Evolutionary dynamics of multidrug resistant Salmonella enterica serovar 4,[5],12:i:- in Australia

Abstract

Salmonella enterica serovar 4,[5],12:i:- (Salmonella 4,[5],12:i:-) is a monophasic variant of Salmonella Typhimurium that has emerged as a global cause of multidrug resistant salmonellosis. We used Bayesian phylodynamics, genomic epidemiology, and phenotypic characterization to describe the emergence and evolution of Salmonella 4,[5],12:i:- in Australia. We show that the interruption of the genetic region surrounding the phase II flagellin, FljB, causing a monophasic phenotype, represents a stepwise evolutionary event through the accumulation of mobile resistance elements with minimal impairment to bacterial fitness. We identify three lineages with different population dynamics and discrete antimicrobial resistance profiles emerged, likely reflecting differential antimicrobial selection pressures. Two lineages are associated with travel to South-East Asia and the third lineage is endemic to Australia. Moreover antimicrobial-resistant Salmonella 4,[5],12:i- lineages efficiently infected and survived in host phagocytes and epithelial cells without eliciting significant cellular cytotoxicity, suggesting a suppression of host immune response that may facilitate the persistence of Salmonella 4,[5],12:i:-.

Fig. 1. Inferred population dynamics of ST34 Salmonella 4,[5],12:i:-.

A Bayesian evolutionary analysis showing the maximum-clade credibility (MCC) tree of ST34 isolates (n = 309) inferred from 2693 SNPs, demonstrating the timing of divergence and geographical spread of Salmonella 4,[5],12:i:- globally. The tips of the phylogeny are colored by the major geographical regions for the reported country of collection (for publicly available data) or by destination of reported travel (if any) for Australian isolates. The three lineages are highlighted on the phylogeny. The posterior support of ≥0.95 for internal nodes is shown with black circles on the node. The time in years is given on the x axis. Phylogeny available in Newick form from microreact (https://microreact.org/project/mfxxBchBsUpsJu7nvfkFw4). B Heatmap that shows the study from which the isolates were sourced and the presence of the ASSuT genotype (defined by any genes mediating antimicrobial resistance to ampicillin, streptomycin, sulfonamides, and tetracycline). C The estimated effective population size through time for each of the three lineages. The shaded area indicates the 95% confidence interval. D Visualization of the most recent common ancestor (MRCA) of the three lineages with highest posterior density (HPD) intervals shown as the bars. E Circular migration diagram of the migration events between the five geographical regions. The size of the colored block denotes the posterior mean number of inferred migration events from the Bayesian phylogeographical analysis and arrows denote directionality. The inset box is the legend for the different attributes.

Fig. 2. The interruption of the phase II flagella differs between lineages.

A The frequencies that each gene in the phase II flagella region appears in each lineage is shown as a heatmap, with dark blue indicating the gene was detected in all isolates and white indicating that the gene was not detected in any. For comparison, genes in the biphasic ST19 S. Typhimurium SL1344 type are shown above. B–D The different interruption patterns of the phase II flagella in Australian Salmonella 4,[5],12:i:- genomes resolved through long-read sequencing are shown in comparison to the biphasic ST19 S. Typhimurium SL1344 type strain. The blue blocks indicate >95% sequence identity. B Two isolates from humans in Lineage 1. C Two isolates from humans and one bovine isolate from Lineage 2, in addition to the publicly available TW-STm6 isolate collected from a pig in 2014 and D Two isolates from humans in Lineage 3.

Fig. 3. Third-generation cephalosporin and colistin resistance profiles differ by lineages and are associated with different plasmids.

A AMR genes mediating resistance to third-generation cephalosporins (3GCs) and colistin was detected in the 309 isolates stratified by lineage membership. The * indicates a close but not exact sequence match. Rarefaction curves of the diversity in the unique AMR gene combinations for each lineage are shown to the right. B Regions encoding AMR in plasmids resolved from long-read data for common AMR profiles that mediate resistance to third-generation cephalosporins and /or colistin in the three lineages. C The plasmid STs in the genomes are shown by lineage membership for the top three plasmid replicons detected from the initial plasmid screen. The * indicates a close but not exact sequence match. Rarefaction curves of the diversity in the unique plasmid replicon combinations for each lineage are shown to the right. AMR antimicrobial resistance.

Fig. 4. Lineage 2 and 3 ST34 Salmonella 4,[5],12:i:- demonstrate increased infectivity and replication in human cell lines.

Differentiated THP-1 (human monocyte) cells (A–D), HT-29 (human intestinal epithelial) cells (E, F), or BJ-5ta (human fibroblast) cells (G, H) were infected at an MOI:10 with selected ST34 Salmonella 4,[5],12:i:- and ST34 matched-Lineage 1 biphasic S. Typhimurium isolates (Supplementary Table 1). THP-1 cells were lysed and intracellular bacteria enumerated as CFU/well at times 0, 6, 12, and 24 hpi. HT-29 and BJ-5ta cells were lysed and intracellular bacteria enumerated as CFU/well at times 0 and 24 hpi. Isolates depicted on the x axis correspond to isolates listed in Supplementary Table 1. Each dot represents CFU/well of a biological replicate (performed in technical duplicate), with error bars indicating ±1 standard deviation of n = 3 or 4 biological replicates. Statistical significance was determined by nested one-way ANOVA with Fisher’s uncorrected multiple comparisons test. MOI multiplicity of infection, hpi hours post-infection, CFU colony-forming units.