Genus-wide Leptospira core genome multilocus sequence typing for strain taxonomy and global surveillance

Abstract

Leptospira is a highly heterogeneous bacterial genus that can be divided into three evolutionary lineages and >300 serovars. The causative agents of leptospirosis are responsible of an emerging zoonotic disease worldwide. To advance our understanding of the biodiversity of Leptospira strains at the global level, we evaluated the performance of whole-genome sequencing (WGS) as a genus-wide strain classification and genotyping tool. Herein we propose a set of 545 highly conserved loci as a core genome MLST (cgMLST) genotyping scheme applicable to the entire Leptospira genus, including non-pathogenic species. Evaluation of cgMLST genotyping was undertaken with 509 genomes, including 327 newly sequenced genomes, from diverse species, sources and geographical locations. Phylogenetic analysis showed that cgMLST defines species, clades, subclades, clonal groups and cgMLST sequence types (cgST), with high precision and robustness to missing data. Novel Leptospira species, including a novel subclade named S2 (saprophytes 2), were identified. We defined clonal groups (CG) optimally using a single-linkage clustering threshold of 40 allelic mismatches. While some CGs such as L. interrogans CG6 (serogroup Icterohaemorrhagiae) are globally distributed, others are geographically restricted. cgMLST was congruent with classical MLST schemes, but had greatly improved resolution and broader applicability. Single nucleotide polymorphisms within single cgST groups was limited to <30 SNPs, underlining a potential role for cgMLST in epidemiological surveillance. Finally, cgMLST allowed identification of serogroups and closely related serovars. In conclusion, the proposed cgMLST strategy allows high-resolution genotyping of Leptospira isolates across the phylogenetic breadth of the genus. The unified genomic taxonomy of Leptospira strains, available publicly at http://bigsdb.pasteur.fr/leptospira, will facilitate global harmonization of Leptospira genotyping, strain emergence follow-up and novel collaborative studies of the epidemiology and evolution of this emerging pathogen.

Fig 1. Phylogeny of representative strains of 42 Leptospira species.

The phylogenetic tree was obtained from the concatenation of 545 amino-acid sequence alignments using IQ-TREE [48] and the maximum likelihood criterion. In addition to previously described Leptospira species, new species (S1 Table) from subclades S2 (L. ilyithenensis and L. kobayashii), S1 (L. bouyouniensis, L. kemamanensis, L. jelokensis, L. bandrabouensis, L. mtsangambouensis, L. perdikensis, L. congkakensis), P2 (L. sarikeiensis, L. dzoumogneensis, L. selangorensis), and P1 (L. putramalaysiae and L. dzianensis) are included. Species are grouped into subgroups (numbers within blocks in the column on the right) within each subclade (colors of blocks; see key).

Fig 2. Geographic origins of the most frequent pathogenic Leptospira species.

Each pie chart corresponds to a given world region. From left to right: Pacific Ocean, North America, Central America, Caribbean, South America, Europe, Africa, Middle East, Indian Ocean, Asia, Southeast Asia, Oceania. The size of the pie charts is proportional to the number of isolates (see key). The ‘Others’ category includes all pathogenic species not individualized in the key. The figure has been generated using R and the package "rworldmap", which is free under the GPL-2 and GPL-3 licenses (https://cran.r-project.org/web/packages/rworldmap/index.html).

Fig 3. Comparison of cgMLST clustering with sequence types of the three classical MLST schemes currently in use.

The inner colors overlaying the strain names represent the species (see colour key). The 4 outer circles represent ST and cgMLST clonal groups (CG) classifications (S1 Table), with alternating shades as the identifiers change along the circle. Yellow circle, scheme MLST1 [45]; green circle: scheme MLST2 [46]; blue circle: scheme MLST3 [29]. The red circle corresponds to the clonal group (CG) cgMLST classification. Clonal groups are defined as groups of allelic profiles differing by no more than 40 allelic mismatches out of 545 loci from any other member of the group. The numbers in the four circles correspond to the main STs/CGs (with more than 3 representative genomes) of the corresponding sector. Red numbers correspond to STs for which all genomes are part of the subclade but whose subclade also contain at least one member of another ST (i.e., paraphyletic STs). Red numbers separated by commas correspond to multiple main STs mixed together.