Combining short- and long-read sequencing unveils geographically structured diversity in Borrelia miyamotoi

Abstract

Borrelia miyamotoi is an emerging Ixodes tick-borne human pathogen in the Northern hemisphere. The aim of the current study was to compare whole genome sequences of B. miyamotoi isolates from different continents. Using a combination of Illumina and PacBio platforms and a novel genome assembly and plasmid typing pipeline, we reveal that the 21 sequenced B. miyamotoi isolates and publically available B. miyamotoi genomes from North America, Asia, and Europe form genetically distinct populations and cluster according to their geographical origin, where distinct Ixodes species are endemic. We identified 20 linear and 17 circular plasmid types and the presence of specific plasmids for isolates originating from different continents. Linear plasmids lp12, lp23, lp41, and lp72 were core plasmids found in all isolates, with lp41 consistently containing the vmp expression site. Our data provide insights into the genetic basis of vector competence, virulence, and pathogenesis of B. miyamotoi.

Keywords: Bacteriology; Geography; Microbial genomics; Phylogenetics; Techniques in genetics.

Graphical abstract

Figure 1

Phylogenetic tree of Borrelia miyamotoi species by core gene alignment A phylogenetic tree of Borrelia miyamotoi strains based on the concatenated alignment of nucleotide sequences from 667 core genes located on the chromosome using Roary. A maximum likelihood tree was constructed by RAxML software using a nucleotide substitution model with a gamma distribution of variable positions (GTR + Γ). The resulting tree was rooted at the branch connecting the European clade to the rest of the tree. Scale bar indicates substitution rates. The figure includes the name of the strain, its origin concerning Borrelia species, geographical location, isolation source and year of collection. ∗ Indicates the 21 Borrelia miyamotoi strains analyzed in this paper.

Figure 2

Plasmid diversity of three reference Borrelia miyamotoi genomes A schematic representation of the plasmid diversity of the three reference Borrelia miyamotoi genomes. Identified are the genes and pseudogenes of PF32, 49, 50, 57/62, vlp and vsp. They are displayed in the order, amount and relative position on the linear and circular plasmids.

Figure 3

Pan-genome analysis of Borrelia miyamotoi performed by PEPPAN pipeline (A) Whole-genome MLST tree based on PEPPAN output data. Scale indicates the number of genes different by alleles. (B) Core-pan plots of studied Borrelia miyamotoi genomes. On the pan-genomic curve (black line with gray confident interval) Heaps' law model was calculated (γ = 0.096 ± 0.010, κ = 1,212,022 ± 29,199), resulting in approximately 7.292 new genes per new genome. On the core genome curve (green line with green confident interval) the power law model was applied (α = 63,115-∞, κ = 1,142,212-∞), resulting in no new core genes by adding new genomes. (C) Upset plot depicting the intersections of a sets of common or different gene clusters (≥10) among the 21 B miyamotoi genomes. (D) Classification of the clusters of the genes based on frequency among 21 B. miyamotoi genomes. (E) Histogram representing the gene cluster frequency spectrum G(k), of the cumulative genomes.

Figure 4

Nucleotide sequences comparison of virulence plasmid lp41 of Borrelia miyamotoi Comparison of the nucleotide sequences of the virulence plasmid cp26 of the B. burgdorferi s.s. strain Bol26 (NC_012512) and the virulence plasmid lp41 of the 21 B miyamotoi isolates originating from America (LB-2001, 410, 483, 1938, 1995, 1999, and CT14D4), Asia (HT31, Izh-4, Izh-5, Izh-14, Izh-16, Yekat-1, Yekat-6, Yekat-17, Yekat-18, Yekat-19, Yekat-21, and Yekat-76), and Europe (ZStruIII14-9 and NL-IR-1). Plain gray blocks indicate similar areas (range of identity 70–100%) between plasmids. The arrows indicate genes and direction of open reading frames. The vmp block of genes are represented by the genes after the expression site, i.e., the expressed vlp gene in brown, the expressed vsp gene in purple, and the non-expressed vmp genes in orange. Other open reading frames are shown in arrows in light blue, guaA and guaB genes in green, and resT genes in dark blue.

Figure 5

Alignment of the lp41 intergenic region upstream of the vmp expression site Alignment of the intergenic region (73 bp) located upstream of the expressed vmp gene on the virulence plasmid lp41 expression site. The areas of a ribosome binding site (RBS), and the "-35" and "-10" sites are indicated in the gray boxes.

Figure 6

Phylogenetic diversity of vlp and vsp genes in B. miyamotoi genomes Phylogenetic diversity of 623 vlp genes (A and B), and of 211 vsp genes (C, D) from 21 B miyamotoi genomes. The maximum likelihood tree was constructed using IQ-TREE v2.0.3. The trees were constructed based on amino acid sequences which contain domains corresponding to the lipoprotein 2 family (IPR000680) and OspC (IPR001800) for the vlp and vsp genes, respectively. According to BIC the WAG+F+G4 model was chosen for vlp genes, and the JTTDCMut+F+I + G4 model for vsp genes. The different subfamilies clades are indicated in panels (A and B) with vlp-α in yellow, vlp-γ in red, vlp-δ in blue, and vlp-β in purple; and in panels (C and D) with vsp-1 in yellow, vsp-2 in red, and vsp-3 in blue. (A and C) Unrooted maximum likelihood tree from the boot.splits.nex output visualized by the SplitsTree software v4. The tree shows support values for all bifurcations, computed as the occurrence frequencies in percentage in the bootstrap trees. The inlet of the tree root is enlarged to allow visualization of the branches, with 100% support values for the vlp genes (black arrows), and for the vsp genes 79–87% support values. (B and D) Midpoint rooted maximum likelihood consensus tree from the boot.contree output. Nodes marked as ‘exp’ are genes located at the expression site on the lp41 plasmid. Colored bars indicate the vmp genes divided in the geographical groups of B. miyamotoi strains. (B) The 149 American vlp genes in red, the 395 Asian vlp genes in green, and the 100 European vlp genes in orange. (D) The 36 American vsp genes in red, the 126 Asian vsp genes in green, and the 49 European vsp genes in orange.