MntR(Rv2788) a transcriptional regulator that controls manganese homeostasis in Mycobacterium tuberculosis

Summary

The pathogenic mycobacterium Mycobacterium tuberculosis encodes two members of the DtxR/MntR family of metalloregulators, IdeR and SirR. IdeR represses gene expression in response to ferrous iron, and we here demonstrate that SirR (Rv2788), although also annotated as an iron-dependent repressor, functions instead as a manganese dependent transcriptional repressor and is therefore renamed MntR. MntR regulates transporters that promote manganese import and genes that respond to metal ion deficiency such as the esx3 system. Repression of manganese import by MntR is essential for survival of M. tuberculosis under conditions of high manganese availability, but mntR is dispensable during infection. In contrast, manganese import by MntH and MntABCD was found to be indispensable for replication of M. tuberculosis in macrophages. These results suggest that manganese is limiting in the host and that interfering with import of this essential metal may be an effective strategy to attenuate M. tuberculosis.

Introduction

Mycobacterium tuberculosis, the causative agent of human tuberculosis, remains one of the most successful pathogens. The emergence of multidrug resistant isolates and the association of Mtb and HIV infection, underscores the need to identify new targets for therapeutic intervention.

Metals are required for all living organisms due to their role in protein structure and enzymatic activity. In response to infection and to combat invading pathogens, vertebrates restrict the availability of transition metals including iron (Fe), manganese (Mn) and zinc (Zn) (Weinberg, 1984, Kehl-Fie and Skaar, 2010, Hood et al., 2012). This process is known as “nutritional immunity”. To overcome metal limitation, bacteria synthesize molecules that allow them to bind and import essential metals from the environment, competing for metal ions with the host. Mycobacterium tuberculosis (Mtb) possesses several metal sensing and acquisition systems, of which the high affinity siderophore-mediated Fe³⁺ uptake systems are the best studied (Neyrolles et al., 2015). In contrast, very little is known regarding uptake systems for other metals. Although metals are essential for viability and growth, when in excess they can be toxic to cells. Iron and copper ions are particularly harmful due to their tendency to catalyze the Fenton reaction leading to the production of damaging oxygen radicals (Stadtman, 1991). Other transition metals can also be toxic due to their adventitious binding and inactivation of proteins (Waldron and Robinson, 2009). Therefore, maintaining metal homeostasis is fundamental for cell survival.

In bacteria, metal-dependent regulatory (metalloregulatory) proteins maintain metal homeostasis by controlling the expression of genes involved in metal import, storage and efflux (Waldron and Robinson, 2009). Metal sensing involves reversible interaction of the regulator with specific metal ions. This interaction generally leads to the activation of DNA binding activity, thereby modulating transcription of the target genes. In bacteria, the best characterized metalloregulatory proteins cluster in seven distinct families. The diphtheria toxin repressor (DtxR) family has representatives that sense either or both Fe²⁺ and Mn²⁺ as co-repressors to promote DNA binding. Mtb has two representatives of the DtxR family: IdeR (iron dependent repressor), which controls iron homeostasis genes and is essential for virulence (Rodriguez et al., 2002, Gold et al., 2001, Pandey and Rodriguez, 2014), and the protein encoded by Rv2788 which has been annotated as SirR (iron repressor), but whose function has not been characterized.

In this study we show that Rv2788 is a manganese transport regulator, and therefore rename this protein as MntR (manganese transport regulator) to be consistent with its characterized orthologs in other organisms (Que and Helmann, 2000, Baumgart and Frunzke, 2015). We show that MntR is the central regulator of Mn²⁺ homeostasis in Mtb and identify an MntR-binding consensus motif present in the promoter of several genes regulated by Mn²⁺. In particular, we show that MntR directly regulates the expression of two Mn²⁺ transporters, mntH whose role in Mn²⁺ uptake has been proposed previously (Agranoff et al., 1999), and MntABCD, an ABC transporter that has not been previously implicated in Mn²⁺ import. We show that MntABCD is highly induced under Mn²⁺ limitation and is necessary for growth under Mn²⁺ deficiency. We found that mutants lacking MntR, which constitutively express Mn²⁺ import functions, are unaffected in growth in macrophages and in mice. Conversely, mutants lacking the two Mn²⁺ transporters were unable to grow intracellularly, indicating that Mn²⁺ import is essential for Mtb replication in macrophages. Together, these results suggest that targeting Mn²⁺ acquisition may be a good strategy for attenuation of Mtb pathogenesis.

Results

MntR is required for manganese homeostasis

The protein encoded by mntR (formerly sirR) is a homolog of the diphtheria toxin regulator DtxR, the prototype of a large family of bacterial metalloregulators. DtxR and its ortholog in Mtb (IdeR) are well characterized divalent metal ion dependent repressors that control iron homeostasis. Other members of the DtxR family, like Bacillus subtilis MntR, function as Mn²⁺ sensors and control Mn²⁺ homeostasis. Metal selectivity in this protein family is influenced by the sequence and geometry of the metal binding site and also the cellular metal environment. Important residues for metal selectivity in members of this family were previously identified (Guedon and Helmann, 2003). In DtxR, Met-10, Cys-102, Glu-105, and His-106 coordinate the regulatory metal. In MntR, Met-10 and Cys-102 are replaced by Asp and Glu, respectively. Alteration of these residues in MntR to those in DtxR results in a mutant protein that responds to both manganese and iron. Like B. subtilis MntR, mycobacterial MntR has Asp and Glu in the metal coordinating site suggesting that it may be a manganese responsive regulator (Fig. 1).

Fig. 1.

Sequence comparison of DtxR-like homologues. Primary structure alignment of DtxR homologues including iron and manganese regulators. Identical and similar positions are indicated by black and gray boxes respectively. Conserved residues involved in the coordination of the regulatory metal are marked by an asterisk (*) under the sequence. Aspartic and Glutamic acid at the metal binding site characteristic of Mn sensors are conserved in Mtb MntR. Mycobacterium tuberculosis (Mtb), Corynebacterium diphtheriae (C.diph), Bacillus subtilis (B.sub) and Staphylococcus epidermidis (S.epi). Mtb SirR was renamed MntR based on the results of this study.

To determine the function of MntR, we introduced a kanamycin resistance cassette into the mntR gene and used allelic replacement to disrupt the chromosomal copy in Mtb as verified by Southern Blot analysis (Fig. S1). The mntR null mutant had similar colony morphology and growth properties to wild type Mtb in standard Middlebrook 7H9 medium (data not shown). Reasoning that MntR might function in control of metal ion transport, we examined the sensitivity of the mutant to metal ions using a defined minimal medium containing a sufficient concentration of essential metals and supplemented with high concentrations of one metal. In these experiments the mntR mutant displayed high sensitivity to Mn²⁺ relative to the wild type (Fig 2A). This phenotype was complemented by a single copy of mntR introduced in an integrative plasmid at a heterologous site (Fig 2B).

Fig. 2. MntR deficiency results in sensitivity to high Mn concentrations in Mtb.

(A) Exponential growing cultures of wild type (filled bars) and mntR mutant (open bars) were diluted to a starting O.D 540 of 0.05 and exposed to increasing concentrations of metal ions. Shown is the O.D 540 of four-day cultures exposed to 200 µM of the various metal ions. Except for Mn²⁺, higher concentrations of other metals did not affect the mutant more than the wild type (data not shown). (B) Growth of M. tuberculosis strains under different manganese conditions. Wild type (squares), mntR mutant (triangles) and the complemented strain (circles) were grown in low Mn²⁺ minimal media (filled symbols) or medium containing 200µM MnCl₂ (open symbols). Growth was monitored by the increase in O.D 540 nm. Data are expressed as the mean ± standard deviations from three biological replicates.

To determine if the increased sensitivity to Mn²⁺ was associated with changes in intracellular Mn²⁺ levels we measured the intracellular content of Mn²⁺ in wild type, mntR mutant, and complemented strains exposed to sufficient (1 µM) or excess (200 µM) Mn²⁺ in the medium. The results demonstrated that the hypersensitivity to Mn²⁺ in mntR was accompanied by a large elevation in the intracellular levels of Mn²⁺ relative to the wild type and complemented strains (Fig 3). In high concentrations of Mn²⁺ the MntR showed reduced viability indicating that MntR is necessary to prevent toxic accumulation of intracellular Mn²⁺ (Fig S2). Conversely, inactivation of mntR did not affect growth under Mn²⁺-limiting conditions (Fig 2B), suggesting that MntR is not needed for efficient Mn²⁺ import in Mtb. The mntR mutant was also highly sensitivity to cadmium Cd²⁺ (Table 2), a metal that is often adventitiously imported through Mn²⁺ uptake channels. These results are consistent with the hypothesis that Mtb MntR is an ortholog of B. subtilis MntR (MntR_BS). MntR_BS represses Mn²⁺ uptake under Mn²⁺ sufficiency by repression of both mntH and mntABCD operons, and its loss causes sensitivity to both Mn²⁺ and Cd²⁺ (Guedon et al., 2003)

Fig. 3. Lack of MntR leads to increased cellular manganese content.

Total manganese in the wild type (WT), mntR mutant and mntR mutant complemented (mntR comp) strains cultured in minimal media containing 1µM (filled bars) or 200µM MnCl₂ (open bars) was measured by ICP-MS as described in experimental procedures. Data are expressed as the mean ± standard deviation from three biological replicates. *P< 0.01.

Table 2. Sensitivity of Mtb strains to oxidative stress and CdCl₂.

Data show the diameters of the zones of growth inhibition in mm produced in the presence of 500 mM hydrogen peroxide (H₂O₂), 5 mM plumbagin or 100 mM cadmium chloride (CdCl₂). Wild type (WT) H37Rv parental strain, the mntR mutant, the mntR mutant complemented (mntR::mntR) and the double mntR-mntA mutant were grown in 7H9 medium to exponential phase and plated 7H10 medium. Data are expressed as the mean±standard deviations from three biological replicates.

	H37Rv (WT)	mntR	mntR ::mntR	mntR-ΔmntA
H2O2	22±1	18±1	24±1	ND
Plumbagin	24±1	42±2	23±2	23±2
CdCl2	21±1	46±2	20±1	33±4

MntR represses mntH

Reasoning that the mntR mutant might have derepressed Mn²⁺ transport we set out to investigate regulation of Mn²⁺ transport by MntR. The Nramp homologue MntH (also known as Mramp) is believed to transport Mn²⁺ in mycobacteria based on its ability to transport Mn²⁺ and other divalent metals when expressed in a heterologous system (Agranoff et al., 1999). A mntH mutant of Mtb was reported to be deficient for growth under iron limitation (Boechat et al., 2002), but the possible role of this protein in Mn²⁺ acquisition in Mtb has not been investigated.

We first examined whether expression of mntH was induced under Mn limitation. Using qRT-PCR we measured mntH transcript levels in wild type and the mntR mutant cultured in Mn²⁺ limited medium and in medium containing 50 µM Mn, which is a high but not toxic concentration for the mutant. Expression of mntH in the wild type strain was ~4 fold higher under Mn²⁺ limitation, while in the mntR mutant mntH was constitutively expressed and in the complemented strain mntH expression was regulated similarly to the wild type (Fig 4A). In addition, an electrophoretic mobility shift assay (EMSA) demonstrated Mn²⁺-dependent binding of MntR to the mntH promoter region (Fig 4B), suggesting that MntR can regulate mntH directly. To determine the contribution of MntH to Mn²⁺ acquisition we generated an mntH null mutant in Mtb H37Rv by allelic replacement, and examined the effect of the mutation. We found that inactivation of mntH did not impact the ability of Mtb to grow under Mn limitation (Fig 4C) indicating that Mtb contains at least one additional Mn²⁺ uptake system.

Figure 4. mntH (Rv0924) is regulated by Mn²⁺ and MntR but is dispensable for growth under Mn²⁺ limitation.

A. Expression of mntH in Mtb strains grown under Mn²⁺ limitation (open bars) or with 50 µM Mn²⁺ (filled bars) was analyzed by qRT-PCR with SYBR green. The data are expressed as the relative quantity of mRNA normalized to 16S ribosomal RNA and presented as the mean ± standard deviations from three biological replicates. *P< 0.01. B. Electrophoretic mobility shift assay (EMSA). MntR was added in increasing concentrations from 0.25 to 1 µM to the biotin labeled DNA probe and where indicated, 0.2 mM MnCl₂ was added. C. Growth of wild type (■) and mntH mutant (ο) under Mn limitation. Growth was monitored by the increase in O.D 540 nm. Data are expressed as the mean ± standard deviations from three biological replicates.

Response of Mtb to manganese limitation

To identify additional Mn²⁺ transporters we conducted transcriptomic analysis of Mtb cultured in low and high Mn²⁺ (Table 3 and 4; Geo Accession GSE70812). 47 genes were significantly upregulated in response to Mn²⁺ limitation (Table 3) and 4 were induced in high Mn²⁺ conditions (Table 4). Many of the genes induced in low Mn²⁺ are organized in clusters representing functional units. These include the entire 11-gene cluster (Rv0280-Rv0291) encoding Esx3, an essential secretion system, and five genes encoding Esat-6 like secreted proteins. A ribosomal protein encoding gene cluster (Rv2055c-Rv2058c) and several PE-PPE proteins were also upregulated. Genes induced in high Mn²⁺ encoded a transcriptional regulator, a chaperone and two unknown proteins.

Table 3. Genes induced under Mn limitation.

A DNA microarray was used to measure mRNA levels in cultures of Mtb H37Rv during exponential growth under Mn deficient (LMn) (< 1 µM MnCl₂) or sufficient HMn (50 µM MnCl₂) conditions (Geo Accession GSE70812). Differences in gene-specific RNA levels of two fold or more are listed. Genes that are separated by < 50 bp and probably coregulated are grouped. The genes are annotated as described by the Pasteur Institute on TUBERCULIST (http://genolist.pasteur.fr/tuberculist).

Rv no.	Gene	mRNA Wt LMn/HMn	Function
Rv0023		3.2±1.0	Transcriptional regulator
Rv0105c	rpmB	2.9±0.9	Ribosomal protein
Rv0106		9.2±1.6	CMP
Rv0280		54.2±29	PPE
Rv0281		5.4±0.5	CMP
Rv0282		6.6±1.5	CHP
Rv0283		8.0±1.2	CMP
Rv0284		7.2±0.8	CMP
Rv0285		6.5±0.8	PE
Rv0286		5.5±0.7	PPE
Rv0287	esxG	5.3±0.5	Esat-6 like protein
Rv0288	esxH	6.0±1.1	EsxH
Rv0289		5.6±0.3	CHP
Rv0290		5.5±0.3	CMP
Rv0291	mycp3	5.4±0.4	Membrane anchored mycosin
Rv0532		2.9±0.8	PE-PGRS
Rv0822c		2.1±0.4	CHP
Rv0823c		3.1±0.6	Transcriptional regulator
Rv1087		2.4±0.1	PE-PGRS
Rv1195		2.5±0.2	PE
Rv1217c		2.1±o.2	Membrane protein ABC transporter
Rv1219c		2.5±0.7	Transcriptional regulator
Rv1283c		50.0±13	ATP binding protein ABC transporter
Rv1361c		2.1±0.4	PPE
Rv1387		2.3±0.02	PPE
Rv1719		2.4±0.4	Transcriptional Regulator
Rv1981c	nrdF1	2.5±0.1	Ribonucleotide reductase
Rv2055c	rpsR2	21.63±2.4	Ribosomal protein
Rv2056c	rpsN2	40.0±24	Ribosomal protein
Rv2057c	L33	37.3±6.4	Ribosomal protein
Rv2058c	rpmB2	44.6±18	Ribosomal protein
Rv2059		10.1±2.1	CHP
Rv2645		2.0±0.2	HP
Rv2933	ppsC	2.2±0.2	Polyketide synthase
Rv2990c		11.8±5.5	HP
Rv3019c	esxR	7.7±2.1	Esat-6 like protein
Rv3020c	esxS	5.1±0.3	Esat-6 like protein
Rv3061c	fadE22	2.1±0.2	Acyl-CoA dehydrogenase
Rv3092c		2.1±0.4	CMP
Rv3094c		2.4±0.7	CHP
Rv3095		2.1±0.2	Transcriptional regulator
Rv3485c		2.4±0.7	Dehydrogenase/reductase
Rv3620	esxW	3.0±0.9	Esat-6 like protein
Rv3653		2.8±1.1	PE-PGRS
Rv3659c		2.7±0.4	CHP
Rv3741c		2.4±0.3	oxidoreductase
Rv3862	whiB6	2.8±0.3	Transcriptional regulator

Table 4. Genes induced under Mn sufficiency.

A DNA microarray was used to measure mRNA levels in cultures of Mtb H37Rv during exponential growth under Mn deficient (LMn) (< 1 µM MnCl₂) or sufficient HMn (50 µM MnCl₂) conditions (Geo Accession GSE70812). Differences in gene-specific RNA levels of two fold or more are listed. The genes are annotated as described by the Pasteur Institute on TUBERCULIST (http://genolist.pasteur.fr/tuberculist).

Rv no.	Gene	mRNA Wt HMn/LMn	Function
Rv2358		4.4±0.6	Transcriptional regulator
Rv3269		3.2±0.8	CHP
Rv3418c	groES	2.5±0.1	chaperonin
Rv3848		7.3±2.8	Membrane protein

The second highest induced gene in response to Mn²⁺ limitation was Rv1283c, which encodes a membrane protein part of the permease component of an ABC transporter. Because manganese specific importers across phylogenetically diverse bacteria predominantly involve members of the ABC transporter family in addition to (Nramp) homologues, we focused our attention on this transporter.

MntABCD, an ABC transporter regulated by MntR

Rv1283c is the first gene in the predicted and here confirmed operon composed of Rv1283c–Rv1280c (http://genolist.pasteur.fr/Tuberculist/ http://genome.tbdb.org) (Fig S3). Rv1280c encodes the substrate binding protein, Rv1281c encodes the ATPase and Rv1282c and Rv1283c encode the membrane permease component of an ABC transporter. Quantitative RT-PCR confirmed induction of Rv1283c in response to Mn²⁺ limitation (Fig. 5A). However, In the absence of MntR, Rv1283c was expressed equally in low and high Mn²⁺, indicating that MntR was required for Mn²⁺ mediated repression of this gene (Fig. 5A). To determine whether MntR regulates Rv1280c-RV1283c directly or indirectly we tested MntR binding to the promoter region upstream of Rv1283c by EMSA. In the presence of Mn²⁺, MntR bound to the Rv1283c promoter region (Fig 5B). The affinity of the interaction (K_d) of MntR with the Rv1283c and mntH promoter regions was comparable: approximately 250 nM MntR was required to retard the mobility of 50% of either labeled DNA fragment (Fig 4 and 5). In contrast, even at a high concentration of 1 µM, MntR did not bind to a non-specific DNA fragment (Fig 5C). To determine the range of metal ions that could activate DNA binding we performed another EMSA using MntR and various divalent cations. MntR bound DNA most efficiently in the presence of Mn²⁺, but low binding was also detected in the presence of Ni²⁺ or Zn²⁺ while no binding occurred in the presence of Cu²⁺ (Fig 6).

Figure 5. Manganese and MntR dependent regulation of Rv1283c in Mtb.

A. Expression of Rv1283c in Mtb strains wild type (WT), mntR mutant and mntR mutant complemented with wild type mntR (mntR comp) grown under Mn limitation (open bars) or with 50 µM (filled bars) manganese was analyzed by qRT-PCR with SYBR green. The data are expressed as the relative quantity of mRNA normalized to 16S ribosomal RNA and presented as the mean ± standard deviations from three biological replicates. *P< 0.01. B. Electrophoretic mobility shift assay (EMSA). MntR was added in increasing concentrations from 0.25 to 1 µM to the biotin labeled DNA probe and where indicated, 0.2 mM MnCl₂ was added. C. EMSA of MntR and a non specific DNA. MntR was added where indicated at a final concentration of 1 µM.

Figure 6. Metal requirement for binding of MntR to DNA.

MntR (0.5 µM) was incubated with the promoter region of 1283c in the presence of different metals, as indicated in the figure. Presence and absence of metal/protein is indicated by (+) and (−) respectively. Shown is one representative experiment repeated three times.

MntABCD promote Mn import in Mtb

To examine the role of Rv1280c-RV1283c in Mn²⁺ homeostasis in Mtb, Rv1280c (mntA) was deleted in both the mntR null mutant and in its parental strain and the resulting mutants were evaluated for Mn²⁺ dependent growth. In the mntR mutant inactivation of the transporter almost fully restored normal resistance to high Mn²⁺ (Fig. 7A), while in the parental strain it led to reduced growth under Mn²⁺ limitation (Fig 7B). Taken together, these results indicate that the ABC transporter encoded by Rv1280c-RV1283c promotes Mn²⁺ uptake in Mtb. Since inactivation of this transporter suppressed the Mn²⁺ hypersensitivity of the MntR mutant, these results also suggest that MntR-mediated repression of this transporter is critical to Mn²⁺ homeostasis in Mtb. In light of these results we have renamed the Rv1280c-RV1283c operon mntABCD, according to the accepted nomenclature for Mn²⁺ transporters.

Fig. 7. Deletion of mntA rescues Mn²⁺ resistance in the mntR mutant while decreases Mn²⁺ import in wild type Mtb.

(A) Wild type (■), ΔmntA (▲), mntR mutant (♦) and a mntR, mntA double mutant (●) were grown in minimal media with 200µM MnCl₂. (B) Mtb wild type (■), ΔmntA (▲), mntH mutant (◊) and mntH-mntA double mutant (○) were grown in manganese deficient medium. Growth was monitored by the increase in optical density at 540nm. Data are expressed as the mean ± standard deviations from three biological replicates. (C) Total manganese content in bacteria cultured in minimal media with 1µM MnCl₂ as measured by ICP-MS. Data are expressed as the mean ± standard deviations from three biological replicates. *P< 0.01.

To better define the contribution of MntABCD to Mn²⁺ transport in Mtb we also deleted mntA in the mntH mutant to assess the impact of lack of both transporters on growth under Mn²⁺ limitation. Inactivation of MntH and MntABCD completely eliminated growth of Mtb under Mn²⁺ limitation (Fig 7B). This phenotype was accompanied by a large reduction of intracellular Mn⁺² levels relative to the wild type strain (Fig 7C), consistent with decreased Mn²⁺ uptake in the double mutant. Growth of Mn²⁺ deficient mntH-mntA was restored by Mn²⁺ indicating that Mn²⁺ limitation is bacteriostatic for Mtb (Fig S4). From these results we conclude that MntH and MntABCD constitute the main Mn²⁺ importers in Mtb.

Identification of other MntR regulated genes

The results of the microarray analysis were validated by qRT-PCR for the genes most highly induced under Mn²⁺ limitation in the wild type Mtb. This analysis confirmed that three components of the esx3 cluster (Rv0283, Rv0284 and Rv0287), the first gene in the ribosomal protein operon Rv2055c–Rv2058c, and Rv2059-Rv2060 were all repressed by Mn²⁺ in an MntR-dependent manner (Fig. 8). Additionally, MntR binding to the promoter region of Rv2055c and Rv2059 was demonstrated in EMSA experiments (Fig S5), indicating that these genes were under the direct control of MntR. In silico analysis revealed a consensus sequence motif in the promoter region of genes that were validated as direct targets of MntR (Fig 9). Truncation of this sequence from the Rv1283c promoter region abolished MntR binding, confirming this sequence as the MntR binding site in this promoter (Fig 10A). Additionally, point mutations in highly conserved nucleotides in this sequence identified several nucleotides critical for MntR binding (Fig 10B).

Fig. 8. Effect of mntR inactivation on the expression of manganese regulated genes.

mRNA transcripts of (A) Rv2055c, (B) Rv2059, (C) Rv0283, (D) Rv0284 and (E) Rv0287 obtained from cells grown in low manganese or in high manganese (50µM MnCl₂) were analyzed by qRT-PCR with SYBR green in Wild type (Wt), mntR mutant and mutant complemented (mntR comp) strains. Open bars represent low manganese and the filled bars represent high manganese. The data are expressed as the relative quantity of the respective mRNA normalized to 16S ribosomal RNA and presented as the mean ± standard deviations from three biological replicates. *P< 0.01.

Fig 9. Conserved MntR binding sequence present in the promoter region of genetically and biochemically verified MntR regulated promoters.

Figure created with WebLogo http://weblogo.berkeley.edu/logo.cgi.

Fig 10. Characterization of the MntR binding sequence.

(A) EMSA of MntR and truncated probes derived from the promoter region upstream of Rv1283c. MntR was incubated with three different probes. Probe a includes the conserved binding sequence (Mn box) plus adjacent sequences upstream and downstream. In probe b the adjacent sequence upstream of the Mn box was removed. Probe c is truncated at the Mn box. MntR DNA binding reactions included 0.5 µM MntR and 0.2 mM MnCl₂ and complexes were resolved on 8% Tris acetate polyacrylamide gel and visualized by SYBER green staining. B. Site directed mutagenesis of the Rv1283c Mn box and effect on MntR binding. Point mutations were introduced in Mn box as indicated (M1-M5) and the effect on MntR binding was examined by EMSA. Replacement of all conserved nucleotides simultaneously abolish MntR shift. Changes A1C, G15T, A18C lead to partial shift whereas the change of T22G had no effect.

Identification of the transcriptional start point (TSP) for several MntR target genes by 5’ RACE indicated that the MntR binding sequence overlapped the −10 sequence motif in the promoters (Fig. 11). These results indicate that MntR binding should occlude access of the RNAP for initiation of transcription. MntR binding consensus sequences were also identified upstream of additional Mn²⁺ repressed genes (Fig S6). Additional validation is required to determine if all these genes are regulated directly by MntR.

Fig. 11.

Regulatory region of selected MntR controlled genes. (A) Regulatory region upstream Rv2058c, (B) Regulatory region of Rv2059, (C) Regulatory region of Rv1283c. Boldface nucleotides represent the putative translational start site, predicted MntR binding sites are underlined, transcriptional start sites are represented as +1 and deduced –10 and –35 sequences are marked and labeled. In each case the MntR binding site overlapps the RNAP recognition sequence −10.

Mn homeostasis and resistance to oxidative stress

To begin to elucidate the basis of Mn²⁺ toxicity in Mtb we investigated the sensitivity of the MntR mutant to various stresses other than Mn²⁺. We found that while the mntR mutant was as sensitive as the wild type strain to hydrogen peroxide it was hypersensitive to the superoxide generator plumbagin. Furthermore, like the hypersensitivity to Mn²⁺, the high sensitivity to superoxide was also suppressed by inactivation of the MntABCD transporter (Table 2). This result suggested that oxidative stress was linked to unrestricted Mn²⁺ uptake and elevated Mn²⁺ in the cell. Superoxide dismutase (SOD) is the main line of defense against superoxide. Mtb has two SODs an Fe-SOD and a Cu/Zn-SOD therefore, we examined whether Mn²⁺ could affect SOD activity. While SOD activity was equivalent in all strains when they were cultured in 1 µM Mn²⁺, the mntR mutant had significantly less SOD activity than the wild type strain when the concentration of Mn²⁺ in the medium was raised to 200 µM and this decrease was fully complemented by mntR (Fig 12). This result indicates that under conditions of high Mn²⁺ in the environment, MntR is required to maintain normal levels of SOD activity.

Fig. 12. MntR deficiency results in reduced SOD activity.

Wild type (Wt), mntR mutant and mntR mutant complemented (mntR comp) strains were grown in minimal media with 1 µM MnCl₂ (open bars) or with 200 µM MnCl₂ (filled bars) and SOD activity was calculated as described in experimental procedures. Data are expressed as the mean ± standard deviations from three biological replicates. *P< 0.01.

Manganese homeostasis and Mtb virulence

To determine the relevance of Mn²⁺ homeostasis to Mtb pathogenesis, we infected THP-1 derived macrophages with the mntR mutant and the Mn²⁺ transporter-deficient strains and evaluated their ability to survive and replicate in these cells. We found that while mntR was dispensable for intracellular growth (Fig 13B), loss of MntABCD reduced Mtb growth and simultaneous inactivation of both MntH and MntABCD eliminated multiplication of Mtb in macrophages (Fig 13A). Dispensability of MntR and the requirement of Mn²⁺ uptake for proliferation suggest limited availability of Mn²⁺ in the THP-1 phagosome.

Fig. 13. Role of Mn²⁺ transport in Mtb growth in macrophages.

THP-1 cells were induced to differentiate into macrophages and infected with A. wild type parental strain (♦), mntA mutant (▲), mntA mutant complemented (Δ), mntH mutant (■), and mntA-mntH double mutant (●) or B. with wild type (●) and mntR mutant (■) as described in experimental procedures. Data shows the number of CFUs recovered from infected macrophages. Data are expressed as the mean ± standard deviations from three wells. The experiment was repeated twice.

Discussion

Increasing evidence supports the importance of metal homeostasis and metalloregulatory proteins in bacterial physiology and pathogenesis. In this work we identified MntR, a member of the DtxR family of metalloregulators conserved in mycobacteria, as a Mn²⁺ dependent transcriptional repressor required for homeostatic regulation of intracellular Mn²⁺ concentration in Mtb. In particular, we found that Mtb lacking MntR accumulated cellular Mn²⁺ and were defective for growth under conditions of high Mn²⁺ availability. We also identified MntR regulated genes including two Mtb Mn²⁺ transporters, MntH and MntABCD and defined an MntR binding consensus sequence. Finally, we found that Mn²⁺ transporters were required for Mtb proliferation within macrophages.

Mn²⁺ is essential for the survival of most bacteria where it serves as a cofactor for numerous oxidoreductases, transferases and hydrolases. However, Mn²⁺ is also potentially toxic when in excess (Fig S2). Increased sensitivity to superoxide in Mtb lacking MntR was Mn²⁺ dependent, correlated with reduced superoxide dismutase activity and was suppressed by inactivation of a Mn²⁺ transporter (Table 2). The main superoxide dismutase in Mtb, SODA is an Fe-SOD (Kusunose et al., 1976) that belongs to the family of Fe-Mn superoxide dismutases which are active only when they bind the right metal (either Fe²⁺ or Mn²⁺) and are very sensitive to inactivation by the non-cognate metal (Aguirre and Cullota, 2012). Based on these observations, we hypothesize that elevation of intracellular Mn²⁺ resulting from mntR inactivation favors Mn²⁺ misincorporation into SODA and consequently, its loss of function. We also postulate that altering the normal content of Mn²⁺ in the cell results in Mn²⁺ outcompeting other metals for protein ligation. Incorporation of Mn²⁺ instead of the cognate metal can affect protein structure and/or inhibit enzymatic activity. This may represent a relevant mechanism of Mn²⁺ toxicity in Mtb.

Our efforts to identify additional Mn²⁺ uptake systems in Mtb led us to examine the role of an ABC transporter highly induced under Mn²⁺ limitation and repressed by MntR which we renamed MntABCD. MntABCD inactivation resulted in Mn²⁺ deficiency in the wild type strain while rescuing Mn²⁺ sensitivity of the mntR mutant. Moreover, simultaneous inactivation of this transporter and MntH eliminated growth of Mtb under Mn²⁺ limitation and reduced intracellular Mn²⁺ concentration. In agreement with previous studies mutation of MntH only, does not reduce Mtb virulence in macrophages or mice (Boechat et al., 2002, Domenech et al., 2002). However, Inactivation of MntABCD reduced growth under Mn limitation and partially attenuated Mtb in macrophages. Collectively these results suggest that although MntABCD and MntH constitute the primary Mn²⁺ transporters in Mtb, MntABCD has a larger impact than MntH in Mn import. MntABCD was previously postulated to function as a dipeptide transporter and mntABCD mutants are more resistant to toxic dipeptides (Green et al., 2000, Flores-Valdez et al., 2009). Our results indicate that MntABCD, participates also in Mn⁺² import. Plant oligopeptide transporters that also mediate metal transport have been described (Lubkowitz, 2011). An interesting possibility is that MntABCD may transport Mn²⁺ in the form of a peptidic chelate similar to phytochelatins used by plants to chelate metals and transport them through peptide transporters. A detailed biochemical characterization and substrate uptake determinations are needed to address this question.

Interestingly, we discovered that the Esx3 secretion system is regulated by Mn²⁺ and MntR. This system was previously found to be repressed by IdeR and by Zur in response to Fe²⁺ and Zn²⁺ respectively (Rodriguez et al., 2002, Maciag et al., 2007). Two promoters have been identified driving the expression of this gene cluster: a promoter proximal to the transcriptional start site that is repressed by IdeR-Fe and a more distal promoter repressed by Zur-Zn (Maciag et al., 2007). A putative MntR binding site is present in between these two promoters. MntR bound to this binding site may impact transcription originating at these promoters or may control an additional promoter, suggesting a complex regulation of Esx3 by Fe²⁺, Mn²⁺ and Zn²⁺ that merits further investigation. Esx3 contributes to Fe and Zn uptake (Serafini et al., 2013) (Siegrist et al., 2009) by mechanisms that are still unknown and its regulation by Mn²⁺ raises the possibility that it also plays a role in Mn²⁺ uptake. In our experiments the double mntH-mntABCD mutant was unable to grow under Mn²⁺ deficiency, suggesting that they are the main Mn²⁺ transporters in Mtb. However, we do not rule out that under particular conditions, not tested here, Esx3 could contribute to Mn²⁺ uptake. It is interesting that EsxH, a protein secreted by the Esx3 system, is involved in impairing phagosome maturation (Mehra et al., 2013). Thus, metal limitation in the phagosome may signal Mtb to activate secretion by Esx3, preventing Mtb trafficking to the lysosome. More comprehensive studies on the response of esx3 to metal ion limitation will enhance the understanding of its role in Mtb pathogenesis.

Ribosomal protein-encoding genes Rv2055c–Rv2058c are induced under Zn²⁺ deficiency (Maciag et al., 2007), and according to our results, also by Mn²⁺ limitation. Comparative genomics of zinc regulons in bacteria have shown that genes encoding paralogs of ribosomal proteins are often induced under zinc deficiency (Panina et al., 2003). It has been suggested that induction of genes encoding paralogs of ribosomal proteins that do not require zinc allows incorporation of these paralogs in a fraction of ribosomes in place of zinc proteins, to maintain ribosome function. Under these circumstances the original metal containing protein can be degraded, with subsequent release of metal for utilization by other proteins. Thus, induction of Rv2055c–Rv2058c may serve this purpose under Zn²⁺ or Mn²⁺ limitation in Mtb.

Manganese acquisition is required for full virulence of several pathogens including Brucella abortus, Yersinia pestis, Streptococcus pneumoniae, Streptococcus pyogenes and Staphylococcus aureus (Anderson et al., 2009, Bearden and Perry, 1999, Berry and Paton, 1996, Dintilhac et al., 1997, Janulczyk et al., 2003, Kehl-Fie et al., 2013). In Mtb, we found MntR to be dispensable for growth both in macrophages (Fig 13) and mice (Fig S7). In contrast, Mn²⁺ transport was required for growth in THP-1 macrophages. Collectively, these results suggest that, Mn²⁺ is limiting in the phagosome and therefore, restriction of Mn²⁺ uptake by MntR is dispensable, while Mn²⁺ import is required for growth. Previous studies that quantify the metal content of phagosomes in INF-γ activated murine macrophages infected with M. avium did not suggest decrease in Mn⁺² availability (Wagner et al., 2005). However, this study was conducted in Nramp1^s murine macrophages. Since THP-1 cells are Nramp1^R, Mn²⁺ restriction in the phagosome may be maintained by Nramp1 mediated efflux of Mn²⁺ out of the phagosome. Future studies will investigate the role of Mn²⁺ uptake in Mtb virulence in the mice model of TB and elucidate the contribution of Mn²⁺ restriction to host defense against Mtb.

Experimental procedures

Bacterial strains, media, and growth conditions

Escherichia coli strains JM109, XL-10 (stratagene) and HB101 used for cloning were grown in Luria-Bertani (LB) media. M. tuberculosis H37Rv and derived strains were maintained in 7H10 agar or 7H9 broth (Difco) supplemented with 0.2% glycerol, 0.05% Tween 80 and ADN (0.5% BSA, 0.2% dextrose and 0.085% NaCl). To grow Mtb under metal defined conditions, minimal medium (MM) was used. MM contains 0.5% asparagine (w/v), 0.5% KH₂PO₄ (w/v), 2% glycerol, 0.05% Tween 80 and 10% ADN. The pH was adjusted to 6.8. To lower the level of trace metal contamination MM was treated with Chelex-100 (Bio-Rad laboratories) according to the manufacturer’s instructions. Chelex was removed by filtration, and before use the medium was supplemented with 40 mg l⁻¹ MgSO₄, 0.5 mg l⁻¹ ZnCl₂, and 50 µM FeCl_3. We refer to this medium as low Mn medium (contains less that 1µM Mn²⁺) and to this medium we add the desired concentrations of MnCl₂ as specified. Where indicated, antibiotics were included at the following concentrations: hygromycin (Hyg) 100 µg ml⁻¹, kanamycin (Kan) 20 µg ml⁻¹, Streptomycin (Sm) 20 µg ml⁻¹, Spectinomycin (Spc) 75 µg ml⁻¹.

DNA manipulation and analysis

Plasmid DNA and PCR products were purified using Qiagen Kits. All modifying and restriction enzymes were obtained from New England Biolabs (NEB). All constructs were verified by DNA sequencing. Analysis of DNA sequences was performed using Vector NTI (Invitrogen).

Generation of Mtb mutants

The mntR (Rv2788) and the mntH (Rv0924) mutants were generated using two-step homologous recombination with sacB counter-selection as described previously (Rodriguez et al., 2002). Plasmids pSM517 and pSM591 were used as recombination substrates to generate the mntR and the mntH mutant respectively (Table 1). These plasmids are derivatives of pSM270 a suicide vector that carries sacB and a Sm resistance cassette in the plasmid backbone (Rodriguez et al., 2002). pSM517 contained mntR, 200 bp upstream and 250 bp downstream with a Kan cassette insertion in the middle of mntR. pSM591 contained mntH interrupted by insertion of a Kan resistance cassette. Single cross over events were selected with Kan and Sm and after amplification without antibiotics, clones that carried out the second cross-over excising the intervening vector sequences including both sacB and the Sm cassette were selected in medium containing Kan and 8% sucrose (Sigma). Kan^R Suc^R Sm^S colonies were examined by PCR or Southern Blot to confirm allelic exchange.

Table 1. Plamids and bacterial strains.

Plasmid	Description	Source
pSM911	200–300bp upstream region of Rv1283 cloned into TOPO vector	This work
pSM912	200–300bp upstream region of Rv2055 cloned into TOPO vector	This work
pSM913	200–300bp upstream of Rv0924 cloned into TOPO vector	This work
pSM914	200–300bp upstream of Rv2059 cloned into TOPO vector	This work
pSM898	Rv1280 cloned into pSM232	This work
pSM871	Upstream and downstream region of Rv1280 cloned into pJSC284 vector	This work
Mtb Strains
ST70	mntR mutant (sirR::kan)	This work
ST74	mntR mutant complemented with wild type copy of mntR	This work
ST98	Rv0924 (mramp) mutant (mramp::kan)	This work
ST251	Rv1280c deletion mutant	This work
ST252	Rv1280c mutant complemented with wild type copy of Rv1280c	This work
ST267	Double mutant of Rv0924 and Rv1280c	This work
ST277	Double mutant of mntR and Rv1280c	This work

The Rv1280 deletion mutant was generated by specialized transduction and allelic exchange as previously (Pandey and Rodriguez, 2012).

A double mutant of mntR-Rv1280c and Rv1280-mntH was generated also by specialized transduction, introducing the Rv1280c deletion into the mntR or the mntH single mutants. Double mutants were selected in 7H10 agar containing Hyg. Each candidate mutant was verified by PCR amplification using primers in the Hyg cassette and the sequence flanking the predicted insertion site. PCR products were purified and sequenced to verify legitimate recombination.

Mutant complementation

To complement the mntR mutant, the wild type mntR gene with its own promoter was PCR amplified using the primers sirKO1 and sirKO4 (Table S1). The PCR product was cloned at the NotI-HindIII sites in pMV306, an integrative plasmid containing a Hyg resistance cassette. The resulting plasmid pSM542, was electroporated into the mntR mutant strain ST70, and transformants were selected on 7H10 containing Hyg. The complementing strain was named ST74 (Table 1).

To complement the Rv1280c mutant, Rv1280c was PCR amplified and cloned in front of the hsp60 promoter in pSM232, a derivative of pMV261 that contains a Sm/Spc resistance cassette.

RNA extraction

Mtb strains were grown to early logarithmic phase in MM without Mn²⁺ or supplemented with 50µM MnCl₂. Cells were collected by centrifugation and pellets were resuspended in 1 ml TRI reagent and immediately transferred to a tube containing 0.5 ml 200 µm zirconia beads (Sigma Aldrich) and disrupted by two 1 min pulses in a BeadBeater. RNA was purified as previously using RNeasy columns following the manufacturer instructions (Qiagen) (Rodriguez et al., 2002). The quality and quantity of purified total RNA was estimated using a Nanodrop spectrophotometer (Nanodrop, Wilmington, DE).

Microarray Analysis

The Mtb DNA microarray consisted of 4295 70-mer oligonucleotides representing the 3924 predicted ORFs of the H37Rv strain (http://www.sanger.ac.uk). The arrays were prepared by spotting oligonucleotides (Tuberculosis Genome Set version 1.0, Operon Biotechnologies) onto poly-l-lysine-coated glass microscope slides, using a GeneMachines Omnigrid 100 Arrayer (Genomic Solutions) and SMP3 pins (Telechem). Total RNA from three independent cultures in the same condition was prepared as described above. Briefly, cDNA was synthesized using random primers and labeled with Cyanine-3 or Cyanine-5 dUTP (PerkinElmer) as described before (Pang et al., 2007) and hybridized to the arrays overnight. After washing, the arrays were scanned with a GenePix4000B scanner (Molecular Devices). The images were processed using GenePix 5.1. Data were filtered by removing all spots that were not above the background noise. Spots were considered to be not sufficiently above background noise for further analysis if the sum of the median intensities of the two channels was less than twice the highest mean background of the chip. The chips were normalized by the print-tip Lowess method (Dudoit and Speed, 2000). The ratio of the mean median intensity of Cy5 over the mean median intensity of Cy3 was determined for each spot and the fold change values were calculated. A one-class SAM analysis (Tusher et al., 2001) was performed with the MEV software (Saeed AI et al., 2003) to find genes with changes that occurred consistently in all replicates. A median FDR (false discovery rate) of zero, delta values ranging from 3.14 to 3.63 and a mean change of at least two fold were considered our cut-off for significant.

The data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus (Edgar et al Ref 2002) and are accessible through GEO Series accession number GSE70812 (http://www.ncbi.njm.nih.gov/geo/query/acc.cgi?acc=GSE70812).

Quantitative real time PCR (qRT-PCR)

Complementary DNA was synthesized from 200 ng of RNA by reverse transcription using random hexamer primers and ThermoScript™ (Invitrogen) following the manufacturer instructions. Control mock reactions without ThermoScriptTM were prepared to exclude significant DNA contamination. After cDNA amplification, the samples were diluted 50-fold before PCR amplification. Quantitative PCR was performed in a Bio-Rad i-cycler using SYBR green qPCR superMix Universal (Invitrogen). PCR conditions were identical for all reactions. The 25 µl reaction mixture contained 6 ng of cDNA template, 12.5 µl of SYBR Green Master Mix, 2 µl primer mix (10 pmol of each primer) and 7.5 µl water. After 10 min at 94°C to activate the DNA polymerase, a set of 40 cycles of 30 s at 95°C, annealing at 55°C for 30 s and elongation at 72°C for 30 s was run. Primers used for qRT-PCR are listed in Table S1. Fluorescence was measured during the annealing step and plotted automatically for each sample. Transcript abundance was normalized to the amount of 16S ribosomal RNA. In order to obtain a standard curve for the RT-PCR, PCR was performed with each primer set by using calibrated amounts of chromosomal DNA and these reactions were run at the same time as the RT-PCR. Standard curves were generated and used to calculate the amount of cDNA for each gene present in the test sample.

Electrophoretic mobility shift assays (EMSA)

Purified Mtb MntR was used in EMSA with DNA probes derived from approximately 200 bp sequences upstream of Rv1283c, Rv2059, Rv2058c and mntH (Rv0924) and obtained by PCR amplification using primers Rv1283 prom.Fw and Rv1283 prom.Rev (for Rv1283c), Rv2058 prom.Fw and Rv2058 prom.Rev (for Rv2058), Rv2059 prom.Fw and Rv2059 prom.Rev (for Rv2059), Rv0924 prom.Rev and Rv0924 prom.Fw (for Rv0924) (Table S1). These PCR fragments were end-labeled with biotinylated ribonucleotides using 3’ End DNA labeling Kit (Thermo Scientific) following the manufacturer’s instructions or, where indicated, were detected by staining with SYBR Green. Binding reactions contained binding buffer (20 mM Tris-HCL pH 8.0, 1 mM DTT, 50 mM KCL, 5 mM MgCl₂, 0.05 mg ml⁻¹ poly(dI-dC), 0.05 mg ml⁻¹ BSA and 10% glycerol), 20 ng DNA probe (8 nM) and purified MntR in a 20 µl final volume. Where specified, various metal salts were added in the reaction at the indicated concentrations. The reactions were incubated for 40 min at room temperature. Samples were separated in an 8% native polyacrylamide gel containing 40 mM Tris acetate pH 8.0. Electrophoresis was conducted at 100 V and 4°C and the gel was transferred to positively charged Nylon transfer membrane (Amersham HybondD™-N⁺) (GE Healthcare) at 15 V for 45 min. Biotinylated DNA probes were detected using enhanced luminal substrate for Streptavidin-horseradish peroxidase (HRP) from Chemiluminescent Nucleic Acid Detection Module (Pierce).

Mutagenesis

Site directed mutagenesis was conducted using XL-Gold Quick change (Stratagene) as recommended by the manufacturer.

Transcriptional start site mapping

The transcriptional start site of Rv1283c, Rv2058c and Rv2059 was identified by 5’ rapid amplification of c-DNA ends (RACE) using 5’/3’ RACE kit (Roche) and gene specific primers as follows: Rv1283 RACE primer1 and Rv1283 RACE primer2 (for Rv1280), Rv2058 RACE primer1 and Rv2058 RACE primer2 (for Rv2055), Rv2059 RACE primer1 and Rv2059 RACE primer2 (for Rv2059) as listed in Table S1 and following the manufacturer instructions.

Stress sensitivity assays

Sensitivity to H₂O₂, plumbagin and CdCl₂ was tested in zone inhibition assays. Mtb strains were grown to logarithmic phase (OD₅₄₀ 0.5) in 7H9. A solution containing approximately 3×10⁷ bacteria was spread evenly on 7H10 agar. A 6.5 mm paper disk saturated with 10 µl of 500 mM H₂O₂, 5mM plumbagin or 200 mM CdCl₂ was placed on the center of the plate. After incubation for 10 days, the diameter of the halo of growth inhibition generated by each compound was measured.

Superoxide dismutase activity assay

Mtb strains were grown to mid logarithmic phase in MM with or without 200 µM MnCl₂ in the presence of 50 µM FeCl₃ and harvested by centrifugation. The pellet was washed twice with cold 50 mM sodium phosphate pH 8.0 and resuspended in the same buffer. Cells were broken by three 30 s pulses in a bead beater with 200 µm zirconia silica beads (Sigma Aldrich) and cell debris were removed by centrifugation. Protein concentration in the lysate was determined using Bradford DC protein assay (BioRad). SOD activity was assayed based on the ability of SOD to inhibit the reduction of nitro-blue tetrazolium (NBT) by superoxide. The method described by Beauchamp and Fridovich (Beauchamp and Fridovich, 1971) was used with minor modifications. Briefly, a reaction mix was prepared by adding 1.2 mM NBT, 14 µM riboflavin and 14 mM tetramethylethylenediamine (TEMED) in 50 mM sodium phosphate buffer pH 8.0. Simultaneously a blank reaction was prepared with 1.2 mM NBT in 50 mM sodium phosphate buffer pH 8.0. Reactions were protected from light. 50 µg of lysate was added to 140 µl of the reaction or blank mix and the volume was taken up to 240 µl with 50 mM sodium phosphate buffer pH 8.0. Reactions were exposed to light for 5 min and absorbance at 560 nm was examined. The percentage inhibition of NBT reduction was calculated. SOD activity was determined based on a standard curve obtained with increasing concentrations of purified SOD (Sigma Aldrich). SOD activity was expressed as SOD units (one unit of SOD inhibits NBT’s reduction by 50%) per mg protein.

Cellular metal measurements

Four milliliters of mid-log-phase cells was harvested and washed once with phosphate-buffered saline (PBS) containing 1 mM nitrilotriacetic acid, followed by two washes with PBS. Cells were resuspended in 400 µl PBS, from which 50 µl was used for an OD₆₀₀ measurement. Ten microliters of 10 mg ml⁻¹ lysozyme was added to the remaining cells and incubated at 37°C for 20 min. Six hundred microliters of 5% HNO₃ with 0.1% (v/v) Triton X-100 was then added, and the samples were boiled at 95°C for 30 min. Lysates were then centrifuged, and the supernatants were used for metal content analysis by inductively coupled plasma mass spectrometry (ICP-MS).

Purification of MntR protein

MntR (SirR) protein was purified from an E. coli overexpressing strain by the TB structural consortium. The tag was removed and the protein was sent to us. We confirmed the purity of the protein by electrophoresis in an SDS poly acrylamide gel and staining with Sypro Ruby. The identity of the MntR protein was verified by LC-MS.

Macrophage infections

Infection of THP-1 human monocytic cells was performed as previously (Rodriguez and Smith, 2006). Briefly 1×10⁵ THP-1 cells per well were induce to differentiate by 24 h treatment with 50 nM phorbol myristate acetate and infected with Mtb strains, pre-grown to logarithmic phase in 7H9 medium. A low multiplicity of infection 1:50 bacterium per macrophages was used. After 4 h of incubation at 37°C in 5% CO₂ atmosphere, extracellular bacteria were removed by three washes with warm PBS. Pre-warm RPMI was added to each well and the plate was incubated at 37°C in 5% CO₂ atmosphere. RPMI was replaced every 48 h. At indicated time points post infection, triplicate wells for each Mtb strain infection were treated with 0.05% sodium dodecyl sulphate (SDS) to lyse the macrophages and the numbers of CFU were determined by plating serial dilutions on 7H10 plates.

Mice infection

For each strain tested a 10 ml bacterial suspension of 1×10⁶ bacilli/ml in saline containing 0.04% Tween 80 was used. Aerosols were generated with a Lovelace nebulizer (In-tox Products, Alburquerque, NM) and C57BL/6 female mice were exposed to the aerosol for 30 minutes. Under these conditions the number of microorganisms detected in the lungs at time 0 (4hr post infection) was approximately 100. At the indicated time points after infection, 3 mice for each strain were sacrificed; lungs, spleen and liver were removed and homogenized in PBS-Tween 80. Dilutions of the homogenates were plated on 7H10 agar to determine number of CFUs.

Live/dead Staining

Bacterial cultures grown in minimal medium with 1 µM Mn²⁺ were exposed to 200 µM Mn for 24 hr. Cells were collected by centrifugation and resuspended in 1 ml of 0.9% saline with 0.2%) tween 20 and stained with Live/Dead BacLight Bacterial viability kit Molecular probes (Life Technology) according to the manufacturer instructions.