Lipopolysaccharide biosynthesis and protein glycosylation in Helicobacter pylori

Glycosylation requires the action of glycosyltransferases (GTs) to recognize both the activated-sugar donor and the corresponding acceptor to form a new glycosidic bond. The glycosyl donors are activated nucleotide diphosphate (or monophosphate) sugars, or dolichol-related lipid groups. The acceptor can be a monosaccharide, or an oligosaccharide, or protein or lipid.

Glycosylation is essential in Helicobacter pylori, as it is known to be involved in lipopolysaccharide (LPS) biosynthesis, peptidoglycan biosynthesis, flagellin glycosylation, and cholesterol glucosylation. As proteins other than flagellins, including RecA, catalase and AlpA/B, have been suggested to be glycosylated, and the structure and biosynthetic pathway for LPS in H. pylori remain to be determined, this study focused on protein glycosylation and elucidation of the structure, biosynthesis and biological function of H. pylori LPS.

Profiling protein glycosylation in H. pylori using bioorthogonal chemistry (click chemistry) was unsuccessful in this study. However, it was demonstrated that RecA in H. pylori is not a glycoprotein, and that the interaction of RecA with different lengths of LPS was shown to explain the molecular weight shift in RecA observed between wild-type H. pylori and associated LPS mutants. This involved experiments including purification and glycol staining of His6-tagged RecA from H. pylori G27 wild-type and LPS mutants, and LPS titration assays that showed a progressive increase in the molecular weight of RecA with up-titration of LPS extracted from H. pylori full-length LPS. As with RecA, a molecular weight shift was also observed for the proteins AlpA/B, SabA, OipA, and CagI. It was inferred that these shifts could also be explained by the protein interacting with different lengths of LPS.

Using a one-step transformation technique (difH/XerH recombination), a panel of 30 mutants, including previously unstudied GTs, was constructed for the elucidation of LPS structure and its biosynthetic pathway in H. pylori. Mutant phenotypes were assessed to investigate LPS size, the presence of Le^x and Le^y, the role of LPS in cell integrity, cell morphology and motility, and the production of outer membrane vesicles (OMVs). The composition of LPS in the mutants was also analysed using mass spectrometry and nuclear magnetic resonance spectroscopy, and the ability of mutants to colonize a mouse model of infection was also studied. The previously unstudied GTs HP1283, HPG27_1230, HP1284, HP0102, HP0805, HP1578 were found to be involved in the biosynthesis of the LPS core region, and the LPS from G27 was found to follow the same general linear structure as strains 26695, SS1, and a serogroup O:3 isolate. The LPS core was found to be essential for the resistance of H. pylori to polymyxin B, and the inner core heptose residues were found to be essential in OMV biogenesis. An additional role for FutA in assembly of the LPS core was suggested by the markedly increased sensitivity of G27ΔfutA to polymyxin B. The LPS size difference (~ 10 kDa) observed between the O-antigen ligase mutant G27ΔwaaL and the Lewis antigen mutant G27ΔHP0826, suggested that the LPS lipid A-core may be smaller than expected, and that the O-antigen might contain more than the Lewis antigen.

It is expected that the structural analysis of the LPS from G27 wild-type and the associated LPS mutants constructed in this study will enable determination of the LPS structure and its biosynthetic pathway in H. pylori, which will be crucial for dissecting the role of LPS in H. pylori pathogenesis and immune escape, and could lead to the development of therapeutic drugs targeting the H. pylori LPS biosynthetic pathway.