Fig 1.
Comparative abundance and origins of complete or partial P. multocida sequences in public databases.
(A) Country and host distribution of whole genome sequences deposited in GenBank, SRA, or characterized in this study (S1 and S2 Tables). (B) Country and host distribution of MLST clonal complexes from 1503 isolates genotyped by the RIRDC scheme, last updated 3rd March 2024.
Fig 2.
Neighbor-joining tree output from the core genome distances of 741 P. multocida produced by PopPUNK using a Bayesian Gaussian Mixture Model fitting.
The inner color strip indicates the core clusters assigned within the sample by the program. The sources of the Australian datasets are represented by the symbols on the outer rims, respectively indicating the current study, SRA, RefSeq/GenBank, and DOIs of recent studies [7,13].
Fig 3.
Maximum likelihood phylogenetic tree from aligned SNP output from parsnp.
The tree was constructed with FastTreeDbl with 100 bootstraps and the GTR substitution model. The tree is mid-rooted, and represents 330 genomes and 77950 polymorphic positions. The color strips, starting from the outer strip, indicate the host species, LPS type and capsule type, respectively. The reference genome for Pasteurella multocida (PM70) is indicated in red. Branch thickness represents bootstrap values.
Table 1.
Australian isolates with unclassified STs from the Multi-host and RIRDC MLST schemes.
Fig 4.
Deep-branched clade extracted from the complete maximum likelihood phylogenetic tree represented in Fig 3, representing 85 genomes.
The color strips, starting from the left strip, indicate the host category, capsule type and LPS type, respectively. Branch thickness represents bootstrap values.
Fig 5.
Minimal spanning tree from PHYLOViZ goeBURST analysis of whole genomes of P. multocida.
Highlighted sections of the analysis: Inset A, branch containing ST20 isolates; Inset B, clade containing most isolates from companion animals and wildlife in the study. The distance labels in Insets A and B indicate the number of different polymorphic positions between two isolates or sets of isolates. The locations of the isolates within goeBURST A-E containing mixed host species are also indicated. A total of 76 groups of isolates with a closest genome displaying 0 to 4 loci differences, representing 525 genomes (S6 table), were considered to represent clonal complexes. Of those, five groups (goeBURST A-E) contained isolates from different host species, suggesting cross-species transmission. Within each group, the genomes had a unique ST and originated from close or identical geographical areas (Table 2).
Table 2.
The genomes of the 5 clonal complexes containing mixed host species visualized by PHYLOViZ goeBURST analysis.
Table 3.
Number of SNP differences in paired P. multocida assembled genomes from MLST groups with putative cross-species transmissions.
Fig 6.
Distribution of mobile genetic elements in 585 P. multocida genomes.
The maximum likelihood phylogenetic tree was constructed from aligned SNPs of all genomes, after correction for recombination, using FastTreeDbl with 100 bootstraps and the GTR substitution model. The tree is mid-rooted; it represents 585 genomes and 58913 polymorphic positions. Phages were predicted with Phigaro and represented by triangles. Plasmids were predicted with PLASMe and represented by circles (Transformer evidence) or squares (BlastN evidence). Conjugative transposons were predicted by BlastN searches against the IceBerg 2.0 database and represented by stars.
Table 4.
Distribution of a cryptic 1805 bp plasmid U51470 amongst P. multocida isolates.
Fig 7.
Distribution of P. multocida subspecies within the population.
A set of 498 genomes from GenBank, supplemented by 84 Australian isolates sequenced in our laboratory (25 recently published ST20 and 59 from this study), was used. The colors yellow, green, and purple indicate the multocida, septica, or gallicida subspecies, respectively. (A) Unrooted maximum likelihood phylogenetic tree built from 585 core SNPs alignments of genomes, as shown in Fig 5. The deep-branched clade is indicated in red. Subspecies mentioned in the GenBank file header are reported by squares. Positive in-silico subspecies PCR results are indicated by triangles. (B) Hierarchical clustering of a matrix of pairwise ANI values calculated from the same set of 585 genomes. The left sidebar indicates the positions of the subspecies mentioned by the GenBank file headers.
Table 5.
Results of AST by microdilution for P. multocida isolates in this study.
Fig 8.
Predicted ARGs in P. multocida genomes from compiled ABRIcate results.
Resistances to aminoglycosides, beta-lactams, phenicols, macrolides, tetracyclines and trimethoprim/sulfamides are respectively represented by purple, green, brown, red, yellow and blue color dots, with sizes proportional to the number of hits within the dataset.
Table 6.
Predicted ARGs in Australian bovine P. multocida ST394 genomes.
Fig 9.
Alignment of the 7.6 kb circular contigs found in two bovine isolates from Victoria and the plasmid pYH12207−3 of Acinetobacter piscicola strain YH12207_T. Genes belonging to the identical similarity group are represented by the same color.