Introduction

Metal ions are essential to the functioning and survival of all known forms of cellular life. Among all trace elements, zinc (Zn) is of particular importance because it functions as a cofactor of many essential enzymes and is, more generally, needed to sustain protein structure [1,2,3]. Zn is found at the top of the Irving-Williams series (Mg < Ca < Mn < Fe < Co < Ni < Cu > Zn) that ranks metal ions according to their ability to form complexes with proteins [4]. This is why, when present in excess inside the cell, Zn tends to outcompete other metals and cause protein mismetallization, the major cause of its cellular toxicity [5]. The intracellular concentration of Zn has therefore to be tightly and finely controlled by complex homeostasis mechanisms. In bacteria, the cytoplasmic Zn concentration is maintained within the range of 10–3 to 10–4 M, making it the second most abundant metal after iron [6].

In the Gram-negative bacterium Pseudomonas aeruginosa, several systems are involved in Zn homeostasis [7, 8]. Zn import into the periplasm is mediated by TonB-dependent importers such as ZnuD. A P-type ATPase and three ABC transporters help with the uptake of Zn from the periplasm into the cytoplasm. In case of metal excess, Zn is exported by a specific P-type ATPase CadA, a Resistance-Nodulation-Cell Division (RND) efflux pump (CzcCBA), and two cation diffusion facilitators (CDF) that play a minor role in Zn resistance [6, 9]. Control of Zn import and export is extremely important for some bacterial pathogens, especially for their interaction with their host, whether an animal [10] or a plant [11]. P. aeruginosa, for example, is able to grow and thrive in Zn-poor environments such as the bloodstream, where trace metals are chelated to prevent microbial growth [6]. Conversely, in a phagolysosome, high concentrations of Cu and Zn have antimicrobial properties and can eliminate the pathogen [12].

Several regulators of Zn homeostasis have been identified in P. aeruginosa. The major regulator of Zn import is the Zinc Uptake Regulator, hereafter Zur, protein (formerly Np20, [13]), a Fur (Ferric Uptake Regulator)-like protein [14]. In the presence of an excess of cytoplasmic Zn, Zur binds to the metal ion, dimerizes, and, in its dimeric form, binds to specific Zur boxes on the DNA to repress genes involved in Zn import [7, 13]. Zn export is controlled by two other regulators: CadR, a MerR-type regulator that activates the transcription of the P-type ATPase CadA [15], and CzcR, a response regulator (RR) of the two-component system CzcRS, which, when phosphorylated by CzcS, activates the transcription of the efflux pump CzcCBA that expels Zn from the cell [16]. Until now, the regulatory networks controlling Zn import and Zn export have been considered essentially independent from one another and no crosstalk between them has been substantiated.

In this study, we discovered that the Zur protein is needed for the full expression of the Zn efflux system when P. aeruginosa switches from a low-Zn environment to an excess-Zn situation. We show that Zur accelerates czcR transcription, and thus indirectly activates the transcription of Zn efflux systems by binding to two DNA sites in a region upstream the czcR translation start site (ATG). This new function makes Zur the keystone of Zn homeostasis in P. aeruginosa, acting not only as a repressor of Zn import but also as activator of Zn export.

Results

Zur binds to the czcRS promoter

The Zur protein, when concentrations of Zn are sufficient or excessive, binds to a 17 bp DNA Zur box to repress genes involved in Zn import [17]. In a global bioinformatics search on the P. aeruginosa genome, we found, as expected, Zur boxes in the promoter regions of genes involved in the import of Zn, but also, intriguingly, upstream of the czcRS genes [8]. This suggested a possible direct regulatory link, mediated by Zur, between the import and export of Zn. In order to validate the in silico analysis, we purified the Zur protein and performed an electrophoretic mobility shift assay (EMSA) on the czcR upstream region. A clear DNA shift was observed in the presence of Zur (Fig. 1A). The interaction is Zn-dependent as no shift was detected in presence of TPEN, a chelator with high affinity for Zn. DNA footprint analysis revealed two boxes located approximately 40 bps upstream of the translation start codon of czcR (Fig. 1B, 1C and S1A).

Fig. 1
figure 1

Zur binds upstream the czcR translation start codon. A) Electrophoretic Mobility Shift Assay (EMSA) using purified Zur protein and czcR-5’UTR DNA region or znuB ORF DNA region as negative control. 25 ng of DNA per reaction was used. The Zur protein, Zn and TPEN (Zn chelator) concentrations are indicated at the top of the figure. B) DNAse I footprinting of czcR promoter in the absence or presence or of 1 µM Zur protein. DNA fragments were analyzed by capillary electrophoresis. The two Zur boxes are indicated by numbers 1 and 2 in red. The sequencing reaction is visible below the figure (seq2). C) Sequence of the two boxes as determined by the footprinting experiments (Fig. 1B coding strand; Fig S1A template strand)

Full size image

The transcription start site of czcR was determined by 5’RACE and it was found 300 bps upstream of the czcR translation start, revealing a 292-nucleotide 5’-UTR (Fig. 2A, S1B). Promoter fusions (-gfp) confirmed that a region longer than 292 bps upstream of the ATG is required to sustain czcRS transcription in the presence of Zn (construct #1 and #2 in Fig. 2B).

Fig. 2
figure 2

GFP reporter of the czcR promoters. A) Map of the czcR promoter and 5’-UTR DNA region with the two Zur boxes, the CzcR DNA binding site (according to [18]) and the czcR and czcC transcription start sites (+ 1). The number represents the nucleotide position relative to the translation start of czcR gene. B) GFP promoter fusions #1 and #2 transformed into the wt PAO1 strain. Cultures were incubated at 37 °C with shaking in a microplate reader and fluorescence measurement was performed every 15 min after the addition of 2 mM ZnCl2 (#1 + Zn, #2 + Zn) or without the addition of Zn (#1 and #2). The fluorescence values are normalized with cell density. Standard deviations of the triplicates are indicated

Full size image

To determine whether the binding at the two Zur boxes was independent, we repeated the band shift assay using different versions of the czcR 5’-UTR DNA region, where box1 (mut1), box 2 (mut2), or both (mut1 + 2) Zur boxes were mutated at key positions (Fig. 3A). A higher shift was obtained when the two Zur boxes were present compared to the presence of only one, and no shift was detected when the two boxes were mutated (Fig. 3B), indicating that the two boxes support the binding of the regulator independently.

Fig. 3
figure 3

Mutations in the Zur boxes and their effects on Zur binding activity. A) Sequences of modified Zur boxes: mut1, mut2 and mut1 + 2 in red. B) Electrophoretic Mobility Shift Assay (EMSA) using purified Zur protein, different czcR 5’-UTR DNA region (czcR) and znuB (negative DNA control). 25 ng of each DNA per reaction was used. The Zur protein and Zn concentrations are indicated at the top of the figure. czcR wt, corresponds to the wt sequence of the czcR 5’-UTR DNA region, czcR mut1 and mut2 possess mutated box 1 or 2, respectively; czcR mut1 + 2 possess the two mutated Zur boxes

Full size image

Zur is involved in zinc resistance

In order to investigate whether Zur had a role in the regulation of Zn export, we characterized the phenotype of a Δzur mutant. We found that, in the absence of Zur, P. aeruginosa is more sensitive to Zn excess than the parental strain (Fig. 4A). The phenotype is milder than in a ΔczcRS mutant, but more severe than in a ΔcadR mutant [15].

Fig. 4
figure 4

Zur is involved in CzcR expression. A) Growth curves of P. aeruginosa wt, Δzur, ΔczcRS and ΔcadR mutants in Zn-depleted medium (30 µM TPEN), in 2 mM and 4 mM ZnCl2, as indicated. B) Expression of czcR measured using qRT-PCR. RNAs were extracted at 0, 15 min, 1 h and 5 h after addition of 2 mM ZnCl2 as indicated. Fold expressions are compared to the wt strain at t0 and normalized by oprF expression. Standard deviations of the triplicates are indicated. Statistical analyses were performed according to the Student’s t-test and p-values are given as follows: ≤ 0.05 (*). C) Western blot analysis of total proteins sampled just before (0 h), or 1 h and 5 h after addition of 2 mM Zn of the wt and Δzur mutant, containing the empty plasmid (pE) or the IPTG inducible zur:6His gene (pzur). Blots were decorated with anti-CzcR, anti-6His for Zur detection and anti-Hsp70 as loading control

Full size image

The effect of the zur deletion on czcR transcription was assessed using qRT-PCR. In the absence of Zur, czcR expression decreased significantly compared to the control (Fig. 4B). The difference was obvious 15 min (400-fold decrease) after Zn addition, still visible after 1 h (230-fold decrease), but no longer present after 5 h of induction. An immunoblot analysis confirmed this expression pattern at the protein level (Fig. 4C). As expected, the delay in czcR expression was abolished when zur:6His (zur containing a C-terminal 6-histidine tag to follow its expression) was overexpressed in trans by the addition of 0.2 mM IPTG (Fig. 4C). It has been shown that the Zn inductions of the CzcCBA efflux pump and the cation diffusion facilitator CzcD were CzcR dependent [7]. In agreement with this, the expression of these two export systems were also reduced in the Δzur mutant (Fig. 5). Conversely, the P-type ATPase CadA, which is known to be overexpressed in the absence of CzcCBA (i.e. ΔczcRS or ΔczcA [15]), showed increased expression in the Δzur mutant (Fig. 5). Altogether, these results strongly suggest that Zur is required to bring about the full expression of czcR and mount the proper response to a Zn excess. Deleting the Zur regulated major Zn import permease ZnuB has no effect on the zur mutant growth phenotype in the presence of Zn (Fig. S2) suggesting a direct effect of Zur on czcR expression. Using transcriptional gfp-fusions of czcC, including wt or mutated Zur boxes, no difference in fluorescence was observed, that Zur affects indirectly the CzcCBA expression, through CzcR (Fig. S3).

Fig. 5
figure 5

Expression of czcC, czcD and cadA measured using qRT-PCR. RNAs were extracted at 0, 15 min, 1 h and 5 h after addition of 2 mM ZnCl2 as indicated. Fold expressions are compared to the wt strain at t0 and normalized by oprF expression. Standard deviations of the triplicates are indicated. Statistical analyses were performed according to the Student’s t-test and p-values are given as follows: ≤ 0.05 (*), ≤ 0.001 (***)

Full size image

Discussion

Zinc is an essential element whose cellular concentration is tightly regulated [19]. It is vital for cells and must always be present in sufficient amounts, but can become toxic as soon as it exceeds a threshold [20]. In the bacterium P. aeruginosa, eight zinc import and four zinc export systems have been described [7]. In case of zinc excess, the Zur protein represses the expression of many import systems, thus preventing too much Zn from entering the cell. In this work we described a novel key function of the Zur protein in P. aeruginosa, being involved not only in the repression of the import of Zn, but also in the early and fast induction of the CzcCBA Zn export system. Zur is therefore a keystone of Zn homeostasis in P. aeruginosa, integrating both the import and export of the ion.

In some other bacteria, Zur also acts as an activator of zinc efflux genes. In Xanthomonas campestris, for instance, Zur binds a GC-rich sequence overlapping the -35 to -10 region of the XC2976 gene, encoding a CzcD homolog, and directly activates its transcription [21]. In Caulobacter crescentus a Zur box, located just upstream the -35 element of a Zn export system operon allows its expression in the presence of Zn [22]. Another situation has been observed in Streptomyces coelicolor, where Zur binds to a Zur box located just upstream the -35 region of the zitB gene, a czcD homolog. Interestingly, at low Zn concentrations, Zur binds as a dimer to this region, weakly activating zitB transcription. As Zn concentration increases, Zur multimerizes on the promoter and increases zitB transcription [23]. Zur also acts as a positive transcriptional regulator of T6SS4 in Yersinia pseudotuberculosis by binding to the -90 DNA region [24]. This secretion system exports YezP, a Zn binding protein implicated in oxidative stress responses [25]. Other processes, not directly related to metal homeostasis, have also been found to be positively regulated by Zur. This is the case in Streptomyces avermitilis where Zur, by binding to the transcription start site, activates the transcription of the aveR gene, inducing the synthesis of the Avermectin secondary metabolites [26]. In the case of P. aeruginosa, we found that the Zur protein binds in a Zn-dependent manner to two DNA regions (Zur box1 and 2) located 200 bp after the czcR TSS, in the 5’-UTR DNA region. This binding has a positive effect on czcR transcription and, in general, on Zur-mediated Zn resistance.

What is the molecular action of Zur for the activation of czcR transcription? Recently numerous two-compoent systems response regulators (TCS-RR) binding sites have been detected in the intergenic region between czcR and czcC [27]. Zur could therefore be involved in the competition with the binding of transcriptional repressors in this region. Fine tuning of the efflux system could be essential in environmental conditions not tested in the laboratory. For instance, rapid changes in Zn concentration are known to occur during infection, because of nutritional immunity and metal excess in the phagolysosome [28, 29]. The tight efflux pump control might also be important in the environment, where rapid changes in metal concentrations might also occur. Interestingly, among the Pseudomonas species analyzed for the presence and location of Zur boxes (i.e. P. chlororaphis, P. fluorescens, P. putida, P. stuzeri and P. syringae), only P. aeruginosa possesses boxes upstream of the czcR gene [8]. This characteristic could be related to its ability to proliferate in presence of high metal concentrations. Thus, in P. aeruginosa, Zur appears to be the central cog in Zn homeostasis since it controls both the import and export of this metal. CzcR is also known to be capable of controlling the expression of the gene coding for the OprD porin, involved in imipenem resistance, or the lasI gene involved in quorum sensing [16, 30]. The importance of Zur in the control of czcR expression highlights a novel and important role of Zur in the interaction between Zn homeostasis, virulence factor expression and antibiotic resistance in P. aeruginosa. Deciphering this novel and important regulation occurring by Zur at the czcR gene vicinity is therefore of prime interest.

Materials and methods

Bacterial strains and culture media

The bacterial strains are listed in supplementary table S1. The modified Luria–Bertani medium (M-LB) used for this work was prepared as mentioned previously [15]. When required, antibiotics were added to the medium at the following concentrations: 50 μg/mL tetracycline (Tc, Axxora), 200 μg/mL carbenicillin (Cb, phytotechlab), 50 μg/mL Gentamycin (Gm, AppliChem) for P. aeruginosa or 100 μg/mL ampicillin (Ap, AppliChem) and 15 μg/mL Tc or Gm for E. coli.

Genetic manipulations

The primers used for cloning are detailed in supplementary table S2. Restriction and ligation enzymes (Promega), or fragment insertion using the Gibson assembly Cloning kit (New England Biolabs) were employed according to the supplier’s instructions. Plasmids were transformed into E. coli DH5α by heat shock, verified by sequencing before being electroporated into the P. aeruginosa wild type or Δzur strains [31]. For transcriptional gfp-fusion assays, the two czcR promoter regions (#1 and #2 in Fig. 2A) were obtained by PCR amplification. Fragments were then digested with KpnI and BglII enzymes and ligated into the corresponding sites of the pBBR1-gfp vector. For Zur6HIS overexpression assays, the full zur gene was amplified by PCR with a reverse primer containing the 6xHIS tag. The fragment was digested with the EcoRI and BamHI enzymes and ligated into the pMMB66EH plasmid. The pME6001-czcRS vector was used as a template for the sequencing reactions of the 5’RACE experiments. For this construction, the full czcRS operon and its promoter region were amplified by PCR and cloned into the pME6001 vector with the XhoI and HindIII restriction sites.. For the Δzur mutant, a DNA product, consisting of the two 500 bp amplicons flanking the zur gene fused together, was obtained by megaprimer amplification, digested with EcoRI and BamHI restriction enzymes and ligated into the pME3087 plasmid according to standard procedure [32]. After verification, the resulting plasmids were transformed into the P. aeruginosa wt strain. Merodiploids were selected as described previously [33] and deletions were confirmed by sequencing.

Growth experiments

To monitor the growth of the wt and the mutant strains, overnight cultures were diluted in M-LB medium to an OD600 of 0.05, supplemented with N,N,N′,N′-tetrakis(2-pyridylmethyl)-ethylenediamine (TPEN; Brunschwig) or ZnCl2 (Fluka) as indicated in Figs. 3A and 4. Cultures were incubated at 37 °C with shaking in a microplate reader (BioTek Instruments) and absorbance at 600 nm was measured every 15 min, for 12 h. For strains carrying the pBBR1 plasmid with gfp fusions #1 or #2, overnight cultures were diluted to an OD600 of 0.1 and cultured for 2.5 h in the presence of 30 μM TPEN before adding ZnCl2. Absorbance at 600 nm and fluorescence at 528 nm were then monitored every 15 min. Arbitrary units indicated in the Fig. 2B correspond to the fluorescence values normalized with the cell density.

Immunoblot analyses

Immunoblot analyses were performed as previously described [30]. Briefly, overnight cultures were diluted to an OD 0.1 and incubated 2.5 h in M-LB supplemented with 30 μM TPEN. Isopropyl-1-thio-D-galactopyranoside (IPTG, Sigma Aldrich) was added at a final concentration of 0.2 mM to the cultures of strains complemented with zur:6His on the pMMB66EH or the empty plasmid. 1 ml was collected and centrifuged immediately prior to ZnCl2 addition (t0) and after several time points, as specified in Fig. 4C. Pellets were solubilized in the appropriate volume of 2 × β-mercaptoethanol gel-loading buffer (an OD600 of 1 gives 0.175 mg/ml of protein) to obtain a final concentration of 2 mg/ml and loaded onto SDS PAGE, using 4–12% precast gels (Thermofisher Scientific). Transfers were performed with iBlot2 transfer stacks (Invitrogen) and nitrocellulose membranes were incubated with the anti-CzcR [34], anti-penta HIS (Invitrogen) and anti-HSP70 [30] antibodies. Blots were revealed by chemiluminescence using the Amersham Imager 680 System.

EMSA

The ability of Zur to bind the 5’-UTR DNA region of czcR, was determined by electrophoretic mobility shift assays (EMSA). The znuB gene, the 5’-UTR of czcR, wt or containing the mutated zur box-es (#mut1, #mut2 or #mut 1 + 2), were amplified by PCR from either the P. aeruginosa wt genomic DNA or from the in vitro synthetized mutated DNA (GeneArt, Thermo Fisher Scientific, Fig. 3A). Primers used for PCR amplification are indicated in supplementary table S2. Amplicons were then purified on agarose gel. Binding assays were performed as previously described [7]. Results were analyzed by staining the gel with 0.1% ethidium bromide and revealing with UV light using a NuGenius instrument.

DNAse I Footprinting assays

The czcR promoter was amplified by PCR from the wild type P. aeruginosa genomic DNA with 5’ Fluorescein (6FAM) primer as indicated in supplementary table S 2. DNA, with the plus or minus strand labeled independently, were gel purified. 50 ng of DNA fragments were mixed with or without 1 µM Zur in 40 μL of EMSA binding buffer [7] supplemented with 5 μM zinc. The reaction was incubated for 30 min at room temperature, followed by partial DNAse I digestion as described previously [15]. The sequence of the pME6001-czcRS plasmid, determined with the same FAM-labeled primers and the Thermo Sequenase Dye kit (Thermofisher Scientific) was used to align the peaks. All fragments were analyzed by capillary sequencing (Microsynth AG, Switzerland) using ILS600 as a size standard, then peaks were assessed using PeakScanner2 software (Thermofisher Scientific).

RNA extraction

Overnight cultures of the P. aeruginosa wt or mutant strains were diluted to an OD600 of 0.1 in M-LB containing 30 μM TPEN and cultured for 2.5 h before being induced with 2 mM ZnCl2. For RT-qPCR, 0.5 ml of culture was mixed with 1 ml of RNA protect (Qiagen) immediately prior to metal addition (t0) and after several time points as indicated in the figures. The P. aeruginosa wt culture intended for the 5’RACE experiment was collected after 5 h of 2 mM ZnCl2 induction. Total RNA extractions were performed as mentioned in previous publications [34]. Briefly, pellets were resuspended in 100 μL Tris–EDTA buffer supplemented with 5 mg/mL Lysozyme (Fluka) and incubated at room temperature for 10 min. The following steps were carried out with the RNeasy mini kit (Qiagen), according to the manufacturer’s directives. Residual genomic DNA was removed by treating with 10 units/sample of RNAse-free RQ1 DNAse (Promega) for 1 h at 37 °C followed by phenol–chloroform extraction and ethanol precipitation. Total RNA was then resuspended in RNAse-free water.

RT-qPCR

Quantitative RT-PCR were performed as previously described [35]. Briefly 500 ng of total RNA were reverse transcribed using random primers (Promega) with the ImProm-II reverse transcriptase (Promega) according to the manufacturer’s instructions. Quantitative PCR was performed on a tenfold dilution of resulting cDNA and using the SYBR Select Master Mix (Applied Biosystem), with the primers listed in table S2. Results were analyzed as formerly mentioned [36] and standardized with oprF.

5’RACE

The transcription starts of the czcR and czcC genes, were determined using the Rapid amplification of cDNA-5’ends (5’RACE) kit (Roche). Briefly, 5 μg of total RNA from the P. aeruginosa wt were reverse transcribed using the specific primers sp1R for czcR mRNA or sp1C for czcC mRNA. Resulting cDNAs were purified with the Wizard SV Gel and PCR Clean-Up System (Promega) and tailing with dATP and TdT. Tailed-cDNA strands were then amplified with the dT-Anchor primer (Roche) and the specific primer sp2R or sp2C (for czcR cDNA and czcC cDNA respectively). A 20-fold dilution of the first PCR was used as template for a second PCR with primers Anchor (Roche) and sp3R for czcR cDNA or sp2C for czcC cDNA. PCR products were then purified by agarose gel electrophoresis and subcloned into a pCR2.1 vector with the TA Cloning kit (Life Technologies) according to the supplier’s instructions. Plasmids isolated from E. coli clones were analyzed by sequencing. The transcription starts of czcR or czcC indicated in Fig. 2A correspond to the first nucleotide following the polyA tail.

Experimental relevance and statistical data

Regarding the graph representations, the mean values of at least three independent experiments are indicated in the figures, along with the corresponding standard deviations. When specified, statistical analyses were performed according to the Student’s t-test and significance p-values were set at p ≤ 0.05(*) or p ≤ 0.001 (***). For other figures, data points represent the mean of triplicate values.