Genome reduction in Leptospira borgpetersenii reflects limited transmission potential

Abstract

Leptospirosis is one of the most common zoonotic diseases in the world, resulting in high morbidity and mortality in humans and affecting global livestock production. Most infections are caused by either Leptospira borgpetersenii or Leptospira interrogans, bacteria that vary in their distribution in nature and rely on different modes of transmission. We report the complete genomic sequences of two strains of L. borgpetersenii serovar Hardjo that have distinct phenotypes and virulence. These two strains have nearly identical genetic content, with subtle frameshift and point mutations being a common form of genetic variation. Starkly limited regions of synteny are shared between the large chromosomes of L. borgpetersenii and L. interrogans, probably the result of frequent recombination events between insertion sequences. The L. borgpetersenii genome is approximately 700 kb smaller and has a lower coding density than L. interrogans, indicating it is decaying through a process of insertion sequence-mediated genome reduction. Loss of gene function is not random but is centered on impairment of environmental sensing and metabolite transport and utilization. These features distinguish L. borgpetersenii from L. interrogans, a species with minimal genetic decay and that survives extended passage in aquatic environments encountering a mammalian host. We conclude that L. borgpetersenii is evolving toward dependence on a strict host-to-host transmission cycle.

Fig. 1.

Chromosome maps of L. borgpetersenii serovar Hardjo strain L550. Functional information for the large (CI) and small (CII) chromosomes is presented in concentric rings (maps are not drawn to scale; replicon sizes are shown in bp). Starting from the outside: genes encoded on the top strand (first ring), genes encoded on the bottom strand (second ring), transposases (third ring), and pseudogenes (fourth ring). Genes are colored according to the corresponding functional categories shown at the bottom. The fifth ring shows GC content deviations from the genomic average calculated by using a window of 10 kb in steps of 1 kb; values greater than the average fall on the outside of the ring, and values less than the average fall on the inside of the ring. The innermost ring shows GC skew calculated by using a window of 10 kb in steps of 1 kb; positive skew is shown in green, and negative skew is shown in purple.

Fig. 2.

Comparison of Leptospira CI replicons. The CI replicons of L. borgpetersenii serovar Hardjo strains L550 and JB197 and L. interrogans serovars Lai and Copenhageni were aligned by MegaBLAST and visualized with ACT (60). Colored lines drawn between two adjacent linearized chromosomes (horizontal black lines) show the location of homologous genes and indicate the same (red) or opposite (black) orientation relative to the chromosome immediately above. Note that the L550 CI replicon aligns with the CI replicons from either L. interrogans serovar with gene scattering similar to that shown for strain JB197.

Fig. 3.

Quantitative analysis of functional groups. Genes were clustered from each Leptospira genome according to the functional groups shown. The percentage of pseudogenes (Left) or paralogs (Right) that comprise the total number of genes within each category is shown. For reference, a bar graph showing the total number of genes in each category is presented in Fig. 5.