Genome sequence of the saprophyte Leptospira biflexa provides insights into the evolution of Leptospira and the pathogenesis of leptospirosis

Abstract

Leptospira biflexa is a free-living saprophytic spirochete present in aquatic environments. We determined the genome sequence of L. biflexa, making it the first saprophytic Leptospira to be sequenced. The L. biflexa genome has 3,590 protein-coding genes distributed across three circular replicons: the major 3,604 chromosome, a smaller 278-kb replicon that also carries essential genes, and a third 74-kb replicon. Comparative sequence analysis provides evidence that L. biflexa is an excellent model for the study of Leptospira evolution; we conclude that 2052 genes (61%) represent a progenitor genome that existed before divergence of pathogenic and saprophytic Leptospira species. Comparisons of the L. biflexa genome with two pathogenic Leptospira species reveal several major findings. Nearly one-third of the L. biflexa genes are absent in pathogenic Leptospira. We suggest that once incorporated into the L. biflexa genome, laterally transferred DNA undergoes minimal rearrangement due to physical restrictions imposed by high gene density and limited presence of transposable elements. In contrast, the genomes of pathogenic Leptospira species undergo frequent rearrangements, often involving recombination between insertion sequences. Identification of genes common to the two pathogenic species, L. borgpetersenii and L. interrogans, but absent in L. biflexa, is consistent with a role for these genes in pathogenesis. Differences in environmental sensing capacities of L. biflexa, L. borgpetersenii, and L. interrogans suggest a model which postulates that loss of signal transduction functions in L. borgpetersenii has impaired its survival outside a mammalian host, whereas L. interrogans has retained environmental sensory functions that facilitate disease transmission through water.

Figure 1. Circular maps of the three L. biflexa replicons.

(1) the coordinates in bp beginning at 0 = oriC; (2) dark pink: genes unique to L. biflexa, not found in L. interrogans serovar Copenhageni and L. borgpetersenii serovar Hardjobovis (identity >40% over 80% of the length of the smallest protein). (3) dark purple: genes found in L. biflexa, L. interrogans and L. borgpetersenii (identity >40% over 80% of the length of the smallest protein). (4) red: genes found in L. biflexa and L. borgpetersenii, but not in L. interrogans (identity >40% over 80% of the length of the smallest protein). (5) brown : genes found in L. biflexa and L. interrogans, but not in L. borgpetersenii (identity >40% over 80% of the length of the smallest protein). (6) blue: genes found in L. biflexa and other sequenced spirochetes (Borrelia afzelii PKo, Borrelia burgdorferi, Borrelia garinii, Treponema denticola and Treponema pallidum) (identity >40% over 80% of the length of the smallest protein ). (7) The innermost ring shows GC skew; positive skew is shown in grey, and negative skew is shown in black.

Figure 2. Venn diagram showing numbers of unique and shared genes amongst L. interrogans, L. borgpetersenii and L. biflexa.

Orthologous CDS were identified in a pair-wise fashion using Whole-Genome Reciprocal Best-Hit BLAST Analysis . Manual curation ensured a one to one relationship for orthologous CDS, particularly in situations where sets of paralogous CDS existed and in addition evaluated the nature of the relationship between CDS with reciprocal best-hits but low expect values. This analysis was performed using the L. interrogans serovar Copenhageni strain Fiocruz, L. borgpetersenii serovar Hardjo strain L550 and L. biflexa serovar Patoc strain Ames genome sequences.

Figure 3. Synteny plot between the five Leptospira genomes.

The line plots were obtained using synteny results between the large CI(A) or small CII(B) chromosomes of L. biflexa serovar Patoc strain Patoc1, L. interrogans serovar Lai strain 56601, L. interrogans serovar Copenhageni strain Fiocruz L1-130, L. borgpetersenii serovar Hardjo strain L550, and L. borgpetersenii serovar Hardjo strain JB197. A line plot (C) compares synteny between L. borgpetersenii serovar Hardjo strain JB197 and L. interrogans serovar Copenhageni strain Fiocruz L1-130. Comparative analysis was performed using the MaGe interface in the SpiroScope database (https://www.genoscope.cns.fr/agc/mage). The minimum size of the synteny groups is set to five genes. In green: synteny groups are organized on the same strand; in red: synteny groups are organized on opposite strands.