The wild, wild west of plasmids: First insights into comparative genomics of Borrelia burgdorferi sensu lato.

Lyme borreliosis is caused by species of the Borrelia burgdorferi sensu lato complex. Genospecies vary in human pathogenicity, reservoir host and tick vector adaptation. What causes these differences is currently unknown, but some answers are likely to be found within the fragmented genomes containing linear and circular plasmids (and even fusions). These plasmids are highly variable and challenging to reconstruct completely. We aimed to analyze plasmid presence/absence and to identify trends in plasmid numbers and/or types associated with specific ecological adaptations or pathogenic potentials. For this, we used the most complete, reliable, and currently available 86 B. burgdorferi s.l. genomes (30 newly sequenced, 56 publicly available) belonging to 21 of the 28 known species. The plasmid types per isolate vary from 6-21 (median: 12) and linear plasmid numbers (3-12, median: 7) are significantly higher than circular plasmids (2-11, median: 5). The only plasmids found in all isolates are lp54 and cp26 and only a few species- or population-specific plasmids are identified, demonstrating the high variability. Species with smaller plasmid counts (median < 11) tend to have fewer lp28 and cp32 plasmids than larger plasmid counts species (median ≥ 15). Significantly higher total and circular plasmid numbers are found in rodents-only compared to birds-only associated species. Interestingly, plasmid numbers are not significantly reduced in species associated with a single reservoir host/vector compared to multiple. Analyses of the plasmid-location of the vls/vlsE antigenic variation system highlighted the dynamic nature of B. burgdorferi s.l. plasmids. Surprisingly, whilst the plasmid repertoire of B. burgdorferi s.l. is very plastic and variable, our analyses rarely reveal plasmids associated with specific ecological adaptations or pathogenic potentials. However, some plasmids were more frequently found in one of the groups and will be of interest for gene-based analyses that will be required for further progress in this research area.