The Structures of Oligosaccharide Bisphosphates Isolated from the Lipopolysaccharide of a Recombinant Escherichia coli Strain Expressing the Gene gseA [3-deoxy-d-manno-Octulopyranosonic Acid (Kdo) Transferase] of Chlamydia psittaci 6BC

Abstract

The lipopolysaccharide from the recombinant strain Escherichia coli F515-140 containing the cloned gene gseA [3-deoxy-D-manno-octulopyranosonic acid (Kdo) transferase] from Chlamydia psittaci 6BC was isolated and sequentially de-O-acylated and de-N-acylated. The products were separated by high-performance anion-exchange chromatography into three fractions, two of which contained a single compound. Their structures were elucidated by high-field NMR spectroscopy as alpha-Kdo-(2-->4)-alpha-Kdo-(2-->6)-beta-D-GlcN-(1-->6)-alpha-D-GlcN 1,4'-P2 (compound 1) (tetrasaccharide bisphosphate) [Holst, O., Broer, W., Thomas-Oates, J. E., Mamat, U. & Brade, H. (1993) Eur. J. Biochem. 214, 703-710] and alpha-Kdo-(2-->4)-[alpha-Kdo-(2-->8)-]-alpha-Kdo-(2-->4)-alpha-Kdo- (2-->6)-beta-D-GlcN-(1-->6)-alpha-D-GlcN 1,4'-P2 (compound 4) (hexasaccharide bisphosphate). The third fraction comprised two pentasaccharide bisphosphates, which could be separated by affinity chromatography using an immobilized monoclonal antibody specific for the trisaccharide alpha-Kdo-(2-->8)-alpha-Kdo-(2-->4)-alpha-Kdo. The bound fraction was identified as alpha-Kdo-(2-->8)-alpha-Kdo-(2-->4)-alpha-Kdo-(2-->6)-beta-D- GlcN-(1-->6)-alpha-D-GlcN 1,4'-P2 (compound 2) [Holst, O., Broer, W., Thomas-Oates, J. E., Mamat, U. & Brade, H. (1993) Eur. J. Biochem. 214, 703-710], whereas the unbound fraction was identified as alpha-Kdo-(2-->4)-alpha-Kdo-(2-->4)-alpha-Kdo-(2-->6)-beta-D-GlcN-(1-->6 )- alpha-D-GlcN 1,4'-P2 (compound 3). This novel Kdo tetrasaccharide extends our knowledge on multifunctional Kdo transferases.

Deletion of the Heptosyltransferase Genes rfaC and rfaF in Escherichia Coli K‐12 Results in an Re‐Type Lipopolysaccharide with a High Degree of 2‐Aminoethanol Phosphate Substitution

“…To further investigate the possibility of phosphate migration, we treated in a first experiment each of the isolated and pure tetrasaccharide trisphosphates 2 and 3 again with 4 M KOH (120°C, 16 h) and separated the samples in HPAE. Both compounds remained unchanged, confirming our earlier findings (Holst et al, , 1995Bock et a]., 1994;Vinogradov et al, 1994) that phosphomonoesters do not undergo phosphate migration under strong alkaline conditions. However, in LPS 3-deoxy-D-munno-octulopyranosate is substituted at position 7 with 2-aminoethyl phosphate and thus, we hypothesized that migration could occur in case of a phosphodiester.…”

“…are summarized in Tables 1-3. The structure of tetrasaccharide bisphosphate compound 1 was readily identified since its elution time in analytical HPAE was identical to that of the same compound isolated from recombinant E. coli strains, and was finally proven by assignment of 'H-NMR and 13C-NMR chemical shift values and J-coupling constants which were similar to those reported [data not shown (Holst et al, , 1995l. The retention times of compounds 2 and 3 differed from those of compound 1 by approximately 2 min and 4 min, respectively.…”

“…It remained unclear from the above described results whether the identified phosphate substitution at either 0 7 or 0 8 of residue D was present in the native LPS or was a product of phosphate migration caused by the strong alkaline conditions used for deacylation. No such migration had previously occurred using our method for the preparation of oligosaccharide monophosphates and bisphosphates (Holst et al, , 1995Bock et al, 1994;Vinogradov et al, 1994). However, in this case two tetrasaccharide trisphosphates were isolated in which the third monophosphate group was localized in neighbouring positions ( 0 7 and 0 8 of the second Kdo).…”

“…Methyl 3-deoxy-D-manno-octulopyranoside 7-(2-acetarnidoethyl phosphate) (Holst et al, 1990) was obtained from D. Charon (Orsay, France). The following methods were performed as published : compositional analysis of LPS (Kaca et al, 1988), determination of the absolute configuration of GlcN and the configuration of Kdo (Kriille et al, 1993(Kriille et al, , 1994, GLC and GLC coupled to mass spectrometry [GLC-MS (Muller-Loennies et al, 1994)], analytical high-performance anion-exchange chromatography [HPAE (Holst et al, 1995)], and high-voltage paper electrophoresis and staining of the pherograms (Kaca et al, 1988). Semipreparative HPAE was performed on a Dionex DX 300 chromatography system using a Dionex CarboPac PA 1 column (9 mmX250 mm) eluted at 4 ml/min using eluents A, 0.1 M NaOH, and B, 1 M sodium acetate in A, and the gradient programme 30% to SO% B in 16 min, then 50% isocratically to 40 min, followed by 80% isocratically to 50 min.…”

Rhizobium rubiT: A Gram‐Negative Phytopathogenic Bacterium Expressing the Lewis B Epitope on the Outer Core of its Lipooligosaccharide Fraction

“…Ka was identified as an O7-substituted unit and its location at O4 of Kb was inferred from the NOE between its H6 with H3 eq of Kb. [8] In the case of Dha, proton H6 was poorly identifiable in the spectrum that was recorded at neutral pH because its signal was very broad; its position was deduced from the TOCSY and ROESY (Figure 4) spectra, and no correlation in the HSQC spectrum was present. The Dha analysis problem was solved by analysing the spectrum at alkaline pH (Table S2, Figure 3 B), in this case, the H6 signal appeared as a broad singlet and it correlated to C6 and exhibited the expected long-range correlation with a carbonyl at 177.0 ppm.…”