Investigation of Zur-regulated metal transport systems reveals an unexpected role of pyochelin in zinc homeostasis

Abstract

Limiting the availability of transition metals at infection sites serves as a critical defense mechanism employed by the innate immune system to combat microbial infections. Pseudomonas aeruginosa exhibits a remarkable ability to thrive in zinc-deficient environments, facilitated by intricate cellular responses governed by numerous genes regulated by the zinc-responsive transcription factor Zur. Many of these genes have unknown functions, including those within the predicted PA2911-PA2914 and PA4063-PA4066 operons. A structural bioinformatics investigation revealed that PA2911-PA2914 comprises a TonB-dependent outer membrane receptor and inner membrane ABC-permeases responsible for importing metal-chelating molecules, whereas PA4063-PA4066 contains genes encoding a MacB transporter, likely involved in the export of large molecules. Molecular genetics and biochemical experiments, feeding assays, and intracellular metal content measurements support the hypothesis that PA2911-PA2914 and PA4063-PA4066 are engaged in the import and export of the pyochelin-cobalt complex, respectively. Notably, cobalt can reduce zinc demand and promote the growth of P. aeruginosa strains unable to import zinc, highlighting pyochelin-mediated cobalt import as a novel bacterial strategy to counteract zinc deficiency. These results unveil an unexpected role for pyochelin in zinc homeostasis and challenge the traditional view of this metallophore exclusively as an iron transporter.

Importance: The mechanisms underlying the remarkable ability of Pseudomonas aeruginosa to resist the zinc sequestration mechanisms implemented by the vertebrate innate immune system to control bacterial infections are still far from being fully understood. This study reveals that the Zur-regulated gene clusters PA2911-2914 and PA4063-PA4066 encode systems for the import and export of cobalt-bound pyochelin, respectively. This proves to be a useful strategy to counteract conditions of severe zinc deficiency since cobalt can replace zinc in many proteins. The discovery that pyochelin may contribute to cellular responses to zinc deficiency leads to a reevaluation of the paradigm that pyochelin is a siderophore involved exclusively in iron acquisition and suggests that this molecule has a broader role in modulating the homeostasis of multiple metals.

Keywords: Pseudomonas aeruginosa; cystic fibrosis; iron transport; metal transport; metallophore; siderophores; zinc transport; zincophores.

Fig 1

Schematic representation of the PA2911-PA2914 and PA4063-PA4066 operons. The corresponding gene ID in the PA14 strain and a brief description of the function are reported for each ORF. Sources are ^aPseudomonas Genome Database (https://www.pseudomonas.com) and ^b(33). The Zur boxes (highlighted sequences) were previously identified by Pederick et al. (34).

Fig 2

(A) Transcription of PA4063 and PA2911 in response to Zn availability. qRT-PCR on PA2911 and PA4063 from wild-type and znuAzrmB mutant strains grown for 20 hours in E-VBMM with or without ZnSO₄ 3 mM, as indicated in the legend. Data are mean values ± SD of triplicates and expressed as relative fold expression (log₂DDCt) compared to the gene expression in the wild-type strain grown in E-VBMM (control). Statistical analyses were performed by two-way ANOVA and Bonferroni’s multiple comparison test. Asterisks indicate statistically significant differences between all samples and the control (*P < 0.05; **P < 0.01; ****P < 0.0001); hash signs indicate statistically significant differences between the znuAzrmB strain grown in E-VBMM and E-VBMM +Zn (^####P < 0.0001). (B) Contribution of PA2914 and PA4065 to PA14 growth under conditions of Zn limitation. Growth curves of wild-type, znuAzrmB, and mutant strains carrying the PA2914 deletion or the PA4065 deletion, as indicated in the legends, grown in E-VBMM. Each symbol indicates the mean ± SD of triplicates, and lines represent nonlinear fit according to the Logistic Growth equation. (C) Effect of PA4065 and PA2914 deletions on intracellular Zn request. The beta-galactosidase activity driven by the zrmA promoter was evaluated after 20 hours of growth in E-VBMM in strains carrying plasmid pzrmApTZ110. Each bar represents the mean ± SD of three biological replicates. Statistical significance was calculated by one-way ANOVA and Bonferroni’s multiple comparison test. Asterisks indicate statistical significance between wild-type (wt) and mutant strains (*P < 0.05; ****P < 0.0001); hash signs indicate statistical significance between PA4065znuAzrmB and the other mutant strains (^#### P < 0.0001).

Fig 3

Molecular modeling of PA4064-PA4065. Molecular representation of the predicted MacB transporter. (A) Side view of the four protein subunits (Q9HWW3 and Q9HWW4) embedded within the lipid bilayer. The extracellular domain (ECD), transmembrane (TM) helices, and nucleotide-binding domain (NBD) are labeled. α-helices are represented as orange spirals, β-strands through violet arrows, while the membrane is shown using a stick representation. (B) Spacefill representation of the transporter, highlighting the arrangement of the four distinct protein subunits, each shown by a different color.

Fig 4

PCH synthesis is downregulated in the znuAzrmB strain. (A) UV-vis absorbance spectra (230–600 nm) of ethyl acetate fractions extracted from the supernatants of cultures grown for 20 hours in VBMM. The arrows denote peaks at 250 and 315 nm wavelength matching the PCH spectrum. (B) Relative fold expression (log₂ΔΔCt) of genes involved in PCH uptake (ftpA and fptX), and synthesis (pchD and pchE), in the znuAzrmB mutant grown in E-VBMM for 20 hours, relative to the PA14 wild type. Data are mean values ± SD of triplicates, and statistical analysis was performed by paired t test. Asterisks indicate statistically significant differences between the znuAzrmB strain and wild type, grown in the same conditions (**P < 0.01; ***P < 0.001; ****P < 0.0001).

Fig 5

Effects of PCH on intracellular Zn request. (A) The zrmA promoter activity was evaluated in strains carrying plasmid pzrmApTZ110, grown in E-VBMM for 20 hours. Each bar represents the mean ± SD of three biological replicates, and statistical significance was calculated by one-way ANOVA and Bonferroni’s multiple comparison test. Asterisks indicate the statistical difference with the znuAzrmB strain (*P < 0.05; **P < 0.01); hash signs indicate the statistical difference between the znuAzrmBpchE and the other strains (^##P < 0.01); § shows statistical difference with the PA4065znuAzrmB strain (^§§§ P < 0.001) (B) The zrmA promoter activity was evaluated in strains carrying plasmid pzrmApTZ110, grown for 20 hours in E-VBMM (-) or supplemented with the ethyl acetate fraction of supernatants from wild-type (wt) or pchE strains (pchE). The decrease in zrmA expressions in cultures treated with supernatants from the pchE strain is likely caused by trace amounts of Zn introduced during the ethyl acetate extraction. Data represent means ± SD of three biological replicates. Two-way ANOVA and Tukey’s multiple comparison tests were performed to assess significant differences (***P < 0.001; ****P < 0.0001; ns: not significant).

Fig 6

Quantification of PCH in cell lysates and cell-free supernatants. Wild-type and mutant strains were grown in VBMM, and PCH production was quantified using the P. aeruginosa ΔpvdAΔpchDΔfpvA PpchE::lux biosensor, as described in Materials and Methods. Data are expressed as µM/OD₆₀₀; each bar is the mean value of three independent experiments ± SD. One-way ANOVA and Tukey’s multiple comparison tests were performed to assess significant differences among wild-type (wt) and mutant strains (***P < 0.001; ****P < 0.0001), znuAzrmB and other mutant strains (^#P < 0.05; ^##P < 0.01; ^####P < 0.0001) and between PA4065 and PA4065znuAzrmB (^§§§§P < 0.0001).

Fig 7

Influence of PCH on intracellular Zn request. The zrmA promoter activity was evaluated in znuAzrmB (left panel), PA2914znuAzrmB (middle panel), and PA4065znuAzrmB (right panel) strains carrying plasmid pzrmApTZ110, grown for 20 hours in VBMM supplemented (+) or not (−) with 1 µM apo-PCH. Each bar is the mean value of three biological replicates ±SD. Two-way ANOVA and Sidak’s multiple comparison tests were performed to assess significant differences, and asterisks indicate significant differences between untreated and PCH-treated samples (**P < 0.01; ****P < 0.0001); ns: not significant.

Fig 8

Intracellular Zn and Co content. ICP-MS analyses of bacteria grown for 20 hours in E-VBMM with trace metals (0.2 µM ZnSO₄ and 0.1 µM FeSO₄, NiSO₄, Co(NO₃)₂, CuSO₄ and MnCl₂). Bars are the mean value of three biological replicates ±SD. Statistical analyses were performed by one-way ANOVA. Asterisks indicate statistically significant differences between mutant strains and wild type (**P < 0.01; ****P < 0.0001); hash signs indicate pairwise statistically significant differences (^####P < 0.0001; ns: not significant).

Fig 9

Effect of pchE on Co supplementation of strains carrying the PA4065 deletion. Growth curves of PA4065znuAzrmB and PA4065znuAzrmBpchE strains in E-VBMM (filled symbols) and E-VBMM + Co(NO₃)₂ 10 µM (empty symbols). Each point indicates the mean value ±SD of triplicates, and lines represent nonlinear fit according to the logistic growth equation. Statistical differences among the end points of each curve were analyzed by two-way ANOVA (*P < 0.05; **P < 0.01; ns: not significant).

Fig 10

Schematic view of P. aeruginosa responses to Zn deficiency. The diagram illustrates the metal transport systems for which a role in the response to Zn deficiency is demonstrated. As in all Gram-negative bacteria, the Zur-regulated ZnuABC transporter has a central role in ensuring the transport of Zn from the periplasm to the cytoplasm (the dashed line through the Outer Membrane (OM) indicates that the entry routes of Zn into the periplasm are not defined). In addition to znuABCD, the operons zrmABCD (responsible for the synthesis and transport of pseudopaline), PA2911-PA2914, and PA4063-PA4066 are also negatively regulated by Zur under Zn-replete conditions and transcribed when the intracellular Zn concentration is low, a condition in which Zur is no longer properly metalated. The zincophore pseudopaline, synthesized in the cytoplasm by ZrmB and ZrmC, is transported into the periplasm by ZrmD and then exported out of the cell by the MexAB-OmpR efflux pump (64). Pseudopaline complexed to Zn or Co re-enters the periplasm via the TonB-dependent Outer Membrane receptor ZrmA. It is currently unknown whether the pseudopaline-bound metal is released in the periplasm or transported into the cytoplasm in complex with the pseudopaline (dashed arrows through the Inner Membrane—IM). Co, which can partially compensate for conditions of severe Zn deficiency, can enter either in a complex with pseudopaline or bound to PCH. The PCH-Co complex can enter the cell either from FptA, the receptor involved in the uptake of the PCH-Fe complex, or through the PA2911-PA2914 import system. The binding of PCH-Co to PchR, the main regulator of PCH synthesis, represses the synthesis of PCH and the FptA-FptX import system (53). By contrast, the expression of PA2911-PA2914 is not repressed by PchR, but induced under conditions of poor Zn availability, thereby allowing the entry of PCH-Co even if fptA-fptX is repressed. To avoid potentially toxic accumulations of PCH and Co within the cell, excess PCH-Co is exported through a pump involving the MacB transporter formed by PA4064/PA4065.