Identification of the Transcriptional Regulator NcrB in the Nickel Resistance Determinant of Leptospirillum ferriphilum UBK03

The nickel resistance determinant ncrABCY was identified in Leptospirillum ferriphilum UBK03. Within this operon, ncrA and ncrC encode two membrane proteins that form an efflux system, and ncrB encodes NcrB, which belongs to an uncharacterized family (DUF156) of proteins. How this determinant is regulated remains unknown. Our data indicate that expression of the nickel resistance determinant is induced by nickel. The promoter of ncrA, designated pncrA, was cloned into the promoter probe vector pPR9TT, and co-transformed with either a wild-type or mutant nickel resistance determinant. The results revealed that ncrB encoded a transcriptional regulator that could regulate the expression of ncrA, ncrB, and ncrC. A GC-rich inverted repeat sequence was identified in the promoter pncrA. Electrophoretic mobility shift assays (EMSAs) and footprinting assays showed that purified NcrB could specifically bind to the inverted repeat sequence of pncrA in vitro; this was confirmed by bacterial one-hybrid analysis. Moreover, this binding was inhibited in the presence of nickel ions. Thus, we classified NcrB as a transcriptional regulator that recognizes the inverted repeat sequence binding motif to regulate the expression of the key nickel resistance gene, ncrA.


Introduction
Metals are essential cofactors for many enzymes in bacterial cells. Nevertheless, many metals become toxic at high ion concentrations, because unlike toxic organic compounds, metals cannot be degraded or modified [1]. The nickel ion, like other metal ions, is essential for bacterial metabolism and becomes toxic at high intracellular concentrations [2]. For example, Escherichia coli can endure the presence of no more than 2 mM Ni 2+ in culture media [3]. Nickel resistance in bacteria is accomplished principally by an operon-encoded and energy-dependent specific efflux system that pumps Ni 2+ out of the cell, thereby lowering the intracellular concentration [4,5]. Several nickel-resistant bacteria have been isolated from heavy metal-contaminated locations, and their nickelresistance systems have been identified. Among these, the best characterized include CnrCBA (Co 2+ and Ni 2+ resistance) and NccCBA (Ni 2+ , Co 2+ , and Cd 2+ ) of Cupriavidus metallidurans CH34 [6,7,8], the nre and ncc determinant (Ni 2+ , Co 2+ , and Cd 2+ ) of Achromobacter xylosoxidans 31A [9,10], and the CznCBA efflux system (Co 2+ , Zn 2+ , and Ni 2+ ) of Helicobacter pylori [11]. A number of new efflux proteins have been identified; for example, RcnA of E. coli (Ni 2+ and Co 2+ resistance) [12,13], the cnr-like operon of Comamonas sp. [14], and mrdH of Pseudomonas putida (Ni 2+ and Co 2+ ) [15].
In a previous study, we identified a metal-resistant bacterium, L. ferriphilum UBK03, and cloned its nickel resistance determinant, including the ncrA, ncrB, ncrC, and ncrY genes. L. ferriphilum is a genus of iron-oxidizing bacteria which play an important role in the industrial bioleaching and biooxidation [15,16,17,18]. The ncrA and ncrC genes encode two membrane proteins that together form an efflux system [3]. NcrB is a cytoplasmic, histidine-rich, 89amino acid (aa) protein of unknown function (Pfam accession no. PF02583) [19]. It contains a conserved 85-aa domain of unknown function (DUF), DUF156, which contains two conserved cysteines and one conserved histidine residue [20]. Similarity analysis revealed that the protein was widely distributed in bacteria [21]. NcrB has been proposed to be a regulator of gene expression [22].
As we know, some nickel responsive regulators (RcnR in rcnR-rcnA efflux system from Escherichia coli [12,23], NikR from Helicobacter pylori [24,25] and Nur from Streptomyces coelicolor [26]) have been well identified and characterized. However, the protein NcrB has no apparent sequence similarity to these known transcript regulators.
In this study, we aimed to elucidate the function of NcrB. Using various approaches, we determined that NcrB binds to an inverted repeat sequence within the pncrA promoter and represses transcription of ncrA, a key gene for bacterial resistance to nickel. Furthermore, NcrB-mediated transcriptional repression was inhibited in the presence of Ni 2+ .

Induction of nickel resistance
The expression of most bacterial metal resistance systems is induced [27]. To investigate whether the nickel resistance determinant of L. ferriphilum UBK03 [3] is also inducible, the effect of nickel on E. coli NR21 growth was assessed. When non-induced E. coli NR21 was exposed to 4 mM NiCl 2 , there was a growth delay of 2 h compared with E. coli NR21 induced with 2 mM NiCl 2 , although the growth yield was unaffected (Fig. 1).
RT-PCR was conducted to confirm that the nickel resistance system is inducible. The transcription of ncrA, ncrB, and ncrC was upregulated in the presence of Ni 2+ (Fig. 2). Moreover, RT-QPCR revealed that the presence of Ni 2+ in culture medium resulted in a 10-fold increase in ncrA transcription. These data suggest that Ni 2+ induces transcription of the nickel resistance system.

Construction of promoter-lacZ fusion plasmids
Analysis of the sequence immediately upstream of ncrA and ncrB revealed the presence of two promoters (pncrA and pncrB, Fig. 3). These regions were inserted into the upstream of lacZ in pPR9TT, a low copy-number lacZ-based promoter probe plasmid [28] to construct the plasmids pPR-pncrA and pPR-pncrB. These two plasmids and pPR9TT (negative control) were transformed into E. coli JM109, respectively. No b-galactosidase activity was detected with pPR9TT in E. coli JM109 (data not shown), whereas about 9 Miller units of b-galactosidase activity were detected in with pPR-pncrA and pPR-pncrB (Fig. 4). These data indicate that pPR-pncrA and pPR-pncrB acted as the constitutive promoters in the absence of the nickel resistance genes (ncrA, ncrB, and ncrC).

Activity of the promoters with different nickel resistant genes
The promoter probe plasmids (pPR-pncrA and pPR-pncrB) and pNRABC, which contains the nickel resistance genes (ncrA, ncrB, and ncrC) [3], were co-transformed into E. coli JM109, and transformants were selected using 50 mg/mL ampicillin and 4 mM Ni 2+ . Both pncrA and pncrB were induced in the presence of Ni 2+ . The results further suggested that NcrA, NcrB, or NcrC may contain a transcriptional regulator. Then, E. coli JM109 cells were transformed with the plasmid pPR-pncrA or pPR-pncrB alone, or co-tranformed with the plasmid pNTA, pNTB, or pNTC (which contained the inserted mutations in ncrA, ncrB, or ncrC by tetracycline box insertion) [3]. Transformants were selected using 50 mg/mL ampicillin and 50 mg/mL tetracycline. Both promoters were constitutively active when co-transformed with mutated ncrB, but not with mutated ncrA or ncrC (Fig. 4). These data suggest that the NcrB protein functions as a transcriptional regulator to regulate the activity of pncrA and pncrB.

Interaction of NcrB with the promoter pncrA
The transcription start point of ncrA was localized at position 44 nt upstream of the potential start codon (ATG) of ncrA by the highresolution S1 nuclease mapping ( Figure 5D). As shown in Fig. 5D, a high GC content and inverted repeat sequence (p1p17) was identified at the downstream of the transcription start point. The possibility of a direct interaction of NcrB with the putative operator in pncrA was assessed in vitro by EMSA. The ncrB gene was ligated into the expression plasmid pET30a(+), purified and assessed by SDS-PAGE. The pncrA fragment was labeled using infrared dye-labeled M13 oligos and purified [29]. The EMSA results showed that NcrB caused a slower movement of labeled pncrA, indicating that NcrB binds to ncrA (Fig. 5A). Moreover, given the large excess of competitor DNA [poly (dI-dC)] or M13 primer in the binding mix, NcrB-pncrA binding must be specific. Binding was significantly reduced in the presence of unlabeled pncrA or the 17-bp inverted repeat. Thus, NcrB could bind pncrA at the 17-bp inverted repeat region.
The DNase I footprinting experiment was also used to determine the binding sites of NcrB in the promoter region of pncrA. As shown in Fig. 5D, a protected region from positions 24 to +25 relative to the transcription start point of ncrA was detected. Moreover, the high GC content and inverted repeat sequence was also located at the protected region. These results indicated that the protein NcrB could bind to the 17-bp inverted repeat region in vitro.

In vivo binding of NcrB to promoter regions
A bacterial one-hybrid system [30,31] was used to test NcrB binding to promoter regions in vivo. For analysis of the NcrB binding site, the sequence of the promoter region (n32p43) was divided into three overlapping fragments (n32p6, n16p22, and p7p43), as shown in Fig. 3B. Five different overlapping fragments and the inverted repeat region (Fig. 3B) from pncrA were inserted respectively into the prey plasmid pH3U3, which has two reporter genes (yeast HIS3 and URA3). These plasmids were used to transform E. coli US0hisB-pyrF-cells, and transformants were screened in medium containing 4.5 mM 5-FOA, 30 mg/mL chloramphenicol, and 20 mg/mL tetracycline. All of the transformants were able to grow on YM plates containing 4.5 mM 5-FOA, indicating that the URA3 reporter gene in the prey plasmid was not expressed. Thus, these regions of pncrA did not self-activate reporter gene expression.
The ncrB gene was then inserted into the bait plasmid pB1H1, forming pB1H1-ncrB. This bait plasmid and the prey plasmids were used to co-transform E. coli US0hisB-pyrF-cells, and transformants were screened on His-selective plates [30,31] containing different concentration of 3-AT. Co-transformed strains containing either n16p22 or the inverted repeat region (p1p17) of pncrA in the prey plasmid were able to grow on 3-AT plates (Fig. 6). However, the strains that contained the n32p6 or p7p43 region, which did not contained an intact inverted repeat, were not capable of growth on 3-AT plates. These results indicate that NcrB binds directly to the 17-bp (G+C)-rich inverted repeat sequence (59-ATCCCCCTGGGGGGGAT-39) in the p1p17 region.
The effect of Ni 2+ on the binding between NcrB and the inverted repeat region (p1p17) To test its effect on the binding between NcrB and the inverted repeat region, Ni 2+ was added to the bacterial one-hybrid system by adding 1 mM Ni 2+ to the 3-AT selective screening medium plate. As shown in Fig. 7, the strains containing the promoter (n32p43) or inverted repeat region (p1p17) in the prey plasmid grew on the 3-AT selective medium plates, but could not grow on the 3-AT plates containing 1 mM Ni 2+ . The presence of 1 mM Ni 2+ in the medium did not affect the growth of E. coli [3,9]. Thus, 1 mM Ni 2+ could disrupt the binding between NcrB and the inverted repeat region (p1p17).

Discussion
In a previous report, we identified the metal-resistant bacterium L. ferriphilum UBK03 and cloned its nickel resistance determinant, which included ncrA, ncrB, ncrC, and ncrY. NcrA contains 10 transmembrane helices and is the foundation of the nickel resistance complex [3]. NcrC, similar to NcrA, is a membrane  protein belonging to the high-affinity nickel transport protein family (Pfam accession no. PF03824) and contributes significantly to nickel resistance, possibly by chelating nickel cations in the cytoplasm [13,21,32].
NcrB is a cytoplasmic, histidine-rich, 89-aa protein of unknown function (Pfam accession no. PF02583). It belongs to the DUF156 group of proteins that are widely distributed in bacteria. Sequence analysis revealed that NcrB contains a leucine zipper domain [33] at the N-terminus. This domain is rich in positively charged amino acid residues, which may contribute to DNA recognition and binding. The C-terminal region is rich in negatively charged amino acid residues, suggesting that this region may be involved in polymerization ( Figure S1). Taken together, these features indicated that NcrB may be a transcriptional regulator. Some authors have speculated that NcrB acts as a regulator of gene expression [22]. In the present study, we identified the regulatory function of NcrB and its recognition site.
NcrB regulated the pncrA and pncrB promoters. NcrB appeared to bind to the GC-rich inverted repeat region of pncrA, which is not present in pncrB (Fig. 3A). NcrB may recognize another region of pncrB and thereby regulate the expression of ncrB and ncrC. The effect of NcrB on pncrB is currently being investigated in our laboratory. Data from a bacterial one-hybrid system indicated that NcrB binds directly to pncrA, and this binding was inhibited by 1 mM Ni 2+ . In an EMSA assay to elucidate the effect of Ni 2+ , the Ni 2+ unexpectedly formed a precipitate in the binding buffer; thus, this assay could not be carried out.
In conclusion, NcrB is a transcriptional repressor of the nickel resistance determinant in L. ferriphilum UBK03. In the presence of low concentrations of Ni 2+ , NcrB binds to the inverted repeat region of pncrA, thereby repressing its function. However, at high concentrations of Ni 2+ , the repression by NcrB is removed, pncrA becomes active, and the ncrA nickel resistance gene is expressed.

Materials and Methods
Bacterial strains and culture conditions Table 1 lists the strains and plasmids used in this study. The E. coli strains were grown aerobically in Luria-Bertani (LB) medium at 37uC with continuous shaking at 200 rpm. For selection of E. coli transformants, ampicillin and kanamycin were added to final concentrations of 100 and 50 mg/ml, respectively. As bacterial one-hybrid system selective media, His-selective (positive) NM medium and 5-FOA-selective (negative) YM medium were used as described previously [30].

RT-PCR and quantitative RT-PCR (QRT-PCR)
The effect of Ni 2+ on ncrA, ncrB, and ncrC expression was assessed by RT-PCR. Cells were grown in the presence or absence of 2 mM Ni 2+ . Total RNAs were extracted using an RNAprep pure bacteria kit (TianGen, China) according to the manufacturer's instructions, and cDNA was synthesized from 4 mg of total RNA using Moloney murine leukemia virus reverse transcriptase (TianGen, China) at 42uC for 50 min. PCR was performed for 30 cycles under the following conditions: denaturation at 94uC for 15 s, annealing at 55uC for 20 s, and extension at 72uC for 20 s. PCR products were analyzed by agarose gel electrophoresis. QRT-PCR was performed using a real-time PCR system (Bio-Rad). SYBR Green master mix (Toyoto, Japan) was used to amplify DNA under the following conditions: initial denaturation at 95uC for 3 min, followed by 40 amplification cycles (15 s at 95uC, 20 s at 55uC, and 20 s at 72uC). Melt curve data were collected using 10-s cycles (55uC for 80 cycles). Duplicate cycle threshold (CT) values were analyzed by the comparative 2 -DDCT method [34]. The relative amount of target mRNA was obtained by normalizing to an ampicillin resistance reference gene.

Bacterial one-hybrid analysis
Bacterial one-hybrid analysis was performed as described previously [30,31] with some modifications. High-efficiency electrocompetent XL1-blue E. coli cells were substituted for E. coli Trans10 (TransGen, China). The promoter ncrA (pncrA) and its deletions were amplified by PCR with primers containing NotI and EcoRI sites, and then cloned into the NotI-EcoRI sites of the reporter plasmid pH3U3. NcrB was amplified by PCR and inserted into the NotI and AvrII sites of the bait plasmid pB1H1. The constructs were verified by sequencing at the State Key Laboratory of Crop Genetic Improvement, Chinese Academy of Agricultural Sciences, Beijing, China. The two plasmids were used to co-transform the selection strain US0 by electroporation [31], and co-transformants were selected using medium that contained 3-AT, chloramphenicol (30 mg/mL), kanamycin (25 mg/mL), and tetracycline (20 mg/mL). Self-activation experiments were performed using selective medium containing 4.5 mM 5-FOA, chloramphenicol (30 mg/mL), and tetracycline (20 mg/mL).

Expression and purification of His-tagged NcrB protein
An ncrB-containing DNA fragment was amplified by PCR using plasmid pNR21as the template and primers pET-ncrB_R and pET-ncrB_F (Table S1). The PCR product was purified using a gel extraction kit (TianGen, China), digested with BamHI and HindIII, and ligated into pET-30a(+) vector. Insertion was confirmed by sequencing. For protein expression, the plasmid was used to transform E. coli BL21 (DE3) cells. Transformants were cultured in LB medium (100 mL), and IPTG (final concentration, 1 mM) was added when the A 600 of the culture reached 0.6. After incubation for 12 h at 16uC, the cells were harvested by centrifugation and washed with lysis buffer (20 mM Tris-HCl, pH 8.0; 4uC). As the N-terminus of recombinant NcrB was fused to a His 6 tag, NcrB was purified using a Ni-NTA His-bind TM resin column (Novagen, San Diego, CA) according to the manufacturer's instructions. The column   eluate was desalted with lysis buffer and stored at 220uC until required. Purified protein was assessed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE). Protein concentrations were quantified using a Bio-Rad protein assay kits II (Bio-Rad Laboratories (Beijing) Ltd. China).

Electrophoretic mobility shift assay (EMSA)
DNA fragments containing different promoter fragments were prepared by PCR using primers ProAF and ProAR (Table S1). The pncrA fragment was labeled using infrared dye-labeled M13 oligos and purified as described previously [29]. Each reaction mixture (20 mL) contained infrared dye-labeled probe and His-NcrB in buffer (10 mM Tris, pH 7.5, 50 mM NaCl, 1 mM dithiothreitol, 0.25% Tween 20, 5 mM MgCl 2 , and 1 mg of poly(dI-dC). The mixture was incubated for 20 min at 37uC in the dark, followed by electrophoresis in an 8% non-denaturing polyacrylamide gel in 0.56 TBE buffer, at 80 V for 50 min with a mini-protein electrophoresis system. The mobility positions of the labeled products in the gel were detected using the Odyssey software package (LI-COR Biosciences UK Ltd., Cambridge, UK).

S1 nuclease protection analysis
Total RNAs of NR21 induced by 4 mM NiCl 2 were extracted using an RNAprep pure bacteria kit (TianGen, China) according to the manufacturer's instructions. The S1 nuclease protection analysis were performed as described previously [35,36]. The pncrA probe was prepared by PCR using the unlabeled primer pncrAmap-F and the 59-end [c-32 P] ATP-labeled primer pncrA-map-R. A DNA sequencing ladder was generated using the same labeled primer with an fmol DNA cycle sequencing kit (Promega). The protected fragments were analyzed on a 6.0% polyacrylamide gel containing 7 M urea.

DNase I footprinting assays
In order to determine the NcrB binding sites in pncrA promoter region, DNase I footprinting assays were carried out as described previously [35,36]. The probe was prepared by labelling the 59 end of pncrA antisense stand using primers pncrA-map-F and pncrA-map-R. The primer pncrA-map-R was 32 P-labelled with T4 polynucleotide kinase before PCR. The footprinting reaction mixture contained 40 000 cpm of 32 P-labelled DNA probe, different concentrations of His-NcrB, 10 mM Tris-HCl (pH 7.5), 2 mM DTT, 0.5 mg/mL calf BSA and 5% glycerol in a total volume of 50 mL. After incubation of the mixture at 25uC for 30 min, 0.4 U DNase I (Promega) was added to the binding mixture. It was further incubated at 25uC for 70 seconds and was stopped by the addition of 50 mL stop solution (20 mM EDTA, pH 8.0) and 100 ml phenol-chloroform (1:1, v/v). DNA fragments in the aqueous phase were precipitated by adding 10 mL ammonium acetate (3 M), 2 mL glycogen (10 mg/mL) and 2.5 vol ethanol, washed with 75% ethanol, dried and directly suspended in 10 ml of 90% formamide-loading gel buffer (10 mM Tris-HCl, pH 8.0, 20 mM EDTA, pH 8.0, 0.05% bromophenol blue, 0.05% xylene cyanol). Samples were then denatured at 95uC for 10 min and 2 mL of each sample was loaded on a 6% polyacrylamide-urea gel. The sequence ladder was same to the ladder in the S1 nuclease protection analysis. After electrophoresis, the gels were dried and exposed to Kodak X-ray film.

Construction of pncrA-lacZ and pncrB-lacZ fusions
The plasmid pPR9TT was used to assess pncrA and pncrB function. The promoter regions pncrA and pncrB were amplified from pNR21 using the primers pPR-pncrA_F, pPR-pncrA_R, pPR-pncrB_R, and pPR-pncrB_F (Table S1). The PCR-amplified DNA fragments were digested with XhoI and PstI and ligated into XhoIand PstI-cleaved pPR9TT, yielding pPR-pncrA and pPR-pncrB. Correct gene insertion was confirmed by DNA sequencing, performed at the State Key Laboratory of Crop Genetic Improvement, Chinese Academy of Agricultural Sciences.
b-galactosidase assay b-Galactosidase activity was measured as described by Miller [37] and expressed in Miller units. The data presented are the results from at least three independent experiments, with a standard deviation of 10%.

Supporting Information
Table S1 Sequence of oligonucleotide primers used in this study. (DOC) Figure S1 The leucine zipper structure of NcrB. The residues below the triangle were the leucines in the leucine zipper structure. (DOC)