The Cation Diffusion Facilitator Family Protein EmfA Confers Resistance to Manganese Toxicity in Brucella abortus 2308 and Is an Essential Virulence Determinant in Mice

Mn nutrition is essential for the basic physiology and virulence of Brucella strains. The results of the study presented here demonstrate that the cation diffusion facilitator (CDF)-type metal exporter EmfA plays critical roles in maintaining Mn homeostasis and preventing Mn toxicity in Brucella and is an essential virulence determinant for these bacteria. EmfA and other cellular components involved in Mn homeostasis represent attractive targets for the development of improved vaccines and chemotherapeutic strategies for preventing and treating brucellosis in humans and animals.

ABSTRACT

The gene designated bab_rs23470 in the Brucella abortus 2308 genome encodes an ortholog of the cation diffusion facilitator family protein EmfA which has been linked to resistance to Mn toxicity in Rhizobium etli. A B. abortus emfA null mutant derived from strain 2308 displays increased sensitivity to elevated levels of Mn in the growth medium compared to that of the parent strain but wild-type resistance to Fe, Mg, Zn, Cu, Co, and Ni. Inductively coupled plasma mass spectroscopy also indicates that the B. abortus emfA mutant retains significantly higher levels of cellular Mn after exposure to this metal than the parent strain, which is consistent with the proposed role of EmfA as a Mn exporter. Phenotypic analysis of mutants indicates that EmfA plays a much more important role in maintaining Mn homeostasis and preventing the toxicity of this metal in Brucella than does the Mn-responsive transcriptional regulator Mur. EmfA is also an essential virulence determinant for B. abortus 2308 in C57BL/6 and C57BL/6^Nramp1+/+ mice, which suggests that avoiding Mn toxicity plays a critical role in Brucella pathogenesis.

IMPORTANCE Mn nutrition is essential for the basic physiology and virulence of Brucella strains. The results of the study presented here demonstrate that the cation diffusion facilitator (CDF)-type metal exporter EmfA plays critical roles in maintaining Mn homeostasis and preventing Mn toxicity in Brucella and is an essential virulence determinant for these bacteria. EmfA and other cellular components involved in Mn homeostasis represent attractive targets for the development of improved vaccines and chemotherapeutic strategies for preventing and treating brucellosis in humans and animals.

INTRODUCTION

Metals are essential trace nutrients for all bacteria. The most widespread use of metals is to support the structure and catalytic function of cellular proteins. Numerous important physiological processes, including DNA synthesis, metabolic pathways, and energy generation, rely on enzymes with metal cofactors to function properly (1). To accommodate the portion of their proteome that requires metal ions, bacteria employ high-affinity metal acquisition systems to import bioavailable metals from their immediate environment. Obtaining sufficient intracellular levels of certain metals such as iron (Fe), manganese (Mn), and zinc (Zn) is especially difficult for bacteria that are mammalian pathogens because these bacteria must overcome the so-called “metal withdrawal defenses” of their hosts, which actively limit the availability of metal nutrients (2). Accordingly, high-affinity Fe, Mn, and Zn acquisition systems typically represent essential virulence determinants for bacterial pathogens (3–6).

Despite their role as critical micronutrients, metals become toxic if accumulated in excess. Metal-induced toxicity is largely the result of surplus cytosolic metal atoms outcompeting other metal species for their native protein binding sites (7). Improper protein metalation can lead to the loss of the affected proteins’ functions, effectively disrupting the many cellular functions carried out by metal-dependent proteins. Computational studies estimate that roughly one-third of bacterial proteins require metal cofactors (8), which further emphasizes the extent to which metal toxicity causes harm. Other mechanisms of metal toxicity are specific to individual metal species. Excess accumulation of Fe, for example, results in the generation of harmful oxidative radicals by means of intracellular Fenton chemistry (9).

Bacteria actively regulate intracellular metal homeostasis by means of high-affinity metal importers, metal-responsive transcriptional regulators, and metal efflux proteins (1). These systems operate in a concerted manner to both maintain sufficient metal nutrition and prevent excess accumulation of intracellular metals. Metal-responsive transcriptional regulators, for instance, control the expression of metal acquisition systems to ensure these systems operate only under conditions of metal deprivation and ensure that metal exporters are only functional under conditions of metal excess. Furthermore, bacterial species encode metal-specific chaperones to facilitate proper protein metalation (10) and storage proteins to protect against uncontrolled reactivity of intracellular metals (11).

The Brucella spp. are Gram-negative bacteria that cause abortion and infertility in animals (12). Humans are also at risk of infection as brucellosis is the world’s most prevalent zoonotic disease, and this disease is endemic in areas of the world where brucellosis in food animals is not controlled by efficient eradication programs (13). The metal manganese (Mn) is an essential micronutrient for Brucella strains (14), presumably because several cellular proteins that are required for the basic physiology and virulence of these bacteria are Mn dependent (15, 16). To resist Mn deprivation, Brucella strains rely on a single high-affinity Mn importer, MntH (17), but the mechanisms that these bacteria use to resist Mn toxicity are unknown. This is an important consideration because Mn, magnesium (Mg), and iron (Fe) can be interchangeable in their interactions with proteins (18), and excess intracellular Mn has been shown to interfere with the physiologic functions of bacterial proteins such as isocitrate lyase and Fur, which require Mg or Fe for their normal enzymatic or regulatory activities (19–21). Excess Mn has also been shown to interfere with Mg transport in bacteria (19, 22). Expression of the mntH gene is repressed by a prototypical Fur-like transcriptional repressor, Mur, in response to increasing intracellular Mn concentrations (23). The role that Mur plays in preventing Mn toxicity in Brucella, however, has not been experimentally determined. The present study sought to understand how Brucella abortus 2308 copes with increasing intracellular Mn and determine the importance of preventing Mn toxicity during host infection.

RESULTS

A B. abortus mur mutant does not exhibit increased sensitivity to excess environmental Mn.

The B. abortus mur mutant EAM001 was grown on medium supplemented with increasing levels of Mn to examine the role of Mur in this bacterium’s resistance to Mn toxicity. As shown in Fig. 1, the B. abortus mur mutant displayed the same level of resistance to excess Mn in the growth medium as the parental 2308 strain. To further examine Mur’s contribution to Mn homeostasis in Brucella, inductively coupled plasma mass spectrometry (ICP-MS) was employed to measure cellular Mn levels of B. abortus 2308 and the mur mutant during growth in a rich medium and after exposure to exogenous Mn in this medium. As shown in Fig. 2, the parent strain and mur mutant exhibited similar cellular levels of Mn under both conditions. These data suggest that Mur plays a limited role in protecting B. abortus 2308 from Mn toxicity.

FIG 1.

B. abortus 2308 does not require Mur for growth in the presence of excess exogenous Mn. Twenty-five-microliter suspensions of B. abortus 2308 and the isogenic mur mutant EAM001 were placed on Schaedler agar plates supplemented with MnCl₂, and growth was observed following 72 h of incubation at 37°C with 5% CO₂. The pictures shown represent growth in a single representative experiment. One plate containing each level of metal was examined in each individual experiment, and the experiment was repeated 3 times with the same results.

FIG 2.

The B. abortus mur mutant contains similar cellular Mn levels as the wild type. The cellular Mn levels of B. abortus 2308 and the mur mutant EAM001 were measured following growth to mid-log phase in brucella broth (A) and a subsequent 2-h exposure to 50 μM MnCl₂ in this medium (B). Cellular Mn content was determined by ICP-MS, and cellular protein levels were determined using the Bradford assay. The data were evaluated using one-way analysis of variance (ANOVA) and Tukey’s multiple-comparison test, and no statistically significant differences were detected between the parent strain and emfA mutant.

An ortholog of the Rhizobium EmfA protein confers resistance to Mn toxicity in Brucella.

The gene designated bab_rs23470 in the B. abortus 2308 genome is predicted to encode a 302-amino-acid protein that shares 64% identity and 76% similarity at the amino acid sequence level with the EmfA protein of Rhizobium etli (24). EmfA is a clade VI member of the cation diffusion facilitator (CDF) family of metal exporters, and phenotypic analysis of an R. etli emfA mutant indicates that this protein plays an important role in protecting this bacterium from Mn toxicity. To determine if the Brucella EmfA ortholog performs a similar function, B. abortus 2308 and the isogenic emfA mutant JEP061 were grown on Schaedler agar plates supplemented with increasing levels of Mn (Fig. 3). The B. abortus emfA mutant exhibited growth restriction at considerably lower levels of exogenous Mn in this assay than the parental 2308 strain, and genetic complementation of the emfA mutant with a plasmid-borne copy of the emfA gene restored the resistance of this mutant to Mn to the same level as that of 2308. Notably, a B. abortus emfA mur double mutant displayed the same level of sensitivity to Mn as the emfA mutant in this assay.

FIG 3.

EmfA protects B. abortus 2308 from Mn toxicity. Twenty-five-microliter suspensions of B. abortus 2308 and derivative strains were placed on Schaedler agar plates supplemented with MnCl₂, and growth was observed following 72 h of incubation at 37°C with 5% CO₂. The pictures shown represent growth in a single representative experiment. One plate containing each level of metal was examined in each individual experiment, and the experiment was repeated 4 times with the same results.

The concentration of Mn that inhibits the growth of the B. abortus emfA mutant in the plate assays shown in Fig. 3 (e.g., >100 μM) is higher than the tissue concentrations of this metal that have been reported for the spleens and livers of C57BL/6 mice, which range from 35 to 90 μM (25, 26). Thus, to assess the comparative resistance of B. abortus 2308 and the emfA mutant to more physiologically relevant levels of Mn, these strains were examined for their capacity to grow in brucella broth and this medium supplemented with 10, 20, and 50 μM MnCl₂. Notably, the emfA mutant exhibited growth restriction in brucella broth containing 50 μM MnCl₂ (Fig. 4) in this assay, but the parental 2308 strain did not.

FIG 4.

The B. abortus emfA mutant is more sensitive to growth restriction by exogenous Mn than the parental 2308 strain during cultivation in brucella broth. The results presented are the OD₆₀₀ values obtained for 3 cultures of each B. abortus strain under each experimental condition after 40 h of growth in brucella broth supplemented with the indicated concentration of Mn in a single experiment. The experiment was repeated 3 times with the same results.

The designation “Emf” stands for efflux of Mn and Fe, and EmfA homologs have been proposed to be able to transport both Mn and Fe out of bacterial cells (24). To examine the possibility that EmfA is able to protect Brucella strains from toxic levels of metals other than Mn, B. abortus 2308 and the emfA mutant were grown on agar plates supplemented with increasing levels of Fe, Mg, Zn, Ni, Co, and Cu. As shown in Fig. 5, both strains displayed equivalent levels of resistance to these metals. This suggests that EmfA plays a specific role in protecting Brucella from Mn toxicity.

FIG 5.

EmfA does not protect B. abortus 2308 from Fe, Mg, Zn, Ni, Co, or Cu toxicity. Twenty-five-microliter suspensions of B. abortus 2308 and the isogenic emfA mutant JEP61 were placed on Schaedler agar plates supplemented with FeCl₃, MgCl₂, ZnSO₄, NiCl₂, CoCl₂, or CuCl₂, and growth was observed following 72 h of incubation at 37°C with 5% CO₂. The pictures shown represent growth in a single representative experiment. One plate containing each level of metal was examined in each individual experiment, and the experiment was repeated 3 times with the same results.

The phenotype of the B. abortus emfA mutant (increased sensitivity to Mn) is consistent with the proposed function of the corresponding gene product as a CDF-type Mn exporter. To evaluate this proposed function more directly, we used ICP-MS to measure cellular Mn levels in B. abortus 2308, the emfA mutant, and complemented mutant during routine cultivation in brucella broth (a rich medium) and following a brief exposure to 50 μM MnCl₂ in this growth medium. The B. abortus emfA mutant exhibited significantly higher levels of cellular Mn following exposure to this metal in brucella broth than either the parent strain or complemented mutant (Fig. 6). It is also interesting to note that the levels of Mn detected in the complemented emfA mutant MJJ012 were significantly lower under these conditions than those found in the parent strain after exposure to exogenous Mn. This is presumably the result of the increased number of copies of emfA in the complemented mutant. Specifically, pMR10 was used for this complementation, and this plasmid is maintained at 2 to 4 copies per genome in Brucella strains (27). Both experimental findings support the proposed function of EmfA as a Mn exporter in Brucella.

FIG 6.

The B. abortus emfA mutant has increased cellular Mn content following Mn challenge compared to that of B. abortus 2308. Cellular Mn levels were measured in B. abortus 2308, the emfA mutant JEP61, and the complemented emfA mutant MJJ012 (ΔemfA^C) following growth to the mid-log phase in brucella broth (A) and following a subsequent 2-h exposure to 50 μM MnCl₂ in this medium (B). ICP-MS was used to measure cellular Mn levels, and the Bradford assay used to determine cellular protein levels. *, P < 0.05; ***, P < 0.005 for comparisons of 2308 versus the emfA mutant or the complemented emfA mutant using one-way ANOVA and Tukey’s multiple-comparison test.

EmfA is required for the wild-type virulence of B. abortus 2308 in mice.

Bacterial CDF-type Mn exporters have been linked to virulence (28–30). The B. abortus emfA mutant also exhibited growth restriction when exposed to a physiologically relevant level of Mn in an in vitro assay (Fig. 4). To assess the importance of EmfA for the virulence of B. abortus 2308, spleen colonization profiles of this strain, the emfA mutant, and complemented mutant were examined in C57BL/6 and C57BL/6^Nramp1+/+ mice (31). The divalent cation transporter Nramp1 has been shown to be important for restricting Mn availability to bacterial pathogens during their residence in host macrophages (32), and long-term intracellular persistence in macrophages is a key determinant of Brucella virulence (33). Notably, the B. abortus emfA mutant displayed attenuation in both strains of mice, but the attenuation was most pronounced in the mice that harbor a functional Nramp1 (Fig. 7).

FIG 7.

The B. abortus emfA mutant displays attenuation in mice. Spleen colonization profiles of Brucella abortus 2308, the emfA mutant, and the complemented emfA mutant (ΔemfA^C) in C57BL/6 and C57BL/6^Nramp1+/+ mice are shown. Ten mice were individually infected with 5 × 10⁴ of each bacterial strain via the intraperitoneal route, and five mice from each experimental group were evaluated at 2 weeks and 5 weeks postinfection. *, P < 0.05; ***, P < 0.005 for comparisons of 2308 versus the emfA mutant or complemented emfA mutant using one-way ANOVA and Tukey’s multiple-comparison test.

DISCUSSION

Many bacteria require both Mn-responsive transcriptional regulators and Mn efflux proteins to maintain proper Mn homeostasis and prevent Mn toxicity (21, 34, 35). To date, Mur is the only Mn-responsive regulator that has been described in Brucella (23). With no evidence suggesting that Mur is required for the resistance of B. abortus 2308 to Mn toxicity, the possibility that this bacterium employs additional factors to prevent the overaccumulation of this metal was examined. The studies reported here demonstrate that the previously uncharacterized B. abortus 2308 bab_rs23470 gene encodes a functional homolog of the Rhizobium Mn exporter EmfA and that this protein is essential for this strain’s wild-type resistance to Mn toxicity.

To our knowledge, only two other EmfA orthologs (R. etli EmfA and Sinorhizobium meliloti SmYiiP) have been characterized, and in all three cases (including this study), the loss of a functional EmfA (or SmYiiP) caused the corresponding mutants to display increased sensitivity to Mn toxicity (24, 36). These proteins belong to the clade VI CDF-type metal exporters described by Cubillas et al. (24). The Emf (efflux of Mn/Fe) designation of the prototype R. etli protein is based on the fact that transcription of the corresponding gene is elevated in response to exposure to both Mn and Fe and that heterologous expression of the R. etli emfA gene provides Fe resistance to an Escherichia coli zitB zntA mntP mutant. However, studies with a purified version of the S. meliloti EmfA homolog (SmYiiP) incorporated into lipid vesicles demonstrated that this protein is a Mn-specific transporter (36), and the observation that R. etli (24), S. meliloti (36), and B. abortus emfA mutants (this study) display increased susceptibility to Mn but not Fe supports this proposition. Consequently, it will be important to use biochemical and genetic analyses to determine whether the Brucella EmfA transports Fe in its native setting.

Mammals employ “metal withdrawal” defenses that are essential for these organisms to combat pathogenic bacteria during infection (2). Accordingly, host factors that sequester free Mn play an important role in the host immune response because they deprive these bacteria of bioavailable Mn in both the extracellular and intracellular environments (37, 38). For Brucella strains, ingestion by host macrophages is an essential component of their pathogenesis, as these bacteria naturally replicate within and require macrophages to sustain host infection (33). The severe attenuation of a B. abortus mntH mutant in mice and cultivated macrophages suggests that this bacterium requires sufficient Mn nutrition throughout the course of an infection (17). The evidence presented in this study indicates that Mn efflux is equally important to B. abortus virulence as Mn acquisition and is also consistent with the fact that Mn efflux proteins are critical virulence determinants for other pathogenic bacteria (28–30).

C57BL/6 mice lack a functional Nramp1 due to a substitution of aspartate for glutamate at amino acid residue 169 of this protein (39). The pronounced attenuation demonstrated by the B. abortus emfA mutant in mice that encode a functional copy of Nramp1 relative to those that do not is of considerable interest, because this finding appears to be in conflict with the proposed role that Nramp1 plays as a host defense against other intracellular pathogens. Specifically, studies have shown that Nramp1-mediated efflux of Mn and other transition metals from the phagosomal compartments of macrophages limits the availability of essential metal nutrients at the host-pathogen interface (32). Thus, the Brucella-containing vacuoles (BCVs) in the macrophages of the C57BL/6^Nramp1+/+ mice would be expected to have less Mn than the BCVs in macrophages from the C57BL/6 mice, a situation where EmfA should not provide a fitness benefit for the brucellae. One possible explanation for this seemingly paradoxical phenotype of the B. abortus emfA mutant in C57BL/6^Nramp1+/+ mice is that the capacity of Nramp1 to transport both Mn and Fe out of the phagosomal compartment may disrupt the Mn/Fe balance in the Brucella cells that inhabit the macrophages of these mice. The intracellular concentrations of these two metals are codependent in other bacteria (40); therefore, it is logical to consider that the Mn efflux activity of B. abortus EmfA would enable these bacteria to reduce intracellular Mn levels in order to maintain a proper balance in this bacterium’s Mn/Fe ratio. This proposal, in part, is supported by the fact that the sensitivity of a Neisseria meningitidis mutant lacking the LysE-type Mn exporter MntX to Mn toxicity depends upon environmental Fe levels (40). Specifically, the sensitivity of this mutant to Mn toxicity becomes more intense at lower Fe levels. It is also important to note that it is presently unclear how important Nramp1 is for the resistance of mice to Brucella infections (41) and that Nramp1 has been shown to play roles in host defense that are not directly linked to its capacity to deprive intracellular pathogens of metals (42). Thus, further studies will be required to accurately assess the relationship between host Nramp1 and bacterial EmfA activity and the impact that this relationship has on Brucella pathogenesis.

The transcription of genes encoding Mn efflux proteins is regulated by Mn-responsive transcriptional regulators in many bacteria (21, 43). The Mn-responsive transcriptional network of these bacteria typically relies on a single “Mn-sensing” transcriptional regulator that inversely regulates genes involved in Mn uptake and efflux. This enables these bacteria to strategically activate or repress these genes in response to Mn availability. Preliminary studies suggest that emfA expression is not Mn responsive in B. abortus 2308, and electrophoretic mobility shift assays indicate that this gene is not a direct target of Mur (M. J. Johnsrude, unpublished data). These experimental findings suggest that the Brucella emfA may be transcribed constitutively in the same fashion as reported for the Streptococcus pneumoniae mntE (44). The efflux activity of many CDF-type metal exporters is regulated at the posttranslational level (45). If the Brucella EmfA is also regulated in this manner, it would conceivably allow these bacteria to coordinate their Mn import and export in response to cellular levels of this metal in lieu of Mn-responsive regulation of emfA transcription.

The lack of a Mn-sensitive phenotype in the B. abortus mur mutant is inconsistent with previous studies examining Mn-responsive transcriptional regulators employed by other bacteria (46, 47). Specifically, the studies presented here indicate that Mur does not play a major role in resisting Mn toxicity in Brucella and raise questions about the precise physiologic function of this transcriptional regulator. Bioinformatics (48) and microarray analyses (J. E. Pitzer, unpublished data) suggest that the Mur regulon in Brucella may be limited to only two genes, mntH and a gene (bab_rs17810 in B. abortus 2308) that encodes an uncharacterized Fur homolog. If this Fur homolog is in fact a metal-responsive transcriptional regulator, it is conceivable that overlapping metal-responsive regulation by this regulator and Mur interferes with an accurate assessment of Mur’s regulatory function. Consequently, it will be important to define the function of the bab_rs17810 gene product and determine how it impacts the regulatory function of Mur to better understand the physiologic roles that these proteins play in Brucella.

MATERIALS AND METHODS

Bacterial strains and culture conditions.

The B. abortus strains used in this study are listed in Table 1. These strains were routinely cultivated on Schaedler agar supplemented with 5% defibrinated bovine blood (SBA) at 37°C with 5% CO₂ or in brucella broth at 37°C with shaking at 250 rpm. Brucella stock cultures were maintained in brucella broth supplemented with 25% glycerol and stored at −80°C.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Genotype or description	Reference or source
Brucella abortus strains
2308	Virulent challenge strain	53
EAM001	2308 Δmur	This study
JEP61	2308 ΔemfA	This study
MJJ012	JEP61 with a plasmid-borne emfA	This study
JEP62	2308 Δmur ΔemfA	This study
Plasmids
pNPTS138	sacB-based gene replacement vector; kanamycin resistance	54
pGEM-T Easy	Cloning vector; ampicillin resistance	Promega
pMR10	Cloning vector; broad-host range; low copy no. (2 to 4 copies per cell); kanamycin resistance	27
pMJ12	pMR10-based plasmid carrying the B. abortus emfA gene; kanamycin resistance	This study

All work with live Brucella strains was performed in a biosafety level 3 laboratory certified by the Centers for Disease Control and Prevention’s Division of Select Agents and Toxins.

Construction of the B. abortus mur, emfA, and mur emfA mutants.

Mutants containing in-frame deletions that fuse the first two codons to the last two codons of the mur (bab_rs23905) and emfA (bab_rs23470) coding regions were constructed from B. abortus 2308, and the genotypes of these mutants were confirmed using the methods described by Caswell et al. (49). This same approach was used to introduce an emfA mutation into the B. abortus mur mutant EAM001. The nucleotide primers used for the construction of the pNTPS138-based plasmids used for making these mutants are shown in Table 2.

TABLE 2.

Oligonucleotide primers used in this study

Primer	Sequence (5′→3′)
emfA up F BamH	GATGGATCCGAAGGCAGGTTGACGATCAGCTTGTTG
emfA down R PstI	GATCTGCAGCCTTCGTCCATACGCTTGCCAATTATTT
emfA up R	TTCGCCTGAGCCAAATTACCATC
emfA down F	GTCCATAATGAACCTTTGGTTATCTTG
emfA comp EcoRI-R	GATGAATTCTGGCTCAGGCGAACGGAACC
emfA promoter PstI F	GATCTGCAGTAACTTGAAGCCAGCCGCAGCCAAACCTT
mur up F PstI	GATCTGCAGCATGGCCATCGGCAATCAGGGACTGAA
mur up R	GTTCATGGATCGATTATGCACTGCATCGCCATCAGTTGC
mur down F	TCCTGACGGAACCAATCCAGGAAGTTGGCGTTTTACGG
mur down R EcoRI	GATGAATTCTCGAACTGCTGATCAATGACG

Genetic complementation of the B. abortus emfA mutant.

The nucleotide primers emfA promoter PstI F and emfA comp EcoRI-R (Table 2) were used to amplify an intact copy of the emfA gene (bab_rs23470) from B. abortus 2308 genomic DNA using PCR, and this DNA fragment was cloned into pMR10 using the procedures described by Caswell et al. (49). The resulting plasmid was given the designation pMJ12, and this plasmid was introduced into B. abortus JEP61 by electroporation (50). The B. abortus JEP61 derivative with the plasmid-borne copy of emfA was designated MJJ012 (Table 1).

Growth of B. abortus strains in metal-supplemented media.

For the metal toxicity assays employing solid media, Brucella abortus strains were cultivated on SBA for 48 h and harvested in phosphate-buffered saline (PBS). The cell suspensions were adjusted to a final cell density of 10⁴ CFU/ml, and 25 μl of the bacterial suspensions was placed onto Schaedler agar plates supplemented with various concentrations of MnCl₂, FeCl₃, ZnSO₄, NiCl₂, MgCl₂, CoCl₂, or CuCl₂. After 72 h of incubation on these plates at 37°C with 5% CO₂, the relative growth of each strain was examined.

For the Mn toxicity assays in liquid media, bacterial cell suspensions grown overnight in brucella broth were adjusted to a cell density of 10⁴ CFU/ml in 5 ml of this medium or this medium supplemented with a final concentration of 10, 20, or 50 μM MnCl₂ in 17-mm by 100-mm snap-cap polystyrene tubes. These tubes were incubated at 37°C with shaking at 250 rpm, and the optical density at 600 nm (OD₆₀₀) of each culture was determined after 40 h of incubation.

Inductively coupled plasma mass spectrometry for measurement of total cell-associated Mn.

Fifty milliliters of brucella broth in 500-ml flasks was inoculated with 10⁴ B. abortus cells, and these cultures were incubated at 37°C with shaking at 250 rpm. When these cultures reached mid-log phase, 5 ml was removed from each culture for determining the baseline cellular concentration of Mn. MnCl₂ was then added to each bacterial culture to a final concentration of 50 μM. The cultures were incubated with shaking for an additional 2 h, and 5 ml of the cultures was collected. Bacterial cells were harvested by centrifugation at 12,000 × g for 10 min at 4°C and washed 1× in 10 ml sterile distilled water (dH₂O), and the resulting bacterial pellets were hydrolyzed in 1 ml nitric acid and stored at −80°C. Prior to removal from the biosafety level 3 (BSL3) laboratory, the hydrolyzed bacterial cell preparations were boiled for 30 min, and portions of the treated cell preparations were incubated in sterile brucella broth for 7 days to ensure that no live organisms remained in these preparations. ICP-MS analysis of samples was performed by the Environmental and Agricultural Testing Service at North Carolina State University using a Perkin Elmer Elan DRCII mass spectrometer. The total cellular protein concentrations of the bacterial cultures before and after treatment with Mn were determined using the Bradford assay.

Experimental infection of mice.

The virulence of the B. abortus strains in mice was evaluated using the methods described by Elzer et al. (51). Briefly, bacterial strains were grown for 3 days on Schaedler agar supplemented with 5% defibrinated bovine blood (SBA) under 5% CO₂, and the bacterial cells were harvested in PBS. Individual mice were inoculated with 5 × 10⁴ brucellae in 100 μl PBS via the intraperitoneal route, and at 2 and 5 weeks postinfection, the mice were euthanized by isoflurane overdose. The number of brucellae present in the spleens of mice was determined by homogenizing these organs and serial dilution and plating of the homogenates on SBA.

Female C57BL/6 mice were purchased from Jackson Laboratories and infected at 4 weeks of age. C57BL/6^Nramp1+/+ mice were obtained as breeding pairs from Renee Tsolis at the University of California at Davis. Both sexes of C57BL/6^Nramp1+/+ mice ranging in age from 4 to 9 weeks were used for the experimental infections and obtained from a breed colony maintained at ECU. The genotype of the C57BL/6^Nramp1+/+ mice was confirmed by PCR analysis of DNA obtained from tail snips (52). The animal use protocols under which these experiments were performed were reviewed and approved by the East Carolina University Animal Care and Use Committee.