Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Feb 1.
Published in final edited form as: Mol Microbiol. 2010 Jan 13;75(4):900–909. doi: 10.1111/j.1365-2958.2009.07025.x

Genetic analysis of petrobactin transport in Bacillus anthracis

Paul E Carlson Jr 1, Shandee D Dixon 1, Brian K Janes 1, Katherine A Carr 1, Tyler D Nusca 1,2, Erica C Anderson 1, Sarra E Keene 1, David H Sherman 1,2, Philip C Hanna 1,*
PMCID: PMC2917102  NIHMSID: NIHMS221984  PMID: 20487286

Summary

Iron acquisition mechanisms play an important role in the pathogenesis of many infectious microbes. In Bacillus anthracis, the siderophore petrobactin is required for both growth in iron depleted conditions and for full virulence of the bacterium. Here we demonstrate the roles of two putative petrobactin binding proteins FatB and FpuA (encoded by GBAA5330 and GBAA4766, respectively) in Bacillus anthracis iron acquisition and pathogenesis. Markerless deletion mutants were created using allelic exchange. The ΔfatB strain was capable of wild-type levels of growth in iron depleted conditions, indicating that FatB does not play an essential role in petrobactin uptake. In contrast, ΔfpuA bacteria exhibited a significant decrease in growth under low iron conditions when compared to wild-type bacteria. This mutant could not be rescued by the addition of exogenous purified petrobactin. Further examination of this strain demonstrated increased levels of petrobactin accumulation in the culture supernatants, suggesting no defect in siderophore synthesis or export but, instead, an inability of ΔfpuA to import this siderophore. ΔfpuA spores were also significantly attenuated in a murine model of inhalational anthrax. These results provide the first genetic evidence demonstrating the role of FpuA in petrobactin uptake.

Keywords: anthrax, iron transport, petrobactin

Introduction

Bacillus anthracis, the causative agent of anthrax, is a Gram-positive, spore-forming bacterium that readily infects a variety of mammals. In the environment, B. anthracis exists primarily as dormant spores, which are the infectious form of this microorganism (Dixon et al., 1999). Spores can enter a mammalian host through three distinct routes: cutaneous, gastrointestinal, or inhalational, each leading to distinct symptoms and disease course (Dixon et al., 1999, Inglesby et al., 1999, Inglesby et al., 2002, Doganay & Welsby, 2006). Infection through the inhalational route leads to the most severe and potentially fatal form of the disease (Dixon et al., 1999, Inglesby et al., 1999, Inglesby et al., 2002, Doganay & Welsby, 2006). Upon entry into the host, spores are phagocytosed by resident macrophages and transported to the regional lymph nodes (Dutz & Kohout, 1971, Dutz & Kohout-Dutz, 1981, Guidi-Rontani et al., 1999, Dixon et al., 2000). While in the intracellular environment, B. anthracis spores germinate, transforming into rapidly dividing vegetative cells (Dixon et al., 2000). It is thought that these bacilli begin to multiply within macrophages prior to the induction of cell death and escape from these immune cells (Dixon et al., 2000). Once out of the macrophages, B. anthracis vegetative cells continue to grow rapidly in the bloodstream. Although symptoms during these early stages are relatively mild, the bacteria soon replicate to very high levels, approaching 108 organisms per milliliter of blood. Production of two binary toxins by these bacteria can lead to serious complications, including hypotension, shock, and eventually death (Dixon et al., 1999).

Iron acquisition mechanisms are vital for the survival and pathogenesis of many infectious microbes, including B. anthracis (Ratledge & Dover, 2000, Faraldo-Gomez & Sansom, 2003, Wandersman & Delepelaire, 2004). In the mammalian host the concentration of free iron is limiting, with most of the iron sequestered into complexes such as transferrin and hemoglobin (Ratledge & Dover, 2000, Glanfield et al., 2007). B. anthracis, and other pathogenic microbes, succeed in such environments by utilizing various iron acquisition methods including the production of siderophores and a variety of iron transport systems (Ratledge & Dover, 2000, Faraldo-Gomez & Sansom, 2003, Glanfield et al., 2007). Siderophores are high-affinity iron binding molecules that are secreted by bacteria to sequester insoluble ferric iron from host complexes (Ratledge & Dover, 2000). B. anthracis produces two different siderophore molecules, bacillibactin and petrobactin (Cendrowski et al., 2004, Koppisch et al., 2005, Wilson et al., 2006). Although the genes encoding bacillibactin biosynthesis are induced during iron starvation, this molecule is not required for growth of B. anthracis under iron limiting conditions (Cendrowski et al., 2004). B. anthracis strains deficient in bacillibactin biosynthesis are fully virulent in a murine model of anthrax infection (Cendrowski et al., 2004). In contrast, petrobactin biosynthesis has been clearly associated with anthrax pathogenesis. Strains unable to synthesize this siderophore exhibit attenuated growth under iron depleted conditions and are highly attenuated for murine virulence (Cendrowski et al., 2004, Abergel et al., 2006, Lee et al., 2007, Pfleger et al., 2007).

Once secreted siderophores bind extracellular iron they must be reacquired for the organism to take advantage of this system. Uptake of iron bound siderophores typically occurs through specific bacterial membrane receptor complexes (Faraldo-Gomez & Sansom, 2003). The specific transport system involved in petrobactin uptake has yet to be identified. Recently, Zawadzka et al. reported the identification of two B. cereus proteins, FatB and FpuA, capable of binding multiple forms of petrobactin in vitro (Zawadzka et al., 2009a). The B. anthracis homologs of these proteins are encoded by genes designated GBAA5330 and GBAA4766, and have high homology to the B. cereus proteins (95% and 99% amino acid identity, respectively). Expression of these genes were also identified as significantly up-regulated by B. anthracis during iron starvation (Carlson et al., 2009). In the current study, we sought to elucidate the role of these two petrobactin binding proteins in the pathogenesis of B. anthracis.

Results

B. anthracis ΔfpuA exhibits a growth defect in iron depleted media

We sought to determine the role of two genes encoding homologs of the B. cereus petrobactin binding proteins FatB and FpuA (encoded by GBAA5330 and GBAA4766, respectively) in B. anthracis iron acquisition. B. cereus homologs of these proteins are able to bind petrobactin in vitro (Zawadzka et al., 2009a); however the function of these proteins in bacterial iron acquisition and pathogenesis remains to be elucidated. To accomplish this mutant B. anthracis strains lacking either GBAA4766 (ΔfpuA), GBAA5330 (ΔfatB), or both genes (ΔfpuA ΔfatB) were created by markerless gene replacement (Fig. 1 and Table 1). PCR was used to screen all mutant strains for the expected gene deletions, as well as the presence of the virulence plasmid pXO1, prior to use in experiments (data not shown).

Figure 1. Construction of B. anthracis mutants.

Figure 1

Open in a new tab

Gene deletion mutants were created by allelic exchange removing either (A) GBAA4766 (fpuA) or (B) GBAA5330 (fatB). For each deletion, the initial 30 and final 30 nucleotides of the gene were fused creating a markerless deletion. Deleted genetic material was replaced by three stop codons, represented by “X” above.

Table 1.

Bacillus anthracis strains used in this study

Strain Relevant Characteristics Reference
Sterne 34F2 Wild type (pXO1+, pXO2−) (Sterne, 1939)
BA850 34F2, ΔasbABCDEF (Lee et al., 2007)
PC101 34F2, ΔfatB This work
PC102 34F2, ΔfpuA This work
PC103 34F2, ΔfpuA ΔfatB This work
Open in a new tab

We hypothesized that if these proteins were involved in the uptake of petrobactin, mutants lacking the respective genes would exhibit significant growth defects in iron depleted media (IDM), similar to a mutant that cannot synthesize petrobactin. To test this, vegetative bacilli of wild-type B. anthracis, ΔfpuA, ΔfatB, and the ΔfpuA ΔfatB double mutant were inoculated into iron-depleted media (IDM) at an initial OD600 of 0.05 and growth was monitored hourly by measurement of optical density at 600nm. A mutant strain deficient for petrobactin biosynthesis (ΔasbABCDEF) was also included as a control (Cendrowski et al., 2004, Lee et al., 2007). Cultures of wild-type B. anthracis are capable of rapid growth under these conditions, reaching high density within four hours (Fig. 2A, solid diamonds). As previously reported, the petrobactin deficient mutant exhibited a growth defect in this medium, emphasizing the importance of this siderophore in the ability of B. anthracis to replicate in low iron conditions (Fig. 2A, open circles) (Cendrowski et al., 2004). The ΔfatB mutant grew as well as wild-type B. anthracis in IDM, suggesting that this gene is not required for growth in low iron conditions (Fig. 2A, solid squares). In sharp contrast, both mutant strains lacking fpuA (ΔfpuA and ΔfpuA ΔfatB) exhibited severe growth defects in IDM, with each growing to levels nearly identical to the petrobactin-deficient strain (Fig. 2A). Introduction of the ΔfpuA mutant allele into the petrobactin biosynthesis deficient strain (ΔasbABCDEF) did not result in a stronger phenotype, further suggesting a role for FpuA in petrobactin dependent iron acquisition (data not shown). It should also be noted that none of the gene deletions studied here resulted in a general growth defect and all mutant strains grew equally well in brain heart infusion broth (Fig. 2B). Importantly, plasmid based expression of fpuA driven by its native promoter resulted in complementation of ΔfpuA as demonstrated by increased growth in IDM (Fig. 2C). The complementation studies were performed at room temperature as the pBKJ236 vector can not replicate at 37°C. These data indicate that FpuA, but not FatB, is necessary for B. anthracis growth in iron depleted conditions and ΔfpuA mutants are unable to efficiently utilize iron at the low levels available in this medium.

Figure 2. Growth of B. anthracis mutant strains in iron depleted media.

Figure 2

Open in a new tab

(A and B) Wild-type (solid diamonds), ΔfatB (solid squares), ΔfpuA (open triangles), ΔfpuA ΔfatB (open squares), and ΔasbABCDEF (open circles) were grown in either iron depleted media (A) or BHI (B). (C) Wild-type (solid diamonds), ΔfpuA (open triangles), ΔfpuA/pfpuA+ were grown in IDM at 26°C. All cultures were inoculated with vegetative bacilli at an initial OD600 = 0.05 and growth was monitored by measuring change in OD600 over time. Data presented are representative of least three individual experiments.

ΔfpuA mutants are able to produce and secrete petrobactin

It was possible that the growth defect observed for the ΔfpuA mutant was due to an unexpected loss of the ability to produce or secrete petrobactin, rather than an inability to import the siderophore. Since strains deficient in petrobactin production can be rescued by the addition of exogenous petrobactin to the media (Lee et al., 2007), we sought to determine if the ΔfpuA mutants could also be complemented in the same manner. Wild-type, ΔfpuA, and ΔasbABCDEF spores were inoculated into IDM with or without supplemental petrobactin (2.5μM) and growth was monitored by increase in optical density (600nm) over a ten hour time course (Fig. 3). These experiments were performed in triplicate in 96-well plates due to the limited availability of purified petrobactin. All strains used in this study germinated with equal efficiencies (data not shown). As expected, wild-type spores exhibited significant growth in IDM with or without supplemental petrobactin. As previously observed, the ΔasbABCDEF strain, while unable to grow in IDM, grew at a rate similar to wild-type in cultures supplemented with petrobactin. Conversely, the ΔfpuA mutant was not rescued by the addition of supplemental petrobactin. This finding is consistent with the hypothesis that this protein is required for the uptake of iron-bound petrobactin molecules.

Figure 3. Addition of supplemental petrobactin does not rescue binding protein mutants.

Figure 3

Open in a new tab

Wild-type (solid diamonds), ΔfpuA (open triangles), and ΔasbABCDEF (open circles) were grown in either (A) iron depleted media or(B) iron depleted media supplemented with 2.5μM petrobactin. Cultures were inoculated with spores and grown in microtiter plates. Data presented are representative of three individual experiments.

If FpuA is indeed the petrobactin receptor, we hypothesized that the ΔfpuA mutant would exhibit increased levels of petrobactin in culture supernatants due to their inability to import this molecule. The Arnow assay was used to measure the concentration of total catechols in culture supernatants (Arnow, 1937). This colormetric assay measures the presence of both petrobactin and its immediate biosynthetic precursor, 3, 4-dihydroxybenzoic acid (DHB) (Cendrowski et al., 2004, Garner et al., 2004). Vegetative bacilli from wild-type, ΔfpuA, and ΔasbABCDEF were inoculated into IDM, as described previously. Supernatants were isolated from each of the cultures hourly for six hours post-inoculation and analyzed for catechol levels (Fig. 4A). No signal could be detected above background from supernatants isolated during the first two hours of growth. Importantly, supernatants from ΔasbABCDEF cultures, which should not contain any petrobactin, did not exhibit a reaction in this assay at any of the time points tested. The lack of detectable signal in supernatants from these mutants suggests that, at least at these timepoints, petrobactin and 3, 4-DHB are the only catechols in the media. Wild-type B. anthracis cultures did show an increase in catechol levels in supernatants over the time course studied. The ΔfpuA mutant exhibited significantly higher levels of extracellular catechols, presumably petrobactin, at all time points when compared to wild-type bacteria (Fig. 4A).

Figure 4. Increased petrobactin levels observed in ΔfpuA strains.

Figure 4

Open in a new tab

(A) Extracellular catechol levels were measured over time in culture supernatants using the Arnow assay. Data were normalized to the OD600 of cultures at each timepoint and are presented as percent of wild-type petrobactin levels at three hours. No signal could be detected at timepoints prior to three hours. Error bars represent standard deviation of triplicate samples within one experiment. Data are representative of four individual experiments. (B) Comparison of petrobactin accumulation by HPLC. Supernatants from wild-type and ΔfpuA cultures that were grown for six hours and normalized to equivalent OD600 were observed by HPLC. A peak corresponding to an authentic petrobactin standard is observed in both WT and ΔfpuA traces (arrow).

Since the Arnow assay is able to detect multiple molecules, we next sought to measure petrobactin levels directly from culture supernatants. Supernatants obtained from both wild-type and ΔfpuA mutant cultures grown for six hours were normalized to equivalent OD600 and examined using high performance liquid chromatography (HPLC). Peaks corresponding in retention time (37.5 min), UV absorbance spectra (peaks at 260 and 290 nm), and mass spectra (m/z: [M+2H]2+= 360 and [M+2H-H2O]2+= 351) to an authentic petrobactin standard were observed in both WT and ΔfpuA traces. HPLC peak integration revealed an approximate 65% increase of petrobactin accumulation per OD600 in ΔfpuA supernatants over that observed in wild type samples (Fig. 4B). Importantly, this peak was not observed in the ΔasbABCDEF strain which is unable to produce petrobactin (data not shown).

In order to examine the possibility that the altered petrobactin levels were due to increased production, we performed quantitative real-time PCR (Q-PCR) to determine levels of asbA expression after six hours of growth in IDM. Loss of fpuA expression in either single or double mutant strains did not lead to a significant increase in the transcription of this gene (log2 [fold change] = 0.6 ± 0.15). While it is more likely that the increase in petrobactin accumulation detected in ΔfpuA culture supernatants was due to the inability of this strain to import iron-bound petrobactin, we can not exclude the possibility that this slight increase in transcription is responsible for this observation.

Gallium associated growth inhibition requires FpuA

One test that has been used to show siderophore transport in other systems is gallium sensitivity. Gallium has been shown to bind to various siderophores, including petrobactin (Banin et al., 2008, Zawadzka et al., 2009b). Gallium is toxic to most bacterial cells and transport of siderophore-gallium complexes across the membrane often leads to cell death (Olakanmi et al., 2000, Banin et al., 2008). It is, therefore, possible to correlate defects in siderophore transport with increased resistance to gallium. This system was used here to more definitively show a decrease in petrobactin in mutant strains. Wild-type, ΔasbABCDEF, and ΔfpuA strains were grown in IDM with or without 20μM gallium sulfate. Growth was monitored by change in optical density (600nm) over time. The addition of gallium to cultures of wild-type B. anthracis led to a significant decrease in growth over time, with a peak inhibition of 40% observed at two hours (Table 2). As would be expected, growth of the petrobactin deficient strain ΔasbABCDEF was not significantly impaired by the addition of gallium to the culture media (Table 2). Finally, deletion of the putative petrobactin binding protein, FpuA, protects B. anthracis from gallium induced growth inhibition, suggesting that this strain is defective in transport of gallium across the membrane. The fact that strains lacking the putative receptor have the same phenotype as those deficient in petrobactin biosynthesis supports a decreased transport of petrobactin in these mutants. These data provide further evidence that FpuA is required for transport of petrobactin into B. anthracis.

Table 2.

Resistance of ΔfpuA to gallium treatment.

hours
1 2 3 4
wild-type 25.0a, b 40.2 33.4 27.5
ΔasbABCDEF 4.6 6.0 8.7 5.0
ΔfpuA 4.6 4.6 4.6 8.0
Open in a new tab
a

% growth inhibition

b

results are representative of 4 independent experiments

Role of FpuA in B. anthracis virulence

Since it is known that mutants unable to synthesize petrobactin exhibit significant attenuation in murine virulence (Cendrowski et al., 2004, Pfleger et al., 2008), we hypothesized that strains unable to import this siderophore would exhibit similar levels of attenuation. To test this hypothesis, we employed a murine model of inhalational anthrax. DBA/2J mice were used for these studies as these mice are known to be susceptible to B. anthracis Sterne strain (Welkos et al., 1986). DBA/2J mice (n = 8) were inoculated intratracheally with wild-type, ΔfpuA, or ΔasbABCDEF spores and monitored over the course of fourteen days. When inoculated with 1.5×105 spores, mice receiving B. anthracis ΔfpuA exhibited increased survival compared to those infected with wild-type spores (Fig. 5). At this dose, all mice infected with wild-type spores succumbed to infection within five days post infection, while 87.5% of mice infected with ΔfpuA (p = 0.0006) survived the full two week experiment, with the only death occurring on the first day post-infection (Fig. 5A). In fact, even at a ten-fold higher dose (1.5×106 spores/mouse), the ΔfpuA mutant showed attenuated murine virulence compared to wild-type spores (Fig. 5B). At this dose, 75% of mice infected with ΔfpuA (p = 0.01) showed no sign of disease after fourteen days, while all of the mice infected with wild-type spores succumbed to infection by day four (Fig. 5B). These results are similar to what was observed for the mutant strain defective in petrobactin biosynthesis (data not shown and (Cendrowski et al., 2004)). In order to calculate an accurate LD50 for both the ΔfpuA and ΔasbABCDEF strains, mice were infected with 1.5×107 spores of each of the mutants. Even at this dose four of eight mice infected with ΔfpuA failed to show any sign of disease during the course of the infection. The LD50 of ΔfpuA was determined to be 2.2×107 spores/mouse, approximately three logs higher than that of wild-type spores (Table 3). Interestingly, when mice were infected with a similar dose of the petrobactin biosynthesis deficient strain, ΔasbABCDEF, 100% of mice succumbed to disease by day four post-infection, indicating that ΔfpuA is slightly more attenuated than this strain (p = 0.0002). In fact, the ΔasbABCDEF mutant spores exhibited an LD50 nearly ten fold lower than that of the ΔfpuA mutants (Table 3).

Figure 5. Attenuation of virulence of the ΔfpuA strain in a murine model of infection.

Figure 5

Open in a new tab

DBA/2J mice were infected by intratracheal infection with WT (filled diamonds) or ΔfpuA (open triangles) spores at 1×105 (A), or 1×106 (B) spores per mouse. Mice were monitored for fourteen days. Survival curves for ΔfpuA were significantly different from wild-type at all doses tested by the log-rank test (p values indicated above).

Table 3.

Attenuated virulence of ΔfpuA in DBA/2J mice.

Strain LD50
wild-type 7.5 × 103
ΔfpuA 2.2 × 107
ΔasbABCDEF 2.7 × 106
Open in a new tab

Discussion

The ability of a microorganism to survive in low iron environments, such as a mammalian host, typically requires specialized systems of iron acquisition and uptake. Although B. anthracis encodes the biosynthesis machinery for two siderophores, only one of these, petrobactin, is required for efficient bacterial growth in low iron conditions and for full virulence in a murine model of anthrax infection (Cendrowski et al., 2004, Lee et al., 2007). Since this siderophore plays such an important role in B. anthracis pathogenesis, we sought to elucidate the role of two proteins recently implicated in petrobactin binding. We created deletion mutants lacking two putative siderophore binding proteins, FatB and FpuA, encoded by GBAA5330 and GBAA4766, respectively. B. cereus homologs of these proteins were recently shown to have the ability to bind to petrobactin, however their role in the pathogenesis of the bacterium remained untested. (Zawadzka et al., 2009a). Here we have demonstrated the role of these genes during growth in low iron conditions as well as their effect on murine virulence.

Despite the ability of both FpuA and FatB to bind petrobactin in vitro, only FpuA appears important for growth of B. anthracis under iron limiting conditions. Mutants lacking fpuA exhibits a severe growth defect when either vegetative bacteria or spores were inoculated into IDM (Figs. 2, 3A). These cultures also exhibit growth kinetics nearly identical to those of B. anthracis strains unable to synthesize petrobactin (Fig. 2, 3A and (Cendrowski et al., 2004, Lee et al., 2007)). Importantly, although vegetative bacilli of these strains (ΔasbABCDEF and ΔfpuA) are capable of some growth in IDM (Fig. 2A), these strains are not capable of outgrowth from spores in this medium (Fig. 3). Although the growth kinetics were slightly slower than wild-type bacteria, replacement of this gene in trans did restore the ability of the bacteria to grow exponentially in IDM (Fig. 2C). It should be noted, a B. anthracis strain with deletions in both putative petrobactin binding proteins (ΔfpuA ΔfatB) grew to the same levels as ΔfpuA single mutants (Fig. 1 and data not shown). In contrast, mutants lacking FatB are not significantly impaired in their ability to grow in IDM (Fig. 2). In fact, growth curves for the ΔfatB mutant in IDM were nearly identical to those of wild-type bacilli (Fig. 2, filled squares). These data indicate that FatB does not play a role in uptake of petrobactin by B. anthracis. Curiously, the FatB homolog in B. subtilis has recently been shown to be required for uptake of petrobactin (Zawadzka et al., 2009b). It should be noted that B. subtilis does not make petrobactin, nor does it have an ortholog of fpuA. A role for this protein in iron metabolism in B. cereus remains to be tested.

Importantly, we have shown that ΔfpuA bacteria are not deficient in petrobactin synthesis or secretion, but rather the growth defect of this strain appears to be caused by an inability to use this siderophore. The addition of supplemental petrobactin had no effect on the growth of the ΔfpuA strain (Fig. 3B). Addition of excess siderophore is able to restore the growth of B. anthracis strains deficient in petrobactin biosynthesis, indicating that the ΔfpuA is phenotypically distinct from these strains. Petrobactin production from ΔfpuA bacteria was conclusively shown using both the Arnow assay and HPLC (Fig. 4). At the time points tested in the Arnow assay, no signal could be detected in the supernatants of strains unable to synthesize petrobactin, raising the possibility that bacillibactin was not being secreted at these time points. This finding is not surprising given the fact that most bacillibactin secretion has been seen following 24 hour cultures and recently it was shown that bacillibactin secretion was undetectable until the bacteria had grown for 10 or more hours (Wilson et al., 2006, Pfleger et al., 2008, Wilson et al., 2009). Although no catechols were detected using the Arnow assay on ΔasbABCDEF supernatants, these molecules were observed from both wild-type bacteria and the ΔfpuA mutant. Also, petrobactin was specifically detected in both wild-type and ΔfpuA supernatants using HPLC. Interestingly, petrobactin levels were significantly higher in supernatants of ΔfpuA cultures than in those of wild-type B. anthracis. It is likely that the ΔfpuA mutant strain is synthesizing petrobactin at levels comparable to wild-type bacteria, as only a minimal increase was observed in the transcription of the petrobactin biosynthetic operon during growth of this strain in IDM. These results suggest that the increased accumulation of petrobactin in the supernatants of ΔfpuA cultures is due to the inability of these strains to import petrobactin, thus trapping the iron-bound siderophore in the extracellular environment and depriving the bacterium of the necessary iron. Strains lacking the putative petrobactin receptor also exhibit a resistance to gallium induced growth inhibition. Since gallium can be bound by petrobactin, these results further support the idea that the ΔfpuA mutant is deficient in petrobactin uptake. Taken together, the inability of supplemental petrobactin to restore the growth of ΔfpuA mutants, the increased accumulation of petrobactin in the ΔfpuA culture supernatants, the increased resistance of ΔfpuA to gallium, and the fact that the homologous protein from B. cereus binds petrobactin, clearly implicate FpuA in the uptake of iron-bound petrobactin.

Although it was clear that ΔfpuA strains were deficient for growth under iron limiting conditions in vitro, it was also important to determine the role of this protein during infection. We have shown that the loss of a single putative petrobactin receptor in the ΔfpuA strain leads to severe attenuation in this infection model (Fig. 5). The LD50 of the ΔfpuA mutant was found to be nearly 3,000 times that of wild-type B. anthracis Sterne (Table 3). Similar to what was observed in the in vitro experiments, no significant difference was seen in the virulence of the double mutant (ΔfpuA ΔfatB) when compared to the mutant lacking only fpuA (data not shown). The level of attenuation observed for these mutants further emphasizes the importance of iron acquisition through petrobactin and its receptor complex in B. anthracis pathogenesis.

The observed attenuation of strains unable to import petrobactin is consistent with published reports on the attenuation of petrobactin biosynthesis mutants (Cendrowski et al., 2004, Pfleger et al., 2008). Although the growth phenotypes of ΔfpuA and ΔasbABCDEF strains appear very similar in IDM, slight differences in murine virulence were observed between these two strains. The virulence of these two mutant strains appears similar at lower doses, with neither strain killing a significant number of the mice tested. A difference in virulence was observed at higher doses however, and the LD50 of ΔasbABCDEF mutants was found to be approximately eight fold lower than that of the ΔfpuA strain (Table 3). These results indicate that ΔfpuA mutants are slightly less virulent than even the petrobactin biosynthesis mutant, leading to the hypothesis that this receptor protein plays a secondary function during mammalian infection. It is possible that this receptor is able to scavenge siderophore molecules produced by other bacterial species, providing the bacterium an increased survival advantage within its host. Uptake of exogenous siderophores has been shown in other bacterial species including B. cereus and Pseudomonas aeruginosa (Ollinger et al., 2006, Greenwald et al., 2009), however the exact mechanism involved in B. anthracis remains to be elucidated.

The data presented here, combined with the fact that FpuA is able to bind petrobactin in vitro (Zawadzka et al., 2009a), implicate this protein as a part of the petrobactin uptake system. FpuA has homology to substrate binding proteins from ABC family transporters (Miethke & Marahiel, 2007). In Gram-positive bacteria, these are membrane bound proteins which interact with a transmembrane permease to transport molecules into the bacterium (Miethke & Marahiel, 2007). The identity of the remaining components of this ABC transport system remains to be elucidated. Although some candidate proteins have been identified (Zawadzka et al., 2009a), further study is required to show a role for these proteins in petrobactin uptake.

Iron uptake systems are thought to be strong candidate targets for vaccine and therapeutic treatments, since they are often implicated in pathogen virulence and their surface localization leaves them exposed to extracellular molecules and the immune system (Glanfield et al., 2007). Studies in E. coli, S. pneumoniae and S. aureus have shown that vaccines targeting iron uptake systems can improve immune responses and disease survival in animal infection models (Jomaa et al., 2006, Kuklin et al., 2006, Alteri et al., 2009). It has been postulated that iron uptake proteins from B. anthracis, including the two examined here, could serve as vaccine targets (Gat et al., 2006). Antibodies to these proteins were present in the serum of infected guinea pigs, indicating that they are expressed by the bacterium during infection (Gat et al., 2006). Since disruption of a single siderophore binding protein/iron uptake system in the ΔfpuA mutant leads to such dramatic phenotypes in B. anthracis growth and virulence it is likely that this protein could be a prime target for the development of new therapeutic measures. The data presented here indicate that a therapeutic blocking this receptor could greatly reduce the bacterium’s ability to replicate within the low iron environment of the human host. Such a treatment could, hypothetically, allow the host immune system to eliminate the pathogen before bacterial replication and toxin production reach lethal levels.

Given that recombinant forms of B. cereus FpuA have been shown to bind petrobactin directly, that these proteins are nearly identical to their B. anthracis homologs, and the genetic evidence provided in this study, it is highly likely that FpuA serves as the receptor for petrobactin. In contrast, our data indicate that FatB does not play a role in petrobactin uptake in B. anthracis. Finally, the severe attenuation of ΔfpuA strains in a murine model of inhalation anthrax raises the possibility that this protein might be a useful target for the development of new vaccines and therapeutics.

Experimental procedures

Bacterial strains and growth conditions

All of the work described in this manuscript was performed using the Sterne 34F2 (pXO1+, pXO2−) strain of B. anthracis. Initial spore stocks were prepared as follows. Strains were grown in modified G medium (Kim & Goepfert, 1974) for three days at 37°C with shaking and spores were prepared as previously described (Passalacqua et al., 2006). Spores were stored at room temperature in sterile water and titered by hemacytometer (spores/ml) for animal studies. For all iron depletion studies, iron depleted media (IDM) was used (Cendrowski et al., 2004). Experimental design for growth curves, Arnow assays, and RNA isolation experiments were done as follows. Spores were germinated in brain heart infusion (BHI) broth and incubated overnight with shaking, at room temperature. The following day, vegetative bacilli were diluted 1:100 in fresh BHI and grown 1–2 hours at 37°C. Actively growing bacilli were then washed 3x with phosphate buffered saline (PBS) (Invitrogen) and 3x with IDM to ensure removal of nutrients and potential iron sources carried over from the BHI. Washed bacteria were then used to inoculate cultures of IDM or iron replete media (IRM) [IDM + 20μM ferrous sulfate] at an OD600 = 0.05. Experiments performed with petrobactin were started with 1×105 spores per well in 96 well plates and performed in a Molecular Devices M2 plate reader. Growth was monitored by change in OD600 over ten hours. For gallium resistance experiments, strains were grown in IDM with or without 20μM gallium sulfate and growth was monitored over time. Percent growth inhibition was calculated as ([(OD600 IDM+gallium)/(OD600 IDM)]*100).

Mutant Construction

Mutant B. anthracis strains were constructed using allelic exchange as previously described (Janes & Stibitz, 2006). The ΔfpuA and ΔfatB single mutants were created in the wild-type Sterne 34F2 background by deletion of either GBAA4766 or GBAA 5330, respectively. The double mutant was created by deleting GBAA4766 in the ΔfatB single deletion strain. The ΔasbABCDEF B. anthracis strain used in these studies has been reported previously (Lee et al., 2007). The sequence of all oligonucleotides used in creating and screening these mutant strains are available on request.

Complementation of gene deletions

fpuA along with the upstream promoter region (120 bp) was cloned into the temperature sensitive plasmid, pBKJ401 (a derivative of pBKJ236 (Janes & Stibitz, 2006)). The resulting vector, pfpuA+, was then introduced to ΔfpuA by conjugal transfer. Conjugates were selected by screening for erythromycin resistance. Because this vector can not replicate at 37°C, all growth of the plasmid containing strain was performed at 25°C.

Real time RT-PCR

cDNA synthesis was performed using Superscript III (Invitrogen) and 1 μg of total RNA. Real time reactions were performed with a 1:500 final dilution of template cDNA. Primer sets were designed for genes indicated using Primer 3 (Rozen & Skaletsky, 2000) and reactions were carried out on an BioRad ICycler real time machine using SYBR Green (BioRad). The acoB gene (GBAA2775) was used as the internal reference, as its expression was previously determined to remain unchanged during iron starvation (data not shown).

Measurement of petrobactin

Extracellular levels of catechols were measured using the Arnow assay (Arnow, 1937). Bacteria were grown in IDM and samples of cultures were removed at the indicated timepoints. Cultures were centrifuged and filtered using 0.2μm filters to remove bacteria. Samples were mixed with equal volumes of 0.5M HCl, nitrate-molybdate reagent (10% sodium nitrate and 10% sodium molybdate), and 1N NaOH. Positive reactions produce a red color and absorbance was determined at 515nm. Samples were normalized to OD600 of the original culture and data are presented as percent of wild-type petrobactin levels at the three hour timepoint.

For direct measurement of petrobactin, supernatants were obtained from three cultures of both wild type Sterne and ΔfpuA following six hours of growth in IDM. All cultures were diluted with IDM to equal OD600 and sterile-filtered. Samples of individual culture supernatants were mixed with nine volumes of LCMS grade methanol and incubated at −20° C for 20 minutes. Methanol crashes were centrifuged at 14,000 × g and 1600 μl of each supernatant was transferred to disposable 1.7 ml tubes for evaporation by vacuum centrifugation. Dried pellets were re-dissolved in 80 μl of ddH2O and directly injected onto a Beckman Coulter System Gold for HPLC analysis. Peaks were separated using a C18 analytical reverse-phase column (Waters XBridge C18, 5 μm, 6 × 250 mm) at a rate of 1 ml/min with the following solvent system: 5% methanol (MeOH) in water with 0.1% formic acid (FA) for 25 minutes then increased to 50% MeOH plus 0.1% FA over the next 28.5 minutes followed by a cleanup step with 100% MeOH plus 0.1% FA for the remaining 10 minutes. Mass spectrometric detection was performed by direct injection of a collected peak onto a Finnigan LTQ Linear Ion Trap (ThermoElectron) scanning in positive mode. An authentic standard of petrobactin was used for comparisons. The A290 peaks corresponding to petrobactin were integrated using the area annotation feature included in the 32 Karat Software package (Beckman Coulter). Peak areas corresponding to petrobactin were normalized with the mean of wild type sample values equaling 100% petrobactin production.

Murine infections

Intratracheal infections of DBA/2J mice (Jackson Laboratories) were performed as previously described (Heffernan et al., 2007). Groups of eight mice were infected with either wild-type or mutant spores at a variety of doses ranging from 1×105 through 1×108 spores per mouse. Mice were monitored over a period of fourteen days. LD50 values were calculated using the Moving Average Interpolation program available at (http://falkow.stanford.edu/whatwedo/software/software.html) (Kim et al., 2003).

Acknowledgments

The authors would like to thank Yousong Ding for assistance with mass spectrometry analysis. This work was sponsored by the NIH/NIAID Regional Center of Excellence for Bio-defense and Emerging Infectious Diseases Research (RCE) Program. The authors wish to acknowledge membership within and support from the Region V ‘Great Lakes’ RCE (NIH award 1-U54-AI-057053). This research was also supported by HHS contract N266200400059C-N01-AI-40059. PEC received funding from training grant T32-AI-07528, “Molecular Mechanisms of Microbial Pathogenesis.”

References

  1. Abergel RJ, Wilson MK, Arceneaux JE, Hoette TM, Strong RK, Byers BR, et al. Anthrax pathogen evades the mammalian immune system through stealth siderophore production. Proc Natl Acad Sci U S A. 2006;103:18499–18503. doi: 10.1073/pnas.0607055103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alteri CJ, Hagan EC, Sivick KE, Smith SN, Mobley HL. Mucosal immunization with iron receptor antigens protects against urinary tract infection. PLoS Pathog. 2009;5:e1000586. doi: 10.1371/journal.ppat.1000586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arnow LE. Colorimetric determination of the components of 3, 4 - dihydroxyphenylalanine-tyrosine mixtures. J Biol Chem. 1937;118:531–537. [Google Scholar]
  4. Banin E, Lozinski A, Brady KM, Berenshtein E, Butterfield PW, Moshe M, et al. The potential of desferrioxamine-gallium as an anti-Pseudomonas therapeutic agent. Proc Natl Acad Sci U S A. 2008;105:16761–16766. doi: 10.1073/pnas.0808608105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Carlson PE, Jr, Carr KA, Janes BK, Anderson EC, Hanna PC. Transcriptional profiling of Bacillus anthracis Sterne (34F2) during iron starvation. PLoS One. 2009;4:e6988. doi: 10.1371/journal.pone.0006988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cendrowski S, MacArthur W, Hanna P. Bacillus anthracis requires siderophore biosynthesis for growth in macrophages and mouse virulence. Mol Microbiol. 2004;51:407–417. doi: 10.1046/j.1365-2958.2003.03861.x. [DOI] [PubMed] [Google Scholar]
  7. Dixon TC, Meselson M, Guillemin J, Hanna PC. Anthrax. N Engl J Med. 1999;341:815–826. doi: 10.1056/NEJM199909093411107. [DOI] [PubMed] [Google Scholar]
  8. Dixon TC, Fadl AA, Koehler TM, Swanson JA, Hanna PC. Early Bacillus anthracis-macrophage interactions: intracellular survival survival and escape. Cell Microbiol. 2000;2:453–463. doi: 10.1046/j.1462-5822.2000.00067.x. [DOI] [PubMed] [Google Scholar]
  9. Doganay L, Welsby PD. Anthrax: a disease in waiting? Postgrad Med J. 2006;82:754–756. doi: 10.1136/pgmj.2005.044487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dutz W, Kohout E. Anthrax. Pathol Annu. 1971;6:209–248. [PubMed] [Google Scholar]
  11. Dutz W, Kohout-Dutz E. Anthrax. Int J Dermatol. 1981;20:203–206. doi: 10.1111/j.1365-4362.1981.tb00461.x. [DOI] [PubMed] [Google Scholar]
  12. Faraldo-Gomez JD, Sansom MS. Acquisition of siderophores in gram-negative bacteria. Nat Rev Mol Cell Biol. 2003;4:105–116. doi: 10.1038/nrm1015. [DOI] [PubMed] [Google Scholar]
  13. Garner BL, Arceneaux JE, Byers BR. Temperature control of a 3,4-dihydroxybenzoate (protocatechuate)-based siderophore in Bacillus anthracis. Curr Microbiol. 2004;49:89–94. doi: 10.1007/s00284-004-4286-7. [DOI] [PubMed] [Google Scholar]
  14. Gat O, Grosfeld H, Ariel N, Inbar I, Zaide G, Broder Y, et al. Search for Bacillus anthracis potential vaccine candidates by a functional genomic-serologic screen. Infect Immun. 2006;74:3987–4001. doi: 10.1128/IAI.00174-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Glanfield A, McManus DP, Anderson GJ, Jones MK. Pumping iron: a potential target for novel therapeutics against schistosomes. Trends Parasitol. 2007;23:583–588. doi: 10.1016/j.pt.2007.08.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Greenwald J, Nader M, Celia H, Gruffaz C, Geoffroy V, Meyer JM, et al. FpvA bound to non-cognate pyoverdines: molecular basis of siderophore recognition by an iron transporter. Mol Microbiol. 2009;72:1246–1259. doi: 10.1111/j.1365-2958.2009.06721.x. [DOI] [PubMed] [Google Scholar]
  17. Guidi-Rontani C, Weber-Levy M, Labruyere E, Mock M. Germination of Bacillus anthracis spores within alveolar macrophages. Mol Microbiol. 1999;31:9–17. doi: 10.1046/j.1365-2958.1999.01137.x. [DOI] [PubMed] [Google Scholar]
  18. Heffernan BJ, Thomason B, Herring-Palmer A, Hanna P. Bacillus anthracis anthrolysin O and three phospholipases C are functionally redundant in a murine model of inhalation anthrax. FEMS Microbiol Lett. 2007;271:98–105. doi: 10.1111/j.1574-6968.2007.00713.x. [DOI] [PubMed] [Google Scholar]
  19. Inglesby TV, Henderson DA, Bartlett JG, Ascher MS, Eitzen E, Friedlander AM, et al. Anthrax as a biological weapon: medical and public health management. Working Group on Civilian Biodefense. JAMA. 1999;281:1735–1745. doi: 10.1001/jama.281.18.1735. [DOI] [PubMed] [Google Scholar]
  20. Inglesby TV, O’Toole T, Henderson DA, Bartlett JG, Ascher MS, Eitzen E, et al. Anthrax as a biological weapon, 2002: updated recommendations for management. JAMA. 2002;287:2236–2252. doi: 10.1001/jama.287.17.2236. [DOI] [PubMed] [Google Scholar]
  21. Janes BK, Stibitz S. Routine markerless gene replacement in Bacillus anthracis. Infect Immun. 2006;74:1949–1953. doi: 10.1128/IAI.74.3.1949-1953.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jomaa M, Terry S, Hale C, Jones C, Dougan G, Brown J. Immunization with the iron uptake ABC transporter proteins PiaA and PiuA prevents respiratory infection with Streptococcus pneumoniae. Vaccine. 2006;24:5133–5139. doi: 10.1016/j.vaccine.2006.04.012. [DOI] [PubMed] [Google Scholar]
  23. Kim CC, Monack D, Falkow S. Modulation of virulence by two acidified nitrite-responsive loci of Salmonella enterica serovar Typhimurium. Infect Immun. 2003;71:3196–3205. doi: 10.1128/IAI.71.6.3196-3205.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kim HU, Goepfert JM. A sporulation medium for Bacillus anthracis. J Appl Bacteriol. 1974;37:265–267. doi: 10.1111/j.1365-2672.1974.tb00438.x. [DOI] [PubMed] [Google Scholar]
  25. Koppisch AT, Browder CC, Moe AL, Shelley JT, Kinkel BA, Hersman LE, et al. Petrobactin is the primary siderophore synthesized by Bacillus anthracis str. Sterne under conditions of iron starvation. Biometals. 2005;18:577–585. doi: 10.1007/s10534-005-1782-6. [DOI] [PubMed] [Google Scholar]
  26. Kuklin NA, Clark DJ, Secore S, Cook J, Cope LD, McNeely T, et al. A novel Staphylococcus aureus vaccine: iron surface determinant B induces rapid antibody responses in rhesus macaques and specific increased survival in a murine S. aureus sepsis model. Infect Immun. 2006;74:2215–2223. doi: 10.1128/IAI.74.4.2215-2223.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lee JY, Janes BK, Passalacqua KD, Pfleger BF, Bergman NH, Liu H, et al. Biosynthetic analysis of the petrobactin siderophore pathway from Bacillus anthracis. J Bacteriol. 2007;189:1698–1710. doi: 10.1128/JB.01526-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Miethke M, Marahiel MA. Siderophore-based iron acquisition and pathogen control. Microbiol Mol Biol Rev. 2007;71:413–451. doi: 10.1128/MMBR.00012-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Olakanmi O, Britigan BE, Schlesinger LS. Gallium disrupts iron metabolism of mycobacteria residing within human macrophages. Infect Immun. 2000;68:5619–5627. doi: 10.1128/iai.68.10.5619-5627.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ollinger J, Song KB, Antelmann H, Hecker M, Helmann JD. Role of the Fur regulon in iron transport in Bacillus subtilis. J Bacteriol. 2006;188:3664–3673. doi: 10.1128/JB.188.10.3664-3673.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Passalacqua KD, Bergman NH, Herring-Palmer A, Hanna P. The superoxide dismutases of Bacillus anthracis do not cooperatively protect against endogenous superoxide stress. J Bacteriol. 2006;188:3837–3848. doi: 10.1128/JB.00239-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pfleger BF, Lee JY, Somu RV, Aldrich CC, Hanna PC, Sherman DH. Characterization and analysis of early enzymes for petrobactin biosynthesis in Bacillus anthracis. Biochemistry. 2007;46:4147–4157. doi: 10.1021/bi6023995. [DOI] [PubMed] [Google Scholar]
  33. Pfleger BF, Kim Y, Nusca TD, Maltseva N, Lee JY, Rath CM, et al. Structural and functional analysis of AsbF: origin of the stealth 3,4-dihydroxybenzoic acid subunit for petrobactin biosynthesis. Proc Natl Acad Sci U S A. 2008;105:17133–17138. doi: 10.1073/pnas.0808118105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ratledge C, Dover LG. Iron metabolism in pathogenic bacteria. Annu Rev Microbiol. 2000;54:881–941. doi: 10.1146/annurev.micro.54.1.881. [DOI] [PubMed] [Google Scholar]
  35. Rozen S, Skaletsky HJ. Primer3 on the WWW for general users and for biologist programmers. In: Krawetz S, Misener S, editors. Bioinformatics Methods and Protocols: Methods in Molecular Biology. Totowa, NJ: Humana Press; 2000. pp. 365–386. [DOI] [PubMed] [Google Scholar]
  36. Sterne M. The immunization of laboratory animals against anthrax. Onderstepoort. J Vet Sci Anim Ind. 1939;13:313–317. [Google Scholar]
  37. Wandersman C, Delepelaire P. Bacterial iron sources: from siderophores to hemophores. Annu Rev Microbiol. 2004;58:611–647. doi: 10.1146/annurev.micro.58.030603.123811. [DOI] [PubMed] [Google Scholar]
  38. Welkos SL, Keener TJ, Gibbs PH. Differences in susceptibility of inbred mice to Bacillus anthracis. Infect Immun. 1986;51:795–800. doi: 10.1128/iai.51.3.795-800.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Wilson MK, Abergel RJ, Raymond KN, Arceneaux JE, Byers BR. Siderophores of Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis. Biochem Biophys Res Commun. 2006;348:320–325. doi: 10.1016/j.bbrc.2006.07.055. [DOI] [PubMed] [Google Scholar]
  40. Wilson MK, Abergel RJ, Arceneaux JE, Raymond KN, Byers BR. Temporal production of the two Bacillus anthracis siderophores, petrobactin and bacillibactin. Biometals. 2009 doi: 10.1007/s10534-009-9272-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zawadzka AM, Abergel RJ, Nichiporuk R, Andersen UN, Raymond KN. Siderophore-mediated iron acquisition systems in Bacillus cereus: Identification of receptors for anthrax virulence-associated petrobactin. Biochemistry. 2009a;48:3645–3657. doi: 10.1021/bi8018674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zawadzka AM, Kim Y, Maltseva N, Nichiporuk R, Fan Y, Joachimiak A, et al. Characterization of a Bacillus subtilis transporter for petrobactin, an anthrax stealth siderophore. Proc Natl Acad Sci U S A. 2009b doi: 10.1073/pnas.0904793106. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES