Spectroscopic characterisation of cell-surface carbo-hydrates: lactonisation of ketodeoxyoctonate

• Published in: Carbohydrate research 1980-05, Vol.81 (2), p.295-303
• Main Authors: Melton, Laurence D., Morris, Edwin R., Rees, David A., Thom, David, Bociek, Stephen M.
• Format: Article
• Language: English
• ISSN: 0008-6215; 1873-426X
• DOI: 10.1016/S0008-6215(00)85660-8

Ketodeoxyoctonate of bacterial cells shows close structural analogy to N-acetylneuraminic acid of animal cell-walls. In neutral solution, their circular dichroism (c.d.) and 300-MHz p.m.r. spectra show the expected similarities, although the proportion of the less-stable anomer (axial carboxyl) is appreciably higher in ketodeoxyoctonate. On acidification, however, ketodeoxyoctonate shows an unexpected reversal of sign in c.d., and the n.m.r. spectrum becomes complex, suggesting the co-existence of more than one molecular species. This inference was confirmed by thin-layer chromatography, which showed two distinct bands under acid conditions, and only one at neutral pH. On partial re-neutralisation (pH 5.5), reversal of the observed spectral changes was slow, indicative of a chemical process. I.r. spectra recorded at acid pH show bands consistent with the presence of an ester or lactone, in addition to the expected carboxyl bands. The magnitude of the c.d. changes on acidification argue against intermolecular ester formation, but are fully consistent with the radical change in sugar-ring geometry to be expected on lactonisation. I.r. and c.d. evidence suggests that the lactone formed spontaneously in acid solution is not the same as that previously identified on heating ketodeoxyoctonate at 95° in O.l M HCl. An alternative 1,5-lactone structure is proposed, which explains the resistance of N-acetylneuraminic acid to lactonisation, since the relevant hydroxyl group is replaced by the acetamido substituent.