Measurement of interglycosidic 3JCH coupling constants of selectively 13C labeled oligosaccharides by 2D J-resolved 1H NMR spectroscopy

Conformation of the O-specific polysaccharide of Shigella dysenteriae type 1: molecular modeling shows a helical structure with efficient exposure of the antigenic determinant alpha-L-Rhap-(1-->2)-alpha-D-Galp

The O-specific polysaccharide of Shigella dysenteriae type 1, which has the repeating tetrasaccharide unit -->3)-alpha-L-Rhap-(1-->3)-alpha-L-Rhap-(1-->2)-alpha-D-Galp-(1-->3)-alpha-D-GlcNAcp-(1--> (A-B-C-D), is a major virulence factor, and it is believed that antibodies against this polysaccharide confer protection to the host. The conformational properties of fragments of this O-antigen were explored using systematic search with a modified HSEA method (GLYCAN) and with molecular mechanics MM3(96). The results show that the alpha-D-Gal-(1-->3)-alpha-D-GlcNAc linkage adopts two favored conformations, phi/psi approximately equal to -40 degrees /-30 degrees (I) and approximately 15 degrees /30 degrees (II), whereas the other glycosidic linkages only have a single favored phi/psi conformational range. MM3 indicates that the trisaccharide B-C-D and tetrasaccharides containing the B-C-D moiety exist as two different conformers, distinguished by the conformations I and II of the C-D linkage. For the pentasaccharide A-B-C-D-A' and longer fragments, the calculations show preference for the C-D conformation II. These results can explain previously reported nuclear magnetic resonance data. The pentasaccharide in its favored conformation II is sharply bent, with the galactose residue exposed at the vertex. This hairpin conformation of the pentasaccharide was successfully docked with the binding site of a monoclonal IgM antibody (E3707 E9) that had been homology modeled from known crystal structures. For fragments made of repetitive tetrasaccharide units, the hairpin conformation leads to a left-handed helical structure with the galactose residues protruding radially at the helix surface. This arrangement results in a pronounced exposure of the galactose and also the adjacent rhamnose in each repeating unit, which is consistent with the known role of the as alpha-L-Rhap-(1-->2)-alpha-D-Galp moiety as a major antigenic epitope of this O-specific polysaccharide.

NMR spectroscopy, molecular dynamics, and conformation of a synthetic octasaccharide fragment of the O-specific polysaccharide of Shigella dysenteriae type 1

A synthetic octasaccharide fragment (2) of the O-specific polysaccharide (1) of Shigella dysenteriae type 1 has been studied as its methyl glycoside by one- and two-dimensional homo- and heteronuclear NMR spectroscopy. Complete 1H and 13C NMR assignments have been generated, and the 13C spin-lattice relaxation times have been measured for the octasaccharide 2. A congener (6) of this octasaccharide containing one D-galactose residue with a specific 13C label at C-1 has been synthesized and used to measure interglycosidic 13C-1H coupling by the 2D J-resolved 1H NMR method. From the NMR data, three types of conformational restraints were developed: (a) 29 inter-residue, distance restraints; (b) 48 intra-residue, ring atom dihedral angle restraints, and (c) one heteronuclear, inter-residue dihedral angle restraint. The use of these restraints in a restrained molecular dynamics computation with simulated annealing yielded a conformation resembling a short, irregular spiral, with methyl substituents on the exterior.

Protein conjugates of synthetic saccharides elicit higher levels of serum IgG lipopolysaccharide antibodies in mice than do those of the O-specific polysaccharide from Shigella dysenteriae type 1

Our development of vaccines to prevent shigellosis is based on the hypothesis that a critical (protective) level of serum IgG to the O-specific polysaccharide (O-SP) domain of Shigella lipopolysaccharide (LPS) confers immunity. The O-SP is a hapten and must be conjugated to a protein to induce serum antibodies. The O-SP of Shigella dysenteriae type 1 (approximately 27 tetrasaccharide repeat units), prepared by acid hydrolysis of the LPS, was bound to human serum albumin (HSA) by multiple point attachment (O-SP-HSA): The molar ratio of HSA to O-SP was 1.0. Synthetic saccharides, composed of one or multiples of the O-SP tetrasaccharide, equipped with a spacer at their reducing end, were bound to HSA by a single point attachment: The average molar ratios of the saccharides to HSA ranged from 4 to 24. Serum IgG anti-LPS, elicited in mice by O-SP-HSA or synthetic tetra-, octa-, dodeca-, and hexadecasaccharide fragments, was measured by ELISA. Outbred 6-week-old female mice were injected s.c. three times at biweekly intervals with 2.5 micrograms of saccharide as a conjugate and were bled 7 days after the second and third injections. Excepting the tetramer, conjugates of the octamer, dodecamer and hexadecamer elicited IgG LPS antibodies after the second injection, a statistically significant rise (booster) after the third injection, and higher levels than those vaccinated with O-SP-HSA (P = 0.0001). The highest geometric mean levels of IgG anti-LPS were elicited by the hexadecamer with 9 chains or 9 moles of saccharide/HSA (15.5 ELISA units) followed by the octamer with 20 chains (11.1 ELISA units) and the dodecamer with 10 chains (9.52 ELISA units). Clinical evaluation of these synthetic saccharides bound to a medically useful carrier is planned.