Fig 1.
Schematic structure of the Y. pseudotuberculosis serotype O:1b LPS core and O-antigen.
The putative identities of the genes required for LPS synthesis are derived from reference [28]. The O-antigen is predicted to be present in bacteria grown in vitro at 21°C but not at 37°C. Those genes highlighted in bold represent genes identified in this study with transposon insertions that result in increased AMC binding.
Table 1.
Independent Tn5 insertion mutants in LPS-related genes with increased AMC binding phenotypes.
Fig 2.
Flow cytometric analysis of CCL28 binding to the surface of wild type and mutant bacteria.
In chemokine binding assays bacteria were incubated with CCL28 followed by biotinylated anti-chemokine antibody and APC-streptavidin conjugates. Negative controls included the omission of chemokine followed by the incubation of bacteria with anti-chemokine antibody followed by streptavidin-APC. Results indicate that wild-type Y. pseudotuberculosis is not readily bound by CCL28 (A). Conversely, transposon mutant insertions in the manC (B) or hldD (C) genes greatly increase the number of bacteria bound by chemokines.
Fig 3.
Role of HldD, WaaF, WaaC in AMC binding.
(A) A mutation in hldD results in significantly greater CCL25 and CCL28 binding (p<0.0001 by Two-way ANOVA), expressed as a percentage of cells that stain positive as measured by flow cytometry. In the hldD::Tn5 mutant background, addition of hldD/waaF slightly reduced CCL28 (**p<0.01) and CCL25 (*p<0.05) binding, whereas the presence of all three genes (hldD/waaF/waaC) completely reduced binding affinity for both CCL25 and CCL28 to wild-type levels (****p<0.0001) when grown at 21°C. (B) Enhanced CCL28 binding due to hldD mutation is also evident in bacteria expressing low levels of O-antigen from cultures grown at 37°C.
Fig 4.
Role of the hldD/waaF/waaC operon in survival against CCL28.
Relative survival of bacteria (expressed as a percentage of the number of live cells counted in the unexposed control of the same strain) in the presence of CCL28 over a 5 h exposure. Asterisks denote strains that were significantly different from the wild type strain at that time point (Two-way ANOVA (****p < 0.0001).
Fig 5.
Role of HldD, WaaF, and WaaC in polymyxin resistance.
Relative survival of bacteria (expressed as a percentage of the number of live cells counted in the unexposed control of the same strain) in the presence of different concentrations of polymyxin B. Asterisks denote that the result obtained was significantly different from the wild type strain at the given concentration by Two-way ANOVA (****p < 0.0001, ***p < 0.001, **p< 0.01, *p < 0.05).
Fig 6.
The hldD::Tn5 mutant shows no O- antigen expression and the presence of a truncated core oligosaccharide. Addition of hldD, hldD/waaF, or hldD/waaF/waaC plasmids cause production of full-length core oligosaccharide and O- antigen to increase. As expected, Y. pseudotuberculosis grown at 37°C does not show O- antigen expression. Sample IDs (from left to right) 1—wild type Y. pseudotuberculosis IP 32953 grown at 21°C, 2—hldD::Tn5 mutant, 3 -hldD+, 4 –hldD/waaF+, 5—hldD/waaF/waaC +, 6—wild type Y. pseudotuberculosis grown at 37°C. LPS was detected by fluorescent staining of carbohydrates.
Fig 7.
Oxidation of surface carbohydrates increases CCL28 binding.
(A) CCL28 binding (fluorescence intensity of the bacterial population after exposure to CCL28) to wild type Y. pseudotuberculosis IP32953 is enhanced by treatment with sodium periodate in a dose-dependent fashion. The binding levels are expressed as the mean (±SD) relative to the 80 mM samples. (B) Removal of surface proteins from the hldD::Tn5 mutant strain by proteinase K treatment decreases CCL28 binding (***p = 0.0001 by unpaired T-test) relative to the untreated samples.