Flagellin glycosylation is a common phenomenon among a wide range of bacterial and archaeal lineages. Furthermore, a wide range of functions have been ascribed to this trait, including the assembly, stabilisation and functioning of the flagellum, attachment, host defence avoidance and virulence. In this study, genetic screening and comparative genomic analyses was used to show that flagellin glycosylation islands are universally present among all studied P. ananatis strains, regardless of their geographic origin or source of isolation. The genomic analyses undertaken also highlighted the high degree of variability in the genetic architecture of this locus. A typing scheme was developed on the basis of the presence/absence of genes in the islands of the compared strains, and the seventeen genome sequenced strains could be discrimanted into four distinct FGI types. These FGI types differed in terms of the glycosyltransferases encoded on the islands, the sugar biosynthetic genes as well as the gene complement involved in additional methylation, acetylation and formylation of the glycans, which likely translates in highly versatile flagellin glycan structures, containing potentially different sugars with different side chains.
By means of PCR assays designed on the basis of differential gene fragments for each of the FGI types identified from the genomic data, a large number of P. ananatis strains isolated from a wide range of environmental sources and geographical locations were screened. The analyses suggested that the diversity of the flagellin glycosylation islands, and thus likely the flagellin glycans themselves, could possibly be capable of numerous functions that stretch beyond the bounds of what was observed here. Furthermore, there was no distinct correlation between the FGI type and the host/geographic locations from which the P. ananatis strains were isolated. This may suggest that unlike in some other phytopathogens, flagellin glycosylation does not play a role in host specificity. A role for flagellin glycosylation in virulence is possible, although FGIs were also observed in several isolates from non-plant sources. It should be tested experimentally, however, if these strains are capable of causing disease symptoms when inoculated into a P. ananatis host plant. The universal presence of FGIs in P. ananatis may also suggest a potential essential role of flagellin glycosylation in the assembly, maintenance or functioning of the flagellum of this species. The global distribution of the different FGI types are very interesting, as this would suggest that the horizontal gene exchange events from FGIs are purported to have been derived, may have occurred multiple times on several different continents. However, the increased globalization of our planet may also have driven the global distribution of P. ananatis strains with the different FGI types. It is thus evident that there are a number of unanswered questions about flagellin glycosylation in P. ananatis. At least one of these, is whether the FGI loci are functional and the flagellin glycans are expressed. Future investigations into this fascinating genetic locus must therefore be undertaken.