Abstract Background The ability of bacteria to chemically modify their cell surfaces through O -linked glycosylation is well characterised in Gram-negative bacteria and includes the modification of flagellin proteins with nonulosonic acids (NulOs). These modifications are widespread and have established roles in flagellar assembly, motility and virulence in various pathogens. However, there are limited documented examples of similar modifications in Gram-positive bacteria, and the significance of these is less well understood. In this study, we expand upon our previous findings that the flagellar biosynthetic locus of several pathogenic Clostridium butyricum strains contains a cluster of genes predicted to be involved in the biosynthesis of NulOs, and examine similar clusters across the whole genus Clostridium to assess the wider occurrence of NulO-mediated flagellin modification. Results Using orthologues of the components of the biosynthetic pathways for the NulOs pseudaminic acid (Pse) and legionaminic acid (Leg) in the pathogenic species Campylobacter jejuni , we predict that the flagellar biosynthetic loci of several pathogenic C. butyricum strains also encode genes for these modifications. To verify this, we demonstrate that the predicted PseB orthologue encoded in the flagellar biosynthesis locus of the pathogenic strain C. butyricum 5521 is able to catalyse the first step of Pse biosynthesis. Furthermore, we show that across the genus Clostridium , genes involved in flagellar glycosylation are located in a hypervariable region (HVR) of the flagellar biosynthetic locus, which is flanked by the flagellar structural genes flgB and fliD . The content of this region differs considerably across strains and species, but remarkably over half of the Clostridium genomes encode NulO biosynthesis genes in these regions. Finally, we present evidence that suggests that this flagellar HVR has evolved independently from the rest of the genome. Conclusions We show that genes required for NulO-mediated flagellar glycosylation are widespread across the genus Clostridium. Our findings have potential applications in the characterisation of pathogenic strains within these species and in the engineering of strains used in industrial biotechnology.
Anderson et al. (Wed,) studied this question.