis of the aldotetrauronic acid after sodium borohydride reduction yielded g a l a c t i t o l , D-mannose and the aldotri- and aldobiouronic acids present in the original hydrolyzate. The analysis of the aldobiouronic acid showed that i t had the same characteristics as the aldobiouronic acid isolated previously. 9 10 Capsular polysaccharide was methylated by the method of Hakomori ' and hydrolyzed. Analysis of the hydrolysate showed the presence of 2,4,6-tri-O-methyl-D-galactose, 4,6-di-0_-methyl-D-mannose, a methylated aldobiouronic acid, and a small quantity of 2,3,4-tri-O-methyl-D-glucuronic acid. The acidic components were isolated using ion exchange resin and hydrolyzed with 3.5 M hydrochloric acid. 2,4,6-Tri-O-methyl-D-galactose was identified as the only neutral sugar, together with 2,3,4-tri-O-methyl-D-glucuronic acid, while the hydrolysis of the 30 carboxyl reduced acidic component gave 2,3,4-tri-O-methyl-D-glucose and 2,4, 6-tri-C^-methyl-D-galactose. Another sample of the ful l y methylated polysaccharide was reduced with lithium borohydride and hydrolyzed. Quantitative and qualitative analysis of the derived a l d i t o l acetates by gas-liquid chromatography showed the presence of 2,4,6-tri-O-methyl-D-galactose, 2,3,4-tri-0-methyl-D-glucose, and 4,6-di-0-methyl-D-mannose in a molar ratio of 2:1:1. In the above cases the identity of the methylated sugars was confirmed by 11 12 g.l.c. and mass spectrometry (m.s.) ' of the derived a l d i t o l acetates by comparison with standard sugars and demethylation and reacetylation to 13 parent sugar acetates. These results show that the aldobiouronic acid is 3-0-(B-D-glucopyranosyluronic acid)-D-galactose which is joined to the main chain where the D-galactose units are substituted at position 3, and the D-mannose units at positions 2 and 3. In order to determine the position of attachment of the aldobio-uronic acid side chain to the main chain the degraded polysaccharide obtained from partial hydrolysis as described above was methylated and hydrolyzed. Examination of the hydrolysate showed the presence of 2,3,4,6-tetra-O-methyl-D-mannose, 2,3,4,6-tetra-0-methy1-D-galactose, 3,4,6-tri-O^methyl-D-mannose, 2,4,6-tri-i0-methy 1-D-galactose and 4,6-di-O-methyl-g-mannose in a ratio of 2:3:9:12:1. The disappearance of 4,6~di-0-methyl-D-mannose and the appearance of 3,4,6-tri-O-methyl-D-mannose indicates that the aldobiouronic acid units are attached to position 3 of D-mannose which in turn must be linked through position 2 in the main chain. 31 The carboxyl reduced polysaccharide was prepared exactly as 14 described for sapote gum using lithium borohydride in tetrahydrofuran. It was then hydrolyzed and the qualitative and quantitative analysis of the hydrolysate showed the presence of D-galactose, D-glucose and D-mannose in a ratio of 2:1:1. The p.m.r. spectrum run in D^O at 95° showed signals at T (4.78 + 4.86), 5.00 and 5.45 i n the ratio 2:1:1 (Table 1). The signals at x 5.45 showed a distinct doublet with a coupling constant of 7 Hz, while that at T 5.00 gave a broad signal of 6 Hz wide 'whichdisappeared when D-glucose was removed by the Smith periodate degradation (discussed l a t e r ) . The methylation data in conjunction with the results of analysis of the acidic oligosaccharides enable the repeat unit of Klebsiella K20 capsular polysaccharide to be written thus: 2 1 3 1 — D-Manp D-Galp — 3 1 D-Galp 3 1 D-GlcAp Confirmation of the above structure was sought by the Smith periodate degradation procedure. Carboxyl reduced polysaccharide was oxidized with periodate when 0.57 mole per hexose unit was consumed. The derived polyaldehyde after reduction with sodium borohydride was partially hydrolyzed by trifluoroacetic acid at room temperature to give glycerol and a degraded polysaccharide which was recovered by d i a l y s i s . 32 The [cc]n of the degraded polysaccharide is + 1 2 9° . P.m.r. of the degraded polysaccharide showed signals at T 4.60, 4.73 and 5.40 i n a ratio of 1:1:1. The signals at T 5.40 showed a distinct doublet with a coupling constant of 7 Hz, a l l other anomeric coupling being small. This indicates there are two a-linkages and one B-linkage. The degraded polysaccharide obtained in the above Smith degradation was then methylated and hydrolyzed. Analysis of the partially methylated sugars showed the presence of 2,3,4,6-tetra-O-methyl-D-galactose, 2,4,6-tri-0_-methyl-D-galactose and 4,6-di-0_-methyl-D-mannose in a ratio of 1.05:0.9:1. These results show that the degraded polysaccharide consists of a D-galactose unit substituted at position 3 and a D-mannose unit substituted at positions 2 and 3 together with a D-galactose residue present as a side chain. The degraded polymer was again subjected to a second Smith degrada-tion in which case 0.8 mole of periodate was consumed per hexose unit. After the usual work up as above only glycerol was found together with 20 a new degraded polysaccharide having [a]^ + 5 7 ° . P.m.r. showed signals at T 4.80 and 5.48 in a ratio of 1:1. The signals at T 5.48 consist of a doublet with coupling constant of 6 Hz while that at T 4.80 has a very small coupling constant. This shows that the new polysaccharide has one a-linkage and one B-linkage. The new polymer was again methylated and hydrolyzed. The analysis of the par t i a l l y methylated sugars indicates the presence of 3,4,6-tri-0_-methyl-D-mannose and 2,4,6-tri-0_-methyl-D-galactose in a ratio of 1:1. This result shows that the new degraded polysaccharide consists of a chain of D-mannose units substituted at position 2 and D-galactose units substituted at position 3. 33 Together with the results obtained previously, the D-galactose unit which is present as a side chain is joined to position 3 of D-mannose which in turn is linked through position 2 in the main chain. In order to determine the sequence of the main chain, the new degraded polysaccharide obtained from the above experiment was again subjected to a third Smith degradation. After the usual work up, glycerol and an oligosaccharide were isolated, the oligosaccharide when reduced with sodium borohydride had R .. 1.05 in solvent B and when hydrolyzed G a l glycerol and D-galactose were obtained in approximately equimolar amount. This finding indicates that the main chain consists of D-galactose units linked (1-K3) and these were flanked by (l->-2) linked D-mannose units, for example 2 1 3 1 — D-Manp ^ - ^ D-Galp = - a = - t The evidence presented shows clearly that the structure of the capsular polysaccharide of Klebsiella K-type 20 is as given previously. The structure found for K20 is different to those previously reported for K2, K8, K9, K21, K24, and K54 and K56 i n that they each have a repeating unit of four or five sugars including a single unit side chain, while that of K20 has a repeat of four sugars including a two unit side chain. 3 4 EXPERIMENTAL General Methods Paper chromatography was carried out by the descending method using Whatman No. 1 paper and the following solvent systems (v/v): (A) ethyl acetate-acetic acid-formic acid-water (18:3:1:4); (B) ethyl acetate-pyridine-water (4:1:1); (C) 1-butanol-acetic acid-water (2:1:1); (D) butanone-water azeotrope + 1% aqueous ammonia; (E) 1-butanol-ethanol-water (4:1:5). Chromatograms were developed with p-anisidine 16 17 trichloroacetate spray or with silver nitrate. ' The cla r i t y of the colors on chromatograms of methylated sugars developed with the former spray is greatly improved by washing the paper under running hot water after heating the chromatogram at 100-110° in the normal manner. Gas-liquid chromatography was carried out on a F and M model 720 dual column instrument f i t t e d with thermal conductivity detectors. The helium flow was 60-80 ml/min with the following .columns: (a) 3% ECNSS-M on Chromosorb W (8 f t x 0.25 in); (b) 5% butanediol succinate on Diatoport S (4 f t x 0.25 i n ) . Circular dichroism spectra were run on a Jasco J-20 automatic recording spectropolarimeter using a quartz c e l l with path length 0.1 cm. Optical rotations were measured at 22 + 2° on a Perkin Elmer model 141 polarimeter. P.m.r. spectra were run on Varian T60 or XL100 instruments. Samples were prepared by dissolving in B^O and freeze drying 2 or 3 times before taking spectra in D20. Tetramethylsilane was used as an external standard. 18 Mass spectra, obtained on individual fractions collected from the gas chromatograph, were run on an MS 902 instrument at 70 eV. The 35 position of methoxyl substitution was determined using the data of 12 Lindberg and coworker, and, where possible, by comparison of mass spectra with those of authentic compounds. Methylation of polysaccharides and oligosaccharides was carried 9 10 out by the method of Hakomori. ' Partially methylated a l d i t o l 19 acetates were demethylated by reaction with boron trichloride. A l l solutions were concentrated on a rotary evaporator under reduced pressure at 40°. D-Glucostat and D-Galactostat reagents were obtained from the Worthington Biochemical Corporation, and 8-D-glucuronidase from Sigma Chemical Company. Isolation and Properties of K20 Capsular Polysaccharide A culture of Klebsiella K20 was obtained from Dr. 0rskov, Copenhagen and was grown on the following medium for 3 days at 2 5 ° : 8 g NaCl, 4 g ^HPO^, 1 g MgS04.7H20, 2 g CaC03> 120 g sucrose and 8 g Bacto Yeast extract in 4 l i t r e s of water. The slime-was--OTllec-ted-afterj 3"days, diluted with water containing 1% phenol, and centrifuged at 60,000 x g for 30 minutes. The clear supernatant f l u i d was poured into ethanol, and the recovered polysaccharide was purified by Cetavlon precipitation. The yield of purified polysaccharide was approximately 3 g per 4 1. of medium and two batches were collected and used for the structural studies of the polysaccharide. The purified polysaccharide was dissolved i n d i s t i l l e d water, deionized with Amberlite IR 120 resin, dialyzed and freeze-dried. The 20 product had [a] +94° (c 0.21, HO), N 0%. The equivalent weight of the 36 polysaccharide was found to be 701 as determined by t i t r a t i o n with 0.03 M sodium hydroxide. The above neutralized polysaccharide was lyophilized. The residue was dissolved in B^O and exchanged twice by lyophilization. The p.m.r. spectrum of a 2% solution i n V^O showed a peak at x 7.80 indicating the 4 presence of acetyl group. Due to the high viscosity of this solution, the magnitude of the HOD peak was so great that d i f f i c u l t y was encountered in determining a ratio between the anomeric protons and the acetyl group, however, integration of the ring protons and the acetyl group showed there is approximately one acetyl per 8 sugar units. The presence of acetate was also confirmed by methanolysis of the polysaccharide and identification of the methyl acetate using the hydroxyl-5 amine-ferric chloride reagent. Attempted location of the O-acetyl group using the procedure of de Belder and Norrman was unsuccessful due to the extremely low yield involved in the reaction with methyl vinyl ether. Hydrolysis of the Polysaccharide Polysaccharide (150 mg) was hydrolyzed with 2 M trifluoroacetic acid (TFA) at 100° for 2 hours. After evaporation, the hydrolysate was found, by paper chromatography in solvent A, to contain D-mannose, D-galactose and a component with the mobility of an aldobiouronic acid (R~ 0.26). Gax The hydrolysate was separated into neutral and acidic fractions using ion-exchange resins (Amberlite IR 120 and Duolite A4). The neutral fraction was analyzed and found to contain D-mannose and D-galactose in a 1:1 ratio as determined by the g.l.c. of the a l d i t o l acetates (column a). G.l.c. of the acetates allowed recovery of the individual compounds 37 and identification of D-mannitol hexaacetate, m.p. 1 1 8 - 1 2 0°, and galactitol hexaacetate, m.p. 1 6 2 - 1 6 4°. A portion of the neutral fraction of the hydrolyzate was tested with D-Galactostat reagent and a positive response confirmed the D-configuration of galactose. The D-configuration of mannose was found by dissolving part of the mannitol hexaacetate collected by g.l.c. in acetonitrile. A positive circular dichroism curve, identical to that given by a standard sample, confirmed the ID-configuration of mannose. 20 The acidic component (60 mg) was found to have [ct]^ +17° (c 0.75, H^ O) and an. equivalent weight of 349 when titrated with 0.03 M sodium hydroxide (C22H20°12 r e clu i r e s 356). The acidic fraction had R^  0.16 in solvent C. A portion (20 mg) of the aldobiouronic acid was dissolved in 3 ml of 0.3 M acetate buffer (pH 4.8), 3 mg of 8-D-glucuronidase was added, and the solution was incubated at 38° for 6 hours. D-Galactose was found, by paper chromatography in solvent B, to be the only neutral sugar present, whereas the presence of D-galactose, D-glucuronic acid, and D-glucuronolactone was demonstrated in solvent A. The same results were obtained by hydrolysis of the aldobiouronic acid (10 mg) by using 3.5 M hydrochloric acid at 100° for 2.5 hours. The aldobiouronic acid (20 mg) was methylated and hydrolyzed with 3 M hydrochloric acid at 95° for 3.5 hours. Paper chromatography in solvent E showed the presence of 2,4,6-tri-O-methyl-D-galactose (R^ 0.66) and 2,3,4-tri-O-methyl-D-glucuronic acid (Rf 0.22). The identity of 2,4,6-tri-O-methyl-D-galactose was confirmed by g.l.c.-m.s. of the a l d i t o l 12 acetate. Another portion of the polysaccharide (20 mg) was hydrolyzed with 0.4M TFA at 95° for 2 hours. The aldobiouronic acid was found to be the only mobile compound on paper chromatography in solvent A. 38 Partial Hydrolysis of Polysaccharide K20 polysaccharide (1 g) was partially hydrolyzed with 0.4 M TFA for 30 hours at 95° using the apparatus described by Galanos, Liideritz 8 and Himmelspach. The degraded polysaccharide in the dialysis sac was kept for the methylation study discussed later. The dialysate was evaporated and separated into neutral and acidic fractions by ion exchange resin. On paper chromatography (solvent C) the acidic fraction was shown to contain an aldobiouronic aaid (R n 0.45) as the major Gax component together with aldotetrauronic acid (R_ 1 0.08) and trace amount of aldotriouronic acid (R„ , 0.30). The acids were separated by Gal paper chromatography (solvent C) and analyzed as follows:-The aldobiouronic acid (80 mg) was found to have identical characters as the aldobiouronic acid isolated previously. P.m.r. spectrum showed signals at T 4.75 and x 5.40 in a ratio of 0.5:1.4 (table 1). The signal at x 5.40 showed a distinct doublet with a coupling constant of 7 Hz. The aldotetrauronic acid (50 mg) had an equivalent weight of 650. The acid (10 mg) was hydrolyzed with 1 M TFA at 95° for 1 hour. Paper chromatography (solvent A) showed the presence of D-galactose as the principle monosaccharide with traces of D-mannose and a series of aldobiouronic, aldotriouronic and aldotetrauronic acids which had the same mobilities as the components of the original mixtures. The aldo-tetrauronic acid (10 mg) was reduced with sodium borohydride and, after removal of the borate, the product was hydrolyzed with 2 M TFA for 1/2 hour. On paper chromatography (solvents A and B), the neutral compound found was galactitol which is confirmed by g.l.c of the a l d i t o l acetate, 3 9 m.p. 1 6 2 - 1 6 4°. Also produced were the same aldobio-aldotrio-and aldotetrauronic acids (solvent C) described above. Methylation Analysis of the Polysaccharide Capsular polysaccharide (90 mg) was methylated by the method of 9 10 Hakomori ' The product was dialyzed against running water overnight and then extracted with chloroform. Concentration of the chloroform solution yielded a product which showed no absorption at 3600 cm The methylated polysaccharide (20 mg) was hydrolyzed with 2 M TFA at 100° for 2 hours. Analysis of the hydrolysate by paper chromatography (solvent E) showed the presence of 2,4,6-tri-O-methyl-D-galactose (R^ 0.66), 4,6-di-0_-methyl-D-mannose (R^ 0.59), a methylated aldobiouronic acid (R^ 0.35), and a small quantity of 2,3,4-tri-0_-methyl-D-glucuronic acid (R^ 0.22). The hydrolysate was separated into neutral and acidic fractions using ion exchange resin (Duolite A4). On paper chromatography, the neutral fraction was shown to contain 2,4,6-tri-Oj-methyl-D-galactose (Rf 0.4, solvent D) and 4,6-di-0-methyl-D-mannose (Rf 0.29, solvent D). It was then reduced with sodium borohydride and after removal of the borate, the product was acetylated i n a sealed tube by using pyridine and acetic anhydride (1:1, 1 ml, 9 5 ° , 15 min). After removal of the solvents, the a l d i t o l acetates were dissolved in a small volume of ethyl acetate and analyzed by g.l.c. on column b at 200° (flow rate 100 ml/min); two peaks were observed. The component eluted f i r s t had a retention time (6.9 min) and mass spectrum identical to authentic 2,4,6-tri-O-methyl-D-galactitol triacetate. The second component had identical characters in retention 40 time (9.4 min) and mass spectrum as authentic 4,6-di-O-methyl-D-mannitol tetraacetate. Demethylation and reacetylation of the partially methylated 13 a l d i t o l acetates, gave the corresponding galactitol hexaacetate, m.p. 166-169° and D-mannitol hexaacetate, 1 1 9 - 1 2 2°. A portion (4 mg) of the acidic fraction was hydrolyzed with 3.5 M HCI for 2 hours at 100°^ after evaporation and neutralization with silver carbonate, the hydrolysate was found to contain 2,4,6-tri-O-methyl-D-galactose (analyzed by g.l.c.-m.s. of the a l d i t o l acetate) as the only neutral sugar, and 2,3,4-tri-C^-methyl-D-glucuronic acid (R^ 0.22 solvent E). Another portion (5 mg) was methanolyzed with 2% methanolic/HCl and reduced with lithium borohydride. After removal of the borate by IR 120 and evaporations with methanol, the product was hydrolyzed (0.5 M H^SO^ for 4 hours). Paper chromatography showed the presence of 2,3,4-tri-0-methyl-D-glucose (R^ 0.62, solvent D) and 2,4,6-tri-0-methyl-D~galacto se (R^ 0.40, solvent D). G.l.c. of the a l d i t o l acetates showed two peaks at 9.5 min and 8.7 min (column b, 200° flow rate 60 ml/min) which were identical i n retention time and mass spectrum to the corresponding authentic a l d i t o l acetates. Another sample of methylated polysaccharide (20 mg) was reduced with lithium borohyride and hydrolyzed. Quantitative analysis of the derived a l d i t o l acetates by gas-liquid chromatography (3% ECNSS-M, 180° to 217° at 2°/min) showed the presence of 2,4,6-tri-0_-methyl-g-galactose (24.5 min), 2,3,4-tri-0-methyl-D-glucose (26 min), and 4,6-di-0-methyl-D-mannose (29.8 min) in a molar ratio of 2:1:1. The identity of the methylated sugars was again confirmed by mass spectrometry of the derived a l d i t o l acetates. * correct nomenclature: 1,3-di-0-methyl-D-mannitol tetraacetate 41 Methylation of Degraded Polysaccharide A portion (120 rng) of the degraded polysaccharide recovered from the partial hydrolysis was methylated and the product was hydrolyzed with 2 M TFA at 100° for 4 hours. After evaporation, the hydrolysate was shown by paper chromatography to contain 3,4,6-tri-O-methyl-D-mannose (Rf 0.58, solvent D) and 2,4,6-tri-O-methyl-D-galactose (Rf 0.38, solvent D) as the major components, together with minor compounds of 2,3,4,6-tetra-0^-methyl-g-galactose and D-mannose and faint traces of methylated aldobiouronic acid (solvent E). The hydrolysate was reduced with sodium borohydride, acetylated and injected into column b at 1 9 0° . 2,3,4,6-Tetra-0_-methyl-D-mannose (retention time 6.5 min), 2,3,4,6-tetra-O-methyl-D-galactose (8.0 min), 3,4,6-tri-0_-methyl-D-mannose (12.0 min), 2,4,6-tri-O-methyl-D-galactose (16.0.min) and 4,6-di-0_-methyl-g-mannose (18.0 min) was found in a ratio of 2:3:9:12:1. The mass spectrum of 3,4,6~tri-0-methyl-D-mannitol acetate was again identical with-an authentic sample. Reduction of the Capsular Polysaccharide Capsular polysaccharide (0.5 g) was converted into the methyl ester 14 propionate and reduced with lithium borohyride to give the reduced polysaccharide (0.4 g), [a]D +95.5° (c 2, water). The reduced polysaccharide (20 mg) was hydrolyzed with 2 M TFA at 100° for 4 hours. On analysis of the hydrolysate by paper chromatography and g . l . c , D-galactose, g-glucose and D-mannose were shown to be present in a ratio of 2:1:1 as determined by g.l.c. of their a l d i t o l acetates. 42 Galactitol hexaacetate had m.p. 166-169°, g-glucitol hexaacetate had m.p. 96-98° and D-mannitol hexaacetate had m.p. 1 1 9 - 1 2 2°. The hydrolysate gave a positive test with D-P;lucostat reagent, thus confirming the D-configuration of D-glucuronic acid. The p.m.r. spectrum of the reduced K20 polysaccharide was run i n Ji^O a t 95° and was shown to give anomeric signals at T (4.78 + 4.86), 5.00 and 5.45 in the ratio of 2:1:1 (Table 1). The signals at x 5.45 showed a distinct doublet with a coupling constant of 7 Hz, while that at T 5.00 was 5-6 Hz wide and showed no distinct s p l i t t i n g . Smith Periodate Degradations and Methylation Studies of the Degraded Polysaccharides  The carboxyl reduced polysaccharide (66 mg) was dissolved in 10 ml water then 20 ml 0.05 M sodium metaperiodate were addedi. After 66 hours, 0.57 mole of periodate per hexose unit was consumed. Ethylene glycol (1.5 ml) was added and the solution was l e f t at room temperature for 1 hour to destroy excess periodate. The product was then dialyzed overnight against running water. Sodium borohydride (0.5 g) was added and the solution was l e f t overnight. The solution was deionized with Amberlite IR 120, freeze-dried, and the product was d i s t i l l e d with several portions of methanol. The derived polyalcohol was dissolved in 5 ml 0.5 M TFA and l e f t at room temperature overnight. On paper chromatography (solvent A) only glycerol was identified (R^ 0.45). The product was dialyzed against 2 1. of d i s t i l l e d water. The dialysate when evaporated was found to contain glycerol by paper chromatography and g.l.c. (retention time of glycerol triacetate was 6.4 min, 1 5 0° , column b). 43 The degraded polysaccharide remaining in the dialysing sac was freeze-20 dried. It had [ct]^ +129° .The.p-.m,rspectrum of the degraded polysaccharide in D20 showed signals at x 4.60, 4.73 and 5.40 i n a ratio of 1:1:1. The signals at x 5.40 showed a doublet with coupling constant of 7 Hz, a l l other coupling constants being small. The degraded polysaccharide was thai methylated by the Halcomori method and hydrolyzed. Paper chromatography showed the presence of 2,3,4,6-tetra-0~methyl-D-galactose (R^ 0.70 solvent D), 2,4,6-tri-O-methyl-D-galactose (Rf 0.40), and 4,6-di-O-methyl-D-mannose (R^ 0.29). The ratio was found to be 1.05:0.9:1 by g.l.c. of the a l d i t o l acetates. 2, 3,4 ,6-Tetra-0_-methyl-D-galactitol acetate had a retention time of 13 min (column b, 160-200° at 2°/min) and had a mass spectrum identical to an authenic sample. Another portion of the carboxyl reduced polysaccharide (0.3 g) was oxidized by periodate as described previously to yield 0.2 g of the degraded polysaccharide (S^). To this degraded polysaccharide S^, 50 ml of 0.04 M sodium periodate was added. Periodate consumption was found to be 0.8 mole per hexose unit after 2 days. . The derived polyaldehyde after reduction with sodium borohydride was partially hydrolyzed with 0.4 M TFA at room temperature overnight. Glycerol was found to be the only component mobile on paper chromatography in solvent A (R^ 0.44). The product was dissolved in 5 ml of water and dialyzed against 1 1. of d i s t i l l e d water. The substance remaining in the dialysis sac was freeze-dried to yield 0.1 g of a new degraded polysaccharide (S2) . The 20 degraded polysaccharide S? had [a]Q +57° ( £ 0 . 3 5 , water). P.m.r. of this polysaccharide was run in D,,0 at 95° and showed signals at x 4.80 and 5.48 in a ratio of 1:1. The signals at x 5.48 consisted of a doublet with A 44 coupling constant of 6 Hz while that at T 4.80 had a very small coupling constant. A portion (27 mg) of the degraded polysaccharide was methylated by the Hakomori method and hydrolyzed (2 M TFA, 1 0 0° , 4 hours). Paper chromatography (solvent D) showed the presence of 3,4,6-tri-O-methyl-D-mannose (R^ . 0.57) and 2,4,6-tri-Oj-methyl-g-galactose (R^ 0.38). The ratio was found to be 1:1 by g.l.c. of the a l d i t o l acetates. 3,4,6-Tri-Oj-methyl-D-mannose had an identical retention time (22 min, column''.' b, 180-200° at 2°/min) and mass spectrum to an authentic sample. The degraded polysaccharide (43 mg) was further degraded by periodate. After the usual work up, the derived polyalcohol was hydrolyzed with 0.5 M TFA at room temperature for 6 hours. The product was then reduced with sodium borohydride. After removal of the borate, paper chromatography in solvents B and C showed the presence of glycerol (R 4.3, solvent B, R 2.1, solvent C) and an unknown component «ai b-ax (R 1 1.05, solvent B, R .. 1.2, solvent C). The unknown component Gal Gal was separated by paper chromatography in solvent B, and then hydrolyzed with 2 M TFA for 2 hours. On paper chromatography in solvents A and B, D-galactose and glycerol (R_ ., 4.6, solvent A; R .. 2.4, solvent B) — Gal Gal were identified. The product was reduced with sodium borohydride and acetylated. G.l.c. of glycerol triacetate (retention time 8 min, column a, 140-200°) and galactitol hexaacetate (retention time 58 min) were found in approximate equimolar amount. 45 Table 1. P.m.r. data on Klebsiella K20 capsular polysaccharide and derived polysaccharides and oligosaccharide Repeating unit of compound T value (coupling constant,Hz) Ratio of integral Proton assignment Man . 3 a 1 Gal 3 3 1 Glc - ^ G a l 4.78 + 4.86 5.00 (§) 5.45 (7) a-Gal 1 3 Man a-Man 1 3 Gal 3-Glc 1 3 Gal 8-Gal 1 2 Man 2 « 1 3 n i 1 — Man Gal 3 a i a 1 Gal 2 vr 1 3 n 1 1 — Man Gal — — a B GlcA i-T-3- Gal-OH 4.63 4.73 5.40 (7) 4.80 5.48 (6) 4.75 5.40 (7) 1 a-Gal 1 3 Man 1 a-Man 1 3 Gal 1 (3-Gal 1 2 Man 1 a-Man 1 3 Gal 1 B-Gal 1 2 Man 0.5 1.4 a-Gal-OH B-Gal-OH B-GlcA — Gal Spectra run in D^ O with external tetramethylsilane (T = 10) at 100 MHz. A* A l l sugars have D-configuration and are pyranose. (§) signal i s 5-6 Hz wide; shows no distinct s p l i t t i n g . 46 Table 2. Methyl ethers from the hydrolysis of methylated Klebsiella K20 polysaccharides. Sugars 2,3,4,6-Tetra-O-methyl-D-mannose 2,3,4,6-Tetra-0_-methyl-D-galactose 2,4,6-Tri-O-methyl-D-galactose 3,4,6-Tri-O-methyl-D-mannose 2,3,4-Tri-O-methy1-D-glucuronic acid 2,3,4-Tri-O-methy1-D-glucose 4,6-Di-0_-methyl~p-mannose Samples  A B C D E+ F+ G+ H 2 3 1 + + + + 12 2 1 1 9 1 + + + 1 + 1 1 1 A, neutral sugars from methylated original K20 polysaccharide; B, aldobiouronic acid fraction from methylated original K20 polysaccharide; C, reduced aldobiouronic acid fraction from methylated original K20 polysaccharide; D, methylated aldobiouronic acid; E, neutral sugars from methylated residual polysaccharide after partial hydrolysis; F, sugars from methylated reduced K20 polysaccharide; G, sugars from 1st Smith degraded polysaccharide; H, sugars from 2nd Smith degraded polysaccharide. + Approximate molar ratios. (+) Signifies present and has no quantitative significance. 47 Table 3. Diagnostic prominent peaks (m/e) in the mass spectra of acetates of methylated a l d i t o l s . Parent sugar m/e  43 45 117 129 145 161 189 205 233 261 2s3,4,6-Me4Gal + + + + + + + 2,4,6-Me3Gal + + + + + + 3,4,6-Me3Man + + + + + 2,3,4-Me3Glu + + + + + + 4,6-Me2Man + + + + + 48 BIBLIOGRAPHY 1. Y.M. Choy and G.G.S. Dutton, J . Ba c t e r i d . , 112, 635 (1972). 2. J.E. Scott, Chem. & Ind. (London), 1568 (1955). 3. Y.M. Choy, G.G.S. Dutton, A.M. Stephen and M.T. Yang, Anal. Lett. 5, 675 (1972). 4. G.M. Bebault, Y.M. Choy, G.G.S. Dutton, N. Funnell, A.M. Stephen and M.T. Yang, J . Ba c t e r i d . , March (1973). 5. M. Abdel-Aker and F. Smith, J . Am. Chem. S o c , 73, 5859 (1951). 6. A.N. de Belder and B. Norrman, Carbohydr. Res., j8, 1 (1968). 7. G.M. Bebault, J . Berry, Y.M. Choy, G.G.S. Dutton, N. Funnell, L.D. Hayward and A.M. Stephen, Can. J . Chem., 5_1, 324 (1973). 8. C. Galanos, 0. Liideritz and K. Himmelspach, Eur. J . Biochem. >3, 332 (1969). 9. S.I. Hakomori, J . Biochem. (Tokyo), 55_, 205 (1964). 10. P.A. Sandford and H.E. Conrad, Biochemistry, 5_, 1508 (1966). 11. H. Bjorndal, B. Lindberg and S. Svensson, Carbohydr. Res., _5, 433 (1967). 12. H. Bjorndal, C.G. Hellerqvist, B. Lindberg and S. Svensson, Angew. Chem. internat. Edit., _9_, 610 (1970). 13. G.G.S. Dutton and Y.M. Choy, Carbohydr. Res., 21, 169 (1972). 14. G.G.S. Dutton and S. Kabir, Anal. Lett., 4-, 95 (1971). 15. I.J. Goldstein, G.W. Hay, B.A. Lewis and F. Smith, Methods Carbohydr. Chem., 5_, 361 (1965). 16. L. Hough, J.K.N. Jones and W.H. Wadman, J . Chem. S o c , 1702 (1950). 17. W.E. Trevelyan, D.P. Proctor and J.S. Harrison, Nature, 166, 444 (1950). 18. G.G.S. Dutton and K.B. Gibney, J . Chromatogr. V7\2, 179 (1972). 19. S. Allen, T.G. Bonner, E.J. Bourne and N.M. Saville, Chem. & Ind., (London), 630 (1958). PART II THE STRUCTURE OF THE CAPSULAR POLYSACCHARIDE FROM KLEBSIELLA K-TYPE 56 50 SUMMARY Methylation, periodate oxidation and partial hydrolysis studies of the capsular polysaccharide of K56 show the structure to consist of a repeating unit 3 1 3 1 3 1 3 1 — D-Glcp ^ D-Galp_ D-Galp_ ^ - ^ D-Galp_ - y ~«V* ' ~*\[ CH--C-C00H _ -7 3 L-Rhap_ with a L-rhamnose side chain which joins a(1+2) to the second D-galactose of the main chain. D i f f i c u l t i e s have been encountered in assigning the L-rhamnose side chain to one of the three D-galactoses, however, with the isolation of the partially methylated oligosaccharides, this problem has been solved. The anomeric linkages were determined by p.m.r. spectroscopy of isolated oligosaccharides, and i n part, by specific enzymes. P.m.r. spectroscopy of the original polysaccharide using the sodium salt in D^ O showed clearly a ratio of one pyruvic acid ketal (CHg, x 8.57) to five anomeric protons (T 4.6 to 5.3), while the ratio of pyruvic acid ketal to L-rhamnose -CH^ is 1:1. Of a l l the Klebsiella capsular polysaccharides so far studied ', this i s the f i r s t investigation of the detailed structure of a Klebsiella capsular polysaccharide which does not contain uronic acid. 51 DISCUSSION > L-Rhamnose has been found present as a constituent i n about one-1 2 third of the 80 Klebsiella ' capsular polysaccharides. Capsular poly-saccharide . from K-type 56 consists of one L-rhamnose unit per five sugar units and constitutes the f i r s t report of these Klebsiella polysaccharides lacking uronic acids. The acidity of the polysaccharide is due only to the pyruvic acid present. The polysaccharide after purification by precipitation with Cetavlon, 2 4 had +79° (c 0.43, water) and an equivalent weight (by titration) of 908. The p.m.r. spectrum of a 2% solution of the sodium salt of the polysaccharide in D^ O run at 95° showed a sharp singlet at T 8.57 3 4 characteristic of the CH^ of a pyruvate ketal ' and a doublet at T 8.73 characteristic of the CH^ of a 6-deoxyhexose.^ The ratio of these two signals is 1:1. The anomeric region of the spectrum suggested that the repeating unit ; consists of five sugar units of which three are linked 3 A-by B- and two by ct-glycosidic bonds. ' Integration of the anomeric and pyruvate signals indicated one pyruvate ketal group to five sugar units. Acid hydrolysis of the polysaccharide showed the rapid liberation of L-rhamnosej the amount of which reached a maximum after 1/2 hour with only traces of D-galactose and D-glucose. After 4 hours L-rhamnose, D-galactose and D-glucose were proved to be in the ratio of 1.1:3:1. This analysis was carried out on the derived a l d i t o l acetates. Samples of these were collected by g.l.c. and measurement of their c d . spectra confirmed the assignment of the L-configuration of rhamnose, and D-configuration of glucose while the D-configuration of 52 galactose was assigned by using the a l d i t o l acetate of the partially methylated sugar (see l a t e r ) . T h e presence of L-rhamnose was confirmed by methanolysis of the polysaccharide to give the methyl ct-L-rhamnopyranoside which crystallized easily on seeding. Partial hydrolysis of the polysaccharide gave a series of neutral oligosaccharides which were separated from the monosaccharides by a 7 8 charcoal column ' and individual components were obtained by Sephadex 9 gel chromatography (G-15). These were shown to be a glucosyl galactose disaccharide, together with related t r i - and tetrasaccharides. •The structures of these oligosaccharides are given in Table 1 together with the chemical shifts of the anomeric protons which enable the nature of the glycosidic linkages to be determined. The structures given were determined by (a) hydrolysis with acid and enzyme, (b) partial hydrolysis and (c) methylation, as described in the experimental .^section. A sample of pure capsular polysaccharide was methylated}^ '"'""'" hydrolyzed and the partially methylated sugars were examined. 2,3,4-Tri-0_-methyl-L-rhamnose, 2,4,6-tri-0_-methyl-D-galactose, 4,6-di-0-methyl-D-galactose and 2-0-methyl-D-glucose were found in a molar ratio 12 of 0.85:2.2:1:1 which was determined by g.l.c. of the a l d i t o l acetates. The identities of the individual partially methylated sugars were determined by comparing with standards using (a) paper chromatography, 13 (b) g.l.c.-m.s.- of the a l d i t o l acetates and (c) demethylation to the 14 parent sugar a l d i t o l acetates. These results show the presence of two D-galactose units substituted at position 3. The isolation of 4,6-di-0-methy1-D-galactose indicates that one of the three-galactoses.'•.is\ a branch p o i n t _ I , t : is substitute.dr_at both the 53 2 and 3 positions, with the L-rhamnose residue present as a side chain. The 2-O^methyl-D-glucose found agrees with the fact that D-glucose i s in the main chain substituted at position 3, with pyruvic acid linked to the 4,6-positions as a ketal. In order to determine the position of attachment of the L-rhamnose units to the main chain a sample of the original polysaccharide was subjected to Smith periodate degradation.^ The polysaccharide consumed 0.45 mole of periodate per hexose unit after 64 hours at 4 ° . The derived polyalcohol was partially hydrolyzed with trifluoroacetic acid (TFA) at room temperature to give 1-deoxy glycerol (or 1,2-propanediol) and a residual polysaccharide. The latter was methylated and hydrolyzed to give 2,4,6-tri-O-methyl-D-galactose and 2-C^-methyl-D-glucose in a ratio of 3:1 indicating that the L-rhamnose side chains are joined to position 2 of D-galactose, which in turn must be linked through position 3 in the main chain. The p.m.r. spectrum at 95° of a 2% solution of the sodium salt of the above residual polysaccharide also showed a sharp singlet at x 8.57 indicating that the pyruvic acid 3 4 ketal ' was resistant to the mild conditions of the Smith periodate degradation. Signals from the anomeric protons appeared at x 4.75, 5.09 to.'; 5.21 which integrated for 9 +(10 + 20)in comparison to the pyruvic acid methyl signal of 30 (arbitrary units). This established a ratio of one pyruvic acid to four sugar residues. The signal.at x 5.09 was shifted to higher f i e l d to x 5.24 when the pyruvic acid ketal was removed by 0.1 M TFA. The disappearance of one ct-anomeric signal indicates that the L-rhamnose i s a-linked to the main chain which is consistent with the result obtained from the change i n optical rotation 54 20 of the polysaccharide [ [alD +79° to +97° (c 1.0, water)] as the L-rhamnose side chain is removed. The methylation data in conjunction with the results of analysis of the oligosaccharides and Smith periodate degradation enable the repeat unit of Klebsiella K 5 6 capsular polysaccharide to be written thus: — D-Glc£ "^p D-Galp_ "^p D~Galp_ p p g-Gal£ ^ j- and a 1 4 \J 6 L-Rhap_ CHpC-COOH The L-rhamnose could be linked (1-^ 2) to either one of the three D-galactose units. It is known that rhamnose linkages"*"^ and galactose linked (l->3)^ '7 are very labile to acid hydrolysis. The assign-ment of the position of the side chain from partial hydrolysis of the original polysaccharide is difficult in the present instance since i t is hard to isolate an oligosaccharide containing ^-rhamnose, and also it is hard to distinguish between the three D-galactoses as they a l l have (l->-3) linkages. However, by partial hydrolysis of the methylated capsular polysaccharide, this problem has been solved. The fully methylated capsular polysaccharide was hydrolyzed with 90% formic acid 18 for 45 minutes at 7 0°. On TLG (solvent D) three spots were obtained (Rf 0.78, 0.72 and 0.66) together with 2 faint spots with Rf 0.47 and 0.25-0.30. The component with R^  0.72 had a mobility identical to 2,3,4-tri-O-methyl-L-rhamnose. A mixture of 2,3,4-tri-O-methyl-L-rhamnose, 2,4,6-tri-O-methyl-D-galactose, 4,6-di-O-methyl-D-galactose and 2-0_-methyl-D-glucose was shown by paper chromatography when the minor component with R^  0.78 was reduced and hydrolyzed. This is probably a 55 repeating unit of the polysaccharide with 2,4,6-tri-O-methyl-D-galactose at the reducing end which when reduced i s not detectable on paper chromatography. The major component (R^ 0.66) was separated by TLC i n solvent D. P.m.r. of this fraction showed a strong signal at x 8.45 indicating a 3 4 pyruvate ketal. ' On hydrolysis, 2,4,6-tri-O-methyl-D-galactose, 4,6-di-O-methyl-D-galactose and 2-0--methyl-D-glucose were obtained in a ratio of 1.5:0.8:1 indicating a mixture of oligosaccharides. On reduction with sodium borohydride, this fraction gave two components on TLC with R^  0.14 and 0.07. The faster component gave on hydrolysis 2,4,6-tri-0-methyl-D-galactitol and 2-0-methy1-D-glucose in a ratio of 1.25:1. On hydrolysis of the slower component, 2,4,6-tri-Gj-methyl-D-galactose, 4,6-di-O-methyl-D-galactose, 2-0_-methyl-D-glucose and 2,4,6-tri-O-methyl-D-galactitol were found in a ratio of 1.1:1:1:1.3. Mien this slow component was methylated and hydrolyzed only 2,3-di-O-methyl-D-glucose and 2,4,6-tri-O-methyl-D-galactose were shown by paper chromatography there was no sign of tetramethyl glucose or galactose. These results indicate that the slow component consists of a partially methylated tetrasaccharide with 2,4,6-tri-O-methyl-D-galactose at the reducing end, with 4,6-£-(l-carboxylethylidene)-2-0_-methyl-D-glucose at the non-reducing end. In order to obtain a partially methylated disaccharide, a portion of the above partially hydrolyzed methylated polysaccharide was hydrolyzed for a further period of 4 hours with 0.5 M TFA at 9 5 ° . Paper chromatography (solvent E) showed the presence of 2,4,6-tri-O-methyl-D-galactose, 4,6-di-O-methyl-D-galactose, 2-0-methyl-D-glucose and a 56 new component with 0.79. The new component was separated by TLC (R^ 0.6 solvent D). Hydrolysis and analysis of the par t i a l l y methylated sugars showed the presence of 2,4,6-tri-0_-methyl-D-galactose and 2-0-methyl-D-glucose in a ratio of 1:1. This i s good evidence of a partially methylated disaccharide. When the disaccharide was reduced with sodium borohydride and hydrolyzed, 2-0-methyl-D-glucose was shown by paper chromatography, while the g.l.c. of the a l d i t o l and aldose acetates showed the presence of 2,4,6-tri-0_-methyl-D-galactitol and 2-0-methyl-D-glucose in a ratio of 1.2:1. Together with the results obtained from the disaccharide, the L-rhamnose side chain could be assigned to the second D-galactose. It was not possible to isolate a partially methylated trisaccharide having 4,6-di-0_-methyl-D-galactose as the reducing end probably because this galactose is linked a(l-K3) which is more stable than the 8(l-+3)^ linkage under the above hydrolysis conditions. The sum of these results firmly establishes the structure written above as representing the repeating unit of the capsular polysaccharide of Klebsiella K56 and this i s the f i r s t report of the detailed structure of Klebsiella capsular polysaccharide which does not contain uronic acid. 57 EXPERIMENTAL General methods are as previously described i n Part I. Isolation and Properties of K56 Capsular Polysaccharide Klebsiella K56 (3534/51) was grown in the medium as for K20 and the harvested capsular polysaccharide and cells (o<2 1.) were diluted 2-fold and centrifuged at 27,000 r.p.m. at 4° for 30 minutes. The supernatant was collected and poured with s t i r r i n g into ethanol (5 volumes). The polysaccharide was collected, dissolved in d i s t i l l e d water and 19 purified through precipitation with Cetavlon giving 8 g of pure acidic 24 + polysaccharide having [ct]D +79 (c 0.43, water) as the Na s a l t . After deionization with Amberlite IR 120 and d i a l y s i s , the equivalent weight was found by t i t r a t i o n with 0.01 M sodium hydroxide (phenol-phthalein) to be 908. A sample of the Na+ salt of the polysaccharide was dissolved in D^ O and exchanged twice by lyophilization. The p.m.r. spectrum of the solution (2% in D^ O) was run at 95° and showed a sharp singlet at T 8.57, a doublet at x 8.73 and five protons in the range x 4.6-5.3 (see Table 1). The ratio between the singlet at x 8.57 and the doublet at x 8.73 was shown by integration to be 1:1 which in turn was shown to be in the ratio of 1:5 with the anomeric protons. The presence of pyruvic acid was also confirmed by hydrolysis of the polysaccharide and identification of the pyruvic acid by paper chromatography (solvent A) using jD-phenylenediamine spray which gave a 20 characteristic fluorescence when examined by u.v. 58 Hydrolysis of the Polysaccharide Polysaccharide (50 mg) was hydrolyzed with 2 M TFA at 95° for periods of 1/2 hour, 1 hour, 2 hours, and 4 hours. The liberation of L-rhamnose was found (paper chromatography) to be a maximum at 1/2 hour with no increase on further hydrolysis. After 4 hours, the hydrolysate was evaporated and the ratio of L-rhamnose, D-galactose and D-glucose was found to be 1.1:3:1 by g.l.c. of the alditol acetates (column a); Galactitol hexaacetate had m.p. and mixed m.p. 162-164° and D-glucitol hexaacetate had m.p. and mixed m.p. 98-99°. The configuration of L-rhamnose and D-glucose was determined by measuring the c.d. curves of the alditol acetates.^ The presence of L-rhamnose was confirmed by methanolysis (3% MeOH/HCl, 2 hours) of the polysaccharide (200 mg) and separation by TLC^ (R^ 0.4, solvent D) of methyl a-Jj-rhamnoside which on recrystallization from ethyl acetate had m.p. and mixed m.p. 108-109°. Methylation Capsular polysaccharide (1 g) was methylated by the Hakomori procedure.^ '''"''" The resulting solution was dialyzed against running tap water and extracted 3 times with chloroform. The dried residue was then fractionated by using petroleum ether-chloroform mixture. The fraction soluble in 70:30 petroleum ether-chloroform showed no hydroxyl absorption in the i . r . and was used for analysis'. The fully methylated polysaccharide (85 mg) was hydrolyzed with 2 M TFA at 100° for 2 hours then at 95° for 16 hours. Trifluoroacetic acid was evaporated and the syrup was shown by paper chromatography (solvent D) 59 to contain four components corresponding to 2,3,4-tri-O-methyl-L-rhamnose, 2,4,6-tri-O-methyl-g-galactose, 4,6-di-O-methyl-D-galactose and 2-0-methyl-D-glucose (Rf values 0.88, 0.42, 0.16, 0.07). The ratio was found to be 0.85:2.2:1:1 as determined by the g.l.c. of the alditol acetates which permitted isolation of 2-0-methyl-D-glucitol pentaacetate as a crystalline derivative having m.p. and mixed m.p. 21 13 54-56°. The other three components were confirmed by m.s. of the individual alditol acetates and demethylation (BCl^) and reacetylation 14 to the parent sugar alditol acetates; m.p. of galactitol hexaacetate 164-167°, m.p. of D-glucitol hexaacetate 9 9° . The c d . spectrum of 2,4,6-tri-O-methyl-g-galactitol acetate is identical to a standard sample thus comfirming the D-conf iguration of D-galactose. Smith Degradation Capsular polysaccharide (220 mg) was dissolved in 100 ml 0.025 M sodium metaperiodate. After 64 hours in the dark at 4° 0.45 mole of periodate had been consumed per sugar unit. Following the addition of ethylene glycol, dialysis, reduction with sodium borohydride, dialysis, deionization, lyophilization and removal of borate the product was hydrolyzed (TFA 0.5 M) at room temperature for 8 hours. Paper chromatography in solvent A showed the presence of 1-deoxyglycerol (1,2-propanediol) having identical value (0.80) and retention time (4.2 min on column a at 90°) with authentic standard. The product in 5 ml was dialyzed against 2 1. of distilled water. The residue was freeze dried to yield a degraded polymer (150 mg). A part (70 mg) of 60 the degraded polysaccharide was methylated (Hakomori procedure) and then hydrolyzed. Paper chromatography (solvents D and E) , 2,4,6-tri-0_-methyl-D-galactose and 2-0-methyl-D-glucose were shown to be the only two sugars present and had a ratio of 3:1 when determined by g.l.c. of the alditol acetates. The two components were again analyzed by the m.s. of the alditol acetates and m.p. (2-0-methyl-D-glucitol pentaacetate m.p. 54-56°). 20 The degraded polysaccharide had [ct]^ +97° (c 1.0, water) and 40 mg of the sodium salt of this polysaccharide was exchanged twice with DrfO by lyophilization. The p.m.r. spectrum was run at 95° in V^O (100%) and showed a singlet at T 8.57 while the doublet at x 8.73 was completely absent. The anomeric signals (Table 1) at x 4.75, 5.09-5.21 were found to be in a ratio of 1:3. The ratio between the CH^  of pyruvate ketal to the anomeric protons was found to be 3:4 showing one pyruvic acid per four sugar units. The anomeric signal at x 5.09 was shifted upfield to 5.24 when the pyruvate ketal group was removed by hydrolysis with 0.1 M TFA at 95° for 1 hour. Partial Hydrolysis of the Polysaccharide K56 polysaccharide (1 g) was dissolved in trifluoroacetic acid (TFA 50 ml, 0.5 M) and the solution was heated on a steam bath for 1 hour. The solution was evaporated and the monosaccharides were separated from the oligosaccharides by a charcoal column (Darco G60, 10 x 3 cm). Monosaccharides were eluted by distilled water (2 1.) and a series of oligosaccharides (mainly d i , t r i and tetrasaccharides) was eluted with 20% ethanol-water (1 1.), yield about 0.3 g. The oligosaccharides 61 (100 mg) were applied to a Sephadex G15 column (110 x 2 cm) and irrigated with water at a flow rate of 4-6 ml per hour. The oligo-saccharides were separated into di, t r i and tetrasaccharides (R 1 Gal values 0.52, 0.23, 0.16, solvent C) and analyzed as follows: A. Analysis of disaccharide (40 mg) P.m.r. (Table 1) of the disaccharide when run in D^O at 95° - . showed i t to be g-'linked- , When 10 mg was hydrolyzed with 2 M TFA for 4 hours, D-galactose and D-glucose were found as the only two sugars present in the ratio of approximately 1:1 (as determined by g.l.c. of the alditol acetates). The same ratio was obtained when another portion (10 mg) was dissolved in 5 ml 0.3 M acetate buffer (pH 5.5) to which 10 mg of 3-D-glucosidase was added and incubated at 38° for 16 hours. The disaccharide (10 mg) was methylated and hydrolyzed, 2,3,4,6-tetra-O-methyl-D-glucose and 2,4,6-tri-O-methyl-D-galactose were found by paper chromatography (solvent D) having R^  0.80 and 0.40 respectively together with small amounts of tri-0_-methyl-D-galactofuranose R^  0.59 (solvent D) which was not determined. This showed that D-glucose is the non-reducing end. The ratio of 2,3,4,6-tetra-O-methyl-D-glucose to 2,4,6-tri-O-methyl-D-galactose was found to be 1:0.9 as determined by g.l.c. of the alditol acetates. B. Analysis of the trisaccharide (20 mg) When the p.m.r. spectrum was run in D20 at 9 5° , anomeric protons were obtained at T 5.29 (Table 1) indicating the sugars are joined by 6-linkages. A portion (5 mg) was hydrolyzed by 0.5 M TFA for 1/2 hour 62 and paper chromatography (solvent C) showed D-galactose, together with the same disaccharide as described above. In order to locate the reducing end, another portion (10 mg) was reduced with sodium borohydride and hydrolyzed (0.5 M TFA, 1 hour) to give galactitol (R„ 0.88 paper chromatography, solvent B) confirmed by g.l.c. of the alditol acetate (m.p. 164-166°) together with the same disaccharide (solvent C) as described above. C. Analysis of tetrasaccharide (25 mg) The anomeric linkages were proved by p.m.r. (at 95° in D20) to contain one a-linkage and two 3-linkages (Table 1). A portion (10 mg) was hydrolyzed by 0.5 M TFA for 1/2 hour. On paper chromatography (solvent C), D-galactose was shown together with the same disaccharide and trisaccharide as described above. Another portion (10 mg) was methylated by Hakomori method and hydrolyzed. 2,3,4,6-Tetra-O-methyl-D-glucose was found by paper chromatography (solvent D, 0.79) together with a major spot of 2,4,6-tri-O-methyl-D-galactose (R^  0.37). The quantity of the partially methylated sugars was insufficient for g.l.c. analysis. Partial Hydrolysis of Methylated Capsular Polysaccharide Methylated capsular polysaccharide (200 mg) was hydrolyzed with 90% formic acid for 45 minutes at 7 0°. TLC of the hydrolysate in solvent D showed the presence of three spots with R^  0.78, 0.72 and 0.66 (major), together with 2 faint spots with R^  0.47 and 0.25-0.30. The component with R^  0.72 had the same characteristics as 2,3,4-tri-0-methy1-L-rhamnose while the component with Rf 0.78 was reduced and 63 hydrolyzed, 2,3,4-tri-0-methy1-L-rhamnose, 2,4,6-tri-O-methyl-D-galactose, 4,6-di-0-methyl-r)-galactose and 2-iO-methyl-D-glucose were shown by paper chromatography,to be present. The hydrolysate (150 mg) was separated by TLC to yield a major component (40 mg) (R^ 0.66, solvent D). P.m.r. of this fraction showed a strong signal at x 8.45. A small portion (7 mg) was hydrolyzed to give 2,4,6-tri-O-methyl-D-galactose, 4,6-di-0_-methyl-D-galactose and 2-0-methyl-D-glucose in a ratio of 1.5:0.8:1 (determined as alditol acetates). On reduction with sodium borohydride, two components were obtained with R^  values 0.14 and 0.07 on TLC in solvent D. The two components were again separated on TLC in solvent D. The faster component (12 mg), on hydrolysis, showed only 2-0j-methyl-D-glucose on paper chromatography but when the hydrolysate was acetylated, 2,4,6-tri-O-methyl-g-galactitol acetate and 2-0-methyl-D-glucose acetates were found in a ratio of 1.25:1. When a portion of the slow moving component (10 mg) was hydrolyzed, the following results were obtained: R^  Paper Chromatography A Molar Solvent D Solvent E T . Ratio min 0.39 0.61 2,4,6-tri-O-methyl-D-galactose 18 (ct)+23 (3) 1.1 0.14 0.46 4,6-di-O-methyl-D-galactose 26(a)+32(3) 1 0.06 0.35 2-0-methyl-D-glucose 36.5(a)+38(3) 1 not detectable by £- 2,4,6-tri-O-methyl-D-galactitol 25 1 ,3 anisidine The slower component (5 mg) was methylated by the Hakomori procedure and on hydrolysis showed (solvent D) 2,4,6-tri-O-methyl-D-_ ECNSS-M column program from 170-220° at 2°/min. T - retention times of sugar acetates. 64 galactose (Rf 0.38) and 2,3-di-O-methyl-D-glucose (Rf 0.27). There is no evidence of the presence of tetramethyl glucose or galactose. The 2,3-di-O-methyl-D-glucose was confirmed by g.l.c.-m.s. of the alditol acetate (retention time 24 min, 215°, column a). Another portion (50 mg) of the original hydrolysate was hydrolyzed for a further period of 4 hours with 0.5 M TFA at 9 5°. By paper chromatography (solvent E)5 2,4,6-tri-O-methyl-D-galactose, 4,6-di-O-methyl-D-galactose, and 2-0-methyl-D-glucose were shown together with a new component with R^  0.79. The new component was separated by TLC (R^ 0.6, solvent D). Yield 15 mg. The new component (5 mg) was hydrolyzed with 2 M TFA at 95° for 16 hours. Paper chromatography (solvent D) showed the presence of 2,4,6-tri-O-methyl-D-galactose (R^ 0.38) and 2-0j-methyl-D-glucose (R^ 0.06). The ratio was found to be 1:1 as determined by g.l.c. of the alditol acetates. The rest was reduced with sodium borohydride and hydrolyzed. On paper chromatography only 2-0-methyl-D-glucose was shown (R^ 0.35, solvent E). The hydrolysate was acetylated and g.l.c. showed the presence of 2,4,6-tri-O-methyl-D-galactitol acetate and 2-0-methyl-D-glucose acetates in a ratio of 1.2:1. The retention time of 2,4,6-tri-O-methyl-D-galactitol acetate is 22 min, while that of 2-0-methyl-D-glucose acetates are 36 min (a) and 38 min (B) when column a is programmed from 170-220°. Table 1. P.m.r. data on Klebsiella K56 capsular polysaccharide and derived oligosaccharides. Repeating unit of compound T-value Ratio of (coupling constant,Hz) integrals Proton assignment Glc Gal-OH 4.64 5.28 (7) 0.5 1.5 a-Gal-OH 8-Gal-OH B-Glc Gal 13 13 Glc Gal Gal-OH 4.63 • 5.29 (7) 0.5 2.5 a-Gal-OH 6-Gal-OH B-Glc — - Gal 3-Gal ^ - ^ Gal 13 13 13 Glc ^  Gal Gal Gal-OH B 8 a 4.63 4.75 5.28 (7) 0.5 1 2.6 a-Gal-OH o-Gal — Gal B-Gal-OH B-Glc — Gal B-Gal — Gal Table 1 (continued) I-CO • A CE3CCOONa 3 ' ' 1 3 1 3 1 3 l - Glc gal Gal — Gal —t a 1 L-Rha CH3CCOONa 3 1 3 13 13 1 — Glc ~ - G al ^ G a l G a l —c 3 3 a i 3 13 13 13 1 — Glc ^TT1 G a l " V 1 Gal Gal - V 3 3 a 3 4.76 5.08 (§) 5.28 (§) 8.57 8.73 (5) 4.75 5.09- 5.21 (§) 8.57 4.75 (3) 5.24 (5-7) 1 3 3 1 3 3 1 a-L-Rha -^-^ Gal ot-Gal — Gal B-Glc — Gal B-Gal — Glc B-Gai — ' Gal CH3 of pyruvate ketal CH3 of L-Rha a-Gal — Gal B-Gal — Glc B-Glc — Gal B-Gal — Gal CH„ of pyruvate ketal 13 a-Gal ±-1 Gal B-Glc — Gal B-Gal — Glc B-Gal — Gal * AA Spectra run in H^O with external tetramethylsilane (x = 10) at 100 MHz. All sugars have D-conf iguration except L-rhamnose and a l l are pyranose. § signals approximately 5-6 Hz wide; shows no distinct splitting. 67 Table 2. Methyl ethers from the hydrolysis of methylated Klebsiella K56 polysaccharides. Sugars Sample  A+ B+ C+ D+ E+ F+ 2,3,4,6-Tetra-O-methyl-D- 1 + glucose 2,3,4-Tri-O-methyl-L- 0.85 rhamnose 2,4,6-Tri-O-methyl-D- 2.2 3 0.9 + 1.1 galactose 4,6-Di-O-methyl-D-galactose 1 1 2,3-Di-O-methyl-D-glucose 2-0-Methyl-D-glucose 1 1 1 1 2,4,6-Tri-O-methyl-D- 1.3 1.2 galactitol _ A, sugars from methylated original K56 polysaccharide; B, sugars from methylated Smith periodate degraded polysaccharide; C, sugars from methylated disaccharide; D, sugars from methylated tetrasaccharide; E, sugars from partially methylated tetrasaccharide; F, sugars from partially methylated disaccharide. t Approximate molar ratios. (+) Signifies present and has no quantitative significance. 68 Table 3. Diagnostic prominent peaks (m/e) in the mass spectra of acetates of methylated alditols. Parent sugar m/e  43 45 117 129 131 139 161 175 189 233 261 333 2,3,4-Me3Rha + + + + + 2,4,6-Me3Gal + + + + + + 4,6-Me2Gal + + + + + + 2,3-Me2Glu + + + 2-MeGlu + + + 69 BIBLIOGRAPHY 1. W. Nimmich, Z. Med. Mikrobiol. Immunol., 154, 117 (1968). 2. W. Nimmich, Acta b i o l . med. germ., 2§_, 397 (1971). '3. G.M. Bebault, Y.M. Choy, G.G.S. Dutton, N. Funnell, A.M. Stephen, and M.T. Yang, J . Bact., March (1973). 4. Y.M. Choy, G.G.S. Dutton, A.M. Stephen, and M.T. Yang, Analytical Letters, 5_, 675 (1972). 5. 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). 6. G.G.S. Dutton and N. Funnell, unpublished results. 7. R.L. Whistler and D.F. Durso, J . Am. Chem. S o c , 72, 677 (1950). 8. W.J. Whelan, J.M. Bailey and P.J.P. Roberts, J . Chem. S o c , 1293 (1953) 9. S.C. Churns, Advan. Carbohydrate Chem., 25, 13 (1970). 10. S.I. Hakomori, J . Biochem. (Tokyo), 55, 205 (1964). 11. P.A. Sandford and H.E. Conrad, Biochemistry, 5_» 1508 (1966). 12. H. Bjorndal, B. Lindberg and S. Svensson, Acta Chem. Scand., 21, I, 1801 (1967). 13. H. Bjorndal, C.G. Hellerqvist, B. Lindberg and S. Svensson, Angew. Chem. internat. Edit., . _9_, 610 (1970). 14. Y.M. Choy, G.G.S. Dutton, Can. J . Chem., Jan. 15 (1973). 15. I.J. Goldstein, G.W. Hay, B.A. Lewis and F. Smith, Methods Carbohydrate Chem., _5, 361 (1965). 16. B. Capon, Chem. Rev. 69(4) , 407 (1969). 17. C.J. Lawson and D.A. Rees, J. Chem. Soc. (C), 1301 (1968). 18. B. Lindberg, J . Lonngren and W. Nimmich, Carbohydr. Res., 23, 47 (1972). 70 19. J.E. Scott, Chem. & Ind. (London), 1568 (1955). 20. M. Duckworth and W. Yaphe, Chem. & Ind. (London), 747 (1970). 21. G.G.S. Dutton and A.M. Stephen, unpublished results. 71 APPENDIX I The Structure of the Capsular Polysaccharide from Klebsiella K-Type 21 Can. J . Chem.', 51, 198 (1973) 198 The Structure of the Capsular Polysaccharide from Klebsiella K-Type 211 Y. M. CHOY AND G. G. A. DUTTON Department of Chemistry, University of British Columbia, Vancouver, British Columbia Received July 27, 1972 Mcthylation, periodate oxidation, and partial hydrolysis studies on the capsular polysaccharide, and on the carboxyl reduced polymer, of Klebsiella l\.2\ show the structure to consist of a repeating unit. 3 „ , . 1 3 . . 1 2 . , 1 3 „ , 1 — D - G l c A p — D-Manp — D - M a n p — D-Galp — 4 a a a p a 1 D-Gal U CH3—C—COOH The anomeric linkages were determined by p.m.r. spectroscopy of isolated oligosaccharides and, in part, by specific enzymes. P.m.r. spectroscopy of the original polysaccharide in methyl sulfoxide-</6 showed clearly a ratio of one pyruvic acid ketal ( C H 3 , x 8.5) to five anomeric protons (t 4.65-5.40). Les mcthodes de mcthylation, oxydation pcriodique et hydrolyse partiellc portees au polysaccharide capsulaire, et an polymere avec le groupe carbo.xyle rcduit, de Klebsiella K21 ont demontre que la structure se compose d'une unite qui se rcpete. 3 ^, * 1 3 w 1 2 1 3 _ , 1 - — D-GlcAp — - D-Manp — D-Manp — D-Galp — 4 a a a J3 a 1 D-Gal n C H 3—C—COOH Les liaisons anomcrcs ont cte distinguees par la spectroscopic r .m.n. des oligosaccharides isoles et, en partie, par des enzymes speeifiques. L a spectroscopic r.m.n. du polysaccharide original dans le methyl sulfoxidc-r/ 6 a etabli le rapport tin a cinq entre le cctal de l'acidc pyruvique ( C H j , T S.5) et les protons anomcrcs (T 4.65-5.40). Can. J. Chem., 51, 198 (1973) Qualitative analyses of the capsular polysac-charides of the 80 serotypes of Klebsiella bacteria have been provided by Nimmich (1,2) who has shown that the great majority contain glucuronic acid in combination with licxose and deo.xyhe.xose sugars. Approximately one half of the polysac-charides also contain pyruvic acid (3) covalently bound (1-carbo.xyethylidene ketals). Despite the amount of qualitative information available on 'Presented in part at the 55th Canadian Chemical Conference of .the Chemical Institute of Canada, Quebec Ci t y , Quebec, June, 1972 and at (he 6th International Carbohydrate Symposium, Madison, Wisconsin, August, 1972. Klebsiella polysaccharides detailed structures are only known for those capsular materials from types K.2 (ref. 4), K5 (rcf. 5), KS (rcf. 6), K9 (ref. 7), K20(ref. 8), K54(rcfs. 9-11), and three related but untyped species of Aerobacter acrogenes (12-14). The structure of the capsular polysac-charide from Klebsiella K21 is now reported. A preliminary report has appeared (15) which also gives other background references. The acidic capsular polysaccharide from Klebsiella K2i was purified by one precipitation with Cetavlon and had [ot]„ 4-130\ Gel filtration and electrophoresis showed the material to be homogeneous. In work with other Klebsiella C H O Y A N D D U T T O N : P O L Y S A C C H A R I D E S 199 capsular polysaccharides it has been found that there is an approximate, inverse relationship between the equivalent weight of the polysaccha-ride and the distance of migration. One may pre-dict that the polysaccharide from Klebsiella K62 will have an equivalent weight about 5S0 (Fig. 1). The p.m.r. spectrum of a 2% solution of the polysaccharide in D 2 0 showed a sharp singlet at T 8.5 indicative of a pyruvic acid ketal (16, 17). Determination of the pyruvic acid :sugar ratio was hampered by the presence of the HOD signal even after several exchanges with D 20. The HOD peak was shifted downfield by the addition of trifluoroacetic acid or upfield by running the spectrum at 95-100° on the polysaccharide in methyl suIfoxide-tf6. In the latter case integration of the signals due to the anomeric protons and that of the methyl at x 8.5 showed there was one pyruvic acid ketal to five sugar residues (Fig. 2). Partial assignment of the anomeric signals may be made using the data of Table 1, which are based on examination of the oligosaccharides obtained in subsequent experiments, and similar data given by Conrad and coworkers (4) and others. These enable one to predict that of the five glycosidic bonds (per repeating unit) only one is of the p-D-configuration (shown subsequently to be p-D-galactose). In later work with other Klebsiella capsular polysaccharides (K20, K24) excellent spectra have been obtained by using the sodium salt of the polysaccharide in D,0 at 95° and dispensing with the use of methyl sulfoxide. Acidic hydrolysisof the polysaccharideshowed, Distance of Migration (cm) FIG. I. Klebsiella capsular polysaccharides; electro-phoresis on cellulose acetate at .100 V Tor 30 min (p// S.8, vcronal-tris butler). Distance of migration is. equivalent weight. K62 migrates 2.4 cm indicating an equivalent weight of 5S0 . by paper chromatography, the presence of D -mannose, o-galactosc, D-glucuronic acid, lactone, and an aldobiouronic acid. Separation of the hydrolyzate into neutral and acidic fractions and g.l.c. of the former, both as trimethylsilyl derivatives and as alditol acetates, gave a ratio of D-mannose to D-galactose of 1:1.2. Collection of the alditol acetates permitted characterization of D-mannitol hexaacetate m.p. 118-121° and galac-titol hexaacetate, m.p. 162°. The aldobiouronic acid was separated from the acidic fraction by paper chromatography and reduction of the carboxyl function followed by hydrolysis gave D-glucose and D-mannose, further characterized as their alditol acetates. The configuration of D -galactose was established by D-galactose oxidase, D-glucuronic acid as D-glucose by D-glucose oxidase, and D-mannose by the circular dichroism curve of D-mannilol hexaacetate (18). Autohydrolysis of the polysaccharide gave pyruvic acid (19) characterized as the 2,4-dini-trophenylhydrazone (20). It was considered significant that the aqueous solution showed the presence of D-galactosc as the only monosaccha-ride together with traces of oligosaccharides. The residual polysaccharide, obtained by dialysis and lyophilization of the aqueous hydrolyzate, was saved for subsequent methylation studies discuss-ed later. A sample of pure capsular polysaccharide was methylated (10, 21) and hydrolyzed to yield a neutral and an acidic fraction after separation on ion-exchange resins. Table 2 (A) gives the neutral sugars obtained which were identified as discussed in the experimental. These results show that D -mannose occurs in the polysaccharide linked in two ways; substituted at position 2 and at position 3. The ratio of 3,4,6-tri-O-mcthyl-D-mannose to the 2,4,6-isomer was 1.7:1, consistent with the concept that part of the mannose remained bound to the uronic acid. Isolation of 2,4,6-tri-O-mcthyl-D-galactose indicates that this unit represents a 3-substituted sugar in the chain. The 2,3-di-O-methyl-D-galactosc cannot represent a branch point since no corresponding quantity of terminal units was found. The pyruvic acid must therefore be linked to D-galactosc as a 4,6-kctal. The acidic fraction of the hydrolyzate was con-verted to the methyl ester glycoside, reduced and hydrolyzed to give 2,4,6-tri-O-mcthyl-D-mannosc and 2-O-melhyl-D-glucosc, Table 2 (8). These results show that the aldobiouronic acid is 3-0-(glucopyranosyluronic acid)-D-mannosc and that 74 200 CAN. J. CHUM. VOL. 51, 1973 TABLE J. P.m.r. data on Klebsiella K21 capsular polysaccharide and derived oligosaccharides Repeating unit or compound T-Valuc* Solvent, (coupling Ratio of temperature (°) constant, Hz) integrals Original polysaccharide Mc2SO-rf6 100 —Gal — - GIcA — Man — Man — Gal— Aldotetrauronic acid GlcA — Man — Man — Gal—OH a a a. D20 95 Oligosaccharide from Smith degradation 13 13 1 Gal — GlcA — Man — P a a DjO 95 CH2OH -H CH2OH 4.65 4.90+4.95 5.05 5.40 4.75 + 4.80 5.00 5.47(8) 4.75 4.95 5.47 (8) 1.2 Proton assignment (all sugars have D-configtiration) 1 3 ot-GIcA — - Man 1 3 a-Man — Gal a-Gal — GlcA 1 2 1 a-Man — Man 1 P-Gal — GlcA 2.5 a-GlcA —Man 1 3 a-Man — Gal a-Gal—OH 1 2 1 a-Man — Man 0.5 P-Gal—OH 0.9 a-GlcA—-Man a-Man 1 CH2OH H CH,OH P-Gal ~ GlcA •External tctramethylsilanc standard for aqueous solutions, internal for methyl sulfoxide. the D-glucuronic acid is a branch point in the chain . In the absence o f tetramethylhexose as terminal groups the side chain attached, to the glucuronic acid must be the 4 ,6 -0- ( l -carbo. \y-ethylidcne)-D-galactose units. These methylat ion data permit the drawing of partial structures such as A and B but do not distinguish between them nor do they establish the posit ion o f attachment o f the side chains nor the mode o f linkage o f the glucuronic acid in the main cha in . 1 3 1 2 . , 1 3 „ , — D-Manp — D-Galp— -D-GlcpA — D-Manp or 1 3 1 2 D-Galp — D-Manp-D-Gal 4 CI 13—C—COOH T h e residual polysaccharide recovered f rom the C H O Y A N D D U T T O N : P O L Y S A C C H A R I D E S 201 FIG. 2. P.m.r. spectrum of Klebsiella K21 capsular polysaccharide in methyl sulfoxide'-*^ at 100°. Tetramethylsilane internal standard (x 10), 100 MHz. TABLE 2. Methyl ethers from the hydrolysates of methylated Klebsiella K21 polysaccharides Sample* Sugars A t B C D E 2,3,4,6-Tetra-O-methyl-D-galactose 3 2,3,4,6-Tetra-O-methyl-D-mannose 1 2,4,6-Tri-O-methyl-D-gainctosc 1.0 23 0.8 2,4,6-Tn-O-mcthyl-D-mannose 0.6 + 24 -f 1.0 3,4,6-Tri-O-methyl-D-mannose 1.0 27 1.1 2,3-Di-O-mcthyi-n-ga lactose 1.0 2 1.0 2,4-Di-O-methyl-D-glucose + 2,6-Di-O-methyl-D-glucose 1.0 2-O-Methyl-D-glucose + •A, neutral sue.ars from methylated original K2I polysaccharide; C, reduced aldobiouronic acid fraction from methylated original K21 polysaccharide; C, neutral sugars from methylated residual polysaccharide after atnohydrnlysis; D. reduced aldobiouronic acid from methylated residual polysaccharide after aulohydrolysis; JE, neutral sugars front methylated reduced K2I polysaccharide. tApproximate molar ratios; (-t ) signifies present in about cquimolecular amounts. autohydrolysis described above was methylated and hydrolyzed to yield, after separation on ion-exchange columns, neutral and acidic frac-tions. T h e composi t ion o f the former is shown in T a b l e 2 (C) and the significant feature is the virtually complete disappearance o f 2 ,3-di -O-methyl - D -galactose and , incidentally, the very small amounts o f tctramcthyl sugar produced. Paper chromatography o f the acidic fraction showed two components in apparently equal quantities with mobilities corresponding to par-tially methylated uronic acid (R, 0.16) and a ldob iouron ic acid (R( 0.23). Reduct ion and hydrolysis o f this mixture yielded 2,4,6-tri-O-methyl -D -mannose and 2 ,4 -d i -O-mcthy l - D -g lu -cosc, Tab le 2 (D) . T h e disappearance of 2 ,3-di -O-methyl - D -galactose in the neutral fraction and the appearance in the acid fraction o f 2 ,4 -d i -O-methy l - D -g lucuronic acid is convincing 'ev idence that the 4 ,6 -0 - ( l -carbo. \ycthy l idcne ) - D -ga lactose units arc attached to posit ion 4 o f D -g lucuronic acid which in turn must be l inked through posi -tion 3 in the main cha in . T h e informat ion thus far obtained was c o n -f irmed, and more accurate quantitative data were obtained, by examinat ion o f thecarboxy l - reduccd polysaccharide. T h i s was prepared exactly as described for sapote gum using l i thium boro -76 C A N . ). C111£M. V O L . 51. 1973 202 hydride in tctrahydrofuran (22). T h i s method has given virtually quantitative reduction with this and several related polysaccharides and is now our preferred method o f achieving such a c o n -version. It should also be noted thai this reduced poly-saccharide and the methylated polymer described above each showed a characteristic sharp singlet ( T ca. 8.6) for the methyl group o f the cor respond-ing pyruvic acid derivative. Th is signal thus affords not only a method for the quantitative estimation o f l -carboxyethy!idcne ketals but also for checking that the related hydroxy isopropy l -idene and 1-carbomethoxycthylidene groups survive subsequent chemical transformations. In the polysaccharide only small changes in chemical shift were detected in these methyl signals (16). Hydro lys is o f the reduced polysaccharide gave D -mannose, D-galactose, and D-glucose in the ratio o f 1:1:0.6 (as alditol acetates). T h e reduced polysaccharide was methylated and gave on hydrolysis the sugars in the propor -tions shown in Tab le 2 (E). T h e format ion o f 2,6-di-O-mcthyl-D -glucose arises f rom the reduced D -glucuronic acid moiety and the quantitative data are consistent with the proposed partial structure. A port ion o f the original capsular polysac-charide was autohydrolyzed at 95° for 3 days (initial p / / 2 . 3 ) using the apparatus described by G a l a n o s et al. (23) which circulates the poly-saccharide solut ion through stages o f heating, coo l ing , and dialysis and recycles unhydrolyzed material. Th is apparatus is a modern improve-ment on the ideas o f Painter (24) and of Perila and B ishop (25) but because o f the close cont ro l that can be exercized high yields o f o l igosaccha-rides may be obtained. A s previously noted little, except D-galactose and pyruvic acid, is liberated on autohydrolysis; therefore the polysaccharide solut ion was made up to 0.125 .4/" sulfuric acid and dialyzed against acid o f the same concentrat ion. Hydro lys is was cont inued for a further 3 days at 95°. T h e dialyzate was neutralized with bar ium carbonate and separated into neutral and acidic fractions. T h e latter was separated on paper to give, as the main component ; an aldotetrauronic acid with lesser amounts o f aldotr i - and a ldo-biouronic acids. Tab le 1 gives the structure o f the aldotetrauronic acid based on the evidence pre-sented in the Experimental . T h e salient points arc that partial hydrolysis o f the aldotetrauronic acid gave D-galactose as the main monosacchar ide and , similarly, hydrolysis after s o d i u m boro-hydr ide reduction yielded galactitol. T h e other products o f the partial hydrolysis cor responded to the aldotr i - and a ldob iouron ic acids isolated. T h e a ldob iouron ic acid was resistant to the action o f p-D -glucuronidasc and gave D -mannose on hydrolysis. Identification of the a ldotetrauro-nic acid enables a decision to be made in favor of structure A. T h e a ldob iouron ic acid 3-0-(c.-D-glucopyranosyluronic ac id ) -o -mannose has been isolated previously f rom Serraiia marceseens (26), Klebsiel la K 2 (ref. 4), and from Pseudomonas aeruginosa (27). Conf i rmat ion o f structure A was sought by periodate oxidat ion o f the original capsu la r poly-saccharide. Periodate consumpt ion o f 0.36 mol per hexose unit was constant after 3 d a y s and the derived po lya lcohol was hydrolyzed b y heating an aqueous solut ion, initial pl/ 2.2, at 9 5 = for 1 h. A separate paper has discussed the wide variety o f condit ions which has been app l i ed in the hydrolysis step of Smith degradat ions and has noted that other authors have experienced diffi-culty in obtaining cleavage o f acetal l inkages in polymers containing uronic acids (28). Hydro lyses have c o m m o n l y been carried out with acid strengths o f 0.05-0.5 A / al though t w o groups have recently reported good results wi th 0.01 M acid (13, 29). Johnson and Percival (30) have utilized the acidity o f sulfated polysacchar ides to effect autohydrolysis o f the p o l y a l c o h o l but the present work appears to be the first instance where the acidity o f the uronic acid has been used to achieve hydrolysis o f the p o l y a l c o h o l in a Smi th degradation. Paper chromatographic examina t ion (solvent A) o f the hydrolyzate showed the presence o f (a) pyruvic ac id , (b) glycerol , (c) D - threitol , (cl) D-galactose (traces only), and (t-) an o l igosacchar ide which had a mobil i ty (in solvent E) intermediate between the a ldobi - and a l d o t r i o u r o n i c acids previously isolated. T h e ratio o f g lycero l to D-threitol was 1.2:1 by g.l .c. o f the acetates. It is pertinent to note here that even under these mi ld condi t ions o f autohydro lys is the pyruvic acid ketal was hydrolyzed. S i m i l a r l y , no condit ions o f partial hydrolysis o f the original polysaccharide could be found w h i c h would permit the isolation o f 4 ,6-c9- ( l -carboxyethyl -idenc)-D-galaclose. Fur thermore , an attempt to achieve this by the use of a-galactosklase was unsuccessful since the enzyme was w i thout action on the original polysaccharide. T h i s a c i d lability C H O Y A N D D U T . O N : P O L Y S A C C H A R I D E S 203 of the pyruvic acid ketal is in direct contrast to the result of G o r i n and Ishikawa (16) who were able to hydrolyze the glycoside methyl 4,6-C1-(l-car-boxyethyl idenc)-c . - D -galactoside without cleav-age of the ketal. T h e difference in stability may be accounted for by the conformat ion o f the ketal or by the proximity o f the D -g lucuronic acid moiety, but no informat ion is as yet available on these points. T h e ol igosaccharide obtained in the Smith degradation was o f sufficient molecular weight to be precipitated when the aqueous solut ion was poured into ethanol. T h e p.m.r. spectrum of the reduced ol igosaccharide showed the presence o f one B - D - l inkage per three anomer ic protons and incubat ion o f the original ol igosaccharide with P-galactosidase yielded D-galactose. W h e n the reduced ol igosaccharide was methylated and hydrolyzed examinat ion by paper chromatogra -phy in solvent D showed 2,3,4,6-tetra-O-mcthyl-D-galaclose to be the major neutral sugar with traces o f 2 ,4 ,6- t r i -O-methyl - D -mannose. T h i s result was confirmed by g.l.c. E x a m i n a t i o n in solvent £ showed also the presence o f two acidic components with mobilities identical to 2 ,4-d i -O-methy l - D -g lucuronic acid and the partially methylated a ldob iouron icac id isolated previously f rom the methylated residual polysaccharide. In neither case was the d i -O-methylg lyccro l sought. T h e sum o f these results firmly establishes structure A as representing the repeating unit o f the capsular polysaccharide of Klebsiella K21 and demonstrates for the first time the mode o f attachment o f the pyruvic acid in this genus. T h e structure found for K21 is similar to those previously reported for K 2 , K 8 , K 9 , K.54 (refs. 4, 6, 7, II) in that they each have a repeating unit of four or five sugars including a single unit side chain . Tha t this is not an invariant rule for Klebsiella polysaccharides is shown by the structure o f K 5 (ref. 5) which has a repeat o f three sugars and no side chain and o f K20( rc f . 8) which has a repeat o f four sugars including a two unit side chain . Ful l details o f these other structures will be published in due course. Experimental General Methods P a p e r c h r o m a t o g r a p h y w a s c a r r i e d o u t b y the d e s c e n d -i n g m e t h o d u s i n g W h a t m a n N o . 1 p a p e r a n d t he f o l l o w i n g s o l v e n t s y s t e m s (V/v): {A) e t h y l a c e t a t e - a c e t i c a c i d - f o r m i c a c i d - w a t e r (18 :3 :1 : • ' ) ; (/*) e l h y l ace ta te -p y r i d i n e - w a t e r ( 4 : 1 : 1 ) ; ( C ) I - b u t a n o l - a c e t i c a c i d -w a t e r (2 :1 : l ) ; ( / ) ) b u l a n o n c s a t u r a t e d w i t h 1% a q u e o u s a m m o n i a ; ( £ ) 1 - b u t a n o l - c t h a n o l - w a t e r (4:1:5). C h r o -m a t o g r a m s w e r e d e v e l o p e d w i t h /> -an is id inc t r i c h l o r o -a c e t i c s p r a y (31) o r w i t h s i l v e r n i t r a t e (32) . T h e c l a r i t y o f the c o l o r s o n c h r o m a l o g r a m s o f m e t h y l a t e d s u g a r s d e -v e l o p e d w i t h the f o r m e r s p r a y is g r e a t l y i m p r o v e d b y w a s h -i n g the p a p e r u n d e r r u n n i n g ho t w a t e r a f t e r h e a t i n g t he c h r o m a t o g r a m a l 1 0 0 - 1 1 0 ' in t he n o r m a l m a n n e r . G . l . c . w a s c a r r i e d o u t o n a n F a n d M m o d e l 7 2 0 d u a l c o l u m n i n s t r u m e n t f i t ted w i t h t h e r m a l c o n d u c t i v i t y d e t e c t o r s . T h e h e l i u m f l o w was 6 0 - 8 0 m l / m i n w i t h the f o l l o w i n g c o l u m n s : (a) 3 % E C N S S - M o n C h r o m o s o r b W (8 ft x 0 .25 i n . ) ; (b) 5 % b u t a n e d i o l s u c c i n a t e o n D i a t o p o r t S (4 ft x 0 .25 i n . ) ; (<) 8 % S E - 5 2 o n D i a i o p o r t S (8 ft x 0 .25 i n . ) . C i r c u l a r d i c h r o i s m s p e c t r a w e r e r u n o n a J a s c o J - 2 0 a u t o m a t i c r e c o r d i n g s p e c t r o p o l a r i m e t e r u s i n g a q u a r t z c e l l w i t h p a t h l e n g t h 0.1 c m . O p t i c a l r o t a t i o n s w e r e m e a s u r e d at 23 ± 2° o n a P e r k i n - E l m e r m o d e l 141 p o i a r i m c t e r . P . m . r . s p e c t r a we re r u n o n V a r i a n T 6 0 o r X L I 0 0 i n s t r u m e n t s . S a m p l e s we re p r e p a r e d b y d i s s o l v i n g i n D 2 0 a n d f reeze d r y i n g 3 o r 4 t i m e s b e f o r e d i s s o l v i n g in D > 0 o r m e t h y l s u l f o x i d c - r / 6 . T e t r a m e t h y l s i l a n e w a s u s e d as a s t a n d a r d , e x t e r n a l l y f o r D 2 0 s o l u t i o n s a n d i n t e r n a l l y f o r m e t h y l s u l f o x i d c - < / 0 . M a s s s p e c t r a , o b t a i n e d o n i n d i v i d u a l f r a c t i o n s c o l l e c t e d (33) f r o m the gas c h r o m a t o g r a p h y w e r e r u n o n a n M S 902 i n s t r u m e n t at 70 e V . T h e p o s i t i o n o f m e t h o x y l s u b s t i t u -t i o n w a s d e t e r m i n e d u s i n g the d a t a o f L i n d b e r g a n d c o w o r k e r s (34) a n d , w h e r e p o s s i b l e , b y c o m p a r i s o n o f m a s s s p e c t r a w i t h t hose o f a u t h e n t i c c o m p o u n d s . M e l h y l a l i o n s o n p o l y s a c c h a r i d e s a n d o l i g o s a c c h a r i d e s w e r e c a r r i e d o u t by the m e t h o d o f H a k o m o r i (9 , 21 ) . P a r t i a l l y m e t h y l a t e d a l d i t o l ace ta tes w e r e d c m c t h y l a i e d b y r e a c t i o n w i t h b o r o n t r i c h l o r i d e (35) . A l l s o l u t i o n s w e r e c o n c e n t r a t e d o n a r o t a r y e v a p o r a t o r in vacuo at a b a t h t e m p e r a t u r e o f 4 0 " . D - G l u c o s t a t a n d u - G a l a c t o s t a t r e a g e n t s w e r e o b t a i n e d f r o m the W o r t h i n g t o n B i o c h e m i c a l C o r p o r a t i o n , P-D-g a l a c t o s i d a s e f r o m K o c h - L i g h t , a n d P - D - g l u c u r o n i d a s e f r o m S i g m a C h e m i c a l C o m p a n y . c / - D - G a l a c t o s i d a s e w a s p r e p a r e d f r o m g r e e n co f f ee b e a n s b y the p r o c e d u r e o f C l a r k e el al. (36) . S a g a v a c 6 F is a p r o d u c t o f S e r a v a c L a b o r a t o r i e s , C a p e T o w n . Preparation anil Properties of K2I Capsular Polysaccharide A c u l t u r e o f Klebsiella K 2 I ( 1 7 0 2 / 4 0 ) w a s o b t a i n e d f r o m D r . O r s k o v , C o p e n h a g e n , as a n a g a r s l a n t a n d w a s g r o w n o n the f o l l o w i n g m e d i u m f o r 3 d a y s at 2 5 " : 8 g N a C I , 4 g K . H I ' O j , I g M g S O W i l 2 0 , 2 g C a C O j , 1 2 0 g s u c r o s e , a n d 8 g B a c l o yeas t e x t r a c t i n 4 I o f w a t e r . C e l l s w e r e h a r v e s t e d a f t e r 3 d a y s , d i l u t e d w i t h w a t e r c o n t a i n i n g 1% p h e n o l , a n d c e n t r i f u g e d a t 19 0 0 0 g . C a p s u l a r p o l y -s a c c h a r i d e w a s o b t a i n e d b y c o n c e n t r a t i n g the a q u e o u s s o l u t i o n , p o u r i n g it i n t o e t h a n o l , a n d p u r i f y i n g the c r u d e m a t e r i a l b y C c t a v l o n p r e c i p i t a t i o n . A s a m p l e o f the s u p e r n a t a n t w a s s h o w n t o c o n t a i n v e r y l i t t l e m a t e r i a l a n d t h e r e f o r e (h is w a s n o ! f u r t h e r e x a m i n e d . The y i e l d o f p o l y -s a c c h a r i d e w a s a p p r o x i m a t e l y 2 g pe r 4 1 o f m e d i u m . T h r e e b a t c h e s w e r e p o o l e d a n d u s e d f o r the s t r u c t u r a l s t u d i e s o f the p o l y s a c c h a r i d e . P u r i f i e d p o l y s a c c h a r i d e (2 g) w a s d i s s o l v e d in d i s t i l l e d w a t e r (2 I), d e i o n i z e d w i t h A m b e r l i t e 1R 120 r e s i n , d i a l y z c d , a n d f r e e / e - d r i c d . T h e p r o d u c t h a d [ - i ] „ + 1 3 0 ' 78 CAN. J. CULM. VOL. 51, 1973 204 (c 0.38, water), ash 0.8%, N 0.3%. The equivalent weight o f the polysaccharide, determined by titration with 0.01 At sodium hydroxide was "155 (calcd. '147). The average molecular weight was cn. 4 x 10* as determined by gel permeation chromatography on Sagavac 6F through the courtesy of Dr. S . C. Churms. Electrophoresis at pll 3 and 8.8 showed the polysaccharide moved as a single component (migration 3.2 cm in 30 min at 300 V at pll 8.8; by courtesy of Or. P. II. Reid and Miss C. Podcr). The p.m.r. spectrum of a 2% solution of D20 showed a sharp singlet at T 8.5 due to the CM 3 of the pyruvic acid ketal group. The relative amount of pyruvic acid was initially obtained by running the spectrum on a sample dissolved in 3 M irifluoroacetic acid in D20. Integration gave 17-18 units for the CH3 signal at T 8.5 and 146 units for the ring protons, excluding anomeric ones. The ratio was confirmed by running the spectrum of a 2% solution o f the polysaccharide in methyl sulfoxide-r/5 at 100'. In this case the HOD peak was shifted upfield so that the anomeric proton signals were clear. These were at T 4.65 (IH); 4.90, 4.95, 5.05 (31-1); 5.40 (11-1). The ratio of anomeric protons to pyruvate CH3 was 5:3. Hydrolysis of the Polysaccharide Polysaccharide (0.1 g) was hydrolyzed with 0.5 Al sulfuric acid at 100' for S h. Neutralization (BaC03) followed by paper chromatography (solvents .-( and B) showed D-mannose and D-galactose, D-glucuronolactonc, D-glucuronic acid, and an aldobiouronic acid. The hydrolyzate was separated into neutral and acidic frac-tions using ion-exchange resins (Amberlite IR 120 and Duolite A4). The neutral fraction contained D-mannose and D-galactose in approximately equal amounts (1:1.2) as determined by g.l.c. of the trimethylsilyl derivatives (column c) and of the alditol acetates (column a). G.l.c. o f the acetates • permitted recovery of the individual compounds and identification of D-mannitol hexaacetate, m.p. and mixed m.p. 118-12P", and galactitol hexaacetate, m.p. 162J (37). A portion of the neutral fraction of the hydrolyzate was tested with D-Ga!actostat reagent and a positive response confirmed the D-contiguration of galactose. Part of the mannitol hexaacetate collected by g.l.c. was recrystallized and dissolved in acetonitriic. A positive circular dichroism curve, identical to that given by a standard sample, confirmed the D-configuration of mannose (18). The acidic fraction was separated into two parts by paper chromatography (solvent A). The slower migrating, major portion had the mobility of an aldobiouronic acid (Rc,3i 0.39) and was converted into the ester glycoside by rcfiuxing overnight with 3% mcthanolic hydrogen chloride. The product was dissolved in tetrahydrofuran and reduced by boiling under reflux overnight with lithium aluminum hydride. Hydrolysis of the neutral disaccharide gave approximately cquimolar amounts of D-glucose and D-mannose as determined by g.l.c. of the alditol acetates. The configuration of the glucose was confirmed by the D-Glucostat reagent. A separate portion (1.5 g) of the polysaccharide was dissolved in water (initial pl/ 2-2.3) and heated for 8 h at 953. Paper chromatography (solvent .-1) of the con-centrated solution showed a fast moving spot, chromato-graphieally identical with pyruvic acid (R, 0.74) and having the same characteristic lluoresccnce as a standard when sprayed with o-phenylencdiamine and examined by u.v. (19). The chromatogram also showed D-galactose as the only monosaccharide and traces of slow moving oligomers. The aqueous solution was extracted with clher and the ether extract was treated with 2,4-dinitrophcnyl-hydrazine to yield the derivative of pyruvic acid, m.p. 21-1-216", undepressed by an authentic sample (20). The autohydrolysis solution was concentrated to 50 ml and dialy/.cd against distiiled water (2 I). The non-dialyzablc material (0.65 g) was recovered by frccze-drying and kept for methylation studies. AIel hy lat ion Analysis Capsular polysaccharide (0.5 g) was methylated to give a product having a methoxyl content of 38.5% and showing no hydroxy! absorption in the i.r. The methoxyl content was not increased by two further treatments with silver oxide and methyl iodide. The fully methylated polysaccharide was dissolved in cold 72% sulfuric acid and after 1 li the solution was diluted to 1 Al and hydro-lyzed at 100^  for S h. After neutralization and evaporation, the syrupy product was separated into neutral and acidic fractions by ion-exchange resins. The neutral portion was fractionated by cellulose column chromatography (38) using butanone-vvater azcotrope and the partially methyl-ated sugars were characterized as follows: Fraction I (SO mg) was shown by paper chromatogra-phy, (solvent D) to contain two components having R, 0.57 and 0.55; the faster mo\ ing one (la) gave a brownish yellow spot when sprayed with p-anisidine while the slower one (\b) gave a pink color. On demethylation, fraction I was shown by paper chromatography (solvents A and 13) to contain only D-mannose. Fraction 1 (20 mg) was reduced by sodium borohydride and, after removal of the borate, the product was acetylated in a scaled tube by using pyridine and acetic anhydride (1:1, 2 ml, I OO", 20 min). After evaporation of the solvents, the alditol acetates were dissolved in a small volume of ethyl acetate and analyzed by g.l.c. on column b programmed from 150-200: at 2i/min; two peaks were observed. The component eluted first, had a retention time of 17.2 min and mass spectrometry showed it to be a 3,4,6-lii-O-methyl hexitol acetate. On demethylation and acetylation, D-mannitol hexaacetate was obtained (g.l.c., m.p., and mixed m.p. 118-121"). The second component had a retention time (19.6 min) identical to that of authentic 2,4,6-lri-O-melhyl-D-mannitol triacetate. A sample collected from the gas chromatograph had m.p. 65-671 undepressed by authentic material (39) and the identity was further confirmed by the mass spectrum. The ratio of the amount of 3,4,6-isomer to that of the 2,4,6-compound was 1.0.6. Fraction II (30 mg) was shown by paper chromatogra-phy to be a single compound (/?, 0.37, solvent I)) and on demethylation, D-galactose was obtained. Crystallization of the syrupy sugar from ether gave 2,4,6-tri-O-mcthyl-D-galactosc m.p. 103-105 (fit. (40) m.p. 102-105 ). The derived alditol acetate had a retention time of 21 min (column a) and the mass spectrum was consistent with a 2,4,6-tri-O-methyl hexitol acetate. Fraction III (35 mg) was shown by paper chromato-graphy to be a single compound and to correspond to 2,3-di-0-mcthyl-i>-galaetose (R, 0.18, solvent D). This assignment was confirmed by the mass spectrum of the C H O Y A N D D U T T O N : P O L Y S A C C H A R I D E S 205 alditol acetate which on dcnicthylation and acetylation gave galactitol hexaacetate, m.p. and mixed m.p. 162°. The acidic fraction was heated under reflux overnight with 2% methanolic hydrogen chloride to give the ester glycoside which was reduced by boiling overnight with lithium borohydride in tctrahydrofuran. The product (20 mg) was hydrolyzed with 0.5 M sulf uric acid for 4 h. Paper chromatography showed the presence of 2,4,6-tri-O-mcthyl-n-mannose (A'f 0.5, solvent 0; 0.SO solvent E) and 2-O-melhyl-D-glucose (A'f 0.07, 0.21). The identities of both partially methylated sugars were confirmed by g.l.c. and mass spectrometry of the alditol acetates. The mannitol derivative crystallized, m.p. 64-66°, and the glucilol derivative was demethylated and acetylatcd to give glucilol hexaacetate m.p. 92-95°. 2-O-Mcthyl-D-glucitoi pentaacetate has been isolated subsequently from Klebsiella K7 methylated polysaccharide and has m.p. 56-57°. Methylation of Degraded Polysaccharide Degraded polysaccharide (0.5 g), recovered from the autohydrolysis, was methylated, the product was dis-solved in 5 ml of 72% sulfuric acid at 0° and then the solution was diluted to 1 M and hydrolyzed at 100' for 8 h. After neutralization the hydrolyzate was separated into acidic and neutral fractions. The latter was shown by paper chromatography (solvent D) and by g.l.c. of the alditol acetates (column b, 160-210° at 27min) to contain •the sugars listed in Table 2 (C) which also gives their approximate molar ratios. Paper chromatography of the acidic fraction (solvent E) showed two components with R, 0.16 and 0.23 and in approximately equal amounts. When sprayed with p-anisidine trichloroacetatc the slower gave the bright red color of a methylated uronic acid and the faster gave a reddish-brown color and was judged to be a partially methylated aldobiouronic acid. A portion of the mixture (20 mg) was converted to the ester glycosides and reduced with lithium borohydride in tetrahydrofuran. Paper chromatography, after acid hydrolysis, showed the presence of 2,4-di-O-methyl-D-glucose (Ii, 0.1S, solvent D) as the major component together with 2,4,6-tri-O-mcthyl-D-mannose. The identity of the former was con-firmed by g.l.c. - mass spectrometry of the alditol acetate and that of the latter by the crystalline alditol acetate m.p. 6-4-66°. Reduction of the Capsular Polysaccharide Capsular polysaccharide (0.4 g) was converted into the methyl ester propionate and reduced with lithium boro-hydride as previously described (22) to give the reduced polysaccharide (0.3 g), {-i)u + I15; (c 0.42, water). The reduced polysaccharide (0.1 g) was hydrolyzed with 0.5 M sulfuric acid at I00! for 8 h to give D-mannose, D-glucosc, and n-galactose (paper chromatography, solvent H). The identities of the sugars were confirmed by g.l.c. of the alditol acetates and collection of the individual fractions afforded D-mannitol hexaacetate m.p. and mixed m.p. 11S-120\ galactitol hexaacetate m.p. and mixed m.p. 160-162", and D-glucitol hexaacetate m.p. and mixed m.p. 92-95'. The ratio of D-mannitol, galacti-tol, and D-glucitol hexaacetatcs was 1:1:0.6. The p.m.r. spectrum of reduced K21 polysaccharide in D20 was shown to contain four anomeric signals at T 4.54, 4.80, 5.00, and 5.15 in the ratio of 1:2:1:1. All the signals showed small coupling constants except that at T 5.15 (7 Hz). The remainder of the reduced polysaccharide (0.2 g) was methylated and hydrolyzed. Table 2 (E) shows the nature and proportion of Ihe methylated sugars found and, in addition to those previously obtained from the methylated original polysaccharide, now included 2.6-di-O-niethyl-D-glucosc which was characterized by g.l.c. -mass spectrometry of its alditol acetate. Partial Hydrolysis of Polysaccharide K21 polysaccharide (1 g) was partially hydrolyzed at p// 2.3 at 95° for 3 days using an apparatus similar to that described by Galanos and coworkers (23). Since only D-galactosc and pyruvic acid were found in the dialyzate the solution was made 0.125 M with sulfuric acid and hydrolyzed at 95° for a further 3 days. The dialyzate was neutralized (BaC03), evaporated, and separated by ion-exchange resins into acidic and neutral fractions. On paper chromatography (solvent C) the acidic fraction was shown to contain an aldotetrauronic acid (R^^ 0.2) as the major component together with small amounts of aldo-biouronic acid (/\Gai 0.70), aldotriouronic acid (Rc>i 0.42), and traces of higher oligomers. The acids were separated by paper chromatography (solvent C) and analyzed as follows. The aldotetrauronic acid (50 mg) had an equivalent weight of 660 (NaOH titration, calcd. 680) and + 117° (c 1.2, water) indicating a-linkages. This was confirmed by the p.m.r. spectrum of a 4% solution in D;0 run at 95° (see Table I). The aldotetrauronic acid (2 mg) was hydrolyzed with 0.5 SI trilluoi oicetic acid at 100° for 1 h. Paper chromatography (solvent C) showed the presence of n-galactose as the principle, monosaccharide with traces of D-mannose and a series of aldobiouronic, aldotriouronic, and aldotelrauronic acids which had the same mobilities as the components of the original mixture. The aldotetrauronic acid (4 nig) was reduced with sodium borohydride and, aflcr removal of borate, the product was hydrolyzed with 2 St trifluoroacetic acid for 2 h. On paper chromatography (solvents A and /?), the neutral compounds found were D-mannose and galactitol. G.l.c. of the alditol acetates of the neutral compounds showed the ratio of galactitol and D-mannitol to be 1:1.3 and permitted characterization of each compound. The acidic sugats were D-glucuronic acid and the aldobiouronic acid. The aldotetrauronic acid (II mg) was methylated and then hydrolyzed by. 2 St trifluoroacetic acid for 2 h. Paper chromatography (solvents D and E) showed the following partially methylated sugars to be present 3,4,6-tri-O-mcihyl-D-mannose (R, 0.58, solvent D), 2,4,6-tri-O-methyl-D-mannose {R, 0.55), 2,4,6-tri-C>-methyl-D-galactose (R, 0.40), 2,3,4-lri-6>-me!hyl-D-glucuronie acid (R, 0.22, solvent /;). and traces of partially methylated aldobiouronic acid (R, 0.2S, solvent /:'). The identities of the neutral sugars were further characterized as the alditol acetates and the ratio was found to be approximately 1:1:1. The aldotriouronic acid (AY,.,, 0.42) had Ihc same mobility on paper as a component obtained by partial hydrolysis of the telrauronic acid and was not further examined. ,80 i CAN. J. CHEM. VOL. 51, 1973 206 The aklobiouronie acid (12 mg) had [i]0 +40° (r 0.62, water). A sample (2 mg) was hydrolyzed with 3 M hydrochloric acid for 2 h. The only neutral sugar found on paper chromatography (solvent li) was D-mannosc. A negative result was obtained when the aldobiouronic acid (4 mg) was incubated with P-D-glucuronidase (3 mg) at pll 4.8 at 381 overnight. Under the same conditions the enzyme hydrolyzed 3-0-(ft-o-glueopyranosyluronic acid)-D-ga!actosc (from Klebsiella K20, rcf. 8) and 6-0-(P-o-glueopyranosy 1 uronic acid)-n-galactose. When a portion (5 mg) of the aldobiouronic acid was methylated and then hydrolyzed by 3.5 M hydrochloric acid at 95° for 3 h paper chromatography showed the presence of 2,4,6-tri-O-methyl-D-mannose and 2,3,4-lri-O-mcthyl-D-glucuronic acid which had the same mobility as a standard (R, 0.22, solvent E). The mannose derivative was further characterized as the alditol acetate, m.p. and mixed m.p. 64-65°. Periodate Oxidation The polysaccharide (0.25 g) was dissolved in aqueous sodium metapcriodate (100 ml, 0.025 M) and 0.36 mol of periodate per hexose Unit was consumed after 3 days. Sodium borohydride (I g) was added and the solution was left overnight. The solution was c .ionized with Amberlite IR 120, freeze-dried, and the product was distilled with several portions of methanol. The polyalcohol was dissolved in water (3 ml) to give a solution of p/f 2.2 which was heated at 95" for 1 h. On paperchromatography (solvent A) the following compounds were identified by comparison with standards: {a) pyruvic acid (Rc 0.74), (b) glycerol (R, 0.44), (c) D-thrcitol (Rf 0.30) (crythritol had R( 0.32), (d) D-galactose (in traces only), and (e) an oligosaccharide which had a mobility intermediate between the aldobiouronic and the aldotriouronic acids in solvent E. The ratio of glycerol to D-threitol was found to be 1.2:1 by acetylation of part of the mixture and g.l.c. (column b, 150-1 S O 3 at 27min) . A control reaction showed that crythritol tetraacetate crystallized very easily while D-threitol tetraacetate remained as a syrup; both acetates had distinctly different retention times (13.5 and 15 min, respectively). The oligosaccharide was obtained by precipitation into ethanol, yield 0.1 g. The product was reduced with sodium borohydride and, after removal of borate, the p.m.r. spectrum was run in D:0 (20 mg/5 ml) at 95". One p-D- and two a-D-linkages were observed (Table 1). The signal at r 5.47 was a doublet with a coupling con-stant of 8 Hz. The reduced oligosaccharide (10 ml) was dissolved in 0.5 M phosphate buffer (3 ml, pit 7.0), P-D-galactosidase (5 mg) was added, and the solution was incubated at 38° overnight. The monosaccharide liberated was D-galactose which was confirmed by g.l.c. of the alditol acetate, m.p. and mixed m.p. 160-162'. The reduced oligosaccharide (50 mg) was methylated followed by hydrolysis with 2 A7 trifiuotoacetie acid for 2 h. Paper chromatography (solvent D) showed, by comparison with a standard, that 2,3,4,6-tctra-O-methyl-D-galactose (Rf 0.68) was the major neutral sugar with traces of 2,4,6-tri-O-methyl-D-mannose. The tctramethyl-galactitol acetate also had the same retention time as authentic material (II min, column a at 200'). In solvent E the acidic components had the same mobilities as 2,4-di-O-mclhyl-D-glucuronic acid (A', 0.16) and 3-0-(2,4-di-C-methyl-a-D-glucopyranosyluronic acid)-2,4,6-tri-0-methyl-D-mannose (R, 0.23) previously isolated from the methylated degraded polysaccharide. We are indebted to Dr. f. Orskov, Copenhagen, for the gift of an authentic culture (1702/49) of Klebsiella K2I, to Dr. P. J. Salisbury of this Department for growing the bacteria, and to Dr. G. H . N. Towers of the Department of Botany, University of British Columbia, for the use of his centrifuge. Gel filtration measurements were kindly made by Dr. S . C. Churms (Cape Town) and electro-phoresis was carried out by Dr. P. li. Rcid and Miss C. Podcr of the Department of Pathology, University of British Columbia. P.m.r. spectra were determined by Miss P. Watson and Mr. A. Brooke of this Department. The financial support of the National Research Council of Canada, and the award of a Killam Fellowship to Y.M.C. arc gratefully acknowledged. 1. W. N I M M I C H . Z. Med. Mikrobiol. Immunol. 154, 117 (1968). ' 2. W. N I M M I C H . Acta Biol. Med. Ger. 26, 397 (1971). 3. W. N I M M I C H . Personal communication. 4. L. C. G A I I A N , P. A. SANDI -ORD, and H . E. C O N R A D . Biochem. 6, 2755 (1967). 5. G. G. S. D U T T O N and M. T. Y A N G . Can. J. Chem. 50, 2382 (1972). 6. I. W. S U T H E R L A N D . Biochem. 9, 2180(1970). 7. B . L I N D B E R G , J. L O N N O R E N , W. N I M M I C H , and J. L. T H O M P S O N . Carbohydr. Res. In press. 8. Y . M. C H O Y and G. G. S. D U T T O N . J. Bacteriol. 112, 635 (1972). 9. 1. W. S U T H E R L A N D and J. F. W I L K I N S O N. Biochem. J. 110,749 (1968). 10. P. A. S A N D F O R D and H . E. C O N R A D . Biochem. 5, 1508 (1966). 11. H . E. C O N R A D , j. R. H A M B U R G , J. D. E P L E Y , and T. J. K I N D T . Biochem. 5, 2S08 (1966). 12. D. E. K O E L T Z O W und H . E. C O N R A D . Biochem. 10, 214(1971), 13. E. C. Y U R E W I C Z , M. A. G H A L A M B O V , and H . C H E A T H . J. Biol. Chem.-246, 5596(1971). 14. M . P. V E N I A M I N . J. Chromatogr. Sci. 8, 173 (1970). 15. G. G. S. D U T T O N and Y . M. C H O Y . Carbohydr. Res. 21, 169 (1972). 16. P. A. J. G O R I N and T. ISHIKAWA. Can. J. Chem. 45 , 521 (1967). 17. C. J. L A W S O N , C. W. M C C L E A R Y , H . (. N A K A D A , D. A. R E E S , I. W . S U T H E R L A N D , and J. F . W I L K I N S O N . Biochem. J. 115, 935 (1969). 18. G. M. B E B A U L T , J. M. B E R R Y , Y . M. C H O Y , G. G. S. D U T T O N , N . F U N N E L L , L. D . H A Y W A R D , and A. M. S T E P H E N . Can. J. Chem. 51, 324 (1973). 19. M. D U C K W O R T H and W . Y A P H E . Chem. ind. (London), 747 (1970). 20. H . T. O P E N S H A W . Qualitative organic analysis. Cam-bridge, 1946. 21. S. H A K O M O R I . J. Biochem. (Tokyo), 55 , 205 (1964). 22. G. G. S. D U T T O N and S. K A B I R . Anal. Lett. 4 , 95 (1971). 23. C. G A L A N O S , O. L U D K R I T Z , and K . H I M M E L S P A C H . Eur. J. Biochem. 8 , 332 (1969). 24. T. J. P A I N T E R. Chem. Ind. (London), 1214 (1960). CHOY AND DUTTON: POLYSACCHARIDES 207 25. O . P E R I L A and C. T . BISHOP. Can. i. Chem. 3 9 , 815 (1961). 26. G . A . A D A M S and R. Y O U N G . Can. J . Chem. 43 , 2940 (1965). 27. H . O . Bouyr.NG, I. BREMNF.R, and B . L I N D B E R G . Acta Chem. Scand. 19, 1003 (1965). 28. G . G . S. D U T T O N and K. B . G I B N E Y . Carbohydr. Res. In press. 29. P. A . J . G O R I N , J . F . T . S P E N C E R , and S. S. B H A T T A C H A R J E E . Can. J. Chem. 47 , 1499 (1969) and refs. therein. 30. P . G . J O H N S O N and E. P E R C I V A L. J. Chem. Soc. C, 906 . (1969). 31. L. H O U G H , J. K. N. JONES, and W . H . W A D M A N . J . Chem. Soc. 1702 (1950). 32. W . E. T R E V E L Y A N , D . P. PROCTOR, and J . S. H A R R I S O N. Nature, 166,444 (1950). 33. G . G . S. D U T T O N and K. B . G I B N E Y . J . Chromatogr. 72 , 179(1972). 34. H . B J O R N D A L , C. G . H E L L E R Q V I S T , B. L I N D B E R G , and S. S V E N S S O N . Angew. Chem. Int. Ed. 9 , 610 (1970). 35. S. A L L E N , T . G . B O N N E R , E. J. B O U R N E , and N. M . S A V I L L E . Chem. Ind. (London), 630 (1958). 36. J . T . R. C L A R K E , L . S . W O L F E , and A. S. P E R L I N . J . Biol. Chem. 246, 5563 (1971). 37. L . H O U G H and A.C. R I C H A R D S O N . In Rodd's Chemis-try of carbon compounds. 2nd ed. Vol. IF . Elsevier, Amsterdam. 1967. p. 22. 38. J . D . G E E R D E S , B. A. L E W I S , R. M O N T G O M E R Y , and F . S M I T H . Anal. Chem. 26, 264 (1954). 39. Y . M . C H O Y and A. M . U N R A U . Carbohydr. Res. 17, 439 (1971). 40. D . J. B E L L and S. W I L L I A M S O N . J. Chem. Soc. 1136 (1938). APPENDIX II The Structure of the Capsular Polysaccharide Klebsiella K-Type 24 Can. J . Chem., in press. 83 T h e S t r u c t u r e 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 of K l e b s i e l l a K - t y p e 24 Y u e n - M i n C h o y , Guy G . S . D u t t o n a n d A l b e r t o M . Z a n l u n g o D e p a r t m e n t o f C h e m i s t r y , T h e 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 , V a n c o u v e r , B r i t i s h C o l u m b i a . [ Q u a n t i t a t i v e a n a l y s i s o f s u g a r c o n s t i t u e n t s a n d p a r t i a l h y d r o l y s i s w e r e d o n e b y D r . A . M . Z a n l u n g o ] . ' ' -[ M e t h y l a t i o n a n a l y s i s a n d p e r i o d a t e d e g r a d a t i o n w e r e d o n e by. Y . M . C h o y ] . On l e a v e f r o m U n i v e r s l d a d T e c h n i c a d e l E s t a d o , S a n t i a g o , C h i l e . Methylation, periodate oxidation and partial hydrolysis studies on the capsular polysaccharide, and on the carboxyl reduced polymer, of Klebsiella K24 show the structure to consist of a repeating unit ? 13 12 13 1 — - D-GlcAp — D-Manp — D-Manp =-=• D-Glcp -~-= Aa~a = *- a =* ** 3 3 i \ D-Manp p The anomeric linkages were determined in Isolated oligosaccharides by p.m.r. spectroscopy which also showed the presence in the polysaccharide of one O-acetyl group per 7-8 sugar residues. The 0,-acetyl is tentatively assigned to one of the D-mannose units. Les methodes de methylation, oxydation periodique et hydrolyse partlelle portees au polysaccharide capsulaire, et au polymere avec le groupe carboxyle reduit, de Klebsiella K24 ont demontre' que la structure se compose d'une unit£ qui se r£p£te. Les liaisons anomeres ont 2 13 12 13 1 — - D-GlcAp — D-Manp ±-=- D-Manp =^=- D-Glcp =—-D-Mang 6te distlngu£es chez des oligosaccharides par la spectroscopic r.m.n. qui a aussi mis en Evidence dans le polysaccharide la presence d'un groupe 0~«ieetyle par rapport a 7-8 sucres. On pense que le groupe 0-acety.!c ^st lie a un des restes de D-mannose. \ - 3 -85 Qualitative analyses of the capsular polysaccharides of the 80 serotypes of Klebsiella bacteria have been provided by Nimmich (1,2). In continuation of our work (3) on the structure of these materials we now report studies on the capsular polysaccharide from Klebsiella K-type 24. The polysaccharide, purified by precipitation with Cetavlon, moved as a single band on electrophoresis on cellulose acetate and 27 had Ict]n +79p and an equivalent weight (by titration) of 786. The p.m.r. spectrum of a 2% solution of the sodium salt of 'the polysaccharide in run at 95° showed a sharp singlet at T 7.8 characteristic of O-acetyl and the absence of pyruvate (4,5). The anomeric region of the spectrum suggested that the repeat unit consists of five sugar units of which three are linked by a-D- and two by g-D-glycosidic bonds' (5). Integration of the anomeric and acetate signals indicated one acetyl group to seven or eight sugar units. The presence of £~acetyl groups was also confirmed by formation of the hydroxamic ester (6). Acid hydrolysis of the polysaccharide showed the rapid liberation of D-mannose and after 4 h D-mannose and D-glucose were proved to be the only neutral sugars present, in the ratio of 2.7:1. This analysis was carried out on the derived alditol acetates samples of which were collected by g.l.c. and measurement of their c d . spectra confirmed the assignment of the D-configuration to both sugars (7). Partial hydrolysis of the polysaccharide gave a series of acidic oligosaccharides which were separated from the neutral sugars by ion-exchange resins and individual components were obtained by paper chromatography. These were shown to be an aldobiouronic acid together - 4 - 86 with related aldotri- and ?!doietra-uronic acids. The structures of these oligosaccharides (as their alditols) are given in Table 1 together with the chemical shift of the anomeric proton(s) which enables the nature of the glycosidic linkages to be determined. The structures given were determined by (a) hydrolysis after conversion to the a ldi to l , (b) hydrolysis after carboxyl reduction and (c) methylation as described in the experimental. The aldobiouronic acid 3-0-(ct-D-glucopyranosyluronic acid)-D-mannose has been found in Klebsiella K2 and K21 and elsewhere (3). When a sample of the fully methylated (8,9) polysaccharide was hydrolyzed and the neutral sugars were examined 2,3,4,6-tetra- and 3,4,6-tri-O-methyl-D-mannose and 2,4,6-tri-()-methyl-D-glucose were found together with a small quantity of 2,4,6-tri-O-methyl-D-mannose. An attempt to analyze this mixture quantitatively as alditol acetates was not satisfactory since on columns of ECNSS-M and butanediol succinate the 2,4,6- and 3,4,6-trimethylhexoses were not resolved. On a column of OS 138 these components gave two major peaks poorly resolved and a minor one (ca 9% of the trimethyl fraction) which was identified by its retention time and mass spectrum as 2,4,6-tri-O-methy1-D-mannose. A good separation of the main components was obtained by chromatographing the trimethylsilyl derivatives (10) of the methylated sugars. This was done using three different columns and authentic standards as shown in Table 2. Under these conditions 2,3,4,6-tetra-fJ-methyl-D-mannose was clearly distinguished from the D-glucose isomer but i t was not possible to estimate accurately the small amount of 2,4,6-tri-O-methyl-D-mannose. The ratio of 2,3,4,6-- 5 - 87 tetra-O^methyl-D-mannose: 3,4,6-tri-O-methyl-D-mannose: 2,4, 6-tri-0>-methy1-D-glucose was 1.1:1:1. A l t e r n a t i v e l y I t has been found (11) that methylated sugars may be separated conveniently as t h e i r acetates. When t h i s method was employed here three main groups of peaks (A; B^, B^', C^, C^) in approximately equal proportions were obtained; JS^ was present in very small amount and was not examined f u r t h e r . Samples were c o l l e c t e d of the other components, deacetylated and examined on paper. Further samples were reduced, acetylated and examined by g.l.c.-m.s. and each of the main components thus obtained was demethylated and converted to the h e x i t o l peracetate. A l l of these r e s u l t s were consistent with the assignment of peaks A, B and C as 2,3,4,6-tetra-O-methyl-D-mannose (A), 3,4,6-tri-£-methyl-D-mannose (B.^ and 2,4,6-tri-£-methyl-D-glucose (C^ and C^). The methylated a l d i t o l acetate obtained from B^ showed an a d d i t i o n a l minor peak whereas both components and gave the same single a d i t o l acetate. This indicates that and C2 are the two anomers of the trimethylglucose and suggests that the small amount of 2,4,6-tri-£-methyl-D-mannose has a s i m i l a r retention time to the 3,4,6-isomer. I t may be noted that i n t h i s instance the mannose ethers have shorter retention times than the glucose compound but f o r the 2,3,6-tri-jO-methyl derivatives the s i t u a t i o n i s reversed (11). These r e s u l t s show the presence of a D-mannose unit substituted at p o s i t i o n 2 and a D-glucose unit substituted at p o s i t i o n 3 together with a D-mannose residue present as a side chain. The small amount of 2,4,6-tri-O-methyl-D-mannose undoubtedly arose by p a r t i a l cleavage of the aldobiouronic acid linkage. Another sample of the fully methylated polysaccharide was reduced with lithium borohydride and hydrolyzed. In addition to the methylated sugars described above a new component having the mobility on paper of 3- or 4-£-methyl-D-glucose was obtained. The mixture of methylated sugars was reduced with sodium borodeuteride (12) and the new component was isolated by g.l.c. of the alditol acetates. The mass spectrum clearly identified this component as a 3~0-methyl hexitol. Furthermore the c d . spectrum of the alditol acetate was identical in sign (negative) with that obtained from authentic 3-()-methyl-p-glucose. This establishes the D-configuration of the glucuronic acid in this polysaccharide (7). In order to determine the position of attachment of the pendant D-mannose units to the main chain a sample of the original polysaccharide was subjected to mild hydrolysis and the recovered polymeric material was methylated, reduced with lithium borohydride and hydrolyzed. Examination of the hydrolyzate by paper chromatography showed, in addition to the compounds previously obtained, a component having the mobility of 3,4-di-fJ-methyl-D-glucose, This was confirmed by g.l.c. separation of the alditol acetates and mass spectrometry of the dimethyl fraction together with demethylation and acetylation to give D-glucitol hexaacetate. The ratio of the t r i - to dimethyl fraction was 3.2:1. This indicates that the D-mannose side chains are joined to position 4 of D-glucuronic acid which, in turn, must be linked through position 2 in the main chain. - 7 - 8 9 The mcthylation data i n conjunction with the r e s u l t s of analysis of the a c i d i c oligosaccharides enable the repeat u n i t of K l e b s i e l l a K24 capsular polysaccharide to be wri t t e n thus: 2 D-GlcAp 1 3 1 2 ; 1 3 1 a a a 3 Confirmation of t h i s structure was sought by the Smith degradation procedure (13). Deacetylated polysaccharide was oxidized with periodate when 0.66 mole per hexose unit was consumed. The derived polyalcohol was p a r t i a l l y hydrolyzed by t r i f l u o r o a c e t l c acid at room temperature to give a mixture of g l y c e r o l and an oligosaccharide which was then reduced with sodium borohydride., The l a t t e r compound was methylated reduced and hydrolyzed to give 2,3,4,6-tetra- and 3,4-di-O-methyl-D-glucose and 2,4,6-tri-0-methyl-D-mannose as expected from the proposed structure; the methylated g l y c e r o l d e r i v a t i v e was not sought. An attempt was made to locate the p o s i t i o n of the 0-acetyl group using the procedure of de Belder and Norrman (14). The polysaccharide was dissolved i n methyl sulfoxide and reacted overnight with methyl v i n y l ether. Excess ether was removed i n vacuo and the polysaccharide was treated d i r e c t l y with dimethsulfinyl anion and methyl i o d i d e . When the product was hydrolyzed and the components were examined by paper chromatography a major spot f o r monosaccharide was obtained together with a f a i n t spot with an greater than a l l monomethylglucose derivatives and which was therefore assumed to be a monomethylmannose. Analysis - 8 - 90 by gj,I.e. of the hydrolyzate, as alditol acetates, suggested that there might be two monomethyl mannoses present. Even i f the experiment was to be repeated on a larger scale to allow characterization of the monomethyl sugars their identification as mannose derivatives would s t i l l not permit any deduction as to which of the three structurally different mannose residues was acetylated in the original polysaccharide. For this reason and because of the low yields obtained this aspect of the structure was not pursued further. The use of methyl vinyl ether as a reagent in locating O-acetyl groups has been applied successfully to lipopolysaccharides ( 1 5 ) which appear to have lower molecular weights than the capsular polysaccharides of Klebsiella 5 (M.W. 5 to 9 x 10 ). This difference in molecular weight may account for the low yield obtained in the present instance; similar difficulties have been experience with other capsular polysaccharides from Klebsiella (16). There is good evidence that these capsular polysaccharid' have a true repeating unit and also that where a sugar residue is in. the form of a pyruvic acid ketal (1-carboxyethylidene derivative) this feature similarly repeats regularly ( 5 ) . In the case of partial acetylation the substitution pattern appears to be less well defined. The evidence presented shows clearly that the structure of the capsular polysaccharide of Klebsiella K-type 24 is as given above with the tentative assignment of some 0-a c e tyl groups on certain of the mannose residues. c • . 91 - 9 -EXPERIMENTAL General methods are as previously described ( 3 ) . I s o l a t i o n and Properties of K24 Capsular Polysaccharide K l e b s i e l l a K24 was grown i n the medium (3) as for K21 and the harvested capsular polysaccharide and c e l l s (1 1/2 1) were di l u t e d 5 f o l d and centrifuged at 27,000 r.p.m. at 20° for 45 minutes. The supernatant was c o l l e c t e d and freeze-dried. A 1% aqueous s o l u t i o n of the product (1 v o l . ) was added with s t i r r i n g to ethanol (6-7 v o l . ) . The polysaccharide was c o l l e c t e dt washed with acetone and a i r d r i e d . Y i e l d ca. 12 g. The crude product was p u r i f i e d through p r e c i p i t a t i o n with Cetavlon, giving 7 g of pure a c i d i c polysaccharide . having [ a ] ^5 +90° (c 0.54, water) as Na+ s a l t ; [a]*7 +79° (c 0.68, water) a f t e r d i o n i z a t i o n with Amberlite IR 120.. Equivalent weight by t i t r a t i o n with 0.03 N sodium hydroxide (phenolphthalein) was 786. Addition of excess sodium hydroxide and back t i t r a t i o n with hydrochloric acid a f t e r s t i r r i n g under nitrogen f o r 50 min showed the presence of ca, 4% 0-acetyl. Electrophoresis at pH 3 and 8.8 showed the polysaccharide moved as a single component. Anal. N , 0%; ash, 0%. A sample of polysaccharide was exactly neutralized and the s o l u t i o n was l y o p h i l i z e d . The residue was dissolved i n D^ O and exchanged twice by l y o p h i l i z a t i o n . The p.m.r. spectrum of the so l u t i o n (ca. 2% i n Vrfi) was run at 95° and showed a sharp peak at x 7.8 and f i v e protons i n the range T 4.5-5.5 (see Table 1 ) . - 10 -92 The presence of acetate was also confirmed by methanolysis of the polysaccharide and i d e n t i f i c a t i o n of methyl acetate using the hydroxylamine-ferric chloride reagent (6,16,17)• Hydrolysis of the Polysaccharide Polysaccharide (25 mg) was dissolved i n s u l f u r i c acid (2.5 ml, N) at room temperature (30 min) and the solu t i o n was heated on a steam bath for 4 h. The s o l u t i o n was neutralized (BaCO^) and the f i l t r a t e was evaporated. On paper chromatography i n solvents A and B spots corresponding to D-glucose, D-mannose and D-glucuronic acid were observed. The hydrolyzate was separated into neutral and a c i d i c f r a c t i o n s using ion-exchange resins and the neutral f r a c t i o n was found to contain D-mannose and D-glucose i n an approximate r a t i o of 2.7:1 as determined by the g . l . c . of the alditol.acetates(column a); D-mannitol hexaacetate had m.p. and mixed m.p. 121.5-122° and D-g l u c i t o l hexaacetate had m.p. and mixed m.p. 9 8 . 5 - 9 9°. P a r t i a l Hydrolysis of the Polysaccharide The polysaccharide (1.0 g) was dissolved i n t r i f l u o r o a c e t i c acid (TFA, 100 ml, 2M) and the solution was heated on a steam bath f o r 4 h. The sol u t i o n was evaporated, dissolved i n 5 ml of water and evaporated again. The r e s u l t i n g syrup was separated by ion-exchange resins i n t o neutral and a c i d i c (eluted with 4 M HOAc) f r a c t i o n s . The l a t t e r was concentrated (300 mg) and chromatography i n solvent A showed four components: (a) D-glucuronic a c i d , (b) aldobiouronic acid (R , 0.43), (c) aldotriouronic acid (R .. 0.13) and -glucose , -glucose - 11 - 9 3 (d) aldotetrauronic acid (R - 0.04). The syrup was separated V -glucose •> r r by preparative paper chromatography on Whatman 3 mm paper in solvent C for 80 h. Yields: (a) aldobiouronic acid, 130 mg; (b) aldotriouronic acid 50 mg; (c) aldotetrauronic acid 25 mg. Structural Analysis of Aldobiouronic Acid (a) Anomeric linkage A sample of aldobiouronic acid (25 mg) was dissolved in water (1 ml) and reduced with sodium borohydride (25 mg) for 8 h. After workup and removal of borate, the syrup was dissolved in D2O (2 ml) and freeze-dried. P.m.r. of this product (in I>2°) showed a doublet at T 4.85, J % 3.5 Hz, indicating an ct-linkage in the disaccharide. Cb) Reducing end A portion of the above reduced aldobiouronic acid was hydrolyzed with trifluoroacetlc acid (2 M) at 100° for 1 h. Uronic acids were removed with anion exchange resin and the neutral fraction was concentrated and acetylated. G.l.c. (column a) gave mannitol hexaacetate m.p. 122°. (c) Carboxyl reduction and hydrolysis of the disaccharide Aldobiouronic acid (20 mg) was dissolved in anhydrous methanol (15 ml) containing 1% hydrogen chloride and the solution was left 12 h at room temperature. The solution was neutralized (Ag^CO^), centrifuged, and evaporated. The resulting syrup was dissolved in the minimum amount of water, followed by the addition of methanol and the solution was added dropwise over 1 h to a stirred solution of sodium borohydride (25 rag) in a mixture of water (0.5 ml) and methanol (2 ml). The - 12 - 94 solution was left 12 h at room temperature nnd. was then carefully neutralized and stirred with Amberlite IR-120 (H ) resin until the solution was free of Na . A portion of the syrup, after removal of borate, was hydrolyzed with trifluoroacetic acid. (2 M) at 100° for 1 h and was shown (solvent A) to give D-glucoeo and D-mannose. (d) Methylation Aldobiouronic acid (12 mg) was methylated according to the procedure of Hakomori and was hydrolyzed (HCI, 3.5 N) to give 2,4,6-tri-O-methyl-D-mannose (Rf 0.55, solvent D) and 2,3,4-tri-£-methyl-D-glucuronic acid (R^ 0.22, solvent E). The former compound was also confirmed by g.l.c.-m.s. of the alditol acetate. Structure of Aldotriouronic Acid (a) Anomeric linkage A sample of aldotriouronic acid (20 mg) was reduced and dissolved in D20. The signals at x 4.7, J = 3.5 Hz (from D-glucuronic acid) . and x 5.0, J = 1.5 Hz indicate ct-linkages (Table 1). (b) Reducing end Hydrolysis of the reduced aldotriouronic acid-showed D-mannitol (paper, solvent B and g.l.c.) and the same aldobiouronic acid (paper, solvents A and C). (c) Methylation The methyl ester methylglycoside (10 mg) was methylated and on hydrolysis showed (solvent D) two overlapping spots corresponding to 3,4,6- and 2,4,6-tri-£-methyl-D-mannose (R^  0.58 and 0.55, respectively). G.l.c. of the trimethylsilyl derivatives on column a at 135° and on column b at 120° confirmed these assignments and showed them to be present in approximately equimolecular amounts. Structure of Aldotetrauronic Acid (a) Anomeric linkage A portion of the aldotetrauronic acid (15 mg) was reduced as for the aldobiouronic acid. The p.m.r. spectrum in D^ O showed the following signals: T 4.70, J = 3.5 Hz and x 4.73, J = 3.0 Hz (2H); x 4.95, J = 1.5 Hz (IH). (b) Reducing end.. A portion of the reduced aldotetrauronic acid was hydrolyzed and paper chromatography (solvent B) showed D-glucitol, confirmed by g.l.c. of the alditol acetate. Also produced were the same aldobi-and aldotriouronic acids (solvent C) described above. Methylation Dry polysaccharide (1.7 g) was dissolved in anhydrous methyl sulfoxide (300 ml) under nitrogen ( 8 , 9 ) . To this solution methyl sulfoxide anion (26 ml, 2 M) was added and the mixture was -stirred for 9 h under nitrogen. Methyl iodide (15 ml) was added at such a rate as to keep the temperature below 20° and the resulting clear solution was stirred for 10 h. The solution was dialyzed against running water for 2 days and concentrated to a syrup which was methylated eight times with Purdie's reagents to give a product showing no - 1 2f) absorption at 3600 cm , [a]* +64.5° (c 1.2, chloroform), OMe, 43.2%. Purification on a silica gel column using chloroform-methanol (95:5) raised the methoxyl content to 44.3%. Methylated polysaccharide (60 mg) was dissolved in trifluoroacetic acid (2 M) and the solution was refluxed for 5 hB Separation on ion-exchange resins gave a neutral fraction which showed on paper chromatography (solvent D) three major components corresponding to 2,3,4,6-tetra-and 3,4,6-tri-fJ-methyl-D-mannose, 21;4,6-tri-£-methyl-D-glucose (R^  values 0.80, 0.58 and 0.50) and a minor component 2,4,6-tri-()-methyl= D-mannose (R^  0.55). A portion (10 mgT of the neutral fraction was reduced, acetylated and examined on column a programmed from 170-194° at 2°/min. Two peaks in the ratio of 1:1.8 were observed at 25.1 and 36.6 tnln. The faster component was tentatively Identified by its retention time and mass spectrum as 2,3,4,6-tetra-0_-methyl-D-mannose. The mass spectrum of the slower material showed lt to be a mixture of 2,4,6-and 3,4,6-trlmethylhexoses which also did not resolve on column b. On a column of OS-138 15% on Gas Chrom Q) operated iso.thermally at 240°. The slower moving fraction was shown to contain one minor and two major components. The minor component crystallized m.p. 63-66° (3)" and was further identified as 2,4,6-tri-0~methyl-D-mannitol triacetate by mass spectrometry; the main peaks were not sufficiently well resolved to permit identification. Resolution of the trimethylhexose fraction was obtained yhen the trimethylsilyl derivatives of the neutral sugars were run on cc limns a, b, and c (samples Injected in hexane, see Table 2 ). * Correct nomenclature: 1,3,5-tri-O-methyl-D-mannitol triacetate. - 15 - 97 When a portion (40 mg) of the neutral methylated sugars was acetylated (pyridine-acetic anhydride 1:1, 100°, 15 min), extracted with chloroform and injected onto column a at 170° the following peaks were obtained: A (5.8 min); (8.8), B2 (10.5); (16.2), C2 (17.6). The component B,> was present in only small amount and this, together with a minor component with a retention time of 7.5 min, was not examined further. Individual fractions were deacetylated (NaOMe, 1 N in MeOH, 25°, 30 min), deionized and examined on paper in solvent D. Components A, B^, and C2 had values of 0.79, 0.57, 0.49 and 0.49. Samples of the individual fractions were also reduced with sodium borohydride, acetylated and examined on column b at 175". Components A', B^ ', C^ ' and C2' were eluted at 7.8, 14.6, 15.8 and 15.8-min, each was examined by mass spectrometry and each was demethylated and the product acetylated. D-Mannitol hexaacetate m.p. 120-122° was obtained from A' and B^1 while C^ ' and C2' yielded D-glucitol hexaacetate m.p. 96-97°. Reduced Methylated polysaccharide Part (100 mg) o f the fully methylated polysaccharide was reduced overnight with l i t h i u m borohydride in tetrahydrofuran and the product was hydrolyzed by refluxing for 2 h with trifluoroacetic acid (2 M). Paper chromatography in solvent D revealed the presence of an extra 98 - 16 -sugar having the m o b i l i t y (R^ 0.08) of a moncnnethylhexosc wK)ch was shown i n Solvent E to have the same ch a r a c t e r ! s i i c s as 3- or 4-0-methyl-D-glucose. A portion (30 mg) of the hydrolyzate. was reduced with sodium borodeuteride, acetylated and separated on column b. Mass spectrometry of the component having a retention time of 38 min showed i t to be a 3-0-methylhexitol and the cd.. spectrum was i d e n t i c a l Me CN with that from 3-0-methyl-D-glucitol pentaac.eta.te (^£2^3 -0.29). Methylated Degraded Polysaccharide Capsular polysaccharide (1 g) was hydrolyzed overnight at 95° with t r i f l u o r o a c e t i c acid (0.4 M) using the apparatus of Galanos and co-workers (18). The recovered polysaccharide (200 mg). was methylated, reduced (LiBH^) and hydrolyzed (TFA, 2 M). Paper chromatography showed the presence of a new sugar having the m o b i l i t y of 3,4-di-O-methyl-p-glucose (R 0.26, solvent D; 0.51, solvent E). This was confirmed by the mass spectrum of the dimethyl f r a c t i o n , separated as a l d i t o l acetates on column a, and by conversion of t h i s f r a c t i o n to D - g l u c i t o l hexaacetate m.p. 9 8° . The c d . curve of 3,4-di-O-methyl-D-g l u c i t o l tetraacetate i s weakly negative. , Smith Degradation Capsular polysaccharide was deacetylated with aqueous sodium hydroxide, dialyzed and l y o p h i l i z e d . The product (130 mg) was dissolved i n sodium metaperiodate (20 ml, 0.05 M) and after 3 days i n the dark at 4° 0.66 mole of periodate had been consumed per sugar u n i t . Following - 17 - 9 9 the addition of ethylene glycol, dialysis, reduction with nod.tun, borohydride, dialysis, d e i o n i z a t i o T i . lyophilization and removal of borate the product was hydrolyzed (TFA, 0.5 M) at room temperature for 8 h. Paper chromatography in solvent A showed mainly glycerol and an oligosaccharide with traces of glucose. A part of the hydrolyz.rte was acetylated and gave a peak corresponding to glycerol triacetate (4 min) on column b at 1 5 0°. The majority of the hydrolyzate was reduced (NaBH^ ) and paper chromatography (solvent A) showed one component having R - 0.37 and glycerol, "galactose The mixture containing the oligosaccharide was methylated (8,9), reduced and hydrolyzed to give (solvent D) components equivalent to 2,3,4,6-tetra-and 3,4-di-£-methyl-D-glucose (Rf 0.80, 0.28) and 2,4,6-tri-£-methyl-D-mannose (R^ 0.56). This result was confirmed by g.l.c.-.m.s. of the alditol acetates (column a, 160-200° at 2°/min) which had retention times of 14.5, 30 and 21 min, respectively. Attempted Location of £-Acetyl Group Capsular polysaccharide (80 mg) was dissolved in methyl sulfoxide (20 ml) and anhydrous £-toluene sulfonic acid (ca. 20 mg) was added. Methyl vinyl ether was passed into the solution until i t turned yellow and the reaction was left overnight. Excess methyl vinyl ether was removed on a rotary evaporator and methyl sulfoxide anion was added directly without isolation of the polysaccharide. The solution was and shaken for 20 h, methyl iodide (5 ml) was added|the shaking continued for 3 h. The solution was dialyzed against running water, polymeric material formed from methyl vinyl ether was extracted with chloroform and the aqueous portion was lyophilyzed. Hydrolysis (TFA. 1 M. 100°) showed (solvent E) a major spot for monosaccharides together NWIth a f a i n t spot moving f a s t e r than a l l monomethyl$|ucose derivatives (R^ 0.35). G.l.c. of the a l d i t o l acetates gave, two small peaks at 21 and 23 min (column b, 190°) i n addition t o h e x i t o l acetates. ACKNOWLEDGEMENTS We are indebted to Dr. I . 0rskov, Copenhagen, for the g i f t of an authentic culture of K l e b s i e l l a K2A (1680/49), to Dr. P.J. Salisbury of t h i s Department' f o r growing the b a c t e r i a , to Drs. J.J.R. Campbell and Dr. J.B. Hudson of the Department of Microbiology, University of B r i t i s h Columbia, for the use of t h e i r c e n t r i f u g e s , and to Dr. C T . Bishop for an authentic sample of 3,4,6-tri-0_-methyl-D-mannose. The f i n a n c i a l support of the National Research Council of Canada and the award of a K i l l a m Fellowship to Y.M.C. are g r a t e f u l l y acknowledged. TABLE 1. P.m.r. data on Klebsiella K24 capsular polysaccharide and derived oligosaccharides. Repeating unit of compcvinc! T-Value (coupling constant, Hz) Ratio of integrals Proton / assignment (al l sugars have D-configuration) GlcA Mannitol a 4.85 (3.5) 1 3 a-GlcA Mannitol GlcA ^ Man Mannitol a a 4.7 (3.5) 5.0 (1.5) 1 1 1 3 a-GlcA Man 1 2 a-Man Mannitol 1 3 12 13 GlcA Man =-=• Man •=-=• Glucitol a a ex 4.7 + 4.73 4.95 (1.5) 1 3 a-GlcA, Man 1 3 a-Man Glucitol a-Man ^  ^  Man 2 A - 1 3 „ 1 2 M 1 3 , . 1 — GlcA Man Man Glc —r-, a a a 8 J 4 B Man 4.6 4.95 (1.5) 5.27 (1.0) 5.31 (6.0) 1 1 1 a - G l c A + — Man 1 3 a-Man Glc 1 2 a-Man Man &-Man — GlcA -i 1 2 0-Glc --=• GlcA i -1 o i— • Spectra run in D^ o with external tetramethylsilane (t= 10) at 100 KHz). 102 TABLE 2. Per ( t r i m e t h y l s i l y l ) derivatives o{ methylated sugars from K l e b s i e l l a K24 capsular polypnceri«iride (a) ECNSS -M Columns (b) BDS (c) SE-52 Compounds Re ten r..i on times (min) Methylated aldoses of K24 14.4 19.8 4.6 7.8 17.6 16.6 21.4 11.3 24.1 3,4,6-Me3-Man 14,4 4.6 17.6 2S4S6-Me3-Glc 19.8 7.8 24.1 2,3,4,6-Me^ -Man 21.4 11.3 16.6 2,3,4,6-Me^-Glc 15.4 (20%) 7.4 (15%) 17.0 (80%) 8.2 (85%) - 1 9 - 1 0 3 1. W. NIMMICH. Z. Med. Mikrobiol. Immunol. 154, 117 (39^8). 2. W. NIMMICH. Acta biol. med. germ. 26, 397 (1971). 3. Y . M. CHOY and G. G. S. DUTTON. Can. J. Chem. in press (Jan. 15 1973). 4. Y. M. CHOY, G. G. S. DUTTON, A. M. STEPHEN and M. T. YANG. Anal. Lett. 5_, 6jS (1972). 5. G. M. BEBAULT, Y. M. CHOY, G. G. S. DUTTON, N. FUNNELL, A. M. STEPHEN and M. T. YANG. J. Bacteri.pl. In press. 6. M. ABDEL-AKER and F. SMITH. J. Amer. Chem. Soc. 73, 5859 (1951). 7. G. M. BEBAULT, J. M. BERRY, Y. H. CHOY, G. G. S. DUTTON, N. FUNNELL, L. D. HAYWARD and A. M. STEPHEN. Can. J. Chem. In press, 8. S. HAKOMORI. J. Biochem. (Tokyo), 55, 205 (1964). 9. P. A. SANDFORD and H. E. CONRAD. Biochem. 5>, 1508 (1966). 10. C. C, SWEELEY, R. BENTLEY, M. MAKITA, and W. W. WELLS. J. Am. Chem. Soc. 85,.2497 (1963). 11. G. M. BEBAULT, G. G. S. DUTTON and R. H. WALKER. Carbohydr. Re's. , • 23, 430 (1972). 12. H. BJORNDAL, C. G. HELLERQVIST, B. LINDBERG, and S. SVENSSON. Angew. Chem. Int. Ed. 9_> 61° (1970). 13. I. J. GOLDSTEIN, G. W. HAY, B. A. LEWIS, and F. SMITH. Methods Carbohydr. Chem. 5>, 361 (1965). 14. A. N. de BELDER and B. N0RRMAN. Carbohydr. Res. _8, 1 (1968). 15. C. G. HELLERQVIST, B. LINDBERG, K. SAMUELSON, and R.R. BRUBAKER. Acta Chem. Scand. '26, 1389 (1972). 16. G. G. S. DUTTON and M. T. YANG. Can. J. Chem. Submitted. 17. J. H. SLONEKER and A. JEANES. Can. J. Chem. 4_0, 2066 (1962). 18. C. GALAN0S, O. LUDERITZ, and K. HIMMELSPACH. Eur. J. Biochem. 8. 332 (1969). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0060204/manifest

Comment

Related Items