Analysis of the Manganese and MntR Regulon in Corynebacterium diphtheriae

ABSTRACT

Corynebacterium diphtheriae is the causative agent of a severe respiratory disease in humans. The bacterial systems required for infection are poorly understood, but the acquisition of metals such as manganese (Mn) is likely critical for host colonization. MntR is an Mn-dependent transcriptional regulator in C. diphtheriae that represses the expression of the mntABCD genes, which encode a putative ABC metal transporter. However, other targets of Mn and MntR regulation in C. diphtheriae have not been identified. In this study, we use comparisons between the gene expression profiles of wild-type C. diphtheriae strain 1737 grown without or with Mn supplementation and comparisons of gene expression between the wild type and an mntR deletion mutant to characterize the C. diphtheriae Mn and MntR regulon. MntR was observed to both repress and induce various target genes in an Mn-dependent manner. Genes induced by MntR include the Mn-superoxide dismutase, sodA, and the putative ABC transporter locus, iutABCD. DNA binding studies showed that MntR interacts with the promoter regions for several genes identified in the expression study, and a 17-bp consensus MntR DNA binding site was identified. We found that an mntR mutant displayed increased sensitivity to Mn and cadmium that could be alleviated by the additional deletion of the mntABCD transport locus, providing evidence that the MntABCD transporter functions as an Mn uptake system in C. diphtheriae. The findings in this study further our understanding of metal uptake systems and global metal regulatory networks in this important human pathogen.

IMPORTANCE Mechanisms for metal scavenging are critical to the survival and success of bacterial pathogens, including Corynebacterium diphtheriae. Metal import systems in pathogenic bacteria have been studied as possible vaccine components due to high conservation, critical functionality, and surface localization. In this study, we expand our understanding of the genes controlled by the global manganese regulator, MntR. We determined a role for the MntABCD transporter in manganese import using evidence from manganese and cadmium toxicity assays. Understanding the nutritional requirements of C. diphtheriae and the tools used to acquire essential metals will aid in the development of future vaccines.

INTRODUCTION

Corynebacterium diphtheriae is the causative agent of the severe respiratory disease diphtheria, which is mediated primarily through the activity of the secreted exotoxin diphtheria toxin (1). Although the requirements for C. diphtheriae to infect the human host are poorly understood, nutrients such as manganese (Mn) are likely critical for survival and colonization. Mn has roles in oxidative stress tolerance and nucleotide metabolism (2, 3). Availability of Mn in the host is restricted by the protein calprotectin, which sequesters free Mn during inflammation (4, 5), and by NRAMP1 (natural resistance-associated macrophage protein 1), which depletes divalent cations from macrophage phagolysosomes (6).

Bacterial pathogens, such as C. diphtheriae, must be able to compete with host mechanisms for Mn sequestration (3, 7). While C. diphtheriae encodes a putative Mn ABC transporter, MntABCD (MntA-D), Mn transport through this system has not been demonstrated, and a phenotype was not described for a C. diphtheriae mutant lacking mntABCD (mntA-D) (8). Studies in other Gram-positive bacteria have demonstrated that Mn acquisition occurs largely through Mn-specific ABC transporters, similar to C. diphtheriae MntA-D, and/or through NRAMP family transporters, usually named MntH, which share similarity to the eukaryotic macrophage protein, NRAMP1 (3). Functional characterization of Mn import, both in vitro and in animal models of infection, has been described in several species, including Bacillus subtilis (9), Bacillus anthracis (10), Streptococcus mutans (11), and Staphylococcus aureus (12–14). In addition to Mn import, several pathogens, including Streptococcus pyogenes (15), Neisseria meningitidis (16), and S. aureus (17), encode mechanisms for Mn efflux to protect the bacteria from toxic levels of Mn (18). To coordinate the expression of Mn import and export, bacteria encode transcriptional regulators, such as the C. diphtheriae Mn transport regulator (MntR), which control the expression of target genes in response to intracellular levels of Mn. Homologs of MntR are widely distributed in bacteria (3) and have been studied in both human pathogens such as Mycobacterium tuberculosis (19), S. aureus (17, 20), and Streptococcus spp. (11, 21, 22), and model bacteria including B. subtilis (23) and Corynebacterium glutamicum (24). In these bacteria, the targets of MntR regulation have been defined; however, limited information is available regarding the MntR regulon in C. diphtheriae.

C. diphtheriae MntR was first identified through homology to the diphtheria toxin repressor (DtxR), the global regulator for iron (Fe) homeostasis in C. diphtheriae (8). MntR has a function similar to that of DtxR, except MntR recognizes Mn instead of Fe to regulate gene expression. Under Mn-replete conditions, Mn-bound MntR binds to a sequence upstream of the C. diphtheriae mntA gene to repress transcription (8). Under Mn-limited conditions, apo-MntR is inactive and fails to block transcription of mntA. The C. diphtheriae mntR gene is the terminal gene in the mntA-R locus, and whether this locus constitutes a single operon or if there are multiple promoters driving transcription is not known (8). DNA binding studies with MntR at the mntA promoter have shown that Mn and various other divalent cations can activate MntR binding to DNA in vitro. DNase I footprinting experiments showed that MntR protected a 73-bp region of the mntA promoter, which was significantly larger than the 30-bp region protected by DtxR at various Fe-regulated promoters in C. diphtheriae. Furthermore, C. diphtheriae strains lacking mntR or mntA grew similarly to the wild-type strain in a metal-depleted medium, suggesting that additional Mn importers are encoded by the C. diphtheriae genome (8).

In this study, we used gene expression arrays to identify new targets of Mn and MntR regulation in C. diphtheriae. We found that MntR regulates 14 genes, located at seven distinct loci, in response to Mn availability. The genes identified include the previously studied mntA-D (8) and iutABCD (iutA-D) (25) operons, which are both predicted to encode metal-specific ABC transporters. While mntA-D was repressed in the presence of Mn, iutA-D was weakly induced by Mn. The genes encoding a predicted Mn-dependent superoxide dismutase, sodA, and siderophore biosynthesis protein, ciuE (26), were also induced in the presence of Mn in an MntR-dependent manner. Mobility shift assays were used to identify five MntR binding sites upstream of Mn/MntR-regulated promoters. Growth of an mntR deletion mutant was inhibited by high levels of Mn, which could be reversed by a deletion of mntA, suggesting that the mntA-D transporter is involved in Mn uptake and is responsible for the Mn toxicity observed in the mntR mutant. Growth studies that examined the toxic effects of cadmium (Cd) provided additional supportive evidence of metal transport by the mntA-D system. This study expands upon our understanding of the C. diphtheriae Mn/MntR regulon and suggests a role for MntA-D as a high-affinity Mn importer in C. diphtheriae.

RESULTS

Identification of Mn/MntR-regulated genes.

To identify C. diphtheriae genes regulated by MntR in response to available Mn, we grew wild-type C. diphtheriae strain 1737 and an isogenic mntR deletion mutant in semidefined metal-limited medium (mPGT) without and with MnCl₂ supplementation at 5 μM (see Fig. S1 in the supplemental material). This level of Mn supplementation was shown to fully repress the Mn- and MntR-regulated C. diphtheriae mntA promoter (8). C. diphtheriae growth profiles were similar with and without Mn supplementation, and RNA used for the gene expression arrays was harvested during logarithmic growth (Fig. S1). Data from gene expression arrays were analyzed as described previously (27).

We identified a total of 14 genes regulated by MntR in response to Mn availability (Table 1). These include genes that encode two putative ABC transporter loci (iutA-D [25] and mntA-D [8]), ribonucleoside reductase proteins (nrdI and nrdF2), a siderophore biosynthesis protein (ciuE) (26), and an Mn-dependent superoxide dismutase (sodA). Notably, expression of iutA-D, ciuE, and sodA was induced by Mn supplementation and was dependent on MntR function, while the remainder of the identified genes were repressed. The array was validated through quantitative PCR (qPCR) of genes that were either repressed or induced by Mn, including piuB, iutA, mntA, nrdI, and sodA (Fig. 1) (R² = 0.9691). To further support the gene expression array results, we cloned mntR into the shuttle vector pKN2.6Z to generate the complementing clone pKNmntR. The pKNmntR plasmid in the ΔmntR strain restored Mn-dependent regulation for piuB, mntA, nrdI, and sodA (Fig. S2). Although ΔC_q values were not statistically different for iutA, a weak increase in expression was observed for the complementing clone in the ΔmntR strain.

TABLE 1.

Genes regulated in response to manganese by MntR^c

Locus	Gene	Protein name	WT ± Mna	ΔmntR:WTb
DIP0124	piuB	PepSY domain-containing protein	−4.79	5.70
DIP0169	iutA	Metal ABC transporter substrate-binding protein	1.37	−1.53
DIP0170	iutB	Metal ABC transporter ATPase	1.54	−1.47
DIP0171	iutC	Metal ABC transporter permease	1.45	−1.50
DIP0172	iutD	Metal ABC transporter permease	1.16	−1.15
DIP0586	ciuE	Siderophore biosynthesis	1.08	−2.06
DIP0615	mntA	Mn ABC transporter substrate-binding protein	−7.16	7.22
DIP0616	mntB	Mn ABC transporter ATPase	−6.44	6.72
DIP0617	mntC	Mn ABC transporter permease	−6.38	6.59
DIP0618	mntD	Mn ABC transporter permease	−3.79	5.17
DIP1923	nrdI	Class Ib ribonucleoside-diphosphate reductase assembly flavoprotein	−7.32	8.15
DIP1924	nrdF2	Ribonucleoside-diphosphate reductase beta chain 2	−6.20	7.01
DIP2261	sodA	Manganese superoxide dismutase	2.04	−2.06
DIP2262		NAD(P)H-dependent oxidoreductase	−2.22	1.95

^aBoldface (negative values) indicates lower expression in +Mn samples (i.e. repressed by Mn). Italics (positive values) indicate increased expression in +Mn samples (i.e. induced by Mn).

^bBoldface (negative values) indicate lower expression in the ΔmntR samples; italics (positive values) indicate increased expression in the ΔmntR samples.

^cLog₂ fold change indicated for all genes. Divided sections indicate separate gene loci.

FIG 1.

Gene expression array and qPCR correlation. Expression of the indicated genes was probed by qPCR for comparison to the microarray. The comparison of relative expression of genes from wild-type C. diphtheriae strain 1737 grown without and with 5 μM MnCl₂ supplementation is shown. Data are the means and standard deviations from the three biological replicates used in the microarray analysis (R² = 0.9691).

Expression of mntR is independent of mntA-D.

A previous report identified promoter elements and an MntR binding site upstream of the mntA gene that were required for the Mn- and MntR-dependent regulation (8). The proximity of mntA-D and mntR suggests that these genes constitute an operon under the control of the single promoter upstream of mntA (Fig. 2A). However, the microarray results indicate that while the mntA-D genes were strongly repressed by Mn in an MntR-dependent manner, expression of mntR was not affected by Mn supplementation (Table S1). To confirm this observation, qPCR was used to probe expression of mntA, mntB, mntD, and mntR in both the wild-type and ΔmntR strains (Fig. 2B). The qPCR results support the finding from the gene expression array and show that genes encoding the putative ABC metal transporter (mntA, mntB, and mntD) are similarly regulated by Mn and MntR, while mntR expression is not affected by Mn supplementation. Despite a short, 14-bp intergenic region between mntD and mntR, mntR is not cotranscribed as part of the mntA-D operon. These results support a recent study that mapped the mntR transcriptional start site 57 bp upstream of the start codon, which is within the mntD coding region (28).

FIG 2.

Further analysis of the mntABCD-R locus. (A) Genetic map of the mntABCD-mntR gene locus with predicted gene products described. The mntR gene is located 14 bp downstream of mntD. The location of transcription start sites (TSS) and direction of transcription are indicated. (B) qPCR was used to measure the relative expression of mntA, mntB, and mntD in the wild type (WT) and ΔmntR mutant in response to Mn supplementation; mntR expression was examined only in the wild type. Data were analyzed using the ΔΔC_q method using gyrB for normalization compared against the wild type with empty vector in the absence of Mn supplementation. *, P < 0.0001 by 2-way ANOVA Holm-Sidak’s multiple comparisons test comparing the wild type with added Mn to that with no added Mn (n = 3).

MntR binds to a conserved DNA motif.

The MntR binding site at the promoter region of mntA was previously characterized through mobility shift assays and DNase I footprinting (8). The footprinting studies showed that MntR protected an approximately 73-bp region, and it was proposed that multiple MntR dimers are involved in the binding to the mntA upstream region. Putative promoter regions (200 bp) for the seven loci identified in Table 1 were amplified by PCR and tested for binding to recombinant MntR using mobility shift assays (Fig. 3A). The mntA promoter region was used as a positive control, and a fragment of the mntR coding region served as a negative control for the binding studies. In the presence of Mn, mobility shifts were observed for the regions upstream of piuB, iutA, nrdI, and dip2262 as well as the mntA positive control. DNA fragments encompassing the promoter regions for ciuE and sodA did not shift in the presence of Mn. To test if MntR binds to these promoter regions in its apo form, EDTA was added to the binding reactions instead of Mn. The addition of EDTA eliminated the binding of MntR to the mntA promoter but did not result in binding of MntR at the promoters for ciuE or sodA, suggesting that MntR does not bind to these upstream regions under the conditions tested. Since the sodA and ciuE genes were found to be induced in the presence of Mn in an MntR-dependent manner, MntR may indirectly regulate the Mn-dependent expression of these genes or bind further upstream.

FIG 3.

MntR binds promoters of target genes. (A) Electrophoretic mobility shift assays were performed with recombinant His-tagged MntR and PCR-amplified putative promoter regions for the indicated genes. MntR was either excluded (−) or added (+) to the binding reaction prior to separation by electrophoresis. Binding reactions shown were performed with MnCl₂ (top) or EDTA (bottom) as noted. (B) Binding reactions were performed using complementary annealed nucleotides encompassing predicted binding sites identified using MEME (29), indicated below, with the most highly conserved bases in boldface. The black arrow head indicates migration of the double-stranded annealed fragment. The gray arrow head indicates migration of biotinylated single-stranded oligonucleotides. Two potential overlapping binding sites were found in the iutA promoter region; key bases for one are in boldface and the others are underlined. (C) The MntR binding consensus sequence based upon sequences that showed binding to MntR. *, iutA sequence was not used to establish the binding consensus.

The promoter regions that bound to MntR were subjected to analysis by the Multiple Em for Motif Elicitation (MEME) algorithm (29). MEME identified a potential 17-bp consensus site in the promoter sequences of mntA, piuB, iutA, nrdI, and dip2262 (Fig. 3B and C); no binding sites were identified in the ciuE and sodA promoter regions. This binding region aligned with a sequence found to be important for Mn-dependent regulation for mntA in the previous study (8) and shares similarity to the MntR binding site found in C. glutamicum (24). We tested the binding of MntR to double-stranded oligonucleotides containing the consensus binding site and surrounding sequence and found that MntR bound the sites for mntA, piuB, nrdI, and dip2262 (Fig. 3B). We did not observe binding for iutA, and it is unclear where MntR binds in the iutA upstream region. Two putative MntR binding sites identified by MEME in the iutA promoter region lacked several of the bases that are conserved in the confirmed binding sites (Fig. 3B and Fig. 4A), and it is possible that failure of MntR to bind to the oligonucleotide for the putative iutA binding site indicates that additional flanking sequences are required for MntR binding. It is also possible that MntR interacts with DNA differently at promoters that are induced by Mn and MntR. In general, binding to the oligonucleotides was not as strong as the observed binding to the larger fragments, which may have contributed to the lack of observed binding of MntR to the iutA promoter region. We submitted the four confirmed binding sequences to MEME and searched the C. diphtheriae NCTC 13129 genome for possible additional MntR binding sites using Find Individual Motif Occurrences (FIMO) (30). Using a stringent search, nine total sites were identified, of which five were the sequences tested; the remaining four possible binding sites occur within coding regions and are unlikely to impact target gene expression (Table S3).

FIG 4.

Promoter fusions for iutA, sodA, and ciuE. (A) Notable elements in the iutA promoter region are indicated, including the DtxR and Zur binding sites that were identified previously (25) and promoter elements (−35, −10, and +1). The region containing two independent, overlapping motifs with similarity to the MntR binding consensus is noted below; MntR binding to this region could not be confirmed using DNA oligonucleotides (see Fig. 5B). (B) Promoter fusions for iutA, sodA, and ciuE were tested for activity in mPGT without or with MnCl₂ supplementation in wild-type C. diphtheriae 1737 and the ΔmntR mutant. The data are the means and standard deviations from four biological replicates. Statistical significance between the corresponding with and without Mn samples determined by multiple paired t tests: *, P < 0.05; **, P < 0.01.

Confirmation of MntR-dependent gene induction using translational fusions.

We observed induction of iutA, ciuE, and sodA in our gene expression data; however, MntR binding was not observed with promoter regions for ciuE and sodA, and the precise binding site for iutA could not be confirmed by sequence analysis or through binding of MntR to oligonucleotides. To further test how Mn and MntR affect iutA, sodA, and ciuE expression, we used beta-galactosidase promoter fusions for each gene. The complete intergenic regions of iutA (173 bp) and sodA (153 bp) were cloned upstream of the lacZ gene in our reporter plasmid, pPSZ, and a short 75-bp putative promoter region for ciuE was tested, as the intergenic region between ciuE and the upstream gene, ciuD, is only 22 bp. Consistent with our gene expression data, we observed a weak but significant increase in promoter activity in response to Mn supplementation for all three promoters in the wild-type strain (Fig. 4B). Induction was not observed when the plasmids were tested in the ΔmntR mutant. Although the mechanism remains unclear, induction of gene expression by Mn and MntR in C. diphtheriae has not been previously established. Further study will be needed to assess whether this induction is an outcome of direct MntR-promoter binding in vivo or an indirect effect of changes in Mn homeostasis.

The role of MntA-D in Mn import.

From our gene expression studies, we identified four gene loci that were repressed by MntR in response to Mn supplementation. Of the loci identified, the products of mntA-D and piuB are potential candidates for Mn importers. The mntA-D genes encode a putative ABC metal transporter that includes a substrate binding protein (MntA), while piuB encodes a hypothetical membrane protein of unknown function. To determine if the products of mntA-D and piuB are involved in Mn transport, we generated several mutants lacking some or all of the mntA-R loci (Fig. 5A) and deleted the piuB gene in both the wild-type strain and in the mntA-D mutant background. We tested the effect of Mn availability on the growth of wild-type C. diphtheriae 1737 and the various mutants in metal-depleted mPGT medium (Fig. 5B). No significant growth differences were observed between the mntA-D mutant and wild type, regardless of Mn supplementation. The deletion of either mntA alone or mntA-D resulted in growth similar to that of the wild type under all Mn levels tested (Fig. 5B), demonstrating that the MntA-D system is not required for wild-type growth under the conditions examined. However, the ΔmntR strain grew very poorly at 100 μM Mn, while growth of the other strains was only minimally affected at this high Mn level. The presence of the cloned mntR gene on plasmid pKNmntR restored wild-type levels of growth to the ΔmntR strain in 100 μM Mn (Fig. 5C). The piuB mutant showed growth similar to that of the wild type except at 1 μM supplementation. However, a piuB mntA-D double mutant was not significantly different from the wild type, suggesting that PiuB does not function in Mn import, and its function in C. diphtheriae remains unknown.

FIG 5.

Growth of the C. diphtheriae wild type and mutant derivatives in Mn. (A) Line diagram indicating the deletions introduced into C. diphtheriae for the mntA-R locus. Dashed lines indicate the region of DNA removed in each respective mutant. (B) Overnight growth of strains in mPGT with MnCl₂ supplementation as indicated, measured by OD₆₀₀ (n = 4). (C) Growth of the wild-type strain carrying empty vector (pKN2.6Z) and the ΔmntR strain with either empty vector or mntR clone (pKNmntR) in mPGT with MnCl₂ supplementation as indicated. Data are the means and standard deviations from biological replicates (n = 3). Statistical significance comparing the corresponding wild-type growth and indicated strain (by color) determined by Holm-Sidak’s multiple-comparison test: *, P < 0.05; **, P < 0.01.

Sensitivity of the mntR mutant to high Mn may be caused by unregulated Mn import and accumulation of toxic Mn levels. This provided an opportunity to assess whether MntA-D has a role in import; if MntA-D is a Mn transporter, then the deletion of mntA-D in addition to mntR may suppress the mntR phenotype. Indeed, the deletion of both mntA-D and mntR (mntA-R) resulted in wild-type levels of growth (Fig. 5B), suggesting that MntA-D serves as an Mn importer under the conditions tested.

The target substrate of the transport system encoded by iutABCD-E is not known; in vitro binding to Mn²⁺ and Zn²⁺ was previously demonstrated for IutA and IutE, but a transport-associated phenotype for the genes was not observed (25). Because the iutA-D locus was weakly induced in the presence of high Mn, it is possible that this transport system serves to detoxify high levels of Mn. We tested a deletion mutant in the iutA-E gene cluster, comparing growth in the absence of added Mn and at 100 μM supplementation, and did not observe significant differences between growth of the ΔiutA-E strain and the wild-type or ΔmntA-D strain under either condition (Fig. S3). A function for the IutA-E transporter in C. diphtheriae remains unknown.

MntA-D contributes to cadmium sensitivity.

To further characterize MntA-D metal import, we examined the uptake of cadmium (Cd) by C. diphtheriae. Cd shares chemical properties with Mn, and because Cd is highly toxic to bacteria, it can be used to test for transport through Mn uptake systems (9, 31, 32). We assessed wild-type C. diphtheriae 1737 and various mutants for their sensitivity to Cd using a disc diffusion method (Fig. 6A) and liquid growth assays (Fig. 6B).

FIG 6.

MntA-D is responsible for C. diphtheriae CdCl₂ sensitivity. (A) Wild-type or mutant strains as indicated were seeded in HIA and filter discs with 10 μl of 1 mM CdCl₂ were placed over the agar. Plates were incubated overnight at 37°C, and the diameters (mm) of the zone of clearance were measured. Data are the means and standard deviations from biological replicates (n = 4); only nonsignificant comparisons are noted. (B) Strains were grown in HIBTW with CdCl₂ added at the final concentrations indicated, and the optical density (OD₆₀₀) was measured following overnight incubation; data are the means and standard deviations from 3 biological replicates. (C) Wild-type or mutant strains with plasmids as indicated were seeded in HIA, and agar assays were performed as described for panel A. (D) Strains with the indicated plasmids were grown in HIBTW with CdCl₂ as for panel B; data are the means and standard deviations from 3 biological replicates. (A and C) Significance for data was assessed by Holm-Sidak’s multiple-comparison test. NS (not significant), P > 0.05; all other comparisons were significant. (B and D) Statistical significance comparing the corresponding wild-type growth and indicated strain (by color) determined by Holm-Sidak’s multiple-comparison test: *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

Filter discs containing CdCl₂ were placed on agar plates in which wild-type and mutant strains were embedded in the agar. The zones of growth inhibition observed following overnight incubation were measured to assess Cd sensitivity. The ΔmntR mutant showed a wide zone of growth inhibition and was significantly more sensitive to Cd than the wild-type strain (Fig. 6A). Mutants in the mnt transport system, mntA-D and mntA, had smaller zones of inhibition than the wild type, suggesting greater tolerance to the toxic metal. Consistent with the results for Mn described above, the sensitivity of the ΔmntA-R mutant to Cd was not significantly different from that of the wild type, which supports a role for the MntA-D transporter in the uptake of Cd. Deletion of piuB had no effect on Cd sensitivity (Fig. S4).

We also assessed Cd sensitivity in liquid assays where the overnight growth of various strains was examined in medium supplemented with CdCl₂ to final concentrations of 0, 1, 3, and 6 μM and measured culture densities after overnight growth (Fig. 6B). The wild-type strain was sensitive to 3 μM CdCl₂, while the mntR mutant showed reduced growth at 1 μM CdCl₂. Both ΔmntA and ΔmntA-D strains tolerated higher concentrations of CdCl₂ than the wild-type strain, suggesting that the MntA-D transporter is primarily responsible for the import of Cd and the resulting toxicity. The increased sensitivity of the mntR mutant to Cd and suppression of the phenotype by the deletion of mntA-D further supports a role for the MntA-D system in the import of metals. Complementation of the mntR and mntA deletion mutants restored wild-type levels of Cd sensitivity in the disc diffusion assay (Fig. 6C) and the liquid assay (Fig. 6D).

DISCUSSION

This report is the first description of an MntR/Mn regulon in a pathogenic Corynebacterium species. We showed that MntR directly binds five promoters to control expression of 12 genes in response to Mn. Additionally, we identified several genes whose expression was altered by the deletion of mntR independent of Mn levels, including several genes in the DtxR and Fe regulon (see Table S1 in the supplemental material), which exhibit reduced expression in the mntR mutant. These DtxR-regulated genes are primarily involved in Fe and heme transport and are located on seven distinct loci (33–36). A possible reason for the repression of these Fe transport genes in the mntR mutant is that high Mn levels caused by the constitutive expression of the mntA-D Mn importer results in activation of DtxR to repress transcription. A prior study demonstrated that DtxR DNA binding activity can be activated in vitro by Mn²⁺ (37). It is also possible that elevated intracellular Mn perturbs intracellular Fe homeostasis and leads to an increase of bioavailable Fe, as proposed by Guedon et al. (38). This increase in available Fe may activate DtxR, leading to repression of genes in the DtxR regulon. Further analysis of our gene expression data shows weak regulation of several genes in response to Mn independent of MntR (Table S2). The mechanism of regulation for this group of genes has not been determined.

The iutA-E gene cluster is predicted to encode an ABC metal transporter with two separate substrate binding proteins, IutA and IutE (25). Although the native metal substrate(s) for the IutABCDE (IutA-E) transporter is not known, the predicted iutA-E gene products show similarity to ABC transporters involved in Mn or Zn uptake, and IutA and IutE bind both Mn²⁺ and Zn²⁺ in vitro. The gene cluster is transcribed as two separate units, with a promoter for iutA-D and a separate promoter for iutE. DtxR and Fe directly repress the iutA-D operon and indirectly induce expression of iutE by an unknown mechanism. Zinc availability weakly regulates iutA-D (Zur binding site noted in Fig. 4A). Changes to protein levels for IutA and IutE were observed in response to Fe and Zn availability (25). In this study, we found that transcription of the iutA-D operon was weakly induced in response to Mn in an MntR-dependent manner, suggesting a possible role in protecting cells against high Mn. However, deletion of the iutA-E locus did not impact sensitivity to high Mn. It is notable that the transcription of the iutA-D operon is affected by three different metals; additional studies are needed to identify the substrates used in vivo by this unusual transporter.

While we did not observe a growth defect in the mntA-D or mntA deletion mutants under Mn-limited conditions, the deletion of mntA-D suppressed the Mn sensitivity of the mntR mutant. This is consistent with phenotypes observed in several other bacteria, including M. tuberculosis (19) and B. subtilis (9), and supports a model in which Mn accumulates to toxic levels in the absence of regulation by mntR through the function of importers such as MntA-D. In M. tuberculosis, deletion of mntR also results in sensitivity to high Mn levels (19). However, M. tuberculosis encodes two Mn import mechanisms regulated by MntR: the ABC transporter MntABCD, and MntH, an NRAMP family protein. Despite the presence of two transporters, deletion of mntA restored wild-type levels of Mn resistance in the mntR mutant. An mntR mntH double mutant was not tested for Mn sensitivity. Deletion of mntA alone resulted in reduced growth under Mn limitation, and the deletion of both mntA and mntH eliminated growth. A blastp search did not identify any MntH homologs in C. diphtheriae (39).

C. glutamicum, a commensal soil bacterium related to C. diphtheriae, encodes an MntR homolog that has 52% amino acid sequence identity to C. diphtheriae MntR. Transcriptome analysis of C. glutamicum showed that MntR regulates at least 11 genes, including several genes for arginine biosynthesis and other metabolic processes (24). The C. glutamicum strain that was tested does not encode MntH or MntA-D homologs; however, C. glutamicum MntR regulates a gene (cg1623) annotated as a Zn transporter of the ZIP family. Baumgart and Frunzke showed that Mn and MntR repress cg1623 expression, and MntR directly binds to the cg1623 promoter region. Although the regulation by Mn and MntR suggest a role in Mn import, a cg1623 deletion mutant did not have a discernible phenotype. Homologs for CG1623 were not identified in C. diphtheriae. Furthermore, the deletion of mntR from C. glutamicum did not increase sensitivity to excess Mn (24). While C. glutamicum is related to C. diphtheriae, mechanisms for Mn import appear to be significantly different.

The MntR/Mn regulon in B. subtilis is perhaps the best-characterized system of Mn regulation in Gram-positive bacteria. The B. subtilis MntR protein was initially identified based upon its homology to C. diphtheriae DtxR (9). An mntR null mutant showed greatly increased sensitivity to Mn and Cd compared to the wild-type strain. Using a transposon mutagenesis approach, Que and Helmann identified an NRAMP family protein homolog, MntH (formerly YdaR), as a potential Mn importer (9). An mntH mntR double mutant was less sensitive to both Mn and Cd than the wild type. Interestingly, the B. subtilis strain deficient in mntH was as tolerant of Cd as an mntH mntA disrupted strain, suggesting that the import of Cd occurred primarily through MntH. The mntH mutant grew poorly under Mn limitation, a phenotype that was suppressed by the deletion of mntR, leading to the identification of the Mn-dependent ABC transporter MntABCD in B. subtilis. An mntR mntH mntA triple mutant was found to be sensitive to Mn, and it was later found that Mn homeostasis in B. subtilis requires not only regulation of import through MntABCD and MntH but also control of export (23). Expression of mneP and mneS, which encode proteins of the cation-diffusion facilitator (CDF) family and are involved in Mn export, are induced by MntR in response to Mn excess. The mneP and mneS genes are only weakly expressed in an mntR mutant. The Mn sensitivity of a B. subtilis mntR mutant was attributed to the unregulated import of Mn through the Mn-specific importers and the inability to efflux Mn due to reduced expression of mneP and mneS. Unlike the phenotype observed in B. subtilis, we were able to fully suppress the Mn sensitivity of the C. diphtheriae mntR mutant through the deletion of the mntA-D transport locus. No Mn-specific efflux proteins have been identified in C. diphtheriae. Furthermore, the C. diphtheriae MntA-D locus appears primarily responsible for the sensitivity to Cd.

Our gene expression study suggests a relatively limited number of genes regulated by Mn in response to MntR. While MntA-D appears to be the primary mechanism for Mn import, deletion of mntA-D had no effect on growth in Mn depleted medium, which suggests that other transport systems, which are not regulated by MntR, may be involved in Mn uptake. Several Fe- and Zn- regulated ABC transporters whose specific metal substrate(s) has not been identified are present in C. diphtheriae and may function to import Mn (27, 36, 40). ABC transport systems in other bacteria are able to transport more than one metal; our data suggest that the MntABCD system can import both Mn and Cd and a recent study of S. aureus provides evidence that MntABC is involved in copper import in addition to Mn (41). While it is not known if Mn is an essential metal for C. diphtheriae survival and colonization, based on what is known about numerous other bacterial pathogens, it seems likely that some level of Mn is required for growth. A low requirement for Mn would be consistent with the relatively few genes identified in the C. diphtheriae MntR regulon.

MATERIALS AND METHODS

Strains, media, and growth conditions.

C. diphtheriae strain 1737 (42) and mutant derivatives (Table 2) were routinely cultured in heart infusion broth with 0.2% (vol/vol) Tween 80 (HIBTW) or on heart infusion agar (1.5% agar) (HIA) at 37°C. Strains were stored at −80°C in heart infusion broth with 20% (vol/vol) glycerol. mPGT medium was prepared as described previously (43). mPGT was supplemented with 0.5 μM FeCl₃. Kanamycin was used at 25 μg/ml and 50 μg/ml for C. diphtheriae and E. coli, respectively. Spectinomycin was used at 100 μg/ml. MnCl₂ supplementation was used as indicated in each experiment.

TABLE 2.

Strains and plasmids used in this study

Strain or plasmid	Description or use	Reference or source
C. diphtheriae strains
1737	Wild type, Gravis biotype, Tox+
1737 ΔmntR	Deletion for mntR (dip0619) in 1737	This study
1737 ΔmntA-D	Deletion for mntA-D (dip0615-618) in 1737	This study
1737 ΔpiuB	Deletion for piuB (dip0124) in 1737	This study
1737 ΔmntA-DΔpiuB	Deletion for mntA-D (dip0615-618) and piuB (dip0124) in 1737	This study
1737 ΔmntA	Deletion for mntA (dip0615) in 1737	This study
1737 ΔmntA-R	Deletion for mntA-R (dip0615-619) in 1737	This study
E. coli strains
NEB 5-alpha competent	Cloning strain	New England Biolabs, Inc.
S17-1 λpir	Mating strain	52
BL21(DE3)	Protein expression strain	53
Plasmids
pKN2.6Z	C. diphtheriae shuttle vector, Knr	54
pKNmntR	pKN2.6Z carrying the C. diphtheriae mntR gene	This study
pKNmntA	pKN2.6Z carrying the C. diphtheriae mntA gene	GenScript
pKΔmntR	Suicide vector for the deletion of mntR	This study
pKΔmntAD	Suicide vector for the deletion of the mntABCD operon	25
pKΔmntA	Suicide vector for the deletion of the mntA operon	GenScript
pKΔmntAR	Suicide vector for the deletion of the mntABCDR locus	GenScript
pSPZ	Carries lacZ; Spcr	44
pSPZsodA153	153 bp upstream of dip2261 fused to lacZ in pSPZ	This study
pSPZiutA173	173 bp upstream of dip0169 fused to lacZ in pSPZ	This study
pSPZciuE75	75 bp upstream of dip0586 fused to lacZ in pSPZ	GenScript
p28-R34H	Overexpression clone for recombinant N-terminally His-tagged MntR	8

DNA cloning and generation of C. diphtheriae mutants.

DNA sequences used in this study were derived from C. diphtheriae strain 1737 and amplified by PCR from genomic DNA or synthesized. The NEBuilder HiFi DNA assembly cloning kit (New England Biolabs, Inc.) was used to insert DNA into pKN2.6Z to generate pKNmntR. pKNmntR contains 200 bases upstream of the mntR start codon and the entire coding region. DNA was ordered from GenScript to generate pKNmntA. pKNmntA contains 150 bases upstream of the mntA start codon and the entire coding region.

Reporter plasmids pSPZ-iutA, pSPZ-sodA, and pSPZ-ciuE contain 173, 153, and 75 bp upstream of the start codons for each respective gene fused to the lacZ start codon in pSPZ (44). Sequences for iutA and sodA were amplified by PCR, while sequence for ciuE was ordered from GenScript. Plasmid sequences were verified by DNA sequencing (Macrogen).

Flanking sequences were amplified by PCR for the deletion of mntR, mntA-D, and piuB and cloned into pK18mobsacB (45). DNA fragments containing flanking sequences for the deletion of mntA and mntA-R were purchased from GenScript and cloned into pK19mobsacB. Generation of C. diphtheriae mutants was done as previously described (46); PCR across the gene locus was used to confirm the deletion. The coding sequence of mntR that remains is ATGCATGCAAAATAGATTCTTAG. The deletion of mntA-D retains the first 10 amino acids of mntA fused to the terminal 28 amino acids of mntD. The deletion of mntA-R retains the first 10 amino acids of mntA fused to the terminal 10 amino acids of mntR. The deletion of mntA retains the first 10 amino acids of mntA fused to the terminal 10 amino acids. The deletion of piuB retains the first 6 amino acids of piuB fused to the terminal 4 amino acids.

RNA extraction.

Preparation of samples for RNA extraction was done as previously described (27). C. diphtheriae strains were grown overnight in mPGT with 1 μM FeCl₃ and antibiotics as needed. Overnight cultures of C. diphtheriae were diluted into mPGT with 1 μM FeCl₃ with antibiotics as needed and grown at 37°C for 2 h. Cells were then diluted to a final optical density at 600 nm (OD₆₀₀) of 0.1 in mPGT with 0.5 μM FeCl₃ supplementation with or without 5 μM MnCl₂ supplementation. After 3.5 h of growth at 37°C, cultures were diluted with 95% ethanol (EtOH)–5% phenol on ice, and cells were collected through centrifugation at 4°C. Cells were lysed by mechanical lysis with lysing matrix B (MP Biomedicals) in phosphate-buffered saline (PBS) with 9.5% EtOH, 0.5% phenol, and 14.3 mM β-mercaptoethanol. TRIzol LS reagent (Thermo Fisher Scientific) was added to cell lysates, and RNA was isolated by following the Direct-zol RNA (Zymo Research) extraction protocol. RNA was treated with the TURBO DNA-free kit (Ambion, Life Technologies) by following the extended treatment protocol. RNA integrity was assessed using an Agilent RNA 6000 Nano chip and Bioanalyzer 2100. RNA samples were stored at −80°C until use.

Agilent single-channel microarray preparation, data collection, and analysis.

Processing of the RNA for microarray hybridization was performed as described previously (27). Custom microarrays were designed based on the C. diphtheriae NCTC 13129 genome (47) using Agilent’s eArray platform. Total RNA collected from 3 independent biological replicates of each strain and condition was used as input for the process. The Agilent One-Color Microarray-Based Exon Analysis, version 2.0, protocol was followed per the manufacturer’s directions. An Agilent G2600D SureScan microarray scanner was used. The median of the gProcessedSignal for each gene probe was normalized by the mean of the entire sample set. Normalized values were log₂ transformed and subjected to the J5 test (48) using the Center for Biologics Evaluation and Research High-Performance Integrated Virtual Environment (HIVE) (49). Genes with a log₂ fold change greater than 1 and a J5 test value greater than 2 were considered significantly regulated.

cDNA synthesis and qPCR.

The ProtoScript II first-strand cDNA synthesis kit (New England Biolabs, Inc.) was used to synthesize cDNA from 1 μg of total RNA. qPCR was performed using the LightCycler 96 (Roche) and Luna universal qPCR master mix (New England Biolabs, Inc.). Primers used for qPCR were designed using Primer3 (50) and are listed in Table S3. Data were analyzed using the ΔΔC_q method, and significance determined using 2-way analysis of variance (ANOVA) and Holm-Sidak’s multiple-comparison test on ΔC_q values. ΔC_q values for gyrB (dip0005) were used for normalization; gyrB was not differentially expressed in our analysis.

Purification of recombinant His-MntR.

Recombinant N-terminally His-tagged MntR was produced using the plasmid p28-R34H, described previously (8). A single colony was used to inoculate an overnight culture of Overnight Express TB medium (Millipore). Bacterial cells were harvested by centrifugation and stored at −80°C prior to affinity purification following the HisTALON gravity column purification protocol for the native purification of proteins (Clontech Laboratories, Inc.). The elution fraction containing His-MntR was dialyzed twice against PBS with 5% Chelex 100 resin (Bio-Rad Laboratories, Inc.). Purified protein was stored at –80°C until use.

EMSAs.

Electrophoretic mobility shift assays (EMSAs) were performed using recombinant His-tagged MntR (8). DNA fragments amplified by PCR with biotinylated primers or short double-stranded DNA generated by annealing two single-stranded DNA oligomers (described previously [27]) were used in binding reactions containing 20 mM Na₂HPO₄, 50 mM NaCl, 2 mM dithiothreitol (DTT), 5 mM MgCl₂, 0.2 μg/μl bovine serum albumin (BSA), 0.03 μg/ml sonicated salmon sperm DNA, and 0.5 mM MnCl₂ in 10% glycerol at pH 7.0. His-MntR was added as indicated at a final concentration of 5 μM. Binding reactions were incubated for 15 min at room temperature; samples were then separated by gel electrophoresis (5% acrylamide with 50 mM Na₂HPO₄ and 1 mM DTT, pH 7.0) and transferred onto a nylon membrane in 45 mM Tris-borate, 1 mM EDTA (0.5 TBE). Biotinylated DNA was detected using the LightShift chemiluminescent EMSA kit (Thermo Scientific) by following the manufacturer’s instructions. Primers used in EMSA experiments are listed in Table S4.

Beta-galactosidase (LacZ) activity assays.

LacZ activity assays were performed as described previously, with modifications (27). C. diphtheriae strains carrying the indicated plasmids were passaged from mPGT (1 μM FeCl₃) into mPGT (0.5 μM FeCl₃) without and with 5 μM MnCl₂ supplementation and grown overnight. The following day, cells were pelleted by centrifugation at 15,000 relative centrifugal force for 2 min at room temperature and treated with 10 mg/ml lysozyme in PBS for 30 min at 37°C. Cells were resuspended in 600 μl PBS, and beta-galactosidase assays were performed as described by Miller (51). Four biological replicates were performed for each strain and plasmid combination.

Cadmium chloride sensitivity assays.

Overnight cultures of each indicated strain were diluted 1:1 with fresh HIBTW and grown for 6 h. An OD₆₀₀ of 0.1 was added to 20 ml of molten HIA. The cell-agar suspension was mixed by pipetting and dispensed into 100-mm petri plates. The agar was solidified at room temperature in a biosafety cabinet, and 10 μl of 1 mM CdCl₂ was pipetted onto a 10-mm Whatman filter disc and placed on the agar surface. The plates were then incubated overnight at 37°C and results read the next morning. The diameter of the zones of clearance surrounding the Whatman filter was measured and reported.

Statistical data analysis.

GraphPad Prism version 9.0 was used for statistical analysis. Specific tests used and P values are indicated in figure legends.

Data availability.

The microarray data and design are available at the Gene Expression Omnibus database under accession number GSE173960 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE173960).