Rsp activates expression of the Cnt system in Staphylococcus aureus

Background The Cnt system is crucial for the optimal import of essential metals in metal-limiting conditions and contributes to virulence in S. aureus. In a screen for regulators of efflux pumps in a phage-based ultra-high-density transposon library, we identified Rsp as a candidate regulator of the cntE gene. Results A two-fold decrease in expression of all genes of the cnt operon was observed by RT-qPCR in the rsp mutant compared to the parental strain, indicating that Rsp acts as an activator of the cnt operon. To determine whether the Rsp activation depends on iron, we compared mutant and parent cnt expression under varying metal conditions. A 2-fold reduction in cnt gene expression was detected in the rsp mutant in TSB, and a slightly smaller decrease (1.9, 1.7, and 1.5-fold changes for cntK, cmtA, and cntE respectively) was observed after addition of dipyridyl. The greatest decrease was seen with addition of FeSO4 (4.1, 5.3 and 6.3-fold changes for cntK, cmtA and cntE respectively). These findings suggest that Rsp activates the cnt operon in low and high iron conditions. To study the relationship between Rsp and the cnt repressors Fur and Zur, we created single and double mutants. Both fur and zur single mutants had significant increases in cnt gene expression compared to the parental strain, as did the fur rsp double mutant. The zur rsp double mutant also had a significant increase in cntK expression and a trend in increases in cntA and cntE expression just below statistical significance. Thus, the ability of Fur and Zur to repress cnt gene expression are not eliminated by the presence of Rsp. However, there were significantly smaller increases in cnt gene expression in the double mutants compared to single mutants, suggesting that Rsp activation can still occur in the absence of these repressors. To determine if Rsp directly modulates expression of cnt genes, incubation of purified Rsp caused a DNA-specific band shift for the cntK and cntA promoters. Conclusions Rsp activation may act to maintain basal cellular levels of staphylopine to scavenge free metals when needed, in addition to metal dependent regulation by Fur and Zur. Supplementary information Supplementary information accompanies this paper at 10.1186/s12866-020-02013-0.


Background
Staphylococcus aureus poses a serious risk to public health due to its prevalence as a colonizer of skin and nares, its ability to cause a wide range of infections, and the increasing incidence of antibiotic-resistant strains [1]. During infection, S. aureus is confronted with a robust host innate immune response and an environment with low availability of free iron, manganese, and zinc [1,2]. To overcome this host defense mechanism, S. aureus has evolved a diverse array of metal acquisition strategies that facilitates its proliferation and pathogenesis in host tissues [2,3]. S. aureus captures free metals from the host through high-affinity transporter systems or small chelating molecules called metallophores [3].
Staphylopine (StP) is a nicotianamine-like metallophore secreted by S. aureus, with a remarkably broad spectrum of metal ligands, including nickel, cobalt, copper, zinc, and iron [4,5]. The nine-gene cntKLMABCDFE operon is required for StP synthesis and trafficking. The first three genes cntK-M are responsible for StP biosynthesis; cntA-F encode the importer of the StP-metal complex; and CntE is involved in the export of StP [4][5][6][7].
The Cnt system is repressed by Fur and Zur in the presence of iron and zinc, respectively, with tight control of the StP biosynthesis genes and a looser control of the genes responsible for its export and recovery [8]. Besides the iron-dependent regulator Fur, an iron-independent regulation of the cnt regulon has been considered but not yet identified [8].
In this study, we identified a novel regulator of the cnt operon that activates the system in addition to the metal-dependent Fur and Zur repressors.

Results
Regulation of the cnt operon by the AraC-type regulator, Rsp Fur is an iron-dependent transcriptional regulator that uses Fe (II) as a cofactor and negatively regulates transcription of iron transport genes by binding to the two Fur boxes present in the cnt operon [8]. Besides the Furdependent regulation, a Fur-independent regulation by iron was also described, but these regulators have not been identified [8], and nothing is known about an ironindependent regulation of the cnt operon. In addition, no activators of the cnt system has been described.
In a screen for regulators of efflux pumps and/or exporters, using ciprofloxacin selection in a previously reported phage-based ultra-high-density transposon library procedure [9] we identified Rsp, as a candidate regulator of the cntE exporter gene. The screen used a selection at varying concentrations of ciprofloxacin, a substrate of both the NorA and NorB efflux pumps, and identified differences in knockouts of genes of transcriptional regulators.
To evaluate further the role of Rsp in cnt expression, a rsp mutant (NE1304) and its parental strain (USA300 JE2) from the Nebraska Transposon Mutant Library [10] were used to measure the expression of all genes of the cnt operon by RT-qPCR. As shown in Table 1, the expression of all genes in cnt operon decreased by an average of two-fold in the rsp mutant compared to the parental strain, being statistically significant for all genes of the operon, indicating that Rsp acts as an activator not only of cntE but also of the entire cnt operon. This finding was further supported by the results that restoration of the presence of Rsp via introducing plasmid encoding rsp gene (pRsp) reestablished the expression of cntK, cntA and cntE genes to the levels of wild type strain (USA300 JE2) with the empty plasmid as the control (Fig. 1).

Rsp activation acts in iron-independent manner
To determine if the Rsp activation acts in ironindependent manner, we measured the level of expression of cntK, cntA and cntE under varying metal conditions. As it is shown in Fig. 2a, the transcript levels of the three genes in the rsp mutant strain decreased in the three conditions tested, rich media (TSB), metaldepleted (TSB + DIP) and metal-replete medium (TSB + DIP + Fe) relative to the wild type strain (USA300 JE2). A 2-fold change was detected in TSB for all the genes of the operon. A slightly smaller decrease (1.9-, 1.7-and 1.5-fold changes for cntK, cmtA and cntE, respectively) was observed under DIP conditions. Additionally, the decrease was amplified with addition of 50 μM FeSO 4 to DIP media (4.1-, 5.3-and 6.3-fold changes for cntK, cntA and cntE respectively) (Fig. 2a), likely due to the maximum level of repression of Fur in an iron-rich media. These data suggest that Rsp can activate the cnt operon under varying levels of iron-dependent repression by Fur, thereby supporting a basal level of the cnt gene expression in the cells. In complementation assays in the iron-depleted media, the decrease in cnt gene expression observed in the rsp mutant strain carrying the empty plasmid pTZN10 (3.3-, 2-and 3.2-fold changes detected for cntK, cmtA and cntE, respectively) was complemented to wild type levels with the plasmid containing rsp (pRsp). (1.4-, 0.92-and 1.2-fold changes for cntK, cntA and cntE, respectively) (Fig. 2B1). The same complementation was observed in the iron-repleted media, when pRsp was introduced into the rsp mutant. Compared to the rsp mutant strain with the empty plasmid pTZN10, the rsp strain with pRsp exhibited changes of 3.47-vs 1.35-fold for cntK, 3.3-vs 0.96-fold for cntA and 2.7-vs 0.9-fold for cntE, further confirming the effect of the Rsp activation in varying iron conditions (Fig. 2B2).
Relationship between the repressors Fur and Zur and the activator Rsp in the regulation of the cnt operon As noted, Fur negatively regulates transcription of the cnt operon in an iron-dependent manner [8]. In  addition, zinc regulation of the cnt operon through the repressor Zur has been previously reported [8]. To determine the possible interaction of Rsp with Fur and Zur in controlling cnt expression, we created single (fur, zur) and double mutants (fur rsp, zur rsp) in a USA300 background and measured the expression of cnt genes. As expected, the expression of cntK, cntA and cntE genes was increased substantially in the fur mutant (23-,14-, and 7-fold, respectively) as well as the zur mutant (15-, 17-, and 13-fold, respectively) (Fig. 3a). Although these increases were reduced in the rsp double mutants (by1.5-to 1.6-fold for fur rsp and by 1.3-to 1.6-fold for zur rsp) relative to single mutants there remained substantial increases in cnt gene expression relative to the parental strain the double mutants (7-to 15-fold for fur rsp and 7-to 12-fold for zur rsp). Thus, repression by Fur and Zur can occur in the presence and absence of Rsp activation, however, was sufficient to significantly partially counter Fur repression for cntK and cntE and to partially counter Zur repression for cntA and cntE (Fig. 3b).
Binding of Rsp to the promoter region of the cnt operon To determine if Rsp acts directly to modulate expression of cnt genes, Rsp was expressed in a pQE-9 (His tag expression vector) previously constructed [11] using E. coli  affinity column, an SDS-PAGE gel indicated a homogenous single protein band (data not shown). Two promoters have been previously determined for the cnt operon, one within the upstream region of cntK (cntK promoter) and the second one within the intergenic region between cntM and cntA genes (named cntA promoter) [8]. Incubation of Rsp with either the 494-bp cntK or the 381-bp cntA promoter fragments resulted in a DNA band shift (Fig. 4). These bands shifts were reduced in the presence of 200-fold excess unlabeled specific promoter region DNA and remained unchanged in the presence of a similar excess of salmon sperm DNA, indicating specific binding to the promoter DNA fragment.

Discussion
The Cnt system is essential for the optimal import of nickel, cobalt, zinc and ferrous iron when these metals are limited, and it contributes to virulence in S. aureus [6,12,13].
Several studies have identified Fur and Zur as repressors of the Cnt system [6][7][8] but, no activators of the operon have been identified. In this study, we identified that the AraC-type global regulator Rsp is involved in the activation of the cnt operon by binding directly to the cntK and cntA promoters. Since both a Fur box and a Zur box are present upstream of cntK and cntA promoters [8], it is possible that Rsp binding involves one of those boxes. However, the expression of the Furregulated sbn operon [14], was not changed in a rsp mutant (data not shown), suggesting that Rsp does not bind to the Fur boxes in general. Another study demonstrated that Rsp upregulates agrA and downregulates ica by direct binding of their promoters [11]. The predicted binding motifs of the agrP2 and the ica promoters, however, are different [11] and at the same time no similar sequence was found in the fnbA promoter, which reportedly also binds Rsp [15]. Those possible binding motifs were also not present in the cnt promoters. These findings favor our hypothesis that Rsp does not bind to the Fur box, and determination of the binding motif of Rsp will await further study. The upregulation of the Cnt system by Rsp was demonstrated to occur in both low and high metal conditions suggesting that Rsp acts to maintain a basal level expression of the cnt operon in the presence of Fur-mediated iron-dependent repression. Notably the effect was greatest when Fur repression was increased under high iron conditions. To evaluate the interaction of Rsp with Fur or Zur repressors, we constructed double mutants. Notably the increases in cnt gene expression levels in fur and zur single were not eliminated in the rsp double mutants. A partial counter effect of Rsp on Fur repression was seen for cntK and cntE and a partial counter effect of Rsp on Zur was seen for cntA and cntE. Thus, Rsp, Fur, and Zur can act independently of each other in affecting levels of cnt gene expression.

Conclusions
In conclusion, we identified for the first time an activator of the Cnt system, Rsp, that acts in both low and high iron conditions and contributes to a basal level of production of the StP synthesis genes and StP importers to scavenge free metals when needed. Rsp acts in addition to the known regulation by repressors Fur and Zur. This tight and complex regulation also could prevent the toxicity due to the StP accumulation by the activation of the expression of the StP exporter cntE gene [16].

Bacterial strains and growth conditions
All the strains and plasmids used in this study are summarized in Table 2. S. aureus strains were routinely cultivated in trypticase soy broth (TSB) and on tryptic soy agar plates (TSA), while E. coli strains were grown in Luria-Bertani broth (LB). Bacteria were grown at 37°C unless stated otherwise. The following antibiotics were obtained from Sigma-Aldrich (St. Louis, MO) and used for plasmid maintenance at the concentrations indicated: ampicillin, 100 μg/ml; kanamycin, 25 μg/ml or 75 μg/ml; and tetracycline, 10 μg/ml.

Construction of S. aureus mutant strains
S. aureus USA300 fur and USA300 zur single mutants and USA300 fur rsp and USA300 zur rsp double mutants strains were created via bacterial phage φ85 transduction from S. aureus Newman fur::tet and S. aureus NCTC8325 zur::tet strains respectively as previously described [5]. Colonies of interest were selected on TSA plates containing sodium citrate (10 μg/ml) and tetracycline (10 μg/ml). Mutant candidates were isolated, and the mutants were confirmed by PCR and DNA sequencing.

Construction of Rsp complementation plasmid
To complement the rsp mutant, the entire rsp gene was amplified from USA300 chromosomal DNA using primers listed in supplementary table (Table S1) and cloned into shuttle vector pTZN10 [17]. The plasmid was extracted from E. coli DH5α, introduced into S. aureus RN4220, then into USA300 mutants, and maintained by the addition of chloramphenicol (10 μg/ml) to the culture medium. The plasmid insert was verified by DNA sequencing.

RNA extraction and relative expression of the cnt operon genes by RT-qPCR
Cultures were inoculated from − 80°C stocks into TSB broth supplemented with the appropriate antibiotic, grown at 37°C overnight, diluted 1:100 into fresh TSB broth with the same antibiotics, and grown for 2 to 3 h to an optical density at 600 nm (OD 600 ) of ̴ 0.5. To measure cnt operon expression under iron-depleted medium, 400 μM dipyridyl (DIP) was added at an OD 600 of ̴ 0.5 with or without the supplementation of 50 μM FeSO 4. RNA Protect Bacteria Reagent (Qiagen) was added to the culture for immediate RNA stabilization prior to harvest the cells. Total S. aureus RNA was isolated using the Qiagen RNeasy mini kit (Qiagen, Valencia, CA) following the manufacturer's instructions and using lysostaphin for cell lysis. Reverse transcription followed by real-time quantitative PCR (RT-qPCR) was performed to determine the expression levels of the cnt operon genes. Comparison was made to the expression level of the housekeeping gene gmk [5]. cDNA was generated by reverse transcription using the Verso cDNA synthesis kit (Thermo Scientific) according to the manufacturer's protocol. SsoFast Eva-Green supermix and the QuantStudio 3 Real-time PCR system (Thermo Scientific) were employed for RT-qPCR using the synthesized cDNA as the template. Primers used are listed in the supplementary table (Table S1). At least three different assays with three independent cultures and RNA extractions were performed for each gene tested. Two-group comparisons of gene expression levels were evaluated by the Mann-Whitney test with the exception of the Students's t-test used for comparison of parental and fur and zur single mutants.
Expression and purification of histidine-tagged Rsp protein E. coli SG13009 (pREP4) cells harboring rsp cloned into pQE-9 vector [11] were grown in LB supplemented with ampicillin 100 μg/ml plus kanamycin 25 μg/ml at 37°C (OD 600 of ̴ 1) and then subjected to induction by 0.5 mM isopropyl-β-d-thiogalactopyranoside (IPTG) for 4 h. After centrifugation at 10,000 rpm for 30 min at 4°C the cells were resuspended in lysis buffer (50 mM NaH 2 PO 4 , pH 8, 300 mM NaCl) at 0.04 volume of the original culture. We added benzonase nuclease (Sigma-Aldrich) and a complete mini protease inhibitor cocktail tablet (Roche) to the lysate for further sonication on ice for 10 cycles of 30 s with 30 s resting periods followed by centrifugation at 10,000 rpm for 30 min at 4°C and filtering the supernatant through a Steriflip-GP, 0.22-μm, polyethersulfone membrane (EMD Millipore, Billerica, MA). Nickel affinity chromatography (GE Healthcare, Marlborough, MA) using 5 ml fractions and incremental concentrations of imidazole was performed for purification of the histidine-tagged Rsp protein. The fractions obtained were run on SDS-PAGE Coomassie gels and Western blotting (Invitrogen, Carlsbad, CA) was performed following the manufacturer's protocols. The appropriate fractions were then cleared of imidazole using PD-10 desalting columns (GE Healthcare, Marlborough, MA) and concentrated on Amicon Ultra-15 (Ultracel 50 K) centrifugal filter units (EMD Millipore). The concentration was measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific).

Electrophoretic DNA mobility shift assays
Using a LightShift chemiluminescent EMSA kit (Thermo Scientific, Wilmington, DE), the binding of Rsp to the labeled 494-bp region upstream of cntK and 381-bp region upstream of cntA was assessed. Both cntK and cntA