Table 1.
Aeromonas veronii group isolates used in this study.
Figure 1.
Symbiotic competence of AVG isolates and ability to grow in blood.
(A). The ability of AVG isolates to colonize the leech. Spontantous antibiotic resistant isolates were coinoculated with the competitor strain (HM21R or HM21S) in a 1∶1 ratio. The antibiotic resistance of the strain is indicated by “R” for rifampin resistant and “S” for streptomycin resistant. The CI [(testoutput/competitorout)/(testinput/competitorinput)] was calculated. Each point represents the CI value from a single competition assay. A CI of 1 (dashed line) indicates that the test strain and competitor strain colonize to equal levels. A CI below 1 indicates that the test strain is outcompeted and has a colonization defect. The CI for each animal is shown. Horizontal lines represent median CI values. Strains with common letters are not statistically different from one another as determined by one-way ANOVA with Tukey's post-hoc test, P<0.05. (B). Blood was coinoculated with an AVG isolate and the competitor strain (HM21R or HM21S) in a 1∶1 ratio. The CI was calculated. A CI of 1 (dashed line) indicates that the test strain and competitor strain proliferate to equal levels. A CI below 1 indicates that the test strain is outcompeted and has a defect in its ability to grow in blood. Horizontal lines represent median CI values. Strains with common letters are not statistically different from one another as determined by one-way ANOVA with Tukey's post-hoc test, P<0.05.
Figure 2.
Virulence of AVG isolates in the Galleria mellonella infection model.
The LD50 was calculated for each isolate 24 h after G. mellonella larvae were inoculated using the Reed-Muench method [59]. Horizontal lines represent median LD50 values. Strains with common letters are not statistically different from one another as determined by one-way ANOVA with Tukey's post-hoc test, P<0.05.
Figure 3.
Gene maps of two differently ordered concatenations of the nine gene families with predicted breakpoints.
GARD inferred recombination breakpoints delimit regions with distinct phylogenies. Significance of breakpoints according to KH test: ***, Ρ<0.01; **, Ρ<0.05; * Ρ<0.1; no asterisk, not significant.
Table 2.
Tests for recombination applied to each gene family.
Figure 4.
Split networks generated using QNet with embedded quartets compatible with the plurality consensus signal.
These evolutionary relationships were inferred from maximum likelihood trees of 100 bootstrap replicates, weighted by frequency, of (A) house-keeping gene sequences (chiA, gyrB, dnaJ, and recA) and (B) T3SS gene sequences (aexT, aexU, ascFG, ascV). Alignments were divided at predicted recombination breakpoints determined by distinct evolutionary histories. Edge lengths are scaled to the number of embedded quartets in agreement with that split and split widths represent a departure from a tree-like signal. In the case of Aeromonas veronii group strains splits with width are likely to be indicative of recombination across that split.
Figure 5.
Bayesian trees inferred from two concatenations of gene family alignments.
A. housekeeping gene sequences (chiA, gyrB, dnaJ, and recA) and B. type III secretion system gene sequences (aexT, aexU, ascFG, ascV). Numbers represent the posterior probability for that split, edge lengths are scaled to average substitutions per site in the posterior distribution.