The RND efflux system ParXY affects siderophore secretion in Pseudomonas putida KT2440

ABSTRACT Tripartite efflux systems transport antimicrobial agents, toxic metabolites, and siderophores from Gram-negative cells into the environment. For example, the main siderophore pyoverdine of the soil bacterium Pseudomonas putida KT2440 is secreted into the environment via the PvdRT-OpmQ and MdtABC-OpmB systems. Here, we looked for efflux systems that might be involved in the secretion of pyoverdine in addition to the latter two systems. Screening of different efflux systems revealed that parX (encoding the periplasmic adapter protein of the ParXY system) is of particular importance for bacterial growth under iron limitation. Further analysis showed that the deletion of parX impairs the production and secretion of pyoverdine, causing the observed growth effect. The effects were dependent on the presence of other tripartite efflux systems and the conditions of iron limitation. The results suggest that ParXY not only plays a role in antibiotic resistance, as shown previously, but also influences the secretion of siderophores in a network of overlapping activities of different tripartite efflux systems. IMPORTANCE Gram-negative bacteria from the Pseudomonas group are survivors in various environmental niches. For example, the bacteria secrete siderophores to capture ferric ions under deficiency conditions. Tripartite efflux systems are involved in the secretion of siderophores, which are also important for antibiotic resistance. For one of these efflux systems, the resistance-nodulation-cell division transporter ParXY from the model organism Pseudomonas putida KT2440, we show that it influences the secretion of the siderophore pyoverdine in addition to its already known involvement in antibiotic resistance. Phenotypically, its role in pyoverdine secretion is only apparent when other pyoverdine secretion systems are inactive. The results confirm that the different tripartite efflux systems have overlapping substrate specificities and can at least partially functionally substitute for each other, especially in important physiological activities such as supplying the cell with iron ions. This fact must be taken into account when developing specific inhibitors for tripartite efflux systems.


Materials included:
Table S1.Strains used in this investigation.
Table S2.Plasmids used in this investigation.
Table S3.Oligonucleotides used in this investigation.Sequences were taken from the homology search at string-db.org.Overall coverage can be found in Table S5.Respective strain and protein identifiers are (1)
Equivalent amino acids are highlighted based on identity (black background) and similarity according to their physicochemical properties (bold).Alignment was performed using the multiple sequence alignment tool from Clustal Omega (default settings) [12,13] and the ESPript 3.0 tool (%Equivalent, 0.7 Global scores, 0.5 Diff. score, coloring B&W) [14].The growth rates were calculated for WT and parX mutant, grown in MH medium supplemented with the antibiotic chloramphenicol (as shown in Fig. 6).Data was calculated for strains in the exponential phase.For statistical analysis, three biological replicates were analyzed using a 2-way ANOVA (Šídák's multiple comparisons tests, α 0.05).Only significant p-values are shown.

Figure S2 :
Figure S2: Impact of the deletion of parX on growth of P. putida KT2440 in the presence of chloramphenicol.

Figure S3 :
Figure S3: Impact of the deletion of parX on colony morphology.

Figure S4 .
Figure S4.Growth rate for complementation of ΔpmΔparX phenotype in CAA medium.

Figure S5 :
Figure S5: Secreted pyoverdine in the supernatant after 24 h in CAA medium.

Figure S6 :
Figure S6: Effect of FeCl3 addition on the growth of different mutants in KB medium supplemented with 2'2-bipyridyl.
Oligonucleotides used in this investigation.Restriction sites are marked in

Figure S4 :
Figure S4: Growth rate for complementation of ΔpmΔparX phenotype in CAA medium.Strains containing pSEVA224 plasmids shown in Fig.4were grown in a 96well plate with CAA medium (Fig.4B) containing 0.5 mM IPTG for gene expression and kanamycin (50 mg per ml) for plasmid maintenance.The growth rate was calculated for strains from Figure4Bin the exponential phase.The triple deletion ΔpmΔparX with the complementation plasmid pSEVA-parX (cyan) does not differ significantly from the double deletion strain with an empty plasmid (dark grey).In contrast, the triple deletion with the empty plasmid (light grey) differs significantly to both tested strains, with a p-value of 0.0052 and 0.0028, respectively.Mean values of a minimum of three biological replicates are shown.For statistical analysis, ANOVA and Tukey's multiple comparisons test (α 0.05) were used.

Figure S5 :
Figure S5: Secreted pyoverdine in the supernatant after 24 h in CAA medium.(A) Cultures from growth curves in Figure 3C were analyzed for their secreted pyoverdine.Mature pyoverdine in the supernatant was quantified by fluorescence measurements.After 24 hours of growth, 1 ml of culture from each flask was taken and centrifuged (3 min, 15 700 x g).Fluorescence was determined with an excitation wavelength of λex: 400 nm and emission wavelength of λem: 455 nm (Tecan infinite® M200 pro plate reader), respectively.Mean values of a minimum of three biological replicates are shown.ANOVA and Tukey's multiple comparisons test (α 0.05) were used for statistical analysis.(B) Cultures were imaged after 24h of continuous shaking at 30°C.The double deletion Δpm (left) shows more yellow-green pyoverdine than the triple deletion ΔpmΔparX (right) by eye.

Figure S6 :
Figure S6: Effect of FeCl 3 on the growth of different mutants in KB medium supplemented with 2'2-bipyridyl.KB medium supplemented with 1 mM Bip was inoculated from overnight culture to yield an OD600 of 0.05 and incubated as described above.(A) Comparison of growth in KB plus 1 mM Bip supplemented with 400 µM FeCl3 or (B) 200 µM FeCl3.OD600 was recorded for 19 h in a CLARIOstar Plus plate reader.(C) The area under the growth curve (AUC) was determined for the conditions from (A) and (B) and various additional conditions (0, 100, and 300 µM FeCl3) and a control medium (ctrl) containing only KB without iron chelator and additional iron.The experiment was performed with at least three biological replicates.

Table S1 2Table S1 .
Strains used in this investigation.

Table S2 .
Plasmids used in this investigation.