|Home | About | Journals | Submit | Contact Us | Français|
The pathways ensuring the efficient uptake of zinc are crucial for the ability of bacteria to multiply in the infected host. To better understand bacterial responses to zinc deficiency, we have investigated the role of the periplasmic protein ZinT in Salmonella enterica serovar Typhimurium. We have found that zinT expression is regulated by Zur and parallels that of ZnuA, the periplasmic component of the zinc transporter ZnuABC. Despite the fact that ZinT contributes to Salmonella growth in media containing little zinc, disruption of zinT does not significantly affect virulence in mice. The role of ZinT became clear using strains expressing a mutated form of ZnuA lacking a characteristic histidine-rich domain. In fact, Salmonella strains producing this modified form of ZnuA exhibited a ZinT-dependent capability to import zinc either in vitro or in infected mice, suggesting that ZinT and the histidine-rich region of ZnuA have redundant function. The hypothesis that ZinT and ZnuA cooperate in the process of zinc recruitment is supported by the observation that they form a stable binary complex in vitro. Although the presence of ZinT is not strictly required to ensure the functionality of the ZnuABC transporter, our data suggest that ZinT facilitates metal acquisition during severe zinc shortage.
Transition metals are essential constituents of a huge number of proteins where they play catalytic or structural functions (4, 50). Therefore, all organisms possess complex machineries to ensure an adequate supply of these elements, while avoiding their potentially toxic intracellular accumulation. A large number of studies have documented the relevance of metals for microbial growth and resistance to a variety of stress conditions (23, 42, 50). In particular, it is well established that the pathways enabling bacteria to recruit metal ions are key for the ability of pathogens to multiply within the host and cause disease (42, 43, 44). The vast majority of studies concerning metal uptake and bacterial pathogenicity have focused on iron, but strong evidence is emerging that the efficient uptake of other transition metals plays an important role in the host-pathogen interaction (24, 54). In particular, recent observations suggest that zinc is not freely available within the host (3). After iron, zinc is the second most abundant transition metal ion in living organisms and plays catalytic and/or structural roles in enzymes of all six classes, several of which play functions essential for cell viability (12). Investigations initially carried out in Escherichia coli and then confirmed in other microorganisms have established that zinc homeostasis is finely controlled by the coordinated activity of import and export systems regulated by Zur and ZntR, two metalloproteins able to regulate gene transcription depending on their metallation state (36, 40). Zur controls the expression of a few genes involved in bacterial response to zinc shortage, whereas ZntR regulates the expression of the zinc efflux pump ZntA. It is worth observing that, while the intracellular zinc concentration is rather constant and independent of the culture medium (close to 200 μM), both of these regulators are able to respond to femtomolar variations in the intracellular concentration of free zinc (38). These observations give emphasis to the dynamic nature of metal homeostasis and suggest that very small alterations in the intracellular zinc concentration may have a relevant influence on cellular physiology.
The metallated form of Zur also influences pathogenicity by the capability to repress the expression of the high-affinity zinc uptake system ZnuABC, a transporter of the ABC family activated in several bacteria in response to zinc deficiency (40). This transporter is constituted by three proteins: ZnuB, the membrane permease; ZnuC, the ATPase component; and ZnuA, a soluble periplasmic protein that captures Zn(II) and delivers it to ZnuB. Different studies have established that the functionality of this transporter is essential to ensure growth of microorganisms in media with little zinc and for bacterial virulence (3, 9, 16, 30, 32, 52). We have recently demonstrated that the expression of znuABC is repressed in Salmonella enterica serovar Typhimurium (hereafter referred to as S. Typhimurium) cultivated in media containing a zinc concentration as low as 1 μM, whereas its expression is strongly activated in bacteria recovered from the spleens of infected mice or from cultured epithelial or macrophage cells (3). Since the zinc concentration within eukaryotic cells is close to 0.2 mM, our studies indicate that the amount of metal effectively available for microorganisms during intracellular infections is very limited and that the ZnuABC transporter is required to ensure the efficient recruitment of zinc within the host.
In addition to the genes encoding the proteins forming the ZnuABC transporter, Zur directly regulates one or more genes encoding paralogs of ribosomal proteins (1, 39, 45). The Zur-regulated ribosomal proteins lack a zinc-binding motif that is present in their paralogs which are normally produced in zinc-replete conditions. The insertion of these proteins in ribosomes during zinc starvation likely facilitates growth by reducing the zinc requirements of bacterial cells. An additional gene (zinT) putatively regulated by Zur was identified in E. coli using a bioinformatics approach (39). zinT, which was formerly known as yodA, was originally identified as a member of the E. coli cadmium stress stimulon (21), and it was proposed that its role could be to decrease the concentration of cadmium ions in E. coli cells during cadmium stress (41). Subsequent studies of E. coli have demonstrated that zinT is modulated in bacterial cells exposed to low pH (8, 27), copper ions (29), or the transition metal chelator N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) (46) or grown in media containing very low levels of zinc (22). In addition, it has been hypothesized that ZinT could play a chaperone function by delivering zinc to other periplasmic zinc-binding proteins, such as Cu,Zn superoxide dismutase (Cu,ZnSOD) (24). This possibility, however, appears to be unlikely, as Cu,ZnSOD is strongly downregulated under conditions of zinc deficiency (2). Although the most recent investigations suggest that ZinT is involved in zinc homeostasis (22, 29, 46), the exact function of this protein has not been elucidated. Interestingly, ZinT, whose three-dimensional structure has been solved in the presence of different metal cofactors (cadmium, zinc, and nickel) (14, 15), shows a very high homology to a domain of AdcA, a component of an ABC transporter involved in zinc acquisition in Streptococcus pneumoniae (19, 39). As the N-terminal portion of AdcA is homologous to ZnuA, this observation strongly suggests that ZinT could cooperate with ZnuA in zinc uptake within the periplasmic space.
To verify this possibility, the role of ZinT has been investigated in S. Typhimurium. Our results demonstrate that ZinT has no role in cadmium resistance and that it participates in the zinc uptake process mediated by ZnuABC.
The S. Typhimurium strains used in this work are listed in Table Table1.1. Cultures were grown aerobically in liquid Luria-Bertani broth (LB) or on LB agar plates at 37°C. For growth under zinc limiting conditions, Vogel-Bonner minimal medium E (MM) (anhydrous MgSO4 [0.04 g/liter], citric acid [2 g/liter], anhydrous K2HPO4 [10 g/liter], NaH4PO4 [3.5 g/liter[, glucose [2 g/liter]), M9 minimal medium (Na2HPO4 [7.52 g/liter], KH2PO4 [3 g/liter], NH4Cl [1 g/liter], NaCl [5 g/liter], MgSO4·7H2O [1.23 g/liter], CaCl2·2H2O [0.007 g/liter], glucose [0.2%]), or Tris minimal medium (Tris-HCl [120 mM] [pH 7.2], K2HPO4 [0.017 g/liter], MgCl2 [2.03 g/liter], NH4Cl [1.06 g/liter], NaSO4 [0.44 g/liter], CaCl2 [0.06 g/liter], NaCl [4.68 g/liter], KCl [1.48 g/liter], glucose [1.98 g/liter]) was employed. For antibiotic selection, agar plates were supplemented with kanamycin (50 μg/ml) or chloramphenicol (30 μg/ml).
All Salmonella knockout mutants and the 3×FLAG epitope-tagged strains were obtained following the one-step inactivation protocol of Datsenko and Wanner (13) and the epitope tagging method described by Uzzau et al. (48), respectively. The oligonucleotides and plasmids used for construction of mutants are listed in Table Table2.2. Each new strain was confirmed by PCR with oligonucleotides annealing upstream or downstream of the mutated allele and an internal primer annealing on the inserted antibiotic resistance cassette. The oligonucleotides used for the construction and verification of each new strain are specified in Table Table1.1. The alleles were then transduced into a clean background by generalized transduction with phage P22 HT 105/1 int-201 (47). In some cases, the antibiotic resistance cassette was removed by the FLP recombinase transiently introduced by electroporation of plasmid pCP20 into the strain. The znuAΔloop mutant was obtained by the Datsenko and Wanner method (13) modified as follows. First, an antibiotic resistance cassette was inserted into the Salmonella chromosome downstream of the znuA gene by electroporating a PCR fragment obtained with oligonucleotides oli167/168 on pKD3 plasmid template. The chromosome of the resulting strain (yebA::cam; strain SA229) was then used as a template for a PCR with primers oli167/163, designed ad hoc to amplify the znuA region with a deletion in the His-rich loop (from nucleotide 411 to nucleotide 480 of the coding sequence) and the downstream antibiotic cassette. The obtained fragment was then electroporated into strain MA6926 carrying pKD46, and recombinants were selected on chloramphenicol selective plates. The deletion of the His-rich loop of znuA was confirmed by nucleic acid sequencing of the mutant strain.
To analyze the accumulation of ZinT, ZnuA, and ZnuB, aliquots of bacterial cultures (approximately 5 × 108 cells) were harvested, lysed by resuspending bacteria in sample buffer containing sodium dodecyl sulfate (SDS) and β-mercaptoethanol, and boiled for 8 min at 100°C. Subsequently, the proteins were run on 12% SDS-polyacrylamide gels and blotted onto a nitrocellulose membrane (Hybond ECL; Amersham). The epitope-flagged proteins were revealed by incubating the nitrocellulose membrane with an appropriate dilution of mouse anti-FLAG antibody (anti-FLAG M2; Sigma) and anti-mouse horseradish peroxidase-conjugated antibody (Bio-Rad), followed by the enhanced chemiluminescence reaction (ECL kit; Amersham).
Each strain was grown overnight in LB broth at 37°C and then diluted 1:500 in fresh LB broth alone or LB broth supplemented with the appropriate concentration of EDTA and/or of metals. The absorbance at 600 nm was monitored every hour for 10 h using a Perkin-Elmer Lambda 9 spectrophotometer.
Overnight cultures of bacteria were diluted in phosphate-buffered saline (PBS) buffer to a final concentration of 104 cells/ml and then mixed in pairs in a 1:1 ratio. Portions (0.2 ml) of each mixture were used to infect 10-week-old female BALB/c mice intraperitoneally. The animals were sacrificed when they exhibited symptoms of terminal septic syndrome (4 or 5 days postinfection). Bacteria recovered from spleens were plated for single colonies, and then 200 colonies were picked on selective plates. The competitive index (CI) was calculated by the formula CI = output (strain A/strain B)/inoculum (strain A/strain B). Statistical differences between outputs and inputs were determined by Student's t test.
The S. Typhimurium znuA and znuAΔloop genes were amplified from chromosomal DNA extracted with ZRfungal/bacterial DNA kit (Zymo Research) from the wild-type and SA233 strains, respectively. In both cases, the primers used were SalZnuAfor (5′-ATAGAATTCCGGGGCTCAATTCAAG-3′) and SalZnuArev (5′-TTTAAGCTTAATCTCCTTTCAGGCAGCT-3′) that amplify the coding sequences plus about 200 bp upstream of the start codon. The purified PCR products were digested with EcoRI and HindIII, ligated into the pEMBL-18 vector (17), obtaining plasmids p18PznuA and p18PznuAΔloop, which were introduced into E. coli DH5α cells. The sequences of the cloned DNA fragments were confirmed by nucleic acid sequencing.
To study the expression of recombinant proteins, cells were grown overnight in LB medium, harvested by centrifugation for 15 min at 8,000 × g, and resuspended in 500 ml of isotonic solution, and the periplasmic proteins were released by osmotic shock as already described (2). After a 20-min centrifugation at 13,000 rpm, the periplasmic proteins contained in the supernatant were applied to a Ni-nitrilotriacetic acid (Ni-NTA) column (Qiagen) preequilibrated with 50 mM sodium phosphate and 250 mM NaCl (pH 7.8) and eluted with a discontinuous gradient of 0 to 250 mM imidazole. ZnuA eluted with 20 to 40 mM imidazole. The protein was further purified by anion-exchange chromatography on a HiLoad Q Sepharose fast-performance liquid chromatography (FPLC) column (Pharmacia Biotech) preequilibrated with 20 mM Tris-HCl (pH 7.0) and eluted using a 0 to 400 mM NaCl linear gradient. The purified protein was concentrated to 30 mg/ml in a buffer containing 20 mM HEPES, 10 mM NaCl, and 5% glycerol (pH 7.0) and stored at −20°C.
Periplasmic extracts containing the ZnuAΔloop protein were initially purified by anion-exchange chromatography on a column equilibrated with 20 mM Tris-HCl (pH 7.0) and eluted using a 0 to 400 mM NaCl linear gradient. Subsequently, the protein was loaded on a cation-exchange HiLoad SP Sepharose column equilibrated with 20 mM sodium phosphate (pH 7.0) and eluted with a 0 to 400 mM NaCl linear gradient. A final chromatographic step was carried out on a HiLoad 26/10 Phenyl Sepharose HP column equilibrated with 30 mM Tris-HCl and 1.5 M (NH4)SO4 (pH 7.0). Proteins were eluted with a linear gradient of 1.5 to 0 M (NH4)SO4. The fractions containing the ZnuAΔloop protein were pooled, and the resulting protein was more than 98% pure, as judged by SDS-PAGE analysis. The purified protein was concentrated to 30 mg/ml in a buffer containing 20 mM HEPES, 10 mM NaCl, and 5% glycerol with pH 7.0 and stored at −20°C.
The S. Typhimurium zinT gene was amplified from chromosomal DNA extracted from the wild-type strain utilizing primers zinT5 (5′-TCCATGGATATTCATTTAAAAAAACTGACAATG-3′) and zinT2 (5′-ATCAAGCTTAATCAGACTTAATGATGTAGCAT-3′). The PCR product was digested with NcoI and HindIII, ligated into the pSE420 vector (Invitrogen), yielding plasmid pSEzinT, and then transformed into E. coli DH5α cells. The sequence of the whole cloned DNA fragment was verified by nucleic acid sequencing.
Cells harboring plasmid pSEzinT were grown in LB medium supplemented with 100 μg/ml ampicillin, and when the absorbance at 600 nm of the culture reached 0.5, protein expression was induced in the culture overnight with 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG). Cells were harvested by centrifugation for 15 min at 5,000 rpm, and periplasmic proteins were extracted by lysozyme treatment (2). Spheroplasts were separated from periplasmic proteins by centrifugation, and the supernatant was applied to a Ni-NTA column preequilibrated with 50 mM sodium phosphate and 250 mM NaCl (pH 7.8) and eluted with a linear gradient of 0 to 500 mM imidazole. ZinT eluted at 250 mM imidazole. Fractions containing ZinT (>98% pure) were pooled, dialyzed against a solution of 20 mM HEPES, 10 mM NaCl, and 5% glycerol (pH 7.0), concentrated to 30 mg/ml, and stored at −20°C.
To analyze the possible formation of a complex between ZnuA and ZinT, the two proteins were mixed in a 2:1 (ZnuA-ZinT) (wt/wt) ratio, corresponding to a 1.4 molar ratio, to favor the formation of complexes. After a 90-min incubation at room temperature, proteins were injected onto a HiLoad 16/60 Superdex 75 gel filtration FPLC column (Amersham Biosciences) equilibrated with 20 mM HEPES and 100 mM NaCl (pH. 7.0). Elution was carried out at room temperature.
Metal-free ZnuA and ZinT were prepared by extensive dialysis against 50 mM sodium acetate buffer and 2 mM EDTA (pH 5.5). The proteins were subsequently dialyzed twice against 50 mM sodium acetate and 0.1 M NaCl (pH 5.5) to remove excess EDTA. Finally, the proteins were dialyzed against 20 mM HEPES and 100 mM NaCl (pH 7.0) to carry out gel filtration experiments. The metal content of the apoproteins was evaluated by atomic absorption using a Perkin-Elmer spectrometer AAnalyst 300 equipped with the graphite furnace HGA-800. The zinc content of the demetallated proteins was below 5%.
A previous study has shown that ZinT is present in some bacterial species as an isolated protein, whereas in other bacteria it is a domain of the AdcA protein involved in zinc uptake (39). The amino acid alignment shown in Fig. Fig.11 shows that the C-terminal portion of AdcA from different bacteria displays high homology with ZinT (53.4% amino acid identity between the overlap region of S. pneumoniae AdcA and mature Salmonella ZinT), whereas its N-terminal portion is similar to ZnuA (26% amino acid identity between the overlap region of the S. pneumoniae protein and mature Salmonella ZnuA). The functional and structural homology between ZnuA and AdcA is confirmed by two additional features: the conservation of the three histidine residues which coordinate the zinc ion in the crystal structures of E. coli (10, 31, 53) and Synechocystis 6803 (5, 51) ZnuA and the presence in the same sequence position of a charged loop rich in acidic and histidine residues (His-rich loop), typical of proteins involved in the transport of zinc (5, 11). Interestingly, the three residues involved in zinc binding in ZinT are strictly conserved in the AdcA proteins, as well as a C-terminal histidine residue, which is likely involved in metal binding (15). However, it should be noted that ZinT differs from the C-terminal domain of AdcA in the presence of a N-terminal histidine-rich domain. The observations that in some bacterial species ZinT is fused to a ZnuA-like protein involved in zinc transport and that zinT and znuA are likely coregulated by Zur (22, 39) strongly suggest that ZinT could participate in the ZnuABC-mediated zinc uptake process. This possibility has been further corroborated by an analysis of the distribution of ZinT, ZnuA (as well as ZnuB and ZnuC), and AdcA in available bacterial genomes (see Table S1 in the supplemental material). This analysis shows that several bacteria possessing the ZnuABC system do not contain ZinT. However, bacterial species lacking a ZnuA homologue do not have a ZinT protein, whereas the occurrence of ZinT is always associated with the presence of ZnuA. These observations support the hypothesis that the role of ZinT is related to the role of ZnuA.
To verify whether ZinT is involved in cadmium resistance, we have constructed an epitope-tagged mutant by introducing a 3×FLAG sequence at the 3′ end of the chromosomal copy of the zinT coding sequence. The PP134 strain (zinT::3×FLAG) shows a growth rate similar to that of the wild-type strain (data not shown).
In line with previous investigations carried out with E. coli, ZinT accumulates in S. Typhimurium grown in LB medium supplemented with 0.5 mM cadmium acetate (Fig. (Fig.2A).2A). However, ZnuA and ZnuB, the periplasmic and transmembrane component of the ZnuABC zinc transporter, respectively, also show comparable increases in protein accumulation in bacteria cultivated in the presence of cadmium (Fig. 2B and C, respectively). This finding suggests that cadmium induces the expression of all Zur-regulated proteins participating in zinc homeostasis.
To better evaluate whether the induction of zinT plays a role in cadmium resistance, we have analyzed the ability of a zinT mutant strain to grow in LB plates containing variable amounts of cadmium. As shown in Table Table3,3, the zinT mutant strain (PP116) exhibited a cadmium susceptibility comparable to that of the wild-type strain, thus confirming recent observations showing that ZinT does not contribute to bacterial growth/survival in the presence of this toxic metal (22, 29). The reduced ability of a znuA mutant strain to grow in the presence of cadmium confirmed that this metal interferes with zinc homeostasis.
Previous studies have shown that ZnuA accumulation is induced by metal-chelating agents (EDTA and TPEN) and repressed by the addition of zinc (3) (Fig. (Fig.3A).3A). To verify whether ZinT accumulation is also modulated by zinc availability, we have grown strain PP134 in the presence of 0.5 mM EDTA with or without equimolar amounts of zinc. As shown in Fig. Fig.3,3, ZinT accumulation is strongly induced by EDTA but is repressed by zinc (Fig. (Fig.3A).3A). The inhibition of ZinT accumulation is specific for zinc, as manganese, iron, and copper have no effect (Fig. (Fig.3B).3B). To verify that zinT and znuA are coregulated by the transcriptional factor Zur, the znuA::3×FLAG and zinT::3×FLAG alleles have been transduced in a strain in which zur had been deleted, obtaining PP131 and PP132 strains, respectively. The accumulation of ZinT and ZnuA in these strains was completely deregulated and insensitive either to EDTA, zinc, or cadmium supplementation (Fig. (Fig.3A).3A). This observation confirms that in Salmonella enterica transcription of zinT is under the control of Zur, as recently observed in E. coli (22, 29).
To better analyze the relationships between ZinT/ZnuA accumulation and zinc availability, we have constructed a double epitope-tagged strain (znuA::3×FLAG zinT::3×FLAG; strain PP141). When this strain was grown in a zinc-depleted medium (MM), both ZnuA and ZinT accumulated at high levels (Fig. (Fig.4A).4A). Similar results were obtained when strain PP141 was grown in defined media of different formulation, i.e., M9 and Tris minimal media (data not shown). However, the two proteins exhibit a slightly different response to zinc availability. In fact, the results shown in the figure show that the addition of 0.5 μM ZnSO4 to the culture medium causes the complete abrogation of ZinT accumulation, but not that of ZnuA (Fig. (Fig.4A),4A), which also shows a low level of accumulation at higher zinc concentrations. In agreement with this observation, ZinT and ZnuA are maximally expressed in bacteria cultivated in LB medium supplemented with 0.5 mM EDTA, but accumulation of ZnuA is observed at EDTA concentrations lower than those required to induce accumulation of ZinT (Fig. (Fig.4B).4B). Moreover, ZinT accumulation in a strain lacking the znuA gene (PP137) is not inhibited by the addition of zinc (Fig. (Fig.4C).4C). In contrast, ZnuA accumulation in a strain lacking the zinT gene (PP128) is comparable to that observed in the wild-type strain (Fig. (Fig.4C4C).
These experiments show that znuA is induced at higher zinc concentrations than those required to activate zinT transcription, indicating that ZnuA is a protein involved in the frontline response to zinc deficiency, whereas ZinT participated in the bacterial response to more severe zinc deficiency. Moreover, the observation that ZinT accumulation cannot be repressed by the external supply of zinc suggests that ZnuA plays a role in zinc import within the cell that cannot be substituted by ZinT.
In agreement with the above reported expression studies, Table Table33 shows that growth of a zinT mutant strain on LB agar plates is inhibited by divalent metal chelators, such as EDTA and TPEN. The growth impairment due to zinT deletion is lower than that observed for a znuA mutant strain, confirming the prominent role of ZnuA in zinc homeostasis.
This observation was confirmed by the results of analysis of the growth curves of the wild-type strain and znuA and zinT mutant strains in the presence of EDTA (Fig. (Fig.5).5). In fact, whereas the growth of all the strains tested is nearly identical in standard LB medium, in the presence of 2 mM EDTA, the growth of the zinT mutant strain is impaired compared to the wild-type strain. The growth defect, however, is lower than that exhibited by the strain lacking znuA. Zinc, but not iron or manganese, supplementation to the growth medium completely abolished this phenotype (data not shown), indicating that it is specifically due to the zinc sequestration ability of EDTA and not to EDTA-induced shortage of other metals. Interestingly, the growth of the zinT znuA double mutant is comparable to that of the znuA mutant strain, confirming that ZinT cannot substitute for ZnuA in the process of zinc import within the cytoplasm.
Previous studies have established that disruption of znuA dramatically decreases Salmonella pathogenicity (3). In this work we have compared the contribution of zinT and znuA to the ability of Salmonella to colonize host tissues by carrying out competition experiments in BALB/c mice. This approach confirmed the relevance of ZnuA in Salmonella infections (Table (Table4).4). In contrast, the ability of the strain lacking zinT (strain PP116) to colonize the spleens of infected mice was comparable to that of the wild-type strain, whereas the zinT mutant (PP116) outcompeted the zinT znuA double mutant (PP118). In addition, we did not observe differences in spleen colonization between the znuA (SA123) and znuABC (SA182) mutant strains. Quite surprisingly, a zinT znuABC mutant strain (PP119) was favored compared to the znuABC mutant strain (SA182). This observation could be suggestive of a detrimental role of ZinT in the absence of ZnuABC, although more studies are required to verify this hypothesis. These experimental results suggest that ZinT facilitates zinc transport through the ZnuABC system but that it has a dispensable role during mouse infections. Moreover, the observation that the disruption of zinT does not attenuate mutant strains lacking a functional ZnuABC transporter (strain SA123 or SA182) provides further support to the hypothesis that ZinT is not involved in a ZnuABC-independent mechanism of zinc transport.
Additional experiments, however, shed more light on the role of ZinT in the mechanism of zinc uptake. The ZnuA protein possesses a charged flexible loop rich in histidines and acidic residues (His-rich loop), whose function has not yet been clarified. It has been hypothesized that the role of this loop could be to increase the ability of ZnuA to sequester zinc in environments with low concentrations of this metal (6, 18), whereas other authors have suggested that it could be a sensor of periplasmic zinc concentration (51). To investigate the role of this His-rich region, we have constructed a znuA mutant strain producing a ZnuA protein devoid of this loop (SA233). As shown in Fig. Fig.6,6, the growth of this mutant strain is not impaired in LB medium containing EDTA, indicating that ZnuA can also mediate zinc transport in the absence of the His-rich region. However, a severe growth defect was observed in a mutant strain expressing a mutated form of ZnuA and unable to produce ZinT. The growth of this mutant, in fact, was comparable to that of the strain lacking znuA. Similar results were obtained in competition experiments. The data reported in Table Table44 show that deletion of yebA, which was required for the construction of the znuAΔloop mutant strain has no effect on Salmonella virulence. Similarly, the znuAΔloop mutant strain was not significantly attenuated in comparison to the wild-type strain, although a slightly higher number of wild-type bacteria were recovered from the spleens of infected mice. In agreement, there was not a significant difference in the spleen colonization ability of a zinT mutant strain (PP125) and of the znuAΔloop strain (SA233). However, the Salmonella strain lacking zinT and expressing the mutated form of znuA (PP130) was attenuated with respect to the strain unable to produce ZinT (PP125). This strain maintained some ability to transport zinc through the ZnuB channel, so it had an advantage over strain PP118, which does not express znuA.
These experiments suggest that ZinT and the His-rich domain of ZnuA are two independent elements which participate in ZnuABC-mediated zinc transport by playing an overlapping role in facilitating zinc recruitment by ZnuA. Although the simultaneous presence of ZinT and the His-rich domain is apparently not indispensable to ensure the functionality of the transporter under the conditions investigated in this work, disruption of each one of the two elements disclose a role for the other element in enhancing the efficiency of zinc uptake.
The above experiments showing that ZnuA and ZinT functionally interact in the mechanism of ZnuABC-mediated zinc import suggest that ZinT might physically interact with ZnuA. To prove this possibility, we have analyzed the abilities of these two proteins to form a complex in vitro. ZinT and ZnuA were cloned, expressed, and purified as described in Materials and Methods. The two proteins were mixed together, incubated for 90 min at room temperature, and then loaded on a gel filtration column. Figure Figure77 shows a comparison of elution of the individual proteins and of the ZnuA-ZinT mixture. ZinT eluted from the column with an apparent molecular mass of 22.3 kDa (peak centered at 75 ml [Fig. [Fig.7A]),7A]), in excellent agreement with the ZinT molecular mass deduced from the amino acid sequence (22.2 kDa). ZnuA eluted with an apparent molecular mass of 34.9 kDa (peak centered at 67 ml [Fig. [Fig.7B]),7B]), a value which is slightly higher than the expected molecular mass (31.5 kDa). When the mixture of ZinT and ZnuA was applied to the column, the elution profiles of the two proteins changed significantly. In fact, in this case maximal ZinT and ZnuA concentration was found in fraction 63 (Fig. (Fig.7C),7C), corresponding to a protein of approximately 48 kDa. The observation that the apparent molecular mass of the two proteins was shifted to values significantly higher than those of the single proteins strongly supports the hypothesis of formation of a binary complex between the two proteins. The elution profile of the complex was rather broad, possibly due to incomplete involvement of the two proteins in complex formation and/or to partial complex dissociation during the elution. However, when the proteins recovered from fraction 63 were concentrated and subjected to a new gel filtration chromatography, the largest part of ZnuA and ZinT still coeluted with a high molecular mass, indicating that the complex between the two proteins is stable (data not shown). The observation that the ZnuA/ZinT complex has an apparent molecular mass lower than that expected (53.7 kDa) is indicative of changes in the hydrodynamic properties of the two proteins following their interaction.
Interestingly, the two apoproteins were not able to form a stable complex (Fig. (Fig.7D).7D). However, the ability of ZnuA and ZinT to stably interact was restored when the two proteins were individually reconstituted with an equimolar amount of zinc before protein incubation (Fig. (Fig.7E).7E). The addition of zinc to a single protein (ZnuA or ZinT) was not sufficient to observe the coelution of the two proteins from the gel filtration column (data not shown). These results suggest that the binding of zinc induces structural rearrangements in ZnuA and ZinT which are necessary for the formation of a stable complex between the two proteins.
To determine whether the histidine-rich region of ZnuA is involved in the formation of a complex between ZinT and ZnuA, similar experiments were carried out using the ZnuAΔloop mutant protein. As shown in Fig. Fig.7G,7G, when these proteins were coincubated, they eluted in the same fractions with apparent molecular masses higher than those of isolated ZnuAΔloop and ZinT proteins. In line with functional studies, this observation demonstrates that the His-rich region of ZnuA is not required for the formation of a ZinT/ZnuA complex in vitro.
The elution profile of ZinT was not altered when the protein was preincubated with bovine serum albumin or Cu,Zn superoxide dismutase (Fig. 7H and I, respectively), thus indicating that the interaction of ZinT and ZnuA is highly specific.
Zinc serves important functions in a wide range of cellular processes; therefore, the mechanisms ensuring highly efficient zinc recruitment are critical for bacterial growth and survival in all the environments where this metal is not abundant. Recent studies have established that zinc is not freely available within host tissues and that the ability of several pathogens to multiply in the infected host is strictly dependent on the zinc transporter ZnuABC. As this is the only high-affinity zinc transporter identified in many bacterial species, these observations suggest that ZnuABC could be an interesting target for novel antimicrobial strategies. To deepen our understanding of the mechanisms governing zinc homeostasis in bacteria, we have investigated the role of ZinT, a poorly characterized protein which has been proposed to be involved in the mechanisms of resistance to toxic metals or to zinc deficiency.
In agreement with recent observations (22, 29), this investigation demonstrates that ZinT has no role in bacterial resistance to cadmium toxicity. In fact, although zinT is strongly induced by this metal, deletion of the gene does not impair bacterial growth in cadmium-containing media. Moreover, cadmium induces the accumulation of ZinT as well as of other proteins involved in zinc transport, i.e., ZnuA and ZnuB. The mechanisms responsible for cadmium toxicity are not completely understood, but they can be largely explained by the ability of cadmium to deplete cells of intracellular glutathione, to react with the sulfhydryl groups of proteins, and to compete with other metals for binding to metalloproteins (7, 49). These results provide novel suggestions to understand the complex molecular basis of cadmium toxicity in bacteria. In fact, cadmium induces the expression of the Zur-regulated genes in bacteria growing in a zinc-replete medium (LB), suggesting that cadmium alters zinc homeostasis in bacteria. Additional studies are required to understand whether this is due to a general ability of cadmium to substitute for the proper metal cofactor in zinc-containing proteins or to a specific effect on Zur (which, upon cadmium binding, might adopt an altered conformation unable to bind to DNA) or on the functionality of the ZnuABC transporter. However, it is worth noting that cadmium binds with high affinity to metal sites characterized by cysteine ligands, such as metallothioneins (26) and zinc-dependent transcription factors (25) and that X-ray absorption studies have shown that zinc binding to Zur involves different cysteine residues (37). This last observation suggests that Zur could be a privileged target for cadmium ions.
All the results described in this work converge in demonstrating that ZinT participates in the process of zinc uptake mediated by ZnuABC. This conclusion is supported by different pieces of evidence. First of all, ZinT can be found, either as an independent protein or as a domain of the AdcA proteins, in bacteria containing ZnuA or homologous periplasmic ligand-binding proteins involved in zinc uptake. In contrast, ZinT is not present in bacteria lacking ZnuA (see Table S1 in the supplemental material). Moreover, in line with recent studies carried out in E. coli (22, 29), we have confirmed the original proposal of Panina and coworkers (39) that zinT is a member of the Zur regulon. We have also demonstrated that zinT is induced under conditions of zinc deficiency (Fig. (Fig.33 and and4)4) and that it contributes to bacterial growth in media with low concentrations of this metal (Fig. (Fig.55 and and6).6). Interestingly, both the analysis of the role of zinT in bacterial growth in vitro (Fig. (Fig.6)6) and the competition experiments in mice (Table (Table4)4) failed to identify a major contribution of ZinT to zinc transport in bacteria lacking znuA or the entire znuABC operon, indicating that the role of ZinT in zinc uptake is dependent on the presence of ZnuA. It is noteworthy that a mutant strain expressing zinT but with the znuABC operon deleted was disadvantaged compared to the zinT znuABC mutant strain (Table (Table4),4), suggesting that in the absence of the ZnuABC transporter, the expression of ZinT is harmful, possibly due to the zinc sequestration ability of this protein, which might decrease zinc import through low-affinity zinc transporters. This possibility is supported by the observation that ZinT accumulates constitutively in bacteria lacking znuA (Fig. (Fig.4C).4C). However, further work is required to verify this hypothesis because the same effect was not observed in the competition between the znuA (SA123) and znuA zinT (PP118) mutant strains. All these observations suggest that ZinT is an additional component of the ZnuABC transporter facilitating zinc recruitment in the periplasmic space. The results reported in Fig. Fig.77 demonstrate that ZinT is also able to form a stable complex with ZnuA in vitro, which we suppose could resemble the structural organization of AdcA proteins. Interestingly, the stability of the complex is dependent on the presence of zinc, suggesting that in vivo the interaction between ZnuA and ZinT could be destabilized upon zinc exchange within the ZnuA/ZinT complex or following zinc release from the complex to ZnuB. The scattered distribution of zinT in eubacteria expressing znuA, the observation that ZnuABC is critical for successful infection in several bacteria lacking zinT, including Haemophilus ducreyi (30), Brucella abortus (52), and Campylobacter jejuni (16), and the results showing that deletion of zinT does not attenuate S. Typhimurium and only marginally decreases bacterial growth in the presence of very high concentrations of chelating agents all together indicate that ZinT has a role in zinc transport which is significantly less important than that of ZnuA. In support of this view, we have observed slight differences in the regulation of zinT and znuA, suggesting that transcription of znuA occurs at zinc concentrations higher than those required to activate zinT expression. In fact, ZinT accumulation is completely repressed in bacteria growing in a minimal medium containing 0.5 μM ZnSO4 or in LB medium supplemented with 0.05 mM EDTA, whereas under the same conditions ZnuA is partially induced (Fig. (Fig.4).4). Such a flexible response to zinc deficiency, likely justified by small differences in the Zur-binding region (data not shown), may provide an explanation for the production of two separate proteins, instead of a single one (AdcA), as in some Gram-positive bacteria. It should be noted that this finding is in apparent contrast with previous transcriptomic studies showing a greater increase in zinT mRNA levels than in znuA mRNA levels in response to TPEN (46) or zinc deficiency (22). However, both these studies were carried out using defined media containing low levels of zinc and Graham and coworkers (22) used a medium containing high concentrations of EDTA. We believe that the media used in these works provided conditions sufficient to induce a significant basal expression of znuA, thus explaining the much greater mRNA induction of zinT upon further depletion of zinc.
Although dispensable for the functionality of ZnuABC, ZinT enhances the ability of bacteria to grow under severe zinc shortage. The presence of another periplasmic protein distinct from ZnuA may be useful to facilitate metal sequestration in this cellular compartment. We have observed that ZinT has a very high affinity for immobilized nickel (see Materials and Methods) possibly due to its N-terminal His-rich domain (Fig. (Fig.1).1). This region of ZinT differs from the ZnuA His-rich loop, as the last motif is characterized by the presence of a very high number of acidic residues, which likely affects the protonation state of the neighboring histidine residues. We suggest that the His-rich domain of ZinT could play a role in facilitating zinc displacement from other proteins and the subsequent entry of the metal into the ZnuABC-mediated process of zinc transport from the periplasm to the cytoplasm.
Some experiments reported in this study also shed new light on the role of the His-rich region of ZnuA. ZnuA possesses a central domain of variable length in different bacterial species which is characterized by the presence of a high number of histidine and acidic residues whose function is not yet known. A few studies have suggested that this loop could enhance zinc binding ability and its subsequent transfer to the primary binding site of ZnuA (6, 18). Alternatively, it has been suggested that it could be a sensor of periplasmic zinc concentration, able to inhibit zinc transfer from ZnuA to ZnuB at high zinc concentrations (51). Interestingly, similar domains are also present in a vast number of eukaryotic zinc transporters (20). Different studies have explored the role of these histidine-rich regions by producing mutant proteins devoid of this protein domain. The results obtained have failed to provide an unequivocal answer to the role of His-rich loops, as it has been proposed that they can contribute to protein stability (33), the process of metal transfer (34), metal selectivity (35), and modulation of protein activity (28). We have observed that an S. Typhimurium strain expressing a ZnuA variant devoid of the His-rich region is not attenuated in infection studies and grows as the wild-type strain does in LB medium containing 2 mM EDTA. This observation apparently argues against a role of this domain in metal recruitment. However, when this mutation was inserted in a strain in which zinT had been deleted, mutant ZnuA proved not to be able to work as well as native ZnuA either in vitro (Fig. (Fig.6)6) or in the infected animal (Table (Table4).4). Although we cannot exclude the possibility that the His-rich region could play additional roles (for example, in facilitating the selection of the correct metal ion), our findings suggest that ZinT and the His-rich region play overlapping roles in increasing the metal binding ability of ZnuA. Under the experimental conditions we have explored in this work, this function can be disclosed only in bacteria lacking both ZinT and the His-rich region of ZnuA, suggesting that they play similar roles. It is likely that the simultaneous presence of two structurally distinct elements with apparently redundant roles could enhance metal recruitment during conditions of severe zinc shortage and maximize zinc import under variable environmental conditions.
This study, which shows that ZinT participates in the ZnuABC-mediated process of zinc transport, provides additional insights into the mechanisms employed by some bacteria to obtain zinc. ZinT is dispensable for the functionality of the transporter, thus explaining its absence in several bacteria, but plays a role auxiliary to that of ZnuA in the recruitment of zinc within the periplasmic space.
This work was partially supported by an Istituto Superiore di Sanità (ISS) grant to A.B. and P.P. and by a grant from the Fondazione Roma to A.B.
Published ahead of print on 22 January 2010.
†Supplemental material for this article may be found at http://jb.asm.org/.