Fig 1.
Simplified theories of QseBC gene regulation in EHEC (A) and UPEC (B).
A. In EHEC, QseC phosphorylates three different RRs KdpE, QseB and QseF that in turn upregulate LEE, flhC and recA genes involved in A/E lesion formation, motility and SOS response, respectively [23]. B. In UPEC, it is theorized that QseB RR acts as a repressor of motility genes and other genes. QseB RR is phosphorylated by the non-cognate HK PmrB and de-phosphorylated by QseC that controls QseB repressor activity [25]. Created with Biorender.com.
Fig 2.
The strategy used for constructing gene deletions showing the histidine kinase (QseC) and response regulator (QseB) deletion as an example.
In histidine kinases an in-frame deletion was inserted to eliminate more than 80% of the cytoplasmic domain that contains the catalytic domains. Sensory domain and transmembrane region were left functional. More than 80% of the response regulators were deleted, including parts of both the receiver domain and DNA-binding domain. In-frame deletions were used to avoid polar effects on expression of the rest of the operon. Created with Biorender.com.
Fig 3.
Growth curves of EPEC O125ac:H6 deletion mutants compared to the WT.
Growth curves represent OD600 measurements recorded every 30 minutes for 15 hours for EPEC O125ac:H6 WT, ΔqseB, ΔqseC, ΔqseBΔqseC, ΔpmrA, ΔpmrB, ΔpmrAΔpmrB, ΔpmrAΔqseB, ΔpmrAΔqseC, ΔpmrBΔqseB, ΔpmrBΔqseC (A) and ΔkdpD, ΔkdpE ΔkdpDΔqseB, ΔkdpDΔqseC, ΔkdpEΔqseB and ΔkdpEΔqseC (B). Error bars represent the Standard Deviation, n = 4.
Fig 4.
Relative motility of QseBC and PmrAB single and double mutants.
A. Relative motility compared to the WT of EPEC O125ac:H6 ΔqseB, ΔqseC, ΔqseBΔqseC, ΔpmrA, ΔpmrB, ΔpmrAΔpmrB, ΔpmrAΔqseB, ΔpmrAΔqseC, ΔpmrBΔqseB, ΔpmrBΔqseC. B. Relative motility compared to the WT of EPEC O125ac:H6 ΔqseB complemented with WT qseB, ΔpmrBΔqseC complemented with WT pmrb ΔqseC complemented with qseC, WT QseBD51A, ΔqseC QseBD51A and ΔqseC PmrBH152A. WT motility is calculated as the relative motility of the WT per plate compared to the average motility of the WT in the experiment. Motility plates that exemplify the seen motility are depicted. * p-value <0.05; ** <0.01; *** <0.001; **** <0.0001 via one-way ANOVA with Tukey’s multiple comparison (A) and via Krukal-Wallis test with Dunn’s multiple comparison (B), n = 8.
Fig 5.
Relative motility of QseBC and KdpDE single and double mutants.
Relative motility compared to the WT of eight replicates is depicted for EPEC O125ac:H6 ΔqseB, ΔqseC, ΔkdpD, ΔkdpE ΔkdpDΔqseB, ΔkdpDΔqseC, ΔkdpEΔqseB and ΔkdpEΔqseC. EPEC O125ac:H6 WT motility is calculated as the relative motility of the WT per plate compared to the average motility of the WT in the experiment. Motility plates that exemplify the seen motility are depicted. * p-value <0.05; ** <0.01; *** <0.001; **** <0.0001 via one-way ANOVA with Tukey’s multiple comparison, n = 8.
Fig 6.
Effect of adrenergic signals (epinephrine), AI-3 and iron in the motility of EPEC O125ac:H6.
A. Chemical structure of AI-3 [36] B. Relative motility compared to the non-treated control of EPEC O125ac:H6 in presence of 500 μM of epinephrine, AI-3 and Fe3+. C. Comparison between the relative motility of EPEC O125ac:H6 mutants compared to the WT in non-treated agar plates (green) with the relative motility of EPEC O125ac:H6 mutants compared to the WT in agar plates containing 500 μM Fe3+ (red). Motility plates of mutants that are affected by the presence of iron are depicted. * p-value <0.05; ** <0.01; *** <0.001; **** <0.0001 via one-way ANOVA with Tukey’s multiple comparisons, n = 5.
Fig 7.
Differential expression of flhC, fliA, ler, recA and qseB in different EPEC O125ac:H6 QseBC, PmrAB and KdpDE mutants.
A. Differential expression according to the luciferase assay of the genes flhC, fliA, ler, recA and bla of all tested mutants. Significant differences are indicated by ++ > 2-fold; + 1-2-fold;. non-significant differences with WT; - 1–0.5-fold and—- < 0.5-fold B. Quantitative PCR of genes ler and espA in low glucose DMEM in EPEC O125ac:H6 H6 ΔqseB and ΔqseC C. qPCR of genes flhC and fliA in EPEC O125ac:H6 ΔqseB, ΔqseC, and ΔpmrAΔqseC to prove that luciferase assay was accurate. D. qPCR in EPEC O125ac:H6 ΔqseC and ΔpmrAΔqseC to check qseB expression in these mutants. * p-value <0.05; ** <0.01; *** <0.001; **** <0.0001 via one-way ANOVA with Tukey’s multiple comparisons, n = 4.
Fig 8.
Effect of mutations in QseBC, PmrAB and KdpDE genes in colistin and polymyxin B resistance.
A. MIC data of colistin and polymyxin B (PBM) for EPEC O125ac:H6 WT, ΔqseB, ΔqseC, ΔqseBΔqseC, ΔpmrA, ΔpmrB, ΔpmrAΔpmrB, ΔpmrAΔqseB, ΔpmrAΔqseC, ΔpmrBΔqseB, ΔpmrBΔqseC., ΔkdpD, ΔkdpE ΔkdpDΔqseB, ΔkdpDΔqseC, ΔkdpEΔqseB and ΔkdpEΔqseC. B. Differential expression using qPCR of colistin and PBM resistance genes arnB and eptA in the mutants EPEC O125ac:H6 WT, ΔqseB, ΔqseC and ΔpmrAΔqseC* p-value <0.05; ** <0.01; *** <0.001; **** <0.0001 via one-way ANOVA with Tukey multiple comparison test, n = 4. C. In the promoter region of the arn operon, binding motifs are found for both QseB (blue) and PmrA (orange). The -35 and -10 sites were predicted using BPROM [42], and a rpoD17 binding site was predicted within the PmrA binding site. The start (ATG, red) of arnB gene is also shown. D. Electrophoretic mobility shift assay (EMSA). To validate the technique a EBNA control provided by the kit is shown. Biotylinated DNA fragment containing the QseB binding site in the arn operon is found to shif with the presence of 3 μg of QseB-P, which is reversed by the presence of unlabeled QseB site. Finally unphosphorylated QseB is found to bind with lower affinity to the QseB binding site.
Fig 9.
Proposed regulation of the TCS KdpDE, PmrAB and QseBC in EPEC.
A. In WT aEPEC, KdpD HK phosphorylates an unknown effector to activate flagella late-regulatory genes. PmrB activates its cognate RR PmrA, and the non-cognate QseB. QseC HK controls the level of activated QseB via de-phosphorylation and sequestration of activated QseB. Unphosphorylated QseB in EPEC WT is able to exert a basal repression of flhC, recA and ler genes. B. In ΔpmrAΔqseC mutant, QseB is phosphorylated via PmrB, and cannot be deactivated by QseC. P~QseB activates the expression of qseBC operon leading to QseB over-expression. In arn operon, PmrA usually blocks transcription, but in its absence phosphorylated QseB leads to activation of colistin resistance genes. Excess of phosphorylated QseB tightly represses flhC and recA genes. Created with Biorender.com.
Fig 10.
Synthetic scheme of autoinucer AI-3.